JOURNAL OF SEDIMENTARY PETROLOGY, VOL. 34, NO. 4, PP.

830-847
FIGS. 1-11, DECEMaER, 1964

GRAIN SIZE REPRESENTATION BY CM PATTERNS AS A

GEOLOGICAL TOOL 1

R. PASSEGA
Consultant Bartlesville, Oklahoma

ABSTRACT
The characteristics of a depositional agent are reflected in the texture of the sediment. This relationship is
particularly evident if the texture is represented by two parameters of the grain size distribution: C the one
percentile and M the median diameter. CM patterns formed by sample points off a deposit are characteristic of
the agent of deposition.
Since the first discussion of CM patterns (Passega 1957) more than ten thousand analyses of sediments rang-
ing in age from Paleozoie to Recent, sampled in basins of all types and sizes, including the Rhine graben,
mobile basins of the Apennines and southeast Iran, and Cratonic Sahara basins, confirmed the conclusions of
the earlier paper. They also disclosed new relationships between CM patterns and sedimentation, valid under
greatly different conditions.
The most notable progress of recent research was the construction of paleobathymetric maps and Iogs show-
lug the depth of ancient seas and the variations of this depth in a stratigraphic section.
Paleobathymetry is geologically important as it seems related to such characteristics of the sediments as
facies, permeability and possibly also petroleum productivity. CM patterns are a means for analyzing trans-
portation mechanisms and determining the mechanisms that form peculiar deposits, such as stratigraphic traps.
Turbidites and certain tractive current deposits have several common characteristics. Their texture shows
that they may have a common origin as wave suspensions.
The many applications of grain size analyses indicate that they are a versatile means of investigation that
discloses the orderly arrangement of a number of geological factors. This order is a fertile field for geological re-
search,

I. INTRODUCTION wide application as a means of geological investi-

For m a n y years grain size analyses were used gation.
by geologists in a t t e m p t s to characterize depo- A clastlc deposit is formed by sediments trans-
sitional e n v i r o n m e n t s of clastic sediments. T h e ported in different ways. In particular, the finest
trend of modern research has been toward a fractions m a y be t r a n s p o r t e d i n d e p e n d e n t l y of
definition of individual samples of a deposit by the coarser particles. Swift s e d i m e n t a r y agents
p a r a m e t e r s of their texture a n d a definition of can be characterized best by p a r a m e t e r s which
the deposit as a whole by the variation of these give more information on the coarsest t h a n on
parameters. U n f o r t u n a t e l y this definition, gen- the finest fractions of their sediments. Two
erally, is not sufficiently accurate to characterize p a r a m e t e r s of t h e grain size distribution of indi-
sedimentary deposits. Furthermore, a consider- vidual samples are particularly significant: C,
able difference between recent and ancient sedi- the one percentile, a n d M, the median diameter.
m e n t s (Passega, 1962) complicates the identifi- I t was shown t h a t C M p a t t e r n s on a logarithmic
cation of ancient environments. diagram are characteristic of the depositional
Textural parameters, however, can give useful a g e n t (Passega, 1937). Since this first discussion a
information a b o u t the e n v i r o n m e n t . F r i e d m a n large n u m b e r of deposits of all types, mostly
(1961) showed t h a t mean, s t a n d a r d deviation European and North-African, were analyzed.
and skewness permitted a distinction between The new d a t a confirm and define with added
beach sands and dune or river sands. He also precision the relationships between C M p a t t e r n s
gave (1962) a useful correlation between sorting, and mode of deposition.
measured by s t a n d a r d deviation, and the deposi- This paper begins with a discussion of t h e
tional e n v i r o n m e n t of sediments. different p a r t s of a C M p a t t e r n in t e r m s of proc-
The present paper approaches t h e relation- esses of t r a n s p o r t a t i o n a n d deposition. T h e
ships between texture and e n v i r o n m e n t some- discussion is illustrated by examples from differ-
w h a t differently. It shows t h a t fairly precise ent environments.
relationships exist between sediment texture a n d T h e discussion of t h e p a t t e r n s is followed by
t r a n s p o r t and t h a t these relationships can have a sections t r e a t i n g the application of the p a t t e r n s
for estimating the d e p t h of ancient seas, for
a Manuscript received January 20, 1964; revised comparing t r a n s p o r t a t i o n by t r a c t i v e a n d tur-
May 10, 1964. bidity currents, for s t u d y i n g the genesis of strati-
 G R A I N S I Z E R E P R E S E N T A T I O N B Y CM PA T T E R N S 831

graphical traps and sedimentary structures.

Sampling techniques are briefly mentioned.
Most of this work was a part of sedimentologic
research projects of AGIP, San Donato Mila-
nese, Italy, and most of the analyses were made
in the laboratories of this company. A part of
this work was sponsored by the companies of
Bureau des Recherches de Pgtrole, Paris, and
Compagnie Franqaise des Pgtroles, Paris, and
particularly by Compagnie des Pgtroles Total
(Libya).
A number of geologists of these companies
participated in the field and laboratory work and
in the interpretation of the data. These geolo-
gists collaborated in the publication of papers or
in the writing of internal company reports and
their names are listed in the bibliography. FIG. l.--Complete CM pattern of
tractive current deposit.
Thanks are due particularly to the ~ollowing
persons in the management of the petroleum
companies who initiated and encouraged sedi- Mayersville and the grain size distribution and
mentologic research: C. Zanmatti, E. Egidi, T. concentration of the load suspended in the river,
Rocco, G. Long, C. Loddo, L. Lncchetti (AGIP), sampled at regular depth intervals, gave good
R. Lgvy, A. Perrodon, G. Lugol, J. J. Burger, C. indications about the mechanism of transport by
Temp~re (B.R.P.), H. Vautrin, P. F. Burollet, tractive currents (Passega 1957).
C. de Lapparent (C.F.P.), R. Byramjee. Some grains are transported only by rolling.
P. H. Kuenen critically reviewed this paper They never are in the suspended load. Just above
and made a number of suggestions. the bed, a part of the suspension is graded; that
is, its maximum and median grain size as well as
l l . CM PATTERNS AND MODE its concentration in the water increase inversely
OF TRANSPORTATION to the height above the bed. This is the part of
CM patterns are constructed by representing the suspension having the greatest concentra-
on a logarithmic diagram approximately 30 tion. In the Mississippi, at Mayersville, the
samples of a deposit by two parameters of their graded suspension is at the most three meters
grain size distribution: C, the one percentile and thick.
M, the median diameter. Each sample should be Above the graded suspension is a more dilute
a deposit of homogenous sedimentation. The 30 suspension having approximately the same con-
samples should represent all textures available. centration and maximum grain size at any point
Bottom currents capable of transporting sedi-
ments are of two types. Tractive currents are
capable of transporting their load either by
rolling or in suspension. Turbidity currents
generally flow during a limited time and are so
rapid that they cannot be followed by rolling
grains. Their load is entirely a suspension load.
Exceptionally, turbidity currents can roll par-
ticles relatively short distances from their origin,
if the currents are formed continuously during a
certain time, or longer distances, if, in an area of
non deposition, the particles are rolled by suc-
cessive currents.
Rivers, marine currents and waves touching
bottom are tractive currents.
~Fig. 1 is the most complete CM pattern of
tractive current deposits, whereas Fig. 2 is a typi-
cal pattern of turbidites (deposits of turbidity
currents). The meaning of these patterns in
terms of meehanism of transport is illustrated by
analysis of a few patterns. FIo. 2.--A typical CM pattern of turbidites of the
The texture of the Mississippi River bed near Marnoso-arenacea formation, Northern Italy.
832 R. PA SSEGA

of the river section, independently of its depth.

