Primates (2005) 46:21–32
DOI 10.1007/s10329-004-0109-5
O R I GI N A L A R T IC L E
Martin Pickford
Incisor-molar relationships in chimpanzees and other hominoids:
implications for diet and phylogeny
Received: 20 February 2004 / Accepted: 13 July 2004 / Published online: 21 October 2004
Ó Japan Monkey Centre and Springer-Verlag 2004
Abstract In chimpanzees, the cutting edge of the incisor
battery is longer in relation to the length of the molar
row than in any other hominoid, extant or fossil, the
only other lineage approaching it being the orangutan.
Apart from their increased mesio-distal dimensions, the
upper and lower incisors of chimpanzees differ in addi-
tional ways from those of almost all other hominoids.
The I2/ is enlarged, so that the difference in size between
it and the central upper incisor is less than it is in the
heteromorphic upper incisors of other hominoids. The
lower incisors are expanded mesio-distally, so much so
that isolated I/2 crowns can resemble upper central
incisors. In chimpanzees the lingual surface of the lower
incisors is generally more procumbent than it is in other
hominoids, which have more vertically oriented incisor
crowns and there is a greater difference in enamel
thickness between labial and lingual sides. The re-ori-
entation of the lower incisor crown is reflected in the
root, which in lateral view is anteriorly concave in
chimpanzees whereas it is more orthogonal or convex in
other hominoids. The molars of chimpanzees, especially
the lowers, have extensive and relatively deep occlusal
basins, and the main cusps are peripheralised and labio-
lingually compressed, making them more trenchant than
those of other hominoids. This paper examines the
incisor-lower molar proportions in extinct and living
hominoids and develops a new hypothesis about the
evolution of the dentition of chimpanzees and links it to
their diet. It also examines the incisor-molar proportions
of hominids and African apes in order to throw light on
the phylogenetic relationships between them. It is shown
that chimpanzees are highly derived in this respect and
that several recent ideas concerning the chimp-like
M. Pickford (&)
Département Histoire de la Terre, UMR 5143,
CNRS, 8, rue Buffon, Paris, 75005, France
E-mail: pickford@mnhn.fr
M. Pickford
Collège de France,
11, Place Marcellin Berthelot, Paris, 75005, France
appearance of the last common ancestor of hominids
and African apes are likely to be incorrect.
Keywords Hominoidea Æ Incisors Æ Molars Æ
Meat-eating Æ Phylogeny
Introduction
It has long been known that, when compared with other
living or extant apes, chimpanzees possess relatively
elongated incisor batteries measured mesio-distally.
Functionally, the elongated incisor battery in chimpan-
zees has often been linked to frugivory (Pilbeam 1969),
but in view of the fact that most fossil hominoids possess
mesio-distally short incisors, such a hypothesis seems to
be an incomplete explanation, since there are good
reasons for thinking that most of the Miocene homi-
noids were frugivorous, as are most of the extant ones
including gibbons and siamangs, and their incisor bat-
teries are not mesio-distally enlarged. This paper ex-
plores the relationships between the length of the lower
incisive cutting edge in various extant and extinct
hominoids and compares it to the length of the molar
row as an independent variable. It then examines the
diets of various extant and fossil hominoids (where
possible) to determine whether there are any significant
differences between chimpanzee diets or ways of pro-
cessing foods and those of other hominoids that could
explain the dental differences.
In all known Miocene hominoids the lower incisors
are relatively small and peg-like, with parallel to sub-
parallel mesial and distal edges, weak or absent lingual
ridges and small or absent basal tubercles (Figs. 1, 2).
The lower incisors are also more vertically oriented than
are those of chimpanzees and the enamel on the labial
and lingual surfaces is almost the same thickness (the
lingual enamel is usually only slightly thinner than the
labial enamel). In both species of chimpanzees, however,
the lower incisors are expanded mesio-distally and their
22
Fig. 1 Oblique view of a cast of a mandible of Proconsul nyanzae/
heseloni to show tall, mesio-distally short, peg-like lower incisors
with reduced basal tubercle and central lingual ridge
Fig. 2 Oblique view of a cast of a lower jaw of Sivapithecus indicus
showing tall, mesio-distally short incisors with no basal tubercle
and lingual ridge
mesial and distal edges are spatulate, so much so that
isolated specimens may be confused with upper incisors.
They also possess strong lingual ridges and large basal
tubercles and the lingual enamel is appreciably thinner
than that on the labial surface (Figs. 3, 4).
Most species of Miocene hominoid (Kenyapithecus,
Nacholapithecus, Afropithecus, Proconsul, Otavipithecus,
Ugandapithecus and many others) have a large spatulate
upper central incisor, but a much smaller more peg-like
lateral one. There is thus marked incisor heteromorphy
in the premaxillary teeth of Miocene hominoids. But
Fig. 3 Lingual view of the lower incisors of a Pan troglodytes
mandible from Mahale (University of Kyoto collection) showing
worn incisors, large basal tubercle and central lingual ridge
Fig. 4 Lingual view of the lower incisors of a P. troglodytes
mandible from Mahale (University of Kyoto collection) showing
the great difference in enamel thickness between lingual and labial
sides of the crown, in this individual enhanced by differentially
greater wear on the lingual side
chimpanzees have enlarged lateral upper incisors which
are spatulate to such an extent that isolated specimens
can be confused with central ones by the uninitiated,
something which is unlikely to occur with Miocene
hominoids. Because of this the sockets for the lateral
incisors are located more laterally than they are in other
hominoids, being positioned lateral to the edge of the
nasal aperture. They also often have lingual ridges,
which are weak or absent in the lateral incisors of
Miocene apes and extant gorillas.
