Ungulate Taxonomy 2011

Ung ulate Ta xonomy
—-1
—0
© 2011 The Johns Hopkins University Press
UNCORRECTED PROOF —+1
Do not quote for publication until verified with finished book. All rights reserved. No portion of this may be reproduced or distributed without permission.
NOT FOR SALE OR DISTRIBUTION
349-47558_ch00_1P.indd i 5/31/11 6:06 PM

-1—
0—
+1— UNCORRECTED PROOF
349-47558_ch00_1P.indd ii 5/31/11 6:06 PM

Ung ulate Taxonomy
colin groves and peter grubb
TH E J OH NS H O P K I N S U N I V ER S I T Y P R ES S | ba lt i m o r e
—-1
—0
349-47558_ch00_1P.indd iii 5/31/11 6:06 PM

All rights reserved. Published 2011
Printed in the United States of America on acid-free paper
9 8 7 6 5 4 3 2 1
The Johns Hopkins University Press

2715 North Charles Street
Baltimore, Maryland 21218-4363
www.press.jhu.edu
[[CIP data to come]]
A catalog record for this book is available from the British

Library.
Special discounts are available for bulk purchases of this book. For
more information, please contact Special Sales at 410-516-6936
or [email protected].
The Johns Hopkins University Press uses environmentally

friendly book materials, including recycled text paper that
is composed of at least 30 percent post- consumer waste,
whenever possible. All of our book papers are acid-free, and
-1— our jackets and covers are printed on paper with recycled
0— content.
349-47558_ch00_1P.indd iv 5/31/11 6:06 PM

Contents
Preface vii
Theory of Ungulate Taxonomy 1
part i PERISSODACTYLA
1 Equidae 13
2 Tapiridae 18
3 Rhinocerotidae 21
part ii ARTIODACTYLA
4 Tylopoda 29
5 Suidae 33
6 Hippopotamidae 54
7 Tragulidae and Moschidae 56
8 Antilocapridae 63
9 Giraffidae 64
10 Cervidae 71
11 Bovidae 108
References 281
Index 000
—-1
—0
349-47558_ch00_1P.indd v 5/31/11 6:06 PM

-1—
0—
349-47558_ch00_1P.indd vi 5/31/11 6:06 PM

Preface
it has now been nearly a century since the last major survey of ungu-
late taxonomy appeared (Lydekker, 1913, 1915a, b; Lydekker & Blaine,
1914a, b). A new treatment would be needed only if, in the meantime,
there have been major advances in knowledge and understanding; and
indeed there have. This is our fi rst major justification for writing this
new book.
In most biota, ungulates are among the most visible fauna on the
landscape. In national parks they are readily spotted and much ad-
mired by visitors, and it is easy to assume that because these animals
are so familiar, everything is known about them, especially how many
species there are. Such an assumption would be wrong. Mammalogists
expect diversity among small mammals, and seek it out, but they have
grown used to accepting that larger ones, because they are thought to
be highly mobile (and indeed they may be), are bound to be much less
diverse and thus will yield up less information about ecosystems, bio-
geography, and so on. It may be taken for granted that an antelope of a
particular “kind” is, to an extent, expendable: if it becomes endangered
in one place, replacements can be found elsewhere and imported in
order to top up the local gene pool—such, at any rate, is the assumption,
but it needs to be tested: are the local gene pools really exchangeable,
or are some of them unique? Taxonomy can go some way toward test-
ing such propositions.
The coming of the molecular revolution has begun to reopen such
questions, but morphological research has often not kept up. Because
both of us have been traditional skin-and-skull taxonomists, we felt
that this was a gap that must be fi lled, and it was only by going through
the different ungulate groups, one by one, that we would fi nd out ex-
actly which gaps remained to be fi lled, and what research still needed
to be done.
We do not intend to present this book as a finished proposition. Time
and again, we remark that “more research is necessary”; we want, above
all, to stimulate more taxonomic research on ungulates. In some cases
we have been able to break new ground, and we hope that these exam-
ples will be examined by colleagues and, if found acceptable, used as
templates for other studies.
We have both benefited enormously from the assistance of museum —-1
curators, discussions with colleagues, and help and advice from families —0
349-47558_ch00_1P.indd vii 5/31/11 6:06 PM

viii preface
and friends—going back at least 40 years. We would

A NOTE ON SYNONYMY
like to take this opportunity to thank (in alphabetical
order): Alexei Abramov, Faris Al-Timimi, Renate An- The synonymies given for most of the taxa we list are
germann, Bruce Banwell, Kurt Benirschke, the late incomplete. Given that full synonymy lists are readily
Biswamoy Biswas, Boeadi, Gennady Boeskorov, Peter available in other sources, we have thought it unnec-
van Bree, Isabella Capellini, Sujit Chakraborty, Lucas essary to repeat them here. In general we include only
Chin, Bill Clark, Cäsar Claude, Juliet Clutton-Brock, those synonyms that are little known or controversial
Gordon Corbet, Loïc Costeur, Woody Cotterill, the or otherwise contextual. Synonymy entries that we
late Peter Crowcroft, Jacques Cuisin, the late Pierre consider doubtful are preceded by a question mark.
Dandelot, J. C. Daniel, Bijan Dareshuri, Shantini Daw-
son, Jim Dolan, Keith Dunmall, Feng Zuojiang, Pruthu
IN MEMORIAM: PETER GRUBB
Fernando, Wolfgang Frey, Val Geist, Alan Gentry,
Anthea Gentry, Arnaud Greth, David Happold, Mal- My friend and colleague Peter Grubb died in late De-
colm Harmon, David Harrison, Larry Heaney, Kris cember 2006 (Groves, 2006b).
Helgen, Mahmoud Reza Hemami, Helmut Hemmer, We were fi rst introduced by the late Peter Jewell
Daphne Hills, the late John Hill, Kes Hillman, the late in 1970, and we repaired to a pub in London to talk
Dick Hooijer, Barry Hughes, Paula Jenkins, the late about our mutual interests. I knew that Peter had
Peter Jewell, Tolga Kankiliç, Mahmoud Karami, Chris- done fieldwork on Soay sheep on the island of Hirta
tian Kern, Jonathan Kingdon, Steve Kingswood, Dieter in the Outer Hebrides, and on giant tortoises on
Kock, Katrin Kohmann, Richard Krafft, the late Arlene Aldabra in the Indian Ocean, so at fi rst I was slightly
Kumamoto, Adrian Lister, Ibnu Maryanto, Frieder surprised that he was interested in mammalian tax-
Mayer, Ji Mazak, the late Vratja Mazák, Erik Meijaard, onomy. It turned out that this was, in fact, a consum-
the late Erna Mohr, Guy Musser, Gertrud Neumann- ing passion for him, one that had overtaken animal
Denzau, William Oliver, Rohan Pethiyagoda, Francis behavior. We started to plan strategies, beginning with
Petter, Pierre Pfeffer, Roberto Portela Miguez, M. K. a research trip around European museums (Berlin,
Ranjitsinh, Tom Roberts, Jan Robovsky, Kees Rook- Bonn, Frankfurt, Leiden, Tervuren) in 1973, after which
maaker, Terri Roth, Klaus Rudloff, Maria Rutzmoser, we separated to take up posts overseas, I in Canberra,
Jaffar Shikari, Chris Smeenk, Kes Smith, Atanas Tcho- he in Accra. From these distant centers we would write
banov, Dirk Thys van den Audenaerde, Louise Tom- to each other, gradually work up papers together, and
sett, Michel Tranier, Marc Vassart, Wang Sung, Wang send each other our sole-authored manuscripts for
Yingxiang, Eileen Westwig, Henning Wiesner, Detlef comment. “Groves and Grubb” alternated with “Grubb
Willborn, Don Wilson, Roland Wirth, Derek Yalden, and Groves” in the literature, and our colleagues got
Yang Chan Man, Yang Qisen, Zainal-Zahari Zainud- the impression that we were making a takeover bid for
din, Tatjana Zharkikh, and the late Klaus Zimmer- the ungulates, perhaps for all the mammals.
mann. Above all, we would like to thank Phyll and The unfortunate thing was that, while my post was
Eileen, who have been with us and supported us all a continuing one, and I happily established my teach-
the time. ing and research profi le at the Australian National
University, Peter’s, at the University of Ghana, was a
fi xed-term appointment. At its termination in the late
WHAT ARE UNGULATES?
1970s, when he returned to the UK, he found a very
“Ungulate” means “having hoofs.” This is nowadays tight market in the university world. Most zoology
taken to imply that part of the terminal phalanx is departments were not especially interested in taxon-
encased in a sturdy hoof, although Lydekker (1913) omy at the time; he did apply for an advertised post in
required, inferentially, only that the toenail be thick- animal behavior, and was shortlisted, but he was, in
ened for it to count as a hoof; hence, his inclusion of effect, not very interested and withdrew. And so it was
elephants and hyraxes in his cata log of ungulates. that he became a schoolteacher.
This book takes ungulates to be members of the He was not unhappy teaching biology, but chafed
Artiodactyla and Perissodactyla; moreover, not all of a little under the restrictions of the system. It meant
the Artiodactyla are ungulates, if the Cetacea should that he had comparatively little time now for his re-
actually be included in the order (as they should be, search, but his friends supported him in his applications
-1— to make the Artiodactyla monophyletic). for leave to make overseas research trips—notably, a
0—
349-47558_ch00_1P.indd viii 5/31/11 6:06 PM

p ref ace ix
very productive one to museums in North America— (Grubb, 1978a), rainforest biogeography (Grubb, 1979,
and his output, plus our joint output, continued, only 1982a, 1990b, 1999b, 2001b), and—throwing light on a
much more slowly than before. As his reputation grew, neglected field—the East African forest biome (Grubb,
he came to be invited to more and more conferences 1985b). There is no doubt that his expertise, perspicac-
and symposia, though I doubt whether his school ever ity, and insights will be greatly missed.
really realized what a gem they had on their hands. When, after his death from cancer, I fi nally began
His first and only trip to China, to take part in the cata- to orga nize a project that we had always had as a fu-
loging of the rediscovered Heude collection (Braun et ture ambition, a book on ungulate taxonomy, I was
al., 2001), was shorter than planned, simply because his invited by his wife Eileen, and his old colleague Barry
employers grumbled at the prospect of him staying Hughes, to go through his papers. To my astonish-
away for more than two weeks. ment, there were manuscripts almost ready to be sent
Different professional colleagues will remember for publication, and acres and acres of notes and mea-
him for different things. In ungulate taxonomy, he surements that he had taken in museums over the
will long be remembered as the person who replaced years but never found time to work up into publish-
assumptions by research and wrote the fi rst sensible able form. Using his data, and incorporating his man-
things for half a century or more on such groups as uscripts (in different stages of completion), I decided
African buff aloes (Grubb, 1972), bushbuck (Grubb, to make this book the culmination of the “Groves and
1985a), or wild pigs (Grubb, 1993b); I myself got straight- Grubb” series. My European research trip in 2009 was
ened out about musk-deer in a revealing paper (Grubb, made thoroughly manageable by the knowledge that
1982b). Wild pig aficionados think of his astonishing a lot of the basic work had already been done by Peter,
demonstration that warthogs are much, much more and I had mainly to fi ll in gaps.
interesting than they had envisaged! Meanwhile, pri- I will miss my old friend. I will miss our commu-
matologists began increasingly to call on his exper- nications by letter and by e-mail, and, even more, I will
tise, such as at an important symposium on primate miss our too-infrequent meetings, usually when I vis-
taxonomy in Orlando, Florida, in 2000. ited London. We would plan to spend some days to-
He was always interested in the bigger picture: gether at the Natural History Museum, but our time
relationships between forest and savanna-living rela- would begin with sitting down together in an Indian
tives in Africa, and interrelationships among African restaurant, shaking hands or slapping palms, and at
forest faunas. His papers on African forest biogeogra- once beginning to talk as if we had never been apart.
phy are by now standard texts: patterns of speciation Colin Groves
—-1
—0
349-47558_ch00_1P.indd ix 5/31/11 6:06 PM

-1—
0—
349-47558_ch00_1P.indd x 5/31/11 6:06 PM

Ung ulate Ta xonomy
—-1
—0
349-47558_ch00_1P.indd xi 5/31/11 6:06 PM

-1—
0—
349-47558_ch00_1P.indd xii 5/31/11 6:06 PM

tendencies” (Simpson, 1961). This is true, in the sense
that it says why species are important in the grand
scheme of things (as the units of biodiversity), but it is
not useful when it comes to actually identifying them
and discriminating them from each other. We are in
need of an operational definition. One of us has argued
extensively (Groves, 2001a, b, 2004) that what is known
as the Phylogenetic Species Concept (PSC) fi lls this
need.
In this book, therefore, we defi ne a species (in sex-
ually reproducing organisms) as follows: “A species is
Theory of Ungulate the smallest population or aggregation of populations
which has fi xed heritable differences from other such
Taxonomy populations or aggregations” (modified following El-
dredge and Cracraft, 1980; Nixon & Wheeler, 1990; and
Christofferson, 1995; also see Groves [2001a, b, and es-
pecially 2004] for discussions of the implications of
this concept).
Note that, under the PSC, species have to be distin-
guished by fixed character states, but—in most versions
of the concept—these character states do not have to
WHAT IS TAXONOMY?
be autapomorphic. It is perfectly feasible that, when a
It is still true that taxonomy is regarded to some ex- parent species splits into two, evolutionary changes
tent as the poor relation of ecology and behavior, but may occur, and even accumulate, in just one of the
it is again becoming more fashionable nowadays, riding, daughter species, the other persisting essentially un-
at least partly, on the coattails of the molecular revo- changed from the parent species; this, indeed, is the
lution. The general public, too, are becoming more basis for progressive clines (see Grubb, 2000a ).
and more aware of biodiversity and get excited about We must admit that, when looking at the mor-
the discovery or identification of new species; taxon- phological or other phenotypic features (behavioral,
omy, of course, is the basis for understanding biodiver- ecological) that differentiate species, it is not always
sity, as was recently emphasized by Cotterill & Foissner possible to demonstrate that the differences are heri-
(2009) in the context of the importance of museum table, but it is important to identify the units of bio-
collections. diversity that are species, so we argue that it is advis-
At its best, a taxonomic arrangement (above the able to start by acting as if they are, until it is shown
species level) is a two-dimensional readout of phylog- otherwise.
eny: taxa of the same rank are sister groups, and as the The advantage of the PSC is that it depends entirely
rank/time depth nexus becomes accepted, the level at on the evidence to hand; there is no extrapolation.
which these sister groups are ranked tells us approxi- The Biological Species Concept (BSC), which was
mately when they separated in evolutionary time. This the dominant concept of (sexually reproducing) spe-
is not always possible: there are only so many taxo- cies for some half century following on the great evo-
nomic ranks to go around (not a problem if we want lutionary synthesis of the 1930s, defi ned species as
to junk ranks altogether and adopt the phylocode, of being “reproductively isolated,” meaning that they did
course—which we do not), splitting may not always not interbreed with one another in a state of nature.
be dichotomous, and we sometimes have evidence Many biologists still use the BSC as their guiding prin-
of past reticulations. Nonetheless, taken cautiously, a ciple, even if only lip service is paid to its primacy. There
taxonomic scheme can perform something of this are, however, two glaring problems with the BSC, and
function. either, by itself, would be sufficient to raise doubts
about its universal applicability.
The fi rst problem concerns allopatric populations,
SPECIES
that is to say, populations that are completely out of
In essence, a species is “a lineage . . . evolving separately contact with one another, because they are separated —-1
from others and with its own evolutionary role and by barriers of some kind (water barriers, tracts of —0
349-47558_ch01_1P.indd 1 5/31/11 7:05 PM

2 Ungulate Taxonomy
unsuitable habitat, mountain ranges). In such cases, the that, in some character state or other, two operational
BSC obliges the taxonomist to hypothesize whether taxonomic units (OTUs) are discrete (i.e., nonoverlap-
interbreeding “might” occur if their ranges were to ping ), then the two OTUs are to be classified as dis-
meet. This results in an impossible situation: how tinct species. Further data may show that there is, in
would one know? The distributions of the black-faced fact, overlap, so that the decision to class them as dis-
impala and the common impala do not overlap; they tinct species was wrong, but the important point is
have no possibility to interbreed in nature, so how that the decision always depends on data, never on
should we classify them? Other criteria began to creep extrapolation or hypothesis. The only point at which
in: are they “different enough” to be ranked as dis- we are forced to extrapolate is in the inference that
tinct species, or not? In short, the BSC offers no guid- the OTUs in question are different populations. Spe-
ance when it comes to allopatric populations, and we cies must be populations in any species concept—
are reduced to untestable hypotheses. Don’t get us there is no getting away from this inference, ever—
wrong; the book by Mayr (1963), premised on the BSC, and unfortunately this is just something that we have
is a remarkable piece of work, and even today one can to live with; once we have made this extrapolation,
learn endlessly from it, but, over time, the species con- we can get back to the dataset without making any
cept it promotes has been shown to have irreparable more hypotheses. If this results in more—perhaps
flaws. many more— species being recognized than previ-
The second problem is, quite simply, that it is wrong. ously, surely it is a small price to pay for exactitude?
What we have called the molecular revolution has This is as close as we can come to putting a fi nger
shown that species actually do interbreed in nature; on the units of biodiversity. The next level— where
yet they nonetheless persist as discrete entities. White- we start deducing how these different units are re-
tail and mule deer are widely sympatric in North lated to each other, what ecological roles they perform,
America and demonstrably form separate entities in and their relationship to geography—is where the ex-
nature, yet they often have each other’s mitochon- citement begins for many workers, and we can only
drial DNA (mtDNA)! Evidently they were intermit- agree that this is a matter of intense interest (see, for
tently interbreeding, undetected by biologists prior example, the geospecies concept in Grubb, 2006). Like-
to the 1980s. We will say more about this below; but, wise, we have to ensure the conservation of these units,
for the moment, the important point is that to say that and, if there are more of them than we formerly thought,
different species are “reproductively isolated” can of- then more care, and probably more finances, will have
ten be shown to be incorrect. to be involved. But fi rst we have to determine what
Paterson (1978, 1980) proposed that a species has the units actually are.
its own Mate Recognition System, and that this is the The characteristics of species under the PSC are
overriding criterion for species rank: what he termed therefore as follows:
the Recognition Species Concept. This puts us back
—They are the terminal points on a cladogram;
into the realm of hypothesis, though in a rather dif-
that is, they are the least inclusive phylogenetic
ferent way from the BSC. In most living species, be-
units.
havioral studies are simply not sufficiently advanced
—They are discrete units; that is, it is below the
to be able to determine, with any degree of certainty,
species level that relationships are reticulate
what these recognition features actually are; in the
(and yet some species are undoubtedly the
case of fossil species, one can never know. It is easier
products of hybridization!— see below).
to have a “good guess” in the case of many ungulates
—They are 100% diagnosable; that is, they have
than, for example, in primates or rodents, but that is
fi xed heritable differences between them,
all it is. There is an intense interest in detecting what
though these may, in fact, be expressed only
sort of interactions seem to matter in a mating con-
in one sex or at one stage in the life cycle.
text, and how these patterns differ in various species,
—They are genetically isolated, though not
but one cannot use these observations as the basis for
necessarily reproductively isolated.
deciding whether they are actually separate species
or not, even if one were certain that one did have com- The criteria by which species may be recognized
plete information. may be morphological, or perhaps physiological or
The Phylogenetic Species Concept does not depend behavioral, or they may be base-pair diff erences in a
-1— on hypothesis, nor on extrapolation; it depends on the DNA sequence. We cannot privilege one source of
0— evidence to hand. If the evidence before us indicates information over another, as long as there is at least
349-47558_ch01_1P.indd 2 5/31/11 7:05 PM

t h eo ry o f ung ul at e t axo no my 3
a presumption that they are heritable (to be tested, if structuring in mtDNA (Lorenzen et al., 2008b). Evi-
feasible). dently, a fairly recent advantageous mutation occurred
in the mitochondrial genome (remembering that mi-
tochondrial genes code for important functions in
MOLECULES AND MORPHOLOGY
metabolism) and moved rapidly through the species.
We have found, and recorded in this book, that in many The third reason, of course, is introgression. Hybrid-
cases the evidences from DNA (the “molecular evi- ization between two species is frequently asymmet-
dence”) and morphology, including morphometrics, rical. This idea was fi rst put forth more than half a
are consilient: they give the same results. In other cases, century ago by Flerov, who described hybrids between
however, they are inconsistent. In former days, when sika and wapiti in N China and the Russian Far East:
DNA sequencing was its infancy and was very like “These hybrids are encountered comparatively fre-
magic to most nonpractitioners, the demonstration quently in the wild state and have long been known
that two hitherto established taxa were not much dif- to the Chinese. The male wapiti during rutting drives
ferent in, say, the control region, was held to require away the weaker spotted deer male and covers his
that they should be synonymized: DNA was “the an- females” (Flerov, 1952:142 [126 of the English transla-
swer.” We now know much better; we have remem- tion]). If this superiority of wapiti stags also applied
bered that phenotypes result from genotypes (plus, over hybrid stags in F1 and subsequent generations—
potentially, from an input from environmental effects), as it well might, since the size of the hinds would in-
so these are also, in their own way, telling us about crease in consecutive backcross generations, and they
genetics. would become accessible only to wapiti stags—then the
There are several reasons why DNA data, especially proportion of sika nDNA would halve in each suc-
mtDNA, might not send precisely the same message cessive generation, until we would we end up with
as morphology. First, we might have the persistence of populations that were effectively wapiti but with sika
ancestral polymorphism in DNA, particularly mtDNA, mtDNA. This effect, known as nuclear swamping,
so that two given populations might not have achieved seems widespread among ruminants, as we will dis-
reciprocal monophyly. R. G. Harrison (1998) has cal- cuss in several places in this book.
culated that, where N is the size of the two populations,
and if N has remained approximately constant over
SUBSPECIES
time, then the probability of reciprocal monophyly in
mtDNA sequences is about 35% after N generations, Subspecies are, in principle, geographic divisions of
57% after 2N generations, and 83% after 4N genera- a species. There are strong differences in the frequen-
tions. Two daughter populations of 10,000 individuals cies of their par ticu lar character states, but subspe-
each— not an inflated figure for many mammalian cies are not absolutely different: they overlap. There-
populations—will therefore need 20,000 generations fore they are, to an extent, arbitrary and must in no
to have an even chance of reciprocal monophyly; if case be reified.
an average artiodactyl generation can be envisaged as In the past, there were too many subspecies: large
about five years, this translates to 100,000 years. On numbers of subspecies are listed for most of the spe-
the other hand, reciprocal monophyly for a nuclear cies in such key checklists as G. M. Allen (1939), Chasen
DNA (nDNA) sequence will be, on average, four times (1940), Ellerman & Morrison Scott (1951), Hall & Kel-
that for an mtDNA sequence and, as morphological son (1959), and Cabrera (1961). These subspecies may
characters result from nDNA (not from mtDNA), cases be differentiated into the Good, the Bad, and the Ugly.
where a population will be homogeneous morpho- The Good subspecies are 100% diagnosable; hence
logically, but similar to its neighbors in mtDNA, will they are actually distinct species masquerading as mere
be uncommon for this reason, but not impossible. subspecies—victims of the general feeling around
This, however, applies only in the neutral evolu- the mid-20th century (and certainly under the influ-
tion case. If selection is involved, all bets are off. Thus ence of the BSC) that taxa should be, if possible, rele-
the second reason why DNA sequences and morphol- gated to the status of subspecies of the nearest species
ogy may be offset is if there has been a selective sweep as long as “they do not occur together” (a form of word-
in one or the other. An example of this might be the ing that constantly recurs throughout the checklist
case of the plains zebra, Equus quagga, in which—despite of Ellerman & Morrison Scott, 1951). As noted above,
the clear geographical differences in morphology (strip- the BSC offers no guidance for the allocation of allo- —-1
ing, size, incisor cupping, mane)— there is no such patric taxa, so that the classification of these taxa is —0
349-47558_ch01_1P.indd 3 5/31/11 7:05 PM

4 Ungulate Taxonomy
forever unfalsifiable— it remains purely a matter of It is surely abhorrent to maintain such an impor-
opinion. tant taxonomic ranking, one whose name is the fi rst
The Bad subspecies are points along a cline, or are word in a binomial, in this unfalsifiable limbo. Modi-
differentiated at very low frequency levels, or simply fying previously expressed ideas going back to Willi
are based on one or two individuals that seemed out- Hennig half a century ago, Groves (2001a, 2004) argued
standing at the time. The samples that they repre- that the genus category must be given some objectiv-
sent may be interesting for population genetics or ity by linking it to time depth and suggested, by us-
in some other respect, but they have no taxonomic ing the “principle of least violence,” that this would
standing. be best put at about the Miocene–Pliocene boundary.
The Ugly subspecies are the ones which are left Such time-depth information is not always—and even
over. Subsequent studies have shown that they can be not usually— available, either from the fossil record
differentiated from other subspecies (i.e., from other or from the molecular clock; nonetheless, it is some-
geographic segments of the same species) at high fre- thing to aim for, and future studies will doubtless
quencies, but they are not absolutely differentiated or modify some generic recognitions that we have used
diagnosable. The dilemma is, what to do with them? here.
There does seem to be some advantage in dignify- As far as the tahr are concerned, Ropiquet & Has-
ing them with a trinomial, especially for conservation sanin (2005) gave the following mean separation times
purposes, but as these Ugly subspecies are arbitrary (under a local clock model): Hemitragus from Capra,
and unfalsifiable, one cannot insist upon it. Nonethe- 4.5 million years ago (Ma); Arabitragus from Ammotra-
less, this is what we have in mind when we recognize gus, 6 Ma; Nilgiritragus from Ovis, 4.5 Ma. Under the
subspecies in this book, unless we specifically state Groves regime, the genus Arabitragus would certainly
that they are “provisional,” be valid; the other two—including Hemitragus itself!—
might be questionable. (Later, however, Ropiquet
[2006:figure 18], in her preferred synthetic tree, put all
GENERA
these separations well into the Pliocene, in which case
Genera are monophyletic groupings of species. If a none of the genera could be maintained).
putative genus is shown to be nonmonophyletic, then Groves (2001a, 2004) likewise argued that families,
it must be either be disbanded (by dividing it up into as well as genera, should have a time depth going back
its monophyletic constituents), or expanded (by includ- to the Oligocene–Miocene boundary.
ing in it the other putative genera whose independent In this book, no formal taxonomic conclusion should
existence rendered it nonmonophyletic in the fi rst be drawn from the fact that, when a genus is large and
place), or narrowed (by taking out all but the nomi- relatively unwieldy in terms of the numbers of spe-
notypical species and placing the former in the gen- cies it contains, we arrange the species according to
era to which they are truly related). species-groups. These are not intended to be monophy-
Ropiquet & Hassanin (2005) found that the three letic units, although they may be in many cases. In-
species of the genus Hemitragus (known as tahr) were stead, they are simply grouping devices, the species
more closely related to other caprins than to each other: within them being united on a phenetic basis.
H. jayakari to the genus Ammotragus; H. hylocrius to The concept of geospecies, first used in ornithology,
Ovis; and H. jemlahicus, the nominotypical species, to has been applied to primates by Grubb (2006), who
Capra. They had three choices. First, they could keep defi ned the concept as follows: “Geospecies are lin-
all three species in Hemitragus, into which they would eages passing through an evolutionary history from
also sink Ovis, Capra, and Ammotragus (the prior avail- the stage when they have lost allopatric sister taxa
able name would actually be either Ovis or Capra). Sec- through extinction, to the stage when they have pro-
ond, they could retain Hemitragus, but for jemlahicus liferated by cladogenesis, but not so far that daughter
alone, and place jayakari in Ammotragus and hylocrius taxa have yet become sympatric.” They are, thus, what
in Ovis. Third, they could again retain Hemitragus to have been called superspecies (monophyletic lineages
contain jemlahicus alone, but erect new genera for the of allopatric species), to which have been added spe-
other two species. They chose the third option (put- cies that stand on their own and do not belong to such
ting jayakari in a new genus Arabitragus and hylocrius diverse clusters. The problem with such concepts is
in a new genus Nilgiritragus), but actually any of the always that they are rather fuzzy at the edges. Taxa
-1— three solutions is, under present circumstances, com- may quite easily fail to be sympatric not because they
0— pletely arbitrary. have not diverged far enough, but because, however
349-47558_ch01_1P.indd 4 5/31/11 7:05 PM

phylogenetically distinct they may be, they may be samples, or calculated one measurement as a percent-
adapted to quite different ecosystems; or they may age of another, or plotted two measurements together
simply have retained their original ecological adapta- on a graph, or—whenever we could—included them
tions so as to exclude even distantly related taxa from in a multivariate analysis.
sympatry. We do not here use the superspecies con- The ideal multivariate method is discriminant anal-
cept, preferring instead to use the phylogeny-neutral ysis. This is a method that requires specimens to be
term species-groups, as explained above. assigned to groups, and it then minimizes intragroup
variation while maximizing intergroup variation. The
major constraint is that there should not be too many
PHENOTYPIC PLASTICITY
variables (measurements, in this case) for the number
In this book, one of the methods that we use a great of specimens per group: as a rule of thumb, most of
deal is multivariate morphometrics, especially cranio- the groups should have more specimens than there
metrics. Geist (1989), however, has particularly urged are variables, or we run the risk of a type I error (i.e., a
that morphometrics is not a valid taxonomic tool, be- false positive).
cause of the ubiquity of phenotypic plasticity. In partic- Sometimes discriminant analysis is not possible,
ular, he has cited the work of Franz Vogt who, in the and we use instead the rather unsatisfactory alterna-
1930s, conducted experiments in red deer and roedeer tive of a principal components analysis, in which speci-
breeding, rearing his unselected breeding groups on an mens are entered without grouping them. Whereas
especially high-quality diet (specifically, pressed sesame- in discriminant analysis the different groups, what-
seed cakes). Vogt found noticeable increases in body ever their size, are typically treated equally, so that
and antler size in both species, and these increases con- small groups are given an equal chance to contribute
tinued over four generations (presumably due to mater- to the fi nal dispersion, in principal components anal-
nal effect); the differences were reversed in the red deer ysis the less well-represented samples are “swallowed
when they are were returned to less high-quality diets. up” by the larger samples and fail to contribute equi-
The degree to which the conclusions from these ex- tably to the dispersion. Principal components analysis
periments can be applied across the board, such as to takes the raw measurements, which are all correlated
the differences between “maintenance” and “dispersal” with each other, and turns them into a number of in-
phenotypes, as Geist has propounded (both in his 1989 dependent variates; separating the input specimens
survey and elsewhere), is arguable, however. Deer, and into groupings is only incidental to the process, and
also pigs, seem especially susceptible to the effects of groupings may not emerge at all. That is why, if at all
different planes of nutrition and other factors (such as possible, samples need to be grouped.
temperature) on growth, but these are much less evi- As long as sample sizes are large enough, the ideal
dent in bovids, beyond some early effects, such as milk is to compare different samples of restricted geographic
supply in single versus twin lambs (see the summaries origin with each other, aggregating those which turn
in J. Hammond et al., 1983). In general, we take the “de- out not to be discrete. Eventually, a picture is arrived
fault” situation to be that wild animals, living in envi- at in which two or more of these aggregations may
ronments in which they can flourish and maintain via- remain, and, if they are discrete and there is morpho-
ble populations, grow along a trajectory to achieve an metric space between them, they then answer to the
adult size and shape where both are predictable from requirements of the PSC. If no discrete groups remain,
their genotype (within reasonably narrow limits). The then there is no morphometric evidence for different
argument maintaining that differences between two species.
populations are entirely a result of environmental ef- In multivariate analysis, it is important that all the
fects, or that environmental factors overwhelm genetic included measurements should be of the same kind,
differences, needs to be substantiated. that is to say, either growth dependent or growth
independent, and these two types of measurements
should not be included simultaneously in the same
MORPHOMETRICS
analysis. Growth-dependent measurements are, in almost
Both of us have taken substantial amounts of skull all of our cases, those of bone and/or horns. They in-
measurements (and sometimes collated flesh measure- crease with age, up to maturity at least (and beyond, in
ments) on most of the taxa treated in this book. We cases of indeterminate growth), and they bear an allo-
have worked out means, standard deviations, and ab- metric relationship to each other. Growth-independent —-1
solute ranges for the measurements for geographic mea surements are, in our cases, those of teeth. Teeth —0
349-47558_ch01_1P.indd 5 5/31/11 7:05 PM

6 Ungulate Taxonomy
do not grow; they erupt and are then of a standard size Sikkim, Tibet, or elsewhere, have been labeled “Ne-
(saving changes with wear, which usually means size pal,” and it is only by looking at the date of acquisi-
reduction due to interproximal attrition), and they bear tion and contextual information that these errors can
no allometric relationship with growth- dependent be corrected. Hodgson himself described quite a large
measurements. It may be of interest to make a bivari- number of new species from Nepal, Sikkim, and Ti-
ate plot between a measurement of skull size and a bet; he also contributed his notes and color plates to
measurement of tooth size, but skull measurements the Museum’s archives, and these are sometimes in-
and tooth measurements should on no account be valuable in fi lling in deficiencies of the collection (see,
combined in a multivariate analysis. for example, Grubb, 1982b).
The latter half of the 19th century was an era of
empire building by Western European nations; espe-
A BRIEF HISTORY OF THE TAXONOMY
cially, this was the period of the notorious “scram-
OF UNGULATES
ble for Africa,” which meant a further influx of large
Ungulate taxonomy has advanced, due to—or, in some mammals— slain by explorers, empire builders, and
cases, one might better say suffered under— a variety the big game hunters who followed in their wake—
of sometimes colorful characters. Many of these also, into the British Museum (Natural History), which, by
of course, contributed to other areas in mammalogy, that time, had become a separate institution. These
including primatology, and their accomplishments were described by Oldfield Thomas, Richard Lydekker,
have been briefly recounted by Groves (2001a, 2008b), and others, with a growing contribution by Reginald
but a few of them will be further highlighted here. Pocock who, after his retirement from the London Zoo
Although Linnaeus (1758, 1766) described quite a (he took care to examine the animals that died there,
few ungulates, the detailed description of species of particularly the ruminants, describing their skin glands,
ungulates did not really begin until the work of Pallas noses, and genitalia), worked on a voluntary basis at
(1766, 1767–1777, 1811), who not only described new the Museum.
species on the basis of his travels throughout the Rus- In France itself, mammalian taxonomy had almost
sian Empire, but gave scientific names to many others, dried up, but in Shanghai a French missionary, Pierre-
based particularly on descriptions by Buffon, who had Marie Heude, became passionately interested in nat-
conscientiously refused to use Linnaean nomencla- ural history, especially large mammals, and was, when-
ture. Pallas was followed, in par tic u lar, by Alfonse ever possible, diverting his attention from spreading
Desmarest (1822), who described new species from the the Word in favor of hunting, collecting live animals,
growing collections of the Paris Museum (they had and getting his brother missionaries and other col-
grown not only as a result of overseas collections by leagues to send him specimens—not only from China
French explorers, but also by the activities of Étienne but from neighboring territories, which he loosely
Geoff roy Saint-Hilaire and others who, as Napoleon and inaccurately referred to as “the Chinese Empire”:
conquered one country after another, had visited the the Philippines, Indochina, Korea, the Russian Far
museums and cabinets of the conquered countries East, and even Japan. He had not the least idea of spe-
and confiscated interesting specimens); and by Charles cies, and described hundreds of spurious ones— often
Hamilton Smith (1827), who likewise described many half a dozen or more from exactly the same locality—
new species largely on the basis of the growing British in the midst of his few genuine ones. For a long time
Museum collections. Between them, these three au- after his death (in 1905), his enormous collection re-
thors described a surprisingly large proportion of the mained underappreciated, and finally was lost to view,
ungulate species that we recognize even today. until it was rediscovered by CPG in 1996, and fi nally
The British Museum collections continued to grow cata loged by us and our colleagues in 2000.
and were described by John Edward Gray—its assidu- In the introduction to our cata log, we noted Heu-
ous, opinionated, and cranky mammal curator—who de’s unchecked excesses of splitting and wrote: “Be-
wrote catalog after catalog from the 1840s to 1870s (he fore we join in the near universal condemnation of this
died in 1873). In the 1830s and 1840s, the Museum’s prolixity, it should be remembered that Heude stood
collections were enormously enriched by specimens at the threshold of a new era in mammalian taxon-
sent by Brian Hodgson, fi rst from Nepal and after- omy, when for the very fi rst time large samples were
wards from a new base in Darjeeling. Unfortunately, becoming available. Used to samples of one or two
-1— his transfer from Nepal remained unremarked, so that specimens per region, from which one or two species
0— many of these specimens, which were actually from would be described, the sudden embarrassment of
349-47558_ch01_1P.indd 6 5/31/11 7:05 PM

riches proved overwhelming for many taxonomists. of conversations with Matschie (we cannot trace that
Heude was not the only one to suffer bemusement” Matschie had previously published the idea), hybrids
(Braun et al., 2001:610), and we went on to instance C. between different species were to be identified by be-
Hart Merriam, the founder of American mammalogy, ing asymmetrical. Again, in the light of modern work
parts of whose record were at least as immoderate. on hybrid developmental instability (Ackermann et al.,
Despite these contributions, the period from about 2006), this is not so unreasonable, but in Zukowsky’s
1890 to 1924 was really the era of Paul Matschie. This view it meant that the specific characters on one side
extraordinary individual, perhaps the last notable were those of one of the parent species, and those on
holdout against the theory of evolution among profes- the other side, of the other parent species. Where
sional zoologists, registered hundreds, even thousands should one expect hybrids? In watersheds, of course;
of specimens, including (and especially) ungulates, and, in his notorious 1910 paper, Zukowsky identified
that were sent to the Berlin Museum by explorers and some hybrids that he claimed came from watersheds.
hunters in the newly acquired German territories Famously, he figured the horns of a buffalo which had
(which, in Africa, were Togo, Cameroon, Tanganyika been shot in the district of Bihe, which he said was in
[“German East Africa”], and Southwest Africa). He the watershed of the N-flowing Cuanza and the SW-
also paid regular visits to the Berlin Zoo and examined flowing Cunene and SE-flowing Cubango (= Oka-
animals there. The result was an enormous profusion vango) rivers. Bihe, now called Kuito (although the
of new species and, when he came to adopt the con- province of which it is the capital is still called Bie [sic]),
cept, subspecies—and an overarching theory, outlined is at 12° 22' S, 16° 55' E, in a highland area that is in-
in his Bemerkungen uber die Verbreitung der Tiere in den deed a watershed, although somewhat to the E of the
Deutschen Schutzgebieten (Matschie, 1910a). For him, sources of all the above-mentioned rivers. The horns
animal distributions were centered on river valleys; a in question are strikingly asymmetrical, the right
rather sensible idea, one might say, but Matschie took horn being directed mainly forward and upward, left
it to extremes. Every major river valley had to have horn much more outward and backward. Zukowsky
its own species in every species-group, and so did identified it as a hybrid between the Cunene and Cu-
many rather minor ones; Matschie strove mightily to bango buffaloes, and mysteriously seemed to know
identify these species and the differences between which horn represented which buffalo, because he
them. The boundaries of their ranges were, of course, made the right horn the type of Bubalus caffer cune-
the watersheds between the river systems. nensis (subsp. nov.) and the left horn the type of Buba-
Every so often there would be an exhibition of game lus caffer cubangensis (subsp. nov.) In the same paper,
trophies in Berlin, and on most occasions Matschie he did it again, describing Bubalus caffer sankurrensis
would attend it and publish a report on it. His reports and Bubalus caffer lomamiensis on another asymmetri-
would be peppered with descriptions of the speci- cal specimen, and Bubalus rufuensis on the left horn of
mens in the exhibition, illustrated by photographs of a specimen of which he identified the right horn as
mounted heads and horns in rows. Often enough, he belonging to the already-described Bubalus schillingsi
would declare many of the trophies to represent new Matschie. Matschie and Zukowsky later coauthored
taxa (species and subspecies) and duly name them, a series of three papers on Lichtenstein’s hartebeest,
even though the specimens concerned were mostly which they divided into a large number of different
destined to disappear into the private collections of species, again naming some of the new species on the
those who had obtained them; in a very few cases, the evidence of one horn or the other of asymmetrical
collectors would generously donate their specimens specimens (Matschie & Zukowsky, 1916, 1917, 1922).
to the Berlin Museum, where at least they would be Matschie got more and more mystical with age, sub-
available in perpetuity, but mostly the only remain- stituting his watershed hypothesis with one called the
ing evidence of these type specimens would be the “elementary areas of distribution,” in which he divided
small photos in Matschie’s reports in the Deutsche the landscape into rectangles whose sides were the di-
Jäger-Zeitung. agonals between every 2° of latitude and longitude
Matschie also had a young admirer, Ludwig Zu- (Matschie, 1920). In one of his last papers (Matschie,
kowsky, who was, at that time, an assistant in the Ber- 1922), he described the distributions of what he re-
lin Zoo, but often came to study specimens in the garded as three different species of kiang, thus: E. kiang
Museum. In what may have been his fi rst published 35/80, E. holdereri 37/100, and E. tafeli 35/99.
paper (Zukowsky, 1910), he took Matschie’s watershed Matschie’s colleagues continued to respect him for —-1
idea to extremes. For example, apparently on the basis his vast knowledge, but had long since ceased to take —0
349-47558_ch01_1P.indd 7 5/31/11 7:05 PM

8 Ungulate Taxonomy
him seriously as a taxonomist. And so it was that, in scientific names, regardless of whether they might, in
part as a reaction, mammalian taxonomy entered a fact, be of mixed (hybrid) origin.
world in which lumping held sway. The Phylogenetic The fi rst paragraph of the ruling states that under
Species Concept, proposed in 1983 (but foreshadowed the plenary power, “it is hereby ruled that the name
in the 1970s), took a long while to enter mammalogy, for each of the wild species listed in (2) and (3) below
although in retrospect its logic should have been more is not invalidated by virtue of being pre-dated by a
obvious; indeed, to a degree it is a return to the tacit name based on a domestic form” (ICZN, Opinion 2027,
species concepts of Oldfield Thomas and Gerritt Miller. 2003:81). The names in question are listed in a further
Although one of us (CPG) was feeling his way toward paragraph. Reference to this ICZN Opinion is made
this concept in the late 1990s, and fi nally published a throughout this book in the appropriate places. The
book employing it (Groves, 2001a), it was only later Opinion ends with the following sentence: “The names
that Peter Grubb (PG), after discussions with CPG listed in the ruling above, which are the fi rst available
and after reading the work of Cotterill (2003, 2005), names in use based on wild populations, apply to wild
also accepted the value of it. Nonetheless, PG decided species and include those for their domestic deriva-
to be conservative in his approach to species listing; tives if these are not distinguishable.”
in his contributions to Wilson & Reeder (Grubb, 2005a, PG always regretted that, under pressure of time,
b), he took the position that he ought to follow a policy he at fi rst did not appreciate the full meaning of the
of adopting the classification from the latest revision Opinion, and continued to use the names given to do-
in each case. mestic animals for their wild ancestors in his contri-
butions to Wilson & Reeder (Grubb 2005a, b). It is most
unfortunate that this two-volume book was commis-
DOMESTICATION
sioned by the Smithsonian Institution Press, who then
Many species of ungulates have been domesticated, suddenly ceased publishing, and major changes were
and the evidence tends to suggest that in many cases not permitted in the interim as new publishers for the
what we view as modern domestic “species” are a mix- photo-ready manuscript were sought. Hence the work,
ture of several different domestication events, or at as fi nally published (by the Johns Hopkins University
least a mixture to which several different wild popu- Press), was perforce not quite up-to-date—including
lations have contributed. Some of what has been pro- in its treatment of domesticated species and their wild
posed on this score may, however, be spurious, if the forebears.
conclusions of Ho et al. (2005) are justified, since these The process of domestication is complex, but in each
authors argued that rates of molecular evolution are case it has apparently followed along somewhat parallel
initially extremely high, and then slow down after 1 lines. For the biology of domestication, and brain-size
million years or so. Molecular-clock calculations of changes in par tic u lar, see especially Hemmer (1983)
the separation time between two domestic breeds of- and Kruska (2005, 2007).
ten go back well into the Pleistocene, implying that In this book, we have decided not to deal with do-
the ancestors of the breeds in question separated well mestic animals in any depth. For completeness’ sake,
before actual domestication, so that they must be de- however, here is a list of the scientific names applica-
scended from different wild ancestors. If the Ho et al. ble to domestic species, together with their probable
model holds, however, then such early dates will need wild ancestors:
to be drastically reduced, and multiple domestication
Equus caballus Linnaeus, 1758, horse—from Equus
models may not always be necessary.
ferus Boddaert, 1785
The nomenclature of domestic “species” has been
Equus asinus Linnaeus, 1758, donkey (ass, burro)—
problematic. After long literature discussions, going
from Equus africanus von Heuglin & Fitzinger,
back at least to Bohlken (1958), the International Com-
1866
mission on Zoological Nomenclature (ICZN) fi nally
Camelus bactrianus Linnaeus, 1758, Bactrian
made a ruling (ICZN, Opinion 2027, 2003). A. Gentry
camel—from Camelus ferus Przewalski, 1878
et al. (2004) explained what this means for the nomen-
Camelus dromedarius Linnaeus, 1758, Arabian
clature of domestic animals: the name given to wild
camel (dromedary)—wild ancestor unknown
populations of species take precedence over those given
Lama glama (Linnaeus, 1758), llama—from Lama
to domesticates. Nonetheless, it is often convenient to
guanicoe (Müller, 1776)
-1— continue to refer to domestic animals by their own
0—
349-47558_ch01_1P.indd 8 5/31/11 7:05 PM

Lama pacos (Linnaeus, 1758), alpaca—from Lama distinct species, is critically endangered in
guanicoe (Müller, 1776) × L. vicugna (Molina, Somalia, but a few hundred are protected in
1782) Eritrea.
Sus domesticus Erxleben, 1777, pig—from Sus Rhinoceros unicornis— slowly increasing in number
scrofa Linnaeus, 1758, and other species under protection.
Bos taurus Linnaeus, 1758, European (humpless) Rhinoceros sondaicus—the second-rarest of all large
cattle—from Bos primigenius Bojanus, 1827 mammals, only ±60 remain, but protected.
Bos indicus Linnaeus, 1758, humped cattle Dicerorhinus sumatrensis— only 200–300 remain,
(zebu)—from Bos namadicus Falconer, 1859 not well protected.
Bos frontalis Lambert, 1804, mithan—from Bos Diceros bicornis—in the 1970s and 1980s, suffered
gaurus Hamilton Smith, 1827 the most astonishing crash in numbers known
Bos javanicus domesticus Gans, 1916, Bali cattle— for any large mammal, from nearly 100,000 to
from Bos javanicus d’Alton, 1823 only 2500 individuals; now very slowly
Bos grunniens Linnaeus, 1766, yak—from Bos increasing under protection.
mutus (Przewalski, 1883) Ceratotherium cottoni— on the verge of extinction,
Bubalus bubalis (Linnaeus, 1758), water buffalo— only 10 known to survive, all in captivity or
from Bubalus arnee (Kerr, 1792) semicaptivity.
Capra hircus Linnaeus, 1758, goat—from Capra Choeropsis heslopi—not reported since 1943.
aegagrus Erxleben, 1777 Dama mesopotamica—very low numbers in the
Ovis aries Linnaeus, 1758, sheep—from Ovis gmelini wild; bred in captivity.
Blyth, 1841 Rucervus schomburgki—presumed extinct since
the 1930s.
Panolia (all species)—persist only in very small
CONSERVATION
populations, except for those on Hainan
We unfortunately live in a time when the preservation Island; P. eldii from Manipur, whose numbers
of the world’s biodiversity is in crisis. While we may were once as low as 14, has, however, in-
only infer that substantial numbers of invertebrates creased to more than 200 under protection.
are going extinct, unremarked, every year as their habi- Cervus pseudaxis—number unknown in the wild,
tats disappear, ungulates are large and visible, and but the amount is certainly very small; bred in
the continuing drastic declines in their numbers and captivity.
ranges are all too obvious. We have said little about Cervus alfredi—number unknown in the wild, but
this under the different species’ headings, but, as one the amount must be extremely small; bred in
of the major aims of this book is to document taxo- captivity.
nomic diversity, we must draw attention to it here. Bos sauveli—probably extinct, but searches for it
Species treated in this book that are in imminent dan- continue.
ger of extinction are listed below. We do not list sub- Bubalus arnee— survives only in small isolates.
species, though some, like Hippotragus niger variani, are Bubalus mindorensis—perhaps 200–300 survive.
also highly endangered in their own right. We have Pseudoryx nghetinhensis— still illegally hunted,
selected only the most critically endangered: despite its small numbers; restricted range and
nominal protection.
Equus przewalskii—persists only in captivity,
Gazella acaciae— only about 20 remain; now
although most of the captive stock consists of
protected.
what is called the B-line, known to be partly
Gazella arabica—not seen since its original
descended from at least two domestic horses;
description, 180 years ago.
the A-line, almost certainly descended from
Gazella bilkis—may be extinct, or nearly so,
pure-blooded Przewalski horses, numbers
because of overhunting.
somewhat over a hundred individuals.
Eudorcas rufina—not recorded since the 1880s, and
Equus africanus—the nominotypical subspecies
presumed extinct.
is possibly extinct, though it may persist as a
Procapra przewalskii— only about 200 remain, in
captive stock (possibly from the hybrid zone?)
small isolated populations around Qinghai
in private hands; subsp. somaliensis, perhaps a
Lake and on the Buha River. —-1
—0
349-47558_ch01_1P.indd 9 5/31/11 7:05 PM

10 Ungulate Taxonomy
Alcelaphus tora—uncertain whether any still exist. D. korrigum (with the dubiously distinct
Beatragus hunteri—500–1000 remaining; declining D. tiang E of the Nile) and the five species
in its original habitat; introduced into Tsavo dotted around the E African landscape.
East National Park, where about 100 persist. —The Serengeti-Mara ecosystem—It contains
Damaliscus selousi—unknown if any still exist. a number of apparently unique species. It is
Oryx dammah— extinct in the wild, though therefore important not only as an example of
breeding well in captivity. a nearly intact E African ecosystem in its own
Oryx leucoryx—well represented in captivity, but right, but also as the sole repository of a range
populations reintroduced into the wild have of endemic species.
been under constant threat from poaching. —The lower Oubangui River—Wide as it is, this
Addax nasomaculatus—reduced to a few small, river appears to be insignificant as a faunal
fragmentary populations. barrier; many forest species (in ungulates as in
Capra walie—found in small fragments, mainly primates) go right across it as if it did not exist.
in Semyen National Park, Ethiopia, where it There is some suggestion that its upper course,
has recently increased to about 500 under mostly a continuation of its major constituent,
protection. the Uele River, was at one time (during the Early
Cephalophus jentinki—threatened by hunting and Holocene?) part of the Shari-Logone system,
habitat loss in its very small range. and has since been captured by the Congo to
Cephalophus adersi— survives in only very small form the lower Oubangui.
numbers in both known populations (Zanzibar, —The Cape region—It has its own range of species,
and Sokoke Forest in Kenya). separate from their congenerics farther N.
—Lower Yangtze River—There is a special
ecosystem S of the lower Yangtze, apparently
BIOGEOGRAPHY OF UNGULATES
centered on Lake Po-yang, with its own
The pictures of the biogeography of ungulates that species (as far as ungulates are concerned) of
emerge from this book will take some time to analyze deer and pigs.
and digest, but we would like to draw attention to a —Russian Far East—Finally, we would like to draw
few fi ndings, some of which were unanticipated: attention yet again, as has been done previously
(especially in Russian works), to the separate
—Vrba’s stenotypic versus eurytopic division—
faunal status of the Russian Far East, including
This, which in hindsight now seems so obvious,
at least parts of the Korean peninsula and the
is abundantly confirmed. We return to this
Manchurian provinces of China.
theme in our introduction to the Bovidae.
—The Sudanic grasslands ecosystem—Many
species are spread throughout this ecosystem, OR GA NI ZATION OF THE
from Senegal in the W to the borders of Ethiopia TAXONOMIC SECTIONS
in the E, with little or no geographic variation,
In each case, after introducing the family, we discuss
except sometimes for minor differentiation W
its constituent genera in turn, listing species (and sub-
and E of the Nile. This contrasts with many of
species, where applicable) either before or after we have
the same species-groups in E and S Africa,
detailed our own research and/or that of others. As
where their ranges are restricted to isolates.
noted in the introduction, we have not invariably listed
An example would be the contrast in the
complete synonymies.
Damaliscus korrigum group between the Sudanic
-1—
0—
349-47558_ch01_1P.indd 10 5/31/11 7:05 PM

part i: PERISSODACTYLA
There are no particular problems in the order Perissodactyla, at least

at the higher levels: there are three families in the living fauna, and, as
far as we know, no one has demurred from this assessment.
All perissodactyls are hindgut fermenters; none has a complex stom-
ach, and all have a large caecum and colon.
From being the most diverse ungulates in the Oligocene, they began
to decline in the Miocene, presumably in the face of competition from
the rising artiodactyls, and today perissodactyls are greatly reduced in
both diversity and abundance.
—-1
—0
349-47558_ch01_1P.indd 11 5/31/11 7:05 PM

-1—
0—
349-47558_ch01_1P.indd 12 5/31/11 7:05 PM

A full synonymy is given, because there is still some
1
misunderstanding about the applicability of the names.
The names ferus and gmelini are objective synonyms,
both being based on a description of steppe tarpans by
Gmelin (see Groves, 1994); the names sylvestris and
sylvaticus have different bases, but both refer to wild
horses from the Bialowieza Forest.
The name Equus ferus Boddaert, 1785, was placed on the
Official List of Specific Names in Zoology by ICZN,
Opinion 2027 (2003).
Equidae Suggestions have been made that forest and steppe tar-
pan were distinct, but there is no objective evidence
for this.
The history of the extinction of this species has been
documented by Heptner et al. (1961), Groves (1994), and
elsewhere.
Equus przewalskii Poliakov, 1881

EQUIDAE GRAY, 1821 Przewalski horse, takhi
The taxonomy of the Equidae is summarized by Groves 1881 Equus przewalskii Poliakov. “In the steppes of eastern
& Ryder (2000). There is only a single living genus in Dzungaria” (Harper, 1940).
the family, but this divides well into three subgenera. 1903 Equus hagenbecki Matschie. Ebi Spring, Njursu, and the
Horses (subgenus Equus) Urungu River.
1909 Equus equiferus Hilzheimer. On this name, frequently
Two species of wild horses survived into modern times, but mistakenly ascribed to Pallas, see Harper (1940).
but one (the tarpan) has been extinct since 1920, the 1946 Equus caballus gutsenensis Skorkowski. Gashun Oasis or
other (the takhi or Przewalski horse) is extinct in the Gutchan Mountains, Mongolia.
wild but survives in a flourishing captive stock (very
little of which, however, is purebred). The tarpan is sup- This species, and its former distribution and the his-
posed to be the ancestor of domestic horses, though tory of knowledge of it, are described in Groves (1994).
this needs to be tested genetically. The evidence of intermediate horses E of the Urals,
Domestic horses, even within a single breed, show on the basis of which Groves (1986) originally pro-
high diversity in mtDNA, but much less in Y-chromo- posed to put this species and the last together as Equus
some DNA (Kavar & Dovc, 2008). The unrooted ferus, is equivocal, and the differences between the
neighbor-joining tree in Kavar & Dovc(2008:figure 7) (now extinct) European wild horse and the (still extant)
shows Przewalski sequences closely associated with Mongolian wild horse are clear cut (see summary in
one domestic cluster (C1), and a sequence from the Groves, 1994).
Urals associated with a different cluster (C4); interest- Asses (subgenus Asinus)
ingly, wild horses from Germany and from Siberia were
not (in most cases) closely associated with any domes- The taxonomy of the Asian forms follows Groves &
tic horse cluster. Mazák (1967), but is updated.
Equus ferus Boddaert, 1785 Equus africanus

tarpan The name Equus africanus Heuglin & Fitzinger, 1866, was
1785 Equus ferus Boddaert. Bobrovsk region, near Voronesh. placed on the Official List of Specific Names in Zoology
1826 Equus sylvestris Brincken. Bialowieza Forest. by ICZN, Opinion 2027 (2003).
1912 Equus gmelini Antonius. Bobrovsk region, near On this species, see Groves (1986, 2002) and Groves
Voronesh. & Ryder (2000), and for the nomenclature, see Groves &
1936 Equus przewalskii sylvaticus Vetulani. Bialowieza Smeenk (2007). Its former distribution in the Middle
Forest. East was discussed by Uerpmann (1987). —-1
—0
349-47558_ch01_1P.indd 13 5/31/11 7:05 PM

14 Perissodactyla
domestic donkeys from all over the Old World.

Equus ?africanus africanus The DNA of the Somali and presumed Nubian wild
von Heuglin & Fitzinger, 1866 asses formed two entirely different clusters, with very
Nubian wild ass strong bootstrap support and Bayesian posterior
Long ears, 182–245 mm; shoulder height 115–121 mm; probability. The DNA of domestic donkeys likewise
hoofs high, narrow. Dorsal stripe always present, nearly formed two distinct clusters, one being intermixed
always complete from the mane to the tail-tuft. Leg- with the reputed Nubian wild ass sequences, the other
stripes, where present at all, restricted to a few bands being close to, but separate from, the Somali wild ass
at the fetlocks. Skull length usually less than in other sequences (both of these domestic groups were rep-
subspecies; diastema short; postorbital constriction resented in all geographic regions except for the
well marked; orbit placed high, generally interrupt- far W of Africa, where only the “Nubian” type was
ing the dorsal profi le. found).
Formerly known from the Atbara region of Sudan, On the face of it, this would be interpreted as
E to the Red Sea Hills, and probably into northernmost meaning that Nubian and Somali wild asses are dif-
Eritrea; not known to survive in the wild. ferent species, and that some domestic donkeys are
Groves (1986, 2002) showed that there are slight descended (at least in the female line) from the Nu-
differences between populations. In the Atbara pop- bian wild ass, while others are from a source similar
ulation, the color is more buff y; the shoulder-stripe but not identical to the Danakil population of the
is thick, well marked, short; the diastema is shorter; Somali wild ass, the two lineages being mostly in-
the occipital crest is narrower; there are never traces termixed in local stocks. But the true identity of
of leg-stripes. In the Red Sea population, the color what was reported as Nubian wild asses must be
is grayer; the shoulder cross is nearly always thin questioned. Beja-Pereira et al.’s (2004) samples (fresh
and poorly expressed, and sometimes absent alto- feces) were obtained from “two different wild herds”
gether; the diastema is longer; the occipital crest is in NE Sudan, 21.08° N, 36.20° E and 20.46° N, 36.49°
broader. E. These localities are somewhat to the N of the re-
Wild asses— probably truly wild, not feral— are corded localities of the Nubian wild ass, which is
found in some parts of the Sahara, and apparently interesting and not necessarily a problem in itself,
resemble this subspecies more. They have not been but, in fact, no evidence of the survival of Nubian
described as a separate taxon. wild asses at all has been forthcoming since the
1930s, the date of the collection of museum speci-
Equus ?africanus somaliensis Noack, 1884 mens by Powell- Cotton and by Mason, and of the
Somali wild ass capture of the Hellabrun-Catskill breeding stock, both
Ears shorter, 187–200 mm; shoulder height 120–125 mm; apparently in the Red Sea Hills just on either side of
hoofs wider, lower. Dorsal stripe often absent, and, the Sudan– Eritrea border. Therefore, it cannot be
when present, generally incomplete and broken at guaranteed that Beja-Pereira et al.’s samples really
some point along the dorsum; shoulder cross, when were from Nubian wild asses, as opposed to feral
present, poorly expressed. Leg-stripes present from donkeys.
above the carpus and tarsus to the hoofs. Skull longer;
diastema long; postorbital constriction less marked; Equus kiang Moorcroft, 1841
orbit low-placed. Apparently longer-legged, shorter- kiang
bodied than E. a. africanus. On this species, and its separation from Equus hemionus,
It is not clear whether the two “subspecies” are dis- with which it had previously been considered conspe-
crete and ought to be regarded as species. cific, see Groves & Mazák (1967). It needs to be reiter-
Central (C) and S Eritrea, the Danakil region of ated that the kiang is strongly and consistently distinct
Ethiopia, Djibouti, and N Somalia. There are slight from other Asian equids.
differences between the Somali population and those
farther W. Equus kiang kiang Moorcroft, 1841
Beja-Pereira et al. (2004) sequenced mtDNA (con- western kiang
trol region) from what they deemed Nubian wild asses, Large size; short nasals; short toothrow; short diastema;
from a zoo sample of Somali wild asses (Berlin Tier- color very dark, especially in winter, with dark areas
-1— park, from stock bred in the Hai Bar Reserve, Israel, predominating on the flanks.
0— originally captured in the Danakil region), and from Ladakh and neighboring parts of SW Tibet.
349-47558_ch01_1P.indd 14 5/31/11 7:05 PM

eq uidae 15
& Denzau (1999), that all wild asses from Mongolia and
Equus kiang holdereri Matschie, 1911 surrounding areas are essentially the same.
eastern kiang Based on 31 microsatellite loci (hereafter referred
Somewhat larger; long nasals; long toothrow; long to as microsatellites), this taxon appears, curiously, to
diastema; color lighter, less red, with the light area of be somewhat closer to E. kiang than to E. h. onager
the underside reaching nearly halfway up the flanks. (Krüger et al., 2005).
E Tibet and W plateau of Sichuan.
Equus hemionus kulan (Groves & Mazák, 1967)
Equus kiang polyodon Hodgson, 1847 kulan, Turkmenistani onager
southern kiang Smaller in size; pale sandy yellow in summer, darker
Very small in size; short occiput. and more disruptively colored in winter; below the
Plateau area of Sikkim and the region just to the N. eye, the border between the colored area of the face
On this taxon, see Neumann-Denzau & Denzau and the white of the interramal region cutting diago-
(2003). nally across the jaw angle; occiput strongly elongated.
Turkmenistan; nowadays restricted to Badkhyz and
Equus hemionus Pallas, 1775 a few populations derived from that reserve.
onager, Asian wild ass Whether E. h. kulan and E. h. onager are really sub-
Dorsal outline of the skull very straight; skull low- species of Equus hemionus needs to be reexamined.
crowned; dark color of the upperside extending well
down the flanks, restricting the white of the underside; Equus hemionus onager Boddaert, 1785
dark hoof-rings present; dorsal stripe extends down Ira nian onager
the tail to the tuft. Medium size; rather pale yellow-brown, less disrup-
tively colored than E. h. kulan in summer; below the
Equus hemionus hemionus Pallas, 1775 eye, the border between the colored area of the face
Mongolian wild ass and the white of the interramal region more or less
following the line of the jaw angle; occipital region
1775 Equus hemionus Pallas. Tarei Nor, Transbaikalia, 50° N,
short.
115° E.
Ira nian plateau.
1911 Equus (Asinus) hemionus bedfordi Matschie. Supposedly
Based on 31 microsatellites, E. h. onager is closely re-
from Kobdo, Mongolia.
lated to E. h. kulan, but is distinct (Krüger et al., 2005).
1911 Equus (Asinus) hemionus finschi Matschie. NE of
Zaisan Nor.
Equus khur Lesson, 1827
1911 Equus (Asinus) hemionus luteus Matschie. Surin Gol, Gansu.
Indian wild ass, khur
Large size; sandy coloration, grading into the light fawn Somewhat smaller in size; skull very high-crowned;
of the underside, especially in summer; light border nasals convex anteriorly, making the dorsal profi le
to the dorsal stripe vague or absent, especially in the of the skull sinuous. Color of the upperside much re-
adults. stricted, with the white of the underparts reaching at
Mongolia, extending N into Transbaikalia and W least halfway up the flanks; no hoof-rings; dorsal stripe
into Xinjiang and SW Siberia in the Semipalatinsk ending at or slightly beyond the base of the tail, not
area. reaching the end.
Groves & Mazák (1967) divided this taxon into two, Presently restricted to the Little Rann of Kutch, but
from N and S Mongolia, respectively (the N form ex- formerly W and NW through the Thar Desert into
tending N into Transbaikalia and W into the Semipala- Baluchistan, as far as Kandahar.
tinsk region), differing in the degree to which they ex-
hibit grading (desertlike) coloration. Neumann-Denzau Equus hemippus I. Geoffroy, 1855
& Denzau (1999) criticized this arrangement, arguing Syrian wild ass, achdari
that all Mongolian hemiones are consubspecific. They Very small in size; intergrading desert coloration; white
also discussed the remarkably disruptively patterned of the underparts restricted to the belly; dorsal stripe
taxon described as Equus onager castaneus, known only extending to the tail-tuft; hoof-rings present. Skull
from a painting, leaving it “in limbo” for future inves- high-crowned; dorsal outline concave.
tigation. For this reason, the full synonymy is given Now extinct. Lived in the desert country of Iraq, —-1
here, indicating, in agreement with Neumann-Denzau Jordan, and Palestine until the 1930s. —0
349-47558_ch01_1P.indd 15 5/31/11 7:05 PM

16 Peri ssodactyla
Groves & Bell (2004) could fi nd no significant geo-

Zebras (subgenus Hippotigris)
graphic variation in this species.
On zebras in general, see Groves & Bell (2004); on
Equus quagga, see Grubb (1981). Equus quagga
An influential paper by Bard (1977) proposed that plains zebra
the striping differences between the species (or species- On the general biology of this species (under the name
groups) of zebra can be explained if the pattern is laid Equus burchellii), see Grubb (1981).
down in the third week of development in E. quagga, in Neuhaus & Ruckstuhl (2002) found year-round re-
the fourth week in the E. zebra group, and in the fifth production in this species in Etosha National Park,
week in E. grevyi. We wonder if this idea might be profit- Namibia; they concluded that, in association with the
ably considered anew at a more detailed taxonomic level. harem mating system, this forced the synchroniza-
Leonard et al. (2005) extracted two short mtDNA tion of the time budget between the males and the
sequences from four preserved specimens of “true females, and might explain the relative lack of sexual
quagga.” All quagga specimens were found to be nested dimorphism in body size.
within the plains zebra clade; indeed, even within the Reynolds & Bishop (2002) found that S and E pop-
S African subclade. Six of specimens of E African ulations differ notably in size; E African populations
plains zebras (E. q. boehmi) formed a separate subclade are far smaller than their fossil predecessors, both Late
to the S African specimens, while the seventh indi- Pliocene and Early Pleistocene. There has been a size
vidual was part of the S subclade, suggesting that the reduction in S Africa as well, but a much less dramatic
N was the original stock, and that the S populations one. The conclusion was that during Late Pliocene
were progressively derived from those farther N, los- and Early Pleistocene times, the E African climate was
ing stripes as they moved S. drier and more seasonal; hence the strong size reduc-
Here, we follow Groves & Bell (2004), wherein full tion of the E African zebras represents adaptation to a
synonymy will be found. warmer, wetter environment, whereas the South Af-
rican environment remained much the same.
Equus zebra Linnaeus, 1758 There is very marked geographic variation in
Cape mountain zebra external features, as well as in skull size and shape
Black stripes broader, wider than the white interspaces. and in the presence or absence of infundibula in the
Relatively small in size; males smaller than the females. lower incisors (Groves & Bell, 2004). In contrast to
Muzzle broader; diastema longer than in E. hartmannae. this, Lorenzen et al. (2008b) were able to fi nd no
Mountainous regions of the S Cape. convincing geographic structuring in either the mi-
tochondrial control region or in any of seven micro-
Equus hartmannae Matschie, 1898 satellites, although there was some separation of north-
Hartmann’s zebra ernmost (E. q. borensis) and southernmost (E. q. quagga)
Black stripes much narrower than the white inter- populations. Evidently a good new model sweeps the
spaces. Size larger, with sexual dimorphism. Muzzle board.
relatively narrow; diastema shorter. In morphologi-
cal characters, this species appears, on the available evi- Equus quagga quagga Boddaert, 1785
dence, to be diagnosably distinct from E. zebra. quagga
Coastal hills of S Angola and N Namibia. Striped with dark brown on a buff y white background
Using both microsatellites and the mitochondrial on the head and neck; flanks darker, yellower, the
control region, Moodley & Hartley (2005) could not stripes fading out along the flanks, sometimes reach-
confirm that the lineages of E. hartmannae and E. zebra ing the haunches; legs white. Skull relatively broad;
were reciprocally monophyletic; nonetheless, all con- occiput narrow.
trol region haplotypes were exclusive to one or the Formerly from S Cape and Free State N to the Vaal-
other (3 to E. zebra, 25 to E. hartmannae). Orange system; now extinct.
Equus grevyi Oustalet, 1882 Equus quagga burchellii (Gray, 1824)

Grévy’s zebra bontequagga, Burchell’s zebra, Damara zebra,
Kenya, NE of the Mt. Kenya region, into S Ethiopia and Zululand zebra
-1— S Somalia, though whether it still occurs anywhere Striped with brownish black or black on an off-
0— but in Kenya is unclear. white ground on the head, the neck, and the flanks;
349-47558_ch01_1P.indd 16 5/31/11 7:05 PM

eq uidae 17
stripes reaching the haunches, and appearing vari-

ably down the upper segments of the limbs, some- Equus quagga crawshayi de Winton, 1896
times with traces below the carpus and the tarsus. Crawshay’s zebra
One or two shadow stripes between the main bold, Striped with black on a striking white ground on the
broad stripes on the haunch. Skull small, narrow, head, the neck, the flanks, the haunches, and the whole
especially the occiput; incisors relatively broad; di- of the limbs, extending unbroken to the hoofs. No
astema longer. shadow stripes. Stripes and interspaces strikingly nar-
The size increases from the SE to the NW of the row, with five or more stripes joining the belly-stripes
range, but the differences are slight. (in other subspecies, the stripes nearly always fewer
Formerly found extensively N of the Vaal- Orange than five). Infundibulum on the lower incisor usually
system, from Etosha and the Kaokoveld to Swazi- reduced to a cup, or absent altogether.
land and KwaZulu-Natal; surviving at the two ends Zambia E of the Luangwa River, Malawi, SE Tan-
of the distribution, but extinct in the midportion. zania, and Mozambique, S at least to Gorongoza.
Traditionally, this extinct midportion has been deemed
to be the “extinct true Burchell’s zebra,” and the sur- Equus quagga boehmi Matschie, 1892
viving (indeed, flourishing) Damara zebra has been Grant’s zebra
known as Equus quagga antiquorum— but there seems Striped with black on a white ground on the head, the
to be no rationale for such a division. As there has neck, the flanks, the haunches, and the whole of the
been such a lot of misunderstanding about this, it is limbs down to the hoofs. Shadow stripes poor or absent.
necessary to reiterate yet again that the type locality Stripes and interspaces broad. Infundibula on the lower
of Hippotigris antiquorum Hamilton Smith, 1841, is incisors usually reduced to a cup, or absent altogether.
the Mafeking district, very close to the Kuruman The smallest subspecies.
district from which Asinus burchellii was itself de- Zambia W of the Luangwa, W at least to Kariba,
scribed; it is therefore very unlikely that the two Shaba, Tanzania (except for the SE), SW Uganda, SW
names designate different taxa, and quite categori- Kenya, E Kenya (E of the Rift Valley), and into south-
cally the name antiquorum is not available for any ernmost Ethiopia, perhaps as far as the Juba River,
putative Namibian subspecies (“Damara zebra”) if Somalia.
such should prove separable. References to an “ex- The Juba population appears to be very small and
tinct true Burchell’s Zebra” are inappropriate, as the maneless, with white ears; better evidence would prob-
subspecies survives not only in Namibia but also in ably establish it as a distinct taxon, for which the name
KwaZulu-Natal. isabella Ziccardi, 1959, is applicable.
Equus quagga chapmani Layard, 1865 Equus quagga borensis Lönnberg, 1921
Chapman’s zebra half-maned zebra
Striped with black on an off-white ground on the head, Resembling E. q. boehmi externally, except that the mane
the neck, the flanks, the haunches, and the upper seg- is tufty or absent in the adult males, but usually pres-
ments of the limbs, broken below the carpus and the ent in the females and the subadults; backs of the ears
tarsus, but generally extending to the hoofs. Shadow white. Larger than E. q. boehmi, and more sexually di-
stripes prominent on the haunches and, usually, on morphic; rather short diastema. Infundibula on the
the neck. Skull large; snout shorter, narrower. lower incisors apparently absent.
“Transvaal” N to Zimbabwe, W to the Okavango, NW Kenya, from Lake Baringo to Karamoja, and
the Caprivi Strip, and S Angola. extreme SE Sudan.
—-1
—0
349-47558_ch01_1P.indd 17 5/31/11 7:05 PM

tion of the land bridge between North and South
2
America and, presumably, the arrival of tapirs in South
America.
Using the cytochrome b gene, Ruiz-García et al. (in
press) found that the three American tapirs do, in fact,
form a clade with respect to the Malayan tapir. They
experimented with different calibration points, fi nd-
ing that the best was a split between T. terrestris and T.
pinchaque at 3 Ma, and between the Asian and Ameri-
can clades at 18 Ma, with T. bairdii splitting from the
other American species at some period in between.
Tapiridae Tapirus Brisson, 1762
This was one of the generic names which A. Gentry (1994)
requested to be conserved, even though other names
from Brisson (1762) were to be rejected.
The facial skeleton is low, so that the nasal cavity

is fairly low; the dorsal profi le of the cranium slopes
TAPIRIDAE GRAY, 1821 upward from the nasals to the crown. The occiput is
On the general biology of this family, see Eisenberg narrow, sloping backward. The dorsal spiral grooves
et al. (1987). A detailed anatomical description and anal- on the nasals are narrow and sharply bordered poste-
ysis of the proboscis of the tapir has been published by riorly. The cartilaginous nasal septum is relatively
Witmer et al. (1999). small, and sits in something of a slot between the max-
All tapirs are externally dark, often with white illae on either side. The anterior end of the premaxilla
rims to the ears, sometimes white around the mouth is not greatly curved down, so that the incisors occlude
and around the hoofs; only in one genus, Acrocodia, is in the plane of the occlusal line of the cheekteeth. The
there what may be described as a color pattern on the ectoloph crest is notched. The MtI rudiment does not
body itself. Infants have a pattern of longitudinal whit- articulate with MtIV.
ish stripes, generally partly or almost entirely broken These tapirs are relatively slenderly built, with a
up into rows of elongated spots. slender and not excessively elongated proboscis. The
Although they are rather similar externally, the body is uniformly gray, brown, or black, though there
skulls of the four living species are very different from may well be (and usually are) white or whitish marks
each other; there are a few differences in the dentition; on the head, the throat, and the legs.
and Acrocodia differs from other tapirs in the structure
of the foot (Radinsky, 1963). Tapirus terrestris (Linnaeus, 1758)
A phylogeny based on the mitochondrial COII South American tapir
gene (Ashley et al. 1996) placed Tapirus terrestris and The skull has a very high, narrow sagittal crest, which
T. pinchaque very close to each other and quite distant rises posteriorly and gives the dorsal outline of the skull
from the other two species that, in one of the two a strongly convex contour. Unusually, the sagittal crest
most parsimonious trees, actually formed a clade to- does not develop from the fusion of the superior tem-
gether (though with only 51% bootstrap support); in poral lines, but emerges from the middle of the brain-
the other tree, they formed successive branches. Ap- case very early in ontogeny, the temporal muscles be-
plying a molecular clock, and calibrating it to the date ing already well developed; with growth, the temporal
of separation between the Tapiridae and the Equidae, muscles continue to enlarge, and the sagittal crest height-
they suggested that (what we here call) Acrocodia sep- ens (Holbrook, 2002). The nasal bones are narrow and
arated from the neotropical tapirs 21 to 25 Ma, and reduced. On the upper molars, both the anterior and
Tapirella separated from the South American spe- the posterior cingula are narrow.
cies 19 to 20 Ma, while the separation between T. ter- The body color is brown, varying from fawn to
restris and T. pinchaque was only about 3 Ma, which, grayish or nearly black, with somewhat lighter cheeks,
as they point out, was not too long after the forma- chin, and throat, and there are often white rims to the
-1—
0—
349-47558_ch01_1P.indd 18 5/31/11 7:05 PM

t ap iridae 19
ears. There is a short, but conspicuous, narrow mane on the second molar, causing the tooth margins to be
from the forehead to the withers. The proboscis is rela- convex.
tively short. Body color is nearly black, with striking white lips
Shoulder height 77–110 cm; weight 180–250 kg; great- and (usually) white ear rims. Pelage is long, thick, and
est skull length 375–407 mm, varying geographically. coarse. The proboscis is fairly short.
Geographic variation remains to be analyzed in T. pinchaque is shorter-legged than T. terrestris, but
detail, but the following patterns can be seen (col- more compactly built, so that the weight is much the
lected data of CPG): same. Height 75–80 cm, weight about 225–250 kg; great-
est skull length 370–380 mm.
1. From E Brazil S to Minas Gerais and Paraguay,
Confi ned to high altitudes (2000– 4000 m) in Co-
and W to Mato Grosso and Bolivia—Body
lombia and Ecuador.
color tends to be pale gray-brown to tawny;
greatest skull length averages about 389 mm Tapirella Palmer, 1903
(N = 5) in Paraguay, rising to 398 (N = 7) to the
The nasal bones are short, but continued anteriorly
NW, in Bolivia. If subspecific status for these
by the nasal septum, which is much more strongly
four geographic forms should turn out to be
ossified than in Tapirus; the edges of the maxillae
warranted, this would be the nominotypical
are dorsally prolonged to support this massive ossi-
Tapirus terrestris terrestris (Linnaeus, 1758), the
fication. The plane of the nasals and their continua-
type locality being Pernambuco.
tion is abruptly stepped down from that of the fron-
2. From Venezuela to the Guyanas, Brazil (Ama-
tals. The temporal lines do not fuse, but form a
zonas), E Ecuador, and Peru—Body color
double sagittal crest. The maxillae and the premax-
tends to be black-brown, with contrastingly
illae are slightly down-bent, so that the incisors oc-
light to white cheeks and throat. Relatively
clude slightly below the plane of occlusion of the
small in size; greatest skull length averages
cheekteeth.
382 mm (N = 32). The name Tapirus terrestris
The anterior premolars are more molarized and
tapir (Erxleben, 1777) may be available for
wider than in other tapirs, and the cingula are nar-
this form if it is a distinct subspecies, as the
row, so that the anterior margins of all three molars
distribution given by Erxleben is “from the
are straight. The ectoloph crest is not notched.
Isthmus of Darién to the Amazon.”
3. N Argentina and Rio Grande do Sul—These
Tapirella bairdii (Gill, 1865)
tapirs tend to be very dark, and large in size;
Baird’s tapir
greatest skull length averages 407 mm (N = 8).
The body color is dark brown; there is a short, barely
The available name for this taxon would be
expressed mane. The proboscis is greatly enlarged (sup-
Tapirus terrestris spegazzinii Ameghino, 1909
ported by the septal ossification). The legs are long; the
(type locality: Río Pescado, Dept. Orán, Salta,
tail is also long, up to 13 cm (not above 10 cm in other
Argentina).
species).
4. W Colombia—Dark to medium brown in
Height up to 120 cm; weight up to 300 kg; greatest
color, and small in size; greatest skull length
skull length 409– 433 mm.
averages 381 mm (N = 5). This putative taxon
From southernmost Mexico, Belize, and Guate-
is Tapirus terrestris columbianus Hershkovitz,
mala SE to W Colombia, where it is apparently sym-
1954.
patric with Tapirus terrestris.
Acrocodia Goldman, 1913
Tapirus pinchaque (Roulin, 1829)
mountain tapir, woolly tapir The skull is very high-crowned; the nasal bones emerge
The braincase is low and flat; the sagittal crest high above the floor of the nasal fossa; the dorsal pro-
forms with age, by the fusion of the superior tempo- file, from the nasals to the occiput, is straight; the supe-
ral lines, remains low, and is not arched. The nasal rior temporal lines do not meet, and the dorsal table
bones are very long, and in the plane of the frontals, between them is low and wide; the occiput is wide
so that there is only a shallow rise posteriorly. The and vertical. As in Tapirus, the cartilaginous nasal sep-
anterior cingulum on the upper molars is wider than tum is small and sits in a slot between the two maxil-
in T. terrestris, as is the posterior cingulum, especially lae. The dorsal depressions of the nasals form deep, —-1
—0
349-47558_ch01_1P.indd 19 5/31/11 7:05 PM

20 Perissodactyla
wide channels that are prolonged back into the fron-

tals as a pair of laterally spiraled scrolls. There is a no- Acrocodia indica (Desmarest, 1819)
ticeable frontolacrimal tubercle. The anterior premo- Malay tapir
lars are not well molarized; the upper molars have The general color is black, with the body, behind the
wide anterior and posterior cingula. The anterior ends shoulders and above the hips, white (except in a rare
of the premaxillae curve down, so that the incisors all-black mutant, known from the Palembang region).
occlude at a level well below the occlusal line of the There is no mane.
cheekteeth. Height 90–105 cm, weight 250–320 kg; greatest skull
In the foot skeleton, the presumed rudiment of the length 410– 444 mm.
fi rst metatarsal is swung well laterally to articulate Sumatra (not the extreme N) and the Malay Pen-
with MtIV as well as MtIII and the entocuneiform insula. Recorded in Laos (Deuve, 1972), and may have
(Radinsky, 1963). survived in Borneo until well into the 20th century
Heavily built, with a long, capacious proboscis. (Cranbrook & Piper, 2009).
-1—
0—
349-47558_ch01_1P.indd 20 5/31/11 7:05 PM

small, the laterals large, procumbent, and pointed,
3
forming formidable tusks. Postglenoid and posttym-
panic processes united below the external auditory
meatus. Occipital plane very wide, forming a flat-topped
triangle in posterior view, and sloping forward in lat-
eral view. Dorsal outline of the cranium deeply con-
cave. Foramen magnum more or less triangular. Na-
sal bones anteriorly pointed. (Following Flower, 1876;
Pocock, 1945; Chakraborty, 1972.) Externally, there
are deep folds in the skin, as follows: along the angle
of the jaw (much deeper in the males than in the fe-
Rhinocerotidae males); two to three vertically around the neck, with
one emerging posteriorly and running horizontally
for some distance, about halfway up the thickness of
the neck; vertically anterior to the foreleg; horizon-
tally around the base of each foreleg; vertically poste-
rior to the foreleg, travelling across behind the shoul-
der to meet its opposite number on the other side;
RHINOCEROTIDAE GRAY, 1821 vertically anterior to the hindleg, crossing in front of
the croup to meet its opposite number, and continu-
The most up-to-date survey of the status of living ous, with one going horizontally across the base of
rhinos is by Amin et al. (2006). On rhinoceros taxon- each hindleg; posterior to each hindleg, going verti-
omy and evolution, see Groves (1997c). cally across behind the croup and above the tail; and,
It was Flower (1876) who laid the basis of modern fi nally, one running horizontally from the posterior
rhinoceros taxonomy by reducing the number of rec- hindleg fold, level with lower surface of the tail base,
ognized species to a provisional six (one of them re- and usually not reaching the anterior hindleg fold.
garded as extremely doubtful) and arranging them Body hairs sparse, but locally detectable in certain
into three genera. Flower also listed and described lights and to the touch. One single nasal horn.
the most cogent characters by which the species can A survey of the Asian rhinos, incorporating much of
be recognized, illustrating some of them. Groves (1971) what was known up to that date, is in Groves (1982a).
elucidated the differences between the horns of the
existing taxa. Rhinoceros unicornis Linnaeus, 1758
There have been many attempts to decipher the Indian one-horned rhinoceros, greater one-horned
exact branching order of the three major lineages: Rhi- rhinoceros
noceros, Dicerorhinus, and African rhinos (Diceros and
Ceratotherium being universally accepted to be branches 1758 Rhinoceros unicornis Linnaeus. Rookmaaker (1998)
of the same major lineage). Groves (1983a), for example, showed that the sources for this species actually
on the basis of a detailed survey of skulls and teeth, included R. sondaicus, as well. The lectotype is the rhino
put Dicerorhinus closer to Rhinoceros. The latest attempt in Dürer’s famous woodcut of 1515.
(Willerslev et al., 2009) was unable to resolve the pat- 1779 Rhinoceros rugosus Blumenbach. For this name, see
tern: the separation times, regardless of whether Rhi- Rookmaaker (2004).
noceros, Dicerorhinus, or the African lineage separated 1875 Rhinoceros jamrachi Jamrach. Manipur.
first, were all between 32 and 30.4 Ma (note that a mean 1970 Rhinoceros unicornis bengalensis Kourist. Bengal.
of 60 Ma was specified for the age of the root [i.e., sepa- Only synonyms not listed in Ellerman & Morrison Scott
ration from other perissodactyls], and used as the (1951) are given here.
calibration point—which some paleontologists might In the upper molars and premolars, the crochet and
think outrageously early!). the crista become united with wear, cutting off an
Rhinoceros Linnaeus, 1758 accessory valley from the medisinus. The posterior
end of the vomer is thickened, fused to pterygoid
A single pair of large, compressed, blocklike upper plates on either side; the mesopterygoid fossa is nar-
incisors, with (at least in the young adults) a much row, the bases of the pterygoid plates almost meet —-1
smaller lateral pair; mandibular central incisors very posteriorly. —0
349-47558_ch01_1P.indd 21 5/31/11 7:05 PM

22 Perissodactyla
The skinfolds are deep, and hang loosely; the hori-

zontal fold emerging posteriorly from the neck folds Rhinoceros sondaicus sondaicus Desmarest, 1822
is short, petering out before reaching posterior fore-
1822 Rhinoceros sondaicus Desmarest. Java; exact locality
leg fold. The body skin is studied with low knobs.
indeterminable (Rookmaaker, 1982).
Sexual dimorphism in the development of the man-
1824 Rhinoceros javanicus E. Geoff roy & F. Cuvier. Java.
dibular and neck folds is considerable: adult males
1827 Rhinoceros camperis Griffith. Java.
develop a noticeable “bib.”
1829 Rhinoceros javanus G. Cuvier. Java.
The upper lip has a short median point, which is
1836 Rhinoceros camperii Jardine. Java.
used for browsing, but more usually the point is
1868 Rhinoceros floweri Gray. Sumatra.
“tucked in,” so that the lip becomes a wide cropping
1868 Rhinoceros nasalis Gray. “Borneo,” probably Java.
organ.
1876 Rhinoceros frontalis von Martens. Error for Rhinoceros
Rookmaaker (1980) has traced the recent distribu-
nasalis.
tion of this species, at least as far as India, Bangladesh,
The full synonymy is given here, following Rookmaaker
and the Indochinese region are concerned. It was
(1983).
known in modern times from Bihar E through N
Bengal and Assam, perhaps just entering Bangladesh, Dentition relatively small; palatine bones 44%–53%
as far E as the Tirap Frontier tract (uncertain), and of the total palate length (posterior to the incisive
possibly even into N Laos. foramina); palatine bones >80% as wide as long, and
usually longer; mandibular corpus not greatly deep-
Rhinoceros sondaicus Desmarest, 1822 ening posteriorly, its depth at the third molar 111%–
Javan one-horned rhinoceros, lesser one-horned 144% of that at the second premolar; on P2, protoloph
rhinoceros fusing with the ectoloph with wear; anterior premo-
The upper molar and premolar crochet and crista al- lars (probably actually milk molars) remaining in
most never fuse. place for much of the animal’s adult life; crochet on
The posterior end of vomer is thin, free from pter- the premolars usually doubled; crista on the premo-
ygoids; the mesopterygoid fossa is wide. (Mainly fol- lars usually much reduced on Sumatran and mainland
lowing Flower, 1876; Pocock, 1945.) individuals, but often quite absent on those from Java,
The skinfolds are much “tighter” than in R. uni- and absent on the molars in all animals.
cornis; the horizontal fold emerging posteriorly from Java, Sumatra, and the Malay peninsula as far N as
the neck folds slants upward, crossing the withers to Tenasserim. Survives, as far as is known, only in Ujung
join its opposite number on the other side, forming Kulon National Park, westernmost Java.
a kind of “saddle” about it. The skin has no raised
knobs, but it is covered by a network of very fi ne Rhinoceros sondaicus annamiticus Heude, 1892
cracks; the head is small in proportion to the bulk of
1892 Rhinoceros annamiticus Heude. Tay-ninh, 700 km from
the body; the females are slightly larger than the
Saigon (see Braun et al., 2001).
males, on average, and the males develop only a
Rookmaaker (1983) gave the corrected reference to this name.
small “bib.”
The upper lip is long and pointed, suitable for Nuchal surface more inclined anteriorly; facial skele-
browsing; the premaxilla does not fuse to the maxilla ton relatively low, with a comparatively deep dorsal
until old age, extending the gape and increasing the concavity; palatine bones 39%– 63% of the total pal-
mobility of the upper lip. ate length; palatine bones 36%– 66% as wide as long;
The horn in the male is smaller, and more slender, mandibular body as in nominotypical R. s. sondaicus;
than in R. unicornis, and the supporting nasal bones in P2, protoloph fusing with the ectoloph with wear,
are narrow; in the females, the horn is very reduced. as in nominotypical R. s. sondaicus; anterior premolars
The characterizations of the subspecies follow (probably actually milk molars) remaining in place
Groves (1967a), and especially Groves & Guérin (1980); for much of the animal’s adult life; crochet on the pre-
see also Groves & Chakraborty (1983). Some of the molars usually not doubled; crista usually absent on
differences appear discrete, and nonoverlapping, and the premolars, but usually present, though small, on the
the Indochinese and Bengal taxa are theoretically can- molars. Long bones apparently longer and more slen-
didates for recognition as full species, but caution is re- der than in nominotypical R. s. sondaicus; medium
-1— quired, since only very small sample sizes are available. metapodials shorter and wider.
0—
349-47558_ch01_1P.indd 22 5/31/11 7:05 PM

rh ino cero t idae 23
Rookmaaker (1980) noted that this subspecies is the Melaka Zoo, has been given by Zainuddin et al.
depicted on a relief in Angkor Wat in Cambodia and (1990).
he gives evidence that it may still have existed in S The skinfolds are tighter than in Rhinoceros, and
Laos at the time of his writing. It possibly still occurs only the fold behind the foreleg travels over the shoul-
in Cat Tien National Park, NE of Saigon. der to join the one on the opposite side. There are two
horns, with the frontal horn placed well behind the
Rhinoceros sondaicus inermis Lesson, 1838 nasal horn; the bases are not continuous. The skin of
the nose is heavily cornified, lacking the wrinkles an-
1838 Rhinoceros inermis Lesson. Sunderbans.
terior to the nostril characteristic of other rhinos, with
Dentition relatively large; palatine bones occupying a only a single deep crease running between the ante-
much larger proportion of the hard palate, >80% of the rior margins of the nostrils from side to side, allowing
total palate length posterior to the incisive foramina; mobility for the upper lip.
palatine bones 50%–80% as wide as long; mandibular Body hair is comparatively long and profuse, com-
corpus noticeably deepening posteriorly, its height at pared with other rhinos.
the third molar 167% of that at the second premolar; on A population aggregation analysis of mtDNA by
P2, protoloph remaining separate from the ectoloph, Amato et al. (1995), using six individuals from Suma-
apparently throughout life; anterior premolars (proba- tra, four from Borneo, and seven from West Malaysia,
bly actually milk molars) shedding before maturity; found extremely low diversity: only a single haplotype
crochet on the premolars usually not doubled; crista on Borneo, one on the mainland, and two in Sumatra.
present on the premolars, absent from the molars. Two variable sites differentiated the Borneo samples
In females, the horn appears to have been absent from the samples from Sumatra plus the mainland.
altogether. Subspecies follow Groves (1967a); see also Groves
Now extinct. Rookmaaker (1980) has recorded its & Chakraborty (1983).
former distribution in India. R. s. inermis was known
from Moraghat in the Bhutan Duars, probably the Sik- Dicerorhinus sumatrensis sumatrensis
kim Terai, the Sundarbans (where its more detailed (Fischer, 1814)
distribution was given by Rookmaaker, 1997), and Chit-
1814 Rhinoceros sumatrensis Fischer. Fort Marlborough,
tagong. In Bangladesh, there is a record from Sylhet.
Bintuhan district, Sumatra.
Dicerorhinus Gloger, 1841 1822 Rhinoceros sumatranus Raffles. Sumatra.
1873 Ceratorhinus blythii Gray. Pegu.
The maxillary incisors are as in Rhinoceros; the man-
1873 Ceratorhinus niger Gray. Sunghi-Njong district, Malacca.
dibular central incisors are lacking, but the laterals
1874 Rhinoceros malayanus Newman, nom. nud.
are tusklike. There is no subaural closure. The occipi-
The full synonymy is given here, following Rookmaaker
tal plane is narrow, somewhat rectangular in poste-
(1983).
rior view, and more or less vertical in lateral view.
The dorsal outline of the cranium is not strongly con- Size large; teeth medium to small; occiput low, narrow.
cave, with a slight convexity on the frontals. The fo- Sumatra and the Malay peninsula, N to Pegu. The
ramen magnum is pear-shaped, its dorsal rim re- only certain records referable to it in the Indochinese
duced upward into a narrow prolongation. The nasal district seem to be Nong Het (Laos), and Lao-Dao
bones are anteriorly pointed. (Following Flower, and Nhatrang (Vietnam), but whether they were re-
1876; Pocock, 1945; Chakraborty, 1972.) ferable to this subspecies is unknown.
A survey of the Asian rhinos, incorporating much
of what was known up to that date, is in Groves (1982a). Dicerorhinus sumatrensis harrissoni
(Groves, 1965)
Dicerorhinus sumatrensis Fischer, 1814
1912 Rhinoceros borniensis Hose & McDougall, nom. nud.
Sumatran rhinoceros
1965 Didermocerus sumatrensis harrissoni Groves. Suan-
The name Rhinoceros crossii Gray, 1854, belongs to this Lambah, Sabah.
species, but to which subspecies is unknown.
Size small; teeth small; occiput narrow, but propor-
On this species in general, see Groves (1982a). A survey tionally high, and forwardly inclined.
of the biology of D. sumatrensis, intensively studied at Borneo. —-1
—0
349-47558_ch01_1P.indd 23 5/31/11 7:05 PM

24 Perissodactyla
Dicerorhinus sumatrensis lasiotis (Sclater, 1872) Diceros bicornis bicornis (Linnaeus, 1758)
1872 Rhinoceros lasiotis Sclater. About 16 hours (as the 1758 Rhinoceros bicornis Linnaeus. As shown by Rookmaaker
elephant marches) S of Chittagong (Harper, 1940). (1998), Linnaeus based this name on specimens both of
this species and of Dicerorhinus sumatrensis. A neotype,
Size very large; teeth proportionately very large;
with the locality being the Cape of Good Hope, was
occiput broad and high.
validly designated by Zukowsky (1965).
The northernmost subspecies, now perhaps ex-
1797 Rhinoceros africanus Blumenbach.
tinct. The westernmost locality was the Sankosh
1836 Rhinoceros keitloa A. Smith. Mafeking.
River, and Rookmaaker (1980) mapped other records
1842 Rhinoceros gordoni Lesson. Near the sources of the
in India and Bangladesh.
Gamka River.
Diceros Gray, 1821 1845 Rhinoceros camperi Schinz; not Jardine, 1836. Cape of
Good Hope.
The incisors are rudimentary (possibly only decidu-
1845 Rhinoceros niger Schinz. Tsondap, Nuuibeb Mountains,
ous) or absent, and the front of both jaws is abbrevi-
S Namibia.
ated. There is no subaural closure. The occipital plane
As before, the full synonymy is given here, following
is narrow, extending backward in lateral view, and
Rookmaaker (1983), who investigated the basis of the
overhanging the occipital condyles posteriorly. The
names given.
nasal bones are much thickened, and truncated anteri-
orly. There are two horns, with the frontal horn placed Very large in size, with large teeth. Mandibular fi rst
immediately behind the nasal horn; the bases of the premolar apparently always absent in the adults. Max-
two horns are sometimes continuous. The upper lip is illary [second, third, and fourth] premolars commonly
long, pointed, and mobile. Body hairs are lacking, ex- possessing a crista. Radius under 80% of the humerus
cept on a few specific places (the tail tip, the ear rims, length (in all other subspecies, it is over 80%). Foreleg
and the base of the horns). slightly shorter than the hindleg; limbs apparently
rather slender. Skin apparently smooth, not deeply
Diceros bicornis (Linnaeus, 1758) folded.
black rhinoceros Formerly from the Cape N to Kuruman, and
The basic outlines of our knowledge of geographic vari- apparently to S Namibia, but not the coastal strip E
ation were summarized by Groves (1967b) and Rook- of the Drakensberg. This subspecies has long been
maaker & Groves (1978). Rookmaaker (1995) proposed extinct; this vital information was inadvertently ab-
that this taxonomy, with later published amendments, sent from Groves (1967b), but was emphasized by
should be used pending future work; he registered a Rookmaaker & Groves (1978) in their general re-
strong objection to the practice, at that time becoming view of this subspecies. Unfortunately, the failure
increasingly common, of using subspecific names for by Groves (1967b) to note that it was extinct has re-
what authors specifically stated were “ecotypes.” Later sulted in the name becoming (spuriously) wide-
(Rookmaaker, 2005), he also deplored the tendency to spread among wildlife workers for the still-extant
“redefine the subspecies of the black rhinoceros on an Namibian population.
ad hoc or geographically limited basis,” insisting that
they must be allocated with reference to real data. Diceros bicornis chobiensis Zukowsky, 1965
Rookmaaker (1983) showed that the names Atelo-
1965 Diceros bicornis angolensis Zukowsky. Virui waterhole,
dus bicornis var. plesioceros and A. b. platyceros Brandt,
Mossamedes, Angola.
1878, belong to this species, but to which subspecies
1965 Diceros bicornis chobiensis Zukowsky. Konsumbia,
is not known; he showed however, that the name A. b.
parent streams of the Loma River, tributary of the
porrhoceros, occurring in the same publication, re-
Cuando River, SE Angola.
ferred to an animal from Upper Nubia; hence, it is a
synonym of D. b. brucii. Somewhat smaller than nominotypical D. b. bicornis.
Harley et al. (2005) showed that, in the main, genetic No crista on the upper premolars, or very minute. Skin
differentiation in microsatellites between the subspe- with deep body folds.
cies was reasonably good, and that all individuals ex- Okavango region.
-1— amined could be ascribed to their own subspecies with Apparently only a single individual is known to
0— some confidence, with the exception of D. b. chobiensis. survive. This individual was studied by Harley et al.
349-47558_ch01_1P.indd 24 5/31/11 7:05 PM

rh ino cero t idae 25
(2005), using microsatellites; it was not well differen- Still smaller (one of the smallest subspecies); relatively
tiated from D. b. minor. broad-skulled. External habitus the same as in D. b.
chobiensis, D. b. minor, and D. b. occidentalis. Mandibular
Diceros bicornis minor (Drummond, 1876) first premolar apparently always absent in the adults.
No crista on the upper premolars, or very minute.
1876 Rhinoceros bicornis major Drummond. Country SE of
Foreleg slightly longer than the hindleg.
Zambezi.
NW Tanzania into E Kenya, including Tsavo and
1876 Rhinoceros bicornis minor Drummond. Zululand.
the N Guaso Nyiro.
1893 Rhinoceros bicornis holmwoodi Sclater. Udulia, 50 mi. S of
Individuals of this subspecies studied by Harley et
Speke Gulf.
al. (2005) were differentiated, in microsatellites, from
1947 Diceros bicornis punyana Potter. Hluhluwe.
those of other subspecies he studied.
1965 Diceros bicornis nyasae Zukowsky; a conditional name,
hence unavailable. N end of Lake Malawi.
Diceros bicornis brucii (Lesson, 1842)
1965 Diceros bicornis rowumae Zukowsky; a conditional
name, hence unavailable. Inland from Mikindani, 1842 Rhinoceros brucii Lesson. Tscherkin, between the Bahr
Tanzania. Salaam and Atbara rivers.
1972 Rhinoceros kulumane Player. 1878 Atelodus bicornis var. porrhoceros Brandt.
1897 Rhinoceros bicornis somaliensis Potocki. “Somaliland.”
Still smaller in size. Externally, characterized by a
? 1947 Diceros bicornis palustris Benzon. Near Aweng, N of
short, compact body, with well- marked skinfolds and
the Lol River, Bahr- el- Ghazal.
a large head. Mandibular fi rst premolar apparently
1965 Diceros bicornis atbarensis Zukowsky. Anseba Valley, Eritrea.
always absent in the adults in E Africa, but retained in
60% of those from Hluhluwe. No crista on the upper Size as in D. b. minor; very narrow across the zygo-
premolars, or very minute. Foreleg slightly longer mata. Mandibular first premolar retained in the adults.
than the hindleg. Crochet on the maxillary premolars simple, not bifid
From KwaZulu-Natal N into NW Tanzania and (unlike most subspecies); these premolars commonly
the SW borders of Kenya. possessing a crista. External appearance, as documented
in photographs, clearly different from those of D. b.
Diceros bicornis occidentalis (Zukowsky, 1922) minor, D. b. occidentalis, and D. b. michaeli, with the skin-
folds much less marked.
1922 Opsiceros occidentalis Zukowsky. Kaokoveld- Cunene
Somalia, the Ogaden and N Ethiopia, Eritrea, and
region.
N Sudan E of the Nile; an isolated population, possi-
Size similar to D. b. minor; much broader across the bly attributable to this taxon, occurred in a small dis-
zygomata. External phenotype resembling D. b. chobien- trict of Bahr-el-Ghazal. Probably extinct.
sis and D. b. minor. No crista on the upper premolars, or
very minute. Diceros bicornis ladoensis Groves, 1967
N Namibia and S Angola.
1965 Diceros bicornis ladoensis Zukowsky; a conditional
This subspecies was not recognized by Groves
name, hence unavailable under Zukowsky’s authorship.
(1967b), but the study of a few more specimens has
Shambe, near Lado, southernmost Sudan.
shown that it can be largely distinguished from D. b.
1967 Diceros bicornis ladoensis Groves.
minor.
Harley et al. (2005) studied microsatellites from 53 Larger than D. b. minor; broad-skulled, especially across
individuals of this subspecies (but unfortunately used the occipital crest. Mandibular fi rst premolar appar-
the designation “D. b. bicornis,” for the reasons given ently always absent in the adults. No crista on the up-
above) from Namibia, and from populations reintro- per premolars, or very minute. Foreleg slightly shorter
duced into South Africa. They were clearly assignable than hindleg.
to their own subspecies, rather than to any other. Kenya Rift Valley NW into S Sudan, E of the Nile.
Diceros bicornis michaeli Zukowsky, 1965 Diceros bicornis longipes Zukowsky, 1949
1965 Diceros bicornis michaeli Zukowsky. Between Engaruka 1949 Diceros bicornis longipes Zukowsky. Mogrum, Chad.
and Serengeti.
A very small subspecies, equivalent in size to D. b. mi- —-1
1965 Diceros bicornis rendilis Zukowsky. N Guaso Nyiro.
chaeli; shorter occipital crest; long distal limb segments; —0
349-47558_ch01_1P.indd 25 5/31/11 7:05 PM

26 Perissodactyla
wide, square base to the horns. Mandibular first pre- and microsatellite analysis (Robinson et al., 2005). It
molar retained in the adults. Crochet on the maxillary was intermediate in head shape between the two; its
premolars simple (as in D. b. brucii), not bifid (unlike mouth was more like that of a white rhino, but its ear
most subspecies); these premolars commonly possess- more like that of a black rhino. To judge by the photos,
ing a crista. the shape of the dorsal outline of the body was inter-
Formerly from SW Chad, Central African Repub- mediate between the two parents, and the deep costal
lic (CAR), N Cameroon, and NE Nigeria. Probably grooving was more like that of a black rhino.
extinct. Groves et al. (2010) reviewed the differences be-
Microsatellites of a single individual of this subspe- tween northern and southern white rhinos and came
cies were studied by Harley et al. (2005), who found it to the conclusion that they are strongly distinct spe-
more strongly differentiated from the other subspecies, with about 1 million years of separation. For con-
cies he studied than these latter were from each other. servation purposes, this conclusion is especially sig-
nificant, and it is possible that, had the nature and
Ceratotherium Gray, 1868
consistency of the differences between the two been
The basic skull characters are as in Diceros. The inci- realized earlier (rather than being hidden as “mere sub-
sors are rudimentary (possibly only deciduous) or ab- species”), more stringent efforts would have managed
sent; the front of both jaws is abbreviated. There is no to save the critically endangered northern species.
subaural closure. The occipital plane is narrow, ex-
tending backward in lateral view, and overhanging Ceratotherium simum (Burchell, 1817)
the occipital condyles posteriorly. The nasal bones southern white rhinoceros
are much thickened, and truncated anteriorly.
1817 Rhinoceros simus Burchell.
Compared with Diceros, the cranium is extremely
1827 Rhinoceros burchellii Lesson.z
elongated; the occipital crest is enormously prolonged
1827 Rhinoceros canus Griffith.
posteriorly; the dorsal outline of the cranium is still
1847 Rhinoceros oswelli Elliot.
less concave; the protoloph and the metaloph on the
1866 Rhinoceros kiaboaba Murray.
upper cheekteeth are strongly curved backward, fus-
1878 Atelodus simus var. camptoceros Brandt.
ing with wear; the cheekteeth are higher-crowned,
1878 Atelodus simus var. prostheceros Brandt.
with much crown cement; the mandibular symphysis
The synonymy includes some names elucidated by
is very broad; the ascending ramus is more backwardly
Rookmaaker (1983).
inclined; there is a pre-sacral eminence, formed by
the anticlinal status of the 17th or 18th thoracic verte- This is the only species of rhinoceros that today can be
bra; there are two horns, the frontal horn placed some- considered reasonably “abundant”; it has been widely
what behind the nasal horn; the bases of the two horns reintroduced over much of its former range S of the
rarely touch; the mouth is broad and blunt, with Zambezi, and beyond it (Kenya).
scarcely any median prolongation of the upper lip;
there is a muscular nuchal hump; copious subcutane- Ceratotherium cottoni (Lydekker, 1908)
ous fat causes atrophy of the body folds and the costal northern white rhinoceros
grooves; the prepuce is translucent, and the penis has
1908 Rhinoceros simus cottoni Lydekker.
eccrine as well as apocrine glands. (Following Groves,
1975b.) Body hair is much reduced, but still detectable A very detailed description of this species, a compari-
in C. simum. son with the southern white rhino, and an argument
Tests on 30 samples from Umfolosi showed that, as to why it has to be considered a distinct species have
as far as microsatellites are concerned, there seemed all recently been given by Groves et al. (2010) and thus
to be a rather low degree of genetic variability, pre- will not be repeated here.
sumably due to the severe population decrease that It is unfortunate that a study such as this was made
had occurred by the beginning of the 20th century. only as this species arrived at the brink of extinction;
A hybrid between a black rhinoceros male and a one hopes that the remnants can still be persuaded to
southern white rhinoceros female was born in a large breed, and that it will not be “saved” by hybridizing it
enclosure in South Africa, and verified by cytogenetics with C. simum.
-1—
0—
349-47558_ch01_1P.indd 26 5/31/11 7:05 PM

part ii: ARTIODACTYLA
Here, we continue to refer to the even-toed ungulates as the order

Artiodactyla, while acknowledging that the cetaceans belong in them
(as a sister group to the Hippopotamidae; see especially Gatesy et al.,
1999). In this, we follow Helgen (2003; see also Asher & Helgen, 2010),
who argued that if the Cetacea were the sister group to the Artiodac-
tyla as such, the name Cetartiodactyla would be appropriate; yet, as
they are deeply nested within the Artiodactyla, there should be no
change of ordinal name (on the precedent of, for example, the Carniv-
ora, which retained that name even with the inclusion of Pinnipedia).
A succinct summary of the phylogeny, including major fossil repre-
sentatives, has been given by A. W. Gentry & Hooker (1988). As far as
ruminants are concerned, there seems to have been a very rapid radia-
tion in the Late Oligocene or Early Miocene, making it very difficult
to decipher the exact interrelationships of the living families (Kraus &
Miyamoto, 1991).
Waddell et al. (1999) classified what they called the Cetartiodactyla
(and we the Artiodactyla) into unranked nested categories. We, more
old fashioned, retain suborders and infraorders.
Similarly, Skinner & Chimimba (2005) used the new divisions, but
at the ordinal level. In the cohort Ferungulata, they recognized three
superorders (Ferae for the orders Pholidota and Carnivora, Paraxonia
[recte Mesaxonia] for the Perissodactyla, and Cetartiodactyla), dividing
the latter into three orders: Suiformes, Whippomorpha, and Rumi-
nantia (no Tylopoda, simply because there are no camels in S Africa).
This incorporates the view, which we endorse, that as long as all groups
are monophyletic, there is no need to make a classification that is strictly
dichotomous.
We prefer to retain the Artiodactyla (as we call it) as an order, be-
cause the molecular evidence suggests that its subdivision into the
modern crown-groups dates from the Cenozoic, and we classify the
Artiodactyla into four suborders. Like Skinner & Chimimba (2005),
we use an old name, Ancodonta, for just the hippos (among the living
fauna, at any rate), but at the infraordinal level. We, like them, use one
of Waddell et al.’s (1999) names, Whippomorpha, for Cetacea plus An-
codonta, but as a suborder. Our fi nal difference from Skinner & Chi-
mimba (2005) is that we use Suina for the pigs and peccaries, rather —-1
—0
349-47558_ch01_1P.indd 27 5/31/11 7:05 PM

28 Artiodactyla
than Suiformes, because the latter name was used in Infraorder Ancodonta
older classifications for this group plus the hippos. Family Hippopotamidae
Our arrangement, therefore, is as follows: Suborder Ruminantia
Infraorder Tragulina
Suborder Tylopoda
Family Tragulidae
Family Camelidae
Infraorder Pecora
Suborder Suina
Family Moschidae
Family Tayassuidae
Family Giraffidae
Family Suidae
Family Antilocapridae
Suborder Whippomorpha
Family Cervidae
Infraorder Cetacea
Family Bovidae
-1—
0—
349-47558_ch01_1P.indd 28 5/31/11 7:05 PM

cacsilensis. He described nominotypical L. g. guanicoe,
4
from Patagonia, as brownish red, with the head, the
cheeks, and the nape light gray. B. González et al.
(2006) surveyed the literature and reported that guana-
cos in Buenos Aires Province have “a cinnamon rufous
half-line on the dorsum,” with lighter flanks, and dirty
white on the abdomen; continental populations of Pa-
tagonia are slightly reddish brown, those N of the Ma-
gellan Straits are yellowish, and those of S Tierra del
Fuego are dark reddish and long-coated. Accordingly,
Krumbiegel’s (1944) characterization is only partly
Tylopoda correct. Those from Chile (L. g. huanacus of Krumbie-
gel, but this name cannot be used; see above) are de-
scribed by Krumbiegel as having a black-gray head,
with dark gray cheeks, and a gray to black nape.
Krumbiegel described his new subspecies L. g. voglii
as softer, lighter in color, sandy (compared with the
strong red-brown in the Patagonian guanacos), with
CAMELIDAE GRAY, 1821 a light gray tone on the cheeks and the nape. He men-
tioned no specimens in museums; therefore, it is quite
Lama G. Cuvier, 1800 unclear what the evidence was supposed to be for the
Guanacos subspecies!
B. González et al. (2006) reviewed geographic varia-
tion in guanacos. Lana guanicoe cacsilensis Lönnberg, 1913
Lama guanicoe guanicoe Müller, 1776 Type locality: Cacsile, Nuñoa, 14.29° S, 70.39° W, Dept. of
guanaco Puno, Peru.
1776 Camelus guanicoe Müller. Placed on the Official List of This was described from a single skull of very small
Specific Names in Zoology by ICZN, Opinion 2027 size. Wheeler (1995) described Peruvian guanacos as
(2003). Krumbiegel (1944) proposed to fi x the type being ochery yellow in color.
locality as Patagonia, but see below. This seems, for the moment, to be a valid subspe-
1782 Lama huanacus Molina. Type locality: Quillota, cies (see below).
32.54° S, 71.16° W (O. Thomas, 1917). Homonym of
guanicoe, according to Cabrera (1961), for the following According to our discriminant analysis of all re-
reason. The International Code of Zoological Nomencla- gions, with the males and the females combined,
ture, Article 33.1, states: “A subsequent spelling of a valid taxa would be (1) Tierra del Fuego, and (2) Pata-
name, if different from the original spelling, is either gonia = Chubut = Bolivia = Chile. These two group-
an emendation, . . . or an incorrect subsequent ings occupy separate regions in multivariate space.
spelling, . . . or a mandatory change.” Art. 33.2.3 states: The small sample from NW Argentina is unclassified,
“Any other emendation [i.e., not ’demonstrably inten- as, in the discriminant analysis, it falls well outside
tional’] is an ’unjustified emendation’ . . . and is a junior any other range. It is not clear whether the type of L.
objective synonym of the name in its original spelling.” g. cacsilensis represents a distinct population of small
1880 Palaeolama mesolithica Gervais & Ameghino. “Prehis- size, or whether it is simply an extremely small indi-
toric” remains from Cañada de Rocha, near Luján, vidual of second taxon.
Buenos Aires Province. According to Cabrera (1961), this High nasal length and breadth, condylobasal
takes precedence over Lama guanicoe voglii (below) if the lengths, and diastema length are what distinguish
guanaco of the pampas is distinct. NW Argentina and Chile specimens from the other
1944 Lama guanicoe voglii Krumbiegel. The pampas of samples; high biorbital breadth and diastema length
Argentina. largely distinguish Patagonia and Chubut specimens
from the other samples. The males have long canine
Krumbiegel (1944) recognized four subspecies of gua- teeth, which the females do not, but there is no indi- —-1
naco: L. g. guanicoe, L. g. huanacus, L. g. voglii, and L. g. cation of any metrical differences between the sexes. —0
349-47558_ch01_1P.indd 29 5/31/11 7:05 PM

30 Artiodactyla
Table 1 Skull measurements for the Lama guanicoe group but those of the other three putative (“Krumbie-
gel”) subspecies were intermixed. There was mix-
Gt l Cb l Nas l Nas br Biorb ture along the borders between L. g. cacsilensis and
the others. The authors suggested that the S (non-
Tierra del Fuego
Mean 307.71 288.50 81.57 61.57 149.00
cacsilensis) guanacos underwent a postglacial range
N 7 6 7 7 7 expansion.
Std dev 8.139 6.626 5.682 5.682 5.132 Wheeler (1995, 1998) said that, even though there
Min 298 281 75 54 139 are no feral llama or alpaca living in the C Andes,
Max 318 299 89 71 155 there is (or was) a group of guanaco-llama hybrids liv-
Patagonia, Chubut, Bolivia ing wild in Cordoba Province, Argentina; these re-
Mean 295.80 274.52 71.48 56.52 141.25
N 55 23 23 23 56
semble guanacos, though some have white areas of
Std dev 13.469 11.253 5.367 3.616 5.734 varying distribution, especially on the head and the
Min 271 253 61 50 120 neck.
Max 331 304 81 62 152
C Chile Vicuñas
Mean 281.00 273.60 68.40 59.00 141.11 We do not accept that vicuñas are generically distinct
N 9 5 5 5 9
Std dev 14.900 17.358 5.550 4.416 5.968
from L. guanicoe. The incisors grow from persistent
Min 260 244 60 53 128 pulps, unlike in guanacos (and llamas, but like some—
Max 302 285 74 64 147 yet not all— alpacas), but this need not indicate a deep
cacsilensis separation and, according to the molecular data (see
Mean 261.00 244.00 — — 128.00 below), does not.
N 1 1 — — 1
NW Argentina Lama vicugna (Molina, 1782)
Mean 318.00 297.50 79.00 61.50 145.67
N 3 2 2 2 3 1782 Camellus [sic] vicugna Molina. Andes of Coquimbo and
Std dev 3.606 — — — 1.155 Copiapó, Chile. This name was placed on the Official
Min 315 290 77 60 145
List of Specific Names in Zoology by ICZN, Opinion
Max 322 305 81 63 147
2027 (2003).
1944 Lama vicugna elfridae Krumbiegel. Type locality:
unknown.
Table 1 gives the measurements for the different
geographic samples. Those from NW Argentina are Wheeler (1995) described the Chilean vicuña as lack-
very large (mean skull length 318.0 mm, N = 3), fol- ing long chest hairs, and having a light beige pelage,
lowed by Tierra del Fuego (307.7 mm, N = 7), the S with much white on the body, coming halfway up
Argentina and Bolivia group (295.8 mm, N = 55), then the flanks and all the way to the crest of the ilium, as
C Chile (281 mm, N = 9), with the only known speci- well as to the anterior portion of the hindlegs.
men of L. g. cacsilensis being the smallest (at 261 mm), Krumbiegel’s L. v. elfridae was described on the
though this is equaled in size by the very smallest basis of living animals and was chiefly distinguished
specimen from C Chile, which measures 260 mm. by its much larger size—guanaco-sized, the shoul-
Condylobasal length is not entirely consistent with der height 92–110 cm (N = 5), whereas he stated that
this: the mean for the Chile series is equal to that of small vicuña (“L. v. vicugna”; Krumbiegel evidently
S Argentina / Bolivia, whereas the greatest length is was unaware of L. mensalis) are 73– 83 cm (N = 8).
very much smaller, indicating a lesser development Three animals of known locality were also referred
of the occipital crest. Similarly, the Chile sample has by Krumbiegel to L. v. elfridae— one from Jujuy, and
shorter but broader nasals, and the biorbital breadth two thought to be from Salta; a specimen cited by
is equal. The NW Argentina sample is comparatively Lydekker from Catamarca was also provisionally re-
narrow. None of the ranges of variation are actually ferred to L. v. elfridae. When we use only animals of
exclusive. known locality, these shoulder height figures be-
Two mitochondrial sequences (complete cyto- come 100–110 cm (N = 2), compared with 73–78 cm
chrome b and partial control region) were studied by (N = 4). Wheeler (1995) described the S vicuña as hav-
Marín et al. (2008). Sequences from the range of L. g. ing a shoulder height of 90 cm on average, which puts
-1— cacsilensis were almost (but not entirely) exclusive, it on the edge of the range given by Krumbiegel (1944)
0—
349-47558_ch01_1P.indd 30 5/31/11 7:05 PM

t yl o p o da 31
for his total L. v. elfridae sample, though admittedly Table 2 Skull measurements for the Lama
below that for individuals of known locality. vicugna group
Krumbiegel also described L. v. elfridae as being
darker red-brown than the more sandy-colored “nom- Gt l Cb l Teeth Biorb
inate” form of vicuña, with the legs red-brown, and
Catamarca
the white strongly set off; whereas in the small race,
Mean 244.00 224.00 64.00 121.00
their color is grading. He described the mane in win- N 1 1 1 1
ter as being unnoticeable, while in summer it was Chile
20 cm long, compared with 35 cm in the small “nomi- Mean 236.00 221.00 66.50 116.50
nate” race. N 2 1 2 2
Later, without explanation, Krumbiegel (1952) re- Min 234 — 66 115
Max 238 — 67 118
garded elfridae as a full species, even seeming to im-
N Peru
ply that it might be closer to the guanaco.
Mean 235.75 216.50 54.33 108.33
Krumbiegel, in fact, correctly described the differ- N 4 2 6 6
ences between the two species of vicuña: large size, Std dev 9.878 — 6.593 24.792
short brisket mane, dark color, white legs. But be- Min 221 211 47 58
cause he thought that the name vicugna denoted the Max 242 222 64 123
Peruvian species, he described what is, in effect, the Peru–Bolivia border area
Mean 228.14 211.14 56.14 118.00
larger S species as new. Therefore, it is evident that L.
N 7 7 7 8
v. elfridae is a synonym of L. vicugna. Std dev 5.900 5.146 3.338 5.127
Min 217 202 52 110
Lama mensalis Thomas, 1917 Max 236 219 61 126
1917 Lama vicugna mensalis Thomas. Incapirca, Junin, Peru.
Described by Wheeler (1995) as being slightly smaller

(70 cm high at the withers, compared with 90 cm in dae) is close to the one skull from Chile (which has no
nominotypical L. v. vicugna); teeth much smaller; color exact locality).
more strongly fulvous. Wheeler (1995) may have been The Chile and Catamarca vicuñas are of great ab-
the first person to focus on the prime distinction of Pe- solute size, contrasting with their low nasal length
ruvian vicuña: they have a long mane on the chest and and diastema length.
the lower throat. The color, as she described it, is dark Table 2 gives skull measurements of the geographic
cinnamon, with white on the underside, the inner as- samples. The skull from Catamarca is 244 mm in great-
pects of the legs, the underside of the tail, and the lower est length, larger than any other. The greatest skull
portion of the face; the eye-rings and ear rims are white. length means of the Chile and N Peru specimens are
Elsewhere in Peru, on the pampas, larger vicuña very close (approximately 236 mm), but the differ-
apparently exist; Wheeler (1995) cited a study by Pau- ence in values for condylobasal length is greater (n.b.,
car et al. (which we have not seen) giving average very small sample sizes!), indicating that the Chile
heights of 90.4 cm for the males and 86.5 cm for the sample has less development of the occipital crest.
females from Pampa Galeras, Peru. The toothrow is much longer in vicuñas from Cata-
marca and Chile than in the two Peru samples. The N
The differences between the Chilean and Peruvian Peru sample has a narrow biorbital breadth, on aver-
vicuñas are great, and, on the evidence, consistent (see age (note, however, the very wide range of variation).
our skulls analyses and the DNA data below). We there- Using the mitochondrial control region, Marín
fore separate them at the specific level. et al. (2007) found N and S vicuñas to be enormously
There were too few specimens of most regional different, with just a small amount of marginal mix-
groups for a discriminant analysis, so we used a prin- ing between them. The N population extended from
cipal components analysis: Peru specimens (includ- Catac in C Peru, SE to Ingenio on the Bolivian border
ing the type of L. v. mensalis) fall outside the general (not far from La Paz), and S along the Chile–Bolivia
Peru–Bolivia scatter, and a young adult from Junin border to Salar Surire. The S population extended from
falls at the edge of it. The Catamarca skull (the one Inta and Cineguillas in far NW Argentina, S along the
which Krumbiegel provisionally referred to L. v. elfri- Argentina– Chile border to San Juan. The authors —-1
—0
349-47558_ch01_1P.indd 31 5/31/11 7:05 PM

32 Artiodactyla
hypothesized that the barrier would be what is called the two domestic species of camel, much greater than
the “dry diagonal” between the N summer rainfall between any two taxa of Lama. If the two genera sepa-
and S winter rainfall areas; a few individuals in the S rated 11 Ma, as they interpret it, then speciation in
population are, however, found within the dry diago- Camelus would have begun in the early Pliocene.
nal. The N population, but not the S, shows the signa-
tures of fairly recent expansion from a refugium. Camelus ferus Przewalski, 1883
Camelus Linnaeus, 1758 1883 Camelus bactrianus ferus Przewalski. Border of the
Kum-Tagh, E of Lob Nor, N of the Altyn-Tagh,
Two quite distinct domestic camel species are known:
Xinjiang.
Camelus bactrianus (the two-humped, or Bactrian,
The name C. ferus Przewalski, 1878, was placed on the
camel) and Camelus dromedarius (the one-humped, or
Arabian, camel). The wild ancestry of the former is
Opinion 2027 (2003).
presumably from an extinct species similar, in some
respects, to C. ferus; the wild ancestry of the latter is On this species, the only living truly wild camel, see
unknown. Schaller (1998), who concluded that it is not as closely
Analysis of the mitochondrial cytochrome b gene related to domestic Bactrian camels as has been
(Stanley et al., 1994) showed a 10.3% difference between thought.
-1—
0—
349-47558_ch01_1P.indd 32 5/31/11 7:05 PM

The cytogenetics of the family have been reviewed
5
by Bosma et al. (1991).
A phylogenetic tree, using the mitochondrial cyto-
chrome b gene, placed Babyrousa as sister to Phaco-
choerus plus Sus (Randi et al., 1996). Within Sus, S. barba-
tus was sister to S. scrofa; within the latter, the Maremma
wild pig, other European wild pigs, and Asian wild and
domestic pigs occupied different positions on the tree,
as calculated by different methods.
Babyrousa Perry, 1811
Suidae Stomach larger than that of the domestic pig, with a
large diverticulum ventriculi; mucus-producing car-
diac glands occupying over 70% of the stomach area,
versus 33% in the pig (Leus et al., 1999).
Three animals from Sulawesi had 2n = 31; most
chromosomes look like those of the domestic pig,
but there appear to be some translocations, and five
The Suiformes (as presently restricted). The two pairs are very unlike those of Sus (Bosma et al.
living families. Genera of the Suidae, including the 1991).
following: What had hitherto been classed as three subspe-
cies of a single species by Groves (1980) were later
—The realization of several species of raised to species rank by Meijaard & Groves (2002).
Babyrousa.
—Porcula, the pygmy hog—Much more distinct Babyrousa babyrussa (Linnaeus, 1758)
than Colin Groves had thought; why he could
have got it wrong. 1758 Sus babyrussa Linnaeus. “Borneo”; corrected to Buru by
—Warthogs—The rediscovery of the desert Groves (1980).
warthog; how Griff Ewer confirmed the reality 1811 Babyrousa quadricornis Perry.
of a neglected myth in the fossil record, and 1827 Babirussa alfurus Lesson. Buru.
how Peter Grubb’s perspicacity brought it into 1920 Babirussa babyrussa frosti Thomas. Taliabu, Sula Islands.
the living fauna. Long, thick body hair; tail-tuft well developed. Skull
—Potamochoerus—Peter Grubb discovered that short, cheekteeth small; frontal furrows deep, sharp-
the bushpig and the red river hog are thor- edged. Upper canine of the males short, slender; alveo-
oughly distinct species; the biogeographic lus forwardly rotated, with the lower canine crossing it
implications of this; what is known of their in lateral view; upper canines generally divergent or
geographic variation. parallel or weakly convergent.
Buru; Taliabu and Sulabesi in the Sula Islands.
SUIDAE GRAY, 1821 Babyrousa togeanensis Sody, 1949

The differences in the heads of pigs (Suidae) and pec- 1949 Babirussa babyrussa togeanensis Sody. Malenge, Togian
caries (Tayassuidae), especially relating to their func- Islands.
tion and their dentition, were reviewed by Herring
(1972). In pigs, the upper canine of the males turns Body hair present, but less long and dense than
outward and/or upward; it is useful in visual display in babyrussa, and paler on the underparts; tail-tuft
and in protection of the face during male–male com- well developed. Skull very large, but the cheekteeth
bat; in peccaries, the upper canine is not sexually di- small, especially the third molars; frontal furrows
morphic, points downward in the usual mammalian shallow, with beveled edges. Upper canine of the
fashion, and is a weapon. males short, slender, somewhat rotated forward, al-
On the Suidae, see Groves (1981d, 1997a, 2007); ways converging.
Groves & Grubb (1993a, b); and Grubb (1993a, b) Known only from Malenge. —-1
—0
349-47558_ch01_1P.indd 33 5/31/11 7:05 PM

34 Arti odactyla
Table 3 confi rms that the males of H. meinertz-

Babyrousa celebensis Deninger, 1909 hageni are very much larger than any other species,
as are the females (to some extent). H. ivoriensis is
1909 Babirussa celebensis Deninger. Pulau Lembeh, N
not distinguished by any single mea surement, only
peninsula of Sulawesi.
by a combination (as we found in our discriminant
1964 Babyrousa babyrussa merkusi de Beaufort, nom. nud.
analyses).
Body hair short, sparse (appearing naked); tail-tuft Therefore, our arrangement is that we recognize
small, sparse. Skull large, with large teeth; frontal fur- three species, of which the first—the E species—is dra-
rows shallow, with beveled edges. Upper canine of the matically different from the other two.
males long, thick, vertically implanted, with the lower
canine not crossing it in lateral view; almost always Hylochoerus meinertzhageni Thomas, 1904
converging.
1904 Hylochoerus meinertzhageni Thomas. Kakamega Forest,
N peninsula of Sulawesi.
Kenya.
Babirussa from C and SE Sulawesi are unclassified.
By far the largest of the three species, especially in
Hylochoerus Thomas, 1904
the males. Tusks in the males flaring very widely;
giant forest hog enamel pillars on the cheekteeth more widely sepa-
Pelage with very stout bristles, always black in color; rated, with more cement on the crown (Grubb, 1993b).
piglets weakly striped, if at all. Nasal disk extremely Mountains W of the Albertine [Western] Rift,
broad. Zygomata thickened and pneumatized, espe- Rwanda, northernmost Tanzania, Uganda, the Ima-
cially in the males, supporting huge, funguslike infra- tong Mountains in S Sudan, the Kenya highlands, and,
orbital warts. No canine apophyses. Cheekteeth hyp- apparently, the Ethiopian highlands (as summarized
sodont, with much crown cement; anterior premolars by Grubb, 1993b).
lost with age. Hylochoerus has the lowest diploid chro- The discoverer of this species was the notori-
mosome number among the Suidae, 2n = 32, although ous Richard Meinertzhagen, recently shown to have
it is based on a single specimen (Bosma et al., 1991). lived a life marked throughout by deceit and hum-
Grubb (1993b) commented on previous taxonomic bug, ranging from continual self-promotion to sci-
arrangements of the genus, and, while placing the taxa entific fraud, and even to sometimes murderous
all in one species, initiated the basic arrangement into treachery (Garfield, 2006). His discovery of the giant
three taxa which we are able to reiterate below, on the forest hog, early in his career, was, however, genuine
basis of new multivariate analyses. enough; presumably, when you are in the presence
We find that the three described western taxa are all of something startlingly new, there is no need for
largely distinct in discriminant analysis, although their pretense.
dispersions approach each other closely, especially H.
ituriensis and H. rimator. The described taxa (H. iturien- Hylochoerus rimator Thomas, 1906
sis, H. rimator, and H. ivoriensis) are distinguished by
1906 Hylochoerus ituriensis Matschie. Ituri Forest.
size, from high to low scores, especially emphasizing
1906 Hylochoerus rimator Thomas. Dja River, Cameroon.
facial length (preorbital and palatal lengths) and mas-
1909 Hylochoerus gigliolii Balducci. Upper Congo.
toid breadth. The sample of H. rimator has a low value
for greatest skull length and preorbital length. Smaller than H. meinertzhageni; tusks less flared in the
When we compare the three E taxa (H. ituriensis males; less crown cement on the cheekteeth.
and H. rimator with E Africa), we fi nd no differences Cameroon–Nigeria border through W-C Africa to
between Uganda and Kenya, or, in Kenya, between the Ituri Forest, N into the CAR and S Sudan, W of the
those from E and W of the Rift Valley. But the two Nile. Apparently absent from S of the Congo River.
easternmost samples as a unit (H. meinertzhageni) are
100% distinguishable from those from C Africa (H. Hylochoerus ivoriensis Bouet & Neuville, 1930
ituriensis and H. rimator), H. meinertzhageni being of
1930 Hylochoerus ivoriensis Bouet & Neuville. Bolobo,
great size, as represented by greatest skull length,
Liberia.
preorbital length, and palatal length, and by breadth
across the canine apophyses. Somewhat smaller than H. rimator; rostrum and pal-
-1— A specimen from Mt. Kahuzi identifies with the E ate short; narrow across the mastoids.
0— African sample, not with H. ituriensis. W African forests, from Liberia to Ghana.
349-47558_ch01_1P.indd 34 5/31/11 7:05 PM

Table 3 Skull measurements for Hylochoerus
Gt l Bas l Pal l Mast br Pal br Zyg test Preorb l Canine br
Males
ivoriensis
Mean 377.69 322.64 232.00 142.73 76.39 202.87 247.76 113.47
N 16 14 17 13 18 15 17 19
Std dev 12.695 10.058 7.778 8.308 2.873 17.435 8.182 9.076
Min 355 305 214 134 72 179 227 99
Max 397 335 243 163 82 235 258 127
rimator
Mean 363.22 325.71 235.90 152.33 79.00 203.56 245.10 112.00
N 9 7 10 9 10 9 10 8
Std dev 14.438 11.295 9.905 7.348 3.559 18.304 11.865 10.198
Min 341 317 221 145 74 171 223 95
Max 388 349 252 166 84 231 261 125
ituriensis
Mean 388.32 343.17 248.59 159.61 79.18 213.94 263.96 122.30
N 28 24 28 27 28 27 28 27
Std dev 13.644 11.251 9.871 12.069 4.701 15.577 9.738 8.826
Min 362 317 230 115 71 187 246 102
Max 415 363 269 177 88 243 282 139
meinertzhageni
Mean 427.65 372.60 276.83 174.06 86.16 245.75 291.29 147.24
N 17 15 18 17 16 18 17 17
Std dev 14.904 18.007 15.085 12.671 4.162 20.611 14.347 12.286
Min 407 328 248 155 78 199 262 127
Max 461 407 310 195 93 281 318 170
Mt. Kahuzi
Mean 425.00 382.00 — — 94.00 277.00 — —
N 1 1 — — 1 1 — —
Females
ivoriensis
Mean 352.29 314.25 221.00 133.33 69.71 174.33 235.14 95.71
N 7 4 6 6 7 6 7 7
Std dev 15.119 12.312 9.423 5.241 1.496 4.502 11.466 5.648
Min 333 299 210 127 67 169 220 85
Max 372 329 234 141 72 180 252 102
rimator
Mean 357.00 320.00 230.50 142.20 75.67 174.33 241.67 93.67
N 5 5 6 5 6 6 6 6
Std dev 17.306 12.288 8.216 3.899 2.582 6.743 9.933 6.250
Min 334 303 222 137 72 164 226 85
Max 377 331 245 147 79 183 252 100
ituriensis
Mean 365.25 325.42 235.33 149.79 77.03 181.43 249.38 103.13
N 20 19 20 19 20 20 20 20
Std dev 13.932 10.516 9.069 5.075 4.544 8.580 11.011 5.094
Min 331 305 214 141 72 165 227 91
Max 391 349 252 161 89 193 272 109
meinertzhageni
Mean 393.50 348.93 259.65 155.79 79.22 197.24 270.57 112.09
N 14 15 13 17 18 17 15 17
Std dev 14.378 16.184 14.185 11.754 6.141 13.722 14.294 10.007
Min 375 315 225 137 64 177 235 86
Max 427 377 278 185 88 235 299 134
—-1
—0
349-47558_ch01_1P.indd 35 5/31/11 7:05 PM

36 Artiodactyla
The differences in both cases were very great, sug-

Porcula Hodgson, 1847
gesting separation during the Late Pliocene.
The resurrection of this genus was recommended
on morphological grounds by Ghosh (1988), and has Phacochoerus aethiopicus (Pallas, 1767)
been supported molecularly (Funk et al., 2007). desert warthog
Only three pairs of mammae, rather than the six Skull shorter and broader than in P. africanus. Zygo-
of Sus; tail rudimentary, about 30 cm, with only 10 or matic arch greatly pneumatized, swollen into a hol-
fewer vertebrae (compared with over 20 in Sus). Nasal low knob just in front of the suture with the squa-
shorter and broader than in Sus; premaxilla rectangu- mous temporal. Sphenoidal pits enormously expanded.
lar, rather than long and narrow, with a tapering na- Upper incisors absent, even in the young; lower inci-
sal branch; orbit situated above the second molar, sors rudimentary or absent. Posterior molars not de-
rather than above the third molar; zygomatic arch veloping roots until the entire crown has come into
not inflated; parietal longer, narrower; dorsal mar- wear.
gin of the skull arched, not straight; maxilla laterally Externally, the warts on the cheeks hook-shaped;
notched. tips of the ears bent backward; suborbital areas are
swollen, pouchlike; head appearing more egg-shaped
Porcula salvania Hodgson, 1847 (d’Huart & Grubb, 2005).
pygmy hog
2n = 38; karyotype said to resemble that of the domes- Phacochoerus aethiopicus aethiopicus
tic pig, except for certain banding differences (Bosma (Pallas, 1767)
et al., 1991). In historic times, this subspecies may have been ap-
No mane; ear small, oval, virtually hairless; with- proximately restricted to the Karoo; Grubb (1993b)
ers lower than the rump. mentioned early specimens from the E Cape, between
the Sondags and Boesmans rivers, and the upper Or-
Phacochoerus F. Cuvier, 1817
ange River.
Distinguished by the disproportionately large head,
slender limbs, and extreme unguligrady. Orbits Phacochoerus aethiopicus delamerei
placed well back on the cranium, and high up; brain- Lönnberg, 1909
case extremely shortened. A pair of deep sphenoidal In NE Africa, the distribution of this species has been
pits behind the internal nares. Cheekteeth strongly mapped by d’Huart & Grubb (2001). It is known from
hypsodont, with numerous, close-packed enamel pil- Somalia, NW almost as far as the borders of Djibouti,
lars; premolars tending to be shed during adult life. the Ogaden, the Webi Shebeyli and Juba rivers, and
Maxilla deep; ascending ramus of the mandible elon- NE Kenya W to about 37° E along the N Guaso Ny-
gated; canine sheaths very wide. Upper canines very iro and 39° E along the Ethiopian border, SE to the
long, retaining their points throughout life; females Lamu district. Only in the Berbera district of Soma-
also have long canines. Pelage very sparse, retaining lia is it approximately sympatric, or parapatric, with
only a prominent, long dorsal crest; often a fringe of P. africanus.
white hairs on the cheeks, especially in the young.
Long, conical warts on the cheeks and on the snout. Phacochoerus africanus (Gmelin, 1788)
It was Ewer (1957) who fi rst proposed, on the common warthog
grounds of fossil material, that there were two dis- 2n = 34 (reported as P. aethiopicus); chromosome 1 ap-
tinct species of warthog in the Pleistocene, and that pears to be a fusion between Sus numbers 13 and 16,
these probably corresponded to different Holocene and chromosome 3 appears to be a fusion between Sus
species as well. This was investigated by Grubb (1993b), numbers 15 and 17 (Bosma et al., 1991).
who found not only that the differences (in the pres- Skull longer, less broad. Zygomatic arch robust,
ence/absence of incisors and the timing of the erup- but less pneumatized. Sphenoidal pits distinct, but
tion of the third molars) are consistent and valid, and not greatly expanded. Two upper incisors present,
that there are other important differences in the but sometimes lost in aged individuals; six lower inci-
skull, but also that warthogs still existing in NE Af- sors. Posterior molars developing roots at the time of
rica belong to the otherwise extinct Cape species. eruption.
-1— Randi et al. (2002) obtained mtDNA and five dif- Externally, the facial warts conical; ears leaf-
0— ferent nuclear loci from both species from E Africa. shaped; suborbital region less enormously swol-
349-47558_ch01_1P.indd 36 5/31/11 7:05 PM

s uidae 37
len; head more diabolo-shaped (d’Huart & Grubb, Face mostly gray, with a broad blackish band over
2005). the muzzle. Usually, body brown or reddish brown,
In NE Africa, P. africanus seems to occur in less grading into a darker color on the belly and the
arid country than P. aethiopicus—hence not, in gen- limbs; occasionally entirely blackish or off-white in
eral, E of about 38° E in Kenya or 42° E in Ethiopia— the males. Skull larger, 345–395 mm (females), 367–
but it also occurs in the Berbera district of Somalia, 415 mm (males).
and here it may be sympatric or parapatric with P. ae- Left bank of the lower Congo River, Katanga, Tan-
thiopicus (d’Huart & Grubb, 2001). zania (N at least to Kilosa), S into Namibia and N South
Africa; a population in the S Cape may be isolated by
Potamochoerus Gray, 1852 a hiatus in KwaZulu in Natal and Transkei. P. l. larva-
Most similar to Sus, but longer-bodied and shorter- tus is widespread (presumably introduced) in Mada-
limbed. Canine apophyses greatly enlarged, with an gascar and on Mayotte, and in Madagascar it appears
additional exostosis above them; together, these sup- to have differentiated into slightly different E and W
port a large rostral wart. No infraorbital warts. Post- forms. It should be noted that the type specimen of
orbital part of the skull elongated, low-crowned; ros- P l. larvatus was from Madagascar.
trum relatively short. Cheekteeth brachyodont, with
thick enamel. Size relatively small, with reduced sex- Potamochoerus porcus (Linnaeus, 1758)
ual dimorphism. red river hog
Bright reddish orange; white dorsal line—not a bris-
Potamochoerus larvatus (F. Cuvier, 1822) tly crest—beginning from behind the head; face with
bushpig a black mask, although with a white muzzle, white
Body hairs bristly, very long, relatively sparse, form- eye-rings, and long white cheek whiskers. Ear ex-
ing a nuchodorsal crest. Head contrasting in color tremely elongated, with a long terminal tuft. Pelage
with the body, but not forming a facial mask. Body bristly on the snout and much of the face, but short,
color often strikingly polymorphic within a single soft, and dense on the body and the forehead. Rela-
population. tively small; skull length 269–378 mm (females), 327–
Five individuals (reported in error as P. porcus) 405 mm (males).
from Kenya and Zimbabwe had 2n = 34, with appar- Rainforest and gallery forest from Senegal to W-C
ent banding differences from other genera. Africa and the DRC. There is little or no detectable
Grubb (1993b) proposed the following three geographic variation (Grubb, 1993b).
subspecies.
Sus Linnaeus, 1758
Potamochoerus larvatus hassama (Heuglin, 1863) Mainly plesiomorphic, compared with other genera
Face white, often with blackish markings in the fe- of the family; large canines in the males, the upper
males and the young males. Adult males black or off- ones curling upward, and the lower ones directed
white. Size small; skull length 327–353 mm (females), laterally; tips of the upper canines wear as they erupt
341–377 mm (males). against the lower canines, so the tips of the upper ca-
Ethiopia, S Sudan, E DRC, Rwanda, Burundi, Ke- nines are blunted, while those of the lower canines
nya, Uganda, N Tanzania; inhabits densely vegetated are sharpened. Canines much smaller in the females,
highland areas (Grubb, 1993b). pointing downward in the upper jaw and upward and
forward in the lower jaw. Large canine apophyses in
Potamochoerus larvatus somaliensis the maxilla in the males. Dentition not reduced; only
de Beaux, 1924 the third molars elongated, with multiple conical cusps
Provisionally recognized as distinct by Grubb (1993b). and cuspules. Externally, the body bristly, with un-
Similar to P. l. hassama, but larger. derwool in temperate-zone forms. Facial warts some-
Forests along the Tana, Juba, and Shebeyli times present.
rivers. There is a great deal of polymorphism in the karyo-
type within this genus, even (and especially) among
Potamochoerus larvatus larvatus (F. Cuvier, 1822) taxa hitherto referred to Sus scrofa. The domestic pig,
as represented by the Swedish landrace, has 2n = 38;
1822 Sus larvatus F. Cuvier. Madagascar.
the Y chromosome is the smallest of the set, and is —-1
1831 Sus koiropotamus Desmoulins. South Africa.
metacentric (Hansen-Melander & Melander, 1974). —0
349-47558_ch01_1P.indd 37 5/31/11 7:05 PM

38 Artiodactyla
Based on cytochrome b sequences, Mona et al. Chinese-derived breeds formed an entirely separate
(2007) found two major clades in this genus: one con- lineage.
taining Sus barbatus and S. verrucosus, the other con- 2n = usually 36. Totaling the figures for Austria,
taining S. philippensis and the scrofa group. Unexpect- Germany, the Netherlands, Switzerland, France, and
edly, samples of S. celebensis occurred in both clades. Spain given by Bosma et al. (1991, table 1) gives 112
Domestic pigs undoubtedly come from several with 2n = 36, 20 with 2n = 37, and 3 with 2n = 38.
different wild sources. This was argued long ago Chromosomes involved in Robertsonian transloca-
by Groves (1981d), and is amply supported by molec- tion are always numbers 15 and 17. The same chro-
ular studies (Larson, Dobney, et al., 2007, and other mosome polymorphism in the same translocations
sources). occur in Lithuania, Belarus, and Central Russia, but
no individual numbers were given, except that the
Sus scrofa Group total number of individuals was 15; and nine individ-
“wild boar” uals from the former Yugoslavia had 2n = 38 (summa-
Distinguished by the relatively short facial skele- rized in Bosma et al., 1991). Note that these figures
ton; comparatively shallow preorbital fossa; and the are different from those available to Groves (1981d),
shape of the male’s lower canine, with the inferior who thought that chromosomal differences, together
surface narrower than the posterior surface. No fa- with an average size difference, supported the separa-
cial warts. tion of an E European S. s. attila from a W European
All members of this group were placed in a single S. s. scrofa.
species by Groves (1981d), but the differences between
some of the taxa in the group are sharper and more Western, Central, and Northern Asia
consistent than previously recognized. Diploid chromosome numbers may be 2n = 36, 37, or
Larson, Dobney, et al. (2007), using the control re- 38 in Azerbaijan (N = 8), Kyrgyzstan (N = 37) and the
gion of mtDNA, found that this species-group has a Russian Far East (N = 20). The distribution of these
“basal comb” (which also included samples of S. cele- figures is not given, but the Robertsonian transloca-
bensis), of which one branch contained all of the Eur- tion involved is between chromosomes 16 and 17,
asian samples. The basic division within Eurasia was that is, different from Europe (Bosma et al., 1991).
India versus the rest; then a basal cluster of general
Eurasian series; with a fi nal, distinctive clade of Euro- Japan
pean and Middle Eastern samples. This certainly im- Bosma et al. (1991) recorded only two karyotyped in-
plies a Southeast Asian diversification for the genus, dividuals from Japan (apparently from the main Japa-
especially as the “basal comb” included samples of S. nese islands), both having 2n = 38.
verrucosus and S. barbatus. Some of the same authors Endo et al. (1995) produced the fi rst detailed report
(Larson, Cucchi, et al., 2007) concentrated on the on the Iriomote wild pigs, which are very small in
Southeast Asian and Pacific region, fi nding links to size. On the basis of an admittedly small amount of
both island and mainland Southeast Asia; they also material, there appears to be no sexual size differ-
studied the shape of the third lower molar, distinguish- ence; three males with third molars in wear averaged
ing a Sulawesi/Philippine form, a New Guinea / Flores 263 mm in “profi le length”; and three females, 261 mm
form, and a China / Southeast Asia form. These molar (calculated from their table 4).
differences were later revisited and refi ned by Cucchi In a later study, Endo et al. (2001) compared man-
et al. (2009). dibles from different parts of the main Japa nese is-
lands; there was a decline in size from N to S.
Europe
Wild pigs from Spain (N = 7) and Italy (N = 1) were Indochina
included in an analysis based on mitochondrial cy- Groves et al. (1997) announced the apparent rediscov-
tochrome b and control region sequences by Alves ery of S. bucculentus, based on the skull of a young in-
et al. (2003), centered mainly around a comparison dividual that resembled the type specimen. The au-
between native Eu ropean and Chinese / Chinese- thors noted, however, that the form of lower canine
derived domestic breeds. The Spanish wild pigs nested in the type specimen was not, in fact, precisely as had
within the Eu ropean domestic breeds, while the been indicated by Heude in his original description.
-1— Italian sample was sister to these; the Chinese and Groves & Schaller (2000) gave further details, reiter-
0—
349-47558_ch01_1P.indd 38 5/31/11 7:05 PM

s uidae 39
ating that the type specimen was not as similar to S. We fi nd a single taxon from all of Burma, Indo-
verrucosus as had been formerly supposed. china, Korea, and China, except in the swamps SW of
Robins et al. (2006) found that this same subadult Shanghai (the S-C China sample), where there is a
specimen ascribed to S. bucculentus had mtDNA al- second taxon, and Heilongjiang (plus Fu Song in
most identical to that of pigs from the Solomon Is- Jilin, plus the Russian Far East), where there is a
lands. Likewise, Mona et al. (2007), using cytochrome third. Overwhelmingly, this differentiation is based
b, found that this specimen was nested deep within on size, with the face longer and the bizygomatic
the S. scrofa group. narrower in the larger taxa (table 4). We propose pro-
When all the male Indochinese skulls in the Heude visionally that these represent distinct species, as
collection were studied by CPG in October 2009, it follows:
was evident that they formed a graded series, with
1. Burma/China—Sus moupinensis Milne-
the type of S. bucculentus at one extreme of, in par tic-
Edwards, 1871. Synonyms as in Ellerman &
u lar, depth of the preorbital fossa and enlargement of
Morrison-Scott (1951), plus Sus canescens
the malar in the facial region. Their sizes were consis-
Heude, 1897, Sus collinus Heude, 1892, Sus
tent within a comparatively narrow range. The shape
coreanus Heude, 1897, Sus meles Sowerby,
of the third lower molar was that of Larson, Dobney,
1917— plus Sus bucculentus Heude, 1892!
et al.’s (2007) “China” type.
2. S-C China—Sus chirodontus Heude, 1888.
A discriminant analysis, using eight craniometric
Synonymy as in Ellerman & Morrison-Scott
variables, was done comparing samples from Burma,
(1951).
Indochina, and the Kienté and Ningkuo region of S-C
3. Heilongjiang / Far East—Sus ussuricus Heude,
China. Skulls from S-C China are well outside the
1888. Synonymy as in Ellerman & Morrison-
range of those from Burma and Indochina; these Chi-
Scott (1951).
nese skulls are low in greatest skull length and palate
length, with relatively wide bizygomatic breadth and
Southeast Asia
postorbital constriction breadth.
Four individuals from Java all had 2n = 38 (Bosma et
The conclusion must be that the skulls from Burma
al., 1991). Because of the uncertainty now surround-
and Indochina are the same taxon, and in subsequent
ing Southeast Asian wild pigs, particularly since the
analyses they were combined; but S-C China (Kienté,
studies of Larson, Cucchi, et al. (2007) and Cucchi
Ningkuo) is different.
et al. (2009), we will not venture a new arrangement
When we added in the S China specimens, we
of wild scrofa-like pigs. For the moment, we propose
found that a single specimen from Zhejiang belongs
that they be classified as S. vittatus, recognizing that
in the S-C China sample, whereas two from Fujian
they do, nonetheless, all share a morphology which
went along with Indochina/Burma. Shaanxi and
is diagnosably distinct from other members of the
Anhui are both at the edge of the Burma/Indochina
group.
range, but not beyond it, whereas the S-C China sam-
While we cannot venture a new classification of the
ple is still quite distinct.
species of the S. scrofa group, we do have fresh evi-
We added in our two skulls from Sichuan: both of
dence, beyond that reported by Groves (1981d), for E
these skulls fall into the same Burma/Indochina
and NE Asia, and mainland Southeast Asia. None-
group, with their long palate and fairly high greatest
theless, we will still make some comments on the
length, contrasted with low bizygomatic breadth and
1981 arrangement, since Groves’s (1981d) subspecies
postorbital constriction (as before).
were of varying quality. Some of them were actually
We next compared all the Chinese samples with
diagnosably distinct; hence recognizable as species.
each other, adding in those from Korea and the Rus-
Some of them were very poor; hence fit only to be
sian Far East. The Anhui and Shaanxi samples fall to-
pushed into synonymy. Finally, some of them re-
gether with Korea, but the Heilongjiang and Russian
main indeterminable, and are listed below with a
Far East sample is entirely separate. A skull from
query:
Fu Song (42.18° N, 127.17° E) falls with Heilongjiang/
Russia. Sus scrofa Linnaeus, 1758 (synonyms Sus attila, Sus
Shanxi identifies with the Shaanxi group. lybicus, Sus algira)—Wild pigs from (most of)
Finally, we performed an analysis combining the Europe, the Middle East as far as the Zagros
groups which fell together (as summarized above). Range, and North Africa. —-1
—0
349-47558_ch01_1P.indd 39 5/31/11 7:05 PM

Table 4 Skull measurements for Sus from mainland Southeast Asia, China, Korea, and the Russian Far East
Gt l Cb l Bizyg Occ br Occ ht Pal l Molars M3 br Lm3 l
Males
Burma/Indochina moupinensis
Mean 379.11 330.52 156.27 76.93 102.36 229.38 73.910 39.471 —
N 27 25 30 27 25 29 29 21 —
Std. dev 14.766 13.326 9.009 8.471 8.371 9.770 4.6876 3.4104 —
Min 354 312 136 61 84 216 63.0 32.8 —
Max 418 364 176 97 116 256 83.0 47.3 —
chirodontus
Mean 417.50 365.75 157.80 84.00 109.75 258.90 79.700 44.725 —
N 10 8 10 10 8 10 10 4 —
Std. dev 7.276 7.126 5.007 5.558 7.833 6.757 2.7101 2.0451 —
Min 408 352 152 73 99 249 74.0 41.8 —
Max 432 378 167 90 122 270 82.0 46.4 —
N China / Korea moupinensis
Mean 397.55 350.45 158.73 78.10 105.09 245.55 75.709 40.467 —
N 11 11 11 10 11 11 11 6 —
Std. dev 18.272 19.351 11.791 11.100 6.848 16.979 3.9612 2.6440 —
Min 362 317 142 66 94 222 70.0 38.0 —
Max 430 380 186 96 115 276 84.0 44.6 —
ussuricus
Mean 452.63 399.50 172.75 93.00 120.67 280.00 87.875 48.350 —
N 8 8 8 3 3 8 8 6 —
Std. dev 20.646 19.034 7.517 5.292 3.215 12.649 4.0861 1.9316 —
Min 423 378 163 89 117 263 79.0 46.7 —
Max 482 431 188 99 123 301 91.0 50.8 —
Females
Burma/Indochina moupinensis
Mean 358.00 317.00 142.33 77.33 90.00 216.00 74.500 — 35.000
N 2 2 3 3 3 2 2 — 1
Std. dev — — 4.509 11.150 8.660 — — — —
Min 354 308 138 69 80 212 69.0 — —
Max 362 326 147 90 95 220 80.0 — —
chirodontus
Mean 357.00 314.67 138.00 69.67 91.00 218.00 74.667 — —
N 3 3 3 3 3 3 3 — —
Std. dev 10.536 13.614 3.606 10.599 5.568 6.245 3.5119 — —
Min 347 304 134 60 86 211 71.0 — —
Max 368 330 141 81 97 223 78.0 — —
N China / Korea moupinensis
Mean 384.00 336.67 148.33 72.33 92.00 232.67 75.667 — 34.833
N 3 3 3 3 3 3 3 — 3
Std. dev 24.637 18.903 6.028 4.041 8.888 16.563 2.8868 — 3.8423
Min 361 322 142 70 85 217 74.0 — 30.4
Max 410 358 154 77 102 250 79.0 — 37.2
ussuricus
Mean 405.80 366.67 152.60 — — 251.00 86.750 — 44.120
N 5 3 5 — — 5 4 — 5
Std. dev 5.933 11.015 8.620 — — 6.124 2.8723 — 2.7842
Min 402 356 139 — — 244 85.0 — 40.5
Max 416 378 160 — — 257 91.0 — 47.3
-1—
0—
349-47558_ch01_1P.indd 40 5/31/11 7:05 PM

s uidae 41
? Sus meridionalis Forsyth Major, 1882—The cies (where previous arrangements have recognized
Sardinian/Corsican wild pigs may or may not only one), but with a large number of subspecies;
be identical to those from S Spain, as proposed whether any of the four species listed above with a
by Groves (1981d), but certainly their affi nities query against them might be recognizable, in some
need further study. sense at least, cannot at the moment be determined.
? Sus nigripes Blanford, 1875—We have seen no
further information on this pig (from Kyrgyz- Warty Pigs
stan and Xinjiang) since that recounted (on the Sus celebensis Müller, 1840
basis of the literature, and of a single speci- Sulawesi warty pig
men) 30 years ago by Groves (1981d). 2n = 38; the Y chromosome is submetacentric and
? Sus sibiricus Staffe, 1922—Likewise, the affi nities larger than in the S. scrofa group, in which it is meta-
of this pig, from Mongolia and Transbaikalia, centric (Bosma et al., 1991).
remain totally obscure, and further study is Small in size, with short legs, short ears, and a
needed. We have seen no specimens; the type long, simply tufted tail; three facial wart pairs in the
description suggests close affi nity, perhaps males, of which the preorbital is the most developed,
identity, with S. nigripes (if that is itself at least until old age, when the gonial wart hypertro-
distinct). phies. Color usually black, often with white or yellow
Sus leucomystax Temminck, 1842—The Japa nese hairs intermixed, but some animals predominantly
wild pig is distinct from any from the main- red-brown or yellowish; underside becoming light yel-
land of Asia. low with age, contrasting sharply with the upperside.
Sus taevanus Swinhoe, 1863—The Taiwan wild Often a yellow snout band. Prime males with conspic-
pig, a very small-sized taxon, is not identifi- uous “toupée” on the crown, abrading with age.
able— on the admittedly slim evidence so far In the molecular plus morphological study by Lar-
available—with any other pig. son, Cucchi, et al. (2007), the third lower molar mor-
? Sus riukiuanus Kuroda, 1924—The Ryukyu pig, phology of S. celebensis was largely similar to that from
as noted above, is distinguished by its very the Philippines, but overlapping with the New Guinea
small size from those on the main Japa nese sample; samples from Flores fall in both the S. celeben-
islands, but it needs close comparison with the sis and New Guinea ranges. Further detailed study by
Taiwan pig. Cucchi et al. (2009) showed that the samples from
Sus davidi Groves, 1981—This species, from Sulawesi, the Philippines, and New Guinea are fully
Rajasthan, Pakistan, and Iran E of the Zagros separable from each other.
Mountains, is distinct, on both cranial and Sulawesi and offshore islands; widely introduced
external characters, from those to the E and W elsewhere in Indonesia, particularly the W Sumatran
of its range. islands, Maluku, and Nusatenggara.
Sus cristatus Wagner, 1839—The distinctive An unexpected finding is that molecular sam-
Indian wild pig is diagnosably separate from ples of this species assort both with the barbatus/
other Sus taxa. On the basis of evidence verrucosus clade and with the scrofa/philippensis clade
accumulated since, Groves (1981d) was (Mona et al., 2007), although no haplotypes are ac-
wrong to include the Burmese pig in the tually identical with any from elsewhere. There is
same taxon: it has a much shorter third indeed some suggestion from the morphology that
molar, a less distinct mane, and a much specimens from N and S Sulawesi may be somewhat
deeper preorbital fossa. Instead, its affi nities different, although related to one another, rather
lie with those from China and Indochina, than nonmonophyletic, as the molecular evidence
as noted above. On the other hand, it was suggests. An explanation for this could be that the
probably not justified to distinguish S. affinis species is at least partly of hybrid origin; nuclear
from S India, which differs only in average DNA sequences are needed to determine the full
size. story.
Taking into consideration the three species dis-
Sus verrucosus Müller, 1840
tinguished in the new craniometric analyses (S.
Javan warty pig
moupinensis, S. chirodontus, and S. ussuricus ), plus S.
2n = 38; the Y chromosome is large and submetacen- —-1
vittatus, we are left with a minimum of eight spe-
tric; the banding pattern of chromosome 10 is specific. —0
349-47558_ch01_1P.indd 41 5/31/11 7:05 PM