This suspension extends in the Mississippi from
the graded suspension to the surface of the river
and is as much as 20 meters thick.
In the Mississippi there is always a uniform
suspension while a graded suspension may be
lacking. When both suspensions are present, the
maximum grain size in the upper part of the
graded suspension is the maximum grain size of
the uniform suspension.
These different transports are well represented
by the different parts of CM diagrams. This can
be shown by comparing sediments and transport
in the Mississippi and other environments of
tractive and turbidity currents. The Mississippi
sedimentation at Mayersville has already been
discussed (Passega 1957) and will only be re-
viewed here. FIG. 3.--A, B, C, grain size cumulative curves of
The CM patterns of all sediments of deposits the Marnoso-arenacea turbidites represented by
of tractive currents that were investigated con- Fig. 2. D cumulative curve of Mississippi sediment of
segment PQ of the CM pattern.
form to the general pattern of Fig. 1, although
they are less complete. The general pattern is
divided into different segments by points N, O, graded suspension. On both patterns, the maxi-
P, Q, R, S. mum grain size of the graded suspension can be
The grain size analyses of the suspended load designated as Cs (see Figs. 1 and 2).
of the Mississippi show that the value "Cs," of C The analyses of the Mississippi sediments
at point Q, corresponds to the largest size trans- showed that particles larger than Cs are found
ported as a graded suspension and that the value only on the bed of the river, never in suspension.
Cu, of C at point R, generally is the largest grain These particles probably are transported by
size transported as a uniform suspension. There- rolling. It was therefore assumed that segment
fore it can be assumed that the sediments trans- PQ of the tractive pattern represents a deposit
ported as a graded suspension form deposits formed by the sediment of point Q, to which
represented by points of segment QR. These are, rolling adds a small amount of grains which
among the points of the complete pattern, those affects the value of C without changing M. Fig.
situated nearest line C = M. They represent the 3-D is a typical curve of this segment character-
sediments having the best sorting, at least in ized by a tail on the coarse side, formed by the
their coarser half. grains that are rolled.
The cumulative curves of the grain size distri- Segment PO reflects an increase in the amount
bution of the samples plotted in segment QR of rolled material. This segment lacking in the
generally end bluntly on the coarse side, sharply Mississippi pattern was found in a number of
defining the value of C. These curves have all other patterns of tractive currents (see for in-
fairly similar shapes so that for arty value of M, stance Fig. 4-F). It is particularly well developed
C is defined within narrow limits. in patterns of sediments of the Adige river which
The graded suspension deposits of tractive will be discussed later.
currents have certain characteristics in common Segment NO of the pattern represents fairly
with turbidity current deposits. Fig. 2 is a typi- well sorted sediments, almost entirely trans-
cal CM pattern of turbidites. As segment QR of ported by rolling as the pebble banks of certain
the tractive pattern, it parallels line C = M. rivers.
The cumulative curves of turbidites are also Why is the passage from transportation as a
similar to those of segment QR. Fig. 3 shows graded suspension to transportation by rolling
three curves of the Marnoso-arenacea Forma- represented by the bend OPQ of the tractive
tion. These curves end bluntly on the coarse side, pattern rather than by a segment OQ parallel to
defining sharply the value of C. As a conse- line C = M ? The answer is that sediments that
quence, the patterns of turbidites show little can be transported as graded suspensions and
dispersion. those easiest to roll are separated by a gap in
The similarity in patterns and cumulative grain sizes.
curves between turbidites and the tractive de- An experiment by HjulstrSm (1939) showed
posits of segment QR suggests that these are that in water moving at a rate of 50 centimeters
characteristics of sediments transported as a a second, grains five millimeters in diameter
 G R A I N S I Z E R E P R E S E N T A T I O N B Y CM PA T T E R N S 833

ADRRIATICSEA OFIONAFORMATION MIDDLEPLIOCENE GALLUP SANDSTONE,

Semments of 2 S r r t , s o b e t h [ T u r b~d~le ) SS~LVO mlS]r, N M

" ' r[111 . . . .

! - ~--- '~ ~ ~TT~ ,. ..... :::-?!f

]l.[i.
IILI
II II Ivl=
i i i ii# ®
nL
,2.~1
,o'!2 / , L l_l.l].ll
I'~'~'i I I II
®

- . . . .

~O~ROW SAND,

TEXAS PA N H AI~ D L E

REPETIO FFORMATtON
(Turbldae on ant~¢Ine )
LOWER
BUSSEXO
F'i_IOC
ENE, i

i i !!!!!!I I~
• l. ° I I I IIIIII - &L ~A LEVEL

$1 S E A BOTTOM

! . ?iiii: o ...... mOLLt~ ~tO,~t~

S L ~ - -
!o.!: Vo .......

.....

. . . .
~n~OEO

LUTITe
SUSP[NSbON ~POStr

U N ~ O R . SUSPE.S,ON ~ p o s l x

M m~c~nt
5L . . . . . LOW ~M~ABIKI~Y A~ER
CONSO~iOATiON

- ~... il2 -- I~PER~OUS K F T E ~ CONSOLIDATION

FIG. 4.--Relationships between transport and shape of deposits illustrated by the CM

pattern and by a schematical cross-section of the deposit.

rolled twice as fast as those of one millimeter. As Adriatic sea. Near Bolzano, where the slope of
Hjulstr6m's curves show, the speed of the cur- the river is 6 : 1000, the Adige receives the turbu-
rent required to transport grains larger than five lent waters of its tributary, the Isareo. Between
millimeters increases with the diameter. There- that point and the sea, tributaries contribute
fore it can be assumed that under given condi- little to the Adige load as they are small and as
tions, grains of an optimum diameter are easier most of them are dammed. In the Po Plain the
to roll than either smaller or larger grains. It is slope of the river is only 0.2 : 1000.
suggested that the value Cr of C for segment O P Bed sediments of the Adige were sampled at
of the tractive pattern corresponds to this opti- stations evenly spaced between the Isarco con-
mum size (see Fig. 1). fluence and the sea, and for each station a CM
The bend OPQ of the pattern means that the pattern was constructed.
grains having diameters intermediate between The value of Cs was found to vary from 2.5
Cs and Cr probably are not abundant in the millimeters at the Isarco's confluence to approxi-
deposit. Assuming that Cr is the easiest size to mately 500 microns in the Po Plain. In the Al-
roll, these intermediate sizes are those that are pine Valley, segment O P of the CM pattern is
too large to be transported as a graded suspen- well developed and for each station is situated at
sion and too small to be rolled easily. The gap in approximately the same place of the diagram,
sizes would indicate that rolling and transporta- the value of Cr being approximately 25 mm.
tion as a graded suspension are independent Cumulative curves representing the average
mechanisms of transport. This was well illus- sample of each station show well the range of
trated by an investigation of the Adige river sizes missing at each station. This gap in sizes
sediments in Northern Italy (Francini and corresponds fairly well to the interval between
Storer, 1961). Cs and Cr and increases as Cs decreases. This
The course of the Adige is 300 kilometers gap marks well the change from transport con-
between the city of Bolzano in the Alps and the trolled by turbulence to transport in which the
834 R. PA SSEGA

