sivapithecus is east and dryopithecus is west, and never

53© 2004 The Anthropological Society of Nippon

ANTHROPOLOGICAL SCIENCE

Vol. 113, 53–64, 2005

Sivapithecus is east and Dryopithecus is west, and never the twain shall meet

DAVID R. BEGUN1*

1Department of Anthropology, University of Toronto, Toronto, Ontario, M5S 3G3, Canada

Received 26 June 2003; accepted 7 March 2004

Abstract Sivapithecus and Dryopithecus are well-described Miocene hominids (great apes andhumans), both known since the 19th century. Over the years these genera have been combined into one(Dryopithecus) or separated up to the subfamily level. Each have been dismissed as interesting sidebranches, hailed as direct ancestors, or recognized as sister clades to one or more clade of extant hom-inid. Here I argue that they are each stem taxa of the two living hominid clades Ponginae and Homi-ninae. A famous poem by Rudyard Kipling tells the tale of a British and Afghan soldier whosedifferences (in ethnicity) obscure their similarities (in character). The relationship between Sivapithe-cus and Dryopithecus is similar. On the one hand, Sivapithecus is restricted to South Asia, has thicklyenameled molars, robust jaws, and superficially baboon-like forelimbs; Dryopithecus is European, hasthinly enameled molars and gracile jaws, with suspensory forelimbs. On the other hand, both are greatapes, both had suspensory adaptations, large brains, and delayed development, and both are closelyrelated to living hominids. Recognition of the likely relations of Sivapithecus and Dryopithecus pro-vides insight into the causes, timing, and paleobiogeography of crown hominid origins.

Key words: hominid origins, Asia, Europe, climate change

Introduction

“Sivapithecus is east and Dryopithecus is west, and neverthe twain shall meet, Till climates change, forests shrink,and hominids retreat. Alas, today they are neither east norwest, on the continents of their birth, Since hominids movedSouth of Cancer from opposite ends of the earth” (shame-lessly modified from Rudyard Kipling’s “The Ballad of Eastand West”, The One Volume Kipling Authorized, Double-day, Doran & Company, Garden City, New York, 1928).

Dryopithecus was first described in 1856 from fossilsfound in the foothills of the French Pyrenees, three yearsbefore Darwin’s publication of “On the Origin of Species”(Lartet, 1856). Lartet and most subsequent workers in the19th century recognized the great ape affinities of this fossiltaxon, and a number noted the resemblances of Dryopithe-cus to African apes (e.g. Gaudry, 1890). Sivapithecus wasfirst definitively described by Lydekker (1879), from fossilsfrom the Potwar Plateau of present day Pakistan, though heoriginally named the taxon Paleopithecus [noting that Pale-opithecus is preoccupied, Pilgrim (1910) introduced thenomen Sivapithecus]. Again, a number of researchers,including Lydekker, noted particular similarities to orangu-tans. Kelley (2002) and Begun (2002) review the history ofthese and other Eurasian hominoid discoveries. To make along story short, once hominoids were recovered in Africa

[reviewed recently by Harrison (2002) and Ward and Duren(2002)] it seemed less likely to many researchers that Eur-asian fossil hominoids could be directly related to livingtaxa, though many continued to believe that Sivapithecus (orRamapithecus) in particular had a specific relationship tohumans (see Appendix for a classification of the Miocenehominoids discussed here).

In the modern era most researchers now agree that Dryo-pithecus and Sivapithecus are great apes, but there is dis-agreement on their relations to living taxa. Most researchersconclude that Sivapithecus is a sister clade to Pongo. A fewresearchers, focusing primarily on differences betweenPongo and Sivapithecus in postcranial anatomy, but also inpart on gnathic differences, have suggested that the fossiltaxon is either a stem great ape or a stem hominoid, withouta direct relationship to Pongo (e.g. Benefit and McCrossin,1995; Pilbeam, 1997). In my view the case for a Pongo–Sivapithecus clade, reviewed briefly below, is considerablystronger. The relations of Dryopithecus are even less wellagreed upon. A number of researchers also feel that Dryop-ithecus is a stem hominid (Andrews, 1992) or hominoid(Benefit and McCrossin, 1995; Pilbeam, 1997). Others seeDryopithecus as a member of a very diverse Pongo cladethat includes most Eurasian hominids, including Ouranop-ithecus and Oreopithecus (Moyà-Solà and Köhler, 1993).Finally, some researchers view Dryopithecus as a member ofthe African ape and human clade, along with Ouranopithe-cus, though these researchers disagree on the precise rela-tions among African apes and humans (e.g. de Bonis andKoufos, 1993, 1997; Koufos, 1995; Begun et al., 1997;Begun, 2001, 2002). I will argue here that most of the mor-

* Corresponding author. e-mail: [email protected]: �1-416-978-8850; fax: �1-416-978-3217

Published online 30 September 2004in J-STAGE (www.jstage.jst.go.jp) DOI: 10.1537/ase.04S008

54 D.R. BEGUN ANTHROPOLOGICAL SCIENCE

phological evidence of Dryopithecus and Ouranopithecuspoints to a sister clade relationship to African apes andhumans.

Both hypotheses of specific phylogenetic links betweenAsian great apes on the one hand and Afro-European greatapes on the other are also consistent with proposed phyloge-netic relations and paleobiogeographic patterns among otherMiocene to recent Eurasian and African mammals. They arealso consistent with most estimates of divergence dates frommolecular data and suggested patterns of dispersal caused bychanging ecological settings. The combined evidence ofmorphology, functional anatomy, phylogeny (morphologyand molecular), climate change, and paleobiogeographyprovide the basis for a model of modern hominid origins.

Eurasian Hominid Origins

Current evidence suggests that Eurasian hominids andmiddle Miocene African hominoids derive from one or moremembers of a group (probably paraphyletic) loosely referredto here as the ‘griphopiths’. This includes Griphopithecusfrom Europe and Turkey, and Equatorius (or Griphopithe-cus), Nacholapithecus and Kenyapithecus from Kenya. Thecf. Griphopithecus from Engelswies, Germany and Gripho-pithecus alpani from Paşalar and Çandır, Turkey are appar-ently the oldest ‘griphopiths’, though some uncertaintypersists with regard to the age of the Turkish samples. Ger-man and Turkish Griphopithecus is currently estimated to be~16–16.5 Ma (Heizmann and Begun, 2001; Begun et al.,2003). African and Slovakian ‘griphopiths’ are younger, at~15 Ma (Steininger, 1999; Ward et al., 1999). African‘griphopiths’ may have evolved from Eurasian Griphopithe-cus, though the fossil record of middle Miocene hominoidsin Africa is very poor and the possibility that in situ ances-tors existed cannot be dismissed. ‘Griphopiths’ from allthree continents are united by a suite of derived charactersthat distinguish them from early Miocene hominoids such asProconsul, including large, thickly enameled molars withlow cusps and restricted dentine penetrance, some reductionin cingulum development and frequency, robust jaws, andcertain modifications to the upper incisors. They are distin-guished from Afropithecus in lacking the autapomorphicfeatures of the anterior dentition, premolars, palate, andmandible (Leakey and Walker, 1997). One ‘griphopith’,Nacholapithecus, more closely resembles Afropithecus inhaving a relatively long and more horizontal nasoalveolarclivus and robust, relatively low-crowned upper canines, butlacks most of the dental features and the deep mandibles ofAfropithecus (Ishida et al., 2004). It may be in the end thatNacholapithecus is more closely related to Afropithecus andis not a ‘griphopith’.

Very little is known of the skull of the ‘griphopiths’beyond the gnathic region, but postcranially they are as agroup broadly similar to early Miocene hominoids (Begun,1992, 2002; Benefit and McCrossin, 1995; McCrossin andBenefit, 1997; Nakatsukasa et al., 1998; Ward and Duren,2002). While all of the later-occurring taxa from East Africahave individual postcranial autapomorphies suggestive ofspecialized adaptations [e.g. elongated, robust forelimbs inNacholapithecus; indications of terrestriality in Equatorius

(McCrossin et al., 1998; Ishida et al., 2004)], none show anyindication of any type of below-branch quadrupedalism as inextant and late Miocene hominoids.