42 Artiodactyla
Skull distinctive, with a greatly elongated facial ippine subspecies being S. b. ahoenobarbus; S. philippen-
skeleton. Sexual dimorphism extreme, the males sis, with subspecies S. p. philippensis (Luzon group), S. p.
being more than twice the weight of the females; small mindanensis (Mindanao group), and S. p. oliveri (Mind-
size of the canines in the females striking. Long- oro group); and S. cebifrons, with subspecies S. c. cebi-
limbed; ears large; tail long, simply tufted; head large frons and S. c. negrinus. Attention was also drawn to the
and heavy in the living animal. Three pairs of warts possibility of the existence of further taxa, especially
on the face in the males: infraorbital (the largest), one from Jolo, in the Sulu Archipelago.
gonial (marked by a tuft of long hair in the young, We performed a discriminant analysis on skulls of
before the wart itself has developed), and rostral. Body Philippine pigs of the philippensis group. On average,
color agouti-reddish. Skull length in the adult males we could distinguish the three groups. Mindoro (now
408– 429 mm. Shape of the third lower molar similar S. oliveri; see below) has a long palate and wide bizy-
to that of S. barbatus, as one would expect, but both, gomatic breadth, contrasting with a short greatest
unexpectedly, also close to Southeast Asian S. scrofa length and narrow postorbital constriction. Mind-
(Cucchi et al., 2009). anao and Mindoro are both distinguished from Lu-
Confi ned to Java and, formerly, Madura. zon by a long palate and wide zygomatic breadth,
contrasting with their low greatest length and espe-
Sus blouchi Groves, 1981 cially low occipital breadth.
Bawean warty pig In tooth measurements, Mindoro is absolutely dis-
S. blouchi is closely related to S. verrucosus, but abso- tinct from the other two, which overlap slightly. Min-
lutely different: much smaller (skull length in an doro has a wide M2 and a fairly long P3, contrasting
adult male only 354 mm), and the pelage yellowish, especially with a short M1. Mindoro and Mindanao
not reddish. differ from Luzon in a wide M3 and a fairly long P 4
Confi ned to Bawean, where its survival prospects and M2, contrasting with a narrow P4. Leyte clusters
seem, ironically, rather better than those of its endan- with Mindanao, and Catanduanes with Luzon.
gered sister species on Java. Comparing members of the S. philippensis group
with a sample from Negros (S. cebifrons), we found
Sus barbatus Müller, 1838 that, as expected, Negros clusters apart from the other
bearded pig two.
2n = 38; the karyotype is said to be similar to the do- We next compared samples of all three Philippine
mestic pig. groups (S. barbatus, S. philippensis, and S. cebifrons), in-
Facial skeleton even more elongated; warts much corporating S. b. ahoenobarbus (now S. ahoenobarbus; see
smaller, with only two pairs (no gonial wart). Long below), and were able to include two skulls from the
dorsal mane. Females are larger, relative to the males, S islands, Tawtawi and Jolo, both of which clustered
than in S. verrucosus. Exceptionally long-legged; ears with those from Palawan.
relatively small; tail long, with a large terminal tuft When we dropped S. cebifrons, as being irrelevant
divided into distinct anterior and posterior parts. to fi nding the position of the pigs from the S islands,
The most recognizable characteristic of this spe- and brought in S. celebensis, Tawitawi appeared clos-
cies is a bushy cheek-beard. Groves (1981d) distinguished est to Mindanao, while Jolo was far from anything.
two subspecies, one from Borneo and one from Suma- Palawan differs greatly from the others in wide bizy-
tra plus the Malay peninsula, largely on the suppos- gomatic breadth and a (fairly) long greatest length,
edly different forms of the beards, but specimens seen contrasting with an extremely narrow and fairly low
since then by CPG do not sustain this difference. occiput. Jolo differs from all the others in its very
Borneo, Sumatra, Bangka, and the Malay penin- short nasals and narrow postorbital constriction, con-
sula. Those from Bangka and S Sumatra are consider- trasting with a long palate and a high occiput.
ably smaller, at least on average, than those from else- Reducing the variable list enabled us to bring in
where, a difference which needs further investigation, two incomplete specimens from Jolo and make a
as does the possibility of differences between migra- group out of it, instead of merely including a single
tory and nonmigratory populations. specimen as an unknown. One Jolo skull groups with
Mindanao; the others are separate.
Philippine Group Ungrouped specimens fall as follows: Tawitawi is
-1— Groves (1997a) revised the Philippine wild pigs, di- close to Jolo (and not too far from Mindanao), while
0— viding them into three species: S. barbatus, the Phil- Basilan is close to Mindanao.
349-47558_ch01_1P.indd 42 5/31/11 7:05 PM

s uidae 43
Table 5 Skull measurements for Philippine pigs
Gt l Cb l Bizyg Occ br Occ ht Nas l Pal l
Mindoro
Mean 340.00 292.33 151.00 85.00 103.00 168.33 208.33
N 3 3 3 3 3 3 3
Std dev 9.539 12.055 2.646 6.557 5.568 6.110 6.658
Min 334 281 149 78 98 163 204
Max 351 305 154 91 109 175 216
Luzon
Mean 325.20 277.77 139.00 81.07 97.23 164.33 195.93
N 15 13 15 15 13 15 15
Std dev 18.339 9.194 6.990 6.923 10.787 13.916 10.660
Min 287 265 125 72 82 139 172
Max 352 294 151 91 118 188 210
Mindanao
Mean 347.78 289.38 146.22 80.67 100.67 176.50 207.22
N 9 8 9 9 6 8 9
Std dev 22.554 11.413 12.538 6.305 15.552 17.163 9.680
Min 313 276 126 71 81 155 195
Max 385 307 162 90 117 207 220
Negros
Mean 315.33 271.50 142.25 81.00 101.00 149.67 185.25
N 6 4 6 4 4 6 6
Std dev 20.887 21.579 7.973 8.602 8.206 15.240 14.511
Min 295 254 132 70 90 134 174
Max 356 303 155 90 109 179 214
Palawan
Mean 369.00 322.25 147.00 68.00 96.50 181.75 239.25
N 4 4 4 4 4 4 4
Std dev 10.100 9.179 7.958 5.477 8.185 7.544 8.958
Min 362 314 139 62 89 171 231
Max 384 334 158 74 108 187 252
Jolo
Mean 332.00 292.67 142.67 75.00 99.00 166.67 204.00
N 1 3 3 1 1 3 3
Std dev — 7.506 3.215 — — 15.308 1.000
Min — 285 139 — — 149 203
Max — 300 145 — — 176 205
Tawitawi
Mean 343.00 301.00 139.00 77.00 101.00 171.00 216.00
Palawan has a very long palate, contrasting with smaller and relatively broader, with a wider, higher-
short nasals and a fairly short basal length; Jolo differs crowned occiput. Tawitawi, compared with Jolo, is
in short nasals and is narrow across the postorbital slightly bigger and narrower; compared with Pala-
processes, with a long condylobasal length. wan, it is smaller, with a wider, higher-crowned oc-
In a discriminant analysis of the lower dentition of ciput; compared with Mindanao, it has a somewhat
several Philippine pigs, all are distinct. Palawan has a shorter greatest length compared with condylobasal
long P2 compared with others. Jolo has a long P2, a length, a narrower bizygomatic breadth and oc-
short M1. ciput, and shorter nasals. As far as the teeth are con-
Table 5 gives the univariate statistics for skull cerned, there is a general trend for S. celebensis and
measurements, and table 6 for mandibular tooth S. cebifrons to be small in size, and S. ahoenobarbus
mea surements. The skull for Jolo is small, like to be large, but the molars and fourth premolar
Luzon, with a narrower, lower-crowned occiput of the Jolo pig are more or less as large as those of —-1
and a wider palate; compared with Palawan, it is S. ahoenobarbus, while the anterior premolar of —0
349-47558_ch01_1P.indd 43 5/31/11 7:05 PM

44 Artiodactyla
Table 6 Length measurements of mandibular cheekteeth for Philippine pigs and Sus celebensis
Lp1 l Lp2 l Lp3 l Lp4 l Lm1 l Lm2 l Lm3 l
Mindoro
Mean 7.900 10.733 11.800 12.900 13.400 18.067 29.333
N 2 3 3 3 3 3 3
Std dev 1.5556 .1155 .2000 .5292 1.9313 1.1590 2.6026
Min 6.8 10.6 11.6 12.5 11.7 17.0 26.8
Max 9.0 10.8 12.0 13.5 15.5 19.3 32.0
Luzon
Mean 8.860 11.000 11.643 12.686 13.343 16.563 26.738
N 5 7 7 7 7 8 8
Std dev 1.2402 .4041 .5884 1.0746 .7613 .6844 1.3469
Min 8.0 10.5 11.0 11.3 12.0 15.8 25.5
Max 11.0 11.8 12.4 14.0 14.1 17.6 29.8
Mindanao
Mean 8.083 10.650 11.633 12.100 13.700 17.783 26.600
N 6 6 6 6 6 6 5
Std dev .3545 .7287 .7033 .7155 .7266 .5913 .9301
Min 7.6 9.8 11.0 11.0 12.7 16.6 25.7
Max 8.5 11.5 12.7 12.9 14.9 18.2 28.0
Negros
Mean 6.967 9.400 10.867 10.800 13.133 16.167 25.067
N 3 2 3 3 3 3 3
Std dev 2.1733 .4243 .5686 .1732 .6110 .2082 2.2301
Min 5.0 9.1 10.4 10.7 12.6 16.0 23.4
Max 9.3 9.7 11.5 11.0 13.8 16.4 27.6
Palawan
Mean 7.400 13.050 14.050 14.250 17.100 19.975 32.125
N 2 2 2 2 2 4 4
Std dev .5657 .0707 .3536 .2121 .1414 1.1266 1.9788
Min 7.0 13.0 13.8 14.1 17.0 19.0 29.9
Max 7.8 13.1 14.3 14.4 17.2 21.0 34.4
Sulawesi
Mean 6.175 9.371 10.314 11.286 12.650 16.571 26.429
N 4 7 7 7 6 7 7
Std dev .9251 .4821 .9173 .9512 1.0075 .6075 2.3705
Min 5.0 9.0 9.0 10.0 11.0 16.0 24.0
Max 7.2 10.0 11.0 13.0 13.9 17.5 30.0
Jolo
Mean 8.333 11.667 12.000 13.500 14.333 17.667 32.333
N 3 3 3 3 3 3 3
Std dev .5774 .5774 .0000 .5000 1.1547 1.5275 .5774
Min 8.0 11.0 12.0 13.0 13.0 16.0 32.0
Max 9.0 12.0 12.0 14.0 15.0 19.0 33.0
S. philippensis, S. oliveri, and the Jolo pig are the larg-

est in the series. Sus philippensis Nehring, 1886
Philippine wild pig
Sus ahoenobarbus Huet, 1888 Black, sometimes with a pale snout band and red
Palawan wild pig patches in the mane. Large crown crest (or “toupée”)
Very small size; darker in color than S. barbatus. in prime males. Inferior surface of the lower canine
S. ahoenobarbus was originally classed as a sub- exceptionally broad.
species of S. barbatus, as, for example by Groves Luzon, Mindanao, and offshore islands.
-1— (1981). Probably there are at least two subspecies to be
0— Palawan, Balabac, and the Calamianes group. included here (as in Groves, 1981d, 1997a); they differ
349-47558_ch01_1P.indd 44 5/31/11 7:05 PM

s uidae 45
on average, but we can fi nd no absolute differentia- —The two long-recognized genera, and their
tion between them. tortuous nomenclatural history.
—The shock of discovering, 30 years ago, that the
Sus oliveri Groves, 1997 genus Catagonus, described as a fossil, is part of
Mindoro wild pig the living fauna.
Long palate; wide bizygomatic breadth; wide M2; —Geographic variation—how many species in
fairly long P3, compared with a short M1. each genus?
S. oliveri was originally described as a subspecies —The problems raised by the description of the
of S. philippensis, but it is, in fact, absolutely distinct. new species Pecari maximus in 2007.
Mindoro.
Chromosomes of peccaries were studied by Be-
nirschke & Kumamoto (1989) and Vassart, Pinton, et al.
Sus cebifrons Heude, 1888
(1994).
Visayan wild pig
Gongora & Moran (2005), using three mitochondrial
Mane long, extending along the back to the rump;
and five nuclear sequences, found that Tayassu and
large facial warts. Skull very short, exceptionally
Catagonus form a clade (with 100% bootstrap, in the case
high-crowned.
of the mitochondrial sequences) with respect to Pecari.
Breeding groups from both Negros (S. c. negrinus
Sanborn, 1952) and Panay (no taxonomic designation) Tayassu Fischer von Waldheim, 1814
have been established in the Philippines, and from
white-lipped peccaries
photographs kindly supplied to CPG by William Oli-
Chromosomes of white lipped peccaries from Brazil
ver, they are undoubtedly different in external appear-
had 2n = 26, but the X chromosome was found to be
ance. There is, however, a problem in representing this
polymorphic (Vassart, Pinton, et al., 1994): that from
difference taxonomically: the prior available name for
São Paulo was telocentric, whereas that from S Ama-
any Visayan pig is Sus cebifrons, described from Cebu, a
zonia was subtelocentric.
population now apparently extinct and known by a
We recognize here only a single species, with four
single adult male skull, but not by its external appear-
subspecies, although, at present, we do not know
ance. It seems, then, that we have to await the demise
their full distributions, nor can we place our small
of one or more of the adult males of the Panay breed-
samples from E Brazil.
ing group before we can place this population, deter-
mining whether it is identical to the Cebu form, or
Tayassu pecari pecari (Link, 1795)
distinct from both this and the Negros form.
1795 Sus pecari Link. “Paraguay”; but, according to
Hershkovitz (1963), Cayenne.
TAYASSUIDAE PALMER, 1897
1921 Tayassu pecari beebei Anthony. Kartabo, Guyana.
The most striking difference between this family
Skins dark, but much more extensively light straw on
(peccaries) and the Suidae (pigs) concerns, fi rst of all,
the underparts than the other geographic samples; dark
the canines, which in the peccaries point downward
color invading the underside only on the chest. White
in both sexes in the normal mammalian fashion; and,
hairs on the body; longer, light bases on the dark hairs,
second, the lack of lateral false hoofs on the hindfeet.
with the skins more brindled than other taxa; inter-
The dentition in the Tayassuidae is very heavy, and
mixed hairs may be whitish, straw-colored, or more
the molars are shorter than in the Suidae; the whole
red-brown, giving the generally darker color. White on
facial skeleton forms a kind of compact, load-bearing
the face fairly developed, but not unduly extensive.
cylinder.
We have seen specimens from Guyana and
Herring (1974) found that most sutures of the pal-
Suriname.
ate and the facial skeleton fuse very early in peccaries,
and suggested that this relates to rooting and mastica-
Tayassu pecari aequatoris (Lönnberg, 1921)
tion, with concomitant needs to reinforce the snout.
On this family, see Grubb & Groves (1993). On the 1921 Dicotyles pecari aequatoris Lönnberg. Gualea, Pichincha
differences between Tayassu and Pecari, see Wood- Province, Ec uador.
burne (1968); on Catagonus, see Wetzel (1977).
Median dorsal mane very long; skins vary, especially
The problems and topics concerning this family —-1
in the length of the light bases of the hairs; some hairs
include: —0
349-47558_ch01_1P.indd 45 5/31/11 7:05 PM

46 Artiodactyla
with light tips, especially on the light nuchal mane. temporomandibular width; NE Peru is the opposite;
Less white on the face than in nominotypical T. p. pe- SE Peru is intermediate.
cari. Size relatively small. The two Peruvian samples and the one from Bo-
E Colombia (Bauco and Bogotá), Ecuador, Peru, livia again overlap, but they differ on average. When
Bolivia; details of the distribution remain to be worked comparing samples from Brazil with those from the
out. Guyanas, it is, curiously, the sample from Pará which
is somewhat separate.
Tayassu pecari albirostris (Illiger, 1815) Table 7 shows that Amazonia and Mato Grosso /
Paraguay have relatively narrow bicanine apophyses
1815 Sus albirostris Illiger. Paraguay.
compared with others; all the South American sam-
1817 Dicotyles labiatus G. Cuvier.
ples have a narrow occiput, compared with Central
Noticeable straw color in the axil and the groin; face American ones; the South Americans (except Colom-
markings extensively white. Hairs with straw-colored bia and Venezuela) tend to be larger than the Central
bases, the rest being black or brown, making it Americans; SE Brazil has a short greatest length com-
dark pepper-and-salt to nearly black overall color pared with condylobasal length (i.e., a less projecting
(the light hair bases very short in skins with the occipital crest); and Pará has large teeth.
latter tone). Skull with relatively narrow bicanine We conclude that, thus far, craniometry contrib-
apophyses. utes rather little toward understanding geographic
Paraguay, Mato Grosso, and Amazonas. variation in white-lipped peccaries, and a great deal
more material is needed.
Tayassu pecari ringens Merriam, 1901
Pecari Reichenbach, 1835
1901 Tayassu albirostris ringens Merriam. Apazote, Campeche
collared peccaries
(Yucatán).
Cope (1889) distinguished two species of collared pec-
1912 Tayassu albirostris spiradens Goldman. Talamanca,
cary, as follows:
Costa Rica.
1. D. tajassu (from S Brazil)—malar crest
Distinguished by the greater extension of the whitish
terminating above the infraorbital foramen;
face markings (the light area including the whole
nasal bones rounded in cross section; fi rst
muzzle, from the tip to midway between the nose
superior premolar tritubercular or rounded
and the eyes, and extending backward along the sides
in outline, premolariform; molars not
of the lower jaw below the ears). Ill-defi ned white
wrinkled.
band above each pair of hind hoofs. Underparts griz-
2. D. angulatus, sp. nov. (from Texas)—malar
zled black and fulvous. Skull relatively small, distin-
crest continuing forward to the base of the
guished by the wide canine apophyses and a short,
canine alveolus; nasal bones pinched or angulate
broad occipital crest.
on the middle line; first superior premolar
quadritubercular, with intermediate tubercles,
For our craniometric study, we analyzed N and S
and quadrate in outline, molariform; molars
groups separately. Among the N groups, Central Amer-
wrinkled.
ica is separate, but the Guyanas (Guyana and Suri-
name) and Colombia are intermixed, differing only He also noted the following:
on average.
The characters cited are constant, although
Central American skulls have wide canine
the amount of angulation of the nasal bones
apophyses and a fairly broad occiput, but a narrow
in D. angulatus is subject to some variation.
bi-temporomandibular joint width, and a relatively
Another character, and generally constant, is
low greatest length compared with condylobasal
the form of the fossa above the diastema. In
length.
D. tajassu it is a narrow groove; in D. angulatus
Among groups in N South America, there is little
it is a wide fossa (Cope, 1889:147).
discrimination: the SE Peru sample overlaps with the
Guyanas, while NE Peru is slightly more separate. These comparisons were based on 16 skulls from S
The Guyanas have a very high greatest length and Brazil and 6 from Texas. We fi nd that, despite Cope’s
-1— high bicanine apophyses and palate breadth, con- small sample sizes, the main differences he elucidated
0— trasting with low condylobasal length and bi- are, indeed, absolutely consistent.
349-47558_ch01_1P.indd 46 5/31/11 7:05 PM

Table 7 Measurements for Tayassu samples
Bican Pal br
Gt l Cb l Bi-THJ apoph Teeth ext Occ br
Guatemala, Belize
Mean 279.00 247.00 113.50 63.33 78.67 53.50 43.17
N 6 6 6 6 6 6 6
Std dev 5.967 4.336 3.886 1.633 2.160 1.049 2.927
Min 270 239 106 61 76 52 40
Max 287 252 116 66 82 55 48
Costa Rica
Mean 274.00 244.00 117.00 60.00 73.00 52.00 39.00
N 1 1 1 1 1 1 1
Colombia
Mean 279.25 243.63 116.75 61.37 77.13 53.13 37.75
N 8 8 8 8 8 8 8
Std dev 8.155 6.232 4.367 7.110 2.850 2.031 4.713
Min 270 237 112 53 73 51 32
Max 291 257 124 70 83 57 45
Venezuela
Mean 270.50 241.00 111.50 59.50 73.50 51.00 36.00
N 2 2 2 2 2 2 2
Min 269 239 111 59 72 50 35
Max 272 243 112 60 75 52 37
Guyana, Suriname
Mean 292.33 247.63 119.78 63.56 79.67 54.56 42.44
N 9 8 9 9 9 9 9
Std dev 10.100 6.140 5.094 4.927 2.739 1.333 2.404
Min 278 241 111 56 77 53 39
Max 307 259 130 73 84 56 46
NE Peru
Mean 282.33 249.00 120.00 59.00 79.00 53.67 38.33
N 6 6 6 6 6 6 6
Std dev 5.715 7.099 2.966 2.191 4.243 1.366 4.633
Min 277 238 117 57 72 52 35
Max 291 258 125 63 85 56 47
SE Peru
Mean 281.71 244.67 117.63 60.37 77.13 53.75 39.86
N 7 6 8 8 8 8 7
Std dev 7.064 8.430 2.326 3.662 3.682 1.753 3.237
Min 273 234 114 55 69 52 36
Max 291 256 120 66 80 56 44
Bolivia
Mean 281.50 245.20 116.63 64.38 78.50 54.88 41.13
N 8 5 8 8 8 8 8
Std dev 12.917 11.077 5.553 3.739 4.140 1.553 3.907
Min 255 230 110 61 69 53 35
Max 295 257 124 70 82 58 46
Amazonia
Mean 288.50 251.00 116.50 58.00 79.00 52.50 41.50
N 2 2 2 2 2 2 2
Min 285 248 116 56 78 52 38
Max 292 254 117 60 80 53 45
Mato Grosso, Paraguay
Mean 288.00 249.00 118.50 59.50 78.50 53.00 40.50
N 2 2 2 2 2 2 2
Min 284 246 116 59 77 52 40
Max 292 252 121 60 80 54 41 —-1
(continued) —0
349-47558_ch01_1P.indd 47 5/31/11 7:05 PM

48 Artiodactyla
Table 7 (continued)
Bican Pal br
Pará
Mean 286.00 254.33 119.67 64.33 81.00 54.67 45.00
N 3 3 3 3 3 3 3
Std dev 7.211 2.309 8.145 8.737 1.000 3.786 5.292
Min 280 253 114 57 80 52 41
Max 294 257 129 74 82 59 51
SE Brazil
Mean 285.80 245.20 119.80 64.40 77.20 53.80 42.60
N 5 5 5 5 5 5 5
Std dev 11.167 4.658 2.049 2.881 2.588 .837 5.505
Min 278 238 118 61 73 53 34
Max 305 249 123 68 80 55 49
Lydekker (1915a) accepted Cope’s two species, but, distribution within that country. In Colombia, skulls
curiously, Lydekker did not mention the most strik- of “N” type are from the following:
ing distinguishing character, that of the malar crest,
Socorré (7.51° N, 76.17° W)
mentioning instead only the characters of the basal
Unguia (8.02° N, 77.05° W)
angulation of the nasals, the palatal ridge, and the
Coloso (9.29° N, 75.21° W)
preorbital depression.
Los Micos (4.58° N, 75.49° W)
We refer here to what Cope called D. angulatus and
Rio Sandó, 100–160 m (5.13° N, 77.05° W)
D. tajassu as the “N” and “S” types, respectively. De-
Old Magdalena Valley
spite the distinctiveness of the two morphological
types, most later authors (probably following Cabrera, Skulls of “S” type are from the following:
1961) have united them, recognizing only a single
Rio Mecaya (0.28° N, 75.20° W)
species of collared peccary.
Rio Guapaya (ca. 5.00° N, 75.00° W)
The recent description of a new species of collared
Rio Saija, Cauca, 100 m
peccary, P. maximus (see van Roosmalen et al., 2006),
César
is difficult to test, given the small number of available
Narino, Candelilla, 200 m (1.27° N, 78.40° W)
specimens from Amazonia; it is not easy to use the
measurements published in that paper, because very Collared peccaries from French Guiana (Vassart,
few of them correspond to those used in the present Pinton, et al., 1994) had 2n = 35, like those previously
study. In their phylogenetic tree, employing several studied from North America, but there were differ-
mtDNA sequences and two short interspersed nu- ences in details: the X chromosome of the North
clear elements (SINEs), P. maximus was sister to three American specimens was metacentric, while that of
other collared peccary specimens: one from Arizona, the French Guiana specimens was acrocentric; and
one from Colombia, and one of unknown origin. As the North American animals had six medium-sized
Colombia is the country where “N” and “S” types metacentric autosomes, while those from French Gui-
both occur, this may be telling us only that P. maxi- ana had only five (one extra autosome therefore being
mus belongs to the “S” group, which we can in- acrocentric).
deed detect from the illustrations accompanying the In striking agreement with the cranial data, both
type description. Much more work is needed on this the mitochondrial and four nuclear sequences divided
question. North American (including Central American) and
We fi nd that “N” and “S” skull types are charac- South American lineages cleanly, with specimens from
teristic of Central American and South American col- Colombia being allocated to both (Gongora et al., 2006).
lared peccaries, respectively, and that they meet in Co- In the mitochondrial tree, two specimens known to be
-1— lombia. Therefore, it is worthwhile examining their from N Colombia, two from C Colombia, and one from
0—
349-47558_ch01_1P.indd 48 5/31/11 7:05 PM

s uidae 49
S Colombia assorted with the N clade; while one from

SE Colombia, one from N Columbia, one from S Co- Unknowns
lumbia, one known merely as being from “Colombia,” We presume that the following specimens belong to
and one known to be from Caqueta assorted with the S previously unnamed taxa:
clade. There was no geographic sorting in the N clade,
—Two skins from Paraguay are very distinctive,
but within the S clade there were significantly different
with the hairs multibanded straw/black,
clusters from Colombia, (N) Bolivia, and Argentina.
giving an overall fairly light tone, but with a
Interestingly, the nuclear network showed that those
solid, thick, black dorsal stripe. Collar barely
Colombian specimens were that were part of the N
marked amid all the long, light hair bases.
clade were well within it, not at its root, implying a re-
One from Sierra Santa Barbara, Jujuy, with a
invasion into Colombia after the two lineages had been
less striking differentiation, but the dorsal
separated for some while.
stripe still well marked; collar much better
Therefore, it is clear that there are at least two spe-
marked.
cies involved in this genus; we fi nd evidence, in fact,
—One skin from San Francisco de Caup, Minas
for at least three (two “N,” one “S”), but there are
Gerais, is different from the others. No
many gaps in our dataset.
marked dorsal stripe; collar virtually absent;
predominance of the whitish bands on the
Pecari tajacu (Linnaeus, 1758)
hairs.
The “S skull-type” species: malar crest directed up-
ward, ending above the infraorbital foramen; nasal
bones more rounded in cross section; fi rst upper pre- Pecari angulatus (Cope, 1889)
molar premolariform; molars not wrinkled.
1889 Dicotyles angulatus Cope. Guadelupe Valley, Texas.
At present we cannot fi nd more than one species
1897 Dicotyles angulatus sonoriensis Mearns. San Bernardino
in South America (Colombia excepted).
Valley, Sonora, Mexico.
1901 Tayassu angulatus humeralis Merriam. Armería, Colima,
Pecari tajacu tajacu (Linnaeus, 1758)
Mexico.
1758 Sus tajacu Linnaeus. Pernambuco, according to
One of two “N skull-type” species. Malar crest slopes
Cabrera (1961).
forward, ending above the canine alveolus; nasal bones
Stated to have the general color being grizzled gray pinched or angulated; fi rst upper premolar quadritu-
or tawny. Six skins from Chapada and Miranda, Mato bercular, with intermediate tubercles, molariform;
Grosso, somewhat similar to the Paraguay and Jujuy molars wrinkled.
series, but the dorsal stripe sometimes less marked; Hair banding blackish and gray or white; head
collar often quite faint, but the collar in one skin even lighter in tone; collar white, but often not marked;
better expressed than in the Jujuy skin. sometimes with a black dorsal stripe. Skull larger and
Skull relatively large, long, narrow, vaulted. narrower than in P. tajacu.
We have seen specimens of this subspecies from Evidently a species characteristic of the dry zone
Mato Grosso, N Argentina, and Paraguay. of N Mexico and the southernmost United States.
Pecari tajacu patira Kerr, 1792 Pecari crassus (Merriam, 1901)

1792 Sus patira and Sus minor Kerr. French Guyana. 1901 Tayassu angulatus crassus Merriam. Metlaltoyuca,
1816 Dicotyles torquatus G. Cuvier. Puebla, Mexico.
1917 Pecari tajacu modestus Cabrera. Rio Napo. 1901 Tayassu angulatus yucatanensis Merriam. Tunkás,
1921 Pecari tajacu macrocephalus Anthony. Guyana. Yucatán.
1901 Tayassu nanus Merriam. Cozumel Island, off
Skins from the Magdalena River and Rio César, Co-
Yucatán.
lombia, and six skins from Guyana, are dark, due to a
1902 Tayassu crusnigrum Bangs. Boquete, Chiriquí,
predominance of dark bands on the hairs. Expression
Panama.
of the collar very good. Dark dorsal stripe not ex-
1917 Pecari angulatus bangsi Goldman. Boca de Cupe, Darión,
pressed, except down the nape. Skull relatively small,
Panama.
short, wide, depressed. —-1
—0
349-47558_ch01_1P.indd 49 5/31/11 7:05 PM

50 Artiodactyla
1926 Pecari angulatus nelsoni Goldman. Huehuetán, Table 8 Discriminant analysis of skulls for Pecari with
Chiapas. N versus S facial crest types in Colombia
1926 Pecari angulatus nigrescens Goldman. Chamelecón,
Cortés, Honduras. Predicted group
membership
“N skull-type,” like P. angulatus. Coat coarser; light hair
bands wide, making the overall color gray or white; N S Total
black dorsal stripe; head lighter; collar various, often Original count N 15 0 15
indistinct. S 0 4 4
A skin from Vera Cruz, however, is darker, only Original % N 100.0 — 100.0
slightly less dark than those from Guyana (see above); S — 100.0 100.0
there is no dorsal stripe, even on the nuchal area; no
Function
noticeable collar.
Presumably, this is the species of the humid rain 1
forest zone of S Mexico and Central America, extend- Gt l −1.866
ing into NW Colombia and Ecuador W of the Andes. Cb l .085
The following two names may be synonymous Bi-THJ .774
with P. crassus: Bican apoph 1.056
Pal br ext −.729
1. ?Pecari tajacu torvus (Bangs, 1898) Occ br .812
—Type locality: Santa Marta, Colombia.
—This name would have priority, if it is truly
identical with the North American humid smaller, wider across the temporomandibular joints,
forest species. and have a larger condylobasal length. Evidently there
2. ?Pecari tajacu niger ( J. A. Allen, 1913) is more differentiation in South America than we are
—Type locality: Esmeraldas, Ecuador. able to document.
— General color is stated to be a nearly E Peru and the Peru highlands groups separate
uniform black. well: the lowland skulls, compared with the highland
ones, have a longer condylobasal length, but a shorter
In an analysis of collared peccary skulls from
greatest length; and are wider across the palate and
North and Central America, Oaxaca plus Veracruz
the temporomandibular joints, but narrower across
(humid zone) versus Texas and N and C Mexico (arid
the canine apophyses.
zone) overlap only slightly. The Texas and C Mexico
A skull from Trinidad is part of the Colombia “S”
skulls are larger, with a somewhat broader occiput
scatter.
and narrower palates.
The Guyanas overlap with the C Peru highlands,
Whether taking the skulls of the “N” and “S” crest
and they only just overlap with E Peru lowlands;
types as a whole, or restricting the comparison to
compared with the latter, they have a larger greatest
those in Colombia (table 8), the “N type” are larger
length, but a smaller condylobasal length, and they
and narrower. Compared with Guatemala/Belize,
are narrower. All of these skulls, as well as the Peru
the Columbian “N type” skulls are not distinguished
highlands ones, differ from the Colombia “S” type in
morphometrically, whereas the Columbian “S types”
their somewhat larger size, but they are narrower
ones are clearly distinguished by being smaller and
across the canine apophyses.
broader. Comparing the two Colombian types with a
Guyana and a single Amazonas skull have a
sample from E Peru, however, the difference re-
larger condylobasal length (especially relative to
mains, but the Peruvian sample is again different
greatest length), and they are narrower across the
from either of the Columbian types. E Peru skulls
canine apophyses and the occiput. Mato Grosso dif-
have a larger condylobasal length, especially com-
fers from Pará in being wider across the temporo-
pared with the greatest length, and are broader across
mandibular joints; Guyana is more like Mato Grosso
the temporomandibular joints but narrower across
in this.
the canine apophyses; they have a somewhat broader
Pará, compared with Mato Grosso, Bolivia, and
palate, but a narrower occiput. The “S type” Colom-
Amazonas, is smaller, and wider across the canine
bian skulls, compared with the “N type,” are far
-1— apophyses and the occiput, compared with the tem-
0—
349-47558_ch01_1P.indd 50 5/31/11 7:05 PM

s uidae 51
poromandibular joints and the palate. Bolivian skulls the state of Mato Grosso but well N of the other
have a larger greatest length, especially compared Mato Grosso specimens) are separate from C Peru
with condylobasal length; they also have a very broad and from the main Mato Grosso / Bolivia sample,
palate and are very narrow across the temporoman- with a long condylobasal length and a short greatest
dibular joints. The single Amazonas specimen is an length, and a broad width across the temporoman-
extreme example of the Mato Grosso morphology. dibular joints, but a narrow one across the canine
Mato Grosso overlaps with Paraguay; Bolivia and apophyses.
São Paulo seem distinct. Bolivian skulls, as before, have Our conclusion is that, in our limited sample, Am-
a long greatest length and a short condylobasal length, azonas and E Peru form a single group; Mato Grosso
a broad palate, and narrow bi-temporomandibular and and Bolivia, together with C Peru, form another
canine apophyses breadths. São Paulo seems to differ group.
from Mato Grosso (N = 2) in having a short greatest Table 9 gives univariate statistics for the skulls
length compared with a fairly long condylobasal length, of collared peccaries. There are no great differences
and a narrow width across the temporomandibular in size (except for the small size of the single avail-
joints and the palate, compared with a greater width able skull from Trinidad). The São Paulo (including
across the canine apophyses and the occiput. Santa Catarina) sample stands out in its narrow width
E Peru and Amazonas (including a skull from across the temporomandibular joints. Samples from
“north of Xavantina, Serra do Roncador,” which is in the United States and the more arid N parts of Mexico
Table 9 Measurements for Pecari samples
Bican Pal br
USA
Mean 236.81 201.31 101.63 59.38 64.38 46.31 35.00
N 16 16 16 16 16 16 16
Std dev 7.378 6.610 5.760 4.080 2.918 1.537 3.706
Min 221 187 93 54 59 43 26
Max 247 210 116 71 70 50 41
Durango, Sinaloa
Mean 245.00 204.17 103.33 58.50 62.50 44.00 37.50
N 6 6 6 6 6 6 6
Std dev 8.198 5.419 3.266 2.665 1.761 1.897 3.564
Min 231 198 99 54 60 42 32
Max 254 214 108 61 65 46 43
Veracruz
Mean 239.00 204.50 104.50 60.00 66.00 46.00 32.00
N 2 2 2 2 2 2 2
Min 238 203 104 57 64 46 32
Max 240 206 105 63 68 46 32
Oaxaca, Tabasco, Guatemala, Belize
Mean 240.78 203.75 101.11 55.56 60.56 43.22 32.44
N 9 8 9 9 9 9 9
Std dev 7.807 6.692 5.600 3.167 3.127 1.481 4.304
Min 229 193 93 49 56 41 27
Max 254 216 109 59 66 46 38
Colombia, “N type”
Mean 236.00 200.35 97.33 55.06 63.11 45.11 34.37
N 16 17 18 18 18 18 16
Std dev 6.552 6.614 3.218 8.299 3.660 2.220 2.306
Min 227 183 92 48 55 42 29
Max 248 211 104 85 69 49 37
(continued) —-1
—0
349-47558_ch01_1P.indd 51 5/31/11 7:05 PM

Table 9 (continued)
Bican Pal br
Colombia, “S type”
Mean 229.50 199.50 98.50 55.00 60.50 40.00 34.50
N 2 2 2 2 2 2 2
Min 226 199 96 52 58 34 34
Max 233 200 101 58 63 46 35
E Peru, E Ec uador
Mean 240.69 209.80 103.14 53.50 66.21 46.86 35.57
N 13 10 14 14 14 14 14
Std dev 7.620 7.193 5.318 4.468 3.068 2.282 2.901
Min 227 202 95 48 61 44 31
Max 254 225 113 63 71 50 41
C Peru
Mean 242.20 205.80 100.00 51.00 62.60 45.80 32.40
N 5 5 5 5 5 5 5
Std dev 7.596 7.050 4.301 5.916 2.302 1.924 2.074
Min 231 197 94 44 59 44 31
Max 251 216 106 58 65 49 36
Trinidad
Mean 217.00 189.00 100.00 49.00 62.00 42.00 28.00
N 1 1 1 1 1 1 1
Guyana, Suriname, Roraima
Mean 244.07 207.93 102.79 53.64 64.86 44.64 30.86
N 14 14 14 14 14 14 14
Std dev 5.427 5.015 6.278 4.272 2.770 1.906 4.092
Min 233 202 87 48 61 41 22
Max 252 219 115 61 70 48 35
Pará
Mean 233.00 197.80 95.00 50.40 61.20 42.60 31.80
N 5 5 5 5 5 5 5
Std dev 7.778 7.430 2.345 3.975 4.438 .548 3.701
Min 224 188 93 47 56 42 26
Max 245 208 99 56 67 43 36
Amazonas
Mean 252.17 216.00 100.00 49.00 66.00 45.00 36.00
N 6 1 1 1 1 1 1
Std dev 11.907 — — — — — —
Min 230 — — — — — —
Max 262 — — — — — —
Mato Grosso, E Bolivia
Mean 231.38 194.25 99.75 53.00 61.50 44.00 29.63
N 8 8 8 8 8 8 8
Std dev 10.127 9.528 4.400 4.957 3.817 2.000 4.809
Min 209 174 94 45 56 40 21
Max 240 206 106 60 67 46 36
N Argentina, Paraguay
Mean 231.00 196.86 102.57 54.88 64.71 45.25 33.20
N 5 7 7 8 7 8 5
Std dev 10.320 7.381 3.823 3.091 2.812 2.188 5.263
Min 222 187 98 50 60 42 28
Max 247 208 110 59 67 50 42
São Paulo, Santa Catarina
Mean 232.00 201.50 95.00 52.50 65.50 45.00 35.50
N 2 2 2 2 2 2 2
-1— Min 229 199 93 52 62 45 35
0— Max 235 204 97 53 69 45 36
+1—
349-47558_ch01_1P.indd 52 5/31/11 7:05 PM

s uidae 53
are noticeable for the wide distance across the canine

apophyses. Catagonus wagneri (Rusconi, 1930)
On this species, see Wetzel (1977). The discovery of
Catagonus Ameghino, 1904 such a large species of living mammal— one, more-
2n = 20; all chromosomes are metacentric, with un- over, that was previously known only as a fossil—was
usually large heterochromatin blocks (Benirschke one of the great biological surprises of the 20th
et al., 1985). century.
—-1
—0
349-47558_ch01_1P.indd 53 5/31/11 7:05 PM

concept, it is irresistible to try to fi nd such potential
6
evidence of affi nities, and of a very ancient aquatic
heritage!
Hippopotamus Linnaeus, 1758
Hippopotamus amphibius is the only survivor of a once-
diverse radiation.
Hippopotamus amphibius Linnaeus, 1758

common hippopotamus
Grubb (1993a) recognized subspecies of this spe-
Hippopotamidae cies, but was inclined to doubt whether there was
really much geographic differentiation; he recom-
mended further morphological study, and this is
still needed.
Mitochondrial control region sequences (Okello
et al., 2005) showed low differentiation among hippo-
potamus populations from Uganda, Kenya, Tanzania,
The Whippomorpha concept. The aquatic adaptations and Zambia, with only weak geographic structuring
of the Hippopotamidae: are they really those of whales and evidence of comparatively recent population ex-
writ small, or is it just the “Aha” effect? Is there signifi- pansion; the most ancestral haplotypes were clustered
cant geographic variation in the common hippo? Is the around Murchison Falls National Park, and the au-
pygmy hippo really a survivor of the fossil genus Hexa- thors related this expansion to the changing river pat-
protodon? The pygmy hippo of the Niger Delta: a dis- terns consequent on the E African tectonic activity of
tinct species. the Middle Pleistocene. We may note that, during the
Two decades of molecular research have now Early Pleistocene, there was at least one other spe-
clearly established that cetaceans (whales and dol- cies, Hippopotamus gorgops, that was probably a com-
phins) are not only part of the Artiodactyla, but are a petitor and was much commoner in the fossil record;
sister group to hippos. This sister-group relationship presumably a rapid expansion of Hippopotamus am-
was formalized by Waddell et al. (1999) under the phibius would have followed on the extinction of its
name Whippomorpha. Skinner & Chimimba (2005) competitor.
split the Artiodactyla into several distinct orders, of Choeropsis Leidy, 1853
which Whippomorpha was one. We gave our reasons
(see Part 2) for preferring to retain the order Artio- Very small in size; anterior ends of the nasals com-
dactyla and recognize Whippomorpha at the subor- paratively long and curved downward; orbit placed
dinal level, dividing the suborder into two in- well below the dorsal outline; palatine with a strong
fraorders, Cetacea (not treated here) and Ancodonta, posterior nasal spine; tympanic bulla inflated; lateral
the latter including (in the living fauna) only the fam- notch posterior to the basilar tubercles; cranial roof
ily Hippopotamidae. curved downward posteriorly; only two mandibular
incisors.
As a result of the work of Coryndon (1978 and else-
HIPPOPOTAMIDAE GRAY, 1821 where), it became usual to assign the living pygmy
Some of the characteristics of members of this family, hippo to the genus Hexaprotodon. Subsequent to the
such as copious body fat, relative hairlessness, com- description of further hippos, however, Boisserie (2005)
pact build, and underwater communication, could be reinstituted the genus Choeropsis, which he consid-
seen either as ad hoc aquatic adaptations in the fam- ered the sister to all other hippos (except probably the
ily, or, nowadays, as inherited from their common fossil genus Saotherium). The diagnosis of the genus
ancestor with Cetacea. We simply know too little given above depends partly on his analysis.
of the fossil history of hippos and their relatives, but G. Corbet (1969) described the Niger Delta taxon
in the light of the relatively new Whippomorpha as a subspecies of C. liberiensis, but noted that 9 of the
-1— 11 character states by which it differed were discrete.
0—
349-47558_ch01_1P.indd 54 5/31/11 7:05 PM

h ip p o p o t amidae 55
While the sample size was small (N = 2– 4, compared

with N = 10–14 [N = 3 in the case of the nasal measure- Choeropsis heslopi Corbet, 1969
ments] of nominotypical C. l. liberiensis), the sheer Niger Delta pygmy hippopotamus
number of the differences, and the distinct gaps be-
1969 Choeropsis liberiensis heslopi Corbet. Omoku, Owerri
tween the two taxa in many of them, give confidence
Province, Nigeria (5° 19' N, 6° 40' E).
that the two really are absolutely distinguishable,
with, as Corbet put it, “no indication of clinal variation Dorsal margin of the zygomatic process of the squa-
within the range of liberiensis affecting the characters mosal not so strongly sinuous, forming only a slight
that distinguish” C. heslopi. We therefore have little concavity; nasals narrow, the smallest width <15% of
hesitation in recognizing two species in this genus. the greatest length, the greatest width 22%–26% of
the greatest length; bullae rounded, the anterior mus-
Choeropsis liberiensis (Morton, 1844) cular processes prominent; wide median concavity
western pygmy hippopotamus in the frontal bones; no (or only a very rudimentary)
Dorsal margin of the zygomatic process of the squa- median process on the posterior margin of the palate;
mosal strongly sinuous, forming a rounded 90° con- P2 narrow, its width 54%–58% of its length, with no
cavity; nasals broad, the smallest width >15% of the posteromedian ridge; no posteromedian ridge on
greatest length, the greatest width 32%–37% of the P3; these two premolars relatively short, their length
greatest length; bullae angular, the anterior muscular only 42%– 45% of the total toothrow; median nasals
processes rudimentary; frontal bones flat, or slightly tips long, extending about 10 mm beyond the adja-
convex in the midline; very long (6–10 mm) median cent concavities; deep ventral grooves on the malars.
process on the posterior margin of the palate; P2 (Following Corbet, 1969.)
wide, its width 67%– 80 % of its length, with a promi- Niger Delta. Unfortunately, pygmy hippos appear
nent posteromedian ridge; posteromedian ridge on not to have been recorded in the Niger Delta since the
P3; these two premolars long, their length 46%– 60 % 1940s.
of the total toothrow; median nasal tips usually short;
ventral grooves on the malars usually shallow. (Fol- It appears that one of the dwarf hippos that are now
lowing G. Corbet, 1969.) known from Madagascar may have persisted until
Sierra Leone to Ivory Coast. the late 20th century (Burney & Ramilisonina, 1999).
—-1
—0
349-47558_ch01_1P.indd 55 5/31/11 7:05 PM

Our arrangement follows Meijaard & Groves
7
(2004a), and we here give only brief summaries of the
fi ndings of that revision.
Tragulus napu Group

The large Malay chevrotains (see Meijaard & Groves,
2004a).
Large size; relatively small auditory bullae (negative
allometry?). Usually a distinct, characteristic pelage.
Tragulus napu (F. Cuvier, 1822)

Tragulidae and Tragulus napu napu (F. Cuvier, 1822)
Moschidae 1822 Moscus napu F. Cuvier. S Sumatra.

1906 Tragulus bancanus Lyon. Bangka.
1935 Tragulus javanicus abjectus Chasen. Sirhassen Island.
Sumatra, except for the N (Medan and presumably

farther N); Malay Peninsula; Pangkor Island; Lang-
The Tragulidae: astonishing diversity in these small- kawi; Serasan; Bangka; Borneo; and Pulau Laut [Laut
sized, little-known animals. Tragulus: why not to try Island].
to shove 50 “subspecies” into just two species. Mos-
chiola as an example of the astonishing biodiversity of Tragulus napu neubronneri Sody, 1931
Sri Lanka. Hyemoschus and its relationship to other Maintained provisionally by Meijaard & Groves (2004a)
mouse deer and to the fossil genus Dorcatherium. as a separate subspecies, as the occipital height and the
mandibular toothrow length appear to be consider-
ably greater than in T. n. napu.
TRAGULIDAE MILNE EDWARDS, 1864 N Sumatra.
The lacrimal orifice is single, elongated, and situated
just inside the rim of the orbit (Leinders & Heintz, Tragulus napu niasis Lyon, 1916
1980). Maintained provisionally by Meijaard & Groves (2004a)
Mouse deer (chevrotains) are sister to all other as a separate subspecies, distinguished from T. n. napu
remnants, and on this basis we separate them into an by the shorter hindfoot.
infraorder Tragulina; other ruminant families belong Nias.
to an infraorder Pecora.
Tragulus napu terutus Thomas & Wroughton, 1909
Tragulus Brisson, 1762 Distinguished by the long, narrow auditory bullae;
This was one of the generic names which A. Gentry (1994)
the wide skull; and the high occiput.
requested to be conserved, even though other names
from Brisson (1762) were to be rejected.
Tragulus napu bunguranensis Miller, 1901
Distinguished from other members of the species by
Chromosomes were studied by Kim et al. (2004), who its large size. Meijaard & Groves (2004a) consider that
compared the results with earlier studies. All Tragulus this is possibly a distinct species.
appear to have 2n = 32. T. kanchil (reported as T. ja- Bunguran, in the N Natuna Islands.
vanicus) from Pahang and Selangor have a metacen-
tric Y chromosome; in the same species from Sabah, Tragulus napu rufulus Miller, 1900
it is subtelocentric. In T. williamsoni from Yunnan,
1900 Tragulus rufulus Miller. Tioman Island.
the Y was submetacentric, and larger than in the Ma-
1903 Tragulus formosus Miller. Bintang.
laysian males. In T. napu from Pulau Tioman [Tio-
1903 Tragulus pretiosus Miller. Lingga.
man Island], the Y was subtelocentric (like T. kanchil
from Sabah) but larger, and the short arm of the X Distinguished by the very bright reddish color; also
-1— chromosome was much shorter than in the javanicus by the nape stripe being absent or, occasionally, pres-
0— group. ent but vague.
349-47558_ch01_1P.indd 56 5/31/11 7:05 PM

t rag ul idae and mo s ch idae 57
Many islands of the Riau Archipelago (Batam,

Galang, Sekikir, Bulan, Bintang, Lingga, Bakong, Se- Tragulus versicolor Thomas, 1910
bangka), and Pulau Tioman [Tioman Island]. See above (for the species-group).
Tragulus napu banguei Chasen & Kloss, 1931 Tragulus javanicus Group
Very small in size; throat pattern distinctive. The small Malay chevrotains (see Meijaard & Groves,
Banggi and Balembangan. 2004a).
Incertae sedis Tragulus javanicus (Osbeck, 1765)

Meijaard & Groves (2004a) suggested that, as had been
Moschus stanleyanus Gray, 1836 (locality
proposed by earlier authors, there may indeed be two
unknown)
distinct mouse deer on Java, but this needs substantial
Tragulus brevipes Miller, 1903
further work.
Tragulus napu amoenus Miller, 1903 (= Tragulus
Java.
napu jugularis Miller, 1903; these two taxa are
evidently color morphs of the same species)
Tragulus kanchil (Raffles, 1821)
Tragulus napu batuanus Miller, 1903
Tragulus kanchil kanchil (Raffles, 1821)
Tragulus pinius Lyon, 1916
Sumatra.
Tragulus russulus Miller, 1903
Tragulus kanchil hosei Bonhote, 1903
Tragulus nigricans Thomas, 1892
1903 Tragulus kanchil hosei Bonhote. Baram, Sarawak.
Upper parts washed with black; neck mixed black and
1935 ?Tragulus kanchil abruptus Chasen. Pulau Subi [Subi
fulvous.
Island].
Balabac.
The following taxa are provisionally recognized NW, W, E, and S of Borneo; Subi Island(?).
as subspecies of this species by Meijaard & Groves
(2004a): Tragulus kanchil klossi Chasen, 1934
Larger, at least on average, than other T. kanchil, and
Tragulus billitonus Lyon, 1906-the allocation to this
apparently absolutely larger than T. k. hosei; pelage
species-group is uncertain
mottled, more like T. napu.
Tragulus napu hendersoni Chasen, 1940—the
At least the C and E parts of Sabah, and the N
allocation to this species-group is uncertain
area of East Kalimantan. The closest locality for T.
Tragulus nigrocinctus Miller, 1906— could be a
k. hosei is the Baram River, E Sarawak; there are
distinct species (Meijaard & Groves, 2004a)
no specimens available for any mouse deer from S
Kalimantan.
Tragulus versicolor Group Possibly a distinct species, according to Meijaard
Skull very small, but relatively wide; relatively wide, & Groves (2004), but this needs further study.
rounded bullae (possibly allometric); very long na-
sals. Hair very coarse; marked contrast between the Tragulus kanchil ravulus Miller, 1903
ochraceous-buff foreparts and the clear gray poste- Paler than T. k. ravus, with a less defi ned nape stripe;
rior half (starting behind the shoulders); no dark me- auditory bullae short, narrow; nasals short; short man-
dian line on the white underside; no dark lines from dible; narrower braincase.
the eye to the nose; absence of the dark collar that is Adang and Rawi islands.
almost universal in other Tragulus. On the throat, the
lateral reddish lines converge posteriorly but do not Tragulus kanchil affinis Gray, 1861
touch, with the central white throat line being con-
1861 Tragulus affinis Gray. Cambodia.
tinuous with the white on the underparts.
1903 Tragulus kanchil pierrei Bonhote. Bienhoa, S Vietnam.
Meijaard & Groves (2004a) consider this Indochi-
nese species as quite distinct from either of the other Craniometrically not distinct from other mainland
two species-groups, and it is not, as is commonly sup- taxa, but externally pale, without blackening on the
posed, a member— even, a subspecies!— of the T. napu neck. —-1
group. Vietnam, Laos, Cambodia, and SE and E Thailand. —0
349-47558_ch01_1P.indd 57 5/31/11 7:05 PM

58 Artiodactyla
Tragulus kanchil ravus Miller, 1902 Moschiola Gray, 1853

1902 Tragulus ravus Miller. Trang, southernmost Type species: Meminna indica (Gray, 1843), by monotypy.
Thailand.
This genus, formerly considered monotypic, was re-
1903 Tragulus lancavensis Miller. Langkawi Island.
cently revised by Groves & Meijaard (2005). There are
Paler than T. k. fulviventer, with a very less distinct nape three (possibly four) distinct species.
stripe.
SW Thailand, C and N Malay Peninsula, Lang- Moschiola indica (Gray, 1852)
kawi and Butang islands.
1852 Meminna indica Gray. E Ghats, India.
Tragulus kanchil augustiae Kloss, 1918 Dull brown; stripes and spots very clear, white; upper
Nape stripe well defi ned. row of spots forming a continuous stripe on the shoul-
Tenasserim, from 10° N to about 16° 30' N. der, but disintegrating into spots halfway back on the
body; other longitudinal stripes broken completely
Tragulus kanchil fulviventer Gray, 1836 into rows of elongated spots; three spot-rows across
the croup and the rump partly fused into stripes;
1940 Tragulus javanicus pumilus Chasen
crown and forehead dark brown; underside pale beige,
Distinguished by its pale fulvous underparts. becoming creamy medially. Size very large; long hind-
Malay Peninsula, approximately S of the Thai– legs; hindfoot length >140 mm. Skull broad, especially
Malay border (including Singapore, the type locality). across the rostrum; zygomatic breadth >50 mm; oc-
cipital height (measured from the basion) >27.8 mm;
Tragulus kanchil everetti Bonhote, 1903 width across the canine alveoli >15.2 mm; least maxil-
lary breadth >14.5 mm; braincase breadth >34.2 mm.
1903 Tragulus kanchil everetti Bonhote. Pulau Bunguran
Much of India, and formerly—but perhaps not sur-
[Bunguran Island] (N Natuna Islands).
viving in recent times?— also in Nepal (Baral et al.,
1903 Tragulus natunae Miller. Bunguran.
2009).
Resembling T. k. hosei, but a much deeper red on the
flanks; teeth larger; bullae smaller. Moschiola meminna (Erxleben, 1777)
The following taxa are provisionally recognized
1777 Moschus meminna Erxleben. Wariayapola, 7° 37' N, 80°
as subspecies of this species by Meijaard & Groves
13' E, 460 m, North Central Province, Sri Lanka.
(2004a):
Color and pattern very like M. indica, but the upper row
Tragulus kanchil anambensis Chasen & Kloss, 1928
of spots not extensively fused into a continuous stripe
Tragulus kanchil carimatae Miller, 1906
on the shoulder. Size small; hindlegs relatively long; hind-
Tragulus kanchil fulvicollis Lyon, 1908
foot length 120–144.5 mm. Zygomatic breadth <50 mm;
Tragulus kanchil insularis Chasen, 1940
occipital height <27.5 mm; width across the canine alveoli
Tragulus kanchil lampensis Miller, 1903
<15 mm; braincase breadth <33 mm.
Tragulus kanchil luteicollisi Lyon, 1906
Dry zone of Sri Lanka.
Tragulus kanchil pallidus Miller, 1901
Tragulus kanchil pidonis Chasen, 1940
Moschiola kathygre Groves & Meijaard, 2005
Tragulus kanchil rubeus Miller, 1903
Tragulus kanchil siantanicus Chasen & Kloss, 1928 2005 Moschiola kathygre Groves & Meijaard. Kumbalgamuwa,
Tragulus kanchil subrufus Miller, 1903 7° 06' N, 80° 51' E, 747 m, Kandy district, Central
Province, Sri Lanka.
Tragulus williamsoni Kloss, 1916
Color a much warmer, more ochery brown; spots on
Very much larger than any other member of this
the stripes yellowed, not white; at least two tolerably
species-group; color fairly similar to T. k. affinis.
complete longitudinal stripes along the flanks, with
Known securely only by the type specimen, from
an elongated spots row between them and two spot
Meh Lem, Muang Pre, Song Forest, far N Thailand,
rows above them; upper stripe curving around on the
18° 25' N, 100° 23' E.
-1— shoulder, becoming continuous with the anterior
0—
349-47558_ch01_1P.indd 58 5/31/11 7:05 PM

transverse stripes; two bold stripes over the haunch,

and a third one farther back under the tail; haunch Hyemoschus aquaticus (Ogilby, 1841)
more densely spotted; crown and nose less darkened; Although subspecies have been described (Hyemoschus
on the underside, a sharp differentiation between a aquaticus aquaticus from Sierra Leone; Hyemoschus
white median strip and the pale ochery of the rest of aquaticus batesi Lydekker, 1906, from Efulen, Camer-
the underside; lower halves of the hindlimbs darker. oon; Hyemoschus aquaticus cottoni Lydekker, 1906, from
Size small; hindlegs relatively short; hindfoot length the Ituri Forest), the small number of available samples
<115 mm. Rostrum breadth narrower, relative to the suggest little in the way of craniometric differentia-
skull length. Zygomatic breadth <51 mm; occipital tion. A single skull from Okuni, in SE Nigeria, has un-
height 25.1–29.2 mm; width across the canine alveoli usually short nasals that widen anteriorly. The nasals
<15; braincase breadth <33 mm. of the westernmost (Sierra Leone, Gambia, Senegal)
Wet zone of Sri Lanka, into the highlands, at least and easternmost (Uele, Ituri Forest) samples are nar-
to the region of Kandy. row, compared with those of intervening localities.
An undescribed species from Nuwara Elieya, in
the cloud forest at over 2000 m, is known from a sin-
MOSCHIDAE GRAY 1821
gle skull, but other specimens are in the process of
study. At any rate, the diversity of mouse deer in Sri As in the Tragulidae, the lacrimal orifice is single,
Lanka, associated with the climatic zones, duplicates elongated, and situated just inside the rim of the or-
that of many other vertebrates. bit, but on its medial border there is a “small protu-
berance” resembling either an incipient or a former
Hyemoschus Gray, 1845
state of division of the orifice (Leinders & Heintz,
African mouse deer, water-chevrotains. 1980). The anterior sagittal gully of the metatarsus is
closed by a bridge distally, as in the Cervidae. The po-
larity of the character states distinguishing the Mos-
Table 10 Measurements for Hyemoschus samples chidae from other (non-tragulid) ruminants is ques-
tionable ( Janis & Scott, 1987).
Gt l Nas l Nas br post Nas br ant A gall bladder is present.
A detailed, morphology-based cladogram placed
W Africa Moschus as sister to the clade containing the Giraffi-
Mean 143.618 45.367 19.350 10.850
dae and the Bovidae and their fossil relatives, not
N 6 6 6 6
Std dev 3.5217 3.9073 2.8219 1.3751 with the Cervidae, as had been previously “tradi-
Min 139.1 39.0 16.4 9.5 tional” (A. W. Gentry & Hooker, 1988). Hassanin &
Max 148.0 49.3 23.9 13.0 Douzery (2003), using both mitochondrial and mo-
Ghana lecular sequences, placed the Moschidae fi rmly as sis-
Mean 144.000 — 25.500 — ter to the Bovidae, with very strong support.
N 2 — 2 —
Min 137.7 — 22.2 — Moschus Linnaeus, 1758
Max 150.3 — 28.8 —
Okuni musk deer
Mean 136.800 40.900 23.300 15.900 Our arrangement follows Groves et al. (1995), but first
N 1 1 1 1 we must acknowledge a competing classification.
Batouri Sokolov & Prikhod’ko (1997) lumped all musk
Mean 143.833 45.267 22.700 11.933 deer into a single species, M. moschiferus, dividing it
N 3 3 3 3
into two groups, one with four subspecies and one
Std dev .8622 3.4962 .7211 .5859
Min 142.9 43.1 22.1 11.5 with three, as follows:
Max 144.6 49.3 23.5 12.6
1. sibirica group
Ituri/Uele
Siberian Moschus moschiferus moschiferus—the
Mean 143.567 46.500 19.933 11.400
N 3 1 3 2 largest skull, decreasing in size from W to E
Std dev 2.8885 — 4.8087 — Verkhoyansk (E Siberia) M. m. arcticus—
Min 141.8 — 15.3 11.3 smaller skull, especially shorter face, narrow
Max 146.9 — 24.9 11.5 interorbital distance, and short diastema
—-1
—0
349-47558_ch01_1P.indd 59 5/31/11 7:05 PM

60 Artiodactyla
Far Eastern M. m. turowi— skull still smaller, lumped taxonomy, stating, for example, that M. ber-
facial part shorter, toothrows shorter, short ezovskii is a color morph of what they referred to
mandible as Moschus moschiferus chrysogaster, the distinction
Sakhalin M. m. sachalinensis—intermediate between them being often due to “mistakes in age-
between Siberia and the Far East, with a long ing.” They did, however, now accept that Korean
upper toothrow, relatively narrow braincase musk deer are closer to those from the Far East and
and interorbital distance, and short mandible Sakhalin.
2. himalaica group
Korean M. m. parvipes— skull smallest, short Moschus moschiferus Linnaeus, 1758
face, narrow interorbital distance Pelage softer, less quilly, than in other species. Indi-
Chinese M. m. chrysogaster—braincase shorter vidual hairs dark, with a white subterminal band.
Himalayan M. m. leucogaster— supraorbital Underside lighter, grayish brown. Neck brownish; head
ridges well developed, the edges protruding more grayish, sometimes either lighter or darker than
beyond the line of the frontals the body. Ears dark brown or black, lighter at the base.
A pair of narrow creamy or white stripes (made up of
While, in many respects, the authors’ work is very
white-tipped hairs) from the chin down the neck to the
careful— studying the inheritance of color features,
chest. Light spots variably visible, prominent in the
and reviewing the growth of the skull and testing dif-
young. Lacrimal length shorter than its height; brain-
ferent regions for significant differences—this revision
case elongated; orbits tubular.
is very clearly overlumped as to species, but apparently
oversplit as to subspecies. For example, inspection of
Moschus moschiferus moschiferus
their table 2 shows that the skulls of the Verkhoyansk
Linnaeus, 1758
sample are statistically different from those from the
three subregions of the main Siberian sample, but the 1758 Moschus moschiferus Linnaeus. “Tataria versus Chinam”;
differences are, in very few cases, even as much as Harper (1940) restricted this to the Russian Altai.
one standard deviation apart. The same is true for 1779 Moschus sibiricus Pallas. Above Abakan, by one of the
comparisons between the Far Eastern and Sakhalin upper tributaries of the Yenisey River.
samples. On the other hand, the differences between 1929 Moschus moschiferus arcticus Flerov. Mt. Toulaja, 460 km
Siberia, on the one hand, and the Far East plus Sakha- N of Verkhoyansk.
lin, on the other, do seem substantial, the latter being
Overall color predominantly dull brown. Skull length
very much smaller: the mean greatest skull length for
>145 mm, usually >150 mm.
the four Siberian populations is 152.4–154.9 mm (the
E Altai and Yenisey, Mongolia; NE Nei Mongol;
standard deviation in the largest sample [65 individu-
NW Heilongjiang.
als from E Sayan] being 3.52), whereas that for the Far
East is 147.3 mm, and that for Sakhalin is 149.8 mm.
Moschus moschiferus parvipes Hollister, 1911
The sample sizes for the himalaica group are admit-
tedly much smaller, with only those taken from Sheng 1911 Moschus moschiferus parvipes Hollister. Mok-po, Korea.
(1992) being numerous; the measurements for the Ko- 1945 Moschus moschiferus turowi Tsalkin. Sikhote-Alin,
rean sample look very much like those from the Far Russian Far East.
East, and it is not clear why they are placed in the S
Darker than nominotypical M. m. moschiferus. Skull
group rather than in the N group (it does not appear
length 141–151 mm; bizygomatic breadth 60– 68.6 mm;
to fit at least one aspect of the defi nition of the S group;
lacrimal height 20–21.9 mm; lacrimal length 12.8–
namely, that the facial part of the skull is greater than
14 mm. Russian Far East, North Korea, S Heilongjing,
50% of the greatest skull length).
E Jilin, and E Liaoning
The same authors published a second paper in
the following year, dealing with chromosomes and
Moschus moschiferus sachalinensis Flerov, 1929
pelage (Sokolov & Prikhod’ko, 1998). Karyotypes
seem to be the same in all musk deer. The authors 1929 Moschus moschiferus sachalinensis Flerov. Sakhalin.
carefully described the pelage of the Siberian musk
Color dark, as in M. m. parvipes. Skull length 147–
deer, and those few skins from China and the Hima-
151 mm; bizygomatic breadth 67.5–70.3 mm. Lacri-
layas available to them, but they again produced a
-1— mal height 19.5–20.3 mm; lacrimal length 14–15 mm.
0—
349-47558_ch01_1P.indd 60 5/31/11 7:05 PM

Skull length 153–160 mm. Brownish yellow, weakly

Moschus chrysogaster Hodgson, 1839 striated. Head gray-brown. Ears brown, rims and in-
The largest species; skull length >150 mm, averaging side grayish white; poorly expressed grayish eye-ring.
155.1 mm. Overall color striated yellowish gray; paler Legs dark. Rump dark. Bases of the dorsal hairs pure
on the flanks; underside reddish creamy gray. Indi- white. No neck-stripe; throat all dark. Interramal re-
vidual hairs brown, with a red-yellow subterminal gion grayish white; underside, from the chest to the
band. Ears light brown, tipped with yellow, gray in- groin, grayish white. Juveniles spotted. Lacrimal length
side. Throat with a single broad, ill-defi ned, creamy longer than its height.
longitudinal band. Legs lighter than the body on the Nepal, Bhutan, and Sikkim; 2500– 4000 m.
lower segments, grayish yellow, somewhat darkened On this species, including Hodgson’s documenta-
down the front surfaces. Rump paler than the back, tion of it, see Grubb (1982b), who, at that time, treated
yellowish. Young animals spotted. Faint dorsal stripe. it as a subspecies of M. chrysogaster.
Hair bases long, milky gray or brown. Winter hairs There is possibly a separate subspecies (or species?)
34– 42 mm on the withers, 55– 64 mm on the rump. from Zhangmu (28°N, 87° E), in the forest zone of
Lacrimal length much longer than its height. Limbs S Tibet: it is dark, without hair-banding, but the bases
elongated; metacarpal length 109–118 mm; metatarsal of the hairs are white; the ears, the limbs, and the neck
length 128–138 mm. are all dark; the posterior aspect of the rump is orange-
On this species, including Hodgson’s own docu- white; and the nasals are anteriorly expanded.
mentation, see Grubb (1982b). From the Khumjung Glacier district in Nepal come
skins related to the species, but differing slightly: the
Moschus chrysogaster chrysogaster Hodgson, neck is somewhat paler; the throat is pale, with a
1839 poorly marked stripe on either side; the chin and the
interramal region are creamy white; the ears are gray
1839 Moschus chrysogaster Hodgson. Probably from the
basally, and black terminally; the buttocks are yellow-
Tibetan plateau, N of the Himalayas.
brown; the legs are mostly black; the chest is black;
Skull length 144–160 mm. More reddish tones; red- and the belly is grayer. In Groves et al.’s (1995) multi-
gold underside, inner surfaces of the limbs, and mid- variate analysis, the combined Zhangmu/Khumjung
line of the throat; orange eye-ring; Throat with white sample separates from the single available specimen
stripes; interramal region white. Limbs dark brown, of M. leucogaster.
becoming light brown on the lower segments; ears There is a further unnamed form, related to this
dark brown, orange-gold internally, at the base, and species, known from Kulu in NW Uttar Pradesh,
along the rims. Black patch on the buttocks. Rostrum at 3000 m or more, and from part of Nepal. This,
less than half the length of the skull. which we call the “pepper-and-salt” form, is brown
Alpine zone of Nepal, Sikkim, Bhutan, and S Ti- or red-agouti, sometimes forming a saddle, often
bet; 2800– 4000 m. with a light spot; throat light, this color often in a
diff use median band; lower limb segments lighter
Moschus chrysogaster sifanicus Buechner, 1891 than the body, because of the white speckling; but-
tocks and tail paler; ears dark gray or brown, often
1891 Moschus sifanicus Buechner.
with a whitish border, and white inside. Hair bases
Size averaging slightly larger; skull length 150–165 mm. yellow.
Throat with a broad creamy band. Limbs lighter; ears
with yellow tips. Rostrum length more than half the Moschus cupreus Grubb, 1982
skull length.
1982 Moschus leucogaster cupreus Grubb. Kashmir, at over
S Gansu, W Sichuan, Qinghai, S Ningxia, SE Ti-
3000 m.
bet, N Yunnan; 3500– 4800 m.
Skull length 150–155 mm. Gray-brown, often vaguely
Moschus leucogaster Hodgson, 1839 spotted, with a conspicuous, coppery-reddish, unspeck-
led dorsal saddle; rump very dark gray; underside light
1839 Moschus leucogaster Hodgson. Probably from the
gray; throat white; lower segments of the limbs whit-
Himalayan slopes of Nepal.
ish. Ears dark brown, white at the base, with frosted
1839 Moschus saturatus Hodgson. Probably from the
rims. Hairs with long white bases; hairs 33–38 mm —-1
Himalayan slopes of Nepal.
—0
349-47558_ch01_1P.indd 61 5/31/11 7:05 PM

62 Artiodactyla
long on the withers, 37–58 mm on the rump. Lacrimal principal components analysis, Li M. et al. (1999)
length longer than its height. found that skulls of this species fall far from those of
Kashmir. any other, and, using cytochrome b, they found con-
On this species, see Grubb (1982b), who, at that time, sistent differences from other species.
treated it as a subspecies of M. chrysogaster. In Groves et SW Anhui; below 500 m.
al.’s (1995) discriminant analysis, this species is quite
separate from all the others, but it is somewhat closer Moschus berezovskii Flerov, 1929
to M. chrysogaster than to the other taxa. Hair bases short, gray-white; throat with three wide
longitudinal stripes, white to orange in color, the lat-
Moschus fuscus Li Z-x., 1981 eral pair running from the jaw angles to the brisket,
the central one shorter, and all three sometimes bro-
1981 Moschus fuscus Li. Babo, Gongshan, 3500 m.
ken; haunch yellower; rump nearly black; limbs dark
Blackish brown all over; much darker than any other down the front surface. Underside yellow. Ear orange
species, with no neck or throat markings, but very at the base, black at the tip, white inside. Length of
occasionally traces of yellowish spotting on the body. the hairs 73– 87 mm on the withers, 45– 68 mm on the
Neck sometimes lighter than the body; throat often rump (varying according to season). Limbs relatively
with two incomplete yellow “collars.” Underside dark. short; metacarpal length 73– 87 mm; metatarsal length
Rump with ochery tones, but the buttocks black. Ju- 103–118 mm (hence the hindlimbs relatively longer
venile hairs agouti-banded; adult hairs solid-colored, than in most other taxa). Lacrimal length shorter than
except for very short white bases. Hairs long, 32–46 mm its height.
on the withers, 51– 63 mm on the rump. Skull length
135–150 mm; nasals <46 mm long. Muzzle short, less Moschus berezovskii berezovskii Flerov, 1929
than half the length of the skull; lacrimal length longer
1929 Moschus berezovskii Flerov.
than its height. Limbs more elongated than in M. ber-
ezovskii, despite the smaller skull; metacarpal length Skull length 142–153 mm. Color very dark, olive-brown,
88–101 mm, metatarsal length 126–135 mm, with the with orange tones.
hindlimbs even more elongated than in M. berezovskii. Sichuan, S parts of Shaanxi, Gansu, and Ningxia,
Hoof elongated, 27–29 mm in one specimen (maxi- SE Qinghai, and SE Tibet; between 500 and 2500 m.
mum in other species, 24 mm).
NW Yunnan (Gongshan, 26° 30' N, 98° 50' E), SE Moschus berezovskii caobangis Dao, 1969
Tibet (Zayu, 28° 25' N, 97° 06' E), and northernmost
1969 Moschus moschiferus caobangis Dao. Caobang Province,
Burma (Dchpu L’kha; Adung-Seingku); 2800– 4200 m.
Vietnam; S Yunnan (Mile and Langcang) and Guangxi;
There is a possibility that this species is the one
50– 400 m.
that has been found at over 4000 m, in the Khumbu
region of Mt. Everest and Tserping in Nepal, Sikkim, The smallest musk deer; skull length 120–135 mm.
and Bhutan. Both the adults and the juveniles resem- Color very light; limbs dark only from the knee and
ble M. fuscus from China and Burma, except that the the hocks downward; buttocks and ear tips still black.
Khumbu specimens have a slightly lighter underside Hair shorter, 21–30 mm on the withers, 38– 47 mm on
and interramal region. the rump.
We have not seen specimens of the following two
Moschus anhuiensis Wang et al., 1982 taxa:
1982 Moschus moschiferus anhuiensis Wang et al. Moschus berezovskii bijiangensis Wang & Li, 1993.
NW Yunnan to Zayu, SE Tibet. Described
A small species; skull length 141–149 mm (mean
as being paler than nominotypical M. b.
143.8 mm). Gray-brown in color, with dense (but not
berezovskii, fulvus-brown, with a grayish white
well-marked) light spots on the body. Neck-stripe
ventral surface, the toothrow averaging
present. Lacrimal length shorter than its height. Ears
slightly shorter.
darker than the body. Hairs long, 38 mm on the with-
Moschus berezovskii yunguiensis Wang & Li, 1993.
ers, 54 mm on the rump.
Yunnan-Guizhou plateau margins, E to
First proposed as a subspecies of M. moschiferus,
Ichang. Described as being somewhat smaller
-1— this taxon was initially transferred to M. berezovskii
than nominotypical M. b. berezovskii.
0— (Groves & Feng, 1986), then given species rank. Using
349-47558_ch01_1P.indd 62 5/31/11 7:05 PM

8
Antilocapridae
The Antilocapridae: last survivors of a once diverse The position of the family Antilocapridae, whether
family. sister to either the Cervidae or the Bovidae, or to a
clade containing both, is still not quite clear. The com-
bined mitochondrial and nuclear results of Hassanin
ANTILOCAPRIDAE OWEN 1841
& Douzery (2003) suggest that it branched off at about
The lacrimal duct opens into two orifices; the infe- the same time as the Giraffidae, or slightly earlier; in
rior one, as in the Cervidae, is situated on the orbital other words, it could even be the earliest branch from
rim, but the superior one is inside the rim (Leinders & the basic ruminant stem.
Heintz, 1980). As in the Cervidae, the anterior sagit-
tal gully of the metatarsus is closed by a distal bridge.
Antilocapra Ord, 1818
Antilocapra americana (Ord, 1815)
The hindfoot articular structure has converged on the
O’Gara (1990) has given a detailed account of this spe-
bovid type, resulting in a marked “pogostick effect”
cies, including whether it contains significant geo-
(Leinders, 1979).
graphic variation (which he tended to doubt).
A gall bladder is present.
On pronghorns, and particularly on the nature of
their appendages, see O’Gara (1990).
—-1
—0
349-47558_ch01_1P.indd 63 5/31/11 7:05 PM

A remarkable morphocline runs from the S spe-
9
cies (such as G. giraffa), via G. tippelskirchi, to the N
species (such as G. camelopardalis), relating to the de-
velopment of the medial ossicone, which gets larger
as one goes N, and is associated with the greater de-
velopment of the hyperossification of the dorsal skull
roof.
Our revisionary work is outlined below, followed
by our taxonomic arrangement.
Skull Comparisons of Males

Giraffidae We compared the three named taxa of N Kenya,
Uganda, and NE DRC (G. reticulata, G. rothschildi,
G. congoensis) by discriminant analysis, using just
four variables. We found that there are overlaps,
but, on the whole G. reticulata separates from the
other two, which do not separate well from each
other. The differences are based mainly on size, es-
Giraffidae. PG’s use of giraffe cranial variation to il- pecially the orbital rim to frontal distance (repre-
lustrate progressive clines. Working out the complex senting the height of the frontal horn), followed by
geographic variation of modern giraffes: skin pattern, the greatest skull length, then (to a much lesser de-
and its relation to African biogeography. The okapi: gree) the mastoid breadth and the condylobasal
early taxonomic misunderstandings, although it is length.
still enigmatic after all this time. The three C and W taxa—G. congoensis, G. roths-
childi, and the C African sample ascribed to G. peralta—
separate completely. Again, the main discriminator is
GIRAFFIDAE GRAY 1821 the orbital rim to frontal distance.
Usually there is one single, very small lacrimal orifice We then ran a discriminant analysis on the S
inside the rim of the orbit, although there may be no groups: G. tippelskirchi, G. thornicrofti, G. giraffa, and
orifice at all (Leinders & Heintz, 1980). G. angolensis. Three specimens of G. giraffa were in-
Molecular data (see, for example, Hassanin & Dou- termixed with the G. angolensis sample, and one was
zery, 2003) tend to place the Giraffidae as sister to the within the G. tippelskirchi dispersion. We found that
Cervidae plus the Moschidae plus the Bovidae, the po- the single available specimen of G. thornicrofti is out-
sition of the Antilocapridae being less secure. side the range of any of the other three. The differ-
ence in the fi rst case relies on small contrasts between
Giraffa Brisson, 1762 a large orbital rim to frontal distance, on the one hand,
This was one of the generic names which A. Gentry (1994) and, on the other, small preorbital length, palate
requested to be conserved, even though other names length, and biorbital breadth. The specimen of G. thor-
from Brisson (1762) were to be rejected. nicrofti has a low orbital rim to frontal distance, and a
low biorbital breadth.
It was shown by Solounias (1999) that to say that the Univariate statistics for the males are given in
giraffe has seven cervical vertebrae is to oversimplify. tables 11 and 12. In size, G. reticulata is noticeably
Essentially, there is an extra cervical vertebra, but smaller than the other species, while G. cameloparda-
what was C8 became thoracicized (the morphology lis and G. giraffa are noticeably larger. The differ-
of the vertebra in question shows this). This has to do ences are not as substantial in the condylobasal
with the position of the neck on the thorax: it enters length as they are for the greatest length, denoting a
the thorax from a superior direction above the scapu- greater development of the occipital crest in larger
lae, instead of from an anterior direction in front of species. G. giraffa has a noticeably wide palate; G. re-
them, and the glenohumeral joint protrudes anterior ticulata is very narrow across the zygomatic arches,
to the thorax, making it possible to breathe while the and also across the orbits; G. thornicrofti also has a
-1— neck is angled downward between the forelimbs for very small biorbital breadth (n.b., only one speci-
0— drinking. men). The N giraffes all have a larger orbital margin
349-47558_ch01_1P.indd 64 5/31/11 7:05 PM

g iraf f idae 65
Table 11 Measurements for giraffe species: males
Orb rim
Gt l Cb l Pal l Teeth Pal br Zyg br Biorb fr
reticulata
Mean 634.78 590.33 340.00 142.73 156.88 215.62 269.33 193.50
N 9 12 9 11 8 8 15 10
Std dev 20.687 20.366 15.819 3.379 12.800 3.662 15.089 18.858
Min 600 566 311 136 148 210 242 161
Max 662 640 357 147 188 222 298 216
camelopardalis
Mean 703.29 642.57 380.14 151.21 165.00 256.10 313.00 263.29
N 7 7 7 7 7 5 6 7
Std dev 24.669 25.376 21.098 3.740 7.095 8.721 15.518 17.802
Min 678 605 350 146 153 248 294 250
Max 746 675 410 155 173 269 339 289
antiquorum
Mean 676.73 634.78 365.00 148.33 152.50 235.92 289.00 228.36
N 11 9 10 12 2 12 13 11
Std dev 26.811 22.225 15.026 5.245 .707 13.153 22.884 20.101
Min 636 605 344 142 152 215 250 198
Max 727 673 391 158 153 256 322 255
peralta
Mean 668.50 612.50 349.25 152.75 153.00 246.33 293.50 193.25
N 4 4 4 4 1 3 4 4
Std dev 15.927 10.878 15.435 5.679 — 35.726 13.077 30.544
Min 655 601 334 145 153 220 283 166
Max 690 627 364 157 153 287 311 224
tippelskirchi
Mean 686.22 637.00 372.73 150.09 157.55 229.77 290.21 185.11
N 18 14 15 17 11 15 19 19
Std dev 38.374 23.085 20.600 6.949 9.606 10.479 21.168 25.486
Min 625 598 336 139 144 212 254 132
Max 752 683 407 162 172 241 331 237
thornicrofti
Mean 667.00 634.00 380.00 155.00 150.00 — 275.00 150.00
N 1 1 1 1 1 — 1 1
giraffa
Mean 698.00 642.75 384.00 152.80 169.00 234.50 293.80 171.33
N 4 4 3 5 3 2 5 3
Std dev 20.543 16.317 6.557 2.775 2.646 13.435 15.563 19.502
Min 670 620 378 150 167 225 276 149
Max 718 656 391 157 172 244 312 185
angolensis
Mean 687.89 639.57 383.88 150.94 155.69 236.71 291.56 164.00
N 9 7 8 9 8 7 9 9
Std dev 27.479 26.676 14.116 6.635 12.056 7.653 17.249 14.950
Min 647 615 358 142 132 225 269 147
Max 737 690 401 162 170 246 320 194
to frontal distance, indicating the development of a a sexual size difference is G. peralta, in which the
large frontal horn; this is far larger in G. cameloparda- females are much the largest of any giraffe species.
lis than in any of the other species, even other N The dramatically large size of the frontal horn (in-
species. dicated by the orbital rim to frontal distance) of the G.
The small size and lack of sexual size difference in camelopardalis males is followed successively by some- —-1
G. reticulata is noticeable; the other species lacking what smaller sizes in G. antiquorum, G. peralta plus G. —0
349-47558_ch01_1P.indd 65 5/31/11 7:05 PM

66 Artiodactyla
Table 12 Measurements for giraffe species: females
Orb rim
Gt l Cb l Pal l Teeth Pal br Zyg br Biorb fr
reticulata
Mean 623.40 582.67 344.40 141.80 148.40 211.00 247.17 144.80
N 5 6 5 5 5 3 6 5
Std dev 26.773 26.319 11.283 9.338 3.847 6.928 18.137 25.791
Min 590 538 334 130 145 207 232 126
Max 659 605 361 152 155 219 283 190
camelopardalis
Mean 616.67 589.50 354.00 142.67 159.00 200.00 263.00 142.00
N 3 2 3 3 2 1 3 3
Std dev 19.035 — 9.165 2.082 — — 20.809 7.810
Min 597 584 344 141 155 — 239 133
Max 635 595 362 145 163 — 276 147
antiquorum
Mean 619.50 576.00 340.00 148.00 148.00 229.00 260.50 142.50
N 2 2 2 2 1 1 2 2
Min 606 566 328 145 — — 255 141
Max 633 586 352 151 — — 266 144
peralta
Mean 607.00 574.50 338.50 143.50 — 206.80 256.25 137.25
N 4 2 2 6 — 5 4 4
Std dev 12.910 6.364 7.778 6.535 — 14.584 25.303 10.782
Min 592 570 333 133 — 186 229 123
Max 622 579 344 153 — 226 290 147
tippelskirchi
Mean 666.50 640.67 369.67 145.75 151.67 — 270.00 182.67
N 4 3 3 4 3 — 3 3
Std dev 53.681 38.031 21.385 5.058 8.083 — 13.892 11.590
Min 602 598 345 140 147 — 254 172
Max 716 671 383 152 161 — 279 195
thornicrofti
Mean 631.00 588.20 356.60 139.17 155.88 219.10 259.33 134.00
N 6 5 5 6 4 5 6 6
Std dev 12.458 19.601 9.965 4.167 1.436 7.603 11.325 7.321
Min 615 565 342 135 154 212 239 121
Max 650 610 368 146 157 228 268 142
giraffa
Mean 620.00 610.00 — 149.00 — 216.50 261.50 —
N 1 1 — 2 — 2 2 —
Min — — — 148 — 216 255 —
Max — — — 150 — 217 268 —
angolensis
Mean 625.00 588.00 352.00 148.20 163.00 223.60 266.00 134.00
N 5 5 4 5 4 5 5 5
Std dev 8.515 14.612 12.193 4.324 3.367 5.550 4.899 3.391
Min 619 572 342 143 159 217 260 130
Max 640 609 368 155 167 231 272 138
reticulata, G. tippelskirchi, and, finally, the S species

(n.b., very small female samples in these latter). The External Characters
females of G. peralta have rather large frontal horns, We obtained a large number of photos of wild gi-
-1— like the males, mirroring the general lack of sexual di- raffes, as well as some photos of captive giraffes of
0— morphism in this species. known origin: from the literature, from our own
349-47558_ch01_1P.indd 66 5/31/11 7:05 PM

g iraf f idae 67
photographs, and from the massive collection of Finally, we measured the length of the head and
Mr. Velizar Simeonovski, for whose generosity we are its height, the latter measurement extending from
extremely grateful. We compared giraffes from dif- the forehead (including the median horn, if present)
ferent areas for a number of features: to the jaw angle, thus expressing height as a percent-
age of length.
ground color
These measurements, of course, could be taken
color of blotches on the neck and the body
only on photographs depicting giraffes standing di-
extent of spotting on the legs
rectly laterally to the camera. In addition, we were
color of the lower limb segments
careful to measure each individual only once; if dif-
pattern above the eye, and along the dorsum nasi
ferent photographs appeared to depict the same indi-
pattern below the eye, and along the lateral
vidual, all photos but one were excluded.
margin of the lower jaw
In an attempt to formalize aspects of color-pattern DNA Revolution
variation, we took measurements from the photo- The study of mtDNA and microsatellites in giraffes
graphs, as follows: by Brown et al. (2007) caused quite a stir in the field,
because the authors proposed that there are a num-
1. We drew a line between the elbow and the
ber of species of giraffes, not merely one, as had been
stifle, extending it forward to the anterior
assumed for so long. Six taxa were studied by them,
outline of the body and backward to the
with each represented by numerous samples obtained
posterior outline of the rump, and counted the
(noninvasively) from the wild: G. reticulata, G. roths-
number of spots anterior to the elbow (that is,
childi, G. peralta, G. tippelskirchi, G. giraffa, and G. an-
along the base of the foreleg), between the
golensis. Each formed a monophyletic clade, charac-
elbow and the stifle (i.e., along the side of
terized by numerous substitutions, and there were
the abdomen), and posterior to the stifle (along
subclades within the N and S groupings that were also
the base of the hindleg).
strongly distinct. Within G. tippelskirchi, there were, in
2. We drew a line across the base of the neck,
fact, somewhat distinctive clades E and W of the Rift
starting at the bump corresponding to the tip
Valley.
of the neural spine of the second rib-bearing
Their phylogeny was as follows:
vertebra (usually identified as T2, but the “true”
T1 according to Solounias, 1999) and extending (angolensis ((tippelskirchi, giraffa) (reticulata (peralta,
forward and downward at the widest section. rothschildi)))),
We counted the number of spots/blotches that
although the position of angolensis as sister to all the
were crossed by this line, and measured the line
other taxa may have been an artifact of the tree-
itself and the length of each blotch. Adding up
building method; more likely, it was sister just to the
the scores of the blotches, we expressed this
other two S taxa.
as a percentage of the line measurement, thus
The authors emphasized that there was no evi-
calculating the proportion of the pelage in this
dence that even neighboring populations of adjacent
region that was covered by the blotches, rather
taxa were exchanging genes, and they concluded
than the ground color.
that the six taxa were behaving more like species
3. We inspected the outline of each of the blotches
than like subspecies. The entire study found only
that were cut by this line (see item #2 above),
two rothschildi/reticulata hybrids and one tippelskirchi/
and counted the number of “bites” (emargina-
reticulata hybrid (meaning pelage types with the
tions) out of the edges of each, in an attempt to
“wrong” mtDNA).
document how stellate each blotch was.
Separation times, according to different models of
the substitution rate, were around a quarter to nearly
In the second and third cases, we chose the base of half a million years within the N and S groups, and
the neck as being the region where we could take half to over one and one-half million years between
these measurements with the greatest degree of ob- the two groups.
jectivity, and we then inspected the pattern over the An almost simultaneous study by Hassanin et al.
whole neck and body, in order to make sure that the (2007)—this time including not only wild giraffes but
neck-base pattern was fully representative of the pat- also zoo specimens of known origin— added further —-1
tern as a whole. populations, as well as a further taxon, G. antiquorum. —0
349-47558_ch01_1P.indd 67 5/31/11 7:05 PM

68 Artiodactyla
The cladogram again divided into N and S groups. possible difference is the extreme darkening with age
The two S subclades were (1) G. angolensis from the in the extant E African population, although we
Lisbon Zoo, and (2) G. tippelskirchi from the Basel know little about the Nubian aging process.
Zoo plus G. giraffa from the Thoiry Zoo and from the
wild. The two N clades were (1) reputed G. peralta Giraffa reticulata de Winton, 1899
from the Vincennes Zoo (the female founder was ul- reticulated giraffe
timately from Negry, N CAR) and from Waza and Very dark above the eye and along the dorsum nasi,
Boubandjida national parks, Cameroon, plus G. anti- with individual spots hardly (or not at all) visible; very
quorum from the Antwerp Zoo (ultimately from near large blotches below the eye. Spots 3 on the lower
Khartoum) and from Zakouma, Chad, and (2) G. per- flanks, 2–3 on the base of the foreleg, 4 on the base of
alta from Kouré, Niger, plus zoo specimens of G. re- the hindleg. Pattern essentially dominated by rectan-
ticulata and G. rothschildi. gular blotches, with the ground color reduced to a net-
Our morphological data strikingly corroborate work of fine lines; 86%–90% of the base of the neck
the DNA data and enable us to produce new covered by blotches, which never have emarginations;
synonymies. neck blotches tending to be in longitudinal rows.
Ground color white, blotches liver-red, darkening to
Northern Giraffes nearly black in old males. Spots on the legs usually
The median horn in the adult male is well expressed, reaching only to the knees, occasionally halfway down
so that the head height is usually greater than 60% of the shanks; below the spotted area, the legs white to
the head length. The borders of the blotches are ochery fawn.
entire, or with (at most) 1–2 extremely small emargi- Intermediacy between G. reticulata and neighboring
nations. There are always 3– 4 blotches on the base of species was recorded by Stott (1959) and Stott & Selsor
the neck. The limbs below the knees and the hocks (1981): on the S bank of the Tana River, 9 mi. SW of Ga-
are white to ochery fawn, with very little spotting. rissa, out of a group of four giraffes, one resembled G.
reticulata, and one G. tippelskirchi, while the other two
Giraffa camelopardalis (Linnaeus, 1758) appeared intermediate; on the other hand, to the N of
Nubian giraffe, Rothschild’s giraffe the river, all giraffes were G. reticulata. Intermediates
between G. reticulata and what the authors referred
1758 Cervus camelopardalis Linnaeus. Sennar; restricted by
to as G. rothschildi were seen on the Laikipia Plateau,
Harper (1940).
“along the Maralal–Rumuruti Road,” and between Ru-
1903 Giraffa camelopardalis rothschildi Lydekker. Guasin-
muruti to the S and the Loroghi Plateau to the N, as far
Gishu Plateau.
as the desert 24 km SE of the S tip of Lake Turkana, and
1904 Giraffa camelopardalis cottoni Lydekker. Koten Plain,
“east to the dry bed of the Barsaloi River.”
S Doborsa country, NE Uganda.
Spots above the eye and along the dorsum nasi tend- Giraffa antiquorum (Swainson, 1835)
ing to be fused; spots below the eye fairly large. Spots Kordofan giraffe
3 on the lower flanks (between the elbow and the
1835 Camelopardalis antiquorum Swainson. Bagger el Homer,
stifle), 3– 4 on the base of the foreleg anterior to the
Kordofan, about 10° N, 28° E (as fi xed by Harper, 1940).
elbow, and 4–5 on the base of the hindleg behind
1903 Giraffa camelopardalis congoensis Lydekker. Dungu, Uele
the stifle. Blotches covering 66%–77% of the base of
Valley, NE DRC.
neck, the rest being ground color. Ground color dirty
creamy; blotches liver-brown, with darkened centers Smudgily spotted, not very dark, above the eye; spot-
in the adult males, the centers tending to become black- ting below the eye, reaching to the mouth corner.
ish with age. Spots on the legs barely reaching the Blotches 3– 6 on the lower flanks, 3– 4 on the base of
knees and the hocks, or not at all. the foreleg, 6–7 on the base of the hindleg (corrobo-
Formerly from the latitude of Khartoum S through rating the description “broken up into a number of
E Sudan and W Ethiopia (W of the highlands); per- very small and irregular” spots [Lydekker & Blaine,
sisting in NW Kenya and E Uganda. 1914b]). Blotches covering 56%–72% of the base of the
As far as the evidence goes, there seems to be no neck; blotches entire, or with (at most) one very small
essential difference between the (presumably ex- emargination. Ground color off-white to pale yellow-
-1— tinct?) N populations and those that survive today in ish; spots medium to dark brown, paler on the body
0— Kenya and Uganda under the name G. rothschildi; a than on the neck. Spots on the legs reaching down to
349-47558_ch01_1P.indd 68 5/31/11 7:05 PM

g iraf f idae 69
the knee and the hock, sometimes slightly below; the foreleg, 5– 9 on the base of the hindleg (very
legs white below the spotted area. occasionally more than this number). Blotches 3– 4,
The Paris Zoo breeding group belongs to this spe- occasionally as many as 6, on the base of the neck;
cies; they were identified by Hassanin et al. (2007) as blotches covering 60%– 80% of the base of the neck;
G. peralta, but they correspond in appearance to G. borders of the blotches deeply emarginate, with up
antiquorum. Thus the distribution extends from Kordo- to 8 “bites” out of the larger blotches. Ground color
fan S to the N CAR and to Zakouma in Chad. Giraffes creamy to fawn; spots medium liver-brown, darken-
in the Garamba National Park, NE DRC, also belong ing with age. Spots on the legs indistinct, reaching
to this species. halfway down the shanks or to the fetlocks; lower leg
light fawn; fetlocks contrastingly white.
Giraffa peralta Thomas, 1908 This is the only S giraffe with noticeable sexual di-
West African giraffe morphism in the development of the median horn:
head height is 52%–58% of its length in the males, but
1908 Giraffa camelopardalis peralta Thomas. Lokojya,
only 40%–55% in the females (the female figure is sim-
Niger– Benue junction, Nigeria.
ilar to that of the N giraffes).
1971 Giraffa camelopardalis renatae Krumbiegel. Hinterland
Giraffes in Tsavo East National Park appeared to
of Lagos, and between Parakau and Moschi (°9° N, 3° E).
be much more variable than those in Tsavo West Na-
Sparsely spotted or smudgy gray, not very dark, above tional Park, Mikumi, Tarangire, or Ruaha. Those in
the eye; spotting below the eye not reaching forward Serengeti are also variable; at these two ends of the
to the mouth region. Blotches 4–5 on the lower flank, range they differ on average, especially in the num-
3–5 on the base of the foreleg, 5–9 on the base of the ber of deep “bites” in the neck spots—up to eight in
hindleg. Ground color predominant in the pattern; Tsavo East, but often only one, and that not very deep,
blotches covering 50%–71% of the base of the neck; in Serengeti. In other words, the Tsavo East giraffes
blotches entire (rounded, rather than blocklike), or are often the most “extreme” in the species. Recall that
with very small emarginations. Ground color white, Brown et al. (2007) found a deep division between hap-
gray-white, or pale yellow; spots gray to dark liver- lotypes of G. tippelskirchi from E and W of the Rift
brown, becoming black-brown with age. Spots on the Valley, so it may be that further study of the external
legs reaching (at most) down to the knee and the characters is required to test this.
hock, the remainder of the leg being white.
Formerly from West Africa, as far E as N Cameroon; Giraffa thornicrofti Lydekker, 1911
giraffes in Waza National Park belong to this species. Luangwa giraffe
Apart from the Cameroon populations, the only sur- Upper half of the face tending not to be very dark, the
viving population of G. peralta appears to be in Niger. spots often only partly fused; spotting below the eye
nearly (or quite) as far forward as the mouth angle.
Southern Giraffes Blotches 5–7 on the lower flanks, 4–5 on the base of
The median horn in the adult male is poorly ex- the foreleg, 7–9 on the base of the hindleg. Blotches
pressed or absent, so that the head height is usually 5– 8 on the base of the neck, covering 53%–77% of the
less than 60% of the head length. The blotches are neck; borders of the blotches often deeply emargin-
stellate, with up to 8 emarginations, which often reach ate, with 1–5 deep “bites.” Ground color creamy gray;
deeply into the blotch. The number of blotches on base spots medium to light brown on the neck, darker on
of the neck varies. The limbs below the knees and the the body. Legs spotted only to the hock and just
hocks tend to be darker, often spotted. above the knee; lower legs pale fawn or ochery; fet-
locks not contrasting.
Giraffa tippelskirchi Matschie 1898 Confi ned to the Luangwa Valley, E Zambia.
Maasai giraffe, vine-leaf giraffe This species is the only one whose mtDNA has
thus far not been studied.
1898 Giraffa schillingsi Matschie. Taveta, SE Kenya.
1898 Giraffa tippelskirchi Matschie. Lake Eyasi, NW
Giraffa giraffa (Boddaert, 1785)
Tanzania.
Cape giraffe
Entire upper face dark; spotting below the eye reach-
1785 Camelopardalis giraffa Boddaert. S of Warmbad, Orange
ing as far forward as the mouth corner or beyond. —-1
River (Rookmaaker, 1989).
Blotches 3–5 on the lower flanks, 3–5 on the base of —0
349-47558_ch01_1P.indd 69 5/31/11 7:05 PM

70 Artiodactyla
1842 Camelopardalis capensis Lesson. Gamma River, Great leg, 4– 6 on the base of the hindleg. Blotches 3– 4 on
Namaqualand, about 27° S, 18° E. the base of the neck, covering 61%– 80% of the neck;
1898 Giraffa infumata Noack. Barotseland, Zambia. borders of the blotches with 1– 6 emarginations, one
1904 Giraffa camelopardalis wardi Lydekker. N Transvaal. or two of which are very deep. Pattern with a “washed-
out” appearance; ground color gray-white; spots gray-
Upper face and dorsum nasi dark to the end of the
brown on the neck, darker on the body. Legs spotted,
snout; spotting below the eye reaching as far forward
reaching to about halfway between the knee and the
as the mouth. Blotches 4–5 on the lower flanks, 3– 6
hock; lower shanks pale to very pale fawn.
on the base of the forelegs, 6– 8 on the base of the
hindlegs. Blotches 3–5 on the base of the neck, cover- Okapia Lankester, 1901
ing 59%– 83% of the neck; borders of the blotches
okapi
with 1–13 emarginations, one or two of them being
The okapi does not share in the bizarre reshuffl ing of
very deep. Ground color off-white or fawny white;
the cervical/thoracic vertebral junction seen in the
spots medium to dark brown. Legs spotted, reaching
giraffe (Solounias, 1999).
all the way down, but fading toward the fetlocks, on
a pale fawn ground.
Okapia johnstoni (Sclater, 1901)
N South Africa, Zimbabwe, Botswana.
On the general biology of this species, see Gijzen
Giraffes in Kruger National Park are, on average,
(1959) and Lindsey et al. (1999).
more closely spotted than those in the Okavango,
There have been a few attempts to split up this
Botswana.
species, all of them based on early, inadequate infor-
mation. Nonetheless, Gijzen (1959) did draw attention
Giraffa angolensis Lydekker, 1903
to locality records S of the Grand Cuvette (the great
Angolan giraffe
bend of the Congo River), and specimens from these
Upper face dark to the end of the snout; partly fused
localities clearly need to be examined and compared
spots below the eye extending to the mouth. Blotches
with those from the region of the Ituri Forest and
3– 4 on the lower flanks, 3– 4 on the base of the fore-
westward, from which the species is well known.
-1—
0—
349-47558_ch01_1P.indd 70 5/31/11 7:05 PM

and the plethora of genera presently recognized in
10
South America, throw into stark relief the need for
objective criteria for the recognition of higher catego-
ries in taxonomy. This fairly long chapter goes through
the two subfamilies— each with two or more tribes,
and each of them with two or more genera— and con-
siders the number of species (and, where necessary,
subspecies) in each.
Groves & Grubb’s (1987) division of the family into
three subfamilies—the Hydropotinae, the Odocoilei-
nae and the Cervinae—was based on the lack of ant-
Cervidae lers in the first, and the way the lateral metacarpals
have been retained in the other two: the Odocoileinae
have kept the distal ends (telemetacarpal); and the Cer-
vinae, the proximal (plesiometacarpal). In plesiometa-
carpal deer, the postglenoid foramen is entirely within
the squamosal; in telemetacarpal deer, the petrous
forms its medial border (Bouvrain et al., 1989); behav-
CERVIDAE GOLDFÜSS, 1820 ioral features separate the two, as well (Cap et al., 2002).
Both of these latter studies also showed that Hydropotes,
The lacrimal duct almost invariably has two orifices antlerless though it is, is part of the odocoileine clade.
on the rim of the orbit, the inferior one being slightly Miyamoto et al. (1990), Li & Sheng (1998), Randi,
anterior to the superior one (Leinders & Heintz, 1980). Mucci, et al. (1998), Hassanin & Douzery (2003), Pitra
The distal end of the anterior sagittal gully of the et al. (2004), and Gilbert et al. (2006), using different
metatarsus is closed by a bridge. DNA sequences, confi rmed the basic division and
Leinders (1979) drew attention to differences in showed that Hydropotes is part of the telemetacarpal
the hindfeet between this family and the Bovidae: in clade.
the Cervidae, the sagittal crests on the distal articular A more recent study of male vocal behavior (Cap
surfaces of the metatarsus are less prominent and less et al., 2008) resulted in a cladogram almost exactly
proximally extended; there is a nonarticulatory sur- duplicating the molecular phylogeny: muntjak are as-
face at the posterior end of the proximal articular sociated with Old World deer; Capreolus and Hydro-
facet of the medial phalanx (“plateau postarticulaire”); potes are in a special clade (contra Cap et al.’s (2002)
and the proximal articular facet of the distal phalanx phylogeny, in which Hydropotes was the sister to other
is much less posteriorly extended and lacks the ante- members of the Capreolinae); and so on.
rior extensor process (“éminence pyramidale”). This, We use the amended nomenclature of Grubb
according to Leinders (1979), results in less flexibility (2000b).
of these joints; that is, there is less of a “pogostick ef-
fect” than in the Bovidae. Subfamily Capreolinae Brookes, 1828
The gall bladder is absent. The name Odocoileinae, used by Groves & Grubb (1987), is
On antlers, their nature, and how they differ from antedated by Capreolinae, and the name Neocervinae
the horns of the Bovidae and the pronghorns of the Carette, 1922, still frequently used for this group, is not
Antilocapridae, see Bubenik (1990). The Cervidae, available, because it is not based on a genus-group name
unlike the Bovidae, possess a number of clearly de- (Grubb, 2000b).
rived character states, quite apart from the kind of
cranial appendages ( Janis & Scott, 1987). Karyotype evolution (Abril, Sarria-Perea, et al., 2010)
The major outlines of the taxonomy of this family seems to make the Rangiferini paraphyletic, in that
have become much clearer lately because of molecu- an initial split separates the Mazama americana group
lar phylogenetic studies, but, at the same time, these from the rest of the subfamily. Evidently there has
have muddied previous assumptions about the inter- been some homoplasy; inspection of these authors’ fig-
relationships of some groups, particularly the genus ure 2 shows that the source of the homoplasy lies in a
Cervus and its close relatives (the Cervini). The consequence of two pericentric inversions that are known
to have considerable potential for parallel occurrence. —-1
troversies over how many genera exist in the Cervini, —0
349-47558_ch01_1P.indd 71 5/31/11 7:05 PM

72 Artiodactyla
by Pitra et al. (2004) was a red brocket (M. americana

Tribe Rangiferini Brookes, 1828 group), so, to that extent, their conclusions are entirely
consistent with those of other authors.
This name, not Odocoileinae, is the prior available name for
We have found that cranial characters strongly
the New World deer (Grubb, 2000b). Rangifer is likely to
support this division.
be the sister taxon to the other genera (Pitra et al., 2004).
We retain the old divisions for the moment, but,
The tribe is distinguished from the Capreolini and the
ultimately, it will almost certainly be necessary to ar-
Alceini by the derived condition of the vomer (com-
range the neotropical deer quite differently. For the
pletely dividing the choanae), by contact between the
red brockets, the name Mazama Rafi nesque, 1817, is
stylohyoid and the paroccipital, and by the widely
available, with the type species Mazama pita, which,
separated antler pedicels (Webb, 2000).
although it cannot be defi nitively allocated (see be-
Detailed information on the neotropical representa- low), certainly refers to a member of the M. americana
tives of this tribe will be found in Duarte & González group. The generic name Subulo Hamilton Smith,
(2010); their evolution, both in the fossil record and in 1827, has no type species, but the fi rst of the three spe-
their adaptive biology, is treated by Merino & Rossi cies standing in that genus was Cervus rufus, likewise
(2010). a member of the M. americana group and probably
Webb (2000) proposed the following morphologi- referring to the same animal as M. pita. The name
cal cladogram: Odocoileus Rafi nesque, 1832, postdates Mazama, and,
given the relatively short time span inferred for the
((Pudu (Rangifer (Hippocamelus))) (Mazama
group as a whole, this name will probably have to be
(Ozotoceros (Blastocerus, Odocoileus)))).
regarded as a ju nior synonym of Mazama (to the dis-
Pitra et al. (2004) proposed the following molecular may of North American mammalogists).
phylogeny: For the brown brockets, the earliest available names
are Passalites Gloger, 1841, and Nanelaphus Fitzinger,
(Rangifer (Blastocerus, Pudu) (Odocoileus, Mazama)).
1873. Both are based on Cervus nemorivagus F. Cuvier,
(The remaining two genera were not available.) None- 1817, and both have been unused for well over a cen-
theless, it now appears that the genus Mazama is poly- tury, certainly rating as nomina oblita. As, however,
phyletic. Duarte et al. (2008) found that mtDNA splits the brown brockets did not split very long ago from
Mazama into two clear groups that assort not with each the other (more purely) neotropical deer, the earliest
other but, respectively, with Odocoileus (the M. ameri- name for the entire group should probably be Hippo-
cana group) and with the other South American deer camelus Leuckart, 1816. Lydekker found this name
(the M. gouazoubira group). The same conclusions, on “highly objectionable” (either because of its mixed
larger samples, have been reached by S. González, derivation, the fi rst half Greek and the second half
Duarte, et al. (2010). In fact, the Mazama sample used Latin, or else because of its general meaninglessness,
Table 13 Skull characters for the two groups of South American deer
M. gouazoubira-Hippocamelus-Pudu-Blastoceros-
M. americana-Odocoileus group Ozotoceros group
Ethmoid vacuity Very large Restricted

Frontolacrimal suture length Much shorter than the frontal margin of the About equal to the length of the frontal margin
ethmoid vacuity of the ethmoid vacuity
Maxillolacrimal suture length Equal to or less than the maxillary margin of Much longer than the maxillary margin of the
the ethmoid vacuity ethmoid vacuity
Median tips of the nasals Much shorter than the lateral prongs Reach anterior to the lateral prongs
Lateral ridges of the diastema Turn in anterior to P2, then curve out very Deeply bowed inward at P2, then bowed
slowly and slightly laterally to the incisive outward, again laterally to the incisive
foramina foramina
I1 More broadened at the tip; I 2 is also Only slightly broadened at the tip, separated by
broadened, so it partly fi lls in the gap a V-shaped gap from the others
-1— Antler texture Roughened Ridged or fluted (exception—Ozotocerus)
0—
349-47558_ch01_1P.indd 72 5/31/11 7:05 PM

cerv idae 73
i.e., “horse-camel”); so that he could use it for as few There is a decline in size of the deer identified as
species as possible, he broke up what would otherwise O. virginianus down the E coast of North America,
have been a single genus into four! culminating in the diminutive deer of the Florida
Keys (currently known as Odocoileus virginianus cla-
Odocoileus Rafinesque, 1832 vium). A study of mtDNA of deer in Florida, including
white-tailed deer, black-tailed deer, and mule deer the Florida Keys deer, found three different haplo-
types, but these corresponded only poorly with the
Grubb (2000b) reviewed the use of the name americanus
described subspecies, including O. v. clavium (Ells-
Erxleben, 1777, in combination with generic names
worth et al., 1994).This brought up the question, fre-
Moschus (now = Mazama americana) and Cervus (now
quently posed in recent years, of whether putative
regarded as an unavailable senior synonym of Odocoileus
subspecies would be better defi ned as possessors of
virginianus).
unique mitochondrial haplotypes rather than, as tra-
Odocoileus is a very difficult genus taxonomically. ditionally defi ned, being based on morphological
In North America there are two species-groups, each characters. It has been argued that the function of the
currently being classed as a single species, but this is subspecies category is to delimit geograph ically re-
almost certainly overlumped. One problem is that stricted lineages. Against this, we would maintain,
white-tailed deer (O. virginianus—if this really is a fi rst, that of course mtDNA is inherited only matri-
single species, rather than a species complex) and lineally, and that the depicted lineages might be en-
mule deer (O. cf. hemionus) are known to interbreed tirely different if Y-chromosome DNA would be stud-
in West Texas. It was discovered 20 years ago that in ied (as in the West Texas hybridization study cited
this region, the two share a mtDNA restriction type above); and second, that, in any case, morphological
characteristic of white-tailed deer, suggesting that in- characters are themselves (broadly speaking) herita-
terbreeding had occurred between male mule deer ble, and that there is value in continuing to recognize
and female white-tailed deer, and that in the hybrid gene pools that are strongly divergent as a whole.
herds, the white-tailed phenotypic characteristics An analogous situation recurs in Venezuela, at the
have been lost by generations of backcrossing with S end of the distribution of white-tailed deer. Molina
mule deer (Carr et al., 1986). A subsequent study & Molinari (1999) recognized three distinct species in
(Cathey et al. 1998) confi rmed that F1 hybrids are rare Venezuela (and suggested that there was evidence for
and found that Y-chromosome DNA, in contrast with a fourth). Their three species are O. margaritae (Mar-
mtDNA, assorted more clearly along species lines. garita Island), O. lasiotis (Mérida, Andean highlands),
Some implications of this hybridization, which actu- and O. cariacou (from the rest of the range). In con-
ally occurs in both directions in different popula- trast, a study of mtDNA by Moscarella et al. (2003)
tions, have been discussed by Bradley et al. (2003). identified four clades, but these corresponded rather
Within what is commonly classified as Odocoileus poorly with the proposed species. It is clearly time for
hemionus, a recent study (Latch et al., 2009) found a new look, utilizing both methods and using the
that there are two distinct clades, corresponding to same samples for both. For the moment, we leave the
what are known as mule deer and black-tailed deer. taxonomy of Odocoileus strictly alone, recognizing
These had originally been described as separate spe- that we have neither the data nor the experience to
cies, O. hemionus and O. columbianus, respectively; deal with it; and we strongly urge North American
they are indeed quite distinct and easily recognizable taxonomists— as well as South American ones—to
phenotypically, but they had been united into one look closely at the genus. The neotropical forms are
species early in the 20th century on the grounds that described by Gallina et al. (2010).
they have been known to interbreed. The DNA fi nd-
ing seems to be good evidence that the two should be Blastocerus Wagner, 1844
re-split; there were, of course, some instances of the
Grubb (2000b) argued that this name is available (antedating
“wrong” DNA clade in a given population on the bor-
the same name by Gray, 1850), and he designated Cervus
ders of the ranges of the two, but this should not be
paludosus Desmarest, 1822 (= Cervus dichotomus Illiger,
taken as evidence that they are one and the same.
1811), as the type.
Within O. hemionus proper, there were seven distinct
mtDNA haplotypes, but these cross-cut described Skull characters are typical for those of the other
subspecies (whose phenotypic distinctiveness needs neotropical endemics and the Mazama gouazoubira —-1
to be retested anyway). group (see below), with a few modifications. The —0
349-47558_ch01_1P.indd 73 5/31/11 7:05 PM

74 Arti odactyla
supraorbital sulcus is extremely long and broad, and 2002) more defi nitively divided the species into five
it contains four fairly large foramina. The unworn ant- subspecies, discussed below.
lers are heavily rugose.
Ozotocerus bezoarticus bezoarticus (Linnaeus, 1758)
Blastocerus dichotomus (Illiger, 1815) S. González et al. (2002) had no comparative material
marsh deer for this subspecies, which appears to be known from
For a full account of this species, see Piovezan et al. a single skull.
(2010).
A study using mtDNA of four populations in the La Ozotocerus bezoarticus celer Cabrera, 1943
Plata Basin (the La Plata Delta, Iberá, S Bolivia, and the Mean condylobasal length 214.8 mm (males), 212.4 mm
Pantanal of SW Brazil) found considerable differences (females).
among them (Márquez et al., 2006). Control region hap- According to S. González, Cosse, et al. (2010), this
lotypes were unique to each region: two cytochrome b is the subspecies found in Buenos Aires Province.
haplotypes were shared between the Pantanal and
Iberá samples, but S Bolivia and the La Plata Delta each Ozotocerus bezoarticus leucogaster (Goldfüss, 1817)
had a unique haplotype, that of the Delta being the Mean condylobasal length 215.0 mm (males), 208.1 mm
most divergent. The authors pointed to the somewhat (females); this is based on 9 males and 20 females, so
divergent biogeographic nature of the Paraná Delta, the apparent size difference between the sexes may be
and estimated the time of separation of this popula- real. In the discriminant analysis of S. González et al.
tion from the others at about 200,000 years ago, which (2002), O. b. leucogaster overlaps greatly with O. b. celer,
suggests the need for a taxonomic reappraisal. even in the analysis of the female skulls, which are
much better differentiated than those of the males.
Ozotoceros Ameghino, 1891 According to S. González et al. (2002), this is the
subspecies found in Mato Grosso and the Argentin-
The prior available name for this genus has frequently been
ian Chaco.
held to be Blastoceros Fitzinger, 1860, a virtual homonym
of Blastocerus Gray, 1850 (recte Wagner, 1844; see above).
Ozotocerus bezoarticus uruguayensis González
Fitzinger was aware of this name, and his version of it is,
et al., 2002
therefore, to be regarded as an unjustified emendation;
hence, it is unavailable as a name for the present genus Type locality: Uruguay, Rocha Department, Sierra de Los
(Grubb, 2000b). Ajos, 33° 50' S, 54° 01' W.
2n = 68, based on specimens from Uruguay and Para- Color light fawn-brown to tawny, bay, and dark cinna-
guay. The X chromosome is metacentric. mon. Antorbital pit unusually long and deep. Breadth
The skull characters are typically those of the neo- of the frontal 73 mm. Mean condylobasal length
tropical endemics (see above). The antlers are curi- 213.8 mm (males), 218.6 (females), but the sample sizes
ous, with little excrescences all over. are small, so the apparent larger size of the females
may, in this case, be illusory.
Ozotocerus bezoarticus (Linnaeus, 1758) E grasslands of Rocha Department.
pampas deer In S. González et al.’s (2002) discriminant analysis
For a full account of this species, see S. González, of the female skulls, the range of variation of O. b.
Cosse, et al. (2010). uruguayensis falls beyond those of others, but the ad-
The question of possible taxonomic variation mittedly small sample size dictates caution.
within this species, or even between related species,
has been much discussed. Using craniometrics, S. Ozotocerus bezoarticus arerunguaensis González
González et al. (1991) found that there are differences, et al., 2002.
at least on average, between samples ascribed to O. b.
Type locality: Uruguay, Salto Department, Arerunguá, El
celer (from the Pampas), a sample from Uruguay, and
Tapado, 31° 41' S, 56° 43' W.
samples ascribed to O. b. leucogaster and O. b. bezoarti-
cus. The authors ended on a cautionary note: “A wide Color from fawn-brown to tawny and cinnamon.
sample of specimens from the entire distribution area Antorbital pit small. Breadth of the frontal 58.5 mm.
-1— would reveal the true subspecifical systematics.” This seems to be the smallest subspecies; mean con-
0— Later, some of the same authors (S. González et al., dylobasal length 208.2 mm (males), 205.3 (females),
349-47558_ch01_1P.indd 74 5/31/11 7:05 PM

cerv idae 75
although the ranges of variation are very wide. In the

discriminant analysis of the female skulls (S. González Mazama Rafinesque, 1817
et al., 2002), the range of variation of O. b. arerun- brockets
guaensis has a small overlap with that of O. b. celer, but Mazama is an even more taxonomically complex ge-
it does not overlap with any other; interestingly, O. b. nus than Odocoileus, with far more species than the
arerunguaensis seems furthest from the other Uru- four or so that had been previously assigned to it (see,
guayan subspecies, O. b. uruguayensis. for example, Medellín et al., 1998, who identified a
NW grassland, near the type locality. further distinct species for Yucatán).
For the moment, we will retain both red and brown
Hippocamelus Leuckart, 1816 brockets in the genus Mazama, while anticipating that
The two species presently allocated to this genus are a totally different classification will have to be adopted
extremely different. in the future. Meanwhile, we would like to pay tribute
to the first reviser, J. A. Allen, whose 1915 monograph
Hippocamelus antisensis (d’Orbigny, 1834) was a model of thoroughness and has a remarkably
taruka modern aspect.
For a full account of this species, see Barrio (2010).
Second lower incisor hardly widened at all. Sharp Group Forming a Clade with Odocoileus
supraorbital sulci, containing three large foramina. (Red Brockets)
Lacrimal fossa uniquely wide. Posterior ends of the According to Duarte et al., 2008, this group divides
nasals (somewhat widened in all genera) remarkably into four subclades: (1) ten individuals of M. ameri-
expanded here, and then rather suddenly constricted cana, (2) Odocoileus, (3) one individual called “Mazama
anteriorly. Unworn antlers deeply fluted, with a long sp.”, and (4) nine individuals of M. americana, one of
anterior branch, somewhat over half the length of the M. sp., M. bororo, and M. nana.
posterior branch. Pelage coarse, grayish, and white Eisenberg (1989), Redford & Eisenberg (1992),
underneath, from the chin down the front to the chest; Eisenberg & Redford (1999), and Varela et al. (2010)
rest of the underside dark brown, but the entire inner gave the following measurements for members of
aspects of the legs, and the outer sides of the legs be- this group, which they mainly treated under the
low the carpus and tarsus, strikingly white; rump ex- species name Mazama americana: weight 34.9 kg, head
tensively white; tail white below. plus body 1154 mm, ear 103 mm, metatarsus 233 mm,
Along the Andes Mountains, extending from N tail 136 mm, height 642 mm (Brazil); total length
Peru through W Bolivia into N Argentina and north- 1200 mm, tail 126 mm, hindfoot 674 mm, ear 99 mm,
ernmost Chile, between 2000 and 3000 m in the S and weight 29 kg (presumably from the N distribution?);
as much as 3800–5000 m in the N part of the range total length 1209 mm, head plus body 1082 mm, tail
(Barrio, 2010). 127 mm, hindfoot 674 mm, ear 99 mm (Argentina);
total length 1200 mm, tail 150 mm, weight 28.9 kg
Hippocamelus bisulcus (Molina, 1782) (N = 11, Paraguay). These measurements therefore
huemul vary geograph ically, but all red brockets can gener-
For a full account of this species, see Vila et al. (2010). ally be distinguished from others by their large size
2n = 70. and, of course, their bright red-brown color, with a
Second lower incisor somewhat wider. Supraorbital white or creamy underside. The hindlegs are often
sulci shallow or flat, each containing just one large fo- very dark, almost black, below the heel to the hoof.
ramen. Lacrimal fossa fairly deep, but narrow. Poste- Only 14% (in a Suriname sample) have canines.
rior ends of the nasals not strongly expanded. Unworn For a full account of this species-group, see Varela
antlers fluted, but not to the extent seen in H. antisensis; et al. (2010).
antlers shorter than in H. antisensis, especially the pos- Duarte (1996, and elsewhere) has recognized three
terior branch. Pelage coarse, dark brown, lighter and species in the americana group: Mazama americana,
more grayish yellow in winter than in summer. Only M. nana, and M. bororo. Differentiating these three
the underside of the tail, the groin, and spots above the phenotypically is problematic, so they will simply
eyes and beside the mouth white. be treated according to locality in our analysis. It is
Andes of S Chile and Argentina, extending to the S clear that, even shorn of these more obviously ex-
limit of the mainland but not to Tierra del Fuego; traneous components, the species Mazama americana —-1
ranging from sea level to 2000 m (Vila et al., 2010). is taxonomically heterogeneous. Karyotypes vary —0
349-47558_ch01_1P.indd 75 5/31/11 7:05 PM

76 Artiodactyla
enormously; for example, Abril, Vogliotti et al. (2010) the body mass. Red body color, however, more ho-
have illustrated two living females that appear al- mogeneous; no dark area on the hindlimbs, except
most identical externally yet differ strongly in karyo- on a line from the posterior region of the ankle to
type: the one from Paraná has 2n = 52, the one from the fi nal third of the metatarsal. White half-moon
Rondônia has 2n = 42, and there is yet more chromo- mark at the base of the ear more marked than in M.
some polymorphism within Brazil (Varela et al., americana.
2010). On the other hand, there is very little protein In many respects, M. bororo resembles captive-bred
polymorphism (Garcia & de Oliveira, 1910). hybrids between M. americana and M. nana.
We lack a sufficient number of specimens to make Atlantic forest of Brazil, the Serra do Mar coastal
large sample sizes for multivariate analysis, except in forests ecoregion (Vogliotti & Duarte, 2010).
the case of the three montane samples, which we fi nd Weight 25.0 kg, head plus body length 1061 mm,
are 100% differentiated. Skull lengths (greatest and ear 94 mm, metatarsus 220 mm, (tail, no data), height
condylobasal) are high in the Peruvian samples, low 575 mm. This compares with M. nana × M. americana
in Colombia (with nasal breadth); nasal breadth and hybrids as follows: weight 22.0 kg, head plus body
biorbital breadth are high, and toothrow length low, length 1054 mm, ear 90 mm, metatarsus 205 mm, tail
in the Peruvian Andes group; these characters are the 112 mm, height 562 mm.
opposite in the Ecuador–Peru border mountains,
with Colombia in between. Mazama nana (Lesson, 1842)
For other geographic regions, although sample 2n = 36–39.
sizes are often small, it is nonetheless worth giving a For a full account of this species, see Abril, Vo-
table of univariate measurements (table 14), because gliotti, et al. (2010).
there are striking differences among the samples; avail- Medium-sized. Weight 15.2 kg, head plus body
able names for par ticu lar regional samples, where 900 mm, ear 91.2 mm, metatarsus 178 mm, tail 88 mm,
they exist, are indicated. Notable variations are as height 482 mm (Duarte et al., 2008); but Eisenberg &
follows: The Darien specimen and the zetta sample are Redford (1996) give total length 853 mm, head plus
the smallest in size, though the Darien skull is broad; body 776 mm, tail 78 mm (all specimens, N = 10), ear
M. americana and the La Macarena skull are unusually 83 mm (N = 9), weight 8.2 kg (N = 1).
large. The Roca Nova skull (M. jucunda) is below the Striking chestnut-brown; venter snow-white. Males
size of any other. The M. whiteleyi mean skull length is with spike antlers. Relatively short-legged. Large pre-
average for the group, but the toothrow is short. The orbital glands.
Meta sample has long nasals for its skull size. SE Brazil, the far NE of Argentina, and adjacent
The antlers are long in M. whiteleyi, M. trinitatis easternmost Paraguay; in moist forest, often in hilly
and La Macarena, medium-sized in M. fuscata, M. terrain.
zetta and the São Paulo skull (which may represent
M. nana); and very small in M. zamora, M. americana, Mazama whitelyi (Gray, 1873)
and Meta.
1898 Mazama tschudii Lydekker, in part; not Wagner, 1855.
We offer the following provisional arrangement,
using relevant species names simply for guidance, Medium-sized (in our admittedly small sample), with
and listing what, from our data seem, to be their out- a very short toothrow. Antlers long.
standing characteristics. S Peru.
Mazama bororo Duarte, 1996 Mazama americana (Erxleben, 1777)

1919 Mazama bororo Miranda-Ribeiro; conditional name, We propose that the following names are synonyms of
made available by its defi nitive use by Duarte (1996). M. americana:
1817 Cervus rufus F. Cuvier, in part; not Illiger, 1815.
For a full account of this species, see Vogliotti & Du-
? 1850 Coassus auritus Gray.
arte (2010), whose description and comparisons are fol-
1872 Homelaphus inornatus Gray.
lowed below.
1915 Mazama americana juruana J. A. Allen.
2n = 32–34.
1915 Mazama americana tumatumari J. A. Allen, in part; skin
Medium-sized. Resembles M. americana in the
only.
-1— reddish color, position of the white body areas, and
0—
349-47558_ch01_1P.indd 76 5/31/11 7:05 PM

Table 14 Univariate statistics for samples of the Mazama americana group
Gt l Cb l Teeth Biorb Nas l Nas post br Antler l
Darien temama 198.00 188.00 61.00 94.00 62.00 26.00 —

N 1 1 1 1 1 1
Manabi fuscata 212.00 199.00 65.00 94.00 — 28.00 69
N 1 1 1 1 — 1 1
Mts of Colombia zetta = gualea
Mean 202.13 190.50 59.78 86.44 61.00 25.67 67.33
N 8 8 9 9 9 9 3
Std dev 8.659 9.150 1.986 3.812 7.762 3.873 22.546
Min 193 177 57 81 50 19 44
Max 215 203 63 90 72 29 89
Zamora zamora
Mean 210.00 199.60 65.20 88.40 65.00 20.40 20
N 5 5 5 5 5 5 1
Std dev 5.292 4.722 2.775 5.177 3.162 1.817 —
Min 203 193 62 81 60 18 —
Max 217 204 69 95 68 22 —
Peruvian Andes whitelyi
Mean 214.75 203.50 60.38 93.38 62.75 24.62 90.33
N 8 8 8 8 8 8 3
Std dev 4.621 5.555 3.543 3.503 4.334 1.923 16.503
Min 208 197 55 89 56 22 74
Max 222 212 65 99 69 28 107
S Bolivia sarae — — 58.00 78.00 — 26.00 —
N — — 1 1 — 1 —
La Macarena 223.00 210.00 64.00 100.00 72.00 31.00 110
N 1 1 1 1 1 1 1
Meta
Mean 214.67 204.00 65.00 91.33 70.50 22.50 9
N 3 3 3 3 2 2 —
Std dev 5.033 6.928 1.732 4.163 — — —
Min 210 200 63 88 69 22 —
Max 220 212 66 96 72 23 —
Guyana Shield americana = tumatumari = juruana
Mean 227.00 216.00 66.50 94.00 67.00 24.00 20
N 2 2 2 2 2 2 —
Min 227 214 61 90 66 24 —
Max 227 218 72 98 68 24 —
Trinidad trinitatis
Mean 210.67 201.67 64.00 91.67 61.67 24.00 92.5
N 3 3 3 3 3 3 2
Std dev 1.528 2.517 2.646 2.517 2.082 .000 —
Min 209 199 62 89 60 24 65
Max 212 204 67 94 64 24 120
Pará bororo?
Mean 216.67 204.00 65.33 91.67 70.33 23.00 —
N 3 3 3 3 3 3 —
Std dev 4.163 5.000 3.786 5.859 3.215 1.732 —
Min 212 199 61 85 68 22 —
Max 220 209 68 96 74 25 —
Roca Nova jucunda — 178.00 58.00 80.00 59.00 19.00 —
N — 1 1 1 1 1 —
São Paulo nana? 207.00 199.00 64.00 91.00 61.00 25.00 56
N 1 1 1 1 1 1 1
—-1
—0
349-47558_ch01_1P.indd 77 5/31/11 7:05 PM

78 Artiodactyla
Karyotype: 2n = 68–70; but in Brazil, 2n = 48–53. Incertae sedis

Antlers small. Preorbital gland small or absent. The following names cannot be allocated at present:
Gestation 218 and 225 days (two records).
Cervus rufus Illiger, 1815. From Mato Grosso,
Guyanas, S and E Venezuela, NE Brazil.
across Paraguay and extreme N of Argentina,
to the Rio Bermejo. Synonyms—Mazama pita
Mazama jucunda Thomas, 1913
Rafi nesque, 1817; Cervus dolichurus Wagner,
1884 Mazama rufus Nehring; not Illiger, 1815. 1844; Mazama rufa toba Lönnberg, 1919.
Mazama rufa rosii Lönnberg, 1919. N Argentina,
Very small in size (in our material).
“from Rio Bermejo to the provinces of
SE Brazil, from São Paulo to Rio Grande do Sul.
Tucumán, Santiago del Estero, and Santa Fe.”
Mazama sarae Thomas, 1925. Mountains of
Mazama zamora J. A. Allen, 1915
extreme S Bolivia and the neighboring part of
Apparently medium-sized; antlers very small.
Argentina.
SE Colombia, Ec uador E of the Andes, and NE
Peru. The following two samples evidently represent
distinct taxa, but no names appear to be available:
Mazama zetta Thomas, 1913
1. medium-sized; long nasals—Meta (antlers very
Small (in our dataset); antlers medium-sized.
small)
Inter-Andean valleys of Colombia.
2. large; very broad, long nasals—La Macarena
(antlers long)
Mazama gualea J. A. Allen, 1915
1915 Mazama fuscata J. A. Allen.
Other Species
A poorly defined, medium-sized species or population. Two remaining species, M. bricenii and M. rufina (de-
Ec uador, W of the Andes, and possibly SW scribed below), which are almost certainly related to M.
Colombia. americana, were not examined by Duarte et al. (2008).
We have only very small samples of these two (N = 3
Mazama temama (Kerr, 1792) and 4, respectively; see table 15). Even with a very high
ratio of variable number to sample size (which risks a
We propose that the following names are synonyms of M.
type I error), they did not separate in our discriminant
temama:
analysis. Given the reported differences between them,
1860 Cervus sartorii Saussure. S Mexico.
however, we do not recommend uniting them, and ta-
1913 Mazama tema reperticia Goldman. Panama, probably the
ble 15 shows that M. bricenii has smaller teeth than M.
extreme N of Colombia.
rufina. A specimen from Cundinamarca may represent
1914 Mazama tema cerasina Hollister. E Chiapas.
a range extension for M. rufina.
? 1959 Mazama americana carrikeri Hershkovitz. 3000–
3900 m, Sierra Nevada de Santa Marta, N Colombia.
Mazama bricenii Thomas, 1898
This is a well-authenticated species; for a full account, For a full account of this species, see Lizcano, Yerena,
see Bello-Gutiérrez et al. (2010). et al. (2010).
Karyotype: 2n = 50 (males), 49 (females). Eisenberg & Redford (1999) gave a skull length of
Very small. Skull broad. Antler length 50–91 mm 159 mm; antlers 5–5.5 cm. Reddish chestnut, with dark
(Mexico, N = 10), 81–96 mm (Central America and Co- brown cheeks; underparts lighter brown, but the ven-
lombia, N = 3). tral surface of the tail white.
Extending from SE Mexico through Central Andes of W Venezuela and neighboring Colombia.
America to NW Colombia (Bello- Gutiérrez et al.,
2010). Mazama rufina (Pucheran, 1852)
For a full account of this species, see Lizcano, Álva-
Mazama trinitatis J. A. Allen, 1915 rez, et al. (2010).
Light cinnamon-rufous, paler sides and belly; tail Rich chestnut-brown above. Face dark brown.
white below. Size medium to large; antlers long. Well-developed preorbital glands. Dark face; outer side
-1— Trinidad. of the legs blackish; ears with some white hair on the
0—
349-47558_ch01_1P.indd 78 5/31/11 7:05 PM

cerv idae 79
Table 15 Univariate statistics for Mazama rufina and M. bricenii
Gt l Cb l Teeth Biorb Nas l Nas post l Antler l
Colombia/Ec uador rufina

Mean 163.50 156.25 53.25 71.00 44.75 19.00 69.00
N 4 4 4 4 4 4 1
Std dev 4.203 3.500 2.630 1.826 4.349 1.414 —
Min 158 151 51 69 41 18 —
Max 168 158 56 73 51 21 —
Mérida bricenii
Mean 160.67 153.33 50.00 67.33 42.67 20.33 —
N 3 3 3 3 3 3 —
Std dev 6.028 7.638 1.000 4.041 2.517 1.155 —
Min 155 145 49 65 40 19 —
Max 167 160 51 72 45 21 —
Cundinamarca
Mean 158.00 149.00 49.00 71.00 41.00 19.00 —
N 1 1 1 1 1 1 —
inside. Fur short and thick, with a woolly undercoat. sulci usually more than 20 mm long (Colombia
Antlers rarely more than 8 cm. and Venezuela); nasals straight (M. permira) or
Premontane forests of S Colombia and Ecuador; slightly and evenly convex in lateral profi le
dense undergrowth of premontane to montane for- (Colombia and Venezuela); narrow postorbital
ests, at 1500–3500 m. constriction; squamosal root of the zygomata
broadly arched (in lateral profi le) above the
Group Forming a Clade with Hippocamelus, glenoid fossa; mesopterygoid fossa V-shaped
Blastocerus, and Ozotoceros (Brown Brockets) —pedicels slender; antlers short to long, straight,
According to Duarte et al. (2008), this divides into four ridged at the base when unworn
subclades: (1) Mazama gouazoubira and Hippocamelus bi-
Gestation 206 days.
sulcus (support values for these two in the same clade
Panama S to N Argentina, including the Chaco; S
are not very high); (2) Blastocerus; (3) Hippocamelus anti-
to Catamarca.
sensis and Ozotocerus; and (4) Mazama nemorivaga.
We analyzed brown brockets in the following two
groupings:
Mazama gouazoubira Group
For a full account of this group, see Black-Décima 1. Montane populations. Our samples fall into
et al. (2010). three groups—(1) Santa Marta, Cauca, Bolivia,
2n = 63–64, differing from all other New World Mérida (up to 2000 m); (2) Loreto (Peru–
deer in having an acrocentric X chromosome. Ecuador border), 100–200 m; (3) Meta (about
Total length 1034 mm (N = 32), head plus body 180 m)—with a specimen from Pasque (La
924 mm, tail 110 (N = 33), hindfoot 268 mm (N = 29), Macarena) that is different from all the others.
ear 107.5 mm (N = 21), weight 16.3 kg (N = 20, Argen- 2. Lowland populations. The Atlantic coast groups
tina, Paraguay, Uruguay). (Pará, São Paulo, Uruguay) fall close together,
Eisenberg (1989), Redford & Eisenberg (1992), Me- while the other lowland populations are separate.
dellín et al. (1998), and Eisenberg & Redford (1999)
We offer the following tentative revision of the M.
gave the following information for this species-group,
gouazoubira group, but it is extremely provisional,
which they all treated as a single species:
and under the same constraints as for the M. ameri-
—smaller and slighter than M. americana cana group.
—gray to gray-brown coat; predominantly gray
in the S parts of the range Mazama cita Osgood, 1912
—supraorbital foramina minute or absent (M.
1833 Cervus humboldtii Wiegmann, nom. nud.
permira; see below) or variable, with associated
1913 Mazama sheila Thomas. —-1
—0
349-47558_ch01_1P.indd 79 5/31/11 7:05 PM

Table 16 Univariate statistics for Mazama gouazoubira
Nas
Gt l Cb l Teeth Biorb Nas l post br Antler l
Santa Marta sanctaemartae

Mean 180.69 171.69 55.69 78.92 55.23 23.62 106.00
N 13 13 13 13 13 13 1
Std dev 3.772 4.803 1.843 4.051 3.632 2.815 —
Min 175 165 53 72 49 19 —
Max 187 180 60 85 61 28 —
Cauca 2000 m
Mean 178.67 167.67 59.00 82.33 52.67 22.00 44.00
N 3 3 3 3 3 3 1
Std dev .577 1.528 2.000 2.517 2.887 .000 —
Min 178 166 57 80 51 22 —
Max 179 169 61 85 56 22 —
Loreto 100 m
Mean 180.50 169.50 55.00 74.50 49.50 16.00 67.00
N 2 2 2 2 2 2 1
Min 174 162 54 71 42 15 67
Max 187 177 56 78 57 17 67
Pasco 1000 m tschudii
Mean — — 60.00 94.00 — 24.00 90.00
N — — 1 1 — 1 1
Bolivia 50 m
Mean 180.00 172.00 54.50 78.00 55.50 21.00 —
N 2 2 2 2 2 2 —
Min 178 170 53 76 54 21 —
Max 182 174 56 80 57 21 —
Mérida, Zulia cita = sheila
Mean 181.50 173.75 57.75 78.25 53.25 21.50 53.50
N 4 4 4 4 4 4 2
Std dev 7.326 7.274 2.754 4.787 3.862 2.380 —
Min 171 163 55 74 51 20 33
Max 188 179 61 85 59 25 74
Pasque
Mean 165.00 157.00 55.00 75.00 45.00 19.00 21.00
N 1 1 1 1 1 1 1
Meta, Caqueta 180 m murelia
Mean 172.75 162.00 52.75 72.00 49.00 17.50 55.00
N 4 4 4 4 4 4 1
Std dev 2.754 2.309 1.258 1.826 2.449 1.291 —
Min 170 160 51 70 46 16 —
Max 176 164 54 74 52 19 —
Suriname nemorivagus
Mean 176.25 167.00 50.25 78.25 52.75 17.25 67.50
N 4 4 4 4 4 4 2
Std dev 5.679 5.944 .500 1.708 .957 1.258 —
Min 172 162 50 76 52 16 58
Max 184 175 51 80 54 19 77
Pará superciliaris
Mean 175.88 166.38 50.88 74.38 53.88 20.13 49.50
N 8 8 8 8 8 8 2
Std dev 8.983 9.501 6.010 6.739 5.384 2.588 —
Min 167 159 44 65 48 16 36
Max 194 186 64 86 64 24 63
-1—
0—
349-47558_ch01_1P.indd 80 5/31/11 7:05 PM

cerv idae 81
Table 16 (continued)
Nas
Gt l Cb l Teeth Biorb Nas l post br Antler l
Mato Grosso 142 m simplicicornis

Mean 183.80 177.40 55.60 76.20 51.60 20.80 92.00
N 5 5 5 5 5 5 1
Std dev 6.380 5.941 2.074 3.114 6.348 1.789 —
Min 174 171 54 71 44 20 —
Max 190 186 59 79 59 24 —
São Paulo
Mean 185.50 175.50 63.00 85.00 57.50 22.50 —
N 2 2 2 2 2 2 —
Min 185 175 63 85 57 22 —
Max 186 176 63 85 58 23 —
Uruguay
Mean 182.50 173.00 52.00 82.50 51.50 22.50 134.50
N 2 2 2 2 2 2 2
Min 180 171 49 80 51 19 123
Max 185 175 55 85 52 26 146
Large size; medium to short antlers. Mazama nemorivaga (F. Cuvier, 1817)
Thomas and J. A. Allen referred M. sheila to the For a full account of this species, see Rossi et al.
americana group, as it is bright rufous; but the skull (2010), who separated it from M. gouazoubira as a dis-
has characteristics of the M. gouazoubira group. The tinct species.
color of M. cita has been noted as being different from Medium-sized; very short toothrow; antlers me-
other brown brockets. dium length.
N of Venezuela. Guyanas, SE of Venezuela, and N of Brazil.
Mazama gouazoubira (G. Fischer, 1814) Mazama permira Kellogg, 1946

1815 Cervus simplicicornis Illiger. Type locality: Isla San Jose, Panama.
1817 Mazama bira Rafi nesque.
We have no experience of this putative species.
1817 Cervus nemorivagus F. Cuvier, in part.
1879 Nanelaphus namby Fitzinger, 1873, nom. nud.
Mazama sanctaemartae J. A. Allen, 1915
1883 Nanelaphus nambi Pelzeln.
Large size; long antlers.
1919 Mazama argentina Lönnberg.
N of Colombia.
1919 Mazama simplicicornis var. kozeritzi Miranda-
Ribeiro.
Mazama superciliaris (Gray, 1852)
1923 Azarina fusca Larrañaga.
Medium-sized; toothrow very short; narrow biorbital
1951 Mazama gouazoubira Hershkovitz.
breadth; broad nasals; antlers medium to short.
Large size; antlers fairly long. C and E Brazil, from Amazonas to Serra dos
S of Brazil, from the state of São Paulo and the Parecis (in Mato Grosso), and the state of Espiritu
Planalto de Mato Grosso; Paraguay; N of Argentina Santo.
as far as Tucumán, Santiago del Estero, and Entre
Rios; Uruguay. Mazama tschudii (Wagner, 1855)
Size apparently medium; long toothrow; antlers fairly
Mazama murelia J. A. Allen, 1915 long.
Small size; short toothrow; narrow biorbital breadth; Peruvian Cordillera.
antlers medium to short. The following samples may represent distinct
SE of Colombia and Ecuador, E of the Andes. taxa, which appear to be without names: —-1
—0
349-47558_ch01_1P.indd 81 5/31/11 7:05 PM

82 Artiodactyla
Uruguay—size large, with wide biorbital breadth,

and antlers very long Pudu Gray, 1852
Cauca—also large and broad, but antlers pudus
short The skull characters are typically those of the neo-
Bolivia—size large, less broad tropical endemics, with modifications. The supraor-
São Paulo— size large, with large teeth and long bital sulci are short, or nearly flat, and contain two
nasals large foramina. For details of the cranial and postcra-
Loreto—size large, with narrow biorbital nial morphology, see the monograph by Hershkovitz
breadth, short narrow nasals, and antlers (1982); for a full account of their biology, see Escamilo
medium et al. (2010) and Jiménez (2010). The two species are
Pasque—size very small, antlers very short Pudu puda (Molina, 1782) and Pudu mephistophiles de
Winton, 1896.
Mazama rondoni Miranda-Ribeiro, 1914
2n = 68–70. Rangifer Hamilton Smith, 1827
We have no experience of this putative species. reindeer (Eu rope), caribou (North America)
Rangifer are currently regarded as a single species, R.
Mazama pandora Merriam, 1901 tarandus, which is broadly divided into woodland,
For a full account of this species, see Weber & Medel- tundra, and High Arctic groups of subspecies. There
lín (2010). is a convincing cline of shorter limbs related to lower
Distinguished as a species in a meticulous study temperature, covering all Rangifer, whether wood-
by Medellín et al. (1998), from whom the description land, tundra, or Arctic (Klein et al., 1987)— giving
below is largely taken. the Arctic island forms, in par ticu lar, a distinctive
Brown to gray-brown, venter paler to whitish. appearance—but there are many other characters,
Large patch of long, dark, stiff hairs on the forehead. varying within the genus, that are not especially re-
Premaxillae comparatively broad, spatulate; postor- lated to climate. The woodland forms (R. t. fennicus
bital constriction broad; posterior half of the nasals in the Old World, R. t. caribou in the New World, in
conspicuously humped; supraorbital foramina large, Banfield’s [1961] classic monograph) are very different
usually opening into long, prominent grooves; poste- in appearance— antler form, color, and build—from
rior margin of the palate usually U-shaped; dorsal the tundra forms (R. t. tarandus in the Old World, R. t.
margin of the squamosal root of the zygoma nar- granti and R. t. groenlandicus in the New World). There
rowly arched above the glenoid fossa. Frontal region is said to be some seasonal overlap in range when
broad, especially in the males. Large auditory bulla. tundra reindeer / caribou enter the N fringes of the
Condylobasal length 161–177 mm (N = 8). Antler pedi- coniferous forest, but how much interbreeding there
cels massive; antlers long, divergent, usually curved, may be is not known. This situation suggests that
sometimes converging at the tips. Antlers heavy, speciation is well underway (or complete), but before
fluted along almost the whole length when unworn. we can contemplate a full taxonomic arrangement,
Antler length 112–142 mm (N = 6). we need to know whether Old and New World wood-
Yucatán peninsula; in S and coastal Campeche, it land forms fall into a clade separate from the tundra
is sympatric with M. americana. forms, or are separately derived from them (or, even,
if the tundra forms are derived from the woodland
Mazama chunyi Hershkovitz, 1959 forms).
For a full account of this species, see Rumiz & Pardo On the other hand, the monophyly of the three
(2010). High Arctic forms has already been investigated: the
Head plus body 70 cm; height 38 cm; antlers 3.5 cm; two New World forms, R. t. pearyi and R. t. eogroen-
weight averaging 11 kg. landicus, do indeed form a clade, one related to the
Short-tailed; brown, with a dark head, legs, and nearest tundra subspecies, R. t. groenlandicus (known
the backs of the ears; underside of the tail white. as barren-ground caribou in Canada); while the Sval-
Pale; white markings on the ear margins and the bard reindeer, R. t. platyrhynchus, though morpholog-
tip of the muzzle. Lacks the dark brown face of M. ically not dissimilar to them, is a clear derivative of
rufina. Old World tundra reindeer (Gravlund et al., 1998).
-1— E premontane forests of Andean Peru and Bolivia; Tundra and High Arctic reindeer are said to be sea-
0— 1500–3200 m. sonally sympatric on some of the Canadian High Arc-
349-47558_ch01_1P.indd 82 5/31/11 7:05 PM

cerv idae 83
tic islands; they are strikingly different in appearance, In sum, there is a clear need for a thoroughgoing
but in this case one wonders how much the differ- revision of this genus.
ences might be due to phenotypic plasticity.
The taxonomy and distribution of the caribou of Tribe Capreolini Brookes, 1828
the Canadian Arctic have been reviewed by D. This tribe contains two genera: Capreolus, the roe-
Thomas & Everson (1982). The Boothia Peninsula deer, and Hydropotes, the Chinese water-deer.
has R. t. groenlandicus. R. t. pearyi occurs on the W
Queen Elizabeth Islands and Parry Islands, and on Capreolus Gray, 1821
Somerset Island and Prince of Wales Island; the cari- roedeer
bou of these two island groups have some differ- Lister et al. (1998) described the morphology of roe-
ences between them. The main differences found in deer as a whole and affi rmed that there are two quite
the study were that R. t. pearyi is smaller, with a distinct species. Randi, Pierpaoli, et al. (1998) studied
much shorter muzzle and much lighter coloration. control region haplotypes of European and Siberian
Strictly speaking, however, the more S island popu- populations of both species, and found them to be
lations were considered to be “intergrades”; these 100% different; the split between them goes back 2 to
apparently spent the winter on the N Boothia Penin- 3 Ma. Moreover, there are two clusters within each
sula and the summer on Prince of Wales and Somer- of the species, opening the possibility that the taxo-
set islands. nomic diversity is still underestimated (Randi, Pier-
There have been several attempts to test the wood- paoli, et al., 1998). The genus needs a thoroughgoing
land versus barren-ground division in North Amer- taxonomic revision, which we are not yet equipped
ica, with different results. Cronin (1992) found shared to undertake.
genotypes, whereas Røed et al. (1991) showed that The western roedeer Capreolus capreolus seems to
they differed in transferrins, and that woodland cari- have extended its range at the expense of the E C. py-
bou were geograph ically variable as well (see also gargus in modern times.
van Staaden et al. [1995], who likewise studied trans-
ferrin frequencies). Flagstad & Røed (2003), on their Capreolus capreolus (Linnaeus, 1758)
minimum spanning network, discovered that almost western roedeer
all samples of caribou formed a single cluster, whereas Karyotype: 2n = 70; there are no B chromosomes
R. t. groenlandicus and R. t. granti clustered in amongst (Baskevich & Danilkin, 1991).
the undifferentiated R. tarandus sample, but tended to A general survey of roedeer in Europe was given
be within their own subclusters. Courtois et al. (2003) by von Lehmann & Sägesser (1981), including a brief
similarly found low gene flow between woodland, comparison with C. pygargus: the skulls of European
barren-ground, and mountain caribou in NE Quebec, roedeer are shorter but relatively broader and higher
using eight microsatellites. than those of C. pygargus; and the ascending branch of
Geist (1998) did not think that the division be- the premaxilla makes a point contact, or none at all,
tween the two was necessarily the most fundamen- with the nasal, whereas in C. pygargus it makes a broad
tal; he argued, rather, that tundra caribou have contact.
evolved more than once from woodland forms, al- Von Lehmann & Sägesser (1981) distinguished two
though in Europe it may be the other way round (i.e., summer pelage types in C. capreolus: “continental”
European forest reindeer derived from tundra forms). (paler; yellowish ochery-colored; longer, stronger hair,
In addition, he noted that there are fundamental dif- with a dark gray to black base) and “Atlantic” (deep
ferences between Eurasian and American taxa in red; thin hair with a white base). The Atlantic type
color, especially the much greater extension of the is seen in Belgium, the Netherlands, and the lower
white neck color onto the shoulder, and the presence Rhine region; the continental type is in, for example,
of an extensive white area on the flanks, separated by the Rothaar Mountains (in N Germany, E of Co-
a clear dark line from a broadly white underside, both logne). Both types occur in France and Spain. Scot-
characteristic of Eurasian taxa. tish animals mix the deep red color of the Atlantic
Indeed, population subdivision has been impor- type with the dark hair bases and strong hair struc-
tant throughout Rangifer history, as was shown by ture of the continental type. Ear length is less in the
Côté et al. (2002), who found that even within Sval- N than in the S. Roedeer from Spain and Switzerland
bard, in that very distinctive taxon, there is some sub- seem to be somewhat bigger than those from Ger- —-1
division of populations. many; those from the Carpathians and Croatia also —0
349-47558_ch01_1P.indd 83 5/31/11 7:05 PM

84 Artiodactyla
seem rather large. The phalanges divide into an E

type, which are thinner and longer, and a shorter, Hydropotes Swinhoe, 1870
thicker W type. Von Lehmann & Sägesser (1981) went This genus, according to multiple datasets (molecu-
on to list 12 subspecies, but without details such as lar and others), is nested deep within the Cervidae,
sample size. and is sister to Capreolus. Accordingly, the old prob-
Lister et al. (1998) noted that E European roedeer lem of whether Hydropotes is primitively antlerless,
are larger than W European roedeer: sample means or has lost them, is decided in favor of the second
for E Europe are 207–230 mm for the males, and for proposition.
W Europe 194–207 mm (the females are correspond-
ingly smaller). Markov et al. (1985) performed a mul- Hydropotes inermis Swinhoe, 1870
tivariate analysis on the limited geographic samples Chinese water- deer
available to them, fi nding a sharp difference between There seems to be no more detailed study of the biol-
the French and Bulgarian samples, on the one hand, ogy of this species than the Wikipedia article (http://
and Ukrainian samples and those from the Baltic en.wikipedia.org/wiki/Water_deer).
countries, on the other. Geist (1998) likewise com- A recent study (Hu et al., 2006) could fi nd no phy-
mented on geographic forms of roedeer, with par tic- logeographic structure in the mtDNA sequences which
u lar reference in this case to a distinctive, apparently they studied.
unnamed, taxon in Garganta, S Spain. Randi, Pierpaoli
et al. (1998) identified genetically unique features in Tribe Alceini Brookes, 1828
the Italian roedeer C. c. italicus versus other European In North America, modern moose appeared only at
samples. the end of the Pleistocene, when they replaced a late-
In a study of roedeer in both France and Sweden, surviving population of the precursor species Alces
Vanpé et al. (2007) found that antler size was corre- latifrons (classed by Azzaroli [1985] as a distinct, en-
lated with age and with body mass, but it was not af- demic North American species, A. scotti).
fected by environmental conditions (except for popu-
lation density, and that only in one of the three Alces Gray, 1821
populations). The authors concluded that antler size moose (North America and international), elk (Eu rope)
is therefore an honest signal of quality. Most authors place all living Alces in a single species,
but Boyeskorov (1997, 1999) has convincingly argued
Capreolus pygargus (Pallas, 1771) that there are, in fact, two species: Alces alces from Eu-
Siberian roedeer rope and W Siberia, and A. americanus from E Siberia
The basic karyotype is 2n = 70, as in C. capreolus, but and North America. The boundary seems to be the
there are varying numbers of B chromosomes. Ac- Enisey River. They differ in chromosome number,
cording to Baskevich & Danilkin (1991), 2n = 1– 4 in color and color pattern, body proportions, skull char-
W Russia as far as Cis-Baikal (Irkutsk), but 2n = 2–14 acters, and antler form. The small moose from the
in the Altai, 2n = 6– 8 in Kazakhstan and Kyrgyzstan, Manchuria/Primoriye region, which Boyeskorov re-
2n = 5– 8 in Mongolia, 2n = 8–12 in Transbaikalia, and ferred to as A. americanus cameloides, has single-palm
2n = 5–10 in the Far East. The authors treated the fi rst antlers like A. alces, rather than the double-palm ant-
(low number) populations as nominotypical C. pygar- lers of the larger A. americanus, and it may rank as a
gus pygargus, and the second (mostly high number) third species. The status of the recently extinct Cauca-
series of populations as C. pygargus tianschanicus. It is sus moose has yet to be settled.
possible that these could indeed be treated as differ- By contrast, Hundertmark, Shields, Bowyer, et al.
ent subspecies, though the Altai would evidently be a (2002) found no such division in cytochrome b se-
transition zone between them, rather than being true quencing; rather, there was a strong division between
C. p. tianschanicus, as they proposed. the three haplotypes in Europe and the four in Asia,
In Randi, Pierpaoli, et al.’s (1998) study, there was with the single haplotype found in North America
a consistent difference between specimens from the being slightly closer to that of Europe. The authors
Kurgan region of W Siberia and those from the did not, however, give the localities of their Asian
Amur region of the Far East. Similarly, animals from haplotypes.
Cheju Island, Korea, formed a distinct clade from In contrast, the localities of origin were given in a
-1— those from the Kurgan and Amur regions (Koh & different study by many of the same authors, using
0— Randi, 2001). the control region (Hundertmark, Shields, Udina, et al.,
349-47558_ch01_1P.indd 84 5/31/11 7:05 PM

cerv idae 85
2002); here there was some, although very little, di- mediate type in 4.57% of them, but apparently never
versity in North America. The minimum-spanning in moose from America or the Far East. The rostrum
tree (their figure 3) showed three phylogroups, with is relatively long; the choanae are broad. Over 90% of
several substitutions between each pair: the fi rst con- the antlers have a double palm, the exception being
tained all the North American samples, plus two those from the Far East and E Mongolia, which nor-
from the Russian Far East and two from Yakutia; the mally have almost unpalmated antlers. The antler
second, the bulk of the E Asian samples; and the third, shaft is longer, averaging 120 mm. There are said to
the European and three E Asian samples. The authors be differences in vocalization, in some proteins, and
suggested that the strong differences—for example, in sex pheromones.
in chromosomes—between the E and W moose must The division between the two species goes back
be of rather recent origin. to Flerov (1931), who distinguished them only at the
subspecific level. For E Siberia, he recognized Alces
Alces alces (Linnaeus, 1758) alces pfizenmayeri Zukowsky, 1910, which, he percep-
This, the moose from Europe and W Asia, at least as tively suggested, “may turn out to be identical with
far E as the Altai, has 2n = 68 (Boyeskorov et al., 1993; those of Alaska A. a. gigas Miller.” He also recognized
Boyeskorov, 1997).The localities where this chromo- what he called “Alces alces subsp.?” from the Amur,
some number has been recorded were mapped by which has the premaxillary characters of E elks but is
Boyeskorov (1999). of small size; in fact, the available name for this kind
Boyeskorov (1999) summarized the overall differ- of moose is A. a. cameloides (Milne-Edwards, 1867).
ences between the two species. The W species has an Geist (1987) also noted that A. americana has a much
evenly brownish overall color, with the legs whitish. larger bell, with a longer rope; the face of the cow is
The body size index (body length as a percentage of reddish brown, whereas in the bull the rostrum is
shoulder height) is 80.8%– 84.5%. In the skull, the black. In principle, there is a light saddle on the back,
posterior end of the nasal branch of the premax- but this is less conspicuous in E North America.
illa is “spoonlike,” although a very small proportion The control region shows low diversity in North
(3.57%) have the pointed “American” type, and an in- America (Hundertmark et al., 2003) and forms a star
termediate type occurs in 11.71%. The rostrum is phylogeny, indicating a recent expansion of popula-
relatively short. The antlers have a single palm tions (the fossil record depicts the modern moose
(80.3%), rarely a double palm, which is known as the entering North America only after the end of the
butterfly type (13.6%), while 6.2% had virtually no Pleistocene).The validity of the described subspecies
palmation. The antler shaft is relatively short, averag- will not, however, depend on the inferred late date
ing 91.3 mm. of their diversification, but rather on whether the
Earlier, Geist (1987) considered the differences differences between them (see Peterson, 1950) char-
between the two types of moose in a different man- acterize a sufficient proportion of their respective
ner. The “European type,” he found, has “noticeably populations.
smaller hindquarters” and their antlers are “best de-
scribed as being of a three-pronged plan: those of the
Subfamily Cervinae Goldfuss, 1820
American-type moose of a four-pronged plan”; this The plesiometacarpal deer; the brow-tine deer (pace
principle was illustrated in his figure 2. Elaphurus!).
On the Old World deer in general, see Groves &
Alces americana (Clinton, 1822) Grubb (1987), Geist (1998), and Grubb (200b).
Moose with 2n = 70 occur in North America and, as
far as Eurasia is concerned, have been confi rmed in C Tribe Muntiacini Knottnerus-Meyer, 1907
Sakha, on the Kolyma River (Boyeskorov et al., 1993;
The name Cervulini Sclater, 1870 (originally Cervulinae),
Boyeskorov, 1997).
has technically priority over Muntiacini, but the former
As summarized by Boyeskorov (1999), the overall
has now been placed on ICZN’s Official Index of Rejected
color is from light brown to black, with legs from
and Invalid Family- Group Names.
light to dark brown or gray. The body size index is
87.8%– 89.6%, but a NE Siberian moose is 94.3%. The The reputed diversity of species in Muntiacus, with
posterior end of the nasal branch of the premaxilla is several new species described over the past 10 years
narrow and pointed; the “spoonlike” type occurs in or so (beginning with the giant muntjac, originally —-1
only 0.57% of the E Siberian specimens, and the inter- placed in a separate genus and named Megamuntiacus —0
349-47558_ch01_1P.indd 85 5/31/11 7:05 PM

86 Artiodactyla
vuquangensis), is getting out of hand. It is surely 1827 Stylocerus Hamilton Smith. Type species: Cervus
time to gather the available specimens of little black muntjak Zimmermann, 1780.
muntjac in one place, compare them morphologically 1836 Prox Ogilby. Type species: Prox moschatus Ogilby,
and morphometrically, and run several mtDNA and 1836.
nDNA sequences. That said, we now agree that the 1843 Muntjacus Gray. Type species: Muntjacus vaginalis Gray,
revision by Groves & Grubb (1990) was somewhat 1843 (= Cervus vaginalis Boddaert, 1785).
overlumped. 1923 Procops Pocock. Type species: Cervulus feae Thomas &
Doria, 1889.
Elaphodus Milne-Edwards, 1871
Antlers longer than in tufted deer, with long pedicels,
tufted deer
extending down onto the face as thick straight “ribs”
1871 Elaphodus Milne-Edwards. Type species: E. cephalophus ending on the upper orbital margins, but not follow-
Milne-Edwards, 1871. ing them; preorbital fossa smaller, less well defi ned;
1874 Lophotragus Swinhoe. Type species: L. michianus preorbital fissure well developed; ears narrower, more
Swinhoe, 1874. pointed; lateral hoofs shorter; frontal tuft, if present,
less thick or matlike.
Chromosomes 48 (at least in the females).
Muntiacus is notorious for having the widest range
Antler pedicels short, very slender, not extending
of chromosome numbers in any mammalian genus,
as more than weak ridges on the face, and continuing
from 2n = 46 in M. reevesi to 2n = 6/7 in M. vaginalis,
along the upper margins of the orbits; extremely large,
with several different numbers in between (M. crini-
well-defi ned preorbital fossa; preorbital fissure poorly
frons and M. gongshanensis, 2n = 8/9; M. feae, 2n = 13/14).
developed or absent; ears broad, rounded; lateral hoofs
The homologies between them were studied by Yang
very long; frontal tuft short but very thick.
et al. (1995) through chromosome painting; these au-
thors showed, for example, that the chromosomes of
Elaphodus cephalophus Milne-Edwards, 1871
M. crinifrons and M. gongshanensis, despite having the
1871 Elaphodus cephalophus Milne-Edwards. Baoxing, same diploid number, actually have several rearrange-
Sichuan. ments between them.
1874 Lophotragus michianus Swinhoe. Ningpo, Chejiang. Lan et al. (1995) compared the four Chinese spe-
1904 Elaphodus ichangensis Lydekker. Ichang, Hubei. cies of Muntiacus using restriction sites of mtDNA.
1904 Elaphodus michianus fociensis Lydekker. “Kohwang near Whether using parsimony or distance methods, “Indian
Ching Feng Ling, about 100 miles northwest of muntjac” (here, M. vaginalis) was sister to the others, fol-
Foochow” (see Groves & Grubb, 1990). lowed by a split between M. reevesi on one branch, and
M. gongshanensis and M. crinifrons on the other.
There is a good deal of geographic variability in Ela-
Amato et al. (2000), however, on the basis of four
phodus, but with so little available material, Groves &
mitochondrial sequences, placed M. reevesi as sister to
Grubb (1990) were unable to make much of it.
the other (mostly mainland) species, followed by a
As noted by Groves & Grubb (1990), there is a con-
split between M. truongsonensis, M. putaoensis, M. roo-
siderable size different between tufted deer from the W
seveltorum, and M. vuquangensis, on the one hand, and
China mountains and N Burma (skull length 181–
M. muntjak, M. gongshanensis, M. crinifrons, and M.
202 mm), and those from the coastal mountains of E
feae, on the other.
China (skull length 171–181 mm); but the type of Elapho-
Pitra et al. (2004) calculated that M. reevesi sepa-
dus michianus fociensis is large, like the W ones. Clearly, a
rated from M. crinifrons plus M. vaginalis somewhat
larger series of specimens must be studied before any
over 3 Ma, while these latter two would have sepa-
conclusion can be reached as to taxonomic diversity.
rated some 2.5 Ma.
There is still much to be learned about this genus.
Muntiacus Rafinesque, 1815
The necessity of breaking up the former M. muntjak
muntjac
into several different species, in addition to the pleth-
1815 Muntiacus Rafi nesque. Type species: Cervus muntjak ora of new species that have been described over the
Zimmermann, 1780. This name has been validated by last 20 years or so, necessitate the genus being divided
ICZN, Opinion 460. in some way; we have divided the species into four
1816 Cervulus de Blainville. Type species: Cervus muntjak informal groups, which do not pretend to be homo-
-1— Zimmermann, 1780. geneous or monophyletic groups.
0—
349-47558_ch01_1P.indd 86 5/31/11 7:05 PM

cerv idae 87
2n = 6 (females), 7 (males).
Muntiacus muntjak Group Antlers generally shorter, 90–125 mm. Dark red-
(Red or Indian Muntjac) dish in general, somewhat darker on the midback;
Large muntjak with low chromosome numbers. nape slightly grayer; forehead and occiput light orange-
It should be noted that karyotypes are known only brown, the rest of the face grayish; ear-backs reddish
for M. muntjak and M. vaginalis; while it may well at the base, the remaining two-thirds dark gray; limbs
be that the other mainland taxa will have identical brown to gray; underside paler; groin and the line on
karyotypes to those of M. vaginalis, it should not be the front of the hindlegs white.
assumed. Nepal, most of E India, to W Burma.
Muntiacus muntjak (Zimmermann, 1780) Muntiacus vaginalis curvostylis (Gray, 1872)

1780 Cervus muntjak Zimmermann. Java. 1872 Cervulus curvostylis Gray. Pachebon, Thailand.
Synonymy as in Groves & Grubb (1990), under Muntiacus 1904 Cervulus muntjac grandicornis Lydekker. Thouagyen
muntjak muntjak. Forest, Amherst district, Burma.
1928 Muntiacus muntjak annamensis Kloss. Langbian.
2n = 8 in the females, 9 in the males.
1988 Muntiacus muntjak menglalis Wang & Groves. Pujiao,
Antlers long, >80 mm, usually >100. Black stripes
Mengla County, Xishuanbana, Yunnan.
on the pedicels exceptionally thick and deep black.
Color deep rufous, the nape and the mid-dorsal re- Much paler than the nominotypical M. v. vaginalis;
gion contrastingly dusky; forehead light cinnamon- legs hardly (or not at all) darker than the body.
rufous, contrasting with a dusky face; ear-backs dark; S and E Burma, Thailand, Laos, Vietnam (except
thighs and shoulders somewhat contrastingly dark the extreme N), southernmost Yunnan.
chestnut-brown, becoming darker and grayer down
the shanks; throat creamy white, becoming light Muntiacus malabaricus Lydekker, 1915
cinnamon-rufous on the rest of the underparts; pubic
1915 Muntiacus muntjak malabaricus Lydekker. Nagarahole.
region and the front of the thighs creamy white. Skull
relatively narrow. Smaller in size than M. vaginalis, with shorter antlers
Java, Bali, Lombok, Borneo, Bangka, Lampung, (<95 mm) and short pedicels; much paler reddish, with
and coastal areas of E Sumatra; Malay Peninsula N to gray tones on the nape and the back; limbs colored as
about Trang. in the body; underside drab; white area on the lower
Groves & Grubb (1990) anticipated the separation limbs prominent, extending around to the front of the
of the Malaysian/Indonesian and the general main- pasterns, with only a narrow band down the limbs.
land forms of red muntjac at the species level. Sri Lanka and W Ghats.
Muntiacus vaginalis (Boddaert, 1785) Muntiacus aureus (Hamilton Smith, 1826)

We have some doubts whether the two subspecies
1826 Cervus aureus Hamilton Smith. “Some part of southern
placed under this name are correctly assigned to a
India,” according to Lydekker (1915a).
single species, although certain characters (shorter
1844 Cervus albipes Wagner. “Bombay and Poona.”
antlers, darker midback, gray nape, orange-brown fore-
1872 Cervus tamulicus Gray. Deccan.
head and occiput, etc.) do unite them. Their colors, on
the other hand, are rather different. Smaller than the other Indian taxa. Pale yellowish,
with a gray nape; forehead and occiput pale orange-
Muntiacus vaginalis vaginalis (Boddaert, 1785) brown, the rest of the face light gray-orange; ear-backs
orange at the base, becoming gray and dark gray on
1785 Cervus vaginalis Boddaert. Bengal.
the tips and the rims; limbs colored like the body; un-
1827 Cervus moschatus Hamilton Smith; not de Blainville,
derside slightly paler; white line down the front of the
1816. Nepal.
thighs to the hocks; antlers short, <100 mm.
1833 Cervus ratwa Hodgson. Nepal.
NW and C India; and on the lower Chindwin in
1839 Cervus melas Ogilby. Himalayas.
Burma.
1844 Cervus stylocerus Schinz.
Groves & Grubb (1990) discussed this species, not-
1846 Prox ratva and Prox albipes Sundevall.
ing its curiously disjunct distribution, but there seems —-1
1846 Stylocerus muntjac Cantor.
to be no doubt about its homogeneity, or its striking —0
349-47558_ch01_1P.indd 87 5/31/11 7:05 PM

88 Artiodactyla
differences from others of this group. Presumably a pressed below the line of the facial profi le; greatest
former continuous range has been interrupted by the skull length 205–224 mm; skull with a convex fa-
spread of M. vaginalis. cial skeleton and a depression between the face and
the interorbital-frontal region; nasals somewhat com-
Muntiacus nigripes G. M. Allen, 1930 pressed, the posterior ends narrow; ethmoid fissure
very small, narrow; forehead very narrow between
1930 Muntiacus muntjak nigripes G. M. Allen. Nodoa, Hainan.
the frontal ridges; preorbital fossa restricted to the
1988 Muntiacus muntjak yunnanensis Ma & Wang. Wokang
lower half of the lacrimal bone. Lateral hoofs long;
Dashan, Menglai, Cangyuan County, W Yunnan,
tail very long, with a long white fringe restricting the
2200 m.
black dorsum to a narrow median area; color black,
Color bright orange to deep red; forehead and occiput with some red tones; frontal tuft bright red, very long,
deep brown or black in color; ear-backs nearly black, thick, hiding the antler pedicels, the red color extend-
except at the bases; shanks, or the whole of the limbs, ing back to the ear-backs and the occiput; underside
gray to dark chestnut to blackish brown; chin and only slightly lighter than the upperside.
throat, axillae, and groin pure white, sending a white W Zhejiang, SE Anhui, extending to Pucheng
line down the inside of the hindlimbs to beyond the County, Fujian.
hocks; white patch on the pasterns; rest of the under-
side a paler version of the upperside. Antlers short, Muntiacus gongshanensis Ma, 1990
but the pedicels comparatively long. Gongshan muntjac
Hainan, northernmost Vietnam (including the
1990 Muntiacus gongshanensis Ma. Mijiao (27° 35' N, 98° 47'
Annamites; see Groves & Schaller, 2000), and Yun-
E), Puladi, Gongshan County, E slope of the N sector of
nan, N of about 23° 10' N.
Gaoligong Mountain, NW Yunnan.
The two described forms— one on the mainland,
Although the 1990 paper describing this species was by Ma,
one on Hainan— differ in size and average color, and
Wang, and Shi, the new species was attributed to Ma
it may be that they should be separated.
alone.
Incertae sedis A medium-sized species; skull length 194.8–205.4 mm.

