A Synopsis of the Phylogeny and Paleobiology of Amphipithecidae, South Asian Middle and Late Eocene Primates RICHARD F

ANTHROPOLOGICAL SCIENCE Vol. 113, 33–42, 2005

A synopsis of the phylogeny and paleobiology of Amphipithecidae, South Asian middle and late Eocene primates RICHARD F. KAY1*

1Department of Biological Anthropology and Anatomy, Box 3170, Duke University Medical Center, Durham NC 27710, U.S.A.

Received 4 July 2003; accepted 7 May 2004

Abstract Amphipithecidae of late middle Eocene to late Eocene of Myanmar and Thailand is a phylogenetically enigmatic group that some place with Anthropoidea and others with Adapoidea. A linkage with adapoids is hard to demonstrate because it relies largely on a series of similarities that are arguably symplesiomorphies of Primates as a whole. The possibility that amphipithecids are specially related to crown anthropoids (e.g. Aegyptopithecus) is suggested by some shared-derived dental and gnathic anatomy. The postcranial anatomy indicates that the amphipithecids, if they are anthropoids, are probably a distantly related stem group outside the clade of African late Eocene-to-Recent anthropoids. Even the stem-group anthropoid status of amphipithecids is not supported by the absence of postorbital closure and enlarged olfactory bulbs, since postorbital closure and reduced olfactory bulbs characterize a more inclusive crown haplorhine clade of Tarsius plus Anthropoidea. An appealing possibility is that amphipithecids are basal haplorhines whose divergence would have predated the Tar- sius–Anthropoidea split. Larger amphipithecids equal or exceed the body size of the largest known Eocene primates. Dental and mandibular anatomy suggests these large-bodied amphipithecids were fruit and hard-object (nut) feeders. A more primitive contemporary amphipithecid, Myanmarpithecus, was smaller, about 1–2 kg, and its cheek teeth suggest a frugivorous diet but do not imply seed eating. The humerus and calcaneus of a large amphipithecid from Myanmar (Pondaungia or Amphipithecus) suggest a slow-moving arboreal quadrupedal locomotion like that of lorises. A talus of an amphipithcid is more suggestive of an active arboreal quadruped.

Key words: Amphipithecidae, Anthropoidea, Eocene, Thailand, Myanmar (Burma)

Introduction suggest a somewhat younger age than the Myanmar beds— 31 to 34 Ma (Benammi et al., 2001)—making it late Eocene Several clades of primate taxa are known from the middle to early Oligocene (Berggren et al., 1995). Thus, in total, and late Eocene of South Asia of which the Amphipithe- these extinct primates overlap and encompass the age distri- cidae is among the most diverse, including Pondaungia, bution of anthropoids described from Fayum province, Amphipithecus, Siamopithecus, and Myanmarpithecus. Egypt, and elsewhere in northern Africa and Arabia. Most of these taxa are known from teeth and jaws although Amphipithecids are known from three or perhaps four several specimens from Myanmar, assignable to Amphipith- species. From Thailand comes Siamopithecus eocaenus. ecus and Pondaungia, are known from parts of the humerus, From Myanmar come Amphipithecus mogaungensis and ulna, calcaneus, talus, face, and circumorbital region. In this Myanmarpithecus yarshensis. A third Myanmar taxon, paper I review the phylogenetic position of amphipithecids Pondaungia, seems to be represented by two species (P. cot- and, in somewhat more detail, comment on the paleobiology teri and P. savagei), although these could also represent a of this interesting clade. single sexually dimorphic species. One argument for the Myanmar amphipithecid specimens come from the former interpretation is that the ratio of the size of the lower Pondaung Formation. A fission-track age of ~37 Ma (late canine (or its root socket) to the dimensions of the molars is middle Eocene) is reported from one level in the Pondaung similar in the larger and smaller specimens. In contrast, Formation (Tsubamoto et al., 2002), concordant with the for- among extant sexually dimorphic primates, there is more mation being overlain by marine beds with foraminifera of size difference between the canines than between the molars late Eocene age (Aung, 1999; Mon, 1999). Thailand of the two sexes. amphipithecids come from coal deposits at Krabi. The mag- All amphipithecid species are known from the maxilla, netostratigraphy and paleofaunas with which it is associated mandible, and teeth, but cranial and postcranial evidence is sparser. Frontal bones of Amphipithecus have been * Corresponding author. e-mail: [email protected] described (Ciochon and Gunnell, 2002; Gunnell et al., 2002; phone: 1-919-684-2143; fax: 1-919-684-8542 Takai et al., 2003; Shigehara and Takai, 2004). Three associ- Published online 11 August 2004 ated postcranial fragments (humerus, ulna, and calcaneus) in J-STAGE (www.jstage.jst.go.jp) DOI: 10.1537/ase.04S005 belonging to a large-bodied species from Myanmar could be