main motive force, due to the drag of the water, sediments transported as a graded suspension.
is horizontal. This parameter will be called index of maximum
What is the ultimate condition of grains of sorting, Im, because it measures the part of the
these intermediate sizes after the decrease in deposit having the best sorting. A parallel to line
turbulence causes them to settle? They probably C = M is drawn so that it divides the graded
form a part of the river bed at certain points and suspension points into two equal numbers. The
gradually are abraded by the rolling load until distance between this llne and line C = M, meas-
reduced to such a size that can again be trans- ured on the M axis and expressed in phi units, is
ported in suspension. Im. Im measured on all CM diagrams available
In some patterns certain points of part NOP ranged from less than one for certain tractive
are dispersed. Values of C are then larger than and turbidity current deposits, to six for a mud
normal for the value of M, never smaller. A few flow.
points of segment PQ may also be dispersed, the A study of turbidites and associated tractive
values of M being smaller but never larger than current deposits disclosed a relationship between
normal for the value of C. This latter dispersion Cs as defined on Figs. 1 and 2 and Im. Fig. 5-A is
is probably due to rolling of single particles on a an Im-Cs diagram of tractive current deposits,
finer deposit previously formed. In sampling, it very probably wave deposits, while Fig. 5-B is
may not be possible to make a distinction be- the diagram of associated turbidites. It can be
tween the early deposit and the few grains rolled seen that the relationship between Im and Cs is
later which alter the value of C. Fortunately almost identical for turbidites and for wave
dispersion is not common and is useful in indicat- deposits.
ing that some grains are rolled. All sediments of a graded suspension, fine as
The blunt end of cumulative curves and the well as coarse, being represented by points situ-
lack of dispersion indicate that the deposits of ated approximately at the same distance of llne
graded suspensions do not contain rolled par- C = M , have the same sorting in terms of Im.
ticles. How can it be explained that turbidity Fig. 5 shows that this sorting is a function of Cs,
currents, sufficiently turbulent to transport in that is of the maximum turbulence when settling
suspension particles that may be larger than one begins. Therefore settling of the fine part of a
millimeter, do not keep rolling part of their load graded suspension is controlled by conditions
while the suspension settles? The probable expla- under which the coarsest particles settle. This
nation is that sedimentation is so rapid that seems to exclude particle by particle settling that
particles are buried almost as soon as they settle would be controlled by a gradual decrease in
on the bottom. The following hypothetical depo- velocity and turbulence of the current and imply
sitional mechanism is suggested. that sedimentation of the fine fraction of the
When a turbidity current gradually loses suspension should be at least in part independent
speed, its turbulence decreases. Before settling, of the conditions under which the coarse part
the suspension moves toward the bottom of the settles.
current. The density of the suspension as well as The analyses of the sediments suspended in
the density stratification increase. According to a the Mississippi at Mayersville give good informa-
known property of fluids the increased density tion about the meaning of segment RS of the
stratification accelerates the decrease in turbu- tractive CM pattern. The maximum grain size of
lence. Sedimentation begins and further in- this segment, Cu, shown on Fig. 1, is the same as
creases the density stratification. This self- the maximum grain size in the uniform suspen-
accelerating process produces a mass sedimenta- sion. This indicates that the sediment of the
tion of the suspension which, because of its deposit, too fine to contain either sediments
momentum, becomes a sand flowing on the bot- transported as a graded suspension or rolled
tom until stopped by the drag. Graded bedding sediments, was transported as a uniform suspen-
probably reflects not a gradual sedimentation sion.
controlled by the gradually decreasing current's Cu is an indicator of the maximum turbulence
velocity but the grain size distribution in the in the current above the zone disturbed by bot-
suspension when sedimentation begins. tom friction, provided all sizes that can be held
It is probable that graded suspensions of trac- in uniform suspension are available. If the CM
tive currents also settle as a whole if their con- diagram shows that some sediments are trans-
centration is sufficient to cause an appreciable ported as a graded suspension, it can be assumed
density stratification. that the uniform suspension will transport the
The assumption of mass settling of a concen- maximum size it is able to keep in suspension.
trated graded suspension is supported by the Near-shore Adriatic sediments, entirely trans-
sorting characteristics of the deposit. A new ported as a uniform suspension (Passega, Bor-
parameter is proposed to measure the sorting of ghetti and Florio, 1962) show as will be men-
 G R A I N S I Z E R E P R E S E N ] ' A T I O N B Y CM PA T T E R N S 835

® ®
LS

44. -

.A :

i !
QQ
2
I

O,S

0
100 ~ 4C0 EO0 ~ tO00 0
C 8 ( whcr.,zm ) C s ( m~rord, )
L m

FIG. 5.--Im-Cs diagram, Middle Miocene, Northera Apennines. A-wave deposits, B-turbidites.

tioned in the next section, that certain sediments uniform suspension, but do not seem to transport
are too fine to be an index of the turbulence of these sediments a long distance. Apparently if
the water. the uniform suspension did not flow on bottom,
The Adriatic sediments just mentioned give the sediments would settle through the under-
some information about the composition and lying quiet water. This characteristic of uniform
settling process of a uniform suspension. Fig. 4-A suspension transport explains the fact that in the
is the CM pattern of a deposit sampled along the Adriatic and in the Thyrrhenian seas, deposits of
25 m isobath. The deposit consists of a mixture uniform suspensions are formed at present only
in various proportions of a well sorted, very fine near shore. Farther off-shore and in deep basins,
grained sand that would be represented on the a few turbidites excepted, the only sediment
pattern by a point near line C = M and of a sedi- known to be deposited at present is lutite.
ment finer than 31 microns, the lutite. In all Uniform suspensions are not all deposited in
samples of the deposit very fine sand and lutite relatively shallow water. The southern flank of
compositions are constant. It is probable that the Northern Apennines trough in which the
when turbulence decreases, a large part of the Marnoso-arenacea turbidites were deposited
well sorted sand of the uniform suspension settles consists of shales with occasional streaks of very
first and is followed by a sediment containing fine sand (Vezzani, Passega, 1963).
decreasing amounts of sand and increasing The sand is represented by a uniform suspen-
amounts of lutite. As sedimentation proceeds, sion pattern. A paleogeographic reconstruction
the median M of the sediment decreases, but the of the basin shows that the turbidity currents
value of C approximately remains that of the which deposited the Marnoso-arenacea Forma-
clean sand. The CM pattern therefore has a tion flowed down along these argillaceous slopes.
shape parallel to the M axis which is a character- Foraminifera paleoecology confirms that the
istic of uniform suspension deposits. slopes were in deep water (M. A. Chierici, per-
Uniform suspensions, although not thoroughly sonal communication). It seems that these slopes
studied yet, seem to be bottom suspensions. were essentially an area of non deposition of
Currents that do not flow along the bottom may turbidites in which however, the currents were
be sufficiently turbulent to support sediments in losing a part of their finest load. These fine sedi-
836 R. PASSEGA

ments probably were held in suspension in the tive and turbidity currents. The corresponding
eddies, formed in the wake of the turbidity cur- deposits are schematically represented. The
rents by the surrounding water, and settled as tractive patterns consist of parts of the complete
thin layers on the mud of the slope. pattern of Fig. 1, showing that these deposits are
Shales form CM patterns such as part T of the product of some of the different mechanisms
Fig. 1. Sediments that form shales probably are of transportation, probably according to the size
transported as a pelagic suspension, floating in of the sediments available in the area.
water not necessarily turbulent near bottom. In some areas the roiling part of the pattern, in
Most of these sediments, finer than 31 microns, other the uniform suspension part are missing.
can be called lutite. The reason for selecting Graded suspension may be missing in patterns
31 microns, which is the dividing size between that show rolling and uniform suspension. Sedi-
medium and coarse silt, is that particles smaller ments formed by pebbles and lutite with no
than 31 microns are so easily kept in suspension graded suspension and almost no uniform sus-
that they can travel long distances as a pelagic pension (not represented on Fig. 4) are known in
sediment. As an example, samples of the Tyr- Northern Italy. Pebbles are either isolated in the
rhenian sea floor taken between Italy and Cor- shale or grouped in banks.
sica at depths ranging from 65 to 1400 meters Patterns intermediate between those of trac-
were, with the exception of some shells, all mud tive and turbidity currents will be discussed in
having the same grain size distribution with section V.
medians of approximately 3 microns and values
of C ranging from 20 to 33 microns. III. PALEOBATHYMETRY
In the area of the Adriatic sea, mentioned
previously, limited by the coast lirte between the Under certain conditions, grain size analyses
towns of Pescara and Ortona, and the 50 m. may be used to estimate the depth of the sea in
isobath it was found that while the percentage of which the sediments were deposited.
lutite in the bottom sediment varies, the com- Storm waves acting on an open sea floor less
position of the lutite and particularly its content than 100 meters deep probably are the predomi-
in clay (particles finer than 4 microns) is con- nant agent of deposition of ancient sands (Pas-
stant. Below 35 meters, the sediment has a sega 1962). Waves are the only agent capable of
maximum grain size of 50 to 65 microns and is forming the sand blankets that during certain
practically all lutite (Passega, Borghetti and periods covered more than 100,000 square kilom-
Florio, 1962). eters of the sea floor.
As only a few examples were studied, it is not While waves of any size may deposit sedi-
known whether all lutite having a constant com- ments, these are reworked during the most vio.
position in a given area was transported in a tent storms. Therefore the sediments that are
uniform suspension or as a pelagic suspension. preserved in the geologic record are those of these
The lutite transported in a graded suspension exceptional storms. Stetson (1938) showed that
was studied in a few turbidites. Preliminary on the Continental Shelf, east of North America,
results indicate that the proportion of lutite in at a depth of 60 to 70 meters, Pleistocene sands
the coarsest part of a graded bed is almost con- are reworked by waves and kept clean of mud.
stant but increases slightly upward with the The water particles of a wave describe an
decrease in median grain size. In the fine upper orbit which is almost circular near surface. If the
part of the graded bed, the proportion of lutite wave "feels" bottom, near bottom the orbit
rapidly increases as the median decreases. The flattens and becomes a segment parallel to the
finest sediments represented on typical turbidite bottom. The alternate movement of water along
patterns, parallel to line C = M, have a value of this segment is a tractive current, and its sedi-
C of approximately 30 microns and are therefore ments form a tractive CM pattern.
entirely formed by lutite. The median of this As the action of waves decreases with increas-
lutite is larger than 20 microns, which means ing depth, the value of Cs, which is a measure of
that more than half of the lutite ranges in size the maximum turbulence, should be an index of
between 20 and 30 microns. In comparison, as the sea depth. The deepest part of the basin
was mentioned, the pelagic lutite of the Thyr- should be indicated by the smallest values of Cs.
rhenian sea has a median of approximately three This assumption was tested by an ecological
microns. study of Pliocene and Miocene foraminifera of
The percentage of lutite in a deposit can be Northern Italy still living in the Adriatic sea
studied by constructing LM sample point pat- (Chierici, Busi and Cita, 1962). Depth intervals
terns where M is the median of the sample and L were defined, applied to the Pliocene and Mio-
the percentage of lutite. cene sediments and compared with the values of
Fig. 4 shows a number of CM patterns of trac- Cs on the diagram of Fig. 6.
 GRAIN SIZE REPRESENTATION B Y CM P A T T E R N S 837