The principle distinguishing feature of the ‘griphopiths’ istheir robust gnathic morphology, and this is probably relatedto a novel adaptation to exploit a wider range of food typesin the slightly more seasonal environments of the middleMiocene of Eurasia (Heizmann and Begun, 2001). Ampleevidence exists to indicate that the climate changes thatwould lead eventually to the disappearance of tropical andsubtropical ecological settings in most of Eurasia werealready under way at the beginning of the middle Miocene(see below).

Eurasian Hominid Diversity

It is intriguing that the ‘griphopiths’ have a European,Asian, and African distribution, and it is tempting to con-clude that this represents the initial separation of both Eur-asian great ape clades and the African ape clade (Figure 1).However there is no evidence for this. Griphopithecusalpani does not share any derived character with any lateMiocene hominid, despite tentative suggestions to the con-trary (e.g. Alpagut et al., 1990). Similarly, Equatorius, Ken-yapithecus, and Nacholapithecus do not share specificsynapomorphies with great apes, again despite a few claimsto the contrary (e.g. McCrossin et al., 1998; Ishida et al.,2004). While the nasoalveolar clivus of Nacholapithecus hasbeen interpreted as elongated and overlapping with the max-illary palatine process, resulting as well in some reduction ofthe incisive foramina, other similarities to great apes arelacking. These premaxillary similarities are more parsimoni-ously interpreted as homoplasies possibly related to theenlarged anterior dentition of Nacholapithecus, apparentlyalso shared by Afropithecus (Ishida et al., 2004). All of thesetaxa are best viewed as broadly ancestral to all hominids.

Because the ‘griphopiths’ are distributed widely both geo-graphically and temporally, it is not clear exactly when andwhere the two main clades of living hominids evolved (Fig-ure 1). However, by ~12.5 Ma both are established in theirrespective regions, Sivapithecus in Chinji Formation depos-its in the Potwar plateau and Dryopithecus from St. Gaudensin France and St. Stefan in Austria (Kappelman et al, 1991;Steininger, 1999). All hominids from the Potwar plateau arecurrently attributed to Sivapithecus (Kelley, 2002), thoughthe earliest specimens from Chinji are much more poorlypreserved than at later Potwar localities, from which the bulkof Sivapithecus is known. This is interesting becauseAnkarapithecus from Anatolia, of equivalent age to much ofthe Sivapithecus sample (from the Nagri Formation, whichis ~10 Ma and occurs in the middle of the Potwar hominidsequence) is related to Sivapithecus but lacks key derivedcharacters and is thus excluded from the Sivapithecus–Pongo clade (Alpagut et al., 1996; Begun and Güleç, 1998;Kappelman et al., 2003). It is possible that the Sivapithecus–Ankarapithecus–Pongo clade evolved somewhere in theeastern Mediterranean–South Asian area and dispersedwithin the region into a number of taxa. If Chinji (the oldest)hominids are the ancestors of this radiation, then they wouldhave to be a different genus from Sivapithecus and Ankarap-

ASIAN AND AFRICAN GREAT APES DIVERGED IN EURASIA 55Vol. 113, 2005

ithecus, its putative descendents (Begun and Güleç, 1998).Only more complete fossils from Chinji age deposits cananswer this problem.

The origins of Dryopithecus are similarly murky. The old-est specimens of Dryopithecus are roughly the same age asSivapithecus, though this is based on less reliable evidencefrom relatively small faunal samples. Middle Miocene Dry-

opithecus is known from Spain (Can Vila, Can Mata, Castelde Barbera), France (St. Gaudens, La Grive), and Austria(St. Stefan), all of which are dated mainly on the basis of theabsence of typical late Miocene taxa [true mice (murids) andthree-toed horses (hipparionines)]. These localities are usu-ally attributed to the European land mammal biochronologiczone known as MN 7/8, which is correlated to localitiesdated between about 14 and 11.5 Ma (Steininger, 1999).Thus they are roughly contemporaneous with earliest Sivap-ithecus, though they could be a bit older or younger. As withSivapithecus, Dryopithecus radiates in situ into at least fourspecies distributed from Spain in the west to Hungary andpossibly Georgia in the east [the possibility of Dryopithecusin East Asia exists but is currently unsubstantiated (Kelley,2002)]. In the late Miocene, hominids diversify at thegeneric level in both Europe and Asia. Between about 10and 7 Ma, Sivapithecus radiates into a diversity of taxa inAsia and Dryopithecus does the same in Europe. Ouranop-ithecus appears in Greece and possibly Anatolia betweenabout 9.5 and 7–8 Ma, possibly as a derived member of theDryopithecus clade (see below). Oreopithecus may be a partof the same radiation (Harrison and Rook, 1997), though inmy view this late Miocene taxon is a distinct clade (Begun,2002). Sivapithecus apparently shares an ancestry withGigantopithecus and Lufengpithecus. Gigantopithecus is aproblematic taxon because it is known only from lower jawsand isolated teeth, the large size of which make them looksimilar. It is quite possible that this includes more than onegenus, one represented by the Miocene specimens (Giganto-pithecus giganteus) and the other by Pleistocene fossils(Gigantopithecus blacki). Once again, more and better pre-served specimens are needed. Lufengpithecus is representedby large samples from Yunnan Province, China, that provideevidence of a distinctive cranial morphology in combinationwith dental and postcranial similarities to Pongo (Schwartz,1997; Kelley, 2002; personal observations). Because Sivap-ithecus more closely resembles Pongo in facial morphologywhile Lufengpithecus more closely resembles Pongo in den-tal morphology, the relations among these taxa are not com-pletely clear, though most researchers agree that all aremembers of the same clade.

Eurasian Hominid Phylogeny

EuropeAs noted earlier, relations among European middle and

late Miocene hominids and between them and living homi-nids are currently debated. Rather than reviewing this litera-ture here [it is reviewed in Begun (1994, 2001, 2002), Begunand Kordos (1997), and Kordos and Begun (2001b)] I willsimply review the evidence that supports the conclusions mycolleagues and I have reached concerning European homi-nids.

In an analysis of 240 characters in fossil and living homi-noids (Begun et al., 1997), more recently expanded to 247characters (Begun, 2001), we found that Dryopithecus aloneor Dryopithecus and Ouranopithecus together are sisterclades to the African apes and humans. The numerous char-acters that represent synapomorphies of the greatapes � Dryopithecus and African apes � Dryopithecus

Figure 1. Paleobiogeography of late early to early middle Miocenehominoids. (A) In the Karpatian, hominoids enter western Eurasiabefore the Langhian transgression temporarily stops land mammalmigrations between Eurasia and Africa. The source may have beenEast Africa (Afropithecus) or Saudi Arabia (Heliopithecus). The earlierhominids (‘griphopiths’) first appear in Europe east of the Rhine gra-ben (dotted line) and in Anatolia. (B) After the Langhian (Badenian)‘griphopiths’ spread and have thus far been recovered in Slovakia andEast Africa. Sometime toward the end of this initial dispersal periodthe clades of living hominids emerged. Pongines may have evolvedfrom western Eurasian ancestors while hominines may have evolvedin situ in the west. In this and in Figure 3 the shaded areas representexposed land surfaces as reconstructed by Rögl (1999).


clades are listed in Table 1. Using a smaller number of char-acters from a restricted anatomical region (20 features of theface), three features have been used to suggest that Dryop-ithecus is more closely related to Pongo (Moyà-Solà andKöhler, 1995). The problem here, beyond their very smallnumber, is that these characters are actually variable in Dry-opithecus (Hungarian Dryopithecus differs in all of thesefeatures from Spanish Dryopithecus), or they are based on areconstruction that can be interpreted differently (Begun andKordos, 1997; Kordos and Begun, 2001b). So they do notcharacterize most known Dryopithecus crania. Furthermore,all of these characters are known to be variable, and thus ofunpredictable phylogenetic significance in hominids gener-ally (Eckhardt and Eckhardt, 1995; Msuya and Harrison,1996).