© 2004 The Anthropological Society of Nippon 33 34 R.F. KAY ANTHROPOLOGICAL SCIENCE either Amphipithecus or Pondaungia (Ciochon et al., 2001; Simons and Rasmussen, 1996). A resolution of this problem Ciochon and Gunnell, 2004). A talus from the Pondaung awaits recovery of new material establishing the orbital Formation also is of a size appropriate to go with Amphipith- structure of eosimiid anthropoids. ecus (Marivaux et al., 2003). One of the amphipithecid frontal fragments mentioned above preserves the dorsal endocranial surface of the olfac- Phylogenetic Position of Amphipithecidae tory fossa and frontal pole of the cerebrum (Takai et al., 2003). The olfactory bulbs apparently were quite large— The phylogenetic position of amphipithecids is debated. within the range of extant strepsirrhines, and comparable to A cladistic analysis of dental, cranial, and postcranial anat- those of Eocene omomyids and adapoids, but distinctly omy by Kay et al. (2004c) examined the possibilities. Con- larger than Fayum anthropoids like Apidium and Aegyptop- sidering the collective evidence, Kay et al. (2004c) found it ithecus. Again, this reinforces the point of view that amphip- to be slightly more parsimonious (owing to dental charac- ithecids are more primitive than Fayum and more recent ters) to root amphipithecids within the anthropoid clade, anthropoids but does not rule out a haplorhine status. especially with Propliopithecidae, the earliest catarrhine In summary, the dental, cranial, and postcranial evidence anthropoids from the African Oligocene. Dental evidence point in very different directions: the dental evidence sug- provides the only support for a crown anthropoid hypothe- gests a link to crown anthropoids; the postcranial evidence sis. Some or all amphipithecids bear a strong resemblance to suggests that amphipithecids could be anthropoids but are propliopithecids, sharing with them deep jaws, spatulate and outside the more inclusive clade of crown and African stem enlarged upper central incisors, robust canines with a anthropoids; the orbital and olfactory anatomy suggests that rounded oval cross-section, P3 with a convex distal margin, the group, while possibly haplorhine is outside the crown p4 with a metaconid but lacking a paraconid, low crowned haplorhine clade. molars without paraconids, with little disparity between trig- A very different phylogenetic placement of the amphipi- onid and talonid heights, and the presence of wear facet X on thecid clade is as a sister group to one or another group of the molars (Ducrocq, 1998, 1999; Jaeger et al., 1998a, b; adapoid primates. The case for a relationship to notharctine Chaimanee et al., 2000; Shigehara et al., 2002; Kay et al., adapoids from North America was made by Gunnell, Cio- 2004c). However, a careful evaluation of the total available chon, and colleagues (Ciochon et al., 2001; Gunnell et al., anatomical evidence leads to the conclusion that a linkage 2002). Such an argument was based mainly on overall large with catarrhine anthropoids is unlikely because it would size and postcranial primitiveness. Additionally, synapo- require many convergences in the orbital anatomy (lack of morphies of the dentition were proposed, especially the pres- postorbital closure, a feature of stem anthropoids). Further- ence of a ‘pseudohypocone’ in amphipithecids and some more, the humerus and calcaneus lack the shared-derived advanced notharctines. Ducrocq (1999) and later Shigehara features characteristic of late Eocene and early Oligocene et al. (2002) reviewed the latter proposed similarity and dis- African oligopithecids, parapithecids, and propliopithecids counted it. (Kay and Williams, 1994; Ciochon and Gunnell, 2004; A possible phyletic link has been suggested between Seiffert et al., 2004). This interpretation is reinforced by a amphipithecids and non-notharctine adapoids of Europe like recently described amphipithecid talus which, while more Adapis, Leptadapis, Pronycticebus, and their relatives on diagnostically anthropoid, does not associate exclusively other continents such as Mahgarita (North America) and with crown anthropoids (Marivaux et al., 2003). Thus, this Aframonius (Africa). Kay et al. (2004c) reported that such a postcranial evidence in total indicates that the amphipithec- link is less parsimonious than the anthropoid hypothesis but ids might be stem anthropoids outside the clade of African has the advantage of not requiring parallel evolution of pos- Eocene-to-Recent crown anthropoids (Kay et al., 2004c). torbital closure. At that time, we indicated our preference for A major challenge for the anthropoid status of amphipith- this second, admittedly less parsimonious, interpretation that ecids comes from orbital and olfactory anatomy. The earliest amphipithecids are adapoids with dental and gnathic conver- primates had a postorbital bar but there was no bony parti- gences toward later larger-bodied Oligocene African anthro- tion between the orbit and the muscles of mastication (Mar- poids. However, as noted above, the recent discovery that tin, 1990). Adapoids, omomyids, and crown strepsirrhines the amphipithecid talus is distinctly anthropoid-like share this condition. In haplorhines, a bony partition sepa- (Marivaux et al., 2003) now leads me to conclude that rates the orbital contents from the muscles of mastication amphipithecids, like eosimiids, may well be stem anthro- (postorbital closure). Amphipithecus appears to have had the poids or at least stem haplorhines. plesiomorphic condition in having a postorbital bar but lacking postorbital closure (Shigehara and Takai, 2004; Takai Amphipithecid Paleobiology and Shigehara, 2004). Many authors argue that postorbital closure evolved just once at the base of the Body size Tarsius Anthropoid clade (e.g. Ross, 1994, 1996). If so, Published estimates of body size in amphipithecids are amphipithecids would be placed outside the crown hap- based on molar size (Table 1) (see also Egi et al., 2004) as lorhine clade, or postorbital closure was secondarily lost in well as estimates from humerus length (Ciochon et al., 2001) amphipithecids. An alternative possibility is that postorbital and talus size (Marivaux et al., 2003). The larger specimens closure evolved convergently in Tarsius and Anthropoidea of Pondaungia (specimens referred by some to the species as argued by Simons, Rasmussen, Beard, and others (Ras- Pondaungia savagei) had a mass of between 7.1 and 8.2 kg. mussen and Simons, 1992; Beard and MacPhee, 1994; Smaller specimens (for some P. cotteri sensu stricto) were Vol. 113, 2005 PHYLOGENY AND PALEOBIOLOGY OF AMPHIPITHECIDAE 35