No forehead tuft. Chin glands rather small. Antler
Muntiacus muntjak guangdongensis Li &Xu, 1996.
length 71.7 mm; pedicels 34.5 mm. Body color and
Dinghushan Reserve, Gaoyiao County,
upperside of the tail dark brown. Neck hair not re-
Guangdong Province, 850 m.
versed; forehead and crown reddish, blackened in
We have not seen specimens of this taxon. The the females. On the skull, the “condylolateral pro-
description states that it is “of medium size”; greatest cess” (as the authors called what seems to be the
skull length 191–205 mm, with an exceptionally wide paroccipital process) larger than in the other spe-
interorbital width. Pale brown-yellow in color, dark- cies; rostrum long, narrow; orbital rims well de-
ening on the midback; flanks and underparts paler to fi ned, with a constriction between the orbit and the
creamy white; black stripes running up the antler lower margin of the preorbital fossa; zygomatic arch
pedicels. In about 50% of the individuals, the anterior angled.
part of the forelimbs and the sides of the hindlimbs Amato et al. (2000) could fi nd no consistent differ-
dark brown. (Following Li J-x. & Xu, 1996.) ences between this species and M. crinifrons in the
four mtDNA sequences they studied. Externally, how-
Muntiacus crinifrons Group ever, M. gongshanensis does differ—it lacks a forehead
Large muntjak with intermediate chromosome tuft, and the tail is said to be not as dark— and the
numbers. karyotype, though basically the same as in M. crini-
frons (8 in the females, 9 in the males), differs in the
Muntiacus crinifrons (Sclater, 1885) nucleolar orga nizing regions, being on chromo-
hairy-fronted muntjac somes 1 and 4, instead of on chromosomes 2 and 4
(Ma et al., 1990); there are also several other chromo-
1885 Cervulus crinifrons Sclater. Ningpo.
some differences.
2n = 8 (females), 9 (males). The distribution is given as the Gaoligong and
-1— Antlers short, up to 52 mm, often unbranched; Biluo mountains. It is doubtless this species, rather
0— pedicels long (40–70 mm) and slender, sometimes de- than M. crinifrons (as reported), that occurs in the
349-47558_ch01_1P.indd 88 5/31/11 7:05 PM

cerv idae 89
Hkakabo-Razi area in Burma, on the border of China. the small muntjac species, and it is possible that this is
at about 30° N, 97° E (Rabinowitz et al., 1998), and it another case of nuclear swamping—in this instance,
is evidently the “northern M. feae” described by Groves of one of the small Indochinese muntjac by a large
& Grubb (1990) from N Burma, W Yunnan, S Tibet, species (proto-vuquangensis).
and possibly even Darjiling.
Muntiacus rooseveltorum Osgood, 1932
Muntiacus feae (Thomas & Doria, 1889)
1932 Muntiacus roose veltorum Osgood. Muong Yo, N Laos
Fea’s muntjac
(12° 30' N, 102° E).
Grubb (1977) used the version feai, but this was unjustified
A small species; agouti-brown pelage; ochraceous
(Grubb, 1990a, 2000b).
crown; white throat; very large mental glands (the
2n = 12, 13, 14 (females), with the same X-autosome glands on the inferior surface of the mandibular
fusion as seen in the M. muntjak group (Soma et al., symphysis).
1983). 2n = 14 (males). Its known distribution, confi ned to Laos, was
Antlers short, fairly stout, usually branched, regu- mapped by Groves & Schaller (2000).
larly cast; pedicels long, very slender; skull broad, DNA from the skin of the type specimen and from
with the maxillae bulging outward; rostrum short, new specimens from Laos placed this species as sister
deep; preorbital fossa large; ethmoid fissure very to putaoensis/truongsonensis.
large; nasals very flat, expanded at the posterior ends;
tail long, black, with a white underside and a long Muntiacus putaoensis Amato et al., 1999
white lateral fringe; lateral hoofs long; dull agouti-
1999 Muntiacus putaoensis Amato et al. Atanga village, 30 km
olive-brown, with contrasting deep blackish limbs;
E of Putao (27° 21' N, 97° 24' E), extreme N Burma.
white ring around the hoofs. Hair reversed on the
midline of the neck. Females larger than the males, Very small in size; pelage chestnut, including the up-
with a long yellow frontal tuft with black stripes; per surface of the tail; pedicels short (under 40 mm)
males similar, but without a tuft. and thin; antlers very short (8–35 mm), unbranched.
According to Amato et al. (2000), this species is Skull length averaging 175 mm; braincase width 45–
sister to the crinifrons/gongshanensis group. 53 mm, mean 47 mm (Amato et al., 1999; Rabinowitz
S Thailand, from about 9° N to more than 14° N. et al., 1999).
A phylogenetic tree based on three mitochondrial
“New Muntjac” Group sequences (Amato et al., 1999) showed this species as
Muntjac of varying size, of unknown karyotype. closely related to M. truongsonensis, but somewhat dif-
Amato et al. (2000) found that, at least on mtDNA ferent; these two belonged in a clade with M. roose-
sequencing, the species of this group do indeed form veltorum, and the next closest species was M. vuquan-
a monophyletic clade. gensis. These four species formed a clade opposed to
one containing the M. muntjak group plus M. feae, M.
Muntiacus vuquangensis (Tuoc et al., 1994) gongshanensis, and M. crinifrons; sister to all these spe-
giant muntjac cies was M. reevesi, although the bootstrap value of
the non-reevesi clade was only 35%.
1994 Megamuntiacus vuquangensis Tuoc et al.
Notes and descriptions of this species were given by Muntiacus truongsonensis (Giao et al., 1998t)
Schaller & Vrba (1996) and Groves & Schaller (2000).
1997 Caninmuntiacus truongsonensis Giao et al. Hien district,
Largest species of muntjac; only species with the
W Quang Nam Province, Vietnam.
antlers considerably longer than the pedicels.
Its distribution was mapped by Groves & Schaller A small, black muntjac; tail short, fairly flat, with long,
(2000). white lateral hairs; pedicels shorter and narrower
Though placed in a separate genus, Megamuntia- than the M. muntjak group; antlers very short and un-
cus, when described, Amato et al. (2000), on the basis branched; female canines almost as long as those of
of mtDNA, found it to belong in a clade with the the males.
other Indochinese species, but it was their sister Because all available cranial material was based
group. This result was somewhat surprising, given on trophy skulls, it was not possible to mea sure —-1
the strong differences between M. vuquangensis and skull length; braincase width averages about 47 mm. —0
349-47558_ch01_1P.indd 89 5/31/11 7:05 PM

90 Artiodactyla
Antler pedicels average 36.7 mm, shorter than any row; frontal tuft absent; mostly reddish yellow, with
other species except M. putaoensis; antler beam aver- some gray overtones; head more reddish than the
ages 10.9 mm, comparable only to M. crinifrons or M. body, relatively large for the body size. In the females,
putaoensis. the frontal region is black, this zone narrowing be-
Its known distribution was mapped by Groves & tween the ears and continuing as a nuchal stripe; oc-
Schaller (2000); it appears to be restricted to the S part ciput orange. In the males, the frontal region bright
of the Annamite range. orange, edged with black stripes, continuing onto the
According to Amato et al. (2000), M. truongsonensis pedicels; orange color extending onto the occiput and
is sister to M. putaoensis. In Rabinowitz et al.’s (1999) the ears; nuchal stripe not connected to the head
table 3, the only significant difference between this markings.
species and M. putaoensis would seem to be the great- China, from about the Yangtze River S to Guang-
est width across the nasals, 22 mm versus 16 ±0.15 mm dong, including Taiwan. Groves & Grubb (1990) re-
in M. putaoensis. marked on how poorly the Taiwan population was
differentiated from any on the mainland.
Muntiacus puhoatensis Chau, 1997
Muntiacus atherodes Groves & Grubb, 1982
1997 Muntiacus puhoatensis Chau. Puhoat area, Que Phong
Bornean yellow muntjac
district, Nghe An Province, Vietnam.
1982 Muntiacus atherodes Groves & Grubb. Near Forest
The diagnosis stated that it is a very small species
Camp 1, Cocoa Research Station, Tawau, Sabah.
of Muntiacus. Weight estimated at 8–15 kg (12 kg on
average). Antlers short (25–26 mm) and unbranched. Very small; light orange-yellow in color, with a dark,
Horn pedicels short (ca. 27–28 mm) and parallel; pedi- diff use dorsal stripe; no frontal tuft; frontal region
cel with a diameter of 7– 8 mm, and covered with blackish, extending onto the inner halves of the ped-
long, thick, yellow and dark brown hair; underparts icels in the males, and continuous with the nuchal
and undertail mixed brown and white. stripe, not divided into separate dark stripes; occiput
The habitat is said to be closed-canopy evergreen not contrastingly orange-colored; antlers tiny, spike-
forest, on steep slopes above about 700 m. like, on very short, slender pedicels. Antlers appar-
ently not shedding, as a rule; very few specimens with
Muntiacus reevesi Group burrs (indicating that the antlers have been shed)
Small muntjac of high chromosome number (where known.
known). Borneo.
On this species, and on the history of knowledge
Muntiacus reevesi (Ogilby, 1839) of the sympatry of the two species in Borneo, see
Chinese muntjac Groves & Grubb (1982).
1839 Cervus reevesi Ogilby. Canton.
Muntiacus montanus Robinson & Kloss, 1918
1871 Cervulus lachrymans Milne-Edwards. Baoxing.
1872 Cervulus sclateri Swinhoe. Ningpo. 1918 Muntiacus muntjak montanus Robinson & Kloss. S
1875 Cervulus micrurus Sclater. Taiwan. Kering, G[unung or Mt.] Kerinci, 2200 m.
1905 Cervulus sinensis Hillzheimer. Hwai Mountains,
Very small in size; greatest skull length 179–198 mm;
Anhui.
antlers usually unbranched, usually not shed, short,
1906 Cervulus reevesi pingshianicus Hillzheimer. Pingshiang,
<100 mm; pedicels very long, 96–102 mm, always con-
C China.
siderably longer than the antlers. Color very dark, a
1910 Cervulus bridgemani Lydekker. Hwai Mountains.
dark chestnut, with blackish speckles; midback dark-
1914 Muntiacus lachrymans teesdalei Wroughton. Tatung,
ened; forehead, pedicels, occiput, and bases of the ears
Yangtze Valley.
orange-brown, the rest of the face brown-gray; backs
2n = 46 in both sexes. of the ears dark; sides and shoulders brown, becom-
Size small; antlers short, 40– 80 mm, often un- ing dark brown on the shanks; throat olive-buff, the
branched, but irregularly shed; pedicels long, broad; rest of the underside reddish white (extending up onto
preorbital fossa very large, often reaching back to the the lower flanks), changing to whitish on the groin;
-1— orbital margin; preorbital fissure very long and nar- feet white, this tone going up the front of the lower
0—
349-47558_ch01_1P.indd 90 5/31/11 7:05 PM

cerv idae 91
half of the limbs; tail blackish brown above; stripe on

the pedicels thick, black. Dama dama (Linnaeus, 1758)
Highlands of Sumatra, from Kerinci N into Aceh. common fallow- deer
Groves & Grubb (1990) considered, and then ten- Populations of this species from Rhodes and from
tatively rejected, the idea that this could be a distinct Turkey (the latter being the last remaining popula-
species, analogous to M. atherodes of Borneo. Recon- tion of the species in its original homeland) form dis-
sideration, and inspection by both of us of photos of tinct clades, with surprisingly high heterozygosity
live animals (from photo traps), leads us to defi ni- (Masseti et al., 2008). Dama schaeferi Hillzheimer,
tively place it as a distinct species, but we still do not 1926, described from N Africa, was actually from It-
know whether it is related to M. atherodes, or M. munt- aly, and it was an ordinary fallow deer (Kock, 2000).
jak, or perhaps occupies an isolated position.
Dama mesopotamica (Brooke, 1875)
Tribe Cervini Goldfuss, 1820 Mesopotamian fallow- deer
The phylogeny of the Cervini has been most recently On this species, see the Wikipedia article (http://en
explored by Pitra et al. (2004). Certainly four genera .wikipedia.org/wiki/Persian _fallow_deer).
(Axis, Rucervus, Dama, and Cervus) should be recog-
nized in the living fauna, but the recent discovery of Axis Hamilton Smith, 1827
a 7- to 9-million-year-old muntjac (Dong et al., 2004) The slender antlers are built on a plesiomorphic
indicates that the calibration chosen by Pitra et al. three-point plan; the anterior ends of the nasal bones
(2004) for their molecular clock—just 7 Ma for the are bifid, the median prong being equal to or longer
muntjac-cervin split—has evidently been set too late, than the lateral one, so that together they form a con-
and we here propose to recognize two further genera, cave anterior margin; the posterior ends form a shal-
Panolia and Elaphurus. low, blunt wedge into the frontals. There are no up-
per canines; and the lower central incisors are very
Dama Frisch, 1775 wide, their width exceeding the combined width of
fallow- deer the two lateral incisors plus the canine.
This genus is immediately distinguishable by the ten- Pitra et al. (2004) found Axis to be the sister ge-
dency to palmation in the antlers; the strongly marked nus to Rucervus, from which it separated some 5 Ma
pygal (rump) bands; the presence of white spots dur- (but, in our estimation, these dates should be pushed
ing all seasons; and the pendulous penis, whose pre- back considerably; see above); their common clade
puce is not closely adherent to the body wall. The na- separated from the Cervus/Dama clade more than
sal bones resemble those of Rucervus. There are no 6 Ma.
upper canines (except as an anomaly). The lower cen-
tral incisors are widened, as in Axis. Axis axis Group
Dama is closer to Cervus than are Rucervus and Axis axis (Erxleben, 1777)
Axis, according to Pitra et al. (2004), who calculated chital, axis, spotted deer
the date of its separation from the Cervus group at over
For the full synonymy of this species, see Groves (2003).
5 Ma. It contains two living species, the European
fallow-deer (D. dama) and the Persian one (D. mesopo- No differences seem to be apparent between Sri
tamica); these are quite distinct, despite the urge of Lankan and mainland populations.
many authors to make them subspecies of a single spe-
cies. The fossil record of the genus does not begin until Axis porcinus Group
the Middle Pleistocene, with D. clactoniana, of which In Pitra et al.’s (2004) analysis, Axis porcinus was placed
the modern D. mesopotamica is essentially a size- in a clade with Cervus timorensis. This position is un-
reduced version, although Pfeffer (1999) (plausibly, in likely, and doubtless results from a misidentification,
our view) also ascribed some Late Pliocene fossils to as argued by Gilbert et al. (2006), who found, as would
the Dama stem. have been expected, that this group sits alongside A.
Mitochondrial D-loop sequence phylogeny axis.
showed that D. mesopotamica and D. dama differ dras- Cranial measurements of the described taxa are
tically (Masseti et al., 2008); the authors estimated a compared in table 17. A. annamiticus is the largest
divergence time of over 400,000 years. species, and A. kuhlii is the smallest (but there is a —-1
—0
349-47558_ch01_1P.indd 91 5/31/11 7:05 PM

92 Artiodactyla
Table 17 Univariate statistics for members of the Axis Axis kuhlii (Müller & Schlegel, 1845)
porcinus group Bawean deer
Notes on A. kuhlii were provided by van Bemmel
Gt l Preorb Biorb Braincase (1948), who was the fi rst to seriously support its close
affi nity with A. porcinus.
porcinus
Unlike A. porcinus, the fawns are, at most, only
Mean 230.50 121.60 104.00 69.00
N 6 5 6 5 faintly spotted.
Std dev 7.817 3.647 2.366 4.000
Min 219 118 101 64 Axis calamianensis Heude, 1888
Max 241 126 108 75 Calamian deer
annamiticus Characterized by a shorter tail than other members
Mean 236.50 122.00 107.50 70.25
of the genus; a narrow skull; a relatively large size;
N 2 2 4 4
Std dev 9.192 9.899 3.317 4.193 and long, slender antlers.
Min 230 115 104 64
Max 243 129 112 73 Rucervus Hodgson, 1838
kuhlii swamp deer, barasingha
Mean 223.50 112.50 102.00 70.00 Rucervus contains the swamp deer (traditionally a sin-
N 2 2 3 3
gle species, R. duvauceli) of Nepal and India, and the
Std dev 14.849 4.950 3.606 6.000
Min 213 109 98 64 presumed-extinct Schomburgk’s deer (R. schomburgki)
Max 234 116 105 76 of Thailand. Pitra et al. (2004) extracted DNA from a
calamianensis Schomburgk’s deer specimen and showed that it is in-
Mean 229.50 120.50 97.00 63.00 deed the sister species of the swamp deer, not the rep-
N 2 2 2 4 resentative of a separate genus (or subgenus) Thaocer-
Std dev 2.121 .707 .000 2.449
vus, as had sometimes been thought. They found that
Min 228 120 97 60
Max 231 121 97 65 Rucervus is most closely related (if distantly) to Axis.
Cranially, Rucervus resembles Dama— and is read-
ily distinguished from the other Cervini—by the
great deal of individual variation in size). A. kuhlii form of the nasofrontal suture: the nasal bones to-
has a very broad skull, and A. calamianensis a very gether make a deep, acute-angled wedge back into
narrow one. the frontals. At their free ends, each has two prongs,
a very long lateral one and a rudimentary median
Axis porcinus (Zimmermann, 1780) one, as in Axis. There are no upper canines.
Indian hog- deer
From Sind E into Nepal and NE India; Sri Lanka, Rucervus duvaucelii (G. Cuvier, 1823)
where it may be introduced. western swamp deer, brasingha
Data on growth, adult body measurements, and
1823 Cervus duvaucelii G. Cuvier. Kumaun.
antlers from the introduced population on Sunday
For the full synonymy of this species, see Groves (1983c).
Island, off Wilsons Promontory (Victoria, Australia),
were provided by Presidente & Draisma (1980). Adult Nasals short, relative to the snout length; rostrum not
males weighed approximately 50 kg, females 35 kg. deep; antlers long, slender, and not compressed or pal-
Antlers achieve full size, at a mean of about 300 mm, mated; little or no sexual dimorphism in size. Tail rela-
around 4 years of age, about two years before the tively long and slim, with prominent white hairs on
body-weight plateau is reached. the underside; ears very large and rounded, with thick,
white hair on the inside; in moult, the white spots in
Axis annamiticus (Heude, 1888) the dorsal region very prominent. Feet splayed, with
Indochina hog- deer bare “heels.”
Axis porcinus and A. annamiticus are cranially very
similar, the latter being very slightly larger, but exter- Rucervus ranjitsinhi (Groves, 1983)
nally they are very different. A. annamiticus is much eastern swamp deer
brighter; more ochery, less gray and buff y; and not
1983 Cervus duvaucelii ranjitsinhi Groves. Guwahati,
-1— speckled (this is due to the fainter banding on the
Assam.
0— hairs). In these features, it resembles A. calamianensis.
349-47558_ch01_1P.indd 92 5/31/11 7:05 PM

cerv idae 93
Elongated nasals; snout short, deep; maxilla narrow; tween the dryland-living thamin of Burma and the
antlers short, thick, branching low down, with an critically endangered sangai, confi ned to the float-
especially shortened anterior branch; antlers some- ing reed beds (phumdi) of Logtak Lake in Manipur.
what compressed, tending to be palmated. Heavily Groves (2006a) suggested that Eld’s deer be reexam-
built; females notably smaller than the males; ears ined, with a view to ascertaining whether they
small, pointed, with very little white hair on the in- ought to be reclassified into two or even three dis-
side; tail short; in moult, the white spots in the dor- tinct species.
sal region less prominent. Feet splayed, with bare While there is a great deal of variability, the skulls
“heels.” of P. thamin have a slight tendency to be narrower
than those of the other two taxa.
Rucervus branderi Pocock, 1943 Table 18 shows that P. siamensis, on average, is
hard-ground barasingha considerably smaller than the other two species; the
table also confi rms the relative narrowness of P. sia-
1943 Rucervus duvauceli branderi Pocock. Mandla, C India.
mensis and P. thamin.
Distinguished from R. duvaucelii and R. ranjitsinhi
by the “well-knit” feet, with hairy pasterns. Size Panolia eldii (McClelland, 1842)
smaller; nasals long; snout short, but the nose not deep; sangai, Manipur Eld’s deer
maxilla somewhat broadened; antlers more as in R. Large; short antlers; bare pasterns (swamp-living).
duvaucelii, but extremely long, many branched, with Although still exceedingly rare, P. eldii has un-
a long brow tine; branching high up the beam; ante- dergone a dramatic improvement in its fortunes
rior branch especially long. since a low point in 1975, when only 14 individuals
were counted; in the 2003 survey, the number was 180
Rucervus schomburgki (Blyth, 1863) (Khaute, 2010).
Schomburgk’s deer
Dark brown; large, multitined antlers. Panolia thamin Thomas, 1918
A presumably extinct deer, known only from the thamin
swampy country of C Thailand. Narrow-skulled, compared with other taxa; large in
size; dark brown in color.
Panolia Gray, 1843 Burma and W Thailand.
Eld’s deer, brow-antlered deer
The Eld’s deer complex constitutes a separate genus,
Panolia, whose affi nities, based on mtDNA, are not Table 18 Univariate statistics for taxa
with Rucervus (to which it has usually been thought of Panolia (adults)
to be related) but with Elaphurus, from which it sepa-
rated some 3.5 Ma, according to Pitra et al. (2004). Gt l Biorb
More likely, this separation would have been consid-
erably earlier (see above). It is cranially distinct from eldi
Mean 329.00 131.33
other Cervini, especially as the posterior ends of the
N 3 3
nasal bones form only a blunt, shallow wedge into Std dev 12.124 7.371
the frontals; whereas anteriorly, each nasal is single- Min 316 123
pointed, the points diverging from one another, Max 340 137
leaving a midline, V-shaped gap— the form usually siamensis
seen in Cervus. The antlers of Panolia are highly dis- Mean 305.33 127.67
N 3 3
tinctive, the brow tine and the beam forming a con-
Std dev 17.786 4.041
tinuous, almost unbroken arc. Traditionally only a Min 286 123
single species, Panolia (formerly Cervus or Rucervus) Max 321 130
eldi, has now been recognized. Balakrishnan et al. thamin
(2003) showed, however, that on mtDNA there is a Mean 333.07 127.82
deep split between a W clade (from Burma and Ma- N 15 17
Std dev 12.372 7.376
nipur) and an E clade (from the Indochinese region
Min 310 117
and Hainan). There are also considerable morpho- Max 353 140 —-1
logical differences between the two, as well as be- —0
349-47558_ch01_1P.indd 93 5/31/11 7:05 PM

94 Artiodactyla
Panolia siamensis (Lydekker, 1915) gued by Pitra et al. (2004), specializations of the dis-
eastern Eld’s deer play organs (antlers, mane, rump-patch, voice) are
Smaller than other taxa; lighter in color; spots visible indicators of habitat and reproductive seasonality,
along the median dorsal region. rather than of phylogenetic affi nity. The three well-
E Thailand, Cambodia, S Laos, perhaps Vietnam, separated clades—whose separation, according to Pitra
Hainan. et al. (2004), took place about 3.5 Ma (but see our re-
marks above)— are as follows:
Elaphurus Milne-Edwards, 1866
1. Tropical clade—Cervus timorensis (rusa), and C.
The apparent affi nities of the E Asian genus Elaphurus
unicolor (sambar) and its relatives.
(Père David’s deer) vary according to which system
2. W temperate clade—C. elaphus and its
is being studied. As listed by Meijaard & Groves
relatives. Pitra et al. (2004) suggested that the
(2004b:table 7), its morphological features (except for
geograph ically isolated C Asian C. yarkanden-
its unique antler conformation) generally recall those
sis (including C. bactrianus?) constitutes a
of Cervus; protein electrophoresis and an nDNA se-
distinct species; the spotted form C. maral,
quence (k-casein) similarly align it with Cervus; yet
from Turkey and Iran, and the small, second-
mtDNA puts it in a clade with Panolia. At present, the
arily simplified N African deer (introduced to
best explanation of this is that Elaphurus originates
Corsica and Sardinia), are further candidates
from an ancient hybridization between stem repre-
for species status.
sentatives of Cervus (males) and Panolia (females)
3. E temperate clade—recently surveyed by
(Meijaard & Groves, 2004b; Pitra et al., 2004). Elaphurus
Groves (2006a). This includes C. albirostris
is known as far back as the late Pliocene (E. bifurcatus)
(white-lipped deer), and a subclade contain-
and, according to the molecular clock of Pitra et al.
ing the C. nippon group (sika) and the large
(2004), its separation from Panolia would date from
Sino-Rosso-American deer (wapiti and shou),
about 3.5 Ma, although (as noted above) we would
which form a progressive cline running
recalculate the separation as being earlier. Deer hy-
SW–NE from C. wallichii (Tibetan shou) via
bridize readily, at least within the same genus, and
C. macneilli (Sichuan shou) and C. xanthopygus
it may be that other species will also turn out to be
(izubra, from Primoriye, Manchuria, E
of hybrid origin, but no case at present seems as
Mongolia) to C. canadensis of the C Asian
plausible as that of Père David’s deer. Y-chromosome
mountains and North America. One of the
DNA analysis in the Cervidae would be of enormous
most unexpected, but most consistently
interest.
corroborated, fi ndings of molecular studies on
deer has been that the wapiti and shou are not
Elaphurus davidianus Milne-Edwards, 1866
E subspecies of C. elaphus, as had always been
Père David’s deer
assumed, but form an entirely separate clade,
Cao (1978) analyzed Holocene sites where remains of
to which sika and white-lipped deer also
this species have been found. His map shows a former
belong.
distribution through what is effectively the coastal
plain of N and C China, from the latitude of Beijing to The fi rst wide-scale analysis of the genus using
Ningbo, somewhat S of the (present-day) mouth of mitochondrial control region was by Randi et al. (2001).
the Yangtze. He concluded that Père David’s deer were The maximum likelihood tree was as follows:
domesticated during the Zhou dynasty (1045–256 BC),
(alfredi ((elaphus—Europe) ((timorensis, unicolor)
but still existed in the wild in reduced abundance dur-
(nippon, canadensis))).
ing the Han dynasty (206 BC–220 AD), after which
they became extinct in the wild, surviving only in the This appears to be the first recognition that taxa
Imperial Hunting Park. hitherto included within Cervus elaphus might not be
monophyletic. For convenience, and because we sug-
Cervus Linnaeus, 1758 gest it may be more complex than simple nonmono-
The genus Cervus is cranially most similar to Pano- phyly, we continue to refer to a catchall “Cervus elaphus
lia, except that the posterior nasal bones form a dis- group,” which includes red deer, shou, and wapiti.
tinct, but relatively short, point. The males possess What follows is largely based on our independently
-1— small canines. Even shorn of some of its erstwhile written summaries (Grubb, 2004b; Groves, 2006a),
0— components, the genus is large and unwieldy. As ar- plus some new research.
349-47558_ch01_1P.indd 94 5/31/11 7:05 PM

cerv idae 95
Cranial material was studied by one or the other

Cervus elaphus Group of us in Philadelphia, London, and the Oswald collec-
Perhaps the most easily identified feature that varies tion near Munich. Our studies were heavily weighted
among members of this group is the rump-patch. toward the more enigmatic taxa, especially those in E
Broadly speaking, there are three types: Asia (the shou group, C. alashanicus, C. xanthopygus);
the largest sample of group 1 specimens was that in
1. European / W Asian type—true red deer.
the Oswald collection from Spain. Our multivariate
This includes Yarkand and Ira nian deer.
analysis found a broad division between those from E
Red-brown / white rump-patch. Tail length in
Europe and Iran, on the one hand, and those from W
most populations 190–220 mm; hair length
Europe plus, unexpectedly, C Asia, on the other.
5– 8 mm; tail length without hair 14–16 mm;
This, in par ticu lar, separates Cervus elaphus from Cer-
tail circumference with hair 13.5–16.5 mm. In
vus pannoniensis (see below).
Iran these figures are less; one specimen has
When the two largest samples, from Spain and
measurements, respectively, of 150 mm, 4 mm,
Hungary, were simply contrasted in a two-way analy-
11 mm, and 13 mm.
sis, there was a slight overlap (1 of 21 specimens from
2. Tibetan group— commonly called shou.
Spain is misclassified with the 8 specimens from Hun-
White rump-patch, with a 2–5 cm broad dark
gary). This again separated Cervus pannoniensis from
rim. Tail much shorter, rather thinner than in
W European red deer, although it says nothing about
group 1. Tail measurements (for C. macneilli
whether the putative taxon C. hispanicus is distinct or
and C. alashanicus) are length 100–120 mm,
not, as no other W samples were available.
hair 3— 4 mm, length without hair 7– 8 mm,
Among group 2 specimens (shou), we fi nd almost
circumference 12.5–13.5 mm.
complete separation: the C. hanglu, C. wallichii, and
3. E Asian / American group—wapiti. Mostly
so-called kansuensis (= C. macneilli; see below) sam-
with a light brown / cream, two-colored
ples are all discrete, mainly on the basis of overall
rump-patch (Altai, Tianshan to America); but
size and nasal length.
the izubra, C. xanthopygus, has a uniform
Next, we examined the position of C. alashanicus:
golden-yellow rump-patch. Tail intermediate
is it more like a shou or a wapiti? (Wapiti are repre-
in length, but in hair length and circumference
sented in our sample by C. xanthopygus, our only rea-
more like group 1. Tail measurements are
sonable wapiti sample). As different variables were
length 130–160 mm, hair 4–11 mm, length
available for different samples, fi rst we used those
without hair 3.5–11 mm, circumference
that would maximize the representation of C. kan-
14–16.5 mm (including C. xanthopygus). Cap
suensis. The three are largely discrete, mainly differ-
et al. (2008) continued to place this species in a
ing in skull breadth and, to a lesser extent, in nasal
clade with C. elaphus, which excluded C.
length.
nippon, although the bootstrap support was
Then we selected variables to maximize the repre-
only 67%, much lower than for the other
sentation of C. alashanicus. Again, all three groups are
cervid clades in the authors’ male-vocalization-
separate, on the basis of size, palate length, and palate
based tree. Schonewald (1994) had previously
breadth. We take this as evidence that C. alashani-
argued strongly for the status of wapiti as a
cus is neither a shou nor a wapiti, but different from
species distinct from European red deer, citing
either.
voice; skull size and shape (premaxilla length,
Finally, we compared different samples of wapiti
interorbital width, palate depth, nasal
(mainly Asian; we had only a single skull of an Ameri-
breadth); the pattern of sexual dimorphism;
can wapiti) with each other. Our samples of C. xantho-
and the apparent reproductive barrier between
pygus and C. asiaticus overlap, whereas C. songaricus
them in New Zealand, where they were
and C. canadensis are distinct, based on a contrast be-
introduced separately and have now met.
tween the skull breadth and the nasal breadth.
Antlers tend to be characteristic for each of these Three phylogenetic studies based on mtDNA ex-
three groups, but they are more variable than the ist. In the fi rst, Li M. et al. (1998) studied five Chinese
rump-patches. In almost all taxa, one can fi nd indi- taxa (C. songaricus, C. macneilli, C. xanthopygus, C. kan-
viduals with one antler elaphoid and the other wap- suensis, and C. alashanicus), using cytochrome b, and
itoid. Geist (1998) also gave behavioral characters dis- found considerable levels of divergence. Although —-1
tinguishing the major species-groups. bootstrap values were relatively low, they were not, —0
349-47558_ch01_1P.indd 95 5/31/11 7:05 PM

96 Artiodactyla
apparently, significantly different from a five-way split, We now present our taxonomic arrangement,
although they favored the view that C. songaricus di- in which craniometric and molecular results are in
verged fi rst from the other four; they put the times of agreement.
divergence around one-half to three-quarters Ma.
Liu et al. (2003), studying a different mix of Chi- Cervus elaphus Linnaeus, 1758
nese taxa, again using the cytochrome b gene, found West Eu ropean red deer
the following: We have insufficient evidence to assess the valid-
ity of Swedish, Norwegian, British, Spanish, and Eu-
((sibiricus, songaricus) (xanthopygus (wallichii
ropean continental taxa within this species. Geist
(macneilli, kansuensis)))).
(1998) argued for the distinctness of C. e. hispanicus
This was on a 5' partial sequence; but the whole se- Hilzheimer, 1909, from S Spain, both in pelage and
quence, available for fewer samples, reversed the or- in a 90° inward direction of the fi fth tine, associ-
der of the fi nal clade, placing C. kansuensis as sister to ated with the curvature of the whole antler (resem-
C. macneilli plus C. wallichii (but the bootstrap values bling C. e. angulatus, a 250,000 BP fossil from Ger-
were not high in either case). many). For Pitra et al. (2004), however, the Spanish
Ludt et al. (2004) sequenced cytochrome b in Cer- red deer is an integral member of the W Eu ropean
vus elaphus (as traditionally defi ned), fi nding that they group.
divide into two extremely divergent clades, a W and The last remaining indigenous red deer of Italy,
an E. They are not each other’s closest relatives, with from Mesola Wood in the Po delta, were described by
the E clade being part of a major branch with the sika Mattioli et al. (2003). Their affi nities to other living
group and Cervus albirostris, exactly as Randi et al. red deer are unclear, but their value for understanding
(2001) had found earlier. Divisions within the two red deer taxonomy and phylogeny is considerable.
“elaphus” clades were as follows:
Cervus pannoniensis Banwell, 1997
1. The W clade divides into a number of distinct
East Eu ropean red deer
subclades. The basic division is between the C
Botezat (1903) named two kinds of red deer in the
Asian group (Cervus bactrianus, C. yarkandensis)
Carpathians of Bukowena (on the Ukraine/Romania
and the European group. The European group
border): C. campestris (a large gray deer from the up-
in turn divides into
land plains) and C. montanus (a smaller dark deer of
—the “African group” (C. barbarus, C. corsicanus),
the slopes and valley bottoms). Both names are un-
—Middle East (C. maral, from Turkey and
available; as pointed out by Grubb (2000b), they are
Iran),
both nomina nuda, as well as being unavailable be-
—Balkans group (Austria, Bulgaria, Hungary,
cause of the previously published Cervus macrotis var.
Romania, and former Yugoslavia, i.e.,
montanus (Caton, 1881), a synonym of Odocoileus hemio-
C. pannoniensis), and
nus, and Cervus campestris F. Cuvier, 1817, a synonym
—W Europe (Sweden, Norway, Spain, France,
of Odocoileus cariacou, respectively. Dobroruka (1960)
Britain, Germany, Poland, and Crimea, i.e.,
was able to distinguish a darker, maned deer in the W
C. elaphus).
Carpathians (which he called C. hippelaphus) and a
2. The E clade has three major divisions, namely
grayer, maneless deer farther E (which he called C.
—E Asia (C. xanthopygus, C. alashanicus),
montanus), but, unlike Botezat, he did not specify habi-
— S Asia (C. wallichii, C. macneilli, C. kansuensis),
tat differences. Could Botezat’s and Dobroruka’s two
and
types be W and E European forms, meeting in the
—N Asia / America (C. songaricus, C. sibiricus,
Carpathians?
C. canadensis).
Hartl et al. (1995) and Ludt et al. (2004) distin-
Bootstrap values are 95%–100% for all of these clades. guished the deer of W Eu rope (including Poland)
The Balkan and Middle Eastern groups cluster to- from those of Austria, Bulgaria, Hungary, Yugo-
gether with 90% bootstrap confidence, and the slavia, Romania, and Turkish Thrace. For the latter,
non–N African group within the European subclade Banwell (1997) provided the new name C. pannonien-
clusters together with 80% bootstrap support. Both sis, giving a detailed description, and Grubb (2004b)
Randi et al. (2001) and Ludt et al. (2004) concluded designated a skull with antlers in Château Cham-
-1— that (at least?) two species should be recognized, but bord (from the Zala district, S Hungary) as the
0— many fewer subspecies. lectotype.
349-47558_ch01_1P.indd 96 5/31/11 7:05 PM

cerv idae 97
form; they live in a similar semidesert habitat in sax-

Cervus maral Gray, 1850 aul and tamarisk forests, and in riparian poplar woods
Turkish/Persian red deer and reed beds.
The Syr Darya and Amu Darya basins.
Type locality: N Persia.
Pitra et al. (2004) confi rmed that C. bactrianus and
Geist (1998), O’Gara (2002), and Oswald (2002) thought C. yarkandensis are, in fact, distinct— becoming sepa-
that the Balkan and Middle Eastern deer are the same, rated about 1 Ma— and placed the separation of their
but, in fact, they differ genetically, and C. maral and joint lineage from that of other “true red deer” at
C. pannoniensis, though related, are quite distinct in nearly 3 Ma.
color, spotting, mane, and antlers (see Banwell, 1997, In a female in the Cologne Zoo, a karyotype of
1998, 2009). In the scheme of Pitra et al. (2004), C. 2n = 68 was recorded by Schreiber (1994), who con-
maral separated from the European and N African trasted it with karyotypes of 2n = 66 and 67 in what
clades around 1.5 Ma. were reputed to be the same taxon in the Beijing Zoo;
C. e. brauneri, described from Crimea, is an enig- (given that they were in a Chinese zoo, these latter
matic taxon. It is a very small form, according to Hept- animals may have been C. yarkandensis.
ner et al. (1961), and Ludt et al. (2004) found Crimean
deer to have W Eu ropean affi nities. In the early to Cervus hanglu Wagner, 1844
middle 20th century, N Caucasus deer were intro- Kashmir stag
duced into Crimea, and the present status of indige- Dark liver-brown in winter, with the legs and the chest
nous deer there is unclear. rather darker, and the face, the neck, and the back
lighter; lighter in summer, sharply contrasting with
Cervus corsicanus Erxleben, 1777 the dark chest and the dark limbs. Like C. wallichii,
Corsico-Sardinian / North African red deer, Barbary stag the summer coat is simply a faded version of the winter
The Corsico-Sardinian deer is an island dwarf with coat, and their antlers are fairly similar; however, C.
reduced antlers, almost certainly derived from an hanglu is smaller than C. wallichii. Only a short mane.
Early Holocene human introduction. The N African Rump-patch narrow, hardly extending beyond the tail
C. barbarus Bennett, 1833, is very close. While not a root, invaded dorsally by a wedge of body color some-
dwarf, C. corsicanus has short tines and no bez tine in times extending onto the tail, and bordered laterally
95% of the cases; molecularly, it is closely related to by a black band confi ned to the lower part. Chin, lips,
the Corsico-Sardinian deer (Pitra et al., 2004), and N and inside of the ears white; belly, groin, and inner sur-
Africa may thus be the source of the island popula- faces of the hindlegs whitish. Metatarsal gland creamy
tions, if the Barbary stag is not itself introduced from to light red. Dark curly hair between the antler pedi-
somewhere else (the Mesola deer [see above] should cels. Only a few pairs of antlers (9 out of 71) have 13 or
be carefully looked at in this regard). Pitra et al.’s (2004) more tines, 34/71 have 11 or 12 tines, and 28/71 have
molecular clock puts their joint separation from the 10 or fewer. The rutting call, in the words of Geist
European group at a little over 1 Ma. In one of the only (1998), begins like red deer, with a closed-lips roar,
major departures from the molecular tree, Cap et al. and then becomes a wapiti-like bugle as the lips are
(2008) found that this species, according to the male opened out.
vocalizations, is sister to a clade containing C. elaphus, Vale of Kashmir and neighboring regions.
C. canadensis, and C. nippon. While this species is dark, with a conservative
rump-patch like C. macneilli, its mtDNA lineage is W
Cervus yarkandensis Blanford, 1892 (specifically, close to C. yarkandensis).
Yarkand stag
A distinctive deer, with a straight antler beam and a Cervus wallichii G. Cuvier, 1823
simple, three-tined crown (see Dolan, 1988; Banwell, Tibetan shou
2009). Yellowish brown in winter, dark gray-brown in sum-
Tarim Basin in Xinjiang. mer, when the winter coat has faded (there is appar-
ently no special summer coat). Rump-patch white, usu-
Cervus bactrianus Lydekker, 1900 ally small, but extending well up onto the croup, often
Bactrian stag divided by a dark line, its border poorly marked (larger
C. bactrianus is described as very similar to C. yarkan- rump-patches appear to predominate at the E and W —-1
densis, with similar antlers and the same rump-patch ends of the distribution, and smaller ones toward the —0
349-47558_ch01_1P.indd 97 5/31/11 7:05 PM

98 Artiodactyla
middle). Mane poor. Tail very short. Antlers simple, C. kansuensis lives at somewhat lower altitudes
usually 5-pointed, angulated at the middle of the beam; (2500–3000 m).
of 14 pairs of antlers, only 2 had more than 10 tines. Dolan (1988) kept C. kansuensis separate from C.
Nasal region broad, presumably reflecting an enlarged macneilli; Geist (1998) argued that they are actually
nasal cavity. synonymous, and that what Dolan had illustrated as
S Tibet, from near Lake Mansarowar, (formerly) S C. macneilli (from the Chengdu Zoo) were actually a
into Sikkim, and E to the Tsari district. group of C. wallichii with a dark median line on the
tail, as they differ from C. macneilli in the much greater
Cervus macneilli Lydekker, 1909 dorsal extension of the rump-patch and the absence
Sichuan shou of a dark dorsal rim. The Chengdu Zoo group were
2n = 68. from E Nangjing Shan, on the Sichuan side, S of Batang
Stags in winter pale creamy gray, with reddish and N of Chubalung (Dolan, pers. comm. to PG), and
tones and a broad, reddish dorsal stripe; hinds redder. so considerably to the S of any C. macneilli; if Geist’s
In summer, somewhat darker and redder. Rump- argument is correct, which seems plausible, then this
patch white, less extensive, divided by a black line represents a considerable E extension of the range of
sometimes extending onto the upper tail surface; C. wallichii.
characteristic black border (4–5 cm wide) to the rump- According to Geist (1998), Dolan’s “kansuensis” is
patch, expanded dorsally into a broad black patch the true C. macneilli (while G. M. Allen’s description
extending up onto the croup. Tail somewhat longer of “kansuensis” actually refers to C. alashanicus).
than in C. wallichii. Antlers usually simple in structure, The molecular analysis of Pitra et al. (2004) placed
and somewhat angulated in the middle of the beam, macneilli/kansuensis and wallichii close together, sug-
but often less markedly so than in C. wallichii. Antlers gesting that their lineage separated from that of their
with disproportionately small third tines, like the closest relatives, C. xanthopygus and C. canadensis,
more wide-spreading antlers of C. wallichii, but of a six- somewhat over 1 Ma.
point plan, unlike wallichii. Oswald (pers. comm. to
PG) has found that 45% of the antlers have elaphoid Cervus alashanicus Bobrinskii & Flerov, 1935
crowns. Massive hoofs. Alashan stag
Sichuan and Gansu, at 3000– 4500 m.
Oswald (pers. comm. to PG) identified the type locality of
In Gansu, a separate taxon has been described,
this taxon as Helanshan, Ninxia.
Cervus kansuensis Pocock, 1912. We have considered
in some detail whether this is really a valid taxon, and Our description of the species is largely from speci-
concluded that it is probably not, for the reasons mens in the Oswald collection (agreeing with the type
given below. description). Tail and large rump-patch brownish white
The type of C. macneilli is from Litang (30° 01' N, or russet-orange, with a fairly wide, indistinct, brown-
100° 16' E). We have seen specimens (N = 19) from ish stripe approaching the root of the tail; almost no
Yekundo (33° 01' N, 96° 44' E), Gur La Pass (Yangtze- darkening along the edge in the type specimen, but
Mekong watershed, 31° 50' N, 96° 40' E), Dzogchen mostly a distinct, dark brown margin (almost inter-
Gomba (32° 07' N, 98° 54' E), and Xiao Sumang (33° rupted anteriorly on each side), continuous with a
30' N, 97° E). well-marked median stripe reaching to the tail base.
The type of C. kansuensis is from Taochow (34° 41' Dark spinal stripe; mane whitish. Antlers with a five-
N, 103° 21' 52' E) in Gansu. We have seen specimens tine plan.
like it from the Peling Mountains; Longshou Shan, There may be intermediate populations between
Gansu; Ikhe-Gol (36° 10' N, 97° 30' E); Qinghai Nan this species and C. macneilli. Six specimens from Daq-
Shan (37° N, 100° E); and the N Tatung Range (37° ing Shan, Inner Mongolia, show the female and im-
15' N, 100° E). There are differences in color, at least mature male pelage; they are identical with female C.
on average, from “true” C. macneilli, but not in the macneilli, except that the rump-patch is orange and
rump-patch or antlers, which are the same. Oswald the amount of dark brown margin varies from none
(pers. comm. to PG) has found deer of this type from to broad and complete. Adult males from this locality
the E slopes of the Tibetan plateau (Qilian Shan) up to have the dark margin incomplete medially, with a
the mountain chain bordering directly onto the separate median dark stripe to the base of the tail. A
-1— Alashan Desert; Oswald stated that 40% have an ela- female from River Long Chang, Shanxi, has the black
0— phoid crown, which is similar to “true” C. macneilli. edge almost complete, but thinner in the midline,
349-47558_ch01_1P.indd 98 5/31/11 7:05 PM

cerv idae 99
with a prominent median stripe to the base of the —Hinds dark brown; rump-patch smaller,
tail. light brown, no stripe; neck hair hardly
longer than elsewhere.
Cervus xanthopygus Milne-Edwards, 1867 — Specimens from Trans-Iliisk Alatau, near
Manchurian wapiti the E end of Issik-Kur (syntypes of C.
Stags creamy fawn in winter, with a dorsal stripe; in songaricus); Alexandrovsk Mountains; lower
summer, light yellow-red, with a dark dorsal stripe; Kok-an Valley, E Tianshan; Tekes Valley;
legs uniform dark brown; neck light brown, the hairs Lake Balikol (46° N, 91° E), E of Urumshi.
with reddish tips. Rump-patch reddish yellow, its 2. Altai
shape and size resembling C. canadensis, but with a —Two types of male coloration occur, one like
complete blackish brown, dark border and divided by the American wapiti, the other grayer, more
a thin, reddish brown stripe. Hinds a more uniform reddish on the head, the neck, the belly, and
gray-brown / light sandy; no mane; large, whitish yel- the feet, the head and the neck an intense
low rump-patch, edged with dark brown only on the brown tone, with some reddish color; dark
haunches, with no median line; legs only slightly brown spine stripe only in the shoulder
darker than the trunk, with a narrow dark stripe down region; rump-patch large and, like the tail,
the front surface; neck brown, with yellow hair tips. bright reddish cream, with a dark brown rim
Tail very short. Antlers complex, the beam somewhat above, an almost black stripe on the sides, and
curved, with seven points. a black median stripe reaching partway to the
Yakutia, Amuria, and Manchuria. tail (this seems variable; some have none).
—Females in winter more monochromatic,
Cervus canadensis Erxleben, 1777 with a distinctly rimmed, white or reddish
Asian/American wapiti, elk (North America) rump-patch and a slight darkening along the
Gray-brown or buff in winter; dark brown in sum- spine.
mer. Rump-patch creamy, very extensive, flaring lat- — Specimens from Altai; W Mongolia; deer
erally on the upper part of the rump, then contract- farms; 95 km NE of Ulan Bator; Barguzin
ing to be confi ned to the inner margins of the upper Ridge, Transbaikalia.
thighs; dark border confi ned to the lower part of the
Our conclusion must be that—though they differ
margin (on the thighs). Antlers complex in structure,
on average from each other and (especially those in
the beam curved, P3 posterior to A3, not mesial to it
the Tianshan) from North American wapiti, so possi-
(as in red deer); distal part of the antler flattened.
bly a separate subspecific status for each of the three
Whether the Asian and American populations can
could be supported—they all belong to C. canadensis.
be distinguished has been much discussed. The names
Schonewald (1994) examined differences among
Cervus sibiricus Schreber, 1784, and Cervus songaricus
North American populations, fi nding that they were
Severtzov, 1873, are available for the Altai and Tian-
defi nable, but that this needs more study.
shan populations, respectively, should they be re-
garded as valid taxa. We describe the more poorly
Cervus albirostris Przewalski, 1883
known Asian wapiti as follows:
white-lipped deer
1. Tianshan For the general biology and distribution of this spe-
— Stags with head and neck brownish gray, cies, see Leslie (2009), who assigned it to a separate
with longer hair than the rest of the body, genus, Przewalskium Flerov, 1930.
each hair with black and pale gray-brown This deer is not related to Rusa, as Flerov (1952)
rings; shoulders, back, side, and thighs thought, but is part of the canadensis-nippon clade, from
brownish gray, shaded with yellow, consid- which it separated somewhat over 2.5 Ma, according to
erably lighter than the neck; yellow rump- Pitra et al. (2004).
patch separated from the body color by a
dark gray stripe, sharply marked toward the Cervus nippon Group
tail, and gradually shading off into the gray sika
of the back; tail light yellow generally, with On this group, see Banwell’s (1999) detailed treatment.
a grayish brown line down the middle; There has been considerable interest in the nature
breast, belly, and legs clear dark brown. and frequency of hybridization of C. nippon with native —-1
—0
349-47558_ch01_1P.indd 99 5/31/11 7:05 PM

100 Artiodactyla
C. elaphus in places to which the former has been in- Wu H. et al. (2004) compared the mitochondrial
troduced in Europe (including the British Isles). Per- control region in sika from four regions in China:
haps the most detailed study is that of Goodman et al. Zhejiang, Jiangxi, Sichuan, and “the northeast of
(1999), who concluded that they are not, in fact, pan- China,” this last sample being taken from “domestic”
mictic, the degree of introgression being apparently (i.e., deer farm) populations. They fell into two very
less than would be expected from current hybridiza- distinct clusters: one consisting solely of the Zhejiang
tion rates; there is significant linkage disequilibrium sample (N = 12), the other containing all the rest. The
within both populations. second cluster was again divided into two subclus-
In their study of hybridization between red deer ters, one consisting of 7 out of the 11 specimens from
and sika in England, Lowe & Gardiner (1975) studied Jiangxi plus 3 of the “NE” specimens, the other con-
a number of different groups of sika. Specimens from sisting of the remaining 4 from Jiangxi, all of the 7
the mainland (Manchuria and farther S in China) from Sichuan, and the remaining 13 from the “NE.”
always separated very clearly from those from Japan This could be taken to mean that the “domestic” pop-
(including Kerama). In their figure 4, a specimen ulations were derived from several sources, including
from Taiwan assorted more closely with those from Sichuan, Jiangxi, and the indigenous “N” population,
Manchuria. or that all of these populations had spread rather re-
Imaizumi (1970), in the course of describing the cently and had not achieved reciprocal monophyly.
new species Cervus pulchellus, proposed a novel Pitra et al. (2004) dated the separation of sika from
method of differentiating species in this group: he the C. canadensis group at about 2.5 Ma; within the sika,
graphed skull measurements against “temperature they found mainland sika (as represented by C. sichuani-
index” (an index derived from the mean tempera- cus) separating from the island forms around 2 Ma; and
tures of the localities of taxa, modified by altitude). the yesoensis/”centralis” clade (equivalent to the N group
The rationale was that a given species would respond of Nagata et al., 1999) was calculated to separate from
to temperature consistently, whereas different spe- the S clade—containing C. nippon from the S part of the
cies would respond differently. Hence the deer from main Japanese islands and the small island forms (C.
most of the Japa nese islands qualified as one species; mageshimae, C. keramae, C. pulchellus)—about 1.5 Ma.
those from the mainland and Hokkaido as a second; Our craniometric studies rely heavily on the
those from Taiwan as a third; and the new species Heude collection in the Academia Sinica, Beijing, of
from Tsushima, C. pulchellus, as a fourth. which the largest sample is from the Goto Archipel-
Mitochondrial D-loop sequences from sika through- ago. They give the following results: Honshu, Goto,
out the Japanese islands were compared by Nagata et Yakushima, Mageshima, and Tsushima samples dif-
al. (1999). Both the neighbor-joining and the maxi- fer largely by size, especially in nasal length and mas-
mum parsimony methods divided them very strongly toid breadth, and, to a lesser degree, in biorbital
(in neighbor-joining, at >90% bootstrap) into two breadth and palate length (braincase length hardly
groups: N and S. The boundary was around the “neck” differs among the samples). The order, from high to
of land in S Honshu constricted by Biwa Ko; samples low, is N Honshu; Goto / S Honshu; Tsushima; and
from Wadayama, Hyogo Prefecture, belong to the N Yakushima/Mageshima. Another sample ordering,
group, and those from Yamaguchi to the S group. Sev- using nasal length and palate length, separates Tsu-
eral things are interesting about the composition of shima, Yakushima, and Mageshima from Honshu
these two clades. First of all, samples from Hokkaido and (especially) Goto. In the main, the small S Hon-
were nested well within the N clade, not sister to shu sample is close to Goto.
N Honshu. Within the S clade, samples from Tsu- When we turn to a comparison of the mainland
shima assorted with those from Yamaguchi; those (mainly E China) versus Japan, the long facial skeleton
from the Goto Islands, with Kyushu; while those from of the E China sample is distinct from all Japanese sam-
Yakushima and Kerama formed distinct subclades, not, ples, but this is not strongly correlated with the overall
however, very strongly distinguished from the rest. size difference that separates Hokkaido / N Honshu
Randi et al. (2001) found that, in fact, Japa nese and from Goto. The skull of a hybrid between a Goto and a
mainland plus Taiwanese sika did not form a clade Hokkaido deer (bred by Heude) is intermediate, though
when using partial control region sequences; the closer to Goto, confirming that the differences between
Japa nese clade was sister to C. canadensis, whereas the them do have a genetic basis. Yakushima and Mages-
-1— non-Japanese clade was sister to the combination of hima skulls are closer to those from Honshu, whereas
0— Japan plus C. canadensis. Tsushima is at the edge of the E China range of varia-
349-47558_ch01_1P.indd 100 5/31/11 7:05 PM

cerv idae 101
tion, and seems to be basically a mainland taxon. Sich- —Solo (the enigmatic Sikelaphus soloensis Heude,
uan and Vietnam skulls are both closest to E China. from Jolo in the Philippines!) has the longest
When we look at the skulls of the females, we fi nd (but narrow) nasals, extremely large teeth, and a
the following results: Of the two skulls from Kerama, very wide (yet not absolutely the widest) palate.
one is closest to Goto, while the measurements of the
The suggested species that emerge from these anal-
other actually fall within the Goto range. Two from
yses are as follows:
Taiwan are intermediate between Goto and E China.
Of the two Sichuan skulls, one falls within the E 1. Hokkaido + N Honshu.
China range ( just), while the other is intermediate 2. S Honshu = Goto = Yakushima = Mages-
between E China and the Far East sample. In absolute hima = Kerama—this taxon is absolutely
size, the sequence is Far East, E China, Goto. The smaller than taxon 1 in greatest length and
Taiwan samples have a particularly high biorbital nasal length.
breadth compared with the skull length. 3. Tsushima—slightly larger than S Honshu,
Summarizing all this by sample (table 19 gives the with absolutely short nasals and a short palate,
univariate statistics for male skulls, and table 20 for and narrow across the mastoids.
female skulls): 4. E China— as large as N Honshu, but with
narrow orbits, a wide braincase, and smaller
—Hokkaido does not differ from N Honshu, but
teeth. Vietnam is at the lower end of the range
S Honshu is smaller (absolutely so in greatest
of E China.
length and nasal length). We have here the
5. Far East = Beijing = Shanxi—very large
basis of a division between the N and S parts
compared with E China, with a broad palate,
of the main Japa nese islands, with the division
wide across the condyles and with a wide
cutting across Honshu, exactly where the
interorbital breadth.
DNA data divide.
6. Sichuan—smaller than the Far East, with short
—Goto hardly differs from S Honshu (in variables
nasals; larger on average than E China, but
where the sample size for S Honshu is reason-
with narrow mastoids.
able), and Kerama, Yakushima, and Mages-
7. Solo—in the size range of N Honshu, with
hima generally fall within the range of Goto.
long, narrow nasals, the largest teeth of all,
On this showing, they can all be placed in the
and a relatively wide palate.
S Japan taxon.
—Tsushima is larger in size than Goto, more like Molecular data, as has been discussed above, make
a small N Honshu, but it has absolutely short exactly the same divisions in cases where samples
nasals and a short palate, and narrow mas- have been available.
toids. It is clearly a distinct taxon, though
more of mainland than of Japan affi nity. Cervus nippon Temminck, 1836
—E China skulls are also much bigger than ones
For the full synonymy of this species, see Groves & Smeenk
from Goto, their size being the same as N
(1978).
Honshu / Hokkaido, but long and relatively
narrow. Skulls from the Far East, compared In all Japa nese sika, the antler velvet is black. Body
with E China, are very large, wide across the color tends to be dark, and there is always a pale facial
condyles, with a broad palate, and, in the chevron. The nighttime vocalization of the stag in
females, with narrow orbits and small teeth; rut is described as being like a donkey’s bray (Ban-
Beijing, Shansi, and Sichuan are also very big. well, 1999).
If these and the Far East specimens are taken Size small, distinguishing it from more N Japa-
as one sample, they are absolutely wider in nese deer; very large neck mane; yellowish white hair
interorbital breadth, though (in the females, around the metatarsal glands; lighter-colored face;
at least) Sichuan is somewhat smaller, with darker winter coat; constant presence of a row of spots
narrower mastoids. on either side of the median dorsal line.
—Vietnam skulls are slightly smaller than E S part of Japan: Honshu S from Biwa Ko, and
China ones, and have narrow nasals, narrow smaller islands to the S.
orbits, and a short palate (none, however, We provisionally recognize a few subspecies, listed
being absolutely different) below. —-1
—0
349-47558_ch01_1P.indd 101 5/31/11 7:05 PM

Table 19 Univariate statistics for male skulls of the Cervus nippon group
Brain- Span
Gt l Nas l Nas br Biorb Interorb case Pal l Mastoid Teeth Pal br condyles
Hokkaido
Mean 293.75 102.50 37.00 130.25 93.00 79.75 167.50 104.25 90.25 86.75 58.25
N 4 4 4 4 4 4 4 4 4 4 4
Std dev 5.560 5.686 5.598 4.113 3.266 4.113 6.403 3.775 4.349 3.096 1.258
Min 290 94 32 125 89 75 163 99 84 84 57
Max 302 106 45 135 97 85 177 108 94 91 60
N Honshu
Mean 292.67 102.17 36.17 133.00 90.00 78.80 166.50 105.40 89.00 87.40 58.00
N 6 6 6 6 5 5 4 5 5 5 4
Std dev 12.011 5.913 3.430 6.325 7.106 5.848 5.916 6.986 1.732 2.608 2.944
Min 277 94 32 122 79 75 160 94 86 84 55
Max 305 109 42 140 98 89 172 113 90 90 61
S Honshu
Mean 246.80 78.80 30.50 111.50 84.00 79.00 157.00 96.00 79.00 84.00 59.00
N 5 5 2 2 1 1 1 1 1 1 1
Std dev 14.550 5.891 6.364 16.263 — — — — — — —
Min 236 72 26 100 — — — — — — —
Max 271 86 35 123 — — — — — — —
Goto Islands
Mean 245.00 75.10 28.70 109.60 69.80 70.80 136.89 85.10 72.60 73.00 48.70
N 9 10 10 10 10 10 9 10 10 10 10
Std dev 9.124 3.695 2.003 2.836 3.425 2.150 5.947 3.071 4.222 2.357 1.829
Min 233 71 26 105 65 68 130 80 67 70 46
Max 261 81 31 113 77 75 145 90 80 78 51
Yakushima
Mean 231.00 72.00 23.00 105.00 72.00 71.00 134.00 70.00 — — 46.00
N 2 — — 1 — — — — — — —
Min 219 — — — — — — — — — —
Max 242 — — — — — — — — — —
Mageshima
Mean 252 77 28 — 80 74 137 77 — — 50
N — — — — — — — — — — —
Tsushima
Mean 271.00 83.00 29.00 — 76.00 73.25 148.50 80.25 — — 56.00
N 1 4 1 — 4 4 4 4 — — 1
Std dev — 2.944 — — 2.708 1.258 3.873 1.708 — — —
Min — 79 — — 74 72 143 78 — — —
Max — 86 — — 80 75 152 82 — — —
Far East
Mean — — — 140.00 92.00 84.00 — 111.00 88.00 91.00 60.00
N — — — 1 1 1 — 1 1 1 1
Beijing
Mean 304.00 105.00 — — 120.00 — — 99.00 — — —
N 1 1 — — 1 — — 1 — — —
Shansi
Mean — 125.00 — — 100.00 — — 84.00 99.00 — —
N — 1 — — 1 — — 1 1 — —
Sichuan
Mean 325.00 114.00 — — 99.00 — — 85.00 — — —
N 1 1 — — 1 — — 1 — — —
E China
Mean 290.93 97.50 33.69 124.00 81.69 79.88 166.71 101.38 85.31 84.94 56.47
N 15 16 16 16 16 16 14 16 16 16 15
Std dev 18.553 10.838 3.554 10.152 8.867 4.365 10.306 10.639 3.842 5.144 3.182
-1— Min 261 76 27 103 63 74 149 81 79 76 51
0— Max 323 115 39 138 91 90 186 115 92 95 62
349-47558_ch01_1P.indd 102 5/31/11 7:05 PM

Brain- Span
Vietnam
Mean 272.00 77.50 25.00 109.50 68.00 74.50 142.50 85.67 85.00 78.00 49.50
N 2 2 2 2 2 2 2 3 2 2 2
Min 255 65 21 99 59 69 117 74 79 72 43
Max 289 90 29 120 77 80 168 101 91 84 56
Solo
Mean — 111.00 33.00 123.00 87.00 79.00 — — 97.00 89.00 —
N — 1 1 1 1 1 — — 1 1 —
Table 20 Univariate statistics for female skulls of the Cervus nippon group
Brain- Span
Goto Islands
Mean 226.75 69.13 24.25 94.38 55.88 66.87 127.57 70.75 68.75 67.88 44.63
N 8 8 8 8 8 8 7 8 8 8 8
Std dev 5.392 3.271 2.121 2.387 1.727 1.553 5.855 2.188 3.615 2.416 1.188
Min 220 64 22 91 53 64 121 68 64 64 43
Max 237 74 27 97 58 69 136 74 76 71 46
Kerama
Mean 221.50 67.00 23.50 92.00 — — — — — — —
N 2 2 2 2 — — — — — — —
Min 210 63 22 88 — — — — — — —
Max 233 71 25 96 — — — — — — —
Tsushima
Mean 243.50 76.50 31.00 105.00 66.50 72.50 138.00 80.00 73.00 80.50 43.50
N 2 2 2 2 2 2 2 2 2 2 2
Min 243 71 27 104 65 72 136 80 70 79 42
Max 244 82 35 106 68 73 140 80 76 82 45
Far East
Mean 303.67 109.00 35.33 122.00 80.67 79.67 176.67 98.00 86.33 87.67 57.00
N 3 3 3 3 3 3 3 3 3 3 3
Std dev 9.074 6.083 2.517 2.646 4.041 3.055 7.024 1.000 3.512 .577 1.000
Min 294 105 33 120 77 77 170 97 83 87 56
Max 312 116 38 125 85 83 184 99 90 88 58
Sichuan
Mean 285.00 93.50 — — 74.00 — — 88.00 — — —
N 2 2 — — 2 — — 2 — — —
Min 279 89 — — 72 — — 87 — — —
Max 291 98 — — 76 — — 89 — — —
E China
Mean 273.83 91.00 29.50 107.50 70.67 74.83 161.50 87.00 83.83 78.33 53.50
N 6 6 6 6 6 6 6 6 6 6 6
Std dev 7.278 5.215 2.074 2.588 3.559 2.639 6.745 4.050 5.419 1.633 2.429
Min 267 84 27 103 67 72 153 82 77 76 50
Max 283 100 32 111 76 79 170 93 90 80 57 —-1
—0
349-47558_ch01_1P.indd 103 5/31/11 7:05 PM

104 Artiodactyla
Cervus nippon nippon Temminck, 1838 For the full synonymy of this subspecies, see Groves &
Smeenk (1978).
1836 Cervus nippon Temminck. “Isles of Japan”; probably the
surrounds of Nagasaki (Groves & Smeenk, 1978).
Cervus aplodontus yesoensis (Heude, 1884)
1884 Sika orthopus Heude. Kobe.
1884 Sika schlegeli (and other names) Heude. Goto Islands. 1884 Sika yesoensis (and other names) Heude. Hokkaido.
1888 Sika paschalis (and other names) Heude. Goto Islands. 1897 Sika dolichorhinus Heude. Hokkaido.
1898 Sikaillus daimius (and other names) Heude. Goto Islands. 1897 Sika rutilus Heude. Hakodate.
For the full synonymy of this subspecies, see Groves &
S Honshu, Shikoku, and Kyushu; Goto Islands.
Smeenk (1978).
Cervus nippon mageshimae Kuroda & Okada, 1951

Cervus pulchellus Imaizumi, 1970
1951 Cervus nippon mageshimae Kuroda & Okada.
1970 Cervus pulchellus Imaizumi. Tsushima Island.
Mageshima.
Distinguished by skull characters; by the long basal
Larger than the deer on Yaku.
segment to the antlers; and by the whitish belly.
Mageshima & Tanegashima.
Cervus hortulorum Swinhoe, 1864
Cervus nippon yakushimae Kuroda & Okada, 1951
1864 Cervus hortulorum Swinhoe. Imperial Summer Palace,
1951 Cervus nippon yakushimae Kuroda & Okada. Yakushima.
Beijing.
On the differences between these small-island sika, 1864 Cervus mantchuricus Swinhoe. Newchwang (= Jingkou),
see Banwell (1999). Manchuria.
1874 Cervus euopis Sclater. Newchwang (= Jingkou), Manchuria.
Cervus nippon keramae Kuroda, 1924 1876 Cervus dybowskii Taczanowski. S Ussuri district,
Manchuria.
1924 Cervus nippon keramae Kuroda. Kerama Island.
1882 Cervus cyclorhinus and Cervus hyemalis Heude. Shandong
Smaller in size; dark, with no spots. As its describer Province.
himself admitted, it may be that the Kerama popula- 1884 Sika grassianus Heude. Tsinglo-hsien (= Echeng), N Shanxi.
tion derives from a medieval introduction. 1884 Sika microspilus Heude. Mukden (= Shenyang), Manchuria.
1894 Sika imperialis Heude. Songari Valley, Manchuria.
Cervus aplodontus (Heude, 1884)
Specimens of what were identified as Cervus nippon
Males with a small neck mane, or none; underside
hortulorum, from Altai deer farms, had 2n = 66 (Grapho-
dark, like the back; often an indistinct row of light
datsky & Radjabli, 1985).
spots along either side of the dorsal stripe. Light eye-
Tail short, 8% of the head plus body length. Neck
rings; facial chevron present in summer, absent in
mane reduced or absent. Winter coat light red-brown
winter. Rump-patch large, rimmed with black; tail
to gray; rump-patch large.
very short, dark. Metatarsal glands dark, elongated,
N China and the Russian Far East.
surrounded by light gray fur. Winter pelage dark gray;
As Geist (1998) has emphasized, the problem with
summer pelage reddish; males spotted in summer,
the nomenclature of this species is that almost none
the females without spots.
of the type specimens were obtained in the wild.
N Japan, N from Biwa Ko.
The type of C. hortulorum, for example, was from the
Provisionally, we recognize the N Honshu and
grounds of the Imperial Summer Palace; those of C.
Hokkaido deer as separate subspecies, but we are not
mantchuricus and C. euopis were obtained in the har-
especially convinced of this distinction.
bor city of Jingkou; and so on. The only present-day
wild populations apparently supported by names are
Cervus aplodontus aplodontus (Heude, 1884)
C. dybowskii and C. imperialis, both from the Ussuri
1884 Sika aplodontus Heude. N of Tokyo. region; the similarity of the type of C. hortulorum to
1884 Sika mitratus Heude. Tokyo. the Ussuri specimens does, however, suggest that it
1884 Sika xendaiensis Heude. Sendai. represents the same population. Possibly C. cyclorhi-
1897 Sika ellipticus Heude. Sendai. nus and C. hyemalis, whose types could not be found
-1— 1897 Sika minoensis Heude. Mino, W of Tokyo. by Braun et al. (2001), were also obtained in the wild.
0—
349-47558_ch01_1P.indd 104 5/31/11 7:05 PM

cerv idae 105
ter states, compared with other deer (including sika):

Cervus sichuanicus Guo et al., 1978 the more complex antlers, including (except in some
individuals of the dwarfed C. corsicanus) the presence of
1978 Cervus nippon sichuanicus Guo et al. Ruoergai district,
a bez tine, and the brow tine being at a right angle or
Sichuan, 2900 m.
more to the beam; the much larger rump-patch; the
Described as a large species, with a short tail; no mane; shorter tail; the smaller rhinarium; and the uniform
median dorsal stripe distinct, dark, bordered on each body color of the adults (with one exception—C. maral
side by a row of distinct white spots. retains not very conspicuous spotting into adulthood).
Sichuan. By contrast, sika have primitive states: three-point-plan
antlers, without a bez tine and with the brow tine being
Cervus taiouanus Blyth, 1860 at an acute angle to the beam; the rump-patch re-
stricted to medial aspect of the buttocks; the tail long;
Synonyms include several names by Heude, 1882 and 1884.
the rhinarium larger; and the adults conspicuously
Clearly spotted in winter in both sexes; in summer, white-spotted. Yet the mtDNA of C. wallichii, C. mac-
bright chestnut, with a prominent dorsal stripe; tail neilli, C. alashanicus, C. xanthopygus, and C. canadensis is
light fawn below, not white as in other species; inside close to that of sika, forming merely a somewhat sepa-
of the ears fawn; males strongly maned. Antlers rather rate subclade. It is particularly noticeable that, morpho-
weak, but with 10 points; color of the velvet pinkish, logically, C. hanglu closely resembles C. wallichii and C.
not black as in other sika. Apparently long-bodied, macneilli, to the extent that Flerov (1952) even synony-
compared with other sika. mized them, yet the mtDNA of C. hanglu is that of the
Taiwan. W (red deer) group. What is the cause of this anomaly?
Our explanation is that hybridization has taken
Cervus pseudaxis Gervais, 1841 place. We propose that shou related to C. hanglu spread
E and then NE through the mountain forests of the
1841 Cervus pseudaxis Gervais in Eydoux & Souleyet. No
Himalayas and the E edges of the Tibetan plateau,
locality; thought to be Vietnam.
meeting indigenous sika populations and interbreed-
1871 Cervus mandarinus Milne-Edwards. Summer Palace,
ing with them asymmetrically— shou stags with sika
Beijing (see below).
hinds, and then with backcross hinds, and so on—
1873 Cervus kopschi Swinhoe. Kien-chang (= Aicheng), Jiangxi.
until the local sika populations had been eliminated
1882 Cervus frinianus (and other names) Heude. Lake Poyang.
by nuclear swamping, but had left their mtDNA in
1884 Sika brachyrhinus (and other names) Heude. Lake Poyang.
the transformed population. Only a single sika taxon,
1888 Sika rivierianus Heude. Lake Poyang.
C. sichuanicus, remains in these mountain forests.
1894 Sika dugenneanus Heude. Mountains of Phu-Lang-
The shou populations continued to spread (and to ac-
Thuong, Tonkin, Vietnam.
quire successively more apomorphic character states,
Externally, somewhat resembling C. taiouanus. Sum- becoming “wapiti”) along the E edge of the plateau,
mer coat bright, winter coat gray to black, spotted in bearing the sika mtDNA, and then spread through
both seasons; dorsal stripe dark; underside light similar country farther N and NW, and to similar for-
brown. Tail relatively long, 13% of the head plus body est types to the NE, ultimately crossing the Bering
length, with a broad, dark dorsal portion; rump-patch Land Bridge into North America.
small, white; mane thick, light-colored; facial chevron
absent or vague. Metatarsal glands small. Inside of the Cervus unicolor Group
ears white. Males darker than the females. rusa and sambar
Vietnam and S China. On the general biology of this species-group, see Ban-
As noted by Geist (1998), the type of C. mandarinus well (2006).
resembles these S populations, despite its having Antlers with only three points; rhinarium extend-
been obtained from the Summer Palace. ing well below the nostrils; tail long and bushy; pel-
age coarse and shaggy.
Addendum to the Cervus elaphus and According to the molecular clock of Pitra et al.
Cervus nippon Groups (2004), this group, the albirostris-canadensis-nippon
A problem remains: what are the interrelationships of group, and the C. elaphus group form a three-way split,
these two species-groups? Morphologically, red deer, dating from about 3.5 Ma. The authors dated the sep- —-1
shou, and wapiti share several evidently derived charac- aration between C. unicolor and C. timorensis at about —0
349-47558_ch01_1P.indd 105 5/31/11 7:05 PM

106 Artiodactyla
1 Ma. These dates, as noted earlier, should be pushed Bengal and Assam, NE into S China (including
back somewhat, because the muntjac-cervine split is Taiwan) and SE to the Malay Peninsula, Sumatra,
now known, from fossil evidence, to be earlier than and Borneo.
was formerly thought. We have relatively little information on geographic
variation in this species, although it is clear that the
Cervus alfredi Sclater, 1870 mainland specimens are of relatively larger size, are
Prince Alfred’s deer somewhat more sexually dimorphic, and have antlers
that are often much longer than those of Sumatra,
1870 Cervus alfredi Sclater. “Manila.”
Borneo, and other islands; but we cannot fi nd any
For the full synonymy of this species, see Grubb & Groves
absolute differences. Groves (2006a) drew attention to
(1983).
the apparent differences of the Chinese animals from S
Color blackish brown, light-spotted on the flanks; dor- Asian and Southeast Asian sambar.
sal band dark, bordered with faint ochre spots. Under-
side, extending to the inner aspects of the limbs and Cervus mariannus Desmarest, 1822
the chin, a sharply contrasting buff y. Lower segments Philippine deer
of the legs paler than the body. Head somewhat paler
1822 Cervus mariannus Desmarest. Guam.
than the body. Pelage thick and soft.
And numerous synonyms by Heude!
Leyte, Cebu, Guimaras, Negros, and neighboring
For the full synonymy of this species, see Grubb & Groves
islands in the C Philippines.
(1983).
According to Randi et al. (2001), this species is sis-
ter to the rest of the genus Cervus; morphologically, 2n = 65.
as noted by Grubb & Groves (1983), it sits very uneas- Deep brown, with a blackish tinge on the back, paler
ily amid the rusa/sambar group. on the cheeks, slightly paler on the underside. Tail white
beneath. Muzzle dark; rest of the head as in the body.
Cervus unicolor Kerr, 1792 Legs paler than the body, especially on the lower por-
Indian sambar tions. Spots present in the fawns, but not in the adults.
2n = 60. Luzon, and introduced to Guam; Mindanao;
Long-legged; color brown, varying from grayish Basilan.
to very dark brown in winter, more yellowish red- There is a possibility of a second species in the
brown in summer; belly darker than the back; tail lowlands of Mindanao, around Sarangati Bay, as de-
long-haired, very bushy, contrastingly darker than the scribed by Grubb & Groves (1983), and this needs
body. Forehead flat; preorbital fossa very deep. Ant- further investigation.
lers robust, relatively short, with a short brow tine;
the two terminal tines tending to form an equal fork, Cervus barandanus (Heude, 1888)
with the posteromedial tine usually being longer. Mindoro deer
(Following van Bemmel, 1949). Fawns spotted.
1888 Ussa barandanus Heude. Mindoro.
Sri Lanka and most of mainland of S Asia.
Sri Lankan specimens average slightly smaller than C. barandanus, for the fi rst time, is regarded here as
the Indian ones. a distinct species, because the width of the braincase,
the mastoids, and the nasals, and the length of the
Cervus equinus (G. Cuvier, 1823) toothrow, are consistently and distinctively less than
Southeast Asian sambar in C. mariannus, and the palate is wider.
Synonymy includes numerous names by Heude!
Cervus nigellus (Hollister, 1913)
2n = 62 (Indochinese sample). Mindanao mountain deer
Smaller than C. unicolor; brow tine longer; antero-
1913 Rusa nigellus Hollister. Mt. Malindang, Mindanao.
lateral tine always continuing the line of the beam,
? 1952 Rusa nigellus apoensis Sanborn. Mt. Apo, Mindanao.
and longer than the posteromedial tine, the latter turn-
ing inward and backward. Tail somewhat more bushy Size very small; pelage soft; higher frequency of neck
than in C. unicolor. Dark brown in winter; lighter, more hair reversal than in lowland deer.
-1— yellowish or reddish, in summer. Fawns spotted at The question of the separation of the mountain
0— fi rst, the spots disappearing rapidly with age. deer from the lowland ones—and, indeed, of those on
349-47558_ch01_1P.indd 106 5/31/11 7:05 PM

cerv idae 107
the two mountains from each other—was discussed relatively long and narrow, not contrasting in color
by Grubb & Groves (1983), who concluded that the with the body. Forehead concave; preorbital fossa
mountain form is “very unlikely to be a distinct spe- less deep. Antlers relatively long and slender; poste-
cies”; but here we are inclined to reverse that decision, rior tine continuing the direction of the beam, and
despite the fact that deer from the lower slopes are generally longer. (Following van Bemmel, 1949.)
larger than those from the upper slopes. Nonetheless, Java and Bali; widely introduced by human agency
we recommend further study. into the E Indonesian islands as far as Timor and the
Moluccas.
Cervus timorensis There appear to have been changes in pelage and
rusa size over the several centuries (up to 3 millennia)
2n = 60. since their introductions, witnessed by the descrip-
Smaller than C. unicolor and C. equinus, but larger tion of several subspecies— conspicuous enough to
than the Philippine forms. Relatively short legged and have been recognized by van Bemmel (1949)—from
long bodied. Tail less bushy than the other species, the deep-water islands of E Indonesia.
—-1
—0
349-47558_ch01_1P.indd 107 5/31/11 7:05 PM

reversal of these characters in the swamp-loving sita-
11
tunga (Tragelaphus spekii).
A gall bladder is present.
Janis & Scott have fully discussed the relationships
of the Bovidae among the ruminants. While undoubt-
edly the family forms a monophyletic clade, “it ap-
pears that no apomorphic characters, apart from the
presence of horn cores, can be used to characterize
the Bovidae” ( Janis & Scott, 1987:47).
The overriding theme in bovid evolution is surely
sexual dimorphism.
Bovidae A widely accepted model of social orga ni zation
( Jarman, 1974) relates feeding style to group size via
the dispersion of food items, habitat choice, and anti-
predator behavior, which, taken together, help to de-
termine the females’ tendency to aggregate; and, there-
fore, the males’ best strategy to secure mating. This,
in turn, relates back to sexual dimorphism.
The Bovidae: the largest, most diverse family of un- Lundrigan (1996) associated horn form with male
gulates. Par ticu lar themes include the role of sexual strategy, specifically, fighting behavior. Short, simple
selection; African biogeography; the recency of spe- horns would be associated with stabbing in fighting
ciation in some cases; gene flow between related but (e.g., nilgai, Oreamnos, most dwarf antelope); long
distinct species; why many species simply must be horns, with wrestling (the larger Tragelaphini) and
split up. The question of a hybrid origin for the genus fencing (the Hippotragini); a long “catching arch” (the
Capra (as proposed by Alexandre Hassanin); are there anteriorly concave midsection of a sigmoid or twisted
any other taxa of hybrid origin lurking within the horn) is especially associated with wrestling (the
Bovidae? Tragelaphini, Nanger, the Reduncini, impala); and thick
horn bases, with ramming during fighting (Cape buf-
falo, thick-horned duiker, Ovis).
BOVIDAE GRAY 1821 Returning again to the sexual dimorphism theme
We here use the nomenclature established by Grubb (2001a). much later, Jarman (2000) plotted sexual dimorphism
(in body weight) against the log of the female’s body
Usually there is one single, circular lacrimal orifice, weight, showing that, on the whole, it is the medium-
inside the orbital rim; there is, however, some varia- sized bovids (peaking at a female weight of about
tion to this pattern in the Bovinae (Leinders & Heintz, 55 kg), not the very large ones, that have the great-
1980). est sexual dimorphism. He concluded that, just as a
The anterior sagittal gulley of the metatarsus re- larger male can monopolize mating, a larger female
mains open over its entire length. can more effectively monopolize resources to the
The sagittal crests on the articular surfaces of the benefit of her philopatric daughters. Estes (2000) like-
distal condyles of the metatarsus are more protrud- wise drew attention to the importance of female fit-
ing, more sharp edged, and more proximately ex- ness, especially in sexually integrated social groups,
tended than in the Cervidae; there is no “plateau where females may mimic males—for example, in
postarticulaire” on the medial phalanx; and the prox- color.
imal articular surface of the distal phalanx is more This is not to say that natural selection itself has
concave and posteriorly extended, with an “émi- not been a powerful force in bovid evolution (as one,
nence pyramidale,” that is, an extensor process on its of course, predicts). Stoner et al. (2003) examined
anterior edge (Leinders, 1979). As that author notes, color and possibly associated characters by optimiz-
these features give the articular surfaces a wider ing character states on a phylogenetic tree. They
range of movement (especially of hyperextension) examined all artiodactyls, although obviously bo-
than in the Cervidae and greater shock-absorbing ca- vids constituted the largest component of their data-
-1— pability, allowing for what he dubs “greater pogostick set. They concluded that stripes and spots really
0— effect.” According to Leinders (1979), there is some do aid in concealment, whereas flank bands, facial
349-47558_ch01_1P.indd 108 5/31/11 7:05 PM

b o v idae 109
coloration patterns, and conspicuous, contrastingly phylogenetically based taxonomic arrangement,

dark or light legs and tails are all communicatively which is here accepted, as follows:
significant.
Subfamily Bovinae Gray, 1821
An analysis of morphological and behavioral
Tribe Bovini Gray, 1821
changes in the Bovidae is in Vrba & Schaller (2000). A
Bos, Bubalus, Syncerus, Pseudoryx
concise account of the evolution of the Bovidae in
Tribe Boselaphini Knottnerus-Meyer, 1907
Africa, with special attention to the fossil record, has
Boselaphus, Tetracerus
been given by A. W. Gentry (1990).
Tribe Tragelaphini Blyth, 1863
Patterns of diversification within restricted taxo-
Ammelaphus, Nyala, Tragelaphus, Taurotragus,
nomic groups follow trends that are, to a consider-
Strepsiceros
able extent, predictable from ecology. Vrba (1980),
Subfamily Antilopinae Gray, 1821
studying fossil bovids—in par ticu lar, comparing the
Tribe Neotragini Sclater & Thomas, 1894
Alcelaphini with the Aepycerotini (which at that time
Neotragus (presumably including Hylarnus,
were considered probable sister groups)—proposed
Nesotragus)
that stenotopic lineages (i.e., those specialized in par-
Tribe Aepycerotini Gray, 1872
ticu lar ecological niches) speciate more than do eury-
Aepyceros
topic lineages (those of more diverse ecological toler-
Tribe Antilopini Gray, 1821
ance), and this provides a potential background
Raphicerus, Antidorcas, Ammodorcas,
against which to analyze diversification patterns in
Litocranius, Saiga, Antilope, Nanger,
living species, including (in this case) bovids. We
Gazella, Eudorcas, Dorcatragus, Madoqua,
have found that stenotopic taxa, such as the Alcela-
Ourebia, Procapra
phini in general, indeed do speciate readily (the genus
Tribe Reduncini Knottnerus-Meyer, 1907
Damaliscus is unexpectedly speciose). Examining eco-
Redunca, Kobus (probably including Adenota,
logical parameters, as given by Estes (1991) and by
Hydrotragus), Pelea
Kingdon (1982a, b), suggests that grysbok are more
Tribe Hippotragini Sundevall, 1845
stenotopic than steenbok; lechwe than waterbuck;
Hippotragus, Addax, Oryx
sable antelope than roan antelope; and lesser kudu
Tribe Alcelaphini Brooke, in Wallace, 1876
than greater kudu or, especially, eland, which is eury-
Alcelaphus, Beatragus, Damaliscus,
topic in the extreme. If Vrba’s (1980) model is correct,
Connochaetes
one should therefore expect greater taxonomic diver-
Tribe Caprini Gray, 1821
sity at the stenotopic end of the scale; and this, we
Pantholops, Oreamnos, Budorcas, Ammotragus,
have found, turns out to be the case.
Arabitragus, Hemitragus, Pseudois, Capra,
Grubb (2000a) drew attention to the existence of
Nilgiritragus, Ovis, Rupicapra, Naemorhedus,
striking morphoclines among African bovids, and
Capricornis, Ovibos
also giraffes. Examples include the genera Oryx (run-
Tribe Cephalophini Blyth, 1863
ning from Arabia to S Africa); Tragelaphus and Synce-
Philantomba, Cephalophus, Sylvicapra
rus (both running from closed habitats in W Africa to
Tribe Oreotragini Pocock, 1910
open habitats in E and S Africa); the Nanger granti
Oreotragus
group (from SE Kenya, in a ring via Ethiopia and Su-
dan to Maasailand); and Ovis and, perhaps, Capra.
Subfamily Bovinae Gray, 1821
Morphological analyses have always supported a
basic division of the family Bovidae into two major The horns are not regularly ridged, but they are usu-
clades: (1) the Bovinae, and (2) the rest, for which the ally strongly keeled and spirally twisted.
name Antilopinae has been generally adopted (see, The mastoid is narrow, and, in the last two man-
for example, A. W. Gentry, 1992). Molecular studies dibular premolars, the paraconid is differentiated
are universal in supporting such a division (see, for from the parastylid (A. W. Gentry, 1992).
example, Allard et al., 1992; Gatesy et al, 1997; Has- Leinders & Heintz (1980) described some varia-
sanin & Douzery, 1999a; Matthee & Davis, 2001; tions in the pattern of the lacrimal orifice in the Bo-
Kuznetsova et al., 2002; Ropiquet, 2006). vini and the Tragelaphini (these are not seen in the
The most complete DNA phylogeny of the family, Boselaphini). There may be a single orifice, as in all
fully compatible with most or all other published other Bovidae, or there may be two, and their posi- —-1
trees, is that of Ropiquet (2006), who proposed a tions are slightly variable, usually inside the orbital —0
349-47558_ch01_1P.indd 109 5/31/11 7:05 PM

110 Artiodactyla
rim (as in most Bovidae), but sometimes on the rim alternative neighbor-joining trees for the Bovini, us-
itself. ing two different methodologies. In one, B. indicus
There is little disagreement that there are three and B. taurus formed a sister clade to B. frontalis plus
tribes, which form almost a three-way phylogenetic B. grunniens, and the two together were the sister
split. In analyzing the COII gene, however, Janeck clade to American bison. The African and Asian buf-
et al. (1996) found that Boselaphus (representing the faloes constituted separate branches, which, together
Boselaphini) and Ammelaphus (representing the Tragela- with the bison/cattle clade, formed a basal trifurca-
phini) formed a clade, but with only 51% bootstrap tion. The other methodology was similar, except that
support, although the association of the buffaloes to B. grunniens was no longer joined to B. frontalis, but
Bos also had only 51% bootstrap support. was sister to a B. frontalis / B. taurus plus B. indicus clade.
Divergence times were calculated, and compared with
Tribe Bovini Gray, 1821 those of Hartl et al. (1988), which appeared slightly
As elucidated by A. W. Gentry (1992), the Bovini have older: in the order of 0.5–1.5 Ma for B. taurus / B. in-
low, wide skulls; horn cores emerging transversely; dicus versus either mithan, yak, or bison; 1–2.61 Ma
internal sinuses in the frontals extending into the versus African buffalo; and 1.9– 4.9 Ma versus Asian
horn cores; a short braincase and widened occiput; buffalo.
molars with large basal pillars and complicated cen- Using multilocus DNA fi ngerprinting, Vasil’ev et
tral cavities; and upper molars with prominent ribs al. (2002) arrived at the following cladogram, based
between the styles. on genetic similarity coefficients:
Groves (1981b) attempted a phylogenetic arrange-
(bonasus (bison ((gaurus, indicus) ( javanicus,
ment based on cranial characters: in the genus Bos, B.
taurus)))).
gaurus was conceived as sister to other species, fol-
lowed by B. javanicus, and there was then a dichot- While it is possible that the inclusion of interspe-
omy between a B. sauveli / B. primigenius clade and cific hybrids in these calculations may have artifi-
a B. bison / B. mutus clade. Aspects of this have since cially brought some species pairs together (B. gaurus
been shown to be incorrect, but the association of bi- and B. indicus, in par tic u lar), inspection of the indi-
son and yak has proved justified. In the main, the vidual values in these authors’ table 3 shows that
phylogeny of the Bovini is depicted in DNA trees as the genetic similarities are more or less as in their
an initial separation between buffaloes and the rest, dendrogram— for example, the genetic similarity
followed among “the rest” by a more or less three- between B. bonasus and other taxa is lower than
way split: true cattle, gaur-banteng-kouprey (the erst- that for B. bison.
while subgenus Bibos), and bison-yak (the erstwhile The Y-chromosome phylogeny of Nijman et al.
subgenera Bison and Peophagus). (2008) found a rather different tree: the two genera of
The dendrogram generated by Hartl et al. (1988), buffaloes were linked (which they often are not in
on the basis of 15 protein loci studied by electrophore- mtDNA phylogeny), and the first (short, but signifi-
sis, gave buffaloes as a clade separate from Bos (includ- cant) branch from the non-buffalo lineage was the
ing the bison). European and American bison formed a yak; then there is a split between bison (both B. bison
clade with almost no difference between the two, and and B. bonasus) and other cattle. This supports the hy-
this was sister to a clade containing the following: pothesis of Verkaar et al. (2004) that Bos bonasus is of
hybrid origin; and Nijman et al. (2008) suggested that
((gaurus, javanicus) (mutus, primigenius)).
the genus Poephagus, for the yak, should be revived.
In all the phylogenetic trees generated by Janeck
et al. (1996), Bos sensu stricto and the bison formed a Cattle Group
clade with respect to the buffaloes; while the precise Bos primigenius Bojanus, 1827
interrelationships of the species varied, B. bonasus was aurochs
always sister to the other species, with strong boot-
The name Bos primigenius Bojanus, 1827 was placed on the
strap support, and B. taurus and B. grunniens always
formed a clade—rather unexpected fi ndings that do
Opinion 2027 (2003).
not cohere with other phylogenetic results.
On the basis of genetic distances calculated from A study of mtDNA sequences from modern domestic
-1— 20 microsatellites, Ritz et al. (2000) constructed two cattle (Loftus et al., 1994) separated taurine and zebu
0—
349-47558_ch01_1P.indd 110 5/31/11 7:05 PM

b o v idae 111
breed-groups very strongly, with an average pair- Table 21 Univariate statistics for Bos gaurus: males
wise distance of 7.86%, not very much less than that
between cattle and bison at 10.62%. The authors ar- Gt occ
Gt l br Teeth Tip–tip Span
gued that, if bison and cattle lineages separated 1–
1.4 Ma, then the two cattle lineages separated 740,000 Southeast Asia
to 1.04 Ma. Mean 581.13 298.72 147.56 569.50 874.76
Cattle horns were used in the early 20th century N 15 18 16 20 21
in Cambodia to create a mysterious animal, known Std dev 24.118 36.950 7.831 153.207 110.226
only by horns and frontlets, called the “linh duong” Min 537 202 127 313 528
Max 626 334 160 941 1088
or “khting vor” (Pseudonovibos spiralis). Both its mtDNA
India
and nDNA are identical to those of domestic cattle, Mean 544.89 293.05 145.30 473.07 827.68
specifically Vietnamese cattle (Hassanin et al., 2001), N 45 38 40 41 41
and the method of faking was discovered by examin- Std dev 20.589 18.271 10.634 135.001 99.261
ing longitudinal sections of the horns (H. Thomas et Min 506 233 130 248 668
al., 2001). A brief history of this episode—which led to Max 586 318 173 964 1180
a spurious species being listed on conservation data-
bases, and even to frantic searches for it in the field—
was recounted by Galbreath & Melville (2003). Table 22 Univariate statistics for Bos gaurus: females
Gaur-Banteng-Kouprey Group Gt occ

Bos gaurus Hamilton Smith, 1827 Gt l br Teeth Tip–tip Span
gaur, seladang
Southeast Asia
1827 Bos gaurus Hamilton Smith. Mainpat, Sarguja Mean 524.40 237.75 142.40 374.71 643.57
Tributary States, approximately 23° N, 83° E (see N 5 4 5 7 7
Std dev 6.877 8.539 10.479 273.229 149.203
Harper, 1940).
Min 516 230 127 90 510
The name Bos gaurus Hamilton Smith, 1827, was placed on Max 531 247 155 765 867
the Official List of Specific Names in Zoology by ICZN, India
Opinion 2027 (2003). It had previously been conserved Mean 539.88 270.38 145.50 417.50 718.50
(because it was threatened by an earlier name; see N 8 8 8 8 8
Groves, 1982c) under ICZN, Opinion 1348 (1985). Std dev 17.956 23.323 8.536 158.073 121.959
Min 522 237 135 64 535
We found no differences among samples of gaur from Max 579 306 157 593 870
different parts of India (S, N, Sikkim). Within South-
east Asia, 83% of the samples from Thailand and 63%
from the Indochinese region are correctly sorted. In this case, the Tripura specimen assorts with India,
Using six cranial variables (greatest skull length, as do specimens from the Upper Chindwin and
nasal length, postorbital breadth, breadth between Mogok; one specimen from Chittagong assorts with
horn bases, and greatest and least occipital breadths), India, and the other with Southeast Asia.
however, showed that there is complete separation Tables 21 and 22 show that Southeast Asian gaur
between the Indian and the Southeast Asian samples tend to be considerably larger than Indian guar, though
(N = 7, 10, respectively). A skull from Tripura assorts there is much overlap. Horns in the males continue
with Southeast Asia. Absolute size is the main dis- this tendency, but in the females it is reversed: both
criminator, with greatest skull length, nasal length, span and tip-to-tip distance are greater for the Indian
and postorbital breadth contrasting with least occipi- than for the Southeast Asian females.
tal breadth. A more complete defi nition of the Gaur, of unknown affi nity, formerly occurred in
boundary between them is needed, but on this cra- Sri Lanka (Groves, 1982c).
nial basis, probably the S and Southeast Asian gaur
would better be classed as two distinct species. Bos gaurus gaurus Hamilton Smith, 1827
The horn variables give a much less complete sep- 2n = 58.
aration: 80% of the Indian specimens are correctly Rather smaller in size than B. gaurus but with rela-
identified, but only 67% of the Southeast Asian ones. tively longer nasals and a wider occiput. —-1
—0
349-47558_ch01_1P.indd 111 5/31/11 7:05 PM

112 Artiodactyla
Bos gaurus laosiensis (Heude, 1901) horn span, but the ranges of variation overlap. Main-
land B. j. birmanicus is not larger in cranial dimen-
For the full synonymy of this subspecies, see Braun et al.
sions than B. j. javanicus, but the horns, on average,
(2001).
are larger in all dimensions.
Vadhanakul et al. (2004) found that a male from Thai-
land had a chromosome number of 2n = 56, with a Bos javanicus javanicus d’Alton, 1823
Robertsonian fusion; but this is not a consistent differ- 2n = 56.
ence between Southeast Asian and Indian gaur, be- Mature bulls black, with horns spreading out and
cause Chaveerach et al. (2007) later found 2n = 57 in then curving up, connected by a cornified band across
another Thai individual. the intercornual ridge; cows and calves ochery-
Larger, with relatively short nasals and a narrow colored, with smaller, crescentic, upstanding horns.
occiput; horn tips less turned in relative to the span. Many populations of the domesticated form of
this species, known as Bali cattle, turn out to have
Bos javanicus d’Alton, 1823 introgressed genes from ordinary (humped) domes-
banteng tic cattle, and the same cannot be excluded for rem-
nant wild populations (Bradshaw et al., 2006). In the
1830 Bos leucoprymnus Quoy & Gaimard. Supposed by Blyth
opinion of those authors, this makes the feral popula-
(1842) to be based on a hybrid, but there is no evidence
tion in the Cobourg Peninsula, Northern Territory,
for this.
Australia, especially valuable, as there is no trace of
1840 Bos sondaicus Schlegel & Müller. Java.
any non-banteng DNA in 54 sampled individuals.
The three banteng “subspecies” do not separate well Nonetheless, the Australian banteng do have domes-
on multivariate analysis, but one problem is that many tic (Bali cattle) ancestors, and they have not redevel-
of the available specimens originated as hunting tro- oped the full wild phenotype; for example, the horn
phies; thus they are very incomplete, and it is not pos- tips of bulls point outward, not inward as in wild
sible to build up a large dataset in which each speci- banteng bulls.
men has all available variables. Additional specimens
might well enable complete separation to be made, but Bos javanicus lowi Lydekker, 1912.
to do so at the moment would be unsafe. Table 23 gives Smaller, with less outcurved horns.
univariate statistics. The Bornean B. j. lowi is consider- There are too few specimens to decide whether
ably smaller than the other two taxa, with a smaller this is a distinct species.
Table 23 Univariate statistics for Bos javanicus: males
Horn l
Gt l Biorb Teeth Tip–tip Span curve
birmanicus
Mean 513.75 228.79 139.78 555.43 829.40 314.58
N 28 38 32 40 42 24
Std dev 23.609 20.287 8.178 142.579 121.935 40.239
Min 465 202 117 205 580 235
Max 566 331 157 855 1080 382
javanicus
Mean 511.55 222.10 137.28 482.31 733.57 295.23
N 47 48 47 42 44 22
Std dev 19.053 9.825 4.610 112.601 88.222 40.959
Min 461 203 128 310 577 226
Max 557 247 148 865 896 362
lowi
Mean 472.00 208.24 139.27 319.85 546.70 269.10
N 3 25 11 33 33 10
Std dev 19.079 11.505 18.639 100.856 85.753 41.816
Min 450 184 119 113 381 204
-1— Max 484 237 191 497 697 318
0—
349-47558_ch01_1P.indd 112 5/31/11 7:05 PM

b o v idae 113
Museum of Bourges, deemed to have been a Cambo-

Bos javanicus birmanicus Lydekker, 1898. dian domestic ox but physically resembling a kou-
prey. Indeed, the specimen clearly was a kouprey, res-
1898 Bos sondaicus birmanicus Lydekker. Burma.
urrecting the possibility, which had been suggested
1909 Bos sondaicus porteri Lydekker. MeWong district, C
half a century before, that domestic kouprey exist.
Thailand.
For the full synonymy of this subspecies, see Braun et al.
Bison-Yak Group
(2001).
Flerov (1965) and Bohlken (1967) made craniological
A study of two males and two females from Cambo- comparisons between the taxa of bison. Flerov de-
dia showed a difference in karyotype from Javan ban- scribed the skull of the European bison in some detail:
teng: 2n = 60 in the Cambodian banteng, contrasting smaller than that of the American bison, with a depres-
with 2n = 56 in the Javan ones (Ropiquet et al., 2008). sion on the forehead behind the nasals and between the
We need to know the consistency of this difference, orbits; the premaxillae narrower; the horn cores pro-
as well as the karyotype of the Bornean taxon, before jecting less backward; the facial skeleton low; the cho-
we can use this information as the basis for a taxo- ana relatively low; the occipital condyles less recurved
nomic regrading. forward at their lateral tips; the basioccipital shortened
In Burma, N Thailand, and the Indochinese parts and narrowing anteriorly; the mandibular condyles
of the range, the bulls are ochery-colored, like the short; the external auditory meatus larger; and the
cows, but in the peninsula many bulls turn black. As (mandibular) incisor-canine row relatively narrow.
this is the external difference that is commonly cited These are the major points of his extremely long and
between mainland and Javanese banteng, it appears detailed description, but their degree of consistency,
that no fi rm differentiation can be made on the spe- and even the number of specimens studied, is unclear
cies level. (of bison in total, living and fossil, he mentioned “more
The skull of the type specimen of Bos sondaicus than 300 specimens of different age and sex”). For the
butleri Lydekker, 1905, often considered a synonym of moment, we list these characters in summary form, in
this taxon, appears to us to be an ordinary domestic the hope that they can be tested in the future.
cow. Bohlken (1967), on the other hand, using mostly
metrical characters, found very few major differences
Bos sauveli Urbain, 1937 between American and European bison, which, in all
kouprey allometric comparisons, tended to fall on the same or
Hassanin & Ropiquet (2004), using both mitochon- parallel slopes; the main difference was the less ro-
drial and nuclear sequences, placed kouprey, gaur, bust horns of American bison, with a correlated lower
and banteng as three approximately equal splits from occipital height.
a common clade.
A brief flurry was caused when Galbreath et al. Bos bonasus Linnaeus, 1758
(2006) found that the kouprey and Cambodian ban- wisent, Eu ropean bison (Polish/lowland bison)
teng contained very similar mtDNA, and suggested
Flerov (1965) opined that Linnaeus described the species
that the kouprey was, in fact, a feral hybrid between
“on the basis of specimens from the Bialowieza Forest.”
banteng and humped cattle, as had been fi rst pro-
posed long before. This, however, was laid to rest by Distinguishing the lowland bison from the Cauca-
Hassanin & Ropiquet’s (2007) study of nDNA, which sian bison, Flerov (1965) described the former as hav-
showed that, on the contrary, it was the Cambodian ing a convex forehead; the preorbital depression is
banteng which had incorporated mtDNA from the less clearly defi ned; the occipital condyles are simpler
kouprey, not vice versa. This makes us wonder in form; and the nasal cavity is relatively more volu-
whether the distribution of the kouprey was at one minous, so that the skull height is slightly greater—he
time much wider, and has been progressively reduced stated that the distance from the lower orbital mar-
by peripheral populations undergoing nuclear gin to the alveolus of the posterior molar is about a
swamping as the banteng spread into their range, quarter of the toothrow length. The nasal process of
with banteng males mating with kouprey females the premaxilla ends sharply, falling short of the nasal
and then with the female hybrids and backcrosses. by more than the length of the fi rst upper molar. The
Moreover, Hassanin et al. (2006) obtained DNA horns of cows are comparatively slightly bent inward, —-1
from a mounted specimen in the Natural History whereas those of bulls are “spiral: they point out, in, —0
349-47558_ch01_1P.indd 113 5/31/11 7:05 PM

114 Artiodactyla
and slightly back,” and the horn cores are directed ung”; the height of the withers of the adult bulls
laterally and slightly back, their tips bent upward. is about 160 cm, as opposed to 185 cm or more for B.
Verkaar et al. (2004) pointed out the anomaly that, bonasus; the hoofs are shorter, the forehoof <85 cm,
whereas the Y-chromosome DNA of this species is as opposed to >90 cm, and the hindhoof <90 cm, as
close to that of Bos bison, mitochondrial sequences opposed to >100 cm (Mohr 1952; Heptner et al., 1961).
align it, if distantly, with the Bos taurus group. After Bohlken (1967) found that the tuber malar is less de-
considering lineage sorting as a possible (if not particu- veloped than in B. bonasus, giving a smaller relative
larly plausible) explanation, they proposed instead that breadth across the facial skeleton; the frontal is nearly
the European bison is a hybrid species, in which bulls flat, as opposed to convex, behind the orbits; and, es-
of one of the European Pleistocene bison species (prob- pecially, the occiput is strongly constricted below the
ably Bos priscus) interbred with cows of an ancient Bos occipital crest. Although all the cited authors regarded
primigenius–like species, then with the hybrid cows, Caucasian bison as only subspecificially distinct from
and so on, until nuclear swamping of the latter had lowland bison, on the evidence they are consistently
occurred, while the original mtDNA was retained in (diagnosably) different. The difference in the occiput
the new hybridogenetic species. We agree that this is is not only relative to overall skull size, but is an abso-
the most plausible explanation. lute univariate difference (occipital constriction breadth
Two well-known breeding lines of this species are in B. caucasicus 101–130 mm (N = 7), compared with B.
preserved in captivity: the lowland and the lowland- bonasus 149–204 mm (N = 9). The horns are also some-
Caucasian lines. The former are descended solely from what smaller and more slender, on average (Bohlken,
bison ultimately from the Bialowieza region, so they 1967).
are the surviving purebred representative of the pres- Meanwhile, Flerov (1965) distinguished B. cauca-
ent species. The latter, the lowland-Caucasian line, are sicus from the European lowland bison as having “al-
descended from Bialowieza bison plus a single Cauca- ways two distensions” at the base of the horn cores;
sian male; that is to say, they represent Bos bonasus the preorbital depression is clearly defi ned; the nasal
with some Bos caucasicus genes. As described by Rau- cavity is smaller, so that the facial skeleton is low; the
tian et al. (2000), there is, in fact, a third line, the high- distance from the lower margin of the orbit to the al-
land line, descended from a mixture between the other veolus of third molar is about “one-third of the
two lines plus three American bison: in 1940, five high- toothrow length” ([sic]; this would seem to make it
land bison were brought to the Caucasian State Nature higher than in B. bonasus!). The nasal process of the
Reserve and bred there in an enclosure; in 1950 the fe- premaxilla ends bluntly in most cases and may con-
males were mated with males from the lowland- tact the nasals, or, if it does not, the distance between
Caucasian line; and fi nally, in 1968, they were all them is less than the length of the first molar. Flerov,
released into the reserve, where they now live wild. like Bohlken (1967), found that the Caucasian bison
Rautian et al. (2000) studied 15 male and 19 female skulls has a much less convex (usually almost flat) forehead,
of highland bison, collected between 1968 and 1994, as and the tuber malar is less developed. Flerov described
well as plentiful material of what we here call B. bonasus the horns thus: those of the females are noticeably
and B. caucasicus. They found, interestingly, that the bent inward; those of the males are more strongly
highland bison differ substantially from the other two, spiral, their cores pointing straight out to the side, or
mostly by being intermediate, but in some cases tend- even slightly forward, and their tips pointed upward.
ing in the direction of B. bison, and—in a few cases— As before, we are uncertain what to make of Flerov’s
deviating in an unexpected direction from all putative descriptions, since it is unclear whether they are con-
parents. This microevolution clearly deserves fuller sistent, and on what samples they are based.
study, but what is not so clear is that those authors Kretzoi (1946) described a new (subfossil) subspe-
were justified in describing a new subspecies, Bison cies, Bison bonasus hungarorum, based on a complete
bonasus montanus, on this mixed population! skeleton of a female and a cranial fragment of a male.
The differences would be fi rst, that the horn cores
Bos caucasicus Satunin, 1904 initially diverge horizontally, and are then slightly
Caucasian bison curved upward, with the tips curved backward; sec-
Differences between this species and Bos bonasus have ond, that the withers are much lower, compared with
been listed as follows: the pelage more clearly extends the pelvis; and third, that the size is small. He stated
-1— down to the fetlocks; the color is more of a warm sepia, that he had examined drawings and paintings from
0— as opposed to “fahlbraun mit ockerbrauner Schattier- the 17th and 18th centuries depicting Hungarian bi-
349-47558_ch01_1P.indd 114 5/31/11 7:05 PM

b o v idae 115
son, and these were consistent. On the other hand, he abundant, but his two photos of southern plains bi-
admitted that some living bison apparently have the son (B. b. bison, in his classification) are most striking,
same horn form, which he took as meaning that and quite unlike most plains bison one would see to-
some ancestry from the old Hungarian bison persists day, which would be B. b. montanae. Clearly much is
in them (!). to be learned about the former geographic variation
The DNA of this species is not known. Extracting of this species, as Krumbiegel indeed admitted in his
both mtDNA and Y-chromosome DNA from pre- paper.
served specimens will help to show whether this spe- The wood bison, B. b. athabascae, is generally rec-
cies, like B. bonasus, is of hybrid origin, or whether it ognized as a subspecies of American bison, but there
is a separate lineage. are complications, which must now be faced.
For Flerov (1965), B. b. athabascae is a large-sized
Bos bison Linnaeus, 1758 and primitive form that retains a great many charac-
American bison (“buffalo” in North America) ters of the extinct Bos (“Bison”) priscus, such as the
more rounded, less widened muzzle; the anterior
1758 Bos bison Linnaeus. C Kansas (fi xed by Hershkovitz,
ends of the premaxillae lacking the bladelike expan-
1957).
sion of B. b. bison; the nasals straight and not narrow-
? 1898 Bison bison athabascae Rhoads.
ing toward the front; the presence of a depression on
1899 Bison occidentalis Lucas. St. Michael, Alaska.
the forehead behind the nasals; the long, thick, less
? 1915 Bison americanus pennsylvanicus Shoemaker, nom. nud.
downcurved horn cores; and the narrow incisor-
1932 Bison bison oregonus Bailey. Malheur Lake, 43° 5' N, 119°
canine row.
W, Oregon.
Collecting data from the literature, Bohlken (1967)
1933 Bison bison haningtoni Figgins. Rock Creek, NE of South
concurred that the skull of B. b. athabascae is a great
Park, Colorado.
deal larger, on average, than that of nominotypical B.
1933 Bison bison septentrionalis Figgins. Near Palmer Merrick,
b. bison, with longer, more spreading horns.
Nebraska.
Any study of the differences between B. b. athabas-
1980 Bison bison montanae Krumbiegel. Montana.
cae, the wood bison, and other American bison is, of
Krumbiegel (1980) gave the following table of the three necessity, somewhat constrained, because the only
subspecies of North American bison that he recog- known population of B. b. athabascae, living in Wood
nized (here somewhat modified): Buffalo National Park in N Canada (spanning the
Alberta–Northwest Territories border), was forever
1. Bison bison athabascae—wood bison. Large;
destroyed by the 1925 introduction by the Canadian
long-bodied, long-legged. Horns long, more
government (apparently in the teeth of much opposi-
curved, tips often pointing forward, horn base
tion from the public) of a large number of plains bison
narrow. Hoof broader. Hair shorter, softer, not
from Montana. The later discovery of a presumed
curly. Color darker, especially the head, legs,
isolated population in the park, at Nyarling River,
and tail.
was supposed to have saved the integrity of B. b. atha-
2. Bison bison montanae—northern plains bison.
bascae, but it is now recognized that even this popula-
Smaller; shorter legs and a lower, longer body.
tion had not escaped gene flow from the introduced
Horns short, thick, tending to be covered by
plains bison, though how much would be in question.
the thick, upwardly directed hair on the top of
In any event, bison from the Nyarling River herd, se-
the head, the tips not pointing forward. Hoof
lected for their athabascae phenotype, were subse-
narrower. Hair longer, denser, more curly;
quently translocated to the Mackenzie Bison Sanctu-
“chaps” (hair on the upper forelegs) longer,
ary (Northwest Territories) and to Elk Island National
thick. Color lighter.
Park (N Alberta). (There is also a herd of plains bi-
3. Bison bison bison—southern plains bison. Small;
son, also originating from Montana, in Elk Island Na-
short-legged, like the previous taxon, but the
tional Park, kept separately from the putative wood
body higher and shorter. Horns longer, less
bison).
covered by long hair on top of the head. Chaps
It is in this context that genetic studies must be
longer and less thick, grading more into the
interpreted. Frequencies of red cell carbonic anhy-
hair of the foreparts.
drase alleles were different in B. b. athabascae (from
The evidence for the differentiation of Krumbie- the Elk Island National Park herd) from plains bison, —-1
gel’s new subspecies B. b. montanae is not all that but not more than the frequencies of plains bison in —0
349-47558_ch01_1P.indd 115 5/31/11 7:05 PM

116 Artiodactyla
Wichita Mountains Wildlife Refuge (Oklahoma) dif- Geist (1991b) took great exception to the idea that
fered from those in Santa Catalina Island Conservancy B. b. athabascae is distinct at all. He applied his fi nd-
in California and the National Bison Range in Mon- ings on deer (Geist, 1989) to bison, claiming that B. b.
tana (these two herds have a common origin). Inter- athabascae is, in fact, only an ecotype—that is to say,
estingly, the presumed hybrids in Wood Buffalo Na- its differences from plains bison are not genetic at all.
tional Park were quite close to the plains bison herd Reviewing many of the features deemed distinctive
in Elk Island National Park, and these, in turn, were by van Zyll de Jong (1986), he found that they are not
rather closer to B. b. athabascae than to other plains consistent, not only as concerns the Nyarling River
bison. herd but even, to some extent, when old pre-1925 pho-
In another study, a genetic comparison of the two tos are studied. The photos of extinct southern plains
groups of bison in Elk Island National Park showed bison from Krumbiegel (1980) and other sources do,
very low differentiation (Bork et al., 1991). Recall that indeed, also bear some similarity to wood bison. In
the plains bison there originate ultimately from Mon- addition, Geist (1991b) maintained that the Nyarling
tana, as did a component of the ancestry of the wood River bison that had been translocated to Elk Island
buffalo population. National Park, and those of their descendants that
A very thorough study on the differentiation be- were translocated elsewhere, were more reminiscent
tween plains bison (nominotypical B. b. bison) and of plains bison and appeared to have converged on
wood bison (B. b. athabascae) was conducted by van them; that is, the differences in size, body shape, and
Zyll de Jong (1986), using both observational data on features of the pelage were actually not hereditary at
living animals and morphometrics on crania. The all, but resulted from environmental plasticity. As for
study was again compromised by the hybridization the skull differences, he attacked morphometrics as
that had taken place, due to the introduction in the being completely invalid, because it did not take phe-
1920s of plains bison to Wood Buffalo National Park. notypic plasticity into account.
Van Zyll de Jong found that many Nyarling River bi- Van Zyll de Jong et al. (1995) responded by reexam-
son skulls indeed were very like old specimens of B. b. ining some of the characteristics that had been previ-
athabascae, though not completely so, and he was ously used by van Zyll de Jong (1986) to differentiate
constrained to selecting specimens. Nonetheless, in B. b. athabascae from B. b. bison, and they scored vari-
many of his craniometric analyses, the two putative ous populations of known origin for the development
taxa were completely separated (see, for example, his of the hump, the cape, the chaps, the frontal display
figure 6, based on six horn measurements and five hair, the ventral neck hair, and the beard. Strong dif-
cranial ones). The differences he found were, essen- ferences between presumably purebred plains bison
tially, that B. b. athabascae were larger than B. b. bison and predominately wood bison remained, casting
in the length and transverse diameter of the horn doubt on the view of Geist (1991) that these were all
cores, posterior cranial width, and basal skull length; subject to environmental modification.
but that the former were smaller in the width of the Given that there are no remaining pure B. b. atha-
skull across the masseteric processes and, curiously, bascae populations, with even the Nyarling River herd
the occipital condyles. Moreover, although based on being mixed to some degree, these positive results do
rather small sample sizes, B. b. athabascae was distin- suggest that there really was such a taxon as Bos bison
guished by the greater length of the long bones and athabascae, with its anteriorly high hump, reduced fore-
the pelvis, and the much greater width of the pelvis leg chaps, and so on. We, very clearly, do not support
across the ischia. B. b. athabascae has/had a “suddenly the disparagement of morphometrics/craniometrics,
tall” hump, which continues farther back along the since we have used this method throughout this book.
spine as well; whereas B. b. bison has a deeper neck, We can fi nd no evidence that (nonpathological) speci-
from which the hump rises slowly, and its hump mens, let alone whole populations, show such all-
does not decline as much behind. There were also pervasive environmental modification as to completely
external differences in B. b. athabascae: the color of override the genetic basis of cranial (or other) shape
the cape; the form of the frontal display hair and differences.
the length of the display hair on the upper forelegs When all is said and done, the subspecies (or spe-
(“chaps”); and the form of the ventral neck mane, cies?) B. b. athabascae no longer exists. The case is
making the beard stand out much more and appear somewhat analogous to that of Bos caucasicus; in both
-1— more pointed. cases, all we have now are some of their genes. The
0—
349-47558_ch01_1P.indd 116 5/31/11 7:05 PM

b o v idae 117
difference is that in the athabascae case we do not a type I error (a false positive). Therefore, a conserva-
know how much of its genome we possess, and how tive course is provisionally followed here.
much the selected lookalikes really are, in any real Analysis of the mitochondrial D-loop and cyto-
sense, wood bison. chrome b gene indicated that river and swamp buffa-
loes formed two distinct clades, separated by up to
Bos mutus (Przewalski, 1883) 270,000 years, indicating that they would have been
yak independently domesticated (Kumar et al., 2007). But
what their respective wild ancestors are is not clear.
1883 Poephagus mutus Przewalski. Nan Shan, at approxi-
There is, of course, a constant worry over the ge-
mately 32° 20' N, 95° E.
netic purity of wild Asian buffalo. Flamand et al.
The name mutus Przewalski, 1883, was placed on the Official
(2003) used microsatellites to examine the status of
List of Specific Names in Zoology by ICZN, Opinion 2027
the wild buffalo of Kosi Tappu Wildlife Reserve (Ne-
(2003).
pal), fi nding evidence for a little (but very little) gene
The complete mitochondrial sequence of domestic yak flow from domestic to wild animals, and rather more
was determined by Gu et al. (2007), who dated the di- gene flow the other way.
vergence between yak and domestic cattle at 4.38–5.3
2 Ma. Schaller (1998) found that, in contrast to wild and Bubalus arnee arnee (Kerr, 1792)
domestic Bactrian camels, wild and domestic yak are
1792 Bos bubalus arnee Kerr.
really not very different genetically.
1852 Bubalus buffelus macrocerus Gray. Nepal.
1912 Bubalus bubalis septentrionalis Matschie. Kukri-Mukri,
Bubalus Hamilton Smith 1827
an island on the seaward side of the Sundarbans.
Asian buffalo
This genus contains at least four living species: wild Relatively small in size; skull length usually <570 mm;
buffalo, tamaraw, and two species of anoa. There may relatively large teeth, >27% of the skull length; horn
even be more. span usually 1000–1200 mm; tip-to-tip distance nearly
Pitra et al. (1997) provided an alternative view of always <80% of the span. Color black, with contrast-
the relationships of the anoa. They sequenced two ingly white lower limbs below the knees and the
nuclear genes (Cyp19, and the promoter region of Lf ) hocks, and a white muzzle; tail reaching to the hocks.
and found that, when Capreolus was taken as an out- Formerly from the Sundarbans SW into Madhya
group, the anoa formed a clade with Boselaphus, Pradesh and Andra Pradesh, and NW into Nepal;
whereas other buffaloes (both Asian and African) now apparently occurs only in the Raipur and Bastar
were, as expected, a sister group to Bos. districts of E Madhya Pradesh and the Kosi Tappu
Reserve, SE Nepal.
Bubalus arnee (Kerr, 1792)
Asian wild buffalo Bubalus arnee fulvus (Blanford, 1891)
1792 Bos arnee Kerr. Koch Behar (as restricted by Harper, 1940). 1891 Bos bubalus var. fulvus Blanford. Mishmi Hills.
The name arnee Kerr, 1792, was placed on the Official List of
Very large size; greatest skull length usually >570 mm;
Specific Names in Zoology by ICZN, Opinion 2027 (2003).
toothrow length approximately 26%–28% of the skull
The karyotype of wild Asian buffalo is unknown, but length; horn span >1100 mm in most specimens; tip-to-
those of its putative descendants, the river and swamp tip distance as in nominotypical B. a. arnee. Color a
breed-groups of domestic buffalo, show differences: the lighter gray or brownish gray, and the limbs less con-
river buffalo has 2n = 50, the swamp buffalo 2n = 48 trastingly white; tail falling short of the hocks.
(Bongso & Hilmi, 1982). Brahmaputra Valley; formerly from Kuch Behar
The taxonomy follows Groves (1996b); in figure 1 to the Mishmi Hills, and S to the Chittagong Hills.
of his paper, the three mainland subspecies separate Still occurs sporadically over most of the Indian por-
cleanly on multivariate analysis, so there is a possibil- tion of the range, and marginally in Bhutan.
ity that they actually should rank as distinct species,
although the analysis was based on 10 skull measure- Bubalus arnee theerapati Groves, 1996
ments and, as only one of the three samples had more
1996 Bubalus arnee theerapati Groves. Mae Wong, Nakhon
than 10 specimens, there is a reasonable possibility of —-1
Sawan, Thailand.
—0
349-47558_ch01_1P.indd 117 5/31/11 7:05 PM

118 Artiodactyla
Relatively small in size; skull length <570 mm; toothrow line down the front and across the pasterns; generally
length 24%–27% of the skull length; horn span usu- a white crescent on the throat; groin light to white.
ally >1200 mm, the tip-to-tip distance >80% of the Tail length <18% of the head plus body length.
span. Groves listed localities for this species throughout
Formerly from the Irrawaddy Delta through Thai- Sulawesi, in more lowland areas.
land to Cambodia and probably Vietnam. Still occurs
in the Huai Kha Khaeng Wildlife Sanctuary, Thai- Bubalus quarlesi (Ouwens, 1910)
land, and in various places in Cambodia. mountain anoa
? 1792 Bos bubalus anoa Kerr. Suppressed by ICZN, Opinion

Bubalus arnee migona Deraniyagala, 1952
1349 (1985).
1952 Bubalus bubalis migona Deraniyagala. Yala National 1910 Anoa quarlesi Ouwens. Mountains of C Toraja, Sulawesi.
Park, Sri Lanka. The name quarlesi was apparently threatened by Bos bubalus
anoa Kerr, 1792 (see Groves, 1982d), but it was conserved
Skull relatively small and not sexually dimorphic,
by ICZN, Opinion 1349 (1985).
<570 mm in both sexes; occipital breadth especially
small, usually <260 mm in both sexes (compared Groves defined the species as being small in size (skull
with >260 mm in the females, 280 mm in the males, length 244–290 mm, toothrow length 65–80 mm); horns
in the mainland taxa). Horns, on average, smaller short, conical, cylindrical, 146–199 mm long; color
in all dimensions than in the mainland taxa; span dark brown to black in the adults, the hair tending to be
<1000 mm. Apparently almost lacking white mark- thick and woolly; legs dark, like the body, with whitish
ings (though in Groves & Jayawardene’s [2009] photos, or yellowish spots above the hoofs, or none at all; no
mud obscures much of the lower part of the limbs). white on the throat; groin light, but not white. Tail
Sri Lanka; now threatened by interbreeding with length <18% of the head plus body length.
domestic buffalo (Groves & Jayewardene, 2009). Groves listed localities for this species throughout
Sulawesi, in mountainous areas.
Bubalus mindorensis Heude, 1888 The relationship between the two species of
tamaraw anoa, and whether there really are only two species,
has been discussed by Schreiber & Nötzold (1995).
Braun et al. (2002) described their rediscovery of the type
It is clear, at any rate, that both the two putative spe-
specimen and gave some historical notes.
cies, especially B. quarlesi, are variable in color pat-
Custodio et al. (1996) gave a full account of this tern, size, length of the horns, and so on; there could
species. well be more taxonomic variation than hitherto
Fossil remains (but from a historic period) of a recognized.
second, even smaller, Philippine dwarf buffalo have Kikkawa et al. (1997), on the basis of mitochondrial
been described from Cebu (Croft et al., 2006). cytochrome b, found differences between what they
called lowland, mountain, and quarles anoas (1.2% be-
Bubalus depressicornis (Hamilton Smith, 1827) tween “mountain” and “quarles,” 3.3%–3.6% between
lowland anoa lowland and either of the other two); the sequence di-
vergence between anoas and water buffalo was itself
1827 Antilope (Anoa) depressicornis Hamilton Smith. Celebes.
only 3.33%. The meanings of “mountain” and “quar-
1853 Oreas platyceros Temminck.
les” are unclear, but it is interesting that the major
1865 Probubalus celebensis Rütimeyer.
differentiation among anoas seemed to be equivalent
1905 Bos depressicornis fergusoni Lydekker. No locality.
to the difference between anoas in general and other
Groves (1969b) defi ned the species as being large in Asian buffalo.
size (skull length 290–300 mm in the females, 298– Schreiber & Nötzold (1995) took a more cautious
322 in the males; toothrow length 82–98 mm); horns view of taxonomy. They pointed out that European
triangular in cross section in the adults, with marked zoos at the time had anoas deriving from four differ-
transfers ridges and an external keel (183–260 mm ent imports: the Berlin/Krefeld, Antwerp, Rotter-
long in the females, 271–373 mm in the males); color dam, and Leipzig groups. The fi rst of these corre-
black in the adults, the hair sparse, the woolly brown spond to “mountain” anoas, the other three to
-1— juvenile coat lost before maturity; lower segments of “lowland” ones, although they also differ somewhat
0— the limbs white or yellowish white, except for a black from each other. The Berlin/Krefeld stock have
349-47558_ch01_1P.indd 118 5/31/11 7:05 PM

b o v idae 119
2n = 44 or 45; the others, 2n = 48 (in one individual, After a period in which every slightly (or not at
2n = 47). The authors had visited Sulawesi and else- all!) differentiated population was given specific (later
where in Indonesia. They reported seeing mountain subspecific) rank, especially by Matschie (1906, 1918a,
anoa trophies for sale, and also a living captive ani- b), Lydekker (1913) united them all as subspecies of a
mal, in the Toraja highlands; probable lowland anoas single species, with a certain amount of cutting down
in the Gunung Tangkoko National Park on the N tip of subspecies. The forest buffalo appeared to such
of Sulawesi, and evidence of both kinds of anoas in or authors as Lydekker to be considerably more diverse
near the Dumoga-Bone National Park. They cited a than the big savanna buffalo (Cape buffalo). Christy
personal communication from Prof. Nawangsari Su- (1924) at fi rst divided up forest buffalo into six distinct
giri that a Lore Lindu anoa, physically resembling the geographic forms, but later lumped them all (Christy,
Berlin/Krefeld stock, had a similar chromosome 1929), including W African Syncerus brachyceros, into a
number (2n = 44), while a much more pale brown one single catchall S. nanus, which he did, however, keep
from the Pompangeo Mountains, E of Lake Poso, had separate from S. caffer. Even between forest and Cape
2n = 46, which is not recorded in any of the European buffaloes, however, a forced rapprochement was in
anoas. In Minahassa, Prof. Sugiri found anoas physi- the air. The single-species model seemed inexorably
cally resembling the Leipzig stock and with the same to fi nd more favor, for whatever reason, and tended
2n = 48, and Schreiber & Nötzold (1995) also de- to be adopted by later authors. It was only in the
scribed having seen a fi lm of anoas in the Gorontalo 1970s that Grubb (1972) began to query the basis of
district that resembled the Leipzig stock. Christy’s (1929) arrangement.
In a later publication, Schreiber et al. (1999) found The taxa of this group can be arranged in a ring-
that the two phenotypes of anoas in the European shaped morphocline, from the C African rainforest
zoos also have extremely divergent cytochrome b via W Africa and Sudan to the E and S African sa-
haplotypes, suggesting a divergence time of 1.25 Ma. vanna (Grubb, 2000a), but within this morphocline
The authors also made distance trees (including some there were nonetheless some sharp breaks.
previously published data), and used their own previ- A study by van Hooft et al. (2002) found that, in
ous work on MHC class II genes, and these gave the mtDNA D-loop hypervariable region I sequences, Af-
same results. Given that some of the studies that they rican buffalo fall into two extremely distinct clades,
incorporated into their analysis had given no evi- separated by a 6.6% sequence divergence. The fi rst
dence that the species of anoa studied were as clade consists entirely of buffaloes from Kenya, Tan-
claimed, they recommended that attention to an ex- zania, Uganda, and S Africa (i.e., Cape buffalo); the
ternal phenotype be given by all future workers— a second, mainly of forest buffaloes from Cameroon,
recommendation that we fully endorse. Gabon, Angola, and (supposedly) Namibia, plus two
buffaloes from Uganda, one from S Africa, and one
Syncerus Hodgson 1847 from Kenya or Tanzania. They also studied the
African buffalo Y-chromosome microsatellite INRA008; in this, all
African buffalo were historically distributed from buffaloes from Kenya, Tanzania, and South Africa
Senegal in the W to Eritrea and Somalia in the E, (N = 66) had haplotype a, while two from Cameroon
and S to the Cape of Good Hope. They seem to have had haplotype b, two from Gabon had c, and one of
been absent only from extremely arid areas, such as the two from “Namibia” had d; of two “forest buf-
Namibia (except the extreme N) and much of Soma- falo” of unknown origin, one had d, and the other
lia (except along the Shebeyli and Juba rivers). They had a different haplotype, e. On the basis of mismatch
have been exterminated from parts of this range, es- distributions, the authors postulated a Middle to Late
pecially around the edges, including Bioko (Butyn- Pleistocene population expansion of Cape buffalo, ex-
ski et al., 1997). They live in savanna, thornbush, panding from E to S Africa.
light forest, rainforest, and mountain habitats. One We would like to comment on this. A few Cape
expects, at any rate, only limited taxonomic diver- buffalo had forest mtDNA haplotypes, but there were
sity in such a eurytopic taxon. Nonetheless, there no “wrong” Y-chromosome haplotypes; this suggests
has always been admitted to be a rather dramatic strongly that the distribution of forest buffalo— and,
contrast between “savanna” buffalo, called Cape buf- presumably, of the rainforest itself—was much wider
falo, which are large, long horned, and black, and in the past, and that vegetation changes (as van Hooft
rainforest buffalo, which are small, short horned, et al. [2002] themselves mention) have enabled Cape —-1
and red. buffalo to expand, replacing forest buffalo populations —0
349-47558_ch01_1P.indd 119 5/31/11 7:05 PM

120 Artiodactyla
by nuclear swamping. This is what one would expect, what larger, with a greater horn span but a lesser cur-
given the huge advantage that Cape buffalo bulls would vature than the others. W Rift forest buffalo are not
surely have over forest buffalo bulls. different from these either. We consider this group,
The association of two buffaloes from Namibia too, as a homogeneous taxon, for which the prior
with forest buffalo rather than with Cape buffalo, in available name is Syncerus nanus.
both mtDNA and Y-chromosome DNA, was raised S. caffer and the S. aequinoctialis / S. brachyceros group
several times in van Hooft et al.’s (2002) paper. The have a greater span and palm width than S. nanus, and
specimens are identified in their table 1 as being from a somewhat greater curve length; while the S. aequi-
the Antwerp Zoo, “one with father from Windhoek noctialis / S. brachyceros group, when compared with
and one with mother from Okahandja”; on the grounds S. caffer, has a smaller horn curve and a somewhat lower
of relevance, presumably the one with the father from span, contrasting with a relatively slightly greater palm
Windhoek was a male used for the Y-chromosome width.
study, while both were used for the mtDNA study. Within W Africa, those from SW Nigeria, Togo,
Actually, buffalo occur in Namibia only in the Caprivi Ghana, and Ivory Coast are larger than those from SE
Strip and W along the Angolan border just into and C Nigeria; and of two specimens from Senegal/
Ovamboland—not nearly as far S as Windhoek or Oka- Gambia, one falls in the same far W group, while the
handja (Shortridge, 1934:440, and see the map oppo- other is rather closer to SE– C Nigeria. A specimen
site his p. 578). Under these circumstances, it is evi- from Garoua (N Cameroon) falls in with SE & C Ni-
dent that the parents of the two Antwerp buffaloes in geria (as discussed below). The far W buffalo differ in
the study must have been obtained from dealers or their wide mastoids and horn span, compared with
from a Namibian zoo; one could not possibly say with- the relatively shorter skull.
out inspecting them personally whether they were Buffalo from the CAR (Ndele, Bangui, 6° N 21° E,
phenotypically Cape or forest buffaloes. 7° N 19° E) are clearly different from those from the
We now report on our craniometric analysis, in Uele district (Pandu, Ibembo, Buta) and those from the
which we divided the very large sample into 19 sub- Ubangui (Dubulu, Bosobolo, Molegbwe, Kekongo,
samples. We found that discrimination on some mea- Libenge, Duma, Budja, Ubangi River) in being larger,
surements, especially horn span, was extreme, and in with a greater horn span compared with horn palm
the end we concentrated on these relatively few mea- width.
surements, which enabled us to incorporate more in- The buffalo from the W Rift fall into three cleanly
dividuals per sample. distinct groups: S. caffer, “intermediates,” and S. na-
The fi rst important conclusion is that Cape buffalo nus. The “intermediates” are somewhat closer to S.
from S Africa, E Africa, and the W Rift Valley (or W caffer but, strikingly, do not overlap with them, nor is
Rift) are all the same—we could not separate them there overlap elsewhere. The differences are all con-
on discriminant analysis. Those from South Africa cerned with absolute size, horn span being most em-
average somewhat greater in the curved length of the phasized, followed by curve length.
horn, and less in horn span, but these differences are The position of the “intermediates” is perhaps
extremely small, and on average only. We consider somewhat unexpected. Rather than spanning the
them a homogeneous taxon, for which the prior morphospace between S. caffer and S. nanus, it would
available name is Syncerus caffer. appear that there is, in fact, a discrete population of in-
All samples that might be classed as S. brachyceros termediates. For these, the name S. mathewsi Lydekker,
and S. aequinoctialis (Sudan; Uele; N Nigeria / Chad) are 1904 (type locality: Virunga volcanoes), is available,
more or less the same. The Uele localities include Ango, with, as its synonym, S. cottoni Lydekker, 1907 (type
Gangala na Bodio, Aku, and Faradje. The type of S. locality: Kasindi, W side of Lake Edward). The locali-
neumanni Matschie, 1906 (from Chagwe [= Kyagwe], ties for the three sharply different taxa in the W Rift
Uganda, just W of the Nile where it leaves Lake Vic- are as follows:
toria), identifies itself with this group in the cranio-
1. Syncerus caffer—Rutshuru (Katsibwe, 0.41° S,
metric analyses.
29.16° E, and S of Lake Edward), Rwindi (0.47°
Forest buffalo from Cameroon (and other W-C Af-
S, 29.18° E), Parc National Virunga (including
rican areas, especially Gabon), S-C DRC, and SW DRC
Kamohorora, 0.41° S, 29.40° E), Ishasha
plus N Angola form another homogeneous group, al-
(0.31°–1.00° S, 29.45° E), Usumbura (3.23° S,
-1— though the SW DRC / Angola ones tend to be some-
0—
349-47558_ch01_1P.indd 120 5/31/11 7:05 PM

b o v idae 121
29.21° E), Semliki (0.45° N, 29.50° E), Gabiro Localities of the skulls used in this analysis are as
River (1.31° S, 30.24° E), Lake Rugwero and follows:
Kataranga (Burundi), and Beni (0.29° N,
—S-C DRC nanus—Itana, Moma, Idembe,
29.29° E)
Eligampangu, Itoko, Walatunga, Kylia,
2. Syncerus mathewsi—Kasindi (0.03° N, 29.41° E),
Kikaku
Semliki, and Virunga volcanoes.
—SW DRC / Angola—Kinshasa, Banana,
3. Syncerus nanus—Watalinga (0.45° N, 29.48° E),
Kwamouth, Mpu, Bas-Congo, Bashongo,
Kylia (Kapanatu), Kikuku (Kombo River, Konto,
Temvo, Matadi, Kunungu, Bokala, “opposite
0.59° S, 29.04° E), and simply “Rutshuru.”
Boka,” Bokoro, Tsangata, Lemba, Lake
In other words, within this relatively restricted re- Leopold, Bokuma, Franz-Joseph Falls,
gion there are three quite distinct taxa. There may or Quingombe, Ambriz, Novo Redondo,
may not be some marginal interbreeding between Braentul, Pama, Binga, Bembe
them, but we have no evidence for it in our dataset.
Samples from N Nigeria and Chad (Bedadi, Mugur,
It is clearly apparent from our data that buffalo
Fort Lamy, Lake Chad, Bahr Salamat, “10° N, 20° E”),
from the SE part of the DRC (mainly Katanga) are fully
from the CAR (N Gaoundere, Kentu, Ndele, Bangui,
S. caffer. Localities for them are Luisa (7.12° S, 23.25°
four out of five of the specimens from Garoua, plus “8°
E), Katombe (9.16° S, 24.40° E), Bushimaie River (6.03°
N, 21° E,” “6° N, 21° E,” and “7° N, 19° E”), and from
S, 23.75° E), and Lukonzolwa (8.45° S, 28.40° E).
SE and C Nigeria (N Bamusu, Dundo, Mbam Geh,
The SE Nigeria sample is not different from that
Katsena, Gabim Ba, Washishi, Mamfe, and one out of
from the rainforest of S Cameroon (i.e., S. nanus), but
five of the specimens from Garoua) are rather different
both are different from the sample from the CAR.
from each other; the first two groups, especially. are
The CAR sample differs from rainforest S. nanus in
different from SE and C Nigeria. These latter have a
its greater span, longer skull length, and greater palm
narrower span, and a lesser curve and palm width,
width. The major part of a sample from Garoua (N
than the first two. As shown previously, they are, in
Cameroon) belongs with the CAR group: four of the
effect, a larger version of S. nanus, while the other two
five available skulls from Garoua fall clearly into that
groups are different, and somewhat (but not com-
group, although the fi fth one does fall in the range of
pletely) different from each other. The prior available
SE Nigeria.
name for the Chad buffalo is S. brachyceros Gray, 1837
W Rift S. nanus are closest to those from S-C DRC,
(synonym S. bornouensis Hamilton Smith, 1842; both
and next closest to S Cameroon; while skulls from
names with Bornu as the type locality).
SW DRC plus N Angola are different, on average.
Univariate statistics for the working groups of Af-
Differences are that those from SW DRC / Angola
rican buffaloes can be found in table 24, summarized
have a greater span, followed by higher values for great-
using the currently available taxonomic names.
est skull length, palm width, and mastoid breadth—
The univariate differences mostly overlap with
contrasting slightly with the not much wider horn
each other. It is on multivariate analysis that we
curve.
found strong, nonoverlapping distinctions between
When we compare this SW DRC / N Angola
taxa. We summarize our fi ndings as follows:
group with South African S. caffer and S-C DRC S.
nanus, we fi nd that, even though this group does dif- —The sample called Syncerus aequinoctialis differs
fer somewhat from rainforest S. nanus, it shows no only a little from S. caffer; the greatest average
tendency to resemble S. caffer, which is much larger difference is in the horn length on the curve.
in all measurements than either of the S. nanus —The sample called S. brachyceros overlaps
groups (most especially in the horn span). The SW strongly with S. aequinoctialis, and is almost
DRC / Angola buffalo are what have been called certainly the same.
“forest-edge nanus,” and are often assumed to be dif- —The sample called S. mathewsi mostly overlaps
ferent from rainforest nanus as a result of gene flow with the two above, so it is “intermediate” in a
from S. caffer. On the basis of this analysis, however, way, but the span, especially, is relatively
there seems to be no evidence that this is what is go- small.
ing on; they are simply forest-edge rather than deep —Excluding its forest-edge component, S. nanus
forest animals, and have adapted accordingly. is absolutely different from any of the above,
—-1
—0
349-47558_ch01_1P.indd 121 5/31/11 7:05 PM

122 Artiodactyla
Table 24 Summary of univariate statistics for the working groups of African buffaloes
Gt l Mastoid br Palm w Span Horn l curve
caffer 448–575 241–316 166–274 725–1340 660–1160

aequinoctialis 449–528 150–317 170–259 562–1030 530–945
brachyceros 433–599 241–266 178–250 740–1110 545–860
Ubangi/Uele/CAR 422– 486 220–252 90–260 370–825 400–700
intermediates
mathewsi 450–508 259–302 167–227 621–745 530–720
† ‡
nanus Central 394– 468* 137–257 105–161 341–655 410–690
nanus West 411– 492 199–276 119–167 345–720 350–635
*Forest-edge nanus (S DRC) goes up to 520.