Apart from the differences in crown morphology, the
lingual surfaces of the lower incisors of chimpanzees are
more procumbent than they are in Miocene hominoids.
They lean anteriorly considerably more than in other

hominoids, forming an angle of about 30° between the
lingual aspects of the crown and the root (Fig. 5). In the
lower incisors of Miocene hominoids the angle is con-
siderably less than 30°, most genera having angles of less
than 5°. Furthermore, in many Miocene hominoids,
both large (Proconsul, Nacholapithecus, Kenyapithecus)
and small (Micropithecus, Dendropithecus, Limnopithe-
cus, Simiolus), the lower incisors are relatively tall and
mesio-distally short, the cervix to apex measurement in
unworn teeth being much greater than the mesio-distal
length or the labio-lingual breadth. In chimpanzees in
contrast, the crowns are mesio-distally much longer,
making them more spatulate. From a Miocene ape
perspective there can be no doubt that the enlarged
spatulate lower incisors of chimpanzees are derived.
Other derived features in lower incisors of chimpanzees
are the relative enlargement of the basal tubercle, the
degree of enhancement of the central ridge, the thinning
of the lingual enamel, the increase in procumbency of
the lingual surface (which increases during the lifetime of
the individual) and the concave profile of the labial side
of the tooth.
Methods
The lengths of the cutting edge of the lower incisors and
the lower molar row were obtained by measuring teeth
in situ in mandibles of extant and fossil hominoids, and
by searching the literature for comparable measure-
ments. The mesio-distal length of each incisor crown was
measured and the four measurements added together to
produce the length of the cutting edge. In specimens
lacking one or two incisors, measurements of the inci-
sors on one side were taken, added together and then
doubled. In a few cases in fossils where there were no
incisors preserved, the incisor alveolar chord was mea-
sured to which was added an allowance for the upward
flaring of the incisor crowns. In chimpanzees differences
of up to 2 mm may occur between measurements of the
Fig. 5 Distal views of the lower incisors of Pan troglodytes, Orrorin
tugenensis (cast) and Gorilla gorilla. Note the labial concavity in the
profile of the Pan incisor and the more upright convex labial profile
of the incisors in Orrorin and Gorilla
23
incisor alveolus chords and the apical incisive cutting
edge, meaning that the margin of error in the latter
method of estimating the length of the incisor cutting
edge is minor, and it does not radically alter the resulting
analysis. Measurements of lower molar row were taken
directly on mandibles containing all three teeth if they
were in correct line, but in specimens with teeth out of
normal position, or still erupting, the mesio-distal length
of individual molars was obtained and the measure-
ments added together. Errors reflected in the latter cases
(only a few specimens) are 1–2 mm, and do not greatly
affect the resulting analyses.
The table of measurements was treated using Excel to
produce bivariate plots of length of incisive cutting edge
against length of lower molar row (Figs. 6, 7, 8).
Results
Analysis of incisor-molar proportions
Figure 6 reveals that most fossil and extant hominoids
possess incisor and molar rows that plot out relatively
narrowly along a regression line. For most of the species
sampled, lower molar row is approximately 1.6 times the
length of the incisive cutting edge. There are exceptions
(Fig. 7), one of which comprises chimpanzees (both
species) which have greatly elongated incisive cutting
edges (the incisive cutting edge is approximately equal in
length to the molar row) and another of which impli-
cates robust australopithecines, which have small inci-
sors and relatively huge molar rows (molar row is 2.75
times the length of the incisive cutting edge in Paran-
thropus; Fig. 8). The orangutan, like the chimpanzee,
deviates from the regression and the length of the molar
row is less than that of incisor cutting edge (Fig. 8). The
mean molar row length is about 1.7 times the incisor
length, and considering body-size, the deviation is less
than that of the chimpanzees. Humans fall slightly on
the low side of the regression (Fig. 8), but this appears to
be mainly due to a slight reduction of the size of the
molars rather than increase in the dimensions of the
incisors which show no sign of being spatulate.
Extant and fossil humans possess lower incisors that
are similar to those of most Miocene apes, and their
incisor-molar relation is close to that of Miocene apes.
Early australopithecines have lower incisors that are
close to or slightly shorter mesio-distally relative to
molar row length than those of Miocene hominoids
(Fig. 8), whereas Paranthropus and Gigantopithecus have
a very short incisor battery relative to length of the
molar row (Fig. 8). As a result they fall way off the
hominoid incisor-molar regression line, but on the
opposite side of it from the chimpanzees.
It is interesting to note that the more carnivorous
common chimpanzee (Pan troglodytes) has the most
elongated and most procumbent incisive cutting edge of
all known hominoids, whether extant or fossil, and the
pygmy chimpanzee which seldom eats meat has less
24
Fig. 6 Bivariate plot of length of cutting edge of lower incisors
(x-axis) on length of lower molar row (y-axis) of various Miocene,
Pliocene, Pleistocene and extant Hominoidea, omitting Pan, Pongo,
robust australopithecines and Gigantopithecus [included in this
figure (numbers of individuals in brackets) are Afropithecus
turkanensis (2), Australopithecus afarensis (3), Australopithecus
africanus (3), Australopithecus anamensis (1), Dendropithecus
macinnesi (1), Gorilla gorilla (44), Griphopithecus alpani (1), Homo
sapiens (10), Homo fossils (7), Hylobates klossi (3), Hylobates lar
(5), Hylobates molloch (2), Hylobates muelleri (4), Kalepithecus
songhorensis (1), Kenyapithecus africanus (2), Limnopithecus evansi
(1), L. legetet (2), Lufengpithecus lufengensis (4), Micropithecus
clarki (1), Nacholapithecus kerioi (2), Nyanzapithecus pickfordi (1),
Orrorin tugenensis (1), Otavipithecus namibiensis (1), Pliopithecus
vindobonensis (1), Proconsul nyanzae/heseloni (6), Rangwapithecus
vancouveringi (1), Simiolus enjiessi (1), Sivapithecus indicus (4),
Symphalangus (10), Turkanapithecus kalakolensis (1), Ugandapithe-
cus major (4)]
elongated incisor cutting edges and the incisors are less
procumbent. The lowest measurements of P. troglodytes
sampled were invariably old individuals with consider-
ably worn incisors, yet the incisive cutting edge was still
greater in length relative to molar row length than in any
other genus of hominoid. There does not appear to be
marked sexual dimorphism in incisor-molar proportions
in chimpanzees despite major dimorphism in canine size
and body weight.