Table 1. Estimated body mass (mean and range) of Amphipithecidae from m1 and m2 measurements Species All primate Prosimian Anthropoid All primate regressiona regressiona regressiona regressionb Pondaungia savagei (or larger morph) 8755 5348 9257 7058 (1787–42899) (1186–25351) (3617–23692) (6377–7894) Pondaungia cotteri (smaller morph) 4609 2993 5264 4970 (1059–20058) (718–13109) (2147–12902) (3611–6123) Amphipithecus mogaungensis 5153 3310 5806 5796 (1160–22890) (784–14700) (2351–14339) (5636–5876) Myanmarpithecus yarshensis — — — 1392 Siamopithecus eocaenus — — — 4932 a Based on m1 area; b based on m2 length. Body weight estimates were calculated from least-squares regression models predicting body mass from m1 area and m2 length found in Conroy (1987) and Kay and Simons (1980). Molar dimensions are from Takai et al. (2001). Specimens used in the analysis are enumerated in Kay et al. (2004b).

between 4.6 and 5.0 kg. If the two taxa prove to be sexual olfactory fossa, which housed the olfactory bulbs, is pre- morphs of one species, males would on average have been as served (Takai et al., 2003; Shigehara and Takai, 2004). The much as 60% larger than females. Such a high level of fossa appears to be larger than in living or fossil anthropoids dimorphism does occur among anthropoids, and might count but similar in relative size to those of living and fossil strep- as an added symplesiomorphy with anthropoids. sirrhines including Leptadapis (Kay and Cartmill, 1977; for The body size of Amphipithecus mogaungensis was discussion see Radinsky, 1977; Takai et al., 2003). The between 5.8 to 6.3 kg and that of Siamopithecus eocaenus implication is that the sense of smell may have been more was approximately 4.9 kg. These estimates place Pondaun- important as a mode of perception and communication in gia, Amphipithecus, and Siamopithecus among the largest amphipithecids than is the case for living anthropoids and of Eocene primates. The only other taxon that approaches them comparable importance to living strepsirrhines (Kay et al., in size is Leptadapis magnus from the late Eocene of 2004a, d). Europe; its size range was up to 8 kg (Gingerich, 1980). No late Eocene African primate approaches these amphipithec- Diet and feeding behavior ids in size (Kirk and Simons, 2000). The upper central incisor of Amphipithecus is proportion- The body size of Myanmarpithecus is estimated to be ately large and spatulate and has a buccolingually broad about 1.2 kg. This animal was comparable in size to Adapis root, suggesting that it was optimized to resist powerful buc- parisiensis (late Eocene, Europe). The body sizes of known colingual stresses engendered when it was used to separate a amphipithecids exceed those of extant insectivorous pri- bite of food (Shigehara et al., 2002). Apparently, Pondaun- mates (Kay, 1975; Gingerich, 1980; Kay and Covert, 1984). gia used its incisors for powerful incision, as do fruit-husk- On this basis we can rule out a primarily insectivorous diet ing anthropoids. for all known amphipithecids. In large-bodied amphipithecids [Amphipithecus, Pondaungia (Figure 1A), and Siamopithecus], the develop- Special senses ment of the shearing crests on the molars, expressed as The orbits of amphipithecids are known only in Amphipi- shearing quotients (SQs, see explanation in Table 2, Table 3) thecus, and even in that taxon, not sufficiently preserved to estimate the orbital diameter. Therefore, we hazard no infer- ence about the activity pattern of Amphipithecus. The absence of postorbital closure in Amphipithecus (see above) allows some inferences to be made about visual acuity. A functional link has been proposed between postorbital closure, a retinal fovea, and the absence of a tapetum lucidum (Cartmill, 1980; Ross and Hylander, 1996; Kirk and Kay, 2004), all adaptations for more acute vision in anthropoids than in extant strepsirrhines whose visual acuity, while excellent by mammalian standards, is less than in anthropoids (Kay and Kirk, 2000; Kirk and Kay, 2004). Therefore the absence of postorbital closure in Amphipithecus suggests this animal did not possess the acute vision present in modern anthropoids (Kay et al., 2004b). Indeed, the preserved parts of the orbits of Amphipithecus resemble those of Lept- adapis, an Eocene adapoid that apparently did not possess Figure 1. Box and whisker plots of the distribution of shearing quotients (SQs) of extant strepsirrhines, extant platyrrhines, and acute vision (Kay and Kirk, 2000). amphipithecids based on the frugivorous strepsirrhine model (after On the ventral surface of one Amphipithecus frontal, the Kay et al., 2004b). See Table 2, Table 3 for data and details. 36 R.F. KAY ANTHROPOLOGICAL SCIENCE