Among the most important apparent synapomorhies of theAfro-European great ape clade are the morphology of thetemporal bone, the premaxilla and the basic architecture ofthe cranium. Although the temporal bone of Ouranopithecusis not known, in Dryopithecus the tympanic and articularportions are fused, the entoglenoid process is strongly devel-oped, and the glenoid fossa is relatively deep. These aredetailed developmental and morphological features sharedwith African apes and humans in a region that has beenfound to have a strong phylogenetic signal in hominids(Brown and Ward, 1988). A new specimen of Dryopithecusfrom Rudabánya, Hungary shows clearly for the first timethat the face of this taxon is klinorhynch, or ventrallyrotated, as in African apes and humans. While it is less wellpreserved, the cranium of Ouranopithecus was probably kli-norhynch as well (Begun, 1994; Begun and Kordos, 1997;Kordos and Begun, 2001b). Ventral rotation of the face rela-tive to the neurocranium in African apes has been related tothe development of the supraorbital structures (supraorbitaltori/supratoral sulci), both of which are also more stronglydeveloped in Dryopithecus and Ouranopithecus than inother Miocene apes, though less strongly developed than inAfrica apes and most humans (Shea, 1988; Begun, 1994).

Finally, three specimens from Rudabánya preserve partialpremaxillae and anterior nasal fossae, all indicating the pres-ence of a stepped subnasal fossa and a true incisive canal(Begun, 1992, 1994; Begun and Kordos, 1997; Kordos andBegun, 2001b). All three specimens (RUD 12, RUD 44/47,and RUD 200) are damaged in this region, but together agood picture of the morphology of the region emerges, espe-cially in light of newly recovered fragments and teeth thatare associated with RUD 44/47 (Kordos and Begun, 2001a;Begun, 2002). The Dryopithecus fossil evidence in itsentirety has not been considered in previous interpretationsof this region (e.g. Ward and Kimbel, 1983; Ward and Pil-beam, 1983; McCollum and Ward, 1997). While other fossiltaxa have been said to demonstrate a stepped subnasal fossaor an elongated premaxilla, none other than Dryopithecusand Ouranopithecus among Miocene apes has the full suiteof features that typify this region in living African apes andin ‘australopithecines’. For example, Afropithecus andNacholapithecus have moderately elongated premaxillae butones that are relatively horizontal, labiolingually com-pressed with little or no overlap with the palatine process.Afropithecus has the typical mammalian pattern of a fenes-

trated palate, as in Proconsul [the premaxilla separated fromthe palatine process by a foramen (fenestra) rather than acanal (Schwartz, 1983)]. In Nacholapithecus there is moreoverlap with the maxilllary palatine process, but the overallconfiguration is nonetheless similar to Afropithecus (Ishidaet al., 2004). In African apes and humans, Dryopithecus, andOuranopithecus the nasoalveolar clivus region consists of anelongated premaxilla that is thick labiolingually and bi-con-cave labially. It overlaps to varying degrees with the palatineprocess of the maxilla but is vertically displaced and not incontact with the palate in the midline, thus contributing to astep into the nasal fossa. Gorilla, Dryopithecus, and Ouran-opithecus have relatively short premaxilla and limited over-lap with the palatine process while Pan andAustralopithecus have more elongated premaxilla with moreextensive overlap. The former condition has been inter-preted as primitive for the African ape and human clade(including the Eurasian fossil taxa) and the latter a synapo-morphy of the Pan–hominin clade (Begun, 1992, 1994).

The remaining synapomorphies of the Afro-Europeanhominid clade are too numerous to summarize here (Table1). It is simply worth noting that these features come frommany regions of the cranium, as do the synapomorphies ofthe great apes found in Dryopithecus. It seems highlyunlikely that they are all functionally or structurally interre-lated (though some may be), and more likely that most rep-resent distinct evolutionary transformations. Given theirnumber and diversity it is therefore very unlikely that allappear in parallel in European and African taxa.

AsiaA comprehensive review of the evidence for an Asian

great ape clade is beyond the scope of this paper, especiallysince several recent reviews are available (Ward, 1997;Kelley, 2002). I touched earlier on the possible relations

Table 1. Great ape and African ape cranio-dental character states of Dryopithecus

Great ape character states African ape character states

Labiolingually thick incisors Bi-convex premaxillaCompressed canines Stepped subnasal fossaElongated premolars and molars Patent incisive canalsM1 � M2 Broad, flat nasal aperture baseNo molar cingula Shallow canine fossaReduced premolar cusp

heteromorphySupraorbital torus

High root of the zygomatic Inflated glabellaElongated midface Frontal sinus above and below

nasionBroad nasal aperture

below the orbitsProjecting entoglenoid process

Reduced midfacial prognathism Fused articular and tympanic temporal

Elongated, robust premaxilla Broad temporal fossaPremaxilla–palatine overlap Deep glenoid fossaShallow subarcuate fossa Elongated neurocraniumEnlarged semicircular canals Moderate alveolar prognathismLarge brain KlinorhynchyHigh cranial base

Modified from Kordos and Begun (2001b).


among Sivapithecus, Ankarapithecus, Lufengpithecus, andGigantopithecus (see Figure 2). I will focus briefly on someproblem areas.

Sivapithecus is widely believed to be the sister taxon toPongo (Pilbeam, 1979, 1982; Andrews and Cronin, 1982;Ward and Pilbeam, 1983; Kelley and Pilbeam, 1986; Brownand Ward, 1988; Andrews, 1992; Ward, 1997; Kelley, 2002).Disagreement comes from two main sources. On the onehand, certain aspects of the morphology of Sivapithecus aredistinctly unlike that of Pongo. On the other hand, for someof these anatomical regions fossils more closely resemblingPongo are known. From the perspective of larger patterns ofdispersals and origins discussed here there are two importantquestions that arise from these objections. Are any of theAsian fossil great apes specifically related to Pongo and ifso, which clade is the most likely sister taxon?

Most of the researchers cited above have noted that theface of Sivapithecus is remarkably similar to that of Pongoin many details of the premaxilla, subnasal fossa, anteriornasal and periorbital regions, temporal bone morphology,dental implantation, and various aspects of dental propor-tions. However, most of these authors and others who havedescribed this material (e.g. Pilbeam et al., 1980; Kelley,1988; Brown, 1997) also note that Sivapithecus gnathic mor-phology is generally more robust, particularly the mandibleand zygomatic, than in similarly sized Pongo (only thelarger specimens of Sivapithecus fall within the range ofPongo, and even here they are for the most part within thefemale Pongo range). The canine crowns tend to be shorterand more robust and the postcanine dentition is large, thicklyenameled with broad, low, rounded cusps, shallow basins,and low dentine penetrance. This suite of characters hasbeen related by many authors to powerful mastication (e.g.Kay, 1981) and in fact occurs in many living and extinct pri-mate clades (Begun, 2001, 2002). In the case of the homi-

nids it is almost certainly the primitive condition, beingfound in the ‘griphopiths’ and to some extent in earlierclades (e.g. Afropithecus), though it recurs periodically(Ouranopithecus, Australopithecus, Gigantopithecus). Thusit is not too surprising that Sivapithecus may retain a numberof primitive features while sharing other characters withPongo, an example of one of the most commonly observedphenomena in paleobiology, mosaic evolution.

Other apparent homoplasies are more difficult to explain.Several authors have noted that Sivapithecus lacks a numberof postcranial characters present in Pongo. While some ofthese can also be attributed to the fact that Pongo and itsancestors have experienced millions of years of selectionsince Sivapithecus (e.g. the absence of a fovea for the liga-mentum teres of the femur in Pongo), others are not so easilydismissed as mosaic evolution. In particular, the upper fore-limb morphology of Sivapithecus is said to be too primitivefor the taxon to be related to Pongo (Benefit and McCrossin,1995; Pilbeam, 1996, 1997). This is based primarily on theobservations that the proximal half of the shaft of thehumerus in Sivapithecus is retroflexed or bent posteriorlyand distally (concave posteriorly), as in early Miocene hom-inoids, monkeys, and indeed most mammals (Pilbeam et al.,1990; Rose, 1997; Madar et al., 2002). The region is alsocharacterized by a very strongly developed platform for theinsertion of the pectoralis major and deltoideus muscles. Incontrast, the proximal humerus of Dryopithecus and all otherhominids is anteroflexed (smoothly concave anteriorly), orstraight, depending on the taxon. The humeral morphologyof Sivapithecus has impressed many researchers for a num-ber of good reasons. Sivapithecus is uniquely different fromall other hominids; it appears to share a morphology withmost other mammals; this morphology of the humerus isassociated with a fundamental aspect of behavioral ecology(retroflexed humeri are associated with pronograde quadru-

Figure 2. Phylogenetic relations and geographic distribution of some of the taxa discussed in the text. To save space, Griphopithecus repre-sents the ‘griphopiths’ Griphopithecus, Equatorius, Kenyapithecus, and Nacholapithecus.


peds and anteroflexed humeri with suspensory hominoids).These considerations have led some to question the phyleticlink between Sivapithecus and Pongo (e.g. Pilbeam, 1997;Larsen, 1998).