Analysis of incisor morphology and orientation
Chimpanzee incisor alveoli appear to have similar ori-
entations to those of other hominoids, but the lingual
surfaces of chimpanzee lower incisors are more pro-
cumbent. This surface forms a more open angle relative
to the cheek tooth occlusal surface than in other homi-
noids. This is achieved in two ways. Firstly, the incisor
crown is more angled (procumbently) on the root than it
is in other hominoids, such that in lateral view the labial
profile of the root-crown margin is less upright to slightly
concave, whereas in other hominoids it is more upright to
slightly convex (Fig. 5). Secondly, there is an increase in
the labio-lingual dimension near the cervix due to
enlargement of the basal tubercle and this has the effect of
increasing the angle of the incline from apex to cervix on
the lingual side of the crown. Thus, in chimpanzees, the
upper and lower incisors form a narrower angle when in
occlusion than, for example, in gorillas, in which the
lower incisors are more vertically oriented (Fig. 10).
Incisor enamel thickness in most hominoids is almost
equal on the lingual and labial surfaces, the lingual en-
amel being only a little thinner than that on the labial
side, or in some cases thicker lingually (Figs. 9, 10, 11,
12). In chimpanzees, however, in particular P. troglo-
dytes, the lingual enamel is thinner lingually than labi-
ally (Fig. 4). Thus, during wear, the lingual surface
abrades faster than the labial side, a result of which is
production of a sharp cutting edge at the apex (Figs. 3,
4). Rodents achieve a similar effect by having no enamel
at all on the lingual surface of the incisor crowns.
Young adult gorillas, orangutans, australopithecines,
Orrorin and many of the Miocene apes that could be
examined produce apical wear facets on the lower inci-
sors which are almost at right angles to the height axis of
 25

Fig. 7 Bivariate plot of length

of cutting edge of lower incisors
(x-axis) on length of lower
molar row (y-axis) of various
Miocene, Pliocene, Pleistocene
and extant Hominoidea,
including Gigantopithecus blacki
(1), Pan paniscus (20), Pan
troglodytes (33), and
Paranthropus boisei (4)

Fig. 8 Bivariate plot of length

of cutting edge of lower incisors
(x-axis) on length of lower
molar row (y-axis) of various
Miocene, Pliocene, Pleistocene
and extant Hominoidea,
omitting Pan and Pongo, but
including robust
australopithecines and
Gigantopithecus and
highlighting Late Miocene
Orrorin, Pliocene
australopithecines and early
Homo

the crown, the wear surface being almost parallel to the
occlusal surface of the cheek teeth (Fig. 13). In chim-
panzees these wear facets are also parallel to the occlusal
surface of the cheek teeth, but because of the procum-
bency of the incisor crowns (Fig. 14), the facets are
oriented at much less than a right angle to the labial
surface of the incisors (Fig. 15). Because of this, and
because of the differences in enamel thickness lingually
and labially, chimpanzee lower incisors have sharper
apical cutting edges than occur in other hominoids. In
26
Fig. 9 Oblique view of a cast of the mandible of Paranthropus
boisei from Peninj, Tanzania, showing normal sized incisors, but an
immensely expanded molar row
Fig. 10 Oblique view of the mandible of Gorilla gorilla in the
Kyoto University collection. Note the tall, but mesio-distally short
incisors
senile hominoid individuals wear facets become less clear
and other abrasion factors come to predominate, with
the result that the incisors look like pegs with rounded
apices. But in chimpanzees, even very aged individuals
may possess sharp cutting edges to their incisors, as long
as some enamel remains labially.
In hominoids, the upper and lower incisors have
slightly different but complementary functions, the up-
per incisors having sharp apical edges and the upright
lower incisor batteries having blunt horizontal wear
facets. In chimpanzees, in contrast, both the lower and
Fig. 11 Oblique view of a cast of the mandible of Homo habilis
from Olduvai Gorge, Tanzania, showing normal incisors and
somewhat expanded lower molars
Fig. 12 Oblique view of mandible of Homo sapiens from Japan,
showing tall, mesio-distally short incisors and slightly reduced
lower molars
upper incisors have sharp cutting edges and are angled
outwards (procumbent). By moving the lower jaw
slightly backwards and forwards chimpanzees achieve a
cutting action which alternates between upper and lower
incisive batteries. Wear scratches on the lingual surface
of chimpanzee incisors are mostly oriented parallel to
the root-apex axis (Fig. 15), suggesting that the principal
movement of the jaws is forwards and backwards, with
very little lateral movement.