Table 2. Relative development of shearing crests on m2 in strepsirrhines and m1 in platyrrhines Group n Mean Standard error Range Leaf-eating strepsirrhines (based on m2 length) 8 16.78 3.82 6.23 to 38.32 Fruit- and gum-eating strepsirrhines 13 0.54 1.79 13.39 to 10.67 Insect eating strepsirrhines 5 22.64 4.19 11.80 to 37.54 Leaf-eating platyrrhines (based on m1 length) 4 21.58 4.95 17.34 to 28.28 Insect eating platyrrhines (based on m1 length) 3 17.68 2.83 12.69 to 22.47 Fruit- and gum-eating platyrrhines (based on m1 length) 8 1.80 1.80 7.22 to 6.54 Seed-eating platyrrhines (based on m1 length) 2 11.35 — 11.91 to 10.80 The estimate of shearing development is based on measurements of six lower molar crests on m1 (in platyrrhines) and m2 (in strepsirrhines) (see Kay, 1975, 1978 for anatomical details). A regression line was fitted to the natural log of molar length (ln m1L or ln m2L) versus the natural log of the sum of the measured crests (ln SH) for frugivorous strepsirrhines. The equation expressing the strepsirrhine line is: ln SH 1.0122(ln m2L) 0.5716. For each strepsirrhine taxon, the expected ln SH was calculated from this equation. The observed (measured) ln SH for each species was compared with the expected and expressed as a residual (Shear Quotient, or SQ): SQ 100 × (observed expected)/(expected). To calcu- late the expected shearing and SQs of platyrrhine species, values for m1 length and sum of m1 shearing crest lengths were substituted in the above formula. Note that the mean SQs for platyrrhine frugivores are very similar to those of strepsirrhine species. Likewise, paired comparisons of mean SQs for leaf-eating strepsirrhines and platyrrhines and insect-eating strepsirrhines and platyrrhines are very similar. These paired similarities indicate that the strepsirrhine model is robust and not skewed as a consequence of using a different tooth (m1 vs. m2), or by phylogenetic effects. Pos- itive SQ values indicate a degree of shearing capacity greater than expected for a frugivorous strepsirrhine. Lepilemur was deleted from the sample because its teeth are exceptionally long and narrow, thus leading to an overestimate of the ‘size’ corrector, and a corresponding underestimate of the SQ (for discussion, see Kay, 1975, 1978).

Table 3. Relative development of shearing crests on m1 or m2 in amphipithecids using the strepsirrhine and platyrrhine models Taxon Specimen n m1 or m1 or SQ from number m2 length m2 shear strepsirrhine model Amphipithecus mogaungensis m1 NMMP 6, 7 1 5.75 9.36 9.95 Amphipithecus mogaungensis m2 NMMP 6 1 5.51 8.93 10.36 Pondaungia cotteri m1 NMMP 30 1 5.51 9.70 2.54 Pondaungia cotteri m2 NMMP 30 1 5.64 9.58 6.10 Pondaungia savagei m2 NMMP 1, 3 2 6.49 9.64 18.00 Myanmarpithecus yarshensis m2 NMMP 8/9/10/11 1 3.97 7.43 3.82 Siamopithecus eocaenus m2 TF 3634 1 5.88 8.61 19.04 The strepsirrhine and platyrrhine models are those described in Table 2.