There is a risk here of throwing the baby out with the bathwater. Anatomical complexes that contribute in an inte-grated fashion to specific adaptations (i.e. suspensory posi-tional behavior) can be modified with various charactersdecoupled from others in response to specific selective pres-sures. Good examples include the australopithecine hip andthe hip of Pleistocene to recent Homo. Both are clearlyadapted to bipedalism and thus integrated tightly to numer-ous other postcranial attributes, but both are easily distin-guished from one another (Lovejoy, 1974; Stern andSusman, 1983). Differences in the paleobiology of eachtaxon accounts, at least in part, for dramatic differences in apart of the anatomy that is otherwise critical to the normalfunctioning of an essential aspect of their adaptation. If onewere to predict that a fossil with a chimpanzee-sized femoralhead could not have been from a biped, one would be wrongin the case of Australopithecus despite strong correlationsand good biomechanical arguments. We know this becausewe have a relatively good knowledge of the anatomy of thehip of Australopithecus. We do not have a good knowledgeof the shoulder of Sivapithecus. The proximal humerus ofSivapithecus, like the proximal femur of Homo, may havebeen modified in response to selection for a biomechanicallydifferent form of an otherwise similar positional behavior,antipronograde quadrupedalism, without losing the adapta-tion entirely.

In the end, Sivapithecus has more hominid-like postcra-nial attributes than not, and those, primarily of the proximalhumerus, that are non-hominid like are probably autapomor-phies of that genus, and have no bearing on the issue of itsrelationship to Pongo (Begun et al., 1997; Rose, 1997). Par-simony and an appeal to other examples in the hominoid fos-sil record both support the phylogeny depicted in Figure 2.

For the purposes of this paper it is not necessary to resolvethe issue of which Asian late Miocene hominid is mostclosely related to Pongo. Recent discoveries of Lufengpithe-cus in Thailand with postcanine teeth more closely resem-bling those of Pongo may help resolve this debate(Chaimanee et al., 2003), especially when cranial and post-cranial material is recovered. Despite the fact that Pongo,Lufengpithecus from China, and cf. Lufengpithecus fromThailand share complexly crenulated molar occlusal sur-faces, the faces of Sivapithecus and Pongo much moreclosely resemble one another than do the faces of Pongo andLufengpithecus. As with European hominids, it seems thatthere is great diversity in this radiation of Asian great apes.Sivapithecus is more likely to be the sister clade to Pongo. Itcould be that attributes characteristic of Pongo in Lufeng-pithecus and cf. Lufengpithecus are emergent in the mor-phology or development of the postcanine dentition of thisclade (they occur occasionally in Sivapithecus as well),providing a plausible explanation for the homoplasticappearance of this feature.

Eurasian Hominid Dispersals and Extinctions

The phylogeny depicted in Figure 2 has implications forthe origin of the modern clades of the great apes andhumans. Cladistic relations among Eurasian and Africanextinct and living hominids suggest that African hominidsderive from a European ancestor and Pongo derives from anAsian ancestor (Begun, 1994, 2001; Begun et al., 1997).This view has been supported by genetic evidence (Stewartand Disotel, 1998) and criticized based on differences ininterpretations of phylogeny discussed above and on per-ceived limitations of the fossil record. With regard to the lat-ter, it has been noted that Africa is huge and barely sampled,especially in the late Miocene, but a large number of lateMiocene localities are in fact known, many preservingpaleoecological indications of forested settings (Begun,2001). The hypothesis that African hominids originated inEurope or western Asia is not based simply on the absenceof evidence from Africa. It is based on two phylogeneticconclusions. Firstly, the hominine clade is present in Europein the late Miocene (Begun et al., 1997; Begun, 2001). Sec-ondly, the few contemporaneous hominoids from Africa arenot cladistically hominid (Begun, 2001). Otavipithecus isclearly primitive (Begun, 1994; Singleton, 2000). Sambu-rupithecus superficially resembles gorillas in a few aspectsof molar morphology but retains many primitive dental andmaxillary characters (Begun, 2001). The isolated teeth fromNgorora have been described as having affinities primarilywith the Proconsuloidea or middle Miocene East Africanhominoids (e.g. Equatorius) (Hill and Ward, 1988; Begun,2001; Hill et al., 2002).

Even if the absence of the hominine clade in Africa beforethe latest Miocene represents a sampling bias, why is itpresent in Europe? If European hominids are not related toAfrican hominids, then why do they share so many charac-ters with African hominids? These may be homoplasies, butas noted above this is highly unlikely. In other words, Eur-asian late Miocene hominids are most reasonably interpretedas hominids: those from Europe are stem hominines (Afri-can apes and humans), and the Asian late Miocene hominidsare stem pongines (Pongo). It is on this basis that I wouldexplain the fact that a crown hominid (related to both Afri-can and Asian hominids) has not yet been found in Africa.On the other hand it may well be that the Afro-Europeanhominine clade was widely distributed across Europe, west-ern Asia, and Africa, but that up until now they have onlybeen found in Europe. It is certainly possible that new fossildiscoveries will show that the African ape and human claderemained in part in Africa; that the clade expanded intoEurope but a core remained in Africa; and that in the endEuropean hominines went extinct and African taxa, cur-rently unknown, are the true ancestors of the African ape andhuman clade.

I have suggested that hominids moved south from Eurasiain response to global climate changes that produced moreseasonal conditions in Eurasia toward the end of theMiocene [Begun (2001) based in part on Quade et al. (1989),Leakey et al. (1996), Cerling et al. (1997) and many others].This scenario seems to hold up reasonably well in the east.Whichever Miocene taxon is ultimately related to Pongo, all


occur north of the current range of the living taxon, consis-tent with a climatic change that made it more challenging forhominids to live north of the Tropic of Cancer. In the west, ifhominines originated in Europe or western Asia, the paleo-biogeographic model is consistent with that from East Asia;they moved south at the end of the Miocene as well. If anunknown core of hominines has yet to be found in Africa,this model needs revision to account for the origin of thehominines (Figure 3).

Corroborating evidenceAmple evidence now exists for a gradual degradation in

climate throughout much of Eurasia in the late Miocene,which with additional factors, mainly tectonic, culminated inthe development of markedly more seasonal conditions anda number of dramatic ecological changes. The most spectac-ular in the Miocene is the Messinian salinity crisis that led tothe dessication of the Mediterranean basin at the end of theMiocene (Hsü et al., 1973; Clauzon et al., 1996; Krijgsmanet al., 1999). Other consequences include the developmentof Asian monsoons, desertification in North Africa, the earlyphases of Neogene polar ice cap expansion and the expan-sion of North American grasslands (Garcés et al., 1997;Hoorn et al., 2000; Zhisheng et al., 2001; Griffin, 2002; Guoet al., 2002; Janis et al., 2002; Liu and Yin, 2002; Wilson etal., 2002). In both Europe and Asia subtropical forestsretreated and were increasing replaced by more open coun-try, including true grasslands and steppes (Bernor et al.,1979; Bernor, 1983; Fortelius et al., 1996; Cerling et al.,1997; de Bonis et al., 1999; Magyar et al., 1999; Solouniaset al., 1999; Fortelius and Hokkanen, 2001). It is importantto note that in some places forests persisted and elsewheremore severe changes occurred, creating a number of refugia,some of which continued to host hominids well into theperiod of climatic deterioration. This is the case for theOreopithecus localities of Tuscany and Sardinia (Harrisonand Rook, 1997). Other well known localities such as Dorn-Dürkheim in Germany retained a strongly forested characterthough they lack hominoids (Franzen, 1997; Franzen andStorch, 1999).