†
Forest-edge nanus goes up to 303.
‡
Forest-edge nanus goes up to 295.
especially in palm width (which distin-

guishes it absolutely from the closest sample, Syncerus brachyceros (Gray, 1837)
S. mathewsi) and in span. Lake Chad buffalo
—The Ubangi/Uele/CAR intermediates are not
1866 Syncerus aequinoctialis (Blyth).
the same as S. mathewsi; the mastoid breadth is
1872 Syncerus centralis (Gray).
absolutely larger, and the specimens are very
1906 Syncerus azrakensis (Matschie).
much more variable, as expected for a hybrid
1906 Syncerus neumanni (Matschie).
swarm. These are the genuine intermediates,
‘ Syncerus solvayi (Matschie).
and constitute almost the only evidence for
intermediacy between any two taxa. Size smaller; horns widely spread laterally, but much
less curved, with the length of each horn noticeably less
The general impression of African buffaloes as a
than the total span; bases of the horns greatly expanded,
single species, grading from large black buffaloes in
nearly meeting in midline, but not forming a convex
the savanna to small red ones in the forest, is mislead-
boss. Color dark brown, usually only slightly reddish,
ing. Forest and savanna buffaloes are sharply distinct,
apparently never quite black. Ears meagerly fringed.
do not grade into one another, and are distributed in a
Ethiopia and Sudan W through the savanna belt
more mosaic fashion than has previously been recog-
to Senegal. In C Africa, N of the Shari River. In E Af-
nized. For example, red buffaloes (of which, unfortu-
rica, this is evidently the common buffalo of most of
nately, no specimens appear to be available) occur in
Uganda: along the Nile to Lake Victoria, and along
some montane forests in Ethiopia (Yalden et al., 1984).
the E side of the Great Lakes at least as far as the Ka-
There are, therefore, four species within the ge-
zinga Channel. It appears to enter Kenya in the Lake
nus Syncerus.
Turkana region.
Grubb (1972) called this taxon S. aequinoctialis, and
Syncerus caffer (Sparrman, 1789)
applied the name S. brachyceros to what we are here
Cape buffalo
regarding as hybrids between S. brachyceros and S.
Size very large; horns widely spread laterally, each
nanus; our study of more abundant material has
horn strongly curving down, below the skull base,
shown that (as noted above) the buffalo found around
and then up, with the curved length of each horn it-
Lake Chad, including the type locality of S. brachy-
self nearly equaling the span; bases of the horns greatly
ceros, is the present species.
expanded, nearly meeting at the midline of the fore-
head, where each base forms a convex boss. Color jet-
Syncerus mathewsi (Lydekker, 1904)
black. Ears meagerly fringed.
Virunga buffalo
The whole of the S and E African parts of the range
of the genus, extending into the Rutshuru plains of E 1907 Syncerus cottoni (Lydekker).
DRC, and N into Kenya, at least as far as Mt. Elgon and
-1— Size similar to S. brachyceros; horns about the same
the N Guaso Nyiro.
0— length, but less widely spread, with the length of each
349-47558_ch01_1P.indd 122 5/31/11 7:05 PM

Table 25 Univariate statistics for working groups of
Syncerus Table 25
Horn l Horn l
Gt l Mastoid Palm w Span curve Gt l Mastoid Palm w Span curve
neumanni DRC nanus

Mean — — 191.00 875.00 710.00 Mean 425.93 227.14 131.24 458.06 503.82
N — — 1 1 1 N 14 14 17 17 17
Sudan aequinoctialis Std dev 11.118 11.935 10.053 53.150 67.489
Mean 502.52 264.09 210.77 855.77 753.10 Min 410 210 119 385 410
N 21 22 22 22 21 Max 448 257 157 592 650
Std dev 18.029 32.607 15.350 79.950 92.053 DRC—Angola nanus
Min 463 150 192 693 640 Mean 436.72 222.77 145.39 481.85 446.56
Max 528 317 259 1030 945 N 46 44 54 55 52
Uele aequinoctialis Std dev 17.855 13.518 36.589 56.398 142.526
Mean 491.93 267.00 198.15 757.79 684.23 Min 394 196 115 362 122
N 14 10 13 14 13 Max 468 251 295 655 690
Std dev 23.950 21.756 20.145 99.091 74.745 W-C Africa nanus
Min 449 236 170 562 530 Mean 420.09 211.26 129.30 431.15 459.58
Max 522 295 233 948 760 N 23 23 33 33 33
W Africa nanus Std dev 14.823 12.509 13.190 55.710 73.300
Mean 454.23 229.64 152.00 527.69 519.09 Min 378 187 105 341 360
N 13 14 14 13 11 Max 441 237 161 580 655
Std dev 24.177 13.675 17.449 68.930 60.449 S DRC nanus
Min 411 212 125 447 415 Mean 492.04 273.43 213.77 828.39 780.48
Max 492 250 185 720 585 N 23 21 22 23 21
E Africa caffer Std dev 16.316 17.113 21.928 91.052 102.187
Mean 506.92 283.84 222.04 1003.25 906.36 Min 463 244 185 623 615
N 25 25 24 24 22 Max 520 303 261 1000 955
Std dev 23.615 15.283 22.162 123.869 112.981 Ubangui intermediates
Min 448 250 187 798 750 Mean 442.67 234.80 161.29 586.21 544.86
Max 558 312 274 1340 1160 N 6 5 34 33 35
S Africa caffer Std dev 17.224 11.520 20.366 120.857 82.091
Mean 497.50 275.00 227.81 913.45 874.08 Min 422 220 124 370 400
N 38 25 27 44 37 Max 464 252 207 825 700
Std dev 18.176 13.216 18.874 84.392 84.271 Uele intermediates
Min 459 241 201 725 720 Mean 445.00 222.50 151.80 498.60 495.00
Max 543 300 258 1150 1060 N 3 2 5 5 5
W Rift caffer Std dev 9.849 .707 15.106 68.039 59.477
Mean 516.52 284.78 222.14 955.11 856.93 Min 437 222 132 405 430
N 25 23 28 27 27 Max 456 223 174 595 590
Std dev 25.810 16.384 19.529 94.331 75.655 W Africa brachyceros
Min 479 251 166 820 660 Mean 443.72 234.96 142.13 514.03 541.92
Max 575 316 249 1150 1020 N 25 25 32 31 13
CAR cf. brachyceros Std dev 19.403 20.163 11.530 93.087 78.357
Mean 467.60 — 184.45 657.89 587.31 Min 411 199 119 345 350
N 10 — 22 19 13 Max 474 276 167 690 635
Std dev 13.753 — 36.471 70.806 49.859 W Rift intermediates
Min 446 — 96 540 450 Mean 478.50 273.00 192.00 670.75 638.75
Max 486 — 260 774 640 N 4 4 4 4 4
N Nigeria and Chad brachyceros Std dev 26.388 20.281 25.377 56.240 90.035
Mean 495.00 251.00 204.50 892.58 750.17 Min 452 259 167 621 530
N 7 4 12 12 12 Max 508 302 227 745 720
Std dev 50.484 12.247 21.682 112.353 89.668
Min 433 241 178 740 545
Max 599 266 250 1110 860
—-1
—0
349-47558_ch01_1P.indd 123 5/31/11 7:05 PM

124 Artiodactyla
horn only somewhat less than the total span; bases of Nigeria, and farther E in the Uele region. Both of
the horns expanded, but flattened, not forming a con- these intermediates are between Syncerus nanus and
vex boss. Color dark brown to black, often with red- S. brachyceros.
dish tones, especially in the cows, which are often
reddish animals with a dark dorsal stripe and dark Pseudoryx Dung et al., 1994
shanks. Ears with a fringe of hair. The discovery of the saola, Pseudoryx nghetinhensis, in
Restricted to a the forested mountainous area the early 1990s (Dung et al., 1993) sent shock waves
from the Virunga volcanoes N along the W side of through both mammalogy and the conservation
Lake Edward. movement, and signaled the start of a series of discov-
As noted by Grubb (1972), these buffalo are, in a eries of unanticipated new mammals in Vietnam and
sense, intermediate between the large savanna buf- Laos.
falo and S. nanus, but further evidence, since that At first there was some discrepancy over the affini-
date, has indicated that they are not simply inter- ties of this new genus. A resolution of this initial-
grades, but form a homogeneous taxon distinct from seeming uncertainty over the proposed relationships—
all others in the region. does it belong to the Bovini or to the Caprini?—was
proposed simultaneously by Gatesy & Arctander
Syncerus nanus (Boddaert, 1785) (2000) and by Groves & Schaller (2000). In the first in-
forest buffalo, red buffalo stance, three DNA sequences (12S and 16S, k-casein
and b-casein) placed Pseudoryx in the Bovinae, most
The synonymy is complex, depending on whether the W
convincingly (with 85% bootstrap support) as sister to
African populations, which are somewhat larger in size,
Bos in the k-casein case (although by a different solu-
with a slightly greater horn span, on average, are
tion, in a nonsense-tree, using a-lectalbumin intron 1).
recognized as a distinct subspecies.
The second paper reinterpreted the morphological
Very small in size; horns short, not much curved lat- data in this context and produced a parsimonious tree
erally, but mainly curved upward, the length greater incorporating Pseudoryx in the Bovini. At again almost
than the total span, with this discrepancy greater in the same time, Hassanin & Douzery (1999b) used both
C than in W African populations. Generally light red- mitochondrial and nuclear sequences to show Pseudo-
dish brown, with dark markings developing on the ryx nested well within the Bovini, slightly closer (but
limbs and the shoulders; sometimes entirely black. with not very strong support values) to Bos than to the
Ears very large and prominent, with heavy fringes on buffaloes; subsequently, Hassanin & Ropiquet (2004)
the lateral (i.e., inferior) edges, the long, thick hair be- concluded that it was one of three basic branches (with
ginning well inside the inner surface of the pinna, Bos and the buffaloes) in the Bovini.
typically in two transverse strips.
Forest belt of W and C Africa. Pseudoryx nghetinhensis Dung et al., 1994
As noted above, there are slight average differences saola, spindlehorn, Vu Quang ox
between W and C African populations, and there are Even today, very little is known about this species;
intermediate populations between S. nanus and S. see Dung et al. (1994) and Schaller & Rabinowitz
brachyceros in S Nigeria, the CAR S of the Shari River, (1995). Groves & Schaller (2000) gave a more detailed
and the Uele district. Grubb (1972) used the name S. description, described and mapped its distribution,
brachyceros for these intermediate populations; abun- and considered its phylogenetic position (see above).
dant new evidence suggests that they are not homoge-
neous, and represent a fairly narrow hybrid zone. Tribe Boselaphini Knottnerus-Meyer, 1907
The two living genera of this tribe do not appear to be
Within each of these species, there is much more ho- very closely related. As far as the skulls are concerned,
mogeneity than had previously been recognized. So A. W. Gentry (1992) could find only a few character
far from relations among them being everywhere states in common: the horn cores are keeled, much in-
clinal, they are mostly sharply distinct where their clined, and inserted well behind the orbits; the mastoid
ranges approach each other (even in the W Rift re- is narrow; and the supraorbital pits are situated far
gion, to which Syncerus mathewsi is restricted); the forward and widely apart above the front of the orbits.
only true intermediates we have found are along the The two genera have very different karyotypes: in
-1— N edge of the C African forest: in N Cameroon, in NE Boselaphus, there is a Y-autosome 14 translocation, as
0—
349-47558_ch01_1P.indd 124 5/31/11 7:05 PM

b o v idae 125
in the Tragelaphini, whereas this is not the case in Tetracerus quadricornis iodes Hodgson, 1847
Tetracerus. Therefore, Benirschke et al. (1980) pro- Similar in size, but with wider nasals; horns smaller;
posed that Boselaphus is sister to the Tragelaphini, or mean length of the posterior horn 73.5 mm, of the
possibly even to the lesser kudu (here, Ammelaphus), anterior horn 20.7 mm. Fawn above, light fawn be-
with which it shares a fusion of the X chromosome low, this tone often confi ned to a very narrow mid-
with the same autosome. line streak; foreleg with (at most) a vague, dark brown
line, often interrupted at the knee; median dorsal re-
Boselaphus de Blainville, 1816 gion not darkened. Nose diff usely darker.
2n = 46 in both sexes; as in the Tragelaphini, the Y Nepal and Champa.
chromosome is fused to autosome 14, but also there
is a fusion of the X chromosome to autosome 14— a Tetracerus quadricornis subquadricornis (Gray, 1843)
character shared, curiously, with Ammelaphus (Be- Size smaller; mean skull length of the males 187 mm.
nirschke et al., 1980). Nasal bones very broad. Only a single pair of horns
at any age; horns rather long, averaging 83.5 mm.
Boselaphus tragocamelus (Pallas, 1766) Red-fawn to more olive; underside whitish at the
nilgai midline, pale yellow laterally; foreleg line very
Groves (2003) could fi nd no marked geographic varia- vague, or absent; median dorsal region slightly dark-
tion in this species, with the possible exception of a ened; midline of the nose sometimes sharply darker.
single unusual cranium from Bengal. Tail short.
S India: Palkonda Hills, Madras, Dharwar,
Tetracerus Leach, 1825 Kuckanalla.
2n = 38; all chromosomes are acrocentric, and there is
no indication that any of them are compound, that is, Tribe Tragelaphini Blyth, 1863
there are no autosome / sex chromosome transloca-
Grubb (2004a) noted that, strictly speaking, the name
tions (Benirschke et al., 1980).
Tragelaphini is antedated by Strepsicerotini Gray, 1846
The karyotype, therefore, is very different from
(published as “Strepsiceriae”).
Boselaphus.
As in the Boselaphini, the horn cores are keeled and
Tetracerus quadricornis (de Blainville, 1816) the cranial roof is horizontal. As in both the Bosela-
chowsingha, four-horned antelope phini and the Bovini, the mastoid is narrow and, in
the last two mandibular premolars, the paraconid is
For the full synonymy of this species, see Groves
differentiated from the parastylid (A. W. Gentry,
(2003).
1992).
Groves (2003) recognized three quite distinct sub- 2n is as follows (males/females): greater kudu,
species. It should be noted that in the two four-horned 2n = 31/32; lesser kudu, 2n = 38/38; nyala, 2n = 55/56;
subspecies, the smaller, anterior horn pair do not de- sitatunga, 2n = 30/30; bushbuck, 2n = 33/34; common
velop until the second year of life. eland, 2n = 31/32; all have a Y-autosome translocation
(Benirschke et al. 1980).
Tetracerus quadricornis quadricornis Benirschke et al. (1980) postulated that the lesser
(de Blainville, 1816) kudu (plus the nilgai) and the nyala separated fi rst,
Relatively large; mean skull length of the males and the original Y chromosome was subacrocentric;
192.9 mm. Nasal bones narrow. Four long horns; mean then a pericentric inversion occurred in the Y chro-
length of the posterior horn 90.7 mm, of the anterior mosome (making it submetacentric), after the separa-
horn 48.6 mm (Rajasthan), 31.7 mm (elsewhere). Yellow- tion of the lesser kudu and the nyala. Benirschke et al.
fawn above, creamy or creamy fawn below, this tone (1980) proposed the following phylogeny:
often confi ned to the midline; forelimbs markedly
((nyala) (lesser kudu, nilgai) ((bushbuck) (bongo)
blackened down the anterior surface; median dorsal
(greater kudu, sitatunga, eland))).
region darker than rest of the upperside; tail long,
with a bushy white tip. Willows-Munro et al. (2005) compared mtDNA
From Rajasthan and Gujarat E to Bengal and S to and nDNA data (the latter being introns). The differ-
Karnataka and Andra Pradesh. ences were as follows: —-1
—0
349-47558_ch01_1P.indd 125 5/31/11 7:05 PM

126 Artiodactyla
1. The lesser kudu, sister to all the rest in the We examined skulls in the Natural History Mu-
mitochondrial tree, was closely associated seum (London) and elsewhere, and found that their
with the nyala in the nuclear tree. morphological relationships cohere very nicely with
2. In the mitochondrial tree, the bushbuck are those of Ropiquet’s molecular-based taxonomy.
sister to the sitatunga plus the bongo, with the
mountain nyala as sister to all three; yet in Nyala Heller, 1912
the nuclear tree, the bushbuck, the mountain 2n = 55 (males), 56 (females); the Y chromosome is
nyala, and the sitatunga plus the bongo are translocated onto chromosome 14, as in other Tragela-
three equal branches. Moodley et al. (2009) phini, but there is no inversion of the Y chromosome,
noted that the bushbuck sequences used by unlike all other Tragelaphini, with the exception of
Willows-Munro et al. (2005) were, in effect, Ammelaphus (Benirschke et al., 1980).
T. sylvaticus. Anterior median tips of the nasals very long; lat-
3. In the mitochondrial tree, the eland, the eral prongs virtually absent. Several supraorbital fo-
greater kudu, and the mountain nyala / bongo/ ramina, but only the anterior ones large; foramina
sitatunga/bushbuck clade are three equal sitting in an elongated depression. Lacrimal vacu-
branches; whereas in the nuclear tree, the ity relatively short, not reaching the maxilla. Malar-
eland is sister to the rest, then the greater maxillary suture as in Tragelaphus, but more deeply
kudu, then the fi nal clade. penetrating anteriorly. Auditory bulla small, angular;
postglenoid foramen an elongated oval shape, with a
As every one of these branches in both trees is
heart-shaped anterior margin. Basioccipital of even
highly supported using Bayesian posterior proba-
width between the tuberosities, which are low, but
bilities and both ML and parsimony bootstraps, the
angular. Mesopterygoid fossa forming a wide U; vo-
branching patterns seem real and presumably de-
mer forming a ridge. Free ends of the paroccipital
note past episodes of hybridization. In par tic u lar,
processes short, slightly curved. Buccal styles of the
the nuclear (but not the mitochondrial) relationship
upper molars very prominent.
of the lesser kudu (here, the genus Ammelaphus) and
According to Ropiquet (2006), Nyala has been a
the nyala (here, the genus Nyala) strongly suggests
separate genus since the Late Miocene.
that the ancestors of the lesser kudu have undergone
nuclear introgression from the proto-nyala.
Nyala angasii Angas, 1849
Hybridization is well known between bongos and
nyala
sitatungas; a less expected hybrid was between an
eland and a kudu. This strange-looking animal, a male, Grubb (2004a) concluded that this name must not be
was illustrated, and its genetics described, by Jorge et al. attributed to Gray, as previously assigned, but to Angas.
(1976). It was sterile.
A SE African species that has no geographic variation,
The genera recognized here follow Ropiquet
as far as is known.
(2006:figure 9), who proposed the following phylog-
eny (using common names):
Tragelaphus de Blainville, 1816
(lesser kudu (nyala (eland, greater kudu) Anterior median and lateral nasal tips more or less
(bushbuck (mountain nyala (sitatunga, bongo)))). equal in length, the median ones being somewhat
longer. One to several foramina in a shallow, elon-
In her figure 18, she indicated a separation time for
gated depression in the frontals. Lacrimal vacuity
the lesser kudu well back in the Late Miocene; and
long, usually reaching the maxilla. Malar-maxillary
of the greater kudu, the eland, and the bushbuck/
suture straight, more or less oblique. Bulla fairly
sitatunga, toward the end of the Late Miocene. The
rounded or somewhat angular, small; postglenoid fo-
nyala is not represented in the figure, but she later
ramen forming an elongated oval. Basioccipital of
specified (Ropiquet, 2006:66) that “le nyala (N. angasii)
even width, or slightly constricted between the tuber-
a divergé des autres Tragelaphini juste après A. imber-
osities, forming distinct, often rounded ridges. Mesop-
bis.” She specifically linked this recognition of several
terygoid fossa forming a wide V or a narrow U; vomer
genera to their “divergence ancienne (Miocène su-
rounded, not ridged. Free ends of the paroccipital pro-
périeur)” (Ropiquet, 2006:67), and we agree with her
cesses long. Buccal styles on the upper cheekteeth very
employment of this criterion.
-1— prominent (the least so in T. scriptus).
0—
349-47558_ch01_1P.indd 126 5/31/11 7:05 PM

b o v idae 127
According to Ropiquet (2006:figure 18), this genus T. locorinae, T. laticeps, T. heterochrous, T.

separated from its sister genera, Taurotragus and Strep- barkeri.
siceros, toward the end of the Late Miocene, and the —E Uganda (highlands), Kenya, S Sudan
species in the genus began to diverge in the early part (highlands), N Tanzania.
of the Late Pliocene. The bushbuck are depicted as —Forms a species-group with T. s. dama and T.
separating first, but Ropiquet did not state which taxon sylvaticus.
of bushbuck was used in the analysis. 7. Tragelaphus scriptus ornatus
—Angola, S DRC, Zambia, N Botswana, W
Tragelaphus scriptus Group Zimbabwe.
bushbuck —Pelage seemed rather more like T. s. scriptus;
Grubb (1985a) recognized the following taxa in E Af- in other respects shown to be close to T. s.
rica, which, at that time he classed as subspecies of sylvaticus.
T. scriptus:
Here, we recognize several species-level taxa
1. Tragelaphus scriptus bor within what has previously been called T. scriptus.
— Synonyms—T. cottoni, T. meridionalis, T. The differences in the scriptus-group taxa may be tab-
dodingae; perhaps T. pictus, T. signatus, T. ulated as follows:
punctatus, T. uellensis. In our craniometric analyses, we can fi nd no dif-
—Lowland savanna and swamps of Sudan and ferences among T. scriptus populations W of the Nile.
N Uganda, across the Albert Nile, but not S Nor are there any differences between T. dodingae and
across the Victoria Nile. T. bor (i.e., the scriptus-like populations E and W of
—Forms a species-group with T. s. scriptus. the Nile), but T. decula is quite different.
2. Tragelaphus scriptus decula In a comparison of montane populations from Su-
— Synonyms—T. multicolor, T. nigrinotatus, T. dan and Ethiopia, T. decula is at the edge of the varia-
fulvoochraceus. tion of T. meneliki; skulls from the distributional area
—Eritrea, Sudan near the Atbara, Ethiopia to assigned to T. powelli are within the range of varia-
about 10° N. tion of T. meneliki, consistent with the evidence from
3. Tragelaphus scriptus meneliki the skins (Grubb, 1985a) that indicates the two are
—Synonym—T. powelli. synonymous. The Sudanese montane population de-
—Higher altitudes of Ethiopia. scribed as T. barkeri is quite distinct, differing from
—Was considered close to T. s. decula. the others in its larger size (mainly in greatest length
4. Tragelaphus scriptus fasciatus and preorbital length; there are almost no differences
—Synonyms—T. olivaceus, T. reidae. in the horns, and very little difference in the bizygo-
—Riverine forest and coastal forest– savanna matic breadth).
mosaic, at low altitudes, from Tanzania and In populations from Sudan and Uganda, T. bor is
Kenya to Somalia and along the Webi separate from the others; T. barkeri and T. dama are
Shebeyli, and perhaps also the Juba River, the same, and differ from T. bor in the longer skull but
into Ethiopia. shorter preorbital length, and the slightly greater
5. Tragelaphus scriptus dama horn span. Of two skulls labeled “S Kordofan,” one is
— Synonyms—T. dianae, T. simplex, T. sassae. evidently an example of T. bor, while the other is
—Ituri, Uganda N to the Victoria Nile, closer to barkeri/dama. A skull from Jebel Manda (N
Rwanda, extreme N of Lake Tanganyika. Bahr-el-Ghazal) is like barkeri/dama; one from Nian-
—Intergrades with the T. s. scriptus group gara represents T. bor.
along a narrow and discontinuous belt from When T. barkeri and T. dama are compared with
Lake Edward to Lake Tanganyika, adjacent E African T. delamerei, they do not separate— hardly
to the W Rift Valley, but no intermediates even on average. But there is good separation be-
are known from the apparent contact area tween T. meneliki and the sample here called T. dela-
W of Lake Albert or in S Sudan. merei, whereas T. fasciatus does not separate as well
6. Tragelaphus scriptus delamerei from T. delamerei; these two E African taxa differ
— Synonyms—T. massaicus, T. haywoodi, T. from T. meneliki by having a longer skull and greater
meruensis, T. tjaderi, T. brunneus, T. eldomae, preorbital length, and a slightly narrower horn span.
—-1
—0
349-47558_ch01_1P.indd 127 5/31/11 7:05 PM

+1—
0—
-1—
Table 26 Differences in the Tragelaphus scriptus group
bor decula meneliki fasciatus dama delamerei ornatus
Neck hair Short Very short Long Very short Long but with a gray collar Long but with a gray Long but with a gray
349-47558_ch01_1P.indd 128
of short hairs collar of short hairs collar of short hairs
Dorsal stripe Well marked, dark Black- brown; often Young males only Brownish, poorly Dark brown in the males; Dark brown in the Dark brown in the
brown, more or less expanded to a blackish defi ned in the often narrow in the males; often narrow males; often narrow
extending to the neck saddle in the females females females in the females in the females
Median crest of Especially in the Often absent Often a few hairs Some hairs anteriorly Usually prominent Few Few
white hairs males posteriorly
Nuchal crest Usually not reversed Usually not reversed Never reversed Usually reversed Reversed Reversed Reversed
Adult males
Body color Light yellowish brown Dull sandy ochre Uniform glossy Grayish ochre or Light grizzled gray-ochre Dark brown in adults, Bright dark
dark (often brownish gray in the adults, yellowish reddish chestnut in red-brown
grayish) brown chestnut in the subadults the young; grizzled
Blackish Present Restricted to the saddle in None None Blacker Blacker Blacker
grizzling the young, dominant in
adults
Underside and Blackish Blackish Darker Darker brown Dark gray in the males, buff Dark blackish brown Darker
upper limbs in the females
Neck Paler, grayer Lighter, often with a dark As in the body As in the body As in the body As in the body As in the body
patch at the base
UNCORRECTED PROOF
Adult females

Body color Yellower; grizzling Light sandy ochre to light Uniform glossy Rich yellow-ochre Yellowish chestnut, not Chestnut, not Chestnut

less evident rufous dark brown obviously grizzled obviously grizzled
Blackish Absent Marked Absent, or reduced None None (more red-ochre None (more red-ochre None (more red-ochre
markings to 2 faint spots below) below) below)
Neck Relatively lighter Lighter As in the body As in the body As in the body As in the body As in the body
Upper lateral 15–30 cm (occasionally Long, or absent Absent Absent Absent Absent Absent
stripe absent or faint)
Lower lateral Usually continuous Absent, or a few spots Absent, or a few Line of scattered spots, Reduced to spots; rump Absent, or just a few Broken up into spots
stripe posteriorly very small rump but many spots on spots faint spots, but usually
spots the haunches some rump spots
Vertical stripes 2– 9, rarely faint, or a None None 3–7 Absent, or faint spots Absent, or faint spots Absent, or faint spots
line of spots
White streaks May be confluent Extensive Often very Yellowish Poorly marked, or absent Poorly marked, or Poorly marked, or
on the lower reduced absent absent
limbs
Horns Straight, smooth Smooth Smooth Widespread, rugose, Massive, widespread Massive, widespread Massive, widespread
well-marked annuli
Keels Virtually absent Present, but weak Present, but weak Pronounced, “folded” Strong Strong Strong
5/31/11 7:05 PM
b o v idae 129
A skull from Tunduru in SE Tanzania is T. fascia- Within the T. sylvaticus clade, the only really sig-
tus (or could perhaps be T. delamerei). One from Mi- nificant division is between T. ornatus (94% support)
kunduni (on the lower Tana) is on the edge of the and the rest (only 83% support). T. meneliki is divided
delamerei range of variation. into two apparently significantly different clades, but
Bringing in the S African taxa, T. ornatus is with only two specimens in each, and is associated
somewhat—but not absolutely— different from T. dela- nonsignificantly (67% support) with a cluster of south-
merei, with T. sylvaticus being intermediate, but they ernmost T. sylvaticus. Haplotypes of T. fasciatus do not
overlap fairly widely. form a monophyletic clade but, instead, recur in two
We tested the affi nities of T. ornatus, which has places within the major sylvaticus clade; we take this
clear white body markings, by bringing in skull mea- to mean that there has been a great deal of gene flow
surements of C African forest bushbuck (T. phalera- between this (phenotypically well-defi ned) coastal
tus) of the T. scriptus group. The T. ornatus sample is species and the more inland T. sylvaticus.
not intermediate, but it is clearly close to T. sylvaticus, Further analysis (Moodley et al., 2009) found that
differing from T. phaleratus by its larger size but not only were there two major clades, but that they
shorter preorbital length, narrower bizygomatic seem to occupy different positions among the Tragela-
breadth, and longer horns. phini. The T. sylvaticus clade was preferentially associ-
We fi nd that a specimen from Butele, 5.02° S, ated with the T. spekii / T. euryceros clade (although, as
23.48° E, represents T. ornatus. the authors admit, not with a very high level of sup-
We cannot split up the C African scriptus-like port), whereas the T. scriptus clade was either the sis-
bushbuck by craniometrics. Comparing the W-C Af- ter group of Nyala (Tragelaphus angasi in the “tradi-
rican (T. phaleratus), C African (T. johannae) and Uele tional” taxonomic arrangement, which the authors
Valley (T. punctatus) populations, there are different used) or else sister to all other Tragelaphini. Whatever
centers of gravity, but no defi nite divisions. the true affi nities of the two bushbuck clades, non-
Table 27 gives univariate statistics for populations monophyly (at least of mtDNA) was clearly favored
of bushbuck. over monophyly.
The study of the mitochondrial control region and
cytochrome b by Moodley & Bruford (2007) showed Tragelaphus scriptus Cluster
considerable genetic diversity among bushbuck, with harnessed bushbuck
rather more significant clades than is indicated by the
morphological data. Tragelaphus scriptus (Pallas, 1766)
The significant separation in their phylogenetic
1766 Antilope scripta Pallas. Senegal.
tree is between the T. scriptus and T. sylvaticus clus-
1844 Antilope leucophaea Forster; not Pallas, 1766 (= the
ters; interestingly, T. decula is well within the former,
blaubok).
and part of the T. bor subclade. The morphological
1882 Tragelaphus gratus Rochebrune; not Sclater, 1880 (a
affi nities of T. decula are so very clearly with the T.
sitatunga). Mouth of the Senegal River, between Cayor
sylvaticus group that a hybrid status seems the most
and Walo.
plausible solution, conjecturing that formerly the Ethi-
1898 Tragelaphus obscurus Trouessart. To replace Tragelaphus
opian plateau was occupied in suitable areas by forest-
gratus.
living bushbuck of the T. scriptus group. As the climate
changed, T. sylvaticus bushbuck moved northward, re- Specimens available to us are rich dark rufous with a
placing the indigenous phenotype by nuclear swamp- blackish suff usion; between 3 and 10 distinct, trans-
ing (the larger, more sturdily horned males dominat- verse white stripes; an upper and a lower white longi-
ing the smaller, weaker-horned indigenous males). tudinal flank band; a circle of white haunch-spots.
Within the T. scriptus clade, Moodley & Bruford Upper part of the limbs blackish, the inner side white;
(2007) found a major split between those from Sierra often a black line right down the front of the fore-
Leone and Guinea-Bissau (T. scriptus) and those from limbs. Underparts black. Pelage fairly long. Females
other localities; the support for the non-scriptus clade paler than the males, marked similarly.
is only 79%, while the two significant subclades within The mtDNA data (Moodley & Bruford, 2007)
it (Ghana to NW DRC [T. phaleratus] and N DRC / Chad/ unite specimens from far W Africa, specifically Sierra
Sudan / NW Uganda [T. bor]) have support values of Leone and Guinea-Bissau, to the exclusion of others.
100% and 90%, respectively. The support value for this clade is 99%. —-1
—0
349-47558_ch01_1P.indd 129 5/31/11 7:05 PM

Table 27 Univariate statistics for the Tragelaphus scriptus group: males
Gt l Bizyg Preorb l Horn l Span Horn br
W Africa scriptus
Mean 237.43 93.89 119.14 232.50 121.75 39.50
N 7 9 7 8 8 2
Std dev 6.901 2.369 5.460 20.529 15.341 —
Min 228 90 112 207 106 37
Max 247 97 127 263 155 42
Lower Volta scriptus
Mean 257.00 99.00 124.00 299.60 160.00 43.00
N 1 1 1 5 4 2
Std dev — — — 14.450 14.445 —
Min — — — 279 148 43
Max — — — 317 180 43
phaleratus
Mean 235.61 93.44 116.91 237.80 125.03 42.00
N 28 39 29 41 37 17
Std dev 7.529 3.331 5.011 25.040 23.410 3.623
Min 217 87 106 185 19 36
Max 245 100 130 290 184 48
bor
Mean 238.14 94.42 118.91 253.43 134.58 44.75
N 22 39 22 42 40 20
Std dev 7.298 2.701 4.492 26.821 18.832 5.911
Min 227 90 112 197 110 34
Max 253 101 129 320 188 56
dodingae
Mean 238.00 95.00 115.00 230.00 127.50 —
N 2 2 2 2 2 —
Min 237 93 113 218 107 —
Max 239 97 117 242 148 —
decula
Mean 223.43 91.44 105.07 250.86 143.67 —
N 7 8 7 7 6 —
Std dev 5.192 2.921 3.297 29.464 14.180 —
Min 219 86 101 212 123 —
Max 234 95 110 305 161 —
powelli
Mean 221.67 90.75 107.00 226.75 147.33 —
N 3 4 3 4 3 —
Std dev 4.041 5.679 1.000 21.469 15.144 —
Min 218 87 106 200 130 —
Max 226 99 108 249 158 —
meneliki
Mean 228.05 92.43 109.16 265.58 150.88 —
N 19 20 19 18 17 —
Std dev 7.051 3.836 4.717 25.308 15.029 —
Min 217 83 102 220 130 —
Max 238 98 119 320 191 —
barkeri
Mean 254.88 101.75 127.13 321.00 178.20 50.67
N 8 8 8 10 10 3
Std dev 4.016 3.151 2.949 43.586 28.177 4.509
Min 248 96 123 276 148 46
Max 262 105 132 404 248 55
-1—
0—
349-47558_ch01_1P.indd 130 5/31/11 7:05 PM

b o v idae 131
Gt l Bizyg Preorb l Horn l Span Horn br
dama
Mean 254.81 100.29 125.76 312.19 175.41 50.05
N 40 64 44 77 76 56
Std dev 9.119 3.809 5.874 39.079 20.854 4.518
Min 235 92 115 202 132 40
Max 274 108 137 420 217 59
delamerei
Mean 253.64 99.59 125.02 322.23 176.05 48.07
N 29 46 32 47 46 27
Std dev 8.612 5.010 5.600 36.756 27.138 4.682
Min 232 89 113 238 135 39
Max 274 110 136 405 263 60
fasciatus
Mean 253.44 98.85 127.00 344.85 202.90 58.00
N 9 10 9 10 10 2
Std dev 9.700 4.972 6.016 46.357 49.492 16.971
Min 236 93 115 261 150 46
Max 271 106 136 420 332 70
ornatus
Mean 243.27 94.98 118.61 300.79 168.36 47.03
N 33 45 33 48 45 32
Std dev 8.255 4.251 5.267 39.295 32.343 3.403
Min 229 84 107 205 109 39
Max 268 106 132 372 309 53
sylvaticus
Mean 243.44 94.32 122.00 295.06 164.59 45.75
N 18 25 10 32 17 8
Std dev 8.016 2.911 3.300 33.069 18.456 1.165
Min 230 89 117 232 130 43
Max 258 102 127 380 219 47
Tragelaphus phaleratus (Hamilton Smith, 1827) Tragelaphus bor Heuglin, 1877

1827 Antilope phalerata Hamilton Smith. W bank of Stanley 1877 Tragelaphus bor Heuglin. Req swamps (restricted by
Falls, Congo. Grubb, 1985a), near Bor, White Nile, S Sudan.
1905 Tragelaphus scriptus knutsoni Lönnberg. Upper Manns 1912 Tragelaphus cottoni Matschie. Farajala, W of Lado, on
Valley, SE Nigeria. the Koda River, White Nile.
1914 Tragelaphus scriptus pictus Schwarz. Duguia, lower Shari 1912 Tragelaphus cottoni dodingae Matschie. Kedef Valley, W
Valley. slope of the Didinga Hills, E of White Nile, Sudan, 4° 20'
1914 Tragelaphus scriptus punctatus Schwarz. Duma, N, 33° 30' E.
Oubangui Valley. 1912 Tragelaphus cottoni meridionalis Matschie. Three days’
1914 Tragelaphus scriptus signatus Schwarz. Tome Valley, near march N of Wadelai, Uganda.
the Gribinge- Oubangui watershed. 1914 Tragelaphus scriptus uellensis Schwarz. Angu, Uele
Valley.
Body color reddish brown, without any blackish suf-
fusion, except on the withers. Upper longitudinal band Color more ochery, less red, with no dark suff usion on
often absent, especially in the females. the withers; transverse stripes less distinct, especially in
Ghana, Togo, Nigeria, Cameroon, and Congo; the old animals; upper longitudinal band short or absent,
easternmost locality is Bolobo, in the far NW of the the lower one generally broken into spots and streaks;
DRC. several haunch spots; white mark on the throat.
The support value for this clade in the Moodley & From Bambio in the CAR, Koumoudjou in Chad, —-1
Bruford (2007) analysis is 100%. and the Uele district in NE DRC to the Victoria Nile —0
349-47558_ch01_1P.indd 131 5/31/11 7:05 PM

132 Artiodactyla
in Uganda and the S Sudan as far E as the Kedef Val- Males dark gray- brown above, gray on the sides; females
ley, well to the E of the Nile. more yellowish; young animals more rufous. About
Support for this clade is 90%. six white stripes, generally distinct; broken longitudi-
nal fl ank- band; many haunch- spots. No black on the
Tragelaphus sylvaticus Cluster crown and the nose; chevron below the eyes in the
keeled bushbuck females.
Coastal forests of E Africa, from the Webi Shebeyli S at least
Tragelaphus decula (Rüppell, 1835) to S Tanzania.
Moodley & Bruford (2007) found that the mtDNA of two
1835 Antilope decula Rüppell. “Abyssinia.”
specimens of this species (one from the Tana River, one
1902 Tragelaphus multicolor Neumann. Awash Valley, SE of
from Mona Mofa in Somalia) clustered with a T.
the Sekuala Range, Shoa.
sylvaticus specimen (ascribed to T. roualeyni) from
1902 Tragelaphus nigrinotatus Neumann. Barsa Valley,
Thabazimbi, South Africa, while two others from Mona
Maleland, N of Lake Stephanie.
Mofa clustered with a number of T. sylvaticus specimens
1912 Tragelaphus decula fulvo-ochraceus Matschie. Dungoler,
from localities in Tanzania, Malawi, and Zimbabwe.
near Lake Tana.
There has evidently been gene flow between T. fasciatus
Color ochery to yellow-brown, often with a black suf- and T. sylvaticus, but its history and direction are hard to
fusion on the back; white markings indistinct, fading work out.
completely in old animals, with the exception of some
haunch spots; two longitudinal bands often visible in Tragelaphus ornatus Pocock, 1900
young animals; white throat-patch present, sometimes
1900 Tragelaphus scriptus ornatus Pocock. Linyanti, Chobe
also a breast-patch. Front of the forelegs black, with
River between Lake Ngami and the Zambezi River.
white on the front of the knees.
Dry lowlands of Ethiopia. Males rich dark rufous, becoming black on the with-
Evidently a species of hybrid origin, because it pos- ers; 6– 8 white stripes; many haunch-spots; longitudi-
sesses the phenotype of the T. sylvaticus group but the nal bands represented only by a row of spots. Outer
mtDNA of the T. scriptus group. Moodley & Bruford’s sides of the legs black above the knees, the inner sides
(2007) specimens were from Lake Abaya and Ghimbi. white, with a broad black “garter” above the knee and
the hock; white stripe from the knees and the hocks
Tragelaphus meneliki Neumann, 1902 to the pasterns. Females light red-brown, with fewer
stripes.
1902 Tragelaphus meneliki Neumann. Gara Mulatta, Burka,
From the Chobe and Okavango rivers E along
and Jaffa ranges, the upper Webi Shebeyli watershed.
the Zambezi and its tributaries, at least as far as
1912 Tragelaphus powelli Matschie. Managasha Forest, W of
Malawi.
Shoa.
Moodley & Bruford (2007) found 94% support for
Dark brown to black in the adult males, with no white this species in general, as represented by their clade
except in the axillae, but with the occasional excep- containing specimens from Zambia, Zimbabwe, Bo-
tion of indistinct white flecks on the haunches and tswana, and Angola; in turn, the Angola and non-
above the hoofs; very occasionally, traces of white Angola clades also had support values of 97% and 98%,
lines down the forelegs. Females lighter, red, with a respectively; the affi nities of these two geographic
tendency for the white marks to be clearer. Coat long, groups need to be investigated morphologically.
full, shining. On both DNA and morphological grounds, this
Highlands of Ethiopia. species is a clear member of the T. sylvaticus cluster,
despite the presence of striping in the adults.
Tragelaphus fasciatus Pocock, 1900
Tragelaphus sylvaticus (Sparrman, 1780)
1900 Tragelaphus scriptus fasciatus Pocock. Sen Morettu,
Webi Valley, SE Ethiopia, 6° 20' N, 42° 50' E. 1780 Antilope sylvatica Sparrman. Groot Vatersbosch district,
1913 Tragelaphus scriptus olivaceus Heller. Maji-Ya- Chumbi, S Cape.
SE Kenya. 1850 Antelopus roualeynei Gordon- Cumming, nom. nud.
1949 Tragelaphus scriptus reidae Babault. Amboni, NE 1891 Tragelaphus scriptus roualeyni Thomas. Bakarikari, near
-1— Tanzania. the sources of the Limpopo River.
0—
349-47558_ch01_1P.indd 132 5/31/11 7:05 PM

b o v idae 133
1900 Tragelaphus delamerei Pocock. Saya Valley, 0° 55' N, 36° Didinga-Marangole mountains, Sudan. It is, in gen-
55' E. eral, impossible to tell a Kenyan from a Cape animal
1902 Tragelaphus massaicus Neumann. Upper Bubu Valley, of the same sex and similar age. Samples in Moodley
NW of Irangi, Tanzania. & Bruford’s (2007) phylogenetic tree come from as far
1902 Tragelaphus scriptus dama Neumann. Kavirondo Gulf. N as the Imatong Mountains, Lango, and Karamoja
1905 Tragelaphus haywoodi Thomas. Nyeri, near Mt. Kenya. (N Uganda) and Irumu, the Ishasha River and the
1908 Tragelaphus sylvaticus meruensis Lönnberg. Between Mt. Semliki district (DRC), down through Kenya and Tan-
Kilimanjaro and Mt. Meru, N Tanzania. zania to South Africa. As with the morphological char-
1909 Tragelaphus tjaederi J. A. Allen. Nakuru. acters, the haplotypes are all intermixed, and no real
1912 Tragelaphus dianae Matschie. Kabakaba, near Mahagi, geographic divisions can be discerned.
N end of Lake Albert, 2° 10' N, 30° 45' E. In a few areas, however, some distinctive morpho-
1912 Tragelaphus dianae simplex Matschie. Ituri Valley, near logical types predominate, and it may be that a few
Kifuku and Irumu. subspecies ought to be recognized. Mt. Elgon indi-
1912 Tragelaphus eldomae Matschie. Eldama Ravine, Mau viduals tend to be more of a dark red-brown in the
Forest, Kenya. males, with almost no traces, even in the females, of
1912 Tragelaphus haywoodi brunneus Matschie. W of white marks, except for some haunch-spots; some
Mt. Kenya. skins can be very similar to the lighter skins of T.
1912 Tragelaphus locorinae Matschie. Mt. Lcorina, 2950 ft., S meneliki. Possibly these should be separated under the
of the Didinga Hills, and N of the Marangole Mountains, name T. heterochrous Lönnberg, 1919.
Sudan. Not “near Narringepur,” as stated by Grubb Those from W Uganda and E DRC are clearly
(1985a); this is the locality of the juvenile female sylvaticus-like in size, horns, short collar, and so on,
specimen given as the type in the printed version of the but they tend to be lighter, with clear white markings
type description, but a note in Matschie’s handwriting and, generally, a white longitudinal band composed
on the Berlin Museum off print of the paper deletes the of a line of flecks. It is possible that they represent a
“Typus” against that specimen and writes it against the population with some gene flow from T. bor. This is
adult male specimen from Mt. Locorina. particularly the case for the type specimen of T. s. sas-
1912 Tragelaphus locorinae laticeps Matschie. NW foot of Mt. sae, from the Sassa (= Ishasha) River, which is ochery
Kadam, in the Debasien Range, 3950 ft., N of Mt. Elgon. brown with a hint of rufous, and faint but distinct
1912 Tragelaphus scriptus makalae Matschie. Makala, S of the white stripes.
Lindi Valley, about 27° 50' N, 0° 35' E. The Imatong Mountains population, described as
1912 Tragelaphus scriptus sassae Matschie. Ishasha River, T. barkeri, and the Didinga-Marangole Mountains
DRC– Uganda boundary S of Lake Edward. population, described as T. locorinae, are of special in-
1919 Tragelaphus scriptus heterochrous Lönnberg. W slope of terest. The Imatongs, or Al Istiwa’iyah, a small, iso-
Mt. Elgon. lated range at 4° 5' N, 32° 51' E, just N of the Uganda
1924 Tragelaphus barkeri Millais. Imatong Mountains, SE border, and the more extensive Didingas farther E
Sudan. (and their S continuation across the border, the Ma-
1961 Tragelaphus insularis or Tragelaphus scriptus insularis rangoles), both have bushbuck that were described as
Zukowsky. Islands in the Rufiji delta. separate species (T. barkeri, T. locorinae). Both of these
species were thought to be “doubtfully distinct from
Older males deep brown to blackish brown, with gray-
T. scriptus bor Heuglin” by G. M. Allen (1939), but they
ish sides, more chestnut above; younger males more
are, in fact, populations of T. sylvaticus in highland
red-brown. Forehead black-brown; nose black, with
forest, surrounded by a sea of harnessed bushbuck, T.
two white suborbital spots. White mark on the throat;
bor (Grubb, 1985a). This was the fi rst indication that
white spots (usually indistinct) on the face and the
what had been taken to be simple geographic varia-
haunches; little or no trace, in the adult males, of trans-
tion within a single species was, in fact, more compli-
verse or longitudinal stripes. Females dark yellow-
cated, involving interdigitation of sharply distinct
brown to more reddish, tending to be lighter on the
species of bushbuck, segregated by habitat.
shoulders and the upper part of the forelegs; often
A dwarf bushbuck was described as T. insularis by
with distinct traces of stripes.
Zukowsky (1961), from islands in the Rufiji delta. It
T. sylvaticus has an enormous range, from the
is impossible to say, from the type description, to
Cape N to the Kenya highlands, Uganda, and NE
which species this might be allied— presumably ei- —-1
DRC, with outlying populations in the Imatong and
ther T. sylvaticus or T. fasciatus. —0
349-47558_ch01_1P.indd 133 5/31/11 7:05 PM

134 Artiodactyla
Table 28 Univariate statistics for the Tragelaphus

Tragelaphus spekii Group spekii group: males
sitatunga
Much larger than bushbuck, with long, robust horns; Gt l Bizyg Preorb l Horn l
particularly characterized by extremely elongated
albonotatus
hoofs, adapted to their marshy habitat; only those of Mean 303.00 122.00 151.00 (366.00)
the dry Sesse Islands, in Lake Victoria, appear to have N 1 1 1 1
shorter, more “normal” hoofs. Angola
The results of our craniometric analyses are given Mean 327.00 133.00 175.00 665.00
below. N 1 1 1 1
Okavango sitatunga are large, but short-faced, Bangweulu
with relatively short horns; the skulls seem closer to Mean 298.33 119.67 159.33 582.33
N 3 3 3 3
those from Uganda than to those from Bangweulu or Std dev 5.859 6.351 3.055 51.598
other S African samples. Min 294 116 156 524
As far as population samples from Lake Victoria Max 305 127 162 622
and elsewhere in Uganda are concerned, skulls from Cameroon
Zinga Island, just offshore and SW of Entebbe, and Mean 304.00 120.38 160.20 509.44
Damba Island, only a little farther offshore, are not N 6 8 5 9
Std dev 10.450 2.326 7.014 50.028
essentially different from specimens from the main- Min 290 116 150 425
land, but one from Nkosi Island, in the Sesse group, Max 316 124 167 568
well offshore of the W side of the lake, is quite Congo
different. Mean 298.00 126.00 152.00 529.50
The Cameroon sample is also somewhat different N 1 1 1 2
from Okavango and Uganda specimens, and Sudan Min — — — 503
Max — — — 556
sample even more so—its large size and long face
Gabon
contrasting with its short horns and narrow skull Mean 323.00 120.00 166.00 554.00
(Cameroon has these features to a lesser extent). N 1 1 1 1
A skull from the Kouyou River, Congo, is more Gambia
like the ones from Bangweulu and Uganda than like Mean — — — 608.00
those from Cameroon (which is the closest sample N — — — 1
geograph ically). A skull from Elephant Bay, Angola, Okavango
is different from all the others, being especially broad Mean 297.50 126.67 150.50 540.33
N 2 3 2 3
for its size. The type of T. albonotatus, from “possibly Std dev — 7.371 — 20.108
either upper Guinea or Angola,” is clearly close to the Min 295 121 147 518
Cameroon sample; the Cameroon and T. albonotatus Max 300 135 154 557
skulls are large and long-faced, but they have con- Sesse Island
trastingly short horns. Mean 270.00 116.67 131.00 495.50
For their size, sitatunga from Cameroon, Sudan, N 2 3 2 2
Std dev — 3.786 — —
and Okavango have comparatively short horns; Ban- Min 265 114 128 481
gweulu, Tanzania, and Uganda sitatunga have rela- Max 275 121 134 510
tively long ones. Sudan
Univariate statistics are given in table 28. The type Mean 326.00 121.00 173.00 533.33
specimen of T. albonotatus (in the Zoologisches Mu- N 1 2 1 3
seum Alexander Humboldt in Berlin) is an extremely Std dev — — — 56.439
Min — 117 — 494
aged animal with strongly worn horns, their tips be- Max — 125 — 598
ing reduced to rounded stubs. Hence the horn length Tanzania
is not comparable to those of other specimens, and is Mean 285.00 116.00 141.00 531.00
enclosed in parentheses. N 1 1 1 1
Okavango (T. selousi) specimens are the same size Uganda
as those from Bangweulu (to which the name T. inor- Mean 286.20 120.17 151.70 553.16
natus presumably applies), but broader and shorter- N 10 18 10 19
-1— Std dev 13.248 3.930 7.258 40.237
0— faced, with shorter horns. Tanzania (T. ugallae) skulls Min 262 112 143 467
+1— Max 304 125 166 608
349-47558_ch01_1P.indd 134 5/31/11 7:05 PM

b o v idae 135
are smaller, compared with Bangweulu ones, and We presume that the name Tragelaphus wilhelmi
much shorter-faced, with shorter horns, thereby some- (Lönnberg & Gyldenstolpe, 1924)—type locality:
what resembling those from Okavango. Lake Bunyonyi, Kigezi, Uganda—is a synonym of T.
Most skulls from Uganda (T. spekii) are also smaller spekii. If so, then this is a species found in suitable ar-
than the Bangweulu ones, and they have similar skull eas almost throughout C and W Uganda.
proportions but slightly shorter horns. Those from
the Sesse Islands (T. sylvestris), though only slightly Tragelaphus sylvestris (Meinertzhagen, 1916)
smaller than other Uganda skulls, have a much shorter Skins seen: 3/2.
face and very short horns. Males with the neck paler, the top of the head red-
Sudan specimens (T. larkenii) are larger than dish, and preorbital marks sometimes faint; females
Uganda ones, the skull being relatively narrower and dull red-brown, with a dark dorsal stripe and no white
the horns shorter. marks. Young animals more reddish; no markings, or
In the Cameroon sample, the size is like the Ban- perhaps a flank band; neck paler; top of the head red-
gweulu specimens, with the same skull proportions dish; dark muzzle. Skull small; facial skeleton very
and very short horns. The Gabon skull (presumably short; horns short. According to the describer, T. syl-
representing T. gratus) does not fall into the range of vestris lacks the elongated hoofs of all other sitatunga.
the Cameroon specimens, but it is larger than any Nkosi Island; apparently a dryland taxon.
from Cameroon, is relatively narrower, and has lon- This is Meinertzhagen’s other contribution to
ger horns. The Congo skull is the same size as those mammalian taxonomy, apart from the giant forest
from Cameroon, but the former is relatively broader hog, and, despite his reputation (Garfield, 2006), we
and, again, has slightly longer horns. Finally, the can fi nd no fault with him in this instance.
skull from Elephant Bay, Angola, is larger than any
from Cameroon—in fact, it is the largest of the en- Tragelaphus larkenii (St. Leger, 1931)
tire set— and has a shorter face and extremely long Skins seen: 4/4.
horns. Males rich dark brown, 4– 8 stripes, usually faint,
Sitatunga habitat, in deep-water swampland, is and a lateral stripe that may be reduced to 3– 4 spots;
discreet and intermittent. It is no surprise that these females dark red-brown, 6–7 faint or broken stripes
animals have differentiated into quite a number of and a lateral stripe or spots; dorsal crest usually
distinct local forms. We propose the following revi- with white in both sexes; much gray in the necks of
sion, based on our analyses (above) plus our study of the females. Young animals rich red-brown, with
skins, but, as noted above, there are hints that the di- 6– 9 stripes and a lateral stripe or spots; dorsal crest
versity may actually be even greater. sometimes white or all dark; neck pale. Large in
size; long face; relatively short horns; relatively nar-
Tragelaphus spekii Speke, 1863 row skull.
Fifty mi. S of Yambio, Sudan. We have seen speci-
1863 Tragelaphus spekii Speke. Karagwe.
mens from the swamps of the White Nile and the
The correct spelling of the specific name, the date of the
Bahr-el-Ghazal, S Sudan.
description, and the person to whom it should be
attributed have all been much discussed; the correct
?Tragelaphus ugallae Matschie, 1913
author should be Speke, the date 1863, the spelling –ii
(Grubb, 2004a). 1913 ?Tragelaphus ugallae Matschie. Sindi, Ugalla, Tanzania.
Preserved skins seen: 13/4 (males/females). Skins seen: 0/2.

Sexes dissimilar in color. Males brown, females Dull yellow-brown, with only vague white mark-
dull reddish brown; males with prominent preorbital ings (in the skin from Tabora), or reddish (a nice red-
spots, occasionally with cheek spots; females with no yellow-gray) with seven clear but ill-bordered stripes
or only faint preorbital markings. Males occasionally and a flank stripe (in the skin from Kigoma); dark
with some white in the crest. Females sometimes dorsal stripe. Skull relatively small; short-faced; rela-
with rump spots or flank spots. Young animals a uni- tively short horns, as in T. selousi from the Okavango.
form yellow-brown or dull reddish brown, with faint We use this name provisionally for specimens
or no white marks; preorbital spots sometimes dis- from Tanzania that we have seen, although the two
tinct in the females; red on the head; neck paler. Skull available skins are more variable than seems likely in —-1
relatively small; horns long. a single taxon. —0
349-47558_ch01_1P.indd 135 5/31/11 7:05 PM

136 Artiodactyla
Clearly, more needs to be known about these sita-

tunga: whether they are all the same or, as seems Tragelaphus euryceros (Ogilby, 1837)
probable, they represent different taxa. bongo
2n = 33 (males), 34 (females), there being a Y-autosome
Tragelaphus gratus Sclater, 1880 translocation; the X chromosome is polymorphic, ei-
ther a large acrocentric (as in other Tragelaphini), or
1880 Tragelaphus gratus Sclater. Gabon.
an apparently uniquely derived submetacentric (Be-
Skins seen: 2/2. nirschke et al., 1982).
Large; pale pink-brown; white dorsal crest; white The bongo, despite its more robust build and lack
lateral stripes and rump spots. A female from Mbi of significant sexual dimorphism (including the pres-
Crater, 15 mi. NE of Bamenda, is dull gray-yellow- ence of horns in the females), is essentially a dryland
brown, with 4 very faint stripes, but the white dorsal sister species of the sitatunga. Their skulls are simi-
stripe is very clear. A male from Tiko, W Camer- lar, and the horns are very alike.
oon, is rich pale rufous, with spots high on the back In our analysis of the skulls of males from W and
rather than stripes and a clear black dorsal stripe. C Africa (see table 29), the three major geographic
The skull, if the Cameroon sample is typical, is some- groups—W-C African, Uele, and W African—separate
what like that of T. larkenii, but with very short horns. nicely. They have a progressively larger size, and
We are unsure if all of these represent the same horns with a somewhat greater span, in that se-
taxon. quence; but the horn length, the tip-to-tip distance,
The type of Tragelaphus albonotatus Neumann, and relative skull breadth are not quite consistent,;
1905—type locality: “possibly either upper Guinea or that is to say, the cline is not a geographic one. The
Angola”—is like T. gratus, but with a longer and wider picture for females is similar (see table 30): W Africa
frontal chevron. The skull shows no difference. are again somewhat broader in biorbital breadth than
Three males from Angola are pale to dark brown, W-C Africa, with a wide tip-to-tip distance; while
with paler hairs in the dorsal crest, and hints of 1–2 Uele are smaller, with a broader span, and a some-
stripes. They evidently represent a different, unde- what greater tip-to-tip distance than the other two
scribed taxon, but more evidence is needed before it groups.
can possibly be described. The univariate statistics for males (table 29) em-
phasize that these are possibly significant differences,
Tragelaphus selousi Rothschild, 1898 although the very small sample sizes in most cases
suggests that the results should be interpreted cau-
1898 Tragelaphus selousi Rothschild. Zambezi Valley; exact
tiously. Skulls of W-C African males are absolutely
locality uncertain (Grubb, 1999a).
shorter than W African ones, but not broader; the
Skins seen: Okavango region 3/7, Bangweulu horns, on average, are shorter, with a narrower span,
region 7/6. but the tip-to-tip distance is roughly the same. Uele is
Okovango skins are brown in both sexes; preor- intermediate in size, with small teeth, and long, di-
bital spots are prominent in the males, faint or obso- vergent horns (like W Africa), but the tip-to-tip dis-
lete in the females; one female has a faint lateral line tance is greater, on average, and the basal diameter is
and rump spots. Young animals have long fur and are larger (very slight overlap).
dull brown, with faint preorbital spots, a dark dorsal E Kenya is also intermediate in size, but nar-
stripe, a faint lateral line, rump spots, and hints of row, with short nasals, and fairly long horns (more
stripes. like W Africa), with a narrower span but divergent
Bangweulu skins are mostly the same, except that tips.
one is pale yellow-brown; but, as shown above, they The single specimen from W Kenya appears larger
differ in skull form (by being longer-faced and rela- than any other specimen, with long, robust, diver-
tively longer-horned) from the Okavango individuals, gent horns with the tips close together.
which may be taken (rightly or wrongly?) as represent- The females (table 30) show no great differences in
ing T. selousi. The type locality of T. inornatus Cabrera, size, implying different degrees of sexual dimor-
1918, is Lake Young, NE Zambia; this may be the name phism, but the two Kenya samples are narrow, with
that would be applicable to the Bangweulu sitatunga if (in W Kenya) short nasals. There are no great differ-
-1— further research finds it to be a distinct species. ences in horn span, and the sample size is too small to
0—
349-47558_ch01_1P.indd 136 5/31/11 7:05 PM

b o v idae 137
Table 29 Univariate statistics for bongo: males
Max Horn l Bas

Gt l Biorb Nas l teeth str Span Tip–tip diam
W Africa
Mean 417.40 156.45 157.86 122.00 637.08 416.00 264.09 84.00
N 5 11 7 2 12 11 11 13
Std dev 12.582 6.346 13.692 .000 47.354 71.208 118.875 10.654
Min 405 150 134 122 558 283 70 66
Max 437 168 172 122 728 525 482 100
W-C Africa
Mean 384.25 156.00 146.50 — 550.00 351.00 260.50 88.00
N 4 6 2 — 6 4 4 4
Std dev 15.586 6.782 16.263 — 32.398 15.979 56.122 5.598
Min 364 150 135 — 505 335 181 80
Max 402 164 158 — 590 370 312 93
Uele
Mean 406.00 159.50 142.00 113.00 653.33 405.50 339.17 99.00
N 4 4 1 2 6 6 6 2
Std dev 28.012 10.279 — 9.899 57.270 50.469 56.782 1.414
Min 365 146 142 106 600 340 260 98
Max 427 171 142 120 760 472 430 100
W Kenya
Mean — 170.00 — — 712.00 455.00 194.00 95.00
N — 1 — — 1 1 1 1
Std dev — — — — — — — —
Min — 170 — — 712 455 194 95
Max — 170 — — 712 455 194 95
E Kenya
Mean 394.00 147.67 145.00 117.50 616.00 385.00 316.00 85.33
N 1 3 2 2 1 3 3 3
Std dev — 5.132 7.071 .707 — 28.583 60.506 7.572
Min 394 142 140 117 616 352 272 80
Max 394 152 150 118 616 402 385 94
guarantee that differences in the tip-to-tip distance

are valid. The skulls in the W Kenya sample have Tragelaphus buxtoni Lydekker, 1910
very robust horns. The tips sometimes cross in the W mountain nyala
Kenya ones, and they always tend to be closer to- T. buxtoni is sister to the sitatunga/bongo clade. The
gether than in the males. horns (males only) are much more open-spiraled,
There is a strong impression of different amounts with a very marked, raised anterior keel, making
of sexual dimorphism, although the very small them look superficially like those of greater kudu, al-
sample sizes necessitate great caution in interpreta- though there are fewer turns and the spiral is even
tion. On the face of it, however, it appears that the more open in T. buxtoni. Body stripes are less well
size difference between the sexes is greatest in the marked than for any other member of the Tragelaph-
Uele region, followed by W Africa, and is least in ini (except for some bushbuck), the main body mark-
W-C Africa. This trend, if valid, seems corre- ings being the white facial chevrons, the white jaw,
lated with habitat: considerable sexual dimorphism the two white throat bands (one on the upper throat,
in more open forest, and hardly any in deep one near the brisket), the white on the inner surfaces
rainforest. of the forelegs, and an upwardly curved row of white
Therefore, we cannot make any secure conclu- spots on the haunch.
sions about taxonomic variation in this species (or Restricted to the forests of the Bale Mountains of
species-group). Ethiopia. —-1
—0
349-47558_ch01_1P.indd 137 5/31/11 7:05 PM

138 Artiodactyla
Table 30 Univariate statistics for bongo: females
Max Horn l Bas

Gt l Biorb Nas l teeth str Span Tip–tip diam
W Africa
Mean 395.75 152.83 153.67 — 580.13 256.13 93.43 60.50
N 4 6 3 — 8 8 7 8
Std dev 10.243 6.911 4.041 — 72.878 25.396 71.666 8.211
Min 383 141 149 — 514 208 0 48
Max 405 161 156 — 740 280 173 70
W-C Africa
Mean 386.67 150.50 151.00 — 535.50 240.25 89.00 60.75
N 3 6 1 — 4 4 4 4
Std dev 6.351 7.635 — — 55.296 7.411 35.449 6.292
Min 383 140 151 — 495 231 47 55
Max 394 162 151 — 617 248 131 68
Uele
Mean 381.60 151.40 — — 568.29 261.57 96.86 —
N 5 5 — — 7 7 7 —
Std dev 3.209 1.949 — — 33.470 32.382 65.226 —
Min 377 150 — — 518 206 23 —
Max 386 154 — — 621 300 207 —
W Kenya
Mean 392.67 144.33 134.67 108.00 564.00 233.33 88.00 67.67
N 3 3 3 1 3 3 3 3
Std dev 8.327 10.017 6.658 — 65.506 4.163 89.370 3.786
Min 386 133 127 108 489 230 11 65
Max 402 152 139 108 610 238 186 72
E Kenya
Mean 384.00 138.00 — 113.00 — 253.00 177.00 58.00
N 1 1 — 1 — 1 1 1
Std dev — — — — — — — —
Min 384 138 — 113 — 253 177 58
Max 384 138 — 113 — 253 177 58
maxillary suture oblique, the suture then angling

Ammelaphus Heller, 1912 forward, then turning and running backward. Audi-
lesser kudu tory bulla rounded; postglenoid foramen huge, an
2n = 38 in both sexes for one female and four male elongated oval. Basioccipital constricted between the
lesser kudu, of unrecorded origin, that were karyo- mesopterygoid plates; mesopterygoid fossa forming a
typed by Benirschke et al. (1980). Unlike other narrow U; vomer forming a sharp ridge; basioccipital
Tragelaphini, both X and Y chromosomes were fused tuberosities high and angular. Free ends of the paroc-
with autosomes. cipital processes long, straight. Buccal styles on the
Benirschke et al. (1980) proposed that lesser kudu upper cheekteeth not as prominent as in other tragela-
separated early from other taxa, possibly along the lin- phines. Horn spiral comparatively open; anterior keel
eage leading to Boselaphus. DNA data (see, for example well marked, somewhat raised from the plane of the
Ropiquet, 2006) confirm that Ammelaphus was the ear- rest of the horn.
liest lineage within the Tragelaphini to separate, al- The two sexes very alike in external features, with
though not that it was part of the Boselaphus lineage. the only noticeable difference being in the overall
Anterior median and lateral nasal tips more or less tone, redder in the females and grayer in the males.
equal in length, with a wide cleft between them. Sev- (The skins described below are males, unless other-
eral supraorbital foramina, in a deep channel, but wise indicated.)
-1— only the anterior one large. Lacrimal vacuity short, Two sets of horns of lesser kudu are known from
0— not reaching the maxilla. Superior part of the malar- the Arabian Peninsula: one from Yemen, and one
349-47558_ch01_1P.indd 138 5/31/11 7:05 PM

b o v idae 139
from Saudi Arabia. It is unknown whether this ani- inner side of the upper arm; striking white
mal is, or was, indigenous, or whether the horns were chevron; 1–2 white spots on each side of the
ultimately brought from Africa (D. Harrison & Bates, jaw angle; ears dark, with white rims; one
1991). upper and one lower white spot, usually fairly
small, on the throat; 7–13, but usually 10–12,
Ammelaphus imberbis (Blyth, 1869) contrasting white stripes on each side, and no
oblique stripes on the haunch. Black, round
1869 Ammelaphus imberbis (Blyth). Somalia.
posterior false hoofs; no (or very little) white
Skins seen: 10. on the pasterns. Brown mane shortening to a
Males paler, grayer than the more S specimens, line, then becoming a mane again on the
with much fewer contrasting white markings. Whole withers and the anterior back, with white hairs
tail with a dark dorsal surface, whereas S specimens in it; mane continuing as a thin white dorsal
with the tail dark only on the tip and at the base (and stripe, and becoming lengthened a bit, with
along the sides). Upper throat spot fairly small; lower more brown on the rump. Tail black at the
one widened to a chevron (usually not widened in S base and on the tuft, white in between and on
skins). Stripes beginning farther back on the shoul- the sides. The localities for the skins are
ders. Body stripes any quantity between 8 and 14, Lotholier, Turkana, Kedef Valley, Karamoja,
usually more than 10, the last 2 in each case being Engaruka, an Omo tributary, Lake Abaya, and
short, somewhat converging below, and going some- Lake Zwai.
what obliquely onto the haunch. Striping very char- 3. Galana (1)—Dark gray–chocolate-brown; 14
acteristic of A. imberbis: the stripes more spaced than stripes; big white spots on the front feet; large
in A. australis, and rather less marked. Hind pasterns white area on the back feet.
generally more fully black, but the anterior ones 4. Iringa (1)— More reddish brown than the
white. previous skins; big white throat spots; 12
N Somalia. We have seen specimens from Orthar, stripes; big white spots on the feet. The
Gurgura, Ginoble, Ulfula, Dire Dawa, and Berbera. locality for the skin is Mtandika, Iringa.
The type specimen of A. imberbis resembles other, 5. Jubaland—There are, curiously, no skins
better-preserved skins of this species, although the available from Jubaland.
skin of the type is much too yellowed to detect its
The skulls fall into three completely separate clus-
original color.
ters: N Somalia / Ogaden, Jubaland, and Turkana/
Abaya. These are 100% distinct. The N Somalia /
Ammelaphus australis Heller, 1913
Ogaden sample represents A. imberbis; the Jubaland
1913 Ammelaphus australis Heller. Longaya, Marsabit. sample is probably A. australis (although, having no
available skins from Jubaland, we cannot be certain
A. australis was said to differ from A. imberbis by being
of this); the third sample would therefore seem to rep-
darker (bright ochery tawny), with no white spot on
resent an undescribed taxon (and, as implied above,
the front of the pasterns; horns shorter.
the skins we have seen from this region are different
In our study, the skins differed slightly according
from the descriptions of the type and paratype of A.
to different localities, and there may be more taxo-
australis). Table 31 gives the univariate statistics. The
nomic variation than hitherto recognized, as follows:
number of female skulls is too small to break the sam-
1. Longaya Water (2 skins, both females; the type ple down by locality; the figures for these skulls are
and paratype of A. australis)—Bright listed merely to give an idea of the amount of sexual
ochraceous-tawny; 12 stripes; no spots on the dimorphism.
front pasterns, but a white spot on the hind The described differences in horn length (A. imber-
pasterns. Tail tawny above; tip seal-brown. bis was stated to be longer than A. australis) do not
Lower throat-spot larger than the upper one. hold; the horns, on average, are considerably longer,
This description is taken from Heller (1913b), and more spreading, in the southernmost sample
because we have seen no specimens from this (Lake Turkana, Lake Abaya)!
region ourselves. The number of horn turns varies by locality:
2. Turkana area (10)—Gray-white on the neck;
N Somalia—5 have 2 turns, 3 have 2.5 turns —-1
body chocolate-brown; black band on the
Jubaland—6 have 2.5 turns, 1 has 3 turns —0
349-47558_ch01_1P.indd 139 5/31/11 7:05 PM

140 Artiodactyla
Table 31 Univariate statistics for lesser kudu
Horn l
Gt l Cb l Gt br Preorb str Tip–tip Span
Males
N Somalia, Ogaden
Mean 319.80 314.40 122.71 176.00 526.62 261.29 290.50
N 5 5 7 6 8 7 6
Std dev 7.014 9.099 3.684 5.254 52.236 62.147 79.490
Min 310 303 118 168 433 187 187
Max 328 327 129 184 588 360 403
Jubaland
Mean 308.67 299.33 119.86 170.17 532.71 257.40 274.80
N 6 6 7 6 7 5 5
Std dev 7.633 6.683 4.525 4.579 42.976 65.653 29.811
Min 296 292 114 166 469 168 238
Max 318 310 128 178 577 315 305
Lake Turkana region
Mean 319.37 310.86 119.45 174.25 559.60 303.50 328.30
N 8 7 11 8 10 8 10
Std dev 7.763 5.640 6.532 5.800 45.957 77.371 52.603
Min 309 303 104 166 508 222 263
Max 332 319 127 181 655 415 415
Females
Total for all regions
Mean 290.67 277.33 110.67 157.33 — — —
N 3 3 3 3 — — —
Std dev .577 3.786 4.041 .577 — — —
Min 290 273 107 157 — — —
Max 291 280 115 158 — — —
Lake Turkana— 4 have 2 turns, 1 has 2.25 turns, 4 their range. In the neighbor-joining haplotype tree,
have 2.5 turns, 2 have 3 turns the single specimen from Samburu was sister to all
of the others (the range of these “others” extending as
Strepsiceros Hamilton Smith, 1827 far NE as Maasailand in N Tanzania); the bootstrap
greater kudu value for the non-Samburu clade was 100%. In this
Anterior lateral nasal tips much longer than the cen- other clade, most of the SW individuals clustered
tral one, with a deep cleft in between. Supraorbital separately from the rest, but the bootstrap values
foramina large, sometimes several, or sometimes all were not high. Although the specimens in the Tanza-
fused into one, in an elongated depression. Lacrimal nian sample clustered together, it was a subclade
vacuity short, not reaching the maxilla. Superior part nested well within the SW– S African sample.
of the malar-maxillary suture oblique backward. There is clearly need for a wider geographic study
then the suture arching forward. Auditory bulla of the greater kudu; for the moment, it is notable that,
deep, rounded in lateral view, the inferior surface an- as in our morphological sample, there appears to be a
gular; postglenoid foramen small and round. Basioc- deep split between those from NE Africa and those
cipital of even width, the tuberosities low, rounded; from farther S.
mesopterygoid fossa forming a wide V; vomer sharply
ridged. Free ends of the paroccipital processes long, Strepsiceros strepsiceros (Pallas, 1766)
curved forward. Buccal styles on the upper cheek-
1766 Strepsiceros strepsiceros (Pallas). SE Cape (Grubb, 1999a).
teeth very prominent.
Nersting and Arctander (2001) studied the mito- Grubb (1999a) noted that there has historically been
-1— chondrial control region of kudu from Namibia NE discontinuity within South Africa for Strepsiceros;
0— to Samburu (N Kenya); that is to say, from much of greater kudu occur in the S Cape, and not again until
349-47558_ch01_1P.indd 140 5/31/11 7:05 PM

b o v idae 141
the N Cape / Transvaal/Botswana/Namibia popula-

tion. Shortridge (1934) found that a skin from Keis- Strepsiceros chora (Cretzschmar, 1826)
kama, E Cape, was considerably darker than those
1826 Strepsiceros chora (Cretzschmar). “Eastern Sudan.”
from Namibia; so did A. Roberts (1951:308), who
1913 Strepsiceros bea Heller. E of Lake Baringo, Kenya.
wrote: “Flat skins which I have seen in shops in the
Great Karas Mountains area of Great Namaqualand Males (20 skins seen): Pale, light gray-brown; big facial
appeared to me to be darker than those from the East.” chevron; long, pale nuchal mane, with dark brown
We ourselves have seen no skins from the S Cape, tips, on both the throat and the nape (in Andal and
apart from the skin of a young female from “Steythers- Eireirib specimens), but only a very short, dark, ex-
ville, Cape Colony,” that is dark red-brown, with 12 cuse for a throat mane, restricted to the lower throat,
stripes, a good mane, and very little black on the feet. in the others (and this is pale in the Blue Nile speci-
men). Only 3–7 stripes. Feet with black or brown pas-
Strepsiceros zambesiensis Lorenz, 1894 terns, and a white band above the hoofs; this region
always whitish above the dark zone. Ear-backs wholly
1894 Strepsiceros zambesiensis Lorenz. Lesuma Forest, on the
black or dark gray (except the rims), or becoming black
Zimbabwe– Botswana border.
distally (in Andal specimens).
1914 Strepsiceros frommi Matschie. 7° 20' S in Tanzania.
Localities of male specimens we have seen: Andal
Males (17 skins seen): grayer to light ochery to red- (Sudan); Eireirib; “Somalia”; Gadabusi (Somalia); Sen-
dish gray-brown, some white hairs mixed in. Ear-tips nar; Blue Nile; Bogos; Arroweina (S Ethiopia); Laiki-
occasionally white; ears otherwise a darker shade of pia (Kenya).
the body color. Very clear white chevron on the face, Females (3 skins seen): Dull to light brown; full
often rather short. Long throat mane, straw-colored facial chevron. In one skin, 5 clear stripes, plus a sixth
or mixed pale red-brown and black, often the whole on the haunch; in another skin, 3 stripes on the
hair tip blackish. Short red-brown nuchal mane, be- haunches, 3 others well spaced farther forward, a
coming a clear white dorsal stripe (with slightly length- vague stripe between these two groups. Dark ear-
ened hairs) at the commencement of the first stripe, backs; dark ring all around the hoofs; no mane, just
but with black hairs sometimes mixed in. Stripes 8–11, very slightly lengthened hair on the nape and the
variable, but tending to be contrasty. Pasterns brown, throat.
usually not black, on the backs. Lower legs more cin- Localities of female specimens we have seen:
namon than the color of the body and the upper Chuowkan (Abyssinia); Andal; “Kenya.”
segments.
Localities from which we have seen male speci- Strepsiceros cottoni Dollman & Burlace, 1928
mens: Hillingdale, Kangongo, Koreis, Bambi, Etosha Males (3 skins seen): Pale brown; facial chevron poorly
(Namibia), Angola, Olifants River (Mpumalanga), expressed. Stripes 4– 8, the dorsal stripe white or light
Mahura, Katumbas (E Zambia), Lengwe (Malawi), brown, the hair somewhat lengthened, with more
“Malawi.” black hair toward the croup. Ear-backs dark gray.
Females, young males (17 skins seen): ochery to Hoofs black, round, very small; this region is always
bright grayish brown; longer, redder hair than the whitish above the dark zone. Mane pale mane, very
adult males. Chevron on the face more poorly devel- short on the nape, and long on the throat.
oped than in the adult males. Very short, straw- Females (1 skin seen, the type): Like S. chora. Light
colored throat mane. No white tips to the ears. Stripes brown; full facial chevron. Stripes 3 on the haunches,
5–10. Nuchal mane longer, reddish, with more black 3 others well spaced farther forward, with a space
hairs mixed in on the withers than in the adult males; between the two groups. Dark ear-backs; dark ring
dorsal stripe vaguely marked in parts, white, except all around the hoofs; no mane, just very slightly
where sporadically interrupted with black in the mid- lengthened hair on the nape and the throat.
dle of the back (after the fourth stripe). Pasterns usu- Aya.
ally black (sometimes not much—they may be nar-
rowly brown). The two N species (S. chora and S. cottoni) are very
Localities of female and young male specimens we alike in many respects, but, on the limited evidence,
have seen are Nubis, Hillingdale, Struispan, Omataka the skins from Chad fall outside the range of varia-
(Namibia), “Namibia,” Okavango, Angola, Banga tion of the more numerous specimens in the sample —-1
(Zimbabwe), Kwa Kzembe (DRC), Kruger, Rupias. from the Horn of Africa. —0
349-47558_ch01_1P.indd 141 5/31/11 7:05 PM

142 Artiodactyla
Table 32 Regional differences in the number of turns small differences in four autosomes ( Jorge & Be-
in the horns of greater kudu nirschke, 1976).
<2.5 2.5 >2.5 Taurotragus oryx oryx (Pallas, 1767)

No stripes in the adults.
S Africa 5 13 5
Kenya, Somalia, Chad 12 7 —
Taurotragus oryx livingstonii Sclater, 1864
About 8 white stripes. As far as we can tell, mainly
from photographs taken in the wild, eland N of
Craniometric analysis is restricted, because of the Zambezi tend to retain their stripes until old
comparatively few complete skulls; most available age.
specimens are trophies, and consist only of frontlets
and horns. We are utterly unable to distinguish any geographic
Skulls from Namibia are larger all around, com- samples of T. oryx by multivariate analysis.
pared with those from Malawi and Somalia. Somali Univariate statistics are presented in table 34. For
skulls are generally small, especially their horn the males, E African eland average smaller than other
length, but the skull breadth is not small. samples, with shorter, more slender, less divergent
When the small Chad sample (N = 4) is com- horns. The W Zambian sample has large teeth, and is
pared with the Somali sample (N = 8), they are also fairly small. South Africa, Botswana, and Angola
completely separated, most strikingly in the horn specimens are large, with large teeth and slender
mea surements. horns; the single pair of South African horns are ex-
Greater kudu show regional differences in the tremely short and not very divergent. Except perhaps
number of turns in the horns. for the E and the South African samples, these differ-
Table 33 gives the univariate statistics for both the ences are not consistent enough for subspecific
males and the females. The Kenya specimens, which recognition.
all come from N Kenya, clearly belong with the So- For the females, the skulls in the E Africa sample
mali sample. The three S samples probably belong are not smaller than those in other samples, but the
together. Thus, three species are recognizable in the shorter, more slender, less divergent horns are similar
main series, and the poorly known S Cape taxon to the horns of the males. The South African (Cape)
surely represents a fourth. specimen is large, and the horns (N = 1!) are again
very short and not very divergent. The Namibia skull
Taurotragus Wagner, 1855 (Grootfontein) has large teeth and slender horns (as
eland does the type specimen for Antilope triangularis Gün-
The eland skull, in many respects, is very like that of ther, 1889, of unknown origin). Except perhaps for
Strepsiceros: anterior lateral nasal tips much longer the E African and the South African samples, these
than the median ones, with a deep cleft in between; differences, again, are not consistent enough for sub-
lacrimal vacuity short; malar-maxillary suture sinu- specific recognition.
ous, forming an anterior-pointing W; auditory bulla
large, its inferior surface slightly angular, with a Taurotragus derbianus (Gray, 1847)
small postglenoid foramen anterior to it; basioccipital giant eland
of even width, with the basioccipital ridges relatively There are no morphometric differences among geo-
low; buccal styles very prominent. Differences: su- graphic samples (table 35), and our inspection of pho-
praorbital foramina in Taurotragus 1–2, very large, in tos shows no consistent external differences, either.
a very deep, wide, and elongated depression; mesop- Thus all giant eland, from Gambia to the Nile, would
terygoid fossa forming a wide U; the free ends of the be taxonomically the same.
paroccipital processes short. Comparing T. derbianus and T. oryx, and combin-
ing all samples into one for each species (table 36), T.
Taurotragus oryx (Pallas, 1767) derbianus averages slightly the smaller of the two spe-
common eland cies in the males (but is very slightly larger in the fe-
2n = 31 (males), with the Y/14 translocation that is males), while the horns are very much larger—
-1— usual in the Tragelaphini. The chromosomes are al- thicker at the base, longer (absolutely so in the case of
0— most identical to those of the greater kudu, with the males), and more divergent.
349-47558_ch01_1P.indd 142 5/31/11 7:05 PM

Table 33 Univariate statistics for greater kudu
Horn l
Gt l Cb l Gt br Preorb str Tip–tip
Males
Cape
Mean 421.67 408.33 166.75 258.00 895.00 708.67
N 3 3 4 2 3 3
Std dev 20.232 13.051 2.500 — 110.340 98.592
Min 409 398 164 221 780 595
Max 445 423 170 295 1000 771
Zambia, Namibia, Zimbabwe
Mean 415.10 398.11 171.14 226.67 951.33 747.77
N 10 9 14 9 15 13
Std dev 11.308 10.374 4.737 8.337 67.969 182.179
Min 394 379 164 210 850 423
Max 428 413 181 241 1110 1034
Malawi, S Tanzania
Mean 415.00 402.00 166.00 227.50 1067.50 803.75
N 4 3 5 4 4 4
Std dev 17.720 15.620 3.317 9.434 47.170 84.200
Min 396 384 161 214 1000 685
Max 435 412 170 234 1100 880
N Kenya
Mean 397.00 — 168.00 212.00 903.00 594.00
N 1 — 2 1 2 1
Min — — 166 — 883 —
Max — — 170 — 923 —
Somali, Eritrea, NE Sudan
Mean 393.44 380.86 165.25 216.44 831.73 630.36
N 9 7 12 9 11 11
Std dev 14.917 15.432 3.769 8.278 114.990 119.576
Min 373 363 160 206 690 426
Max 419 403 170 230 1110 810
Chad, W Sudan
Mean 375.17 358.25 159.86 205.67 780.50 614.83
N 6 4 7 6 6 6
Std dev 12.352 8.221 5.146 11.843 84.033 139.481
Min 358 351 152 188 636 510
Max 393 370 167 221 850 877
Females
Cape
Mean 391.00 372.00 155.50 — — —
N 2 1 2 — — —
Min 387 — 151 — — —
Max 395 — 160 — — —
Zambia, Namibia, Zimbabwe
Mean 375.89 362.33 145.70 210.22 — —
N 9 9 10 9 — —
Std dev 12.624 11.269 6.038 9.550 — —
Min 354 345 134 196 — —
Max 391 378 156 223 — —
Mean 375.89 362.33 145.70 210.22 — —
Malawi, S Tanzania
Mean 378.00 357.00 146.00 209.00 — —
N 1 1 1 1 — —
(continued)
—-1
—0
349-47558_ch01_1P.indd 143 5/31/11 7:05 PM

144 Artiodactyla
Horn l
Gt l Cb l Gt br Preorb str Tip–tip
N Kenya, Somali, Eritrea, NE Sudan

Mean 367.00 551.67 146.33 202.00 — —
N 3 3 3 3 — —
Std dev 11.533 353.670 4.041 1.732 — —
Min 358 342 142 200 — —
Max 380 960 150 203 — —
Chad, W Sudan
Mean 362.00 350.00 146.00 200.50 — —
N 2 2 2 2 — —
Min 354 346 145 197 — —
Max 370 354 147 204 — —
The name “giant” eland is a bit of a misnomer; the back of palate, and the lack of differentiation of
presumably it is the enormous horns that have led to the parastylid from the paraconid on the two poste-
it being awarded this name. rior lower premolars.
Subfamily Antilopinae Gray, 1821 Neotragus Hamilton Smith, 1827
In his morphologically based cladogram, A. W. Gen- The distribution and systematics of the dwarf ante-
try (1992) found the Cephalophini to be sister to all lope (Neotragus species) are not well known. Although
other groups, followed by Neotragus (as represented we here continue to unite them all into a single ge-
by N. moschatus). There was then a split between the nus, as is conventional, whether they are all truly
Antilopini and others; he was thus probably the closely related to one another has never been tested,
earliest to recognize the affi nity of antelope (such just merely assumed.
as the Alcelaphini and the Hippotragini) with the The dwarf antelope are among the smallest ungu-
Caprini. lates. They have relatively prognathous skulls, with
On the basis of partial mitochondrial sequences elongated premaxillae; very brachyodont dentition,
(12S and 16S rRNA), and with an unavoidably limited with relatively small premolars; and the horns are not
Chinese-centered dataset, Lei, Jiang, et al.’s (2003) erect, but directed back in the plane of the face. The
neighbor-joining tree of this subfamily was as genotypic species, N. pygmaeus, differs from all other
follows: bovids in the frequency with which it retains milk
canines.
(Neotragus (Oreotragus (Ourebia, Procapra, Saiga,
Gazella/Eudorcas, Antidorcas/Madoqua, Antilope,
Neotragus pygmaeus (Linnaeus, 1758)
Raphicerus)) ((Damaliscus, Oryx/Hippotragus)
royal antelope
(Ovis, Pantholops, Pseudois/Capra/Hemitragus))).
1758 Capra pygmaea Linnaeus. “Guinea, India” (W Africa).
Kuznetsova & Kholodova (2003), however, also
1777 Antilope regia Erxleben. Senegal.
using 16S as well as a nuclear fragment (b-spectrin),
1827 Antilope opinigera Lesson. Coasts of Guinea and Loango.
found that the data supported, at most, a weak asso-
1851 Nanotragus perpusillus Gray. No locality.
ciation of the Neotragini with the Antilopini, so these
authors instead placed the Neotragini in a clade with Size small, shoulder height to 25 cm (Lydekker 1913;
the Aepycerotini, which is also the conclusion of Rop- Ansell 1971); weight 1.8–2.3 kg (five males), 3.1 kg
iquet (2006) and others. (two adult females).
Lateral hoofs said to be present, but rudimentary
Tribe Neotragini Sclater & Thomas, 1894 (Urbain & Friant 1942; Dragesco et al, 1979); we
A. W. Gentry (1992) noted many apparent primitive found no trace in the specimens we examined. Horns
-1— retentions in this tribe, such as the low infraorbital smooth and short (to about 38 mm, according to An-
0— foramen, the posterior level of the median indent at sell, 1971).
349-47558_ch01_1P.indd 144 5/31/11 7:05 PM

Table 34 Univariate statistics for common eland (Taurotragus oryx)
Max Bas
Gt l Cb l Biorb teeth diam Horn l Tip–tip
Males
S Africa
Mean 511.00 484.00 213.00 152.00 79.00 433.00 303.00
N 1 1 1 1 1 1 1
Bakwena
Mean 508.00 490.00 198.00 148.00 77.00 676.00 436.00
N 1 1 1 1 1 1 1
Angola
Mean 490.00 471.00 198.00 148.00 78.00 580.00 402.00
N 1 1 1 1 1 1 1
S Mozambique
Mean 488.29 467.60 203.14 139.00 87.00 589.00 316.86
N 7 5 7 7 7 7 7
Std dev 13.937 9.017 2.854 4.435 4.690 37.216 14.622
Min 463 457 200 130 80 519 300
Max 506 479 207 144 94 638 337
Zimbabwe
Mean 496.00 459.00 204.50 141.00 91.50 654.00 312.50
N 2 1 2 2 2 2 2
Min 491 — 203 132 89 650 280
Max 501 — 206 150 94 658 345
W Zambia
Mean 480.33 468.00 200.29 148.14 85.83 645.83 385.33
N 6 3 7 7 6 6 6
Std dev 14.597 15.588 17.689 9.805 6.824 45.429 94.299
Min 459 450 182 135 73 595 257
Max 500 477 236 161 93 700 515
Malawi, S Tanzania
Mean 491.14 467.40 201.78 139.43 83.56 664.33 353.22
N 7 5 9 7 9 9 9
Std dev 11.611 8.849 6.741 7.458 2.920 43.629 85.130
Min 474 459 191 126 80 603 176
Max 508 477 210 151 88 740 480
E Africa
Mean 477.38 457.43 201.11 138.22 80.89 565.78 272.33
N 8 7 9 9 9 9 9
Std dev 20.311 16.481 8.638 7.412 7.356 53.399 43.709
Min 450 439 190 126 69 512 181
Max 515 485 214 146 88 661 322
Females
Cape
Mean 480.00 456.00 192.00 144.00 74.00 542.00 341.00
N 1 1 1 1 1 1 1
Type of triangularis
Mean — — 155.00 — 56.00 795.00 563.00
N — — 1 — 1 1 1
Grootfontein
Mean 453.00 — 184.00 150.00 58.00 648.00 233.00
N 1 — 1 1 1 1 1
S Mozambique
Mean 445.20 442.75 183.80 132.00 64.40 628.20 349.00
N 5 4 5 5 5 5 5
Std dev 34.981 7.136 6.760 9.592 7.503 93.919 149.355
Min 384 434 177 123 54 533 212 —-1
Max 470 451 193 145 73 773 570 —0
(continued) —+1
349-47558_ch01_1P.indd 145 5/31/11 7:05 PM

Max Bas
Gt l Cb l Biorb teeth diam Horn l Tip–tip
Zimbabwe
Mean 461.00 440.00 189.00 138.00 63.00 644.00 361.00
N 1 1 1 1 1 1 1
W Zambia
Mean 456.67 442.00 185.29 135.83 61.33 642.00 385.33
N 6 5 7 6 6 6 6
Std dev 6.713 16.016 10.452 4.262 5.007 41.323 97.867
Min 444 427 165 130 56 598 230
Max 463 469 196 143 68 704 506
Malawi, S Tanzania
Mean 439.00 418.50 176.80 133.75 57.20 673.00 278.00
N 5 2 5 4 5 5 5
Std dev 13.096 — 6.458 4.272 4.324 114.645 132.293
Min 427 417 170 128 51 535 175
Max 460 420 185 137 63 823 500
E Africa
Mean 441.90 427.00 179.80 131.00 60.78 601.33 305.50
N 10 6 10 9 9 9 8
Std dev 22.218 19.411 12.674 8.703 7.839 47.326 46.448
Min 402 404 153 114 54 524 246
Max 473 455 201 140 79 646 376
Table 35 Univariate statistics for Taurotragus derbianus
Gt l Cb l Biorb Max teeth Bas diam Horn l Tip–tip
Males
Sudan
Mean 471.00 455.50 195.75 140.67 100.67 866.00 691.25
N 3 2 4 3 3 4 4
Std dev 15.100 13.435 .957 2.309 10.263 54.437 91.587
Min 455 446 195 138 92 814 623
Max 485 465 197 142 112 922 822
Central Africa
Mean 477.75 464.75 200.00 140.00 94.71 878.71 665.14
N 4 4 5 4 7 7 7
Std dev 18.337 14.245 7.810 5.477 4.751 47.419 188.552
Min 460 452 187 134 89 800 483
Max 502 479 207 147 100 933 985
West Africa
Mean — — — — 91.00 810.00 737.00
N — — — — 2 2 2
Min — — — — 91 800 724
Max — — — — 91 820 750
Females
Central Africa
Mean 450.71 439.57 182.88 138.00 69.63 718.63 431.75
N 7 7 8 8 8 8 8
Std dev 16.101 11.013 5.489 4.781 7.425 81.367 131.947
Min 418 419 171 133 55 579 211
-1— Max 471 455 189 145 77 796 575
0—
349-47558_ch01_1P.indd 146 5/31/11 7:05 PM

b o v idae 147
Table 36 Univariate statistics for the two species of Taurotragus
Gt l Cb l Biorb Max teeth Bas diam Horn l Tip–tip
Males
oryx
Mean 486.61 466.04 202.26 140.89 84.27 611.46 329.89
N 33 24 38 36 37 37 37
Std dev 16.105 13.637 9.512 9.504 6.323 65.562 73.738
Min 450 439 182 112 69 433 176
Max 515 490 236 161 97 740 515
derbianus
Mean 474.86 461.67 198.11 140.29 95.58 864.23 684.23
N 7 6 9 7 12 13 13
Std dev 16.036 13.441 5.988 4.112 6.543 49.954 143.507
Min 455 446 187 134 89 800 483
Max 502 479 207 147 112 933 985
Females
oryx
Mean 447.38 435.58 181.23 134.11 61.17 634.45 335.32
N 29 19 31 27 29 29 28
Std dev 21.316 16.567 10.938 7.876 6.693 77.171 109.664
Min 384 404 153 114 51 524 175
Max 480 469 201 150 79 823 570
derbianus
Mean 450.71 439.57 182.88 138.00 69.63 718.63 431.75
N 7 7 8 8 8 8 8
Std dev 16.101 11.013 5.489 4.781 7.425 81.367 131.947
Min 418 419 171 133 55 579 211
Max 471 455 189 145 77 796 575
General color dull grayish brown, fi nely speckled more rarely so in the adults; palatine foramina long;
with golden-brown (speckling absent or weak on the lateral palatine notches extending farther forward, to
neck and the blaze), shading quite sharply to bright the front edge of the third upper molar. N. pygmaeus
orange-fulvous, without hair-banding, on the rump, resembles N. batesi and differs from N. moschatus in
the tail, the margins of the white belly-patch, the being less prognathous; and in having both the ante-
forelegs, the side of head (well below the eye), and the rior rim of the orbit farther forward, above the last
underside of the neck; dark brown on the pasterns premolar, and the auditory bullae large and inflated.
and the digits, and the fronts of the forelegs from the Known only from the forests of upper Guinea W
carpus downward; pale patches on the digits (almost of the Volta River, from Sierra Leone to Ghana.
white in the juveniles); white belly-patch narrow, ex-
tending down to the hock and the carpus on the in- Neotragus batesi (De Winton, 1903)
sides of the limbs; separate white throat-patch; tail dwarf antelope
colored like the upperparts above, but white below;
1903 Neotragus batesi De Winton. Efulen, Cameroon.
ears with whitish marks along the margin near the
1906 Hylarnus harrisoni Thomas. Semliki Forest, DRC.
base, and white inside.
N. pygmaeus differs in cranial features from N. Said to differ from N. pygmaeus in size: nearly half
batesi and N. moschatus in having the free anterior as large again as in the royal antelope, according
margin of the premaxilla bent and proximally more to Lydekker (1915a), and shoulder height 30 cm, ac-
vertical, as well as in having a median spinous pro- cording to Ansell (1971), but mean mea surements
cess; no maxillary-premaxillary fissure; nasals nar- for the males and the females, respectively, are
row at the base, not reaching back to the point of the 109 cm and 111 cm (total body length), compared with
lacrimofrontal suture on the orbital rim; minute 120 cm and 114 cm for N. pygmaeus, suggesting that —-1
maxillary canines frequently present in the juveniles, Lydekker (1915a) exaggerated the size differences. —0
349-47558_ch01_1P.indd 147 5/31/11 7:05 PM

148 Artiodactyla
Cranial measurements are very close in the two spe- are not significant, and the cranial characters of the
cies. Dragesco et al. (1979) have provided measure- holotype, described by O. Thomas (1906), are of an
ments from 138 adult animals: shoulder height (with individual nature: the maxillo-premaxillary vacuity
standard deviation) was 303 ±27.0 mm in 62 males, is very large in the holotype, but values for the series
310 ±22.6 mm in 75 females; weights were 2.26 ±0.46 kg of specimens overlap quite considerably: 6.5–13 mm
in 54 males, 2.53 ±0.45 kg in 66 females, but not neces- long in 20 W-C African N. batesi (mean 9.45 mm), and
sarily heavier than N. pygmaeus, for which the data 5.5–15 mm in 23 E-C African harrisoni (mean 9.76 mm).
are scanty. The nasals in the holotype are broad at the base, but
Said to differ from N. moschatus in the presence of nasals tend to broaden with age, as is also the case with
rudimentary lateral hoofs (Rode 1943), but these are N. moschatus.
not present in the specimens we have seen, nor were N. batesi has a discontinuous distribution, having
they in those examined by Dragesco et al. (1979). been recorded from three areas: (1) in the Owerri re-
General color superficially like N. pygmaeus, but gion of Nigeria, between the Niger and Cross rivers;
with less contrast between the dorsum and the lighter (2) in Cameroon S of the Sanaga River and E of 10° E,
flanks; speckling of the hairs coarser and more con- extending into N Gabon and the Congo Republic (ab-
trasting; speckling extending onto the blaze and the sent in Mimongo, MBigou, Koula Mouton, Mouila,
neck and, very fi nely, onto the ears; speckling more Fougamou, Fernan Vaz, and neighboring areas of S
olive-ochre than orange-ochre; blaze narrower, with Gabon [Malbrant & Maclatchy]); and (3) in the DRC
the transition from the dark blaze to the lighter face between the Congo River and the Rift Valley high-
color occurring above the eye; anterior surfaces of lands E of 24° E. (Beni, Epulu River, and Semliki For-
the forelegs and the metatarsi above the pasterns usu- est, S to Shabunda; Kibale, Kalinzu, and Mara-
ally darker than the body. magambo forests in Uganda [Kingdon 1982a]).
N. batesi differs cranially from N. pygmaeus, and
resembles N. moschatus, in having the premaxilla Neotragus moschatus Group
straight along its anterior margin and at a greater an- suni
gle from the vertical, with the median process usu- The suni is larger than the other species of Neotragus:
ally absent (present as a tiny spicule in a few speci- skull more than 110 mm in greatest length, compared
mens); a vacuity present between the maxilla and the with less than 110 mm in the other species; shoulder
premaxilla; nasals broad at the base, extending back height 1314 in. (= 3335 cm) in N. moschatus and 1415 in.
to the level of the lacrimofrontal suture on the orbit (= 3538 cm) in N. livingstonianus, according to Lydekker
rim; canines absent at all ages; palatine foramina short; (1915a); horns relatively longer, not below 60 mm in
lateral palatine notches not extending so far forward, length, and ringed.
only reaching the level of the hind edge of the third Suni differ from N. pygmaeus and N. batesi in hav-
upper molar. N. batesi differs from both N. pygmaeus ing a more prognathous skull, with the front border
and N. moschatus in the absence of an ethmoid fissure of the orbit farther back, above the fi rst or second up-
(Ansell 1971; Dragesco et al. 1979). per molar; preorbital fossa relatively larger; auditory
Dragesco et al. (1979) detected significant differ- bullae small, not inflated. Resemblances to N. batesi
ences between the sexes in several measurements, have already been discussed.
including mass and limb length. Mean head plus Sexual dimorphism in suni is possibly less than in
body length for the males was 98% of that for the fe- N. batesi (at least); skull length is virtually identical in
males; mean body width, 89%; and mean skull length the two sexes.
(the precise nature of the measurement was not Two main groups of taxa may be recognized: the
given) 97%, compared with 98% for our data from smaller animals from N of the Zambezi (N. kirchen-
the DRC specimens. paueri and N. moschatus) and the larger ones from S of
Geograph ical variation was implied by the recog- that river (N. livingstonianus and N. zuluensis). N. liv-
nition of the species (or subspecies) harrisoni. This ingstonianus differs from the nominate group not
form was described as being less contrasting in color- only in size, but also in the simplification of the band-
ation than typical N. batesi, but the holotype skin is ing pattern of the hairs, so that the pelage is streaked
not accessible, as it was returned to the donor after it rather than speckled. These two species, from N and
had been described. A more extensive series now S of the Zambezi River, have been confused, yet they
-1— available indicates that harrisoni cannot be distin- were generally treated as separate species until the
0— guished from N. batesi. Differences in measurements intervention of Ellerman et al. (1953).
349-47558_ch01_1P.indd 148 5/31/11 7:05 PM

b o v idae 149
Table 37 Skull measurements for taxa referred to Neotragus
Skull l Biorb br Horn l
Mean Range N Mean Range N Mean Range N
pygmaeus 103.6 99–108 19 46.8 47–50 21 25.5 18–35 5

batesi 104.4 99–109 35 47.6 47–51 36 31.4 26–38 7
moschatus 115.6 113–114 5 56.5 55–58 10 57.0 57–57 3
kirchenpaueri 118.0 113–120 22 58.3 50–57 22 67.2 63–82 14
livingstonianus 124.6 122–129 15 57.7 55–61 14 73.5 65–87 6
There is considerable variation in color in the N. Confi ned to forests in the highlands of Kenya E of
moschatus group, with the darkest specimens coming the Gregory Rift Valley, and in N Tanzania: Ab-
from montane localities (inland in Kenya and N Tan- erdares Escarpment, Mt. Kenya, Kikuyu and Langata
zania) and paler animals from lower altitudes (coastal near Nairobi, Mt. Kiliminjaro, Mt. Meru, Ngorong-
Kenya and Tanzania, Zanzibar, and possibly also in oro (Swynnerton & Hayman, 1951), Phillipshof (=
Mozambique). These are clearly distinct, with no in- Magamba) (G. M. Allen & Loveridge, 1927), and the
termediates, although no cranial measurements dif- Olmoti and Empakaai craters (Frame, 1982). N. kirch-
ferentiate them (there may be a difference in horn enpaueri is not positively recorded from isolated
length, however; see table 37). A few skins from Ma- mountain forests in S Tanzania (Kungwe, Ufipa, Po-
lawi present a zoogeographic problem. roto, Rungwe, Uluguru or Udzungwa), though the
species is said to occur in the Iringa district (Swyn-
Neotragus kirchenpaueri (Pagenstecher, 1885) nerton & Hayman, 1951). It also occurs in the high-
lands of Malawi.
1885 Nesotragus kirchenpaueri Pagenstecher. Great Arusha,
Tanzania.
Neotragus moschatus (Von Dueben, 1846)
1913 Nesotragus moschatus akeleyi Heller. SE slope of Mt.
Kenya, 7000 ft. 1846 Nesotragus moschatus Von Dueben in Sundevall.
Chapani Islet, Zanzibar, Tanzania.
2n = probably 52 (Kingswood, Kumamoto, Charter &
1913 Nesotragus moschatus deserticola Heller. Maji-Ya- Chumvi,
Jones, 1998). A captive group of suni, whose origin
Kenya.
was from stock caught both in KwaZulu-Natal and
on Mt. Kenya, had 2n = 52–56; certain indications, in- Very much lighter in color, and somewhat less strongly
cluding the relatively high perinatal mortality of cy- speckled, than N. kirchenpaueri; blaze not blackish;
totypes 53, 54, and 55, suggested to the authors that ventral tone more sandy.
these latter numbers were due to hybridization be- Thirty skins from Malawi have been examined by
tween the two original stocks. us, and most cannot be separated from the E African
Coloration dark, dull brown, speckled dorsally material. But four British Museum skins (Mpezo;
with browny buff, shading to dull sandy along the hills near Blantyre; two without exact locality) are
white belly-patch, and warmer sandy on the lower very dark, indistinguishable from N. kirchenpaueri.
neck, in both cases with loss of the dark bands on the Apparently N. moschatus shows as wide a range of
hairs; light fawn patch at the base of the ears; tail not color within Malawi as it does in Kenya and Tanza-
contrasting with the body, colored like the back, but nia, but over a much smaller area (these enigmatic
darkening toward the tip, with its unbanded hairs, specimens were discussed by Grubb, 1989). As the lo-
and white below; legs from the hock and the top of calization of the dark specimens is poorly known, at
the forelimb light sandy brown, like the lower neck; present it is not possible to relate color variation to
pasterns and digits dark brown, with a dark line run- habitat, but it may be that these dark skins could be
ning up to the carpus in some specimens; ears gray- representatives of a further taxon, if they are not
brown; blaze chestnut, with a dark brown to black identical to N. kirchenpaueri. This latter species, then,
streak on the muzzle; top of the head dark brown, appears to be polytopic, with an exceptionally discon- —-1
with chestnut speckling, or wholly deep chestnut. tinuous distribution. —0
349-47558_ch01_1P.indd 149 5/31/11 7:05 PM

150 Art iodactyla
Coastal Kenya and Tanzania (Maji ya Chumvi; whole body faintly tipped with dark brown; throat,
Mombasa Island; Sokoke Forest; Takaungu); Zanzi- belly, and groin to below the hock white; axilla to the
bar (including Chwaka and Dimani, per Rutzmoser, carpus white.
pers. comm. to PG), Chapani Islet (Von Dueben, 1846); Presumably intergrades with N. l. zuluensis in Mo-
Bawani Islet (Sclater & Thomas, 1895); Boydu Island, zambique; said to range S to about the Vila Pery and
Rufiji River delta; Sambala; Kilwa, Lindi, Liwale, Beira districts (Smithers & Tello, 1976); a specimen
and Mikindani districts (Rushby & Swynnerton, from Masangena, Save River, on the S border of these
1946); Kidenge, Uzaramo; Mpwapa; Tununguo, Ukami districts, is N. l. livingstonianus, but so is one from
(Matschie, 1895); Mafia Island; Mbulu and Iringa dis- Gazaland, still farther to the S (Smithers & Tello,
tricts (Swynnerton & Hayman, 1950); inland from 1976). Suni occur in Zimbabwe (Chorley, 1956; Child
Kilwa; Mikindani; Morogoro; S presumably to Mo- & Savory, 1964), although the taxonomic status for
zambique (from which the species is recorded from N the Zimbabwe suni is yet to be confi rmed.
of the Zambezi by Smithers & Tello [1976], but no
specimens from there are available for study), and Neotragus livingstonianus zuluensis
from there into S Malawi. Believed to have been in- (Thomas, 1898)
troduced to the Grave Islets off Zanzibar (Sclater &
1898 Nesotragus livingstonianus zuluensis Thomas. Umkuja (=
Thomas, 1895), namely Bawani and the type locality
Mkuzi) Valley, Zululand.
of the species, Chapani; not present on Pemba.
The Zanzibar population cannot be distin- 2n = 56 (Kingswood, Kumamoto, Charter & Jones,
guished in color or in size from some mainland pop- 1998).
ulations, so it cannot be regarded as a separate insu- Compared with N. l. livingstonianus, greater con-
lar subspecies. trast between the dorsum and the haunches, which
are gray; black hair tips overlap, forming dark streams
Neotragus livingstonianus (Kirk, 1865) on the rump; dark rump streams flowing together
Livingston’s suni into the blackish brown color of the tail; tail white
Pelage not obviously speckled; pale band to the below; white belly-patch extending up the throat
hair just below the long dark tip, but much narrower (Zululand), or a separate white patch on the throat
than in N. kirchenpaueri and N. moschatus, grading (Coguno); lower throat and hind shanks sandy buff,
evenly into the somewhat darker shaft; general color less rich in tone than in N. moschatus; back of the ears
much browner than in N. moschatus, but with greater dark gray. These are strong tendencies, but thus far it
contrast between the dorsum and the haunches; tail is not possible to draw absolutely sharp distinctions
dark above, white below; flanks shading to buff y, between N. l. zuluensis and N. l. livingstonianus.
without dark hair tips, in the vicinity of the white Mozambique S to the Hluhluwe River, Zululand
belly-patch; muzzle paler, black on the top; forehead (Ansell, 1971); said to range N toward the Vila Pery
deeper toned. and Beira districts (Smithers & Tello, 1976), but may
Known only from S of the Zambezi. Consistently not occur N of the Save River (A. Roberts, 1915; Pien-
larger than the two E. African species. aar, 1963; Rautenbach, 1982); other localities in Natal
mapped by Mentis (1974).
Neotragus livingstonianus livingstonianus
(Kirk, 1865) Tribe Aepycerotini Gray, 1872
Like the Alcelaphini, the Aepycerotini have trans-
1865 Nesotragus livingstonianus Kirk. Shupanga,
verse ridges on the horn cores, the face is elongated,
Mozambique.
the braincase is inclined, the zygomatic arch is ex-
In coloration, resembles N. l. zuluensis through being panded below the orbit, there is no ethmoid fissure,
superficially nonspeckled, but N. l. livingstonianus is there are strong longitudinal ridges on the basioccipi-
less streaked with black, the thighs less contrastingly tal, the central incisors are expanded, and the paraco-
grayish, and the tail not so dark. being speckled gray nid and metaconid are fused on the posterior lower
above, and white below; general color light fulvous- premolar. For these reasons, A. W. Gentry (1985)
brown, with very fine, light ticking, particularly on the placed Aepyceros in the Alcelaphini, although he later
hindquarters, formed by the pale pigment of the hair (1992) reversed his opinion.
-1— shaft lightening just below the dark tip; muzzle paler, On the basis of both a mitochondrial (16S) and a
0— black on the top; forehead deeper toned; hairs on the nuclear (b-spectrin) sequence, Kuznetsova & Kholo-
349-47558_ch01_1P.indd 150 5/31/11 7:05 PM

b o v idae 151
dova (2003) supported an association between Aepy- than the latter were from each other. This is a very
ceros and Neotragus. interesting result, as far as evolutionary theory goes—
centrifugal speciation is strongly indicated—but it
Aepyceros Sundevall, 1847 seems impossible to make a taxonomic arrangement
impala from it. At the moment, therefore, no subspecies are
Impala have a gland on the forehead in both sexes; in recognized in the common impala.
the males, these swell and increase in activity during
the rut, whereas in the females they do not change sea- Tribe Antilopini Gray, 1821
sonally; metatarsal glands likewise show no seasonal A. W. Gentry (1992) noted character states shared by
changes (Welsch et al., 1998). The males have a thick- all members of this tribe, including those “dwarf an-
ened dermal shield, covering at least the shoulders, pre- telope” that he included in it for the fi rst time: an ex-
sumably related to their fighting style ( Jarman, 1972). panded lacrimal bone; absence of basal pillars on the
The phylogeny by Nersting and Arctander (2001), molars; very wide temporal ridges; palatal ridges
using the mitochondrial control region, separated touching one another; fusion of the metaconid and
most samples of the two species; in pairwise com- the entoconid on the posterior lower premolar; a lat-
parisons, the two species were separated by an aver- erally swollen femoral head; and an expanded hypo-
age of 9.8 substitutions, whereas only an average of conulid on the posterior lower molar that is not offset
2.6 substitutions was found among A. melampus. One labially and has a central cavity.
haplotype from A. petersi, however, was nested very What was perhaps the first character-based phylog-
deeply within the A. melampus clade, presumably a eny of the Antilopini (as the group was restricted at
result of recent interbreeding. that time, i.e., without the “dwarf antelope” and the
saiga) was that of Eff ron et al. (1976), based on chro-
Aepyceros petersi Bocage, 1879 mosomes. Their proposed phylogeny was as follows:
black-faced impala
((Antidorcas) (Litocranius) (Antidorcas, Eudorcas)
Although there has been widespread translocation of
((Antilope, Nanger) Gazella)).
A. melampus into game farms in Namibia, on the bor-
ders of Etosha National Park, a study of eight micro- Notably, the Antilope-Nanger-Gazella clade was char-
satellite markers found no evidence for introgression acterized by an X-autosome translocation and many
(Lorenzen & Siegismund, 2004), so any interbreed- Robertsonian fusions among the autosomes.
ing, as inferred from the fi nding of Nersting & Arc- Rebholz & Harley (1999) sequenced the mitochon-
tander (2001— see above), must be very infrequent. drial cytochrome b and cytochrome c oxidase III genes,
The differentiation between A. petersi and A. melam- producing the following neighbor-joining phylogeny:
pus, in both mtDNA and microsatellites, is very high.
(Raphicerus (Ourebia, Madoqua (Antidorcas ((Eudorcas,
Nanger) (Antilope, Litocranius, Gazella))))).
Aepyceros melampus (Lichtenstein, 1812)
common impala The parsimony cladogram was slightly different:
Lorenzen, Arctander, et al. (2006), using mtDNA se-
((Raphicerus, Ourebia, Madoqua) ((Litocranius,
quences and microsatellites, found some evidence of
Antidorcas) (Eudorcas (Nanger (Antilope (Gazella)))))).
genetic differentiation between S and E African sam-
ples of A. melampus. It is fair to say, however, that most of the nodes in the
Bastos-Silveira & Lister (2007) made a metrical parsimony cladogram that were unresolved in the
analysis of skulls and horns, with an assessment of pel- neighbor-joining cladogram had very low support.
age color and patterning. They found that A. petersi is
thoroughly distinct, and they were also inclined to Raphicerus Hamilton Smith, 1827
recognize a South African, an E African, and a Zam-
The early literature on South African members of this genus
bian/Malawi subspecies. Their maps of coat color and
was surveyed by Rookmaaker (1988), who pointed out
facial patterns showed that, among A. melampus, there
that the presently used name Raphicerus campestris
was no significant geographic variation, except that
Thunberg, 1811, is actually antedated by Antelope dama
the Zambian sample did tend to stand out somewhat.
var. rupestris Forster, 1796, and Raphicerus melanotis
Again, in their principal components analysis, the
Thunberg, 1811, is antedated by Antelope dama var.
Zambia sample (the one in the middle) was differenti- —-1
melanotis Forster, 1790. He argued that although
ated much more from the S and E African samples —0
349-47558_ch01_1P.indd 151 5/31/11 7:05 PM

152 Art iodactyla
Forster’s two names are not clearly available in the sense Raphicerus colonicus Thomas & Schwann, 1906
of the International Code of Zoological Nomenclature, Limpopo grysbok
there is room for argument; he therefore recommended Size larger than R. melanotis and R. sharpei, but the
continued use of Thunberg’s names. skull relatively narrower; preorbital fossa wider; na-
sals broaden posteriorly, nearly crowding out the lac-
The two species-groups are distinguished by body
rimal vacuity; bullae little inflated. Numerous white
build (relative length of the legs), pelage, size, horn
hairs interspersed among the browner ones, like the
size, and skull characters. Craniometrically, size, espe-
other two species, but the overall color richer, darker ;
cially the large greatest skull length compared with
legs lighter, more reddish; underparts, including the
condylobasal length (indicating greater development
throat, more buff, with the boundary line more indis-
of the occipital crest), is the main discriminator. The
tinct on the sides of the belly. Horns similar in size to
two groups are very different, and it may well be that
R. sharpei. No lateral hoofs.
they will be found to be generically distinct.
South Africa (N of the S coastal strip) and Zim-
babwe; probably also S Mozambique, S of the
Raphicerus melanotis Group
Zambezi.
grysbok
Relatively short legs; small in size, with the rump
Possible Other Raphicerus Species
higher than the withers; short horns; preorbital fossa
A fourth, unnamed, species, from W and C Zambia,
wider; nasals posteriorly broader, nearly crowding out
included in table 38, is easily separable craniometri-
the lacrimal vacuity; bullae little inflated. We can di-
cally from all other grysbok, although the sample sizes
vide the species in the R. melanotis group into two
are small. Skull length is somewhat greater than in R.
subgroups: the R. sharpei subgroup, lacking lateral
sharpei, nasal breadth is much greater than either R.
hoofs, and R. melanotis itself, which possesses them.
sharpei or R. colonicus, and the teeth are smaller than
in either of them.
Raphicerus melanotis (Thunberg, 1811)
Cape grysbok
Raphicerus campestris (Thunberg, 1811)
1790 Antelope dama var. melanotis Forster, nom. nud. (see steenbock
Rookmaaker, 1988). More lightly built, taller, and longer-bodied than the
1796 Antelope dama var. grisea Forster; not Boddaert, 1785. three (four?) grysbok species, but with the same high
1811 Antilope melanotis Thunberg. Cape of Good Hope. rump and short face. Pelage not noticeably white-
1816 Antilope grisea Cuvier. ticked; broad white eye-rings. Horns longer, sharply
1822 Antilope rubro-albescens Desmoulins. upright. Lateral hoofs present.
Craniometrically, there are no absolute divisions
Color reddish brown, with white hairs intermingled,
between any of the geographic samples within this
becoming yellower on the cheeks, the legs and the sides
widespread species (table 39), and those differences in
of the neck and the body; a white patch on the throat.
external features are slight. Therefore, the described
Muzzle brown above, darkest in the midline, but with-
taxa are subspecifically distinct, at best.
out a defined nose-patch. Sides of the muzzle dirty white.
Lateral hoofs present, unlike the other species of
Raphicerus campestris campestris (Thunberg, 1811)
grysbok.
S of the Karroo, as far as S KwaZulu-Natal. 1783 Antilope grimmia Sparrman; not Capra grimmia
Linnaeus, 1758.
Raphicerus sharpei Thomas, 1897 1790 Antelope dama Forster; not Antilope dama Pallas, 1766.
Sharpe’s grysbok 1796 Antelope dama var. rupestris Forster, nom. nud. (see
Color like R. melanotis, and similar to R. melanotis in Rookmaaker, 1988).
most other respects, but the ears large and thinly haired 1811 Antilope campestris Thunberg.
at the back, mostly white, but black along the margins; 1811 Antilope capensis Thunberg; not P. L. S. Müller, 1766.
underparts whiter; legs lighter reddish. A striking dif-
General color orange-rufous; cheeks and legs
ference is that there are no lateral hoofs. Tail short, col-
ochraceous-tawny. Large white area on the under-
ored above like the back, but white below. Horns much
parts, not extending down the inside of the limbs;
shorter than in R. melanotis; teeth smaller.
-1— white patches on the throat less extensive than in the
Zambia E of the Luangwa; Malawi; presumably N
0— other subspecies. Muzzle dark brown. Dark brown
Mozambique? © 2011 The Johns Hopkins University Press
349-47558_ch01_1P.indd 152 5/31/11 7:05 PM

b o v idae 153
Table 38 Univariate statistics for the Raphicerus melanotis group
Gt l Cb l Gt br Nas l Nas br Max teeth Horn l
melanotis
Mean 137.79 129.58 67.86 37.29 17.21 44.14 75.20
N 7 6 7 7 7 7 5
Std dev 4.222 5.886 1.345 1.976 3.026 .900 8.927
Min 132 121 66 35 14 43 65
Max 143 136 70 40 23 45 86
sharpei group
colonicus
Mean 132.40 126.00 64.00 35.80 14.50 42.17 46.00
N 5 5 6 5 5 6 2
Std dev 2.793 2.828 2.449 1.924 1.658 1.835
Min 129 123 60 34 13 40 44
Max 135 129 67 39 16 45 48
sharpei
Mean 127.25 120.75 63.75 37.25 14.25 42.50 —
N 4 4 4 4 4 4 —
Std dev 1.500 1.708 3.775 1.500 1.500 1.291 —
Min 126 119 60 36 13 41 —
Max 129 123 69 39 16 44 —
W + C Zambia
Mean 128.80 121.60 63.60 38.40 16.80 40.80 43.67
N 5 5 5 5 5 5 3
Std dev 2.683 2.608 2.302 3.209 .837 1.643 5.686
Min 126 119 61 34 16 39 39
Max 132 125 66 42 18 43 50
horseshoe-like patch on the crown. Tail short, colored Apparently from KwaZulu-Natal into the E parts
above like the back, and white below. of the former Transvaal.
The limited evidence suggests that this subspecies
may be smaller, but with larger teeth, than others as- Raphicerus campestris capricornis Thomas &
cribed to this species. Schwann, 1906
S and SW Cape Province, according to A. Roberts Amber-brown, more rufous than R. c. fulvorubescens,
(1951). and with the face much lighter. White extends con-
tinuously from the throat to the chin, and more ex-
Raphicerus campestris fulvorubescens tensively elsewhere below.
(Desmoulins, 1822) According to the limited dataset, specimens
Rather lighter than R. c. campestris, near “amber from the topotypical region seem to have shorter
brown.” White of the throat extending almost to the horns.
chin; lower lip whitish; white on the underparts ex- Between the Limpopo and the Zambezi rivers, ac-
tending down inside of the upper part of the thighs cording to A. Roberts (1951).
and the forelegs.
Albany district, according to A. Roberts (1951). Raphicerus campestris neumanni (Matschie, 1894)
1908 Raphicerus neumanni stigmaticus Lönnberg. Lake Natron.
Raphicerus campestris natalensis
(Rothschild, 1907) No dark, crescentic coronal mark. General color paler.
White eye-rings large, complete. More white on the
1914 Raphicerus campestris zuluensis Roberts. Umfolosi.
margins of the ears. White more clearly defined on the
Color much the same as in R. c. capricornis; R. c. zu- underside, also on the lips.
luensis particularly distinguished by the much longer Occupies the species range N of the Zambezi, into —-1
horns. S Kenya. —0
349-47558_ch01_1P.indd 153 5/31/11 7:05 PM

154 Artiodactyla
Table 39 Univariate statistics for Raphicerus campestris subspecies
Gt l Cb l Gt br Nas l Nas br Max teeth Horn l
campestris
Mean 137.00 — 66.75 36.50 13.00 50.80 95.00
N 2 — 2 1 1 2 1
Min 135 — 65 — — 50 —
Max 139 — 69 — — 52 —
fulvorubescens
Mean 145.50 137.00 73.00 45.00 20.00 47.00 101.50
N 1 1 1 1 1 1 1
zuluensis
Mean 143.44 135.89 68.72 39.78 14.56 46.50 115.00
N 9 9 9 9 9 9 3
Std dev 3.005 2.607 2.623 4.644 2.800 2.179 10.817
Min 139 131 66 34 11 43 103
Max 147 139 74 47 20 50 124
natalensis
Mean 141.75 135.25 67.50 40.00 13.25 46.00 113.50
N 2 2 2 2 2 2 1
Min 141 133 66 36 13 43 —
Max 143 138 70 44 14 49 —
capricornis
Mean 144.42 136.50 70.33 42.58 14.42 46.58 79.38
N 6 6 6 6 6 6 4
Std dev 6.583 5.941 1.602 4.576 1.594 1.744 11.309
Min 134 126 68 37 13 45 66
Max 152 143 72 50 17 49 92
steinhardti
Mean 142.07 135.00 66.71 44.42 13.00 45.64 93.00
N 7 7 7 6 6 7 1
Std dev 2.950 3.096 1.997 4.005 .894 1.994 —
Min 138 131 64 38 12 43 —
Max 147 140 69 50 14 49 —
kelleni
Mean 142.91 134.82 68.09 42.00 15.91 45.64 86.00
N 11 11 11 11 11 11 6
Std dev 3.390 3.920 2.427 2.898 2.256 1.567 9.143
Min 138 129 63 38 13 43 75
Max 148 142 71 46 21 47 98
neumanni
Mean 141.82 133.91 69.36 43.36 16.36 46.00 100.22
N 11 11 11 11 11 11 9
Std dev 2.089 2.071 2.014 4.388 2.157 2.490 9.615
Min 139 131 66 35 14 41 90
Max 146 139 72 49 21 50 116
Over the whole of this wide area, we can fi nd no Orange River N to Ovamboland, throughout the
noticeable geographic variation. Kalahari and Botswana, and into Angola.
This is certainly the most distinctive of the de-
Raphicerus campestris steinhardti (Zukowsky, 1924) scribed subspecies, both in the pelage and the skull.
A. Roberts (1951) gives the full synonymy for this
Antidorcas Sundevall, 1847
subspecies.
springbok
-1— Paler than other subspecies; white parts extensive, like 2n = 56 in all three taxa of this genus; the X chromo-
0— R. c. capricornis. some is a large acrocentric, and the Y chromosome is
349-47558_ch01_1P.indd 154 5/31/11 7:05 PM

b o v idae 155
a small metacentric, the only metacentric chromo- ability of winter dietary protein, raising the question
some in the genome (Robinson & Skinner, 1976). The of the degree to which this difference in body mass is
karyotype is close to that of the reconstructed ances- due to genetic factors or to phenotypic plasticity.
tral caprine type, and quite unlike that of any “ga- The general taxonomy here follows Groves (1981c),
zelle” (Vassart, Séguéla, et al., 1995). except that the three subspecies recognized therein
Horn rings of the males very prominent, widely are raised to species rank, as they differ 100%.
spaced, circular (all around the horn), connected by
longitudinal ridges. Horn cores circular in cross sec- Antidorcas marsupialis Zimmermann, 1780
tion, with traces of the rings on them, and with broad, Rich chestnut-brown; flank-stripe deep brown; pygal-
deep grooves up the posterior surface. Anterior pre- stripe well marked. Face-stripes very thin, not dark;
molar (P2) very reduced, generally absent in the man- forehead brown, fawn, or white, this color not extend-
dible; styles on the other premolars small, but well de- ing in front of the level of the eyes, and (if brown or
veloped on the molars; molars elongated, especially fawn) not sharply bordered anteriorly. Nose white, or
the distal selene, with its flat buccal wall. Mandibular with a brown smudge. Horns exceptionally short in
P4 elongate, with the lingual wall open mesially, nearly the females, only 60%–70% of the length of those of
closed distally; mandibular molars elongate, their buc- the males.
cal midwall complexities developed into stylids, their S of the Orange River, from the NE Cape to the
buccal valleys deep, but their lingual valleys hardly Free State and the Lombard and Kimberley districts.
developed; distal lobe on the posterior lower molar
well individualized. Supraorbital foramina flush with Antidorcas hofmeyri Thomas, 1926
the forehead; lacrimal bone extending forward and Light fawn; lateral band dark brown to nearly black;
downward onto the face, and into the preorbital fossa; pygal band medium brown, thin. Face-stripes thin,
ethmoid fissure short; lacrimal bone with fenestrae; dark brown. Forehead anteriorly extending in front of
median nasal tips very long, the lateral prongs very the level of the eyes, brown or pale fawn, with no an-
short. Basioccipital short and broad; only a narrow terior border, so the transition to the white of the face
channel between the anterior basioccipital tuberosi- not sharply demarcated. Dark smudge on the nose
ties; anterior tuberosities small, the posterior ones varying from dark to pale to absent altogether. Larger
represented only by low ridges, and the longitudinal than the other two species, to judge from the skulls.
ridges between the anterior and the posterior pairs Horns of the females clearly annulated, large, 77%–
barely indicated; pterygoid plates evenly sloping. Cra- 85% the length of those of the males.
nial profi le short, high; facial profi le straight; horns N of the Orange River, from the Upington district
placed slightly behind the posterior margin of the or- (N Cape) via the Sandfontein district through Bo-
bit, sloping back at a 45° angle in the males; tips of tswana to Namibia.
the premaxillae well above palatal level; preorbital
foramen rounded, above the posterior premolars; pala- Antidorcas angolensis Blaine, 1922
tal surface of the diastema raised well above the plane Brown-tawny; lateral band nearly black; pygal band
of the palate; auditory meatus oblique, fairly rounded. very dark brown. Face-stripes thick, rich dark brown,
No glandular knee-tufts; preorbital glands present; extending two-thirds of the way to the muzzle. Fore-
foot glands present; inguinal glands absent; rhinarium head, down to in front of the eye level, medium brown,
restricted to a median line; one pair of mammae; tail bordered anteriorly by a dark brown edge; rest of the
relatively long. (Following Groves, 2000.) face sharply white, except for a brown spot on the nose.
In a survey of this genus, Lange (1970) pointed out Horns of the females angulated, relatively large, 87%
the reduced sexual dimorphism of Antidorcas in com- of the length of those of the males.
parison with Gazella. Angola, N to the latitude of Benguela.
Robinson (1979) found strong and significant size
differences between the two N taxa (designated A. m. Ammodorcas Thomas, 1891
angolensis and A. m. hofmeyri; note that his Kaokoveld Females without horns. Horn rings of the males very
sample actually represents A. m. hofmeyri) and the S prominent, widely spaced, angulated in front, much
nominotypical A. m. marsupialis: the mean body mass flatter behind than in front, connected by longitudinal
of the males of the first two taxa was 42.0 and 41.6 kg, ridges. Horn cores rather compressed, with traces of
respectively, and of A. m. marsupialis, 31.2 kg. Body rings on them, and with broad, deep grooves up both —-1
mass was found to be highly correlated with the avail- the anterior and the posterior surfaces. Anterior —0
349-47558_ch01_1P.indd 155 5/31/11 7:05 PM

156 Art iodactyla
premolar (P2) very small, oval in both jaws; styles on shallow pit; lacrimal bone short, not extending into
the other premolars small, but well developed on the the preorbital fossa; ethmoid fi ssure long, extending
molars; molars elongated, especially the distal selene, some way down the lateral edge of the nasal; lacrimal
with its flat buccal wall. Mandibular P4 elongate, with bone with numerous fenestrae; median nasal tips very
the lingual wall open both mesially and distally; man- long, the lateral prongs well developed. Basioccipital
dibular molars elongate, their buccal midwall convex- long and narrow; only a narrow channel between the
ities developed into stylids, their buccal valleys deep, anterior basioccipital tuberosities; anterior tuberosities
but their lingual valleys hardly developed; distal lobe small, the posterior ones represented by bulky ridges,
on the posterior lower molar simple. Preorbital fossa and longitudinal ridges between the anterior and the
shallow. Supraorbital foramina flush with the fore- posterior pairs barely indicated; pterygoid plates with a
head; lacrimal bone extending onto the face, but not large hamuli. Cranial profile very long and low; facial
into the preorbital fossa; ethmoid fissure short; lacri- profile a convex T; horns placed directly above the or-
mal bone with numerous fenestrae; median nasal tips bit, rising at an angle of nearly 90°; tips of the premax-
very long, the lateral prongs fairly long. Pterygoid illae at or below the occlusal level; preorbital foramen
plates evenly sloping. Cranial profi le very long and slitlike, above the anterior premolar; palatal surface
low; facial profile straight; horns placed slightly behind of the diastema in the plane of the palate; auditory
the posterior margin of the orbit, sloping at a 60° angle meatus vertical-oval. Glandular knee-tufts present;
in the males; tips of the premaxillae at or below the oc- preorbital glands present; foot glands present; inguinal
clusal level; preorbital foramen slitlike, above the ante- glands absent; rhinarium extending slightly along the
rior premolar; palatal surface of the diastema in the dorsal rims of the nostrils; two pairs of mammae; tail
plane of the palate; auditory meatus vertically oval. relatively long. (Following Groves, 2000).
Glandular knee-tufts and preorbital glands present; Taxonomic characters follow Grubb (2002), but
foot and inguinal glands absent; rhinarium restricted the two taxa recognized therein as subspecies are
to the median line; two pairs of mammae; tail very here recognized as full species, because their ranges
long. (Following Groves, 2000). of variation are discrete.
Ammodorcas clarkei (Thomas, 1891) Litocranius walleri (Brooke, 1878)

dibatag
1878 Gazella walleri Brooke. Grubb (2002) designated
On the general biology of this species, see Schomber
Chisimaio (0° 23' S, 42° 34' E) as the restricted type locality.
(2007); on its distribution and advances in its conser-
vation, see Wilhelmi et al. (2007). Size small; skull length of the males 211–241 mm
A. clarkei is essentially a species of restricted distri- (mean 231.6 mm, N = 25), of the females 207–212 mm
bution in the Somali arid zone. (mean 210.9 mm, N = 5) (Grubb, 2002). A band of re-
versed hairs, directed toward the head, on the dorsal
Litocranius Kohl, 1886 midline of the neck.
gerenuk NE Tanzania through Kenya to Galcaio in E-C
Horns absent in the females. Horn rings flattened, Somalia, N of the Webi Shebeyli.
very close set, circular (all around the horn), con-
nected only by very poor longitudinal ridges. Horn Litocranius sclateri Neumann, 1899
cores very compressed, with only irregular, shallow,
1899 Lithocranius [sic] sclateri Neumann. Berbera, Somalia.
short grooves, but with deep grooves up both the an-
terior and the posterior surfaces. Anterior premolar Size large; mean skull length of the males 248–259 mm
(P2) not reduced; styles on the other premolars very (mean 251.2 mm, N = 10), of the females 226–231 mm
prominent, producing a “winged” appearance, but less (mean 227.3 mm, N = 4). Other cranial measurements,
well developed on the molars; molars not elongated, especially of the facial skeleton, much larger than in
the selenes with concave buccal walls. Mandibular pre- L. walleri, with their ranges not overlapping. No hair
molars short, with the lingual wall open both mesi- reversal on the neck.
ally and distally; mandibular molars short, their buccal No color differences can be detected between L.
midwall convexities not developed into stylids, their sclateri and L. walleri.
lingual valleys well developed; distal lobe on the lower NW Somalia: Berbera district and W, to just over
-1— molar simple, very small. Supraorbital foramina in a the Ethiopian border; Djibouti.
0—
349-47558_ch01_1P.indd 156 5/31/11 7:05 PM

b o v idae 157
netic study (Hassanin & Douzery, 1999a; and other

Saiga Gray, 1843 sources).
saiga The extreme specializations of the nasal cavity
Females hornless. Horns of the males amber-colored, have been studied by Clifford & Witmer (2004). The
not blackish like other antilopins, their rings very cavity is enormously enlarged and inflatable, with a
prominent, widely spaced, circular (all around the lateral vestibular recess, apparently entirely neomor-
horn), connected only by very poorly expressed lon- phic, and a largely membranous nasal septum, in which
gitudinal ridges. Horn cores circular in cross section. there is a large patch of tissue acting as a baffle, con-
Anterior premolar (P2) very reduced, generally ab- trolling air flow.
sent in the mandible; styles on the other premolars Characters of the two very distinctive taxa largely
small, but well developed on the molars; molars follow Bannikov (1963).
short. Mandibular P 4 elongate, with the lingual wall
closed both mesially and distally; mandibular molars Saiga tartarica (Linnaeus, 1766)
elongate, their buccal valleys deep, but their lingual
1766 Antilope saiga Pallas.
valleys hardly developed; distal lobe on the lower
1766 Capra tartarica Linnaeus. Ural steppes of Russia.
molar well individualized. Nasal bones very reduced,
1767 Antilope scythica Pallas.
above a very deep nasal cavity. Supraorbital foramina
1827 Antilope colus Hamilton Smith.
flush with the forehead; lacrimal bone extending for-
ward and downward onto the face, but not into the Horns thick, waxy-colored, semitranslucent, with 12–
preorbital fossa; ethmoid fissure absent, probably cor- 20 strongly marked rings; horns 25–33 mm thick at
related with the extremely short nasal bones; lacri- the base, 280–380 mm long. Nasal opening enormously
mal bone with no fenestrae; median nasal tips very enlarged, raised. Summer coat yellowish red, paler on
long, the lateral prongs very short. Basioccipital short the flanks, the underside white, with darker zones on
and broad, with a broad smooth channel between the the shoulders and the loins; winter coat very light,
anterior basioccipital tuberosities; anterior tuberosities gray-colored. Condylobasal length of the males 222–
small, ridged, the posterior ones represented only by 250 mm, of the females 205–209 mm.
low ridges, and the longitudinal ridges between the Historically found from the borders of the Carpath-
anterior and the posterior pairs barely indicated; ptery- ians E to Dzungaria; now reduced to small fragments,
goid plates with large hamuli. Cranial profile very mainly in Kazakhstan, but as far W as the right bank of
short and high; facial profile convex, the facial skeleton the Volga River.
very deep; horns placed slightly behind the posterior
margin of the orbit, sloping back at a 60° angle; tips of Saiga mongolica Bannikov, 1946
the premaxillae at or below the occlusal level; preor-
1946 Saiga mongolica Bannikov. Tshargin Gobi, Mongolia.
bital foramen slitlike, above the posterior premolars;
palatal surface of the diastema in the plane of the pal- Horns relatively short, thin, with weakly expressed
ate; auditory meatus obliquely oval. Glandular knee- rings; bases of the horns up to 28 mm thick; horn
tufts present; preorbital, foot, and inguinal glands all length at most 220 mm; frontals more sloping anteri-
present; rhinarium absent; two pairs of mammae; tail orly, with the nasal opening less raised than in S. tar-
very short. (Following Groves, 2000). tarica. Summer coat sandy gray; dorsal region not
Saiga was formerly included in the Caprini, but A. darkened, but the brown spot in the lumbar region
W. Gentry (1992) pointed out that it shares some very large and sharply bordered. Skull length 203–237 mm.
cogent character states with acknowledged Antilo- E of the Mongolian Altai, in the basins of the W
pini: the horn core diameter is enlarged, the inter- lakes in Mongolia, from nearly 50° N to 48° N: Kirghiz
frontal suture is complicated, the anterior basioc- Nor SE to the Tshargin Gobi, and periodically to Orok
cipital tuberosities are strong and localized (as in Nor. S. mongolica was included as a subspecies of the
some of the Caprini), there is a central cavity on the fossil S. borealis by Baryshnikov & Tikhonov (1994).
hypoconulid of the posterior lower molar, the digi-
tal flexor ridge on the tibia is laterally placed, and Antilope Pallas, 1766
the lateral tubercle on the radius is high-placed. The 2n = 30 (females), 31 (males).
cranial roof, however, is not strongly inclined. Gentry’s Antilope emerges as sister to Gazella (that is to say,
insight has been vindicated by molecular phyloge- it is closer to Gazella than are Eudorcas and Nanger) in —-1
—0
349-47558_ch01_1P.indd 157 5/31/11 7:05 PM

158 Art iodactyla
most phylogenies (Rebholz & Harley, 1999), although cluding Eudorcas as a species of Gazella, they regarded
the earliest proposed phylogeny placed it as sister to blackbuck as being actually nested within Gazella).
Nanger (Eff ron et al., 1976). Its detailed karyotype, in
fact, more closely resembles that of the Gazella sub- Antilope cervicapra (Linnaeus, 1758)
gutturosa group than that of other gazelles (Vassart, blackbuck
Séguéla, et al., 1995).
The full synonymy of this species will be found in Groves
Horn of the males fairly prominent, widely spaced;
(1982b, 2003).
rings circular (all around the horn), connected by
poorly expressed longitudinal ridges. Horn cores The taxonomy follows Groves (1982b, 2003). There is
nearly circular in cross section, with no traces of a possibility that the two recognized subspecies may
rings on them, and with one groove up the posterior actually be distinct species, but more needs to be
surface and one up the anterior surface. Anterior pre- known about their individual variations.
molar (P2) reduced in the mandible, not in the max-
illa; styles prominent on the other premolars and on Antilope cervicapra cervicapra (Linnaeus, 1758)
the molars; molars elongated, especially the distal Smaller, with short, fi ne hair; dark color of the up-
selene, with its concave buccal wall. Mandibular P 4 perside running all down the limbs to the hoofs;
short, with the lingual wall open mesially, nearly white eye-ring narrowed above the eye. Horns rela-
closed distally; mandibular molars elongate, their tively short, not very divergent, with a relatively open
buccal midwall convexities not developed into styl- spiral.
ids, their buccal valleys shallow, but their lingual val- E and S of the Delhi region, as far S as Madras and
leys more developed; distal lobe on the lower molar Karnataka, and E to Bengal.
well individualized. Supraorbital foramina in shallow
pits; lacrimal bone extending forward and downward Antilope cervicapra rajputanae Zukowsky, 1927
onto the face, and into the preorbital fossa; ethmoid Larger, with longer, roughened pelage; males, in the
fissure long, extending some way down the lateral breeding season, with a gray sheen; shanks largely
edge of the nasal; lacrimal bone with a single fenestra; white, with little or no extension of the dark color
median nasal tips shorter than the lateral prongs. Basi- from the upper limb segments; white eye-ring broad
occipital somewhat elongated; only a narrow channel all around the eye. Horns tending to be longer, more
between the anterior basioccipital tuberosities; ante- divergent, and more closely spiraled.
rior tuberosities large, ridged, the posterior ones more W of the Delhi region, to Saurashtra and Vado-
bulky, and the longitudinal ridges between the ante- dara, Amritsar, and into Pakistan.
rior and posterior pairs barely indicated; pterygoid
plates evenly sloping. Cranial profi le short, high; fa- Nanger Lataste, 1885
cial profi le straight; horns placed slightly behind the 2n = 40.
posterior margin of the orbit, sloping back at a 45° There is little similarity between this genus and
angle in the males; tips of the premaxillae well above any other in chromosomes, except that, like Eudorcas,
palatal level; preorbital foramen rounded, above the Nanger has a Y-autosome translocation, as well as the
posterior premolars; palatal surface of the diastema X-autosome translocation characteristic of Gazella
raised somewhat above the plane of the palate; audi- and Antilope (Vassart, Séguéla, et al., 1995).
tory meatus oblique, fairly rounded. Glandular knee- Differs from Antilope as follows: horn cores more
tufts present; preorbital glands present, even hyper- compressed; several shallow grooves up the posterior
trophied; foot and inguinal glands absent; rhinarium surface of the core, instead of long, deep, broad ones;
extends all along the dorsal rims of the nostrils; one molar styles less well developed; distal selenes of the
pair of mammae; tail short. (Following Groves, 2000). molars with a broad, flat buccal wall, instead of being
Schreiber et al. (1997) found low protein polymor- concave; anterior lower premolar less reduced; lin-
phism in this genus compared with Gazella, Eudorcas, gual stylids of the lower molars less well developed;
and Antidorcas, and they ascribed this to a history of buccal valleys of the lower molars deeper; lacrimal
rapid evolution of its highly autapomorphic phenotype, bone usually with no fenestra; pterygoid plates with
aided by strong selection of a few dominant males, due large hamuli; cranial profi le more elongated; horns
to its lekking behavior, since its common ancestor was directly above the orbits, sloping back less; premax-
-1— with Gazella (since the authors were, at the time, in- illa tips at or below the occlusal level; preorbital fora-
0—
349-47558_ch01_1P.indd 158 5/31/11 7:05 PM

b o v idae 159
Table 40 Univariate statistics for the Gazella subgutturosa group: males
Horn l Tip–tip Span Base br Nas br ant Nas br post Gt l Braincase l
marica
Mean 270.188 101.50 168.118 56.278 22.682 22.500 181.333 100.200
N 16 14 17 18 11 12 9 15
Std dev 22.0884 41.210 39.6562 2.4448 3.1167 2.2764 6.9101 3.8210
Min 234.0 50 97.0 50.0 17.0 20.0 172.0 92.0
Max 312.0 178 228.0 61.0 28.0 28.0 193.0 107.0
Iraq
Mean 271.875 108.00 181.375 59.700 23.333 22.800 185.500 104.500
N 8 7 8 10 3 5 2 6
Std dev 16.7369 38.066 25.1733 1.7029 1.5275 2.2804 6.3640 4.0373
Min 251.0 46 152.0 57.0 22.0 20.0 181.0 100.0
Max 296.0 148 228.0 63.0 25.0 25.0 190.0 112.0
Caucasus subgutturosa
Mean 282.000 107.00 196.000 67.500 23.500 26.000 215.000 116.500
N 1 1 1 2 2 2 2 2
Std dev — — — .7071 .7071 1.4142 5.6569 3.5355
Min 282.0 107 196.0 67.0 23.0 25.0 211.0 114.0
Max 282.0 107 196.0 68.0 24.0 27.0 219.0 119.0
Iran seistanica
Mean 297.333 129.56 200.778 63.950 25.125 26.313 208.500 113.000
N 9 9 9 10 8 8 6 8
Std dev 21.4418 71.544 57.2904 2.5653 1.4577 1.6677 6.5345 3.9641
Min 246.0 56 151.0 60.0 22.0 24.0 199.0 106.0
Max 318.0 245 298.0 69.0 27.0 28.0 219.0 118.0
diversicornis
Mean 308.835 142.68 219.218 74.400 25.664 27.125 203.779 112.035
N 17 17 17 17 14 12 14 17
Std dev 26.4785 53.170 37.4524 1.8894 1.4302 1.4169 4.2392 3.3429
Min 229.7 78 151.7 70.5 24.0 25.0 195.8 104.7
Max 331.0 250 302.0 77.4 29.5 29.5 210.4 117.3
yarkandensis
Mean 280.778 92.93 175.875 69.000 27.310 28.840 212.375 112.230
N 9 7 8 8 10 10 8 10
Std dev 32.9802 36.788 26.9997 5.9761 1.4364 1.2607 5.7056 4.8369
Min 213.0 41 128.0 64.0 25.0 27.0 201.0 106.0
Max 318.0 137 214.0 79.0 29.0 31.0 220.0 122.0
hillieriana
Mean 258.252 109.47 163.737 64.380 27.284 29.332 203.763 107.391
N 21 18 19 25 25 25 19 23
Std dev 16.4623 35.398 28.0888 4.3091 1.5491 1.8443 5.5587 4.3142
Min 233.0 39 120.0 58.0 25.0 26.0 194.0 99.0
Max 310.0 176 213.0 79.0 31.0 33.0 218.5 116.0
men farther forward; glandular knee-tufts hyper- cialized, with strong and progressive pattern reduc-
trophied; preorbital glands less enlarged; rhinarium tion, the pattern tending to remain more strongly ex-
extending only slightly along the dorsal rims of the pressed in the females than in the males. He pointed
nostrils; females with horns. (Following Groves, 2000). out a relationship between specialization in horn form
The anterior basioccipital tuberosities tend to be small, and pattern reduction.
as in Gazella (A. W. Gentry, 1964). In the phylogenies of Rebholz & Harley (1999), N.
Lange (1971) argued that the pelage pattern of this soemmerringi and N. dama always formed a sister clade
genus (which he deemed a subgenus of Gazella) is spe- to N. granti. —-1
—0
349-47558_ch01_1P.indd 159 5/31/11 7:05 PM

160 Artiodactyla
1906 Gazella granti lacuum Neumann. Suksuk River, S of

Nanger granti Group Lake Zwai, Ethiopia.
Grant’s gazelle 1913 Gazella granti raineyi Heller. Isiolo, N Guaso Nyiro,
Furley (1986) recorded gestation length as 198–199 Kenya.
days; the fi rst fertile mating of a male at 450 days; and
Skins seen: 4/6.
the fi rst fertile mating of a female at 420– 450 days,
Flank band in the males usually somewhat darker,
but occasionally as early as 210–270 days.
in the females, very faintly darker. Instead of a broad
Grubb (2000a) argued that the species of this
pygal band extending all down the rump (as in N.
group form a ring-shaped morphocline, from N. pe-
granti), the rump band narrow, extending only half-
tersi on the Kenya coast via Sudan/Ethiopian N. no-
way down; color a shade darker.
tata (designated N. brighti in that paper) to N. granti
Grubb (1994) found that the type of Gazella granti
and, fi nally, N. robertsi. Lorenzen et al. (2008a) stud-
notata is somewhat different from the types and topo-
ied mtDNA in this group and found three recipro-
typical series of G. g. brighti and G. g. lacuum, and G. g.
cally monophyletic clades, differentiated by numer-
raineyi; it may represent “a unique localized popula-
ous substitutions.
tion, possibly now extinct.” For the moment, we retain
all these names in synonymy.
Nanger granti (Brooke, 1872)
All of the specimens we have seen are from N of
1872 Gazella granti Brooke. W Kinyenye, Ugogo, Tanzania. the Kenya highlands: Marsabit Road; N Guaso Nyiro;
1903 Gazella granti robertsi Thomas. Mwanza, Speke Gulf, Lorian Swamp; Ngare Ndare; Laikipia; Lake Baringo;
Lake Victoria, Tanzania. Rumuruti; Turkwell River. Looking at published
1913 Gazella granti roosevelti Heller. Kitanga Farm, Athi Plains. maps, in historical records there always seems to
have been a geographic discontinuity between N. no-
Skins seen: 12/8, as well as observations of numerous
tata and N. granti (Grubb, 1994).
living animals.
White rump-patch large, undivided; whole of the
Nanger petersi Günther, 1884
tail base white; lateral extension of the rump-patch
intruding largely into the body color, thus extending 1884 Gazella petersii Günther. Gelidja, near the mouth of the
beyond the dark pygal band. Flank band in the males Osi and Tana rivers.
colored like the dorsum, or somewhat darker, or dark 1887 Gazella granti var. geliadjiensis Noack. Gelidja.
but fading anteriorly, usually present in the females 1913 Gazella granti serengetae Heller. Taveta, SE Kenya.
(occasionally fading forward), but sometimes faint, or
Color darker than others. No dark lateral band in
even absent.
the adult males. White rump-patch relatively small,
In populations to the W of the Rift Valley (described
divided above by a broad band extending onto and
as a subspecies, G. g. robertsi), the horns are strongly
along the upper surface of the tail. Lateral prolonga-
outcurved and divergent, with the tips backwardly
tion of the rump-patch narrower, and intruding, to a
curved, in about 50% of the adult males, but as there
smaller extent, into the body color, thus scarcely, if at
is apparently no other difference, and this unusual
all, overhanging the dark pygal band. Horns short,
horn type does not characterize a preponderance
not very divergent, not backward-curved. Skull small,
of the population, the subspecies is not recognized
with a relatively narrow nasal opening.
here.
Kenya coast, extending N into southernmost So-
We have seen specimens from Lake Elmenteita,
malia. There have been noticeable range changes in
Gilgil, Lake Nakuru, Nairobi, Simba Station, Luke-
the recent past.
nya, Athi River, Sultan Hamud, Taveta, S Guaso Ny-
Gazella granti serengetae Heller, 1913 (type locality:
iro, Serengeti plains.
Taveta), was described as being dark cinnamon or
fawn on the back; in two males, the fawn color ex-
Nanger notata (Thomas, 1897)
tended backward as a narrow stripe through the mid-
1897 Gazella granti notata Thomas. W slope of the Loroghi dle of the white rump-patch and onto the upper sur-
Mountains, Kenya. face of the tail, of which only the basal third is white,
1901 Gazella granti brighti Thomas. 150 mi. E of Lado (5° 20' the remainder being black. The females lack this divi-
N, 34° 5' E), Sudan. sion. According to Roosevelt & Heller (1914), G. g.
-1—
0—
349-47558_ch01_1P.indd 160 5/31/11 7:05 PM

b o v idae 161
serengetae differs from N. petersi in the cinnamon streak

being narrow, and in having larger and less parallel Nanger soemmerringi (Cretzschmar, 1826)
horns. Actually, G. g. serengetae is clearly a synonym Soemmerring’s gazelle
of N. petersi— straight, parallel horns; reduced croup- 2n varied, from 2n = 34 to 2n = 39, for 7 male and 20
patch, with dark dorsal color bisecting the croup-patch female individuals in two US zoos (Benirschke et al.,
and joining the tail. Yet today, N. granti is the species 1984). The source of the animals was unclear; perina-
that occurs at Taveta, and, according to Lorenzen et al. tal mortality was very high, and whether the stock
(2008a), mtDNA from gazelles at the type locality re- was actually mixed—perhaps even with a member of
sembles what they call Gazella granti robertsi, (i.e., N. the N. granti group, as the authors inferred—was not
granti). Hence, it appears that there has been a shift in known.
distribution since the early 20th century, when G. g. Furley (1986) recorded gestation length as 195–210
serengetae was collected. days, and the age of a female at earliest fertile mating
Four males in Tervuren—from the Ziwani Swamp, as 540–570 days.
Tsavo— are variable. Three have a slight wedge back This (presumed) species is found throughout the
toward the tail, and that of the fourth reaches the tail lowlands of the Ethiopian region and into Eritrea,
base; this also applies to one from Pika Pika (3° 49' S, Somalia, and far E Sudan. We have little data on the
38° 52' E), and one from Ubwegetwe, Tanzania. These potential taxonomic diversity within the species, but
suggest a population with some gene flow from N. we note that, given the barriers formed by the Ethio-
granti. In addition, Lorenzen et al. (2008a) noted that pian highlands, the probability of significant differen-
one of 11 samples from Mkomazi, in NE Tanzania tiation is great.
near the Kenya border, had mtDNA not like N. granti
(as in the other 10), but like N. petersi, corroborating all Nanger dama Pallas, 1766
of the above indications of the latter’s former occur- dama gazelle
rence far to the SW of its present range.
1766 Antilope dama Pallas. Senegal.
The study of mitochondrial control region se-
As reviewed by Harper (1940), anywhere in the Lake Chad
quences by Arctander et al. (1996) found astonishingly
region cannot be the type locality, because in the time of
large degrees of separation between these three spe-
Buffon (whose descriptions Pallas used), Lake Chad was
cies, comparable to the distance between any of them
unknown.
and N. soemmerringi. There was a tendency for robertsi
samples to form a subcluster within N. granti. A pre- Gestation length is recorded as 174–202 days in N. d.
liminary report on this work appeared in the Antelope dama, 201 days in N. d. mhorr, and 210–225 days in
Specialist Group’s Gnusletter, and occasioned some N. d. rufi collis. The fi rst fertile mating of a male was
surprise, which was not, in fact, warranted (Grubb, recorded at 120 days; of a female, 610 days (Furley,
1994). 1986).
Lorenzen et al. (2008a), working on a larger num- The revision by Perez (1984), based on study of 50
ber of samples, likewise separated these three species museum skins and 150 living animals (including the
cleanly; this indicated to those authors that at least N. skins of some of the latter on their demise), divided
petersi should be regarded as a distinct species, differ- this species into three subspecies:
ing from the other two by an amount similar to that
1. Nanger dama mhorr (synonyms Gazella dama
differentiating them from N. soemmerringi. They drew
permista, Gazella dama lozanoi)— Ranging
attention to the opening up of Tsavo East National
from S Morocco to Rio de Oro, Senegal, and
Park during the late 19th and early 20th centuries by
the Niger River. Upper parts reddish brown,
elephant action, bringing into contact N. petersi and N.
reaching from the head to the middle of the
granti, previously separated by Acacia-Commiphora
croup, and down the hindlegs at least to the
woodland, as reported by Leuthold (1981), who docu-
hocks; the underside, the rest of the legs, and
mented a case of intermale combat between the two
the croup white (sending a V-shaped wedge
species. The range of N. petersi has continued to
forward into the haunch), separated from the
spread; it is now the predominant (or only) species of
reddish parts by a clear line of demarcation.
Nanger in Tsavo East National Park (Bruce Patterson,
The reddish line on the thigh is broadened,
pers. comm. to CPG), thus evidently reestablishing
restricting the white posteriorly to the
itself in much of its pre-20th-century range. —-1
—0
349-47558_ch01_1P.indd 161 5/31/11 7:05 PM

162 Artiodactyla
croup-wedge, and anteriorly to the anterior stripe on the limbs is only on the lower
edge of the stifle. The reddish line on the segments, if at all.
upper segment of the forelegs may or may not
Our data, collected over the years, would tend more
be connected with that of the body color.
to support the arrangement of Perez (1984). There is a
There are a darker and a lighter morph, the
decided difference between those W of Lake Chad
latter being the one described as N. d. lozanoi.
and those to the E, with an admittedly wide area of
2. Nanger dama dama (synonym Gazella dama
intermediacy within the Republic of Chad. But the
damergouensis)—Ranging from about 7° E to
boundaries in the W between N. d. dama and N. d.
Chad. Distinguished from N. d. mhorr by the
mhorr is incorrect. The specimens we have seen
absence of the thick reddish thigh-block;
from Senegal are indeed like Buffon’s plate—which
instead, a thin reddish line runs back horizon-
constitutes the type of N. d. dama— and speci-
tally from the lower part of the body-color
mens corresponding to Gazella dama permista also
block, then turns downward onto the upper
occur in W Africa, and even in the Chad zone of in-
thigh. Perez argued that Buffon’s “Nanguer,”
termediacy; the dark N. d. mhorr and intermixed Ga-
on which the name Nanger dama was founded,
zella dama lozanoi types do not occur farther S than
could not have been from Senegal, as his
the W Sahara. Our arrangement, therefore, is listed
illustration does not correspond with speci-
below.
mens known to have come from Senegal. Yet,
as noted above, a more E place of origin is very
Nanger dama dama (Pallas, 1766)
unlikely.
3. Nanger dama rufi collis— E of about 15° E. 1766 Antilope dama Pallas. Probably Senegal (Harper, 1940).
The reddish tones are mostly restricted to 1833 Antilope nanguer Bennett. Senegal.
the neck, sometimes nearly reaching to the 1847 Antilope dama var. occidentalis Sundevall. Senegal and
croup, but always restricted to the upper part Morocco.
of the body, and the line of demarcation with 1906 Gazella dama permista Neumann. Senegal.
the (often red-toned) zone on the underparts 1907 Gazella mhorr reducta K. Heller. Locality unknown.
is diagonal. 1921 Gazella dama damergouensis Rothschild. Takoukout,
Damergou, Niger River.
The following year, a very similar arrangement
1935 Gazella dama weidholtzi Zimara. Hombori, Mali.
was produced by Drüwa (1985), who likewise argued
that the name N. d. dama could not refer to the Sene-
Nanger dama mhorr (Bennett, 1833)
gal taxon, and saw N. d. mhorr as extending E as far as
the Niger River. Uniquely, however, he synonymized 1833 Antilope mhorr Bennett. Wednun, near Tafi lat, Mogador,
N. d. ruficollis with N. d. dama, thus reducing the sub- Morocco.
species to two: 1934 Gazella dama lozanoi Morales Agacino. Cape Juby, Rio
de Oro, W Sahara.
1. Nanger dama mhorr—Mountain- and grass-
steppes of W Africa. Dark, with a marked
Nanger dama ruficollis (Hamilton Smith, 1827)
brown tone to the colored areas; broad, brown
stripes extend from the flank zone down the 1827 Antilope ruficollis Hamilton Smith. Dongola, Sudan.
legs. There is variation in the size of the 1833 Antilope addra Bennett. Nubia and Kordofan.
rump-patch, the face pattern, the extension of 1847 Antilope dama var. orientalis Sundevall. Sennar.
the brown tone onto the belly, and of the flank
stripes down the legs. Synonyms would Gazella de Blainville, 1816
include not only Gazella dama lozanoi but also smaller gazelles
G. d. permista and G. d. damergouensis. Differs from Antilope as follows: horn rings usually
2. Nanger dama dama— Dry grass and desert somewhat angulated in front, and usually better ex-
zones of the C and E Sahel. The brown body pressed on the front surfaces of the horns; horn cores
color is reduced, to a greater or lesser extent, of the males slightly more compressed; molar styles
to a saddle; the behind parts and the sides of less developed; anterior lower premolar less reduced;
the body are contrastingly white. A clear preorbital fossa deeper; lacrimal bone extending well
-1— flank stripe can extend as a rudiment, from into preorbital fossa; ethmoid fissure shorter; basioccipi-
0— the saddle to the haunch, but the brown tal longer and narrower, anterior tuberosities smaller,
349-47558_ch01_1P.indd 162 5/31/11 7:05 PM

b o v idae 163
more triangular, the posterior ones less bulky, the lon- cial skeleton somewhat elongated. Valvular preorbital
gitudinal ridges between them more expressed; horns glands hypertrophied.
of the males more upright; preorbital foramen farther
forward; preorbital glands less hypertrophied. (Fol- Gazella leptoceros (F. Cuvier, 1842)
lowing Groves, 2000). slender-horned gazelle
Groves (1969a) and Lange (1972) placed G. pelzelni, Lange (1972) made G. leptoceros conspecific with G. sub-
G. saudiya and G. bennetti in G. dorcas, but Lange (172) gutturosa. Interestingly, in R. Hammond et al.’s (2001)
transferred G. cuvieri from G. rufifrons (here, Eudorcas phylogeny, G. marica assorted with G. leptoceros, rather
rufifrons), whereas Groves (1969a,) had placed it, in than with G. subgutturosa.
a moment of mental aberration, as a subspecies of Gestation length is 156–169 days, and the age of a
G. gazella, thereby returning to the old habitat- female at fi rst fertile mating can be as early as 152
dependent model of Ellerman & Morrison Scott days (Furley, 1986).
(1951). Lange (1972) also made G. leptoceros conspe-
cific with G. subgutturosa. Gazella cuvieri Ogilby 1841
Vassart, Granjon, et al. (1994) calculated an un- Cuvier’s gazelle
rooted cladogram on the basis of allele frequencies of 2n = 32 (females) and 33 (males), with an X-autosome
16 genetic loci. G. dorcas appeared very different from translocation like most gazelles; the karyotype
members of the G. gazella group. The use of allele strongly resembles G. leptoceros, with a slight differ-
frequencies may not, however, be a satisfactory way ence in the banding pattern on the short arm of the X
of reconstructing phylogeny above the species level (Kumamoto & Bogart, 1984).
(see, for example, J. P. Grobler & van der Bank, Gestation length is 165 days; the age of a male at
1993). first fertile mating is given as 180 days, or alternatively
Rebholz & Harley (1999) found that in their at 360–540 days; the age of a female at first fertile mat-
neighbor-joining tree, there were three clades in this ing, 422 days; the frequency of twinning is about 50%
genus: the G. bennetti group, the G. subgutturosa (Furley, 1986).
group, and a G. dorcas / G. spekei / G. gazella group. R.
Hammond et al.’s (2001) phylogeny of the major taxa Gazella subgutturosa and Related Taxa
of the genus was as follows: goitered gazelles
Wurster (1972) reported 2n = 30 (females) and 2n = 31
((subgutturosa (bennetti, leptoceros/marica)) (gazella
(males) in “Persian gazelles,” Kingswood and Kuma-
(saudiya, dorcas/pelzelni))).
moto (1988) found the same for five males and four
The present and former distribution of gazelles in females specified as G. subgutturosa, and Shi (1987) re-
the Middle East was discussed by Uerpmann (1987), ported the same for what was identified as G. subgut-
with notes on their taxonomy. turosa from China.
Kingswood & Kumamoto (1988) also karyotyped
Gazella subgutturosa Group 38 males and 33 females of what were identified as
Females often hornless. Rings on the horns of the Gazella subgutturosa marica, fi nding 2n = 31, 32, and
males somewhat less close-set than in the G. dorcas 33 (males), and 2n = 30, 31, and 32 (females). These
group, but more than in the G. bennetti group, and authors argued that the stock may ultimately have
equivalent to the G. gazella group, as well as less an- been founded by a mixed stock of G. marica and G.
gulated in front than either of the other two groups subgutturosa, and they noted, in addition, that the
and more equal in prominence from back to front, presumed G. marica were also less fertile than cap-
with even poorer longitudinal ridges between them; tive stocks of G. subgutturosa. Granjon et al. (1991)
horn cores more nearly circular in cross section, with karyotyped 11 males and 19 females of G. marica
a deep groove running up the posterior surface. Up- from a captive stock in Saudi Arabia, fi nding 2n = 33
per molars with more convex buccal walls. Preorbital in all males, but 2n = 32 and 2n = 31 (in 4 out of 20) in
fossa very deeply excavated. Supraorbital foramina the females; they considered this to be support
2–3, in deeper pits than the G. dorcas or the G. bennetti for Kingswood & Kumamoto’s (1988) hypothesis, al-
groups. Nasofrontal suture transverse, or slightly though the presence of the four 2n = 31 females seems
W-shaped. Lacrimal bone with a fenestra. Median na- to indicate that, even though the hypothesis may be
sal tips shorter than the lateral prongs. Anterior basi- correct, there may have been a little mixture here, —-1
occipital tuberosities smaller than other Gazella. Fa- as well. —0
349-47558_ch01_1P.indd 163 5/31/11 7:05 PM

164 Artiodactyla
This species complex has been fully described by closest to Turkmenistan, although only the last of
Kingswood & Blank (1996). It has a characteristic en- these is actually within the Turkmen range.
largement of the larynx, which is more prominent in The Turkmenistan group is differentiated from
the males; horns of the males are close together at the the Gobi sample by the thick horn bases and the long
bases, long and lyrate, but (in most taxa) usually ab- braincase, contrasting with the narrow posterior na-
sent in the females. Facial markings and the flank sals and the relatively short skull. The Yarkand group
band fade with maturity; adults of most taxa have a tends to be long-skulled but relatively narrow, with a
face that is essentially white. In the skull, the palate relatively short braincase, but the differences from
and the biorbital breadth tend to be greater than for the Gobi sample are not consistent, and there is no
other taxa; the lacrimal pits are very deep. evidence that they are not conspecific.
In our craniometric comparisons, in the SW mem- On horn measurements, we get much the same pic-
bers of this group there is a clear distinction between ture. A specimen from Zaisan Nor falls in the range of
those from Iran and the others, but the samples from the Turkmenistan group.
Arabia, on the one hand, and from Iraq and Khuzestan, Accordingly, the arrangement of the taxa of this
on the other, though they differ in color, do not differ group is given below.
strongly craniometrically. The difference is largely in
size; while palate breadth, braincase breadth, and Gazella marica Thomas, 1897
braincase length are hardly different, greatest skull Arabian sand gazelle
breadth and breadth across the base of the horns ac- Smallest species, but comparatively robustly built;
count for the size difference (skull length was not in- whitish or pale yellow-brown in color, with only ru-
cluded in the analysis, because of the high number of dimentary face and body markings.
skulls with broken premaxillae). From the Arabian desert, apparently preferably
In the N-C samples, there is (quite unexpectedly) a living in sandy areas.
clear distinction between Iran and Turkmenistan.
We have two available “Caucasian” specimens, but Gazella subgutturosa (Güldenstaedt, 1780)
they are actually from different areas: one, inside the Persian gazelle
Ira nian range of variation, is from Baku in Azerbai- Reddish or grayish sandy above, white below and on
jan, just NW of the Ira nian border, whereas the other, the buttocks. Young animals with a white stripe from
from Tbilisi (i.e., the topotypical area for G. subgut- above the eye to the nostril, and a narrow black line
turosa), seems rather different, distinguished from the below, but this nearly or entirely disappears in the
Ira nian sample by its long, relatively narrow brain- adults. Females usually (but not invariably) hornless.
case and broad palate. Those in the Turkmenistan Kumerloeve (1969) illustrated Turkish examples of
group have very high values for the basal breadth of this species and gave a map of the distribution of ga-
the horns, contrasting with the relatively narrow pal- zelles in SE Turkey, although probably not all of these
ate and the short braincase. These fi ndings are unex- refer to G. subgutturosa. Karami et al. (2002) have de-
pected, considering that previously, the Turkmeni- tailed the distribution of this species in Iran.
stan and Tbilisi gazelles had been regarded as The gazelles of the Caucasus may be different from
consubspecific with those from Iran. those coming from the main Iranian plateau; they
In our four horn variables, we again fi nd a clear dif- sometimes average larger, with relatively short horns,
ference in Iran versus Turkmenistan. The only Cauca- and, according to Heptner et al. (1961), are darker, a
sian specimen (Tbilisi) this time assorts with Iran. brownish gray, with a clear, dark flank stripe and a
The Turkmenistan gazelles have a wide tip-to-tip in- heavy tail-tuft.
terval, contrasting with a relatively narrow span.
When the comparisons are moved farther to the Gazella gracilicornis Stroganov, 1956
NE, skulls from the Gobi are quite separate from Sandy brown to sandy gray in summer; lighter in
those from Turkmenistan, and those from Yarkand winter. Very wide across the bases of the horns; rela-
(in Xinjiang) overlap with the Gobi sample (one speci- tively narrow palate; short braincase.
men, from the Cherchen River, W of Kunlun, ap- Deserts of Kazakhstan, Turkmenistan, and
proaches Turkmenistan, but does not overlap with it). Uzbekistan.
Individual skulls from Caidam, Semiretch, and Har G. gracilicornis was regarded as a synonym of G.
-1— Nor (38.20° N, 97.30° E, on the borders of Caidam) are subgutturosa by Groves (1969a), but the new cranio-
0—
349-47558_ch01_1P.indd 164 5/31/11 7:05 PM

b o v idae 165
metric information reported here indicates that it is lying at the same level as the foramen ovale (A. W.
absolutely distinct. Gentry, 1964).
Alados (1987) took 53 mea surements of the skulls
Gazella yarkandensis Blanford, 1875 and the jaws of 10 OTUs of the G. dorcas group, plus
G. saudiya (group S) and “G. gazella arabica” (group
1875 Gazella subgutturosa var. yarkandensis Blanford.
Ar; actually a mixture of G. cora and G. erlangeri).
Yarkand, Xinjiang.
The OTUs within G. dorcas were as follows: P I (G.
1894 Gazella hillieriana and Gazella mongolica Heude.
pelzelni from Berbera), P II (G. pelzelni from Dan-
Mongolia.
akil), I (G. isabella), L (“G. littoralis”), N (“G. neglecta”),
? 1900 Gazella subgutturosa sairensis Lydekker. Saiar
D I (G. dorcas from Darfur and Kordofan), D II (G.
Mountains, Xinjiang.
dorcas “from Hoggar and nearby northern areas”);
? 1931 Gazella subgutturosa reginae Adlerberg. W Caidam.
and A I (G. dorcas from Rio de Oro). Unfortunately,
Color much the same as in G. gracilicornis. Relatively the samples were far from clear cut: the samples
narrow across the horn bases; long skull; short brain- listed as G. isabella and “G. littoralis” were nearly co-
case; very broad posterior nasals. terminous in their geographic scatter; the G. neglecta
We have been shown photographs from the wild sample was specifically from Hoggar, but so were
of gazelles in the Saiar Mountains, and they are con- some in the D II sample; and the D II sample also
spicuous by their strong coloration and strong, dark included specimens from the Maghreb, Egypt, and
body markings; we think it improbable that the name Sudan. The measurements were subjected to a sys-
Gazella subgutturosa sairensis is, in fact, a synonym of tem of gap-coding and treated cladistically. In the
this species. Our material from Caidam, however, is overall cladogram, groups I and L assorted together
too meager to allow us to make any fi rm decision on with Ar, with N as their sister group; on the other
the status of Gazella subgutturosa reginae. branch, successively, were P I, P II, D II, A I, and S. If
S was taken as “ancestral,” the branching order was
Gazella Species (Undescribed) A I, P I, D II, P II, N, L, I, and Ar. If Ar was taken as
Gazelles in Iraq, in Anatolia, and in Iran (on the W side “ancestral,” the branching order was I, L, then a
of the Zagros) are much smaller than those of the Ira- three-way split of N, PII plus PI, and DII plus a group-
nian plateau, with their horns having a narrow span and ing of S plus A I.
tips that are very close together, and stronger markings It was most unfortunate that this potentially inter-
(flank-stripes and face-stripes) on their bodies. esting analysis was marred by the intermixing of
samples. In this light, it is unsurprising that G. isabella
Gazella dorcas Group and “G. littoralis” always came out close together—
Females with comparatively long, but thin, horns. because they are, in essence, the same sample. More
Horn rings less prominent than for other gazelles, but significant is the fi nding that they cluster close to the
somewhat expressed in the females. Horn cores of the cora/erlangeri combination; this rapprochement has
males more or less circular in cross section (the ratio of been noticed previously. Also significant is the fi nd-
the transverse to the anteroposterior diameter of the ing that the two G. pelzelni samples always came out
pedicels is given by A. W. Gentry [1964] as 58– 66 mm, together, and were not close to G. isabella (pace
with a mean of 61.7 mm); no anterior groove on the Groves, 1981a). The fairly consistent association of A I
core, but one running up the posterior surface (these (G. dorcas from Rio de Oro) with G. saudiya is puz-
distinctions, first made by Ducos [1968], were con- zling, and perhaps reflects convergence in their ex-
fi rmed by Tchernov et al. [1987], and they have also treme desert adaptations.
been noted by one of us [CPG]). Horns of the adult fe-
males usually long, well formed, and slightly ringed. Gazella dorcas (Linnaeus, 1758)
Ascending branch of the premaxilla making a sub- dorcas gazelle
stantial contact with the nasals. Nasals tending to Gestation is given as 164–174 days; the earliest age for
broaden from the posterior to the anterior ends. Pos- a male at fi rst fertile mating, 589 days; the age of a fe-
teriorly, the nasals together forming a V, penetrating male at fi rst fertile mating, anything from 192 to 599
into the frontals. days; pregnancy is rare (Furley, 1986).
Anterior basioccipital tuberosities tending to be Tchernov et al. (1987) found that the presence of
small, but larger than the ridges behind them, and G. dorcas in the Sinai and Israel dates from after the —-1
—0
349-47558_ch01_1P.indd 165 5/31/11 7:05 PM

166 Artiodactyla
Pre-Pottery Neolithic B period, when, with increas- Gazella saudiya Carruthers & Schwarz, 1935
ing aridification, G. dorcas largely replaced the G. ga- Saudi gazelle
zella group in that region. A small species, with poorly expressed patterning
Groves (1981c) recognized five subspecies, apart and long horns—nearly straight, barely turned in at
from G. pelzelni. These will be maintained here, but the tips—in both sexes.
they need restudy, given that different populations G. saudiya is securely known from only two re-
within each can be almost as different from each gions: the type locality (Dhalm) and nearby localities
other as the “central” populations of each subspecies. (Alam Abyadh, Wadi Markha, Arq Abu Da’ir, Ru-
waik, Taraf Al Ain, Wadi Naq’a, Sirr al Yamani), all
Gazella dorcas dorcas (Linnaeus, 1758) roughly 150 mi. NE of Mecca, collected during the
Rich fawn; flank band well expressed, pygal band 1930s; and Kuwait, collected in 1934. The Kuwait skin
clear; dark lateral face-stripe tending to be suff used is darker, grayer than the type series from Dhalm,
with black. Size very small; skull broad, with a short though the skulls are alike.
rostrum. Horns long in both sexes, lyrate, with nu- It seems unlikely that either population still exists.
merous close rings, in the males. Nonetheless, there are two captive breeding groups
W desert of Egypt, as far W as Tunisia. that have been identified as being of this species: one
in Al Ain Zoo (Abu Dhabi, United Arab Emirates),
Gazella dorcas osiris Blaine, 1913 the other in Al-Areen (Bahrain). One of us (CPG) saw
Color very pale fawn; dark flank band not well both of these groups in the early 1990s and he thought
marked, but bordered above with a very pale band; that they were both G. saudiya, though there were
pygal stripe poorly marked; face markings paler, with differences between the two (the Al-Areen gazelles
no nose-spot; light face-stripes pure white; ears very are larger and more disruptively colored, for exam-
pale; little or no darkening on the knee tufts. Skull ple). Stock from both of these collections have since
small; horns long and narrowly ringed in the males, been shared with the King Khalid Wildlife Research
relatively long in the females. Centre (KKWRC), Saudi Arabia. A third captive
Saharan form, from W of the Nile to the Atlantic. group, in Al Wabra, Qatar, has also been identified as
“Saudi dorcas,” although in this case, CPG identified
Gazella dorcas massaesyla Cabrera, 1928 them as G. dorcas isabella and as having probably not
Richer, more ochery color; nose-spot poorly ex- been obtained in Saudi Arabia at all.
pressed in the adults; skull rather narrow across the Rebholz et al. (1991) karyotyped two males and
horn bases; horn tips less inturned. three females from the Al-Areen collection at the
Apparently the Maghreb: coastal ranges N of the KKWRC. 2n = 47 in the females, and 2n = 51 and 52 in
Atlas Mountains, along the Mediterranean. the two males. There was conspicuous variation in
chromosomal morphology. In at least one of the
Gazella dorcas isabella Gray, 1846 males, two Y chromosomes were present, indicating
Relatively large; dark brown-gray, with reddish tones; the presence of the usual X-autosome translocation.
dark markings, including the nose-spot, well ex- The authors noted that the karyotypes appear to be
pressed. Horns of the males relatively short and stout, very similar to those of the G. bennetti–group gazelles
with the tips strongly turned inward; horns of the fe- illustrated by Furley et al. (1988).
males less elongated than in taxa from W of the Nile. Kumamoto et al. (1995) karyotyped five female
N Eritrea; Sudan, E of the Nile; up the Red Sea and two male specimens, also from Al-Areen and the
coast to the Sinai, and into the deserts of S Israel. KKWRC. 2n = 46, 48, 49 or 50 in the females, and
2n = 49 and 53 in the two males; hey had the usual
Gazella dorcas beccarii de Beaux, 1931 X-autosome translocation. The authors considered it
Richer and more chestnut than G. d. isabella; flank likely that some hybridization with G. bennetti might
band darker; extremely large in size. have occurred in this stock. They noted, interestingly,
This is an enigmatic taxon, known from only one that in two cases the karyotypes were identical with
skull and two skins from the Upper Baraka district in those of two G. bennetti–group gazelles from Iran.
Eritrea, approximately 15° 30' N, 38° E. G. d. beccarii R. Hammond et al. (2001) managed to extract sam-
needs further examination, preferably including field ples of the mitochondrial cytochrome b gene from
-1— observations. museum specimens known or presumed to be of
0—
349-47558_ch01_1P.indd 166 5/31/11 7:05 PM

b o v idae 167
this species, and to compare them with the samples deeper pit for the supraorbital foramina; the more
from the “Saudi gazelles” in the Arabian collections arch-shaped nasofrontal suture; the longer median na-
(Al Wabra, Al Areen, Al Ain Zoo, and the KKWRC). sal tips, usually longer than the lateral prongs; and the
The authors also sampled other taxa, both from the less high and less rounded cranial profi le.
KKWRC and from museum material. The results were Even under the Biological Species Concept, it came
surprising. The captive breeding-groups identified by to be realized that this species-group cannot be united
earlier authors as “saudiya” proved to be quite different in a single species. Groves (1985) wrote of “darker,
from the genuine (i.e., the museum) saudiya—the for- straighter-horned individuals” of what he called G. g.
mer had sequences typical of either G. subgutturosa cora predominating along the Red Sea coast, and later
(three out of the four Al Areen specimens) or of “G. ben- Groves (1996a) separated this population as a full spe-
netti” (actually, G. fuscifrons in our present taxonomic cies (G. erlangeri), without yet associating the small,
scheme: the fourth Al Areen animal, plus all three from dark Thumamah gazelle with it (see below).
Al Ain). A specimen in the Harrison Museum, identi- In R. Hammond et al.’s (2001) cytochrome b tree,
fied as G. saudiya, proved to be G. marica. Meanwhile, members of the G. gazella group (museum specimens
the three Al Wabra specimens—which had been infor- of G. muscatensis, plus a living KKWRC specimen of
mally ascribed to G. saudiya, simply because they were uncertain affi nities) formed a sister clade to G. dor-
rumored to come from Saudi Arabia, not because they cas / G. pelzelni plus G. saudiya.
resembled G. saudiya phenotypically (see above)— Table 41 gives the univariate statistics for males of
proved to be G. dorcas, exactly as they always appeared this species-group.
to be.
The real G. saudiya turned out, after all, to be re- Gazella gazella (Pallas, 1766)
lated to G. dorcas. And, alas, they are totally extinct, mountain gazelle
unless some of the captive stock contain their genes 2n = 34 (females), 35 (males) (Greth et al., 1993).
(recall that only mtDNA was sequenced). Horn cores of the males elliptical in cross section;
broad groove running up the anterior surface of the
Gazella pelzelni Kohl, 1886 core, frequently with a second groove, less marked,
Pelzeln’s gazelle just medial to it, and with another groove running
Very large in size for the G. dorcas group; horns long, up the posterior surface (Ducos, 1968; Tchernov,
with the tips hardly (or not at all) turned in; horns not 1987). Horns of the females short (84–153 mm), nearly
widely flared, and sparsely ringed; color bright sandy smooth, and fragile, easily broken in life. Ascending
ochre, with a well-marked, dark lateral flank band branch of the premaxilla only just contacting the na-
and dark face-stripes; dark knee-tufts; always at least sals, or not at all. Nasals usually broader posteriorly,
a trace of a dark nose-spot. and not penetrating deeply between the frontals.
N Somalia and the Danakil country of Ethiopia Coronal suture double-bowed. Light facial stripes
and S Eritrea (see below). off-white, edged below with dark or black; midface
Groves (1981c) included this species in G. dorcas, dark, usually with a blackish nose-spot. Body gray-
on the grounds that specimens from the Danakil brown, with a thick, dark stripe separating this color
country are intermediate between it and “true” G. from the white underparts, and a paler stripe above
dorcas, but G. pelzelni is, in fact, absolutely different the dark one. Premolar length 37.5%– 41% of the
from any of them, and the Danakil specimens may whole toothrow; lingual styles and valleys on the
actually be a hybrid population. cheekteeth angular; talonid of the third lower molar
Living animals, seen in Al Wabra, seem noticeably short.
more slender and long-legged than G. dorcas. A relatively stockily built gazelle; weight (in kg),
Two museum specimens from Danakil, whose expressed as a percentage of the head plus body
partial cytochrome b sequences were analyzed, fell length (in cm), averaging 23.6% in nine specimens.
into the G. dorcas clade (R. Hammond et al., 2001). Limbs relatively short, with short distal segments: tibia
averaging 126.1% of the femur; metatarsal, 101.1% of
Gazella gazella Group the femur; hindleg, 301.9% of the skull length (N = 21
Differs from the G. dorcas group in the more weakly or 22). Ear relatively short, its length averaging 11.7%
developed horns of the females; the more prominent, of the head plus body length (N = 15), and characteris-
less close-set horn rings of the males; the (usually) tically held upright, more or less parallel with the —-1
—0
349-47558_ch01_1P.indd 167 5/31/11 7:05 PM

