Table Of ContentDavid R.Begun 1,Erksin Güleç 2 & Denis Geraads 3
1 University of Toronto
2 Dil ve Tarih Cografya Fakültesi,Sihhiye,Ankara
3 CNRS - UPR 2147,Paris
Dispersal patterns of Eurasian hominoids:
implications from Turkey
Begun, D.R., Güleç, E. & Geraads, D., 2003 - Dispersal patterns of Eurasian hominoids: implica-
tions from Turkey - in: Reumer, J.W.F. & Wessels, W. (eds.) - DISTRIBUTIONANDMIGRATIONOF
TERTIARYMAMMALSINEURASIA. AVOLUMEINHONOUROFHANSDEBRUIJN- DEINSEA10: 23-39
[ISSN 0923-9308] Published 1 December 2003
Akey to understanding the complex pattern of hominoid dispersals, origins, and extinction is the
fossil evidence from Turkey. Miocene localities in Turkey are known to contain taxa from across
Eurasia and Africa, and even North America. Three or four separate hominoid clades are identified
in Anatolia, more than in any other place except Kenya. Each of these clades has its closest affini-
ties with hominoids from different continents. Hominoids disperse into Eurasia at the beginning of
MN 5, ahead of the Langhian transgression, probably as a result of the key adaptation of thick
occlusal enamel and robust jaws and teeth, first observed in Afropithecusand Heliopithecusfrom
Kenya and Saudi Arabia. Subtropical and mainly forested conditions encourage rapid dispersal and
diversification of hominoids in Eurasia, while trends toward more open conditions in sub-Saharan
Africa lead to the reduction in diversity and eventual extinction of African pre-modern hominoids.
Toward the end of the Vallesian, Eurasian hominoids undergo a series of extinction events that
begin in the northwestern end of their range (Dryopithecusfrom Western and Central Europe) and
end near the end of the Miocene in Southern China (Lufeng and Hudieliangzi, Yunnan Province).
Of the last surviving Eurasian hominoids, the Yunnan taxa appear to have affinities to Pongowhile
the latest forms from Europe and western Asia are most closely related to the African apes and
humans. Localities in Turkey document the earliest evidence of hominoids in Eurasia, the spread
of hominoids across Eurasia, and, perhaps, the return to Africa in the Turolian of the common
ancestor of the African apes and humans.
Correspondence: David R. Begun1, Department of Anthropology, University of Toronto, Toronto,
M5S 3G3, Canada, [email protected]; Erksin Güleç, Dept of Anthropology, Dil ve Tarih
Cografya Fakültesi, Sihhiye - 06100 Ankara, Turkey, [email protected];
Denis Geraads, CNRS - UPR 2147 - 44 rue de l'Amiral Mouchez - 75014 PARIS, France, dge-
[email protected]. (1To whom correspondence should be addressed)
Keywords: Griphopithecus, Ankarapithecus, Miocene, Hominids, Africa, Asia, Europe.
INTRODUCTION associated with dispersals of other mammal-
At least three and perhaps four clades of ian taxa, circum-Mediterranean changes in
hominoids occur in Turkey between about sea levels during the middle and late
16.5 and 7 Ma. These clades represent sepa- Miocene, and global climate changes toward
rate dispersal events at various times between the end of the Miocene.
Africa, Asia, Europe and Anatolia, and can be The first hominoids in Turkey appear about
23
DISTRIBUTIONANDMIGRATIONOFTERTIARYMAMMALSINEURASIA DEINSEA10,2003
16 to 16.5 Ma, and are among the earliest postcanine teeth. This morphology is associa-
outside of Africa (Begun et al.2002). They ted with robust mandibles and some reduc-
are represented by a large sample of isolated tion in the relative size of the anterior denti-
teeth and a few larger fragments from the tion (incisors and canines). Most researchers
western Anatolian site of Pasalar, near Bursa, assign these samples of hominoids to the
and by a single mandible from the central genus Griphopithecus(Fig. 1) (Pas¸alar and
Anatolian site of Çandır, north of Ankara. Of Çandır) or cf. Griphopithecus (Engelswies)
close to the same age is the hominoid from (Heizmann 1992; Heizmann et al. 1996;
Engelswies in southern Germany (Heizmann Andrews et al. 1996; Güleç & Begun 2002;
& Begun 2001). All of these hominoids share Heizmann & Begun 2001). Recently some
characters of the dentition that distinguish researchers have suggested that a second
them from early Miocene hominoids from genus of hominoid with close affinities to
East Africa, and link them to more modern Kenyapithecus (Fig. 1) from Fort Ternan,
hominoids. The principal synapomorphy of Kenya may be represented in the sample from
middle Miocene Eurasian and more modern Pa¸salar (Ward et al. 1999; Kelley et al. 2000;
hominoids is thick occlusal enamel on the Begun 2000). Most researchers hold that
Figure 1 A phylogeny of some hominoid taxa modified from Begun (2001).Griphopithecusis the outgroup to the Euhominoidea,
the clade that includes all living hominoids and all Eurasian fossil hominoids.The node leading to Kenyapithecus,Hylobates,
Oreopithecusand the Hominidae is unresolved.