For the most part though, hominids disappeared fromEurope and western Asia in the late Miocene, but not all atonce. There is a gradient of extinctions of forest forms fromwest to east, corresponding to the gradient of appearance ofmore open country faunas from west to east (Bernor et al.,1979; Fortelius et al., 1996; Begun, 2001). Between about12 and 10 Ma Dryopithecus disappeared from localities inEurope, becoming very rare by 9.5 Ma in Spain and Ger-many, and there represented only by one or two specimens.This is despite the persistence locally of apparently suitableecological settings such as at Dorn-Dürkheim (Franzen,1997). This wave of extinctions ends coincident with animportant faunal event in western Europe known as the mid-Vallesian event, when a major turnover of terrestrial faunasled to the widespread extinction of local taxa generallyattributed to the development of more open conditions(Moyà-Solà and Agustí, 1990; Fortelius et al., 1996). Theyoungest specimens possibly attributable to Dryopithecusare so far east that they are actually Asian, though still westof Sivapithecus localities. These teeth are currently assigned

to Udabnopithecus from the 8–8.5 Ma locality of Udabno inGeorgia (Gabunia et al., 2001).

In the eastern Mediterranean hominids persisted to theend of this time period and perhaps somewhat beyond.Ouranopithecus in Greece is mainly known from the end ofthe hominid reign in Europe, and may be a terminal taxon ofthe Dryopithecus clade (Begun and Kordos, 1997). In Ana-

Figure 3. Paleobiogeography of late Miocene hominids. (A) Atthe end of the middle Miocene into the late Miocene hominines diver-sify in western Eurasia and pongines in South and East Asia, whilehominids in Africa become extinct. The one exception to the title ofthis paper may be in central Anatolia (doubled ended arrow) whereone clade of pongine (Ankarapithecus) appears after the earliestappearance of the clade in South Asia, though in fact no hominines areknown from Anatolia at this time. (B) After ~9 Ma hominids progres-sively go extinct in most of Eurasia, starting in the west and movingeast. During this time hominines first appear in Africa, and probably inSoutheast Asia as well.


tolia at the eastern edge of the faunal province that includesGreece and the eastern Mediterranean [the Greco-Iranianprovince (de Bonis et al., 1999)] a new, very large hominidresembling Ouranopithecus may be as young as 7–8 Ma inage (Sevim et al., 2001). During this time, forest taxa wereincreasingly replaced by more open country forms. This istrue of virtually all mammalian orders. Among the primates,hominoids declined and cercopithecoids were on theincrease (Andrews et al., 1996). Grazing ungulates andgrassland or dry ecology adapted micromammals alsobecome more common (Fortelius et al., 1996; Agustí et al.,1999; de Bonis et al., 1999; Solounias et al., 1999).

The dispersal of late Miocene faunas between Eurasia andAfrica is complex and has intercontinental sources. Thoughthe dominant signature was the rise of open country taxa,forest forms persisted and dispersed as well, including hom-inids. Among the more open country forms horses dispersedfrom North America to the Old World, and modern bovidsand giraffids appear to have dispersed from Europe to Africa(Dawson, 1999; Solounias et al., 1999; van der Made, 1999;Agustí et al., 2001). Among the more closed-setting mam-mals, hippos moved from Africa to Europe, and pigs of vary-ing ecological preferences moved from Asia to Europe andAfrica (Fortelius et al., 1996; van der Made, 1999). Smallcarnivores (mustelids, felids and viverrids), larger carni-vores (ursids, hyaenids), porcupines, rabbits, and chalico-theres, most of which also prefer more closed settings, alsodispersed from Eurasia to Africa (Leakey et al., 1996; Gins-burg, 1999; Heissig, 1999; van der Made, 1999; Winkler,2002). In sum, phylogenetic analysis suggests that the ances-tors of many African mammals had appeared in the Valle-sian of Eurasia. Ecologically many of these new taxa wereadapted to more open country conditions (hypsodonty, cur-soriality, etc.), but faunal exchanges among the three OldWorld continents also involved forest or wetter ecologyforms (hippos, some suids, primates, carnivores, rodents,and chalicotheres), which is consistent with the evidence ofclimate change at that time. Many taxa dispersed south intoAfrica as conditions continued to deteriorate, leading to theMessinian crisis; among these taxa were probably the ances-tors of the African apes and humans. While the scenario thatwould have hominid ancestors leave Africa in the earlyMiocene only to return to Africa as hominines in the lateMiocene may seem unnecessarily complicated, this is pre-cisely what seems to have occurred in several mammalianlineages, including those represented by late Miocene Afri-can species of Orycteropus, several small carnivores, thehippo Hexaprotodon, and possibly the proboscideans Anan-cus, Deinotherium, and Choerolophodon (Leakey et al.,1996; Ginsburg, 1999; Heissig, 1999; van der Made, 1999;Boisserie et al., 2003; Werdelin, 2003). This number is likelyto increase as our knowledge of fossil mammal lineagesexpands. Finally, one could well ask how Dryopithecus, asubtropical forest great ape, or Ouranopithecus, a massive-jawed, more open woodland great ape could have made it toAfrica to evolve into the thinly enameled ancestors ofhumans such as Ardipithecus [although there are severalcandidates for earliest human, a discussion of which isbeyond the scope of this paper, Ardipithecus ramidus is inmy view the only taxon with clear and unambiguous affini-

ties to later hominins (Begun, 2004)]. They did not. Dryo-pithecus and Ouranopithecus are sister taxa to the ancestorof the African ape and human clade. The actual last commonancestor of that clade is not known. However, the distribu-tion of characteristics of modern and fossil hominines sug-gests that this last common ancestor would most likely havebeen woodland or forest dependent and but strongly terres-trial, that is, a knuckle-walker (Richmond et al., 2001).

Conclusions

The phylogenetic divide between Asian and Afro-Euro-pean great apes is deep. The east and west have been sepa-rated for at least 14–16 Ma. On the other hand, both lineagesresponded in similar ways to the changing ecological set-tings of late Miocene Eurasia. Both for the most partremained true to their forest ecology origins, and bothmoved south as the forests retreated in that direction.

Acknowledgments

I am grateful to the organizers of the international sympo-sium “Asian Paleoprimatology: Evolution of the TertiaryPrimates in Asia” for inviting me to participate and to theeditors of Anthropological Science for producing this spe-cial volume. Special thanks go to Masanaru Takai as editorof this special volume for all his hard work. I am also grate-ful to Masato Nakatsukasa and an anonymous reviewer forthoughtful comments that improved this manuscript.

References

Agustí J., Cabrera L., Garcés M., and Llenas M. (1999) Mammalturnover and global climate change in the late Miocene terres-trial record of the Vallès Penedès basin (NE Spain). In: AgustíJ., Rook L., and Andrews P. (eds.), Hominoid Evolution andClimate Change in Europe, Volume 1: The Evolution of Neo-gene Terrestrial Ecosystems in Europe. Cambridge UniversityPress, Cambridge, pp. 397–412.

Agustí J., Cabrera L., and Garcés M. (2001) Chronology and zoo-geography of the Miocene hominoid record in Europe. In: deBonis L., Koufos G., and Andrews P. (eds.), Hominoid Evolu-tion and Environmental Change in the Neogene of Europe,Volume 2: Hologeny of the Neogene Hominoid Primates ofEurasia. Cambridge University Press, Cambridge, pp. 2–18.

Alpagut B., Andrews P., and Martin L. (1990) New Miocene hom-inoid specimens from the middle Miocene site at Paşalar.Journal of Human Evolution, 19: 397–422.

Alpagut B., Andrews P., Fortelius M., Kappelman K., Temizsoy I.,and Lindsay W. (1996) A new specimen of Ankarapithecusmeteai from the Sinap formation of central Anatolia. Nature,382: 349–351.

Andrews P. (1992) Evolution and environment in the Hominoidea.Nature, 360: 641–646.

Andrews P. and Cronin J. (1982) The relationships of Sivapithecusand Ramapithecus and the evolution of the orang-utan.Nature, 297: 541–546.

Andrews P., Harrison T., Delson E., Bernor R.L., and Martin L.(1996) Distribution and biochronology of European andSouthwest Asian Miocene Catarrhines. In: Bernor R.L., Fahl-busch V., and Mittmann H.-W. (eds.), The Evolution of West-ern Eurasian Neogene Mammal Faunas. Columbia UniversityPress, New York, pp. 168–295.

Begun D.R. (1992) Miocene fossil hominids and the chimp–


human clade. Science, 257: 1929–1933.Begun D.R. (1994) Relations among the great apes and humans:

new interpretations based on the fossil great ape Dryopithe-cus. Yearbook of Physical Anthropology, 37: 11–63.

