The Pleistocene Boundary and The Beginning of The Quaternary (John A. Van Couvering, 2001)

The Pleistocene Boundary and the
Beginning of the Quaternary
WORLD AND REGIONAL GEOLOGY SERIES

Series Editors: M. R. A. Thomson and J. A. Reinemund
This series comprises monographs and reference books on
studies of world and regional geology. Topics include comprehensive studies of key geological regions and the results of recent
IGCP projects.
Geological Evolution of Antarctica - M. R. A. Thomson,

J. A. Crame & J. W. Thomson (eds.)
2 Permo-Triassic Events in the Eastern Tethys - W. C. Sweet,

Yang Zunyi, J. M. Dickins & Yin Hongfu (eds.)
3
The Jurassic of the Cricum-Pacific - G. E. G. Westermann

(ed.)
4 Global Geological Record of Lake Basins, Vol. 1 E. Gierlowski-Kordesch & K. Kelts (eds.)
5 Earth's Glacial Record - M. Deynoux, J. M. G. Miller,
E. W. Domack, N. Eyles, I. Fairchild & G. M. Young
(eds.)
6
Tertiary Basins of Spain: The Stratigraphic Record of

Crustal Kinematics - P. F. Friend and C. J. Dabrio (eds.)
The Quaternary History of Scandinavia - J. Donner
The Tectonic Evolution of Asia - A. Yin and

M. Harrison (eds.)
The Pleistocene Boundary and the Beginning of the
Quaternary - J. A. Van Couvering (ed.)
The Pleistocene
Boundary and the
Beginning of the
Quaternary
Edited by
JOHN A. VAN COUVERING
American Museum of Natural History, New York
Final report of the International Geological Correlation

Program-Project 41: Neogene/Quaternary Boundary
Coordinator: Ksenia V. Nikiforova,

Geological Institute of the Russian Academy of Sciences,
Moscow
Associate Editors:
Emiliano Aguirre,
National Museum of Natural Sciences, Madrid
Mikhail N. Alekseev,
Geological Institute of the Academy of Sciences of Russia,
Moscow
Giancarlo Pasini,
Institute of Marine Geology of the Italian National Research
Council, Bologna
mim
CAMBRIDGE
UNIVERSITY PRESS
PUBLISHED BY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE

The Pitt Building, Trumpington Street, Cambridge, United Kingdom
CAMBRIDGE UNIVERSITY PRESS
The Edinburgh Building, Cambridge CB2 2RU, UK
40 West 20th Street, New York NY 10011-4211, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
Ruiz de Alarcon 13, 28014 Madrid, Spain
Dock House, The Waterfront, Cape Town 8001, South Africa
http://www.cambridge.org
Cambridge University Press 1997
This book is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.
First published 1997
First paperback edition 2004
Typeset in Times Roman
A catalogue record for this book is available from the British Library
Library of Congress Cataloguing-in-Publication Data
The Pleistocene boundary and the beginning of the Quaternary /
edited by John A. Van Couvering : associate editors, Giancarlo
Pasini, Emiliano Aguirre, Mikhail Alekseev.
p. cm. (World and regional geology series)
ISBN 0 521 34115 9 (hardback)
1. Pliocene-Pleistocene boundary. 2. Stratigraphic correlation.
I. Van Couvering, John A. II. Series: World and regional geology.
QE695.5.P57 1997
551.7'92-dc20
95-26529
CIP
ISBN 0 521 34115 9 hardback
ISBN 0 521 61702 2 paperback
9 World and regional geology series
Contents
List of contributors
vii
Preface: the new Pleistocene
Part IV The Pleistocene boundary in regional sequences

xi
11
Foreword
RUGGIERI, SAMUELE SARTONI, AND RODOLFO SPROVIERI
12
Part I Definition of the base of the Quaternary

International Geological Correlation Program, Project
41: "Neogene/Quaternary Boundary"
3
GEDALIAHU GVIRTZMAN, GUIDO M. MARTINOTTI, AND SHIMON
13
The Pliocene-Pleistocene transition in the Iberian

Peninsula
169
14
Biostratigraphy and calibrated climatic chronology of

the Upper Pliocene and Lower Pleistocene of
France
178
15
The Plio-Pleistocene of England and Iceland
Part II Characterization of the Pleistocene boundarystratotype
EMILIANO AGUIRRE
The Pliocene-Pleistocene boundary-stratotype at Vrica,

Italy
15
GIANCARLO PASINI AND MARIA LUISA COLALONGO
The magnetostratigraphy of the Vrica section, Italy, and

its correlation with the Plio-Pleistocene of the Boso
Peninsula, Japan
46
HISAO NAKAGAWA, NOBUAKI NIITSUMA, TOSHIAKI TAKAYAMA,
JEAN CHALINE
16
The Neogene-Ouaternary boundary in The

Netherlands
185
17
The Tertiary-Quaternary boundary in western

Germany
191
WALDO H. ZAGWIJN
TOKUNAGA, HIROSHI KITAZATO, TOYOSABURO SAKAI, AND

ITARU KOIZUMI
Comparison of the laminated units at Vrica and deep-sea

sapropels from the eastern Mediterranean
57
ISABELLA RAFFI AND ROBERT THUNELL
Calcareous nannofossil biochronology and the PliocenePleistocene boundary

63
GUNTHER VON DER BRELIE, KARL BRUNNACKER, WINFRJED

HINSCH, AND HEINZ TOBIEN
18
WIEGANK
79
19
ANDREW J. M. PALMER
The Plio-Pleistocene of Hungary
206
ANDREW A. R6NAI
20
Part HI The paleontological context of the Pleistocene

boundary
The Pliocene-Pleistocene boundary in Romania
216
CONSTANTIN GHENEA
21
The Pliocene and Pleistocene of the European part of

the Commonwealth of Independent States
221
22
The N/Q boundary in Asian Russia and

Tadjikistan
227
23
The Pliocene-Pleistocene boundary in the Indian

subcontinent
232
24
The Pliocene-Pleistocene boundary in Japan: the

Osaka Group, Kinki district
239
The Pliocene-Pleistocene boundary in deep-sea

sediments
87
KSENIA V. NIKIFOROVA
WILLIAM A. BERGGREN AND LLOYD H. BURCKLE, JR.
Late Cenozoic changes of flora in extra-tropical Eurasia

in the light of paleomagnetic stratigraphy
104
VLADIMIR P. GRICHUK
MIKHAIL N . ALEKSEEV
114
EMILIANO AGUIRRE, ELEANORA A. VANGENGEIM,
M. V. A. SASTRY
JORGE MORALES, MARINA V. SOTNIKOVA, AND

VLADIMIR S. ZAZHIGIN
Human evolution in the Plio-Pleistocene interval
The Pliocene-Pleistocene boundary in eastern

Germany
201
HANS-DIETRICH KAHLKE, LOTHAR EISSMANN, AND FRIEDRICH
DOMENICO RIO, ISABELLA RAFFI, AND JAN BACKMAN
Plio-Pleistocene mammal faunas: an overview
183
RICHARD W. HEY
YASUMOCHI MATOBA, MOTOYOSHI ODA, SHIGEMOTO
EMILIANO AGUIRRE
Stratigraphy of the Plio-Pleistocene sequence of the

Mediterranean coastal belt of Israel and its implications
for the evolution of the Nile Cone
156
MOSHKOVITZ
KSENIA V. NIKIFOROVA AND MIKHAIL N . ALEKSEEV
10
141
NAKAGAWA, GIANCARLO PASINI, DOMENICO RIO, GIULIANO
Diatom microfossils from the Vrica section
The Pliocene-Pleistocene boundary in Italy
AUGUSTO AZZAROLI, MARIA LUISA COLALONGO, HISAO
xix
129
MINORU ITIHARA, SHUSAKU YOSHIKAWA, AND TADAO

KAMEI
Contents
VI
25
The Pliocene-Pleistocene boundary in Japan:

stratigraphy in the Boso Peninsula, central Japan
28
244
HISAO NAKAGAWA, NOBUAKI NIITSUMA,
29
The Pliocene-Pleistocene boundary in New

Zealand
273
30
The Pliocene-Pleistocene boundary in continental

sequences of North America
278
TAKASHI MITSUNASHI, MOTOYOSHI ODA,
ANTHONY R. EDWARDS
TOSHIAKI TAKAYAMA, AND TOYOSABURO SAKAI
247
Plio-Pleistocene reference sections in Indonesia

FRANQOIS SEMAH
26
The base of the Quaternary in China
27
Plio-Pleistocene deposits and the Quaternary boundary

in sub-Saharan Africa
254
EVERETT H. LINDSAY
H. BASIL S. COOKE
Index
ZHANG SHOUXIN
291
264
Contributors
Emiliano Aguirre
H. Basil S. Cooke
Museo Nacional de Ciencias Naturales CSIC

Jose Gutierrez Abascal 2
28006 Madrid, Spain
2133, 154th Street

White Rock, B.C., Canada V4A 4S5
Mikhail N. Alekseev
Laboratory of Quaternary Geology

Geological Institute, Russian Academy of Sciences
Pyzhevsky per. 7
Moscow 109017, Russia
Anthony R. Edwards
New Zealand Geological Survey

P.O. Box 30368
Lower Hutt, New Zealand
Lothar Eissmann
Augusto Azzaroli
Museo di Geologia e Paleontologia

Via La Pira 4
50121 Firenze, Italy
Jan Backman
Department of Geology
University of Stockholm
S-106 Stockholm, Sweden
Universitat Leipzig, Sektion Physik

Wissenschaftsbereich Geophysik
TalstraBe 35
0-7010 Leipzig, Germany
Constantin Ghenea
Bd. Banu Hanta 1

Bl. IB Apt. 63
Bucarest, Romania
William A. Berggren
Woods Hole Oceanographic Institution

Woods Hole, MA 02543, USA
Karl Brunnacker
Geologisches Institut der Universitat zu Koln

Zulpicher Str. 49
DW-5000 Koln, Germany
Vladimir P. Grichuk
Paleobotanical Institute
Russian Academy of Sciences
Pzyhevsky per. 7
Gedaliahu Gvirtzman
Lamont-Doherty Geological Observatory

Palisades, NY 10964, USA
Geological Survey of Israel

30 Malkhei Israel Street
Jerusalem 95501, Israel
Jean Chaline
Richard W. Hey
Centre de Paleontologie analytique et Geologie sedimentaire (U.R.A.

CNRS 157)
Laboratoire de Prehistoire et de Paleoecologie du Quaternaire de
l'E.P.H.E.
Centre des Sciences de la Terre
6 Boulevard Gabriel
21000 Dijon, France
Harleton Lodge
Bromsash, Ross-on-Wye
Herefordshire HR9 7SB, United Kingdom
Lloyd H. Burckle, Jr.
Maria Luisa Colalongo
Istituto di Geologia e Paleontologia

Dipartimento di Scienze Geologiche
Universita di Bologna
Via Zamboni 67
40127 Bologna, Italy
Winfried Hinsch
Geologisches Landesamt Schleswig-Holstein

Mercatorstr. 7
DW-2300 Kiel, Germany
Minoru Itihara
Department of Geosciences
Osaka City University
558 Osaka, Japan
vn
Contributors
Hans-Dietrich Kahlke
Motoyoshi Oda
Institut fur Quartarpalaontologie

SteubenstraBe 19a-21
0-5300 Weimar, Germany
Kumamoto University
860 Kumamoto, Japan
Tadao Kamei
Andrew J. M. Palmer
Department of Geology and Mineralogy

Kyoto University
606 Kyoto, Japan
University of South Carolina
Columbia, SC 29208, USA
Hiroshi Kitazato
Giancarlo Pasini
Institute of Geosciences
Shizuoka University
422 Shizuoka, Japan
Itaru Koizumi
Department of Geology and Mineralogy

Hokkaido University
060 Sapporo, Japan
Everett H. Lindsay
University of Arizona
Tucson, AZ 85721, USA
Istituto di Geologia Marina del C.N.R.

Via Gobetti 101
Isabella Raffi
Servizio Geologico Nazionale

Largo S. Susanna 13
00187 Roma, Italy
Domenico Rio
Dipartimento di Geologia, Paleontologia e Geofisica

Universita di Padova
35137 Padova, Italy
Guido M. Martinotti

Yasumochi Matoba
Institute of Mining Geology

Akita University
010 Akita, Japan
Takashi Mitsunashi
Shimane University
690 Matsue, Japan
Jorge Morales
Museo Nacional de Ciencias Naturales CSIC

Jose Gutierrez Abascal 2
28006 Madrid, Spain
Shimon Moshkovitz

Hisao Nakagawa
Institute of Geology and Paleontology

Tohoku University
980 Sendai, Japan
Nobuaki Niitsuma
Andrew A. Ronai
Magyar Allami Foldtani Intezet

Nepstadion lit. 14
Postafiok 106
H 1442 Budapest, Hungary
Giuliano Ruggieri

Universita di Palermo
Palermo, Sicilia, Italy
Toyosaburo Sakai
Department of Earth Sciences

Utsunomiya University
321 Utsunomiya, Japan
Samuele Sartoni

Universita di Bologna
Via Zamboni 67
M. V. A. Sastry
Geological Survey of India
A-6 Atreya Apartments, 44
15th Cross, Malleswaram
Bangalore-560003, India
Francois Semah
Institute of Geosciences
Shizuoka University
422 Shizuoka, Japan
Institut de Paleontologie Humaine

1 rue Rene Panhard
75013 Paris, France
Ksenia V. Nikiforova
Marina V. Sotnikova
Geological Institute
Pyzhevsky per. 7
Pyzhevsky per. 7
Contributors
Rodolfo Sprovieri
Giinther von der Brelie

Universita di Palermo
Palermo, Sicilia, Italy
Marmstorferweg 46
DW-2100 Hamburg-Harburg, Germany
Friedrich Wiegank
Toshiaki Takayama
Kanazawa University
920 Kanazawa, Japan
Zentralinstitut fur Physik der Erde

Telegrafenberg
0-1500 Potsdam, Germany
Shusaku Yoshikawa
Robert Thunell
Department of Geological Sciences

University of South Carolina
Columbia SC 29208, USA
Osaka City University
558 Osaka, Japan
Waldo H. Zagwijn
Heinz Tobien (deceased)
Palaontologisches Institut der Universitat Mainz

Saarstr. 21
DW-6500 Mainz, Germany
Geological Survey of The Netherlands

Spaarne 17
2000 AD Haarlem, The Netherlands
Vladimir S. Zazhigin
Shigemoto Tokunaga
Palynosurvey Co., Ltd.

103 Tokyo, Japan
Pyzhevsky per. 7
John A. Van Couvering
Micropaleontology Press
American Museum of Natural History
New York, NY 10024, USA
Eleanora A. Vangengeim
Pyzhevsky per. 7
Zhang Shouxin
Institute of Geology
Academia Sinica
P.O. Box 634
Beijing 100029, People's Republic of China
IX

Nailing down the golden spike
The goal of the International Geological Correlation Program,

Project 41 (IGCP-41), was to propose a formal definition for the
base of the Pleistocene Series - by international agreement the
lower (and only) boundary of the Quaternary subsystem - in
accordance with the resolution passed by the XVIII International Geological Congress (IGC) in London in 1948 (King and
Oakley, 1950), as well as to characterize and globally correlate
this boundary. After years of study and discussion, as described
by K. V. Nikiforova and M. N. Alekseev in the first chapter of
this volume, the members of IGCP-41 agreed on a boundary
defined by a physical reference point or "golden spike" within
the outer-shelf marine section exposed at Vrica, near Crotone,
Calabria. As this book demonstrates, the members of IGCP-41
were able to recognize this level in marine and nonmarine
sections around the world. The Vrica boundary-stratotype for
the Pleistocene boundary was ratified and adopted by the
international authorities and has gained approval as a global
boundary-stratotype section and point (GSSP) (Cowie and
Bassett, 1989).
The participants in IGCP-41 may take pride in the establishment, at long last, of an internationally accepted and unambiguous definition for the base of the Pleistocene. With regard to this
momentous advance, this book has three aims: (1) to document
the Vrica boundary-stratotype; (2) to show how the boundarystratotype can be characterized in terms of the paleontological
and paleoenvironmental record, radiometric dating and magnetostratigraphy; and (3) to describe the worldwide stratigraphic
context of the boundary.
corrected in the text to the current orbitally tuned calibration of

the geomagnetic polarity time scale (GPTS) (Table 1), although
some figures could not be revised.
In editing this work, I have held to the general principle that
authors should be allowed to speak with their own voices in
regard to matters of taste such as "planktic" versus "planktonic,"
"ostracod" versus "ostracode," and "-logic" and "-graphic" (as in
geologic and stratigraphic) versus "-logical" and "-graphical." As
for the problems introduced by my attempts to standardize the
terms and usages in this volume, I am grateful for the patience of
my colleagues.
Modern stratigraphic philosophy is based on the principle that
base defines boundary, which is to say that the base or beginning
of each unit also defines the top of the preceding unit (Salvador,
1994). In view of this, the terms "Pliocene-Pleistocene boundary" or "Neogene/Quaternary boundary" actually mean the
same as "Pleistocene boundary" or "Quaternary boundary," and
both usages are found in this work as space or taste dictates.
With regard to punctuation, many authors have used the virgule,
as in "Pliocene/Pleistocene," rather than the dash, as in
"Pliocene-Pleistocene," to express the boundary between two
units such as these. The older term is written first, as is
stratigraphically logical, but unfortunately the line of the virgule
symbolizes a boundary, and its slant might be taken to suggest
that the older unit rests on the younger. For this reason, I have
preferred the dash where there was a choice. The term "PlioPleistocene" is restricted herein to mean a body of strata or time
that includes both Pliocene and Pleistocene parts.
Is there a "Neogene-Quaternary" boundary?
Editorial history and notes on style. The great majority of

manuscripts for this book were in hand by the end of 1984,
although a few solicited chapters and revisions were submitted in
1987. Progress in preparing the book was regrettably slow, but all
manuscripts were circulated for updating in 1992, and most
chapters (notably Chapter 2, on Vrica itself, and Chapter 10, on
the rapidly changing context of human evolution) also incorporate even more recent information. In particular, age determinations derived from the paleomagnetic time scale have been
The International Union of Geological Sciences (IUGS) currently recognizes the units "Paleogene," "Neogene," and "Quaternary" as the three periods making up the Cenozoic era (Cowie
and Bassett, 1989). The expression "Neogene-Quaternary," or
its abbreviation "N/Q," which appears throughout this work to
indicate the chronostratigraphic boundary defined by the base of
the Pleistocene epoch, is therefore correct according to international guidelines. The philosophy and history behind the
Xll
terminology suggest, however, that this arrangement is vulnerable and may be changed in the future.
It has been pointed out by Berggren and Van Couvering (1974,
1979) that the original mid-nineteenth-century meaning of
"Neogene," which Moritz Homes (1853) coined for the marine
fossils of the Vienna Basin, does not agree with the sense of the
term as used by M. Gignoux (1913) in the first decade of this
century. I am greatly indebted to Prof. F. Steininger (personal
communication, 1991) for pointing out that Homes based his
term on the "meiocan" and "pleiocan" local faunal units outlined
by Bronn in 1838. Admittedly Bronn (1838) was extending to the
Vienna Basin the terms "Miocene" and "Pliocene," which had
just been introduced by Charles Lyell (1833), but Homes,
writing in 1853, made it clear that it was Bronn's units, rather
than Lyell's (contra Harland et al., 1990), that he considered to
be basic to the Neogene. In so doing, Homes ignored Lyell's
post-1838 revisions regarding the limits of his epochs, not to
mention Beyrich's studies, which were soon to lead to the formal
proposal of the Oligocene in 1855. Thus, the original Neogene,
like the original Miocene and Pliocene on which Bronn based his
units, encompassed the entire post-Eocene fossil record. Homes
specifically included collections from the glacial "Loss-" and
"Diluvial-Bildungen" in Austria. Furthermore, in correlating the
Vienna Basin Neogene assemblages with Mediterranean faunas,
Homes (1853, p. 809) mentioned collections from Rhodes,
Crete, and Sicily that would now be dated to the Pleistocene.
The term "Neogene" was thereafter much neglected until
Gignoux (1913) reintroduced it. In its revised form, it appears to
have been nothing more than a casual literary convenience
(Gignoux, 1955, p. 467), with no definition beyond the bald
statement that it consisted of the Miocene and Pliocene epochs
together. Despite the superficial resemblance of this diagnosis to
the original, Gignoux's meaning for "Neogene" was actually
quite different, because Homes did not recognize the Pleistocene
as separate from the Pliocene, whereas Gignoux, writing 60 years
later, took this separation for granted. Furthermore, although
Homes, like Lyell, defined his unit in terms of the fossil record,
unlike Lyell he seems to have ignored its application to geologic
time, as is suggested by his choice of "-gen" rather than "-cen" as
an ending. Gignoux made no mention of Hornes's faunal
definition (indeed, he made no mention of Homes at all) in
treating the Neogene as a compound of epochs, or chronostratigraphic units. In that rather careless way the Pleistocene
(and Oligocene, for that matter) was excluded from the Neogene
without discussion, and it was in this "modern" sense that the
term was used by IGCP-41 and eventually by the IUGS. Taking
all this into consideration, Berggren and Van Couvering (1974,
1979) contended that the Neogene, in its original sense, was a
biochronological unit irrelevant to the time-rock unit of the
Quaternary, and chronologically overlapping it.
No post-Neogene unit is specified in the definition of Neogene
given by Gignoux, let alone by Homes. On the other hand, the
most appropriate term for the unit preceding the Quaternary is,
of course, "Tertiary" and if the one is valid it is difficult to see
why the other is not. Indeed, the 1948 commission (King and
Oakley, 1950) stated that it was the Tertiary-Quaternary

boundary which would be identified with the PliocenePleistocene boundary in its recommendation.
Gignoux's revised concept has been accepted by those who
advocate that "Tertiary" should be dispensed with, and that
"Neogene" and "Paleogene" should be used in its place. Harland
et al. (1990) proposed to effect this change by elevating both
"Tertiary" and "Quaternary" to the functionless rank of sub-era,
where both antique concepts could be expected to fade away. To
follow the Neogene, they introduced "Pleistogene" in place of
"Quaternary" as the final system/period of the Cenozoic. On the
other hand, as noted earlier, the approach of the IUGS (Cowie
and Bassett, 1989) is to leave "Quaternary" as is, sequential to
the Neogene, although it makes a somewhat inappropriate name
for a third (alas, not fourth!) division of the Cenozoic.
A third proposal for the name of a unit to follow the Neogene
(sensu Gignoux), with more historical and aesthetic merit than
either of the preceding two, is the term "Anthropogene," as
described in K. V. Nikiforova's Foreword to this volume.
Though it originated as a conceptual definition of a geological
interval, it is no different from Paleogene and Neogene in that
regard, and in the end, as a chronostratigraphic unit, it would be
defined by the base of the Pleistocene, just as the equivalent
Pleistogene or Quaternary would.
New geochronology at the Pleistocene boundary
Upper limit of the Olduvai subchron. Pasini and Colalongo

(Chapter 2, this volume) detail a minor but important change in
the position of the Pleistocene boundary-stratotype at Vrica with
regard to the conventional paleomagnetic model. The age of the
top of the Olduvai normal-polarity subchron has been difficult to
fix, because of a mixed-polarity "flutter" in the transition
(Hilgen, 1991). Analysis of closely spaced samples by Zijderveld
et al. (1991) in the Vrica section showed that a thin normalpolarity zone is present above marker e, which in turn is slightly
above the reversal which was identified as the "top of the
Olduvai" by Tauxe et al. (1983) and Nakagawa et al. (Chapter 3,
this volume; Pasini and Colalongo, Chapter 2, this volume). This
brief normal-polarity interval is not to be confused with the Cobb
Mountain event (Turrin et al., 1994) or with other short normalpolarity excursions in the upper Matuyama chronozone which
have been noted by various workers (Azzaroli et al., Chapter 11,
this volume), including the N3 unit above marker s at Vrica
(Tauxe etal., 1983).
With this new analysis, a slight modification is required for
paleomagnetic characterization of the Pleistocene boundarystratotype at Vrica. Although evidence for a complex structure
in the upper Olduvai has been noted in a few instances
(Zijderveld et al., 1991, p. 711), most published paleomagnetic
profiles that show the Olduvai subchronozone have been in more
condensed (or less closely sampled) sections, in which the short
reversed section that contains marker e at Vrica was not
resolved. Thus, as Zijderveld et al. (1991) reasoned, the thin
normal-polarity zone above the main zone has always been

included in the Olduvai wherever this upper zone has been
sampled, and to consider it now as a separate subchron is not
useful or practical. In view of this, the Vrica boundary should
now be described as being "just below" rather than "just above"
the top of the Olduvai. The time difference is not great (Table 1),
and all of the associated biochronological markers cited in this
work remain valid (Pasini and Colalongo, Chapter 2, this
volume).
Table 1. Comparison of the paleomagnetic time scale in use 10
years ago and the current, orbitally tuned time scale.
Age, Ma
(1985)
Age, Ma
(1995)
Matuyama-Brunhes
0.73
0.780
Jaramillo top
Jaramillo base
0.90
0.97
0.990
1.070
1.186
1.65
1.770
1.785
1.796
1.815
1.950
Magnetostratigraphic
boundary
*Cobb Mt. (Gils*)

Olduvai top
Vrica zone (3 top
-> P/P boundary
Vrica zone p base
Olduvai base
1.65
0-65)
1.82
Astronomically tuned time scale. Mathematical models of longterm variations in global insolation values, as predicted on the
basis of harmonics in the obliquity of the earth's axis and the
ellipticity and precession of its orbit, have been refined with the
aid of modern computers (Berger and Loutre, 1988). This has
brought fresh success in calibrating the proxy records of
astronomically forced climatic periodicities to the magnetostratigraphy and biostratigraphy in marine deposits (Hilgen
and Langereis, 1989; Shackleton, Berger, and Peltier, 1990;
Shackleton et al., 1995a). The application of the new astronomically calibrated time scale to the Vrica section is discussed by
Pasini and Colalongo (Chapter 2, this volume), but has not been
consistently applied in the descriptions of other Plio-Pleistocene
sequences in this volume. Table 1 is a reference to the "old"
versus "new" values, as an aid to the reader.
Standard global Plio-Pleistocene chronostratigraphic
units
The adoption of a stratigraphically defined lower boundary of

the Pleistocene opens the possibility for an improved worldwide
standard chronostratigraphy for the Plio-Pleistocene interval.
Global standard stages have recently been established or
proposed, based on the same marine sequences in southern Italy
that comprise the type area of the Pleistocene. Figure 1
summarizes the consensus from a recent meeting of Italian
stratigraphers (Van Couvering, 1995).
Calabrian versus Selinuntian. For some time now, the lowermost

stage of the Pleistocene has been the subject of controversy. The
1948 resolution (King and Oakley, 1950) invoked a basic
principle
of chronostratigraphy in advocating that "the PlioceneReunion top
2.140
1.99
Pleistocene boundary should be based on changes in marine
2.02
2.150
Reunion base
faunas," and to that end recommended that "the lower Pleistocene should include as its basal member in the type area the
Gauss-Matuyama
2.581
2.48
Calabrian formation (marine)" and that "according to evidence
given this usage would place the boundary at the horizon of the
2.92
3.040
Kaena top
first indication of climatic deterioration in the Italian Neogene
3.110
2.99
Kaena base
succession" (King and Oakley, 1950, p. 214).
Gignoux did not specify a type section for the Calabrian.
3.08
3.220
Mammoth top
According to correspondence seen by G. B. Vai (personal
3.18
Mammoth base
3.330
communication, 1994), he purposely declined to do so on the
grounds that it was impossible to consider any single exposure to
be an adequate example of the complete "cycle." In order to
Note: The earlier time scale (Berggren, Kent, and Van convert the Calabrian to a standard chronostratigraphic unit,
Couvering, 1985) gives values that average 6% older than the Selli (1971) selected one of Gignoux's described sections, Santa
orbitally tuned time scale (Cande and Kent, 1995) at each point. Maria di Catanzaro, as the stage stratotype, but this only proved
The Cobb Mountain subchron is dated according to Turrin et al. Gignoux's point. Ruggieri and Sprovieri (1977), as discussed by
(1994). The age limits for the Vrica interval (3 were calculated Rio, Raff], and Backman (Chapter 5, this volume) and Azzaroli
from the upper and lower age limits of the Olduvai subchron, et al. (Chapter 11, this volume), demonstrated that the Santa
assuming a constant sedimentation rate in the Vrica sequence Maria di Catanzaro section represents only the uppermost part
(Pasini and Colalongo, Chapter 2, this volume). In 1985 (column of Gignoux's (and the 1948 subcommission's) concept of the
1) the accepted age of 1.65 Ma for the top of the Olduvai Calabrian and is stratigraphically far above levels where it might
subchron was erroneously applied to the top of the normal zone be expected that the evidence for the end-Pliocene climate
change would be found. Furthermore, micropaleontological
below interval (3 at Vrica.
XIV
CO
Q-
iS
C/3
Ma o
GSSP to be proposed
(Taranto Area, Puglie)
in relation to the base of the
"Strombus bed"
1
0.1 0.2
0.30.4-
0 5Z
0.6-
0.9
10
z
cc
LU
1
<
LU
Z
LU
O
GSSP to be proposed
(Montalbano lonico, Basilicata)
close to isotopic stage 25 and/or
small Gephyrocapsa/P lacunosa
nannofossil zonal boundary
O
CO
_
LU
-1
CU
ICIILI/
0.8-
>-
cc
<
z
cc
CD
1.4
CO
z
<
MIL
07
LU
15
<
z
<
z
cc
GSSP to be proposed
(Montalbano lonico, Basilicata or
Vrica, Calabria)
close to FO T. trucatulinoides
excelsa
GSSP proposed
(Vrica, Calabria)
Pasini & Colalongo, 1994
LU
z
CO
1.9-
GSSP ratified
(Vrica, Calabria)
Bassett, 1985
Aguirre& Pasini, 1985
2.0-
z
<
2.1 -
LU
22
LU
LU
if)
<
2.3
O
O
LU
2.4-
LU
LU
O
~
Z
2.5
2.6
GSSP proposed
(S. Nicola, Sicily)
Rioetal, 1994
K
1 Unnamed (Tarentian?); 2: Tyrrhenian
Figure 1. Chronostratigraphic units in the PlioPleistocene sequence of

the Gulf of Taranto, southern Italy. Locations for GSSPs at the bases of
the newly proposed Gelasian and Ionian stages and those of the
revalidated Calabrian and its substages are shown according to the consensus at the 1994 Bari workshop (Van Couvering, 1995).
correlations have indicated that this section is in fact equivalent

to strata in Sicily that had been regarded as typical of the piano
siciliano of Doederlein since the nineteenth century (Ruggieri
and Sprovieri, 1976; Rio et al., Chapter 5, this volume). Because
Gignoux (1913) considered that the Sicilian "cycle" followed that
of the Calabrian, the validity of the Santa Maria di Catanzaro
stratotype came into question, and thus, by definition, the
validity of the term "Calabrian" itself (Ruggieri and Sprovieri,
1977, 1979; Ruggieri, Rio, and Sprovieri, 1984).
The approach of IGCP-41 was to defer the problem presented
by the stratotype of the Calabrian Stage and to follow the spirit,
rather than the letter, of the 1948 IGC resolution in proposing to

locate the Pliocene-Pleistocene boundary at Vrica. The strata in
the Vrica section are in the same sub-Apennine formation that is
partially exposed at Santa Maria di Catanzaro and Le Castella
(Haq, Berggren, and Van Couvering, 1977) to the west, and
although Vrica was not mentioned by Gignoux, it clearly
demonstrates the concept of a change from warm- to coldclimate deposition at the end of the Pliocene, which was the basis
of his Calabrian Stage.
In consideration of their view that the Calabrian was an
"invalid junior synonym," Ruggieri and Sprovieri (1979) proposed the creation of two new pre-Sicilian stages/ages for the
lowest Pleistocene, the Santernian and Emilian, based on
sections in the Santerno River basin near Bologna. Those stages,
together with the Sicilian, were considered to be components of
a new Lower Pleistocene superstage, the Selinuntian, in place of
the Calabrian sensu largo. Later, Ruggieri et al. (1984) revised
the status of the Selinuntian, defining it as a single, composite
stage/age made up of the Santernian, Emilian, and Sicilian at
substage/chronozone rank. This position was also taken by Rio
et al. (Chapter 5, this volume) and Azzaroli et al. (Chapter 11,
this volume). In effect, the Selinuntian defined in this way
occupies the same interval, and carries the same meaning, as the
historical concept of Calabrian.
The question is not yet fully resolved, but many Italian
stratigraphers involved in the debate now favor a return to the
historical usage (Van Couvering, 1995) in which the status of the
Calabrian is restored as the lowest stage of the Pleistocene in
Italy, by recognizing that its lower boundary-stratotype is
established at Vrica. One primary reason is the fact that,
according to international guidelines (Salvador, 1994), this
recognition may be merely a formality. Two points must be kept
in mind: that chronostratigraphic units are fundamentally hierarchical, and that a boundary-stratotype may be defined separately
from a body-stratotype. Few stages have such explicit hierarchical status as the Calabrian, which was specified as the basal unit
of the Pleistocene in the same international resolution that
justified the proposals made by IGCP-41. We must therefore
agree, according to the concept of hierarchy outlined in the
international guidelines, that any definition of the base of the
Pleistocene Series cannot exist apart from the base of the
Calabrian Stage. The basal limit of the Calabrian Stage would
thus have been removed axiomatically from the base of the bodystratotype at Santa Maria di Catanzaro to a boundary-stratotype
at the level of the claystone overlying marker e at Vrica, by the
act of proposing this level as the definition for the base of the
Pleistocene Series. It should be noted that the problem of
synchronicity or "priority" between the body-stratotype of the
Calabrian at Santa Maria di Catanzaro and the typical Sicilian at
Ficarazzi disappears when the Sicilian is considered as a
substage, as proposed by Ruggieri et al. (1984).
New stages and boundary proposals. Recent advances in micropaleontology, magnetochronology, and stable-isotope cyclostratigraphy in the upper Cenzoic marine strata in southern Italy
xv
support the recognition of new Pliocene and Pleistocene global Ficarazzi (Sicily), which is near the base of the "small"
chronostratigraphic units in the same context as the Pleistocene Gephyrocapsa zone.
boundary-stratotype. Rio, Sprovieri, and Thunell (1991) have
conclusively demonstrated that in the upper Pliocene, the
Lowering the Quaternary
stratotype of the Piacenzian Stage is erosionally truncated near
the Gauss-Matuyama boundary, some 0.8 m.y. prior to the base Years of effort have culminated in the establishment of an
of the Pleistocene as presently recognized. Under the interna- internationally accepted Pleistocene boundary-stratotype coincitional guidelines (Salvador, 1994), chronostratigraphic "gaps" dent with a major climatic downturn. This level, near the top of
such as that are eliminated by considering the top of a unit to be the Olduvai subchron, with an age of about 1.8 Ma, is (by
defined by the base of the next succeeding unit, so that the definition) the Quaternary boundary as well. It is now well
Piacenzian Stage could be considered to continue up to the base known that the shift from the equable, stable climates of the
of the lowest stage in the Pleistocene. (An alternative proposal, early Pliocene to the intensely seasonal and highly cyclic climates
to move the base of the Pleistocene downward to fill the gap, is of the late Pleistocene and Recent proceeded episodically, with
not in accord with the guidelines, as discussed later). Rio, progressive step-like increments, from the middle Miocene
Sprovieri, and Di Stefano (1994) pointed out that the climatic maximum to modern "fully glacial" conditions by about
paleobiological and paleoclimatic characteristics of the gap in 0.4 Ma (Thunell and Williams, 1983). Major steps toward the
question, as seen in the well-studied section at Monte San Nicola present glacial-climate condition have been confirmed in highin Sicily, distinguish it from the typical Piancenzian. Those quality deep-sea records at 3.2 Ma and 2.5 Ma (Shackleton, Hall,
authors proposed a new stage, the Gelasian, to embody the and Pate, 1995b), prior to the 1.8-Ma downturn, and other major
Upper Pliocene and to more clearly document the sharp change steps at 0.9 Ma and 0.4 Ma (Shackleton et al., 1990; Hilgen,
from warm-climate to cold-climate conditions at the level of the 1991). A detailed review is beyond the scope of this brief
Vrica boundary.
commentary, but it is safe to say that whereas a given step or two
The Ionian Stage has been proposed for the Middle Pleisto- may be strikingly distinct in one or more of the proxy records
cene by a group led by Neri Ciaranfi at Bari University (Van that have been developed - stable isotopes, vertebrate fossils,
Couvering, 1995), to be based on the upper part of a thick land microflora and macroflora, marine planktic and benthic
sequence of highly fossiliferous Lower and Middle Pleistocene microfossils, marine and continental epiglacial sediments - no
marine clayey silts at Montalbano Ionico in southern Basilicata one climatic downturn stands out from all the rest on a consensus
(Ciaranfi et al., 1994). Preliminary studies of the sequence have basis. Because the step at 1.8 Ma is not universally dominant, the
shown continuous sedimentation from the middle Matuyama to effects of the two most closely bracketing steps have also been
the middle Brunhes, in the context of calcareous nannoplankton advocated as criteria for the Pliocene-Pleistocene boundary.
zones of "large" Gephyrocapsa, "small" Gephyrocapsa, and
For example, Nikiforova describes, in the Foreword to this
Pseudoemiliana lacunosa (Rio, Raffi, and Villa, 1990). In order volume, how researchers in eastern Europe and Russia have long
to avoid the problems raised by the fact that cold-climate held (with some justification) that the establishment of lowland
lowstands are erosional in shallow-marine and continental glaciers on the European continent at about 0.9 Ma (i.e., isotope
deposits, the boundary-stratotype of the Ionian has been stage 24) (Hilgen, 1991) marked the proper beginning of the
proposed at the level of the last warm, transgressive event prior Pleistocene, in accord with the original intentions of nineteenthto the Menapian "glacial Pleistocene," at the 0.9-Ma paleo- century stratigraphers. Itihara et al. (Chapter 24, this volume)
climatic step, and the end of the Villafranchian (Azzaroli et al., show that Japanese stratigraphers have also dated the base of the
Chapter 11, this volume). This is isotope stage 25, just above the Pleistocene to that time, coincident with the extinction of typical
Jaramillo and correlative to the base of the P. lacunosa zone Pliocene macroflora.
(Castradori, 1993), which is well documented at Montalbano
On the other hand, the glacial maximum at about 2.5 Ma (i.e.,
Ionico.
isotope stage 100) (Shackleton et al., 1995b) appears to have
Boundary-stratotypes for the Santernian, Emilian, and Sicil- been the first to have noticeably impacted the continental and
ian, which have been designated as Lower Pleistocene substages shallow-marine environments, and it was also the earliest cold(Ruggieri et al., 1984), have been proposed in southern Italy climate phase in which the limits of ice-rafted debris (IRD)
(Figure 1). The base of the Santernian substage is identified, by expanded beyond the Arctic Ocean and into the northern
hierarchic necessity, with that of the Calabrian Stage and that of Atlantic and northern Pacific. Workers in China, as noted by
the Pleistocene Series at the Vrica GSSP. The boundary- Zhang (Chapter 26, this volume), were accustomed to using the
stratotype of the Emilian has been defined by Pasini and earliest loess, together with evidence for the development of the
Colalongo (1994) at a point 71 m stratigraphically above the Villafranchian mammal fauna, as indicators that the Pleistocene
Pleistocene boundary in the Vrica section, at the level of epoch began at levels now dated to 2.5 Ma. Likewise, evidence
Hyalinea baltica FAD and the base of the "large" Gephyrocapsahas been cited for distinct changes in continental mammal faunas
zone. A boundary-stratotype for the Sicilian has not been at that time in western Europe (Azzaroli et al., Chapter 11, this
designated, but Ruggieri et al. (1984) suggested the first volume), North America (Lindsay, Chapter 30, this volume),
appearance level of Globorotalia truncatulinoides excelsa FAD atand Africa (Cooke, Chapter 27, this volume). Although the next
XVI
major cold-climate period, at about 1.8 Ma, was more intense,

its effects were less dramatic, in the context of a biosphere that
had already been "winterized" to a significant extent.
Historically, as Berggren and Van Couvering (1979) described,
agreement on the Pliocene-Pleistocene boundary was a vexatious problem for many years, because researchers in different
fields, like the blind men examining the elephant, were unable to
reconcile their separate understandings. The physical reference
point, or "golden spike," adopted by the 1948 London IGC was a
logical and practical solution, with the goal of shifting the
argument from the debating room to the outcrop. Nevertheless,
the ideology of the ice ages continues to exert a fascination. The
primary example is opposition to the Vrica definition (e.g.,
Zagwijn, 1992; Sibrava, 1992), on the ground that the coldclimate maximum in the 2.5-Ma step, being of greater extent and
severity than those known earlier in Norway and Iceland, was the
true "first glacial." It is worth noting, in this regard, that the
extent of ice-rafting and the latitudinal distribution of indicator
biota suggest that at its coldest, the 2.5-Ma "glacial" was warmer
than the present-day interglacial. Furthermore, between 2.5 and
1.8 Ma, the Tiglian phase (approximately equivalent to the
marine Gelasian Stage) could be characterized as "the last
preglacial," because it was characterized by global climates that
were significantly warmer and more equable than those in any
Pleistocene cycle that followed. The conundrum faced by the
climatically correct is thus to decide whether it offends more that
the Pliocene should include an interval of relatively cold climate
if the Pliocene-Pleistocene boundary is at 1.8 Ma, or that the
Pleistocene should include a clearly pre-glacial interval if the
boundary is at 2.5 Ma.
That such considerations should be brought forward as reason
to question the 1948 resolution and to reopen the boundary wars
is, however, quite another matter. There is no necessary
relationship between chronostratigraphy and past climate (Salvador, 1994), even though the identity of the Pleistocene epoch has
been gravely debated as if it were a paleoclimatic unit ever since
Edward Forbes (1846) noticed that it was also a time of "ice
ages." The truth is that whereas it is pleasing, and even logical,
to make an epoch boundary coincident with a notable
geohistorica! event such as a change in global climate, it cannot
be made a requirement in chronostratigraphy. After three
decades of conscientious and painstaking study, an international
agreement has been reached to the effect that the intent of the
1948 resolution has been fulfilled with the selection of a physical
boundary-stratotype at Vrica. It is our hope that this volume will
demonstrate that in addition to being legitimate, the Vrica
boundary-stratotype is in fact as practical and appropriate as any
other.
References
Aguirre, E., and Pasini, G. 1985. The Pliocene-Pleistocene

boundary. Episodes 8:116-120.
Bassett, M. G. 1985. Towards a common language in
stratigraphy. Episodes 8:87-92.
Berger, A. L., and Loutre, M. F. 1988. New isolation values for

the climate of the last 10 Myr. Louvain-la-Neuve: Univ.
Cathol. Louvain, Inst. Astron. Geophys. G. Lemaitre.
Berggren, W. A., Kent, D. V, and Van Couvering, J. A. 1985.
Neogene geochronology and chronostratigraphy. In The
chronology of the geological record, ed. N. J. Snelling, pp.
211-250. Geological Society of London, Memoir 10.
Oxford: Blackwell.
Berggren, W. A., and Van Couvering, J. A. 1974. The Late
Neogene: biostratigraphy, geochronology and paleoclimatology of the last 15 million years. The Hague: Elsevier.
Berggren, W. A., and Van Couvering, J. A. 1979. The
Quaternary. In Treatise on invertebrate paleontology, Part
A: introduction, ed. R. A. Robison and C. Teichert, pp.
A505-A543. Boulder: Geological Society of America.
Bronn, H. G. 1838. Lethaea geognostica, oder Abbildungen und
Beschreibungen der fur die Gebirgs-Formationen bezeichnendsten Versteinerungen. Zweiter Band: das Kreide und
Molasse-Gebirge enthaltend, pp. 545-1157. Stuttgart: E.
Schweizbart's Verlag.
Cande, S. C , and Kent, D. V. 1995. Revised calibration of the
geomagnetic polarity time scale for the late Cretaceous
and Cenozoic. /. Geophys. Res. 100:6093-6095.
Castradori, D. 1993. Calcareous nannofossil biostratigraphy and
biochronology in eastern Mediterranean deep-sea cores.
Riv. Ital. Paleontol Strat. 99:107-126.
Ciaranfi, N., et al. 1994. Escursione B3: la sezzione di
Montalbano Jonico in Basilicata. Guide alia escursione,
Congresso della Societa Geologico Italiana, Bari (1994).
Quaderne Bibl. Prov. Matera 15:117-156.
Cowie, J. W., and Bassett, M. G. 1989. International Union of
Geological Sciences 1989 stratigraphic chart. Episodes Yl:
(unpaginated insert).
Forbes, E. 1846. On the connection between the distribution of
the existing fauna and flora of the British Isles and the
geological changes which have affected their area, especially
during the epochs of the Northern Drift. Memoir no. 1.
London: Geological Survey of Great Britain.
Gignoux, M. 1913. Les formations marines pliocenes et
quaternaires de lTtalie du Sud et de la Sicile. Univ. Lyon,
Ann., n.s. l(36):l-633.
Gignoux, M. 1955. Stratigraphic geology (translation of the fifth
edition). San Francisco: Freeman.
Haq, B. U , Berggren, W. A., and Van Couvering, J. A. 1977.
Corrected age of the Plio-Pleistocene boundary. Nature
269:483-488.
Harland, W. B., Armstrong, R. L., Cox, A.V, Craig, L. E.,
Smith, A.G., and Smith, D. G. 1990. A geologic time scale
1989. Cambridge University Press.
Hilgen, E J. 1991. Astronomical calibration of the Gauss to
Matuyama sapropels in the Mediterranean and implications for the Geomagnetic Polarity Time Scale. Earth
Planet. Set Lett. 104:226-244.
Hilgen, F I , and Langereis, C. G. 1989. Periodicities of CaCO3
cycles in the Pliocene of Sicily: discrepancies with the
quasi-periods of the Earth's orbital cycles? Terra Nova
1:409-415.
Homes, M. 1853. Mittheilungen an Professor Bronn gerichtet.
N Jb. Mineral. Geol. Geogn. Petrefakt. 1853:806-810.
King, W. B. R., and Oakley, K. P. 1950. Report of the
Temporary Commission orr the Pliocene-Pleistocene
Boundary, appointed 16th August, 1948. In International
Geological Congress, Report of the Eighteenth Session,
Great Britain 1948, Part I. General Proceedings, ed. A. J.
Butler, pp. 213-214. London: Geological Society.
xvii
Lyell, C. 1833. Principles of Geology: being an attempt to

(Calabria-Italy). A potential Neogene/Quaternary boundexplain the former changes in the Earth's surface, by
ary stratotype. Giorn. Geol. 41:181-204.
reference to causes now in operation, vol. 3. London: John Shackleton, N. J., Berger, A., and Peltier, W. R. 1990. An
alternative astronomical calibration of the lower PleistoMurray.
cene time scale based on ODP Site 677. Trans. R. Soc.
Pasini, G., and Colalonga, M. 1994. Proposal for the erection
Edinburgh, Earth Sci. 81:251-261.
of the Santernian/Emilian boundary-stratotype (lower
Pleistocene) and new data on the Pliocene/Pleistocene Shackleton, N. J., Baldauf, J. G., Flores J.-A., Iwai, M.,
Moore, T. C , Raffi, I., and Vincent, E. 1995a. Biostratiboundary-stratotype. Soc. Paleontol. ItaL, Boll. 33:101
graphic summary for Leg 138. In Proceedings of the
120.
Ocean Drilling Program: scientific results, vol. 138, ed. N.
Rio, D., Raffi, I., and Villa, G., 1990. Plio-Pleistocene
G. Pisias et al., pp. 517-536. College Station, TX: Ocean
calcareous nannofossil distribution patterns in the western
Drilling Program.
Mediterranean. In Proceedings of the Ocean Drilling
Program: scientific results, vol. 107, ed. K. A. Kastens, J. Shackleton, N. J., Hall, M. A., and Pate, D. 1995b. Pliocene
stable isotope stratigraphy of Site 846. In Proceedings of
Mascle, et al., pp. 513-533. College Station, TX: Ocean
the Ocean Drilling Program: scientific results, vol. 138, ed.
Drilling Program.
N. G. Pisias et al., pp. 337-356. College Station, TX:
Rio, D., Sprovieri, R., and Di Stefano, E. 1994. The Gelasian
Ocean Drilling Program.
Stage: a proposal of a new chronostratigraphic unit of the
Pliocene Series. Riv. ItaL Paleontol. Strat. 100:103-124. Sibrava, V. 1992. Should the Pliocene-Pleistocene boundary be
Rio, D., Sprovieri, R., and Thunell, R. 1991. Pliocene-Lower
lowered? Sver. Geol. Undersok., Ser. Ca 81:327-332.
Pleistocene chronostratigraphy: a re-evaluation of Mediter- Tauxe, L., Opdyke, N. D., Pasini, G., and Elmi, C. 1983. Age of
ranean type sections. Geol. Soc. Am., Bull. 103:1049the Plio-Pleistocene boundary in the Vrica section,
1058.
southern Italy. Nature 304:125-129.
Ruggieri, G., Rio, D., and Sprovieri, R. 1984. Remarks on the Thunell, R. C , and Williams, D. R 1983. The stepwise
development of Pliocene-Pleistocene paleoclimatic and
chronostratigraphic classification of Lower Pleistocene.
paleoceangraphic conditions in the Mediterranean: oxygen
Soc. Geol. ItaL, Boll. 53:251-259.
isotopic studies of DSDP sites 125 and 132. Utrecht
Ruggieri, G., and Sprovieri, R. 1977. A revision of Italian
Micropal. Bull. 30:111-127.
Pleistocene stratigraphy. Geol. Romana 16:131-139.
Ruggieri, G., and Sprovieri, R. 1979. Selinuntiano, nuovo Turrin, B. D., Donnelly-Nolan, J. M., and Hearn, B. C. 1994.
40
Ar/39Ar ages from the rhyolite of Alder Creek, Califorsuperpiano per il Pleistocene inferiore. Soc. Geol. ItaL,
nia: age of the Cobb Mountain Normal-Polarity Subchron
Boll. 96:797-802.
revisited. Geology 22:251-254.
Salvador, A. (ed.) 1994. Internationalstratigraphic guide: a guide
to stratigraphic classification, terminology, and procedure,Van Couvering, J. A. 1995. Setting Pleistocene marine stages.
2nd ed. Boulder: Geological Society of America.
Geotimes 40:10-11.
Selli, R. 1971. Calabrian. In Stratotypes of Mediterranean Zagwijn, W. J. 1992. The beginning of the Ice Age in Europe and
its major subdivisions. Quatern. Sci. Rev. 11:583-591.
Neogene stages, ed. G. C. Carloni et al., pp. 55-64.
Zijderveld, J. D. A., Hilgen, R J., Langereis, C. G., Verhallen,
Giornale di Geologia 37, estratto, fasc. II.
P. J. J. M., and Zachariasse, W. J. 1991. Integrated
Selli, R., Accorsi, C. A., Bandini Mazzanti, M., Bertolani
magnetostratigraphy and biostratigraphy of the upper
Marchetti, D., Bigazzi, G., Bonadonna, K G . , Borsetti,
Pliocene-lower Pleistocene from the Monte Singa and
A. M., Cati, R, Colalongo, M. L., D'Onofrio, S.,
Crotone areas in Calabria, Italy. Earth Planet. Sci. Lett.
Landini, W., Menesini, E., Mezzetti, R., Pasini, G.,
107:697-714.
Savelli, C , and Tampieri, R. 1977. The Vrica Section
Foreword
Quaternary deposits are everywhere. In mountainous areas they

range from thin soils and scree through ephemeral accumulations, such as talus fans and stream alluvium, to sometimes quite
long-lived deposits like peat, valley fills, lake beds, and glacial
moraines. The acidic ground water and soils of mountains,
however, leave few fossils, apart from pollen and macrofloral
impressions. In lowland river valleys and coastal margins,
continental Quaternary deposits are thicker, more widespread,
and more fossiliferous. Even so, they are lithologically and
stratigraphically discontinuous on the larger scale. Only in the
marine environment, particularly on the floors of the deep ocean
basins, are essentially uninterrupted and highly fossiliferous
sequences of Quaternary strata found over wide regions.
Because the boundary between land and sea has changed very
little, relatively, during the geologically brief period of the
Quaternary, the early, classic studies of this crucially important
interval in the earth's history were restricted to the relatively
discontinuous continental deposits. The number of studies of
marine Quaternary deposits has increased greatly during the past
few decades, however, and they have provided us with a new and
much better documented basis from which to establish the
position of the lower boundary of the Quaternary.
As far back as 1760, the Italian geologist Arduino singled out
Quaternary strata as a distinct group of geological deposits
characterized by lack of induration. The concept of a Quaternary
time, as the most recent part of geological history, was suggested
in 1825 by Desnoyers, and in 1839 Charles Lyell introduced a
new term, "Pleistocene," for the epoch of youngest marine
faunas. By the middle of the nineteenth century it was evident
that much of the Quaternary was characterized by ancient glacial
deposits; at the same time, the distinct paleofaunal change from
Pliocene to Pleistocene was understood to be caused by the
increased influence of glacial environments. Thus, "Pleistocene"
and "Quaternary" came to be considered as coeval terms for the
glacially dominated period that ended the Cenozoic.
In 1846, Forbes recommended that use of "Pleistocene" be
restricted to exclude the most recent post-glacial strata, which he
denoted by the term "Holocene." Eventually the term "Pleistocene" came to be used by most stratigraphers in that way, as a
name for the deposits of the "Ice Age," with an upper limit
equivalent to the ending of glacial conditions in temperate

lowlands, presently dated to about 12,000 years ago.
The lower limit of the Quaternary is still subject to dispute,
and some researchers do not even acknowledge its existence,
because of the opinion that it is based on the same antiquated
preconceptions as the Tertiary, not to speak of the Primary and
Secondary. These workers consider the Cenozoic era to be
directly divided into constituent epochs, such as the Pleistocene,
or that the Pleistocene epoch is included in an extended Neogene
sub-era. A second group (including some of the same researchers) would also eliminate the Holocene as a separate, independent epoch. The work of IGCP-41, on the other hand, is based
on the prevailing opinion that the Quaternary should be
considered an independent period of the Cenozoic and that its
lower limit, the boundary between the Neogene and the
Quaternary (the N/Q boundary), is equivalent to the boundary
between the Pliocene and the Pleistocene.
In Russian, the terms "Post-Tertiary" and "Post-Pliocene," as
synonyms for the Quaternary, were widely used until the 1920s.
In 1919, A. P. Pavlov proposed to replace the name "Quaternary" with "Anthropogene," according to the view that the main
event in the organic history of the final system/period of the
Cenozoic was the appearance of humans. In actual fact, as E.
Aguirre points out in this volume (Chapter 10), the evolution of
Homo erectus in Africa at about 1.8 Ma is almost coincident with
the age of the N/Q boundary at Vrica, as recommended by
IGCP-41. The name "Anthropogene" accorded well with the
preceding "Paleogene" and "Neogene," which had already
replaced "Tertiary" in the Soviet schema, and it is now
commonly used throughout central and eastern Eurasia. In 1963
the names "Quaternary" and "Anthropogene" were officially
approved as equivalents by the Interdepartmental Stratigraphic
Committee of the USSR, in the MSC (modern stratigraphic
code) (Zhamoida et al., 1977). The term "Anthropogene" has an
obvious priority over the recently proposed "Pleistogene" of
Harland and others (Van Couvering, Preface, this volume).
Also, in consideration of priority, one should remember that
"Pleistocene" was long applied in the Soviet Union according to
a strict interpretation of Forbes's definition, which restricts the
term to the time of fully glacial conditions. It is now known that
in mainland North America, continental ice sheets appeared

during the climatic deterioration at the beginning of the
presently defined Pleistocene, and even before that in Greenland and Iceland, whereas in subpolar Eurasia (to which Forbes
was referring) the first fully glacial conditions are identified with
the Menapian (classic Mindel) paleoclimatic unit, now dated as
being of early Middle Pleistocene age. The pre-Menapian
deposits corresponding to the Lower Pleistocene of western
Europe are therefore equivalent to the Eopleistocene ("dawn of
Pleistocene") in the former USSR.
The definition of boundaries is a major issue in the debates on
Quaternary geology. The first step toward fixing the place of the
boundary between the Pliocene and the Pleistocene was the
organization of a special temporary commission at the XVIII
International Geological Congress in London in 1948. The
commission, affirming that the base of the Quaternary was
equivalent to the base of the Pleistocene, outlined three basic
criteria for the placement of the boundary: (1) the boundary
should be based on a faunal change in a section of marine
deposits, (2) the boundary should be located in the classic
territory of Quaternary marine deposits of Italy, the area in
which these principles can be best applied, and (3) the boundary
should be placed at a horizon demonstrating the first indication
of a deterioration of climate in the Italian deposits, as evidenced
by the first appearance of (unspecified) "northern guests" in the
Mediterranean Sea. Based on those criteria, the commission
recommended that the lower boundary of the Pleistocene should
be drawn at the base of the Calabrian Stage in Italy. That
recommendation was approved and adopted in the closing
session of the XVIII International Geological Congress.
The commission was aware of the fact that in proposing the
Calabrian Stage, Gignoux (1913) had referred to a number of
sections throughout Italy that evidenced a change from warm,
Pliocene conditions in the Mediterranean to a significantly colder
climate. That environmental change was documented by the first
appearance in Italian sections of animals that clearly originated
in the North Atlantic bioprovince. The first species identified as
a "northern guest," and still the most popular example, is the
mollusk Cyprina islandica, now generally referred to the genus
Arctica, which today lives in shallow subarctic waters above the
60-m isobath. Its value for close correlation is limited, and
stratigraphers therefore put forward a boreal benthonic foraminifer also found in the Calabrian, Hyalinea baltica, as a deep-water
indicator for the base of the Pleistocene (and also Quaternary).
Evidence of glacial climates can now be documented by
characteristic changes in many different assemblages of microorganisms and in the composition of fossil-spore spectra.
Originally it was assumed that the earth had been free from
lowland ice sheets until it suddenly entered a worldwide ice age
in the last part of the Cenozoic. With the aid of modern studies
of microfossil groups, however, we can recognize diachronous
and recurring appearances of cold-climate conditions in the
Mediterranean Sea. Application of this type of data to the
history of the late Cenozoic confirms that the deterioration in
climate throughout the Neogene was in fact a succession of
gradually intensifying global cycles, with repeated coolings

having taken place in the Mediterranean long before the first
appearance of the truly boreal elements that are designated as
indicative of the beginning of the Quaternary. In addition, Italian
geologists have recently shown that the appearances of the two
most often cited "cold guests," Arctica islandica and Hyalinea
baltica, were not synchronous and that H. baltica is certain to
have appeared later than A. islandica.
Among the typical Calabrian sections, Gignoux (1913) included Santa Maria di Catanzaro (Calabria), and that section
was later designated by Selli (1971) as the neostratotype. As far
back as 1961, Ruggieri proposed a new stratigraphic zonation of
the Italian Pliocene and early Pleistocene, with two zones: a
Lower Pliocene zone with two subzones (A and B), assigned to
the early Pliocene, and an upper zone also divided into two
subzones (C, with A. islandica, and D, with A. islandica plus
Hyalinea baltica). Ruggieri considered subzone D to be the base
of the Pleistocene because it coincided with the first appearance
of H. baltica, with subzone C representing the late Pliocene.
Later on, Ruggieri and Sprovieri (1976) recognized that subzone
C, with its famous "cold guest," was actually equivalent to the
lowermost Pleistocene, corresponding with the accepted general
concept of the Calabrian Stage. Subzone D was coeval with the
beginning of the Emilian Stage, and the overlying Sicilian Stage
was characterized by the two species named earlier plus
Globorotalia truncatulinoides excelsa.
As is now well known, the same authors (Ruggieri and
Sprovieri, 1977) later felt it necessary to abandon the name
"Calabrian" for the lowermost Pleistocene stage when it was
discovered that the Calabrian of Santa Maria di Catanzaro
appeared to be synchronous with the typical Sicilian, and the
Calabrian Stage defined in that section by Selli (1971) was, in
their view, therefore an invalid "synonym" of the type Sicilian
proposed by Doederlein in 1872 in a section at Ficarazzi
(Palermo). In place of the Calabrian, Ruggieri and Sprovieri
(1977) proposed a new stage, the Santernian, defined in the
Santerno Valley tributary to the Po in northern Italy.
At the IGCP/Plio-41 symposium in Italy in 1975, R. Selli and
G. Pasini proposed the section at Vrica, 4 km south of Crotone
(Calabria), as a candidate to serve as the boundary-stratotype of
the N/Q or Pliocene-Pleistocene boundary. Study of this section
by the working group of IGCP Project 41 and INQUA
Subcommission 1-d, "Pliocene-Pleistocene Boundary" (Selli et
al., 1977), showed that the section met all the requirements of
the modern stratigraphic codes for such a boundary (Hedberg,
1976; Zhamoida et al., 1977). Further investigations have only
strengthened the acceptance of the Vrica section as a location for
the Pliocene-Pleistocene boundary-stratotype, or neostratotype
according to the Soviet rules of stratigraphic nomenclature
(Zhamoida et al., 1977). Detailed descriptions of the section are
given in publications of Aguirre and Pasini (1985) and Nikiforova
(1985), as well as in various chapters in this volume.
The three stages Santernian, Emilian, and Sicilian were united
by Ruggieri and Sprovieri (1979) into one superstage, Selinuntian, although the short durations of those units prompted those
Foreword
authors subsequently to reconsider them as substages or chronozones of a Selinuntian Stage (Ruggieri, Rio, and Sprovieri,
1984). Most recently there has been general agreement (Preface,
this volume) to restore the Calabrian as the lowest stage in the
Pleistocene, according to the historical argument that the
boundary-stratotype for this stage was automatically established
at the same lithologic point as the definition of the PliocenePleistocene boundary-stratotype.
Thus, the section at Vrica proposed by Selli and Pasini in 1975,
and thoroughly studied afterward, meets all the international
requirements for adequate definition of the boundary-stratotype.
If we understand the restored Calabrian to embody the original
Calabrian concept of Gignoux, then the Vrica proposal also meets
every condition set by the 1948 London IGC for definition of the
Pliocene-Pleistocene boundary.
It should be noted in conclusion that the most urgent task now
facing the INQUA Commission on Stratigraphy is standardization of Quaternary stratigraphical and geochronological units
and their terminology, as concerns both the name of the system
itself and its subdivisions. As the stratigraphic range of the
Quaternary does not exceed one normal biozone, that of
Globorotalia truncatulinoides, recognition of series, stages, and
chronozones within the Quaternary may not be justified. In the
Stratigraphic Code of the USSR (Zhamoida et al., 1977),
subdivisions even finer than zones are described, some of which
(division, link) have been included in the standard stratigraphic
scale and may be appropriate as Quaternary subunits. These
problems are not proper subjects for this volume, of course, but
are considerations that arise now that the work on IGCP Project
41 is completed.
References
Doederlein, P. 1872. Note illustrative della carta geologica del

Modenense e del Reggiano. Memoria tersa. Modena:
Gaddi.
Forbes, E. 1846. On the connection between the distribution of
the existing fauna and flora of the British Isles and the
geological changes which have affected their area, especially
during the epochs of the Northern Drift. Memoir no. 1.
London: Geological Survey of Great Britain.
Gignoux, M. 1913. Le formations marines pliocenes et
quaternaires de lTtalie du Sud et de la Sicile. Univ. Lyon,
Ann., n.s. l(36):l-633.
Harland, W. B., Armstrong, R. L., Cox, A. V., Craig, L. E.,
Smith, A. G., and Smith, D. R. 1990. A geologic time scale
1989. Cambridge University Press.
Hedberg, H. D. (ed.) 1976. International stratigraphic guide.
New York: Wiley.
Nikiforova, K. V. 1985. The boundary between the Neogene and
Quaternary: where to draw it? (in Russian). Nature and
Resources 21:35-38.
Ruggieri, G., Rio, D., and Sprovieri, R. 1984. Remarks on the
Soc. Geol. ItaL, Boll. 53:251-259.
Ruggieri, G., and Sprovieri, R. 1976. La definizione dello
stratotipo del Piano Siciliano e le sue consequenze. Riv.
Min. Siciliana 26:1-7.
Ruggieri, G., and Sprovieri, R. 1979. Selinuntiano, nuovo
superpiano per il Pleistocene inferioro. Soc. Geol. ItaL,
Boll. 96:797-802.
Selli, R. 1971. Calabrian. In Stratotypes of Mediterranean
Neogene stages, ed. G. C. Carloni et al., pp. 55-64.
Giornale di Geologia 37: Estratto, fasc. II.
Marchetti, D., Bigazzi, G., Bonadonna, F. G., Borsetti,
A. M., Cati, F , Colalongo, M. L., D'Onofrio, S.,
Savelli, C , and Tampieri, R. 1977. The Vrica section
(Calabria-Italy). A potential Neogene/Quaternary boundary stratotype. Giorn. Geol. 41:181-204.
Zhamoida, A. I., Kovalevsky, O. P., Moisejeva, A. L., and
Yarkin, V I. 1977. Stratigraphic code of the USSR (in
Russian and English). Leningrad: VSEGEI.
Parti
Definition of the base of the Quaternary
International Geological Correlation Program, Project 41:

"Neogene/Quaternary Boundary"
KSENIA V. NIKIFOROVA and MIKHAIL N. ALEKSEEV
Formation of the project
Research on Project 41, "Neogene/Quaternary Boundary," was

initiated in 1974 under the auspices of the International
Geological Correlation Program at the second session of the
IGCP board held in Vienna. At the same session, a separate
proposal by the Geological Survey of India to define this
boundary in the Indo-Pakistani subcontinent was included in
IGCP Project 41 (Sastry, Chapter 23, this volume).
The first session of the working group for Project 41 was held
in Barcelona in September 1974. In attendance were E. Aguirre
(Spain), A. Azzaroli (Italy), M. Alekseev (USSR), W. A.
Berggren (USA), H. B. S. Cooke (Canada), L. K. Gabunia
(USSR), C. Ghenea (Romania), J. Michaux (France), K. V.
Nikiforova (USSR), A. Ronai (Hungary), M. V. A. Sastry
(India), R. Selli (Italy), K. O. Lange (UNESCO), and R.
Gonzales (Spain). At the conclusion of that meeting, it was
agreed that the membership of the Project 41 working group
would be as follows:
K. V. Nikiforova (chairman), Geological Institute,
USSR [now Russian] Academy of Sciences, Moscow, Russia
E. Aguirre (treasurer), Ciudad Universitaria, Madrid,
Spain
M. Alekseev (secretary), Geological Institute, USSR
[now Russian] Academy of Sciences, Moscow,
Russia
A. Azzaroli, Museum of Geology and Paleontology,
Florence, Italy
L. Benda, Niedersachsisches Landesamt fur Bodenforschung, Hannover, [West] Germany
W. A. Berggren, Woods Hole Oceanographic Institution, Woods Hole, MA, USA
H. B. S. Cooke, Dalhousie University, Halifax, N.S.,
Canada
L. K. Gabunya, Institute of Paleobiology, Academy of
Sciences of the Georgian SSR [now Georgian
Academy of Sciences], Tbilisi, Georgia
E. D. Gill, University of New South Wales, Canterbury,
Australia
C. Ghenea, Geological Institute, Bucharest, Romania

H. M. S. Hartono, Geological Research and Development Centre, Bandung, Indonesia
R. W. Hey, Cambridge University, Cambridge, UK
J. C. Ingle, Stanford University, Stanford, CA, USA
M. Itihara, Osaka City University, Osaka, Japan
H.-D. Kahlke, Institute of Quaternary Paleontology,
Weimar, [East] Germany
J. Michaux, University of Montpellier, Montpellier,
France
H. Nakagawa, Tohoku University, Sendai, Japan
R. Pascual, National University, La Plata, Argentina
K. Prasad, Geological Survey of India, Bangalore,
India
A. Psilovikos, Aristotle University of Thessaloniki,
Greece
A. A. Ronai, Hungarian Geological Institute, Budapest, Hungary
M. V. A. Sastry, Geological Survey of India, Calcutta,
India
R. Selli, Institute of Geology, Bologna, Italy
N. J. Shackleton, Cambridge University, Cambridge,
UK
P. Vella, Victoria University, Wellington, New Zealand
The members of the working group agreed that the main task
of Project 41 would be detailed studies of the stratigraphy of
Pliocene and Lower Quaternary deposits developed on continents, islands, and ocean floors, aimed at definition and
worldwide recognition of the Neogene-Quaternary boundary.
Establishment of that boundary would be of great importance for
the universal stratigraphic scale and for strengthening the
standardized basis for mineral prospecting, geological mapping,
map legends, neotectonics, geomorphology, and so forth.
In accordance with the adopted program, the group agreed
that various methods should be used for solving this problem:
biostratigraphic, climatostratigraphic, and physical methods such
as paleomagnetic, paleotemperature, and radiometric age determinations. Studies of marine and continental faunas and floras
and of climatic changes at various intervals of the Pliocene and
Ksenia V. Nikiforova and Mikhail N. Alekseev
Quaternary would provide the research data for paleogeographic

reconstructions of certain time sections. Of particular importance would be the use of physical methods that could provide
more precise dating, applicable to the problems of synchroneity
and metasynchroneity of geological events on a regional or
global scale.
Work on Project 41 would elucidate the geological history and
structure of continental shelves that hold oil and gas prospects.
Continental and near-shore deposits of Quaternary age include
placers of precious metals and rare ores. Superficial deposits of
tropical continents include large-scale oxide ore bodies and
concentrations of heavy minerals.
The project was conceived as an interdisciplinary study.
Specialists from various fields - geology, geomorphology, paleoclimatology, paleontology, micropaleontology, archeology, anthropology, chemistry, and physics - took part. Close cooperation with international and national organizations such as IUGS
(International Union of Geological Sciences) and INQUA
(International Quaternary Association) was deemed essential. In
this respect the project conforms fully to the aims and goals of
UNESCO (United Nations Educational, Scientific, and Cultural
Organization).
Background
The position of the N/Q (Neogene-Quaternary) boundary had
already been discussed at several meetings of the IUGS and
INQUA, as well as in international symposia and colloquia
convened especially to study this problem. The problem of
defining the boundary was first debated at the XVIII International Geological Congress in London in 1948, which recommended that the deposits of the Calabrian Stage of Italy (in
which the northern immigrants Arctica islandica and Hyalinea
baltica appear for the first time) be defined as the basal member
of the Lower Pleistocene in the Quaternary system. The
Villafranchian was considered to be the terrestrial equivalent of
the Calabrian. Later it was found that the term "Villafranchian"
applied to a much wider time interval and that its lower twothirds belongs to the Pliocene, if the Calabrian represents the
oldest part of the Pleistocene.
At the international colloquium organized by the INQUA
Subcommission 1-d, "Pliocene/Pleistocene Boundary," held in
the USSR in 1972, the following recommendation was adopted:
"the lowermost level in the section of Le Castella in Calabria
with remains of Hyalinea baltica (Schrotter) [is] selected as the
initial definition of the base of the Pleistocene in a marine section
of the Mediterranean basin." The colloquium also noted the
need for more exact correlation between marine and continental
deposits and the establishment of stratigraphic analogues of the
Calabrian in other territories, including sections of sea-floor
sediments. The recommendation of the 1972 colloquium in the
USSR served as the basis for the subsequent recommendations
of the XXIV International Geological Congress in Montreal in
1972.
It must be noted here that in subsequent studies by Italian
stratigraphers it has been shown that H. baltica unquestionably

appears in the southern Italian sequences much later than
Arctica islandica, at a level that has been identified with the base
of the Emilian substage. The first appearance of A. islandica in
Italian strata is still considered to be an accurate indication of the
earliest part of the Pleistocene.
The next international colloquium on the Pliocene-Pleistocene boundary organized by INQUA Subcommission 1-d was
held in Christchurch, New Zealand, in December 1973 during
the IX INQUA Congress. That congress provided an important
opportunity for firsthand examination of the Upper Pliocene and
Lower Pleistocene deposits, as well as the position established
for the N/Q boundary in New Zealand and Australia. Some of
the papers read at that symposium were published in Bulletin 13
of the Royal Society of New Zealand, Quaternary Studies, in
1975.
Organization and objectives of Project 41
Plan of work
The basic goals of the international working group on IGCP
Project 41, "Neogene/Quaternary Boundary," were as follows:
1. analysis of the available data and solution of approachable problems through investigations carried out by the
national working groups, or in cooperation with scientists from other countries involved in these activities
2. field studies regarding new objectives and critical
sequences proposed by national working groups
3. field investigations, consultations, and exchange of
knowledge on the geology of Pliocene and Lower
Pleistocene deposits in the developing countries in
order to stimulate scientific work
4. interdisciplinary studies in regions of difficult access,
with priority given to studies of abyssal oceanic
sediments, in consultation with the INQUA Subcommission on Stratigraphy of Deep-Sea Sediments
The field-team concept
The field-team approach was adopted as the best method to
study important local stratigraphy and correlation of Pliocene
and Lower Quaternary deposits. The following are some of the
particular sections studied by the national working groups:
Italy: stratotypical Calabrian and Villafranchian sections; review of the present state of investigations of
the problem "N/Q Boundary" in the type area.
Japan: key sections of the Pliocene and Lower Quaternary deposits of the Akita, Kinki, and Kanto
regions.
USSR: nearly complete sections of Pliocene and Lower
Pleistocene deposits in the European part of the
former USSR and in middle Asia.
United States: key sections of the Pliocene and Lower
IGCP. 41: Neogene/Quaternary Boundary
Pleistocene marine and continental deposits of

California and the interior areas.
India: field conference, study of Pliocene and Lower
Pleistocene deposits of the Siwaliks.
Correlation
The program of stratigraphic studies was planned to correlate in
time and space the Pliocene and Lower Quaternary deposits in
various paleogeographical provinces of different continents and
islands, and on the ocean floor, as follows:
Deposits in continents and islands. Elaboration of stratigraphy
and correlation of deposits located in different paleogeographic
provinces, as well as correlations between marine and continental deposits, would pose a problem. The stratigraphic sections in
Italy, England, and The Netherlands had been the best studied in
the regions of western, central, and northern Europe, although
the sections in western and eastern Germany would also be of
importance.
In eastern Europe, successions in the European part of the
USSR, southern Moldavia, Ukraine, Azerbaijan, and Georgia
had been well studied. The sections in Romania, Yugoslavia, and
Hungary would be very important. The main problems in
Europe appeared to involve correlation of marine and continental deposits, using a combination of methods such as radiometric
and paleomagnetic analyses, and coordination with the sections
of subaqueous and subaerial series.
In Asia, determination of the N/Q boundary would involve
studies in the sections of Turkmenia, middle Asia, Turkey,
Transuralian Russia, western and eastern Siberia, the northeastern USSR, the Far East, the Indian subcontinent, Burma,
Indonesia, China, and the Japanese islands. The principal
problems in those regions would also lie in the correlation of
marine and continental deposits, as well as volcanogenic
formations of boreal and tropical zones, and in the definitions of
the nature of sedimentation, erosional phases, and sea-level
fluctuations.
Africa would require study of the sections of continental fossilbearing deposits of Pliocene and Lower Quaternary age in the
African rift zone. The main task here would be to detail the
chronostratigraphic correlations of the Rift Valley sections with
the climatically influenced sections of the southern Mediterranean coast (Algeria, Tunisia, Morocco) and those of the
northern coast of the Mediterranean.
In the case of North America, the most important objectives
would be comparisons of the marine sediments of California and
the Gulf of Mexico with the continental sediments. It would be
necessary to correlate the older glacial deposits and to work out
the detailed stratigraphic schemes based on the use of a number
of methods.
In South America, objectives would include thorough studies
of the stratigraphy of Pliocene and Lower Pleistocene deposits of
the Pampas, radiometric ages for volcanogenic deposits, elaboration of the stratigraphy of Plio-Pleistocene marine sediments,
and, above all, correlation of climatic fluctuations between the

southern and northern tropical zones on the South American
continent.
Australia and New Zealand would offer an opportunity for
thorough study of the Pliocene and Lower Quaternary sections
of marine, continental, and volcanogenic deposits, thus providing the data to work out a climatostratigraphic scheme for the
deposits of those ages. A more precise definition of the positions
of datum planes with regard to global chronostratigraphy would
be the final objective.
Deposits of ocean basins. The objectives of studies in the ocean
basins would be analyses of the precise stratification of abyssal
deposits and establishment of the boundary of the Globorotalia
tosaensis and Globorotalia truncatulinoides zones, as well as the
stratigraphic positions of other oceanic fossils in the Pacific,
Indian, and Atlantic oceans. A further major objective would be
the zonation and correlation of sections of Upper Pliocene and
Lower Pleistocene deposits at high, middle, and low latitudes.
Establishment of stratigraphic analogues of the G. tosaensis and
G. truncatulinoides zones, as well as nannoplankton and radiolarian zones typical of middle latitudes, would be required for
deposits laid down in high latitudes. Contributing to that effort
would be studies of the paleomagnetism, radiometric age, and
paleotemperatures of such deposits.
History of the working group activities
During 1975, national and regional working groups were

organized and began to implement research activities in the
German Democratic Republic, the Federal Republic of Germany, India, Spain, Japan, Italy, the USA, and the USSR, and
international groups began work on the magnetostratigraphy and
stratigraphy of deep-sea Pliocene and Lower Quaternary deposits. They were joined by working groups in Australia and New
Zealand in 1976.
The Italian national working group carried out extensive
studies on the establishment of a stratotypical section in marine
deposits in southern Italy, as well as research on Pliocene and
Lower Quaternary continental deposits.
The symposium of the working group in Italy, 1975
The second symposium was held in Bologna, Italy, in October
1975, and the presented papers were published in the Giornale di
Geologia, vol. 41 (1977). The meeting was attended by 47
representatives from 12 countries (United Kingdom, France,
Federal Republic of Germany, The Netherlands, Hungary,
Canada, Spain, Italy, Japan, Yugoslavia, the USA, and the
USSR). Twenty-five contributions from various countries were
delivered at the Bologna session, most of which were preliminary
to the chapters in this volume.
The group spent six days making excursions to the sections of
Pliocene and Quaternary deposits in central Italy (basins of the
Santerno, Arno, and Tiber rivers) and in Calabria. The
participants agreed that there were some difficulties in the

published interpretation of the Santa Maria di Catanzaro
sections, as well as in the Le Castella outcrop, with regard to
locating a boundary-stratotype. It was noted that the Vrica
outcrops of the Crotone area, not far north of Le Castella in
Calabria, form a section about 300 m thick in which there are no
apparent interruptions in sedimentation. It was already known
that sedimentation at Vrica had taken place under conditions of
rather deep water, as compared with those farther to the south,
and that changes of both foraminiferal and ostracod complexes
could be observed in the section (Pasini et al., 1975). The
participants decided to study the Vrica section in detail and to
submit the results at the session of the working group in 1977 at
the X INQUA Congress in Birmingham. The working group
members also suggested that by the end of 1977 the Vrica section
would have been sufficiently well studied to enable them to
recommend it as a potential stratotype of the N/Q boundary.
Paleontological and radiometric studies of the Vrica section were
to be carried out by the Italian national working group, and the
paleomagnetic determinations by Japanese scientists.
To that end, the conference adopted a resolution to the effect
that the working group of IGCP-41, "Neogene/Quaternary
Boundary," had drawn the attention of the appropriate Italian
authorities to the great importance of the geological sections in
Calabria for the establishment of a time scale for use all over the
world. Thick deposits at Vrica, near Crotone, were currently
being investigated and promised to yield results of major
importance to this purpose.
More detailed evaluations of the small and large mammalian
markers, to establish a reliable subdivision of the continental late
Pliocene and early Pleistocene, would be necessary. It was
recommended that an international meeting of mammalogists be
held in the western Mediterranean region to achieve a common
understanding of the detailed mammalian biostratigraphic subdivisions that could be relevant to the aim of Neogene/Quaternary
correlation.
The IGCP-41 working group symposium in Japan,
1976
The third IGCP-41 symposium took place as a joint meeting of the
IGCP-41 working group and the IUGS working group on the
Pliocene-Pleistocene boundary in May 1976, during the First
International Congress on Pacific Neogene Stratigraphy, in
Tokyo.
The participants at the symposium examined the key sections
of the Pliocene and Lower Quaternary on the Boso Peninsula
and in the Kinki and Kakegawa regions. The PlioceneQuaternary sequences of Japan are key sections for eastern Asia
and the North Pacific region and can be correlated with the
stratotypical section of the same age in Italy.
In a number of districts on Honshu Island, three groups of
deposits have been defined: the Osaka, Kobiwako, and Tokai
groups. In the Osaka group, a little below ash bed Ma-0, an
uncomformity indicating a transgression was discovered, with a
radiometric age of 1.6-1.7 Ma. This transgression could perhaps

correspond to the Calabrian transgression of Italy and, consequently, to events near the N/Q boundary.
The fresh-water Kobiwako group is subdivided into two parts.
Early regression in the basin is estimated to have occurred
between 1.7 and 2.0 Ma, followed by the formation of Biwa
Lake. Thus, the origin of this lake may be coincidental with the
Neogene-Quaternary boundary as well. The Tokai group is also
subdivided into two parts. The earlier part is close in age to the
early Kobiwako group, but the later part is much younger than
Biwa Lake.
Technical reports at that session were devoted mainly to the
micropaleontology of the N/Q boundary level in deep-sea
records.
Round-table meeting in Spain, 1976
In order to help correlate the N/Q boundary in continental
environments by means of land-mammal biostratigraphy, the
divisions and nomenclature for a mammal biostratigraphic scale
were adopted in a round-table meeting in Madrid, 1976, by
representatives of the various IUGS and IGCP projects for the
Neogene. Those standards were published in Trabajos Sobre
Neogene-Cuaternario, No. 7, Madrid, Instituto Lucas Mallada
CSIC.
Birmingham, 1977
A joint meeting of the INQUA subcommission on the PliocenePleistocene boundary and the IGCP working group on Project 41
was held in Birmingham, England, during the X INQUA
Congress. Papers presented at this meeting were published with
the proceedings of the congress.
The participants of the meeting heard accounts of the ongoing
studies of the N/Q boundary conducted during the intercongress
period 1974-1977, prepared by the chairmen of the subcommission and the working group, as well as reports on the studies that
were carried out in various countries. The participants were
informed that a considerable amount of progress had been made
during this period, including collection of abundant material on
the geology, biostratigraphy, climatostratigraphy, and, to a lesser
extent, the radiometric ages of Pliocene and Lower Quaternary
deposits of the various continents, islands, and oceans.
General agreement among the investigators was obtained on
the following matters:
1. The Neogene-Quaternary boundary should be drawn
in accordance with general stratigraphic principles (i.e.,
based on changes in the open-ocean marine faunas).
2. Italy should be the stratotypical area for determination
of the N/Q boundary.
3. Detailed correlations between marine and continental
deposits will be necessary in order to establish the
correlation of the N/Q boundary. Data on paleomagnetic stratigraphy and the radiometric ages of deposits
must also be taken into consideration. The boundary
defined in marine sections in Italy must be taken as a

basis, and then correlated with the continental sequences. These analyses can provide reliable grounds
for global correlation.
4. The participants at the meeting emphasized the necessity for further micropaleontological investigations
aimed at finding more reliable criteria for definition of
the Neogene-Quaternary boundary.
In addition, the participants at the meeting concentrated on
the considerable advances in our knowledge in the fields of
biostratigraphy and magnetostratigraphy and agreed that the use
of radiometric dating would be one of the most important
methods for wide-range correlation, although such data were not
yet available for many parts of the world. The participants
expressed a desire for this method to be used more extensively in
the work of Project 41. The participants at the meeting
supported the suggestion to take the Vrica section as the N/Q
boundary-stratotype and recommended that a detailed, complex
study of this section be proposed.
The meeting approved the membership of an editorial board
for the final report, to consist of E. Aguirre, M. Alekseev, W. A.
Berggren, F. P. Bonadonna, H. B. S. Cooke, R. W. Hey, H.
Nakagawa, K. V. Nikiforova, G. Pasini, and R. Selli. At that
time, the board recommended further study of the following
stratigraphic and chronological levels:
1. The boundary between the Gilbert and Gauss paleomagnetic epochs (3.3-3.5 Ma) that tentatively corresponds to the base of the Astian and Piacenzian of the
Italian sections and the Akchagylian in the USSR
2. The beginning of the Matuyama paleomagnetic epoch
(approximately 2.5 Ma), corresponding to the boundary of the Lower and Middle Villafranchian and the
base of deposits containing the fauna of the Khaprovian
complex of mammals in the USSR
3. The beginning of the Olduvai event of normal magnetization, which, at that time, was thought to be dated to
approximately 1.79 Ma and to be related to a shift to
faunas with arctic elements (mollusks, ostracodes, and
foraminifers), as well as to the base of the Upper
Villafranchian in Italy, the Apsheronian in the USSR,
and deposits containing the mammalian fauna of the
Odessa complex in the USSR
4. The boundary between the Brunhes and Matuyama
paleomagnetic epochs (0.7-0.8 Ma), which is tentatively correlated with the base of the Cromerian in The
Netherlands and, in the USSR, with the base of the
Bakunian deposits, as well as with the base of the series
of continental deposits enclosing the mammalian fauna
of the Tiraspolian complex
IGCP review, 1977
During the meeting of the governing board of the IGCP in Paris
in March 1977 it was decided that it was time for an appraisal of
the program as a whole. The board requested all project leaders

to submit reports to the Secretariat describing the progress and
achievements of their projects. Statements for 1973-1977 were
presented and published in a special issue of Geological
Correlation in 1978.
At that time, four different concepts regarding the placement

of the lower boundary of the Quaternary were noted:
1. At the location of the cooling event below the
Piacenzian in Italian sections, which was related to the
Globorotalia miocenica zone of Bolli and Premoli Silva,
as well as to the Upper Ruscinian and Lower Villafranchian mammalian faunas, and which is close to the
boundary between the Gilbert and Gauss paleomagnetic epochs.
2. At the location of the significant climate changes that
marked the base of the Middle Villafranchian, coinciding approximately with the boundary between the
Gauss and Matuyama paleomagnetic epochs.
3. At the base of the Calabrian Stage, which contains
arctic elements (beds with Arctica islandica). (At that
time, the horizon was believed to be correlated to the
base of the Globorotalia truncatulinoides zone and the
beginning of the Olduvai paleomagnetic event, as well
as to Upper Villafranchian faunas of Italy.)
4. At the base of the Cromerian in The Netherlands (base
of the "glacial Pleistocene" in Europe), which is close
to the boundary between the Matuyama and Brunhes
paleomagnetic epochs.
It was pointed out that only two of those suggested boundaries
(1 and 3) could be seriously regarded as definitions of the
Neogene-Quaternary boundary, because only they were in
accord with the general principle of placing such a boundary at a
widely recognized horizon in marine sequences, in this case
involving planktonic organisms that could allow widespread
correlation. The other two suggested boundaries were in
nonmarine or local sequences defined by a climatic change.
It was firmly established by the 1948 London IGC resolution
that the stratotype sections for the Neogene-Quaternary boundary should be in Italy and should be situated within Mediterranean marine deposits of the Calabrian in relation to the
appearance of cold-water Atlantic immigrants. The IGCP-41
working group noted that the Vrica section met all of the
necessary requirements to be chosen as a stratotype; it is
characterized by continuously deposited strata laid down under
bathyal conditions and is rich in fossil organisms, including
planktonic groups.
By the time of the Paris IGCP board meeting in 1977, research
had been expanded considerably within the framework of the
project. For example, additional studies in Java had been
initiated with the help of Japanese working group members.
Also, the Indian working group had organized studies on the
stratigraphy of the Pliocene and Lower Quaternary deposits in
the Nicobar and Andaman islands and had extended its interests
to remote northern and northeastern regions of the country.
research of IGCP Project 41. Considerable attention was given

to the studies produced by Project 41 on stratigraphic correlation
Findings in southern Tadjikistan in 1977 were the subjects of an between the sedimentary basins of the ESCAP region. The
IGCP-41 symposium. The group studied the following reference developing countries of the ESCAP region were invited to take
sections: Akjar, Karamaidan, Karatau, Lakhuti, Khonako, and part in implementation of Project 41 research, and there were
Kuruksai. The papers presented during the meeting were devoted discussions with the professional staff of the Geological Survey
to the stratigraphic subdivision of the Upper Pliocene-Quater- of Indonesia about possible specific research projects. Various
nary deposits in Tadjikistan, their paleomagnetic characteristics, regional meetings were held to discuss the N/Q boundary
their fauna and flora, a comprehensive analysis of the loess-soil problem in the USSR (in Bashkiria, Transcaucasia, the lower
formations, and studies of the newest tectonic movements and the Volga River basin, etc.), in addition tofieldconferences, drilling,
structural characteristics of the Upper Cenozoic molasse. The and field work in the USSR, Hungary, India, Japan, Indonesia,
problems of the geology of Paleolithic deposits and the ecology of Germany, Spain, Greece, and the USA. New groups of
ancient humans in Tadjikistan were examined. Additionally, organisms were used for definitions of stratigraphic boundaries
papers were presented on the stratigraphy of Pliocene and in this interval. The number of countries participating in the
Quaternary deposits, the flora and fauna of Kirgizia, Uzbekistan, work increased, and the developing countries began to particiTurkmenia, and Kazakhstan, and the problems of stratigraphy of pate more actively in the work of Project 41. Also in 1978, the
the Pamir glacial deposits and their correlation with glacial INQUA subcommission on European stratigraphy met to discuss
sequences in middle Asia. The presented papers appeared in the the problem of the N/Q boundary. The almost complete absence
Proceedings of the second symposium on the Neogene/Quaternary of correlations between the stratigraphy of continental deposits
and the oceanic zonation was widely discussed. That may have
boundary (Nikiforova and Dodonov, 1980).
In earlier years, the stratigraphic subdivisions of the Upper influenced the majority of the western European scientists to
Pliocene and Quaternary deposits had been based mainly on conclude that the N/Q boundary should be positioned at the base
geological-geomorphological criteria, together with paleontolog- of the Middle Villafranchian (about 2.5 Ma), because that level
ical and archeological data. Practically no analyses on stratig- was marked by an intensive cooling of the climate. However,
raphy of overlying deposits and their correlation with alluvial that cooling is not strongly reflected in the deep-sea sediments,
complexes had been carried out. More recently, considerable and significant changes in the assemblages of marine organisms
success has been achieved in studies of the Upper Cenozoic have not been recorded at that level.
deposits in Tadjikistan by applying interdisciplinary techniques to
micropaleontological (i.e., palynological), paleopedological, paActivities in 1979
leomagnetic, and thermoluminescent methods. Thus, the Tadjik
depression has become an important reference region for solving A field conference held in the northwestern part of India in 1979
various problems of Quaternary stratigraphy in Central Asia.
was devoted to the problems of stratigraphy, paleontology,
Participants at the meeting agreed that interdisciplinary tectonics, and volcanism of the Upper Pliocene-Lower Pleistostudies of the Upper Cenozoic deposits of Tadjikistan should be cene, as well as to the magnetostratigraphy and geochronology
continued and widened to establish the climatostratigraphic of the Pliocene-Quaternary transitional interval. The presentaessence of the previously mentioned alluvial, alluvial-proluvial, tions were given at sessions on (1) stratigraphy, (2) paleontology
and paleogeography, (3) tectonics, volcanism, magnetostratigand glacial deposits.
raphy, and geochronology, and (4) paleoanthropology and
archeology. All reports were later published by the Geological
Activities in 1978
Survey of India (Sastry et al., 1981). The scientific sessions were
During 1978, a working group in Greece joined Project 41 under followed by excursions to sections of the Siwalik series in the
the coordination of A. Psilovikos. A group in Bangladesh also Himachal Pradesh and Jammu provinces and the Karewa series
joined the project, with a new branch of the Geological Survey in Kashmir.
of Bangladesh created to study Quaternary geology. Additional
The editorial board met in Madrid in February 1979 to discuss
studies in Java were continued by Indonesian and Japanese the preparation of the final report on Project 41: clarification of
members of the working group. The work of Project 41 will be the program, its mode of presentation, and the date of its
important in developing countries of Asia and the Pacific region, completion. It was decided to include J. A. Van Couvering
complementing IGCP Project 32, "Stratigraphic Correlations (USA) and M. V. A. Sastry (India) on the editorial board. Dr.
between Sedimentary Basins of the ESCAP Region." In 1977, Van Couvering was named to be editor of the final report.
Dr. H. F. Doutch, an expert from the ESCAP Secretariat,
At the International Pacific Science Congress in Khabarovsk,
participated in the Tadjikistan symposium.
USSR, in 1979, the secretary of the central working group, M. N.
In November 1978, the 15th session of the United Nations Alekseev, reported on the activities of Project 41. Reports
Committee for Coordination of Joint Prospecting for Mineral devoted to the N/Q boundary problem on the eastern coast of
Resources in Offshore Areas of ESCAP met in Bangkok and Asia and in Siberia were presented and discussed at a section on
considered the possibility of cooperation with the stratigraphic Upper Cenozoic stratigraphy. An excursion to the Kolyma
Symposium in Tadjikistan (USSR), 1977
IGCR 41: Neogene/Quaternary Boundary
lowland was organized to study the stratigraphy of the Quaternary. Also in 1979, M. N. Alekseev convened a meeting at
Sangiran, Java, to review cooperative work between the Geological Survey of Indonesia and Japanese scientists on a detailed
study and geological survey of the critical area in central Java,
emphasizing correlations between continental and marine PlioPleistocene deposits. That led to a more precise understanding of
the positions of stratigraphic boundaries in the portion of the
section intermediate between the Pliocene and the Quaternary.
Further physical research into the problem of the N/Q
boundary was carried out by the national working groups of
Hungary, India, Indonesia, the German Democratic Republic,
Spain, Italy, Greece, the USA, the USSR, and some other
countries. The Project 41 leaders maintained close contact with
the investigators of Project 32, "Stratigraphic Correlation
between Sedimentary Basins of the ESCAP Region," Project 25,
"Stratigraphic Correlation of the Tethys-Paratethys Neogene,"
and Project 114, "Biostratigraphic Datum-Planes of the Pacific
Neogene." Considerable micropaleontological research was conducted by the national working group of Italy to document the
Vrica section (Italy) as a key stratotype section for the N/Q
boundary. In addition, magnetostratigraphic studies by H.
Nakagawa revealed a normal-polarity interval in the predominantly reverse-polarized section that potentially can be considered to represent the Olduvai episode of the Matuyama
geomagnetic epoch near the N/Q boundary. Samples were
collected for further determination of the radiometric ages of the
ash interlayers. A meeting of the Spanish national working group
was held to discuss new research on selecting the local
parastratotype section. The coastal area from the Portuguese
frontier around to the Ebro delta was investigated, and studies of
new sites with mammalian fauna and detailed geomorphological
study in La Mancha indicated that the Campo de Calatrava
volcanic formations in western La Mancha could be used for
radiometric age determinations to help date the Spanish N/Q
sequence. It was recommended that magnetostratigraphic investigations be carried out in the southern Meseta and in the Baza
basin and in the territories of the Cadiz coast, Murcia-Alicante,
and the Ebro River.
During 1979, Project 41 engaged 19 regional working groups
to conduct broad investigations in the territory of the USSR. The
reports of those regional groups were discussed at a meeting of
the Soviet working group in March 1979. The Hungarian
working group drilled research boreholes in the Great Hungarian Plain, with the goal of interdisciplinary geological analysis.
They encountered Quaternary and Upper Pliocene sequences
that in some places were continuous and complete, without
stratigraphic lapses. The borehole samples were investigated
from many points of view, including magnetostratigraphy, and
the preliminary results were published.
Activities in 1980
A meeting of the Project 41 working group was held during the
26th session of the International Geological Congress in Paris,
jointly with INQUA Subcommission 1-d on the PliocenePleistocene boundary. Twenty-five delegates from various countries attended the meetings. The recent activities of the working
group were presented by the leader of the project, K. V.
Nikiforova. That was followed by a report on the previous two
years of work by the INQUA subcommission and the IUGS
working group on the Pliocene-Pleistocene boundary, and a
report on the results of a mail consultation conducted by its
chairman, E. Aguirre. The current state of research on the
sections at Le Castella, Santa Maria di Catanzaro, and Vrica was
discussed by G. C. Pasini, and an introduction to the report on
Plio-Pleistocene datum levels in the deep sea was given by J. A.
Van Couvering. After discussions on the reports, the meeting
unanimously adopted the following principles:
1. The lower boundary of the Quaternary would have to
be established in accordance with the general principles
of stratigraphy (i.e., the decision must conform to the
guidelines recommended by the International Commission on Stratigraphic Classification) (Hedberg, 1976).
2. The recommendation of the IGC (London, 1948)
should be slightly modified to state that the boundary
must be designated as a stratigraphic plane (boundarystratotype) in a continuous sequence of open-marine
deposits.
3. The 1948 recommendation is understood to mean that
the base of the Quaternary (viz., the Pliocene-Pleistocene boundary) should be defined by the base of the
Calabrian Stage in southern Italy. It was therefore
recommended that the Calabrian Stage should be
redefined (taking into account modern research, which
indicated that the Catanzaro section is not satisfactory
to express Gignoux's concept of the Calabrian Stage) to
make its base unambiguous.
4. Multiple criteria should be used in selecting a stratigraphic plane for the base of the Calabrian and thus the
N/Q boundary-stratotype; that is, all the available
evidence that could help wide-range correlations should
be taken into account. The positions corresponding to
the N/Q (Pliocene-Pleistocene) boundary in other
areas would have to be determined by working out local
stratigraphic scales and correlating them to the stratotype section. For compilation of the local scales, a
synthesis of biostratigraphic, climatologic, magnetostratigraphic, and radiometric techniques should be
used. Special attention would have to be paid to the
current difficulties of identifying the boundary, by
means of correlation, in different latitudes and in
continental sequences.
As a first criterion, it was proposed that the N/Q boundary
could be placed in the Vrica section at the FAD (first-appearance
datum) of the "cold guest" ostracode Cytheropteron testudo,
whatever its paleoclimatic significance might be. At the time of
that meeting it was thought that the first C. testudo could be
found in the Vrica section 10 m above sapropel e, but in later
10
studies (Pasini and Colalongo, Chapter 2, this volume) it was

established that this FAD actually occurs somewhat earlier.
An alternative possibility for selecting a level in the Vrica
section as the N/Q boundary-stratotype would be to place it
within the stratigraphic interval between the level of sapropel e
and the level of volcanic ash ra, as close as possible to a
paleomagnetic reversal. It was recommended that the final
decision on the exact placement of the N/Q boundary-stratotype
be deferred until the paleomagnetic record of the Vrica section
could be further investigated. F. P. Bonadonna agreed to
coordinate that work, which was to include a new detailed
sampling and study of the section, with the cooperation of
scientists from several institutions and countries.
Studies within the framework of the national working groups
were continued throughout 1980. A field conference devoted to
the problems of biostratigraphy and magnetostratigraphy was
held in Spain, with scientists from Spain, Italy, USA, USSR, and
France taking part. The most suitable paleontologically characterized Plio-Pleistocene sections were selected, with the aim of
paleomagnetic investigations. In 1980 the working group in
Japan summarized its results from thorough studies of sections of
the Boso Peninsula (Honshu). The first appearance of Globorotalia truncatulinoides was established in the middle part of the
Kiwada Formation. Japanese and Indonesian scientists continued studies of Pliocene-Quaternary sections in the central part
of Java.
Scientists in the People's Republic of China managed to
establish that the loessic series of Malan and Lishi belong to the
Brunhes zone of positive paleomagnetic polarity. On the basis of
paleomagnetic measurements, the Wucheng loess was attributed
to the Matuyama reversed-polarity zone, with the upper part of
these loessic series falling in the Upper Matuyama.
The USA working group concentrated on studies of sections in
Arizona and California, in preparation for a field conference at
Tucson, Arizona, in March-April 1981.
In the USSR, important studies were carried out in preparation for the scientific excursions of the XI INQUA Congress in
1982, primarily in middle Asia, Moldavia, Transbaikalia, and
Yakutia. In the eastern provinces of the USSR, local biostratigraphic subdivisions and the alluvial series on the high terraces of
the Lena River were correlated to the general magnetostratigraphic scale, and many studies dealt with climatostratigraphic
interpretations in both the European and the Asian parts of the
USSR.
Activities in 1981
The meeting in Tucson was followed by afieldconference on the
key sections of Pliocene and Pleistocene deposits in Arizona and
California, coordinated by the USA working group. Members
from China, Hungary, India, Italy, Spain, the USA, and the
USSR took part. The meeting was held jointly with IGCP
Project 128, "Late Cenozoic Magnetostratigraphy," and the
INQUA subcommission on the Pliocene-Pleistocene boundary.
At the Tucson meeting, thefindingsfrom ongoing studies were

discussed, and further work on the Vrica section was proposed,
to be carried out during 1981 by scientists from Italy and the
USA. Following the Tucson meeting, magnetostratigraphic
investigations, further collection of radiometric samples, and
detailed micropaleontological studies in the Vrica section were
carried out. Paleontological studies of the continental deposits of
northern Italy were included as stratigraphic analogues of the
Pliocene-Lower Pleistocene marine beds.
During 1981, the Spanish working group continued biostratigraphic study of some key sections of Neogene-Quaternary age
to develop correlations to the stratotypical section of Italy. Work
also continued in various regions of the European and Asian
parts of the USSR to compile a series of key sections with clearly
distinguishable chronologic, biostratigraphic, and magnetostratigraphic evidence from the Olduvai event to the Gilbert-Gauss
boundary. A significant scientific event in this respect was the
Ail-Union Meeting on Quaternary Research at Ufa, Kuibyshev,
in August 1981, with part of the program devoted to the problem
of the N/Q boundary. The geological excursions included a
special visit to Pliocene and Lower Pleistocene deposits in
Bashkiria and Kuibyshev Zavolzhie. In addition, the fifth
meeting of the Soviet working group, held in April 1981,
recommended additional research, including drilling in the deltas
of the Dniestr, Don, Dnieper, Lena, and other rivers.
The national working group of China studied the stratigraphic
subdivision of the Pliocene-Quaternary deposits in the regions
of Pingliang, Xifeng, and Wuchi. In the course of studying the
Plio-Pleistocene deposits of the Beijing plain, a biological
boundary corresponding to the magnetostratigraphic boundary
between the Matuyama and Gauss paleomagnetic zones was
established. According to the Chinese working group, the N/Q
boundary in China should be drawn at this level, now dated to
2.5 Ma. Future studies were to be carried out in the western part
of the North China Plain.
Adoption of a proposal
A meeting of the IGCP-41 working group was held in Moscow
during the XI INQUA Congress in 1983, jointly with the INQUA
subcommission on the Pliocene-Pleistocene boundary. The
purpose of the meeting was to consider the progress made toward
implementation of Project 41, the results of studies on the N/Q
boundary-stratotype, the problems of correlation of the N/Q
stratotype with key N/Q sequences in various parts of the world,
and preparation of the final report. At that meeting it was
proposed, discussed, and adopted that the Vrica section, located
in Calabria, be designated as the Pliocene-Pleistocene boundarystratotype section.
The Vrica section, located in Calabria, southern Italy, had
been extensively described by Selli et al. (1977) and Colalongo,
Pasini, and Sartoni (1981), among other reports (Pasini and
Colalongo, Chapter 2, this volume). That Plio-Pleistocene
section satisfied all the internationally accepted guidelines for an
adequate boundary-stratotype: good vertical development (more

than 300 m), complete exposure, stratigraphical continuity,
abundance and variety of well-preserved fossils, facies (bathyal
marine sediments) favorable for recognizing time-significant
biohorizons in long-distance correlation, no structural complication or metamorphism, suitability for magnetostratigraphic
investigations, and accessibility. The Vrica section offered the
possibility of selecting a Pliocene-Pleistocene boundary-stratotype that was consistent with the original concepts of Pliocene
and Pleistocene, as described by Lyell and as elaborated by
generations of later workers. Moreover, the magnetostratigraphy of the section had already been determined, and some
radiometric ages of several included ash levels had been
obtained, as reported by Nakagawa et al. (1980), Nakagawa
(1981), and Tauxe et al. (1983), as reviewed by Nakagawa et al.
(Chapter 3, this volume).
At the 1983 meeting, the members of IGCP-41 and INQUA
Subcommission 1-a (i.e., 1-d) initially proposed to the INQUA
Commission on Quaternary Stratigraphy that the PliocenePleistocene boundary-stratotype be defined in the Vrica section.
The base of the bed of silty-marly claystone conformably
overlying the sapropelic bed e at the Vrica section, where this
level is exposed in profile B, as described by Colalongo et al.
(1982), was selected as the boundary marker point or "golden
spike," for many reasons. One of those was the fact that several
paleontologic markers (first and last occurrences of microfossils)
straddling the proposed boundary-stratotype occur in Italian
shallow-water sections near the first appearance of Arctica
islandica, which is one of the main criteria for locating the base
of the Pleistocene. The biostratigraphy at the selected boundary
level has been reviewed in detail (Pasini and Colalongo, Chapter
2, this volume). It was noted that these paleontologic markers
are found in the same order and in the same position with respect
to the Olduvai subchron in the Vrica section and in oceanic deepsea cores. Furthermore, the Mediterranean first appearance of
A. islandica, historically the index for the beginning of the
Calabrian (and, in present terms, the Santernian as well)
(Preface, this volume), clearly postdates the upper levels of the
Piacenzian stratotype, as defined by Colalongo, Elmi, and
Sartoni (1974). According to modern stratigraphic guides, this
means that the base of the lowest Pleistocene stage, if defined in
a level that accurately reflects the appearance of A. islandica in
Italian sequences, will also define the top of the Piacenzian,
according to the principle of "base defines boundary," at a level
that essentially accords with the previously accepted upper limit
of the Piacenzian. This condition appears to be met in the Vrica
definition, although A. islandica itself is not autochthonous in
this deep-sea section. Therefore, the recommended PliocenePleistocene boundary-stratotype will coincide with the top of the
Piacenzian, the youngest stage of the Pliocene, as well as with
the established marine biostratigraphic concept for the beginning
of the Pleistocene in Italy.
With the presentation of that final report, the work of the
IGCP Project 41 was brought to a successful close.
11
Subsequent activities
Acceptance of the Vrica section as a stratotype for establishing

the N/Q boundary was provisionally approved by members of
the working groups of IGCP Project 41 and the INQUA
Subcommission 1-d on the Pliocene-Pleistocene boundary. The
final decision was submitted to the IUGS Commission on
Stratigraphy and then to the XXVII International Geological
Congress in 1984. In 1985, the complete Vrica proposal was
published (Aguirre and Pasini, 1985) together with the announcement (Bassett, 1985) that the terms of the proposal had been
formally ratified by the ICS (International Commission on
Stratigraphy) as of May 31, 1985. The Vrica boundary-stratotype
was subsequently adopted by the IUGS Executive as a global
stratotype section and point (GSSP) (Cowie et al., 1986) and was
included as such in the IUGS 1989 Global Stratigraphic Chart
(Cowie and Bassett, 1989).
The subject of continental analogues to the earliest marine
Pleistocene beds was not addressed in the proposal to IUGS. At
present, it can be considered that the basal marine Pleistocene
(e.g., the Santernian of the Po Valley) (Ruggieri, Rio, and
Sprovieri, 1984) relates only to the upper Villafranchian. The top
of the Olduvai paleomagnetic zone coincides very closely with
the appearance of the Late Villafranchian mammalian fauna in
the continental deposits of the Mediterranean, as exemplified in
the fauna of the Olivola and Tasso intervals (Azzaroli et al.,
1988; Azzaroli et al., Chapter 11, this volume). The Olivola
fauna is presently attributed to the beginning of the Eburonian
Stage in The Netherlands; in turn, the beginning of climatic
cooling is recorded in faunal and floral markers seen in the
Eburonian, as well as in the Baventian in East Anglia and in the
lower Apsheronian in the former USSR.
References

Azzaroli, A., De Giuli, C , Ficarelli, G., and Torre, D. 1988.
Late Pliocene to early mid-Pleistocene mammals in Eurasia: faunal succession and dispersal events. Palaegeogr.
Palaeoclimatol. Palaeoecol. 66:77-100.
Bassett, M. G. 1985. Towards a "common language" in
Colalongo, M. L., Elmi, C , and Sartoni, S. 1974. Stratotypes of
the Pliocene and Santerno River section. Bur. Rech. Geol.
Min., Mem. 75 2:603-624.
Colalongo, M. L., Pasini, G., Pelosio, G., Raffi, S., Rio, D.,
Ruggieri, G., Sartoni, S., Selli, R., and Sprovieri, R. 1982.
A proposal on the Neogene-Quaternary boundary. Geogr.
Fisica Dinam. Quater. 5:59-68.
Colalongo, M. L., Pasini, G., and Sartoni, S. 1981. Remarks on
the Neogene-Quaternary boundary and the Vrica section
(Calabria, Italy). Soc. Paleontol. Ital, Boll. 20:99-120.
Cowie, J. W., and Bassett, M. G. 1989. International Union of
Geological Sciences 1989 stratigraphic chart. Episodes
12(2): (unpaginated insert).
Cowie, J. W., Ziegler, W., Boucot, A. J., Bassett, M. G., and
12
the Plio-Pleistocene boundary in the Vrica section,

Remane, J. 1986. Guidelines and statutes of the Internasouthern Italy. Nature 304:125-129.
tional Commission on Stratigraphy (ICS). Forschungsinst.
Senckenb. Cour. 100:53-107.
Hedberg, H. D. (ed.) 1976. International Stratigraphic Guide: a
guide to stratigraphic classification, terminology and proceAppendix: Selected publications of the USSR working
dure. New York: Wiley.
group
Nakagawa, H. 1981. Neogene-Quaternary boundary and correlation of the Vrica section. In Proceedings of the NIQ Alekseev, M. N., Arkhipov, S. A., and Deviatkin, E. V (eds.)
1981. The boundary between the Neogene and Quaternary
Boundary Field Conference, India, 1979, ed. M. V. A.
systems (unpublished collection of papers of the USSR
Sastry et al., pp. 107-111. Calcutta: Geological Survey of
scientists). Moscow: International Geological Correlation
India.
Program.
Nakagawa, H., Niitsuma, N., Takayama, J., Tokunaga, S.,
Kitazato, H., and Koizumi, I. 1980. Preliminary results of Alekseev, M. N., and Nikiforova, K. V. (eds.) 1987. Granitsa
mezhdu neogenovoj i chetvertichnoj sistemami v SSSR [The
magneto- and biostratigraphy of the Vrica section (Calaboundary between the Neogene and Quaternary in the
bria, Southern Italy). In Proceedings of the second
USSR]. Moscow: Nauka.
symposium on the Neogene/Quaternary boundary, USSR,
1977, ed. K. V. Nikiforova and A. Y. Dodonov, pp. 145- Gabunia, L. K., and Vekua, A. K. 1979. The terrestrial
156. Moscow: Akademia Nauk.
mammals of the Pliocene and the early Pleistocene and
Nikiforova, K. V., and Dodonov, A. Y , (eds.) 1980. Proceedings
the boundary between the Neogene and the Quaternary
of the second symposium on the Neogene/Quaternary
system in Georgia (abstract). In Field conference,
boundary, USSR, 1977. Moscow: Akademia Nauk.
Neogene/Quaternary boundary, ed. V S. Krishnaswamy,
Pasini, G., Selli, R., Tampieri, R., Colalongo, M. L.,
p. 49. Paris: International Union of Geological Sciences.
D'Onofrio, S., Borsetti, A. M., and Cati, F. 1975. The Nikiforova, K. V. 1980. Projekt "Granitsa mezhdu neogenom i
Vrica section. In The NeogeneQuaternary Boundary: II
chetvertichnoy sistemoy" [Project "The boundary between
symposium (Bologna-Crotone), excursion guide-book, ed.
the Neogene and Quaternary system"]. Akad. Nauk SSSR,
R. Selli, pp. 62-72. Bologna: Editografica Rastignano.
Vestn. 1980:98-100.
Ruggieri, G., Rio, D., and Sprovieri, R., 1984. Remarks on the Nikiforova, K. V. 1985. The boundary between the Neogene and
Quaternary: where to draw it? Nature and Resources
Soc. Geol. Ital, Boll. 103:251-259.
21:35-38.
Sastry, M. V. A., Kurien, T. K., Dutta, A. K., and Biswas, S. Nikiforova, K. V. 1988. Project 41: Neogene-Quaternary
(eds.) 1981. Proceedings of the NIQ Boundary Field
boundary. Geol. Correl., special issue 1983 7:50-51.
Conference, India, 1979. Calcutta: Geological Survey of Nikiforova, K. V, and Alekseev, M. N. 1989. Sovremennoje
India.
sostojanie problemy granitsy mezhdu neogenovoj i chetverSelli, R., Accorsi, C. A., Bandini Mazzanti, M., Bertolani
tichnoj sistemami [The present state of the problem: the
boundary between the Neogene and Quaternary systems].
Inst. Geol., Akad. Nauk SSSR, Trudy 431:227-252.
Landini, W., Menesini, E., Mezzetti, R., Pasini, G., Nikiforova, K. V, Alekseev, M. N., Krasnov, I. I., et al. 1985.
Nizhnaha granitsa chetvertichnoy (antropogenovoy) sis(Calabria-Italy). A potential Neogene/Quaternary boundtemy [The lower boundary of the Quaternary (Anthroary stratotype. Giorn. Geol. 41:181-204.
pogene) system]. Akad. Nauk SSSR, Izvestia, ser. geol.
Tauxe, L., Opdyke, N. D., Pasini, G., and Elmi, C. 1983. Age of
7:9-14.
Part II
Characterization of the Pleistocene boundary-stratotype
The Pliocene-Pleistocene boundary-stratotype at Vrica, Italy

GIANCARLO PASINI and MARIA LUISA COLALONGO
the Pliocene-Pleistocene boundary, so consequently the type

section should be chosen in the marine sequences . . . of
The Plio-Pleistocene sequence of deep-water marine beds at Southern Italy."
Vrica, Calabria, Italy, was proposed by INQUA Subcommission
Accordingly, among the sections located in Calabria, three
1-d, "Pliocene/Pleistocene Boundary," and by the working group were considered as candidates for the location of the Plioceneof IGCP Project 41, "Neogene/Quaternary Boundary," as the Pleistocene boundary: (1) Santa Maria di Catanzaro, (2) Le
location for a physical boundary-stratotype for the base of the Castella, and (3) Vrica. Studies by Colalongo, Pasini, and
Pleistocene, according to modern chronostratigraphic guide- Sartoni (1981) indicated that neither the Santa Maria di
lines. The recommended level has been adopted formally by the Catanzaro section nor the Le Castella section was suitable as a
IUGS (Nikiforova and Alekseev, Chapter 1, this volume). That standard section for defining the Pleistocene boundary-stratosection is herein characterized in terms of sedimentology, type, because of evidence that each contains a hiatus correspondpaleoecology, biostratigraphy, biochronology, and magnetostra- ing to the uppermost Pliocene and the lowermost Pleistocene.
tigraphy, based on a decade of studies by stratigraphers in Thus, the only one of those sections suitable for defining the
different countries. According to criteria established by prior Pleistocene boundary-stratotype was at Vrica, where a continurecommendations, the physical location of the Pleistocene ous, highly fossiliferous bathyal sequence is exposed.
boundary-stratotype is identified with the base of the claystone
The Vrica section is located about 4 km south of the town of
conformably overlying marker bed e of the Vrica section. This Crotone (Figure 2.1). Like the areas of Le Castella and Santa
level is very close to the Olduvai normal-polarity subzone and is Maria di Catanzaro, which are 19 km to the southwest and 55 km
approximately coeval with the beginning of a cold-climate phase to the west-southwest, respectively, the Vrica area is an
marked by the first appearance of the "northern guest" Arctica emergent portion of the Crotone sedimentary basin. According
islandica, a mollusk confined to boreal waters during interglacial to Barone et al. (1982), the Crotone basin is a graben-like
periods.
structure containing postorogenic sediments deposited since
Tortonian times on top of the nappes making up the emerged
Calabrian Arc. After the early Pleistocene, a large portion of the
Background
Crotone basin emerged north of the Catanzaro isthmus, exposAt the XVIII International Geological Congress in London in ing Miocene sediments and a thick Plio-Pleistocene sequence.
1948, the council unanimously accepted the recommendation of
The Plio-Pleistocene sediments of the Marchesato peninsula
the temporary commission to advise on the question of the (Figure 2.1), including the Vrica area, are very gently folded,
definition of the Pliocene-Pleistocene boundary, which made the with axes oriented approximately north-south; the folds are cut
following essential points: "The Commission considers that the by normal faults (Selli, 1977). The Vrica section is located in a
Pliocene-Pleistocene boundary should be based on changes in regular monocline, dipping 5-15 westward (Figure 2.2).
marine faunas. . . . The classic area of marine sedimentation in
Italy is regarded as the area where this principle can be
Physical stratigraphy of the Stuni-Vrica sequence and
implemented best" (King and Oakley, 1950).
the Vrica section
Following that recommendation, during the joint meeting in
India in 1979 of the working group for IGCP Project 41, In the extensive badlands area located between Crotone and C.
"Neogene/Quaternary Boundary," and the INQUA Subcommis- Colonne (Figures 2.1 and 2.4) a well-exposed Plio-Pleistocene
sion 1-a (since changed to 1-d), "Pliocene/Pleistocene Bound- sequence crops out. The lowermost part of this sequence is
ary," it was further resolved that "the territory of Southern Italy formed by gray claystones of the "Semaforo Formation" of Selli
(the region of Calabria) is selected as type area for establishing (1977) (Figures 2.2 and 2.3), which is middle Pliocene in age. At
Introduction
15
16
Giancarlo Pasini and Maria Luisa Colalongo
16*150
Figure 2.1. Geologic map of the

Marchesato Peninsula, an emerged
portion of the Crotone sedimentary
basin. The submerged part of the basin is located southeast of this peninsula: 1, continental sediments (Pleistocene and Holocene); 2, sediments of
the Upper Pleistocene marine terraces; 3, Middle-to-Upper Pliocene
and Lower Pleistocene sediments; 4,
Tortonian-Lower Pliocene sediments; 5, rocks of the substrate; 6,
main faults; 7, possible faults. (From
Tortorici, 1982, courtesy of Societa di
Geologica Italiana.)
Figure 2.2. Geology of the Vrica

area: 1, beach sand and dunes; 2,
calcarenites and sands of the
"Milazzian" marine terrace; 3, gray
silty-marly claystones, with intercalated sapropelic layers (Lower Pleistocene); 4, first appearance of
Hyalinea baltica, 35 m above the m
volcanic ash layer (maximum age
1.99 Ma); 5, gray silty-marly
claystones, with intercalated
sapropelic layers (Lower Pleistocene, Upper and Middle Pliocene);
6, lower volcanic ash layer (/..)>
dated to 2.22 Ma; 7, gray
claystones (Middle Pliocene); 8,
faults; 9, strike and dip; 10,
component-sections measured and
sampled by Pasini and co-workers
(Figure 2.3, second column); 11,
component-sections measured and
sampled by Nakagawa and coworkers (Figure 2.3, third column).
(Adapted from Selli et al., 1977.)
the top of these claystones is the middle Pliocene volcanic ash

horizon La. (=lower ash), with a thickness of about 20 cm. The
overlying portion of the sequence, about 380 m thick, is formed
by gray silty-marly claystones with several shale layers and a few,
rare sandy horizons, belonging to the Upper Pliocene-Lower
Pleistocene "Tripolacea Formation" and "Papanice Formation"
of Selli (1977). Approximately 100 m below the top of the
17'|oo'
17*|1Q'
CROTONE
VRICA
fCROPANI
MARINA
SECTION
LE CASTELLA
SECTION
Hi
]JA
sequence there is a second volcanic ash horizon, layer m, varying

in thickness between 2 and 7 cm. The lower and middle parts of
this sequence are well exposed in gullies in the Stuni region, and
the upper part crops out near the site of Vrica, a ruined
farmhouse used as a geodetic point (Figure 2.2). Thus the StuniVrica sequence, as the full section is called, is a composite of
correlated component-sections from different areas (e.g., Fig-
Plio-Pleistocene boundary-stratotype, Vrica
COLUMNAR
SECTION
L I T HO LO GY
COMPONENT-SECTIONS
ace.to
this paper
450
ace to
Nakagawa
et al.1980
G& I pumice block

"| volcanic ash layer
| sandy layer
U ^ J sapropelic layers (shale layers)

I or tripoli ( 58, 59 and Sio only)
C
400
17
JA
1 sapropel - clay layer
03
"""7! laminated siltstone
silty marly claystone or

clayey siltstone (below
component- section A)
CO
0
350
B
c
300
o
o
o
JB
the component-section below the

wavy lines is not correctable with
the component-section above them
through lithological elements
03
250
the component - section below

the horizontal line is easily corre.
latable with the component-section
above it through lithological elements (groups of shale layers,
sandy layers etc)
LU
200
150
JC
56
S7
100
58
59
Sio
1
50
JD
m0
ures 2.5-2.7). The Stuni-Vrica sequence has been measured and

sampled independently by Italian and Japanese geological teams
(Pasini et al., 1975; Selli et al., 1977; Nakagawa, 1981; Nakagawa
et al., 1980; Nakagawa et al., Chapter 3, this volume).
The Italian geological team has used the name "Vrica Section"
in a restricted sense to mean the middle and upper parts of the
Stuni-Vrica sequence (Figure 2.3), specifically a stratigraphic
interval 306 m thick measured from 125 m below the marker bed
a to the exposed top. The level of the Pleistocene boundary-
Figure 2.3. The Stuni-Vrica sequence and the Vrica section.

The component-sections selected
by Pasini et al. (1975), and described in this chapter are compared with those selected by
Nakagawa et al. (1980), marked
originally as A, B, C, and D,
but shown here as JA, JB, JC,
and JD, for clarity. The columnar section from zero to 140 m
is taken from Nakagawa et al.
(1980), and that from 140 m up
to the top is from Selli et al.
(1977).
stratotype is approximately in the middle of this measured

sequence, about 140 m below the top.
The Vrica stratigraphic section, from which the Italian team
has collected 275 samples, is represented in three partly
overlapping component-sections identified as A, B, and C
(Figures 2.2-2.7). The correlations between the componentsections are based on conspicuous and clearly identifiable
groupings of sapropelic shale layers. The beds of the lowermost
45 m of component-section A (from the sapropel-clay layer
18
Pliocene and the lower part of the Pleistocene. This sequence

includes the Pliocene-Pleistocene boundary level according to
the criteria presented here, and its upper part has been
accurately correlated to the Vrica section by Zijderveld et al.
(1991). The same authors also correlated the uppermost part of
the Vrica section to the Crotone section (located in the environs
of this town), where about 40 m of strata overlying marker bed t
are exposed.
Sedimentology and paleoecology
The Vrica section, 306 m thick, is a well-exposed, continuous
sequence of silty-marly claystones sparsely intercalated with
laminated sapropelic shales and fine sands, with one volcanic ash
layer (Figure 2.3). As noted, 275 samples were collected for
biostratigraphic and sedimentologic studies.
Silty-marly claystones
Figure 2.4. Aerial photograph of the Vrica outcrop area. For the locations of the component-sections A, B, and C, compare this photograph
with Figures 2.19 and 2.20. (I.G.M.I. photo no. 227/10145, 1955, reproduced with permission of the S.M.A. in publication no. 231, of April 7,
1970.)
down to the base: see Figure 2.3) are not well exposed, and the
section is synthesized by correlation between scattered outcrops
of claystones.
The Japanese geologists initially reconstructed a PlioPleistocene sequence in the Vrica area starting from a level 67
m below the La. (lower ash) horizon up to the top, for a total
thickness of about 450 m, in four component-sections D, C, B,
and A in ascending stratigraphic order (Nakagawa, 1977, 1981;
Nakagawa et al., 1980) (Figure 2.8). In Figures 2.2 and 2.3, and
throughout this chapter, the Japanese component-sections are
given here as JD, JC, JB, and JA, to distinguish them from the
Italian A, B, C, and D component-sections. The componentsections JB and JA were measured in the same gullies where
the Italian team measured the component-sections B and C
(Selli et al., 1977) and therefore relate to the same boundarystratotype section and point.
In recent years, the Vrica section has been extended downward and upward. At Monte Singa, Calabria, about 100 km
southwest of Vrica, a thick sequence of marine marls and clays
with sapropelic layers spans the middle and upper parts of the
The Vrica section is dominated by silty-marly claystones, gray or

blue-gray in color, obscurely bedded, and containing 60.078.9% clay, 20.5-38.3% silt, and 0.7-9.3% fine to very fine
sand. Its carbonate content ranges from 14.5% to 25.1%. These
sediments were formed mainly from clay minerals and the
remains of calcareous nannoplankton and foraminifera (with
planktic species more abundant than benthic). Variable, though
subordinate, amounts of volcanic glass and detrital minerals
(feldspars, micas, glauconite) are present. Rarely, the claystones
show surfaces and beds with red staining, probably due to
limonite. Other claystone beds show burrows filled with indurated clay (often reddish), and some are distinguished by
abundant brachiopods, echinoid fragments, or pteropods. Wellpreserved mollusks, mostly small gastropods, are rare; a few
solitary corals, otoliths, and fish teeth have also been collected.
In addition to calcareous nannofossils and foraminifera, the
well-preserved microfauna includes abundant ostracodes. The
presence within the claystones and throughout the section of
psychrospheric ostracodes such as Agrenocythere pliocenica,
Bathycythere vanstraateni, Zabythocypris antemacella, and Bythoceratina scaberrima mediterranea (Colalongo and Pasini, 1980a),
deep-water benthic foraminifera including Articulina tubulosa,
Bolivina albatrossi, Cassidulina carinata, Discospirina italica,
Eggerella bradyi, Gyroidina soldanii, Hoeglundina elegans,
Planulina wuellerstorfi, Karreriella bradyi, Pleurostomella alternans, Pullenia bulloides, and Rhabdammina linearis (D'Onofrio, 1981), and the deep-water mollusks Propeamussium duodecimlamellatum and Malletia excisa (Selli et al., 1977) indicates
that the deposition of the clays took place in a bathyal
environment, in water depths of approximately 500-800 m. This
paleodepth range is in agreement with the interpretation of the
fish fauna, as discussed later.
Shale layers (saprop elites)
Fourteen conspicuous layers of thinly laminated shales are
interbedded with the massive claystones of the Vrica section.
19
Figure 2.5. Portion of component-section A of the Vrica section, located

above the sapropel-clay layer (see Figure 2.3).
These shale layers are useful marker beds; their thicknesses,

measured in the line of section (Figure 2.2), are approximately as
follows (from top down):
shale layer t
shale layer s
shale layer r
shale layer q
shale layer p
shale layer o
shale layer n
shale layer h
shale layer /
shale layer e
shale layer d
shale layer c
shale layer b
shale layer a
75 cm
65 cm
35 cm
80 cm
55 cm
55 cm
100 cm
90 cm
115 cm
190 cm
155 cm
340 cm
115 cm
70 cm
The shale units are gray-pink in color and show undisturbed

primary laminae parallel to the bedding that measure from a few
millimeters to less than 1 mm in thickness. The composition of

the shales is 56.7-78.3% clay, 20.6-36.8% silt, and 0.9-11.3%
fine sand, with a carbonate content ranging from 11.3% to
19.2%.
In our opinion (Pasini and Colalongo, 1982) the laminites of
the Vrica area should be interpreted as sapropelic layers and
possibly as true sapropels (Kidd, Cita, and Ryan, 1978) from
which some organic carbon has been removed by surface
weathering. Like typical sapropels (Sigl et al., 1978), the Vrica
organic laminites appear to have been formed during short
periods of oxygen deficiency in the bottom water, most probably
during intervals of poor ventilation related to overwhelming
surface productivity, judging from the following observations.
First, because primary laminae deposited on well-oxygenated
bottoms are quickly destroyed by mud-eating organisms, such
laminae can be preserved only where such organisms are rare or
absent because of anoxic conditions at the sediment interface. In
the Vrica laminites, a very few clay-filled burrows, all starting
from the overlying claystones, indicate that mud-eating organisms that lived after the deposition of the sapropelite shales
Figure 2.6. Component-section B of the Vrica section (in the foreground). Marker beds c, d,
e, / , and h are indicated. On the right side, the uppermost part of component-section A and
the Costa Tiziana Hotel are visible.
21
Figure 2.7. Component-section C

of the Vrica section, exposed along
the ravine to the left.
sometimes deepened their burrows down into the shale layers.

Second, the planar and undisturbed lamination also suggests the
absence of bottom currents. Finally, the virtual absence within
the shale units of the benthic epifauna that is common in the
enclosing mudstones is another indication of a severe oxygen
deficit in bottom waters. According to D'Onofrio (1981), the
presence of dwarfed, rare bolivinids and brizalinids is due to "the
presence of oxygen-depleted waters" during deposition of the
shale layers.
Correlative evidence comes from the presence of fish remains
in all of the sapropelic layers examined (9 out of the 15 shale
layers, b to p) (Landini and Menesini, 1978a). The abundance
and good preservation of the fish fossils indicate the absence of
scavengers and aerobic bacteria (Brongersma-Sanders, 1957). In
addition, the presence of minerals that are formed through
alteration of iron sulfides, such as gypsum, goethite, limonite,
and (rarely) jarosite, suggests that the Vrica laminites contained
primary sulfides, strongly indicating euxinic conditions. It may
be that some of these layers are coeval (considering the
geographic location of the area) with the Pliocene and Quaternary sapropels of the eastern Mediterranean (Stanley, 1978;
Kidd et al., 1978), but Raffi and Thunell (Chapter 4, this
volume) were not able to identify any of the Vrica sapropelic
layers with Plio-Pleistocene sapropels in cores from DSDP
(Deep Sea Drilling Project) site 125.
The fish species identified by Landini and Menesini (1978a)
from the sapropelic layers today live exclusively or preferentially
at depths exceeding 500 m, which supports the evidence from the
benthic microfauna that the Vrica sediments were deposited in a

bathyal environment.
The sapropelic laminites of the Vrica section have been further
studied by Howell, Rio, and Thunell (1990), Hilgen (1991), and
Lourens et al. (1994). In the two latter papers the sapropelites
were calibrated to astronomically forced climate cycles.
Sandy layers
About 100 m below the top of the Vrica section, three closely
spaced sandy layers (g, /, and /) are interbedded with the
claystones. These layers are pale gray in color, weathering to
reddish. The thicknesses of the sandy layers in the line of section
are as follows:
sandy layer / 3-6 cm
sandy layer / 3-6 cm
sandy layer g 6-10 cm
These layers, which are very useful (kilometers wide) marker
beds, contain 1.0-3.5% fine sand (3 to 2 <\> units), 37.1-60.4%
very fine sand (4 to 3 <$> units), 15.2-21.1% silt, and 22.0-40.8%
clay. The carbonate content ranges from 12.3% to 14.3%.
The sandy layers are composed of calcareous nannoplankton,
foraminifera, volcanic glass (more or less altered), clay minerals,
and relatively coarse-grained terrigenous debris, including
quartz (mostly metamorphic), altered plagioclases, micas, and
sparse lithic fragments. The terrigenous debris, mostly of
metamorphic origin, probably came from the Paleozoic crystal-
COLUMNAR
SECTION
Abundance
Lithology
clayey siltstone
laminated
j---.- sand
siltstone
layer
of
Pollen
and
spore
0 < I < 1 '

1< < 5
tripoli
.....
tuff
u
Coiling ratio
L 100 V. left
R 100 V. r i g h t
5^
< 10
10 <
< 20
20 < ^
< 50
50 < A
Occurrence of
Foraminifera and
calcareous
nannoplankton
very rare
Figure 2.8. Stratigraphic distributions of foraminifera, calcareous nannofossils, pollen,

and spores in the Stuni-Vrica sequence, according to Nakagawa and co-workers.
(Adapted from Nakagawa, 1981, courtesy of the Geological Survey of India.)
line rocks of the Sila massif, about 40 km west-northwest of the

Vrica area.
Volcanic ash layer m and pumice block
Layer m, a 2-7-cm-thick silt-textured layer of pale gray color
(reddish where weathered), crops out on the line of componentsection B about 6 m above the top of the sapropelic shale layer h,
and 2.5 m above the sandy layer /. Macroscopically it is very
similar to the sandy layers g, i, and / and is also a useful marker.
It has the following granulometry: 3.9% fine sand, 10.9% very
fine sand, 51.4% silt, and 33.8% clay; the carbonate content is
relatively low, at 3.3%.
Mineralogically, layer m is made up of about 80% very fresh,
colorless, often fibrous or vesicular volcanic glass, with small
amounts of fresh, euhedral to subhedral andesinic plagioclases,
well-preserved biotite, altered micas, amorphous ferric aggregates, microfossils, and carbonate fragments. The glass is
chemically rhyo-dacitic, with a refraction index of 1.512 0.001
(Savelli and Mezzetti, 1977). Plagioclase and biotite crystals are
included within the glass shards in some samples. The abundance
of glass and the freshness of the volcanic materials, including the
glass itself, indicate a primary ashfall that was essentially
synchronous with eruption.
A block of pale-gray pumice, about 30 cm in diameter and
perfectly preserved, was delivered to one of the authors (G.P.)
by a quarryman from Crotone who had extracted it from a clay
quarry some 3 km north of component-section C and about 1 km
south of the center of Crotone, near the coastal road. In that
quarry the volcanic ash layer m, the three sandy layers g, /, and /,
and the sapropelic shale layers f to p crop out. The quarryman
was able to show the exact spot where the pumice block was
found, about 1 m above the volcanic ash layer m (Selli et al.,
1977) (Figure 2.3).
The pumice block consists of fresh, colorless glass (refraction
index = 1.500 0.001) and includes phenocrysts of hornblende
and plagioclase (Savelli and Mezzetti, 1977; Obradovich et al.,
1982). The mineralogy and chemistry of the pumice are unlike
those of the underlying volcanic ash layer m, which contains
biotite and plagioclase. Instrumental neutron activation analysis
(INAA) of the glass fraction indicates that the pumice is in fact
very similar to the thick volcanic ash layer in the lower part of the
Stuni-Vrica sequence (layer La. in Figure 2.3) (Obradovich et
al., 1982).
Biostratigraphy and biochronology
The Vrica section is very rich in fossils. Groups that have been
studied include calcareous nannoplankton, planktic and benthic
foraminifera, ostracoda, mollusks, fish, and pollen, as well as
diatoms (Palmer, Chapter 6, this volume). Besides these groups,
silicoflagellates, radiolarians, octocorals, brachiopods, pteropods, and echinoderms are present.
In this section we consider the biostratigraphy and the
biochronology of the studied groups, in particular the fossils
23
whose first appearances (FA) or last occurrences (LO) in the

Mediterranean and also in the extra-Mediterranean regions are
considered by most authors as key events in late Pliocene or
early Pleistocene time. These biotic events can now be related to
the boundary-stratotype (sensu Hedberg, 1976) of the Pleistocene as adopted in the Vrica section.
Calcareous nannoplankton
The calcareous nannoplankton of the Vrica section have been
studied by Nakagawa (1981), Nakagawa et al. (1980), Cati and
Borsetti (1981), Raffi and Rio (in Pasini and Colalongo, 1982),
Backman, Shackleton, and Tauxe (1983), Lourens et al. (1994),
and Rio, Raffi, and Backman (Chapter 5, this volume) (Figures
2.8-2.11).
Rio et al. (Chapter 5, this volume) emphasize that because of
reworked specimens, the extinction levels of species in the
section can be accurately identified only through careful
semiquantitative or quantitative analysis (Raffi and Rio, cited in
Pasini and Colalongo, 1982; Backman et al., 1983). The latter
authors have also standardized the informal morphometric
taxonomy of the Gephyrocapsa group as used by Rio (1982), in
place of the historically varied taxonomic concepts that have
made the studies in previous publications difficult to compare.
Among the major nannofossil events identified by Rio et al.
(Chapter 5, this volume) in the Vrica section, the most
important, from both the stratigraphic and biochronologic points
of view, are (from younger to older) as follows: the first
appearance of Gephyrocapsa spp. with coccoliths larger than 5.5
fim (i.e., "large" Gephyrocapsa); the extinction of Calcidiscus
macintyrei;
the first a p p e a r a n c e of Gephyrocapsa
oceanica
s.L,
sensu Rio (1982) (i.e., "medium-sized" Gephyrocapsa, between

4.0 and 5.5 /JLHI; teste D. Rio); the synchronous extinctions of
Discoaster brouweri and D. triradiatus. The relationships of
these events to one another and to the magnetostratigraphy are
the same in the Vrica section as in cores from the Mediterranean,
the Atlantic, the Pacific, and the Caribbean.
Lourens et al. (1994) calibrated the Lower Pleistocene
sapropels of the Vrica-Crotone composite section (as discussed
earlier) to the astronomical record in order to obtain an
accurate, high-resolution chronology for the Pleistocene part of
this section. Those authors analyzed oxygen isotopes from
microfossil samples and used their new age model to construct a
218
O curve. They concluded that "identification of oxygen isotope
stages in the Vrica/Crotone composite and their correlation to
obliquity is consistent with the astronomical calibration of these
stages proposed by Shackleton et al. (1990)" (Lourens et al.,
1994). In addition, those authors carried out a biostratigraphic
study of the calcareous nannofossils and selected planktic
foraminifera in the Vrica-Crotone composite section to correlate calcareous plankton events to the oxygen-isotope stages.
The stratigraphic positions of the nannofossil events indicated by
Lourens et al. (1994) are very similar to those shown herein
(Figure 2.9).
Raffi et al. (1993) studied the Plio-Pleistocene nannofossil
24
biostratigraphy of DSDP site 607 (midlatitude North Atlantic)

and Ocean Drilling Program (ODP) site 677 (eastern equatorial
Pacific) and pointed out that "the successful astronomical
calibration of oxygen isotope stratigraphies from Deep Sea
Drilling Project site 607 and Ocean Drilling Program site 677 in
the Matuyama Chron [as discussed by Shackleton et al., 1990]
permits calibration of the biostratigraphic events to these
uniquely resolved isotope chronologies." In ODP Leg 138 sites
(eastern equatorial Pacific), Shackleton et al. (1995a) dated
numerous first and last occurrences of Neogene microfossils
belonging to different groups (nannofossils, radiolaria, planktonic foraminifera, diatoms) based on the astronomically calibrated time scale of Shackleton et al. (1995b).
The results obtained by Lourens et al. (1994), Raffi et al.
(1993), and Shackleton et al. (1995a), as they pertain to the
calcareous nannofossil biostratigraphy in the Vrica section, will
be discussed in order of increasing age. Ages shown in italics are
those calculated for the Vrica section by Lourens et al. (1994).
Event 4:1.56 Ma(?). First appearance (FA) of "large" Gephyrocapsa. This event (Figures 2.9, 2.11) occurs at isotope stage
boundary 53/52 in the Vrica section, dated to 1.56 Ma, and
practically in the same position in the Singa section (Lourens et
al., 1994). In DSDP site 607, the FA of "large" Gephyrocapsa
occurs in the top part of stage 49, dated to 1.479 Ma; in ODP site
677 this event is recorded in the top part of stage 48, dated to
1.457 Ma (Raffi et al., 1993). In ODP Leg 138 sites the estimate
for the time of this event is 1.44 Ma (Shackleton et al., 1995a).
According to Berger et al. (1994), the FA of "large" Gephyrocapsa occurs in the western equatorial Pacific at 1.515 0.025
Ma. If we accept the age estimate of Lourens et al. (1994), this
event would have occurred in the Mediterranean about 80 k.y.
earlier than in the midlatitude North Atlantic. According to F. J.
Hilgen (personal communication), one option would be to
accept that the "large" Gephyrocapsa FA was in fact earlier in
the Mediterranean than in the open oceans; a second option
would be a hiatus in both the Vrica and Singa sections between
the LO of C. macintyrei and the FA of "large" Gephyrocapsa.
Considering that the C. macintyrei LO has almost the same age
in the Vrica section as in the Singa section (Lourens et al., 1994),
it would be required that such a hiatus would have to represent
precisely the same time interval in both sections, which are about
100 km apart and had different sedimentation rates. In our
opinion, the presence of such a hiatus is very unlikely, and the
noted event probably was diachronous between the Mediterranean and the open oceans.
Legend
1 specimen
Rare
Scarce
Abundant /
COLUMNAR
SECTION
* i z i '
<
LU 300
Z
LU
o 250
I
(D
LU
a.
200
LU
150
z
LU
O
100
I
50
CL
mO
Event 3: 1.67 Ma. Extinction of Calcidiscus macintyrei. This Figure 2.9. Ranges of selected nannofossil species in the Vrica section,
event is recorded in the Vrica section at stage boundary 59/58. In according to I. Raffi and D. Rio (in Pasini and Colalongo, 1982).
DSDP site 607 it occurs at stage boundary 58/57, dated to 1.640
Ma; in ODP site 677 it is recorded in the top part of stage 55,
dated to 1.597 Ma (Raffi et al., 1993). The estimate for the
extinction of C. macintyrei in ODP Leg 138 sites is 1.58 Ma
(Shackleton et al., 1995a). In ODP site 806 this event is dated to
1.627 0.025 Ma (Berger et al., 1994).
No. of
0 BROUWERI mm2
as
10
1.5
25
No Of
0. BP0UWER1 m m 2
a BROUWERI
No of
NO. OF PREH. SELLU
WR TRIRADIATUS CMAQNTYREI mm2 PLIOCENE
(A/.50I
OSCQASTERS
5mm 2
0 20 40%
10 15
0 20 40% 0
No Of
C MAONTYREl mm
20
J 0
40
H SELLU
300 f 1
Myr
145 Myr
200.=
0. BROUWERI
VAR TRIPADI AT
^ . 1 4 5 Myr
liool
1
Figure 2.10. Quantitative nannofossil stratigraphy for the Vrica section
and piston core V28-239 (western equatorial Pacific). On the left are data
from the Vrica section, starting with the lithologic section of Selli et al.
(1977) and the magnetostratigraphy of Tauxe et al. (1983). The abundances
of Discoaster brouweri, D. brouweri var. triradiatus relative to all forms of
D. brouweri, and Calcidiscus macintyrei each show a clear upper limit. PrePliocene discoasters maintain a uniform reworked abundance throughout
the section, and the proportion of Helicosphaera sellii relative to all
helicosphaerids also does not show any drop in abundance in the upper
part of the section. On the right side, the sequence in piston core V28-239 is
from Backman and Shackleton (1983), with magnetostratigraphy from
Shackleton and Opdyke (1976). (From Backman et al., 1983, with permission of the editors of Nature.)
20
after the reversal, within the basal Olduvai. Both the disappearance event and the reversal boundary are assigned age estimates,
however, of 1.95 Ma [according to] Shackleton et al. (1990)." In
ODP site 677 the extinction levels for D. brouweri and D.
triradiatus are not clear, but the estimate for the extinction of D.
brouweri in ODP Leg 138 sites is 1.96 Ma (Shackleton et al.,
1995a).
According to Nakagawa (1981), Raffi and Rio (in Pasini and
Colalongo, 1982), Backman et al. (1983), and Rio et al. (Chapter
5, this volume), Helicosphaera sellii is present throughout the
Event 1:1.95 Ma. Extinction of Discoaster brouweri. This event Vrica section (Figures 2.8-2.11), and therefore the top of this
is recorded in stage 72 in the Vrica section (Lourens et al., 1994) section is older than the regional extinction of this species.
(Figure 2.4), practically in coincidence with the lower boundary According to Lourens et al. (1994), the H. sellii LO occurs in the
of the Olduvai normal-polarity subzone, as defined by Zijder- Crotone section about 35 m above marker bed t, at the isotope
veld et al. (1991). In DSDP site 607 it occurs at stage boundary stage 38/37 boundary.
72/71, dated to 1.950 Ma, simultaneously with the extinction of
Discoaster brouweri and D. triradiatus. Raffi et al. (1993, pp.
Planktic foraminifera
399-400) affirm in this regard that "the results from core V28239 [western equatorial Pacific] and those from DSDP site 606 The planktic foraminifera assemblages of the Vrica section are
[subtropical North Atlantic] (Backman and Pestiaux, 1987) . . . very diverse and abundant, in genera as well as species, and are
seem compatible with those from site 607, which indicate that generally very well preserved. Figure 2.12 shows the stratiboth D. brouweri and D. triradiatus have their last occurrence graphic distributions of selected planktic foraminifera in the
virtually at the Olduvai . . . [lower] boundary, although this Vrica section, based on studies by Colalongo and Sartoni (in Selli
disappearance event probably occurred a few thousand years et al., 1977), Colalongo et al. (1980, 1981, 1982), Pasini and
Event 2: 1.71 Ma. First appearance of Gephyrocapsa oceanica
s.l., sensu Rio (1982) (i.e., medium-sized Gephyrocapsa between
4.0 and 5.5 ^m). This event in the Vrica section is documented
immediately below marker bed g, just above the Olduvai
subzone (Figures 2.9 and 2.18), and occurs within stage 60 both
here and in DSDP site 607, where it is dated to 1.700 Ma (Raffi
et al., 1993). In ODP Leg 138 sites the estimate for the time of
this event is 1.69 Ma (Shackleton et al., 1995a), and in ODP site
806 it is 1.664 0.025 Ma (Berger et al., 1994).
O
3Z
"
s
q
J_H.BALTICA
DISCOASTER
BROUWERI
TRIRADIATUS
D.BROUWERI
VAR,
50/<
BACKMAN ET A L .
10%
CALCIDISCUS
SELLII
HELICOSPHAERA
<
-J
<
PQ
cc
ce
cj
CO
cc
\-
MACINTYREI
y
S /
GEPHYROCAPSA
<
Ln
CD
GROUP
CJ
<
CD
CO'
/
/
CD
CJ
CD
<
CO
IN
/z.
CO
<
<
RAFFI AND RIO NAKAGAWA ET AL CATI AND BORSETTI
S5
11 I
_J
1 ^
//
//
//
/.
//
OCEANIC
SAMPLE
POSITIONS
_]
CQ
ce:
CO
CJ
<
<
<
<
OC CQ
<
<
__
TI AND EORSET
en
APERTA
en
7-r
^ ^ ^ ^ - ^ ^ ^
77
CARIBBE
ERICSON
MARGERE
oc
CD
J _ C . TESTUDO
h
'g
e
d
c
j
CKMAN ET
CD
300-
250-
200-
150-
/ /
/ /
/ /
/ /
FFI AND
""'
cc:
<c
c_>
cJ
cc:
CD
=
1
CKMAN ET AL.
TI AND B
en
<
u_
u_
S5
c_>
oo
CD
1
C_)
100-J
J _ G . INFLATA
FFI AND RIO
KAGAWA E
ce:
ce:
CD
CD
CD
CD
*->
OO
=1
=c
__
50-
TI AND BORSET
<
CARIBBE
OCEANIC
cS
CD
z
CD
O
CJ
UJ
CM
ce:
CQ
cc:
CD
CQ
C_J
CD
cc:
UJ
oo
c_>
oo
ca
FFI
CKMAN ET
BIOSTRATIGRAPHY
<->
g
oo
UJ
ca
AND RIO
FFI
en
en
oo
o
<c
., .1
<c
1
s:
UJ
CD
UJ
ca
oo
cc:
CM
oo
0 ^
KAGAWA E
TI AND B
AND
RIO
D
ROUW
oo
11 1
CD
oo
OO
oo
ca
CD"
OO
PSEU 30EMILIA
ELLII
HEL ICO SPH

IDES
GARTNE
SELLI
HELICOPO
MACINTYRE
(ADA AND
GEPHYR OCAPSA
)IDES CN
ca
UJ
YREI
THU
BROU
MARTIN
BROU
KAGAWA E T AL.
CKMAN ET AL.
K C.
03
Is
ii
Cfl / - ^
8 i
5l
as
2 5
*-
en x:
JO
3D
''v.v.'
DER
1
i
UJ
e
d
150 c
UJ
<
ENELL
a.
c
GLOBOROTALIA Y
:ALID/
C A L ABRA
GLOBIGERII
cr
CD
a
O
IGER INA
5
z
NEO
o
o
a.
CD
O
BIGE RIN
<
OBI
DES
DES
CO
DES
IGER INA
CD
ION
ION
oo:
<
o
UI
CD
o
o
cr
cr
ION I
lldO\
OBIIGERI
IGERI
- -
CD
in
no-
=
o
I19
In
LID
z
Ul
OBI GERII
OBI GERI
in
9(1
8
o
in
IGERI
TIC
_J
X
UJ
BLI QUU
cr
IDA
JQlJUS
:MUS
UJ
IViOb109
-J
in
3
<
UI
BIG ERINO IDES
INFI.ATA
IA
100
BOI ROTALI A OSC ITANS
CL
<
CD
50
<
GLOBIGERIf
-7
ID
CL
z
cr
a
"ZL
LU
GITAL
NE0GL0B0CiUADR INA
right - coiling (
1 3 1
CL 200
Tj
GLOBlGERIf
CL
o
c
GLOBIGERII
in
3
LICA
til.)
h-
3
a
30DVII
o 250
in
UJ
GLOBIGERINOIDES
iL)
r
q
CALI
DIGIT ATA
UJ
RTREI
PULLENIAT INA s
i/
TLA NTIS
300
FORAMINIFERA
PAC HYDI
UJ
PLANKTIC
GLOBIGERII
AGE
COLUMNAR
SECTION
27
mO
Figure 2.12. Distribution of selected planktic foraminifera in the Vrica section. Data are from M.-L. Colalongo (in Selli et al., 1977) and Colalongo
et al. (1981).
Colalongo (1982), as well as on recent research carried out by

one of us (M.L.C.). Planktic foraminifera at Vrica have also
been studied by Nakagawa et al. (1980) and Nakagawa (1981), as
indicated in Figure 2.8, as well as by Spaak (1983), Zijderveld et
al. (1991), Sprovieri (1993), and Lourens et al. (1994).
According to Colalongo et al. (1984), among the planktic
foraminifera events recorded in the Vrica section, the most
reliable for interregional correlation and dating are the
Globorotalia inflata FA, the beginning of dominantly sinistral
coiling in Neogloboquadrina pachy derma, and the Globigerina
cariacoensis FA.
Globorotalia inflatafirstoccurs in the Vrica section between 62
and 67 m below marker bed a. Lourens et al. (1992) recorded the
first appearance of G. inflata in the upper Singa section in oxygen-
isotope stage 78, which those authors date to 2.076 Ma.