DIAGRAM SEA DEPTH -'C~ values of C for each pattern, of 64 and 62 mi-
crons.
i The other patterns, of which one is shown on
Fig. 4-A, are p a t t e r n s of sediments entirely
~oe,,~ ,i t r a n s p o r t e d as a uniform suspension. S e d i m e n t s
of graded suspensions are absent.
aoo • Cu, the m a x i m u m value of C of each p a t t e r n ,
regularly decreases as the d e p t h increases from
•oo '\'-": ¢'x i 10 to 30 meters. However, the relationship be-
tween Cu and sea d e p t h is not the same as the
r e l a t i o n s h i p shown by Fig. 6. For a given depth,
Cu is m u c h lower for the Adriatic s e d i m e n t t h a n
i \ \ ,,. on Fig. 6. This is due to the grain size of t h e
sediments supplied b y streams to t h e Adriatic,
tap i i'" which is much finer t h a n the m a x i m u m size t h a t
can be t r a n s p o r t e d as a uniform suspension. As
~'PTH IN METERS was mentioned, u n d e r these conditions Cu can
FIG. 6. Cs-sea depth diagram. Bars correspond to not be used as a b a t h y m e t r i c indicator.
the formations sampled and indicate the depth inter- The grain size distribution in the Adriatic near
vals defined by ecology of Foraminifera. shore water is the result of selective t r a n s p o r t a -
tion and will be discussed in the next section.
Fig. 6 shows an orderly a r r a n g e m e n t of the In t o d a y ' s oceans, where grain size of the
ecological bars t h a t confirms the decrease of Cs sediments is known, these seem to be t r a n s p o r t e d
with an increasing sea depth. By drawing a line as a uniform suspension. In some of these areas,
t h a t intersects all bars, a relationship between particularly near shore, a part of the s e d i m e n t is
Cs and sea d e p t h is obtained t h a t gives the ap- rolled, b u t graded suspension generally is lack-
proximate d e p t h of the Pliocene a n d Miocene ing. Although the total n u m b e r of analyses of
seas of N o r t h e r n Italy corresponding to a n y recent marine sediments usable to construct C M
value of Cs. p a t t e r n s is still too small to prove it, it seems
As the m a x i m u m size of s t o r m waves was n o t t h a t uniform suspension probably is the pre-
the same in all seas, the above relationship does d o m i n a n t mechanism of sand t r a n s p o r t in pres-
not indicate the absolute d e p t h of o t h e r seas. ent seas. It should be noted t h a t the position of
However it still gives relative values t h a t p e r m i t the m u d line a t a d e p t h of between 30 a n d 40
a reconstruction of the s u b m a r i n e topography. meters in the Adriatic is the result of selective
CM p a t t e r n s of wave deposits seem to show a t r a n s p o r t a t i o n and is not an indication t h a t
relationship between Cu and Cs, the m a x i m u m waves c a n n o t be active below 30 meters. As was
grain size of the uniform a n d graded suspensions. mentioned, Pleistocene sands are reworked on
This relationship is indicated on Fig. 6 by a line the c o n t i n e n t a l shelves by waves a t a much
Cu-sea d e p t h t h a t permits the estimation of t h e greater d e p t h (see also Passega 1962).
sea d e p t h when Cu is known b u t Cs is not. How- In some areas of present seas, sands are suffi-
ever, Cu represents the m a x i m u m turbulence of a ciently coarse to form a more complete p a t t e r n .
uniform suspension only if all sizes t h a t can be Fig. 7, for instance, is t h e p a t t e r n of calcareous
t r a n s p o r t e d in uniform suspension are available. sands near the shore of Florida, discussed by
It can be assumed t h a t these sizes are available if Vause (1959). On Fig. 7, the value of Cs is ap-
the C M p a t t e r n indicates t h a t a part of t h e proximately 1.8 millimeter, which is not far from
sediment is sufficiently coarse to be t r a n s p o r t e d the value indicated b y the c h a r t of Fig. 6. A
as a graded suspension. comparable result was obtained by analyses of
Do sediments of the present oceans conform to calcareous sands from the Bahmas.
the assumed relationship between Cs a n d sea In s u m m a r y , on a open sea floor, sediments
d e p t h ? Sediments of the Adriatic Sea floor men- t r a n s p o r t is to a large e x t e n t controlled by the
tioned before were sampled between the cities of composition of the supply. W h e r e the supply is
Pescara and Ortona, a n d between the 10 and 50 sufficiently coarse to be t r a n s p o r t e d as a graded
meters isobaths. A total of approximately 250 suspension the C M p a t t e r n s are a means of
samples were subdivided according to t h e sea estimating the sea depth.
d e p t h and seven CM p a t t e r n s were drawn for t h e In present seas, graded suspensions seem to be
isobaths of 10, 15, 20, 25, 30, 40 a n d 50 meters rare, p r o b a b l y because rivers, as shown by the
(Passega, Borghetti and Florio, 1962). sediments of the Adige, far less t u r b u l e n t t h a n
T h e p a t t e r n s of 40, a n d 50 meters are typical storm waves supply the seas with fine sediments
shale p a t t e r n s showing respectively m a x i m u m t h a t are t r a n s p o r t e d by marine agents as uni-
838 R. P A S S E G A

N ~ UT~OLOGY ~ET.OORAP.~ WWd~ ra~L~r~ Aoc

-~. .(.i'(.~1 ~::i:::::::::..