Begun D.R. (2001) African and Eurasian Miocene hominoids andthe origins of the Hominidae. In: de Bonis L., Koufos G., andAndrews P. (eds.), Hominoid Evolution and EnvironmentalChange in the Neogene of Europe, Volume 2: Phylogeny ofthe Neogene Hominoid Primates of Eurasia. Cambridge Uni-versity Press, Cambridge, pp. 231–253.

Begun D.R. (2002) European Hominoids. In: Hartwig W. (ed.),The Primate Fossil Record. Cambridge University Press,Cambridge, pp. 339–368.

Begun D.R. (2004) The earliest hominins—is less more? Science,303: 1478–1480.

Begun D.R. and Güleç E. (1998) Restoration of the Type and Pal-ate of Ankarapithecus meteai: taxonomic, phylogenetic, andfunctional implications. American Journal of PhysicalAnthropology, 105: 279–314.

Begun D.R. and Kordos L. (1997) Phyletic affinities and func-tional convergence in Dryopithecus and other Miocene andliving hominids. In: Begun D.R., Ward C.V., and Rose M.D.(eds.), Function, Phylogeny, and Fossils: Miocene HominoidEvolution and Adaptations. Plenum Press, New York, pp.291–316.

Begun D.R., Ward C.V., and Rose M.D. (1997) Events in homi-noid evolution. In: Begun D.R., Ward C.V., and Rose M.D.(eds.), Function, Phylogeny, and Fossils: Miocene HominoidOrigins and Adaptations. Plenum Press, New York, pp. 389–415.

Begun D.R., Geraads D., and Güleç E. (2003) The Çandır homi-noid locality: implications for the timing and pattern of homi-noid dispersal events. Courier ForschungsinstitutSenckenberg, 240: 251–265.

Benefit B.R. and McCrossin M.L. (1995) Miocene Hominoids andhominid origins. Annual Review of Anthropology, 24: 237–256.

Bernor R.L. (1983) Geochronology and zoogeographic relation-ships of Miocene Hominoidea. In: Corruccini R.S. and Cio-chon R.L. (eds.), New Interpretations of Ape and HumanAncestry. Academic Press, New York, 21–64.

Bernor R.L., Andrews P.J., Solounias N., and van CouveringJ.A.H. (1979) The evolution of ‘Pontian’ mammal faunas:some zoogeographic, paleoecologic and chronostratigraphicconsiderations. Annales Géologiques Pays Hellèniques, 7thInternational Congress on Mediterranean Neogene, Athéns, 1:81–89.

Boisserie J.-R., Brunet M., Andossa L., and Vignaud P. (2003)Hippopotamids from the Djurab Pliocene faunas, Chad, Cen-tral Africa. Journal of African Earth Sciences, 36: 15–27.

Brown B. (1997) Miocene hominoid mandibles: functional andphylogenetic perspectives. In: Begun D.R., Ward C.V., andRose M.D. (eds.), Function, Phylogeny, and Fossils: MioceneHominoid Evolution and Adaptations. Plenum Press, NewYork, pp. 153–171.

Brown B. and Ward S. (1988) Basicranial and facial topography inPongo and Sivapithecus. In: Schwartz J.H. (ed.), Ourang-utanBiology. Oxford University Press, New York, pp. 247–260.

Cerling T., Harris J.R., MacFadden B.J., Leakey M.G., Quade J.,Eisenmann V., and Ehleringer J.R. (1997) Global vegetationchange through the Miocene/Pliocene boundary. Nature, 389:153–158.

Chaimanee Y., Jolly D., Benammi M., Tafforeau P., Duzer D.,Moussa I., and Jaeger J.-J. (2003) A middle Miocene homi-noid from Thailand and orangutan origins. Nature, 422: 61–65.

Clauzon G., Suc J.-P., Gautier F., Berger A., and Loutre M.-F.(1996) Alternate interpretation of the Messinian salinity cri-sis: controversy resolved? Geology, 24: 363–366.

Dawson M.R. (1999) Bering down: Miocene dispersals of landmammals between North America and Europe. In: Rössner G.and Heissig K. (eds.), The Miocene Land Mammals ofEurope. Verlag Dr. Friedrich Pfeil, München, pp. 473–483.

de Bonis L. and Koufos G. (1993) The face and mandible of Oura-nopithecus macedoniensis: description of new specimens andcomparisons. Journal of Human Evolution, 24: 469–491.

de Bonis L. and Koufos G. (1997) The phylogenetic and functionalimplications of Ouranopithecus macedoniensis. In: BegunD.R., Ward C.V., and Rose M.D. (eds.), Function, Phylogeny,and Fossils: Miocene Hominoid Origins and Adaptations.Plenum Press, New York, pp. 317–326.

de Bonis L., Bouvrain G., and Koufos G. (1999) Palaeoenviron-ments of late Miocene primate localities in Macedonia,Greece. In: Agusti J., Rook L., and Andrews P. (eds.), TheEvolution of Neogene Terrestrial Ecosystems in Europe.Cambridge University Press, Cambridge, pp. 413–435.

Eckhardt J.T. and Eckhardt R.B. (1995) Ancient trans-speciespolymorphisms shared by African and Asian Hominoideapreclude assignment of Dryopithecus to a Pongo clade.American Journal of Physical Anthropology, Supplement 20:86.

Fortelius M. and Hokkanen A. (2001) The trophic context of hom-inoid occurrence in the later Miocene of western Eurasia: aprimate-free view. In: de Bonis L., Koufos G.D., and AndrewsP. (eds.), Hominoid Evolution and Climatic Change inEurope, Volume 2: Phylogeny of the Neogene Hominoid Pri-mates of Eurasia. Cambridge University Press, Cambridge,19–47.

Fortelius M., Werdelin L., Andrews P., Bernor R.L., Gentry A.,Humphrey L., Mittmann H.-W., and Viranta S. (1996) Provin-ciality, diversity, turnover and paleoecology in land mammalfaunas of the later Miocene of Western Eurasia. In: BernorR.L., Fahlbusch V., and Mittmann H.-W. (eds.), The Evolu-tion of Western Eurasian Neogene Mammal Faunas. Colum-bia University Press, New York, pp. 414–448.

Franzen J.L. (1997) Die Säugetiere aus dem Turolium von Dorn-Dürkheim (Rheinhessen, Deutschland), Teil 1: Carnivora,Proboscidea (Tetralophodontidae), Perissodactyla (Rhinocer-otidae, Equidae), Artiodactyla (Suidae). CourierForschungsinstitut Senckenberg, Band 197, Senckenber-gische Naturforschende Gesellschaft, Frankfurt am Main.

Franzen J.L. and Storch G. (1999) Late Miocene mammals fromCentral Europe. In: Agustí J., Rook L., and Andrews P. (eds.),Hominoid Evolution and Climatic Change in Europe, Volume1: The Evolution of Neogene Terrestrial Ecosystems inEurope. Cambridge University Press, Cambridge, pp. 165–190.

Gabunia L., Gabashvili E., Vekua A., and Lordkipanidze D. (2001)The late Miocene hominoid from Georgia. In: de Bonis L.,Koufos G., and Andrews P. (eds.), Hominoid Evolution andClimatic Change in Europe, Volume 2: Phylogeny of the Neo-gene Hominoid Primates of Eurasia. Cambridge UniversityPress, Cambridge, pp. 316–325.

Garcés M., Cabrera L., Agustí J., and Parés J.M. (1997) Old Worldfirst appearance datum of ‘Hipparion’ horses: late Miocenelarge-mammal dispersal and global events. Geology, 25: 19–22.

Gaudry A. (1890) Le Dryopithèque. Mémoires de la SociétéGéologique de France, 1: 1–11.

Ginsburg L. (1999) Order Carnivora. In: Rössner G.E. and HeissigK. (eds.), The Miocene Land Mammals of Europe. Verlag Dr.Friedrich Pfeil, München, pp. 109–148.

Griffin D.L. (2002) Aridity and humidity: two aspects of the lateMiocene climate of North Africa and the Mediterranean.Palaeogeography, Palaeoclimatology, Palaeoecology, 182:65–91.

