2 re Fe ane & eer y
San p eM AIRS IA ES 8 VE

G
Sie Naaret ase alte PPLE pie
6S Sofie

Softeec ttt net eatint pete ee A
ere eae .

ip erat trntyye = peg , 5

AyrenEne te rt ‘ : oP oe 7 fs ;
¥ ti raf x - roe On rt Mem nha estes ~ r * . ae hat beth * + ihe
i ie eb my FP OOS F H RS ‘ : tp at : siya eines

- eRe street arth Mah bo 3 hetater tate hsthalhs Wer tevhe Mey bee baton Mahe ban stale

4 PDP RTS ee es atti ot eyes betwee te Sahel

ah - -< whe Bers ie othe Paty i shete ts teh

Ppt taptp ts Mahe ie eM aihe Sethe

r. et
Th ekg regi chet tit”

Fete eee Gren ieee
ee Ne tng iterate tt the Toth ds Aesth Nine wet ot MP2 a
nt =i , “a Sage tm

et IIe a
ep candied ea ehA uy 7 Rate Ng Tac? Mag ath ge arte ™<
te Vite ea

Matta hat og tye
foren etait pte Vata + vy
ne Vian cataging an att <= Se
v D waie ata i
ate ee ears
ett N tye hata

Preteen ee
A a

nape? cs ah ee gh
. Fatereneree
bn Sa

Jur

pads Saree

fg AIT ee
el ae IY

SS ee ee ee ee PS eS ©
~~ tee

Postilla

PEABODY MUSEUM OF NATURAL HISTORY

YALE*UNIVERSITY
NEW HAVEN, CONNECTICUT, U.S.A.

Number I11 23 October 1967

A FOSSIL COLOBUS SKULL FROM THE SUDAN
(PRIMATES, CERCOPITHECIDAE)

ELwyn L. SIMONS

DEPARTMENT OF GEOLOGY AND PEABODY MUSEUM
OF NATURAL History, YALE UNIVERSITY

ABSTRACT

A fossil monkey skull recovered from the “Old Alluvium” of
presumed Pleistocene age near Wad Medani in the Central Sudan
is described. This find represents a variant of Colobus polykomos
abyssinicus, a living species whose habitat is restricted to forest
canopy. Since this skull is not abraded and is unlikely to have
been water-transported more than a few kilometers to its site of
burial, it provides evidence that continuous forest cover once
reached northwest across the Sudan to the vicinity of Wad Medani.

i)

Postilla YALE PEABODY MUSEUM No. 111

INTRODUCTION

A. fossil cercopithecoid monkey skull assignable to the sub-
family Colobinae and apparently representing a variety of the
living subspecies Colobus polykomos abyssinicus (Oken) was dis-
covered in Nile alluvial deposits of the Sudan in the vicinity of
Wad Medani, Blue Nile, in 1938. This find was made by Mr. W. P.
Archdale, at that time with the Sudan Geological Survey. In Feb-
ruary 1940, the skull was sent by Gerald Andrew, Government
Geologist of the Sudan, to Professor Otto Zdansky, then Chairman
of the Geology Department, Cairo University. Although this
cranium, thickly encrusted with matrix, was partly cleaned by
Dr. Zdansky, it was left undescribed by him. Later he gave the
specimen to Y. S. Moustafa, who took over Dr. Zdansky’s duties
in Vertebrate Paleontology at Cairo University in 1950. The skull
was recently (1963) presented by Dr. Moustafa to the Peabody
Museum at Yale University.