/-
s

I I

°"° ":~ .:';"-'l ;!!::iN'

:':':':':':':':

Fm 7--CM pattern of calcareous sands deposited

at depths between 8 and 18 meters near the coast of
Florida.

form suspensions, or with rolled grains. In some

ancient seas conditions were different. In the
Sahara, transgressive Paleozoic seas found on the
shield an abundant supply of sediments formed !~:!:i~'2:?i::':?.::
by weathering of granite and spread these sedi- :-- :-:-_--:-::_-:-:-:-:-
ments over extensive areas mostly as graded
suspensions.
The Cs-sea depth relationship makes it pos-
sible to log the variations of the depositional sea
depth in a stratigraphic section. Fig. 8 shows a
paleobathymetric and petrographic log of the ..... : y ::iiiiiii
iii:
Paleozoic section in the Koufra basin of Libya. .... ~ ! i:i:i:!:i:i:i:
At a given point, the sea depth remained con- :~ ;.,' ,. I ' . ' . -'..:.:.'.:~

stant during long periods, probably because of an ~i.= i.; i:i:::::::::i:::!

equilibrium between supply of sediments and
.... .i~ :;~,:'i
, .::'::':':':':'::
: : :.:.:.:.:~
subsidence. The sharp depth change that marks
the contact between Acacus and Tadrart forma-
tions can be correlated in a large part of Libya.
As the constancy of the sea depth at any point L E ~ E N D
during fairly long periods is common, it is pos-
sible to construct paleobathymetric maps repre-
senting the submarine topography during these
periods.
Fig. 9 is a paleobathymetric map of the Mio- PALEOBATHYMETRIC AND PETROGRAPHIC LOG,
cene sea in the Northern Apennines (Vezzani and DOHONE AREA, KUFRA BASIN, LIBYA.
Passega 1963) constructed by using surface and
well data. Other paleobathymetric maps were
constructed ; particularly, a map of the Oligocene
sea in a part of the Rhine Graben (Gottis and
Passega, 1960), a map of the Lower Devonian of FIG. 8.--Paleobathymetric logs of Paleozoic
the Algerian Sahara (Burollet, Byramjee, Men- sections, Koufra basin, Libya.
 G R A I N S I Z E R E P R E S E N T A T I O N B Y CM PA T T E R N S 839

--~Ip.3zo~rE~A~G~O~E
MIOCENE PALEOBATHYMETRIC M A P
~ ~ :..2:;T.T': . . d NORTHERN APENNINES

PARMA

REOGIO E ..f4om / "40 METERS PALIrOISOBATH

~-~m~,

BOLOGNA

Rm*e, ......

FIG. 9.--Miocene Paleobathymetric map, Northern Apennines, Italy.

nig and Passega, 1963), maps of the Miocene and will cause them to be deposited closer to shore
Pliocene in Southeast Iran (Cardello and Passega than the latter.
1962), and a map of the Middle Plioeene on the The decrease in grain size of wave sediments as
Adriatic coastal region of Italy (Rizzini, Cardello the water depth increases tends to decrease the
and Passega, 1962). permeability of the sediments after they become
The interest of paleobathymetry as a geologi- consolidated. This trend, however, is countered
cal parameter is that it is related to some of the to some extent by the relationship between maxi-
principal characteristics of the sediments. The mum sorting and Cs shown by Fig. 5-A. As the
rules that control the general migration of clastic sediments become finer, their sorting improves.
sediments on a sea floor are not yet known, but This explains how certain very fine sediments as
sea depth and composition of the supply seem to those of the Acacus formation in Northwest
be major factors of this migration. Lateral facies Libya, deposited at a depth of approximately 50
changes affecting in some areas sections several meters, can have high permeabilities.
hundred or even several thousand meters thick The presence of source beds of petroleum may
generally can be related to flexures of the sub- also be related to the depositional depth of these
marine topography caused and maintained by beds. Analyses of a number of formations, con-
structural features. sisting of alternating sands and shales, show that
Sediments transported in suspension by waves the formations that include source beds formed
become finer as depth increases, but pebbles or CM patterns having values of Cs lower than 500
coarse sands can be rolled by waves on silt or stiff microns. In shallow open seas, the bottom water
clay at depths probably approaching the wave agitation probably supplies too much oxygen for
base. However, as will be seen later, rolled sedi- the preservation of organic matter. It is possible
ments are more difficult to transport than sus- that as far as petroleum exploration is concerned,
pension sediments and selective transportation these shallow sediments should be considered
840 R. P A S S E G A

almost as unfavorable source rocks as continen- Rolled sediments are next in order of difficulty
tal beds. of transport. They are present only in the Alpine
At this point it may he noted that the conclu- part of the Adige's course.
sions of this section contrast with the opinion of Sediments transported as a graded suspension
oceanographers who draw their experience from travel in the Adige farther than the rolled
present seas. Oceanographers generally consider sediments.
that, turbidites excepted, an ocean transports In the Po Valley part of the Adige's course,
sand not more than a few miles from shore and where the slope of the river bed is 0.2:1000 as
not deeper than 30 meters. They suggest that far as the mouth of the river, all deposits are uni-
ancient widespread sands are distributed by wide form suspension sediments.
movements of the shoreline across the area. The Adriatic bottom samples that were dis-
This theory fails to explain sections such as the cussed show that lutite, transported as a pelagic
several hundred meters thick Devonian section suspension is the clastic sediment easiest to
of the Sahara consisting of alternating sand- transport.
stones and shales, containing bryozoa and other Sediments transported by waves in the sea,
marine organisms, and extending over enormous probably are only in part deposited in the same
areas. The texture of the sands in terms of CM general order as the sediments of the Adige. The
patterns is uniform in an)' locality but gradually submarine topography is an important factor of
varies across the area (Burollet el al., 1963). transport that is not yet well understood. Cer-
During a transgression or a regression, for any tain mechanisms of transport apparently move
position of the shore line, according to oceanog- sediments preferentially in zones of high turbu-
raphers' ideas, only a prism of sand a few miles lence, that is on highs. Other mechanisms of
wide and less than 30 meters thick could be transport, probably affected by gravity, move
deposited. Wide lateral movements of the shore sediments down slope into the depressions of the
line could therefore only deposit a veneer of sea floor.
sediments that could not account either for the The properties of the different modes of trans-
sand thickness in the Sahara or for its textural port known at present will be reviewed in an
characteristics. attempt to link them to the geometrical char-
It may also be noted that sand associated with acteristics of the deposits. The examples dis-
a shore line is for the most pat t moved by long- cussed are illustrated by Fig. 4.
shore currents that form sedimentary trends Transportation by uniform suspension was
parallel to the coast. Crossbedding of the Sahara studied, as mentioned, in the Adriatic Sea. A
sands indicates a movement perpendicular to the CM pattern of the deposits is shown on Fig. 4-A.
coast toward the deepest part of the basin. The concentration in suspended matter in a
Differences in the characteristics of the sand suspension 10 to 20 meters high is too low to be
supply ind ueing a different action of gravity seem much affected by gravity and the sediment fol-
to the writer to explain the difference in trans- lows the movement of the water in which it is
port between the Sahara Devonian sea and pres- suspended. Along this part of the Adriatic coast
ent oceans. the general transport of water is due to waves
and long shore currents. Strong currents directed
IV. G E O M E T R Y OF SAND B O D I E S
seaward are not indicated by the distribution of
AND T R A N S P O R T
sediments.
An understanding of the relationships that It generally is recognized that in waves that
may link geomet W of sand bodies and mechan- touch bottom near shore the lowest water has a
ism of transport of the sand may help the search net movement shoreward. This movement is
for stratigraphic traps. With the available data compensated for, higher above bottom, by a net
a first idea about these relationships can be seaward flow. Therefore the coarsest part of the
formed. suspension which tends to stay the longest in the
The variations of the Adige's river sediments bottom water tends to move shoreward, while
along the course of the river make it possible to finer particles will move seaward. This selective
list these sediments in order of difficulty of their action of waves keeps the coarsest part of the
transport. uniform suspension near shore, whereas most of
As was mentioned in another seetion, the most the lutite is deposited in deeper water. In the
difficult sediments to transport have sizes inter- area studied, the mudline is approximately 30
mediate between the sizes that are easily rolled meters deep. This depth, determined by the
and those transported as graded suspension. As selective transport j ust described, is not the wave
the turbulence of the river decreases, the range base.
of these intermediate sizes gradually increases How uniform suspensions moved in ancient
toward the fine sizes. open seas is not known, but it can be assumed
 GRA I N S I Z E R E P R E S E N T A T I O N B Y CM PA T T E R N S 841