Guo Z.T., Ruddiman W.F., Hao Q.Z., Wu H.B., Qiao Y.S., ZhuR.X., Peng S.Z., Wei J.J., Yuan B.Y., and Liu T.S. (2002)


Onset of Asian desertification by 22 Myr ago inferred fromloess deposits in China. Nature, 416: 159–163.

Harrison T. (2002) Late Oligocene to middle Miocene catarrhinesfrom Afro-Arabia. In: Hartwig W. (ed.), The Primate FossilRecord. Cambridge University Press, Cambridge, pp. 311–338.

Harrison T. and Rook L. (1997) Enigmatic anthropoid or misun-derstood ape: the phylogenetic status of Oreopithecus bam-bolii reconsidered. In: Begun D.R., Ward C.V., and RoseM.D. (eds.), Function, Phylogeny and Fossils: Miocene Hom-inoid Origins and Adaptations. Plenum Press, New York, pp.327–362.

Heissig K. (1999) Family Chalicotheriidae. In: Rössner G.E. andHeissig K. (eds.), The Miocene Land Mammals of Europe.Verlag Dr. Friedrich Pfeil, München, pp. 189–192.

Heizmann E. and Begun D.R. (2001) The oldest European homi-noid. Journal of Human Evolution, 41: 463–481.

Hill A. and Ward S. (1988) Origin of the Hominidae: the record ofAfrican large hominoid evolution between 14 my and 4 my.Yearbook of Physical Anthropology, 31: 49–83.

Hill A., Leakey M., Kingston J.D., and Ward S. (2002) New cerco-pithecoids and a hominoid from 12.5 Ma in the Tugen Hillssuccession, Kenya. Journal of Human Evolution, 42: 75–93.

Hoorn C., Ohja T., and Quade J. (2000) Palynological evidence forvegetation development and climatic change in the Sub-Himalayan Zone (Neogene, Central Nepal). Palaeogeography,Palaeoclimatology, Palaeoecology, 163: 133–161.

Hsü K.J., Cita M.B., and Ryan W.B.F. (1973) The origin of theMediterranean evaporites. Initial Reports of the Deep SeaDrilling Project, 13: 1203–1231.

Ishida H., Kunimatsu Y., Takano T., Nakano Y., and NakatsukasaM. (2004) Nacholapithecus skeleton from the middleMiocene of Kenya. Journal of Human Evolution, 46: 1–35.

Janis C.M., Damuth J., and Theodor J.M. (2002) The origins andevolution of the North American grassland biome: the storyfrom the hoofed mammals. Palaeogeography, Palaeoclimatol-ogy, Palaeoecology, 177: 183–198.

Kappelman J., Kelley J., Pilbeam D.R., Sheikh K.A., Ward S.,Anwar M., Barry J.C., Brown B., Hake P., Johnson N.M.,Raza S.M., and Shah S.M.I. (1991) The earliest occurrence ofSivapithecus from the middle Miocene Chinji Formation ofPakistan. Journal of Human Evolution, 21: 61–73.

Kappelman J., Richmond B.G., Seiffert E.R., Maga A.M., andRyan T.M. (2003) Hominoidea (Primates). In: Fortelius M.,Kappelman J., Sen S., and Bernor R. (eds.), Geology andPaleontology of the Miocene Sinap Formation, Turkey.Columbia University Press, New York, pp. 90–124.

Kay R.F. (1981) The nut-crackers—a theory of the adaptations ofthe Ramapithecinae. American Journal of Physical Anthro-pology, 55: 141–1151.

Kelley J. (1988) A new large species of Sivapithecus from theSiwaliks of Pakistan. Journal of Human Evolution, 17: 305–324.

Kelley J. (2002) The Hominoid Radiation in Asia. In: Hartwig W.(ed.), The Primate Fossil Record. Cambridge UniversityPress, Cambridge, pp. 369–384.

Kelley J. and Pilbeam D.R. (1986) The Dryopithecines: taxonomy,comparative anatomy, and phylogeny of Miocene large homi-noid. In: Swindler D.R. and Erwin J. (eds.), Comparative Pri-mate Biology, Volume 1: Systematics, Evolution andAnatomy. Alan R. Liss, New York, pp. 361–411.

Kordos L. and Begun D.R. (2001a) Fossil catarrhines from the lateMiocene of Rudabánya. Journal of Human Evolution, 40: 17–39.

Kordos L. and Begun D.R. (2001b) A new cranium of Dryopithe-cus from Rudabánya, Hungary. Journal of Human Evolution,41: 689–700.

Koufos G.D. (1995) The first female maxilla of the hominoidOuranopithecus macedoniensis from the late Miocene of

Macedonia, Greece. Journal of Human Evolution, 29(4):385–389.

Krijgsman W., Hilgen F.J., Raffi I., Sierro F.J., and Wilson D.S.(1999) Chronology, causes and progression of the Messiniansalinity crisis. Nature, 400: 652–655.

Larson S.G. (1998) Parallel evolution in the hominoid trunk andforelimb. Evolutionary Anthropology, 6(3): 87–99.

Lartet E. (1856) Note sur un grand singe fossile qui se rattache augroupe des singes superieurs. Comptes Rendus Hebdoma-daires des Séances de l’Académie des Sciences, 43: 219–223.

Leakey M.G. and Walker A. (1997) Afropithecus: function andphylogeny. In: Begun D.R., Ward C.V., and Rose M.D. (eds.),Function, Phylogeny and Fossils: Miocene Hominoid Evolu-tion and Adaptations. Plenum Press, New York, pp. 225–239.

Leakey M.G., Feibel C.S., Bernor R.L., Harris J.M., Cerling T.E.,Stewart K.M., Storrs G.W., Walker A., Werdelin L., and Win-kler A.J. (1996) Lothagam: a record of faunal change in theLate Miocene of East Africa. Journal of Vertebrate Paleontol-ogy, 16: 556–570.

Liu X. and Yin Z.-Y. (2002) Sensitivity of East Asian monsoon cli-mate to the uplift of the Tibetan Plateau. Palaeogeography,Palaeoclimatology, Palaeoecology, 183: 223–245.

Lovejoy C.O. (1974) The gait of australopithecines. Yearbook ofPhysical Anthropology, 17: 147–161.

Lydekker R. (1879) Further notices of Siwalik Mammalia. Recordsof the Geological Survey of India, 7: 33–52.

Madar S.I., Rose M.D., Kelley J., MacLatchy L., and Pilbeam D.R.(2002) New Sivapithecus postcranial specimens from theSiwaliks of Pakistan. Journal of Human Evolution, 42: 705–752.

Magyar I., Geary D.H., and Müller P. (1999) Paleogeographic evo-lution of the late Miocene Lake Pannon in Central Europe.Palaeogeography, Palaeoclimatology, Palaeoecology, 147:151–167.

McCollum M.A. and Ward S.C. (1997) Subnasoalveolar anatomyand hominoid phylogeny: evidence from comparative ontog-eny. American Journal of Physical Anthropology, 102: 377–405.

McCrossin M.L. and Benefit B.R. (1997) On the relationships andadaptations of Kenyapithecus, a large-bodied hominoid fromthe middle Miocene of eastern Africa. In: Begun D.R., WardC.V., and Rose M.D. (eds.), Function, Phylogeny and Fossils:Miocene Hominoid Origins and Adaptations. Plenum Press,New York, pp. 241–267.

McCrossin M.L., Benefit B.R., and Gitau S.N. (1998) Functionaland phylogenetic analysis of the distal radius of Kenyapithe-cus with comments on the origin of the African ape andhuman clade. American Journal of Physical Anthropology,Supplement 26: 158–159.

Moyà-Solà S. and Agustí J. (1990) Bioevents and mammal succes-sions in the Spanish Miocene. In: Lindsay E.H., Fahlbusch V.,and Mein P. (eds.), European Neogene Mammal Chronology.Plenum Press, New York, pp. 357–373.

Moyà-Solà S. and Köhler M. (1993) Recent discoveries of Dryop-ithecus shed new light on evolution of great apes. Nature,365: 543–545.

Moyà-Solà S. and Köhler M. (1995) New partial cranium of Dryo-pithecus Lartet, 1863 (Hominoidea, Primates) from the upperMiocene of Can Llobateres, Barcelona, Spain. Journal ofHuman Evolution, 29: 101–139.

Msuya C.P. and Harrison T. (1996) The circumorbital foramina inprimates: a phylogenetic perspective. Kaupia-DarmstädterBeiträge zur Naturgeschichte, 6: 103–109.