According to Sprovieri (1993), who found that fluctuations in
planktic foraminifera abundances in Mediterranean Plio-Pleistocene sequences were in phase with astronomically forced oxygenisotope stages, the G. inflata FA in all studied sequences is
coincidental with oxygen-isotope stage 80. Sprovieri affirmed that
the most reliable estimate for the age offluctuation80 is 2.13 Ma,
obtained by using the time scale of Hilgen (1991). On the other
hand, Raymo et al. (1989) found the G. inflata FA within isotope
stage 78 at DSDP sites 607 and 609, and perhaps slightly earlier at
DSDP site 552A; those three sites are located in the North
Atlantic, at latitudes of 41, 50, and 56, respectively.
The beginning of dominant sinistral coiling in Neogloboquadrina pachy derma occurs about 178 m above the base of the Vrica
28
section and is practically coincident with the top of the Olduvai

subzone as defined by Zijderveld et al. (1991) (Figure 2.18). This
event is also almost coincidental with the top of the Olduvai in
sites 650 and 651 of ODP Leg 107 (Tyrrhenian Sea) and in DSDP
sites 552, 607, 609, and 610 in the North Atlantic, whereas in
DSDP site 611 in the North Atlantic it slightly predates the top of
the Olduvai (Raymo et al., 1989; Channel et al., 1990).
According to Lourens et al. (1994), in the Vrica section the first
substantial increase in sinistrally coiled neogloboquadrinids
occurs in isotope stage 64 at an age of 1.80 Ma, very close to the
top of the main normal-polarity zone of the Olduvai, which is
dated by these authors at 1.79 Ma. In the Singa section, Lourens
et al. (1992) located that event in the same position as in the
Vrica section. Sprovieri (1993) noted that the first common
occurrence of left-coiling specimens of N. pachyderma in the
Vrica section and other Mediterranean sequences is coincident
with the abundance fluctuation linked to oxygen-isotope stage
64. According to Raymo et al. (1989), in DSDP sites 607, 609,
and 552A (North Atlantic) the first abundant occurrence of leftcoiled N. pachyderma is also found within isotope stage 64. The
beginning of dominant sinistral coiling in N. pachyderma has
been used as a criterion for recognizing the Pliocene-Pleistocene
boundary in northern Italy in the Santerno section (Colalongo,
1968), the Rio Vendina-Crostolo section (Colalongo, Cremonini, and Sartoni, 1978), and the Po plain (Dondi and Papetti,
1968), in southern Italy in the Pisticci section of Puglie (Lentini,
1971), and in other sequences.
The first appearance of Globigerina cariacoensis is in the Vrica
section about 2 m below the midpoint of marker bed /. Lourens
et al. (1994) dated the midpoints of marker beds/and e at 1.738
and 1.809 Ma, respectively. By linear interpolation between the
ages of these two sapropels, an age of about 1.744 Ma is obtained
for the G. cariacoensis FA in the Vrica section, which occurs in
this section within isotope stage 62. The G. cariacoensis FA is
recorded in sediments considered basal Pleistocene in many
other sections, such as the Capo Rossello-Punta Piccola section
and the Monte San Nicola section in southern Sicily (Rio,
Sprovieri, and Raff], 1984), the Santerno section of northern
Italy, sections in the Marche region of central Italy and sections
in the Calabria region of southern Italy according to our own
unpublished data, in DSDP site 132 of the Tyrrhenian Sea
(Colalongo et al., 1981), and in DSDP site 125 of the Ionian Sea
(Rio et al., Chapter 5, this volume).
DSDP site 132, and sections sampled in the Romagna and

Marche regions (M.L.C., unpublished data).
196 m: last occurrence of Globigerinoides obliquus extremus, an
event recorded in sediments considered to be basal Pleistocene
in age in DSDP site 132 and in the regions of Romagna, Marche,
and Calabria (M.L.C., unpublished data).
769 m: last occurrence of dextrally coiled Neogloboquadrina
atlantica, and, at 156.5 m (ca.), first occurrence of Globigerina
digitata digitata. According to R. Sprovieri (personal communication), those two events are recorded close to the PliocenePleistocene boundary, as presently understood, in the Monte San
Nicola section and in the Capo Rossello-Punta Piccola section.
According to our unpublished data (M.L.C.), these two events
are also recorded close to the Pliocene-Pleistocene boundary in
DSDP site 132 (Tyrrhenian Sea).
65 m: first appearance of Globorotalia oscitans, an event
recorded in sediments considered "upper Pliocene" in age (sensu
Colalongo and Sartoni, 1979, and Iaccarino and Salvatorini,
1982) in the Capo Rossello-Punta Piccola section, the Monte
San Nicola section and DSDP site 125 (Sprovieri, 1993), the Rio
Vendina-Crostolo section (Colalongo et al., 1978), the Marche
region (D'Onofrio, 1968), and, according to our unpublished
data (M.L.C.), DSDP site 132 and in the Calabria region.
40 m: first appearance of Globorotalia umbilicata. The appearance of this taxon is recorded in Pliocene sediments, close to the
G. inflata FA, in the Capo Rossello-Punta Piccola section, the
Monte San Nicola section, and DSDP site 125 (Sprovieri, 1993).
According to our unpublished data (M.L.C.), this relationship
also occurs in DSDP site 132 and Italian land sections in the
Romagna, Marche, and Calabria regions.
The ages for most of the aforementioned planktic foraminifera
events in the Vrica section can be obtained from the accurate
astronomical calibration of sapropels in this section made by
Lourens et al. (1994). It should be noted that several of the
planktic foraminifera events noted here had not been widely
mentioned in earlier publications on the Mediterranean successions. That possibly was because their importance for Mediterranean biostratigraphy had not been clearly recognized, even
though the species are not uncommon in most samples of the
appropriate ages.
Other planktonic foraminifera events

The following are recorded in the Vrica section at the indicated
approximate stratigraphic distances above the base of the
section:
197 m: first appearance of Globigerinoides tenellus, an event
recorded in sediments assigned to the basal Pleistocene in the
Caraffa di Catanzaro section of Calabria (Pasini, Selli, and
Colalongo, 1977a), the Capo Rossello section (Sprovieri, 1978),
the Rio Vendina-Crostolo section (Colalongo et al., 1978),
Benthic foraminifera
The benthic foraminifera assemblages of the Vrica section are
also abundant and diverse and are generally very well preserved
(D'Onofrio, 1981). The stratigraphic distributions of selected
benthic foraminifera in the Vrica section, according to D'Onofrio, are shown in Figure 2.13. Nakagawa (1981) and Nakagawa
et al. (1980) reported only the range of Hyalinea baltica among
the benthics in the Vrica section (Figure 2.8).
According to Colalongo et al. (1984), among the benthic
CO LUMNAR
SECTION
300
BENTHIC
F O R A M I N I F E R A
t
s
q
250
200
150
= |S
100
II"
29
Tiepido, Emilia (Rio et al., Chapter 5, this volume), and in

sections at Santerno (Colalongo, 1968), Monte Cassiano,
Marche (D'Onofrio, 1968), Caraffa di Catanzaro (Pasini et al.,
1977a), Capo Rossello (Sprovieri, 1978), and several other
Italian sections. Furthermore, the H. baltica FA occurs in
sediments assigned to the lower part of the Pleistocene in the Po
plain of northern Italy (Dondi and Papetti, 1968) and in the
Puglie region of southern Italy and in Sicily (Wezel, 1968).
The H. baltica FA is synchronous with the "large" Gephyrocapsa FA in the Vrica section (Figure 2.18) and in the Capo
Rossello-Punta Piccola section (Rio et al., Chapter 5, Figure
5.5, this volume). It slightly predates this nannoplankton event
in the Tiepido section and slightly postdates it in the Santerno
section (Rio et al., Chapter 5, Figure 5.5, this volume). Both of
these latter sections are characterized by very high sediment
accumulation rates.
216 m: first appearance of Bulimina etnea. This event is recorded
between marker beds m and n. It is also seen in sediments assigned
to the basal Pleistocene at Santerno (Colalongo, 1968) and at
Caraffa di Catanzaro, Calabria (Pasini et al., 1977a), and in
Sicilian sections at Agrigento (Sprovieri, 1968), Monte Navone
(Di Geronimo, 1969), and Capo Rossello (Sprovieri, 1978). The
B. etnea FA occurs immediately above the Calcidiscus macintyrei
LO in the Vrica section (Figure 2.18), but below that datum,
between it and the G. oceanica s.l. FA, in the Capo RosselloPunta Piccola section (R. Sprovieri, personal communication).
160 m: first appearance of Uvigerina bradyana. This event is

recorded immediately below marker bed d. According to
Colalongo (1968) and R. Sprovieri (personal communication),
z
^
the U. bradyana FA approximates the Pliocene-Pleistocene
boundary in the Santerno section (Romagna, northern Italy) and
i- i
in the Capo Rossello-Punta Piccola section (Sicily), respectively.
mO
In the Vrica section (Figure 2.18) and in the Capo RosselloPunta Piccola section (R. Sprovieri, unpublished data) the U.
Figure 2.13. Distribution of selected benthic foraminifera in the Vrica
bradyana FA occurs between the Discoaster brouweri LO and
section. (Adapted from d'Onofrio, 1981, courtesy of Giornale di
the Globigerina cariacoensis FA.
Geologia.)
According to Ruggieri and Sprovieri (1977), the H. baltica FA
marks the base of the "Emilian stage," which these authors
foraminifera events observed in the Vrica section, the three most consider to be the second of three chronostratigraphic units
important are located as follows (in meters above the base of the making up the Italian Lower Pleistocene. According to Rio,
section, approximately located):
Ruggieri, and Sprovieri (1982), recent advances in nannofossil
biostratigraphy allow these units to be recognized, with approxi239 m: first appearance of Hyalinea baltica. In the Vrica section, mate limits, in extra-Mediterranean regions. In view of the
this event is evident between marker beds o and/?. According to unsatisfactory exposures at this level in the Santerno River
Nakagawa (1981) and Verhallen (1991), the first occurrence of section, we have proposed (Pasini and Colalongo, 1994) to
H. baltica is between marker beds p and q (Figure 2.8). This define the boundary-stratotype and GSSP for the base of the
discrepancy is very probably due to the fact that whereas we Emilian substage at a level 2 m below marker p in the Vrica
collected 22 samples between marker beds o and/?, the Japanese section, where H. baltica is first observed.
team and Verhallen collected only a few samples from this
stratigraphic interval, and by accident did not find any H. baltica
Ostracodes
in their samples.
This first appearance of Hyalinea baltica is recorded in The ostracofauna of the Vrica section is very well preserved and
sediments considered Lower Pleistocene in age in the section at rich in species (Colalongo and Pasini, 1980a). Figure 2.14 shows
50
30
COLUMNAR
SECTION
II'
i
iiiii
a: a.
l! 1 B
j
CM!
lm
J00"
. .
Figure 2.14. Distribution of presumed autochthonous ostracodes in the Vrica section. (Adapted from Colalongo and Pasini, 1980a, courtesy of
Giornale di Geologia.)
below layer e, respectively, are important biostratigraphic

markers. In the Italian land sections, the Z. antemacella FA is
contemporaneous with or slightly predates the G. inflata FA
(G.P., unpublished data); in ODP site 654 (Tyrrhenian Sea) this
ostracode event is recorded a little above the G. inflata FA
(Colalongo et al., 1990). In addition, the M. adriatica FA has
been recognized by the present authors in Upper Pliocene
sediments of the Marche region.
No ostracode extinction events are recorded in this lower
interval except for that of Agrenocythere pliocenica, which
disappears about 14 m below layer e. That was a local event
that appears to have predated the final extinction of this species
in the Mediterranean basin, because one of us (M.L.C.) has
Ostracode events below marker bed e. Between the base of the found ostracofaunas rich in A. pliocenica (with larval stages) in
Vrica section and marker bed e, the Zabythocypris antemacella younger samples, including material from Le Castella, that
FA and the Macrocypris adriatica FA, at about 69 m and 4 m contain Globorotalia truncatulinoides excelsa - compare Colathe stratigraphic distributions of the ostracodes that we consider
to be autochthonous, and Figure 2.15 shows the distributions of
the ostracodes that we consider to be displaced from shallowwater sediments contemporaneous with the Vrica sediments
(Colalongo and Pasini, 1980a). Besides the forms listed in these
figures, numerous specimens of Krithe spp. and Parakrithe spp.
are present throughout the section.
From Figures 2.14 and 2.15 we get the impression of a sharp
change in the ostracode fauna close to marker bed e. Below this
marker bed, the fauna consists of a comparatively small number
of species that continue to the top of the section, and above this
level many first appearances are recorded.
31
COLUMNAR
SECTION
3J2I2
, t o ; t n | tfl j
2501
mo
CO CO
;
^
200}
m
o i* GO O)
750
100*
50
i
mO*
Figure 2.15. Distribution, in the Vrica section, of the ostracodes displaced from shallow-water marine sediments considered to have been
contemporaneous with the sediments of the Vrica section. (Adapted

from Colalongo and Pasini, 1980a, courtesy of Giornale di Geologia.)
ance of the boreal clam Arctica islandica in Italian shallowmarine sediments (Ruggieri, 1977b; Pelosio, Raff], and Rio,
1980; Colalongo et al., 1981). The appearance of this famous
"northern guest" in the Mediterranean has long been one of the
Ostracode events above marker bed e. The first appearance of main criteria for the beginning of the Pleistocene, and the C.
Cytheropteron testudo, considered as a "northern guest" by testudo FA has therefore also been considered as a marker for
Ruggieri (1977a, 1980), is about 9 m above the top of marker bed the base of the Pleistocene (Colalongo and Sartoni, 1977; Pelosio
e. Until the end of 1982 the C. testudo FA in Italian bathyal et al., 1980; Colalongo and Pasini, 1980a; Colalongo et al., 1981,
sediments was considered approximately coeval with the appear- 1982; Pasini and Colalongo, 1982). According to more recent
longo (1965) with Pasini et al. (1977b) - and also in samples

from the Marche region containing H. baltica (Colalongo,
Nanni, and Ricci Lucchi, 1979).
32
data, however, C. testudo first occurs in Sicily near the base of

the Upper Pliocene (i.e., Gelasian Stage of Rio, Sprovieri, and
Di Stefano, 1994).
Among the species listed in Figures 2.14 and 2.15, the
following appear in the Italian region in Lower Pleistocene
strata: Cytheropteron datum (Colalongo et al., 1979); C.
rotundatum and C. punctatum (M.L.C., unpublished data);
Buntonia textilis (Ruggieri, 1980); Procytherideis subspiralis
(G.P., unpublished data); Microcytherura nigrescens (Ruggieri,
1976); Callistocythere praecincta (Ciampo, 1976); Semicytherura
calabra and Poly cope demulderi (Ruggieri, 1980); Callistocythere
rastrifera (Ruggieri et al., 1976); Triebelina raripila and Cytheromorpha nana (Ruggieri, 1980); Leptocythere ramosa (G.P.,
unpublished data); and Semicytherura quadridentata (Ruggieri,
1976). All of these species appear in the Vrica section between 9
and 82 m above marker bed e.
In our study of the ostracodes of the Vrica section (Colalongo
and Pasini, 1980a) we erected several new species which first
appear above marker bed e. The first occurrences of some of
these species, namely Typhlocythere ovata, T. carinata, Cytheropteron pseudoalatum, Typhloeucytherura calabra, "Bythoceratina"
poligonia, Saida limbata, Cluthia praekeji, Tuberculocythere
quadrituberculata, T. batrachoides, Neocytherideis vricae, Bythocythere elliptica, Pedicythere polita, Cluthia undata, and Ruggieriella decemcostata, were later recognized by the present
authors (unpublished data) in Lower Pleistocene sediments of
different Italian regions, in some places associated with Hyalinea
baltica. In addition, we pointed out (Colalongo and Pasini,
1980a) that Cytheropteron garganicum, C. monoceros, Loxoconchidea minima, and Eucythere pubera, previously found only in
Holocene sediments, appear in the Vrica section a little above
the H. baltica FA.
Mollusks
The stratigraphic distribution of selected mollusks in the Vrica
section, according to Tampieri (in Selli et al., 1977), is shown in
Figure 2.16. Among these, only Pseudoamussium septemradiatum seems to be a true "northern guest" (e.g., Pelosio and Raffi,
1973; Ruggieri and Sprovieri, 1977). The paleoclimatic significance of presently boreal taxa in the bathyal (psychrospheric)
sequences of the Mediterranean Lower Pleistocene (such as the
Vrica section) is still an open problem, however, according to S.
Raffi (personal communication). In any case it must be pointed
out that in the Vrica section the P. septemradiatum FA slightly
postdates the Gephyrocapsa oceanica s.l. FA, which in other
Italian shallow-water sections approximates the first appearance
of Arctica islandica (Rio et al., Chapter 5, Figure 5.5, this
volume).
According to the mollusk distribution scheme of Ruggieri
(1962), the Hinia turbinellus LO and the Tunis contigua LO
seem to occur in the Italian sections close to the PliocenePleistocene boundary. Ruggieri and Sprovieri (1977) note also
that the extinction of Gimnobela brevis pliorecens occurs in Italy
in the lowermost Pleistocene.
COLUMNAR
SECTION
Figure 2.16. Distribution of selected mollusks in the Vrica section.

(Adapted from R. Tampieri, in Selli et al., 1977, courtesy of Giornale di
Geologia.)
Fish
The sapropel layers b, c, d, e, f, h, n, o, and p have yielded an

abundant ichthyofauna of generally well preserved remains
(Landini and Menesini, 1978a,b). In all, more than 700 fossil
fishes have been identified to species (Table 2.1). All species
survive to the present in the Mediterranean, except for Engraulis
encrasicholus macrocephalus and Tavania crotonensis. Of the
living forms, Cyclothone pygmaea, Chauliodus sloanei, Hygophum hygomi, Lobianchia dofleini, and Lampanyctus crocodilus
live exclusively or preferentially at depths exceeding 500 m.
Pollen
According to Accorsi, Bertolani Marchetti, and Bandini

Mazzanti (1978; Selli et al., 1977), the pollen of the Vrica section
generally shows marked prevalence of terminocrats (Figure
33
Table 2.1 Ichthyofauna of some sapropelic layers of the Vrica section

LAYERS OF ORIGIN
PRESENT GEOGR. DISTRIB.
11ST OF SPECIES
Engraulis cncrasicholus
macrocepbalus
Engraulis sp.
Cyclothone braueri
Cyclothone pygmaca
Cyclothone spp.
Maurolicus muelleri
Vinciguerria poweriae
Vinciguerria attenuata
Ichthyococcus ovatus
Argyropelecus hemigymnus
Chauliodus sloanei
Electrona rissoi
Hygophum hygomi
Hygophum benoiti
Lobiancbia dofleini
Lampanyctus crocodilus
Lampanyctus pusillus
Ceratoscopelus madercnsis
Diapbus rafinesquei
Diaphus holti
Benthosema glaciate
Belone belone
55
13 8
10 11
23 10
16 6
3
6
2
13
2
34 16
32 6
39 14
11 12
3
3
9
3
1
5
3
6
3
18 5
15 6
42 17
41 6
1
1
1
11
2
1
1
5
1
1
2
2
1
8
3
1
1
2
1
1
16
1
38 10
1 7
3 2
2 1
3
1
1
1
1
1
Micromesistius poutassou
Gadiculus argenteus argenteus
Microichthys coccoi
Stephanolepis sp.
Tavania crotonensis
Total
sum
63
1
108
108
173
100
3
1
1
17
2
9
3
5
69
13*
14
9
3
1
1
2
1
3
1
I
Med. Med. Atl. Atl.

E
W NE NW
44-
44-
444-
44444-
4
4
4444-
4444
+
44444-
44-f44-f
44444444
4444-
4-
4-
+
+
Atl.
SE
Atl. Pacif. Ind.

SW
4-
4-
4-
4-
+
44444-
+ 44-
444-
4-
4,
4-
4
_
>
4-
4-
4- +
.
4-
4-
Symbols: (+) species of wide geographic distribution; (-) species of restricted geographic distribution; (?) uncertain
presence.
Source: Adapted from Landini and Menesini (1978a,b), courtesy of Societa di Paleontologia Italiana.
2.17). Among these, Pinus (as P. diploxylon and P. haploxylon) beds / and h should correspond to renewed cooling. The pollen
predominates, in comparison with other genera of mountain diagram between marker bed h and the top of the section
conifers (e.g., Picea, Cedrusy Tsuga, Podocarpus). Mediocrats indicates a cool climate, with minor fluctuations.
are represented mainly by Caryay Pterocarya, Corylus, Quercus, According to Nakagawa et al. (1980; Chapter 3, this volume),
Ulmus, Zelkova, Carpinus, Tilia, Castanea, Liriodendron, and the pollen show abundant conifers, especially Pinus, throughout
Liquidambar. Taxodiaceae (Taxodium type) are present only in the section (Figure 2.8). Pollen of broadleaf genera range from
some intervals and in small percentages; they probably corre- 10% to 30%. Those authors comment that the "paleoclimate is
spond to coastal forests. Herbaceous taxa are neither frequent considered to be moderate to cool temperate, and a gradual
nor significant. According to the aforementioned authors and to cooling is suggested by the upward increase of Abies, Picea and
C. A. Accorsi (personal communication), the pollen diagram Tsuga, and decrease of Podocarpus, Taxodiaceae, Carpinus,
indicates a cool climate for the time interval represented by the Juglans, Pterocarya and Ericaceae." In particular, according to
sediments between the base of the section and marker bed a. those authors, it is possible to recognize unstable forest around
Between marker bed a and marker bed /, mediocrat pollen marker bed m and the beginning of a slight climatic deterioration
species are in general much more abundant, and terminocrats above that level.
scarcer, suggesting a milder climate. The sharp increase in
Combourieu-Nebout, Semah, and Djubiantono (1990) carried
terminocrats and the decrease in mediocrats between marker out detailed pollen analyses of 60 samples from the Vrica
34
- r
q
I
S
p
c
E
Figure 2.17. Synthesis of pollen spectra from the Vrica section, with lithology to left and
pollen diagrams to right: Gr.
1, Taxodiaceae, Engelhardtia,
Palmae, etc.; Gr. 2, Cathaia;
Gr. 3, QuercuSy Carya, UlmusZelkova, Carpinus, etc; Gr. 4,
Pinus and indeterminable
Abietaceae; Gr. 5, Tsuga; Gr.
6, Cedrus; Gr. 7, Abies and
Picea; Gr. 8, taxa not classified; Gr. 9, Mediterranean
xerophytes (Olea, Phillyrea,
Pistacia, etc.); Gr. 10, openvegetation herbaceous plants;
Gr. 11, Artemisia and
Ephedra. (From CombourieuNebout et a!., 1990, with permission of Academie des Sciences de Paris.)
* V * W
h
f
N
E
ml TTi
30-03=31
--
r-1 r~r
1 1
PTffT]
i'
* 1
C
E
N
J|
->
^ ^
10 K^
"-"
, J-
1 10 m
percentages calculated on the total of the pollen
CHDma,
^-i samid
taminites
! * * ! volcanic ash
section, 5 of which were collected from the crucial interval

between marker beds e and/(Figure 2.17). According to those
authors, in the lower part of the sequence (from the bottom to
sample 20, located about 15 m below marker bed a) there are
high percentages of altitudinal elements (Tsuga, Cedrus, Abies,
and Picea) and of herbaceous taxa. This association indicates a
relatively cool period, which on the basis of biostratigraphic and
chronostratigraphic data those authors ascribed to the "Praetiglian glacial phase" typified in northern Europe. Above sample
20, herbaceous taxa decrease, and Cathaia and Taxodiaceae
begin to increase. From marker a to marker e, a long period
dominated by forest in a subtropical to temperate-warm climate
is indicated by relatively high proportions of Taxodiaceae, Engelhardtia, Palmae, Cathaia, Quercus, Carya, Ulmus-Zelkova, and
Carpinus. This interval has been correlated by CombourieuNebout et al. (1990) to the "Tiglian interglacial phase." From
marker bed e and continuing to the top of the section, the pollen
spectra recorded by those authors show a steady increase in herbaceous taxa (especially Artemisia) and a corresponding decrease in forest taxa, with progressive enrichment of altitudinal
trees. This interval, which is considered by Combourieu-Nebout
et al. (1990) to have been a period of cool and xeric climate, is
ascribed to the "Eburonian glacial phase." The most important
WF1 Gr.l
B Gr.7
r^
O
Gr.2 ^
Gr.3
Gr.8 flUD Gr.9
percentages calculated excluding Pmus from the total
EEB Gr 4 C D Gr.5 ffl|

C D Cr-0 [ 2 Gr.l I
Gr.6
point in that detailed research is that the major floristic shift that
those authors identified as the transition from the "Tiglian interglacial phase" to the "Eburonian glacial phase" corresponds to
the Pleistocene boundary-stratotype at the top of marker bed e.
K/Ar and fission-track dating
The volcanic ash layer La.

This layer, 20 cm thick, crops out about 65 m above the bottom
of the component-section JD of the Japanese team (Nakagawa et
al., Chapter 3, this volume) and at least 75 m below the base of
the Vrica section as defined here (Figure 2.3). This layer is
grayish in color and is characterized by the presence of
hornblende, plagioclase, zircon, and glass shards (Obradovich et
al.,1982).
Foraminifera assemblages examined by one of us (M.L.C.)
from the siltstone immediately below and above the La. layer
are typical of the Globorotalia crassaformis subzone (i.e., upper
part of the Globorotalia ex gr. G. crassaformis zone), which
corresponds to the lower portion of the Upper Pliocene (i.e.,
Gelasian Stage of Rio et al., 1994). In addition, the LO of
dominant left-coiled Neogloboquadrina atlantica is found about
9 m below the La. layer. Zijderveld et al. (1991) noted the

LO of N. atlantica (which, in our opinion, means left-coiled
N. atlantica) in the Singa section at a level corresponding to
2.41 Ma.
Separated fractions from the La. horizon were dated by
Obradovich et al. (1982) to 2.22 0.03 Ma (K/Ar, hornblende),
2.0 0.16 Ma (fission-track, zircon), and 2.2 0.2 Ma (fissiontrack, glass). Obradovich et al. (1982) also demonstrated that the
age estimates previously obtained for this volcanic ash of 3.1 and
3.4 Ma, reported by Selli (1970), were incorrect.
35
Magnetostratigraphy
In 1977, H. Nakagawa presented a preliminary version of the

magnetostratigraphy of the Stuni-Vrica sequence at the joint
meeting of INQUA Subcommission 1-a and the IGCP-41
working group in Birmingham, and later at the NeogeneQuaternary boundary symposium in Dushanbe (Nakagawa et
al., 1980). The Japanese team subsequently reported revised
polarity sequences (Nakagawa, 1981, 1982), and other measurements on the Stuni-Vrica sequence were published by Tauxe et
al. (1983). Nakagawa et al. (Chapter 3, this volume) repeated
their paleomagnetic sampling and analysis of the Stuni-Vrica
Volcanic ash layer m
sequence with new equipment for their final version of the
magnetostratigraphy of this sequence.
This layer (Figure 2.3) is 2-7 cm thick and crops out about 6 m
The biostratigraphy of that part of the Stuni-Vrica sequence
above the top of sapropelic layer h. The m ash was first dated at
stratigraphically below the Vrica section, as defined by Selli et al.
2.07 0.33 Ma by Bigazzi and Bonadonna (in Selli et al., 1977),
(1977) and in this chapter, has not yet been studied in sufficient
at 2.2 0.2 Ma (K/Ar, hornblende) by Savelli and Mezzetti
detail to evaluate the physical continuity of this part of the
(1977) and Savelli (in Selli et al., 1977), and at 2.5 0.1 Ma (K/
section. For this reason, with regard to the data presented by
Ar, glass) by Boellstorff (1977). Obradovich et al. (1982)
Nakagawa and co-workers, we have taken into account only the
demonstrated that all of those age analyses were erroneous
paleomagnetic information that is inarguably within the Vrica
because of misidentification of samples or because of the
section, in sections JA and JB (Figures 2.3 and 2.18).
unsuitability of the material used for dating. In re-dating the m
Nakagawa et al. (Chapter 3, this volume) correlated the
ash, Obradovich et al. (1982) found two distinct fractions of
normal-polarity zone of the lower part of the Vrica section,
older detrital biotite mixed with the primary volcanogenic
including two short reversed-polarity intercalations separated
biotite. The amount of detrital biotite ranged from 0.91% to
by a short intermediate-polarity interval (Figure 2.18), with
4.72% of total biotite, related inversely to the grain size, so that
the Olduvai subchron. Their correlation is supported by the
the coarsest fraction yielded the youngest K/Ar age: 1.99 0.08
planktic microfossil events Gephyrocapsa caribbeanica FA,
Ma. Even the coarse fraction, however, was contaminated with
Discoaster brouweri LO, Globigerinoides obliquus LO, Gephyreworked detrital biotite, and it can be concluded that the age of
rocapsa oceanica FA, and Calcidiscus macintyrei LO in or near
the m volcanic ash is certainly younger than 1.99 Ma. The m ash
the normal-polarity zone of the Vrica section. These are
cannot be dated by fission-track analysis because the glass shards
events that are recorded in or near the Olduvai subchron in
are unsuitable and uranium-retentive minerals such as zircon and
deep-sea sections, and to some extent in Japanese land
apatite are not present. It might be possible to obtain a more
sections of the Boso Peninsula (Itihara et al., Chapter 24, this
precise age determination for this ash by analyzing a small
volume).
sample of hand-picked large euhedral grains in a more sensitive
Tauxe et al. (1983) studied almost all the Vrica section, but
system (J. D. Obradovich, personal communication, 1993).
they omitted the lower part of component-section A because that
part was synthesized from scattered outcrops, and they discarded
all samples above marker bed s because they were judged to be
Pumice block
unsuitable for paleomagnetic polarity measurements. According
A whole-rock K/Ar age of 2.0 0.1 Ma was obtained by Savelli to Tauxe et al. (1983), two normal-polarity zones are recognizand Mezzetti (1977; Selli et al., 1977) on the pumice block found able in the Vrica section (Figure 2.18). The lower zone (N1-N2),
about 1 m above volcanic ash layer m. It must be remembered interrupted by a short reversed-polarity intercalation, is located
that whole-rock K/Ar analyses of glassy materials such as in the lower part of the section. The younger normal-polarity
pumice, especially those exposed to marine conditions, are zone (N3) was observed immediately below marker bed s. Tauxe
highly unreliable (Dalrymple and Lanphere, 1969). Subse- et al. (1983) correlated the N1-N2 normal-polarity interval with
quently, however, Obradovich et al. (1982) obtained a mineral the Olduvai subchron because it lies between the Discoaster
K/Ar age of 2.35 0.16 Ma for hornblende from the pumice brouweri LO at its base and the Calcidiscus macintyrei LO above
its top, as discussed earlier (Figures 2.10 and 2.18). With regard
block.
As noted earlier, the mineralogy and chemistry, as well as the to the normal-polarity zone N3, the top of the Vrica section does
K/Ar age, of the detached pumice block differ conspicuously not extend to levels as young as the lower boundary of the zone
from those of the autochthonous m ash. Obradovich et al. (1982) of "small" Gephyrocapsa, which is below the base of the
interpret this "as indicating that the pumice block has been Jaramillo subzone (Rio et al., Chapter 5, this volume) (Figures
reworked from an older pumice layer and thus has no relevance 2.2-2.4). The top of the Vrica section is therefore older than the
Jaramillo, and the N3 zone, if confirmed, must represent a
to the age of the m marker bed."
MAGNETOSTRATIGRAPHY
VRICA
DSTRACODA MOLLUSCA According to Tauxe et al. (1983 )
F0RAM.
F0RAMINIFE RA
NANNOPLANKTON
SECTION
BENTHIC
P L A N KT1 C
CALCAREOUS
300
cr
a
LU
TATUM
ETNE A
D
CD
TF
1 1 1
PSE
_L
no
Undefined
m0
OBLIOL US
RRIS
E X T f'EMU
ACOST
D
ERINO IDES
HEL CO SPHAE
SCUS
5T E^R
>TER
CALC
DISC
VO
o 9
DISC
50
CL
SELLII
BRO U W E R I
o S
o o
z o
SON*
ar. T R I R A CJ I A T U
p
I
<
<
H>
I*j
1
I=>
_J
JELL
100
X
// \\
LU
<
Reversed
SUBCHRONS
P L I C)RECE
150
REV
Normal
o
o
o
o
rpr 6 / Pleistoce 76 bOlin dsry - stfL tnt\stna
LU
SHORT NORMAL
EVENT BETWEEN
THE OLDUVAI
AND THE JARA -
CYTHER
POLARITY
TENELL
AR A C O E N
DA
ALIC
OBIGERINA
\i
h>
9-
g
<
z
I
12
<
z
or
.OBIGERINA
200
z
&
Intermediate
z
-J
JTER
BBEAN
ANICA
a.
*-
RTlCULlNA
LU
OBIGERINA
layer
Si It y marly
claystone
o
n
TATA
hCO
D
<
-OBIGERINA
250
7 1 Sapropel clay
E P H Y R OCA
L>
Sapropehc
layers
OAMUS SlUK
<
ALA
Sandy layer
LU
pr
RON
Volcanic ash layer
Accordig to
Accordig to
Nakagawa
ZijdervekJ et al 1991
( 1983 pers. com.)
and to Nagakava
et al. (in this vol.)
ADIAT
LITHOLOGY
(0
Figure 2.18. Distribution, in the Vrica section, of the most significant planktic and benthic
organisms from the biostratigraphic and biochronologic points of view. To the right, magnetostratigraphy of the Vrica section according to Tauxe et al. (1983), and according to H.
Nakagawa (personal communication, 1983) and Nakagawa et al. (Chapter 3, this volume).
Regarding the latter paleomagnetic log, we have considered only the information from
component-sections JA and JB (Figure 2.3).
normal-polarity interval between the Olduvai and the Jaramillo

subchrons.
Zijderveld et al. (1991) carried out a closely spaced
paleomagnetic sampling in the Vrica section, with a sampling
density much higher than that of Tauxe et al. (1983), by use of a
motorized corer to collect unweathered samples. The magnetostratigraphy of the Vrica section proposed by those authors is
shown in Figure 2.18. According to Zijderveld et al. (1991), a 10m-thick re versed-polarity interval, designated here as interval (3,
includes the Pleistocene boundary-stratotype. Interval (3 is
located above the 45-m-thick normal-polarity interval a, and
below the 5-m-thick normal-polarity interval y, also as designated here. After reviewing the subject in detail, Zijderveld et
al. (1991, p. 711) concluded that the base of the normal-polarity
interval a represents the lower boundary of the Olduvai
subchron, and the top of the normal-polarity interval y above
sapropel e "most probably represents the upper Olduvai polarity
transition as generally found elsewhere." That interpretation
places the short reversed-polarity interval (3 in the top part of the
Olduvai subzone. Zijderveld et al. (1991) did not find the
normal-polarity zone N3 reported by Tauxe et al. (1983) below
marker s at the top of the Vrica section, nor in the fresh rocks of
the parallel Crotone section; according to Zijderveld et al. (1991,
p. 712), "given the very weathered state of the top part of the
Vrica section these normal polarities are almost certainly due to
secondary magnetizations related to the weathering and do not
represent an extra normal subchron."
The presence of a reversed-polarity interval within the upper
Olduvai had previously been recognized in the Boso Peninsula in
central Japan (Nakagawa et al., 1982), in core V20-109 from the
northern Pacific Ocean (Ninkovitch et al., 1966), in several
deep-sea cores of the Indian Ocean (Opdyke and Glass, 1969), in
loessic deposits on the Chinese Loess Plateau (Heller et al.,
1991), and in other sequences.
Adoption of the Vrica section
The Vrica section was first described in the guidebook for the
symposium on the Neogene/Quaternary boundary in Bologna
and Crotone in October 1975 (Pasini et al., 1975). The members
of the IGCP Project 41 working group, after visiting the main
Plio-Pleistocene sections of Calabria, recognized that "there are
some difficulties in the interpretation of the Santa Maria di
Catanzaro and Le Castella sections as [Pleistocene] boundary
stratotype [sections]" and that the Vrica section represented a
potential Pleistocene boundary-stratotype section (Selli and
Cati, 1977, pp. 29-30).
The results of further research on the Vrica section were
presented in 1977 by one of us (G.P.) with C. Savelli and R. Selli
(in Selli et al., 1977) and by H. Nakagawa at the joint meeting of
INQUA Subcommission 1-a and the IGCP Project 41 working
group during the X INQUA Congress in Birmingham. At the
close of that joint meeting, a resolution was passed, emphasizing
that "the participants of the Meeting support the suggestion to
37
take the Vrica Section (Crotone in Italy) as the Neogene/

Quaternary boundary stratotype [section], and find it necessary
to further a detailed complex study of this section."
At the next joint meeting of the IGCP-41 working group and
the INQUA Subcommission on the Pliocene-Pleistocene boundary, held in Chandigarh and Srinagar (India) in OctoberNovember 1979 (Sastry et al., 1981), it was concluded that in view
of the resolution adopted by the 1948 International Geological
Congress in London, a physical reference point for the PliocenePleistocene boundary should be chosen in the marine sequences
of southern Italy, and that "the Vrica Section in Calabria, Italy,
described by Selli et al. (1977) appears more suitable [than the
Santa Maria di Catanzaro and Le Castella sections for defining
the Pleistocene boundary-stratotype] because it meets the
general criteria enumerated above," and furthermore that "the
works of Selli et al. [1977], Nakagawa et al. [1980], and Arias et
al. [1980] show that the Vrica deposits contain many elements of
value for wide range correlation."
Essentially the same conclusions were reaffirmed in subsequent meetings, as further work on the Vrica section continued
to demonstrate its suitability for the boundary-stratotype. One of
us (G.P.) presented a report describing and correlating the
principal Plio-Pleistocene sections of Calabria (including the
Vrica section) and DSDP site 132 (Tyrrhenian Sea) at the XXVI
International Geological Congress, Paris, 1980 (Colalongo et al.,
1981). At a joint meeting of INQUA Subcommission 1-a and the
IGCP-41 working group during this congress, the propositions
from that report were accepted, and it was resolved that "the N/
Q [Neogene/Quaternary] boundary [stratotype] should be placed
in the Vrica Section." Again, at the joint international field
conference of IGCP Project 41 ("Neogene/Quaternary Boundary") and Project 128 ("Late Cenozoic Magnetostratigraphy") in
Arizona and California, March-April 1981, a resolution was
passed that "the Vrica Section in Calabria is the most suitable
candidate for the P/P [Pliocene-Pleistocene] boundary stratotype," and that "paleontologic information already available
from Vrica offers a sound basis for world-wide correlations in
marine sections."
At the next joint meeting of the INQUA Subcommission 1-a
and of the IGCP-41 working group at the XI INQUA Congress,
in Moscow, August 1982, the members of these groups and
outside specialists conducted a lengthy and detailed discussion
of the Pliocene-Pleistocene boundary problem and the biostratigraphy, biochronology, magnetostratigraphy, paleoclimatology, radiometric ages, and sedimentology of the Vrica
section. At the end of the meeting, an overwhelming majority
of those present approved a formal proposal in favor of the
Vrica stratotype.
Thefinalwording of the proposal of the ICS working group on
the Pliocene-Pleistocene boundary concerning the definition of
the Pliocene-Pleistocene boundary-stratotype, as described in
this volume, was edited during a joint meeting of the ICS
working group on the Pliocene-Pleistocene boundary and the
IGCP-41 working group in Madrid, May 23, 1983.
38
Location of the Pleistocene boundary-stratotype in the

Vrica section
As is well known, the proposal by the temporary commission to

advise on the question of the definition of the PliocenePleistocene boundary, which was accepted as read by the council
of the XVIII International Geological Congress in London,
1948, emphasized that the Pliocene-Pleistocene boundarystratotype should be placed not only to coincide with the base of
the Italian Calabrian Stage but also "at the horizon of the first
indication of climatic deterioration in the Italian Neogene
succession" and "should be based on changes in marine faunas."
Given the state of knowledge at the time and the composition of
the temporary commission, there can be no question that the "first
indication of climatic deterioration" must be understood as an
allusion to the entrance of the first of the well-known "northern
guests" into Mediterranean faunas. This is made doubly clear in
the follow-up report organized by the Italian Geological Society at
the XIX International Geological Congress, in Algiers, 1952,
which was requested by the council of the London congress in
order to select "a type locality for the precise definition of the
boundary." In that report, four sections were described: the
Monte Mario section near Rome (Blanc, Tongiorgi, and Trevisan,
1954), the Castell'Arquato section near Piacenza (Di Napoli,
1954), the Santerno section near Imola (Ruggieri, 1954), and the
Val Musone section near Ancona (Selli, 1954); see also Azzaroli et
al. (Chapter 11, this volume). In that report, the PliocenePleistocene boundary was placed to coincide in each section with
the first occurrence of "northern guests," and each section was
considered to be an example of the Calabrian Stage. It is therefore
the appearance of the first "northern guest" in the Italian PlioPleistocene sections that is the historical criterion used to mark the
Pliocene-Pleistocene boundary, in conformity with the intention
of the London 1948 resolution.
In 1977, proposals for the location of the Pliocene-Pleistocene
boundary-stratotype in the Vrica section were first advanced.
Selli et al. (1977) proposed to locate the Pliocene-Pleistocene
boundary within "Unit Y" of the Vrica section, an informal
biostratigraphic unit proposed by Selli et al. (1977) that extends
from the Cytheropteron testudo FA (between marker beds e and
f) to the Hyalinea baltica FA (between marker beds o and p).
Accordingly, various authors, including Colalongo and Sartoni
(1977), Nakagawa et al. (1980), Colalongo and Pasini (1977a,
1980a,b), Pelosio et al. (1980), Colalongo et al. (1980, 1981,
1982), and Nakagawa (1981), have proposed that the Pleistocene
boundary-stratotype should be equated with different horizons
located between marker beds e and m of the Vrica section.
In regard to these proposals, Colalongo and Pasini (1980a) and
Colalongo et al. (1980) recommended that the Pleistocene
boundary-stratotype be placed, in any event, within the e-m
interval because that is where the most pronounced, if not
precisely coincident, faunal changes in both planktic and benthic
microfossils are recorded. That recommendation was approved
at the joint meeting of IGCP-41 and INQUA Subcommission 1-a
at the 1980 Paris International Geological Congress.
At that point, the problem was to select a physical horizon

within the e-m interval that could serve as the definitive
Pleistocene boundary-stratotype. The 1982 joint meeting of
INQUA Subcommission 1-a and IGCP-41 in Moscow recommended the physical horizon immediately below the first
occurrence of the cold-adapted ostracode Cytheropteron testudo
(between marker beds e and /), but, as noted earlier, that was
soon rendered unacceptable by the discovery that this ostracode
also was present in the Mediterranean basin during the middle
Pliocene.
Summary of the final proposal
After discussion with members of IGCP-41 and preliminary
approval at the 1968 Moscow INQUA Congress, in 1983 the ICS
Working Group on the Pliocene/Pleistocene Boundary presented
its final report to the parent ICS Subcommission on Quaternary
Stratigraphy, recommending "the base of the claystone conformably overlying sapropelic marker bed e of the Vrica section . . . as
the Pliocene!Pleistocene boundary-stratotype."
The document, "Proposal of the ICS Working Group on the
Pliocene/Pleistocene Boundary," prepared in 1983 by E. Aguirre
and G. Pasini as chairman and secretary, respectively, of the
IGCP-41 working group, made three important points in support
of the Vrica section as the boundary-stratotype section for the
base of the Pleistocene epoch. Those points, which are even
more pertinent in view of more recent developments, are
reviewed and amplified as follows:
Point 1. The base of the claystone overlying marker bed e of the
Vrica section is an appropriate location for the boundary. In fact,
the base of the claystone unit is close to calcareous microplankton events such as the Gephyrocapsa oceanica s.l. FA,
Globigerinoides obliquus extremus LO, the beginning of dominant left-coiled Neogloboquadrina pachyderma, and the Globigerina cariacoensis FA, which in other Italian sections (e.g.,
Santerno, Stirone) have been identified close to the first
appearance of the shallow-water mollusk Arctica islandica
(Colalongo, 1968; Pelosio et al., 1980; Azzaroli et al., Chapter
11, this volume; Rio et al., Chapter 5, this volume). In
particular, Rio et al. (Chapter 5, this volume) state that "the
stratigraphically lowest level where A. islandica is represented
seems to be in the Castell'Arquato and the Stirone sections. No
reliable nannofossil data are available from the former section.
In the Stirone section [near Parma], we have observed that the
appearance of A. islandica slightly predates the first occurrence
of G. oceanica s.l. . . . close to the top of the Olduvai subchron."
In the Vrica section, the base of the claystone overlying marker
bed e is similarly positioned, being only 99 k.y. older than the
first occurrence of G. oceanica s.l. ("medium-sized" Gephyrocapsa of Lourens et al., 1994) and close to the top of the Olduvai
subzone (Figure 2.18). We may consider that the base of the
claystone overlying marker bed e is penecontemporaneous with
the first appearance of A. islandica in Italy.
It is well known that the entrance of A. islandica, the most
famous among the earliest "northern guests," into the Mediterranean has historically been the main criterion for recognizing the
beginning of the Pleistocene in Italy. Thus, the designation of the
base of the claystone overlying layer e as the Pleistocene
boundary-stratotype places the boundary very close to its
traditional statigraphic position, thereby avoiding serious upset
of the geological literature and maps, in accordance with the
recommendations of the International Stratigraphic Guide (Hedberg, 1976; Salvador, 1994). Furthermore, the paleomagnetic
reversal at the top of the Olduvai subzone, very close to bed e, is
clearly associated with evidence of major climatic deterioration
in various stratigraphic sections from around the world (as
exemplified by reports in this volume).
39
Point 3. Marker bed e and its contact with the overlying claystone
are very well exposed in the Vrica section and in the surrounding
area, and abundant fossils are present to facilitate correlation,
again in accordance with the recommendations of the International Stratigrahic Guide (Hedberg, 1976; Salvador, 1994).
Approval of the final proposal by the ICS and the
IUGS
Thefinalversion of the document, "Proposal of the ICS Working

Group on the Pliocene-Pleistocene Boundary," was presented
for discussion at the ICS business meeting in Moscow on August
13, 1984. In a postal ballot, the 25 voting members of the
commission approved the proposal by a vote of 20 to 1, with 4
abstentions. Following a report on behalf of the ICS by M. G.
Point 2. Multiple, reinforcing criteria offer a sound basis for Bassett and J. W. Cowie to a meeting of the full IUGS Executive
worldwide correlation of the proposed boundary horizon. The Committee in Rabat, Morocco, on February 10, 1985, the
main calcareous nannofossil events associated with the base of proposal was submitted in writing to the IUGS Executive for a
the claystone bed overlying marker bed e in the Vrica section, postal ballot. On May 31, 1985, the secretary-general of IUGS,
such as the Discoaster brouweri LO, Gephyrocapsa oceanica s.l. R. Sinding-Larsen, informed the ICS that the proposal had
FA, and Calcidiscus macintyrei LO (Figure 2.18), are found in received majority support from the IUGS Executive. That
the same order and in the same position with respect to the announcement of support represented formal ratification of the
Olduvai subzone in oceanic deep-sea sediments as well (Rio, proposal by the IUGS (Bassett, 1985, p. 91).
1982; Backman and Shackleton, 1983; Backman et al., 1983; Rio
et al., Chapter 5, this volume). Similarly, the base of the
Dating the boundary
claystone bed overlying marker bed e is close to planktic
foraminifera events such as the Globigerinoides obliquus In order to calculate the age of the Pleistocene boundaryextremus LO and the beginning of dominant left-coiling in stratotype, we take into account only the astronomically caliNeogloboquadrina p achy derma (Figure 2.18) that have been brated ages of the two boundaries of the Olduvai subzone. In the
widely used in long-range correlations. Pollen analysis of the time scale of Shackleton et al. (1990), the top of the Olduvai
Vrica section (Combourieu-Nebout et al., 1990) demonstrates a subchron is placed in isotope stage 63, dated to 1.77 Ma, and the
clear correlation with the major Plio-Pleistocene paleoclimatic lower boundary of the subchron in isotope stage 71, dated to 1.95
phases in the continental environments of Europe and, by Ma. According to Hilgen (1991), the top of the Olduvai falls
extension, other continental regions. In terms of magne- within stage 64 and is dated to 1.79 Ma, and the base is dated to
tostratigraphy, the base of the claystone overlying marker bed e 1.95 Ma. Tiedemann, Sarnthein, and Shackleton (1994) reis a little below the top of the Olduvai subzone, and in high- corded the upper boundary of the Olduvai in isotope stage 63,
resolution sections can be even more precisely correlated within dated to 1.78 Ma, and the lower boundary in stage 71, dated to
the thin, distinctive re versed-polarity interval recognized in the 1.94 Ma. Lourens et al. (1994) located the upper boundary
Vrica section by Zijderveld et al. (1991) (Figure 2.18). As for the within oxygen-isotope stage 64, dated to 1.79 Ma, and the base
stable isotopes, according to Lourens et al. (1994), the Pleisto- within stage 72, dated to 1.94 Ma. In the time scale of Shackleton
cene boundary-stratotype is located at oxygen-isotope stage et al. (1995a), the upper and lower boundaries of the Olduvai
boundary 65/64. This determination unambiguously positions the subchron are dated to 1.770 Ma and 1.950 Ma, respectively. The
boundary-stratotype with respect to the precisely calibrated age values of Shackleton et al. (1995a) were also adopted by
record of orbitally forced climatic variation that has now been Cande and Kent (1995) in the new GPTS (geomagnetic polarity
observed in all the world's oceans and is being extended to many time scale). We have taken into account the magnetostratigraphy
land sections.
of Zijderveld et al. (1991) (Figure 2.18) and have calculated the
Considering these data, the proposed Pleistocene boundary- age of the base of the claystone conformably overlying marker
stratotype can be easily identified on micropaleontologi- bed e, the Pleistocene boundary-stratotype, by linear interpolacal, palynological, magnetostratigraphic, and cyclostratigraphic tion between the ages of the bottom and the top of the Olduvai
grounds in both marine and nonmarine Plio-Pleistocene se- subzone (Hilgen, 1991). With reference to the ages of these
quences. Consequently, the base of the claystone overlying boundaries proposed by Shackleton et al. (1990), Hilgen (1991),
marker bed e satisfies the basic requirement for a chronostrati- Tiedemann et al. (1994), Lourens et al. (1994), and Shackleton
graphic boundary, that is, its suitability for worldwide recogni- et al. (1995a), we obtain for the Pleistocene boundary-stratotype
tion, as recommended in the International Stratigraphic Guide the respective ages of 1.796, 1.813, 1.803, 1.811, and 1.796 Ma;
the last value is the same as in the revised GPTS of Cande and
(Hedberg, 1976; Salvador, 1994).
40
Figure 2.19. Field-guide itineraries for component-sections A and

C of the Vrica section (Figures
2.2 and 2.3), indicated by arrows
marked with A and C, respectively; 1, roads suitable for normal cars; 2, impassable to cars;
3, the component-sections. (Reproduced with permission of the
Istituto Geografico Militare, authorization no. 2568 of April 1,
1987. Dai tipi dell'Istituto
Geografico Militare, autorizzazione n. 2568 in data 1/4/1987.)
Kent (1995). All of these ages are the same, at 1.8 Ma, when
rounded up to one decimal place.
Appendix: accessibility of the Vrica section
The Vrica section is located about 4 km south of the town of

Crotone (Figures 2.1 and 2.19).
The component-sections A, B, and C in the Vrica section
(Figures 2.2 and 2.3) can be easily reached by automobile from
Crotone. Good boots are the only climbing equipment required
to traverse the exposures. The outcrops of the Vrica section are
being granted the status of nature preserves, with legal protection against intentional damage.
Component-section A (Figure 2.19)
This section can be reached from the parking lot at the Costa
Tiziana Hotel, 3 km south of Crotone. This hotel is located a few
km 0
0.5
tens of meters west (inland) of the coastal road, via a short

driveway.
The base of component-section A is located about 250 m south
of the hotel, at the confluence of two normally dry ravines. The
trail from the hotel to this point is not well marked but may be
followed easily with the aid of the map shown in Figure 2.19.
The traverse from the base of component-section A to the
"sapropel-clay layer" (Figure 2.3) follows up the valley that
extends southward from this confluence. The stratigraphy of this
lower part, measuring about 45 m in thickness, has been
reconstructed by geometric and lithologic correlations of scattered outcrops on both sides of the valley. About 100 m southsouthwest of the base of the section is a small earthen checkdam, and some 200 m farther, on the west side of the seasonally
flooded plain behind the dam, is the mouth of an east-west gully
(Figure 2.5). Just at the mouth of the gully, about 2 m above the
plain, a clearly laminated shale layer about 30 cm thick crops
41
Figure 2.20. Field-guide itinerary

for component-section B of the
Vrica section (Figures 2.2 and 2.3),
indicated by arrows marked with
B. The road from Crotone to the
Costa Tiziana Hotel is shown in Figure 2.19; symbols as in Figure 2.19.
(Reproduced with permission of the
Istituto Geografico Militare, authorization no. 2568 of April 1,
1987. Dai tipi dell'Istituto
Geografico Militare, autorizzazione n. 2568 in data 1/4/1987.)
out. This is the "sapropel-clay layer," 45 m above the base of the

Vrica section (Figure 2.3). This shale layer is normally deeply
weathered and not conspicuous, but it can easily be located with
the help of a rock hammer.
From this point, the section is followed westward up the side
gully for several tens of meters. The first conspicuous sapropelic
shale layer, cropping out near the head of the gully, is marker
bed a (Figure 2.3).
By climbing to the ridge above the gully (Figure 2.5) and
following the ridge line to the southwest for a few tens of
meters, the sapropelic marker beds b, c, d, and e can
be recognized according to the thicknesses of the respective
layers and the interbedded claystones, as described previously
(Figure 2.3).
Component-section B (Figure 2.20)
This section may be reached by continuing past the mouth of the
east-west gully in which the upper portion of section A crops out
and proceeding south-southwest up the dry ravine that feeds into
the impounded plain.
An easier way, however (Figure 2.20), is to continue driving
past the Costa Tiziana Hotel and Ca Donato on the coastal road,
to a crossroads about 8 km from Crotone. There, turn right and

drive past Ca Santo Spirito and Campione, and about 2.5 km
past Campione turn right again and go about 1.5 km to the end of
the country road that crosses the locality Parasinaci. Park and
walk northwest along the northern or northeastern margin of
marine terrace remnants. About 1 km northwest of the end of
the country road a northeast-trending ravine cuts sharply into the
terrace surface; the cliff on the opposite side of this ravine
exposes component-section B (Figure 2.6). To reach the base of
component-section B, go downhill (north-northeast) along a
gentle slope, as far as the normally dry bed of the ravine.
Marker bed a, at the base of component-section B (Figure
2.3), crops out in the ravine near a curve about 40 m downstream
from a major confluence. In some seasons it may be covered by
alluvium and will not be easy to find. All the other marker beds
of component-section B, however, are always clearly recognizable and can be identified (as outlined in the earlier section
"Sedimentology and paleoecology") by their thicknesses and
internal spacing (Figure 2.3). To traverse component-section B
starting from the base at marker a, follow the ravine upstream to
the point where it crosses the tuff layer of marker bed m. From
this point onward, the section must be traversed on the steep
western slope of the ravine. The uppermost recognizable marker
42
bed is sapropelic layer p, cropping out a few meters below the

crest of the the slope.
Component-section C (Figure 2.19)
This section can be reached from the top of component-section B
by following the map shown in Figure 2.19. An easier route is to
go by car on the Via Cutro from the center of Crotone (Figure
2.19), turning left about 1.8 km from town on a country road that
is about 300 m beyond the first leftward curve. Follow this road
for 3.8 km, as far as a gentle bend to the right some 100 m before
reaching a bridge. Immediately beyond the bend, turn left on the
road to Ca Tuvolo (Figure 2.19), a little house near the road. At
Ca Tuvolo turn again to the left and follow an uneven road for
about 300 m, to the base of an earth dam. From this point walk
along the western edge of the pond behind the dam to its
northernmost point (Figure 2.19), where a gully and a small
valley oriented northeast-southwest are found. About 300 m up
this valley from the edge of the pond, a gully in the western side
of the valley exposes component-section C (Figure 2.7).
Marker bed o, at the base of component-section C (Figure
2.3), crops out a few meters above the point where the gully joins
the main streambed. Within the gully, all the other marker beds
can be recognized, with the marker bed t a few meters below the
upper end.
The topmost part of the continuous Plio-Pleistocene marine
sequence consists of about 2 m of grayish claystones, overlying
marker bed t. Weathered red sands (not shown in the columnar
sections of this chapter) overlie these claystones. These sands are
part of a marine terrace, a large remnant of which is preserved at
the top of the gully.
Acknowledgments
We are especially indebted to the late Professor Raimondo Selli
(University of Bologna), who devoted much of his time to the
problem of the Pliocene-Pleistocene boundary and promoted
the first research on the Vrica section. C. Elmi (University of
Bologna) and R. Sprovieri (University of Palermo) kindly
permitted the use of their data on the Stuni section and on the
distribution of planktic and benthic foraminifera in other Italian
sections. We are also grateful to S. Raffi (University of Urbino)
for very useful information and discussion regarding the mollusks, and to R. Sartori (University of Bologna) for his assistance
in the x-ray diffractometer analyses.
We would like to acknowledge useful comments on an early
version of this chapter from W. A. Berggren (Woods Hole
Oceanographic Institution), H. Nakagawa (Tohoku University),
D. Rio (University of Padova), and S. Sartoni and G. B. Vai
(University of Bologna).
Finally, we wish to thank R. Tampieri (University of Turin), W.
Trentini, and R. Gavaruzzi, who collaborated in measuring and
sampling the Vrica section, and also G. Busatti, L. Casoni, V.
Landuzzi, E. Lipparini, A. Magagnoli, G. Marozzi, M. Mengoli,
and G. Zini (Istituto di Geologia Marina of the C.N.R.), and P.
Ferraresi, P. Ferrieri, and R. Gamberini (University of Bologna)

for their invaluable technical support. The officers of the Societa
di Paleontologia Italiana, the Giornale di Geologia, and the
Academie des Sciences de Paris graciously permitted us to
reproduce figures that first appeared in their journals.
This research was supported by the Istituto di Geologia
Marina of the C.N.R. (contribution no. 600) and by a grant from
the Ministerio della Pubblica Istruzione (40%) to S. Sartoni.
Postscriptum
The original manuscript of this chapter was delivered in 1984,
and partly revised in 1987. The section headed "Approval of the
final proposal by the ICS and the IUGS" was added in 1991, and
in 1995 the chapter was modified in view of important new data.
A complete revision, however, was not done.
References
Arias, C , Azzaroli, A., Bigazzi, G., and Bonadonna, F. 1980.
Magnetostratigraphy and Pliocene-Pleistocene boundary
in Italy. Quat. Res. 13:65-74.
Backman, J., and Pestiaux, P. 1987. Pliocene discoaster abundance variations, Deep Sea Drilling Project site 606:
biochronology and paleoenvironmental implications. In
Initial Reports of the Deep Sea Drilling Project, vol. 94,
part 2, ed. W. F. Ruddiman, R. B. Kidd, et al., pp. 903910. Washington, DC: U.S. Government Printing Office.
Backman, I , and Shackleton, N. J. 1983. Quantitative biochronology of Pliocene and early Pleistocene calcareous
nannofossils from the Atlantic, Indian and Pacific Oceans.
Mar. Micropal. 8:141-170.
Backman, I , Shackleton, N. J., andTauxe, L. 1983. Quantitative
nannofossil correlation to open ocean deep-sea sections
from the Plio-Pleistocene boundary at Vrica, Italy. Nature
304:156-158.
Barone, A., Fabri, A., Rossi, S., and Sartori, R. 1982.
Geological structure and evolution of the marine areas
adjacent to the Calabria Arc. Earth Evol Sci. 3:207-221.
Bassett, M. G. 1985. Towards a "common language" in
Berger, W. H., Yasuda, M. K., Bickert, T., Wefer, G., and
Takayama, T. 1994. Quaternary time scale for the Ontong
Java Plateau: Milankovitch template for Ocean Drilling
Program site 806. Geology 22:463-467.
Bertolani Marchetti, D., Accorsi, C. A., and Bandini Mazzanti,
M. 1978. Primi dati palinologici sulla serie marina pliopleistocenica di Vrica presso Crotone (Calabria). Giorn.
Botan. Ital. 4:296.
Blanc, A. C , Tongiorgi, E., and Trevisan, L. 1954. La limite
Plio-Pleistocene dans la coupe de Monte Mario (environs
de Rome). In Congres Geologique International, Comptes
Rendus de XIX Session, Alger 1952, vol. 15, pp. 218-228.
Boellstorff, J. 1977. The Pliocene/Pleistocene boundary in the
Central Great Plains, U.S.A. In X IN QUA Congress,
Abstracts, p. 37.
Brongersma-Sanders, M. 1957. Mass mortality in the sea. Geol.
Soc. Am., Mem. 67:941-1010.
geomagnetic polarity time scale for the Late Cretaceous
43
Cati, R, and Borsetti, A. M. 1981. Calcareous nannoplankton Colalongo, M. L., Pasini, G., and Sartoni, S. 1981. Remarks on
biostratigraphy of the Vrica Section (Calabria, Southern
the Neogene/Quaternary boundary and the Vrica section
Italy). Giorn. Geol. 43:365-384.
(Calabria, Italy). Soc. Paleontol. ItaL, Boll. 20:99-120.
Channel, J. E. T., Rio, D., Sprovieri, R., and Glacon, G. 1990. Colalongo, M. L., and Sartoni, S. 1977. Globigerina calabra
Biomagnetostratigraphic correlations from Leg 107 in the
nuova specie presso il limite Plio-Pleistocene della
Tyrrhenian Sea. In Proceedings of the Ocean Drilling
Sezione Vrica (Calabria). Giorn. Geol. 42:205-220.
Program: Scientific Results, vol. 107, ed. K. A. Kastens et Colalongo, M. L., and Sartoni, S. 1979. Schema biostratigrafico
al., pp. 670-682. College Station, TX: Ocean Drilling
per il Pliocene e il Basso Pleistocene in Italia. Contr. Prel.
Program.
Realiz. Carta Neotett. ItaL 251:645-654.
Ciampo, G. 1976. Ostracodi pleistocenici di Cala Bianca Combourieu-Nebout, N., Semah, R, and Djubiantono, T. 1990.
(Marina di Camerota - Salerno). Soc. Paleontol. ItaL,
La limite Pliocene-Pleistocene: precisions magnetostratiBoll 15:3-23.
graphiques et climatiques par 1'etude serie de la coupeColalongo, M. L. 1965. Gli ostracodi della serie de Le Castella.
type de Vrica (Crotone, Italie). C. R. Acad. Sci. Paris, ser.
Giorn. Geol. 33:83-123.
2 311:851-857.
Colalongo, M. L. 1968. Cenozone a foraminiferi e ostracodi nel Dalrymple, G. B., and Lanphere, M. A. 1969. Potassium/argon
Pliocene e basso Pleistocene della serie del Santerno e
dating. Principles, techniques and applications to geochrodeirAppennino romagnolo. Giorn. Geol. 35:29-61.
nology. San Francisco: Freeman.
Colalongo, M. L. 1977. The Vrica section (Calabria, Italy): II. Di Geronimo, I. 1969. La sezione stratigrafica Plio-Pleistocenica
Planktonic foraminifera as indicators of the Neogenedi Monte Navone (Piazza Armerina, Enna). Soc. Gioenia
Quaternary boundary. In X INQUA Congress, BirmingSci. Nat., Atti 6:81-146.
ham, Abstracts, p. 29.
Di Napoli, E. 1954. La limite Plio-Pleistocene dans la coupe de
Colalongo, M. L., Cremonini, G., and Sartoni, S. 1978. La
Castell'Arquato (Plaisance). In Congres Geologique Intersezione stratigrafica di Rio Vendina (Messinianonational, Comptes Rendus de XIX session, Alger 1952, vol.
Pleistocene, Reggio Emilia). Soc. ItaL Set Nat. Museo
15, pp. 229-234.
Civ. Stor. Nat. Milano, Atti 119:61-76.
Di Stefano, E., and Rio, D. 1981. Biostratigrafia a nannofossili e
Colalongo, M. L., Nanni, T., and Ricci Lucchi, R 1979.
biocronologia del Siciliano nella localita tipo di Ficarazzi
Sedimentazione ciclica nel Pleistocene anconetano. Geol.
(Palermo, Sicilia). Ateneo Parmense, Acta Nat. 17:97-111.
Romana 18:71-92.
Dondi, L., and Papetti, I. 1968. Biostratigraphical zones of Po
Colalongo, M. L., Padovani, A., Sartoni, S., Tampieri, R.,
Valley Pliocene. Giorn. Geol. 35:63-98.
D'Onofrio, S., Elmi, C , Francavilla, R, Manzoni, M., D'Onofrio, S. 1968. Biostratigrafia del Pliocene e Pleistocene
Poluzzi, A., and Russo, A. 1972. Biostratigrafia e
inferiore delle Marche. Giorn. Geol. 35:99-114.
cronostratigrafia del Pliocene. Soc. Geol. ItaL, Boll. D'Onofrio, S. 1981. I Foraminiferi bentonici della Sezione Vrica
91:489-509.
(Calabria, Italia). Giorn. Geol. 43:327-364.
Colalongo, M. L., and Pasini, G. 1977. The Vrica section Hedberg, H. D. (ed.) 1976. The International Stratigraphic
(Calabria, Italy): III. Ostracoda distribution and their
Guide: a guide to stratigraphic classification, terminology
importance for the Neogene/Quaternary boundary. In X
and procedure. New York: Wiley.
INQUA Congress, Birmingham, Abstracts, p. 90.
Heller, F , Liu, X. M., Liu, T. S., and Xu, T. C. 1991. Magnetic
Colalongo, M. L., and Pasini, G. 1980a. La Ostracofauna della
susceptibility of loess in China. Earth Planet. Sci. Lett.
sezione Vrica in Calabria (con considerazioni sul limite
103:301-310.
Neogene/Quaternario). Soc. Paleontol. ItaL, Boll. 19:44- Hilgen, F. J. 1991. Astronomical calibration of Gauss to
126.
Matuyama sapropels in the Mediterranean and implication
Colalongo, M. L., and Pasini, G. 1980b. Planktonic foraminifera
for the Geomagnetic Polarity Time Scale. Earth Planet.
as markers for the Neogene/Quaternary boundary. In
Sci. Lett. 104:226-244.
XXVI International Geological Congress, Paris, Abstracts,Howell, M. W., Rio, D., and Thunell, R. C. 1990. Laminated
vol. 2, p. 644.
sediments from the Vrica section (Calabria, S. Italy):
evidence for Plio-Pleistocene climatic change in the MediRuggieri, G., Sartoni, S., Selli, R., and Sprovieri, R. 1982.
terranean region. Palaeogeogr. Palaeoclimatol. Palaeoecol.
The Neogene/Quaternary boundary definition: a review
78:195-216.
and proposal. Geogr. Fisica Dinam. Quat. 5:59-68.
Iaccarino, S., and Salvatorini, G. 1982. A framework of
planktonic foraminiferal biostratigraphy for Early MioColalongo, M. L., Pasini, G., Poluzzi, A., and Sprovieri, R.
cene to Late Pliocene Mediterranean area. Paleontol.
1990. Relationship between the benthic foraminifers and
Strat. Evol. 2:115-126.
the ostracodes in the Pliocene-Pleistocene Tyrrhenian
deep-sea record (ODP Leg 107, site 654). In Proceedings Kidd, R. B., Cita, M. B., and Ryan, W. B. F. (1978).
Stratigraphy of eastern Mediterranean sapropel sequences
of the Ocean Drilling Program: Scientific Results, vol. 107,
recovered during DSDP leg 42A and their paleoenvironed. K. A. Kastens et al., pp. 479-493. College Station,
mental significance. In Initial reports of the Deep Sea
TX: Ocean Drilling Program.
Drilling Project, vol. 42, part 1, ed. K. J. Hsu, L.
Colalongo, M. L., Pasini, G., Raffi, L, Rio, D., Sartoni, S., and
Montadert, et al., pp. 421-443. Washington, DC: U.S.
Sprovieri, R. 1984. Biochronology of the Italian marine
Government Printing Office.
Pliocene and lower Pleistocene. In Proceedings of the
XXVII International Geological Congress, vol. 3, pp. 109- King, W. B. R., and Oakley, K. P. 1950. Report of the
Temporary Commission on the Pliocene-Pleistocene
127.
Boundary, appointed 16th August, 1948. In International
Geological Congress, report of the eighteenth session, Great
the Neogene/Quaternary boundary and the Vrica section.
Britain, 1948, Parti. General proceedings, ed. A. J. Butler,
(Unpublished preprint). XXVI International Geological
pp. 213-214. London: Geological Society.
Congress, Paris 1980, Sect. S.08.
44
Landini, W., and Menesini, E. 1978a. L'ittiofauna pliopleistocenica della sezione Vrica (Crotone-Calabria). Soc.
Paleontol. ItaL, Boll. 17:143-175.
Landini, W., and Menesini, E. 1978b. Una nuova forma di
Teleosteo (Tavania crotonensis n. gen., n. sp.) del Pliocene
superiore italiano. Soc. Paleontol. ItaL, Boll. 17:257-261.
Lentini, F. 1971. La sezione plio-pleistocenica di Pisticci sul
bordo appenninico della fossa Bradanica. Soc. Gioenia
Sci. Nat., Atti 7:109-192.
Lourens, L. J., Hilgen, F. J., Gudjonsson, L., and Zachariasse,
W. J. 1992. Late Pliocene to early Pleistocene astronomically forced sea surface productivity and temperature
variations in the Mediterranean. Mar. Micropal. 19:49-78.
Lourens, L. J., Hilgen, F. J., Raffi, I., and Vergnaud-Grazzini,
C. 1994. Early Pleistocene chronology of the Vrica section
(Calabria, Italy). In Astronomical forcing of Mediterranean climate during the last 5.3 million years, ed. L. J.
Lourens, pp. 129-145. Wageningen: Grafisch bedrijf
Ponsen & Looijen (thesis, University of Utrecht).
Nakagawa, H. 1977. Preliminary report on the magneto- and
biostratigraphy of the Vrica section, Calabria, Italy.
Preprint, X INQUA Congress, Birmingham.
Nakagawa, H. 1981. Neogene/Quaternary boundary and correlation of Vrica Section. In Field Conference, NeogeneQuaternary Boundary, India, 1979: Proceedings, ed.
M. V. A. Sastry et al., pp. 107-111. Calcutta: Geological
Survey of India.
Nakagawa, H. 1982. Correlation of the Neogene-Quaternary
transition between Calabria (Italy) and Boso Peninsula
(Japan). In XI INQUA Congress, Moscow, 1982, Abstracts, vol. 2, p. 209. Moscow: Akademia Nauk SSSR.
Nakagawa, H., Niitsuma, N., Mitsunashi, T., Oda, M., Takayama, T., and Sakai, T. 1982. Plio-Pleistocene stratigraphy
in Boso Peninsula, central Japan. In The third report on the
Pliocene-Pleistocene Boundary in Japan, ed. M. Itihara
and Y. Kuwano, pp. 76-80. Kyoto: Dogura and Co.
Nakagawa, H., Oda, M., Sakai, T., Yoshida, K., Asano, K.,
Niitsuma, N., Takayama, T., Tokunaga, S., Kitazato, H.,
and Koizumi, I. 1980. Preliminary results of magneto- and
Italy). In Proceedings of the Second Symposium on the
NeogenelQuaternary Boundary, USSR, 1977, ed. K. V.
Nikiforova and A. Y. Dodonov, pp. 145-156. Moscow:
Akademia Nauk.
Ninkovitch, D., Opdyke, N. D., Heezen, B. C , and Foster, J.
H. 1966. Paleomagnetic stratigraphy, rates of deposition
and tephrachronology in North Pacific deep sea sediments.
Earth Planet. Sci. Lett. 1:476-492.
Obradovich, J. D., Naeser, C. W., Izett, G. A., Pasini, G., and
Bigazzi, G. 1982. Age constraints on the proposed PlioPleistocene boundary stratotype at Vrica, Italy. Nature
298:55-59.
Opdyke, N. D., and Glass, B. P. 1969. The paleomagnetism of
sediment cores from the Indian Ocean. Deep-Sea Res.
16:249-261.
Pasini, G., and Colalongo, M. L. 1982. Status of research on the
Vrica section (Calabria, Italy), the proposed Neogenel
Quaternary boundary-stratotype section, in 1982. Report
presented at XI INQUA Congress, Moscow, 1982. Bologna: Pitagora-Tecnoprint.
Pasini, G., and Colalongo, M. L. 1994. Proposal for the erection
of the Santernian-Emilian boundary-stratotype (lower
Pleistocene) and new data on the Pliocene-Pleistocene
boundary-stratotype. Soc. Paleontol. ItaL, Boll. 33:101120.
Pasini, G., Selli, R., and Colalongo, M. L. 1977a. The section of

Caraffa di Catanzaro (Calabria, Italy). Giorn. Geol.
41:430-435.
Pasini, G., Selli, R., Colalongo, M. L., andD'Onofrio, S. 1977b.
The section of Le Castella. Giorn. Geol. 41:437-440.
Pasini, G., Selli, R., Tampieri, R., Colalongo, M. L.,
D'Onofrio, S., Borsetti, A. M., and Cati, F. 1975. The
Vrica section. In The Neogene-Quaternary Boundary, II
Symposium (Bologna-Crotone), Excursion Guide-Book,
ed. R. Selli, pp. 62-72. Bologna: Editografica Rastignano.
Pasini, G., Selli, R., Tampieri, R., Colalongo, M. L., D'Onofrio, S., Borsetti, A. M., and Cati, F. 1977c. The Vrica
Section. Giorn. Geol. 41:449-458.
Pelosio, G., and Raff], S. 1973. Considerazioni sul limite PlioPleistocene della serie del T. Crostolo (preappennino
reggiano). UAteneo Parmense, Acta Nat. 2:39-63.
Pelosio, G., Raffi, S., and Rio, D. 1980. The Plio-Pleistocene
Boundary controversy. Status in 1979 in the light of
International Stratigraphic Guide. In Volume speciale in
memoria di Sergio Venzo, ed. F. Petrucci and R. Cavazzini,
pp. 131-140. Parma: Grafiche STEP.
Raffi, I., Backman, J., Rio, D., and Shackleton, N. J. 1993. PlioPleistocene nannofossil biostratigraphy and calibration to
oxygen isotope stratigraphies from Deep Sea Drilling
Project site 607 and Ocean Drilling Program site 677.
Paleoceanography 8:387-408.
Raffi, I., and Rio, D. 1979. Calcareous nannofossil biostratigrpahy of Deep Sea Drilling Project site 132, Leg 13
(Tyrrhenian Sea, Western Mediterranean). Riv. ItaL
Paleontol. Strat. 85:127-172.
Raffi, I., and Rio, D. 1980a. Nannoplancton calcareo del
Pliocene superiore-Pleistocene inferiore del Fiume Santerno. Paleontol. Strat. Evol. Quaderno 1:187-195.
Raffi, I., and Rio, D. 1980b. II limite Pliocene/Pleistocene nella
serie del T. Crostolo-Vendina (provincia di Reggio
Emilia, Italia settentrionale). Paleontol. Strat. Evol.
Quaderno 1:169-175.
Ray mo, M. E., Ruddiman, W. F., Backman, J., Clement, B. M.,
and Martinsson, D. G. 1989. Late Pliocene variation in
northern hemisphere ice sheets and North Atlantic deep
water circulation. Paleoceanography 4:413-446.
Rio, D. 1982. The fossil distribution of coccolithophore genus
Gephyrocapsa Kamptner and related Plio-Pleistocene
chronostratigraphic problems. In Initial reports of the Deep
Sea Drilling Project, vol. 68, ed. W. L. Prell, J. V. Gardner,
et al., pp. 325-343. Washington, DC: U.S. Government
Printing Office.
Rio, D., Ruggieri, G., and Sprovieri, R. 1982. Remarks on the
chronostratigraphic classification of lower Pleistocene. In
XI INQUA Congress, Moscow, 1982, Abstracts, vol. 1, p.
265.
Rio, D., Sprovieri, R., and Di Stefano, E. 1994. The Gelasian
stage: a proposal of a new chronostratigraphic unit of the
Pliocene series. Riv. ItaL Paleontol. Strat. 100:103-124.
Rio, D., Sprovieri, R., and Raffi, I. 1984. Calcareous plankton
biostratigraphy and biochronology of the Pliocene-lower
Pleistocene succession of the Capo Rossello area (Sicily).
Rivalenti, C. 1995. II diagramma pollinico di una sezione nelle
argille marine della valle del Lamone (Ravenna, Nord
Italia): un contributo alia correlazione del limite PliocenePleistocene. Unpublished Ph.D. thesis, Universita di
Firenze.
Ruggieri, G. 1954. La limite entre Pliocene et Quaternaire dans
la serie Plio-Pleistocene du Santerno. In Congres Geo-
45
logique International, Comptes Rendus de XIX Session,

cene timescale based on ODP site 677. Trans. R. Soc.
Algerl952, vol. 15, pp. 235-240.
Ruggieri, G. 1962. La serie marina pliocenica e quaternaria dellaShackleton, N. J., Crowhurst, S., Hagelberg, T., Pisias, N. G.,
and Schneider, D. A. 1995b. A new late Neogene time
Romagna. Forli: Camera di Commercio, Industria e
scale: application to Leg 138 sites. In Proceedings of the
Agricultura di Forli.
Ocean Drilling Program: Scientific Results, vol. 138, ed. N.
Ruggieri, G. 1976. La ostracofauna pleistocenica della Falesia di
Cinisi (Sicilia). Soc. Paleontol. ItaL, Boll. 15:85-106.
Drilling Program.
Ruggieri, G. 1977a. Nuovi Ostracodi nordici nel Pleistocene
Shackleton, N. J., and Opdyke, N. D. 1976. Oxygen-isotope and
della Sicilia. Soc. Paleontol. ItaL, Boll. 16:81-85.
paleomagnetic stratigraphy of Pacific core V28-239, late
Ruggieri, G. 1977b. Individuazione del Pleistocene nella Sezione
Pliocene to latest Pleistocene. Geol. Soc. Am., Mem.
di Castrocaro (Forli). Soc. Geol. ItaL, Boll. 95:881-890.
145:449-464.
Ruggieri, G. 1980. Sulla distribuzione stratigrafica di alcuni
Ostracodi nel Pleistocene italiano. Soc. Paleontol. ItaL, Sigl, W., Chamley, H., Fabricius, F , Giroud D'Argoud, G., and
Miiller, J. 1978. Sedimentology and environmental condiBoll. 19:127-135.
tions of sapropels. In Initial reports of the Deep Sea
Ruggieri, G., Buccheri, G., Greco, A., and Sprovieri, R. 1976.
Drilling Project, vol. 42, part 1, ed. K. J. Hsu et al., pp.
Un affioramento di Siciliano nel quadro della revisione
445-465. Washington, DC: U.S. Government Printing
della stratigrafia del Pleistocene inferiore. Soc. Geol. ItaL,
Office.
Boll. 94:889-914.
Ruggieri, G., and Sprovieri, R. 1977. A revision of Italian Spaak, P. 1983. Accuracy in correlation and ecological aspects of
the planktonic foraminiferal zonation of the MediterraPleistocene stratigraphy. Geol. Romana 16:131-139.
nean Pliocene. Utrecht Micropal. Bull. 28:1-160.
Salvador, A. (ed.) 1994. International Stratigraphie Guide: a guide
to stratigraphic classification, terminology and procedure,Sprovieri, R. 1968. La Serie plio-pleistocenica di Agrigento.
Giorn. Geol. 35:295-301.
2nd ed. Boulder, CO: Geological Society of America.
Sastry, M. V. A., Kurien, T. K., Dutta, T. K., and Biswas, S. Sprovieri, R. 1978. I Foraminiferi benthonici della sezione PlioPleistocenica di Capo Rossello (Agrigento, Sicilia). Soc.
(eds.) 1981. Field conference, Neogene-Quaternary boundPaleontol. ItaL, Boll. 17:68-97.
ary, India 1979, Proceedings. Calcutta: Geological Survey
of India.
Sprovieri, R. 1993. Pliocene-early Pleistocene astronomically
forced planktonic foraminifera abundance fluctuations and
Savelli, C , and Mezzetti, R. 1977. The Vrica Section (Calabria,
chronology of Mediterranean calcareous plankton bioItaly): IV. K/Ar age and petrographic characters of the
events. Riv. ItaL Paleontol. Strat. 99:371-414.
"volcaniclastic" layer. In X INQUA Congress, Birmingham, Abstracts, p. 400.
Stanley, D. J. 1978. Ionian Sea sapropel distribution and late
Quaternary paleo-oceanography in the eastern MediterraSelli, R. 1954. La limite Plio-Pleistocene dans les environs
nean. Nature 274:149-152.
d'Ancona (Marche). In Congres Geologique International,
Comptes Rendus de XIX Session, Alger 1952, vol. 15, pp. Tauxe, L., Opdyke, N. D., Pasini, G., and Elmi, C. 1983. Age of
241-247.
Selli, R. 1970. Report on the absolute age. Giant. Geol. 35:51-59.
Selli, R. 1977. The Neogene/Quaternary boundary in the Italian Tiedemann, R., Sarnthein, M., and Shackleton, N. J. 1994.
Astronomic timescale for the Pliocene Atlantic 218O and
marine formations. Giorn. Geol. 41:81-104.
dust flux records of Ocean Drilling Program site 659.
Paleoceanography 9:619-638.
A. M., Cati, F , Colalongo, M. L., D'Onofrio, S., Tortorici, L. 1982. Analisi delle deformazioni fragili dei sedimenti postorogeni della Calabria. Soc. Geol. ItaL, Boll.
100:291-308.
(Calabria-Italy). A potential Neogene/Quaternary Bound- Verhallen, P. J. J. M. 1991. Late Pliocene to early Pleistocene
Mediterranean mud-dwelling foraminifera; influence of a
ary Stratotype. Giorn. Geol. 41:181-204.
changing enivironment on community structure and evoluSelli, R., and Cati, F. (eds.) 1977. Proceedings of the 2nd
tion. Utrecht Micropal. Bull. 40:1-220.
Symposium on the Neogene-Quaternary Boundary.
Giorn. Geol. 41:1-459.
Wezel, F C. 1968. Le cenozone del Pliocene superiorePleistocene inferiore in Sicilia e Lucania. Giorn. Geol.
Shackleton, N. X, Baldauf, J. G., Flores, J.-A., Iwai, M., Moore,
35:437-448.
T. C , Raffi, I., and Vincent, E. 1995a. Biostratigraphic
summary for Leg 138. In Proceedings of the Ocean Drilling Zijderveld, J. D. A., Hilgen, F. J., Langereis, C. G., Verhallen,
Program: Scientific Results, vol. 138, ed. N. G. Pisias et
Program.
Shackleton, N. I , Berger, A., and Peltier, W. R. 1990. An
107:697-714.
alternative astronomical calibration of the lower Pleisto-
The magnetostratigraphy of the Vrica section, Italy, and its

correlation with the Plio-Pleistocene of the Boso Peninsula, Japan
HISAO NAKAGAWA, NOBUAKI NIITSUMA, TOSHIAKI TAKAYAMA, YASUMOCHI MATOBA, MOTOYOSHI
ODA, SHIGEMOTO TOKUNAGA, HIROSHI KITAZATO, TOYOSABURO SAKAI, and ITARU KOIZUMI
Introduction
During the XI INQUA Congress in Moscow in 1982, the

INQUA Subcommission 1-d, "Pliocene-Pleistocene Boundary,"
and the working group of IGCP Project 41, "Neogene/
Quaternary Boundary," jointly proposed that the section at
Vrica in Calabria, southern Italy, be adopted as the PliocenePleistocene boundary-stratotype. With the acceptance of that
proposal by the IUGS International Commission on Stratigraphy
(Nikiforova and Alekseev, Chapter 1, this volume), the type
horizon designated at Vrica has been extended to PlioPleistocene sections around the world.
On the other side of the globe from Vrica, many marine
sedimentary sections that are continuous through the PliocenePleistocene transition are well exposed along the Pacific coast.
Among these the Boso Peninsula in central Japan is one of the
best places to study the Pliocene-Pleistocene transition (Itihara
et al., Chapter 24, this volume). In recent years we have worked
to correlate the Plio-Pleistocene sections of Calabria with those
of the Boso Peninsula on the basis of geomagnetic reversals and
planktonic microfossil events. We began our study of the Vrica
section in 1976, shortly after the initial reports from the Italian
team led by Raimondo Selli (1975; Selli et al., 1977). In 1977, in
meetings at Birmingham, England, and at Dushanbe, Tadjikistan, USSR, we responded to the urgent request of the IGCP-41
working group and reported on a preliminary study of the
paleomagnetic polarity sequence of the Vrica section, although
the sampling interval had been inadequate to satisfactorily
resolve the magnetostratigraphy (Nakagawa et al., 1980). Subsequently, we reported a revised Calabrian Plio-Pleistocene
polarity sequence at a field conference on the NeogeneQuaternary boundary held at Chandigarh, India, in 1979
(Nakagawa, 1981) and in a symposium on the lower boundary of
the Quaternary, held during the XI INQUA Congress in Moscow
in 1982 (Nakagawa, 1982).
Since that time, we have repeated our paleomagnetic analyses
with new equipment, having revisited Vrica in 1983 for further
field work and sampling. As a result we were able to improve
significantly on our preliminary reports, particularly in regard to
biostratigraphic and lithostratigraphic correlations to the paleo46
magnetic record. The magnetostratigraphic data presented here

(Figures 3.4, 3.5, 3.7, 3.8, 3.10) are calibrated according to the
time scale of Berggren, Kent, and Van Couvering (1985); the
same data using orbitally tuned calibration (Zijderveld et al.,
1991) give ages approximately 6% older.
Vrica section
In the vicinity of the Vrica triangulation station south of

Crotone, in the type area of the classic Calabrian Stage (Figure
3.1), fossiliferous marine Plio-Pleistocene strata crop out along
the valleys cut into the "Milazzian" and younger terraces and on
the sea-cliffs (Pasini and Colalongo, Chapter 2, this volume).
Figure 3.2 shows the current status of our field observations in
this area, with the locations and horizons of samples taken for
laboratory analysis of paleomagnetism and microfossils.
The section consists mostly of clayey siltstone, but includes
many layers of sand and sapropelic clays and a few layers of
volcanic ash, which are useful marker horizons. Some of these
layers are interbedded at short intervals, forming distinctive
striped zones of which the intervals defined by markers S9-S9t,
S4-S8, SI-S3, b-e, f'-m2, and qt are the most prominent. The
proposed boundary-stratotype horizon is at the base of the shale
layer overlying sapropelic marker layer e, in the notation of
Pasini and Colalongo (Chapter 2, this volume). The distinctive
stratigraphy of the b-e and f'-m2 zones helps to locate the
boundary in the field. Principal distinguishing characteristics
include the fact that the distances between markers b, c, d, and e
are almost equal, that b and d are bluish gray and c and d are
yellow from a distance, and that/and h are gray and thicker than
the other layers in the zone f'-m2.
The geologic structure of the Vrica section is simple (Figures
3.2 and 3.3). In general the beds dip gently, about 10 westsouthwest. A normal fault strikes N40W near sample localities 31
and 47, with a downthrow to the west of about 10 m. The section
at Vrica lies on the eastern limb of a syncline, the western limb of
which exposes the Plio-Pleistocene section at Le Castella, about
20 km farther to the south.
Figure 3.4 shows the geomagnetic polarity sequence of the
Vrica section together with magnetic measurements of the
Magnetostratigraphy of the Vrica section
47
F. Neto
S. Leonardo
(Costa Tiziana]
(Casa Rossa)
C.DoMto
Capo Colonna
Semaforo
Parasinaci
Capo Rizzuto
10 km
Figure 3.1. Index map of
Calabria.
samples. After initial magnetic examination, we applied thermal gradual cooling, which coincides with the cooling of sea water
demagnetization at 200C and alternating field demagnetization indicated by the coiling change in N. pachyderma.
at 15 mT to all samples. Intensity and direction of the remanent
Colalongo et al. (1981, 1982) and Pasini and Colalongo (1982)
magnetization were measured with a ring-core magnetometer. reported the appearance of the "cold guest" Cytheropteron
The polarity sequence is based on the six most reliable testudo between the sapropelic layers e and/(8 m above e and 12
measurements at each sampled horizon. Stable reversed polarity m below/, near sample localities 10-12). Pasini and Colalongo
is dominant in the section, but a normal-polarity zone with a (Chapter 2, this volume) noted that although this event is no
short reversed intercalation occurs in the middle part. Intermedi- longer considered as a reliable index to the moment at which a
ate polarity values are found in a horizon in the lower part of the glacial climate first affected the Mediterranean, there are further
section and in a thicker zone in the upper part. The normal- reasons to correlate this approximate level at Vrica to the
polarity zone is correlated with the Olduvai normal-polarity initiation of Pleistocene conditions in temperate latitudes.
subchron by the fossil evidence described later; additional Therefore, the recommendations of INQUA Subcommission 1-d
normal polarities, observed in a thin zone just above this by and the IGCP-41 working group that designated the PlioceneZijderveld et al. (1991), are probably also part of the Olduvai as Pleistocene boundary at the top of the marker layer e at Vrica
observed in more condensed marine sections.
(Nikiforova and Alekseev, Chapter 1, this volume; Pasini and
Figure 3.5 shows the stratigraphic distribution of selected Colalongo, Chapter 2, this volume) remain valid.
planktonic microfossils, pollen, and spore grains in the same
section. The index horizons for regional correlation are (1) the
Boso Peninsula
lowest horizon of Gephyrocapsa aperta, (2) the lowest horizon of
Gephyrocapsa caribbeanica, (3) the highest horizon of Globigeri- The Boso Peninsula is situated to the southeast of Tokyo
noides obliquus, (4) the highest horizon of Cyclococcolithus [= (Figure 3.6). A fossiliferous marine Plio-Pleistocene section is
Calcidiscus] macintyrei, (5) the horizon of D-S (dextral-to- well exposed in the river valleys cut into the uplands and on
sinistral) coiling change in Neogloboquadrina pachyderma, and the sea-cliffs in the central to northern part of the peninsula.
(6) the highest horizon of Helicosphaera sellii.
The sediments are mostly sandstone and siltstone, interbedded
No remarkable changes are recognized in pollen flora, but with many layers of volcanic ash, which are valuable correlaincreases in Abies, Picea, and Tsuga and simultaneous decreases tion markers (Figure 3.7) (Itihara et al., Chapter 24, this
in Podocarpus and Taxodiaceae in the upper part indicate volume).
Nakagawa et al.
48
COSTA TIZIANAV
* Reported by
Pasini and
Colalongo,
1982
laminated sapropeiic sandy

siltstone
layer
layer
TOO-l
'
Contour and height in meters

above sea level roughly
estimated
Cliffs and float stones

Valleys, valley flat and pond
4t\<
Talus cone
Road and bridge
Trails
Earth dam, water tanks
and manhole
Figure 3.2. Geologic route map and columnar section for the Vrica
area. The marker horizons a-u are indicated by capital letters A-U for
49
clarity (see also Figures 3.4 and 3.5). Neogene-Quarternary boundary

stratotype.
50
Nakagawa et al.
100
200
300
400 m
Sample
jjL
001
.01
.1T 1 10
I
I
I
m
400 - i
300-
200-
Intensity
(A/m)
06
05
04
03
02
01
07
08
09
10
11
12
13
14
15
X 1O"
Declination
Inclination
Pole Position
N
0
Ei
W*
I I
II
II
S D
180 90
I I
90 90
I T
90
i r
I r
I-
19
29^
30
47
46
45
31
100-
32
33
34
35
36
26
25
24
23
Figure 3.4. Magnetostratigraphy of the Vrica

section.
0-1
i
L_
51
r-750'
-/00
---SiiV^-^----
21
- 50
0
Figure 3.3. Projected cross section of the Vrica area.
Li t h o l o g y
clayey siltstone
laminated siltstone
^ c 5 *
v - j sandy layer
^
-2 S a
:r 5 ." ,*i rf aJSo.,? ^." Ui,? orr o.*
III,
sapropelic layer
-9 z
I 1 3
^ -j ui o o O Z i
I T
T" f 1 * f
lu
"
Coiling
L
ratio
100*/.
100 V.
left
right
Occurrence
:.
of
Foraminifera
iil
: i
and
calcareous
nannoplankton
very rare
Abundance
(i
of
;..
pollen
and
spore
0 <
| < 1
7.
1< < 5
5 < < io
10 < #
< 20
20 < 0
< 50
50 <
I;
11
Figure 3.5. Magnetostratigraphy and biostratigraphy of the Vrica
section.
4 4
4 *
Polarity
normal
Nakagawa et al.
52
Miura
10
20
JO
40^m
Figure 3.6. Index map of the

Boso Peninsula.
In the geomagnetic polarity sequence of the Boso Peninsula,

reversed polarity is dominant in the Plio-Pleistocene section, but
there are three zones, plus several thinner horizons, of normal
polarity (Niitsuma, 1976; Nakagawa and Niitsuma, 1977). In
these same strata, planktonic microfossil index horizons remarkable for their use in long-distance correlation include (1) the
lowest horizon of Globorotalia tosaensis, (2) the lowest horizon
of Globorotalia truncatulinoides, (3) the highest horizon of
Globigerinoides obliquus, (4) the highest horizon of Discoaster
brouweri, (5) the lowest horizon of Gephyrocapsa caribbeanica,
(6) the lowest horizon of Eucyrtidium matuyamai, (7) the
horizon of S-D (sinistral-to-dextral) coiling change of Pulleniatina spp., (8) the lowest horizon of Gephyrocapsa oceanica, (9)
the highest horizon of Helicosphaera sellii, and (10) the highest
horizon of Eucyrtidium matuyamai.
Benthic foraminiferan and molluscan faunas indicate changes
in sea-water temperatures. The indicated changes were moderate in the transition from Pliocene to Pleistocene, although the
general trend was toward gradual cooling in the upper part of the
section.
The transitional zone from Pliocene to Pleistocene is exposed

along the upper streams of the Obitsu, Yoro, and Isumi rivers in
the southeastern part of the peninsula (Figure 3.6). The area is
near the Pacific coast, and modern geological maps have been
published (Mitsunashi et al., 1961, 1976b), as has an excursion
guidebook in English (Mitsunashi, Nakagawa, and Suzuki,
1976a).
Correlation between the Vrica area and the Boso
Peninsula
The Vrica and Boso sections have yielded only a few planktonic
microfossil species in common, but the sequence of index horizons
noted earlier has been correlated to magnetostratigraphic and
biostratigraphic sequences in many deep-sea sediment cores.
Referring to the evidence in deep-sea sections, we can correlate
the microfossil and geomagnetic horizons between Calabria and
the Boso Peninsula as indicated in Figure 3.8. In this interpretation the normal-polarity zone in the lower part of the Kiwada
Formation of the Boso Peninsula is correlated with the Olduvai
LITHOSTRATIGRAPHY
Terr, f. %
Shimosa G.
Kasamori
MAGNETOSTRATIGRAPHY
BIOSTRATIGRAPHY
BENTHIC FORAMINIFERAL
ZONE
PLANKTONIC MICROFOSSIL
Relative
water
temperature
MAGNETO-
Nonionella stell a
COOL\
_\ W/
WARM - Crybroelphidium alavatum
Cassidulina subglobosa
Conan
.
Kakinokidai^ Coiling
direction
Kokumoto
Cassidulina
Uvigerina
Umegase
Bulimina
subaavinata
akitaensis
aauleata
Bulirrina-Bolivina
Bolivina
(U)
a piss a
Otadai
Bulirr,ina-Bo livina
(L)
f> t i I o,? torn at la

kctienziensis
Kiwada
Bulimina
stviata
E
p.
p,
en
Gyvoidina
Ohara
CO
Katsuura
CO
cf. orbiculari
Ptero
Namihana
lyroidina-MeIonic
Lithology
Geomagnetic polarity
r-1000
Kiyosumi
Amatsu
- 500
si ltstone
normal
sandy siltstone
reversed
sandstone
indefinite
conglomerate
alternation of sandstone and siltstone
intraformational deformation
marker tuff
%. terrestrial deposits
Figure 3.7. Magnetostratigraphy and biostratigraphy of the Plio-Pleistocene of the Boso Peninsula.
Nakagawa et al.
54
Detrital
remanent
magnetization
Composite section of
Sediment
grains
?deposited
suspending
fixed
preserved
Figure 3.9. Illustration of time difference between sediment grains and

detrital remanent magnetization.
Hyalinea balthica
Cytheroptern testudo
Gephyrocapsa oceanica
Gephyrocapsa caribbeanica
Eucryrtidium matuyamai
Globorotalia truncatulinoides
Gephyrocapsa aperta
Globorotalia tosaensis
Emt
Hst
CmT
GoT
DbT
LAD
LAD
LAD
LAD
LAD
Eucrytidium matuyamai
Helicosphaera sellii
Cyclococcolithus macintyrei
Globigerina obliqua
Discoaster brouweri
HbB
CtB
GocB
GcB
EmB
GtrB
GaB
GtoB
FAD
FAD
FAD
FAD
FAD
FAD
FAD
FAD
NpRL
NLR
Change from R(ight) to L(eft) coiling in Neogloboquadrina pachyderma

Change from L to R coiling in Pulleniatina species
Figure 3.8. Correlation between the Vrica and Boso sections.
normal subchron of the Matuyama reversed chron in the

standard PMTS (paleomagnetic time scale) (Berggren et al.,
1985) applied in deep-sea cores, and also with the normalpolarity zone identified to this paleomagnetic unit in the middle
part of the Vrica section, just below the marker bed e.
The full sequence of microfossil index horizons linked with the
paleomagnetic reversal sequence in deep-sea sediments is in
rough agreement with the somewhat less complete records of the
Boso and Vrica sections. The relationships of microfossil

biochronologic events to geomagnetic reversals differ in detail,
however, between the deep-sea synthesis and these two exposed
sections. Undoubtedly, geographic, oceanographic, and ecological factors influenced local occurrences of fossils. Besides, if the
age or equivalent horizon is measured using the geomagnetic
reversal sequence as a standard, some time difference is a
physical necessity. This is because the fixation of detrital
geomagnetic orientation in marine deposits does not take place
until the material is buried to about 40 cm (Niitsuma and Ku,
1977). The time required for this depth of burial is, of course,
much longer in slowly deposited sediments. The age of a
sediment grain, or a microfossil specimen, is therefore synchronous not with the age of the remanent magnetization in the
horizon in which it occurs but with that of a horizon about 40 cm
farther below it (Figure 3.9). Thus the apparent microfossil
"date" of a paleomagnetic horizon is always somewhat younger
in deep-sea sediments than in the more rapidly deposited
shallow-marine equivalents.
With these factors taken into account, we can more accurately
correlate the Plio-Pleistocene sections of Calabria to those of
the Boso Peninsula. The Pliocene-Pleistocene boundary designated in the Vrica section, at the top of the sapropelite e,
corresponds in age to a horizon between key tuffs Kd38 and
Kd20 of the Kiwada Formation in the Boso Peninsula (Figure
3.8). The fission-track age of the Kd23 tuff in this interval is 1.6
0.2 Ma (Kasuya, 1990). Figure 3.10 shows further correlation
of the Plio-Pleistocene sections in the Japanese islands and
Taiwan. Nakagawa (1981) presented a correlation of magnetostratigraphic zones between Vrica and some other sections in
Italy, according to the previously published biostratigraphy of
Italian authors. Azzaroli et al. (Chapter 11, this volume) have
reinterpreted the paleomagneticfindingsaccording to new micropaleontological evidence.
ApB
EmB
GaB
GcB
GIB
GocB
GtoB
GtrB
MeB
NaB
NkB
PIB
SdB
FAD
FAD
FAD
FAD
FAD
FAD
FAD
FAD
FAD
FAD
FAD
FAD
FAD
Asterorotalia pulchella
Eucyrtidium matuyamai
Gephyrocapsa aperta
Gephyrocapsa caribbeanica
Globorotalia inflata
Globorotalia truncatulinoides
Mesocena elliptica
Neogloboquadrina asanoi
Neogloboquadrina kagaensis
Pseudoemiliana lacunosa
Sphaeroidinella dehiscens
CmT
DbT
DkT
DpT
EmT
GfT
GoT
GtoT
HsT
MeT
NaT
NkT
RpT
SdT
LAD
LAD
LAD
LAD
LAD
LAD
LAD
LAD
LAD
LAD
LAD
LAD
LAD
LAD
Cyclococcolithus macintyrei
Discoaster brouweri
Denticulopsis kamtschatica
Discoaster pentaradiatus
Globigerinoides fistulosus
Globigerinoides obliquus
Helicosphaera sellii
Mesocena elliptica
Neogloboquadrina asanoi
Neogloboquadrina kagaensis
Reticulofenestra pseudoumbilicata
Sphaeroidinella dehiscens
ST
LAD
Sphaeroidinellopsis
TcT
LAD Thalassiosira convexa
spp.
NpRL
PLR
PRL
Horizon of R(ight) to L(eft) coiling change in Neogloboquadrina pachyderma

Horizon of L to R coiling change in Pulleniatina sp.
Horizon of R to L coiling change in Pulleniatina sp.
KA
OH
Katsuura
Ohara
KU
OU
Kurotaki
Oura
NA
Namihana
55
' UKinawa ^
f Ryukyu
Islands
Figure 3.10. Correlation of the Plio-Pleistocene sections in the Japanese

islands and Taiwan. (Courtesy of Dr. T.-Y. Huang)
Nakagawa et al.
56
1:50,000. Geological maps of oil and gas fields of Japan,

no. 84. Tokyo: Geological Survey of Japan.
Nakagawa, H. 1981. Neogene-Quaternary boundary and correlation of Vrica section. In Field conference, Neogene/
We are indebted to the late Professor R. Selli for general
Quaternary boundary, India, 1979, Proceedings, ed. M. V.
information on the Plio-Pleistocene sections in Italy and for
A. Sastry et al., pp. 107-111. Calcutta: Geological Survey
helpful suggestions, and we are grateful to Professor M. L.
of India.
Colalongo and Dr. G. Pasini for stratigraphic information on the Nakagawa, H. 1982. Correlation of the Neogene-Quaternary
transition between Calabria (Italy) and Boso Peninsula
Vrica section. Professors K. Asano, N. Kitamura, Y. Takaya(Japan). In XI INQUA Congress, Moscow, 1982, Abnagi, and T. Saito very kindly made valuable suggestions to
stracts, vol. 2, p. 209.
improve our work, and Dr. T.-Y. Huang graciously offered the
Nakagawa, H., and Niitsuma, N. 1977. Magnetostratigraphy of
newest columnar section of western Taiwan for correlation. The
the Late Cenozoic of the Boso Peninsula, central Japan.
work was supported by the Ministry of Education, Science and
Quat. Res. 7:294-301.
Culture of the Japanese government and by the Japan Society for Nakagawa, H., Niitsuma, N., Takayama, T., Tokunaga, S.,
Kitazato, H., and Koizumi, I. 1980. Preliminary results of
the Promotion of Science.
magneto-and biostratigraphy of the Vrica section (Calabria, southern Italy). In Proceedings of the Second
Symposium on the Neogene/'Quaternary Boundary, USSR,
1977, ed. K. V. Nikiforova and A. Y. Dodonov, pp. 145References
156. Moscow: Akademia Nauk.
Niitsuma, N. 1976. Magnetic stratigraphy in the Boso Peninsula.
Berggren, W. A., Kent, D. V, and Van Couvering, J. A. 1985.
J. Geol. Soc. Japan, 82:163-181.
Neogene geochronology and chronostratigraphy. In The Niitsuma, N., and Ku, T.-L. 1977. Happenings in biosphere at
Chronology of the Geological Record, ed. N. J. Snelling,
geomagnetic reversals (in Japanese). Kagaku [Science]
pp. 211-250. Geological Society of London, Memoir 10.
47:671-678.
Oxford: Blackwell.
Vrica section (Calabria, Italy), the proposed Neogene/
Ruggieri, G., Sartoni, S., Selli, R., and Sprovieri, R.
1982. The Neogene/Quaternary boundary definition: a
presented at XI INQUA Congress, Moscow, 1982. Boloreview and proposal. Geogr. Fisica Dinam. Quat. 5:
gna: Pitagora-Tecnoprint.
59-68.
Selli, R. (ed.) 1975. The Neogene-Quaternary Boundary, II
Symposium (Bologna-Crotone), Excursion Guide-Book.
the Neogene/Quaternary boundary and the Vrica section
Bologna: Editografica Rastignano. (Also, 1977, Giorn.
(Calabria, Italy). Soc. Paleontol ItaL, Boll. 20:99-120.
Geol. 41:385-458.)
Kasuya, M. 1990. Fission-track ages of tuff layers related to the Selli, R., Accorsi, C. A., Bandini Mazzanti, M., Bertolani
Pliocene-Pleistocene boundary on the Boso Peninsula,
Marchetti, D., Bigazzi, G., Bonadonna, E G . , Borsetti,
Japan. Quat. Res. 33:86-93.
Mitsunashi, T., Nakagawa, H., and Suzuki, Y. (eds.) 1976a. First
International Congress on Pacific Neogene Stratigraphy,
Tokyo, 1976. Guidebook for Excursion 2: Boso Peninsula.
(Calabria-Italy). A potential Neogene/Quaternary BounTokyo: Geological Survey of Japan.
dary-stratotype. Giorn. Geol. 41:181-204.
Mitsunashi, T., Nasu, N., Nirei, H., et al. 1976b. Geological map Zijderveld, J. D. A., Hilgen, F. J., Langereis, C. G., Verhallen,
of Tokyo Bay and adjacent areas, 1:100,000. Miscellaneous
Maps Ser. 20. Tokyo: Geological Survey of Japan.
Mitsunashi, T., Yazaki, K., Kageyama, K., Shimada, T, Ono,
E., Yasukuni, N., Makino, T., Shinada, Y, Fuji war a, K.,
107:697-714.
and Kamata, S. 1961. Geological map of Futtsu-Otaki,
Acknowledgments
Comparison of the laminated units at Vrica and deep-sea sapropels

from the eastern Mediterranean
ISABELLA RAFFI and ROBERT THUNELL
Introduction and background
Sapropels - highly organic, oxygen-reduced layers - in deepwater strata are similar in origin to organic-rich laminite units in
upper-slope sediments and are clear manifestations of major
changes in the climatic-oceanographic regime of the Mediterranean region. The periodic deposition of organic-rich facies at
Vrica may have been related to the intensification of climatic
oscillations in the late Pliocene, as seen in the deep-sea record.
In this chapter, we review the conditions of sapropelic deposition
and the relationship between the upper-slope and deep-sea
events in the Ionian Basin during the later Pliocene and early
Pleistocene.
Laminated sedimentary units are common features in marine
sequences from Miocene to Pleistocene age in the marginal
circum-Mediterranean region. For example, laminated sediments have been described from Algeria (Anderson, 1933),
Spain (Geel, 1978), Sicily (Ogniben, 1957; Brolsma, 1978;
Meulenkamp et al., 1978; McKenzie, Jenkyns, and Bennet,
1979; Gersonde, 1980), southern Italy (Martina et al., 1979),
northern Italy (Sturani and Sampo, 1973), Morocco (Bizon,
Muller, and Vergnaud-Grazzini, 1979), Cyprus (Bizon et al.,
1979), and the Ionian Islands of Greece (Heimann, Just, and
Muller, 1979; Thomas, 1980; Spaak, 1983). Many, but not all, of
these laminated layers are rich in biogenic silica because of high
abundances of diatoms. These diatomaceous laminites are often
collectively referred to as "tripoli."
A number of different mechanisms have been proposed to
explain the formation of Neogene laminites in marginal settings.
The diatom-rich laminites of Morocco (Bizon et al., 1979) and
Sicily (McKenzie et al., 1979; Gersonde, 1980; Van der Zwaan,
1982) have been attributed to increased productivity brought on
by increased upwelling of nutrient-rich waters. Likewise, Van
der Zwaan (1979) also concluded that increased productivity was
responsible for the late Miocene diatomites in the Falconara
section of Sicily. However, he attributed the increased productivity to an increase in the runoff of nutrient-rich continental water.
In addition, it has generally been assumed that these laminated
units were deposited in oxygen-deficient or anoxic bottom waters
(Van der Zwaan, 1982, 1983; Spaak, 1983).
Whereas Neogene laminites have been found in marginal

basins in both the eastern Mediterranean and the western
Mediterranean, laminated deep-water sediments of equivalent
age are found only in the eastern basin of the deep Mediterranean (Kidd, Cita, and Ryan, 1978; Cita and Grignani, 1982;
Thunell, Williams, and Belyea, 1984). These "deep-sea laminites" are black, organic-rich units, commonly referred to as
"sapropels." By definition, sapropels contain more than 2%
organic carbon by weight (Olausson, 1961; Kidd et al., 1978).
Lithologically similar units containing only 0.5% to 2.0% organic
carbon are called "sapropelic layers."
Sapropel formation in the deep eastern Mediterranean is most
commonly ascribed to the periodic development of a low-salinity
surface layer that produced a density stratification and caused
bottom waters to become oxygen-deficient (Olausson, 1961;
Ryan, 1972; Thunell, Williams, and Kennett, 1977; Stanley and
Blanpied, 1980; Rossignol-Strick et al., 1982; Thunell, Williams,
and Cita, 1983). Calvert (1983) and Sutherland, Calvert, and
Morris (1984) have argued convincingly that density stratification alone cannot account for the high organic-carbon content of
deep-sea sapropels found in the eastern Mediterranean. Those
authors suggest that the fresh-water runoff that formed lowsalinity surface layers was also very rich in nutrients, and that
caused a significant increase in productivity coincident with, and
contributing to, oxygen minima below the euphotic zone.
Evidence for increased productivity has been found in carbonisotope studies of late Pleistocene sapropels (Thunell and
Williams, 1982).
The question arises as to what relationship, if any, exists
between these deep-sea sapropels and the laminites from
marginal basins of the Mediterranean. It is obvious that the real
distributions of these two facies are different, since laminites are
found in both the eastern basin and the western basin. Also, as
Spaak (1983) has pointed out, laminite deposition in marginal
areas occurred much more frequently than sapropel formation in
the deep eastern basin. While there have been a number of
geochemical studies of deep-sea sapropels (Sigl et al., 1978;
Calvert, 1983; Sutherland et al., 1984), very little is known about
the organic geochemistry of laminites, making it very difficult to
compare the two. In fact, virtually no information is available on
57
Isabella Raffi and Robert Thunell
58
the organic-carbon content of Neogene laminites from around

the Mediterranean.
The objective of the study reported here was to make direct
comparisons between the laminated units found in the Vrica
section and the deep-sea sapropels recovered from DSDP site
125 on the Mediterranean Ridge. We attempted to evaluate the
stratigraphic relationship of these facies, as well as to make
preliminary comparisons of these units from the viewpoint of
their organic geochemistry. By doing so, we should be able to
determine whether or not it is appropriate to refer to the
laminites at Vrica as "sapropels" (Selli et al., 1977; Colalongo et
al., 1981, 1982).
VRICA
LITHOLOGIC
LOG
300
250
200
Laminated layers in the Vrica section
In the Vrica section, 14 finely laminated layers are intercalated

in the silty and marly limestone succession (Figure 4.1). These
laminated units, together with sandy beds and volcanic ash
layers, have been used as marker horizons for reconstructing
the three component sections that make up the entire sequence
at Vrica (Pasini et al., 1975; Selli et al., 1977). Biostratigraphic
and chronostratigraphic studies indicate that there are five
Upper Pliocene laminites and nine Lower Pleistocene laminites
(Figure 4.1).
The laminated units at Vrica are considered to have been
deposited in an oxygen-deficient, stagnant environment and
have been referred to as sapropels (Pasini and Colalongo, 1982).
Unlike many of the other Neogene laminated units found
throughout the marginal Mediterranean, the laminites at Vrica
contain very few diatoms and radiolaria (Pasini and Colalongo,
1982), although they do contain dwarfed benthic foraminifera
assemblages, as compared with the abundant and diverse fauna
found in the surrounding claystone. In addition, more than 700
well-preserved specimens of fish have been identified from these
layers (Landini and Menesini, 1977, 1978). The benthic
foraminifera assemblages, and also the fossil fish, suggest a
depositional depth of more than 500 m for this sequence. The
undisturbed condition of the fish and the sedimentary laminae is
indicative of an abiotic, and thus anoxic, bottom environment.
CO
cc
LLJ
hLU
PLIOCENE/PLEISTOCENE
BOUNDARY
150
Q.
UJ
Q
100
[|
50
Fine Sand
|y v| Volcanic Ash
pB^
Laminated Layers
Figure 4.1. Stratigraphic column for the Vrica section, showing the positions of various lithologic marker beds and the position of the PliocenePleistocene boundary (top of laminated layer e).
Biostratigraphic correlation of laminites and sapropels
DSDP site 125 is the only eastern Mediterranean deep-sea

succession in which Lower Pleistocene sapropels have been
recovered within a rather complete sedimentary sequence.
Despite some anomalous sedimentation patterns above and
below these Lower Pleistocene sapropels (Raffi and Sprovieri,
1984), the nannofossil data for site 125 and the Vrica section
establish a good biostratigraphic correlation of the two sequences (Figure 4.2).
In the Vrica section (Rio et al., Chapter 5, this volume),
laminated layers a through / were deposited after the last
occurrence of Discoaster brouweri and before the first appearance of Gephyrocapsa oceanica s.l. (i.e., the Crenalithus
doronicoides zone). In this part of the sequence, laminate e has
been proposed as the physical basis for the Pliocene-Pleistocene

boundary-stratotype at Vrica (Pasini and Colalongo, Chapter 2,
this volume). Unfortunately, no sapropels are present in the
equivalent stratigraphic interval at site 125 (Figure 4.2), where
the C. doronicoides zone is represented by a very condensed
section.
Laminite h at Vrica occurs within the Lower Pleistocene
Calcidiscus macintyrei zone, and again there are no stratigraphically equivalent sapropels at site 125 (Figure 4.2). The remaining
seven laminated units at Vrica (layers n-t) are dispersed
throughout the upper 73 m of the section, within the
Vrica laminates and deep-sea sapropels
59
laminites, generally was in the range of 0.4-0.5% (Figure 4.3).

In contrast, the organic-carbon content was nearly doubled
within the laminite layers, reaching values of 0.8-1.0%. In
accordance with the current definition, the three studied
laminites cannot be considered true sapropels, because they do
not contain more than 2% organic carbon by weight. However, if
organic-carbon content is the only criterion considered, then
these three laminites can be classed as sapropelic layers.
The relationship between the organic carbon in the Vrica
sediments and the C/N (carbon/nitrogen) ratios is illustrated in
Figure 4.4. These data clearly indicate that the C/N ratios
increase with increasing organic-carbon content (i.e., that the
laminated sediments typically have higher C/N ratios). This is the
same basic trend observed by Calvert (1983) in a study of late
Pleistocene deep-sea sapropels.
The observed trends in C/N ratios may be explainable in a
number of ways. Most simply, they may be reflecting two
different sources of organic matter, since high C/N ratios are
considered to be characteristic of terrestrial, rather than marine,
organic matter (Trask, 1953; Bordovskiy, 1965). Using that
criterion, Sigl et al. (1978) concluded that there was a large
terrestrial component in the organic matter of Pliocene and
Pleistocene sapropels from the eastern Mediterranean. However, as pointed out by Calvert (1983), C/N ratios can also be
artifacts of diagenesis. As organic matter decomposes with
50
burial, nitrogen is preferentially lost relative to carbon, and this
process is accentuated in environments characterized by a high
accumulation rate (Muller, 1977). That may have been the
situation at Vrica, where sediments accumulated at rates of up to
Figure 4.2. Biostratigraphic correlation of sapropel layers in DSDP site
44 cm/1,000 years (Rio et al., Chapter 5, this volume). Thus,
125 and laminated layers in the Vrica section. The biostratigraphic data
without determining the extent of alteration of the organic
for site 125 are from Raffi and Sprovieri (1984), and those for Vrica are
from Rio et al. (1984) and Rio et al. (Chapter 5, this volume).
matter, it is not possible to conclude unequivocally whether the
organic matter in the laminites is primarily from a marine or
terrestrial source.
Helicosphaera sellii zone (Figure 4.2). It is within this Lower
Previous geochemical and sedimentological studies of deepsea sapropels (Kidd et al., 1978; Muller, 1978; Sigl et al., 1978;
Pleistocene interval that sapropels are also found at site 125.
Cita and Grignani, 1982; Calvert, 1983) provided the basis for
comparison with the laminites we examined from Vrica. The
Geochemical comparison of laminites and sapropels
biggest difference was in the organic-carbon content of the
One Upper Pliocene (layer e) and two Pleistocene laminites laminites, which fell within the range for sapropelic sediments
(layers / and h) were selected for geochemical study. A series of (0.5-2.0% organic C), as opposed to the much higher values in
samples was collected across each of these layers, and the total the sapropels. Despite that difference, the "sapropelic laminites"
organic-carbon and carbonate contents were measured. The and the deep-sea sapropels were similar in that both were
findings from the determinations of organic-carbon and carbon- characterized by high C/N ratios (Sigl et al., 1978; Calvert,
ate contents are presented in Figure 4.3. A consistent and 1983).
systematic decrease in carbonate content and an increase in
Another similarity between the laminites and the sapropels
organic-carbon content were found within each of the laminites was that the carbonate contents in both units were significantly
(Figure 4.3). Both the basic patterns and the absolute values lower than in the surrounding sediment (Calvert, 1983). Such a
were similar across the layers. In sediments enclosing the decrease in carbonate content could be due to (1) a decrease in
laminites, carbonate values ranged from 16% to 20%, while the production of calcareous plankton, (2) an increase in
within the laminites those values decreased to between 12% and carbonate dissolution, and/or (3) an increase in the input of
14%. Within 50 cm above the top of each laminite, the carbonate noncalcareous sediments. With regard to the first explanation, a
content returned to the value typically found just below the decrease in productivity during the deposition of either laminites
layer.
or sapropels would be at odds with the models that have been
The organic-carbon content, both above and below each of the proposed for their formation. In particular, stable-isotope
60
ORGANIC C (%)
200
199 -
200h
195 -
190 -
0.0 0.2 0.4 0.6 0.8 1.0
20
i
10 12 14 16 18
20
z\
193 s.
191 -
LU
12 14 16 18
i
192 -
185 -
10
- 6
5
4
3
2
199.
g
t
iiiiiiiiiii
CARBONATE(%)
0.0 0.2 0.4 0.6 0.8 1.0
190 -
180 -
175 0.0
171 -
170 -
170 169 168 167 -
165 d
0.2 0.4 0.6 0.8
1.0
9
R
6
5
- 4
- 2
^1
iiPilllt
Figure 4.3. Vrica laminates. Results of analyses of organic-carbon
and carbonate contents across laminated layers e, / , and h at Vrica.
160-
studies (Thunell and Williams, 1982) and geochemical studies

(Calvert, 1983) indicate a general increase in productivity during
the deposition of late Pleistocene sapropels. The second explanation, that of dissolution, is inconsistent with the observation that
calcareous microfossils in both types of facies are not appreciably
less well preserved within these layers than in surrounding
sediments (Sigl et al., 1978; Thunell and Williams, 1982). We
consider, therefore, that the decrease in carbonate content
within laminites and sapropels may have been due to dilution by
noncalcareous sediments. Increased clastic input associated with
enhanced runoff from the continents could be responsible for the
observed decrease in carbonate, particularly in the marginal
settings where most sapropelites (= laminites) are found.
Increased runoff also would have been a source for the
additional nutrients needed to raise productivity and increase the
flux of organic carbon to the seafloor. An increase in runoff also
may have resulted in the input of higher levels of terrestrial
LAMINATED LAYERS
organic matter, which could be at least a minor component of the

relatively high total organic carbon in the laminated layers.
Discussion
The deposition of laminites in marginal settings and of sapropels

in the deep eastern basins undoubtedly was related to major
changes in the climatic-oceanographic regime of the Mediterranean region. Thunell et al. (1984) have demonstrated that the
latest Pliocene and early Pleistocene sapropels were fundamentally different from earlier Miocene and Pliocene sapropels and
that the formation of the more recent sapropels was climatically
modulated. In particular, the alternating glacial-interglacial
climatic conditions that began in the late Pliocene (Thunell and
Williams, 1983; Shackleton et al., 1984,1995) may have been the
triggering mechanism for both laminate and sapropel formation.
These glacial-interglacial climatic changes most certainly had an
Vrica laminates and deep-sea sapropels
15
SAPROPELIC MARL
MARL
o
o
10
CO
DC
.o.o
0 - Laminated
- Nonlaminated
02
0.4
0.6
OB
1.0
Organic C (wt %)
Figure 4.4. Relationship between organic-carbon content and C/N ratio
in laminated and nonlaminated sediments from Vrica. The distinction
between marls and sapropelic marls is based on the definition that
sapropelic marls contain between 0.5% and 2.0% organic carbon (Kidd
et al., 1978).
impact on the discharge of fresh water into the Mediterranean. It

has been suggested that increased discharge from the Nile, due
to increased precipitation, periodically altered the surface-water
conditions in the eastern Mediterranean (Rossignol-Strick et al.,
1982). Increased runoff would serve to establish sapropelproducing conditions in two ways: (1) The surface-water
productivity would be increased because of increased nutrient
input from the continents, and that, in turn, would increase the
flux of organic matter to the seafloor. (2) The reduction of
surface-water salinities would inhibit vertical overturn of the
water column, resulting in oxygen depletion in the bottom waters
and an environment favorable for preservation of organic
matter.
Acknowledgments
The authors are indebted to D. Rio, R. Sprovieri, and E. Tappa
for their assistance in carrying out the sampling at Vrica. We
thank E. Tappa and B. Linsley for analytical assistance, and J.
Morris for use of his laboratory facilities. This work was partially
supported by NSF grants OCE-81117007 and INT-8400206 and
by NATO grant 870119 to R. Thunell.
References
Anderson, R. 1933. The diatomaceous and fish-bearing Beida
Slate of Algeria. /. Geol. 41:673-698.
Bizon, G., Muller, C , and Vergnaud-Grazzini, C. 1979.
Paleoenvironmental conditions during the deposition of
Late Miocene diatomaceous sediments in Morocco and
Cyprus. Ann. Geol. Pays Hell, tome hors ser. 1:113-128.
Bordovskiy, O. K. 1965. Sources of organic matter in marine
basins. Mar. Geol. 3:5-31.
61
Brolsma, M. J. 1978. Quantitative foraminiferal analysis and

environmental interpretation of the Pliocene and topmost
Miocene on the South coast of Sicily. Utrecht Micropal.
Bull. 18:1-121.
Calvert, S. E. 1983. Geochemistry of Pleistocene sapropels and
associated sediments from the eastern Mediterranean.
OceanoL Acta 6:255-267.
Cita, M. B., and Grignani, D. 1982. Nature and origin of Late
Neogene Mediterranean sapropels. In Nature and origin of
Cretaceous carbon-rich fades, ed. S. O. Schlanger and M.
B. Cita, pp. 165-196. New York: Academic Press.
and proposal. Geogr. Fisica Dinam. Quater. 5:59-68.
the Vrica section (Calabria, Italy). Soc. Paleontol. Ital.,
Boll. 20:99-120.
Geel, T. 1978. Messinian gypsiferous deposits of the Lorca Basin
(province of Murcia, SE Spain). Soc. Geol. Ital, Mem.
16:369-385.
Gersonde, R. 1980. Palaoekologische und biostratigraphische
Auswertung von diatomeen Assoziationen aus dem Messinium aus dem Caltanisetta-Beckens (Sizilien) und einiger
Vergleichsprofile in SO-Spanien, NW-Algerien und auf
Kreta. Unpublished thesis, University of Kiel.
Heimann, L., Just, J., and Muller, C. 1979. Sedimentological and
paleoecological aspects of Messinian strata on the Ionian
Islands (Greece). Ann. Geol. Pays Hell., tome hors ser.
1:501-514.
Kidd, R. B., Cita, M. B., and Ryan, W. B. F. 1978. Stratigraphy
of Eastern Mediterranean sapropel sequences recovered
during Leg 42A and their paleoenvironmental significance. In Initial Reports of the Deep Sea Drilling Project,
vol. 42, ed. K. J. Hsu, L. Montadert, et al., pp. 421-443.
Washington, DC: U.S. Government Printing Office.
Landini, W, and Menesini, E. 1977. L'lttiofauna Plio-Pleistocenica della sezione della Vrica (Crotone-Calabria) (Note
preliminare). Soc. Tosc. Sci. Nat., S.A., Atti. 84:1-14.
Landini, W, and Menesini, E. 1978. L'lttiofauna Plio-Pleistocenica della sezione della Vrica (Crotone-Calabria). Soc.
Paleontol. Ital., Boll. 17:143-175.
Martina, E., Casati, P., Cita, M. B., Gersonde, R., D'Onofrio,
S., and Bossio, A. 1979. Notes on the Messinian
stratigraphy of the Crotone Basin, Calabria (Italy). Ann.
Geol. Pays Hell., tome hors ser. 2:755-766.
McKenzie, J., Jenkyns, H., and Bennet, G. 1979. Stable isotopic
study of the cyclic diatomite-claystones from the Tripoli
Formation, Sicily: a prelude to the Messinian Salinity Crisis. Palaeogeogr. Palaeoclimatol. Palaeoecol. 29:125-141.
Meulenkamp, J. E., Schmidt, R., Tsapralis, V, and Van der
Zwaan, G. J. 1978. An empirical approach to paleoenvironmental analysis. Kon. Ned. Akad. Wetens., Proc, ser. B
81:339-363.
Muller, J. 1978. Carbonate content, carbonate mineralogy, and
bulk mineralogy of DSDP Leg 42A samples. In Initial
Reports of the Deep Sea Drilling Project, vol. 43A, ed.
W. B. F. Ryan, K. J. Hsu, et al., pp. 1211-1220.
Muller, P. J. 1977. C/N ratios in Pacific deep-sea sediments:
effect of inorganic ammonium and organic nitrogen
compounds sorbed by clays. Geochim. Cosmochim. Acta
41:165-116.
Obradovich, J. D., Naeser, C , Izett, G., Pasini, G., and Bigazzi,
G. 1982. Age constraints on the boundary stratotype at
Vrica, Italy. Nature 298:55-59.
62
Ogniben, L. 1957. Petrografia delle Serie Solififera Siciliana e Sigl, W., Chamley, H., Fabrocius, R, D'Argoud, G., and Muller,
considerazioni geologische relative. Serv. Geol. Ital,
J. 1978. Sedimentology and environmental conditions of
sapropels. In Initial Reports of the Deep Sea Drilling ProjMem. Descr. Delia Carta Ital 33:1-275.
ect, vol. 42, ed. K. J. Hsu, L. Montadert, et al., pp. 445Olausson, E. 1961. Studies of deep-sea cores. Rep. Swed. Deepsea Exped. 1947-1948 8:353-391.
465. Washington, DC: U.S. Government Printing Office.
Pasini, G., and Colalongo, M. L. 1982. Status of research on the Spaak, P. 1983. Accuracy in correlation and ecological aspects of
Vrica section (Calabria-Italy), the proposed Neogenel
the planktonic foraminiferal zonation of the MediterraQuaternary boundary stratotype section, in 1982. Report
nean Pliocene. Utrecht Micropal Bull 28:1-160.
presented at XI INQUA Congress, Moscow, 1982. Bolo- Stanley, D. J., and Blanpied, C. 1980. Late Quaternary water
exchange between the eastern Mediterranean and the
Pasini, G., Tampieri, R., Colalongo, M. L., D'Onofrio, S.,
Black Sea. Nature 285:537-541.
Borsetti, A. M., and Cati, F. 1975. The Vrica Section. In Sturani, C , and Sampo, M. 1973. II Messiano inferiore in facies
diatomitica del bacino terziaro Piemontese. Soc. Geol
The Neogenel Quaternary boundary: II symposium (Bologna-Crotone) excursion guidebook, ed. R. Selli, pp. 62Ital, Mem. 12:335-349.
Sutherland, H. E., Calvert, S. E., and Morris, R. J. 1984.
72. Bologna: Editografica Rastignano.
Geochemical studies of the recent sapropel and associated
Raffi, I., and Sprovieri, R. 1984. Calcareous plankton in DSDP
sediment from the Hellenic Outer Ridge, Eastern MediterHole 125-Leg 13 (Ionian Sea-Eastern Mediterranean: a
ranean. Mar. Geol. 56:79-92.
review). Soc. Paleontol. Ital, Boll. 23:57-73.
Rio, D., Sprovieri, R., and Raffi, I. 1984. Calcareous planktonic Thomas, E. 1980. Details of Uvigerina development in the
Cretan Mio-Pliocene. Utrecht Micropal Bull. 23:1-167.
biostratigraphy and biochronology of the Pliocene-Lower
Pleistocene succession of the Capo Rossello area (Sicily). Thunell, R. C , and Williams, D. F. 1982. Paleoceanographic
Mar. Micropal 9:135-180.
events associated with Termination II in the Eastern
Mediterranean. Oceanol Acta 5:229-233.
Rossignol-Strick, M., Nesteroff, W., Olive, P., and VergnaudGrazzini, C. 1982. After the deluge: Mediterranean Thunell, R. C , and Williams, D. F. 1983. The stepwise
stagnation and sapropel formation. Nature 295:105-110.
paleoceanographic conditions in the Mediterranean: oxyRyan, W. B. F. 1972. Stratigraphy of Late Quaternary sediments
gen isotopic studies of DSDP sites 125 and 132. Utrecht
in the eastern Mediterranean. In The Mediterranean Sea: a
Micropal Bull 30:111-128.
natural sedimentation laboratory, ed. D. J. Stanley, pp.
149-169. Stroudsburg, PA: Dowden, Hutchinson, & Ross, Thunell, R. C , Williams, D. F , and Belyea, P. 1984. Anoxic
events in the Mediterranean Sea in relation to the evolution
Inc.
of late Neogene climates. Mar. Geol. 59:105-134.
Marchetti, D., Bigazzi, G., Bonadonna, K G . , Borsetti, Thunell, R. C , Williams, D. F , and Cita, M. B. 1983. Glacial
anoxia in the eastern Mediterranean. /. Foram. Res.
13:283-290.
Savelli, C , and Tampieri, R. 1977. The Vrica section Thunell, R. C , Williams, D. R, and Kennett, J. P. 1977. Late
(Calabria, Italy), a potential Neogene/Quaternary boundQuaternary paleoclimatology, stratigraphy and sapropel
history in eastern Mediterranean deep-sea sediments.
Shackleton, N. J., Backman, J., Zimmerman, H., Kent, D. V,
Hall, M. V, Roberts, D. G., Schnitker, D., Baldauf, J. Trask, P. D. 1953. Chemical studies of the sediment of the
G., Desprairies, A., Homrighanson, R., Huddleston, P.,
western Gulf of Mexico. M.I.T Pap. Phys. Oceanogr.
Keene, J. B., Kaltenback, A. J., Krumsiek, K. A. D.,
Meteorol. 12:49-120.
Morton, A. C , Murray, J. W., and Westberg-Smith, J. Van der Zwaan, G. J. 1979. The pre-evaporite late Miocene
1984. Oxygen isotope calibration of the onset of ice rafting
environment of the Mediterranean: stable isotopes of
and the history of glaciation in the North Atlantic region.
planktonic foraminifera from section Falconara, Spain.
Nature 307:620-623.
Kon. Ned. Akad. Wetens., Proc, ser. B 82:487-502.
Shackleton, N. J., Hall, M. A., and Pate, D. 1995. Pliocene Van der Zwaan, G. J. 1982. Paleoecology of Late Miocene Mediterranean foraminifera. Utrecht Micropal. Bull 25:1-202.
the Ocean Drilling Program: scientific results, Leg 138, ed.Van der Zwaan, G. J. 1983. Quantitative analyses and the
reconstruction of benthic foraminiferal communities.
Utrecht Micropal. Bull 30:49-69.
Calcareous nannofossil biochronology and the Pliocene-Pleistocene

boundary
DOMENICO RIO, ISABELLA RAFFI, and JAN BACKMAN
Introduction
Calcareous nannofossils are microscopic skeletal elements of

calcite produced by planktic unicellular algae, occurring abundantly in marine sediments from the Jurassic to the present. The
rapid evolution of the nannofossils through time has been used to
establish highly resolved biostratigraphic zonations, particularly
for the Cretaceous and Cenozoic eras (e.g., Martini, 1971;
Thierstein, 1976; Roth, 1978; Bukry, 1978). The zonal boundaries are defined by extinctions or first appearances of individual
species, and such datum events have been found to have
occurred synchronously on a global scale, when calibrated
against oxygen-isotope stratigraphy (e.g., Thierstein et al., 1977)
or against magnetostratigraphy (e.g., Backman and Shackleton,
1983).
Nannofossils are well represented in the Pliocene and Lower
Pleistocene marine deposits in Italy, and therefore they provide a
means by which this regional geologic record can be correlated
with extra-Mediterranean areas (Figure 5.1). This possibility of
correlation is important because the type regions of the Pliocene
and the Pleistocene are located in Italy (Lyell, 1833; Berggren,
1972).
In this chapter we shall briefly review the development of
nannofossil biostratigraphy at the Pliocene-Pleistocene transition, as derived from deep-sea sediments and regions in Italy.
Second, we shall present the biochronology of critical nannofossil datum events. Third, we shall insert the regional Italian
lithostratigraphic record into the established nannofossil
biostratigraphic-biochronologic framework and estimate the age
of the physical horizon that has been proposed by INQUA
Subcommission 1-d, "Pliocene/Pleistocene Boundary," to be
used as the definition of the Pliocene-Pleistocene boundary in
the Vrica section, Calabria.
This chapter was submitted essentially in its present form
in 1985, when two of us (D.R. and I.R.) were at the Istituto di
Geologia, Universita di Parma. Although much work has
subsequently been done on Italian Plio-Pleistocene stratigraphy (Nikiforova, Foreword, this volume), only a few
changes have been necessary to update the biostratigraphic data
herein.
Nannofossil biostratigraphy at the Pliocene-Pleistocene

transition
Biostratigraphic studies of calcareous nannofossils originated

from the work of Bramlette and Riedel (1954). Within a decade,
the development of Pliocene and Pleistocene nannofossil biostratigraphy was widely recognized. Ericson, Ewing, and Wollin
(1963) and Riedel, Bramlette, and Parker (1963) characterized
the extinction of Discoaster brouweri as a useful datum event in
the deep-sea sediments of different oceans. Akers (1965)
observed that the evidence of that extinction event in the
northern Gulf of Mexico had been preceded by the closely
spaced extinctions of Discoaster surculus, D. pentaradiatus, and
D. variabilis. Most notably, however, Ericson et al. (1963)
correlated the extinction level of Discoaster brouweri to the
Pliocene-Pleistocene boundary. Although that correlation was
immediately questioned (Riedel et al., 1963; Akers, 1965), it has
been widely used since then to separate the Pliocene from the
Pleistocene in deep-sea sediments. A study by Mclntyre, Be, and
Pretiskas (1967) indicated that the first appearances of species of
the genus Gephyrocapsa followed shortly after the extinction of
D. brouweri and that those gephyrocapsid events could also be
used for approximate determination of the Pliocene-Pleistocene
boundary.
The first tentative nannofossil zonations for the Upper
Pliocene and the Lower Pleistocene were proposed by Hay et al.
(1967) and Boudreaux and Hay (1967, 1969). Gartner (1969)
proposed a zonation that later was incorporated in the Cenozoic
"standard zonation" of Martini (1971). In a comprehensive zonal
scheme for the entire Cenozoic, Bukry (1973) utilized the
Gephyrocapsa group to refine biostratigraphic subdivisions in
the interval of the Pliocene-Pleistocene boundary, while also
noting that Calcidiscus macintyrei disappeared shortly after the
D. brouweri extinction. The revised Pleistocene zonation of
Gartner (1977) disregarded the Gephyrocapsa events and instead
subdivided the Lower Pleistocene interval, after the extinction of
D. brouweri, in the ascending order of (1) the extinction of C.
macintyrei, (2) the extinction of Helicosphaera sellii, and (3) the
upper limit of a short stratigraphic interval strongly dominated
by small Gephyrocapsa spp.
63
Domenico Rio, Isabella Raffi, and Jan Backman
64
12
16
20
CASTELL'ARQUATO
44
40
S.MARIA dl CcutanzaAo AFRICA

CAST EL LA
FICARAZZI
SAW NICOLA
36
KmO
100
200
Figure 5.1. Italian sections discussed in the text.

Gartner's (1977) set of events (including the D. brouweri
extinction) can be readily identified in Mediterranean deep-sea
sediments (Raffi and Rio, 1979). In Italian land sections,
however, consistent identification of these datum events has
been difficult because of severe sediment reworking and,
particularly in northern Italian sections, because of the fact that
both C. macintyrei and D. brouweri declined to low abundances
prior to their extinctions (Rio et al., 1982). Quantitative
approaches can be used, however, to overcome most of the
reworking problem (Backman and Shackleton, 1983). On the
other hand, the use of first appearances, rather than extinctions,
minimizes the influence of reworking on the biostratigraphic
resolution. This led Haq, Berggren, and Van Couvering (1977),
Gartner (1977), and Raffi and Rio (1979) to make renewed
studies of the biostratigraphic distribution of the Gephyrocapsa
group of species, which, although beset by problems of taxonomy and nomenclature, are abundant in Italian land sections.
This group also underwent marked evolutionary changes in the
critical time interval, thus offering the opportunity for high-level
biostratigraphic control. Rio (1982), in summarizing the work on
Lower Pleistocene gephyrocapsids, indicated four biostratigraphically useful events, in ascending order: (1) the first
appearance of Gephyrocapsa oceanica s.L, (2) the first appearance of Gephyrocapsa spp. larger than 5.5 /xm, (3) the beginning
of dominance of small Gephyrocapsa spp., and (4) the end of

dominance of small Gephyrocapsa spp.
Finally, Backman and Shackleton (1983) demonstrated that
the increase in abundance of Discoaster brouweri triradiatus
relative to D. brouweri represents a distinct biostratigraphic
signal in sediments of late Pliocene age. They also showed that
these Discoaster forms had virtually the same extinction times.
Discoaster brouweri triradiatus shows rare and scattered occurrences in the Pliocene, except during the last few hundred
thousand years of its range, and therefore is less likely to have
been reworked.
The zonations and the set of events discussed are summarized
in Figure 5.2.
Nannofossil biochronology
The biochronologic resolution that can be achieved by means of
calcareous nannofossils at the transition from the Pliocene to the
Pleistocene is determined by eight of nine datum events reported
in Figure 5.2. All but H. sellii provide datum events usable over
wide geographic areas, including the Mediterranean region.
Backman and Shackleton (1983) have shown that H. sellii is
virtually valueless for long-distance correlations, and this species
is therefore excluded from further biochronologic discussion.
The remaining eight datum events have been determined in
sedimentary sequences that have established magnetostratigraphies, allowing their ages to be derived from interpolations of
sediment accumulation rates in reference to the magneticreversal boundaries. Therefore, the precision of each age
estimate depends partly on the accuracy of the time scale used
and partly on uncertainties in estimates of sediment accumulation rates, which increase as a function of increasing distance
from the nearest magnetic-reversal boundary. The error in our
age estimates probably does not exceed 0.05 Ma in seven of the
eight cases.
In a series of studies, Backman and Shackleton (1983),
Backman, Shackleton, and Tauxe (1983), and Backman et al.
(1984) worked out a reliable nannofossil biochronology in which
a sequence of datum events can be shown to follow a predictable
pattern in numerous deep-sea cores for which there are
paleomagnetic controls, and which represent a wide range of
geographic locations.
Thefirstdatum event is defined as the increase in abundance of
D. brouweri triradiatus relative to D. brouweri to values in excess
of 20%. The higher proportion of the former lasted uninterruptedly until the simultaneous extinctions of both forms. The
duration of this signal may range between approximately 0.1 and
0.2 m.y., since the available data do not allow a better age
determination of its beginning. The synchronous extinctions of D.
brouweri and D. brouweri triradiatus occurred immediately prior
to the Olduvai subchron. These events took place at 1.89 Ma,
according to the time scale of Berggren, Kent, and Van Couvering
(1985) used in preparing this chapter, corrected to 2.04 Ma in the
most recently revised orbitally tuned time scale (Shackleton et al.,
1995). The available data indicate that these Discoaster events
Calcareous nannofossils, Vrica
HAY et alii
1967
65
BUKRY 1973 - 1978

GARTNER 1969
RAFF I and RIO 1979
DATUM
GARTNER 1977
MARTINI 1971
OKADA and BUKRY 1980
P.LACUNOSA
S.PULCHRA
+A.S.G.
+A.S.G.
SMALL
GEPHYROCAPSAE
SHALL
CN
14 a
GEPHYROCAPSA
OCEANICA
PSEUDOEMILIANIA
P.LACUNOSA
NN 19
E. OVATA
GEPHYRXAPSA
+H.s.
anall Gephyrocapsa*
-x- H.sellii
H.SELLII
H.SELLII
CN
13 b
CN
13 a
G.CARIBBEANICA
~T~ C.macintyrei
G.CARIBBEANICA
C.MACINTYREI
G.o.
E.ANNULA
_I_G.o
C.DORONICOIDES
+D.b.
T T
DISCOASTER
DISCOASTER
BROUWERI
BROUWERI
CN 12
CN
12 d
CN
12 c
D.BROUWERI
+D.p.
C.MACINTYREI
+D.p.
+D.p.
D. PENT/HIATUS
NN 17
-*
_1_ Large (>5.5ûm)

Gephyrocapsa
CN13
C.MACINTYREI
NN 1 8
End Acme
Small Gephyrocapsa
Beginning Acme
A.S.G.
D.BROUWERI
EVENTS
MEDITERRANEAN ZONATION
D.s
D.PENTARADIATUS
D.SURC
D.brouweri +
D.brouweri var.
triradiatus
Beginning Acme
D.brouweri var.
triradiatus
D.PENTWfllATUS
Figure 5.2. Principal nannofossil zonation schemes proposed for the Upper Pliocene and Lower Pleistocene. Note that the range of Helicosphaera sellii is valid only for Mediterranean sections. Abbreviations in
Figure 5.2 refer to binomens of index taxa in right-hand column. Other

symbols are as follows: A.S.G., acme of small gephyrocapsids; *, first
appearance; +, last appearance or extinction.
occurred at the same time in all low-latitude and mid-latitude

oceans, and in the higher latitudes of the North Atlantic as well.
Such was also the case with the extinction of Calcidiscus
macintyrei, although that event occurred later than the Olduvai
subchron. These three datum events are easily identified from a
taxonomic point of view, adding to their value as biochronologic
indices. We have therefore used the extinction events of these
forms for deducing sediment accumulation rates in the sequence
provided by DSDP site 132 in the Tyrrhenian Sea, in which
magnetostratigraphy is lacking, in order to derive age estimates
for the Gephyrocapsa group in that sequence. The C. macintyrei
event has also been used in analyzing core V12-18 and DSDP site
502B in order to account for the accumulation rate changes
occurring between the Olduvai and Jaramillo subchrons.
The biochronology of datum events provided by the genus
Gephyrocapsa is complicated by the fact that unambiguous
identification of gephyrocapsid species is difficult because of
profound intra-species and inter-species morphologic variations.
Only by applying rigorous morphometric criteria (Rio, 1982) has
it been possible to consistently distinguish certain transition
events (as discussed earlier) in the group's evolution using lightmicroscope techniques. In order to estimate ages for the four
Gephyrocapsa events in sequences with established magnetostratigraphies, we have studied their stratigraphic distributions
in core V28-239 from the western equatorial Pacific (Backman
and Shackleton, 1983), in core V12-18 from the southwestern
Atlantic (Backman and Shackleton, 1983), and in DSDP site
502B in the Caribbean (Prell et al., 1982).
The findings from core V28-239 are presented in Figure 5.3.
Of all the material studied, this core provides the best sequence
as far as continuity in deposition and control of sedimentation

rates are concerned. The first appearance of G. oceanica s.l.
occurs over a 30-cm interval just above the Olduvai subchron,
and the first appearance of Gephyrocapsa spp. larger than 5.5
fjum occurs at a level about halfway between the Olduvai and
Jaramillo (1.34 Ma uncorrected; 1.44 corrected). Shortly below
the base of the Jaramillo subchron, a profound change is
recorded in the Gephyrocapsa assemblage, characterized by
disappearance of large and normal-sized Gephyrocapsa spp. This
event marks the beginning of an interval in which, virtually, only
Gephyrocapsa spp. smaller than 3.5 jLtm are present. This
dominance of small Gephyrocapsa spp. comes to an end close to
the top of the Jaramillo subchron, being just within it at low
latitudes, and just above it at middle and higher latitudes in the
Atlantic, as well as in the Mediterranean (Castradori, 1993).
Gartner (1977) first observed this interval and introduced a
"small Gephyrocapsa Zone," which he defined as "the interval
from the highest occurrence of Helicosphaera sellii to the highest
level of dominantly small Gephyrocapsa." Gartner considered
that the uppper limit of the stratigraphic interval characterized
by the acme of small Gephyrocapsa spp. was "very sharp," and
its lower limit less sharp "but still readily identifiable." Rio's
(1982) results indicate, however, that the lower limit of the small
Gephyrocapsa spp. acme is no less distinct than its upper limit.
The data sets obtained from core V12-18 and from DSDP site
502B (Figure 5.4) provide age estimates for the gephyrocapsid
datum events that are, without exception, in good agreement
with those obtained from V28-239, indicating synchronous
occurrences in the western equatorial Pacific, the southwestern
Atlantic, and the Caribbean Sea.
66
AGE (Ma)
1.5
2.0
1.15 Ma
1.0
End small >

Geph. spp.
N.
10 Beginning acme smatrV

Gephyrocapsa
spp
12 FA Gephyrocapsa
-
spp. > 5.5 |Jm
LO C. macmtyrei
FA
G. oceamca
si.
16 -
18 -
Figure 5.3. Sediment accumulation rate for core

V28-239, plotted against nannofossil-datum
chronology.
LO D- brouwen
and D.
triradlatus
20 -
Age estimates for Gephyrocapsa events were derived also

from DSDP site 132 by linear interpolation among the extinction
events for Pseudoemiliana lacunosa (Thierstein et al., 1977;
Raffi and Rio, 1979), C. macintyrei, and D. brouweri/D.
brouwen triradiatus (depth positions used for the three latter
events are those reported by this study). At DSDP site 132, the
end of dominance of small Gephyrocapsa spp. could not be
determined, because it occurs in an unrecovered interval (Raffi
and Rio, 1979). Recently, it has been widely observed in piston
cores from the Ionian Sea and Levantine Basin (Castradori,
1993). The ages obtained for the remaining Gephyrocapsa events
are in agreement with those obtained from the extraMediterranean sequences (Table 5.1).
The "northern guests" and the definition of the

Pliocene-Pleistocene boundary
The definition of the Pliocene-Pleistocene boundary (its assignment to a physical horizon) in the stratotype section had to be
such that, besides being amenable to long-distance correlation, it
would respect the historical concepts of the Pliocene and the
Pleistocene in order to maintain stratigraphic stability. As
discussed by Pelosio, Raffi, and Rio (1980), the PliocenePleistocene boundary traditionally had been placed at a level at
which marine faunal elements that at present are restricted to the
boreal bioprovince began to occur as "northern guests" in the
Mediterranean (specifically the Italian) geologic record. Indeed,
that concept was the main reason for a concentration of studies
of the Calabrian Stage (the traditionally accepted first stage of
Nannofossil biostratigraphy of the Italian surface
the Pleistocene) at Santa Maria di Catanzaro (Gignoux, 1913;
exposures
Selli, 1971), of the top of the Piacenzian (the generally accepted
A correlation chart for classical Italian land sections (including stage for the Upper Pliocene) at Castell'Arquato (Barbieri,
stratotypes) and DSDP site 132 is presented in Figure 5.5. This 1971; Colalongo, Elmi, and Sartoni, 1974), of the proposed
figure shows the critical biostratigraphic and lithostratigraphic Pliocene-Pleistocene boundary-stratotype at Le Castella (Pelocharacters on which the chronostratigraphy of the Pliocene and sio et al., 1980), and of the subsequently proposed boundarythe Pleistocene is based. Two topics that deserve attention are, stratotype in the Vrica section (Colalongo et al., 1982; Pasini and
first, the timing of the appearance of the "northern guests" in the Colalongo, Chapter 2, this volume). It is important to note that
Mediterranean and, second, the prior lithostratigraphic defini- Lyell's (1833) original concept of the Pleistocene, based on fossil
tions of the Pliocene-Pleistocene boundary.
molluscan faunas in Italian strata with no less than 70% of
67
AGE (Ma)
10
5 2
\
o
2:
760 - 060
\
End acme small \
\
j Oephyrocapsa spp.
6 2
2.0
o
2:
o
2:
1.10 - 1.1
1.5
M
\
Beginning acme small

25-
Gephyrocapsa spp.
\
o
I
a_
a
to
a
30 -
8 2
FA
Oephyrocapsa spp. > 5.5
Mm
X
hCL
UJ
Q
LO C. macintyrei
9 2
35-
FA
G. oceanica
si
40-
Figure 5.4. Sediment accumulation rate for

DSDP site 502B plotted against nannofossildatum chronology (uncorrected time scale).
species living in the region to this day, actually corresponds fairly

well to the situation at the time of arrival of the "northern
guests" in the Mediterranean (Pelosio et al., 1980).
The first faunal element that was proposed as a "northern
guest," and indeed the most "popular" one, was the shallowwater pelecypod Arctica islandica. That form today has specific
ecologic demands in that it thrives only in water depths less than
50-60 m, and consequently its fossils have only a limited

stratigraphic applicability. For that reason stratigraphers were
forced to use other criteria in order to recognize the base of the
Pleistocene in sections where A. islandica was missing. As a
substitute, they used benthic forms like the foraminifer Hyalinea
baltica (Trevisan and Di Napoli, 1938) and the ostracode
Cytheropteron testudo (Ruggieri, 1950,1973), which retained the
68
FICARAZZI
10-
DSDP Site 132
SANTA MARIA
di Catanzaro
m40
FA G.TRUNCATULINOIDES EXCELSA
45100-
"G-G' "
BED
40-
LE CASTELLA
50-
50-
m25 FA G,TRUNC.EXCELSA
15 -
CAPO ROSSELLO
250 -
M.te S.NICOLA
MARKER BED
H.B.
.60-
*G.P.
65-
, P/P BOUNDARY
DEFINITION '
200-
150 -
100-
Symbols
A - F A HYALINEA BALTICA
* - INCREASE OF G.PACHYDERMA "LEFT"
^ _ F A ARCTICA ISLANDICA
%-CCCURRENCE OF G . TRUNCATUL I NO IDES
TRUNCATULINOIDES
80 -|
50-
BEGINN
69
ING ACME SMALL GEPHYROCAPSA SPP.
SANTERNO
TIEPIDO
2000 -
CROSTOLO
500 AH.B.
1300
SA SPP.
STIRONE
1500-
'A.I.
r
CASTELL ARQUATO
1000-
100 -
50-
A.I.
'H.B.
800 A
"
100
A.I.
10 1000
0-
50
800 -
Figure 5.5. Correlation

of classic Italian land sections to DSDP site 132
(note key to abbreviations). FA, first
appearance.
70
Table 5.1. Locations and previous studies of sections utilized for Gephyrocapsa spp. biochronology
Water
depth
Section
Location
Lat.
Long.
(m)
Main studies
V28-239
Equatorial
Pacific
315'N
15911'E
3,490
DSDP hole 502B
Colombia Basin,
Caribbean Sea
Southwestern
Atlantic
Mediterranean,
Tyrrhenian Sea
1129'N
7923'W
3,051
Shackleton and Opdyke

(1977), Backman and Shackleton (1983)
Prell et al. (1982)
2842'S
3434'W
4,021
4015'N
1125'E
2,813
V12-18
DSDP site 132
Backman and Shackleton

(1983)
Ryan and Hsu (1973), Raffi
and Rio (1979)
"northern immigrant" concept. Evidence of climatic deteriora- Gephyrocapsa spp. are present, including transitional forms
tion, such as the coiling change of Neogloboquadrina pachy- indicating the first appearance of G. oceanica s.l. This suggests
derma (Dondi and Papetti, 1968) and the extinction of that the appearance of A. islandica is close to the top of the
Taxodiaceae (Lona, 1962), also have been used as substitutes for Olduvai subchron in this section. It is noteworthy that in both the
the recognition of the Pliocene-Pleistocene boundary. The Stirone and the Castell'Arquato sections the first appearance of
rationale underlying those choices was based on the misconcept A. islandica postdates the first occurrence of Globorotalia inflata.
that the earth passed from ice-free to glaciated conditions at the In Mediterranean sections the latter event occurs close to the first
Pliocene-Pleistocene boundary. Such a threshold in the earth's appearance of Globorotalia truncatulinoides truncatulinoides,
history ought to have been detectable in every environment and which has been determined to have occurred just below the
therefore should have been correlatable as a time horizon. Olduvai subchron both in the Mediterranean (Rio et al., 1984a,b)
However, recent work has shown that the climatic history is and in the equatorial oceans (Rio, Fornaciari, and Raffi, 1990;
much more complicated, and glacial conditions probably were Shackleton etal., 1995).
brought on through a series of climatic steps (e.g., Shackleton
The INQUA Subcommission 1-d, "Pliocene/Pleistocene
and Kennett, 1975; Thunell and Williams, 1983). The appear- Boundary," adopted the proposal that the base of the shale bed
ance of the "northern guests" in the Mediterranean during the overlying sapropelic layer e in the Vrica section should be used
Pleistocene was gradual and discontinuous, occurring at different for definition of the Pliocene-Pleistocene boundary. This level
times for different taxa. In particular, the appearances of A. is close to the top of the Olduvai subchron, according to
islandica and H. baltica did not represent a single event, and magnetostratigraphy and biostratigraphy (as discussed later),
since both were environmentally controlled, these forms do not suggesting that the boundary-definition proposal is appropriate
allow reliable correlations even within a small region. The in historical terms.
correlations that are based on planktic forms (Figure 5.5)
Prior to the introduction of the Vrica section as the location of
illustrate this point. Note the scattering of the appearances of A. the Pliocene-Pleistocene boundary-stratotype, two other definiislandica and H. baltica relative to the proposed correlations. It tions were in use: the base of the Calabrian Stage, provisionally
also appears from Figure 5.5 that the first appearance of H. defined in the Santa Maria di Catanzaro section (Selli, 1971),
baltica was clearly later than that of A. islandica. H. baltica and the so-called marker bed in the Le Castella section
occurs close to the appearance of large Gephyrocapsa spp. (Berggren and Van Couvering, 1974; Haq et al., 1977).
(larger than 5.5 /mi), whereas A. islandica occurs below the first
appearance of G. oceanica s. I.
The Santa Maria di Catanzaro section
If we want to adhere to the original concept that the base of the
Pleistocene is to be tied to the first appearance of A. islandica in Although there is a general agreement that the Santa Maria di
Italian sections, we need to locate, in a deeper-water boundary- Catanzaro section is unsuitable for definition of the base of the
stratotype section, a physical horizon that is time-equivalent with Pleistocene (Haq et al., 1977; Pelosio et al., 1980; Colalongo et
thefirstappearance of A. islandica in a shallow-water section. The al., 1982), this section needs some discussion because the
stratigraphically lowest level where A. islandica is present seems Calabrian Stage was first introduced (Gignoux, 1913) with
to be represented in the Castell'Arquato and the Stirone sections. reference to this section, among many others. It was this section,
No reliable nannofossil data are available from the former section. however, and in particular the G-G' bed shown in a diagram by
In the Stirone section, we have observed that the appearance of A. Gignoux (1913), that had been provisionally designated to
islandica slightly predated the first occurrence of G. oceanica s. I.represent the stratotype base of the Calabrian Stage (Selli, 1971;
Below the levels containing the first A. islandica, abundant cf. Berggren and Van Couvering, 1979). However, the G-G' bed
is well above the first Mediterranean occurrence of Globorotalia

truncatulinoides excelsa (Colalongo, Pasini, and Sartoni, 1981),
an event which occurred close to the beginning of dominance of
small Gephyrocapsa spp. and which also marks the base of the
Sicilian Stage (Di Stefano and Rio, 1981) or, preferably, substage
(Ruggieri, Rio, and Sprovieri, 1984). The Calabrian, as exemplified at Santa Maria di Catanzaro, is thus completely overstepped
by the Sicilian (Figure 5.5). Because the Sicilian was introduced
earlier (Doederlein, 1872), it follows that the Calabrian is a
junior synonym of the Sicilian and, as such, is an invalid stage
(Ruggieri and Sprovieri, 1977). Note also in Figure 5.5 how the
G-G' bed represents a chronohorizon well above the base of the
Pleistocene.
The Le Castella section
Many authors, according to the review by Rio et al. (1974), have
used a sandy level, named "marker bed" by Emiliani, Mayeda,
and Selli (1961), in the Le Castella section as defining the
Pliocene-Pleistocene boundary. This level is unsuitable as a
boundary definition because there is a hiatus immediately below
it. The "marker bed" level is a key horizon associated with the
first local occurrence of H. baltica, an event of somewhat
uncertain value for recognizing the base of the Pleistocene (as
discussed earlier). On biostratigraphic grounds, Haq et al. (1977)
considered that the "marker bed" level in the Le Castella section
is associated with the top of the Olduvai subchron and therefore
proposed that this geomagnetic reversal should be considered as
an index for the Pliocene-Pleistocene boundary. While this
association has since been confirmed at Vrica (as discussed
later), the data presented in Figure 5.5 and Table 5.3, together
with the nannofossil biochronology previously discussed (Table
5.2), indicate that the age of the Le Castella "marker bed" is
younger than 1.48-1.44 Ma, because of the presence of
Gephyrocapsa spp. larger than 5.5 /urn.
The Vrica section
71
cus. Secondary elements are represented by Cyclolithella annula,

Rhabdosphaera spp., Umbilicosphaera spp., Pontosphaera spp.,
and Scyphosphaera spp. The genus Discoaster is represented by
rare reworked Tertiary specimens, and by rare D. brouweri and
D. brouweri triradiatus in the lower part of the section. The
genus Ceratolithus is practically absent from the section, as is the
case in most Upper Pliocene and Lower Pleistocene sediments
within the Mediterranean area.
Of the previously discussed nannofossil biochronologic indications, all except the interval of dominantly small Gephyrocapsa
spp. are present in the Vrica section (Figure 5.6). Each of these
is briefly discussed here.
Last occurrence of Discoaster brouweri. Except for Nakagawa
and co-workers, all authors have considered the extinction of
D. brouweri in their range charts. No discrepancy is noted in
the positions of this datum event when the different sampling
intervals are taken into account. Combining the Raffi and Rio
data with those of Backman et al. (1983) results in a location
between 101 and 105 m for this datum event in the Vrica
section.
Discoaster brouweri triradiatus abundance interval. Backman et
al. (1983) provided counts of the abundance of this form in the
Vrica section and showed that its last occurrence coincides with
that of D. brouweri, in agreement with the findings of Backman
and Shackleton (1983) from extra-Mediterranean areas and the
findings from other Mediterranean sections (Rio, 1982). The
increase in abundance of D. brouweri triradiatus relative to D.
brouweri in the Vrica section begins between 20 and 35 m,
according to Raffi and Rio's sampling.
Last occurrence of Calcidiscus macintyrei. The findings of Raffi
and Rio, on one hand, and Backman and co-workers, on the
other, are in full agreement regarding the extinction level of C.
macintyrei (considering the different sampling intervals). Cati
and Borsetti suggest a slightly higher extinction level, and
Nakagawa and co-workers report C. macintyrei throughout the
Vrica section. These different ranges are best explained by
reworking, as demonstrated by Backman et al. (1983). By
combining the sampling levels of Backman and colleagues and
those of Raffi and Rio, the last occurrence of C. macintyrei in the
Vrica section can be located to 216 m.
Calcareous nannofossils have been studied in the Vrica section

by Nakagawa (1977, 1981), Nakagawa et al. (1980), Cati and
Borsetti (1980), Raffi and Rio (in Pasini and Calalongo, 1982),
and Backman et al. (1983). The sample levels studied and the
main findings reported by those authors are shown in Figure 5.6.
Although the nannofossil assemblages in the Vrica section are
variable in terms of abundance and preservation state, many
samples contain abundant nannofossils showing good or excel- Range of Helicosphaera sellii. All nannofossil biostratigraphers
lent preservation. Reworked input appears to have been except Cati and Borsetti consider H. sellii to be present
relatively stable (Backman and Shackleton, 1983), and the thoughout the Vrica section. Cati and Borsetti have suggested
extinction levels of Pleistocene species can be accurately that this species is absent from the lowermost and uppermost
identified through the use of quantitative data and by taking into parts of the section. The reason for this discrepancy in
account the backgound "noise" of reworking.
observation is unknown. Neverthless, we consider that the top of
The most abundant elements of the assemblages are repre- the Vrica section is below the regional last occurrence of H.
sented by Gephyrocapsa spp., Pseudoemiliana lacunosa (with sellii. In other Mediterranean sections the disappearance of H.
both circular and elliptical morphotypes), Calcidiscus spp., sellii occurs together with the temporary disappearance of
Syracosphaera spp., Helicosphaera spp., and Coccolithuspelagi-normal-sized Gephyrocapsa spp. (which is the beginning of the
72
Table 5.2. Sections studied in this chapter
Section
Field
stratigraphic
information
Nannofossil
data
Other
information
Castell'Arquato
(Piacenza)
Pelosio et al. (1980)
Stirone
(Parma)
Bertolani Marchetti et al.