168 Artiodactyla
Table 41 Univariate statistics for desert members of the Gazella gazella group: males (measurements for
G. farasani are from Thouless & al Bassri 1991)
Base br Nas ant Nas post Nas l Gt br Pal br Gt l Braincase Horns Tip–tip Span
acaciae
Mean 66.33 19.33 25.67 54.40 80.40 48.00 195.60 58.33 236.80 111.00 130.20
N 3 3 3 5 5 3 5 3 5 3 5
Std dev 3.055 1.528 3.512 2.966 1.673 4.359 3.507 2.082 30.343 10.583 5.718
Min 63 18 22 50 79 45 192 56 208 99 123
Max 69 21 29 58 83 53 200 60 277 119 135
cora
Mean 64.70 20.13 22.39 46.39 81.04 45.46 184.57 59.75 217.93 89.15 133.41
N 27 23 23 23 27 26 21 24 27 27 27
Std dev 2.163 1.866 1.751 5.263 2.312 2.121 5.861 2.364 16.920 18.340 12.879
Min 61 16 19 38 77 39 173 55 189 56 115
Max 69 23 26 58 87 49 195 65 255 129 170
dareshurii
Mean 63.20 20.36 22.64 39.27 78.60 43.90 175.10 54.00 202.00 110.71 149.71
N 10 11 11 11 10 10 10 10 7 7 7
Std dev 3.615 .924 1.027 4.429 2.171 1.524 9.480 1.247 12.028 20.702 24.088
Min 55 19 21 33 75 42 152 52 188 89 117
Max 67 22 24 47 81 47 185 56 225 143 175
erlangeri
Mean 60.00 21.80 23.60 42.60 81.20 48.60 176.60 55.80 212.60 109.80 120.60
N 5 5 5 5 5 5 5 5 5 5 5
Std dev 3.536 1.483 2.608 7.701 4.266 3.362 5.320 4.147 22.300 22.797 16.041
Min 55 20 22 35 75 45 169 52 177 85 96
Max 64 24 28 54 87 53 184 62 232 137 135
farasani
Mean 57.8 — — — — 46.5 165.6 51.0 193.2 81.2 86.0
N 12 — — — — 4 3 7 6 6 6
Std dev 2.8 — — — — 2.1 6.7 1.3 22.5 9.1 4.1
karamii
Mean 58.00 21.00 22.00 41.00 84.00 49.00 173.00 58.00 228.00 190.00 210.00
N 1 1 1 1 1 1 1 1 1 1 1
muscatensis
Mean 62.00 20.25 23.25 36.25 79.33 44.67 172.00 57.67 188.40 86.40 132.40
N 6 4 4 4 6 6 1 6 5 5 5
horns (all of these figures from Mendelssohn et al., The three desert taxa are distinct from each other
1997). on multivariate analysis, and univariate statistics are
Details of the biology of this species have been given in table 41.
given by Mendelssohn et al. (1995). G. erlangeri is smaller and relatively broader than
G. gazella has flourished under protection in N Is- G. acaciae and G. cora, its nasals narrower anteriorly;
rael and the Golan Heights. Its survival in Lebanon G. acaciae is larger and relatively narrower, its nasals
and Syria is unknown, but Dr. Tolga Kankiliç has wider posteriorly and narrower anteriorly. (Note, of
kindly forwarded photos (pers. comm. to CPG) of a course, that the sample size of G. acaciae is small).
thriving population of G. gazella at Hatay, in Turkey G. erlangeri has a wider tip-to-tip interval and a
near the Syrian border, just inland from the Mediter- narrower span (i.e., the horns are straighter and less
ranean coast. divergent), and the horns are narrower across the
bases than the other two; G. acaciae averages wider
Other Species of the Gazella gazella Group across both the tips and the bases.
-1— Notes on the distribution and present status of the Ara- The three Persian Gulf taxa can now be brought
0— bian peninsular species are given by Greth et al. (1993). into the comparisons. Of the skulls in the sample of
349-47558_ch01_1P.indd 168 5/31/11 7:05 PM

b o v idae 169
G. muscatensis, most are incomplete (primarily lack- pared with span, and especially much wider across
ing the premaxillae). G. muscatensis and G. karamii have the horn cores—than in any member of the G. subgut-
wide skulls and short nasals, and are perhaps slightly turosa group.
narrower across the horn bases; G. dareshurii is the op- Dollman (1927) described Gazella arabica hanishi
posite, with G. erlangeri in between. G. erlangeri and G. from Great Hanish Island, in the S Red Sea, describ-
karamii have nasals that are wide anteriorly, compared ing it as differing from “the typical Gazella arabica ar-
with posteriorly, and have narrow horn bases. abica” in its larger, more sharply defi ned nose-spot; at
As far as horn variables are concerned, G. musca- the base of horns, there “are two black spots, which
tensis and G. dareshurii have relatively short horns, extend down the face almost as far as the dark nasal
with a wide span, but they are narrow across the tips marking.” Gazella arabica hanishi is said to be less red-
(the horns are initially bowed outward, then con- dish, and the flank stripe is more developed, ap-
verge toward the tips) and have relatively wide bases; proaching G. erlangeri, but hanishi is less rufous and
G. dareshurii averages larger than G. muscatensis in all has a larger nose-spot. The type of G. a. hanishi was
horn dimensions. G. karamii is also large in all dimen- the mounted head of an adult male in the collection
sions, but it has a large tip-to-tip interval compared of the Duke of York: horns 9 in. long (= 229 mm), tip-
with the span, and long horns compared with their to-tip distance 4-1/8 in. (= 124 mm). The following
basal breadth. year, Dollman (1928) redescribed the specimen, and
In summary, the cranial and horn differences this time he illustrated it. A second specimen was
among these desert gazelles are as follows (mention- mentioned, but not pictured; this was said to have
ing mainly the extremes in each case): horns 6-3/4 in. (= 173 mm) long.
It is somewhat difficult to judge this described
—In size, acaciae > cora > others, but there are
taxon without seeing the type specimen, which is
overlaps between neighboring taxa.
not, at present, in the Natural History Museum in
—G. karamii is exceptionally narrow across the
London, although it may be in a private royal collec-
horn bases, compared with a wide braincase.
tion somewhere in Britain. It is clearly not G. cora,
—G. acaciae has nasals that are exceptionally
because the horns are nearly straight ( just somewhat
narrow anteriorly and wide posteriorly.
inturned at the tips) and the face is much too dark. G.
—G. dareshurii and G. muscatensis are alike in
arabica and G. bilkis are dark, but, at least in the latter,
having very short nasals.
the light face-stripes are much more infused with
—G. karamii and G. erlangeri both have a wide
ocher, and the midface is so dark that any nose-spot
palate.
cannot be detected. In the type (and only known
—G. muscatensis has comparatively short horns;
specimen) of G. arabica, the horns are 272 mm long;
G. karamii and G. acaciae have long horns.
their span is 102 mm; the tip-to-tip distance is 92 mm;
—G. muscatensis and G. cora have a very narrow
and they have 15 rings. In G. bilkis, the horns are 204–
tip-to-tip distance, especially compared with
254 mm long in the males; the greatest horn span is
the horn span; G. acaciae, and especially G.
86–102 mm; the tip-to-tip distance is 52– 86 mm; and
erlangeri, have a fairly narrow span, with less
there are 12–15 rings (N = 2). The horns in both are
inturning of the tips.
shorter and more robust than those of the specimen
Relationships among these species are not geocli- in Dollman’s photo; there are fewer and less promi-
nal. Probably G. cora and G. acaciae are more closely nent rings; and the tips do not turn in, as they do for
related to each other, while the other four seem to the specimen in the photo. The closest resemblance
form a second morphological cluster, though all are of the Hanish gazelle is with G. erlangeri, which has a
diagnosably distinct, and we recognize them all as similarly dark face, but with a large, clearly defi ned
full species here. nose-spot; long, slender horns (length 177–232 mm),
G. karamii has perhaps a superficial similarity to tending to turn in at the tips (span 96–135 mm, versus
the G. subgutturosa group in its skull, in that it has a tip-to-tip distance 85–137 mm); and, usually, fewer
rather wider skull, especially the palate, compared than 15 prominent rings.
with the G. gazella group, though the widths are not
as great as in the G. subgutturosa group. The horns Gazella acaciae Mendelssohn et al., 1997
G. karamii, however, are characteristically those of acacia gazelle
the G. gazella group; indeed, they are very like those A very slender, rangy gazelle; weight (in kg), expressed —-1
of G. dareshurii—much wider between the tips, com- as a percentage of the head plus body length (in cm), —0
349-47558_ch01_1P.indd 169 5/31/11 7:05 PM

170 Artiodactyla
only 16%–18% (N =3). Distal limb segments extremely less so in the paler ones. Weakly marked transition
elongated: tibia length 135.7% of the femur length, on the haunch from the darker fawn above to the
metatarsal length 114.9% of the femur length; hind- paler fawn below, continuing down the hindlegs. Lit-
leg length 306.8% of the skull length (N = 1). Neck tle or no pygal band. Forehead and nose varying from
long. Ear long, 11.4%–13.5% of the head plus body dark chestnut to merely a slightly darker fawn; nose-
length (N = 3), reaching to the nostrils when folded spot, if any, vague; forehead often sprinkled with
forward on the flat skin, and broad, in life held slant- whitish hairs; face-stripes clear, broad, white, bor-
ing sideways at a 45° angle. Tail long, bushy; body dered below with a dark brown line. Tail relatively
color dark earth-brown, lighter in summer than in long and thin. Ear, when folded forward, reaches
winter; pygal stripe black; white of the rump reach- halfway or completely to the nostrils.
ing relatively high up on either side of the tail; facial
markings strongly expressed, the dark stripes broad, Gazella erlangeri Neumann, 1906
the midface less dark than in G. gazella, the light face- Arabian coastal gazelle
stripes nearly white, with a very broad, conspicuous, 2n = 34 (females), 35 (males), as in G. cora and G. ga-
black nose-spot; white stripe on the inner surface of zella (Greth et al., 1993).
the hindleg reaching down to the hoof. (Following Overall color very dark gray-brown; legs a lighter,
Mendelssohn et al., 1997). more ochery brown, with a sharp transition between
Nasopremaxillary contact longer than usual in this and the darker upper-body color on the haunch;
this species-group. Premolars relatively long. Tym- face dark, with a large, clearly defi ned nose-spot,
panic bullae large. these being the only real differences in color from G.
Tchernov et al. (1987) could not examine horn cores muscatensis. Tail with a thick, black tuft. Ear long,
of this species, so these authors were unable to confirm reaching halfway between the eye and the nostril
that this was the species that inhabited the Sinai until when folded forward. Long, slender horns (length
the increasing aridification after the Pre-Pottery Neo- 177–232 mm), tending to turn in at the tips (span 96–
lithic B period. At any rate, after that time, a member 135 mm, versus tip-to-tip distance of 85–137 mm), and
of the G. gazella species-group was replaced by G. dor- usually fewer than 15 prominent rings. Females with
cas in the Sinai and Israel. relatively long horns. Very wide palate.
Found sporadically along the W Arabian coast,
Gazella cora Hamilton Smith, 1827 from Aden an unknown distance N along the Red
Arabian desert gazelle (sometimes known, inappropri- Sea coast.
ately, as “mountain gazelle”) The description of Gazella gazella farasani Thou-
less & al Bassri (1991), from Great Farasan Island and
Groves (1983b) showed that the name arabica, “traditionally”
the nearby island of Zifaf, is extraordinarily reminis-
used for this species, is actually not applicable. The
cent of G. erlangeri, and photos seen by CPG reinforce
earliest available name seems to be Antilope cora
this impression. The means and the standard devia-
Hamilton Smith, 1827, although, given the uncertainty,
tions given by Thouless & al Bassri (1991) have been
it is probably appropriate that in the near future a
added to table 41; they denote an animal clearly
neotype should be selected.
smaller, on average, than G. erlangeri, and—provided
2n = 34 (females), 35 (males) (Greth et al., 1993). always that this difference is not purely a result of
Limbs elongated— especially the hindlimb, which phenotypic plasticity—indicate a valid subspecies.
is 317.6% the length of the skull—but less elongated Vassart, Granjon, et al. (1994), however, found that G.
distally than in G. acaciae: tibia length 125% of the g. farasani clustered more closely with G. cora than with
femur length; metatarsal length 105% of the femur G. erlangeri on the basis of allele frequencies. In this
length (N = 1 ). These figures are not very different study and a subsequent one (Vassart, Granjon, et al.,
from G. gazella. Ear 119% of the head plus body length 1995), G. erlangeri was found to be fully monomor-
(N = 3), more or less like G. gazella, but held slanting phic (17 individuals); the same study found its karyo-
outward like G. acaciae. (Following Mendelssohn type identical to other species of the G. gazella group.
et al. 1997).
Color medium to very light fawn; flank-stripe Gazella muscatensis Brooke, 1874
varying from a thin black line to just a very slightly muscat gazelle
-1— darker zone, with a lighter zone above it on the Very small, like G. erlangeri; but the horns very slen-
0— flanks, which is more marked in the darker skins but der, bowed outward and then inward at the tips. Fe-
349-47558_ch01_1P.indd 170 5/31/11 7:05 PM

b o v idae 171
males with comparatively long horns, like G. erlangeri. tions between the dark and light zones on the flanks
Very dark gray-brown, often with an almost silvery and haunch; poorly marked on the flanks, with only a
sheen, resembling G. erlangeri very closely in color. slightly darkened pygal band; but a clear dark nose-
Flank-stripe in the form of a conspicuous darkening spot and dark forehead. The horns of the males are
along the flank, with a lighter, pinky gray zone above thick, straight, and well ringed, only slightly diver-
it. Pygal stripe a clear, dark rim to the white buttocks. gent; those of the females are straight, slender, and
Very sharp, oblique transition on the haunch between well formed. Evidently, the gazelles here are of the G.
the dark body color above it and the fawn-colored bennetti group, perhaps closest to G. shikarii, but ap-
haunch and hindleg. Forehead dark chestnut; nose- pearing to be more brightly colored.
stripes white, becoming pure white above and behind
each eye, bordered below by a smudgy dark stripe. Gazella bilkis Groves & Lay, 1985
Nose-spot smudgy, small, black. Ear-backs dull fawn; Bilkis gazelle
ear, when folded forward, reaching halfway between Size very large; horns relatively short, straight, and
the eye and the nostril. Tail with a long, thick, black relatively poorly ringed in the males, but very large
tuft. Nasal bones very short. and ringed in the females. Color very dark, sharply
Along the Batinah coast of Oman. paler on the flanks, the haunches, and the legs; flank-
stripe very thick, black, with a thin red line below;
Gazella dareshurii Karami & Groves, 1993 white on the inside of the thigh, continuing down the
Farrur gazelle shank. Horns 204–254 mm long in the males; greatest
Most closely related to G. muscatensis, but the horns of horn span 86–102 mm, tip-to-tip distance 52– 86 mm;
the males longer; horns of both sexes broad across the 12–15 rings (N = 2).
base; skull narrower; nasal bones posteriorly narrow. Greth (1992) recorded seeing a pair of horns in the
Ascending branch of the premaxilla apparently never National Museum in Taez, Yemen, which were nearly
reaching the nasals. Photographs of these animals in straight, “about 25–28 cm” long, with a very narrow
the wild show a pale, sandy brown animal. span and 14–15 rings. This seems most consistent
As far as is known, G. dareshurii is restricted to the with G. bilkis, but there was no evidence of the spe-
island of Farrur ( Jazireh-e-Forur), 26° 30' N, 54° 30' cies’ survival in Yemen.
E, in Iran. Karami & Groves (1993) discussed its pos-
sible existence on the mainland. Gazella arabica Lichtenstein, 1827
large Farasan gazelle
Gazella karamii Groves, 1993
Groves (1983b) showed that this name does not belong to
Bushehr gazelle
the “common” Arabian gazelle, but to a gazelle
Distinguished by its small size; very narrow across
formerly(?) existing on the Farasan Islands.
the bases of the horns; palate very broad. The skin of
the type, found during a visit by CPG to Berlin in Large, very dark, with a red line between the dark
October 2009, is as follows: color very dark, resem- flank band and the white of the underside, and long,
bling G. muscatensis, and with the same well-tufted very straight horns. In the type (and only known
tail; face-stripes off-white (more strikingly white specimen), the horns 272 mm long; the span 102 mm;
above the eye), bordered below by a thin black band; the tip-to-tip distance 92 mm; 15 rings.
nose-spot small, dark, smudgy; forehead sprinkled Farasan Islands; possibly extinct.
with white hairs.
The type, a zoo specimen that died in 1928, was Gazella spekei Blyth, 1863
caught E of Borazjan, near Bushehr, on the Persian Speke’s gazelle
Gulf. CPG, during his earlier visit to Berlin in the In the phylogeny of Eff ron et al. (1976), G. spekei
1980s, when he fi rst studied the skull, must have been is placed as sister to the G. dorcas plus G. gazella
having an off day (or a hangover), because he later clade; 2n = 33 (males), 32 (females). We provisionally
described it as a subspecies of G. bennetti! (and for convenience) place it here in the G. gazella
Photos taken in the wild by Dr. M. R. Hemami group.
(copies kindly sent to CPG) of gazelles in the Nay- Horn pedicels much more compressed than in the
band Protected Area, 27° 22' N, 52° 38' E, about G. dorcas group; ratio of the transverse to the antero-
250 km S by SE of Bushehr, show a brightly colored, posterior diameter averaging 61.7 mm, with a range of —-1
reddish sandy gazelle, with poorly expressed transi- 58– 66 mm (A. W. Gentry, 1964). Anterior basioccipital —0
349-47558_ch01_1P.indd 171 5/31/11 7:05 PM

172 Art iodactyla
tuberosities tending to resemble those of the G. ben- When we look at horn length versus skull length for
nettii group in size and position. the males, it is clear that G. fuscifrons has relatively
Gestation is 169–190 days (mean 178.6 days); the short horns compared with its general skull size,
age of the male at fi rst fertile mating, 480 days; of the whereas the opposite is true for others, especially G.
female, 504 days, but as early as 240 days (Furley, salinarum.
1986).
Gazella shikarii Groves, 1993
Gazella bennetti Group western jebeer
chinkara Color light, resembling G. subgutturosa, with which it
2n =50 (females), 51 (males). Furley et al. (1988) show is sympatric in W and N Iran. In fact, presumably for
that the chromosomes of the G. bennetti group (based this reason, this species had long been misidentified
on specimens of unknown origin, but very likely from as G. subgutturosa (Groves, 1993a ). Horns strikingly
Pakistan) are quite different from those of the G. dor- long in the females; more divergent in both sexes
cas, the G. gazella, and the G. subgutturosa groups. than in other species of the G. bennetti group.
In a study by Kumamoto et al. (1995), three The Ira nian plateau, from the Tehran region E to
females from Pakistan had 2n = 51, and a further the Touran Biosphere Reserve (35° 39' E, 56° 42' E)
female (of unknown origin) had 2n = 50, while a and S to Deh Kuh (27° 53' N, 54° 22' E). Animals from
male had 2n = 52; a male from Iran had 2n = 49, and a Semnan and Kerman provinces, where the range ap-
female had 2n = 52. The Pakistani series was poly- proaches that of G. fuscifrons, are entirely typical, and
morphic for a Robertsonian translocation between do not approach the latter in appearance (Karami
chromosomes 8 and 14; the Ira nian sample was poly- et al., 2002).
morphic for three independent Robertsonian translo-
cations. Chromosomes 22 and 25 were metacentric in Gazella fuscifrons Blanford, 1873
the Ira nian specimens and acrocentric in the Paki- eastern jebeer
stani ones. Dark in color; prominent flank band and face-stripes;
The G. bennetti group differs from the G. dorcas conspicuous blackish nose-spot; very dark color on
and the G. gazella groups in their horn rings: more the forehead. Skull and horns slightly smaller in the
widely spaced, not angulated, much more prominent males, but much smaller in the females, than in G.
posteriorly as well as anteriorly, and not connected shikarii; horns, however, very much less divergent.
with the longitudinal ridges. Horn tips pointing SE Iran, into the Makran coastal region of Paki-
somewhat outward. Supraorbital foramina 2–3. Naso- stan. A specimen from 100 km S of Lar (27° N, 54° 20' E),
frontal suture, as in the G. dorcas group, tending to be Hormozgan Province, where its range closely ap-
triangular in shape, penetrating well into the fron- proaches that of G. shikarii, is entirely typical of the
tals. Glandular knee tufts hypertrophied. species, and does not approach the latter in appear-
Anterior basioccipital tuberosities fairly small, as ance (Karami et al., 2002).
in the G. dorcas group; longitudinal ridges behind
them tending to be as bulky as the tuberosities them- Gazella bennettii (Sykes, 1831)
selves; tuberosities lying behind the level of the fora- Deccan chinkara (following Groves, 2003)
men ovale (A. W. Gentry, 1964). Dull reddish brown; median dorsal region and lower
Our discriminant analysis, using a combination of flanks abruptly darker, tawny; coat longer in winter
nine cranial variables (greatest skull length, brain- than in summer. Horn tips of the males slightly less
case length, preorbital length, braincase breadth, pal- divergent, compared with the span, than in G. fusci-
ate breadth, biorbital breadth, nasal length, posterior frons; horns of the females much shorter; nasals
nasal breadth, and anterior nasal breadth) distin- longer.
guished all taxa, except for an overlap between G. Deccan Plateau and the Ganges Valley.
shikarii and G. salinarum.
The largest species, G. salinarum, may have a Gazella christyi Blyth, 1842
slightly greater sexual dimorphism than the other Gujarat chinkara (following Groves, 2003)
species. Differences in horn length among the fe- Very pale, almost silvery drab brown, with only the
males are much more interesting. The horns in the very restricted median dorsal and lower flank zones
-1— females of the two Indian species are very short; in slightly darker. Little seasonal difference in coat
0— G. shikarii the females have much longer horns. length. Males larger, with longer and, on average,
349-47558_ch01_1P.indd 172 5/31/11 7:05 PM

Table 42 Univariate statistics for the Gazella bennetti group
Horn l Tip–tip Span Nas l Gt l Gt br
Males
bennettii
Mean 256.75 120.68 127.84 54.31 181.75 84.00
N 20 19 19 13 8 19
Std dev 31.817 33.657 31.360 5.345 7.996 4.096
Min 177 57 70 47 173 77
Max 338 181 194 67 198 93
christyi
Mean 272.94 129.06 136.15 54.09 191.14 85.27
N 33 33 33 11 7 33
Std dev 25.747 29.830 26.680 4.847 2.610 2.491
Min 217 45 81 47 187 81
Max 350 181 190 63 195 91
fuscifrons
Mean 254.45 116.36 126.09 51.76 185.27 84.48
N 22 22 22 17 15 21
Std dev 28.892 30.335 30.638 4.409 4.008 2.400
Min 194 55 41 44 178 81
Max 304 169 173 59 192 89
salinarum
Mean 283.47 140.36 145.36 57.22 198.00 87.00
N 15 14 14 9 7 15
Std dev 16.115 35.454 36.097 3.383 5.292 1.813
Min 254 96 107 52 190 84
Max 305 235 242 61 205 91
shikarii
Mean 263.50 134.25 150.75 53.67 192.50 85.75
N 4 4 4 3 4 4
Std dev 25.878 31.117 27.921 3.215 3.317 2.217
Min 247 99 124 50 188 83
Max 302 174 188 56 196 88
Females
bennettii
Mean 116.86 53.33 63.67 50.62 179.50 80.20
N 7 3 3 8 6 10
Std dev 18.032 9.292 6.506 5.397 7.817 2.098
Min 101 47 57 44 172 78
Max 149 64 70 61 190 85
christyi
Mean 118.00 87.00 91.00 46.00 184.00 81.00
N 3 1 1 3 2 4
Std dev 23.516 — — 7.810 — 2.000
Min 102 — — 41 184 78
Max 145 — — 55 184 82
fuscifrons
Mean 139.90 58.33 72.67 49.20 181.10 81.45
N 10 6 6 10 10 11
Std dev 60.154 15.201 15.642 5.160 4.630 2.544
Min 0 38 52 43 174 77
Max 230 77 94 58 188 86
salinarum
Mean 136.00 53.00 58.00 55.50 191.50 82.67
N 3 1 1 2 2 3
Std dev 29.462 — — — — —
Min 118 — — 55 186 79
Max 170 — — 56 197 85 —-1
(continued) —0
349-47558_ch01_1P.indd 173 5/31/11 7:05 PM

174 Artiodactyla
Horn l Tip–tip Span Nas l Gt l Gt br
shikarii
Mean 188.00 85.50 67.00 60.00 191.00 80.50
N 2 2 2 1 2 2
Min 182 78 57 — 191 79
Max 194 93 77 — 191 82
more divergent horns than G. bennetti; females hardly nal ridges behind them. Auditory bullae compara-
different in size, their horns short, as in G. bennetti, tively small. Preorbital fossa very large, its inferior
but much more divergent (at least in the single avail- edge being directed obliquely downward, not anteri-
able specimen). orly, as in Gazella. Supraorbital pits larger than in Ga-
Thar Desert, Gujarat, and probably Rajasthan. zella. (These skull characters are as fi rst described by
A. W. Gentry, 1964).
Gazella salinarum Groves, 2003 Groves (1979a) included the species of Eudorcas in
Salt Range gazelle Gazella, and even made them conspecific with G. cu-
Color rich tobacco-brown, with no contrasting zone vieri (!). He also included the E African gazelles (here,
on the back, but a contrasting flank band. Largest E. thomsoni and relatives) in this species, as well as NE
species of the G. bennetti group in the males (the fe- African E. tilonura, but not the enigmatic E. rufina.
males are the same size as the females of G. shikarii), Lange (1972) did not agree that E. thomsoni and E.
with the longest horns, the tips not inturned, with nasalis belong to E. rufifrons, drawing attention, in
the tip-to-tip distance thus about the same as the par ticu lar, to the very long premaxillary contact with
span; horns of the females much shorter than in G. the nasal, which sometimes excludes the maxilla
shikarii (about the same size as those of the consider- from any contact, and the idiosyncratic form of the
ably smaller G. fuscifrons), and straight, like those of nasofrontal suture. After some discussion, however,
the males. Especially long nasal bones. Lange concurred with Groves (1979a) in making ti-
Salt Range, E to Delhi. lonura a subspecies of E. rufifrons. He also argued that
E. rufina is related to the E. rufifrons group.
Eudorcas Fitzinger, 1869
2n = 58. Like Nanger, there is not only the usual ga- Eudorcas rufifrons (Gray, 1846)
zelle X-autosome translocation, but also a Y-autosome red-fronted gazelle
translocation (Vassart, Séguéla, et al., 1995). There is Reddish to ochery in color, with a thin, black flank-
little, if any, difference in chromosomes between E. band, bordered below by a band of the body color (or
rufifrons and E. thomsoni. slightly redder), separating it from the white of the
Differs from other genera of the Antilope/Nanger/ underside. White of the face most clearly expressed
Gazella clade as follows: horn rings extremely promi- as a broad, white eye-ring, with a less well-marked
nent, widely spaced; horn cores more compressed than stripe continuing down to the nose, and little expres-
the others, except for Nanger; molars not elongated, sion of the usual gazelline dark stripe below that.
the styles more developed; lingual wall of the man- Ratio of the transverse to the anteroposterior diame-
dibular P4 open mesially, but fully closed distally; lac- ter of the horn cores averaging 69.1 mm, with a range
rimal bone more extensive on the face than in Gazella; of 61–75 mm (A. W. Gentry, 1964).
nasofrontal suture with an anterior V in the midline; The growth and physical characteristics of E. r.
median nasal tips unusually long, much longer than kanuri in Waza National Park, Cameroon, have been
the lateral prongs; pterygoid plates with large ham- recorded by Nchanji & Amubode (2002). Color red-
uli, as in Nanger; horns slope back at a 45° angle in the dish, deeper in the midfacial region. Mean shoulder
males, as in Antilope; rhinarium restricted to the me- height 69 cm in both sexes (N = 95); body weight
dian portion of muzzle. (Following Groves, 2000). 29.7 kg. Horn length averaging 39.9 cm; interestingly,
Anterior basioccipital tuberosities spaced much virtually the same in the two sexes. One pair of in-
-1— more widely apart than in Gazella, without longitudi- guinal glands; interdigital glands in both feet.
0—
349-47558_ch01_1P.indd 174 5/31/11 7:05 PM

b o v idae 175
Groves (1975a) found that E. r. kanuri is different and Nanger; premolar to molar row length 51– 64 mm
from both the smaller, more reddish, more sexually (mean 57.3 mm), while being generally above 60 mm in
dimorphic W nominotypical E. r. rufifrons and the E. rufifrons and the other gazelle genera. (These differ-
larger, more ochery, less sexually dimorphic E E. r. ences were first demonstrated by A. W. Gentry [1964]).
laevipes. The three subspecies are marginally distinct, Compared with E. nasalis, E. thomsoni is consider-
and restudy is needed to confi rm their status. ably larger, with the horns spreading more in the males,
and almost no white on the face, whereas the nose-spot
Eudorcas tilonura (Heuglin, 1869) is less conspicuous and the eye stripes are less marked
Eritrean gazelle (Brooks, 1961).
Distinguished by its very small size, and much re- Furley (1986) gives gestation as about 180 days in
duced sexual dimorphism; horns more slender, much the wild, but there are records of up to 230 days in
shorter in the males, and suddenly hooked in toward captivity; the earliest age of the male at fi rst fertile
the tips. More rufous in color than E. rufifrons; no nose- mating is 480 days; of the female, 204–364 days; twin-
spot; pygal band absent; knee-brushes larger; light lat- ning is extremely rare.
eral face-stripes nearly obsolete, except around the E of the Rift Valley in Kenya and Tanzania, S to
eye. the Arusha district, and turning SW to Lake Eyasi,
Sudan, E of the Nile and N of the Setit, as far NE as the Wembere Plains, and Shinyanga (Brooks, 1961).
the Bogos River in Eritrea.
Eudorcas nasalis (Lönnberg, 1908)
Eudorcas albonotata (Rothschild, 1903) Serengeti Thompson’s gazelle
Mongalla gazelle Horns of the females very short, only 32% as long as
Size similar to E. rufifrons, but the skull narrower, and those of the males. Size much smaller in both sexes
the horns of the females much shorter, averaging than in E. thomsoni, but the horns not much less; horn
only 50% the length of those of the males (compared tips less widely apart, on average. Brooks (1961) re-
with 60%–70% in E. rufifrons and E. tilonura). Nose- corded that in the field, E. nasalis appears redder than
spot usually present; knee-brushes large; forehead E. thomsoni, but this difference could not be recog-
often entirely white; flank band very wide, with only nized in museum skins. Face-stripes darker; nose-
a slight rufous stripe below it, if that; pygal band spot much more prominent. White crown, extending
present. Compared with E. thomsoni and E. nasalis, as a V down the center of the forehead, often making
horns of the males shorter and curved forward more the face virtually white, except for the dark eye
at the tips. stripes and nose-spot (Brooks, 1961); in this, E. nasalis
Upper Nile, S of the Sudd, and N of the Uganda resembles E. albonotata.
border. The Serengeti ecosystem, extending into the Ke-
nya Rift Valley.
Eudorcas thomsoni (Günther, 1884)
eastern Thomson’s gazelle Eudorcas rufina (Thomas, 1894)
Premaxilla making an extremely long contact with red gazelle
the nasal, generally longer than that of the maxilla
1894 Gazella rufina Thomas. Algiers (purchased).
(generally shorter in E. rufifrons). Horns of the fe-
1895 Antilope (Dorcas) pallaryi Pomel. Oran (purchased).
males only 39% as long as those of the males. Horn
pedicels much more compressed than in E. rufifrons; E. rufina is still very difficult to classify, but it does pos-
ratio of the transverse to the anteroposterior diame- sess many of the features of the genus: compressed
ter averaging 62.8 mm, with a range of 57– 69 mm. horns with very strong, well-spaced ridges, sloping
Posterior expansion of the nasals greater than in E. back at a 45° angle; long median nasal tips; well-
rufifrons, reaching across the area of the ethmoidal developed styles on the molars; lower margin of the
fissure. Nasofrontal sutures strongly concave, sud- preorbital fossa slanting inferiorly. On the other hand,
denly extending backward in the midline, together the anterior basioccipital tuberosities larger than in
making a backward-pointing “relief nib.” Nasals lon- other species, with developed ridges behind them.
ger in the females than in the males (not so in E. rufi- As far as is known, E. rufina, which presumably
frons). Length of the premolar row noticeably shorter formerly lived in N Algeria (although even this is not
than in E. rufifrons or, indeed, in the genera Gazella known for certain), is extinct. —-1
—0
349-47558_ch01_1P.indd 175 5/31/11 7:05 PM

176 Art iodactyla
Madoqua saltiana (Desmarest, 1816)

Dorcatragus Noack, 1894 Salt’s dik- dik
Apparently more closely related to dik-diks than to
1860 Antilope saltiana Desmarest. 1816. Abyssinia.
gazelles, but slightly larger, without an inflated muzzle
1909 Madoqua cordeauxi Drake-Brockman. Dire Dawa, 9° 35'
and no crown tuft; no face-glands; hoofs with small
N, 41° 52' E.
posterior pads; sexes the same size; horns extremely
long for the small size of the skull; skull with very Largest of the species in the M. saltiana group. Color
large auditory bullae. rather drab; upper parts agouti–reddish gray; legs
sandy, not agouti.
Dorcatragus megalotis (Menges, 1894) From the Atbara River to the southernmost
beira Red Sea hills of Sudan, through Eritrea and Ethio-
Color a delicate reddish gray, with contrastingly och- pia N of the highlands, as far SE as the Chercher
ery legs and underside; noticeable dark, gazelline Mountains.
flank-band; head ochery, with white eye-rings.
Along the Somali coast from the Horn of Africa W Madoqua phillipsi Thomas, 1894
to just over the Ethiopian border, and N into SE Dji- Phillips’s dik- dik
bouti, where there is an isolated population in the
1894 Madoqua phillipsi Thomas. Dobwain, 40 mi. S of
Randa region and somewhat farther E, just N of the
Berbera, N Somalia, ca. 10° N, 45° E.
Gulf of Tajoura (Künzel & Künzel, 1998).
1909 Madoqua phillipsi gubanensis Drake-Brockman. Hul
Kaboba, Golis foothills, 35 mi. S of Berbera, 10° 03' N,
Madoqua Ogilby, 1837
45° 07' E.
dik- dik
Tiny antelope with very swollen nasal cavities, short Averaging considerably smaller than M. saltiana;
nasals, and somewhat shortened premaxillae; males somewhat paler and more contrastingly colored. Up-
with small horns, females hornless; tuft of hair on per parts pale gray–agouti, with pale orange shoulder
the crown; females in most species larger than the flares; shoulder flares seeming to be less marked to-
males. ward the W of the range and much more marked
This genus is sometimes divided into two differ- toward the E.
ent genera, or at least subgenera: a brightly colored N Somalia, from the border with Ethiopia and Dji-
group with a less swollen nasal region, and a more bouti, S at least to the Golis Mountains.
dull-colored group with an extreme nasal enlargement
(the putative genus, or subgenus, Rhynchotragus). Here, Madoqua hararensis Neumann, 1905
they are regarded simply as two species-groups, but Harrar dik- dik
more detailed study is needed.
1905 Madoqua hararensis Neumann. Kumbi, Ennia-
Gallaland, 8° 09' N, 41° 39' E.
Madoqua saltiana Group
Small dik-diks (often separated as the subgenus Mado- Size as in M. phillipsi; brighter and redder than M.
qua), all with a gray-agouti back and upper flanks, phillipsi or M. saltiana. Much darker than most speci-
sharply distinct from the reddish-colored lower mens of M. phillipsi; flanks and shoulder-flares bright
flanks, underside, and legs, this reddish area gener- reddish brown, the reddish tint spreading to the back
ally extending to the shoulders (the “shoulder flares” to give a ginger-agouti tone.
of Yalden, 1978). Apparently with an inland range, S of the Golis
Yalden recognized M. piacentinii as a distinct spe- Mountains and E of the Chercher Mountains, as far
cies, and all other taxa as subspecies of M. saltiana. A S as the Webi Shebeyli River. Specimens of M. hara-
close reading of Yalden’s paper indicates that the rensis were obtained with M. phillipsi at Daber-
taxa relegated to subspecies status are, in fact, ex- hick, at the N end of the Golis Mountains, and with
tremely sharply differentiated, though they form an M. swaynei at Milmil, 60 km E of Harar; these rec-
interesting geocline (see his figure 5). Therefore, we ords may indicate sympatry or parapatry, although
recognize a number of different species. Our descrip- Yalden (1978:262) suggests that, alternatively, this
tions (below) are closely based on those in Yalden might indicate “a single species, somewhat variable
-1— (1978). in color.”
0—
349-47558_ch01_1P.indd 176 5/31/11 7:05 PM

b o v idae 177
Madoqua swaynei Thomas, 1894 It was pointed out by Ryder et al. (1989) that groups
Swayne’s dik- dik of dik-dik (referring to breeding-groups in the San
Diego Zoo), which may be almost indistinguishable
1894 Madoqua swaynei Thomas. Type purchased in Berbera,
externally, may nonetheless have quite different
but almost certainly from somewhere S of the River Web
karyotypes and so would not be interfertile.
(Yalden, 1978).
Of three M. guentheri specimens, two had 2n = 50,
1905 Madoqua erlangeri Neumann. Near Sheik Hussein,
and the third had 2n = 48, the difference being a sim-
7° 36' N, 40° 40' E.
ple Robertsonian fission/fusion; offspring had, as ex-
1922 Madoqua citernii de Beaux. Dolo, 4° 11' N, 42° 05' E.
pected, 2n = 49. But it was dik-diks ascribed to M.
Size apparently as in M. phillipsi. Upper parts “grey kirkii that appeared most bewildering. The authors
brown agouti, the plain sandy color being confi ned to divided them into two cytotypes, A and B. Cytotype
the legs, virtually no shoulder flares” (Yalden, 1978:252); A had 2n = 46; B had 2n = 47, differing by a transloca-
thus somewhat similar to M. saltiana in its dull color tion between the X chromosome and chromosome
tones. 10 and, in addition, by several inversions and tandem
Yalden (1978) discussed the problem of identifying fusions. The fi rst group had been imported from an
the “true” M. swaynei, not least because its actual type unknown location in Kenya; the second, from Ga-
locality is unknown. He mapped this species from rissa (and a further specimen was later imported from
the Webi Shebeyli S to beyond the Juba River; the Mbalambala, along with a group of M. guentheri). At-
range is apparently limited to the W by the aridity of tempts to hybridize them resulted in fertile females,
the Ogaden. but sterile males.
The karyotype of M. damarensis (at that time con-
Madoqua lawrancei Drake-Brockman, 1926 sidered a subspecies of M. kirkii) was studied by Ku-
Lawrance’s dik-dik mamoto et al. (1994), who assigned it to a further
type (cytotype D), differing from cytotype A by a
1926 Madoqua phillipsi lawrancei Drake-Brockman. Eil Hur,
tandem fusion and a pericentric inversion, and from
near Obbia, Somalia, 5° 00' N, 48° 17' E.
cytotype B by a pericentric inversion, two additions/
Size as in M. phillipsi. Body color pale silvery agouti; deletions of heterochromatin, and the lack of an
legs and shoulder flares deep reddish orange, sharply X-autosome translocation.
distinct from the back color. The M. kirkii and the M. guentheri groups separate
Mapped by Yalden (1978) from the N and C parts completely when we compare male skulls. They dif-
of the E coastal region of Somalia, its distribution to fer in orbitonasal length compared with greatest skull
the W apparently limited by arid country. length, biorbital breadth, and nasal length; the major
difference between the M. guentheri and the M. kirkii
Madoqua piacentinii Drake-Brockman, 1911 groups is the much longer premaxilla, and the shorter
silver dik- dik orbitonasal distance, of the former (in this relation-
ship, M. damarensis is not different from the M. kirkii
1911 Madoqua piacentinii Drake-Brockman. Gharabwein,
group). The taxa within each group differ in the de-
near Obbia, Somalia, 5° 25' N, 48° 25' E.
grees of contrast between greatest length, biorbital
The smallest species, averaging somewhat smaller than breadth, nasal length, and premaxillary length against
M. phillipsi. Body silver, more finely agouti-banded than orbitonasal length and (to a slight degree) nasal
other species; legs pale sandy; no shoulder flares. Dis- length.
tinctive black rim around the back surface of the ears. In the M. guentheri group, there is an absolute dif-
A few localities on the C Somali coast, nearly as far ference between M. guentheri and M. smithii. Within
S as the Webi Shebeyli. the former, the Kenyan and Torit samples differ from
the Ethiopian sample in their large size, but essen-
Madoqua kirkii Group tially do not differ from each other. Hence we can
This is the group sometimes separated as the genus recognize just two species, with their distributions as
(or subgenus) Rhynchotragus, distinguished by its very follows:
enlarged proboscis, with an enlarged nasal cavity,
1. Madoqua guentheri—most of the Ethiopian
underlain by extremely short nasals and short, curved
range, including M. guentheri Thomas, 1894;
(S-shaped) premaxillae. —-1
—0
349-47558_ch01_1P.indd 177 5/31/11 7:05 PM

178 Art iodactyla
M. wroughtoni Drake-Brockman, 1909; and M. shoulders. White on the limbs extending to the knee,
hodsoni Pocock, 1926. and (usually) beyond the hock.
2. Madoqua smithii—Lake Stephanie S to Lake NW Tanzania. We have seen specimens from Mwa-
Baringo, W to the Nile. nza, Olduvai, Engaruka, 25 mi. E of Ngorongoro, Mto
wa Mbu, Kondoa, Dodoma, Bugogo, Irangi, Tabora.
The female samples differ in much the same way
as the males.
Madoqua hindei Thomas, 1902
In the M. kirkii group, three groupings clearly
Hinde’s dik- dik
emerge: Somalia; Kenya Rift Valley and highlands plus
Tanzania; and E Kenya plus Tsavo, although there is 1913 Rhynchotragus kirki nyikae Heller.
also some difference between the NE Kenya and Tsavo
Unlike any of the other E African species, the males
samples. We can recognize the following four species,
larger than the females. Red-olive; dorsum redder,
one of them distinct on pelage, the other three on
more fi nely speckled. Transition zone on the flank
pelage plus craniometrics:
very narrow. Hair bases darker; black band incom-
1. Somalia— specimens from N and E of the plete; light band red-yellow. Less white on the under-
Tana River are poorly distinguished from side, the limbs, and the face. Becoming more red, less
specimens from here. This is the “true” olive, toward the midback. Chestnut not extending to
Madoqua kirkii. the shoulder; legs paler chestnut (restricted in its ex-
2. Tsavo and elsewhere in E Kenya—the prior tent). White on the legs reaching less far down. Muz-
available name is Madoqua hindei. zle paler, whitish. Forehead and tuft marbled with
3. Kenya and the N Tanzania Rift Valley, black. Eye-ring reaching farther in front of the eye.
including the Serengeti ecosystem—Madoqua SE Kenya, including the Ngong Hills, Tsavo,
cavendishi. Ukamba, and the foot of Mt. Kilimanjaro. The range
4. W Tanzania (mainly S of item #3 above)—this is probably bounded by either the Tana River or the
species, distinguished by pelage characters Athi-Galana River, the coast, the Pare-Usambaras,
only (the skull mostly resembles item #1 and the Rift highlands.
above), would bear the name Madoqua thomasi.
5. Skulls from the San Diego Zoo of the known Madoqua kirkii (Günther, 1880)
cytotype were kindly forwarded to CPG by Kirk’s dik- dik
Steve Kingswood and the late Arlene Kuma-
1912 Rhynchotragus cavendishi minor Lönnberg.
moto. Discriminant analysis showed clearly
that cytotype A is the Rift Valley species (i.e., Females somewhat larger than the males. Much paler
M. cavendishi); cytotype B is (almost certainly) (yellow-olive), but the black speckling varying. Transi-
the Tsavo species (i.e., M. hindei). tion zone on the flanks very narrow. Hair bases very
pale; black band very pale; light band very pale. Nose
Skull lengths for taxa of the two groups confi rm
white. Forehead and tuft marbled with black. Eye-ring
that M. guentheri is much smaller than M. smithii, and
reaching farther in front of the eye. Legs paler chestnut
that M. cavendishi is the largest of the M. kirkii group,
(restricted in its extent); white on the limbs reaching
followed by M. hindei, with M. thomasi and M. kirkii
less far.
being equal as the smallest species.
N Guaso Nyiro to Lamu, into Somalia as far as
Mogadishu; the range is probably E and N of either
Madoqua thomasi (Neumann, 1905)
the Tana River or the Athi-Galana River.
Thomas’s dik- dik
1918 Madoqua cavendishi sensu Hollister. Madoqua cavendishi Thomas, 1898

Cavendish’s dik- dik
Females slightly larger than the males. Rabbit-gray,
with red tones. Flank transition zone usually narrow. 1909 Madoqua langi J. A. Allen.
Hair bases pinkish white; dark band long, sharp, black
Largest of the E African species (greatest skull length
or brownish ; light band yellow or red-yellow. Head
>113 mm, versus <113 mm in the other species) and,
chestnut. Forehead and tuft chestnut, often mixed
uniquely, lacking sexual dimorphism in size. General
with black and off-white. Eye-rings usually going right
-1— color gray; hair bases more whitish; black band short,
around the eyes. Rump and thighs more gray. Legs
0— inconspicuous; light band yellow. Transition zone on
chestnut, with the chestnut zone extending to the
+1—
349-47558_ch01_1P.indd 178 5/31/11 7:05 PM

b o v idae 179
the flank wide. On the head, only the nose chestnut. Eye- Gentry (1992), some resemblances to the Reduncini,
ring reaching farther in front of the eye. Forehead and which must be interpreted as convergences; namely,
tuft marbled with black. Legs pale chestnut (restricted the bare glandular patch below the ear, and the deep
in its extent); white on the limbs reaching farther down. labial valley on the posterior lower premolar.
Kenya’s Rift Valley to the Serengeti ecosystem: We recognize four species, each with a wide range;
Lake Elmenteita; Enterit River, S end of Lake Nak- although all are rather variable in pelage and other
uru; Lake Naivasha; Kijabe; Kedong River; Olorge- characters, we cannot seem to make any convincing
sailie; Amala River; Loita Plains; 13 mi. NW of Narok; divisions within any of them. Undoubtedly there is
Sotik, S Guaso Nyiro; Grumeti River; Banagi. much to be learned about this genus, and our arrange-
ment must be taken as provisional.
Madoqua damarensis (Günther, 1880) Table 43 gives the univariate statistics, and table
Damara dik- dik 44 lists the characteristics of each species. Full syn-
Females considerably larger than the males (appar- onymies are not given here.
ently to a greater extent than any of the E African spe-
cies); general size range about the same as in M. caven- Ourebia quadriscopa (Hamilton Smith, 1827)
dishi. Uniquely, foot glands absent, but a rudimentary Skull quite distinct (see table 43); color varies, but
skinfold between the toes (the glands present in all quite distinct; apparent difference between the sexes
other dik-diks). Anterior part of the body paler than in in the presence or absence of the dark crown mark.
the E African members of the M. kirkii group; crest W Africa; exact E boundary is not known.
between the horns more blackish in color.
N and NW Namibia, and SW Angola Ourebia montana (Cretzschmar, 1826)
1907 Ourebia goslingi Thomas & Wroughton.
Madoqua guentheri Thomas, 1894
1912 Ourebia montana aequatoria Heller.
Gunther’s dik- dik
1913 Ourebia gallarum Blaine.
M. guentheri and M. smithii differ from all previous
1921 Ourebia montana ugandae de Beaux.
Madoqua species in their extremely reduced premax-
1926 Ourebia pitmani Ruxton.
illae and nasals, with a greatly enlarged proboscis. Vir-
tually no sexual dimorphism in skull size, although in Color varies; dark crown mark variable; considerably
external measurements, the females noticeably larger larger than O. quadriscopa, but the horns not longer.
than the males. Greatest skull length 98–109 mm. Tail sometimes darker, but not usually black (unlike
Finely speckled gray; not much chestnut on the shoul- other species); only the Ankole population (“O. pit-
der; underparts pinkish buff. Forehead and crest mani”) sometimes with a black tail.
strongly marbled with black. From N Nigeria E into Ethiopia, S into Uganda.
Lowlands of S and E Ethiopia, NE from Lake
Stephanie, and Somalia (except the coast). Ourebia hastata (Peters, 1852)
1895 Neotragus haggardi Thomas.
Madoqua smithii Thomas, 1901
1905 Ourebia kenyae Meinertzhagen.
Smith’s dik- dik
1908 Ourebia cottoni Thomas & Wroughton.
Closely related to M. guentheri, with which it shares
1922 Ourebia rutilus Blaine.
extreme cranial and nasal specializations and a lack
of sexual dimorphism, but much larger; greatest skull Distinguished particularly by the much heavier band-
length 109–120 mm. General color more reddish; under- ing of richer, more yellow hairs; tail black, even on
parts buff y white. the underside. Size the same as in O. montana, but the
From Lake Stephanie S to Lake Baringo, and W to horns considerably longer.
the Nile. Kenya S to Mozambique and E to Angola.
Ourebia Laurillard, 1842 Ourebia ourebi (Zimmermann, 1783)

oribi Skull quite distinct. Remarkable as the only Ourebia
This genus is unusual among the Antilopini in that species with the males larger than the females, in-
the lingual walls on the lower molars are not flat, the stead of the females being larger. Color more rufous,
nasal tips lack lateral prongs, and the premaxilla fails without conspicuous banding. —-1
to contact the nasals. There are also, as noted by A. W. S of the Zambezi. —0
349-47558_ch01_1P.indd 179 5/31/11 7:05 PM

Table 43 Univariate statistics for Ourebia
Gt l Biorb Max teeth Pre orb Horn l
Males
quadriscopa
Mean 156.875 69.500 45.333 83.800 102.000
N 4 4 6 5 1
Std dev 2.8395 .5774 2.2509 1.3038 —
Min 154.0 69.0 43.0 82.0 —
Max 160.0 70.0 48.0 85.0 —
goslingi
Mean 169.182 74.245 48.327 92.833 97.822
N 11 11 11 6 9
Std dev 4.4118 3.5472 2.4617 4.6224 14.6423
Min 166.0 69.0 42.5 89.0 81.5
Max 180.5 79.0 50.8 102.0 132.0
aequatoria
Mean 169.325 74.289 49.050 91.188 108.550
N 20 19 20 8 20
Std dev 5.1126 3.3221 2.7140 5.1543 16.0361
Min 159.5 69.0 44.0 81.0 79.5
Max 179.0 81.5 53.0 99.0 143.0
montana
Mean 167.414 74.773 257.657 90.440 106.306
N 35 33 35 25 36
Std dev 5.0534 2.8009 1234.0872 3.9773 12.2102
Min 155.0 70.5 44.0 82.0 84.0
Max 176.5 81.5 7350.0 99.5 134.0
gallarum
Mean 167.889 77.306 50.184 88.100 105.222
N 18 18 19 10 18
Std dev 4.6827 3.3437 1.9380 4.0947 11.7301
Min 160.5 71.0 45.0 83.0 83.0
Max 175.5 82.5 52.0 95.0 126.0
cottoni
Mean 170.886 76.913 49.205 86.800 120.667
N 22 23 22 10 21
Std dev 2.8323 1.9344 2.0334 13.1322 11.8894
Min 166.0 74.0 45.0 50.0 97.0
Max 178.5 83.0 53.0 96.0 145.5
pitmani
Mean 167.750 74.711 48.275 91.500 113.906
N 16 19 20 17 16
Std dev 4.3321 2.1751 1.8812 5.5199 13.4959
Min 161.0 70.0 45.0 85.5 88.0
Max 176.0 77.5 52.0 109.0 130.0
kenyae
Mean 168.083 77.571 48.643 88.167 123.571
N 6 7 7 6 7
Std dev 1.3571 2.2809 1.4920 1.5055 5.6600
Min 166.0 75.0 46.0 85.5 116.0
Max 169.5 81.0 50.5 90.0 134.0
haggardi
Mean 162.600 73.600 49.800 87.900 122.900
N 5 5 5 5 5
Std dev 6.1887 1.4318 2.0187 3.8471 11.9185
Min 155.0 71.5 47.0 84.0 108.0
-1— Max 169.0 75.5 52.0 94.0 137.0
0—
349-47558_ch01_1P.indd 180 5/31/11 7:05 PM

hastata
Mean 173.250 75.583 50.250 95.667 105.800
N 6 6 6 6 5
Std dev 2.8240 2.3752 2.4444 2.4014 16.8434
Min 168.0 72.0 45.5 93.0 83.0
Max 175.0 79.0 52.0 99.0 126.0
W Zambia
Mean 168.350 73.156 50.563 91.375 123.206
N 10 16 16 8 17
Std dev 3.1096 1.9470 2.1282 2.1835 14.3745
Min 163.0 70.5 45.0 87.0 100.0
Max 174.0 77.0 53.0 94.0 157.0
rutila
Mean 169.000 74.750 50.417 92.667 139.643
N 4 6 6 3 7
Std dev 2.0412 1.2550 2.8708 1.5275 17.7570
Min 167.5 73.5 46.0 91.0 125.5
Max 172.0 77.0 55.0 94.0 177.0
ourebi
Mean 166.667 75.500 49.333 89.750 117.000
N 3 3 3 2 3
Std dev .5774 1.8028 .5774 — 17.5214
Min 166.0 74.0 49.0 86.5 99.0
Max 167.0 77.5 50.0 93.0 134.0
Females
quadriscopa
Mean 162.611 71.125 46.500 88.222 —
N 9 8 9 9 —
Std dev 4.0603 3.7201 2.4495 2.5386 —
Min 156.0 64.0 43.0 85.0 —
Max 167.0 76.0 50.5 92.0 —
goslingi
Mean 173.500 72.542 49.731 95.200 —
N 12 12 13 5 —
Std dev 5.4897 1.9005 3.8058 2.7749 —
Min 166.0 69.0 42.0 92.0 —
Max 180.0 76.0 54.0 99.0 —
aequatoria
Mean 172.818 71.091 50.292 94.071 —
N 11 11 12 7 —
Std dev 4.4737 2.3856 3.5958 2.7903 —
Min 164.0 66.5 45.0 90.0 —
Max 179.0 74.5 56.0 98.5 —
montana
Mean 171.214 71.133 49.200 91.611 —
N 14 15 15 9 —
Std dev 6.1354 2.0131 2.2975 9.6559 —
Min 157.5 67.0 45.5 67.0 —
Max 179.0 74.0 53.0 101.0 —
gallarum
Mean 170.500 72.786 48.214 92.667 —
N 7 7 7 6 —
Std dev 7.3144 1.8225 2.0587 5.1737 —
Min 161.5 71.0 46.5 86.5 —
Max 185.0 75.5 52.5 102.0 —
—-1
(continued) —0
349-47558_ch01_1P.indd 181 5/31/11 7:05 PM

182 Artiodactyla
cottoni
Mean 173.750 73.000 50.000 89.250 —
N 6 6 6 2 —
Std dev 4.6449 .9487 2.0736 1.0607 —
Min 167.0 72.0 48.0 88.5 —
Max 179.0 74.0 53.0 90.0 —
pitmani
Mean 172.080 82.056 48.800 94.556 —
N 10 9 10 9 —
Std dev 4.2931 34.1462 2.1499 3.0867 —
Min 165.0 68.5 46.0 90.0 —
Max 180.0 173.0 51.0 100.0 —
kenyae
Mean 174.250 74.500 48.000 92.250 —
N 2 2 2 2 —
Std dev 6.7175 3.5355 2.1213 1.7678 —
Min 169.5 72.0 46.5 91.0 —
Max 179.0 77.0 49.5 93.5 —
haggardi
Mean 165.000 75.000 46.500 89.000 —
N 1 1 1 1 —
Std dev — — — — —
Min 165.0 75.0 46.5 89.0 —
Max 165.0 75.0 46.5 89.0 —
hastata
Mean 174.500 70.000 49.750 97.750 —
N 2 2 2 2 —
Std dev 2.1213 .0000 .3536 1.0607 —
Min 173.0 70.0 49.5 97.0 —
Max 176.0 70.0 50.0 98.5 —
W Zambia
Mean 173.625 73.125 49.400 95.333 —
N 4 4 5 3 —
Std dev 4.8713 1.4361 3.4533 4.3108 —
Min 168.5 71.5 45.0 91.5 —
Max 179.5 75.0 53.0 100.0 —
rutila
Mean 173.000 71.833 50.500 95.667 —
N 3 3 3 3 —
Std dev 5.2915 1.2583 .5000 2.3094 —
Min 167.0 70.5 50.0 93.0 —
Max 177.0 73.0 51.0 97.0 —
ourebi
Mean 163.500 75.833 50.000 94.250 —
N 3 3 3 2 —
Std dev 15.1740 1.6073 2.6458 1.0607 —
Min 146.0 74.0 47.0 93.5 —
Max 173.0 77.0 52.0 95.0 —
gitudinal ridges. Horn cores not compressed. Ante-

Procapra Hodgson, 1846 rior premolar (P2) large in both jaws; styles on the
Females without horns. Horn rings of the males other premolars and on the molars small; molars not
-1— prominent, closely spaced, equally prominent on the elongated. Mandibular P 4 elongate, with the lingual
0— back and the front, connected by only very poor lon- wall closed both mesially and distally; mandibular
349-47558_ch01_1P.indd 182 5/31/11 7:05 PM

b o v idae 183
Table 44 Characteristics of each species of oribi
quadriscopa montana hastata ourebi
Color Gray-fawn to deep Dark clay to orange-buff Rich yellow- Bright sandy
orange ochre rufous
Speckling? Speckled Speckled Heavy banding Uniform
Crown mark Small in the females, Varies from conspicuous — Small, conspicuous
absent in the males to absent
Tail Black Darker, at most Black, including the Black
underside
Horns — — Compressed, ridged —
Horn length 102 106 124 117
Skull length, males 159 169 168 167
Skull length, females 163 174 174 164
Tail length — 83–95 — 75
molars not elongate, their buccal midwall convexities P. picticaudata is evidently found all over the Ti-
angular, but not developed into stylids, their lingual betan plateau, and overlaps with P. przewalskii in the
valleys hardly developed; distal lobe on the posterior valley of the Buha River, which flows into Qinghai
lower molar simple. Preorbital fossa totally absent. Su- Lake. No geographic differentiation could be detected
praorbital foramina flush with the forehead; lacrimal in the skulls or the horns by Groves (1967c), but mito-
bone extending onto the face; ethmoid fissure very chondrial control region and cytochrome b sequences
long, extending along the sides of the nasals; lacrimal strongly differentiated five specimens from the Ruoer-
bone with no fenestrae; median nasal tips very long, gai Nature Reserve, Sichuan, from 41 specimens from
the lateral prongs absent. Pterygoid plates with large Tibet and Qinghai Lake (Zhang & Jiang, 2006). The
hamuli. Cranial profile fairly short and high; facial pro- Sichuan clade had a bootstrap value of 91% in the
file convex or straight; horns placed well behind the control region tree and 96% in the cytochrome b tree;
posterior margin of the orbit, sloping at a 60° angle (in the non-Sichuan clade had bootstraps of 100% in both
P. gutturosa, P. przewalskii), or at >60° (in P. picticau- cases.
data); tips of the premaxillae well above the occlusal
level; preorbital foramen rounded, above the mid- Procapra przewalskii (Büchner, 1891)
dle premolars; auditory meatus oval. Glandular knee- Przewalski’s gazelle
tufts absent in P. picticaudata, but present in P. guttur- Much larger than P. picticaudata; rostrum relatively
osa; preorbital glands present only in P. gutturosa; foot short, zygomata less wide, braincase less steeply de-
and inguinal glands present; rhinarium absent; two scending posteriorly. Horns relatively short, curved
pairs of mammae; tail extremely short. (Following backward and inward at the tips. As far as is known,
Groves, 2000). complement of glands as in P. picticaudata.
On this genus, see Groves (1967c). In the 1990s, this species was known only from
In apparent conflict with morphological data, mo- the N and E shores of Qinghai Lake (Kukunor), and a
lecular data (mitochondrial 12S and 16S rRNA se- small area (Bird Island) at the W end ( Jiang et al., 1995);
quences) found that P. przewalskii and P. gutturosa form in the mid-2000s, two small populations were identi-
a clade with respect to P. picticaudata, with 100% boot- fied S of the lake, and along the Buha River (“Bukhain
strap support (Lei, Hu, et al., 2003). Gol” of early Russian authors), which supplies the lake
from the W.
Procapra picticaudata Hodgson, 1846 Lei, Hu, et al. (2003) studied the hypervariable re-
Tibetan gazelle gion of the mtDNA control region from all four of the
Smallest species; relatively long rostrum; broad zygo- surviving population isolates that were known at that
mata; short braincase; narrow nasal bones. Horns very time: Bird Island (now a peninsula as a result of the re-
long, slender, curved in only one sagittal plane. No cent progressive shrinkage of Qinghai Lake), Hudong-
preorbital glands or carpal tufts; no inguinal glands; Ketu, Yuanzhe, and Shadao- Gahai. The Bird Island —-1
glands behind the horns. population is at the W end of the lake; the second and —0
349-47558_ch01_1P.indd 183 5/31/11 7:05 PM

184 Artiodactyla
third populations are at its E end; and the fourth pop- glands behind the horns. Large throat pouch in the
ulation is on its N shore. Present distances between males.
them are not great, but they are strongly isolated, A detailed description of this species is given by
and the degree to which the isolation is entirely due Sokolov & Lushchekina (1997).
to the intrusion of residential areas and roads, as op-
posed to being partially natural, is unclear. In the Tribe Reduncini Knottnerus-Meyer, 1907
minimum-spanning network, the Shadao-Gahai pop-
Reduncini Knottnerus-Meyer, 1907, is antedated by several
ulation proved to be central, with the other three ra-
other names, some of them ranking as nomina oblita, but
diating out from it: Yuanzhe separated from it by
one other, Peleini Gray, 1872 (original spelling
4 substitutions, Hudong-Ketu by 6, and Bird Island
“Peleadae”), may be a threat to priority. Grubb (2004a)
by 14. The authors discussed reasons for the high de-
suggested retaining Reduncini on the grounds of
gree of differentiation, concluding that an interrup-
familiarity, pending a submission to the ICZN.
tion of gene flow between them by human activity,
and a founder effect, due to the reduction in num- The horn cores are transversely ridged; the temporal
ber, were most likely. It is, nonetheless, most strik- lines are often closely approximated on the skull roof;
ing that the Hudong-Ketu and Yuanzhe populations, there are prominent maxillary tuberosities; the cheek-
just 9 km apart, not only shared no haplotypes, and teeth are relatively small, but they have basal pillars
had several fi xed differences between them, but and (on the upper molars) ribs between the styles.
were related only through one of the Shadao- Gahai It is evident that the reedbuck/kob/lechwe/water-
haplotypes. buck group belongs in a clade with the Antilopini, the
Alcelaphini, and the Hippotragini, but constitutes the
Procapra przewalskii przewalskii (Büchner, 1891) sister group to them (Matthee & Robinson, 1999a;
Ropiquet, 2006). The only demurral is Kuznetsova et
1891 Procapra przewalskii przewalskii (Büchner). Bukhain Gol
al. (2002), who found them in a clade with the Cephalo-
(= Buha River), which runs into Qinghai Lake.
phini, supported by bootstrap values of >90% in the
Populations still persist in this general region, but they combined 12S and 16S sequences.
are critically endangered. Within the Reduncini, cytochrome b sequences
gave two clear clades, corresponding to the genera
Procapra przewalskii diversicornis (Stroganov, 1959) Kobus and Redunca, with 98% and 97% bootstrap sup-
port, respectively (Birungi & Arctander, 2001). When
1959 Gazella przewalskii diversicornis Stroganov. Oasis of
Pelea was included, it formed a sister clade to Kobus
Sin-Zhin-Pu, Gansu.
plus Redunca, but the latter clade had only 53% boot-
Size larger; horns more slender, strongly divergent, strap support.
with the tips only a little incurved; color darker.
This is known only from two collections: that Redunca Hamilton Smith, 1827
from the type locality, obtained by Pyotr Kozlov in reedbucks
1909, and another, evidently from nearby, by Fenwick Distinguished from Kobus by the long diastema; the
Owen in 1913. The differences from nominotypical P. very short midsection of the horns; the presence of a
p. przewalskii are very great, and do not overlap. One subauricular gland (marked by a patch of bare skin);
would like to classify them as separate species (and and a wholly bushy tail, lacking a smooth, elongate sec-
this might be justified), but recent evidence ( Jiang tion (Vrba et al., 1994). The face–braincase angle is low,
Zhigang, pers. comm. to CPG) indicates that matters the basicranial tuberosities are wide, preorbital glands
are not quite so simple. are absent, foot glands are present, and inguinal glands
are present (very strongly developed).
Procapra gutturosa (Pallas, 1777) The species / species-groups of this genus form a
Mongolian gazelle morphocline from R. fulvorufula through R. redunca to
2n = 60, all being acrocentrics; this is almost identical R. arundinum, with R. cottoni forming a branch diverg-
to the saiga, except that in the latter, the Y chromo- ing from R. redunca (Grubb, 2000a).
some has very short arms (Soma et al., 1979). On the basis of cytochrome b sequences, R. fulvoru-
Much larger than either P. picticaudata or P. fula was found to be sister to R. redunca plus R. arundi-
-1— przewalskii; small preorbital glands; carpal tufts, num (Birungi & Arctander, 2001); note that the old
0— which may be glandular; inguinal glands large. No three-species arrangement was used by these authors.
349-47558_ch01_1P.indd 184 5/31/11 7:05 PM

b o v idae 185
Redunca nigeriensis Blaine, 1913

Redunca redunca Group Nigerian reedbuck
common reedbucks Light fulvous-fawn, slightly darker along the mid-
The reedbucks in the R. redunca group are relatively line of the back. Paler flanks merge into the white
large in size, with small bullae and relatively small of the underparts. Pale dusky stripe down the front
orbits; the horns are evenly concave forward. This of the lower portion of the forelegs. Hair short,
group is commonly divided into two species, R. arundi- close.
num (corresponding to the first two species in the Nigeria (Sokotu, Wase, Ibi), Chad.
present arrangement, R. occidentalis and R. arundinum)
and R. redunca (the remaining species). There is, in fact, Redunca cottoni Rothschild, 1902
less overall difference between R. arundinum and R. Sudan reedbuck
redunca than this traditional arrangement might sug- Very long, spreading, but thin horns the most distinc-
gest, despite the fact that there is overlap between them tive feature of R. cottoni, the only overlaps with other
in S Tanzania. What has generally been called R. r. reedbucks being provided by old individuals, with
cottoni (here ranked as a full species) is, in its main worn horn tips. Color light. Stripe extending down
features, very like R. arundinum, to the extent that the front of the leg pale, usually mouse-gray.
Lydekker & Blaine (1914a:209) even referred a speci- Sudan, on both sides of Nile, from Bahr-el-Ghazal
men of R. cottoni, from Wau, Bahr-el-Ghazal, to R. to the Dinder and Setit rivers, S to Mongalla.
arundinum.
Redunca bohor Rüppell, 1842
Redunca arundinum (Boddaert, 1785) eastern bohor reedbuck
southern reedbuck
1900 Cervicapra redunca wardi Thomas.
1785 Redunca arundinum (Boddaert). “Cape.” 1913 Cervicapra bohor ugandae Blaine.
1913 Redunca redunca tohi Heller.
Light grayish fawn, grizzled with brown, with a ful-
vous vtinge, especially on the head and the neck.; Grizzled yellowish fawn; dark limb markings sometimes
base of the ears white; round forepart of the eyes present. Underparts white, sharply defined. Fairly close
whitish; chin, upper throat, and underparts white; in general characters to R. nigeriensis, but separated geo-
forelegs generally black in front, from the knee to graphically by the very different R. cottoni.
the hoof; hindlimbs frequently with black markings Ethiopia, Kenya, Uganda.
on the lower part of the shanks. Tail thick, bushy.
E Cape and the E and N provinces of South Africa, Another Species in the Redunca redunca Group
Zimbabwe, N Namibia, and W Zambia (W of the There is evidently an undescribed further species from
Muchinga Escarpment), extending into SW Tanzania. Tanzania (Kigoma S to the Ruaha Plains, Lake Rukwa),
distinguished by its strong sexual size difference. R.
Redunca occidentalis (Rothschild, 1907) arundinum also occurs in this region.
Malawi reedbuck
Paler and grayer; pale rusty gray on the limbs, the Redunca fulvorufula Group
tail, and the body. mountain reedbucks
R. occidentalis has a longer greatest skull length Size small; auditory bullae very large; orbits large, tu-
compared with condylobasal length than does R. bular; horns slender, short, with only weak rings.
arundinum, indicating a larger occipital crest; skull
relatively narrow; nasals longer and broader. Redunca fulvorufula (Afzelius, 1815)
Malawi; Zambia E of the Muchinga Escarpment.
1815 Redunca fulvorufula (Afzelius). E Cape.
Redunca redunca Pallas, 1767 Grizzled grayish fawn, tinged with rufous, especially
western bohor reedbuck on the head and the neck; chin, upper throat, under-
Very small; relatively broad skull; horns short, not parts (extensively), and inner limbs white.
spreading widely; short-horn tips hooked strongly S region of the E Cape, and, farther N, in KwaZulu-
forward and inward. Hair relatively long. Dark yel- Natal, Free State, and the former Transvaal, just
lowish fawn. No dark markings on the limbs. over the borders into SE Botswana, Lesotho, and S —-1
W Africa (Senegal, Gambia, Guinea Bissau). Mozambique. —0
349-47558_ch01_1P.indd 185 5/31/11 7:05 PM

Table 45 Univariate statistics for skulls in the Redunca redunca group
Bas l Gt l Biorb Zyg br Nas l Nas br Max teeth
Males
W Africa
Mean 206.00 226.60 97.88 87.52 86.80 19.66 54.86
N 5 5 5 5 5 5 5
Std dev 4.301 2.702 1.171 2.890 6.181 1.711 1.816
Min 201 224 96 85 77 17 52
Max 211 231 99 91 93 22 57
Nigeria
Mean 221.86 245.78 105.00 93.70 95.80 20.50 59.55
N 7 9 11 10 10 10 11
Std dev 4.914 6.591 2.530 4.990 8.135 .707 3.267
Min 216 238 101 84 82 19 55
Max 229 256 109 103 110 21 65
Chad
Mean 216.40 235.40 99.40 89.33 99.00 20.60 57.00
N 5 5 5 3 5 5 5
Std dev 5.857 7.987 4.615 4.041 6.595 2.510 2.739
Min 208 226 93 87 89 19 53
Max 222 245 106 94 107 25 60
Sudan W of Nile
Mean 227.31 251.23 105.15 96.15 100.15 22.46 59.92
N 13 13 13 13 13 13 13
Std dev 7.994 8.457 5.320 3.671 8.019 2.570 1.801
Min 213 237 91 91 91 18 57
Max 240 262 111 101 112 26 63
Sudan E of Nile
Mean 231.00 254.00 103.54 95.92 106.80 23.33 60.92
N 11 12 13 13 10 12 12
Std dev 8.922 10.180 3.332 3.883 8.677 2.146 4.719
Min 215 235 98 91 97 20 51
Max 241 267 108 104 120 28 70
Ethiopia
Mean 226.55 249.09 108.25 96.73 97.45 21.27 61.36
N 11 11 12 11 11 11 11
Std dev 5.520 9.534 3.166 4.474 5.184 2.195 2.942
Min 218 230 104 87 91 19 59
Max 233 260 115 104 110 27 68
SW Uganda
Mean 221.46 244.56 104.70 96.46 97.54 21.59 59.00
N 24 26 27 28 28 27 28
Std dev 5.672 6.800 3.383 2.782 5.935 1.760 2.293
Min 210 234 96 92 84 18 54
Max 233 261 113 102 107 25 63
E Uganda, W Kenya
Mean 231.30 249.73 106.92 98.58 95.92 23.25 60.67
N 10 11 12 12 12 12 12
Std dev 19.189 7.268 3.476 2.843 6.640 1.288 3.312
Min 211 234 101 94 82 21 55
Max 282 262 114 105 104 25 65
E Kenya
Mean 227.00 249.50 107.00 97.50 96.50 20.50 56.50
N 2 2 2 2 2 2 2
Min 225 246 104 95 96 20 54
Max 229 253 110 100 97 21 59
-1—
0—
349-47558_ch01_1P.indd 186 5/31/11 7:05 PM

b o v idae 187
Bas l Gt l Biorb Zyg br Nas l Nas br Max teeth
Tanzania
Mean 234.63 257.00 109.44 99.22 104.56 21.89 60.44
N 8 8 9 9 9 9 9
Std dev 9.334 12.444 5.570 4.024 7.892 2.147 4.447
Min 220 240 98 91 97 18 53
Max 251 280 119 104 123 25 68
Females
Nigeria
Mean — — 108.00 96.00 — — —
N — — 1 1 — — —
Chad
Mean 210.00 233.00 84.00 86.00 84.00 18.00 58.00
N 1 1 1 1 1 1 1
Ethiopia
Mean 216.00 244.50 102.00 94.50 98.00 21.50 61.50
N 2 2 2 2 2 2 2
Min 216 237 97 91 96 21 58
Max 216 252 107 98 100 22 65
SW Uganda
Mean 220.00 234.67 92.33 92.00 84.33 21.00 58.33
N 3 3 3 3 3 3 3
Std dev 16.371 6.028 3.512 6.557 3.786 3.464 .577
Min 206 229 89 86 80 19 58
Max 238 241 96 99 87 25 59
E Uganda, W Kenya
Mean 221.25 246.00 101.38 94.25 97.25 21.33 60.00
N 8 8 8 8 8 9 9
Std dev 14.878 15.043 5.999 4.464 9.377 2.179 3.122
Min 195 223 91 89 86 18 56
Max 243 269 110 100 113 25 64
E Kenya
Mean 206.00 227.00 92.00 92.00 93.00 17.00 62.00
N 1 1 1 1 1 1 1
Tanzania
Mean 211.50 235.50 95.00 94.00 86.50 19.00 61.00
N 2 2 2 2 2 2 2
Min 207 227 93 91 82 18 60
Max 216 244 97 97 91 20 62
It seems likely that two species are actually and the proportions of their skulls so different that I
represented under the Redunca fulvorufula heading: cannot see why they should be regarded as of the same
two specimens (one incomplete) from the Cape are species: they could well have evolved separately from
smaller than the others; thus somewhat resembling the species occurring in the plains in their respective
R. chanleri. areas” (A. Roberts, 1951:293). The difference in cranio-
metrics is, in fact, mostly one of smaller size, but with
Redunca chanleri (Rothschild, 1895) relatively longer horns. Similar suggestions have been
made in the past about other stenotopic species with
1895 Redunca chanleri (Rothschild). Kenya.
consequently sporadic, isolated distributions, such as
Color lighter and grayer than in R. fulvorufula, with sitatunga. This may or may not be justified, but it is
very little reddish tint; size smaller, with relatively necessary to be cautious, because of what we now
longer horns. “The break in distribution between the know about the shifting distributions of habitats —-1
southern and northern mountain reedbucks is so wide within Africa in the past. —0
349-47558_ch01_1P.indd 187 5/31/11 7:05 PM

Table 46 Univariate statistics for horns in the Redunca redunca group: males
Horn l curve Span Tip–tip Bas diam Horn l str
W Africa
Mean 195.00 135.20 82.00 39.60 140.50
N 5 5 5 5 4
Std dev 25.000 31.396 31.757 2.074 25.489
Min 160 102 57 38 114
Max 230 187 127 43 175
Nigeria
Mean 226.00 186.50 130.10 42.70 177.63
N 10 10 10 10 8
Std dev 19.692 27.355 36.214 3.302 20.199
Min 185 132 90 39 133
Max 250 226 191 48 193
Chad
Mean 216.40 163.20 155.40 38.67 173.75
N 5 5 5 3 4
Std dev 17.743 82.087 24.027 2.082 8.057
Min 200 19 118 37 167
Max 240 217 184 41 183
Sudan W of Nile
Mean 270.00 254.00 176.58 42.17 204.08
N 12 12 12 12 12
Std dev 27.303 40.779 58.405 2.480 19.148
Min 215 194 77 38 176
Max 310 330 282 46 236
Sudan E of Nile
Mean 267.67 261.58 209.08 41.92 203.09
N 12 12 12 12 11
Std dev 39.032 55.590 57.270 2.109 27.566
Min 195 170 100 39 147
Max 330 357 287 46 247
Ethiopia
Mean 214.50 196.78 165.22 40.50 172.40
N 10 9 9 10 10
Std dev 18.326 24.278 32.026 1.958 12.465
Min 185* 151 126 37 153
Max 240 228 210 43 194
SW Uganda
Mean 196.05 153.90 109.38 40.46 154.26
N 19 21 21 24 23
Std dev 19.118 22.246 32.549 3.551 17.353
Min 160 104 41 36 115
Max 220 189 162 47 184
E Uganda, W Kenya
Mean 215.00 163.60 111.70 45.10 168.09
N 11 10 10 10 11
Std dev 24.393 22.653 27.841 4.149 23.437
Min 185 130 68 39 132
Max 265 198 152 51 214
E Kenya
Mean 195.00 206.00 171.00 39.00 150.00
N 1 1 1 1 1
Tanzania
Mean 224.38 177.14 128.86 41.88 179.50
N 8 7 7 8 8
Std dev 21.287 26.448 27.631 2.031 23.028
Min 200 154 84 39 145
-1— Max 260 226 164 45 208
0—
*This includes some aged specimens with very worn tips.
+1—
349-47558_ch01_1P.indd 188 5/31/11 7:05 PM

b o v idae 189
Table 47 Univariate statistics for the Redunca fulvorufula group
Gt l Bas l Biorb Nas l Nas br Max teeth Horn l str Span Tip–tip Horn l curve
Males
Cape
Mean 223.00 198.00 101.50 90.00 23.00 58.50 127.00 108.00 86.50 152.50
N 1 1 2 2 2 2 2 2 2 2
Min — — 100 88 22 58 125 102 84 150
Max — — 103 92 24 59 129 114 89 155
fulvorufula
Mean 231.60 205.93 105.77 88.86 23.07 54.70 156.57 128.33 — —
N 15 14 15 7 7 15 14 6 — —
Std dev 3.180 2.786 4.292 5.336 1.644 2.313 20.879 22.250 — —
Min 226 201 99 81 21 50 134 102 — —
Max 238 211 113 98 26 59 213 156 — —
chanleri
Mean 218.35 192.94 98.65 92.50 22.00 55.50 136.88 127.50 83.50 185.00
N 10 9 10 2 2 10 8 2 2 2
Std dev 5.869 4.333 2.109 — — 3.171 27.368 — — —
Min 209 185 96 89 19 50 113 95 53 130
Max 229 199 103 96 25 60 199 160 114 240
adamauae
Mean — 180.25 90.50 — — — 127.50 — 65.00 —
N — 2 2 — — — 2 — 1 —
Min — 180 90 — — — 125 — 65 —
Max — 181 92 — — — 130 — 65 —
Mongalla
Mean 210.00 184.00 93.00 — — 54.00 119.00 — — —
N 1 1 1 — — 1 1 — — —
Females — — — — — — — — — —
fulvorufula
Mean 228.83 200.83 98.50 90.50 20.25 56.33 — — — —
N 6 6 6 2 2 6 — — — —
Std dev 2.317 1.722 2.881 2.121 2.475 3.327 — — — —
Min 226 199 94 89 19 50 — — — —
Max 233 204 101 92 22 59 — — — —
chanleri
Mean 230.00 — — — — — — — — —
N 1 — — — — — — — — —
Through the mountainous areas of Kenya and practically parallel or only slightly divergent, and very
Ethiopia, and into N Tanzania. little recurved forward. Compared with nominotypi-
cal R. f. fulvorufula, the skull much broader, with
Redunca adamauae Pfeffer, 1962 somewhat longer nasals and frontals. In the only two
Size still smaller. A taxonomically undescribed speci- specimens available at the time of the type descrip-
men from Mongalla is cranially intermediate be- tion, there are only two premolars in the upper jaw,
tween R. adamauae and R. chanleri. whereas all other reedbuck, of this group or any
Pfeffer (1962) described R. adamauae as a new sub- other, have three (the lower jaw of R. adamauae has
species (Redunca fulvorufula adamauae) of smaller size three).
than the others, having a bright reddish coloration, Pfeffer opined that the distribution of R. adamauae
with multibanded, red-tipped hairs; red color becom- was confi ned to the Massif of Adamawa (he gave an
ing more yellowish on the flanks, and white on the altitude of 1923 m), where it is sympatric with a mem-
belly and the internal aspects of the upper limb seg- ber of the R. redunca group, which he identified as R. —-1
ments. Horns more slender than other reedbuck, nigeriensis. —0
349-47558_ch01_1P.indd 189 5/31/11 7:05 PM

190 Artiodactyla
Kobus leche Gray, 1850

Kobus A. Smith, 1840 red lechwe
Distinguished from Redunca by the relatively much Color light to bright red-brown, paler on the sides of
shorter diastema; the long, sweeping midsection of the neck and the flanks. Ear-backs light red-brown.
the horns; the absence of the subauricular gland; and a Foreleg-stripe broad, very dark from the hoof to the
smooth, elongate section on the tail (Vrba et al., 1994). elbow. White throat stripe continuous with the white
The face–braincase angle is high, the basicranial tuber- hair of the chin and the underside.
osities are narrow, and foot glands are present, while A very wide distribution in the wetlands of Zim-
preorbital glands and inguinal glands vary in differ- babwe, Botswana, and much of Zambia.
ent groups of species.
On the basis of complete mitochondrial cytochrome Kobus anselli Cotterill, 2005
b sequences, Birungi & Arctander (2001) found that Upemba lechwe
this genus falls into two clades: one for the lechwe, Color light to bright red-brown, paler red on the sides
and one for waterbuck plus kob. The names Onotra- of the neck and the flanks. Underside, lips, chin, and
gus, Kobus, and Adenota are available for these three around the eyes and the ears white. White of the inner
groups, respectively, if they should prove generically and anterior aspects of the ears extending onto the ear-
or subgenerically distinct. backs. Dark striping along the anterior foreleg re-
duced to a thin stripe on the lower leg, or even just a
Kobus leche Group knee-patch. White throat stripe reduced or absent.
lechwe Relatively small in size; gracile. Skull small and nar-
Members of this group are distinguished from the row; horns more slender than in other taxa, but rela-
Kobus kob group by the much longer midsection of the tively long. (Following Cotterill, 2005).
horns; the longer, rougher pelage; the lack of a preor- Wetlands of the Upemba region, Katanga.
bital gland; and the presence of inguinal glands (Vrba
et al., 1994). Within the group, K. megaceros is sister to Kobus kafuensis Haltenorth, 1963
the S clade, within which K. smithemani is sister to K. Kafue Flats lechwe
leche plus K. kafuensis, with 92% support for the latter Adult males developing large, dark shoulder-patches
clade (Birungi & Arctander, 2001). at the top of the thick black leg-stripes. White throat
For convenience, a few measurements from Cot- stripe continuous with the white chin and the white
terill (2005) are tabulated below: underside. Size relatively large; horns extremely large.
There are strong average differences (note partic- Restricted to the Kafue Flats, Zambia.
ularly the enormous horns of K. kafuensis, and the
narrowness of the skull in K. anselli), but the ranges Kobus smithemani Lydekker, 1900
overlap at their extremes. Nonetheless, the four well- black lechwe
represented species (i.e., excluding K. robertsi) sepa- Horns relatively short; tips more divergent. Body in
rated very cleanly on discriminant analysis (Cotterill, the adult males (in maximum development over 4
2005:figure 6). years of age) suff used with dark hairs on the face, the
Table 48 Measurements for lechwe, from Cotterill (2005)
Horn l Tip–tip Gt skull l Zyg br
anselli 546 315 280.6 118.8

410– 667 (15) 194– 474 (15) 269–292 (10) 114–125 (10)
kafuensis 712 443 300.3 126.4
588– 812 (75) 181–756 (74) 282–317 (54) 116–134 (73)
leche 575 292 298.1 124.6
500– 694 (39) 125–539 (39) 287–320 (28) 116–131 (34)
smithemani 529 359 286.0 119.1
412– 600 (37) 208–517 (35) 271–303 (28) 108–127 (32)
type of robertsi 449 254 281.3 113.5
-1—
0—
349-47558_ch01_1P.indd 190 5/31/11 7:05 PM

b o v idae 191
neck, the shoulders, and the flanks. White throat stripe Kobus kob Group
continuous with the white chin and the white under- kob
side. Size smaller than K. leche and K. kafuensis, and The horn core midsection is shorter than in the K. leche
somewhat larger than K. anselli. group; preorbital glands are present; the inguinal
Lake Bangweulu and the Bangweulu district, NE glands are more strongly developed; and the adult pel-
Zambia. age is short and smooth (Vrba et al., 1994). Because our
arrangement is new, we here give full synonymies.
Kobus robertsi Rothschild, 1907
Roberts’s lechwe Kobus kob (Erxleben, 1777)
Dark shoulder-patches in the adult males, extending Buffon’s kob
onto the throat. Continuous white throat-stripe,
1777 Antilope kob Erxleben. Upper Guinea, “ad Senegal.”
extending forward to the chin and backward to the
1827 Antilope adenota Hamilton Smith. Senegal.
underside. On the evidence of a single skull, size
1827 Antilope forfex Hamilton Smith. Gambia.
small (about the same as K. anselli); horns very short.
1840 Kobus adansoni A. Smith. W Africa.
Formerly from a small area in NE Zambia; now
1842 Antilope annulipes Gray. Gambia.
apparently extinct.
1869 Adenota buffonii Fitzinger. Senegambia.
1869 [Pseudokobus forfex] fraseri Fitzinger. W Africa.
Kobus megaceros (Fitzinger, 1855)
1899 Cobus cob Lydekker.
Mrs. Gray’s waterbuck
1899 Cobus nigricans Lydekker. Kafari, 80 mi. in [i.e., to the]
The females and the juveniles (and the subordinate
W of Freetown, Sierra Leone.
adult males) fawn, but the (dominant) adult males
1905 Adenota koba Neumann.
blackish. Large white patch on the shoulders; white on
1914 Adenota kob riparia Schwarz. Kpandu district, W Togo.
the lips, the underside of the jaw, the midline of the ab-
domen, and the inner aspect of the limbs. Hair reversed Foot glands absent. Color orange-fulvous or tawny.
along the midline of the neck. Slightly smaller than the Whitish ring around each eye, and another around
S species of lechwe, but with extremely long horns. the base of each ear. Ears fulvous on the back, with in-
As shown by Vrba et al. (1994), K. megaceros has a distinct black tips. Indistinct blackish stripe down the
higher face–braincase angle than other lechwe or front of the legs, usually interrupted by a white band
kob, more similar to that of waterbuck. just above the hoofs. Leg lines thin. Smaller than other
Falchetti et al. (1995) studied the relationship be- taxa of the K. kob group, though with slight overlaps in
tween color pattern and dominance in the males in individual measurements.
the Rome Zoo. The males are sandy fawn, like the W Africa, more or less W from Lake Chad. Sam-
females, until 2 years of age, after which they start to ples from far W Africa, Togo, and Nigeria cannot be
turn black, and the nape, the neck, and the area be- distinguished from each other either externally or
tween the horns all begin to become lighter; the craniometrically.
white areas contrast strongly with the remaining The type of Cobus nigricans (from Sierra Leone)
blackish body areas. The authors pointed out that has a more dusky color than usual, with the middle
such a color pattern not only acts as a visual symbol and posterior part of the back being chocolate-brown.
in itself, and emphasizes a male’s threat display, but This coloration is an approach toward the typical
also makes for an easy assessment of horn size. Three black color of the dominant male in K. thomasi popu-
bachelor males, housed together, changed their color lations, and thus is perplexing. Intuitively, the strong
between 2.6 and 3 years of age, but two of them, after sexual dimorphism seen in K. thomasi, as well as in its
losing fights, returned to the subadult coloration; one sympatric congener K. megaceros (perhaps even mim-
of them later regained the adult color after being sep- icking it?), would be a derived condition, responding
arated from the others. In a second enclosure, two to sexual selection; if so, its occurrence in the far W
young males, housed with an adult male as well as of Africa might be interpreted as a homologous muta-
females and other young males, “acquired a dull adult tion. Alternatively, the blackened tone might once
coloring after three years,” but lost it again after a have characterized all of the K. kob group, having
short while; when the adult male died, one of them been lost in all but K. thomasi; Kingdon (1997) implied
reacquired the adult coloration. that the black coloration might be selected against,
Swamps of the White Nile, Sudan. because it is more conspicuous to predators. —-1
—0
349-47558_ch01_1P.indd 191 5/31/11 7:05 PM

192 Artiodactyla
Kobus loderi Lydekker, 1900 Craniometrically, Uganda kob are distinct from
Loder’s kob those from W Sudan and C Africa, mainly in their
small, less spreading horns in comparison with their
1900 Cobus vardoni loderi Lydekker. Type locality:
slightly larger skull size; the horns, however, are
unknown.
stouter.
1905 Adenota pousarguesi Neumann. Sanaga Valley.
1913 Adenota kob adolfi-friderici Schwarz. W side of the
Kobus leucotis (Lichtenstein & Peters, 1854)
mouth of the Shari River.
white- eared kob
1913 Kobus kob alurae Heller. Radio Camp, Lado.
1913 Kobus kob bahrkeetae Schwarz. Bahr Keeta, Upper Shari 1854 Antilope leucotis Lichtenstein & Peters. Sobat River,
River district. Sudan.
1913 Kobus kob neumanni Rothschild. Lake Albert. 1854 Kobus leucotis (Lichtenstein & Peters). White Nile,
1913 Kobus kob ubangiensis Schwarz. Duma, near Libenge, 100 km S of Khartoum.
Oubangui. 1863 Adenota kul Heuglin. Plains of the Sobat River.
1914 Kobus kob adolfi Lydekker & Blaine. 1863 Adenota wuil Heuglin. Sobat River plains.
1899 Adenota nigroscapulata Matschie. Bahr- el- Gebel, W of
Tawny; thick, deep, black line down the front of the
Mongalla.
foreleg below the elbow; thinner black line down
1906 Cobus vaughani Lydekker. Wau, Bahr- el- Ghazal.
the hind shank; black line crossing the anterior pas-
1913 Adenota kob notata Rothschild. Gebel Achmed Agha,
terns, separating the white of the pasterns into two
upper White Nile.
spots. Leg lines varying from thick to thin, ending
fairly abruptly on the foreleg, just below the level of Adult males deep black; females and juveniles, and
the stifle. Ears with the inside tending to be white. some adult males, fawn, like other species of Kobus.
From the Benue River, via the Shari River, to the Large white patch around the eyes and the ears, and,
upper Bahr-el-Ghazal region. as a continuum, through the muzzle, the chin, the up-
Samples in Frankfurt that are referable to Schwarz’s per throat, the chest, and the inner sides of the upper
various described taxa plus Adenota pousarguesi, to portion of the limbs; white eye– ear zone separated
which are added a few from corresponding localities from the white chin–muzzle–throat zone by just a
in the Berlin and Powell-Cotton museums, cannot be relatively thin, but well-marked, black stripe, extend-
distinguished from one another; in craniometrics, ing from the black midface to the side of the neck.
they overlap widely, and, in the absence of pelage dif- Females and other non-black individuals with more
ferences, they are here synonymized. white around the eye than in other species of Kobus,
Some of the E specimens, such as the type of K. k. with this continuing forward on the snout; compara-
ubangiensis, may, however, be darker, with a distinct tively light brown leg-stripe reaching the shoulder and
black suff usion; whitish on the outer half of the ears; going forward to the neck to a variable extent; how-
and have a more conspicuous white eye-ring. Horn ever, a similar black patch on the stifle not connected
span is slightly less in C African than in Sudanese spec- to the hindleg line.
imens. There is, thus, some slight clinal variation. On the evidence of very few individuals, the horns
extremely long and widely spreading, in the upper
Kobus thomasi Sclater, 1896 range of other kob.
Uganda kob Along the White Nile, N from the Uganda border
and E to about the Ethiopian border.
1896 Cobus thomasi Sclater. Berkeley Bay, Lake Victoria, on
Selous (1908) thought the black coloration in the
the Uganda– Kenya border.
adult males was seasonal; Roosevelt and Heller (1914)
Large in size. Markings very distinct, deep black, but thought that it was age related, because black and
not very extensive (ending in a point, at the level of the non-black were equally common during the same sea-
stifle, on the legs). White area around the eyes very son. Analogous to K. megaceros, one presumes that this
much larger than in most specimens of K. loderi, tend- coloration is more likely to be an aspect of dominance.
ing to become buff below the eye. Backs of the ears The hypothesis irresistibly arises that K. leucotis is a
showing a tendency to whiteness. mimic of the somewhat larger, much longer-horned
From the Guasin-Gishu Plateau W through Uganda K. megaceros, which lives in much the same region.
-1— as far as the Semliki Plains. Semliki kob have perhaps There is a magnificent photo ( http://i.livescience
0— less white on the ear than those elsewhere. .com/images/070612_kob_herd_02.jpg), taken in Boma
349-47558_ch01_1P.indd 192 5/31/11 7:05 PM

b o v idae 193
National Park, about 6° N, in Sudan near Ethiopian 5 out of the 40 K. leucotis specimens were part of clade
border, of about 30 animals, mostly or entirely males, 1, and 13 out of the 17 K. thomasi ones from Murchi-
including four strikingly black males with white ear– son Falls were part of clade 2. Out of the 15 samples
eye and muzzle–throat–inside limb zones; three or of K. thomasi from Semliki, geograph ically between
four that are less black, mainly with a deep red-brown Murchison Falls and Queen Elizabeth National Park,
tinge; and others that are tawny, but with the same 9 fell into clade 1, and 6 into clade 2. The authors ex-
very white markings. At least one young male can be plained this by noting that the Kobus kob group was,
seen in which the white markings are muted and less at times during the Pleistocene, divided into W and E
distinct. African refuges (proto-kob and proto-leucotis). Later,
the W population dispersed into E Africa as K. thom-
Revision of the Kobus kob Group asi. Finally, the migratory K. leucotis spread S, hybrid-
Comments on the table leading to our new revision izing with N populations (i.e., Murchison and, to a
of this group are as follows: lesser extent, Semliki) of K. thomasi.
We might propose further that continued back-
—When differentiating the W Sudan versus the
crossing of the hybrids with indigenous K. thomasi
C Africa (both K. loderi) versus the Uganda (K.
would serve to genetically swamp any trace of the K.
thomasi) samples with horn as well as cranial
leucotis phenotype, though the continued predomi-
measurements, Uganda is distinct ( just),
nance of leucotis microsatellites in Murchison Falls
while Sudan and C Africa are intermixed. The
suggests selection one way or the other— either to
types of K. k. pousarguesi and K. k. ubangiensis,
eliminate the leucotis external characters or to main-
as well as a specimen from Bahr-el-Arab, are
tain the latter’s microsatellites. If Kingdon’s (1997)
all within the C African range of variation.
inference is correct, that the black coloration is too
Uganda is distinguished from the other
conspicuous to predators, then there would addition-
samples by skull size, including both zygo-
ally be selection against it outside the White Nile
matic breadth and greatest length. The W
region.
Sudan sample of K. loderi is distinguished from
the C African one by its horn span and
Kobus vardoni (Livingstone, 1857)
(slightly) lower palate length and greatest
puku
length.
Foot glands present. Ingles (1965) recorded that the
—Horn measurements (length [straight], span,
young males have two pairs of mammae in front of
tip-to-tip, and basal diameter) distinguish the
the scrotum and (laterally placed) inguinal glands,
two E Sudan specimens (K. leucotis) strongly
whereas in the females, one pair was anterior and one
from the 10 from W Sudan.
posterior to the inguinal glands. Color golden–buff y
Table 49 gives the univariate statistics for the yellow, brownish only at the hoofs, and whitish inside
group. The overall small size of K. kob can be seen. the ears, around the eyes, the upper lips, the chin, and
Using the mitochondrial control region, Birungi the upper throat, and again on the underside of body
& Arctander (2000) found that K. thomasi is paraphy- and inside the upper part of the legs. Females with a
letic with respect to K. kob. Lorenzen et al. (2007) brownish crown. Black ear-tips small.
compared mtDNA control region sequences and Birungi & Arctander (2001) found very little differ-
seven microsatellites in three of the four taxa recog- ence from the K. kob group in either cytochrome b or
nized here: K. kob (Ghana and Ivory Coast), K. thomasi two nuclear pseudogene sequences, and they even
(Queen Elizabeth National Park, Semliki, and Mur- suggested that members of this group could be only
chison Falls National Park) and K. leucotis (Sudan, subspecifically distinct. Earlier, the same authors
both E and W of the Nile). The microsatellites gave (Birungi & Arctander, 2000) had even found puku
some differentiation between the three species, ex- paraphyletic with respect to kob.
cept that K. thomasi from Murchison Falls fell into the Ngamiland, Zambia, to Malawi.
K. leucotis grouping (note that no microsatellites were Table 50 gives the univariate statistics for K.
available from Semliki). The control region haplo- vardoni. Data for females are more abundant than
types fell into two strongly supported clades: clade 1 those for males. No geographic variation can be de-
contained both K. thomasi from Queen Elizabeth Na- tected, with the possible exception of the very broad
tional Park and K. kob, and clade 2 contained K. leuco- skull and short nasal bones of the single specimen of —-1
tis. There were, however, some “wrong” associations: each sex from Ifakara. —0
349-47558_ch01_1P.indd 193 5/31/11 7:05 PM

Table 49 Univariate statistics for the Kobus kob group
Bas l Gt l Pal l Zyg br Nas l Teeth Span Tip–tip Bas diam Horn l str
Males
W Africa
Mean 243.42 268.94 143.12 99.26 99.06 67.83 249.10 184.20 43.48 299.31
N 12 17 17 19 16 18 21 20 21 16
Std dev 6.748 9.236 4.428 3.445 8.752 2.550 33.602 44.915 4.589 31.419
Min 234 253 134 95 87 64 186 100 35 254
Max 254 283 148 105 119 73 312 268 55 346
C Africa
Mean 262.55 290.70 154.17 104.88 109.35 73.25 305.39 237.74 48.50 385.00
N 29 30 30 32 31 32 33 31 32 30
Std dev 6.495 8.078 5.120 3.536 7.387 3.894 31.282 63.493 4.565 44.830
Min 252 272 142 98 92 65 234 147 41 298
Max 281 312 167 111 122 81 377 377 66 457
W Sudan
Mean 263.36 292.23 152.27 107.31 108.00 72.69 353.77 262.46 49.85 417.62
N 11 13 11 13 12 13 13 13 13 13
Std dev 9.678 8.584 6.051 3.326 5.815 3.401 46.332 57.851 3.805 27.807
Min 250 273 141 99 98 67 280 193 44 373
Max 283 305 162 111 120 78 433 400 57 470
Uganda
Mean 269.29 298.82 155.33 111.45 112.10 73.64 306.67 223.78 53.11 384.00
N 7 11 9 11 10 11 9 9 9 9
Std dev 12.175 10.088 6.538 4.344 9.351 3.613 42.796 87.692 3.444 21.131
Min 257 285 146 105 96 68 256 90 47 347
Max 287 312 167 120 122 79 376 376 58 410
E Sudan
Mean 260.00 285.00 150.50 107.00 109.33 73.67 381.33 291.00 50.00 462.00
N 2 3 2 3 3 3 3 3 3 3
Std dev — 12.000 — 2.000 9.238 2.082 17.039 85.539 2.646 11.136
Min 248 273 149 105 104 72 365 207 47 450
Max 272 297 152 109 120 76 399 378 52 472
Females — — —
W Africa
Mean 227.25 257.25 142.25 95.25 88.75 63.25 — — — —
N 4 4 4 4 4 4 — — — —
Std dev 7.089 8.808 2.872 .957 11.295 2.363 — — — —
Min 217 245 138 94 75 60 — — — —
Max 232 266 144 96 98 65 — — — —
C Africa
Mean 242.00 269.00 149.00 102.00 96.00 69.00 — — — —
N 1 1 1 1 1 1 — — — —
W Sudan
Mean 243.00 277.00 153.00 99.00 110.00 72.50 — — — —
N 2 2 2 2 2 2 — — — —
Min 240 275 150 98 107 68 — — — —
Max 246 279 156 100 113 77 — — — —
Uganda
Mean 248.67 277.33 149.67 105.67 103.50 71.00 — — — —
N 3 3 3 3 2 3 — — — —
Std dev 9.504 7.234 3.055 3.055 13.435 2.000 — — — —
Min 239 269 147 103 94 69 — — — —
Max 258 282 153 109 113 73 — — — —
-1—
0—
349-47558_ch01_1P.indd 194 5/31/11 7:05 PM

b o v idae 195
Table 50 Univariate statistics for Kobus vardoni
Horn l Horn l
Gt l Cb l Gt br Nas l Nas br Max teeth curve str
Males
Ngamiland
Mean 303.00 282.00 112.00 116.00 23.50 75.00 362.50 362.50
N 1 1 1 1 1 1 2 2
Min — — — — — — 347 347
Max — — — — — — 378 378
W + C Zambia
Mean — — 113.00 111.00 28.00 73.00 480.00 400.00
N — — 1 1 1 1 1 1
Luangwa + Malawi
Mean — — — 129.00 33.00 74.00 360.00 302.00
N — — — 1 1 1 1 1
Ifakara
Mean 303.00 286.00 119.00 97.00 31.00 73.00 430.00 373.00
N 1 1 1 1 1 1 1 1
Females
Ngamiland
Mean 288.33 273.00 102.83 107.67 22.00 72.67 — —
N 3 3 3 3 3 3 — —
Std dev 1.528 5.000 5.795 7.371 2.000 3.786 — —
Min 287 268 98 102 20 70 — —
Max 290 278 109 116 24 77 — —
W + C Zambia
Mean 280.17 266.83 106.33 105.20 24.00 73.33 — —
N 6 6 6 5 5 6 — —
Std dev 5.419 4.535 3.559 9.680 3.674 6.121 — —
Min 276 261 102 92 21 65 — —
Max 291 274 111 117 30 78 — —
Luangwa + Malawi
Mean 278.00 266.50 104.00 109.67 22.67 74.67 — —
N 3 2 3 3 3 3 — —
Std dev 6.245 3.536 6.245 9.609 2.082 2.517 — —
Min 271 264 99 101 21 72 — —
Max 283 269 111 120 25 77 — —
Ifakara
Mean 283.00 267.00 112.00 104.00 23.00 70.00 — —
N 1 1 1 1 1 1 — —
Vrba et al. (1994), who listed these characters, main-

Kobus ellipsiprymnus Group tained that waterbuck are paedomorphic in the char-
waterbuck acters of the horns, the short braincase, the small ba-
Members of the K. ellipsiprymnus group have more sicranial tuberosities, the absence of inguinal glands,
slender horn cores, with a lower angle between the their small group size, and their lack of migration.
bases, than in other species-groups; the braincase
is shorter; the anterior basicranial tuberosities are Kobus ellipsiprymnus (Ogilby, 1833)
narrower; the horn cores do not form a distinct angle ellipsen waterbuck
from the base to the midsection; the midsection of 2n = 50–52 (Kingswood, Kumamoto, Charter, Aman,
the horns is long, as in the K. leche group; the face– et al., 1998). The sample of 26 individuals was poly-
braincase angle is high, as in K. megaceros. Preorbital morphic for a centric fusion between chromosomes
glands, foot glands, and inguinal glands are absent. 7 and 11; a fusion between 6 and 18, for which K. —-1
—0
349-47558_ch01_1P.indd 195 5/31/11 7:05 PM

196 Artiodactyla
defassa was polymorphic, was fi xed in the sample of less; in W Africa, 76 mm less; and in Ankole, Laikipia,
K. ellipsiprymnus. and Serengeti, 95–100 mm less.
White ring around the rump. A study of waterbuck by Lorenzen, Simonsen, et
Craniometrically, specimens from South Africa, al. (2006) found that mitochondrial control region hap-
Malawi, and Somalia can be absolutely distinguished lotypes sort into four clusters: two consist entirely of
on very small samples, but larger samples may alter K. defassa (W and E African samples were not distin-
this. Specimens from Gorongoza and from two locali- guished), one contains both K. ellipsiprymnus and hy-
ties in Kenya (Tsavo and Thika) fall broadly within brids from Samburu and Nairobi National Park, and
the range of those from Malawi. the fourth contains samples of all three groups (K. de-
Those from South Africa are the largest, followed fassa, K. ellipsiprymnus, and the hybrids). There was
by those from S Tanzania; the others (Gorongoza, Ma- very little geographic structuring within either of the
lawi, E Kenya) are small. South African skulls are ex- two species.
ceptionally broad— all others (except Gorongoza) are
below the South African range. Pelea Gray, 1851
Those from Kenya have very small teeth. This genus is certainly a sister group to Kobus and
Redunca. Horns very slender, upright, and straight,
Kobus defassa (Rüppell, 1835) with no division into basal, midsection, and terminal
defassa waterbuck portions; braincase short; face–braincase angle low;
2n = 53–54 (Kingswood, Kumamoto, Charter, Aman, basicranial tuberosities small, narrow; foot glands
et al., 1998). The sample of 26 individuals—most strongly developed; other skin glands absent; tail like
from zoos but including three wild individuals from Redunca.
Lake Nakuru—was polymorphic for a centric fusion
between chromosomes 6 and 18; a fusion between 7 Pelea capreolus (Bechstein, 1800)
and 11, for which K. ellipsiprymnus was polymorphic, rhebok
was fi xed in the sample of K. defassa. Small in size, with a pale gray, woolly coat; rhinarium
Very similar to K. ellipsiprymnus, but the entire back somewhat swollen, extending back behind the nos-
of the rump white, instead of having a white ring. trils. Horns short, spikelike, nearly straight or slightly
SW Tanzanian skulls are large and short-faced, bent back.
with short nasals. Zambian skulls are small and long- Found spottily from the Cape, NE through the
faced, with long nasals. Those from Serengeti are Free State, through the high veldt of the NE parts of
small, with short horns, a narrow palate, and a long South Africa.
snout. Those from Laikipia in Kenya are small, but
with a broad palate and a wide horn span. Sudan, An- Tribe Hippotragini Sundevall, 1845
kole, and Rutshuru skulls are large but relatively
The author and date of this name are corrected after Grubb
short-faced, differentiating them somewhat from oth-
(2004a).
ers, including those from W Africa.
Table 51 gives univariate statistics for all water- The females are almost or quite as well-horned as the
buck. Despite its larger range, K. defassa seems less males; the horns, which are ringed, are generally more
geograph ically variable in skull measurements than or less parallel; the frontal sinuses extend into the horn
K. ellipsiprymnus. pedicels; the cheekteeth are hypsodont; the premo-
Differences in the horns are more noticeable than lars are not very reduced; and the molars have basal
in the skulls, though horn measurements are rarely pillars.
different more than on average. Those from Rutshuru
and Ankole are very long (627–796 mm), and those Hippotragus Sundevall, 1845
from both Mongalla and from Zambia/Angola are
Grubb (2004a) gave something of the complex history of
very short (504– 656 mm). Horn span is widest in Ethio-
this name.
pia/Sudan and Ankole, averaging 559 and 528 mm. re-
spectively; all others have an average horn span of un- Hippotragus contains the largest members of the tribe,
der 500 mm. Zambia/Angola animals have tips that do with upright horns curving back in a close arc, so that
not turn in much: the mean tip-to-tip distance is only the tips point downward in large males. There is al-
-1— 43 mm, less than the mean span. In Ethiopia/Sudan, ways a mane along the neck, reaching to a variable
0— Rutshuru. and Mongalla, tip-to-tip distance is 49 mm distance along the back.
349-47558_ch01_1P.indd 196 5/31/11 7:05 PM

Table 51 Univariate statistics for waterbuck: males
Gt l Gt br Preorb Nas l Nas br Max teeth Pal br Horn l str Horn span Tip–tip
S Africa ellips
Mean 413.40 172.67 — 170.67 43.67 106.83 — — — —
N 5 6 — 6 6 6 — — — —
Std dev 16.365 3.559 — 9.416 3.488 4.956 — — — —
Min 398 168 — 157 39 100 — — — —
Max 438 177 — 186 49 114 — — — —
Gorongoza ellips
Mean 384.00 168.00 223.00 136.00 44.00 100.00 94.00 522.00 383.00 247.00
N 1 1 1 1 1 1 1 1 1 1
Malawi ellips
Mean 394.00 158.00 250.00 174.00 40.00 100.00 94.00 475.00 397.00 293.00
N 1 1 1 1 1 1 1 1 1 1
S Tanzania ellips
Mean 400.00 158.50 230.50 151.00 41.50 108.00 93.00 546.00 357.00 195.00
N 1 2 2 2 2 2 2 1 1 1
Min — 156 220 143 41 107 85 — — —
Max — 161 241 159 42 109 101 — — —
Tsavo ellips
Mean 382.00 162.00 230.00 162.00 40.00 99.00 101.00 — — —
N 1 1 1 1 1 1 1 — — —
Thika ellips
Mean 376.67 159.67 227.33 145.33 34.00 93.00 92.67 496.00 436.00 403.00
N 3 3 3 3 3 3 3 2 2 2
Std dev 18.583 7.024 12.014 6.028 1.000 10.149 1.155 — — —
Min 356 153 215 139 33 82 92 493 435 401
Max 392 167 239 151 35 102 94 499 437 405
W Africa, W Sudan defassa
Mean 393.56 155.95 242.17 153.21 41.74 99.26 98.21 614.94 475.88 400.19
N 18 19 18 19 19 19 19 17 16 16
Std dev 20.999 6.753 16.336 9.549 3.694 4.241 6.294 67.389 66.369 107.767
Min 375 140 226 138 36 93 86 520 363 210
Max 471 166 290 181 48 111 113 750 592 546
Ethiopia, NE Sudan defassa
Mean 393.50 160.00 234.33 156.86 40.43 98.57 96.86 643.00 558.60 510.20
N 6 7 6 7 7 7 7 6 5 5
Std dev 5.206 3.266 3.204 5.460 6.655 5.740 2.673 53.062 86.382 131.551
Min 384 155 229 151 32 91 93 554 453 343
Max 398 164 237 167 51 106 101 687 682 682
Rutshuru defassa
Mean 403.40 165.60 239.80 160.30 47.30 100.60 100.10 692.83 528.13 478.64
N 10 10 10 10 10 10 10 12 8 11
Std dev 11.433 4.326 8.664 8.858 5.012 7.090 5.820 49.883 50.654 73.320
Min 389 157 228 146 40 88 88 627 467 374
Max 422 170 251 173 53 107 111 770 635 635
Ankole defassa
Mean 402.71 162.86 242.57 159.00 45.57 105.00 101.29 702.00 510.50 414.25
N 7 7 7 7 7 7 7 5 4 4
Std dev 6.800 2.478 4.894 6.377 4.353 3.830 3.684 59.317 23.700 56.287
Min 396 160 238 150 40 100 96 636 482 356
Max 415 166 252 168 54 110 106 796 535 490
Mongalla, Karamoja defassa
Mean 387.00 155.00 232.67 145.50 39.25 101.50 91.00 564.50 426.00 367.67
N 3 4 3 4 4 4 4 4 3 3
Std dev 13.892 6.880 2.517 8.544 3.202 6.608 6.164 73.903 90.714 138.609
Min 378 150 230 135 37 92 87 504 326 226 —-1
Max 403 165 235 155 44 106 100 656 503 503 —0
(continued) —+1
349-47558_ch01_1P.indd 197 5/31/11 7:05 PM

198 Artiodactyla
Gt l Gt br Preorb Nas l Nas br Max teeth Pal br Horn l str Horn span Tip–tip
Serengeti defassa
Mean 419.44 164.18 253.00 158.73 43.73 107.91 102.91 630.17 474.57 375.50
N 9 11 10 11 11 11 11 6 7 6
Std dev 16.576 5.724 11.284 13.229 4.077 4.929 4.614 43.088 51.626 83.131
Min 398 152 239 144 39 98 97 580 407 260
Max 446 172 272 183 50 116 114 680 546 469
Laikipia, Baringo defassa
Mean 409.20 161.40 253.40 166.40 45.00 98.20 103.00 612.67 477.00 381.67
N 5 5 5 5 5 5 5 3 3 3
Std dev 14.307 2.302 10.761 9.762 3.674 5.848 5.050 26.388 40.632 115.578
Min 398 158 242 158 41 90 97 595 432 260
Max 433 164 271 183 50 106 107 643 511 490
Zambia, Angola defassa
Mean 389.78 159.70 234.56 157.30 42.10 102.50 98.00 569.29 467.00 424.14
N 9 10 9 10 10 10 9 7 7 7
Std dev 33.815 10.904 25.100 19.015 6.027 9.443 10.025 39.225 79.484 122.380
Min 304 135 170 111 33 79 79 515 393 291
Max 425 171 256 178 52 110 113 611 594 594
region contrastingly black, with white preorbital

Hippotragus leucophaeus (Pallas, 1766) streaks and a white muzzle. Ears very long, the tips
blaubok downturned and strongly tufted.
Smallest species of the genus. As we have only very old H. equinus is traditionally divided into a number of
(late 18th century), presumably faded, skins, the color, subspecies, but the exact number is disputed. Avail-
beyond vaguely “blue-gray,” is difficult to reconstruct, able names, and their approximate type localities, are
including its degree of disruptive patterning. Only two as follows:
skulls are known (Groves & Westwood, 1995).
Aegocerus koba (Gray, 1872). Gambia.
H. leucophaeus seems to have had a wide distribu-
Antilope equina (Desmarest, 1804). South Africa.
tion at one time; Loubser et al. (1990) argue that paint-
Egocerus equinus scharicus (Schwarz, 1913). Lower
ings of it occur in a rock shelter near Ficksburg in the
Shari River.
Caledon River Valley in the Free State. These authors
Hippotragus bakeri Heuglin, 1863. Sennar,
also mapped its former distribution, as known by ar-
Sudan.
chaeological deposits along the coast of the W Cape,
Hippotragus equinus cottoni Dollman & Burlace,
as far N as Elands Bay, and they drew attention to an
1928. Cuanza River, Angola.
eyewitness description of what must have been this
Hippotragus equinus dogetti de Beaux, 1921.
species as late as 1853, in the Bethlehem region, not far
Gondokoro, White Nile.
N of Ficksburg.
Hippotragus equinus gambianus Sclater & Thomas,
A short cytochrome b sequence was extracted
1899. Gambia.
from the skin of H. leucophaeus by Robinson et al.
Hippotragus langheldi Matschie, 1898. Tabora,
(1996). Whether by parsimony, neighbor-joining, or
Tanzania.
maximum-likelihood, phylogenetic analysis (using
Hippotragus rufopallidus Neumann, 1899. Upper
Damaliscus as the outgroup) made this species sister
Bubu River, Tanzania.
to H. equinus plus H. niger.
The applicability of Gray’s name is in some question;
Hippotragus equinus (Desmarest, 1804) it is based on “le koba” of Buffon, which in fact re-
roan antelope ferred to both roan antelope and korrigum, in both
Largest species of the genus, but with relatively short, cases from Senegal. In recent times, however, most
-1— strongly curved horns. Overall body color fawn, authorities have felt at ease with its use for the W Af-
0— varying from nearly white to strongly reddish; facial rican roan antelope.
349-47558_ch01_1P.indd 198 5/31/11 7:05 PM

b o v idae 199
Matthee & Robinson (1999b), in their DNA phylog- S forms tend to be paler; the two W African forms
eny, found that roan antelope divide into two clades: H. are much redder in color, on average; H. e. bakeri is
e. koba and the rest. The non-koba clade had 100% boot- more distinctively colored. Forehead (and to some ex-
strap support in the maximum-likelihood and neighbor- tent face) color is sexually dimorphic in H. e. langheldi.
joining trees, but only 82% in the parsimony tree. Like- At the moment, we are not inclined to divide up H.
wise, Alpers et al. (2004), using mitochondrial control equinus. The geographic variations in color and skulls
region sequencing and microsatellite genotyping, found are not well-enough defi ned, and, with the exception
the major division to be between W Africa (Benin, of the W African form, the DNA haplotypes are all
Ghana, and Senegal) and the rest; samples from Kruger intermixed. It is possible that a W African taxon could
National Park, Namibia, Zambia (both E and W of the be recognized, distinguished solely by mtDNA, but
Muchinga Escarpment) and various parts of Tanzania more research needs to go into this.
were scattered through the E/S clade; and, interestingly,
the single specimen from Cameroon was linked to this Hippotragus niger Group
portion, not to the W African one, although it was sable antelope
slightly outside the E/S group (their figure 3). Sable antelope are rather smaller than roan antelope,
The results of our craniometric analysis are as but they have much larger horns, which are also
follows: more sexually dimorphic: those of the males are very
Skulls from South Africa, Mozambique, Zimba- long, and form what is more or less a semicircle, the
bwe, Angola, Katanga, and (W and C) Zambia do not tips pointing downward; while those of the females
separate. Skulls from S Tanzania and Uganda are are much shorter, and make only a relatively short
slightly more separated, but there is still no absolute arc. The juveniles are bright sandy fawn in color, with
difference. Maasailand skulls (H. e. langheldi) are larger a white underside, a white muzzle, and white stripes
in both length and breadth, but greatest length is large from near the bases of the horns down toward the
compared with condylobasal length (i.e., there is more muzzle. Both sexes darken with age, becoming nearly
occipital crest) and with breadth across the horn bases. black in all of the males, and in the females of some
Skulls from the White Nile (H. e. doggetti), and from populations; in other populations, the degree of dark-
Eritrea and W Ethiopia (H. e. bakeri), are somewhat ening is much less (see below).
different, on average. J. H. Grobler (1980) has studied growth in this spe-
There is little difference among populations in con- cies in Zimbabwe. Females in good condition weighed
dylobasal length, but only Ankole and W Africa fail to 160–180 kg, and more if in “very good condition” or
show a marked median occipital protuberance (giving pregnant; males, 180–200 kg, but with their weight
rise to a reduced greatest skull length); might this be seeming to respond more to their general condition.
related to their smaller horns? The average width Both sexes became sexually mature before physical
across the horn bases is smaller in H. e. doggetti than in maturity; males reached physical maturity at 7– 8
other populations. years of age, and females at 5– 6 years. Full dental erup-
Sexual dimorphism in horn size is enormous in tion was complete at 44 months. The horns of the fe-
H. e. doggetti (though sample sizes are small, as N = 4 males appeared to reach their asymptotic length,
[males], 2 [females]), very large in H. e. bakeri and H. e. some 670– 690 mm, at about 8 years of age; and those
scharicus, much less in others, and apparently nonex- of the males, circa 940–970 mm, at about the same
istent in H. e. koba (again, note the small sample sizes, time.
as N = 6, 4). All sable antelope are usually placed in a single
Sexual dimorphism in skull length is greatest in species, but we fi nd good reason to divide them into
H. e. doggetti (N = 6, 4), though this is not as marked two, H. niger and H. roosevelti; not, however, to divide
as it is for horn length; it is unexceptional in others. them into more than two, the described taxa H. n.
As far as external characters are concerned, we kirkii, H. n. variani, and H. n. anselli not being fully
fi nd the following to vary significantly individually, diagnosably different from each other and from the
but not geographically: length of the mane, ranging nominotypical H. n. niger (though there is only a little
from just behind the shoulders to well along the back; overlap in the case of H. n. variani).
foreleg-stripe from poorly marked to very thick and Matthee & Robinson (1999b), using the mtDNA
black, either extending to the knee, or not. control region, found that sable antelope form two
The following external features do vary geograph- 100% supported clades: H. roose velti, and the rest. —-1
ically, but they are not highly consistent: Within the non-roosevelti clade, samples of H. n. niger, —0
349-47558_ch01_1P.indd 199 5/31/11 7:05 PM

Table 52 Univariate statistics for roan antelope
Horn Gt l Cb l Orb Bases One base
Males
equinus
Mean 518.70 450.52 435.59 180.20 135.19 51.04
N 63 84 78 92 90 74
Std dev 46.543 15.635 14.009 7.711 8.333 5.313
Min 489 401 397 158 118 37
Max 606 486 467 197 155 67
langheldi
Mean 539.00 436.71 415.50 183.00 131.43 47.60
N 6 7 6 7 7 5
Std dev 57.124 11.011 4.637 4.546 6.949 2.608
Min 488 428 408 177 120 46
Max 650 460 422 191 143 52
Ankole
Mean 497.33 439.20 429.38 178.30 133.10 48.20
N 6 10 8 10 10 5
Std dev 42.387 14.987 12.850 4.900 7.978 5.020
Min 428 411 414 171 117 42
Max 553 460 453 188 144 56
bakeri
Mean 555.33 451.36 435.92 179.31 133.25 52.44
N 9 14 13 16 16 16
Std dev 63.250 20.075 19.619 7.002 5.905 4.427
Min 460 425 411 165 124 47
Max 645 493 474 187 145 62
doggetti
Mean 509.50 437.75 431.67 174.80 124.20 48.40
N 4 4 3 5 5 5
Std dev 76.405 17.914 21.197 6.140 10.474 7.537
Min 418 415 409 166 111 37
Max 605 458 451 182 140 57
scharicus
Mean 530.72 456.10 444.29 182.06 132.89 50.91
N 18 29 28 35 35 34
Std dev 31.470 15.488 13.410 6.301 6.597 4.323
Min 476 426 412 159 110 41
Max 584 481 469 192 145 59
koba
Mean 481.75 440.86 426.50 172.70 128.70 49.11
N 4 7 6 10 10 9
Std dev 25.025 8.275 7.450 5.982 7.212 5.904
Min 449 431 416 164 119 40
Max 505 455 435 183 140 55
Females
equinus
Mean 483.68 447.70 432.96 175.38 122.85 41.59
N 37 47 45 56 55 46
Std dev 44.412 11.779 11.076 5.839 5.596 3.429
Min 391 422 408 162 110 35
Max 608 469 455 187 137 49
langheldi
Mean 504.40 433.80 423.40 172.00 121.83 41.50
N 5 5 5 6 6 4
Std dev 69.230 15.659 17.009 12.586 6.113 4.041
Min 443 415 401 149 116 38
-1— Max 593 458 448 185 130 47
0—
349-47558_ch01_1P.indd 200 5/31/11 7:05 PM

Horn Gt l Cb l Orb Bases One base
Ankole
Mean 494.00 433.71 422.29 168.29 121.71 39.67
N 1 7 7 7 7 6
Std dev . 17.270 15.798 7.825 7.181 4.082
Min 494 405 400 155 113 36
Max 494 463 449 180 134 47
bakeri
Mean 490.00 447.67 432.67 166.00 114.00 36.00
N 2 3 3 3 2 2
Std dev 52.326 7.638 7.767 10.149 8.485 4.243
Min 453 441 424 155 108 33
Max 527 456 439 175 120 39
doggetti
Mean 362.00 425.33 412.83 167.00 118.33 38.50
N 2 6 6 2 6 6
Std dev 35.355 21.323 18.335 11.314 6.623 2.811
Min 337 395 383 159 108 36
Max 387 457 434 175 125 42
scharicus
Mean 487.82 447.14 434.36 173.64 121.50 42.30
N 11 14 14 14 14 10
Std dev 53.671 14.469 15.599 5.786 6.992 4.877
Min 396 420 405 161 109 33
Max 572 470 460 184 131 49
koba
Mean 493.83 438.25 425.33 170.67 118.89 41.00
N 6 8 6 9 9 5
Std dev 38.851 14.840 15.371 7.778 3.790 2.550
Min 435 415 405 153 114 37
Max 533 466 449 178 126 43
Table 53 External features of roan antelope, describing inconsistent geographic variation
N Color Face Forehead
S Africa equinus, 42 Whitish with red tones Chocolate to black Light brown to black, especially
cottoni to medium brown-gray in the females
or red-brown
C + N Tanzania, 16 Very pale red-fawn to Males: black Females: usually Males: black or chestnut
Kenya langheldi grayish or olive-brown more dark brown (sometimes a few brown hairs)
Females: always chestnut
Ankole 4 Pale roan or gray-fawn Black-brown to very black Chestnut (both sexes)
E of White Nile 10 Light fawn to gray / rusty Dark brown to black Male (N = 1): redder, browner
doggetti gray or red-brown Females and sex uncertain: tan
to nearly black
Dinder, Setit, Eritrea 8 Pale to darker red to Dark chocolate to blackish Brown to nearly black
bakeri honey-colored
Shari scharicus 11 Pale red to gray-red- Deep brown to very black Chestnut to dark brown
tawny
W Africa koba 7 Pale red-brown to sandy Dark brown to black Brown to chestnut
red-gray —-1
—0
349-47558_ch01_1P.indd 201 5/31/11 7:05 PM

202 Artiodactyla
H. n. kirkii, and H. n. variani were intermixed (no perhaps the Angolan one as well; the Kenya one is
samples of H. n. anselli were studied). Interestingly, in very different. The W Tanzania population may be
view of subsequent fi ndings (those of Pitra et al., 2002, a hybrid swarm, just as was indicated by the DNA
2006), all their samples of H. roosevelti were from W data.
Tanzania. Table 55 presents the results of our craniometric
Pitra et al. (2002) sequenced the control region of analysis.
95 individuals. Like Matthee & Robinson (1999b), they In the skull, as in the external features, H. n. niger,
found a wide separation between H. roosevelti (from H. n. kirkii, and H. n. anselli differ on average only. H.
the Shimba Hills and three localities in E Tanzania: n. kirkii averages larger than H. n. niger or H. n. anselli;
Sadani, Songea, and Selous), on the one hand, and the H. roosevelti is very small, as is the W Tanzania sam-
S African samples (Zambia, Malawi, Katanga, and ple; H. n. variani averages larger than other taxa, but
farther S), on the other. They also found an additional, it is not outside the range. Horns of H. n. kirkii aver-
even more divergent clade, restricted to W Tanzania, age considerably smaller than those of H. n. niger, but
from Wembere S to Rungwe; interestingly, 4 out of the only horn of H. n. variani available to be mea-
the 40 samples from W Tanzania assorted not with sured by us is not quite as large as the largest horn of
this clade, but with the S African clade. The authors H. n. niger. The horns of H. roosevelti are very small
found the same divisions in their more limited sam- (on average), and an especially notable feature of H.
ple of cytochrome b. To explain this, they suggested roose velti is the narrow distance across the horn bases.
that there was an initial division between a W Tanza- H. n. kirkii and H. n. anselli differ from H. n. niger in
nian population and the rest, followed by the subse- their high condylobasal length compared with great-
quent division of “the rest” into an E coastal popula- est length (presumably this relates to the develop-
tion and a S population; after that, individuals from ment of the occipital ridge), and the wide distance
the S population spread N and introgressed into the across their horn bases compared with biorbital
W Tanzanian population. breadth. H. n. niger and H. n. kirkii differ from H. n.
The Pitra et al. (2002) study did sample a few speci- anselli in their wide biorbital breadth compared with
mens from Luangwa (presumably representing H. n. small greatest length, and the small distance across
anselli), which nested well within the S clade. The au- their horn bases. The two available W Tanzanian
thors did not sample any specimens of H. n. variani. skulls fall (more or less) within the range of H. n. roo-
Later, part of the same team (Pitra et al., 2006) sevelti from Kenya.
managed to obtain samples of H. n. variani, which As far as skull length is concerned, the size in S
formed a distinct subclade within the S clade. (The African taxa increases in the series niger-kirkii-variani;
other two clades were as before). The S clade, more H. n. anselli (including the Tendaguru sample) is
clearly than in the earlier study, proved to have distinct smaller again, and H. roosevelti (also the enigmatic W
subclades, corresponding approximately to Zambia, Tanzanian sample) is extremely small, though in size
Zimbabwe–Botswana–Namibia, and W Tanzania H. it does overlap with other samples. The limited data
n. variani, although only the last of these was strongly also suggest (for future investigation) that H. roose-
supported (98% bootstrap). The finding that the H. n. velti (if it includes the W Tanzanian sample) is less
variani sequences form a sister group to the minor sexually dimorphic than other taxa.
component of the W Tanzanian taxon may possibly Consequently, we classify sable antelope as given
suggest a source for the latter. below.
Unlike roan antelope, sable antelope have very
defi nite, geograph ically varying morphological fea- Hippotragus niger (Harris, 1838)
tures, which may be conveniently summarized in the southern sable
following table: Size large, horns long and wide across the bases; fe-
In this table, premaxilla type b means that the as- males becoming dark at maturity, either black or dark
cending branch of the premaxilla barely makes con- red-brown, but never golden-red.
tact with the nasal bone; type c means that it forms a
substantial suture with the nasal. Hippotragus niger niger (Harris, 1838)
Evidently, according to this summary table, there Females, and both adult and young males, very dark
are at least four distinct taxa here: S of Zambezi, Zam- red-brown, nearly but not quite black. Estes (2000)
-1— bia/Katanga/Malawi / SE Tanzania, Angola, and Ke- engagingly calls this the “black black sable.”
0— nya. The first two seem different on average only, and S of the Zambezi.
349-47558_ch01_1P.indd 202 5/31/11 7:05 PM

b o v idae 203
Table 54 Summary of the geographically varying features of sable antelope
Premax
N b c Color of males Color of females Face stripes Literature on female color
South Africa 7/6 5 2 Rich deep Deep red-brown to Normal; occasionally Deep chestnut-brown,
red-brown to black-brown vague (N = 1) verging on black
very dark
black-brown
Zimbabwe 3/1 4 Deep reddish Deep reddish Normal (N = 3); narrow, Many females nearly as
brown-black to black-brown invaded (N = 1) black as the males
purplish-black (Victoria Falls NP,
Zambia: Ansell 1971)
W Zambia 6/5 6 Deep brown-black Fairly light Normal (N = 4); fairly Deep chestnut to dark
to nearly black red-brown to deep obfuscated (N = 3); brown (Kafue NP: Ansell
red-brown almost gone (N = 1) 1971); tawny (Batoka:
Ansell 1971)
Katanga 3/2 2 11 Deep blackish Dark red-brown, Only eye-spots (N = 1); —
red-brown to often with just traces on the nose
black blackish tones (N = 2); complete, but
yellowed (N = 2)
Malawi 5/— 9 2 Deep brown — Normal, even very broad Brown (E Province,
Zambia: Ansell 1971)
Iringa, Tunduru 1/2 9 13 Nearly jet-black Deep brown, often Normal —
(SE Tanzania) with blackish
tones
Rukwa, W 6/5 2 Blackish or dark Darkish golden- Thin, disappearing ( just) —
Tanzania chestnut-brown brown to light before the muzzle or
reddish-brown halfway down; or to
the muzzle but much
yellowed
Kenya 1/6* 1 1 Blackish Light golden-brown — Chestnut; face stripes
buff y yellowish (Shimba
Hills)
Angola 9/4 1 13 Black Deep chestnut- to Only eye-spots (N = 4); or —
blackish-brown vague lines (N = 9)
*Based on one museum skin (the type specimen, female) and a wild group seen in the Shimba Hills.
Hippotragus niger kirkii Gray, 1872 Hippotragus niger variani Thomas, 1916
Color of the females usually rich chocolate-brown, giant sable
darker than in nominotypical H. n. niger, with a Larger and longer-horned on average than other S
few darker females present in the W part of the sable; white face-stripe restricted to a white oblong
distribution. in front of the eye, or very vaguely continuing to the
N of the Zambezi, in Katanga and Zambia W of snout. Females usually more chestnut-colored than in
the Luangwa. other H. niger subspecies.
Now restricted to two reserves in Angola. They
Hippotragus niger anselli Groves, 1983 have had to be reintroduced to one of the reserves
As described by Groves (1983d), H. n. anselli differs after the remnant population of females, male-
strongly on average, but not absolutely, from other S sa- deprived, apparently all hybridized with a male roan
ble antelope. Color of the females resembling that of H. antelope.
n. kirkii; white face-stripes averaging broader than other
taxa; skull averaging narrower than other taxa; premax- Hippotragus roosevelti (Heller, 1910)
illa making a much shorter suture with the nasal. Roosevelt’s sable
Zambia E of the Luangwa, Malawi, N Mozam- Noticeably smaller than other sable, with shorter horns, —-1
bique, SE Tanzania (Tendaguru region). and less wide across the horn bases; females almost —0
349-47558_ch01_1P.indd 203 5/31/11 7:05 PM

Table 55 Univariate statistics for the Hippotragus niger group
Max Max Mand Mand

Horn l Gt l Cb l Orb teeth premolar teeth premolar Bases One base
Males
niger
Mean 765.96 422.58 408.39 155.69 111.07 45.75 117.33 46.83 128.67 53.51
N 24 33 18 45 28 20 6 6 30 23
Std dev 99.405 16.072 14.809 5.608 4.422 2.337 4.033 2.483 7.136 5.018
Min 555 394 387 145 104 40 114 44 115 42
Max 1060 457 438 168 120 50 125 51 141 61
kirkii
Mean 708.27 442.27 419.36 159.09 114.94 49.06 123.00 49.33 132.89 52.89
N 15 15 11 23 18 17 3 3 19 18
Std dev 60.879 15.632 11.952 7.982 6.530 2.221 2.646 2.082 6.871 4.639
Min 555 413 397 137 102 45 121 47 122 43
Max 790 465 432 183 124 53 126 51 145 60
anselli
Mean — 425.26 406.00 153.58 113.71 48.00 121.33 47.50 125.86 49.57
N — 23 5 24 7 7 6 6 7 7
Std dev — 12.632 6.364 5.875 4.112 3.162 3.445 2.588 5.429 3.409
Min — 396 397 140 108 44 118 45 116 46
Max — 448 412 164 120 53 127 52 132 55
Tendaguru
Mean 732.00 424.23 402.54 156.93 108.64 43.14 114.00 43.50 131.71 52.17
N 10 13 13 14 14 7 4 4 14 6
Std dev 48.605 13.492 9.527 4.969 5.583 3.132 4.690 2.887 4.665 1.722
Min 667 394 380 149 97 39 109 40 125 50
Max 820 447 417 165 115 47 120 47 141 55
W Tanzania
Mean 720.00 410.67 396.00 160.33 107.00 46.50 117.50 46.00 124.75 48.67
N 3 3 3 3 4 2 2 2 4 3
Std dev 24.576 5.859 3.606 6.429 1.414 — — — 4.425 2.309
Min 692 404 392 153 106 46 115 44 120 46
Max 738 415 399 165 109 47 120 48 129 50
roose velti
Mean 678.00 407.60 373.00 154.17 107.33 41.33 115.00 43.00 120.33 47.20
N 4 5 1 6 6 3 1 1 6 5
Std dev 33.596 14.170 — 5.193 3.327 2.517 — — 8.959 5.215
Min 641 388 — 149 103 39 — — 109 43
Max 715 422 — 164 112 44 — — 134 55
variani
Mean 980.00 453.50 438.14 160.50 116.58 47.27 126.00 47.50 132.93 55.23
N 1 10 7 14 12 11 2 2 14 13
Std dev — 11.674 11.067 4.398 5.728 2.936 4.243 .707 5.717 6.796
Min — 430 422 153 104 41 123 47 123 45
Max — 467 456 168 123 51 129 48 143 69
Females
niger
Mean 623.80 401.17 395.50 146.41 109.54 44.46 115.50 44.50 113.27 42.42
N 10 23 8 27 13 13 4 4 11 12
Std dev 48.540 14.825 17.889 5.679 5.043 3.017 4.123 2.646 5.815 2.392
Min 558 376 361 136 102 40 110 41 106 38
Max 732 434 414 158 118 50 120 47 125 45
kirkii
Mean 611.57 422.60 409.00 153.30 110.14 49.17 111.33 48.67 117.50 41.00
N 7 5 2 10 7 6 3 3 8 8
Std dev 71.575 17.111 4.498 5.460 4.355 14.572 2.517 2.563 2.268
-1— Min 462 407 400 148 103 44 95 46 114 37
0— Max 665 444 418 160 117 55 123 51 121 44
+1—
349-47558_ch01_1P.indd 204 5/31/11 7:05 PM

b o v idae 205
Max Max Mand Mand

Horn l Gt l Cb l Orb teeth premolar teeth premolar Bases One base
anselli
Mean — 415.83 401.00 147.58 113.00 45.80 119.75 45.75 113.20 39.60
N — 12 5 12 5 5 4 4 5 5
Std dev — 12.918 8.276 5.760 6.892 2.387 7.500 3.862 6.535 2.793
Min — 393 389 140 103 42 110 40 103 37
Max — 438 412 162 120 48 128 48 120 43
Tendaguru
Mean 634.09 409.38 389.69 150.23 108.15 44.00 113.67 44.67 115.08 44.00
N 11 13 13 13 13 3 3 3 13 3
Std dev 35.498 6.475 6.473 5.876 3.760 3.000 4.041 1.155 4.804 4.000
Min 570 401 377 138 101 41 109 44 108 40
Max 682 419 402 160 115 47 116 46 123 48
W Tanzania
Mean 581.50 397.50 379.50 148.00 106.67 44.33 112.50 43.50 113.33 41.67
N 2 2 2 2 3 3 2 2 3 3
Std dev 16.263 — — — 6.110 5.686 — — 3.055 4.726
Min 570 391 376 147 100 38 107 39 110 38
Max 593 404 383 149 112 49 118 48 116 47
roose velti
Mean 516.00 — — 152.00 108.00 44.00 — — 112.00 43.00
N 1 — — 1 1 1 — — 1 1
variani
Mean — — — 150.00 114.33 46.33 127.00 48.00 112.67 38.33
N — — — 2 3 3 1 1 3 3
Std dev — — — — 3.215 1.528 — — 4.041 .577
Min — — — 148 112 45 — — 108 38
Max — — — 152 118 48 — — 115 39
invariably a relatively light golden-red color, although

Estes (2000) illustrates a very dark female. Oryx de Blainville, 1816
There are also appears to be an enigma with re- oryx
gard to the W Tanzanian sable, which may be a mix- Long-bodied and relatively short-legged, compared
ture between H. roosevelti and a now-vanished popu- with Hippotragus; horns long, straight or only slightly
lation related to H. n. variani. curved, going back and continuing the line of the face
instead of being upright. Color always light, usually
Addax Rafinesque, 1815 with a black facial mask and a black line separating
The spiral-horned representative of the tribe; stockily the darker flanks from the white underside.
built, with extremely large hoofs. Table 56 gives the univariate statistics for the genus.
The species of Oryx can be arranged in a morpho-
Addax nasomaculatus (de Blainville, 1816) cline, from primitive to derived, in horn length and
addax in skull breadth (associated with body size), running
A. nasomaculatus was described in some detail by from Arabia through E Africa into S Africa, branch-
Dolan (1966), in the context of the establishment of ing to give off the Saharan O. dammah (Grubb, 2000a).
successful captive breeding groups and of a studbook. Iyengar et al. (2006), using the entire control re-
He also briefly discussed the question of whether there gion sequence with addax as the outgroup, found that
could be geographic variation across the Sahara, con- the gemsbok is sister to the Arabian plus scimitar-
cluding, on the basis of an examination of material in horned oryx, with 95% bootstrap support for the non-
the American Museum of Natural History, as well as of gemsbok clade; using a larger dataset, but with only —-1
living animals, that there is no evidence for it. 750 control region nucleotides, the gemsbok was —0
349-47558_ch01_1P.indd 205 5/31/11 7:05 PM

Table 56 Univariate statistics for the genus Oryx
Gt l Biorb Teeth Horn l Horn br Tip–tip Rings
type of beisa
Mean 376.00 147.000 99.000 — — — —
N 1 1 1 — — — —
Awash
Mean 375.17 147.857 105.063 820.00 114.00 188.00 20.00
N 6 7 8 8 1 1 1
Std dev 8.424 4.4508 4.5233 53.809 — — —
Min 362 140.0 97.0 738 — — —
Max 387 154.0 110.0 901 — — —
gallarum
Mean 374.07 149.102 103.491 745.05 108.54 170.61 21.17
N 46 54 54 40 14 14 12
Std dev 11.328 6.4832 6.6115 47.866 7.672 48.939 —
Min 347 134.0 90.0 654 95 72 17
Max 398 165.0 139.0 857 121 244 24
callotis
Mean 377.40 157.361 105.944 743.42 127.50 293.00 16.25
N 15 18 18 19 5 4 4
Std dev 9.156 8.0308 4.1759 70.206 10.308 81.273 —
Min 358 144.0 98.0 643 118 231 13
Max 393 175.0 113.0 895 143 409 23
Angola gazella
Mean 409.33 158.000 121.250 892.75 127.63 362.50 19.75
N 3 4 4 4 4 4 4
Std dev 11.504 2.5820 3.2016 58.864 3.351 40.927 —
Min 398 155.0 119.0 818 124 306 17
Max 421 161.0 126.0 946 132 402 24
Namibia gazella
Mean 431.24 167.087 115.833 895.74 133.31 398.33 20.13
N 21 23 21 19 8 6 8
Std dev 11.924 5.6722 6.7088 86.700 6.447 71.676 —
Min 403 158.0 104.5 705 125 319 11
Max 453 175.0 126.0 1060 145 514 25
Kalahari gazella
Mean 425.50 164.875 116.833 933.00 131.50 477.50 20.67
N 8 12 9 13 6 6 6
Std dev 11.339 6.2126 5.2915 65.986 6.340 89.063 —
Min 401 157.0 108.0 794 123 319 17
Max 438 177.0 123.0 1010 142 545 27
Nata River gazella
Mean 363.00 156.333 99.000 857.00 122.33 318.00 22.00
N 2 3 2 3 3 3 3
Std dev — 5.1316 — 184.854 9.251 154.234 —
Min 349 152.0 93.0 703 117 224 19
Max 377 162.0 105.0 1062 133 496 27
leucoryx
Mean 310.67 124.250 96.250 610.71 85.00 — 30.33
N 6 8 8 7 4 — 3
Std dev 35.814 10.4745 7.4785 260.401 5.050 — —
Min 270 108.0 84.0 89 80 — 27
Max 375 138.5 105.0 890 91 — 34
dammah
Mean 367.86 142.357 101.167 916.00 105.63 — 37.00
N 7 7 6 7 4 — 4
Std dev 10.254 2.4785 9.5376 65.087 4.423 — —
-1— Min 354 138.0 92.0 820 99 — 34
0— Max 380 145.5 113.0 1000 108 — 41
+1—
349-47558_ch01_1P.indd 206 5/31/11 7:05 PM

b o v idae 207
sister to the Arabian oryx, but in this case the boot- were reported as being from “eastern Africa, possibly
strap support was low. Kenya,” it seems more likely that they belonged to
Masembe et al. (2006) found that mtDNA (control the present species.
region and cytochrome b) sequences distinguish O. Paler, a purer gray; this color extending below the
callotis strongly from O. gallarum (which they referred black flank band, and therefore not a boundary be-
to as Oryx beisa beisa), except for a single haplotype tween the body color and the white underside. Face a
from Samburu, which assorted with O. callotis. Two little paler than the neck, or even whitish. Facial and
widely divergent lineages of the control region oc- gorge bands separated (sometimes the two nasal
curred in the Samburu sample of O. gallarum; the au- bands joined); median bands black; nose streak dif-
thors interpreted this as a relic of secondary contact fuse. Legs white, with only a trace of stripes. Flank
between two formerly isolated populations. band 20– 44 mm thick (12 skins), much as in the
Awash taxon. Dorsal stripe 30– 43 mm thick (12 skins),
Oryx beisa (Rüppell, 1835) but generally very vaguely expressed, and totally lack-
Beisa oryx ing in one specimen.
Skins seen: 8. The localities of specimens seen by us are N Guaso
More ochraceous (a pinkish wash); this color not Nyiro, Guasinarok, Kahlbin, Ngare Ndare, Guida, Lai-
extending below the black flank band, which forms kipia, Nyeri, and Archer’s Post.
the boundary between the gray flanks and the white
belly. Face colored like the body; facial bands gener- Oryx callotis Thomas, 1892
ally joined up with each other and with the gorge fringe- eared oryx
band. No ear tufts. Flank band 39–58 mm thick in 2n = 58; the karyotype is identical to that of O. galla-
four skins. Dorsal stripe 56 and 71mm thick in two rum (Kumamoto et al., 1999).
males, and 31 and 46 mm in two females; this stripe Darker, duller, and browner than in O. gallarum.
sometimes extending fully up the neck, or for three- Face deep ochery, except for the white muzzle stripes.
quarters of the neck, or just to the withers. Toothrow Ears with long tufts or tassels. Usually no connection
short, compared with other taxa (on the evidence of a between the nasal and the median face bands. Legs
single skull). not paler below the carpus/tarsus. Flank band 30 and
Somalia and extending into Ethiopia. The localities 44 mm thick in two skins. Dorsal stripe very reduced,
for specimens seen by us are Hullieh, Mersi, Haud, only 25 and 30 mm thick in two skins, confi ned to
Bale, Berbera. the rump only, or very faint. Skull, though the same
There is a fairly distinctive form, closest to this overall size as in other NE African species, compara-
species, in the Awash Valley of Ethiopia. It differs in tively broad. Horns relatively short, but very thick at
that the face is deeper ochraceous in color; the legs the base; tips comparatively wide apart.
are paler; the flank band in 8 skins is only 23– 45 mm We have seen specimens from S of Mt. Longido,
thick; the dorsal stripe is 24– 40 mm thick; and the and from 100 mi. S of Kilimanjaro.
toothrow is not shortened.
The horns, compared with O. gallarum and with Oryx gazella (Linnaeus, 1758)
most (but not all) O. callotis, are relatively very long for gemsbok
the skull length. Very unfortunately, the only avail- 2n = 56; differing from O. beisa and O. callotis in hav-
able skull of O. beisa (the type) lacks the horn sheaths, ing a centric fusion between chromosomes 2 and 17
so it cannot be compared with others. (Kumamoto et al., 1999).
Flank band 119–229 mm thick. Dorsal stripe 90–
Oryx gallarum Neumann, 1902 116 mm thick; in two juveniles, this thickness only 54
and 74 mm. Skull larger than in the NE African spe-
1910 Oryx annectens Hollister.
cies. Horns thick, like those of O. callotis, and longer
Probably Oryx gazella subcallotis Rothschild, 1921, is another
than in the NE African species, with the tips farther
synonym.
apart.
Skins seen: 16. Samples from Angola, Namibia, and the Kalahari
2n = 58; differing from O. gazella in lacking the are, in general, very much alike, but three specimens
centric fusion between chromosomes 2 and 17 (Ku- from the Nata River (far NE Botswana, near the Zim-
mamoto et al., 1999). The animals these authors sam- babwe border) are unexpectedly small (see table 56), —-1
pled were identified as Oryx gazella beisa, but as they with small teeth and less spreading horns. —0
349-47558_ch01_1P.indd 207 5/31/11 7:05 PM

208 Artiodactyla
analysis, Vrba (1979) proposed that the genus Beatra-

Oryx dammah Cretzschmar, 1826 gus should not be recognized, but should instead be
scimitar-horned oryx sunk into Damaliscus; and that Lichtenstein’s harte-
2n = 56–58; polymorphic for a centric fusion between beest was not part of Alcelaphus, but was more closely
chromosomes 2 and 15 (Kumamoto et al., 1999). related to Connochaetes, so she revived the genus Sig-
Very pale species, with “washed-out” markings; moceros for it. Later, Vrba (1997) reconsidered this, in
white, with light reddish face markings and a reddish part on the basis of a much fuller fossil record; she re-
neck, this latter color continuing as a vague stripe along stored the genus Beatragus, and sunk Sigmoceros back
the lower flanks and onto the haunches. Skull compa- into Alcelaphus, proposing the following splitting times
rable in size to that of most of the NE African species, based on the fossil record:
but with rather small teeth. Horns more curved than
Alcelaphus versus Connochaetes— slightly <3 Ma
in any other species, and very slender, with more rings.
Alcelaphus/Connochaetes versus Beatragus—
Sahara; extinct in the wild, but breeding well in
somewhat >5 Ma
captivity.
Alcelaphus/Connochaetes/Beatragus versus
Damaliscus— 6 Ma
Oryx leucoryx (Pallas, 1777)
Arabian oryx This order of splitting does not agree with molecular
2n = 57–58; polymorphic for a rare (two heterozygotes evidence (see above).
in 77 specimens) centric fusion between chromosomes
18 and 19 (Vassart et al., 1992; Kumamoto et al., 1999). Alcelaphus de Blainville, 1816
Vassart et al. (1992) noted that the translocation occurs hartebeest
only in herds originating from Qatar, and ultimately One of the key early 20th-century papers on bovid
from the Rub Al Khali. No genetic difference was de- taxonomy, and one that has survived the test of time
tected by Vassart et al. (1991) between those from N with very little necessary modification, is Ruxton &
and S Arabia. Schwarz’s (1929) revision of hartebeest. They may have
White, but the typical oryx markings black and recognized somewhat too many taxa, and they may
well expressed. Skull and teeth much smaller than in have been wrong in ascribing almost all hartebeest to a
any other Oryx species; horns shorter and more slen- single species (Schwarz was exhibiting his basic ten-
der. Number of horns rings intermediate between O. dency toward lumping), but their analysis of hybrid-
dammah and the other Oryx species. ization, in par ticu lar, was ahead of its time.
Formerly ranged throughout the Arabian penin- As noted above, Vrba (1997) reversed her former
sula, N to Jordan. Became extinct in the wild, but re- decision to recognize the genus Sigmoceros for Lich-
introduced from captive stocks. tenstein’s hartebeest. Matthee & Robinson (1999a)
confi rmed, on molecular grounds, that Lichtenstein’s
Tribe Alcelaphini Brooke in Wallace, 1876 hartebeest does not constitute a distinct genus, but is
part of Alcelaphus.
This name is not to be displaced by the slightly earlier
Some of the species of this genus form a morpho-
Connochaetini Gray, 1872 (original spelling “Conno-
cline as far as the elongation of the horn pedicels is
chetidae”), as noted by Grubb (2004a).
concerned: A. cokii to A. lichtensteini on one branch,
The females are almost as well horned as the males. and to A. lelwel on another (Grubb, 2000a).
The horn cores tend to be transversely ridged; there is Arctander et al. (1999) took control region samples
a postcornual fossa; extensive frontal sinuses reach from all geographic groups of hartebeest, except for
into the horn pedicels; the brain case is short and A. buselaphus and A. tora. They found a clean division
strongly bent downward; the temporal ridges diverge between three quite distinct clades, all with high boot-
posteriorly; the facial skeleton is extremely elongated; strap values: clade I, A. major; clade II, an E African
the zygomatic arches are deepened beneath the or- group (A. lelwel, A. cokii, A. swaynei); and clade III, a S
bits; the palatal foramina are wide; the teeth are very and SE group (A. caama, A. lichtensteini). Within clade
hypsodont; and the premolar rows are shortened. II, samples of the three taxa were partly separate: two
This tribe contains four genera, of which Conno- haplotypes of A. swaynei were nested within A. cokii,
chaetes is sister to the other three (Matthee & Robert- whereas the other five formed a separate subclade, and
-1— son, 1999a; Ropiquet, 2006). On the basis of a cladistic three of A. cokii (out of quite a large number) nested
0—
349-47558_ch01_1P.indd 208 5/31/11 7:05 PM

b o v idae 209
within A. lelwel. Within clade III, the two component ger pedicels to deflect the blows. The only other high
species were entirely separate. correlation was between horn length and mean annual
A further study (Flagstad et al., 2001), with larger, rainfall; evidently, male hartebeest grow relatively lon-
more diverse samples (including some from museums), ger horns in more productive environments.
expanded somewhat on this. These authors found the We give full synonymies in this genus, except for
same division between S, E, and W lineages; the E and the two southernmost species, for which there is no
W lineages together formed a major clade, with 91% taxonomic dispute.
bootstrap support against the S clade. Within the E lin-
eage, the three available specimens of A. tora formed Alcelaphus buselaphus (Pallas, 1766)
their own clade, the others being the main A. swaynei Bubal hartebeest
clade and a mixed swaynei/cokii/lelwel clade, within
1766 Antilope buselaphus Pallas. Probably Morocco.
which each of the three species tended to occupy a sep-
1767 Antilope bubalis Pallas. A renaming.
arate subclade, but with a considerable number of
1836 Bubalis mauritanicus Ogilby.
“wrong” allocations. In addition, there was some para-
1891 Alcelaphus bubalinus Flower & Lydekker.
phyly between E and W lineages, but not between
1914 Bubalis bubastis Blaine. Abadiyeh, Egypt.
them and the S lineage. The authors suggested that this
favored an E African origin for the crown-group of the Males of this species with the shortest horns of all,
genus, and that the modern E species derived from a though not more slender than in A. tora, A. swaynei,
refugium within the range of present-day A. lelwel. and A. cokii (Cappellini & Gosling, 2006). Greatest
The studies of Cappellini (2007; Cappellini & Gos- skull length 421 mm in a single male; horn span 97%
ling, 2006) have added a great deal to our understand- of the basal length in the males; least frontal width
ing of taxonomic variation in Alcelaphus, particularly as 71%–72% of the biorbital width in two males (Ruxton
related to sexual dimorphism and sexual selection. & Schwarz, 1929).
Hartebeest males fight by dropping to their knees and Formerly lived in the Maghreb (Morocco, Algeria,
battering each other, and, finally, by interlocking horns and Tunisia N of the Atlas Mountains), and during
and wrestling. The elongated horn pedicels apparently dynastic times in Egypt. Hartebeest of unknown af-
displace the impact of the blows away from the brain- fi nities formerly occurred in the Palestine/Israel re-
case. Cappellini (2007) divided the species, broadly gion (Uerpmann, 1987).
speaking, into heavily armed (A. lelwel, A. lichtensteini,
A. caama, and A. major) and lightly armed (A. buselaphus, Alcelaphus major (Blyth, 1869)
A. tora, A. swaynei, and A. cokii) groups. All species are western hartebeest
sexually dimorphic, but to varying degrees: in skull
1869 Boselaphus major Blyth. Gambia (fi xed by Schwarz,
length, there is very little in any (of the extant) species;
1920).
in pedicle height, there is little in A. tora or A. cokii,
1914 Bubalis luzarchei Grandidier. Nieri Ko River, near the
more in A. major, and still more in the other four extant
junction with the Gambia River, Senegal.
species; in horn length, little in A. tora or A. caama, enor-
1914 Bubalis major invadens Schwarz. Garoua, Benue River.
mous in A. lichtensteini, and intermediate in other spe-
1914 Bubalis major matschiei Schwarz. Kpandu, Togo.
cies; in horn circumference, much greater in A. lichten-
steini and somewhat greater in A. caama and A. lelwel Horns strongly twisted, U-shaped from in front. Great-
than in other species; and, finally, in skull weight—an est skull length 470–528 mm in the males; horn span
unusual metric, not commonly taken, but very reveal- 64%– 83% of the basal length in the males; least fron-
ing in this instance—low in A. cokii, A. tora, and A. ma- tal width 76%– 87% of the biorbital width (Ruxton &
jor, and very great in the other four species. The length Schwarz, 1929). Color uniform tan-brown, with no
of the available breeding season in a given species was dark markings but, occasionally, a thin white band be-
negatively correlated with pedicle height, skull weight, tween the eyes. In Cappellini & Gosling’s (2006) analy-
and (though not strongly) horn circumference. In other sis, this species comes out as “middling” in all respects
words, the shorter the breeding season, the more in- (relative horn length, horn circumference, pedicle
tense the male competition, so those hartebeest living height, skull weight).
in more seasonal environments, with more constrained From Senegal in the W to the sources of the Log-
breeding seasons, would have heavier skulls (and, per- one River, somewhat to the S of the W end of the dis-
haps, thicker horns) to deliver stronger blows, and lon- tribution of A. lelwel. —-1
—0
349-47558_ch01_1P.indd 209 5/31/11 7:05 PM