are poorly developed. The molar shear development among ecus, and Pondaungia are not fused, but the symphyseal sur- these amphipithecids is similar to that of living strepsir- face is rugose with substantial relief (Figure 3). In spite of rhines and platyrrhines that have low-fiber diets (fruit- and the lack of fusion, this arrangement would have allowed the seed-eaters) and far less than molar SQs of living species efficient transfer of muscle forces from the balancing to the that have higher-fiber diets. Shearing development in large- working side of the mandible. In light of the large spatulate bodied amphipithecids is especially similar to that of extant upper incisors, poorly developed molar shearing, and (in seed-eating platyrrhines (Figure 2). Pondaungia) thick molar enamel, this symphyseal construc- Myanmarpithecus has better developed shearing than tion suggests hard or tough-object feeding. other amphipithecids; nevertheless, it exhibits far less shearing than do extant folivorous strepsirrhines and platyrrhines. Locomotion For this taxon, a mixed diet is inferred, consisting primarily Humeral, ulnar, and calcaneal remains of a single individ- of fruit with substantial components of leaves or insects as a ual (NMMP 20) and a talus of another are referable to protein source. Amphipithecidae. NMMP 20 was a slow-moving arboreal The cheek teeth of living primates and other mammals quadruped like a modern lorisid and the late Eocene Euro- that specialize in eating hard seeds or in splitting open tough, pean adapoids Adapis and Leptadapis (Dagosto, 1983), but hard fruits often have thick enamel (Kay, 1981). While pre- quite unlike the middle Eocene North American adapoids cise measurement of enamel thickness is not possible for any Notharctus or Smilodectes (Dagosto, 1993; Schmitt, 1996). amphipithecid, one broken specimen of Pondaungia In the proximal humerus the tuberosities are positioned well (NMMP 12) reveals that its enamel is very thick (Figure 2B, below the summit of the articular surface of the humeral C). This suggests that Pondaungia incorporated substantial head. In active arboreal and terrestrial quadrupedal primates quantities of very hard objects such as seeds encased in hard (AAC), and primates that engage in vertical clinging behav- shells in its diet, a conclusion that is consistent with infer- iors (in association with leaping) (hereafter VCL; Figure ences about powerful incisal biting and poorly developed 4C), the tuberosities for insertion of the rotator cuff muscles molar shearing. rise to the same level as, or are superior to, the summit of the The mandibular symphyses of Siamopithecus, Amphipith- glenohumeral articular surface. In contrast, living slow- Vol. 113, 2005 PHYLOGENY AND PALEOBIOLOGY OF AMPHIPITHECIDAE 37

Figure 3. Mandibular symphysis of Amphipithecus mogaungensis in medial view showing the rugose unfused symphyseal surface (after Kay et al., 2004b).

Figure 2. Pondaungia dentition. (A) Occlusolateral view of Pondaungia cotteri (NMMP 1), showing the poorly developed shearing crests and crenulated enamel characteristic of the species; (B) Pondaungia cotteri (NMMP-KU0003), medial view of M1 and broken M2; and (C) traced outline of the occlusal surface and exposed den- tine-enamel junction along the natural break of M2 (after Kay et al., Figure 4. Medial views of the proximal humeri of (A) Nycticebus 2004b). coucang, a slow-moving arboreal quadruped; (B) NMMP 20; and (C) Propithecus verreauxi, a vertical clinging and leaping species. The dashed line illustrates the position of the tuberosities. The median ori- entation of the glenohumeral articular surface is depicted by the solid moving arboreal quadrupeds (Figure 4A) have the tuberosi- line (after Kay et al., 2004b). ties positioned below the summit, that is to say, the humeral head projects distinctly superior to the tuberosities (Jolly, 1967; Gebo, 1988; Rose, 1988; Harrison, 1989). more cranially) than in most AAQ and VCL primates As illustrated in Figure 4, the proximal articular surface of (Schmitt, 1996). the humeral head of NMMP 20 is oriented cranially in a NMMP 20 has a nearly equal mediolateral breadth rela- manner most like that of a loris or Alouatta, the howler mon- tive to its proximodistal length and is relatively flat when its key (Schon-Ybarra, 1998). This goes along with the posi- depth is compared either with its proximodistal length or tioning of the tuberosities below the articular summit to mediolateral length. Primates that engage in habitual VCL indicate that the shoulder joint was optimized for overhead are separated from AAQ species and extant cautious arbo- reaching and bridging (Jolly, 1967; Walker, 1974; Gebo, real quadrupeds as well (Loris, Nycticebus, Perodicticus, 1988) and had reduced stability in protracted positions and Alouatta) (Schmitt, 1996; Kay et al., 2004b). In indices (Rose, 1989; Schmitt, 1996). In contrast, the articular sur- of humeral head shape, NMMP 20 falls among the slow- face is oriented more acutely to the long axis of the shaft (i.e. moving arboreal quadrupeds (Figure 5). 38 R.F. KAY ANTHROPOLOGICAL SCIENCE