that if they were not too dense they followed the Marnoso-arenacea Formation of the Northern
water movement. It is possible that waves con- Appenines showed the gradual change in char-
centrated the coarsest fractions on highs in rela- acteristics from wave to turbidity current de-
tively shallow water, even at considerable dis- posits (Vezzani and Passega, 1963). However it
tances from shore. should be noted that waves are not the only
Sediments transported as uniform suspensions agent that forms turbidity currents. In the Po
may be sand, silt or lutite. After consolidation Valley basin a precise reconstruction of the
according to their grain size, the deposits may Tortonian (Middle Miocene) shore line was
either form reservoirs or permeability barriers. possible as the shore was formed by the front of
A deposit of uniform suspension may form a allochthonous formations in places more than
permeability barrier to sediments deposited by one thousand meters thick. Turbidity currents
other transport mechanisms, but within deposits probably originated right at the shore line, de-
of uniform suspension, lateral changes suf- scended on a steep shaly flank of the basin and
ficiently sharp to form permeability barriers are filled the deepest part of the Tortonian trough
not known and seem unlikely. A uniform sus- with molasse like turbidites (undaturbidites).
pension being high extends over low submarine Paleoecology of Foraminifera (M. A. Chierici,
rises on which the deposit probably tends to be personal communication) indicates that the
coarsest. Under these conditions, no strati- depth of the trough was several hundred meters.
graphic trap should be expected. It is assumed that the currents were originated
Turbidity currents are formed by graded sus- by streams heavily ladden with Miocene and
pensions, but waves may also carry sediments as Oligocene sand which constituted a considerable
a graded suspension. As was mentioned, deposits part of the allochthonous. These streams could
of the two agents are fairly similar. This sup- have been similar to those that, as will be men-
ports the assumption (Duly, 1936; Passega, tioned later, transported as graded suspensions
1962) that turbidity currents may originate as a some of the Swiss Molasse sediments (see also
wave suspension. A difference between the Rizzini, Passega, 1964).
graded suspensions of waves and turbidity cur- Graded suspensions under the action of
rents is that the suspensions supported by wave gravity tend to fill depressions or channels of the
turbulence do not require any net lateral move- sea floor. If the supply of sediment is sufficient,
ment, whereas, the turbulence of the turbidity they form extensive blankets that become thin-
current suspension is generated by bottom fric- ner on highs.
tion that necessitates a high current velocity. Should a rise of the sea floor, formed by mate-
As turbidity currents, graded wave suspensions rial difficult to erode, be higher than the profile
may be sufficiently concentrated to be subjected formed by deposition of the sand of a graded
to gravity. If gravity can overcome the shore- suspension, this high will stay bare of sand and
ward transport of water by shoaling waves, the a pinch-out will result. This pinch-out can be
suspension flows down along the basin's slope. formed at any depth and is not necessarily, as is
The lower Devonian sands that cover a large at times thought, an ancient shore line.
part of the Algerian Sahara are an example of As a graded suspension is transported near
graded suspension deposit probably formed by bottom, it is more sensitive than a uniform sus-
waves. Sandstones interbedded with shlales con- pension to a low relief of the sea floor.
taining bryozoa and other marine organisms Fig. 4-B is a reservoir of the San Roque field,
form a section several hundred meters thick. Venezuela, formed by graded suspensions, prob-
CM patterns of the sands indicate transport as ably turbidites, deposited in an erosion channel.
graded suspensions with some rolling. A paleo- The coarsest material is in the channel and is
bathymetric map was constructed (Burollet fairly permeable, while the sediment deposited
et al., 1963). Abundant crnssbedding shows uni- outside the channel is very fine sand, silt and
directional flow of the transporting currents lutite and is impervious. The lateral facies change
toward the deepest part of the basin. forms a stratigraphic trap.
Numerous examples indicate that transport of Fig. 4-C is a turbidite that forms a reservoir in
ancient sands as graded suspensions was the Los Angeles Basin. It seems that in this
common. basin turbidites generally are more permeable on
If the density of the wave suspension and the anticlines than in synelines. It is tentatively sug-
slope are sufficient, the suspension may gather gested that sorting may be a controlling factor
speed, the part of the suspension formed in of their permeability and that according to the
shallower more turbulent water being coarser relationship of Fig. 5-B the fine turbidites on the
and denser flows under the slower deeper water crest of anticlines are better sorted than the
suspension and gradually the wave suspen- coarse turbidites deposited in the synclines.
sion becomes a turbidity current. Study of the Fig. 4-D is a typical Lower Pliocene blanket
842 R. P A S S E G A

sand of the Pc Valley, which shows transporta- gravels rolled on rigid mud accumulate along
tion mostly as a graded suspension. Pliocene and lines of structural flexure to form stratigraphic
Miocene sediments of the Po Valley are char- traps and gas fields. Graded suspension is absent,
acterized by facies changes on the flanks of uniform suspension almost absent. The closure
structural highs. Some beds are shale on highs, is formed by the lutite.
sand on the flanks. This very probably is due to These few examples show that stratigraphic
the low relief formed by the highs on the sea traps formed by lateral facies changes of clastic
floor. Mud, apparently sufficiently rigid to be re- sediments may be subdivided into a small num-
sistant to eromon even shortly after deposition, ber of types and that these types are related to
was not affected by wave action, while any sand the transport of the sediments. An understand-
deposited on the high was swept down to the ing of the style of sedimentation prevailing in
profile of sand deposition. A number of Po Valley an area may indicate the type of stratigraphic
gas fields owe their closure to these facies changes. trap that may be found.
The field of Busseto illustrated by Fig. 4-D is a
structural nose closed by a permeability barrier.
v. SEDIMENTARY STRUCTURES AND TRANSPORT
Flowing graded wave suspensions commonly
roll sediment. Rolled grains are abundant in the Relationships between sedimentary structures
blanket sands of the Sahara basins, but their and transport cart be studied by means of CM
proportion in the sediment decreases down- patterns. A pattern of the deposit as a whole is
stream. Fig. 4-E is an example of transport by constructed and a sample taken on the structure
rolling and by graded suspension. Similarly to is plotted on the pattern. The segment of the
the sand of Fig. 4-D this sand sharply changes pattern in which the point falls indicates the
up-dip to shale, forming a permeability barrier mechanism of transport. Systematic investiga-
and a stratigraphic trap. tions of sedimentary structures were not made,
As rolled grains cannot move as fast as a swift but a few observations will be reported.
current, they tend to be by-passed. This particu- Crossbedding often seems to correspond to
larly applies to turbidity currents. segment P Q of CM patterns, that is to be
Graded suspensions probably do not every- formed by graded suspension and rolling. The
where smoothly fill submarine depressions. gap in sizes between the grains rolled and those
Should they encounter a bar or a channel of the transported in suspension may be a factor of
marine floor, they may settle rapidly and form crossbedding. South of the Polignac Basin,
lenses having surface slopes up to the angle of Algerian Sahara, more than 80 percent of the
repose of sand. This type of sedimentation that Lower Devonian beds are crossbedded. Farther
seems to form certain stratigraphic traps is not north, away from the source, crossbedding is less
well understood. abundant and cores of a Libyan well were mas-
Grains can be rolled when graded suspension is sive sand represented by segment Q R and two
lacking. On low submarine slopes they do not thin beds crossbedded represented by point Q.
seem to feel the effect of gravity and tend to The decrease in crossbedding away from the
move with the bottom water that generally keeps source probably is due to a decrease in the pro-
them on submarine highs. portion of sediment that was rolled.
Fig. 4-F represents a stratigraphic trap formed Oscillation ripple marks are abundant in near
by the Gallup sand at Bisti, New Mexico. This shore areas of present oceans where transport by
elongated field probably follows a line of flexure. uniform suspension is predominant. However,
Sediments of the field are rolled sand. Sediments data is not yet sufficient to establish a relation-
down-dip are uniform suspension deposits and ship between ripple marks and transport by
up-dip, uniform suspension deposits with some means of CM patterns.
rolled grains. The sand was probably rolled to Massive sand showing no stratification seems
the line of flexure and accumulated at the edge to have been transported in suspension, either
of the deeper water. The uniform suspension graded or uniform.
forms a stratigraphic closure. The accumulated Graded sands are a typical deposit of graded
rolled sand probably causes by its roughness suf- suspensions. Graded bedding is not an exclusive
ficient turbulence to keep the sand clean from feature of turbidites. Bersier (1958) reports that
suspension sediments. near Lausanne almost every bed of Molasse for-
Fig. 4-G also represents a stratigraphic trap mation, probably deposited by rivers, is graded.
formed at the edge of a flexure by rolled material, CM patterns constructed for some of these beds
uniform suspension and lutite forming the show transport as graded suspensions.
closure. The sedimentary structures so typical of the
Finally, in the Po Valley, Lower Pliocene soles of turbidites can be explained by the as-
 G R A I N S I Z E R E P R E S E N T A T I O N B Y CM P A T T E R N S 843
sumption, discussed before, of mass sedimenta- a cause of convolute bedding. Silt and very fine
tion of a swiftly flowing graded suspension and sand, because of their high surface-to-mass ratio,
formation of a mass of flowing sand. The sand may be metastable when just deposited so that
flows at first with the speed of the turbidity cur- loading or vibrations might induce spontaneous
rent, but its speed is rapidly checked by drag on liquefaction (Terzaghi, 1960, Williams, 1960 and
the underlying beds. It was mentioned that sort- Dott, 1963).
ing of the sands is a function of the current At this point it may be mentioned that the
velocity when sedimentation starts, as shown by assumption of flute casts being carved by mass
Fig. 5-B. sedimentation changes their paleogeographic
Load casts, erosion, deformation by flowage of meaning. Flute casts would be a depositional
shale underlying the turbidites are self explana- feature indicating only the direction of the slope
tory. Flute casts are probably abraded by t h e at the point of deposition. A trough could be fed
flowing sand and almost immediately filled so with sediments laterally and show only flute casts
that their finely sculptured features are pre- parallel to its axis.
served. Angular fragments of intraformational Fig. 10 illustrates a peculiar sedimentary de-
shale are, in all probability, the result of under- posit of Civago, Northern Apennines (see loca-
cutting of thin shale beds more resistant to ero- tion on Fig. 9). This deposit, common in the area
sion than the underlying sand. shown on Fig. 9, has characteristics intermediate
Cross- and convolute laminations and slump between those of wave and those of turbidity
structures are common in the upper part of turbi- current deposits. For this reason, Rizzini and
bidites. These structures may be attributed to Passega (1964), suggested that it be called
the action of the water, freed from most of its "undaturbidite." Similar sediments had previ-
suspension but still carrying silt. Williams (1960) ously been called fluxoturbidites (Dzulynski,
suggests that "spontaneous liquefaction" may be Ksiazkiewicz and Kuenen, 1959).