GEOLOGICAL AGE

The age of this fossil within the Pleistocene is not known but
the deep burial in the Old Alluvium of the Gezira Plain, the per-
mineralization of the bone and the surrounding dense concretion
all attest to an age prior to the Recent. Some minor morphological
differences from a considerable series of Colobus polykomos and
C. badius at the American Museum of Natural History, New York,
exist and although not warranting taxonomic distinction at present
these do suggest considerable antiquity for the skull. Even if it is of
early Pleistocene age there is no a priori reason why this skull
should differ in any conveniently definable set of morphological
features from living Black Colobus. For example, in the case of
another primate genus, Simpson (1965) has recently reported
that specimens of fossil galagos from Bed I, Olduvai Gorge,
Tanzania, are extremely similar to the two living subspecies,
Galago senegalensis moholi and G. s. braccatus, of that region
although they are about 1.75 million years older. He remarked
(pS) ez

“I do not mean to suggest that the fossils belong to the
living subspecies of the same region, or even that such sub-

A FOSSIL Colobus SKULL FROM THE SUDAN 3

species are definable and valid as now recognized...... The
fact that the fossils cannot be definitely distinguished from
living Galago senegalensis does not amount to absolute deter-
mination that they are of that species. Larger series, especially
with upper dentitions and skulls, might well demonstrate a
distinction, but the parts now known in the Olduvai Bed I
form do suffice to indicate that it is in any event very closely
related to the living lesser Galago of the same general region.”

The question of absolute age of this skull of Colobus is con-
sequently of some interest. At least one method of geochemical
dating may apply in this case and this is currently being investi-
gated further at Yale. If this skull contains measurable uranium
and if the bulk of this uranium entered the skull shortly after
burial so that the time of mineralization of the bone was short
compared with subsequent time then a possibility exists that this
skull could be dated by the Thorium 230/Uranium 238 method.
This method has already been utilized for fossil bones as reported
by V. V. Cherdyntsev (1965: Engl. transl).

ECOLOGICAL SIGNIFICANCE

The skull exhibits no abrasion or other evidence of significant
fluviatile transportation so it must be assumed that the occurrence
of this fossil in the Old Alluvium near Wad Medani appears to
establish the fact that within only a few kilometers of that city,
dense forest conditions once existed. Colobus monkeys are entirely
confined to canopy forest habitat and do not normally travel on
the ground even for short distances. As Schenkel and Schenkel-
Hulliger (1966) have recently reported, their observation of
Colobus polykomos in Kenya indicates that even in partly cut-
over forest these monkeys when seen on or near the ground pre-
ferentially leap or walk on fallen branches. Thus range extension
for these primates can be accomplished only by travel through
virtually continuous forest canopy.

This specimen is therefore of particular zoogeographic inter-
est, for it was found about 400 km. west of the present possible
range in Ethiopia of members of this genus (see Figure 1). Within
the Sudan itself one has to look 1,050 km. to the south in the

4 Postilla YALE PEABODY MUSEUM No. 111

Imatong mountains to find present-day Colobus. The prior exist-
ence of Colobus polykomos abyssinicus in the vicinity of Wad
Medani thus strongly suggests a much wetter climate in the Central
Sudan formerly and implies that there once existed continuous
forest cover, perhaps running several hundred kilometers west-
ward from the highland forests now existing in the Lake Tana
region of Abyssinia at the headwaters of the Blue Nile.

ep 5 Massawo
Locality of fossil mee
Colobus skull.

KORDOFAN

(Present western :
limit of forest habitat KEEN EATS
suitable for Co/obus in Zee 5A

5 Ethiopia. z
fizzy Present range of
Colobus in the Sudan

Butl Imatong Mts
utler, |966 25 35

) 00. 200. 300. 400. +500 600 700 800 KILOMETRES

FIGURE 1

COLOBINE FOSSILS

Most fossil monkeys so far recovered from Pliocene-Pleis-
tocene localities in Africa (species of Simopithecus, Cercopithe-
coides, Gorgopithecus, among others) have been ranked with the