(1979)
Pelosio et al. (1980),

Rio et al. (this chapter)
Crostolo
(Reggio Emilia)
Barbieri and Petrucci

(1967)
Raffi and Rio (1980a)
Tiepido
(Modena)
Annovi et al. (1979)
Raffi and
Rio (1980b), Rio et al.
(this chapter)
Santerno
(Bologna)
Colalongo et al. (1974)
Raffi and Rio (1980c),

Rio et al. (this chapter)
Monte San Nicola

(southern Sicily)
Rio et al. (1984a)
Rio et al. (1984a)
Capo Rossello
(southern Sicily)
Rio et al. (1984b)
Rio et al. (1984b)
Rossellian superstage stratotype section (Cita

and Decima, 1975)
Ficarazzi
(Palermo, Sicily)
Di Stefano and Rio

(1981)
Di Stefano and Rio

(1981)
Sicilian stratotype section (Ruggieri and

Sprovieri, 1977), proposed Lower-Middle
Pleistocene boundary-stratotype section
(Ruggieri etal., 1984)
Vrica
(Calabria)
Selli et al. (1977)
Raffi and Rio (in Pasini

and Colalongo, 1982),
Backman and Shackleton
(1983)
Pliocene-Pleistocene (P/P) boundarystratotype section
Le Castella
(Calabria)
Raffi and Rio (1980d),

Raffi and Rio (1980d)
Proposed P/P boundary-stratotype section

(Pelosio etal., 1980)
Santa Maria di Catanzaro

(Calabria)
Piacenzian stratotype section (Barbieri, 1971)
Santernian and Emilian stratotype sections

(Ruggieri and Sprovieri, 1977)
Calabrian stratotype section
interval of dominantly small Gephyrocapsa spp.) and the ric) taxonomy of Rio (1982). It should be emphasized that we
appearance of Globorotalia truncatulinoides excelsa (Raffi and have applied this informal taxonomy on all material studied, thus
Rio, 1979; Di Stefano and Rio, 1981). Neither of these two avoiding taxonomy and nomenclature problems when evaluating
events is recorded in the Vrica section.
the biochronology of the datum events provided by the
Gephyrocapsa group.
Gephyrocapsa group. Range charts for the Gephyrocapsa group
have been used by all authors who have studied the Vrica section
Nannofossil biochronology of the Vrica section and
except Backman et al. (1983). Identifications in this group have
the age of the proposed Pliocene-Pleistocene
been unstable because of taxonomy and nomenclature problems,
boundary-stratotype
and thus it is not surprising that conflicting findings have been
reported by different authors. In order to apply consistent The INQUA Subcommission 1-d, "Pliocene/Pleistocene Boundtaxonomic concepts, we have adopted the informal (morphomet- ary," meeting in Madrid in 1983, proposed that the base of the
73
Table 5.3. Summary offield stratigraphic and biostratigraphic studies used in the compilation of Figure 5.5
Age estimates (Ma) for nannofossil datum events around the Plio-Pleistocene boundary
Datum event
End of acme of small
Gephyrocapsa spp.
Beginning of acme of small
Gephyrocapsa spp.
First appearance of
Gephyrocapsa spp. > 5.5 pin
First appearance of
Gephyrocapsa oceanica s.l.
Last occurrence of
Calcidius macintyrei
Last occurrence of
Discoaster brouweri
Last occurrence of
D. brouweri var. triradiatus
Increase in proportion of
D. brouweri var. triradiatus
V12-18
(southwestern
Atlantic)
DSDP 132
(Mediterranean
Sea)
V28-239 (western
eq. Pacific)
DSDP 502B
(Caribbean Sea)
0.93-0.95
0.90-0.94
1.13-1.15
1.10-1.11
1.31-1.32
1.32-1.34
1.29-1.31
1.32-1.36
1.57-1.61
1.55-1.56
1.55-1.59
1.56-1.62
1.08-1.14
Synchronous worldwide: 1.45-1.46 Ma+

Synchronous in low-latitude and midlatitude oceans, and high latitudes in North Atlantic:
1.89 Ma (Backman and Shackleton, 1983)
Synchronous in low-latitude and midlatitude oceans, and high latitudes in North Atlantic:
189 Ma (Backman and Shackleton, 1983)
Occurs in all Discoaster-bearing sediments: 2.0-2.1 Ma (Backman and Shackleton, 1983)
stratum overlying layer e in the Vrica section should be used for

definition of the Pliocene-Pleistocene boundary. As noted by
Nikiforova and Alekseev (Chapter 1, this volume), that
proposal has now been adopted by the IUGS authorities. This
level is only 8 m below the level where the cold-water ostracode
Cytheropteron testudo is first recorded, which previously was
proposed as a boundary definition (Colalongo et al., 1982; cf.
Pasini and Colalongo, Chapter 2, this volume). Considering
the high rates of sediment accumulation in the Vrica section,
the two biochronohorizons are virtually indistinguishable (Figure 5.7).
The last occurrence of D. brouweri (and D. brouweri
triradiatus) is associated with the bottom of the Olduvai
subchron, as in all low-latitude and mid-latitude deep-sea cores.
The first appearance of G. oceanica s.l. occurs immediately
above the Olduvai subchron, which is closely followed by the
extinction of C. macintyrei and the first appearance of
Gephyrocapsa spp. larger than 5.5 /xm. The sequencing of these
datum events and their estimated ages, in terms of sedimentation rates relative to the magnetostratigraphy at Vrica, correspond precisely with the findings from the Pacific, the Caribbean, and the Atlantic (Shipboard Scientific Party, 1992), giving
the marker bed e an unarguable biostratigraphic, biochronologic, and magnetostratigraphic context. From this context, the
proposed Pliocene-Pleistocene boundary can be accurately
identified in both marine and continental geologic records,
within the Mediterranean as well as in extra-Mediterranean
areas.
Conclusions
Several calcareous nannofossil datum events in Upper Pliocene

and Lower Pleistocene marine sediments are well documented
with regard to the magneostratigraphic time scale, from which
their ages are estimated. All datum events observed in the
investigated material occur without notable diachronism from
the global standard.
A cursory investigation of the biostratigraphy and lithostratigraphy of the proposed Pliocene-Pleistocene boundary in the
Vrica section affirms that it is historically appropriate in that it
respects the concepts of the Pliocene and the Pleistocene as
originally conceived by Lyell (1833).
As presently defined, the Calabrian Stage is a junior synonym
for the Sicilian, and furthermore its base is unsuitable for
definition of the base of the Pleistocene, according to the
paleoclimatic concept that has guided the international effort
begun at the London (1948) International Geological Congress.
The "marker bed" in the Le Castella section that had been
proposed previously to be the definition of the PliocenePleistocene boundary (Haq et al., 1977; Pelosio et al., 1980)
cannot be used for this purpose because there is a hiatus
immediately below it. Moreover, the only key event associated
with the "marker bed" is the first (apparent) local occurrence of
Hyalinea baltica, which is an unreliable and biostratigraphically
variable event. Calcareous nannofossils occur abundantly in the
proposed Vrica stratotype section, and all critical Upper
Pliocene and Lower Pleistocene nannofossil datum events that
II
1 11 1 1 1 1 11
BROUWERI
1 11
1
1
1
CJi
DISCOASTER
C 0 CC 0 L I T HU S
BROUWERI
CN 12a
MACINTYREI
CALCIDISCUS
DISCOASTER
CN 12
BROUWERI
DISCOASTER
NN 18
BROUWERI
DISCOASTER
J_ G . I N F L A T A
CJ
1 II 1
II
Q.
II 1
(t>
o
o
H
m
CO
'
II
HIM
fl
1 1 1
>
DO
>
SELLII
to
O
1 1 1 11
11 1
1971
DTH
1070
ly/y
RAFFI AND RIO
CATI AND BORSETTI
NAKAGAWA ET A L .
BACKMAN ET AL,
RAFFI AND RIO
CATI AND BORSETTI
BACKMAN ET A L .
AND RIO
CATI AND BORSETTI
NAKAGAWA ET AL.
L I TH 0 L0 G Y
AIIO
1977
KArrl AND KIU
DACCT
GARTNER
OKADA AND BUKRY 1980
MARTINI
i I I I ii mm Him i i i inn M I i ii 111 II RAFFI
1 1
<J
.; o
rvo
CJi
HELICOSPHAERA
ro
O
C.mCINT.
SELLII
LACUNOSA
HELICOPONTOSPHAERA
ZED
CN 14
N N 19
OCEAN ICA
PSEUDOEMILIANIA
G N
+ GEPHYROCAPSA
L AC UN 0 SA
R E C 0
I I i II ii
I M i
urn
1 1
1 1
1 III II II II
O)
MACINTYRE
NOT
DORONICOIDES
cus
CRENALITHUS
P ELA G
DORONICOIDES CN 13
CYCLOCOCCOLITHINA
CRENALITHUS
P S E U D O E M I L I A N I
CO
rn
-o
CD
O
GO
DO
CL
o'
HELICOSPHAERA
SELLII
GEPHYROCAPSA
GROUP
RAFFI AND RIO NAKAGAWA ET AL CATI AND BORSETTI
Figure 5.6. Summary of nannofossil biostratigraphy in the

Vrica section.
75
76
AGE (Ma)
1.5
1.3
1.1
OLDUVAI
300\
> UL c m / k y r
N3
\
250-
\
FA Gep'hyrocoipsa spp. > 5.5 urn
o
v cm/kyr->s^^18 cm/kyr
LU
L0 C. macintyrei
200 LU
FA 0. oceanic a s.l.
TOP OF LAYER e
N2
CL
LU
Q
150-
25
N1
Figure 5.7. Biochronology of the

Pliocene-Pleistocene boundary in
the Vrica section (uncorrected time
scale).
L0
D. brouweri
100-
can be used as biochronologic indications are present. By

combining these with the established magnetostratigraphy of
the section, the Pliocene-Pleistocene boundary in the Vrica stratotype can be correlated to most other marine sections by
means of biostratigraphy, biochronology, and magnetostratigraphy. Magnetostratigraphy also allows correlation to continental biochronology.
Italian and Mediterranean sections was provided through various

CNR grants during the years 1978-1982 to D. Rio and through
MPI grants during 1982-1984 to G. Pelosio. Financial support
for the work of J. Backman was provided by the Swedish Natural
Science Research Council.
References
Akers, W. H. 1965. Pliocene-Pleistocene boundary, northern

Gulf of Mexico. Science 149:741-742.
We express our thanks to K. V. Nikiforova, leader of IGCP Backman, J., Roberts, D. G., Schnitker, D., et al. 1984.
Cenozoic calcareous nannofossil biostratigraphy from the
Project 41, "Neogene-Quaternary Boundary," who invited us to
northeastern Atlantic Ocean - Deep Sea Drilling Project
contribute this chapter. We are grateful to G. Pasini for providing
Leg 81. In Initial Reports of the Deep Sea Drilling Project,
samples from the Vrica section and for his help and cooperation
vol. 81, pp. 403-428. Washington, DC: U.S. Government
during discussions on the problems of Pliocene and Pleistocene
Printing Office.
stratigraphy. Many of the findings presented here resulted from Backman, J., and Shackleton, N. J. 1983. Quantitative biocooperative projects with our colleagues R. Sprovieri, M. L.
chronology of Pliocene and Early Pleistocene calcareous
nannoplankton from the Altantic, Indian and Pacific
Colalongo, and S. Sartoni, to all of whom we express our sincere
Oceans. Mar. Micropal. 8:141-170.
thanks. We are obliged to the core curators at the DSDP facilities
Backman, J., Shackleton, N. I , andTauxe, L. 1983. Quantitative
at La Jolla (W. R. Reidel) and at Lamont-Doherty Geological
Observatory, New York (F. McCoy), for their assistance.
from Plio-Pleistocene boundary at Vrica, Italy. Nature
Financial support for nannofossil biostratigraphic studies in
304:156-158.
Acknowledgments
77
Barbieri, F. 1971. Piacenzian. In Stratotypes of Mediterranean Dondi, L., and Papetti, I. 1968. Biostratigraphical zones of the
Neogene Stages, ed. G. C. Carloni et al., pp. 147-155.
Po Valley Pliocene. Giorn. Geol. 25:63-68.
Emiliani, C , Mayeda, T., and Selli, R. 1961. Paleotemperature
Giornale di Geologia 37, estratto, fasc. II.
analysis of the Plio-Pleistocene section at Le Castella,
Berggren, W. A. 1972. A Cenozoic time scale, some implications
Calabria, southern Italy. Geol. Soc. Am., Bull. 72:679for regional geology and paleobiogeography. Lethaia
5:195-215.
688.
Berggren, W. A., and Van Couvering, J. A. 1974. The Late Ericson, D. B., Ewing, M., and Wollin, G. 1963. PlioPleistocene boundary in deep-sea sediments. Science
Neogene. Palaeogeogr. Palaeoclimatol. Palaeoecol. 16:1
139:727-737.
215.
Berggren, W. A., and Van Couvering, J. A. 1979. The Gartner, S. 1969. Correlation of Neogene planktonic foraminifera and calcareous nannofossil zones. Gulf Coast Assoc.
Quaternary. In Treatise on Invertebrate Paleontology, Part
A: Introduction, ed. R. A. Robison and C. Teichert, pp.
Geol. Soc, Trans. 19:585-599.
Gartner, S. 1977. Calcareous nannofossil biostratigraphy and
revised zonation of the Pleistocene. Mar. Micropal. 2:1Berggren, W. A., Kent, D. V, and Van Couvering, J. A. 1985.
25.
chronology of the geological record, ed. N. J. Snelling, pp. Gignoux, M. 1913. Les formations marines pliocenes et
211-260. Geol. Soc. London, Mem. 10. Oxford: Blackwell.
quaternaires dTtalie du Sud et de la Sicile. Univ. Lyon,
Bertolani Marchetti, D., Accorsi, C. A., Pelosio, G., and Raff],
Ann., n.s. 36: 1-693.
S. 1979. Palynology and stratigraphy of the Plio-Pleiso- Haq, B. U., Berggren, W. A., and Van Couvering, J. A. 1977.
tocene sequence of the Stirone river (Northern Italy).
Pollen et Spores 21:149-167.
269:483-488.
Boudreaux, J. E., and Hay, W. W. 1967. Zonation of the latest Hay, W. W., Mohler, H. P., Roth, P. H., Schmidt, R. R., and
Pliocene-Recent interval. Gulf Coast Assoc. Geol. Soc,
Boudreaux, J. E. 1967. Calcareous nannoplankton zonaTrans. 17:420-443.
tion in the Cenozoic of the Gulf Coast and CaribbeanAntillean area and transoceanic correlation. Gulf Coast
Boudreaux, J. E., and Hay, W. W. 1969. Calcareous nannoAssoc. Geol. Soc, Trans. 17:438-480.
plankton and biostratigraphy of the Late PliocenePleistocene sediments in Submarex cores. Rev. Esp. Lona, F. 1962. Prime analisi pollinologiche sui depositi terziari
quaternari di Castell'Arquato; reperti di vegetazione da
Micropal. 1:249-262.
clima freddo sotto le formazioni calcarea ad Amphistega.
Bramlette, M. N., and Riedel, W. R. 1954. Stratigraphic value of
Soc Geol. Ital, Boll 81:89-91.
discoasters and some other microfossils related to recent
Lyell, C. 1833. Principles of geology: being an attempt to explain
coccolithophores. /. Paleontol. 28:385-403.
Bukry, D. 1973. Low latitude coccolith biostratigraphic
the former changes in the Earth's surface, by reference to
causes now in operation, vol. 3. London: John Murray.
zonation. In Initial Reports of the Deep Sea Drilling
Project, vol. 15, ed. N. T. Edgar, J. B. Saunders, et al., Martini, E. 1971. Standard Tertiary and Quaternary calcareous
pp. 685-703. Washington, DC: U.S. Government Printnannoplankton zonation. In Proceeding of the 2nd Planking Office.
tonic Conference, Roma, vol. 2, ed. A. Farinacci, pp. 739Bukry, D. 1978. Biostratigraphy of Cenozoic marine sediments
785. Rome: Ediziones Tecnoscienza.
by calcareous nannofossils. Micropaleontology 24:44-60. Mclntyre, A., Be, A. W. H., and Pretiskas, R. 1967. Coccoliths
Castradori, D. 1993. Calcareous nannofossil biostratigraphy and
and the Plio-Pleistocene boundary. In The Quaternary
history of the ocean basins. Progress in oceanography, vol.
biochronology in eastern Mediterranean deep-sea cores.
Riv. Ital. Paleontol. Strat. 99:107-126.
4, ed. M. Sears, pp. 3-24. Oxford: Pergamon Press.
Cati, R, and Borsetti, A. M. 1980. Calcareous nannoplankton Nakagawa, H. 1977. Preliminary report on the magneto- and
biostratigraphy of the Vrica section (Calabria, southern
biostratigraphy of the Vrica section, Calabria, Southern
Italy. Preprint, X INQUA Congress.
Italy). Giorn. Geol. 43:365-384.
Cita, M. B., and Decima, A. 1975. Rossellian: proposal of Nakagawa, H. 1981. Neogene/Quaternary boundary and correlation of Vrica Section. In Field conference, Neogene/
superstage for the marine Pliocene. In VI Congress of the
Quaternary boundary, India, 1979, Proceedings, ed. M. V
Regional Committee for Mediterranean Neogene StratiA. Sastry et al., pp. 107-111. Calcutta: Geological Survey
graphy, Bratislava, 1975, Proceedings, vol. 1, ed. J. Senes
of India.
and J. Slavik, pp. 217-227. Bratislava: Akad. Vied.
Colalongo, M. L., Elmi, C , and Sartoni, S. 1974. Stratotypes of Nakagawa, H., Oda, M., Sakai, T., Yoshida, K., Asano, K.,
Niitsuma, N., Takayama, T., Tokunaga, S., Kitazato, H.,
Pliocene and Santerno River section. Bur. Rech. Geol.
and Koizumi, I. 1980. Preliminary results of magneto- and
Min., Mem. 78:603-624.
Italy). In Proceedings of the Second Symposium on the
Neogene/'Quaternary Boundary, USSR, 1977, ed. K. V
Nikiforova and A. Y. Dodonov, pp. 145-156. Moscow:
and a proposal. Geogr. Fisica Dinam. Quat. 5:59-68.
Akademia Nauk.
the Neogene/Quaternary boundary and the Vrica section Pasini, G., and Colalongo, M. L. 1982. Status of research on the
(Calabria, Italy). Soc. Paleontol. Ital, Boll. 20:99-120.
Di Stefano, E., and Rio, D. 1981. Biostratigrafia a nannofossili e
presented at INQUA XI Congress, Moscow, 1982. Bolobiocronologia del Siciliano nella localita tipo di Ficarazzi
(Palermo, Sicilia). Ateneo Parmense, Acta Nat. 17:97-111.
Doederlein, P. 1872. Note illustrative della carta geologica del Pelosio, G., Raffi, S., and Rio, D. 1980. The Plio/Pleistocene
boundary controversy: status in 1979 at the light of
Modenense e del Reggiano. Menoria tersa. Modena:
International Stratigraphic Guide. In Volume speciale in
Gaddi.
78
memoria di Sergio Venzo, ed. F. Petrucci and R. Cavazzini,

nean upper Pliocene geologic record. Micropaleontology
30:121-137.
Prell, W. 1982. Oxygen and carbon isotope stratigraphy for the Rio, D., Sprovieri, R., and Raffi, I. 1984b. Calcareous plankton
biostratigraphy and biochronology of the Pliocene-Lower
Quaternary of Hole 502B: evidence for two modes of
isotopic variability. In Initial Reports of the Deep Sea
Drilling Project, vol. 68, ed. W. L. Prell, J. V. Gardner, et
al., pp. 465-474. Washington, DC: U.S. Government Roth, P. H. 1978. Cretaceous nannoplankton biostratigraphy
and oceanography of the northwestern Atlantic Ocean. In
Printing Office.
Initial Reports of the Deep Sea Drilling Project, vol. 44, ed.
Prell, W. L., Gardner, J. V., et al. (eds.) 1982. Initial Reports of
the Deep Sea Drilling Project, vol. 68. Washington DC:
W E. Benson, R. E. Sheridan, et al., pp. 731-758.
U.S. Government Printing Office.
Raffi, I., and Rio, D. 1979. Calcareous nannofossil bio- Ruggieri, G. 1950. Gli ultimi capitoli della storia geologica della
stratigraphy of DSDP site 132-Leg 13 (Tyrrhenian SeaRomagna. Studi Romagnoli 1:303-311.
Western Mediterranean). Riv. Ital. Paleontol. Strat. Ruggieri, G. 1973. Gli ostracodi e la stratigrafia del Pleistocene
marino Mediterraneo. Soc. Geol. Ital., Boll. 92:213-232.
85:127-172.
Raffi, I., and Rio, D. 1980a. II limite Plio-Pleistocene nella serie Ruggieri, G., Rio, D., and Sprovieri, R. 1984. Remarks on the
della Torrente Crostolo-Torrente Vendino (Provincia di
Reggio Emilia-Italia Settentrionale). In Paleontologia,
Soc. Paleontol. Ital., Boll. 53:251-259.
Stratigrafia, Evoluzione, vol. 1, ed. C. Petronio, pp. 165- Ruggieri, G., and Sprovieri, R. 1977. A revision of Italian
167. Roma: Consiglio Nazionale della Recherche.
Raffi, I., and Rio, D. 1980b. Biostratigrafia a nannofossili Ryan, W. B. F , and Hsu, K. J. (eds.) 1973. Initial Reports of the
Deep Sea Drilling Project, vol. 13, no. 2. Washington, DC:
calcarei, biocronologia, e cronostratigrafia della serie del
Torrente Tiepido (Subappenino romagnolo, Provincia di
U.S. Government Printing Office.
Modena). Ateneo Parmense, Acta Nat. 16:19-31.
Raffi, I., and Rio, D. 1980c. Nannoplancton calcareo del
Pliocene superiore-Pleistocene inferiore della serie del
Giorn. Geol. 37, estratto, fasc. II.
flume Santerno. In Paleontologia, Stratigrafia, Evolu- Selli, R., Accorsi, C. A., Bandini Mazzanti, M., Bertolani
zione, vol. 1, ed. C. Petronio, pp. 187-195. Roma:
Consiglio Nazionale della Recherche.
Raffi, L, and Rio, D. 1980d. Nuove osservazioni sulla disLandini, W., Menesini, E., Mezzetti, R., Pasini, G.,
tribuzione del nannoplancton calcareo nella serie di Le
Castella (Calabria). In Paleontologia, Stratigrafia, Evolu(Calabria-Italy). A potential Neogene/Quaternary boundzione, vol. 1, ed. C. Petronio, pp. 181-186. Roma:
Shackleton, N. J., Baldauf, J. G., Flores, J.-A., Iwai, M., Moore,
T. C , Raffi, I., and Vincent, E. 1995. Biostratigraphic
Riedel, W. R., Bramlette, M. N., and Parker, F. L. 1963.
summary for Leg 138. In Proceedings of the Ocean Drilling
"Pliocene-Pleistocene" boundary in deep-sea sediments.
Program, Scientific Results, Leg 138, ed. N. G. Pisias et
Science 140:1238-1240.
Rio, D. 1974. Remarks on Late Pliocene-Early Pleistocene
Program.
calcareous nannofossil stratigraphy in Italy. Ateneo Parmense, Acta Nat. 10:409-449.
Shackleton, N. J., and Opdyke, N. D. 1977. Oxygen isotopes and
Rio, D. 1982. The fossil distribution of coccolithophore genus
paleomagnetic evidence for early northern hemisphere
Gephyrocapsa Kamptner and related Plio-Pleistocene
glaciation. Nature 270:216-219.
chronostratigraphic problems. In Initial Reports of the Shipboard Scientific Party. 1992. Explanatory notes. In Proceedings of the Ocean Drilling Program: initial reports, vol.
Deep Sea Drilling Project, vol. 68, ed. W. L. Prell, J. V.
Gardner, et al., pp. 325-343. Washington, DC: U.S.
138, part 1, ed. L. Mayer et al., pp. 175-235. College
Station, TX: Ocean Drilling Program.
Rio, D., Fornaciari, E., and Raffi, I. 1990. Late Oligocene Thierstein, H. R. 1976. Mesozoic calcareous nannoplankton
through early Pleistocene calcareous nannofossils from
biostratigraphy of marine sediments. Mar. Micropal.
1:325-362.
western equatorial Indian Ocean (Leg 115). In Proceedings of the Ocean Drilling Program: scientific results, vol. Thierstein, H. R., Geitzenauer, K. R., Molfino, B., and
Shackleton, N. J. 1977. Global synchroneity of Late
115, ed. R. A. Duncan et al., pp. 175-235. College
Quaternary coccolith datum levels: validation by oxygen
Station, TX: Ocean Drilling Program.
isotopes. Geology 5:400-404.
Rio, D., Palmieri, G., Raffi, I., and Villa, G. 1982. Classificazione biostratigrafica dei sedimenti marini Plio-Pleisto- Thunell, R. C , and Williams, D. F. 1983. The stepwise
cenici del bacino Padano basata sul nannoplancton
calcareo: correlazione con le biozone a Foraminiferi e con
paleoceanographic conditions in the Mediterranean: oxyla cronostratigrafia. In Guida alia geologia del margine
gen isotopic studies of DSDP sites 125 and 132. Utrecht
appenninicopadano, ed. G. Cremonini and F. R. Lucchi,
pp. 123-129. Roma: Societa Geologica Italiana.
Trevisan, L., and Di Napoli, E. 1938. Siciliano e Calabriano
Rio, D., Sprovieri, R., Di Stefano, E., and Raffi, I. 1984a.
nella Sicilia sud-occidentale. Giorn. Sci. Nat. Econ.
Globorotalia truncatulinoides (d'Orbigny) in the MediterraPalermo 39:1-39 (Mem. 8).
Diatom microfossils from the Vrica section

ANDREW! M. PALMER
Introduction
The marine strata exposed at Vrica, in Calabria, were proposed

by Selli et al. (1977) as the Pliocene-Pleistocene boundarystratotype section because those strata, originally deposited at
depths between 500 and 800 m, offered the potential for
correlation of oceanic microfossils with established datum levels.
Backman, Shackleton, and Tauxe (1983) and Rio, Raffi, and
Backman (Chapter 5, this volume) correlated the calcareous
nannofossils from Vrica to those recovered from deep-sea piston
core V28-239, taken at a depth of 3,490 m on the Solomon Rise
in the western equatorial Pacific. In that biostratigraphy,
supported by observations from other cores, the first appearance
of Gephyrocapsa s.l. just above the Olduvai subchronozone and
the global extinction of Calcidiscus macintyrei at a slightly higher
level are the closest available approximations of the boundary,
which is located within the uppermost part of the Olduvai at
approximately 1.8 Ma (Pasini and Colalongo, Chapter 2, this
volume), and not above the subchron, according to the data of
Tauxe et al. (1983) (Figure 6.1).
Diatom assemblages of Plio-Pleistocene age have been
reported from deep-sea sites around the globe, with emphasis on
the Pacific Ocean (Burckle and Todd, 1976; Barron, 1980a,b;
Koizumi, 1985). More detailed studies in the northwest Pacific
(Koizumi, 1985, 1992), Kamchatka (Gladenkov, 1994), northeast and southern Pacific (Schrader, 1973, 1976), and the
Southern Ocean (McCollum, 1975; Gombos, 1976) support the
regional analysis. Research by Burckle and co-workers (e.g.,
Saito, Burckle, and Hays, 1975; Burckle and Opdyke, 1977;
Burckle and Trainer, 1979) was based on integration of diatom
datum levels and magnetostratigraphy. Burckle and Trainer
(1979) noted four features in central Pacific diatom assemblages
(Figure 6.2) which have proved to be of widespread significance
(Shackleton et al., 1995a,b). These are as follows:
praebergonii and the first appearance of Pseudoeunotia

doliolus
4. the last appearance of Rhizosolenia praebergonii var.
robusta just above the Olduvai subchronozone
In the study reported here, I examined sediment samples from
the Vrica section for diatom assemblages that might be useful in
correlating onshore (marine) sediments with the plankton
stratigraphy established in offshore (marine) sediments, following the example set by Koizumi, Barron, and Harper (1980).
Sample material
Twenty-four sediment samples from three laminated sapropelic

units (e, /, and h) (Raffi and Thunell, Chapter 4, this volume) of
the Vrica section (Figure 6.3) were prepared for diatom analysis.
The material was dispersed in hydrogen peroxide and dilute
hydrochloric acid, washed by a simple settling technique to
minimize loss of any opal, and subsequently examined in a highrefractive-index mounting medium under oil immersion at
1,000 x. Two samples were found to be diatomaceous, both from
level e. The lower level was distinguished here as e5, and the
upper as e6. Fecal pellets were not common, and the samples
were not difficult to disaggregate. Backman et al. (1983) refer to
evidence that "the whole section is marked by very extensive . . . influx of [reworked] Palaeogene and Cretaceous nannofossils." Diatoms are subject to the same general taphonomic
constraints as calcareous nannofossils, and the possibility of
redeposited specimens was kept in mind during this study.
Observations of assemblage composition
Two sets of observations were made: a quantitative count of 186

specimens from sample e6 in a larger population randomly
1. the marked decline of Coscinodiscus nodulifer var. mounted upon a microscope slide, and a qualitative examination
of correlative samples mounted on SEM (scanning electron
cyclopus starting below the Olduvai subchronozone
2. the consistent abundance of Rhizosolenia praebergonii microscope) stubs. Under light microscopy, fourteen taxa plus
var. robusta beginning just below the base of the undifferentiated spores and "unknowns" collectively constituted
more than 95% of the siliceous micro fossil sample, as indicated
Olduvai subchronozone
3. within the Olduvai, the last appearance of Rhizosolenia in Table 6.1. All of the fossil diatoms belonged to species living in
79
80
Andrew J. M. Palmer
^mi&Mi:
propoctd Plio-PUistocene
boundary stratotype
Figure 6.1. Geologic setting of the

Vrica section: (a) location map
and (b) stratigraphy, from
Colalongo et al. (1982); (c) magnetostratigraphy of the Vrica section, from Tauxe et al. (1983).
The boundary proposed in 1982 is
shown at the first appearance of
Cytheropteron testudo, whereas
the boundary that was finally
adopted is at the top of level e.
The N1-N2 normal-polarity intervals are assigned to the Olduvai
event (Tauxe et al., 1983;
Nakagawa et al., Chapter 3, this
volume).
modern marine environments, as noted in the table (Schrader

and Matherne, 1981). Specimens of the colonial forms Fragilaria
and Melosira probably originated in coastal environments and
were transported offshore; their recovery in samples dominated
by planktonic diatoms was not surprising (Hendey, 1964). The
recovery of the colonial diatoms was remarkable, however, in
that they were observed as intact chains, whereas adjacent
samples were barren of any diatoms. Furthermore, there was no
R2
evidence of reworked pre-late Cenozoic diatoms in the prepared

sample. Unfortunately, however, none of the taxa that have been
considered here as key, magnetostratigraphically correlated
markers were found. Several palynomorphs (including Pinus
spp.) were observed, consistent with the details discussed by Selli
et al. (1977).
SEM micrographs revealed that some specimens had suffered
solution along exposed edges, particularly Cosdnodiscus. While
Diatoms of the Vrica section
8l- 0
Rhizosolenia
praebergonii
POB
*4O
30
Rhizosolenia proebergonn
vor robusta
10%
10%
81
Figure 6.2. Integrated Plio-PIeistocene magnetostratigraphy and marine diatom stratigraphy from the equatorial Pacific. Note the level of
Cosanodiscus nodulifer
vor cyclopus
Pseudoeunotia do/io/us
Nitzschia
the Olduvai normal polarities, marked with the letter O on the depth
scale. (From Burckle and Trainer, 1979, with permission.)
Table 6.1. Major and minor components of a diatom assemblage (n = 186) from Vrica level e6
Taxon
Trophic Group
Count
Percentage 0
Coscinodiscus obscurus
Coscinodiscus radiatus
(spores, undifferentiated)
Rhizosolenia hebetata f.
semispina
Thalassiosira eccentrica
Coscinodiscus oculus-iridis
Coscinodiscus normani
Melosira sp. (chains in girdle
view)
Fragilaria sp.
Thalassiothrix mediterraneana
Nitzschia sp.
Coscinodiscus stellaris
Actinoptychus senarius
(unknowns)
Thalassiosira lineata
Hemidiscus cuneiformis
ocean plankton
ocean plankton
plankton
temperate plankton
62
23
18
15
33.3
12.4
9.7
8.1
6.8%
4.8%
4.3%
3.9%
neritic plankton
ocean plankton
plankton
sessile**
12
6
6
6
6.5
3.2
3.2
3.2
3.5%
2.5%
2.5%
2.5%
neritic plankton
neritic plankton
ocean plankton
ocean plankton
neritic plankton
5
5
5
4
4
4
3
3
2.7
2.7
2.7
2.2
2.2
2.2
1.6
1.6
2.3%
2.3%
2.3%
2.1%
2.1%
2.1%
1.8%
1.8%
fl
plankton
ocean plankton
The calculated percentage is indicated at the 95% level of confidence.

Melosira chains also found in plankton.
Source: Adapted from Galehouse (1971).
jouseoe
20%
82
Andrew J. M. Palmer
No.
-6
5
4
2 00
197'
-2
-1
19 3'
4
-3
191
*17 0<
22), undifferentiated spores, Rhizosolenia hebetata forma semispina, and Thalassiosira eccentrica. Other diatoms noted are
Coscinodiscus asteromphalus (cf. Hustedt, 1971, figure 250;
Hendey, 1964, pi. 24) and Coscinodiscus normani (cf. Hendey,
1964, p. 80; Hustedt, 1971, as C. rothi var. normani, figure 213;
Schmidt et al., 1972, as Coscinodiscus fasciatus). The assemblage
does not appear to contain any of the equatorial Pacific taxa
discussed by Burckle and Trainer (1979), such as Coscinodiscus
nodulifer var. cyclopus, Rhizosolenia praebergonii (with the
variety robusta), and Pseudoeunotia doliolus. Correlation between the diatom biostratigraphy associated with the Olduvai
subchron in Pacific Ocean cores and that of the Vrica sample e6
is therefore impossible. The diatom assemblages in the deep-sea
sapropels of the eastern Mediterranean examined by Schrader
and Matherne (1981) were dominated by such taxa as Rhizosolenia calcaravis, Thalassionema nitzschioides, and Thalassiosira
oestrupii. Again, there was a poor match between those
elaborately quantified diatom assemblages and the diatoms
recovered from the laminates of the Vrica section.
Conclusions
Although diatom biostratigraphic correlation of the Vrica

section with other Plio-Pleistocene deep-sea sequences is
severely limited by the absence of key taxa, some paleoenvironmental inferences are possible. At modern rates of surface
production, diatoms are rarely preserved in the bottom sedi167
ments of the Mediterranean (Schrader and Matherne, 1981),
apparently because the water is not sufficiently close to
saturation with respect to opaline silica. It can be inferred that
similar conditions prevailed during much of the time represented
laminated
in the Vrica sediments. The anomalous recovery of diatom
microfossils in Vrica samples e6 and e5 suggests two possible
Figure 6.3. Stratigraphic distribution of diatom samples from the Vrica
explanations: either an increase in surface productivity or a
section. Thicknesses (in meters) accord with the section measured by
major change in the water chemistry of the Crotone Basin, either
Selli et al. (1977).
of which would have led to a transient saturation with dissolved
silica. The first explanation is favored by a considerable body of
information suggesting that surface production in the Mediterrathe overall dissolution was not severe, most of the sieve plates on nean was sharply increased during deposition of the laminated
sediments (Raffi and Thunell, Chapter 4, this volume), whereas
the Coscinodiscus specimens were heavily eroded.
The diatom assemblage from sample e6 at Vrica is overwhelm- there is no good evidence in favor of transient major changes in
ingly dominated by oceanic and/or neritic plankton, with a minor the global deep-sea silica budget during that interval.
The low diversity (not more than 15 distinct taxa in my count)
contribution from coastal benthic diatoms. On the basis of these
limited observations, it is not possible to reach conclusions about indicates, however, that conditions apparently did not support
preferential dissolution. However, it can be assumed that the very complex original populations. Thalassionema nitzschioides
dissolution occurred either within the water column or at the is a dissolution-resistant diatom (Schrader and Matherne, 1981)
sediment-water interface during sediment accumulation, rather sometimes used informally as an indicator of upwelling (L. H.
than as a diagenetic change after burial. The generally good Burckle, personal communication, 1982). Its absence from Vrica
condition of the specimens, however, and the intact "chains" of sample e6 suggests, but of course does not prove, that an episode
colonial forms suggest that environmental conditions during of upwelling was not the direct cause of the increased production
accumulation of the e6 laminates were anomalous, in that the indicated for this sample.
diatoms were relatively undisturbed before fossilization.
The interpretation of diatom microfossils from Vrica sample
The diatom assemblage in sample e6 is dominated by e6 is incomplete and probably will remain so until correlative
Coscinodiscus obscurus (cf. Hustedt, 1971,figure224), Coscino- recoveries are made. Nonetheless, it is fair to state that (1) there
discus radiatus (cf. Hustedt, 1971, figure 225; Hendey, 1964, pi. is no apparent component of redeposited Paleogene diatoms, (2)
1
60
Diatoms of the Vrica section
83
nally published 1930-1959, Leipzig: Akademische Verlagsgesellschaft.

Jouse, A. P. 1973. The Late Cenozoic diatom stratigraphy of
Sites 183-193, Leg 19. In Initial reports of the Deep Sea
Drilling Project, vol. 19, ed. J. S. Creager et al., pp.
Office.
Jouse, A. P. (ed.) 1977. Atlas of microorganisms in bottom
Acknowledgments
sediments of the oceans (diatoms, radiolarians, silicoflagellates, coccoliths). Moscow: Nauka.
The author wishes to thank R. C. Thunell for providing both the Koizumi, I. 1973. The late Cenozoic diatom stratigraphy of sites
samples whose study is reported herein and helpful editorial
183-193, Leg 19. In Initial reports of the Deep Sea Drilling
Project, vol. 19, ed. J. S. Creager et al., pp. 805-855.
remarks. The Belle W. Baruch Institute provided analytical
laboratory facilities. Dr. Tom Borg, Director of the Electron
Koizumi, I. 1975. Neogene diatoms from the western margin of
Microscopy Laboratory, School of Medicine, University of South
the Pacific Ocean, Leg 31, Deep Sea Drilling Project. In
Carolina, graciously permitted use of their SEM specifically for
Initial reports of the Deep Sea Drilling Project, vol. 31, ed.
this project, for which the author is most grateful.
D. E. Karig, J. C. Ingle, et al., pp. 799-807. Washington,
DC: U.S. Government Printing Office.
Koizumi, I. 1985. Diatom biochronology for late Cenozoic
northwest Pacific. 7. Geol Soc. Japan 91:195-211.
References
Koizumi, I. 1992. Diatom biostratigraphy of the Japan Sea. In
Proceedings of the Ocean Drilling Program: scientific
Backman, J., Shackleton, N. I , andTauxe, L. 1983. Quantitative
results, vol. 127/128, pt. 1, ed. K. A. Pisciotto et al., pp.
248-249. College Station, TX: Ocean Drilling Program.
Koizumi, I., Barron, J. A., and Harper, H. E. 1980. Diatom
304:156-158.
correlation of Legs 56 and 57 with onshore sequences in
Barron, J. A. 1980a. Upper Pliocene and Quaternary diatom
Japan. In Initial reports of the Deep Sea Drilling Project,
biostratigraphy of Deep Sea Drilling Project Leg 54,
vol. 56/57, ed. E. Honza et al., pp. 687-693. Washington,
tropical eastern Pacific. In Initial reports of the Deep Sea
Drilling Project, vol. 54, ed. B. R. Rosendahl et al., pp.
455-485. Washington, DC: U.S. Government Printing McCollum, D. W. 1975. Diatom stratigraphy of the Southern
Office.
Ocean. In Initial reports of the Deep Sea Drilling Project,
Barron, J. A. 1980b. Lower Miocene to Quaternary diatom
vol. 28, ed. D. E. Hayes et al., pp. 515-572. Washington,
biostratigraphy of Leg 57, off Northeastern Japan, Deep
Sea Drilling Project. In Initial reports of the Deep Sea Saito, T, Burckle, L. H., and Hays, J. D. 1975. Late Miocene to
Drilling Project, vol. 56/57, ed. E. Honza et al., pp.
Pleistocene biostratigraphy of equatorial Pacific sedi664-685. Washington, DC: U.S. Government Printing
ments. In Late Neogene epoch boundaries, ed. L. H.
Office.
Burckle and T. Saito, pp. 226-244. New York: MiBurckle, L. H., and Opdyke, N. D. 1977. Late Neogene diatom
cropaleontology Press.
correlations in the circum-Pacific. In Proceedings of the Schmidt, A., et al. 1972. Atlas der Diatomaceenkunde.
First International Congress on Pacific Neogene StratiKoenigstein/Taunus: Otto Koeltz. Originally published
graphy, Tokyo, 1976, ed. T. Saito and H. Ujiie, pp. 2551874-1959, Leipzig.
284. Tokyo: IUGS Regional Committee on Pacific Neo- Schrader, H. J. 1973. Cenozoic diatoms from the northeast
Pacific, Leg 18. In Initial reports of the Deep Sea Drilling
gene Stratigraphy.
Project, vol. 18, ed. L. D. Kulm, R. von Huene, et al., pp.
Burckle, L. H., and Todd, A. 1976. Correlation of late Neogene
sections on the Noto and Oga peninsulas, Japan. In
Office.
Progress in micropaleontology, ed. Y. Takayanagi and T.
Schrader, H. J. 1976. Cenozoic planktonic diatom biostratiSaito, pp. 20-26. New York: Micropaleontology Press.
graphy of the southern Pacific Ocean. In Initial reports of
Burckle, L. H., and Trainer, J. 1979. Middle and late Pliocene
the Deep Sea Drilling Project, vol. 35, ed. C. D. Hollister,
diatom datum levels from the central Pacific. MicropaleonC. Craddock, et al., pp. 605-640. Washington, DC: U.S.
tology 25:281-293.
Gladenkov, A. Y. 1994. Diatom assemblages from the PliocenePleistocene boundary beds in Kamchatka, Russia. Mi- Schrader, H. J., and Matherne, A. 1981. Sapropel formation in
cropaleontology 40:79-94.
the eastern Mediterranean Sea: evidence from preserved
Gombos, A. M. 1976. Paleogene and Neogene diatoms from the
opal assemblages. Micropaleontology 27:191-203.
Falkland plateau and Malvinas Outer Basin. In Initial Selli, R., Accorsi, C. A., Bandini Mazzanti, M., Bertolani
reports of the Deep Sea Drilling Project, vol. 36, ed. P. F.
Barker et al., pp. 575-687. Washington DC: U.S. GovernA. M., Cati, E, Colalongo, M. L., D'Onofrio, S.,
ment Printing Office.
Hendey, N. I. 1964. An introductory account of the smaller algae
(Calabria-Italy), a potential Neogene/Quaternary boundof British coastal waters, Part V, Bacillariophyceae. Lonary stratotype. Giorn. Geol 41:181-204.
don: Her Majesty's Stationery Office. Reprinted Amsterdam: A. Asher & Co.
Shackleton, N. J., Baldauf, J. G., Flores, J.-A., Iwai, M.,
Hustedt, F. 1971. Die Kieselalgen Deutschlands, Osterreichs
Moore, T. C , Raffi, I., and Vincent, E. 1995a. Biostratiund der Schweiz. In Kryptogamenflora, ed. L. Rabengraphic summary for Leg 138. In Proceedings of the
horst. Leuterhausen: Strauss u. Cramer GmbH. OrigiOcean Drilling Program, scientific results, vol. 138, ed.
the dominance of marine plankton is consistent with a 500-800m water column as inferred by Selli et al. (1977), (3) there is no
immediate correlation with deep-sea diatom biostratigraphy, and
(4) the assemblage represents a substantial short-term increase in
latest Pliocene sea-surface productivity.
84
Andrew J. M. Palmer

Shackleton, N. I , Crowhurst, S., Hagelberg, T., Pisias, N. G.,
and Schneider, D. A. 1995b. A new late Neogene time
scale: application to Leg 138 sites. In Proceedings of the
Ocean Drilling Program, scientific results, vol. 138, ed. N.

Drilling Program.
Part III
The paleontological context of the Pleistocene boundary
The Pliocene-Pleistocene boundary in deep-sea sediments

WILLIAM A. BERGGREN and LLOYD H. BURCKLE, JR.
sea cores over large latitudinal spans of the world's oceans. In

This chapter is a summary of the major micropaleontological Figure 7.2, we have plotted the better-documented datum levels
events that are represented in the stratigraphic record of deep- that reflect changes in calcareous and siliceous plankton over the
sea sediments over the past 2.5 m.y. The main focus is on those past 2.5 m.y. Owing to the sequential and relatively rapid
events associated with the Olduvai subchron, inasmuch as this is changes that occurred in Pliocene paleogeography and climatolthe time interval in which the Pliocene-Pleistocene boundary ogy and, concomitantly, biogeography, some of these datum
has now been located by formal action of the IUGS. Our events are geographically restricted, whereas others are mondial
discussion concentrates on three areas of particular interest: in nature (Figure 7.2). It will be seen that the Olduvai subchron
biostratigraphy, the history of oxygen-isotope variations, and is bracketed by a number of biostratigraphic events in both
calcareous and siliceous plankton.
carbonate stratigraphy.
In terms of planktonic foraminifera, the Olduvai subchron is
bracketed by the LADs of Globorotalia miocenica and
Biostratigraphy
Globorotalia exilis (tropical Atlantic region only) in the late
Pliocene, the FAD of Globorotalia truncatulinoides and the
Boundary-stratotype section
LAD of Globorotalia limbata just below the base of the
Current studies on the proposed Vrica boundary-stratotype Olduvai, the LAD of Globigerinoides obliquus extremus near
(Pasini and Colalongo, Chapter 2, this volume; Nakagawa et al., the top of the Olduvai, and the LAD of Globigerinoides
Chapter 3, this volume; Rio, Raffi, and Backman, Chapter 5, fistulosus coincident with or a short distance above the Olduvai
this volume; Palmer, Chapter 6, this volume) indicate that this (Figure 7.2). In North Atlantic sediments, the Pliocenedefinition places the Pliocene-Pleistocene boundary just above Pleistocene boundary, as recognized here, would appear to be
(or just within) the top of the Olduvai subchron of the Matuyama within Ericson's Zone Q, which is characterized by the absence
chron, with an estimated age of 1.81 Ma (Zijderveld et al., 1991; or low values of the Globorotalia menardii-G. tumida complex
Pasini and Colalongo, Chapter 2, this volume), and that this level (Figure 7.3).
Coiling patterns in the genus Pulleniatina have been found
is bracketed below by the extinction, or LAD (last-appearance
datum), of Discoaster brouweri and its triradiate variety, and useful in determining the biostratigraphy of Plio-Pleistocene
above by the FAD (first-appearance datum) of Gephyrocapsa deep-sea sequences (Saito, 1976). A continuous sequence can be
oceanica s.l. and the LAD of Calcidiscus macintyrei. The same followed from the late Miocene to the present day in the
relationships are seen in deep-sea cores (Figure 7.1). In equatorial Pacific, whereas in the equatorial Atlantic the genus
foraminifera biostratigraphy, the boundary-stratotype is almost Pulleniatina was absent from about 3.8 Ma to 2.6 Ma, owing, no
exactly coincident with the FAD of Globigerina cariacoensis. doubt, to the closure of the Panama seaway in mid-Pliocene
Other correlated events are the shift from dextral to sinistral time. In late Pliocene time there was general synchrony in coiling
coiling in most Neogloboquadrina pachyderma, the FAD of patterns between the Atlantic and Pacific forms (Figure 7.4),
Globorotalia umbilicata, and (at a slightly earlier age) the last which is useful in interoceanic correlation. The Olduvai
occurrence of Neogloboquadrina atlantica and first occurrence of subchronozone (a time-rock unit, as distinct from the time terms
"chron" and "subchron") is essentially bracketed by two peaks of
Globigerina digitata digitata.
sinistral coiling (L5 at the base and L4 at the top in the Pacific,
and AL2 at the base and AL1 at the top in the Atlantic). The
The deep sea
Pliocene-Pleistocene boundary, recognized here at the top of the
A number of biostratigraphic events have now been reliably Olduvai subchron, corresponds to a pronounced peak in dextral
correlated to the Cenozoic paleomagnetic stratigraphy in deep- coiling (L4 and L3) in the equatorial Pacific and to the beginning
Introduction
87
500-
No of
D BROUWERI mm 2
D BROUWERI
No of
VfiRTRIRADIATUS CMACJNTYREI mm 2
NO OF PREPLIOCENE
DISCOASTERS
05
10
15
20
40%
10
15
5 mm 2
H SELL/1
(A/=50)
0
N3
r
q
No of
D BROUWERI mm' 2
20 40%
40
No of
H SELLII
C MACINTYREI m m 2 [/V=100()/V=500W]
20
80
40
10%
ft*
b98
10-
Myr
p
1
n
m
400
R4
f
-
166
N2
R3
c
- b
D BROUWERI
N1
VAR TRIP ADIATUS
300
200 {
20
100 j
Figure 7.1. Quantitative nannofossil stratigraphy for the Vrica section and piston core
V28-239 (western equatorial Pacific). On the left are data from the Vrica section, starting with the lithologic section of Selli et al. (1977) and the magnetostratigraphy of Tauxe
et al. (1983). The abundances of Discoaster brouweri, D. brouweri var. triradiatus relative to all forms of D. brouweri, and Calcidiscus macintyrei each show a clear upper
limit. Pre-Pliocene discoasters maintain a uniform reworked abundance throughout the
section, and the proportion of Helicosphaera sellii relative to all helicosphaerids also
does not show any drop in abundance in the upper part of the section. The filled circles
between 270 and 345 m indicate samples that were prepared and counted in duplicate.
On the right side, the sequence in piston core V28-239 is from Backman and Shackleton (1983), with magnetostratigraphy from Shackleton and Opdyke (1976). (From
Backman et al., 1983, with permission of the editors of Nature.)
COILLING DIRECTION TRENDS

OF Pulleniatino
% RIGHT COILING
INDO-PACIFIC
SOUTHERN
REGIONS
1.2-
EQUATORIAL
REGIONS
Pulleniatina
NORTH PACIFIC
NORTH ATLANTIC
50
0
100
ATLANTIC
50
100
finalis
Helicopontosphoera sellii
Helicopontosphoera sellii
L3
Calcidiscus macintyrei
Rhizosolenia barboi
Vobigerinoides fistulosus Rhizosolenia curvirostris

Rhizosolenia praebergonin t \
vor. robusta
\>
Clathrocyclas bicornis
Pterocanium prismatium
L5
Pseudoeunotia doliolus
Eucyrtidium calvertense
Lamprocylas heteroporoi^
Globoquodrina asanoi
ofooororo/to truncatuiinoiaes
Discoaster brouwerii* t^
Coscinodiscus kolbei
Vobigerinoides obliquus
extremus
id truncatu
ts/er brouwern
Globorotalia exilis
Coscinodiscus vulnificus
^Gephyrocapsa aperta
L6
Thalassiosira convexa
frGephyrocapso aperta
Globorotalia miocenica \
Reappearance
of
Hetotholus vema
and D. surculus
and D. surculus
Desmospyris spongiosaj(\
Cosmiodiscus insignis-^ Stichocorys peregrina
X
Discoaster tamalis
X
Summary diagram for all microfossil datum levels in the Pliocene-Pleistocene boundary interval.
Pulleniotino
First
Appearance
Last
Appearance
90
William A. Berggren and Lloyd H. Burckle, Jr.
ERICSON'S
ZONES
DEPTH TIME
cm
my
PLANKTONIC FORAMINIFERA
-100
i1
1
N0RMAI
-300
O.69<
-400
-500
0.95"
ARAMIL
0.89.
1o
x r
-200
.ARIT
c
Spho<1 roi fl
g^
\ ^
Y
X
w
PERCENTAGE
MENARDII-COMPLEX
this paper
-600
a.
ex
o
|
o
o
5
z
EVI
ATUYAI
1
2E
1.61.
1/9
c
2.00
rin<
o
w
S
E
1100
oboro
o
*o
1200
1292
z;
-V
-900
1000
bliqi
1_
o
- !
800
lob
ex
otdl
Figure 7.3. Magnetic and faunal stratigraphy. The

solid curve is a plot of the percentage of
Globorotalia menardii complex versus depth in centimeters. Ericson's zones are defined by the presence
or absence of G. menardiiG. tumida species and
subspecies. The paleomagnetic normal-polarity
events are represented by solid black, and the
re versed-polarity events by white. The appearances
and extinctions of planktonic foraminifera species
are represented by the labeled vertical lines. The
horizontal dashed lines emphasize key biostratigraphic horizons. Between 800 cm and 1,000 cm,
the dashed lines are the key faunal boundaries that
define the Pliocene-Pleistocene boundary. The line
through the top of the P-zone marks the Discoaster
extinction boundary. The P-zone, as originally defined (Briskin and Berggren, 1975, fig. 1) by the
first abundant G. menardii-tumida peak in the
Olduvai subchron, is actually of latest Pliocene age
in terms of the chronology presented here. (From
Briskin and Berggren, 1975, courtesy of Micropaleontology Press.)
i n c o t u l i noidti
700
2.20-
.
5
1
10
I
15
20
Figure 7.4. Time (m.y.) versus percentage of right-coiling individuals in Pulleniatina populations from 10 low-latitude deep-sea cores. Time lines are drawn according to the
paleomagnetic-reversal sequence established for each core. Prominent left-coiling intervals
are numbered L1-L9, AL1, and AL2. The upper Gilbert and Gauss intervals in core
RC12-66 are low in carbonate because of dissolution effects. (From Saito, 1976, fig. 2,
courtesy of the Geological Society of America.)
92
the Pliocene-Pleistocene boundary in the Vrica section as well as

in deep-sea cores (Backman, Shackleton, and Tauxe, 1983,
MY
INDO-PACIFIC
ATLANTIC
figure 1; Shackleton et al., 1995).
o
BP
0
50
100
Quantitative data on the LADs for several late Pliocene
50
100
50I
100
I
calcareous nannoplankton have been delineated by Backman
and Shackleton (1983) and are shown here (Figures 7.6 and 7.7)
by way of reinforcing this approach as a powerful tool in
magnetobiochronologic research.
0 5In diatom-bearing sediments of the deep sea, the Olduvai
subchronozone has been identified in the Pacific, equatorial
Atlantic, and Southern oceans. In the equatorial Pacific, the
Olduvai subchronozone is characterized by a number of first and
last appearances of diatoms (Burckle, 1972, 1977). The LAD of
Thalassiosira convexa occurs in the lower Matuyama, about onethird of the way from the top of the Gauss to the base of the
Olduvai, whereas the first specimens of Pseudoeunotia doliolus
in most deep-sea cores are usually found in the lower part of the
Olduvai subchronozone. The actual FAD for this species appears
to be just below the Olduvai (Shackleton et al., 1995), but it
tends to be very rare in the lowest part of its range. The LAD of
Rhizosolenia
praebergonii var. praebergonii is just above the
ALK
base of the Olduvai, while very slightly above its top is the LAD
20of R. praebergonii var. robusta; for a summary of these data, see
Figures 7.2 and 7.8. In the equatorial Atlantic, Plio-Pleistocene
diatom data from DSDP site 397 were reported by Burckle
REAPPEARANCE OF
(1979), who found the LAD of Thalassiosira convexa just above
25Pulleniotmo
the Gauss chronozone, but found no other datum levels nearer
to the boundary (Figure 7.9).
U
Few quantitative data of that age are available for diatoms
O
across this boundary in the equatorial Pacific. Burckle (1971)
30noted some evidence for a slight cooling trend throughout the
Olduvai, a point that was also made by Mclntyre, Be, and
Pretiskas (1967). Burckle and Trainer (1979) concluded that
DISAPPEARANCE OF
climatic fluctuations played an important role in the equatorial
Pullematmo
Pacific in dictating diatom first and last appearances during the
35late Pliocene-early Pleistocene.
Paleomagnetically dated deep-sea cores of that age are not
L9
available from high northern latitudes in the Pacific, but rich
assemblages of boreal and cold-temperate marine diatoms occur
40J
in the Olkhov Formation of eastern Kamchatka, dated to
Figure 7.5. Generalized trends of coiling direction for Pulleniatina popuMatuyama age. According to Gladenkov (1994), the Olkhov
lations since 4 Ma in the Indo-Pacific and Atlantic faunal provinces.
beds represent subarctic shelf environments, with winter ice
(From Saito, 1976, fig. 3, courtesy of the Geological Society of Amercover, in which the level of the Vrica boundary in the upper
ica.)
Olduvai subchronozone is associated with the LADs for
Thalassiosira antiqua and Neodenticula koizumii, above the FAD
of what was to remain a dextral coiling pattern in the equatorial of Pyxidicula californica, and just below the FADs for Simonseniella curvirostris and Thalassiosira pacifica. This relationAtlantic (Figure 7.5).
Quantitative studies on the calcareous nannoplankton (Back- ship correlates to the uppermost part of the Neodenticula
man and Shackleton, 1983) have refined the application of the koizumii zone, just below the boundary with the zone of
datum levels in this group to biostratigraphic problems. The Actinocyclus oculatus, as defined in deep-sea cores from the
extinction of the Discoaster brouweri group, just slightly below northwestern Pacific and the Sea of Japan (Koizumi, 1985,1992).
the Olduvai subchronozone, has essentially verified and given In some parts of the northwestern Pacific, the last occurrence of
greater precision to previous studies (Haq et al., 1977) that had Rhizosolenia praebergonii var. robusta is available to help
suggested an association of these two events. The LAD of identify the immediate post-Olduvai levels, but this datum tends
Calcidiscus macintyrei is also seen to be stratigraphically above to be diachronous in the higher latitudes.
PERCENT RIGHT
COILING
Plio-Pleistocene boundary in the deep sea
NUMBER OF D B R O U W E R I () A N D
REWORKED DISCOASTER | . | m m 2
0
20
40
NUMBER OF D B R O U W E R I
T R I R A D I A T E FORM | % )
0
20
93
() A N D
REWORKED DISCOASTER ()
40
10
20
30 0
100
mm2
200
300
TRIRADIATE FORM
0
20
(%)
40
RC11 -209
NUMBER OF D BROUWERI mm
0
20
40
TRIRADIATE FORM
0
20
(%)
NUMBER OF D. BROUWERI
40
10
20
30 0
100
mm
200
300
1100-
Figure 7.6. D. brouweri abundances in four mid- and low-latitude Pacific cores. For each core, the abundance of D. brouweri is shown on
the left, and the relative abundance of triradiate forms on the right.
Open circles indicate the abundance of reworked Pliocene discoasters.

(From Backman and Shackleton, 1983, fig. 4, with permission of
Elsevier Publishing Co.)
In the Southern Ocean, the Olduvai subchronozone has been

identified in a number of piston cores and DSDP sites, in which it
is characterized by the LAD of Cosmiodiscus insignis just above
the top of the Gauss chronozone, the LAD of Coscinodiscus
vulnificus midway between the top of the Gauss chronozone and
the base of the Olduvai subchronozone, the LAD of Coscinodiscus kolbei just below the Olduvai, and the LAD of Rhizosolenia
barboi just above the Olduvai (Donahue, 1970; McCollum,
1975; Cieselski, 1983). One of the major problems in detailing
diatom datum events through the Plio-Pleistocene interval in the
Southern Ocean is the uncertainty concerning the effects of the
different water masses on these datum events. Most of the same
biostratigraphic criteria apply to the sub-Antarctic region, but
we know less about the region south of the polar front.
Our knowledge of radiolarian datum levels in the PliocenePleistocene boundary interval of paleomagnetically dated cores
is largely derived from the classical studies of Hays and coworkers (Hays, 1965, 1970; Opdyke et al., 1966; Hays and
Opdyke, 1967; Hays et al., 1969; Saito, Burckle, and Hays,

1975). The most significant datum in the equatorial Pacific is the
widely recorded extinction of Pterocanium prismatium near the
Pliocene-Pleistocene boundary. Hays et al. (1969) and Saito et
al. (1975) reported that it was absent from the upper part of the
Olduvai subchron (Figure 7.2), while Shackleton et al. (1995)
placed this datum just above its top. The latter authors also
noted the FAD of Anthocyrtidium angulare coincident with the
LAD of Globigerinoidesfistulosusat the upper boundary of the
Olduvai subchronozone, and the LAD of Anthocyrtidium
jenghisi coincident with the FAD of Globorotalia truncatulinoides at about 0.5 m.y. earlier than its base. The other
significant radiolarian biostratigraphic events in the vicinity of
the boundary are the last appearance of Stichcorys peregrina in
the upper part of the Gauss chron and the successive last
appearances of Lamprocyrtis heteroporos, Pterocorys minithorax, and Theocorythium vetulum in the interval between the
Olduvai and Jaramillo subchrons, the last through evolutionary
94
JMBER
OF 0 BROUWERI
mm
JMBER OT 0 ASYMMETRICUS m m '
NUMBER OF 0 TAMALIS
mm
NUMBER OF 0 VARIABIUS GROUP m m '
NUMBER OF 0 SURCIRUS n
NUMBER OF 0 PENTARADIATUS mm
Figure 7.7. Abundances of Discoaster species in V28-179, central equatorial Pacific, in the interval from the top of the Olduvai to the base of
the Gauss (see Preface, this volume, for modern time scale values): A,
abundance of D. brouweri; B, abundance of D. triradiatus; C, abundance of D. tamalis; D, relative abundance of D. tamalis; E, abundance
of the D. variabilis group (including D. challenged and D. decorus); F,
abundance of D. surculus; G, abundance of D. pentaradiatus. The sampling interval is 5 cm, but the data are presented on a time-scale based
on paleomagnetic data. Note scale difference between D. brouweri and
the other discoasters. (From Backman and Shackleton, 1983, fig. 6,
with permission of Elsevier Publishing Co.)
replacement by Theocorythium trachelium (Nigrini, 1970; Johnson and Knoll, 1975; Shackleton et al., 1995).
In the northern Pacific, three radiolarian datum levels are
recognized in the Pliocene-Pleistocene boundary interval (Hays,
1970) (Figure 7.2). The first, the LAD of Eucyrtidium elongatum
peregrinum, occurs in the middle of the Gauss chronozone, as it
does in the central Pacific. The second is the LAD of
Lamprocyrtis heteroporos near the base of the Olduvai
subchronozone, although, as discussed earlier, the southern
population of this species in the equatorial Pacific became extinct
at a later date, as was also the case in the high latitudes of the
Southern Ocean (Hays, 1970). Third, Eucyrtidium matuyamai
evolved from E. calvertense near the base of the Olduvai
subchron. Hays (1970) notes that in the Southern Ocean, E.
calvertense disappears in the lower part of the Olduvai subchron
at approximately the same level where it gives rise to E.
matuyamai in the northern Pacific.
In the Southern Ocean, Hays (1965), Opdyke et al. (1966),
and Hays and Opdyke (1967) reported a number of radiolarian
events near the Pliocene-Pleistocene boundary interval. Desmospyris spongiosa and Helotholus vema disappear just above the
Gauss chronozone, and Clathrocyclas bicornis disappears near
the top of the Olduvai subchronozone. In terms of assemblage
zones, the Pliocene-Pleistocene boundary would fall within the
X Zone of Hays (1965). Hays and Opdyke (1967) also reported a
sharp decrease in the abundances of the more widely ranging,
presumably less cold-tolerant radiolarians in the Southern Ocean
just above the Olduvai.
Stable-isotope record
It is well known that in the Southern Hemisphere, Antarctica has

had a continental-ice record extending as far back as the middle
Miocene, probably into the late Oligocene, and, according to
some interpretations of the stable-isotope record, as far back as
the Eocene-Oligocene transition. In contrast, the biostratigraphic studies of North Atlantic deep-sea cores suggest that the
initiation of ice rafting and thus Northern Hemisphere glaciation
95
S l 8 0 %o P.D.B.
45
40
3 5
30
I T
Rhizosolenia praebergonii
var. robusta (1 )
I B
I T
Pseudoeunotia doliolus (2)

Rhizosolenia praebergonii (2)
I TR Coscinodiscus nodulifer
var cyclopus (2)
I BR Coscinodiscus nodulifer
| >C
vor
cyclopus
(2)
^Rhizosolenia praebergoniiR praebergonii
var robusta ( 2 )
1 2
I TR Coscinodiscus nodulifer
var
I T
cyclopus
(2)
Th convexa var aspinosad)
1 3
I BR Coscinodiscus nodulifer
I g var cyclopus(2)
wx robusta (2)
I T
Nitzschio jouseae (1)
I B
(1)
I BR Thalassiosira convexa (3)
19
Thalassiosira convexa (3)
IB
Thalassiosira
convexa (3)
LU
CO
20
IT
Figure 7.8. Magnetostratigraphy

and diatom datum levels for Upper Pliocene-Lower Pleistocene
sediments of the equatorial Pacific. (From Burckle and Trainer,
1979, courtesy of Micropaleontology Press.)
NORTH PACIFIC
DATUM LEVELS
(after Koizumi, 1975
Burckle & Opdyke, 1977)
DSDP
SITE 397
Paleomagnetic
Stratigraphy
u-
m.
X'X*
x-.-.-x-x
..%:...
>X\\vX- v X
iiii
HIM
CO
UJ
XvXvX X v
X
2
QC
CD
-LAD Rhizqsolenia
curvirostris
HIP
-LAD Nitzschia reinholdii-
BRUNHES
96
- LA D Nitzschia reinholdii
!::::
-LAD Mesocena ellip tica
-LAD Mesocena ellipticavX*xvx*x-x*Xv;
XvX'Xv
X*M*
X'X'X'X*
:gj
X*X*"
x*x
MATU
XvXvX
XvXvX
XvXvX
liiixB
MATUYAMA
100-
Rhizosolenia
-LAD curvirostris
-LAD Thalassiosira convex a -
200-
-LAD Thalassiosira convexa
||
occurred at about 3 Ma (Berggren, 1972). That remained the

orthodox opinion for nearly a decade. A biostratigraphic restudy
of North Atlantic core material (Backman, 1979) has more
recently suggested that the initiation of glaciation may have
occurred closer to 2.5 Ma than 3.0 Ma, corroborated by oxygenisotope records in the equatorial Pacific (Shackleton and
Opdyke, 1977; Shackleton et al., 1995) and the North Atlantic
(Shackleton et al., 1984, 1990).
It is now apparent that although there was marked climatic
variability on a global scale in the early-middle Pliocene, global
climate cycles did not induce sea-level glaciation until about 2.5
Ma. That conspicuous climatic threshold corresponds temporally
to the Pretiglian cool-climate interval in Europe, which saw
significant vegetational changes just after the Gauss/Matuyama
polarity reversal, and that change has been interpreted by some
(e.g., Zagwijn, 1974) as indicative of the Pliocene-Pleistocene
boundary. In fact, the Pretiglian cold-climate peak seems to have
been a stepping-stone midway on the descent from preglacial
conditions in the mid-Pliocene (3.0 Ma) to fully glacial conditions in the mid-Pleistocene (0.9 Ma). As discussed elsewhere
(Hays and Berggren, 1971; Berggren and Van Couvering, 1979;
Aguirre and Pasini, 1985; Shackleton et al., 1990), that climatic
event was not of the same age as the Lyellian boundary
recognized by the International Geological Congress in 1948
(King and Oakley, 1950) and formally adopted at Vrica (Selli,
1977; Cowie and Bassett, 1989).
iliiH
CO
CO
GAU
Figure 7.9. Magnetostratigraphy and

diatom datum levels for Upper
Pliocene-Lower Pleistocene sediments
for DSDP site 397 in the equatorial Atlantic. (From Burckle, 1979, courtesy
of the Offshore Drilling Program.)
GAUSS
The main features of the late Pliocene-Pleistocene oxygenisotope record include:

1. minor fluctuations prior to 2.4 Ma
2. a relatively major (approximately 1%) increase in 218O
of ocean waters at about 2.5 Ma, corresponding to a
major expansion of glacial (year-round) ice in the
Northern Hemisphere and an expansion to the limits of
drifting ice in the northern oceans
3. stronger cycles since that time, with fluctuations in
oxygen-isotope ratios on the order of 1% and periodicities that, at least over the past million years, have
been correlated to periodic variabilities in the rotation
and orbital path of the earth (Milankovitch cycles)
(Shackleton et al., 1990, 1995).
An inspection of late Pliocene-Pleistocene oxygen-isotope
records (Figures 7.10 and 7.11) shows two climatic regimes
characterized by average glacial-interglacial variations in
oxygen-isotope ratios, the first (between 2.5 and 0.9 Ma) varying
by about 1%, and the second (0.9 Ma to present day) by more
than 1%. The change to more strongly contrasting values in the
middle and late Pleistocene (i.e., after the Jaramillo subchron)
indicates a change to more severe glacial conditions during
climatic lows. In terms of the Pliocene-Pleistocene boundary,
oxygen-isotope variations during the approximately 220-k.y.-
0.50
97
0.00
Figure 7.10. Correlation of oxygen-isotope stratigraphy between the Atlantic (DSDP site 502B, and piston cores P6304-9 and P6408-9) and
the Pacific (cores V28-238 and V28-239). Interglacial isotope stage
numbers are circled. Where present, magnetostratigraphic boundaries
long Olduvai subchron correspond to approximately three

stadial-interstadial cycles. According to previous interpretations
of the Vrica section (Aguirre and Pasini, 1985), the PliocenePleistocene boundary placed just above the top of the Olduvai
interval would fall within an interstadial period (Shackleton et
al., 1984,1995) (Figure 7.12), equivalent to oxygen-isotope stage
63 (Shackleton et al., 1990, figure 4). If the boundary is placed
within the uppermost part of the Olduvai subchronozone,
however, according to revised magnetostratigraphy at Vrica
(Zijderveld et al., 1991), it will fall instead in the cold-climate
maximum of oxygen-isotope stage 64 (Shackleton et al., 1990,
figure 4).
are shown. Cores P6304-9 and P6408-9 are from Emiliani (1966,
1978), and cores V28-238 and V28-239 are from Shackleton and
Opdyke (1976, 1977). Site 502B is from Prell (1982, fig. 2).
Carbonate stratigraphy
Cyclic fluctuations in the carbonate contents of deep-sea

sediment cores have long been known to correlate with, and
reflect, climatic oscillations on a global scale. In general, high
carbonate values correspond to glacial intervals, and low
carbonate values correspond to relatively warm interglacial
intervals. When correlated by means of biostratigraphy and
calibrated to a geochronologic system by means of paleomagnetic stratigraphy, carbonate stratigraphy has become a powerful
tool for late Neogene global correlations, as summarized by
Vincent (1981).
98
CaCO3
30
40
50
60
518O
70
80
90
4.0
3.0
2.0
1.0
-1.0
-2.0
M21
150-
GU3
200
250-
300-
300
350
400-
I Unrecovered interval
Figure 7.11. Magnetostratigraphy, carbonate stratigraphy, and isotope

stratigraphy at DSDP site 157. Magnetostratigraphy derived from correlations with paleomagnetically dated cores (Kaneps, 1973). Carbonate
data from Bode and Cronin (1973). Isotope data from Keigwin (1979)
on Uvigerina. Note the coincidence, as at site 310, of low-carbonate

event M21 with the enrichment in 18O just above the Matuyama-Gauss
boundary. (From Vincent, 1981, fig. 4, courtesy of the Offshore Drilling Program.)
A sequence of eight carbonate cycles (corresponding to

glacial-interglacial episodes) was identified in the equatorial
Pacific (Hays et al., 1969) throughout the Brunhes magneticpolarity chronozone. Periodicities ranged from about 75,000
years in the late Brunhes to over 100,000 years in the early
Brunhes. In the Matuyama chron (down to the top of the
Olduvai subchron) seven more cycles were identified, with
average, but more irregular, periodicities of 100,000 years. A
numerical system was established, with odd and even numbers

denoting low (interstadial warm climate) and high (stadial cold
climate) carbonate intervals, respectively. Carbonate cycles were
recognized down to the Gilbert C event in the early Pliocene.
That sequence was subsequently extended down into the late
Miocene by Saito et al. (1975). Kaneps (1973) extended the
nomenclature system below the mid-Pliocene (GU3) to the early
Gilbert and informally numbered 25 carbonate peaks and valleys
'
' l '
'
99
60m
90-
IJO
140
. : . . .
L47
D.P.
D.B.
24
D.3.
D.T.
3.0
ii
Figure 7.12. Combined magnetic

and oxygen-isotope records for
DSDP site 552A. Top: Magnetic
record for site 552A. Demagnetized inclinations are shown only
for apparently undisturbed parts
of the cores (the data from core
11 suggest the presence of some
unrecognized disturbance). Bottom: Upper panel shows the
oxygen-isotope record of site
552A, cores 7-12. The plotting
scale is linear between magneticreversal horizons. The lower
panel shows the oxygen-isotope
record in Pacific core V28-179
for comparison. Vertical lines
show horizons used for time control as follows: top of the Olduvai
subchron at 1.66 Ma, base of the
Matuyama reversed chron at 2.4
Ma, top of the Kaena reversed
subchron at 2.92 Ma, base of the
Gauss normal chron at 3.40 Ma.
Nannofossil extinction horizons determined on both cores are indicated as follows: DT, Discoaster
tamalis; DS, D. surculus; DP, D.
pentaradiatus; DB, D. brouweri.
(From Shackleton et al., 1984,
figs. 2 and 4, with permission of
the editors of Nature, London.)
to
100
if
r"\/*
O
0
0
to
D.P.
24
D.3.
D.B.
D.T.
SdO
24
1.7
14
1.9
2JO
2.1
24
24
1A
24
24
2.7
24
AGE, MILLION YEARS
2.0
IJO
t.1
100
100
Figure 7.13. Correlations between the carbonate stratigraphy in site

502 and the oxygen-isotope stratigraphy in core V28-239, and between
the carbonate stratigraphies in site 503 and core RC11-209 and the
oxygen-isotope stratigraphy in core V28-239. The oxygen-isotope stages
are modified from Shackleton and Opdyke (1976), and the carbonate
cycles are from Hays et al. (1969). (From Gardner, 1982, fig. 10, courtesy of the Offshore Drilling Program.)
with lowercase-letter designations. That informal system was

replaced by a formal nomenclature in which carbonate events
were designated according to their association within a particular
magnetic-polarity chron (Dunn and Moore, 1981). Dunn and
Moore extended their carbonate stratigraphy down to magnetic
chron 9 and gave alternating lowercase notations to carbonate
minima (a, c, e, etc.) and maxima (b, d, f, etc.).
In the western Caribbean (DSDP site 502) and eastern
equatorial Pacific (DSDP site 503), a high-resolution carbonate
stratigraphy has been established for the late Pliocene and
Pleistocene (Gardner, 1982) (Figure 7.13) in which carbonate
variations are linked with the oxygen-isotope stages of Shackle-
ton and Opdyke (1976), and their system of informal designation

has been extended into the early Pleistocene. In this scheme,
lowercase letters are used, each preceded by the uppercase
abbreviation for the associated magnetic-polarity chron (i.e., Mb
= Matuyama b). Similar high-resolution carbonate stratigraphies
have been developed for the Pacific Ocean (Figure 7.14).
In terms of the Pliocene-Pleistocene boundary, it can be seen
from the data summarized by Vincent (1981) and Gardner (1982)
that the Olduvai subchronozone is bracketed by a carbonate low
(M17) at the top and a carbonate peak (M18) at the base, in the
terminology of Hays et al. (1969). The Pliocene-Pleistocene
boundary, if located just above the top of the Olduvai subchron,
DSDP SITE 310

CaCO3
20
40
(%)
60
80
RC11-209
CaCO
100
V24-59
CaCO,
20
40
60
80
B3B5
^
B i3-<cz:
315=
^
Ml
10-
=*
-,.
4-
M17-<!
M19-
20\13
Collosphaera sp. A -*- Buccinosphaera invaginata

11
T
Axoprunum angelinum
13
T
Pseudoemiliania lacunosa
14
B
Collosphaera tuberosa
15
B
Collospaera sp. A
16
T
Nitzschia reinholdii
18
T
Mesocena eliptica
:=
8-
GI5==m
| =
n GI7GI9=
30-
Biostratigraphic Correlations (numbers follow listing in Table 12; T = top; B = bottom):

2
3
4
5
6
8
9
GU1
G U 3 ^ T Z ^ ^
\18
M21
<r^
- ^.
M21
T
Anthocyrtidium angulare
Mesocena elliptica
B
Theocorythium vetulum - * 7". trachelium
T
T
Pterocanium prismatium
B
Pseudoeunotia doliolus
"
12-
Figure 7.14. Correlation of Pleistocene carbonate stratigraphy between

DSDP site 310 (Hess Rise) and equatorial piston cores. Numbers on the
right-hand side of each sequence refer to sub-bottom depths in meters.
(From Vincent, 1981,fig.6, courtesy of the Offshore Drilling Program.)
102
Burckle, L. H. 1979. Validation of middle Pliocene to Pleistocene paleomagnetic reversal record using diatom and
silicoflagellate datum levels. In Initial reports of the Deep
Sea Drilling Project, vol. 47, ed. U. von Rad, W. B. F.
Ryan, et al., pp. 479-480. Washington, DC: U.S. Government Printing Office.
Burckle, L. H., Morley, I , Koizumi, I., and Bleil, U. 1985. Late
Neogene biostratigraphic and paleomagnetic correlations
Acknowledgments
between the equatorial and northwest Pacific. In Initial
reports of the Deep Sea Drilling Project, vol. 86, ed. G. R.
The authors would like to thank J. D. Hays (Lamont-Doherty
Heath, L. H. Burckle, et al., pp. 781-786. Washington,
Geological Observatory) and W. Prell (Brown University) for
their helpful comments and review of an earlier draft of this Burckle, L. H., and Opdyke, N. D. 1977. Late Neogene diatom
chapter. The efforts of the U.S. working group of IGCP Project
correlations in the circum-Pacific. In Proceedings of the 1st
International Congress on Pacific Neogene Stratigraphy, ed.
41 have been supported by a grant from the National Science
Y. Takayanagi and T. Saito, pp. 255-284. Tokyo: IUGS
Foundation (Earth Sciences Division). This is WHOI ContribuRegional Committee on Pacific Neogene Stratigraphy.
tion no. 5675.
Burckle, L. H., and Opdyke, N. D. 1984. Late Miocene/earliest
Pliocene diatom correlations in the north Pacific. Mar.
Micropal. 9:212-227.
References
Burckle, L. H., and Trainer, J. 1979. Late Pliocene diatom
datum levels in the central Pacific. Micropaleontology
25:281-293.
Backman, J. 1979. Pliocene biostratigraphy of DSDP sites 111 Cieselski, P. 1983. The Neogene and Quaternary diatom
and 116 from the North Atlantic Ocean and the age of
biostratigraphy of sub-Antarctic sediments, Deep Sea
northern hemisphere glaciation. Stockholm Contrib. Geoi
Drilling Project, Leg 71. In Initial reports of the Deep Sea
32:117-137.
Drilling Project, vol. 71, ed. W. Ludwig, V. A. KrasheninBackman, J., and Shackleton, N. J. 1983. Quantitative bionikov, et al., pp. 635-666. Washington, DC: U.S. Governchronology of Pliocene and early Pleistocene calcareous
ment Printing Office.
nannofossils from the Atlantic, Indian and Pacific Oceans. Cowie, J. W., and Bassett, M. G. 1989. International Union of
Geological Sciences 1989 stratigraphic chart. Episodes 12:
Backman, J., Shackleton, N. J., andTauxe, L. 1983. Quantitative
nannofossil correlation to open ocean deep-sea sections Denizot, G. 1957. Lexique stratigraphique internationale, Volume
from the Plio-Pleistocene boundary at Vrica. Nature
1, fascicule 4a VII. Tertiare: France, Belgique, Pays Bas,
304:156-158.
Luxembourg. Paris: Centre National de la Recherche
Berggren, W. A. 1972. Late Pliocene-Pleistocene glaciation. In
Scientifique.
Initial reports of the Deep Sea Drilling Project, vol. 12, ed. Donahue, J. 1970. Diatoms as Quaternary biostratigraphic and
A. S. Laughton, W. A. Berggren, et al., pp. 953-963.
paleomagnetic indicators in high latitudes of the Pacific
ocean. Unpublished dissertation, Columbia University.
Berggren, W. A. 1975. The Pliocene-Pleistocene boundary in Dunn, D. A., and Moore, T. C , Jr. 1981. Late MiocenePliocene (magnetic epoch 9-Gilbert magnetic epoch)
deep-sea sediments: status in 1975. Geoph. GeoL, ser. 2
calcium carbonate stratigraphy of the equatorial Pacific
41:375-384.
Ocean. Geol. Soc. Am., Bull. 92:408-451.
Berggren, W. A., and Van Couvering, J. A. 1979. Quaternary. In
Treatise on Invertebrate Paleontology. Part A: Introduction,Emiliani, C. 1966. Paleotemperature analysis of Caribbean cores
ed. R. A. Robison and C. Teichert, pp. A505-A543.
P6304-8 and P6304-9 and a generalized temperature
Boulder: Geological Society of America.
curve for the past 425,000 years. /. Geol. 74:109-126.
Briskin, M., and Berggren, W. A. 1975. Pleistocene stratigraphy Emiliani, C. 1978. The cause of the ice ages. Earth Planet. Sci.
and quantitative paleo-oceanography of tropical North
Lett. 37:349-352.
Atlantic core V16-205. In Late Neogene epoch bound- Gardner, J. V. 1982. High resolution carbonate and organicaries, ed. T. Saito and L. H. Burckle, pp. 167-198. New
carbon stratigraphies for the late Neogene and Quaternary
York: Micropaleontology Press.
from the western Caribbean and eastern equatorial Pacific.
Bode, G. W., and Cronin, D. S. 1973. Carbon and carbonate
In Initial reports of the Deep Sea Drilling Project, vol. 68,
analyses, Leg 16. In Initial reports of the Deep Sea Drilling
ed. W. L. Prell, J. V. Gardner, et al., pp. 455-464.
Project, vol. 16, ed. T. H. van Andel, G. R. Heath, et al.,
pp. 521-528. Washington DC: U.S. Government Printing Gignoux, M. 1950. Geologie stratigraphique, 4th ed. Paris:
Office.
Masson et Cie.
Burckle, L. H. 1971. Late Cenozoic diatoms from the eastern Gignoux, M. 1955. Stratigraphic Geology (translation of the 5th
equatorial Pacific. Unpublished Ph.D. thesis, New York
ed.). San Francisco: Freeman.
University.
Gladenkov, A. Y. 1994. Diatom assemblages from the PlioceneBurckle, L. H. 1972. Late Cenozoic planktonic diatom zones
Pleistocene boundary beds in Kamchatka, Russia. Microfrom the eastern equatorial Pacific. In First International
paleontology 40:79-94.
Symposium on Fossil Marine Diatoms, ed. R. Simonsen, Haq, B. U , Berggren, W. A., and Van Couvering, J. A. 1977.
pp. 217-245. Nova Hedwigia, vol. 39. Bremerhaven.
Burckle, L. H. 1977. Pliocene and Pleistocene diatom datum
269:483-488.
levels from the equatorial Pacific. Quat. Res. 7:330-340. Hays, J. D. 1965. Radiolaria and late Tertiary and Quaternary
would correspond to the carbonate minimum M15, but if located
within the latest part of the Olduvai subchron it would correlate
with the more pronounced minimum M17; both are within the
interval of low carbonate values corresponding to oxygen stage
Mb of Gardner (1982).
103
isotopic variability. In Initial reports of the Deep Sea

history of Antarctic seas. Biology of the Antarctic Seas, II.
Drilling Project, vol. 68, ed. W. L. Prell, J. V. Gardner, et
Antarct. Res. Series 5:125-184.
al., pp. 465-474. Washington, DC: U.S. Government
Hays, J. D. 1970. Stratigraphy and evolutionary trends of
Printing Office.
radiolaria in north Pacific deep-sea sediments. In Geological investigations of the North Pacific, ed. J. D. Hays, pp. Rossignol-Strick, M. 1983. African monsoons, an immediate
response to orbital insolation. Nature 304:46-49.
185-218. Memoir 126. New York: Geological Society of
Saito, T. 1976. Geologic significance of coiling direction in the
America.
planktonic foraminifera Pulleniatina. Geology 4:305-309.
Hays, J. D., and Berggren, W. A. 1971. Quaternary boundaries
and correlations. In Micropaleontology of the oceans, ed. Saito, T., Burckle, L. H., and Hays, J. D. 1975. Late Miocene to
Pleistocene biostratigraphy of equatorial Pacific sediB. M. Funnell and W. R. Riedel, pp. 669-691. Cambridge
ments. In Late Neogene epoch boundaries, ed. T. Saito and
University Press.
L. H. Burckle, pp. 226-244. New York: MicropaleonHays, J. D., Imbrie, J., and Shackleton, N. J. 1976. Variations in
tology Press.
the earth's orbit: pacemaker of the ice ages. Science
Selli, R. 1977. The Neogene/Quaternary boundary in the Italian
194:1121-1132.
marine formations. Giorn. Geol. 41:81-105.
Hays, J. D., and Opdyke, N. D. 1967. Antarctic radiolaria,
magnetic reversals, and climatic change. Science 158:1001- Selli, R., Accorsi, C. A., Bandini Mazzanti, M., Bertolani
1011.
Hays, J. D., Saito, T., Opdyke, N. D., and Burckle, L. H. 1969.
Landini, W, Menesini, E., Mezzetti, R., Pasini, G.,
Pliocene-Pleistocene sediments of the equatorial Pacific:
their paleomagnetic, biostratigraphic and climatic record.
(Calabria-Italy). A potential Neogene/Quaternary boundGeol. Soc. Am., Bull. 80:1481-1513.
Hoernes, M. 1856. Die fossilen Mollusken des Tertidr-Beckens
von Wien, Abhandlungen no. 3. Wien: K.-K. geologisches Shackleton, N. J., Backman, J., Zimmerman, H., Kent, D. V.,
Hall, M. V., Roberts, D. G., Schnitker, D., Baldauf, J.
Reichanstalt.
G., Desprairies, A., Homrighanson, R., Huddleston, P.,
Johnson, D. A., and Knoll, A. H. 1975. Absolute ages of
Keene, J. B., Kaltenback, A. J., Krumsiek, K. A. D.,
Quaternary radiolarian datum levels in the equatorial
Morton, A. C , Murray, J. W., and Westberg-Smith, J.
Pacific. Quat. Res. 5:99-110.
1984. Oxygen-isotope calibration of the onset of ice-rafting
Kaneps, A. 1973. Carbonate chronology for Pliocene deep-sea
and history of glaciation in the North Atlantic region.
sediments. In Initial reports of the Deep Sea Drilling Project,
Nature 307:620-623.
vol. 16, ed. T. van Andel, G. R. Heath, et al., pp. 873-881.
Shackleton, N. J., Baldauf, J. G., Flores, J.-A., Iwai, M.,
Moore, T. C , Raffi, L, and Vincent, E. 1995. BiostratiKeigwin, L. 1979. Late Cenozoic stable isotope stratigraphy and
graphic summary for Leg 138. In Proceedings of the
paleoceanography of DSDP sites from the east equatorial
Ocean Drilling Program, scientific results, vol. 138, ed.
and central North Pacific Ocean. Earth Planet. Sci. Lett.
45:361-382.
King, W. B. R., and Oakley, K. P. 1950. Report of the
Temporary Commission on the Pliocene-Pleistocene Shackleton, N. J., Berger, A., and Peltier, W. R. 1990. An
alternative astronomical calibration of the lower PleistoBoundary, appointed 16th August, 1948. In International
cene timescale based on ODP site 677. Trans. R. Soc.
Geological Congress, Report of the Eighteenth Session,
Great Britain 1948, Part I. General Proceedings, ed. A. J.
Butler, pp. 213-214. London: Geological Society.
paleomagnetic stratigraphy of Pacific core V28-239: late
Koizumi, I. 1985. Diatom biochronology for late Cenozoic
Pliocene to late Pleistocene. In Investigations of Late
Northwest Pacific. /. Geol. Soc. Japan 91:195-211.
Quaternary paleoceanography and paleoclimatology, ed.
Koizumi, I. 1992. Diatom biostratigraphy of the Japan Sea. In
R. M. Cline and J. D. Hays, pp. 449-464. Memoir 145.
Proceedings of the Ocean Drilling Program: scientific
New York: Geological Society of America.
results, vol. 127/128, pt. 1, ed. K. A. Pisciotto, et al., pp.
Shackleton, N. J., and Opdyke, N. D. 1977. Oxygen isotopes and
248-249. College Station, TX: Ocean Drilling Program.
paleomagnetic evidence for early northern hemisphere
McCollum, D. 1975. Diatom stratigraphy of the Southern
glaciation. Nature 270:216-219.
Ocean. In Initial reports of the Deep Sea Drilling Project,
vol. 28, ed. D. E. Hays, L. A. Frakes, et al., pp. 515-571. Tauxe, L., Opdyke, N. D., Pasini, G., and Elmi, C. 1983. Age of
Mclntyre, A., Be, A. W. H., and Pretiskas, R. 1967. Coccoliths
and the Pliocene-Pleistocene boundary. In Progress in Vincent, E. 1981. Neogene carbonate stratigraphy of Hess Rise
(Central North Pacific) and paleoceanographic implicaoceanography, vol. 4, ed. M. Sears, pp. 3-25. New York:
tions. In Initial reports of the Deep Sea Drilling Project,
Pergamon Press.
vol. 62, ed. J. Thiede and T. Vallier, pp. 571-606.
Nigrini, C. A. 1970. Radiolarian assemblages in the North Pacific
and their application to a study of Quaternary sediments in
core V20-130. In Geological investigations of the North Zagwijn, W. H. 1974. The Pliocene-Pleistocene boundary in
western and southern Europe. Boreas 3:75-97.
Pacific, ed. J. D. Hays, pp. 139-183. Memoir 126. New
Zijderveld, J. D. A., Hilgen, F. J., Langereis, C. G., Verhallen,
York: Geological Society of America.
Opdyke, N. D., Glass, B., Hays, J. D., and Foster, J. 1966. A
paleomagnetic study of Antarctic deep-sea cores. Science
154:349-357.
Prell, W. 1982. Oxygen and carbon isotope stratigraphy for the
107:697-714.
Quaternary of Hole 502B: evidence for two modes of
Late Cenozoic changes of flora in extra-tropical Eurasia in the light of

paleomagnetic stratigraphy
VLADIMIR P. GRICHUK
Introduction
Most studies of the problem of stratigraphic resolution in the

transition zone between the Pliocene and the Pleistocene have
been based on geology, mammals, malacofauna, or marine
microfauna. Paleobotany has been used to a much lesser extent,
and mainly for interpretation of the vegetational changes
resulting from continental glaciations. Good examples of this
approach are the studies of Sue and Zagwijn (1983) and Lona
and Bertoldi (1973). Both in those studies and in the majority of
others, evolution of flora has been considered on a quite limited
scale and as a rule has concerned only evolutionary sequences in
the development of individual species (mainly of water plants
and bog plants). Such studies seldom have been concerned with
overall changes in regional floras.
The paleobotanic data on the deposits of the late Cenozoic
period show, quite reliably, that the flora of the Neogene system,
which was of a subtropical character, was considerably richer
than that of the Quaternary. The latter, in the whole extratropical space of the Northern Hemisphere, had the properties of
a temperate-type flora. That gives us grounds to believe that
studies of the processes by whichfloristictypes change may yield
quite important insights into stratigraphic problems at the border
between the Neogene and the Quaternary.
Unfortunately, a detailed characterization of the successive
paleofloras of the late Cenozoic period is possible, at present, in
only a few well-studied regions. There is, however, one circumstance that can greatly help in obtaining the data necessary for
such an investigation. During the late Cenozoic period, the
vegetation cover over the greater part of extra-tropical Eurasia
was clearly of forest character. In botanic studies, all of the main
types of extra-tropical forests (boreal, nemoral, subtropical
shrub-arboreous) usually are described according to tree species.
According to data reported by Sokolov and Svyazeva (1965),
however, the characteristic differences, in terms of both systematics and diversity, are already accounted for when dendroflora
are classified at the genus level. Genera are also more certainly
identifiable, and more widespread, than any of the species of
which they are composed. The genus is therefore the most
appropriate basis for a reliable and regionally applicable charac104
terization of the late Cenozoic forest floras. Furthermore, studies

of fossil dendroflora at the genus level can considerably broaden
the usefulness of paleobotanic materials, in which palynology
based on the determination of tree pollen at the genus level rather
than the species level occupies the central place.
Paleofloral associations and climate change
In this review we shall consider data from a few regions within

the extra-tropical part of Eurasia. These regions have been
chosen according to the following two criteria: (1) adequate
sample availability across the Pliocene-Pleistocene boundary
interval and (2) the availability, in the given region, of
paleomagnetic definitions, which make it possible to establish
reliable correlations between the characterized horizons and the
general paleomagnetic scale. The latter requirement, being
absolutely necessary, has limited this study to the six regions
described herein (Figure 8.1).
The genera of dendroflora in the Upper Cenozoic deposits of
these regions have been compared with the present distributions
of all the species in each genus. In this way it is possible to group
the genera according to the modern areas in which they are most
abundant and most diversified, which for the most part reflects
their basic adaptive characteristics. The use of this system allows
us to follow the process of change in the compositions of dendrofloras and to interpret the differentiation of floristic adaptations
during the late Cenozoic period in terms of climatic change.
The geographic associations of genera (Grichuk, 1959) according to this method of organization are as follows:
1.
2.
3.
4.
5.
6.
Panholarctic
American-Eurasiatic
American-Mediterranean-Asiatic
American-Eastasiatic
Eastasiatic
North American (abbreviated as N.A. or A. in some
tables)
7. Tropical, including subgroups of pluricontinental and
South Asiatic-American (abbreviated as Tr. in some
tables)
Macujruo
Figure 8.1. Map of regions for which the changes in the genus diversity of Upper
Cenozoic dendroflora were analyzed (see Tables 8.1-8.6): 1, The Netherlands; 2, the
Bashkirian foreland of the Urals; 3, Primoriye; 4, northern Italy; 5, the Kura lowland
and the southeastern foothills of the High Caucasus; 6, the Pamirs.
106
Vladimir P. Grichuk
The Netherlands
Table 8.1 shows data on changes in the regional dendroflora in
the period from the middle Pliocene (Brunssumian) to the
Holocene. In this table, as in the others in this chapter, only
horizons with thermophilic flora are shown. These horizons
correspond to the subdivisions of the Pliocene and interglacial
epochs distinguished in the stratigraphic scheme of Holland
(Zagwijn, 1963). Data on the flora of glacial horizons are not
shown, because those floras are not important in showing the
evolution of forest floras.
South Ural forelands and Bashkiria
Changes at the genus level in the late Cenozoic dendroflora in
this region are shown in Table 8.2. In this region, all the Pliocene
horizons, as well as those corresponding to the interglacial
epochs distinguished in the stratigraphic scheme of Yakhimovich
(1970; Yakhimovich and Suleimanov, 1981), are present. Detailed paleomagnetic studies have been undertaken within the
Bashkirian foreland, making possible comparison of the stratigraphic horizons with the paleomagnetic scale (Yakhimovich and
Suleimanov, 1981). Long-term paleocarpological and palynological studies have provided materials that fully highlight the
changes in composition of the dendroflora through the late
Cenozoic deposits.
Primoriye
The materials from the Primoriye territory are relatively limited.
The most diverse and consistent are those for the late Cenozoic
in the south of the region, as published in two generalized
monographs (Korotkiy, Karaulova, and Troitskaya, 1980;
Golubeva and Karaulova, 1983). Table 8.3 shows Upper
Cenozoic horizons in the stratigraphic scheme of Korotkiy
(Korotkiy et al., 1980). Unfortunately, there have been no
systematic paleomagnetic studies within Primoriye, but the
studies by Alekseev (1978) and Korotkiy et al. (1980) make it
possible to establish a definite link between the stratigraphic
scheme in this region and the paleomagnetic scale.
Northern Italy
The paleofloristic history in this region is presented in Table 8.4,
based on the stratigraphic scale of Selli (1967). The paleomagnetic investigations undertaken in Italy have pertained mainly to
its southern part, with only a few for northern Italy. Nevertheless, biostratigraphic analysis makes it possible to compare the
stratigraphic horizons with the paleomagnetic scale (Ryan,
1973). In this region, the paleobotanical material consists mainly
of fossil pollen, and we have detailed descriptions of macrofloral
remains (leaves, seeds, etc.) only for the Astian Stage (i.e.,
continental Upper Pliocene) and certain parts of the Calabrian
Stage.
Kura lowland and southeastern Greater Caucasus

foothills
Data on this area are given in Table 8.5, based on the
stratigraphic scale of Isaeva-Petrova (1972). Paleomagnetic
research here has been extensive, and there are no doubts as to
the comparison of the stratigraphic sequence with the paleomagnetic scale (Grishanov et al., 1983). The abundant paleobotanic
material has allowed us to document the late Cenozoic
dendrofloral history in detail.
We cannot dwell on the voluminous body of information
concerning the western Caucasus region in this chapter. In
particular, the situation in the Kolkhida, or Chalcedonia, with its
enormously rich relict flora, makes it hardly comparable with
other territories. However, similar general tendencies are quite
clearly reflected here as well.
The Pamirs
Generalizations about the paleobotany of the Pamirs are difficult
because of many controversies about Upper Cenozoic stratigraphy (Pakhomov, 1980). The compilation in Table 8.6, which
reflects the changes of the fossil dendroflora in the Pamir
deposits, is based on the stratigraphic scheme of Chediya (1971).
The great volume of paleomagnetic data from this region makes
it possible to link this succession quite definitively to the
paleomagnetic scale (Dodonov, 1980). At the generic level, the
late Cenozoic history of the continental dendroflora is fairly well
known (although with a certain degree of fragmentation), mainly
through the palynological studies of Pakhomov (1980, 1983).
Summary
The data given in Tables 8.1-8.6 show that one clear and sharp
change was associated with or was quite close to the
paleomagnetic reversal at the Gauss-Matuyama boundary,
coeval with the transition from Reuverian to Tiglian paleofloras
in Holland. Another sharp change between "Astian" (i.e.,
Piacenzian) and Calabrian paleofloras in Italy is close to the
Olduvai subchronozone. In eastern Europe, the most important
reorganization was associated with the middle horizons of the
Akchagylian. Research on the history of the flora of the southern
European part of the former USSR has shown that this
phenomenon was associated with the establishment there of a
climate with winter periods characterized by temperatures below
freezing (Grichuk, 1959). In that period, representatives of
tropical and North American genera completely disappeared
except for certain species of North American origin that were
preserved in the southern and western Caucasus.
In the Lower Pleistocene stratigraphic horizons that overlie
beds with lower Matuyama paleomagnetic polarity there is only a
gradual reduction in the other groups that are now extinct in the
respective regions. The final disappearances of genera belonging
to presently "alien" groups (now occurring in restricted or
displaced ranges, as compared with their Pliocene distributions)
Late Cenozoic changes of Eurasian macroflora
107
Table 8.1. Dendroflora of Upper Cenozoic warm-climate horizons of The Netherlands

BRUNHES
MATUYAMA
GAUSS
iOlduvai
GENERA
Brunssum
Waal
Reuver
Praetiglian
Cromer
HolsteinTreene
Holocene
Eem
Saale Warthe Weichsel
Menap
Pinus
Salix
Myrica
Alnus
Rhamnus
Cornus
Sambucus
Viburnum
Picea
Abies
Carpinus
Corylus
Fagus
Quercus
Ulmus
Ilex
Acer
Tilia
Fraxinus
Pyrus
Vitis
Staphylea
Celtis
Juglans
Castanea
Ostrya
Pterocarya
Rhus
Zelcova
Liquidambar
Aesculus
Stirax
Diospyrus
Elaeagnus
Parthenocissus
Tsuga
Carya
Magnolia
Liriodendron
Pyrularia
Nyssa
Stewartia
Fothergillia
Meliosma
Barchemia
Torre va
Eucommia
Sciadopitys
Phyllodendron
Actinidia
Halesia
Pseudolarix
Corylopsis
Cunninghamia
Cyclocaria
Glyptostrobus
Schizandra
Taxodium
Sequoia
Simplocos
Alangium
Number of genera
Floral groups
61
51
37
33
25
23
20
18
III.
Note: Cold-climate intervals are indicated in italics. Correlation of the paleofloral units of The
Netherlands with the paleomagnetic scale is according to Zagwijn (Chapter 16, this volume).
108
Vladimir P. Grichuk
Table 8.2. Dendroflora of Upper Cenozoic warm-climate horizons of the Bashkirian piedmont of the Urals
BRUNHES
MATUYAMA
GAUSS
lOlduvai
GENERA
II
1,2
zr
o
7,8
5,6
3,4
9, 10
11, 12, 13
11. 14, 15
18
15
Pinus
Abies
Picea
Salix
Populus
Betula
Alnus
Prunus
Sambucus
Viburnum
Larix
Swida
Myrica
Corylus
Quercus
Ulmus
Acer
Tilia
Fraxinus
Carpinus
Ilex
Fagus
Taxus
Celtis
Pterocarya
Juglans
Zelcova
Elaeagnus
Cerasus
Vitis
Paliurus
Tsuga
Chamaecypris
Carya
Abelia
Aralia
Liriodendron
Actinidia
Ealauterococcus
Phyllodendron
Weigelia
Number of genera
Floral groups
41
25
23
23
19
15
III.
Note: Correlation of the stratigraphic sequence (Yakhimovich, 1970) to the paleomagnetic scale is according
to the data of Yakhimovich and Suleimanov (1981). The main fossiliferous sections are as follows: 1,
Simbugino; 2, Belekes, Kumurly, Khabarovka; 3, Voevodskoye (lower flora); 4, Nagayevo, Tukayevo; 5,
Akkulayevo; 6, Chiki-Anachevo; 7, Chui-Atasevo; 8, Tirlan'; 9, Baisakal; 10, Afonasovo; 11, Voevodskoye
(upper flora); 12, alluvium of Terrace IV, Belaya River basin; 13, Gremyachy Creek; 14, Minueshty Creek;
15, alluvium of Terrace II, Belaya River basin; 16,floodplainalluvium of Belaya River basin.
109
Table 8.3. Dendroflora of Upper Cenozoic warm-climate horizons of the Primoriye region, eastern Siberia
GAUSS
BRUNHES
MAT.
iOlduv.
GENERA
(Q
Lower
Suifun
Suite
N1/2
Upper
Suifun
Suite
N1/2
1,2
no
I
CD
p
m
Upper
Krasnozvetna
N2/2
Ussuriya
Q1/1
5,6
3,4
Sungach
Q
3/ll
Khankay
5,10,11
5,9
7,8
Nakhodkin
Chernoruch'y
5,12
Holocene
Q, v
13, 14
Pinus
Abies
Picea
Larix
Betula
Alnus
Myrica
Carpinus
Corylus
Quercus
Ulmus
Tilia
Fraxinus
Acer
Fagus
Castanea
Ilex
Juglans
Syringa
Pterocarya
Zelcova
Rhus
Ostrya
Liquidambar
Celtis
Morus
Aralia
Tsuga
Carya
Torreya
Nyssa
Phyllodendron
Kalopanax
Weigelia
Cryptomeria
Sciadopitys
Glyptostrobus
Engelhardtia
Ginkgo
Taxodium
Sequoia
Planera
Number of genera
Floral groups
42
36
30
28
I.
19
23
II.
17
17
16
III.
Note: Correlation of the stratigraphic sequence of Korotkiy et al. (1980) to the paleomagnetic scale is according to the
data of Alekseev (1978) and Korotkiy et al. (1980). The main fossiliferous sections are as follows: 1, Perevoznaya Bay;
2, Povorotny Cape; 3, Krasnozvetna Series of Tokhtin depression; 4, Spassk-Dalny; 5, Ussuri-Khankai depression; 6,
Bolshaya Ussurska River; 7, Melgunovka River mouth; 8, interfluve of Sungach and Ussuri rivers; 9, Terney village;
10, Vostok Bay; 11, Tumangan River; 12, Belaya Scala Bay; 13, Tal'ma Lake; 14, Amur Bay.
110
Vladimir P. Grichuk
Table 8.4. Dendroflora of Upper Cenozoic warm-climate horizons of northern

Italy
GAUSS
BRUNHES
MATUYAMA
iOlduvai
GENERA
f
5 a
Astian
Calabr.
(Lower)
DonauGunz
GiinzMindel
MindelRiss
RissWiirm
Holocene
Pinus
Abies
Populus
Salix
Betula
Alnus
Rhamnus
Viburnum
Picea
Larix
Taxus
Carpinus
Corylus
Fagus
Quercus
Ulmus
Ilex
Acer
Tilia
Fraxinus
Ostrya
Castanea
Vitis
Diospyrus
Cedrus
Pterocarya
Zelcova
Juglans
Aesculus
Laurus
Liquidambar
Platanus
Amygdalus
Tsuga
Carya
Pseudotsuga
Magnolia
Nyssa
Benzoin
Sophora
Sapindus
Borchemia
Eucommia
Keteleeria
Cephalotaxus
Ginkgo
Planera
Taxodium
Asimina
Ptelea
Geonoma
Ficus
Pheobe
Litsea
Cassia
Machaerium
Celastrus
Sterculia
Terminalia
Leuconthoe
Porana
Persea
Eugena
Appolonias
Boscia
Pittosporum
Combretum
Number of genera
65
41
33
30
28
24
23
Floral suites
Note: Correlation of the stratigraphic sequence of Selli (1967) to the

paleomagnetic scale is according to the data of Ryan (1973). The main sections
are as follows: 1, Asti; 2, Mongardino; 3, Principe; 4, Stirone; 5, Leffe; 6,
Pianura Padana; 7, Ca Marcozzi; 8, Padova; 9, Pianico-Selleri; 10, Polgaria.
111
Table 8.5. Dendroflora of Upper Cenozoic warm-climate horizons of the Kura depression and the southesat
piedmont of the Urals
MATUYAMA
GAUSS
BRUNHES
lOlduvai
GENERA
Productivna
Akchagyl
Apsheron
2,3
o.
CD
Z3
S-S"
O
'
Khazar
Baku
6,8
6,7
4,5
Khvalyn
Holocene
9, 10, 11
Pinus
Salix
Populus
Betula
Alnus
Rhamnus
Lonicera
Picea
Abies
Myrica
Carpinus
Corylus
Fagus
Quercus
Ulmus
Acer
Tilia
Fraxinus
Ilex
Celtis
Pyrus
Rhus
Vitis
Punica
Elaeagnus
Pterocarya
Ostrya
Parrotia
Castanea
Juglans
Buxus
Vitis
Morus
Zelcova
Cedrus
Carya
Tsuga
Nyssa
Libocedrus
Aralia
Paulovnia
Thuja
Platicarya
Taxodium
Sequoia
Cinnamomum
Persea
Number of genera
Floral suites
46
35
38
I.
28
25
II.
21
20
III.
Note: Correlation of the 1963 MSC (Modern Stratigraphic Code) standard sequence of the USSR to the
paleomagnetic scale is according to data of Grishanov et al. (1983). The main sections are as follows: 1, Baku or
Bakinsky Archipelago, western Caspian; 2, Shirak steppe; 3, Kvabebi; 4, Lengibiz Ridge; 5, Oblivnoi Island; 6,
Tagirkent; 7, Divichi; 8, Kysyl-Burun; 9, Kudialchai; 10, Shura-Ozen; 11, Binagady.
112
Vladimir P. Grichuk
Table 8.6 Dendroflora of Upper Cenozoic warm-climate or interglacial sequences of the Pamir Range,
Tadjikistan
BRUNHES
MATUYAMA
GAUSS
Olduvai!
Polizak
GENERA
"Gan"
subtillite
Nizhnekilimbin
Kokbai
intergl.
3,4,5
Dushanbe
llyak
Kuruksay
Akdzhar
intergl.
Early
Altyndar
6,7
Late
Altyndar
Amudar
Holocene
9, 10
Juniperus
Ephedra
Salix
Betula
Pinus
Picea
Alnus
Rosa
Comus
Abies
Rhododendron
Populus
Hippophae"
Corylus
Quercus
Ulmus
Acer
Tilia
Fraxinus
Carpinus
Fagus
Ilex
Pistacia
Rhus
Elaeagnus
Cedrus
Juglans
Celtis
Berberis
Platanus
Tamarix
Pterocarya
Zelcova
Ostrya
Morus
Vitis
Zygophyllum
Liquidambar
Tsuga
Carya
Menispermum
Fothergillia
Platycaria
Engelhardtia
Cerdidiphyllum
Glyptostrobus
Corylopsis
Taxodium
Altingia
Sabal
No. of genera
Floral groups
51
40
36
30
II.
26
25
III.
Note: Correlation of the stratigraphic sequence of Chediya (1971) to the paleomagnetic scale is
according to data of Dodonov (1980). The main sections are as follows: 1, Orta-Uchkul' (lower
levels); 2, Khyrga-Dara; 3, Orta-Uchkul' (upper levels); 4, Kokdzhar-Uchkul'; 5, Khiriak-Dara; 6,
Akdzhar; 7, Karatau-1; 8, Lakhuti; 9, Ogzikichik; 10, Khudzhi; 11, Shugnou.
113
palynology (in Russian), ed. V. P. Grichuk, pp. 193-212.

Moscow: Inst. GeoL, Akad. Nauk SSSR.
Korotkiy, A. M., Karaulova, L. P., and Troitskaya, T. S. 1980.
Quaternary deposits of the Primorie: stratigraphy and
paleogeography (in Russian). Novosibirsk: Nauka.
Lona, R, and Bertoldi, R. 1973. La storia del Plio-Pleistocene
Italiano in alcuna sequenze vegetazionali lacustri e marine.
Acad. Nazion. Lincei., Atti, Ser. 8 9(3): 1-46.
Pakhomov, M. M. 1980. Paleogeographical aspects of the history
of the mountain vegetation of Central Asia (on the
example of the Pamir-Alaya (in Russian). Botanicheskii
Zhurnal 65:1138-1148.
Pakhomov, M. M. 1983. New data on the paleogeography of the
loess soil series of Central Asia (in Russian). Akad. Nauk
SSSR, Doklady 270:967-972.
Ryan, W. B. F. 1973. Paleomagnetic stratigraphy. In Initial
reports of the Deep Sea Drilling Project, vol. 13, no. 2, ed.
References
W. B. F Ryan and K. J. Hsu, pp. 1380-1387. Washington,
Selli, R. 1967. The Pliocene-Pleistocene boundary in Italian
Alekseev, M. N. 1978. Anthropogene of Eastern Asia. Stratimarine section and its relationship to continental stratiggraphy and correlation (in Russian). Moscow: Nauka.
raphies. In Progress in oceanography, vol. 4, ed. M. Sears,
Chediya, O. K. 1971. South-central Asia in the most recent epoch
pp. 67-86. New York: Pergamon Press.
oftectonism (in Russian). Frunze: Him.
Sokolov, S. Y., and Svyazeva, O. A. 1965. Chorology of ancient
Dodonov, A. E. 1980. Principles of the stratigraphic subdivision
plants of the USSR. Komarov lectures, XVII (in Russian).
of the upper Pleistocene-Quaternary deposits of TadjiMoscow: Nauka.
kistan. In Boundaries of Neogene and Quaternary systemsSue, J. P., and Zagwijn, W. H. 1983. Plio-Pleistocene correla(in Russian), ed. K. V. Nikiforova and A. E. Dodonov,
tions between the northwestern Mediterranean region and
pp. 12-22. Moscow: Nauka.
northwestern Europe according to recent biostratigraphic
Golubeva, L. V., and Karaulova, L. P. 1983. Vegetation and
and palaeomagnetic data. Boreas 12:153-166.
climatostratigraphy of Pleistocene and Holocene of the Yakhimovich, V. L. 1970. Cenozoic of Bashkirian cis-Urals, vol.
southern Far East of the USSR (in Russian). Moscow:
II, part 3 (in Russian). Moscow: Nauka.
Nauka.
Yakhimovich, V. L., and Suleimanov, F. I. 1981. MagneGrichuk, V. P. 1959. Lower boundary of the Quaternary period
tostratigraphic section of Pliocene and lower Pleistocene
(system) and its stratigraphic position on the Russian
infraglacial zone of the cis-Urals. Commission for RePlain. Inst. GeoL, Akad. Nauk SSSR, Trudy 77:5-90.
search on the Quaternary Period (in Russian). Akad.
Grishanov, A. N., Eremin, V. N., Imnadze, Z. A., Kitovani,
Nauk SSSR, Byul. 51:31-37.
T. G., Kitovani, Sh. K., Molostovskyi, E. A., and Zagwijn, W. H. 1963. Pleistocene stratigraphy in the NethTorozov, R. I. 1983. Stratigraphy of upper Pliocene and
erlands, based on changes in vegetation and climate.
Kon. Ned. Genoots. Mijn. GeoL, Verh., GeoL Ser. 21:
lower Pleistocene deposits of Guria (Western Georgia)
173-196.
according to paleontological and paleomagnetic data (in
Russian). Komiss. Izuch. Chetvert. Perioda Akad. Nauk Zagwijn, W. H. 1974. The Pliocene-Pleistocene boundary in
western and southern Europe. Boreas 3:75-97'.
SSSR, Byul. 52:18-28.
Isaeva-Petrova, L. S. 1972. Reconstruction of the altitudinal Zhamoida, A. I., Kovalevsky, O. P., Moisejeva, A. L., and
vegetation zones of the eastern part of the Greater
Yarkin, V. I. 1977. Stratigraphic code of the USSR (in
Caucasus in Apsheronian time (in Russian). In Pleistocene
Russian and English). Leningrad: VSEGEI.
are observed somewhat above levels containing the BrunhesMatuyama boundary.
At the stratigraphic level corresponding to the Olduvai event,
sharp changes in dendroflora at the specific level have not been
recorded. On the other hand, at the genus level all Eastasiatic
genera, with the exception of Eucommia, disappeared from
northern Italy and Holland at that time, including such cosmopolitan American-Eastasiatic genera as Taxodium, Torrea,
Magnolia, and Nyssa. On the whole, those changes were similar
in scale to the later major floristic changes seen during the
Pleistocene.
Plio-Pleistocene mammal faunas: an overview

EMILIANO AGUIRRE, ELEANORA A. VANGENGEIM, JORGE MORALES, MARINA V. SOTNIKOVA, and
VLADIMIR S. ZAZHIGIN
Introduction
The evidence from mammalian biostratigraphy with regard to

the boundary between the Neogene and the Quaternary, or,
more precisely, the base of the Pleistocene, can be analyzed in
various ways. Long-established regional biostratigraphic scales
are still widely used, although for the most part they are based on
uneven, inadequate evidence and thus are subject to differing
and unreliable interpretations. In seeking for greater reliability,
vertebrate paleontologists in the past few decades have attempted to subdivide the geological time scale into probabilistic
mammal ages and mammal zones, such as the MN (Mammal
Neogene) zones of Mein. That approach is equally subject to
imprecision and subjective bias (De Bruijn et al., 1992;
Fahlbusch, 1991), and scholars may, consciously or unconsciously, constrain the paleofaunal changes to coincide with a
particular magnetic reversal, a climate event, or any other
significant, even preselected, level.
The objective approach is to begin with the ages of local faunal
horizons that are directly based on radiometric dates or
paleomagnetic analysis and build a biochronologic framework on
this ground (e.g., Lindsay et al., 1987). Interpolations are
legitimate, as are correlations based on faunal similarities, when
the limits of probable error are realistically included. In cases
where a local fauna presents many first appearances or last
appearances, it is reasonable to suspect a lack of information in
the preceding or following time interval, respectively. When the
sampling factor is corrected, if unusual numbers of earliest and
latest occurrences are still seen at one level, a genuine faunal
overturn can be inferred. Normally, such overturns are more
likely to have been the result of shifting paleobiogeographic
boundaries (due either to shifts in regional environmental
conditions or to relaxation of previous physical barriers) than to
have been coincidental speciation or extinction events in the
endemic regional fauna.
Plio-Pleistocene mammalian faunal succession in
Europe
Gauss chronozone. The pre-Pleistocene local faunas of fossil

mammals from the older part of the Gauss chron in western
114
Europe include those at Layna, with evidence of a change from

warm to cold climate, Las Higueruelas, dated to 3.2 Ma or
earlier (Aguirre and Morales, 1990), and Triversa (De Giuli et
al., 1983). These are succeeded by the l.f. (local fauna) of
Vialette, dated to about 3 Ma, and its correlative at Villarroya.
In central Europe, the Hajnacka l.f. is also referred to the Gauss
(Fejfar, Heinrich, and Heintz, 1990), as are the Ripa Skortselskaya l.f. (Konstantinova, 1967) and the Tulucesti-Cernatesti
group of localities (Ghenea, 1981) in the southeast region of
Europe (Figure 9.1).
Changes in the rodent paleofaunas in the middle or, perhaps,
the earlier part of the Gauss chronozone are evident in the local
faunas of Arondelli and Hajnacka in Europe and at Betekia in
western Asia. In eastern Europe, Dolomys and Borsodia are
characteristic of this interval, with the latter genus also known in
Siberia and Transbaikalia. Species of Mimomys (M. hintoni, M.
polonicus, M. stehlini) were widely distributed at the beginning
of that time throughout Eurasia. Among large mammals, several
last occurrences characterize the record of that time interval,
while the FAD (first-appearance datum, or lowest stratigraphic
occurrence) of Mammuthus occurs with Equus in Ripa
Skortselskaya (Konstantinova, 1967). The genus Equus has been
reported from Malusteni in Romania (Ghenea, 1981; Chapter
20, this volume) and Etulia in Moldava, both localities being
assigned to younger Gilbert ages (i.e., 4 Ma, at least). No other
localities with Equus older than 3 Ma are known in Europe or in
Asia, nor even in North America, where the genus appears to
have had its origin (Lindsay, Chapter 30, this volume). This
evidence suggests either that Equus evolved in the lower Danube
basin more than 1 m.y. before it appeared anywhere else or that
the attribution of the Romanian and Moldavan sites to the
Gilbert is wrong.
The earliest occurrences of Equus in western Europe are
documented in the Rincon 1 local fauna (Alberdi et al., 1982)
and in the Roccaneyra and Etouaires local faunas in France,
dated to 2.5 Ma. A later occurrence was reported from the
Montopoli l.f. in Italy, which was placed in the lowest level of the
Matuyama chron (Lindsay, Opdyke, and Johnson, 1980), and the
Huelago-Carretera l.f. of Spain (Alberdi et al., 1989) is probably
of similar age. In these two latter faunas, Mammuthus is also
Plio-Pleistocene mammal faunas
present. The FAD of Mammuthus in western Europe is in the

Valdeganga l.f., in a horizon geographically and chronologically
close to Rincon 1 (Alberdi et al., 1982). The dispersal of the true
elephants and horses into this part of Europe is thus certain to
have occurred before 2.6 Ma.
Among a number of faunal events in these horizons we should
mention the LAD (last-appearance datum, or highest stratigraphic occurrence) of Ursus minimus, Cervus perrieri, and
Arvernoceros ardei and the FAD of Procamptoceras and
Gallogoral. The changes in ruminant populations can be dated to
just before the Gauss-Matuyama paleomagnetic reversal. The
local faunas of Kaltensundheim and Stranzendorf C illustrate the
same time slice for central Europe, the former with normal
magnetization, and the latter virtually coincident with the
Gauss-Matuyama reversal, according to Rabeder (1981; von der
Brelie et al., Chapter 17, this volume).
Lower Matuyama chronozone. The fossil record for mammals in
Europe between the base of the Matuyama chronozone and the
Reunion excursions (2.6-2.2 Ma) is poor. During much of that
time span, marine regression, tectonism, and erosion reduced
the sedimentary record over wide areas of the continental
regions. On the other hand, a number of reliably dated postReunion sites are known. These include the following: Seneze,
with reverse polarity, above a normal (Reunion?) excursion,
which would indicate an age of 2 Ma or less (Boeuf, 1990); Le
Coupet, dated to 1.92 Ma (Bonifay, 1991); Chilhac, which is not
much younger than 1.85 Ma (Boeuf, 1990). Saint-Vallier and
Puebla de Valverde have yielded fossil assemblages no younger
than those from the dated sites and thus should be dated to
approximately 1.7 or 1.8 Ma. The preservation of the SaintVallier fossils in a loessic formation is indicative of cold
(Eburonian?) climate. In eastern Europe, the Liventsovka l.f.
(Virina, Dobrodeev, and Faustov, 1971) and the lower part of
the Kryzhanovka section (Tretyak and Volok, 1974) have a
reversed remanent paleomagnetic polarity that corresponds to a
part of the Matuyama between Reunion and Olduvai. The
Graunceanu l.f. and Tetoiu l.f. in Romania (Bolomay, 1965;
Radulesco and Samson, 1990) are pre-Olduvai as well.
Several biochronologically important events in the evolution
and dispersal of Arvicolidae occurred during that time, which we
can consider as latest Pliocene, but the dating is not completely
established. Repenning (1987) estimated an age of 2.6 Ma for the
migration of Clethrionomys, Lagurodon, and Pliomys into
Europe; further documentation of the actual FAD of Pliomys
comes from the lower horizons of Liventsovka (Aleksandrova,
1976), which are significantly older than 2.1 Ma and perhaps
near the base of the Matuyama, and from Weze, which is
considered to be older than 3.0 Ma. Evolution in Mimomys
produced M. pliocaenicus near the time of the Gauss-Matuyama
reversal (Vangengeim, 1977; Nikiforova and Aleksandrova,
1987), according to the conventional correlation of the base of
the Khaprovian "complex" in which the FAD of this species in
European Russia is recorded. Chaline (1986; Chapter 14, this
volume), however, has assigned that event to a level within the
115
lower Matuyama. The material referred to Mimomys cf. M.

pliocaenicus at Rincon 1, Spain, is assigned to the later Gauss (as
mentioned earlier); this species is well documented in all
horizons at Liventsovka and Kislang, together with M. reidi, and
also at Gundersheim.
In the interval between the Reunion and the Olduvai
subchronozones, the faunae are distinguished by several LADs
and FADs and by a number of taxa that occur exclusively in this
particular group of sites. The last occurrences of Mastodon,
Anancus, Hipparion, Nyctereutes, Gazella, Gazellospira, and
Gallogoral are noted here, and the new cervid taxa "Cervus"
rhenanus, Pseudodama pardinensis, and Pseudodama philisi did
not survive into the Olduvai subchronozone. In addition, the
LAD of Croizetoceros ramosus in western Eurasia appears to be
at Chilhac. Most of the Tegelen faunal succession can be
correlated with this assemblage (Masini and Torre, 1990),
although uppermost Tegel levels probably date to the lower part
of the Olduvai subchronozone.
Upper Matuyama chronozone. Evidence of rapid replacement in
arvicolid species starts with the FAD of Allophaiomys in the
upper Tegel deposits of The Netherlands (Van Kolfschoten,
1990), in the lowermost Olduvai subchronozone. Such replacements appear to have taken place almost simultaneously in North
America (Lundelius et al., 1987; Repenning, Fejfar, and
Heinrich, 1990; Repenning and Brouwers, 1992). The species A.
deucalion occurs slightly later in strata dating to the upper
Olduvai at Villany 5 and Kadzielna (Vangengeim, 1977), and
possibly at Orce (as discussed later). The FAD of Mimomys
tornensis, whether at Kadzielna or at the age-equivalent site of
Almenara-Casablanca (Esteban Aenlle and Lopez Martinez,
1987), the FAD of Mimomys pusillus in the upper horizon at
Liventsovka, the FAD of Villanyia exilis in Villany 5, and the
FAD of Mimomys ostramosensis in the lower horizons of
Kolinany also occur within the age limits of the Olduvai
subchronozone (cf. Rabeder, 1981).
Among large mammals, the earliest occurrences of Panthera
gombaszoegensis, Canis etruscus, Pachycrocuta brevirostris,
Pseudodama nestii, and Eucladoceros dicranios have been
reported from Olivola (Azzaroli et al., 1986), jointly with the last
occurrences of Chasmaporthetes and Procamptoceras. The precise age of Olivola has been difficult to establish; De Giuli and
Masini (1986) proposed a post-Olduvai age, on faunal evidence
that it postdates the Tegelen fauna, in rough correlation with
climatic cooling and the Aullan diastrophism. This is, however,
probably equivalent to the Eburonian cold-climate phase that
actually begins in the uppermost Olduvai subchron (Pasini and
Colalongo, Chapter 2, this volume). In the AlmenaraCasablanca cave filling, where Mimomys tornensis is found
together with M. medasensis and Mimomys aff. M. rex, we find
the latest occurrence of Gazellospira torticornis, together with
early occurrences of Pachycrocuta brevirostris, Canis etruscus,
and Pseudodama nestii (Soto and Morales, 1985). At this site, a
species of Ovibovini, more progressive than the Megalovis from
Seneze and resembling the Praeovibos from Sinzelles (Moya-
116
Aguirre et al.
Magnetic Western
Europe
Ma scale
oTi
0.6
0.7
C-Europe
& Balkans
Eastern
Europe
Betfia 5
KARAI-DUBJNA
South Africa
N. Africa North+West
Mid.-East Asia
C-East
Africa
glo.73 ATAPUERCA 3
0 . 8 Ml 1
OLDUVAI IV
LAKHUTI2
Oler
VIATKINO
SOLILHAC
UO.88
H O 94 VALLONET
0.9
1
1.1
Monte Peglia
Pirro
Imola
Venta Micena
1.2
Swartkrans 3
NOGAISK-2
UNTERMASS FIELD SENNAYA BALKA
1
1
Swartkrans 2
Ubeidiya
1
D. Altenburg.4
Betfia 2
OLDUVAI III
LAKHUT11
A
1
1.3
Cueva Victoria
1.4
SINZELLES
Casa Frata
Brielle
1.5
Psekups
Upper
D. Altenburg 2
Vcelare 3B1
KRYZHANOVKA Sterkfontein 5
Dmanisi
Kromdraai A
ATn- J
OLDUVAI II u A
Hanech
1
CHARI
A
1
SHUNGURA. L A
Kizikha
Osztramos 3
1.6
1
.;.;J.;.;J.;.;.;.;.;.
Olivola
II
LJi.72 TEGELEN 6
1.7
1.8
1 9
1 a
Kotiriaf>vS::::'
ORCE2
Almenara
KADZIELNIA
Villany 5
CHILHAC
COUPET
Kislang
SLATINA 2
iSwartk-raRSit::-: .QLQU.VAI.-JI-.FTV^V
SHUNGURA. J A
Kromdraai B3
1
OKOTE
A
1
OLDUVAI lllowA
SHUNGURA. H A
KBS_
A
OLDUVAI 1
A
LIVENTSOVKA
Khapry
KURUKSAI
PODPUSK
LEBYAZHIE
2.04 SENEZE
2.1
Saint Vallier
RIPPERSROD 4
2.2
SHUNGURA. G A
2.3
2 4
Mm a
2 5 "U2.47
at a W
aC a O
2.7
2.8
2.9
3
o ^
O.I
3.2
GOMARETI
MONTOPOLI
ROCCANEYRA STRANZENDC)RF
RINCON 1
C
Kotlovina?
ETOUAIRES
KALTENSUNDH.
KALOCHORO A
SHUNGURA. E A
Sterkfonte in 4
u. SHUNGURA. C
U3.I6
TULUCESTl
PJ3.07
BETEKE
ATnKADAHADAR A Brimba
2.88
U2.96
SHUNGURA. D A
BURGI Tuff-u A
Villarroya
VIALETTE
HIGUERUELAS
'riversa
RIPA SKORTSELSKAYA
HAJNACKA
MAKAPAN 3
LOMEKWI
A L. Ichkei 1
SHUNGURA. C A
DENEN DORA
TULU BOR
A
SHUNGURA. B A
Ichai
India
c
o>
c
Conglomerate
tit
1
GONGWAG-LING
O PearletteAshCUDAHY
Hartford Ash
COURTLAND
CANAL
IRVINGTON
8
Rock Creek
Q.
^"
CO
CD
JETIS
KEDUNG-BRU~BUS
TRINIL
M
P V
r A
r R
T T T
Upper
Arroyo seco
EL MUELLE
Lower
Arroyo Seco
C/ S/4/Af
Dongcun
Top of up.
TT nihowan
Warm
TARUA
co
CO
Upper
Yushe III
C
CO
Toledo X Ash max

* s S Pearlette Ash
SAPPA
1
E. Mediterrar 1.1
Land Bridge
I
(LB 1.2
Fisherman's
Warm
Cliff ?
1
<D
1
1
1
1
a. SATIR
HI
endemic
Arvicolidae
.-77
YOUNGER
LACUSTRINE, Youhe
Lower
Yushe III
FSD Equus
in India
Lower.
Lari'ar
Smilpdop ...
VALCnO'-'-'-'-
Corbicula Bed
WELLSCH
VALLEY
1
1
i
C. R. GIDLEY
BORCHERS 7
B PearlettQ Ash
1.4
Bioevent
Aridity
1.5
Cold, Low SL
1.6
Mon
Mira
VCasKfTvir-'-'-'-'-'
FVOUSXtJ\lû. . . . . .
1.3
CO (0
CO
C/)
* -
D>
CO
0.7
0.8
Cold
Compress.
Diastrophism 0.9
Cooling
V
Li.
0.6
Ma
0.5
CO
Boulder
Australia
events
& N . Guinea
South
America
North
America
09
CL
CO
111
Central Java
uja
o
a China
r-
117
SMEATON
::::::::::V::-:
Unconformity 1.7
Cooling
Endemism
11 a SJ
ft
Diastrophism
Panama LB
Behring L B V 1.9
Aridity
Diastrophism 2
Cold, Low SL
2 1
^
KAREWA Gr
FSD Equus
in Kashmir
Soricini India \
Illl LOWER
LACUS
TRINE, Youhe
"QUUS
Lower/Green
Nihowan
Yushe II
CO
ttttt
Figure 9.1. Pliocene-Pleistocene mammal correlations.
2.2
'Torn
Base of
Pala nkarina
Ensenadan
2.3
Ay. UBERTAD/
S.s MONTE BLANCO Top of \ /
Cool
2.4
Vorohue
ce Age, V
Continent
i
CITA CANYON
2.5
accretion
GRANDVIEW/
c
D
anama LB 2.6
CO
Warm
AWE
Land
Bridge)
c
2.7
Behring LB
CO
2.8
<D
SANDPCHNT
Early
3lancan
POST RANCH
5
rohi
Late
Blancan
D"
.ow sea level

Diastrophism
BARRANCA
DELOSLOBOS
2.9
"oo ling
S-ice sheet
arosion
Diastrophism
3.1
3.2
118
Aguirre et al.
Sola and Menendez, 1986), is also noteworthy. The Almenara- and a fossil Homo (Dzaparidze et al., 1991), has a local fauna in
Casablanca l.f. is therefore most probably close to Olivola and which most elements (other than the bovids) are held in common
Kadzielna in age, and possibly precedes Kolinany. All these with the localities mentioned earler for the upper Matuyama
localities probably should be dated to the upper part of the time span, above the Olduvai subchron.
Olduvai subchron, if not to the top.
The local small-mammal faunas of Brielle (Van der Meulen
Normal magnetic polarity, possibly representing the Olduvai and Zagwijn, 1974), Neuleiningen-15 (Fejfar and Heinrich,
subchronozone, has also been reported from Orce 2, a horizon in 1981, given as Neuleiningen-11), and Bagur (Lopez, Michaux,
the Baza Basin in southern Spain that yields a faunule that and Villalta, 1976), as well as the local faunas of Psekups,
includes Mimomys ostramosensis, together with Mimomys Akkulaevo, and the "Odessan Complex" of eastern Europe
pusillus, Allophaiomys cf. A. deucalion, Apodemus mystacinus, (Nikiforova, Chapter 21, this volume), may correspond to the
Castillomys crusafonti, Gazellospira torticornis, and Leptobos younger, relatively more stable part of this interval during the
etruscus (Agusti et al., 1987; Aguirre, 1989b). The closest transition from cool climate to the warmer Waalian phase.
correlation of this fauna is to Kolinany, close to the top of the
A succeeding interval of more rapid change began at about 1.2
Olduvai subchron. A similar situation has been suggested for the Ma, the approximate age of Betfia 2, Deutsch-Altenburg, and
Kotsakuri l.f. of Georgia (Gabunia and Vekua, 1981).
Mas Rambault. A number of local faunas have been dated by
All of these faunal horizons together represent a faunal interpolation in paleomagnetic profiles, such as the following:
sequence that gives a fairly good picture of the diversity of large the central European site of Untermassfeld, with reversed
and small mammals in Europe at the end of the Olduvai polarity that is either just older or just younger than the Jaramillo
subchron, and thus the time of transition from the Pliocene to subchron; Sennaya-Balka, Nogaisk, and Vallonet, all of which
the basal Pleistocene. The accelerated pace of evolution in are contemporary with the Jaramillo (deLumley et al., 1988;
mammalian faunas seen in this interval may represent a true Markova, 1990); and Solilhac, which also has been referred to
faunal overturn, or an accelerated tempo of replacement, during the Jaramillo (Bonifay, 1991), but which more probably is related
to a post-Jaramillo normal-polarity excursion. The youngest part
this transition.
As a continuation of the same rapid pace of adaptation, some of the Matuyama, from levels just prior to the Jaramillo up to the
of the arvicolids that appeared in the lower Olduvai subchron Brunhes reversal, also includes the undated but correlative local
became extinct near its top, such as Villanyia exilis, or slightly faunas of Betfia 2, Les Valerots, Monte Peglia, Cava Pirro,
above this level, such as Allophaiomys deucalion, more or less Imola, and Venta Micena (Agusti, 1986; Agusti et al., 1987; De
simultaneously with the first appearance of Allophaiomys Giuli, Masini, and Torre, 1990).
pliocaenicus. The occurrences of this taxon at Vcelare 3 Bl and
In the short interval from 1.2 Ma to the beginning of the
Kamyk are probably its earliest record (Nadachowski, 1990), and Matuyama at 0.78 Ma, important changes occurred in mamit is widespread and abundant throughout Europe in other sites malian microfaunas. The last Mimomys with closed-root incisors
that date to just above the Olduvai subchron (Chaline, Chapter (M. savini) appears in Betfia 2 and in the upper Kryzhanovka
14, this volume), at the beginning of an important evolutionary level, dated to about 1.2 Ma, as well as Deutsch-Altenburg 4 and
lineage for the Quaternary (Van der Meulen, 1973).
Cueva Victoria. In the same faunas, we find evidence of
Almost at the same time, a number of changes occurred cladogenesis with the earliest appearance of Microtus (subgenus
among the European large-mammal assemblages. The evidence Pitymys) and related genera such as Stenocranius and Iberomys.
comes from such sites as Sinzelles, with a date of 1.4 Ma The earliest occurrences of Microtus sensu stricto and Pitymys in
(Bonifay, 1991); Cueva Victoria in southern Spain, which is central and western Asia are penecontemporaneous, in Itantsa
similar to Sinzelles, or perhaps younger (Ferrandez et al., 1989); and Razdolie, respectively (Vangengeim, 1977; Zazhigin, 1980).
and Casa Frata, an Italian locality stratigraphically equivalent to Beginning with the pre-Jaramillo levels at sites such as Betfia 2,
the ancient Tasso faunal assemblage, which has been dated to be Venta Micena, Monte Peglia, and Les Valerots we also find new
older than Sinzelles and younger than Olivola (De Giuli and species of Allophaiomys: A. laguroides, A. nutiensis, and A.
Masini, 1986). The immigrations of Hippopotamus and Cervalces burgondiae (Van der Meulen, 1973; Chaline, 1986). Finally, at
are recorded at Tasso and Sinzelles, the latter also including the approximately the same time, the dispersal of Ellobius and the
earliest record of Praemegaceros in western Europe. The Cueva immigration of Eolagurus are recorded in the Nogaisk 1 l.f.
Victoria l.f. is important for yielding the earliest record of (Vangengeim, 1977; Markova, 1990).
Dolichodoriceros (= Praedama) and of Cervus ex gr. C. elaphus In large-mammal faunas dating to the upper Matuyama, the
(Azanza and Sanchez, 1990). Fossil canids questionably attrib- most notable change is the regional diversification of Mammuuted to Canis falconeri or Xenocyon are found at Casa Frata and thus meridionalis into progressive varieties (Aguirre, 1972): M.
Cueva Victoria, and the last mention of Leptobos etruscus also m. tamanensis from Sennaya Balka and many other localities in
comes from these localities. The local faunas of Tetoiu 2 Ukraine, southern Russia, Georgia, and Romania, M. m.
(Radulesco and Samson, 1990) and Lybakos (Steensma, 1988) in vestinus from Imola and Pirro in Italy, M. m. wuesti from central
the Balkan basins correspond to this part of the Matuyama; Europe, and perhaps other varieties. Remains of these progresTetoiu 2 records the last occurrence of Megalovis. The new site sive varieties have frequently been confused with M. trogonof Dmanisi, in Georgia, with occurrences of Bos, Capra, Ovis, therii. As an immigrant from Asia, Elasmotherium caucasicum
119
occurs in the upper Kryzhanovka l.f., shortly before the been unfounded. Replacements at the species and variety levels
Jaramillo (Tretyak and Volok, 1974); Dicerorhinus etruscus must in particular groups within diverse areas more probably were
be read for D. hemitoechus at Solilhac. Equus suessenbornensis influenced by minor climatic cycles and vegetational changes.
and Equus altidens replace the last representatives of the E.
Several species were consistently present in African faunas
stenonis group, Dama ex. gr. D. dama replaces Dama (Pseudo- between 3 Ma and 0.74 Ma, among them the two living
dama) nestii, and Cervalces latifrons replaces C. gallicus; at the rhinoceroses, the extinct Hipparion (Stylohipparion) lybicum,
same time, Megacerini and modern Capreolus become abun- the southern zebra Equus capensis, several bovids such as
dant. Bison schoetensacki and Soergelia also appear before the Antidorcas recki and Oreotragus major, and extant carnivores,
Matuyama-Brunhes reversal in some of these sites. The first including leopard, cheetah, serval, caracal, striped hyena, and
species of Crocuta, new canids such as Canis arnensis, Canis spotted hyena. In faunas collected from levels of middle Gauss
mosbachensis, and Canis tamanensis, and ursids such as Ursus age, about 3-2.7 Ma, such as Makapansgat Mb 3 (McFadden,
deningeri appear. The last occurrence of European Acinonyx is Brock, and Partridge, 1979), Shungura B, Denen Dora in the
noted at Le Vallonet, and the last Megantereon is found in Hadar Formation, and the Tulu Bor level in Koobi Fora, we find
Untermassfeld.
the last occurrences of many Lower Pliocene taxa that characterWarm-climate episodes were followed by the Menapian ize older associations at Langebaanweg, Kanapoi, Laetolil, and
cooling trend near the time of the Jaramillo and by an increase in Usno: Theropithecus darti, Australopithecus afarensis, Ancylocompressive tectonics in various regions, together with uplifting therium hennigi, Gazella vanhoepeni, and Simatherium kohllarand erosion, as, for instance, the Cassian unconformity in Italy. seni (Vrba, 1982). At slightly higher levels, for instance in the
The succeeding faunas were but slightly modified from those of lower Burgi Member of Koobi Fora, in Shungura C and D of the
the Olduvai-Jaramillo interval, as, for instance, in Akhalkalaki Omo Basin, and in Sterkfontein 4, the most notable events are
(Vekua, 1987), Karai Dubina (Markova, 1990), Petropavlovka, the last Nyanzachoerus and the earliest Metridiochoerus.
and Huescar (Alberdi et al., 1989), which are just slightly older
The next interval showing faunal changes in eastern Africa is
than the Matuyama-Brunhes transition, and in Atapuerca TD3 bracketed between 2.0 and 1.75 Ma and includes local faunas
(Gil, 1987; Aguirre, 1989a), a faunal horizon in which that from Shungura G to H, Olduvai Bed I-lower Bed II, and the
paleomagnetic reversal is recorded.
upper Burgi and the lower KBS members at Koobi Fora. First
occurrences during this interval are the African elephant
Loxodonta africana in Shungura H and the earliest warthog,
Pliocene-Pleistocene mammalian successions in Africa
Phacochoerus antiquus, in lower Bed II at Olduvai (Cooke and
The African mammal assemblages seem to have been more Maglio, 1972), and the last Notochoerus are found at Shungura
stable than those of other continents. In the somewhat and Koobi Fora.
discontinuous record of southern Africa, accelerated faunal
There are no known fossil-bearing beds within the 2.0-1.75change has been reported between 3 and 2.5 Ma approximately, Ma interval in southern Africa, but several first and last
again between 2 and 1.5 Ma, and finally at about 1 Ma (Turner, occurrences have been noted in the fauna from Swartkrans 1 (the
1990b). In East Africa, the more nearly continuous, ra- "hanging breccia" included) and in the Kromdraai B-East 3,
diometrically calibrated sequences, such as the Shungura Forma- roughly dated between 1.6 and 1.7 Ma (Partridge, 1982; Brain et
tion north of Lake Turkana, also display evidence of faunal al., 1988). Among the earliest appearances there are those of
change. Cooke (1978) recognized such changes within Member Oreotragus major, Redunca cf. R. arundinum, Antidorcas
C (prior to 2.6-2.7 Ma) and within Member G (approximately 2 australis, Rabaticeras arambourgi, Equus quagga, and PaMa). Those suggested changes do not coincide precisely with the ranthropus robustus, and the last occurrences of Chasmainterpretation of Maglio (1972) regarding the Koobi Fora porthetes silberbergi and Hipparion steytleri (Vrba, 1982; Klein,
Formation, northeast of Lake Turkana, where he proposed three 1984). The upper part of Olduvai Bed II, above Tuff IIA, is of
zones: Zone 3, bracketed between the Tulu Bor Tuff (3.35 Ma) similar age, 1.67-1.65 Ma, with the first occurrences of Elephas
and the base of the KBS Tuff (1.95 Ma); Zone 2, from that level recki stage III, Damaliscus niro, Pelorovis olduwayensis, and
up to the base of the Okote Tuff (1.64 Ma); and Zone 1, from several suid taxa, and the last records of Homo habilis,
that level to the top of the preserved section. No major changes Parmularius braini, Parmularius angusticornis, and Mammuthus
in bovid faunal composition were recognized by Cooke (1978) in africanavus in East Africa (Cooke, 1978).
the transition between Maglio's zones 2 and 1, around 1.6 Ma.
Further first and last stratigraphic data have been noted in
Near to that level, however, Maglio placed the evolutionary Kromdraai A, involving a number of carnivores (Turner, 1990a),
change between the Elephas recki varieties 2 and 3, which and in Sterkfontein Mb 5, with an indirectly estimated age of
coincides with the subdivision between lower Olduvai Bed II and about 1.5 Ma. At that time, a number of new suid taxa appeared
its middle-upper part.
in South African assemblages, including the Phacochoerini
Overall, the major compositional changes of the southern and Afrochoerus, Tapinochoerus, and Orthostonyx (Cooke and
east-central fossil faunas of Africa appear to have been related to Maglio, 1972). Members 2 and 3 of Swartkrans, nearing 1 Ma,
continent-wide turnovers, and the doubts expressed by Turner show minor changes (Vrba, 1982). In fact, the fossil record in
(1990b) about the possibility of correlation would seem to have Africa between the middle of the early Pleistocene (1.25 Ma)
120
Aguirre et al.
Erbaeva, and Pospelova, 1976). Local faunas of the Dodogol

horizon in Transbaikalia and Kizhikia have been correlated with
the Odessan faunal complex of eastern Europe by the presence
of Allactaga, Lagurodon, and Allophaiomys; Dodogol also
contains Eolagurus. The Razdolie l.f. of Siberia is correlated
with the Tamanian complex (Vangengeim, 1977; Erbaeva, 1986).
In Tadjikistan, the Lakhuti 1 small mammal fauna precedes the
Jaramillo, and the Lakhuti 2 l.f., with large mammals, shortly
Pliocene-Pleistocene mammals of Asia and Australasia
precedes the Matuyama-Brunhes reversal (Dodonov, 1986;
Low sea levels associated with glacial maxima at the end of the Sotnikova and Vislobokova, 1990).
Olduvai and again above the Jaramillo would have favored
Faunas in the Plio-Pleistocene Kopali Formation in Kazakhmigrations between southwestern Asia and southern Europe, stan can be correlated to European and other western Asian sites
between eastern Africa and Indo-Pakistan via southern Arabia, by virtue of cosmopolitan rodent taxa that occur jointly with
and between the mainland of eastern Asia and the half- endemic species (Tjutkova, 1991). The Kazakhstan assemblage
submerged peninsulas of Japan, the Philippines, Taiwan, and from Iliyskaya correlates with Kislang and Villany 3 and 5
Indonesia, to say nothing of the Beringia lowlands.
because of the presence of Mimomys newtoni and diverse species
In extreme southwestern Asia, the faunal association at Tell of Villanyia. The Tsharynskaya unit is probably younger than 1.7
'Ubeidiya in Israel includes a few taxa that also occur in the East Ma, considering the presence of Allophaiomys pliocaenicus
African Lower Pleistocene levels, such as Kolpochoerus oldu- together with diverse species of Allactaga, Ellobius, and
vaiensis, Hippopotamus gorgops, Pelorovis olduwayensis, and Allocricetulus, a combination that in Europe dates to about 1.6
Crocuta crocuta, and one, Equus cf. E. tabeti, in common with Ma. The units Jalanashkaya and Kopalinskaya are younger,
the North African site of Ai'n Hanech. The majority, however, between 1.4 Ma and 0.8 Ma.
have been identified with European species (Tchernov, 1988).
Marked ecological barriers between Europe and Asia, as well
Most of these species disappeared from Europe at about 0.8 Ma, as between diverse Asian regions, are reflected in many
but the 'Ubeidiya fauna includes a few that have been reported taxonomic differences among the ungulates at the genus level.
in Europe only at sites older than 1.6 Ma, plus several others that Dispersal of the carnivore guild (sensu Turner, 1990b) was not
did not appear in Europe until after 0.8 Ma. Tchernov (1988), impeded by ecosystem fragmentation to the same extent, and so
following the advice of F. C. Howell, referred to species of fossil carnivores provide an important basis for correlation
Lagurodon in estimating the correlation of 'Ubeidiya to an age throughout the whole Eurasian continent (Sotnikova, 1991). The
between 1.4 Ma and 0.8 Ma. Another criterion of this age is the local faunas of Dongcun in the Chifeng Beds (Lu, 1991) and
presence of Mammuthus (= Archidiskodon) meridionalis Nalaika in Mongolia (Sotnikova, 1991) include taxa in common
tamanensis, which is part of the diversified group of meridional with western Asian and (to a lesser extent) European localities of
mammoths found throughout Eurasia, from Britain to Japan, in younger Matuyama age. The local faunas of Dodogol and
the time span between the Jaramillo (1.1 Ma) and levels that date Zasuhino 2, in Transbaikalia, have a number of taxa in common
to slightly younger than 0.6 Ma. The combination of both with European localities considered to date to the post-Olduvai
criteria, taking into account the fact that the number of taxa interval, as well as several large mammals also known from the
shared in common with European sites younger than 1.1 Ma later part of the Nihewan faunal assemblage in China. A closer
greatly exceeds the number in common with sites dated to resemblance to the upper Nihewan sequence is found in the local
between 1.4 and 1.1 Ma, suggests that 'Ubeidiya has an age close fauna from the upper Zasuhino 3 horizon (Erbaeva, 1986;
to 1.0 Ma. The species in common with East African assem- Vangengeim, Erbaeva, and Sotnikova, 1990).
blages at 'Ubeidiya agree with the more restricted time span.
The upper Nihewan Xiaodukou beds, with a post-Hipparion
It is easy to correlate the mammal assemblages from the fauna and a considerable Paleolithic tool industry, is positioned
western regions of Asia with the calibrated European sequences just below the Jaramillo (Li and Wang, 1982), almost coincident
because of the high number of taxa common to both sides of the with the transition from Wucheng loess to Lishi loess, and is
Urals (Vangengeim, 1977). Considerable effort has been devoted dated to 1.11 Ma (Liu and Ding, 1983; Heller and Wang, 1991;
during the past two decades to fine-tuning the correlation of the Zhang, Chapter 26, this volume). The lower member of the
local faunas of eastern Europe and northern Asia to the Yuanmou Formation has a fauna generally characteristic of
paleomagnetic scale. The Betekia l.f. in western Siberia corre- middle Villafranchian age; paleomagnetic analyses indicate that
sponds to the Gauss chronozone (Gnibidenko and Pospelova, this unit extends up from the Gauss-Matuyama boundary to
1981), with faunal correlations to the Chikoi l.f. in Transbaikalia normal-polarity sediments, the correlation of which is uncertain
(Vangengeim, 1977). The Kuruksai l.f. in Tadjikistan has been because of compressed and possibly incomplete stratigraphy.
referred to the lower Matuyama, with two horizons straddling The upper Yuanmou member, or Shangnabang unit, is above the
the Reunion (Dodonov, 1986). Podpusk and Lebiazhie in normal-polarity strata and has a small fauna with Stegodon
western Siberia correspond to horizons with reverse polarity orientalis and Homo erectus (Zhang, Chapter 26, this volume).
immediately below the Olduvai normal subzone (Gnibidenko, According to Liu and Ding (1983) and Jiang, Sun, and Liang
and the middle Pleistocene (0.8 Ma) is relatively poor and
discontinuous. In North Africa, late Pliocene and early Pleistocene faunas are becoming better known, but still seem to be
quite discontinuous (Geraads, 1987); the Pliocene-Pleistocene
formations, which include the sites of Ai'n Boucherit and Ain
Hanech, deserve further study.
121
(1988), it belongs totally to the Brunhes, but it could reasonably

be placed in the latest Matuyama above the Jaramillo, between
1.0 and 0.8 Ma. Either interpretation of the paleomagnetic
record will be consistent with the faunal record of the upper
Yuanmou, which is definitely younger than the pre-Jaramillo
Xiaodukou faunas at Nihewan (Li, 1983b). Consequently, the
age of the Transbaikalian Zasuhino 2 l.f. may be approximately
1.8-1.6 Ma, whereas the Zasuhino 3 l.f. and the similar Dongcun
l.f., equivalent to the upper Nihewan Xiaodukou levels, are
closer to 1.2 Ma.
High rates of faunal change, including rapid spreading of
several taxa through Eurasia and their subsequent isolation and
diversification, were the rule during the youngest part of the
Matuyama, beginning just before 1.2 Ma and ending just after
0.8 Ma. Although Paleolithic tools indicate the presence of
humans during that interval, the oldest known remains of Homo
in China are specimens attributed to H. erectus from Gongwangling, Shaanxi province, almost at the top of the Matuyama, and
perhaps (depending on interpretation) also at Yuanmou. Although the Gongwangling l.f. shares many elements with that of
Zhokoudian and other mid-Pleistocene sites (Li, 1983b), its most
probable correlation, from the broad paleofaunal viewpoint, is
to the earliest mid-Pleistocene, prior to the Matuyama-Brunhes
reversal. The close similarities with the Jetis paleofauna, with its
reversed magnetic polarity (Zhou, Yanxian, and Wang, 1982),
support this interpretation.
presence of the defining subspecies, as well as several endemic

cervids, suggesting a period of insularity and high sea level in
Japan. Finally, the unit M. px.
(Mammuthus
meridionalis
proximus) is characterized by a progressive form of M.

meridionalis that appears in the Jaramillo and is replaced by M.
trogontherii in the early part of the Brunhes, in Japan as well as in
Europe.
Indo-Pakistan. There are diverse opinions as to the correct

dating for the transition from Tatrot to Pinjor mammal ages in
the Upper Siwalik sequence of northwestern India. The Tatrot
and Pinjor paleofaunas are broadly distinguished by the presence
of Hipparion and the absence of Equus in the former, and the
presence of Equus and the absence of Hipparion in the latter
(West, 1981). Several suid species that are found in Tatrot
assemblages do not occur in the Pinjor, which is marked by the
first regional appearances of Bubalus, Hemibos, and Bos
(Badam, 1979; Yokoyama, 1981). The earliest true elephants,
assigned to Elephas planifrons, are recognized already in the
Tatrot and consequently precede Equus in this area. Yokoyama
(1981) placed the base of the Pinjor within a normal-polarity
interval; he identified it as the Olduvai, but that is somewhat
questionable. Ranga Rao et al. (1981) noted that the evidence
supported two possible correlations, the other being to a level
within the Gauss closer to 3 Ma. In earlier works, a pre-Olduvai
Pinjor faunal horizon was noted by Dodonov, Pevzner, and
Penkova (1979), and the change from Tatrot to Pinjor assemJapan. Kamei and Otsuka (1981) established a succession of blages was dated to the uppermost Gauss, at about 2.5 Ma (now
mammal units for Japan, based on subspecies of proboscideans 2.7), by Opdyke et al. (1979) in Pakistan. The definition of the
and controlled by abundant paleomagnetic data from the basal Pinjor levels on the basis of the FAD of Equus is cited in all
continental strata, that affords correlations to other Asian time of those contradictory opinions.
scales. Kamei and Otsuka (1981) identified a major turnover in
Kotlia (1991) pointed to the arrival of Equus in Kashmir at 2.2
their oldest zone, called the S. s. (Stegodon sugiyamai) zone, Ma, significantly earlier than the FAD of this taxon in the
which dates from about 3 Ma to the base of the Olduvai Siwaliks, according to Yokoyama (1981). Kotlia noted that the
subchronozone. The turnover corresponds to the boundary first appearances of arvicolids in Kashmir, at 2.4 Ma, were
between the Daodi and Nihewan {sensu stricto) local faunas in followed by a period of local evolution in that group, until about
northern China and to the end of the Tatrot in India. In each 1.6 Ma. The arvicolid immigration in Kashmir coincided with the
instance these changes have been dated to the early Matuyama. appearances of Soriculini and Beremendini in the Indian
The strongly marked turnover between the early Nihewan faunas foothills.
and the typical levels, such as Xiashagou and Danangou (middle
Yushe III in earlier notation), characterized by the expansion of Java. In a thorough review of the mammal paleofaunas of Java
Indo-Malaysian elements and the appearance of western taxa and their stratigraphic position, Sondaar (1984) proposed the
such as Rusa, Eucladoceros, and Bison appears to be of faunal sequence Satir, Ci Saat, Trinil, and Kedung Brubus, in
immediate pre-Olduvai age in the mainland faunas (Zheng and ascending order, the last of which includes the Jetis l.f. (or
Cai, 1991); in Japan, that turnover has been correlated with the Djetis). Paleomagnetic studies in the area by Semah (1986)
median part of the S. a. (Stegodon akashiensis) zone, which resulted in somewhat different age estimations than Sondaar's.
contains the Olduvai subchron and a part of the later Matuyama A more consistent framework for the dating of formations and
with reverse polarity. J. A. Van Couvering (personal communica- fossil assemblages in central Java was established as a result of
tion) noted that the delay of this large-mammal turnover to the multidisciplinary work on the Kendeng Group in the Sangiran
end of the Olduvai in Japan may reflect the fact that migration to area by Watanabe and Kadar (1985), with further refinements by
those islands from mainland faunas is dependent on glacially Sudijono (1987) and Semah (Chapter 28, this volume).
lowered sea levels, such as at the beginning of the Pleistocene.
The earliest known Javanese fossil mammals appear to have
The S. a. unit is succeeded by the M. s. (Mammuthus lived under insular conditions; this is the Satir or upper
meridionalis shigensis) mammal unit, which extends to the base Kaliglagah fauna, from lacustrine beds in the lower part of the
of the Jaramillo subchron. This fauna is distinguished by the Sangiran Formation, just above the normal Olduvai subchrono-
Aguirre et al.
122
zone. The earliest Satir fauna is therefore assigned an age of

about 1.5 Ma. In the upper part of the "Black Clays" of the
upper Sangiran, the Ci Saat l.f. represents a faunal change, with
the presence of new immigrants. That change has been correlated with the Jaramillo, above a horizon dated at 1.16 Ma. The
base of the Kabuh Formation is formed by the "Grenzbank"
conglomerates containing the Trinil l.f., similar to that of Ci
Saat. Farther east, on the Solo River, the Bapang Formation
(equivalent to the Kabuh of Sangiran Dome) starts above the
Jaramillo and extends up to levels dated to 0.5 Ma within the
Brunhes. In the Bapang section, the Kedung Brubus and Jetis
local faunas come from the middle of the unit, in a sequence of
beds dated by a tektite fission-track age of 0.78 Ma and reverse
polarity to the very top of the Matuyama chron (Aimi and Aziz,
in Watanabe and Kadar, 1985, pp. 155-168).
Australia. Late Cenozoic fossil land vertebrates are known from
sites in all parts of Australia, and in New Guinea as well. One of
the earliest, the Awe l.f. of New Guinea, was assigned the early
Pliocene by Plane (1967). In a comprehensive review of
Australo-Papuan vertebrate paleontology, Rich et al. (1988)
dated the Awe to an age between 3 and 2.5 Ma. Other Pliocene
sites are the Chinchilla l.f. of Queensland, indirectly dated to an
age younger than 4 Ma, and the Smeaton l.f., from beds
overlying a basalt dated to 2.1 Ma. The Palankarinna l.f. of
South Australia is recognized as younger than the Awe, but older
than definitively Pleistocene assemblages, and the Morwell,
Kanunka, and Bone Gulch are other local faunas given a
transitional late Pliocene-early Pleistocene status. The local
fauna of Fisherman's Cliff is more certainly attributed to early
Pleistocene.
Changes in the late Pliocene and early Pleistocene

mammal faunas of the Americas
North America. The Plio-Pleistocene interval in North America

is documented in the fossil vertebrate faunas of the classic
Blancan and Irvingtonian mammal ages, knowledge of which has
been greatly advanced in the past few decades (Lindsay, Chapter
30, this volume). Thirteen distinct fossiliferous horizons have
been synchronized with the paleomagnetic scale in San Pedro
Valley, Arizona, between the Gilbert-Gauss boundary and the
late Matuyama below the Jaramillo (Johnson, Opdyke, and
Lindsay, 1975), resulting in an average interval between local
faunas of 150 k.y. About 20 different horizons have yielded
fossils of latest Blancan and Irvingtonian mammal ages in the
paleomagnetically calibrated Vallecito Creek section of AnzaBorrego, in southern California (Downs and White, 1960;
Lindsay et al., 1987), with an interval of less than 100 k.y. as an
average. In addition, numerous fossiliferous sections in the High
Plains of Kansas and Texas have now been correlated according
to new evidence from paleomagnetism and tephrochronology
(Lindsay, Johnson, and Opdyke, 1976; Lindsay, Chapter 30, this
volume). Churcher (1983) has published a similar review of