aooo
I
I f" \
4/ I
2ooo
/
I ///'I ' /
I.!/ : /

.14" •
i el -
F/
a
t •
/ "1 //

L LI
t 8 e]

--- C$

v i/ ........
//

/
L
i

Mlmitroni| ~0 30

FIG. 10.--Undaturbidites, Civago, Northern Apennines.

844 R. PA SSEGA

Undaturbidites are either sandstones inter- It is interesting to note that in the area of
bedded with shale, as those of Fig. 10, similar to Fig. 9 the wave deposits that laterally grade into
the normal Marnoso-arenacea turbidites but undaturbidites were mostly transported as uni-
more lenticular and showing fewer flute casts, or form suspensions or by rolling, not as graded
thick massive lenticular sands, in places filling suspensions. It seems probable that these are
scour channels, with few shale beds, fairly similar residual deposits left by the turbidity currents.
to wave deposits, but showing some flute casts Having lost their graded suspension fraction, the
and some convolute bedding. deposits formed by sediments not sensitive to
The location of undaturbidites in the area gravity became "stabilized" and remained on
represented on Fig. 9 is intermediate between the submarine highs.
location of wave and of turbidity current de- Fig. 11 illustrates a Pennsylvaniau deposit of
posits. Gradual lateral changes between unda- the Northeast Oklahoma platform. The Torpedo
turbidites and wave deposits can be seen. sandstone forms blanket and channel sands. The
Patterns of undaturbidites such as the pattern CM pattern shows that in the area where this
of Fig. 10 are intermediate between tractive and sandstone was sampled it was transported as a
turbidity current patterns, and commonly are graded suspension. As indicated by the photo-
formed by several segments, all parallel to line graphs, the deposit was the result of mass sedi-
C = M, but situated at different distances from mentation and the lower part was deformed by
this line. Some of these patterns show a consider- sliding while the deposit was still in process of
able scatter on their coarse side. formation. Photo A shows lower beds bent in the
It is suggested that undaturbidites are the de- form of balls about one meter in diameter and
posits of certain turbidity currents which still upper beds undisturbed. Photo B shows con-
reflect the grain size distribution of the tractive torted beds, irregularly in contact with upper
current sediments from which they originated. undisturbed beds. The sliding is well illustrated
The suspension has not yet been reorganized so by Photo C which shows the sole of the beds of
as to deposit normal turbidites such as those of Photo B. It may be mentioned that in this area,
Fig. 2. While the maximum grain size tends to flute casts and grove casts, although rare, are
be proportional to the median, the median is still, known.
as an average, the median of the tractive current Similar sedimentary structures of Pennsyl-
sediments. From this peculiar grain size distri- vanian sandstones, near Tulsa, Oklahoma, are
bution certain characteristics of the current can attributed by Dott (1963) to plastic flow. Plastic
be inferred. flow is a logical consequence of mass sedimenta-
As was mentioned, the density stratification of tion of a swiftly flowing graded suspension.
a current tends to decrease its turbulence. For The sands of Fig. 11, as those of Fig. 10, can
equal turbulence, a greater density stratification be considered as deposits intermediate between
requires a greater bottom friction and therefore, typical wave and normal turbidity current
a greater current velocity. It seems logical to deposits.
assume that sediments represented by segments In summary, CM patterns give fairly precise
situated at different distances from line C = M information on the genesis and meaning of sedi-
were transported as suspensions having different mentary structures.
densities. The interface between these suspen-
VI. S A M P L I N G A N D ANALYSES
sions should correspond to a marked density
stratification. The currents therefore should be Sampling and analyses are very likely the
more violent and less stable than normal main cause of dispersion of the points in a CM
turbidity currents. pattern. Sampling and analyses techniques will
This conclusion is supported by observation be discussed briefly.
on undaturbidites environment. As mentioned Samples must be as homogeneous as possible,
by Rizzini and Passega (1964), undaturbidites and must be analyzed individually. A mixed
commonly were formed during periods of orog- sample formed by a shale that contains paper
eny, when sand was supplied in large amounts to thin streaks of sand would have a value of C
the steep slopes of the basin. determined by the sand and the median of the
Undaturbidites are more likely than normal shale. As can be seen on the pattern of Fig. 1,
turbidites to contain pebbles. The smallest the representative point of the mixed sample
pebbles are very likely transported in suspension would fall far outside the tractive pattern. A
by the dense and violent current, the largest are sample should represent a sedimentation unit
probably rolled. Gravity may be a factor in and be the product of a period of homogeneous
pebbles transportation, as it may help successive sedimentation. The shale and sand mentioned
currents in rolling the pebbles on steep slopes. should be sampled separately.
 GRA I N S I Z E REPRESENTA TION B Y CM PA TTERNS 845

D4

10 M t m i © r o ~ t l 20

FIG. 11.--Torpedo Sandstone, Bartlesville, Oklahoma.