Perhaps because the apical edge of chimpanzee lower
incisors has taken on a cutting function, the crown
morphology of their lower incisors has converged to-
wards that of the upper incisors. Thus, uniquely among
African hominoids, the first and second lower incisors of
chimpanzees are spatulate rather than peg-like, and
possess well-developed lingual ridges and clear mesial

Fig. 13 Lingual view of lower incisors of Pan troglodytes, Orrorin
tugenensis (cast) and Gorilla gorilla to show the spatulate shape of
the Pan incisor compared with the more parallel-sided incisors of
Orrorin and Gorilla. Note also the large basal lingual pillar and
central lingual ridge in the Pan incisor
Fig. 14 Oblique view of a mandible of Pan troglodytes from
Mahale, Tanzania, stored at Kyoto University, to show the
procumbent incisor row and the relation between the incisors and
molars
and distal ridges (Fig. 16) which are unknown or only
weakly evident in other hominoids, apart from some
individuals of Pongo (Fig. 17). The convergence in form,
angulation and size of the lower incisors to the upper
ones is so striking that on more than one occasion the
author has found museum specimens in which chim-
panzee lower incisors have been glued into upper incisor
sockets. The opposite does not happen so frequently
because central upper incisor roots are appreciably
bigger than lower incisor alveoli and they won’t fit into
them but it does happen with the lateral upper incisor.
27
Fig. 15 Lingual surface of lower incisor of Pan troglodytes from
Mahale, Tanzania, showing wear and scratches. Apical is upper
right, and cervical is lower left of the image
Fig. 16 Oblique view of a cast of a mandible of Pan paniscus from
Congo, stored at Tervuren, Belgium, showing spatulate lower
incisors with enlarged basal lingual pillars and central lingual ridges
Discussion
Meat-eating adaptations in the dentition of chimpanzees
The question naturally arises about the function of en-
larged and spatuliform lower incisors with permanently
sharp cutting edges that occur in chimpanzees. If frugi-
vory is not the complete answer, then what are chim-
panzees consuming that requires a long, sharp cutting
incisive edge? A possible answer is meat. If so, then do
they show any other dental adaptations to meat-eating?
Meat procurement and canine function
Traditionally the differences in dimensions and shape
between the canines of male and female chimpanzees
28
Fig. 17 Oblique view of a mandible of Pongo pygmaeus stored at
Kyoto University. Note the somewhat spatulate incisors that form
a long cutting edge relative to the dimensions of the molar row
have been explained in terms of sexual selection and
anti-predator defence. More recently it has been ob-
served that chimpanzees may on occasion kill conspe-
cifics, the canines being used to inflict serious wounds to
the head and body of the victim (Watts 2004). The same
explanations have been applied to orangutans and
gorillas as well as many of the Miocene hominoids
(Pickford 1986a–1986c). However, most large bodied
Miocene hominoids, including Afropithecus, Ugandapi-
thecus, Kenyapithecus, Nacholapithecus, Ouranopithecus,
Ankarapithecus, Oreopithecus, Sahelanthropus and Si-
vapithecus, have large canine roots but stubby, relatively
blunt crowns which undergo apical wear and generally
have poorly developed or absent honing action against
the lower third premolar. There is good reason to believe
that most, if not all, of these Miocene large hominoids
were frugivorous (Oreopithecus may have been more
folivorous than the others).
Male chimpanzees differ from the bulk of Miocene
hominoids by their possession of a well-developed
honing action between the upper canine and the lower
third premolar which maintains a sharp distal cutting
edge and pointed apex in the canine. In order for this
action to be effective, the upper canine has not only
become higher crowned and mesio-distally elongated
when compared to those of most Miocene apes, but also
its postero-lingual surface has been hollowed out when
compared to the stubbier, bulkier, shorter canine crowns
of most Miocene hominoids. There were concomitant
changes in the structure of the p/3 in male chimpanzees,
the antero-labial part of the crown becoming larger and
the enamel extending further rootwards than is the case
in most Miocene apes. From a Miocene ape perspective,
therefore, the large, sharp canines of male chimpanzees
are likely to represent a derived morpho-functional
complex, and not a primitive holdover from the ances-
tral condition.
Several lineages of large frugivorous hominoids sur-
vived for millions of years with short stubby canine
crowns, suggesting that the presence of the sharp, labio-
lingually compressed, pointed canines that occur in male
chimpanzees cannot be explained either solely by sexual
selection or by anti-predator defence. Other selective
factors must have been involved, very probably of a
functional nature. Since the primary function of teeth is
related to diet, a dietary explanation is called for.
Most of the animal prey eaten by chimpanzees is
killed by adult males, which usually bite the victim with
the canines, although other methods are sometimes
used, such as flailing the body against the ground, al-
though even in these cases biting with the canines is
usually also employed. Thus, it is likely that the
impressive canine crown height and the fully developed
C-p/3 honing that occurs in chimpanzees is related pri-
marily to the procurement of meat rather than to sexual
selection, anti-predator protection or the killing of
conspecifics, and if so, then the latter may have played a
subsidiary role in the selection of canine morphology.
Meat ingestion and incisor function
It has been said on many occasions that the enlarged
incisor battery of chimpanzees is functionally related to
their frugivorous diet (Pilbeam 1969). However, a mo-
ment’s thought will show that this hypothesis is incom-
plete, because there is good reason to consider that most
of the Miocene apes were frugivores, as are most extant
apes, yet not a single one of them possesses enlarged
spatulate procumbent lower incisors such as those
present in chimpanzees. There is thus no correlation
between frugivory in hominoids and the presence of
enlarged spatulate procumbent lower incisors.