Canadian sites and their correlations with those to the south.
From those studies it is clear that the chronologic ranges of
individual taxa in different regions of the North American
continent are diachronous, due to paleoecological and paleoclimatic factors. Scholars therefore prefer speaking of lowest and
highest occurrences of taxa within local or regional stratigraphic
formations, expressed here by the terms FAD (first-appearance
datum) and LAD (last-appearance datum). Note that the use of
LSD (lowest stratigraphic datum) favored by Lindsay (Lindsay
et al., 1987; Lindsay, Chapter 30, this volume) is confusing,
because "L" is also used for "last" in other stratigraphic
abbreviations.
Dates for subdivision of the Blancan and for the BlancanIrvingtonian transition have been controversial. Webb (1984)
and other scholars have argued that the late Blancan should
begin just after the Sand Point l.f., equivalent to the Kaena
subchron at about 3.3 Ma. Others, however, would locate the
type Monte Blanco level, with reversed magnetic polarity of the
earliest Matuyama, in the early Blancan. A compromise has
been suggested by those who favor the base of the late Blancan
being traced just below the Monte Blanco bed, thus placing the
division closer to the Gauss-Matuyama reversal (Lundelius et
al., 1987), at 2.6 Ma. Criteria that have been proposed for the
beginning of Irvingtonian time include the immigration of
Mammuthus, the FAD of Ondatra and Smilodon, and the
replacement of archaeolagine rabbits (Hypolagus) by the modern leporines such as Sylvilagus and Lepus. The latter event is
evidenced almost simultaneously in the Curtis Ranch l.f. of San
Pedro Valley, just below the Olduvai at about 2.0 Ma, and in the
Borchers l.f. of Kansas immediately above the Pearlette B Ash,
dated to 1.9-2.0 Ma. The first Smilodon in Vallecito Creek falls
between 1.7 and 1.6 Ma, slightly above the Olduvai, and the first
Mammuthus appears to date to about the same time. In this
context, a strictly defined boundary would be difficult to
establish, but the more extensive small-mammal record suggests
a Blancan-Irvingtonian transition that would be older than, but
close to coincident with, the Pliocene-Pleistocene boundary at
the top of the Olduvai (Lindsay, Chapter 30, this volume).
Lundelius et al. (1987) revised the division of the Irvingtonian,
as earlier proposed by Schultz et al. (1978), into three parts: in
ascending order, the Sappan, Cudahayan, and Sheridanian. The
base of the Sheridanian is defined by the Pearlette O Ash, dated
to 0.61 Ma, and this unit therefore belongs totally to the middle
Pleistocene. The Cudahayan is considered to include the Rock
Creek l.f. and the Irvington l.f., the type of the Irvingtonian
mammal age, with reverse magnetic polarity, interpreted as a
part of the Matuyama above the Jaramillo. Also included in the
Cudahayan is the Courtland Canal l.f., which underlies the
Hartford Ash, dated to 0.74 Ma, and is thus close to the
Matuyama-Brunhes boundary. The Sappan starts with the
Borchers and Kentuck local faunas immediately overlying the
2.0-Ma Huckleberry Ridge or Pearlette B Ash in Kansas, while
the type Sappa l.f. of Nebraska directly underlies the Mesa Falls
or Pearlette S Ash, dated to 1.27 Ma. Thus, the Sappan-to-
Cudahayan transition would be placed between 1.2 and 1.0 Ma,

close to the warm-climate interval preceding the Jaramillo
subchron, and the base of the Sappan may be older than the
A number of significant faunal replacements that occurred
during the late Gauss are recorded in strata younger than 3 Ma:
for instance, the FADs of Glossotherium at several sites and of
Glyptotherium in Great Plains faunas, marking South American
connections, and the FADs of Tremarctos, Paramylodon, and
Tetrameryx in southwestern sites such as Anza-Borrego. Datum
events in faunas of early Matuyama age in the western USA are
the LADs of Hypolagus and Nannippus and the FADs of
Ondatra and Nothriotheriops between 2.4 and 2.2 Ma. The FAD
of Sylvilagus at 1.9 Ma is followed by the appearance of
Mammuthus meridionalis and Allophaiomys at the end of the
Olduvai, in the Wellsch Valley l.f. of Canada, and the appearance of Smilodon in Anza-Borrego. In this same time span, the
LADs of Hypolagus, Borophagus, and Pliophenacomys are seen
in the Wellsch Valley fauna, and the last Equus and Paradipodomys occur in Anza-Borrego. The Stegomastodon LAD in
the Great Plains and in the Gilliland local fauna of Texas occurs
near the end of the Sappan, while Cervus and Rangifer first
appear in the Cudahayan (Lundelius et al., 1987).
Repenning (1987) proposed somewhat different subdivisions
of the Blancan and the Irvingtonian, based on faunal changes
(immigrations, replacements) in microtine rodents, which he
correlated with parallel events in Europe (Lindsay, Chapter 30,
this volume, Figure 30.3). The Blancan IV unit, between events
5 (3.2 Ma) and 6 (2.6 Ma), displays stability in the endemic
populations of the western USA. The Blancan V, between events
6 and 7 (1.9 Ma), is marked by the immigration of Synaptomys
(s.s.) and Synaptomys (Mictomys), synchronously with the
appearances of Clethrionomys, Pliomys, and Lagurodon in
Europe. The Irvingtonian I unit, between events 7 and 8 (0.85
Ma), is characterized by the immigration of Microtus, Allophaiomys, Phenacomys, and Proneoflber to North America and
the contemporaneous appearances of Microtus, Allophaiomys,
Pity mys, and Dicrostonyx in Europe. Event 8 itself is marked by
the immigration of Clethrionomys and Pitymys into western
USA faunas, which Repenning correlated with the dispersal into
eastern Europe of the genera Eolagurus and Lagurus and with
the maximum southern extension of the range of Microtus
species.
South America. The general lithologic uniformities of the Upper
Cenozoic continental formations of South America, together
with the high degree of endemism in the successive vertebrate
paleofaunas and the latitudinal and geographic environmental
diversity, make it difficult to correlate the local biostratigraphic
divisions to the global scale. Even though the abundance and
diversity of fossil mammals in South America offer an outstanding opportunity for studies on biodiversity and evolution, their
distribution through time has been subject to diverse interpretations, so that many geologists lack confidence in their utility as
stratigraphic tools and prefer to group them in parastratigraphic
123
units. Tuning the South American mammal successions to the

magnetostratigraphic scale, as recommended by Pascual and
Fidalgo (1972), has met with partial success for the Ensenadan
sequence of Buenos Aires. Paleomagnetic studies of beds
attributed to the lower part of the pre-Ensenadan Uquia
Formation, from which sparse remains representing not more
than nine mammalian genera have been obtained, indicate an
age between 3.4 and 2.5 Ma (Orgeira, 1987); the upper part is
not fossiliferous. Considering the Uquian to be poorly documented and its typical fauna to be of Chapadamalalan age, Cione
and Tonni (1995) have proposed Marplatan as a new southern
South American continental stage for the interval prior to the
Ensenadan. The unit is based on abundant fossils representing
more than 120 genera collected from strata exposed in sea-cliffs
near Mar del Plata, some 1,800 km south of Buenos Aires, and
directly overlying the beds with the type fauna of the
Chapadamalalan. A major faunal turnover, involving a great
extinction of autochthonous South American taxa, attributed to
climate change rather than to invading taxa from North America,
is clearly evident at the top of the Marplatan. The small-mammal
evidence suggests a correlation of the Marplatan with the upper
part of the North American Blancan, and reversed paleomagnetic polarities in the upper part of the typical Marplatan, at
Barranca de los Lobos, most probably reflect the lower
Matuyama chronozone (Orgeira, 1990; Cione and Tonni, 1995).
The upper beds of the Uquia Formation may also correlate to
this level (Marshall et al., 1982; Orgeira, 1987). On that basis, it
can be suggested that the beginning of the Ensenadan mammal
age, with its well-recognized faunal turnover, should be dated to
the upper part of the lower Matuyama.
Determining the duration of the Ensenadan is complicated by
uncertainty as to definition of the Ensenadan-Lujanian boundary. This is rooted in the placement of the Arroyo Seco unit in
the Barranca de los Lobos sequence near Mar del Plata, in which
the Matuyama-Brunhes reversal has been observed. If the
whole Arroyo Seco is included into the Lujanian South
American Land Mammal Age, as suggested by Tonni et al.
(1992), then the conspicuous faunal overturn at the base of the
Lujanian must be located in the upper part of the Matuyama. On
the other hand, MacFadden et al. (1983), in a study of the
paleomagnetic profile of a composite section near Tarija,
Bolivia, and other related sequences with several fossiliferous
horizons, placed the base of the Ensenadan not in the early
Matuyama but instead just below the Jaramillo, close to 1.2 Ma,
and the Ensenadan-Lujanian transition within the Brunhes. The
latter correlation would be acceptable only if the Arroyo Seco
l.f. were retained in the upper Ensenadan.
Subdivision of the Ensenadan was long ago proposed by
Ameghino (fide Pascual and Fidalgo, 1972) on the basis of an
interbedded marine tongue near the middle of the type sequence. A more detailed analysis of differences between the
faunas collected in the upper and lower sections is desirable; in
modern studies, the Ensenadan usually is treated as a uniform
faunal unit (Tonni et al., 1992), without consideration of even
minor changes.
Aguirre et al.
124
The South American Ensenadan of the type area and the

North American Irvingtonian are similar in that both begin close
to (but below) the Olduvai subchron and in the fact that both
record high rates of exchange between North America and South
America. As Cione and Tonni have shown, most of the North
American invaders attributed to "Uquian" in earlier reports
were actually part of the great Ensenadan exchange that
followed some time after the mass extinction of autochthonous
forms that opened the age. With regard to that extinction event,
it should be noted that a pronounced austral cooling occurred
just above the Gauss-Matuyama reversal, at about 2.5 Ma
(Tonni et al., 1992), but that glacial episode appears to have been
too early to have been synchronized with the early Ensenadan
faunal crisis. No further step in cooling is recorded prior to a
paleoclimatic deterioration near the Jaramillo, at about 1 Ma,
which seems to have been too late. It is possible, therefore, to
imagine that the great basal-Ensenadan extinction event might
be close in age to the Pliocene-Pleistocene boundary, but the
evidence for that is far from conclusive.
Conclusion
The peaks of faunal overturn and the intervals of stability were
approximately synchronous in the mammal communities of
different continents during the Pliocene and early Pleistocene. It
can be inferred, not surprisingly, that the tempo of dispersal and/
or evolutionary change in mammalian communities was influenced by major changes in the global environment.
In relating faunal overturns to geological and climatic events,
one has to consider that the interdependence and development
of those phenomena were not uniform. Occasional global events
that lead to migration and dispersal may affect faunal stability
and produce major departures from patterns of "normal"
ecological dynamics. A "dispersal event" actually involves
several concepts. In the first case, it can mean the immigration of
one or more taxa into areas where they are pre-adapted for
success, as in the exchanges via the filter of Central America or
Beringia. In the second case, it may mean the expansion of
endemic lineages that have evolved adaptations to exploit the
resources of a larger region, such as the spread of Allophaiomys,
Microtus, Bison, Equus, and the progressive forms of Mammuthus meridionalis. The third case may arise from the effects of a
geographic expansion of favorable conditions. In the second
case, the effect is enhanced if it coincides with the development
of corridors or land bridges, but in each of the noted examples
the dispersal would have come about through the same favorable
adaptation: increased hypsodonty, with continued enamel production in the teeth-forming tissues. The success of such
adaptations depends on vegetation, and that is favored by
expansion of suitable plant associations, in this case the grasses
of the family Poaceae; and that, in turn, is also an effect of
climate change. Thus, there is an overlap with the third type of
dispersal. Other instances of the third case are the dispersal of
the carnivore "guild" (Turner, 1990b; Masini and Torre, 1990),
that of the new cervids, Cervalces and the Megacerini, and that
of Hippopotamus. All of those would have been predictable,
taking into account the lowered sea levels and dominantly mild
and moist climate up to the end of the Eburonian.
With respect to the stratigraphic questions, it is necessary to
recognize that all of the earlier described examples of faunal
overturns represent accelerated or powered processes that were
not instantaneous, but time-transgressive events, the same as
other geodynamic and paleogeographic changes. The type of
events that are closest to a geological "instant" are the magnetic
reversals. The ages and successions of land-mammal faunas are
now better calibrated and more precisely understood, so that
faunal changes, whether major or minor, global or regional, can
be tuned to the astronomically forced climate cycles, paleomagnetic reversals, paleogeographical and tectonic events, as well as
the global stratotype sections and points (GSSPs).
The basal boundary of the Pleistocene, as proposed by Aguirre
and Pasini (1985) and adopted in the Global Stratigraphic Scale
(Cowie and Bassett, 1989), is very close to the faunal overturns
represented in Allophaiomys event in Eurasia, to the dispersal of
the carnivore guild mentioned earlier, and to the changes in
African fauna seen in Olduvai Bed II. The faunal horizons closest
in time to these events are as follows: Olivola in northern Italy;
Barranca de los Conejos, close to Orce 2 in Spain; Villany 5 in
Hungary; Kadzielna in Ukraine; Shungura J7, the lowest part of
the Okote beds, and the earliest faunal levels in Olduvai Bed II in
East Africa; the Smilodon FAD in Vallecito Creek, southern
California; the top of the Uquia Formation, Argentina.
Acknowledgments
The authors are deeply indebted to many colleagues who have
shared their invaluable knowledge and advice. We wish to thank
in particular Dr. Ksenia V. Nikiforova and also the officers of the
INQUA and of the IUGS International Commission on
Stratigraphy, who have supported and encouraged our efforts
over the past two decades.
References
Aguirre, E. 1972. Utilization of proboscidians in Pleistocene
stratigraphy. In International colloquium on "The boundary between Neogene and Quaternary," collected papers,
vol. 3, ed. M. N. Alekseev et al., pp. 3-13. Moscow:
Nauka.
Aguirre, E. 1989a. Vertebrados del Pleistoceno continental. In
Map a del Cuaternario de Espana, escala 1:1.100.000, ed.
E. Llorente, pp. 47-68. Madrid: Instituto Tecnologico
Geominera de Espana.
Aguirre, E. 1989b. El limite inferior del Pleistoceno. In Mapa
del Cuaternario de Espana. Escala/scale 1:1.000.000, ed.
E. Llorente, pp. 87-94. Madrid: Instituto Tecnologico
Aguirre, E., and Morales, J. 1990. Villafranchian faunal record
of Spain. Quartdrpaldontologie (Berlin) 8:7-12.
125
Agusti, J. 1986. Synthese biostratigraphique du Plio-Pleistocene Cowie, J. W., and Bassett, M. G. 1989. International Union of
de Guadix-Baza (province de Granada, Sud-Est de
Geological Sciences 1989 stratigraphic chart. Episodes 12:
l'Espagne). Geobios 19:505-510.
Agusti, J., Moya-Sola, S., Martin-Suarez, E., and Marin, M. De Bruijn, H., Daams, R., Daxner-Hock, G., Fahlbusch, V.,
1987. Faunas de mamiferos en el Pleistoceno inferior de la
Ginsburg, L., Mein, P., and Morales, J. 1992. Report of
region de Orce (Granada, Espana). Paleontol. EvoL,
the RCMNS Working Group on fossil mammals, ReisenMem. especial 1:73-86.
burg 1990. Newsl. Strat. 26:65-118.
Alberdi, M. T., Alcala, L., Azanza, B., Cerdeno, E., Mazo, De Giuli, C , Ficarelli, G., Mazza, P., and Torre, D. 1983.
Confronto tra successioni marine e continetali del Pliocene
A. V., Morales, J., and Sese, C. 1989. Consideraciones
e Pleistocene inferiore in Italia e nell'area mediterranea.
biostratigraficas sobre la fauna de Vertebrados fosiles de la
Cuenca de Guadix-Baza (Granada, Espana). In Geologia
Soc. Paleontol. ItaLf Boll. 22:323-328.
y paleontologia de la Cuenca de Guadix-Baza, ed. M. T. De Giuli, C , and Masini, F. 1986. Late Villafranchian faunas in
Alberdi and F. P. Bonadonna, pp. 347-355. Trabajos sobre
Italy: the Casa Frata local fauna (Upper Valdarno,
Tuscany). Palaeontogr. ItaL 74:1-9.
Neogeno-Cuaternario, Mem. 11. Madrid: Consejo Superior de Investigaciones Cientificas.
De Giuli, C , Masini, F , and Torre, D. 1990. The latest
Alberdi, M. T., Arias, C , Bigazzi, G., Bonadonna, F. P., Leone,
Villafranchian faunas in Italy: the Pirro Nord local fauna
G., Lopez, N., Michaux, J., Morales, J., Robles, F , and
(Apricena, Gargano). Quartdrpaldontologie (Berlin) 8:2934.
Soria, D. 1982. Nuevo yacimiento de moluscos y vertebrados del Villafranquiense de la Cuenca del Jucar deLumley, H., Kahlke, H.-D., Moigne, A.-M., and Moulle, P.-E.
1988. Les faunes de grandes mammiferes de la Grotte du
(Albacete, Espana). In Colloque "Le Villafranchien
Vallonnet, Roquebrune-Cap. Martin, Alpes-Maritimes.
mediterraneen," Lille 9-10 Dec. 1982, Actes, vol. 1, pp.
UAnthropologie 92:465-496.
255-271. Paris: CNRS.
Aleksandrova, L. P. 1976. Rodents of Anthropogene of the Dodonov, A. E. 1986. Anthropogene of Southern Tadjikistan (in
Russian). Moscow: Nauka.
European part of the USSR (in Russian). Moscow: Nauka.
Azanza, B., and Sanchez, B. 1990. Les Cervides du Pleistocene Dodonov, A. E., Pevzner, M. A., and Penkova, A. M. 1979.
Several dates on geological structure and paleomagnetism
moyen d'Atapuerca (Burgos, Espagne). Quaternaire
of Late Pliocene-Quaternary deposits of northern and
(Lille) 1:197-212.
central India (in Russian). Komiss. Izuch. Chetvert.
Mammal succession of the Plio-Pleistocene in Italy. Soc.
PeriodaAkad. Nauk. SSR, Byul. 49:105-115.
Geol. ItaL, Mem. 31:213-218.
Downs, T., and White, J. A. 1960. A vertebrate faunal
Badam, G. L. 1979. Pleistocene fauna of India. Pune: Deccan
succession in superposed sediments from late Pliocere to
College, Postgraduate Research Institute.
middle Pleistocene in California. In XXIII International
Boeuf, O. 1990. Originalite et importance de la faune PlioGeological Congress, Proceedings, vol. 10, pp. 41-47.
Pleistocene de Chilhac (Haute-Loire, France). Quartdr- Dzaparidze, V., Bosinski, G., Bugianisvili, T., Gabunia, L.,
paldontologie (Berlin) 8:13-28.
Justus, A., Klopotovskaja, N., Kvavadze, E., LordkipaBolomay, A. 1965. Die Fauna zweiter villafrankischer Fundstelnidze, D., Majsuradze, G., Mgeladze, N., Nioradze, M.,
len in Rumanien. Geol. Ges. DDR, Bericht 10:77-99.
Pavlenisvili, E., Schmincke, H. U., Sologasvili, D. Z.,
Bonifay, E. 1991. Les premieres industries du Sud-Est de la
Tustfabramasvili, D., Tvalkrelidze, M., and Vekua, A.
France et du Massif Central. In Les premiers europeens,
1991. Der altpalaolitische Fundplatz Dmanisi in Georgien
ed. E. Bonifay and B. Vandermeersch, pp. 63-80. Actes
(Kaukasus). Romisch-Germ. Zentralmus. Mainz, Jahrb.
du 114e. Congres National des Societes Savantes, Paris,
36:67-116.
1989. Paris: Ed. CTHS.
Erbaeva, M. 1986. The Late Cenozoic faunistic complexes of
Brain, C. K., Churcher, C. S., Clark, J. D., Grine, F. E.,
Transbaikalia with special reference to micromammalia.
Shipman, P., Susman, R. L., Turner, A., and Watson, V.
Quartdrpaldontologie (Berlin) 6:25-28.
1988. New evidence of early hominids, their culture and Esteban Aenlle, J., and Lopez Martinez, N. 1987. Les
environment from the Swartkrans cave, South Africa. S.
Arvicolides (Rodentia, Mammalia) du Villanyen recent de
Afr. J. Sci. 84:828-835.
Casablanca I (Castellon, Espagne). Geobios 20:591-623.
Chaline, J. 1986. Continental faunal units of the Plio-Pleistocene Fahlbusch, V. 1991. The meaning of MN-zonation: considerof France. Soc. Geol. ItaL, Mem. 31:175-183.
ations for a subdivision of the European continental
Churcher, C. S. 1983. Faunal correlations of Pleistocene deposits
Tertiary using mammals. Newsl. Strat. 24:159-173.
in Western Canada. In Correlation of Quaternary chronolo- Fejfar, O., and Heinrich, W.-D. 1981. Zur biostratigraphischen
gies, ed. W. C. Mahaney, pp. 145-148. Toronto: York
Untergliederung des kontinentalen Quartars in Europa
University.
anhang von Arvicoliden (Rodentia, Mammalia). Eclogae
Cione, A. L., and Tonni, E. P. 1995. Chronstratigraphy and
Geol. Helvet. 74:997-1006.
"Land-Mammal ages" in the Cenozoic of southern South Fejfar, O., Heinrich, W.-D., and Heintz, E. 1990. Neues aus den
America: principles, practices, and the "Uquian" probVillafranchium von Hajnacka bei Filakovo (Slovakei,
lem. /. Paleontol. 69:135-159.
CSSR). Quartdrpaldontologie (Berlin) 8:47-70.
Cooke, H. B. S. 1978. Faunal evidence for the biotic setting of Ferrandez, C., Perez Cuadrado, J. L., Gibert, J., and Martinez,
Early African hominids. In Early Hominids of Africa, ed.
B. 1989. Estudio preliminar de los sedimentos de relleno
C. Jolly, pp. 267-281. London: Duckworth.
carstico de Cueva Victoria (Cartagena, Murcia). In Los
Cooke, H. B. S., and Maglio, V. J. 1972. Plio-Pleistocene
restos humanos de Orce y Cueva Victoria, ed. J. Gibertstratigraphy in East Africa in relation to proboscidean and
Clols et al., pp. 379-394. Barcelona: Institut Paleontologic
suid evolution. In Calibration of Hominoid Evolution, ed.
Dr. M. Crusafont.
W. W. Bishop and J. A. Miller, pp. 303-329. New York: Gabunia, L. K., and Vekua, A. K. 1981. Terrestrial mammals
Wenner-Gren Foundation.
of the Pliocene and the Early Pleistocene and boundary
126
Aguirre et al.
between Neogene and Quaternary in Georgia, USSR. In

Pliocene dispersal of the horse Equus and Late Cenozoic
Proceedings, Field Conference Neogene-Quaternary Bounmammalian dispersal event. Nature 287:135-138.
dary, India 1979, ed. M. V. A. Sastry et al., pp. 45-48. Lindsay, E. H., Opdyke, N. D., Johnson, N. M., and Butler,
Calcutta: Geological Survey of India.
R. F. 1987. Mammalian chronology and the magnetic
Geraads, D. 1987. Dating the Northern African cercopithecid
polarity time scale. In Cenozoic mammals of North
fossil record. J. Hum. Evol. 2:19-27.
America, ed. M. O. Woodburne, pp. 269-284. Berkeley:
Ghenea, K. 1981. Remarks on the Pliocene/Lower Pleistocene
University of California Press.
Stratigraphy in Romania. In Proceedings, Field Confer- Liu, T., and Ding, M. 1983. Discussion on the age of "Yuanmou
ence Neogene-Quaternary Boundary, India 1979, ed.
Man." Ada Anthropol. Sin. 2:40-48.
M. V. A. Sastry et al., pp. 49-54. Calcutta: Geological Lopez, N., Michaux, J., and Villalta, J. F. 1976. Rongeurs et
Survey of India.
lagomorphes de Bagur-2 (Province de Girone, Espagne).
Gil, E. 1987. Taxonomia y bioestratigrafia de micromamiferos
Nouveau remplissage de fissure du debut du Pleistocene
del Pleistoceno Medio, especialmente roedores, de los
Moyen. A eta Geol. Hisp. (Barcelona) 11:46-54.
rellenos karsticos de la Trinchera del ferrocarril de Lu, Y. 1991. The Pleistocene stratigraphy of Chifeng region,
Atapuerca (Burgos). Unpublished doctoral thesis, UniverInner Mongolia. In XIII INQUA Congress, Beijing, 1991.
sidad Zaragoza, Faculdad de Ciencias.
Abstracts, p. 221. Beijing: Science Press.
Gnibidenko, Z. N., Erbaeva, M. A., and Pospelova, G. A. 1976. Lundelius, E. L., Churcher, C. S., Downs, T., Harrington,
Paleomagnetism and biostratigraphy of upper Cenozoic
C. R., Lindsay, E. H., Schultz, G. E., Semken, H. A.,
deposits in western Transbaikalia (in Russian). In
Webb, S. D., and Zakrzewski, R. J. 1987. The North
American Quaternary sequence. In Cenozoic mammals of
Paleomagnetism of Mesozoic and Cenozoic of Siberia and
North America, ed. M. O. Woodburne, pp. 211-235.
the Far East (in Russian), ed. E. E. Fatiadi, pp. 75-95.
Berkeley: University of California Press.
Novosibirsk: Institute of Geology and Geophysics, Siberian Branch of USSR Academy of Sciences.
MacFadden, B. J., Siles, O., Zeitler, P., Johnson, N. M., and
Campbell, K. E. 1983. Magnetic polarity stratigraphy of
Gnibidenko, Z. N., and Pospelova, G. A. 1981. Paleomagnetism
the Middle Pleistocene (Ensenadan) Tarija Formation of
of Betekeian sedimentary formation in North Kazakhstan
Southern Bolivia. Quat. Res. 19:172-187.
(in Russian). Geol. Geophiz. 9:111-119.
Heller, F., and Wang Junda. 1991. Magnetism of Quaternary Maglio, V J. 1972. Vertebrate faunas and chronology of hominid
bearing sediments east of Lake Rudolf, Kenya. Nature
sediments: loess in China. In Special Proceedings: Review
Reports. XIII INQUA Congress, Beijing, 1991, pp. 88-97.
250:19-23.
Beijing: Science Press.
Markova, A. K. 1990. The sequence of Early Pleistocene smallmammal fauna from the South Russian Plain. QuartarpaldJiang N., Sun R., and Liang Q. 1988. Strates cenozoiques
tardives et fossiles humains dans le basin de Yuanmou,
ontologie (Berlin) 8:131-152.
Yunnan, Chine. UAnthropologie 92:939-944.
Marshall, L. G., Berta, A., Hoffstetter, R., Pascual, R., Reig,
Johnson, N. M., Opdyke, N. D., and Lindsay, E. H. 1975.
O., Bombin, M., and Mones, A. 1984. Mammals and
Magnetic polarity stratigraphy of Pliocene-Pleistocene
stratigraphy: geochronology of the continental mammalterrestrial deposits and Vertebrate faunas, San Pedro
bearing Quaternary of South America. Paleovertebrata,
Valley, Arizona. Geol Soc. Am., Bull. 86:5-12.
Mem. Extr. pp. 1-76.
Kamei, T., and Otsuka, H. 1981. The Plio-Pleistocene Marshall, L. G., Butler, R. F , Drake, R. E., and Curtis, G. H.
1982. Geochronology of type Uquian (Late Cenozoic)
stratigraphy of Japan in relation to proboscidean evolution.
land mammal age, Argentina. Science 216:986-989.
In Proceedings, Field Conference Neogene-Quaternary
Boundary, India 1979, ed. M. V. A. Sastry et al., pp. 83-88. Masini, F , and Torre, D. 1990. Large mammal dispersal events
at the beginning of the Late Villafranchian. In European
Calcutta: Geological Survey of India.
Neogene mammal chronology, ed. E. H. Lindsay et al.,
Klein, R. G. 1984. The large mammals of southern Africa: Late
pp. 131-138. New York: Plenum Press.
Pliocene to Recent. In Southern African prehistory and
paleoenvironments, ed. R. G. Klein, pp. 107-146. Rotter- McFadden, P. L., Brock, A., and Partridge, T. C. 1979.
dam: Balkema.
Palaeomagnetism and the age of the Makapansgat
Konstantinova, N. A. 1967. Quaternary of Southern Moldavia
hominid site. Earth Planet. Sci. Lett. 44:373-388.
and Southwestern Ukrania (in Russian). Geol. Inst. Acad. Moya-Sola, S., and Menendez, E. 1986. Los Artiodactilos
Nauk USSR, Trudy 173:1-137.
(Bovidae y Cervidae, Mammalia) del Pleistoceno inferior
Kotlia, B. S. 1991. Neogene-Quaternary rodent biochronology
de Europa Occidental: ensayo de sintesis. Paleontol. Evol.
of Kashmir Basin, India. In XIII INQUA Congress,
(Sabadell) 20:289-295.
Beijing, 1991, Abstracts, p. 175. Beijing: Science Press.
Nabel, P. E., and Valencio, D. A. 1981. La magnetoestratigrafia
del Ensenadense de la ciudad de Buenos Aires: su
Li Huamei and Wang Junda. 1982. Magnetostratigraphic study
significado geologico. Asoc. Geol. Argentina Rev. 36:7of several typical geologic sections in North China. In
Quaternary geology and environment of China, pp. 33-38.
18.
Nadachowsky, A. 1990. Lower Pleistocene rodents of Poland:
Beijing: Ocean Press.
Faunal succession and biostratigraphy. QuartdrpaldontoLi Yi. 1983a. Le paleolithique inferieur en Chine du Nord.
logie (Berlin) 8:215-224.
L Anthropologie 87:185-189.
Li Yi. 1983b. A propos de l'age geologique de l'Homme de Nikiforova, K. V, and Aleksandrova, L. P. 1987. Stratigraphic
scheme of Late Pliocene and Eopleistocene sediments of
Yuanmou. L Anthropologie 87:271-272.
the European part of the USSR (in Russian). In The
Lindsay, E. H., Johnson, N. M., and Opdyke, N. D. 1976.
boundary between Neogene and Quaternary systems of the
Preliminary correlation of North American land Mammal
USSR (in Russian), ed. M. N. Alekseev and K. V.
Ages and geomagnetic chronology. Univ. Mich. Pap.
Nikiforova, pp. 8-13. Moscow: Nauka.
Paleontol. 12:111-119.
Lindsay, E. H., Opdyke, N. D., and Johnson, N. M. 1980. Opdyke, N. D., Lindsay, E., Johnson, G. D., Johnson, N.,
127
Tahirkheli, R. A. K., and Mirza, M. A. 1979. Magnetic Sotnikova, M. V., and Vislobokova, I. A. 1990. Pleistocene
mammals from Lakhuti, Southern Tajikistan, U.S.S.R.
polarity stratigraphy and vertebrate paleontology of the
Quartdrpaldontologie (Berlin) 8:237-244.
Upper Siwalik subgroup of Northern Pakistan. Palaeogeogr. Palaeoclimatol. Palaeoecol. 27:1-34.
Soto, E., and Morales, J. 1985. Grandes mamiferos del
Orgeira, M. J. 1987. Estudio paleomagnetico de sedimentos del
yacimiento villafranquiense de Casablanca 1, Almenara
Cenozoico tardio en la costa atlantica bonaerense. Asoc.
(Castellon). Estud. geol. 41:243-249.
Steensma, K. J. 1988. Plio/Pleistozane Grosssaugetiere (MamGeol. Argentina Rev. 42:362-376.
malia) aus dem Becken von Kastoria/Grevena, Sudlich
Orgeira, M. J. 1990. Paleomagnetism of Late Cenozoic fosvon Neapolis-NW Griechenland. Unpublished dissertasiliferous sediments from Barranca de los Lobos (Buenos
tion, Technischen Universitat Clausthal.
Aires province, Argentina). The magnetic age of the
South American land-mammal ages. Phys. Earth Planet. Sudijono. 1987. Status of the mammalian biostratigraphy of Java
with special reference to the Sangiran area. In Proceedings
Int. 64:121-132.
of the First International Colloquium on Quaternary
Partridge, T. C. 1982. The chronological positions of the fossil
Stratigraphy of Asia and Pacific Area. Osaka, 1986, ed. M.
hominids of Southern Africa. In ler Congres International
Itihara and T. Kamei, pp. 270-289. Osaka: INQUA
de Paleontologie Humaine, Nice 1982, Pretirage, vol. 2, ed.
Science Council of Japan.
H. deLumley and M. A. deLumley, pp. 617-675. Nice:
Tchernov, E. 1988. La biochronologie du site d'Ubeidiya (vallee
Centre National de Recherche Scientifique.
du Jourdain) et les plus anciens hominides du Levant.
Pascual, R., and Fidalgo, F. 1972. The problem of the PlioL'Anthropologie 92:839-861.
Pleistocene Boundary in Argentina (South America). In
International Colloquium "The Boundary between Neo- Tjutkova, L. A. 1991. Late Neogene and Early Quaternary
rodents from the Southeastern Kazakhstan (USSR). XIII
gene and Quaternary," Collection of Papers, vol. 2, ed.
M. N. Alekseev et al., pp. 205-262. Moscow: Nauka.
Congress INQUA, 1991, Beijing. Abstracts, p. 357.
Plane, M. D. 1967. Stratigraphy and vertebrate fauna of the
Otibanda Formation, New Guinea. Austral. Bur. Min. Tonni, E. P., Alberdi, M. T , Prado, J. L., Bargo, M. S., and
Cione, A. L. 1992. Changes of mammal assemblages in the
Resources, Geol. Geoph. Bull. 86:50-64.
Pampean region (Argentina) and their relation with the
Rabeder, G. 1981. Die Arvicoliden (Rodentia) aus dem Pliozan
Plio-Pleistocene boundary. Palaeogeogr. Palaeoclimatol.
und dem alteren Pleistozan von Niederosterreich. Beitr.
Palaeoecol. 95:179-194.
Pal. Osterr. (Wien) 8:1-343.
Radulesco, C , and Samson, P. 1990. The Plio-Pleistocene Tretyak, A. N., and Volok, Z. E. 1974. Paleomagnetic essay of
Pliocene-Quaternary-stratified sediments of USSR terrimammalian succession of the Oltet Valley, Dacic Basin,
tory (in Russian). Geophys. Coll. (Kiev Naukova Dumka)
Romania. Quartdrpaldontologie (Berlin) 8:225-232.
58:50-55.
Ranga Rao, A., Khan, K. N., Venkatachala, B. S., and Sastri,
V. V. 1981. Neogene/Quaternary Boundary and the Turner, A. 1990a. Late Neogene/Lower Pleistocene Felidae of
Africa: evolution and dispersal. Quartdrpaldontologie (BerSiwaliks. In Proceedings. Field Conference Neogenelin) 8:247-256.
Quaternary Boundary, India, 1979, ed. M. V. A. Sastry et
al., pp. 131-142. Calcutta: Geological Survey of India.
Turner, A. 1990b. The evolution of the guild of larger terrestrial
Repenning, C. A. 1987. Biochronology of the microtine rodents
carnivores during the Plio-Pleistocene in Africa. Geobios
of the United States. In Cenozoic mammals of North
23:349-368.
America, ed. M. O. Woodburne, pp. 236-268. Berkeley: Van der Meulen, A. J. 1973. Middle Pleistocene smaller
University of California Press.
mammals from the Monte Peglia (Orvieto, Italy) with
Repenning, C. A., and Brouwers, E. 1992. Late Pliocene-Early
special reference to the phylogeny of Microtus (ArviPleistocene ecologic changes in the Arctic Ocean bordercolidae, Rodentia). Quaternaria (Roma) 17:1-144.
Van der Meulen, A. J., and Zagwijn, W. H. 1974. Microtus
land. U.S. Geological Survey, Bulletin 2036.
(Allophaiomys) pliocaenicus from the lower Pleistocene
Repenning, C. A., Fejfar, O., and Heinrich, W.-D. 1990.
near Brielle, the Netherland. Scripta Geol. 21:1-12.
Arvicolid rodent biochronology of the Northern Hemisphere. In International symposium: evolution, phylogeny Vangengeim, E. A. 1977. Paleontologie foundation of the
and biostratigraphy of arvicolids (Rodentia, Mammalia),
Anthropogene stratigraphy of Northern Asia (on mammals)
ed. O. Fejfar and W.-D. Heinrich, pp. 385-415. Prague:
(in Russian). Moscow: Nauka.
Geological Survey.
Vangengeim, E. A., Erbaeva, M. A., and Sotnikova, M. V.
Rich, T., Archer, M., Plane, M., Flannery, T., Pledge, N., Hand,
1990. Pleistocene Mammals from Zasuhino, Western
S., and Rich, P. V. 1988. Australian Tertiary mammal
Transbaikalia. Quartdrpaldontologie (Berlin) 8:257-264.
localities. In The fossil vertebrate record of Australasia, ed. Van Kolfschoten, T. 1990. Review of the Pleistocene arvicolid
faunas from the Netherlands. In International symposium:
P. V. Rich and R. M. Thompson, pp. 527-570. Clayton:
Monash University Publications Committee.
evolution, phylogeny and biostratigraphy of arvicolids
(Rodentia, Mammalia), ed. O. Fejfar and W.-D. Heinrich,
Schultz, C. B., Martin, L. D., Tanner, L. G., and Corner, R. G.
1978. Provincial land mammal ages for the North Ameripp. 255-274. Prague: Geological Survey.
can Quaternary. Nebraska Acad. Sci., Transact. 5:59-64. Vekua, A. 1987. The lower Pleistocene mammalian fauna of
Akhalkalaki (southern Georgia, USSR). Palaeontogr. Ital.
Semah, F. 1986. Le peuplement ancien de Java. Ebauche d'un
74:63-96.
cadre chronologique. UAnthropologie 90:359-400.
Sondaar, P. 1984. Faunal evolution and the mammalian Virina, E. I., Dobrodeev, O. P., and Faustov, S. S. 1971.
Paleomagnetism and correlation of new sediments of the
biostratigraphy of Java. Forschungsinst. Senckenb., Cour.
southern Russian Plains (in Russian). In Problems of
69:219-235.
Sotnikova, M. V. 1991. Early Quaternary carnivora of Asia. XIII
Pleistocene subdivision (in Russian), pp. 45-56. Moscow:
Congress INQUA, Beijing, 1991. Abstracts, p. 337.
Nauka.
Vrba, E. S. 1982. Biostratigraphy and chronology, based
128
Aguirre et al.
particularly on bovidae, of southern African hominidassociated assemblages. In ler Congres International de

Paleontologie Humaine, Nice 1982, Pretirage, vol. 2, ed.
Watanabe, N., and Kadar, D. (eds.) 1985. Quaternary Geology
of the hominid fossil bearing formations in Java. Special
Publication no. 4. Bandung: Geological Research and
Development Centre.
Webb, D. 1984. Ten million years of mammal extinctions in
North America. In A Prehistoric Revolution, ed. P. S.
Martin and R. G. Klein, pp. 189-210. Tucson: Univ.
Arizona Press.
West, R. M. 1981. Plio-Pleistocene fossil vertebrates and
biostratigraphy, Bhittanni and Marwat ranges, North-West
Pakistan. In Proceedings, Field Conference NeogeneQuaternary Boundary, India, 1979, ed. M. V. A. Sastry et
al., pp. 211-215. Calcutta: Geological Survey of India.
Yokoyama, T. 1981. Palaeomagnetic study of Tatrot and Pinjor

Formations, Upper Siwaliks east of Chandigarh, NorthWest India. In Proceedings, Field Conference NeogeneQuaternary Boundary, India, 1979, ed. M. V. A. Sastry
et al., pp. 217-220. Calcutta: Geological Survey of
India.
Zazhigin, B. S. 1980. Late Pliocene and Anthropogene rodents of
the South of Western Siberia (in Russian). Moscow: Nauka.
Zheng Shaohua and Cai Baoqua. 1991. Fossil micromammals
from the Dongguo section of Dongyaozitu, Yuxian county,
Hebei Province (in Chinese, with English summary). In
Contributions to INQUA XIII, Beijing, 1991, pp. 100-131.
Zhou, M., Yanxian, L., and Wang, L. 1982. Chronology of the
Chinese fossil hominids. In ler Congres International de
Paleontologie Humaine, Nice 1982, Pretirage, vol. 2, ed.
10
Human evolution in the Plio-Pleistocene interval

EMILIANO AGUIRRE
Introduction
For many years, the main problem in relating the PliocenePleistocene boundary to human evolution has been the controversy over "Tertiary man." For some, this has had serious
philosophical implications arising from the definition of the
Quaternary period in anthropological terms. The development
of modern stratigraphical standards, however, makes this a
merely nominal question, because such boundaries are now
based entirely on geological principles. At the same time, the
terms "Tertiary" and "Quaternary," having been drawn from
primitive concepts of stratigraphy that once also distinguished a
"Primary" and a "Secondary," have been called into question.
Nevertheless, the chronology of early humans is pertinent to the
goal of this project, for other than merely historical reasons.
The search for fossil Hominidae in the past three decades has
produced an immense amount of geological and paleontological
evidence. In Africa, particularly, well-exposed and relatively
thick Plio-Pleistocene sections yielding abundant remains of
Hominidae and a wide diversity of other fossils have been very
well characterized in terms of biochronology, and also with
reference to magnetostratigraphy and isotopic dates. Similar
data, if not so well developed, have been recovered from other
areas as a direct result of paleoanthropological studies.
It is now clear that key events in human evolution occurred
within the time interval that also encompassed the PliocenePleistocene boundary (i.e., within the limits of the Matuyama
paleomagnetic chron, from 2.6 to 0.8 Ma) (Figure 10.1). It was
during that time that the genus Homo first appeared in Africa,
giving rise to the key species Homo habilis Leakey, Tobias, and
Napier, at about 1.8 Ma, almost exactly coincident with the
beginning of the Pleistocene. At some later time, perhaps very
soon after its evolution in Africa at about 1.6 Ma, the clade of
Homo ergasterlerectus began to disperse into southern Eurasia.
Africa
Intensive research over the past few decades has uncovered a

nearly continuous record of hominid evolution in Africa during
the period in question that conclusively demonstrates the origin
of our genus in the mammalian fauna of that continent. No less

than four separate areas have yielded numerous fossil hominids
in stratified context, each with abundant associations of tools and
animals, and three of which are also well suited for geochronologic analysis.
Turkana Basin
The sediments deposited in the margins and tributary valleys of

the Lake Turkana depression in northernmost Kenya and
southern Ethiopia contain an incredibly rich fossil record
covering middle Pliocene to middle Pleistocene time, as well as
earlier (i.e., Lothagam, Kalodirr) and later (Kibish, Galana Boi,
Guomde) sites of interest to human evolution. Three major
collecting areas in this basin are geographically separate but
stratigraphically correlative, as demonstrated by identification of
numerous tuff horizons that extend throughout the basin (Feibel,
Brown, and McDougall, 1989). White (1988) has published a
comprehensive inventory of the hominid remains from the
Turkana Basin, identified to generic level, within the time span
between 2.6 Ma and 1.4 Ma.
Shungura Formation. In the lower Omo valley, north of Lake
Turkana in Ethiopia, a nearly continuous and outstandingly
fossiliferous section, with many radiometric, paleomagnetic, and
geochemical calibration controls, has been mapped by teams led
by F. Clark Howell (Coppens et al., 1976). The Shungura section
is divided into members, lettered A to L, each defined by a
volcanic tuff at its base and including many other tuffs within its
thickness. The age determinations for the Shungura tuffs are
analytically reliable and stratigraphically consistent. The most
pronounced faunal change is observed in the interval between
tuffs C and G (Aguirre et al., Chapter 9, this volume), dated to
approximately 2.85 Ma and 2.33 0.03 Ma, respectively.
Supporting dates have been obtained on tuffs D and F within the
interval, dated 2.52 0.11 and 2.36 0.05 Ma (Brown et al.,
1985a,b; Feibel etal., 1989).
Shungura Members C, D, E, and F are highly fossiliferous,
each with more than 10 localities that have yielded fossil
hominids. Of hominids, the fossil material includes isolated teeth
129
C-E.ETHIOPIA
W-TURKANA E-TURKANA
(CHAVAILLON
(BROWN e. a. (BROWN e. a. 0M0SHUNGURA S2)
85)
'85)
TANZANIA
(LEAKEY,
S-AFRICA
Ma (VRBA, 1982) ISAAC '82)
Mauer
0.5
0.6
Ternifine
0.7
0.8
CO
<
CQ
11
CC
1.3
O Tuff 110 Z
Tuff IIC
Chari Tuff
<&*
<
>
WT-15000
Malbe Tuff
KBS Tuff
2.3
Kalochoro Mb
Y
s
C^ ^/^ Iffy T/N ^^ ^^\ 1 ^X
2.5 oiGrktontGin
|
Lokalalei Mb BurgiTuff
WT-17000
w
GTuff
0 FTuff 5
*
ETuff
. D Tuff
* BK-T2Tuff
2.7
9 ft
.omekwi Mb
2.9
Makapan4 A
3
3.1
Makapan3 A
Kada Hadar Mb
Hasuna Tuff CTuff
AL-288
v
Denen Dora Mb
BIOTuff AL-333
T
AL-266
T
AL-128
T
\
SH Cinerite
3.2
3.3
#Toroto
# TuluBorTuff
Sidi-Hakoma
3.4
3.5
DKhabilis
Homosp.
Paranthropus boise
O P. robustus
P.aethiopicus
0 Paranthropus sp.
A Australopithecus sp.
A A. africanus
A.afarensis
vHominidaegeasp.
^ Stone tools
H2 Tuff
'
Ogol Lavas
Dmanisi
Homo erectus
2.2
o c
IB
2.4
M
y
H
Gombore
; Okote tuffs KTuff

J7 Tuff
Notao Mb
Tuff IIA
Tuff ID
Tuff IB
LU
Member
i l l ^^ i l l ^^^y 1
2 1
Cueva Victoria??
< GarbalV
LTuff
Mb
Tuff II B
1.8
1.9
"
"
"
1
1
Nariokotome
1.5 Swartkrans *
O
S2.6
Trinil
S4.S5
S1
Ubeidiya
-z.
Unnamed Tuf
DC
Member III
LJLJ
1.6 1
Kromdraai
1.7 B3
A tuff top
Z)
1.2
CTuff
c
Gombore IIC o
BTuff
<
LU
Swartkrans
OH
1.1 2
1.4 Sterkfontein
Gongwangling
z S12.17.3
Member IV
Swartkrans
0.9
Magnetic
scale
EURASIA
INDONESIA
Top, upper
Laetolil Beds y
Figure 10.1. Principal occurrences of human fossils and paleolithic artifacts in the Old World during the late Pliocene and early Pleistocene.
The paleomagnetic scale on the right shows the Matuyama chron be-
Tuff
Mb
11
Hi
tween 0.7 and 2.5 Ma, with the Olduvai subchron normal-polarity interval at about 1.8 Ma.
Human evolution in the Plio-Pleistocene
131
range that should be considered as an independent species,

Homo rudolfensis Alexeev, 1986. Other crania and mandibles
from just below the KBS Tuff (i.e., KNM-ER 1813, 3735), as
well as many others from the KBS Member above the tuff, were
retained by Wood in H. habilis. The alternative view can also be
supported, that is, that the differences between the nearly coeval
1470 rudolfensis and 1813 habilis crania are sufficiently explained
by sexual dimorphism, whereas differences seen in the other
specimens from near the KBS Tuff can be ascribed, quite
reasonably, to intraspecific variation.
In this regard, a new mandible from the Chiwondo Beds of
Malawi, UR-501, is said to be closely comparable to the
specimen KNM-ER "1483" (yel 1482) of the lower Koobi Fora
(T. Bromage, personal communication, 1992). The Malawi
specimen is also of roughly the same age as "1483," to judge by
faunal context, but whether these older specimens are correctly
Homo, and not Paranthropus or even a directly ancestral
Australopithecus, will be difficult to determine without a better
definition of the cladogenetic event. In any event, many
characteristics in these specimens are at the extremes of their
ranges of variation for H. habilis.
In the upper Koobi Fora, Paranthropus boisei is recognized in
many fossils from the KBS and Okote members up to but not
above the Chari Tuff, at 1.4 Ma. As to human remains, a fossil
skull, KNM-ER 3733, from the upper KBS Member just below
the Okote Tuff complex, is the earliest specimen to be identified
to Homo erectus; its stratigraphic position is dated slightly older
East Turkana. Radiometrically dated tuffs in the Koobi Fora and than 1.6 Ma. Within the Okote Member, coeval with the
Guomde formations east of Lake Turkana also cover the time youngest P. boisei, we find a cranium (KNM-ER 3883) and
interval we are concerned with. Several depositional hiatuses are mandibles (KNM-ER 992, 730) of other H. erectus.
recognized in the Koobi Fora sequence: an unconformity in the
The East Turkana sequence has been correlated on the basis of
strata between the Toroto Tuff (3.32 0.02 Ma) and the Burgi physical stratigraphy of airfall tuffs, as well as by radiometric
Tuff (2.6 Ma), and a larger hiatus representing most of the time dates, paleomagnetic reversals, and paleontology, with the
between the Burgi Tuff and the KBS Tuff (1.88 Ma) (Brown et Shungura Formation north of Lake Turkana, with West Turkana,
al., 1985b; Feibel et al., 1989). The upper part of the Koobi Fora with the Awash region in the Ethiopian Rift Valley, and with
sequence, above the KBS Tuff, is calibrated on the basis of the Olduvai-Laetoli in Tanzania (Cooke, Chapter 27, this volume).
Okote complex of tuffs (1.6-1.5 Ma) and the Chari Tuff (1.39
0.01 Ma), another tuff dated at 1.25 0.02 Ma, and at the top West Turkana. In the Nariokotome III site west of Lake Turkana,
the Silbo Tuff (0.74 0.01 Ma) (Brown et al., 1985b; Feibel et a nearly complete skeleton of an adolescent Homo erectus was
al.,1989).
discovered in 1985 (Brown et al., 1985a), with additional
Several localities in the lower part of the Koobi Fora, at levels elements found in later field seasons (Leakey and Walker, 1989).
straddling the Burgi Tuff, have yielded fossil hominids, but most The cranium, mandible, and most of the postcranium are
of the extremely abundant and generally well-preserved fossil exceptionally well preserved, allowing accurate estimation of the
hominids in East Turkana occur in the upper part of the Koobi ratio of brain size to body weight and other pertinent compariFora Formation, from just below the KBS Tuff up to the Chari sons to modern humans. Tuff correlations (Brown et al., 1985b)
complex. In the lower Koobi Fora, both Paranthropus and suggest an age close to 1.6 Ma for this specimen. At a lower
Homo are recognized in beds dating to about 2.5 Ma (White, level, within the Lomekwi Member, remains of Paranthropus
1988, table 1), though according to my 1977 notes on the KNM aethiopicus are dated by tuff correlations to an age of 2.6 Ma,
collection, the specimen numbers of KNM-ER 1482 on a closely contemporaneous with the other occurrences of this
mandible assigned to Homo and KNM-ER 1483 on a mandibular species in the Shungura and Koobi Fora formations (Walker et
body without tooth crowns assigned to Paranthropus were al.,1986).
transposed in that publication. Wood (1992) considers mandibles
KNM-ER 1482 and 1483, and also the well-preserved crania Lake Baringo. A hominid temporal bone from the Lake Baringo
KNM-ER 1470 and 3732 just below the KBS Tuff and dating to basin, in the upper or type Chemeron Beds, has been assigned to
about 2.0 Ma, to be among a number of specimens in this age the genus Homo (H. rudolfensis Alexeev, according to Wood,
(more than 130 in all), four mandibles, two fragmentary skulls,
and several postcranial bones. Studies of these fossils are still
largely unpublished. The earliest indications of tool-making, on
stone and on bony material, also occur near the base of Member
E, at 2.6 Ma. An assemblage of pebble andflaketools was found
near the top of the same member, and artifacts become abundant
at the base of Member F. It is noteworthy that the hominid
remains from the Shungura below Member E are presently
assigned to Paranthropus aethiopicus (Coppens) by Kimbell et
al. (1988), whereas the earliest known remains of humans,
classified as Homo rudolfensis Alexeev, are known from Koobi
Fora, Chemeron, and Malawi only from beds dated to 2.6 Ma or
younger, as discussed later. Paranthropus boisei (Leakey and
Leakey) is recognized within Member E, and the first fossil
remains attributable to Homo occur at the top.
The upper part of the Shungura Formation starts with Tuff G
(2.33 Ma) and ends with Member L (ca. 1.4 Ma); members H, J,
and K are much poorer in fossil hominid remains than the lower
members. Within Member H, Tuff H-2 is dated 1.88 0.02 Ma,
and Tuff L has a date of 1.39 Ma (Brown et al., 1985b). The top
of the Olduvai subchron is found in Member J, and Ceding and
Brown (1982) correlated Tuff J-7 with a tuff in the Okote
complex at Koobi Fora, with an age measured at the time close
to 1.6 Ma. Hominid fossil remains from Member G include two
fragmentary skulls, a maxilla, 4 mandibles, 45 isolated teeth, and
several postcranial bones.
Emiliano Aguirre
132
1992). This specimen is dated to about 2.45 Ma (Hill et al.,

1992), approximately coeval with earliest remains and artifacts
from the Turkana Basin. Even earlier artifacts, dated to about
2.85 Ma, are known from the Hadar Formation in Ethiopia, as
discussed later.
Awash Valley
In central Ethiopia, the time interval of interest here is
represented by two areas in the Awash Valley. The first area is
that of the exposures of the Sagantole and Hadar formations,
with a formational contact dated by the presence of the
Lokochot (Shungura A) Tuff, at 3.47 Ma, and the Tulu Bor
(Shungura B) Tuff, at 3.35 Ma, in the Sidi Hakoma Member of
the lower Hadar Formation (Haileab and Brown, 1992).
Australopithecine fossils are known from the Sagantole Formation at Belohdelie, dated to about 3.8 Ma, but otherwise all the
human remains come from the middle part of the Hadar
Formation in the Denen Dora and lower Kada Hadar members,
in the interval between the base of the Denen Dora at 3.1 Ma
and the BKT Tuff dated to 2.85 Ma (Haileab and Brown, 1992).
Site AL 333 in the Denen Dora has yielded more than 300
specimens, attributed to a minimum number of 13 individuals,
showing differences that can be considered as a close approximation of a speciation event in human evolution. From AL 288, just
above an unconformity near the base of the Kada Hadar
Member, the skeleton nicknamed "Lucy" has been assigned to
Australopithecus afarensis, but it could reasonably be assigned to
Homo instead. The earliest dated assemblage of stone artifacts in
East Africa derives from the lower part of the Kada Hadar
Member, just below the 2.85-Ma BKT Tuff. This tuff was
previously K/Ar-dated to 2.63 0.05 Ma by Taieb and Tiercelin
(in Chavaillon, 1982), and to 2.58 0.23 Ma by fission-track
dating (Chavaillon, 1982; Harris, 1986).
The record in this region is continued at Melka-Kunture,
farther to the west, which is also well calibrated with paleomagnetic and radiometric dates (Chavaillon, 1982). The lower
levels, with reversed polarity, are correlated to the upper
Matuyama above the Olduvai normal. The oldest sites contain
tools of mixed Acheulean-Olduwan aspect, and at Gombore I-B
part of a human humerus is dated to approximately 1.6 Ma.
Higher sites, such as Gombore II and Garba XII, have yielded
extremely abundant tools of Acheulean traditions, and other
human fossils have been found at Garba IV (1.35 Ma) and again
at Gombore II-C between Tuff B and Tuff C, dated 0.84 and 0.80
Ma, respectively. The fossil hominids of Gombore I, Garba IV,
and Gombore II are attributed to Homo erectus. Finally, Tuff D,
dated 0.74 Ma, includes an artifact assemblage of evolved
Acheulean tradition.
Olduvai Gorge
In central northern Tanzania, the Rift Valley encloses a broad
sedimentary basin adjacent to an array of young volcanoes. The
only place in which the Plio-Pleistocene sediments of this basin
are deeply exposed is the unique canyon of Olduvai Gorge.

Older Pliocene beds exposed in shallow gullies to the south at
Laetoli, which have yielded australopithecine remains and
footprints, as well as abundant mammalian fossils, appear to be a
continuation of the Olduvai sequence (Cooke, Chapter 27, this
volume).
The lower units (called "beds" by convention) of the Olduvai
Gorge sequence, including most of the section and most of the
fossil-bearing outcrops, have long been calibrated by radiometric
dates, most recently by Walter et al. (1991). An age of 0.62 Ma
has been estimated for the top of Bed IV (Leakey and Hay,
1982), while the earliest Bed I sediments at Olduvai, below the
Lower Basalt in the upper gorge, date to approximately 2.0 Ma
(Walter et al., 1991). The boundaries of the paleomagnetically
normal strata representing the magnetostratotype of the Olduvai
subchronozone overlap the interval from Tuff I-A to Tuff I-F,
with fossiliferous deposits beginning at the level of Tuff I-B. The
radiometric dating of these tuffs by Walter et al. (1991) suggested
a duration for the Olduvai subchron between 2.0 and 1.75 Ma,
which compares well to the astronomically tuned calibration of
1.95 to 1.79 Ma (Preface, this volume). Bed II, above Tuff I-F,
extends up to about 1.5 Ma. Olduvai Bed IV (0.83-0.62 Ma)
straddles the Brunhes-Matuyama magnetic reversal, which is
recorded in Tuff IV-B.
The type specimen and associated material of Homo habilis, as
well as specimens assigned to Australopithecus boisei, come from
Bed I and lower Bed II. In particular, well-preserved remains of
H. habilis, including KNM-OH7, OH8, OH13, OH24, OH35,
and OH48, come from various levels, which range from just
beneath Tuff I-B (1.95 Ma) to just below Tuff II-B (1.7 Ma). All
of these fossils are younger than the hominids of Shungura E-G
(before 2.0 Ma) and probably those found below the KBS Tuff of
East Turkana (before 1.9 Ma), whereas on present dating they
are slightly older than the earliest Homo at Melka Kunture. The
skull of OH9, attributed to Homo erectus, was discovered out of
stratigraphic context but is inferentially related to a level in
upper Bed II, above Tuff II-D. It is younger, in any case, than
the erectus skull, KNM-ER 3733, in the upper KBS Member at
Koobi Fora if the assumed dates are correct. The variability in
dating does not preclude the possibility that the last H. habilis
and the earliest H. erectus (i.e., H. ergaster of some authors)
specimens are coincident. It is likely that representatives of the
H. erectus clade were in fact present in the vicinity of the Rift
Valley before 1.5 Ma, and perhaps before 1.7 Ma, according to
the evidence from dated specimens at Olduvai and East Turkana.
The robust australopithecines are not found above the top of
Bed II, about the same age as the Chesowanja specimen (1.5
Ma); their last appearance is thus above the level of the robust
specimen found at Peninj, west of Lake Natron (about 1.6 Ma),
and below the level of the Homo erectus specimens associated
with the Chari Tuff (1.4 Ma) at Koobi Fora and Shungura in the
Turkana Basin.
There is a general lacuna in the East African hominid record
from the level of upper Bed II until the uppermost Upper
Pleistocene strata of Kanam (Lake Victoria), dated to about 1.1
Ma, Olduvai Bed IV, Olorgesailie (0.95 Ma), and Guomde (East
Turkana), dated to about 1.0 Ma.
South Africa
The chronology of Plio-Pleistocene fossil Hominidae from the
South African caves is less precise than that in East Africa, with
no radiometric dates and only a debatable paleomagnetic record
from Makapansgat. Two alternative age sequences have been
proposed for the fossils from the younger "Australopithecine
caves." The first is correlated in a morpho climatic approach, set
out by Partridge (1982), while the second follows faunal
correlations with East African sequences by Cooke (1978), Vrba
(1982), and others. The two approaches almost coincide in their
estimations for the most likely age of the main level of
Sterkfontein (St. 4, with Australopithecus africanus), suggesting
an age range between 2.8 Ma and 2.4 Ma. The upper level of
Sterkfontein (St. 5, with Homo cf. H. erectus) is estimated to
have an age between 1.6 and 1.4 Ma, if not slightly younger,
which would correlate to a level in upper Olduvai Bed II.
The dates that have been suggested for the other sites are less
consistent, regardless of the approach. Makapansgat Member 3,
the earliest australopithecine level, has been assigned an age of
just over 3 Ma by Partridge (1982), based on the interpretation
that middle Gauss (Mammoth and Kaena) paleomagnetic
polarities are recorded in the overlying Member 4, also with
hominid remains. The paleomagnetic profile, however, could be
interpreted as the upper part of the Gauss, with an age between
2.8 and 2.4 Ma, like the St. 4 main level at Sterkfontein. All of
the fossil hominines from Makapansgat are commonly attributed
to A. africanus, but Aguirre (1970) suggested that the lower jaw
of an immature individual resembles Paranthropus in some
features. This question should now be reassessed in view of the
discovery of a primitive species of that genus, P. aethiopicus,
from beds of that age in the Turkana Basin.
The age of the deposits of Swartkrans 1, with the robust
australopithecine Paranthropus and indications of early Homo,
should be 1.9-1.6 Ma, according to the faunal correlations, just
as at Olduvai, whereas Partridge (1982) estimates it to be
between 1.55 and 1.2 Ma, and thus significantly younger than the
youngest known Australopithecus from the well-dated East
African hominid record. The age of Swartkrans 2, with H.
erectus, but no Paranthropus, should be 1.0 Ma, according to
Vrba (1982), but Partridge (1982) estimates it as much younger,
corresponding to early middle Pleistocene. Finally, two age
estimates for Member B of Kromdraai, with robust australopithecines, are 2.1-1.8 Ma (Vrba, 1982) or 1.25-1.0 Ma
(Partridge, 1982).
Northwest Africa
Among the fossil Hominidae known from Middle Pleistocene
sites in Northwest Africa, the skull and associated material of the
"advanced Homo erectus" of Tighenif (also known as Palikao
133
and Ternifine), Algeria, must be mentioned, because the most

recently estimated age there is 0.7 Ma or even older (Hublin,
1985). Other specimens of younger age, between 0.5 and 0.4 Ma,
have been found in coastal-plain terraces on the Moroccan coast
at Sale, Rabat, Sidi Abderrhaman, and Thomas Quarry (Casablanca), having features attributed to the transition from erectus
to sapiens.
Dispersal event in Eurasia
The earliest human fossils outside of Africa, in order of

increasing distance from mainland Africa, are the cranial
fragments of Tell 'Ubeidiya in Israel, which in fact may be
considered paleogeographically as an outlying part of the AfroArabian biotic province, the mandible recently discovered at
Dmanisi, Georgia, the skull from Gongwangling (and possibly
the teeth from Yuanmou) in China, and a partial cranium and
two mandibles from the lowest levels at Sangiran, Java. All of
these are assigned ages older than the Matuyama-Brunhes
reversal, although questionably so in the case of Gongwangling,
and much more questionably in the case of Sangiran. Most
modern authors, however, believe that all the fossil hominids
from Sangiran and Trinil date from the Matuyama, and a recent
redating of the lower Sangiran (Swisher et al., 1994) has
suggested that the Java record of erectus may be as old as in
Africa. Overall, the Eurasian finds indicate that the earliest
members of this species expanded rapidly into a much wider
geographic range than had their African precursors, although
perhaps in tropical areas before more temperate ones. Two
localities in Spain, Orce and Cueva Victoria, have also been
suggested to yield fossil hominids older than 1 Ma, but the
evidence is not constraining in either of these two cases (Aguirre,
Chapter 13, this volume).
Because their fossil remains are so rare, the dispersal of
humans in Eurasia has been amplified appropriately by studies of
archeological material, including tools and living sites. The most
acceptable conclusion from this kind of evidence points to the
early Pleistocene, just prior to 1 Ma, as the time when humans
first became established outside of Afro-Arabia, in the warmwinter zones of Eurasia from the Near East to India, Java, and
northern China. In agreement with older dates in Java (Swisher
et al., 1994), fossil remains from the pioneer population, at
present known only from the Caucasus (Dmanisi) and central
Java, show plesiomorphic features in common with the earliest
populations of Homo erectus (i.e., the earliest-evolving stem
distinct from H. habilis) known in Africa. A later dispersal
introduced humans into western Eurasia close to the beginning
of the Brunhes chron (0.7 Ma), again documented primarily
from archeological records. That dispersal involved a population
of more advanced character, whose oldest fossil remains (e.g.,
Atapuerca, Arago, Petralona) have close similarities to East
African material of the same age: the later part of the early
Pleistocene to the early part of the middle Pleistocene (Aguirre
et al., 1980; Arsuaga et al., 1993).
Emiliano Aguirre
134
1983), but the paleomagnetic interpretation and fauna are also

consistent with a post-Jaramillo age, near 1.0 Ma. Finally, with
In Tell 'Ubeidiya, Israel, human fossil remains assigned to H. regard to archeological evidence, in the Nihewan Basin of Hebei
erectus have been found in association with early Acheulean province, northern China, the occurrence of abundant paleotools. The site has been dated, with some difficulty, according to lithic implements associated with late Villafranchian mammal
uncertain correlation of the mammalian fauna. Most likely the remains at Donggutuo and several other pre-Jaramillo sites
age of the 'Ubeidiya humans could be placed in the later part of (Schick et al., 1991) appears to record human activity in that
region at least as early as 1.2 Ma.
the early Pleistocene (Tchernov, 1987) at about 1.1 Ma.
It is noteworthy that the ground-dwelling ape Gigantopithecus
What may be the oldest human fossil in mainland Eurasia was
continued
to live for a short time in southern Asia (and maybe
discovered at Dmanisi, Georgia, in 1991 (Dzaparidze et al.,
also
in
Java)
sympatrically with the earliest populations of Homo
1991). The specimen is a nearly complete, undeformed adult
erectus.
There
may have been little competition between them in
mandible that presents a mixture of progressive features, such as
terms
of
ecological
specialization and use of resources, as in the
a molar series decreasing in size anteriorly, along with other
limited
interaction
between
Homo and the other great apes, but
features that are more primitive. It resembles Homo habilis and
perhaps
the
analogy
should
be drawn with Paranthropus,
australopithecines in possessing such plesiomorphic features as a
another
large
bipedal
hominoid
that
became extinct shortly after
narrow mandibular body with slightly divergent horizontal
the
first
Acheulean
tool-users
appeared
in Africa.
branches, molar rows converging posteriorly, robust Ml with
conspicuous cingulum, and an elongated P3 with a poorly
developed lingual cuspule. A sinuosity in the vertical anterior
Java
profile of the symphyseal region creates a depression resulting in
a chin-like prominence which resembles a feature seen in some
There are two opinions concerning the age of the "pithecanthroaustralopithecines, in contrast to a true chin.
The fossil-bearing horizon at Dmanisi overlies a basalt dated pines," the original Homo erectus from Indonesia, even though
to 1.8 0.1 Ma. The fauna associated with the mandible has the stratigraphy is fairly well established. The precise horizons
been assigned to the Apsheronian Stage, or Lower Pleistocene in from which the older collections came have been carefully
the Soviet schema (Nikiforova, Chapter 20, this volume). The identified, and an overall stratigraphic framework is generally
fauna is of Upper Villafranchian character, with Mammuthus accepted for the relative ages of the fossils recovered in different
meridionalis meridionalis, Canis etruscus, Bos, and pre-caballinelocalities. The key unit is the Grenzbank, a conglomerate facies
horses. Although it is not as old as the lower Apsheronian of variable thickness that rests on the Sangiran Black Clays and
Kocachuri local fauna in which Leptobos rather than Bos is still that underlies the Kabuh Formation at Sangiran on the upper
present, the Dmanisi assemblage is also not younger than the Solo River (Semah, Chapter 28, this volume). The Black Clays
levels at about 1.1 Ma in which advanced subspecies of M. correlate to the Pucangan Formation and to the Ci Saat faunal
meridionalis begin to appear. In other words, on preliminary assemblage, according to Leinders et al. (1985). Those authors
assessment of faunal evidence, the age of the mandible appears further correlated the Trinil fauna found along the middle Solo
to lie between 1.5 and 1.2 Ma, and if that should be River to the level of the Grenzbank and just below, and they
substantiated, it will be further evidence of the dispersal of assigned the Kedung Brubus fauna to the Kabuh Formation.
Opinions differ as to the relative ages of the local stratigraphic
Homo out of Africa very shortly after the evolution of the erectus
units,
although recent opinion has held that the dating of a few
clade.
decades ago was erroneously old. Magnetostratigraphic research
(Shimizu et al., 1985; Semah, 1986; Semah, Chapter 28, this
volume)
strongly supports an upward revision for the earliest
China
faunal levels, although the most recent dating (Swisher et al.,
Paleomagnetic data suggest that the fossil skull of H. erectus 1994) is contradictory.
According to one alternative interpretation of paleomagnetic
from Gongwangling, Lantian area, has a pre-Brunhes age older
than 0.8 Ma (Wu, 1985; Aguirre et al., Chapter 9, this volume), data, as discussed by Semah (1986), there is enough uncertainty
although Kahlke (1986) placed it younger than that, according to about the presence of reversed-polarity strata in the lower
faunal correlation. Zhou (1990) proposed an age of 1.15 Ma, Kabuh that it is reasonable to place the entire Kabuh Formation
through correlation with the L15 loess, which agrees well with in the Brunhes magnetic epoch, with the Grenzbank unit almost
the dating of archeological material in the lower Nihewan coincident with the Matuyama-Brunhes reversal. The age of the
sequence (as discussed later). Fossil teeth ascribed confidently to earliest hominid specimens of Java, those from the underlying
erectus from Yuanmou, Yunnan, were initially assigned an age of Sangiran Black Clays, would thus be no older than the latest
more than 1.6 Ma (yel 1.8) on the basis of paleomagnetic Matuyama, above the Jaramillo, and most of the classic
analysis. On faunal grounds, the underlying normal-polarity "pithecanthropines," which come from the uppermost Sangiransediments have been reassigned to the Brunhes, so that the Grenzbank-lower Kabuh interval and the age-equivalent Trinil
Yunnan fossils may be younger than 0.8 Ma (Liu and Ding, strata, would be dated between 0.6 and 1.0 Ma.
Middle East
On the other hand, Shimizu et al. (1985) found that the top of
the Jaramillo is clearly evidenced just below the basal Grenzbank, indicating that the controversial interpretations of reversed polarity in the Grenzbank and lower Kabuh are probably
valid. In this view, the range of typical "pithecanthropines" in
Java extended from the Matuyama-Brunhes reversal down to a
level well below the Jaramillo, as discussed later. The dating of
Swisher et al. (1994) on a tuff just above the oldest specimens at
Sangiran (S27 and S31) indicates an age of 1.66 Ma, well down in
the upper Matuyama. They also dated hornblende from normally magnetized volcaniclastic deposits at the site of the
Modjokerto skull at Perning, chemically comparable to material
from within the skull itself, at 1.81 Ma, consistent with an age
within the Olduvai subchron and thus slightly older than the
oldest dated erectus-grade hominids in Africa.
The biostratigraphic and radiometric data appear to be in
better agreement with the older interpretation, rather than with
the younger interpretation. In the biostratigraphic analysis of
Leinders et al. (1985) the Kedung Brubus local fauna in the
lower Kabuh Formation is correlated to an age older than 0.7
Ma. The fossils from Trinil and those of the Grenzbank at
Sangiran are placed at the level of the Jaramillo subchron (1.1
Ma), while the material from the top of the Black Clays and the
Ci Saat fauna are dated to about 1.3 Ma, or (pace Swisher et al.,
1994) even older. Itihara, Kadar, and Watanabe (1985) noted
that in regard to the three pumice tuffs identified within the
Kabuh Formation (their Bapang),fission-trackages by Suzuki et
al. (1985) dated the Upper Tuff to the lowermost Brunhes, and
the Middle Tuff at 0.78 Ma to the latest Matuyama. That dating
agrees with the determination that the 0.71-Ma tektites found at
different levels in the middle Kabuh are in situ just above the
Middle Tuff and that the Brunhes-Matuyama boundary is
therefore correctly placed between these horizons.
135
on the inner surface is quite similar in shape to one characteristically found in the Equidae (S. Moya-Sola and J. Agusti, in
Gibert-Clols et al., 1989), and very uncommon in humans.
Nevertheless, immunological assay of the specimen has shown a
closer affinity to humans than to horses (Gibert-Clols et al.,
1989, pp. 225-228).
The Cueva Victoria fossil, found in a cave setting, is a second
phalanx of the fifth finger (Gibert-Clols et al., 1989, pp. 395406) that looks definitely human and cannot be attributed to any
other known primate. Although it was not collected in situ, all
the evidence leads to the conclusion that the Cueva Victoria
phalanx came from a bone breccia that completely filled the cave
in the early Pleistocene. The very abundant faunal remains,
which are ascribed to hyena accumulation, are well correlated to
the Sinzelles faunal assemblage (1.3-1.4 Ma) by Moya-Sola and
Menendez (1986; Agusti, 1986; Aguirre et al., Chapter 9, this
volume). Fossils of an undescribed giant baboon have been
recovered from the Cueva Victoria breccia (J. Moya-Sola,
personal communication), but the possibility that this is the
proper attribution for the phalanx can be positively ruled out
(Perez-Perez, 1989).
Conclusion
Evolution of the genus Homo can be closely tied to the PliocenePleistocene boundary, located in the uppermost part of the
Olduvai subchron. The earliest known species, H. rudolfensis,
was replaced by H. habilis within the time interval of the
Olduvai, and shortly after it ended, the H. erectus clade evolved
in Africa and rapidly spread into southeastern Asia. The
dispersal of humans into southwestern Eurasia was not much
later. The present state of knowledge therefore suggests that the
first major dispersal of humans outside of Africa took place close
to the Pliocene-Pleistocene boundary.
Southern Europe
The attribution of a skull fragment and a fragment of bone from
the locality of Venta Micena, near Orce (Gibert-Clols et al.,
1989), to the genus Homo is questionable, although the
stratigraphic position is well established in the Baza endorheic
sequence (Anadon et al., 1987; Aguirre, Chapter 13, this
volume). This level corresponds to the median third of the early
Pleistocene and has a rich fauna younger than that of Sinzelles
(1.3-1.4 Ma) and older than that of Vallonet (0.95 Ma) (MoyaSola et al., 1981). By interpolation, the specimen can thus be said
to date from approximately 1.1 Ma, with an error margin of as
much as 10%, which is about the same age and uncertainty as for
the Tell 'Ubeidiya remains. The fossil in question is a small
fragment of a cranial vault, including the lambda region, and
what arguably may be a section of the coronoidal suture. The
bone is very thin, with a very poorly developed pneumatized
layer. That would indicate a very immature human, but it is
inconsistent with a partial fusion of the sutures. Moreover, the
shape of the suture is highly uncommon in humans, but is more
often found in several other orders of Mammalia. A crestal ridge
References
Aguirre, E. 1970. Identification de "Paranthropus" en Makapansgat. In Cronica del XI Congreso Nacional de

Arqueologia, Merida 1969, ed. F. Beltran, pp. 98-124.
Zaragoza: Asociacion Espanola de Arqueologia.
Aguirre, E., deLumley, M. A., Basabe, J. M., and Botella, M.
1980. Affinities between the mandibles from Atapuerca
and l'Arago, and some East-African fossil hominids. In
Proceedings of the 8th Panafrican Congress of Prehistory
and Quaternary Studies, Nairobi, 5-10 September 1977,
ed. R. E. Leakey and B. A. Ogot, pp. 171-174. Nairobi:
TILLMIAP, National Museums of Kenya.
Agusti, J. 1986. Synthese biostratigraphique du Plio-Pleistocene
de Guadix-Baza (province de Granada, Sud-est de
L'Espagne). Geobios 19:505-510.
Agusti, J., Moya-Sola, S., Martin-Suarez, E., and Marin, M.
1987. Faunas de Mamiferos en el Pleistoceno inferior de la
region de Orce (Granada, Esparia). In Geologia y
Paleontologia del Pleistoceno inferior de Venta Micena, ed.
S. Moya-Sola et al., pp. 73-86. Paleontologia i Evolucio,
Memoria especial, no. 1. Sabadell: Institut de Paleontologic Dr. M. Crusafont.
136
Emiliano Aguirre
Anadon, P., Julia, R., De Decker, P., Rosso, J. C , and SoulieMarsche, I. 1987. Contribution a la paleolimnologia del
Pleistoceno inferior de la Cuenca de Baza (sector OrceVenta Micena). In Geologia y Paleontologia del Pleistoceno inferior de Venta Micena, ed. S. Moya-Sola et al., pp.
35-72. Paleontologia i Evolucio, Memoria especial, no. 1.
Sabadell: Institut de Paleontologic Dr. M. Crusafont.
Arsuaga, J. L., Martinez, I., Garcia, A., Carretero, J. M., and
Carbonnel, E. 1993. Three new human skulls from the
Sima de los Huesos, Middle Pleistocene site in Sierra de
Atapuerca, Spain. Nature 362:534-537.
Brown, F. H., Harris, J. M., Leakey, R. E. L., and Walker, A.
1985a. Early Homo erectus skeleton from West Lake
Turkana, Kenya. Nature 316:788-791.
Brown, F. H., McDougall, I., Davies, T., and Maier, R. 1985b.
An integrated Plio-Pleistocene chronology for the
Turkana Basin. In Ancestors: the hard evidence, ed. E.
Delson, pp. 82-90. New York: Liss.
Ceding, T. E., and Brown, F. H. 1982. Tuffaceous marker
horizons in the Koobi Fora region and the Lower Omo
Valley. Nature 299:216-221.
Chavaillon, J. 1982. Position chronologique des hominides
fossiles d'Ethiopie. In ler Congres International de
Paleontologie humaine. Pretirage, vol. 2, ed. H. deLumley
and M. A. deLumley, pp. 166-791. Nice: Centre National
de la Recherche scientifique.
Cooke, H. B. S. 1978. Faunal evidence for the biotic setting of
early African hominids. In Early hominids of Africa, ed.
C. Jolly, pp. 267-281. London: Duckworth.
Coppens, Y., Howell, F. C , Isaac, G. L., and Leakey, R. E. F.
(eds.) 1976. Earliest man and environments in the Lake
Rudolf Basin. University of Chicago Press.
Dzaparidze, V., Bosinski, G., Bugianisvili, T , Gabunia, L.,
Justus, A., Klopotovskaja, N., Kvavadze, E., Lordkipanidze, D., Majsuradze, G., Mgeladze, N., Nioradze, M.,
Pavlenisvili, E., Schmincke, H. U., Sologasvili, D. Z.,
Tustfabramasvili, D., Tvalkrelidze, M., and Vekua, A.
1991. Der altpalaolitische Fundplatz Dmanisi in Georgien
(Kaukasus). Romisch-Germ. Zentralmus. Mainz, Jahrb.
36:67-116.
Feibel, C. G., Brown, F. H., and McDougall, I. 1989.
Stratigraphic context of fossil hominids from the Omo
Group deposits, northern Turkana Basin, Kenya and
Ethiopia. Am. J. Phys. Anthropol 78:595-622.
Gibert-Clols, J., Campillo-Valero, D., and Garcia-Olivares, E.
(eds.) 1989. Los restos humanos de Orcey Cueva Victoria.
Sabadell: Institut Paleontologic Dr. M. Crusafont.
Haileab, B., and Brown, F. H. 1992. Turkana Basin-Middle
Awash Valley correlations and the age of the Sagantole
and Hadar formations. /. Hum. Evol. 22:453-468.
Harris, J. 1986. Decouverte de materiel archeologique olduwayen dans le Rift de l'Afar. VAnthropologie 90:339-357.
Hill, A., Ward, S., Deino, A., Curtis, G. H., and Drake, R.
1992. Earliest Homo. Nature 355:719-722.
Hublin, J. J. 1985. Human fossils from the North African Middle
Pleistocene and the origins of Homo sapiens. In Ancestors:
the hard evidence, ed. E. Delson, pp. 283-288. New York:
Liss.
Itihara, M., Kadar, D., and Watanabe, N. 1985. Concluding
remarks. In Quaternary geology of the hominid fossil
bearing formations in Java, ed. N. Watanabe and D.
Kadar, pp. 367-378. Special Publication no. 4. Bandung:
Geological Research and Development Centre.
Johanson, D. C. 1989. A partial Homo habilis skeleton from
Olduvai Gorge, Tanzania: a summary of preliminary
results. In Hominidae, ed. G. Giacobini, pp. 155-166.

Milan: Jaca Book.
Kahlke, H. D. 1986. Correlations biostratigraphiques entre les
sites a hominides anciens d'Eurasie. LAnthropologie
90:401-405.
Leakey, M. D., and Hay, R. L. 1982. The chronological position
of the fossil hominids of Tanzania. In ler Congres
International de Paleontologie humaine. Pretirage, vol. 2,
ed. H. deLumley and M. A. deLumley, pp. 753-765. Nice:
Centre National de la Recherche scientifique.
Leakey, R. E., and Ogot, B. A. (eds.) 1980. Proceedings of the 8th
Panafrican Congress of Prehistory and Quaternary Studies,
Nairobi, 5-10 September 1977 Nairobi: TILLMIAP, National Museums of Kenya.
Leakey, R. E., and Walker, A. 1989. Early Homo erectus from
West Lake Turkana, Kenya. In Homindae, ed. G.
Giacobini, pp. 209-216. Milan: Jaca Book.
Leinders, J. J. M., Aziz, F , Sondaar, P. Y, and de Vos, J. 1985.
The age of the hominid-bearing deposits of Java: state of
the art. Geol. Mijnb. 64:167-173.
Liu Tungsheng and Ding Menglin. 1983. Discussion on the age of
"Yuanmou Man." Acta Anthropol. Sin. 2:40-48.
Maglio, V. J. 1972. Vertebrate faunas and chronology of
hominid-bearing sediments east of Lake Rudolf, Kenya.
Nature 239:379-385.
Matsu'ura, S. 1982. A chronological framing for the Sangiran
hominids: fundamental study by the fluorine dating
method. Nat. Sci. Museum, Tokyo, Bull. D 8:1-53.
Moya-Sola, S., Agusti, J., Gibert, J., and Pons-Moya, J. 1981. El
yacimiento cuaternario de Venta Micena (Espana) y su
importancia dentro de las asociaciones faunisticas del
Pleistoceno inferior europeo. Paleontol. Evol. (Sabadell)
16:39-53.
Moya-Sola, S., and Menendez, E. 1986. Los Artiodactilos
(Bovidae y Cervidae, Mammalia) del Pleistoceno inferior
de Europa Occidental: ensayo de sintesis. Paleontol. Evol.
20:289-295.
hominids of Southern Africa. In ler Congres International
de Paleontologie humaine. Pretirage, vol. 2, ed. H.
deLumley and M. A. deLumley, pp. 617-675. Nice:
Centre National de la Recherche scientifique.
Perez-Perez, A. 1989. La falange de Cueva Victoria: analisis
discriminante y afiliacion taxonomica. In Los restos
humanos de Orce y Cueva Victoria, ed. J. Gibert-Clols et
al., pp. 407-414. Paleontologia i Evolucio, Memoria
especial no. 1. Sabadell: Institut Paleontoligic Dr. M.
Crusafont.
Schick, K., Toth, N., Wei Qi, Clark, J. D., and Etler, D. 1991.
Archaeological perspectives in the Nihewan Basin, China.
/. Hum. Evol. 21:13-26.
Semah, F 1986. Le peuplement ancien de Java: ebauche d'un
cadre chronologique. L Anthropologie 90:359-400.
Shimizu, Y, Mubroto, B., Siagian, H., and Untung, M. 1985. A
paleomagnetic study in the Sangiran area. In Quaternary
geology of the hominid fossil bearing formations in Java,
ed. N. Watanabe and D. Kadar, pp. 275-307. Special
Publication 4. Bandung: Geological Research and Development Centre.
Swisher, C. C , III, Curtis, G. H., Jacob, T , Getty, A. G.,
Suprijo, A., and Widiasmoro. 1994. Age of the earliest
known hominids in Java, Indonesia. Science 263:11181121.
Suzuki, M., Wikarno, Budisantoso, Saefudin, I., and Itihara, M.
1985. Fission track ages of pumice tuff, tuff layers, and
137
javites of hominid fossil bearing formations in Sangiran Walter, R. C , Manega, P. C , Hay, R. L., Drake, R. E., and
Curtis, G. H. 1991. Laser-fusion 40Ar/39Ar dating of Bed
area, Central Java. In Quaternary geology of the hominid
I, Olduvai Gorge, Tanzania. Nature 354:145-149.
fossil bearing formations in Java, ed. N. Watanabe and D.
Kadar, pp. 310-357. Special Publication 4. Bandung: White, T. D. 1988. The comparative biology of "robust"
Geological Research and Development Centre.
Australopithecus: clues from context. In Evolutionary
Tchernov, E. 1987. The age of the 'Ubeidiya Formation, an early
history of the "robust" Australopithecines, ed. F. E. Grine,
Pleistocene hominid site in the Jordan Valley, Israel. Isr. J.
pp. 449-483. New York: Aldine de Gruyter.
Earth Sci. 36:3-30.
Wood, B. 1992. Origin and evolution of the genus Homo. Nature
Vrba, E. S. 1982. Biostratigraphy and chronology, based
355:783-790.
particularly on Bovidae, of southern hominid-associated Wu Rukang. 1985. New Chinese Homo erectus and recent work
assemblages. In ler Congres International de Paleontologie
at Zhoukoudian. In Ancestors: the hard evidence, ed. E.
humaine. Pretirage, vol. 2, ed. H. deLumley and M. A.
Delson, pp. 245-248. New York: Liss.
deLumley, pp. 707-752. Nice: Centre National de la Zhou, Y. 1990. The living environment of human fossils in north
Recherche scientifique.
China and the Asia monsoon variation. In INQUA
Walker, A., Leakey, R. L., Harris, J. M., and Brown, F. H.
International symposium on Quaternary stratigraphy and
1986. 2.5 Myr Australopithecus boisei from west of Lake
events in Eurasia and the Pacific region, Yakutsk, USSR,
1990. Abstracts (preprint). Moscow: Akademia Nauk.
Part IV
The Pleistocene boundary in regional sequences
11
The Pliocene-Pleistocene boundary in Italy

AUGUSTO AZZAROLI, MARIA LUISA COLALONGO, HISAO NAKAGAWA, GIANCARLO PASINI,
DOMENICO RIO, GIULIANO RUGGIERI, SAMUELE SARTONI, and RODOLFO SPROVIERI
Introduction
Italy is one of the primary sources of data for the Pliocene and
Pleistocene. The Italian coastal margins were affected by
tectonism during the Pleistocene more than any other part of
western Europe, and as a result the Upper Cenozoic marine
sequences that crop out along the coast in Sicily and in the
mainland Apennine foothills (Figure 11.1) have been the
subjects of classic studies on the chronostratigraphy of that age.
Additionally, Plio-Pleistocene strata with stratigraphic thicknesses exceeding 8,000 m have been sampled in boreholes in the
subsurface of the central Po Valley. Finally, Pliocene and Lower
Pleistocene nonmarine strata are common in the montane basins
of the Italian peninsula.
Since the very beginning of stratigraphy as a science, the
abundance of Cenozoic fossils, particularly in the marine record,
has attracted attention to the Italian sections. Lyell's original
designations of the "older" and "younger" Pliocene (Lyell,
1833), the latter of which became the Pleistocene, relied heavily
on the malacofaunas of the northern Apennines. For the same
reason, all standard Pliocene and Pleistocene marine chronostratigraphic terms used in global correlation are based on the
biostratigraphy of the Italian sequences. In recognition of this
historical importance, the XVIII International Geological Congress in London in 1948 confirmed a general consensus that the
Pliocene-Pleistocene boundary should be defined in the exposed
Italian marine strata.
In this chapter we shall review the stratigraphic data available
from the marine and continental records of Italy, as background
for defining the Pliocene-Pleistocene boundary. The known
Italian Plio-Pleistocene sections are so numerous that it would
be impossible to treat them all in this brief review, and therefore
we shall consider only those sections that are well documented
and stratigraphically significant and that have been historically
important in subdividing the Pliocene and the Pleistocene.
Biostratigraphic and chronostratigraphic framework
Several fossil groups (mollusks, ostracodes, foraminifers, calcareous nannofossils, pollen, etc.) are biostratigraphically use-
ful in the Italian marine record. The most detailed and most
reliable correlations are obtained through planktic foraminifera
(Pasini and Colalongo, Chapter 2, this volume) and calcareous
nannofossils (Rio, Raffi, and Backman, Chapter 5, this volume). The other groups are less reliable for long-distance
correlations, although they can provide additional stratigraphic
controls.
The biostratigraphic schemes based on planktic foraminifera
and calcareous nannofossils, as discussed by the aforementioned authors, are shown in Figure 11.2. Figure 11.2 also
shows the informal biostratigraphic subdivisions of the Lower
Pleistocene of Ruggieri et al. (1976). Zones C, D, and E of
Ruggieri and co-workers are defined, in ascending order, by the
successive entrances into the Mediterranean basin of Arctica
islandica (or other time-equivalent "northern guests"), Hyalinea
baltica, and finally Globorotalia truncatulinoides excelsa. The
chronostratigraphic subdivisions adopted for the Lower Pleistocene (Figure 11.2) are discussed in Ruggieri et al. (1984) and
Rio et al. (1991). In particular, for the recognition of the
Pliocene-Pleistocene boundary, we have used the definition
proposed by INQUA Subcommission 1-d, "Pliocene/Pleistocene Boundary," and adopted by action of the IUGS, namely,
the base of the claystone conformably overlying sapropelic layer
e of the Vrica section in Calabria (Pasini and Colalongo,
In reviewing the individual sections, we shall also discuss the
available magnetostratigraphy in light of the biostratigraphic
framework presented earlier. Cross-checks of magnetostratigraphy and biostratigraphy are crucial in the Italian sections
considered here, because all of them are erosionally truncated
and lack radiometric control. Therefore, only calibrated biostratigraphic events can allow correlation of the local magneticpolarity reversals to the standard MPTS (magnetic polarity time
scale). The chronometry followed here was originally developed
according to the MPTS of Mankinen and Dalrymple (1979), but
their values have been translated throughout this text according
to the newly developed orbital time scale (Preface, this volume).
The biostratigraphic events reported in Figures 11.2 and 11.9
have been tied to geomagnetic polarity, following Rio et al.
(Chapter 5, this volume).
141
Azzaroli et al.
142
CASTELL' ARQUATO
SANTERNO
S. MARIA Dl CATANZARO
d VRICA
LE CASTELLA
MONASTERACE
100
200 Kn
Figure 11.1. Locations of the Italian sections discussed in this chapter.
Marine record
We note, in reference to the marine criteria for subdividing the

Upper Cenozoic of Italy, that Globorotalia truncatulinoides
excelsa Sprovieri, Ruggieri, and Unti, 1981, is a subspecies
representing the most advanced morphotypes of the species and
is restricted to the later Pleistocene of the Mediterranean. The
first occurrence of the typical G. truncatulinoides truncatulinoides in the Mediterranean is recorded at the stratigraphic level
where Globorotalia inflata also first appears (Rio et al., 1984b).
This stratigraphic level is considered to be equivalent to the base
of the Upper Pliocene by scholars who use a threefold
subdivision of the Pliocene (Colalongo, Elmi, and Sartoni,
1974). In the earlier literature, of course, the excelsa forms were
not distinguished from the earlier, more typical subspecies.
Northern Italy
Santerno Valley. In the Santerno Valley (Bologna province),
between the villages of Tossignano and Imola, a thick sequence
of marine Pliocene and Lower Pleistocene sediments crops out in
two measured sections: the Santerno River section and the West
Santerno section.
The Santerno River section (Figure 11.3) has been sampled
along the bed of the river except where alluvium or debris

requires a lateral shift to the slopes of the valley. The sediments
are poorly stratified gray-blue silty clays and sandy interbeds of
turbiditic origin (Colalongo et al., 1982b). The sedimentary
environment is bathyal in the lower and middle parts; at the top
of the section, gray and yellow sands, more or less cemented, are
indicative of a littoral-sublittoral environment (Colalongo et al.,
1982b). The total thickness of the section is 2,600 m, according to
Cremonini, Elmi, and Monesi (1969). Abundant macrofauna
and microfauna have been reported in biostratigraphical studies
by Ruggieri (1975), Colalongo (1968), Colalongo et al. (1974),
Padovani and Tampieri (1970, 1974), Francavilla (1974), Raffi
and Rio (1980a), Rio, Sprovieri, and Raffi (1984a), and Rio et
al. (1991).
The Santerno River section is the most informative among the
marine sections studied so far in northern Italy, and has been
designated by Ruggieri and Sprovieri (1977) as the stratotype
section of Santernian and Emilian chronostratigraphic units. The
Santernian and Emilian, together with the Sicilian Stage of the
Palermo area, were united as substages of the new Selinuntian
Stage (Ruggieri et al., 1984), which those authors considered to
be a replacement for the Calabrian Stage as a primary
subdivision of the Lower Pleistocene.
The Pliocene-Pleistocene boundary traditionally has been
traced in the Santerno section at the first appearance of the
pelecypod Arctica islandica. According to the log of the Santerno
section published by Cremonini et al. (1969), the level at which
Ruggieri (1957) reported that event is 1,510 m above the base of
the Pliocene, but subsequently Padovani and Tampieri (1970)
showed the presence of this index fossil as low as 1,460 m.
Actually, in the deep-water sediments of the Santerno River
section, the shallow-water pelecypod A. islandica must have
been reworked, and its appearance would have been delayed
with respect to its first appearance in shallow water, such as in
the Stirone section in western Emilia (Rio et al., Chapter 5,
Figure 5.5, this volume). In fact, in the Stirone section, the first
occurrence of A. islandica predates the first occurrence of
Gephyrocapsa oceanica s.L, whereas in the Santerno River
section the first occurrence of A. islandica is recorded about 150
m above the first occurrence of G. oceanica s. I.
Rio et al. (Chapter 5, this volume) have reported the
appearance of G. oceanica at around 1,300 m in the Santerno
River logged section. According to M. L. Colalongo (unpublished data), the first occurrence of Globigerina cariacoensis and
the last occurrence of Globigerinoides obliquus extremus are at
about 1,440 m. These same three events are recorded close
together, although in a different order, in the Vrica section just
above the top of level e, the adopted definition for the PliocenePleistocene boundary. Accordingly, the Pliocene-Pleistocene
boundary of Vrica correlates in the Santerno River section
somewhere around 1,300 m in the columnar log of Cremonini et
al. (1969).
Magnetostratigraphy of the Santerno River section. The earliest
analysis of the magnetostratigraphy of the Pliocene-Lower
Plio-Pleistocene of Italy
CHRONOSTRATI
SERIES
EPOCH
STAGE
AGE
B I O S T R A T I G R A P H Y
GRAPHY
PLANKTONIC
NANNOFOSSILS
POLARITY
TIME SCALE
SUBTAGE M a
Z.
BI OZONE
SZ.
BIOEVENT
ZONES OF
ZONAL BOUNDARY
BIOEVENT
ZONAL BOUNDARY
RUGGIERI AND
BIOEVENT
SPROVIERI
1976
GLOBOROTALIA
TRUNCATULINOIDES
EXCELSA
end Acme Smalt
SMALL
GEPHYROCAPSA
FORAMINIFERA
BIOZONE
SYRACOSPHAERA
PULCHRA
ZONE E
6pp.
kbtginrung Acme.
Small Gzphyfiocap6a
6pp.
HELICOSPHAERA
SELLII
-1.5
143
>5.5 fun
Gzpf
6pp.
G.
G.
exceJUa
exceJL6a
GLOBIGERINA
ZONE D
HyaJUnea
balXica
CARIACOENSIS
ZONE C
~J C.
C.MACINTYREI
jG.oczaru.ca
6.1.
f G. COAAJX.C0 <M>U>
kfictica
CRENALITHUS
DORONICOIDES
GLOBOROTALIA
INFLATA
J G.inileuta
DISCOASTER
BROUWERI
GLOBOROTALIA
CRASSAFORMIS
DISCOASTER
PENTARADIATUS
Figure 11.2. Chronostratigraphy and biostratigraphy adopted in this chapter.
Pleistocene portion of the Santerno River section was presented

in several papers by Nakagawa and co-workers; see Nakagawa et
al. (1975) and the sources cited therein. Kukla, Collins, and
Bender (1979) proposed a reinterpretation of those data, but
unfortunately the two teams had measured the total thickness of
the section differently, and both sets of measurements differed
from those reported by Cremonini et al. (1969), to which our
biostratigraphic data are referred.
In Figure 11.9 we show the biostratigraphic data as interpolated between fixed points in the different versions of the log of
the section (i.e., the first local occurrence of A. islandica, and
the chronostratigraphic boundaries). Although this implies that
the positions of the biostratigraphic events are rather approximate, it does not affect our discussion of the magnetostratigraphic interpretations.
The previous interpretations proposed by Nakagawa et al.
(1975) and by Kukla et al. (1979) are compared in Figure 11.9.
Apart from thickness problems, there are other discrepancies.
For instance, Kukla et al. (1979) do not show the many short,
normal-polarity events recorded by Nakagawa et al. (1975) in the
reversed interval below SAN 1 and above SAN 2. This may, of

course, be due to better cleaning to remove outcrop remagnetization, or it may indicate overlooked data.
Correlation of the paleomagnetism of this section to the
standard paleomagnetic time scale is particularly difficult, as
testified by conflicting opinions among workers (cf. Nakagawa et
al., 1975; Kukla et al., 1979; Arias et al., 1980). Though we are
unable to present here a definitive interpretation, we shall
discuss the published interpretations in the light of the constraints suggested by biostratigraphy.
To begin with, the upper normal-polarity magnetozone,
named SAN 1 by Kukla et al. (1979), was correlated to the
Brunhes chron by Nakagawa et al. (1975) and to the Olduvai
subchron by Kukla et al. (1979). Biostratigraphic evidence
indicates than neither of those interpretations is tenable.
At the top of the Santerno River section, a vertebrate fauna
assigned to the late Villafranchian Farneta unit has been
recovered (Azzaroli and Berzi, 1970). This faunal unit is older
than 0.8 Ma (Arias et al., 1980), which conflicts with possible
correlation of the normal magnetozone SAN 1 to Brunhes (less
Azzaroli et al.
144
CHRONO
STRATIGRAPHY
BIO
STRATIGRAPHY
BIOSTRATIGRAPHIC
INFORMATIONS
2500
8
<
o
\
- 2000
GO
1 H.
>5.5 ym
Gzphyn.oca.p6OL 6pp.
1500
_L ~T" G. o btiquu6
G.
avUa.coe.n6AJ> f G.oczanica
G.in^laXa
[_
MIOC, MESS
6.1.
suggest a correlation between the SAN 2 magnetozone of

Santerno River and the (upper) N3 magnetozone of the Vrica
section. Above this level, a normal-polarity interval close to the
first local appearance of Arctica islandica (Figure 11.9) was
correlated to the Olduvai subchron by Nakagawa et al. (1975)
and to the Reunion subchron by Kukla et al. (1979). Biostratigraphy indicates that both correlations are doubtful. In particular,
we note that the first occurrence of G. oceanica s.l., which has
been observed above the top of the Olduvai (Rio et al., Chapter
5, this volume), is seen below the normal-polarity interval
referred to in the Santerno River section. Furthermore, the first
occurrence of Globigerina cariacoensis and the last occurrence of
Globigerinoides obliquus extremus, also below the normalpolarity interval under discussion, are above the N1-N2 magnetozones of the Vrica section (Figure 11.9), which in turn have
been definitively correlated to the Olduvai subchron (Tauxe et
al., 1983; Pasini and Colalongo, Chapter 2, this volume).
Strangely, no long, normal-polarity interval correlatable to the
Olduvai subchron has been identified below the first occurrence
of G. oceanica s.l. in the Santerno section.
The lower part of the Santerno section, magnetozone SAN 3
of Kukla et al. (1979), shows normal paleomagnetic polarity and
has been correlated to the Gauss chron. This correlation is in
agreement with our biostratigraphic data. It is interesting that
Nakagawa et al. (1975) reported a reverse-polarity interval at the
base of the Santerno River section, which should be correlated to
the uppermost Gilbert chron if the section is continuous. I. Raffi
(unpublished data) recorded the first occurrence of Pseudoemiliana lacunosa a few meters above the base of the Pliocene in
the Santerno River section, in the sampling of Colalongo et al.
(1974); that event was calibrated to the top of the Gilbert by Rio
(1982) and to the Gilbert-Gauss boundary by Poore et al.
(1983).
P.lacuno6a
West Santerno section. Colalongo et al. (1982b) studied this

section
in the watershed between the Santerno and Sellustra
|
| MARLY CLAY
valleys.
The main stratigraphic features are reported in Cola|; :^:| SAND
longo et al. (1982b). The Pliocene-Pleistocene boundary can be
placed by correlation with the Pliocene-Pleistocene boundaryFigure 11.3. The Santerno River section.
stratotype in the Vrica section by means of planktic foraminifera
bioevents. In fact, the first occurrence of Globigerina cariacoensis, the last occurrence of Globigerinoides obliquus
than 0.78 Ma). On the other hand, below the SAN 1 magextremus, and the increase in abundance of left-coiling Neonetozone the successive appearances of Gephyrocapsa oceanica
globoquadrina pachyderma have been recorded close to each
s.l. and "large" (more than 5.5 /xm in diameter) Gephyrocapsa
other and in the same relative order as in the Vrica section,
spp. have been detected by Raffi and Rio (1980a); both of those
where those events are close to the level adopted as the
bioevents are known to have occurred in the late Matuyama,
definition of the Pliocene-Pleistocene boundary.
above the Olduvai and below the Jaramillo (Rio et al., Chapter
5, this volume). Taken together, these findings strongly conflict
with the correlation of magnetozone SAN 1 to either the Olduvai CastelVArquato section. In the Val d'Arda hills, near the towns of
or the Brunhes; the most probable correlation is, by default, the Vernasca, Lugagnano, and Castell'Arquato (Piacenza province),
Jaramillo (ca. 1.0 Ma). The great thickness of the SAN 1 normal- a thick sequence of Pliocene sediments is exposed that includes
polarity zone suggests the possibility, however, of secondary the Piacenzian (Upper Pliocene) stratotype. Pleistocene sedimagnetization in this interval of the Santerno River section.
ments crop out along the Riorzo and Bertacca streams west of
The biostratigraphic correlations reported in Figure 11.9 Castell'Arquato. The complete section (Figure 11.4) has been
LIMESTONE
CHRONOSTRA
TIGRAPHY
>-
<
a: cr
BIOSTRATIGRAPHIC
LJ
O
u_
- - H-
ITHO LOG
fAGE
OR AM
IOST
CO
U - CQ
INFORMATIONS
CO
z:
SELINUN
TIAN
CARIACO.
ERIES
on
CO
>LEISTO
CENE
1
>
<
ETER
CD
- 700 4
<
'
INF
- 600
CD
LU
<_>
<c
LU
Li.
<
CO
CO
<
WJ
M H
i-
Q_
- 500
RATC
DRMIS
<c
'
'
a! co
'
tr>
LU
- 400
CD
C_>
'
300
<
Q-
^^
<
_j
o
H
O
z>
Qu
<c
CD
LU
|
- 200
TAI
<->
<C
a:
<
CD
a:
<
- 100
z:
145
according to Colalongo et al. (1974). The correlations among

these components have been locally difficult and may be
incorrect.
Mollusca are very abundant in the outcrops on the left slope of
the Arda valley (Monte Giogo, Monte Padova, Monte Falcone,
Rio Cravalese), and corals, bryozoa, brachiopods, and echinoderms are also present. Benthic foraminifera are abundant
throughout the whole section, while ostracodes and planktic
foraminifera are much scarcer. Nannofossils are abundant, but
mainly have been reworked (Barbieri and Rio, 1974; Rio et al.,
1988). All things considered, it is difficult to obtain a detailed
and reliable biostratigraphy based on calcareous microplankton.
Different biostratigraphic zonations of the Castell'Arquato
section have been published, based on the mollusks (Padovani
and Tampieri, 1974), ostracodes (Colalongo and Russo, 1974),
and benthic and planktic foraminifera (Barbieri, 1967; Colalongo et al., 1974). It is possible, nevertheless, that all those
biostratigraphic efforts failed to identify hiatuses and correlation
errors due to the desultory presence of the main index fossils
(Colalongo et al., 1974).
According to Colalongo et al. (1974), the Pliocene-Pleistocene
boundary in the Castell'Arquato section (i.e., the upper
boundary of the Piacenzian Stage) is drawn in coincidence with
the first occurrence of the "northern guest" Arctica islandica.
As shown by Rio et al. (Chapter 5, this volume), this level
seems to be at least roughly correctable to the boundarystratotype of the Pliocene-Pleistocene boundary in the Vrica
section. A major unconformity, however, truncates the Piacenzian stratotype, below the shallow-marine deposits with
Pleistocene fauna. According to Rio et al. (1994), this gap,
which could be considered as an undefined Upper Pliocene
chronostratigraphic interval, is represented by strata exposed at
Monte San Nicola in southern Sicily that they have nominated
as the stratotype of the new Gelasian Stage.
vD
- 0
L ___ | SAND
ET^Z] CLAY AND

l_=^J SAND
|I
H | SILTY
SILTY CLAY
CLAY
Central Italy
FÎ
CONGLOMERATES
CALCARENITES
Figure 11.4. The Castell'Arquato section.
studied by several authors (Albertelli and Mazzei, 1963; Dondi,

1963; Barbieri, 1967, 1971; Colalongo et al., 1974; Rio et al.,
1988).
The lithological sequence of the Castell'Arquato section,
shown in stratigraphic order, is as follows:
sands (100 m)
calcarenites with interbedded silty clays (180 m)
clays and sands (140 m)
silty clays, with a basal conglomerate (ca. 320 m)
As a matter of fact, the Castell'Arquato section is a composite
of different subsections: 9 according to Barbieri (1967), and 10
Vallebiaja section. Named for a farm on the hills southwest of

Fauglia, Livorno district, this site is mentioned only because it is
one of the typical sections of the Calabrian, according to
Gignoux (1913, p. 300). Actually, Vallebiaja is not a section, but
only a richly fossiliferous outcrop with a molluscan assemblage of
Santernian age. As it happens, that particular outcrop no longer
exists, and exposures in the area are so poor that the section may
be considered completely unusable for biostratigraphic analysis.
Monte Mario section. The section at Monte Mario, on the right
bank of the Trevere River in the city of Rome, is also one of the
typical sections of the Calabrian, according to Gignoux (1913, p.
286). This section consisted of scattered exposures of fossiliferous sands belonging to the earliest Pleistocene, with the
best outcrop at the locality called "La Farnesina." At the present
time, the entire area has been built over because of the
expansion of the city, and none of the old sites are accessible.
146
Azzaroli et al.
Musone section. The Musone is a small river near the village of

Recanati, not far from Ancona. The section was synthesized by
Selli (1954, p. 242), on the basis of a few outcrops, as a late
Pliocene to early Pleistocene sequence. Even at the time of
Selli's research, exposures in the area were generally poor, and
in recent years several of the outcrops he described have been
buried or destroyed.
CHRONO
STRATIGRAPHY
BIOSTRATI
GRAPHY
BIOSTRATIGRAPHIC
INFORMATIONS
>5.5 fan Ge.phyioaa.pia ipp.

HyaJLinza baLUca
Southern Italy
Monasterace section. This section crops out along the southern
slope of Monte Rosito, a few kilometers north of Monasterace
village on the Ionian side of Calabria, near Punta Stilo. It is
noteworthy as one of the sections of the Calabrian described by
Gignoux (1913, p. 48), but the local Plio-Pleistocene sequence is
poorly exposed.
In the middle part of this section, where there are some scanty
outcrops, about 20 m of blue marls can be sampled. From the
presence of Globorotalia inflata, Globorotalia truncatulinoides
truncatulinoides, Globorotalia tosaensis, and Discoaster brouweri
these strata must be ascribed to the uppermost Pliocene. Greco,
Ruggieri, and Sprovieri (1974) erroneously referred these marls
to the Pleistocene because of the presence of G. truncatulinoides,
which at that time was considered to have been restricted to the
Quaternary.
Transgressively above the Upper Pliocene blue marls lies a
sequence, about 100 m thick, that begins in a fossiliferous
conglomerate with Arctica islandica, followed by interbedded
sands and sandy clays, in which G. truncatulinoides excelsa is
present. This transgressive sequence must be ascribed to a
separate sedimentary cycle of late Emilian-Sicilian age (Figure
11.2), in a geologic and stratigraphic setting very similar to that
in the Santa Maria di Catanzaro area (Sprovieri, D'Agostino,
and Di Stefano, 1973). This being the case, we may conclude that
the Monasterace section, even though indicated by Gignoux
(1913) as one of the typical examples of the Calabrian, represents
only the uppermost part of the Lower Pleistocene.
_J_ G. ocia/Uca
i.l.
_1_ G. pachydzAma. "le.(
J_
P. ICULI.
iti:
~V
-\ SEMIENDURATED CHALK
SILTY MARL WITH
Fe - Mg BROWN HORIZONS
Capo Rossello section. One of the most spectacular Upper

Miocene to Lower Pleistocene sections in the Mediterranean
crops out at Capo Rossello, along the sea-cliff westward from
Agrigento (Sicily). From bottom to top, the section (Figure 11.5)
consists of white pelagic marly limestones (Trubi Formation),
marls, marly clays, sandy clays (Monte Narbone Formation), and
a thick turbiditic calcarenite bed (Agrigento Formation). In the
classic Capo Rossello section, as described by Cita and Gartner
(1973), there is a sedimentary hiatus at the Trubi-Monte
Narbone transition. The missing part is well exposed nearby,
however, at the locality of Punta Piccola (Rio et al., 1984a).
The sediments of this section, about 360 m thick, are bathyal
in the lower part to neritic in the upper part and are very rich in
microfossils, mainly foraminifera and calcareous nannofossils.
The Capo Rossello section has been extensively studied in recent
years by Cita and Gartner (1973), Cita and Decima (1975),
Figure 11.5. The Capo Rossello section.
Gartner (1977b), Sprovieri (1978), and Rio et al. (1984a), among

others. Initially, Cita and Decima (1975) placed the PliocenePleistocene boundary just below the base of a calcarenite marker
where the lowest (resedimented) specimens of Arctica islandica
occur in the section, at a level containing the first appearance of
Hyalinea baltica. Subsequently, Sprovieri (1978) recognized the
Pliocene-Pleistocene boundary by correlation with the Vrica
section, on the basis of the first appearance of Globigerina
cariacoensis and of the beginning of abundant left-coiling
specimens of Globoquadrino pachyderma, about 80 m below the
calcarenite bed. In addition, Raffi and Rio (1980a) and Rio et al.
(1984a) noted the first occurrence of Gephyrocapsa oceanica s.l.
about 65 m below that marker. In conclusion, the Capo Rossello
CHRONOSTRATI
GRAFHY
BIOSTRATI
GRAPHY
BIOSTRATIGRAPHIC
INFORMATIONS
SZ
NOT
ZONED
_- 15
_J_ G. ViunccucuJU.noidzA
-MARKER BED,,
H.baUlca
>5.5 /
cap*a 6pp
I
| : :;|
CLAY
[^jjr]
DIATOMITE
SAND
^ ^
HIATUS
G. obtiquuA
Figure 11.6. The Le Castella section.
section displays many very interesting features: (1) It represents

an exceptionally long time interval, from the base of the Pliocene
to the early Pleistocene, with continuous sedimentation in facies
particularly suitable for paleontological studies. (2) It is directly
correlatable with the Vrica section in the segment straddling the
Pliocene-Pleistocene boundary. (3) According to preliminary
studies, it seems to be suitable for paleomagnetic investigations.
147
Colalongo et al. (1981, p. 105). Unfortunately, many other

workers who have studied the Le Castella area have included
those parts of the section, extending for several tens of meters
below and above the "marker bed," that are distorted and
duplicated by faults and outcrop slumps.
Earlier, the Le Castella section was selected by Emiliani et al.
(1961) for studying the paleoclimatic variations near the
Pliocene-Pleistocene boundary, located by those authors at the
base of the "marker bed" and practically in coincidence with the
first appearance in that section of the "cold guest" Hyalinea
baltica. The Le Castella section was subsequently proposed (at
the VII INQUA Congress, Denver, 1966) as a suitable section
for representing the Pliocene-Pleistocene boundary-stratotype
and was accepted as such by numerous authors (e.g., Rio, 1974;
Haq, Berggren, and Van Couvering, 1977).
In recent years, however, the consensus has grown that the
"marker bed" of Le Castella is unsuitable as a physical stratotype
of the Pliocene-Pleistocene boundary (Pelosio, Raffi, and Rio,
1980; Colalongo et al., 1981, 1982b). The main reasons are (1)
that the part of the section in the transition from Pliocene to
Pleistocene is too short, (2) that the first local occurrence of
Hyalinea baltica, which justifies the "marker bed" definition, is
not the best criterion for recognizing the basal Pleistocene in
Italy (Rio et al., Chapter 5, this volume), and (3) that
biostratigraphic analyses (Raffi and Rio, 1980b; Colalongo et al.,
1981) indicate the presence of a hiatus just below the "marker
bed," as shown by a coalescence of biostratigraphic events that
normally are widely spaced.
These findings indicate an Emilian age, probably late Emilian,
for the "marker bed," in the chronostratigraphic subdivision of
Ruggieri et al. (1984). In particular, the appearance of
Globorotalia truncatulinoides excelsa 11 m above the "marker
bed" (Colalongo et al., 1981) is indicative of the base of the
Sicilian substage (Ruggieri and Sprovieri, 1977; Ruggieri et al.,
1984).
Magnetostratigraphy of the Le Castella section. Nakagawa and

co-workers (1975; Chapter 3, this volume) have reported on the
magnetostratigraphy of a Le Castella section more than 300 m
thick, sampled along a track where several long intervals are
covered by debris and where faults have been recognized. As
noted earlier, the Le Castella section is reliably continuous only
he Castella section. The Le Castella section (Figure 11.6) crops over a short interval below and above the "marker bed" (Figure
out in the Crotone sedimentary basin in south-central Calabria, 11.6), and we therefore consider only the magnetostratigraphy of
about 20 km southwest of the city of Crotone. It is about 30 m this interval (Figure 11.9). Biostratigraphic correlations (Figure
thick, consisting of a clayey sequence, with some diatomaceous 11.9) indicate that the normal-polarity interval (here named C2)
interbeds and two sandy beds, the thicker of which is well known below the "marker bed," in the interval above the last
in the literature as the "marker bed" (Emiliani, Mayeda, and occurrence of Discoaster brouweri and below the first occurrence
Selli, 1961). Selli et al. (1977, p. 198) studied the part of the of Gephyrocapsa oceanica s. I., can be correlated to the Olduvai
section "corresponding to the slope located near Telegrafo, subchron and to magnetozone N2-N3 of the Vrica section
comprising the 11 m under the sand marker bed and the 17 m (Tauxe et al., 1983). The normal-polarity interval above the
above it. . . . As we get off this slope the exposure of the section "marker bed" (here named Cl) can be correlated to Vrica
becomes discontinuous, being frequently interrupted by several magnetozone N3 on biostratigraphic evidence, as shown in
slumps and big blocks fallen from the overlying calcarenites; the Figure 11.9. The short N3 magnetozone, as discussed by Tauxe et
area also presents some faults." For further discussion, see al. (1983) and Rio et al. (Chapter 5, this volume), cannot be
Azzaroli et al.
148
Station
ScrraMv/a
SE
CHRONOSTRATI
GRAPHY
STRATIGRAPHIC
Fio. 5. Coupe des collines de S. Maria di Catanzaro

(Echelle des hauteurs ct des longueurs, environ i/a.~>.ooo)
3. Graviers et sables tres fossilifercs
I
2. Sables argileux et argiles
I
.
G. - Bane de gres calcaire tres fossilifere a [ p J l o C y n e supencur
Cyprina islandica.
\
i. Argiles plastiques
Pliocene ancien.
INFORMATIONS
LU
LU
UJ
h-
Figure 11.7. Stratigraphic cross section of the hills south of Santa Maria di Catanzaro, according to Gignoux (1913). In this drawing, the
"Pliocene superieur" is the equivalent of the Calabrian. (From
Gignoux, 1913, p. 22.)
correlated to the Jaramillo subchron and might better be

correlated to one of the short normal-polarity intervals recognized between the Olduvai and the Jaramillo by Cooke (1981),
Mitsunashi et al. (1982), Nakagawa et al. (1982), and Suzuki and
Manabe (1982).
"G" BED
Santa Maria di Catanzaro section. In the neighborhood of Santa

Maria di Catanzaro, Calabria, the Plio-Pleistocene sequence is
subdivided into a lower clayey unit (A) and an upper sandy unit
(B). The upper unit includes a distinctive yellowish, macrofossilrich calcarenite bed termed the G-G' horizon by Gignoux (1913)
4 G.
(Figures 11.7 and 11.8).
Unit A is securely dated to the Pliocene. The "northern
guests" Arctica islandica and Hyalinea baltica occur from the
A.ûlandLcca, G.zznoJULuA and
H.battlca fanom the. bate. o&
very base of unit B (Sprovieri et al., 1973), indicating that all of
xho. 6<iczion
this unit should be ascribed to the Pleistocene (Figure 11.2). The
two units have an unconformable contact described either as a
COVERED
: : ;,.-| FINE SAND
CALCARENITE
transgressive surface (Sprovieri et al., 1973) or as a set of faults
(Pasini, Selli, and Colalango, 1977). Unit A is not represented,
~ - I | SANDY CLAY
COARSE SAND
however, in the classic "Santa Maria di Catanzaro section" of
Gignoux (1913), which crops out along the road from the village Figure 11.8. The Santa Maria di Catanzaro section.
of Santa Maria di Catanzaro up to the Colle Santa Maria (Pasini
et al., 1977). The traverse crosses unit B and the G-G' horizon,
but nevertheless this section (Figure 11.7), like every other authors since Emiliani et al. (1961) to define the base of the
section in the vicinity, presents unfavorable features, including Calabrian at Catanzaro according to the intentions of Gignoux
unexposed intervals, possible hiatuses, and tectonic distur- (1913), is actually younger than the base of the Sicilian Stage, as
bances. A synthesis is possible only through correlating a great marked by the first appearance of G. truncatulinoides excelsa.
many separate segments, with undue chances of error.
Consequently, the Calabrian Stage, at least in the restricted
As noted, Arctica islandica and Hyalinea baltica are present sense of the Catanzaro "neostratotype" with the G-G' bed at the
from the base of the "classic" section. In that section, Ruggieri base, as proposed by Selli (1971), is the same age as part of the
and Sprovieri (1977) reported that Globorotalia truncatulinoides Sicilian Stage, previously erected in 1872 by Doederlein. The
excelsa occurs at least 50 m below the G-G' bed, and on the basis term "Calabrian" under that definition is therefore equivalent to
of the biostratigraphic scheme in Figure 11.2, that leads to the earlier term for this chronostratigraphic interval in Italy and
several conclusions. First, the whole "classic" Santa Maria di should be regarded as invalid. The question of replacing
Catanzaro section is not only Pleistocene in age but also younger "Calabrian" as the name for the lowermost Pleistocene levels,
than the lowermost part of zone D of Ruggieri et al. (1976). This which are not defined by the Santa Maria di Catanzaro
is confirmed by the presence, from the base of the section, of neostratotype, or of amending the present definition is currently
Bolivinita quadrilatera and Coryphostoma karrerianum, which in being debated (Rio, Sprovieri, and Thunell, 1991; Van
Italy always appear above the first appearance of H. baltica. Couvering, Preface, this volume).
Second, the G-G' bed, which has been considered by most
Although Nakagawa and co-workers observed normal and
reversed magnetozones in the Santa Maria di Catanzaro section,

in work reviewed by Nakagawa (1981), Watkins et al. (1974)
were unable to identify any variations in depositional polarity in
samples from the same section, which appeared all normal. The
latter authors demonstrated the presence of a precipitate phase,
suggesting a secondary overprint, and as a result of this
uncertainty we did not show magnetostratigraphy of this section
in the correlation chart of Figure 11.9.
149
end of the Globorotalia puncticulata biozone in marine micropaleontology (De Giuli et al., 1983). Lindsay et al. (1980)
estimated an age for this level at about 3 Ma, based on
correlations to a short, normal-polarity paleomagnetic event
between the Kaena and Mammoth reversed chrons in the Gauss
epoch.
Montopoli event. The first major dispersal event took place

during the early Villafranchian, between the Triversa and
Vrica section. This section (Figure 11.9), about 4 km south of Montopoli faunal units. It was characterized by the disappearCrotone, is ideally suited for the Pliocene-Pleistocene boundary- ance of a warm-forest assemblage (Mammut borsoni, Tapirus
stratotype and has been approved as such by the International arvernensis, Ursus minimus, Sus minor) and its replacement by
Commission on Stratigraphy of the IUGS. The reader is referred an open-forest or bushland assemblage with the first appearances
to Aguirre and Pasini (1985) and to Chapters 2-5 in this volume of Equus, Gazella, and Mammuthus. The Montopoli unit falls
within the time interval of the Globorotalia crassaformis plankfor details.
tonic foraminifera zone (De Giuli et al., 1983). Lindsay et al.
(1980) dated the Montopoli-type fauna to the beginning of the
Conclusion: the Italian marine record
Matuyama paleomagnetic epoch, slightly younger than 2.6 Ma.
Most of the Italian sequences reviewed in this chapter are That date finds support in a similar faunal assemblage at the
unsuitable for defining the Pliocene-Pleistocene boundary- Rincon 1 site, in the Jucar Basin of eastern Spain, which Alberdi
stratotype, for a variety of reasons: (1) because their exposures et al. (1982) have dated to 2.5 Ma. The arrival of elephants in
are of poor quality (Vallebiaja, Monte Mario, Musone, and Italy, however, was somewhat earlier than in the type Montopoli
Santa Maria di Catanzaro); (2) because they do not extend into fauna in the lower Valdarno of Tuscany, as evidenced by an
the Pliocene (Vallebiaja and Monte Mario); (3) because they occurrence in underlying strata at Laiatico.
lack the lowermost Pleistocene (Monasterace, Le Castella, and
The first dispersal event coincided with a marked cooling
Santa Maria di Catanzaro); (4) because they consist of several recorded in marine sequences at about 2.4 Ma, the most severe
different component-sections of unreliable continuity and con- cooling of the Pliocene, in a series of steadily intensifying climate
tain a poor planktic assemblage (CastelFArquato). Only four of cycles leading up to the Pleistocene conditions (Thunell and
the reviewed sections (Vrica, Santerno River, West Santerno, Williams, 1983; Rio et al., 1991).
and Capo Rossello) are sufficiently well exposed, fossiliferous,
A second dispersal event, at the beginning of the Saint-Vallier
and continuous to be considered for the Pliocene-Pleistocene unit, is less well marked. In Italy, that event is poorly
boundary-stratotype. The northern Italian Santerno sections, represented because it coincided with the Aquatraversan phase
however, are highly terrigenous, and calcareous plankton (in of sea-level lowering (Ambrosetti et al., 1972), a time when the
particular, the nannofossils) are much less well represented than Italian peninsula was generally subjected to erosion. On the
in the southern Italian sections.
other hand, richer faunas have been described at Saint-Vallier
and Chilhac in France and at La Puebla del Valverde in Spain.
The Tegelen fauna of The Netherlands can also be ascribed to
Continental record
this unit, as can the older of the two faunas from the rich site of
Seneze (Aguirre et al., Chapter 9, this volume). The transition
Villafranchian mammal stratigraphy and dispersal
from the Montopoli to the Saint-Vallier faunal unit is not sharp,
events
but is marked by the replacements of Dicerorhinus jeanvireti,
Detailed correlations of the typical Villafranchian faunas of Italy Mammuthus gromovi, and Equus livenzovensis by Dicerorhinus
to European and Asian mammal faunas and to marine and etruscus, primitive Mammuthus meridionalis, and Equus stenopaleomagnetic stratigraphy have been presented by Azzaroli nis, respectively, in what may have been evolutionary succes(1977, 1983) and Azzaroli et al. (1982, 1986) (Figure 11.10). The sions. Other characteristic forms in the new fauna were Cervus
Villafranchian has been subdivided into six faunal units, charac- rhenanus Dubois, 1905 (= C. philisi Schaub, 1941), Eucladoterized from oldest to youngest by the sites at Triversa, ceros teguliensis (Dubois), 1905 (= E. senezensis Deperet and
Montopoli, Saint-Vallier, Olivola, Tasso, and Farneta, altogether Mayet, 1910), and sporadic early appearances of Sus strozzii.
representing the time from 3.0 Ma to approximately 1.0 Ma. Two The Saint-Vallier unit also contains the final forms of Gazella,
major faunal turnovers - the "dispersal events" of Lindsay et al. the cervid Croizetoceros, and the canid Nyctereutes in European
(1980) - occurred at the beginning of the Montopoli unit and at faunas. The absence of Hipparion, last known from Etouaires in
the end of the Villafranchian. Less dramatic but still significant the later Montopoli fauna, is also notable.
faunal changes distinguish other unit boundaries.
In the Tegelen clays of The Netherlands, pollen complexes
The Triversa unit, at the base of the Villafranchian, can be indicating a generally warm and equable climate, with lowdated to the transition from early to middle Pliocene time, at the amplitude climatic fluctuations, span the early Matuyama from
GEOMAGNETIC POLARITY
TIME SCALE
(Mankmen & Dalrymple 1979)
LE CASTELLA
GEPHYROCAPSA FO
MARKER BED
55
MACINTYREI LO
PROPOSED P/P
BOUNDARY DEFINITION
(+)
BROUWERI * D" TRIRADIA TUS" L 0
Globorotalia truncatulinoides excelsa FO
Hyaline a bait tea FO
Globigerina
Globigerinoides
3fc Arctica
MAGNETIC
cariacoensis FO
obliquus
ex tremus
islandica FLO (First
LO
Local Occurrence)
POLARITY
normal
reversed
intermediate
undefined
\-\-) maximum age for G crassaformis gr. FO
Figure 11.9. Biostratigraphic correlations for the Vrica section, the Le

Castella section, and the Santerno section. FO, first occurrence; LO,
last occurrence. The paleomagnetic logs are as follows: 1, Tauxe et al.
(1983); 2, Nakagawa et al. (Chapter 3, this volume); 5, Nakagawa et al.

(1977); 6, Nakagawa et al. (1975); 7, Kukla et al. (1979).
151
u
*ine
M a m m a l s
Pollen
2
W.Europe
E. Europe
W.CuroDe
India
M A
tandar
<o
NE A s i a
Major
events
Erosionai
phases
Ostlan
UJ
V
ClOmtAAAA
Nomntanan
UJ o
TAjuupoLLan
Ranucclo
p ^
o5 O
o
o
P.t Gal.rU
Utnaplan
V)
Do^liara
Parncta
Avvicola
riaminian
o
O
Chu.Uoc.(i^-^.n.'. ""*
End V i l l a f r .
Cattian
TamanLan
Taaao
O.
EbuAoniA*
Ollvola
OdtAmOn
Wolf
mm ^m mm ^ a
X
o
O .
S.V.lli.r
w
Aullan
mm im mm mt
a.
4
Eleohant
Equua
Rcuvvuan
Trlvra
Acqufltraversan
Moldavian
w
Z
U
CO
Leptoooe
P L I 0
3-
about 2.2 Ma to the top of the Olduvai normal at 1.8 Ma (Sue and
Zagwijn, 1983; Zagwijn, Chapter 16, this volume). The maar
deposits at Seneze have a pollen spectrum with similar features
and also contain a mammal fauna that shares many characteristic
elements with the Saint-Vallier unit (e.g., Nyctereutes megamas-
Figure 11.10. Chronostratigraphy of

mammalian faunas, paleofloras, and
physiographic events in Eurasia (uncorrected time scale).
toides, Eucladoceros teguliensis, Cervus rhenanus, and Croizetoceros ramosus). The Seneze fauna is generally considered to be
younger than the fauna at Saint-Vallier (Meon et al., 1979), and in
fact it includes more progressive taxa, which nevertheless are
older than the comparative forms in the next (Olivola) unit. A
152
Azzaroli et al.
Farneta-like fauna is the younger fauna at Seneze noted earlier,

with Cervalces gallicus, Equus bressanus, Canis arnensis, and
possibly Equus stehlini. Megalovis latifrons may also belong to
this level, rather than to the older fauna in the maar deposits.
Fairly rich Upper Villafranchian faunas occur at several sites in
Olivola event. The transition to the Olivola unit is more sharply the Forest Bed series of East Anglia, beneath the Cromer Forest
marked, in a changeover termed the "wolf event" by Azzaroli Bed. Those sites have been divided into Pastonian and pre(1983). This is characterized by the appearance of the "wolf" Pastonian, but unfortunately it is impossible in virtually all
Canis etruscus, as well as Pachycrocuta brevirostris, Leptobos instances to know from which level the collected material was
etruscus, and cervids such as Dama nestii and Eucladoceros recovered (Stuart, 1982). At East Runton, the Forest Bed fauna
dicranios (including E. ctenoides), the latter two as evolutionary occurs alone, but at Bacton, Sidestrand, Overstrand, Mundlesey,
successors to Cervus rhenanus and Eucladoceros teguliensis, and possibly Pakefield the fossils are found together with
respectively. Scattered occurrences of the mastodont Anancus younger Cromerian (or, better, Galerian) material (Azzaroli,
arvernensis in the beds of the Montevarchi Group in the Upper 1953; Aguirre et al., Chapter 9, this volume). Dama nestii,
Valdarno sequence provide evidence that this typically Pliocene Eucladoceros dicranios, Equus stenonis, and Mimomys pliocaenispecies survived for a short time after the end of the Saint-Vallier cus indicate a general late Villafranchian age, while Mammuthus
time. The type specimen of Mimomys pliocaenicus (Major), meridionalis cromerensis, Eucladoceros tetraceros, Cervalces
1889, comes from this unit in the Upper Valdarno or (less gallicus, and Equus bressanus suggest latest Villafranchian (i.e.,
Tasso or Farneta) levels.
probably) from the overlying Tasso unit.
There has been no calibration of the Olivola in Tuscany itself,
despite several attempts to secure paleomagnetic or pollen data, End-Villafranchian event. The most dramatic dispersal event
but sediments that yield the Olivola-type fauna in northwestern occurred at the end of the Villafranchian, with the transition to
Tuscany are high-energy deposits indicative of cold, wet climatic the Galerian faunas of the early middle Pleistocene (Ambrosetti
conditions and lowered sea level in the Aullan erosional phase et al., 1972). That event, which affected not only Italy but all of
(Arias et al., 1980). The Aullan, in all likelihood, was coincidental Eurasia (Aguirre et al., Chapter 9, this volume), was the "endwith the Eburonian cold phase documented in The Netherlands Villafranchian event" of Azzaroli (1983). The faunal list com(Zagwijn, Chapter 16, this volume), which succeeded the Tiglian piled by Azzaroli et al. (1982) for Italy alone lists 17 extinctions
warm climate at the end of the Olduvai normal-polarity chron. and 17 new taxa at the beginning of the Galerian, with 4 lineages
The beginning of the Olivola faunal unit, which in classic terms that survived the transition undergoing marked evolutionary
marks the beginning of the Middle Villafranchian (Azzaroli, changes, and only 4 that survived into the Galerian, if only for a
1977), thus closely approaches the Pliocene-Pleistocene bound- brief time, with little or no change.
ary defined by the marine sequence at Vrica. De Giuli et al. (1983)
The extent of the end-Villafranchian turnover is clearly
proposed a correlation of the Olivola unit with the Globigerina apparent from these statistics, but from these changes other
cariacoensis biozone.
changes are also apparent. The fact most deserving of emphasis
is that the end-Villafranchian event was marked by new types of
Tasso and Farneta events. Succeeding units of the later Villa- adaptations unknown in earlier faunas. The development of
franchian are less sharply defined. The next younger Tasso unit is hypselodont (ever-growing) teeth in the microtine rodents
characterized by the following immigration events: Canis arnen- Microtus, Arvicola, and Pity mys may have been no more than
sis, a jackal or coyote; Canis falconeri, possibly a primitive the natural outcome of a trend that had already begun in the
"hunting dog" or lycaonid; "Leptobos" vallisarni, a somewhat Pliocene, but among the ungulates the appearance of giant
aberrant species of this genus with a massive skull; a poorly herbivores, such as the megacerid Cervalces latifrons and heavyknown ovibovine (Borselli et al., 1980); and Hippopotamus bodied species of Bison, Bos, Praeovibos, and Ovibos, repreantiquus. The caballine Equus stehlini in that fauna may have sents, in totality, an entirely new response to the environment.
evolved in situ from E. stenonis.
The end-Villafranchian turnover was not instantaneous, but it
The Farneta unit is even less well known. A distinguishing took place within a relatively short time span that in Europe
element is the highly derived elephant, Mammuthus meridionalis seems to have been confined to about 70 k.y., within or just
vestinus, in Italy, which has its close relatives M. m. tamanensis in above the Jaramillo subchron, between 1.0 and 0.9 Ma. It was
southern Russia and M. m. cromerensis in the East Anglian accompanied by major changes in the vegetation of the region
faunas of Great Britain. The local fauna at Imola, in northern from Europe to Japan (Grichuk, Chapter 8, this volume) and
Italy, is characterized by progressive subspecies of Mammuthus great increases in loess deposition in central Asia and the
Danube and Rhine basins. A major shift in stable-isotope values
and Eucladoceros associated with Hippopotamus.
There are two similar faunas in the Massif Central of France, in marine environments of this age indicates a sharp cooling at
one at the Creux de Peyrolles, near Perrier, with Eucladoceros approximately the same time (Shackleton and Opdyke, 1976;
tetraceros, Dicerorhinus etruscus, and the still poorly known Thunell and Williams, 1983; Shackleton et al., 1995), together
Cervus perolensis (Bout and Azzaroli, 1953). The other French with sea-level lowering and erosion evidenced in the Cassian
date between 2.0 and 1.8 Ma is the only plausible age for the fauna
of the Seneze maar deposits. It must be noted, again, that Bout
(1970) reported fossils from slopes above the maar deposit, which
accounts for the mixture of two distinct paleofaunas at this site.
event in the Tiber delta (Ambrosetti et al., 1972; Azzaroli,

1983).
153
Vernasca-Castell'Arquato including the Piacenzian stratotype (Piacenza Province). Soc. Ital. Sci. Nat., Mem.
15:145-163.
Barbieri, F. 1971. Piacentian. In Stratotypes of Mediterranean
Summary of the continental record
Giorn. Geol. 37, estratto, fasc. II.
In conclusion, the Villafranchian and Galerian continental Barbieri, R, and Rio, D. 1974. Calcareous nannoplankton from
the Piacenzian (Late Pliocene) of Western Emily. L'Ateneo
faunas in Italy show evidence of two major turnovers, the
Parmense, Acta Nat. 10:29-42.
elephant-Egwws event at about 2.5 Ma and the end-VillafranchBorselli, V, De Giuli, C , Ficcarelli, G., and Mazzini, M. 1980.
ian event at about 0.9 Ma, both of which were approximately
Casa Frata: una localita fossilifera del Villafranchiano
coeval with major climate and ocean-level minima. The
superiore presso Terranova Bracciolini (Arezzo) nel
Valdarno Superiore. Soc. Paleontol. Ital., Boll. 19:254Pliocene-Pleistocene boundary, coincident with the beginning of
258.
the Olivola unit in the Valdarno sequence, was also marked by
significant changes in the land-mammal faunas, but those Bout, P. 1970. Absolute ages of some volcanic formations in the
Auvergne and Velay areas and chronology of the European
changes were not accompanied by changes in the environment
Pleistocene. Palaeogeogr. Palaeoclimatol. Palaeoecol. 8:
comparable to the changes that accompanied the two major
95-106.
Bout, P., and Azzaroli, A. 1953. Stratigraphie et faune du Creux
events.
de Peyrolles (Puy-de-Dome). Ann. Paleontol. 38:37-56.
Brolsma, M., and Meulenkamp, J. E. 1973. Benthonic foraminiferal assemblages from the Calabrian deposits of
References
Santa Maria di Catanzaro. Newsl. Stratigr. 3:1-24.
Aguirre, E., and Pasini, G. 1985. The Pliocene-Pleistocene Cita, M. B., and Decima, A. 1975. Rossellian: proposal of
superstage for the marine Pliocene. In VI Congress of the
Alberdi, M. T , Arias, P., Bigazzi, G., Bonadonna, F. P., Leone,
Regional Committee for Mediterranean Neogene StratiG., Lopez, N., Michaux, J., Morales, J., Robles, R, and
graphy, Bratislava, 1975, Proceedings, vol. 1, ed. J. Senes
Soria, Y. D. 1982. Nuevo yacimiento de molluscos y
and J. Slavik, pp. 217-227. Bratislava: Akad. Vied.
vertebrados del Villafranquiense de la Cuenca del Jucar Cita, M. B., and Gartner, S., Jr. 1973. The stratotype Zanclean:
(Albacete, Espafia). In Actes de la Colloqium "Le
foraminiferal and nannofossil biostratigraphy. Riv. Ital.
Villafranchien mediterraneen," Lille, 9-10 Dec. 1982, vol.
Paleontol. Strat. 79:503-558.
1, pp. 255-271. Lille: Lille Univ., Fac. Sci. (CNRS).
Colalongo, M. L. 1968. Cenozone a foraminiferi e ostracodi nel
Albertelli, L., and Mazzei, R. 1963. Geologia del settore
Pliocene e basso Pleistocene della serie del Santerno e
Vogherese-piacentino. Geol. Soc. Ital, Boll. 81(4):20-44.
dell'Appennino romagnolo. Giorn. Geol. 35:29-61.
Ambrosetti, P., Azzaroli, A., Bonadonna, F. P., and Folieri, M. Colalongo, M. L., Elmi, C , and Sartoni, S. 1974. Stratotypes of
1972. A scheme of Pleistocene chronology for the
the Pliocene and Santerno River Section. Bur. Rech. Geol.
Tyrrhenian side of Central Italy. Geol. Soc. Ital, Boll.
Min., Mem. 78:604-624.
91:169-184.
Colalongo, M. L., and Pasini, G. 1980. La ostracofauna del
Arias, C , Azzaroli, A., Bigazzi, G., and Bonadonna, F. P. 1980.
Vrica in Calabria (con considerazioni sul limite Neogene/
Magnetostratigraphy and Pliocene-Pleistocene boundary
Quaternario). Soc. Paleontol. Ital, Boll. 19:44-126.
in Italy. Quaternary Res. 13:65-74.
Ruggieri, G., Sartoni, S., Selli, R., and Sprovieri, R.
Azzaroli, A. 1953. The deer of the Weybourn Crag and Forest
1982a. The Neogene/Quaternary boundary: a review and a
Bed of Norfolk. Brit. Mus. (Nat. Hist.), Bull. Geol. 2:196.
proposal. Geogr. Fis. Dinam. Quat. 5:59-68.
Azzaroli, A. 1977. The Villafranchian State in Italy and the Plio- Colalongo, M. L., Pasini, G., Raffi, I., Rio, D., Sartoni, S., and
Sprovieri, R. 1984. Biochronology of the Italian marine
Pleistocene Boundary. Giorn. Geol. 41:61-79.
Pliocene and Lower Pleistocene. In XXVII International
Azzaroli, A. 1983. Quaternary mammals and the end-VillaGeological Congress, Moscow. Proceedings, vol. 3, pp.
franchian dispersal event - a turning point in the history
109-127. Moscow: Nauka.
of Eurasia. Palaeogeogr. Palaeoclimatol. Palaeoecol. 44:
117-139.
the Neogene-Quaternary boundary and the Vrica Section.
Azzaroli, A., and Berzi, A. 1970. On an Upper Villafranchian
Soc. Paleontol. Ital, Boll. 20(2):99-120.
fauna at Imola Northern Italy, and its correlation with the
marine Pleistocene sequence of the Po Plain. Palaeontogr. Colalongo, M. L., Ricci Lucchi, R, Guarnieri, P., and Mancini,
E. 1982b. II Plio-Pleistocene del Santerno. In Guida alia
Ital. 66:1-12.
geologia del margine appenninico-padano, ed. G. CremoAzzaroli, A., De Giuli, C , Ficcarelli, G., and Torre, D. 1982.
nini and P. Ricci Lucchi, pp. 161-166. Bologna: Pitagora
Table of the stratigraphic distribution of terrestrial faunas
Tecnoprint.
in Italy from the Pliocene to the early Middle Pleistocene.
Geogr. Fis. Dinam. Quat. 5:55-58.
Colalongo, M. L., and Russo, A. 1974. Stratotypes of the
Azzaroli, A., De Giuli, C , Ficcarelli, G., and Torre, D. 1986.
Pliocene and Santerno River section, comparison and
Mammal succession of the Plio-Pleistocene in Italy. Soc.
correlations on the ostracofaunas. Bur. Rech. Geol Min.,
Geol. Ital., Mem. 31:213-218.
Mem. 78:595-601.
Azzaroli, A., and Napoleone, G. 1982. Magnetostratigraphic Cooke, H. B. S. 1981. Age control of Quaternary sedimentary
investigation of the Upper Sivaliks near Pinjor, India. Riv.
climatic record from deep boreholes in the Great HungarItal. Paleontol. Stratigr. 87:739-762.
ian Plain. In Quaternary Paleoclimate, ed. W. C. Mahaney,
Barbieri, F. 1967. The Foraminifera in the Pliocene section
pp. 1-12. Norwich: Geo Abstracts Ltd.
154
Azzaroli et al.
aries, ed. T. Saito and L. H. Burckle, pp. 64-70. New

Cremonini, G., Elmi, C , and Monesi, A. 1969. Osservazioni
York: Micropaleontology Press.
geologiche e sedimentologiche su alcune sezioni PlioPleistoceniche dell'Appennino romagnolo. Giorn. Geol. Nakagawa, H., Niitsuma, N., Mitsunashi, T., Oda, M.,
Takayama, T., and Sakai, T. 1982. Plio-Pleistocene
35:85-96.
stratigraphy in Boso Peninsula, central Japan. In The third
De Giuli, C , Ficcarelli, G., Mazza, P., and Torre, D. 1983.
Confronto tra successioni marine e continental! del Plioreport on the Pliocene-Pleistocene Boundary in Japan, ed.
M. Itihara and Y. Kuwano, pp. 81-86. Kyoto: Dogura and
cene e Pleistocene inferiore in Italia e nell'Area mediterranea. Soc. Paleontol. ItaL, Boll. 22:323-328.
Co.
Doederlein, P. 1872. Note illustrative della carta geologica del Padovani, A., and Tampieri, R. 1970. Ricerche sui Molluschi
Modenese e del Reggiano. Memoria no. 3. Roma: Servizo
Plio-Pleistocenici della Valle del Santerno (Appennino
romagnolo). Giorn. Geol. 35(l):301-309.
Geologica Italiana.
Dondi, L. 1963. Nota paleontologica stratigrafica sul pedeappen- Padovani, M. A., and Tampieri, R. 1974. Stratotypes du
Pliocene et coupes du fleuve Santerno confrontes sur la
ninopadano. Soc. Geol. ItaL, Boll. 81:113-229.
base de la Malacofaune. Bur. Rich. Geol. Min., Mem.
Emiliani, C , Mayeda, T., and Selli, R. 1961. Paleotemperature
78:647-654.
analysis of the Plio-Pleistocene section at Le Castella,
Calabria, Southern Italy. Geol. Soc. Am., Bull. 72:679- Pasini, G., and Selli, R. 1977. The section of Le Castella. Giorn.
Geol. 41(l-2):424-430.
688.
Francavilla, F. 1974. Donnees palynologiques et paleoclimato- Pasini, G., Selli, R., and Colalongo, M. L. 1977. The S. Maria di
Catanzaro Section. Giorn. Geol. 41(1-2):424-430.
logiques de la coupe du Santerno (Italie). Bur. Reck. Geol.
Pelosio, G., Raffi, S., and Rio, D. 1980. The PlioceneMin., Mem. 78:633-637.
Pleistocene boundary controversy. Status in 1979 at the
Gartner, S. 1977a. Calcareous nannofossil biostratigraphy and
light of International Stratigraphic Guide. In Volume
revised zonation of the Pleistocene. Mar. Micropal. 2:125.
dedicato a Sergio Venzo, ed. F. Petrucci and R. Cavazzini,
Gartner, S. 1977b. Calcareous nannofossil biostratigraphy of the
Monte Narbone Formation at the type-locality of the Poore, R. Z., Tauxe, L., Percival, S. F., Jr., LaBreque, J. L.,
Rossellian superstage. Riv. ItaL Paleontol. Strat. 83:179Wright, R., Peterson, N. P., Smith, C. C , Tucker, P., and
180.
Hsu, K. J. 1983. Late Cretaceous-Cenozoic magneGignoux, M. 1913. Les formations marines pliocenes et
tostratigraphy and biostratigraphy at South Atlantic
quaternaires de l'ltalie du Sud et de la Sicilie. Ann. Univ.
Ocean: DSDP Leg 73. Palaeogeogr. Palaeoclimatol.
Lyon, n.s. 1(36): 1-633.
Palaeoecol. 42:127-149.
Greco, A., Ruggieri, G., and Sprovieri, R. 1974. La sezione Raffi, I., and Rio, D. 1979. Calcareous nannofossil biostratigraphy of DSDP Site 132, Leg 13 (Tyrrhenian Seacalabriana di Monasterace (Calabria). Soc. Geol. ItaL,
Western Mediterranean). Riv. ItaL Paleontol. Strat. 85:
Boll. 93:151-179.
127-172.
Haq, B. U , Berggren, W. A., and Van Couvering, J. A. 1977.
Corrected age of the Pliocene-Pleistocene boundary. Raffi, I., and Rio, D. 1980a. Nannoplancton calcareo del
Nature 269:483-488.
Pliocene superiore-Pleistocene inferiore della serie del
Kukla, G., Collins, B. P., and Bender, M. L. 1979. Radiometric
flume Santerno. In Paleontologia, Stratigrafia, Evoluage of the Arctica islandica boundary in Italy: 2 my. Ann.
zione, vol. 1, ed. C. Petronio, pp. 187-195. Roma:
Geol. Pays Hellen., tome hors ser. 2:699-709.
Lindsay, E. H., Opdyke, N. D., and Johnson, N. M. 1980. Raffi, I., and Rio, D. 1980b. Nuove osservazioni sulla disPliocene dispersal of the Horse Equus and late Cenozoic
tribuzione del nannoplancton calcareo nella serie di Le
mammalian dispersal events. Nature 287:135-138.
Castella (Calabria). In Paleontologia, Stratigrafia, Evoluzione, vol. 1, ed. C. Petronio, pp. 181-186. Roma:
Lyell, C. 1833. Elements of Geology, vol. 2. London: John
Murray.
Mankinen, E. A., and Dalrymple, G. B. 1979. Revised Rio, D. 1974. Remarks on Late Pliocene-Early Pleistocene
geomagnetic polarity time scale for the interval 0-5 Ma
calcareous nannofossil stratigraphy in Italy. Ateneo
B.P. /. Geophys. Res. 84:615-626.
Parmense, Acta Nat. 10:409-449.
Meon, H., Ballesio, R., Guerin, C , and Mein, P. 1979. Rio, D. 1982. The fossil distribution of Coccolithophore Genus
Approche climatologique du Neogene superieur (TorGephyrocapsa Kamptner and related Plio-Pleistocene
tonien a Pleistocene moyen ancien) d'apres les faunes et
chronostratigraphic problems. In Initial Reports of the
les fiores d'Europe occidentale. Mus. Nat. Hist. Natur.,
Deep Sea Drilling Project, vol. 68, ed. W. L. Prell, J. V.
Mem. (B) 27:182-195.
Gardner, et al., pp. 325-343. Washington DC: U.S.
Mitsunashi, T., and Uonuma Hills Collaborative Research
Group. 1982. The Pliocene-Pleistocene boundary in the Rio, D., Sprovieri, R., and Di Stefano, E. 1994. The Gelasian
Niigata sedimentary basin. In The third report on the
Stage: a proposal of a new chronostratigraphic unit of the
PliocenePleistocene boundary in Japan, ed. M. Itihara
Pliocene Series. Riv. ItaL Paleontol. Strat. 100:103-124.
and Y. Kuwano, pp. 28-41. Kyoto: Dogura and Co.
Rio, D., Sprovieri, R., Di Stefano, E., and Raffi, I. 1984b.
Nakagawa, H. 1981. Neogene-Quaternary boundary and correlaGloborotalia truncatulinoides (d'Orbigny)in the Mediterration of Vrica section. In Field conference Neogene/
nean upper Pliocene geological record. Micropaleontology
Quaternary boundary, India, 1979. Proceedings, ed. M. V.
30:121-137.
A. Sastry et al., pp. 107-111. Calcutta: Geological Survey Rio, D., Sprovieri, R., and Raffi, I. 1984a. Calcareous plankton
of India.
biochronology of the Pliocene-Lower Pleistocene succesNakagawa, H., Niitsuma, N., Kimura, K., and Sakai, T. 1975.
sion of the Capo Rossello area, Sicily. Mar. Micropal.
Magnetic stratigraphy of Late Cenozoic Stages in Italy and
9:135-180.
their correlatives in Japan. In Late Neogene epoch bound- Rio, D., Sprovieri, R., Raffi, L, and Valleri, G. 1988.
155

Biostratigrafia e paleoecologia della sezione stratotipica
del Piacenziano. Soc. Paleontol. Ital, Boll 27:213-238.
Rio, D., Sprovieri, R., and Thunell, R. 1991. Pliocene-lower Shackleton, N. J., and Opdyke, N. D. 1976. Oxygen isotope and
paleomagnetic stratigraphy of Pacific core V28-239: late
Pleistocene chronostratigraphy: a re-evaluation of MediPliocene to late Pleistocene. In Investigations of Late
terranean type sections. Geol Soc. Amer., Bull. 103:
Quaternary paleoceanography and paleoclimatology, Mem1049-1058.
oir 145, ed. R. M. Cline and J. D. Hays, pp. 449-464. New
Ruggieri, G. 1957. Nuovi dati sul contatto Pliocene-Calabriano
York: Geological Society of America.
nella sezione del Santerno (Imola). Giorn. Geol. 26:81Sprovieri, R. 1978. I Foraminiferi bentonici della sezione Plio89.
Pleistocenica di Capo Rossello (Agrigento, Sicilia). Soc.
Ruggieri, G. 1975. Revisione della Ostracofauna marina
Paleontol Ital, Boll. 17:68-97.
Quaternaria di Imola (Bologna). Riv. Espanola Micropal.
Sprovieri, R., D'Agostino, S., and Di Stefano, E. 1973.
6:419-446.
Giacitura del Calabriano nei dintorni di Catanzaro. Riv.
Ruggieri, G., Buccheri, G., Greco, A., and Sprovieri, R. 1976.
Ital Paleontol Strat. 79:127-140.
Un affioramento di Siciliano nel quadro della revisione
della stratigrafla del Pleistocene inferiore. Soc. Geol. Ital., Sprovieri, R., Ruggieri, G., and Unti, M. 1980. Globorotalia
truncatulinoides excelsa n. subsp., foraminifero plancBoll. 94:889-914.
tonico guida per il Pleistocene inferiore. Soc. Geol Ital,
Boll 99:3-11.
Stuart, A. J. 1982. Pleistocene vertebrates in the British Isles.
Soc. Geol. Ital, Boll 53:251-259.
London: Longmans.
Ruggieri, G., and Sprovieri, R. 1976a. Ricerche sul Siciliano di
Palermo: le argille del Fiume Oreto. Soc. Geol. Ital, Boll. Sue, J. P., and Zagwijn, W. H. 1983. Plio-Pleistocene correlations between the northern Mediterranean region and
84:1613-1622.
northwestern Europe according to recent biostratigraphic
Ruggieri, G., and Sprovieri, R. 1976b. La definizione dello
and palaeoclimatic data. Boreas 12:153-166.
stratotipo del Piano Siciliano e le sue consequenze. Riv.
Suzuki, K., and Manabe, K. 1982. Pliocene-Pleistocene chronolMin. Siciliana 26:151-153.
ogy of the Yamato Group of Aizu Basin, Northeast
Honshu, Japan. In The third report on the PliocenePleistocene stratigraphy. Geol. Romana 16:131-139.
Pleistocene Boundary in Japan, ed. M. Itihara and Y.
Savage, D. E., and Curtis, G. H. 1970. The Villafranchian
Kuwano, pp. 18-27. Kyoto: Dogura and Co.
Stage-Age and its radiometric dating. Geol Soc. Am.,
Bull 124:207-231.
the Pliocene-Pleistocene boundary in the Vrica section,
Selli, R. 1954. La limite Plio-Pleistocene dans les environs
d'Ancona (Marche). In XIX Congres Geologique InternaThunell, R. C , and Williams, D. F. 1983. The stepwise
tionale, Alger. Comptes Rendus, vol. 15, pp. 241-247.
paleoceanographic conmditions in the Mediterranean:
oxygen isotopic studies of DSDP sites 125 and 132. Utrecht
Giorn. Geol 37, estratto, fasc. II.
Marchetti, D., Bigazzi, G., Bonadonna, K G . , Borsetti, Venzo, S. 1975. Remarks on the stratigraphic and paleontoA. M., Cati, F., Colalongo, M. L., D'Onofrio, S., Landini,
logical sequence of the proposed Pliocene-Pleistocene
W., Menesini, E., Mezzetti, R., Pasini, G., Savelli, C , and
boundary type-section of Le Castella. Ateneo Parmense,
Tampieri, R. 1977. The Vrica Section (Calabria - Italy).
ActaNat. 11:423-447.
A potential Neogene/Quaternary Boundary Stratotype. Watkins, N. D., Western, D. R., and Kennett, J. P. 1974.
Giorn. Geol 41:181-204.
Paleomagnetism of the type Pliocene-Pleistocene boundary section of Santa Maria di Catanzaro, Italy, and the
Shackleton, N. J., Hall, M. A., and Pate, D. 1995. Pliocene
problem of post-depositional precipitation of magnetic
minerals. Earth Planet. Sci. Lett. 24:113-119.
the Ocean Drilling Program: scientific results, Leg 138, ed.
12
Stratigraphy of the Plio-Pleistocene sequence of the