24
BEGUN et al.:Eurasian hominoids dispersal
Kenyapithecus is a more modern hominoid locality of Pa¸salar and a great deal has been
than Griphopithecus (Fig. 1), sharing with written about the age and paleoecology of
later Miocene hominoids compressed canines this locality. Less is published about Çandır,
and a higher position of the root of the zygo- although the analysis of this locality is now
matic process of the maxilla (Harrison 1992; complete (Güleç et al. 2002; Begun et al.
Begun 2001; but see McCrossin & Benefit 2002). Results of these analyses suggest a
1997 for an opposing view). If Kenyapithecus slightly different picture from that presented
or a related taxon is also present at Pa¸salar it by many but not all researchers on the
would represent evidence of a second clade Pa¸salar site (e.g., Bernor & Tobien 1990;
of middle Miocene hominoid in Turkey. Andrews 1990; Sen 1990). Here we briefly
The second or third clade of hominoid from review the major conclusions of our research
Turkey is represented by Ankarapithecus at Çandır inso far as it relates to an under-
(Fig. 1) from the Sinap Formation of central standing of the relations of Anatolian middle
Anatolia. Ankarapithecusis known from a Miocene hominoids to other hominoids, and
partial mandible, a male maxilla with a por- the timing of hominoid dispersal events.
tion of the face, and a female partial skull The holotype of Griphopithecus darwini is
(Ozansoy 1965; Alpagut et al.1996; Begun a lower molar from Devínská Nová Ves,
& Güleç 1998). The evolutionary relations of Slovakia (Abel 1902). The long history of
Ankarapithecus are being debated, but the this nomen is reviewed elsewhere (Andrews
most comprehensive analysis to date indica- et al. 1996; Güleç & Begun 2002). The pri-
tes a close relationship to Asian great apes mate mandible from Çandır (Fig. 2) is the
(see below). There is no evidence of a direct holotype of Griphopithecus alpani (Tekkaya
relationship to the middle Miocene homi- 1974). It is indistinguishable from the
noids of Turkey. Slovakian sample of four isolated teeth,
Finally, another hominoid clade is represen- mostly because the latter is inadequate for
ted in Turkey by a newly discovered maxilla detailed comparisons. Most researchers have
from Çorakyerler, in the northern Çankırı- also concluded that the Çandır taxon is the
Çorum basin near the town of Çankırı (Sevim same as the more abundant of the two homi-
et al.2001). While the relations of this taxon noid taxa from Pa¸salar (Alpagut et al. 1990;
remain to be worked out, it has affinities with Ward et al. 1999; Güleç & Begun 2002).
European and African late Miocene to Griphopithecus is characterized by thickly
Pliocene taxa and not to the Asian great apes. enameled molars with broad, rounded cusps,
Each of these fossil hominoids is associated shallow occlusal basins, blunt occlusal crests
with faunas and paleoecological settings that and a shallow topography of the enamal-den-
are distinct from the others, and each proba- tine junction. The Çandır specimen preserves
bly represents a separate migration rather most of the mandibular corpus, which is low
than in situ evolution of Anatolian homi- and robust in transverse dimensions, both of
noids. For the remainder of this paper we will which are functionally consistent with thickly
describe in more detail evidence for the evo- enameled molars. It also preserves the premo-
lutionary relations of Anatolian hominoids lars, which have a simple occlusal morpholo-
and their implications for understanding bro- gy, and the alveoli for the canines and incis-
ader issues of migration and dispersal of ors, which must have been quite small. The
hominoids across Eurasia and to and from symphysis is strongly reinforced by superior
Africa. and inferior transverse tori (Güleç & Begun
2002). The sample of mostly isolated teeth
GRIPHOPITHECUS with a few gnathic fragments from Pa¸salar is
The earliest hominoids from Anatolia are broadly consistent with this morphology, and
known primarily from the middle Miocene together these samples indicate that the first
25
DISTRIBUTIONANDMIGRATIONOFTERTIARYMAMMALSINEURASIA DEINSEA10,2003
hominoid from Anatolia was a hard object hominoids, which range from France east to
feeder with dental characters most similar to India and Pakistan and south to Namibia, and
middle and late Miocene Sivapithecusfrom most Çandır and Pasalar taxa, at least at the
South Asia, Ouranopithecusfrom Greece, genus level, are found in all three Old World
and Pliocene hominids from East and South continents in the middle part of the Miocene.