Because chimpanzee upper and lower incisors are
procumbent, when they are in occlusion they meet at an
angle that ensures maintainance of a sharp leading edge
in both upper and lower batteries, even in quite worn
specimens. In other non-human hominoids and homi-
nids, the incisors meet at a more upright angle, with the
result that the lower incisors become blunt with use.
Thus, the occlusal wear of chimpanzee incisors main-
tains a sharp cutting edge, quite different from the
blunter wear produced in most hominoids including
Pongo.
Observations on meat-eating in chimpanzees reveal
that they use their incisors a great deal, especially to rip
and tear skin from the cadaver. They have been ob-
served to use a twisting action in which the skin of the
prey is clamped between the upper and lower incisor
rows while the parts of the prey remaining outside the
mouth are twisted manually until the skin inside the
mouth parts from it (T. Nishida, personal communica-
tion). The presence of sharp apical edges of the lower
and upper incisors undoubtedly increases the efficiency

of this method of processing food, providing a positive
selection pressure for this kind of incisor morphology. In
meat-eating chimpanzees, much of the carnassial func-
tion is performed by the incisors, albeit far less well than
the perfected carnassial function in carnivores.
Meat chewing and molar function
The molars of chimpanzees appear to show some features
that could be related to meat consumption. Most large
bodied Miocene and extant hominoids possess bundont
cheek teeth with large, inflated, voluminous cusps which
restrict the size of the occlusal basins and foveae. Quite a
variety of them also have cingular structures. Gorillas
have higher crowned cheek teeth than Miocene homi-
noids, with prominent crests between the cusps that have
been interpreted in terms of enhanced herbivory in this
genus, probably correctly so, even though gorillas do
consume large quantities of fruit when it is available.
Chimpanzees, in contrast, have retained lower crowned
molars, but the volume of the main cusps is reduced and
they are more peripherally positioned than they are in
most Miocene hominoids. Part of the reduction in cusp
volume seems to have been achieved by a reduction of
enamel thickness, which is usually considered by dental
morphologists to be ‘‘thin’’ (Kono 2003). Chimpanzee
molars also have high dentine penetrance, meaning that
the dentine relief is relatively great beneath the main
cusps. As the crowns wear, they produce dentine lakes at
earlier stages than for example occur in teeth with low
dentine penetrance, such as typify australopithecines,
sivapithecines, Orrorin and Homo.
In combination, all this has two main effects on the
morphology of chimpanzee molars. Firstly, the cusps,
especially those in the lower molars, are somewhat more
medio-laterally compressed (Fig. 18), and are therefore
sharper than they are in most Miocene hominoids, and
they remain sharp into quite old individuals with deeply
worn teeth, and secondly, the occlusal basins and foveae
are expanded, those in the lower molar forming an al-
most uninterrupted basin from mesial to distal. This
‘‘hollowing out’’ and deepening of the occlusal surface,
not only of the lower molars, but also to some extent of
the uppers, has the functional effect of increasing the
relief of the peripheral cusps, making them, especially
the lingual ones, more trenchant. The mesial and distal
crests that descend from the apices of the buccal cusps in
upper molars, and lingual cusps in lower molars (and
also the p/4; Fig. 18), produce a battery of sharp-edged
cusps that enhance the efficiency of the molars for slicing
meat and other foodstuffs. The molars of chimpanzees
are by no means carnassials—the main meat-cutting
function is performed by the incisors—but they do show
some morphological features that can be interpreted in
terms of increasing the efficiency of the meat chewing
process.
In summary, not only do chimpanzee canines show
features that enhance prey procurement, but also their
29
Fig. 18 Oblique view of adult female chimpanzee mandible (Pan
troglodytes) from Mahali, Tanzania, showing the trenchant aspect
of the lingual cusps in the lower molars (and p/4; arrows),
morphology which is here interpreted as being an adaptation to
meat-eating (note also the procumbent incisors)
incisors possess features that increase the efficiency of
meat ingestion, and their molars have morphology that
increases the efficiency of meat chewing. The canines are
frequently used to kill prey, the incisors function as
carnassials, albeit poorly when compared to the carn-
assials of most carnivores, and the molars are employed
in chewing the flesh before it is swallowed.
Meat-eating in chimpanzees
It is known that common chimpanzees from various
parts of Africa hunt, kill and consume a variety of
medium sized mammals (and even birds) on a regular
basis. At Kibale, Uganda, for example, chimps from one
community killed and consumed so many red colobus
that they have caused a serious diminution in the colo-
bus population (Mitani and Watts 2001; Mitani et al.
2002; Watts and Mitani 2002).
Observers at Ngogo, Uganda (Watts and Mitani
2002), recorded 111 hunting episodes by chimpanzees, of
which 80 were successful during which 292 kills oc-
curred. In addition they observed 16 meat-eating and 4
carcase-carrying episodes during the same periods.
Animals killed and eaten during the survey included 258
red colobus (Procolobus pennanti: body weight 11 kg for
males, 6 kg for females), 11 black-and-white colobus
(Colobus guereza), 10 red-tailed guenons (Cercopithecus
ascanius), 7 red duiker (Cephalophus monticola), 2 blue
duiker (Cephalophus callipyga), and one each of bushpig
(Potamochoerus porcus), mangabey (Lophocebus albige-
na), blue monkey (Cercopithecus mitis) and Guinea fowl
(Guttera pulcherani). Given the difficulties of observing
even habituated chimpanzees, these figures are surely an
underestimate of the total number of hunts and kills by
the Ngogo group (Watts and Mitani 2002) especially
30
because some of the species hunted (bushpig, duiker,
red-tailed guenon) are killed on an opportunistic basis
and are seldom observed (recorded as meat-eating or
carcase-carrying episodes, rather than hunts).