Mediterranean coastal belt of Israel and its implications for the
evolution of the Nile Cone
GEDALIAHU GVIRTZMAN, GUIDO M. MARTINOTTI, and SHIMON MOSHKOVITZ
Introduction
The Plio-Pleistocene sedimentary sequence of the coastal plain

and the continental shelf of Israel is an integral part of the Nile
Cone, which has been building up since the end of the Messinian
Event. The Yafo Formation of the Saqiye Group (Pliocene to
early Pleistocene) and the Hefer Formation of the Kurkar Group
(Pleistocene), which constitute this sequence, are subdivided
into 15 correlatable members. Nine biostratigraphic datum levels
have been established within the sequence, which permits
correlations to the Plio-Pleistocene events in the central
Mediterranean and Italian sequences.
Background
Our compilation of the marine Plio-Pleistocene sequence of the
coastal belt of Israel is based mostly on our previous studies and
on additional examinations of new boreholes. The lithostratigraphy and mapping, including correlations of wells and outcrops,
interpretations of seismic surveys, subcrop mapping, and sedimentology, were published by Gvirtzman (1969a,b, 1970, 1983)
and Gvirtzman and Buchbinder (1969). Other publications
include studies of biostratigraphy and molluscan correlation by
Moshkovitz (1961, 1963, 1968), studies of nannofossil biostratigraphy by Ehrlich and Moshkovitz (Ehrlich and Moshkovitz,
1978; Moshkovitz and Ehrlich, 1980), studies of foraminiferal
biostratigraphy by Martinotti (1981a-c, 1986), and studies of
combined biostratigraphy of foraminifera and nannofossils by
Moshkovitz and Martinotti (1979). In addition, compilations and
regional syntheses were prepared by Gvirtzman and Buchbinder
(1977, 1978) and by Gvirtzman et al. (1984).
The Nile Cone
The Plio-Pleistocene sequence is an integral part of the Nile
Cone (Figure 12.1). The eastern lobe of the cone is formed by
sediments which are transported by the Nile River and are
carried by anticlockwise longshore currents into the southeastern
corner of the Mediterranean (Emery and Bentor, 1960; Emery
and Neev, 1960; Neev et al., 1976; Nir, 1984). This lobe is
156
distributed over the coastal plain, the continental shelf and the
continental slope of Israel.
The present Nile Delta and Nile Cone began forming at the
end of the Miocene, when the Mediterranean sea level was
restored and sediment from the Nile began to backfill the deeply
dissected canyon that had been incised to the Messinian (latest
Miocene) base level. Thus, the sediments of the delta and cone
overlie Messinian evaporites in most of the area. The sedimentary sequence of the delta and cone is composed of three
successive complexes (Rizzini et al., 1978; Gvirtzman and
Buchbinder, 1978; Said, 1981). In the deposits of the Israel
coastal plain, these are recognized as the basal post-evaporite
shallow-water Afiq Formation, the deep-water Yafo Formation,
and the overlying shallow-water Hefer Formation (Figure 12.2).
The Afiq is composed of coarse elastics that fill erosional
channels, and it does not form a continuous sheet. This complex
is found only at the southern margin of the area, near the
continent. The middle complex, or Yafo, forms most of the
volume of the delta and the cone and is distributed throughout
the entire area. This complex includes open-sea and hemipelagic
clays and silts in foreset beds. The Hefer strata at the top are
composed of coarse, mainly sandy elastics distributed (in the
same manner as the Afiq) only in the southern coastal plain and
near the continent.
Four lithologic components make up the sediments (Gvirtzman, 1970):
1. Clays (smectite, kaolinite, and illite) and quartz particles, most of which were transported by the Nile River.
This component accounts for most of the volume of the
delta and the cone.
2. Biogenic detritus formed in the marine environment,
including microfossil oozes (planktonic foraminifera
and nannofossils) and shells of benthic foraminifera,
siliceous microfossils, and mollusks.
3. Eolian dust from the Sahara and Sinai deserts.
4. Coarse detritus, including conglomerates, transported
by small rivers other than the Nile.
Features such as salt tectonics, salt flows, and diapirs are
common in the continental slope in areas where thick Messinian
Plio-Pleistocene of Israel
157
ERATOSTHENES
SEAMOUNT/
Figure 12.1. Physiographic elements in the southeastern Mediterranean (Ross and Uchupi, 1977;
Gvirtzman and Buchbinder, 1978;
Said, 1981) and location map of
the studied area. Section A - B - C
is illustrated in Figures 12.3 and
12.5. Data on boreholes 1-18 are
given in the Appendix.
evaporites underlie the Nile Cone. These features cause huge

growth faults, gigantic rotational slump structures, and collapse
structures in the clayey sediments of the cone (Neev et al., 1976;
Ross and Uchupi, 1977; Almagor and Garfunkel, 1979).
Gravitites (gravity-induced sediments) and suspensites (hemipelagic-sapropelic sediments) are being deposited today, as
during the late Quaternary, over most of the cone area
(Maldonado and Stanley, 1978). After a study of clay minerals
from those sediments, Maldonado and Stanley (1981) concluded that the smectite, the dominant clay mineral, is
deposited by the mass flow of water from the Nile River,
whereas the kaolinite is windblown, originating in North Africa,
and the illite and chlorite have a northwestern provenance. It is
highly probable that sediments of the same type and composition have been deposited during the entire history of the Nile
Delta and Cone.
Lithostratigraphy
The findings from our study are presented in stratigraphic

sections of 18 representative boreholes or wells selected along a
line following the present shoreline of the Mediterranean and
parallel to the depositional strike (Figure 12.1). An up-to-date
lithostratigraphic subdivision of these wells, together with
distribution charts for selected index fossils, has been prepared
from the borehole data shown in the Appendix. The new
lithostratigraphic nomenclature is illustrated in Figure 12.2;
detailed correlations of the upper part of the sequence, including
the occurrences of the most significant fossils, are shown in
Figure 12.3. The sequence from the Jaffa 1 borehole, which
constitutes the reference section for the Plio-Pleistocene in

Israel (Gvirtzman, 1970; Gvirtzman et al., 1984), is illustrated in
Figure 12.4, and a regional correlation chart of the basin is
shown in Figure 12.5.
Two stratal groups are included in the Plio-Pleistocene
sequence: the Saqiye and the Kurkar. The Saqiye Group
(Gvirtzman and Reiss, 1965; Gvirtzman, 1970) is dated from the
late Eocene to the early Pleistocene and is composed mostly of
open sea marls and clays. Its upper part (Figure 12.2), in
ascending order, includes the Mavqiim Formation (Gvirtzman,
1970), composed of anhydrite, gypsum, and salt of late Miocene
(Messinian) age, the Afiq Formation (Gvirtzman, 1970;
Gvirtzman and Buchbinder, 1978), composed of conglomerates,
variegated shales, sandstones, siltstones, and marls of latest
Messinian (latest Miocene) age, and the Yafo Formation
(Gvirtzman and Reiss, 1965), composed of marine marls and
clays of Pliocene and early Pleistocene age. The Kurkar Group
(Gvirtzman, 1969b; Gvirtzman et al., 1984), which ranges from
early Pleistocene to Holocene in age (Figure 12.2), is composed
of calcareous sandstone (known by the local name of "kurkar"),
reddish clayey-silty sandstones (known by the local name of
"hamra"), siltstones, marls, conglomerates, dark swamp claystones, and unconsolidated dune sands. In the west, near the
present shoreline, this group is represented by the Hefer
Formation (Gvirtzman et al., 1984), composed of marine
calcareous sandstones, eolianites, hamras, siltstones, and marls.
To the east, the Kurkar Group in the coastal plain is represented,
in ascending order, by the Pleshet Formation (Issar, 1961, as
emended by Gvirtzman and Buchbinder, 1969) composed of
marine calcareous sandstones, by the Ahuzam Formation (Issar,
Gvirtzman, Martinotti, and Moshkovitz
158
;HA a HEDERA i
of which are recycled Cretaceous and Tertiary fossils. Seismic

records from the western coastal plain and from the continental
shelf show huge growth faults and rotational slump structures in
the Yafo Formation (Almagor and Garfunkel, 1979). The
formation is subdivided into three members.
Lower Member. The Yafo Lower Member, introduced herein,
has its type section in the Jaffa 1 borehole in the depth interval
between 650 and 1,212 m (Figures 12.2 and 12.5). It is composed
of relatively carbonate-rich marls, with abundant planktonic
foraminifera and calcareous nannofossils. Its thickness changes
from about 100 m at the margin of the basin to about 700 m at its
center (Figure 12.5). The member overlies the Mavqiim and the
Afiq formations and in some places rests on a truncated surface
of older formations. The top of the Lower Member is eroded
along the margin of the basin. In the center of the basin, the
member is overlain by the Middle Member of the Yafo
Formation. The boundary between the Lower and the Middle
members is somewhat transitional, and therefore it is not always
well defined.
Figure 12.2. Stratigraphic nomenclature of the PlioPleistocene sequence of the coastal plain of Israel (modified from Gvirtzman et al.,
1984). Lithology: 1, marine calcareous sandstones (kurkar); 2,
eolianites; 3, clayey-silty reddish sandstones (hamra); 4, marine shale;
5, dark clayey swamp deposits with plant remains; 6, marine silts,
clays, and calcareous sandstones; 7, loose dune sands; 8, conglomerates; 9, coquina beds; 10, anhydrites. Biostratigraphic markers: A,
Amphisorus sp.; H, Hyalinea baltica; G, Gephyrocapsa oceanica; P, Pyramidella plicosa; D, Discoaster spp.; S, Sphaeroidinellopsis seminulina;
Gm, Globorotalia margaritae.
1961) composed of fluvial conglomerates, and by the Rehovot

Formation (Issar, 1961) composed of hamras, eolianites, and
dark swamp clays. The Hefer Formation is a lateral equivalent of
these three formations. The main difference between the Saqiye
and the Kurkar groups is that the Saqiye is composed mainly of
marls, clays, and silts, whereas the Kurkar is made up mainly of
sandstones.
Yafo Formation
The Yafo Formation varies in thickness from a few tens of meters
in the east to about 2,000 m in the continental shelf. The type
section of the formation is in the Jaffa 1 borehole in the depth
interval between 217 and 1,212 m (Figure 12.4). The marls of the
formation are composed of smectite, kaolinite, and illite (2070%) and of silt-size and fine-sand-size quartz particles (510%), most of which are derived from the Nile River
(Gvirtzman, 1970). The marls also include biogenic calcite (2575%), mainly derived from microfossils and nannofossils, some
Middle Member. This member, also introduced herein, has its

type section in the Jaffa 1 borehole in the depth interval between
310 and 650 m. It is relatively rich in clays and marls and low in
carbonates, and while poor in planktonics, redeposited nannofossils are abundant (Moshkovitz and Ehrlich, 1980). Its thickness
varies from zero at the margin of the basin to about 500 m in the
center. The member conformably overlies the Lower Member
and is overlain by the Petah Tiqwa Member, the latter horizon
being marked by a strong seismic reflector just above the
boundary (remark no. 25 in the Appendix). From seismic
records it is evident that most of the growth faults and the slump
structures are confined within the Middle Member. The seismic
reflector near the base of the Petah Tiqwa Member appears to be
a smoothed surface on the distorted blocks and rotated slabs of
the Middle Member. Some of the faults, however, also displace
the overlying Petah Tiqwa Member.
Petah Tiqwa Member. The Petah Tiqwa unit of Gill (1965) is the
upper member of the Yafo Formation, with its type section in the
Jaffa 1 borehole, in the depth interval between 217 and 310 m,
and is composed of carbonate-rich marls. In most of the area,
three distinct coquina beds composed of fine sands and
carbonate-rich marls alternate with clayey beds, with a distinct
signature that can be recognized clearly in the electric logs
throughout the entire basin. The lowermost coquina bed is rich
in loose, well-preserved mollusks. The thickness of the member
varies from about 10 m to about 100 m.
Five electric-log markers in the Petah Tiqwa, known in
ascending order as BPT, PT3, PT2, PT1, and A, can be followed
throughout most of the area. These electric markers (Figures
12.3 and 12.4) have been used for precise correlation. With the
Petah Tiqwa Member a regional shallowing of the entire basin is
introduced. Minor sands presage the shift to dominantly sandy
lithologies in the overlying members of the sequence.
-N-
-s-
-N-
Figure 12.3. Upper Pliocene to

Pleistocene correlations in
boreholes along the shoreline of Israel (partly from Gvirtzman et
al., 1984). See Figure 12.1 for locations of the correlation profiles,
and Appendix for borehole data.
Lithologic notation as in Figure
12.2. Boundary types: 1, lithostratigraphic boundary; 2,
biostratigraphic boundary; 3, first
occurrence (FO) datum; 4, last occurrence (LO) datum. The points
marked A, PT1, PT2, PT3, and
BPT are electric-log markers in
the Petah Tiqwa Member. Biostratigraphic markers: A,
Amphisorus sp.; H, Hyalinea baltica; G, Gephyrocapsa oceanica; P,
Pyramidella plicosa; D, Discoaster
spp.; S, Sphaeroidinellopsis
seminulina; Gm, Globorotalia margaritae.
160
Q
u.
UI -J
\NETANYA/GIVAT OLGA
KEFAR VITI
HERZLIYYA
T
^POLEG
u.
UJ
X
/tAESAREA\
UPPER
BAT YAM
/NIZZANIM\
L. BAT YAW
T If
A
(1
^^
IB -
ASHDOO
0-
2
UI
Z
1 50
li 5c
-PTi
-PT 2
u
O
CO
UI
_1
UI
^ !
DAN
PETAH
TIQWA
lui
(_)
UJ
oc
UI
^
1
Formation
NO.
2
2
TAARUKHA
TEL AVIV
_
8I0STR
E DISTRIBUTION
ZONES
X
2-J 2"
<
H- OF 5ELECTE
3UJ
cr
a. INDEX FOSSIL
z
o
UI
Q
Q -J u.
z
HOL
CD
2
a:
OC
to
- cc
oc <
cc
UI
YROCA t
BALT
o
f
Q_
a.
:=rlf
"PT 3
>30_
7)
ID
D -
T
p
1
AMALIS
-I-
_4^
MIDDLE
STE
(to
_j
a.
(D
ATE
D
80_0^
UI
* "
Qi
-I:
CP
IOCE
650
_j
900"
-
~z<
LOWER
-i-
100_0
~~
>:
-~~
-~~
~_
~-~
-_
1
*.
O
|
NHYDRITE
1
S|
ro
_
-
vi
2
Z
CO
^~
1100"
_
-
st~
-_~
~
~
~~
UJ
2
2
2
d
Nl
_J
CC
<
UI
UJ
2
UI
CL
e>
a:
UI
2
2
1. siltstone, marls, marine calcareous sandstones rich in

benthic fauna and mollusks of the Dan Member
2. marine marls and clays of the Nizzanim Member
3. marine calcareous sandstones, including beach rocks
and coquina beds, of the Lower Bat Yam Member, the
Upper Bat Yam Member, the lower part of the
Ashdod Member, and the lower part of the Herzliyya
Member
4. eolianites of the upper part of the Ashdod Member, the
upper part of the Herzliyya Member, the Giv'at Olga
Member, and the Tel Aviv Member
5. reddish, clayey-silty sandstones ("hamra") of the Yad
Mordekhay Member, part of the Caesarea Member,
and layers in the Poleg, Kefar Vitkin, and Netanya
members.
6. dark swampy clays with plant remains of part of the
Caesarea Member
The Hefer Formation is a product of six sedimentary cycles
(Issar, 1961; Gvirtzman et al., 1984) resulting from the eustatic
oscillations of Quaternary sea levels. The sedimentary cycles,
from bottom to top, are represented by the members as
follows:
=)
UI
v>
- -JT
1200
_
Ul
2:
=:|
-i"
~_~
_-_
-i-
DDI
:-:
LIO
<
i-z
-i~
o
h-
LIN A Z.
-:~
SIS_
UI
The type section of this formation (Gvirtzman et al., 1984) is in

the Jaffa 1 borehole (Figure 12.4). The Hefer conformably
overlies the Petah Tiqwa Member in the center of the basin and
rests unconformably on older truncated formations in the distal
parts of the basin, with a thickness varying from about 70 m in
the east to about 180 m in the west. The formation is subdivided
into 13 members (Gvirtzman et al., 1984) (Figure 12.2), one of
which, the Yad Mordekhay Member, wedges out in the
westernmost belt of the coastal plain and therefore is not
included in the correlation chart of Figure 12.3.
The following lithologies are found in the various members of
the Hefer Formation (Figure 12.2):
2
UI
130j0
Figure 12.4. Reference type section of the Plio-Pleistocene sequence of the

coastal plain of Israel, in the Jaffa 1 borehole (no. 9 in Figure 12.1 and in
the Appendix). The distribution chart of index fossils and biostratigraphic
datum levels for the Plio-Pleistocene sequence are based on data compiled
cycle 1: Petah Tiqwa-Dan-Yad Mordekhay

cycle 2: Lower Bat Yam-Nizzanim-Upper Bat YamCaesarea
cycle 3: Ashdod-Poleg
cycle 4: Herzliyya-Kefar Vitkin
cycle 5: Giv'at Olga-Netanya
cycle 6: Tel Aviv
The Ta'arukha Member (Horowitz, 1979) and the Hadera
Member (Horowitz, 1979) are two generations of sand dunes
which cover part of the coastal plain. The Ta'arukha Member has
been partly stabilized since the Middle Bronze Age (Gvirtzman
herein (see remarks 17,18, and 19 in the Appendix) and data from
Gvirtzman (1970), Gvirtzman et al. (1984), Martinotti (1981a),
Moshkovitz (1963), Moshkovitz and Ehrlich (1980), Reiss (1964), and Reiss
and Issar (1961). Nannofossil zones after Martini (1971).
161
-N-
HADERA. TA'ARUKHA.TEL AVIV, NETANYATS
Afiq Fm.
(Messinian)
Mavqi im F
(Messinian)
Figure 12.5. Cross section A-B-C of the full Upper Miocene to Pleistocene basinal section in the coastal plain of Israel. Locations of boreholes
as in Figure 12.1, and lithostratigraphy, boundaries, and
biostratigraphic markers denoted as in Figures 12.3 and 12.4. Note

gaps and unconformities in the northern margin (Carmel block) and
the southern margin (Gaza-Be'eri block) of the basin.
et al., 1984). The Hadera Member is a result of the rejuvenation

of the sand migration since the time of the Byzantines and
Crusaders (Gvirtzman et al., 1984).
rately determined. Another problem was that the Upper

Pliocene and the Pleistocene sequences in Israel are relatively
poor in planktonics, and correlations between the eastern and
western Mediterranean deposits of this age are not always well
founded.
In this chapter, we refer only to occurrences of index fossils
whose stratigraphic distributions have been well established, in
spite of the previously mentioned limitations. These are as
follows: the LO (last occurrence) of Globorotalia margaritae; the
LO of Sphaeroidinellopsis seminulina; the LO of Discoaster spp.
(including D. tamalis, D. surculus, D. pentaradiatus, and D.
brouweri, all of which disappear from our area at about the same
time because of local paleoecological conditions); the FO (first
occurrence) and LO of Pyramidella plicosa, the FO of
Gephyrocapsa oceanica, single occurrences of Hyalinea baltica,
and the FO and LO of an Amphisorus sp. On the other hand,
some fossils that occur only rarely and whose exact stratigraphic
ranges still are in doubt were not included, in spite of their
biostratigraphic importance. Those include Globorotalia puncticulata, G. crassaformis, G. bononiensis, G. inflata, Reticulofenestra pseudoumbilicata, Pseudoemiliana lacunosa, and Cyclococcolithus macintyrei, among others.
Biostratigraphy
The distribution and zonations of the marine invertebrates of

the Yafo and Hefer formations are well established in the Israeli
coastal plain, based mostly on subsurface information from
boreholes. The foraminiferal biozonation has been published by
Reiss and Issar (1961), Reiss (1964, 1966), Perath (1965), Reiss
and Gvirtzman (1966a,b), Derin and Reiss (1973), Moshkovitz
and Martinotti (1979), and Martinotti (1981a,b). The biostratigraphy of the mollusks, which occur mainly in the Petah
Tiqwa and Dan members, was established by Moshkovitz
(1961, 1963, 1968), and nannofossil studies were published by
Ehrlich and Moshkovitz (1978), Moshkovitz and Martinotti
(1979), and Moshkovitz and Ehrlich (1980). One of the main
problems encountered in all those studies was the lack of
good continuous cores. The cutting samples from the boreholes were highly contaminated by uphole caving, and the
first occurrences of index fossils could not always be accu-
162
nean, such as DSDP site 132 (Raffi and Rio, 1979), at Vrica
(Backman et al., 1983; Rio, Raffi, and Backman, Chapter 5, this
volume), and at Capo Rossello (Rio et al., 1984; Chapter 5, this
LO of Globorotalia margaritae. This datum level is well
volume), the discoasterids show successive extinctions which can
established in many wells in Israel and has been correlated with
be correlated with the extra-Mediterranean biozones. It therethe global last appearance of the same taxon (Martinotti, 1981b):
fore seems that the disappearance of the discoasterids from our
Lower Member of the Yafo Formation.
area is a result of three factors which are locally interrelated:
global cooling, a lowered sea level, and an increase of freshwater
LO of Sphaeroidinellopsis seminulina. This well-established discharge, with simultaneous increase in the flux of continental
datum level is correlated with the last appearance of the taxon in detritus from the Nile. Thus, shallowing, increasing erosional
the deepest basins of the Mediterranean (cf. Martinotti, 1981b, forces, turbulence, and dilution in this region led, simulta1986): Lower Member of the Yafo Formation.
neously, to disappearance of the discoasterids, a decrease of
planktonic foraminifera, the dominance of some benthonic
LO of Discoaster spp. In the subsurface of the coastal plain and foraminifera, and better preservation of allochthonous siliceous
continental shelf in Israel, this datum level is related to an microorganisms.
ecological event, because all of the species of this genus disappear
at a certain level in the sequence in all the boreholes studied in FO and LO of Pyramidella plicosa. This gastropod is the most
Israel (Ehrlich and Moshkovitz, 1978; Moshkovitz and Mar- significant taxon in a mollusk assemblage (Moshkovitz, 1963)
tinotti, 1979; Moshkovitz and Ehrlich, 1980, p. 13). The dis- that forms the widespread lower coquina bed in the Petah Tiqwa
appearance of the discoasterids from our area occurred during Member (Gvirtzman, 1970). It is always found between the BPT
NN-16 and therefore not later than 2.6 Ma, according to the inter- and PT3 E-log markers (Figures 12.3 and 12.4) and is a good tool
polated time scale of Rio, Sprovieri, and Raffi (1984, table II). for local correlations. The so-called Pyramidella plicosa zone
This disappearance level, which is found in the Yafo Lower includes macrofossils such as Pyramidella plicosa Brn., DenMember, is interpreted in our area as being related to a major talium michelottii Hoernes, D. sexangulum acutangularis Cocc,
change in the discharge of the Nile River in the middle-to-late and D. ?passerinianum Coss., of typical Pliocene age. Because
Pliocene. The extinction of discoasters coincides with a sharp the assemblage also contains some forms representing extant
reduction in planktic/benthic ratios and a marked increase in lineages, it has been interpreted to be of late Pliocene age
reworked nannofossils of Upper Cretaceous-Lower Tertiary (Moshkovitz, 1963, 1968). Unfortunately, biozones of this kind
age, together with the appearance of marine and fresh-water cannot be correlated with the global stratigraphy of the Pliosiliceous microfossils. This evidence points to abundant influx of Pleistocene (Berggren et al., 1980). The fact, however, that this
alluvial material from the Nile (Ehrlich and Moshkovitz, 1978). molluscan assemblage is always found in layers above the NN-16
At about the same level, we observe a strong increase of shallow- zone and below the first occurrence of the early Pleistocene
water benthic foraminifera, with "many Ammonia and elphidiids Gephyrocapsa oceanica is confirmatory evidence of the late
of robust size" noted by Perath (1965, p. 19), indicating a Pliocene age of P. plicosa. Furthermore, in the Ashqelon 2
reduction in water depth. A parallel decrease or absence of borehole (no. 13, Figures 12.1, 12.3, and 12.5), the last
discoasterids in the NN-16 zone in the western Mediterranean occurrence of P. plicosa was found at about 304 m, significantly
was noted by Muller (1979), who associated that phenomenon below the first occurrence of Globorotalia inflata at 250 m
with the glaciation of the Northern Hemisphere. According to (Martinotti, 1981b). Although G. inflata is very rare in our area
Rio et al. (1984), marked changes in the late Pliocene, at about and has not been established as an index fossil, as discussed
2.4-2.6 Ma, occurred in the western Mediterranean when earlier, the possibility that these observations represent the
discoasters faded out (with the exception of D. brouweri). Those regional last occurrence of P. plicosa below the regional first
floral events were also related to a severe cooling, in this case occurrence of G. inflata should not be excluded.
one of the precursor cycles (Rio et al., 1984, referring to Thunell
FO of Gephyrocapsa oceanica. This datum level was found
and Williams, 1983). Initiations of major global climatic changes during our investigations in most of the wells studied in the upper
and high-latitude glaciation at about 2.4-2.6 Ma have been part of the Petah Tiqwa Member, slightly below the PT1 electricnoted by many authors, both in the Northern Hemisphere log marker (Figure 12.3; see also remarks 13 and 19 in the
(Berggren, 1972; Berggren and Van Couvering, 1974; Shackle- Appendix). This datum level is correlated with the global firstton et al., 1984) and in the Southern Hemisphere (Stipp, appearance datum of G. oceanica. Detailed evolutionary studies
Chappell, and McDougall, 1967), as well as in the western of this species (Samtleben, 1980) and its significance for
Mediterranean (Keigwin and Thunell, 1979; Thunell and Wil- Pleistocene stratigraphy was recently summarized by Perchliams, 1983; Sue, 1984). Stipp et al. (1967) noted a major drop in Nielsen (1985, p. 511). In the Mediterranean basin, its first
sea level as a result of the advances of polar glaciers at that time, appearance was recorded in the Vrica section slightly above the
and Vail and Hardenbol (1979) also reported a fall in sea level Olduvai magnetic event (Rio et al., Chapter 5, this volume).
during the NN-16 zone time interval.
Accordingly, it might prove to be a useful tool for both
Nevertheless, in several locations in the western Mediterra- stratigraphic and paleoecologic purposes in our area.
Datum levels
Qccurrences of Hyalinea baltica. Along the line of correlation

shown in Figure 12.3, this fossil was found in onlyfiveboreholes,
occurring only rarely (remarks 15, 18, 23, and 27 in the
Appendix). In four of the boreholes, its specific stratigraphic
position is in the upper part of the Dan Member, above the FO
of G. oceanica. In the fifth, its stratigraphic position is unknown
(remark 27 in the Appendix). This fossil is one of the so-called
Ml-zone assemblage, found in boreholes near Tel Aviv (Reiss
and Issar, 1961). Its absence higher in the section in the Lower
Bat Yam Member might be related to a facies change. This fossil
is the only representative in Israel of the "nordic guests" in the
Mediterranean, and it is found in an assemblage characterized as
the "Calabro-Sicilian" stage (Reiss and Issar, 1961). As in the
western Mediterranean area, such as the Vrica section (Pasini
and Colalongo, 1982; Ruggieri, Rio, and Sprovieri, 1984, p.
257), this fossil occurs in Israel slightly above the FO of G.
oceanica in the upper Dan Member. Therefore, in spite of its
rarity in Israel, it seems probable that its occurrence is very close
to the first appearance of the species.
163
appears that the invasion into the eastern Mediterranean of the

Senegalian warm-water fauna lasted during the time of stages 7
and 5.
Evaluation of datum levels and correlations
The nine datum levels defined by the distribution of seven index

fossils in the Plio-Pleistocene marine record in the coastal belt of
Israel are indicated in the type section of the Jaffa 1 borehole
(Figure 12.4). Three of these datum levels are of global
significance, namely: G. margaritae LO, S. seminulina LO, and
G. oceanica FO (no. 1, 2, and 6, respectively). The other six
datum levels are considered as ecologic, climatic, or sedimentologic "events" and can be used for intra-Mediterranean or
intra-basinal correlations. The term "event" is used here for
designating episodes in a sequence of significant physiographic
changes in the huge "natural sedimentation laboratory of the
Mediterranean Sea" (Stanley, 1972). This usage follows another
well-established term, in the same context and for the same
basin: the late Miocene "Messinian Event." Although all of the
FO and LO of Amphisorus sp. Avinmelech (1952) included a Plio-Pleistocene events are believed to have happened almost
Marginopora sp. and Amphisorus hemprichi in a detailed list of synchronously throughout the Mediterranean, they were not
microfossils found in the Quaternary of the coastal plain. An synchronous in the same sense as an evolutionary event. For
unidentified species of Amphisorus was also recorded by Itzhaki instance, the local LO of Discoaster spp. (datum level 3) reflects
(1960) from strata that he attributed to the Tyrrhenian. Reiss and an ecologic event related to global glaciation and changes in the
Issar (1961) regarded Amphisorus to be a synonym of Mar- Nile. Three other datum levels are related to circumginopora and characteristic of their Mp and Mpa biostratigraphic Mediterranean climatic events of faunal migration. The occurzones. Those zones were later equated to the Ashdod and rence of Hyalinea baltica (datum level 7) is connected with the
Herzliyya members of the Hefer Formation (Gvirtzman et al., invasion of "nordic guests" from the North Atlantic Ocean into
1984). Following Reiss and Issar (1961), this fossil was recorded the Mediterranean Sea. The FO and LO of Amphisorus (datum
as Marginopora (Michelson, 1970; Issar and Kafri, 1972; levels 8 and 9) were related to a warm-water faunal invasion of
Gvirtzman et al., 1984), but Hottinger (1977, p. 100) and, more the Mediterranean from the coasts of western Africa. Two other
recently, Z. Reiss (personal communication, 1987) have restored datum levels (4 and 5) are reflections of a limited local
the identity of this fossil as Amphisorus, an amendment followed sedimentological event on the Levant shoreline, where a rather
here. Amphisorus is found in the marine calcareous sandstones thick coquina bed is characterized by the presence of Pyraof the aforementioned two members, usually in the lower part midellaplicosa (FO and LO in the same bed).
below eolianites. Issar and Kafri (1972) found the gastropod
Strombus bubonius in the Galilee coastal plain, in AmphisorusThe Pliocene-Pleistocene boundary
bearing calcareous sandstones that subsequently were named the
Herzliyya Member (Gvirtzman et al., 1984). The sedimentary Two possibilities for a biostratigraphic identification of this
cycles of Ashdod (which contains only an Amphisorus sp.) and of boundary in Israel can be considered: (1) a boundary based on
Herzliyya (which contains both an Amphisorus sp. and Strombus the first-occurrence datum of G. oceanica, as proposed at the
bubonius) represent in Israel the "warm-water Senegalian fauna INQUA Moscow Congress of 1982 (Pasini and Colalongo, 1982;
of the Tyrrhenian Stage," which is well known from Quaternary cf. Pelosio, Raffi, and Rio, 1980); or (2) the level suggested in
shorelines all through the Mediterranean basin. Thus, stages the draft proposal of INQUA Subcommission 1-d on the
such as the "Tyrrhenien inferieur," "Eutyrrhenien," and Pliocene-Pleistocene boundary (Aguirre and Pasini, 1985),
"Neotyrrhenien," of Bonifay (1975), or the "shorelines with based on the detailed study of the Vrica stratotype (Selli et al.,
Strombus fauna" (Butzer, 1975), should be considered, in our 1977; Pelosio et al., 1980; Obradovich et al., 1982; Pasini and
opinion, as reflecting a circum-Mediterranean "Tyrrhenian Colalongo, 1982; Chapter 2, this volume; Backman et al., 1983;
Event," based on stratigraphic relationships, paleoclimatic con- Tauxe et al., 1983). That latter proposal defined the Pliocenesiderations, and human artifacts.
Pleistocene boundary at the base of the claystone layer overlying
Correlations between the Ashdod and the Herzliyya members sapropelic bed e in the Vrica section, approximately 22 m below
and the oxygen-isotope stages 7 and 5, respectively, of Emiliani the FO datum of G. oceanica and close to the top of the Olduvai
and Shackleton (1974) were recently suggested (Gvirtzman, normal event.
The first option would locate the boundary in the Petah Tiqwa
1980; Gvirtzman et al., 1984). From these correlations, it
164
southern marginal areas of the basin, Figure 12.5). A westward

tilt of the entire coastal plain (Neev et al., 1978; Gvirtzman,
Dickenstein, and Croker, 1980) and truncation in the eastern
coastal plain resulted in unconformities between the Lower and
Petah Tiqwa members of the Yafo Formation. Other results of
the westward tilt were slumps and growth faults in the eastern
lobe of the Nile Cone. Deformation occurred mostly in the
The Mediterranean Pleistocene
Middle Member of the Yafo Formation. Alternations of Upper
In spite of numerous studies, the regional chronostratigraphy of Pliocene coquina beds with sand units and muds, as a result of
the Mediterranean Pleistocene remains unresolved. The Cala- regional shallowing, are recorded in the Petah Tiqwa Member.
The dominance of clays and silts in the depositional supply, as
brian, Sicilian, and Tyrrhenian stages have been difficult to
identify and correlate in terms of modern global chronostratig- seen in the Saqiye Group, gave way to a dominantly sandy supply
raphy (Berggren et al., 1980), although some progress has been for the Kurkar Group during the early Pleistocene. Though some
made recently (Preface, this volume). For the time being, sand units are found in the Petah Tiqwa Member, and some mud
however, it seems that Gephyrocapsa oceanica, Hyalinea baltica, units are found in the Dan Member, the overall change in
and the Amphisorus sp. in our area are useful for intra- lithology is quite marked.
An early Pleistocene invasion of the Mediterranean by "nordic
Mediterranean correlations. G. oceanica is the only fossil which
can also be used for global correlation. The possible correlation guests" from the North Atlantic coasts is documented in the
of Pleistocene sedimentary cycles of Israel with the global upper part of the Dan Member by the occurrence of Hyalinea
oxygen-isotope stages, as mentioned earlier, seems to hold baltica. An early Pleistocene renewal of the earlier block
movements and of the westward tilt is deduced from the
promise for future work in this regard.
erosional surface found below the Lower Bat Yam Member in
the Carmel area. During that pre-Bat Yam erosional phase, the
Event stratigraphy and correlation with the Nile Delta Ahuzam conglomerate, a fluvial sediment which covers the
The lithostratigraphic sequences of the coastal plain of Israel, eastern coastal plain (Gvirtzman et al., 1984) and the lower
from the Messinian evaporites upward, in conjunction with the Shefala and cuts valleys into the upper Shefala (Sneh and
biostratigraphic datum levels and their chronostratigraphic Buchbinder, 1984), was deposited. The Yad Mordekhay hamra
interpretations, have permitted detailed correlation within the was also deposited during the phase between the Dan and the
Nile Cone sequence in the coastal belt of Israel (Figure 12.5) and Lower Bat Yam members (Gvirtzman et al., 1984) (Figure 12.2).
also, to some extent, within the entire Mediterranean basin. A A late Pleistocene invasion of West African "warm-water
sequence of events can be deduced from the regional correlation Senegalian fauna" into the Mediterranean, and its extinction
chart (Figure 12.5), which in turn will throw new light on the thereafter, is documented by the occurrence of Amphisorus and
evolution of the Nile Delta and the Nile Cone. The reconstructed Strombus bubonius in the Ashdod and the Herzliyya members.
The Plio-Pleistocene sequence of the Nile Delta, described by
events, in their stratigraphic order, are summarized in the
Rizzini et al. (1978) and by Said (1981), is correlated with the
following scheme.
Evidences of desiccation, sebkha deposits, salt lakes, and sequence of the coastal belt of Israel, using mainly lithostratievaporite accumulations are recorded in the Upper Miocene graphic criteria, as follows:
Mavqiim Formation in Israel (Figure 12.2). Late Miocene
erosion, evidenced in a buried topography of canyons, channel
1. The Mavqiim Formation and the Rosetta Formation
fill, and drainage patterns, can be deduced from study of the
are time-equivalents, both composed of Messinian
Afiq Formation and from interpretation of seismic records for
evaporites.
the coastal plain of Israel. The widespread early Pliocene
2. The Afiq Formation and the Abu Madi Formation
transgression is connected with the beginning of the buildup of
represent the fluvial coarse elastics found at the base of
the modern Nile Delta and Nile Cone, in the form of basinal and
the Nile Delta and Cone in the nearshore area.
hemipelagic muds, deposited as deltaic foreset beds, recorded in
3. The Lower and Middle members of the Yafo Formation
the Yafo Lower Member. A middle-late Pliocene (ca. 2.6 Ma)
and the Kafr el Sheikh Formation are the main body of
dilution of the planktonic biomass, together with the extinction
the foreset muds in the Nile Delta and Cone.
of the discoasterids, probably resulted from global cooling and a
4. The Petah Tiqwa Member and the El Wastini Formachange in the discharge of the Nile River. That extinction event
tion represent the transition from the foreset mud units
is recorded in the upper part of the basinal foreset beds of the
of the Nile Delta to the overlying topset sand units.
Lower Member of the Yafo Formation.
Both formations are composed of alternating beds of
muds and sands.
Late Pliocene differential movements, in which the Carmel
and the Gaza-Beeri blocks were uplifted and eroded, while the
5. The Hefer Formation and the Mit Ghamr Formation
central basin was downfaulted, took place prior to the deposition
represent the topset sand beds of the Nile Delta in the
of the Petah Tiqwa Member (notably in the northern and
nearshore area.
Member, as seen in the Jaffa 1 type section (Figure 12.4),
approximately between the PT1 and the PT2 electric-log
markers, at a depth of about 234 m. The second option would
place the boundary also in the Petah Tiqwa Member, between
the PT2 and PT3 electric-log markers, at a depth of about 255 m.
165
Gill, D. 1965. Subsurface geology of the Petah-Tiqwa area (Up.

Cretaceous-Pleistocene) (in Hebrew). Tel Aviv: Israel
National Oil Company.
The authors wish to thank Ze'ev Reiss of the Hebrew University
of Jerusalem and Binyamin Buchbinder of the Geological Survey Gvirtzman, G. 1969a. The Saqiye Group (Late Eocene to Early
Pleistocene) in the Coastal Plain and Hashephela regions,
of Israel for their useful comments. Thanks are also due to Be vie
Israel, volume 2. Geol. Surv. Israel, Bull. 51:1.
Katz for editing the manuscript, to Rivka Backman and Malka Gvirtzman, G. 1969b. Subsurface data on the Saqiye Group (Late
Eocene to Early Pleistocene) in the Coastal Plain and
Wohl for typing the manuscript, and to Sa'adya Levy and Arie
Hashephela regions, Israel. Report OD/1/69. Jerusalem:
Peer for drafting the figures. This study is a result of Project
Geological Survey of Israel.
28977 of the Geological Survey of Israel.
Gvirtzman, G. 1970. The Saqiye Group (Late Eocene to Early
Pleistocene) in the Coastal Plain and Hashephela regions,
Israel. Report OD/5/67 (in Hebrew, with English sumReferences
mary). Jerusalem: Geological Survey of Israel.
Aguirre, E., and Pasini, G. 1985. The Plio-Pleistocene bound- Gvirtzman, G. 1980. Possible correlation between the Kurkarary. Episodes 8:116-120.
Hamra cycles of the Quaternary of the Coastal Plain of
Israel and the global-oceanic-climatic oxygen isotope
Almagor, G., and Garfunkel, Z. 1979. Submarine slumping in
continental margin of Israel and northern Sinai. Am.
cycles. In Program and Abstracts, 37th Bat-Sheva SemiAssoc. Petrol Geol, Bull 63:324-340.
nar on approaches and methods in paleoclimatic research
with emphasis on aridic areas. Jerusalem and Rehovot,
Avinmelech, M. 1952. Contribution to the knowledge of the
October 1980, pp. 46-47. Rehovot: Weizmann Institute
Quaternary oscillations of the shoreline in Palestine. In
of Science.
Atti del Congreso Internazionale Preistorico e Protoistorico
Gvirtzman, G. 1983. Quaternary coastal terraces and prehistoric
Mediterraneo, 1950, pp. 31-41. Firenze.
sites in Lebanon: correlation with sedimentary cycles in
Backman, J., Shackleton, N., and Tauxe, L. 1983. QuantitaIsrael and chronological interpretation. Integrating Artive nannofossil correlation to deep-sea sections from
chaeological and Geological data in Quaternary studies (in
Plio-Pleistocene boundary of Vrica, Italy. Nature 304:
156-158.
Hebrew). In Annual Meeting of the Israel Society for
Quaternary Research, Jerusalem, October 1983, ed. Z.
Berggren, W. A. 1972. Late Pliocene-Pleistocene glaciation. In
Begin and D. Ginzburg, pp. 14-20. Jerusalem: Geological
Initial Reports of the Deep Sea Drilling Project, vol. 12, ed.
Survey of Israel.
A. S. Laughton, W. A. Berggren et al., pp. 953-963.
Gvirtzman, G., and Buchbinder, B. 1969. Outcrops of Neogene
formations in the central and southern Coastal Plain,
Berggren, W. A., Burckle, L. H., Cita, M. B., Cooke, H. B. S.,
Hashephela and Be'er Sheva regions. Geol Surv. Israel,
Funnell, B. M., Gartner, S., Hays, J. D., Kennett, J. P.,
Bull. 50:1-76.
Opdyke, N. D., Pastouret, L. P., Shackleton, N. J., and
Takayanagi, Y. 1980. Towards a Quaternary time scale. Gvirtzman, G., and Buchbinder, B. 1977. The desiccation events
in the eastern Mediterranean during Messinian times as
Quatern. Res. 13:277-302.
compared with other Miocene desiccation events in basins
Berggren, W. A., and Van Couvering, J. A. 1974. The Late
around the Mediterranean. In International symposium on
Neogene: biostratigraphy, geochronology and paleoclimathe structural history of the Mediterranean basin. Split,
tology of the last 15 million years in marine and continental
Yugoslavia, ed. B. Biju-Duval and L. Montadert, pp. 411
sequences. Amsterdam: Elsevier.
420. Paris: Editions Technip.
Bonifay, E. 1975. "L'era quaternaire": definition, limites et
subdivisions sur la base de la chronologie mediter- Gvirtzman, G., and Buchbinder, B. 1978. The Late Tertiary of
the coastal plain and continental shelf of Israel and its
raneenne. Soc. Geol France, Bull 17:380-393.
bearing on the history of the eastern Mediterranean. In
Butzer, K. W. 1975. Pleistocene littoral-sedimentary cycles of the
Initial reports of the Deep Sea Drilling Project, vol. 42, pt.
Mediterranean Basin: a Mallorquin view. In After the
2, ed. D. A. Ross, Y. D. Neprochnov et al., pp. 1195australopithecines: stratigraphy, ecology and cultural change
1222. Washington DC: U. S. Government Printing Office.
in the Middle Pleistocene, ed. K. W. Butzer and G. LI. Isaac,
Gvirtzman, G., Dickenstein, J., and Croker, P. 1980. Oil
pp. 25-71. Chicago: Aldine.
prospects of the Sharon area and recommendation for the
Derin, B., and Reiss, Z. 1973. Revision of marine Neogene
drilling of Ga'ash 2. Report 80/14. Tel Aviv: Oil Explorastratigraphy in Israel. Isr. J. Earth Sci. 22:199-210.
Ecker, A., and Olsina, A. 1978. The geology of the Quaternary in
tion (Investments) Ltd.
the proposed nuclear power plant site, Nizanim, Israel. Gvirtzman, G., and Reiss, Z. 1965. Stratigraphic nomenclature
in the Coastal Plain and Hashephela regions. Report OD/1/
Report Hydro/1/78.1. Jerusalem: Geological Survey of
Israel.
65. Jerusalem: Geological Survey of Israel.
Ehrlich, A., and Moshkovitz, S. 1978. Distribution of calcareous Gvirtzman, G., Shachnai, E., Bakler, N., and Hani, S. 1984.
nannofossils and siliceous microfossils in the PlioStratigraphy of the Kurkar Group (Quaternary) of the
Pleistocene sediments of the Mediterranean coast and
coastal plain of Israel. Geol Surv. Israel, Current Res.
offshore of Israel (a preliminary study). Isr. J. Earth Sci.
1983-84:70-82.
27:65-71.
Horowitz, A. 1979. The Quaternary of Israel New York:
Emery, K. O., and Bentor, Y. 1960. The continental shelf of
Academic Press.
Israel. Geol. Surv. Israel, Bull. 26:25-41.
Hottinger, L. 1977. Distribution of larger Peneroplidae, Borelis
and Nummulitidae in the Gulf of Elat, Red Sea. Utrecht
Emery, K. O., and Neev, D. 1960. Mediterranean beaches of
Israel. Geol Surv. Israel, Bull. 26:1-24.
Emiliani, C , and Shackleton, N. J. 1974. The Brunhes Epoch: Issar, A. 1961. Geology of the subsurface water resource of the
isotopic paleotemperatures and geochronology. Science
Shefela and Sharon areas (in Hebrew), Report P.M. 307.
183:511-514.
Tel Aviv: Tahal Water Planning for Israel, Ltd.
Acknowledgments
166
Issar, A., and Kafri, U. 1972. Neogene and Pleistocene geology

of the western Galilee coastal plain. Geol. Surv. Israel,
Bull 53:1-17.
Itzhaki, Y. 1960. Pleistocene shorelines in the coastal plain of
Israel. Geol. Surv. Israel, Bull. 32:1-9.
Keigwin, L. D., Jr., and Thunell, R. C. 1979. Middle Pliocene
climatic change in the Western Mediterranean from faunal
and oxygen isotopic trends. Nature 282:292-296.
Klang, A., and Gafsou, R. 1981. Synthetic seismogram, Shiqma
1. Report no. SS/854/81 prepared for Oil Exploration
(Investments) Ltd. Tel Aviv: Institute for Petroleum
Research and Geophysics.
Maldonado, A., and Stanley, D. J. 1978. Nile Cone depositional
processes and patterns in the Late Quaternary. In Sedimentation in submarine canyons, fans and trenches, ed. D. J.
Stanley and G. Kelling, pp. 239-257. Stroudsburg, PA:
Dowden, Hutchinson and Ross, Inc.
Maldonado, A., and Stanley, D. J. 1981. Clay mineral distribution patterns as influenced by depositional processes in the
southeastern Levantine Sea. Sedimentology 28:21-32.
Martini, E. 1971. Standard Tertiary and Quaternary calcareous
nannoplankton zonation. In Proceeding of the 2nd Planktonic Conference, Roma, vol. 2, ed. A. Farinacci, pp. 739785. Rome: Ediziones Tecnoscienza.
Martinotti, G. M. 1981a. A new evaluation of the biostratigraphy of the Late Eocene to Pleistocene sequence in
the Jaffa 1 well. Geol. Surv. Israel, Current Res. 1980:
53-57.
Martinotti, G. M. 1981b. Biostratigraphy and planktonic
foraminifera of the Late Eocene to ?Pleistocene sequence
in the Ashqelon 2 well (southern coastal plain, Israel).
Rev. Espan. Micropal. 13:343-381.
Martinotti, G. M. 1981c. Notes on the stratigraphy of the Saqiye
Group as determined from foraminifera (first report).
Report P/3/81. Jerusalem: Geological Survey of Israel.
Martinotti, G. M. 1986. Renaming the Mediterranean Sphaeroidinellopsis subdehiscens Zone. Newsl. Strat. 16:49-55.
Michelson, H. 1970. Geology of the Carmel coastal plain, Report
HG/70/25 (in Hebrew). Tel Aviv: Tahal Water Planning for
Israel Ltd.
Moshkovitz, S. 1961. The tracing of the Plio-Pleistocene
boundary by means of Mollusca (in Hebrew). In Proceedings of the Symposium of the Pleistocene Research Society
in Israel (in Hebrew), pp. 1-5. Jerusalem: Hebrew
University of Jerusalem.
Moshkovitz, S. 1963. The mollusca in the upper part of the
"Saqiye beds" (Upper Neogene-Lower Pleistocene) in the
Central Coastal Plain of Israel. Isr. J. Earth Sci. 12:97-146.
Moshkovitz, S. 1968. The Mollusca in the marine Pliocene and
Pleistocene sediments of the southeastern Mediterranean
Basin (Cyprus-Israel). Unpublished Ph.D. thesis, Hebrew
University, Jerusalem.
Moshkovitz, S., and Ehrlich, A. 1980. Distribution of the
calcareous nannofossils in the Neogene sequence of the
Jaffa I borehole, central coastal plain. Report PD/1/80.
Jerusalem: Geological Survey of Israel.
Moshkovitz, S., and Martinotti, G. M. 1979. The biostratigraphy
of the Pliocene-Pleistocene sequence of some boreholes in
the Caesarea Netanya areas (calcareous nannofossils,
planktonic foraminifera and molluscs). Isr. J. Earth Sci.
28:110-127.
Muller, C. 1979. Nannoplankton. In Bizon, G., Report of the
working group on micropaleontology. In Proceedings of
the VII Congress, Regional Committee on Mediterranean
Neogene Stratigraphy, Athens, 1979, ed. G. Bizon et al.,
pp. 1335-1364. (Ann. Geol. Pays Hellen, tome hors ser.,

fasc. 3).
Neev, D., Almagor, G., Arad, A., Ginzburg, A., and Hall, J. K.
1976. The geology of the southern Mediterranean. Geol.
Surv Israel, Bull 68:1-52.
Neev, D., Shachnai, E., Hall, J. K., Bakler, N., and BenAvraham, Z. 1978. The young (post lower Pliocene)
geological history of the Caesarea structure. Isr. J. Earth
Sci. 27:43-64.
Nir, Y. 1984. Recent sediments of the Israel Mediterranean
continental shelf and slope. Unpublished Ph.D. thesis,
University of Gothenburg.
Obradovich, J. D., Naeser, C. W., Izett, G. A., Pasini, G., and
Bigazzi, G. 1982. Age constraints on the proposed PlioPleistocene boundary stratotype at Vrica, Italy. Nature
298:55-59.
Quaternary boundary-stratotype section, in 1982. Preprint: INQUA Sixth Congress, Moscow.
Perch-Nielsen, K. 1985. Cenozoic calcareous nannofossils. In
Plankton stratigraphy, ed. H. M. Bolli et al., pp. 427-554.
Cambridge University Press.
Pelosio, G., Raffi, S., and Rio, D. 1980. The Plio-Pleistocene
boundary controversy. Status in 1979 at the light of
International Stratigraphic Guide. In Volume dedicato a
Sergio Venzo, ed. F. Petrucci and R. Cavazzini, pp. 131
140. Parma: Grafiche STEP.
Perath, I. 1965. Microfaunas of the Jaffa I well, benthonic
foraminifera. Report OS/4/65. Jerusalem: Oil Companies'
Micropaleontological Laboratory.
Raffi, I., and Rio, D. 1979. Calcareous nannofossil biostratigraphy of DSDP Site 132-Leg 13 (Tyrrhenian Sea-Western
Mediterranean). Riv. Ital Paleont. Strat. 85:127-172.
Reiss, Z. 1964. Microfaunas of the Jaffa I well part I: planktonic
foraminifera. Report Pal/1/64. Jerusalem: Geological Survey of Israel.
Reiss, Z. 1966. Recent advances in marine Late Paleogene and
Neogene Stratigraphy of Israel. Isr. J. Earth Sci. 15:8-26.
Reiss, Z., and Gvirtzman, G. 1966a. Subsurface Neogene
stratigraphy of Israel. In Proceedings of the 3rd Session of
the Regional Committee on Mediterranean Neogene
Stratigraphy, Bern, ed. C. W. Drooger et al., pp. 314-346.
Leiden: E. J. Brill.
Reiss, Z., and Gvirtzman, G. 1966b. Borelis from Israel. Eclog.
Geol Helvet. 59:437-447.
Reiss, Z., and Issar, A. 1961. Subsurface Quaternary correlations in the Tel-Aviv region. Geol. Surv. Israel, Bull
32:10-26.
Rio, D., Sprovieri, R., and Raffi, I. 1984. Calcareous plankton
biostratigraphy and biochronology of the Pliocene-lower
Rizzini, A., Vezzani, R, Cococcetta, V., and Milad, G. 1978.
Stratigraphy and sedimentation of a Neogene-Quaternary
section in the Nile Delta area. Marine Geol. 27:327-348.
Ross, D., and Uchupi, E. 1977. Structure and sedimentary
history of southeastern Mediterranean Sea-Nile Cone
area. Am. Assoc. Petrol. Geol, Bull. 61:872-902.
chronostratigraphic classification of the Lower Pleistocene. Soc. Geol Ital, Boll. 103:251-259.
Said, R. 1981. The geological evolution of the River Nile. New
York: Springer-Verlag.
Samtleben, C. 1980. Die Evolution der Coccolithophoriden-
Gattung Gephyrocapsa nach Befunden in Atlantik.

Paldont. Z. 54:91-127.
Selli, R., Accorsi, C. A., Bandini Mazzanti, M., Bertoclani
Marchetti, D., Bigazzi, G., Bonatfdonna, F. G., Borsetti,
A. M , Cati, R, Colalongo, M. L., D'Onotffrio, S.,
(Calabria-Italy). A potential Neogene/Quaternary boundary stratotype. Giorn. Geol. 41:181-204.
Shackleton, N. J., Backman, J., Zimmerman, H., Kent, D. V.,
Hall, M. A., Roberts, D. G., Schnitker, D., Baldauf,
J. B., Desprairies, A., Homrighansen, R., Huddleston, P.,
Keene, J. B., Kaltenback, A. J., Krumsiek, K. A. O.,
Morton, A. C , Murray, J. W., and Westberg-Smith, J.
1984. Oxygen isotope calibration of the onset of ice-rafting
and history of glaciation in the North Atlantic region.
Nature 307:620-623.
Sneh, A., and Buchbinder, B. 1984. Miocene to Pleistocene
surfaces and their associated sediments in the Shefela
region, Israel. Geol. Surv. Israel., Current Res. 198384:60-64.
Stanley, D. J. (ed.) 1972. The Mediterranean Sea-a natural
sedimentation laboratory. Stroudsburg: Dowden, Hutchinson & Ross, Inc.
Stipp, J. J., Chappell, J. H. A., and McDougall, I. 1967. K/Ar
Age estimate of the Pliocene-Pleistocene boundary in
New Zealand. Am. J. Sci. 265:462-474.
Sue, J.-P. 1984. Origin and evolution of Mediterranean vegetation and climate in Europe. Nature 307:429-432.
Tauxe, L., Opdyke, N. D., Pasini, G., and Elmi, G. 1983. Age of
Thunell, R. C , and Williams, D. F. 1983. The stepwise
paleoceanographic conditions in the Mediterranean: oxygen isotopic studies of DSDP Sites 125 and 132. In
Reconstruction of marine paleoenvironments, ed. J. E.
Meulenkamp, pp. 111-127. Utrecht: Utrecht Micropaleontology Bulletins.
Vail, P. R., and Hardenbol, J. 1979. Sea-level changes during the
Tertiary. Oceanus 22:71-79.
Appendix
The following remarks refer to the wells denoted in Figures 12.1,

and 12.3-12.5. Numbers 1-18 refer to the 18 well sites studied
for this chapter, and numbers 19-27 refer to details of the
lithology and biostratigraphy of certain wells, as indicated.
1. Caesarea IEC 3. Lithostratigraphy after Gvirtzman et
al. (1984).
2. Caesarea IEC 1. Lithostratigraphy prepared for this
chapter.
3. Hadera 1. Biostratigraphy previously published.
4. Netanya IEC 1. Biostratigraphy according to the
authors.
5. Netanya IEC 4. Biostratigraphy after Moshkovitz and
Martinotti (1979).
6. Ga'asha 1. Biostratigraphy after Martinotti (1981c).
7. Reading 33/1A. Biostratigraphy after Martinotti
(1981b).
167
8. Tel Aviv 32/A. Biostratigraphy after Martinotti

(1981a).
9. Jaffa 1. Biostratigraphy after Moshkovitz (1963).
10. Dan Biyyuv 14. Biostratigraphy after Reiss and Issar
(1961).
11. Bene Darom. Biostratigraphy after Reiss (1964).
12. Nizzanim G/2. Biostratigraphy after Perath (1965).
13. Ashqelon 2. Biostratigraphy after Moshkovitz and
Ehrlich (1980), who noted a single occurrence of
Gephyrocapsa oceanica in core no. 1 (205 m depth).
However, as a result of recent detailed study, we noted
the first occurrence of G. oceanica in this well at a
depth of 224 m, very close to the electric-log marker
PT1.
14. Ashqelon G/O. Biostratigraphy after G. M. Martinotti
(in Moshkovitz and Ehrlich, 1980, table 2).
15. Shiqma SB/12. Hyalinea baltica was not found in this
well. However, Z. Reiss (in Moshkovitz, 1961) found
this fossil in the nearby Reading 33/0 well (410 m to the
northwest) in the interval 164-181 m. That interval was
projected to the correlative depth interval of this well,
as illustrated in Figure 12.3.
16. Shiqma S/103. This well is included in this report to
show the lithostratigraphic correlation according to the
authors.
17. Shiqma 1. The presence of Amphisorus was not
determined in this well. The presence of this fossil in
the depth interval cited (Figures 12.3 and 12.4) was
projected from several nearby wells, according to
information from Reiss and Issar (1961) and also from
the archives of the Geological Survey of Israel.
18. Nezarim 1. Hyalinea baltica was reported by Martinotti
(in Moshkovitz and Ehrlich, 1980, table 2) as a single
occurrence in core no. 1 (205 m). We were unable to
find any other specimens at or below that level, despite
careful examination.
19. In the Jaffa 1 well (no. 9), the ranges of distribution of
Gephyrocapsa oceanica, Hyalinea baltica, and Amphisorus, as shown in Figure 12.4, are the maximum ranges
according to correlations and compilations of all the
studied boreholes, as shown in Figure 12.3.
20. In the Bene Darom well (no. 11), no electric log is
available for the upper part or for the complete
sequence of this well. The lithostratigraphic correlation
is based only on well cuttings.
21. In the Ashqelon 2 well (no. 13), no electric log is
available for the upper part of this well, and the well
cutting samples are not reliable.
22. In the Ashqelon G/O well (no. 14), the presence of
Amphisorus was not determined, but is assumed
because of occurrences at equivalent depths in the two
nearby wells Nizzanim 10/0 (980 m to the northeast)
and OBS-1 (360 m to the northeast). Data from Ecker
and Olsina (1978).
168
23. In the Nizzanim G/2 well (no. 12), Hyalinea baltica

occurs at depths between 199 and 200 m and is also
found in the nearby Nizzanim B-76-8 well (860 m to the
southeast) at a depth of 192 m, both in the upper part of
the Dan Member.
24. The Jaffa 1 well (no. 9) is the type section of the Yafo
Formation (Gvirtzman, 1970) and of the Hefer Formation (Gvirtzman et al., 1984).
25. In the Shiqma 1 well (no. 17), acoustic and density logs
for this well were converted into a synthetic seismo-
gram (Klang and Gafsou, 1981). A significant and

continuous seismic horizon from the Yafo Formation
was identified in the synthetic seismogram near the
base of the Petah Tiqwa Member in this well.
26. In the Nezarim 1 well (no. 18), Sphaeroidinellopsis
seminulina was not found at the depth interval of 149200 m in this well, probably because of ecological
factors. However, the 5. seminulina zone is tentatively
recognized from the occurrence of G. padana without
G. margaritae (Martinotti, 1981c).
13
The Pliocene-Pleistocene transition in the Iberian Peninsula

EMILIANO AGUIRRE
Introduction
The transition from the late Neogene to the Pleistocene in the
Iberian Peninsula was characterized by widespread erosion on
the Meseta and its margins because of structural deformation
associated with subsidence over wide areas, both along the
Mediterranean shorelands and in the intramontane basins along
the Betic front. Marine deposition was rare and was restricted to
littoral fringes. Volcanic activity in the southeast, and also in the
southern Meseta, decreased after the end of the Miocene. The
paleomagnetic record is extremely limited. However, there are
significant sites in continental environments with both small and
large vertebrates (Figure 13.1) that provide the material for
interregional correlations.
Stratigraphic sequences
Bay of Cadiz
to reversed polarity in the lowest unit (TP-1), which may

correspond to the middle Gilbert chron.
Puerto Real section. The quarries beside the highway near
Puerto Real (3632'20"N; 229'W) offer the most nearly complete section in this area. At the base, a regressive sequence
termed PR-1 (Puerto Real Unit 1) is seen, with offshore sands
changing to offshore marginal calcarenites with Globorotalia
crassaformis', this cycle ends with a shore-edge conglomerate. A
roughly transgressive sequence, PR-2, follows, with minor
oscillations dominated by lagoon and barrier-island facies; it
ends with shallow-marine conglomerates. The upper part is
affected by continental karstification, and that level is followed
by a new cycle, PR-3, with conglomerates, sands, and lagoon
deposits. Within this last unit, in a horizon of yellow sands, a
fragment of a tooth of Mammuthus meridionalis has been found.
The sequence ends in a karstified surface buried by an extensive
regional terrace.
The Bay of Cadiz, in the area between the Guadalquivir

depression and the western margin of the Subbetic realm, was
first inundated during the late Miocene Tortonian transgression.
The subsidence of the bay was related to the advance of the
Subbetic olistostrome to the north-northwest, and its present
configuration was influenced by a normal fault along an eastnortheast axis.
El Aculadero section. The Plio-Pleistocene sequence at El

Aculadero (3634'45"N; 233'50"W) starts with the facies known
as the Ostionero (viz., "conglomerado ostionero" or oysterbearing conglomerate). Common fossils in this basal sequence
are Chlamys varia, Chlamys opercularis, Chlamys glabra, Pecten
jacobaeus, Pecten maximus, Ostrea lamellosa, and Flabellipecten
planomedius. The assemblage is indicative equally of late
Torre del Puerco section. This section (3619'20"N; 229'45"W) Pliocene or early Pleistocene age. The surface of the Ostionero is
begins with marls and marly limestones of the TP-1 (Torre karstified, and pockets are infilled with red marine sands, with
Puerco Unit 1) containing Globorotalia margaritae, overlain by subaerial soils at the top. The section continues upward into
TP-2 marly sands and bioturbated offshore horizons with G. marine-brackish lagoonal silts with interfingering sands (washcrassaformis. Near its top, G. margaritae, G. crassaformis, andover fans) indicating breakup of a barrier island during a
Turborotalia puncticulata disappear, but Globigerinoides extre- transgressive oscillation. The end of that episode is marked by a
mus and Turborotalia humerosa are still present. The TP-3 marine conglomerate. During the following hiatus, a new karst
shallow conglomerates that close this cycle yield abundant was developed, infilled and overlain by a terrace deposit or glacis
rhodoliths of Lithotamniaeae, with Flabellipecten planomedius, containing Early Paleolithic tools of Olduwan tradition (Querol
Pseudoamussium calcatur, Pecten jacobaeus, and Ostrea lamel-and Santonja, 1983, figure 1, A-I; figure 5, LP). The top layer is
losa. The top of the series is continentally karstified, and it fersialithic soil.
disconformably underlies a lower to early middle Pleistocene
terrace with Olduwan-like pebble tools. The paleomagnetic Cantera de la Florida. In the exposures of an abandoned quarry
record shows a gradual shift from reversed to normal and again at La Florida, near El Puerto de Santa Maria (Cadiz), Zazo et al.
169
170
Emiliano Aguirre
CANTABRIC FRINGE
CAMPO DE CALATRAVA
Higueruelas (
MAR MENOR
Cueva Victoria
ALMERIA LfTTORAL
Figure 13.1. Index map of Iberian

Pliocene-Pleistocene study areas.
(1985) identified the paleochannels of a former southern outlet

of the Guadalquivir River, to which the Guadelete was tributary
in early Pleistocene time. The earliest quartzite pebbles deposited by the Guadalquivir in fluviomarine strata at the La Florida
mouth, according to those authors, were very near the base of
the Pleistocene.
between the two assemblages, that is, just below the first
appearance of Globigerina hessi.
Sanlucar de Barrameda and Moguer. At several localities near

Moguer, Huelva, and Sanlucar de Barrameda, Torcal et al.
(1991) have studied the connection between the PliocenePleistocene boundary and climate changes, taking into account
Cerro de los Mdrtires. Diaz, Parra, and Benot (1988) recognized sedimentological evidence and local stratigraphy in the Bay of
the Pliocene-Pleistocene boundary in marine strata cropping out Cadiz. Within the Arenas de Bonares, correlated to the regional
at Cerro de los Martires, south of San Fernanda, Cadiz, where Arenas Rojas or "Red Sands" formation, those authors identialternating yellowish sands and blue clays contain two planktonic fied a lower horizon with illite, smectite, and dominant
foraminifera faunas. The lower fauna is characterized by an calcalkaline feldspar, representing subtidal deposition, and an
assemblage of Globorotalia tosaensis, Globorotalia inflata, upper sandy horizon with dominant potassic feldspar and
Globorotalia crassula crassula, Globorotalia crassula viola, kaolinite, without smectite, which represents an intertidal
Globorotalia ex gr. G. crassaformis, Globorotalia praehirsuta, regime. The Pliocene-Pleistocene boundary is close to the
Globorotalia bononiensis, Turborotalia quinqueloba, Globigerinatransition between these two suites, which could be attributable
uvula, Globigerina apertura, Globigerina ex gr. G. bulloides, to local diastrophism or to a climate change (Torcal et al., 1991),
Globigerinoides conglobatus, Globigerinoides elongatus, Globige-so that the epoch boundary could in fact be located within the
rinoides obliquus obliquus, Globigerinoides obliquus extremus upper unit, near its base.
(very rare), Globigerinoides ruber, Neogloboquadrina acostaensis, Neogloboquadrina humerosa, Neogloboquadrinapachyderma Summary of major features. In the Cadiz area, sedimentation
(dextral), and Globorotalia truncatulinoides. The upper fauna was discontinuous during the time in question, and a hiatus in
includes Globigerina hessi, Globigerina aff. G. calida, Globiger-deposition is almost universal in the upper part of the sequence.
ina aff. G. digitata, Globigerinoides tenellus, NeogloboquadrinaThe Upper Pliocene (Piacenzian) probably is represented by the
dutertrei, and sinistrally coiled Neogloboquadrina pachy derma. lower Ostionero and the succeeding erosional interval. The
The Pliocene-Pleistocene boundary is correlated to the transition overlying Arenas Rojas is known all along the Atlantic shore of
Plio-Pleistocene of the Iberian Peninsula
the southern Iberian Peninsula, from Algarve, Portugal, to

Algeciras on the Mediterranean side of the Gibraltar Strait. The
formation may represent estuarine bars developed at the same
time as some of the lower Ostionero; it should be noted that on
those estuarine sands a primary red soil is developed. The
succeeding transgressive marine sediments, including correlatives such as the TP-4 and PR-2 (and, locally, later Ostionerofacies deposits), probably are early Pleistocene in age.
A final regressive episode, marked by the deposition and
karstification of the lower conglomerate at El Aculadero and the
upper conglomerate at Puerto Real, represents another major
feature. I suggest, with some reservation, that those terminal
processes belonged to the general lowering of sea level and the
severe climate changes that marked the early middle Pleistocene
(i.e., Menapian) glacial maxima. A change from tensional to
compressive style in regional tectonism is conspicuous near the
top of the lower Ostionero beds. The new compressive phase,
which has lasted to the present, is characterized by strike-slip
faults, both right- and left-lateral (Benkhelil, 1976; Zazo et al.,
1977; Zazo, 1989).
The evidence strongly suggests that the Pliocene-Pleistocene
boundary, as defined in the Vrica section (Aguirre and Pasini,
1985), can be correlated to a lowering of the sea level that is
reflected in the karstic erosion of the late Pliocene Ostionero
Conglomerate and in the change from shallow-marine to
estuarine facies in a number of localities. In other locations, the
Pliocene-Pleistocene boundary appears to be correlated to a
level within the Arenas Rojas, the base of which is timetransgressive in the Bay of Cadiz region from Huelva to
Algeciras (Zazo, 1989).
171
sea level) ranging from +65 m in Alicante to +90 m in Almeria.

According to Goy et al. (1992), the ages of these highest terraces
appear to be early Pleistocene. They are preceded either by
regressive Pliocene marine deposits or by major unconformities
of tectonic origin. An exception is the Torre vie ja-Mar Menor
district, where terraces are not evident because of continuous
subsidence throughout the Pleistocene.
The Spanish Meseta
The best correlations between the Pliocene-Pleistocene
boundary-stratotype at Vrica and the continental strata of Spain
are found in the rana terrace formations on the margins of the
Meseta, in the Orce deposits of the Baza Basin, and (potentially)
in the cave fillings of Casablanca. In those deposits, paleomagnetic and paleontologic evidence and rare radiometric age
determinations have been integrated in a framework of regional
geodynamic history.
La Rana. The local lithofacies known as La Rana is a cobble

piedmont sheet with pseudogley soil. Although the perfect
synchrony of all rana sheets can be debated, it is currently agreed
that the rana lithology is related to neotectonic uplifts that ended
basinal sedimentation in the Meseta region (Perez Gonzalez and
Gallardo, 1987). Tentatively, La Rana is dated older than 2 Ma;
it is followed in the Meseta by two or more erosional events
marked by terraces at +200 and +180 m ASL, probably of latest
Pliocene age. Following the incision of these transitional forms,
the drainage of the Meseta region to the ocean began, and the
graded river-valley terraces started to develop. The number of
terrace levels attributed to the early Pleistocene in the major
valleys
of the Meseta is four to six, starting with landforms at
Mediterranean shorelines
elevations of +155 m above present grade; the next lowest, a
The Cantabric fringe. According to Hoyos Gomez (1989), the terrace of +80-70 m, is usually judged to be pre-Cromerian.
Rana deposits occur in separated areas of the Spanish Meseta,
transition from Pliocene to Pleistocene in the Cantabric coastal
always
in marginal areas bordering the neighboring mountain
region was linked to the development of the extensive erosion
ranges.
According to Molina (1975), in the Campo de Calatrava
surface known as the Rasa. The main erosional phase took place
of
the
Guadiana
River basin the middle Pliocene (S-1) surface
in the later Pliocene, as indicated by the fact that the final surface
was
moderately
deformed
prior to the development of a new
of the Rasa is closely echoed in the earliest Pleistocene drainage
surface
at
lower
elevations
during
a period of warm climate with
and seems to have lasted through most of the early Matuyama,
between 2.6 and 2.0 Ma. After formation of the Rasa surface, marked seasonality. That new surface was covered by rana
the initial incision of the present river systems and the materials laid down in a series of anastomosing channels under
development of a karstic erosion surface, with abundant decalcifi- even more pronounced seasonal climate conditions, as indicated
cation clays and stream deposits in caves, date from a period by caliche crust formed on the deposit surface. That surface,
clearly antecedent to the severe climate shifts that began in the named S-2 by Molina (1975), was then dissected, and the new
middle Pleistocene. Hoyos Gomez (1989) concluded that the landscape was covered by glacis (wind-sorted gravel terraces),
Pliocene-Pleistocene boundary cannot be traced closer than this considered by Molina (1975) to represent the earliest deposits of
very approximate indication, because of the absence of more the Pleistocene. The V-III eruptive episode in the Calatrava
volcanic field appears to have begun during the rana deposition
precise dating.
and caliche formation, but after reaching a maximum during the
dissection of the S-2 surface the eruptions ceased before
Iberian coastline. In the littoral of the east coast of the Iberian deposition of Lower Pleistocene alluvial sediments (Lopez-Ruiz
Peninsula and in the Balearic archipelago, the highest well- et al., 1993). In the volcanic edifice of Cabezo de Segura,
preserved marine terraces are at elevations above MSL (mean magnetostratigraphic studies by Calvo Sorando et al. (1992) have
Emiliano Aguirre
172
shown that a deposit correlative to the local rana is interbedded

between two volcanic layers with reversed remanent polarity and
that the lower volcanic layer rests on a surface that corresponds
to the upper "polygenetic surface" of the Meseta and the Iberic
Chain.
A geomorphologic analysis of rana formations along the
southern margin of the Sierra de Somo and the Sierra de Ayllon
in the Cordillera Central led Perez Gonzales and Gallardo (1987)
to conclude that repeated rana erosional and depositional
processes followed the Ibero-manchega II uplift and shortly
preceded the earliest (highest) valleys of the Jarama and Sorbe
river systems. Perez Gonzales and Gallardo (1987) offered no
conclusion as to the relationship of the rana to climate change,
but in my opinion the concentration of quartzite cobbles in the
rana deposits is best explained as the result of active gelifraction
on tectonized quartzites in the newly uplifted sierras. According
to M. Hoyos Gomez (personal communication, 1991), the
general mode of rana deposition is in the form of a regional
glacis, not as debris cones or alluvial fans; this is logically
explained as a reflection of specific climate conditions, as
suggested by Molina (1975), and not as the result of tectonism
alone.
After accounting for the effects of local diastrophism, the rana
deposits described in the two studies just cited would appear to
be products of arid, highly seasonal climates in which soil
formation and fluvial erosion were less active, and deflation was
more active, than in prior or following regimes. The conditions
for rana formation ended with renewed erosion and indications
of a moister climate, which could be related to global cooling.
Such events at about 2.1 Ma, and at about 1.8 Ma (the end of the
Olduvai normal-polarity subchron), may have been implicated,
and a correlation with the Aullan unconformity of Tuscany
(Italy) is thus a tentative possibility. In this hypothesis, La Rana
of Calatrava and the foothills of the Somo and Ayllon sierras
would mark the very end of the Pliocene, just prior to the
There is no evidence that ranas in other areas on the Meseta
borderlands differ significantly in age from the foregoing. The
only difference is that these accretion surfaces were built up with
materials other than quartzite cobbles, but that must be expected
in areas where the source rocks were different. Correlation of
ranas deserves detailed study as a means of tracing the PliocenePleistocene boundary in a large area of central Spain.
The Mediterranean shoreline

In the Mediterranean littoral of the Iberian Peninsula and in the
Balearic Islands the uppermost traces of sea level are found at
+65 m ASL in Alicante and Murcia, and at +90 m ASL in
Almeria. These beach lines represent early Pleistocene sea
levels, being younger than regressive deposits and erosion of
Pliocene age. In the Torre vie ja-Mar Menor basin, on the other
hand, Lower Pleistocene deposits were laid down under conditions of continuing subsidence (Goy et al., 1992).
Radiometric dates
The stratigraphic sections studied in the Iberian Peninsula for the

purposes of contributing to IGCP Project 41 are not well
documented with radiometric dates. Among the very few datable
volcanic layers that could be of interest for the period in question
are the San Isidro Basalt (2.8-2.6 Ma) at Rambla de la Auia and
Venta del Lirioh near Cartagena, Murcia province, and the
Hostalrich (3.08 Ma) and San Corneli (1.99 Ma) basalts at La
Selva, Gerona province. Unfortunately, the stratigraphic positions of these dated rocks in the respective regional sequences
are as yet quite uncertain.
In Campo de Calatrava, a region of late Cenozoic basaltic
volcanism in Ciudad Real province, the Higueruelas mammal
site yields remains of Anancus arvernensis, Hipparion rocinantis,
Cervus cf. C. cusanus, Cervus cf. C. perrieri, and Gazella cf. G.
borbonica, in a fluvial deposit that includes coarse volcanic
debris. No dates have been obtained for the co-deposited
volcanic fragments, but the fauna corresponds to zone MN-16,
with approximate age limits of 3.5 and 2.6 Ma. In support of this
estimate, basalt flows that unconformably overlie the Higueruelas beds have been given a K-Ar age of 2.3 0.4 Ma (R. E.
Drake and J. A. Van Couvering, personal communication).
More recently, M. Hoyos Gomez (personal communication,
1992) dated the Higueruelas beds to the Gauss chron, according
to a radiometric age of about 3 Ma on the overlying lavas,
accompanied by paleomagnetic analyses that indicated normal
polarity in the fossil bed and reversed polarity in strata
immediately below. As noted earlier, Molina (1975) and LopezRuiz et al. (1993) concluded that these Upper Pliocene Calatrava
volcanics had been erupted during the development of the
regional erosion surface that is capped by the terminal Pliocene
rana deposits, whereas Calvo Sorando et al. (1992) have
suggested that some of these volcanics were of the same age as
the rana formation. Evidently, the reduction of the landscape
and the gradual accumulation of rana carapace were parts of a
long-term geomorphological process that in the Calatrava region
was periodically interrupted by volcanism.
Plio-Pleistocene mammalian faunas
Elephants and horses appear for the first time in the strata of the
Jucar River valley at levels corresponding to the small-mammal
horizons Valdeganga I and II (Mein, Moissenet, and True, 1978).
Equus stenonis and Mammuthus ex gr. meridionalis are in this
site associated with Mimomys medasensis, Mimomys capettai,
Stephanomys balcellsi, Castillomys crusafonti crusafonti, and Sus
strozzii, an assemblage which corresponds to MN-16 age (i.e.,
from 3.5 to 2.6 Ma). Rincon 1 appears to be similar in age, with
Oryctolagus sp., Eliomys intermedius, Mimomys capettai, Castillomys crusafonti crusafonti, Nyctereutes megamastoides, Pachycrocuta sp. (of larger size), Equus stenonis, Dicerorhinus sp.,
Gazella borbonica, and other artiodactyl genera. Paleotemperature determinations and correlation of the Rincon section with
the normally magnetized lower third of the red clays and
Age Magnetic
Ma scale
0.5
0.6
0.7
0.8
i
I
0.9
1
1.1
>
V
Subsidence in
Baetics & other
areas
Tlting in Meseta
1.5
1.9
2
2.1
2.2
2.3
2.4
Aa
2.5
G ^H
3.1
3.2
3.3
3.4
3.5
M I
G H
Gi I J
Pirro
D. Altenburg 4
SEINZELLES
Cortijo Don Alfonso

Barranco Conejos
CERRO
DEL PALO
Lacustrine in
Tectonic
reactiV Southern Meseta
Humid
vation
^
Warm
Iberomanchega
phase 1
V
High SL
V
Tectonic activ.
Boulder Conglomerate (Panjab)
Microtus &
Bison events
V
Megacehni, Cams
Hippopotamus
events
Tectonic activ.
Almenara
Valdeganga IV
0RCE2
CHILHAC
Valdeganga III
COUPET
Puebla de
SENEZE
Valverde
ST-VALLIER
Fuentes Nuevas 1
V
Allophaiomys
event
Land
V
Bridges
\y
Erosion surfaces
SL lowering
Erosion in
highlands
Uplift
Huelago-C
Escorihuela
RINCON 1
Valdeganga 1,2
RINCON 2, 3
ROCCANEYRA
VIALETTE
oL Tail
Kvabebi
Triversa
Tectonic
activity
a
b
TORRE DEL
PUERCO III
sandstones of Villalgordo suggest a maximum age in the very

latest Gauss chron (Alberdi et al., 1982) for Rincon 1 and
Valdeganga III faunas, in the later part of MN-16 (Figure 13.2).
A later regional fauna, of MN-17 age, is typified in Spain by
anama
Land Bridge
VillalbaAlta
LAYNA
Rising SL
Mammuthus
EquusFSD
Villarroya
HIGUERUELAS
V
Cold in highland
High SL warm
Wide-range
events
Olivola
n
n +
y
Lacustrine
+ fluviatile
deposition
Venta Micena 1,2
Incarcal
Dissection of
major present
Tectonic
valleys
reactivation
v Penetrative
CABEZO surface
SEGURA RANA
Iberoman- .
chega
^ RANA
3
phase 2
olygenetic
surface
LowSL
V
Severe erosion
most obvious
n highlands
Tlting
CERRO
PELADO Peneplanation
Sussenborn
ATAPUERCA TD6
West-Runton
TD5
TD4
ATAPUERCA TD3
Huescar
Lachar
SOULHAC
VALLONNET
Bagur 2
Cueva Victoria
1.8
2.9
3
River dissection
terraces
Continental uplift
1.4
2.8
Tectonic
reactivation
1.3
2.6
2.7
SL raise
^ 3 LowSL
2> o
SL raise
w
O H
1.2
1.6
1.7
Correlated
sites
Tectonic
Volca D e P s i t s Climate Mammal sites
Marine sites
processes nism Landforms
173
Tectonic
activity
Figure 13.2. Correlation of main

vertebrate fossil localities and geologic features of Spain, from the
Lower Pliocene (lower Gauss) to
the Middle Pleistocene (Matuyama-Brunhes boundary).
the Puebla de Valverde local fauna (Heintz, 1978). That site, in

Teruel province, yields Paradolichopithecus arvernensis, Macaca
sp., Nyctereutes megamastoides, Vulpes alopecoides, Ursus etruscus, Hyaena perrieri, Chasmaporthetes lunensis, Megantereon
174
Emiliano Aguirre
megantereon, Acinonyx pardinensis schaubi, Lynx issiodorensis, gombaszoegensis, Dicerorhinus hemitoechus, Equus cf. E.
Mammuthus meridionalis, Dicerorhinus etruscus, Equus stenonisstehlini, Megacerini, "Cervus" sp., and Bison schoetensacki, an
vireti, Croizetoceros ramosus, "Cervus"philisi, Eucladocerossene-assemblage similar to that of the lower Galerian of Italy. Near
zensis, Gazella borbonica, Gazellospira torticornis, and Gallo- the base of this fossiliferous sequence, a geomagnetic-polarity
goral meneghini. This is a typical Middle Villafranchian assem- reversal identified as the Matuyama-Brunhes transition has
blage (Heintz, 1978), or later Villanyan, in the modern usage. recently been identified (Garacedo, Soler, and Chicharro,
In the upper horizons of the Jucar canyon, Albacete province, personal communication, 1991).
the Valdeganga III site yields Mimomys aff. M. medasensis,
Mimomys rex, Stephanomys progressus, Castillomys crusafonti
Plio-Pleistocene faunas of the Betic depressions
subsp. indet., Micromys aff. M. minutus, Apodemus aff. A.
dominans, Eliomys aff. E. quercinus, Parapetenya hibbardi, The Guadix and Baza basins are linked tectonic depressions
Sorex subminutus, Desmana nehringi, Prolagus calpensis, presently drained by the Guadalquivir River. In late Neogene
Oryctolagus aff. O. laynensis, and Equus stenonis subsp. indet. and early Pleistocene time they were closed endorheic basins
(Mein et al., 1978). The assemblage appears to correlate best to thatfilledwith lacustrine deposits. Large and small mammals are
the characteristic later Villanyan faunas of Villany 3 and Kislang. found throughout the sequence, and in the Baza Basin in
The karst filling in the Casablanca quarry, near Almenara, particular they constitute an almost continuous record from the
Castellon province, has yielded abundant large- and small- early Pliocene to the early middle Pleistocene.
mammal fossils, including Prolagus calpensis, Desmana inflata, In the time span relevant to the Pliocene-Pleistocene boundMyotis cf. M. myotis, Miniopterus aff. M. schreibersi, Rhinolo- ary, the sites with the richest fossil records are (in ascending
phus cf. R. mehelyi, Eliomys sp., Stephanomys progressus, chronostratigraphic order) as follows: Carretera de Huelago
Apodemus cf. A. mystacinus, Apodemus aff. A. occitanus, (correlative to Montopoli and Etouaires); Fuentes Nuevas 1
Castillomys crusafonti subsp. indet., Mimomys tornensis, M. (correlative to Puebla de Valverde, in Teruel province, and to
medasensis, Mimomys aff. M. rex, Ursus etruscus, PachycrocutaSaint-Vallier) (Aguirre et al., Chapter 9, this volume); Orce D
brevirostris, Felis sp. indet., Dicerorhinus cf. D. etruscus, Equusand Orce 1 (correlative to Seneze, Chilhac, and Le Coupet); Orce
stenonis subsp. indet., Cervus philisi, Gazellospira torticornis, 2 (correlative to Olivola, Montousse, and other sites placed in the
and Ovibovini gen. indet. (Soto and Morales, 1985; Esteban lower Mimomys ostramosensis zone of Chaline, Chapter 14, this
Aenlle and Lopez Martinez, 1987). The rodent assemblage is volume); Barranco de los Conejos (correlative to sites in the
comparable to that of Villany 5 and Kadzielna, and rather more upper ostramosensis zone), which Agusti, Moya-Sola, and Ponsto the latter, while the large-mammal assemblage is closely Moya (1987b) have equated with Casa Frata; Venta Micena 1 and
correlative to that of Olivola.
2 (correlative to early Biharian faunas at Betfia 2, Betfia 13, and
The MN-17 mammal assemblages are followed by a new Altenburg 2); Barranca Leon, Fuentes Nuevas 3, and Canada de
faunal complex, characterized in Bagur 2 karst filling by Murcia 1 (correlative to levels slightly younger than Venta
Allophaiomys pliocaenicus, Lagurus pannonicus, Pliomys episco- Micena); Huescar 2 and 3 and Puerto Lobo 1 (correlative to Le
palis, Ungaromys sp., Castillomys crusafonti, Apodemus aff. A. Vallonet on the basis of Allophaiomys nutiensis); Huescar 1
mystacinus, Apodemus aff. A. sylvaticus, Eliomys aff. E. (correlative to Atapuerca TD3-TD6, with small-mammal faunas
intermedius, Prolagus cf. P. calpensis, and Oryctolagus cf. O. assigned to the zone of Pitymys gregaloides and Mimomys savini);
lacosti (Lopez Martinez, Michaux, and Villalta, 1976). Allo- Cullar de Baza (correlative to later Cromerian and later Galerian
phaiomys is also recognized in younger horizons of the Baza on the basis of Arvicola cantiana and Mammuthus trogontherii).
basin and in the Venta Micena local fauna discussed later.
Also to be mentioned are the Cueva Victoria fauna from La
A very extensive detritic formation in the central basin of the Union, near Cartagena, which is slightly older than the Venta
Rio Guadiana is found at the present river level, east and west of Micena fauna and correlates to the Sinzelles faunal level (1.3 to
Ciudad Real. From a stock-pond excavation, this deposit yielded 1.4 Ma), and Lachar in the Granada Basin, which is well
Mammuthus meridionalis, Eucladoceros dicranios, Leptobos sp., correlated to the Solilhac faunal level.
Hippopotamus major, and Equus mosbachensis. This association
The Plio-Pleistocene faunal succession of the Betic basins
can be considered transitional to the early middle Pleistocene, or begins with Carretera de Huelago, from which Alberdi and
Cromerian-Galerian faunas. The identification of Mammuthus Bonnadonna (1989) reported Mammuthus meridionalis, Equus
meridionalis tamanensis at a quarry on a terrace near Fuensanta, stenonis livenzovensis, Dicerorhinus cf. D. etruscus, Leptobos cf
in the Jucar Valley, together with Hippopotamus major (Aguirre, L. elatus, Gazella borbonica, Gazellospira torticornis, Ovibovini
1989b), conveys the possibility of recognizing the Tamanian stage cf. Hesperoceros merlae, Croizetoceros ramosus, Eucladoceros
in Spain.
cf. E. senezensis, Cervidae gen. indet., Mimomys cf. M. reidi,
The lower half of the known cave-fill section at Atapuerca and Stephanomys sp.
Burgos (Atapuerca I of Aguirre et al., 1990) yields Mimomys
The Fuentes Nuevas 1 site yields Mimomys cf. M. reidi,
savini, Pliomys episcopalis, and Pity mys gregaloides. Atapuerca Castillomys crusafonti, Apodemus cf. A. dominans, Equus
I also yields the large mammals Crocuta intermedia, Panthera stenonis aff. E. s. vireti, and Gazella borbonica. The two teeth
175
and metatarsal of the equid from Fuentes Nuevas resemble most north, and it is possible that this district is beyond the
closely those of the variety found at Puebla de Valverde and paleogeographic limit for exact correlation of the PlioceneSaint-Vallier.
Pleistocene boundary on the basis of central European microtine
At Orce 2, the fossil list includes Drepanosorex sp., Mimomys evolution.
ostramosensis, Mimomys pusillus, Castillomys crusafonti, Micro- Outside of the Guadix-Baza basin, but also in the southeasttus (cf. Allophaiomys) sp., Apodemus aff. A. sylvaticus, ern part of Spain, the site of Cueva Victoria is a cave complex
Apodemus mystacinus, Eliomys aff. E. quercinus, Galemys that was completely filled with hyena debris in the early
kormosi, Gazellospira torticornis, and Leptobos etruscus. The Pleistocene. Most of the infilling bone-breccia was later eroded,
co-occurrence of the last two taxa is known elsewhere only at but the remaining material includes remains of Crocidura sp.,
Olivola. Sediments at Orce 2 show normal paleomagnetic Erinaceus sp., Prolagus calpensis, Oryctolagus cf. O. lacosti,
polarity (Agusti et al., 1987a). An assemblage similar to Orce 2 Myotis sp., Rhinolophus euryhale, Rhinolophus cf. R. mehelyi,
is recorded at Barranco de los Conejos from beds 8a and 9, Miniopterus sp., Allophaiomys chalinei, Eliomys quercinus,
corresponding to the lower part of unit b in the informal notation Apodemus mystacinus, Vulpes sp., Canis etruscus, Equus
of Anadon etal. (1987).
stenonis, Equus sp., Mammuthus meridionalis, Dolichodoriceros
The extensive outcrops at Venta Micena 1 and 2 yield fossils savini, "Cervus" elaphoides, Hemitragus sp., Ovibovini gen.
from stratigraphic unit c of Anadon et al. (1987). As described by indet., Caprini gen. indet., Bovini gen. indet., and Papionini cf.
Agusti et al. (1987a), the list from Venta Micena 2 includes Papio sp. The presence of Homo sp. has been reported from
Desmana sp., Microtus (Allophaiomys) pliocaenicus, Apodemus Cueva Victoria, and from Venta Micena as well, but identificaaff A. mystacinus, Castillomys crusafonti, Eliomys intermedius, tion remains doubtful (Gibert-Clols et al., 1989). Cueva VictoHystrix major, Prolagus calpensis, Oryctolagus cf. O. lacosti, ria, on present evidence, is best correlated to Sinzelles; Venta
Micena is not much younger.
Ursus etruscus, Canis etruscus mosbachensis, Vulpes preglacialis,
Cuonpriscus, Xenocyonsp., Homotherium latidens, Megantereon The fauna at Huescar 1, in the upper part of the Baza Basin
cultridens adroveri, Pachycrocuta brevirostris, Panthera cf. P. sequence, includes many bird fossils. Mammals listed by Alberdi
gombaszoegensis, Lynx sp., Meles sp., Mammuthus (nee Ar- et al. (1989) are Soricidae gen. indet., Eliomys quercinus,
chidiskodon) meridionalis, Equus stenonis granatensis, Dicerorhi-Apodemus sp., Castillomys crusafonti, Mimomys savini, Pity mys
nus etruscus brachicephalus, Hippopotamus incognitus, Praemegagregaloides, Microtus brecciensis, Oryctolagus sp., Lepus cf. L.
ceros solilhacus, "Cervus" elaphoides, Praeovibos sp., Capra granatensis, Canis etruscus, Panthera gombaszoegensis, Homoalba, Soergelia minor, Bison sp., Caprini gen. indet., Testudo sp., therium sp., Elephas (Palaeoloxodon) antiquus, Equus stenonis,
and Lacerta sp.
Equus suessenbornensis, Dicerorhinus etruscus brachycephalus,
Barranca Leon has yielded fossils from at least three levels: L2 Hippopotamus major, and Praemegaceros cf. P. solilhacus. This
and L3, in unit d of Anadon et al. (1987), and LI, which is in assemblage is comparable to the early Galerian, or, in other
upper d and lower e. Another site in the lower part of unit e is the words, close to the older faunas at Atapuerca, and thus to the
rich horizon of Orce 7. The fossils from all these levels may be transition between early and middle Pleistocene. The fauna at
regarded as a single assemblage consisting of Microtus Lachar, in the Granada Basin, is a slightly older assemblage and
(Allophaiomys) pliocaenicus, Apodemus mystacinus, Apodemus includes Mammuthus meridionalis, Equus stenonis granatensis,
aff. A. sylvaticus, Castillomys crusafonti, Eliomys intermedius, Dicerorhinus etruscus, Bison sp., Bovidae gen. indet., PraemegaHippopotamus sp., "Cervus" elaphoides, Bison sp., Capra alba, ceros sp., Capreolus sp., and Dama cf. D. clactoniana (Aguirre,
Soergelia minor, Equus stenonis, and Mammuthus meridionalis. 1989b; B. Azanza and B. Sanchez, personal communication,
Orce 3, at a slightly higher stratigraphic horizon, is relatively 1991). The Lachar fauna is very similar to that of Solilhac, which
poor, but it includes a Galemys sp. and a Mimomys aff. M. is to say close in time to the transition from the terminal early
Pleistocene paleofauna, such as Vallonet, to earliest Cromerian
savini.
An unresolved question is whether or not Mimomys pusillus and Galerian assemblages.
has been identified correctly from Orce 2, in unit b, and the
In sum, the abundance of rich fossil sites, representing in one
correlative site of Valdeganga IV. If so, its appearance is stratigraphic sequence almost every recognized faunal event
anomalous with regard to the first occurrence of Microtus straddling the Pliocene-Pleistocene boundary, suggests the value
(Allophaiomys) pliocaenicus in Venta Micena at the lower of the Guadix-Baza basin for illustrating the Plioceneboundary of unit c. In regions to the north of the Pyrenees, the Pleistocene transition in the Iberian Peninsula and the ease with
appearance of M. pusillus is delayed until after the first which this biostratigraphic succession can be correlated in the
appearance of M. pliocaenicus, at a level that is recognized as regional synthesis. I therefore expressly propose here that the
equivalent to the top of the Olduvai normal-polarity subchron basin-filling sequence of the Guadix-Baza basin should be
(Chaline, Chapter 14, this volume). The apparent presence of adopted as a parastratotype area for the Pliocene-Pleistocene
M. pusillus in normally polarized (Olduvai?) strata at an older boundary in continental sediments, with the expectation that this
level than M. pliocaenicus indicates that one or the other had a will encourage stratigraphers working in this region to identify
different time range in southern Spain than in the regions farther and propose a formal reference section.
Emiliano Aguirre
176
Cuaternario, No. 11. Madrid: Consejo Superior de Investigaciones Cientificas.

This chapter represents the final report of the Spanish working Anadon, P., Julia, R., De Dekker, P., Rosso, J.-C, and SoulieMarsche, I. 1987. Contribution a la paleolimnologia del
group on the Pliocene-Pleistocene boundary. In addition to the
Pleistoceno inferior de la Cuenca de Baza (sector Orceauthor, the members are as follows: I. Agusti, M. T. Alberdi, C.
Venta Micena). In Geologia y paleontologia del Pleistoceno
Arias, J. Bech, G. Bigazzi, F. P. Bonnadonna, J. Civis, C.
inferior de Venta Micena, ed. Moya-Sola et al., pp. 35-72.
Paleontologia i Evolucio, Memoria especial, no. 1.
Dabrio, J. Gilbert, E. Gil, J. L. Goy, G. Leone, N. Lopez, A.
Maldonado, E. Molina, J. Morales, S. Moya-Sola, A. PerezBenkhelil, J. 1976. Etude neotectonique de la termination
Gonzalez, J. de Porta, M. Ramos, F. Robles, A. Ruiz Bustos, J.
occidentale des Cordilleres Betiques (Espagne). These 3e.
A. Santos, C. Sese, N. Sole, D. Soria, E. Soto, T. Torres, and C.
cycle (unpublished), Universite de Nice.
Zazo. I am indebted to my colleagues in this working group for Calvo Sorando, J. P., Hoyos Gomez, M., Morales Romero, J.,
Ordonez Delgado, S., et al. 1992. Estratigrafia, sediassistance in the preparation of this summary, and to Dr. Caridad
mentologia y materias primas minerales de Neogeno de la
Zazo in particular.
Cuenca de Madrid. In /// Congreso geologico de Espana y
VIII Congreso latinoamericano de geologia, Salamanca,
1992. Excursiones, pp. 139-179. Madrid: ENADIMSA.
References
Diaz, M. G., Parra, A., and Benot, C. 1988. Datos preliminares
Aguirre, E. 1989a. El limite inferior del Pleistocene In Map a del
sobre materiales Plio-Pleistocenicos en San Fernando
Cuaternario de Espaha. Escalalscale 1:1.000.000, ed. E.
(Cadiz). In TV Jornadas de Paleontologia. Resumenes de
Llorente, pp. 87-94. Madrid: Instituto Tecnologico Geolas Comunicaciones, p. 37. Salamanca: Universidad de
minera de Espana.
Salamanca.
Aguirre, E. 1989b. Vertebrados del Pleistoceno continental. In Esteban Aenlle, J., and Lopez Martinez, N. 1987. Les
Mapa del Cuaternario de Espana. Escalalscale 1:1.000.000,
Arvicolides (Rod., Mamm.) du Villanyen recent de
ed. E. Llorente, pp. 47-68. Madrid: Instituto Tecnologico
Casablanca I (Castellon, Espagne). Geobios 20:591-623.
Gibert-Clols, J., Campillo-Valero, D., and Garcia-Olivares, E.
Aguirre, E., Arsuaga, J. L., Bermudez de Castro, J. M.,
(eds.) 1989. Los restos humanos de Orce y Cueva Victoria.
Carbonell, E., et al. 1990. The Atapuerca sites and their
Goy, J. L., Zazo, C , Somoza, L., Dabrio, C. J., Bardaji, T., and
hominids. Human Evolution 5:55-73.
Silva, P. 1992. Evolution of Quaternary shorelines in the
area connecting the Atlantic and Mediterranean. In
boundary. Episodes 8(2):116-120.
Impacts of tectonics on Quaternary coastal evolution, ed.
Agusti, J., Moya-Sola, S., Martin-Suarez, E., and Marin, M.
Y. Ota. Quaternary International, vols. 15-16, special
1987a. Faunas de mamiferos en el Pleistoceno inferior de
issue.
la region de Orce (Granada, Espana). In Geologia y
paleontologia del Pleistoceno inferior de Venta Micena, ed.Heintz, E. 1978. La fauna villafranchienne de La Puebla de
Valverde, Teruel, Espagne. Composition cualitative et
Moya-Sola et al., pp. 73-84. Paleontologia i Evolucio,
Memoria especial, no. 1. Sabadell: Institut Paleontologic
quantitative. Geol. Medit. 5:227-280.
Dr. M. Crusafont.
Hoyos Gomez, M. 1989. La Cornisa Cantabrica. In Mapa del
Cuaternario de Espaha. Escalalscale 1:1.000.000, ed. E.
Agusti, J., Moya-Sola, S., and Pons-Moya, J. 1987b. La sucesion
Llorente, pp. 105-118. Madrid: Instituto Tecnologico
de mamiferos en el Pleistoceno inferior de Europa:
proposition de una nueva escala biostratigrafica. In
Geologia y paleontologia del Pleistoceno inferior de VentaLopez Martinez, N., Michaux, J., and Villalta, J. F. 1976.
Rongeurs et Lagomorphes de Bagur-2 (Province de
Micena, ed. Moya-Sola et al., pp. 287-295. Paleontologia i
Girone, Espagne). Nouveau remplissage de fissure du
Evolucio, Memoria especial, no. 1. Sabadell: Institut
debut du Pleistocene moyen. Acta Geol. Hispan. 11:
Paleontologic Dr. M. Crusafont.
46-54.
Alberdi, M. T., Alcala, L., Azanza, B., Cerdeno, E., Mazo,
A. V., Morales, J., and Sese, C. 1989. Consideraciones Lopez-Ruiz, J., Cebria, J. M., Doblas, M., Oyarzun, R.,
Hoyos, M., and Martin, C. 1993. The late Cenozoic
biostratigraficas sobre la fauna de vertebrados fosiles de la
alkaline volcanism of the central Iberian Peninsula
cuenca de Guadix-Baza. In Geologia y paleontologia de la
(Calatrava Volcanic Province, Spain): intraplate vulcuenca de Guadix-Baza, ed. M. T. Alberdi and F. P.
canism related to extensional tectonics. /. Geol. Soc.
Bonnadonna, pp. 347-355. Trabajos sobre NeogenoLondon 150:915-922.
Cuaternario, No. 11. Madrid: Consejo Superior de InMartin-Serrano, A., and Molina, E. 1989. Montes de Toledo y
vestigaciones Cientificas.
Extramadura. In Mapa del Cuaternario de Espana. Escalal
Alberdi, M. T., Arias, C , Bigazzi, G., Bonadonna, F. P., Leone,
G., Lopez, N., Michaux, J., Morales, J., Robles, R, and
scale 1:1.000.000, ed. E. Llorente, pp. 182-200. Madrid:
Soria, D. 1982. Nueva yacimiento de moluscos y verteInstituto Tecnologico Geominera de Espana.
brados del Villafranquiense de la Cuenca del Jucar Mein, P., Moissenet, E., and True, G. 1978. Les formations
(Albacete, Espana). In Actes de la Colloqium "Le
continentales du Neogene superieur des vallees du Jucar et
Villafranchien mediterraneen," Lille, 9-10 Dec. 1982, vol.
du Cabriel au NE d'Albacete (Espagne): biostratigraphie
1, pp. 255-271. Lille Univ., Fac. Sci. (CNRS).
et environnement. Lab. Geol. Fac. Sci. Lyon, Docum.
72:99-147.
Alberdi, M. T., and Bonnadonna, F. P. 1989. La fauna de
mamiferos del Plioceno superior de la Carretera de Molina, E. 1975. Estudio del Terciario superior y del Cuaternario
del Campo de Calatrava (Ciudad Real). Trabajos sobre
Huelago, Cuenca de Baza. In Geologia y paleontologia de
Neogeno-Cuaternario, No. 3. Madrid: Consejo Superior
la cuenca de Guadix-Baza, ed. M. T. Alberdi and F P.
de Investigaciones Cientificas.
Bonnadonna, pp. 117-133. Trabajos sobre NeogenoAcknowledgments
177
Perez Gonzalez, A., and Gallardo, J. 1987. La Rana al sur de Zazo, C. 1989. Los depositos marinos cuaternarios en el Golfo
Somosierra y Sierra de Ay lion: un piedmonte escalonado
de Cadiz. AEQUA Monografia 1:113-122.
del Villafranquiense medio. Geogaceta (Madrid) 2:29-32. Zazo, C , Goy, J. L., Dabrio, C , Civis, J., and Baena, J. 1985.
Paleogeografia de la desembocadura del Guadalquivir al
Querol, M. A., and Santonja, M. 1983. Elyacimiento de cantos
comienzo del Cuaternario (provincia de Cadiz, Espana).
trabajados de El Aculadero (Puerto de Santa Maria,
In Actas i reunido do Quaternario Iberico, vol. 1, ed. M.
Cadiz). Excavaciones Arqueologicas en Espana, no. 130.
Ramos, pp. 461-472. Lisboa: Instituto Nacional de InMadrid: Ministerio de Cultura.
vestigagao Cientifica.
Soto, E., and Morales, J. 1985. Grandes mamiferos del
yacimiento villafranquiense de Casablanca I, Almenara Zazo, C , Goy, J. L., Hoyos, M., Meco, J., User a, J., Garcia
(Castellon). Estudios geol. 41:243-249.
Vicente, J., Galvan, I , and Aguirre, E. 1977. El corte de
Puerto Real y el problema del limite Plio-Pleistoceno en
Torcal, L., Marfil, R., and Zazo, C. 1991. El transito Pliola Bahia de Cadiz. Trabajas sobre Neogeno-Cuaternario,
Pleistoceno en el extremo occidental de la provincia de
No. 6, pp. 319-336. Madrid: Consejo Superior de InHuelva (Espana). Datos petrologicos y mineralogia de arvestigaciones Cientificas.
cillas. R. Soc. Esp. Hist. Nat. (Sec. Geol.), Bol. 86:65-79.
14
Biostratigraphy and calibrated climatic chronology of the Upper

Pliocene and Lower Pleistocene of France
JEAN CHALINE
using multivariate analysis of detailed dental morphology

(Chaline, 1983; Chaline and Laurin, 1986), which shows that
In the years following the consensus on the Pliocene-Pleistocene
these species-level taxa correspond in their successive degrees of
boundary in France (Chaline and Michaux, 1972a) and the
evolution to a single anagenetic or diachronic lineage evolving in
observation (Chaline, 1977) that the appearance of Alloplace in Eurasia, that is, western Europe to central Siberia. This
phaiomys pliocaenicus in European mammal faunas is synchroallows, for the first time, establishment of a high-resolution
nous with the Neogene-Quaternary boundary proposed by
mammal biostratigraphy, with the advantages that the remains are
IGCP-41 at the top of the Olduvai event, there have been several
abundant in sites very close to one another in time and space, and
further developments. In this chapter we review, in succession,
that geochronologic data are available to calibrate this evolutionthe following: the new "rodent biostratigraphy" and the correlaary sequence and to discriminate variable evolutionary rates
tions that it implies; geochronologic data permitting new
(Chaline and Laurin, 1986; Chaline and Farjanel, 1990).
calibrations; new palynological and faunal data of significance
for paleoclimatic correlations in northern Europe; and a reevalGeochronological data
uation of the classic sites with respect to the NeogeneQuaternary boundary.
Four sites have yielded radiometric dates which calibrate
evolutionary stages in the Mimomys occitanus davakosi-savini
lineage.
"Rodent biostratigraphy"
Introduction
In the classic sites of Plio-Pleistocene age, the large mammals

have provided the basis for a biostratigraphy that approaches a
succession of assemblage-zones or cenozones. In practice,
however, the true affiliations of the large mammal species from
the various levels are seldom known for certain, because it is
only rarely that sufficient numbers of individuals are found in a
given assemblage to permit reliable determinations of the
intraspecies and interspecies variations. The same is not true for
rodents, which are notably very abundant in cave fillings, but can
also occur in equal numbers in fluviolacustrine deposits. The
best-documented rodent lineage in the interval under discussion
is that which began with Mimomys occitanus, with seven species
ranging from the middle Pliocene to middle Pleistocene (Chaline
and Michaux, 1972a,b).
The Mimomys occitanus davakosi-Mimomys savini
lineage
Perrier-Etouaires (Auvergne). This site has yielded a tooth of

Mimomys pliocaenicus polonicus from the mammal fauna found
between the Roccaneyra Basalt and the chalky layer np.l. The
earliest analyses, by Curtis (Bout, 1966), suggested dates of 3.9
and 3.3 Ma, respectively, for the enclosing strata. A subsequent
date obtained by Lippolt (Bout, 1966) raised the age of this
fauna to 3.1 Ma.
More recently, Chambaudet and Couthures (1981) obtained
an age of 3.14 Ma based on fission-track analysis of volcanic
sphene. In a subsequent study, which took into account new
stratigraphic observations, Pastre, Chambaudet, and Couthures
(1983) proposed a revised age for the fauna of 2.25 0.25 Ma.
Considering the results of paleomagnetic analyses, which show a
normal polarity for the np.l level (Semah and Biquand, 1978),
the Perrier-Etouaires fauna most probably should be situated at
the end of the Gauss normal chron at about 2.6 Ma. Note that
recent palynological correlations (discussed later), which indicate that the ancestral Mimomys occitanus dates from no more
than 3.0 Ma, support this estimate.
The phylogeny of the Mimomys taxa that make up the evolutionary succession occitanus davakosi-occitanus occitanus-occitanus
hajnackensis-pliocaenicus polonicus-pliocaenicus pliocaenicus- Le Coupet (Auvergne). This site has been dated at the locality of
pliocaenicus ostramosensis-savini is based on a biometric study Saint Georges d'Aurac. A few teeth of Mimomys pliocaenicus s.
178
Plio-Pleistocene of France
str. recovered from a volcanic tuff beneath the Coupet basalt

flow were dated by Savage and Curtis (Chaline and Michaux,
1972a) at 1.92 Ma. The petrography of the fossiliferous sediment
indicates that it is virtually contemporaneous with the flow that
overlies it.
179
flora and below one of Eburonian age. Finally, a core of SimardMeix-Pernod contained a tooth of M. p. ostramosensis or M.
savini in a bed above the Eburonian level, in a paleofloral
context that may represent the beginning of the Waalian
(Chaline and Farjanel, 1990).
For the southern part of western Europe, Sue (1980) has
Valerots a Nuits-Saint-George (Cote d'Or). This cave filling has proposed correlations between micromammal zones according to
yielded a fauna with Mimomys savini, together with advanced palynological evidence (Sue and Zagwijn, 1983). Zone MN-15 of
Mein (1975), with Mimomys occitanus, would thus range
Allophaiomys pliocaenicus (Chaline et al., 1985).
between 3.0 and 2.65 Ma, with the fossil mammal site of Sete at
Courterolles (Yonne). A fauna similar to the preceding, with 2.8 Ma (Aguirre et al., Chapter 9, this volume; Azzaroli et al.,
evolved Allophaiomys pliocaenicus (Brochet, Chaline, and Chapter 11, this volume).
Poplin, 1983), is also found in cave deposits. These faunas with
evolved A. pliocaenicus are best placed between the "Waalian" Faunal evidence. At Perrier-Etouaires, the well-known largeand "Cromerian I" interglacials, in the Menapian cold-climate mammal fauna (Bout, 1960) includes a tapir, a forest pig, and
numerous cervids (Heintz, 1970), together with a Taxodium flora
interval above the Jaramillo normal subchron.
that suggests a partly forested landscape of interglacial
(Reuverian?) type, as further indicated by the presence of
Climate chronology of the late Pliocene and early
Hystrix. At Saint-Vallier (Drome), the loess fauna (Viret, 1954),
Pleistocene of France
of (upper?) Pretiglian age, contains an incisor of Ochotona
Calibrated biostratigraphy can be linked with contemporaneous associated with remains of Mimomys pliocaenicus polonicus.
climate history. By incorporating data from sedimentology, Faunas come from various levels at Chagny, but the tapir is
faunal analysis, and palynology, a climatochronology can be unquestionably from a basal layer (Reuverian?), whereas the
constructed in which the geologic histories of northern and rodents come from the upper levels, with teeth of M.
southern Europe can be compared. The presence of rodent pliocaenicus ostramosensis transitional to M. savini and teeth of a
remains together with those of large mammals in classic species referred to Allophaiomys; the correlation is possibly to
assemblages (Perrier-Etouaires, Chagny, Saint-Vallier, Le Cou- the final part of the Eburonian.
pet, etc.) makes it possible to replace the earlier chronology of
The site of Montousse 5 (Clot et al., 1975) contains, together
the Villafranchian with a new system and to reinterpret the with M. pliocaenicus ostramosensis, some teeth of a lemming in
faunas from a paleoecological point of view.
the genus Lemmus (the most southerly known occurrence) and
of Macaca florentina. The correlation is to the end of the
Stratigraphic evidence. From the stratigraphic point of view, Eburonian glacial-climate phase or the beginning of the Waalian
studies in the south of France by Michaux (1980) and Sue (1980) interglacial.
and in the Bresse basin, between Dijon and Lyon, by the
In the Bresse basin, sediments at Commenailles, Magny,
"groupe Bresse" have provided information concerning the later Montagny, and Vonnas contain remains of desmanids, which
Pliocene. The discovery of a fluvioglacial gravel interstratified imply running water. At Montagny, the tortoises suggest a
with the Bresse marls (Senac, 1981), situated below the Vonnas temperate climate. The correlation of Commenailles is to the
horizon with Mimomys pliocaenicus, is of particular importance. Pretiglian, Montagny and Magny to phase C4 of the Tiglian, and
This gravel could be related either to the Pretiglian or to the Vonnas to the Eburonian (Chaline, 1984; Chaline and Farjanel,
Eburonian, but in any event it represents the most ancient well- 1990). At Trevoux-Reyrieux, in the southern part of the Bresse
dated indication offluvioglacialconditions in western Europe.
basin, the presence of a southern vole, Mimomys capettai,
Again in Bresse, at Montagny-les-Beaune, beds with M. indicates an interglacial phase (Reuverian?).
pliocaenicus pliocaenicus and remains of tortoises are overlain by
a stratum with strongly wind-sculpted calcareous blocks, implyOther elements of climate correlation in Europe
ing a period of denuded and arid soils, which in these latitudes
would occur during cold phases (i.e., Eburonian).
In the stratigraphy of the Mimomys occitanus davakosi-savini
lineage, evidence from other parts of Europe can be applied to
Palynological evidence. Drill cores in the Bresse basin have the developments in France. At Wolfersheim (Tobien, 1952) the
produced rodent remains (Chaline, 1984) from within palynolog- fauna has a tropical aspect, with a tapir (Tapirus arvernensis),
ical sequences that are readily correctable with those of flying squirrels (Petaurista), a cercopithecid (Macaca), Prolagus,
northern Europe (Farjanel, 1985). A core from Servignat- and tortoises. All of these elements are of interglacial type;
Montmain yielded a tooth of M. pliocaenicus in a paleofloral according to Brunnacker and Boenigk (1976), Wolfersheim is
context of interglacial type, probably Tiglian. A core from dated to the Gauss normal chron. According to Kowalski (1960),
Labergement-les-Seurre at La Mare-du-Bois produced a tooth of the type site of Mimomys pliocaenicus polonicus, at RebieliceM. pliocaenicus ostramosensis from a bed overlying a Tiglian Krolewski, also contains the most primitive known lemming
180
Jean Chaline
m.y.
PALEOMA
B. R
GNETISM
PROPOSEDCORRELATIONS MAMMAL
RODENT ZONES
BRESSE
Genus
Diachron
LOCALITIES
WITH THE NETHERLANDS ZONES
Subdiachron
Mimomys savini
MN17
Mimomys savini
Waalian
Simard-'Meix-Pernot'
Mimomys pliocaenicus ostramosensis
Mimomys pliocaenicus ostramosensis
Labergement
Charrey
Q1
MP2
j..
MP1JE
L4| '<*
Eburonian
'1
Tiglian
Mimomys pliocaenicus pliocaenicus
Mimomys pliocaenicus polonicus
L2
Bragny
Geanges
BIV
Broin
Commenailles
Bagnot
Bin
St-loup-La Salle
S2 5
I to l u
Cessey-Sur-Tille
MN16
Praetiglian
Reuverian
Mimomys occitanus hajnackensis
MN15
O
Figure 14.1. Eurasiatic PlioPleistocene correlations between rodents of the

Mimomys occitanusM.
savini lineage and
palynostratigraphy.
(Adapted from Chaline and
Farjanel, 1990.)
|M
Mimomys
occitanus occitanus
Mimomys occitanus davakosi
Brunssumian
{Synaptomys europaeus), the spermophile Citellus polonicus,

and the most primitive member of the avian genus Lagopus,
attesting to a cold-steppe regime, probably Pretiglian.
In The Netherlands, the interglacial-type fauna at Tegelen,
with Macaca and Mimomys pliocaenicus, is in the Tegelen C5
level, in re versed-polarity sediments between the normal paleomagnetic events attributed to the Reunion subchron and the
Olduvai (as "Gilsa") by Freudenthal, Meijer, and Van der
Meulen (1976; see also Zagwijn, Chapter 16, this volume), and
thus about 2.0 Ma in the orbitally calibrated time scale. Deposits
correlated to the earliest Eburonian at Brielle yield primitive
Allophaiomys pliocaenicus and the collared lemming Dicrostonyx (Van der Meulen and Zagwijn, 1974).
In central Europe, Janossy and Van der Meulen (1975) have
described the type of Mimomys pliocaenicus ostramosensis at
Osztramos, Hungary, together with Allophaiomys pliocaenicus
and Lemmus lemmus, in an assemblage correlative with Eburonian conditions. Finally, at Neuleiningen II (Malec and Tobien,
1976) and at Schernfeld (Dehm, 1962), a similar association with
Lemmus indicates the same correlation, despite the fact that
Allophaiomys has not yet been discovered at Schernfeld.
biologic, and climatic schema that covers not only French

territory but also all of Europe and northwestern Asia as far as
western Siberia (Zazhigin, 1980). In the latter region, the cited
Mimomys species and subspecies are considered to be taxonomic
synonyms of the indicated characterizing species of the biozones.
The Upper Pliocene and Lower Pleistocene European biozonation (Figure 14.1) is founded on the evolutionarily successive
morphospecies Mimomys occitanus, M. pliocaenicus, and M.
savini, which are considered to represent three diachrons (a spatiotemporal chronomorphocline unit characterized by a degree of
evolution). These diachrons are further divided into nine subdiachrons (Chaline, 1983; Chaline and Farjanel, 1990) as follows:
Chronobioclimatic synthesis in Europe and the

Quaternary boundary
Mimomys occitanus occitanus bio zone. Type locality: Sete

(Herault, France).
The European record of the Mimomys occitanus davakosi-savini

lineage (Chaline and Laurin, 1986) supports a chronologic,
Mimomys occitanus davakosi bio zone. Type locality: Ptolemais 3

(Macedonia, Greece).
Localities: Spain (Villalba Alta, Arquillo); France
(Serrat d'en Vacquer) (P. Mein and J. Michaux,
personal communication); and Russia (Ob Plateau
of Siberia, with the type of M. antiquus, a synonym
of M. occitanus).
Age range: 4.0 to 3.6 Ma.
Localities: Slovakia (Ivanovce B, with the type of

Mimomys hassiacus atavus, a synonym of M.
Plio-Pleistocene of France
occitanus); Hungary (Osztramos 9, with the type of

Mimomys silasensis, a synonym of M. occitanus).
181
Servignat-Montmain, and Simard-Meix Pernod);

Austria (Deutsch Altenburg 2C); Slovakia (Mokra
1); Hungary (Osztramos 3); Romania (Betfia 2);
Poland (Kamyk, Zaba Cave).
Mimomys occitanus hajnackensis biozone. Type locality: Hajnacka (Slovakia). Because the name stehlini used previously was
ambiguous with respect to lineage, it was decided at the Lower Mimomys savini.and Allophaiomys pliocaenicus biozone.
Rohanov conference (Czechoslovakia) in 1987 to replace it with Type locality: Piispokfurdo (Romania).
the name hajnackensis. This taxon is without doubt related to the
Localities: France (Mas Rambault, La Sartanette);
preceding lineage as indicated.
Spain (Bagur 2); Austria (Deutsch Altenburg 4B4C); Romania (Betfia 1, 3, 7).
Localities: Spain (Escorihuela); Italy (San Giusto);
Germany (Wolfersheim); Poland (Weze); Hungary
(Csarnota 2); Moldavia (Kotlovina); Ukraine (Kryzhanovka, Kujalnick, Uryv 1); Russia (Akkulaevo, in Upper Mimomys savini and Microtus burgondiae-Allophaiomys
Bashkiria, with the type of M. gracilis akkulaevae, a nutiensis biozone. Type locality: Les Valerots (France).
synonym of M. o. hajnackensis).
Localities: England (Westbury 1); France (CourteAge range: 3.0 to 2.6 Ma.
rolles); Spain (Venta Micena, La Victoria); Italy
(Soave, Monte Peglia, Selva Vecchia); the former
Mimomys pliocaenicus polonicus biozone. Type locality: RebieYugoslavia (Marjan); Czech Republic (Chlum 6,
lice 1 (Poland).
Vcelare 4, Holjstein); Poland (Kozi Grzbiet);
Ukraine (Nogaisk, Khairy); Russia (Tichonovka,
Localities: France (Perrier-Etouaires, Saint-Vallier, and
near
Vladivostok).
Bresse Valley sites of Cessey-sur-Tille, Beaune
Age
range:
1.1 to 0.8 Ma.
BIIIa-BIVa, Commenailles, Broin, Chagny 2, Bagnot, and Satin-Loup-la-Salle); Italy (Arondelli);
Moldavia (Kotlovina); Ukraine (Kujalnick, KryzhaIdentification of the Quaternary boundary
novka, Uryv 2); Russia (Liventsovka, in the Don
basin, with M. livenzovicus, a synonym of M. On the basis of studies in the Calabrian marine sequence of Italy
by INQUA Subcommission 1-d, "Pliocene-Pleistocene Boundpliocaenicus polonicus).
ary," and by IGCP Project 41, "Neogene/Quaternary Boundary"
(see Foreword, this volume), the physical stratotype of this
boundary
in the section at Vrica has been defined at a level
Mimomys pliocaenicus pliocaenicus biozone. Type locality:
which
equates
with the lowest (first) appearance of species
Castelfranco (Italy).
indicating a cold climate. According to current calibrations, that
Localities: England (Norwich and Weybourne Crag); cold-climate phase commenced at 1.8 Ma, slightly below the top
The Netherlands (Tegelen); France (Saint Georges of the Olduvai event (Pasini and Colalongo, Chapter 2, this
d'Aurac and Bresse Valley sites of Vonnas, La- volume). It follows that this cold phase in the European
bergement-les-Seurre, Comblanchien, Montagny- continental sequences must be identified with the Eburonian
les-Beaune, Charrey, Bragny, and Geanges); Poland paleoclimatic stage of the lower Rhine basin (Zagwijn, Chapter
(Kadzielna, Kielniki); Moldavia (Kotlovina); Russia 16, this volume).
(Platovo 1 in the Russian plain; Lebiaje, Podpusk,
In accepting this correlation to the Eburonian as a first
and Kizhikia in southern Siberia).
approximation, subject to the support of complementary data,
the Neogene-Quaternary limit in France and in Eurasia should
therefore be identified in small-mammal biochronology primarily
Lower Mimomys pliocaenicus ostramosensis biozone. Type on the basis of the occurrence of Allophaiomys pliocaenicus,
locality: Osztramos 3 (Hungary).
alone or jointly with Mimomys pliocaenicus ostramosensis
Localities: France (Montousse 5 and the Bresse Valley (Chaline, 1977). In France, that was the period when the infilling
site of Demigny); Germany (Schernfeld, Neuleinin- of the Bresse basin was completed, with the overlying deposits
showing the effects of the climate change to the Waalian intergen); Poland (Kamyk).
glacial, as for instance at Chagny 1 (Chaline and Farjanel, 1990).
Upper Mimomys pliocaenicus ostramosensis and Allophaiomys
deucalion biozone. Type locality: Villany 5 (Hungary).
Localities: The Netherlands (Brielle); France (Seneze,
Balaruc 1, and Bresse Valley sites of Chagny 1,
References
Bout, P. 1960. Le Villafranchien du Velay et du Bassin

hypselodont moyen et superieur de VAllier. Le Puy: Imp.
Jeanne d'Arc.
182
Jean Chaline
Bout, P. 1970. Absolute ages of some volcanic formations in the

Hajnacka und Ivanovce (Slowakei) C.S.S.R.: Microtidae
und Cricetidae. N Jb. Geol. Paldont., Abh. 112:48-82.
Pleistocene. Palaeogeogr. Palaeoclimatol. Palaeoecol. 8:Freudenthal, M., Meijer, T., and Van der Meulen, A. J. 1976.
95-106.
Preliminary report on a field campaign in the continental
Brochet, G., Chaline, J., and Poplin, F. 1983. Courterolles
Pleistocene of Tegelen (The Netherlands). Scripta Geol.
(Yonne), une faune interglaciaire a Hippopotame (Waa3:1-27.
lien?) et une microfaune steppique a Allophaiomys Heintz, E. 1970. Les Cervides villafranchiens de France et
(Menapien?) du Pleistocene inferieur. Soc. Prehist.
d'Aurac. Manuscripts, sect. C, no. 22. Paris: Museum
France, Bull. 16:15-18.
national d'Histoire naturelle.
Brunnacker, K., and Boenigk, W. 1976. Uber den Stand der pala- Janossy, D., and Van der Meulen, A. J. 1975. On Mimomys from
omagnetischen Untersuchungen in Pliozan und Pleistozan
Osztramos 3, North Hungary. Kon. Ned. Akad. Wetens.,
der Bundesrepublik Deutschland. Eisz. Gegenw. 27:1-17.
Proc, Ser. 5 78:381-391.
Chaline, J. 1977. Les evenements remarquables de l'histoire Kowalski, K. 1960. Pliocene insectivores and rodents from
Plio-Pleistocene des Campagnols dans l'hemisphere nord,
Riebelice Krolewskie (Poland). Acta zool. Cracov. 5:155essai de correlation avec la limite Plio-Pleistocene etablie
201.
dans le niveaux marins d'ltalie. Giorn. Geol. 41:123-129. Malec, E, and Tobien, H. 1976. Die Saugetierreste-fiihrenden
Chaline, J. 1983. Les roles respectifs de la speciation quantique
Spaltenfullungen des alteren Pleistozans von Neuleiningen
et diachronique dans la radiation des Arvicolides: consebei Griinstadt (Pfalz). Mainzer Geowiss. Mitt. 5:129-134.
quence au niveau des concepts. Cent. Nat. Rech. Sclent., Mein, P. 1975. Biozonation du Neogene mediterranean a partir
Coll. Int. 330:83-89.
des mammiferes. In Report on the activity of the Working
Chaline, J. 1984. La sequence des rongeurs de Bresse, en tant
Groups, 1971-1975, ed. J. Senes. Bratislava: Akad. Vied.
que reference biostratigraphique et paleoclimatique. Michaux, J. 1980. Rapport Languedoc-Roussilon (stratigraphie).
Geologiede la France 3:251-268.
Assoc. Fr. Etud. Quatern., Bull, Suppl. n.s. 1:314-316.
Chaline, J., and Farjanel, G. 1990. Plio-Pleistocene rodent Pastre, J. E, Chambaudet, A., and Couthures, J. 1983. Noubiostratigraphy and palynology of the Bresse Basin,
veaux elements d'interpretations stratigraphique et geodyFrance, and correlations within western Europe. Boreas
namique des formations plio-Pleistocenes de Perrier
19:69-80.
(Massif Central, France). C. R. Acad. sci. Paris, Ser. D
Chaline, J., and Laurin, B. 1986. A test of phyletic gradualism:
296:79-84.
the Plio-Pleistocene Mimomys (Hintonia) occitanus- Semah, E, and Biquand, D. 1978. Mise en evidence d'une
ostramosensis lineage (Arvicolidae, Rodentia). Paleoinversion de polarite magnetique au sein des sediments
biology 12(2):203-216.
villafranchiens de Perrier (Massif Centrale): interpretation
Chaline, J., and Michaux, J. 1972a. Evolution et signification
stratigraphique. C. R. Acad. sci. Paris, Ser. D 286:829-832.
stratigraphique des Arvicolides du genre Mimomys dans le Senac, P. 1981. Le remplissage detritique Plio-Pleistocene de la
Plio-Pleistocene de France. C. R. Acad. Sci. Paris, ser. D
Bresse du Nord, ses rapports avec la Bresse du Sud.
268:3029-3032.
(Sedimentologie, paleogeographie). Unpublished these
Chaline, J., and Michaux, J. 1972b. An account of the Plio3e. cycle, Universite de Dijon.
Pleistocene rodents' fauna of central and western Europe Sue, J. P. 1980. Contribution a la connaissance du Pliocene et
and the question of Pliocene-Pleistocene boundary. In
du Pleistocene inferieur des regions mediterraneennes
International Colloquium "The boundary between Neogene
d'Europe occidentale par l'analyse palynologiques des
and Quaternary," ed. M. N. Alekseev et al., pp. 46-53.
depots du Languedoc-Roussillon et de la Catalogne.
Moscow: Nauka.
Unpublished these doctoral, Universite de Montpellier.
Chaline, J., Renault-Miskovsky, J., Brochet, G., Clement-Dels, Sue, J. P., and Zagwijn, W. H. 1983. Plio-Pleistocene correlaR., Jammot, D., Mourer-Chavre, C , Bonvolot, J., Lang,
tions between the N.W. Mediterranean and N.W. Europe
J., Leneuf, N., and Pascal, A. 1985. L'aven des Valerots
according to recent biostratigraphic and paleoclimatic
(Nuits-San-Georges, Cote d'Or), site de reference du
data. Boreas 12:153-156.
Pleistocene inferieur. Rev. Geol. Dynam. Geogr. Phys. Tobien, H. 1952. Die oberpliozane saugetierfauna von Wolfer26:109-118.
sheim, Wetterau. Deutsch. geol. Ges., Zeits. 104:180-191.
Clot, A., Chaline, J., Heintz, E., Jammot, D., Mourer, C , and Van der Meulen, A. J., and Zagwijn, W. H. 1974. Microtus
Rage, J. G. 1975. Montousse 5 (Hautes Pyrenees), un
(Allophaiomys) pliocaenicus from the lower Pleistocene
nouveau remplissage de fissure a faunes de vertebres du
near Brielle, The Netherlands. Scripta Geol. 21:1-12.
Pleistocene inferieur. Geobios 9:511-514.
Van Montfrans, H. M. 1971. Paleomagnetic dating in the North
Dehm, R. 1962. Alpleistozane Sauger von Schernfeld bei
Sea basin. Unpublished dissertation, Amsterdam UniverEichstatt in Bayern. Bayer. Staats. Paldont. Hist. Geol.
sity.
2:17-62.
Viret, J. 1954. Le loess a banes durcis de Saint Vallier (Drome), et
sa faune de mammiferes villafranchiens. Nouvelles ArFarjanel, G. 1985. La flore et le climat du Neogene et du
chives no. 4. Lyon: Museum d'Histoire naturelle de Lyon.
Pleistocene de Bresse (France) d'apres Vanalyse pollinique.
Implications chrono-stratigraphiques, Document 97. Paris: Zazhigin, V. S. 1980. Late Pleistocene and Anthropogene rodents
of southern Western Siberia (in Russian). Trudy, no. 339.
Bureau de Recherches Geologigues et Mineralogiques.
Moscow: Akad. Nauk SSSR, Geol. Inst.
Fejfar, O. 1961. Die Plio-Pleistozanen wirbeltierfaunen von
15
The Plio-Pleistocene of England and Iceland

RICHARD W. HEY
Plio-Pleistocene in England
The so-called Crag formations of East Anglia are the only British
examples of a sedimentary sequence that is partly Pliocene and
partly early Pleistocene. They are composed of marine nearshore and estuarine shelly sands and clays, in some horizons with
good pollen assemblages. The oldest unit, the Coralline Crag,
has always been accepted as Pliocene. Early attempts to locate
the base of the Pleistocene in that sequence were guided by the
belief that its position would be indicated by evidence of a sharp
drop in temperature. Thus, at the International Geological
Congress in 1948, it was recommended (King and Oakley, 1949)
that in England the Neogene-Quaternary boundary should be
placed in the East Anglian sequence at the base of the Butleyan
Red Crag, the highest and youngest of Harmer's (1902) three
Red Crag "zones." A number of authors (e.g., Boswell, 1952)
subsequently proposed that it should be lowered to the base of
the Red Crag, on the assumption that the transition from the
underlying Coralline Crag indicated a major climatic deterioration. On the premise that the boundary was associated with the
Olduvai normal paleomagnetic event, Hey (1977) and Funnell
(1977) independently concluded that in East Anglia it could lie
above the top of the Red Crag.
It must be emphasized that the paleomagnetic data available
from the East Anglian succession are scanty (Van Montfrans,
1971, pp. 100-101), and materials suitable for radiometric dating
appear to be absent. Any attempt to locate the boundary in the
East Anglian succession must depend on paleontological correlations with the succession in the Rhine basin, which is both more
nearly complete and more securely related to the paleomagnetic
time scale (Zagwijn, Chapter 16, this volume; von der Brelie et
al., Chapter 17, this volume).
Age of the Red Crag
A borehole at Stradbroke, 30 km north-northeast of Ipswich,
passed through nearly 70 m of marine sediments, of which the
uppermost 45 m were assigned palynologically to the Ludhamian
Stage, whereas the lower beds, resting directly upon the Chalk
Formation, yielded a pollen sequence unlike any recorded
elsewhere in Britain (Beck, Funnell, and Lord, 1972). Funnell

(1977) found that the foraminifera of the pre-Ludhamian beds
indicated a correlation, on the one hand, with the typical Red
Crag and, on the other hand, with foraminifera zone FA2 of The
Netherlands. This is the equivalent of the nonmarine Reuverian
Stage, which appears to lie entirely within the Gauss normal
epoch and has long been accepted as Pliocene (Zagwijn and
Doppert, 1978, figures 2 and 3; Zagwijn, Chapter 16, this
volume). The pollen of the Stradbroke pre-Ludhamian suggests
a similar correlation (R. G. West, personal communication,
1981). That evidence made it virtually certain that the whole of
the Red Crag should be relegated to the Pliocene.
An objection to that conclusion arose from the claim that the
Red Crag had yielded remains of the elephant Mammuthus
("Archidiskodon") meridionalis (Stuart, 1974). In the Dutch
succession, that species appears to be unknown below the upper
Tiglian, almost two full stages above the Reuverian (Azzaroli,
1970, p. 113; Zagwijn, 1974, p. 86). In Italy, similarly, it is known
only from the Upper Villafranchian (Azzaroli et al., Chapter 11,
this volume). Such evidence suggests that M. meridionalis first
appeared in Europe only in the latest Pliocene, not before 2 Ma,
in which case its presence in the Red Crag would imply that the
unit should be of Pleistocene age. Dr. A. J. Stuart (personal
communication) has, however, pointed out to the writer that the
determination of fragmentary remains of "Archidiskodon" is a
matter of some difficulty and that the identification of the Red
Crag material should be considered provisional.
In the Dutch succession, according to the latest available
information (Gibbard et al., 1991; Zagwijn, Chapter 16, this
volume), the top of the Olduvai, and hence the NeogeneQuaternary boundary, lies at the base of the Eburonian Stage, if
not in the uppermost part of the underlying Tiglian. In East
Anglia, the Crag sediments immediately succeeding the Red
Crag (e.g., Norwich Crag, Icenian Crag, etc.) have been divided
by West (1961) into a number of stages (from lowest to highest):
Ludhamian, Thurnian, Antian, Baventian, Beestonian, and
Pastonian. On palynological grounds, Zagwijn (1974) correlated
zone TC3 of the mid-Tiglian with the Ludhamian, whereas the
foraminifera evidence suggested to Funnell (1977) that the
Ludhamian correlated to zone FA1 in the foraminifera zonation
183
Richard W. Hey
184
of The Netherlands, which is placed at the top of the Tiglian

(Zagwijn and Doppert, 1978). Those two lines of evidence
therefore combine to suggest that the Ludhamian should join the
Red Crag in the Pliocene. The most recent correlation, however,
between the British and Dutch stages, as summarized by
Gibbard et al. (1991, figure 8), indicates that the horizon
equivalent to the base of the Eburonian lies in the erosional
interval which everywhere marks the transition between the base
of the Beestonian and the underlying deposits. Thus, not only
the Red Crag but also most (or perhaps all) of the Norwich Crag
would be of Pliocene age.
The Pliocene-Pleistocene boundary in Iceland
Rocks described as Plio-Pleistocene, dated to between 3.1 and
0.7 Ma, appear at outcrop over some 25,000 km2 of Iceland
(Saemundsson, 1980). They are predominantly volcanic, but
incude minor intercalations of sediments, including tillites.
Several local successions have now been described in detail, with
paleomagnetic and some radiometric dates, and these studies
suggest that at least six glacier-forming episodes had already
taken place in Iceland before 2 Ma, and thus before the
beginning of the Pleistocene (Albertsson, 1981; Einarsson and
Albertsson, 1988).
The Plio-Pleistocene sequence of the Tjornes peninsula in
northeastern Iceland is unique in comprising considerable
thicknesses of fossiliferous marine strata. The sequence begins
below the base of the Gauss normal chronozone and continues
upward into the Brunhes. Paleomagnetic data are plentiful, but
reliable K-Ar ages are regrettably few, the basalts in the
sequence being deficient in potassium and in many cases being
altered as well. The stratigraphy of the relevant portion of the
Tjornes sequence is as follows, after Albertsson (1978, figure 2):
6. Mana basalts
5. Breidavik Beds (4 marine sedimentary beds, 4 tillites, 3
basalt flows)
4. . Strangarhorn basalt
3. Furuvik Beds (1 marine sedimentary bed, 2 tillites, 1
basalt flow)
2. Furuvik lower basalt
1. Hvalvik basalt
A reasonably reliable age of 2.4 Ma has been obtained for the
Hvalvik basalt, which has reversed remanent polarity and
evidently lies within the lowest part of the Matuyama
chronozone. A less reliable isochron age of 1.2 Ma for the base
of the Mana basalts at the top of the sequence (Albertsson, 1978)
suggests that the Neogene-Quaternary boundary may lie between those two levels. The lowest basalt of Furuvik is normally
polarized, and thus could lie either within the Reunion
subchronozone or, less probably, in the Olduvai subchronozone.
The first interpretation is favored by Einarsson and Albertsson
(1988), who show that it is consistent with the estimated average
rate of accumulation in Tjornes during the Matuyama and
Brunhes chrons. It may be noted that the lowest basalt of

Furuvik is directly overlain by the lower tillite of the Furuvik
Beds, which not only is the oldest tillite in the Tjornes succession
but also appears to be the earliest evidence of sea-level glaciation
in Iceland.
References
Albertsson, K. J. 1978. Um aldur jardlaga a Tjornesi.
Ndtturufraedingnum 48:1-8.
Albertsson, K. J. 1981. On Tertiary tillites in Iceland. In: Earth's
pre-Pleistocene glacial record, ed. M. J. Hambrey and
W. B. Harland. Cambridge University Press.
Azzaroli, A. 1970. Villafranchian correlations based on large
mammals. Giorn. Geol. 35:111-131.
Beck, R. B., Funnell, B. M., and Lord, A. R. 1972. Correlation
of Lower Pleistocene Crag at depth in Suffolk. Geol. Mag.
109:137-139.
Boswell, P. G. H. 1952. The Pliocene-Pleistocene boundary in
the east of England. Geol Assoc. Proc. 63:301-312.
Einarsson, T., and Albertsson, K. J. 1988. The glacial history of
Iceland during the past three million years. In The past
three million years: evolution of climatic variability in the
North Atlantic region, ed. N. J. Shackleton et al., pp. 227234. London: The Royal Society.
Funnell, B. M. 1977. Plio-Pleistocene foraminifera of the North
Sea basin. In IN QUA Tenth Congress, Birmingham.
Abstracts, p. 151.
Gibbard, P. L., West, R. G., Zagwijn, W. H., Balson, P. S.,
Burger, A. W., Funnell, B. M., Jeffery, D. H., de Jong, J.,
Van Kolfschoten, T., Lister, A. M., Meijer, T., Norton,
P. S. P., Preece, R. C , Rose, X, Stuart, A. J., Whiteman,
C. A., and Zalasciewicz, J. A. 1991. Early and middle
Pleistocene correlation in the southern North Sea basin.
Quat. Sci. Rev. 10:23-52.
Harmer, F. W. 1902. A sketch of the later Tertiary history of East
Anglia. Geol. Assoc. Proc. 17:416-479.
Hey, R. W. 1977. The N/Q boundary in the Netherlands,
England and Iceland: recent developments. Giorn. Geol.
41:35-38.
King, W. B. R., and Oakley, K. P. 1949. Definitions of the
Pliocene-Pleistocene boundary. Nature 163:186-188.
Saemundsson, K. 1980. Outline of the geology of Iceland. In
Geology of the European countries. Denmark, Finland,
Iceland, Norway, Sweden; Geologie du pays Europeens:
Danemark, Finlande, Islande, Norvege. 26e. Congres
international geologique, Juillet 7-17 1980, Paris, ed.
Comite national Francais de Geologie, pp. 136-157. Paris:
Dunod and Cie.
Stuart, A. J. 1974. Pleistocene history of the British vertebrate
fauna. Biol. Rev. 49:225-266.
Van Montfrans, H. M. 1971. Paleomagnetic dating in the Noith
Sea basin. Earth Planet. Sci. Lett. 11:226-236.
West, R. G. 1961. Vegetational history of the early Pleistocene of
the Royal Society Borehole at Ludham, Norfolk. R. Soc.
London, Proc, Sect. B 155:437-453.
Zagwijn, W. J. 1974. Variations in the climate as shown by pollen
analysis, especially in the lower Pleistocene of Europe. In
Ice Ages: ancient and modern, ed. A. E. Wright and F.
Moseley. Liverpool: Seal House Press {Geol. J., Spec.
Issue, no. 6).
Zagwijn, W. J., and Doppert, J. W. C. 1978. Upper Cenozoic of
the southern North Sea basin: paleoclimatic and paleogeographic evolution. Geol. Mijnbouw 57:577-588.
16
The Neogene-Quaternary boundary in The Netherlands

WALDO H. ZAGWIJN
Introduction
The lithostratigraphy, biostratigraphy, and chronostratigraphy of

Plio-Pleistocene deposits of The Netherlands can be related to
paleoclimatic evidence for recognizing the Pliocene-Pleistocene
boundary in this area. Such data have contributed to correlation
between the stratigraphic sequence in The Netherlands and the
proposed Neogene-Quaternary boundary-stratotype in the
Vrica section of Calabria, Italy.
The Plio-Pleistocene sequences of strata in The Netherlands
are apparently the most nearly complete within the southern
North Sea basin, as they were deposited in an area of persistent
strong subsidence. Moreover, interfingering of marine and
nonmarine beds is common in this region.
The marine deposits contain microfauna that derived mostly
from rather shallow water, with planktonic foraminifera being
virtually absent, thus preventing direct correlations with the
Mediterranean region or deep-sea sediments. The mollusk
faunas are likewise quite different from those encountered in
more southern regions of Europe. Mammal remains characteristic of the Villafranchian have been reported from some of the
marine beds, especially from those of East Anglia, but again,
those discoveries do not permit precise correlation with the
Mediterranean area.
From about the middle part of the Pliocene strata upward, the
marine faunas in the North Sea basin show reductions in the
number of species, but increases in the numbers of cold-water
specimens. During that time, periodic immigrations of borealarctic mollusks that today do not occur in the southern North Sea
basin played an important role. The continental sequences,
however, have provided the most detailed paleoclimatic and
paleomagnetic information, which has led to a better understanding of the changes in the marine environment and to better
correlations with the Mediterranean realm.
The Netherlands continental sequence
Since the beginning of the century, the Reuver Clay and the
Tegelen Clay, two clay units of fluvial origin, have been
recognized in the southeastern part of The Netherlands. The
latter, in particular, has attracted much attention because of its

abundant faunal remains (Dubois, 1905). In addition, both clay
units have proved to be rich in fossilized fruits and seeds (Reid
and Reid, 1907, 1915). From this evidence, a Quaternary age
was proposed for the Tegelen Clay, a point of view initially
disputed (Tesch, 1909; cf. Van Baren, 1920), but eventually
accepted by all parties (Tesch, 1928, 1934).
The paleobotanical data, in particular, have contributed to
clarification of the stratigraphy of these beds. The macroflora of
the Reuver Clay is rich in species, about 50% of which are
related to present-day floras in eastern Asia and North America,
according to modern interpretations (Grichuk, Chapter 8, this
volume). The flora from the Tegelen Clay also contains a number
of extra-European exotics, although much fewer (i.e., hardly
15%). Moreover, quite a few modern species of Holarctic
distribution, which are absent from the Reuver Clay flora, are
common in the younger Tegelen flora (Reid and Reid, 1915;
Zagwijn, 1959).
This change in floral composition can be traced through pollen
analysis, as well as in macrofloral paleobotany. The palynological
evidence shows that the Reuver Clay contains many taxa
(Sequoia, Taxodium, Sciadopitys, Nyssa, Liquidambar, and
others) that are absent from the Tegelen Clay. In the latter, only
Tsuga, Pterocarya, Carya, Eucommia, and some other exotics
that are relicts from the Upper Tertiary appear to have survived
(Florschutz and Someren, 1950; Zagwijn, 1960, 1963).
These two paleofloras represent temperate, summer-rainfall
associations, suggesting intervals of relatively warm climate,
which have been termed Reuverian and Tiglian. Pollen analysis
has shown that the marked changes in paleofloral composition
between these two clay formations were caused by a severe coldclimate phase, the Pretiglian, during which all warmth-loving
trees disappeared temporarily or permanently from the area, and
the forest became more open, with the appearance of heaths and
sphagnum bogs (Zagwijn, 1960). Although it has now been recognized that other cold-climate phases preceded that stage, none of
them had such marked effects on the regional flora, nor did the
forest actually open up to any notable extent. In other words, the
Pretiglian was the first time during the late Cenozoic that the
timberline was temporarily shifted south of The Netherlands.
185
186
Waldo H. Zagwijn
After the Tiglian, which was a rather long interval with several
alternating cooler and warmer phases (Zagwijn, 1963), a second
phase of deforestation associated with an intense cold climate
followed, the Eburonian cold stage. The Eburonian was followed by another relatively moderate interglacial complex with
at least one phase of minor cooling, the Waalian, and subsequent
to that came the Menapian cold stage, which is very well
expressed in palynological data.
Following the Menapian, the two interglacials of the Bavelian
age are characterized by occurrences of Pterocarya, Carya,
Eucommia, Tsuga, and other trees which were also commonly
present during the preceding Tiglian and Waalian temperate
forest stages (Zagwijn and Doppert, 1978; de Jong, 1987). Beds
representative of those interglacials are overlain by beds of the
"Cromerian Complex," considered to be of middle Pleistocene
age.
MN-17 of Mein (1975). In Eburonian deposits of the western

Netherlands, Allophaiomys pliocaenicus and Dicrostonyx have
been found (Van der Meulen and Zagwijn, 1974). This zone is
considered to be indicative of earliest Pleistocene age in the view
of many vertebrate paleontologists (Aguirre et al., Chapter 9,
this volume; Chaline, Chapter 14, this volume).
Mammalian paleontology. The large-mammal fossils of the

Tegelen Clay, the most significant of which are Anancus
arvernensis, Elephas (Mammuthus) meridionalis, Dicerorhinus
etruscus, Equus robustus, Sus strozzii, and Leptobos, have been
recovered primarily from the upper part of the Tiglian (pollen
zones TC5 and TC6). Some species, including Anancus
arvernensis, have been found in beds dated as young as earliest
Eburonian (Kortenbout van der Sluys and Zagwijn, 1962).
A rich fauna of small mammals has been found in beds of TC5
age (Freudenthal, Jeijer, and Van der Meulen, 1976). Biostratigraphically, the fauna of these beds, as well as those of earlier
Tiglian age, belong to the Mimomys pliocaenicus zone or zone
Mollusk Zone C: Nassarius propinquus-Lentidium complanatum

assemblage zone. This zone is characterized by a very distinctive
assemblage. Many species are abundant only in this zone, and
several become extinct in the overlying zone.
The Netherlands marine sequence
In 1896, Harmer introduced the Amstelian Stage, based on a

mollusk fauna considered to be younger than that of the
Scaldisian in Belgium and older than that of the Icenian Norwich
Crag. In the Amstelian, an immigration of boreal-arctic species
is widely recognized, among which are Serripes (Cardium)
groenlandicus, Yoldia arctica, and others. At that time, the
Amstelian was considered to be of Upper Pliocene age. Tesch
(1934), however, concluded that a marine fauna containing such
Paleomagnetic investigations. Studies by Van Montfrans (1971) boreal-arctic mollusks, underlying the "pre-glacial fluviatile
and Boenigk, Koci, and Brunnacker (1979) have indicated that beds" of the western Netherlands, should be included in the
below the Matuyama-Brunhes boundary, which is in the lower Pleistocene. He referred those deposits to "Icenian" age, while
part of the "Cromerian Complex," the reversed remanent explicitly stating that the "true Amstelian" (because it was
magnetization of the Matuyama is predominant in the sediments supposedly Pliocene) should be below those beds. Nevertheless,
down to the Reuver Clay. That clay has proved to be of normal that usage clearly equated the greater part of Harmer's
polarity throughout, even where it is at its thickest, and therefore "Amstelian" with The Netherlands "Icenian," and that resulted
it is considered to date to the Gauss normal-polarity chron. Some in confusion as subsequent investigators continued to use
consistent subzones of normal polarity can be observed in the "Amstelian" in a different way than originally intended (Figure
predominantly reversed-polarity interval between the Reuverian 16.2). In view of this confusion, it is preferable to abandon the
and the "Cromerian Complex." One normal-polarity zone in the stage name Amstelian.
upper part of the Tiglian and the lowermost part of the Eburonian
At present, four mollusk assemblage zones (Spaink, 1975) are
is identified with the Olduvai normal-polarity subchron, which identified in the Plio-Pleistocene marine beds of The Nethermeans that the Eburonian cold stage began at about 1.8 Ma, in lands (in stratigraphic order, with oldest below):
the present calibration. The normal-polarity intervals correlative
with the Reunion normal subzones are in the lower part of the Mollusk Zone A: My a arenaria-Hydrobia ulvae assemblage
Tiglian (Van Montfrans, 1971; Urban, 1978). Further investiga- zone. Beds of this zone are poor in species, and many of those
tions by Boenigk et al. (1979) and Urban (1978) indicate that the species are presently living in the area under discussion. Influxes
Gauss-Matuyana boundary (2.60 Ma) is within the Reuverian C of land and fresh-water species, combined with shallowing
zone, the uppermost and coolest part of the Reuverian warm- marine conditions, characterized conditions in that interval.
climate paleoclimatic stage. This means that the Pretiglian cold
stage can be dated between about 2.5 and 2.3 Ma (shown with Mollusk Zone B: Serripes groenlandicus-Yoldia lanceolata assemolder calibration in Figure 16.1), in synchrony with observations blage zone. This zonal assemblage contains many boreal-arctic
of late Pliocene cooling in the marine Piacenzian Stage of Italy species that are lacking in older zones; the two nominate species
(Preface, this volume; Azzaroli et al., Chapter 11, this volume). are predominant.
Mollusk Zone D: Turritella triplicataYoldia semistriata assem-
blage zone. This zone is extremely rich in species, though

relatively few of them are restricted to this zone.
According to present knowledge, Mollusk Zone B essentially
coincides with the Amstelian, as defined by Harmer, though
Harmer also included beds whose warmer-climate faunas we
would assign to Mollusk Zone A.
Plio-Pleistocene of The Netherlands
187
in
<
MEN API AN
20*C
Pt
UJ
POSITION
OF SOME
QCUJ
STRATI G R A P H I C A L L Y
ujoing
IMPORTANT
^ < ^ O\
MAMMAL
ISI
O o in
FINOS
MARl
ASSi
(Spa
in
cc
1 MILL,ION
UJ
INTEIRPRI
PALE OMA
MEAS
(van I
Boeni
10*
POLL ENZ
ESTIMATED
STAGES
(Bastd on Polltn MEAN SUMMER
Analysis) TEMPERATURE
in
UJ
TIC
c? S -
D
(
we
W-B
W-A
Uj
UJ
03
01
Uj
Uj
o
UJ
in
ct
yJ
TC6
TC5
>
UJ
'I
<
>
a
o
1,6-
ELEPHAS
MERIDIONALIS
MIMOMYS
PLIOCAENICUS
ETC. .
11
<
z
o -2-
TC1
TB
UJ
<
a:
o
> <
z
1.8-
QOO
TCAb
ANANCUS
ARVERNENSIS
TAPIR US
ARYERNENSIS
UJ
o
UJ
*~
MIMOMYS
PLIOCAENICUS
-2,1-
<
YAY:
TA
Ui
H
<4J
oo
:
\IBR0N
Pt1
**.
R-B
- CASSIDULIN
(F
HANNAI
If
UJ
in
in
<
Ui
In the foraminifera assemblages, the first appearance of

Elphidiella hannai (Cushman), which previously was suggested
to represent the Pliocene-Pleistocene boundary (Lagaaij, 1952;
Van Voorthuysen, 1957) or Elphidiella hannai-Cribrononion
excavatum zone (Doppert, 1975, 1980), is definitely lower than
the base of Mollusk Zone B (i.e., the Amstelian, sensu Harmer).
According to Doppert (1975), zone FA can be subdivided into
two subzones, FA1 and FA2. The younger of these, FA1, is
definitely poorer in species than FA2. The beginning of the
narrow zone of the arctic species Elphidium oregonense in the
middle part of zone FA straddles the FA1/FA2 boundary. That
TURRI1
TRIPLI
YOL,
SEMIST
(MOL
/
/
/1
LA
11
R-C
NORMAL
REUVERIAN
10IN0
Pt2
PRAETIGLIAN
LPHIDI
TNSE
o I
V
)
/
TIGLIAN
UJ
I-2K
(>
LIMA
(FA 1) \
5:
ALLOPHAIOMYS
DICROSTONYX
LEN FAL
EBURONIAN
a:
TUM (F
EB
SER RIPES
0R0ENI AN DIC
YC)LDIA
LANC EOLAl
(M OL.B)
in
Q
to
Q
NASSARIUS
PROPINQUUS
LENTIDIUM
COMPLANATU
(MOL. C)
WAAL/AN
Figure 16.1. Marine and

nonmarine stratigraphy of the
southern part of the North
Sea basin in the Upper Pliocene to Lower Pleistocene interval, from onshore sites in
the Rhine and Scheldt deltas
of The Netherlands. Note that
the acme of the cold-adapted
foraminiferan Elphidium oregonense in the Pretiglian interval corresponds to the first
well-documented glacial influences on deep-sea deposits in
the northern Atlantic, just
above the Gauss-Matuyama
transition. The level of the
Pliocene-Pleistocene boundary at Vrica, at the top of the
Olduvai normal-polarity interval, corresponds to the beginning of a later cycle of coldclimate conditions at the base
of the Eburonian palynostage.
event has been proposed as the marine equivalent to the base of

the palynologically established Pretiglian cold stage (Van
Voorthuysen, Toering, and Zagwijn, 1972; Zagwijn, 1974a) and
also correlates approximately with the base of Mollusk Zone B,
but Van Voorthuysen et al. (1972) believed that the beds
belonging to zone FA1 were of Pleistocene age.
Correlation of marine beds in the North Sea basin
In Figure 16.2, marine beds and biozones are compared for
different parts of the North Sea basin. As stated earlier, recent
Waldo H. Zagwijn
188
THE
NETHERLANDS
Figure 16.2. North Sea basin

correlations. The Netherlands
Plio-Pleistocene succession is
compared with earlier stratigraphic concepts and with other
successions in eastern England
and northern Belgium. Note
that the correlation of the East
Anglian succession, shown here
according to the state of knowledge in 1984, has since been revised (Gibbard et al., 1991) to
move the Norwich Crag and the
equivalent climate stages below
the Eburonian, which is represented by an erosional interval
(see also von der Brelie et al.,
investigators in The Netherlands have used Harmer's stage

names of 1896, but in a somewhat different manner. Part of the
Red Crag of East Anglia, at Butley, has a mollusk fauna
characteristic of the Serripes groenlandicus-Yoldia lanceolata
Mollusk Zone B of The Netherlands. A Pretiglian age, therefore, seems likely. Other Red Crag deposits, in particular the
Walton Crag, are older (Harmer, 1896).
The Ludhamian and Baventian stages were established within
the "Icenian Crag" according to pollen analysis (West, 1961),
and each reflects a warmer-to-colder trend. According to
Zagwijn (1974b), they can be correlated with the upper Tiglian
(pollen zone TC3 to TC5) and the Eburonian (Spaink and
Norton, 1967), rather than with the Tiglian, Eburonian,
Waalian, and Menapian (West, 1961).
The Norwich Crag (Formation), which is another unit
containing a major component of arctic mollusk species,
according to present views dates to the Baventian and probably
also to the older Ludhamian (West, 1980). It also includes
deposits correlated to a younger cold stage, or stages, called
Pre-Pastonian. Until recently, the accepted palynological correlations (Zagwijn, 1974a) indicated that the Ludhamian should
be correlated to the middle Tiglian (Figure 16.2), so that the
Baventian-age deposits seemed to be of the same age as the
Eburonian. In a study that became available only while this
chapter was being written, Gibbard et al. (1991) showed that
the most reasonable correlation of the Eburonian to the East
Anglian sequence is to the cold-climate deposits of Beestonian
age, which unconformably overlie the typical Norwich Crag
wherever they are in contact. In that view, the Baventian and
Tiglian are considered equivalent.
The Belgian sequence, and particularly the foraminifera
assemblages, were reexamined by Laga (1972), who established
that there is good agreement with the sequence in The
Netherlands. The palynological results (Hacquaert, 1961) clearly
indicate that the Merksem Beds should be included into the
topmost part of the Reuverian (C). The Oorderen Beds,
USE OF
HARMER'S
STAGE NAMES BY LA
TER DUTCH
INVESTIG.
A10 30
formerly named Kallo Beds, are of Reuverian B age (Zagwijn,

1959; Hacquaert, 1961).
Correlation with the Mediterranean area and the
Neogene-Quaternary boundary-stratotype
Palynological investigations by Sue (Cravatte and Sue, 1981; Sue

and Cravatte, 1982) have revealed the existence of an Upper
Neogene steppe-flora phase in the western Mediterranean,
beginning just before the beginning of the Globorotalia inflata
planktonic foraminiferal zone (2.5-2.4 Ma), and followed by a
Mediterranean forest phase. Because Pleistocene cold-climate
phases in more northerly parts of Europe clearly correlate with
intervals of steppe conditions in the western Mediterranean, it
can be considered certain that the first steppe phase of the
western Mediterranean, coinciding with the lower part of the G.
inflata zone, correlates with the Pretiglian of The Netherlands
(Sue and Zagwijn, 1982). In both areas, the cyclic climatic
changes of Pleistocene type evidently began in the lowermost
Matuyama magnetozone, around 2.5 Ma, coincident with a
marked cooling event in late Piacenzian time in the world
oceans, and from a strictly paleoclimatic point of view it might be
preferable to draw the lower boundary of the Pleistocene at that
level (Zagwijn, 1992).
The boundary-stratotype section adopted at Vrica, however,
was chosen according to a different view, which includes other
stratigraphic criteria (Pasini and Colalongo, Chapter 2, this
volume). The Neogene-Quaternary boundary at the top of
marker e is definitely younger than the base of the Globorotalia
inflata zone, which is more than 228 m below the proposed
boundary in the Vrica section. If the paleomagnetic data are
used for correlation, according to current evidence (Zijderveld
et al., 1991) the proposed N/Q boundary is just below the top of
the Olduvai normal magnetic subchronozone and is therefore
coeval with the lower boundary of the Eburonian cold stage in
The Netherlands.
Plio-Pleistocene of The Netherlands
189
Reid, C , and Reid, E. M. 1907. The fossil flora of Tegelen-surMeuse, near Venlo, in the Province of Limburg. Kon.
Akad. Wetens., Verh. 13:1.
Boenigk, W., Koci, A., and Brunnacker, K. 1979. Magne- Reid, C , and Reid, E. M. 1915. The Pliocene Flora of the
Dutch-Prussian Border. Rijksopspor. Delfst., Meded. 6:1tostratigraphie im Pliozan der Niederrheinischen Bucht.
N Jb. Geol. Palaont., Mh. 1979:513-528.
178.
Cravatte, J., and Sue, J. P. 1981. Climatic evolution of North- Spaink, G. 1975. Zonering van het mariene Onder-Pleistoceen en
Plioceen op grond van molluskenfauna's. In Toelichting bij
Western Mediterranean area during Pliocene and Early
Pleistocene by pollen-analysis and forams of drill Autan 1.
Geologische Overzichtskaarten van Nederland 1:600.000,
ed. W. H. Zagwijn and C. J. van Staaduinen. Haarlem:
Chronostratigraphic correlations. Pollen et Spores 23:247Rijks Geol. Dienst.
258.
de Jong, J. 1987. Climatic variability during the past three million Spaink, G., and Norton, P. E. P. 1967. The stratigraphical range
years as indicated by vegetational evolution in northwestof Macoma balthica (L.) (Bivalvia, Tellinacea) in the
ern Europe, and with emphasis on data from The
Pleistocene of the Netherlands and Eastern England.
Netherlands. In The past three million years: evolution of
Geol. Sticht., Meded. 18:39-44.
climate variability in the North Atlantic region, ed. N. J. Sue, J.-P, and Cravatte, J. 1982. Etude palynologique du
Shackleton, pp. 193-207. Cambridge University Press.
Pliocene de Catalogne (nord-est de l'Espagne). Apports a
Doppert, J. W. C. 1975. Foraminiferenzonering van het Nederla connaissance de l'histoire climatique de la Mediterlandse Onder-Kwartair en Tertiair. In Toelichting bij
ranee occidentale et implications chronostratigraphiques.
Geologische Overtfzichtstfkaarten van Nederland 1:600.000,
Paleobiol. continent. 12:1-31.
ed. W. H. Zagwijn and C. J. van Staaduinen. Haarlem: Sue, J.-P, and Zagwijn, W. H. 1982. Correlations in the PlioRijks Geol. Dienst.
Pleistocene between the Mediterranean and N. W. EuDoppert, J. W. C. 1980. Lithostratigraphy and biostratigraphy of
rope. In IN QUA Eleventh Congress, Moscow. Abstracts,
marine Neogene deposits in the Nederlands. Rijks Geol.
vol. l , p . 299.
Dienst, Meded. 32:255-311.
Tesch, P. 1909. De Klei van Tegelen, een onderdeel der
Dubois, E. 1905. Over een equivalent van het Cromer forestbed
"Kieseloolithstufe." Kon. Ned. Aardr. Gen., Tijdschr., 2e
in Nederland. Kon. Akad. Wetens., Versl. Meded. 13:243,
ser. 26:573.
453.
Tesch, P. 1928. La separation stratigraphique PlioceneFlorschiitz, E, and Van Someren, A. M. H. 1950. The
Pleistocene en Europe. In Comptes Rendus de la Reunion
Palaeobotanical Boundary Pliocene-Pleistocene in the
Geologique International. Copenhague 7925:183-188.
Netherlands. In International Geological Congress, Report Tesch, P. 1934. De opeenvolging van de oud-pliosceene lagen in
of the 18th Session, Great Britain, 1948, vol. 9, ed. A. J.
Nederland. Kon. Ned. Aardr. Gen., Tijdschr. 1934:649Butler, pp. 40-46. London: Geological Society.
675.
Freudenthal, M., Jeijer, T., and Van der Meulen, A. J. 1976. Urban, B. 1978. Vegetationsgeschichtliche Untersuchungen zur
Preliminary report on a field campaign in the continental
Gliederung des Altquartars der Niederrheinischen Bucht.
Pleistocene of Tegelen. Scripta Geol. 34:1-33.
Geol. Inst. Univ. Koln, Sonderv. 34:1-165.
Gibbard, P. L., West, R. G., Zagwijn, W. H., Balson, P. S., Van Baren, J. 1920. De Boden van Nederland, vol. 1. Amsterdam.
Burger, A. W., Funnell, B. M., Jeffery, D. H., de Jong, J., Van der Meulen, A. J., and Zagwijn, W. H. 1974. Microtus
Van Kolfschoten, T., Lister, A. M., Meijer, T., Norton,
(Allophaiomys) pliocaenicus from the Lower Pleistocene
near Brielle, The Netherlands. Scripta Geol. 21:1-45.
P. S. P., Preece, R. C , Rose, J., Stuart, A. J., Whiteman,
C. A., and Zalasciewicz, J. A. 1991. Early and middle Van Montfrans, H. M. 1971. Paleomagnetic dating in the North
Pleistocene correlation in the southern North Sea basin.
Quat. Sci. Rev. 10:23-52.
Van Voorthuysen, J. H. 1957. The Plio-Pleistocene boundary in
the North Sea Basin. Geol. Mijnb. 19:263-266.
Hacquaert, N. 1961. Palynologisch onderzoek van de cenozoische
mariene zanden (Scaldisien en Merxemien) van het Van Voorthuysen, J. H., Toering, K., and Zagwijn, W. H. 1972.
Hansadok te Antwerpen. Natuurw. Tijdschr. 42:65-112.
The Plio-Pleistocene Boundary in the North Sea Basin;
Harmer, F. W. 1896. On the Pliocene deposits of Holland and
revision of its position in the marine beds. Geol. Mijnb.
their relation to the English and Belgian crags, with a
51:627-639.
suggestion for the establishment of a new zone "Amstel- West, R. G. 1961. Vegetational history of the Early Pleistocene
ian" and some remarks on the geographical conditions of
of the Royal Society Borehole at Ludham, Norfolk. R.
the Pliocene Epoch in Northern Europe. Geol. Soc.
Soc. London, Proc, Sect. B 155:437-453.
London, Quart. J. 52:748-782.
West, R. G. 1980. The pre-glacial Pleistocene of the Norfolk and
Kortenbout van der Sluys, G., and Zagwijn, W. H. 1962. An
Suffolk coasts. Cambridge University Press.
introduction to the stratigraphy and geology of the Tegelen Zagwijn, W. H. 1959. Zur stratigraphischen und pollenanalyclay-pits. Geol. Sticht., Meded., n.s. 15:31-37.
tischen Gliederung der pliozanen Ablagerungen im
Laga, P. G. 1972. Stratigrafie van de mariene Plio-Pleistocene
Roertal-Graben und Venloer Graben der Niederlande.
Fortschr. Geol. Rheinl. Westf. 4:5-26.
afzettingen uit de omgeving van Antwerpen met een
bijzondere studie van de foraminiferen. Unpublished Zagwijn, W. H. 1960. Aspects of the Pliocene and Early
doctoral dissertation, Universiteit van Leuven (Louvaine).
Pleistocene vegetation in the Netherlands. Geol. Sticht.,
Lagaaij, R. 1952. The Pliocene Bryozoa of the Low Countries.
Meded., Ser. C-III-15:1-78.
Geol Sticht., Meded., Ser. C 5:1-233.
Zagwijn, W. H. 1963. Pollen-analytic investigations in the
Mein, P. 1975. Biozonation du Neogene mediterranean a partier
Tiglian of the Netherlands. Geol. Sticht., Meded., n.s.
des Mammiferes. In Report on activities of R.CM.N.S.
6:49-71.
working groups, ed. J. Slavik, pp. 78-81. Bratislava: Zagwijn, W. H. 1974a. Variations in climate as shown by pollen
Akad. Vied.
analysis, especially in the Lower Pleistocene of Europe. In
References
190
Waldo H. Zagwijn
Ice Ages: ancient and modern, ed. A. E. Wright and F.