Africa (Australopithecusand early Homo). Many migration routes between South and
This is probably the primitive morphological East Asia, Europe and Africa appear to have
complex for hominids (great apes and passed through Anatolia, contributing to the
humans) and represents the earliest appearen- unique character of this assemblage compa-
ce of these characters. Given its primtive red to faunas "typical" of specific bioge-
nature, the relations of Griphopithecusare ographic provinces (Geraads et al.2002).
unclear, except that it is derived relative to Reconstructing the timing of the original dis-
early Miocene hominoids such as Proconsul persal of hominoids into Eurasia, and its cor-
and primitive relative to most middle and late relation to other events, such as climate and
Miocene and recent great apes (Andrews et sea level changes and the biochronology of
al. 1996; Güleç & Begun 2002). In Begun other taxa, depends on the assessment of the
(2002) Griphopithecusis placed in its own age of Çandır and Pasalar.
family (Griphopithecidae). Andrews et al. Based in large part on their preliminary
(1996) classification of Griphopithecusand analysis of the Pa¸salar fauna, Bernor &
other middle-late Miocene hominoids is para- Tobien (1990) concluded that Pa¸salar and
phyletic by the authors’admission and is thus Çandır belong in MN 6, with Pa¸salar slightly
not considered further here. older than Çandır. However, most of the large
Central Anatolia is near the geographic mammals from Çandır and Pa¸salar that also
center of the distribution of middle Miocene occur in Europe first occur there by MN5 or
before (Begun et al.2002). Of the large
mammals that have an FAD in MN 6, nearly
all are known only from Turkey (mainly
Pa¸salar and Çandır). Griphopithecus alpani
and Orycteropus senihave congeners or close
relatives from pre-MN 6 localities in
Germany and East Africa (Güleç & Begun
2002; Van der Made 2002), and the rhinos
from Çandır appear to be closely related to
MN 4-5 taxa (Aceratheriumto Plesioacera-
theriumfrom MN 4-MN 5 and Beliajevinato
Hispanotherium, which disappears in late
MN5 or early MN6 in Europe, (Geraads &
Saraç 2002). Bunolistriodonbecomes extinct
before MN6 in Western Europe. Turcocerus,
which is given an MN 6 FAD based on its
occurrence in Anatolia, is not known from
Europe at all, while Giraffokeryxand
Hypsodontus, both un-known from Western
Europe but with records in Eastern Europe,
have MN 5 distributions. Heteroproxis
known from MN 5-MN 6 in Western Europe.
Figure 2 Lingual (a) and occlusal (b) views of the Çandır Only three large mammal taxa from Pa¸salar
mandible.Note the massive corpus,doubled symphyseal tori
or Çandır are considered to be MN 6 taxa as
and broad,rounded molar cusps.
26
BEGUN et al.:Eurasian hominoids dispersal
determined by their presence in localities assign to a species.
other than Pa¸salar or Çandır. These are In sum, few taxa from Çandır and Pa¸salar
Percrocuta miocenica, Listriodon splendens have a FAD in MN 6 as determined by their
and Anchitherium aurelianese hippoides. presence outside Turkey (3 of 25 large mam-
Each of these taxa is represented at Çandır by mals and 2 of 19 small mammals), and most
small numbers of isolated teeth (and some have a FAD in MN 5 or earlier (Begun et al.