Although the total annual meat consumption per
capita is difficult to calculate precisely, for the Ngogo
community it was estimated to be about 15–20 kg for
adult males (range 2–40 kg) and about 10 kg for females
(Watts and Mitani 2002). At Kanywara (also in Kibale
Forest, Uganda) hunting is less frequent that at Ngogo,
but at Mahale (Tanzania) and Taı̈ (Ivory Coast, West
Africa) it is more frequent, but apparently less successful
in terms of numbers of kills per hunt. But even so, with
up to five times as many hunting episodes over similar
observation periods the quantity of meat intake by male
chimpanzees at Taı̈ was greater than at Ngogo (Boesch
and Boesch 2000). At Mahale, daily intake of meat
ranged between 194 gm for alpha males, to 38 gm per
adult non-alpha males to 37 gm for females. This works
out to about 70 kg of meat per year for alpha males and
13.5 kg for other males and females. This compares with
about 110 kg of meat per year for hunter-gatherers in
Africa (Hosaka 2000). The Mahale M group as a whole
is estimated to have obtained 655 kg of meat during
1993.
Hunting frequency by chimpanzees varies seasonally
(Watts and Mitani 2002), but during the 11-month
(1998–1999) study period, there were 59 hunts observed
at Ngogo, 43 of which were successful with multiple kills
per hunt. If allowance is made for hunts that went
unobserved, including opportunistic capture of prey,
then it is reasonable to conclude that at Ngogo, chim-
panzees are hunting about 70 times per year, or on
average 6 times per month or thereabouts. At Taı̈ where
hunting frequency was reported to be five times greater
than at Ngogo, chimpanzees are hunting on average
almost every day of the year. However, because several
hunts can take place in a single day, this does not mean
that Taı̈ chimpanzees are literally hunting every day of
the year. At Ngogo, hunting has been described as
occurring in ‘‘binges’’, and these greater than average
hunting frequencies are interspersed with periods during
which little or no hunting takes place, notably through
periods when there is little ripe fruit to eat. Mitani et al.
(2002) summarise chimpanzee hunting frequency well
when they write that ‘‘Hunts of red colobus (by chim-
panzees) occur frequently, on an average of 4–10 times
per month’’.
Given the importance of hunting and meat-eating
among chimpanzees, it would be surprising if they
showed no dental adaptations to this common behav-
iour. It is the conclusion of this paper that chimpanzees
do indeed possess dentitions adapted for meat-eating, of
which long, permanently honed upper canines in males
predominate in meat procurement (females rarely pro-
curing prey), incisor batteries which are elongated,
procumbent and permanently sharp in both sexes are
largely employed for ingesting meat, and molar rows
with peripheralised, labio-lingually compressed cusps
and voluminous and deep occlusal basins in both sexes
are employed for processing meat once it is in the
mouth. In a functional sense, then, the incisor battery of
chimpanzees can be viewed as being functionally
equivalent to the carnassials in carnivores and creo-
donts, but located at the front of the mouth rather than
inside it.
Pygmy chimpanzees consume appreciably less meat
than common chimpanzees (Badrian and Malenky
1984), and it is thus not surprising to find that even
though their incisors are more spatulate than those of
other hominoids, they are less so than those of common
chimps which do eat greater quantities of meat.
The fact that chimpanzees from West Africa, Uganda
and Tanzania (i.e. the furthest extensions of the species
range) all frequently hunt and eat meat suggests that it is
a widespread propensity of the species, and that it is
therefore most probably an ancient acquisition.
Meat-eating in early hominids?
An interesting observation is that Orrorin, australopi-
thecines and early Homo are similar to Early and
Middle Miocene hominoids in possessing small, narrow,
parallel- to sub-parallel-sided, peg-like, vertically ori-
ented lower incisors, without marked central lingual
ridges or enlarged basal tubercles and lingual enamel
that is almost as thick as that on the labial surface. The
molar morphology of these genera is bunodont, with
restricted occlusal basins, rounded cusps and low den-
tine penetrance, suggesting a crushing and grinding type
of dentition rather than a slicing one, as in chimpan-
zees. From this it is postulated that they were unlikely
to have been eating significant quantities of meat. It
seems probable that hominids only started to include
meat in their diet on a regular basis once they could
prepare it by cutting it into small pieces while it was still
outside the mouth. This extra-oral meat preparation
was possible only after sharp-edged tools had been in-
vented, from which it follows that regular meat-eating
among hominids may have started as late as the Upper
Pliocene. Naturally, if sharp-edged tools are found
earlier that 2.6–3 Mya, then this would push back the
date of the onset of meat consumption by early homi-
nids. Furthermore, meat consumption in extant humans
is overwhelmingly accompanied by cooking. Unlike
chimpanzees, humans do not generally obtain much
nutritional benefit from eating raw meat (Wrangham
2002).
The common ancestor of humans and chimps:
was it chimp-like?
For the past 35 years most molecular biologists have
placed the dichotomy of chimps and humans later than
that of gorillas and humans (Hasegawa and Yano 1984;
Hasegawa et al. 1985; Gagneux et al. 1999; Kumar and

Hedges 1998; Sarich and Wilson 1967; Sibley and Ahl-
quist 1984; Stauffer et al. 2001; Wrangham and Pilbeam
2001; Arnason et al. 1996, 1998, 2000; Janke and Ar-
nason 2002).