Moseley. Liverpool: Seal House Press. {Geol. /., Spec.
Issue, no. 6).
Zagwijn, W. H. 1974b. The Pliocene-Pleistocene boundary in
Zagwijn, W. H. 1992. The beginning of the Ice Age in Europe
and its major subdivisions. Quat. Sci. Rev. 11:583-591.
Zagwijn, W. H., and Doppert, J. W. C. 1978. Upper Cenozoic of
the southern North Sea Basin Palaeodimatic and Paleogeographic evolution. Geol. Mijnb. 57:577-588.
107:697-714.
17
The Tertiary-Quaternary Boundary in western Germany

GUNTHER VON DER BRELIE, KARL BRUNNACKER, WINFRIED HINSCH, and HEINZ TOBIEN*
Introduction
Deposits of late Pliocene and early Pleistocene age are found in

several parts of western Germany (Figure 17.1), with marine
strata principally in the northern coastal plain of northwestern
Germany, and estuarine and nonmarine deposits in the Rhine
Graben and other structural depressions in the North German
Plain. Among these the marine record concerns mostly Pliocene
deposits, whereas the lower Rhine Basin contains a lengthy and
apparently continuous record of sedimentation during that time
period. In the upper Rhine Graben and other inland areas, the
Plio-Pleistocene record is not as comprehensive, being represented by localized and irregularly condensed sequences, which
still afford some important correlation data.
The authors were responsible for the sections of this chapter as
follows: K. Brunnacker, geological and paleomagnetic data; the
late Professor H. Tobien, vertebrate paleontology; G. von der
Brelie, palynology; W. Hinsch, marine paleontology.
Geological background
Northwestern Germany. In the coastal sequence, the stratigraphic units of most interest are the Kaolinite Sand (Weyl,
Rein, and Teichmuller, 1955) and the Lieth series (Menke,
1975). Particular sections have been described from Oldenswort
(Menke, 1975), Weser-Bergland (Benda et al., 1968), and
Ostfriesland (Meyer, 1981).
Lower Rhine Basin. In this region, the stratigraphic sequence,
divided into cyclic units of transgressive-regressive couplets, is
denominated A/a to D/d, from oldest to youngest (Boenigk et
al., 1974), and appears to correspond with the record from the
Rhine delta in The Netherlands (Zagwijn, 1974; Zagwijn,
The lower strata, unquestionably Pliocene in age, have been
correlated according to molluscan faunas and palynology, analyzed petrographically, and inserted into the paleomagnetic scale
by Boenigk and co-workers (Boenigk et al., 1974; Boenigk,
1978a; Boenigk, Koci, and Brunnacker, 1979). Two heavymineral suites are clearly distinguished (Table 17.1). The
deposits in the Al beds (clay horizon A of the Ville district, as

well as the overlying peat layer) contain a suite of stable heavy
minerals, such as tourmaline, zircon, and staurolite, and
correlate with the upper Reuverian (Reuverian B-C) of The
Netherlands climatostratigraphic sequence (Zagwijn, 1974). The
younger A2 clays are characterized by unstable heavy minerals
(epidote, garnet, green hornblende, and alterite). An important
molluscan fauna has been found in the A2 horizon (Boenigk et
al., 1974), which clearly represents a Pliocene assemblage. The
A2 horizon has normal paleomagnetic polarities in the lower part
and reversed polarities in the upper part.
The succeeding stratum, termed the Oldest Pleistocene in
historical literature, is the bl Gravel (Figure 17.2), the lithology
of which resembles the Pliocene quartz gravels, or Kieseloolithschotter (Schnutgen and Brunnacker, 1977). The Bl Clay follows
above, with evidence for relatively warmer conditions described
as the Fortuna Oscillation by Urban (1978). The upper part of
this sequence records a change back to normal paleomagnetic
polarity. At the top of this sequence, a major erosional surface is
the main discordance in the Plio-Pleistocene sequence of this
area.
The overlying b2 Gravel (Figure 17.2) has the more variegated
appearance of the typical Quaternary "colored gravels" (Schnutgen and Brunnacker, 1977). Nevertheless, both the bl and b2
gravels are assigned to the Pretiglian cold-climate interval. The
overlying B2 Clay contains a "Quaternary" (i.e., modernized)
warm-climate molluscan fauna (Lozek, in Boenigk, Kowalczyk,
and Brunnacker, 1972) and evidence of a corresponding warmerclimate flora, including fruits and seeds (E. K. Kempf, in
Boenigk et al., 1972). The presence of Fagus (beech) pollen
indicates the Tiglian-A warm period (Urban, 1978). The B2 Clay
shows reversed paleomagnetic polarity.
Above that level, the variegated quartz-poor c Gravel is
succeeded by the C Clay with another thermophilic molluscan
fauna and flora (Lozek, in Boenigk et al., 1972; Urban, 1978).
This clay also shows reverse magnetization (A. Koci, in Boenigk
etal.,1979).
In the d Gravel horizon, the relatively sand-rich deposits
contain the earliest glacially striated erratics. In the overlying D
Clay, the heavy-mineral association is similar to that of the B
191
von der Brelie, Brunnacker, Hinsch, Tobien
192
horizon, and the molluscan fauna is still of the warm "Quaternary" nature (Lozek, in Boenigk et al., 1972); the paleoflora is
noteworthy for the presence of some holdover "Tertiary"
elements (Kempf, in Boenigk et al., 1972; Urban, 1978).
Samples of the D Clay from the Ville sequence have shown
normal paleomagnetic polarities.
Petrographic analyses (Schniitgen, 1974; Boenigk, 1978b)
provided the basis for a separate classification of the gravel
layers as low-stand terraces, or "main terraces," classified under
the terms HT1 through HT4. The molluscan fauna of the Upper
Pliocene in the Bl Clay is sharply different from the warmclimate fauna in the B2 Clay horizon, which according to pollen
analysis (Urban, 1978) can be correlated to the lowest horizon of
the Tiglian complex in The Netherlands sequence (Zagwijn,
1960, 1974; Zagwijn, Chapter 16, this volume).
\ 9
Figure 17.1. Geologic provinces of western Germany.
Vogelsberg. According to geobotanical results, the Liegendton

("under-clay") and the lignites of the Wohnbach mining area,
situated in the Horloff Graben, belong to the lower part of the
Upper Pliocene (either uppermost Brunssumian or Reuverian
A) and therefore are younger than earlier assumed (Boenigk et
al., 1977). The mammal fauna at Wolfersheim, found at the top
of the Liegendton, is classified as late Ruscinian, MN-15 (as
discussed later), and is therefore older than the lower Villafranchian (e.g., Etouaires) and younger than the typical
Ruscinian (e.g., Montpellier). Evidence for several alternating
cold- and warm-climate periods has been found in the overlying
clay-silt layers, the earliest of which corresponds to the normally
magnetized Pretiglian horizons (Boenigk et al., 1974).
Table 17.1. Stratigraphic correlation of the Pliocene and Quaternary in western Germany
Toringian
MQ2
MQ1
KorlichEand F
Hohen-Sulzen
Weissenburg 7
Neuleinlngen 15
Late Villanyian
MN17
Gundersheim 2
Erpflngen 2
Moggaster Hohle
Schernfeld
Deinsdorf, Schumbach
Early Villanyian
MN16
Frechen
Late Ruscinian
MN15
Gundersheim 3, 4
Early Ruscinian
MN14
Herbolzheim
Turolian
MN13
Pleistocene
Biharian
Pliocene
Late Miocene
Plio-Pleistocene of western Germany
Other inland sections. The Upper Pliocene and Lower Pleistocene strata have also been studied in Nordhessen Buchenau
(Leschick, 1951), in the Kelsterbach Terrace complex of the
Rhine/Main area and the Upper Rhine Graben (Semmel, 1974;
von der Brelie, 1974), and in other deposits of the Upper Rhine
Graben (Bartz, 1976, 1982). The age of the Steinbach deposits
has been discussed by Kolumbe (1963) and Frenzel (1973), and
that of Jockgrim by Guenther and Mai (1977), Peters (1965), and
Koci, Schrimer, and Brunnacker (1973).
Table 17.2. Succession of mammalian local faunas in

western Germany
Biharian
Late Villanyan
Middle Villanyan
Mammalian faunas
Early Villanyan
Ruscinian. The Herbolzheim local fauna, of early Ruscinian age,

has long been known from afissurefillingin Dogger limestones
north of Freiburg im Breisgau, at the eastern border of the Rhine
Graben (Figure 17.1) (Tobien, 1951). The association includes 11
small and large taxa, among which are Anancus arvernensis,
"Sus" minor, Canis donnezani, Sciurus n. sp., Vulpes sp., and
Lutra n. sp., which indicate MN-14 (early Ruscinian) age (Mein,
1989). The riverside forest environment indicated by Sciurus,
Lutra, and a suid is similar to the typical Montpellier
asssemblage itself (Tobien, 1975), although Herbolzheim was
listed erroneously under "Csarnotian" or late Ruscinian.
Upper Villafranchian:
Farneta, Tasso,
Olivola faunas (= MQ-1)
-N/Q boundary-
Biostratigraphy
Mammalian local faunas in the interval of the PliocenePleistocene boundary are categorized in the mammal ages
Ruscinian, Villanyan, and Biharian in the current biostratigraphic concept for central and western Europe. The Villafranchian continental stage corresponds, more or less, to the Villanyan
and Biharian together, as discussed later.
In this concept, the Ruscinian and Villanyan represent the
Pliocene, and the Biharian the lower part of the Pleistocene. The
Pliocene-Pleistocene boundary is drawn between the Villanyan
and the Biharian at a level close to the top of the Olduvai normalpolarity subchron, corresponding to the decision of the Neogene-Quaternary Boundary Commission (Fejfar and Heinrich,
1989), with an age of 1.81 Ma (Pasini and Colalongo, Chapter 2,
this volume). The biochronological subdivisions of these ages are
based on the mammal zones proposed by Mein (1989, p. 73), in
which the Ruscinian is equivalent to Mammalian Neogene (MN)
zones MN-14a, -14b, -15a, and -15b, and Villanyan to MN-16a,
-16b, and -17. The Biharian is equivalent to the Mammalian
Quaternary (MQ) zone MQ-1. The terminal Miocene faunas are
assigned to the Turolian mammal age, zone MN-13 (Table 17.1).
For details, see Fejfar and Heinrich (1989, figures 1 and 2).
Table 17.2 shows the classic Villafranchian Stage of the Italian
continental sequence in place of the mammal ages Villanyan and
Biharian. Correlation between the Villafranchian subdivisions in
Italy and these mammal ages, which are typified mainly in central
and eastern European faunas, is obviously not perfect, given the
present state of knowledge. According to Masini and Torre (1989),
the Italian Plio-Pleistocene biostratigraphy should be compared
with that of the regions to the north as seen in Table 17.2.
193
Late Ruscinian
Upper Middle Villafranchian:

Saint-Vallier fauna (= MN-17)
Lower Middle Villafranchian:
Montopoli fauna (MN-16b)
Lower Villafranchian:
Triversa fauna (MN-16a)
Perpignan (Csarnotian):
Serrat d'en Vaquer (= MN-15)
In Rheinhessen (Figure 17.1), east of Eppelsheim, fissure

fillings in Upper Oligocene limestones at Gundersheim (Heller,
1936) have yielded two local faunas of small mammals of different ages. The older is known from an isolated block from the
quarry floor, Gundersheim-Findling (Storch and Fejfar, 1989) or
Gundersheim 4 (Fejfar and Storch, 1990), and also from samples
in "Fissure nr. 4" at Gundersheim 3 (von Koenigswald and
Tobien, 1990, p. 234). The Findling block has produced a rich
micromammal fauna with 27 taxa, correlated to MN-15b (late
Ruscinian) age with eastern European and Siberian affinities.
A similar attribution is indicated for the rich collection from
Wolfersheim in the Vogelsberg area, which contains large and
small mammals (tapirs, several deer, Dicerorhinus, Anancus,
Mammut borsoni, Sus minor, and several beavers, as well as
microtines such as Mimomys occitanus and M. gracilis, sciurids,
cynomorphs, and insectivores, among them Desmana). The assemblage points to a swampy forested milieu (Tobien, in Boenigk
etal., 1977, p. 65; von Koenigswald and Tobien, 1990, p. 235). As
noted earlier (Table 17.2), the palynological age of the Wolfersheim bone-bearing bed in the upper Liegendton is uppermost
Brunssumian or Reuverian A, whereas the late Ruscinian mammal faunas of Gundersheim 1 and Wolfersheim belong to zone
MN-15 (Fejfar and Heinrich, 1989; Mein, 1989, p. 79).
Villanyan. An early Villanyan level, characterized by the MN-16
marker taxon Mimomys polonicus, is represented in river deposits
on top of lignites in the open-pit quarries of Frechen, near Koln
(Figure 17.1). The fossiliferous level is close to the GaussMatuyama paleomagnetic reversal at 2.48 Ma (Preface, this volume). According to palynological analysis, the lignite deposit corresponds to Reuverian B (Van Kolfschoten and Van der Meulen,
1986; von Koenigswald and Tobien, 1990). The boundary between Reuverian A and B may thus be correlated to the
Ruscinian-Villanyan boundary (Table 17.2), of later Pliocene age.
Numerous late Villanyan sites are known. At Gundersheim, yet
another micromammal assemblage, initially described by Heller
194
(1936) and later separated into two units, Gundersheim 1 and 2,

by Kretzoi (1962), has been revised by Storch and Fejfar (1989).
There are two views on the age of this assemblage: either it is an
early MN-17 fauna, with persistent primitive arvicolids, or it is a
mixture from two different levels (MN-16a and MN-17). In any
event, the former age assignment of late Ruscinian (MN-15) age
for Gundersheim 1 (Kretzoi, 1962; Tobien, 1980) is invalid. Fejfar
and Heinrich (1989, figure 2) have placed the entire assemblage,
as "Gundersheim 2," in MN-17 (i.e., late Villanyan).
A cave deposit at the entrance of the Baren Hohle near
Erpfingen, 50 km south of Stuttgart in the Swabian Alps, is
termed Erpfingen 2, not to be confounded with the nearby,
younger fissure fillings Erpfingen 1 and 3 of Heller (1958). This
site has produced large and small mammals that document a late
Villanyan age (Mein, 1989; von Koenigswald and Tobien, 1990).
The assemblage is closely comparable to the late Villanyan
Tegelen local mammal fauna in The Netherlands (Zagwijn,
Chapter 16, this volume). Other late Villanyan, MN-17 local
faunas come from fissure fillings in Upper Jurassic limestones at
Moggaster Hohle, Schernfeld, Deinsdorf, and Schambach in the
Franconian Alb (Figure 17.1) (Fejfar and Heinrich, 1989).
Present consensus identifies the Eburonian cold-climate phase
as the equivalent of the earliest Pleistocene, as defined in the
Vrica boundary. The late Villanyan local faunas of Gundersheim
and the Baren Hohle, being of the same age as the Tegelen local
fauna and its accompanying Tiglian warm flora underlying the
Eburonian levels, would therefore represent the last warmclimate interval before the beginning of the Quaternary.
Nieuwied basin near Koblenz, between Mainz and Cologne

(Figure 17.1), lie above the Matuyama-Brunhes paleomagnetic
reversal, at 0.78 Ma (Preface, this volume). The fauna of these
levels indicates a late Biharian age (von Koenigswald and
Tobien, 1990). The younger faunas (i.e., Toringian of Fejfar and
Heinrich, 1989) are beyond the scope of this chapter.
Flora*
Pollen and spores are of particular importance for stratigraphical

classification of the Pliocene and early Pleistocene, and thus as
well for determination of the boundary between the Tertiary and
Quaternary. Microflora can be found in almost all organogenic
deposits, and therefore many localities are known; macroflora,
on the other hand, are more seldom found. Zagwijn (1960,1963,
1974) divided the quite considerable Upper Miocene, Pliocene,
and early Pleistocene sequence in The Netherlands into seven
named units based on the changes in the palynoflora: from
bottom to top, Susterian; Brunssumian (with subunits A, B, C);
Reuverian (with subunits A, B, C); Pretiglian; Tiglian (with
subunits A, B, C); Eburonian; and Waalian. Zagwijn's classification serves today as the basis for all pollen-analysis research into
the question of the Tertiary-Quaternary boundary in central
Europe.
As for the fossil leaf and fruit floras in western Germany, there
are few sites. In the Reuverian there are Massenhausen near
Munich (Jung, 1963), Frankfurt am Main (Madler, 1939), and
Willershausen (Straus, 1967). For Tiglian macroflora, the bestknown site is Schwanheim, near Frankfurt (Baas, 1932). The
Biharian. Whereas the Villanyan local faunas are dominated by best assemblages are those from the clays of Reuver and Tegel
the arvicolid Mimomys, in the Biharian the first species of the themselves, situated in The Netherlands (Zagwijn, Chapter 16,
advanced arvicolid Microtus begin to appear in northern this volume).
European mammal communities, together with related forms. In
In regard to the microflora, for several decades the question of
western Germany the most significant sites of this age are the the Tertiary-Quaternary boundary has been linked with debates
fissure fillings of Weissburg 7 in Jurassic limestone of the about the significance of major climate changes evidenced in the
Swabian Alb (Figure 17.1) and Neuleiningen 15 (Fejfar and paleofloras of the lands bordering the North Sea basin. TerresHeinrich, 1989; listed as Neuleiningen 11) in Aquitanian trial deposits of the Reuverian Pliocene are always easily
limestones near Worms, Rheinpfalz (Table 17.1).
identifiable, with a dominance of distinctively Tertiary floral
Whereas the MN-17 fissure fillings with their warm-climate elements such as Sciadopitys, cf. Sequoia, Taxodiacea, Tsuga,
faunas represent the latest Pliocene (Villanyan) levels, these Nyssa, Liquidambar, Carya, Pterocarya, Castanea, Eucommia,
MQ-1 cold-climate faunas clearly represent the early Biharian, Symplocaceae (Symplocoipollenites rotundus and S. vestibulum),
and their correlation with the Eburonian (Fejfar and Heinrich, and the form-species Cupuliferoidaepollenites quisqualis, C.
1989, p. 103) identifies them with the earliest Pleistocene. The fallax, Tricolporopollenites exactus, and Araliaceoipollenites edEburonian of the lower Rhine Basin is represented in the mundi. By contrast, the warm periods of the succeeding
deposits of the "Hauptterrase 1" or "main terrace 1" (Figure paleofloral stages (from Tiglian to Waalian) are characterized by
17.2), so it can be assumed that the entire sequence below the microflora in which only a few typical Tertiary elements are to be
Hauptterrase 1 from Tonhorizon ("clay horizon") D down to found, such as Tsuga, Carya, Pterocarya, Ostrya, Castanea,
Tonhorizon Al are of pre-Biharian or Villanyan, late Pliocene Philodendron, and Eucommia. The decimation of the flora
age. The relationship to the Olduvai paleomagnetic event is during the Pretiglian cold period, as Zagwijn (1960) showed,
involved not only the disappearance of trees adapted to warmobvious (Figure 17.2).
Among later Biharian sites is Hohen-Siilzen, near Worms climate conditions but also a marked thinning of forest density.
(Rheinhessen, Table 17.1), where the gastropod marls contain Because of this marked change, the base of the Pretiglian was
mammalian fossils, inter alia at least 15 microtines, including
*This contribution reflects the state of knowledge up until 1982. The
three relict species of Mimomys (von Koenigswald and Tobien, author regrets that illness has prevented a thorough revision to include
1990). Beds E and F in the loess section of Karlich in the more recent data.
Kaena
GILBERT
Ville
CD
Frechen-lntergl III
a.
first Blocks of Drift

quat. Mollusc-Fauna
Multi-colored Pebbles
B1
Light grey Clay
Peat
Wetterau -
Clay Horizon A1
Main Brown
Frechen-lnterqll^
Fortuna-Osc
?
Quartz-Gravel
Reuver. C-Flora
plioc. Mo lusc-Fauna
A2
e
Frechen-lnterglH ^
Reuver. I 3-Flora
instable
Heavy Min
stable
C
Reuvenum
Coal
J Underlying Clay
e )
B2
Fauna of Wolfersheim
Brunssumian Flora
o
o
upper
Green Silt
Csarnotium
[ = younger
Ruscinium ]
o
o
Mammal
(Biharium)
(Weilerswist)
o
o
o
S 1
O
Mammoth
and
HI o o o o a
GAU
C/,
2.5-
Flora
( I ower)
ooooo
| Clay Horizon
(o) Gravel d
8 Clay Horizon
Gravel c
o
o Clay Horizon
Gravel b2
Clay Horizon
Gravel b1
'8
o
o Clay Horizon
o
o
Climate- Indicators
V i l t a f r a nchiun
'UYAMA
Olduvai
Horloff-Graben/Vogelsberg
Main Terraces
Series
Ville/Lower Rhine
Local Paleomagnetic Classification
195
Brunssumium
Susterium
Ruscinium
holder
Ruscinium ]
- normal
o - rtvers
magnetized
Figure 17.2. Stratigraphy in the terrestrial Tertiary-Quaternary boundary interval of western Germany
long used in this area as the equivalent of the TertiaryQuaternary boundary (Zagwijn, 1974).
Important sections with microflora of Plio-Pleistocene age are
found principally in the coastal regions, such as East Friesland
(Meyer, 1981) and Schleswig-Holstein (Menke, 1975), and in
structural depressions such as the lower Rhine Basin, the
Leinetal Graben (Benda and Liittig, 1968), the Rhein-Hesse
depression (including the Horloff Graben, the Seligenstadter
depression, and the Rhine/Main area), and the Upper Rhine
Graben.
periods are named Nordende, Ellerhoop, Tornesch, Uetersen,

and Pinneberg. Correlation of the Tornesch and the Waalian
paleoflora seems very likely, and the early warm-climate periods
correspond to the Tiglian complex. In The Netherlands and in
the lower Rhine Basin, there are as yet no known parallels to
Menke's youngest two warm-climate periods. Core-drilling at
Oldenswort 9 showed Lieth-series sediments overlying the
Reuverian and older Pliocene sections (Menke, 1975).
Lower Rhine palynofloras. It has been known for some time that
in the lower Rhine area there is a very thick and important
Coastal-plain palynofloras. The most important and best-studied sequence of Pliocene and Pleistocene deposits (Van der Vlerk
sequence is at Lieth near Hamburg, in Schleswig-Holstein, and Florschiitz, 1953). The problem of the Tertiary-Quaternary
where Menke (1975) described a sequence of paleofloras transition in the vegetational history has been discussed by
indicating five warm-climate periods separated by layers indicat- Urban (1978), who investigated profiles in the open-pit lignite
ing cold-climate periods. From bottom to top, the warm-climate mines on the Ville at Fortuna, Garsdorf, and Frechen, as well as
196
in clay pits in the Bruggen area near the German/Dutch border,

and by von der Brelie (1981), who analyzed samples from
boreholes drilled in the lower Rhine Basin. Urban discovered
that the Frechen I interglacial, characterized by a Fagus-Tsuga
association, is equivalent to Tiglian A, and in the Peter van Eyck
clay pit near Bruggen she found a previously unknown interglacial which she correlated with the Tiglian C. A Waalian
correlative may also be present in the latter section. On the basis
of the borehole evidence, clear parallels can be found between
the Tiglian and younger microfloras in the German and Dutch
parts of the lower Rhine Basin. The parallels in the Reuverian
microfloras are also in good agreement (von der Brelie, Hager,
and Kothen, 1981).
Paleofloras in other structural depressions. The Tertiary and
Quaternary sediments in the Horloff Graben are of great
importance (Boenigk et al., 1977). At Wolfersheim, a wellknown middle Pliocene (Csarnotian) mammalian fauna has been
collected from beneath the lignites, as discussed earlier. The
lignites, formerly thought to be of Brunssumian age, must now
be assigned a stratigraphic position of Reuverian A. In the lower
Rhine area the pollen analytical investigations of von der Brelie
et al. (1981) have shown that high proportions of Sequoia-like
pollen, previously thought to be an indicator of Brunssumian
floras, may also occur in the Reuverian A horizon. The lignites
of the Wohnbach open pit therefore can be dated as Reuverian A
and B, which is in better agreement with the mammalian
biochronology.
The Schwanheim site in the Rhine/Main area is well known
(Baas, 1932). Evidence for at least two warm periods of Tiglian
aspect can be found there (von der Brelie, 1974; Semmel, 1974),
and clays and paleosols with similar pollen assemblages can be
found in many places intercalated with fluvial deposits in the
northern and middle parts of the Upper Rhine Graben (von der
Brelie, in Bartz, 1976, 1982). The relics of some Pliocene pollen
forms allow these palynofloras to be distinguished from later
deposits, but attempts to define a more exact biostratigraphy
have been unsuccessful because of the discontinuity of exposures. Even the well-known clay deposits of Jockgrim and
Rheinzabern, although generally thought to belong to the early
Quaternary in the northern European climatic sequence (i.e.,
post-Pretiglian), have not been satisfactorily dated (Peters,
1965). Recent palynological investigations, together with regional geological considerations (Bartz, 1982), indicate that
Jockgrim is almost certainly younger than the Tiglian.
The deposits from Uhlenberg, near Augsburg in southern
Germany, remain to be mentioned (Scheuenpflug, 1979;
Schedler, 1979). A Tiglian age is indicated by the occurrences of
that adequate information exists to identify the TertiaryQuaternary boundary, whether at the base of the Pretiglian or at
a higher level, such as the base of Eburonian cold-climate phase
(Zagwijn, Chapter 16, this volume). In the Upper Rhine Graben
(Bartz, 1976, 1982) and in the Horloff Graben (Gruschkau,
1962; Boenigk et al., 1977) the deposits of that interval are
generally very poorly exposed, and because of slow and erratic
tectonic subsidence rates they have a complicated sedimentation
geometry; the stratigraphy, therefore, is almost impossible to
correlate. It must also be taken into consideration that the
deposits usually are somewhat condensed and normally exhibit
relatively thin interglacial vegetation phases.
Continental molluscan faunas
The upper Reuverian molluscan fauna in the A horizons of the
Ville shows a distinct similarity to the molluscan assemblage of
Hauterives, a later Ruscinian (medial Pliocene) site in southern
France (Strauch and Schlickum, in Boenigk et al., 1974; Mein,
1989). In the past, Reuverian flora were consistently classified as
Upper Pliocene, but the present mid-Pliocene attribution is more
in accord with the molluscan evidence. The mollusks in the B2
Clay horizon from the Ville demonstrate an astonishing faunal
change, similar to the paleofloral change at that level, as noted
earlier. According to Lozek (in Boenigk et al., 1972) the fauna
exhibits distinct warm-climate elements, but without any of the
earlier warm-climate taxa of the Reuverian. This may be seen as
an indication of an unprecedented severity of climate change and
a consequent high rate of extinction in northern European
molluscan faunas during the intervening cold-climate episodes
represented by gravels bl and b2; as noted earlier, pollen
analysis equates these gravels with the Pretiglian, and the B2
Clay with the Tiglian.
Paleomagnetism
The samples analyzed for paleomagnetism (Figure 17.2) from

the upper Brunssumian up to and including the Reuverian C
horizons show normal magnetization, attributed to the Gauss
normal chron. The polarity reversal in the uppermost Reuverian
C (i.e., in the A2 Clay of the Ville, in the lower Rhine Basin) is
therefore the beginning of the Matuyama reversed-polarity
chron. Considering the paleomagnetic sequence in the overlying
Pretiglian and Tiglian in The Netherlands (Zagwijn, Chapter 16,
this volume), it may be that the normal polarities of the Olduvai
subchron in the upper Tiglian are those recorded in the D Clay of
the lower Rhine.
Tsuga, Carya, Pterocarya, and Castanea.

Marine younger Neogene
Summary. Knowledge of the late Pliocene and early Pleistocene

plaeofloras in the southern part of western Germany remains
fragmentary, and most sites cannot be exactly (jated according to
available geological or palynological data (Frenzel, 1973). Until
now, it is only in Schleswig-Holstein and the lower Rhine area
Pre-Pleistocene sedimentation. The main part of the Neogene

(Table 17.3) can be divided into three sedimentation cycles in
this region: (1) an early Miocene cycle (23-17 Ma), with
subsidence in the Vierlandian (tmiv) and a regression, with
197
Table 17.3. Stratigraphic correlation of the Neogene and early Quaternary in the North Sea margins
Epoch
PLEIST.
PLIOCENE
Climate
Stage
BELGIUM
NETHERLANDS
Waalian
Waalian
Eburonlan
Eburonian
Tigllan
DClay
CClay
B2 Clay
Tiglian
Praetlgllan
(Lieth Series)
Mol B
(Praetigilian)
Reuverian
Brunsummian
Susterian
MIOCENE
NV/EST GERMANY,
JUTLAND
(Middle)
(Early)
Merxemian
Mol C
(u. Reuver)
Scaldisian
Mol D1
Kattendijkian
Mol D 2
Mol E
(Delden)
Mol F 1
Baventian
Thurnian
Ludhamian
Scaldisian
Morsumian
Syltian IV
Syltian III
Syltian II
Syltian I
Gramian II
Gramian I
Langenfeld. IV
Langenfeldian III
Langenfeldian II
Langenfeldian I
Diestian s.s.
Deurnian
Mol F 3 - 5
(Eibergen)
Reinbeckian
Oxlundian
Hemmoorian
Behrendorfian
Mol F 2
(Zenderen)
Mol G
(Stemerdink)
Anversian
Houthalian
Edegemian
Mol H
(Miste)
Vierlandian
OLIGOCENE
Neochattian
fluviatile sands from the east, in the Hemmoorian (tmih); (2) a

later Miocene Elbe cycle (17-5 Ma), with Reinbekian subsidence (tmir) reaching a maximum at the tmirltmil boundary
(Levensau and Liineburg subzones, Tostedt Member and
Eibergen Member), and with regression after the late Langenfeldian (tmild).
In regard to the third cycle, the paleogeography and biofacies
distributions of the succeeding Gramian and Syltian are discussed in Hinsch (1990). During the Syltian, marine strata of the
younger Syltian and Morsumian were restricted to the west (e.g.,

the Morsum Cliff on Sylt, and Wursterheide, near Cuxhaven) by
prograding fluviatile sedimentation of the Oldesloe Formation.
The surface exposures at Morsum Cliff (Hinsch, 1984, 1985),
which have been deformed by glacial tectonics, have been
correlated to undisturbed well sections on Sylt by Hinsch (1990).
There and at Wursterheide (Hinsch, 1989) the SyltianMorsumian boundary is observed in marine strata, in which the
Syltian molluscan fauna is very rich (Hinsch, 1977) and the
198
Morsumian is rather poor. Well-preserved calcareous molluscan

fossils of both Syltian and Morsumian age were noted (Hinsch,
1991) in the offshore well R-l (5500'N, 649'E). The Morsumian
represented in the mammalian biostratigraphy at a level between

the Gundersheim 2 Villanyan level and the transitional
Villanyan-Biharian assemblages found in the Neuleiningen,
fauna of R-l, with Spisula arcuata, Cingula inusitata, Telasco Schernfeld, and Deinsdorf local faunas. In marine molluscan
syltensis, Uzita serrata, Fusiturris Helena, Mangelia nysti, and faunas, that level can be equated with the local extinction of
Philbertia perpulchra, can be correlated to the Kattendijkian.
Acila cobboldiae in offshore boreholes. On the other hand, in
Later Pliocene (Scaldisian) faunas have been found in offshore terms of the molluscan paleontology in onshore marine deposits,
wells: 89/5 (5407'N, 219'E), with the Acila-Astarte association as well as for palynology and paleomagnetic data, the Tiglianoverlain by beds with the Cerastro derma hostiei biofacies, and Eburonian boundary, which presently equates to the PlioceneH15/2, west of Helgoland (5410'N, 7E), which records the first Pleistocene boundary, is not well represented in the geologic
appearance of Dosinia imbricata.
record of this area.
An Icenian, or middle Tiglian, molluscan fauna of the
Bramertonian biofacies, with Acila cobboldiae, was also found in
References
the 89/5 borehole. The Bridlingtonian biofacies, with Tridonta
borealis withami, also probably of Icenian age, was reported by Baas, J. 1932. Eine friihdiluviale Flora im Mainzer Becken.
Hinsch (1991) from well 89/3 (5427'N, 547'E).
Zeitschr. Bot. 25:289-371.
The base of the Quaternary, or tpllqp boundary, was assumed Bartz, J. 1976. Quartar und Jungtertiar im Raume Rastatt. Geol.
L.-AmtBaden-Wurtt., Jb. 18:121-178.
by Lagaaij (1952) and Glibert and de Heinzelin (1955), following
Bartz, J. 1982. Quartar und Jungtertiar II im Oberrheingraben
Wirtz (1949), to be located above the Morsumian at the base of
im Grossraum Karlsruhe. Geol. Jb. A63:1-237.
the Merxemian-Waltonian. This is equivalent to a level within Benda, L., von Gaertner, H. R., Herrmann, R., Liittig, G.,
the Reuverian. Zagwijn (1974), considering paleoclimatic eviStreif, H., Vinken, R., and Wunderlich, H. G. 1968.
Kanozoische Sedimente in tektonischen Fallen und Subdence, proposed that the boundary be placed instead between
rosionssenken in Siid-Niedersachsen. Deutschl. geol. Ges.,
the Reuverian and Pretiglian, or equivalent to the top of the
Zeitschr. 117:713-726.
Merxemian-Waltonian. Finally, the definition in the Vrica Benda, L., and Liittig, G. 1968. Das Pliozan von Allershausen
stratotype in Calabria would result in a boundary between
(Soiling, Niedersachsen). Palaeontographica, Abt. B 1968:
Tiglian and Eburonian pollen-floral stages (Zagwijn, Chapter
221-236.
16, this volume). Neither of the latter two boundary definitions Boenigk, W. 1978a. Zur Ausbildung und Entstehung der
jungtertiaren Sedimente in der Niederrheinischen Bucht.
can be represented by any known marine faunas from onshore
Kolner Geograph. Arb. 36:59-68.
exposures in northwestern Germany.
Boenigk, W. 1978b. Gliederung der altquartaren Ablagerungen
On the other hand, in the western part of the southern North
in der Niederrheinischen Bucht. Fortschr. Geol. Rheinl.
Sea, the Pliocene and most Pleistocene interglacials are repreWestf. 28:135-212.
sented by marine horizons with abundant molluscan fauna. With Boenigk, W., Koci, A., and Brunnacker, K. 1979. Magnetostratigraphie im Pliozan der Niederrheinischen Bucht.
regard to the approved new definition of the PlioceneN. Jb. Geol. Paldont., Mh. 1979:513-528.
Pleistocene boundary, the horizon which corresponds to the Boenigk, W., Kowalczyk, G., and Brunnacker, K. 1972. Zur
Tiglian-Eburonian boundary can be recognized by the disappearGeologie des Altestpleistozans der Niederrheinischen
ance of Acila cobboldiae in the North Sea record. Reworked
Bucht. Deutsch. Geol. Ges., Zeitschr. 123:119-161.
shells from the Tiglian, as well as underlying Pliocene strata, can, Boenigk, W., von der Brelie, G., Brunnacker, K., Kempf, E. K.,
Koci, A., Schirmer, W., Stadtler, G., Streit, R., and
however, be found in the transgressive interglacials, as, for
Tobien, H. 1977. Jungtertiar und Quartar im Horloffexample, in basal Eemian sediments in borehole 89/2 (54N,
Graben/Vogelsberg. Hess. L.-Amt Bodenf., Abh. 75:1-90.
5E), as well as in borehole 89/4.
Boenigk, W., von der Brelie, G., Brunnacker, K., Koci, A.,
Schlickum, W. R., and Strauch, F. 1974. Zur PliozanPleistozan-Grenze im Bereich der Ville (Niederrheinische
Bucht). Newsl. Strat. 3:219-241.
Conclusions
Brunnacker, K. 1975. Der stratigraphische Hintergrund von
Floral and mammalian biostratigraphic data from the Ville and
Klimaentwicklung und Morphogenese ab dem hoheren
Pliozan. Z. Geomorph., N.F, Suppl. Bd. 23:82-106.
from Vogelsberg related to the Tertiary-Quaternary boundary
are synthesized in Figure 17.2. The Ruscinian-Villanyan bound- Fejfar, O., and Heinrich, W. 1989. Murold biochronology of the
Neogene and Quaternary. In European Neogene Mammal
ary is correlated with the boundary between Reuverian A and B.
Chronology, ed. E. H. Lindsay et al., pp. 91-117. Paris:
According to the integration of the floral sequence into the
North Atlantic Treaty Organization, A.S.I. Series no. 180.
paleomagnetic scale, the Reuverian-Pretiglian boundary, with Fejfar, O., and Storch, G. 1990. Eine pliozane (oberruscinische) Kleinsaugerfauna aus Gundersheim, Rheinthe beginning of severe cold-climate conditions seen in many
hessen
1. Nagetiere: Mammalia, Rodentia. Senckenb.
stratigraphic sequences of western Germany, would be approxiLeth. 71:139-184.
mately 2.5 Ma, close to the Gauss-Matuyama boundary. The Frenzel, B. 1973. Some remarks on the Pleistocene vegetation.
next major refrigeration, at the Tiglian-Eburonian boundary,
Eisz. Gegenw. 23/24:281-292.
dating to the top of the Olduvai normal-polarity subchron, is Glibert, M., and de Heinzelin, J. 1955. La faune et l'age Miocene
199
superieur des Sables de Deurnes. Inst. Roy. Sc. Nat. Belg., Menke, B. 1975. Vegetationsgeschichte und Florenstratigraphie
Nordwestdeutschlands im Pliozan und Fruhquartar. Mit
Bull., 31(71):l-27; 31(72):1-12.
einem Beitrag zur Biostratigraphie des Weichsel-FriihglazGruschkau, H. 1962. Pollenanalytische Untersuchungen der
Wetterauer Braunkohle und iiber holarktische Ericaceen.
ials. Geol. Jb. A26:3-148.
Unpublished dissertation, Universitat Giessen.
Meyer, K.-J. 1981. Zur Stratigraphie des kontinentalen Pliozans
Guenther, E. W., and Mai, H. 1977. Die pleistozanen Schichten
in NW-Deutschland mittels pollenanalytischer Untersuchvon Jockgrim in der Rheinpfalz. Naturw. Ver. Schlesw.uingen. Newsl. Stratigr. 10:1-19.
Holst., Schr. 47:5-24.
Peters, I. 1965. Zur Altersstellung der Tofe und Gyttjen von
Heller, F. 1936. Eine oberpliozane Wirbeltierfauna aus RheinhesHerxheim, Jockgrim und Rheinzabern in der Vorderpfalz.
sen. N. Jb. Min. Geol. Pal., Bell. Bd. 76B:99-160.
Eisz. Gegenw. 16:121-131.
Heller, F. 1958. Eine neue altquartare Wirbeltierfauna von Schedler, J. 1979. Neue pollenanalytische Untersuchungen am
Erpflngen (Schwabische Alb). N. Jb. Geol. Palaont., Abh.
Schieferkohlevorkommen des Uhlenberges bei Dinkelscherben (Schwaben). Geol. Bavar. 80:165-182.
107:1-101.
Hinsch, W. 1977. Die Molluskenfauna des Syltium vom Morsum- Scheuenpflug, L. 1979. Der Uhlenberg in der ostlichen IllerLech-Platte (Bayerisch-Schwaben). Geol. Bavar. 80:159Kliff. Naturw. Ver. Schlesw.-Hoist., Schr. Mr:39-56.
164.
Hinsch, W. 1984. In Das Neogen im Raum Sylt: Exkursionsfiihrer
Erdgeschichte des Nordsee und Ostseeraumes, pp. 217- Schniitgen, A. 1974. Die Hauptterrassenfolge am linken Niederrhein auf Grund der Schotterpetrographie. Unpublished
248. Universitat Hamburg, Geologisch-palaontologischer
dissertation, Universitat Koln.
Institut.
Hinsch, W. 1985. Fossillagerstatten des marinen Neogen auf Schniitgen, A., and Brunnacker, K. 1977. Zur KieselschieferFiihrung in Schottern am Niederrhein. Decheniana 130:
Sylt. Fossilien von Sylt 1:117-131.
Hinsch, W. 1986a. Lithologie, Stratigraphie und Palaogeo293-298.
graphie des Neogens in Schleswig-Holstein. Beitr. Reg. Semmel, A. 1974. Der Stand der Eiszeitforschung im RheinGeol. Erde 18:22-38.
Main-Gebiet. Rhein-Main. Forsch. 78:9-56.
Hinsch, W. 1986b. The Northwest-German Tertiary basin Storch, G., and Fejfar, O. 1989. Gundersheim-Findling, a
Miocene and Pliocene. Beitr. Reg. Geol. Erde 18:679-699.
Ruscinian rodent fauna of Asian affinities from Germany.
Hinsch, W. 1989. Die Molluskenfauna und sonstige Mesofauna
In European Neogene Mammal Chronology, ed. E. H.
in der Forschungsbohrung Wursterheide. Geol. Jb. A l l l :
Lindsay et al., pp. 405-412. Paris: North Atlantic Treaty
233-286.
Organization, A.S.I. Series No. 180.
Hinsch, W. 1990. Subdivision and palaeogeography of the Straus, A. 1967. Zur Palaontologie des Pliozans von WillershauGramian and Syltian stages (Late Miocene) in Schleswigsen. Berliner Natur-hist. Ges., Abh. 111:15-24.
Holstein and Wursten (NW Germany). Ten. Res. 11:159- Tobien, H. 1951. Die Aufzeichnungen H. G. Stehlin's iiber die
177.
pliozanen Saugerreste von Herbolzheim bei Freiburg i. Br.
Hinsch, W. 1991. Molluscan faunes in sections of Southern North
Baden, geol. Landesamt., Mitt. Bl. 1950:78-84.
sea. Report SNP BH 89 (unpublished).
Tobien, H. 1975. Pleistozane Warmzeiten und Saugetiere in
Jung, W. 1963. Blatt- und Fruchtreste aus der Oberen SiisswasserEuropa. Quartdrpaldontologie 1:221-233.
molasse von Massenhausen, Kreis Freising (Oberbayern). Tobien, H. 1980. Saugerfaunen von der Grenze Pliozan/
Palaeontographica, Abt. B 112:119-166.
Pleistozan in Rheinhessen, 1: die Spaltenfiillungen von
Koci, A., Schrimer, W., and Brunnacker, K. 1973. PalaomagGundersheim bei Worms. Mainzer geowiss. Mitt. 8:209netische Daten aus dem mittleren Pleistozan des Rhein218.
Main-Raumes. N Jb. Geol. Palaont., Mh. 1973:545-554. Urban, B. 1978. Vegetationsgeschichtliche Untersuchungen zur
Kolumbe, E. 1963. Die interglazialen und interstadialen AblagerGliederung des Altquartars der Niederrheinischen Bucht.
ungen von Steinbach bei Baden-Baden. Oberrhein. geol.
Univ. Koln, Geol. Inst., Sonderveroff. 34:1-165.
Abh. 12:25-43.
Van der Vlerk, I. M., and Florschiitz, F. 1953. The palaeontoKretzoi, M. 1962. Fauna und Faunenhorizont von Csarnota.
logical base of the subdivision on the Pleistocene in the
Fold. Intezet. Evi Jelentesi 1959:297-395.
Nederlands. Kon. Ned. Akad. Wetens., Afd. Naturk. 1
Kretzoi, M. 1965. Die Nager und Lagomorphen von Voigstedt in
Reeks 20:1-58.
Thiiringen und ihre chronologische Aussage. Palaont. Van Kolfschoten, T., and Van der Meulen, A. J. 1986. Villanyan
Abh., Abt. A2, 1965:587-660.
and Biharian mammal faunas in The Netherlands. Soc.
Lagaaij, R. 1952. The Pliocene Bryozoa of the Low Countries.
Geol. Ital. Mem. 31:191-200.
Geol. Sticht., Med., C-V 5:1-233.
von der Brelie, G. 1974. Pollenanalytische Untersuchungen an
Leschick, G. 1951. Mikrobotanisch-stratigrahische Untersuchwamrzeitlichen Sedimenten in den Terrassen des
ung der jungpliozanen Braunkohle von Buchenau (Krs.
Untermainz-Gebietes. Rhein-Main. Forsch. 78:83-99.
Hiinfeld). Palaeontographica, Abt. B 92:1-51.
von der Brelie, G. 1981. Die Grenze Pliozan/Pleistozan und das
Madler, K. 1939. Die pliozane Flora von Frankfurt am Main.
Tiglium in der Erst-Scholle (Niederrheinische Bucht) aus
Senckenb. naturf. Ges., Abh. 46:1-202.
palynologischer Sicht. Univ. Koln, Geol. Inst., SonderMasini, E, and Torre, D. 1989. Large Mammal dispersal events
veroff. 41:29-42.
at the beginning of the Late Villafranchian. In European von der Brelie, G., Hager, H., and Kothen, H. 1981. Neue
Neogene Mammal Chronology, ed. E. H. Lindsay et al.,
Gesichtspunkte zur pollenstratigraphischen Gliederung
pp. 131-138. Paris: North Atlantic Treaty Organization,
des Pliozans in der Niederrheinischen Bucht. Fortschr.
A.S.I. Series No. 180.
Geol. Rheinl. Westf 29:265-274.
Mein, P. 1989. Updating of MN Zones. In European Neogene von Koenigswald, W., and Tobien, H. 1990. Important Arvicolid
Mammal Chronology, ed. E. H. Lindsay et al., pp. 131faunas from the late Pliocene to early Holocene in western
138. Paris: North Atlantic Treaty Organization, A.S.I.
Germany. In International Symposium. Evolution, PhylogSeries No. 180.
eny and Biostratigrapahy of Arvicolids (Rodentia, Mam-
200
malia), ed. O. Fejfar and W.-D. Heinrich, pp. 231-253.

Prague: Geological Survey.
Weyl, R., Rein, U , and Teichmiiller, M. 1955. Das Alter des
Sylter Kaolinsandes. Eisz. Gegenw. 6:5-15.
Wirtz, D. 1949. Die Fauna des Sylter Crag und ihre Stellung im
Neogen der Nordsee. Geol. Staatsinst. Hamburg, Mitt.
19:57-76.
Zagwijn, W. H. 1960. Aspects of the Pliocene and Early
Pleistocene vegetation in The Netherlands. Geol. Stichting, Meded., Serie CIII-j 15:1-78.
Zagwijn, W. H. 1963. Pollen-analytic investigations in the
Tiglian of The Netherlands. Geol. Stichting, Meded., n.s.
16:49-69.
Zagwijn, W. H. 1974. The Pliocene-Pleistocene boundary in
18
The Pliocene-Pleistocene boundary in eastern Germany

HANS-DIETRICH KAHLKE, LOTHAR EISSMANN, and FRIEDRICH WIEGANK
Introduction and geological background
In the late Neogene, the sea progressively retreated from wide

areas of the northwestern lowlands of the North German-Polish
coastal basin. Large river systems that developed in the exposed
areas drained generally toward the west, laying down great
amounts offluviatile-limnicsediments. The lithology of pebbles
contained in these sediments indicates that the major source area
was in what is now the eastern Baltic Sea, between central
Sweden and the Baltic countries. On the other hand, the
presence of lydite (radiolarian chert) demonstrates that some
material was also derived from the south, primarily the Bohemian Massif (Hucke, 1928; Ahrens et al., 1968; Duphorn et al.,
1973; Ahrens and Lotsch, 1976). In the southern parts of
Thuringia, in the area of the Thiiringer Wald, there was limited
late Neogene sedimentation in subsident areas (Figure 18.1).
Northern lowlands and northeastern region
In the northern lowlands, Lower Pliocene terrestrial deposits

appear to have been extensive, but only in southwestern
Mecklenburg have erosion relicts been preserved. As far as we
know, deposits of Plio-Pleistocene transitional series can be
found only in the southwestern parts of this area (Ahrens et al.,
1968).
The best-known locality of this type is Ruterberg, where the
lower horizons of the sequence (Bergton-Diatomit-Silbersand
complex) are followed discordantly by the Loosener Kiese
(Gehl, 1958; von Bulow, 1964), a unit characterized by a
mixture of "southern" pebbles (Kieselschiefer, etc.) and silicified rocks and intensely weathered granite from the north.
The Bergton-Diatomit-Silbersand complex was formerly correlated by Krutzsch (1959) to the Reuverian or a still earlier stage
on the basis of pollen, but more recently this unit has been
correlated with the Upper Miocene (Lotsch, 1981; Krutzsch,
1988). The Loosener Kiese deposits have been correlated with
the kaolin sands of the Isle of Sylt by Gehl (1958), who
correlated both units to the Upper Pliocene, whereas von
Biilow (1964) suggested an early Pleistocene age (cf. von der
Brelie et al., Chapter 17, this volume). The deposits of the
Loosener Kiese provide a fossil flora (H. Mai, in Ahrens, et al.,

1968) showing some affinities with the Nordhausen association.
It was recently referred to a time interval characterized as
"post-Reuverian" but not typical Tiglian (TC5) by Krutzsch
(1988).
Pliocene gravels and climate events
In eastern Mecklenburg and in northeastern Brandenburg

numerous deposits of quartz sands, with pebbles of silicified
rocks of Baltic origin, were originally referred to the Pliocene
(Hucke, 1928) or even the Pleistocene (Stolley, 1900) because
the pebble association was similar to the kaolin sands of the Isle
of Sylt. In contrast, Berger (1941), Quitzow (1953), and Ahrens
and Lotsch (1976) suggested a Miocene age for these deposits
because of the mode of deposition.
Recently, deposits of quartz sand showing the typical "Pliocene" pebble association have been found, unambiguously dated
to the Miocene, in large lignite pits much farther south in the
area of Cottbus, Finsterwalde, and Lubben. From the lower part
of the quartz-sand series (0.5 m above the second Lausitz lignite
seam) in the lignite open pit at Seese, there is a diverse fossil
flora which belongs to the Miocene floral zone VIII (Mai, 1967).
The quartz-sand series at Petersdorf and Finkenheerd is also
middle Miocene age (Ahrens and Lotsch, 1976). The use of the
Baltic pebble association as a criterion of Pliocene age clearly is
not justified (Krause, 1933; Berger, 1941; Quitzow, 1953).
Quartz-lydite gravels, reflecting a southern origin, are quite
widely distributed in the Lausitz area, and to a lesser extent in
northeastern Brandenburg. Mielecke (1929, 1934) interpreted
them as a southern Pliocene facies correlated with the northern,
supposedly Pliocene quartz-sand facies described earlier. To
account for those deposits, Genieser (1955) suggested a Pliocene
"Senftenberger Elbelauf" (a channel of the Elbe River containing the Senftenberg Elbe gravels) and a succeeding pre-glacial
"Bautzener Elbelauf."
According to Ahrens et al. (1968) and Ahrens and Lotsch
(1976), characteristic Senftenberger Elbelauf deposits with
quartz-lydite gravels can be dated not only to the Upper Pliocene
or Lower Pleistocene but also as far down as the first Lausitz
201
Kahlke, Eissmann, and Wiegank
202
(D
C
3 CO
Q)
12
C </>
o c
C 13
CD
>
_g m
I IS g
U_ Z S 5
Cromer#Komplex
JL
073
CromerKomplex
12
Horschlitt
Porsten
Artern
o
Menapium
<9y
0,9'.
1,0
nr
Unstrut
Wehlen 1
25
Untermanfeld
jungerer
Zersatz- - -
15
Domsen
MahlTs\@
Eburonium
Untermaflfeld^
DD D
Gerstungen \
Mittlerer Tonkopf
Schwallungen
grobschotteralterer
Komplex ~~
C6
Breitungen
Ddnischer
Berg
Elbe A2
C5
Tiglium
rr
Zersatz der
Zersatzkiese
2,12
2U
KleingieDhubel
RippersrodaNv
Nordhausen ^
Zersatzkiese
Bittstedt
Figure 18.1. Pliocene and Pleistocene localities in eastern Germany.

Paleomagnetic interpretation by F.
Wiegank (1982): 1, normal polarity; 2, reverse polarity; 3, position
of the section with normal polarity; 4, position of the section with
reverse polarity; 5, hiatus; 6,
paleomagnetic anomalies; 7, fossil
fauna; 8, fossil flora. The ZersatzGrobschotter-Komplex comprises
highly weathered coarse gravels
and boulders deposited in coldclimate conditions, the Zersatzkiese consists of highly weathered
gravels, and the Tonig-kohliges
Oberpliozan is composed of clay
and lignite deposits of the Upper
Pliocene. The terms "alterer" and
"jiingerer" refer to older and younger, respectively.
Praetiglium
Elbe
Klotzsche
Sulzfeld\
Reuvenum
tonig-kohliges
Oberpliozan
(altensundheim'^v
Berga t
2,92 I
Haselbach
Brunssumium
315
3 5-1
Plio-Pleistocene of eastern Germany
lignite layer in the Rauno sequence of that area. As noted earlier,

the Lausitz lignites are of Miocene age, and in fact the lower
horizons of the quartz-lydite gravels contain microflora in clay
and lignite beds at Klettwitz, Rauno, and Welzow that correspond
tofloralzone XIII of the Upper Miocene (Mai, 1967; Ahrens and
Lotsch, 1976). Nearly 98% of the Senftenberger Elbe gravels are
composed of quartz and other weathering-resistant components,
and the remaining feldspathic cobbles are heavily decomposed.
The stable heavy minerals also predominate over the unstable
ones (Genieser and Diener, 1958; Wolf, 1980).
In the area of Ottendorf-Okrilla, where such gravels are locally
up to 40 m thick, a peculiar characteristic is the occurrence of
oversized "drift boulders," which are unknown in early and
middle Tertiary deposits of the southern edge of the northern
lowlands. The drift boulders, moreover, increase in number from
bottom to top. Numerous possible "cryogene structures" have
been reported, such as the wedge-like structures interpreted by
Genieser (1955) and Prager (1976) as fossil ice wedges. This is
difficult to reconcile with the warm-temperate TaxodiumLiquidambar-Ulmus macrofloral association from silty layers in
the same gravel body. Parrotia fagifolia and P. pristina are also
common (Jahnichen, 1968; Walther, in Kube, 1979).
According to the fossil flora and the terrace stratigraphy, an
early Pliocene age for the Senftenberger Elbe gravels is the most
probable, although a middle Miocene to late Miocene age cannot
be excluded (Ahrens et al., 1968). The cryo-structures may
reflect temporary climatic cooling. Wolf (1980) recently distinguished two members within the Senftenberger Elbe gravels: a
"240-m" member and an "Al" member. According to paleomagnetic investigations (Wiegank, 1982), the Al gravel body has
primary reversed polarity, in agreement with a correlation (Wolf,
1980) to the Pretiglian.
Lower Pleistocene gravels
The next in the sequence of Elbe gravels, the deposits of the
Bautzener and Schildauer Elbe channels, follow valley stream
systems that are clearly seen in the hilly areas. They are very
readily distinguishable from the Senftenberger gravel type in
having only 50-80% quartz and often 20-30% poorly resistant
feldspathic material. Similarly, the unstable heavy-mineral component predominates over the resistant.
Cryoturbation structures and ice-wedge pseudomorphs produced by intense frost activity are recognized in these gravels
from several different localities (Schubert, 1980). The gravels of
the Bautzener and Schildauer Elbe channels (A2 and E valley
levels) are deposits of a glacially influenced fluviatile cycle (Wolf,
1980). Fine-grained sediments of the A2 sequence at Kleingiesshiibel are normally polarized, and Wiegank (1982) postulated
that sedimentation occurred during either the Reunion subchron
or the Olduvai subchron of the lower Matuyama.
In the area west of the Elbe River, four individual gravel
terraces are older than the Elsterian glaciation. Those terraces
are clearly distinguishable, because of their heavy minerals and
gravel composition, from the quartz gravels mentioned earlier.
In that group, with the exception of the youngest terrace, which
203
has been correlated with the early Elsterian, the older gravel
bodies formerly were interpreted as either pre-glacial or Pliocene
in age. In all those gravel bodies, permafrost patterns, including
ice wedges, are developed (Eissmann, 1975, 1981). Today we
correlate those three earlier accumulations with the Briiggen,
main Eburonian, and Menapian cold-climate phases. The
Briiggen has been identified with the Pretiglian by some
researchers, and by others with the early Eburonian. The postBruggen limnic sequence of Zeuchfeld-Borntal, however, with
Fagotia acicularis, Valvata naticina, Lithoglyphus pyramidatus,
and Discus perspectivus (Mania, 1973), is correlated with the late
Pliocene Tiglian climate stage of the lower Rhine Basin.
Thuringian basin
In the southwestern parts of eastern Germany, between the Harz

and the Thuringer Wald mountains, very localized late Neogene
sedimentation is known. In contrast to the northern lowlands, all
recorded deposits of the southern area are of late Pliocene-early
Pleistocene age. Despite the considerable distances between the
sites, these uppermost Pliocene and lower Pleistocene sequences
show a general correspondence of characters because they were
accumulated under similar conditions, albeit in isolated closed
depressions.
The lignite layers of Gerstungen (Mai and Walther, 1988),
Kranichfeld (Mai, 1965; Mai and Walther, 1988), Oberzella, and
Berga (Krutzsch and Majewski, 1965; Mai and Walther, 1988) are
characterized by fossilflorascorrelated with the Dutch Reuverian
or an even older stage (Brunssumian?). The key taxa here are
mainly exotic elements like Nyssa, Sciadopitys, Platycaria,
Liquidambar, Sequoia, Taxodium, and Symplocos. The basin
clays of Berga, containing a "Reuver flora," are normally
polarized and have been correlated by Wiegank (1982) with the
upper Gauss chron. The microflora of Gerstungen, with Tsuga,
Ilex, Carya, Pterocarya, Sciadopitys, and Nyssa and other
palynomorphs, is characterized by a still greater number of exotic
elements, again suggesting correlation with the Dutch Reuverian
or, according to Mai and Walther (1988), possibly with the
Brunssumian, because of a greater number of ancient ("Miocene") elements in that macroflora. A similar but much richer
microflora is known from the borehole of Gorsbach, near Kelbra,
showing high facies differences in comparison with the nearby
"Reuver" (or older?) flora of Berga (Krutzsch and Majewski,
1965). In the lignite horizons of Kranichfeld (Mai, 1965; Mai and
Walther, 1988) a fossil macroflora has been described, with about
47% of its taxa being exotic. The last appearances of some of
these taxa are in deposits of Reuverian age.
The Mastodon-Schotter (gravels with mastodont remains) of
Siilzfeld, with Mammut borsoni, Anancus arvernensis, Tapirus
arvernensis, and Dicerorhinus sp., are normally polarized and
have been correlated by Wiegank (1982) with the upper Gauss
chron. Based on pollen analysis of samples from the Mammut
borsoni horizon, from the doline filing of Kaltensundheim near
Meiningen, Ukrainzeva (in Kahlke and Ukrainzeva, 1986)
suggested a correlation of that horizon with the Dutch Pretiglian,
but Krutzsch (1988) postulated a much earlier correlation (lower
204
Kahlke, Eissmann, and Wiegank
Villafranchian climatic minimum). According to Mai and

Walther (1988), the fossil macroflora of Kaltensundheim suggests the "first real intra-late-Pliocene phase of cooling and
impoverishment of the flora, because this flora with 65 species is
not poor in species, but poor in exotic elements." The
mammalian assemblage from the deposit suggests a later
Pliocene (early Villafranchian) age, on the basis of Hypolagus
sp., Mammut borsoni, and a primitive cervid (Bdhme, 1963).
The horizon has normal remanent paleomagnetism (Wiegank,
1982) and has been correlated with the upper Gauss chron.
In the later group, the fossil macroflora of Nordhausen (Mai,
1965; Mai and Walther, 1988) comprises, in additon to 39 species
found in the recent European flora, stratigraphically important
exotic taxa such as Salvinia cerebrata, Carpolithus (Epipremnum) ornatus, and Decodon bashkiricus, as well as Spirematospermum wetzleri, previously found only in Pliocene deposits. In
the pollen spectra of Nordhausen (Erd and Majewski, in Ahrens
et al., 1968), Pinus, Alnus, Tsuga, Picea, Graminaceae, and
Osmunda are found regularly. Other taxa, like Abies, Quercus,
and Tilia, rarely exceed 1%. Apart from the relatively high
Tsuga pollen levels, the other exotic taxa recorded are
Sciadopitys and Eucommia. According to pollen analyses, Erd
(in Ahrens et al., 1968) correlated the fossiliferous deposits of
Nordhausen with the Tiglian A, whereas Mai and Walther (1988)
suggested an "uppermost Pliocene" level according to the
macrofloral record. The fossiliferous layers show predominantly
reversed remanent polarity and have been correlated with the
lower Matuyama (pre-Olduvai) interval by Wiegank (1982).
North of the Thiiringer Wald, mainly in the area of Ohrdruf on
the terrain known as Ohrdrufer Muschelkalkplatte, a formation
of highly weathered gravels is known as the Zersatz-Kies
complex (Ziegenhardt, 1965). The most important locality for
those highly weathered gravel deposits north of the Thiiringer
Wald is the Rippersroda II horizon (Mai, Majewski, and Unger,
1963; Schramm, 1964; Unger, 1964, 1974). In general, the
Zersatz-Kies complex discordantly overlies the Triassic limestone (Muschelkalk) of this area. Southeast of Rippersroda,
however, the heavily weathered gravels discordantly cover the
limnic filling of a doline, which includes the Rippersroda I
lignite. The fossil flora from the Rippersroda I lignite layers,
with 33% exotic species such as Tsuga, Carya, and Sciadopitys
(Mai et al., 1963; Mai and Walther, 1988), is associated with the
Rippersroda II fauna of fossil mammals from the overlying
Zersatz-Kies complex (Dietrich, 1953; Kahlke, 1968), which
includes the following species: Mimomys sp., Trogontherium
minor, Anancus arvernensis, Dicerorhinus etruscus, Eucladoceros ernestii, Cervidae gen. et sp. indet. (small cervid), and
Leptobos sp.
Following intensive erosion and deep weathering of the
Zersatz-Kies complex in the area north of the Thiiringer Wald,
the Zersatz-Grobschotter complex (highly weathered coarse
gravels and boulders) was deposited (Ziegenhardt, 1965), with
evidence of having been accumulated during cold conditions.
Similar coarse gravels of the Werra River area were also
accumulated during cold conditions, as proved by Siegel (1959).
The local extreme thicknesses of the layers, as well as the mode

of deposition, have been explained as the result of syngenetic
and postgenetic subrosion of the deeper horizons (Siegel, 1959),
and Ellenberg (1969) suggested that these Werra gravels were
the equivalents of the Zersatz-Grobschotter complex north of
the Thiiringer Wald. The Basiskiese ("basal gravels") of
Voigtstedt, near Sangerhausen, are thought to have accumulated
at about the same time (Cepek, 1968).
The weathered coarse gravels of the Zersatz-Grobschotter
complex north of the Thiiringer Wald and their equivalents of the
Werra River system (Danischer Berg, Breitungen, Gerstungen,
Schwallungen, Mittlerer Tonkopf) have been correlated according to lithostratigraphic and magnetostratigraphic criteria with
the Matuyama epoch between the Olduvai and the Jaramillo
subchrons by Wiegank (1982).
Pliocene-Pleistocene boundary horizons of eastern
Germany
The proposal made by the INQUA Subcommission 1-d and the

IGCP-41 working group at the 1982 INQUA meeting in Moscow,
to define the Neogene-Quaternary boundary-stratotype in the
Vrica section, has been approved by international bodies
(Foreword, this volume). The boundary is placed at the base of
the shale layer overlying sapropelic layer e, within the uppermost
part of the Olduvai event, according to present information
(Pasini and Colalongo, Chapter 2, this volume). The few
deposits correlative to this level in eastern Germany indicate that
the boundary should be drawn between (a) the level of
Nordhausen and Rippersroda II (correlated to the late Pliocene
according to fossil macroflora and fossil mammals, respectively)
and (b) the Zersatz-Grobschotter complex of the Thiiringer
Wald and its equivalents in the Werra River system (lowermost
Pleistocene). In the area of the Elbe River the A2 valley fill
("A2-Talboden") is correlated with the uppermost Pliocene, and
the E valley fill ("E-Talboden") corresponds to the lowermost
Pleistocene.
References
Ahrens, H., and Lotsch, D. 1976. Zum Problem des Pliozans in

Brandenburg. Jb. Geol. 7/8( 1971/1972):277-323.
Ahrens, H., von Biilow, W., Cepek, A. G., Erd, K., Jahnichen,
H., Krutzsch, W., Lotsch, D., Mai, D. H., and
Steinmiiller, A. 1968. Zur Plio/Pleistozan-Grenze in der
Deutschen Demokratischen Republik. In Reports of the
23rd Session, International Geological Congress, Czechoslovakia, 1968. Proceedings of Section 10, Tertiary!
Quarternary Boundary, pp. 65-77. Bratislava: Geologicky
Vied.
Berger, F. 1941. Kritisches zur Altersbestimmung des ostdeutschen Tertiars. Deutsch. geol. Ges., Zeitschr. 93:323356.
Bohme, G. 1963. Uber den Skelettfund eines Pliocerviden aus
dem Pliozan von Kaltensundheim/Rhon. Paldontol. Abh.
1:353-373.
Cepek, A. G. 1958. Zum Pliozan in Brandenburg. Geol. Ges.
DDR, Ber. 3:158-170.
Plio-Pleistocene of eastern Germany
Cepek, A. G. 1968. Quartar. In Grundriss der Geologie der

Deutschen Demokratischen Republik, pp. 305-420.
Dietrich, W. O. 1953. Neue Funde des etruskischen Nashorns in
Deutschland und die Frage der Villafranchium-Faunen.
Zeitschr. GeoL 2:417-430.
Duphorn, K., Grube, R, Meyer, K. D., Streif, H., and Vinken,
R. 1973. Pleistocene and Holocene (area of the Scandinavian glaciation). Eisz. Gegenw. 23/24:222-250.
Eissmann, L. 1975. Das Quartar der Leipziger Tieflandsbucht
und angrenzender Gebiete um Saale und Elbe. Modell
einer Landschaftsentwicklung am Rande der europaischen
Kontinentalvereisung. Schriftenr. GeoL Wiss. 2:1-263.
Eissmann, L. 1981. Periglaziare Prozesse und Permafroststrukturen aus sechs Kaltzeiten des Quartars. Altenburger
Naturwiss. Forsch. 1:1-171.
205
Krutzsch, W., and Majewski, J. 1965. Die mikrobotanische

Datierung des Tertiarvorkommens von Oberzella (Blatt
Vacha/Siidthur.). Geol. Inst. Berlin, Mitt. Zentral. 1:65-70.
Kube, W. 1979. Ein neuer Fund pflanzenfiihrender Tone aus
den Terrassenschottern bei Ottendorf-Okrilla. Veroff.
Kamentz: Museum Westlausitz, Veroff. 3:9-22.
Lotsch, D. 1981. TGL Standard Tertidrstratigraphie DDR, Nr.
25: 234/08. Berlin.
Mai, D. H. 1965. Eine pliozane Flora von Kranichfeld in
Thuringen. GeoL Inst. Berlin, Mitt. Zentral. 1:37-64.
Mai, D. H. 1967. Die Florenzonen, der Florenwechsel und die
Vorstellungen iiber den Klimaablauf im Jungtertiar der
Deutschen Demokratischen Republik. Zentral. Geol. Inst.
Berlin, Abh. 10:55-81.
Mai, D. H., Majewski, J., and Unger, K. P. 1963. Pliozan und
Altpleistozan von Rippersroda in Thuringen. Zeitschr.
Geol. 12:765-815.
Mai, D. H., and Walther, H. 1988. Die pliozanen Floren von
Eissmann, L., and Miiller, A. 1979. Leitlinien der Quartarentwicklung im Norddeutschen Tiefland. Zeitschr. GeoL
Wiss. 7:451-463.
Thuringen. Quartdrpaldontologie 7:55297.
Ellenberg, J. 1969. Die geologisch-morphologische Entwicklung
des sudwest-thuringischen Werragebietes im Pliozan und Mania, D. 1973. Palaookologie, Faunenentwicklung und StratiQuartar. Dissertation, Universitat Jena.
graphie des Eiszeitalters im mittleren Elbe-Saalegebiet auf
Gehl, O. 1958. Das Profil von Riiterberg, ein Beitrag zur PlioGrund von Molluskengesellschaften. Zeitschr. Geol. 21,
Pleistozan-Grenze in Mecklenburg. GeoL Ges. DDR, Ber.
Beiheft 78/79:1-175.
Mielecke, W. 1929. Beitrag zur Kenntnis der von Gerollen
3:153-154.
Genieser, K. 1955. Ehemalige Elbelaufe in der Lausitz. Zeitschr.
nordischer Herkunft freien Kiese in der Niederlausitz.
GeoL 4:223-279.
Zeitschr. Geschiebeforsch. 5:132-143.
Genieser, K., and Diener, I. 1958. Versuch einer Altersdeutung Mielecke, W. 1934. Uber Kieselsauregelite aus dem Pliozan der
der vor- und fruheiszeitlichen Elbelaufe auf Grund neuer
Niederlausitz. Zeitschr. Geschiebeforsch. 10:111-117.
Froschungsergebnisse. Univ. Berlin Wiss. Zeitschr. 6, Prager, F. 1976. Quartare Bildungen in Ostsachsen. Staatl.
Museum Mineral. GeoL Dresden, Abh. 25:125-217.
math.-nat. Reihe 5:477-487.
Hucke, K. 1928. Zur Verbreitung des Pliocans in Norddeutsch- Quitzow, H. 1953. Altersbeziehungen und Flozzusammenhange
in der jiingeren Braunkohlenformation nordlich der
land. Preuss. geol. Landesanst., Jb. 49:413-426.
Mittelgebirge. Geol. Jb. 68:27-132.
Jahnichen, H. 1968. Pflanzenfiihrende Tone aus dem Obermiozan von Ottendorf-Okrilla (Bezirk Dresden). Deutsch. Schramm, H. 1964. Untersuchungen zur Stratigraphie und
Akad. Wiss. Berlin, Monatsber. 10:919-920.
Palaogeographie altpleistozaner Schotterablagerungen im
Kahlke, H. D. 1968. Vertebratenstratigraphie zur Plio/Pleistozannordlichen Vorland des Thuringer Waldes. Dissertation,
Universitat Jena.
Grenze. In Reports of the 23rd Session. International
Geological Congress, Czechoslovakia, 1968. Proceedings ofSchubert, G. 1980. Ein synchroner Taschen- und Tropfenboden
Section 10, TertiaryI Quaternary Boundary, pp. 27-39.
in praelsterkaltzeitlichen Flusschottern ("Bautzener ElbeBratislava: Geologicky Vied.
lauf") der Lausitz. Zeitschr. Geol. Wiss. 8:1345-1348.
Kahlke, H. D. 1977. The N/Q-Boundary. Territories: Federal Siegel, R. 1959. Untersuchungen zur Talgeschichte der oberen
Werra zwischen Themar und Bad Salzungen. Dipl. Arbeit,
Republic of Germany, German Democratic Republic,
Universitat Jena.
People's Republic of Poland (terrestrial sequences).
Steinmuller, A. 1974. Tertiare. In Geologie von Thuringen, ed.
Giorn. GeoL, Ser. 2 41:165-177.
Kahlke, H. D., Eissmann, L., and Wiegank, F. 1984. Die Neogen/
W. Hoppe and G. Seidel, pp. 717-742. Leipzig.
Quartar-Grenze. Territorium der Deutschen Demokrati- Stolley, E. 1900. Geologische Mitteilungen von der Insel Sylt. II.
Cambrische und silurische Gerolle im Miozan. Arch.
schen Republik. Beitrag zum IGCP-Projekt 41: N/QAnthropol. Geol. Schleswig-Holsteins 4:1-49.
Grenze. Zeitschr. Angew. GeoL 30:44-48.
Kahlke, H. D., and Ukrainzeva, V. V. 1986. Pozdnepliocenovaya Unger, K. P. 1964. Die Gliederung der altpleistoanen Sedimente
flora, rastitel'nost i fauna yuga Tyuringii (Ikrug Zul',
und zur Frage der Plio/Pleistozan-Grenze in Thuringen. In
Vlth International Congress on the Quaternary, Warsaw
GDR). [Late Pliocene flora, environment, and fauna of
1961. Reports, vol. 1, pp. 349-356.
southern Thuringia [District Suhl, GDR]). Bot. Zhurn.
Unger, K. P. 1974. Quartar. In Geologie von Thuringen, ed. W.
71:16-22.
Hoppe and G. Seidel, pp. 742-781. Leipzig,
Krause, P. G. 1933. Das Pliozan Ostpreussens und seine
Beziehungen zum nordwestdeutschen und westdeutschen von Biilow, W. 1964. Der Tropfenboden von Ruterberg (SWMecklenburg). Zeitschr. Geol. 13:361-363.
Pliozan. Preuss. geol. Landesanst., Abh. n.f. 144:1-71.
Krutzsch, W. 1959. Sporen vom "Schizea-pusilla-Charakter" im Wiegank, F. 1982. Ergebnisse magnetostratigraphischer Untersuchungen im hoheren Kanozoikum der DDR. Zeitschr.
Pliozan von Riiterberg (= Wendisch-Wehningen). Arch.
GeoL Wiss. 10:737-744.
Naturfr. Mecklenburg 5:36-55.
Krutzsch, W. 1988. Kritische Bemerkungen zur Palynologie und Wolf, L. 1980. Die elster- und praelsterkaltzeitlichen Terrassen
der Elbe. Zeitschr. geol. Wiss. 8:1267-1280.
zur klimastratigraphischen Gliederung des Pliozans bis
tieferen Altpleistozans in Slid-, Siidwest-, Nordwest- und Ziegenhardt, W. 1965. Saxonischer Schollenbau und junge
Krustenbewegungen im nordlichen Vorland des Thuringer
pro parte Mitteleuropa sowie die Lage der Pliozan/
Waldes zwischen Ohra und Ilm. Dissertation, Hochschule
Pleistozan-Grenze in diesem Gebiet. Quartdrpaldontologie
Architektur u. Bauwesen, Universitat Weimar.
7:7-51.
19
The Plio-Pleistocene of Hungary

ANDREW A. RONAI
Introduction
Recent studies and investigations have agreed that paleomagnetic polarity changes are the most expedient criteria for
establishing the main divisions of Quaternary stratigraphy and
for tracing the "N/Q" limit between Neogene and Quaternary
(actually, the Pliocene-Pleistocene boundary). In Hungary,
there are two viewpoints as to where to mark the N/Q boundary:
at the Gauss-Matuyama paleomagnetic boundary (Ronai,
1984), dated to 2.5 Ma, or at the top of the Olduvai event (Pasini
and Colalongo, Chapter 2, this volume), dated to 1.8 Ma. Each
of these reversals is close to the initiation of a period of global
cooling.
If we accept the paleomagnetic polarity changes as signposts of
stratigraphic boundaries, it is still necessary to reach consensus
on which of the cooling events should be recognized as the lower
boundary of the Quaternary. Both reversals have been recognized in Hungary, but most Hungarian scientists have long
favored the view that the beginning of the Quaternary should be
determined by a major stratigraphic and paleontological change,
which we now know to have been almost coincident with the
Gauss-Matuyama paleomagnetic boundary - see Ronai (1984)
and the references cited therein.
There are several reasons to prefer this date, but the main one
is that it corresponds to the most impressive change in the
Carpathian Basin during the past 5-10 million years, namely, the
epirogenetic uplift of that great territory in the middle of the
European continent, and the regression of the Pannonian Lake,
comparable to regression in the Dacic Basin (Ghenea, Chapter
20, this volume). This event changed the geomorphology of the
entire region and seems also to have been coincident with the
starting point of a new tectonic cycle.
(Ronai, 1972; Franyo, 1977). The research was amplified to

include Quaternary tectonic movements, with the ultimate goal
of demonstrating the late Neogene and Quaternary geological
history of the Great Hungarian Plain.
The project was organized around the mapping of three depth
horizons (Franyo, 1977). The first horizon consisted of the
superficial and near-surface sediments, as revealed by a network
of shallow boreholes, up to 10 m in depth, drilled on a regular
1.5-km grid that covered most of the plain. Field data were
mapped at a scale of 1:50,000 and published in atlases at a scale
of 1:200,000, in which agrogeologic, hydrogeologic, and engineering geologic data and test results were also plotted. As many
as 18-20 geologic map sheets were prepared for each mapped
quadrangle.
The second depth horizon was studied in medium-depth (1001,500 m) boreholes in two linear arrays at right angles. The holes
were drilled to investigate the thickness and stratigraphy of the
Quaternary down to the Neogene-Quaternary boundary. In that
part of the project the Pliocene and Pleistocene aquifers were
identified and tested, and the core samples were analyzed for
paleotectonic information. The boreholes were then equipped
with artesian piezometers to monitor selected aquifers; three to
four wells were completed at each location in order to make
simultaneous observations of ground water at different depths.
Finally, the third depth horizon of the investigation consisted
of pre-Pliocene deposits involved in the floor of the modern
basin. At its inception in 1964, the project was scheduled to take
21 years; after systematic progress, the fieldwork and laboratory
analyses were completed by 1984, and the final atlases were in
press by 1985.
History of the Pannonian Basin
Geological investigations in the Great Hungarian Plain
Because of growing interest in problems of agrogeology, engineering geology, hydrogeology, and environmental protection, a
very detailed and thorough program of mapping the younger
deposits of the Pannonian Basin was started by the Quaternary
Department of the Hungarian Geological Institute in 1964
206
The geological evolution of the modern Pannonian Basin

(Figure 19.1) began in the middle Miocene, with contemporaneous uplift of the Carpathian mountain arc and subsidence of the
enclosed lowlands. The newly defined basin was occupied by an
arm of the Paratethys continental sea, connected to the protoMediterranean Tethys Ocean. In that constantly subsiding
regime, some 2,000-3,000 m of shallow-marine shelf sediments
Plio-Pleistocene of Hungary
tic '
>
^%|d)
m;
*
:#;
... ^
K^
SR
207
ill
MB
i
%
**'ywOj i
BL X -
p
|f \ \
''
1
i ^>
M^
*>
"<>
IIIIMI
'.I
100
200
300 km
Figure 19.1. The Hungarian Great Basin, with the area of the
Pannonian lake in Pliocene time indicated by the hatched area.
of the Lower Pannonian Stage were deposited, consisting of

mixed conglomerates, sandstones, marls, and clays.
Near the end of the Lower Pannonian time, the connection to
the Tethys became more restricted, and all of the Paratethys,
from central Asia to southern Germany, became brackishmarine. In the Pannonian Basin, isolation became complete, and
the Lower Pannonian sea was transformed to the Upper
Pannonian fresh-water lake. During that phase, some 1,0002,000 m of lacustrine sediments were deposited in the basin. In
the lower levels, those strata consist of rapidly alternating sand,
fine sand, silt, and clay; the higher levels are dominantly
variegated clays, representing the Levantine substage (Nikiforova, Chapter 21, this volume).
In economic terms, the lower part of the Upper Pannonian
sequence contains the known oil and gas reserves of the Great
Plain, as well as large volumes of saline, locally thermal water. In
the upper part of that same sequence, brown coal and lignite
deposits occur in the marginal facies near the edges of the plain.
At about 2.5 Ma, subsidence slowed and was overtaken by
deposition as the lake began tofillin from the margins, forming a

broad land surface traversed by a stream network flowing from
the surrounding highlands toward a central sink (Figure 19.1). In
different parts of that central area, some 600-700 m of fluviatile
sediments were deposited. Before that time, the Pannonian
Basin had subsided at approximately the same rates over very
large areas, as shown by the fact that the limnic layers are
horizontal and that their stratigraphy correlates in detail throughout major portions of the basin. The beginning of fluvial
sedimentation, on the other hand, was marked by a change in
tectonism and the formation of deep local depressions of smaller
extent. From the beginning those depressions were filled with
fluvial sediments, consisting of coarse sands and gravels in some
and fine-grained silts and clays in others. The lake bottom was
tectonically dismembered, and local areas subsided at varying
rates (Figure 19.2).
In detailed paleontological studies devoted to identification of
the Pliocene-Pleistocene boundary, a persistent problem was the
biotic sterility of thick sedimentary sequences in the upper levels
Andrew A. Ronai
208
yJJJUlti/LJLLLU
100-
500-
1000
Figure 19.2. Quaternary sedimentation in the Hungarian Great Basin. Limnic sediments of the
Lower, Middle and Upper Pliocene (PI1-PI3) and fluvial sediments of the lowest, lower, middle, and upper parts of the Pleistocene (Q!-Q4) are shown in N-S
and E-W sections. According to
paleomagnetic analysis, the "lowest Pleistocene" (Qt) is older than
the Vrica boundary.
50
Upper Pleistocene
PI3
Uppermost Pliocene
Middle Pleistocene
Pl2
Upper Pannonian (Pliocene)
Lower Pleistocene
Pl-j
L. Pannonian (Mio-Pliocene)
Lowermost Pleistocene
(Fluvial)
of the Upper Pannonian lake deposits, which prevented any

indication of climate changes. On the other hand, paleomagnetic
analyses of the core samples from two boreholes in the Koros
Basin represented a significant scientific achievement - see
Cooke (1981) and the Appendix to this chapter. The Koros Basin
is one of the deep local depressions where sedimentation was
essentially continuous throughout the Quaternary, so that a highresolution picture of paleomagnetic reversals can be traced back
(Limnic)
to 5 Ma. In those cores, the change from limnic sedimentation to

fluviatile was almost precisely synchronous with the GaussMatuyama reversal, falling at 2.4 Ma.
The technically determined shoaling of the Pannonian lake
appears to have been a period of distinct climate change as well.
The disappearance of the Pannonian lake was aided by an
extremely arid climate, as indicated by the sterile layers at the
top of the lacustrine sequence, in which no trace of vegetation or
animal life can be found, as contrasted with the rich mollusk and
ostracode faunas and the abundant pollen in the strata below and
above the sterile layers.
The paleontological evidence shows that at the beginning of
the fluviatile deposition, the climate became more humid,
although remaining quite warm; precipitation rates rose, and
vegetation again covered the terrain, becoming increasingly
opulent up to the level of the Olduvai subchronozone. At that
time, the climate began to alternate between cold and warm,
with strong fluctuations in precipitation, but none of those
changes had an effect on vegetation comparable to the transition
in the earliest Gauss (Ronai, 1970) (Figure 19.3). Paleontologically, it has not been possible to differentiate any major biotic
changes that would allow identification of internal, sub-epoch
boundaries of the Pleistocene, such as the four major glacial
intervals observed and demonstrated in the Alps. Only two
major biostratigraphic units can be distinguished in the postGauss interval, with the boundary marked by the transition to
more variable climates at the top of the Olduvai subchronozone.
The great change in geomorphology, tectonism, and climate at
the Gauss-Matuyama boundary supports the Hungarian opinion
that the disappearance of the Pannonian lake should mark the
end of the Pliocene and that the beginning of the fluvial
accumulation should mark the start of the Quaternary, coincident with the Gauss-Matuyama reversal. In this schema, the
climatic and faunal changes at the top of the Olduvai
subchronozone divide the older Pleistocene from the younger,
colder part, similar to a long-held opinion about the divisions of
the Quaternary in the former Soviet Union (Nikiforova, Chapter
21, this volume).
209
along the North African coastline, and in the upper Matuyama it

became more like that of Italy or Greece. For the Brunhes, we
find a cool, continental climate, which changed finally to a
climate like that of Scandinavia nowadays (Ronai, 1970; Kretzoi
and Pecsi, 1979). Overall, erosion was generally slow during the
warmer climates and faster during the cooler, so that the lower
part of the sequence has a higher proportion of silty and clayey
layers in the core profiles, and the sediments deposited during
the later period are more sandy and gravelly.
In the 800-k.y. interval of relatively warm and quite humid
climate in the first part of the Matuyama, we find seven major
climate cycles, according to variations in the indicators of
temperature and humidity. From the Olduvai to the beginning
of the Brunhes, six additional climate cycles can be distinguished according to pollen statistics, followed by four cycles in
the Brunhes. As for the latter 10 cycles, the average climate
was cool to cold; four cycles were relatively humid, and six were
arid.
The Koros Basin boreholes. Two of the boreholes in the Koros

Basin, Devavanya (1,116 m) and Veszto (1,200 m), are
noteworthy for their paleomagnetic data (Cooke, Hall, and
Ronai, 1980; Appendix, this chapter). The Koros Basin is the
second deepest Plio-Pleistocene depression in the Great Plain,
with Paleozoic basement at depths between 3,000 and 4,000 m
overlain by the full Lower Pannonian, Upper Pannonian, and
Quaternary sequence. The Quaternary (i.e., fluviatile) sequence, 200-500 m thick, consists mostly of silts and clays, with
occasional intercalations of thin sand beds. The paleomagnetic
stratigraphy in the Devavanya and Veszto boreholes indicates
that the Koros sedimentary series represents the whole fluviatile
Sedimentation cycles and climate change. The beginning of the Quaternary, without any appreciable hiatus (Cooke et al., 1980).
fluvial sedimentation at the Gauss-Matuyama boundary can be In these boreholes, the measurements indicated all of the known
detected in most boreholes on the Pannonian plain by a shift to polarity reversals of the international paleomagnetic scale, and
cyclical variations in granulometry (Ronai, 1984). Whereas the the distances between the paleomagnetic polarity changes
lake sediments below that boundary are characterized by rapid measured on the cores were closely proportionate to the dated
and sharply bounded oscillations between sandy and clayey time intervals between the polarity reversals.
layers, in the fluvial sequence one can detect sediment cycles in
Core samples were taken at 1-m intervals, with 20-cm-offset
which the granulometry changes gradually from coarse (gravel or backups in some sections. Samples were sent to the paleomagsand) to fine (silt or clay) and again gradually back to coarse. netic laboratory at Dalhousie University, Halifax (Cooke et al.,
Although cyclic sedimentation may have resulted from irregular, 1980). The results of the analyses, correlated to the analyses of
step-wise shifts in the rates of subsidence in the basin, other sedimentology, paleontology, and palynology in the cores, were
evidence from the magnetostratigraphic calibration (discussed initially presented at the INQUA Congress in Birmingham in
later) suggests that the granulometric variations reflect alterna- 1977, at the Paleomagnetic Symposium in Budapest-Szeged in
tions in precipitation and climate during the Quaternary that 1979 (Cooke et al., 1980), and at the International Field
influenced the quantity and coarseness of the river deposits Conference of the International Geological Correlation Program
(Figure 19.4).
(Projects 41 and 128) in Tucson in 1981.
At first, these data were used to designate major subdivisions
Palynological studies of the interval from the present back to
the Gauss-Matuyama boundary in the Great Hungarian Plain of the Quaternary, based on the historic concept that the
have provided remarkably detailed data showing 17 complete Quaternary was equivalent to the full sequence of fluviatile
cold/warm climatic cycles, or 35 fluctuations in total (Ronai, deposition in the Pannonian Basin (Ronai, 1968; Kretzoi and
1970, 1984). The succession of climates in the Pannonian Basin Pecsi, 1979). The beginning of that phase, and thus the beginning
demonstrates a steady shift from southern to northern character- of the Pleistocene, was dated at about 2.4 Ma in the old
istics (Ronai, 1972). In the lower Matuyama interval, the climate calibration (Preface, this volume) at the Gauss-Matuyama
of the Pannonian Basin was generally like the present climate boundary. The Olduvai normal-polarity event was used as an
210
Figure 19.3. PlioPleistocene stratigraphy in

three key boreholes in the
Hungarian Great Basin.
The clay fraction is shown
by the black profile, increasing to the right. Magnetostratigraphy in the
Devavanya and Veszto
boreholes is compared to
paleoclimatic data from the
Jaszladany borehole.
Andrew A. Ronai
1000-
1100-
1200-
Proportion of
the clay fraction
211
J A S Z L A D A N Y
CUMAIE
GlANUlOMfllY
50
Iclc T T U w
|D|H D|H|O H
sous
KX>-
UNIT S
Q4
: N
n
H3^
1 1
~
D>
CUMATIC
11
MOl
o3 - 3
o3 - 2
iJ4t
M
w
- .
:l
-JWI
o? _7
n i
O| - J
D
4
n n
n
D
T1
2 - 3
>
- 1
2 _
O,
o,
Q.
- f
Oi
'
- 3
- 2
... U - J
T - Ttmpi
index for dividing between the "Oldest" and the "Old" (or
Lower) Pleistocene, and the Matuyama-Brunhes paleomagnetic
reversal was used to divide the Lower from the Middle
Pleistocene (Ronai, 1984). Between the Middle and Upper
Pleistocene there are no paleomagnetic events to serve as
boundary markers. As noted earlier, this schema is inconsistent
with the currently adopted criterion of the Vrica bed e, which
locates the base of the Pleistocene to be equivalent to the coldclimate maximum at the end of the Olduvai event. As noted
earlier, this level has been the boundary between Oldest
Pleistocene and Old Pleistocene in the conventional Hungarian
usage.
Sedimentation rates determined from the paleomagnetic data
indicate that subsidence was almost steady at 0.2 mm/year during
the fluviatile interval in the Koros Basin, despite the changes in
sediment types. That evidence confirmed that the variations in
granulometry probably were almost completely independent of
tectonism and therefore should be considered as closely reflecting the changes in environmental conditions.
Figure 19.4. Summary of data in the upper part of the

Jaszladany borehole. The temperate-dry interval of cycle 3 is
approximately equivalent to the Olduvai subchron. (From
Cooke, 1981,fig.2, with permission.)
References
Cooke, H. B. S., Hall, J. I , and Ronai, A. 1980. Paleomagnetic,

sedimentary and climatic records from boreholes at
Devavanya and Veszto, Hungary. Ada Geol. Hung.
24:89-109. {Proceedings of conference and field workshop
on the stratigraphy of loess and alluvial deposits in
Hungary, ed. M. Pecsi.)
Franyo, F. 1977. Exploratory drillings on the Great Hungarian
Plain from 1968 to 1975. Budapest: Geogr. Rev. 1977:6071.
Kretzoi, M., and Pecsi, M. 1979. Pliocene and Pleistocene
development and chronology of the Pannonian Basin. Ada
Geol. Hung. 23:3-33.
Ronai, A. 1968. The Pliocene-Pleistocene boundary in the
Hungarian Basin. Ada Geol. Hung. 12:219-230.
Ronai, A. 1970. Lower and Middle Pleistocene flora in the
Carpathian Basin. Palaeogeogr. Palaeoclimatol. Palaeoecol. 8:265-285.
Ronai, A. 1972. Quaternary sedimentation and climatic history of
the Great Hungarian Basin. Budapest: Magyar Allami
Foldtani Intezet Evkonyve.
212
Andrew A. Ronai
D E V A V A N Y A
V E S Z T O
TIME
STAilt
SCALE
-to
-40
INCLINATION
>40
GANUlOMfTt
-to
so
my
STAilE
INC 11 NAT ION
Figure 19.5. Correlation of

fluvial cycles and magnetostratigraphy in the
Devavanya and Veszto
boreholes. (From Cooke,
1981, fig. 3, with
permission.)
Ronai, A. 1982. Magnetostratigraphy of Pliocene-Quaternary

sediments in the Great Hungarian Plain. Earth Evol. Sci.
3-4:165-267.
Ronai, A. 1984. The development of the Quaternary geology in
Hungary. Acta Geol. Hung. 28:75-90.
Appendix: Quaternary record from deep boreholes in
the Great Hungarian Plain*
H. BASIL S. COOKE
The Carpathian Basin has been subsiding continuously since the
mid-Tertiary, so that the Great Hungarian Plain is underlain by
several thousand meters of Quaternary, Pliocene, and Miocene
sediments. The Pliocene beds were laid down in a great
"Pannonian" lake and are overlain by 200-600 m of fluviatile
sediments, long regarded by Hungarian geologists as representing the Quaternary. Ten sedimentological cycles have been
recognized in these "Quaternary" deposits, and a borehole at
Jaszladany in 1964/65 yielded a good pollen record showing
evidence for at least 35 climatic fluctuations.
The earlier climate was warm, changing to temperate in the
middle "Quaternary" and to cold in the later "Quaternary," but
good dating criteria have been lacking. Two recent boreholes at
Devavanya and Veszto have provided a high-resolution paleomagnetic record in which magnetostratigraphic boundaries can
be drawn and used to provide time controls. The first two fluvial
cycles prove to be older than the Olduvai event and in terms of
present usage would fall in the Upper Pliocene. Both holes have
provided fair mollusk and ostracode faunas and moderate pollen
records that make possible broad correlations with Jaszladany.
The warm climate dominated until approximately 1.8 Ma, as
in the marine record. Further studies are needed before a
detailed evaluation of the climatic changes is possible, but the
prospects are excellent.
The Jaszladany borehole
In 1964/65 one of the early deep wells was drilled at Jaszladany,

in the northern part of the basin (Figure 19.3) and reached 950
m. It penetrated the entire "Quaternary" sequence, 420 m thick,
consisting of fine-grained sand, silt, and clay, completely devoid
of gravel or even coarse sand. The underlying Pannonian beds
comprised variegated clays to a depth of 730 m, below which
there were alternations of sand, silt, and clay. The sediments
were relatively rich in paleontological material, including mollusks, ostracodes, and even some vertebrates, as well as pollen
and spores, so that the locality has proved to be useful as a
* Adapted from: Cooke, H. B. S. 1981. Age control of Quaternary reference standard.
sedimentary/climatic record from deep boreholes in the Great Hungarian
In his analysis of the results, Ronai (1969) drew attention to
Plain. In Quaternary Paleoclimate, ed. W. C. Mahoney, pp. 1-12.
the
existence of ten major cycles of deposition within the
Norwich: Geo Abstracts Ltd. Thefigurenumbers and reference citation
"Quaternary" sequence. Each cycle begins with coarse sedistyle have been changed to conform to this work-Editor.
D E V A V A N Y A
V E S Z T O
CLIMATE
POLLEN
SAMPLES
C c T T WW
DH DH D H
CLIMATE
POLLEN
SAMPLES
CYCLES
C C T T ww
DH DHDH
FLUVIAL
HI
. . .
Hi
MM
m
100m-
"U
.
200-
200-
M
z
*-
OEP
ments; the material then becomes progressively finer, although

with fluctuations, until almost half of the sediment is fine clay.
Thereafter it coarsens gradually, again with fluctuations, until
the commencement of the next cycle. This is readily seen in
Figure 19.4, which shows the variations in grain size in the
Jaszladany core and the nomenclature of major cycles. There are
also some overall trends in the deposition which have led to
recognition of three lithological divisions: a Lower Phase
embracing cycles 1,2, and 3, a Middle Phase comprising cyles 4,
5, 6, and 7, and an Upper Phase comprising cycles 8, 9, and 10.
The boundaries between them are placed at 275 m and at 95 m.
These depositional cycles probably were related to tectonic
factors that had a general impact on the basin as a whole. The
delicate balance between subsidence and sediment supply had
led to occasional subaerial exposure and the formation of minor
soil horizons, shown in the second column of Figure 19.4, but
they do not represent more than localized pauses of short
duration in what is essentially an unbroken succession. There are
also three periods of peat formation displayed in this hole, and in
the vicinity of the 400-m level lignitic horizons occur. The
present-day sedimentation in the vicinity of Jaszladany is so
similar to that shown by the borehole that it is reasonable to
consider that the depositional environment has not changed
significantly. Geodetic observations have demonstrated that
subsidence of 5 mm occurred in a decade, and on that basis
Ronai (1969) estimated that deposition of the entire 420-m
fluvial succession occupied at least 1.3 to 1.4 m.y., a figure
compatible with assignment of the whole unit to the Quaternary.
213
m
300-
300-
. . .
400-
400-
Devavanya and Veszto boreholes

Figure 19.5 summarizes the sedimentological and paleomagnetic
data for the upper half of the Devavanya and Veszto boreholes
and attempts to correlate them with the "standard" paleomagnetic time scale. The boundary between the Brunhes normal and
Matuyama reverse epochs and that between the Matuyama and
the Gauss are readily located in the plot of stable inclinations,
and the Olduvai event is very clearly defined in both holes. The
Jaramillo event is also recognizable, although there is a
magnetically disturbed or oscillatory zone below it that is not
usually seen in the marine records or in the lava sequences.
Dotted lines have been drawn from these known age levels to the
appropriate depth positions in the respective cores, and they
serve to provide some means for estimating average sedimentation rates between them. Very fortunately, the paleomagnetic
record shows a good deal offinestructure, including a number of
short-period excursions not normally resolved. Several of these
can be matched in the two cores and provide additional
intermediate controls on the intercorrelation.
Interpretation
Turning now to the sedimentological record,fluvialcycles can be
distinguished in each of the holes, and the boundaries between
them can be transferred to the time scale by drawing links
parallel to the dotted lines of the paleomagnetic control points.
500-
500-
Poor
fa,,
C
T
W
0
H
"
-
Cold
Tatpro
Woim
O.y
Humid
- - -
-\
Ia
I
Figure 19.6. Paleoclimatic inferences from pollen samples in the Devavanya

and Veszto boreholes. (From Cooke, 1981,fig.4, with permission.)
The findings are in close accord for all the fluvial-cycle

boundaries except that between cycle 6 and cycle 7, where there
is a slight mismatch, albeit within the time span of the Jaramillo
event. But this means that it is possible to estimate with
reasonable accuracy the most probable ages of the boundaries
between thefluvialcycles.
If the present concept of the Neogene/Quaternary boundary as
lying just above the Olduvai event is confirmed (Haq, Berggren,
and Van Couvering, 1977), then fluvial cycles 1 and 2 will belong
to the Upper Pliocene, rather than to the Quaternary, as has
been assumed hitherto.
Unfortunately, both Devavanya and Veszto have furnished
rather scanty pollen records, with counts in the order of tens of
grains per gram of sediment, and only a few with hundreds of
grains per gram. Nevertheless, it is possible to make reasonable
inferences about the climates in parts of the cores, and these are
shown in Figure 19.6. Fluvial cycles 1 to 3 show generally warm
Andrew A. Ronai
214
PALEOMAGNETIC
TIME SCALE
CLIMATE
Figure 19.7. Depth/age plots

for fluvial cycles in
Jaszladany, Devavanya, and
Veszto boreholes, and
paleoclimatic inferences derived from pollen analysis.
(From Cooke, 1981, fig. 5,
with permission.)
conditions, cycles 4 and 5 are temperate, and cycles 7, 8, 9, and

10 indicate a cool environment; data for the general character of
the climatic changes leave little doubt that they parallel those of
the Jaszladany core and strengthen the belief that the fluvial
cycles can be matched over a large part of the basin.
Figure 19.7 has been drawn on the assumption that the fluvial
cycles match and that the age boundaries are reasonable
approximations. For each of the three boreholes under consideration, the depths to the fluvial-cycle boundaries have been
plotted against the estimated ages derived from the paleomagnetic data. The resultant curves show the apparent relative
changes in sedimentation rates for each cycle and are in good
agreement. It would appear that there was an above-average rate
of sedimentation at Jaszladany in cycle 4 and a below-average
rate in cycle 1. There are also rather sharp changes in the
sedimentation rates in all three holes below the b/a-cycle
boundary [= Gauss-Matuyama boundary-editor], also noted in
the earlier interpretation of the whole paleomagnetic record
(Cooke et al., 1980). It must be pointed out that the relative
linearity of the curves for Devavanya and Veszto may in part be
an artifact of the procedure by which the cycle boundaries were
derived by interpolation from the paleomagnetic control points;
however, the boundary values would change very little if an
attempt were made to "straighten" the Jaszladany curve. It is
also the case that soil horizons are much less apparent in the
Devavanya and Veszto cores, so that the Jaszladany sedimentation rate may indeed have been a little more variable. It is also
clearly impossible, without more control points, to take any
account of the variations in actual sedimentation rates that

probably accompanied the changes in grain size within the
cycles. Accordingly, it is considered that the age estimates for the
boundaries of the fluvial cycles are not likely to be seriously in
error.
On the right side of Figure 19.7, the climatic inferences for
each of the three boreholes have been located in relation to their
positions in the fluvial cycles. While they do not match in detail,
they confirm Ronai's (1969) limits for the three major climatic
divisions. Thus the warm climate dominated until approximately
1.5 Ma [= 1.7 Ma in the orbitally tuned time scale-editor], and
the onset of consistently cool conditions began at about 0.9 Ma
[= 1.0 Ma], just above the time of the Jaramillo event, although
there is evidence for a cold spell at about 1.25 Ma [= 1.33 Ma].
This is in remarkably good agreement with the broad aspects of
faunal and isotopic changes in the oceans, and it is worth quoting
from the abstract to an important paper by Ruddiman (1971) on
cores from the equatorial Atlantic:
When applied to two cores containing 1.8 m.y. of
equatorial sedimentary history, this analysis pinpoints
two prominent, large-scale climatic shifts: (1) at 1.3 Ma
BP, the mean climatic situation deteriorated, and short
but severe cold pulses began to punctuate the previous
moderate warmth of the late Matuyama; (2) following
900,000 Ka, the duration of cold intervals increased.
Prior to the Jaramillo, no cold pulse exceeded 30,000
yrs; three post-Jaramillo cold intervals ranged in
215
Fink, I , and Kukla, G. J. 1977. Pleistocene climates in Central

Europe: at least 17 interglacials after the Olduvai Event.
Quat. Res. 7:363-371.
Haq, B., Berggren, W. A., and Van Couvering, J. A. 1977.
Corrected age of the Pliocene/Pleistocene boundary.
Indeed, it is tempting to try to integrate the climatic inferences
Nature 269:483-488.
from the three Hungarian boreholes and then to compare them Kukla, G. J. 1975. Loess stratigraphy of Central Europe. In:
with Ruddiman's core, when a plausible correlation can be
After the Australopithecines, ed. K. W. Butzer and G. LI.
Isaac, pp. 99-188. The Hague: Mouton Publishers.
made. However, in point of fact the Hungarian observations are
on too coarse a scale for such a comparison to be justifiable at Mankinen, E. A., and Dalrymple, G. B. 1979. Revised
geomagnetic polarity time scale for the interval 0-5 Ma
the present time. It is clear that if a new core could be obtained
BP. /. Geophys. Res. 84:615-626.
from the pollen-rich Jaszladany area, a closely spaced sampling McDougall, I. 1977. The Earth: its origin, structure and
program might yield a record as important as the remarkable one
evolution. London: Academic Press.
from the Grande Pile peat bog in France (Woillard, 1975, 1978). Ronai, A. 1969. Eine vollstandige Folge quartare Sedimente in
Ungarn. Eisz. Gegenw. 20:5-34.
It has also been demonstrated very clearly that the traditional
"four glacial" concept of the European Pleistocene is a gross Ronai, A. 1970. Lower and Middle Pleistocene flora in the
Carpathian Basin. Palaeogeogr. Palaeoclimatol. Palaeooversimplification, and the story told by the loess stratigraphy of
ecol. 8:265-285.
Central Europe is much more complicated, with 17 interglacial Ronai, A. 1972. Quaternary sedimentation and climatic history of
the Great Hungarian Basin. Budapest: Magyar Allami
cycles since the Olduvai event (Kukla, 1975; Fink and Kukla,
Foldtani Intezet Evkonyve.
1977). Doubtlessly, closer study of the Hungarian profiles will
Ruddiman, W. F. 1971. Pleistocene sedimentation in the
eventually resolve some of these problems.
equatorial Atlantic: stratigraphy and faunal paleoclimatology. Geol. Soc. Am., Bull. 82:359-362.
Steiger, R. H., and Jaeger, E. 1977. Subcommission on
References
geochronology: convention on the use of decay constants
in gas- and cosmochronology. Earth Planet. Sci. Lett.
Cooke, H. B. S., Hall, J. J., and Ronai, A. 1980. Paleomagnetic,
36:359-362.
sedimentary and climatic records from boreholes at Woillard, G. M. 1975. Recherches palynologiques sur le PleistoDevavanya and Veszto, Hungary. Ada Geol. Hung.
cene dans l'Est de la Belgique et dans les Vosges
24:89-109. (Proceedings of conference andfieldworkshop
Lorraines. Acta Geogr. Lovainiensis 14:1-168.
on the stratigraphy of loess and alluvial deposits in Woillard, G. M. 1978. Grande Pile peat bog: a continuous pollen
Hungary, ed. M. Pecsi.)
record for the last 140,000 years. Quat. Res. 9:1-21.
duration from about 50,000 to 150,000 yrs. The shortest
and most recent of these correlates with the Wisconsin
glaciation.
20
The Pliocene-Pleistocene boundary in Romania

CONSTANTIN GHENEA
Introduction
In recent years, detailed geological studies of the PlioceneLower Pleistocene interval in Romania, including modern
biostratigraphic analysis, have been combined with paleomagnetic data. At present, the chronology of the interval between
3.8 and 1.0 Ma in the Dacic Basin is well known, correlations
being possible not only with the Euxinic Basin (Ukraine) but also
with western Europe.
The Dacic (or Dacian) Basin is the name given to a vast
sedimentary basin bounded by the southern Carpathians, the
Balkans, and the Danube (Figure 20.1). In the middle Pliocene,
about 3.8 Ma, the salinity of the waters filling that basin was
reduced, a trend that became more pronounced in the late
Pliocene. Strata with fresh-water fauna were deposited, and at
certain levels beds rich in fossil mammal faunas were deposited.
This is the Romanian Stage, and data obtained in the past few
years are basic to correct interpretation of the stratigraphy of this
interval.
In most of the Dacic Basin, deposition of continental molasse,
known as the Cindesti Formation, began in the late Pliocene and
continued into the early Pleistocene. This formation occurs
throughout the basin, except for the western part, where
lacustrine conditions prevailed. Recently, the stratigraphic limits
of the Cindesti deposits were correlated to the paleomagnetic
scale, but no stratigraphic subdivisions have been identified
within this complex.
Above the Cindesti Formation, in the central and southern
parts of the Dacic Basin, deposits of a mixed, fluviolacustrine
regime developed in the early and middle Pleistocene; paleontological evidence at various levels permits their correlation with
the formations of the Euxinic Basin (Table 20.1).
Chronostratigraphy of the 3.8-1.0-Ma interval
The major part of the Plio-Pleistocene interval is occupied by

the Romanian Stage, from about 3.8 Ma to 1.8 Ma. This term
was first designated by Krejci Graf and redefined by Mihaila and
Andreescu (Andreescu, 1981) for the period of freshening water
in the Dacic Basin. The brackish-water mollusk faunas disappeared and were replaced by fresh-water lacustrine faunas.
216
The following biostragraphy is based on fresh-water mollusk

faunas studied by Andreescu (1981). These have been correlated
with paleomagnetic data by Ghenea et al. (1982; Alekseeva et
al., 1981), with the paleomagnetic boundary ages revised in this
chapter according to the orbitally tuned calibration of Cande and
Kent (1992).
Romanian Stage
Lower Romanian. In the Carpathian Bend, near Beceni, an
alternation of sands, clays, and marls with lignite intercalations
about 250 m thick conformably overlies the deposits of the
Upper Dacian Stage (Andreescu, 1981). The fossil content is
mainly smooth unionids of the slanicensis, sturdzae, and
brandzae groups, as well as Viviparus bifarcinatus. The Lower
Romanian has the same lithologic and paleontological features in
the central part of the Dacic Basin and in its western extremity,
where lignitic strata are also present.
On the Moldavian Plateau, afluviolacustrinesequence equivalent to the Lower Romanian contains a mammalian fauna
characterized by the following taxa: mastodons Anancus
arvernensis and Zygolophodon (= Mammut) borsoni; the
earliest arvicolid, Promimomys moldavicus; horses of the genus
Equus together with the genus Hipparion; and Tapirus
arvernensis and Sus provincialis. A mollusk fauna with Psilunio
sibinensis occurs in the same formations.
In the Brasov intermontane basin, a lacustrine sequence
consisting of marls, clays, and sands, with lignite intercalations,
is also considered to be equivalent to the Lower Romanian. At
Capeni and Virghis, local mammal faunas include Anancus
arvernensis, Zygolophodon (= Mammut) borsoni, Dicerorhinus
cf. D. leptorhinus, Equus (Macrohippus) sylvanum, Sus provincialis, Protarctos boeckhi, Parailurus anglicus, Prospalax priscus,
Mesopithecus monspessulanus, and others (Samson and Radulescu, 1973).
The mollusk and mammal faunas found in the Lower
Romanian can be correlated with the lower part of the
Moldavian complex in the former Soviet Union. This is in
agreement with recent paleomagnetic analyses, which have
Plio-Pleistocene of Romania
L G
Figure 20.1. Localities mentioned in the text.
established that the formations of the Lower Romanian were

deposited between 3.8 Ma and 3.4 Ma (Table 20.1).
217
tions in the Liquidambar pollen and the Myricaceae-Cyrillaceae association. The second, between strata VII and VIII, is
marked by the maximum development of the MyricaceaeCyrillaceae association. The palynomorph suite in the upper
part of the Romanian is characterized by the dominance of
conifers such as Picea, Cedrus, Pinus, Tsuga, Abies, and Larix,
indicating cold climate (Roman, in Marinescu, Ghenea, and
Papaianopol, 1981), or, more probably, higher elevations in the
source areas from which the sediments were derived, as the
geological evidence suggests.
In the Middle Romanian, fossil mammal remains have been
found at several locations in the Dacic Basin, including the
deposits of Cernatesti and Tulucesti. In addition to the two
mastodons Anancus arvernensis and Zygolophodon borsoni, the
faunas include some typical middle Villafranchian species, such
as Dicerorhinus etruscus, Equus stenonis, Cervus issiodorensis,
and the earliest elephant, "Archidiskodon rumanus" which,
properly speaking, is a species of Mammuthus (Aguirre et al.,
Chapter 9, this volume). In the Brasov intermontane basin,
remains of Dicerorhinus jeanvireti and D. etruscus were mentioned by Samson and Radulescu (1973), within the "Iaras Sand
Formation," which is assigned to the Middle Romanian.
Middle Romanian. The Middle Romanian is marked by a major Upper Romanian. In the western part of the Dacic Basin,
change in the fresh-water mollusks: the appearance and further lacustrine conditions are still represented in the Upper Romadevelopment of the Levantine thermophilous fauna (Tshepalyga, nian strata, consisting of a clayey-sandy sequence with a fresh1972), characterized by numerous genera and subgenera of water mollusk fauna. The latter is marked by the disappearance
sculptured unionids. At Craiova in the western Dacic Basin, of most of the typical "Levantine" genera and species of the
sands and clays with sculptured unionids form the stratotype of Middle Romanian and by the appearance of new genera, such as
the Levantine Stage. Current knowledge allows us to consider Bogatschevia and new species such as Ebersininaia milcovensis,
that this stage, as previously used in the relevant Romanian E. geometrica, E. struevi, Unio kujalnicensis, and Rugunio
riphai.
literature, is the equivalent of the entire Middle Romanian.
In the Olt Valley, at Slatina, the sequence of clays, clayey
The typical Middle Romanian in the western part of the Dacic
Basin, in the Jiu Valley, contains an extremely varied and rich sands, and gravels representing the Middle and Upper Romamollusk fauna consisting of Rugunio lenticularis, Potamides nian is highly fossiliferous. Paleontological evidence (mollusks,
porumbarui, P. herjeui, Cuneopsidea recurvus, C. vukotinovici, micromammals) and paleomagnetic correlations provide a deC. sculpta, C. iconominanus, C. beyrichi, C. doljensis, Rytia tailed stratigraphy across the Pliocene-Pleistocene boundary
bielzii, R. brandzae, R. slavonica, Wenziella clivosa, W. sub- (Andreescu et al., 1981). The Upper Romanian is represented
clivosa, W. cymatoidea, W. gorjensis, Rugunio condai, R. by strata with Ebersininaia milcovensis, E. geometrica, and
turburensis, R. mojsvari, R. pilari, Pristinunio pristinus, P. Rugunio riphai. The boundary between the strata with Rugunio
davilai, Cyclopotomida munieri, C. pannonica, Psilunio craioven-riphai and those with Unio apscheronicus seems to correspond to
the base of the Olduvai subchronozone.
sis, P. altercarinatus, Ebersininaia stefanescui, Viviparus bifarcinatus, V. stricturatus, V. rudis, V. turgidus, V. strossmayerianus, V. The Upper Romanian strata of the Olt Valley also contain a
craiovensis, and V. mammatus, among others. The biozonation rich fauna of fossil mammals. The large mammals include an
of the Middle Romanian is based on the unionid fauna archaic elephant, which, according to Radulescu and Samson (in
(Andreescu, 1981), and the definition of zones and subzones Andreescu et al., 1981), is similar to Mammuthus (= "Arpermits good correlations with the equivalent formations in the chidiskodon") gromovi and is characteristic of the Kotlovina
Ukraine (Tshepalyga, 1972).
level in the former Soviet Union.
Small mammals are present at several levels in the Olt Valley,
In the western extremity of the Dacic Basin (the Jiu-Danube
interfluve), deposition of coal began with the Dacian Stage of at Slatina and Cherlesti (Feru, Radulescu, and Samson, 1978).
the middle Pliocene, continued through the whole Romanian, The lower level (Slatina 1) is characterized by Desmana kormosi,
and ended in the early Pleistocene. Sporopollen sequences Apodemus sp. 1, Dolomys milled subsp. 1, and Mimomys minor
within the coal sequence show two characteristic reference subsp. 1; D. milleri dominates this association. The upper level
horizons. The first, at the level of stratum VI (the Dacian- at Slatina (Slatina 2) contains Desmana nehringi, Talpa fossilis,
Lower Romanian boundary), is marked by substantial reduc- Beremendia fossidens, Leporidae cf. Hypolagus brachygnathus,
Constantin Ghenea
218
Table 20.1. Upper Pliocene and Lower Pleistocene chronostratigraphy in Romania

U. S . S . R. jWEUROft
D
CHR0N0- LfTHOSTRA
STRAT (GRATCRAPHC
PHIC
SUBOIVI
UNITS
SIONS
1
B
C
I
B A S I N
0
M O L L U S C S
CRONOSTRATIGRAPHIC SUBOISIONS
M A M M A L S
M l C R 0 2B I 0
CASPIAN
BASIN
Corbicula fluminalis
0.7
COMPLEX
1.0
JARAMILLO
1,2
Tragontherium
boisvilletti
boisvilleiii
Bogatschevia sturi
z
o
BOSERNITIAN
Archidiskodon
meridionalis
1,6
meridionalis
Umo apscheronicus
Tragontherium
boisviltetti
dacicum
OOMASKINIAN
QL
<
1,8
Rugunio riphaei
2.0
Unio'kjialnicensis Archidiskodon
feunion
2.2
Ebersininaia
2.8
3.4
3.6
3.8
AKKULAEVIAN
Ebersinmaia
milcovensis
2.6
3,2
gromovi
geometrica
2.U
3,0
Mimomys polonicus
FERLADANIAN
M. pliocaenicus
KRIJANOVSKIAN
KAEHA O
Cuneopsidea
iconomianus
Archidiskodon
rumanus
Rytia bielzi
Anancus
arvernensis
Pnsfinunio pristinus
Zygolophodon
borsoni
Rytia brandzae
Vi v iparus
bifarci n a t us
s m o o t h unionids
Anancus
arvernensis
Zygolophodon
borsoni
Tapirus
arvernensis
Mimomys
G I gr. mi nor
CISTOPOLIAN
Pliomys
hungaricus
VESELOVIAN
UPPER
PORATIAN
<
o
o
KAGULIAN
Promimomys
1
2
KUCHURGANIAN
For Romanian stage-Andreescu 1981

Samson a n d R a d u l e s c u 1981
Apodemus sp., Dolomys milleri milleri, and Mimomys minor deposition. The stratigraphic position of the Cindesti Formation
subsp. 2. The reduction of D. milleri and the prevalence of is difficult to establish because paleontological evidence is
extremely scarce (mainly, isolated remains of Mammuthus
Mimomys are characteristic.
At Cherlesti, 13 km north of Slatina, the Upper Pliocene meridionalis). Nevertheless, the stratigraphic limits have been
(between 2.0 and 1.8 Ma) includes a faunal association of estimated through paleomagnetic profiles in sections where
Desmana nehringi, Allactaga ucrainica, Dolomys milleri milleri,sedimentation was continuous during the Romanian-early PleisMimomys ex gr. M. polonicuspliocaenicus, and Mimomys tocene interval. In the Carpathian Bend zone, coarse-grained
minor subsp. 2 (Radulescu and Samson, in Andreescu et al, intercalations have been assigned to the "Cindesti Formation" as
far back as the interval between 2.7 Ma and 2.5 Ma (Alekseeva
1981).
etal.,1981).
Cindesti Formation. At the beginning of late Romanian time, the Stratigraphic relations in the Carpathian region indicate that
continental molasse known as the Cindesti Formation, made up the Cindesti type of deposition continued up into the early
of alternating gravels, conglomerates, sands, and clays, devel- Pleistocene over a wide area. In that respect, the previous
oped over a vast area in the Dacic Basin. In many places, its location of the Pliocene-Pleistocene boundary within the
thickness reaches hundreds of meters, with a maximum of 1,000 Cindesti Formation has no important geological or climatic
m in the Carpathian Bend zone.
significance. The boundary near the top of the Olduvai event, at
The extent and thickness of the Cindesti Formation in the about 1.80 Ma (Pasini and Colalongo, Chapter 2, this volume),
Carpathian foreland are the results of massive uplift of the however, seems to have been associated with evidence of
Carpathians and the resulting intensification of erosion and important changes.
Plio-Pleistocene of Romania
Correlation of Middle Romanian and Upper Romanian
219
Azzaroli et al., Chapter 11, this volume; Nikiforova, Chapter 21,

this volume). A younger level, with a cold-climate fauna
indicated by Canis etruscus, would be equivalent to the Olivola
phase in the transition to Upper Villafranchian (Azzaroli et al.,
In some areas of the southwestern Dacic Basin, the upper part
of the "Fratesti Formation" consists of sandy deposits with claymarly interbeds, the so-called Uzunu Beds, showing a mollusk
fauna including Bogatschevia sturii, Corbicula fluminalis,
Pisidium amnicum, Viviparus craiovensis, V. diluvianus, V.
geticus, V. romaloi, Litoglyphus naticoides, Planorbis planorbis,
and others. The presence of the fauna with Bogatschevia sturii
allows the correlation of the Uzunu Beds to the Boshernitskan
horizon of the Pleistocene in the former Soviet Union, dated to
about 1.3-1.2 Ma.
The biostratigraphic divisions of the Romanian based on the

fossil mollusk and mammal faunas and their calibration with the
paleomagnetic scale now in use allow detailed correlations with
the Upper Pliocene and the Lower Pleistocene of the Euxinic
and Caspian basins. Thus, the Lower Romanian-Middle Romanian boundary is almost the same age as the division between the
Gilbert and Gauss magnetic epochs, at about 3.6 Ma, providing a
criterion that can be used for the Dacic Basin. The Middle
Romanian, which is correlated by its mollusk faunas with the
Kagulian, upper Poratian, and Veselovian horizons in the former
Soviet Union, also corresponds in its mammal fauna to the
Middle Villafranchian (zone MN-16) of the Mediterranean
basin. The Upper Romanian corresponds to the upper part of
the Akchagylian and to the Chistopolian, Akkulaevian,
Krizhanovskan, and Ferlandanian horizons in the Ukraine; its
Problems of the Pliocene-Pleistocene boundary in
base is considered to be at about 2.6 Ma, near the GaussRomania
Matuyama boundary. The upper boundary of the Romanian,
represented by the boundary between the strata with Rugunio
riphai and the overlying strata with Unio apscheronicus, coin- In the past, the main problem in establishing the Pliocenecides with the base of the Olduvai event of the Matuyama epoch. Pleistocene boundary in the Dacic Basin of Romania concerned
disagreements about correlating Villafranchian mammal faunas
to the Calabrian marine deposits, considered to represent the
Pleistocene base. With respect to subdivisions of the VillafranchRegional correlation of the Romanian Lower
ian,
which are well represented in Romania, the PliocenePleistocene
Pleistocene boundary has been recognized variously at about 3.8
In the western part of the Dacic Basin, the lacustrine conditions, Ma, with the appearance of the first Villafranchian elements, at
evidenced by strata with unionids and fossil mammals, seem to about 2.6 Ma, with the appearance of the earliest Middle
have ended at the Unio apscheronicus level, close to 1.9 Ma. In Villafranchian faunas, or at about 1.8 Ma, with the appearance
the central and southern parts of the Dacic Basin, as already of mammal assemblages comparable with the Seneze and Olivola
stated, that is equivalent to the lower alluvial deposits of the faunal phases, termed "upper Middle Villafranchian" by
Cindesti Formation, which continue into the Pleistocene. In the Azzaroli et al. (Chapter 11, this volume), but historically
eastern part of the Dacic Basin, drillings in the Bucuresti area considered as Upper Villafranchian in the Romanian literature.
Applying the recommendations of the working group for
have shown the presence, under younger formations, of 100-200
m of lacustrine beds similar to those of the western Dacic Basin, IGCP-41 to Romania, the acceptability of these various proposat depths ranging between 100 and 300 m. Those accumulations als can be summed up as follows:
were given the name of "Fratesti Beds" by Liteanu in 1953
(Marinescu et al., 1981). Regarding the paleontological content,
1. The first Lower Villafranchian fauna (upper Ruscinianthe Fratesti interval is characterized by rich mammal faunas
Csarnotian transition) is slightly older than the Gilbertwhich allow correlations to the European scale. The best-known
Gauss boundary. That paleomagnetic reversal, at 3.6
deposit is situated at Tetoiu, better known in the literature as
Ma, corresponds to the boundary between the Lower
Bugiulesti. There, the mammal association is marked by the
and Middle Romanian and was associated with an
presence of a single elephant, Mammuthus meridionalis, toimportant change in the fresh-water mollusk fauna to
gether with Paradolichopithecus geticus, Equus (Allohippus)
include a multitude of sculptured unionids representing
stenonis, Dicerorhinus etruscus, Ursus etruscus, Nyctereutes,
a markedly thermophilous fauna, and obviously demonTrogontherium boisvilleti, Megalovis, Leptobos, Libralces galstrating a climatic optimum. It is worth noting that the
licus, and Canis etruscus (Samson and Radulescu, 1973).
mammal faunas at Beresti-Malusteni and CapeniVirghis, in which the appearances of the "VillafranchThis ensemble might seem to be derived from two levels. An
ian" elements are considered to mark an important
earlier warm-climate part, with the last Nyctereutes, could be
change, are situated within the Lower Romanian, below
correlated to the older part of the Odessan complex of the
the Gilbert-Gauss boundary.
Euxino-Caspian realm and with the younger part of the middle
Villafranchian, for example the Seneze fauna and its correlatives
2. The first Middle Villafranchian fauna is approximately
in western Europe, which under current definition are placed in
coeval with the Gauss-Matuyama boundary, at 2.6
the latest Pliocene (Aguirre et al., Chapter 9, this volume;
Ma, and thus with the boundary between the Middle
220
Constantin Ghenea
Romanian and Upper Romanian. It was marked by

cold climate throughout the Paratethys, manifested by
the disappearance of the thermophilous mollusk fauna
with sculptured unionids. Within the interval between
2.7 Ma and 2.5 Ma, important orogenic movements in
the Carpathians brought about the deposition of the
Cindesti continental molasse over a vast area of the
Carpathian foreland. It may also be supposed that the
intensification of erosion and the deposition of large
amounts of materials indicate a great increase in
runoff as a result of higher precipitation in the cold
climate.
3. Although originally considered to be older, the first
cold-climate mammal faunas of the Upper Villafranchian (as this term is understood in Romania) and the top
of the Upper Romanian are now placed at the top of
the Olduvai subchronozone, approximately 1.77 Ma.
At present that is the official criterion adopted in
Romania for the base of the Pleistocene, although
recent studies suggest that the Pliocene-Pleistocene
boundary-stratotype at Vrica, Italy, is in fact slightly
older, with an age of about 1.80 Ma within the upper
Olduvai (Pasini and Colalongo, Chapter 2, this volume). The concept appears to be the most conventional, since it is not based on a climatic or
biostratigraphic criterion, but on correlation to a fixed
point, the Vrica stratotype.
In most of the Dacic Basin, the designated Upper RomanianPleistocene boundary falls within the typically undifferentiated
and generally unfossiliferous conglomerates and sands of the
Cindesti Formation, with the result that in those deposits
geological mapping of the boundary is an insoluble problem. On
the other hand, in western Romania, and again in the "Fratesti
Beds" of the Bucuresti area, biostratigraphic sequences of Upper
Villafranchian mammals and paleomagnetic data allow correlation to the presently defined Pliocene-Pleistocene boundary.
References
Alekseeva, L. I., Andreescu, I., Bandrabur, T., Cepaliga, A.,

Ghenea, C , Mihaila, N., and Trubihin, V. 1981. Correlations on the Pliocene and Lower Pleistocene Deposits of the
Dacic and Euxinic Basins. In 12th Congress of the CarpathoBalkan Association, Bucharest, Romania: Abstracts, pp.
29-30. Bucuresti: Institul Geologia i Geofisica.
Andreescu, I. 1981. Biochronologie et Chronostratigraphie du
Pliocene superieur et du Pleistocene inferieur du Bassin
Dacique. In 12th Congress of the Carpatho-Balkan Association, Bucharest, Romania: Abstracts, pp. 30-31. Bucuresti: Institul Geologia i Geofisica.
Andreescu, I., Radulescu, C , Samson, P., Cepaliga, A., and
Trubihin, V. 1981. Chronologie des formations PlioPleistocenes de la zone de Slatina (Bassin Dacique),
Roumanie. Inst. Speleol. E. Racovitza, Trav. 20:101-128.
Cande, S. C , and Kent, D. V. 1992. A new geomagnetic polarity
time scale for the Late Cretaceous and Cenozoic. /.
Geophys. Res. 97:13917-13951.
Feru, M., Radulescu, C , and Samson, P. 1978. Biostratigraphie
(micro-mammiferes) des depots Plio-Pleistocenes de la
zone de Slatina. Inst. Speleol. E. Racovitza, Trav. 17:1229.
Ghenea, C , Andreescu, I., Bandrabur, T., Cepaliga, A.,
Mihaila, N., and Trubihin, V. 1982. Bio and magnetostratigraphic correlations on the Pliocene and Lower
Pleistocene Formations of the Dacic Basin and the Brasov
Depression (Romania). Bucuresti: St. Techn. E c , Comm.
Geol.,E.2.
Marinescu, E, Ghenea, C , and Papaianopol, I. 1981. Stratigraphy of the Neogene and the Pleistocene Boundary. In
12th Congress of the Carpatho-Balkan Association, Bucharest, Romania, Guidebook 20, pp. 1-110. Bucuresti:
Institul Geologia i Geofisica.
Samson, P., and Radulescu, C. 1973. Les Faunes des
mammiferes et la limite Pliocene-Pleistocene en Roumanie. Inst. Speleol. E. Racovitza, Trav. 12:191-228.
Tshepalyga, A. L. 1972. Neogene-Quaternary boundary according to the data of terrestrial molluscs (in Russian). In
International colloquium on the problem "The Boundary
between the Neogene and Quaternary", ed. M. N. Alekseev
et al., pp. 42-57. Moscow: Acad. Nauk SSSR.
21
The Pliocene and Pleistocene of the European part of the

Commonwealth of Independent States
Introduction
Within western Russia and the adjoining states, the Upper

Pliocene and Quaternary (Anthropogene) sequences are abundantly documented in terms of fossil mammals, marine and
fresh-water mollusks, foraminifera, ostracodes, macro- and
microflora, and remains of human occupation, as well as
extensive data on neotectonics, lithology, and paleomagnetism.
For many years, however, the stratigraphers in that vast region
did not share a unified point of view on the location of the lower
boundary of the Quaternary. Although the majority considered
that the lower boundary of the Quaternary should be correlated
to the base of the Bakuan Stage, some favored the base of the
Akchagylian Stage, whereas others preferred the base of the
Apsheronian.
At a joint meeting of the IGCP-41 working group and INQUA
Subcommission 1-d at the XI INQUA Congress in Moscow in
1984, the proposal was adopted to place the PliocenePleistocene boundary, and thus the Neogene-Quaternary (N/Q)
boundary, in a physical reference point, or boundary-stratotype,
at Vrica, Calabria, located at the base of the claystone layer
conformably overlying sapropelic marker e in component-section
B of Selli et al. (1977) (Aguirre and Pasini, 1985). According to
Tauxe et al. (1983), with slight modification by Zijderveld et al.
(1991), that level is close to the top of the Olduvai normalpolarity subchron. Following that concept, the N/Q boundary
should be located somewhere in the lower part of the
Apsheronian beds of the Caspian Basin and their stratigraphic
equivalents. That boundary level has now been officially
accepted by stratigraphic workers in the former USSR.
The Eopleistocene. As is well known, the "Eopleistocene," as
used throughout the former USSR and in most of eastern Europe,
corresponds to the Lower Pleistocene in western European and
North American stratigraphy. According to the correlation of the
Vrica definition, the lower part of the Apsheronian regiostage
contains the base of the Eopleistocene and the boundary of the
Quaternary. That boundary concept was used in preparing a
detailed chronostratigraphic scheme for late Cenozoic deposits in
European Russia and adjacent states (Nikiforova et al., 1976,
1980; Nikiforova and Alexandrova, 1991). This scheme can be

successfully applied to most, if not all, of northern Eurasia,
although with different degrees of precision.
A correlation of the Vrica boundary-stratotype to a level
within the lower half of the Apsheronian, rather than at its very
base, is supported by paleomagnetic analysis. Gurarij, Pevzner,
and Trubikhin (1973) correlated the normally magnetized beds
within the lower parts of the Apsheronian sequences in
Azerbaijan and Turkmenia with the Olduvai paleomagnetic
event, which is identified with the stratigraphic interval just
below the proposed boundary-stratotype at Vrica (Pasini and
Colalongo, Chapter 2, this volume). Similar correlations have
been made by Pevzner and Tshepalyga (1970) through
paleomagnetic analysis of matrix adhering to Apsheronian
mammal fossils collected from the terraces of large rivers in
Moldavia and in southern Ukraine (the Dniester, Danube, and
Prut rivers), as well as from material from other regions.
Additional information comes from the work of V. M. Trubikhin
(Nikiforova, 1987).
The southern part of the region is quite favorable for
recognizing the lower boundary of the Quaternary. Although
direct comparisons of marine deposits in the Caspian and Black
Sea basins with those in the Mediterranean are hampered by the
dissimilarities of the marine invertebrate faunas, correlations can
be made with the help of the mammal and fresh-water mollusk
faunas, augmented by radiometric dating, stable-isotope curves,
and particularly paleomagnetic sequencing.
Pre-Quaternary stratigraphy
The Permanent Quaternary Commission of the former USSR

Interdepartmental Stratigraphic Committee suggested new terminology for the subdivision of chronozones, which in the Soviet
scheme have the scale of epochs. In English, these units, in order
of descending hierarchy, are divisions (i.e., sub-chronozones);
links; over-steps (i.e., super-steps); and steps. The latter two are
the equivalents of over-horizons and horizons in regional
biostratigraphic schemas. In this terminology, the Upper Pliocene is considered to be a division (i.e., the "upper division of
the Pliocene")- The lower division of the Pliocene includes the
221
222
Cimmerian, or at least its upper part, the Kamyshburunsky, and and Prolagurus (Lagurodon) arankae (Azzaroli et al., 1988;
the Panticapeisky horizons (Semenenko and Pevzner, 1979).
Chaline, Chapter 14, this volume). The beginning of the
The Upper Pliocene division is composed of two links. The Eburonian is located in The Netherlands just at the top of the
lower link, equivalent to the lower Akchagylian, corresponds to Olduvai event (Zagwijn, 1974, 1985; Zagwijn, Chapter 16, this
continental facies with the Moldavian fossil mammal complex. In volume). According to this interpretation, leading from biothose deposits, the key mammal taxa are Anancus arvernensis, stratigraphy to magnetostratigraphy, the Domashkinian coolDicerorhinus jeanvireti, archaic elephants of the genus Archi- climate deposits thus closely coincide with the strata at the base
diskodon (or Mammuthus according to some specialists; see of the Quaternary in the marine sequence at Vrica.
Aguirre et al., Chapter 9, this volume), species of the micro tine
The upper link of the Eopleistocene is composed of the
rodents Dolomys and Pliomys, and in the uppermost levels the uppermost middle Apsheronian and the upper Apsheronian,
first Mimomys (e.g., M. polonicus) (L. P. Alexandrova, in correlative to the Tamanian complex of mammals in continental
Nikiforova, 1987; Chaline, Chapter 14, this volume). The upper facies of European Russia and the adjacent states. That interval
link includes the middle and upper Akchagylian, which corre- in western Europe is represented by the transitional or
spond to continental facies with Khaprovian mammal assem- Epivillafranchian sequences, and in western Siberia by the
blages. The key forms of the Khaprovian are Archidiskodon Razdolinian complex. In those deposits the key taxa are
gromovi, diverse species of Mimomys, including M. pliocaenicus, Archidiskodon meridionalis tamanensis and the related western
and in upper Khaprovian levels the first Villanyia, as V. laguro- subspecies cromerensis and vestinus (Aguirre et al., Chapter 9,
dontoides. The boundary between the Moldavian and Khapro- this volume; Azzaroli et al., Chapter 11, this volume), and
vian is close to the Gauss-Matuyama boundary, at about 2.6 Ma. Microtus (Pitymys) appears in the small-mammal faunas. The
The Moldavian mammal assemblages are similar to those of boundary between these two links coincides approximately with
the late Ruscinian (Csarnotian) and Lower Villafranchian faunas the base of the Jaramillo subchron.
of southwestern Europe and to the Reuverian of the northern
The boundary at the top of the Eopleistocene, formerly considEuropean scale. In western Siberia, the Betekian complex ered to be the base of the Pleistocene and the beginning of the
appears to correspond to the uppermost phase of the Moldavian Quaternary by many workers in the former USSR, lies at the base
complex (Veselovsky horizon), but older phases are missing. of the Tyurkanskaya suite, at the beginning of the Bakuan
Deposits with Khaprovian mammals are mostly correlative to the regiostage. In continental facies, that is correlative to the deposits
upper part of the Lower Villafranchian unit in western Europe with the earliest elements of the Tiraspol faunistic complex (i.e.,
(characterized by the fauna of Montopoli) and possibly, in the faunas with Microtus raticepoides), equivalent to the latest
uppermost Khaprovian, to part of the Middle Villafranchian. It Biharian and earliest Cromerian of western Europe (L. P. Alexis probable, however, that the FAD (first-appearance datum) of androva, in Nikiforova, 1987). Strata of this age have negative
Archidiskodon meridionalis, seen in the Psekups fauna, is remanent polarity at Solilhac and Tiraspol and are thus older than
younger than the Khaprovian, although older than the Odessan the Brunhes-Matuyama. At an earlier level, the "Betfia" phase of
complex. The first appearance of this proboscidean is a marker early Biharian age marks the end of the Villafranchian of western
for the Middle Villafranchian in western Europe, as at Saint- Europe; it is characterized by cold-adapted assemblages which
Vallier, Seneze (lower level), and Tegelen (Azzaroli et al., 1988; appear to correspond to a glacial-climate (Menapian) phase in the
Aguirre et al., Chapter 9, this volume; Azzaroli et al., Chapter uppermost Matuyama, above the Jaramillo.
11, this volume). In western Siberia the Podpusk-Lebyashinsky
The Russian Pleistocene (equivalent to the Middle and Upper
faunal complex is roughly equivalent to the Middle Villafranch- Pleistocene in western Europe and North America) consists of
ian (Alekseev, Chapter 22, this volume).
three links. The lower link, as found throughout the region west
of the Urals and in the Caspian and Black Sea basins, is characterized by the main part of the Tiraspol fossil mammal complex. In
Eopleistocene and Pleistocene stratigraphy
the main Tiraspol fauna there are at least two subdivisions in
In the Eopleistocene, two links are distinguished. The lower link northern Eurasia, with the lower part equal to the Bakuan
embraces the lower Apsheronian and most of the middle regional stage and the "true" Cromerian or Giinz-Mindel
Apsheronian, corresponding approximately to the continental interglacial interval of western Europe, and the upper, or Oksky,
facies with the Odessan complex of mammals, in which part of a cold-climate faunal interval corresponding to the
Archidiskodon meridionalis meridionalis is a key element. The Elsterian or Mindel glacial-climate period of western Europe,
presence of the first rootless-toothed voles, Allophaiomys and the beginning of which is dated at about 0.54 Ma. Thus, in the
Prolagurus, indicates that the Odessan complex is equivalent to scheme of the European part of the former USSR, at least eight
the early Upper Villafranchian. In particular, we may correlate horizons (or steps) are distinguishable in the Lower Pleistocene.
the fauna of the Domashkinian horizon at the base of the
Odessan complex to the earliest level of the Upper VillafranchThe Anthropogene question
ian, represented by the Olivola fauna, to the lower Eburonian
climatic stage of The Netherlands, and to the Kizikhan complex The earliest evidence of humans, or hominids, is of special
of western Siberia, by the presence of Allophaiomys pliocaenicus significance because Soviet scientists have long emphasized a
European Russia, Ukraine, and Moldava
connection between the Quaternary sub-period and human

activity. On the African continent, the oldest pebble-tool
cultures date to more than 2.3 Ma, and the earliest acknowledged remains of true Homo, presently classified as H. rudolfensis, are even older (Aguirre, Chapter 10, this volume).
Outside of Africa, however (i.e., north of the Levant), the
appearance of humans in the form of Homo erectus is evidenced
by tools or fossil remains from sites at Sangiran in Java,
Gongwangling and Nihewan in China, Dmanisi in Georgia, and
possibly Orce and Cueva Victoria in Spain. All of those southern
Eurasian sites are older than the Jaramillo subchron, and the
oldest may be almost as old as the first evolutionary appearance
of the H. erectus clade in Africa at about 1.6 Ma (Aguirre,
Chapter 10, this volume). Although the rarity of such sites leaves
a small margin of uncertainty, we may be justified to conclude
that Homo did not become established in (southern) Eurasia
until after the beginning of the Quaternary, as presently defined,
and before the Quaternary boundary previously adopted in the
former USSR at the base of the Bakuan regiostage in the
uppermost Matuyama, above the Jaramillo.
Climate changes
Although correlating the deposits of the Pliocene and Lower

Quaternary according to continental glacial episodes is a
debatable means of dating, climatic fluctuations during that
interval are recognized by all scientists.
In marine deposits, the late Pliocene (Piacenzian) of southern
Europe is characterized by warm-climate invertebrates, although
with some colder fluctuations near the Gauss-Matuyama boundary. In terms of the late Pliocene mammalian faunas, the first
deterioration of the climate was seen in the early Villafranchian,
and climatic oscillations tended to become more significant
during the middle Villafranchian. An appreciable cooling in the
beginning of the middle Villafranchian was marked by a major
erosional phase, the Aquatraversan, in central Italy (Bigazzi,
Bonadonna, and Iaccarino, 1973) and by sharp changes in
vegetation distinguished by the regional extinction of Taxodium,
Sequoia, and Sciadopitys, marking the base of the Tiberian
paleoflora (Lona and Bertoldi, 1973). In The Netherlands, the
Pretiglian deposits with "cold" flora, dated paleomagnetically to
just above the Gauss-Matuyama boundary, correspond to that
time (Zagwijn, 1985; Zagwijn, Chapter 16, this volume). In the
Soviet schema, the late Pliocene cold period corresponded to a
cold phase in the middle Akchagylian, represented by the
Chistopol horizon of the Kinel Formation of the Middle Volga
and Kama regions. The deposits in that horizon are characterized by taiga flora (Nikiforova et al., 1976).
A lesser cooling seems to be reflected in the specific features
of the Tiglian B horizon in The Netherlands and the lower
Kuyalnik beds from the Odessa region (Kryzhanovka horizon)
containing a cold-adapted mollusk fauna.
The severe climatic deterioration associated with the classic
concept of the Calabrian is correlative to the "cold" fauna of
Olivola (Azzaroli et al., Chapter 11, this volume) and the "cold"
223
(upper) faunal levels at Seneze (Bout, 1970) at the beginning of

the Upper Villafranchian and also to "cold" flora of the Rhine
Valley Eburonian, both correlated to the top of the Olduvai
subchron (Zagwijn, 1974, 1985). As noted earlier, in our region
that cooling corresponds to the very beginning of the
Apsheronian (Domashkinian horizon), represented by the
Paludina beds with a boreal molluscan fauna overlying the
marine upper Akchagylian in the Lower Volga region. The lower
Apsheronian of the Duzdag has also a similarly boreal fauna
(Nikiforova et al., 1976).
In conclusion, the base of the Upper Villafranchian in Italy,
the base of the Apsheronian in the western part of the former
USSR, and the base of the Eburonian cold-climate stage of the
North Sea Basin correlate in terms of paleoclimate, paleontology, and paleomagnetism to the base of cold-climate marine
deposits at Vrica, Italy, that conform to the traditional concept of
the Calabrian Stage designated by the London Commission
(IGC 1948) to define the beginning of the Pleistocene. This level,
which is close to the top of the Olduvai normal-polarity chron,
represents the beginning of the Quaternary wherever it can be
identified, and the cooling at that time can be traced on a global
scale.
Following the cold-climate Domashkinian (lowermost Apsheronian) horizon, the beginning of the middle Apsheronian is
characterized by impoverished spore-pollen spectra in which the
broad-leafed species decline and disappear. The pollen indicate
a warming in the upper part of the middle Apsheronian which
correlates with the warm Baventian stage of The Netherlands. The cooling of the Morozovka horizon of the upper
Apsheronian (marked by the complete disappearance of pollen
from broad-leaf trees and the predominance of pollen from
steppe vegetation) corresponds to the Linge and Dorst glacial
episodes.
Invertebrate biostratigraphy
Climaticfluctuationsare also reflected in the molluscan faunas in

continental deposits in the southern European parts of the
former USSR, which correlate with the Akchagylian and
Apsheronian (Table 21.1). According to Tshepalyga (1972), the
two major stages, called by that author the Levantine and the
Pleistocene, can be distinguished according to the evolution of
molluscan faunas in the East European province. Characteristic
of the Levantine stage are the appearances of new genera and
subgenera. The most significant changes in the evolution of
fresh-water mollusks are seen in the Kishlitzian (Kimmerian)
and upper Poratian horizons, including the appearance and
development of sculptured forms (thermophilic fauna) during
the early and middle Akchagylian (early Villafranchian). By the
midpoint of the middle Akchagylian (beginning of the middle
Villafranchian) the fauna had already changed. The Levantine
elements had disappeared, and the typical boreal fauna without
"warm" elements had come into being, as seen in the Chistopol
horizon. By the end of the middle Akchagylian transgression
climax, the thermophilic molluscan fauna had returned, only to
224
Table 21.1. Mollusk zonation of the Plio-Pleistocene sequence in southwestern Russia and adjoining states
Epoch
LOWER
PLEISTOCENE
Stage
Bakuan
Complex
Biozone
Kolkotovian
Pseudunio moldavica
Platovian
Viviparus pseudoachatinoides
Mikhailovian
Crassiana crassoides
Morozovian
Potomida litoralis
Kosnitsian
Pseudosturia candata
Nesmeyanian
Bogatschevia scutum
Boshernitsian
Bogatschevia sturi
Domashkinian
Unio apsheronicus
Polivadinian
Bogatschevia tamanensis
Kryzhanovian
Unio kujalinicensis
Akkulaevian
Erbersininaia
Age (Ma)
0.8
EOPLEISTOCENE
Apsheronlan
-1.6
UPPER PLIOCENE
Akchagyllan
-2.4
Simbuchinian
Potomida bashkirica
Poratian
Rugunio ienticularis
3.3
MIDDLE PLIOCENE
Klmmerlan
Kimmerian
Unio sturdzae
Source: Adapted from Tshepalyga (1972).
be replaced once more in the beginning of the late Akchagylian

by cold-adapted, non-sculptured forms. At the end of the late
Akchagylian, the Levantine species of Unionidae and Viviparidae, particularly Potomida tamanensis, reappeared once
more, but with less emphasized sculpture.
The beginning of the Apsheronian (late Villafranchian) was
characterized by the complete absence of Levantine elements
(Domashkinian fauna), reflecting the sharp cooling associated
with the beginning of the Quaternary. The subsequent outburst
of speciation and sculptured forms (Boshernitskan fauna with
Bogatschevia sturi) in the early Apsheronian is evidence for a
new amelioration of the climate. The cooling trend of the middle
Apsheronian (Kosnitsa fauna, with Pseudosturia caudatd) was
succeeded by a short-term warming, which in turn was followed
by cooling and the development of the Morozovian fauna, during
which the last Levantine elements became extinct (Tshepalyga,
1972). The succeeding warm period at the end of the
Apsheronian saw the development of the Kolkotovian assemblage, most species of which are living today.
Geological history
The climatic fluctuations discussed earlier are clearly reflected in

the superficial deposits widely distributed in the southern
European parts of the former USSR. The oscillations of climatic
conditions between dry-and-cold and wet-and-warm during the
late Pliocene and early Quaternary are marked by alternating
loamy, frequently loess-like deposits and fossil-soil horizons
(Nikiforova etal., 1976).
According to students of molluscan biostratigraphy (Alizade
et al., 1972; Tshepalyga, 1972), significant changes in molluscan
faunas and in geotectonic regimes occurred in both the Black Sea
and the Caspian Sea regions at closely correlative levels of the
late Pliocene and early Quaternary:
1. Akchagylian-Kuyalnikan (predominance of the Pliocene fauna)
2. Apsheronian-Gurian (mostly Pliocene fauna; appearance of earliest Quaternary mollusk lineages)
European Russia, Ukraine, and Moldava
3. Bakuan-Chaudan (predominance of Quaternary mollusks; rare Pliocene relicts)

Alizade et al. (1972) considered that the last stage should be
equivalent to the base of the Quaternary. It can be seen,
however, that the Quaternary fauna began to appear in the
second stage, which coincides with the customary (if not
stratotypical) Calabrian, or Selinuntian, of Italy, defined by the
Vrica level at which the lower boundary of the Quaternary is
drawn.
The character of the Apsheronian fauna suggests that the
salinity of the Apsheronian basin was similar to that of the
present-day Caspian, although some parts of the basin may have
had lower salinity because of major rivers discharging into the
sea. The flora of the Apsheronian is indistinguishable from the
modern flora (Bogachev, 1936). The climate of that time also
seems to have differed only slightly from today's climate.
Intense volcanic activity was characteristic of the Apsheronian, as indicated by abundant interlayers of volcanic ash brought
far to the east of the eruption centers by winds and sea currents.
In the beginning of the Apsheronian, there were also significant
movements of the earth's crust. The Apsheronian sea had lower
salinity than the Akchagylian, despite the reduction in its surface
area as a result of the tectonic movements; the depth of the basin
was also increased (Andrusov, 1923). As an example, in the
Apsheron peninsula, relatively shallow-water Akchagylian deposits are succeeded by deeper-water Apsheronian sediments,
demonstrating subsidence of the seafloor at that time. Some
subsidence was also characteristic of the Manysh area. Maximum
tectonism was recorded at the beginning of the Apsheronian, in
both the Caspian and the Black Sea regions.
Thus, the transition from the Akchagylian to Apsheronian was
accompanied by major tectonism that caused the Caspian Basin
to become smaller and deeper. Mass appearances of Limnaea
and, more frequently, Dreissena mollusks seem to have resulted
from freshening of the water at the same time. In the middle
Apsheronian, the Caspian-type fauna penetrated the Euxinian
region through the Manysh strait, and a similar but impoverished
fauna inhabited the Black Sea basin in the late Apsheronian
(Koleshnikov, 1940).
Renewed tectonism at the end of the Apsheronian changed
the configuration of the Caspian Basin, downwarping the
northern part. That opened a new connection between the
Caspian and Euxinian basins through the Manysh strait and
resulted in immigration of characteristically Bakuan Cardiidae
from the Didacna crassa group.
The Gurian beds in the Black Sea basin are the equivalents of
the Apsheronian of the Caspian region. The Gurian flora of
Georgia differs sharply from the preceding Pliocene flora, in that
evergreens were replaced by deciduous species that are still
common in Transcaucasia. The Gurian basin appears to have
been located mainly inside the present Black Sea basin; for that
reason, Gurian beds are relatively scarce, occurring mainly in
western Transcaucasia and in the northwestern part of the Kerch
(Chersonese) Peninsula.
225
References
Alizade, K. A., Ali-Zade, A. A., Fedorov, P. V., Gadzhiev,
D. V., Vekilov, B. G., and Asadullaev, E. M. 1972.
Position of Neogene-Quaternary boundary according to
the data of Ponto-Caspian brackish water fauna (in
Russian). In International colloquium on the problem
"The Boundary between the Neogene and Quaternary" (in
Russian), ed. M. N. Alekseev et al., pp. 1-26. Moscow:
Acad. Nauk SSSR.
Andrusov, N. I. 1923. Apsheronian stage (in Russian). Geol.
Com., Proc, n.s. 110:1-294.
Late Pliocene to early mid-Pleistocene mammals in Eurasia: faunal succession and dispersal events. Palaeogeogr.
Bigazzi, G., Bonadonna, F., and Iaccarino, S. 1973. Geochronological hypothesis on Plio-Pleistocene boundary in Latium
region (Italy). Soc. Geol. It., Boll. 92:391-422.
Bogachev, V. V. 1936. Freshwater and terrestrial molluscs from
Upper Tertiary deposits of Kura river basin (in Russian).
Trudy, vol. 13. Baku: Akad. Nauk Azerbaijanskii.
Bout, P. 1970. Absolute ages of some volcanic formations in the
Pleistocene. Palaeogeogr. Palaeoclimatol. Palaeoecol. 8:
95-106.
Gurarij, G. Z., Pevzner, M. A., and Trubikhin, V. M. 1973.
Paleomagnetic scale of Late Cenozoic deposits of Caspian
basin. In Proceedings of IX Conference on permanent
geomagnetic field of magnetic rocks, (in Russian), vol. 3,
pp. 133-143. Baku: Akad. Nauk Azerbaijanskii.
Koleshnikov, V. P. 1940. Middle and upper Pliocene of the
Caspian region. In Stratigraphy of the USSR, Volume 12:
Neogene, pp. 407-476. Moscow: Leningrad Academy of
Science Publishing House.
Lona, F., and Bertoldi, R. 1973. La storia del Plio-Pleistocene
italiano in alcune Sequenze vegetazione ali lacustri e
marine. Mem. Cl. Sci. Fis. Mat. Natur., Ser. VIII 11:1-45.
Nikiforova, K. V. 1987. Extent of the Eopleistocene (in
Russian). Akad. Nauk SSSR, Izv., ser. Geol. 10:79-92.
Nikiforova, K. V., and Alexandrova, L. P. 1991. Late Pliocene
and Anthropogene stratigraphy, chronology and correlation of the events of Europe and North America connected
with the changes in the geosphere and biosphere (in
Russian). In Pliocene and Anthropogene paleogeography
and biostratigraphy (in Russian), pp. 99-123. Moscow:
Nauka.
Nikiforova, K. V., Ivanova, I. K., and Kind, N. V. 1987. Urgent
problems of the Quaternary chronostratigraphy (in Russian). In New data on the Quaternary geochronology (in
Russian), pp. 15-23. Moscow: Nauka.
Nikiforova, K. V., Krasnov, I. I., Aleksandrova, L. P., Vasiliev,
Y. M., Konstantinova, N. A., and Tshepalyga, A. L. 1976.
Climatic fluctuations and detailed stratigraphy of Upper
Pliocene-Lower Pleistocene deposits of the USSR South
(in Russian). In Geology of Quaternary deposits (in
Russian), ed. K. V. Nikiforova et al., pp. 101-119. XXV
Session IGC, Paris, Papers of Soviet Geologists. Moscow:
Nauka.
Nikiforova, K. V., Krasnov, I. I., Aleksandrova, L. P., Vasiliev,
Yu. M., Konstantinova, N. A., and Tshepalyga, A. L.
1980. Chronostratigraphic scheme of the Late Cenozoic of
the European part of the USSR (in Russian). In Quater-
226
nary geology and geomorphology, distant soundings (in

Russian), pp. 55-67. XXVI Session IGC, Sydney, Papers
of Soviet Geologists. Moscow: Nauka.
Pevzner, M. A., and Tshepalyga, A. L. 1970. Paleomagnetic
investigations of Pliocene-Quaternary Dnestr River terraces (in Russian). Akad. Nauk USSR, Dokl 194:43-46.
Landini, W., Menesini, E., Mezzeti, R., Pasini, G.,
(Calabria-Italy). A potential Neogene-Quaternary boundary-stratotype. Giorn. Geol. 41:181-204.
Semenenko, V. N., and Pevzner, M. A. 1979. Correlation of
Upper Miocene and Pliocene of Ponto-Caspian area
according to biostratigraphic and paleoclimatic data (in
Russian). Akad. Nauk SSSR, Izv., ser. geol. 1:67-92.
Tauxe, L., Opdyke, N. D., Pasini, G., and Elmi, C. 1983. The
paleomagnetism of the Vrica section (Calabria, Italy), the
proposed Pliocene/Pleistocene boundary-stratotype section. Nature 304:125-129.