postcrania, in the case of Anchitherium, 2002). Given the extinction of Keramidomys
which however differ from those of the same thaleri and Megacricetodon collongensis in
subspecies at Sansan) (Güleç & Geraads MN 5 and that all other taxa known outside
2002; van der Made 2002; Nagel 2002). Turkey originate in MN 5 or earlier, it is like-
Among the micromammals from Çandır ly that the few apparently MN 6 taxa from
only the flying squirrels have a FAD from Çandır and Pasalar actually originated in MN
Europe later than MN 5. Forsythiais said to 5 or earlier in Turkey and do not appear in
have an MN 7/8 distribution in Europe (De Europe until MN 6. We conclude therefore
Bruijn et al.2002; De Bruijn 1999) but has that Çandır and Pasalar belong to MN 5,
been recently reported in MN 6 at Sansan which we believe to be between 15 and 17.3
(Mein 1999). The generally primitive nature Ma in Turkey and central Europe (Steininger
of the micromammal fauna from Çandır and 1999; Begun et al.2002; see an extensive
Pa¸salar however suggests that those taxa may discussion in the latter on the issue of the age
have come to Europe from Western Asia. of MN 5). Based on faunal similarities to
Cricetodon candirensisis only known from localities from central Europe and somewhat
Çandır, and De Bruijn et al.(1993) and De ambiguous paleomagnetic results from Çandır
Bruijn & Ünay (1996) suggest that the (Krijgsman 2002) we place this locality at
Cricetodonfrom MN 4 in Turkey is ancestral about 16.2 Ma, and Pa¸salar slightly older, fol-
to that from Spain, and that a migration event lowing Bernor and Tobien (1990). With
took place in MN 5. Of the two widespread regard to other hominoid localities, Çandır
and biostratigraphically important cricetini, and Pa¸salar are roughly contemporaneous
Megacricetodon collongensisis extinct by with Engelswies and older than Klein
MN 6, and Democricetodonaff. D. gaillardi Hadersdorf and Devínská Nová Ves (Fig. 3)
from Çandır is thought to be primitive relati- (Begun et al.2002; Heizmann & Begun
ve to the sample from Sansan (De Bruijn et 2001).
al., 2002). Desmanodon is extinct in Europe The Gomphotherium Landbridge came into
before MN 6 and Dinosorexhas a range from being at about 18.5 Ma and appears to have
MN 4 to MN 9 in Europe (Ziegler 1999). been crossed in two phases preceding the
Keramidomys thaleriis extinct by MN 6, and Langhian Transgression that temporarily
in fact is primarily known from MN 5 (De separated Eurasia and Africa (Rögl 1999).
Bruijn et al.2002 and pers. comm., Engesser Mastodons appear to cross into Eurasia from
1999). Eomyops from Çandır is identified as Africa first, followed by Deinotherium. It
E. cf. E. catalaunicus(MN 9), but only 2 may be that primates accompanied the latter
teeth are known from Çandır and the genus is into Eurasia at about 17 to 16.5 Ma. This
notoriously slow to evolve (De Bruijn, per- would correspond roughly to the base (low
sonal communication). Eomyops is thought to stand) of TB 2.2 (Fig. 3) of the sea level
first occur in Europe in MN5, though cycles of Haq et al.(1988) (Woodburne &
Engesser (1999) suggests that it may have Swisher 1995). At this time many of the
been present as early as MN 2/3. Engesser mammals known from Çandır Pa¸salar crossed
(1999) also stresses the slow pace of evolu- into Eurasia from Africa (Hominoidea,
tion in the molars of Eomyops, suggesting Deinotherium, Bunolistriodon, Chalico-
that the two teeth from Çandır are difficult to theriinae, Brachypotherium, Orycteropus), or
27
DISTRIBUTIONANDMIGRATIONOFTERTIARYMAMMALSINEURASIA DEINSEA10,2003
into Africa from Eurasia (hypodontines, Protictitherium, Percrocuta, Pseudaelurus,
Amphicyon, Hemicyon) (Fig. 4). Starting Ischrictis, eomyids, sciurids, and glirids). A
also at this time and continuing into the mid- number of these taxa also appear to disperse
dle Miocene taxa present at Çandır moved into North America at this time as well
between Western Asia and other parts of (petauristids, eomyids and Pseudaelurus
Eurasia (Democricetodon, Megacricetodon, (Woodburne & Swisher 1995). All these path-
Cricetodon, Tamias, Pliospalax, Alloptox, ways potentially take land mammals through
Beliajevina, Tethytragus, Turcocerus, Central Anatolia.
Schizochoerus, petauristids, hominoids) or This scenario of dispersal events differs
from Europe to Western Asia (Palaeomeryx, somewhat from others that have been recent-
Heteroprox, Hoploaceratherium, ly proposed, but mainly in combining impor-
Figure 3 Chronology of important middle Miocene hominoid localities in Eurasia.Chron (left column),and Mediterranean and
Central Paratethys ages (right column) correlations from Steininger (1999) and references therein (see also Begun et al.(2002)
and Heizmann & Begun (2001)).