210 Artiodactyla
As in A. tora, horn circumference in low, relative to

Alcelaphus lelwel (Heuglin, 1877) skull length, in the males; horn length relatively low;
lelwel hartebeest pedicle length low in the females (as low as in A. lich-
tensteini), but not especially so compared with the
1877 Acronotus lelwel Heuglin. Jur River (fi xed by Schwarz,
males in other Alcelaphus species (Cappellini & Gos-
1920).
ling, 2006). Greatest skull length 409– 472 mm in the
1892 Bubalus jacksoni Thomas. Between Lake Victoria and
males; horn span 114%–148% of the basal length in
Lake Naivasha.
the males; least frontal width 64%– 69% of the bior-
1904 Bubalus jacksoni insignis Thomas. Maanja River, W of
bital width (Ruxton & Schwarz, 1929). Much deeper
Kampala.
in color than other E African species, with striking
1905 Bubalis niediecki Neumann. Gelo River, Upper Sobat
black markings on the face and the body.
River, Ethiopia.
Formerly from N Somalia through the Ethiopian
1912 Bubalis lelwel roosevelti Heller. Gondokoro, Sudan.
Rift Valley, at least as far S as Lake Zwai.
1913 Bubalis lelwel tschadensis Schwarz. Ketekma, E of
Flagstad et al. (2000) found that the surviving pop-
Tchekna, Bagirmi.
ulations of this species still retain significant amounts
1914 Bubalis lelwel modestus Schwarz. Bahr Keeta, Upper Shari.
of genetic diversity (mtDNA and microsatellites), and
Horns forming a narrow, upright V from in front. that two of the now-isolated surviving populations
Greatest skull length 455–562 mm in the males; horn have diverged considerably.
span 52%–76% of the basal length in the males; least
frontal width 68%– 86% of the biorbital width (Rux- Alcelaphus cokii Günther, 1884
ton & Schwarz, 1929). Color more reddish than in kongoni
neighboring species.
1884 Alcelaphus cokii Günther. Mlali Plains, E of Mpapwa,
The males in this species have relatively longer
Tanzania.
pedicels than in any other Alcelaphus species, and a
1914 Bubalis cokei sabakiensis Zukowsky. Athi Plains, Kenya.
heavy skull relative to its length (Cappellini & Gos-
1914 Bubalis cokei schillingsi Zukowsky. Lake Jipe, SE of Mt.
ling, 2006).
Kilimanjaro.
From Lake Chad E to E Sudan and possibly just
1914 Bubalis cokei schulzi Zukowsky. Olossirva Plateau, NE of
into Ethiopia, SE to the borders of the E African Rift
Ngorongoro.
Valley, S to the Guasin-Gishu Plateau in Kenya and to
1914 Bubalis cokei tanae Matschie & Zukowsky. N of Nairobi.
Ankole in Uganda.
1914 Bubalis cokei wembaerensis Zukowsky. Mkalamo,
Wembere Steppe, Tanzania (selected by Ruxton &
Alcelaphus tora Gray, 1873
Schwarz, 1929).
tora hartebeest
1916 Bubalis deckeni Matschie & Zukowsky. Middle Rufu
1873 Alcelaphus tora Gray. Dembelas, Bogos, Eritrea. River, Tanzania.
1916 Bubalis oscari Matschie & Zukowsky. Mt. Gurui,
Horn circumference low, relative to the skull length,
Tanzania (selected by Ruxton & Schwarz, 1929).
in the males (Cappellini & Gosling, 2006). Greatest
skull length 480– 487 mm in three males; horn span Size small; greatest skull length 419– 460 mm in the
122%–128% of the basal length in the males; least males; horn span 97%–128% of the basal length in
frontal width 65%– 67% of the biorbital width (Rux- the males; least frontal width 57%–72% of the bior-
ton & Schwarz, 1929). bital width (Ruxton & Schwarz, 1929). Color brown-
Formerly more or less along the Sudan–Ethiopia ish. As in A. tora and A. swaynei, horn circumference
border from Lake Turkana N to W Eritrea. May now low, relative to skull length, in the males; horns rela-
be extinct. tively short, as in A. swaynei (Cappellini & Gosling,
2006).
Alcelaphus swaynei (Sclater, 1892) An E African species, found from Lake Naivasha
Swayne’s hartebeest and the S flanks of Mt. Kenya SE through the S half of
Tsavo East National Park, and SW to the Mara; in
1892 Bubalis swaynei Sclater. Haud, 100 mi. from Berbera,
Tanzania as far S as the coast opposite Zanzibar, the S
Somalia.
border running W to the Wembere Plains, then NW
1905 Bubalis noacki Neumann. Suksuki River, S of Lake
-1— and N to Speke Gulf and the Serengeti. N of the W
Zwai, Ethiopia.
0— part of this range, from Lake Nakuru N through the
349-47558_ch01_1P.indd 210 5/31/11 7:05 PM

b o v idae 211
Rift Valley and W to Kavirondo and to the foot of Mt. Lake Victoria to the S part of the Rift Valley zone,
Elgon, there is (or was) a zone claimed to consist they tend to be more like A. cokii.
mainly of hybrids with A. lelwel. The region from the E side of Lake Turkana N to
the Dinder River (Ethiopia/Sudan border) was identi-
Alcelaphus lichtensteini (Peters, 1852) fied by Ruxton & Schwarz as a hybrid zone between
Lichtenstein’s hartebeest A. lelwel and A. tora, and they ascribed the following
Horn circumference, relative to skull length, in the named taxa to such hybrids:
males greater than in any other Alcelaphus species.
Bubalis neumanni Rothschild, 1897. E shore of
Greatest skull length 458–502 mm in the males; horn
Lake Turkana.
span 67%– 88% of the basal length in the males; least
Bubalis tora digglei Rothschild, 1913. Ofat River,
frontal width 79%– 86% of the biorbital width (Rux-
near Keili, upper Blue Nile.
ton & Schwarz, 1929).
Bubalis tora rahatensis Neumann, 1906. Shunfar
Characteristic species of the Brachystegia zone, from
Ambu and Shimmerler Jowee, W Ethiopia.
the border with A. cokii (and perhaps interdigitating
with it; see Kingdon, 1982b) to the Zambezi River in One of us (PG) has been inclined to think that the
the W and E, Zimbabwe in the E, and (formerly) to name Bubalis neumanni may actually designate a valid
KwaZulu-Natal, Mpumalanga, and Limpopo prov- taxon, rather than a hybrid population. This question
inces of South Africa. needs further investigation.
Ruxton & Schwarz (1929) maintained that the type
Alcelaphus caama (G. Cuvier, 1816) of Bubalus rothschildi Neumann, 1905 (Adoshebai Val-
red hartebeest ley, N of Lake Stephanie, Ethiopia), is probably a hybrid
Much the longest horns relative to skull length, and a between A. lelwel and A. swaynei, although they could
very heavy skull relative to its length, more so even not be certain, because the type skin had been mislaid;
than in A. lelwel (Cappellini & Gosling, 2006). Great- the description mentions the lack of black markings on
est skull length 464–505 mm in the males; horn span the face and the shoulders.
87% of the basal length in a single male, least frontal
width 75%–79% of the biorbital width (Ruxton & Beatragus Heller, 1912
Schwarz, 1929). 2n = 44 (Kumamoto et al., 1996).
From the Angolan border S to the Cape and SE Large preorbital glands.
to the Free State. Eliminated from most of this
range by the early 20th century, but now widely rein- Beatragus hunteri (Sclater, 1889)
troduced through much of it. herola
Butynski (2000) has given a thorough description of
Alcelaphus Hybrids this species, and what follows is largely based on his
Ruxton & Schwarz (1929) showed that hartebeest in account. Smaller and lighter than a hartebeest, with a
the NE African Rift Valley are a hybrid swarm be- more horizontal back. Horns not raised on a pedicle;
tween A. cokii and A. lelwel (Alcelaphus buselaphus jack- horns slender and lyrate, with long, upright tips. Pel-
soni in their arrangement): in absolute skull length, as age tawny, often with rufous tones, tending to darken
well as in relative horn span, they cover part of the with age, especially in the males; forehead, withers,
range of each of the parent species, and are more vari- dorsum, and lower segments of the limbs darker; un-
able than either. The authors maintained that the derparts whitish tawny; inside of the ears and most of
taxa described below actually came from the hybrid the long tail white, the terminal tuft mixed white and
zone: black; narrow, white eye-rings, connected by a white
chevron across the face. Shoulder height 95–100 cm;
Bubalis cokii kongoni Heller, 1912. Loita Plains, S
horn length of the males 450– 600 mm, of the females
Guaso Nyiro River.
350– 490 mm; horn span of the males 350– 400 mm, of
Bubalis lelwel keniae Heller, 1912. 20 mi. NE of
the females 240–330 mm; tip-to-tip distance about the
Nyeri, near Mt. Kenya.
same as the span.
Bubalis nakurae Heller, 1912. Lake Nakuru.
Butynski (2000) gave the known distribution (in
The hybrids, Ruxton & Schwarz (1929) stated, are about the middle of the 20th century) as between the
predominantly of the A. lelwel type to the N, from Tana and Juba rivers; on the Tana River, it extended as —-1
Lake Baringo to Mt. Kenya; whereas to the S, from far N as Garissa, and in Somalia to the Lag Der region —0
349-47558_ch01_1P.indd 211 5/31/11 7:05 PM

212 Artiodactyla
inland from Afmadu. By 1996, the range in Kenya had light saddle; less extensively white underparts. Skull
shrunk to about 42% of its former extent; its fate in So- length 304–328 mm; horn length 335–384 mm (both
malia is unknown. In 1963, and again in 1996, a number sexes); horn span 209–268 mm (both sexes).
of herola were introduced into Tsavo East National From the Karroo N to Botswana.
Park.
Damaliscus lunatus Group
Damaliscus Slater & Thomas, 1894 2n = 36 (for specimens from Transvaal; hence D. luna-
2n = 36 in D. lunatus and D. jimela, and 2n = 38 in D. pyg- tus); the X chromosome submetacentric (as analyzed
argus and D. phillipsi. The karyotype differs from that in Kumamoto et al., 1996).
of Beatragus by the fusion of acrocentric chromosomes Much larger species; (usually) dark midfacial strip;
13 and 15, and 20 and 22, and the presence of autoso- no white on the legs.
mal chain complexes, one having 5 chromosomes
and the other having 12 or 13 (Kumamoto et al., 1996). Damaliscus lunatus (Burchell, 1824)
Robinson et al. (1991) studied the chromosomes of hy- western tsessebe
brids between Damaliscus and Alcelaphus; these hybrids Smaller of the two species. Horns forming a semilu-
were sterile. nate profile. Color generally midbrown or pale brown-
In the control region phylogeny of Arctander et al. tan; facial blaze usually dark brown; shoulder- and
(1999), the D. lunatus group and E African D. jimela haunch-patches mostly light gray to warm gray.
formed quite separate clades, though within the for- Horn length of the males 256–391 mm, of the fe-
mer, D. lunatus and D. superstes were not entirely sepa- males 271–318 mm. Tip-to-tip distance 84– 415 mm
rate (Peter Arctander, pers. comm. to CPG). The sepa- (males), 190–364 (females); horn span 277– 448 mm
ration of D. pygargus from the other two clades was (males), 260– 430 mm (females). Greatest skull length
very deep. 365– 429 mm (both sexes); maxillary toothrow 80.7–
102.7 mm (both sexes). (Following Cotterill, 2003.)
Damaliscus pygargus Group Whether it is worth dividing this species into two
2n = 38; a fusion between chromosomes 18 and 24, subspecies is moot. On discriminant analysis (Cot-
and one of the chromosome chains having only 12 terill, 2003:figures 4c and 4d), Botswanan and Zim-
members instead of 13, differentiates the karyotype babwean samples have distinctly different “centres of
from that of D. lunatus and D. jimela (Kumamoto et al., gravity.”
1996). Zimbabwe, Botswana, the Caprivi Strip of Namibia.
Very small species, with a broadly white midfacial
region and white on the legs. Damaliscus superstes Cotterill, 2003
Bangweulu tsessebe
Damaliscus pygargus Pallas, 1767
2003 Damaliscus superstes Cotterill. Muku Muku Flats, 12°
bontebok
21' S, 30° E, SW Bangweulu Flats.
The reasons why Antilope dorcas Pallas, 1766, is to be
Larger than D. lunatus; horns thicker and wider spread,
rejected were given by Rookmaaker (1991) and supported
with broader pedicles. Horns growing symmetrically
by Grubb (2004a).
outward, with the tips curving inward, forming a
Face between the eyes white down to the snout; rump sphere. Color chocolate-brown, darker than D. luna-
white, extending around the base of tail to the legs and tus; facial blaze black; shoulder- and haunch-patches
the underside; a light saddle contrasting with the dark mostly dark gray.
brown pelage. Skull length 293–335 mm; horn length Horn length of the males 351– 400 mm, of the fe-
320–356 mm (males), 291–331 mm (females); horn span males 304–395 mm. Tip-to-tip distance 220–234 mm
222–246 mm (males), 197–227 mm (females). (males), 200–335 mm (females); horn span 378–446 mm
Known only from the S Cape. (males), 342–430 mm (females). Greatest skull length
390–435 mm (both sexes); maxillary toothrow 88.9–
Damaliscus phillipsi Harper, 1939 105.3 mm (both sexes). (Following Cotterill, 2003.)
blesbok The two species of tsessebe separate well on dis-
Face between the eyes white; snout white; buttocks criminant analysis (Cotterill, 2003:figures 4c and 4d).
-1— white, but not extending to the back of the rump; no Bangweulu district, NE Zambia.
0—
349-47558_ch01_1P.indd 212 5/31/11 7:05 PM

b o v idae 213
in D. tiang than in other taxa, although (on the more

Damaliscus korrigum Group limited data) there also appears to be a considerable
2n = 36 in D. jimela, differing from that of the D. pygar- sexual size difference in the horns of D. topi.
gus group by having one of the chain complexes in- Among females, the curiously small size of female
volving 13 chromosomes instead of merely 12; and D. selousi, barely larger than D. topi, is unexpected.
differing from D. lunatus in the X chromosome being Like the male skulls, the single specimen of a female D.
acrocentric (Kumamoto et al., 1996). eurus skull has long, broad nasals. Table 58 shows that
Since the two sexes seem very alike, we attempted the most approximated horn tips are in D. jimela. The
a discrimination between them craniometrically; N group has very robust horns. Relationships in horns
86.6% of the original grouped cases (85.4% of the cross- among the females are much like those in the males.
validated grouped cases) were correctly classified (47 Our division of this species-group into a surpris-
males, 35 females). Sexing depends strongly on the di- ingly large number of species is given below.
ameter of the horn bases, contrasted with the horn
span and the face length. Damaliscus korrigum (Ogilby, 1837)
Our taxonomic analyses gave the results described korrigum
below (see table 57). Bright orange-bay. Legs from the knees and the hocks
Broadly speaking, they divide into a N group (D. to the hoofs cinnamon; above, body color dark ashy
korrigum, D. tiang), characterized by extremely ro- brown, which fades on the shoulders and the haunches
bust horns, and the rest. into narrow, reddish gray patches, suff used with an
The three C African taxa distinguished by Schwarz ashy sheen. Facial blaze blackish gray. Horns very long
(1920), using primarily the Frankfurt series, but in- and robust, with the tips somewhat converging; skull
corporating a few other specimens from other collec- relatively broad.
tions, are not well differentiated. While 90% of the A wide distribution, from Senegal E to the Nile.
original grouped cases were correctly classified, this is As far as we can detect, there are no striking differ-
almost entirely dependent on the small sample sizes of ences, and certainly nothing that is in any way consis-
D. korrigum purpurescens (N = 5) and D. k. lyra (N = 2), tent, among different populations, even in C Africa
although the sample size for D. k. korrigum (N = 13) is (see above).
somewhat more substantial; and only 45% of the cross-
validated cases were correctly classified. We conclude Damaliscus tiang (Heuglin, 1863)
that no more than a single taxon can be recognized in tiang
C Africa. This, in turn, is not distinct from samples Color reddish bay, suff used with a reddish purple
from Nigeria and Senegal. bloom. Legs bright cinnamon. Shoulder- and haunch-
Sudan D. tiang is slightly more extreme than W-C patches ash gray, with a reddish tinge. Facial blaze
African D. korrigum; D. selousi, from the Guasin-Gishu, blackish gray, with a reddish tinge. In skull and horn
averages larger, with somewhat longer and narrower characters, slightly more extreme than D. korrigum,
nasals and slightly longer, less spreading horns. D. with somewhat more sexual dimorphism.
jimela is absolutely distinct from the N taxa in its Apparently E of the Nile, extending into W Ethio-
shorter, less robust horns, with slightly more divergent pia. D. tiang is undoubtedly closely related to D. kor-
tips, and it is narrower across the zygomatic arches. rigum; we have only a limited number of specimens
The more S groups, from E Africa, are better dis- from E Sudan, and it is possible that further material
tinguished from one another than are the N groups. may show that D. tiang is just an extreme clinal vari-
D. eurus differs from D. jimela and D. ugandae, particu- ant of D. korrigum.
larly in its very long, broad nasals. D. ugandae is wider
across the zygomatic arches, and has longer, more Damaliscus selousi Lydekker, 1907
spreading horns than D. jimela. The coastal species, Guasin- Gishu topi
D. topi, is much smaller and shorter horned, with a Males averaging larger than D. korrigum and D. tiang;
narrower span, than the others. nasals longer, narrower; horns slightly shorter, less
Sexual size difference is greatest in D. selousi, then spreading; sexual dimorphism the greatest in the
D. ugandae, then D. topi; both sexes of D. topi are much D. korrigum group, the females being unexpectedly
smaller than the others, which are all fairly similar in small in size. A pygargus-like white facial blaze has
size. The sexual size difference in the horns is greater been recorded in this species. —-1
—0
349-47558_ch01_1P.indd 213 5/31/11 7:05 PM

Table 57 Univariate statistics for the skulls of taxa in the Damaliscus korrigum group
Cb l Gt l Preorb Pal l Zyg br Nas l Nas br Teeth
Males
korrigum
Mean 391.42 402.42 266.00 221.00 132.09 166.15 36.04 93.80
N 33 36 35 29 37 37 37 37
Std dev 9.788 10.979 8.971 6.308 4.870 9.896 2.593 3.890
Min 374 383 248 210 120 144 31 84
Max 410 431 293 233 143 188 43 100
tiang
Mean 391.25 403.00 267.75 221.50 132.25 169.25 33.75 100.25
N 4 4 4 4 4 4 4 4
Std dev 8.302 6.377 3.096 3.109 .957 6.752 2.500 6.131
Min 382 398 265 219 131 163 31 94
Max 402 412 272 226 133 178 37 108
selousi
Mean 391.00 401.80 267.40 218.14 133.42 169.00 31.64 95.09
N 9 5 5 7 12 10 11 11
Std dev 9.539 12.458 10.213 7.426 4.188 11.116 3.529 5.029
Min 373 384 255 208 128 152 26 87
Max 400 414 283 230 140 184 38 102
jimela
Mean 383.32 393.00 261.33 214.28 128.00 169.95 36.32 92.68
N 19 9 9 18 19 19 19 19
Std dev 13.051 10.283 8.337 7.315 6.307 8.141 2.810 4.498
Min 363 376 246 202 118 153 32 86
Max 403 410 274 227 143 182 42 104
ugandae
Mean 395.63 409.94 273.44 226.13 136.67 177.88 36.13 96.67
N 16 17 16 15 18 16 16 18
Std dev 8.763 8.467 8.718 17.402 7.647 9.715 3.793 7.268
Min 374 392 258 207 128 161 30 83
Max 407 420 284 284 156 196 45 109
eurus
Mean 397.00 415.33 277.33 228.50 133.67 191.67 38.00 97.67
N 3 3 3 2 3 3 3 3
Std dev 17.349 19.553 12.583 7.778 2.309 9.292 1.732 5.686
Min 377 395 264 223 131 181 37 93
Max 408 434 289 234 135 198 40 104
topi
Mean 370.20 377.00 249.60 208.20 125.80 155.20 31.80 90.00
N 5 5 5 5 5 5 5 5
Std dev 12.029 12.708 9.529 7.563 1.304 8.228 1.304 4.950
Min 355 365 239 199 124 144 30 83
Max 386 393 261 216 127 167 33 96
Females
korrigum
Mean 381.71 394.48 262.95 217.47 130.80 158.90 34.65 94.14
N 17 21 21 19 20 20 20 21
Std dev 9.904 12.852 11.373 7.268 4.753 15.307 2.033 5.859
Min 370 378 249 207 124 127 32 78
Max 405 426 290 231 139 202 39 109
tiang
Mean 382.67 391.67 259.33 212.67 130.00 156.67 34.00 91.00
N 3 3 3 3 3 3 3 3
Std dev 18.448 20.817 13.650 12.503 7.000 11.060 4.583 1.732
-1— Min 367 375 247 200 125 145 29 90
0— Max 403 415 274 225 138 167 38 93
349-47558_ch01_1P.indd 214 5/31/11 7:05 PM

b o v idae 215
Cb l Gt l Preorb Pal l Zyg br Nas l Nas br Teeth
selousi
Mean 362.50 — — 205.00 123.00 147.00 29.50 91.50
N 2 — — 2 2 2 2 2
Min 360 — — 205 122 147 29 90
Max 365 — — 205 124 147 30 93
jimela
Mean 374.86 399.33 270.17 211.00 125.67 163.89 33.11 93.44
N 7 6 6 7 9 9 9 9
Std dev 21.836 12.044 12.703 8.406 4.272 15.439 2.934 6.579
Min 356 383 251 202 120 145 29 84
Max 420 417 288 225 134 198 38 104
ugandae
Mean 378.60 394.82 263.50 216.13 130.09 164.82 33.40 91.82
N 10 11 10 8 11 11 10 11
Std dev 10.047 15.329 14.246 6.770 7.162 10.989 4.300 5.326
Min 361 367 242 209 119 147 28 80
Max 401 420 286 228 146 181 40 98
eurus
Mean 383.00 402.00 272.00 215.00 127.00 175.00 35.00 94.00
N 1 1 1 1 1 1 1 1
topi
Mean 355.50 365.50 240.50 201.00 122.50 150.25 30.00 93.50
N 4 4 4 4 4 4 4 4
Std dev 4.796 7.047 7.000 6.055 3.000 9.179 .000 5.260
Min 351 359 233 194 119 140 30 86
Max 361 373 249 207 125 161 30 98
Known only from the Guasin-Gishu Plateau. The some respects intermediate between D. jimela and D.
species may be extinct. ugandae in skull characters, but with very long, broad
nasals.
Damaliscus ugandae Blaine, 1914 S Tanzania, in the Lake Rukwa / upper Ruaha re-
Uganda topi gion. It is unknown if there are any populations be-
Color much darker, maroon suff used with an ashy tween this area and the Serengeti ecosystem, where
sheen. Legs deeper cinnamon below the knees and D. jimela occurs.
the hocks; upper halves of the legs blue-black. Shoul-
der- and haunch-patches larger in area, steel-gray. Fa- Damaliscus jimela (Matschie, 1892)
cial blaze blue-black, but occasional pygargus-like Serengeti topi
white facial blazes have been recorded. Color not unlike D. topi (below), but shoulder-patches
Relatively smaller than D. korrigum and D. tiang but, larger, and shoulder- and haunch-patches more clearly
like them, comparatively broad-skulled, with relatively set off from the body color. Horns shorter, less robust
long, spreading horns; horns of D. ugandae, however, than other Damaliscus species, with the tips not very
considerably less robust. Sexual dimorphism relatively divergent; skull very long, narrow.
great, second only to D. selousi. Restricted to the Serengeti-Mara ecosystem.
Ankole district of SW Uganda, extending to the
Rutshuru Plains over the border in the DRC and to Damaliscus topi Blaine, 1914
Akagera National Park in Uganda. coastal topi
Color darker and richer than in D. tiang, heavily suf-
Damaliscus eurus Blaine, 1914 fused with a mauve bloom, becoming lighter on the
Ruaha topi belly. Facial blaze blackish gray, with a reddish tinge,
Size and color as in D. ugandae, but becoming lighter; and sprinkled with white hairs. Much smaller and —-1
bright reddish bay in the posterior dorsal region. In shorter-horned, with a narrower span, than other —0
349-47558_ch01_1P.indd 215 5/31/11 7:05 PM

216 Artiodactyla
Table 58 Univariate statistics for horn measurements

for northern taxa of Damaliscus Table 58
Span Tip–tip Diam Horn l str Span Tip–tip Diam Horn l str
Males jimela
korrigum Mean 230.00 148.40 53.17 348.17
Mean 314.46 234.83 68.01 485.50 N 5 5 6 6
N 35 35 36 34 Std dev 14.816 33.329 5.345 19.094
Std dev 29.955 50.275 5.279 46.300 Min 218 111 50 322
Min 242 106 57 377 Max 251 191 64 367
Max 375 329 84 553 ugandae
tiang Mean 263.40 184.70 53.70 363.90
Mean 331.50 200.25 68.20 525.20 N 10 10 10 10
N 4 4 5 5 Std dev 13.159 20.870 4.945 21.037
Std dev 51.733 49.742 12.398 56.962 Min 236 141 49 330
Min 271 137 51 468 Max 284 210 67 389
Max 389 257 78 600 eurus
selousi Mean 215.00 122.00 51.00 348.00
Mean 254.14 190.29 61.71 406.71 N 1 1 1 1
N 7 7 7 7 topi
Std dev 16.906 30.707 6.047 34.067 Mean 209.25 134.25 46.50 390.00
Min 217 145 53 383 N 4 4 4 4
Max 266 228 69 478 Std dev 7.136 18.464 1.915 11.916
jimela Min 201 109 44 374
Mean 234.33 154.33 57.67 336.89 Max 218 153 48 402
N 9 9 9 9
Std dev 10.296 25.199 3.640 19.310
Min 217 116 51 305
Max 254 185 63 377
ugandae species of the D. korrigum group, but sexual dimor-
Mean 275.25 190.83 62.92 377.23 phism greater than the other E African species, espe-
N 12 12 13 13 cially in horn size.
Std dev 13.752 21.114 6.103 23.682
Coastal E Africa, from the Juba region of Somalia
Min 245 149 49 345
Max 302 220 71 421 S at least to Malindi. The range is isolated from that
eurus of other taxa.
Mean 258.67 194.00 62.33 329.67
N 3 3 3 3 Connochaetes Lichtenstein, 1814
Std dev 20.133 21.284 1.528 32.146 wildebeest, gnu
Min 240 175 61 293
2n = 58; the X chromosome is a large acrocentric, the
Max 280 217 64 353
Y chromosome a small acrocentric. On the basis of
topi
Mean 224.40 179.60 55.20 405.00 mtDNA, a divergence date between C. taurinus and C.
N 5 5 5 5 gnou of somewhat over 1 Ma was calculated, assum-
Std dev 16.742 24.846 5.718 15.379 ing a divergence rate of 2% per million years (S. W.
Min 199 142 46 379 Corbet & Robinson, 1991).
Max 239 203 61 419
Quite different in appearance and posture
Females from other genera of the tribe: the head is large, held
korrigum
low, with a convex nose on which there is a brush of
Mean 278.00 181.19 56.00 453.67
N 21 21 21 21 short, stiff hair, and huge valvular nostrils. The
Std dev 24.187 29.223 2.933 26.949 horns are thickened and expanded at the base, and
Min 207 125 51 402 oriented either sideways or forward, curved down-
Max 334 232 63 514 ward and then up, not upright like other genera.
tiang Usually, two species are recognized, C. gnou and
Mean 257.33 177.00 59.00 436.00
C. taurinus, but within the latter there are actually
N 3 3 3 3
-1— Std dev 49.943 110.937 9.849 46.228 four sharply distinct species: C. taurinus, C. johnstoni,
0— Min 228 94 48 383 C. albojubatus, and C. mearnsi. Interestingly, C. mearnsi,
+1— Max 315 303 67 468
349-47558_ch01_1P.indd 216 5/31/11 7:05 PM

b o v idae 217
like Damaliscus jimela, is apparently endemic to the ish blue-gray in C. taurinus), with the face black (the
Serengeti-Mara ecosystem. forehead brownish white in the juveniles).
We thus recognize five species altogether. According to Roosevelt and Heller (1914), C.
mearnsi “differs from albojubatus by the decidedly
Connochaetes gnou (Zimmermann, 1777) darker legs, which are olive brown or sepia in old
white-tailed wildebeest, black wildebeest males, and somewhat lighter in females. Body tends
Small; black; white tail; forward-swung horns. to become darker or quite blackish on chest, shoul-
Formerly from the Karroo N to the Orange and ders, and sides.”
Vaal rivers; now preserved in only a few farms and For Estes (1969), C. mearnsi (which he called “C.
reserves. hecki”) is the only geographic form that really stands
out: smaller, darker, with a longer tail; horns with a
Connochaetes taurinus Group more pronounced boss and a shorter spread; territo-
Arctander et al. (1999) found almost complete separa- rial males more vociferous, “with a call of a different,
tion between the four species in the C. taurinus group. less metallic timbre.”
The authors also had two samples from the Luangwa In C. albojubatus, and C. mearnsi, the mane along
Valley (the taxon described as C. cooksoni), of which the nape hangs limp, whereas it is erect in C. taurinus.
one clustered with C. mearnsi, the other with C. john- Estes had not, at the time, studied any populations of
stoni. Two C. mearnsi (out of a large sample) clustered C. johnstoni. From photos (and as later specified by
with C. albojubatus, and one C. taurinus clustered with Estes, 1991), the mane in C. johnstoni stands up like
C. mearnsi. that of C. taurinus, and it is lighter gray in color, with
Table 59 gives univariate statistics for wildebeest a brownish tinge, the legs are light-colored, the tail is
of the whole C. taurinus group (we did not especially short. A white crescent across the face below the eyes
consider C. gnou). often occurs in C. johnstoni; indeed, this was the chief
The smallest species in skull length is C. mearnsi, distinguishing feature mentioned in the type descrip-
but the skull is hardly narrower than in the other spe- tion, but its presence is inconstant.
cies of Connochaetes. The longest nasals are in C. john-
stoni; the other species are all about the same. The Connochaetes taurinus (Burchell, 1824)
two S species (C. taurinus and C. johnstoni) have much brindled wildebeest, blue wildebeest
wider spans than the two N ones (C. albojubatus and
1893 Connochaetes reichei Noack.
C. mearnsi), and C. mearnsi has an extremely narrow
1925 Connochaetes mattosi Blaine.
span, although the distance between the tips is not
1933 Connochaetes borlei Monard.
much different (i.e., the tips do not turn inward as
much). N South Africa, presumably S Mozambique, Zimba-
In an effort to quantify the relative degree of down- bwe, Botswana, W Zambia.
turn in the horns, we sat each skull on a table and mea-
sured, where possible—which was only in a relatively Connochaetes johnstoni Sclater, 1896
few individuals—the distance from the bottom of the Johnston’s wildebeest
curve of one horn to the table top. This averages only
1911 Connochaetes rufijianus de Beaux.
45 mm in C. taurinus (N = 10) and 50 mm in C. johnstoni
1914 Connochaetes cooksoni Blaine.
(N = 3), but 69 mm in C. albojubatus and C. mearnsi
(N = 5, in total). In C. taurinus, the horns can actually E of the Luangwa River, Zambia; Malawi, N Mozam-
sweep down to touch the table top; in C. johnstoni, the bique, SE Tanzania.
minimum distance is 35 mm; and in the other two, the
minimum is 58 mm. Size and shape relationships Connochaetes albojubatus Thomas, 1892
among females are very much as they are in the males, eastern white-bearded wildebeest
but C. mearnsi females have exceptionally short nasals.
1905 Connochaetes hecki Neumann.
The span is least in the single female specimen of C.
1919 Connochaetes babaulti Kollman.
johnstoni.
The supposed taxon C. babaulti was described on the basis of
In the two S species, the beard is black; in the two N
an unusual individual with a black beard.
ones, it tends to be creamy white. All have a pattern of
thin, dark stripes on the neck, fading out along the Athi Plains in S Kenya, S to Tanzania, and W to the N —-1
flanks. The color is some shade of gray (generally dark- shores of Lake Tanganyika. —0
349-47558_ch01_1P.indd 217 5/31/11 7:05 PM

218 Artiodactyla
Table 59 Univariate statistics for the Connochaetes taurinus group
Gt l Gt br Nas l Teeth Span Tip–tip
Males
taurinus
Mean 475.93 182.76 216.66 108.81 641.15 381.51
N 41 42 41 40 41 35
Std dev 22.517 7.556 18.920 6.706 53.860 56.936
Min 433 163 186 92 519 270
Max 524 199 291 123 740 478
johnstoni
Mean 498.73 184.73 234.55 112.44 658.73 387.30
N 11 11 11 9 11 10
Std dev 16.402 4.077 14.201 5.981 26.127 60.988
Min 470 176 215 104 628 281
Max 520 191 252 124 718 459
albojubatus
Mean 467.40 182.44 209.13 105.67 594.50 355.78
N 10 9 8 9 10 9
Std dev 13.786 10.806 12.722 4.123 61.562 64.591
Min 442 159 193 100 488 294
Max 493 194 234 112 670 476
mearnsi
Mean 449.22 179.44 212.17 103.00 543.78 342.22
N 9 9 6 9 9 9
Std dev 10.779 8.457 8.134 3.464 28.146 43.746
Min 430 160 201 95 502 283
Max 458 189 222 106 600 443
Females
taurinus
Mean 467.77 171.77 211.62 102.58 493.38 275.25
N 13 13 13 13 13 8
Std dev 15.216 4.323 13.131 27.948 61.630 60.363
Min 445 164 190 11 333 215
Max 490 182 231 120 588 379
johnstoni
Mean 480.00 176.00 218.00 103.00 418.00 175.00
N 1 1 1 1 1 1
albojubatus
Mean 456.00 167.00 215.00 98.00 439.00 369.00
N 1 1 1 1 1 1
mearnsi
Mean 421.20 162.00 185.00 99.40 432.20 267.40
N 5 5 4 5 5 5
Std dev 13.627 5.701 8.287 6.950 20.548 29.821
Min 406 154 179 90 414 225
Max 442 169 197 105 462 299
Connochaetes mearnsi (Heller, 1913) herds, both from Ngorongoro Crater, supposedly
Serengeti white-bearded wildebeest from the N and S parts of it, respectively(!), though
there does not seem to be any real evidence that this
1913 Connochaetus albojubatus henrici Zukowsky.
was so.
1913 Connochaetus albojubatus lorenzi Zukowsky.
1913 Connochaetus albojubatus schulzi Zukowsky.
Tribe Caprini Gray, 1821
-1— The putative taxon C. henrici was from the Seren- Characterized by the short and anteroposteriorly
0— geti; C. lorenzi and C. schulzi were based on captive compressed metapodials, the (almost always) horned
349-47558_ch01_1P.indd 218 5/31/11 7:05 PM

b o v idae 219
females, the absence of lateral prongs on the nasal by Hassanin et al. (1998), but in this case Oreamnos
tips, the fusion on the posterior premolar of the meta- also formed part of the Ovis clade. In addition, these au-
conid and the entoconid, the strong mesostyles on the thors studied Ammotragus and Pseudois, which formed
upper molars, and the presence of small additional part of the Capra clade.
cavities on the lingual side of the upper molars. (Fol- A study of a few taxa in the Caprini by chromo-
lowing A. W. Gentry, 1992.) some painting (Huang et al., 2005) confirmed that se-
The Caprini is now taken to include Pantholops, as row (Capricornis) and goral (Nemorhaedus) form a group
well as those genera formerly included in a tribe of when compared with goats and sheep, and these au-
their own, the Rupicaprini, which, in fact, have little thors traced the chromosome changes that have oc-
in common with each other except in lacking the cra- curred in the evolution of members of the serow/goral
nial specializations of the “central” genera—Ovis, group.
Capra, and their relatives. Lalueza-Fox et al. (2005), in the course of a study
In Ovis, Capra, and related genera, the morphol- on the extraction of mitochondrial nuclear sequences
ogy of the skull is correlated with that of the horns, from the remains of Myotragus, generated an entire
which, in turn, is derived from the different fight- phylogeny of the Caprini, as follows (adopting their
ing styles of the males, according to Schaffer & Reed Bayesian tree for the concatenated mitochondrial
(1972). The extension of the sinuses is likewise: segments):
extensive throughout the frontals, with complex
((((Capra, Hemitragus), Pseudois) Ammotragus
septa, in Ovis, Hemitragus, Ammotragus, and Pseudois,
(Ovis, Myotragus) Oreamnos (Rupicapra (Capricornis,
but much less so in Capra, except that the sinuses ex-
Ovibos) Budorcas) Pantholops)).
tend throughout the horns, but lack complex septa,
in C. aegagrus. In other words, there is a virtual five-way split sub-
On the basis of chromosome morphology, Bunch sequent to the separation of Pantholops. But the au-
& Nadler (1980) postulated the following phylogeny thors went on from this to propose—using the
of those caprines studied by them: 5.35 Ma separation of Myotragus from Ovis (based
on geological data)— a general dating of the well-
(Capra (Pseudois (Hemitragus
supported nodes on the tree: the Ovis canadensis
(Ammotragus (Ovis))))).
group from other Ovis at 2.2 ±0.3 Ma; Rupicapra pyre-
Hartl et al. (1990) studied proteins of the Caprini naica from R. rupicapra at 1.6 ±0.3 Ma; Capra ibex from
and the Rupicaprini simultaneously, discovering that C. pyrenaica at 0.6 ±0.1 Ma, and these from other
relationships among the genera of these two putative Capra (except for C. sibirica, which separated much
tribes were more complicated than had hitherto been earlier) at 1.5 ±0.2 Ma; and, fi nally, Pantholops from all
supposed. The authors generated an interesting clado- other Caprini at 6.2 ±0.4 Ma.
gram (their figure 4), which showed the following Ropiquet & Hassanin (2005) showed that the genus
branching pattern: Hemitragus is polyphyletic; they restricted the genus to
the type species, Hemitragus jemlahicus, and erected
(Ovis ((Rupicapra, Oreamnos) (Hemitragus
two new genera (Nilgiritragus and Arabitragus) to con-
(Ammotragus (Capra))))).
tain the other two species. They proposed that the
P. Groves & Shields (1996) sequenced the mitochon- Caprini had begun to diversify between 8.7 and
drial cytochrome b gene of the Caprini, with par ticu- 11.9 Ma, that is, much earlier than the date proposed
lar reference to Ovibos and Budorcas, with a view to by Lalueza-Fox et al. (2005); the next split was between
testing whether these two genera, in par ticu lar, are Rupicapra and the rest, at between 7.5 and 10.4 Ma; fol-
related. Though different methods of analysis gave lowed by a division into two clades: Ovis plus Nilgiri-
somewhat different trees, a consensus of six trees, ac- tragus (they did not study Myotragus), and Capra plus
cepting as significant nodes only those with 67% or all remaining genera (Budorcas, Oreamnos, Ovibos, Am-
more support, was as follows: motragus, Arabitragus, Pseudois, Hemitragus, and Capra).
Later, the same authors (Ropiquet & Hassanin, 2006),
(Saiga (Nemorhaedus, Capricornis, Ovibos, Oream-
in the context of demonstrating the hybrid origin
nos) (((Hemitragus, Capra) (Ovis aries (Budorcas
of Capra, proposed some new dates, differing only
(Ovis canadensis group)))).
slightly from the previous ones.
Much the same type of arrangement—with Budorcas Finally, Ropiquet (2006) gave an updated version —-1
as sister to Ovis, and Hemitragus to Capra—was found of this, recognizing the initial split (after the separation —0
349-47558_ch01_1P.indd 219 5/31/11 7:05 PM

220 Artiodactyla
of Pantholops) as an undifferentiated comb with six wise) of subspecies, but note that further morphologi-
branches: Oreamnos, Budorcas, (Ammotragus, Arabitragus, cal study, as well as DNA study, is needed.
Hemitragus, Pseudois, Capra), (Nilgiritragus, Ovis), Rupi-
capra, and (Nemorhaedus, Capricornis, Ovibos). The six- Budorcas Hodgson, 1850
way split is dated at about 8.5 Ma, and the split from takin
Pantholops at about 9.5 Ma. Diversification within each Though it has generally been linked with Ovibos into a
of the three multigeneric branches of the six-way split common subfamily or tribe, Budorcas is more closely
began in the Pliocene, mostly at the beginning of this related to Ovis, while Ovibos is sister to Nemorhaedus (P.
period. Groves & Shields, 1997).
Our descriptions and analysis below are based on
Pantholops Hodgson, 1834 observations not only of museum specimens, but of
A. W. Gentry (1992) was the first to definitively sepa- living, known-age animals, especially in the Shang-
rate this genus from Saiga, with which it had been tra- hai Zoo breeding farm and the Tierpark in Berlin.
ditionally placed, and include it in the Caprini. He There are striking age changes in both pelage and
pointed out that Pantholops shares various features horns; these are far more marked in B. taxicolor and B.
with this tribe, in particular, the long, mediolaterally tibetanus than in the other two taxa.
compressed, keeled horn cores; the frontal sinuses; the Table 60 gives the univariate statistics for samples
absence of lateral prongs on the nasal tips; the zygo- of Budorcas. Skull and horn characters appear to offer
matic arch expanded below the orbit; and the weak very little differentiation; perhaps the main varying
distal flanges of the metatarsals. His demonstration character concerns the nasals, which are, on average,
that Pantholops is a caprin has since been corroborated shorter and less arched (“Pal nas ht,” i.e., the height of
by Gatesy et al. (1997), followed by Lei, Jiang, et al. the most convex part of the nasals above the palate)
(2003), Ropiquet & Hassanin (2005), and all other mo- in B. bedfordi than in others.
lecular studies that included this genus. Pantholops is, Li M. et al. (2003) studied the molecular phylogeny
nonetheless, in an isolated position; thus it is sister to of the genus, using portions of the control region and
the rest of the Caprini. cytochrome b. They had one sample of B. taxicolor from
Gongshan; nine of B. tibetana from Sichuan and four
Pantholops hodgsonii (Abel, 1826) from Bailongjiang in Gansu; and 26 of B. bedfordi from
chiru Shaanxi. The three taxa separated well (with B. taxi-
The available information on this monotypic genus color as sister to the other two), except that the Gansu
has been summarized by Leslie & Schaller (2008). sample of B. tibetana, curiously, fell into the B. bedfordi
clade; but, as the authors pointed out, Bailongjiang is
Oreamnos Rafinesque, 1817 actually on the W edge of the Qinling Range, and takin
2n = 42, with 18 metacentrics and 22 acrocentrics, from there have, at times in the past, been classified as
quite different from the high numbers (with few or B. bedfordi (although the authors gave no phenotype
even no acrocentrics) of other “rupicaprins” (Soma & data for the specimens which yielded the samples).
Kada, 1986). Based on our morphological research, backed up
The characters of this genus were described in de- by the DNA study of Li M. et al. (2003), we classify the
tail by Dolan (1963). genus into four species.
Oreamnos americanus (de Blainville, 1816) Budorcas taxicolor Hodgson, 1850

Rocky Mountain goat Mishmi takin
A full description of this species—its general biology, Juveniles, up to about 6 months of age, light brown,
distribution, behavior, ecology, and genetics—will be with an extensive black zone on the haunches, the
found in Rideout & Hoff mann (1975). belly, the legs, the lower neck, and the head, especially
Dolan (1963) discussed the validity of the described the face. Black extends with age to the whole of the
subspecies, recognizing only O. a. kennedyi as differ- face, under the neck, and to the lower flanks; mixed,
ent from O. a. americanus, distinguished mainly by its brindled zone between these areas and the lighter
much larger size, constricted forehead in front of the (pale yellow) tone of the upper parts. Skulls of the
horn bases, and perhaps more outwardly curved males much larger than those of the females; horns
-1— horns. We take no position on the existence (or other- also sexually dimorphic.
0—
349-47558_ch01_1P.indd 220 5/31/11 7:05 PM

Table 60 Univariate statistics for Budorcas
Gt l Biorb Span Tip–tip Teeth Braincase Nas l Nas br Pal nas ht
Males — — — — — — —
bedfordi
Mean 414.33 181.83 384.67 295.67 114.00 93.50 131.50 54.00 118.50
N 3 3 3 3 2 2 2 1 2
Std dev 25.027 11.751 71.501 35.642 — — — — —
Min 390 170 303 256 112 88 117 — 114
Max 440 194 436 325 116 99 146 — 123
Burma
Mean 414.00 198.00 409.50 326.00 111.00 102.00 159.00 — 139.00
N 1 1 2 2 1 1 1 — 1
Min 414 198 393 302 111 102 159 — 139
Max 414 198 426 350 111 102 159 — 139
taxicolor
Mean 394.50 190.17 371.80 247.40 118.00 100.20 155.17 75.50 137.50
N 2 6 10 10 5 5 6 6 6
Std dev — 11.197 24.535 53.066 3.391 4.087 6.616 7.662 7.342
Min 384 168 341 160 114 93 145 62 126
Max 405 198 411 318 122 103 162 85 146
tibetana
Mean 428.17 192.63 406.09 299.45 119.20 95.80 145.25 68.33 140.80
N 12 16 11 11 5 5 4 3 5
Std dev 18.235 11.087 42.801 47.744 6.017 3.564 11.615 15.885 13.953
Min 392 170 298 226 109 92 135 50 121
Max 448 209 459 366 124 101 160 78 156
whitei
Mean 420.00 176.19 319.56 247.17 120.16 103.80 145.90 67.00 123.00
N 2 7 8 6 5 5 5 2 3
Std dev — 11.745 22.579 24.815 3.777 13.554 8.706 — 14.177
Min 413 161 297 202 114 97 132 52 112
Max 427 197 357 275 124 128 156 82 139
Females
bedfordi
Mean 394.40 166.80 301.75 204.00 119.67 92.00 130.00 57.00 115.00
N 5 5 4 4 3 3 3 3 2
Std dev 2.881 4.087 5.377 19.026 2.887 1.732 2.646 .000 —
Min 391 162 295 185 118 90 128 57 115
Max 398 172 307 230 123 93 133 57 115
Gongshan
Mean 376.50 168.00 279.00 199.50 109.00 87.00 129.50 — 159.00
N 2 2 2 2 2 2 2 — 2
Min 375 166 276 192 105 86 128 — 122
Max 378 170 282 207 113 88 131 — 196
taxicolor
Mean — 176.00 281.33 215.33 117.00 87.00 137.00 58.00 121.00
N — 1 3 3 1 1 1 1 1
Std dev — — 8.505 2.517 — — — — —
Min — — 275 213 — — — — —
Max — — 291 218 — — — — —
tibetana
Mean 394.00 162.40 287.00 201.75 119.00 90.50 125.00 107.00
N 3 5 4 4 1 2 1 — 1
Std dev 3.606 6.950 47.117 23.200 — — — — —
Min 391 153 218 175 — 89 — — —
Max 398 168 324 231 — 92 — — —
whitei
Mean 408.00 172.40 280.75 149.00 120.80 — 140.75 — —
N 1 2 2 1 1 — 2 — —
Min — 168 278 — — — 140 — — —-1
Max — 177 284 — — — 142 — — —0
—+1
349-47558_ch01_1P.indd 221 5/31/11 7:05 PM

222 Artiodactyla
Mishmi Hills, northernmost Burma, to Tengyueh

in Yunnan. Specimens from Gongshan are consider- Ammotragus lervia Pallas, 1777
ably smaller, but otherwise fit into this taxon. This species (if it is a single species) appears to be re-
stricted to mountain ranges in the Sahara and the
Budorcas whitei Lydekker, 1907 Maghreb, a distribution which would suggest that
Bhutan takin there should be strong geographic differentiation; but
Juveniles and adults both similar to B. taxicolor juve- we have little knowledge of the described species
niles in pelage; horns seemingly remain small, with a and/or subspecies, and we know only that it requires
narrow span, through life; males small in size; teeth investigation in the future.
relatively large.
Bhutan and SE Tibet. Arabitragus Ropiquet & Hassanin, 2005
Arabian tahr
Budorcas tibetanus Milne Edwards, 1874 2n = 58.
Sichuan takin Braincase behind the horn very short, descending
Golden-brown when young, the dark areas much less at about a 45° angle. Tympanic tube very long, point-
extensive than in B. taxicolor, and dark brown rather ing laterally, as in Ammotragus; external auditory me-
than black. Color continues to darken with age; dark atus a very large, vertical oval. Frontal region very
golden-brown by the fourth year, with a conspicuous convex. Orbital rim merging with the zygoma. Naso-
pale saddle, the dark brown zones extending onto the frontal suture very broad, often slightly concave anteri-
lower flanks, the lower part of the legs, and between orly. Nasal bones relatively short, with very small lat-
the eyes; neck mane present in the adults. Black on eral processes. Frontolacrimal suture running medially,
the nose becoming extremely conspicuous with age. then angled, then running forward. Ethmoid fissure
Aged females more contrasty than the males, the present, as in Ammotragus. Nasal branch of the premax-
light areas becoming pale yellow-gray and the dark illa very narrow, widely separated from the nasal.
areas black. Colors somewhat darker in winter, The Arabian tahr is sister to Ammotragus, not to
browner, with the dark areas more blackish, than in other species of tahr, as was previously assumed (Ropi-
summer. Sexual dimorphism obvious in the skull and quet and Hassanin, 2005).
the horns.
Sichuan. Arabitragus jayakari (Thomas, 1894)
1894 Hemitragus jayakari Thomas. Jebel Taw, Oman.
Budorcas bedfordi Thomas, 1911
golden takin Diminutive species; skull length 201–219 mm; horns
Young animals light golden-brown, with dark zones longer, more slender, and less ridged than in Hemitra-
only on the haunches; face wholly light-toned. With gus, but much shorter than in Ammotragus. Hair long
age, color a somewhat more red-gold tone, developing on the flanks and the nape in the males; nape hair
a paler golden saddle, but the face, in particular, never developing in the males into a black-tipped dorsal
becoming dark. Nasal region less inflated than in other crest, this crest pale in the females. Hair much more
samples of Budorcas, and less sexually dimorphic. shaggy in winter, forming ruff s on the legs. Light
Qinling Range. sandy brown; underside buff y whitish. Midline of the
face broadly black; black stripe from the eye to the
Ammotragus Blyth, 1840 mouth; pale brown stripe in between.
Barbary sheep Mountain ranges of E Oman, barely extending into
2n = 58. the United Arab Emirates.
Braincase descending steeply behind the horn.
Tympanic tube long, pointing laterally; external au- Hemitragus Hodgson, 1841
ditory meatus relatively small, fairly circular. Fron- Himalayan tahr
tals somewhat convex. Orbital margin merging with 2n = 48.
the zygoma. Nasofrontal suture broad, often concave Braincase behind the horn descending steeply,
backward. Anterior nasals long, with a trace of the almost vertically. Frontals slightly convex; zygoma
lateral processes. Frontolacrimal suture angulated, as separated from the orbit by a sharp orbital rim, with
-1— in Arabitragus. Ethmoid fissure present. Nasal branch a trough below. Frontolacrimal suture forming a
0— of the premaxilla narrow, but approaching the nasal. simple curve. No ethmoid fi ssure. Nasal branch of
349-47558_ch01_1P.indd 222 5/31/11 7:05 PM

b o v idae 223
the premaxilla very broad, forming a broad suture Arid, grassy slopes of the upper Yangtze Gorge,
with the nasal. Underside of the tail naked, like a below the forest zone, from a little N of Batang and
goat. extending for an unknown distance to the S.
Habitats of plateau and dwarf blue sheep are sepa-
Hemitragus jemlahicus (Hamilton Smith, 1827) rated by a belt of subtropical forest, at about 1000 m,
Large; heavily built; very sexually dimorphic. Horns and this almost (or completely) isolates the two popu-
compressed, nearly in contact at the base and then lations. Sequences of the mitochondrial control re-
diverging, with the anterointernal edge raised into a gion in four specimens of each showed an absolute
sharp keel. Body hair long, shaggy, especially on the difference of more than 12%; if a rate of divergence of
forequarters in the males, but short on the head. 6%–12% per million years is followed, then they sep-
Color reddish or dark brown, darkening in old males; arated 1–2 Ma (Feng et al., 2001). The same authors
mane of old males becoming whiter. found that, in contrast, ZFY intron sequences were
Himalayas, from Kashmir E to Sikkim. not completely separate, which suggests that the oc-
casional male greater blue sheep may make his way
Pseudois Hodgson, 1846 through the forest belt and then, dominating dwarf
blue sheep blue sheep rams, mate with the dwarf ewes.
2n = 54.
Taxonomy follows Groves (1978a). Capra Linnaeus, 1758
goats
Pseudois nayaur (Hodgson, 1833) The species of the genus Capra divide easily into
greater blue sheep three phenetic groups: true goats, markhor, and ibex.
Large size; horns initially with backward-pointing They may be distinguished as follows:
tips, then, from about 10 years of age in adult males,
1. True Goats—The horns are sickle-shaped,
curving around in a semicircle; long fur, with much
with a plane to heteronymous spiral, slender
underwool in winter; winter coat brown-gray, with a
and bilaterally compressed, with a sharp,
bluish cast; dark brown face-mask; black stripes down
irregular keel along the front edge, tending to
the foreleg and the hind shanks, sometimes down the
break off between the annual growth rings; at
whole hindleg, and along the flanks; underside white,
its insertion, the keel extends downward,
occasionally golden-brown.
below the plane of the horn base; the cores are
Skull length of the males in different populations
smooth, with a strong anteromedian keel, and
averaging 252–266 mm; horn span in different popu-
compressed, the transverse diameter of the
lations averaging 596–734 mm; basal diameter of the
core being about 70% of the anteroposterior
horn core >80 mm, usually >90 mm.
diameter. The facial profi le is strongly concave
Plateau country of the S and E rims of the Tibetan
overall; the base of the nasals and the frontals
Plateau, from E Kashmir—via Nepal, Bhutan, the Hi-
rises up to form a sharp “forehead” anterior to
malayan regions of India, and S Tibet—to Yunnan,
the orbits, with a convexity at the base of the
Sichuan, and Gansu.
horns, so that the upper rims of the orbits do
not interrupt the dorsal profi le, and the upper
Pseudois schaeferi Haltenorth, 1963
margin of the rostrum is concave. The skull is
dwarf blue sheep
relatively narrow; biorbital breadth is about
Males only half the weight of P. nayaur, the females
50% of the skull length. The nasals are short
less reduced in size; horns more upright, with simpler,
(about 31% of the skull length) and expanded
upturned tips, not forming semicircles; shorter fur,
posteriorly (their posterior breadth is about
with less underwool; drabby in winter, with a silvery
44% of their length), forming a triangular
sheen; dark markings poorly developed, except in old
suture with the frontals; their tips, in lateral
males; limbs dark overall; agouti bands on the hairs
view, are horizontal. The ethmoid fissure
short, soon wearing off in the winter coat; underside
remains wide throughout life. The premaxilla
less white.
forms only a short (about 10 mm) suture with
Skull length of the males averaging 236.3 mm
the nasal; the premaxillary tips are not (or
(maximum 247 mm); horn span averaging 543 mm
hardly) expanded. The braincase is very long
(maximum 566 mm); basal diameter of the horn core —-1
and horizontal; the supramastoid crest is not
<80 mm. —0
349-47558_ch01_1P.indd 223 5/31/11 7:05 PM

224 Artiodactyla
flared. The toothrows are gently outbowed. long (about 50 mm) suture with the nasals; the
The median palatine notch tends to be narrow, premaxillary tips are not expanded. The
extending forward to about halfway along M3, median palatal notch is fairly narrow and
about the same distance as the lateral notches. reaches forward to halfway along M3, much
The basicranium is narrow; the pterygoid anterior to the level reached by the lateral
plates are straight and vertical (as also in Capra notches. The paroccipital processes are rather
nubiana); the paroccopital processes are fairly shorter; the supramastoid crest is not flared.
long, reaching to the level of the lower margin Externally, markhor are characterized by their
of the condyles. Externally, there is (in wild long coats, with a throat fringe as well as a
forms) a strongly expressed body pattern, with beard, and poorly expressed dark markings.
a dark shoulder-stripe and dark face markings, They are among the largest species of the
as well as the usual black-and-white leg genus, and have among the longest horns.
markings. True (wild) goats are among the 3. Ibex—Ibex skulls have some similarities to
smallest species of the genus, but their horns markhor, as listed above, but they differ in
are among the longest. many respects. The horns are not spiraled, or
2. Markhor—There is no precornual convexity, not tightly so, and typically are broad-fronted,
and the facial profi le is not markedly concave with regularly spaced knots on the frontal
in front of the orbits; the orbits usually inter- surface; the cores lack keels or grooves (though
rupt the dorsal profi le. The nasals are longer, there is comparatively little difference, in some
sloping downward, and tending to turn down cases), and they are hardly or not at all com-
at the tips. The braincase is short, sloping pressed, the transverse breadth of a core always
downward somewhat. The palate is broad, so being more than 74% of its anteroposterior
that the toothrows tend to be bowed outward diameter. There is no convexity at the horn
more. The basicranium is broad, especially base, but the facial profi le is concave in the
across the condyles, with small auditory interorbital region (the orbits may or may not
bullae. In all these features, markhor resemble interrupt the line of the profi le). The skull
ibex; but there are, of course, many differ- vertex is between the horn cores. The form
ences. The horns are spirally twisted and of the nasals and the premaxillae varies; the
flattened, with a rounded anteromedial nasopremaxillary suture is intermediate in
surface and a sharper posterior keel; insertion length between true goats and markhor. The
of the horns is not in one plane, but folded over ethmoid fissure is narrow, but usually remains
the vertex of the skull; the cores are com- open. The supramastoid crest is flared.
pressed, their transverse basal diameter only
60% of the anteroposterior diameter, and their
How Many Species?
median surfaces are longitudinally grooved (as
There is presumably only one (wild) species in the
also in Capra pyrenaica), bounded anteromedi-
true goat group: the earliest name for any member of
ally by a weak keel and posteromedially by a
this group is Capra hircus Linnaeus, 1758, and that for
stronger keel; the frontal-core angle is more
a wild goat is Capra aegagrus Erxleben, 1777; wild and
obtuse than in any other species. The facial
domestic goats are here kept apart.
profi le is nearly straight (or, at most, gently
At the moment only a single species of markhor is
concave) from the horn bases to the interor-
recognized: Capra falconeri (Wagner, 1839), but this
bital region. The vertex of the skull is rela-
may have to be modified.
tively far back, at the posterior margin of the
The number of species of ibex has long been dis-
horn core bases. The biorbital breadth is about
puted. Generally, Western authors followed the lead
52% of the skull length (as in Capra ibex). The
of Ellerman & Morrison-Scott (1951) in assigning all
nasals are relatively short (about 37% of the
except the Spanish ibex, Capra pyrenaica Schinz, 1838,
skull length), but broad (the breadth 36% of
and the Daghestan tur (which Ellerman & Morrison
the length). The ethmoid fissure is narrow,
Scott called Capra caucasica Gueldenstaedt & Pallas,
because the frontals intrude into it, often
1783) to a single species, Capra ibex Linnaeus, 1758.
becoming obliterated with age by the intru-
Heptner et al. (1961) split the ibex group into seven
-1— sion of the frontal between the nasal and the
species: C. pyrenaica, C. ibex, C. cylindricornis (the Da-
0— lacrimal bones. The premaxilla forms a very
349-47558_ch01_1P.indd 224 5/31/11 7:05 PM

b o v idae 225
ghestan tur, the species incorrectly called C. caucasica The most distinct of the N ibex is C. pyrenaica, in
by Ellerman & Morrison-Scott), the true C. caucasica which the horns are rounded-rectangular in cross
(the Kuban tur), C. sibirica, C. nubiana, and C. walie. G. section, but with a well-marked posterointernal keel;
B. Corbet (1978) and Valdez (1985) proposed compro- their anterointernal ridge is poorly marked, or has
mise schemes, recognizing both C. cylindricornis and rounded knobs that do not extend laterally across the
C. caucasica as distinct species, but retaining C. sibirica, front surfaces; the horns have a homonymous horn
C. nubiana, and C. walie in C. ibex. curl, like C. falconeri and C. cylindricornis, and they
C. ibex of the Alps and C. sibirica of the C Asian tend to be lyrate in anterior view; the horn bases are
mountains are undeniably more similar to each other expanded; the horn cores are less straight than in
than they are to the intervening Caucasian forms (the other ibex, curving out laterally; the skull is very
two species of tur); what unite alpine and Siberian broad, its breadth 56% of its length; alone among
ibex, in fact, are primitive (for ibex) features of the ibex, the orbits of C. pyrenaica do not interrupt the
horns, while C. caucasica and C. cylindricornis are suc- profi le line; the nasals are as broad as in C. aegagrus
cessively more derived. Moreover, while the two spe- (45% of their length in the males); the nasopremaxil-
cies of tur undoubtedly do share derived features of lary suture is long, about 20 mm; the ethmoid fissure
the skull and the horns (C. caucasica is an ideal inter- is open, if narrowly, right down to the premaxilla; the
mediate between the alpine/Siberian type of ibex and median palatal notch is often well in front of the lat-
the highly derived C. cylindricornis), they would ap- erals; and the paroccipital processes are long.
pear to be reproductively isolated (see below). On Other N ibex have longer, unexpanded nasals,
these grounds alone, there is little alternative to rec- which make a rounded suture with the frontals; the
ognizing all seven as distinct species: they are all di- skull is less broad; the nasopremaxillary suture is
agnosably distinct, and the molecular data (see below) usually shorter; and the paroccipital processes are
make the multispecies solution mandatory. short. These ibex divide into a more primitive group
Ibex separate fairly well phenetically into a S and a (C. ibex, C. sibirica) and a more derived group (C. cau-
N group. The distinctive Spanish ibex, C. pyrenaica, casica, C. cylindricornis). The highly derived (Cauca-
fits, if somewhat uneasily, into the latter. sian) group has a shorter braincase than other ibex,
S ibex have a longer, less-sloping braincase; the except for C. sibirica; the skull breadth is 53%–54%
horn cores are less expanded (the transverse diameter of the length (as in C. sibirica; in C. ibex, it is 52%);
is <80% of the anteroposterior diameter); the horns the nasals are broad (the breadth is 39% of the length,
are rounded-rectangular in cross section, and some- compared with 36% in the alpine and Siberian ibex),
what compressed laterally, with prominent and dis- pointed both anteriorly and posteriorly; the median
crete transverse knobs across their anterior surface; palatal notch is very broad and short, not reaching
the nasals are short (only 34% of the skull length), the back of the toothrow or extending in front of
somewhat oval, with parallel borders, and usually the laterals; the horns are divergent and, above all,
run horizontally before turning down at the tips; the horn bases are rounded in cross section and
the toothrows are less convex; in most cases the me- enormously expanded, the transverse basal breadth
dian palatal notch does not reach farther forward of a core being over 90% of the anteroposterior
than the laterals; the biorbital breadth is about 50% breadth.
of the skull length, as in C. aegagrus; the premaxilla The largest and longest-horned species are C. sibirica
has a long suture with the nasal; and the paroccipi- and C. walie. The two Caucasian species are equally
tal processes are elongated. Males in S ibex have a large, but they have relatively short, massive horns. C.
tendency to develop a boss on the median frontal su- ibex and C. pyrenaica are medium-sized species, with
ture, just below the insertion of the horns; all males horns that are usually not spectacular, though occa-
have it in C. walie, but it only develops in old males sional specimens of the latter may develop quite ex-
in C. nubiana. traordinary horns. Finally, C. nubiana is the smallest
N ibex vary between species, but all have a shorter, species in the genus, yet it possesses relatively long
steeper braincase than S species; the nasals slope evenly horns.
downward (as in C. falconeri); the skull is broader across
the orbits, the supraorbital foramina, the zygomatic Phylogeny of Capra
arches, and the basioccipital region; and the dorsal pro- It is extremely difficult to work out a simple morpho-
fi le runs fairly straight from the horns to the interor- logical cladogram for Capra species. Points to be borne —-1
bital region. in mind are: —0
349-47558_ch01_1P.indd 225 5/31/11 7:05 PM

226 Artiodactyla
—The C. nubiana / C. walie group have horns problem—while, at the same time, raising other
similar to C. ibex and C. sibirica, but their skulls phylogenetic controversies—by combining mitochon-
are very different: the narrow skull with the drial cytochrome b sequences with Y-chromosome
long braincase would be plesiomorphic if C. sequences. The authors confirmed previous fi ndings
aegagrus is treated as the outgroup, but, as the that in the mitochondrial tree, Capra sibirica is sister to
horn forms could be a heterochronic growth all other species; but in the Y-chromosome tree, C. ae-
series (see below, under C. ibex), it might also be gagrus and C. falconeri belong to one clade, and all ibex
that C. aegagrus is morphologically a special to the other, the maximum-likelihood bootstrap val-
derivative from the S ibex group. Possibly ues being 98% in the first case, and 92% in the second.
supporting the latter hypothesis are the limb The best explanation of this is that the initial distribu-
markings (which might, however, also be tion of the species conformed to the mtDNA data, and
symplesiomorphic— see below) and the was then altered by the spread of proto-sibirica, which
forehead convexity, and perhaps the scimitar overwhelmed most other species, leaving only C. ae-
horn type (but see below). gagrus and C. falconeri intact. When looking at the
—The resemblance of the of the horns of C. morphology of the “ibex” species, we can, in fact, see
cylindricornis to Pseudois has often been that there could not have been total nuclear swamp-
noticed and commented upon; but for the fact ing; indeed, there was less and less C. sibirica influence
that the two Caucasian species are remarkably farther and farther W, where the ibex become pro-
similar cranially, it would be reasonable to see gressively less sibirica-like in their horns, with more
the E Caucasian tur as a sister species to all and more retention of the aegagrus-like color pattern.
other Capra. A hypothesis to account for this We can judge the S ibex—C. nubiana, at any rate—to
anomaly will be put forward below. be genetically about 50% proto-sibirica and 50%
—C. pyrenaica, C. cylindricornis, and C. falconeri proto-aegagrus.
all have a homonymous horn curl; this appears According to the mitochondrial data, the relation-
to be the correct interpretation, although the ships of the non-sibirica ibex would be as follows:
direction of the spiral can be extraordinarily
((nubiana ((pyrenaica, ibex) (caucasica,
difficult to figure out (whereas true goats
cylindricornis))).
usually have heteronymous horns, even
in most domestic breeds with corkscrew We will consider some of these relationships below.
horns). It is likely that extreme spiraling,
of either type, is apomorphic. Other Caprini, in Capra aegagrus Erxleben, 1777
the main, also have homonymous horns, so it wild goat, bezoar
seems probable that mild homonymy
The name Capra aegagrus Erxleben, 1777, was placed on the
is primitive for Capra—but whether it is
truly primitive to sickle-shaped horns is moot.
Opinion 2027 (2003).
—Probably the strong leg markings (with white
Various subspecies have been described for this species,
pasterns and knees) of C. aegagrus and the S
which are listed below, with their type localities.
ibex are primitive; they occur outside the
genus in Pseudois. 1777 Capra aegagrus Erxleben. Daghestan district,
Caucasus.
Thus we simply leave the groupings (as outlined above)
1875 Capra blythi Hume. Sind, Pakistan.
as phenetic groups, and, for the present, avoid any at-
1905 Capra persica Matschie. Laristan, SW Iran.
tempt to reconstruct their phylogeny.
1907 Capra florstedti and Capra cilicica Matschie. Bulghar
Curiously, the molecular evidence does little to
Dagh, Turkey.
resolve the phylogeny of Capra, it being complex in
1918 Capra hircus neglectus Zarudny & Bilkevitch. Seistan,
the extreme. According to the mtDNA sequencing
E Iran.
study of Manceau, Després, et al. (1999), Capra sibirica
1950 Capra aegagrus turcmenicus Zalkin. Kopet-Dagh, S
forms a separate clade from all other members of the
Turkmenia.
genus, which, in contrast, shows no special clustering
between any two species. The placement of C. sibirica A further subspecies that has usually been accepted is
as a sister species to all others is unexpected and coun- the Cretan wild goat (Capra aegagrus cretica). The ori-
-1— terintuitive, but Pidancier et al. (2006) resolved this gin of this putative taxon was surveyed by Bar- Gal
0—
349-47558_ch01_1P.indd 226 5/31/11 7:05 PM

b o v idae 227
et al. (2002), with the aid of cytochrome b and D-loop compared with 149.8%; but the ranges for the horn
sequencing. Cretan wild goats cluster quite clearly characters overlap considerably. Therefore, at present
with domestic goats, especially the Baladi breed of we fi nd no evidence, based on skull or horn charac-
Lebanon, and not with wild bezoar; that is to say, the ters, for dividing Capra aegagrus into subspecies, but it
Cretan taxon is a feral goat, but of very ancient ori- may be that a combination of characters would suc-
gin, and presumably extremely valuable as indicating ceed in separating some populations.
something of the morphology of the earliest Middle A study by Zeder (2001) of a large collection of
Eastern domesticated goats. skulls from different regions of Iran found that there
Zalkin (1950) gave measurements of bezoar goats is a noticeable reduction in size, as represented by
from the Caucasus and from the Kopet-Dagh, from metacarpal distal breadth, from N to S, especially in
which it emerges than the only difference between the males. There seems to be a much greater differ-
them is the more slender horns of the latter. He gave the ence in the males between the N Zagros Range and
following statistics (with the mean and the minimal- other parts of the Zagros than between those from C
maximal range in mm): and S Zagros, whereas, in so far as there are differ-
ences in the females, it is the S Zagros females which
1. Skull length (males)
stand out. These data seem to confl ict somewhat with
—Caucasus = 263.4, standard deviation 9.50,
our own observations on skulls; further study of ad-
248–278 (N = 11)
ditional skulls would show either that those from the
—Kopet-Dagh = 261.6, standard deviation 8.45,
S Zagros (and presumably the Makran and Pakistan)
250–279 (N = 11)
are smaller than those from farther N, or else that the
2. Skull length (females)
S populations have more gracile metapodials for the
same general skull size.
215–233 (N = 6)
It is reported in the literature that goats from Sind
are lighter in color than those from Iran, on average;
224–234 (N = 6)
that those from Turkey are darker still, brown or red-
3. Basal circumference of horn (around the
brown rather than yellow-brown or red-gray, and
sheath)
that their flank- and shoulder-bands are broader. The
few localized skins and living specimens we have
200–250 (N = 15)
seen would tend to confi rm this, but it is on average
only that they differ, and the silvery white coat of old
180–230 (N = 37)
age is found in the males from all three areas.
On the basis of the horn circumference figures, Zalkin
erected the subspecies Capra aegagrus turcmenicus. Capra falconeri (Wagner, 1839)
Our own figures confi rm and extend these fi nd- markhor
ings; four male skulls from Sind have a mean skull
There are, again, a large number of subspecific names
length of 261.2 mm; a male from Mazanderan (Alborz
available in this species.
Mountains) is 260 mm, one from Fars is 261 mm, and
two labeled just as “Persia” are 269–286 mm. A skull
1839 Aegoceros (Capra) falconeri (Wagner). Astor, Kashmir.
from the Makran (SE Iran), however, is only 250 mm
1842 Capra megaceros Hutton. Kandahar, Afghan istan.
long—unusually small, at the lower limit of the range
1875 Capra jerdoni Hume. Suleiman Range, Punjab, Pakistan.
given by Zalkin (1950). A female from Shiraz, in SW
1898 Capra falconeri cashmiriensis Lydekker. Pir-Panjal
Iran, has a skull length of 200 mm, and one from Sind
Range, Kashmir.
is 229 mm.
1945 Capra falconeri heptneri Zalkin. Dashtidjum district,
We measured the horn basal circumference around
Tadjikistan.
the core, not the sheath, and found a mean of 186.6 mm
1945 Capra falconeri ognevi Zalkin. Kuh-I-Tang Range,
for 10 specimens from Turkey and the Caucasus,
Uzbekistan.
166.7 mm for 3 from the Kopet-Dagh, 181.8 mm for 4
1950 Capra falconeri gilgitensis Cobb, nom. nud. Gilgit.
from Sind, and 175.0 mm for 2 from S Iran.
1958 Capra falconeri chitralensis Cobb. Chitral.
Horns from the Kopet-Dagh and Sind average lon-
ger (1125 and 1115 mm, respectively) than those from Schaller & Khan (1975) and Schaller (1977) made criti-
the Caucasus and Turkey (923 mm), and their curve cal observations on subspeciation in this species. They —-1
length is 162.7% of their direct base-to-tip distance, found, for example, that within Kashmir (including —0
349-47558_ch01_1P.indd 227 5/31/11 7:05 PM

228 Artiodactyla
Gilgit, Chitral, Astor, and the Pir-Panjal Range) a va- large samples; if it should prove that three distinct
riety of horn types, though always of the open-spiraled species exist, then this will have great significance for
type, can be found in any one locality; similarly, the conservation.
desert ranges of S Pakistan and Afghan istan share a
more tightly spiraled type of horn, which also varies The Chiltan Goat
greatly. They proposed that there are really only two Schaller & Khan (1975) maintained that this screw-
subspecies in the Indian and Pakistani parts of the horned goat (Capra falconeri chialtanensis Lydekker,
range; occasionally, the “wrong” type of spiral can 1913), from the Chiltan and neighboring ranges in
even be found between these two subspecies, though Pakistan, is a localized morph of C. aegagrus. T. Rob-
they can be overwhelmingly separated. For the Kash- erts (1977) and Valdez (1985), on the other hand, ar-
mir type, the name Capra falconeri falconeri has prior- gued that it is a hybrid between bezoar and markhor.
ity; for the S type, Capra falconeri megaceros. In our The only thing that is agreed on is that it is not a form
sample, and adding measurements from the litera- of markhor, as Lydekker thought.
ture (especially Heptner et al.,1961, who found no rea- Schaller & Khan (1975) reported that the screw-
son to separate the two [former] Soviet forms from horned type occurs in the same herds with ordinary
each other), we fi nd the following data (males only): bezoar goats on several ranges, and that bezoar and
From these figures it would appear that the C. f. markhor coexist without interbreeding on at least
heptneri has remarkably short horns; C. f. falconeri has two other ranges. T. Roberts (1977) maintained that
more curled horns and a wider flare; and C. f. megace- there is evidence that markhor formerly occurred on
ros has the tight curl and narrow tip-to-tip distance of at least one of the ranges where the Chiltan goat is
C. f. heptneri, but the very long horns of C. f. falconeri. found; according to his observations, the Chiltan
Some of the above ranges do not overlap, but in- goat has some markhor-like features, such as the lack
creased sampling might well cause them to do so; of a dark face pattern. Valdez (1985) depicted three
however, on the face of it, it does seem likely that we specimens and noted their resemblance to experi-
might well have three separate species. mental goat-markhor hybrids, one of which he also
According to Schaller (1977), the Kashmir taxon illustrated.
would have longer hair, especially in winter, and a We have seen two specimens of this form: the
longer ruff than C. f. megaceros (an extremely long ruff type (a frontlet and horns) and a more complete tro-
also occurs in C. f. heptneri, the taxon that is much the phy skull (British Museum specimen BM 67.795), lack-
commonest in captivity). ing only premaxillae. The latter is, overall, markhor-
Hammer et al. (2008) sequenced the fi rst hyper- sized, and thus is bigger than any C. aegagrus: the
variable segments of the mitochondrial control re- biorbital breadth is 146 mm, at the lower end of the
gion in zoo markhor, and they showed that C. f. hept- markhor range but outside that of the bezoar (in
neri is paraphyletic with respect to C. f. megaceros. bezoar, the maximum for this measurement is
More worryingly, they found that many zoo markhor 144 mm, and it is usually below 140 mm). Nasal length
are introgressed with domestic goat DNA. is 88 mm; the smallest markhor measures 89 mm,
Therefore, we provisionally retain all markhor in while the largest bezoar is 82 mm. For what it is worth,
a single species, but urge a detailed comparison of the cranial capacity of the Chiltan goat is consider-
ably greater than any C. aegagrus, although we have
not measured the capacities of any markhor for
comparison.
Table 61 Horn length for markhor: males We tend to agree with T. Roberts (1977) and Valdez
(1985) that the Chiltan goat is a natural hybrid. Valdez
falconeri megaceros heptneri
(1985) depicted a Chiltan goat that approaches the be-
zoar type of horns, and it is presumably a backcross.
Horn length
curve 708–1355 (20) 755–1130 (12) 750–900
straight 801–1082 (13) 755–782 (12) 580–730 Capra pyrenaica Schinz, 1838
Ratio, 66–80 (14) 76–87 (12) 76–93 Spanish ibex
straight:curve This distinctive species today lives only in Spain. C.
Tip–tip distance 695–1321 (17) 518–711 (10) 210–600
pyrenaica formerly occurred in Portugal, but it is now
Ratio, 85–133 (14) 60–85 (10) 36–82?
-1— tips:straight length extinct there. Engländer (1986) described its occur-
0— rence in SW France in the early Upper Pleistocene,
349-47558_ch01_1P.indd 228 5/31/11 7:05 PM

b o v idae 229
and its replacement in the Latest Pleistocene by C. Although we have seen very few specimens,
ibex. from the above evidence we think that all the varia-
Five subspecific names are available: tion in C. pyrenaica can be ascribed to a major N– S
cline, with some additional microgeographic varia-
Capra lusitanica Schlegel, 1872. N Portugal (now
tion. Probably no subspecies can be recognized
extinct). On the history and availability of this
nomenclaturally.
name, see Almaça (1992).
We should comment, however, on the head of a
Capra pyrenaica Schinz, 1838. Pyrenees.
Pyrenean male (the only one we have seen from this
Capra pyrenaica cabrerae Camerano, 1917. Sierra
population) in the British Museum: at the base, the
Morena.
horns are broader anteroposteriorly than in any other
Capra pyrenaica nowaki Wyrwoll, 1999. This
specimen, with prominent round knots along an an-
name was to replace Capra pyrenaica hispanica
teromedian ridge (but not extending across the front
Schimper, 1848, which was supposedly from
of the horns, as in C. ibex). If enough localized mate-
the Sierra Nevada, but the type series of
rial becomes available, another look at infraspecific
hispanica is, in fact, mixed, and included a
variation would be advisable.
specimen from the Pyrenees, according to
Wyrwoll (1999).
Capra nubiana F. Cuvier, 1825
Capra pyrenaica victoriae Cabrera, 1911. Sierra de
Nubian ibex
Gredos.
Smallest species of the genus.
Engländer (1986) gave numbers to show that there We have seen skulls from Israel, the Sinai, and Su-
is a progressive reduction in size from N to S: head dan, as well as an incomplete skull from Aden, S Ye-
plus body length averages 148 cm in nominotypical C. men; and living specimens from Israel and Sudan.
p. pyrenaica, 142 cm in C. p. lusitanica, 135.5 cm in C. p. Characteristic measurements of the males are as
victoriae, and 121 cm in “C. p. hispanica” (or C. p. nowaki; follows:
see above). The Portuguese form (C. p. lusitanica) had There is thus a cline in size from N to S, but a
the shortest, stoutest horns (mean length 44.5 cm, basal fairly clean difference in tip-to-tip distance between
circumference 23.5 cm), the Pyreneean C. p. pyrenaica skulls from Israel and the Sinai, on the one hand, and
the longest (at 80 cm), and the other two intermediate Sudan, on the other: the tips turn markedly inward in
(73–73.4 cm), while the basal circumference is only the former, less so in the latter. An incomplete skull
18.9 cm in “C. p. hispanica” and 20–22 cm in the other from Aden appears to be at least as large as the aver-
two. The S “C. p. hispanica” and Portuguese C. p. lu- age Sudan specimen.
sitanica have the narrowest horn spans (29.4 and 25 cm, Adult males in Israel and the Sinai are light brown,
respectively), with C. p. pyrenaica and C. p. victoriae with red, yellow, or whitish tones; there is a weak
much wider (at 42 and 47 cm, respectively). Engländer brown shoulder-band, a blackish dorsal stripe, a dark
(1986) noted that a peculiar, somewhat sheep-like horn flank band, a white belly, and white wrist marks. The
morph (in which the horns are curled around into a females are lighter in color. Sudanese specimens are
near circle) occurs frequently on some, but not all, of dark brown, at least in the adults, with no lighter tones
the ranges in the far S; and a horn type that resembles on the body; the shoulder band is dusky; the dorsal
C. ibex, except for the absence of knots, can be found on stripe is black; the flank-band, on the contrary, is barely
the Sierra Blanca. There seems to be a cline in color, visible; the belly is white or off-white; and the wrist is
the N ibex being darker, with a more strongly ex- white, as in other populations. The females are barely
pressed pattern of dark markings; although in the Si-
erra de Gredos, in C Spain, the dark markings may be
often so strong that in summer they seem almost Table 62 Measurements of Nubian ibex: males
black. Engländer (1986) stated that on the Sierra de
Gredos, in C Spain, there are ibex corresponding to all Skull length Horns, tip–tip
four of Cabrera’s described subspecies.
On the basis of mtDNA sequencing, Manceau, Mean Std dev N Mean Std dev N
Crampe, et al. (1999) found that three specimens from
Israel 244.5 8.81 4 336.8 171.92 4
the Pyrenees clustered together, in contrast to speci-
Sinai 250.2 10.03 5 345.7 225.50 3
mens from other localities; the support value for the Sudan 258.1 8.81 10 447.7 158.04 11 —-1
non-Pyreneean clade, however, was only 73%. —0
349-47558_ch01_1P.indd 229 5/31/11 7:05 PM

230 Artiodactyla
lighter in color than the males. We have seen no spec- and C. sibirica in captivity shows that the first-year
imens from S Arabia. horns of the males resemble the C. aegagrus form, that
Granjon et al. (1990) confirmed that Sudanese spec- is, narrow, with a prominent anterior keel; the horn
imens are much darker than those from Egypt and then expands, especially posterolaterally, reducing
the Dead Sea region. A single specimen from Saudi the keel to an anteromedial angle, as in C. ibex; then,
Arabia had different alleles at the MOD (malic-enzyme) around sexual maturity, the anterolateral expansion
locus when compared with 20 specimens from Sudan, takes place to give the adult a broad-fronted horn.
Egypt, and the Dead Sea region. Therefore, either C. ibex can be viewed as a neotenous
Primarily on the basis of color, we recommend derivative of the more “advanced” ibex, or else the
recognition of two subspecies: Capra nubiana nubiana broad-fronted species are seen as hypermorphic. C.
F. Cuvier, 1825, from Sudan; and Capra nubiana sinait- aegagrus would perhaps be easier to fit in at the base of
ica Ehrenberg, 1833 (synonym—Capra arabica Rüp- the latter series, rather than at the top of the former.
pell, 1835), from Israel and the Sinai. It is possible that With age, the horns of C. ibex do not become broad-
Aegoceros beden (Wagner, 1835) (synonym—Capra men- fronted, but the knots become worn down.
gesi Noack, 1896), from Saudi Arabia, may represent a Alpine ibex are one of the great success stories of
third subspecies, but evidence is lacking. Given, how- wildlife conservation; as told by Grodinsky & Stuewe
ever, that the color of the adult males appears, on lim- (1987), their numbers have recovered from a low of
ited evidence, to be quite different between the Suda- about 50 in 1821 to some 22,000 today.
nese and the Israeli/Sinai specimens, it is possible
that we have different species; resolution of this ques- Capra sibirica (Pallas, 1776)
tion is also hampered by the lack of material from S Siberian ibex
Arabia. Large. Differing from C. ibex in the more tubular
orbits; median palatal notch not extending forward
Capra walie Rüppell, 1835 to the third molar, or in front of the laterals; ethmoid
walia ibex fi ssure commonly obliterated with age. Horn bases
This species, from Ethiopia, is the most endangered in about equally expanded (the transverse diameter
the genus. Its taxonomic status is slightly equivocal. C. 82%– 84% of the anteroposterior diameter), but
walie shares many characters with C. nubiana, as listed broad-fronted.
earlier, but differs from it characteristically: it is much There are, in Ellerman & Morrison-Scott’s (1951)
larger (among the largest of the genus); the nasal list, no fewer than 16 names available for subspecific
breadth is 40% of the length (compared with 38% in C. differentiation in this species; Zalkin (1949) reduced
nubiana); the nasofrontal suture is rounded, rather these to three or four, at the same time adding two
than triangular; and the nasopremaxillary suture is more. Zalkin’s list is as follows:
much longer, about 30 mm instead of 18 mm. The
Capra sibirica alaiana Noack, 1902. Tienshan
horn cores are more expanded, their transverse width
system, including Transalai and Alai-Pamir.
79.5% of their anteroposterior diameter. In C. walie, all
Synonyms—Capra sibirica merzbacheri Leise-
adult males have a marked convexity on the interfron-
witz, 1906; Capra sibirica transalaiana and Capra
tal suture; this develops only in old age in C. nubiana.
sibirica almasyi Lorenz, 1907.
In general, C. walie may be viewed as a hypermorphic
Capra sibirica dementjevi Zalkin, 1949. Kuen-Lun
derivative of C. nubiana; that is, a logical prolongation
Range.
of the growth-processes of the latter. Considering the
Capra sibirica formosovi Zalkin, 1949. Talas-Alatau
ease of distinguishing the two, C. walie has to be re-
and other W Tienshan ranges.
tained as a full species.
Capra sibirica hagenbecki Noack, 1903. Gobi Altai
and Transaltai Gobi.
Capra ibex Linnaeus, 1758
Capra sibirica sakeen Blyth, 1842. Kashmir. Zalkin
alpine ibex
had few specimens from this part of the
Medium-sized species, characterized by its horns,
species’ range, but he thought this probably
with the anteroexternal corner rounded, thus having a
represented a distinct subspecies. Synonyms—
“beveled-off ” appearance, unlike the squared-off form
probably Capra sibirica skyn (Wagner, 1844);
(with the anteroexternal angle as well developed as
Capra sibirica dauvergnii Sterndale, 1886; Capra
-1— the anterointernal angle) of C. nubiana, C. walie, and C.
sibirica wardi Lydekker, 1900; Capra sibirica
0— sibirica. Observation of the growth series of C. nubiana
349-47558_ch01_1P.indd 230 5/31/11 7:05 PM

b o v idae 231
Table 63 Characteristic measurements for Capra sibirica sibirica, C. s. alaiana, and C. s. sakeen
sibirica alaiana sakeen
Mean Std dev N Mean Std dev N Mean Std dev N
Skull length
males 270.0 9.25 17 292.0 8.95 20 293.1 9.11 20
females 247.0 6.55 11 254.0 4.45 11 — — —
Nasal length, males 86.6 5.10 16 103.9 6.80 18 107.0 4.32 7
Horn (males), basal circumf 20.7 1.36 17 23.0 2.02 43 — — —
pedri Lorenz, 1906; Capra sibirica filipii Cam- skull length of 230 mm (more than two standard de-
erano, 1911. viations below C. s. sibirica). he metrical evidence for
Capra sibirica sibirica (Pallas, 1776). Russian Altai, the validity of C. s. hagenbecki would, therefore, seem
Sayan, and neighboring ranges in Mongolia. equivocal.
Synonyms—Capra sibirica lydekkeri Roths- Zalkin’s (1949) evidence for his subspecies C. s. de-
childi, 1900; Capra sibirica altaica and Capra mentjevi consisted of two female skulls, much smaller
sibirica fasciata Noack, 1902; Capra sibirica than any other female specimens. We measured a
lorenzi Satunin, 1905. male skull from “near Yarkand,” which suggests the
Kuen-Lun Range from which C. s. dementjevi was de-
There is also the name Capra sibirica hemalayanus
scribed; its measurements fall well within the range
Hodgson, 1841, said to be from Nepal, but the species
of C. s. alaiana.
does not occur there.
Externally, nominotypical C. s. sibirica is described
Heptner et al. (1961) made C. s. formosovi a syn-
as dirty white or yellowish white, and dark along the
onym of C. s. alaiana, commenting that specimens
flanks, the lumbars, the shoulders, the breast, and the
with and without a light saddle spot (its absence be-
anterior legs. The light area forms a saddle, which is
ing the distinguishing feature of C. s. formosovi) occur
separated by the dark shoulder zone from a light cer-
in the same populations in the Tienshan system.
vical spot. Zoo specimens we have seen, mostly known
The following table contains some of the character-
to be from the Altai, are of this type. C. s. hagenbecki is
istic measurements given by Zalkin (1949) for C. s. si-
described as uniform gray-brown; that is, lacking the
birica and C. s. alaiana; we have studied and measured
dark/light contrasts of C. s. sibirica. If this difference
specimens of both (but mainly of C. s. alaiana), which
holds true, which would mean that the presence or
tend to confi rm his fi ndings, and we include in the
absence of the saddle is less variable in Siberian/Mon-
table statistics for C. s. sakeen (from Kashmir, Ladakh,
golian ibex than in those from the Tianshan system,
Pamir, and Afghan istan) taken by ourselves:
then C. s. hagenbecki would be a valid subspecies.
Zalkin (1949) also mentioned that the horns of
C. s. alaiana is described as dark cinnamon-brown,
the males (measured along the curve) are normally
with a light saddle extending to the haunches; again,
>1 m long in C. s. alaiana, but <1 m in C. s. sibirica
the saddle is separate from the cervical spot. The rela-
(though his table shows that actually only 4 out of 6
tively few museum skins seen by us are of this type.
alaiana are >1 m). In our sample of C. s. alaiana, how-
As noted above, the absence of the light saddle is
ever, 10 out of 13 have horns >1 m long (making a
said to predominate in some ranges, and to be less
total of 14 out of 19), and only 2 in our sample of 8 C.
frequent in others, so that C. s. formosovi is a polymor-
s. sibirica are >1 m (one of them only just); in our
phic variant, rather than a valid subspecies.
sample of C. s. sakeen, the horns are usually <1 m
Zalkin (1949) suggested that C. s. sakeen is distin-
(only 6 exceptions in 28 specimens). Thus C. s. sakeen
guished by the broad saddle-spot being connected to
would seem, in general, to be large like C. s. alaiana,
the cervical spot, leaving only the haunches dark. But
but shorter-horned.
some specimens we have seen have the cervical and
G. M. Allen (1940) gave (incomplete) measure-
saddle spots separated by the dark shoulder zone.
ments for two specimens from the Gobi Altai: a male
Capra sibirica thus has two reasonably well-defined
with a skull length of 265 mm (thus only slightly be- —-1
subspecies (C. s. sibirica and C. s. alaiana), a third (C. s.
low the mean for C. s. sibirica), and a female with a —0
349-47558_ch01_1P.indd 231 5/31/11 7:05 PM

232 Artiodactyla
sakeen) that is probably recognizable (note that, if We are uncertain as to what to make of this rede-
it is combined with C. s. alaiana, the name C. s. sa- scription of geographic variation, which includes fea-
keen has priority), and two (C. s. hagenbecki and C. s. tures and differences not known to Zalkin (1949),
dementjevi) whose validity must remain uncertain for who apparently remains the last author to perform a
the moment. primary systematic revision.
Fedosenko & Blank (2001) recognized the follow-
ing subspecies: Caucasian Ibex (Capra caucasica and Capra
cylindricornis)
—Capra sibirica sibirica— Small; skull length
Capra caucasica and Capra cylindricornis are evidently
256–293 mm (males), 239–260 mm (females);
very closely related. They have very short, steeply
horn length 110–117 cm (males; further on it
sloping braincases; extremely broad skulls; broad na-
says <100); girth at the horn base 18.5–24 cm;
sals that are pointed at both ends; and the median pala-
body mass up to 103 kg. Nasals short, their
tal notch is exceptionally broad, reaching to the hind
length usually <100 mm, 28%–37% of the
surface of the third molars, as do the lateral notches.
skull length (males), or 32%– 42% (females).
They are almost equally large in size (skull length
Nasofrontal suture anterior to the anterior
about 294–296 mm, equal to C. sibirica and C. walie),
edge of the orbit, or at the same level. Males
with comparatively short, extremely thick-based
(in winter) relatively light, the back, the sides,
horns. The color in both is a medium brown, with
and the upper part of the neck mainly dirty
little color contrast.
white or yellowish white. Dark and light color
The main difference is in the horns of the males.
patches bright, with contrasting borders
In C. caucasica (the Kuban tur), the horns resemble a
(Zalkin, 1950; Sobanskiy, 1988).
shorter, thicker-based, more divergent form of those of
—Capra sibirica hagenbecki— Similar in size and
C. ibex; like the latter, the anterolateral edge is beveled
build, but with relatively longer horns (males
off. The tip-to-tip distance in the males of C. caucasica
74–139 cm, females 16–38 cm); ridges in the
averages 380 mm; of C. cylindricornis, 540 mm. In C.
middle portion of the horns larger. Color more
caucasica, the young males have prominent anterior
gray and pale brown, without a light saddle;
knobs, but these wear down seemingly earlier than in
old males sometimes becoming nearly white
C. ibex. In C. cylindricornis (the Daghestan tur) the
(Lydekker, 1913; Bannikov, 1954).
horns curl inward, then upward again at the tips, in an
—Capra sibirica alaiana— Bigger; skull length
almost sheep-like fashion, which is, however, some-
276–310 mm (males), 248–262 mm (females).
what foreshadowed in C. caucasica. The transverse di-
Nasals long, their length >100 mm, 32%– 40%
ameter of the horn core base is 99% (±5.0%) of the an-
of the skull length for the males, 34%– 43%
teroposterior diameter in C. cylindricornis (and may
for the females. Nasofrontal suture higher
exceed it), while in C. caucasica this figure is 92%
than the anterior edge of the orbit. Horns
(±10.3%), the second highest in the genus.
length usually >100 cm, and up to 147 mm;
There are some differences in the skull; mainly,
girth at the horn base 20–28 cm. Body mass
that in C. caucasica the ethmoid fi ssure appears to
of the males up to 130 kg. Adult males (in
remain open throughout life, while in C. cylindricor-
winter) relatively dark, a dark brown-gray,
nis it is often obliterated. Also, the coronal suture is
with less contrast in the dark and light
nearly straight in C. caucasica, but projects forward at
patches. Light saddlelike spot on the dorsum
an angle in C. cylindricornis; the highest point of the
variable in form and size (Heptner et al.,
skull is farther to the back of the horn cores in C cylin-
1961).
dricornis; and the parietal has a depression in the lat-
—Capra sibirica sakeen—Like C. s. alaiana, but the
ter (Weinberg, 2002); the length of the premaxillary
horns relatively short (males 102–140 cm),
suture (with the nasals) measures 33.3 ±5.32 mm in
more massive at the base, the girth up to
C. caucasica, but only 21.3 ±8.46 mm in C. cylindricor-
30 cm. Males weighing less, up to 90 kg. Light
nis; the nasals are longer in C. cylindricornis, 125.2
in color; back and sides pale brown or creamy
±7.29 mm (41% of the skull length), while in C. cauca-
white; dorsal stripe pale brown; other parts
sica they are only 107.3 ±8.08 mm (37% of the skull
light yellowish brown; beard and tail black-
length). Heptner et al. (1961) state that the Daghestan
brown (Lydekker 1913; Heptner et al. 1961;
-1— tur has a shorter braincase, but we do not fi nd this in
T. Roberts 1977; Schaller 1977).
0— our data.
349-47558_ch01_1P.indd 232 5/31/11 7:05 PM

b o v idae 233
In the phylogenetic tree of Manceau, Després, et al. designed for alternative styles of battering and for
(1999), six samples of C. caucasica formed a separate taking the force of blows, and thus are not so very
clade from three samples of C. cylindricornis, which, closely related. We therefore propose the following
curiously, were sister to a clade of two out of four scenario.
studied specimens of C. aegagrus. C. ibex once ranged up to the W Caucasus. The
C. caucasica inhabits approximately the W third of Caucasian population became isolated there by cli-
the Caucasus, C. cylindricornis the E two-thirds. Wein- matic change, perhaps being several times cut off
berg (2002) specified that the E limit of C. cylindricornis and then rejoined in successive glacial/interglacial
is 48° 31' E, in Azerbaijan, whereas the W limits are cycles, and its genome was periodically swamped by
less well defi ned. According to Heptner et al. (1961), the larger C. cylindricornis males dominating the
the ranges of the two species overlap between about smaller C. ibex males. Males formed a fitness series:
40° E and 43° 30' E, and here, on the S slopes of the cylindricornis > hybrids > ibex. In this scenario, C. cau-
Caucasus, they hybridize. Hybrids were said by these casica represents a stabilized hybrid population, with
authors to be rather infrequent, and Weinberg (2002) the W end (“C. severtzovi”) retaining more C. ibex char-
even seemed to doubt whether hybridization really acters than the E end (“C. caucasica”). The present situ-
occurs; evidently, there is no hybrid population, and ation would be the latest cycle, with C. cylindricornis
all hybrids, it is claimed, conform to a type, implying, genes again introgressing the W population. We sug-
if true, that there is little or no backcrossing. More- gest that the two Caucasian forms have to hybridize,
over, Heptner et al. (1961) reported no evidence of because, in occupying much the same altitudinal
hybrids from the N slopes. As for the previous de- range, they must compete. Goats, on the other hand,
scriptions of several taxa in the Caucasus, Heptner et al. with different altitudinal preferences, will not com-
put this down to variability not within a hybridizing pete; thus they do not hybridize, even though they
and backcrossing continuum, but within each of the surely come within “cruising distance” of each other (C.
two species themselves; thus C. severtzowi, C. raddei, aegagrus with C. nubiana or with the turs; C. falconeri
and C. dinniki would be synonyms of C. caucasica, and with C. sibirica). The other case within the genus where
C. pallasii of C. cylindricornis. two species with the same altitudinal preferences meet
The results of our work conform to the implica- is that of C. aegagrus and C. falconeri; and here, too (as
tions of Vereschagin’s (1959) claim that C. ibex, C. cau- we argue above), they hybridize.
casica, and C. cylindricornis represent a kind of trans- There is a clear need for a behavioral field study,
formation series. Moreover, the maturation of horns preferably accompanied by genetics, to determine the
in C. caucasica males is reminiscent of the transition extent of hybridization, the question of backcrossing,
from a triangular cross section and prominent knots, and, ultimately, the issue of gene exchange. As far as
which are Ovis gmelini characters, to a rounded cross conservation is concerned, the need is to ensure, as far
section and crowded transverse ridges in the ontog- as possible, the security of both species, as well as the
eny of argali (the Ovis ammon group). So C. caucasica overlap zones, on both the N and S flanks of the Cauca-
could represent a phylogenetic intermediate between sus. Whatever the real genetic situation, we are evi-
C. ibex and C. cylindricornis. dently in the presence of a very interesting and poten-
A quite different interpretation is also possible. tially significant biological phenomenon, which must
First, it is difficult to believe that the hybrids are of not be allowed to disappear unstudied.
lowered fertility when one bears in mind the records
of hybridization elsewhere (Vereschagin, 1959; Valdez, Capra caucasica Güldenstaedt & Pallas, 1783
1985). Second, it is not absolutely clear that there are Kuban tur
only three morphs (parental forms and hybrids). There
1783 Capra caucasica Güldenstaedt & Pallas. Between the
seems to be a widespread belief in geographic varia-
Malka and Baksan rivers, NE of Mt. Elbrus, N Caucasus.
tion within C. caucasica—Vereschagin (1959), for ex-
1888 Capra severtzowi Menzbier. W of Elbruz and S of
ample, treats true C. caucasica as transitional between
Teberda, Great Caucasus Range.
“C. severtzovi” and C. cylindricornis— but not within
1901 Capra raddei Matschie. Upper course of the Ingur River,
the much more widespread C. cylindricornis. C. ibex and
S slope of the Great Caucasus Range, Elbruz region.
C. cylindricornis do not, in general, look like sister spe-
1905 Capra dinniki Satunin. W extremity of the Great
cies, or like the starting point and end point of a transi-
Caucasus Range; probably the upper reaches of the
tion series. These two species have much more the ap- —-1
Beloe, according to Heptner et al. (1961).
pearance, as far as their horns are concerned, of being —0
349-47558_ch01_1P.indd 233 5/31/11 7:05 PM

234 Artiodactyla
Capra cylindricornis (Blyth, 1841) sis to O. ammon, but, when plotting the posterior skull
Daghestan tur breadth against the skull length, the morphocline is
from gmelini/vignei to ammon, the O. canadensis group
1841 Aegoceros pallasii Rouillier; not Schinz, 1838. Caucasus.
falling above the line (i.e., broader skulls), with O. nivi-
1841 Ovis cylindricornis Blyth. Caucasus.
cola still broader yet (Grubb, 2000a).
Pees & Hemmer (1980) found that specimens of
Nilgiritragus Ropiquet & Hassanin, 2005
the O. ammon group (the precise taxon was not stated)
Nilgiri tahr
have a larger cranial capacity relative to the other Old
2n = 58.
World wild sheep.
Horns parallel, somewhat curved back, trans-
For our craniometric analyses, we used mainly
versely wrinkled, the inner surface nearly flat, the
the dataset of Pfeffer (1967), with some extra speci-
outer surface highly convex; low keel on the antero-
mens measured by ourselves.
internal edge; posterior surface rounded. Braincase
Craniometric analyses divide Old World wild sheep
descending steeply behind the horns, as in Ovis. Tym-
into three major groups: (1) O. gmelini plus O. vignei;
panic tube very short, thick. Frontals flat. Trough
(2) O. ammon; and (3) O. nivicola. In some of our analy-
between the body of the zygoma and the orbital rim.
ses, we utilized this division. There are, on the other
Nasofrontal suture V-shaped. Anterior nasals very
hand, four karyogroups: (1) O. gmelini, (2) O. vignei, (3)
long, pointed. Frontolacrimal suture forming a sim-
O. ammon, and (4) O. nivicola. We tested these divisions
ple curve. Ethmoid fissure absent, as in Ovis. Nasal
as well. We find that, comparing skulls of the O. gmelini
branch of the premaxilla broad, the ends squarely
and the O. vignei groups, the latter have a longer fron-
near the nasal, but not contacting it.
tal arc compared with the frontal chord, and a higher
The Nilgiri tahr is sister to Ovis, not to the other
and somewhat narrower and shorter skull. Comparing
species of tahr, as was previously assumed (Ropiquet
skulls of the O. vignei / O. gmelini group with those of
& Hassanin, 2005).
the O. ammon group, the latter are much larger and
broader-skulled, with shorter foreheads.
Nilgiritragus hylocrius (Ogilby, 1837)
Apart from our original analyses, we have made
1837 Kemas hylocrius Ogilby. Nilgiri Hills. much use of Heptner et al. (1961), Valdez (1982), Geist
(1991a) and Vorobeev & van der Ven (2003).
Dark yellow-brown above, paler below; dark dorsal
Wild sheep are very habitat specific, favoring hilly,
stripe. Males becoming much darker with age, develop-
rolling country, though not too steep (at least in the
ing a grayish saddle. Greatest skull length about 290 mm.
Old World). Thus it is not surprising that there are
Hill ranges of the W Ghats of India, above the
quite a large number of taxa, and the differences be-
treeline.
tween them, as far as the evidence goes, are quite
discrete and consistent. Therefore, we have no option
Ovis Linnaeus, 1758
but to recognize a number of species.
sheep
Sheep constitute the culmination of the broad-based-
Ovis gmelini Group
horn method of fighting (Schaffer & Reed, 1972), with,
in the males, thick horn bases and horns curled in The name Ovis orientalis Gmelin, 1774, was placed on the
nearly, or completely, or more than in a full circle, Official List of Specific Names in Zoology by ICZN,
varying from homonymous to heteronymous. Brain- Opinion 2027 (2003), but as it seems to refer to a hybrid
case descending steeply behind the horns. Ethmoid (Valdez, 1982), the name nonetheless is unusable (Art. 17.2).
fissure absent. Anterior nasals very long, pointed.
An enormous amount has been written about tax- Ovis gmelini Blyth, 1841
onomic variation in Ovis; it is very difficult to synthe- red or Armenian sheep
size this and put it together with our own, much
1856 Ovis anatolica Valenciennes.
more limited analysis.
1899 Ovis ophion urmiana Günther.
When the transverse diameter of the horn is plot-
1919 Ovis ophion armeniana Nasonov.
ted against the longitudinal diameter, the species and
species-groups of this genus form a morphocline from 2n = 54 (NW Iran, Marakan Protected Region, 38°
O. gmelini and O. vignei through O. dalli and O. canaden- 50' N, 45° 11' E; Koyun-Daghi Island, Lake Urmiah,
-1—
0—
349-47558_ch01_1P.indd 234 5/31/11 7:05 PM

b o v idae 235
38° 28' N, 45° 37' E). Valdez et al. (1978) karyotyped tional Park, near Shiraz, and the Hormud Protected
5 males and 5 females from Mooteh at 2n = 54, identi- Area (27° 35' N, 54° 05' E). Karyotype 0.1, from a
fying them as O. isphahanica, but Valdez (1982) later zoo.
reidentified them as of this taxon. Horns homonymous. Frontal surface flat, with
Skull length 246–264 mm (males; mean 251 mm), sharp angles. All females seemingly horned.
222–238 mm (females; mean 231mm). Lighter color, with less red than in O. gmelini and
Horns supracervical, with a negative bend (heter- O. isphaganica; straw-brown in summer, brownish in
onymous), exceptionally almost in one plane (weakly winter. White saddle-patch present, as in O. gmelini
perverted). Horn length up to 670 mm, rarely slightly and O. isphaganica. Short black ruff, restricted to the
more; basal circumference 220–270 mm. Frontal sur- lower neck and the breast. Occasionally a dark
face tapering upward somewhat; outer ridge usually shoulder-patch in winter in the males. Like O. ispha-
indistinct. Fronto-orbital edge and frontal surface ganica, with small teeth and shorter, thinner horns
rounded; frontonuchal edge sharp. Hornless females than O. gmelini.
common.
Reddish rusty to rusty cinnamon, darker in win- Ovis vignei Group
ter; light patch of varying size usual on the sides of urial
the males. Muzzle white; lower parts of the limbs
white, with a black streak down part of the front. Chest Ovis vignei Blyth, 1841
and underparts with a short tress of elongated dark to Ladakh urial
blackish hair, not reaching up to the throat and the Skull length 257–272 mm (males); facial part of the
lower jaw. skull relatively long (58%– 62.5% of the maximum
Transcaucasus, Turkey, NW Iran (vicinity of Ta- skull length). Anterior profi le quite highly flexed; rel-
briz), and W Iran in the Zagros, to at least the C Za- atively high braincase behind the horns.
gros. Valdez (1982) identified those of Mooteh as be- Horns almost in one plane (slightly perverted / supra-
longing to this taxon. cervical). Frontal surface of the horn tapering upward
significantly; outer rib variously developed, but usu-
Ovis isphaganica Nasonov, 1910 ally smooth. Transverse knobs relatively large; tem-
Esfahan sheep poral and frontal regions relatively convex.
2n = 54 (Murche Khort, 140 km NW of Esfahan, 33° Brown or rusty cinnamon; no light spots on the
38' N, 50° 46' E). If Valdez (1982) is correct in stating sides. Tress on the chest and the neck better devel-
that the Mooteh sheep are O. gmelini, then these oped, forming something of a beard on the throat
karyotype records would refer to that species, and and the angles of the jaw; bib white; ruff black.
not to O. isphaganica. Wakhan and Ishkashim in Tajikistan, and across
Horns cervical, with the tips growing toward the the Afghan istan border; Ladakh, Kashmir.
neck; fronto-orbital edge rounded; frontonuchal edge
sharp. Females at least sometimes horned. Ovis punjabiensis Lydekker, 1913
Males (in winter) with a full-length black ruff ex- Punjab urial
tending to the brisket; otherwise, resembling O. gmelini Cervical horns; redder color than in O. vignei.
in being bibless; white saddle-patch; lower parts of the Salt and Kala Chitta ranges
limbs white, with a black streak running down part of Our analysis fi nds that O. punjabiensis is quite dis-
the front; muzzle white; chin white. Shorter teeth and tinct from O. vignei, and evidently has much in com-
shorter, thinner horns than O. gmelini. mon with O. bochariensis (perhaps symplesiomorphic:
Valdez (1982) saw this taxon in Kolah- Gazi and a greater skull height and long nasals, contrasting
Tange-Sayad (SE and SW of Esfahan); evidently they with a lower skull length and breadth). An ungrouped
are not the taxon occurring in Mooteh, N of skull from Faraza, near Tach (39.13° N, 68.48° E, in
Esfahan. the mountains N of Dushanbe), identifies with the O.
vignei sample. The type of O. blanfordi Hume, 1877
Ovis laristanica Nasonov, 1909 (Bolan Pass, near Quetta, near the southernmost
Laristan sheep Afghan istan border with Pakistan), is “ultra-bo-
2n = 54 (Schmitt & Ulbrich, 1968), confi rmed by chariensis / punjabiensis,” possibly representing a fur-
Valdez et al. (1978) in specimens from Bamou Na- ther taxon, but this needs investigation. —-1
—0
349-47558_ch01_1P.indd 235 5/31/11 7:05 PM

236 Artiodactyla
Ovis bochariensis Nasonov, 1914 Ovis cycloceros Hutton, 1842

Bukhara urial Turkmenian sheep
For Valdez (1982), this is a synonym of Ovis cycloceros, but we 1850 Ovis arkal Eversmann.
are able to differentiate the two cleanly in our analyses. 1852 Ovis arkar Brandt.
1905 Ovis vignei varenzovi Satunin.
2n = 58 (Pyandzh Kara-Tau, 37°–38° N, 69°–70° E;
Sar-Sorak Range, 38°–39° N, 69°–70° E). 2n = 58 (Kosh Yeilagh Protected Region, 50 km ENE
Skull length 232–255 mm (males), 218–240 (females). of Shahrud; extreme E Alborz, Mohammad Reza Shah
Horns usually in one plane, or close to it (per- Wildlife Park, 37° 20' N, 56° 07' E; Badkhyz Reserve,
verted); sometimes positive (homonymous). Outer Turkmenistan).
rib quite faint; transverse folds small. Horn length up Skull length 257–297 mm (males; mean 272 mm),
to 73 cm, rarely longer; basal circumference 220– 231–263 (females; mean 249 mm). Facial part elongated,
270 mm (mean 24.5 mm). averaging over 60% of the maximum length of the
Rusty cinnamon; no light spots on the sides. Tress skull. External nares longer and broader than in O. ara-
on the chest and the neck well developed, forming bica; nasals narrow. Profile behind the horns sloping
something of a beard on the throat and the face of the more steeply.
lower jaw. Horns homonymous (defi nite positive curvature);
Right bank of the Amu-Darya and lower Pyan- exceptionally, bent in a single or almost a single plane
dzha rivers, E to Darvaza. (highly perverted); markedly larger and heavier than
in neighboring species. Horn length up to 920 mm,
Ovis arabica Sopin & Harrison, 1986 basal circumference 240–300 mm (mean 260 mm).
Oman wild sheep Outer rib well developed; frontal surface flat or
Skull length 243 mm (males). Skull relatively broad slightly concave, not tapered outward, or only poorly
and short; nasals relatively long (39% of the skull length, so; temporal surface flat; transfer skulls prominent,
compared with 37.6% in O. punjabiensis, and 37% in O. well projected.
vignei and O. bochariensis), braincase narrow (28%, com- Yellowish rusty; no light spots on the sides. Tress
pared with 31.2% and 30%), toothrow short (25%, com- luxuriant; light-colored beard on the throat and the
pared with 31.2% and 29%). base of the lower jaw; white bib; black neck ruff.
Horns everted, twisted spirally. Horn length Mountains of extreme S Turkmenistan (Karabil’,
720 mm, basal girth 200 mm, horn spread 500 mm. Chengurets Mountains E of Kushka, Badkhyz, Gyaez’-
Triangular in cross section; frontal surface narrow; Gyadyk), Kopet-Dagh, Great Balkhan, E coast of
nuchal surface concave, bases close together. Mild an- the Caspian Sea, Mangyshlak, Ustyurt, NE Iran,
gulation on the external edge of the horns, and an even Afghan istan.
poorer one on the internal edge. Horn color grayish Mangyshlak sheep are often separated from Kopet-
brown. Dagh sheep as a subspecies, Ovis cycloceros arkal Evers-
Body rusty reddish, with reddish spots on the mann, 1850 (e.g., by Zalkin, 1951); they differ primar-
cheeks. Beard long, white, and narrow, extending as far ily in the absence of animals with perverted horns, and
as the corners of the lower jaw, where it is bordered on by having a more intense beard, both of which are
both sides with dark brown streaks. Saudi Arabian merely extreme developments of O. cycloceros.
specimen (D. L. Harrison & Bates, 1991:figure 273) with Our analysis fi nds that O. cycloceros, O. bocharien-
a clear dark line on the flanks; white underside; white sis, and O. vignei are, in fact, distinct. The O. vignei
below the knees and the hocks. sample includes specimens from montane Tadjiki-
Wadi Kharbora and Hatta, Oman; and near stan; SW lowland Tadjik specimens are within O. bo-
Sharawrah, on the SE edge of the Rub-al-Khali, Saudi chariensis; and our O. cycloceros specimens are from
Arabia. Uerpmann (1987) discussed the distributional within the range assigned to O. c. arkal. Skull length
problems raised by the discovery of O. arabica. Its skull, and breadth, and the frontal arc, contrasted with the
in our analysis, is classified very strongly with the O. low frontal chord, distinguish O. cycloceros, especially
vignei group, even though it seems counterintuitive from O. bochariensis; greater skull height and nasal
that it should belong to this species-group, and not to length plus the frontal arc, contrasted with the low
the O. gmelini group, which lives on the opposite side frontal chord, distinguish O. vignei from both the
-1— of the Persian Gulf. others.
0—
349-47558_ch01_1P.indd 236 5/31/11 7:05 PM

b o v idae 237
Ovis severtzovi Nasonov, 1914 to be a urial as verified by Sapozhnikov (1976),” and

Kyzylkum sheep O. severtzovi indeed fall in with the O. vignei group in
2n = 56, as in argali (Lyapunova et al., 1997; Bunch our craniometrics. On the other hand, for Vorobeev
et al., 1998). & van der Ven (2003), it is an argali; this is supported
Skull length 260–270 mm (males, mean 265 mm). by its karyotype, as well as some other characteristics,
Lacrimal depression poorly developed. such as the poor development of the bib in young ani-
Horns homonymous, resembling O. cycloceros. mals (developing, however, with age), and the waxy-
Maximum horn length 730 mm, basal circumference colored horns. Considering its distribution, between
230–260 mm (mean 240 mm). Fronto-orbital and the ranges of these two major species-groups, and the
frontonuchal edges sharp. Vorobeev & van der Ven possibility that it may not be entirely isolated from
(2003)—who treated O. severtzovi as an argali— stated either of them, we suggest that O. severtzovi is a (per-
that the horns are black, whereas in “other argali” fectly valid) species of hybrid origin.
they are wax-colored; in fact, it seems that the horns
are waxy-colored when clean, but become dark be- Existing Hybrid Populations
cause of ingrained dirt from wear and use (Atanas Alborz Red Sheep Hybrid (Ovis gmelini × Ovis
Tchobanov, pers. comm. to CPG). cycloceros arkal)
Darker; grayish cinnamon to chestnut or grayish 2n = 54 or 55 (Imperial Reserve, just E of Teheran, 35°
brown. Tress weakly developed, short, not reaching 41' N, 51° 34' N); 2n = 55–58 (Parvar Protected Re-
the base of the head; beard absent. Rump-patch very gion, 50 km N of Semnan, 36° 06' N, 53° 35' E). Valdez
pronounced. Slightly paler on the neck; grayish brown et al. (1978) found that in Parvar, one individual had
on the flanks; belly and rump white. Mane white, 2n = 55, eight had 2n = 56, seven had 2n = 57, and eight
tinged with gray. Head darker than the neck, with had 2n = 58; in the Imperial Reserve, five had 2n = 54
white face markings; legs dirty white, with dark, red- and nine had 2n = 55—that is to say, there was no evi-
dish brown stripes. Young animals redder. dence of any deficiency of heterozygotes.
High altitude locations in the Kyzylkum Desert, The name given to this hybrid population is Ovis
especially Aktau and Nuratau. The Nuratau Ridge, in orientalis Gmelin, 1774, described from the mountains
the Kyzylkum, is an extreme NW spur of the W Pamir- of Gilan and Mazanderan provinces, Persia (restricted
Alai Range that is inhabited by O. bochariensis; it is sepa- by Nasonov [1923] to the E part of the Alborz Moun-
rated from the Kara-Tau by the Golodneya Steppe and tains). The subspecies Ovis orientalis erskinei Lydekker,
the Syr Darya River. 1904, was described from the same general area. Val-
Vorobeev & van der Ven (2003) treated O. se- dez (1982:100) illustrated a skull in the Zoological Mu-
vertzovi as an argali, but they admitted that it is some- seum of the Academy of Sciences in St. Petersburg
what similar to a urial in several characters: darker (specimen ZMAS 937), collected by Gmelin, which
than other argali found in Kyrgyzstan, with only the may be the type, or at any rate from the type series; it
tip of the nose being light; belly noticeably lighter, has a sharp fronto-orbital edge, yet supracervical
grayish; virtually no white spots on the face; long horns. Valdez specified that homonymous, supracer-
hair on the neck light, short; long white ruff (or white vical, and intermediate horn types all occur, espe-
interspersed with dark hairs); and white bib. Lyapu- cially in the vicinity of Teheran; farther W, where the
nova et al. (1997) found that in a 4-year-old ram, the hybrid O. orientalis blend with pure Armenian mou-
bib was poorly developed, like an argali; whereas in a flon (O. gmelini), supracervical horns predominate; to
6-year-old ram, it was prominent and snow-white, the E, where they blend with Transcaspian O. c. arkal,
sharply contrasting with the background coat, which the horns tend to be homonymous. A saddle-patch,
was blackish on the anterior side of the forelegs and variable in size, may be present or absent; the neck
brown on the shoulders. ruff is highly variable, and may be either similar to
The weak growth of the tress resembles argali; the that of O. gmelini or bibless, or with either a full-length
skull and horns resemble the urial species O. cycloceros, black bib or a white one and a long black ruff.
but, as noted by Lyapunova et al. (1997), the horns ex- In our craniometric analysis, three skulls of “Al-
tend from the skull at a lower angle than in urial. borz hybrids” are classified strongly with the O. vignei
O. severtzovi has been shuffled back and forth be- group, and a further Alborz skull is marginal between
tween the urial and the argali groups. Valdez (1982) the O. gmelini and the O. vignei groups (O. c. arkal be-
stated that “its skull characteristics clearly establish it longing to the latter); while, of two skulls from “North —-1
—0
349-47558_ch01_1P.indd 237 5/31/11 7:05 PM

238 Artiodactyla
Persia,” one identifies with each group. Of two skulls Horns homonymous, up to 1020 mm long; basal cir-
from Shahroud, 36.25° N, 55.00° E (in the far E Al- cumference 285–335 mm (mean 310 mm). Geist (1991a)
borz, SE of Gorgan), one is more vignei, and the other noted that the horns of old rams have a well-developed
is marginal between the O. vignei and the O. gmelini combat edge (as they do in bighorn sheep).
groups. Darker than O. cycloceros, brownish, with a cinna-
mon or rusty tinge. Tress poorly developed; elongated
Kerman Hybrid (O. laristanica × O. cycloceros) hair (up to 11 cm) only on the chest and the lower por-
2n = 54 and 55 (Khabr-va-Rouchoun Wildlife Refuge, tion of the neck.
28° 45' N, 56° 23' E; Kavir National Park). In our Geist (1991a) described O. nigrimontana as being
analysis, we had only a skull from Panjgur (Mekran), dark brown above, lighter on the neck, with the head
26.58° N, 64.06° E ( just E of the Pakistani border darker than the neck color; the ruff creamy white
with Iran), which is well within the morphological (but comparatively short-haired and gradually blend-
range of the O. gmelini group. ing into the darker shoulder and body hair); a dis-
The horns are homonymous, and the frontal sur- tinctly set-off white rump-patch that does not sur-
face is flat, with sharp angles. In winter, their color is round the tail; a white belly, set off by a dark flank
darker than in O. laristanica, with a white saddle- stripe; a dark rostrum; and a long tail.
patch. The Kerman hybrids are bibless. A black ruff is Karatau, 43° N, 70° E°.
present, but it is variable in length: in some it extends Based on the mtDNA control region, Wu C-h. et al.
the length of the neck, while in others it is limited to (2003) found that O. nigrimontana is the sister clade to
the lower half. other argali (but the bootstrap for the other clade is
Kerman Province in the Khabr-va-Rouchoun only 73%).
Wildlife Refuge and E and S, at least to the Pakistan O. nigrimontana is the other species, apart from O.
border. severtzovi, whose allocation to a species-group has
been regarded as equivocal, though in this case, it is
Ovis ammon Group more evidently an argali. To test its affinities, and those
argali of O. severtzovi, we performed a special analysis. O. se-
A study of the mitochondrial control region in the vertzovi fell in with O. vignei, and O. nigrimontana with
argali (O. ammon) group (Wu C-h. et al., 2003) found O. polii—but, in both cases, toward the edges of these
that, using domestic sheep as outgroups, O. nigrimon- respective dispersions.
tana was the fi rst to separate, followed by a four-way
split: O. severtzovi; O. darwini; O. “adametzik”(?) plus Ovis karelini Severtzov, 1873
O. sairensis (i.e., O. collium; one specimen of each); and Tianshan argali
O. hodgsoni plus O. dalailamae (three and two samples
1873 Ovis heinsii Severtsov.
respectively, forming somewhat separate clades).
1898 Ovis sairensis littledale Lydekker.
This group of sheep was revised in detail by Geist
1913 Ovis ammon humei Lydekker.
(1991a), and his revision is broadly followed here, ex-
1914 Ovis polii karelini forma melanopyga Nasonov.
cept that we treat all the subspecies as distinct species,
1929 Ovis polii nassonovi Laptev.
since their differences are discrete and consistent.
2n = 56.
Ovis nigrimontana Severtzov, 1873 Skull length 305–335 mm (males, mean 324 mm).
Karatau argali Facial part longer, higher, and broader, 60%– 63% of
2n = 56 (in zoo specimens from Karatau; Schmitt & the maximum skull length. Nares longer and broader.
Ulbrich, 1968), as in other argali; but the Y chromo- Interpterygoid fossa broad; condyles very large. Skull
some is acrocentric, not metacentric as in other Ovis, behind the horns projecting a little behind the line of
including other argali (Bunch, 2000). the back of the bony cores; upper profi le steep.
Skull length 285–315 mm (males, mean 303 mm), Horn length up to 1290 mm; basal circumference
258–274 mm (females, mean 268 mm). Skull profi le 330– 400 mm (mean 360 mm). Terminal portion of the
steeper behind the horns; occipital projects a little be- horns relatively weakly bent sideways, rising up al-
yond the line of the back of the bony horn cores; facial most parallel to the base. Outer ridge not so distinct;
part higher and broader; prealveolar portion of the base of the horn blunted.
-1— upper jaw somewhat longer; nares and nasal bones Lighter, either brownish yellow or brownish cin-
0— broader. namon. Tress absent, but the hair on the front of the
349-47558_ch01_1P.indd 238 5/31/11 7:05 PM

b o v idae 239
chest and the base of the neck slightly elongated (up part of the horns highly twisted outward. Geist (1991a)
to 10 cm). Geist (1991a) specified that O. karelini differs characterized O. polii as having well-developed combat
from O. polii by having, in winter, a distinct, dark ridges that are triangular in cross section; the horns
flank stripe segregating the dark body hair from the are normally widely flaring, relatively thin, and light
light belly hair, with the ruff long-haired and white. in color, and, as life progresses, they break, rather than
Vorobeev & van der Ven (2003) described males in “broom,” as do those of other argali.
winter as mainly gray-colored, light brown on the Dark grayish reddish brown, sometimes chocolate.
head, while the neck, the legs, and the belly are beige; Tress on the neck and the chest slightly larger, reaching
in summer, males are bright red to dirty brown. 18 cm. Geist (1991a) described the color as light reddish
Tianshan S to Naryn and the Fergana Range, W to brown to chocolate-brown, usually pale grayish brown
the W extremity of the mountains, N to the Chu-Ili or buff. Secondary rump-patch present (i.e., a rump-
Mountains. patch extending onto the thighs, the hindlegs, the belly,
the chest, and the forelegs, turning these parts white);
Ovis collium Severtzov, 1873 tail length 10 cm, without hair, and without a median
Kazakhstan argali dark stripe. Light flank stripe. Face light. No ruff on the
males in summer, nor at any season on the females.
1898 Ovis sairensis Lydekker.
Vorobeev & van der Ven (2003) described the win-
1923 Ovis polii collium forma albula Nasonov.
ter coat as white–grayish brown, with the back brown-
1923 Ovis polii collium forma obscura Nasonov.
white to pale brown; old animals have white faces.
Size as in O. karelini. Ascending processes of the pre- The legs were grayish in 5- to 7-year-old males. Young
maxillae very long, extending posteriorly between the animals were more creamy colored.
nasals and the maxillae, their ends separated from the Pamir, E extremity of Alai, the Aksai Plateau and the
lacrimals only by a short distance. Chatyrkel’ region, Kokshaal to Khan-Tengri in the N,
Outer ridge of the horn more faintly developed and Dzhungar Alatau. Also occurs in the Chinese sector
than in O. karelini; temporal surface more convex. of Tianshan and the mountains adjoining the Pamirs.
Geist (1991a) noted that the horns are similar in shape Very closely related to O. karelini. Geist (1991a)
to those of O. polii, but arise from the skull at a steeper stated that “the segregation in pelage and horn char-
angle, tend to be shorter and thicker, with increased acteristics may not be complete: there appear to be
rounding of the combat edges, and have a basal cir- enough differences in pelage, horn, and qualitative
cumference of 330– 400 mm. skull characteristics.” We tested this: are O. karelini and
Dark, vivid, brownish cinnamon (light-colored O. collium different from O. polii? Not when skulls alone
animals also known). Geist (1991a) described O. col- are considered, but both are quite distinct when the
lium as darker than O. karelini, with the rostrum gray horns are taken into consideration. O. polii and O. col-
rather than white. lium have a much shorter horn core length—vis-à-vis
Kazakhstan W to about 67°–70° E, E to the Irtysh, sheath length, core circumference, and skull breadth—
S to the Saur and Tarbegatai (including in NW Dzun- compared with O. karelini. O. collium—compared with
garia), and the N shore of Lake Balkhash. O. polii and, to a lesser extent, with O. karelini—has a
Geist did not separate this taxon from O. karelini, lower, narrower skull vis-à-vis its length.
but our craniometric analyses distinguish it clearly
by the much shorter horn cores compared with the Ovis hodgsoni Blyth, 1841
sheaths, and by the relatively smaller core circumfer- Tibetan argali
ence; in addition, the skull is narrower and lower, com-
1841 Ovis ammonoides Hodgson.
pared with O. karelini.
1873 Ovis blythi Severtzov.
1874 Ovis brookei Ward.
Ovis polii Blyth, 1841
1888 Ovis dalai-lamae Przewalski.
Pamir argali, Marco Polo sheep
1892 Ovis henrii Milne-Edwards.
Skull length 328–355 mm (males, mean 341mm), 283–
1913 Ovis poli [sic] adametzi Kowarzik.
305 mm (females, mean 295 mm). Skull behind the
horns projecting even less. Upper profi le steeper. In- Skull length 357–359 mm.
terpterygoid fossa broader. Horns in old rams with a less-developed combat
Horn length sometimes reaching 1900 mm; basal edge than in O. polii, less everted, and usually broomed —-1
circumference 350– 400 mm (mean 370 mm). Terminal instead of broken. —0
349-47558_ch01_1P.indd 239 5/31/11 7:05 PM

240 Artiodactyla
Uniformly brown (no light saddle-patch between Small in size; skull length 325–357 mm (mean 339 mm).
the shoulders), but in spring, the haunches appearing Skull foreshortened; nasals noticeably short. Horns
lighter than the torso. Long-haired, light-colored ruff; massive, heavily crenellated, broomed, and tight.
rump-patch, distinctly set off in the upper parts, sur- For Geist (1991a), “two extremes [of color] are
rounding the tail; tail tiny, about 2.5–5.8 cm; dark, par- lumped under ’Gobi argali.’ ” One, in winter, is dark
tial flank band separating the light belly hair from the chocolate-brown, with a gray-brown muzzle and face,
dark body hair; distinct segregation into light and dark the muzzle barely lighter than the rest of the face. The
fields of hair on the hindlegs and the forelegs, with light portions (the lower limbs and the underside) are
(normally) a dark stripe from the chest to the hoofs. buff to light brown, and blend into the dark body hair.
Whole Tibetan plateau. The other, in winter, has dark brown hair, light lower
limbs, a large rostral spot and muzzle, and large light
Ovis ammon Linnaeus, 1758 areas behind the shoulders and on the withers. The
Altai argali ruff is extensive, but spotted with dark patches.
Tail length, minus hair, 7.5–10 cm. Overall, O. dar-
1785 Ovis argali Boddaert.
wini is very similar to O. ammon, except for the light
1873 Ovis argali altaica Severtsov.
ruff and smaller size.
1873 Ovis argali dauricia Severtsov.
Low, isolated, dry desert ranges, such as the Gobi
1873 Ovis argali mongolica Severtsov.
Altai, Yabarai, to about 112° E.
1923 Ovis ammon przevalski Nasonov.
Is O. darwini distinct? We performed a craniomet-
2n = 56 (Gorno-Altaisk Autonomous District, W spurs ric analysis: it is. Compared with both O. ammon and
of the Chikhachev Mountains, 50° 03' N, 89° 25' E). O. hodgsoni, O. darwini has a shorter, narrower, but
Karyotype 1.0 (Mongolian Altai, 46° 30' N, 93° 00' E). relatively somewhat higher skull.
Skull length 345–395 mm (males, mean 368 mm),
305–337 mm (females, mean 318 mm). Skull very simi- Ovis jubata Peters, 1876
lar to O. polii, differing in the longer anteronasal profile Shansi argali
and the greater width of the brain case. Skull length 315–335 mm.
Horn length up to 1600 mm; basal circumference Horn length up to 320 mm, basal circumference
400–550 mm (mean 440 mm). Outer rib usually 400–500 mm.
weakly developed, sometimes disappearing in the Dark fawn-gray; light to dark ruff, twice as long as
basal portion. Temporal surface convex, especially at the body hair, blending into the dark shoulder hair;
the base, with two weak combat ridges. Horn surface well-developed dorsal neck mane. Rump-patch large,
strongly wrinkled. white to buff, distinctly bordered; shoulders and back
In summer, dark brown to almost white, with rufous flecked with white in very old rams; belly light, but
patches; in winter, dirty white. In summer, the rump- grading into the dark body hair; lower front legs
patch diffuse, sometimes absent; in winter, the rump- creamy; distinct stripes down the forelimbs and the
patch distinct, bordered dorsally, lacking a secondary hindlimbs. Tail length 7.5 mm, without hair, but with
expansion on the thighs. Light underside often blending a dark median stripe.
into the dark body hair. Tail long, with a dark stripe on Shansi, Hebei, Inner Mongolia, Shensi.
the top, sometimes surrounded by a rump-patch. Lower We have no experience of this taxon, and thus rely
limbs light, sometimes with scattered dark hair. Occa- on Geist’s (1991a) description and analysis.
sionally a light rostral spot; muzzle and lower jaw light.
Russian and Mongolian Altai, Tuva. Formerly Ovis nivicola Eschscholtz, 1829
found in Trans-Baikalia (this is, in fact, the type local- snow sheep
ity), but now extinct there. 2n = 52 (Korobitsyna et al., 1974).
O. nivicola are related to O. dalli and O. canadensis
Ovis darwini Przewalski, 1883 from North America, but the evidence clearly indi-
Gobi argali cates that they form a strongly distinct species.
Bunch (2006), on the basis of mitochondrial cyto-
1873 Ovis argali mongolica Severtsov.
chrome b sequences, timed the separation of North
1913 Ovis kozlovi Nasonov.
American sheep (O. canadensis and O. dalli) from other
1919 Ovis comosa Hollister.
-1— sheep, including O. nivicola, at more than 3 Ma. Interest-
1936 Ovis ammon intermedia Gromova.
0— ingly, in the majority-rule tree (his figure 3), O. nivicola
349-47558_ch01_1P.indd 240 5/31/11 7:05 PM

b o v idae 241
was on the O. ammon / O. gmelini branch, not with the 2. Okhotsk snow sheep, Ovis nivicola alleni
North American sheep, although the support values Matschie, 1907 (synonyms—Aegoceros (Ovis)
were not high; nonetheless, it is of note that one of the montanus, O. middendorfi, O. n. potanini)
four snow sheep sequences, the one from the Taygonos —Size, skull, and horns similar to O. n. nivicola.
Peninsula, was further up the same clade, while the — Color usually dark, though lighter individu-
other three (from the Maimandzhinskii Range and als are known. Head, neck, and chest
Cape Yevreinova, farther to the W) were sister to the brown; tip of the muzzle whitish, nose-
rest of this clade. The author proposed that the basal bridge black. Forehead to occiput white.
Ovis had 2n = 52, like O. nivicola (and also Budorcas), Body lighter than the neck. Dark band along
from which, independently, were derived the O. ca- the spine. White spot on the flanks behind
nadensis / O. dalli number of 2n = 54, the O. ammon the axilla. Dark band running along the
group’s 2n = 56, and the O. gmelini / O. vignei progres- lower flanks from the elbow to the groin
sion of 2n = 54 and 58. invariably distinct in light-colored animals.
Chernyavsky (1962) found the following: Rear half of the abdomen white.
—Stanovoi Range, Dzhugdzhur Range, SW
1. In O. canadensis, the relative length of the
parts of the Kolyma Range, Taygonos Peninsula.
nasals is greater than in O. dalli and O. nivicola.
3. Yakutian snow sheep, Ovis nivicola lydekkeri
2. The skull of O. nivicola is significantly wider in
Kowarzik, 1913 (synonyms—O. lenaensis, O.
the orbits than in the two American species,
n. lydekkeri forma albula, O. n. lydekkeri forma
at 49.2% of the basal length (minimum 47%),
obscura)
compared with 44.9% (O. dalli) and 44.5% (O.
—Bony horn core shorter (150–210 mm, mean
canadensis) (maximum in both, 46%).
175 mm), thinner; basal circumference
3. In O. nivicola, the rostral and occipital parts are
250–280 mm (mean 260 mm). Horns much
relatively broader than in the North American
shorter (up to 910 mm) and lighter; basal
species.
circumference up to 250–330 mm (mean
4. Unlike the American species, the white
290 mm).
rump-patch in O. nivicola does not extend onto
— Color very light, sometimes almost white or
the back above the base of the tail.
yellowish white. Dark brown shades on the
In the Late Pleistocene and Early Holocene, snow cheeks, the neck, the chest, and the front of
sheep were found as far SE as the Russian Far East and the legs. Dark band along the spine. Axillae
as far N to the Yana River delta, SW into W Siberia brown; brown bands running posteriorly.
(Boyeskorov, 2001). The shortness of their extremities Underside white, or with some slight
shows that, in contrast to the O. ammon group, snow darkening in the chest. Front of the nose,
sheep are adapted to montane rather than rolling or the forehead, the top of the head, the tips of
plateau habitats. Nonetheless, in some regions in S Si- the ears, and the upper neck light or
beria, they were possibly sympatric. whitish. No dark band on the nose-bridge.
Interestingly, considering the enormous distribu- —From the Lena River in the W (region of the
tional range of the species, the described subspecies, Verkhoyansk Range and other ranges), N parts
listed below, seem to be rather weak (following Hept- of the Kolyma Range, and the Anadyr region.
ner et al., 1961; and other sources): 4. Putorana snow sheep, Ovis nivicola borealis
Severtzov, 1873
1. Kamchatka snow sheep, Ovis nivicola nivicola
—Very little known; dark, fairly uniform,
Eschscholtz, 1829 (synonym—O. storcki)
closely resembling O. n. nivicola.
— Skull length 265–300 mm (males, mean
—Isolated in Putorana (Noril’).
280 mm), 245–265 mm (females, mean
5. Chukotka snow sheep, Ovis nivicola tschuktcho-
255 mm). Horn sheath 170–270 mm (mean
rum Zheleznov, 1994
216 mm); basal circumference 250–325 mm
—According to Železnov (1994:169), “both sexes
(mean 300 mm). Horn core maximum
lack the yellow color and males are without
110 mm; basal circumference 290–360 mm.
the big white stain which is characteristic of
— Color uniform, fairly dark, without signifi-
the Okhotsk [O. n. alleni] and Yakutsk [O. n.
cant lightening on the flanks and the head.
lydekkeri] subspecies. The color is generally —-1
—Kamchatka.
bright with brownish-grey shades.” The —0
349-47558_ch01_1P.indd 241 5/31/11 7:05 PM

242 Artiodactyla
ProOutfitters website (!) says O. n. tschuk-

tchorum has a heavier, lighter coat [light in Ovis canadensis Shaw, 1804
color?] of soft, woolly hair. Železnov (1994) bighorn sheep
also said the “craniometric measurements” Cowan (1940) recognized several subspecies. Ramey
are different, but no measurements appear (1995), using morphometrics, proteins, and mtDNA,
to be given in the paper, which is mainly in could fi nd no difference between the four supposed
Russian (with an English abstract). subspecies of desert bighorn, and recognized only
—“It occurs in the Anadyr district on the O. c. nelsoni. While bighorn from the Rocky Moun-
Chukot Peninsula, without the Koriacky tains (O. c. canadensis) are certainly larger than those
Mountains; from hilly country in an from the desert ranges (O. c. nelsoni), the size over-
elevation of 500 m to mountains of an laps greatly, and other differences are not consis-
elevation of 1300 m” (Železnov, 1994:169). tent, so we question whether there are any subspe-
6. Koryak snow sheep, Ovis nivicola koriakorum cies at all.
Chernyavski, 1966
—Very similar in coloration to O. n. nivicola, but Ovis dalli Nelson, 1884
about 30% smaller; the white muzzle and dall sheep
rump-patch not as pronounced. Compact in Skull longer and wider than in O. canadensis; horns
size, with shorter legs in proportion to the thinner. Color white, or black, or particolored.
body than its S cousin (O. n. nivicola). The S males (the putative subspecies O. d. stonei) have a
ProOutfitters website notes that individuals larger skull size than N males; the white N form has a
found at the N extremity of the Koryak Range basal length of 235–256 mm, the black S form 258–274
exhibit characteristics of O. n. tschuktchorum. (Cowan, 1940).
7. Kodar snow sheep, Ovis nivicola kodarensis A study of the mtDNA control region by Loehr
Medvedev, 1994 et al. (2006) found that most of the population of O.
—The original description gives the following dalli in British Columbia possessed haplotypes evi-
characters: “The suture between os presphe- dently derived from the O. canadensis population of
noidale and os sphenoidale is absent. Os the Canadian Rockies, although long enough ago
lacrimale is very narrow and, in adult (the authors proposed that this would be prior to the
individuals, this bone is separated from last glacial maximum) that the bighorn-derived hap-
nasal bones by connections of appendixes lotypes in British Columbian O. dalli are somewhat
[shoots?] of maxillar and frontal bones. They different, forming their own cluster. They also
have narrow meatus acusticus, long and pointed out that the coloration of O. dalli changes
massive tooth row. Males have a nearly oval clinally across the range, from the pure white form
transverse section of the processus cornutus in the N to the dark O. d. stonei form in the S; that is
in contrast to Yakutian [O. n. lydekkeri] and to say, there is, in fact, no division between the two
Okhotian [O. n. alleni] subspecies which have near-homogeneous populations. Therefore, we are
a rounded-triangular section. Width of skull reluctant to recognize any taxonomic distinction,
is maximal for Siberian snow sheep in Asia both because the color changes clinally, and the pre-
but the length is comparatively short [?].” cise relationship of color to size is unclear; and be-
—The Kodar mountain ridge, E of Lake cause the molecular evidence, cited above, suggests
Baikal, is “a narrow and non-isolated area” that the larger size of the S populations may relate to
(G. G. Boyeskorov, pers. comm. to CPG). gene flow from O. canadensis.
Are the taxa of the O. nivicola group distinct? The
Myotragus Bate, 1909
results of our analyses are equivocal. They seem to
Balearic sheep
show that no, on the characteristics of horns, they are
We list this genus here because it survived into the
not distinct; and that taking the skull plus the horns
Holocene. General information about the terminal
together, O. n. lydekkeri differs, on average, from the
species, Myotragus balearicus Bate, 1909, and its evolu-
others by its greater horn sheath compared with horn
tion, was given by Bover & Alcover (1999). Its extinc-
core circumference, and by its shorter, less thick horns.
tion is put between 3700 and 2030 BC on Mallorca, be-
Table 64 gives the univariate statistics for the skulls
-1— tween 10,000 and 1930 BC on Menorca, and between
and horns of non–North American male Ovis.
0— 3650 and 300 BC on Cabrera (Bover & Alcover, 2003).
349-47558_ch01_1P.indd 242 5/31/11 7:05 PM

Table 64 Univariate statistics for the skulls and horns of Ovis: males
Horn Horn
Max Horn core Horn sheath
Skull l Nas l Fr chord Fr arc teeth Skull ht Skull br core l circumf sheath l circumf
gmelini
Mean 252.14 93.29 97.14 127.86 74.71 129.00 132.57 310.00 185.00 587.14 238.57
N 7 7 7 7 7 6 7 3 4 7 7
Std dev 5.900 4.716 6.122 4.880 2.360 4.690 3.101 26.458 12.910 64.991 17.491
Min 246 87 86 120 71 123 128 290 170 460 220
Max 264 98 105 135 78 135 137 340 200 670 270
Alborz hybrids
Mean 263.14 104.29 93.00 128.86 75.33 139.17 140.50 363.33 206.67 795.00 248.33
N 7 7 6 7 6 6 6 3 3 6 6
Std dev 9.494 5.407 6.325 7.883 6.154 6.145 5.541 41.633 5.774 66.858 9.309
Min 252 97 85 118 67 129 134 330 200 690 240
Max 281 113 103 140 84 146 147 410 210 890 260
isphaganica
Mean 250.33 93.00 93.25 117.50 70.86 131.50 130.88 230.00 165.00 401.40 210.00
N 6 8 8 2 7 8 8 1 1 5 2
Std dev 15.042 2.507 5.007 — 2.911 6.141 4.883 — — 21.767 .000
Min 225 87 88 110 67 123 125 — — 370 210
Max 262 95 100 125 74 145 138 — — 430 210
laristanica
Mean 238.50 87.00 87.50 114.00 70.50 126.50 127.00 300.00 165.00 480.00 190.00
N 2 2 2 2 2 2 2 2 2 1 1
Min 225 80 83 103 68 110 114 240 145 — —
Max 252 94 92 125 73 143 140 360 185 — —
arkal
Mean 273.39 104.83 94.12 133.00 75.79 148.00 144.47 348.33 199.84 739.72 260.50
N 36 35 24 26 33 29 34 18 19 29 30
Std dev 7.530 6.128 6.707 10.855 3.305 6.897 5.440 50.322 16.863 116.820 18.200
Min 252 95 81 112 70 136 132 275 170 490 208
Max 297 119 108 150 81 162 155 480 225 990 295
bochariensis
Mean 250.40 93.30 91.13 125.50 73.10 135.67 130.90 310.83 195.38 610.56 248.89
N 10 10 8 8 10 9 10 6 8 9 9
Std dev 12.937 5.078 5.668 9.562 3.348 7.714 9.098 48.828 20.028 91.598 23.289
Min 233 86 83 115 69 124 118 250 173 510 220
Max 272 102 100 140 81 144 151 380 230 730 300
vignei
Mean 260.90 95.67 103.00 139.50 96.60 127.33 138.70 348.75 288.33 688.33 254.17
N 10 9 8 8 10 9 10 4 6 9 6
Std dev 9.158 5.723 9.472 9.739 43.482 31.468 5.539 76.635 209.396 127.892 15.943
Min 245 87 92 130 72 72 131 285 180 515 235
Max 275 103 117 158 179 163 148 460 715 930 275
blanfordi
Mean 260.00 109.00 107.00 155.00 72.00 148.00 134.00 — — 750.00 —
N 1 1 1 1 1 1 1 — — 1 —
punjabiensis
Mean 245.00 91.33 89.67 123.17 70.17 136.17 132.17 331.67 182.50 695.00 240.00
N 6 6 6 6 6 6 6 3 4 3 1
Std dev 5.514 5.125 3.559 12.090 3.920 3.869 6.555 106.105 17.059 107.587 —
Min 238 81 83 100 66 132 122 245 163 585 240
Max 252 94 93 135 77 142 142 450 202 800 240
arabica
Mean 243.00 96.00 — — 60.00 — 133.00 — 200.00 720.00 —
N 1 1 — — 1 — 1 — 1 1 — —-1
(continued) —0
349-47558_ch01_1P.indd 243 5/31/11 7:05 PM

Horn Horn
severtzovi
Mean 265.00 83.40 96.00 — 72.60 143.00 141.40 300.00 190.00 636.00 245.00
N 2 5 5 — 5 1 5 3 3 5 5
Std dev 7.071 10.668 3.742 — 3.362 — 3.647 45.826 5.000 85.615 13.229
Min 260 70 93 — 69 — 138 260 185 550 230
Max 270 93 102 — 76 — 147 350 195 730 265
nigrimontana
Mean 300.17 116.50 104.00 — 85.00 162.33 161.67 337.50 237.50 908.00 308.00
N 6 6 6 — 5 6 6 4 4 5 5
Std dev 10.128 10.075 7.823 — 2.550 4.274 2.251 23.979 10.408 108.028 19.558
Min 285 102 98 — 83 157 158 315 225 770 285
Max 315 129 119 — 89 168 164 370 250 1020 330
karelini
Mean 324.74 124.05 117.73 168.11 92.32 175.17 175.53 374.38 278.50 1033.61 351.39
N 19 19 11 9 19 18 19 8 10 18 18
Std dev 10.545 8.727 7.989 11.050 4.230 8.024 8.815 30.640 24.043 124.615 22.932
Min 303 108 102 150 85 153 160 300 240 740 300
Max 338 140 126 190 102 186 191 395 320 1270 400
polii
Mean 341.45 132.52 128.16 194.06 90.37 177.14 184.58 412.50 311.67 1286.79 383.19
N 31 25 19 17 27 22 31 8 12 28 26
Std dev 11.051 9.042 8.255 17.444 3.553 28.673 5.772 78.921 21.794 224.715 24.516
Min 306 115 116 165 83 120 174 300 275 780 325
Max 355 152 152 235 96 214 198 515 345 1595 430
collium
Mean 315.00 117.00 107.00 — 86.71 162.83 180.86 319.00 257.00 1037.86 —
N 7 7 1 — 7 6 7 5 5 7 —
Std dev 13.976 12.369 — — 4.309 20.134 7.925 112.606 73.621 89.063 —
Min 290 97 — — 79 122 165 125 130 920 —
Max 335 138 — — 91 175 188 420 320 1150 —
ammon
Mean 366.42 142.61 — — 95.59 191.95 197.32 393.33 340.56 1115.56 451.00
N 19 18 — — 17 19 19 9 9 18 15
Std dev 12.144 10.013 — — 3.792 12.860 4.900 31.623 21.279 153.392 30.308
Min 345 125 — — 90 171 190 360 300 820 400
Max 388 158 — — 101 214 208 440 370 1450 505
darwini
Mean 298.33 94.33 111.33 — 89.67 138.67 166.00 245.00 193.33 643.33 240.00
N 3 3 3 — 3 3 3 3 3 3 3
Std dev 11.547 15.822 4.509 — 9.292 18.583 5.000 118.216 56.862 218.880 80.467
Min 285 77 107 — 79 118 161 120 130 440 150
Max 305 108 116 — 96 154 171 355 240 875 305
hodgsoni
Mean 352.00 136.58 140.00 — 89.62 190.83 189.88 365.00 338.75 1068.75 431.91
N 16 12 2 — 13 6 16 4 4 12 11
Std dev 10.602 16.082 — — 5.026 12.844 11.848 20.817 26.575 64.390 20.413
Min 335 110 136 — 80 173 150 340 310 1010 410
Max 375 162 144 — 97 204 203 390 370 1250 480
nivicola
Mean 281.78 90.22 115.56 170.00 74.35 151.59 173.09 219.48 302.10 813.87 327.17
N 32 32 9 1 26 29 33 29 31 31 30
Std dev 5.751 4.591 3.941 — 3.019 5.729 7.217 21.016 22.278 123.772 18.555
-1— Min 272 79 112 — 68 140 158 180 250 600 285
Max 296 100 123 — 80 170 187 270 335 1110 360
0—
349-47558_ch01_1P.indd 244 5/31/11 7:05 PM

b o v idae 245
Horn Horn
lydekkeri
Mean 277.15 90.25 110.50 145.00 75.42 144.54 169.00 175.56 258.89 708.46 296.54
N 13 12 6 1 12 13 13 9 9 13 13
Std dev 9.608 5.723 9.503 — 3.728 4.875 6.856 18.105 22.469 123.075 22.303
Min 258 81 103 — 70 138 158 150 230 530 265
Max 291 98 127 — 80 153 178 210 290 910 325
alleni
Mean 274.33 89.67 105.00 — 76.00 144.67 165.33 188.33 268.33 676.67 305.00
N 3 3 1 — 3 3 3 3 3 3 3
Std dev 12.014 6.658 — — 3.464 7.638 9.292 24.664 33.292 205.994 25.981
Min 262 82 — — 74 138 155 160 230 460 275
Max 286 94 — — 80 153 173 205 290 870 320
type of 281 — — — 70 — — 169 — 865 312
lenaensis
mean of 288 90 — — 80 — 167 186 — 1000 320
kodarensis
mean of 269 81 114 — 82 — 167 186 281 713 306
koriakorum
Almost certainly, it became extinct as a consequence drew attention to the distinctiveness and continued
of the earliest human occupation of the islands. survival of the Abruzzo chamois, which, in a brief re-
Although at first Lalueza-Fox et al. (2002), using a view of the genus later (Lovari, 1987), he recognized as
338 bp segment of the mitochondrial cytochrome b a species distinct from the Alpine chamois but conspe-
gene, placed Myotragus as part of a three-way split with cific with that of the Pyrenees, thus attributing two
Ovis and Budorcas, bootstrap support for this place- species to the genus: R. pyrenaica (with subspecies R. p.
ment was below 50%. A later study (Lalueza-Fox et al., parva, R. p. pyrenaica, and R. p. ornata) and R. rupicapra
2005) managed to sequence the entire 1143 bp of the (with subspecies R. r. cartusiana, R. r. rupicapra, R. r. tat-
cytochrome b gene, as well as fragments of 12S and the rica, R. r. carpatica, R. r. balcanica, R. r. caucasica, and R.
nuclear 28S rDNA gene, and on this basis achieved r. asiatica). The two species recognized by him are dif-
higher support for a sister-group relationship just with ferent in cranial features and behavior, as well as gene-
Ovis. They pointed out that the geological date of the tically (Nascetti et al., 1985). Except for the white face
isolation of the Balearics—hence, of Myotragus—is and the white upper throat, with a black line running
5.35 Ma, and on this basis they suggested that the dates from the horn through the eye to the snout, R. rupi-
of certain other splits could be calculated (see above, capra is very dark, and nearly black in winter; whereas
under the Caprini). R. pyrenaica has large, light areas on the neck (separated
from the white throat by a strong black stripe, running
Rupicapra de Blainville, 1816 from the ear to the base of the foreleg) and on the
chamois haunch. In summer, R. rupicapra is lighter, but retains
2n = 58, only one autosomal pair being metacentric the black legs; whereas R. pyrenaica is medium grayish
(Soma & Kada, 1986). brown all over, except (again) for the black legs. The ge-
Relatively slenderly built; short, upright, nearly netic divergence between R. rupicapra and R. pyrenaica
smooth black horns, characteristically hooked back- is considerable; the genetic divergence between the two
ward. Gazelle-like, black-and-white pattern on the putative subspecies R. p. pyrenaica and R. p. ornata is
face; body dark (but varying), always with a sharply considerably less, although these subspecies appear to
white underside. have been separated already by the beginning of the
Until quite late in the 20th century, most authors last ice age (Masini & Lovari, 1988).
recognized only a single species, Rupicapra rupicapra, Study of allelic variation in microsatellites (Pérez —-1
typified by the chamois of the Alps. Then Lovari (1977) et al., 2002) confi rmed that the deepest divergence —0
349-47558_ch01_1P.indd 245 5/31/11 7:05 PM

Table 65 Univariate statistics for Rupicapra: skulls
Gt l Par br Fr br Orb vert Orb horiz Nas l Nas br Max teeth
Males
cartusiana
Mean 212.63 61.00 110.67 36.56 37.89 69.00 19.78 62.22
N 8 9 9 9 9 9 9 9
Std dev 3.068 1.323 2.236 1.236 .782 6.325 1.481 2.108
Min 207 60 107 35 37 58 18 59
Max 216 63 114 39 39 77 22 65
rupicapra
Mean 213.14 63.43 110.14 37.29 39.14 65.83 19.86 58.29
N 7 7 7 7 7 6 7 7
Std dev 4.981 1.512 2.610 .756 .690 6.676 1.345 2.563
Min 205 61 107 36 38 57 18 53
Max 220 65 114 38 40 74 21 61
tatrica
Mean 213.00 — 110.00 — — — — —
N 1 — 1 — — — — —
balcanica
Mean 217.00 62.00 106.17 37.33 39.33 67.00 20.75 59.29
N 2 6 6 3 3 4 4 7
Std dev 2.828 2.098 3.371 1.528 .577 2.582 1.258 .756
Min 215 60 101 36 39 64 19 58
Max 219 65 111 39 40 70 22 60
asiatica
Mean 211.00 — — — — — 28.67 62.67
N 3 — — — — — 3 3
Std dev .000 — — — — — .577 2.082
Min 211 — — — — — 28 61
Max 211 — — — — — 29 65
caucasica
Mean 209.00 64.00 108.25 38.00 37.00 71.25 — 58.50
N 3 4 4 4 4 4 — 4
Std dev 3.464 1.414 2.217 .816 1.414 3.403 — 3.000
Min 207 62 106 37 36 67 — 55
Max 213 65 111 39 39 74 — 61
carpatica
Mean 224.00 64.67 115.00 37.00 41.00 69.00 20.67 60.33
N 2 3 3 3 3 3 3 3
Std dev .000 1.528 2.646 1.000 1.000 6.245 1.528 3.215
Min 224 63 112 36 40 62 19 58
Max 224 66 117 38 42 74 22 64
parva
Mean — — 100.00 — — 50.00 — 53.00
N — — 1 — — 1 — 1
pyrenaica
Mean 201.50 59.50 105.25 35.50 37.00 67.25 18.50 54.75
N 4 4 4 4 4 4 4 4
Std dev 3.697 1.291 2.062 2.082 1.414 3.304 .577 2.062
Min 197 58 103 33 35 63 18 52
Max 205 61 108 38 38 71 19 57
ornata
Mean 204.00 64.50 109.00 35.00 36.00 62.50 — 56.00
N 2 2 1 1 2 2 — 2
Std dev 2.828 .707 — — 2.828 3.536 — .000
Min 202 64 — — 34 60 — 56
-1— Max 206 65 — — 38 65 — 56
0—
349-47558_ch01_1P.indd 246 5/31/11 7:05 PM

b o v idae 247
Gt l Par br Fr br Orb vert Orb horiz Nas l Nas br Max teeth
Females
cartusiana
Mean 206.80 61.20 104.40 38.00 37.80 64.20 19.40 60.40
N 5 5 5 5 5 5 5 5
Std dev 4.266 1.643 1.342 1.414 .837 4.764 .548 2.702
Min 201 59 103 37 37 59 19 57
Max 211 63 106 40 39 69 20 63
rupicapra
Mean 208.10 62.64 105.36 37.18 38.55 65.91 19.27 60.00
N 10 11 11 11 11 11 11 11
Std dev 6.641 1.963 3.906 1.328 1.293 4.482 1.272 3.066
Min 200 59 98 35 36 58 17 55
Max 222 65 112 39 40 73 21 66
tatrica
Mean 211.00 — 106.00 — — — — —
N 1 — 1 — — — — —
balcanica
Mean 208.67 61.67 104.33 36.00 39.00 69.00 20.00 60.20
N 3 3 3 1 2 2 2 5
Std dev 3.055 5.508 6.351 — .000 16.971 .000 2.168
Min 206 56 97 36 39 57 20 58
Max 212 67 108 36 39 81 20 63
asiatica
Mean 211.00 — — — — — — 63.00
N 1 — — — — — — 1
caucasica
Mean 207.00 65.00 106.00 38.00 37.50 69.50 — 58.00
N 1 2 2 2 2 2 — 2
Std dev — 4.243 8.485 .000 .707 3.536 — 1.414
Min 207 62 100 38 37 67 — 57
Max 207 68 112 38 38 72 — 59
carpatica
Mean 216.00 65.00 110.00 35.00 39.00 66.00 19.00 59.00
N 1 1 1 1 1 1 1 1
pyrenaica
Mean 197.00 60.18 101.82 34.73 36.36 64.00 17.09 55.91
N 10 11 11 11 11 11 11 11
Std dev 3.771 1.079 2.994 1.421 1.120 5.348 .539 1.868
Min 189 58 98 31 34 55 16 51
Max 202 62 108 36 38 72 18 58
ornata
Mean 202.00 60.67 102.33 35.33 35.67 63.00 — 56.67
N 1 3 3 3 3 3 — 3
Std dev — 1.155 3.215 .577 .577 1.000 — 2.517
Min 202 60 100 35 35 62 — 54
Max 202 62 106 36 36 64 — 59
was indeed between the R. rupicapra group and the R. samples from the Cantabrian Mountains were quite
pyrenaica group. Within the R. rupicapra group, the di- distinct from those from the Pyrenees, although the
versity within the Alps encompassed that among other specimen from the Apennines was more distinct still.
localities, although samples from the Carpathians, the Pemberton et al. (1989) examined the distinctive-
Tatra, the Balkans, and the Caucasus all occupied their ness of the Chartreuse chamois (R. cartusiana) from —-1
own distinct clades; within the R. pyrenaica group, the Alpine chamois (R. rupicapra) in allelic variation —0
349-47558_ch01_1P.indd 247 5/31/11 7:05 PM

248 Artiodactyla
Table 66 Univariate statistics for Rupicapra: horns Table 66
Horn Horn
Horn br Horn base Horn br Horn base
Horn l tips base ap diam Horn l tips base ap diam
Males rupicapra
cartusiana Mean 199.22 86.00 22.56 20.11
Mean 236.50 90.13 27.87 26.00 N 18 18 18 18
N 8 8 8 8 Std dev 21.945 22.526 1.790 1.875
Std dev 16.000 39.219 2.100 1.195 Min 168 47 19 17
Min 220 56 25 24 Max 251 131 26 24
Max 260 155 31 28 balcanica
rupicapra Mean 243.00 101.00 24.00 22.00
Mean 237.29 96.14 29.36 26.64 N 1 1 1 1
N 14 14 14 14 asiatica
Std dev 10.186 16.129 1.906 1.737 Mean 180.00 75.00 — —
Min 218 75 26 24 N 1 1 — —
Max 259 135 33 30 caucasica
balcanica Mean 185.00 80.00 18.00 15.00
Mean 235.33 101.50 26.50 23.75 N 2 2 2 2
N 3 2 4 4 Std dev 183 48 16 14
Std dev 10.970 — 3.109 3.403 Min 187 112 20 16
Min 223 89 22 19 carpatica
Max 244 114 29 27 Mean 250.00 97.00 25.00 25.00
asiatica N 1 1 1 1
Mean 197.00 59.50 — — pyrenaica
N 3 2 — — Mean 78.44 36.88 16.44 14.56
Std dev 14.799 — — — N 9 8 9 9
Min 180 58 — — Std dev 11.148 4.224 1.509 1.740
Max 207 61 — — Min 55 30 13 11
caucasica Max 93 42 18 17
Mean 220.75 86.25 24.00 22.00 ornata
N 4 4 4 4 Mean 126.00 — 25.50 22.00
Std dev 12.121 12.764 1.155 1.414 N 2 — 2 2
Min 203 72 23 20 Min 117 — 22 19
Max 229 103 25 23 Max 135 — 29 25
carpatica
Mean 232.50 84.00 28.50 26.00
N 2 2 2 2
Std dev 228 76 28 25
across 10 different nuclear loci (a further 45 loci
Min 237 92 29 27 proved to be monomorphic). The Chartreuse cham-
parva ois was more distinct from the Alpine populations
Mean 180.00 9.00 — — than they were from each other, although there were
N 1 1 — — no fi xed differences. Unfortunately, individuals
pyrenaica from the Alpine population had already been intro-
Mean 98.67 45.67 19.67 17.67 duced into the Massif de Chartreuse, and the au-
N 3 3 3 3
Std dev 3.215 10.970 1.528 2.082
thors suggested that, if possible, contact between
Min 95 33 18 16 the introduced and indigenous populations should be
Max 101 52 21 20 prevented.
ornata Lovari & Scala (1980) compared the putative taxa
Mean 141.00 — 29.00 24.00 of Rupicapra by both principal component and dis-
N 1 — 1 1 criminant analyses, concentrating on comparing the
Females taxa R. pyrenaica and R. ornata with each other and
cartusiana with other taxa, using the measurements for all but
Mean 207.80 93.60 24.20 21.20
N 5 5 5 5
the nominate R. r. rupicapra in the tables in Couturi-
-1— Std dev 9.338 23.104 .837 1.643 er’s (1938) book. Even using simple principal compo-
0— Min 197 59 23 19 nents, R. pyrenaica and R. ornata were strongly differ-
+1— Max 218 117 25 23
349-47558_ch01_1P.indd 248 5/31/11 7:05 PM

b o v idae 249
entiated from other taxa, but differed only on average posture, and distinctive courtship patterns. It is not
from each other. Using discriminant analysis, how- adapted to the cold environments of R. rupicapra (Ma-
ever, R. pyrenaica, R. ornata, R. carpatica, R. balcanica, sini & Lovari, 1988).
and a R. caucasica/cartusiana pairing were all distinct Pyrenees.
at the one-standard-deviation level on the skulls, but
not on the horns, where only R. pyrenaica was dis- Rupicapra ornata Neumann, 1899
tinct (there was no sample for the horns of R. ornata). Using simple principal components, Scala & Lovari
Lovari & Scala (1984), using museum samples, some (1984) were able to distinguish samples of R. ornata
of the measurements in Couturier (1938), and fresh and R. pyrenaica fully (despite their evident close rela-
hunter-killed samples, compared horn measurements tionship) on the basis of seven skull measurements,
of R. asiatica and found them indistinguishable from but the two overlapped on the basis of seven horn
other Rupicapra taxa. measurements. The authors found no overlap in the
Information on the described taxon R. r. tatrica is length of the frontomaxillary fontanelle: 16–21.6 mm
difficult to obtain, although a few comparisons have in R. ornata, but 3–10 mm in R. pyrenaica.
been made by Hrabe & Koubek (1984), who found Abruzzi.
that there was considerable sexual dimorphism, and
that, in general, the Tatra form is larger than that Rupicapra parva Cabrera, 1911
from the Alps, including an introduced population of R. parva is, as its name suggests, much smaller than
Alpine chamois in the Jeseniky Mountains of the the other two SW Mediterranean species. Despite our
Czech Republic; on the other hand, in R. r. tatrica the limited dataset, R. parva does seem likely to be dis-
nasals are narrower in both sexes, and they are also tinct, especially given its striking separation from R.
shorter in the males. pyrenaica.
For our analyses, we, like Lovari & Scala (1980), Cantabrian Mountains. The very existence of
used the datasets of Couturier (1938). chamois in this region seems to be poorly known, and
Sexual dimorphism is considerable in all taxa (Tur- more attention should be paid to R. parva for conserva-
key and the Caucasus may not be as dimorphic, but the tion purposes, among others.
sample size is very small). Although there is only one
Carpathian individual of each sex, each is very large Rupicapra rupicapra Linnaeus, 1758
(the male is the largest in the entire series). Pyrenees
1938 Rupicapra rupicapra cartusiana Couturier.
chamois are, sex for sex, smaller than any other except
the Abruzzi ones, which average slightly larger. The Chartreuse chamois (R. r. cartusiana) overlaps in
The differences between Pyrenees and Abruzzi all described character states with chamois from the
chamois and the others are very clear on the charac- Alps (R. rupicapra), even if they are somewhat different
ters of their horns. The single specimen from Santander on average. We cannot comment on the distinctiveness
is much larger than those from the Pyrenees and of the taxon Rupicapra rupicapra tatrica Blahout, 1972,
Abruzzi. Those from Turkey appear rather small, fol- which may or may not be different from the Alpine
lowed by the Caucasus specimens. The difference be- chamois (the available measurements, however, are
tween Pyrenees chamois and the rest in the relative very different from those of the Carpathian species,
distance between the horn tips is dramatic; the R. carpatica).
Santander specimen has very long horns, with the tips Alps and neighboring ranges.
very close together. On our discriminant analysis us-
ing just horns, the divisions are Pyrenees plus Abru- Rupicapra carpatica Couturier, 1938
zzi, Carpathians, and the rest. This large, very dark chamois differs absolutely from
We propose the arrangement given below, much other taxa of the genus. Again, despite the relative pau-
of which is provisional, although some defi nite con- city of data, we have no doubt that R. carpatica is a
clusions are possible. thoroughly distinct species.
Carpathians.
Rupicapra pyrenaica Bonaparte, 1845
Apart from its distinctive pelage, R. pyrenaica differs ?Rupicapra balcanica Bolkay, 1925
from R. rupicapra in its smaller size, the consistent ab- Evidence as to the distinctiveness of this taxon is
sence of an ethmoid vacuity, the greater angle of the lacking, and more specimens need to be studied. —-1
horn cores to the frontals, the presence of a conflict Balkan Mountains. —0
349-47558_ch01_1P.indd 249 5/31/11 7:05 PM