Nakatsukasa M., Yamanaka A., Kunimatsu Y., Shimizu D., andIshida H. (1998) A newly discovered Kenyapithecus skeletonand its implications for the evolution of positional behavior inMiocene East African hominoids. Journal of Human Evolu-tion, 34: 659–664.

Pilbeam D.R. (1979) Recent finds and interpretations of Miocene


hominoids. Annual Review of Anthropology, 8: 333–352.Pilbeam D.R. (1982) New hominoid skull material from the

Miocene of Pakistan. Nature, 295: 232–234.Pilbeam D.R. (1996) Genetic and morphological records of the

Hominoidea and hominid origins: a synthesis. MolecularPhylogenetics and Evolution, 5: 155–168.

Pilbeam D.R. (1997) Research on Miocene hominoids and homi-nid origins: the last three decades. In: Begun D.R., Ward C.V.,and Rose M.D. (eds.), Function, Phylogeny and Fossils:Miocene Hominoid Evolution and Adaptations. PlenumPress, New York, pp. 13–28.

Pilbeam D.R., Rose M.D., Badgley C., and Lipschultz B. (1980)Miocene Hominoids from Pakistan. Postilla, 181: 1–94.

Pilbeam D.R., Rose M.D., Barry J.C., and Shah S.M.I. (1990) NewSivapithecus humeri from Pakistan and the relationship ofSivapithecus and Pongo. Nature, 384: 237–239.

Pilgrim G. (1910) Notices of new mammalian genera and speciesfrom the Tertiaries of India. Records of the Geological Surveyof India, 40: 63–71.

Quade J., Cerling T.E., and Bowman J.R. (1989) Development ofAsian monsoon revealed by marked ecological shift duringthe latest Miocene. Nature, 342: 163–166.

Richmond B.G., Begun D.R., and Strait D.S. (2001) Origin ofhuman bipedalism: the knuckle-walking hypothesis reconsid-ered. Yearbook of Physical Anthropology, 44: 70–105.

Rögl F. (1999) Circum-Mediterranean Miocene Paleogeography.In: Rössner G. and Heissig K. (eds.), The Miocene LandMammals of Europe. Verlag Dr. Friedrich Pfeil, München,pp. 39–48.

Rose M.D. (1997) Functional and phylogenetic features of theforelimb in Miocene hominoids. In: Begun D.R., Ward C.V.,and Rose M.D. (eds.), Function, Phylogeny and Fossils:Miocene Hominoid Evolution and Adaptations. PlenumPress, New York, pp. 79–100.

Schwartz J.H. (1983) Palatine fenestrae, the orangutan and homi-noid evolution. Primates, 24: 231–240.

Schwartz J.H. (1997) Lufengpithecus and Hominoid phylogeny.Problems in Delineating and evaluating phylogenetically rele-vant characters. In: Begun D.R., Ward C.V., and Rose M.D.(eds.), Function, Phylogeny, and Fossils: Miocene HominoidEvolution and Adaptations. Plenum Press, New York, pp.363–388.

Sevim A., Begun D.R., Güleç E., Geraads D., and Pehlevan Ç.(2001) A new late Miocene hominid from Turkey. AmericanJournal of Physical Anthropology, Supplement 32: 134–135.

Shea B.T. (1988) Phylogeny and skull form in the hominoid pri-mates. In: Schwartz J. (ed.), Orang-Utan Biology. OxfordUniversity Press, New York, pp. 233–245.

Singleton M. (2000) The phylogenetic affinities of Otavipithecusnamibiensis. Journal of Human Evolution, 38: 537–573.

Solounias N., Plavcan J.M., Quade J., and Witmer L. (1999) Thepaleoecology of the Pikermian Biome and the Savanna myth.In: Agusti J., Rook L., and Andrews P. (eds.), The Evolutionof Neogene Terrestrial Ecosystems in Europe. CambridgeUniversity Press, Cambridge, pp. 436–453.

Steininger F.F. (1999) Chronostratigraphy, Geochronology and

Biochronology of the Miocene ‘European Land MammalMega-Zones (ELMMZ)’ and the Miocene ‘Mammal-Zones(MN-Zones)’. In: Rössner G.E. and Heissig K. (eds.), TheMiocene Land Mammals of Europe. Verlag Dr. FriedrichPfeil, München, pp. 9–24.

Stern J.T. and Susman R.L. (1983) The locomotor anatomy of Aus-tralopithecus afarensis. American Journal of PhysicalAnthropology, 60: 279–317.

Stewart C.-B. and Disotell T.R. (1998) Primate evolution—in andout of Africa. Current Biology, 8: 582–588.

van der Made J. (1999) Intercontinental relationship Europe–Africa and the Indian subcontinent. In: Rössner G.E. andHeissig K. (eds.), The Miocene Land Mammals of Europe.Verlag Dr. Friedrich Pfeil, München, pp. 457–472.

Ward S.C. (1997) The taxonomy and phylogenetic relationships ofSivapithecus revisited. In: Begun D.R., Ward C.V., and RoseM.D. (eds.), Function, Phylogeny and Fossils: Miocene Hom-inid Origins and Adaptations. Plenum Press, New York, pp.269–290.

Ward S.C. and Duren D.L. (2002) Middle and late Miocene Afri-can hominoids. In: Hartwig W. (ed.), The Primate FossilRecord. Cambridge University Press, Cambridge, pp. 385–397.

Ward S.C. and Kimbel W.H. (1983) Subnasal alveolar morphologyand the systematic position of Sivapithecus. American Jour-nal of Physical anthropology, 61: 157–171.

Ward S.C. and Pilbeam D.R. (1983) Maxillofacial morphology ofMiocene hominoids from Africa and Indo-Pakistan. In: Cor-ruccini R.L. and Ciochon R.S. (eds.), New Interpretations ofApe and Human Ancestry. Plenum Press, New York, pp. 211–238.

Ward S.C., Brown B., Hill A., Kelley J., and Downs W. (1999)Equatorius: a new hominoid genus from the middle Mioceneof Kenya. Science, 285: 1382–1386.

Werdelin L. (2003) Mio-Pliocene Carnivora from Lothagam,Kenya. In: Leakey M.G. and Harris J.M. (eds.), Lothagam:The Dawn of Humanity in Eastern Africa. Columbia Univer-sity Press, New York, 261–328.

Wilson G.S., Barron J.A., Ashworth A.C., Askin R.A., Carter J.A.,Curren M.G., Dalhuisen D.H., Friedmann E.I., Fyodorov-Davidov D.G., Gilichinsky D.A., Harper M.A., HarwoodD.M., Hiemstra J.F., Janecek T.R., Licht K.J., OstroumovV.E., Powell R.D., Rivkina E.M., Rose S.A., Stroeven A.P.,Stroeven P., van der Meer J.J.M., and Wizevich M.C. (2002)The Mount Feather Diamicton of the Sirius Group: an accu-mulation of indicators of Neogene Antarctic glacial and cli-matic history. Palaeogeography, Palaeoclimatology,Palaeoecology, 182: 117–131.

Winkler A.J. (2002) Neogene paleobiogeography and East Africanpaleoenvironments: contributions from the Tugen Hillsrodents and lagomorphs. Journal of Human Evolution, 42:237–256.

Zhisheng A., Kutzbach J.E., Prell W.L., and Porter S.C. (2001)Evolution of Asian monsoons and phased uplift of the Hima-laya-Tibetan plateau since late Miocene times. Nature, 411:62–66.


Appendix. Classification of hominoids included in this analysis

Hominoidea‘Eohominoidea’

ProconsulidaeProconsul

AfropithecidaeAfropithecusHeliopithecus

‘Eohominoidea’ indet. (‘griphopiths’)Griphopithecuscf. GriphopithecusEquatoriusNacholapithecusKenyapithecus

‘Euhominoidea’Hominidae

HomininaeDryopithecusOuranopithecusPanGorillaHomo

PonginaeSivapithecusLufengpithecusPongoGigantopithecusAnkarapithecus

Not all possible ranks are named here. The ranks ‘Eohominoidea’and ‘Euhominoidea’, hominoids of primitive and modern aspect,respectively, are informally used here. See text.

sivapithecus is east and dryopithecus is west, and never

Documents