Because of the supposed closeness of the molecular
composition of humans and chimps and because many
researchers seem to think of the chimpanzee as a prim-
itive great ape, and of humans as highly derived, there
has been a strong tendency to visualise the common
ancestor of African apes and humans as being chimp-
like. The most widely publicised hypothesis about hu-
man origins for the past 30 years is that humans and
chimps share a common ancestor that in many ways is
thought to have looked either like a bonobo (Zihlmann
1979; Zihlman et al. 1978) or like the common chim-
panzee (Wrangham and Pilbeam 2001). Arguments that
chimpanzees are not primitive in their cranio-dental and
post-cranial anatomy (Latimer et al. 1981) usually fall
on deaf ears.
Yet in terms of incisor-molar relations, chimpanzees
are markedly different from all known Miocene homi-
noids and humans, and they are remarkably divergent
even from Pliocene and Pleistocene australopithecines
and early Homo. This indicates that chimpanzee denti-
tions are highly derived, at least in their incisor-molar
relations, and are not primitive. Indeed, for this partic-
ular feature, it is humans that display the plesiomorphic
condition.
Thus, even if chimpanzees are indeed the extant sister
group of hominids, in their incisor-molar relationships
they do not resemble the last common ancestor at all,
diverging well away from all known hominoids other
than orangutans. It is much more likely that the last
common ancestor possessed incisor-molar relations that
were close to the Miocene hominoid regression line
(Figs. 6, 7, 8). In this respect it is interesting to note that
Orrorin falls on the regression line and is close to the
scatter of early australopithecines, early Homo and
modern humans. In australopithecines, the incisor-mo-
lar relations diverge more and more above the regression
line with the passage of time (the incisors remain almost
the same length, but the molars increase in size greatly),
whereas in chimpanzees, the incisor-molar relations di-
verge markedly to the right of the regression line (inci-
sors became spatulate while the molars remained more
or less the same size, or were slightly reduced in size). It
is interesting to note also that Gigantopithecus plots out
close to robust australopithecines in terms of its incisor-
molar proportions, providing an Asian version of this
evolutionary trend, just as the orangutan provides an
Asian version of the chimpanzee-like incisor-molar
trend. It is probable that these two Asian examples
evolved in parallel to those in Africa, and do not denote
a close phyletic relationship between australopithecines
and gigantopithecines on the one hand, and chimps and
orangutans on the other.
In their incisor-molar relationships, chimpanzees are
highly derived hominoids, and it is therefore probably
erroneous to think of them as closely resembling the last
31
common ancestor, an approach that is all too evident in
many palaeoanthropological papers of the past three
decades, and more. It is much more likely that the last
common ancestor of australopithecines, humans and
chimpanzees was a genus with incisor-molar relations
similar to those of Orrorin and extant humans. The
gorilla is not far from this bauplan in terms of its incisor-
molar relations, but it is a much larger species, and has
derived morphology in other ways (molar morphology
among others).
Conclusions
Examination of the relationships between the incisor
battery and the molar row of hominoids, both fossil and
extant, reveals that there are only two genera with
elongated incisal cutting edges relative to the length of
the molar row (Pan and Pongo). All other hominoids
studied, except for robust australopithecines and Gi-
gantopithecus, have molar rows that are about 1.6 times
the length of the incisive cutting edge. In the austra-
lopithecines and Gigantopithecus the molar row is
greatly enlarged relative to the incisive cutting edge
(molar row is 2.75 times as long as the incisor battery),
the opposite of the case in chimps (the molar row is
almost the same length as the incisor battery). The
‘‘normal’’ (1.6) relationship holds for genera ranging in
age from Early Miocene to Recent, and from Africa,
Europe and Asia, and through a range of body sizes
from Micropithecus to Gorilla. From this it is deduced
that, as concerns the incisor-molar relationship, chimps
and orangs are highly derived by expansion of the inci-
sive cutting edge, while Paranthropus and Gigantopithe-
cus are highly derived by the relatively great expansion
of their molar rows. It is thus an error to consider
chimpanzees as primitive hominoids. In fact, in the
incisor-molar relation, it is humans that are primitive
and chimps derived. From this it is suggested that it is
erroneous to think of the common ancestor of chimps
and humans as ‘‘chimpanzee-like’’, as has been pub-
lished on many occasions.
Meat-eating behaviour is compatible with the pres-
ence of a permanently sharp incisive battery and with
trenchant lingual cusps in the lower molar rows. What
seems clear is that chimpanzees are using their anterior
and posterior dentitions in ways that no other homi-
noids have ever done, whether Miocene or Recent, with
the sole possible exception of Pongo.
Humans in contrast, appear to be comparable to
most other hominoids from the Miocene in terms of
their incisor-molar relations, their thick enameled mo-
lars with bunodont cusps and so on. In fact extant hu-
mans lie slightly to the right side of the regression line
(Figs. 6, 7, 8) but not far enough to be excluded from it
in the way that chimpanzees are. This means that in
these features, fossil and extant species of Homo are
more likely to be plesiomorphic than apomorphic in
their incisor-molar relationships.
32
Robust australopithecines are greatly divergent from
all other hominoids except Gigantopithecus, by their
enlarged molar rows relative to the size of their anterior
dentitions, making them highly apomorphic and thus
unlikely to be ancestral to Homo. Gracile australopi-
thecines, in contrast, are closer to the Miocene homi-
noids in terms of their incisor-molar relations, but some
populations (species) diverge from the regression line
towards the robust australopithecine type, and because
of this they are also unlikely to represent ancestors
of Homo, unless one pleads a reversion of the Paran-
thropus-like dental trends that they exhibit in a nascent
form.