Figure 6. Morphology of the NMMP 20 humerus. (A) Humeral Figure 5. Comparative morphology of the proximal humerus. reconstruction after Ciochon et al. (2001) and the outline of the natural Means and ranges of (A) humeral head shape, and (B) humeral head mid-shaft cross-section showing the distribution of cortical bone and inflation in slow climbers (SC), active arboreal quadrupeds (AAQ), medular cavity. (B) Anteroposterior values for the quantity K (relative and vertical clingers and leapers (VL). Humeral head shape measure- cortical thickness, described in the text and Table 4) for samples of ments are the length of the base (cord) of a contour along the central active arboreal quadrupeds, slow climbers, and vertical clingers and proximodistal (superior-inferior) surface of the humeral head (PD), the leapers (data from Runestad, 1994). same for a central mediolateral contour (ML), and the central height of the ML contour from its base (AH). The sample includes five species of active arboreal quadrupeds, six species of vertical clingers and leapers, and four species of slow moving quadrupeds (for details see Kay similarities to lorises. In particular, the morphology of the et al., 2004b). radiohumeral and ulnohumeral joints are well suited to pro- vide stability in habitually flexed postures and shows the greatest functional similarity to extant lorises and howler monkeys (Alouatta). The capitulum is large and rounded, The cortical walls of the humeral shaft of NMMP 20 (Fig- and distinctly separated from the trochlea by a narrow, shal- ure 6) were exceptionally thick compared with most living low zona conoidea. The capitulum has a well-developed strepsirrhines and most closely approximate those of slow- anterolateral flange (‘tail’) with a distinct and deep ridge moving arboreal quadrupeds (Kay et al., 2004b). The inter- (Figure 7). This design provides stability for the radius in nal and external dimensions of the humeral shaft give an flexed postures and is best developed in slow climbers like estimate of cortical area (CA); the ratio of internal diameter Nycticebus and Perodicticus (Szalay and Dagosto, 1980; to external diameter (K) can also be estimated (Table 4). Val- Rose, 1993). A distinct capitular ‘tail’ with a deep ridge is ues for CA and K in NMMP 20 are larger for this specimen also found in Alouatta (Schon-Ybarra, 1998). Taken as a than for any leaper or active arboreal quadruped of similar whole, the large, round capitulum, truncated cone-like tro- body size (Kay et al., 2004b). CA of NMMP 20 is most sim- chlea, and the strong capitular tail suggest cautious arboreal ilar to Adapis and Leptadapis (Runestad, 1994) and extant quadrupedalism. This is consistent with the clear signal from slow-climbing lorises (Runestad, 1997). The values for K the proximal humerus and humeral shaft. show that this specimen has extraordinarily thick cortical The NMMP 20 calcaneus does not exhibit the extreme bone, approached only by lorises and Daubentonia. distal elongation of Tarsius. Its proportions are more remi- The elbow joint of NMMP 20 shows additional functional niscent of African late Eocene to early Oligocene anthro- Vol. 113, 2005 PHYLOGENY AND PALEOBIOLOGY OF AMPHIPITHECIDAE 39