28
Begun et al.:Eurasian hominoids dispersal
Figure 4 Paleogeography and hominoid migration routes in pre-Langhian MN 5. Map modified from Rögl (1999).Tailless catar-
rhine symbols represent localities in Anatolia (Pasalar and Çandır) and Germany (Engelswies).Squares represent localities in
Kenya (Kalodirr) and Saudi Arabia (Ad Dabtiyah) from which early Miocene hominoids with thickly enameled molars are known.
tant dispersal events (late Burdigalian and scenario well.
early Serravallian events of Rögl 1999, or TB Hominoids are rare in late early or early
2.2-2.5 cycle events of Van der Made 1999). middle Miocene faunas (Pa¸salar being a nota-
This is a consequence of the earlier age we ble exception), and may have dispersed from
attribute to the Çandır fauna, but it is also Africa toward the end of this period of inter-
consistent with the data on dispersals continental migration events, in early to mid
between the Old World and North America, MN 5. They appear at about the same time in
which are concentrated at this time in the ear- Turkey (Çandır and Pa¸salar) and Germany
lier parts of the TB 2 supercycle (Woodburne (Engelwies) at 16 to 16.5 Ma (Fig. 3), and
& Swisher 1995). It is also consistent with somewhat later in East Africa (Maboko,
the model proposed by Pickford & Morales Kipsarimon, Nachola), and the Vienna Basin
(1994), who correlate faunal exchanges (Devinská Nová Ves [Neudorf] Sandberg and
between African and Eurasia and increases in Klein Hadersdorf) (Fig. 3) (Matsuda et al.
generic diversity, to shifts in the northern 1986; Pickford 1986; Heizmann 1992;
boundary of the tropical biogeographic realm. Heizmann & Begun 2001; Sawada et al.
Their early peak at about 17 to 18 Ma fits our 1998; Ward et al. 1999; Rögl 1999; Begun et
29
DISTRIBUTIONANDMIGRATIONOFTERTIARYMAMMALSINEURASIA DEINSEA10,2003
Figure 5 Paleogeography and hominoid migration routes in post-Langhian MN 5-6. Map modified from Rögl (1999).Tailless
catarrhine symbols represent localities in Anatolia (Pasalar and Çandır) and Slovakia (Devínská Nová Ves).The ape grasping a
branch represents localities in the Siwalik Hills.The precise pattern of dispersal among these localities is not completely clear,
though the Anatolian and German sites are almost certainly older than the Slovakian one.Siwalik hominids may originate from a
Mediterranean source or from an African one.
al. 2002). It is possible that the first homi- wise exclusively Eurasian Pliopithecoidea
noid of modern dental aspect, Gripho- (Begun 2002b). Both primates from Fort
pithecus, evolved in Eurasia and that the pre- Ternan (cf. Limnopithecusand Kenya-
sence of dentally very similar taxa in East pithecus) could be immigrants from Eurasia,
Africa 1 Ma or more later represents an early of which there appear to have been several.
Serravallian return of hominoids to Africa, Along with other Western Asian taxa present
contemporaneous with the spread of at Çandır (Tethytragus, Turcocerus,
Griphopithecus to the Vienna Basin (Begun Schizochoerus, Pliospalax, Cricetodon,
2000, 2001) (Fig. 5). This migration would Cricetini, and possibly the flying squirrels)
correspond to the TB 2.5 lowstand (Fig. 3). hominoids radiate from the northern and eas-
Interesting in this regard also is the fact that tern Mediterranean and diversify into a num-
another middle Miocene primate, represented ber of clades from China to Spain (Begun
by one specimen from Fort Ternan, KNM-FT 2002a).