Acknowledgements I thank members of the Kenya Palaeontology
Expedition for their help in the field, in particular Dr. Brigitte
Senut and Mr. Kiptalam Cheboi. Research permission was ac-
corded by the Ministry of Education, Research and Technology,
Kenya. Funds were provided by the Collège de France (Prof. Y.
Coppens), the Laboratoire de Paléontologie (Prof. P. Taquet), the
French Ministry of Foreign Affairs (Commission de Fouilles) and
the CNRS (Projet PICS). I am particularly keen to thank the
Community Museums of Kenya (Mr. E. Gitonga) for their help
and cooperation and Prof. H. Ishida for inviting me to spend time
in his laboratory as visiting professor at Kyoto University. I also
thank the Primate Research Institute, Inuyama, and Dr. T. Nish-
ida, Kyoto University for access to collections in their care.
References
Arnason U, Gullberg A, Janke A, Xiufeng Xu (1996) Pattern and
timing of evolutionary divergences among hominoids based on
analyses of complete mtDNAs. J Mol Evol 43:650–651
Arnason U, Gullberg A, Janke A (1998) Molecular timing of Pri-
mate divergences as estimated by two nonprimate calibration
points. J Mol Evol 47:718–727
Arnason U, Gullberg A, Schweitzer Burguette A, Janke A (2000)
Molecular estimates of Primate divergences and new hypotheses
for Primate dispersals and the origin of modern humans.
Hereditas 133:217–228
Badrian N, Malenky R (1984) Feeding ecology of Pan paniscus in
the Lomako Forest, Zaire. In: Susman RL (ed) The pygmy
chimpanzee. Plenum, New York, pp 275–300
Boesch C, Boesch H (2000) The chimpanzees of the Taı̈ forest:
behavioural ecology and evolution. Oxford University Press,
Oxford
Gagneux P, Wills C, Gerloff U, Tautz D, Morin P, Boesch C,
Fruth B, Hohmann G, Ryder O, Woodruff D (1999) Mito-
chondrial sequences show diverse evolutionary histories of
African hominoids. Proc Natl Acad Sci USA 96:5077–5082
Hasegawa M, Yano T (1984) Phylogeny and classification of
Hominoidea as inferred from DNA sequence data. Proc Jpn
Acad 60:389–392
Hasegawa M, Kishino H, Yano T (1985) Dating of the human-ape
splitting by a molecular clock of mitochondrial DNA. J Mol
Evol 22:160–174
Hosaka K (2000) Mammal hunting and meat-eating of chimpan-
zees. Anthropol Sci 108:72
Janke A, Arnason U (2002) Primate divergence times. In: Galdikas
B, Briggs N, Sheeran L, Shapiro G, Goodall J (eds) All apes
great and small. African apes, vol 1. Kluwer/Plenum, New
York, pp 18–25
Kono RT (2004) Molar enamel thickness and distribution pattern
in extant great apes and humans: new insights based on a 3-
dimensional whole crown perspective. Anthropol Sci 112:121–
146
Kumar S, Hedges S (1998) A molecular timescale for vertebrate
evolution. Nature 392:917–920
Latimer B, White T, Kimbel W, Johanson D, Lovejoy O (1981)
The pygmy chimpanzee is not a living missing link in human
evolution. J Hum Evol 10:475–488
Mitani J, Watts D (2001) Why do chimpanzees hunt and share
meat? Anim Behav 61:915–924
Mitani J, Watts D, Muller M (2002) Recent developments in the
study of wild chimpanzee behavior. Evol Anthropol 11:9–25
Pickford M (1986a) On the origins of body size dimorphism in
primates. Human Evol 1:77–90
Pickford M (1986b) Sexual dimorphism in Proconsul. Human Evol
1:111–148
Pickford M (1986c) Sex differences in higher primates: a summary
statement. In: Pickford M, Chiarelli AB (eds) Sexual dimor-
phism in living and fossil primates. Il Sedicesimo, Firenze, pp
191–199
Pilbeam D (1969) Tertiary Pongidae of East Africa: evolutionary
relationships and taxonomy. Bull Peabody Mus Nat Hist 31:1–
185
Sarich V, Wilson A (1967) Immunological time scale for hominid
evolution. Science 158:1200–1202
Sibley C, Ahlquist J (1984) The phylogeny of the hominoid pri-
mates, as indicated by DNA–DNA hybridization. J Mol Evol
20:2–15
Stauffer RL, Walker A, Ryder O, Lyons-Wailer M, Hedges S
(2001) Human and ape molecular clocks and constraints on
paleontological hypotheses. Am Genet Assoc 92:469–474
Watts D (2004) Intracommunity coalitionary killing of an adult
male chimpanzee at Ngogo, Kibale National Park, Uganda. Int
J Primatol 25(3):507–521
Watts D, Mitani J (2002) Hunting behavior of chimpanzees at
Ngogo, Kibale National Park, Uganda. Int J Primatol 23:1–28
Wrangham R (2002) Fall back foods and cooking in human evo-
lution. In: Evolution of the apes and the origin of human
beings. International Symposium, Centre of Excellence, Inuy-
ama, 14–17 November 2002, p 29
Wrangham R, Pilbeam D (2001) African apes as time machines. In:
Galdikas B, Briggs N, Sheeran L, Shapiro G, Goodall J (eds)
All apes great and small.African apes, vol 1. Kluwer/Plenum,
New York, pp 5–17
Zihlman AL (1979) Pygmy chimpanzee morphology and the
interpretation of early hominids. S Afr J Sci 75:165–167
Zihlman A, Cronin J, Cramer D, Sarich V (1978) Pygmy chim-
panzee as a possible prototype for the common ancestor of
humans, chimpanzees and gorillas. Nature 275:744–746
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.