250 Artiodactyla
serows); long frontal processes running lateral to the

Rupicapra asiatica Lydekker, 1908 nasals, reaching forward to the maxillae, but some-
times separated by a long lacrimal tongue. Infraor-
Rupicapra tragus caucasica Lydekker, 1910, almost certainly is
bital foramen just behind the level of P2, and halfway
a synonym, but the evidence in both cases is poor.
up the height of the rostrum. Supraorbital foramina
On geographic grounds, it does seem probable that R. generally single, or with a tiny accessory opening (but
asiatica, the chamois of Turkey and the Caucasus, is commonly double, the two openings equal in size, in
distinct from others, but more evidence is needed. Chinese goral). Palatine foramina behind the level of
the second molars. Temporal fossa short and high. Dis-
Nemorhaedus Hamilton Smith, 1827 tance from the base of the horn cores to the margin of
goral the orbit equal to or greater than the diameter of the
orbit, forming short, stout, tapering pedicels. Rostrum
The spelling of the name Nemorhaedus is questionable. Corbet
more or less parallel-sided; tips of the premaxillae
& Hill (1992:270) wrote: “Naemorhedus Smith, 1827. . . .
broad.
Subsequent variation of spelling such as the etymologi-
Horns broad at the base, sometimes hardly ringed
cally correct Nemorhaedus are unjustified emendations;
at all, but the rings more prominent in Chinese goral;
there is no evidence of an error in the original publica-
individual and, especially, age variations makes this
tion.” Pending a submission to the ICZN (in view of the
latter difference rather hard to appreciate.
use of several different spellings in the literature), we will
Preorbital gland very small, not invaginated. Foot
retain the familiar spelling, but it should be noted that it is
gland small, but with a long, back-bent duct.
not technically beyond reproach.
A discriminant analysis of all goral samples, us-
2n = 55 for N. baileyi (Soma & Kada, 1986; as “Nemorhae- ing just four cranial variables (minimized, in order
dus goral cranbrooki”), which, of course, also implies the to be able to include a maximum number of crania),
existence of 2n = 54 and 56, there being no sex chromo- fi nds that there is overlap between all taxa, but the
some / autosome translocation; 2n = 55 for N. griseus; results are sufficient give an outline of affi nities
and 2n = 56 for N. goral. within the genus, with a general differentiation be-
Nemorhaedus and Capricornis, with which it has tween Himalayan, Russian, Chinese, Burmese, and
been universally linked— even, at times, being placed red goral.
in the same genus (Groves & Grubb, 1990)—were as- We then separated the Chinese skulls into three
sociated by A. W. Gentry (1992) with Ovibos, with groups: W mountains, for which the name Nemorhae-
which they share some unusual features among the dus griseus is available; E and C China (prior available
Caprini: notably, the cranial roof is not inclined, and name, Kemas arnouxianus); and Yunnan, bordering on
the ethmoid fissure is lost. the region of Burma, whence Nemorhaedus evansi was
Skull generally more apomorphic than any of described. Most of the Yunnan skulls are from
the serows. Lacrimal fossa small or absent. Lacrimal Tengyueh (= Tengchong, 25° 01' N, 98° 28' E, 1580 m);
bone narrowed dorsoventrally. Facial angle steeper two are from Huiyao (not found); and one is from
than in serows. Zygomatic arch slanting upward Likiang (= Dayan, 26° 48' N, 100° 16' E, 2394 m). Be-
along the anteroinferior orbital margin and closely cause Likiang is so much farther NE (at least from
following the lacrimal/malar suture, ending at a mas- Tengyueh), we decided to keep this skull separate
seteric knob on the maxilla. Nuchal surface lower from the others. N. griseus and so-called Kemas arnoux-
and more horizontal (corresponding to the greater ianus are larger than the Yunnan goral and N. evansi.
cranial kyphosis) than in serows; frontoparietal angle The Likiang skull is large, like N. griseus. Taking this
usually 120°–130°, parieto-occipital angle 135°–150°. question even further with a discriminant analysis,
Nasal process of the premaxilla narrows posterosupe- we found no evidence, at least as far as the cranium is
riorly along the maxillary margin. Nasals relatively concerned, for differentiation between gorals from W
long, separated from the maxilla by a hiatus (except and from E-C China, but N. evansi is distinct; there
in red goral); smaller splint bones (interstitials) some- seems to be no justification for separating the Yunnan
times present alongside the nasals, but not fi lling in goral from N. evansi, except that the single skull from
the lacuna. No marked lateral nasal processes. Naso- Likiang is more like N. griseus.
frontal suture very narrow, arch-shaped backward, The taxonomic differentiation is obviously very
-1— sometimes with a small median wedge on the fron- marked, but given the (slight) morphometric overlap,
0— tals (except in red goral, where it resembles that of the evidence of the external characters is crucial in
349-47558_ch01_1P.indd 250 5/31/11 7:05 PM

b o v idae 251
determining whether the various gorals rank as dis- Bagaili, E Kumaun; Sikkim; Apoon and Deora,
tinct species. Gorkha, Nepal; Katmandu; near Masuri, Garwhal,
In discriminant analysis, the Chinese and Russian Uttar Pradesh; Naini Tal; Molta; Shadra.
Far East taxa are totally separate, and thus should be Hodgson’s and Hardwicke’s Antilope goral, from
classified as distinct species: N. griseus and N. caudatus, “Himalayal range and mts. of Nepaul frontier” was
respectively. The gorals of Shanxi, sometimes sepa- “grey mouse-color (but almost white about the lower
rated as a taxon vidianus, fall into the range of N. griseus. part of the neck and throat), and darker . . . inclining
Thus, this brings the number of recognizable goral taxa to ferrugineous about the legs.” The type was in the
to three: N. griseus, N. caudatus, and N. evansi. Barrackpore menagerie; it had been sent from Kath-
The Himalayan (N. goral, N. bedfordi) and red (N. mandu, and, given this information, Lydekker (1906)
baileyi) gorals are of rather small size, below that of used the name Naemorhedus goral for the E (brown)
(most specimens of) Chinese taxa, but averaging larger form and named the W (gray) form N. bedfordi. Pocock
than N. evansi. The horns of the Himalayan goral(s), (1913) argued that, on the basis of the description,
however, are as long as those of the Chinese goral (N. Lydekker had used Hardwicke’s name wrongly, so Po-
griseus); only N. evansi and N. baileyi are short-horned. cock renamed the E (brown) form N. hodgsoni. Consid-
As far as the lacrimal index is concerned, N. griseus ering that the type locality of N. goral is Kathmandu,
and N. evansi are alike in having relatively deep, short where only brown goral occur, and Hardwicke’s de-
lacrimals. This offers some support to the separation scription of the color could just about fit either brown
of a Himalayan taxon and of red goral as separate en- or gray gorals (though not the extremes), we think
tities; the number of recognizable taxa of goral there- that Lydekker’s interpretation is more likely.
fore now rises to five; whether or not there are two Ward (1925) placed the boundary between the
distinct Himalayan taxa will depend on pelage analy- gray and the brown gorals at roughly the Sardah
sis (below). River between Uttar Pradesh and Nepal; this seems
We fi nd that six taxa show consistent phenotypic too far E, as specimens from Kumaun, Uttar Pradesh,
differences: the two Himalayan gorals, the Chinese are all of the present (brown) goral species. The Sutlej
goral, the Burmese goral, the long-tailed goral, and River seems a more likely boundary.
the red goral. Hence, we recognize six species. Using
craniometry, they cluster into four groups: (1) the two Nemorhaedus bedfordi (Lydekker, 1906)
Himalayan species, N. goral and N. bedfordi; (2) the Chi-
1906 Urotragus bedfordi Lydekker. Dharmsala.
nese N. griseus and the Burmese N. evansi; (3) N. cauda-
tus; and (4) the single skull representing N. baileyi. Skins seen: 13.
Gray to gray-brown to yellow-gray. Legs lighter,
Nemorhaedus goral (Hardwicke, 1825) yellower, with a dark brown line down the front, fad-
Himalayan brown goral ing on the pasterns. Underside off-white. Chin creamy,
this tone reaching the whole way down the throat.
1825 Antilope goral Hardwicke. Kathmandu.
Merest trace (if that) of a dorsal stripe. Forehead some-
1908 Naemorhedus hodgsoni Pocock. Sikkim.
times slightly darker than the head.
Skins seen: 21; as well as observations of about 10 liv- Chamba; Kulu, 7500 ft.; Jagatagukh (or Jagathsukh),
ing specimens in the Delhi Zoo. Kulu; Kishtwar; Manali, 4 mi. S of the Kulu Valley;
Color either medium brown, with black tips to the Kathai Nullah, Kaj-I-Nag, 7000 ft., Kashmir; Palum-
hairs, giving a rabbity effect; or slightly grayer, to pire, Jhelum Valley, Kashmir; N bank of Jhelum River,
gray-brown; or pale or dark fawn. Legs browner, to W of Srinagar; Dharmsala; Wardwan, Kashmir; Lo-
very bright tan, or white on the forelegs only. Under- huMang cliffs, Wardwan Valley, Kashmir; Sind Val-
side paler gray. Throat and chin variably white, some- ley, Kashmir.
times interrupted under the jaw; lips white. Dorsal Pocock (1910a) said that both gray and brown go-
stripe usually weak, sometimes fading behind the ral occur in Nepal, but there seems to be no evidence
withers. Tail short, the distal half black. Short-legged; for this; likewise, Pocock said they both occur at
short-eared. Chamba, but all of the eight skins sent to the British
No nasal interstitials in 25 skulls; no trace of a lac- Museum by Rodon—whom Pocock cited in evidence
rimal fossa. for this statement—are gray. Tom Roberts (pers.
Localities of specimens seen by us: Wangdou comm. to PG) saw goral in March in C Nepal and —-1
Potrang, 4500 ft., Bhutan; Ramchu, 11,000 ft., Nepal; thought they looked exactly similar in coat color to the —0
349-47558_ch01_1P.indd 251 5/31/11 7:05 PM

252 Artiodactyla
goral in Pakistan, with which he was familiar from juveniles); nose zone clearer. Hair fairly long, coarse
his 34 years of experience. He found that the latter in all seasons.
tended to look gray in their summer coat and brown Localities of specimens seen by us: Ichang; Sich-
in winter, with the males grayer than the females. uan, 6000–7000 ft.; Mupin (= Baoxing), Sichuan, 30°
The evidence of the museum skins, which are of both 29' N, 102° 49' E; O-er, near Wei Chow; Wen-chwan
sexes and from different seasons, does, however, indi- Hsien, Si-ho, Sichuan; Wen Chuan, Sichuan, 31° 20' N,
cate a consistent difference between W (mainly Kash- 103° 31' E, and vicinity (Tsao Po, 15 mi. to the SW;
mir) and E Himalayan goral. As noted above (in the Chengou Forks and Chen Lliang, both 30 mi. to the
morphometrics part of the introductory section to W); Loung-Ng’ou-fou, Sichuan; Dun Shih Goh, Sich-
Nemorhaedus), there are no evident skull or horn dif- uan, 30° 25' N, 102° 50' E; Gan Yang Go, Sichuan, 30° 20'
ferences, and their taxonomic separation depends on N, 102° 30' E; Kuan-Hsien; Ta-tsien-pou (= Tatsienlu, Si-
pelage characters alone, which, however, is consis- chuan, 30° 03' N, 102° 13' E); “Korea”; “E Tibet”; Yenping,
tent in our material. Fujian; Patungxian, Yangtze Gorges, Hubei; Wa Shan,
Liujiang, and Batang, Sichuan; Beijing; 30 mi. S of
Nemorhaedus griseus Milne-Edwards, 1871 Fengxiangfu, Shaanxi, 10,000 ft.; Mt. Hoaysan, 3943 m,
Chinese goral Shaanxi; Paochi, Shanxi; Kwei hua cheng, Shanxi;
Tung-ling, Hebei; Mai Tai Chao, Shanxi; S of Fengsian-
1871 Nemorhedus griseus Milne-Edwards. Moupin, Sichuan.
gfu, Shaanxi, 34° 24' N, 107° 29' E; Likiang, 27° 04' N,
1874 Antilope (Nemorhedus) cinerea Milne-Edwards. Moupin,
100° 12' E; Yenping, Fujian, 26° 38' N, 118° 10' E; Tai
Sichuan.
Pei Shan district, Shaanxi, 33° 56' N, 107° 41' E; Liu-
1888 Kemas arnouxianus Heude. Jixien, Zhejiang.
Tsuen, Shaanxi, 34° 02' N, 108° 55' E; Mun San Tsai;
1890 Kemas henryanus Heude. Ichang, Hubei.
Kuei-hua-cheng; Chian Shien; Lianghokou; Hokow,
1894 Kemas aldridgeanus Heude. Yitchang, Hubei.
30° 01' N, 101° 08' E.
1894 Kemas fantozatianus Heude. Mountains of Kiun-
Other records: Taukwan, Sichuan ( Jacobi, 1923);
tscheou, right bank of the Middle Han River, Hubei.
Gego, 55 mi. SEE of Tatsienlu, 6000 ft. (Orr, 1938);
1894 Kemas iodinus Heude. E Sichuan.
near Ichang, Patung Hsien, and the Yangtze Gorges,
1894 Kemas niger, Kemas fargesianus, Kemas initialis, and
Hubei; Wa Shan, Liuyang, and Batang, Sichuan (G.
Kemas versicolor Heude. Chenkouting, Sichuan.
M. Allen, 1940).
1894 Kemas vidianus and Kemas galeanus Heude. Yu Ho
The foot gland was bigger in a goral from Shaanxi
Mountains, S Shaanxi.
than in one from the Himalayas (Pocock, 1918).
1894 Kemas xanthodeiros and Kemas pinchonianus Heude.
Interstitials occur in about one-third to one-half of
W Sichuan.
the skulls (Ichang 0/6, Sichuan 7/18, Shaanxi 0/1);
Skins seen: 27; as well as observations of about 20 even without interstitial bones, the narial notch is not
living specimens in Chinese zoos. usually as deep as in either species of Himalayan go-
Dark to light gray, with varying amounts of brown ral (N. goral and N. bedfordi). The skull, on average, is
and a varying black overlay (more so in the adults deeper than in the two Himalayan goral species, and
than in the juveniles, and more in the males than in the dorsal outline rises above the orbits, rather than
the females); range of colors actually much the same being at about the same level. Usually there is at least
as in N. goral. Front and inner side of the shanks very a trace of a lacrimal fossa (taking the form of a fi ne
pale, whitish ( juvenile) or yellowish (adult), from line of minute pits in the bone); the supraorbital fo-
slightly above the knees or slightly below the hocks, ramina is commonly double, the two openings equal
often with a dark line down the front; thigh color in size. The horn rings tend to be more prominent
sending a wedge down the outer side to halfway down than in the Himalayan gorals; the horn tips are hardly
the shank; underside light gray (but the groin and the downturned.
axillae, as usual, white); throat white, with a golden Out of the 14 species described by Heude and the
tone, not extending much to the interramal, but often two by Milne-Edwards, Sowerby (1917) retained five:
visible on the lateral surface of the neck; chin dark; Nemorhedus griseus (Moupin [= Baoxing], Shaanxi, Hu-
dorsal stripe generally clear, thick, dark gray-brown bei); Kemas niger (Che-ko-ting), with smaller horns; N.
or black, but not always prominent; tail very black, cinereus (also from Moupin), with a heavier skull; K.
short, but bushy for most of its length; forehead brown henryanus (Ichang), with a wider skull; and K. arnoux-
-1— or black-brown in the adults ( just a brown line in the ianus (Zhejiang), isolated from the others.
0—
349-47558_ch01_1P.indd 252 5/31/11 7:05 PM

b o v idae 253
G. M. Allen (1930) recognized the gorals of Shanxi Thus, there is no reason to think that Shanxi and Bei-
(Kweihwacheng—actually in Inner Mongolia—and jing goral are even subspecifically different from N. gri-
Maitaichao, 45 mi. E of Paotowchen), and also of seus, let alone intermediate between the latter and N.
Tungling (in Hebei), as a distinct form, to which unfor- caudatus.
tunately he applied the name Nemorhedus caudatus. He Nor is there any reason to separate goral from E
said that N. griseus has a shorter coat than N. caudatus, China from those from Sichuan. In the discriminant
slightly darker, but differing on average only; and that analysis using Sichuan (Nemorhaedus griseus), coastal
the throat has a whitish patch, with a narrow, ochery (Kemas arnouxianus), and Yunnan and Burmese (N.
border like N. caudatus, extending nearly to the lips. evansi) samples, the K. arnouxianus skulls fell within
He likewise separated Kemas henryanus (Ichang, Hu- the dispersion of N. griseus; on the other hand, Yunnan
bei, Fujian) from Nemorhedus griseus (Sichuan, Yun- skulls (except for Likiang) are separate, and the small
nan), as having a darker neck and a smaller, but bright Burmese sample fell at the edge of the Yunnan one.
orange throat-patch.
Later, G. M. Allen (1940) substituted the earlier Nemorhaedus evansi (Lydekker, 1906)
name Kemas arnouxianus for K. henryanus, and admit- Burmese goral
ted that Nemorhedus griseus is “very little different”
1906 Urotragus evansi Lydekker. Mt. Victoria, Arakan.
from what he called N. caudatus. He distinguished the
two S races by saying that in N. griseus, the throat- Skins seen: 24.
patch is whitish, with a narrow, pale ochraceous bor- Very light, fawn or somewhat brown; legs in the
der, extending nearly to the lips; whereas in Kemas ar- Burmese and the Thai specimens mostly golden to
nouxianus, the throat is uniformly pale orange, except creamy gold, with a darker wedge (as in the Sichuan
in a skin from Fujian. specimens), or browner, with a thick black line down
We think that Allen did not take individual vari- the front; underside light gray-fawn; throat white, with
ability into account; in our samples, the full range of a variable golden tone, extending nearly to the chin, the
throat-patch color, from a small yellow to a large golden tone fading forward. Dorsal stripe usually fairly
white patch, is seen in Ichang skins. prominent, medium dark brown; tail mostly dark
The thicker pelage in Shanxi goral (Allen’s N. cauda- brown. Forehead clear medium brown, extending to
tus) is also variable. It is worth noting that Allen had no the midline of the nose. Hair short in available speci-
specimens of the real N. caudatus available to him; he mens, but there may be seasonal variation.
simply assumed that goral from Shanxi are the same as Much smaller than N. griseus.
those from the Russian Far East. If Shaanxi goral really Localities of specimens seen by us: Byingyi, Kyin-
are distinct, the name for them would be N. vidianus mana (Ryinmama?) district, Upper Burma; Dawna
Heude, but there is no available name for those from Range, Amherst; Doi Angka, 56 km SW of Chiengmai,
Shanxi or from the neighborhood of Beijing. Thailand; Raheng, 16° 50' N, 99° 05' E; Pakokku dis-
Seventeen Sichuan skulls, with teeth in full wear, trict, Upper Burma (W of Mt. Victoria, 21° 17' N, 93°
ranged from 214 to 225 mm; seven younger skulls, 53' E); Arakan; Thaton district, 16° 50' N, 97° 18' E;
with wear not yet completed on the last molars, Xiaojelu and Tengyueh (25° 00' N, 98° 30' E), Yunnan;
from 205 to 221 mm. An adult skull from Shanxi was Hui-yao; Hsiso-ke-la, Yunnan.
209 mm, outside the Sichuan dimensions. This might Other localities: Kloss (1923) recorded goral in
seem suggestive, but two Shanxi skulls with incom- Thailand at the Me Ping rapids, 17° 30' N.
plete tooth wear were 206 and 212 mm— and a Brit- Lydekker (1906) distinguished N. evansi from the
ish Museum skull from Paochi, Shanxi, measured gray Himalayan goral (N. bedfordi) by its more brown-
226 mm. ish gray color, with no white on the cheeks; no dark
As described above (in the morphometrics part of mark on the upper surface of the muzzle; and the legs
the introductory section for the genus), we wished to brown behind, rufous in front. The general color and
test whether the Shanxi and Beijing samples are the leg color are valid differences. Pocock (1913) syn-
distinct—perhaps intermediate between N. griseus onymized N. evansi with N. griseus, from which, how-
and the true N. caudatus—so we performed discrimi- ever, the former differs by its (apparently consistent)
nant analyses on skull measurements. The skulls of N. extremely short hair; average lighter color, and the less
caudatus fell outside the range of N. griseus, whereas contrasting color of the tail and the face; and its much
the Shanxi/Beijing skulls fall well within the latter. smaller size. —-1
—0
349-47558_ch01_1P.indd 253 5/31/11 7:05 PM

254 Artiodactyla
renaming of the animal which Radde (1862) described

Nemorhaedus caudatus Milne-Edwards, 1867 from Amurland and mistakenly called Antilope crispa
long-tailed goral Temminck.
In our small sample, the Ussuri goral are paler
1862 Antilope (Caprina) crispa Radde; not Temminck, 1845.
than those from Beijing, with a long-haired gray tail,
Lagar River, a tributary of the Amur River.
and the shanks are dark down the front, contrasting
1867 Antilope caudata Milne-Edwards. Renaming of Radde’s
with the white pasterns. In the Beijing goral, includ-
Antilope crispa.
ing the ones described by Pocock (1908a) and Milne-
1894 Kemas raddeanus Heude. Amur River.
Edwards (1867), the pelage is somewhat darker and the
Skins seen: 8. tail hair shorter and black, while the shanks and feet
Gray-fawn to brown, but paler than N. griseus, and are all the same light color in front—in all these fea-
with no black overlay. Legs creamy to yellow-gray, tures resembling N. griseus, to which we assign them.
with less of a sharp transition to body color than in N. For Adlerberg (1932), the Ussuri goral was a dis-
griseus; dark line down the front of the foreleg. Un- tinct species, distinguished by its skull characters and
derside pale fawn or gray; groin and axillae white; by its paler color, whiter throat-patch, very long tail
chest darker; throat broadly white, extending forward hairs (which are mixed with white), thick underfur,
to the chin, with no golden tone or edging; lips white, and long fur, even in summer. Heptner et al. (1961)
this color extending to the mouth angles; forehead and said that in the Russian Far East, three color types oc-
nose usually not very dark; whole head, especially the cur: gray, foxy, and a very rare white form.
muzzle, more brownish; often a gazelle-like stripe. Dor- Volf (1976) recorded that crossbreeds between N.
sal stripe usually inconspicuous. Tail with a very long griseus and N. caudatus are sterile, so he supported the
brush, usually black nearly to the base, reaching to the view that they are separate species. A pair (identified
hocks or below, the length of this hair giving N. caudatus as Kemas raddeanus) sent from the Berlin Tierpark had
its long-tailed aspect, the tail itself being actually some- longer horns than a pair identified as N. griseus from
what shorter than in other species (see table 68). Fur Yunnan (although, from the description, it seems more
longer, softer than in others; underwool ashy brown, likely that these were N. evansi); the Tierpark pair were
differing from the long, pale, buff y, dark-tipped con- gray, rather than having a short brown coat, and they
tour hairs. Ears short (see table 68). had a white tail 20 cm long, as opposed to the short,
Localities of specimens seen by us: Ussuri; I-mien- dark tail of the Yunnan pair. Although there might be
po, Heilongjiang, 45° 03' N, 128° 04' E; Kentei Moun- other reasons for the hybrid sterility, this is still
tains, Heilongjiang. suggestive.
Also from Syrun-Bulyk Ridge, SE Mongolia, fide
Adlerberg (1932). Nemorhaedus baileyi Pocock, 1914
For this goral the name Kemas raddeanus has often red goral
been used, under the misapprehension, which may de-
1914 Nemorhaedus baileyi Pocock. Dre, Yigrong Tso, Po Me,
rive from Heude (via Sowerby), that the type locality of
9000 ft.
N. caudatus was Beijing. This, in turn, has led to specula-
1961 Nemorhaedus cranbrooki Hayman. 8000 ft., in the Adung
tion on whether the gorals from near Beijing and Shanxi,
Valley, Upper Burma.
and from Shaanxi (Heude’s N. vidianus), might be inter-
mediate between the Amur form and N. griseus. Thus Our description is based on the type of N. baileyi; the
Sowerby (1917) said K. raddeanus from the Amur region syntypes of N. cranbrooki; a mounted skin in the Kun-
is very large, lighter and grayer than his NE China spec- ming Institute of Zoology; and about a dozen living
imen, which he called N. caudatus, and the long tail is specimens in the San Diego and Shanghai zoos.
white (black in N. caudatus); Howell (1930) said K. radd- Warm brown to foxy red, with no black ticking;
eanus has a white tail-tip, more black on the forelegs, flanks paler; legs like the body; a mere trace of a dark
and a heavier pelage; and G. M. Allen (1930) described line on the knees; underside paler; axillae and groin, or
N. caudatus as being slightly paler than other Chinese the groin only, white; throat and interramal region a
races, with longer underwool in winter, and the gray of paler, yellower version of the body color, sometimes
the back extending to base of the tail (his sample was a small white or yellowish patch on the throat. Nose
from Tungling [E Tombs], Hebei, and Shanxi). slightly darker than the rest of the head. Dorsal stripe
-1— The type locality of N. caudatus, however, is the usually clear, but thin, brown. Tail short, the tuft occu-
0— Bureja Mountains, Amurland, and it is quite clearly a pying half the length; legs and ears short (see table 68).
349-47558_ch01_1P.indd 254 5/31/11 7:05 PM

b o v idae 255
Localities of specimens seen by us: Dre, Yigrong directly to the outside (said to be more goral-like in
Tso, Po Me (30.25° N, 95° E); Gongshan; Adung Val- Japa nese serow [Pocock, 1919]). Lacrimal fossa pres-
ley, 8000 ft., Burma (28.10° N, 97.40° E). ent. Lacrimal bone shorter and higher than in go-
Its distribution extends to Lashio, and perhaps even rals. Forehead swollen and convex. Zygomatic arch
to Loikaw (19.30° N), according to Hla Aung (1967). continuing below the orbit and onto the face as a
In the single available complete skull, the nasals are sharp crest. Superior nuchal crest following a steep
not bordered laterally by a lacuna. Instead, they are course down the side of the occiput; inferior edges of
posteriorly expanded, with a serow-like nasofrontal su- the occipital condyles above the level of the maxillary
ture, and form a suture with the maxilla as far forward toothrow. Frontoparietal angle 141°–155°; parieto-
as the level of the P3/4 margin; lateral to them are long occipital angle 120°–140°. Nasal process of the pre-
splint bones, which fill in the normal nasomaxillary maxilla broad, ending bluntly, in front of the level of
lacuna. A lacrimal fossa is present, though shallow, the anterior premolar. Nasals with short lateral pro-
with a trace of a sharp upper rim. cesses, and with the usual posterior expansion; nar-
An incomplete skull in the Bombay Natural His- ial notch in front of the level of P2; nasal forming a
tory Society, described by Hayman, could not be found suture with the maxilla at this notch, but farther
during CPG’s visit. back the two bones merely lying adjacent, with no
The type of N. baileyi has a small but clear white suture; lacuna commonly occurring at the level of
patch on the throat; that of N. cranbrooki does not. A the lacrimomaxillary suture. Infraorbital foramen
specimen in San Diego Zoo, from N Burma, had an level with P2 and low down, on the horizontal plane
upper throat that was lighter than the surrounding of the palate. Supraorbital foramina always dividing
coat, but not clearly white; the mounted specimen into at least two (sometimes as many as five). Pala-
in Kunming, from Gongshan, has a small but no- tine foramina opposite M 2 or M1. Temporal fossa
ticeable whitish yellow patch. There is only a trace long and low. External auditory meatus a vertically
of a dorsal stripe in the N. baileyi skin; this is much elongated oval. Nasal length relatively short; Tai-
clearer in the type and paratypes of N. cranbrooki and wanese serow have especially short, broad nasal
in the San Diego Zoo animal, and strikingly clearly bones. Distance from the base of the horn cores to
marked in the Kunming specimen from Gongshan. the margin of the orbit less than the diameter of the
One of the three paratypes of N. cranbrooki is orbit. Rostrum swollen in the anterior maxillary re-
darker and browner than the others, and has a more gion, and tapering toward the narrow free ends of
strongly marked dorsal stripe. the premaxillae.
Considering the variability in the zoo specimens, Horns narrowing gradually and evenly along
and the improbability of there being two red gorals their length, closely ribbed with flat rings to more
living so close to each other in the Tibet/Yunnan/ than halfway along.
Burma borderlands, we think it most likely that the We divide the species into two informal groups.
type of N. baileyi and the syntypes of N. cranbrooki
represent one and the same taxon, in its summer and Capricornis sumatraensis Group
winter coats, respectively. U Tun Yin (1967) cited Fran- Preorbital gland very deep. Lacrimal bone relatively
cis Kingdom-Ward’s report of two “foxy red” gorals long, low. Skull elongated, prognathous. Skull
in the Tsangpo Gorge, bringing goral of the N. cran- pinched-in under the orbits. Parietals flattened. Skull
brooki description to the vicinity of the N. baileyi type breadth less than 42% of the length. Cheekteeth
locality. U Tun Yin also mentioned seeing them at 30%–30.6% of the skull length. Horns tend to go back
14,000 ft. on the Diphuk La Pass, between the Adung in the plane of the forehead or up at somewhat of an
Valley and Rima in Tibet. Bailey, the collector of angle. Color black or red, with white facial markings
the type specimen, said that people in the valley of the and at least some development of a mane.
Zayul Chu, a Tibetan tributary of the Tsangpo, wore
bright foxy red goral-skin coats. Capricornis sumatraensis (Bechstein, 1799)
Sumatran serow
Capricornis Ogilby, 1837
1799 Antilope sumatraensis Bechstein. Sumatra.
serow
1814 Antilope interscapularis Lichtenstein. Sumatra.
Preorbital gland very deep, invaginated (in the Hi-
1900 Nemorhaedus swettenhami Butler. Larut Hills, Perak.
malayan serow; less deep in the Japa nese serow). —-1
1908 Capricornis sumatraensis robinsoni Pocock. Selangor.
Foot gland bigger than in gorals, but communicating —0
349-47558_ch01_1P.indd 255 5/31/11 7:05 PM

256 Artiodactyla
2n = 46 (Sumatran specimens) 1908 Capricornis sumatraensis jamrachi Pocock. Kalimpong,

Skins seen: 21. near Darjeeling.
Jet-black; no white bases to the hairs, except on 1908 Capricornis sumatraensis rodoni Pocock. Chamba,
the mane and along the midback on either side of the Punjab.
(black) dorsal stripe; mane varies from mostly white,
Skins seen: 20, including the types of Antilope thar,
or golden buff y white, to black, with only a few
Capricornis sumatraensis humei, C. s. jamrachi, C. s.
white hairs. Underside not white; short, white jaw
rodoni.
streaks quickly becoming red and petering out back-
Jet-black, with straw or white hair bases that show
ward. Legs black, but with distinct (varying) reddish
through; black tips of the hairs becoming redder
tones toward the hoofs; sometimes with white hairs.
lower down the flanks, but sometimes acquiring a
Nose not white, reddish above the lips, from the nos-
buff y tone as the black tips wear off. Long mane
trils to the jaw angles; face slightly redder; lips
mixed black and white, varying, but never with the
creamy white. Long-legged; ears relatively short (see
white predominating. Sharply cream-buff below; in-
table 68).
ner surfaces of the thighs creamy, continuous with
Horns go back in the plane of the face; forehead
the light underside. Broadly white over the nose, or
much more swollen and convex than Indian or Chi-
only on the lip margins; tan patch behind the white.
nese serows. Nasofrontal suture tending to go straight
Face sometimes lighter than the body. White extend-
across.
ing along the jaw lines, backward in a V, or the inter-
Localities of specimens seen by us: Biserat, Batu,
ramal region completely white, or occasionally nearly
Selangor, S Opis Setul (2° 48' N, 101° 55' E) and Sungei
absent. Black tips on the legs becoming red lower
Letue in the Malay Peninsula; “W Siamese Malay
down, and the pure white hairs increasingly intermix,
States”; Kerinci, at 5600 ft., Padang Besi, Aceh, and
with the legs thus becoming creamy white shortly be-
Solon in Sumatra; W Sumatra.
low the knees and the hocks, sometimes with a buff y
We include the Malay Peninsula serow in this spe-
line down the front. Dorsal stripe sometimes visible,
cies, as the skins we have seen are indistinguishable
but not usually markedly so.
from other C. sumatraensis. Note, however, that the
Horns going up at an angle; forehead flatter than
skull of the type of C. s. robinsoni, the only Malaysian
in Sumatran (C. sumatraensis) or Burmese (C. rubidus)
skull we have been able to fi nd, does not fall into the
serow.
range of variation for Sumatran skulls, but instead is
Localities of specimens seen by us: Kashmir (Dachi-
closer to C. maritimus. This may indicate gene flow
gam, Pir Panjal; Warapash [Sindh Valley]; Chasma
between the two species, although the skins show no
Shahi, 8000 ft.); Rampur, Sutlej Valley; W Garwhal;
intermediacy.
Kulu; Katnauli, Kumaun, 6500 ft.; Mussoorie; Cheena
Butler (1900) distinguished Nemorhaedus swetten-
Peak near Naini Tal, 20° 22' N, 79° 26' E; Nepal; Yo-
hami from Capricornis sumatraensis by its jet-black legs.
davru, Nepal; Kursiong, Darjeeling; Kalimpong, near
Pocock (1910b) said C. s. robinsoni apparently dif-
Darjeeling; Sikkim; Tong Chu Valley, Bhutan; Lushai
fers from C. s. sumatraensis in the mane of the former
Hills, Assam.
being more crestlike instead of matlike, and in hav-
Choudhury (1997) gave localities in the following
ing mixed black and white hairs; but he recorded C. s.
districts for Assam, all on the N bank of the Brahma-
robinsoni from Perak, and a specimen resembling C. s.
putra River: Kokrajhar, Sonitpur, Lakhimpur, Dhem-
sumatraensis from the Siamese frontier—probably, as
aji. It “usually occurs in the higher hills, above 300 m
he suggested, all three names are synonyms. Dolan
elevation, but may come down to 100 m in winter. It
(1963), too, noted that C. s. robinsoni falls in the varia-
occurs from tropical rainforest and deciduous forest
tion of Nemorhaedus swettenhami.
in Assam to subtropical and temperate forests, both
conifers and broadleaf, in Arunachal Pradesh.”
Capricornis thar (Hodgson, 1831)
Himalayan serow
Capricornis milneedwardsi David, 1869
1831 Antilope thar Hodgson. Nepal Himalayas. white-maned serow
1832 Antilope bubalina Hodgson. Nepal.
1869 Capricornis (Antilope) milne-edwardsi David. Moupin,
1842 Nemorhaedus (Kemas) proclivus Hodgson. Alternative for
Sichuan.
Antilope thar.
-1— 1871 Nemorhedus edwardsii David.
1908 Capricornis sumatraensis humei Pocock. Kashmir.
0— 1888 Capricornis argyrochaetes Heude. Tchou-ki, Zhejiang.
349-47558_ch01_1P.indd 256 5/31/11 7:05 PM

b o v idae 257
1894 Capricornis maxillaris Heude. Chaohing, Zhejiang. 1923); Tienwan, 50 mi. SEE of Tatsienlu, 5000 ft., Sich-
1894 Capricornis platyrhinus, Capricornis cornutus, Capricornis uan (Orr, 1938); Wa Shan and Tatsien Lu, Sichuan
erythropygius, Capricornis microdontus, Capricornis (G. M. Allen, 1940).
ungulosus, Capricornis nasutus, Capricornis vidianus, Dolan (1963) described age changes, noting that
Capricornis fargesianus, Capricornis brachyrhinus, they account for some of the differences between the
Capricornis pugnax, Capricornis longicornis, and Capricor- described taxa that we have placed as synonyms. He
nis tchrysochaets Heude. Sichuan. noted that young animals can be black; adults,
1899 Capricornis collasinus Heude. Kwantung. “strong brown”; and older animals tend to have more
1921 Capricornis osborni Andrews. Hui-yao, 20 mi. from white in the mane. He found no consistent differ-
Tengueh, Yunnan. ences between Sichuan and C China, so that C. argy-
1930 Capricornis sumatraensis montinus Allen. Likiang Range, rochaetes falls in the variation of C. milneedwardsi.
Snow Mountains, Yunnan.
Capricornis maritimus Heude, 1888
Skins seen: 50, including the syntypes of Nemorhedus
Indochinese serow
edwardsii and the types of Capricornis osborni and Cap-
ricornis sumatraensis montinus. 1888 Capricornis maritimus Heude. Along bay.
Very long, lank hair. Black, tending to reddish, es- 1894 Capricornis rocherianus and Capricornis benetianus
pecially on the flanks, the rump, and the tail; much Heude. Along bay.
white showing through from the white hair bases. 1897 Capricornis marcolinus Heude. Tonkin.
Head browner. Mane varying, but largely silvery. Un- 1898 Capricornis berthetianus Heude. Tonkin.
derside light. Nose not white, but the upper lip white, 1899 Capricornis gendrelianus Heude. Tonkin.
with a broad white jaw-streak, quickly becoming 1919 Capricornis sumatraensis annectens Kloss. Koh Lak.
golden-brown backward, and ending before the go-
Skins seen: 19.
nion. Smudgy light patch on the upper throat. Legs
Black or dark brown, with long, white hair bases,
increasingly mingled with foxy red down the upper
thus appearing gray or brindled; mane mixed with
segments; shanks pure reddish; sometimes a sharp
white or pale yellow-buff hairs, the white bases often
demarcation at the knee. Dorsal stripe present.
making the mane predominately white. Underside
Skull sometimes with a sharp upper “shelf ” to the
not white, often black. Lips white, often back to the
preorbital fossa, which is lacking in other species.
level of the eye, with a white moustache mark; short,
Localities of specimens seen by us: NE Peling
white or golden-brown jaw streaks; white or golden-
Mountains (Min Shan, beyond Choni [or Jonê] and
brown hairs on the throat, sometimes forming a big
Archuen, Gansu, near the Sichuan border, ca. 34° N,
patch. Legs jet-black on the upper half, reddish tan or
103° E; see Wallace, 1913); Wen-Chwan-Hsien (= Wen-
creamy white on the lower half, with a sharp division
qanxian), Si-ho, Sichuan; Loung-Ng’ou-Fou, Sichuan;
in between; sometimes a white or red patch on the
Towquan, 60 mi. N of Kanshien, Sichuan; Maykeng,
carpus, and a black mark or mingling or line on the
near Fatung (or Tatung), 32° N, 117° E; Chengou
lower shanks. Dorsal stripe black, usually thin, but
Creek, Sichuan, 31° 20' N, 103° 00' E; Muli, Mt. Gibbok
sometimes thick. Long ears (see table 68).
(or Gibboh), Sichuan; Dan Shi Go, Sichuan, 30° 25' N,
Localities of specimens seen by us: Trang, Thai-
102° 50' E; Moupin (= Baoxing), 30° 29' N, 102° 49' E;
land; Ta Chang Tai, W Thailand, 16° 51' N, 99° 03' E;
Gan Yang Go, Sichuan, 30° 20' N, 102° 30' E; Liang-
Dawna Range (Amherst and Thaton); SW Siam (col-
hokou, 29° 07' N, 108° 45' E; Leh; Tatsienlu, 30° 03' N,
lected by Gairdner, so probably the Ratburi district);
102° 13' E; Tao River Valley, Gansu, Tibet border;
Koh Lak; Maung Pre, N Thailand; Vinh; Nhatrang;
Kegma, Hui-yao (Huaiyao) and Likiang (Lijiang), Yun-
Ninh Binh, Tonkin, 20° 14' N, 106° 00' E; Langson,
nan; May Keng (32° N, 117° E); Mahkawng Ga, 4000 ft.
Tonkin; Baie d’Along (topotype), Tonkin; Than Hoa, N
(26.49° N, 98.10° E); Mokanshan, Zhejiang, 30° 26' N,
Annam, 19° 49' N, 105° 48' E; Mozok (Mozo, Mogok?),
119° 47' E; Tunglu, Zhejiang, 29° 47' N, 119° 37' E;
upper Burma; Bernardmyo, Ruby Mines, 5000 ft.; N
Chungan Hsien, Fujian, 28° 02' N, 117° 53' E; Yenping
Shan states, 50 mi. E of Maymyo, 3500 ft.; Hpawshi, ca.
Mountains, Fujian, 26° 38' N, 118° 10' E; Fengsiang
26° 05' N, 98° 35' E; Ngawchaung, Burma, 1800 ft.;
Fu, 34° 24' N, 107° 29' E; 45 mi. S, Shaanxi.
near Moulmein (Winyaw River, 16° 19' N, 97° 58' E);
Other records: “between Ssi-Gu River and Cheicho
Gwongye; Pabya Hills, 600 ft. (76 mi. SSW); Dorngin
River, Ssi-Gu district,” Gansu (Buchner, 1891); moun-
Hill (N of Moulmein); Zwagaben Mountain, or the —-1
tains near Taukwan and Kwanhsien, Sichuan ( Jacobi,
Duke of York’s Nose (25 mi. N); Tho Toungyen Valley, —0
349-47558_ch01_1P.indd 257 5/31/11 7:05 PM

258 Artiodactyla
above Myawadi, 16° 42' N, 98° 30' E; Mt. Mouleyit; 18° 30' N, 99° 20' E, Thailand, and noted that the
Tenasserim; Pegu. shanks were rufous. Hanitsch (1918) saw a serow from
Sowerby (1917) described the Tonkin serow—for Phu’ong Mai, 13.30° N (Annam) with a rufous fore-
which he used Heude’s name C. rocherianus instead of head, shaggy white and black-and-white hairs, and legs
the earlier C. maritimus—as being smaller than the Si- that were rusty below the knee. Milner (1921) con-
chuan serow (C. milneedwardsi), with a proportionately firmed that in the Ruby Mines Hills, serow are black
deeper skull; horns that are smaller than the Chinese with red legs; in the Tharawaddy district, Pegu, he
forms; black, inclined to blue-black on the body; the recorded them as being black with red legs and no
face brown; creamy white legs; and no white or creamy white anywhere, even on the lips or the muzzle.
hairs in mane. The Maung Pre skin (in the Singapore collection)
Irwin (1914) hunted serow in Thailand, from 15° lacks white hair bases on the haunches and the shoul-
N (above Lopburi) to 11° 48' N ( just below Koh Lak ders, so it is blackish red overall ; it becomes red or
on the peninsula). He stated that there were two kinds red-brown on the thighs, gradually growing paler
of serow in the country, with or without the rufous down the leg, to become white on the pasterns.
color, and he never heard of the rufous color occur- In addition, these is a skin from the Nam Tamai
ring N (error for S?) of 12° 40' N, beyond which the Valley (27° 42' N, 97° 54' E). One might expect this to
serow only have black and whitish hairs in the pelt. be C. rubidus, as it is so close to the Adung Valley, but
He described a skin from Koh Lak as black, grizzled on is C. maritimus. The skin is gray, with a black dorsal
the back by the whitish hair bases; the mane long and stripe; a prominent white gape streak; a white chest
matlike, with hairs 12 in. long, and gray, due to the and axillae; whitish buff shanks, with a brown median
mixture of black and white hairs; the throat reddish stripe, becoming reddish where the white meets the
gray; the lower lip and chin white, with some reddish black of the upper limb; and the hindlimb reddish
hairs on the head; and the shanks rusty brown. He also below the hock and inside, up to the groin.
mentioned another skin, where the shanks were red-
dish gray, owing to the mixture of red and white hairs. Capricornis rubidus Blyth, 1863
Gairdner (1915) shot a serow E of Si-sa-wad, in the Burmese red serow
Sai Yoke district of Ratburi, Quaa Yai River, at about
1863 Capricornis rubida Blyth. Arakan Hills.
14° 30' N. It had a black mane with white hair bases;
lips white; rufous markings on the base of the ears, Skins seen: 6.
the muzzle, the corners of the mouth, and the throat; Red-brown, with the hair bases black (merely red-
and shanks that were rufous, with a blackish median dish tones on the neck and the flanks), except for the
line. Myitkyina skin, best described as “red-toned black.”
Kloss (1919) named Capricornis sumatraensis annec- Mane very short, dark red. Underside white, the throat
tens from Koh Lak, and also received specimens of it and the interramal region usually white, but red in
from Si-sa-wad, Quaa Yai River, W Siam. He described Adung, and creamy red in Arakan; white over the
it as having the lower limbs largely rufous, but this nose. Legs red or buff y red, with a brown line in the
color not extending above the knee and the hock, un- front. Dorsal stripe black. Hair very short.
like C. milneedwardsi; white hair bases, so the body Tail very short (see table 68). Horns and forehead
color looks grizzled; the mane white at the base; a somewhat intermediate between C. sumatraensis and
large throat-patch, continuous with the white of the C. milneedwardsi.
lips; and a black chin. These are precisely the charac- Localities of specimens seen by us (all in Burma):
ters of C. maritimus. But in the Koh Lak peaks, Kloss Adung Valley, 28.10° N, 97.40° E, 9000 ft.; 150 mi. N
said, this form lives in association with animals in of Myitkyina; Mahtun (26.06° N, 95.58° E), 3000 ft.;
which the lower legs are almost entirely black; this Arakan, including Paletiva; Mahkawng Ga, Upper
suggests that there may be either hybridization or, Burma, 26° 49' N, 98° 10' E. U Tun Yin (1967) recorded
less likely, sympatry with C. sumatraensis. red serow from the following areas in Buma: Upper
A specimen collected by Gairdner at Ta Chang Tai Chindwin, Pakokku, Maymyo, Mu, Arakan Hill
(666 ft., E side of the Me Thot [or Me Taw] River, above tracts, Laukhaung and Putao, and limestone hills
Raheng, ca. 16° 50' N) was black, with the limbs rufous in the Salween Valley. He reported black serow
below the knee or the hock (Chasen & Kloss, 1930). with red legs from Mogok, Mongmit, Maymyo, and
-1— Gyldenstolpe (1916) obtained a serow from Koon Tan, Tharrawaddy, and black serow with black legs from
0—
349-47558_ch01_1P.indd 258 5/31/11 7:05 PM

b o v idae 259
Wetwun, Maymyo, and other localities in the N Shan Strongly red in color, with gray tones; hairs red,
states. with long white bases, not black bases, as in C. rubi-
Blyth (1863) described “C. rubida” from the Arakan dus. Legs with white knees and fetlocks. Rim of the
Hills as “of a red brown color, with black dorsal list; the lower jaw white, continuous with a white patch on
hair shorter than of the others.” The type is reputed to the throat. Underside a dirty grayish fawn. Long red
be no longer in existence, but there is a skin (no. 18039) mane. Apparently short-tailed (see table 67).
in the Bombay Natural History Society collection Localities of specimens seen by us: Longpa, 26°
from Paletiva, Arakan, which has the typical short red 15' N, 95° 25' E, Naga Hills; Mishmi Hills; Minbu,
hair with short black bases. Pocock (1910b) quoted “Mr. 20° 04' N, 94° 04' E, 2000 ft. (these are all full skins, in
Bird,” who said that red serow are common in the the Natural History Museum, London); and Mokokc-
limestone hills of the Salween Valley. hung (a mounted head in the Oxford University Mu-
There is evidently some sympatry, with possibly seum). There are also five skulls without skins, which
some interbreeding, between C. rubidus and C. mariti- are from the same general area, and we suppose that
mus. Milner (1921) saw red serow in the Maigthon they belong to the same species: three from the Garo
Hills, Mu Forest division; on some hills they tended to Hills, 2000 ft., Assam; and two from the Mishmi Hills,
be black, with reddish legs, and he did see an all-black one of them specified as being from 4100 to 6000 ft.
one. Peacock (1933) said that in the Chindwin drain- (i.e., 1230–1830 m.).
age, only red serow appear to be known, though he Choudhury (1997) stated that this form (which he
shot one there that was rather reddish purple; he saw called C. rubidus) is found to the S of the Brahmaputra
another that was sepia-brown, and still others that River, whereas black serow are found to its N. He gave
were dark brown. He noted that along the Irrawaddy the following districts: Tinsukia, Dibrugarh, Sibsagar,
River above Mandalay, the black form is the usual Nagaon, Morigaon, Kamrup, Karbi Anglong, N Cachar
one, while farther down, in the Toungoo and Salween Hills, Cachar, Hailakandi, Karimganj. It “occurs from
districts, both red and black serows occur; near Mogok as low as 100 m to the highest parts of Assam (Laika
the black form is commoner, though a red one was area in Barail Range, 1959 m), from tropical rainforest,
shot there. In Tenasserim, they are all black. deciduous forest, abandoned jhums (shifting cultiva-
C. rubidus is not closely related to any other spe- tion) to subtropical broadleaf forest on the Barails.”
cies. The shorter mane and the lack of white hair Choudhury (2003) presented an excellent color photo of
bases distinguish it from C. thar, C. maritimus, and C. a female and young animals of this species.
milneedwardsi, as well as from the undescribed species Pocock (1910b) quoted Annandale and Bentham
listed below. (in litt.), who said there were skins in Calcutta from
Shillong and Chittagong (CPG could not fi nd these
Capricornis Undescribed Species skins during a visit to the Zoological Survey of India
Assam red serow in Calcutta) with hairs that were white at base, with
We have seen four skins from the India/Burma bor- red tips, thus having no black in them; white knees
der region that cannot be assigned to any known spe- and fetlocks; a white jaw joining the white throat-
cies of serow, and evidently represent an undescribed patch; and, below, the body a dirty grayish fawn color.
species. This would extend the distribution to the S.
Table 67 Differences between Assam red, Burmese red, and Himalayan serow
Himalayan (C. thar) Assam red Burmese red (C. rubidus)
Overall color Jet-black, reddish tinge on the Red Red-brown (more red on the neck
flanks and the flanks)
Hair bases Straw-colored or white White Black
Legs Red/white; creamy white below Knees and fetlocks whitish All red or buff y red, with a
the knees/hocks brown line in the front
Nose, jaw, throat-patch White White White, creamy red, or red
Underside Sharply creamy buff Dirty grayish fawn White
Mane Long, black/white mixed Long, red Very short, dark red
—-1
—0
349-47558_ch01_1P.indd 259 5/31/11 7:05 PM

260 Artiodactyla
Whereas one might expect the Assam red serow upper neck, and the crown of the head. Short-tailed
to be either close to Burmese C. rubidus or else a series and short-eared (see table 68).
of hybrids between C. rubidus and C. thar, in craniomet- Locality of a specimen seen by us: 15 mi. E. of Ya-
rics they lie well away from C. rubidus and fully within magata, Mt. Gannto, Ou Mountains, Honshu. Most
the range of C. thar. The differences between the three specimens are ticketed merely “Japan.”
in external characters are given in table 67; again, the Jass & Mead (2004) have given full information on
Assam red serow resembles C. thar rather than C. rubi- this species.
dus. The Assam red serow is best interpreted as the
sister species of C. thar, and the erythrism that links it Capricornis swinhoii Gray, 1826
to C. rubidus would presumably be convergent ( just Formosan serow
possibly resulting from ancient gene flow?). In this re-
1862 Capricornis swinhoii Gray. Taiwan.
gard, it is of great interest that the erythristic goral,
Nemorhaedus baileyi, is found (at higher altitudes) in Skins seen: 6.
the same general area. Preorbital and foot glands present, but not de-
scribed in detail (Dien, 1963).
Other Serow Compared with the C. sumatraensis group, the lac-
Capricornis crispus (Temminck, 1845) rimal bone shorter and higher, with an anterosuperior
Japa nese serow angulation; facial angle steeper; crista facialis perhaps
more steeply downturned; nasofrontal suture penetrat-
1845 Antilope crispa Temminck. Honshu.
ing the frontals to some extent; some orbital tubularity;
1894 Capricornis pryerianus Heude. Tokyo.
skull broad. In these characters, C. swinhoii resembles
1898 Capricornulus saxicola Heude. Honshu.
C. crispus, but it differs strongly in several respects: na-
2n = 50. sal process of the premaxilla reaching much closer to
Skins seen: 8; as well as observations of about 10 the nasals; infraorbital foramen on the horizontal plane
living specimens in the Beijing and Berlin zoos. of the palate; temporal fossa long and low; parietals
Compared with the C. sumatraensis group, the pre- very strongly flattened; cheekteeth smaller. Like Japa-
orbital gland less deep; foot gland said to be more nese serow (C. crispus), the horns directed more up-
goral-like. Lacrimal fossa occupying more of the lac- ward, but resembling the C. sumatraensis group in form.
rimal bone; lacrimal bone shorter and higher, with an Hair short, thick, coarse; much woolly underfur;
anterosuperior angulation; facial angle steeper; crista uniform black-brown, brown, reddish brown, or dark
facialis more steeply downturned; nasal process of the brown, with a grayish or reddish tinge; face gray-
premaxilla ending only slightly in front of the level of brown. Chin, throat, and an area from the posterior
the anterior premolar; infraorbital foramen somewhat jaw up to the ears pale yellowish, russet, or reddish
above the horizontal plane of the palate; nasofrontal brown; underside of the neck yellow-bay; black zone
suture penetrating the frontals to some extent; short in the interramal space. Front of the legs, from the
frontal processes running lateral to the nasals, not knee and the hock to the hoof, brown or black, this
reaching the maxillae; some orbital tubularity; pari- area becoming broader near the hoof; reddish brown
etals often strongly vaulted; skull breadth more than at the back of the legs. Underside slightly lighter, with
42% of the length; cheekteeth very large, averaging a reddish tinge in the middle; hinder part of the belly
33% of the skull length. Horns directed more upward, creamy, sometimes suff used with red, going halfway
broader at the base, the rings more prominent. down the inside of the thighs. Dorsal stripe thin or
Long white underwool; guard hairs white, with unclear; slightly lengthened nuchal stripe.
black or red-brown tips, but these wearing off in the Infants entirely fawn-brown; hair short; white
course of a season, with the overall tone sometimes streak behind the corner of the mouth; white patch
becoming whiter or grayer; underside white; dark on the throat; faint dorsal stripe.
patch on the withers; thin, wavy, black dorsal stripe. Localities of specimens seen by us (all in Taiwan):
Face blackish brown on the top and the sides of the Nan-Tou Hsien; Wu-Liu Tiung, Shing-I (purchased in
muzzle and around the eyes; white cheeks, lips, and Puli); Wu Lai; I-lan I-lan; Puli.
preorbital glands; white around the rhinarium. Legs
blackish brown, with or without a white ring around Table 68 gives external measurements for both serow
-1— the hoofs. Long, fluff y hair around the cheeks, the and goral taxa. The short ears of Capricornis suma-
0—
349-47558_ch01_1P.indd 260 5/31/11 7:05 PM

b o v idae 261
Table 68 External measurements for serow and variation. For example, in his table of measurements,
goral taxa the range of variation of 16 skulls less than 8 years of
age from N Grant Land (on Ellesmere Island), referred
Tail Hindfoot Ear to O. m. wardi, nearly all completely cover those of the
other putative subspecies. The main exceptions are as
Serows
follows:
sumatraensis 7.8–11 .0 (4) 26.7 (1) 7.4–11.3 (2)
maritimus 7.1–10.3 (4) 21.4–24.5 (2) 12.3–13.2 (3) —horn breadth at the base—O. m. wardi 131–
milneedwardsi 4.9–12.3 (7) 22.7–28.9 (8) 11.7–13.3 (6)
219 mm, O. m. moschatus 201–250 mm,
rubidus 5.1 (1) 26.9 (1) 11.3 (1)
red/white 7.0 (1) 25.5 (1) 11.4 (1) O. m. niphoecus 192–241 mm
swinhoii 6.2–13.7 (7) 22.9–28.0 (3) 10.7–15.2 (7) —mastoid breadth as a percentage of the
crispus 5.7–5.7 (2) 27.0–27.1 (2) 8.9–9.5 (2) basal length—O. m. wardi 38.3%– 41.9%,
Gorals O. m. moschatus 37%– 40.7%, O. m. niphoecus
caudatus 9.3 (1) — 10.9 (1) 30.7%–39%
griseus 10.1–18.0 (8) 20.9–26.9 (8) 12.0–14.2 (6)
evansi 12.7–14.7 (9) 25.5–28.0 (9) 11.1–14.0 (9) In the photographs, slight differences can be de-
goral 13.3–13.3 (2) 15.9–23.3 (2) 11.1 (1) tected in the amount of white on the forehead and
baileyi 7.6–10.3 (2) 20.5–23.4 (2) 9.9–10.3 (2)
between the horns; but these, too, seem to be quite
variable. The question needs to be reexamined; in the
meantime, we recognize no subspecies in just a sin-
gle valid species.
traensis, compared with most of the other serows, are
very noticeable. Among goral, Nemorhaedus evansi Tribe Cephalophini Blyth, 1863
has very long distal extremities, as represented by the duikers
hindfoot length, while N. baileyi has a short tail and
Grubb (2004a) noted that, strictly speaking, the name
short ears.
Cephalophini Blyth, 1863, is antedated by Sylvicaprini
Sundevall, 1845 (as “Sylvicaprina”), but Art. 35.5
Ovibos de Blainville, 1816
stipulates that an older name should not displace a
musk- ox
younger but universally used one.
1816 Ovibos de Blainville. Type species: Bos moschatus
Duikers mostly possess preorbital, foot, and inguinal
Zimmermann, 1780.
glands; agouti-banded hairs; relatively short limbs,
1911 Bosovis Kowarzik.
with the withers lower than the rump; a spatulate
Very broad horn base, covering most of the postor- I1, but with the other incisiform teeth narrow and
bital region. Horns abruptly curving down in both keeled; brachyodont cheek teeth; and relatively
sexes, curving up again toward the tip; whitish, very poorly developed jaw muscles, especially the masse-
rough, fibrous basally, smooth at the tip. Limbs short, ter, with the areas of origin and insertion on the cra-
stout; tail vestigial; pelage very full, with long, coarse nial bones relatively small—in par ticu lar, the origin
overhair, nearly reaching the ground; very long, of the masseter does not approach the preorbital
thick, woolly underfur. Orbits strongly tubular. Rhi- fossa, so that the two are not separated by a promi-
narium narrow. (Following J. A. Allen, 1913.) nent ridge. The origin of the main head-flexing mus-
cle is not marked by basilar tuberosities; the mastoid
Ovibos moschatus (Zimmermann, 1780) is narrow; and the foramen ovale is extremely small
(a character shared with Oreotragus). Duikers have a
1780 Bos moschatus Zimmermann. Hudson Bay.
long skull, with a relatively stout rostrum, nontubu-
1900 Ovibos moschatus wardi Lydekker.
lar orbits, and the braincase not strongly flexed back
1905 Ovibos moschatus niphoecus Elliot.
on the facial part of the skull; the nasal bone contrib-
1910 Ovibos moschatus mackenzianus Kowarzik. Great Slave
utes to the wall of the rostrum and is firmly sutured
Lake.
to the maxilla. The horns are short and straight; the
1910 Ovibos moschatus melvillensis Kowarzik.
horn cores are keeled, inserted behind the orbits,
J. A. Allen’s (1913) monograph, while thorough and and much inclined. There is little size difference
remarkably detailed, is not convincing about geographic between the sexes, and quite often females average —-1
—0
349-47558_ch01_1P.indd 261 5/31/11 7:05 PM

262 Artiodactyla
slightly larger than the males. (Following Grubb & Taxonomic and geographic variation in Sylvicapra
Groves, 2001.) concerns the following features: overall body tone,
On an analysis of 12S and 16S sequences, Kuznetsova degree of speckling, amount of darkening on the
et al. (2002) favored an association of the Cephalophini mid-dorsal region, tone and development of the face-
with the Reduncini. This association was not strongly blaze, amount of black on the tail, color of the pat-
supported, however, when, a nuclear sequence (b- terns, color and development of the stripe on the
spectrin) was later analyzed (Kuznetsova & Kholodova, forelegs, and the degree and tone of whitening on the
2003). underparts; overall size, length of the tail, length and
In the morphologically based cladogram of Grubb shape of the ears; relative skull breadth, elongation of
& Groves (2001), the fi rst split separates Cephalophus the premaxillae, and development of the nasals. This
from a branch which combines Sylvicapra and Philan- variation is, however, not at all marked (with a few
tomba. Within Cephalophus, the monophyly of the C. exceptions); one is as likely to fi nd clinal variation
sylvicultor and the C. ogilbyi groups is supported. within a single subspecies as to fi nd “stepped” varia-
Within the C. silvicultor group, there is a C. dorsalis ver- tion between subspecies. Variation in the skull char-
sus jentinki/sylvicultor/spadix split. C. adersi was the sis- acters is very great within a given population, so that,
ter species to a branch containing both the C. ogilbyi while it can be broadly stated that the SW races are
and the C. natalensis clades in the C. ogilbyi group. C. the largest, and the N ones the smallest, it is in no
niger was the first species to separate from the C. ogilbyi way possible to distinguish a par ticu lar taxon on
clade. C. rufilatus was a sister species to the C. nigrifrons skull characters alone.
group, followed by C. leucogaster, and then by C. na- We divide the genus into three species; in effect,
talensis and C. harveyi. this simply splits the two most strikingly different
The phylogeny of Jansen van Vuuren & Robinson taxa from all the rest. To that extent, it is unsatisfac-
(2001), based (in effect) on the mitochondrial cyto- tory. Reexamination of some of the subspecies we have
chrome b and 12S rRNA genes, is almost exactly that of placed in S. grimmia may well show that there are a
Grubb & Groves (2001), based on morphological char- few further species to be recognized.
acters. The phylogeny of the three duiker genera, how-
ever, is slightly different: Sylvicapra grimmia (Linnaeus, 1758)
common bush duiker
(Philantomba (Sylvicapra, Cephalophus)).
The “subspecies” of S. grimmia, if that is what they
Within Cephalophus there are three major species- are, can be divided, broadly speaking, into a rela-
groups, plus two “left over” (C. zebra and C. adersi); these tively primitive, unspecialized group (S. g. caff ra, S. g.
latter two groups sometimes assort with each other, hindei, S. g. nyansae, S. g. madoqua, S. g. campbelliae, and
but more usually appear as separate major lineages of the Afroalpine forms), and a more derived group. The
their own. members of the fi rst group have retained the agouti
Our arrangement more or less follows that of hair banding: the hair bases are gray (usually light
Grubb & Groves (2001), except for the recognition of gray, but darker in the Afroalpine races), followed by
diagnosably distinct “subspecies” as full species. a brown band (varying from a pale straw color in
most to nearly white in S. g. caff ra, with the band nar-
Sylvicapra Ogilby, 1837 row in S. g. caff ra, and broad in S. g. madoqua), and then
Medium sized; slenderly built, with long, slender legs a dark tip, which is usually quite long. In the second
and particularly elongated distal segments; long neck; group, this pattern is modified, usually by a lighten-
broad pelvis, with an elongated and widely spaced is- ing of the dark zones and a darkening (or reddening)
chia; pelage primitively agouti-patterned, usually of the pale zones; by the common elimination of the
strongly so; coronal tuft long, narrow, dominated by dark tips; and, therefore, by a lack of overall speck-
the forwardly directed hairs. Skull with a straight ling. Speckling in this group is always seen—variably
profi le, an enlarged narial opening, somewhat tubu- developed—in the juveniles, and it fades to a greater
lar orbits, and an oblique occipital plane; nasals with or lesser extent with maturity. The members of the
long median processes; median palatal notch very unspecialized group are really all very much alike;
broad, V-shaped. Horns (present in the males only) subspecies are difficult to distinguish, and even such
long, close together, laterally flattened, rising at a geograph ically distant forms as S. g. caff ra (SE Africa)
-1— steep angle from the skull, close behind the orbits. and S. g. madoqua (Ethiopia) are dismayingly similar
0—
349-47558_ch01_1P.indd 262 5/31/11 7:05 PM

b o v idae 263
in appearance. Very minor differences separate most Sylvicapra grimmia caffra Fitzinger, 1869
of the various forms, and the validity of some of them
1869 Sylvicapra mergens caff ra Fitzinger. “Kaffi rland”; Natal,
is questionable. The various Afroalpine taxa, some
according to A. Roberts (1951).
not described, differ in minor (but apparently consis-
1871 Grimmia irrorata Gray. Natal.
tent) ways from one another and from their low-
1926 Sylvicapra altifrons noomei Roberts. Maputa River, S
land relatives; the small degree of difference is read-
Mozambique.
ily explicable if one calculates a maximum of 10,000
1926 Sylvicapra grimmia transvaalensis Roberts. Rustenberg
years for the total separation of the E African alpine
district, Northwest Province, South Africa.
zones from the lowland savannas. The various taxa
of the specialized group differ more from one an- Well-speckled fawn-brown to gray-yellow; mid-dorsal
other, and they are not to be regarded as a mono- zone darker; facial blaze nearly black, well marked up
phyletic assemblage. to the level of the eyes, then fading on the forehead;
stripe on the foreleg varying in width, but always well
Sylvicapra grimmia grimmia (Linnaeus, 1758) expressed, extending from the hoof to the knee; simi-
lar but shorter stripe present on the hindleg. Underside
1758 Capra grimmia Linnaeus. Cape Town (restricted by O.
mainly light buff, but the belly white, at least in the
Thomas, 1911).
midline, this color extending to the groin and the inner
1811 Antilope nictitans Thunberg. Cape of Good Hope.
sides of the hindlimbs right down to the hoofs, then
1816 Antilope mergens Desmarest. Cape of Good Hope.
reappearing on the axillae and the inner sides of the
1816 Cemas cana Oken; nomenclature in Oken’s work has
forelimbs, and on the interramal region. Tail nearly all-
been declared unavailable by the ICZN.
black on the upper surface, white below. Pasterns black.
1827 Antilope burchelli Hamilton Smith. W side of Caff raria;
In general, giving the appearance of a brown animal
fi xed by A. Roberts (1951) at Zwartwater Poort, Albany
sprinkled with a whitish color; on individual hairs, the
district.
subterminal pale band very narrow, the brown zone
1827 Antilope platous Hamilton Smith. “Vicinity of the
below it very broad, the gray bases rather dark.
Gareep” (i.e., the Orange River), in Hamilton Smith’s
KwaZulu-Natal, the former Transvaal, S Mozam-
contribution to Griffith’s Cuvier’s Animal Kingdom
bique (Coguno).
(4:260), but later (5:345) changed to “mountains on the
west side of Caff raria.”
Sylvicapra grimmia steinhardti Zukowsky, 1924
1827 Antilope ptoox Hamilton Smith. Cape of Good
Hope. 1924 Sylvicapra grimmia cunenensis Zukowsky. Otjonganga,
SE of the Omuhonga Mountains, N Kaokoveld,
On the evidence of eight skulls, this would seem to be
Namibia.
the largest subspecies (greatest skull length 186–207 mm).
1924 Sylvicapra grimmia omurabae Zukowsky. Otjomikambo,
The Leiden skins (4 adult, 4 juvenile) confirm the de-
Grootfontein district.
scription given by A. Roberts (1951), who described it as
1924 Sylvicapra grimmia steinhardti Zukowsky. Otjikuara,
“dull ochraceous tawny,” with the development of a
source of the Hoamib River, Namibia.
medium amount of speckling; facial blaze incomplete,
1924 Sylvicapra grimmia ugabensis Zukowsky. Goreis, 45 km
not extending to the intercornual crest; white of the un-
W of Outjo, N Damaraland, Namibia.
derparts restricted to the throat and the inside of the up-
1926 Sylvicapra grimmia bradfieldi Roberts. Quickborn Farm,
per half of the legs; foreleg-stripe well developed. S. g.
N of Okahandja, Damaraland.
grimmia thus closely resembles S. g. steinhardti (below),
1942 Sylvicapra grimmia vernayi Hill. Kaotwe Pan, Kalahari,
differing essentially in its darker ground color with more
Botswana.
speckling, and probably in its larger size.
S Cape, South Africa. Very pale, sandy-colored or fawn; immature animals
The W “Caff raria” (Albany district, Algoa Bay) darker, grayer. Speckling very reduced, with the con-
population seem to be essentially intergrades be- trast between the light and the dark bands obliterated,
tween S. g. grimmia and S. g. steinhardti. A. Roberts and very few black tips, even in the mid-dorsal region.
(1951), describing the “Caff raria” population as S. g. Facial blaze failing to reach even the level of the eyes;
burchelli, said that they are grayer and less tawny than forehead pale ochre. Leg-stripes broad, black, reaching
S. g. grimmia, but that they have the same absence of the knees and the hocks; pasterns black. Underside off-
white on the belly. white; chest and throat buff y; only the upper parts of —-1
—0
349-47558_ch01_1P.indd 263 5/31/11 7:05 PM

264 Artiodactyla
the inner limb surfaces white, the rest pale buff. Tail side of the limbs; occasionally with a throat streak
black for the terminal two-thirds. (commoner in S specimens); pasterns black; tail black
Namibia; S as far as Port Nolloth, W Cape, South for at least the terminal half, sometimes nearly entirely
Africa; N to Namburi and Cahama, Angola; E to Ka- so. Females considerably larger than the males.
zungula, the Chobe district, and Kuka Pan, Botswana. N of the Zambezi River in Mozambique and Ma-
A specimen from Bambei, Okovango, represents a lawi, Zambia E of the Luangwa Valley, N through
gradation toward S. g. splendidula (below). Tanzania (except the far NW) to the Kenya coast, in-
land as far as Voi, and the Juba River, Somalia.
Sylvicapra grimmia splendidula (Gray, 1871) S specimens are more reddish, with a longer, darker
facial blaze, and larger; N specimens tend to be very
1871 Grimmia splendidula Gray. St. Paul de Loanda, Angola.
pale, and smaller.
1894 Cephalophus grimmia flavescens Lorenz. Victoria Falls.
The controversial taxon Cephalophus walkeri is a
1899 Cephalophus leucoprosopus Neumann. St. Paul de Loanda.
melanistic example of S. g. orbicularis (Grubb, 1988);
1919 Sylvicapra grimmia uvirensis Lönnberg. Uvira and
hence, it can be assigned to this subspecies.
Baraka, W of the N end of Lake Tanganyika.
Light, a bright reddish ochre; black tips prominent on Sylvicapra grimmia hindei (Wroughton, 1910)
the midback, but not elsewhere; face-stripe generally
1910 Cephalophus abyssinicus hindei Wroughton. Fort Hall
short, as in S. g. steinhardti; leg-stripes only sometimes
(Murango), Kenya.
reaching the knee and the hock, but well marked; tail
usually black for the terminal three-quarters, with a An “unspecialized” form, like S. g. caff ra, with much
black line extending forward to the root; underside as speckling (especially in the mid-dorsal region) on an
in S. g. steinhardti, but with the white on the inner sur- ochery ground; facial blaze quite bold, broad and dark,
faces of the limbs extending right to the hoofs, and extending to the crest; leg-stripes unclear, often failing
commonly with a whitish streak from the interramal to reach the knee and the hock; underside buff y white;
region to the throat; paler region of the muzzle and the throat wholly buff, with no white line; white on the
circumocular zone often quite white, clearly marked inner sides of the limbs failing to reach the hoofs; pas-
off. Size less than in S. g. steinhardti; females larger than terns dark brown, not black; tail dark brown above
the males. for the terminal three-quarters. Very small (greatest
Angola, from the Mupa district N to Kunungu, the skull length 159–171 mm), broad-skulled.
DRC, and across the river to Odzala, Congo Republic; Kenya highlands, E of the Rift Valley, as far E as
SE to the S end of Lake Tanganyika; NE to Uvira; in Machakos and Sultan Hamud, and N to the latitude
Zambia, W of the Luangwa Valley, N as far as Mpika; of Mt. Kenya; S as far as Moshi, Tanzania.
Zimbabwe.
Sylvicapra grimmia nyansae Neumann, 1910
Sylvicapra grimmia orbicularis (Peters, 1852)
1910 Sylvicapra abyssinica nyansae Neumann. Kavirondo.
1852 Antilope altifrons Peters. Sena.
Very like S. g. hindei, but brighter in color, with less
1852 Antilope ocularis Peters. Substitute for Antilope orbicularis.
mid-dorsal darkening and whiter underparts; facial
1852 Antilope (Cephalophus) orbicularis Peters. Sena, N bank of the
blaze deep brown rather than blackish; tail less than
Zambezi River, Mozambique (fixed by Ellerman et al., 1953).
half blackened; median line running partially down
1906 Cephalophus walkeri Thomas. Tuchila River, Shire
the throat from the interramal region. White down
highlands, Malawi.
the inner limbs tending to reach farther toward the
1910 Cephalophus abyssinicus shirensis Wroughton. Zomba,
hoofs. Size small. Ears shorter.
Malawi.
Kenya W of the Rift Valley, from S Ewaso Ngiro N
1913 Sylvicapra grimmia deserti Heller. Voi, Kenya.
to Torit in Sudan and Roseires in Ethiopia, on the
Light fawn, without much ochery tinge; black tips in Blue Nile; into the Rift Valley at Elmenteita; W to the
the mid-dorsal region, grayer toward the haunches; fa- Busoga district, Uganda.
cial blaze narrow, usually not reaching above the eyes,
but sometimes vaguely continuing to the crest; foreleg- Sylvicapra grimmia altivallis Heller, 1912
stripe narrow, often reaching above the knee, but the
1912 Sylvicapra grimmia altivallis Heller. Kinangop Peak,
-1— hindleg-stripe not reaching the hock; underparts white
Aberdares, 10,500 ft. (= 3200 m).
0— to buff y, this tone reaching halfway down the inner
349-47558_ch01_1P.indd 264 5/31/11 7:05 PM

b o v idae 265
Grayer than S. g. hindei, with heavier speckling, a more specimens, we have studied a skull in Nairobi and
marked face-stripe (thick, black, complete), black pas- two skins— one immature, and one mature but lack-
terns, and a drabber tone on the underparts; ear short, ing the head—in London. This still does not amount
blunt; nasal bones relatively short; fur long, thick. to material sufficient to characterize a new subspe-
S. g. altivallis shows its relationship to S. g. hindei— cies, although it is surely distinct.
which lives at the foot of the mountains whose alpine Like S. g. altivallis, this alpine-zone duiker seems
zones S. g. altivallis inhabits, and from which S. g. alti- fairly clearly derived from S. g. hindei, with its gray-
vallis is evidently derived—in its dull color, the midbrown color; thick, black facial blaze; reduced black on
back darkening, the amount of black on the tail, the the tail; and off-white underparts not extending down
form of the leg-stripes, and the absence of a white the inner sides of the limbs. It differs from S. g. altivallis
throat-streak. in its usually heavier speckling, with less black on the
Afroalpine zone of the Aberdares and also, proba- tail; the foreleg-stripe that usually reaches the knee;
bly, of Mt. Kenya (per a specimen in the Nairobi Mu- the “squared-off ” ears (King 1975); the shorter tail; and
seum). On Mt. Kenya, S. g. hindei reaches up to the even shorter nasals than in S. g. altivallis.
edge of the forest belt, at 1980 m, and the two taxa are
isolated from one another by unsuitable habitat (mon- Sylvicapra grimmia madoqua (Rüppell, 1835)
tane forest).
1835 Antilope madoqua Rüppell; not rendered unavailable by
A. madoka Hamilton Smith (= Madoqua saltiana, a dik- dik
Sylvicapra grimmia lobeliarum Lönnberg, 1919
[Neotragini]), contra O. Thomas [1892]). Galla, W of
1919 Sylvicapra grimmia lobeliarum Lönnberg. Lobelia zone of Massawa (selected by Grubb & Groves [2001]); and the
Mt. Elgon. Kullaf Mountains.
1892 Cephalophus abyssinicus Thomas. “Abyssinia.”
Pinkish gray–toned, with heavier speckling than S. g.
nyansae; face-stripe thick, chocolate-brown; whole Generally gray-buff to brownish ochre, with heavy
upperside of the tail black; pasterns black; foreleg- speckling; mid-dorsal zone darkened; face-stripe com-
stripe usually extending to the knee; no white streak plete in the highland forms, less so in the lowland spec-
down the throat. Long, thick fur. The only com- imens (Harrar, Abaya, Gazgay); tail usually nearly all-
plete skull—the type—is extremely small, smaller black above, but the black much reduced— on the tip
than any skull of any race so far (this diminution is alone—in the Harrar specimens; foreleg-stripe diff use,
equaled only by a few specimens of S. g. madoqua), even occasionally absent, but, when present, usually
but this and three similar-sized but incomplete skulls reaching the knee; tending to be buff y-toned in the
in the British Museum are comparatively very broad; white areas underneath, but purer white in the Harrar,
no other subspecies shows such an extreme length– Lake Abaya, and Lake Tsana specimens; no throat
breadth relationship. Nasals seem short, but not as streak; no distal extension of the white color on the
shortened as in S. g. altivallis. Ear short, but normally inner surfaces of the limbs, except in a few individu-
pointed. als from Shoa and one from Gazgay. Fur generally
Afroalpine zone of Mt. Elgon. (On the lower slopes, thick and long, especially in the highland areas
below the montane forest belt, S. g. nyansae reaches el- (Semyen, Gojjam, Shoa, Arussi). Skull as short as those
evations of up to 1830 m.). of the various Kenya taxa, or shorter; skull relatively
In most characters, S. g. lobeliarum seems the most broad, but not as much as in S. g. lobeliarum; short na-
distinctive of the Afroalpine forms. It has some fea- sals; tail apparently fairly long.
tures recalling S. g. nyansae—a less-darkened facial Ethiopian highlands, both E and W of the Rift
blaze, a rather bright color, a tendency for the white of Valley; from as high as 2990 m in the Sahatu Moun-
the underside to extend down the inner side of the tains to as low as 610 m at Hawash; it is not certain,
limbs—but S. g. lobeliarum differs in such features as however, that these all belong to one taxon.
the pinkish tone of the fur, the blackening of the whole S. g. madoqua is a quite variable form, difficult to
upper surface of the tail, the long leg-stripes, and the defi ne, although the various component samples do
lack of a throat-streak. differ as a whole from the various Kenya taxa and from
S. g. campbelliae. The Ethiopian taxon (S. g. madoqua)
A Possible Further Sylvicapra grimmia Subspecies differs from the Kenya nonalpine forms in its grayer
This is the duiker of the alpine zone of Mt. Kiliman- color, heavier speckling, less developed facial blaze, —-1
jaro, described by King (1975). In addition to King’s and longer foreleg-stripe, although all these features —0
349-47558_ch01_1P.indd 265 5/31/11 7:05 PM

266 Artiodactyla
are subject to variation. It differs from the Afroalpine

taxa of the Aberdares, Kilimanjaro, and Elgon in its Sylvicapra pallidior Schwarz, 1914
grayer tone, less developed facial blaze, and brown Sahel bush duiker
(not black) pasterns. It differs from S. g. campbeIIiae in
1914 Sylvicapra grimmia pallidior Schwarz. Mani, Lower
its less developed facial blaze and more vaguely devel-
Shari River.
oped leg-stripe, as well as in its (usually) smaller size.
This variability, and the rather poor degree of dif- Pale buff, more weakly speckled, but still with mid-
ferentiation, doubtless reflects the fact that this dorsal darkening; facial blaze extending to the crest,
largely Afroalpine subspecies is not isolated by exten- but more diff use than in S. g. campbelliae; tail dark
sive forest belts from its lower-living relatives, unlike above for over half of its length; foreleg-stripe long,
the Aberdares, Kilimanjaro and Elgon forms. The but diff use; white always extending down the inside
Lake Abaya, Hawash, Gazgay, and Harrar popula- of the limbs to the hoofs. Ears longer. Size very small
tions may be reinvasions of lower habitats, as they (greatest skull length 158–165 mm).
bear little relationship to other low-altitude taxa S or Sahel zone, from Mani (furthest W locality) as far
W of the Ethiopian highlands—with the possible ex- E as Gallabat, on the borders of the Ethiopian high-
ception of Gazgay. Detailed descriptions of variations lands; N as far as Jebel Marra, Darfur; S as far as Fort
will be found in Grubb & Groves (2001). Archambault.
As far as the skulls as concerned, the range of val- Several localities are quite close to the distribu-
ues is small; again, nothing clear cut can be said about tional range of S. g. campbelliae, but the two taxa are
it. In the males, there seems to be a division in size always readily distinguishable, and the savanna-sahel
between those from Shoa and Harrar, and those from changeover always comes between them. (It is not
Arussi, Semyen, and Gojjam— a nonsensical result known whether any form of bush duiker extends into
geographically, and doubtless the result of parallelism. the Sahel W of Lake Chad.)
Tail length appears greatest in the Shoa specimens,
and least in those from Semyen; there may be some- Sylvicapra coronata (Gray, 1842)
thing in this, as the Semyen Massif is the highest of all, crowned bush duiker
while Shoa is the least high of the various ranges.
1842 Cephalophora coronata Gray. “Western Africa.”
Sylvicapra grimmia campbelliae (Gray, 1843) Bright orange-yellow in color, with no speckling
in the adults (hardly detectable even in the juveniles),
1843 CephaIophus campbelliae Gray. “Sierra Leone”; more
but the mid-dorsal region a somewhat darker reddish
likely Nigeria (Pocock, 1910b).
shade; facial blaze deep red, not extending to the crest;
1912 Sylvicapra grimmia roose velti Heller. Rhino camp, NW
tail black only at the tip; pasterns brown; foreleg-stripe
Uganda.
very indistinct, not reaching the knee; hardly ever even
Dull ochery to gray-buff, heavily speckled, and darker a trace of a stripe on the hindlegs; underside yellowish
in the mid-dorsal zone; facial blaze dark, usually ex- white, with no throat streak; only the upper halves of
tending to the crest; tail black, nearly to the root in E the limbs whitish on their inner surfaces. Skull small
specimens, but less and less black toward the W, so (greatest skull length 161–169 mm), but not with the
that the skin from Ejura, Ashanti, has hardly any black allometrically determined broadness and short-faced
at all. Pasterns brown. Foreleg-stripe well marked, characteristics of small individuals of other races; skull
reaching the knee. Buff y white to white below; no curiously narrow, with long premaxillae.
throat streak; usually no extension of the white color This very distinctive species has been recorded
down the inner surfaces of the limbs. Skull fairly small, only from a very small area on the borders of Guinea,
with much the same size range as S. g. madoqua; aver- Guinea-Bissau, Senegal, and Gambia.
aging large in the SE—Burundi, Ankole— and again
in the Shari district, and smaller in size to the NE and Philantomba Blyth, 1840
the W.
1840 Philantomba Blyth. Type species: Antilope philantomba
Savanna country, from Burundi and Karagwe in
Hamilton Smith, 1827; = P. maxwelli (Hamilton Smith,
the SE, N to Bahr-el-Ghazal, W via NE Congo, south-
1827), by tautonomy.
ernmost CAR, S Chad, N Cameroon, and Nigeria to
1852 Guevi Gray. Type species: Antilope maxwelli Hamilton
-1— Ghana (Ashanti) and Burkina Faso (Fada-N’Gourma).
Smith, 1827.
0—
349-47558_ch01_1P.indd 266 5/31/11 7:05 PM

b o v idae 267
Very small size; head plus body length <750 mm; small size and relatively short nasals—is due to
thick mat of hair on the forehead, darker and browner independent insular dwarfi ng.
than the rest of the face; often a buff or red stripe
above each eye; coat identical in color in the adults and Philantomba walteri Colyn et al., 2010
the juveniles, with no agouti pattern and no strong Verheyen’s duiker
contrast between the shaft and the tip; color pattern Color rather light, with the flanks tending to
on the body distinctive, dark on the dorsum, with the buff y, rather than to gray. Hair on the midline of the
flanks usually markedly lighter, sometimes a sharply neck always reversed. Horns of the males usually
marked-off white strip on the underside; dorsal tone longer; females always with horns, although the
darkening fairly rapidly on the croup and the tail, horns much smaller than in the males. Nasal bones
with the sharpest transition from the dorsal to the shorter.
lateral tone taking place on the haunch. Skull small, From the W bank of the Niger River to Benin and
delicate; constricted rostrum and anterior nasal re- Togo.
gion, with the anterior width of the nasals only half
that at the nasofrontal suture; orbits somewhat tu- Philantomba monticola Group
bular; occipital plane oblique; preorbital fossa shal- blue duiker
low, but very extensive; horns small, delicate, up- Smaller size than in the P. maxwelli group; color
turned at the tips, very reduced (or even absent) in brighter, darkening on the rump and the tail, with a
the females. sharply bordered transition to a lighter shade on the
haunch; flanks and legs noticeably lighter than the
Philantomba maxwelli Group back. Cheeks less lightened than in the P. maxwelli
Differs from the P. monticola group in various group, with less contrast between the midface and the
characters: size much larger; tail longer (106–160 mm); cheeks. Tail shorter, <122 mm, and almost always
zygomatic breadth greater; orbital borders less protu- <100 mm. Skull shorter, narrower, but with somewhat
berant; palate broader; free ends of the nasals broader; tubular orbits; rostrum sharply constricted, with the
no sharp break on the haunches between the dark palate and the snout narrower, and the nasals narrow-
croup and the light color of the flanks and the lower ing strongly toward the tips.
haunches; cheeks lighter; superciliary streaks more Unlike Sylvicapra, Philantomba is quite easy to di-
prominent, and much deeper; foot gland larger. vide into different species, not only separating P. max-
Members of this group are apparently less special- welli (as is universally done), and the newly described
ized than the P. monticola group, except in their feet P. walteri, but separating what has hitherto been re-
and, possibly, their larger size. garded as a single species (P. monticola) into a number
of species.
Philantomba maxwelli (Hamilton Smith, 1827) The various species comprising the P. monticola
Maxwell’s duiker group can be divided into two major groups: a gray-
Color light gray-yellow-ochre to a rather dark gray- legged one and a red-legged one. The first group con-
brown, slightly browner on the middle of the back; tains P. congicus, P. melanorheus, P. aequatorialis (in-
yellow-white below, with the groin white. Hair very cluding P. musculoides), P. sundevalli, and P. lugens. In
rarely reversed on the neck. Females usually horn- these, the legs are much the same color as the body,
less; horns of the males comparatively short. Females even though there is still a well-marked transition of
average larger than the males. lighter tones on the haunch; also, the hair on the mid-
Ghana, Ivory Coast, Liberia, Sierra Leone, Guinea, line of the neck is often reversed. The second group
and Guinea-Bissau to Senegal (type locality: Bignona) contains P. simpsoni, P. anchietae, P. defriesi, P. hecki, P.
and Gambia (type of Philantomba whitfieldi Gray, bicolor, and P. monticola. In these, the legs and haunches
1850). are distinctly red-toned; the fur is thicker and softer;
Grubb & Groves (2001) recognized the duikers and the neck hair is never reversed. As in P. maxwelli,
from Yatward and Sherbro islands—the one in the there are populations in which the females are pre-
Rokelle River, the other farther down the coast— as a dominantly hornless; as in P. musculoides, those in
distinct subspecies, Philantomba maxwelli danei (Hin- which no female skull examined had horns; and, as in
ton, 1920). But now it seems likely that what the sam- P. aequatorialis, those in which hornless females pre-
ples from the two separate islands have in common— dominated in most series. —-1
—0
349-47558_ch01_1P.indd 267 5/31/11 7:05 PM

268 Artiodactyla
Gray-Legged Group Philantomba aequatorialis sundevalli (Fitzinger, 1869)

Philantomba congicus (Lönnberg, 1908) Light in color, brownish; hardly any color differen-
west-central African blue duiker tiation between the dorsum and the flanks; croup not
Dorsum bright, strong gray-brown to black; flanks black, but still with quite a sharp transition between
pale grayish, the haunches browner; clear, dark the two colors; haunch somewhat pinkish; white be-
brown, horizontal stripe on the haunches marking low. Size small; horns fairly long in both sexes.
the sharp transition from the blackish brown of the Pemba, Mafia, and probably Zanzibar; also the E
croup to the paler haunch. Underside broadly white. African coast.
Cheeks red-tinged.
Median nuchal hair reversal is usual. Philantomba lugens (Thomas, 1898)
From the E bank of the Niger River SE to the Congo Tanzanian highland blue duiker
River, and across the Oubangui River as far E as Lisala. Very dark, gray-brown, without much contrast be-
tween the flanks and the dorsum; croup black, but the
Philantomba melanorheus (Gray, 1846) haunches so comparatively dark that the transition
Bioko blue duiker line not obvious; legs dark, except for a light streak
Related to P. congicus, but even more contrasty, with a down the front; underside gray or gray-fawn, with (at
black dorsum and pinkish gray flanks; fur long and most) only a narrow streak of white. Horns very long.
thick; very small in size; horns very long. Hair generally reversed on the midline of the neck.
Median nuchal hair reversal is usual. Highlands of Tanzania, from Usambaras S to
Bioko Island. Irangi, and to the borders of Malawi at Tukuyu; also
Uhehe, Kigoma district.
Philantomba aequatorialis (Matschie, 1892)
East African blue duiker Red-Legged Group
A drab-brownish species, with very restricted black Philantomba simpsoni (Thomas, 1910)
on the croup; light gray below, with a tolerably sharp Simpson’s blue duiker
transition on the haunches. Nasal bones short. Mid-dorsal region blackish brown, restricted to a broad
band in the middle of the back; flanks red-brown, pal-
Philantomba aequatorialis aequatorialis ing to a broadly white belly; haunches redder than the
(Matschie, 1892) flanks, with a poorly marked transition stripe; legs
Drabber and browner than P. congicus; black on the brownish, pale, but not white, on the inner surfaces.
croup more restricted; underside usually light gray; Stripe above each eye well marked. Size small (greatest
transition on the haunches less sharp than in P. congi- skull length 114–135 mm); horns of the males fairly
cus, but still very noticeable. Nasal bones, on average, long, but not those of the females (the females occasion-
longer. Females usually hornless. Females tend to av- ally lack horns). Premaxillae often failing to contact the
erage slightly larger than the males. nasals (27 out of 45 specimens), and, in any case, never
From the Uele district into Uganda, at least to Bu- making more than a point contact with them.
nyoro, and probably as far as the Nile; NE to the Ima- Between the lower Congo and Kasai rivers.
tong Mountains in Sudan; S in the DRC, to 4° S. Dui- There are specimens apparently intermediate be-
kers in the Parc National de l’Upemba are intermediate tween P. simpsoni and P. anchietae from the Kwa-
between P. a. aequatorialis and P. defriesi, while in the mouth region, S of the lower Kasai; between P. simp-
Kisangani region one fi nds intermediates with P. soni and P. defriesi from the Luluabourg district; and
simpsoni. between P. simpsoni and P. aquatorialis from the Ki-
sangani district.
Philantomba aequatorialis musculoides (Heller, 1913)
Darker than P. a. aequatorialis on the dorsum, the Philantomba hecki (Matschie, 1897)
flanks somewhat lighter, with this tone extending Malawi blue duiker
farther up the sides; haunch-stripe less well marked. Back light fawn-gray; flanks light red; croup medium
Nasals still shorter. Females apparently always to dark brown (not black), but with a fairly sharp tran-
hornless. sition line to the red-fawn of the haunches; legs reddish
Forest regions from E Uganda into Kenya, as far E tan; underside white, at least in the midline. Skull large
-1— as the Rift Valley. (greatest skull length 130–135 mm); nasals long.
0—
349-47558_ch01_1P.indd 268 5/31/11 7:05 PM

b o v idae 269
Malawi; Zambia E of the Luangwa Valley; N

Mozambique. Cephalophus Hamilton Smith, 1827
Intermediates between P. hecki and P. lugens occur Distinguished from Sylvicapra and Philantomba by its
at Kalambo Falls and at Kasanga. heavy build, convex skull profi le, nontubular orbits,
vertical occipital plane, and depressed horns. The
Philantomba bicolor (Gray, 1863) frontal tuft is shorter and broader-based than in Sylvi-
Zulu blue duiker capra, not forming a thick frontal mat of hair as in
Darker than P. hecki, but not as dark as P. lugens; flanks Philantomba.
normally dark rufous-orange; throat with an orange
tinge; underside broadly white. Small size (greatest Cephalophus zebra Group
skull length 119–130 mm); nasals short. Lacking certain derived characters of the other
KwaZulu-Natal, N to the Zambesi in suitable for- species-groups (their deeper preorbital fossae, greater
ested areas. convexity in the skull profile, a larger tuft, longer
horns), and with numerous specializations of its own (a
Philantomba monticola (Thunberg, 1789) very broad muzzle; dorsal flattening; extremely de-
Cape blue duiker pressed horns, forming grooves in the parietals for
Somewhat light gray or fawn; legs reddish tan; their reception; medially ridged horns; a rounded me-
haunches only slightly reddened, with very little tran- dian palatal notch; zebra-like stripes on the back; a lack
sition between the haunches and the croup; tail not of agouti banding, even when young; special hair-
very dark. Size as in P. bicolor (greatest skull length tufts on the hocks).
115–126 mm); horns longer.
E and W Cape. Cephalophus zebra Gray, 1838
banded duiker, zebra duiker
Philantomba anchietae (Bocage, 1879)
Grubb (2004a) discussed the nomenclature in some depth.
Angolan blue duiker
Back pale gray-brown; flanks gray, with a hint of red; Size small to medium; light reddish yellow, becom-
legs pale red-fawn; haunches red posteriorly, grading ing darker and browner on the neck and the middle
into the gray of the flanks anteriorly. Tail black, but of the back; 12 or 13 black stripes from behind the
this color not extending very far onto the croup; well- shoulders to the tail root, backwardly convex on
marked transition on the haunches. Underside broadly the midback, but becoming forwardly kinked (and
gray-white, this color extending halfway down the browner) on the flanks, the anterior and posterior 2–3
legs. Perhaps the largest species overall; greatest skull stripes becoming increasingly faint. Underside a paler
length 131–138 mm. version of the upperside. Legs redder, with black tufts
N Angola, from about the Cubal River to Dalla on the shanks; pasterns black. Tail with black inter-
Tando, near the DRC border. mixed above, and with white below. Face, and the
very small coronal tuft, maroon; nose with blackish
Philantomba defriesi (Rothschild, 1904) tones. Horns very depressed, forming grooves in
Zambian blue duiker the parietals for their reception; midline of the pari-
Related to P. hecki, but paler and more contrasty; dorsal etals and the posterior frontals convex; horns up-
gray zone dark gray, quite distinct, but restricted to the turned at the very tips. Skull with a very broad snout;
mid-dorsal region; flanks pale red-fawn; haunch no- forehead nearly flat; preorbital depressions relatively
ticeably reddish, as are the legs; croup and tail dark, tiny.
the haunch transition well marked; wholly white be- The extremely restricted distribution: part of Si-
low, with the white extending down the inner aspect erra Leone, much of Liberia, and part of Ivory Coast.
of the upper half of the legs, unlike P. anchietae. Size
very large; greatest skull length 126–140 mm. Cephalophus silvicultor Group
Zambia, W of the Muchinga Escarpment; Ka- Distinguished from C. zebra by its larger preorbital
tanga, N to Lukonzolwa and Kinda. fossae, somewhat more convex skull profi le; lack of
Specimens from the Parc National de l’Upemba parietal grooves; a narrower, deeper median palatal
and the surrounding area are evidently hybrids be- notch; longer and less depressed horns; and other fea-
tween P. defriesi and P. aequatorialis. tures. Distinguished from the C. ogilbyi group by its —-1
—0
349-47558_ch01_1P.indd 269 5/31/11 7:05 PM

270 Artiodactyla
untufted tail; a lack of sexual dimorphism in the skull Cameroon, from Mt. Cameroon S and E; all the
form; a less convex skull profile; a deeper median pala- forests from the Atlantic coast E through the DRC to
tal notch, more depressed and cylindrical (not flat- the Ituri Forest.
tened) horns, without much sexual dimorphism; and Apparently only sporadically found S of the Congo
weakly annulated hair. Distinguished from both C. ze- River. Machado (1969) recorded C. castaneus from N
bra and the C. ogilbyi group by its distinctive juvenile Angola.
pelage, which is dark brown overall, each hair being There is considerable geographic variation within
conspicuously agouti-patterned (with a buff y band), C. castaneus, but no clear divisions can be made. The
and showing a pronounced color change during matu- width of the dorsal stripe varies somewhat: specimens
ration. The six species assigned to this species-group from N Cameroon (Yoko, etc.) average only 45.7 mm;
show a mosaic of characters; four of them (i.e., except those from the Cameroon coast, 60.6 mm; but from all
for C. dorsalis and C. castaneus) have shortened neck hair other localities (including Mt. Cameroon), at least
covering a dermal shield, and there seems little alterna- 80 mm. The forehead, while usually pure red, may be
tive but to assume that this has been independently blackened; the proportion of specimens with black
evolved in C. jentinki and in C. silvicultor / C. spadix (as foreheads falls from W to E, from 10 out of 30 in Cam-
well as in those species of the C. ogilbyi group that sport eroon (but 3 out of 5 in the Sangha River district) to
it), since C. jentinki and C. dorsalis are otherwise so only 1 out of 9 in E and S DRC.
clearly related (see Kuhn, 1968). In the C. silvicultor Note that the type of Cephalophus castaneus arrhenii
group, the development of the nuchal dermal shield Lönnberg, 1917 (from Beni, the Semliki Valley, Zaire),
seems to be size related, or related to the absence of is not a member of this species, but instead is an ex-
much sexual size difference in the horns, or both. ample of C. leucogaster.
The relative abundance of C. castaneus falls mark-
Cephalophus dorsalis Gray, 1846 edly from W to E. The apparently localized nature of
western bay duiker its range in the DRC leads to much uncertainty as to
Skull profi le evenly convex; supraorbital foramina in its exact distribution there.
grooves; nasal bones ending in median points, without
lateral processes; mesopterygoid fossa very narrow; CephaIophus jentinki Thomas, 1982
maxillae outbowed anteriorly. Color bright chestnut Jentink’s Duiker
red in the adults; midfacial blaze and coronal tuft usu- Remarkably particolored. Body medium bluish gray
ally black, or black-suff used; size small (head plus (composed of very short white hairs, mixed in with
body length 760– 850 mm, greatest skull length 156– slightly longer dark brown ones that have light tips and
174 mm). Black dorsal stripe from the nape to the tail- bases); limbs and an extension across the shoulders
tuft, usually broadest on the withers; underside like white; head and neck black-brown. Dark bands down
the upperside, except for a black patch on the chest; the anterior surfaces of the limbs; hocks the same color;
legs brown; tail white below. Juveniles colored quite chest black in the midline. Hair coarser on the trunk,
differently (see above). Midfacial hair parting of the more silky on the foreparts. Neck very short-haired,
juveniles also generally seen in the adults. Inguinal with (apparently) a dermal shield. Lips and muzzle
pouches very deep. white, with (as in juvenile C. dorsalis) dark spots at the
W of the Adamawa highlands, W to Sierra Leone bases of the facial vibrissae. Juveniles, and apparently
and Guinea. also young adults, with rather long, lateral hoofs. Coro-
There appears to be a distributional hiatus be- nal tuft very short, black. Horns more depressed than
tween Dahomey Gap and Cross River, Nigeria. Unex- in C. silvicultor. Ears not furry inside, like C. dorsalis and
pectedly, the barrier separating C. dorsalis from C. C. castaneus.
castaneus seems to be the Adamawa highlands, not Known only from Liberia, the neighboring part
the Niger River, as one might expect. of Ivory Coast, and Freetown Peninsula, Sierra
Leone.
Cephalophus castaneus Thomas, 1892
Larger than C. dorsalis (head plus body length 883– Cephalophus silvicultor (Afzelius, 1815)
1032 mm, greatest skull length 185–202 mm); darker western yellow-backed duiker
in color; crest and face usually a deep red, not black; Largest species of duiker. Skull length >263 mm. Gen-
-1— width of the dorsal stripe usually greater. eral color some shade of gray-brown; hairs with lighter
0—
349-47558_ch01_1P.indd 270 5/31/11 7:05 PM

b o v idae 271
bases. Yellow or creamy white triangle on the back,

beginning in a point at about the level of the forelegs, Cephalophus silvicultor ruficrista Bocage, 1869
then gradually broadening backward, and ending in Skins seen: 35.
a straight line on the apex of the croup; fairly well- Size as in C. s. 1ongiceps, with merely a slightly
defi ned, sparsely haired area behind this triangle, shorter face; color lighter (light to dark brown), usu-
reaching to the root of the tail; whitish and brown ally without gray tones; dorsal triangle broader than
hairs mixed at the base of the tail; lateral to this in in most individuals of C. s. longiceps, and light in color
some (but not all) individuals, possibly some small, (white to dull gold); haunch spots always present.
triangular haunch marks, colored like the dorsal tri- S of the Congo River into Angola and Zambia.
angle, and with their anterior vertices contacting the
posterior vertices of the latter. Posterior surface of the Cephalophus curticeps Grubb & Groves, 2001
rump and the thighs nearly black. Sides of the neck eastern yellow-backed duiker
lighter than the general body color; midface dark Skins seen: 11.
brown; cheeks and muzzle whitish; groin whitish. Much smaller than C. silvicultor; greatest skull
Coronal tuft maroon, fairly well developed. No pale length <260 mm; color darker; dorsal triangle nar-
vibrissal patches; no midfacial parting at any age; no rower and darker, gold to very dark golden-brown;
elongation of the lateral hoofs. Dermal neck shield haunch spot well developed in nine skins, a trace in
extremely well developed. the other two, but never actually absent.
From the far W of Africa E to the Albertine Rift, Rwanda, Burundi, and W Kenya (Mau Forest, Keri-
and S into Zambia. cho, Molo, Masi).
The skull of C. silvicultor differs from those of C.
jentinki, C. dorsalis, and C. castaneus in having a more Cephalophus spadix True, 1890
convex frontal region, making the dorsal profi le Abbott’s duiker
more sinuous; the horns, however, are less depressed. Somewhat smaller than C. curticeps, and very similar
The supraorbital foramina are not sunk into chan- to the latter, but lacking the dorsal triangle. Very dark,
nels. The lateral nasal processes are usually conspic- blackish brown, with a hint of red; dorsal stripe vague,
uous, and may articulate with the premaxilla. The dark; underside reddish, except for the dark midline;
choanae and mesopterygoid fossa are broader. The throat and cheeks light gray; groin white; coronal tuft
rostrum is narrower, but relatively shorter, than in C. large, maroon. Skull differing from C. curticeps in its less
jentinki. The horns, equal in size in the two sexes, are convex frontals, shorter rostrum, very reduced auditory
often quite noticeably downcurved. Generally, me- bullae, narrower (V-shaped) median palatal notch, and
dian pillars develop on the buccal sides of the somewhat narrower mesopterygoid fossa. Juvenile pel-
cheekteeth. age lacking the conspicuous dorsal stripe of C. silvicultor
and C. curticeps, despite the fact that what trace of it
Cephalophus silvicultor silvicultor there is remains in the adults. Occasionally with rudi-
(Afzelius, 1815) ments, or precursors, of the C. silvicultor / C. curticeps
Very large; color medium brown, with grayish tones; dorsal markings: traces of white at the base of the tail, or
dorsal triangle very broad, creamy to ochre-yellow, an elongate white spot on either side of the tail base.
with no haunch spot. Tanzania, from Kilimanjaro and the Usambaras S
Sierra Leone, Liberia, Ghana, Togo, and Lagos (W through the Ulugurus, Uzungwas, and S highlands
Nigeria). to the N tip of Lake Malawi (Poroto Mountains,
Mt. Rungwe, Mfrika Scarp).
Cephalophus silvicultor longiceps Gray, 1865
Skins seen: 72. Cephalophus ogilbyi Group
Somewhat smaller; color tends to be darker, to Distinguished from the C. zebra and the C. silvicultor
blackish brown; dorsal triangle considerably nar- groups by a greater degree of sexual dimorphism, in-
rower, creamy to deep dull gold; haunch spot absent cluding longer, thicker horns and a thickening of the
in 44 skins, a trace in 20, well developed in 8. bone in the frontal region in the males; the mesop-
From the Cross River district to Cameroon, Gabon, terygoid fossa broader, with the median palatal notch
the Congo River mouth, and the Congo Republic E to usually at about the same level as the lateral ones, or
the Uele and Ituri districts and into S Sudan. slightly anterior to them; the horns, and their cores, —-1
—0
349-47558_ch01_1P.indd 271 5/31/11 7:05 PM

272 Artiodactyla
mediolaterally flattened and conical; the horn sheaths legs colored like the body; foreleg with a dark line
striated, with massive transverse ridges; the adult and from the carpus to the hoof; neck somewhat darker,
the subadult growth stages always clearly distinct. The grayer, with frequent hair reversal dorsally. Juveniles
tail is tufted, sometimes very strongly so; the color pat- speckled. Tail-tuft very large. Median nuchal hair-
tern of the juveniles is similar to that of the adults, stream (where it occurs) reversed over a wide median
though often darker, and lacks the conspicuous speck- zone. Juveniles less strongly speckled. Skull with an
ling of the C. silvicultor group. extreme inflation of the forehead behind the nasofron-
The species in the C. ogilbyi group are very diverse. tal suture, making a marked frontal boss; frontal boss
Some have a dermal neck shield and shortened neck reduced laterally, so the supraorbital foramina not
hair, presumably in parallelism with members of the roofed over; zygomatic arch more curved than in C.
C. silvicultor group; these C. ogilbyi–group species also niger, flaring out in the middle; superior temporal line
have a deep and sharply margined preorbital fossa, and a less developed; rostrum not swollen outward; preor-
heavily developed, bipartite tail-tuft. The other species bital fossa smaller; nasals shorter, making a shorter su-
in the C. ogilbyi group lack the above characters; in- ture with the premaxillae.
stead, they have developed an interfrontal groove, and Bioko and the opposite mainland on the Nigeria/
the adults have lost the juveniles’ agouti pattern of hairs. Cameroon border (Cross River district; Rumpi Hills).
Insular specimens tend to have an even more ex-
Cephalophus niger Gray, 1846 treme frontal boss than the mainland ones.
black duiker
Color dark blackish mahogany-brown; individual Cephalophus crusalbum Grubb, 1978
hairs with a straw-colored base, the rest of the hair white-legged duiker
black; midback zone darker, due to the dark brown Smaller; legs white from the knee and the hock to the
bases of the hairs, this band extending from the with- hoof; dorsal stripe broader (25– 60 mm); reversal of
ers to the rump, including the upper part of the but- the hair-stream on the neck (where it occurs) over a
tocks. Yellow stripe down the inner side of the upper much narrower zone. Juveniles strongly speckled.
part of the forelimbs. Forehead, nose, and coronal tuft Skull with a less exaggerated frontal boss.
reddish mahogany; coronal tuft with much black in the Gabon, from the Fernan Vaz district S, perhaps
center. Sides of the face and the underside of the neck pale even to Mayombe, and inland to the Forêt des Abeilles,
gray-brown. Tail black, with a broad, light gray, C Gabon, where it is sympatric with C. callipygus.
paintbrush-shaped tuft. Juveniles similar in color, but Grubb (1978b) described C. crusalbum as a subspe-
with less red about the face, and paler below. Neck hair cies of C. ogilbyi, comparing it not only with other re-
very short in the adults, with a thickened skin; no reversal puted taxa of that species, but also with C. leucogaster,
of the hair tracts. Preorbital pits not very deep, but extend- to which specimens of it had been ascribed in the
ing forward to the premaxillae, and downward to the past.
molar alveoli; well-defined groove between the frontals;
frontals thickened and convex, but not excessively so. Su- Cephalophus brookei Thomas, 1893
praorbital foramina often partly roofed over by the lateral Brooke’s duiker
extension of the frontal thickening. Zygomatic arch Very similar to C. ogilbyi, but of a paler, dull golden
straight, not flared in the middle. Superior temporal line color, becoming red-gold on the dorsum, with a
well marked, extending out (in the adults) into a ridge be- broader (27– 66 mm) black line down the back, indis-
tween the horn core and the postorbital bar. tinct in front of the shoulders and narrowing to a thin
Sierra Leone E to the Niger River. line on the rump; face and coronal tuft red-gold; un-
Jansen van Vuuren & Robinson (2001) found that derside pale yellow; legs colored like the body, but
C. niger is sister to the rest of the C. ogilbyi group in the inner side of the forelegs with whitish hairs. Neck
their cytochrome b tree, but is more closely associ- with extremely short hair, reversed (in most speci-
ated (although with low bootstrap support) with the mens) over a fairly wide zone, starting from a whorl
C. silvicultor group in their 12 S tree. on the withers. Juveniles strongly speckled. Skull
with the forehead not so strongly inflated; muzzle
Cephalophus ogilbyi Waterhouse, 1838 bowed outward on either side; preorbital fossa small,
Ogilby’s duiker shallow (restricted by the bowing of the rostrum);
-1— Large; bright ochery color; dorsal stripe dark, nar- premaxillae widely suturing with the relatively very
0— row, only 11–30 mm broad, extending onto the tail; long nasals.
349-47558_ch01_1P.indd 272 5/31/11 7:05 PM

b o v idae 273
Known only from a few localities in Ghana, W of dered, ending on the root of the tail; size intermedi-
the Volta; Liberia; and one locality in Guinea. ate between C. johnstoni and C. weynsi.
Forests of the Nile– Congo divide, Rwanda.
Cephalophus callipygus Peters, 1876
Peters’s duiker Cephalophus adersi Thomas, 1918
Color as in C. ogilbyi, but becoming bright red on the Aders’s duiker
loins; face and crest rich reddish; underside yellow- Color very light ochery, becoming more reddish on
ish, but the chin and the throat white; dorsal stripe the nape. Underside white; white stripe across the up-
distinct, from the withers to the tail-tuft, becoming per thigh, joining the white of the groin; lower seg-
very wide on the croup and including the buttocks ments of the limbs marked with white spots. Coronal
and the hind surface of the hindlimbs, as far down tuft dark chestnut, as is the face. Hair reversed on the
as the hocks. Forelegs diff usely darkened down the midline of the nape; neck hair not shortened. Light
front; hindlimbs black below the hocks. Hair in the bases of the hairs very long; hairs very soft. Size very
midline of the neck reversed over a narrow zone in small; nasals broadening distally; nasal notch coincid-
about half the specimens. Midline of the chest and ing laterally with the upper margin of the preorbital
the belly dark. Inguinal pouches less clearly present fossa. Other skull characters resembling those of Ceph-
than in other members of the C. ogilbyi group. Skull alophus leucogaster, C. arrhenii, and C. rufiIatus (see be-
resembles that of C. ogilbyi, with a strongly developed low): deep preorbital fossae, with a fairly sharply de-
frontal boss. fined upper margin (though not as marked as in other
Forests of Cameroon, S and E of the Adamawa three species); no interfrontal groove; no marked fron-
highlands, Gabon (except the coastal district), Rio tal convexity.
Muni, the Congo Republic, and the CAR. Zanzibar, and the Arabuko-Sokoke Forest in Kenya.
C. adersi is a species of uncertain and isolated af-
CephaIophus weynsi Thomas, 1901 fi nities; it may even be distantly related to C. zebra,
Weyns’s duiker rather than to the present species-group ( Jansen van
Color chestnut-brown; midback region darker and Vuuren & Robinson, 2001).
redder than the flanks, forming an ill-defi ned dorsal
stripe; darkened zone on the withers; underside pale Cephalophus leucogaster Gray, 1873
ochery; groin white. Coronal tuft red. Limbs dark western white-bellied duiker
brown-gray in color, this color often reaching as Reddish ochery color, becoming darker and redder on
high as the shoulders and the stifle. Tail with a black the back; fairly thick, well-defi ned, black dorsal stripe,
line down the upper surface. Hair-reversal zone on usually complete from the occiput to the tail-tuft,
the neck narrow. Inguinal pouches well developed. the maximum breadth 69 mm, but averaging 36–
Skull resembles C. callipygus, but without a conspic- 47 mm in different regions. Face, in the midline, col-
uous boss. ored like the dorsum (or blacker); crest a mixed black
DRC, E to the C African Rift Valley, and to the and dark red. Legs dark gray. Underside sharply
Imatong Mountains, Sudan. There is an isolated pop- white, as are the hind surfaces of the buttocks. Tail-
ulation on Mt. Kabobo in the SE part of the DRC. tuft white, mixed with black, and very large. Hair
short and fi ne, not shortened on the neck. Skull as in
Cephalophus johnstoni Thomas, 1901 C. adersi, but with a more sharply defi ned upper rim
Johnston’s duiker to the preorbital fossa, a shorter rostrum, and a dis-
Differs from C. weynsi in its much smaller size; skull tinct frontal boss. Horns very short in both sexes.
length measurements do not even overlap (Groves & From the Cameroon/Gabon/Congo region, S and
Grubb, 1974). W of the Adamawa highlands and W of the Congo-
Forests from the Rutshuru district through Oubangui river system.
Uganda to the W side of the Kenya Rift Valley.
Cephalophus arrhenii Lönnberg, 1917
Cephalophus lestradei Groves & Grubb, 1974 eastern white-bellied duiker
Lestrade’s duiker Color paler, browner. Dorsal stripe broader and more
Resembles C. weynsi and C. johnstoni, but very dark in diff use, its minimum breadth 67 mm, but averaging
color, a dark olive-gray-brown, with varying reddish 109 mm. Females, on average, slightly larger than the —-1
tones; dorsal stripe well marked, but not sharply bor- males. —0
349-47558_ch01_1P.indd 273 5/31/11 7:05 PM

274 Artiodactyla
NE of the DRC, from Bondo to the upper Ituri On Mt. Kenya and in the Aberdares, it has not
region. been recorded above 1800 m.
Cephalophus rufiIatus Gray, 1846 Cephalophus nigrifrons Gray, 1871

red-flanked duiker black-fronted duiker
Light reddish, with a striking, broad gray dorsal Very close to C. harveyi, but distinguished by several
stripe and facial blaze; legs gray. Underside and the characters: its narrower skull, with the muzzle pinched
sides of the face ochery. Coronal tuft black. Rostrum in, the nasals obscuring the maxillae in dorsal view;
long, narrow; frontals evenly convex as far forward the much deeper preorbital fossae; the outbowed pre-
as the proximal nasals, and as far back as the bases of maxillae; the more posteriorly restricted frontal boss;
the horns; horn bases raised up; supraorbital foram- the less frequent nasopremaxillary contact; the elon-
ina reduced to a single pair. Very small species. Males gated hoofs; and the thicker, coarser hair, especially
larger than the females (unusual among duikers), long on the withers. Some populations with noticeable
with a more specialized skull. inguinal pouches, lacking in C. natalensis or C. harveyi.
Guinea and Gambia, along the forest-edge zone as Skull length >169 mm.
far E as Yambio in Sudan and Arua in the West Nile Rainforest, including montane forest, from the
district, Uganda. Cameroon/Gabon region across the DRC to the Great
A forest-edge or bush-and-thicket species, not Lakes region and SW Uganda.
found in deep forest. An isolated population in the Niger delta, 500 km
from any other known occurrence, is indistinguish-
Cephalophus nataIensis A. Smith, 1834 able from C. nigrifrons (Powell & Grubb, 2002).
Natal red duiker
Pale orange-ochre to reddish chestnut; legs slightly Cephalophus nigrifrons nigrifrons Gray, 1871
grayer, or not at all darker than the body; midline of Shining chestnut brown; forelegs black as far up as
the face somewhat darker, grayer. Coronal tuft a the elbow; hindlegs variable, but generally black up to
deep red, mixed with black. Neck washed with gray, the hock. Chest broadly black. Noticeable black sprin-
but the hair not shortened. Ears gray-black for most kling on the withers, sometimes light or sometimes
of their length. Tail thin, with a big tassel at the tip; very heavy. Chin yellow. Forehead generally black, the
tail red basally, black distally, and white below. Mid- black continuing onto the facial blaze, but sometimes
back zone darker, more chestnut-colored than the becoming chestnut.
flanks. Underparts light ochery, tending to white in SE Nigeria S to Mayombe, E across the Congo
the groin, the axillae, and the interramal region. Size River to the upper Ituri district, S to the Franz Joseph
small; skull length 153–169 mm; skull broad; frontal and Guillaume falls, SE to Kinda (Shaba).
convexity not strongly developed, and not involving Skins from the Mt. Tshiaberimu area (Butembo,
the horn pedicels. Moera, etc.) are coarser-haired, with the chin tending
Coastal galley and montane forests, from Durban, to be white, rather than yellow, and the forelimbs very
via Hluhluwe and SE Transvaal, Mozambique, and black, thus tending toward C. n. kivuensis.
the Shire Valley (Malawi), to the coast of S Tanzania,
as far N as Kilwa. Cephalophus nigrifrons kivuensis Lönnberg, 1919
Distinguished by its generally more contrasting col-
Cephalophus harveyi Thomas, 1893 oration: the whole of the limbs gray-black; the tail
East African red duiker with much black at the base; the hairs, especially in
Brighter, deeper in color than C. natalensis, a reddish the dorsal region, with a longer, darker (gray-black)
chestnut to ochery brown; legs dark gray to brown- basal zone; and the facial blaze very black and broad.
ish black; midfacial zone blackish, usually right up to Hair not especially long, but thick and coarse. Chin
the coronal tuft. Crest with more black in it. Size, on pale reddish, but varying toward whitish in speci-
average, larger; skull length 162–184 mm. mens from the highest altitudes. Smaller than C. n.
From C Tanzania (Dar-es-Salaam, Mpwapwa, Kon- nigrifrons.
doa) N, via the montane forests of Tanzania (Usam- Virunga volcanoes and the surrounding area in
baras, Kilimanjaro, Meru) and Kenya (Mt. Kenya, Ab- Rwarda, Burundi (as far S as Bururi), Uganda (Shonga
-1— erdares, E Rift Scarp to 1800 m) and the coastal forests, Hill, Rukiga) and the DRC (Masisi, Ngoma, Mt.
0— as far as the Juba River, Somalia. Kahuzi).
349-47558_ch01_1P.indd 274 5/31/11 7:06 PM

b o v idae 275
Cephalophus hypoxanthus Grubb & Groves, 2001

Itombwe duiker Tribe Oreotragini Pocock, 1910
Distinguished from C. nigrifrons by its paler, light yel- This tribe is notable for the very upright horn inser-
lowish chestnut color; the limbs hardly darkened, gray- tions, the absence of lateral prongs on the nasal tips,
brown at most, this color not extending to the elbow and the very short metapodials. It was perhaps unex-
on the forelimbs, and, on the hindlimbs, forming (at pected that molecular studies should reveal that its
most) an infusion on the shanks. Chin white. Hair closest affi nities are with the Cephalophini; the rela-
fairly long, but soft, the hair being rather coarse in tionship is not close, however, and klipspringers are
highland populations of C. nigrifrons. phylogenetically isolated.
Appears to be restricted to the Itombwe Moun-
tains, W of the N end of Lake Tanganyika. Oreotragus A. Smith, 1834
klipspringers
Cephalophus rubidus Thomas, 1901 As might be expected from their ecology, which is
Rwenzori duiker restricted to rocky outcrops, the distribution of klip-
A very beautiful duiker. Hair long, thick, deep rufous springers is very spotty, and there are quite distinc-
in color, lacking any black sprinkling; chin white; tive species in different parts of the range. Some
limbs very black. Tail bushy. klipspringers are distinguished by different pelage
Confi ned to the upper forested slopes of the Rwen- features, but in many cases the pelage is hardly differ-
zoris; recorded from 1300 to 4200 m. ent in widely separated species, and the degree and
Jansen van Vuuren & Robinson (2001) found C. type of sexual dimorphism, as well as skull and tooth
rubidus to be part of the C. ogilbyi clade; this is such a sizes, are more diagnostic.
surprising result that it must surely be questioned as The degree and nature of sexual dimorphism in
to whether the specimen from which they obtained the different species appear to correlate with aspects
their sample really was C. rubidus. of ecology and social behavior, particularly including
the size of the pair territories, as listed by S. Craig
Cephalophus fosteri St. Leger, 1934 Roberts (1996).
Mt. Elgon forest duiker
Very small in size; skull length 153–166 mm. Brown- Oreotragus oreotragus (Zimmermann, 1783)
ish; face and sides of the neck reddish, not gray; chin Cape klipspringer
white, this color extending to the interramal zone. Uniform yellow, speckled with brown; hairs mainly
Hair bases pinkish white. Coronal tuft very short, white, but some dark brown; tips yellow. Underside
completely black. Fur thick and coarse; tail very bushy. pale (with white-tipped hairs) only on the midline;
Forests of Mt. Elgon, at 2400–3400 m. chin and throat light yellowish. Large, dark brown
patch above each hoof. Ears relatively short, whitish,
Cephalophus hooki St. Leger, 1934 with a thick black line on the rim. Forehead and oc-
Mt. Kenya forest duiker ciput reddish brown. Sexes the same size.
Very small, though slightly larger than C. fosteri; skull Horns particularly short. Males the largest among
length 163 mm (N = 1). Dull chestnut or reddish gray; the klipspringers; females (because of the usual reverse
face and sides of the neck grayer; chin reddish white, sexual dimorphism in the genus) only middling sized.
this color often (but not always) extending into the S coastal regions and W Cape.
interramal region; hair bases pinkish gray; subtermi-
nal bands dark brown, not quite black. Facial blaze Oreotragus transvaalensis Roberts, 1917
bordered by a red stripe. Coronal tuft short, com- Transvaal klipspringer
pletely black. Fur thick, coarse; tail bushy, with an Bright golden-yellow, with strongly contrasting white
enormous tuft. underparts. Feet above the hoofs broadly brown. Ears
Highland forests of Mt. Kenya and the Aberdares, short, 80– 84 mm.
above 2400 m. Horns particularly long. Males medium sized; fe-
The forehoof is 28– 44 mm long (mean 34 mm, males tending to be large.
N = 8), resembling that of C. fosteri and C. nigrifrons KwaZulu Natal to the E and W Transvaal.
(28–37 mm, mean 33 mm, N = 10); while in C. harveyi, Populations of this species in the Drakensberg, at
a rather larger animal, it measures 21–29 mm (mean Gamka and at Springbok, had territory sizes of 10, 15, —-1
26 mm, N = 18). and 49 ha respectively (S. C. Roberts, 1996). There —0
349-47558_ch01_1P.indd 275 5/31/11 7:06 PM

276 Artiodactyla
was a territory size of 21.5 ha in the Sentinel district O. aceratos is very similar to O. centralis, from
in Zimbabwe, just N of the Limpopo, for a population which it differs in the smaller size of the females, and
that may also belong to O. transvaalensis. the (on average) slightly larger teeth.
Oreotragus stevensoni Roberts, 1946 Oreotragus schillingsi Neumann, 1902

Stevenson’s klipspringer Maasai klipspringer
Much duller in color than O. transvaalensis; head, espe- Generally very similar to O. saltatrixoides; very rich
cially, much darker. Extensive white underparts, simi- orangey color, with contrastingly gray legs, dark pas-
lar to O. transvaalensis. Marks above the hoofs variable. terns, and a darker midback. A rare variant, occurring
Horns particularly long. Sexes apparently the in some areas, whitish / lemon yellow, especially on
same size; one of the smallest forms in both sexes. the foreparts. Thighs clear gray or rufous, differing
W Zimbabwe. markedly in color from the body.
According to S. C. Roberts (1996), a population Uniquely among klipspringers, horns present in
of this species at Metobo had a mean territory size the females; as Roosevelt & Heller (1914) state, females
of 5.5 ha. are “as well horned as males.” Females much larger
than the males; overall, the males very small, but the
Oreotragus tyleri Hinton, 1921 females the largest in the genus, equal to those in O.
Angolan klipspringer centralis.
S. C. Roberts (1996) described the distribution of
1924 Oreotragus oreotragus cunenensis Zukowsky.
this species as being in Kenya, S of Mt. Kenya; in
1924 Oreotragus oreotragus steinhardti Zukowsky.
Uganda, as far W as the Dodoth Hills; and in Tanza-
Wholly pale sandy ochre (dark yellow, speckled with nia, S to Rukwa.
brown) above; contrast occurring only between the S. C. Roberts also noted that the territories of O.
speckled dorsum and the uniform color of the legs schillingsi, studied both in Tsavo and in Gilgil, average
(sometimes gray, sometimes not). Underparts conspic- less than 2 ha, whereas in all other Oreotragus species
uously, broadly, uniformly white. Conspicuously dark they are much greater than this. As he argued, this
brown above the hoofs. Ears yellow, the insides with lends support to the female-competition hypothesis.
yellowish white hairs. Differs from the S and E forms
in having longer ears and hindfeet. Kenya Highlands Form
Females larger than the males; overall, medium Duller color than the Ankole form; darker, not yellow.
sized in both sexes. Females apparently smaller than males (but we
The Kaokoveld and S Angola. have seen only one female!). This size feature is per-
A population of O. tyleri has the largest known ter- haps identical to the Ankole form (below).
ritories: 100 ha (S. C. Roberts, 1996).
Oreotragus aureus Heller, 1913
Oreotragus centralis Hinton, 1921 golden klipspringer
Zambian klipspringer General color golden-yellow, lighter than O. saltatrix-
Deep rufous to more yellow-gray, legs contrastingly oides. Crown of the head rufous, in marked contrast
gray. No black above the forehoofs. Some of the more to the body color. Color difference between the body
orange individuals very like O. aceratos; others with and the legs very slight, but the rufous crown well
quite a dazzling yellow-and-black effect. marked. Hoof-spot large.
Females larger than the males; overall, the males Males larger than the females (unusual in Oreotra-
medium sized, the females fairly large. gus); overall, both sexes fairly small.
Zambia, and probably extending into neighboring N Guaso Nyiro.
countries.
Ankole Form
Oreotragus aceratos Noack, 1899 More speckled, but gold-orange, not— as in O.
southern Tanzanian klipspringer saltatrixoides— drab gray. Neck, shoulders, and head
Forequarters ochery, only the hindquarters olive, bay-yellow; snout a contrasting gray. Golden, espe-
with the two zones quite well separated; legs gray. cially in the midneck and slightly down the midline.
-1— Sexes apparently the same size. Some variability in color (e.g., in the contrast of the
0— S Tanzania. yellow parts); one specimen with a rich brown crown.
349-47558_ch01_1P.indd 276 5/31/11 7:06 PM

Table 6 9 Univariate statistics for Oreotragus Table 6 9
Skull l Skull br Teeth Horn l Skull l Skull br Teeth Horn l
Males W Zambia
porteousi Mean 141.000 81.000 51.000 107.000
Mean 135.000 78.250 49.500 99.000 N 1 1 1 1
N 4 4 4 2 stevensoni
Std dev 4.0825 3.3040 2.6458 — Mean 135.900 80.700 47.700 104.625
Min 131.0 74.0 47.0 89.0 N 5 5 5 4
Max 139.0 82.0 53.0 109.0 Std dev 2.5100 1.7176 2.8636 11.8980
Bogos / N Sudan Min 134.0 79.0 43.0 88.0
Mean 130.000 78.000 46.000 95.000 Max 140.0 83.0 50.0 115.0
N 1 1 1 1 tyleri
saltatrixoides Mean 139.562 80.591 49.955 77.550
Mean 137.438 82.250 47.500 84.286 N 8 11 11 10
N 8 8 8 7 Std dev 3.8493 4.1402 2.7790 28.4590
Std dev 1.7816 2.3452 2.6049 9.1645 Min 133.0 70.0 45.0 —
Min 135.5 78.5 44.0 72.5 Max 145.0 85.0 53.0 104.0
Max 141.0 85.0 51.0 100.0 transvaalensis
Ankole Mean 142.667 84.000 49.667 109.167
Mean 141.500 79.000 51.500 91.500 N 3 3 3 3
N 2 2 2 2 Std dev 4.5092 4.3589 1.2583 20.0271
Min 141.0 79.0 51.0 81.0 Min 138.0 79.0 48.5 87.5
Max 142.0 79.0 52.0 102.0 Max 147.0 87.0 51.0 127.0
NW Kenya oreotragus
Mean 137.750 77.500 49.900 93.100 Mean 147.000 85.000 50.333 83.667
N 4 5 5 5 N 2 3 3 3
Std dev 2.2174 4.4721 2.0433 9.8133 Std dev — 4.5826 2.0817 8.0829
Min 135.0 71.0 47.0 85.0 Min 143.0 81.0 48.0 75.0
Max 140.0 83.0 52.5 109.0 Max 151.0 90.0 52.0 91.0
aureus Females
Mean 139.714 79.286 48.643 91.714 Bogos / N Sudan
N 7 7 7 7 Mean 131.750 78.000 45.000 —
Std dev 2.9560 2.5797 .8997 5.8228 N 2 2 2 —
Min 134.5 75.0 47.5 83.0 Min 131.5 76.0 41.0 —
Max 144.0 82.5 50.0 99.0 Max 132.0 80.0 49.0 —
Kenya highlands somalicus
Mean 141.265 81.222 49.763 90.136 Mean 143.333 79.333 48.667 —
N 17 18 19 14 N 3 3 3 —
Std dev 3.6406 3.1165 2.6372 12.9468 Std dev 2.5166 1.5275 2.5166 —
Min 132.5 76.5 45.0 67.0 Min 141.0 78.0 46.0 —
Max 147.0 86.0 57.0 112.5 Max 146.0 81.0 51.0 —
schillingsi saltatrixoides
Mean 136.750 77.500 51.000 84.625 Mean 138.300 82.200 48.900 —
N 4 4 6 4 N 5 5 5 —
Std dev 1.7078 2.5166 1.6733 1.8875 Std dev 2.4393 1.8908 1.2450 —
Min 135.0 75.0 49.0 82.5 Min 135.5 80.5 47.0 —
Max 139.0 81.0 53.0 87.0 Max 142.0 85.0 50.0 —
aceratos Ankole
Mean 141.000 82.000 53.000 92.000 Mean 145.333 79.667 50.667 —
N 1 1 1 1 N 3 3 3 —
centralis Std dev 3.7859 4.9329 2.8868 —
Mean 140.875 79.063 50.312 97.875 Min 141.0 74.0 49.0 —
N 16 16 16 16 Max 148.0 83.0 54.0 —
Std dev 5.0050 3.1931 2.3301 16.4879 NW Kenya
Min 132.0 71.0 46.0 74.0 Mean 134.000 78.000 52.000 —
Max 154.0 83.0 54.0 125.0 N 1 1 1 —
(continued)
—-1
—0
349-47558_ch01_1P.indd 277 5/31/11 7:06 PM

278 Artiodactyla
Table 6 9 Females much larger than the males (but specimens

of only two males, and three females, available from
Skull l Skull br Teeth Horn l this region); overall, both sexes fairly large in size.
Apart from the size difference, both this and the Kenya
aureus
Mean 137.286 78.357 49.000 —
highlands form could be synonymous with O. aureus.
N 7 7 7 —
Std dev 2.2704 1.6511 1.5275 — Oreotragus saltatrixoides (Temminck, 1853)
Min 134.0 76.0 47.0 — Ethiopian klipspringer
Max 141.0 80.0 51.0 — Golden color, especially deeper on the neck; speck-
Kenya highlands ling less extensive on the rump.
Mean 133.000 82.500 46.000 —
N 1 1 1 —
Females somewhat larger than the males; overall,
schillingsi
both sexes fairly small. Teeth very small.
Mean 146.667 80.625 52.000 81.00 Ethiopia.
N 3 4 4 4 A population of this species at Sankeber had a
Std dev .5774 3.0923 1.4142 11.460 mean territory size of 8.1 ha (S. C. Roberts, 1996).
Min 146.0 76.0 51.0 67
Max 147.0 82.5 54.0 95
Bogos / N Sudan Form
aceratos
Mean 141.667 76.000 49.333 —
General color uniform grayish olive. Base of the hairs
N 3 3 3 — whitish, especially on the back. Underparts white.
Std dev 2.0817 1.0000 2.0817 — Black patch on front of the feet, above the hoofs. Thighs
Min 140.0 75.0 47.0 — only a little lighter than the back. White spot on the
Max 144.0 77.0 51.0 — outer side of the ear.
centralis Particularly long horns. Both sexes extremely
Mean 146.714 79.000 50.143 —
N 7 7 7 -
small, the smallest of the genus. Teeth very small.
Std dev 3.1997 1.7321 4.0591 — Animals from Bogos and Sudan are most similar
Min 141.0 77.0 46.0 — to O. somalicus, but they differ in such features as size
Max 151.0 81.0 55.0 — and horn length.
W Zambia
Mean 146.000 82.000 52.000 — Oreotragus somalicus Neumann, 1902
N 1 1 1 —
Somali klipspringer
stevensoni
Mean 136.000 79.000 48.500 —
An unusually homogeneous species, the coloration in
N 1 1 1 — almost all individuals being very much the same: yel-
tyleri lowy olive; brown on the crown; legs contrastingly
Mean 144.313 80.813 50.889 — gray in front; underparts white; but individuals vary
N 8 8 9 — in having from very little black to almost all black
Std dev 3.5349 1.4126 1.9808 — above the hoofs. Hardly any differentiation of the
Min 139.0 79.0 47.0 —
Max 151.0 83.0 54.0 —
dorsal pelage from the front to the back. Bases of the
transvaalensis
hairs, especially on the back, reddish gray or reddish
Mean 149.500 82.500 50.000 — brown.
N 2 2 2 — Females medium sized (no male skulls available).
Min 142.0 81.0 48.0 — Teeth small.
Max 157.0 84.0 52.0 — N Somalia.
oreotragus
Mean 143.125 86.875 48.833 —
N 4 4 3 —
Oreotragus porteousi Lydekker, 1911
Std dev 5.2341 4.2106 1.4434 — 1921 Oreotragus oreotragus hyatti Hinton.
Min 138.0 82.0 48.0 —
Max 149.0 91.5 50.5 — Color varying; body color dull yellowy; speckling
down the midline; crown and upper muzzle more
golden; ears gray; upper forelegs paler; lower forelegs
gray. Closely resembling O. saltatrixoides and O. schil-
-1— lingsi in pelage; variation within each of these species
0— greater than between them.
349-47558_ch01_1P.indd 278 5/31/11 7:06 PM

b o v idae 279
Distinguished by the particularly long horns. sexual dimorphism of O. schillingsi (despite the small
Males rather small in size (no female skulls available). sample sizes of the latter, the figures for the two
Small teeth. sexes are so extremely different that this conclusion
Only found on the Jos Plateau, N Nigeria. is quite evidently justified).
We have no information about the reputed popu- Relative horn length is horn length as a per-
lation in the CAR. centage of skull length. Thus, particularly long
horns characterize O. porteousi, apparently the Bo-
The large size of Oreotragus transvaalensis, and, to gos/Sudan form (N = 1), O. stevensoni, and O. trans-
some extent, of O. oreotragus, as well as the small size vaalensis. Particularly short horns characterize O.
of O. porteousi, are all noteworthy, as is the extreme oreotragus.
—-1
—0
349-47558_ch01_1P.indd 279 5/31/11 7:06 PM

-1—
0—
349-47558_ch01_1P.indd 280 5/31/11 7:06 PM

References
Abril, V.V., J.A. Sarria-Perea, D.S.F. Vargas-Munar & Territory. Proceedings of the Boston Society of Natural
J.M.B. Duarte. 2010. Chromosome evolution. In J.M.B. History 38:4l3– 441.
Duarte & S. González (eds.), Neotropical Cervidology: Allen, J.A. 1913. Ontogenetic and other variations in
Biology and Medicine of Latin American Deer. musk-oxen, with a systematic review of the musk-ox
Jaboticabal, Brazil and Gland, Switzerland: Funep, in group, recent and extinct. Memoirs of the American
collaboration with IUCN [International Union for Museum of Natural History (n.s.) 1:103–226.
Conservation of Nature], 18–26. Allen, J.A. 1915. American deer of the genus Mazama.
Abril, V.V., A. Vogliotti, D.M. Varela, J.M.B. Duarte & Bulletin of the American Museum of Natural History
J.L. Cartes. 2010. Brazilian dwarf brocket deer Mazama 34:521–553.
nana (Hensel 1872). In J.M.B. Duarte & S. González Almaça, C. 1992. Notes on Capra pyrenaica lusitanica
(eds.), Neotropical Cervidology: Biology and Medicine Schlegel, 1872. Mammalia 56:121–123.
of Latin American Deer. Jaboticabal, Brazil and Gland, Alpers, D.L., B.J. van Vuuren, P. Arctander & T.J.
Switzerland: Funep, in collaboration with IUCN, Robinson. 2004. Population genetics of the roan
160–165. antelope (Hippotragus equinus) with suggestions for
Ackermann, R.R, J. Rogers & J.M. Cheverud. 2006. conservation. Molecular Ecology 13:1771–1784.
Identifying the morphological signatures of hybrid- Alves, E., C. Óvilo, M.C. Rodríguez & L. Silió. 2003.
ization in primate and human evolution. Journal of Mitochondrial DNA sequence variation and
Human Evolution 51:632– 645. phylogenetic relationships among Iberian pigs and
Adlerberg, G. 1932. Critical review of the genera Nemorhae- other domestic and wild pig populations. Animal
dus H. Smith and Capricornis Ogilby. Izvestiya Genetics 34:319–324.
Akademii Nauk SSSR, Otdeleniye Matematicheskikh i Amato, G., D. Wharton, Z.-Z. Zainuddin & J.R. Powell.
Estestvennykh Nauk (7) 1932:259–285. 1995. Assessment of conservation units for the
Alados, C.L. 1987. A cladistic approach to the taxonomy of Sumatran rhinoceros (Dicerorhinus sumatrensis). Zoo
the dorcas gazelles. Israel’s Journal of Zoology Biology 14:395– 402.
34:33– 49. Amato, G., M.G. Egan & A. Rabinowitz. 1999. A new
Allard, M.W., M.M. Miyamoto, L. Jarecki, F. Kraus & species of muntjac, Muntiacus putaoensis (Artiodactyla:
M.R. Tennant. 1992. DNA systematics and evolution Cervidae) from northern Myanmar. Animal Conserva-
of the artiodactyl family Bovidae. PNAS: Proceedings tion 2:1–7.
of the National Academy of Science USA Amato, G., M.G. Egan & G.B. Schaller. 2000. Mito-
89:3972–3976. chondrial DNA variation in muntjac: evidence for
Allen, G.M. 1930. Bovidae from the Asiatic expeditions. discovery, rediscovery, and phylogenetic relationships.
American Museum Novitates 410:1–11. In E.S. Vrba & G.B. Schaller (eds.), Antelopes, Deer,
Allen, G.M. 1939. A checklist of African mammals. and Relatives. New Haven: Yale University Press,
Bulletin of the Museum of Comparative Zoology 285–295.
83:1–763. Amin, R., K. Thomas, R.H. Emslie, T.J. Foose & N. van
Allen, G.M. 1940. The Mammals of China and Mongolia, Strien. 2006. An overview of the conservation status of
vol. 2. New York: American Museum of Natural and threats to rhinoceros species in the wild. Inter-
History. national Zoo Yearbook 40:96–117.
Allen, G.M. & A. Loveridge. 1927. Mammals from the Ansell, W.F.H. 1971. Artiodactyla (excluding the genus —-1
Uluguru and Usambara Mountains, Tanganyika Gazella). Part 15 in J. Meester & H.W. Setzer (eds.), The
—0
349-47558_ch01_1P.indd 281 5/31/11 7:06 PM

282 ref e r e n c e s
Mammals of Africa: An Identification Manual. aegagrus cretica). Journal of Zoology, London

Washington, DC: Smithsonian Institution Press, 184. 256:369–377.
Arctander, P., P.W. Kat, R.A. Aman & H.R. Siegismund. Barrio, J. 2010. Taruka Hippocamelus antisensis (d’Orbigny
1996. Extreme genetic differences among populations 1834). In J.M.B. Duarte & S. González (eds.), Neo-
of Gazella granti, Grant’s gazelle, in Kenya. Heredity tropical Cervidology: Biology and Medicine of Latin
76:465– 475. American Deer. Jaboticabal, Brazil and Gland,
Arctander, P., C. Johansen & M.-A. Coutellec-Vreto. 1999. Switzerland: Funep, in collaboration with IUCN,
Phylogeography of three closely related African bovids 77– 88.
(tribe Alcelaphini). Molecular Biology and Evolution Baryshnikov, J.F.& A. Tikhonov. 1994. Notes on skulls
16:1724–1739. of Pleistocene saiga of northern Eurasia. Historical
Asher, R.J. & K.M. Helgen. 2010. Nomenclature and Biology 8:209–234.
placental mammal phylogeny. BMC [BioMedCentral] Basilio, R.P.A. 1952. La vida animal en la Guinea Espa-
Evolutionary Biology 10:102. ñola. Madrid: Instituto de Estudios Africanos.
Ashley, M.V., D.J. Melnick & D. Western. 1990. Conserva- Baskevich, M.I. & A.A. Danilkin. 1991. Cytogenetic
tion genetics of the black rhinoceros (Diceros bicornis), variability among representatives of the genus
1: evidence from the mitochondrial DNA of three Capreolus (Artiodactyla, Cervidae). In F. Spitz, G.
populations. Conservation Biology 4:71–77. Janeau, G. Gonzalez, et al. (eds.), Proceedings of the
Ashley, M.V., J.E. Norman & L. Stross. 1996. Phylogenetic International Symposium “Ongulés/Ungulates 91.”
analysis of the Perissodactylan family Tapiridae using Paris: Société Française pour l’Étude et la Protection
mitochondrial cytochrome c oxidase (COII) sequences. des Mammifères, 123–127.
Journal of Mammalian Evolution 3:315–326. Bastos-Silveira, C. & A.M. Lister. 2007. A morphometric
Azzaroli, A. 1985. Taxonomy of Quaternary Alcini assessment of geograph ical variation and subspecies in
(Cervidae, Mammalia). Acta Zoologica Fennica impala. Journal of Zoology, London 271:288–301.
170:179–180. Bechstein, J.M. 1799. Thomas Pennant’s Allgemeine
Balakrishnan, C.N., S.L. Monfort, A. Gaur, L. Singh & Übersicht der vierfüssign Thiere. Weimar:
M.D. Sorensen. 2003. Phylogeography and conserva- Industrie- Comptoirs.
tion genetics of Eld’s deer (Cervus eldi). Molecular Beja-Pereira, A., P.R. England, N. Ferrand, S. Jordan, A.O.
Ecology 12:1–10. Bakhiet, M.A. Abdalla, M. Mashkour, J. Jordana, P.
Banfield, A.W.F. 1961. A revision of the reindeer and Taberlet & G. Luikart. 2004. African origins of the
caribou, genus Rangifer. National Museum of Canada domestic donkey. Science 304:178.
Bulletin 77:1–137. Bello- Gutiérrez, J., R. Reyna-Hurtado & W. Jorge. 2010.
Bannikov, A.G. 1954. Die Säugetiere der Mongolischen Central American red brocket deer Mazama temama
Volksrepublik. Moscow: Verlag Akademie der (Kerr 1792). In J.M.B. Duarte & S. González (eds.),
Wissenschaften der USSR. Neotropical Cervidology: Biology and Medicine of
Bannikov, A.G. 1963. Die Saiga-Antilope (Saiga tartarica Latin American Deer. Jaboticabal, Brazil and Gland,
L.). Die Neue Brehm-Bücherei No. 320. Wittenberg- Switzerland: Funep, in collaboration with IUCN,
Lutherstadt, Germany: A. Ziemsen. 166–171.
Banwell, D.B. 1997. The Pannonians—Cervus elaphus van Bemmel, A.C.V. 1948. A further note on Axis
pannoniensis— a race apart. Deer 10:275–277. (Hyelaphus) kuhlii (Müller & Schlegel) (Ungulata,
Banwell, D.B. 1998. Identification of the Pannonian or Cervidae). Treubia 19:403– 406.
Danubian red deer: a maraloid—Cervus elaphus van Bemmel, A.C.V. 1949. Revision of the rusine deer in
pannoniensis. Deer 10:495– 497. the Indo-Australian archipelago. Treubia 20:191–262.
Banwell, D.B. 1999. The Sika. Auckland: Halcyon. Benirschke, K. & A.T. Kumamoto. 1989. Further studies
Banwell, D.B. 2006. The Rusa, the Sambar, and the on the chromosomes of three species of peccary.
Whitetail. Auckland: Halcyon. Advances in Neotropical Mammalogy 1989:309–316.
Banwell, D.B. 2009. The Red Deer, Part I. Auckland: Benirschke, K., D. Rüedi, H. Müller, A.T. Kumamoto, K.L.
Halcyon. Wagner & H.S. Downes. 1980. The unusual karyotype
Bard, J.B.L. 1977. A unity underlying the different zebra of the lesser kudu, Tragelaphus imberbis. Cytogenetics
striping patterns. Journal of Zoology, London and Cell Genetics 26:85– 92.
183:527–539. Benirschke, K., A.T. Kumamoto, G.N. Esra & K.B.
Baral, H.S., K.B. Shah & J.W. Duckworth. 2009. Crocker. 1982. The chromosomes of the bongo,
A clarification of the status of Indian chevrotain Taurotragus (Boocerus) eurycerus. Cytogenetics and Cell
Moschiola indica in Nepal. Vertebrate Zoology Genetics 34:10–18.
59:197–200. Benirschke, K., A.T. Kumamoto, J.H. Olsen, M.M.
Bar- Gal, G.K., P. Smith, E. Tchernov, C. Greenblatt, Williams & J. Oosterhuis. 1984. On the chromosomes
-1— P. Ducos, A. Gardeisen & L.K. Horwitz. 2002. Genetic of Gazella soemmeringi Cretzschmar, 1826. Zeitschrift
0— evidence for the origin of the agrimi goat (Capra für Säugetierkunde 49:368–373.
349-47558_ch01_1P.indd 282 5/31/11 7:06 PM

ref erences 283
Benirschke, K., A.T. Kumamoto & D.A. Meritt. 1985. Bourlière et J. Verschuren. Brussels: Institut des Parcs
Chromosomes of the Chacoan peccary, Catagonus Nationaux du Congo Belge.
wagneri (Rusconi). Journal of Heredity 76:95– 98. Bouvrain, G., D. Geraads & Y. Jehenne. 1989. Nouvelles
Birungi, J. & P. Arctander. 2000. Large sequence diver- données relatives à la classification des Cervidae
gence of mitochondrial DNA genotypes of the control (Artiodactyla, Mammalia). Zoologischer Anzeiger
region within populations of the African antelope, kob 223:82– 90.
(Kobus kob). Molecular Ecology 9:1997–2008. Bover, P. & J.A. Alcover. 1999. The evolution and
Birungi, J. & P. Arctander. 2001. Molecular systematics ontogeny of the dentition of Myotragus balearicus Bate,
and phylogeny of the Reduncini (Artiodactyla: 1909 (Artiodactyla, Caprinae): evidence from new
Bovidae) inferred from the analysis of mitochondrial fossil data. Biological Journal of the Linnaean Society
cytochrome b gene sequences. Journal of Mammalian 68:401– 428.
Evolution 8:125–147. Bover, P. & J.A. Alcover. 2003. Understanding Late
Black-Décima, P., R.V. Rossi, A. Vogliotti, J.L. Cartes, Quaternary extinctions: the case of Myotragus
L. Maffei, J.M.B. Duarte, S. González & J.P. Juliá. 2010. balearicus (Bate, 1909). Journal of Biogeography
Brown brocket deer Mazama gouazoubira (Fischer 1814). 30:771–781.
In J.M.B. Duarte & S. González (eds.), Neotropical Boyce, W.M. 1999. Population subdivision among desert
Cervidology: Biology and Medicine of Latin American bighorn sheep (Ovis canadensis) ewes revealed by
Deer. Jaboticabal, Brazil and Gland, Switzerland: mitochondrial DNA analysis. Molecular Ecology
Funep, in collaboration with IUCN, 190–201. 8:99–106.
Blyth, E. 1842. Report of Curator for April, 1842. Journal Boyeskorov, G. G. 1997. [Chromosomal differences in
of the Asiatic Society of Bengal, 11:444– 455. moose (Alces alces L., Artiodactyla, Mammalia)].
Blyth, E. 1863. Cata logue of Mammals in the Museum of Genetika 33:974– 978. [In Russian; English translation
the Asiatic Society of Bengal. Calcutta: Government in Russian Journal of Genetics, 33:825– 828.]
Printers. Boyeskorov, G.[G.] 1999. New data on moose (Alces,
Bohlken, H. 1958. Zur Nomenklatur der Haustier. Artiodactyla) systematics. Säugetierkundliche
Zoologischer Anzeiger 160:167–168. Mitteilungen 44:3–13.
Bohlken, H. 1967. Beitrag zur Systematik der rezenten Boyeskorov, G.G. 2001. On the systematics and distribu-
Formen der Gattung Bison H. Smith, 1827. Zeitschrift tion of sheep of the genus Ovis (Artiodactyla, Bovidae)
für Zoologisches Systematik und Evolutionsforschung in eastern Siberia and the Far East in the Pleistocene
5:54–110. and Holocene. Zoologicheskii Zhurnal 80:243–256.
Boisserie, J.-R. 2005. The phylogeny and taxonomy of Boyeskorov, G.G., M.V. Shchelchkova & Yu.V. Revin. 1993.
Hippopotamidae (Mammalia: Artiodactyla): a review Karyotype of moose (Alces alces L.) from north-eastern
based on morphology and cladistic analysis. Zoological Asia. Doklady Rossiyskoy Akademii Nauk
Journal of the Linnaean Society 143:1–26. 329:506–508.
Boisserie, J.-R. 2007. Family Hippopotamidae. In D.R. Bradley, R.D., F.C. Bryant, L.C. Bradley, M.L. Haynie &
Prothero & S.E. Foss (eds.), The Evolution of Artio- R.J. Baker. 2003. Implications of hybridisation between
dactyls. Baltimore: Johns Hopkins University Press, white-tailed deer and mule deer. Southwestern
106–119. Naturalist 48:654– 660.
Bork, A.M., C.M. Strobeck, F.C. Yeh, R.J. Hudson & R.K. Bradshaw, C.J.A., Y. Isagi, S. Kaneko, D.M.J.S. Bowman &
Salmon. 1991. Genetic relationship of wood and plains B.W. Brook. 2006. Conservation value of non-native
bison based on restriction fragment length polymor- banteng in northern Australia. Conservation Biology
phisms. Canadian Journal of Zoology 69:43– 48. 20:1306–1311.
Bongso, T.A. & M. Hilmi. 1982. Chromosome banding Braun, A., C.P. Groves, P. Grubb, Q-s. Yang & L. Xia.
homologies of a tandem fusion in river, swamp, and 2001. Cata logue of the Musée Heude collection of
crossbred buffaloes (Bubalus bubalis). Canadian Journal mammal skulls. Acta Zootaxonomica Sinica
of Genetics and Cytology 24:667– 673. 26:608– 660.
Bosma, A.A., N.A. de Haan & A.A. Macdonald. 1991. The Braun, A., C.P. Groves & P. Grubb. 2002. Rediscovery of
current status of cytogenetics of the Suidae: a review. the type specimen of Bubalus mindorensis Heude, 1888.
Bongo: Zoo-Report, Berlin 18:258–272. Mammalian Biology 67:246–249.
Botezat, E. 1903. Gestaltung und Klassifi kation des Brisson, M.-J. 1762. Regnum Animale in Classes IX
Geweihe des Edelhirsches, nebst einem Anhange über distributum, sive Synopsis Methodica, 2nd ed. Lugduni
die Stärke der Karpathenhirsche und die zwei Rassen Batavorum [Leiden]: Theodorum Haak.
derselben. Gegenbaurs Morphologisches Jahrbuch Brooks, A.C. 1961. A Study of the Thomson’s Gazelle
32:104–158. (Gazella thomsonii Günther) in Tanganyika. London:
Bourlière, F. & J. Verschuren. 1960. Introduction à Her Majesty’s Stationery Office.
l’Écologie des Ongules du Parc National Albert. Fasc. 1 Brown, D.M., R.A. Brenneman, K.-P. Koepfl i, J.P. —-1
of Exploration du Parc National Albert, Mission F. Pollinger, B. Mila, N.J. Georgiadis, E.E. Louis Jr., G.F. —0
349-47558_ch01_1P.indd 283 5/31/11 7:06 PM