A FOSSIL Colobus SKULL FROM THE SUDAN 5

cercopithecine division of Old World monkeys. However, some
early colobines may also exist among known African materials.
Considering the different ecologic requirements of members of these
two subfamilies, the taxonomic allocation of certain fossil species
requires serious attention. DeVore and Washburn (1963) stated
that Cercopithecoides williamsi from the Pleistocene of South
Africa is a colobine monkey. Stromer (1913) figured a partial
maxillary dentition from the Egyptian Pliocene (Wadi Natrun)
which he regarded as a semnopithecine, now colobine, monkey.
As the dentition is much worn, little that is diagnostic remains
and the subfamilial allocation is consequently uncertain. Libypi-
thecus markegrafi, which Stromer described from this same site, is
based on a good skull which has generally been taken to represent
a primitive baboon. Piveteau (1957), however, included his dis-
cussion of the Natrun skull of L. markgrafi among his section on
colobine monkeys. More recently, Verheyen (1962) has devised
an index, interorbital width x 100/length from glabella to pros-
thion, in which almost all cercopithecine and colobine monkeys
differ. In this index Libypithecus falls in the same general zone
as species of Papio and Macaca and well away from those of
Mesopithecus and Cercopithecoides. Macinnes (1943:148, pl. 24)
discussed fossils of a cercopithecine species, which he regarded as
belonging to Mesopithecus, from Miocene deposits on Rusinga
and Kiboko Islands, Kiverondo Gulf, Lake Victoria, Kenya. The
type species of genus Mesopithecus comes from the European
early Pliocene. Pikirmi faunas of Greece and Turkey has also
been generally treated as belonging to the colobine (= semnopi-
thecine) group (Gaudry, 1862; Piveteau, 1957). Such an assign-
ment was strongly questioned by Patterson (1954) but a careful
investigation of the subfamilial assignment of Mesopithecus from
the Pontian of southern Europe has yet to be undertaken. Since
MaclInnes (1943) published on this African early Miocene mon-
key, however, additional material has been found that has been
under study by Von Koenigswald at Utrecht. He has recently
pointed out (personal communication, 1966) that this undescribed
material from Kenya is not Mesopithecus but represents a new
genus and species — the oldest undoubted monkey. His analysis
of this species, when published, may help to clarify the place and
time of differentiation of the two cercopithecoid subfamilies.

6 Postilla YALE PEABODY MUSEUM No. 111

LOCALITY AND LITHOLOGIC ASSOCIATION

The Sudanese fossil monkey skull was found weathering out
of a bluff of alluvial deposits on the east bank of the Blue Nile
15 km SSE of Wad Medani, latitude 14° 18 1%’ N, longitude
33° 37’ E, approximately 1.5 km upstream from Abu Suheli
Gubba (an existing landmark on the west bank). It was discovered
lying in a position which indicated derivation from a dark clay
horizon. These clays, locally called “Old Alluvium,” are often
heavily impregnated with fresh-water limestone (kankar) nodules
or beds, as are the sands and gravelly sands underlying the clays.
On discovery, the skull reported here was also covered with a
kankar concretion. These dark clays compose the “cotton-soil” of
the Gezira Plain, the so-called ‘“‘aeolian clays” of early authors,
and are very thick (12-20 meters). In this region the river flows
in a valley probably 30 meters deep below the Gezira Plain, with
about four or five meters of cracked soil and aeolian clay overly-
ing the Old Alluvium from which the skull was derived. In some
areas, the Old Alluvium is cut into Pliocene sediments but its
temporal position in the Pleistocene is at present unknown.

SYSTEMATICS

Order Primates, Linneus, 1758
Suborder Anthropoidea Mivart, 1864
Family Cercopithecidae Gray, 1821
Subfamily Colobinae Elliot, 1913
Genus Colobus Illiger, 1811

Colobus polykomos abyssinicus (Oken) 1816

Bisures) 2253

Verheyen (1962) in the most recent revision of the genus
Colobus distinguished three species of Black Colobus: C. poly-
komos, C. abyssinicus, and C. satanas. Schwartz (1929) regarded
these three varieties as members of one species, a view continued
by Fiedler (1956) and a usage followed herein.

This skull resembles female skulls of the black Colobus sub-
species group, particularly Colobus polykomos abyssinicus, much
more closely than those of the Red Colobus, C. badius. However,
it is noticeably different from C. p. abyssinicus in several details,

A FOSSIL Colobus SKULL FROM THE SUDAN fh

some of which suggest a certain degree of primitiveness compared
to the present-day East African varieties. These features of dif-
ference from typical members of modern Colobus polykomos
subspecies include a slightly smaller cranial vault, well-defined
frontal bosses, more quadrate orbits, palate bowed out more
widely, development of a deep groove on the posterior face of
upper canines, and a heavy ridge running posteroventrally on the
lateral external walls of the orbits. Inasmuch as the variation of
such cranial structures in Colobus is broad and only this single
specimen of unknown age is available from the Wad Medani
region, it does not seem advisable to name a new subspecies in
this case.