Tshepalyga, A. L. 1972. Neogene-Quaternary boundary according to the data of terrestrial molluscs. In International
colloquium on the problem "The Boundary between the
Neogene and Quaternary," ed. M. N. Alekseev et al., pp.
46-57. Moscow: Acad. Nauk SSSR.
Van Montfrans, H. M. 1971. Paleomagnetic dating in the North
Zagwijn, W. H. 1974. The Pliocene-Pleistocene boundary in
western and southern Europe. Boreas 3:75-97.
Zagwijn, W. H. 1985. An outline of the Quaternary stratigraphy
of the Netherlands. Geol. Mijnb. 64:17-24.
107:697-714.
22
The N/Q boundary in Asian Russia and Tadjikistan

MIKHAIL N. ALEKSEEV
Introduction
On the Siberian Platform (i.e., the region extending from the

Urals eastward to the Pacific Ocean), biostratigraphic, radiometric, and paleomagnetic scales have been applied to
Pliocene and Pleistocene deposits to correlate them not only with
the stratotype sequence of Vrica but also with key sequences of
the European part of the former USSR and in Europe.
Stratigraphic summary
Nine areas are particularly helpful in studying the PlioPleistocene succession of this region. These are western Siberia,
Transbaikalia, Olkhon Island in Lake Baikal, the upper basin of
the Lena River, central Yakutia, the Kolyma Basin, Kamchatka,
and the Primorie/Priamurie region (i.e., the Pacific coast
provinces), all in eastern Russia, and the Tadjik Depression of
Tadjikistan (Figures 22.1 and 22.2). There are difficulties,
however, including the unequal extent of exploration, inadequate exposures, and the vast area of territory in Siberia affected
by permafrost. As a result, few sequences are adequately known.
Western Siberia
ria developed taiga vegetation, and the north developed foresttundra and tundra associations.
The overlying Podpusk-Lebyaginsk layers show wide adaptive
radiations of Equus species, the dispersal of Archidiskodon (or
Mammuthus according to some specialists) (Aguirre et al.,
Chapter 9, this volume) and Eucladoceros, reductions and
eventual disappearances of thermophilous molluscan faunas in
this region, and the initial appearances of recent species. Major
changes in the entire environment are registered at the transition
from the Podpusk-Lebyaginsk to the overlying Kizikhinsk layers,
that is, at the boundary of the Pliocene and Eopleistocene
(between 2.0 Ma and 1.7 Ma). That time was marked by intense
deterioration of the climate, as indicated by the appearance of
tundra and forest-tundra assemblages on the drainage divides
and fir-tree forests in the valleys of southwestern Siberia.
Rootless-toothed voles such as Allophaiomys, Prolagurus, and
Eolagurus appeared in the rodent fauna for the first time. Thus,
the boundary between the Podpusk-Lebyaginsk and Kizikhinsk
layers reflects the usual small-mammal biostratigraphic changes
at the boundary between the Neogene and Quaternary (Aguirre
et al., Chapter 9, this volume; Chaline, Chapter 14, this
volume).
Evidence of some warming is found in the upper part (late
Eopleistocene) of the Kochkov Formation, and that was again
followed by renewed cooling, with the maximum climatic
deterioration seen in the Shaitan horizon, in an interval
coinciding approximately with the Mindel glacial period in
Europe.
The Plio-Pleistocene deposits of southwestern Siberia, around

the headwaters of the Ob River, can be considered of paramount
importance, as they include faunistic complexes that support
extensive biostratigraphic correlations. The Kochkov Formation,
which covers the Upper Pliocene and Eopleistocene, from about
2.4 Ma to 0.6 Ma, contains a series of sediments with four faunal
complexes (Figure 22.2). The layers of Betekian age at the base
Tadjik Depression
contain remains of Hipparion sp., Paracamelus gigas, Paracamelus praebactrianus, Trogontherium minor, Promimomys In Tadjikistan, strata relating to the Upper Pliocene and Lower
gracilis, Mimomys polonicus, Mimomys hintoni, Villanyia Pleistocene in the Tadjik Depression and in the eastern Pamirs
petenyii, Villanyia steklovi, and so on (Zazhigin, 1980).
provide evidence of three major geological boundaries, dated to
The Betekian, according to the paleobotanical and geological about 3.6, 2.2, and 0.7-0.8 Ma by paleomagnetic analysis. The
data of Volkova and Baranova (1980), was characterized by a level corresponding to the Olduvai paleomagnetic interval (1.95fairly warm climate and forest vegetation that spread far beyond 1.77 Ma) is located in the lower part of the middle unit, the
northwestern Siberia. In the final stages of the Betekian (2.4 Ma) Kayrubak suite (Dodonov, 1980); however, a more distinct
the evidence indicates considerable cooling; southwestern Sibe- biostratigraphic boundary is somewhat earlier, at about 2.2 Ma,
227
Mikhail N. Alekseev
228
Figure 22.1. Key areas of

Neogene and Quaternary deposits in the Siberian Platform region: 1, Western Siberia; 2,
Transbaikalia; 3, Olkhon Island
in Lake Baikal; 4, upper Lena
River drainage; 5, central
Yakutia; 6, Kolyma Basin; 7,
Kamchatka; 8, Priamurie/
Primorie; 9, Tadjik depression.
at the transition between the Kayrubak and the underlying

Kuruksay suite.
Transbaikalia
deposits with a Dodogolian mammalian fauna, characterized by

Citellus sp., Villanyia laguriformis, another Villanyia sp., and a
Prosiphneus sp., collected from the red clays and loams exposed
in the lower part of the Dodogol sequence.
The stratigraphically overlying sediments of slope-proluvial
origin contain fossil mammals of the Itantsa complex. They
include Equus ex gr. E. sanmeniensis Teilhard & Piveteau,
Villanyia sp., Mimomys sp., Prosiphneus sp. ex gr. P. youngi,
and others. The Itantsa complex is correlated with the Tamanian
mammalian complex of eastern Europe (Vangengeim, 1977),
where Tamanian-age deposits are considered to be the lowermost part of the Quaternary sequence.
In southeastern Siberia and Transbaikalia, Vangengeim (1977)

and Zazhigin (1980) distinguished the Tchikoian faunal complex,
which correlates with the Betekian and the lower parts of the
Podpusk-Lebyaginsk complexes of western Siberia. Abundant
remains of large mammals and rodents have been found in the
key section at Beregovaya in a layer of reddish, clayey sands.
The fauna includes Hipparion sp. ex gr. H. houfenense Teilhard
& Young, Hipparion tchikoicum Ivanijev, Dicerorhinus sp.,
Gazella cf. G. sinensis Teilhard & Piveteau, Antilospira sp.,
Lake Baikal
Paleotragus sp., Canis cf. C. chihlienensis var. minor Teilhard &
Piveteau, Nyctereutes cf. N. sinensis (Schlosser), Euryboas cf. E. Four mammalian and five molluscan complexes have been
lunensis Camp, Felis (Lynx) shansius Teilhard, Acinonyx sp., identified in the Pliocene-Lower Pleistocene deposits of Olkhon
Mimomys cf. M. reidi Hinton, Mimomys minor Fejfar, Villanyia Island in Lake Baikal (Logachev, 1982). Environmentally
sp., Sinocastor sp., and Prosiphneus sp. ex gr. P. praetingi induced changes in the composition of mammalian and mollus(Vangengeim, 1977). That complex can be dated to the begin- can faunas were closely parallel and clearly reflect the environning of the Middle Villafranchian (Lower Eopleistocene and mental and magnetic changes of the proposed boundary between
beginning of the Middle Eopleistocene). The red formation in the Neogene and the Quaternary.
the lower part of the Tologoi section, which also belongs to the
At the base, the contact between the colored sediments of the
Tchikoian horizon, is normally magnetized and belongs to the Olkhon Island suite and the underlying clays of the Kharanskian
Gauss paleomagnetic epoch, according to Gnibidenko, Erba- suite has been dated to 3.0-3.5 Ma, according to paleomagnetic
yeva, and Popelova (1976). The boundary between the Pliocene measurements. The Sasinsk deposits at the base of the Olkhon
and Eopleistocene in Transbaikalia is drawn at the base of the Island suite contain the Saraian mammalian faunal complex,
Asian Russia and Tadjikistan
PALEQMAGNETIC
SCALE
WESTERN
SIBERIA
TRANSBAIKALIA
QLKHDN ISLAND
LAKE BAIKAL
CENTRAL
YAKUTIA
UPSTREAM
DF LENA RIVER
229
PRIMDRIE oL
PRIAMURIE
KDLYMA BASIN
TAJIK
DEPRESSION
STRATIBRAPJ
HID
i
UNITS
BASALTS OF
GAMCHEN RAN9E
ALLUVIUM wrrw
SHAITAN
HORIZON
KIZIKH1AN
BESS
PROLUVIUM WITH
TOLQGOI FAUNAL
COMPLEX
AKANIAN
BEOS
RED BROWN LOA NYURGANJAN

SUITE
MS WITH FDSSILS
OF ITANSA FAU
NAL COMPLEX
BETEKFJAN
BESS
IS
KANGILSK
FORMATION
SASIN
FORMATION
OLIOR
FORMATION
ALLUVIUM OF
HIGH TERRACES
(LENA,ALDAN,
VILYUY RIVERS)
UPPER RED
FORMATION
CHIKO1 FORMATION
VAKHSH
COMPLEX
ALLUVIUM WTI
ALDAN FAUNA
MANZURKA
AND AN6A
FORMATIONS
KHARANTSYNTAN
SUITE
PDDPUSKLEBIAZHJAN
BEDS
BROWN LOAMS
IULTSK VOLGA
NIC COMPLEX
LOWER RED
FORMATION
DYGDAL
FORMATION
TUMROK
COMPLEX
RED-COLOURED
FORMATION
KAYRUBAK
SUITE
KUTUJAKH
BEDS
SCHAPINSK
FORMATION
KURUKSAV
SUfTE
LIMIMTEVAJAM
FORMATION
FERRUGENEOUS
SANBSOF MAMMOTH HILL,
SUYFUN
SUITE
P0LI2AK
SUITE
Figure 22.2. Correlation of Pliocene and Lower Pleistocene key sequences in Asian Russia and Tadjikistan.
with elements close to the Lower Pliocene fauna of the Pavlodar

suite and the "Pontian" of China. At the same time, the presence
of Microtocoptes and Microtodon in the Saraian fauna makes it
similar to the Novostanichian fauna, dated to the Middle
Pliocene in western Siberia. Thus, the age of the Saraian fauna in
the Baikal area is estimated to be Lower-Middle Pliocene. The
Saraian climate was warm and dry, probably steppes with spots
of bushes and forest vegetation.
The paleomagnetic level corresponding to the beginning of
the Olduvai event is found above that horizon, in the upper
part of the Kharanskian suite. It is associated with the first
appearances of the late rodent species of the VillanyiaMimomys assemblage. Resting on eroded Kharanskian strata,
beds containing the Eolagurus-Lagurodon fauna, with single
species of Allophaiomys and Mimomys, characterize the
Njurganian suite, which embraces the Eopleistocene and
Pleistocene. The 0.78-Ma level, at the transition from the
Matuyama to the Brunhes, is recorded only by polarity changes,
without any marked faunal change.
Lena River basin
In central Yakutia, the sequences in the Lower Aldan depression, the valleys of the middle Lena River, and the lower course
of the Vilyuy River provide significant data for regional
correlation.
In the upper part of the Mammoth Hill section on the Aldan
River, an alluvial layer termed the "ferruginous sands" is
assigned to the Pliocene, based on palynology in overlying and
underlying deposits (Baranova et al., 1976). Paleomagnetic
studies (Minjuk, 1982) indicate that the ferruginous sands were
deposited in the Gilbert epoch. In the palynoflora, the main
Angiospermae pollen are Betula and Alnus, while the Gymno-
spermae consist mainly of Pinus and Picea. That layer is widely

exposed in the Quaternary terrace deposits of the Aldan River,
and a layer with similar palynologic and paleomagnetic character
can be traced in the subsurface levels of the Lower Aldan
depression.
Sands of the Dygdal suite that have the normal polarity of the
Gauss epoch occur in the downwarped central part of the Lower
Aldan depression.
In the Lena River valley, upstream of Yakutsk, normally
magnetized alluvial deposits occur in elevated terraces, between
100 and 120 m above base level. Eopleistocene deposits, with
remains of Alces latifrons, Bison aff. B. schoetensacki, Equus sp.
ex gr. E. sanmeniensis, and Trogontherium cf. T. cuvieri, were
found resting on the bedrock of the Aldan River terraces. The
reversed polarity of those beds corresponds to the Matuyama
epoch. Although the section is of insignificant thickness, the
boundary between the Neogene and Quaternary, as related to
the Olduvai event, must be drawn within that interval. More
precise identification of the boundary position is a matter for the
future.
Sequences in the Aldan and Vilyuy rivers also contain the
boundary between the Matuyama and Brunhes epochs. In the
Vilyuy basin there are alluvial deposits in the 50-60-m terrace
that contain a fauna close to that of the Tiraspolian gravel, and in
which the basal Brunhes paleomagnetic reversal has been
identified. The same relationships between paleomagnetic
events and the distribution of Tiraspolian fossils have been
established in the Kolkotova Balka sequence in Moldavia.
Kolyma Basin
In northeastern Russia, in the Beringian province east of the
Yakut basin, 50-60% of the early Pliocene floras show features
230
Mikhail N. Alekseev
that connect them with the seeds, leaves, and pollen of living
deciduous flora (Volkova and Baranova, 1980). In the interval
between 3.5 Ma and 3.0 Ma, when there was a decrease in land
area and the Bering land bridge was inundated, the region's
vegetation was transformed into something similar to forest
tundra.
In the upper layers of the Kutujakh (or Kututyak) Formation,
E. I. Virina (in Volkova and Baranova, 1980) identified an
interval of normal polarity in the zone of reversed magnetization
of the Matuyama epoch that is considered to represent the
Olduvai event. The lower Kutujakh Formation yielded rare
remains of a fossil mammal fauna similar to that of the
Khaprovian Podpusk-Lebyaginsk assemblage, together with the
earliest distinct signs of cryoturbated soil of the Hypoarctic zone
with perannual permafrost. The pre-Quaternary intensification
of Arctic basin glaciation, at about 2.5 Ma, probably can be
referred to this period.
In the Kolyma lowland, the type section of the Olior (Olyor)
suite documents the upper Matuyama, including the Jaramillo
subchron, and the lowest part of the Brunhes (Sher, 1987). The
boundary between Eopleistocene and "true" or glacial Pleistocene could be drawn, in the type Olior, between the layers with
an early Olior small-mammal fauna, correlated to the Tamanian
complex, and the layers with a late Olior small-mammal fauna,
correlated to the Tiraspolian complex.
Kamchatka
Within the Kamchatka-Chukotka region, age calibrations of
magnetostratigraphic reversals indicate that the Gilbert-Gauss,
Gauss-Matuyama, and Matuyama-Brunhes chron boundaries
can be recognized, as well as the Olduvai subchron. The earliest
reversal found in the sequences on the Pacific coast of Asia, and
the least pronounced, occurs in the volcano-sedimentary sequences of the Kamchatka region in the upper part of the
Schapinsk suite (Pevzner, 1972), which according to paleomagnetic evidence dates within the Gilbert and Gauss epochs.
According to Shantser (in Alekseev et al., 1979), a volcanogenic
layer within the suite has a K/Ar age of approximately 3.8 Ma.
The Gauss-Matuyama boundary and a normally magnetized
zone corresponding to the Olduvai event are recognized in the
marine sequence of Kamchatka. In the lower portion of the
Olkhovian sequence, the boundary between Neogene and
Quaternary at the top of the Olduvai subchronozone is very well
characterized in terms of marine micropaleontology, with clear
changes in the subarctic North Pacific diatom floras and
foraminiferal faunas (Gladenkov, 1994; Gladenkov et al., in
press). In the continental sequence, the Olduvai paleomagnetic
zone is recorded at the boundary between the Schapinsk suite
and the Tumrosk volcanogenic complex.
The stratigraphic level corresponding to the transition from
the reversed polarity of the Matuyama epoch to the normal
polarity of the Brunhes epoch is recorded in the lower part of the
shield and stratovolcanic lava sequence overlying the Tumrosk
volcanogenic complex. It may also be recorded in the sequences
of alluvial and lacustrine deposits of the upper Olkhovian suite of

Kamchatka. In the Karagian marine deposits, the MatuyamaBrunhes boundary lies in the lower part of the arctic-boreal
molluscan complex.
Priamuriel Primorie
In the Pacific margin or far-eastern region of Russia, levels with
the age of the Olduvai and Matuyama-Brunhes paleomagnetic
reversals are clearly recognized. The oldest Upper Cenozoic
stratigraphic level in the region is found in the EvoronChukchagyr depression, where the Pliocene is recognized by a
reduction of the Turgai paleofloral elements in the pollen
complex and an increase of boreal elements. In the Primorie
province, that level is established at the top of the lower subsuite of the Suifun suite. The upper, normally magnetized part of
the Suifun suite is referred to the Gauss paleomagnetic epoch.
In the Priamurie and Primorie territories, the magnetostratigraphic horizon corresponding to the Olduvai event is in the
base or lower part of the reversely magnetized, red-colored layer
that overlies the Suifun suite. The red layer corresponds in
general to the Matuyama paleomagnetic chron, and in the area
of Spassk-Dalny the first considerable cooling is recorded
approximately at the level of the Olduvai event. A more
pronounced cooling is recorded close to the transition between
reversely magnetized red-colored strata and overlying normally
magnetized brown-colored beds, marking the transition from the
Matuyama to the Brunhes epoch.
References
Alekseev, M. N., Golubeva, L. V., Karaulova, L. P., Petrov,

O. M., and Shantser, A. Y. 1979. Problem of the NeogeneQuaternary Boundary on the Pacific Coastal Area of Asia.
In XIV Pacific Science Congress, Abstracts of papers, vol.
B-III(2), ed. N. A. Shilo, pp. 7-9. Moscow: VINITI.
Baranova, Yu. P., Il'Jinskaja, I. A., Nikitin, V. P., Pneva, G. P.,
Fradkina, A. F., and Schvareva, N. Ya. 1976. The Miocene
of the Mamontova Gora: stratigraphy and paleoflora (in
Russian). Moscow: Nauka.
Dodonov, A. Y. 1980. Principles of stratigraphic subdivision of
Upper Pliocene to Quaternary deposits of Tajikistan (in
Russian). In Proceedings of the Second Symposium on the
NeogeneI Quaternary Boundary, USSR, 1977 (in Russian),
ed. K. V. Nikiforova and A. Y. Dodonov, pp. 12-31.
Moscow: Akademia Nauk.
Gladenkov, A. Yu. 1994. Diatom assemblages from the
Pliocene-Pleistocene boundary beds in Kamchatka, Russia. Micropaleontology 40:79-94.
Gladenkov, Yu. B., Basilian, A. E., Bylinskaya, M. E., and
Gladenkov, A. Yu. (in press). Biota of transitional
Pliocene-Pleistocene layers of Kamchatka Region (diatoms, mollusca, foraminifera) (in Russian). Stratigraph.
Geol. Correl.
Gnibidenko, Z. N., Erbayeva, M. A., and Popelova, G. A.
1976. Paleomagnetism and biostratigraphy of Upper
Cenozoic deposits in western Transbaikalia (in Russian).
In Paleomagnetism of Mesozoic and Cenozoic of Siberia
and the Far East (in Russian), ed. E. E. Fatiadi, pp. 75-95.
Asian Russia and Tadjikistan
Novosibirsk: Institute of Geology and Geophysics, Siberian Branch of USSR Academy of Sciences.
Logachev, N. A. 1982. Baikal Region. Guidebook for excursion
A-13, C-13, XIINQUA Congress. Moscow: Nauka.
Minjuk, P. S. 1982. Paleomagnetic studies of the upper part of
section of the Mamontova Gora on the Aldan River (in
Russian). In Geology of the Cenozoic of Yakutia (in
Russian), ed. A. F. Fradkina, pp. 22-27. Yakutsk: Siberian
Branch Academy of Sciences, Yakut Department.
Pevzner, M. A. 1972. Paleomagnetism and stratigraphy of
Pliocene-Quaternary deposits of Kamchatka (in Russian).
Moscow: Nauka.
231
Sher, A. V. 1987. Olyorian land mammal age of northeastern

Siberia. Palaeontogr. Ital. 74:97-112.
Vangengeim, E. A. 1977. Paleontologic foundation of the
Anthropogene stratigraphy of Northern Asia: on mammals
(in Russian). Moscow: Nauka.
Volkova, V. S., and Baranova, Yu. P. 1980. The Pliocene-Early
Pleistocene climatic changes in North Asia (in Russian).
Geol. Geophys. 7(247):43-52.
Zazhigin, V. S. 1980. Late Pliocene and Anthropogene rodents of
southern Western Siberia (in Russian). Trudy Inst. Geol.,
no. 339. Moscow: Akad. Nauk SSSR.
23
The Pliocene-Pleistocene boundary in the Indian subcontinent

M. V. A. SASTRY
Introduction
In 1973, the Geological Survey of India proposed a project on

the Tertiary-Quaternary boundary to cover the global perspective of the boundary between the Pliocene and Pleistocene. This
proposal was merged with the "Neogene/Quaternary Boundary"
project initiated at about the same time by the Geological
Institute of the USSR Academy of Sciences, thus giving rise to
IGCP Project 41 (Nikiforova and Alekseev, Chapter 1, this
volume).
The Indian national working group for IGCP-41 was formed in
1974 with M. V. A. Sastry as convenor and (from March 1981)
with A. Ranga Rao as co-convenor. Members were as follows:
B. S. Towari, Panjab University; V. V. Sastri, Oil and Natural
Gas Commission, Dehra Dun; D. Niyogi, Indian Institute of
Technology, Kharagpur; S. N. Rajaguru, Deccan College, Pune;
M. S. Srinivasan, Banaras Hindu University; K. N. Prasad, K. K.
Verma, and A. K. Dutta, Geological Survey of India. The
working group functioned under the guidance of the director
general of the Geological Survey of India, who was also the
chairman of the Indian National Committee for the IGCP. A
major contribution was the 1979 field conference on the
Neogene-Quaternary boundary as it related to the Siwalik and
the Karewa deposits of India, which stimulated much new work
(Sastry et al., 1981). The present report contains the results of
progress since that field conference.
Neogene and Quaternary deposits are widely exposed in the
foothills of the Himalaya, in the Vale of Kashmir, in the coastal
regions of the Indian peninsula, and in the Andaman Islands.
While the Himalayan foothills, Kashmir, and peninsular India
have continental deposits, the Andaman Islands preserve a
good sequence of deep-water marine facies. Isolated shallowwater marine outcrops of latest Cenozoic age are also found
along the coasts of Kutch-Saurastra, Kerala, and Coromandel
(Figure 23.1).
Andaman-Nicobar Islands
An almost continuous sequence of Neogene and Quaternary

deep-water marine sediments is exposed in the Andaman group
232
of islands, in the Bay of Bengal some 1,300 km away from the

mainland of India. Good sequences can be found in many of the
islands, but the one on the west coast of Neil Island is relatively
accessible and contains a rich and well-preserved foraminiferal
fauna that has been taken as a reference section for the
Neogene-Quaternary boundary (Srinivasan, 1981).
The late Pliocene and early Pleistocene are identified as the
Taipian and Shompenian stages, respectively, based on occurrences of planktonic foraminifera. The Taipian consists of a 15m-thick section of dark gray, highly calcareous, silty mudstone.
A rich assemblage of planktonic foraminifera in the Globorotalia
tosaensis tenuitheca zone, equivalent to zone N.21 of Banner and
Blow (1965), has been recovered from these beds.
Pleistocene strata of the Shompenian Stage overlie the
Taipian, with a possible disconformity, and consist of about 45 m
of moderately hard, buff-colored, lithic grainstone made up of
fairly well-sorted fragments of foraminifera, corals, and algae,
mixed with angular quartz grains. The microfaunal assemblage is
comparatively poor, but is distinguished from the underlying
Taipian by the presence of the initial phylogenetic form of
Globorotalia truncatulinoides. This horizon is taken to delineate
the Pliocene-Pleistocene boundary in the Andaman region.
Figure 23.2 shows foraminiferal ranges of the sequence.
No serious attempts had been made to study the nannofossils,
diatoms, and radiolaria from these horizons when this chapter
was prepared, and data on palaeomagnetism and radiometric
ages were not available.
The Karewa of Kashmir
North of the Siwaliks, and separated from them by the Pir Panjal
Range, the Kashmir Valley contains a thick pile of sediments of
lacustrine, glacial, and fluvial origin capped by loessic strata.
Those sediments are known as the "Karewa" and are divisible
into lower and upper formations.
The Lower Karewa consists of faulted and folded beds
composed of clay, shale, sands, boulder conglomerates, and
lignite beds, with plant and vertebrate fossils throughout. The
Upper Karewa, on the other hand, has thick horizontal beds of
233
Plio-Pleistocene of the Indian subcontinent
SIGNIFICANT
FORAHINIFERA
LATE PLIOCENE
TAIPIAN STAGE
EARLY PLEISTOCENE
SHOMPENIAN STAGE
GLOBOROTALIA TRUNCATULINOIDES
^NEOCENE/QUATERNARY DEPOSITS
G.TOSAENSIS
TENUITHECA
G.TOSAENSIS-TRUNCATULINOIDES PLEXUS
G CRASSAFORMIS
GLOBIGERINA OECORAPERTA
NEOGLOBOOUADRINA OUTERTREI ANDAMANICA
N. DUTERTREI DUTERTREI
GLOBIGERINOIDES OBLIQUES S L .
G ..FISTULOSUS
SPHAEROIDINELLOPSIS SPP.
GLOBOQUADRINA
ISLAND
i
ALTISP1RA
Figure 23.2. Ranges of some significant Plio-Pleistocene foraminifera in

the Andaman-Nicobar Islands. (Adapted from Srinivasan, 1981.)
0%OT BLAIR
Pinjor boundary in the Siwaliks. On the other hand, Roy (1975)

preferred to place the Pliocene-Pleistocene boundary at a higher
level, between Lower Karewa (Hirpur) and Upper Karewa
(Nagum) formations, based on evidence from studies of
tectonism and fresh-water diatoms.
Figure 23.1. Neogene and Quaternary deposits in India. The PliocenePaleomagnetic studies were carried out on samples from the
Pleistocene boundary sequences discussed in the text are (a) Neil Island,
Hirpur and Nagum formations by Agrawal et al. (1981). Three
in the Andaman-Nicobar archipelago, with deep-water marine sediments, (b) Kashmir, with lacustrine, glacial, and fluvial sediments, and
distinctive conglomerate horizons within the Hirpur Formation
(c) Pinjor (fresh-water molasse of the Siwaliks).
served as markers to control paleomagnetic sampling of intervening muds, sands, and silts (Figure 23.3). The paleomagnetic data
indicate that the Nagum strata, above the Hirpur, are in the
gravel, calcareous clay with marl bands, and loamy clay, with Brunhes normal-polarity chron. Reversed polarities of the
Matuyama chron were detected in a 10-m-thick deposit below
very few fossils.
Lydekker (1883) equated the Karewa with the Upper Siwalik Conglomerate III, and normal polarities indicative of the Gauss
(discussed later) and assigned it a Pliocene age. De Terra and chron were found in sand and mud deposits below and above
Paterson (1939) recognized four major glacial advances in the Conglomerate II. The Gilbert re versed-polarity chron is evident
Karewa. They correlated the occurrence of Elephas hysudricus above Conglomerate I. According to those authors, the
in the Lower Karewa with the Pinjor of the Siwaliks and included Pliocene-Pleistocene boundary, taken at the top of the Olduvai
it in the Pleistocene. Wadia (1951) argued that the basal beds of subchron, can be recognized between Conglomerates III and II
the Lower Karewa might extend down to "Pontian" (i.e., Upper in the Hirpur Formation (Figure 23.3).
Miocene) levels. On the basis of a diatom study, Roy (1975)
From the foregoing it is evident that the Pliocene-Pleistocene
thought that the Lower Karewa could be Mio-Pliocene, and boundary lies within the Hirpur Formation significantly above the
Bhatt and Chatterji (1981) preferred to assign a Plio-Pleistocene level where Bhatt and Chatterji (1981) put it (i.e., equivalent to
age to the entire Karewa deposits.
Conglomerate II), and well below the Hirpur-Nagum boundary
The Karewa is divided into two parts, the lower termed the where Roy (1975) estimated it. However, the paleomagnetic
Hirpur Formation and the upper termed the Nagum Formation. sampling was widely spaced, and the studies are still preliminary,
Elephas hysudricus and Equus sivalensis occur within the Hirpur so the exact position of the Neogene-Quaternary boundary in the
Formation, at Sombur and Shopian (Bhatt and Chatterji, 1981). Karewa cannot be stated with precision.
Those authors recognized three zones in the Hirpur Formation
and placed the Neogene-Quaternary boundary immediately
The Siwaliks
above the conglomerate in their zone 2, based on the evidence of
vertebrate fossils and on the premise that the Neogene- In the foothills of the Himalaya in northwestern India and on the
Quaternary boundary should be isochronous with the Tatrot- Potwar Plateau in Pakistan there are thick, fresh-water molasse
I N D I A N
O C E A N
M. V. A. Sastry
234
coarser from the lower to the upper parts of the sequence. The
following is the broad classification of the Siwalik Group
(Pilgrim, 1913; Acharyya, Dutta, and Sastry, 1979):
Upper Siwalik sub-group
Middle Siwalik sub-group

Lower Siwalik sub-group
MUD
Boulder Conglomerate
Formation
Pinj or Formation
Tatrot Formation
Dhok Pathan Formation
Nagri Formation
Chinji Formation
Kamlial Formation
The Potwar Plateau has well-developed Siwalik deposits, and the

type sections of Kamlial, Chinji, Nagri, Dhok Pathan, and Tatrot
are described from that area, whereas the Upper Siwalik beds
are better developed in India. At the Pinjor type section, an
excellent sequence of Tatrot, Pinjor, and Boulder Conglomerate
formations is exposed. Good sections are available east of
Chandigarh, Masol (Tatrot-Pinjor), and Nadah (Pinjor-Lower
Boulder Conglomerate) (Figure 23.4).
SAND =
"MUD
Early studies
SAND
<
a:
a.
MUDS
BASEMENT
Figure 23.3. Schematic lithologic section of the Karewa. The PliocenePleistocene boundary is surmised to lie between the Hirpur and Nagum
formations (i.e., below the Conglomerate III horizon). (Adapted from
Agrawal et al., 1981.)
deposits up to 5,000 m thick, known as the Siwaliks. The Siwalks

are well known for their continuity and their abundant and
diversified mammalian faunas, spanning a period from the
Middle Miocene to the Lower Pleistocene. In terms of lithology,
the beds show great variation laterally and become progressively
The continuity of the record of fossils and environment in the

Siwaliks led earlier workers to undertake detailed studies in
mammalian biostratigraphy. In addition to interpretations of
stratigraphy and climatic changes based on fossils, research on
the Pliocene-Pleistocene boundary was also part of these
studies.
Pilgrim's 1913 classification of the Siwaliks became the basis
for later workers. He compared the faunas with those of Europe,
taking the Pontian assemblage to be transitional between
Miocene and Pliocene. Based on the occurrence of Elephas
planifrons, he assigned a Middle Pliocene age to the Pinjor and
considered both the Upper Pinjor and the overlying Boulder
Conglomerate zones to be uppermost Pliocene. Earlier,
Lydekker (1883), on the evidence of Equus namadicus and
Elephas cf. E. namadicus, had suggested that the Boulder
Conglomerate zone was Lower Pleistocene.
Matthew (1929) advocated a younger age, assigning Lower
Pleistocene to the Upper Siwalik and Middle Pliocene to the
Dhok Pathan. He considered that the mammal fauna recorded a
time break equivalent to the Upper Pliocene between the Dhok
Pathan and the Tatrot. This was supported by De Terra and
Paterson (1939) on the evidence of conglomerate at the base of
the Tatrot, indicating a period of erosion. The apparent
coincidence of those two lines of evidences gave strong support
to the idea that the base of the Upper Siwaliks should be
equivalent to the base of the Pleistocene. De Terra and Teilhard
de Chardin (1936), Lewis (1937), Colbert (1935, 1942), Paterson
(1941), and Movius (1944) emphasized the occurrence in the
Tatrot of Archidiskodon (i.e., Mammuthus, according to Aguirre
et al., Chapter 9, this volume) and advocated from this that the
Tatrot should be assigned a place within the Pleistocene. That
was in line with Haug's proposition (1911) that the Pleistocene
K>
ISKm
was characterized by the first appearances of Elephas (as

primitive forms that were later redesignated as Archidiskodon,
and now Mammuthus), Leptobos, and Equus simultaneously on
several continents. Hooijer and Colbert (1951) carried out a
statistical analysis of the fauna on the premise that the Tatrot and
Pinjor faunas could be correlated to a Villafranchian age
between the Pliocene Dhok Pathan and Pleistocene Pinjor.
Pilgrim (1944) maintained that the Tatrot and Pinjor rocks
were equivalent to Astian and Villafranchian, respectively, of
Europe. He included the Pinjor beds in the Pliocene because of
the fact that Elephas and Bos had definite origins in or near
India, and their appearance in India should therefore have taken
place in the Pliocene, much earlier than in Europe. Because of
the presence of Hipparion in the Pinjor beds, the close
relationship between Pinjor and Villafranchian faunas, and the
fact that the glacial formation of the Bain Boulder Bed overlies
the Pinjor, Pilgrim concluded that the Pliocene-Pleistocene
boundary should also lie above the Pinjor.
Wadia (1951) favored Pilgrim's view that glaciation had begun
earlier in Europe than in Asia and that basing the advent of the
Pleistocene on the first glaciation was not important. However,
he advocated a transition from the pre-glacial Pinjor to the
glacial Boulder Conglomerate zone. Gill (1951) suggested the
existence of a regional unconformity in northwestern India
representing a Siwalik phase of orogeny that commenced in postPinjor times. He suggested that glaciation began earlier than the
Siwalik orogeny and the uplift of the Pir Panjal Range.
235
Figure 23.4. Upper Siwalik area. Pliocene and

Pleistocene strata are best exposed between Pinjor
and Nadah, east of Chandigarh.
Later studies
Sahni and Khan (1968) recorded Leptobos from the Tatrot and
indicated that the first appearances of that genus and Archidiskodon in India marked the end of the Pliocene, because
those genera were indigenous to India. Further, the immigrations of Equus, Bubalus, Hipselephas, and Rhinoceros indicated, according to them, the beginning of the Pleistocene in
India. They equated the Tatrot with the Upper Pliocene
(Astian) and the Pinjor with the Lower Pleistocene (Villafranchian). Prasad (1974) showed that of 26 vertebrate genera
present in the Tatrot, 8 were holdovers from the Dhok Pathan,
9 were newcomers, and the remaining 9 persisted into the
Pinjor. From the faunal data, he argued that the Tatrot might
be the end of the Pliocene.
Balasundaram and Sastry (1972) contended that the Pinjor
was of Pliocene age, based on tectonic, climatic, and paleontological evidence. Sastry and Dutta (1977a,b) observed that the
upper part of the Pinjor is conspicuously marked by the
simultaneous development of rhythmic alternations in the ratios
of pebbles to sand and silt and the absence of fossils. That
horizon, according to them, represents drastic changes in
climate, life, and environment, related to the initiation of a final
phase of Himalayan orogeny, and is the appropriate location for
the Neogene-Quaternary boundary.
A systematic regional mapping of the Siwaliks, coupled with
lithological and heavy-mineral analysis by the ONGC (Oil and
236
M. V. A. Sastry
Natural Gas Commission), has shown that a boundary drawn on

the basis of any one criterion does not precisely coincide with
boundaries drawn on other criteria. Using the first appearance of
Equus in the Pin j or as an index to the Pleistocene, Ranga Rao et
al. (1981) drew the Pliocene-Pleistocene boundary between the
Tatrot and the Pinjor.
As part of IGCP Project 41, an international field conference
on the Neogene-Quaternary boundary in the Siwaliks was
organized in 1979 (as noted earlier), in which leading workers
on the subject took part and examined important sections. It
was not possible, however, to arrive at definitive results
regarding the placement of the boundary, and the conference
group concluded that further paleomagnetic and radiometric
studies were needed.
Based on paleomagnetic studies, M. N. Alekseev (personal
communication) suggested that the Tatrot is likely to correspond
to the Gauss normal-polarity chron, although reversed remanent
magnetization in its upper part correlates to the early part of the
Matuyama chron. The uppermost part of the Pinjor has been
found to be normally magnetized, indicating the Olduvai
subchron (Sastry and Dutta, 1977a,b). Yokoyama's (1981)
paleomagnetic results (following vertebrate fossil analysis) agree
that the Olduvai event is reflected by a normally polarized zone
within the Pinjor.
Azzaroli and Napoleone (1982) measured a well-developed
300-m section at Nadah exposing the Pinjor and Lower Boulder
Conglomerate formations. The base of the Pinjor is not seen in
the Nadah section, but is presumed to be 4 m below the base of
the exposed section, beneath river terraces. The sedimentation
in this section suddenly became conglomeratic during the
Jaramillo normal-polarity event, indicating renewed erosional
activity in the Himalayan belt, coeval with significant environmental changes in Europe related to the Mindel glacial maximum. The bulk of the Siwalik fauna disappeared between the
Olduvai and Jaramillo events, but those authors concluded that
the top of the Olduvai event was closely equivalent to the contact
of the Pinjor and the Boulder Conglomerate.
Thus, differing opinions regarding the basic criteria and the
emphasis to be placed on certain groups of animals in deciphering the Neogene-Quaternary boundary have resulted in at least
three different placements: (1) at the base of the Tatrot
(Matthew, 1929; Colbert, 1935; Lewis, 1937; De Terra and
Paterson, 1939; Hooijer and Colbert, 1951); (2) at the base of the
Pinjor (Sahni and Khan, 1968; Prasad, 1974; Badam, 1979; and
many others); and (3) at the top of the Pinjor (Pilgrim, 1944;
Gill, 1951; Balasundaram and Sastry, 1972; Sastry and Dutta,
1977a; Azzaroli and Napoleone, 1982). It should be noted that
the advocates of placements (1) and (2), as well as early
advocates of (3), were employing criteria other than the
recommendation of the 1948 London IGC committee, which was
to refer the boundary to a physical reference point at the base of
the Calabrian Stage in recognition of a change to colder climate
conditions in the Italian marine sequences, at a level that
subsequently was determined to be close to the top of the
Olduvai event.
Recent studies in Pakistan
On the Potwar Plateau, a multidisciplinary study, including

biostratigraphy, paleomagnetism, and radiometric dating, has
been conducted by several teams of workers. Opdyke et al.
(1979) investigated eight separate stratigraphic sections. Two
prominent bentonitized tuffs were also recorded in some
sections, and radiometric dates obtained by fission-track analysis
on zircons from those tuffs have enabled correlations to the
standard GPTS (geomagnetic polarity time scale) of the Pliocene
and Pleistocene (Cande and Kent, 1995).
The important findings include the observation that the
boundary between the Tatrot and the Pinjor, defined faunally as
the first simultaneous occurrences of Equus-Elephas-Bos (sensu
Haug, 1911) and cervids with antlers, should be dated to about
2.5 Ma, and thus is very near the Gauss-Matuyama paleomagnetic boundary (2.6 Ma). Other findings were that the Pinjor
faunal zone extends to the base of the Brunhes chron, contra
Azzaroli and Napoleone (1982), and that the PliocenePleistocene boundary (top of the Olduvai event) is therefore
located within the Pinjor faunal zone (although possibly not
within the lithologic limits of the Pinjor Formation at the type
Tatrot section). The fossil mammal faunas do not show any
change, except for the extinction of the giraffid Sivatherium, at
the level of that boundary in the Potwar area (Figure 23.5).
More recently, Barry, Lindsay, and Jacobs (1982), in their
biostratigraphic zonation of the Siwaliks on the Potwar Plateau,
proposed an Elephas planifrons interval-zone and recorded
within it several characteristic mammalian taxa such as Elephas
planifrons, Stegodon sp., Equus sivalensis, Hexaprotodon
sivalensis, Proamphibos lachrymas, and cervids. The magneticpolarity correlations indicate that the top of the zone is very close
to the Pliocene-Pleistocene boundary level. Bentonitized tuffs in
the Campbellpore area have been dated to 1.61 Ma (Johnson et
al., 1982), while the paleomagnetic transition from normal to
reversed polarity below the tuff is recognized to be the reversal
that marks the top of the Olduvai subchron, whose inferred age
is about 1.8 Ma (Cande and Kent, 1995; Pasini and Colalongo,
In the trans-Indus region, the Marwat Formation is rich in the
fossils of typical Pinjor assemblages and has a well-developed
lithostratigraphic unit, the "Bain Boulder Bed," whose glacial
origin is debated (West, 1981). The boulder bed, however, is
equated with the Olduvai event and also marks an important
point in the tectonofluvial history of the region.
Analysis
The Tatrot-Pinjor boundary is based essentially on faunal
transitions, and the Pliocene-Pleistocene boundary has been
estimated at that level by several workers. Investigations in
Pakistan by Opdyke et al. (1979), Azzaroli and Napoleone
(1982), Barry et al. (1982), and Johnson et al. (1982), however,
have shown that the Tatrot-Pinjor boundary is correctable with
the Gauss-Matuyama paleomagnetic boundary at about 2.6 Ma.
UJ
rity
300
o
"o
0.
Epoc
>
+90
fy
.90
S Pol
NADAH
237
UJ
X
z
NADAH
LR
BOULDER
200
F?i
CD
Jaramillo
CONGLI
P
I
N
.100
a:
0-73
1-67
Olduvai
1-87
201
2-12 Rtunion
0
R
2-48.
2 92 Katna
_0m
TATROT
If we accept that the top of the Olduvai event is the indicator for
the Pliocene-Pleistocene boundary, the Dhok Pathan-Tatrot
and the Tatrot-Pinjor concepts become irrelevant.
The change to the Pinjor fauna from that of the underlying
Tatrot is not significant either in variety or in abundance, but on
the other hand the rich Pinjor fauna suddenly disappears at the
level of the Boulder Conglomerate. The ratio of pebbles to sand
and silt in the sediments and the rates of deposition continued
without much change from the Tatrot to the Pinjor, but, again,
not into the Boulder Conglomerate. Thus the continuity of the
fauna and the similarity of depositional and climatic conditions
support a close relationship between the Tatrot and the Pinjor.
At the end of the Pinjor, the change in sedimentation suggests
a sudden increase in erosion due to a combination of tectonism in
the adjoining mountains, and also possibly some climate change,
at about 1.0 Ma (Jaramillo), if not 0.7 Ma or later (lower
Brunhes). The high-energy sedimentary environment was much
less favorable for the preservation of fossils, and in fact fossils
are rare if not absent in the Lower Boulder Conglomerate, which
may account for much of the apparent faunal change at that
level. The marked changes at the Pinjor-Boulder Conglomerate
contact are unrelated to the top of the Olduvai subchronozone
Figure 23.5. Lithology and

paleomagnetic polarity in the
Nadah section. (Adapted from
Azzaroli and Napoleone, 1982.)
and should not be used to mark the Pliocene-Pleistocene

boundary.
In the Indian region, nevertheless, further systematic collecting, coupled with detailed paleomagnetic and possibly radiometric investigations, will be necessary before a final conclusion
can be reached on the location of the Pliocene-Pleistocene
boundary in the continental Siwalik deposits. It is not possible at
this stage to establish a local reference or parastratotype to mark
the Neogene-Quaternary boundary in those deposits.
References
Acharyya, S. K., Dutta, A. K., and Sastry, M. V. A. 1979.
Siwalik stratigraphy and its bearing on the Main Boundary
Fault. Geol. Surv. India, Misc. Publ 41(l):67-79.
Agrawal, D. P., Bhatt, D. K., Sheela Kusumgar, and Pant, R. K.
1981. The Neogene/Quaternary boundary in India: a
review. Indian Acad. Sci., Proc. (Earth Planet. Sci.)
90:111-123.
Azzaroli, A., and Napoleone, G. 1982. Magnetostratigraphic
investigation of the Upper Siwaliks near the Pinjor, India.
Riv. Ital. Paleontol. 87:739-762.
Badam, G. L. 1979. Pleistocene fauna of India. Pune: Deccan
College.
238
M. V. A. Sastry
Balasundaram, M. S., and Sastry, M. V. A. 1972. Plio- Opdyke, N. D., Lindsay, E., Johnson, G. D., Johnson, N.,
Tahirkheli, R. A. K., and Mirza, M. A. 1979. Magnetic
Pleistocene boundary in sediments of Indian Subcontipolarity stratigraphy and vertebrate paleontology of the
nent. Geol. Surv. India, Rec. 107:54-72.
Upper Siwalik Subgroups of Northern Pakistan. PaleoBanner, R T., and Blow, W. H. 1965. Progress in the Planktonic
geogr. Paleoclimatol. Paleoecol. 27:1-34.
foraminiferal biostratigraphy of the Neogene. Nature
208:1164-1166.
Paterson, T. T. 1941. On a world correlation of the Pleistocene.
Roy. Soc. Edinb., Trans. 49:373-422.
Barry, J. C , Lindsay, E. H., and Jacobs, L. L. 1982. A biostratigraphic zonation of the middle and Upper Siwaliks of Pilgrim, G. E. 1913. Correlation of Siwaliks with mammal
the Potwar Plateau of Northern Pakistan. Palaeogeogr.
horizons of Europe. Geol. Surv. India, Rec. 43:264-326.
Pilgrim, G. E. 1944. The lower limit of the Pleistocene in Europe
and Asia. Geol. Mag. 81:28-30.
Bhatt, D. K., and Chatterji, A. K. 1981. A recent analysis of
Neogene/Quaternary transition in the Kashmir Region. In Prasad, K. N. 1974. Vertebrate fauna from Perim Island,
Proceedings of the N/Q Boundary Field Conference, India,
Gujarat. Paleontol. Ind., n.s., 41.
1979, ed. M. V. A. Sastry et al., pp. 11-14. Calcutta: Ranga Rao, A., Khan, K. N., Venkatachala, B. S., and Sastri,
Geological Survey of India.
V. V. 1981. Neogene/Quaternary boundary and the
Siwalik. In Proceedings of the N/Q Boundary Field
Conference, India, 1979, ed. M. V. A. Sastry et al., pp.
geomagneic polarity time scale for the Late Cretaceous
131-142. Calcutta: Geological Survey of India.
Colbert, E. H. 1935. Siwalik mammals in the American Museum Roy, D. K. 1975. Stratigraphy and paleontology of the Karewa
of Natural History. Am. Phil. Soc, Trans., n.s. 36:1-401.
Group of Kashmir. Geol. Surv. India, Misc. Publ. 24:204221.
Colbert, E. H. 1942. The geological succession of the
Proboscidea. In The Proboscidea, vol. 2, ed. H. F. Osborn. Sahni, M. R., and Khan, E. 1968. Boundary between the Tatrot
New York: American Museum of Natural History.
and Pinjaur. Paleontol. Soc. India, J. 5:29-30.
De Terra, H., and Teilhard de Chardin, T. 1936. Observation on Sastry, M. V. A., and Dutta, A. K. 1977a. Review on Neogene/
Quaternary boundary in India. Giorn. Geol., ser. 2
the Upper Siwalik formation and later Pleistocene deposits
61:331-340.
in India. Am. Phil. Soc, Proc. 76:791-822.
De Terra, H., and Paterson, T. T. 1939. Studies on the Ice Age in Sastry, M. V. A., and Dutta, A. K. 1977b. Neogene/Quaternary
boundary in the Siwalik. Paleontol. Soc. India, J. 20:320India and associated human cultures. Carnegie Inst. Wash.,
326.
Proc. 493:1-354.
Gill, W. D. 1951. The stratigraphy of the Siwalik Series in the Sastry, M. V. A., Kurien, T. K., Dutta, A. K., and Biswas, S.
(eds.) 1981. Proceedings of the N/Q Boundary Field
northern Potwar, Pakistan. Geol. Soc. London, Quart. J.
Conference, India, 1979. Calcutta: Geological Survey of
107:375-394.
India.
Haug, E. 1911. Traite de Geologie, II: les periodes geologique.
Srinivasan, M. S. 1981. The Neogene/Quaternary boundary in
Paris: Mouton.
the marine sequences of Andaman-Nicobar Islands, NorthHooijer, D. A., and Colbert, E. H. 1951. A note on the
ern Indian Ocean. In Proceedings of the N/Q Boundary
Pliocene-Pleistocene boundary in the Siwalik Series of
Field Conference, India, 1979, ed. M. V. A. Sastry et al.,
India. Am. J. Sci. 244:533-538.
pp. 169-175. Calcutta: Geological Survey of India.
Johnson, G. D., Zeitler, P., Naeser, C. W., Johnson, N.,
Summers, D. M., Frost, C. D., Opdyke, N. D., and Wadia, D. N. 1951. The transitional passage of Pliocene into the
Tahirkheli, R. A. K. 1982. The occurrence and fission
Pleistocene in the north-western Himalayas. In XV111
track ages of late Neogene and Quaternary volcanic
International Geological Congress, Reports, vol. 11(K),
sediments, Siwalik Group, northern Pakistan. Palaeopp. 43-48.
geogr. Palaeoclimatol. Palaeoecol. 37:63-93.
West, R. M. 1981. Plio-Pleistocene fossil vertebrates and
biostratigraphy, Bhittanni and Marwat Ranges, NorthLewis, G. E. 1937. A new Siwalik correlation. Am. J. Sci.
west Pakistan. In Proceedings of the N/Q Boundary Field
33:191-204.
Conference, India, 1979, ed. M. V. A. Sastry et al., pp.
Lydekker, R. 1883. The geology of the Kashmir and Chamba
211-215. Calcutta: Geological Survey of India.
territories and the British district of Khagan. Geol. Serv.
India, Mem. 22:1-344.
Yokoyama, T. 1981. Paleomagnetic study of the Tatrot and the
Matthew, W. D. 1929. Critical observation upon Siwalik mamPin j or Formations, upper Siwaliks, east of Chandigarh,
mals. Am. Mus. Nat. Hist., Bull. 56:437-560.
north-west India. In Proceedings of the N/Q Boundary
Field Conference, India, 1979, ed. M. V. A. Sastry et al.,
Movius, H. L. 1944. Early man and Pleistocene stratigraphy in
southern and eastern Asia. Harvard Univ., Peabody Mus.
Papers 10(3): 1-113.
24
The Pliocene-Pleistocene boundary in Japan: the Osaka Group,

Kinki district
MINORU ITIHARA, SHUSAKU YOSHIKAWA, and TADAO KAMEI
Introduction
sedimentary cycles. The Osaka Group is well exposed in hilly

areas around Osaka, Kyoto, and the Nara and Harima basins, in
the central part of the Kinki district. The sequence, 200-1,500 m
or more thick, unconformably overlies Miocene and pre-Tertiary
basement rocks and is discordantly overlain by terrace deposits.
The Osaka Group has been folded and faulted, with maximum
fault throws estimated to be 200 m to more than 500 m. The sum
of the displacements that have taken place in these basins since
Pliocene time is called the Rokko movement (Ikebe, 1956).
The Osaka Group is divisible by means of key tuff layers, 3-4
m in maximum thickness, and the marine clay beds, 10 m or so in
maximum thickness (Itihara, 1961; Itihara et al., 1975) (Figure
24.2). Among these key horizons with regional significance, the
most informative are as follows: the Fukuda, Yellow, Pink,
Azuki, and Sakura tuff layers, and the 12 marine clays identified
as Ma-1, MaO, and Mai to MalO, in ascending stratigraphic
order (Figure 24.2).
The INQUA Subcommission 1-d on the Pliocene-Pleistocene

boundary organized the working group on the NeogeneQuaternary boundary as the action body for Project 41 of the
International Geological Correlation Program (IGCP) in 1974.
As a research and contact body corresponding to the INQUA
subcommission, the Japanese National Committee for Quaternary Research in the Science Council of Japan organized the
Japanese National Subcommission on the Pliocene-Pleistocene
Boundary, chaired by M. Itihara. The Japanese national
subcommission has functioned simultaneously as the Japanese
national working group for IGCP-41 for investigations of the
boundary problem in Japan.
In August 1982, the Japanese national working group for
IGCP-41 presented its third report at the XI INQUA Congress
in Moscow (Itihara and Kuwano, 1982). That report, however,
covered too many areas (Figure 24.1) to be included in this final
volume. Therefore, for this purpose, it was decided to select two
representative studies: this chapter, on the Pliocene-Pleistocene
Biostratigraphy
boundary in the Osaka Group, as a typical section of lacustrine
and fluvial sediments with marine intercalations, and Chapter Plant and animal fossils are abundant in the Osaka Group.
25, on the Plio-Pleistocene stratigraphy in the Boso Peninsula, Among them, the most important are plant megafossils from
many horizons and proboscidean fossils from a rather small
as a typical section of marine sediments.
number of horizons.
Background and stratigraphic framework
Deposits of Pliocene and Pleistocene age are well exposed in

Japan (Figure 24.1). In the Osaka Group in the Kinki district of
central southern Honshu, strata that correlate well to the interval
containing the Pliocene-Pleistocene boundary have been studied
or discussed by Itihara (1961), Ishida et al. (1969), Itihara and
Kamei (1970, 1982), Itihara et al. (1973, 1975, 1984), and
Maenaka et al. (1977). This chapter describes the present status
of the problem of the Pliocene-Pleistocene boundary in this
well-described sequence.
The Plio-Pleistocene Osaka Group (Osaka Group Research
Group, 1951) is composed mainly of unindurated sand, gravel,
and clay of lacustrine and fluvial origin, intercalated with a
number of tuff layers and 12 marine clays that are the top units of
Plant megafossils
The taxonomy of plant megafossils was studied in detail by Miki
(1941a,b, 1948). In a detailed review of the stratigraphy of the
plant-bearing strata, Itihara (1961) concluded that the lowermost
part of the Osaka Group was deposited during a warm-climate
period characterized by dominance of the Metasequoia flora.
Subsequent climatic deterioration led to a decline of the
Metasequoiaflora,which eventually became extinct by the time
of a cold-climate horizon just below the Azuki tuff layer, at
about 0.9 Ma, associated with the Jaramillo subchron (Figure
24.2). Although that was followed by climatic amelioration,
Itihara (1961) suggested that "in the Osaka Group, the beginning
of the age of extinction of the Metasequoia flora, i.e. the
239
Itihara, Yoshikawa, and Kamei
240
45
-40
Shimane
Boso
Saijo
-35
Kuchinotsu
-30
.* ^ O k i n a w a Is.
250
Figure 24.1. Distribution of PlioPleistocene sediments in Japan.
. 4 Miyako Is.
125
130
beginning of climatic deterioration, should be inferred as the

Pliocene-Pleistocene boundary."
The Metasequoiaflorain the Osaka Group is characterized by
an assemblage of Metasequoia, Glyptostrobus, Sequoia, Pinus
koribai, Pinus fujii, Juglans cinerea var. megacinerea, Liquidambar, Ginkgo, Pseudolarix, Keteleeria, Nyssa, and other
elements. As shown in Figure 24.1, the Metasequoiaflorastill
had the essential characters of the Tertiary, even though Nyssa,
Ginkgo, Pseudolarix, and Liquidambar disappeared during its
later stages (Itihara et al., 1975). It can be inferred that climate
fluctuations were relatively feeble during the time of deposition
of the lowermost part of the Osaka Group.
The beginning of significant climatic deterioration is evidenced
in the Osaka Group not only by the extinction of the
135
140
145
Metasequoiaflorain the lower part of the Osaka Group but also

by the successive appearances of Menyanthes trifoliata and the
subalpine species Pinus koraiensis and Picea maximowiczii in the
interval between a horizon below the Kamifukuda tuff layer and
a horizon below the Mai bed (Kokawa, 1959; Itihara, 1961;
Ibaragi Research Group, 1966; Yoshikawa, 1973; Itihara et al.,
1975; Momohara, 1990). Thus, the first indication of that
remarkable floral change, the extinction of the Metasequoia
flora, is seen in an interval between the Fukuda and Kamifukuda
tuff layers, and the last relics of the Metasequoia flora, such as
Metasequoia and Picea koribai, disappear just below the Azuki
tuff layer (i.e., just below the Ma3 marine bed). Those events
are shown in Figure 24.1 as being just above the Olduvai and
Jaramillo subchrons, respectively. After the extinction of the
Nyssa n i
Ginkgo bilob
Ketel
Pseudolanx kaempferi
Liquidambar formosana
PmZs fujii I 7 i i 71
Sequoia êmpervirens i i
Glyptostrobus pensilis i i
Juglansj:inerea var. meg
Metasequoia disticha i i
P/cea Koribai
II
Senriyama Formation
I I
It
G R O U P
8
e
>.3
03
03
Divisions
Chronostr
J...B . l . l U _ L ^
CD
CD
sted
Corr lat ions
242
Itihara, Yoshikawa, and Kamei
Metasequoia flora, a new kind of flora appeared. For instance, S. elephantoides has been revised to Stegodon shinshuensis, a
Juglans mandschurica replaced Juglans cinerea var. megacinerea,species closely related to Stegodon zdanskyi of northern China.
and then was replaced in turn by Juglans sieboldiana (Nirei, At the same time, the two species of stegodont, S. akashiensis
and S. sugiyamai, which characterize the succeeding zones in the
1975).
Climate fluctuations, indicated by alternations of cold-climate lowermost Osaka Group, are both recognized to be conspecific
megafloras and warm-climate megafloras, intensified after the with Stegodon aurorae (Kamei, 1991). Finally, Dubrovo (1981)
disappearance of the Metasequoia flora (i.e., at the time of has maintained her opinion that both Mammuthus paramammondeposition of the upper part of the Osaka Group), implying the teus shigensis and M. armenaicus proximus are synonymous with
beginning of glacial conditions. This is exemplified by occur- Palaeoloxodon naumanni.
rences of Larix gmelini, Pinus koraiensis, and Oxycoccus
According to Kamei, zone 1 is characterized by the presence
palustris in the interval between the Ma6 and Ma7 beds (cold), of elements found in the Gaozhuang fauna of the Yushe Basin in
followed by Syzygium buxifolium and Podocarpus nagi in the northern China, approximately correlatable with zones MN-14
Ma8 bed (warm), then Pinus koraiensis and Picea maximowiczii and MN-15 of the western European scale (Qiu, 1989). The
from the interval between the Ma8 and Ma9 beds (cold), and faunas of zones 2 and 3 are more endemic, with remnants of the
finally Pinus koraiensis from the horizon above the Ma 10 bed zone 1 assemblage, but are enhanced by immigrants found also
(cold). The warm-climate species Paliurus nipponicus and in the Nihewan fauna of northern China, such as Elaphurus,
Sapium sebiferum are associated with the successive marine beds Axis, Rusa, and so forth. Zone 4 contains transitional and
Ma3 through Ma8, supporting the conclusion that high sea levels endemic forms of temperate-forest affinities, including Apodocumented by the marine clay beds in the Osaka Group were demus argenteus. Zone 5 is made up almost equally of extinct
synchronous with warm-climate conditions.
and living taxa, with temperate-forest elements predominant,
but with a few immigrants from warm-temperate faunas, such as
Stegodon orientalis and Rhinoceros sinensis, which are abundant
Pollen
in the Wanhsien fauna of southern China. The giant crocodilian
Tai (1973) subdivided the Osaka Group into the Metasequoia and "Tomistoma" machikanense from just below the Ma8 level
Fagus zones, with the boundary at the base of the Ma3 bed. belongs to the zone 5 assemblage. Another crocodilian is also
According to Tai, the onset of climatic deterioration can be found below the Mai horizon.
detected within the B subzone of the Metasequoia zone,
It is worth notice that the level of occurrence of the deer
especially at the top part of that subzone. It is noticeable that the Elaphurus of the Nihewan fauna in zone 3 corresponds to the
top part of the B subzone nearly corresponds to the transition final extinction of the Metasequoia flora.
between the period of flourishing and the period of reduction
and eventual extinction of Metasequoia flora, as identified by
Magnetostratigraphy
Itihara (1961). The floral succession seen in the palynological
analyses closely parallels that of plant megafossil studies.
Paleomagnetic studies of the Osaka Group by Torii, Yoshikawa,
and Itihara (1974), Maenaka et al. (1977), Maenaka (1983), and
Itihara et al. (1984) are summarized in Figure 24.1. The data
Mammalia
suggest that the Osaka Group represents the time from the
The Osaka Group has long been noted for its fossil mammals Gauss chron to the early part of the Brunhes chron. One subzone
(Ikebe, Chiji, and Ishida, 1966; Kamei and Setoguchi, 1970; of normal paleomagnetic polarity, from the Mai bed to the
Kamei and Otsuka, 1981). Kamei (1984) found that the Osaka Komyoike tuff layer, has been thought to correspond to the
Group and its correlative, the Kobiwako Group, could be Jaramillo subchron. Another subzone of normal polarity, from
subdivided into the following mammalian zones, in ascending about 2 m above the Shimogaito tuff layer to about 5 m above
the Mitsumatsu tuff layer, has been correlated with the Olduvai
order:
subchron.
8. Sus scrofa and Cervus (Sika) nippon zone (Holocene)
7. Mammuthus primigenius zone (latest Pleistocene)
Radiometric dating
6. Palaeoloxodon naumanni zone
5. Stegodon orientalis zone
Radiometric age determinations for the tuff layers of the Osaka
4. Mammuthus paramammonteus shigensis-Mammuthus
Group, mainly using the fission-track method (Nishimura and
armenaicus proximus zone
Sasajima, 1970; Matsuda, 1980), have been reexamined by
3. Stegodon akashiensis zone
Suzuki (1988), with findings that appear to be substantially
2. Stegodon sugiyamai zone
different from the previously reported ages. It appears that on
1. Stegodon cf. S. elephantoides zone (middle Pliocene)
the basis of direct radiometric dating, the age span of the Osaka
All of these zones were founded on mammals from the Osaka Group probably ranges from about 3.0 Ma to 0.3 Ma and that the
Group except for the earliest (zone 1) and latest (zones 7 and 8). age of extinction of the Metasequoiaflora,the level at which the
It should be noted that at present, the taxonomy of Stegodon cf. Pliocene-Pleistocene boundary has been correlated in earlier
Japan: Osaka Group
243
studies (e.g., Itihara, 1961), was in fact about 0.9 Ma. That age is Kamei, T., and Otsuka, H. 1981. The Plio-Pleistocene
stratigraphy of Japan in relation to Proboscidean evolusomewhat younger than the age for the top of the normaltion. In Proceedings of the Field Conference on Neogenepolarity zone identified as the Olduvai subchron, in which the
Quaternary Boundary, India, 1979, ed. M. V. A. Sastry,
Vrica definition is located. It should be noted that the Vrica
horizon appears to correspond fairly closely to the level between Kamei, T., and Setoguchi, T. 1970. Some remarks on mamthe Fukuda and Kamifukuda tuffs at which the Metasequoia flora
malian faunas in the early Pleistocene (in Japanese, with
English abstract). Daiyonki-Kenkyu [Quaternary Refirst began to decline.
search] 9:158-163.
Kokawa, S. 1959. Plant remains around Nishinomiya City and
their floral transition (in Japanese). Nishinomiya-shi-shi
References
[History of Nishinomiya City] 1:265-285.
Maenaka, K. 1983. Magnetostratigraphic study on the Osaka
Dubrovo, I. A. 1981. Die fossilien Elefanten Japans. QuartdrGroup, with special reference to the existence of pre- and
paldontologie 4:49-84.
post-Jaramillo episodes in the late Matuyama Polarity
Epoch. Hanazono Univ., Kenkyu-Kiyo [Research Journal]
Ibaragi Research Group. 1966. The Osaka Group of Fukui
14:1-65.
Area, north of Ibaragi, Osaka Prefecture and occurrence
of Elephas shigensis (in Japanese with English abstract). In Maenaka, K., Yokoyama, T , and Ishida, S. 1977. Paleomagnetic
stratigraphy and biostratigraphy of the Plio-Pleistocene in
Prof. S. Matsushita Memorial Volume, ed. N. Ikebe et al.,
the Kinki District, Japan. Quat. Res. 7:341-362.
pp. 117-130. Kyoto: Kyoto University.
Ikebe, N. 1956. Cenozoic geohistory of Japan. In Proceedings of Matsuda, T. 1980. A fission-track date of the Pumice Tuff bed.
(In Iida, Y, Unconformity of the Early Pleistocene in the
the 8th Pacific Science Congress,, vol. 2, pp. 446-456.
Osaka Group of Sen-nan Area, south of Osaka) (in
Quezon City: University of the Philippines, Department of
Japanese with English abstract). News of Osaka MiGeology.
crop aleontologists 8:5.
Ikebe, N., Chiji, M., and Ishida, S. 1966. Catalogue of the late
Cenozoic Proboscidea in the Kinki District, Japan. Osaka Miki, S. 1941a. On the change of flora in eastern Asia since
Tertiary Period (I). Japan. J. Botany 11:237-303.
City Univ., J. Geosci. 9:47-48.
Ishida, S., Maenaka, L., and Yokoyama, T. 1969. Paleomagnetic Miki, S. 1941b. Floral remains of the conifer age at Manchidani
chronology of volcanic ash of the Plio-Pleistocene series in
near Nishinomiya, Japan. Japan. J. Botany 11:377-383.
Kinki District, Japan. Geol. Soc. Japan, J. 75:183-197.
Miki, S. 1948. Floral remains in Kinki and adjacent districts since
Itihara, M. 1961. Some problems of the Quaternary sedithe Pliocene with description 8 new species (in Japanese,
mentaries in the Osaka and Akashi Areas, Japan. Osaka
English abstract). Kobutsu to Chishitsu [Mineralogy and
City Univ., Inst. Polytech., J, Ser. G 5:13-30.
Geology] 9:105-144.
Itihara, M., and Kamei, T. 1970. The Osaka Group (in Momohara, A. 1990. Plio-Pleistocene plant biostratigraphy of
Japanese). Kagaku [Science] 40:282-291.
the Osaka Group, with reference to extinction of plants (in
Itihara, M., and Kamei, T. 1982. The Pliocene-Pleistocene
Japanese). Assoc. Quat. Res. Japan, Progr. Abstr. Ann.
Boundary in the Osaka Group, Japan. In The Third Report
Mtg. 30:96-97.
on the Pliocene-Pleistocene Boundary in Japan. Japanese Nirei, H. 1975. A classification of fossil walnuts from Japan.
Osaka City Univ., J. Geosci. 19:31-62.
National Working Group of the IGCP Project No. 41
Neogene-Quaternary Boundary, ed. M. Itihara and Y. Nishimura, S., and Sasajima, S. 1970. Fission-track age of
Kuwano, pp. 42-48. Osaka: Osaka City University.
volcanic ash-layers of the Plio-Pleistocene series in Kinki
Itihara, M., and Kuwano, Y. (eds.) 1982. The Third Report on the
District, Japan (in Japanese with English abstract).
Pliocene-Pleistocene Boundary in Japan. Japanese National
Chikyu-Kagaku [Earth Science] 24:222-224.
Working Group of the IGCP Project No. 41 Neogene- Osaka Group Research Group. 1951. The Osaka Group and the
Quaternary Boundary. Osaka: Osaka City University.
related Cenozoic formations (in Japanese). ChikyuItihara, M., Kamei, T., Mitsunashi, T., Suzuki, K., and Kuwano,
Kagaku [Earth Science] 6:49-60.
Y. 1973. The basis of the Plio-Pleistocene Boundary in Qiu, Z. 1989. The Chinese Neogene mammalian biochronologyJapan. Osaka City Univ., J. Geosci. 16:25-49.
its correlation with the European Neogene mammalian
Itihara, M., Yoshikawa, S., Inoue, K., Hayashi, T., Tateishi, M.,
zonation. In European Neogene mammalian chronology,
and Nakajima, K. 1975. Stratigraphy of the Plioed. E. H. Lindsay et al., pp. 527-556. New York: Plenum
Pleistocene Osaka Group in Sennan-Senpoku Area, south
Press.
Suzuku, M. 1988. Fission track ages of the Quaternary tuff layers
of Osaka, Japan. Osaka City Univ., J. Geosci. 19:1-29.
Itihara, M., Yoshikawa, S., Kawabe, T., and Mitamura, M.
(in Japanese, with English abstract). Geol. Soc. Japan,
1984. On so-called Shiba Unconformity in the drainage
Mem. 30:219-221.
area of the River Tsuda, Kishiwada City, Osaka. Magne- Tai, A. 1973. A study on the pollen stratigraphy of the Osaka
Group, Pliocene-Pleistocene deposits in the Osaka Basin.
tostratigraphy and fission track age of the Osaka Group (in
Kyoto Univ., Fac. ScL, Mem., Ser. G 39:123-165.
Japanese, with English abstract). Chikyu-Kagaku [Earth
Torii, M., Yoshikawa, S., and Itihara, M. 1974. Paleo-magnetism
Science] 38:1-16.
on the water-laid volcanic ash layers in the Japan.
Kamei, T. 1984. Fossil mammals: Lake Biwa and fossil mamRockmagnet. Paleogeophys. 2:34-37.
mals. In Lake Biwa, ed. S. Horie, pp. 475-495.
Dordrecht: W. Junk Publishers.
Yoshikawa, S. 1973. The Osaka Group in the southeast of Osaka
Kamei, T. 1991. Japanese probscidean fossils (in Japanese).
(in Japanese, with English abstract). Geol. Soc. Japan, J.
Tokyo: Tsukiji Shokan Co. Ltd.
79:33-45.
25
The Pliocene-Pleistocene boundary in Japan: stratigraphy in the

Boso Peninsula, central Japan
HISAO NAKAGAWA, NOBUAKI NIITSUMA, TAKASHI MITSUNASHI, MOTOYOSHI ODA, TOSHIAKI
TAKAYAMA, and TOYOSABURO SAKAI
Outline of geology
The Boso Peninsula is situated in the southern part of the Kanto

region in central Honshu, between two major tectonic provinces
of the Japanese Islands: northeastern Honshu, extending to the
north, and southwestern Honshu, extending to the west. Thick
marine sediments accumulated in a depositional basin that
appeared in the middle Pliocene in the southern part of the Kant
region; this Plio-Pleistocene section is well exposed in the Boso
Peninsula.
The Tertiary and Pleistocene deposits in the Boso Peninsula
consist of the Mineoka, Hota, Miura, Kazusa, and Shimosa
groups, in stratigraphic order, together with younger terrace
formations and volcanic-ash layers. Each of the groups is
unconformable with the others. Planktonic foraminifera indicate
that the Miocene-Pliocene transition is in the uppermost part of
the Amatsu Formation.
After deposition of the Anno Formation of the uppermost
Miura Group, the area was uplifted, and a new basin, in which
the Kazusa Group was deposited, appeared to the north of the
emerged area. In the central-to-eastern parts of the peninsula,
the Kazusa Group comprises the Kurotaki, Katsuura, Namihana, Ohara, Kiwada, Otadai, Umegase, Kokumoto, Kakinokidai, Chonan, Mandano, and Kasamori formations, in ascending order. Those formations consist of sandstone and siltstone,
with pyroclastic intercalations, and the classification of the
formations is based on the predominant lithology in the
interbedded rocks.
The pyroclastic intercalations of the Kazusa Group have been
carefully traced in the Boso, Choshi, and Miura peninsulas,
where they are used as key beds. Lateral changes in lithology and
local stratigraphy have been well established (Mitsunashi and
Yazaki, 1958, 1961; Mitsunashi et al., 1959, 1961, 1976a,b, 1979;
Yazaki and Mitsunashi, 1962; Mitsunashi and Yazaki, 1968;
Ishiwada et al., 1971). The Kazusa Group has yielded many
fossils, including mollusks, brachiopods, bryozoans, corals,
echinoids, foraminifera, calcareous nannoplankton, and plants.
Land-mammal fossils have also been found in the Umegase and
younger formations.
The Shimosa Group is distributed in the northern part of the
244
peninsula, where it underlies the Shimosa Plateau. The Shimosa

Group is made up of the Jizodo, Yabu, and Narita formations,
though many other subdivisions have been proposed for this
group. The Jizodo Formation overlies the Kasamori Formation,
with parallel unconformity. In the western area, fossil valleys
were found at the base of the Jizodo Formation. The Shimosa
Group consists largely of loose sands, but includes pebbly sand
and clayey silt layers, representing sedimentary cycles. Generally, each of the cyclic deposits consists of basal coarse sands,
lower-to-middle cross-laminated sands, and upper sandy-toclayey silts. Slight sedimentary breaks are recognized at the
bottoms of some cyclic layers. Each of the formations of the
Shimosa Group includes one to several sedimentary cycles.
The Shimosa Group has yielded abundant molluscan and
other fossils. The molluscan assemblages indicate that the
environment varied between littoral and neritic under the
influence of cool, moderate, and warm currents and brackish
water (Oyama, 1952; Hatai, 1958; Ogose, 1961; Aoki and Baba,
1980). Oxygen-isotope paleotemperatures obtained from massspectroscopic analysis of mollusk shells from the Shimosa Group
vary between 10C and 20C (Masuda and Taira, 1974).
Magnetostratigraphy and biostratigraphy
The Neogene and Pleistocene of the Boso Peninsula have been

classified into magnetozones according to the stratigraphic
sequence of polarity measurements in detrital remanent magnetization (Nakagawa, Niitsuma, and Hayasaka, 1969; Niitsuma,
1976; Nakagawa and Niitsuma, 1977). The magnetozones are
designated by letters, in alphabetical order downward from the
youngest, with the subzones in each magnetozone identified by
a numeral following the letter of the magnetozone (Figure
25.1).
The distributions of planktonic microfossils have been examined in the Neogene and Pleistocene sections of the Boso and
Choshi peninsulas by various authors (Takayama, 1967, 1973;
Sakai, 1972; Koizumi and Kanaya, 1976; Oda, 1977). The
Choshi Peninsula is situated in the northeastern part of the PlioPleistocene basin in which the Kazusa and Shimosa groups were
deposited, and the section exposed in the Choshi Peninsula is
Formation
Kokumoto
POLLEN
BENTHIC FORAMINIFERAL
ZONE
PLANKTONIC MICROFOSSIL
"Terr. f. *
Shimosa G.
Kasamori
Mandano *
Chonan
Kakinokidai
MAGNETOSTRATIGRAPHY
BIOSTRATIGRAPHY
LITHOSTRATIGRAPHY
Marker
Relative
water
temperature
Nonionella Stella
COOL\ WARM _
Crybroelphidium clavatum
Cassidulina subglobosa
Coiling
direction
Cassidulina subcarinata
Uvigenna akitaensis
Umegase
Bulimina aculeata
Bulimina-Bolivina (U)
Bolivina spissa
Otadai
Bulimina-Bolivina (L)
Stilostomella
ketienziensis
Kiwada
Bulimina striata
tiu
e
Gyroidma cf. orbiculans
a.
LUI
Namihana
i
ster
Ohara
Gyroiaina-Melonis
Katsuura
Q
a!
Kurotaki
-m
Lithology
Anno
m
r1000
siltstone
Geomagnetic polarity
^H
sandy siltstone
sandstone
Kiyosumi
- 500
Amatsu
Figure 25.1. Stratigraphy of the Plio-Pleistocene series of the Boso Peninsula.
normal
reversed
IIIIIJI indefinite
conglomerate
alternation of sandstone and siltstone
intraformational de formation
marker tuff
% terrestrial deposits
MAMMAL
TECTONICS
MAGNETO-
Nakagawa et al.
246
equivalent to that of the Boso Peninsula. Several pryoclastic key

beds of the Kazusa Group have been traced directly from the
Boso Peninsula to the Choshi Peninsula, allowing correlation of
the stratigraphic section with the magnetozones. The PlioPleistocene sediments of the Choshi Peninsula have yielded both
calcareous and siliceous microfossils, whereas the sediments of
the Boso Peninsula are poor in siliceous microfossils.
Figure 25.1 shows the lithostratigraphy, biostratigraphy, and
magnetostratigraphy for the Neogene and Pleistocene of the
Boso Peninsula. The stratigraphic distribution of the siliceous
microfossils is based mainly on occurrences in the Choshi
Peninsula (Sakai, 1972; Koizumi and Kanaya, 1976). The benthic
foraminifera zones of Aoki (1968) and the distribution of pollen
by Onishi (1969) are included in the same figure.
In the Kazusa Group, the benthic foraminifera assemblage
gradually changes upward and also westward to that of a shallowwater environment, indicating that the basinfilledfrom the west.
Cold water influenced the middle part of the Kiwada, the upper
part of the Umegase, and the lower part of the Kasamori
formations. Among the planktonic foraminifera, cold-water
species are recognized in the transitional part from the Otadai to
Umegase formations and in the upper part of the Umegase
Formation.
Thus, the lower part of the Kiwada Formation, which is
correlated with the early part of the Olduvai normal-polarity
subzone in deeper-water sediments, was deposited in a warmwater environment, but the end of the Olduvai in the middle
Kiwada saw cold-water conditions develop.
References
Aoki, N. 1968. Benthonic foraminiferal zonation of the Kazusa
Group, Boso Peninsula: vertical faunal change. Paleontol.
Soc. Japan, Trans. Proc, n.s. 70:238266.
Aoki, N., and Baba, K. 1980. Pleistocene molluscan assemblages
of the Boso Peninsula, central Japan. Univ. Tsukuba, Inst.
GeoscL, Sci. Rep., sec. B 1:107-148.
Hatai, K. 1958. Boso Peninsula, Chiba Prefecture. In Jubilee
publication in commemoration of Prof. H. Fujimoto, ed.
H. Shibata, pp. 183-201. Tokyo: Geological Institute of
Tokyo University of Education.
Ishiwada, Y., Mitsunashi, T , Shinada, Y., and Makina, T. 1971.
Geological map of the oil and gas fields of Japan, no. 10.
Mobara 1:15,000. Tokyo: Geological Survey of Japan.
Koizumi, I., and Kanaya, T. 1976. Late Cenozoic marine diatom
sequence from the Choshi district, Pacific coast, central
Japan. In Progress in Micropaleontology, ed. Y Takayanagi and T. Saito, pp. 144-159. New York: Micropaleontology Press.
Masuda, F., and Taira, K. 1974. Oxygen isotope paleotemperature measurements on Pleistocene molluscan fossils
from the Boso Peninsula, central Japan. Geol. Soc. Japan,
J. 80:97-106.
Mitsunashi, T., et al. 1979. Explanatory note of geological map of

Tokyo Bay and adjacent areas, 1:100,000. Miscellaneous
Map Series 20 (in Japanese, with English abstract). Tokyo:
Geological Survey of Japan.
Mitsunashi, T , Nakagawa, H., and Suzuki, Y (eds.) 1976a.
First International Congress on Pacific Neogene Stratigraphy, Tokyo, 1976. Guidebook for Excursion 2: Boso
Peninsula. Tokyo: Regional Committee on Pacific Neogene Stratigraphy.
Mitsunashi, T., Nasu, N., Nirei, H., et al. 1976b. Geological map
of Tokyo Bay and adjacent areas, 1:100,000. Miscellaneous
Map Series 20. Tokyo: Geological Survey of Japan.
Mitsunashi, T., Yasukuni, N., and Shinada, Y. 1959. Stratigraphical section of the Kazusa Group along the shores of
the rivers Yoro and Obitsu. Geol. Surv. Japan, Bull.
10:83-98.
Mitsunashi, T., and Yazaki, K. 1958. The correlation between
the Cenozoic strata in Boso and Miura peninsulas by the
pyroclastic key beds (no. 1). Japan Assoc. Petrol. Tech., J.
23:16-22.
Mitsunashi, T, and Yazaki, K. 1968. Geological map of the oil
and gas fields of Japan, 6, Miura Peninsula, 1:25,000.
Tokyo: Geological Survey of Japan.
Mitsunashi, T., Yazaki, K., Kageyama, K., Shimada, T, Ono,
E., Yasukuni, N., Makino, T., Shinada, Y , Fujiwara, K.,
and Kamata, S. 1961. Geological maps of the oil and gas
fields of Japan, no. 4, Futtsu-Otaki, 1:50,000. Tokyo:
Nakagawa, H., and Niitsuma, N. 1977. Magnetostratigraphy of
the Late Cenozoic of the Boso Peninsula, central Japan.
Quaternary Res. 7:294-301.
Nakagawa, H., Niitsuma, N., and Hayasaka, I. 1969. Late
Cenozoic geomagnetic chronology of the Boso Peninsula.
Geol. Soc. Japan, J. 75:267-280.
Niitsuma, N. 1976. Magnetic stratigraphy in the Boso Peninsula.
Geol. Soc. Japan, J. 82:163-181.
Oda, M. 1977. Planktonic foraminiferal biostratigraphy of the
Late Cenozoic sedimentary sequence, central Honshu,
Japan. Tohoku Univ., Sci. Repts., 2ndser. (Geol) 48:1-72.
Ogose, S. 1961. Some considerations concerning the thermal
changes indicated by the molluscan fossils from the Upper
Pliocene and the Lower Pleistocene strata in the Boso
Peninsula, South Kanto. Geol. Soc. Japan, J. 67:655-669.
Onishi, I. 1969. Pollen flora of the Kazusa Group in the Boso
Peninsula, Japan (in Japanese). Chikyu Kagaku [Earth
Science] 23:236-242.
Oyama, K. 1952. On the fossil molluscan community of Tyonan
and Kasamori formations exposed between Mobara and
Turumai. Japan Assoc. Petrol. Tech., J. 17:59-67.
Sakai, T. 1972. Radiolarian biostratigraphy of the Upper
Cenozoic in the Choshi district, central Honshu, Japan.
Unpublished Ph.D. thesis, Tohoku University.
Takayama, T. 1967. First report on nannoplankton of the Upper
Tertiary and Quaternary of the southern Kwanto Region,
Japan. Geol. Bundesanst. Wien, Jb. 110:169-198.
Takayama, T. 1973. On the distribution of calcareous nannoplankton in the Cenozoic of Japan. Geol. Soc. Japan,
Mem. 8:45-63.
Yazaki, K., and Mitsunashi, T. 1962. Geological map of the oil
and gas fields of Japan, 3, Yokosuka, 1:15,000. Tokyo:
26
The base of the Quaternary in China

ZHANG SHOUXIN
Introduction
The Upper Pliocene and Lower Pleistocene sediments in eastern
China include various types of deposits, providing an ideal
region for study of this part of the geologic record. The cave
fillings, loess deposits, and strata of fluviolacustrine, coastalplain, and marine origin combine to reveal the biological history,
topography, climate, and active tectonics during that interval.
This chapter presents an overview of the sections that are the
most significant for separating the Pliocene and Pleistocene.
Starting in 1949, most Chinese geologists adopted the proposal
made at the 1948 XVIII International Geological Congress in
London to place the Pliocene-Pleistocene boundary at the first
immigration of "cold guests" into the marine faunas of the
Mediterranean region, exemplified by the paleontology at the
base of the Calabrian Stage of Italy, and to consider that level
correlative to the base of the Italian Villafranchian Stage in
continental deposits. Because the Plio-Pleistocene vertebrate
faunas in China have long been famous and better known
(Teilhard de Chardin and Piveteau, 1930) than the marine
sequences, the continental Villafranchian concept, rather than
the marine Calabrian concept, was widely adopted as the basis
for recognition of the base of the Pleistocene in China. It is now
recognized that in Italy the earliest Villafranchian is almost twice
as old as the base of the Calabrian Stage (Azzaroli et al., Chapter
11, this volume). The Chinese workers have maintained,
however, that the Villafranchian definition is in agreement with
the first cold-climate faunas of both marine and continental
environments in that region, coinciding closely with the GaussMatuyama paleomagnetic reversal at about 2.6 Ma.
Nonmarine stratigraphic sections

The Lochuan Loess section
The Lochuan Loess occupies a basin typical of the many large
basins in the center of the loess district in northern China. The
Lochuan section is located at Heimugou, about 6.5 km south of
the capital town of Lochuan county, Shaanxi province. Loess
accumulated in that area continuously to a thickness of 136 m
over an uneven erosional surface cut into the Hipparion Red

Clay Formation. By lithostratigraphic classification, Liu and
Chang (1961) divided the Lochuan Loess into three subunits: the
Wucheng Loess, the Lishi Loess, and the Malan Loess.
Fossils found in the lower or Wucheng Loess subunit are
mainly mammals (Figure 26.1). The species Prosiphneus intermedius and Hipparion (Proboscidipparion) sinense, found in
the unconformably underlying Hipparion Red Clay, do not
persist into the Wucheng Loess. At the same time, we note the
first appearance of Myospalax tingi and Allocricetus (Zheng et
al., 1985) in the lowest part of the loessic deposits. Magnetostratigraphic study of the Lochuan Loess sections shows that
the Wucheng Loess subunit predates the Olduvai subchron of the
Matuyama chron (Heller and Liu, 1982). Other studies in the
Lochuan area have identified the Gauss-Matuyama boundary
just below the base of the Wucheng Loess (Zheng et al., 1985).
Nihewan Series
The Yangyuan and Yuxian basins are intermontane basins in the
northern part of Hebei province, where Pliocene and Pleistocene
fluviolacustrine sediments are classified as the Nihewan Series.
The "Nihewan fauna" described by Teilhard de Chardin and
Piveteau (1930) comes from the sands and clays of this unit in the
vicinity of Nihewan and Xiaodukou, but recent investigations
have discovered additional faunal remains ranging from the base
to the top of the 90-m-thick sequence. Faunal and paleomagnetic
studies indicate that those strata represent a span of time that
overlaps both the customary boundary in China and the revised
boundary as recommended by IGCP-41. In modern studies, the
lower part of the original Nihewan Formation has been
recognized as a separate formation according to lithologic and
biostratigraphic evidence, so that the entire sequence is properly
termed the Nihewan Series.
At the margin of the basin, near Dongyaozitou, Yuxian, the
Nihewan Series rests unconformably on the upper Hipparion
Red Clay (there named the Yuxian Formation), in which
Hipparion houfense, Antilospiroides houfense, Sinoryx, Gazella
blacki, and Viverra have been found (Wang and He, 1982; Zheng
and Cai, 1991). The lowermost Nihewan Series, resting on the
247
Zhang Shouxin
248
GAUSS
Old. MATUYAMA
Jar.
BRUNH.
Lochuan Loess Series

Wuchenq
Lower Lishi
Red clay
X""-X
Y_.........
Prosiphneus intermedius
Myospalax chaoyatseni
Myospalax fontanierii
Microtus brandtoides
Allocricetulus ehiki
Ochotonoides complicidens
Proboscidipparion sinensis
X
.........................
.......
XX
x
x~-.-
Nihewan Series
Daodi I
(1)
Nihewan
2)
(3) (4)
X
K P U l
Xiaodukou
(5) (6 )
(7)
X
X
X
X
X
X
X
Figure 26.1. Plio-Pleistocene

mammal biostratigraphy, on the
basis of selected taxa from
paleomagnetically analyzed sections in the plateau loess of
Lochuan county andfluviatiledeposits in the Nihewan region,
northern China. Many recorded
taxa are omitted from this summary. Data on the Lochuan Loess
adapted from Zheng et al. (1985),
and data on Nihewan Series
adapted from Zheng and Cai
(1991). Faunal levels: 1, Daodi; 2,
Dongyaozitou; 3, Danangou; 4,
Xiashagou; 5, Donggutuo; 6,
Xiaodukou; 7, Xujiawa. The
Xiashagou l.f. is from reoccupied
sites of typical "Nihewan fauna"
of Teilhard de Chardin and
Piveteau in the Yangyuan Basin
(R. H. Tedford, personal communication, 1993). The Donggutuo
and Xiaodukou l.f. are from exposures in the interfluve between
the Yangyuan and Yuxian basins,
and the remaining local faunas
are from the Yuxian Basin (Zheng
and Cai, 1991).
X
X
X
X?
X X
X
X
X
X
X
X
X
X
X
X?
X X
X X
X
X
X X
X X
X X
X
X
X
X
X
X
X
X
eroded Yuxian surface, consist of brown and yellow sands and

clays, overlain by strikingly cross-bedded gravels. Originally
assigned to the Nihewan Formation, those two units were named
the Dongyaozitou Formation by Tang and Ji (1983), but more
recently the basal brown and yellow sands have been designated
the Daodi Formation by Du et al. (1988), and the cross-bedded
gravels are recognized as the base of the Nihewan Formation
sensu stricto.
Several sites in the Daodi Formation have produced arichand
apparently homogeneous fauna (Cai, 1987; Zheng and Cai,
Myospalax tingi
Myospalax fontanieri
Mimomys orientalis
Mimomys cf. M. youhenicus
Prosiphneus sp.
Pltymys cf. P. hintonl
Microtus cf. M. ratticepoides
Allophalomys cf A. pliocaenicus
Pliopentalagus nlhewanensis
Hypolagus schreuderl
Ochotona minor
Ochotona nihewanica
Ochotonoides complicidens
Meles chiai
Eucyon minor
Acinonyx pleistocaenicus
Acinonyx jubatus
Lynx variabilis
Hyaena licenti
Vulpes chikushanensis
Canis chihliensis
Zygolophodon sp.
Elephas namadicus
Proboscidipparion sinensis
Hipparion cf. H. houfense
Equus sanmenensis
Coelodonta antiquitatis
Paracamelus sp.
Muntiacus bohlini
Axis shansius
Rusa elegans
Census elaphus
Megaloceros sanganhoensis
Gazella sinensis
Bison palaeosinensis
Bos primigenius
1991) termed the Daodi fauna. The small-mammal assemblage is

characterized by Mimomys orientalis and Prosiphneus {? = P.
paratingi) and includes representatives of Sorex, ?Beremendia,
Eucastor, Nannocricetus, Germanomys, Orientalomys, Apodemus, Mus, Rattus, Paralactaga, Sminthoides, Hypolagus
schreuderi, Pliopentalagus nihewanensis, Ochotona cf. O.
lagreliiy O. minor, and O. erythrotis {? = O. nihewanica). Large
mammals include Lynx variabilis, Hipparion cf. H. houfense,
Proboscidipparion sinense, Paleotragus progressus, Antilospira
yuxianensis, Axis shansius, and an undetermined elephantid.
Base of the Quaternary in China
249
The Dongyaozitou local fauna, restricted from the original sense

Tang and Ji (1983) believed that the Dongyaozitou fauna was
of Tang (1980) according to Zheng and Cai (1991), occurs at the clearly transitional between the fauna of the Hipparion Red
top of the Daodi Formation; it differs from the Daodi fauna, Clay, considered as later Pliocene in China, and the classic
with Mimomys cf. M. youhenicus in place of M. orientalis. Nihewan faunal unit, which has been dated to the Pleistocene
Occurrences of Nyctereutes cf. N. sinensis, Zygolophodon, according to the Chinese usage. The mammals of the
Coelodonta antiquitatis, Paracamelus, and Gazella sinensis, so Dongyaozitou fauna (of Tang and Ji, 1983) can be correlated
far unrecorded from the Daodi fauna, are noted at that level. with the Youhe faunal unit of the lower Sanmen Formation at
The only known equid is Proboscidipparion sinense, with no Weinan, Shaanxi, and also can be correlated to the later part of
evidence of either Hipparion or Equus. A possible record of the the Montopoli unit (Etouaires fauna) of the early Villafranchian,
latter, cited by Du et al. (1988), is considered invalid (R. H. dated to the earliest part of the Matuyama in Europe. The classic
Tedford, personal communication, 1992). The fossil fish Nihewan fauna contains Myospalax tingi, which indicates a
Pungitius nihowanensis is also first recorded from this level (Tang correlation of the lower Nihewan Formation to the Wucheng
and Ji, 1983).
Loess in Shaanxi (Zheng et al., 1985); Allophaiomys cf. A.
In the cross-bedded gravels at the base of the Nihewan pliocaenicus, Pity mys cf. P. hintoni, and Elephas (PalaeoloxoFormation s.s. at Dongyaozitou, and in beds just above the don) namadicus are elements in common between the Nihewan
gravels, Zheng (1981) and Li (1984) discovered a younger fauna fauna and the pre-Olduvai Villafranchian in central Asia and
called the Danangou local fauna, in which the first reliably Europe (e.g., Saint-Vallier). The presence of Microtus, Bos, and
identified remains of Equus are found. There, E. sanmenensis Cervus in the fauna collected at Donggutuo and Xiaodukou is
occurs together with Proboscidipparion sinense, Ochotona consistent with the late Villafranchian age suggested by the prenihewanica, Orientalomys nihowanicus, Canis chihliensis minor,Jaramillo paleomagnetism in those strata.
Meles chiai, Coelodonta antiquitatis, and Gazella sinensis. From
present indications (Zheng and Cai, 1991) that fauna is essenSanmen Formation
tially the same age as the classical Nihewan fauna of Teilhard de
Chardin and Piveteau (1930); R. H. Tedford (personal communi- The fluviolacustrine sediments developed in the valleys of the
cation, 1992) notes that 80% of the Danangou species-level taxa Huang He (Yellow River) and Wei He have been named the
(including both equids) are also found in the larger Nihewan Sanmen Formation, after Sanmenxia, Henan province. The
fauna, as summarized by Zheng and Cai (1991). Key small- mammalian fossils in the Sanmen Formation are similar to those
mammal taxa from the Nihewan fauna include Myospalax tingi, from the Nihewan Formation. The lower part of the Sanmen
Pity mys cf. P. hintoni, Allophaiomys cf. A. pliocaenicus, and Formation in Weinan, Shaanxi, contains a mammalian fauna
Ochotonoides complicidens. Among the large mammals, aside known as the Youhe faunal unit, with Mammuthus {"Arfrom those previously noted, which also occur at Danangou, the chidiskodon") youheensis, Hipparion houfense, Rhinocerotidae,
presence of Equus teilhardi, Elephas (Palaeoloxodon) nama- Cervavitus sp., Sus subtriquetra, Nyctereutes sinensis, Mimomys
dicus, Elaphurus bifurcatus, Eucladoceros boulei, and Rusa youhenicus, Cricetulus sp., and Ochotonoides cf. O. complielegans is of some significance.
cidens. The Youhe faunal unit, as noted, compares most closely
A widespread unconformity separates the beds with the typical with the Dongyaozitou faunal unit below the typical Nihewan
Nihewan fauna from upper levels, where fossils at Donggutuo Formation in the Yuxian Basin.
(Wei, Meng, and Cheng, 1985) and at Xiaodukou (Wei, 1983)
document the initial occurrences of Microtus raticepoides,
Yuanmou section
Myospalax fontanieri, Acinonyx jubatus, Megaloceros, Bos
primigenius, and Cervus elaphus. At Donggutuo and nearby The Yuanmou section is exposed on the southern bank of the
sites, abundant stone tools have also been discovered (Wei, 1985; Jinsha River, in an intermontane basin in northeastern Yunnan
Wei et al., 1985; Schick et al., 1991), apparently among the province. In that basin, Pliocene and Pleistocene fluviolacustrine
earliest evidences of human occupation in eastern Asia (Aguirre, sediments some 600 m in thickness consist mainly of sands and
gravels, as well as sandy clays with lignite beds. Beginning in
Paleomagnetic analysis of the Nihewan beds indicates that the 1938, with discovery of teeth of Equus yunnanensis, the
upper Daodi Formation, at the approximate level of the Yuanmou beds were taken as the type section for the Lower
Dongyaozitou local fauna, contains the Gauss-Matuyama bound- Pleistocene in southern China, equivalent to the Nihewan Series
ary (Du et al., 1988). The lower Nihewan Formation containing of northern China. In 1961, Zhou Ming-zhen concluded that the
the Danangou l.f. and the classic Nihewan faunas shows the faunal assemblage of the Shagou Lignite, found in the lower part
reversed paleomagnetic polarity of the lower Matuyama chron of the Yuanmou section, could be correlated to the Dhok Pathan
(Dong et al., 1986). Above the stratigraphic hiatus in the middle fauna of the Siwaliks, based on the presence of Enhydriodon cf.
of the Nihewan Formation, the upper Nihewan faunas and E. falconed, and that the Shagou level was therefore of "late
Paleolithic artifacts occur in beds just below the Jaramillo Pliocene" age. The Yuanmou section was subsequently divided
subchron. Strata recording the Olduvai subchronozone are not by You et al. (1978) into three formations: a lower (Shagou), a
middle (Yuanmou), and an upper (Shangnabang).
present, because of the mid-Nine wan hiatus.
Zhang Shouxin
250
GAUSS
Old.
MATUYAMA
Lower Sanmen - Youhe
Jar.
BRUNHES
1
Mimomys youhenicus
Ochotoinoides complicidens
Mammuthus youheensis
Hipparion houfense
Equuus sanmenensis
Hyaena licenti
Nyctereutes sinensis
Euctenoceras sp.
Bison paleosinensis
X
X
X
X
X
Shagou
Yuanmou Series
(condensed)
Yuanmou
Shangnabang
X
X
X
X
X
X
X
X
X
X
X
X
X*
X
X
X
Enhydriodon cf. E. falconieri

Hipparion sp.
Chilotherium sp.
Dicerorhinus sp.
Stegolophodon banguaensis
Stegodon yuanmouensis
Stegodon orientalis
Hyaena licenti
Felis tigris
Felis pardus
Vulpes chikushanensis
Canis yunnanensis
Equus yunnanensis
Rhinoceros cf R. sinensis
Sus scrofa
Muntiacus cf. Mbohlini
Rusa stehlini
Homo erectus
Maritime Plains section (subsurface)

X
X
X
X
X
X
X
X
X
X
X
X
Figure 26.2. Plio-Pleistocene

biostratigraphy according to selected
taxa from paleomagnetically analyzed sections in Sanmenxia (Shaanxi) and
Yuanmou (Yunnan), and from boreholes
S-5 and Bo-3 on the northern coastal
plain, Shunyi county, near Beijing.
X
X
X
X
Globigerina bulloides
Globigerina pachyderma
Globorotalia puncticulata
Globigerinoides trilobus
Turborotaiia continuosa
Turborotalia anfracta
Hyalinea balthica
Buccella frigida
Elphidiella arctica
Coccolithus pelagicus
Emiliana huxleyi
Pseudoemiliana lacunosa
Gephyrocapsa protohuxleyi
llyocypris kaifengensis
Leucocythere gongheensis
At present, the Shagou Formation has yielded a fossil pardus, and Rhinoceros cf. R. sinensis, which are unknown from
assemblage that indicates late Pliocene-early Pleistocene (i.e., the older levels. The rare mammals from the Shangnabang
early Villafranchian) age, with Hipparion, Dicerorhinus, Chilo- horizon indicate correlation to the Middle Pleistocene Stegodon
therium yunnanensis, the proboscideans Serridentinus, Stegolo- orientalis fauna of southern China.
phodon, and Stegodon, and Enhydriodon species (Figure 26.2).
Paleomagnetic studies by Li et al. (1976) and Cheng et al.
In the overlying Yuanmou Formation, which You et al. (1978) (1977) agreed on placing the boundary between the Gauss and
believed to be restricted to early Pleistocene age, a different Matuyama chrons at the boundary between the Shagou and
assemblage is recorded, with a few holdovers from Shagou time Yuanmou formations, which is consistent with the overall
(most notably Stegodon), but with many new taxa, such as Equus Villafranchian character of the Yuanmou vertebrate assemblage.
yunnanensis, Sus scrofa, Felis (Panthera) tigris, Felis (Panthera)Fossil teeth of Homo erectus from the Shangnabang faunal level
were initially assigned an age of more than 1.6 Ma, on the basis
of paleomagnetic analysis, but Liu and Ding (1983) and Jiang,
Sun, and Liang (1988) have argued that the normal-polarity
sediments below the beds from which these specimens were
collected correspond to the Brunhes, not the Olduvai, and that
the Shangnabang fossils are therefore younger than 0.7 Ma. In
that view, the stratigraphic interval yielding Yuanmou fauna
(i.e., the Yuanmou Formation only) spans the entire Matuyama,
from 2.6 to 0.7 Ma, and thus it is a possibility that the Yuanmou
fauna is mixed from levels of pre-Olduvai and post-Olduvai age.
The stratigraphy, however, is rather unclear, and as Aguirre et
al. (Chapter 9, this volume) have pointed out, the faunal and
paleomagnetic data are also consistent with correlation of the
normal-polarity strata below the Shangnabang fauna to the
Jaramillo event at about 1.0 Ma. The Shangnabang H. erectus
might thus come from uppermost Matuyama strata.
Many Chinese vertebrate paleontologists consider that it is
suitable to place the faunal levels of early Villafranchian age in
the early Pleistocene. In the Nihewan and equivalent fluviolacustrine sequences in northern China, there are thus two choices
for the reference point for the N/Q boundary (the base of the
Quaternary), according to the preferred concept of the Pleistocene in that region. One option is to place it at the base of the
Nihewan Formation {sensu strictd), or Bed 6, at the firstappearance datum of Equus sanmenensis, and the other is in
Nihewan Bed 4, at the last-appearance datum of Zygolophodon
and the fish Pungitius nihowanensis. In both options, the N/Q
boundary is close to the Gauss-Matuyama boundary at 2.6 Ma.
However, it is also admitted that the Youhe-Dongyaozitou
mammalian faunas could mark the Upper Pliocene, in view of
the present decision to define the N/Q boundary at Vrica, at a
level equivalent to the top of the Olduvai normal chron and thus
well above the lower Nihewan, lower Sanmen, and lower
Yuanmou levels.
The maritime-plain sections
Plio-Pleistocene sediments in the North China Plain consist

mainly of fluviolacustrine deposits interbedded with marine
layers. Drilling in these strata has shown that the PlioPleistocene interbeds of marine and nonmarine sediments were
deposited in coastal environments.
251
of calcareous nannofossils in the same bed includes Cyclococcolithus leptoporus, Coccolithus pelagicus, Emiliana huxleyi, Pseudoemiliana cf. P. lacunosa, Gephyrocapsa oceanica, and G.
protohuxleyi. A similar marine interval has been observed in
many borehole sections east of Beijing, and paleomagnetic study
has shown those occurrences to be correlative to the marine bed
in the S-2 core section.
Chinese paleontologists recognize that the calcareous nannofossils of the Beijing plain can be correlated to the entire range of
the NN19 through NN21 nannofossil zones, which encompasses
the Quaternary. Therefore, they prefer to place the PliocenePleistocene boundary at the base of the Hyalinea-Globigerina
assemblages, as seen in the S-2 core, essentially coincident with
the Gauss-Matuyama polarity reversal at 468 m.
Bo-3 borehole section. In a 600-m core from the Bo-3 borehole
(3915'N, 11830'E), in the northern coastal plain, three major
paleomagnetic polarity zones were recognized by Li and Wang
(1982). The first zone showed normal polarity from the surface
to a depth of 171 m, with the exception of a few reversed samples
from 15 m and 103 m. From 171 m to 493 m, the polarity of the
remanent magnetization in the samples is reversed, and in the
normal-polarity zone below 493 m there is a zone of mixed
polarities from 572 m to 588 m. From the different lines of
evidence, these three polarity zones appear to correspond to the
Brunhes normal, Matuyama reversed, and Gauss normal chrons,
respectively.
In the lower part of the core, three marine beds occur in an 80m interval. At a core depth of 505 m, about 15 m below the
assumed Gauss-Matuyama boundary, an ostracode assemblage
is found that represents a transition from the older Leucocythere
gongheensis assemblage to the succeeding Ilyocypris assemblage.
The exact position for the N/Q boundary, equivalent to the
boundary in Chinese continental sequences, should thus be at a
depth of 515 m, at the first appearance of Ilyocypris kaifengensis.
In the same Bo-3 borehole, there is a marked change in pollen
flora at a depth of 464 m, from a mixed conifer and broad-leaf
forest assemblage with Ulmus, Pinus, Betula, and others to a dry
steppe assemblage with Chenopodia ceae and Artemisia. According to the magnetostratigraphy described earlier, the changes in
ostracoda and vegetation observed in the Bo-3 core happened
near the Gauss-Matuyama boundary. Those data roughly agree
with the results obtained from the S-5 core section, in which the
layer bearing Hyalinea baltica is just above the GaussMatuyama reversal, at about 2.3 Ma.
In 1978, Zhang and co-workers studied the 600-m Ca-13 core
section. Brunhes, Matuyama, Gauss, and Gilbert polarity zones
were recorded. Those authors put the lower boundary of the
Pleistocene at the Mammoth subchron of the Gauss normalpolarity chron.
S-5 borehole section. Important information has come from

examination of the main stratigraphic features of samples taken
from the S-5 borehole in Shunyi county, near Beijing (Wang and
He, 1982; An et al., 1979). Jurassic basement was encountered at
793.7 m. The Matuyama-Brunhes geomagnetic-polarity reversal
was placed at a depth of 160 m, and the Gauss-Matuyama
reversal at a depth of 468 m. At 428 m, equivalent to an age of
about 2.3 Ma, a marine bed 10 m thick is composed of gray-green
to blue-gray medium-coarse silty sand, with thin layers of sandy
Marine section
clay. That marine layer yields benthic and planktonic foraminifera such as Globigerina bulloides, G. pachyderma, Hyalinea Pliocene and Pleistocene sediments in the South China Sea and
baltica, Buccella frigida, and Elphidiella arctica. An assemblagearound Taiwan Island belong to the neritic environment.
252
Zhang Shouxin
China (in Chinese). Chinese Acad. Geol. Sci., Bull.

Research in Taiwan shows that the first-appearance datum levels
15:149-155.
for the planktonic foraminifera Globorotalia truncatulinoides
and the calcareous nannofossil Gephyrocapsa oceanica, together Du H.-J., Wang A.-D., Zhao Q.-Q., and Cai Baoqua. 1988. A
new late Pliocene stratigraphic unit in the Nihewan region:
with the last-appearance datum of Discoaster brouweri, are more
the Daodi Formation (in Chinese, with English summary).
or less synchronous. These events may be correlated with the
Earth Sci. J., China Univ. Geosci. 13:561-568.
lower boundary of the G. truncatulinoides zone (N22 zone) and Heller, R, and Liu Tungsheng. 1982. Magnetostratigraphical
dating of loess deposits in China. Nature 300:431-433.
the lower boundary of the Pseudoemiliana lacunosa zone (NN19
zone) of the Mediterranean (Rio et al., Chapter 5, this volume). Jiang N., Sun R., and Liang Q. 1988. Strates cenozoiques
tardives et fossiles humains dans le basin de Yuanmou,
Yunnan, Chine. L'Anthropologie 92:939-944.
Li Huamei and Wang Junda. 1982. Magnetostratigraphic study
Conclusions
of several typical geologic sections in North China. In
QRAC: Quaternary geology and environment of China, ed.
The beginning of loess deposition in China, which coincided with
Liu Tungsheng, pp. 33-37. Beijing: China Ocean Press.
major changes in the large- and small-vertebrate faunas equivaLi Pu, Chien Fang, Ma Hsinghua, Pu Chingyu, Hsing Lisheng,
lent to the Middle Villafranchian, dates to the same level as the
and Chu Shichang. 1977. Preliminary study on the age of
Pretiglian cold-climate phase in western Europe and to the
Yuanmou Man by paleomagnetic technique. Sci. Sinica
10:646-663.
earliest loessic deposits in Stranzendorf, Austria, and in
Tadjikistan (Sibrava, 1992). Following Teilhard de Chardin and Li Yi. 1984. The early Pleistocene fossil mammals of Danagou,
Yuxian, Heibei (in Chinese, with English abstract). Vert.
Piveteau (1930), this level was adopted many years ago by
Palas. 22:60-68.
Chinese stratigraphers as a logical position for the Pliocene- Liu Tungsheng and Chang T.-H. 1961. The Huangtu (loess) of
Pleistocene boundary. The resolution of the 1948 IGC that
China. In Report of 6th INQUA Congress (Warsaw), vol.
6, pp. 503-524. Warsaw: Polish Academy of Science.
established the criteria for defining the Pliocene-Pleistocene
boundary in Italy considered that the change in continental Liu Tungsheng and Ding Menglin. 1983. Discussion on the age of
"Yuanmou Man." Acta Anthropol. Sin. 2:40-48.
climates in the middle Villafranchian was coeval with the
Schick, K., Toth, N., Wei Qi, Clark, J. D., and Etler, D. 1991.
appearance of cold-climate fossils in the marine deposits of the
Archaeological perspectives in the Nihewan Basin, China.
Calabrian Stage. Although this correlation appears to be
/. Hum. Evol. 21:13-26.
incorrect for the Mediterranean Basin, it describes the situation Sibrava, V. 1992. Should the Pliocene-Pleistocene boundary be
lowered? Sver. Geol Undersok., Ser. Cfl 81:327-332.
in China rather accurately. Chinese micropaleontologists observed the appearance of the Hyalinea baltica-Globigerina Tang Yingjun. 1980. Note on a small collection of Early
Pleistocene mammalian fossils from northern Hebei (in
pachyderma cold-climate assemblage in northern China to
Chinese, with English abstract). Vert. Palas. 18:314-322.
coincide with the Gauss-Matuyama boundary, and therefore to Tang Yingjun and Ji Hongxiang. 1983. A Pliocene-Pleistocene
coincide with the mid-Villafranchian change in mammal faunas.
transitional fauna from Yuxian, northern Hebei (in Chinese, with English abstract). Vert. Palas. 21:245-254.
That microfossil datum is found above the Olduvai normalTeilhard
de Chardin, P., and Piveteau, J. 1930. Les mammiferes
polarity subchron in the upper Matuyama in Mediterranean
fossiles de Nihowan (Chine). Ann. Paleontol. 19:1-122.
sections, including sections of Calabrian age (Pasini and Wang Naiwen and He Xixian. 1982. Discovery of calcareous
Colalongo, Chapter 2, this volume; Azzaroli et al., Chapter 11,
nannofossils in Beijing plain and its significance for
this volume). The difference in the times of these two placements
stratigraphy, palaeogeography and palaeoclimatology. In
QRAC: Quaternary Geology and environment of China,
indicates that the first appearances of Hyalinea baltica and
ed.
Liu Tungsheng, pp. 105-108. Beijing: China Ocean
Globigerina pachyderma in China (or elsewhere in Asia) were
Press.
the first evolutionary appearances of those taxa, and their first
Wei Qi. 1983. A new Megaloceros from Nihowan beds (in
appearances in the Mediterranean were delayed by regional
Chinese, with English abstract). Vert Palas. 21:87-95.
environmental effects.
Wei Qi. 1985. Palaeoliths from the lower Pleistocene of the
Nihewan beds in the Donggutuo site (in Chinese, with
English abstract). Acta Anthropol. Sinica 4:289-300.
Wei Qi, Meng Hao, and Cheng Shengquan. 1985. New
References
paleolithic site from the Nihewan (Nihowan) beds (in
Chinese, with English abstract). Acta Anthropol. Sinica
An Zhisheng, et al. 1979. Magnetostratigraphy of the core S-5
4:223-232.
and the transgression in the Beijing area during the Early
You Yuzhu, Liu Houyi, et al. 1978. Later Cenozoic strata in
Matuyama Epoch. Geochemica 4:343-346.
Baguo Basin, Youanmou, Yunnan. Collected papers in
Cai Baoqua. 1987. A preliminary report on the late Pliocene
stratigraphy and paleontology, Chinese Academy of Scimicromammalian fauna of Yangyuan and Yuxian, Hebei
ences, no. 7. Beijing: Geology Press.
(in Chinese, with English abstract). Vert. Palas. 25:124Zhang Hongcai, et al. 1978. Paleomagnetic study of two sedi136.
ment cores from the northern coastal region of China.
Cheng Guoliang, Lin Jinlu, et al. 1977. Discussion of the age of
Oceanol. Limnol. Sinica 9:188-192.
Homo erectus youanmouensis and the event of Early
Zheng Shaohua. 1981. New discovered small mammals in the
Matuyama. Sci. Geol. Sinica 1:12-22.
Nihewan bed (in Chinese, with English abstract). Vert.
Dong M.-N., Wang Y.-S., et al. 1986. A study of the Nihewan
Palas. 19:348-358.
beds of the Yangyuan-Yuxian Basin, Hebei Province,
253
Zheng Shaohua and Cai Baoqua. 1991. Fossil micromammals Zheng Shaohua, Yuan B., Gao F.-Q., and Sun F.-Q. 1985.
from the Dongguo section of Dongyaozitu, Yuxian county,
Paleovertebrate biology and palaeoecology of the Loess in
Hebei Province (in Chinese, with English summary). In
China (in Chinese). In Loess and the environment (in
Contributions to INQUA XIII, Beijing, 1991, pp. 100-131.
Chinese), ed. Liu Tungsheng, pp. 113-140. Beijing:
Science Press.
27
Plio-Pleistocene deposits and the Quaternary boundary in

sub-Saharan Africa
H. BASIL S. COOKE
Introduction
According to the definition of the lower boundary of the
Pleistocene epoch at Vrica (Pasini and Colalongo, Chapter 2,
this volume), the top of the Olduvai subchron can be used as a
convenient approximation of the Quaternary boundary in
nonmarine areas. In Africa, several highly fossiliferous stratigraphic sequences are known in which the limits of the Olduvai
subchron can be placed, either by direct paleomagnetic observations or through good radiometric control. The faunal associations in those sequences can then be used for correlating other
deposits in which paleomagnetic or radiometric age data are not
available. The major sequences with geochronological control
are in East Africa, whereas the occurrences in southern (and
northern) Africa are generally deficient with regard to isotope
age determinations and paleomagnetic data.
East African Plio-Pleistocene deposits
The highland areas of East Africa, including Ethiopia, have been
strongly affected by tectonic and volcanic activity since at least
the early Miocene, and the rift valleys and downwarped areas
have provided unusually good traps for the accumulation of
fossil-bearing sediments, in many cases associated with lavas or
volcanic tuffs that furnish radiometric or fission-track ages.
Geochemical "fingerprinting" of tuffs has also been employed
successfully as a correlation tool. The broad geological framework has been described by a number of specialists, and the
fossil mammals have been treated extensively (Maglio and
Cooke, 1978).
The most important deposits of Plio-Pleistocene age occur in
six areas: (1) the Awash Valley and adjoining areas in northcentral Ethiopia, (2) the lower Omo Basin in southwestern
Ethiopia, north of Lake Turkana, (3) the areas flanking the
northern half of Lake Turkana, Kenya (the Koobi Fora area to
the east and the Nachukui area to the west), (4) the Olduvai
Gorge, Laetoli, and Lake Natron areas in northern Tanzania, (5)
the Lake Baringo area in north-central Kenya, (6) western
Kenya and the Western, or "Albertine," Rift in Uganda, and the
Upper Semliki area in Zaire. The principal localities are shown
254
in Figure 27.1. Magnetostratigraphic sequences that include the

Matuyama (Plio-Pleistocene) interval have been most carefully
worked out in the Omo Basin and Olduvai and are reasonably
well established at Koobi Fora and in parts of the Awash and
Baringo successions.
Awash Valley, Ethiopia

A portion of the Middle Awash region, opening into the Afar
depression, is flanked by sediments ranging in age from midMiocene to Holocene. The formations recognized in that
sequence have been designated the Awash Group by Kalb et al.
(1982), although the term has not been widely used by others.
The Hadar Formation, some 200 m thick, contains important
hominid fossils, and for that reason it has been described in
detail (Aronson et al., 1977; Johanson, Taieb, and Coppens,
1982; Tiercelin, 1986), and the fauna has been analyzed (White,
Moore, and Suwa, 1984). Although the exact age range of the
Hadar Formation was a subject of controversy for some years
(Brown, 1982; Aronson, Walter, and Taieb, 1983), the modern
consensus is that the unit ranges between 3.5 Ma and 2.9 Ma
(Haileab and Brown, 1992; Walter, 1994) and thus is clearly midPliocene in age. Substantiation comes from geochemical correlations of the Sidi Hakoma Tuff (SHT) in the upper Hadar with (1)
the Tulu Bor Tuff at Koobi Fora, interpolated to 3.39 Ma in the
orbitally tuned time scale (Walter and Aronson, 1993), (2) Tuff B
in the Shungura Formation of the Omo area, and (3) marine tuffs
in the Gulf of Aden (Sarna-Wojcicki et al., 1985).
The Matabaietu Formation in the Awash Group is not dated
radiometrically, and its stratigraphic relationship to the Hadar
Formation is not established, but it has a younger, apparently late
Pliocene, fauna. It is overlain by the Wehaietu Formation, also
undated, which has a fauna indicative of a time range beginning in
the later part of the early Pleistocene (Kalb and Mebrate, 1993).
The Bodo member, in the lower part of that formation, has
yielded Acheulean artifacts and the cranium of an archaic Homo
(Conroy et al., 1978; Kalb et al., 1980). There are other significant
paleoanthropological occurrences in the same valley (Clark et al.,
1984) and on the plateau near Gadeb (Clark, 1987).
Quaternary in sub-Saharan Africa
255
CHESOWANJA
CHEMERON*VLoke Baringo
Mt
Kenya i
NAIROBI |
OLORGESAILIE^t
PLIOCENE-PLEISTOCENE
FOSSIL
LOCALITIES
j ^ Foss iI
Towns
loca I ity
Lake
Natron
High peaks
"*-* International boundary
OLDUVAI^t
LAETOLI *
Terraces on the upper Awash River, 50 km south of Addis

Ababa at Melka-Kunture in the central Ethiopian Rift, have
yielded important artifact assemblages on occupation floors, as
well as hominid and other (as yet largely undescribed) mammal
fossils, referable to eight cultural levels spanning the entire
Pleistocene (Chavaillon et al., 1979). Radiometric dating of the
tuffs and magnetostratigraphy at Melka-Kunture (Cressier,
1980) indicate that the oldest levels are within the midpart of the
Matuyama re versed-polarity zone, from 1.6 Ma to about 1.0 Ma.
Chavaillon et al. (1979) stressed the similarity between the
KilimanjaV'
MOSHI
Figure 27.1. Principal fossil localities of Plio-Pleistocene age in East

Africa. The inset shows the major
localities in Ethiopia.
Melka-Kunture "Olduwan" culture and that of Beds III at

Olduvai Gorge, but the reversed polarities of the sediments in
Melka-Kunture Beds 1 and 2 suggest that they are equivalent
only to the later "Oldowan" cultural levels in upper Bed II
(discussed later). Evidently, the base of the sequence is very
close to the Pliocene-Pleistocene boundary, as defined at Vrica.
Another Ethiopian Plio-Pleistocene sequence was described
at Konso-Gardula (Asfaw et al., 1992). The upper levels of this
section, dated within the range 1.50-1.40 Ma, contain an
abundant fauna, with early Homo erectus and Acheulean-type
256
H. Basil S. Cooke
tools, closely comparable to Olduvai Bed II. The Chari Tuff of

Koobi Fora (1.39 Ma) is recognized at the top of the fossiliferous
sequence, while the level of the proposed base of the Pleistocene
may be interpolated to lie somewhat below the fossil beds and
slightly above an exposure of the KBS Tuff (1.88 Ma). The PlioPleistocene interval is also documented in Djibouti, at the mouth
of the Afar Rift, where a diverse large-mammal fauna at Anabo
Koma, with Olduvai Bed I affinities, is dated to 1.9 Ma (de Bonis
et al., 1988), and an Acheulean butchering site at Barogali, with
remains of Homo erectus, has been given an age of 1.5 Ma
(Amosseetal., 1991).
the physical stratigraphic problems, and as a result, some of the

key horizons were initially miscorrelated (Ceding et al., 1979;
Brown and Ceding, 1982). On the basis of geochemical
"fingerprinting," which identifies key tuffs occurring in both the
Koobi Fora and Shungura formations, Brown and Ceding (1982)
proposed a revised scheme that has culminated in a major
revision of the stratigraphic nomenclature (Brown and Feibel,
1986). This is backed by a substantial number of radiometric
dates (McDougall, 1985; Feibel et al., 1989). Paleomagnetic data
initially were somewhat inconclusive (Brock and Isaac, 1976),
but now confirm the revised stratigraphy and correlations
(Hillhouse, Ceding, and Brown, 1986; Feibel et al., 1989).
The Koobi Fora hominids and a portion of the mammalian
Lower Onto River basin, Ethiopia
fauna have been described in monographs (Leakey and Leakey,
The Omo River drains into the north end of Lake Turkana 1978; Harris 1983). The Okote and Morutot tuffs (parts of a
(formerly Lake Rudolf). The lower reaches of the river within closely spaced complex that also includes the lower Koobi Fora
Ethiopia are flanked by more than 800 m of tilted and faulted Tuff) are radiometrically dated to 1.64 and 1.65 Ma, respectively,
fossiliferous sediments of the Shungura Formation; there are and are geochemically correlated with tuffs in the lower part of
also two isolated areas where the Usno and Mursi formations are Member J of the Shungura Formation, just above the Olduvai
exposed (Brown, de Heinzelin, and Howell, 1970). The geologi- subchron and thus close to the Pliocene-Pleistocene boundary
cal setting has been described in detail (de Heinzelin, 1983), and (Figure 27.2). The KBS Tuff, which is dated firmly at 1.88 Ma
the fauna has been reviewed (Coppens and Howell, 1985). and is physically correlated with Shungura Tuff H-2, is at the
Numerous tuffs are interbedded with the Shungura strata, the base of the Olduvai normal-polarity interval at Omo.
most prominent of which are used as marker horizons to divide
Prospecting on the west side of Lake Turkana has disclosed a
the sequence into alphabetic members, with the respective sequence of deposits generally similar to those of the Omo and
"marker" at the base of the member. Many radiometric ages Koobi Fora areas, and that has led to the discovery of a
between 5 Ma and 0.5 Ma and a detailed paleomagnetic study remarkable partial skeleton of an australopithecine (Walker et
provide a very well controlled section that has become the al., 1986). Named the Nachukui Formation, the deposits are
reference standard for the Pliocene and early-to-middle Pleisto- divided into members by prominent tuffs, several of which can
cene of East Africa (Brown et al., 1985; Feibel, Brown, and be correlated geochemically with those of the Koobi Fora
McDougall, 1989). The top of the Olduvai subchron almost Formation (Harris, Brown, and Leakey, 1988). The Natoo
coincides with Tuff J of the Shungura Formation, so that the Member has the Lower Koobi Fora Tuff at its base - one of the
boundary between Members H and J (there is no member I) is units of the Okote complex, dated to 1.63 Ma - and thus nearly
effectively very close to the Pliocene-Pleistocene boundary. The coincides with the Pliocene-Pleistocene boundary (Figure 27.2).
Jaramillo subchron has not been recorded and is considered to be The paleogeography of the whole Turkana Basin has been
slightly younger than the top of Member L. The succession is set outlined (Brown and Feibel, 1988) and shows that the basin was
out in the general correlation diagram in Figure 27.2.
not occupied continuously by a lake.
In the northeast corner of the basin, near the Fejej police post,
a fossiliferous sequence has been described by Asfaw et al.
(1991, 1993) straddling the Pliocene-Pleistocene boundary.
Olduvai, Laetoli, and Lake Natron
Locality FJ-5, with an abundant fauna, is dated to 1.7 Ma by the
presence of the Koobi Fora Orange Tuff, while FJ-1, with both North of Lake Eyasi in Tanzania are two important sets of beds
abundant fossils and tools, is dated to 1.9 Ma on the basis of its representing a single succession from the Lower Pliocene to the
relationship to a tuff identified with Shungura Tuff H-l (Asfaw et uppermost Pleistocene. The classic section at Olduvai Gorge
al.,1993).
comprises about 100 m of sediments divided by Reck (1914) into
a series of "Beds," numbered I, II, III, IV, and V. The geology
has been discussed in detail by Hay (1976). The Olduvai beds
Lake Turkana basin
were the first sedimentary strata in Africa to be dated raThe fossil-rich Koobi Foora Formation is exposed in extensive diometrically (Leakey, Evernden, and Curtis, 1961) and the first
badlands on the northeast side of Lake Turkana (Bowen and to be analyzed for remanent paleomagnetism. The originally
Vondra, 1973). The aggregate thickness of the formation is about measured ages for the upper and lower limits of the normally
450 m, but there are breaks within the sequence and wide magnetized zone in Bed I and lowermost Bed II, termed the
alluviated gaps between different exposure areas that render "Olduvai event" by Gromme and Hay (1971), were 1.65 Ma and
correlation difficult (White et al., 1981). Tuffs are as common as 1.8 Ma, respectively, based on 57 K/Ar ages. These limits have
in the Shungura Formation, but their use has been hindered by been revised by subsequent laser-fusion Âr/Âr analysis to 1.75
Age
m.y.
MIDDLE & UPPER

AWASH & AFAR
OMO BASIN
|
KIBISH
WEST
TURKANA
EAST
TURKANA
OLDUVAI
LAKE
NATRON
LAETOLI
LAKE
BARINGO
W
E
WESTERN
RIFT
OTH ER
FM
Silbo
MASEKl
/
Bed IV I
a
0
Bed HI
SAILIE
NDUTU
6
1/5
Chari
Upper
Bed II
Okote
uplex
Lr. Bed II
KBS
Bed I
tûr
..P
*'
r
Burg1..
.. c
"u 19/26"
- u 10
Tulu Bor
Lokochot
- K E Y -
Basal
f
F o o t p r i n t
t u f f
"Cinder
M a r k e r
t u f f s
Tuff"
MURSI
FM
Figure 27.2. Provisional correlation of the main fossiliferous stratigraphic units of PlioPleistocene age in East Africa. Most of the sequences are controlled by radiometric dates and/or
paleomagnetic records (age values are those of the referenced studies; for orbitally tuned values,
see the Preface to this volume). Faunal data are also employed in correlation between sites.
258
H. Basil S. Cooke
and 2.0 Ma by Walter et al. (1991), closely comparable to the

values obtained by astronomical calibration, as noted in the
Preface to this volume.
In the type locality for the Olduvai subchron, the top is now
identified within the Lemuta Member, which forms the upper
part of Lower Bed II (Michael, 1992). The Lemuta Member
represents an accumulation of largely eolian material deposited
during a relatively dry interval and truncated by a regional
disconformity marking the base of Upper Bed II. The marked
change in the fauna above the Lemuta Member was associated
with a change in the environment, but that does not seem to have
been a regional trend. At Laetoli, to the south of Olduvai, the
Laetolil Formation contains important hominid fossils, as well as
a hominid trackway preserved just below an ash horizon dated at
3.6 Ma (Leakey, 1981; Drake and Curtis, 1987; Hay, 1987). The
lower part of the succeeding Ndolanya unit is barren, but the
upper part has a good fauna, overlain unconformably by the
Ogol lavas, dated to 2.4 Ma. The unconformably overlying
Naibadad beds have a fauna comparable with that of Olduvai
Upper Bed II. The Pliocene-Pleistocene boundary would thus
occur in the Laetoli section within the erosion interval below the
Naibadad beds.
Some 150 km to the northeast of Olduvai Gorge lies Lake
Natron, on the west side of which there is a series of largely
lacustrine sediments known as the Peninj Group, from which a
fine australopithecine mandible and numerous artifacts were
collected in the lower unit, the Humbu Formation (Isaac, 1967).
That unit contains a basalt within a normal-polarity interval,
whereas the sediments above and below it are magnetically
reversed. Thouveny and Taieb (1987) suggest that the basalt was
extruded during the Olduvai subchron, which is in agreement
with fossil evidence that the main Peninj fauna, just above the
normal-polarity zone, is comparable to that of the middle of Bed
II at Olduvai (Gentry and Gentry, 1978; Geraads, 1987). A
basalt near the base of the overlying Moinik Formation is fairly
well dated at 1.38-1.33 Ma. It is thus probable that the main
fauna of the Peninj Group is early Pleistocene in age, dating to
about 1.6 Ma.
al., 1971; Bishop, Hill, and Pickford, 1978), as well as artifacts

from an occupation site with suggestive traces of fire (Gowlett et
al., 1981). Both the Chemoigut and the Chesowanja beds are
Lower Pleistocene.
Western Kenya, Uganda, and Zaire
At the foot of the Homa volcano in the Winam (= Kavirondo)

Gulf in western Kenya, Pliocene and Pleistocene strata crop out
in lakeside gullies at Kanam and Kanjera. One of the Kanam
sections, Kanam West, has yielded a distinctive fauna that
Pickford et al. (1990) considered to be earliest Pliocene and
correlative to the lower part of the Kaiso Series (discussed later).
Strata resting on the Pliocene have a mammal fauna correlative
to upper Bed II at Olduvai and include remains of Homo erectus.
The Kaiso Formation occurs in a belt east of the Semliki River
from the type area on Lake Mobutu (formerly Lake Albert) to
eastern Lake Rutanzige (formerly Lake Edward) (Bishop,
1965), as well as in the Kazinga Channel (Krommenhoek, 1969)
and on the northwestern side of Lake Mobutu at Sinda-Mohari
in Zaire. The deposits are mainly ferruginous and calcareous
oolitic beds deposited in well-aerated fresh-water swamps; they
have a good molluscan fauna in which Gautier (1970) recognized
7 or 8 age-related associations. The various collections of fossil
mammals from a number of localities originally were divided into
an "earlier" and a "later" faunal association (Cooke and
Coryndon, 1970), but recently Pickford et al. (1988, 1990) have
shown that the sequence is more complex. Two "earlier" units,
the Nkondo Beds and Warwire Beds, containing faunas similar
to that of Kanam West, have been distinguished with ages
estimated at around 5-3 Ma. Two of the Turkana Basin tuffs, the
Lomugol (3.60 Ma) and the Lokochot (3.40 Ma), have been
identified in the Warwire Beds (Pickford et al., 1991). Overlying
these deposits unconformably, the Kaiso Village beds, correlative to the Kyeoro and Hohwa formations, have a fauna
equivalent to, or slightly older than, that of Olduvai Bed I. The
succeeding Museta beds contain fossils and artifacts comparable
to those in Upper Bed II at Olduvai (Pickford et al., 1990), and a
tuff layer in correlative strata at Kagusa has been chemically
identified as the lower Natoo Tuff of West Turkana, dated to
about 1.5 Ma (Pickford et al., 1991). The Museta beds,
Lake Baringo basin
representing the Lower Pleistocene in Uganda, appear to be part
The important sequence exposed on both sides of Lake Baringo of what was originally named the "Epi-Kaiso" or the Kanda
consists of about 3,000 m of well-dated volcanics with seven Formation (Cooke and Coryndon, 1970). In Zaire, flanking the
intercalated fossil-bearing units ranging in age from Miocene to upper Semliki River, artifacts have been found in the Lusso
mid-Pleistocene (Chapman and Brook, 1978; Hill, Curtis, and beds, which correspond faunally to the Kaiso Village beds,
Drake, 1986). The lower part of the Chemeron Formation has a whereas the overlying Semliki beds are probably equivalent to
fauna that is probably early Pliocene to mid-Pliocene in age, but the Katanda Formation (Boaz, 1990). The Pliocene-Pleistocene
finds of a younger fauna suggest that the formation may range up boundary, based on the foregoing correlations of fossils and
to the equivalent of lower Bed II of the Olduvai sequence, of tuffs, is just older than the Museta beds, while the Kaiso Village
earliest Pleistocene age. On the east side of Lake Baringo, the and Lusso beds are of late Pliocene age.
Chemoigut Formation has furnished Oldowan tools and a
hominid fragment. It is overlain by the Chesowanja Formation,
Southern African Plio-Pleistocene deposits
the lower basalt of which is dated at 1.42 Ma (Hooker and Miller,
1979). The Chesowanja beds are of limited extent, but have In northern Malawi, the Chiwondo beds (Clark, Stevens, and
furnished a partial cranium of an australopithecine (Carney et Coryndon, 1966) contain numerous horizons with fossil mam-
mals that suggest a time range from the early or middle Pliocene
to the earliest Pleistocene (Kaufulu, Vrba, and White, 1981;
Bromage, Schrenk, and Juwaweyi, 1995). They are overlain
unconformably by the unfossiliferous but artifact-rich Chitimwe
beds of middle or late Pleistocene age; artifacts have been
recovered from the paleosurface at the contact, but despite
hopeful claims (Kaufulu and Stern, 1987; Clark, 1990), none
have been clearly documented from within the Chiwondo beds
(Juwaweyi and Betzler, 1995). Analysis of the age-diagnostic
large-mammal taxa in the Chiwondo fossil assemblages, constrained by geology, indicates three main paleontological levels:
unit 2, with a fauna dated to 4.0 Ma or earlier; unit 3A, with
fossils (including a mandible of primitive Homo) which indicate
ages between 3.76 Ma and 2.0 Ma or younger; and unit 3B, with
a fauna of 1.6 Ma to 1.5 Ma (Bromage et al., 1995; Betzler and
Ring, 1995). The Pliocene-Pleistocene boundary appears to fall
between 3A and 3B.
259
adhering to the north wall of the original cave, and a "Lower

Bank" unit on the floor. The space was then filled by the brown
breccia facies of Member 2 and calcified before another cycle of
erosion cut a deep gully, which was then filled with Member 3.
Although there were some changes in the fauna and the artifacts,
those three members do not appear to differ very greatly in age.
Similar cycles led to the formation of Member 4, with Middle
Stone Age tools, and Member 5, of late Pleistocene or Holocene
age. At Kromdraai, there are two sites, a "faunal site"
(Kromdraai A) and the "australopithecine locality" (Kromdraai
B), which are not of the same age. Excavations at Kromdraai B
disclosed two areas, not demonstrably connected, now termed
Kromdraai B West (KBW) and Kromdraai B East (KBE). At
KBE, Partridge (1982) named five members, of which Member 3
is the source of the hominid material and of the fauna (Vrba,
1982).
The faunas from all the sites have been discussed by Brain
(1981), although not from the point of view of dating. Various
analyses of the faunas have indicated that the Makapansgat and
Transvaal caves
Sterkfontein are the oldest and correspond best with the lower
The so-called cave breccias of the Transvaal and northern Cape part of the Shungura Formation in East Africa, whereas the
Province are important sources of fossil mammals, especially of Swartkrans Member 1 and Kromdraai KBE faunas have closer
the extinct hominids known as the Australopithecinae. The first affinities with that of Bed II at Olduvai or that of the upperspecimen to be found was a fissure filling in a secondary middle part of the Shungura succession (Cooke, 1974; Vrba, 1975,
limestone tufa in the Buxton-Norlim quarry at Taung, some 130 1976, 1988; Harris and White, 1979; Delson, 1984, 1988). The
km north of Kimberley; the age probably is late Pliocene, but magnetic properties of the cave breccias are far from ideal for
dating and correlation are difficult (Cooke, 1990). The major paleomagnetic studies, but results at Makapansgat (McFadden
localities are in the Sterkfontein area, 50 km west of Johannes- and Brock, 1984) show reversed polarity in Member 1 and lower
burg, where the three most important sites (Sterkfontein, 2, whereas upper 2 is normal. Member 3 (which is the main source
Swartkrans, and Kromdraai) are separated by only 1-2 km. of the fossils) is magnetically poor, but Member 4 shows two
Another significant occurrence is at Makapansgat, 250 km narrow reversed zones followed by a normal sequence. In
northeast of Johannesburg. All those Transvaal breccias are conjunction with the faunal evidence, that suggests an age in the
essentially infillings of subsurface caves and fissures, into which middle of the Gauss normal-polarity chron involving the Kaena or
external soil was carried and then firmly cemented. Erosion has Mammoth subchron, providing an age close to 3 Ma. However,
removed most of the former cave roofs, so that the pink and the possibility of lateral facies variation makes precision impossibrown breccias, locally rich in bone, are now exposed at the ble. Samples from Sterkfontein Members 2, 3, and 4 are
surface. The process of formation has been clearly set out by magnetically rather unsatisfactory. They show mainly normal
polarity, but there are several intermediate or reversed polarities,
Brain (1958, 1993).
The stratigraphy of the Sterkfontein deposit was formalized by and the bestfitis probably in the middle to upper part of the Gauss
Partridge (1978), who distinguished several successive members (Jones, Brock, and McFadden, 1986). The samples from
(1-4), starting with a basal travertine, as well as a younger unit Swartkrans are unreliable. Kromdraai KBE shows predominantly
(Member 5) separated by a substantial interval. Deposits of reversed polarity and is likely to lie within the Matuyama chron
unknown age have been reported from deeper levels in the cave (Jones etal., 1986).
system (Wilkinson, 1985). At Makapansgat, Partridge (1979)
A tentative correlation of the breccia sites is given in Figure
proposed a division similar to (but not correlated with) that of 27.3. The Pliocene-Pleistocene boundary of Vrica would seem
Sterkfontein, again with four successive members and a later to lie close to Member 5 at Makapansgat and Sterkfontein, and
fifth unit. Maguire (1985) suggested that in some cases facies perhaps within Swartkrans Member 1. Kromdraai KBE may be
changes have been confused with stratigraphic succession. The very late Pliocene, and Sterkfontein and Makapansgat midcomplexity of the cave deposits is well illustrated at Swartkrans, Pliocene.
where two members, separated by extensive erosion, were
originally distinguished (Butzer, 1976; Brain, 1981). Subsequent
Climatic and environmental changes
work has led to the recognition of five units (Brain, 1985, 1988,
1993; Brain et al., 1988). Member 1, which contains abundant In sub-Saharan Africa as a whole, the fossil faunas represent
hominid and other fossils, as well as "Developed Oldowan" varying aspects of the mosaic of grassland, open bush, and
tools, was heavily eroded, leaving a "Hanging Remnant" unit gallery forest that characterizes the region today. In the
H. Basil S. Cooke
260
Age
m.y.
OMO
STERKFONTEIN
MAKAPANSGAT
SWARTKRANS
KROMDRAAI
OTHER
FLORISBAD
BROKEN
HILL
KBE
D
4
..4.-.
Figure 27.3. Tentative correlation of

the main fossiliferous sequences or sites
in southern Africa. There are no reliable radiometric ages, and the correlation and relative dating rest mainly on
faunal comparisons with the wellcontrolled East African sequences (typified by Omo). The available
paleomagnetic evidence is subject to
ambiguous interpretations.
collections from each location there are fluctuations in the

proportions of animals that preferred one or another of those
three habitats, and there were changes in the local ecological
conditions from time to time in each area. Although they
probably were due to climate changes, there is no clear evidence
of major cyclic events. In many of the major faunal groups, the
most marked faunal turnover took place at approximately 2.52.4 Ma. The small mammals at Omo show a rapid shift from
more mesic to more xeric conditions within Member D
(Wesselman, 1984), and that is consistent with the general nature
of the change at that time. Palynological data support the
observed trend (Bonnefille, 1976, 1983, 1984; Bonnefille and
Vincens, 1985). The Bovidae have provided useful analyses, and
the case for a global 2.5-Ma "event" reflected in the continental
environments of Africa is well made by Vrba (1988, 1992). The
genus Equus made its first appearance slightly later, around 2.3
Ma. Bovidae have also been used to provide evidence for major
changes in local habitat preference and ecology at that time
(Vrba, 1982, 1985, 1992; Shipman and Harris, 1988; Bromage et
al., 1995). There were minor faunal turnovers at about 1.8 Ma
and again at 1.5 Ma, but nothing that can be said to mark the
Pliocene-Pleistocene boundary as a period of significant change.
There is also isotopic evidence from carbonates for dramatic
changes in rainfall in the Lake Turkana region between 2.0 and
^7-\
-1-7 2>I-
1.8 Ma, and at Olduvai a minor rainfall change at 1.8 Ma, and a
major one between 0.5 and 0.6 Ma (Cerling, Hay, and O'Neil,
1977). The change at Olduvai at 1.8 Ma was approximately
synchronous with the aforementioned faunal change, but the
latter seems to have been due to local rather than regional
trends, and its coincidence with the Quaternary boundary
adopted at Vrica probably does not signal an important event in
Africa, unlike the climatic and faunal changes at 2.5 Ma.
References
Amosse, J., Boisaubert, J.-L., Bouchez, R., Bruandet, J.-E,

Chavaillon, J., Faure, M., Guerin, C , Jeunet, A., Ma,
J.-L., Nickel, B., Piboule, M., Poupeau, G., Rey, P., and
Warzama, S. A. 1991. Le site de depecage Pleistocene
anciens a Elephas recki de Barogali (Republique de
Djibouti): chronologies relatives et datation par RPE et
spectrometrie gamma d'email dentaire. Cahiers Quatern.
16:379-399.
Aronson, J. L., Schmitt, J. J., Walter, R. C , Taieb, M.,
Tiercelin, J. J., Johanson, D. C , Naeser, C. W., and Nairn,
A. E. M. 1977. New geochronological and paleomagnetic
data for the hominid-bearing Hadar Formation of Ethiopia. Nature 267:323-327.
Aronson, J. L., Walter, R. C , and Taieb, M. 1983. Correlation
of Tulu Bor Tuff at Koobi Fora with the Sidi Hakoma Tuff
at Hadar (with reply by F. H. Brown). Nature 306:209210.

Asfaw, B., Beyene, Y., Semaw, S., Suwa, G., White, T., and
Woldegabriel, G. 1991. Fejej: a new paleoanthropological
research area in Ethiopia. /. Hum. Evol. 21:137-143.
Asfaw, B., Beyene, Y., Semaw, S., Suwa, G., White, T., and
Woldegabriel, G. 1993. Tephra from Fejej, Ethiopia - a
reply. /. Hum. Evol. 25:519-522.
Asfaw, B., Beyene, Y , Suwa, G., Walter, R. C , White, T.,
Woldegabriel, G., and Yemane, T. 1992. The earliest
Acheulean from Konso-Gardula. Nature 360:732-735.
Betzler, C , and Ring, U. 1995. Geology of the Malawi Rift:
facies and stratigraphy of the Chiwondo Beds, northern
Malawi. J. Hum. Evol. 28:23-36.
Bishop, W. W. 1965. Quaternary geology and geomorpholoy of
the Albert-Edward rift valley. Geol. Soc. Am., Sp. Pap.
84:293-321.
Bishop, W. W., Hill, A., and Pickford, M. 1978. Chesowanja: a
revised geological interpretation. In Geological background to fossil man, ed. W. W. Bishop, pp. 309-327.
Edinburgh: Scottish Academic Press.
Boaz, N. (ed.) 1990. Evolution of environments and Hominidae
in the African Rift valley. Memoir no. 1. Norfolk: Virginia
Museum of Natural History.
Bonnefille, R. 1976. Palynological evidence for an important
change in the vegetation of the Omo basin between 2.5
and 2 million years ago. In Earliest man and environments
in the Lake Rudolf Basin, ed. Y Coppens et al., pp. 421431. Chicago: University of Chicago Press.
Bonnefille, R. 1983. Evidence for a cooler and drier climate in
the Ethiopian uplands towards 2.5 M.Y. ago. Nature
303:487-491.
Bonnefille, R. 1984. Palynological research at Olduvai Gorge.
Nat. Geogr. Soc, Res. Repts. 17:227-243.
Bonnefille, R., and Vincens, A. 1985. Apport de la palynologie
a l'environment des hominides d'Afrique orientale. In
L Environment des hominides au Plio-Pleistocene, pp.
237-278. Paris: Fondation Singer-Polignac.
Bowen, B. E., and Vondra, C. F. 1973. Stratigraphical relationships of the Plio-Pleistocene deposits, East Rudolf,
Kenya. Nature 242:391-393.
Brain, C. K. 1958. Some aspects of the South African australopithecine sites and their accumulations. In Early
hominids of Africa, ed. C. Jolly, pp. 130-161. London:
Duckworth.
Brain, C. K. 1981. The hunters or the hunted? Chicago: Chicago
University Press.
Brain, C. K. 1985. Cultural and taphonomic comparisons of
hominids from Swartkrans and Sterkfontein. In Ancestors:
the hard evidence, ed. E. Delson, pp. 72-75. New York:
Alan Liss.
Brain, C. K. 1988. New information from the Swartkrans cave of
relevance to "robust" australopithecines. In Evolutionary
history of the "robust" australopithecines, ed. F. E. Grine,
pp. 311-316. New York: Aldine de Gruyter.
Brain, C. K. (ed.) 1993. Swartkrans-a cave's chronicle of early
man. Monograph 8. Pretoria: Transvaal Museum.
Brain, C. K., Churcher, C. S., Clark, J. D., Grine, F. E.,
Shipman, P., Susman, R. L., Turner, A., and Watson, V.
1988. New evidence of early hominids, their cultures and
environment from Swartkrans cave, South Africa. S. Afr.
J. Sci. 84:828-835.
Brock, A., and Isaac, G. LI. 1976. Reversal stratigraphy and its
application at East Rudolf. In Earliest man and environments in the Lake Rudolf Basin, ed. Y Coppens et al., pp.
148-162. Chicago: University of Chicago Press.
261
Bromage, T. G., Schrenk, F., and Juwaweyi, Y. M. 1995.