Grether, D.K. Jacobs & R.K. Wayne. 2007. Extensive Cao K-q. 1978. On the time of extinction of the wild
population genetic structure in the giraffe. BMC mi-deer in China. Acta Zoologica Sinica 24:289–291.
Biology 5:57. doi:10.1186/1741-7007-5-57. Cap, H., S. Aulagnier & P. Deleporte. 2002. The phylogeny
Bubenik, A.B. 1990. Epigenetic, morphological, physi- and behaviour of the Cervidae (Ruminantia: Pecora).
ological, and behavioural aspects of evolution of horns, Ethology, Ecology & Evolution 14:199–216.
pronghorns, and antlers. In G.A. Bubenik & A.B. Cap, H., P. Deleporte, J. Joachim & D. Reby. 2008. Male
Bubenik (eds.), Horns, Pronghorns, and Antlers. New vocal behaviour and phylogeny in deer. Cladistics
York: Springer-Verlag, 1–113. 24:917– 931.
Buchner, E. 1891. Die Säugethiere der Gansu-Expedition Cappellini, I. 2007. Dimorphism in the hartebeest. In D.J.
(1884– 87). Mélanges Biologiques Tirés du Bulletin de Fairbairn, W.U. Blankenhorn & T. Szekely (eds.), Sex,
l’Académie Impériale des Sciences de St. Pétersbourg Size, and Gender Roles: Evolutionary Studies of Sexual
13:143–164. Size Dimorphism. New York: Oxford University Press,
Bunch, T.D. 2000. Cytogenetics, morphology, and 124–132.
evolution of four subspecies of the giant sheep argali Cappellini, I. & L.M. Gosling. 2006. The evolution of
(Ovis ammon) of Asia. Mammalia 64:199–207. fighting structures in hartebeest. Evolutionary
Bunch, T.D. 2006. Phylogenetic analysis of snow sheep Ecology Research 8:997–1011.
(Ovis nivicola) and closely related taxa. Journal of Carr, S.M., S.W. Ballinger, J.N. Derr, L.H. Blankenship &
Heredity 97:21–30. J.W. Bickham. 1986. Mitochondrial DNA analysis of
Bunch, T.D. & C.F. Nadler. 1980. Giemsa-band patterns of hybridization between sympatric white-tailed deer and
the tahr and chromosomal evolution of the tribe mule deer in West Texas. PNAS: Proceedings of the
Caprini. Journal of Heredity 71:110–116. National Academy of Science USA 83:9576– 9580.
Bunch, T.D., W.C. Foote & J.J. Spillett. 1976. Transloca- Cathey J.C., J.W. Bickham & J.C. Patton. 1998. Introgres-
tions of acrocentric chromosomes and their implica- sive hybridization and non-concordant evolutionary
tions in the evolution of sheep. Cytogenetics and Cell history of maternal and paternal lineages in North
Genetics 17:122–136. American deer. Evolution 52:1224–1229.
Bunch, T.D., S-q. Wang, R.S. Hoff mann, Y-p. Zhang, A-h. Chakraborty, S. 1972. On some cranial features of the
Liu, S-y. Lin & W. Wang. 1997. Diploid chromosome living-nation genera of family Rhinocerotidae
number and karyotype of the dalai-lama argali (Ovis (Mammalia: Perissodactyla). Proceedings of the
ammon dalailamae Przewalski, 1888). Encyclia: Journal Zoological Society, Calcutta 25:123–128.
of the Utah Academy of Sciences, Arts and Letters Chasen, F.N. 1940. A handlist of Malaysian mammals.
74:255–263. Bulletin of the Raffles Museum, Singapore, Straits
Bunch, T.D., N.N. Vorontsov, E.A. Lyapunova & R.S. Settlements 15:1–209.
Hoff mann. 1998. Chromosome number of Severtzov’s Chasen, F.N. & C.B. Kloss. 1930. On mammals from the
sheep (Ovis ammon severtzovi): G-banded karyotype Raheng district, western Siam. Journal of the Siam
comparisons within Ovis. Journal of Heredity Society, Natural History Supplement 8:61–78.
89:266–269. Chau, B. 1997. Another new discovery in Vietnam.
Bunch, T.D., S. Wang, R. Valdez, R.S. Hoff mann, Y. Zhang, Vietnam Economic News 47:46– 47.
A. Liu & S. Lin. 2000. Cytogenetics, morphology, and Chaveerach, A., W. Kakampuy, A. Tanomtong & W.
evolution of four subspecies of the giant sheep argali Sangpakdee. 2007. New Robertsonian translocation
(Ovis ammon) of Asia. Mammalia 64:199–207. chromosomes in captive Thai gaur (Bos gaurus readei).
Burney, D.A. & Ramilisonina. 1999. The Kilopilopitsofy, Pakistan Journal of Biological Science 10:2185–2191.
Kidoky, and Bokyboky: accounts of strange animals Chernyavsky, F.B. 1962. On the systematic relationships
from Belo-sur-Mer, Madagascar, and the megafaunal and history of the snow sheep of the Old and New
“extinction window.” American Anthropologist Worlds. Bulletin of the Moscow Society of Naturalists,
100:957– 966. Biology 67:17–26.
Butler, A.L. 1900. On a new serow from the Malay Chernyavsky, F.B. 2004. On the taxonomy and history of
Peninsula. Proceedings of the Zoological Society of bighorn sheep (Pachyceros subgenus, Artiodactyla).
London 1900:675– 676. Zoologicheskii Zhurnal 83:1059–1070.
Butynski, T.M. 2000. Taxonomy and distribution of the Child, C. & C.R. Savory. 1964. The distribution of large
hirola antelope. Gnusletter 19, 2:11–17. mammal species in Southern Rhodesia. Arnoldia
Butynski, T.M., C.D. Schaaf & G.W. Hearn. 1997. African (Rhodesia) 1(14):1–15.
buffalo Syncerus caffer extirpated on Bioko Island, Chorley, J.K. 1956. The distribution of Livingston’s suni in
Equatorial Guinea. Journal of African Zoology 111:57– 61. Southern Rhodesia. Proceedings and Transactions of
Cabrera, A. 1961. Cata logo de los Mamiferos de America the Rhodesia Scientific Association 44:63.
del Sur. Buenos Aires: Editora Coni. Choudhury, A. 1997. Checklist of the Mammals of Assam.
-1— Cansdale, G.S. 1960. Animals of West Africa. Longmans: Guawahati, India: Gibbon Books and Assam Science
0— London. Technology and Environment Council.
349-47558_ch01_1P.indd 284 5/31/11 7:06 PM

ref erences 285
Choudhury, A. 2003. Status of serow (Capricornis suma- scientific knowledge. Biodiversity and Conservation
traensis) in Assam. Tigerpaper 30, 2:1–2. 19:291–303. doi:10.1007/s10531- 009-9721-4.
Christofferson, M.L. 1995. Cladistic taxonomy, phyloge- Courtois, R., L. Bernatchez, J.-P. Ouellet & L. Breton.
netic systematics, and evolutionary ranking. System- 2003. Significance of caribou (Rangifer tarandus)
atic Biology 44:440– 454. ecotypes from a molecular genetics viewpoint.
Christy, C. 1924. Big Game and Pygmies. London: Conservation Genetics 4:393– 404.
MacMillan. Couturier, M.A.J. 1938. Le Chamois Rupicapra rupicapra
Christy, C. 1929. The African buffaloes. Proceedings of (L). Grenoble: B. Arthaud.
the Zoological Society of London 1929:445– 462. Cowan, I. McT. 1940. Distribution and variation in the
Clifford, A.B. & L.M. Witmer. 2004. Case studies in novel native sheep of North America. American Midland
narial anatomy, 3: structure and function of the nasal Naturalist 24:505–580.
cavity of saiga (Artiodactyla: Bovidae: Saiga tatarica). Cozens, A.B. 1951. The ranges of the dwarf and royal
Journal of Zoology, London 264:217–230. antelopes. Nigerian Field l6:419.
Cobb, E.H. 1958. The markhor. Oryx 4:381–382. Cranbrook, Earl of & P.J. Piper. 2009. Borneo records of
Collins, W.B. 1956. The tropical forest: an animal and Malay tapir, Tapirus indicus Desmarest: a zooarchaeo-
plant association. Nigerian Field 21:427. logical and historical review. International Journal of
Cope, E.D. 1889. On the Mammalia obtained by the Osteoarchaeology 19:491–507.
Naturalist Exploring Expedition to southern Brazil. Croft, D.A., L.R. Heaney, J.J. Flynn & A.P. Bautista. 2006.
American Naturalist 23:128–150. Fossil remains of a new, diminutive Bubalus (Artio-
Corbet, G.B. 1969. The taxonomic status of the pygmy dactyla: Bovidae: Bovini) from Cebu Island, Philip-
hippopotamus, Choeropsis liberiensis, from the Niger pines. Journal of Mammalogy 87:1037–1051.
Delta. Journal of Zoology, London 158:387–394. Cronin, M.A. 1992. Intraspecific mitochondrial DNA
Corbet, G.B. 1978. The Mammals of the Palaearctic variation in North American cervids. Journal of
Region: A Taxonomic Review. London: British Mammalogy 73:70– 82.
Museum (Natural History). Cronin, M.A. 2003. Research on deer taxonomy and its
Corbet, G.B. & J. Clutton-Brock. 1984. Appendix: relevance to management. Ecoscience 10:432– 442.
taxonomy and nomenclature. In I.L. Mason (ed.), Cucchi, T., M. Fujita & K. Dobney. 2009. New insights into
Evolution of Domesticated Animals. New York: pig taxonomy, domestication, and human dispersal in
Longman, 434– 438. Island East Asia: molar shape analysis of Sus remains
Corbet, G.B. & J.E. Hill. 1992. The Mammals of the from Niah Caves, Sarawak. International Journal of
Indomalayan Region: A Systematic Review. London: Osteoarchaeology 19:508–530.
Natural History Museum Publications and Oxford Custodio, C.C., M.V. Lepiten & L.R. Heaney. 1996. Bubalus
University Press. mindorensis. Mammalian Species 520:1–5.
Corbet, S.W. & T.J. Robinson. 1991. Genetic divergence in Dalquest, W.W. 1965. Mammals from the Save River,
South African wildebeest: comparative cytogenetics Mozambique, with descriptions of two new bats.
and analysis of mitochondrial DNA. Journal of Journal of Mammalogy 46:254–264.
Heredity 82:447– 452. Davis, E.B. 2007. Family Antilocapridae. In D.R.
Coryndon, S.C. 1978. Hippopotamidae. In V.J. Maglio & Prothero & S.E. Foss (eds.), The Evolution of Artio-
H.B.S. Cooke (eds.), Evolution of African Mammals. dactyls. Baltimore: Johns Hopkins University Press,
Cambridge, MA: Harvard University Press, 483– 495. 227–240.
Côté, S.D., J.F. Dallas, F. Marshall, R.J. Irvine, R. Langvatn De Beaux, O. 1911. Über einige Antilopen aus dem
& S.D. Albon. 2002. Microsatellite DNA evidence for Rufijitale. Zoologischer Anzeiger 38:575–582.
genetic drift and philopatry in Svalbard reindeer. Deraniyagala, P.E.P. 1972. The destruction of some
Molecular Ecology 11:1923–1930. existing species of Ceylon’s endemic mammals. Loris
Cotterill, F.P.D. 2003. Insights into the taxonomy of 2:242–244, 250.
tsessebe antelopes Damaliscus lunatus (Bovidae: Desmarest, A.G. 1822. Mammalogie, ou Description des
Alcelaphini) with the description of a new evolution- Espèces de Mammifères, 2ème Partie. Paris: Veuve
ary species in south-central Africa. Durban Museum Agasse.
Novitates 28:11–30. Deuve, J. 1972. Les Mammifères du Laos. Vientiane:
Cotterill, F.P.D. 2005. The Upemba lechwe, Kobus Ministry of National Education.
anselli: an antelope new to science emphasises the Dien Z-m. 1963. The Formosan serow (Capricornis swinhoii
conservation importance of Katanga, Democratic Gray). Quarterly Journal of the Taiwan Museum
Republic of the Congo. Journal of Zoology, London 16:97–100.
265:113–132. Ditchkoff, S.S., R.L. Lochmiller, R.E. Masters, S.T.
Cotterill, F.P.D. & W. Foissner. 2009. A pervasive Hoofer & R.A. Van den Bussche. 2001. Major-
denigration of natural history misconstrues how histocompatibility-complex-associated variation in —-1
biodiversity inventories and taxonomy underpin secondary sexual traits of white-tailed deer (Odocoileus —0
349-47558_ch01_1P.indd 285 5/31/11 7:06 PM

virginianus): evidence for good-genes advertisement. Eisenberg, J.F. 1989. The Northern Neotropics: Panama,
Evolution 55:616– 625. Colombia, Venezuela, Guyana, Suriname, French
Dobroruka, L.J. 1960. Der Karpatenhirsch, Cervus elaphus Guiana. Vol. 1 of Mammals of the Neotropics.
montanus Botezat 1903. Zoologischer Anzeiger Chicago: University of Chicago Press.
165:481– 483. Eisenberg, J.F. & K.H. Redford. 1999. The Central
Dolan, J.M. 1963. Beitrag zur systematischen Gliederung Neotropics: Ecuador, Peru, Bolivia, Brazil. Vol. 3 of
des Tribus Rupricaprini, Simpson 1945. Zeitschrift für Mammals of the Neotropics. Chicago: University of
Zoologisches Systematik und Evolutionsforschung Chicago Press.
1:311– 407. Eisenberg, J.F., C.P. Groves & K. MacKinnon. 1987.
Dolan, J.M. 1966. Notes on Addax nasomaculatus (de Tapire. In W. Keienberg (ed.), Grzimeks Enzyklopädie
Blainville, 1816). Zeitschrift für Säugetierkunde Säugetiere. 6 vols. Munich: Kindler, 4:598– 608.
31:23–31. Eldredge, N. & J. Cracraft. 1980. Phylogenetic Patterns
Dolan, J.M. 1988. A deer of many lands— a guide to the and the Evolutionary Process. New York: Columbia
subspecies of the red deer Cervus elaphus. ZooNooz University Press.
62(10):4–34. Ellerman, J.R. & T.C.S. Morrison-Scott. 1951. Checklist of
Dollman, J.G. 1927. A new race of Arabian gazelle. Palaearctic and Indian Mammals, 1758 to 1946.
Proceedings of the Zoological Society of London London: British Museum (Natural History).
1927:1005. Ellerman, J.R., T.C.S. Morrison-Scott & F.W. Hayman.
Dollman, J.G. 1928. A new gazelle shot by HRH the Duke 1953. Southern African Mammals, 1758 to 1951: A
of York, K.G. Natural History Magazine 1, 5:130–131. Reclassification. London: British Museum (Natural
Dong, W., Y. Pan & J. Liu. 2004. The earliest Muntiacus History).
(Artiodactyla, Mammalia) from the Late Miocene of Ellsworth, D.L., R.L. Honeycutt, N.J. Silvy, J.W. Bickham
Yuanmou, southwestern China. Comptes Rendus & W.D. Klimstra. 1994. Historical biogeography and
Palevol 3:379–386. contemporary patterns of mitochondrial DNA
Dragesco, J., F. Feer & J. Genermont. 1979. Contribution variation in white-tailed deer from the southeastern
à la connaissance de Neotragus batesi de Winton, 1903 United States. Evolution 48:122–136.
(position systématique, données biométriques). Endo, H., M. Kurohmaru & Y. Hayashi. 1995. An
Mammalia 43:71– 81. osteometrical study of the cranium and mandible
Drüwa, P. 1985. Die Damagazelle (Gazella dama ssp. of Ryukyu wild pig in Iriomote Island. Journal of
Pallas, 1767), einige Beiträge zur allgemeinen Biologie, Veterinary Medical Science 56:855– 860.
Haltung und Zucht im Zoologischen Garten. Endo, H., Y. Hayashi, M. Sasaki, Y. Kurosawa, K. Tanaka
Zoologischer Garten (n.f.) 55:1–28. & K. Yamazaki. 2001. Geograph ical variation of
Duarte, J.M.B. 1996. Guia de identificação de cervídeos mandible size and shape in the Japa nese wild pig (Sus
brasileiros. Jaboticabal, Brazil: Funep. scrofa leucomystax). Journal of Veterinary Medical
Duarte, J.M.B. & S. González (eds.). 2010. Neotropical Science 63:815– 820.
Cervidology: Biology and Medicine of Latin American Engländer, H. 1986. Capra pyrenaica Schinz,
Deer. Jaboticabal, Brazil and Gland, Switzerland: 1838— Spanischer Steinbock, Iberiensteinbok. In F.
Funep, in collaboration with IUCN. Niethammer & F. Krapp (eds.), Handbuch der
Duarte, J.M.B. & W. Jorge. 2003. Morphologic and Säugetiere Europas, vol. 2. pt. 2. Wiesbaden, Germany:
cytogenetic description of the small red brocket Aula, 405– 420.
(Mazama bororo Duarte, 1996) in Brazil. Mammalia Escamilo B., L.L., J. Barrio, J. Benavides F. & D.G. Tirira.
67:403– 410. 2010. Northern pudu Pudu mephistophiles (de Winton
Duarte, J.M.B., S. González & J.E. Maldonado. 2008. The 1896). In J.M.B. Duarte & S. González (eds.), Neotropi-
surprising evolutionary history of South American cal Cervidology: Biology and Medicine of Latin
deer. Molecular Phylogenetics and Evolution 49:17–22. American Deer. Jaboticabal, Brazil and Gland,
Ducos, P. 1968. L’ Origine des Animaux Domestiques en Switzerland: Funep, in collaboration with IUCN,
Palestine. Publications de l’Institut de Préhistoire de 133–139.
l’Université de Bordeaux No. 6. Bordeaux: Delmas. Estes, R.D. 1969. Behavioural study of East African
Dung, V.V., P. Giao, N. Chinh, T. Tuoc, P. Arctander & ungulates, 1963–1965. National Geographic Society
J. MacKinnon. 1993. A new species of living bovid from Research Reports, 1964 Projects:45–57.
Vietnam. Nature 363:443– 445. Estes, R.D. 1991. The Behaviour Guide to African
Dung, V.V., P. Giao, N. Chinh, D. Tuoc & J. MacKinnon. Mammals. Berkeley: University of California Press.
1994. Discovery and conservation of the vu quang in Estes, R.D. 2000. Evolution of conspicuous colouration in
Vietnam. Oryx 28:16–21. the Bovidae: female mimicry of male secondary
Eff ron, M., M.H. Bogart, A.T. Kumamoto & K. Benirschke. characters as catalyst. In E.S. Vrba & G.B. Schaller
-1— 1976. Chromosome studies in the mammalian (eds.), Antelopes, Deer, and Relatives. New Haven: Yale
0— subfamily Antilopinae. Genetica 46:419– 444. University Press, 234–246.
349-47558_ch01_1P.indd 286 5/31/11 7:06 PM

ref erences 287
Evangelista, P., P. Swartzinski & R. Waltermire. 2007. A Frade, F. & J.A. Silva. 1981. Mamiferos de Mocambique
profi le of the mountain nyala (Tragelaphus buxtoni). (colecção do Centro de Zoologia). Garcia da Orta, Série
African Indaba 5:1– 46. www.africanindaba.co.za. de Zoologia 10:1–11.
Ewer, R.F. 1957. A collection of Phacochoerus aethiopicus Frame, G.W. 1982. Wild mammal survey of Empakaai
teeth from the Kalkbank Middle Stone Age site, central Crater area, Tanganyika Notes Records 88– 89:41–55.
Transvaal. Palaeontologia Africana 5:5–20. Frechkop, S. 1943. Mammifères. Fasc. 1 of Exploration du
Falchetti, E., A. Ceccarelli & C. Mantovani. 1995. Parc National Albert, Mission S. Frechkop (1937–1938).
Relationship between dominance/subordination and Brussels: Institut des Parcs Nationaux du Congo
colouring patterns in Kobus megaceros (Bovidae, Belge.
Reduncinae) captive males. Gnusletter 14, 1:5. Furley, C.W. 1986. Reproductive parameters of African
Fedosenko, A.K. & D.A. Blank. 2001. Capra sibirica. gazelles: gestation, fi rst fertile matings, fi rst parturi-
Mammalian Species 675:1–13. tion, and twinning. African Journal of Ecology
Feer, F. 1979. Observations écologiques sur le néotrague 24:121–128.
de Bates (Neotragus batesi de Winton, 1903, Artiodac- Furley, C.W., H. Tichy & H.-P. Uerpmann. 1988. Systemat-
tyla, Ruminantia, Bovidae) du nordest du Gabon. La ics and chromosomes of the Indian gazelle, Gazella
Terre et La Vie 33:159–239. bennetti (Sykes, 1831). Zeitschrift für Säugetierkunde
Feng, J., C. Lajia, D.J. Taylor & M.S. Webster. 2001. 53:48–54.
Genetic distinctiveness of endangered dwarf blue Funk, S.M., S.K. Verma, G. Larson, K. Prasad, L. Singh, G.
sheep (Pseudois nayaur schaeferi): evidence from Narayan & J.E. Fa. 2007. The pygmy hog is a unique
mitochondrial control region and Y-linked ZFY intron genus: 19th century taxonomists got it right fi rst time
sequences. American Genetic Association 92:9–15. round. Molecular Phylogenetics and Evolution
Flagstad, Ø. & K.H. Røed. 2003. Refugial origins of 45:427– 436.
reindeer (Rangifer tarandus L.) inferred from mitochon- Gairdner, K.G. 1915. Addition to the mammalian fauna of
drial DNA sequences. Evolution 57:658– 670. Katburi. Journal of the Natural History Society of Siam
Flagstad, Ø., P.O. Syvertsen, N.C. Stenseth, J.E. Stacy, I. 1:252–255.
Olsaker, K.H. Røed & K.S. Jakobsen. 2000. Genetic Galbreath, G.J. & R.A. Melville. 2003. Pseudonovibos
variability in Swayne’s hartebeest, an endangered spiralis: epitaph. Journal of Zoology, London
antelope of Ethiopia. Conservation Biology 14:254–264. 259:169–170.
Flagstad, Ø, P.O. Syvertsen, N.C. Stenseth & K.S. Galbreath, G.J., J.C. Mordacq & F.H. Weiler. 2006.
Jakobsen. 2001. Environmental change and rates of Genetically solving a zoological mystery: was the
evolution: the phylogeographic pattern within the kouprey (Bos sauveli) a feral hybrid? Journal of Zoology,
hartebeest complex as related to climatic variation. London 270:561–564.
Proceedings of the Royal Society of London, B Gallina, S., S. Mandujano, J. Bello, H.F. López Arévalo &
268:667– 677. M. Weber. 2010. White-tailed deer Odocoileus virginia-
Flamand, J.R.B., D. Vankan, K.P. Gairhe, H. Duong & nus (Zimmermann 1780). In J.M.B. Duarte & S. González
J.S.F. Barker. 2003. Genetic identification of wild Asian (eds.), Neotropical Cervidology: Biology and Medicine
water buffalo in Nepal. Animal Conservation of Latin American Deer. Jaboticabal, Brazil and Gland,
6:265–270. Switzerland: Funep, in collaboration with IUCN,
Flerov, C.C. 1931. A review of the elks or moose (Alces 101–118.
Gray) of the Old World. Comptes Rendus de Garcia, J.E. & E.J.F. de Oliveira. 2010. Biochemical
l’Académie des Sciences de l’URSS 1931:71–74. genetics. In J.M.B. Duarte & S. González (eds.),
Flerov, C.C. 1952 [English translation 1960]. Mammals: Neotropical Cervidology: Biology and Medicine of
Musk Deer and Deer. Vol. 1, no. 2 of Fauna of USSR Latin American Deer. Jaboticabal, Brazil and Gland,
Mammals. Moscow & Leningrad: Academy of Sciences Switzerland: Funep, in collaboration with IUCN,
of the USSR [English translation by Israel Program for 27–30.
Scientific Translations, Washington, DC: National Garfield, B. 2006. The Meinertzhagen Mystery: The Life
Science Foundation and Smithsonian Institution]. and Legend of a Colossal Fraud. Washington, DC:
Flerov, C.C. 1965. Comparative craniology of recent Potomac Books.
representatives of the genus Bison. Bulletin of the Gatesy, J. & P. Arctander. 2000. Hidden morphological
Moscow Naturalists Society, Biology Division 70:41–54. support for the phylogenetic placement of Pseudoryx
Florescu, A., J.A. Davila, C. Scott, P. Fernando, K. Kellner, nghetinhensis with bovine bovids: a combined analysis
J.C. Morales, D. Melnick, P.T. Boag & P. van Co- of gross anatomical evidence and DNA sequences from
everden de Groot. 2003. Polymorphic microsatellites in five genes. Systematic Biology 49:515–538.
white rhinoceros. Molecular Ecology Notes 3:344–345. Gatesy, J., G. Amato, E. Vrba, G. Schaller & R. DeSalle.
Flower, W.H. 1876. On some cranial and dental characters 1997. A cladistic analysis of mitochondrial ribosomal
of the existing species of rhinoceroses. Proceedings of DNA from the Bovidae. Molecular Phylogenetics and —-1
the Zoological Society of London 1876:443– 457. Evolution 7:303–319. —0
349-47558_ch01_1P.indd 287 5/31/11 7:06 PM

Gatesy, J., M. Milinkovitch, V. Waddell & M. Stanhope. of muntjac (Artiodactyla: Muntiacidae) from central
1999. Stability of cladistic relationships between Vietnam, and implications for conservation. Animal
Cetacea and high-level Artiodactyla taxa. Systematic Conservation 1:61– 68.
Biology 48:6–20. Giere, P., C. Freyer & U. Zeller. 1999. Opening of the
Geist, V. 1987. On the evolution and adaptations of Alces. mammalian vomeronasal organ with respect to the
Swedish Wildlife Research Supplement 1:11–23. Glires hypothesis: a cladistic reconstruction of the
Geist, V. 1989. Environmentally guided phenotype Therian morphotype. Mitteilungen aus dem Museum
plasticity in mammals and some of its consequences to für Naturkunde in Berlin, Zoologische Reihe
theoretical and applied biology. In M.N. Bruton (ed.), 75:247–255.
Alternative Life-History Styles of Animals. Dordrecht: Gijzen, A. 1959. Das Okapi: Okapia johnstoni (Sclater). Die
Kluwer Academic, 153–176. Neue Brehm-Bücherei No. 231. Wittenberg-
Geist, V. 1991a. On the taxonomy of giant sheep (Ovis Lutherstadt, Germany: A. Ziemsen.
ammon Linnaeus, 1766). Canadian Journal of Zoology Gilbert, C., A. Ropiquet & A. Hassanin. 2006. Mitochon-
69:706–723. drial and nuclear phylogenies of Cervidae (Mammalia,
Geist, V. 1991b. Phantom subspecies: the wood bison Bison Ruminantia): systematics, morphology, and biogeogra-
bison “athabascae” Rhoads 1897 is not a valid taxon, but phy. Molecular Phylogenetics and Evolution 40:101–117.
an ecotype. Arctic 44:283–300. Gongora, J. & C. Moran. 2005. Nuclear and mitochondrial
Geist, V. 1998. Deer of the World. Mechnicsburg, PA: evolutionary analyses of collared, white-lipped, and
Stackpole Books. Chacoan peccaries (Tayassuidae). Molecular
Gentry, A. 1994. Case 2928: Regnum Animale . . . , ed. 2 Phylogenetics and Evolution 34:181–189.
(M.-J. Brisson, 1762); proposed rejection, with the Gongora, J., S. Morales, J.E. Bernal & C. Moran. 2006.
conservation of the mammalian generic names Phylogenetic divisions among collared peccaries (Pecari
Philander (Marsupialia), Pteropus (Chiroptera), Glis, tajacu) detected using mitochondrial and nuclear
Cuniculus, and Hydrochoerus (Rodentia), Meles, Lutra, sequences. Molecular Phylogenetics and Evolution
and Hyaena (Carnivora), Tapirus (Perissodactyla), 41:1–11.
Tragulus and Giraffa (Artiodactyla). Bulletin of González, B.A., R.E. Palma, B. Marín & J.C. Marín. 2006.
Zoological Nomenclature 51:135–146. Taxonomic and biogeograph ical status of guanaco
Gentry, A., J. Clutton-Brock & C.P. Groves. 2004. The Lama guanicoe (Artiodactyla, Camelidae). Mammal
naming of wild animal species and their domestic Review 36:157–178.
derivatives. Journal of Archaeological Science González, S., A. Gravier & N. Brum-Zorrilla. 1991. A
31:645– 651. systematic subspecifical approach on Ozotoceros
Gentry, A.W. 1964. Skull characters of African gazelles. bezoarticus (Linn. 1758) (pampas deer) from South
Annals and Magazine of Natural History (13) America. In F. Spitz, G. Janeau, G. Gonzalez, et al.
7:353–382. (eds.), Proceedings of the International Symposium
Gentry, A.W. 1985. The Bovidae of the Omo group “Ongulés/Ungulates 91.” Paris: Société Française pour
deposits, Ethiopia. In F.C. Howell & Y. Coppens (eds.), l’Étude et la Protection des Mammifères, 129–132.
Les Périssodactyles, les Artiodactyles (Bovidae). Vol. 1 González, S., F. Álvarez-Valin & J.E. Maldonado. 2002.
of Les Faunes Plio-Pleistocènes de la Basse Vallée de Morphometric differentiation of endangered pampas
l’Omo (Ethiopie). Paris: Éditions du CNRS [Centre deer (Ozotocerus bezoarticus), with description of new
National de la Recherche Scientifique], 119–191. subspecies from Uruguay. Journal of Mammalogy
Gentry, A.W. 1990. Evolution and dispersal of African 83:1127–1140.
Bovidae. In G.A. Bubenik & A.B. Bubenik (eds.), González, S., M. Cosse, F.G. Braga, A.R. Vila, M.L.
Horns, Pronghorns, and Antlers. New York: Springer- Merino, C. Dellafiore, J.L. Cartes, L. Maffei & M.G.
Verlag, 195–227. Dixon. 2010. Pampas deer Ozotocerus bezoarticus
Gentry, A.W. 1992. The subfamilies and tribes of (Linnaeus 1758). In J.M.B. Duarte & S. González (eds.),
the family Bovidae. Mammal Review 22:1–32. Neotropical Cervidology: Biology and Medicine of
Gentry, A.W. & J.J. Hooker. 1988. The phylogeny of the Latin American Deer. Jaboticabal, Brazil and Gland,
Artiodactyla. In M.J. Benton (ed.), Mammals. Vol. 2 Switzerland: Funep, in collaboration with IUCN,
of The Phylogeny and Classification of the Tetrapods. 119–132.
Oxford: Clarendon Press, 235–272. González, S., J.M.B. Duarte & J.E. Maldonado. 2010.
Ghosh, M. 1988. The craniology and dentition in the Molecular phylogenetics and evolution. In J.M.B.
pigmy hog, with a note on the generic status of Porcula Duarte & S. González (eds.), Neotropical Cervidol-
Hodgson, 1847. Records of the Zoological Survey of ogy: Biology and Medicine of Latin American Deer.
India 85:245–266. Jaboticabal, Brazil and Gland, Switzerland: Funep, in
Giao, P.M, D. Tuoc, V.V. Dung, E.D. Wikramanayake, G. collaboration with IUCN, 12–17.
-1— Amato, P. Arctander & J.R. Mackinnon. 1998. Goodman, S.J., N.H. Barton, G. Swanson, K. Abernethy &
0— Description of Muntiacus truongsonensis, a new species J.M. Pemberton. 1999. Introgression through rare
349-47558_ch01_1P.indd 288 5/31/11 7:06 PM

ref erences 289
hybridisation: a genetic study of a hybrid zone between Groves, C.P. 1975a. Notes on the gazelles, 1: Gazella
red and sika deer (genus Cervus) in Argyll, Scotland. rufifrons and the zoogeography of central African
Genetics 152:355–371. Bovidae. Zeitschrift für Säugetierkunde 40:308–319.
Granjon, L., M. Vassart & A. Greth. 1990. Genetic Groves, C.P. 1975b. Taxonomic notes on the white
variability in the Nubian ibex. Mammalia 54:665– 667. rhinoceros Ceratotherium simum (Burchell, 1817).
Granjon, L., M. Vassart, A. Greth & E.-P. Cribiu. 1991. Säugetierkundliche Mitteilungen 23:200–212.
Genetic study of sand gazelles (Gazella subgutturosa Groves, C.P. 1978a. The taxonomic status of the dwarf
marica) from Saudi Arabia: chromosomal and isozymic blue sheep (Artiodactyla: Bovidae). Säugetierkundliche
data. Zeitschrift für Säugetierkunde 56:169–176. Mitteilungen 26:177–183.
Graphodatsky, A.S. & S.I. Radjabli. 1985. Chromosomes Groves, C.P. 1978b. The extinct Cape rhinoceros, Diceros
of three cervid species (Mammalia). Zoologicheskii bicornis bicornis (Linnaeus, 1758). Säugetierkundliche
Zhurnal 64:1275–1279. Mitteilungen 26:117–128.
Gravlund, P., M. Meldgaard, S. Pääbo & P. Arctander. Groves, C.P. 1980. Notes on the systematics of Babyrousa
1998. Polyphyletic origin of the small-bodied, (Artiodactyla, Suidae). Zoologische Mededelingen
High-Arctic subspecies of tundra reindeer (Rangifer 55:29– 46.
tarandus). Molecular Phylogenetics and Evolution Groves, C.P. 1981a. Notes on the gazelles, 2: subspecies
10:151–159. and clines in the springbok (Antidorcas). Zeitschrift für
Gray, J.E. 1843. List of the Specimens of Mammalia in the Säugetierkunde 46:189–197.
Collection of the British Museum. London: British Groves, C.P. 1981b. Systematic relationships in the Bovini
Museum (Natural History). (Artiodactyla, Bovidae). Zeitschrift für Zoologisches
Gray, J.E. 1871. The Chinese long-tailed goat antelope Systematik und Evolutionsforschung 19:264–278.
(Urotragus caudatus). Annals and Magazine of Natural Groves, C.P. 1981c. Notes on the gazelles, 3: The dorcas
History (4) 8:371–372. gazelles of North Africa. Annali del Museo Civico di
Greth, A. 1992. Yemen gazelles. CBSG [Conservation Storia Naturale di Genova 83:455– 471.
Breeding Specialist Group] News 3, 2:18. Groves, C.P. 1981d. Ancestors for the Pigs: Taxonomy and
Greth, A., D. Williamson, C. Groves, G. Schwede & M. Phylogeny of the Genus Sus. Technical Bulletin No. 3.
Vassart. 1993. Bilkis gazelle in Yemen— status and Canberra: Australian National University, Research
taxonomic relationships. Oryx 27:239–244. School of Pacific Studies, Department of Prehistory.
Grobler, J.H. 1980. Body growth and age determination of Groves, C.P. 1982a. Asian rhinoceroses— down but not
the sable Hippotragus niger niger (Harris, 1838). Koedoe out. Malayan Naturalist 36:11–22.
23:131–156. Groves, C.P. 1982b. A note on geographic variation in
Grobler, J.P. & F.H. van der Bank. 1993. Genetic the Indian blackbuck (Antilope cervicapra Linnaeus,
diversity and differentiation of the three extant 1758). Records of the Zoological Survey of India
southern African species of the subfamily Hippotragi- 79:489–503.
nae (family: Bovidae). Biochemical Systematics and Groves, C.P. 1982c. Bos gaurus H. Smith, 1827 (Mammalia,
Ecology 21:591–596. Artiodactyla): proposed conservation. Bulletin of
Grodinsky, C. & M. Stuewe. 1987. With lots of help, alpine Zoological Nomenclature 39:279–280.
ibex return to their mountains. Smithsonian Magazine Groves, C.P. 1982d. Antilope depressicornis H. Smith, 1827,
18, 9:68–77. and Anoa quarlesi Ouwens, 1910 (Mammalia, Artiodac-
Groves, C.P. 1967a. The rhinoceroses of Southeast Asia. tyla): proposed conservation. Bulletin of Zoological
Säugetierkundliche Mitteilungen 15:221–237. Nomenclature 39:281–282.
Groves, C.P. 1967b. Geographic variation in the black Groves, C.P. 1983a. Phylogeny of the living species of
rhinoceros. Zeitschrift für Säugetierkunde rhinoceros. Zeitschrift für Zoologisches Systematik
32:267–276. und Evolutionsforschung 21:293–313.
Groves, C.P. 1967c. On the gazelles of the genus Procapra Groves, C.P. 1983b. Notes on the gazelles, 4: the Arabian
Hodgson, 1846. Zeitschrift für Säugetierkunde gazelles collected by Hemprich and Ehrenberg.
32:144–149. Zeitschrift für Säugetierkunde 48:371–381.
Groves, C.P. 1969a. The smaller gazelles of the genus Groves, C.P. 1983c. Geographic variation in the baras-
Gazella de Blainville, 1816. Zeitschrift für Säuget- ingha or swamp deer (Cervus duvauceli). Journal of the
ierkunde 34:38– 60. Bombay Natural History Society 79:620– 629.
Groves, C.P. 1969b. Systematics of the anoa (Mammalia, Groves, C.P. 1983d. A new subspecies of sable antelope,
Bovidae). Beaufortia 17:1–12. Hippotragus niger (Harris, 1838). Revue Zoologique
Groves, C.P. 1971. Species characters in rhinoceros horns. Africaine 97:821– 828.
Zeitschrift für Säugetierkunde 36:238–252. Groves, C.P. 1985. An introduction to the gazelles.
Groves, C.P. 1974. A note on the systematic position of the Chinkara 1, 1:4–10.
muntjac (Artiodactyla, Cervidae). Zeitschrift für Groves, C.P. 1986. The taxonomy, distribution, and —-1
Säugetierkunde 39:369–372. adaptations of recent equids. In R.H. Meadow & H.P. —0
349-47558_ch01_1P.indd 289 5/31/11 7:06 PM

Uerpmann (eds.), Equids in the Ancient World. Groves, C.P. 2000. Phylogenetic relationships within
Wiesbaden, Germany: Reichert, 11– 65. recent Antilopini (Bovidae). In E.S. Vrba & G.B.
Groves, C.P. 1989a. Feral mammals of the Mediterranean Schaller (eds.), Antelopes, Deer, and Relatives. New
islands: documents of early domestication. In J. Haven: Yale University Press, 223–233.
Clutton-Brock (ed.), The Walking Larder. London: Groves, C.P. 2001a. Primate Taxonomy. Washington, DC:
Unwin Hyman, 46–58. Smithsonian Institution Press.
Groves, C.P. 1989b. A cata logue of the genus Gazella. In Groves, C.P. 2001b. Why taxonomic stability is a bad i
A. Dixon & D. Jones (eds.), Conservation and Biology dea, or why are there so few species of primates
of Desert Antelopes. London: Christopher Helm, (or are there?). Evolutionary Anthropology
193–198. 10:192–198.
Groves, C.P. 1989c. The gazelles of the Arabian peninsula. Groves, C.P. 2002. Taxonomy of living Equidae. In P.D.
In A.H. Abu-Zinada, P.D. Goriup & I.A. Nader (eds.), Moehlman (ed.), Equids: Zebras, Asses, and Horses.
Wildlife Conservation and Development in Saudi Gland, Switzerland: IUCN, 94–107.
Arabia. Publication No. 3. Riyadh, Saudi Arabia: Groves, C.P. 2003. Taxonomy of ungulates of the Indian
National Commission for Wildlife Conservation and subcontinent. Journal of the Bombay Natural History
Development, 237–248. Society 100:341–362.
Groves, C.P. 1993a. The chinkara (Gazella bennetti) in Iran, Groves, C.P. 2004. The what, why, and how of primate
with the description of two new subspecies. Journal of taxonomy. International Journal of Primatology
Sciences of the Islamic Republic of Iran 4:166–178. 25:1105–1126.
Groves, C.P. 1993b. Testing rhinoceros subspecies by Groves, C.P. 2005. Domestic and wild mammals: naming
multivariate analysis. In O.A. Ryder (ed.), Rhinoceros and identity. In A. Minelli, G. Ortalli & G. Sangha
Biology and Conservation. San Diego: Zoological (eds.), Animal Names. Venice: Istituto Veneto di
Society of San Diego, 92–100. Scienze, Lettere ed Arti, 151–157.
Groves, C.P. 1994. Morphology, habitat, and taxonomy. In Groves, C.P. 2006a. The genus Cervus in eastern Eurasia.
L. Boyd & K.A. Houpt (eds.), Przewalski’s Horse: The European Journal of Wildlife Research 52:14–22.
History and Biology of an Endangered Species. Albany: doi:10.1007/s10344- 005- 0011-5.
State University of New York Press, 39–59. Groves, C.P. 2006b. Tribute to Peter Grubb. Suiform
Groves, C.P. 1995a. On the nomenclature of domestic Soundings 6, 2:4.
animals. Bulletin of Zoological Nomenclature Groves, C.P. 2007. Current views on the taxonomy and
52:137–141. zoogeography of the genus Sus. In U. Albarella, K.
Groves, C.P. 1995b. Microtaxonomy and its implications Dobney, A. Ervynck & P. Rowley- Conwy (eds.), Pigs
for captive breeding. In U. Ganslosser, J.K. Hodges & and Humans: 10,000 Years of Interaction. Oxford:
W. Kaumanns (eds.), Research and Captive Propaga- Oxford University Press, 15–29.
tion. Fürth, Germany: Filander, 24–28. Groves, C.P. 2008a. Family Cervidae. In D.R. Prothero &
Groves, C.P. 1996a. Taxonomic diversity in Arabian S.E. Foss (eds.), The Evolution of Artiodactyls.
gazelles: the state of the art. In A. Greth, C. Magin & Baltimore: Johns Hopkins University Press, 249–256.
M. Ancrenaz (eds.), Conservation of Arabian Gazelles. Groves, C.P. 2008b. Extended Family: Long Lost Cousins.
Riyadh, Saudi Arabia: National Commission for Arlington, VA: Conservation International.
Wildlife Conservation and Development, 8–39. Groves, C.P. & C. Bell. 2004. New investigations on the
Groves, C.P. 1996b. The taxonomy of the Asian wild taxonomy of the zebras genus Equus, subgenus
buffalo from the Asian mainland. Zeitschrift für Hippotigris. Mammalian Biology 69:182–196.
Säugetierkunde 61:327–338. Groves, C.P. & S. Chakraborty. 1983. The Calcutta
Groves, C.P. 1997a. Taxonomy of wild pigs (Sus) of the collection of Asian rhinoceros. Records of the
Philippines. Zoological Journal of the Linnaean Zoological Survey of India 80:251–263.
Society 120:163–191. Groves, C.P. & Feng Z-j. 1986. The status of musk deer
Groves, C.P. 1997b. The taxonomy of Arabian gazelles. In from Anhui Province, China. Acta Theriologica Sinica
K. Habibi, A.H. Abuzinada & I.A. Nader (eds.), The 6:101–106.
Gazelles of Arabia. Publication No. 29. Riyadh, Saudi Groves, C.P. & P. Grubb. 1974. A new duiker from
Arabia: National Commission for Wildlife Conserva- Rwanda. Revue Zoologique Africaine 88:189–196.
tion and Development, 24–51. Groves, C.P. & P. Grubb. 1982. The species of muntjac
Groves, C.P. 1997c. Die Nashörner— Stammesgeschichte (genus Muntiacus) in Borneo: unrecognised sympatry
und Verwandtschaft. In Die Nashörner: Begegnung in tropical deer. Zoologische Mededelingen
mit urzeitlichen Kolossen. Fürth, Germany: Filander, 56:203–216.
14–32. Groves, C.P. & P. Grubb. 1985. Reclassification of the
Groves, C.P. 1999. The advantages and disadvantages of serows and gorals (Nemorhaedus: Bovidae). In S. Lovari
-1— being domesticated (a keynote address). Perspectives in (ed.), The Biology and Management of Mountain
0— Human Biology 4:1–12. Ungulates. London: Croom Helm, 45–50.
349-47558_ch01_1P.indd 290 5/31/11 7:06 PM

ref erences 291
Groves, C.P. & P. Grubb. 1987. Relationships of living deer. Groves, C.P., Y-x. Wang & P. Grubb. 1995. Taxonomy of
In C. Wemmer (ed.), Biology and Management of the musk deer, genus Moschus (Moschidae, Mammalia).
Cervidae. Washington, DC: Smithsonian Institution Acta Theriologica Sinica 15:181–197.
Press, 21–59. Groves, C.P., G.B. Schaller, G. Amato & K. Khounbouline.
Groves, C.P. & P. Grubb. 1990. Muntiacidae. In G.A. 1997. Rediscovery of the wild pig Sus bucculentus.
Bubenik & A.B. Bubenik (eds.) Horns, Pronghorns, and Nature 386:335.
Antlers. New York: Springer-Verlag, 132– 68. Groves, C.P., P. Fernando & J. Robovsky. 2010. The sixth
Groves, C.P. & P. Grubb. 1993a. The suborder Suiformes. rhino: a taxonomic re-assessment of the critically
In W.L.R. Oliver (ed.), Pigs, Peccaries, and Hippos: endangered northern white rhinoceros. PLoS One 5,
Status Survey and Conservation Action Plan. Gland, 4:e9703. doi:10.1371/journal.pone.0009703.
Switzerland: IUCN, SSC [Species Survival Commis- Groves, P. & G.F. Shields. 1996. Phylogenetics of the
sion] Pigs and Peccaries Specialist Group, 1– 4. Caprinae based on cytochrome b sequence. Molecular
Groves, C.P. & P. Grubb. 1993b. The Eurasian suids Sus and Phylogenetics and Evolution 5:467– 476.
Babyrousa: taxonomy and description. In W.L.R. Oliver Groves, P. & G.F. Shields. 1997. Cytochrome b sequences
(ed.), Pigs, Peccaries, and Hippos: Status Survey and suggest convergent evolution of the Asian takin and
Conservation Action Plan. Gland, Switzerland: IUCN, Arctic musk- ox. Molecular Phylogenetics and
SSC Pigs and Peccaries Specialist Group, 107–111. Evolution 8:363–374.
Groves, C.P. & C. Guérin. 1980. Le Rhinoceros sondaicus Grubb, P. 1972. Variation and incipient speciation in the
annamiticus (Mammalia, Perissodactyla) d’Indochine: African buffalo. Zeitschrift für Säugetierkunde
distinction taxonomique et anatomique; relations 37:121–144.
phyletiques. Geobios 13:199–208. Grubb, P. 1977. Notes on a rare deer, Muntiacus feai. Annali
Groves, C.P. & D.L. Harrison. 1967. The taxonomy of the del Museo Civico de Storia Naturale de Genova
gazelles of Arabia. Journal of Zoology, London 81:202–207.
152:381–387. Grubb, P. 1978a. Patterns of speciation in African
Groves, C.P. & J. Jayewardene. 2009. The wild buffalo of mammals. Bulletin of Carnegie Museum of Natural
Sri Lanka. Taprobanica 1:56– 62. History 6:152–167.
Groves, C.P. & M. Karami. 1993. A mammal species new Grubb, P. 1978b. A new antelope from Gabon. Zoological
for Iran: Gazella gazella Pallas, 1766 (Artiodactyla: Journal of the Linnaean Society of London 62:373–380.
Bovidae). Journal of Sciences of the Islamic Republic of Grubb, P. 1979. Refuges and dispersal in the speciation of
Iran 4:81– 89. African forest mammals. Paper presented at the
Groves, C.P. & D.M. Lay. 1985. A new species of the genus Association of Tropical Biologists, 5th International
Gazella from the Arabian Peninsula. Mammalia Symposium, 8–13 February 1979, La Guaira,
49:27–36. Venezuela.
Groves, C.P. & V. Mazák. 1967. Taxonomic problems of Grubb, P. 1981. Equus burchellii. Mammalian Species
Asiatic wild asses: with the description of a new 157:1– 9.
subspecies. Zeitschrift für Säugetierkunde 32:321–355. Grubb, P. 1982a. Refuges and dispersal in the speciation of
Groves, C.P. & E. Meijaard. 2005. Interspecific variation in African forest mammals. In G.T. Prance (ed.),
Moschiola, the Indian chevrotain. Raffles Bulletin of Biological Diversification in the Tropics. New York:
Zoology 12:413– 421. Columbia University Press, 537–553.
Groves, C.P. & O.A. Ryder. 2000. Systematics and phylog- Grubb, P. 1982b. The systematics of Sino-Himalayan musk
eny of the horse. In A.T. Bowling & A. Ruvinsky (eds.), deer (Moschus), with par ticu lar reference to the species
The Genetics of the Horse. New York: CABI, 1–24. described by B.H. Hodgson. Säugetierkundlich
Groves, C.P. & G.B. Schaller. 2000. The phylogeny and Mitteilungen 30:127–137.
biogeography of the newly discovered Annamite Grubb, P. 1985a. Geograph ical variation in the bushbuck
artiodactyls. In E.S. Vrba & G.B. Schaller (eds.), of eastern Africa (Tragelaphus scriptus, Bovidae). In
Antelopes, Deer, and Relatives. New Haven: Yale K.-L. Schuchmann (ed.), African Vertebrates: Systemat-
University Press, 261–282. ics, Phylogeny, and Evolutionary Ecology. Bonn:
Groves, C.P. & C. Smeenk. 1978. On the type material of Zoologisches Forschungsinstitut und Museum
Cervus nippon Temminck, 1836: with a revision of sika Alexander Koenig, 11–27.
deer from the main Japa nese islands. Zoologische Grubb, P. 1985b. The biogeographic significance of forest
Mededelingen 53:11–28. mammals in eastern Africa. In E. Van der Straeten,
Groves, C.P. & C. Smeenk. 2007. The nomenclature of the V.N. Verheyen & F. De Vree (eds.), Proceedings of the
African wild ass. Zoologische Mededelingen Third International Colloquium on the Ecology and
81:121–135. Taxonomy of African Small Mammals. Koninklijk
Groves, C.P. & C.R. Westwood. 1995. Skulls of the Museum voor Midden-Afrika Zoologische Weten-
blaauwbok, Hippotragus leucophaeus. Zeitschrift für schappen No. 237. Tervuren, Belgium: Koninklijk —-1
Säugetierkunde 60:314–318. Museum voor Midden-Afrika, 75– 85. —0
349-47558_ch01_1P.indd 291 5/31/11 7:06 PM

Grubb, P. 1986. Notes on some antelope subspecies. lato). In C. Oswald. (ed.), Symposium on Red Deer
Gnusletter 1986, Jan:3– 4. Taxonomy. Ebersberg, Germany: Christian Oswald-
Grubb, P. 1988. The status of Walker’s duiker, a rare Stiftung, 12–30.
antelope from Malawi. Nyala 12:67–72. Grubb, P. 2005a. Order Perissodactyla. In D.E. Wilson &
Grubb, P. 1989. The systematic status of the suni D.M. Reeder (eds.), Mammal Species of the World: A
(Neotragus moschatus) in Malawi. Nyala 13:21–27. Taxonomic and Geographic Reference, 3rd ed., 2 vols.
Grubb, P. 1990a. Cervidae of Southeast Asia. In G.A. Washington, DC: Smithsonian Institution Press,
Bubenik & A.B. Bubenik (eds.), Horns, Pronghorns, 1:629– 636.
and Antlers. New York: Springer-Verlag, 169–179. Grubb, P. 2005b. Order Artiodactyla. In D.E. Wilson &
Grubb, P. 1990b. Primate geography in the Afrotropical D.M. Reeder (eds.), Mammal Species of the World: A
forest biome. In G. Peters & R. Hutterer (eds.), Taxonomic and Geographic Reference, 3rd ed., 2 vols.
Vertebrates in the Tropics. Bonn: Zoologisches Washington, DC: Smithsonian Institution Press,
Forschungsinstitut und Museum Alexander Koenig, 1:637–722.
187–214. Grubb, P. 2006. Geospecies and superspecies in the
Grubb, P. 1990c. List of deer species and subspecies. Deer African primate fauna. Primate Conservation
8:153–155. 20:75–78.
Grubb, P. 1993a. The Afrotropical hippopotamuses Grubb, P. In press. Evolutionary geography of modern
Hippopotamus and Hexaprotodon: taxonomy and African mammals. In T. Butysnki (ed.), Vegetation,
description. In W.L.R. Oliver (ed.), Pigs, Peccaries, and Climate, and Geology of Africa: Evolution of African
Hippos: Status Survey and Conservation Action Plan. Mammals; Primates. Vol. 1 of The Mammals of Africa.
Gland, Switzerland: IUCN, SSC Pigs and Peccaries London: Academic Press.
Specialist Group, 41– 43. Grubb, P. & A.L. Gardner, 1998. List of species and
Grubb, P. 1993b. The Afrotropical suids Phacochoerus, subspecies of the families Tragulidae, Moschidae, and
Hylochoerus, and Potamochoerus: taxonomy and Cervidae. In C. Wemmer (ed.), Deer: Status Survey and
description. In W.L.R. Oliver (ed.), Pigs, Peccaries, and Conservation Action Plan. Gland Switzerland: IUCN,
Hippos: Status Survey and Conservation Action Plan. 6–16.
Gland, Switzerland: IUCN, SSC Pigs and Peccaries Grubb, P. & C.P. Groves. 1983. Notes on the taxonomy of
Specialist Group, 66–79. the deer (Mammalia, Cervidae) of the Philippines.
Grubb, P. 1994. Notes on Grant’s gazelle. Gnusletter 13, Zoologische Anzeiger ( Jena) 210:119–144.
1–2:4–5. Grubb, P. & C.P. Groves. 1990. Muntiacidae. In G.A.
Grubb, P. 1999a. Types and type localities of ungulates Bubenik & A.B. Bubenik (eds.), Horns, Pronghorns,
named from southern Africa. Koedoe 42:13– 45. and Antlers. New York: Springer-Verlag, 134–168.
Grubb, P. 1999b. Evolutionary processes implicit in Grubb, P. & C.P. Groves. 1993. The neotropical tayassuids
distribution patterns of modern African mammals. In Tayassu and Catagonus: taxonomy and description. In
T.G. Bromage & F. Schrenk (eds.), African Biogeogra- W.L.R. Oliver (ed.), Pigs, Peccaries, and Hippos: Status
phy, Climate Change, and Human Evolution. New Survey and Conservation Action Plan. Gland,
York: Oxford University Press, 150–164. Switzerland: IUCN, SSC Pigs and Peccaries Specialist
Grubb, P. 2000a. Morphoclinal evolution in ungulates. In Group, 5–7.
E.S. Vrba & G.B. Schaller (eds.), Antelopes, Deer, and Grubb, P. & C.P. Groves. 2001. Appendix C: revision and
Relatives. New Haven: Yale University Press, 156–170. classification of the Cephalophinae. In V.J. Wilson,
Grubb, P. 2000b. Valid and invalid nomenclature of living Duikers of Africa: Masters of the African Forest Floor.
and fossil deer, Cervidae. Acta Theriologica Bulawayo: Chipangali Wildlife Trust, 703–728.
45:289–307. Grubb, P. & W. Oliver. 1991. Pigs and Peccaries Specialist
Grubb, P. 2001a. Review of family-group names of living Group report: A forgotten warthog; more bush pigs
bovids. Journal of Mammalogy 82:374–318. but fewer forest hogs. Species 17:61– 62.
Grubb, P., 2001b. Endemism in African rain forest Grubb, P. & C.B. Powell. 2002. Range extension of
mammals. In W. Weber, L.J.T. White, A. Vedder & L. black-fronted duiker (Cephalophus nigrifrons Gray 1871,
Jaughtoh Treves (eds.), African Rain Forest Ecology Artiodactyla, Bovidae): fi rst records from Nigeria.
and Conservation: An Interdisciplinary Perspective. Tropical Zoology 15, 1:89– 95.
New Haven: Yale University Press, 88–100. Grubb, P. & V.J. Wilson. 1990. Classification of antelopes
Grubb, P., 2002. Types, type locality, and subspecies of the adopted for the antelope survey. In R. East (ed.), West
gerenuk Litocranius walleri (Artiodactyla: Bovidae). and Central Africa. Part 3 of Antelopes: Global Survey
Journal of Zoology, London 257, 4:530–543. and Regional Action Plans. Gland, Switzerland: IUCN,
Grubb, P. 2004a. Controversial scientific names of African 3– 4.
mammals. African Zoology 39, 1:91–109. Grubb, P., R. East & V.J. Wilson. 1989. Classification of
-1— Grubb, P. 2004b. Nomenclature, subspeciation, and antelopes adopted for the antelope survey. In R. East
0— affi nities among elaphine deer (Cervus elaphus sensu (ed.), Southern and South- Central Africa. Part 2 of
349-47558_ch01_1P.indd 292 5/31/11 7:06 PM

ref erences 293
Antelopes: Global Survey and Regional Action Plans. Hammond, J., J.C. Bowman & T.J. Robinson. 1983.
Gland, Switzerland: IUCN, 3– 4. Hammond’s Farm Animals. London: Edward Arnold.
Grubb, P., T.S. Jones, A.G. Davies, E. Edberg, E.D. Starin Hammond, R.L., W. Macasero, B. Flores, O.B. Moham-
& J.E. Hill. 1998. Mammals of Ghana, Sierra Leone, med, T. Wacher & M.W. Bruford. 2001. Phylogenetic
and the Gambia. St. Ives, UK: Trendine. reanalysis of the Saudi gazelle and its implications for
Grubb, P., A.M. Lister & S.R.M. Sumner. 1998. Taxonomy, conservation. Conservation Biology 15:1123–1133.
morphology, and evolution of Eu ropean roe deer. In Hanitsch, R. 1918. On a serow from Annam. Journal
R. Andersen, P. Duncan & J.D.C. Linnell (eds.), The of the Straits Branch of the Royal Asiatic Society
European Roe Deer: The Biology of Success. Oslo: 78:59– 65.
Scandinavian University Press, 23– 46. Hansen-Melander, E. & Y. Melander. 1974. The karyotype
Grubb, P., O. Sandrock, O. Kullmer, T.M. Kaiser & F.J. of the pig. Hereditas 77:149–158.
Schrenk. 1999. Relationships between eastern and Hardwicke, T. 1925. Descriptions of two species of
southern African mammal faunas. In T.G. Bromage & antelope from India. Transactions of the Linnaean
F. Schrenk (eds.), African Biogeography, Climate Society 14:518–529.
Change, and Human Evolution. New York: Oxford Harley, E.H., I. Baumgarten, J. Cunningham & C. O’Ryan.
University Press, 253–267. 2005. Genetic variation and population structure in
Grubb, P., S.C. Kingswood & D.P. Mallon. 2001. Classifica- remnant populations of black rhinoceros, Diceros
tion of antelopes adopted for the antelope survey. In bicornis, in Africa. Molecular Ecology 14:2981–2990.
D.P. Mallon & S.C. Kingswood (comps.), North Africa, Harper, F. 1940. The nomenclature and type localities of
the Middle East, and Asia. Part 4 of Antelopes: Global certain Old World mammals. Journal of Mammalogy
Survey and Regional Action Plans. Gland, Switzerland: 21:191–203, 322–332.
IUCN, 7– 9. Harris, J.M. 2007. Superfamily Suoidea. In D.R. Prothero
Grubb, P., C.P. Groves & C.B. Powell, 2003. Duikers and & S.E. Foss (eds.), The Evolution of Artiodactyls.
dwarf antelopes: new or uncertain records. In A. Baltimore: Johns Hopkins University Press, 130–150.
Plowman (ed.), Ecology and Conservation of Small Harrison, D.L. 1967. Observations on a wild goat, Capra
Antelope. Fürth, Germany: Filander, 127–140. aegagrus (Artiodactyla: Bovidae) from Oman, E.
Gu Z-l., X-b. Zhao, N. Li & C-x. Wu. 2007. Complete Arabia. Journal of Zoology, London 151:27–30.
sequence of the yak (Bos grunniens) mitochondrial Harrison, D.L. & P.J.J. Bates. 1991. The Mammals of
genome and its evolutionary relationship with other Arabia, 2nd ed. Sevenoaks, UK: Harrison Zoological
ruminants. Molecular Phylogenetics and Evolution Museum.
42:248–255. Harrison, R.G. 1998. Linking evolutionary patterns and
Guérin, C. 1965. Gallogoral (nov. gen.) meneghinii (Rüti- processes: the relevance of species concepts for the
meyer, 1878): Un Rupicapriné du Villafranchien study of speciation. In D.J. Howard & S.H. Berlocher
d’Europe Occidentale. Documents des Laboratoires de (eds.), Endless Forms: Species and Speciation. Oxford:
Géologie de Lyon No. 11. Oxford University Press, 19–31.
Guo Z-p., E-y. Chen & Y-z. Wang. 1978. A new subspecies Hartl, G.B., R. Göltenboth, M. Grillitsch & R. Willing.
of sika deer from Sichuan—Cervus nippon sichuanicus 1988. On the biochemical systematics of the Bovini.
subsp. nov. Acta Zoologica Sinica 24:187–192. Biochemical Systematics and Ecology 16:575–579.
Gyldenstolpe, N. 1916. Zoological Results of the Swedish Hartl, G.B., H. Burger, R. Willing & F. Suchentrunk. 1990.
Zoological Expedition to Siam 1911–1912 and 1914– On the biochemical systematics of the Caprini and the
1915: 5, Mammals II. Kungliga Svenska Vetenskap- Rupicaprini. Biochemical Systematics and Ecology
sakademien Handlingar No. 57, 2. Stockholm: 18:175–182.
Almqvist & Wiksell. Hartl, G.B., K. Nadlinger, M. Apollonio, G. Markov, F.
Hall, E.R. & K.R. Kelson. 1959. The Mammals of North Klein, G. Lange, S. Findo & J. Markowski. 1995.
America. New York: Ronald Press. Extensive mitochondrial DNA differentiation among
Haltenorth, T. 1963. Klassifi kation der Säugetiere: European red deer (Cervus elaphus) populations:
Artiodactyla. Handbuch der Zoologie No. 8(32), pt. 1, implications for conservation and management.
18. Berlin: De Gruyter. Zeitschrift für Säugetierkunde 60:41–52.
Hamilton Smith, C. 1827. Order VII—Ruminantia Hassanin, A. & E.J.P. Douzery. 1999a. The tribal radiation
(Pecora, Lin.). In vols. 4 and 5 of E. Griffith, W. of the family Bovidae (Artiodactyla) and the evolution
Pidgeon & C. Hamilton Smith, Cuvier’s Animal of the mitochondrial cytochrome b gene. Molecular
Kingdom [t.p. of vol. 5: Synopsis of the Species of the Phylogenetics and Evolution 13:227–243.
Class Mammalia]. London: Whittaker, 33– 428. Hassanin, A. & E.J.P. Douzery. 1999b. Evolutionary
Hammer, S.E., H.M. Schwammer & F. Suchentrunk. 2008. affi nities of the enigmatic saola (Pseudoryx nghetinhen-
Evidence for introgressive hybridisation of captive sis) in the context of the molecular phylogeny of
markhor (Capra falconeri) with domestic goat: cautions Bovidae. Proceedings of the Royal Society of London, —-1
for reintroduction. Biochemical Genetics 46:216–226. B 266:893– 900. —0
349-47558_ch01_1P.indd 293 5/31/11 7:06 PM

Hassanin, A. & E.J.P. Douzery. 2003. Molecular and Collections, 61, No. 22. Washington, DC: Smithsonian
morphological phylogenies of Ruminantia and the Institution.
alternative position of the Moschidae. Systematic Hemmer, H. 1983 [English translation 1990]. Domestica-
Biology 52:206–228. tion: The Decline of Environmental Appreciation.
Hassanin, A. & A. Ropiquet. 2004. Molecular phylogeny of Brunswick, Germany: Vieweg [English translation by
the tribe Bovini (Bovidae, Bovinae) and the taxonomic Neil Beckhaus, Cambridge: Cambridge University
status of the kouprey, Bos sauveli Urbain 1937. Molecu- Press].
lar Phylogenetics and Evolution 33:896– 897. Heptner, V., A.A. Nasimovich & A.G. Bannikov. 1961
Hassanin, A. & A. Ropiquet. 2007. What is the taxonomic [English translation 1988]. Artiodactyla and Perisso-
status of the Cambodian banteng and does it have close dactyla. Vol. 1 of Mammals of the Soviet Union.
genetic links with the kouprey? Journal of Zoology, Moscow: Vysshaya Shkola [English translation,
London 271:246–252. Washington, DC: Smithsonian Institution Libraries
Hassanin, A., E. Pasquet & J.-D. Vigne. 1998. Molecular and National Science Foundation].
systematics of the subfamily Caprinae (Artiodactyla, Herring, S.W. 1972. The role of canine morphology in the
Bovidae) as determined from cytochrome b sequences. evolutionary divergence of pigs and peccaries. Journal
Journal of Mammalian Evolution 5:217–236. of Mammalogy 53:500–512.
Hassanin, A., A. Seveau, H. Thomas, H. Bocherens, Herring, S.W. 1974. A biometric study of suture fusion and
D. Billiou & B.X. Nguyen. 2001. Evidence that the skull growth in peccaries. Anatomy and Embryology
mysterious “linh duong” (Pseudonovibos spiralis) is not a 146:167–180.
new bovid. Comptes Rendus de l’Académie des Hershkovitz, P. 1957. The type locality of Bison bison
Sciences (Paris), Sér. 3, Sciences de la Vie 324:71– 80. Linnaeus. Proceedings of the Biological Society of
Hassanin, A., A. Ropiquet, R. Cornette, M. Tranier, P. Washington 70:31–32.
Pfeffer, P. Candegabe & M. Lemaire. 2006. Has the Hershkovitz, P. 1963. The nomenclature of South
kouprey (Bos sauveli Urbain, 1937) been domesticated in American peccaries. Proceedings of the Biological
Cambodia? Comptes Rendus de l’Académie des Society of Washington 76:85– 88.
Sciences (Paris), Biologies 329:124–135. Hershkovitz, P. 1982. Neotropical deer (Cervidae), part 1:
Hassanin, A., A. Ropiquet, A.-L. Gourmand, B. Chardon- pudus, genus Pudu Gray. Fieldiana Zoology (n.s.) 11:1–86.
net & J. Rigoulet. 2007. Mitochondrial DNA variability Heude, P.-M. 1894a. Notes sur le genre Capricornis (Ogilby,
in Giraffa camelopardalis: consequences for taxonomy, 1836). Mémoires Concernant l’Histoire Naturelle de
phylogeography, and conservation of giraffes in west l’Empire Chinois 2:222–233.
and central Africa. Comptes Rendus de l’Académie des Heude, P.-M. 1894b. Notes sur le genre Kemas (Ogilby, 1836).
Sciences (Paris), Biologies 330:265–274. Mémoires Concernant l’Histoire Naturelle de l’Empire
Hayman, R.W. 1961. The red goral of the north-east Chinois 2:234–245.
frontier region. Proceedings of the Zoological Society Heude, P.-M. 1897a. Notes sur le quatrième capricorne de
of London 136:317–324. Tonkin. Mémoires Concernant l’Histoire Naturelle de
Helgen, K.M. 2003. Major mammalian clades: a review l’Empire Chinois 3:151–155.
under consideration of molecular and palaeontological Heude, P.-M. 1897b. Capricornes de Se-tchouan. Mémoires
evidence. Mammalian Biology 68:1–15. Concernant l’Histoire Naturelle de l’Empire Chinois
Heller, E. 1910. A new sable antelope from British East 3:196–198.
Africa. Smithsonian Miscellaneous Collections 54, Heude, P.-M. 1898. Capricornes de Moupin, 1. Mémoires
No. 6. Washington, DC: Smithsonian Institution. Concernant l’Histoire Naturelle de l’Empire Chinois
Heller, E. 1912. New genera and races of African ungu- 4:1– 9.
lates. Smithsonian Miscellaneous Collections 60, No. 8. Heude, P.-M. 1899. Capricorne de la Chine méridionale.
Washington, DC: Smithsonian Institution. Mémoires Concernant l’Histoire Naturelle de l’Empire
Heller, E. 1913a. New races of antelopes from British East Chinois 4:211.
Africa. Smithsonian Miscellaneous Collections 61, Hillman-Smith, A.K.K. & C.P. Groves. 1994. Diceros
No. 7. Washington, DC: Smithsonian Institution. bicornis. Mammalian Species 455:1– 8.
Heller, E. 1913b. New antelopes and carnivores from Hla Aung, S. 1967. Observations on the Red goral
British East Africa. Smithsonian Miscellaneous Nemorhaedus cranbrooki and the Burmese takin
Collections 61, No. 13. Washington, DC: Smithsonian Budorcas t. taxicolor at Rangoon Zoo. International Zoo
Institution. Yearbook 7: 225–256.
Heller, E. 1913c. New races of ungulates and primates Ho, S.Y.W., M.J. Phillips, A. Cooper & A.J. Drummond.
from equatorial Africa. Smithsonian Miscellaneous 2005. Time dependency of molecular rate estimates
Collections 61, No. 17. Washington, DC: Smithsonian and systematic overestimation of recent divergence
Institution. times. Molecular Biology and Evolution 22:1561–1568.
-1— Heller, E. 1914. Four new subspecies of large mammals Hodgson, B.H. 1831. Contributions in natural history.
0— from equatorial Africa. Smithsonian Miscellaneous Gleanings in Science 3:320–324.
349-47558_ch01_1P.indd 294 5/31/11 7:06 PM

ref erences 295
Hodgson, B.H. 1832. Mammalia and birds collected in Imaizumi, Y. 1970. Description of a new species of Cervus
Nepal by B. H. Hodgson. Proceedings of the Zoologi- from the Tsushima Islands, Japan, with a revision of
cal Society of London, part 2 (1832):10–16. the subgenus Sika based on clinal analysis. Bulletin of
Hodgson, B.H. 1834. On the Mammalia of Nepal. the National Science Museum (Tokyo) 13:185–194.
Proceedings of the Zoological Society of London, Ingles, J.M. 1965. Zambian mammals collected for the
part 2 (1834):95– 99. British Museum (Natural History) in 1962. Puku
Holbrook, L.T. 2002. The unusual development of the 3:75– 86.
sagittal crest in the Brazilian tapir (Tapirus terrestris). International Commission on Zoological Nomenclature
Journal of Zoology, London 256:215–219. [ICZN], Opinion 460. 1957. Validation under the
Honey, J.G. 2007. Family Camelidae. In D.R. Prothero & plenary powers of the generic name of “Muntiacus”
S.E. Foss (eds.), The Evolution of Artiodactyls. Rafi nesque, 1815, and designation for the genus so
Baltimore: Johns Hopkins University Press, 177–188. named of a type species in harmony with accustomed
van Hooft, W.F., A.F. Groen & H.H.T. Prins. 2002. usage (Class Mammalia). Opinions and Declarations
Phylogeography of the African buffalo based on Rendered by the International Commission on
mitochondrial and Y-chromosomal loci: Pleistocene Zoological Nomenclature 15:457– 474.
origin and population expansion of the Cape buffalo ICZN, Opinion 1348. 1985. Bos gaurus H. Smith, 1827
subspecies. Molecular Ecology 11:267–279. (Mammalia, Artiodactyla): conserved. Bulletin of
Howell, A.B. 1930. Mammals from China in the Zoological Nomenclature 42:279–280.
collection of the United States National Museum. ICZN, Opinion 1349. 1985. Antilope depressicornis H. Smith,
Proceedings of the United States National Museum 1827 and Anoa quarlesi Ouwens, 1910 (Mammalia,
75:1– 82. Artiodactyla): conserved. Bulletin of Zoological
Hrabe, V. & P. Koubek. 1984. Craniometrical characteris- Nomenclature 42:281–282.
tics of Rupicapra rupicapra tatrica (Mamm., Bovidae). ICZN, Opinion 1894. 1998. Regnum Animale . . . , ed. 2
Folia Zoologica 33:73– 84. (M.-J. Brisson, 1762): rejected for nomenclatural
Hu J.,.S-g. Fang & Q-h. Wan. 2006. Genetic diversity of purposes, with the conservation of the mammalian
Chinese water deer (Hydropotes inermis inermis): generic names Philander (Marsupialia), Pteropus
implications for conservation. Biochemical Genetics (Chiroptera), Glis, Cuniculus, and Hydrochoerus
44:156–167. (Rodentia), Meles, Lutra, and Hyaena (Carnivora),
Huang, L., W. Nie, J. Wang, W. Su & F. Yang. 2005. Tapirus (Perissodactyla), Tragulus and Giraffa (Artiodac-
Phylogenomic study of the subfamily Caprinae by tyla). Bulletin of Zoological Nomenclature 55:64–71.
cross-species chromosome painting with Chinese ICZN, Opinion 2027. 2003. Case 3010: usage of 17 specific
muntjac paints. Chromosome Research 13:389–399. name was based on wild species which are pre-dated
Huang, L., J. Wang, W. Nie, W. Su & F. Yang. 2006. by or contemporary with those based on domestic
Tandem chromosome fusions in karyotypic evolution animals (Lepidoptera, Osteichthyes, Mammalia);
of Muntiacus: evidence from M. feae and M. gongshanen- conserved. Bulletin of Zoological Nomenclature
sis. Chromosome Research 14:637– 647. 60:81– 84.
d’Huart, J.-P. & P. Grubb. 2001. Distribution of the Irwin, A. J. 1914. Notes on the races of serow, or goat-
common warthog (Phacochoerus africanus) and the antelope, found in Siam. Journal of the Natural
desert warthog (Phacochoerus aethiopicus) in the Horn History Society of Siam 1, 1:19–26.
of Africa. African Journal of Ecology 39:156–169. Iyengar, A., F.M. Diniz, T. Gilbert, D. Woodfi ne,
d’Huart, J.-P. & P. Grubb. 2005. A photographic guide J. Knowles & N. Maclean. 2006. Structure and
to the differences between the common warthog evolution of the mitochondrial control region in oryx.
(Phacochoerus africanus) and the desert warthog (Ph. Molecular Phylogenetics and Evolution 40:305–314.
aethiopicus). Suiform Soundings 5, 2:4– 8. Jacobi, A. 1923. Zoologische Ergebnisse der Walter
Hundertmark, K.J., G.F. Shields, R.T. Bowyer & C.C. Stötznerschen Expeditionen nach Szetschwan,
Schwartz. 2002. Genetic relationships deduced from Osttibet, und Tschili auf Grund der Sammlungen und
cytochrome b sequences among moose. Alces Beobachtungen Dr. Hugo Weigolds: Mammalia.
38:113–122. Abhandlungen und Berichtung des Museen für
Hundertmark, K.J., G.F. Shields, I.G. Udina, R.T. Bowyer, Tierkunde und Völkerkunde Dresden 16:1–22.
A.A. Danilkin & C.C. Schwartz. 2002. Mitochondrial Janeck, L.L., R.L. Honeycutt, R.M. Adkins & S.K. Davis.
phylogeography of moose (Alces alces): Late Pleistocene 1996. Mitochondrial gene sequences and the molecular
divergence and population expansion. Molecular systematics of the artiodactyl subfamily Bovinae.
Phylogenetics and Evolution 22:375–387. Molecular Phylogenetics and Evolution 6:107–119.
Hundertmark, K.J., R.T. Bowyer, G.F. Shields & C.C. Janis, C.M. & K.M. Scott. 1987. The interrelationships of
Schwartz. 2003. Mitochondrial phylogeography of higher ruminant families with special emphasis on the
moose (Alces alces) in North America. Journal of members of the Cervoidea. American Museum —-1
Mammalogy 84:718–728. Novitates 2893:1– 85. —0
349-47558_ch01_1P.indd 295 5/31/11 7:06 PM

Jansen van Vuuren, B. & T.J. Robinson. 2001. Retrieval of Kingdon, J. 1979. Large Mammals. Vol. 3B of East African
four adaptive lineages in duiker antelope: evidence Mammals: An Atlas of Evolution in Africa. London:
from mitochondrial DNA sequences and fluorescence Academic Press.
in situ hybridisation. Molecular Phylogenetics and Kingdon, J. 1982a. Bovids. Vol. 3C of East African
Evolution 20:409– 425. Mammals: An Atlas of Evolution in Africa. London:
Jarman, P.J. 1972. The development of a dermal shield in Academic Press.
impala. Journal of Zoology, London 166:349–356. Kingdon, J. 1982b. Bovids. Vol. 3D of East African
Jarman, P.J. 1974. The social organisation of antelope in Mammals: An Atlas of Evolution in Africa. London:
relation to their ecology. Behaviour 48:215–266. Academic Press.
Jarman, P.J. 2000. Dimorphism in social Artiodactyla: Kingdon, J. 1997. The Kingdon Field Guide to African
selection upon females. In E.S. Vrba & G.B. Schaller Mammals. San Diego: Academic Press.
(eds.), Antelopes, Deer, and Relatives New Haven: Yale Kingswood, S.C. & D.A. Blank. 1996. Gazella subgutturosa.
University Press, 171–179. Mammalian Species 518:1–10.
Jass, C.M. & J.I. Mead. 2004. Capricornis crispus. Mamma- Kingswood, S.C. & A.T. Kumamoto. 1988. Research and
lian Species 750:1–10. management of Arabian sand gazelle in the USA. In A.
Jiang Z-g., Z-j. Feng, Z-w. Wang, L-w. Chen, P. Cai & Y-b. Dixon & D. Jones (eds.), Conservation and Biology of
Li. 1995. Historical and current distributions of Desert Antelopes. London: Christopher Helm, 212–226.
Przewalski’s gazelles. Acta Theriologica Sinica Kingswood, S.C., A.G. Kumamoto, S.J. Charter, R.A.
15:241–245. Aman & O.A. Ryder. 1998. Centric fusion polymor-
Jiménez, J.E. 2010. Southern pudu Pudu puda (Molina phisms in waterbuck (Kobus ellipsiprymnus). Journal of
1782). In J.M.B. Duarte & S. González (eds.), Heredity 89:96–100.
Neotropical Cervidology: Biology and Medicine of Kingswood, S.C., A.T. Kumamoto, S.J. Charter & M.L.
Latin American Deer. Jaboticabal, Brazil and Gland, Jones. 1998. Cryptic chromosomal variation in suni
Switzerland: Funep, in collaboration with IUCN, Neotragus moschatus (Artiodactyla, Bovidae). Animal
140–150. Conservation 1:95–100.
Jorge, W. & K. Benirschke. 1976. Banding and meiotic Klein, D.R., M. Meldgaard & S.G. Fancy. 1987. Factors
chromosome studies in a male eland. Genen en determining leg length in Rangifer tarandus. Journal of
Phaenen 19:7–10. Mammalogy 68:642– 655.
Jorge, W., S. Butler & K. Benirschke. 1976. Studies on a Kloss, C.B. 1919. On mammals collected in Siam. Journal
male eland × kudu hybrid. Journal of Reproduction and of the Natural History Society of Siam 3:333– 407.
Fertility 46:13–16. Kloss, C.B. 1923. The goral in Siam. Journal of the Natural
Karami, M. & C.P. Groves. 1993. A mammal species new History Society of Siam 6, 1:135–137.
for Iran: Gazella gazella Pallas, 1766 (Artiodactyla: Kock, D. 2000. The fallow deer Dama schaeferi Hilzheimer,
Bovidae). Journal of Sciences of the Islamic Republic of 1926 (Mammalia: Cervidae), enigmatic and forgotten.
Iran 4:81– 89. Zoology in the Middle East 21:9–11.
Karami, M., M.R. Hemami & C.P. Groves. 2002. Taxo- Koh, H.S. & E. Randi. 2001. Genetic distinction of roedeer
nomic, distributional, and ecological data on gazelles (Capreolus pygargus Pallas) sampled in Korea. Mamma-
in Iran. Zoology in the Middle East 26:29–36. lian Biology 66:371–375.
Kavar, T. & P. Dovc. 2008. Domestication of the horse: Korobitsyna, K.V., C.F. Nadler, N.N. Vorontsov & R.S.
genetic relationships between domestic and wild Hoff mann. 1974. Chromosomes of the Siberian snow
horses. Livestock Science 116:1–14. sheep, Ovis nivicola, and implications concerning the
Khaute, L.M. 2010. The Sangai: The Pride of Manipur. origin of Amphiberingian wild sheep (subgenus
Delhi: Kalpaz. Pachyceros). Quaternary Research 4:235–245.
Kikkawa, Y., H. Yonekawa, H. Suzuki & T. Amano. 1997. Kraus, F. & M.M. Miyamoto. 1991. Rapid cladogenesis
Analysis of genetic diversity of domestic water among the pecoran ruminants: evidence from
buffaloes and anoas based on variations in the mitochondrial DNA sequences. Systematic Zoology
mitochondrial gene for cytochrome b. Animal 40:117–130.
Genetics 28:195–201. Kretzoi, M. 1946. On Bison bonasus hungarorum n. ssp.
Kim, K.S., K. Tanaka, D.B. Ismail, S. Maruyama, H. Annales Historico-Naturales Musei Nationalis
Matsubayashi, H. Endo, K. Fukuta & J. Kimura. 2004. Hungarici 39:105–191.
Cytogenetic comparison of the lesser mouse deer Krüger, K., C. Gaillard, G. Stranzinger & S. Rieder. 2005.
(Tragulus javanicus) and the greater mouse deer Phylogenetic analysis and species allocation of
(T. napu). Caryologia 57:229–243. individual equids using microsatellite data. Journal of
King, D.G. 1975. The Afro-Alpine grey duiker of Kiliman- Animal Breeding and Genetics 122, Suppl. 1:78– 86.
jaro. Journal of the East Africa Natural History Society Krumbiegel, I. 1944. Die neuweltlichen Tylopoden.
-1— and National Museum, 152:1– 9. Zoologischer Anzeiger 145:45–70.
0—
349-47558_ch01_1P.indd 296 5/31/11 7:06 PM

ref erences 297
Krumbiegel, I. 1952. Lamas. Die Neue Brehm-Bücherei No. Lalueza-Fox, C., B. Shapiro, P. Bover, J.A. Alcover & J.
54. Wittenberg-Lutherstadt, Germany: A. Ziemsen. Bertranpetit. 2002. Molecular phylogeny and evolution
Krumbiegel, I. 1980. Die unterartliche Trennung des of the extinct bovid Myotragus balearicus. Molecular
Bisons, Bison bison (Linné, 1788), und seine Rückzüch- Phylogenetics and Evolution 25:501–510.
tung. Säugetierkundliche Mitteilungen 28:148–160. Lalueza-Fox, C., J. Castresana, L. Sambietro, T. Marquès-
Kruska, D.C.T. 2005. On the evolutionary significance of Bonet, J.A. Alcover & J. Bertranpetit. 2005. Molecular
encephalization in some eutherian mammals: effects of dating of caprines using ancient DNA sequences of
adaptive radiation, domestication, and feralization. Myotragus balearicus, an extinct endemic Balearic
Brain Behavior and Evolution 65:73–108. mammal. BMC Evolutionary Biology 2005, 5:70. doi:
Kruska, D.C.T. 2007. The effects of domestication on brain 10.1186/1471-2148-5-70.
size in the evolution of ner vous systems. In J.H. Kaas Lan H., W. Wang & L-m. Shi. 1995. Phylogeny of
(ed.), Mammals. Vol. 3 of Evolution of Ner vous Muntiacus (based on mitochondrial DNA restriction
Systems. London: Elsevier, 143–153. maps). Biochemical Genetics 33:377–388.
Kuhn, H.-J. 1968. A provisional checklist of the mammals Lange, J. 1970. Ein Beitrag zur phylogenetischen Stellung
of Liberia. Senckenbergiana Biologica 46:321–340. de Springbockes (Antidorcas marsupialis Sundevall,
Kumamoto, A.T. & M.H. Bogart. 1984. The chromosomes 1847). Zeitschrift für Säugetierkunde 35:65–75.
of Cuvier’s gazelle. In O.A. Ryder & M.L. Byrd (eds.), Lange, J. 1971. Ein Beitrag zur systematischen Stellung der
One Medicine: Essays in Honour of Kurt Benirschke. Spiegelgazellen (genus Gazella Blainville, 1816
Berlin: Springer-Verlag, 101–108. subgenus Nanger Lataste, 1885). Zeitschrift für
Kumamoto, A.T., S.C. Kingswood & W. Hugo. 1994. Säugetierkunde 36:1–18.
Chromosomal divergence in allopatric populations of Lange, J. 1972. Studien an Gazellenschädeln: ein Beitrag
Kirk’s dik-dik, Madoqua kirkii (Artiodactyla, Bovidae). zur Systematik der kleineren Gazellen, Gazella (de
Journal of Mammalogy 75:357–364. Blainville, 1816). Säugetierkundliche Mitteilungen
Kumamoto, A.T., S.C. Kingswood, W.E.R. Rebholz & 20:193–249.
M.L. Houck. 1995. The chromosomes of Gazella Larson, G., K. Dobney, U. Albarella, M. Fang, E. Matisoo-
bennetti and Gazella saudiya. Zeitschrift für Smith, J. Robins, S. Lowden, H. Finlayson, T. Brand, E.
Säugetierkunde 60:159–169. Willerslev, P. Rowley- Conwy, L. Andersson & A.
Kumamoto, A.T., S.J. Charter, M.L. Houck & M. Frahm. Cooper. 2007. Worldwide phylogeography of wild boar
1996. Chromosomes of Damaliscus (Artiodactyla: reveals multiple centers of pig domestication. Science
Bovidae): simple and complex centric fusion rearrange- 307:1618–1621.
ments. Chromosome Research 4:614– 621. Larson, G., T. Cucchi, M. Fujita, E. Matisoo-Smith, J.
Kumamoto, A.T., S.J. Charter, S.C. Kingswood, O.J. Ryder Robins, A. Anderson, E. Rolett, M. Spriggs, G.
& D.S. Gallagher Jr. 1999. Centric fusion differences Dolman, T. Djubiantono, P. Griffi n, M. Intoh, E.
among Oryx dammah, O. gazella, and O. leucoryx Keane, P. Kirch, K-t. Li, M. Morwood, L.M. Pedriña,
(Artiodactyla, Bovidae). Cytogenetics and Cell P.J. Piper, R.J. Rabett, P. Shooter, G. Van den Bergh, E.
Genetics 86:74– 80. West, S. Wickler, J. Yuan, A. Cooper & K. Dobney.
Kumar, S., M. Nagarajan, J.S. Sandhu, N. Kumar, V. Behl 2007. Phylogeny and ancient DNA of Sus provides
& G. Nishanth. 2007. Mitochondrial DNA analyses of insights into Neolithic expansion in Island Southeast
Indian water buffalo support a distinct genetic origin Asia and Oceania. PNAS: Proceedings of the National
of river and swamp buffalo. Animal Genetics Academy of Science USA 104:4834– 4839.
38:227–232. Latch, E.K., J.R. Heffelfi nger, J.A. Fike & O.E. Rhodes Jr.
Kumerloeve, H. 1969. Bemerkungen zum Gazellen- 2009. Species-wide phylogeography of North American
Vorkommen im südöstlichen Kleinasien. Zeitschrift mule deer (Odocoileus hemionus). Molecular Ecology
für Säugetierkunde 34:113–120. 18:1730–1745.
Künzel, T. & S. Künzel. 1998. An overlooked population von Lehmann, E. & H. Sägesser. 1981. Capreolus capreolus
of the beira antelope Dorcatragus megalotis in Djibouti. Linnaeus, 1758—Reh. In J. Niethammer & F. Krapp
Oryx 32:75– 80. (eds.), Paarhufer, Artiodactyla: Suidae, Cervidae,
Kuznetsova, M.V. & M.V. Kholodova. 2003. Revision of Bovidae. Vol. 2, pt. 2 of Handbuch der Säugetiere
phylogenetic relationships in the Antilopinae subfam- Europas. Weisbaden, Germany: Aula, 233–266.
ily on the basis of the mitochondrial rRNA and Lei R-h., Z-g. Jiang, Z. Hu & W-l. Yang. 2003. Phylogenetic
b-spectrin nuclear gene sequences. Doklady Biological relationships of Chinese antelopes (subfamily Antilopi-
Sciences 391:333–336. nae) based on mitochondrial ribosomal RNA gene
Kuznetsova, M.V., M.V. Kholodova & A.A. Luschekina. sequences. Journal of Zoology, London 261:227–237.
2002. Phylogenetic analysis of sequences of the 12S and Lei R-h., Z. Hu, Z-g. Jiang & W-l. Yang. 2003. Phylogeog-
16S rRNA mitochondrial genes in the family Bovidae: raphy and genetic diversity of the critically endangered
new evidence. Genetika 38:942– 950. Przewalski’s gazelle. Animal Conservation 6:361–367. —-1
—0
349-47558_ch01_1P.indd 297 5/31/11 7:06 PM

Leinders, J.J.M. 1979. On the osteology and function of the Synonymis, Locis. 10th ed., rev., vol. 1. Stockholm:
digits of some ruminants and their bearing on Laurent Salvius.
taxonomy. Zeitschrift für Säugetierkunde 44:305–318. Linnaeus (von Linné), C. 1766. Systema Naturae per
Leinders, J.J.M. & E. Heintz. 1980. The configuration of Regna Tria Naturae, secundum Classes, Ordines,
the lacrimal orifices in pecorans and tragulids Genera, Species, cum Characteribus, Differentiis,
(Artiodactyla, Mammalia) and its significance for the Synonymis, Locis. 12th ed., rev., vol. 1. Stockholm:
distinction between Bovidae and Cervidae. Beaufortia Laurent Salvius.
30:155–162. Lister, A.M, P. Grubb & S.R.M. Sumner. 1998. Taxonomy,
Leonard, J.A., N. Rohland, S. Glaberman, R.C. Fleischer, morphology, and evolution of Eu ropean roe deer. In
A. Caccone & M. Hofreiter. 2005. A rapid loss of R. Andersen, P. Duncan & J.D.C. Linnell (eds.), The
stripes: the evolutionary history of the extinct quagga. European Roe Deer: The Biology of Success. Oslo:
Biology Letters 1:291–295. Scandinavian University Press, 23– 46.
Leslie, D.M. 2009. Przewalskium albirostre. Mammalian Liu X-h., Y-q. Wang, Z-q. Liu & K-y. Zhou. 2003.
Species 849:7–18. Phylogenetic relationships of Cervinae based on
Leslie, D.M. & G.B. Schaller. 2008. Pantholops hodgsonii. sequence of mitochondrial cytochrome b gene.
Mammalian Species 817:1–13. Zoological Research 24:27–33.
Leslie, D.M. & K. Sharma. 2009. Tetracerus quadricornis Lizcano, D.J., S.J. Álvarez & C.A. Delgado-V. 2010. Dwarf
(Artiodactyla: Bovidae). Mammalian Species 843:1–11. red brocket deer Mazama rufina (Pucheran 1951). In
Leus, K., G.P. Goodall & A.A. Macdonald. 1999. Anatomy J.M.B. Duarte & S. González (eds.), Neotropical
and histology of the babirusa (Babyrousa babyrussa) Cervidology: Biology and Medicine of Latin American
stomach. Comptes Rendus de l’Académie des Sciences Deer. Jaboticabal, Brazil and Gland, Switzerland:
(Paris), Sér. 3, Sciences de la Vie 322:1081–1092. Funep, in collaboration with IUCN, 177–180.
Leuthold, W. 1981. Contact between formerly allopatric Lizcano, D.J., E. Yerena, S.J. Álvarez & J.R. Dietrich. 2010.
subspecies of Grant’s gazelle (Gazella granti Brooke Mérida brocket deer Mazama bricenii (Thomas 1908). In
1872) owing to vegetation changes in Tsavo National J.M.B. Duarte & S. González (eds.), Neotropical
Park, Kenya. Zeitschrift für Säugetierkunde 46:48–55. Cervidology: Biology and Medicine of Latin American
Li J-x. & L-h. Xu. 1996. A new subspecies of the Indian Deer. Jaboticabal, Brazil and Gland, Switzerland:
muntjac (Muntiacus muntjak) in Guangdong, China. Funep, in collaboration with IUCN, 181–184.
Acta Theriologica Sinica 16:25–29. Loehr, J., K. Worley, A. Grapputo, J. Carey, A. Veitch &
Li M. & H-l. Sheng. 1998. MtDNA deference and molecular D.W. Coltman. 2006. Evidence for cryptic glacial
phylogeny among musk deer, Chinese water deer, refugia from North American mountain sheep
muntjak, and deer. Acta Theriologica Sinica 18:184–191. mitochondrial DNA. Journal of Evolutionary Biology
Li M., X-m. Wang, H-l. Sheng, H. Tamate, R. Masuda, J. 19:419– 430.
Nagata & N. Ohtaishi. 1998. Origin and genetic Loftus, R.T., D.E. MacHugh, D.G. Bradley, P.M. Sharp &
diversity of four subspecies of red deer (Cervus elaphus). P. Cunningham. 1994. Evidence for two independent
Zoological Research 19:177–183. domestications of cattle. PNAS: Proceedings of the
Li M.,Y-g. Li, H-l. Sheng, H. Tamate, R. Masuda, J. Nagata National Academy of Science USA 91:2757–2761.
& N. Ohtaishi. 1999. The taxonomic status of Moschus Lönnberg, E. 1917. Mammals Collected in Central Africa
moschiferus anhuiensis. Chinese Science Bulletin by Captain E. Arrhenius. Kungliga Svenska Vetenskap-
44:719–722. sakademien Handlingar No. 58. Stockholm: Almqvist
Li M., F-w. Wei, P. Groves, Z-j. Feng & J-c. Hu. 2003. & Wiksell.
Genetic structure and phylogeography of the takin Lorenzen, E.D. & H.R. Siegismund. 2004. No suggestion
(Budorcas taxicolor) as inferred from mitochondrial of hybridisation between the vulnerable black-faced
DNA sequences. Canadian Journal of Zoology impala (Aepyceros melampus petersi) and the common
81:462– 468. impala (A. m. melampus) in Etosha National Park,
Li Z-x. 1981. On a new species of musk deer from China. Namibia. Molecular Ecology 13:3007–3019.
Zoological Research 2:157–161. Lorenzen, E.D., B.T. Simonsen, P.W. Kat, P. Arctander &
Lichtenstein, H. & W. Peters. 1855. Über Antilope leucotis H.R. Siegismund. 2006. Hybridisation between
Licht. Pet. In Über Neue Merkwürdige Säugethiere des subspecies of waterbuck (Kobus ellipsiprymnus) in zones
Königlichen Zoologischen Museums. Berlin: Akad- of overlap with limited introgression. Molecular
emie der Wissenschaften, 14–20. Ecology 15:3787–3799.
Lindsey, S.L., M.N. Green & C.L. Bennett. 1999. The Lorenzen, E.D., P. Arctander & H.R. Siegismund. 2006.
Okapi: Mysterious Animal of Congo-Zaire. Austin: Regional genetic structuring and evolutionary history
University of Texas Press. of the impala Aepyceros melampus. Journal of Heredity
Linnaeus (von Linné), C. 1758. Systema Naturae per 97:119–132.
-1— Regna Tria Naturae, secundum Classes, Ordines, Lorenzen, E.D., R. de Neergaard, P. Arctander & H.R.
0— Genera, Species, cum Characteribus, Differentiis, Siegismund. 2007. Phylogeography, hybridisation, and
349-47558_ch01_1P.indd 298 5/31/11 7:06 PM

ref erences 299
Pleistocene refugia of the kob antelope (Kobus kob). Lydekker, R. 1915a. Artiodactyla, Families Cervidae
Molecular Ecology 16:3241–3252. (Deer), Tragulidae (Chevrotains), Camelidae (Camels
Lorenzen, E.D., P. Arctander & H.R. Siegismund. 2008a. and Llamas), Suidae (Pigs and Peccaries), and Hippo-
Three reciprocally monophyletic mtDNA lineages potamidae (Hippopotamuses). Vol. 4 of Cata logue of
elucidate the taxonomic status of Grant’s gazelles. the Ungulate Mammals in the British Museum.
Conservation Genetics 9:593– 601. London: British Museum (Natural History).
Lorenzen, E.D., P. Arctander & H.R. Siegismund. 2008b. Lydekker, R. 1915b. Perissodactyla (Horses, Tapirs,
High variation and very low differentiation in Rhinoceroses), Hyracoidea (Hyraxes), Proboscidea
wide-ranging plains zebra (Equus quagga): insights from (Elephants): With Addenda to the Earlier Volumes.
mtDNA and microsatellites. Molecular Ecology Vol. 5 of Cata logue of the Ungulate Mammals in the
17:2812–2824. British Museum. London: British Museum (Natural
Loubser, J., J. Brink & G. Laurens. 1990. Paintings of the History).
extinct blue antelope, Hippotragus leucophaeus, in the Lydekker, R. & G. Blaine. 1914a. Artiodactyla, Family
eastern Orange Free State. South African Archaeologi- Bovidae, Subfamilies Bubalinae to Reduncinae
cal Bulletin 45:106–111. (Hartebeests, Gnus, Duikers, Dik-Diks, Klipspringers,
Lovari, S. 1977. The Abruzzo chamois. Oryx 14:47–50. Reedbucks, Waterbucks, etc). Vol. 2 of Cata logue of the
Lovari, S. 1987. Evolutionary aspects of the biology of Ungulate Mammals in the British Museum. London:
chamois, Rupicapra spp. (Bovidae, Caprinae). In H. British Museum (Natural History).
Soma (ed.), The Biology and Management of Capricor- Lydekker, R. & G. Blaine. 1914b. Artiodactyla, Family
nis and Related Mountain Antelopes. London: Croom Bovidae, Subfamilies Aepycerotinae to Tragelaphinae
Helm, 51– 61. (Pala, Saiga, Gazelles, Oryx Group, Bushbucks, Kudus,
Lovari, S. & C. Scala. 1980. Revision of Rupicapra genus, 1: Elands, etc.), Antilocapridae (Prongbuck), and
a statistical re-evaluation of Couturier’s data on the Giraffidae (Giraffes and Okapi). Vol. 3 of Cata logue of
morphometry of six chamois subspecies. Bolletino the Ungulate Mammals in the British Museum.
Zoologico 47:113–124. London: British Museum (Natural History).
Lovari, S. & C. Scala. 1984. Revision of Rupicapra genus, 4: Ma S-l., Y-x. Wang & L-m. Shi. 1990. A new species of the
horn biometrics of Rupicapra rupicapra asiatica and its genus Muntiacus from Yunnan, China. Zoological
relevance to the taxonomic position of Rupicapra Research 11:48–53.
rupicapra caucasica. Zeitschrift für Säugetierkunde Machado A. de Barros. 1969. Mamíferos de Angola Ainda
49:246–253. Não Citados ou Pouco Conhecidos. Publicações
Lowe, V.P.W. & A.S. Gardiner. 1975. Hybridisation Culturais da Companhia de Diamantes de Angola,
between red deer (Cervus elaphus) and sika deer (Cervus Serviços Culturais, Dundo-Lunda-Angola No. 46.
nippon) with par ticu lar reference to stocks in N.W. Lisbon: Companhia de Diamantes de Angola, 93–232.
England. Journal of Zoology, London 177:553–566. Malbrant, R. & A. Maclatchy. 1949. Mammifères. Vol. 2 of
Ludt, C.J., W. Schroeder, O. Rottmann & R. Kuehn. 2004. Faune de 1’Équateur Africain Français. Paris:
Mitochondrial DNA phylogeography of red deer Lechevalier.
(Cervus elaphus). Molecular Phylogenetics and Evolu- Manceau, V., L. Després, J. Bouvet & P. Taberlet. 1999.
tion 31:1064–1083. Systematics of the genus Capra inferred from mito-
Lundrigan, B. 1996. Morphology of horns and fighting chondrial DNA sequence data. Molecular Phylogene-
behaviour in the family Bovidae. Journal of Mammal- tics and Evolution 13:504–510.
ogy 77:462– 475. Manceau, V., J.-P. Crampe, P. Boursot & P. Taberlet. 1999.
Lyapunova, E.A., T.D. Bunch, N.N. Vorontsov & R.S. Identification of evolutionary significant units in the
Hoff mann. 1997. Chromosome sets and the taxonomy Spanish wild goat, Capra pyrenaica (Mammalia,
of Severtsov wild sheep (Ovis ammon severtzovi). Artiodactyla). Animal Conservation 2:33–39.
Russian Journal of Zoology 1:387–396. Marcot, J.D. 2007. Molecular phylogeny of terrestrial
Lydekker, R. 1900. Great and Small Game of India. artiodactyls: confl icts and resolution. In D.R. Prothero
London: Rowland Ward. & S.E. Foss (eds.), The Evolution of Artiodactyls.
Lydekker, R. 1905. The gorals of India and Burma. Baltimore: Johns Hopkins University Press, 4–18.
Zoologist 9:81– 84. Marín, J.C., C.S. Casey, M. Kadwell, K. Yaya, D. Hoces,
Lydekker, R. 1906. The white-maned serow. Proceedings J. Olazabal, R. Rosadio, J. Rodriguez, A. Spotorno,
of the Zoological Society of London 1905, 2:329–331. M.W. Bruford & J.C. Wheeler. 2007. Mitochondrial
Lydekker, R. 1913. Artiodactyla, Family Bovidae, phylogeography and demographic history of the
Subfamilies Bovinae to Ovibovinae (Cattle, Sheep, vicuña: implications for conservation. Heredity
Goats, Chamois, Serows, Takin, Musk- Oxen, etc.). 99:70– 80.
Vol. 1 of Cata logue of the Ungulate Mammals in the Marín, J.C., A.E. Spotorno, B.A. González, C. Bonacic,
British Museum. London: British Museum (Natural J.C. Wheeler, C.S. Casey, M.W. Bruford, R.E. Palma & —-1
History). E. Poulin. 2008. Mitochondrial DNA variation and —0
349-47558_ch01_1P.indd 299 5/31/11 7:06 PM

systematics of the guanaco (Lama guanicoe, Artiodac- Matschie, P. 1901a. Über Kaukasische Steinbocke.
tyla: Camelidae). Journal of Mammalogy 80:269–281. Sitzungsberichte der Gesellschaft Naturforschender
Marino, M.L. & R.V. Rossi. 2010. Origin, systematics, and Freunde zu Berlin 1901:27–33.
morphological radiation. In J.M.B. Duarte & S. Matschie, P. 1901b. Wilde Pferde im Park des Hrn.
González (eds.), Neotropical Cervidology: Biology and Falz-Fein in Askania-Nova (Sudrussland). Illustrirte
Medicine of Latin American Deer. Jaboticabal, Brazil Zeitung No. 3010 (7 Marz 1901):366–367.
and Gland, Switzerland: Funep, in collaboration with Matschie, P. 1903. Giebt es in Mittelasien mehrere Arten
IUCN, 2–11. von echten Wildpferden? Naturwissenschaftliche
Markov, G.G., A.A. Danilkin, C. Gerasimov & K. Wochenschrift, n.f. 2, 49:581–583.
Nikolov. 1985. A comparative craniometric analysis of Matschie, P. 1906. Einige noch nich beschriebene Arten
Capreolus capreolus. Doklady Akademii Nauk SSSR des afrikanischen Büffels. Sitzungsberichte der
282:489– 493. Gesellschaft Naturforschender Freunde zu Berlin
Márquez, A., J.E. Maldonado, S. González, M.B. Beccaceci, 1906:161–179.
J.E. Garcia & J.M.B. Duarte. 2006. Phylogeography and Matschie, P. 1907. Übersicht über die vom verfasser aus
Pleistocene demographic history of the endangered seiner Reise gesammelten Tiere. In P. Niedieck (ed.),
marsh deer (Blastocerus dichotomus) from the Río de la Kreuzfahrten im Beringmeer. Berlin: Paul Parey,
Plata Basin. Conservation Genetics 7:563–575. 233–247.
Masembe, C., V.B. Muwanika, S. Nyakaana, P. Arctander Matschie, P. 1910a. Bemerkungen über die Verbreitung
& H.R. Siegismund. 2006. Three genetically divergent der Tiere in den Deutschen Schutzgebieten. Berlin:
lineages of the oryx in eastern Africa: evidence for an published privately.
ancient introgressive hybridisation. Conservation Matschie, P. 1910b. Die von Herrn Major P. H. G.
Genetics 7:551–562. Powell- Cotton gesammelten Rassen des Wasserbockes
Masini, F. & S. Lovari. 1988. Systematics, phylogenetic (Kobus). Sitzungsberichten der Gesellschaft Natur-
relationships, and dispersal of the chamois (Rupicapra forschender Freunde zu Berlin 1910:411– 429.
spp.). Quaternary Research 30:339–349. Matschie, P. 1911a. Ueber einige von Herrn Dr. Holderer
Masseti, M., E. Pecchioli & C. Vernesi. 2008. Phylogeogra- in der sudlichen Gobi und in Tibet gesammelte
phy of the last surviving populations of Rhodian and Säugetiere. In K. Futterer (ed.), Zoologie (Nachtrag).
Anatolian fallow deer (Dama dama dama L., 1758). Vol. 3, pt. 5 of Durch Asien. Berlin: Reimer, 4–29.
Biological Journal of the Linnaean Society 93:835– 844. Matschie, P. 1911b. Zoologische Ergebnisse der Expedition
Matschie, P. 1892. Einige Säugethiere von Deutsch- Ost- des Herrn Hauptmann a. D. Fromm 1908/09 nach
Afrika. Sitzungsberichte der Gesellschaft Natur- Deutsch- Ostaafrika, 4: Mammalia (Gattung Kobus).
forschender Freunde zu Berlin 1892:130–140. Mitteilungen aus dem Zoologischen Museum in Berlin
Matschie, P. 1893. Über die Verbreitung der zur Gattung 5:554–575.
Oryx: Blainv. gehörigen Antilopen. Sitzungsberichte Matschie, P. 1912a. Beuteltücke aus fernen Ländern: die
der Gesellschaft Naturforschender Freunde zu Berlin achtzehnte deutsche Gewehausftellung. Deutsche
1893:101–105. Jäger-Zeitung 59:64– 83.
Matschie, P. 1895. Die Säugethiere Deutsch- Ost-Afrikas. Matschie, P. 1912b. Gazella (Nanger) soemmerringii sibyllae
Berlin: Reiner. subsp. nov. Sitzungsberichte der Gesellschaft Natur-
Matschie, P. 1898a. Einige anscheinend noch nicht forschender Freunde zu Berlin 1912:260–270.
beschrieben Säugethiere aus Afrika. Sitzungsberichte Matschie, P. 1912c. Die von Herrn Major P. H. G.
der Gesellschaft Naturforschender Freunde zu Berlin Powell- Cotton gesammelten Rassen der Gattung
1898:75– 81. Tragelaphus. Sitzungsberichte der Gesellschaft
Matschie, P. 1898b. Die geographische Verbreitung der Naturforschender Freunde zu Berlin 1912:544–567.
Tigerpferde und das Zebra des Kaoko-Feldes in Matschie, P. 1913. Eine neue Form der Elenantilope: Oreas
Deutsch- Sudwest Afrika. Sitzungsberichte der oryx niediecki nov. subsp. Sitzungsberichte der
Gesellschaft Naturforschender Freunde zu Berlin Gesellschaft Naturforschender Freunde zu Berlin
1898:169–181. 1913:249–258.
Matschie, P. 1898c. Eine neue Abart von Hippotragus bakeri Matschie, P. 1914. Eine neue Art der Kudu-Antilope.
Heugl. Sitzungsberichte der Gesellschaft Natur- Sitzungsberichte der Gesellschaft Naturforschender
forschender Freunde zu Berlin 1898:181–182. Freunde zu Berlin 1914:383–393.
Matschie, P. 1899. Eine anscheinend neue Adenota vom Matschie, P. 1916. Die richtige Benennung der Kuhanti-
Weissen Nil. Sitzungsberichte der Gesellschaft lope von Baunza. Sitzungsberichte der Gesellschaft
Naturforschender Freunde zu Berlin 1899:15. Naturforschender Freunde zu Berlin 1916:295.
Matschie, P. 1900. Ueber Equus penricei, Thos. Sitzungsb- Matschie, P. 1918a. Die Büffel des Ituri-Urwaldes.
erichte der Gesellschaft Naturforschender Freunde zu Sitzungsberichte der Gesellschaft Naturforschender
-1— Berlin. 1900:231. Freunde au Berlin 1918:1–5.
0—
349-47558_ch01_1P.indd 300 5/31/11 7:06 PM

ref erences 301
Matschie, P. 1918b. Ein Schwarzbüffel des Kafue- Gebietes. Mendelssohn, H., C.P. Groves & B. Shalmon. 1997. A new
Sitzungsberichte der Gesellschaft Naturforschender subspecies of Gazella gazella from the southern Negev.
Freunde zu Berlin 1918:133–140. Israel Journal of Zoology 43:209–215.
Matschie, P. 1920. Tierwelt der deutschen Schutzgebiete. Mentis, M. 1974. Distribution of some wild animals in
In H. Schnee (ed.), Deutsches Kolonial-Lexikon, vol. 3. Natal. Lammergeyer 20:1– 68.
Leipzig: Quelle & Meyer, 483– 494. Merino, M.L. & R.V. Rossi. 2010. Origin, systematics, and
Matschie, P. 1922. Bemerkungen uber einige tibetanische morphological radiation. In J.M.B. Duarte & S.
Säugetiere. Sitzungsberichte der Gesellschaft Natur- González (eds.), Neotropical Cervidology: Biology and
forschender Freunde au Berlin 1922:65–75. Medicine of Latin American Deer. Jaboticabal, Brazil
Matschie, P. & L. Zukowsky. 1916. Die als Sigmoceros and Gland, Switzerland: Funep, in collaboration with
bezeichnete Gruppe der Kuhantilopen. Sitzungsberi- IUCN, 2–11.
chte der Gesellschaft Naturforschender Freunde zu Métais, G. & I. Vislobokova. 2007. Basal Ruminants. In
Berlin 1916:188–211. D.R. Prothero & S.E. Foss (eds.), The Evolution of
Matschie, P. & L. Zukowsky. 1917. Die als Sigmoceros Artiodactyls. Baltimore: Johns Hopkins University
bezeichnete Gruppe der Kuhantilopen. Sitzungsberi- Press, 189–212.
chte der Gesellschaft Naturforschender Freunde zu Milne-Edwards, A.E. 1867. Observations sur quelques
Berlin 1917:527–550. mammifères du nord de la Chine. Annales des Sciences
Matschie, P. & L. Zukowsky. 1922. Die als Sigmoceros Naturelles, Zoologie (5) 7:375–377.
bezeichnete Gruppe der Kuhantilopen. Sitzungsberi- Milner, C.E. 1921. Distribution of serow in Burma. Journal
chte der Gesellschaft Naturforschender Freunde zu of the Bombay Natural History Society 28:267–268.
Berlin 1922:79–144. Min, M-s., H. Okumura, D-j. Jo, J-h. An, K-s. Kim, C-b.
Matthee, C.A. & S.K. Davis. 2001. Molecular insights into Kim, N-s. Shin, M-h. Lee, C-h. Han, I.V. Voloshina &
the evolution of the family Bovidae: a nuclear DNA H. Lee. 2004. Molecular phylogenetic status of the
perspective. Molecular Biology and Evolution Korean goral and Japa nese serow based on partial
18:1220–1230. sequences of the mitochondrial cytochrome b gene.
Matthee, C.A. & T.J. Robinson. 1999a. Cytochrome b Molecules and Cells 17:365–372.
phylogeny of the family Bovidae: resolution within Miyamoto, M.M., F. Kraus & O.A. Ryder. 1990. Phylogeny
the Alcelaphini, Antilopini, Neotragini, and Tragela- and evolution of antlered deer determined from
phini. Molecular Phylogenetics and Evolution mitochondrial DNA sequences. PNAS: Proceedings of
12:31– 46. the National Academy of Science USA 87:6127– 6131.
Matthee, C.A. & T.J. Robinson. 1999b. Mitochondrial Mohr, E. 1952. Der Wisent. Die Neue Brehm-Bücherei
DNA population structure of roan and sable antelope: No. 74. Wittenberg Lutherstadt, Germany: A. Ziemsen.
implications for the translocation and conservation of Molina, M. & J. Molinari. 1999. Taxonomy of Venezuelan
the species. Molecular Ecology 8:227–238. white-tailed deer (Odocoileus, Cervidae, Mammalia),
Mattioli, S., R. Fico, R. Lorenzin & G. Nobili. 2003. based on cranial and mandibular traits. Canadian
Messola red deer: physical characteristics, population Journal of Zoology 77:632– 645.
dynamics, and conservation perspectives. Hystrix: Mona, S., E. Randi & M. Tommaseo-Ponzetta. 2007.
Italian Journal of Mammalogy 14:87– 94. Evolutionary history of the genus Sus inferred from
Mayr, E. 1963. Animal Species and Evolution. Harvard: cytochrome b sequences. Molecular Phylogenetics and
Belknap Press. Evolution 45:757–762.
Medellín, R.A., A.L. Gardner & J.M. Aranda. 1998. The Monfort, A. & N. Monfort 1973. Quelques observations
taxonomic status of the Yucatán brown brocket, sur les grands mammifères du Parc National de Tai
Mazama pandora (Mammalia: Cervidae). Proceedings (Côte d’Ivoire). La Terre et La Vie 27:499–506.
of the Biological Society of Washington 111:1–14. Moodley, Y. & M.W. Bruford. 2007. Molecular biogeogra-
Meijaard, E. & C.[P.] Groves. 2002. Upgrading three phy: towards an integrated framework for conserving
subspecies of babirusa (Babyrousa sp.) to full species Pan-African biodiversity. PLoS ONE 2 (5):e 454.
level. Asian Wild Pig News 2, 2:33–39. doi:10.1371/journal.pone.0000454.
Meijaard, E. & C.P. Groves. 2004a. A taxonomic revision Moodley, Y. & E.H. Harley. 2005. Population structuring
of the Tragulus mouse-deer (Artiodactyla). Zoological in mountain zebras (Equus zebra): the molecular
Journal of the Linnaean Society 140:63–102. consequences of divergent demographic histories.
Meijaard, E. & C.P. Groves. 2004b. Morphometric Conservation Genetics 6:953– 968.
relationships between South-east Asian deer (Cervidae, Moodley, Y., M.W. Bruford, C. Bleidorn, T. Wronski, A.
tribe Cervini): evolutionary and biogeographic Apio & M. Plath. 2009. Analysis of mitochondrial DNA
implications. Journal of Zoology, London 263:179–196. data reveals non-monophyly in the bushbuck (Tragela-
Mendelssohn, H., Y. Yom-Tov & C.P. Groves. 1995. Gazella phus scriptus) complex. Mammalian Biology
gazella. Mammalian Species 490:1–7. 74:418– 422. —-1
—0
349-47558_ch01_1P.indd 301 5/31/11 7:06 PM

Moscarella, R.A., M. Aguilera & A.A. Escalante. 2003. Nguyen An Quang Ha. 1997. Another new animal
Phylogeography, population structure, and implica- discovered in Vietnam. Vietnam Economic News
tions for conservation of white-tailed deer (Odocoileus 38:46– 47.
virginianus) in Venezuelan. Journal of Mammalogy Nijman, I.J., D.C.J. van Boxtel, L.M. van Cann, Y.
84:1300–1315. Marnoch, E. Cuppen & J.A. Lenstra. 2008. Phylogeny
Nadler, C.F., K.V. Korobitsina, R.S. Hoff mann & N.N. of Y chromosomes from bovine species. Cladistics
Vorontsov. 1973. Cytogenetic differentiation, 24:723–726.
geographic distribution, and domestication in Nixon, K.C. & Q.D. Wheeler. 1990. An amplification of
Palaearctic sheep (Ovis). Zeitschrift für Säuget- the phylogenetic species concept. Cladistics 6:211–223.
ierkunde 38:109–125. O’Gara, P.W. 1990. The pronghorn (Antilocapra americana).
Nagata, J., R. Masuda, H.B. Tamate, S-i. Hamasaki, K. In G.A. Bubenik & A.B. Bubenik (eds.), Horns,
Ochiai, M. Asada, S. Tatsuzawa, K. Suda, H. Tado & Pronghorns, and Antlers. New York: Springer-Verlag,
M.C. Yoshida. 1999. Two genetically distinct lineages 231–264.
of the sika deer, Cervus nippon, in Japa nese islands: O’Gara, B.W. 2002. Taxonomy. In D.E. Toweill & J.W.
comparison of mitochondrial D-loop region sequences. Thomas (eds.), North American Elk: Ecology and
Molecular Phylogenetics in Evolution 13:511–519. Management. Smithsonian Institution Press: Washing-
Nascetti, G., S. Lovari, P. Lanfranchi, C. Berducou, S. ton, DC, 3– 66.
Mattrucci, L. Rossi & L. Bullini. 1985. Revision of Ohtaishi, N. & Y. Gao. 1990. A review of the distribution
Rupicapra genus, 3: Electrophoretic studies demon- of all species of deer (Tragulidae, Moschidae, and
strating species distinction of chamois populations of Cervidae) in China. Mammal Review 20:125–144.
the Alps from those of the Apennines and Pyrenees. In Okello, J.B.A., S. Nyakaana, C. Masembe, H.R. Siegis-
S. Lovari (ed.), The Biology and Management of mund & P. Arctander. 2005. Mitochondrial DNA
Mountain Ungulates. London: Croom Helm, 56– 62. variation of the common hippopotamus: evidence for
Nasonov, N. 1923. Geographiceskoe Rasprostranenie any recent population expansion. Heredity 95:206–215.
Dikih Baranov Starogo Sveta [Distribution Orr, R.T. 1938. Mammals from Sikang, China. Proceed-
Géographique des Mouton Sauvages du Monde ings of the California Academy of Sciences 23:307–310.
Ancien]. St. Petersburg, Russia: Akademii Nauk SSR. Oswald, C. 2002. A commentary on the systematics of red
Nchanji, A.C. & F.O. Amubode. 2002. The physical and deer Cervus elaphus L. Deer 12:84– 89.
morphological characteristics of the red-fronted Pallas, P.S. 1766. Miscellanea zoologica, quibus novae
gazelle (Gazella rufifrons kanuri Gray 1846) in Waza imprimis atque obscurae animalum species describun-
National Park, Cameroon. Journal of Zoology, London tur et observationibus iconibusque illustrantur. The
256:505–509. Hague: Kessinger.
Nersting, L.G. & P. Arctander. 2001. Phylogeography and Pallas, P.S. 1767–1777. Spicilegia zoologica, quibus novae
conservation of impala and greater kudu. Molecular imprimis et obscurae animalum species iconibus,
Ecology 10:711–719. descriptionibus atque commentariis illustrantur.
Neuhaus, P. & K.E. Ruckstuhl. 2002. The link between Berlin: Gottl. August Lange.
sexual dimorphism, activity budgets, and group Pallas, P.S. 1811. Zoographia Rosso-Asiatica, 3 vols.
cohesion: the case of the plains zebra (Equus burchelli). St. Petersburg: Academiae Scientarum.
Canadian Journal of Zoology 80:1437–1441. Paterson, H.E.H. 1978. More evidence against speciation
Neumann, O. 1896. Description of a new species of by reinforcement. South African Journal of Science
antelope from East Africa. Proceedings of the 74:369–371.
Zoological Society of London 1896:192–194. Paterson, H.E.H. 1980. A comment on “mate recognition”
Neumann, O. 1898. On a new antelope of the genus systems. Evolution 34:330–331.
Hippotragus. Proceedings of the Zoological Society of Peacock, E.H. 1933. A Game Book for Burma and
London 1898:850– 851. Adjoining Territories. London: H.F. & G. Witherby.
Neumann, O. 1905. Über neue Antilopen-Arten. Sitzungs- Peden, D.G. & G.J. Kraay. 1979. Comparison of blood
berichte der Gesellschaft Naturforschender Freunde characteristics in plains bison, wood bison, and their
zu Berlin 1905:88– 97. hybrids. Canadian Journal of Zoology 57:1778–1784.
Neumann, O. 1906. Gazellan und Kuh-Antilopen. Pees, W. & H. Hemmer. 1980. Hirngrösse und Aktivität
Sitzungsberichte der Gesellschaft Naturforschender bei Wildschafen und Hausschafen (Gattung Ovis).
Freunde zu Berlin 1906:236–248. Säugetierkundliche Mitteilungen 28:39– 45.
Neumann-Denzau, G. & H. Denzau. 1999. Wildesel. Pemberton, J.M., P.W. King, S. Lovari & B. Bauchau. 1989.
Stuttgart: Jan Thorbecke. Genetic variation in the Alpine chamois, with special
Neumann-Denzau, G. & H. Denzau. 2003. The southern reference to the subspecies Rupicapra rupicapra
kiang Equus kiang polyodon. Journal of the Bombay cartusiana, Couturier, 1938. Zeitschrift für Säuget-
-1— Natural History Society 100:322–340. ierkunde 54:243–250.
0—
349-47558_ch01_1P.indd 302 5/31/11 7:06 PM

ref erences 303
Perez, M.C. 1984. Revision der Systematik von Gazella Pocock, R.I. 1910a. The serows, gorals, and takins of
(Nanger) dama. Zeitschrift des Kölner Zoo 27:103–107. British India and the Straits Settlements. Journal of the
Pérez, T., J. Albornoz & A. Domínguez. 2002. Phylo- Bombay Natural History Society 19:807– 821.
geography of chamois (Rupicapra spp.) inferred from Pocock, R.I. 1910b. On the specialized cutaneous glands of
microsatellites. Molecular Phylogenetics and Evolution ruminants. Proceedings of the Zoological Society of
25:524–534. London 1910:840– 986.
Perret, J.-L. & V. Aellen. 1956. Mammifères du Cameroun Pocock, R.I. 1913. The serows, gorals, and takins of British
de la collection J.-L. Perret. Revue Suisse de Zoologie India and the Straits Settlements, part 2. Journal of the
63:395– 450. Bombay Natural History Society 22:296–319.
Peterson, R.L. 1950. A new subspecies of moose from Pocock, R.I. 1914. Description of a new species of goral
North America. Occasional Papers of the Royal (Nemorhaedus) shot by Captain F. M. Bailey. Journal of
Ontario Museum of Zoology 9:1–7. the Bombay Natural History Society 23:32–33.
Pfeffer, P. 1962. Un cobe de montagne propre au Cameroun: Pocock, R. I. 1918. On some external characters of
Redunca fulvorufula adamauae subspecies nova. ruminant Artiodactyla, part 2: the Antilopinae,
Mammalia 26:64–71. Rupicaprinae, and Caprinae, with a note on the penis
Pfeffer, P. 1967. Le mouflon de Corse (Ovis ammon musimon of the Cephalophinae and Neotraginae. Annals and
Schreber, 1782): position systématique, écologie et Magazine of Natural History (9) 2: 125–144.
éthologie comparées. Mammalia 31, Suppl.: 1–262. Pocock, R.I. 1919. On some external characters of
Pidancier, N., S. Jordan, G. Luikart & P. Taberlet. 2006. ruminant Artiodactyla, part 2. Annals and Magazine of
Evolutionary history of the genus Capra (Mammalia, Natural History (9) 11:125–144.
Artiodactyla): discordance between mitochondrial Pocock, R.I. 1945. Some cranial and dental characters of
DNA and Y-chromosome phylogenies. Molecular the existing species of Asiatic rhinoceroses. Proceed-
Phylogenetics and Evolution 40:739–749. ings of the Zoological Society of London 114:437– 450.
Pienaar, U. de V. 1963. The large mammals of the Kruger Powell, C.P. & P. Grubb. 2002. Range extension of
National Park—their distribution and present-day black-fronted duiker (Cephalophus nigrifrons Gray 1871,
status. Koedoe 6:1–37. Artiodactyla, Bovidae): fi rst records from Nigeria.
Piovezan, U., L.M. Tiepolo, W.M. Tomas, J.M.B. Duarte, Tropical Zoology 15:89– 95.
D. Varela & J.S. Marinho Filho. 2010. Marsh deer Presidente, P.J.A. & M. Draisma. 1980. Hog deer on
Blastocerus dichotomus (Illiger 1815). In J.M.B. Duarte & Sunday Island, part 1: analyses of body measurements
S. González (eds.), Neotropical Cervidology: Biology taken in 1978–1979. Australian Deer 5, 3:11–21.
and Medicine of Latin American Deer. Jaboticabal, Prothero, D.R. 2007. Family Moschidae. In D.R. Prothero
Brazil and Gland, Switzerland: Funep, in collaboration & S.E. Foss (eds.), The Evolution of Artiodactyls.
with IUCN, 66–76. Baltimore: Johns Hopkins University Press, 221–226.
Pitra, C., R. Fürbass & H.-M. Seyfert. 1997. Molecular Rabinowitz, A., G. Amato & U Saw Tun Khaing. 1998.
phylogeny of the tribe Bovini (Mammalia: Artiodac- Discovery of the black muntjac, Muntiacus crinifrons
tyla): alternative placement of the anoa. Journal of (Artiodactyla, Cervidae), in north Myanmar. Mamma-
Evolutionary Biology 10:589– 600. lia 62:105–108.
Pitra, C., A.J. Hansen, D. Lieckfeldt & P. Arctander. 2002. Rabinowitz, A., Than Myint, Saw Tun Khaing & S.
An exceptional case of historical outbreeding in Rabinowitz. 1999. Description of the leaf deer (Muntia-
African sable antelope populations. Molecular Ecology cus putaoensis), a new species of muntjac from northern
11:1197–1208. Myanmar. Journal of Zoology, London 249:427– 435.
Pitra, C., J. Fickel, E. Meijaard & C.P. Groves. 2004. Radde, G. 1862. Säugetierfauna. Vol. 1 of Reisen in den
Evolution and phylogeny of Old World deer. Molecular Süden von Ost-Siberien: im auftrage der Kaiserlichen
Phylogenetics and Evolution 33:880– 895. Russischen Gesellschaft ausgeführt in den Jahren
Pitra, C., P. Vaz Pinto, B.W.J. O’Keefe, S. Williams-Munro, 1855–59. Moscow: Buchdrückerei der Kaiserlichen
B. Jansen van Vuuren & T.J. Robinson. 2006. DNA-led Universität.
rediscovery of the giant sable antelope in Angola. Radinsky, L.B. 1963. The Perissodactyl hallux. American
European Journal of Wildlife Research 52:145–152. Museum Novitates 2145:1– 8.
Pocock, R.I. 1908a. Notes on some species and geograph i- Rahm, U. 1961. Esquise mammalogiques de basse Côte
cal races of serows (Capricornis) and gorals (Nemorhae- d’Ivoire. Bulletin de l’Institut Fondamental d’Afrique
dus), based upon specimens exhibited in the Society’s Noire 23:1229–1265.
Gardens. Proceedings of the Zoological Society of Rahm, U. 1966. Les mammifères de la forêt équatoriale de
London 1908:173–202. l’est du Congo. Annales de la Musée Royale de
Pocock, R.I. 1908b. On the generic names of the l’Afrique Centrale 149:39–121.
rupicaprine ruminants known as serows and gorals. Ramey, R.R. 1995. Mitochondrial DNA variation,
Annals and Magazine of Natural History (8) 1:183–188. population structure, and evolution of mountain sheep —-1
—0
349-47558_ch01_1P.indd 303 5/31/11 7:06 PM

304 refe r e n c e s
in the south-western United States and Mexico. Roberts, A. 1913. The collection of mammals in the
Molecular Ecology 4:429– 439. Transvaal Museum registered up to the 31st March
Randi, E., V. Lucchini & C.H. Diong. 1996. Evolutionary 1913, with descriptions of new species. Annals of the
genetics of the Suiformes as reconstructed using Transvaal Museum 4:65–107.
mtDNA sequencing. Journal of Mammalian Evolution kmc
3:163–194. Roberts, A. 1915. Additions to the collections of mammals
Randi, E., N. Mucci, M. Pierpaoli & E. Douzery. 1998. in the Transvaal Museum. Annals of the Transvaal
New phylogenetic perspectives on the Cervidae Museum 4:116–124.
(Artiodactyla) are provided by the mitochondrial Roberts, A. 1951. The Mammals of South Africa, ed. R.
cytochrome b gene. Proceedings of the Royal Society Bigalke, V. FitzSimons & D.E. Malan. Johannesburg:
of London, B 265:793– 801. Trustees of “The Mammals of South Africa” Book
Randi, E., M. Pierpaoli & A. Danilkin. 1998. Mitochon- Fund.
drial DNA polymorphism in populations of Siberian Roberts, S.C. 1996. The evolution of hornedness in female
and European roe deer (Capreolus pygargus and ruminants. Behaviour 133:399– 442.
Capreolus capreolus). Heredity 80:429– 437. Roberts, T.J. 1977. The Mammals of Pakistan. London:
Randi, E., N. Mucci, F. Claro-Hergueta, A. Bonnet & E.J.P. Ernest Benn.
Douzery. 2001. A mitochondrial DNA control region Robins, J.H., H.A. Ross, M.S. Allen & E. Matisoo-Smith.
phylogeny of the Cervinae: speciation in Cervus and 2006. Sus bucculentus revisited. Nature 440: E7.
implications for conservation. Animal Conservation Robinson, T.J. 1979. Influence of a nutritional pa rameter
4:1–11. on the size differences of the three springbok subspe-
Randi, E., J.-P. d’Huart, V. Lucchini & R. Aman. 2002. cies. South African Journal of Zoology 14:13–15.
Evidence of two genetically deeply divergent species of Robinson, T.J. & J.D. Skinner. 1976. A karyological survey
warthog, Phacochoerus africanus and P. aethiopicus of springbok subspecies. South African Journal of
(Artiodactyla: Suiformes) in East Africa. Mammalian Science 72:147–148.
Biology 67:91– 96. Robinson, T.J., D.J. Morris & N. Fairall. 1991. Interspecific
Ranjitsinh, M K. 1989. The Indian Blackbuck. Dehradun, hybridisation in the Bovidae: sterility of Alcelaphus
India: Natraj. buselaphus × Damaliscus dorcas F1 progeny. Biological
Rautenbach, I.L. 1982. Mammals of the Transvaal. Pretoria: Conservation 58:345–356.
Ecoplan. Robinson, T.J., A.D. Bastos, K.M. Halanych & B. Heizig.
Rautian, G.S., B.A. Kalabushkin & A.S. Nemtsev. 2000. A 1996. Mitochondrial DNA sequence relationships of
new subspecies of the European bison, Bison bonasus the extinct blue antelope Hippotragus leucophaeus.
montanus ssp. nov. (Bovidae, Artiodactyla). Doklady Naturwissenschaften 83:178–182.
Biological Sciences 375:636– 640. Robinson, T.J., V. Trifonov, I. Espie & E.H. Harley. 2005.
Rebholz, W.[E.R.] & E. Harley. 1999. Phylogenetic Interspecific hybridisation in rhinoceroses: confi rma-
relationships in the bovid subfamily Antilopinae based tion of a black × white rhinoceros hybrid by karyotype,
on mitochondrial DNA sequences. Molecular fluorescence in situ hybridisation (FISH), and microsat-
Phylogenetics and Evolution 12:87– 94. ellites analysis. Conservation Genetics 6:141–145.
Rebholz, W.E.R., D. Williamson & F. Rietkerk. 1991. Roche, J. 1971. Recherches mammalogiques en Guinée
Saudi gazelle (Gazella saudiya) is not a subspecies of forestière. Bulletin du Museum National d’Histoire
dorcas gazelle. Zoo Biology 10:485– 489. Naturelle (3) 16:737–781.
Redford, K.H. & J.F. Eisenberg. 1992. The Southern Cone: Rode, P. 1943. Mammifères Ongulés de l’Afrique Noire.
Chile, Argentina, Uruguay, Paraguay. Vol. 2 of Vol. 2 of Faune de l’Empire Français. Paris: Larose.
Mammals of the Neotropics. Chicago: University of Røed, K.H., M.A.D. Ferguson, M. Çrête & T.A. Bergerud.
Chicago Press. 1991. Genetic variation in transferrin as a predictor for
Reynolds, S.C. & L.C. Bishop. 2002. Cranio-dental differentiation and evolution of caribou from eastern
variability in modern and fossil plains zebra (Equus Canada. Rangifer 11:65–74.
burchellii Gray 1824) from east and southern Africa. In Rookmaaker, L.C [= K.]. 1980. The distribution of the
M. Mashkour (ed.), Equids in Time and Space: Papers rhinoceros in eastern India, Bangladesh, China, and
in Honour of Véra Eisenmann. Oxford: Oxbow, 49– 60. the Indochinese region. Zoologischer Anzeiger
Rideout, C.B. & R.S. Hoff mann. 1975. Oreamnos america- 205:253–268.
nus. Mammalian Species 63:1– 6. Rookmaaker, L.C [= K.]. 1982. The type locality of the
Rieppel, O. 2009. Species as a process. Acta Biotheoretica Javan rhinoceros (Rhinoceros sondaicus Desmarest,
57:33– 49. 1822). Zeitschrift für Säugetierkunde 47:381–382.
Ritz, L.R., M.-L. Glowatzki-Mullis, D.E. MacHugh & Rookmaaker, L.C [= K.]. 1983. Historical notes on the
C. Gaillard. 2000. Phylogenetic analysis of the tribe taxonomy and nomenclature of the recent Rhinocero-
-1— Bovini using microsatellites. Animal Genetics tidae (Mammalia, Perissodactyla). Beaufortia 33,
0— 31:178–185. 4:37–51.
349-47558_ch01_1P.indd 304 5/31/11 7:06 PM

ref erences 305
Rookmaaker, L.C [= K.]. 1988. The scientific names of the González (eds.), Neotropical Cervidology: Biology and
South African steenbok and grysbok (Raphicerus Medicine of Latin American Deer. Jaboticabal, Brazil
campestris and R. melanotis). Mammalia 52:213–217. and Gland, Switzerland: Funep, in collaboration with
Rookmaaker, L.C. [= K.] 1989. The Zoological Exploration IUCN, 202–210.
of Southern Africa, 1650–1790. Rotterdam: A.A. Rössner, G.E. 2007. Family Tragulidae. In D.R. Prothero
Balkema. & S.E. Foss (eds.), The Evolution of Artiodactyls.
Rookmaaker, L.C [= K.]. 1991. The scientific name of the Baltimore: Johns Hopkins University Press, 213–220.
bontebok. Zeitschrift für Säugetierkunde 66:190–191. Roth, T.L., H.L. Bateman, J.L. Kroll, B.G. Steinetz & P.R.
Rookmaaker, L.C [= K.]. 1995. Subspecies and ecotypes of Reinhart. 2004. Endocrine and ultrasonographic
the black rhinoceros. Pachyderm 20:39– 40. characterization of a successful pregnancy in a
Rookmaaker, L.C. [= K.]. 1997. Records of the Sundarbans Sumatran rhinoceros (Dicerorhinus sumatrensis)
rhinoceros (Rhinoceros sondaicus inermis) in India and supplemented with a synthetic progestin. Zoo Biology
Bangladesh. Pachyderm 24:37– 45. 23:219–238.
Rookmaaker, L.C [= K.]. 1998. The sources of Linnaeus on Ruiz- García, M., C. Vásquez, S. Sandoval, A. Castellanos,
the rhinoceros. Svenska Linnésallskapets Årsskrift, F. Kaston, B. de Thoisy & J. Shostell. In press.
Uppsala 1996–1997:61–80. Phylogeography of the mountain tapir (Tapirus
Rookmaaker, L.C. [= K.]. 2004. Rhinoceros rugosus— a name pinchaque) and of the Central American tapir (Tapirus
for the Indian rhinoceros. Journal of the Bombay bairdii) and the origins of the three Latin American
Natural History Society 101:308–310. tapirs by means of mtCyt-b sequences. Molecular
Rookmaaker, L.C. [= K.]. 2005. The black rhino needs a Phylogeny and Evolution.
taxonomic revision for sound conservation. Interna- Rumiz, D.I. & E. Pardo. 2010. Peruvian dwarf brocket
tional Zoo News 52:280–282. deer Mazama chunyi (Hershkovitz 1959). In J.M.B.
Rookmaaker, L.C [= K.]. & C.P. Groves. 1978. The extinct Duarte & S. González (eds.), Neotropical Cervidology:
Cape rhinoceros, Diceros bicornis bicornis (Linnaeus, Biology and Medicine of Latin American Deer.
1758). Säugetierkunde Mitteilungen 26:117–126. Jaboticabal, Brazil and Gland, Switzerland: Funep, in
Roosevelt, T. & E. Heller. 1914. Life-Histories of African collaboration with IUCN, 185–189.
Game Animals, 2 vols. New York: Charles Scribner’s Rushby, C.C. & C.H. Swynnerton. 1946. Notes on some
Sons. game animals of Tanganyika Territory. Tanganyika
van Roosmalen, M.G.M., L. Frenz, P. van Hooft, H.H. de Notes and Records 22:14–26.
Iongh & H. Leirs. 2006. A new species of living peccary Ruxton, A.E. & E. Schwarz. 1929. On hybrid hartebeests
(Mammalia: Tayassuidae) from the Brazilian Amazon. and on the distribution of Alcelaphus buselaphus.
Bonner Zoologische Beiträge 55:105–112. Proceedings of the Zoological Society of London
Ropiquet, A. 2006. Étude des radiations adaptatives au 1929:567–583.
sein des Antilopinae (Mammalia, Bovidae). PhD diss., Ryder, O.A., A.T. Kumamoto, B.S. Durrant &
Université Paris 6. K. Benirschke. 1989. Chromosomal divergence and
Ropiquet, A. & A. Hassanin. 2004. Molecular phylogeny of reproductive isolation in dik-diks. In D. Otte & J.A.
caprines (Bovidae, Antilopinae): the question of their Endler (eds.), Speciation and Its Consequences.
origin and diversification during the Miocene. Journal Sunderland, MA: Sinauer Associates, 208–225.
of Zoological Systematics and Evolutionary Research Scala, C. & S. Lovari. 1984. Revision of Rupicapra genus, 2:
43:49– 60. A skull and horn statistical comparison of Rupicapra
Ropiquet, A. & A. Hassanin. 2005. Molecular evidence for rupicapra ornata and R. rupicapra pyrenaica chamois.
the polyphyly of the genus Hemitragus (Mammalia, Bolletino Zoologico 51:285–294.
Bovidae). Molecular Phylogenetics and Evolution Schaffer, W.M. & C.A. Reed. 1972. The co- evolution of
36:154–168. social behaviour and cranial morphology in sheep
Ropiquet, A. & A. Hassanin. 2006. Hybrid origin of the and goats (Bovidae, Caprini). Fieldiana, Zoology
Pliocene ancestor of wild goats. Molecular Phylogene- 61:1– 62.
tics and Evolution 41:395– 414. Schaller, G.B. 1977. Mountain Monarchs: Wild Sheep and
Ropiquet, A., M. Gerbault-Seureau, J.L. Deuve, C. Gilbert, Goats of the Himalaya. Chicago: University of Chicago
E. Pagacova, N. Chai, J. Rubes & A. Hassanin. 2008. Press.
Chromosome evolution in the subtribe Bovina Schaller, G.B. 1998. Wildlife of the Tibetan Steppe.
(Mammalia, Bovidae): the karyotype of the Cambo- Chicago: University of Chicago Press.
dian banteng (Bos javanicus birmanicus) suggests that Schaller, G.B. & S. Khan. 1975. The status and distribution
Robertsonian translocations are related to interspecific of markhor (Capra falconeri). Biological Conservation
hybridization. Chromosome Research 16:1107–1118. 7:185–198.
Rossi, R.V., R. Bodmer, J.M.B. Duarte & R.G. Trovati. Schaller, G.B. & A. Rabinowitz. 1995. The saola or
2010. Amazonian brown brocket deer Mazama spindlehorn bovid Pseudoryx nghetinhensis in Laos. —-1
nemorivaga (Cuvier 1817). In J.M.B. Duarte & S. Oryx 29:107–114. —0
349-47558_ch01_1P.indd 305 5/31/11 7:06 PM

Schaller, G.B. & E.S. Vrba. 1996. Description of the giant Smithers, R.H.N. & J.L.P. Lobão Tello 1976. Check List
muntjac (Megamuntiacus vuquangensis) in Laos. Journal and Atlas of the Mammals of Moçambique. Salisbury:
of Mammalogy 77:675– 683. Trustees of the National Museums and Monuments of
Schmitt, J. & F. Ulbrich 1968. Die Chromosomen Rhodesia.
verschiedener Caprini, Simpson, 1945. Zeitschrift für Sobanskiy, G.G. 1988. [The game animals of the Altai
Säugetierkunde 33:180–186. Mountains.] Novosibirsk, Russia: Nauka. [In Russian.]
Schomber, H. 2007. Giraffengazelle und Lamagazelle, 2nd Sokolov, V.E. & A.A. Lushchekina. 1997. Procapra
ed. Hohenwarsleben, Germany: Westarp gutturosa. Mammalian Species 571:1–5.
Wissenschaften. Sokolov, V.E. & V.I. Prikhod’ko. 1997. [Taxonomy of the
Schonewald, C. 1994. Cervus canadensis and C. elaphus: musk deer Moschus moschiferus (Artiodactyla, Mamma-
North American subspecies and evaluation of clinal lia)]. Ivestiia Akademii Nauk, Seriia Biologicheskaia
extremes. Acta Theriologica 39:431– 452. [Biology Bulletin] 24:557–566. [In Russian.]
Schouteden, H. 1944–1946. De zoogdieren van Belgisch- Sokolov, V.E. & V.I. Prikhod’ko. 1998. [Taxonomy of the
Congo en van Ruanda-Urundi. Annales de la Musée du musk deer Moschus moschiferus (Artiodactyla, Mamma-
Congo Belge, C: Zoologie (2) 3:1–576. lia)]. Ivestiia Akademii Nauk, Seriia Biologicheskaia
Schreiber, A. 1994. A female Bukhara deer Cervus elaphus [Biology Bulletin] 25:28–36. [In Russian.]
bactrianus Lydekker, 1900 with 68 chromosomes. Acta Solounias, N. 1999. The remarkable anatomy of the
Theriologica 39:99–102. giraffe’s neck. Journal of Zoology, London
Schreiber, A. & G. Nötzold. 1995. One EEP, but how many 247:257–268.
anoas? In F. Rietkerk, K. Brouwer & S. Smits (eds.), EEP Solounias, N. 2007a. Family Giraffidae. In D.R. Prothero
[European Endangered Species Program]Yearbook & S.E. Foss (eds.), The Evolution of Artiodactyls.
1995/96. Amsterdam: EAZA [European Association of Baltimore: Johns Hopkins University Press, 257–277.
Zoos and Aquaria] Executive Office, 419– 424. Solounias, N. 2007b. Family Bovidae. In D.R. Prothero &
Schreiber, A., P. Fakler & R. Østerballe. 1997. Blood S.E. Foss (eds.), The Evolution of Artiodactyls.
protein variation in blackbuck (Antilope cervicapra), a Baltimore: Johns Hopkins University Press, 278–291.
lekking gazelle. Zeitschrift für Säugetierkunde Soma, H. & H. Kada. 1986. Evolutionary pathway of
62:239–249. chromosomes of the Capricornis. In O.A. Ryder &
Schreiber, A., I. Seibold, G. Nötzold & M. Wink. 1999. M.L. Byrd (eds.), One Medicine: Essays in Honour of
Cytochrome b gene haplotypes characterize chromo- Kurt Benirschke. Berlin: Springer-Verlag, 109–118.
somal lineages of anoa, the Sulawesi dwarf buffalo Soma, H., T. Kiyokawa, K. Matayoshi, I. Tarumoto, M.
(Bovidae: Bubalus sp.). Journal of Heredity 90:165–176. Miyashita & K. Nagase. 1979. The chromosomes of the
Schwarz, E. 1926. Huftiere aus West und Zentralafrika, Mongolian gazelle (Procapra gutturosa), a rare species of
Ergebnisse der 2 Deutschen Zentral Afrika Expedition antelope. Proceedings of the Japan Academy, Ser. B:
1910–1911. Zoology 1:832–1044. Physical and Biological Science 55:6– 9.
Sclater, P.L. & O. Thomas 1895. The Book of Antelopes, Soma, H., H. Kada, K. Matayoshi, Y. Suzuki, C. Meckvi-
vol. 2 [of 4]. London: R.H. Porter. chal, A. Mahannop & B. Vatanaromya. 1983. The
Scoazec, J.-Y. 1966. An overlooked name of an African chromosomes of Muntiacus feae. Cytogenetics and Cell
ungulate: Connochaetes taurinus babaulti Kollman, 1919. Genetics 35:156–158.
Mammalia 60:155–157. Soma, H., H. Kada & K. Matayoshi. 1987. Evolutionary
Selous, F.C. 1908. African Nature Notes and Reminis- pathways of karyotypes of the tribe Rupicaprini. In H.
cences. London: MacMillan. Soma (ed.), The Biology and Management of Capricor-
Shackleton, D.M. (ed.). 1997. Wild Sheep and Goats and nis and Related Mountain Antelopes. London: Croom
Their Relatives: Status Survey and Conservation Helm, 62–71.
Action Plan for Caprinae. Gland, Switzerland: IUCN. Sopin, L.V. & D.L. Harrison. 1987. On the taxonomic
Sheng, H-l. 1992. Deer in China. Shanghai: East China position of the Oman wild sheep. Zoologicheskii
Normal University. Zhurnal 65:952– 955.
Shi, L. 1987. Recent trends in mammalian cytogenetics in Sowerby, A. de C. 1917. On Heude’s collections of pigs,
China. La Kromosomo (2) 45:1458–1467. sika, serows, and gorals in the Sikawei Museum,
Shortridge, G.C. 1934. The Mammals of South West Shanghai. Proceedings of the Zoological Society of
Africa: A Biological Account of the Forms Occurring in London 1917:7–26.
that Region, 2 vols. London: William Heinemann. van Staaden, M.J., M.J. Hamilton & R.K. Chesser. 1995.
Simpson, G.G. 1961. Principles of Animal Taxonomy. New Genetic variation of woodland caribou (Rangifer
York: Columbia University Press. tarandus) in North America. Zeitschrift für Säuget-
Skinner, J.D. & C.T. Chimimba. 2005. The Mammals of ierkunde 60:150–158.
the Southern African Region, 3rd ed. Cambridge: Stanley, H.F., M. Kadwell & J.C. Wheeler. 1994. Molecular
-1— Cambridge University Press. evolution of the family Camelidae: a mitochondrial
0—
349-47558_ch01_1P.indd 306 5/31/11 7:06 PM

ref erences 307
DNA study. Proceedings of the Royal Society of Thouless, C.R. & K. al Bassri. 1991. Taxonomic status of
London B, 256:1– 6. the Farasan Island gazelle. Journal of Zoology, London
Stockley, C.H. 1923. On the forms of the Himalayan serow 223:151–159.
Capricornis sumatraensis. Journal of the Bombay Tun Yin, U. 1967. Wild Animals of Burma. Rangoon:
Natural History Society 29:824– 827. Rangoon Gazette.
Stoner, C.J., T.M. Caro & C.M. Graham. 2003. Ecological Tuoc, D., V.V. Dung, S. Dawson, P. Arctander & J.
and behavioural correlates of coloration in artiodac- MacKinnon. 1994. Vù môt loài mang lon moi ph{t hiên
tyls: systematic analyses of conventional hypotheses. o Viêt Nam. Thông Tin Khoa Hoc Ky Thuât 3:4–12.
Behavioural Ecology 14:823– 840. Uerpmann, H.P. 1987. The Ancient Distribution of the
Stott, K.[W]. 1959. Giraffe intergradation in Kenya. Ungulate Mammals in the Middle East. Beihefte zum
Journal of Mammalogy 40:251. Tübinger Atlas des Vorderen Orients, Reihe A,
Stott, K.W. & C.J. Selsor. 1981. Further remarks on giraffe Naturwissenschaften, No. 27. Wiesbaden, Germany:
intergradation in Kenya and unreported marking Ludwig Reichert.
variations in reticulated and Maasai giraffes. Mamma- Urbain, A. & H. Friant. 1942. Recherches anatomiques sur
lia 45:261–263. l’antilope royale. Archives du Museum d’Histoire
Sugiri, N. & N. Hidayet. 1996. The diversity and Naturelle, Paris (6) 18:167–179.
haematology of anoa from Sulawesi. Paper presented Vadhanakul, N., M. Tansatit & W. Tunwattana. 2004.
at the Population and Habitat Viability Assessment Karyotype of a Thai gaur sire of Khao Kheow Open
Workshop for the Anoa (Bubalua depressicornis, Zoo. Journal of the Thai Veterinary Medical Associa-
Bubalus quarlesi), 22–26 July 1996, Bogor, Java Barat, tion under Royal Patronage 54:35– 45.
Indonesia, IUCN/SSC Conservation Breeding Valdez, R. 1982. The Wild Sheep of the World. Mesilla,
Specialist Group. NM: Wild Sheep and Goat International.
Swynnerton, C.H. & R.W. Hayman. 1951. A checklist of Valdez, R. 1985. Lords of the Pinnacles: Wild Goats of the
the land mammals of the Tanganyika Territory and the World. Mesilla, NM: Wild Sheep and Goat
Zanzibar Protectorate. Journal of the East Africa International.
Natural History Society 20:274–392. Valdez, R., C.F. Nadler & T.D. Bunch. 1978. Evolution of
Tchernov, E., T. Dayan & Y. Yom-Tov. 1987. The paleoge- wild sheep in Iran. Evolution 32:56–72.
ography of Gazella gazella and Gazella dorcas during the Vanpé, C., J.-M. Gaillard, P. Kjellander, A. Mysterud, P.
Holocene of the southern Levant. Israel Journal of Magnien, D. Delorme, G. van Laere, F. Klein, O.
Zoology 34:51–59. Liberg & A.J.M. Hewison. 2007. Antler size provides an
Thomas, D.C. & P. Everson. 1982. Geographic variation in honest signal of male phenotypic quality in roe deer.
caribou on the Canadian Arctic islands. Canadian American Naturalist 169:481– 493.
Journal of Zoology 60:2442–2454. Varela, D.M., R.G. Trovati, K.R. Guzmán, R.V. Rossi &
Thomas, H., A. Seveau & A. Hassanin. 2001. The J.M.B. Duarte. 2010. Red brocket deer Mazama
enigmatic new Indochinese bovid, Pseudonovibos americana (Erxleben 1777). In J.M.B. Duarte & S.
spiralis, an extraordinary forgery. Comptes Rendus de González (eds.), Neotropical Cervidology: Biology and
l’Académie des Sciences (Paris), Sér. 3, Sciences de la Medicine of Latin American Deer. Jaboticabal, Brazil
Vie 324:81– 86. and Gland, Switzerland: Funep, in collaboration with
Thomas, O. 1892. On the Mammalia Collected by Signor IUCN, 151–159.
Leonardo Fea in Burma and Tenasserim. Viaggio di Vasil’ev, V.A., E.P. Steklenev, E.V. Morozova & S.K.
Leonardo Fea in Birmania e Regioni Vicine, 41. Genoa: Semyenova. 2002. DNA fi ngerprinting of individual
Istituto Sordo-Muti. species and intergeneric and interspecific hybrids of the
Thomas, O. 1898. On the Zululand form of Livingstone’s genera Bos and Bison, subfamily Bovinae. Russian
antelope (Nesotragus livingstonianus). Annals and Journal of Genetics 38:415– 420.
Magazine of Natural History (7) 2:317. Vassart, M., L. Granjon & A. Greth. 1991. Genetic variabil-
Thomas, O. 1906. On a new pygmy antelope obtained by ity in the Arabian oryx. Zoo Biology 10:399– 408.
Col. J. J. Harrison in the Semliki forest. Annals and Vassart, M., L. Granjon, A. Greth, J.F. Asmodé & E.P.
Magazine of Natural History (7) 18:148–150. Cribiu. 1992. Genetic research on captive Arabian oryx
Thomas, O. 1911. The mammals of the tenth edition of in Saudi Arabia. In F. Spitz, G. Janeau, G. Gonzalez,
Linnaeus: an attempt to fi x the types of the genera and et al. (eds.), Proceedings of the International Symposium
the exact bases and localities of the species. Proceed- “Ongulés/Ungulates 91.” Paris: Société Française pour
ings of the Zoological Society of London 1911:120–158. l’Étude et la Protection des Mammifères, 595–598.
Thomas, O. 1917. Preliminary diagnoses of the new Vassart, M., L. Granjon, A. Greth & F.M. Catzefl is. 1994.
mammals obtained by the Yale National Geographic Genetic relationships of some Gazella species: an
Society Peruvian expedition. Smithsonian Miscella- allozyme survey. Zeitschrift für Säugetierkunde
neous Collections 63, 4:1–3. 59:236–245. —-1
—0
349-47558_ch01_1P.indd 307 5/31/11 7:06 PM

Vassart, M., A. Pinton, A. Séguéla & C. Dutertre. 1994. Waddell, P.J., N. Okada & M. Hasegawa, 1999. Towards
New data on chromosomes of peccaries. Mammalia resolving the interordinal relationships of placental
58:500–507. mammals. Systematic Biology 48:1–5.
Vassart, M., L. Granjon & A. Greth. 1995. Genetic study of Wallace, H.F. 1913. The Big Game of Central and Western
Gazella gazella: chromosomal and allozyme data. China. London: John Murray.
Comptes Rendus de l’Académie des Sciences (Paris), Ward, A.E. 1925. The mammals and birds of Kashmir and
Sér. 3, Sciences de la Vie 318:27–33. the adjacent hill provinces, part 2. Journal of the
Vassart, M., A. Séguéla & H. Hayes. 1995. Chromosomal Bombay Natural History Society 30:118–131.
evolution in gazelles. Journal of Heredity 86:216–227. Webb, S.D. 2000. Evolutionary history of New World
Vereschagin, N.K. 1959 [English translation 1967]. The Cervidae. In E.S. Vrba & G.B. Schaller (eds.), Ante-
Mammals of the Caucasus. Leningrad: Izd-vo lopes, Deer, and Relatives. New Haven: Yale University
Akademii Nauk SSSR [English translation by A. Press, 38– 64.
Lerman & B. Rabinovich, Jerusalem: Israel Program Weber, M. & R.A. Medellín. 2010. Yucatán brown brocket
for Scientific Translations]. deer Mazama pandora (Merriam 1901). In J.M.B. Duarte
Verkaar, E.L.C., I.J. Nijman, M. Beeke, E. Hanekamp & & S. González (eds.), Neotropical Cervidology: Biology
J.A. Lenstra. 2004. Maternal and paternal lineages in and Medicine of Latin American Deer. Jaboticabal,
cross-breeding bovine species: has wisent a hybrid Brazil and Gland, Switzerland: Funep, in collaboration
origin? Molecular Biology and Evolution 21:1165–1170. with IUCN, 211–216.
Vila, A.R., C.E. Saucedo Galvez, D. Aldridge, E. Ramilo & Weinberg, P.J. 2002. Capra cylindricornis. Mammalian
P.C. González. 2010. South Andean huemul Hippocam- Species 695:1– 9.
elus bisulcus (Molina 1782). In J.M.B. Duarte & S. Welsch, U., G. van Dyk, D. Moss & F. Feuerhake. 1998.
González (eds.), Neotropical Cervidology: Biology and Cutaneous glands of male and female impalas (Aepyceros
Medicine of Latin American Deer. Jaboticabal, Brazil melampus): seasonal activity changes in secretory
and Gland, Switzerland: Funep, in collaboration with mechanisms. Cell and Tissue Research 292:377–394.
IUCN, 89–100. Wetzel, R.M. 1977. The Chacoan peccary Catagonus
Vogliotti, A. & J.M.B. Duarte. 2010. Small red brocket deer wagneri (Rusconi). Bulletin of the Carnegie Museum of
Mazama bororo (Duarte 1996). In J.M.B. Duarte & S. Natural History 3:1–36.
González (eds.), Neotropical Cervidology: Biology and Wheeler, J.C. 1995. Evolution and present situation of the
Medicine of Latin American Deer. Jaboticabal, Brazil South American Camelidae. Biological Journal of the
and Gland, Switzerland: Funep, in collaboration with Linnaean Society 54:271–295.
IUCN, 172–176. Wheeler, J.C. 1998. Evolution and origin of the domestic
Volf, J. 1976. Some remarks on the taxonomy of the genus camelids. Alpaca Registry Journal 3, 1:1–15.
Nemorhaedus H. Smith 1827 (Bovidae: Rupicaprini). Wilhelmi, E., H.Y. Kaariye & S. Hammer. 2007. Das Al
Vestnik Ceskoslovenska, Spolia Zoologia 40:75– 80. Wabra Wildlife Preservation Dibatag-Projekt
Von Dueben, C. 1846. In C.J. Sundevall, Ny Antelop-art. Beobachtungen in der Region Ogaden, Südost-Äthi-
Öfversigt af Kongliga Vetenskaps-Akademiens opien. ZGAP [Zoologische Gesellschaft für Arten- und
Förhandlingar, Stockholm 3, 7:221–222. Populationsschutz] Mitteilungen, 23:9–12.
Vorobeev, G.G. & J. van der Ven 2003. Looking at Willerslev, E., M. Gilbert, J. Binladen, S. Ho, P. Campos,
Mammals in Kyrgyzia, Central Asia. Bishkek, A. Ratan, L. Tomsho, R. de Fonseca, A. Sher, T.
Kyrgyzstan: PDC. Kuznetsova, M. Nowak-Kemp, T. Roth, W. Miller & S.
Vrba, E.S. 1979. Phylogenetic analysis and classification of Schuster. 2009. Analysis of complete mitochondrial
fossil and recent Alcelaphini (Mammalia: Bovidae). genomes from extinct and extant rhinoceroses reveals
Biological Journal of the Linnaean Society 11:207–228. lack of phylogenetic resolution. BMC Evolutionary
Vrba, E.S. 1980. Evolution, species, and fossils: how does Biology 9:95. doi:10.1986/1471-2148-9-95.
life evolve? South African Journal of Science 76:61– 84. Willows-Munro, S., T.J. Robinson & C.A. Matthee. 2005.
Vrba, E.S. 1997. New fossils of Alcelaphini and Caprinae Utility of nuclear DNA intron markers at lower
(Bovidae: Mammalia) from Awash, Ethiopia, and taxonomic levels: phylogenetic resolution among nine
phylogenetic analysis of Alcelaphini. Palaeontologica Tragelaphus spp. Molecular Phylogenetics and Evolu-
Africana 34:127–198. tion 35:624– 636.
Vrba, E.S. & G.B. Schaller. 2000. Phylogeny of Bovidae Wilson, D.E. & D.M. Reeder. 2005. Mammal Species of
based on behaviour, glands, skulls, and postcrania. In the World, 3rd ed., 2 vols. Baltimore: Johns Hopkins
E.S. Vrba & G.B. Schaller (eds.), Antelopes, Deer, and University Press.
Relatives. New Haven: Yale University Press, 203–222. Wilson, V.J. (ed.). 2002. Duikers of Africa: Masters of the
Vrba, E.S., J.R. Vaismys, J.E. Gatesy, R. DeSalle & K-y. African Forest Floor. Bulawayo, Zimbabwe: Chipan-
Wei. 1994. Analysis of paedomorphosis using allo- gali Wildlife Trust.
-1— metric characters: the example of Reduncini antelopes Witmer, L.M., S.D. Sampson & N. Solounias. 1999. The
0— (Bovidae, Mammalia). Systematic Biology 43:92–116. proboscis of tapirs (Mammalia: Perissodactyla): a case
349-47558_ch01_1P.indd 308 5/31/11 7:06 PM

ref erences 309
study in novel narial anatomy. Journal of Zoology, Železnov, N.K. 1994. Obca snežná (Ovis nivicola Esch.)
London 249:249–267. v sebrobuychodnej Sibíri— Snow sheep (Ovis nivicola
Woodburne, M O. 1968. The cranial myology and Esch.) in north-eastern Siberia. Folia Venatoria
osteology of Dicotyles tajacu, the collared peccary, and 24:161–169.
its bearing on classification. Memoirs of the Southern Zhang F-f. & Z-g. Jiang. 2006. Mitochondrial phylogeogra-
California Academy of Sciences 7:1– 46. phy and genetic diversity of the Tibetan gazelle
Wu C-h., Y-.p. Zhang, T.D. Bunch, S. Wang & W. Wang. (Procapra picticaudata): implications for conservation.
2003. Mitochondrial control region sequence Molecular Phylogenetics and Evolution 41:313–321.
variation within the argali wild sheep (Ovis ammon): Zukowsky, L. 1910. Halbseitige Bastarde bei geogra-
evolution and conservation relevance. Mammalia phischen Rassen von Grosswild aus freier Wildbahn.
67:109–118. Zoologischer Beobachter der Zoologische Garten:
Wu H., Q-h. Wan & S-g. Fang. 2004. Two genetically Zeitschrift für Beobachtung, Pflege, und Zucht der
distinct units of the Chinese sika deer (Cervus nippon): Tiere 51:259–272.
analyses of mitochondrial DNA variation. Biological Zukowsky, L. 1913a. Ergänzungen zu meinen Arbeiten
Conservation 119:183–190. über Connochaetes albojubatus Ths. und Eudorcas
Wurster, D.H. 1972. Sex-chromosome translocations and thomsoni Gthr. Archiv für Naturgeschichte 80:142–146.
karyotypes in bovid tribes. Cytogenetics 11:197–207. Zukowsky, L. 1913b. Ueber einige anscheinend neue
Wyrwoll, T.W. 1999. Eine Neubeachreibung des Süd- Rassen von Connochaetus albojubatus Thomas aus
spanischen Steinbocks. Säugetierkundliche Mitteilun- Deutsch- Ostafrika. Archiv für Naturgerschichte
gen 44:93– 98. 79:76– 91.
Yalden, D.W. 1978. A revision of the dik-diks of the Zukowsky, L. 1924. Ein Wort über die Notwendigkeit der
subgenus Madoqua (Madoqua). Monitore Zoologico systematischen Bearbeitung der Wisentreste. Pallasia
Italiano (n.s.) Supplement 11, 10:245–264. 2, 1:1–11.
Yalden, D.W. 2007. Obituary—Peter Grubb (1944–2006). Zukowsky, L. 1928a. Nachtrag zu den Bemerkingen über
Mammal News 147:4. die rassenweise Verschiedenheit der Hirschziegenanti-
Yalden, D.W., M.J. Largen & D. Kock. 1984. Cata logue of lope. Carl Hagenbecks Illustrierte Tier- und Menschen-
the mammals of Ethiopia, 5: Artiodactyla. Monitore welt 2:147.
Zoologico Italiano (n.s.) Supplement 19, 4:67–221. Zukowsky, L. 1928b. Bemerkungen über die rassenweise
Yang F., N.P. Carter, L. Shi & M.A. Ferguson-Smith. 1995. Verschiedenheit der Hirschziegenantilope. Carl
A comparative study of karyotypes of muntjacs by Hagenbecks Illustrierte Tier- und Menschenwelt
chromosome painting. Chromosoma 103:642– 652. 2:124–127.
Zainuddin, Z.-Z., M.-T. Abdullah & M.-F.M. Suri. 1990. Zukowsky, L. 1961. Über eine verzwergte Inselform des
The husbandry and veterinary care of captive Buschbockes, Tragelaphus scriptus, in Tanganjika.
Sumatran rhinoceros at Zoo Melaka, Malaysia. Zoologischer Garten (n.f.) 26:54–55.
Mammalian Nature Journal 44:1–19. Zukowsky, L. 1965 [t.p. 1964, published 1965]. Die
Zalkin, V.I. 1949. Sistematika sibirskikh gornykh kozlov Systematik der Gattung Diceros Gray, 1821. Der
[Systematics of Siberian ibex]. Biulleten Moskovskogo Zoologische Garten (n.f.) 30:1–178.
Obshchestva Ispytatelei Prirody, Otdel Biologicheskii van Zyll de Jong, C.G. 1986. A Systematic Study of Recent
54:3–21. [In Russian]. Bison, with Par ticu lar Consideration of the Wood
Zalkin, V.I. 1950. [Subspecies of wild goat]. Comptes Bison (Bison bison athabascae Rhoads, 1898). National
Rendus de l’Académie des Sciences de l’URSS 22:325. Museum of Natural Sciences, Publications in Natural
[In Russian]. Sciences No. 6. Ottawa: National Museums of Canada,
Zeder, M.A. 2001. A metrical analysis of a collection of National Museum of Natural Sciences.
modern goats (Capra hircus aegagrus and C. h. hircus) van Zyll de Jong, C.G., C. Gates, H. Reynolds & W. Olsen.
from Iran and Iraq: implications for the study of 1995. Phenotypic variation in remnant populations of
caprine domestication. Journal of Archaeological North American bison. Journal of Mammalogy
Science 28:61–79. 76:391– 405.
—-1
—0
349-47558_ch01_1P.indd 309 5/31/11 7:06 PM

-1—
0—
349-47558_ch01_1P.indd 310 5/31/11 7:06 PM

Ungulate Taxonomy 2011

Uploaded by

Copyright:

Available Formats

Ungulate Taxonomy 2011

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ungulate Taxonomy 2011

Uploaded by

Copyright:

Available Formats

Ung ulate Ta xonomy

349-47558_ch00_1P.indd i 5/31/11 6:06 PM

349-47558_ch00_1P.indd ii 5/31/11 6:06 PM

349-47558_ch00_1P.indd iii 5/31/11 6:06 PM

The Johns Hopkins University Press

[[CIP data to come]]

A catalog record for this book is available from the British

The Johns Hopkins University Press uses environmentally

349-47558_ch00_1P.indd iv 5/31/11 6:06 PM

Theory of Ungulate Taxonomy 1

349-47558_ch00_1P.indd v 5/31/11 6:06 PM

349-47558_ch00_1P.indd vi 5/31/11 6:06 PM

349-47558_ch00_1P.indd vii 5/31/11 6:06 PM

and friends—going back at least 40 years. We would

349-47558_ch00_1P.indd viii 5/31/11 6:06 PM

349-47558_ch00_1P.indd ix 5/31/11 6:06 PM

349-47558_ch00_1P.indd x 5/31/11 6:06 PM

349-47558_ch00_1P.indd xi 5/31/11 6:06 PM

349-47558_ch00_1P.indd xii 5/31/11 6:06 PM

349-47558_ch01_1P.indd 1 5/31/11 7:05 PM

349-47558_ch01_1P.indd 2 5/31/11 7:05 PM

349-47558_ch01_1P.indd 3 5/31/11 7:05 PM

349-47558_ch01_1P.indd 4 5/31/11 7:05 PM

349-47558_ch01_1P.indd 5 5/31/11 7:05 PM

349-47558_ch01_1P.indd 6 5/31/11 7:05 PM

349-47558_ch01_1P.indd 7 5/31/11 7:05 PM

349-47558_ch01_1P.indd 8 5/31/11 7:05 PM

349-47558_ch01_1P.indd 9 5/31/11 7:05 PM

349-47558_ch01_1P.indd 10 5/31/11 7:05 PM

There are no particular problems in the order Perissodactyla, at least

349-47558_ch01_1P.indd 11 5/31/11 7:05 PM

349-47558_ch01_1P.indd 12 5/31/11 7:05 PM

Equus przewalskii Poliakov, 1881

Equus ferus Boddaert, 1785 Equus africanus

349-47558_ch01_1P.indd 13 5/31/11 7:05 PM

domestic donkeys from all over the Old World.

349-47558_ch01_1P.indd 14 5/31/11 7:05 PM

349-47558_ch01_1P.indd 15 5/31/11 7:05 PM

Groves & Bell (2004) could fi nd no significant geo-

Equus grevyi Oustalet, 1882 Equus quagga burchellii (Gray, 1824)

349-47558_ch01_1P.indd 16 5/31/11 7:05 PM

stripes reaching the haunches, and appearing vari-

349-47558_ch01_1P.indd 17 5/31/11 7:05 PM

The facial skeleton is low, so that the nasal cavity

349-47558_ch01_1P.indd 18 5/31/11 7:05 PM

349-47558_ch01_1P.indd 19 5/31/11 7:05 PM

wide channels that are prolonged back into the fron-

349-47558_ch01_1P.indd 20 5/31/11 7:05 PM

349-47558_ch01_1P.indd 21 5/31/11 7:05 PM

The skinfolds are deep, and hang loosely; the hori-

349-47558_ch01_1P.indd 22 5/31/11 7:05 PM

349-47558_ch01_1P.indd 23 5/31/11 7:05 PM

349-47558_ch01_1P.indd 24 5/31/11 7:05 PM

349-47558_ch01_1P.indd 25 5/31/11 7:05 PM

349-47558_ch01_1P.indd 26 5/31/11 7:05 PM

Here, we continue to refer to the even-toed ungulates as the order

349-47558_ch01_1P.indd 27 5/31/11 7:05 PM

349-47558_ch01_1P.indd 28 5/31/11 7:05 PM

349-47558_ch01_1P.indd 29 5/31/11 7:05 PM

349-47558_ch01_1P.indd 30 5/31/11 7:05 PM