In fact, naming subspecies among Tertiary (or earlier) mam-
mals is generally inadvisable. This is because any “‘subspecific”
metrical or morphological differences noted between two or more
type specimens might equally well have been specific. Since it is
impossible to test interfertility in fossils representing populations
which might be either species or subspecies, these two taxonomic
levels cannot be clearly distinguished in fossil samples. Subspecific
names of fossil vertebrates often become mere “handles” for one
or more individual specimens which seem different, and are thus
applied in a different manner from common usage of subspecies
by most neontologists. In addition, there is disagreement as to
whether subspecies can truly be defined even among _ living
organisms.

MORPHOLOGICAL DESCRIPTION

GENERAL MORPHOLOGY. The fossil skull (YPM 19063) is remark-
ably well preserved. It is uncrushed and nearly complete, with
only the tympanic bullae, zygomatic arches, incisors, and left
canine missing. The left orbital wall shows slight damage postero-
laterally, and the distal portions of the nasal bones are partly
broken off. The upper incisors and the left canine were lost before
fossilization. The specimen seems to have suffered little deforma-
tion except for a slight displacement internally of part of the
occipital beneath the upper margin of this bone.

Judging from the relative cranial proportions, this skull (YPM
19063) probably belonged to a young adult. The third upper
molar is almost at the level of the other molars, which would

8 Postilla YALE PEABODY MUSEUM No. 111

indicate the animal was not fully mature. Other evidences of
(individual) age are that the frontoparietal suture is not entirely
closed and the interparietal is open. Across the frontals, the dis-
tance between the two supratemporal ridges (temporal crests)
is about 25 mm. In many full adults of black Colobus species
these crests run closer to the midline.

The size of the canine and gracile skull morphology suggest that
the individual represented is female. The base of the canine is not
particularly large in cross-sectional area and its root is comparatively
short. In addition, the rostral face of the maxilla lacks the canine-
root buttress which normally accompanies the relatively larger
canine root of male Colobus. Although the tip of the right canine
is chipped off, projection of the lateral planes of the basal portion
of this tooth suggests that it did not extend much beyond the
general level of the upper tooth row. If the specimen under con-
sideration were male, the canine would have to be much larger
and longer than it is.

SKULL ROOF. This skull shows a remarkably short face. As in
present day Colobus polykomos the nasal bones are also com-
paratively short. Although broken distally, the curvature of the
dorsal surface of the nasals indicates these bones were turned
slightly upward at their ventral extremity. The interfrontal suture
is closed, but in its position a shallow groove running anteropos-
teriorly separates two slightly elevated frontal bosses. These two
bosses reflect the modeling of the frontal lobes of the brain some-
what more clearly than is usually the case in existing members of
Colobus. This lower vaulting of the frontal may also indicate a
slightly smaller volume for the frontal lobes than is typical of
present-day Colobus skulls of comparable size. The superorbital
ridges are fairly well marked, and posterior to them is a distinct
transverse depression. In modern species of Colobus the frontal
portion of the braincase is generally more expanded anteriorly.
In this character, the Abu Suheli skull seems perceptibly more
primitive.

Slightly anterior to the point of juncture of the frontal and
the interparietal suture, the skull vault is depressed below the
general dorsal are of the braincase. The maximum diameter of the
skull in the parietal region is 55.7 mm, as compared with the

B.

FIGURE 2 — YPM 19063
A. Facial view, X approx. 1.25.
B. Right lateral view, x 1.

FIGURE 3 — YPM 19063
AcmPalatesssaliai-
B. Dorsal view, x 1.