20, attributed to cf. Limnopithecus legetet Evidence from Anatolia, central Europe and
(Andrews 1978) has affinities with the other- East Africa suggests that hominoids of
30
BEGUN et al.:Eurasian hominoids dispersal
modern aspect originate in Eurasia from a is much larger than Griphopithecus. In addi-
more primitive African form about 17 Ma. tion the mandible is taller and less robust
Several early Miocene hominoids from this than in Griphopithecus alpaniand the sym-
time, such as Afropithecusfrom northern physeal morphology is very different (Begun
Kenya and Heliopithecusfrom Saudi Arabia, & Güleç 1998). Ankarapithecusmolars lack
share features with the earliest Eurasian cingula that sometimes appear on Gripho-
hominoids, particularly the presence of thick pithecusmolars, and are very similar in
occlusal enamel, that may have provided one morphology and dental proportions to
of them with the key adaptation necessary to Sivapithecusand Ouranopithecus, two taxa
exploit the more northerly subtropical envi- with which Ankarapithecushas been confu-
ronments. Between the late Burdigalian sed in the past (Andrews & Tekkaya 1980;
(Karpatian) dispersals and the late Miocene Martin & Andrews 1984). Like both of these
fewer migration events between Eurasia and taxa, it is likely that Ankarapithecuswas also
Africa are recorded although several lineages a hard object feeder.
of rodents move from Asia to Africa, as does Unlike Griphopithecus, the face of Ankara-
Percrocutaand possibly Kenyapithecus(Tong pithecusis well preserved and reveals a num-
& Jaeger 1993, Van der Made 1999, Begun ber of unambiguous synapomorphies both of
2000 and above). It is during this time that the Hominidae and the Ponginae (the Asian
the hominids (great apes) evolve in situin great ape clade including Sivapithecusand
Eurasia. Pongo) (Fig. 1). The maxilla is massive with
roots of the zygomatic processes placed high
ANKARAPITHECUS on the alveolar process, and a broad nasal
The earliest hominid from Turkey is Ankara- aperture at its base, both being hominid char-
pithecusfrom deposits at Sinap Tepe, acters (Fig. 6). The premaxilla is robust and
Kayıncak, and Delikayıncak Tepe, near elongated, as in hominids, but lacks the high-
Yassıören, Central Anatolia. Unlike the mid- ly derived morphology of Sivapithecus.
dle Miocene localities, the Yassıören locali- Details of the nasal fossa also distinguish
ties are well dated biostratigraphically and by Ankarapithecusfrom Pongo and Sivapithecus
paleomagnetic correlation to about 10 Ma (Begun & Güleç 1998). Like Ouranopithecus
(Sen 1991; Alpagut et al.1996; Pekka the premaxilla is biconvex and less horizon-
Lunkka et al.1999). They are thus roughly tally oriented that in Sivapithecus(Fig. 6).
contemporaneous with Dryopithecusfrom the Like Sivapithecusthe zygoma are broad and
Vallesian of central Europe and Spain and laterally flared, and face anteriorly. The mid-
Sivapithecusfrom the Potwar Plateau face is prognathic and not flat as in
(Pilbeam et al.1979; Kordos 1982, 1991; Sivapithecusand Pongo, but very tall, as in
Raza et al.1983; Barry et al.1985; the Asian taxa. The orbits are squared, not
Kappelman et al.1991; Agustí et al.1996). elongated as in Sivapithecus, and the interor-
Ankarapithecus meteaiis the only species of bital space is relatively narrow, again, like
the genus, and is known cranially from three Asian great apes (Alpagut et al.1996; Begun
specimens, but two of them are relatively & Güleç 1998). The nasal bones are extreme-
complete. Anumber of postcrania have also ly elongated, as in Sivapithecus. Supraciliary
been reported but they are unpublished, other arches contour the superior edges of the
than a brief reference to their relative robusti- orbits, but they do not meet in the midline to
city and indications of some degree of adap- form a true torus. They are robust but closely
tation to terrestrial locomotion (Köhler et al. resemble those of large Pongo, and to a lesser
1999). Like Griphopithecus, Ankarapithecus extent Sivapithecusand large Cebus (Alpagut
is thickly enameled with low, rounded cusps, et al.1996; Begun & Güleç 1998). These
blunt crests and shallow basins (Fig. 6), but it arches have been interpreted as supraorbital
31
DISTRIBUTIONANDMIGRATIONOFTERTIARYMAMMALSINEURASIA DEINSEA10,2003
Figure 6 (a): Frontal view of the palate of a male of Ankarapithecusmeteai.Note the broad and widely flared zygoma.
(b): Lateral view.Note the midfacial prognathism,with the nasal aperture margin projected anteriorly relative to the zygomatic.
(c): Occlusal view.
32
Description:2002). They are represented by a large sample of isolated teeth and a few larger fragments from the western Anatolian site of Pasalar, near Bursa, Ward, S., Anwar, M., Barry, J. C., Brown, B., Hake,. P., Johnson, N. M., Raza, S. M., & Shah, S. M. I.,. 1991 - The earliest occurrence of Sivapithecus