Paleobiogeography of the Malawi Rift: age and vertebrate
paleontology of the Chiwondo Beds, northern Malawi. /.
Hum. Evol. 28:37-58.
Brown, R H. 1982. Tulu Bor Tuff at Koobi Fora correlated with
the Sidi Hakoma Tuff at Hadar. Nature 300:631-633.
Brown, F. H., and Ceding, T. E. 1982. Stratigraphic significance
of the Tulu Bor Tuff of the Koobi Fora Formation. Nature
299:212-215.
Brown, F. H., de Heinzelin, J., and Howell, F. C. 1970. Pliocene/
Pleistocene formations in the lower Omo basin, southern
Ethiopia. Quaternaria 13:247-268.
Brown, F. H., and Feibel, C. S. 1986. Revision of lithostratigraphic nomenclature in the Koobi Fora region, Kenya.
Geol. Soc. Lond., J. 143:297-310.
Brown, F. H., and Feibel, C. S. 1988. "Robust" hominids and
Plio-Pleistocene paleogeography of the Turkana Basin,
Kenya and Ethiopia. In Evolutionary history of the
"robust" australopithecines, ed. F. E. Grine, pp. 325-341.
New York: Aldine de Gruyter.
Brown, F. H., McDougall, I., Davies, T., and Maier, R. 1985.
An Integrated Plio-Pleistocene chronology for the Turkana Basin. In Ancestors: the hard evidence, ed. E.
Delson, pp. 82-90. New York: Alan R. Liss.
Brown, F. H., Shuey, R. T., and Croes, M. K. 1978. Magnetostratigraphy of the Shungura and Usno Formations, southwestern Ethiopia: new data and comprehensive reanalysis.
Geophys. J. Roy. Astron. Soc. (Lond.) 54:519-538.
Butzer, K. W. 1976. Lithostratigraphy of the Swartkrans
Formation. S. Afr. J. Sci. 72:136-141.
Carney, J., Hill, A., Miller, J. A., and Walker, A. 1971. Late
australopithecine from Baringo District, Kenya. Nature
230:509-514.
Cerling, T. E., and Brown, F. H. 1982. Tuffaceous marker
horizons in the Koobi Fora region and the lower Omo
valley. Nature 299:216-221.
Cerling, T. E., Brown, F. H., Cerling, B. W., Curtis, G. H., and
Drake, R. E. 1979. Preliminary correlations between the
Koobi Fora and Shungura Formations, East Africa. Nature
279:118-121.
Cerling, T. E., Hay, R. L., and O'Neil, J. R. 1977. Isotopic
evidence for dramatic climatic changes in East Africa
during the Pleistocene. Nature 267:137-138.
Chapman, G. R., and Brook, M. 1978. Chronostratigraphy of
the Baringo basin, Kenya. In Geological background to
fossil man, ed. W. W. Bishop, pp. 207-223. Edinburgh:
Scottish Academic Press.
Chavaillon, I , Chavaillon, N., Hours, F., and Piperino, M. 1979.
From the Oldowan to the Middle Stone Age at MelkaKunture (Ethiopia): understanding cultural changes.
Quaternaria 21:87-114.
Clark, J. D. 1987. Transitions: Homo erectus and the Acheulian:
the Ethiopian site of Gadem and the Middle Awash. /.
Hum. Evol. 16:809-826.
Clark, J. D. 1990. The earliest cultural evidences of hominids in
southern and south central Africa. In From apes to angels:
essays in anthropology in honor of Phillip V. Tobias, ed. G.
Sperber, pp. 1-15. New York: Wiley-Liss.
Clark, J. D., Asfaw, B., Assefa, G., Harris, J. W. K., Kurasina,
R., Walter, R. C , White, T. D., and Williams, M. A. J.
1984. Paleoanthropological discoveries in the Middle
Awash Valley, Ethiopia. Nature 307:423-428.
Clark, J. D., Stevens, E. A., and Coryndon, S. C. 1966.
Pleistocene fossiliferous lake beds of the Malawi (Nyasa)
Rift: a preliminary report. Amer. Anthropol., Spec. Publ.
68:46-87.
262
H. Basil S. Cooke
Conroy, G. C , Jolly, C. I , Cramer, D., and Kalb, J. E. 1978. Haq, B. U , Berggren, W. A., and Van Couvering, J. A. 1977.
Newly discovered fossil hominid skull from the Afar
Corrected age of the Pliocene/Pleistocene boundary.
depression, Ethiopia. Nature 275:67-70.
Nature 269:483-488.
Cooke, H. B. S. 1974. Plio-Pleistocene deposits and mammalian Harris, J. M. (ed.) 1983. Koobi Fora Research Project, Volume 2.
The fossil ungulates: Proboscidea, Perisodactyla and
faunas of eastern and southern Africa. In Ve Congres
Suidae. Oxford: Clarendon Press.
Neogene Mediterraneen Lyon, 1971, pp. 88-108. Memoir
Harris, J. M., Brown, F. H., and Leakey, M. G. 1988.
78. Paris: Bur. Rech. Geol. Min.
Cooke, H. B. S. 1990. Taung fossils in the University of
Stratigraphy and paleontology of Pliocene and Pleistocene
California collections. In From apes to angels: essays in
localities west of Lake Turkana, Kenya. Contributions in
Science, no. 399. Los Angeles: Natural History Museum of
anthropology in honor of Phillip V. Tobias, ed. G. Sperber,
Los Angeles County.
pp. 119-134. New York: Wiley-Liss.
Cooke, H. B. S., and Coryndon, S. C. 1970. Pleistocene Harris, J. M., and White, T. D. 1979. Evolution of the PlioPleistocene African Suidae. Transactions, no. 691. Philadelmammals from the Kaiso Formation and related deposits
phia: American Philosophical Society.
in Uganda. In Fossil vertebrates of Africa, vol. 2, ed. L. S.
B. Leakey, pp. 107-224. London: Academic Press.
Hay, R. L. 1976. Geology of the Olduvai Gorge. Berkeley:
Coppens, Y., and Howell, F. C. (eds.) 1985. Les Faunes PlioUniversity of California Press.
Pleistocene de la Basse Vallee de VOmo (Ethiopie). VolumeHay, R. L. 1987. Geology of the Laetoli area. In Laetoli: a
2: Les fclephantides (Mammalia, Proboscidea)v Paris:
Pliocene site in northeast Tanzania, ed. M. D. Leakey and
Centre National de Recherches Scientifiques.
J. M. Harris, pp. 23-47. Oxford: Clarendon Press.
Cressier, P. 1980. Magnetostratigraphie du gisement Pleistocene Hill, A., Curtis, G., and Drake, R. 1986. Sedimentary
de Melka-Kunture (Ethiopie). Datation des niveaux
stratigraphy of the Tugen Hills, Baringo, Kenya. In
Sedimentation in the African Rifts, ed. L. E. Frostick et al.,
oldowayens at acheuleeens. Unpublished These geopp. 285-295. London: Geological Society (Special Publicaphysique. Strasbourg: Universite Louis Pasteur.
tion 25).
de Bonis, L., Geraads, D., Jaeger, J.-J., and Sen, S. 1988.
Vertebres du Pleistocene de Djibouti. Soc. geol. Fr., Bull., Hillhouse, J. W., Ceding, T. E., and Brown, F. H. 1986.
Magnetostratigraphy of the Koobi Fora Formation, Lake
ser. 8 4:323-334.
Turkana, Kenya. /. Geophys. Res. 91:11,581-11,595.
de Heinzelen, J. 1983. The Omo Group: archives of the
International Omo Research Expedition. Annales, Ser. 8, Hooker, P. J., and Miller, J. A. 1979. K-Ar dating of Pleistocene
Sciences geologique, no. 85. vol. 1, text; vol. 2, maps.
fossil hominid site at Chesowanja, North Kenya. Nature
Tervuren: Musee Royal de l'Afrique Centrale.
282:710-712.
Delson, E. 1984. Cercopithecid biochronology of the African Isaac, G. LI. 1967. The stratigraphy of the Peninj Group - early
Plio-Pleistocene: correlation among eastern and southern
middle Pleistocene formation west of Lake Natron,
hominid-bearing localities. Forschungs-Inst. Senckenb.,
Tanzania. In Background to evolution in Africa, ed. W. W.
Cour. 69:199-218.
Bishop and J. D. Clark, pp. 229-257. Chicago: University
Delson, E. 1988. Chronology of South African australopith site
of Chicago Press.
units. In Evolutionary history of the "robust" aus- Johanson, D. C , Taieb, M., and Coppens, Y. 1982. Pliocene
tralopithecines, ed. F. E. Grine, pp. 317-324. New York:
hominids from the Hadar Formation, Ethiopia (1973Aldine de Gruyter.
1977): stratigraphic, chronologic, and paleoenvironmental
Drake, R., and Curtis, G. H. 1987. Geochronology of the
contexts, with notes on hominid morphology and systemLaetoli fossil localities. In Laetoli: a Pliocene site in
atics. Amer. J. Phys. Anthropol. 57:373-402.
northeast Tanzania, ed. M. D. Leakey and J. M. Harris, Jones, D. L., Brock, A., and McFadden, P. L. 1986.
pp. 48-52. Oxford: Clarendon Press.
Paleomagnetic results from the Kromdraai and SterkfonFeibel, C. S., Brown, F. H., and McDougall, I. 1989. Stratitein hominid sites. S. Afr. J. Sci. 82:160-163.
graphic context of fossil hominids from the Omo Group Juwaweyi, Y M., and Betzler, C. 1995. Archaeology of the
deposits, northern Turkana Basin, Kenya and Ethiopia.
Malawi Rift: the search continues for Early Stone Age
Am. J. Phys. Anthropol. 78:595-622.
occurrences in the Chiwondo Beds, northern Malawi. /.
Hum. Evol. 28:115-116.
Gautier, A. 1970. Fossil fresh water mollusca of the Lake AlbertLake Edward Rift (Uganda). Annales, Ser. 8, Sciences Kalb, J. E., and Mebrate, A. 1993. Fossil elephantoids from the
Geologiques, no. 67. Tervuren: Musee Royal de l'Afrique
hominid-bearing Awash Group, Middle Awash Valley,
Centrale.
Afar Depression, Ethiopia. Am. Philos. Soc, Trans.
Gentry, A. W., and Gentry, A. 1978. Fossil Bovidae (Mammalia) of Olduvai Gorge, Tanzania. Part II. Br. Mus. Nat. Kalb, J. E., Oswald, E. B., Tebedge, S., Mebrate, A., Tola, E.,
Hist., Geol. Ser., Bull. 30:1-63.
and Peak, D. 1982. Geology and stratigraphy of Neogene
Geraads, D. 1987. La faune des depots Pleistocenes de l'ouest du
deposits, Middle Awash Valley, Ethiopia. Nature 298:17Lac Natron (Tanzanie); interpretation biostratigraphique.
25.
Sci. Geol., Bull. 40:167-184.
Kalb, J. E., Wood, C. B., Smart, C , Oswald, E. B., Mebrate, A.,
Gowlett, J. A. J., Harris, J. W. K., Walton, D., and Wood, B. A.
Tebedge, S., and Whitehead, P. 1980. Preliminary geology
1981. Early archaeological sites, hominid remains and
and palaeontology of the Bodo D'Ar site, Afar, Ethiopia.
traces of fire from Chesowanja, Kenya. Nature 294:125Palaeogeogr. Palaeoclimatol. Palaeoecol. 30:107-120.
129.
Kaufulu, Z. M., and Stern, N. 1987. The first stone artifacts to
Gromme, C. S., and Hay, R. L. 1971. Geomagnetic polarity
be found in situ within the Plio-Pleistocene Chiwondo
epochs: age and duration of the Olduvai normal-polarity
Beds in northern Malawi. /. Hum. Evol. 16:129-1 A0.
event. Earth Planet. Sci. Lett. 10:179-185.
Kaufulu, Z., Vrba, E. S., and White, T. D. 1981. Age of the
Haileab, B., and Brown, F. H. 1992. Turkana Basin-Middle
Chiwondo Beds, Northern Malawi. Transv. Mus., Ann.
Awash Valley correlations and the age of the Sagantole
33:1-8.
and Hadar formations. /. Hum. Evol. 22:453-468.
Krommenhoek, W. 1969. Mammalian fossils from the Pleisto-
263
cene of Lake Edward and the Kazinga Channel. Uganda J.

paleoecology of Australopithecus boisei in eastern Africa.
In Evolutionary history of the "robust" australopithecines,
33:79-84.
ed. F. E. Grine, pp. 342-382. New York: Aldine de
Leakey, L. S. B., Evernden, J. R, and Curtis, G. H. 1961. The
Gruyter.
age of Bed I Olduvai Gorge, Tanganyika. Nature 191:478479.
Thouveny, N., and Taieb, M. 1987. Etude paleomagnetique des
formations du Plio-Pleistocene de la region de la Peninj
Leakey, M. D. 1981. Tracks and tools. Roy. Soc. Lond., Phil.
(Ouest du Lac Natron, Tanzanie): limites de InterTrans. B292:95-102.
pretation magnetostratigraphique. Sci. Geol., Bull. (StrasLeakey, M. G., and Leakey, R. E. F. (eds.) 1978. Koobi Fora
bourg) 40:57-70.
Research Project, Vol. 1: The fossil hominids and an
introduction to their context, 1968-1974. Oxford: Claren- Tiercelin, J. J. 1986. The Pliocene Hadar Formation, Afar
don Press.
Depression of Ethiopia. In Sedimentation in the African
Rifts, ed. L. E. Frostick et al., pp. 221-240. Special
Maglio, V. I , and Cooke, H. B. S. (eds.) 1978. Evolution of
Publication 250. London: Geological Society.
African mammals. Cambridge, Mass.: Harvard University
Press.
Vrba, E. S. 1975. Some evidence of chronology and palaeoecology of Sterkfontein, Swartkrans and Kromdraai from
Maguire, J. M. 1985. Recent geological, stratigraphic and
the fossil Bovidae. Nature 254:301-304.
palaeontological studies at Makapansgat Limeworks. In
Hominid evolution: past, present and future, ed. P. V. Vrba, E. S. 1976. The fossil Bovidae of Sterkfontein, Swartkrans
Tobias, pp. 151-164. New York: Alan Liss.
and Kromdraai. Memoir 21. Pretoria: Transvaal Museum.
McDougall, I. 1985. K-Ar and40Ar/39Ar dating of the hominid- Vrba, E. S. 1982. Biostratigraphy and chronology, based particularly on Bovidae, of southern African hominid-associated
bearing sequence at Koobi Fora, Lake Turkana, northern
assemblages: Makapansgat, Sterkfontein, Taung, KromKenya. Geol. Soc. Am. Bull. 96:159-175.
draai, Swartkrans; also Elandsfontein (Saldanha), Broken
McFadden, P. L., and Brock, A. 1984. Magnetostratigraphy at
Hill (now Kabwe) and Cave of Hearths. In Ier Congres
Makapansgat. S. Afr. J. Sci. 80:482-483.
International de Paleontologie humaine, pretirage, vol. 2,
Michael, W. 1992. Etude magnetostratigraphique des formations
ed. H. deLumley and M. A. deLumley, pp. 707-797. Nice:
du Plio-Pleistocene des Gorges d'Olduvai (Tanzanie).
Centre National de la Recherche scientiflque.
Univ. d'Aix-Marseilles, MemoiresD.E.A. Prehist. Anthropol. Technol.
Vrba, E. S. 1985. Ecological and adaptive changes associated
with early hominid evolution. In Ancestors: the hard
Partridge, T. C. 1978. Re-appraisal of lithostratigraphy of
evidence, ed. E. Delson, pp. 63-71. New York: Alan Liss.
Sterkfontein hominid site. Nature 275:282-287.
Partridge, T. C. 1979. Re-appraisal of lithostratigraphy of Vrba, E. S. 1988. Late Pliocene climatic events and hominid
Makapansgat Limeworks hominid site. Nature 279:484evolution. In Evolutionary history of the "robust" aus488.
tralopithecines, ed. F. E. Grine, pp. 405-426. New York:
Aldine de Gruyter.
hominids of southern Africa. In Ier Congres international de Vrba, E. S. 1992. Mammals as a key to evolutionary theory. /.
paleontologie humaine, pretirage, vol. 2, ed. H. deLumley
Mammal. 73:1-28.
and M. A. deLumley, pp. 617-676. Nice: Union Interna- Walker, A., Leakey, R. E. R, Harris, J. M., and Brown, F. H.
tionale des Sciences Prehistorique et Protohistorique.
1986. 2.5-Myr. Australopithecus boisei from west of Lake
Pickford, M. 1988. Implications de la succession des mollusques
fossiles du bassin du lac Albert (Ouganda). C. R. Acad. Walter, R. C. 1994. Age of Lucy and the First Family: singleSci. Paris, ser. 7/305:317-322.
crystal 40Ar/39Ar dating of the Denen Dora and lower
Kada Hadar members of the Hadar Formation, Ethiopia.
Pickford, M., Senut, B., Poupeau, G., Brown, R, and Haileab,
Geology 22:6-10.
B. 1991. Correlation of tephra layers from the Western
Rift Valley (Uganda) to the Turkana Basin (Ethiopia/ Walter, R. C , and Aronson, J. L. 1993. Age and source of the
Kenya) and the Gulf of Aden. C. R. Acad. sci. Paris, ser.
Sidi Hakoma Tuff, Hadar Formation, Ethiopia. /. Hum.
7/313:223-229.
Evol. 25:229-240.
Pickford, M., Senut, B., Semmanda, I., Elepu, D. E., and Walter, R. C , Manega, P. C , Hay, R. L., Drake, R. E., and
Obwona, P. 1988. Premiers resultats de la mission de
Curtis, G. H. 1991. Laser-fusion 40Ar/39Ar dating of Bed
l'Uganda Palaeontology Expedition a Nkondo (Pliocene
I, Olduvai Gorge, Tanzania. Nature 354:145-149.
du Bassin du Lac Albert, Ouganda). C. R. Acad. Sci. Wesselman, H. B. 1984. The Omo micromammals. In ContribuParis, ser. II 306:315-320.
tions to vertebrate evolution, vol. 7, ed. M. Hecht and F. S.
Szalay, pp. 97-112. Basel: Karger.
Pickford, M., Senut, B., Tiercelin, J. J., Kasande, R., and
Obwona, P. 1990. Resultats de la troisieme mission White, H. I , Burggraf, D. R., Bainbridge, R. B., and Vondra,
C. F. 1981. Hominid habitats in the Rift Valley; Part 1. In
(1988) de l'Uganda Palaeontology Expedition: regions de
Hominid sites: their geologic settings, ed. G. Rapp and C.
Nkondo-Sebugoru et de Hohwa (Bassin du Lac Albert,
Ouganda). C. R. Acad. Sci. Paris, ser. //310:737-744.
F. Vondra, pp. 57-61. Special Symposium 63. Washington:
American Association for the Advancement of Science.
Reck, H. 1914. Erste vorlaufige Mitteilung iiber den Fund eines
fossilen Menschenskelets aus Zentral-afrika. Ges. natur- White, T. D., Moore, R. V, and Suwa, G. 1984. Hadar
forsch. Freunde Berlin, Sitzungsb. 3:81-95.
biostratigraphy and hominid evolution. /. Vert. Pal. 4:575583.
Sarna-Wojcicki, A. M., Meyer, C. E., Roth, P. H., and Brown,
F. H. 1985. Ages of tuff beds at East African early hominid Wilkinson, M. J. 1985. Lower-lying and possibly older fossites and sediments of the Gulf of Aden. Nature 313:306siliferous deposits at Sterkfontein. In Hominid evolution:
308.
past, present and future, ed. P. V. Tobias, pp. 165-170.
New York: Alan Liss.
Shipman, P., and Harris, J. M. 1988. Habitat preference and
28
Plio-Pleistocene reference sections in Indonesia

FRANCOIS SEMAH
Introduction
Quaternary research in Indonesia developed rapidly beginning in

the late 1970s because of the activities of Indonesian, French,
Dutch, and Japanese teams (e.g., Saint-Marc and Suminta, 1979;
Semah et al., 1980; De Vos et al., 1982; Sartono, Semah, and
Djubiantono, 1984; Watanabe and Kadar, 1985; F. Semah,
1986). Understandably, research focused mainly on Java, where
sites with human remains have been known for more than a
century. Comparable information on the Plio-Pleistocene strata
in other areas of Indonesia is not yet available to give a complete
synthesis of the region, and this chapter is therefore restricted to
the island of Java.
The manuscript for this chapter was initially submitted in 1984,
with revisions to certain essential points in 1991. The time-scale
values have been corrected in the text according to orbitally
tuned calibration, as discussed in the editor's Preface. It was not
possible, however, to include a balanced review of all the work
that has been published in the past decade, and this report must
therefore be considered as a reflection of the state of the art in
the mid-1980s.
28.2). The most nearly complete Plio-Pleistocene sections are

exposed in the central depression (Solo zone) and near the axial
anticlinorium (Bogor zone, North Seraju and Kendeng ranges).
We shall describe here two of the areas which illustrate the
Pliocene-Pleistocene boundary, namely the Bumiayu area of
central Java south of Tegal, and the Surakarta region of central
Java on the upper Solo River.
The stratigraphic succession in these two areas is shown in
Table 28.1. Exposures, paleontology, and other features are
discussed next under separate headings.
Elevation of the Southern Mountains of Java had already
taken place before the late Cenozoic (Van Bemmelen, 1949).
The main factor in the Plio-Pleisticene history and stratigraphy
of central Java was the emergence of the axial anticlinorium, for
instance, the Kendeng Hills (e.g., Saint-Marc, Paltrinieri, and
Situmorang, 1977). The present consensus among geologists
working in Java is that tectonic uplift progressed gradually from
west to east, with most of the island being emergent by the end of
the Pliocene. Other important factors in the stratigraphy were
eustatic shifts in sea level, beginning about 2.5 Ma, and local
activity in the inner volcanic arc, which began at least 2 m.y. ago
(Bandetetal., 1989).
Geological background
Exhaustive field work by Dutch geologists during the first half of
the century was fairly well summarized by Van Bemmelen (1949)
and Marks (1957). This report follows the standard terminology
of Marks (1957) for Javanese formations, in order to maintain a
connection with previous works, without implying acceptance of
the chronostratigraphic meanings and correlations proposed by
earlier authors. In our view, the assignment of formation names
has been overdone in publications dealing with Javanese
stratigraphy, particularly with regard to the Plio-Pleistocene.
Recent work has shown that many stratigraphic units have such
limited geographic extensions, or vary in age so greatly from
place to place, that they hardly deserve the name of "formation."
This is particularly the case in the Sangiran Dome area
(discussed later), where lateral facies changes are very abrupt.
Java can be divided into several structural units, roughly
oriented E-W, following the axis of the island (Figures 28.1 and
264
Bumiayu area, western central Java
At the foot of the North Seraju mountains, in the Bumiayu

region, sections in the valleys of the Kali Glagah and Ci Saat
rivers exemplify the sequences of the Plio-Pleistocene interval.
Resting on Miocene and Lower Pliocene units, the Kalibiuk
Formation, with marine mollusks, is followed by the Kali
Glagah-Mengger-Gintung synorogenic series, in Ter Haar's
terminology (1934) (Figure 28.3). The Kalibiuk-Kali Glagah
boundary is located, in the Ci Saat Valley, at the first occurrence
of continental sediments. The Kali Glagah and Mengger
formations contain sandy, conglomeratic, and clayey layers,
whereas the Gintung has a mainly conglomeratic and volcanic
breccia aspect.
At least two vertebrate horizons have been recognized in that
area, which is the type for the so-called Kaliglagah fauna. The
lower one is situated a few tens of meters above the Kalibiuk-
- 6S
JAVA SEA
7S
- 8S
North Coast belt
INDIAN OCEAN
Rembang-Madura hills
Randublatung depression
_ 9S
Central anticlinorium
[yr'.vll
|
ri
\
||
>
]
Central depression
Southern Mountains
Volcanoes
* 1 : Gemolong
Bringinan
JAVA SE-A
Onto
Sangiran
* 2 : Trini1
* 3 : Kedungbrubus
* 4 : Penning and Mojokerto
INDIAN OCEAN
100 km
Figure 28.1 (top). Structural geology of Java. (Adapted from Van Bemmelen, 1949.)
Figure 28.2 (bottom). Locations of sections discussed in this chapter: 1, Gemolong, Bringinan, Onto,
Sangiran; 2, Trinil; 3, Kedung Brubus; 4, Perning and Mojokerto.
Francois Semah
266
Table 28.1. Summary of Plio-Pleistocene stratigraphy in the studied areas of central Java
West Central Java (Bumiayu): Ci Saat, Kali Glagah valleys
Gintung
Conglomerate.
Mengger
Continental sand, gravel, clay.
Kali Glagah
Continental sand, gravel, clay.
Kali Biuk
Marine sands and clays.
East Central Java (Surakarta): Solo River basin, Sangiran dome

Notopuro
Kabuh
Volcanic breccias.
Tuffaceous sands, clays, gravels; basal Grenzbank
conglomerate.
Pucangan
Black estuarine and paludal clay, basal lahar over

Corbicula beds.
Kalibeng
Blue clays, with littoral deposits at top.
Globigerina marls
Open-marine deposits.
Kali Glagah boundary (von Koenigswald, 1933, 1934; Marks,

1957), while the upper one is less clearly localized in the
stratigraphy. According to Sondaar (1981, 1984), the upper
horizon reflects several faunal exchanges with the mainland that
are not evident in the lower one. That might indicate one or
more glacial intervals, with sea level decline and exposure of
Sundaland interconnections.
Paleobotanic data presented by A.-M. Semah (1986) show the
dominance of a mangrove-forest association in the upper
Kalibiuk unit, with a maximum in the lignitic layer which marks
the Kalibiuk-Kali Glagah boundary along the Ci Saat River. The
pollen data show a trend to a drier climate in the upper Kalibiuk
beds; in the lower Kali Glagah, whereas the persistence of
mangrove elements indicates proximity of the shoreline, the
pollen spectra indicate development of a true rain forest on the
higher ground.
Paleomagnetic studies in the Ci Saat Valley (F. Semah, 1983,
1986) have shown a long reversed sequence (100 m thick) in the
upper Kalibiuk marine layers. This long interval appears to
correlate with the early Matuyama chron, on the basis of
preliminary micropaleontological data (Sumarso and Suparyono,
1974). Several normal sequences that appear within the Kali
Glagah series (F. Semah, 1983) are tentatively assigned to the
normal events (Reunion and Olduvai subchrons) which interrupted the reversed polarity of the Matuyama chron between
2.15 and 1.77 Ma. Higher in the section, reversed polarities

persist through the Mengger up to the Gintung unit. Up to now,
no direct geochronological dating has been carried out in the
Bumiayu area, but the paleomagnetic interpretation indicates
that the lower of the two vertebrate-fauna horizons in the Kali
Glagah dates to the early Matuyama, between 2.1 and 1.8 Ma.
Surakarta (Solo) area, eastern central Java
South of the Kendeng Hills, near the town of Surakarta, the Solo
central depression contains giant volcanoes active from the late
Pleistocene to the present, and also deeply eroded and faulted
domal uplifts that are the remnants of much older volcanic
complexes, such as Sangiran, Gemolong, Bringinan, and Onto.
In the uplifted domes, exposed Plio-Pleistocene sections are
divided into the so-called formations of Kalibeng, Pucangan,
Kabuh, and Notopuro, following the classic terminology (Marks,
1957) in that part of Java.
The Solo "formations" have little to do with the type sections
defined some 100-150 km to the northeast (Duyfjes, 1936), and
stratigraphic correlations based strictly on these lithologies are
extremely problematical. For instance, blue marine clays, which
are commonly attributed to the "Upper Kalibeng," have a late
Pliocene age at Sangiran, but were formed in other parts of the
Solo area at various times from the earliest Pleistocene
AGE/POLARITY
m.y.
BUMI A Y U
GEMOLONG
SANGIRAN
PAGEREJO
JENGGLONG
ONTO
KALI ONTO
BAPANG
CI SAAT
SUNGAI
PUREN
SOUTH KRISIK
CENGKLIK
basis of ""N
1st normal
sequence
(1.9/2.1)-/
basis of
the normal
sequence
(1.9/2.1-
Figure 28.3. Proposed correlations of Plio-Pleistocene sequences of central eastern Java.

For locations, see Figure 28.2. (Adapted from F. Semah, 1984b.)
BRINGINAN
268
Francois Semah
recognized a faunal succession within the Pucangan black clays,

equivalent to the upper Kali Glagah succession of the Bumiayu
area, with the so-called Satir and Ci Saat faunas. The basal
Grenzbank and the lower part of the Kabuh formation (in the
terminology of De Vos et al., 1982) yield a Trinil fauna, and
examples of the Kedung Brubus fauna have been collected from
the upper Kabuh.
Hominid remains from Sangiran are not precisely located in
the stratigraphy. The oldest, including the Sangiran "Meganthropus," appear in the upper Pucangan formation (Sartono, 1982)
and might be more than 1 m.y. old. A few hominid remains are
found in the Grenzbank, although that level yields very
abundant fossils of other mammals, and most of the hominid
specimens are from the lower and middle Kabuh formation. Of
the approximately 18 hominids from Sangiran, all are classed as
Homo erectus, in the wide sense.
According to A.-M. Semah (1982, 1984, 1986), the palynological record begins in the upper Kalibeng blue clays in the
Sangiran and Onto Domes, with indications of mangrove forest
and thus a nearby shoreline. Basal Pucangan layers show a
Stratigraphy
marked impoverishment of the flora, possibly due to the effects
The upper Kalibeng beds in the Sangiran-Gemolong area begin
of repeated volcanism. The palynology of the remainder of that
with blue clays containing shallow-marine mollusks and thin
formation indicates renewed diversification of near-shore swamp
tuffaceous beds. Those rest on open-marine lower Kalibeng
vegetation and a land flora that alternated between tropical rain
Globigerina marls, seen only in the Gemolong Dome (Reinhold,
forest and more open, seasonal forest (possibly corresponding to
1937). In the Sangiran area, the blue clays pass upward into
glacial and interglacial climates, respectively). The Kabuh beds
sandy littoral deposits including calcareous Turritella sands and
have an overall palynological association indicative of open
Balanus limestones, capped by a bed rich with the fresh-water
forest and seasonal rainfall. It is not definitely known whether
(swamp) mollusk Corbicula.
those variations are to be connected with local phenomena
In the Sangiran section, lahars (volcanic mudflow breccias) (tectonism, volcanism) or with slight climatic changes. Semah
mark the boundary between the Kalibeng and Pucangan beds. and Rahardjo (1984) interpreted the change in palynoflora at the
The lahars, which cover an irregular topography and range from base of the Kabuh formation to have an environmental cause,
0.5 m to more than 50 m thick (Indonesia-Japan Research parallel with the change in sedimentology, in agreement with
Cooperation Program, 1979), played an important part in Sondaar (1984), who also found indications of a drier climate in
ponding up the drainage to form a lagoon in the Solo area. The correlative layers. The Australasian tektite event, which appears
lagoonal deposits of the lower Pucangan beds reflect low-energy, to correlate with the middle Kabuh (discussed later), is associpoorly oxygenated sedimentation dominated by bluish-gray and ated in deep-sea cores with paleoclimatic indicators of glacioblack clays, with a fauna consisting mainly of fresh-water eustatic sea-level lowering and cold climate (Schneider, Kent,
mollusks {Corbicula, Viviparus, Unio), together with estuarine andMello, 1992).
intercalations characterized by marine mollusks and diatoms
(Reinhold, 1937; Itihara et al., 1985a) and several fine-grained
tuffaceous layers. The upper Pucangan beds consist of poorly
Paleomagnetic studies
stratified black clays. A drastic change is seen at the top of the
Pucangan unit, with the transition to high-energy fluviatile F. Semah (1983; Semah et al., 1980; Sartono et al., 1984) found
sedimentation of the Kabuh beds, mainly sands, gravels, and that the Notopuro and Kabuh formations were of normal
polarity and should be related to the Brunhes epoch. That
tuffs. Conspicuous cliff-forming conglomerates, or Grenzbanks
conclusion was based on study of several parallel Kabuh
("boundary beds"), mark the end of marine/lagoonal sedimentasections. Measurements made after careful thermal cleaning
tion at the base of the Kabuh in the Sangiran Dome. The Kabuh
have shown several mixed and even reversed samples, mainly in
beds are unconformably overlain by the Notopuro volcanic
the Jengglong and Pagerejo sections (F. Semah, 1986), but the
breccias and lahars.
polarities in those samples show little stratigraphic continuity or
statistical consistency. The transition from predominantly rePaleontology
versed to predominantly normal paleomagnetic polarities found
The oldest vertebrate remains in the Sangiran Dome seem to by these authors near the top of the Pucangan unit at Bapang
come from the basal Pucangan lahar, which includes much sandy (southwestern part of the Sangiran Dome) is interpreted to be a
clastic material (Van Es, 1931). Above that level, Sondaar (1984) reflection of the Matuyama-Brunhes boundary.
(Gemolong section) to the beginning of the middle Pleistocene
(Kaliuter section, 10 km north of Gemolong) (Djubiantono and
Semah, 1991). These sediments derive from the shallow-marine
environments that persisted in the Solo area until the final uplift
of the Kendeng Hills. It is obvious that to group such different
exposures on a geological map as a regional formation would be
illogical and misleading, but rather than create a separate
formation name in each exposure, we use these "formation"
names in the sense of facies units. It should be noted, however,
that ItiKara, Kadar, and Watanabe (1985a) have proposed new
names for units based on type sections in the Sangiran Dome,
namely the Puren, Sangiran, Bapang, and Pohjajar formations.
The whole of the stratigraphic sequence is not exposed in
every dome, and for the sake of clarity we shall discuss here a
synthetic section (Figure 28.3). Besides the fundamental publications noted earlier, detailed sections and descriptions of the Solo
area can be found in the works of F. Semah (1983, 1986) and
Watanabe and Kadar (1985).
Plio-Pleistocene of Indonesia
On the other hand, the report of Shimizu et al. (1985, p. 286),

which actually refers to work completed in 1981, showed a
consistent mixed-polarity interval in the lower Kabuh and the
uppermost Pucangan. That would suggest that the base of the
Brunhes might be located in their analyzed section at a level in
the middle Kabuh, very close to levels that have been radiometrically dated to about 0.71 Ma, which is in fact not greatly
different from the age of the Matuyama-Brunhes boundary
(0.78 Ma). The need for caution in lithostratigraphic correlations
in continental-volcanic sequences, even locally, has been emphasized (Lizon-Sureau, 1979; A.-M. Semah, 1984), and there is the
strong possibility that the observed Pucangan-Kabuh boundary,
with its Grenzbank facies, may actually be diachronous with
regard to the Matuyama-Brunhes polarity boundary.
The Jaramillo subchron has not been recognized in the
Sangiran Dome. In the Onto dome, however, F. Semah (1983)
observed a short doublet of normal polarity in the upper part of
the local Pucangan facies that possibly could be evidence of that
episode. F. Semah et al. (1980) found consistent reversed
polarities in samples taken below the upper beds of the Pucangan
unit in Sangiran and referred those levels to the Matuyama in the
post-Olduvai interval, a conclusion further supported by the
work of Shimizu et al. (1985).
In the basal Pucangan lahars at Sangiran the polarities are not
consistent, being reversed in some sections and normal in others.
It would seem clear that the volcanism, which is evidenced in all
parts of the Dome, was diachronous with regard to the upper
boundary of the Olduvai (F. Semah, 1986). Laterally consistent
normal polarities, seen in samples from a significant interval in
the upper Kalibeng formation (F. Semah, 1983; Shimizu et al.,
1985), are correlated with the Olduvai event. Samples in the
lower part of the blue Kalibeng clays show reversed polarity,
corresponding to the early part of the Matuyama chron. Finally,
normal-polarity samples have been taken from the base of the
blue clays and down into the Globigerina marls, which may
represent the top of the Gauss chron.
Radiometric ages
Radiometric dating at Sangiran is summarized in Table 28.2. It is
beyond the scope of this chapter to discuss all of the results, but a
few comments are in order. It must be noted that the Sangiran
sequence, despite its formation in a very active volcanic area,
includes almost no tuff deposits suitable for radiometric dating.
The age determinations we have been able to obtain are
therefore not as well substantiated as would be desirable.
Overall, the dating is in accordance with the interpretations of
the paleomagnetic succession. Among the fission-track ages
obtained by different teams in the tuffaceous layers of the
Pucangan facies, only the determinations by Suzuki et al. (1985)
require consideration, as the other published fission-track age
values would correlate the reversed-polarity Pucangan sediments
with the normal-polarity Brunhes epoch. As regards the Kabuh
facies, the most coherent dates have been obtained by FT
(fission-track) and K/Ar (potassium-argon isotope ratio) analysis
269
of tektite samples. All of the tektites dated by those two methods

are close to 0.7-0.8 Ma, even though several of the dated tektites
are from different levels. The Kabuh tektites have been
geochemically correlated to a microtektite horizon found in
deep-sea cores about 12 k.y. below the Matuyama-Brunhes
boundary (Glass and Heezen, 1967; Schneider et al., 1992). The
specimens found in the Kabuh facies at Sangiran are of large size
and rather rare. Only three specimens are reported to have been
found in situ in the middle Kabuh beds (Itihara et al., 1985a),
while the others were from higher in the sequence. Those
heavier stones have been considered by most workers to
represent specimens that remained at the primary tektite
horizon, but it is possible that all of the Kabuh tektites were
reworked from an even older level, and the dating of the middle
Kabuh on this basis is not yet established beyond question.
An age of 0.8 Ma for the middle Kabuh is too old if the tektite
horizon is within the Brunhes, according to one possible
paleomagnetic interpretation (i.e., F. Semah, 1983), but it could
agree with the findings of Shimizu et al. (1985), as noted
previously. Itihara et al. (1985a) justified placing the MatuyamaBrunhes boundary close to the level of the lowest-occurring
tektites in the middle Kabuh in the northern Sangiran exposures,
according to the fact that Curtis (1981) obtained an average K/Ar
age of 0.83 Ma on 4 middle Kabuh pumices, and Suzuki et al.
(1985) obtained an FT date of 0.78 Ma on zircon from the Kabuh
Middle Tuff, just below the lowermost tektite occurrences (Table
28.2); in modern time-scale calibrations, that dating supports
correlation to the Matuyama-Brunhes (0.78 Ma) reversal more
strongly than before.
In the long reversed-polarity interval of the Pucangan, Suzuki
et al. (1985) obtained stratigraphically consistent FT ages
between 1.16 Ma and 1.51 Ma from pyroclastic zircon. This is
appropriate for the upper Matuyama below the Jaramillo, as
noted earlier. An age of 2.99 Ma has been obtained for the lower
Kalibeng within the basal normal-polarity interval, as is appropriate for the upper Gauss. Finally, Table 28.1 shows the only age
proposed for the Notopuro lahars, at 0.25 Ma. Overall, the
dating cited here is consistent with interpretations that make the
Olduvai normal event, 1.95 Ma to 1.77 Ma, equivalent to the
normal-polarity interval in the upper Kalibeng formation.
Micropaleontology
Micropaleontological data are not as numerous in the Surakarta
region as in the Bumiayu region. According to Kadar (Watanabe
and Kadar, 1985), the foraminifera of the upper blue clays of the
Kalibeng date to the late Pliocene near the top of Zone N.21 in
the planktonic foraminifera zonation. That is not in contradiction of the interpretation that the Olduvai event, with the
Pliocene-Pleistocene boundary near its top, is represented in the
shallow-water sediments above the blue clays, at the top of the
Kalibeng facies. In eastern Java, biostratigraphy of the
Globigerina marls (Saint-Marc and Suminta, 1979) near Ngawi
shows that the deposition of those marls lasted until the end of
the Pliocene, in contrast to the central area, because of the
Table 28.2. Radiometric ages from Sangiran
No.
Sample
Age. Ma
Method
Reference
1.
Notopuro (pumice)
0.25 +/- 0.07
FT
Suzuki etal. 1985
2.
Kabuh (tektite)
0:67 +/- 8.3%
FT
Nishimura et al. 1980.
3.
Kabuh (tektite)
0.70 - 0.72
FT
Suzuki etal. 1985
4.
Kabuh (tektite)
0.71 +/- 0.03
K/Ar
Orchiston & Siesser, 1982
5.
Kabuh (pumice, mean value

0.83
K/Ar
Jacob & Curtis, 1971; Jacob, 1975.
of 4 ages)
6.
Kabuh (tuff)
0.78+/-0.15
FT
Suzuki etal. 1985
7.
Kabuh (tuff)
0.47 +/- 0.02
FT
8.
Kabuh (tuff)
0.5 +/- 0.02
FT
9.
Kabuh (pumice)
1.05+/-0.1
K/Ar
Obradovich & Naeser, 1981
10.
Kabuh (pumice)
1.6+/-0.7
FT
Obradovich & Naeser, 1981
11.
Pucangan
0.4-0.5
FT
Orchiston & Siesser, 1982
12.
Pucangan (tuff)
0.57 +/- 0.32
FT
13.
Pucangan (tuff)
0.67 +/- 0.04
FT
14.
Pucangan (tuff)
1.16+/-0.24
FT
Suzuki etal. 1985
15.
Lower Pucangan (tuff)
1.49 +/- 0.32
FT
Suzuki etal. 1985
16.
Lower Pucangan (tuff)
1.51 +/- 0.25
FT
Suzuki etal. 1985
17.
Pucangan (basal lahar)
2.0 +/- 0.6
K/Ar
Nishimura et al. 1981
18.
Upper Kalibeng (tuff)
2.99 +/- 0.06
FT
Suzuki etal. 1985
Table 28.3. Radiometric ages from eastern Java localities
No.
Sample
19.
Pati-Ayam (Muriah basalt)
20.
Pati-Ayam (basal Slumprit

breccias)
Age, Ma
Method
Reference
0.50 +/- 0.02
K/Ar
Koenigswald, 1964
0.85 +/- 0.02
K/Ar
Bandetetal. 1989
21.
Trinil ("Pucangan" lahar)
0.5 +/- 0.3
K/Ar
Bartstra, 1978
22.
Perning/Modjokerto (tuff)
1.9+/-0.4
K/Ar
Jacob & Curtis, 1971
23.
Kedungbrubus (Pucangan
1.91
K/Ar
Jacob, 1978
1.87 +/- 0.04
K/Ar
Bandetetal. 1989
andesite)
24.
Kedungbrubus/Gunung
Butak
(top lower breccias)
Plio-Pleistocene of Indonesia
progressive emergence. That observation is consistent with the

placement of the Gauss-Matuyama boundary near the top of the
Globigerina marls in the Gemolong Dome.
Conclusion
In this chapter we have treated two lithological sequences that
include the Olduvai subchron. The adopted level for the
Pliocene-Pleistocene boundary at Vrica is associated with the
uppermost part of this normal-polarity event, at about 1.8 Ma
(Pasini and Colalongo, Chapter 2, this volume). In the Bumiayu
area, the boundary probably lies within the continental
synorogenic sediments of the Kali Glagah unit. In the Surakarta
region, it is situated slightly below the apparently diachronous
boundary between the littoral Kalibeng facies and the lahars
produced by the volcanism that marked the initiation of the
brackish-lagoonal and swampy facies of the Pucangan unit.
Global climate changes of the Pleistocene, indicated by successive periods of isolation and mixing in the vertebrate fauna that
appear to be related to glacio-eustatic shifts in sea level, may be
evident in the Bumiayu sequence.
Several other areas in eastern Java are likely to include the
same boundary, according to preliminary dating (Table 28.3) in
fossiliferous sites like Perning (Mojokerto), Kedung Brubus,
Pati-Ayam, and Trinil.
References
Bandet, Y., Semah, R, Sartono, S., and Djubiantono, T. 1989.
Premier peuplement par les mammiferes d'une region de
Java East, a la fin du Pliocene: age de la faune du Gunung
Butak, pres de Kedungbrubus (Indonesie). C. R. Acad.
Sci. Paris, ser. 7/308:867-870.
Bartstra, G. J. 1978. The age of the Djetis beds in East and
Central Java. Antiquity 52:56-58.
Curtis, G. H. 1981. Man's immediate forerunners: establishing a
relevant timescale in anthropological and archaeological
research. R. Soc. London, Phil Trans. B292:7-20.
De Vos, J., Sartono, S., Hardjasasmita, S., and Sondaar, P.-Y.
1982. The fauna from Trinil, type locality of Homo erectus,
are-interpretation. Geol. Mijnb. 61:207-211.
Djubiantono, T., and Semah, F. 1991. Lower Pleistocene
marine-continental transitional beds in the Solo depression
and their relation to the environment of the Pucangan
hominids. Indo-Pacific Prehist. Assoc. Bull.
Duyfjes, J. 1936. Zur Geologie und Stratigraphie des Kendenggebietes zwischen Trinil und Soerabaja (Djawa). De Ing.
Ned. Ind. 8:136-149.
Glass, B. P., and Heezen, B. C. 1967. Tektites and geomagnetic
reversals. Scientific American 217:32-38.
Indonesia-Japan Research Cooperation Program. 1979. Stratigraphy and geological structure in the central part of the
Sangiran Dome. Geol. Res. Dev. Centre, Bull. (Bandung)
2:55-61.
Itihara, M., Kadar, D., and Watanabe, N. 1985a. Concluding
remarks. In Quaternary geology of the hominid fossil
bearing formations in Java, ed. N. Watanabe and D.
Kadar, pp. 367-378. Special Paper 4. Bandung: Geological
Research and Development Centre.
Itihara, M., Wikarno, and Kagemori, Y. 1985b. Tektites from
271
the Sangiran area. In Quaternary geology of the hominid

Jacob, T. 1972. The absolute date of the Djetis beds at
Modjokerto. Antiquity 46:148-155.
Jacob, T. 1975. L'Homme du Java. La Recherche 6:1027-1032.
Jacob, T. 1978. New finds of lower and middle Pleistocene
hominines from Indonesia and an examination of their
antiquity. In Early Paleolithic in South and East Asia, ed.
F. Ikawa-Smith. Den Haag: Edit. Mouton.
Jacob, T., and Curtis, G. H. 1971. Preliminary potassium/argon
dating of early man in Java. Univ. Calif, Archaeol. Res.
Facility, Contrib. 12:1-50.
Lizon-Sureau, B. 1979. Essai de reconstitution de paleoenvironnement du Pithecanthrope de Java au Pleistoceene moyen
dans la region de Sangiran (Java Central) par les methodes
d'analyse. Unpublished these doctorat d'Etat, Universite
de Paris IV.
Marks, P. 1957. Stratigraphie lexicon of Indonesia. Publikasi
Keilmuan Djawatan Geologi Bandung, Seri Geologi, no.
31.
Matsu'ura, S. 1982. A chronological framing for the Sangiran
hominids - fundamental study by the fluorine dating
method. Nat. Sci. Mus. Tokyo, Bull, D 8:1-53.
Nishimura, S., Ikeda, T., and Yokoyama, T. 1981. A fundamental investigation on the fission track method" (in Japanese). Report for the grant-in-aid for scientific research (in
Japanese) 1981:5-11.
Nishimura, S., Thio, K. H., and Hehuwat, F. 1980. Fission
track ages of the tuffs of the Pucangan and Kabuh
formations and the tektites at Sangiran (Central Java). In
Physical geology of the Indonesian arcs, pp. 72-80.
Bandung: LIPI.
Obradovich, J. D., and Naeser, C. W. 1981. Indonesia: age of
Homo erectus from Java. U.S. Geological Survey, Prof.
Paper 1275:285-286.
Orchiston, D. W., and Siesser, W. G. 1982. Chronostratigraphy
of the Plio-Pleistocene fossil hominids of Java. Modern
Quat. Res. Southeast Asia 7:131-149.
Reinhold, T. 1937. Fossil diatoms of the Neogene and their zonal
distribution. Geol. Mijnb. Genootsch., Verh., Geol. Ser.
1937:1-133.
Saint-Marc, P., Paltrinieri, F., and Situmorang, B. 1977. Le
Cenozoique dTndonesie occidentale. Soc. Geol France,
Bull, Ser. 719:125-133.
Saint-Marc, P., and Suminta. 1979. Biostratigraphy of late
Miocene and Pliocene deep water sediments of Eastern
Java. /. Foram. Res. 9:106-117.
Sartono, S. 1982. Characteristics and chronology of early man in
Java. In Congres International de Paleontologie humaine,
pretirage,, vol. 2, ed. H. deLumley and M. A. deLumley,
pp. 495-541. Nice: Centre National de la Recherche
scientifique.
Sartono, S., Semah, F., and Djubiantono, T. 1984. Paleomagnetic results from the Gemolong and Bringinan Domes
(Central Java) - correlations with the Sangiran section.
Jur. Tek. Geologi I. T.B., Bui (Bandung) 13:7-21.
Schneider, D. A., Kent, D. V., and Mello, G. A. 1992. A detailed
chronology of the Australasian impact event, the BrunhesMatuyama geomagnetic polarity reversal, and global climate change. Earth Planet. Sci. Lett. 111:395-405.
Semah, A.-M. 1982. A preliminary report on a Sangiran pollen
diagram. Modern Quat. Res. Southeast Asia 7:165-170.
Semah, A.-M. 1984. Palynology and Javanese Pithecanthropus
paleoenvironment. In G. H. R. von Koenigswald Memo-
272
Francois Semah
rial Symposium, ed. J. L. Franzen and P. J. Andrews, pp.

245-252. Courier, vol. 69. Frankfurt: Forschungs-Institut
Senckenberg.
Semah, A.-M. 1986. Le milieu naturel lors de premier
peuplement de Java: resultats de 1'analyse pollinique.
Unpublished these doctorat d'Etat es Sciences, Universite
de Provence.
Semah, A.-M., and Rahardjo, A. T. 1984. Paleoenvironnement
des sites a hominides de Java. In Table-Ronde "L'homme
fossile et son environnement a Java (Manusia purba dan
lingkungannya di pulau Jaw a," Marseille, resultats et
communications, ed. H. deLumley et al., pp. 37-38. Paris:
Museum Nationale d'Histoire Naturelle and Bandung:
Puslit Arkenas.
Semah, F. 1983. Stratigraphie et paleomagnetisme du Pliocene
superieur et du Pleistocene de Vile de Java (Indonesie).
Travaux, no. 20. Paris: Museum Nationale d'Histoire
Naturelle, Laboratoire de Prehistoire.
Semah, F. 1984. The Sangiran Dome in the Javanese PlioPleistocene chronology. In G. H. R. von Koenigswald
Memorial Symposium, ed. J. L. Franzen and P. J.
Andrews, pp. 245-252. Courier, vol. 69. Frankfurt:
Forschungs-Institut Senckenberg.
Semah, F. 1986. Le peuplement ancien de Java - ebauche d'un
cadre chronologique. UAnthropologie 90:359-400.
Semah, R, Sartono, S., Zaim, Y., and Djubiantono, T. 1980.
Premiers resultats concernant l'etude paleomagnetique de
la partie Ouest de dome de Sangiran (Java, Indonesie). C.
R. Acad. Sc. Paris, ser. D, 290:477-480.
Shimizu, Y., Mubroto, B., Siagian, H., and Untung, M. 1985. A
paleomagnetic study in the Sangiran area. In Quaternary
geology of the hominid fossil bearing formations in Java,
ed. N. Watanabe and D. Kadar, pp. 275-307. Special
Paper 4. Bandung: Geological Research and Development
Centre.
Sondaar, P.-Y. 1981. The Geochelone faunas of the Indonesian
archipelago and their paleogeographical and biostratigraphical significance. Modern Quat. Res. Southeast Asia
6:11-120.
Sondaar, P.-Y. 1984. Faunal evolution and the mammalian

biostratigraphy of Java. In G. H. R. von Koenigswald
Memorial Symposium, ed. J. L. Franzen and P. J.
Andrews, pp. 219-235. Courier, vol. 69. Frankfurt:
Forschungs-Institut Senckenberg.
Sumarso and Suparyono. 1974. A contribution to the
stratigraphy of the Bumiayu area. Bandung: Ikatan Ahli
Geol. Indonesia.
Suzuki, M., Wikarno, Budisantoso, Saefudin, I., and Itihara, M.
1985. Fission track ages of pumice tuff, tuff layers, and
javites of hominid fossil bearing formations in Sangiran
area, Central Java. In Quaternary geology of the hominid
Ter Haar, C. 1934. Geologische kaart van Java - Blad 58 (Bumiajoe). Batavia: Dienst Mijnbouw Nederlandse Indien.
Van Bemmelen, R. W. 1949. The Geology of Indonesia and
adjacent archipelagoes, vol. 1. Den Haag: Martinus
Nijhoff.
Van Es, L. J. C. 1931. The age of Pithecanthropus. Den Haag:
Martin Nijhoff.
von Koenigswald, G. H. R. 1933. Beitrag zur Kenntnis der
fossilen Wirbeltiere Java, Teil I, We tens. Meded. no. 23.
Batavia: Dienst Mijnbouw Nederlandse Indien.
von Koenigswald, G. H. R. 1934. Zur Stratigraphie des
Javanischen Pleistocan. De Ingin. Ned. Ind., Sect. 4
11:185-201.
von Koenigswald, G. H. R. 1964. Potassium/argon dates and
early man: Trinil. In Reports of the 6th INQUA Congress,
Warsaw, 1961., vol. 4, pp. 325-327. Warsaw: Polska
Akademia Naukam.
Watanabe, N., and Kadar, D. (eds.) 1985. Quaternary geology of
the hominid fossil bearing formations in Java. Special
Paper 4. Bandung: Geological Research and Development
Centre.
29
The Pliocene-Pleistocene boundary in New Zealand

ANTHONY R. EDWARDS
Introduction
Castella, Santa Maria di Catanzaro, and Vrica sequences,

Edwards et al. (1981) noted that "at least four different
correlations . . . can be supported."
Data published on the biostratigraphy and paleomagnetic
characterization of the proposed stratotype boundary at Vrica
(Backman, Shackleton, and Tauxe, 1983; Tauxe et al., 1983;
Zijderveld et al., 1991) provide a new opportunity to correlate
between New Zealand and Italy. According to the most recent
analyses, the boundary-stratotype at Vrica occurs just below the
top of the Olduvai normal subchron at approximately 1.81 Ma.
Since 1953 the Pliocene-Pleistocene boundary in New Zealand

has been set at the base of the Nukumaruan (regional) Stage
because that coincides with the abrupt first appearance of the
subantarctic bivalve Chlamys delicatula (Hutton) in marine
sequences of central New Zealand. However, it is now estimated, on the basis of biostratigraphic, magnetostratigraphic,
and radiometric data, that the Nukumaruan commenced at about
2.4 Ma and ended about 1.3 Ma and that the Olduvai subchron
approximately coincides with the middle of the Nukumaruan. In
the stratotype Nukumaruan, and in correlative sections, a
Nonmarine sequences
marked lithologic change (interpreted as the result of glacioeustatic regression) following the first appearance of Gephyro- On North Island, the nonmarine rocks attributed to the
capsa sinuosa Hay and Beaudry may be correlative with the top Nukumaruan Stage include lowland clastic deposits that are
of the Olduvai and with the Vrica boundary-stratotype.
similar to parts of the upper Nukumaruan (including the
stratotype) in the Wanganui Basin (Figure 29.2). On South
Island, uplift of the main mountain ranges during the Pliocene
Historical background
and Pleistocene resulted in many local accumulations, sometimes
In New Zealand, the regional stages for Upper Pliocene to more than 1,000 m thick, of gravel (now deeply weathered) and
Middle Pleistocene strata, in upward sequential order, are finer-grained alluvium.
Waipipian, Mangapanian, and Nukumaruan. The Nukumaruan
Palynomorphs are used to assign those deposits to regional
Stage is sometimes divided into the Hautawan and (above) stages, although correlation with the marine strata on which the
Marahauan substages. More or less continuous marine and stages are based is imprecise. In particular, the stratigraphic
nonmarine sequences extend from the Pliocene into the Pleisto- range of the "Hautawan" (lower Nukumaruan) palynomorph
cene in various parts of the country. Internal correlation between assemblage zone is uncertain. Its lower limit is probably in the
the sequences is poor, and for external paleontological correla- upper Mangapanian (Hornibrook, 1981, pp. 280-281), and its
tion, for example with Italy, only the marine sequences are of upper limit somewhere near the middle of the Nukumaruan (D.
value. Modern reviews of the Pliocene-Pleistocene boundary C. Mildenhall, personal communication). Sequences in the
have been published by Vella (1975), Jenkins (1975), Hornibrook northwestern region of South Island are important for their
(1976), Edwards, Hornibrook, and Te Punga (1981), and Te inclusion of glacial deposits attributed to the Ross and Porika
Punga (1981).
glaciations. Those of the Ross Glaciation (Gage, 1945; Bowen,
A cooling trend indicated by the arrival of the bivalve Chlamys 1967) are within gravel closely overlying a lignite that has a
delicatula (Hutton), a member of the circum-subantarctic C. palynoflora similar to those in the mostly pre-"Hautawan"
patagonica species group, was an important factor in the decision Moutere Gravel of Nelson (Mildenhall and Suggate, 1981). The
of New Zealand stratigraphers in 1953 to adopt the base of the gravel at Ross lies, apparently conformably, about 30 m above a
Nukumaruan Stage as the base of the Pleistocene. Since then, thin shell bed with Waipipian Stage mollusks (Suggate et al.,
despite evidence of earlier coolings and inadequate correlation 1978), so that the Ross Glaciation probably occurred in
with the boundary in Italy, there has been a reluctance to change Waipipian or early Mangapanian time. At Timaru in South
the assignment. Even with substantial information on the Le Canterbury, cool palynofloras, possibly coeval with the Ross
273
Anthony R. Edwards
274
UJI
100
t\
200 km
I
Kaweka yjfig Hawke Bay

WANGANUI BASIN v .
7^(
4(S
DSDP Site
284
Figure 29.1. Distribution of marine

Nukumaruan strata in New Zealand (black is exposed; stippled pattern is subsurface) and locations of
the stratigraphic columns in Figure
29.2 as follows: A, Nukumaru
Beach; B, Mangaopari Stream; C,
Totara Road; D, Devils Elbow. Also
shown is DSDP site 284. Small
outliers of marine strata occur near
Castlepoint, Kaweka, and Waipara;
important nonmarine sequences occur near Porika, Ross, and
Timaru.
^f^4 c WCasttepoint
Tasman
Sea
i-
)x
W Parnassus
Ross/
/^Waipara
Pacific
Ocean
_44_
/
/
Glaciation, overlie warmer palynofloras of Waipipian age (D. C.

Mildenhall, personal communication) and underlie a reversepolarity basalt (Gair, 1961) with a whole-rock K/Ar date of 2.47
0.38 Ma (old constants; Mathews and Curtis, 1966).
Glacial deposits from the Porika Glaciation overlie the
Moutere Gravel and contain "Hautawan" palynomorphs (Mildenhall and Suggate, 1981); thus the Porika Glaciation probably
occurred in late Mangapanian or early Nukumaruan time, coeval
with thefirstarrival of Chlamys delicatula in marine sequences of
New Zealand.
The marine Nukumaruan
Nukumaruan marine strata are well developed in central New

Zealand (Figure 29.1), where basins continued to subside
through the Pliocene and into the Pleistocene. Later uplift and
erosion have provided many good sections, especially in the
southern part of North Island. In the Wanganui Basin, up to 900
m of very shallow-marine sandstone, mudstone, coquina, and
limestone are followed by up to 100 m of nonmarine (and rare
marginally marine) clastic strata. In the eastern basins, the
Nukumaruan strata typically consist of 100-600 m of more or less
cyclically deposited shallow-marine mudstone, shelly sandstone,
and limestone; small areas dominated by deep-water conglomeratic sandstone, alternating sandstone/mudstone, or massive mudstone are also known. Most sections contain minor breaks in
sedimentation and lithologic changes attributed to glacio-eustatic
changes (Beu and Edwards, 1984).
The four on-land sequences summarized in Figure 29.2 are as
follows:
~~
1. Nukumaru Beach, Wanganui Basin (Fleming, 1953;

Beu and Edwards, 1984): stratotype of the Nukumaruan Stage, very shallow marine, becoming nonmarine upsection.
2. Mangaopari Stream, southern Wairarapa (Kennett,
Watkins, and Vella, 1971; Beu and Edwards, 1984): the
best-understood late Cenozoic section in New Zealand.
3. Totara Road, southern Hawkes Bay (Hornibrook,
1981): located near a paleo-seaway between the eastern
and western basins.
4. Devils Elbow, northern Hawkes Bay (Hornibrook,
1981; Beu and Edwards, 1984): a sequence showing
well-developed lithologic cycles.
Together, those four sequences are fairly representative of the
shallow-marine Nukumaruan; although a few deep-water sequences are known, they are all incomplete, poorly exposed, or
technically disturbed.
Biostratigraphy
The New Zealand Upper Pliocene-Middle Pleistocene regional

stages are based on the mollusk-rich shallow-marine strata of the
Wanganui Basin. Many of the key mollusks have wide but
somewhat patchy distributions elsewhere in New Zealand. In the
deeper-water strata of inland eastern North Island, the molluscan biostratigraphic scheme can be related to a recently
developed microfossil scheme only partly recognizable in the
Wanganui Basin (Hornibrook, 1981; Beu and Edwards, 1984).
The resultant integrated biostratigraphy provides a reliable
275
Plio-Pleistocene of New Zealand
VRICA,
ITALY
V28-239,
EQUATORIAL
PACIFIC
DSDP 284,
TASMAN SEA
DEVILS
ELBOW
TOTARA
ROAD
MANGAOPARI
STREAM
NUKUMARU
BEACH
400 m
1 UNCONFORMITY
| MUDSTONE
[;:\\
SANDSTONE
E -= emergence
[rV-j LIMESTONE
fc\<</j CROSS-BEDDING
r"1 CONGLOMERATE
marine incurston
Waipuru Ash (undated)
C.d.
Chlamys delicatula
shale layer (Vrica)
Figure 29.2. Primary stratigraphic correlations between the

Nukumaruan strata of four New Zealand on-land sequences
(Hornibrook, 1981; Beu and Edwards, 1984) and the Italian Vrica sequence (Backman et al., 1983; Tauxe et al., 1983); the short normal
interval above bed e observed by Zijderveld et al. (1991) is not shown.
means of recognizing the New Zealand stages beyond their

stratotypes, although not without some difficulties.
There are few widely distributed biostratigraphic events that
are reliable for correlation purposes. In upward stratigraphic
order, the Nukumaruan bioevents are as follows:
1. First appearance of Chlamys delicatula (Hutton): traditionally accepted, but not without controversy, as indicating the base of the Nukumaruan in many sequences.
2. First appearance of Globorotalia crassula Cushman and
Stewart: currently the most accurate means of recognizing the base of the Nukumaruan, but possibly unreliable in some shallow-marine sequences.
3. Last appearance of predominantly dextral populations
of Globorotalia crassaformis (Galloway and Wissler).
4. First appearance of Gephyrocapsa sinuosa Hay and
Beaudry: very useful in a wide variety of lithofacies.
The positions of these events in four representative New
Zealand sequences are given in Figure 29.2. Other relevant
bioevents are as follows:
The upper bathyal sequence at DSDP site 284 is adapted from Kennett
et al. (1975) and Hornibrook (1982), modified according to new data.
The abyssal sequence of core V28-239 is adapted from Shackleton and
Opdyke (1976) and Backman and Shackleton (1983).
5. First appearance of typical populations of Globorotalia

truncatulinoides (d'Orbigny): useful in some sequences
where it occurs between bioevents 3 and 4 mentioned
earlier. This species gradually evolved from G.
tosaensis Takayanagi and Saito during late Mangapanian and early Nukumaruan times. In subtropical-totropical environments, it first occurs just above the
Gauss-Matuyama boundary (Barron et al., 1985).
6. Last appearance of Discoaster brouweri Tan Sin Hok:
not useful for internal correlation purposes; inferred
from a few very rare and sporadic occurrences to be
between bioevents 3 and 4. This bioe vent occurs not far
below the Olduvai subchronozone in Pacific cores
(Backman and Shackleton, 1983).
7. Last appearance of Calcidiscus macintyrei (Bukry and
Bramlette): possibly a useful late Nukumaruan bioevent, but not yet well documented; definitely lies
above bioe vent 4 in section C. In Pacific cores it is
recorded just above the Olduvai subchronozone (Backman and Shackleton, 1983), as well as in the Mediterra-
276
Anthony R. Edwards
nean (Rio, Raffi, and Backman, Chapter 5, this

volume).
8. Last appearance of Helicosphaera sellii (Bukry and
Bramlette): a very useful but imprecisely located late
Nukumaruan bioevent; definitely lies above bioevent 4
in sections A and C. In the deep sea it is found above
event 7, close to the Jaramillo subchronozone (Gartner,
1977; Raffi, Rio, and Backman, Chapter 5, this
volume).
For further information on these and numerous other bioevents in New Zealand, see Suggate et al. (1978), Hornibrook
(1976, 1981), Beu, Grant-Taylor, and Hornibrook (1977),
Hoskins (1982), and Beu and Edwards (1984).
The three radiometric dates shown against the Nukumaru
Beach column in Figure 29.2 were obtained from elsewhere in
the Wanganui Basin and positioned by lithostratigraphic correlation (Seward, 1976; Beu and Edwards, 1984). The lowest, from
the Ohingaiti Ash, is a zircon fission-track date (Seward, 1979);
the middle date is the K/Ar age of a pebble collected from a
slightly higher (possible Nukumaru Limestone equivalent?)
horizon (Mathews and Curtis, 1966); the highest, from the
Mangahou Ash, is a glass-shard fission-track date (Seward,
1974).
Nukumaruan paleomagnetic observations have been made
only at Mangaopari Stream (Kennett et al., 1971); the interpretation in Figure 29.2 is according to A. R. Edwards (in
Hornibrook, 1981; Beu and Edwards, 1984). The combined
paleontologic, paleomagnetic, and radiometric data are consistent with the Nukumaruan Stage commencing at about 2.4 Ma
and ending at about 1.3 Ma (Beu and Edwards, 1984).
improvements in our understanding of the microbiostratigraphy

of this interval, thanks to Backman and colleagues, as summarized by Rio, Raffi, and Backman (Chapter 5, this volume). Beu
and Edwards (1984), using inferences of glacio-eustatic changes,
have estimated where the boundary should lie in three of the
New Zealand sequences shown in Figure 29.2. Their estimates
are at about 30 m (Devils Elbow), 45 m (Mangaopari Stream),
and 50 m (stratotype of the Nukumaruan Stage at Nukumaru
Beach) above the first appearance of G. sinuosa; all three
estimates fall at obvious lithologic changes.
Acknowledgments
I thank my colleagues, Dr. A. G. Beu, Mr. D. C. Mildenhall,

Mr. G. H. Scott, and Dr. R. P. Suggate, for constructive
criticisms of this review. The manuscript was typed by Brenda
Gray, and the figures were drafted by Shirley Rogers.
References
Backman, J., and Shackleton, N. J. 1983. Quantitative biochronology of Pliocene and early Pleistocene calcareous
nannofossils from the Atlantic, Indian and Pacific Oceans.
Backman, J., Shackleton, N. J., andTauxe, L. 1983. Quantitative
304:156-158.
Barron, J., Nigrini, C. A., Poujos, A., Saito, T., Theyer, R,
Thomas, E., and Weinrich, N. 1985. Synthesis of biostratigraphy, central equatorial Pacific, Deep Sea Drilling
Project Leg 85: refinement of Oligocene to Quaternary
biochronology. In Initial reports of the Deep Sea Drilling
Project, vol. 85, ed. L. Mayer, F. Theyer, et al., pp. 905934. Washington, DC: U.S. Government Printing Office.
External correlations
Beu, A. G., and Edwards, A. R. 1984. New Zealand Pleistocene
and late Pliocene glacio-eustatic cycles. Paleogeogr. PalaeoMagnetostratigraphy appears to be a straightforward method to
climatol. Palaeoecol. 46:119-142.
correlate between the Mangaopari Stream and Vrica sequences
(Figure 29.2). The identification of the Olduvai subchron in New Beu, A. G., Grant-Taylor, T. L., and Hornibrook, N. de B. 1977.
Nukumaruan records of the subantarctic scallop Chlamys
Zealand is still largely dependent on biostratigraphy (Beu and
delicatula and crab Jacquinotia edwardsii in central Hawkes
Edwards, 1984), as noted earlier.
Bay. New Zeal J. Geol Geophys. 20:217-248.
In contrast to the magnetostratigraphic correlation, biostrati- Bowen, F. E. 1967. Early Pleistocene glacial and associated
deposits of the West Coast of the South Island, New
graphic correlations between New Zealand and Vrica are
Zealand. New Zeal. J. Geol. Geophys. 10:164-181.
indirect. The first correlation is between New Zealand and the
Edwards, A. R., Hornibrook, N. de B., and Te Punga, M. T.
nearby upper bathyal DSDP site 284; the second is between
1981. In pursuit of the Pliocene-Pleistocene boundary.
DSDP site 284 and Vrica via the geographically intermediate
Geol. Soc. New Zeal. Newsl. 52:21-28.
abyssal core V28-239 (Figure 29.2). The change in the order of Fleming, C. A. 1953. The geology of Wanganui Subdivision.
New Zeal. Geol. Surv. Bull, n.s. 52:1-362.
the LAD (last-appearance datum) of C. macintyrei and the LAD
Gage,
M. 1945. The Tertiary and Quaternary Geology of Ross,
of H. sellii between core V28-239 and DSDP site 284 is not
Westland. R. Soc. New Zeal., Proc. 75:138-159.
regarded as critical. The LAD of H. sellii probably is the less Gair, H. S. 1961. Drillhole evidence of the Pliocene-Pleistocene
reliable of these two bioevents in the Pacific (Backman and
boundary at Timaru, South Canterbury. New Zeal. J.
Shackleton, 1983).
Geol. Geophys. 4:89-97.
From that, the proposed Pliocene-Pleistocene boundary- Gartner, S. 1977. Calcareous nannofossil biostratigraphy and
revised zonation of the Pleistocene. Mar. Micropal. 2:1stratotype (the base of the claystone resting on bed e in section B
25.
at Vrica) can be correlated to an undefined level within the Hornibrook, N. de B. 1976. Globorotalia truncatulinoides and
upper part of the Nukumaruan and above the first occurrences of
the Pliocene-Pleistocene boundary in northern Hawkes
G. sinuosa in the New Zealand region, largely because of
Bay, New Zealand. In Progress in micropaleontology:
Plio-Pleistocene of New Zealand
selected papers in honor of Prof. Kyoshi Asano, ed. Y.

Takayanagi and T. Saito, pp. 83-102. New York: Micropaleontology Press.
Hornibrook, N. de B. 1981. Globorotalia (planktic Foraminiferida) in the Late Pliocene and Early Pleistocene of New
Zealand. New Zeal J. Geoi Geophys. 24:263-292.
Hornibrook, N. de B. 1982. Late Miocene to Pleistocene
Globorotalia (Foraminiferida) from DSDP Leg 29, Site
284, Southwest Pacific. New Zeal. J. Geol. Geophys.
25:83-99.
Hoskins, R. (ed.) 1982. Stages of the New Zealand marine
Cenozoic: a synopsis. Report 107. Lower Hutt: New
Zealand Geological Survey.
Jenkins, D. G. 1975. The Pliocene/Pleistocene boundary in New
Zealand and Australia. In Late Neogene epoch boundaries,
ed. T. Saito and L. H. Burckle, pp. 94-100. Spec. publ. 1.
New York: Micropaleontology Press.
Kennett, J. P., Houtz, R. E., and others (eds.) 1975. Initial
277
Seward, D. 1974. Age of New Zealand Pleistocene substages by

fission-track dating of glass shards from tephra horizons.
Earth Planet. Sci. Lett. 24:242-248.
Seward, D. 1976. Tephrostratigraphy of the marine sediments in
the Wanganui Basin, New Zealand. New Zeal. J. Geol.
Geophys. 19:9-20.
Seward, D. 1979. Comparison of zircon and glass fission-track
ages from tephra horizons. Geology 7:479-482.
paleomagnetic stratigraphy of Pacific core V28-239, Late
Pliocene to latest Pleistocene. Geol. Soc. Am., Mem.
145:449-464.
Suggate, R. P., Stevens, G. R., andTe Punga, M. T. (eds.) 1978.
The geology of New Zealand, vol. 2. Wellington: Government Printer.
Reports of the Deep Sea Drilling Project, vol. 29. Washing- Te Punga, M. T. 1981. The Pliocene/Pleistocene boundary and
the Nukumaruan Stage, New Zealand. New Zeal. Geol.
ton, DC: U.S. Government Printing Office.
Surv., Rep. 96:1-19.
Kennett, J. P., Watkins, N. D., and Vella, P. 1971. Paleomagnetic
chronology of Pliocene-Early Pleistocene climates and the Vella, P. 1975. The boundaries of the Pliocene in New Zealand.
In Late Neogene epoch boundaries, ed. T. Saito and L. H.
Plio-Pleistocene boundary in New Zealand. Science 171:
Burckle, pp. 85-93. Spec. publ. 1. New York: Mi276-279.
cropaleontology Press,
Mathews, W. H., and Curtis, G. H. 1966. Date of the PliocenePleistocene boundary in New Zealand. Nature 212:979- Zijderveld, J. D. A., Hilgen, F. J., Langereis, C. G., Verhallen,
980.
Mildenhall, D. C , and Suggate, R. P. 1981. Palynology and age
of the Tadmor Group (Late Miocene-Pliocene) and
Porika Formation (early Pleistocene), South Island, New
107:697-714.
Zealand. New Zeal. J. Geol. Geophys. 24:515-528.
30
The Pliocene-Pleistocene boundary in continental sequences of

North America
EVERETT H. LINDSAY
published a comprehensive summary of the North American late

Stratigraphic sections coeval with the Pliocene-Pleistocene Cenozoic vertebrate faunal sequence, in which the mammal
boundary section at Vrica have been securely identified in two record was correlated with the four classic North American
long terrestrial sequences in the southwestern USA: the San glacial ages (Nebraskan, Kansan, Illinoian, and Wisconsinan). In
Pedro Valley sequence in Arizona and the Anza-Borrego that report, the Pliocene-Pleistocene boundary was placed
Badlands sequence in southern California. Those sequences, within, rather than at the end of, the Blancan, because late
with excellent chronologic resolution afforded by magneto- Blancan faunas appeared to represent a shift to colder climates
stratigraphy, biostratigraphy, and isotope dating, provide a (Figure 30.1). The chronologic-climatic framework for that
framework for interpreting the evolutionary and paleoclimatic report was based primarily on the faunal sequence in southwestchanges in North America at the Pliocene-Pleistocene boundary, ern Kansas and northwestern Oklahoma, which was established
as well as for interpreting the sequence and effects of interconti- by Hibbard and his students. A key part of that framework was
the correlation of the distinctive and widespread Pearlette Ash,
nental dispersal events in the late Cenozoic.
which was recognized in post-glacial sediments attributed to the
retreat of the Kansan ice sheet. The warm-adapted Cudahy
Glacial climate change and continental Plio-Pleistocene fauna, from below the Pearlette Ash, was held to be an early
correlations
Irvingtonian assemblage and was correlated with late Kansan
In the late 1930s, the so-called Wood Committee (Wood et al., warming trends. The cold-adapted Borchers fauna, from above
1941) assigned the Blancan Provincial Land Mammal Age to the the Pearlette Ash, was interpreted as late Irvingtonian and was
late Pliocene and inferred that the Pliocene-Pleistocene bound- correlated with the end of the Yarmouth interglacial.
ary was located at the end of the Blancan. Although no formal
Work since 1965 (e.g., Izett, 1981) has shown that the "Pearlette
Pleistocene land mammal age was proposed in the 1941 report, a Ash" above the Cudahy fauna is a volcanic unit different from the
list of Pleistocene mammals was given, for which the term "Pearlette Ash" below the Borchers fauna, with a radiometric age
"Rancholabrean" (characterized by the famous remains from the more than 1 m.y. younger, and that the Borchers fauna is older
Rancho La Brea tar pits in Los Angeles) later came into general (not younger) than the Cudahy fauna. In fact, three "Pearlette
use. In 1951, D. E. Savage identified the Irvington fauna of the Ash" units are now identified, based on stratigraphy near the
San Francisco Bay area as the type for a new "Irvingtonian" land source area in Yellowstone Park, Wyoming: the Lava Creek b ash
mammal age, for early Pleistocene, post-Blancan and pre- (equivalent to Pearlette type O), dated to 0.6 Ma; the Mesa Falls
Rancholabrean fossil mammal faunas. Thus, by the early 1950s, ash (Pearlette type S), dated to 1.3 Ma; and the Huckleberry
three late Cenozoic land mammal ages (Blancan, Irvingtonian, Ridge ash (Pearlette type B), dated to 2.0 Ma (Izett, 1981). The
and Rancholabrean) had been identified in the late Cenozoic of Lava Creek b ash overlies the Cudahy fauna, and the Huckleberry
North America, and the Pliocene-Pleistocene boundary (in Ridge ash underlies the Borchers fauna. The Cudahy fauna is now
North American vertebrate paleontology) was placed at the considered late Irvingtonian, and the Borchers fauna is considBlancan-Irvingtonian boundary. Nevertheless, the age and even ered latest Blancan or earliest Irvingtonian.
the terms of recognition for the epoch boundary in North
At the time, Hibbard suspected that the 1965 framework might
American contintental sequences remained controversial, be- be wrong (Zakrzewski, 1975, p. 123), and by 1972 he openly
cause the criteria for its basic definition in the Old World were in questioned his 1965 correlations with the glacial sequence
conflict (Berggren and Van Couvering, 1979).
(Skinner and Hibbard, 1972, pp. 131-136). For one thing, Izett et
Despite controversies, improvements in the biostratigraphic al. (1970) had already noted the occurrence of two superposed
and chronostratigraphic resolution of the late Cenozoic continen- "Pearlette ashes" at Cerro Summit, Colorado, and for another,
tal record continued, and in 1965 Hibbard and co-workers Skinner and Hibbard (1972, p. 136,fig.60) had come to recognize
Introduction
278
Plio-Pleistocene of North America
MAMMALIAN
CLIMATES
GLACIAL
FAUNAS
EPOCHS
AGES
AGES
Microthermal, subhumid,
continental
Jingle bob
UJ
Mesothermal, humid
maritime
OC
O
X
hispidus
\
^
*
o
Mt
Scott
maritime ?
Illinoian
Microfhermal, subhumid,
continental?
Butler
Spring
UJ
o
Yarmouth
prosior
caliche bed
Semiarid
Mesothermal, subhumid,
Borchers
maritime
R V I NG
Pearle te ash bed

maritime
Cu d ahy
Crotaphytus collaris
Phrynosoma modestum
Notiosorex crawfordi
Dosypterus
gollineri
Sorex arcticus ^ x
Terrapene llanensis
S. cinereus
^
Blarina b. carolinensis
S. palustris
v
^
Oryzomys fossilis
Microtus pennsylvanicus
v
Perco flavescens
^
Ambystoma, neotenic
c
Sorex cinereus
'
/
'
Ambystoma, metamorphosed
/
Geochelone
o |
^
UJ
^S
*'
/ Ambystoma. metamorphosed
/
^
Terrapene llanensis
Oryzomys fossilis
/
/ Geochelone
i1
Terrapene llanensis
\
.
,
Holbrookia texana
maritime
Quarry
tr
o_
Sigmodon
(
,.
Microtus
pennsylvanicus f
^ --
UJ
allopatric \
Ambystoma, neotenic
Sorex cinereus
Citellus richardsoni
Sangomon
Cragin
<
species now
Sorex cinereus
(D
Mild - winter
elements
Many formerly sympatric
Mesothermal, semiarid,
Jones
SHIFTS
Cool - summer
elements
continental
Wisconsin
icn
FAUNAL
AND
SIGNIFICANT
GEOLOGIC DATA
Living
RECENT
279
Sigmodon hilli
Spilogale cf. S. ambarvalis
Ambystoma. neotenic
.
.
Sorex cinereus
S. lacustris
S. megapalustns
.
^
Mrcrosorex pratensis
\
_
, .
*
%
Synaptomys (Mictomys) 1
'
Microtus paroperarius /
_ s
No
o
c
climatically
Se ger
significant
'
Sonders
maritime
Af tonion
Some warm-
Mesothermal,
maritime
Deer Park
o
Sigmodon intermedius
Bensonomys meadensis
Pliolemmus antiquus
elements from
Ambystoma, metamorphosed
'
temperate and
subtropical
Rexroad fauna
species
'
|
>
Geochelone
Pliopotamys
meadensis
Pliolemmus antiquus
z
No
Nebraskan
significant
CD
UJ
climatically
Un named
_J
species
r
B end er
maritime
*i
l
Mammalian
not
fauna
published
UJ
o
o
-1
colich
maritime
Figure 30.1. Late Cenozoic faunal shifts and inferred climatic changes
in the southern Great Plains. Note the Pearlette Ash bed between the
1 Geochelone rexroadensis
^ 1 Notiosorex jacksoni.
Nerterogeomys
minor
Bassariscus
cosei
Baiomys spp.
1 Sigmodon intermedius
Cudahy and Borchers faunas. (From Hibbard et al., 1965, figure 2,

with permission of Princeton University Press.)
280
Everett H. Lindsay
Figure 30.2. Correlation of Blancan deposits and faunas in the Great

Plains region and Idaho, according to Skinner and Hibbard (1972, figure 6). Note that the Sand Draw and Broad water faunas of Nebraska
are correlated to a warm interval near the Pliocene-Pleistocene boundary. Radiometric age calibration according to Berggren et al. (1985).
that there were in fact diverse, warm-adapted late Blancan faunas

(e.g., Sand Draw in northeastern Nebraska) at relatively high
latitudes. That was sufficient evidence for those authors to
conclude that most, if not all, Blancan faunas in the Great Plains
area were older than the "first glacial climates" (Figure 30.2).
Unfortunately, Hibbard and most other students of North
American vertebrate paleontology, up to and including Kurten
and Anderson (1980), were locked into a rigid paleoclimatic
concept in which the earliest evidence of any climatic deterioration was considered, by definition, as Nebraskan and, by
definition, the beginning of the Pleistocene. As an example, the
Broadwater fauna of southwestern Nebraska, a late Blancan
warm-adapted fauna very similar to the Sand Draw fauna of
Skinner and Hibbard (1972), was considered to be of early
Pleistocene age by Schultz, Lueninghoener, and Frankforter
(1951), in part because a gravel bed interpreted as a Nebraskan
till underlies the strata yielding the fauna.
I heartily endorse the conclusions of Harland et al. (1982, p. 43)
regarding the unreliability of Pleistocene glacial sequences:
"Attempts to make a stratigraphic sequence out of successive
cycles of continental glaciation have proved to be totally
inadequate . . . and . . . continental classifications have been
shown to be oversimplified and incomplete, based to a large
extent on erroneous concepts." Oxygen-isotope changes in
oceanic plankton of deep-sea cores represent at least 23 globally
significant intervals of cooling during the past 1 m.y. (Shackleton
and Opdyke, 1973, 1976; Harland et al., 1982; Thunell and
Williams, 1983). Correlation of North American continental
glaciation with oceanic cooling intervals is still uncertain in detail;
to maintain integrity and accuracy, correlations of North Ameri-
can Quaternary land mammal faunas must be based on superposition, morphologic evolution of common taxa, isotope dating, and
magnetic-polarity sequences, not on climatic inferences.
Present state of North American Plio-Pleistocene
correlations
Among the relatively recent attempts to correlate the North
American Quaternary mammal faunas are the following: those
of Repenning (1979, 1987; Repenning, Fejfar, and Heinrich,
1990; Repenning and Brouwers, 1992), based on the evolution
and dispersal of microtine rodents; that of Schultz, Martin,
Tanner, and Corner (1978), based on the grouping of faunas
according to taxonomic similarity; and that of Johnson, Opdyke,
and Lindsay (1975), based on magnetostratigraphic correlations
tied to detailed biostratigraphy in the southwestern USA. It
should be pointed out that those and most other studies of North
American Quaternary mammals (e.g., Kurten and Anderson,
1980) have included Blancan as well as Irvingtonian and
Rancholabrean land mammal ages, since at least part of the
Blancan has frequently been considered Pleistocene.
Microtine rodent event-stratigraphy
Repenning (1979) divided the Blancan intofivesequential faunal
intervals, and the Irvingtonian and Rancholabrean into two
intervals each, characterized by the appearances of immigrant
species of microtine rodents closely related to European faunas,
or characterized by newly evolved species. In a follow-up,
Repenning (1987) expanded on that study with some slight
revisions (Figure 30.3). His chronologic sequence was based on a
Ma
WEST OF DENVER
M I C R O T I N E
EAST OF DENVER
F A U N A ,
U . S . A .
(NEW IMMIGRANTS ARE CAPITALIZED)
EUROPE
AGE
MICRO
HRON
MEGA I
* TEICHERT, CA
0.5
DOWNEY DUMP. I
SNOWVILLE, UT
H LIVERMORE, CA
UPPER TECOPA, CA
BARREL SPRINGS, CA
HIGH ALAMOSA, CO
URRIETA, CA
IRVINGTON,
* CENTERVILLE. CA
MICROTUS PENNSYLVANICUS. M. MEXICANUS, M. MONTANUS, LAGURUS ap.. M. callfornicua,

Synaptomya (S.) auatralla, S. (M.) maltonl-boraalia, Naollbar laonardl, Clathrlonomya p.,
Padomya ochrogaatar, P. llananala. Ondatra nabraakanala, Phanacomys as., Pltymya app.,
TEOUISOUINAHUA, MEX
SANOALL, TX
EZABECK. KS
KANOPOLIS. KS
SLATOM, TX
ANGUS. NE
x CUMBERLAND CAVE. MD
CUDAHY. KS o TOBIN, KS *VERA, TX
WILSON VALLEY, KS
_, ~ B e e l , -,
CONARD FISSURE, AR R 0 C K CREEK. TX
JOCOTOPEC, JAL
FYLLAN CAVE, TX
* LITTLE SIOUX COUNTY. IA
CLETHRIONOMYS cf. GAPPERI, PITYMYS MEADENSIS, P. MCKNOWNI, MICROTUS PAROPERARIUS,

P. cumbarlandanala, Mlcrotua callfornicua, Synaptomya (S.) cooparl, S. (M.) kanaaaanala,
S. (M.) maltoni, Naoflbar laonardl, Ondatra annactana, Padomya llananala,
Allophalomya gulldayl, Atopomya taxanala, A. aalvallnua
0.5
MICROTUS CALIFORNICUS, ALLOPHAIOMYS cl. PLIOCAENICUS, PHENACOMVS p..
1.0
+ SAPPA, NE
GILLILAND. TX
Ondatra annactana, Pllophanacomya oabornl
+ BORCHERS, KS
SYNAPTOMYS (MICTOMYS) VETUS, S. (SYNAPTOMYS) RINKERI, S. (Mlctomya) landaal.

Ondatra Idahoanala, Mlmomya (Ogmodontomya) monohoni, Mlmomya (Ophlomya) parvua,
M. (Op.) maadanala, Nabraakomya mcgrawi, Pliopotamya maadanaia, Pllolammua antiquua,
Pliophanacomya oabornl
OLIVE DELL. CA
1.5
^CURm RANCH. AZ ' ^ V J V A ^ I
2.0
DECEIT,
THAYNE,
BIG SPRINGS, NE
MULLEN, NE (In part)
BEAVER, UT
* HIGH III RANCH, Ai
CALIFORNIA WASH , AZ
ELK HILLS, C
WILD HORSE BUTTE, ID GRAND VIE
2.5
WHITE ROCK, KS
BLANCO. TX
*BOYLECDW|TCH
TUSKER. AZ
ARROYO SECO. CA
Mimomya (Coaomya) prlmua, M. (Ogmodontomya) poaphagua, M. (Ophiomya) taylcn-parvua,

M. (Op.) maadanala, Pliopotamya maadanala, Pliopotamya minor,
Pllophanacomya primaavua -oabornl, Pllolammua antlquua
FLATIRON BUTTE, ID
CLARKDALE, AZ
3.0
BROAOWATER, NE
MENDEV.L RANCH. AZ
SAND POINT. ID
JNJ50N,_AZ
3.0
* COSO MTS. CA
RED CORRAL. TX
PLIOPOTOMYS MINOR. PHopotamy
PANACA, NV
DUNCAN, AZ
3.5
"'"' " "

SAN JOAQUIN, CA
. ID|
TAUNTON. WA
I. (Op.) magllll, Pllolammua antlquua,
x SAND DRAW, NE
K DEER PARK, KS
Mlmomya (Ogdodontomya) poaphagua, M. (Coaomya) prlmua, M. (Ophlomya) mcknightMaylorl,

inacomya flnnayl, Pllolammua antlquua, Nabraakomya raxroadantta
x REXROAD 3, KS
ORCHARO,
.VERfit AZ.1
4.0
x FOX CANYON, KS
_ Zl I"_ J I l _ _ _
WHITE BLUFFS. WA
-4.5
MIMOMYS (COSOMYS) SAWROCKENSIS, M. (OPHIOMYS) MCKNIGHTI, NEBRASKOMYS Cl.

REXROADENSIS, Pllophanacomya wllaonl
CONCHA, CHIHUAHUA
MAXUM, CA
UPPER ALTl
-5.0
-5.5
3.5
Pllophonacomya primaavua, Nabraakomya mcgrawl
REXROAD 2, KS
I. ID
4.0
-2.0
YEPOMERA, CHIHUAHUA
LINO COULEE, WA
MT EDEN, CA
MCKAY, OR
CHRISTMAS VALLEY, OR
g o
Propllophanacomya parkarl
5.0
PROMIMOMYS MIMUS
5.5
WARREN, CA
-6.0
6.0
-
nm
WHITE CONE. AZ
6.7
Figure 30.3. Correlation of microtine dispersal events in North America to subdivisions of the
land mammal ages. (From Re penning, 1987,figure1, courtesy of U.S. Geological Survey.)
Mlcrotoacoptaa Mbbardi, Qontodontomya dlalunctua
Everett H. Lindsay
282
very comprehensive study of this widespread and abundantly

fossiliferous group and, in collaboration with O. Fejfar and
others, on correlations with European faunas (Repenning et al.,
1990). Repenning's division of Quaternary faunas is a great leap
forward toward the chronologic resolution of North American
Pleistocene mammal faunas, although I regard the separation of
the early Blancan into five parts to be excessive, when three are
all that may reasonably be called for. Also, Repenning's dating
of Blancan faunas, based on correlations to Europe, conflicts
with the direct evidence of local age determinations (Lindsay,
Johnson, and Opdyke, 1975; Lindsay, Opdyke, and Johnson,
1984), which date the base of the Blancan to about 4.3 Ma,
rather than 4.8 Ma.
r^-i
HOLOCENE
tt
CO
<
(Late Wisconsinan)
WISCONSIN GLACIAL
Z
U
! tt j
^
(Early Wisconsinan)
SANGAMON INTERGLACIAL
Sheridanian
Port Konnody l.f.

Cudahy l.f.
Conard Fissure l.f.
o
z
Faunal-assemblage subdivisions
Schultz et al. (1978) provided names for the faunal assemblages
of Plio-Pleistocene age that are, in effect, subdivisions of the
conventional Blancan and Irvingtonian mammal ages (Figure
30.4). Early Blancan faunas were grouped into the Rexroadian
sub-age (considered pre-Blancan by Kurten, 1972), and later
Blancan faunas into the Senecan sub-age. Early Irvingtonian
faunas were grouped into the Sappan sub-age, and later
Irvingtonian faunas into the Sheridanian sub-age; the Cudahy,
Port Kennedy, and Conard Fissure faunas were left, however, as
an unnamed middle Irvingtonian group. Rancholabrean faunas
were not discussed, although early and late Wisconsinan
intervals, as well as a Sangamon interglacial division, were
illustrated as Rancholabrean (Schultz et al., 1978,fig.1). In this
chapter, species differences (Smilodon californicus vs. S. fatalis,
and Castoroides kansensis vs. C. nebraskensis) separate Rancholabrean from Irvingtonian faunas. The appearances of
modern species of microtines, and the earliest certain record of
Bison, distinguish Sheridanian from Sappan faunas, and the
appearances of the vole Allophaiomys and the lemming Synaptomys (Mictomys) distinguish Sappan from Senecan faunas.
Schultz et al. (1978) also questioned whether or not Mammuthus
appeared at the beginning of the Irvingtonian (Sappan).
Schultz et al. (1978) vacillated on placing the PliocenePleistocene boundary between the Blancan and Irvingtonian, or
between the egregious Kimballian (= early Hemphillian) (Lindsay et al., 1984) and Blancan. Schultz and Hillerud (1978)
reiterated the Great Plains Pleistocene mammal sequence of
Schultz et al. (1951), based on the concept of a five-terrace
sequence cut into and built on Pliocene rocks (Figure 30.5). The
oldest terrace (Terrace 5) includes the late Blancan Broadwater
formation and fauna (noted earlier), underlain by a gravel
interpreted as a Nebraskan glacial till; that was given an inferred
age of 3.2 Ma, which may or may not be the same age as the
beginning of the trend toward Pleistocene climates noted at
about that time in mid-Pliocene marine sequences (Thunell and
Williams, 1983; Shackleton et al., 1984, 1985; Shackleton, Hall,
and Pate, 1995). A critical examination of Schultz and Hillerud
(1978) (Figure 30.5) suggests that terraces 1, 2, and 3 are
Sappan
(NEW NAME)
Senecan
(NEW NAME)
2
^
Hay Springs l.f.

Rushvillo l.f.
Gordon l.f.
Angus l.f.
Irvington l.f.
Rexroadian
KIMBALLIAN/HEMPHILLIAN
CLARENDONIAN
VALENTINIAN
f Sappal.f.
| Wat hen a l.f.
Java 1. f.
Grandview l.f.
White Rock l.f.
Dixon l.f.
Seneca l.f.
' Broadwater l.f.
Lisco l.f.
Blanco l.f.
Hagerman l.f.
Sand Draw l.f.
PROVINCIAL AGES
OF LOCAL FAUNAS
FROM OGALLALA
GROUP SEDIMENTS
Figure 30.4. Provincial land mammal age subdivisions for the North
American Quaternary and late Tertiary proposed by Schultz et al.
(1978, figure 1). (Courtesy of Nebraska Academy of Sciences.)
Holocene. All of the significant Pleistocene climatic events must

therefore be recorded in Terrace 4.
Magnetostratigraphically controlled biochronology
Magnetostratigraphy, in combination with biostratigraphy, in

thick, continuous sequences offers a third approach to the
analysis of North American Pleistocene terrestrial deposits and
faunas.
San Pedro Valley, Arizona
In the San Pedro Valley sequence of Arizona, the magnetostratigraphy published by Johnson et al. (1975) identified 11
reversal events in 150 m of strata in the St. David Formation that
could be correlated with the interval from late Gilbert to
Brunhes in the magnetic-polarity time scale (Figure 30.6). Also,
four biochronologically significant faunal datum events were
identified in the biostratigraphy of the vertebrate fossil sites. The
first three of these events, the Sigmodon medius LSD (lowest
Complex
of local fills, loesses)
silts, paleosols, and
khhorizons
High Plains Surface

(Nebraska to Texas)
T-5
200-50 m
Diagrammatic Interpretation of the Terrace Sequence

in the Central Great Plains
(Showing Stratigraphk Positions of Various Age Indicators)
Valentine Fm.
( m a x l thickness 7 S m ) i n :
10,500- 12,5O0y.BP
(time of valley erosion)
Figure 30.5. Diagrammatic interpretation of the sequence of post-Kimballian terrace
fills developed in the Great Plains of Nebraska. (From Schultz and Hillerud, 1978,
figure
C. tertrand SchuHx and John M. HHIerud
1, courtesy of Nebraska Academy of Sciences.)
Everett H. Lindsay
284
DECLINATION
400CLAYSTONE
f~|
GRANITE WASH
B |
MARL I I I
SANOSTONE
270*
360*
90*
180*
270*
-120
300- - 9 0
H
OLDUVAI
SILTSTONE
H |
-60
TUFF P H I
Figure 30.6. Variation of magnetic declination with stratigraphic level at Curtis Ranch. In
this and other magnetic-polarity
columns, the black sections represent normally magnetized strata,
and the white sections represent
reversely magnetized strata. Eight
fossil levels in the San Pedro Valley sequence are indicated by
bone symbols to the left of the
lithologic column. These fossil levels, named in order of increasing
age, are Prospect, Glyptotherium
(of Gidley), Gidley level, Johnson
Pocket, Cal Tech, Horsey Green
bed, Honey's Hummock, and Bonanza. (From Johnson et al.,
1975, with permission of the Geological Society of America.)
PALEOSOL
FOSSILS
100- - 3 0
stratigraphic datum) at the base of the Mammoth subchronozone of the Gilbert chronozone, the Nannippus HSD
(highest stratigraphic datum) at the base of the Matuyama
chronozone, and the Ondatra idahoensis LSD near the Reunion
subchronozone of the lower Matuyama, are associated with the
Blancan land mammal age. The youngest datum event, the
Lepus LSD at the base of the Olduvai subchronozone, has been
associated with the Curtis Ranch fauna of earliest Irvingtonian
land mammal age. Unfortunately, the taxon previously recorded
as Lepus in the Curtis Ranch fauna is now identified as a large
species of another rabbit, Sylvilagus (White, 1991). A smaller
species of Sylvilagus appears stratigraphically lower in the San
Pedro Valley strata in the California Wash fauna, associated
with the Ondatra idahoensis datum event (Lindsay et al., 1990).
Perhaps a more significant biochronologic event is the replacement of archaeolagine rabbits (e.g., Hypolagus and Pewelagus)
by leporine rabbits (e.g., Sylvilagus and Lepus) in the late
Blancan (White, 1987). It appears the San Pedro Valley faunal
KAENA
3
I n
0--0
UJ
COCHITI
3
-100 -J30
sequence documents the transition from the Blancan to the

Irvingtonian land mammal age, in which the Curtis Ranch
fauna represents the latest Blancan or earliest Irvingtonian
faunal interval. The Curtis Ranch fauna is securely placed at
the base of the Olduvai subchron and therefore (according to
the Vrica calibration) represents the latest Pliocene fauna in
that sequence.
Anza-Borrego, California
In the Anza-Borrego sequence in southern California, Opdyke
et al. (1977) identified 11 magnetic reversals through 4,000 m of
strata in the Palm Springs and upper Imperial formations
(Figures 30.7 and 30.8). The Anza-Borrego magnetic sequence is
almost a duplicate of the San Pedro Valley magnetic sequence,
although with a much higher average sedimentation rate (1.2 m/
k.y. vs. 0.04 m/k.y.) (Figures 30.6 and 30.9). Johnson et al.
(1983) extended the Anza-Borrego paleomagnetic sequence
285
FAUNAS
DECLINATION ()
M.
FT.
DATUM PLANES
?Euceratherium LSD
Smilodon LSO
0T 0
2000
I Odocoileus LSO
lypolagus HSD
2000-
4000-
Tremarctos LSO
f. Equus LSO
1000
6000
4000-
Geomys LSD
80006000
2000
8000
10.000
12.00010.000 -3000
Ptiohippus HSD
-14000
Figure 30.8. Mammal datum planes in the Anza-Borrego Badlands sequence. (From Opdyke et al., 1977, with permission of the University
of Washington Press.)
12.000
1-4000
ANZA BORREGO
Blancan-Irvingtonian boundary
The transition from the Blancan to the Irvingtonian land

mammal age is identified by the appearance of Ondatra and
perhaps Sylvilagus in the San Pedro Valley sequence and by the
appearance of cf. Odocoileus and Smilodon in the Anza-Borrego
sequence. Those interpretations suggest that the Blancandown through the Imperial Formation, after removing a mag- Irvingtonian boundary should be placed below but near to the
netic overprint in the samples by thermal cleaning. They were base of the Olduvai subchronozone, at approximately 2.1 Ma in
able to identify 17 magnetic reversals (including the Nunivak the orbitally tuned calibration. Thus, the Blancan-Irvingtonian
subchron of the Gilbert magnetic chron), and with that control boundary is slightly older than the Pliocene-Pleistocene boundthey were able to demonstrate a change in the sedimentation rate ary, as identified at Vrica within the uppermost part of the
from 5.5 m/k.y. to 0.5 m/k.y. in an interval of 3.4 m.y. during Olduvai (Pasini and Colalongo, Chapter 2, this volume).
deposition of a deltaic complex in the Anza-Borrego area
The beginning of the Irvingtonian land mammal age is best
(Figure 30.9).
characterized by the appearance of Eurasian immigrants, espeVertebrate fossils in the Anza-Borrego Badlands have been cially Mammuthus. Perhaps the earliest North American record
assigned to Blancan and Irvingtonian land mammal ages (Downs of Mammuthus is from the Wellsch Valley fauna, which has been
and White, 1968). Eight local datum events were identified in the correlated to a level within the Olduvai magnetic subchron. To
Anza-Borrego faunal sequence by Opdyke et al. (1977) and date, Mammuthus has not been recorded in the Curtis Ranch
dated according to paleomagnetic interpolation (Figure 30.9). sequence in the San Pedro Valley nor in the paleomagnetically
The lower six datum events (cf. Pliohippus HSD, Geomys LSD, dated sequence of the Anza-Borrego Badlands. The current
cf. Equus LSD, Tremarctos LSD in the Gilbert chronozone, and consensus of North American vertebrate paleontologists (cf.
the HSD of Hypolagus and LSD of cf. Odocoileus in the lower Lundelius et al., 1987) is to consider the Blancan-Irvingtonian
Matuyama chronozone) are associated with Blancan land mam- boundary to be coincident with the base of the Olduvai
mal age, while the two highest faunal datum events {Smilodon subchron. This has the effect of placing the Olduvai-age Wellsch
LSD and 1 Euceratherium LSD near the top of the Olduvai Valley and Curtis Ranch faunas, probably the earliest known
subchronozone in the upper Matuyama) are associated with Irvingtonian faunas, in the latest part of the Pliocene, just below
the level of the Vrica definition.
Irvingtonian land mammal age.
Figure 30.7. Variation of magnetic declination with stratigraphic level
in the Anza-Borrego Badlands, California. (From Opdyke et al., 1977,
with permission of the University of Washington Press.)
Everett H. Lindsay
286
VGP LATITUDE
4--
2--
Figure 30.9. Extended Anza-Borrego

Badlands sequence, with calibrated
magnetic-polarity time scale of
Mankinen and Dairymple (1979) on
the right. (From Johnson et al.,
1983, with permission of the Geological Society of America.)
Sequence of late Cenozoic mammal faunas in North

America
Additional paleomagnetically dated sequences from other areas

of the western USA have contributed to a rigorous interpretation
of late Cenozoic faunal changes in North America. Those
additional paleomagnetic sequences include Meade County,
Kansas (Lindsay et al., 1975), and the Snake River plain of
Idaho (Neville etal., 1979).
Lindsay et al. (1984) completed a summary of North American
late Cenozoic mammal faunas that includes the faunas mentioned earlier, as well as the Chamita section of New Mexico
(MacFadden, 1977), the Verde section of Arizona (Bressler and
Butler, 1978), the Wikieup section of Arizona (MacFadden,
Johnson, and Opdyke, 1979), and several previously unpublished

sections. Figures 30.10 and 30.11, taken from that study,
illustrate the correlations and sequences of those faunas, which
extend the correlation of North American faunas into the late
Miocene and Hemphillian land mammal age. The Hemphillian
land mammal age includes the interval from about 4.5 Ma to 8
Ma, correlative with magnetic chrons 5, 6, and 7, and half of
chron 4 (Gilbert). The Blancan land mammal age includes the
interval between about 1.9 Ma and 4.5 Ma, correlative with the
lower Matuyama, the Gauss, and the upper half of the Gilbert
magnetic chrons. The Blancan land mammal age thus includes
most of the Pliocene, although the early Pliocene extends into
late Hemphillian age, and the latest Pliocene is equivalent to the
basal Irvingtonian.
287
A/Vl
HAGERMAN
VERDE
CHIHUAHUA
HEMPHILL
i xx 4.4-4.8
QUIBURIS
WIKIEUP
CHAMITA
I xx 5.5-5.9
5.3-5.7
xx 4.2 -6.3
VAA;
Figure 30.10. Correlation of Blancan and Hemphillian fossil sequences.

Magnetic-polarity time scale on the left; positions and error limits of
isotopically dated ashes are shown by the symbol XX. Bone symbols
indicate faunal levels. Hagerman section from Neville et al. (1979), with
Gidley Horse Quarry level; Ringold section from Lindsay et al. (1984),
with level of Ophiomys mcknighti; Verde section from Bressler and Butler (1978), with level of Clarkdale fauna (above) and Verde fauna (be-
Ma
MPTS
LSD
low); Chihuahua section from Lindsay et al. (1984), with level of

Concha fauna (above) and Yepomera fauna (below); Quiburis section
from Lindsay et al. (1984), with position of Camel Canyon fauna indicated by lower bone symbol; Wikieup section from MacFadden et al.
(1979), with level of Wikieup fauna; Chamita section from MacFadden
(1977), with level of Rak Camel Quarry; Hemphill section from Lindsay et al. (1984), with level of White Cone fauna.
HSD
-0
o- I
I -
Mammuthus ( 0
2-
-Dipodomys(A), Smilodon (C)

-Steqomastodon (A,C)
_
. y, . , A X
' ..
.' .
fHypolagus(T ,Nannippus(T ,
-Ondatra (A), Sylv.lagus (A,C)
, J Borophagusm.Rhynchother.um (T)
JSynaptomys (A),Erethizon(A),Neochoerus(A),
LProdipodomys(A)
LGlyptotherium (T),Glossotherium (T),Stegomastodon(T)
rChasmaporthetes(K),Odocoileus(K) J PratilepusiI),
3-
4-
5-
6-
7-
i Neotoma (0
BvfEquustD.Trigonictis (I),Castor(I),Ursus(I),
\Ophiomys (W), Pliopotamys(I),Thomomys(I)
Pliophenacomys(A,M),Geomys (M)
Paenemarmota (M), Prosigmodon(M),Notolagus(M)BNr'Agriotherium(M),Prosthennops(M)
LDinohippus(M), Aphelops (M)
LDin
-Onohippidium (A)
I^PIesiogulo (A), Pliotaxidea (A),

V.Sphenophalus(A)
ilS
-\
\rPediomeryx(T),Mylagaulus
\Perognathoides(A)
(A,T,NM)
v |'Calomys(Bensonomys)(A),Paronychomys(A)
V Agriotherium(T),Rhynchotherium (T),
i . lProdipodomys(A),Galushamys(A)
UPlesiogulo (NM), Pliotaxidea (NM),
V Dinohippus interpolatus(NM), Astrohippus ansae(NM)
loipoides (NM), Hypolagus vetus (NM)
-2
-3
-4
-5
-6
-7
Figure 30.11. North American late
Cenozoic mammal-datum events,
based on data summarized by Lindsay et al. (1975, 1984).
288
Everett H. Lindsay
Lindsay, E. H., Opdyke, N. D., and Johnson, N. M. 1984.

Blancan-Hemphillian land mammal ages and late Cenozoic mammal dispersal events. Ann. Rev. Earth Planet.
Sci. 12:445-488.
Lindsay, E. H., Smith, G. A., Haynes, C. V, and Opdyke, N. D.
1990. Sediments, geomorphology, magnetostratigraphy
and vertebrate paleontology in the San Pedro Valley,
Arizona. /. Geol. 98:605-619.
Lundelius, E. L., Churcher, C. S., Downs, T., Harington, C. R.,
Lindsay, E. H., Schultz, G., Semken, H. A., Webb, S. D.,
and Zakrzewski, R. J. 1987. The North American QuaterReferences
nary sequence. In Cenozoic mammals: their temporal
Berggren, W. A., and Van Couvering, J. A. 1979. The
record, biostratigraphy, and biochronology, ed. M. O.
Quaternary. In Treatise on Invertebrate Paleontology, Part
Woodburne, pp. 211-235. Berkeley: University of CaliforA: Introduction, ed. R. A. Robison and C. Teichert, pp.
nia Press.
MacFadden, B. J. 1977. Magnetic polarity stratigraphy of the
Berggren, W. A., Kent, D. V., and Van Couvering, J. A., 1985.
Chamita Formation Stratotype (Mio-Pliocene) of north
central New Mexico. Am. J. Sci. 277:769-800.
chronology of the geological record, ed. N. J. Snelling, pp. MacFadden, B. J., Johnson, N. M., and Opdyke, N. D. 1979.
211-250. Geol. Soc. London, Memoir 10. Oxford: BlackMagnetic polarity stratigraphy of the Mio-Pliocene
well Scientific Publishers.
mammal-bearing Big Sandy Formation of western AriBressler, S. L., and Butler, R. F. 1978. Magnetostratigraphy of
zona. Earth Planet. Sci. Lett. 44:349-364.
the late Tertiary Verde Formation, central Arizona. Earth Mankinen, E. A., and Dalrymple, B., 1979. Revised geomagPlanet. Sci. Lett. 38:319-330.
netic time scale for the interval 0 to 5 m.y. /. Geophys. Res.
84B:615-626.
Downs, T., and White, J. A. 1968. A vertebrate faunal
succession in superposed sediments from late Pliocene to Neville, C , Opdyke, N. D., Lindsay, E. H., and Johnson, N. M.
1979. Magnetic stratigraphy of Pliocene deposits of the
middle Pleistocene in California. In Proceedings of the
Glenns Ferry Formation and its implications for North
XXIII International Geological Congress, Prague. Section
American mammalian biostratigraphy. Am. I. Sci. 279:
10, Tertiary Quaternary Boundary, pp. 41-47. New York:
503-526.
Academic Press.
Harland, W. B., Cox, A. V, Llewellyn, P. G., Pickton, C. A. G., Opdyke, N. D., Lindsay, E. H., Johnson, N. M., and Downs, T.
1977. The paleomagnetism and magnetic polarity stratigSmith, A. G., and Walters, R. 1982. A geologic time scale.
raphy of the mammal-bearing section of Anza-Borrego
Cambridge: Cambridge University Press.
State Park, California. Quat. Res. 7:316-329.
Hibbard, C. W. 1970. Pleistocene mammalian local faunas of the
Great Plains and central lowland provinces of the United Repenning, C. A. 1979. Faunal exchanges between Siberia and
North America. Can. J. AnthropoL, special AMQUA issue
States. In Pleistocene and recent environments of the
central Great Plains, pp. 395-433. Lawrence: University of
1:37-44.
Repenning, C. A. 1987. Biochronology of the micro tine rodents
Kansas Press.
of the United States. In Cenozoic mammals: their temporal
Hibbard, C. W, Ray, C. E., Savage, D. E., Taylor, D. W., and
record, biostratigraphy, and biochronology, ed. M. O.
Guilday, J. E. 1965. Quaternary mammals of North
Woodburne, pp. 236-268. Berkeley: University of CaliforAmerica. In The Quaternary of the United States, ed. H. E.
nia Press
Wright and D. G. Frey, pp. 509-525. Princeton: Princeton
Repenning, C. A., and Brouwers, E. 1992. Late Pliocene-Early
University Press.
Pleistocene ecologic changes in the Arctic Ocean borderIzett, G. A. 1981. Volcanic ash beds: recorders of upper
land. U.S. Geol. Survey, Bull. 2036:1-37.
Cenozoic silicic pyroclastic volanism in the western United
Repenning, C. A., Fejfar, O., and Heinrich, W.-D. 1990.
States./. Geophys. Res. 86:10200-10222.
Arvicolid rodent biochronology of the Northern HemiIzett, G. A., Wilcox, R. E., Powers, H. A., and Desborough,
sphere. In International symposium: evolution, phylogeny
G. A. 1970. The Bishop ash bed, a Pleistocene marker
and biostratigraphy of arvicolids (Rodentia, Mammalia),
bed in the western United States. Quat. Res. 1:121-132.
Johnson, N. M., Officer, C. B., Opdyke, N. D., Zeitler, P. K.,
ed. O. Fejfar and W.-D. Heinrich, pp. 385-418. Prague:
and Lindsay, E. H. 1983. Rates of subsidence and uplift of
Geological Survey.
the Vallecito-Fish Creek Basin, western Imperial Valley, Savage, D. E. 1951. Late Cenozoic vertebrates of the San
California. Geology 11:664-667.
Francisco Bay Region. Univ. Calif. Publ. Geol. Sci.
Johnson, N. M., Opdyke, N. D., and Lindsay, E. H. 1975.
28:215-314.
Magnetic-polarity stratigraphy of Pliocene-Pleistocene Schultz, C. B., and Hillerud, J. M. 1978. Climatic change and the
terrestrial deposits and vertebrate faunas, San Pedro
extinction of large mammals during the Quaternary. Nebr.
Valley, Arizona. Geol. Soc. Am., Bull. 86:5-12.
Acad. Sci., Trans. 6:95-105.
Kurten, B. 1972. The Age of Mammals. New York: Columbia Schultz, C. B., Lueninghoener, G. C , and Frankforter, W. D.
University Press.
1951. A graphic resume of the Pleistocene of Nebraska
Kurten, B., and Anderson, E. 1980. Pleistocene mammals of
(with notes on the fossil mammalian remains). Nebr. State
North America. New York: Columbia University Press.
Mus., Bull. 3:1-41.
Lindsay, E. H., Johnson, N. M., and Opdyke, N. D. 1975. Schultz, C. B., Martin, L. D., Tanner, L. G., and Corner, R. G.
Preliminary correlation of North American land mammal
1978. Provincial land mammal ages for the North Ameriages and geomagnetic chronology. Univ. Mich. Pap.
can Quaternary. Nebr. Acad. Sci., Trans. 5:59-64.
Paleontol. 12(3):111-119.
Shackleton, N. J., Backman, J., Zimmerman, H., Kent, D. V, et
Chronostratigraphic ordering of these faunas, based on
magnetostratigraphy, biostratigraphy, and radiometric dating,
provides a framework for interpreting faunal and climatic
changes in North America during the late Cenozoic. It also
allows correlation with similar faunal changes in other parts of
the world and facilitates identification of the timing and
sequence of intercontinental dispersal events by land mammals.
289
al. 1984. Oxygen isotope calibration of the onset of ice- Thunell, R. C , and Williams, D. F. 1983. The stepwise
rafting and history of glaciation in the North Atlantic
region. Nature 307:620-623.
paleoceanographic conditions in the Mediterranean: oxygen isotopic studies of DSDP sites 125 and 132. Utrecht
Shackleton, N. J., Hall, M. A., and Pate, D. 1995. Pliocene
the Ocean Drilling Program: scientific results, Leg 138, ed.White, J. A. 1987. The Archeolaginae (Mammalia, LagoN. G. Pisias et al., pp. 337-356. College Station, TX:
morpha) of North America, excluding Archaeolagus and
Panolax. J. Vert. Paleontol 7:425-450.
Shackleton, N. J., and Opdyke, N. D. 1973. Oxygen isotope and White, J. A. 1991. North American Leoprinae (Mammalia,
paleomagnetic stratigraphy of equatorial Pacific core V28Lagomorpha) from Late Miocene (Clarendonian) to latest
238: oxygen isotope temperatures and ice volumes on a
Pliocene (Blancan). J. Vert. Paleontol. 11:67-89.
5
6
Wood, H. E., Chaney, R. W., Clark, J., Colbert, E. H., Jepsen,
10- and 10- year scale. Quat. Res. 3:39-55.
G. L., Reeside, J. B., and Stock, C. 1941. Nomenclature
and correlation of the North American continental Terpaleomagnetic stratigraphy of Pacific core V28-239, late
tiary. Geol. Soc. Am., Bull. 52:1-48.
Pliocene to latest Pleistocene. Geol Soc. Am., Mem.
145:449-464.
Zakrzewski, R. J. 1975. Pleistocene stratigraphy and paleontology in Western Kansas: the state of the art, 1974. Univ.
Skinner, M., and Hibbard, C. W. 1972. Early Pleistocene preglacial and glacial rocks and faunas of north-central
Mich. Pap. Paleontol. 12:121-128.
Nebraska. Amer. Mus. Nat. Hist., Bull. 148:1-148.
Index
0.9 Ma date for Pliocene-Pleistocene

boundary, see Menapian concept;
Cromerian (end-Villafranchian) concept
boundary, see GSSP; PliocenePleistocene boundary; Vrica section
boundary, see Pretiglian concept; Middle
Villafranchian concept; Paratethys tectonic concept
boundary see Lower Villafranchian concept
40
AT/39AT age, see K/Ar dating
Abies (fir), 33, 47, 107-112, 204, 217
Acheulean type stone tools, 132, 134, 249,
255-256; see also artifact
Acila cobboldiae, last appearance as criterion for Pliocene-Pleistocene boundary
in North Sea, 198
Acinonyx (cheetah), 119, 174, 228, 249
Afrochoerus, 119
Akchagylian stage, 7, 111, 219, 221-225
Akkulaevo (Russia), 108, 118, 181, 219
Alces (moose), 229
Allactaga (jerboa), 120, 218
Allocricetus, 120, 247
Allophaiomys, 115, 118, 120, 123-124, 174175, 179-180, 186, 222, 282
Almenara-Casablanca (Spain), 115, 118,
171
Alnus (alder), 107-112, 204, 229
Amphisorus zone in Israel, 163; see also
Tyrrhenian stage
Amstelian stage, 186-187
Anabo Koma (Djibouti), 256
Anancus ("mastodon"), 115, 172, 186, 194,
204, 216, 222
antelopes, see Gazella; Gazellospira;
Antilospira; Antilospiroides; Gallogoral;
Sinoryx
Anthropogene, xii, xix, 221-223
Antian stage, 183
Antilospira, 228, 248
Antilospiroides, 247
Anza-Borrego (California), 122-123, 284285
Apodemus (field mouse), 118, 172-175,

217-218, 242, 247
Apsheronian stage, 7, 11, 108, 111, 122,
221-225
Aquatraversan erosional interval, 149, 223;
see also mid-Pliocene cold climate;
Pretiglian glacial phase; sea-level change
Arago (Spain), 133, 135
Archidiskodon, 120, 183, 222, 227, 234234; see also Mammuthus
Arctica islandicafirstappearance in Mediterranean as criterion for PliocenePleistocene boundary, xx, 4, 10, 15, 38,
67-70, 141-146, 148
argon-argon age, see K/Ar ages
Arialaceoipollenites, 195
Arondelli (Italy), 181
Artemisia, 34
artifact, 121, 131-134, 169, 249, 255-259
Arvicola (water vole), 174
arvicoline rodents, see Allophaiomys;
Arvicola; Cleithrionomys; Lagurodon;
Pliomys; Villanyia
Astian stage, 7, 106
astronomical time scale, see orbital time
scale
Atapuerca (Spain), 119, 133, 174-175
Aullan erosional interval, 115, 152, 172; see
also Eburonian glacial phase; sea-level
change
aurochs, see Bos
Australian Plio-Pleistocene localities, 122
Australopithecus, 119, 131-133
Awash section (Ethiopia), 132, 254-256
Axis (axis deer), 242, 248
baboons, see Parapapio; Theropithecus; see
also monkeys
Bakuan or Bakunian stage, 7, 111, 221
Barogali (Djibouti), 256
base of Pleistocene, see PliocenePleistocene boundary
bats, see Miniopterus; Myotis; Rhinolophus
Baventian stage, 11, 183, 188, 197
Baza Basin (Spain), 118, 171, 174-175
bears, see Protarctos; Tremarctos; Ursus
beavers, see Castoroides; Eucastor;
Sinocastor; Trogontherium
Beestonian stage, 183, 197

Beremendia, 217, 247
Betekian faunal complex, 120, 222, 227
Betfia (Romania), 118, 181, 222
Betula (birch), 107-112, 229
Biharian land mammal age, 193-195, 222
Bison, 119, 121, 124, 174-175, 229, 282
Blancan land mammal age, 122-123, 278286
Bodo (Ethiopia), 255
Bogatschevia, 217, 224
Bos (aurochs), 118, 121, 134, 249
Boshernitskan mollusk fauna, 224
Boulder Conglomerate stage, Siwaliks,
234-236
bovines, see Bison; Bos; Bubalus; Hemibos;
Leptobos
Bresse basin (France), 179-181
Brielle (Netherlands), 118, 180-181
Briiggen climate phase, 203
Brunssumian climate phase, 106-107, 192,
195, 203
Bubalus (water-buffalo or carabao), 121
buffalo, see Bison; Bubalus
Bugiulesti, see Tetoiu
Calabrian Stage, as basis of PliocenePleistocene boundary, xiii-xvi, xx, 4, 46,
66, 106, 142, 145-149, 164, 225, 247; see
also Le Castella; Santa Maria di
Catanzaro; Selinuntian Stage
calcareous nannoplankton, PlioPleistocene, 23-25, 39, 63-76, 250
Calcidiscus macintyrei, 23, 24, 39, 58, 63-
67,71,73,79,87,275
canids, see Canis; Cuon; Nyctereutes;
Vulpes
Canis (wolf, jackal, coyote), 115, 118-119,
134, 149-153, 194, 219, 228, 249
Canis etruscus as indicator of PliocenePleistocene boundary, 118, 219
Capo Rossello section (Sicily), 146-147
Capra (goat), 118, 174-175
Capreolus (roe-deer), 119
carabao, see Bubalus
carbonate cycles in deep sea related to glacial climate cycles, 97-102
Carpinus, 107-112
291
292
Carya (hickory), 33, 107-112, 185, 195,

197, 203
Casa Frata (Italy), 118
Casablanca, see Almenares-Casablanca
Castanea (chestnut), 33, 107-112, 195, 197
Castell'Arquato section (Italy), 38, 144-145
Castillomys, 118, 172-175
Castoroides, 282
cattle, see bovines
Cave bear, see Ursus
caves, see Transvaal caves
Cedrus (cedar), 33, 110-112, 217
Cervalces, 118-119, 124, 153
cervids, see Axis; Capreola; Cervalces;
Cervus; Dama; Elaphurus; Eucladoceros;
Megaloceros; Odocoileus; Praedama;
Rusa
Cervus (stag or elk), 115, 123, 153, 172,
217, 249
Chagny (France), 181
Chasmaporthetes, 115, 119
cheetah, see Acinonyx
Chemeron (Kenya), 131, 258
Chesowanja (Kenya), 132, 258
Chikoi, see Tchikoi
Chilhac (France), 115
Chilotherium, 250
Chistopol (Russia), 223
Chiwondo (Malawi), 131
Chlamys delicatula (southern scallop) as indicator of glacial climate, 273
Citellus (ground squirrel), 180
Clethrionomys, 115, 123
climate, see carbonate cycles; earliest
glaciation; glacial climate development;
interglacial climate; mid-Pliocene coldclimate interval; oxygen-isotope data;
paleobotanic data; sapropelites
Coelodonta ("woolly rhinoceros"), 249
coiling direction in planktonic foraminifera,
28, 39, 47, 52, 87, 144
"cold guests" as Pleistocene index, see Arctica islandica; Cytheropteron testudo;
Hyalinea baltica
Coralline Crag (England), 183
Cornus (dogwood), 107, 112
Corylus (hazelnut), 33, 107-112
coyote, see Canis
Cribrononion, 35
Crocidura (white-toothed shrew), 174
Crocuta (spotted hyena), 119-120, 174
Croizetoceros, 115, 149, 174
Cromerian climate phase, Cromerian stage,
"Cromerian complex", 7, 107, 186
Cromerian (end-Villafranchian) concept of
Pliocene-Pleistocene boundary, 7, 222223, 239-240; see also Metasequoia extinction
Cromerian mammal faunas, 174, 222
Csarnota (Hungary), 181, 219
Cueva Victoria (Spain), 118, 135, 175
Cuon (dhole or red dog), 175
Cupuliferoidaepollenites, 195
cyclostratigraphy, see orbital time scale
Cytheropteron testudo, 9, 31, 38, 47, 67
Index
Dacic or Dacian Basin, 216-220

Dama (fallow deer), 119, 153, 175
Daodi fauna (China), 248-249
deer, see cervids
Desmana (water shrew), 175, 194, 217-218
dhole or red dog, see Cuon
diatoms, Plio-Pleistocene, 79-82, 92-93,
230
Dicerorhinus (black rhino), 119, 149, 172175, 186, 194, 204, 216-217, 219, 228
Dicrostonyx (collared lemming), 180, 186
Didacna crassa as Bakuan marker, 225
Diluvial, as part of original Neogene, xii
Discoaster brouweri, 25, 29, 52, 58, 63-66,
71,73,87,146-147,275
discoaster extinction datum, 25, 63, 71, 87,
162, 275
Djetis (Java), 121
Dmanisi (Georgia), 118
Dodogol (Siberia), 120, 228
dogs, see canids
Dolomys (snow vole), 114, 217-218, 222
Domashkinian horizon, 222, 224
Dongcun (China), 120
Donggutuo (China), 134, 249
Dongyaozitou (China), 249
Dreissena mass appearance near PliocenePleistocene boundary, 225
earliest glaciation, evidence for, xv-xvi, xx,
96, 189, 230, 235, 273-274, 280; see also
glacial climate development; ice ages;
Menapian concept; Pliocene-Pleistocene
boundary; Pretiglian concept; Ross
glaciation
E - L - E datum, see Elephas-LeptobosEquus datum
East Turkana (Kenya), 131
Ebersininaia, 217
Eburonian glacial climate phase, Eburonian
stage, 11, 34, 115, 152, 179-181, 183184, 188, 195-196, 203, 222
Eemian interglacial climate phase, Eemian
stage, 107
Elaphurus (Pere David's deer), 242, 249
Elasmotherium, 118
elephants, see Elephas; Mammuthus;
Palaeoloxodon; Archidiskodon
Elephas (Indian elephant), 119, 233-236,
249
Elephas-Leptobos-Equus datum, 114, 153,
235; see also Equus-Elephas-Bos datum;
Middle Villafranchian concept
Elephas planifrons zone below PliocenePleistocene boundary, 236
Eliomys, 172-175
Ellerhoop warm-climate period, 196
Ellobius (mole-lemming), 118, 120
Elphidiella, 186
Elphidium oregonense as indicator for 2.5Ma cold-climate conditions in North Sea
basin, 187
Elsterian glacial climate phase, Elsterian
stage, 203, 222
Emilian stage, xiv, xx, 29, 142
Enhydriodon, 249-250
Ensenadan land mammal age, 123-124
Eolagurus, 118, 227, 229
Eopleistocene, xx, 221, 228, 230
Equus (horse), 114-115, 118-124, 149, 152,
172-175, 186, 216, 219, 227-229, 233236, 249-250, 285
Equus-Elephas-Bos datum of Haug as indicator of Pliocene-Pleistocene boundary,
236; see also Elephas-Leptobos-Equus
datum
Erpfingen (Germany), 194
Etouaires (France), 114, 178, 181
Eucastor, 248
Euceratherium lowest stratigraphic datum,
285
Eucladoceros, 115, 121, 151-152, 172, 174,
204, 249
Eucommia (chinese elm), 107, 110, 185,
195, 204
Euryboas, 228
Fagus (beech), 107-112, 191, 195, 242
Farneta faunal unit, 143, 152, 194
faunal turnover, mammals, 114, 119, 123,
152-153, 236, 259-260
felids, see Felis; Panthera; Lynx; see also sabertooths, cheetah
Felis, 175, 228, 249-250
first glaciation, see earliest glaciation
fish, Vrica section, 21, 32
fission-track ages, 34-35, 54, 122, 135, 178,
242, 269, 276
Flabellipecten planomedius, 169
foraminifera, benthic, Plio-Pleistocene, 2829, 52, 145, 148, 230. 250
foraminifera, planktonic, Plio-Pleistocene,
25-28, 87-92, 142, 161, 169-170, 232,
250
Forest Beds (England), 152
fox, see Vulpes
Galerian land mammal age, 152, 174-175
Gallogoral 114-115, 174-175
Garba (Ethiopia), see Melka-Kunture
Gauss-Matuyama boundary as criterion for
Pliocene-Pleistocene boundary, see
Pretiglian concept; Middle Villafranchian
concept
Gazella (gazelle), 115, 119, 149, 228, 247249
Gazellospira, 115, 118, 175
Gelasian stage, xv, 34, 145
geomagnetic time scale, see paleomagnetic
data
Geomys (gopher) lowest stratigraphic datum, 285
Gephyrocapsa aperta, 47
Gephyrocapsa caribbeanica, 35, 47, 52
Gephyrocapsa group, xv, 23, 35, 63-66, 7 1 73,79
Gephyrocapsa oceanica, 23-24, 35, 38, 52,
64-65, 70, 144, 146-147, 162-163, 167,
250
Gephyrocapsa sinuosa first appearance as in-
Index
dicator of Pliocene-Pleistocene boundary

in the southern ocean, 273, 275
Germanomys, 247
Gigantopithecus, 134
Gilbert chron, see magnetostratigraphy;
paleomagnetic data
Gingkgo, 109-110, 240
glacial climate development, xv, 33, 70, 96100, 149, 162, 191, 209, 223-224, 282; see
also carbonate cycles, oxygen-isotope
data
glacial climates, see Pretiglian; Eburonian;
Menapian; Weichselian
glacial deposits, 191, 203, 230, 235, 273274; see also loess
glacio-eustatic, see sea-level change
Globigerina cariacoensis, 28, 29, 38, 87, 144
Globigerina digitata, G. d. digitata, 28, 87,
170
Globigerina hessi as Pliocene-Pleistocene
boundary indicator in Spain, 170
Globigerinoides fistulosus, 87
Globigerinoides obliquus, G. o. extremus,
28, 38, 47, 52, 87, 144, 169-170
Globigerinoides tenellus, 28
Globoquadrina pachyderma, see
Neogloboquadrina pachy derma
Globorotalia crassaformis, 34, 169-170, 275
Globorotalia inflata, 27, 28, 142, 146, 170,
188
Globorotalia menardii-G. tumida complex
as criterion for Pliocene-Pleistocene
boundary in North Atlantic, 87
Globorotalia miocenica, 7, 87
Globorotalia oscitans, 28
Globorotalia tosaensis, 5, 52, 146, 275
Globorotalia truncatulinoides, G. t. excelsa,
xv, xx, 5, 7, 10, 30, 52, 71-72, 87, 93,
141-142, 146-148, 170, 232, 251, 275
Globorotalia umbilicata, 28, 87
Glossotherium, 123
Glyptostrobus, 240
Glyptotherium, 123
Gombore (Ethiopia), see Melka-Kunture
gomphothere, see Anancus; Stegomastodon;
Serridentinus
Gongwangling (China), 121, 134
Grenzbank (Java), 122, 134
GSSP (global-boundary stratigraphic section and point) definition of PliocenePleistocene boundary, xi, 11, 29
Gundersheim (Germany), 194
Gurian stage, 224-225
Hadar Formation (Ethiopia), 132
Hajnacka (Slovakia), 114, 181
hares, see Lepus; Oryctolagus; Prolagus;
Pliopentalagus; Hypolagus; see also rabbits
Hautawan stage, 273-27r4
Hauterives (France), 197
Helicosphaera sellii, 25, 47, 52, 59, 63-66,
71, 73, 275
Hemibos, 111
Hesperoceros, 11A
Hexaprotodon, 236
Higueruelas (Spain), 172
Hipparion, 115, 119, 124, 172, 216, 227,
247-249
Hippopotamus, 118, 120, 153, 174-175
hippos, see Hippopotamus; Hexaprotodon
Holsteinian interglacial flora, 107
hominids, see Australopithecus;
Gigantopithecus; Homo erectus; Homo
ergaster; Homo habilis; Homo rudolfensis; Meganthropus; Paranthropus;
Pithecanthropus
Homo erectus, xix, 118, 120-121, 129-135,
175, 223, 250, 255-258, 268
Homo ergaster, 129, 132
HomoMbilis, 119, 129-135
Homo rudolfensis 131, 135, 223
horses, see Equus; Hipparion; Nannippus;
Pliohippus; Proboscidipparion
Huescar (Spain), 175
hunting-dog, see Lycaon
Hyalinea baltica, first appearance in Mediterranean, xv, xx, 4, 29, 32, 38, 67-70,
73, 141, 147-148, 163-164, 167-168; first
appearance in China, 250-251
hyenas, see Chasmaporthetes; Crocuta;
Pachycrocuta
Hypolagus, 211, 247, 284-285
Hystrix (Old world porcupine), 175, 179
ice ages, xvi, xix, 70
ice-rafted debris, xv
Icenian Crag, 181
Icenian stage, 186, 198
Ilex (holly), 107-112, 203
Iliyskaya (Kazakhstan), 120
INQUA (Internation Quaternary Association), 4, 6, 8-11, 37-38, 46-47, 70, 147,
221,239
intercontinental migrations of mammals during ice ages, 121, 123, 266, 271, 280, 282
interglacial climates, see Brunssumian;
Reuverian; Tiglian; Waalian; Holsteinian; Eemian
Ionian stage, xv
IRD, see ice-rafted debris
Irish elk, see Megaceros
Itantsa faunal complex, 228; see also
Tamanian
Irvingtonian land mammal age, 122-123,
278-286
IUGS (International Union of Geological
Sciences), xii, 11, 39
Ivanovce (Slovakia), 180
jackal, see Canis
Jaramillo subchron, see magnetostratigraphy; paleomagnetic data
Java Man, see Homo erectus
Juglans (walnut), 107-112, 240-242
K/Ar ages, 34-35, 129-135, 172, 178-179,
184, 255-259, 269, 276
Kabuh Formation (Java), 121, 134, 268-270
Kadzielna (Poland), 124, 181
293
Kaena subchron, see magnetostratigraphy;

paleomagnetic data
Kaiso (Uganda), 258
Kaliglagah (Java), 121, 264-266
Kaltensundheim (Germany), 115, 203
Kamyk (Poland), 181
Kanam (Kenya), 258
Kanjera (Kenya), 258
Karewa section (Kashmir), 7, 233-234
Kazakhstan Plio-Pleistocene faunas, 120
KBS Formation; KBS Tuff (Kenya), 119,
256
Kedung Brubus (Java), 121, 268
Khaprovian faunal complex, 7, 115, 222
Kimballian stage, 282
Kislang (Hungary), 115
Kizhinskian or Kizhikan faunal complex
(western Siberia) 120, 181, 222
Kolinany (Russia), 118
Kolkotovian mollusk assemblage, 224
Koobi Fora (Kenya), 131, 255-256
Kotlovina (Moldavia), 181
Kromdraai (South Africa), 119, 133, 259
Kryzhanovka (Ukraine), 115, 181, 219, 223
Kujalnik (Ukraine), 181, 223
Kujalnikan or Kuyalnikan stage, 224
Kuruksai (Tadjikistan), 8, 112, 120
Kutujakh or Kututyak (Siberia), 230
Laetolil (Tanzania), 258
Lagurodon, 115,120,123
Lagurus (steppe lemming)
Lakuti or Lakhuti (Tadjikistan), 8
Larix (larch), 108-111, 217, 242
Lausitz lignites (Germany), 201-203
Le Castella section (Calabria), 6, 9, 15, 71,
147-148
Le Coupet (France), 118, 179
lemmings, see Dicrostonyx, Ellobius,
Eolagurus, Lagurodon, Lagurus,
Lemmus, Prolagurus, Synaptomys
Lemmus (lemming), 179
leopard, see Panthera and Felis
Leptobos, 118, 152, 174-175, 186, 204, 219,
235
Lepus (hare), 122-123, 175, 284
Levantine mollusk assemblage, 217, 223
Lieth series and paleoflora (Germany), 195
lignites, 201-203
lion, see Panthera
Limnaea mass appearance near PliocenePleistocene boundary, 225
Liquidambar (sweet gum), 33, 109, 112,
185, 195, 203, 240
Liriodendron (tulip magnolia), 33, 108
Lishi loess (China), 10, 120, 247
Liventsovka (Russia), 115, 181
Lochuan Loess fauna, 248
loess (loss), 10, 115, 120, 247, 251
Lower Villafranchian concept for PliocenePleistocene boundary, 7, 219
Loxodonta, 119
Ludhamian stage, 183-184, 188, 197
Lutra (otter), 194
Lynx, 174, 228, 248; see also Felis
294
Macaca (macaque), 179

Macrocypris adriatica, 30
magnetostratigraphy in Plio-Pleistocene sections, 35, 46-54, 76, 93, 129-132, 142144, 147-148, 242, 244, 250, 255-258,
281-286; see also paleomagnetic data
Magnolia, 107, 110, 113
Makapansgat (South Africa), 133, 259
Malan loess mammalian fauna (China), 10,
247
Malusteni (Romania), 114
Mammoth subchron, see magnetostratigraphy; paleomagnetic data
mammoth, see elephants
Mammut ("woolly mastodon"), 149, 194
Mammuthus (mammoth elephant), 114
115, 118-119, 122-123, 149, 152, 172175, 183, 217-219, 242, 282, 285
mammutids, see Mammut; Zygolophodon
Mangapanian stage, 273
Marplatan land mammal age, 123
Mastodon, 115
Mastodon-Schotter (gravels), 203
Matuyama chron, see magnetostratigraphy;
paleomagnetic data
Megaloceros, 249
Megalovis, 115, 219
Megantereon, 119, 175
Meganthropus, 268; see also Homo erectus
Meles (badger), 175, 249
Melka Kunture (Ethiopia), 132, 255
Menapian concept of Pliocene-Pleistocene
boundary at 0.9 Ma, xv, 7, 152-153; see
also Cromerian concept
Menapian glacial climate phase, Menapian
stage, xv, xx, 119, 171, 188
Merxemian stage (Merksem Beds), 188, 198
Mesopithecus, 219
Metasequoia (dawn redwood), 239-242
Metasequoia extinction as criterion for 0.9-Ma
Pliocene-Pleistocene boundary, 239-240
Metridiochoerus, 119
mice, see Apodemus; Mas
microtine rodents, see Microtus; Mimomys;
Pitymys
microtine rodent event-stratigraphy, 280
Microtodon, 229
Microtocoptes, 229
Microtus (meadow vole), 118, 123-124,
152, 174-175, 194,222,249
mid-Pliocene cold climate interval, xv, 7, 8,
10, 149, 162, 164, 188, 197, 203-204, 223,
227, 230, 250; see also Pretiglian concept;
Middle Villafranchian concept
Middle Villafranchian concept of PliocenePleistocene boundary at 2.5 Ma, 7, 10,
219,247,250-251,260
migration, see intercontinental migrations
Mimomys, 114-115, 118, 172-175, 178181, 186, 194, 196, 204, 218, 222, 227229, 249
Mimomys zonation, 178-181
Mindel glacial, xx; 119, 222
Miniopterus (long-fingered bat), 174-175
Miocene stratigraphy, North Sea basin, 197
Index
MN (Mammal Neogene) zonation, 114,

179, 186, 192-195, 219
Modjokerto (Java), 135
Moldavian faunal complex, 120, 222
moles, see Talpa; Myospalax
mollusks, fresh-water, Plio-Pleistocene,
191-192, 197, 203, 216-219, 223-224
mollusks, marine, Plio-Pleistocene, 18, 32,
52, 145, 169, 186, 197-198
Monasterace section (Italy), 146
monkeys, see Macaca; Mesopithecus;
Paradolichopithecus; see also baboons
Montalbano Jonico section (Italy), xv
Monte Mario section (Rome), 38, 145
Monte Peglia (Italy), 118
Monte San Nicola section (Sicily), xv
Monte Singa section (Italy), 18
Montopoli (Italy), 114
Montopoli faunal unit, 149, 194, 222
moose, see A Ices
Morozovian mollusk assemblage, 224
Morsumian stage, 197
MQ (Mammal Quaternary) zone, 192, 195;
see also MN zone
Mus (mouse), 248
musk-ox, see Hesperoceros; Megalovis;
Praeovibos
muskrat, see Ondatra
Musone section (Italy), 146
My a (steamer clam), 186
Myospalax, 247
Myotis (little brown bat), 174-175
Nannipus, 282
Nannocricetus, 247
nannofossils or nannoplankton, see calcareous nannoplankton
Nebraskan glacial age as criterion for
Pliocene-Pleistocene boundary, 280
Neogene, definition of, xii
Neogloboquadrina atlantica, 28, 34
Neogloboquadrina pachyderma, 27-28, 38,
47, 146, 170, 251
Neuleiningen (Germany), 118, 180, 195
Nihewan faunas (China), 120-121, 134,
247-249
Njurgan or Nyurgan (Siberia), 228-229
Nogaisk (Ukraine), 118, 181
Nordende warm-climate period, 196
Nordhausen (Germany), 204
Northern guests, see Arctica islandica;
Cytheropteron testudo; Hyalinea baltica
Norwich Crag (England), 181
Notochoerus, 119
Nukumaruan stage, 273-274
Nunivak subchron, see magnetostratigraphy; paleomagnetic data
Nyanzachoerus, 119
Nyctereutes, 115, 149-150, 219, 228, 249
Nyssa (swamp gum), 107-108, 111, 113,
185, 195, 203, 240
Nyurgan, see Njurgan
Ochotona (pika), 179, 248
Ochotonoides, 249
Odessan faunal complex, 7, 118, 222

Odocoileus (American deer), 285
Okote Formation, Okote Tuff (Kenya),
119, 131
Oldowan-type tools, 169, 255-258; see also
artifact
Olduvai Gorge (Tanzania), 119, 132, 256258
Olduvai subchron, fossil mammal context,
115-118, 120-123, 241, 245, 248 (fig.
only)
Olduvai subchron, micropaleontological
context, 25, 27-28, 54, 73, 87-94, 230,
245, 273
Olduvai subchron, paleobotanical context,
186, 240-242
Olduvai subchron, see magnetostratigraphy;
paleomagnetic data
Olduvai subchron, stable isotope context,
94-97, 260
Olduwan, see Oldowan
Olior or Olyor (Siberia), 230
Olivola (Italy), 115, 124, 152
Olivola faunal unit, 11, 151-153, 194, 219,
222
Olkhon Island, Lake Baikal (Russia), 228
Olkhov beds (Siberia), diatom microflora
bracketing Pliocene-Pleistocene boundary, 92, 230
Ondatra (muskrat), 122, 282, 285
orbital time scale (orbitally tuned ages) xiii,
39, 141
Orce (Spain), 118, 124, 171, 175
Orientalomys, 247
Orthostonyx, 119
Orwellian stage, 197
Oryctolagus, 174-175
ostracodes, Plio-Pleistocene, 18, 29-30,
145, 250
Ostrea lamellosa, 169
Osztramos (Hungary), 181
otters, see Lutra; Paralutra; Enhydriodon
Ovis (sheep), 118
Oxycoccus (cranberry), 242
oxygen-isotope data, xv, 23-25, 39, 94-97,
244, 280
Pachycrocuta, 115, 152, 172, 175
Palaeoloxodon, 175, 242, 249
paleobotanic data, 32-34, 105-113, 185186, 192, 195-196, 203-204, 212-214,
223, 229-230, 240, 242
Paleolithic, see Acheulean; Oldowan
paleomagnetic data, xi, 35, 39, 54, 120,
122-123, 141, 186. 196, 210, 223, 227230, 236-237, 239-242, 247, 256, 266,
268-269, 275
Paleotragus, 228, 247
Palikao, see Tighenif
Paliurus (christ's-thorn), 242
palynologic data, see paleobotanic data
pandas, see Parailurus
Pannonian stage, 209
Panthera (leopard, lion, tiger), 115, 175,
249-250
Index
Paradolichopithecus, 219
Parailurus, 216
Paracamelus, 227, 249
Paralactaga, 248
Paranthropus, 119, 131, 133
Parapapio, 119
Paratethys tectonic-sedimentologic concept
of Pliocene-Pleistocene boundary, 209210, 220
Pecten jacobaeus, 169
Pelorovis (giant sheep), 119-120
Peninj (Tanzania), 132, 258
Perrier, see Peyrolles
Perrier-Etouaires, see Etouaires
Petaurista (flying-squirrel), 179
Petralona (Greece), 133
Peyrolles (France), 152
Phacochoerus (wart hog), 119
Phenacomys, 123
Phyllodendron, 107-109
Piacenzian stage, xv, 7, 11, 66, 145, 223; see
also Castell'Arquato
Picea (spruce), 33-34, 47, 107-112, 204,
217, 229, 240-242
pigs, see suids
pikas, see Ochotona; Ochotonoides
Pinjor stage, 121, 234-236
Pinneberg warm-climate period, 196
Pinus (pine), 33-34, 107-112, 204, 217,
229, 240-242
Pithecanthropus, 134; see also Homo erectus
Pitymys (pine mouse), 152, 222, 249; see
also Microtus
Platycarya, 111-112, 204
Pleistocene boundary, definition of, xi, xix,
4, 7, 9, 11, 15, 38, 66, 70, 224, 250; see
also Cromerian (end-Villafranchian) concept; earliest glacial climate; Paratethys
tectonic concept; Pretiglian concept;
Lower Villafranchian concept; Middle
Villafranchian concept
Pleistocene boundary, local recognition of,
96,141,164,181,184,188,198,204,211,
219,221,225,250-251,256-259,271,276
Pleistogene, xii
Pliohippus highest stratigraphic datum, 285
Pliomys, 115, 222
Pliopentalagus, 247-248
PMTS (paleomagnetic time scale), see
paleomagnetic data
Podocarpus (black pine), 33, 47, 242
Podpusk (Siberia), 120, 181
Podpusk-Lebyaginsk faunal complex, 120,
222, 227, 230
pollen, see paleobotanic data
porcupine, see Hystrix
Porika glaciation (New Zealand), 273
Potamides, 217
potassium-argon age, see K/Ar age
Praedama, 115
Praeovibos, 115, 175
Pretiglian (Praetiglian) cold-climate phase,
Pretiglian stage, 34, 96, 178-179, 185189, 195-198, 223; see also mid-Pliocene
cold-climate interval
Pretiglian concept of Pliocene-Pleistocene

boundary, xv, 7, 188, 198, 206, 208-209,
251; see also Middle Villafranchian concept; Paratethys tectonic concept
Proamphibos, 236
Proboscidipparion, 247-249
Procamptoceras, 115
Prolagurus, 222
Prolagus, 179
Promimomys, 227
Prosiphneus, 228-229, 247-248
Prospalax, 216
Protarctos, 216
Psekups (Ukraine), 118, 222
Pseudodama, 115
Pseudoemiliana lacunosa, xv, 66, 71, 144
Pseudolarix, 107, 240
Pseudotsuga (Douglas fir), 110
Psilunio, 216-217
Pterocanium prismatium extinction datum,
93
Pterocarya (wing-nut), 108-112, 185, 195
Puebla de Valverde (Spain), 115
Pulleniatina, coiling direction, 118
Pyramidella plicosa zone in Israel, 162-163
Quaternary, definition of, xii, xv, xix-xx, 4,
6-7,9
Quercus (oak), 33, 34, 107-112, 204
Rabat (Morocco), 133
rabbits, see Sylvilagus, Hypolagus
radiolarians, Plio-Pleistocene, 92-94
Rancholabrean land mammal age, 278
Rangifer (reindeer), 123
rats, see Rattus; Sigmodon
Rattus (rat), 248
Rebielece (Poland), 181
Red Crag (England), 183, 188
Reunion subchron, see magnetostratigraphy; paleomagnetic data
Reuver (Netherlands), 185
Reuverian climate phase, Reuverian stage,
107, 183, 185, 188, 191, 194, 196, 203
Rhinoceros, 242, 250
rhinoceroses, see Chilotherium; Coelodonta;
Dicerorhinus; Elasmotherium;
Euceratherium; Rhinoceros
Rhinolophus (horseshoe bat), 174-175
Rhizosolenia praebergonii var. robusta last
appearance datum as indicator of
Rhododendron, 112
Rincon (Spain), 114-115
Rippersroda (Germany), 204
Roccaneyra (France), 115
Ross glaciation (New Zealand), 273
Rugunio, 111, 219
Rusa, 121,242,249
Ruscinian land mammal age, 7, 193-194
sabertooths, see Homotherium;
Megantereon; Smilodon
Saint-Vallier (France), 115, 149, 181, 222
Saint-Vallier faunal unit, 149, 194
295
Salix (willow), 107-108, 110-112

Salvinia (sage), 34
San Pedro Valley (Arizona), 122
Sangiran (Java), 121-122
Santa Maria di Catanzaro (Calabria), xiii,
6,9,70-71, 148-149
Santernian stage, xiv-xvi, xxi, 11, 142
Santerno section (Po Valley, Italy), 38, 142144
Sapium (Chinese tallow), 242
sapropelites, Plio-Pleistocene, 18-21, 5761,79-82
Scaldisian stage, 197
Schernfeld (Germany), 180, 194
Sciadopitys (umbrella-pine), 110, 185, 203,
223
Sciurus (tree-squirrel), 194
sea-level change as glacio-eustatic effect,
120-121, 149, 162, 171-172, 230, 264,
271,276
Selinuntian stage, xiii-xvi, xxi, 142, 225
Semliki (Zaire), 258
Seneze (France), 115, 149, 152, 219, 222
Sequoia (redwood), 107, 109, 111, 185, 195,
197, 223
Serridentinus, 250
Serripes groenlandicus (Greenland venus
clam), 186
Sete (France), 179
shrews, see Sorex; Desmana; Crocidura;
Drepanosorex; Beremendia
Shungura (Ethiopia), 119, 124, 129-131,
255-256
Sicilian stage, xiii-xv, xx, 71, 73, 142, 164
Sigmodon (cotton rat), 282
Simatherium, 119
Singa, see Monte Singa
Sinocastor, 228
Sinoryx, 247
Sinzelles (France), 118
Sivatherium extinction as indicator for
Smilodon (sabertooth), 122-123, 282, 285
Sminthoides, 248
Solilhac (France), 118
Solo (Java), 122
Sorex (long-tailed shrew)
South American Plio-Pleistocene localities,
123-124
Sphaeroidinellopsis seminulina last occurrence datum in Israel, 161-162
squirrels, see Citellus; Sciurus
stable isotopes, see oxygen-isotope data
Stegodon, 120, 121, 236, 242, 250
Stegodon zones in Japan, 121, 242
stegodonts, see Stegodon; Stegolophodon
Stegolophodon, 250
Stegomastodon, 123
Sterkfontein (South Africa), 119, 133, 259
Stirone section (Italy), 70, 142
stone tools, see artifact
Stuni-Vrica sequence, see Vrica section
suids, see Afrochoerus; Metridiochoerus;
Notochoerus; Nyanzachoerus;
Phacochoerus; Sus
296
Suifun suite (Siberia), 230

Sus (pig), 149, 172, 186, 216, 249
Susterian climate phase, Susterian stage,
196-197
Swartkrans (South Africa), 119, 133, 259
Syltian stage, 196-197
Sylvilagus (cottontail rabbit), 122-123
Symplocoipollenites, 195
Synaptomys (bog lemming), 123, 180, 282
Taiwan, Plio-Pleistocene microfaunas, 56
Talpa (mole), 217
Tamanian faunal complex, 222, 228
Tapinochoerus, 119
Tapirus (tapir), 149, 179
Tasso faunal unit (Italy), 152, 194
Tatrot stage, 121, 234-236
Taung (South Africa), 259
Taxodium (cypress), 33, 47, 107, 109-113,
185, 196, 203, 223
Taxodium assemblage extinction as indicator of 2.5-Ma cold climate conditions in
western Europe, 113, 196
Tchikoian or Chikoian faunal complex (Siberia), 228
Tegelen or Tegel (Netherlands), 115, 153,
180, 185, 222; see also Tiglian
Tell 'Ubeidiya (Israel), 120, 135
Ternifine, see Tighenif
Tetoiu or Bugiulesti (Romania), 118, 219
Thalassiosira convexa last appearance datum above Gauss, 92
thermophilic flora, 106
thermophilic molluscs, 217, 223-224; see
also Levantine mollusc fauna
Theropithecus, 119
Thurnian stage, 183, 197
Tichonovka (Russia), 118, 181
tiger, see Panthera and Felis
Tighenif (Morocco), 133
Tiglian climate phase, Tiglian stage, xvi, 34,
107, 179, 183, 185-186, 191, 195-198,
203-204, 223
Tilia (linden), 33, 107-112, 204
Tiraspolian faunal complex, 7, 222, 229
Tjornes section (Iceland), 184
Index
Toringian mammal age, 193

Tornesch warm-climate period, 196
Torrea (torrey pine), 107, 109, 113
tortoises, as climate indicators, 179
Transvaal caves (South Africa), 119
Tremarctos (spectacled bear) lowest stratigraphic datum, 285
Tricolporopollenites, 195
Trinil (Java), 121
Triversa faunal unit, 149
Trogontherium, 219, 229
Tsuga (hemlock), 33, 47, 107-112, 185, 195,
197, 203, 217
tuff, volcanic, see volcanic ash layer
Tulucesti (Romania), 217
Turkana Basin (Kenya-Ethiopia), 119, 129,
131,235-236
Turritella (turret snail), 186
Tyrrhenian stage, 163
Ubeidiyah, see Tell 'Ubeidiya
Uetersen warm-climate period, 196
Ulmus (elm), 33-34, 107-112, 203
Unio (freshwater mussel), 219
unionid mollusks, see Psilunio; Rugunio;
Unio
Untermassfeld (Germany), 118-119
Uquian land mammal age, 123-124
Ursus (bear), 119
Uvigerina bradyana first appearance as indicator of Pliocene-Pleistocene boundary
in Mediterranean, 29
Val Musone section (Ancona, Italy), 38
Valdarno (Italy), 149, 152
Valdeganga (Spain), 115, 175
Vallebiaja section (Italy), 145
Vallecito Creek, see Anza-Borrego
Vallonet (France), 118
Venta Micena (Spain), 118, 135, 174-175,
181
Villafranchian land mammal age, 4, 7, 120,
149-153, 192, 217, 222
Villany (Hungary), 115, 124, 174
Villanyan land mammal age, 174-175, 193195
Villanyan-Biharian boundary equivalent to

Villanyia, 118, 222, 227-229
Vitis (grape), 107-108, 110-112
Viverra (civet), 247
Viviparus (freshwater snail), 216-217, 219
volcanic ash layers near PliocenePleistocene boundary, 34-35, 47, 54, 119,
122, 129, 131-132, 239, 245, 254-258,
269, 278
voles, see Allophaiomys; Arvicola;
Dolomys; Microtus; Mimomys;
Promimomys
Vrica section (Italy), xii-xvi, xix-xxi, 6, 911, 15-42, 46-47, 52-54, 57, 71-72, 7982, 87, 97, 149, 181, 185, 188, 219
Vulpes (fox), 194
Waalian interglacial climate phase, Waalian
stage, 107, 118, 179, 195-196
Waipipian stage, 273-274
Waltonian stage, 197-198
wart hog, see Phacochoerus
water-buffalo, see Bubalus
West Turkana, see Turkana Basin
Weybourne Crag (England), 181
Weze (Poland), 180
wolf, see Canis
Wolfersheim (Germany), 179, 192, 194
Woolly mastodon, see Mammut
Woolly rhinoceros, see Coelodonta
Wucheng loess (China), 120, 247
Xiaodukou (China), 120-121, 249
Yoldia (pea clam), 186
Yuanmou (China), 120-121, 134, 249-250
Yuxian or Yushe Basin (China), 121, 248249
Zabythocypris antemacella, 30
Zasuhino (Siberia), 120-121
Zelkova, see Ulmus
Zersatz-Kies (Germany), 204
Zygolophodon, 203, 219, 249; see also
Mammut

The Pleistocene Boundary and The Beginning of The Quaternary (John A. Van Couvering, 2001)

Uploaded by

Copyright:

Available Formats

The Pleistocene Boundary and The Beginning of The Quaternary (John A. Van Couvering, 2001)

Uploaded by

Document Information

Original Title

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

The Pleistocene Boundary and The Beginning of The Quaternary (John A. Van Couvering, 2001)

Uploaded by

Copyright:

Available Formats

The Pleistocene Boundary and the

Beginning of the Quaternary

WORLD AND REGIONAL GEOLOGY SERIES

Geological Evolution of Antarctica - M. R. A. Thomson,

2 Permo-Triassic Events in the Eastern Tethys - W. C. Sweet,

The Jurassic of the Cricum-Pacific - G. E. G. Westermann

Tertiary Basins of Spain: The Stratigraphic Record of

The Quaternary History of Scandinavia - J. Donner

The Tectonic Evolution of Asia - A. Yin and

Final report of the International Geological Correlation

Coordinator: Ksenia V. Nikiforova,

PUBLISHED BY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE

Part IV The Pleistocene boundary in regional sequences

JOHN A. VAN COUVERING

RUGGIERI, SAMUELE SARTONI, AND RODOLFO SPROVIERI

Part I Definition of the base of the Quaternary

GEDALIAHU GVIRTZMAN, GUIDO M. MARTINOTTI, AND SHIMON

The Pliocene-Pleistocene transition in the Iberian

Biostratigraphy and calibrated climatic chronology of

The Plio-Pleistocene of England and Iceland

Part II Characterization of the Pleistocene boundarystratotype

The Pliocene-Pleistocene boundary-stratotype at Vrica,

The magnetostratigraphy of the Vrica section, Italy, and

The Neogene-Ouaternary boundary in The

The Tertiary-Quaternary boundary in western

TOKUNAGA, HIROSHI KITAZATO, TOYOSABURO SAKAI, AND

Comparison of the laminated units at Vrica and deep-sea

Calcareous nannofossil biochronology and the PliocenePleistocene boundary

GUNTHER VON DER BRELIE, KARL BRUNNACKER, WINFRJED

The Plio-Pleistocene of Hungary

Part HI The paleontological context of the Pleistocene

The Pliocene-Pleistocene boundary in Romania

The Pliocene and Pleistocene of the European part of

The N/Q boundary in Asian Russia and

The Pliocene-Pleistocene boundary in the Indian

The Pliocene-Pleistocene boundary in Japan: the

The Pliocene-Pleistocene boundary in deep-sea

WILLIAM A. BERGGREN AND LLOYD H. BURCKLE, JR.

Late Cenozoic changes of flora in extra-tropical Eurasia

EMILIANO AGUIRRE, ELEANORA A. VANGENGEIM,

JORGE MORALES, MARINA V. SOTNIKOVA, AND

Human evolution in the Plio-Pleistocene interval

The Pliocene-Pleistocene boundary in eastern

DOMENICO RIO, ISABELLA RAFFI, AND JAN BACKMAN

Plio-Pleistocene mammal faunas: an overview

YASUMOCHI MATOBA, MOTOYOSHI ODA, SHIGEMOTO

Stratigraphy of the Plio-Pleistocene sequence of the

KSENIA V. NIKIFOROVA AND MIKHAIL N . ALEKSEEV

NAKAGAWA, GIANCARLO PASINI, DOMENICO RIO, GIULIANO

Diatom microfossils from the Vrica section

The Pliocene-Pleistocene boundary in Italy

AUGUSTO AZZAROLI, MARIA LUISA COLALONGO, HISAO

MINORU ITIHARA, SHUSAKU YOSHIKAWA, AND TADAO

The Pliocene-Pleistocene boundary in Japan:

HISAO NAKAGAWA, NOBUAKI NIITSUMA,

The Pliocene-Pleistocene boundary in New

The Pliocene-Pleistocene boundary in continental

TAKASHI MITSUNASHI, MOTOYOSHI ODA,

TOSHIAKI TAKAYAMA, AND TOYOSABURO SAKAI

Cobb Mt. (Gils)