A FOSSIL Colobus SKULL FROM THE SUDAN 9

width across the postorbital bar, 44.3 mm. The constriction behind
the orbits is more pronounced than in most living Colobus poly-
komos abyssinicus.

A medial suborbital wing of the jugal (malar) contacts the
lacrimal, thus excluding the maxillary facial surface from the
orbital rim. As is typical of Cercopithecidae, the lacrimal foramen
is located well within the orbit. On each side there appears to
be but a single infraorbital foramen. There is no evidence of the
occurrence of a supraorbital foramen on either side of the skull.
Just below the median extremities of the supraorbital rims there
are indications of the presence of a pair of venous foramina
located on either side of the interorbital septum. The orbits are
nearly circular in Outline, with greatest transverse diameter in each
approximately 23.8 mm, and greatest vertical diameter, 22.8 mm.

A somewhat unusual structural feature is a crest which runs
ventrally for about 10 mm on the temporal face of the postorbital
plate, arising at the point of junction of sutural contact between
the external angular process of the frontal with the jugal (malar).
This crest fades smoothly into the surface of the postorbital plate,
forming a shallow fossa anterior to it. At this point, as a con-
sequence of the presence of this subsidiary crest, the dorsoexternal
corner of the orbit is comparatively massive. As is typical of
Colobinae, the parietal in this skull joins an upward extension of
the alisphenoid at the pterion, thus forming a sphenoparietal
suture. Among cercopithecine monkeys these two cranial elements
are commonly separated by a forward extension of the squamosal
which makes sutural contact with the frontal.

PALATE. The hard palate extends beyond the posterior end of
the tooth row. The posterior choanae are comparatively narrow
transversely, and, as in most Anthropoidea, open a short distance
posterior to a transverse line across the posterior faces of the
third molars. The palatal surface is flat and lies nearly 10 mm
above the lingual bases of the teeth. Since the skull shows no
signs of having been laterally compressed, the position of the
U-shaped dental arcade may be safely assumed to be undisturbed.
The outward bowing of the cheek teeth and the relatively small
teeth are resemblances to Colobus polykomos polykomos rather
than to Colobus polykomos abyssinicus. Since in most other fea-

10 Postilla YALE PEABODY MUSEUM No. 111

tures, YPM 19063 is more like C. p. abyssinicus, this suggests,
perhaps, that the population represented by this specimen was
more generalized than either of these two present-day subspecies.

occipuT. The supraoccipital ridge (lambdoidal crest) is rather
weakly developed; it does not project markedly at the inion which
is slightly displaced to the right of the midline, thus offsetting the
supraoccipital suture from the latter. The supraoccipital surface
seems to have been fractured along a curved line which runs
parallel to the supracccipital ridge. It also appears that the frac-
tured occipital surface ventral to the fracture had slipped under the
fixed dorsal remainder of the bone, thus decreasing the exposed
area of the occiput above the foramen magnum. This fracturing
is most probably related to the factor that caused the loss of the
tympanic bullae.

DENTITION. The dental arcade exhibits the U-shaped outline
characteristic of Old World monkeys. The four upper incisors
are missing; however, the situation of their alveoli indicates they
were procumbent, and the lateral incisors were separated basally
at least by a slight diastema from the canines. There is no diastema,
however, between the canine and the upper third premolar.

The cross-section of the canine at its base is roughly oval, with
a relatively broad front curvature. There is a distinct ridge which
runs along the lingual surface of the canine, tapering posteriorly.

The third and fourth upper premolars, P* and P*, are bicuspid
and although the protocone of P* is somewhat damaged, this tooth
appears to be slightly narrower transversely than P*. At the poste-
rior end of P* there is a posterior fovea, bounded by a distinct
posterior crescentic ridge connecting the apices of the protocone
and ectocone. The apices of the two cones of P* indicate greater
wear than those of P*.

The first upper molar, M'‘, is a simple quadritubercular tooth,
with the two lingual cusps — protocone and hypocone — con-
siderably worn compared with the labial cusps — metacone and
paracone. There is no distinct cingular development on either side
of M' or the other cheek teeth. Foveae are present on the anterior
face of the paracone and the posterior face of the metacone.
Pronounced transverse ridges connect the tips of |) the protocone

A FOSSIL Colobus SKULL FROM THE SUDAN ila

and paracone and 2) the metacone and hypocone. These ridges
are not interrupted by wear.