A CM pattern generally is formed by 20 to 30 The grain size of sandstones cemented with

samples of a deposit representing all textures silica and of calcarenites is measured on thin
from the finest to the coarsest. sections cut along the bedding plane.
Surface sections are ideal for sampling as the Thin section measurements give CM points
sediments can be seen and selected. Each sample that are different from those of the sieves. For-
generally represents a bed three or four centi- mulas for translating thin section into sieve
meters thick and a CM pattern represents a measurements were used without success, but
section a few meters thick. tests made by analyzing a sandstone with cal-
In a stratigraphic section of several hundred careous cement by sieve and on thin sections
meters, CM patterns are spaced according to the showed that without any transformation of the
variations of the lithologic characteristics and of thin section measurements, these give the same
the shape of the patterns. value of Cs as the sieves.
Well cores give good patterns, but all textures The thin section technique consists of measur-
of the strata rarely are cored. Drill samples can ing the minimum diameter of 300 grains,
be used only if each individual cutting can be selected at random, with a point-counter in art
treated as a sample. This is possible when the area of uniform texture. Grains finer than 31
cutting is a sandstone or a limestone fragment microns are not measured. Cu therefore rarely
that can be analyzed on a thin section. can be determined.
Samples that can be reduced to grain size are CM patterns, constructed by measuring on
sieved after treatment by hydrogen peroxide and thin sections the maximum grain size and esti-
hydrochloric acid, the openings of the finest mating the median, are too scattered to be
sieve being of 31 microns. When the median is usable.
smaller than 31 microns, the fine fraction is dis-
persed and analyzed by pipette. vii. CONCLUSION
Analyses by sedimentation tubes are not as A elastic deposit is formed by fractions trans-
precise as analyses by sieves for the construction ported and deposited by several transport
of CM patterns mechanisms largely independent of each other.
846 R. P A S S E G A

It is not easy to recognize the sediment deposited extent the characteristics of the deposits when
by each mechanism in the grain size distribution the sedimentation style, that is, the sizes avail-
of the sediment as a whole. However, a consider- able and the transport mechanisms, is known.
able difference in the action of these transports, This helps the search for stratigraphic traps.
which causes the separation of their sediments Waves probably are the predominant deposi-
by gaps of missing sizes makes it possible to tional agent of ancient marine sands. The deter-
analyze the transport when suitable grain size mination in wave deposits of the maximum grain
parameters are used. size transported in suspension, permits the esti-
Sediments are transported by rolling or in mation of the depth of the depositional sea.
suspension. Suspensions are subdivided into Paleobathymetric maps and logs can be con-
graded bottom, uniform bottom, and pelagic structed.
suspensions. Paleobathymetry is a particularly important
Transport of a sediment depends upon the geological parameter as it is related to such
sizes available, and characteristics of the de- characteristics of the sediments as facies and
posits depend upon the transport. The role of permeability. Source rocks of petroleum seem
gravity varies greatly with the mechanism of also to be deposited preferentially in water
transport and causes some deposits to be formed deeper than 40 meters. A sea represented by a
in depressions of the submarine topography few isobaths is a rather simple picture that is of
while others are formed on highs. considerable help as a starting p o i n t / o r the in-
In present seas, suspensions are predominantly vestigation of complex geological regions.
uniform and pelagic in nature, and sand gener- In summary, grain size analysis is a practical
ally is deposited near shore. In many ancient geological tool that can supply relatively quickly
seas blanket sands transported as graded suspen- a number of interrelated geologic parameters.
sions were deposited in the low areas of the sea These parameters are of considerable help in
floor, recognizing the broad lines of the regional geol-
The close relationships between transport and ogy as well as in the detailed study of local sedi-
deposits makes it possible to foresee to a certain mentary trends.

REFERENCES
BERSIER, ARNOLD, 1958, SEquences D~tritiques et Divagations FIuviales: Eclogae Geologieae Helvefiae,
v. (I.e.) 51, No. 3, p. 854-893.
BUROLLET,P. F., ByRAMJEE, R. S., MENmO, J. J., AND PASSEOA,R., 1963, R61e des Etudes S&llmentologlques
dans la Conduite de I'Exploration: Presented to the Sixth Petroleum World Congress, Frankfort, June,
1963, Sec. 1, Paper 13, France, p. 1-10.
CAPa~ELLO, R., AND PASSEGA,R., 1962, Miocene and Pllocene Lateral Facies Changes, Makran Iran: Unpub-
lished Report AGIP.
CmERIeI, M. A., BUSl, M. T., AND CITA, M. B., 1962, Contribution/l une Etude ~cologique des Foraminlf~res
dans la Mer Adriatlque: Revue de Micropal., v. 5, No. 2, p. 123-142.
DALY, R. A., 1936, Origin of Submarine Canyons: Am. Jour. Sci., v. 31, p. 401-420.
DOTT, R. H., JR., 1963, Dynamics of Subaqueous Gravity Depositlonal Processes: Amer. Assoc. Petroleum
Geologists Bull., v. 47, No. I, p. 104---128.'
DZt~INSKI, STANISLAW,KSIAZKIEWICZ,MARIAN,AND KEUNE~, P. H., 1959, Turbidites in Flysch of the Polish
Carpathian Mountains: Geol. Soc. America Bull., v. 70, p. 1089-1118.
F~a~NCINL F., AND STORER, D., 1962, I Sedlmentl del Flume Adlge: Unpublished Report AGIP.
FRIEDMAN, G. M., 1961, Distinction between dune, beach, and river sands from their textural characteristics:
Jour. Sedimentary Petrology, v. 31, No. 4, p. 514.-529.
--, 1962, On Sorting, Sorting Coefficients, and the Log-normallty of the Graln-Size Distribution of Sand-
stones: Jour. Geology, v. 70, No. 6, p. 737-753.
CGoT'rxs, C., AND PASSEGA,R., 1960, Synth~se G~ologlque et S&llmentologique de Basse Alsace: 1960, Unpub-
lished Report PREPA.
HIULSTRO~, FILIP, 1939, Transportation of Detritus by Moving Water: in Recent Marine Sediments Sym-
poslum, Amer. Assoc. Petroleum Geologists Publ., p. 5-31.
PASSEGA, R., 1957, Texture as characteristic of Clastlc Deposition: Amer. Assoc. Petroleum Geologists Bull.,
v. 41, No. 9, p. 1952-1984.
--, 1962, Problem of Comparing Ancient with Recent Sedimentary Deposit: Amer. Assoc. Petroleum
Geologists Bull., v. 46, No. 1, p. 114-118.
PASSECA, R., BORGHETrI, G., AND FLORIO, G., 1962, Sedimenti dell'Adriatico al largo di Pescara: Unpublished
Report, AGIP.
RIzzlm, A., CA~.DELLO,R., ANDPASSEGA,R., 1962, Sedimentazlone e Paleogeografia del Pllocene nella Zona d[
S. Salvo: Unpublished Report AGIP.
RIzzxm, A., ANn PASSEGA,R., 1964, Evolution de la S~dlmentation et Orog~n~se, Vall~e du Santerno (Apen-
nin Septentriona]) in Developments in Sedlmentology-3-Turbldltes: Elsevier, New York. 2(64 p.
STETSON, H. C., 1938, The Sediments of the Continental Shelf off the Eastern Coast of the United States:
Papers in Phys. Oceanog. and Meteorol., Mass. Inst. Technol., and Woods Hole Oceanog. Inst., v..5, No. 4.
 G R A I N S I Z E R E P R E S E N T A T I O N B Y CM P A T T E R N S 847

TERZAGHI, K., 1956, Varieties of Submarine Slope Failures: 8th Texas Soil Mechanics and Foundation Engi-
neering Conf. Proc., Austin, 41 p.
VxusE, J. E., 1959, Underwater Geology and Analyses of Recent Sediments off the Northwest Florida Coast:
Jour. Sedimentary Petrology, v. 29, No. 4, p. 555-563.
VEZZANI, F., 1962, I Terreni Oligo-miocenici della Formazione Collesano (Sicilia Nord-Orientale). Sedimenta-
zione e Paleogeografia: Unpublished Report AGI P.
V~zzxr~t, F., AND PASSECA, R., 1963, Applicazione di Nuovi Metodi Sedimentologici allo Studio dell'Appennino
Settentrionale: Boll. Soc. Geol. Ital., v. 82, p. 3-48.
WILUXMS, E., 1960, Intra-stratal Flow and Convolute Folding: Geol. Mag., v. 97, p. 208-214.