The second upper molar, M?, is the largest of the three molars.
It conforms closely to the basic pattern displayed by M', except
that the crest which connects metacone and hypocone is less
clearly expressed.

The third upper molar, M*, is the second largest molar; it
generally resembles M? except for a slight shortening of its antero-
posterior diameter. The labial groove between paracone and meta-
cone is more deeply incised in this tooth than in either M* or M’.
The crest that connects metacone and hypocone in M' and M? is
broken in M®* between the two cusps. In this respect, the arrange-
ment of cusps on M* deviates from the bilophodont pattern of the
other two molars.

ACKNOWLEDGEMENTS

Research reported here was facilitated by grants from the
Wenner-Gren Foundation of New York and by grants from the
National Science Foundation (Nos. 6-18102 and GP-433). I have
also to thank Y.S. Moustafa for supplying locality data and other
preliminary notes on the specimen we had compiled jointly in
Cairo in 1961. Photographs for figures 2 and 3 were prepared by
John Howard of the Peabody Museum staff.

REFERENCES

Butler, H. 1966. Some notes on the distribution of primates in the Sudan.
Folia Primatologica: 4:416-423.

Cherdyntsev, V. V. 1956. Abundance of chemical elements [in Russian].
State Publishing House of Technical-Theoretical Literature, Moscow.
[English edition, 1961, Univ. of Chicago Press, 304 p.}.

DeVore, I., and S. W. Washburn. 1963. Baboon ecology and human evolu-
tion, p. 335-367. Jn Howell and Bourliere [ed.| African ecology and
human evolution. Aldine Press, Chicago.

Fiedler, W. 1956. Ubersicht iiber das system der Primates. /n Primatologia
1: 1-266, S. Karger.

Edinger, T. 1938. Mitteilungen iiber Wirbeltierreste aus dem Mittelpliocan
des Natrontales (Aégypten). 9 Das Gehirn des Libypithecus. Zentral-
blatt flr Mineral., Ser. B, p. 122-128.

Gaudry, A. 1867. Animaux fossiles et géologie de l’Attique. Paris, 475 p.;
Atlas, 75 pls. (1862-1867).

{2 Postilla YALE PEABODY MUSEUM No. 111

MaclInnes, D. G. 1943. Notes on the East African Miocene Primates. East
Afr. and Uganda Nat. Hist. Soc. 17 (3 and 4): 141-181.

Piveteau, J. 1957. Primates, Paleontol. humaine. Traiteé de Paléontol.,
Masson et Cie., Paris, 7: 1-675.

Schenkel, R., and L. Schenkel-Hulliger. On sociology of free-ranging Colo-
bus. In Conference Report of the First International Primatological
Society meetings, July 1966, S. Karger (in press).

Schwartz, E. 1929. On the local races and distribution of the Black and
White Colobus monkeys. Zool. Soc. London, Proc. pt. Il: 585-598.
Simpson, G. G. 1965. Order: Primates. Jn Olduvai Gorge 1965, Vol. 1,

L. S. B. Leakey, Cambridge Univ. Press, pp. 1-117.

Stromer, E. 1913. Mitteilungen iiber die Wirbeltierreste aus dem Mittel-
pliocin des Natrontaies (Aégypten). 1. Affen. 2. Raubtiere. Zeitschr.
deutsch. geol. Ges. vol. 65, no. 3, p. 350-372.

Verheyen, W. N. 1962. Contribution a la craniologie compareé des Pri-
mates: les genres Colobus Uliger 1811 et Cercopithecus Linne. 1758.
Ann. Mus. Roy. de l’Afrique Centrale, Ser. in 8°, Sci. Zool. no. 105,
pp. 1-255.

=

+ ar etc Da t-te dap erernareaeaoer steele

a
Me

¥

NEL We

Steet dee dete Neaeoard