Table 4. Humeral cortical properties

Species Body Cortical K a-p K m-l Loco- weight area motion Cheirogaleus major 436 5.60 0.71 0.58 AAC Cheirogaleus medius 180 3.14 0.72 0.67 AAC Varecia variegata 3000 26.98 0.70 0.69 AAC Otolemur 1200 12.47 0.65 0.63 AAC crassicaudatus Lemur catta 2423 15.45 0.71 0.70 AAC Nycticebus coucang 658 12.00 0.56 0.59 SC Perodicticus potto 860 13.80 — — SC Loris tardigradus 300 3.50 0.56 0.53 SC Daubentonia 2700 26.40 0.60 0.66 SC madagascarensis Propithecus verrauxi 5794 33.70 0.65 0.63 VCL Indri indri 7500 37.94 0.66 0.67 VCL Avahi laniger 1175 9.28 0.70 0.68 VCL NMMP 20 5500 39.00 0.52 0.46 — Notharctus 2700 24.00 — — — Smilodectes 2400 25.00 — — — Leptadapis 10000 72.00 — — — Adapis 1 2000 20.00 — — — Adapis 2 3500 34.00 — — — Cortical area in NMMP 20 was measured using the methods of Runestad (1994, 1997). The quantity ‘K’ is the ratio of internal diameter to external diameter estimated following the methods of Currey and Alexander (1985) and Demes and Jungers (1993). Data for extant species and adapoids from Runestad (1994). AAQ, active arboreal quadrupeds; VCL, vertical clingers and leapers; SC, slow arboreal clamberers. poids and Eocene North American and European adapoids. These findings are not especially informative as to locomotion. Extant primates exhibiting a similar degree of foot elongation have a wide spectrum of locomotion. The talus of a large-bodied amphipithecid (NMMP 39) not associated with dental remains has recently been described by Marivaux et al. (2003). Using size, Marivaux et al. allocate this specimen to Amphipithecus. However, given the size overlap between Amphipithecus and smaller specimens of Pondaungia, allocation of NMMP 39 to either taxon would appear plausible. Marivaux et al. (2003) reported that NMMP-39 exhibits a straight and moderately long talar neck, a fairly high talar body, and parallel-sided medial and lateral trochlear facets (not wedged-shaped) with circular rims, all of which features might be indicative of leaping. However, they note that (1) the trochlea is quite flattened rather than deeply grooved Figure 7. Distal humeral fragment of NMMP 20, posterior (A), as in most leapers, where only one primary plane of move- anterior (B) and distal (C) views. ment is needed at the talocrural joint; (2) the medial trochlear rim has a long radius of curvature; and (3) the posterior trochlear shelf is absent. These three features suggest much more mobility at the tibiotalar joint than typically Conclusions seen in specialized leaping primates. They suggest that Amphipithecidae is a phylogenetically enigmatic group NMMP-39 might have belonged to an arboreal quadruped, that some place among Anthropoidea and others with not an extreme leaper. Adapoidea. Dental evidence is often cited in support of the Thus there is a discrepancy between the slower quadrupe- anthropoid status of amphipithecids. As noted above, dalism inferred from the humerus (NMMP 20) and active amphipithecids bear a strong resemblance to propliopithec- above-branch quadrupedalism inferred from the talus ids of the African Oligocene, However, it appears that many (NMMP-39). A plausible but presently untestable possibility of the apparent crown anthropoid dental synapomorphies is that we might be dealing with the skeletal remains of two seem to have evolved in several lines of anthropoids. Dis- different amphipithecid genera. similarities between the amphipithecid humerus and calcaneus and those of late Eocene and early Oligocene African oligopithecids, parapithecids, and propliopithecids suggest 40 R.F. KAY ANTHROPOLOGICAL SCIENCE amphipithecids may instead lie outside a more encompass- Kay et al. (2004b) on the paleobiology of this interesting ing clade of these stem anthropoids (see reviews by Ciochon clade. I am indebted to the coauthors of those papers for and Gunnell, 2004; Seiffert et al., 2004). An amphipithecid many of the insights presented here: J. Perry, C. Ross, D. talus shares a number of derived similarities with anthro- Schmitt, N. Shigehara, M. Takai, N. Egi, C. Vinyard, and B. poids from the late Eocene and early Oligocene of Africa but Williams. I thank N. Shigehara, M. Takai, H. Hongo, and M. not exclusively with crown anthropoids. At the very least, Hoffman of the Primate Research Institute of Kyoto Univer- this postcranial evidence indicates that the amphipithecids sity for their invitation to attend the conference held in may be stem anthropoids but are outside the clade of the Inuyama, Japan in 2003 at which I presented a version of this African Eocene-to-Recent anthropoids. paper. Postorbital closure is thought to be a key adaptive innova- tion related to visual acuity, often assumed to have evolved only once at the base of the Tarsius Anthropoid clade (e.g. References Ross, 1994, 1996). The frontal bone of Amphipithecus indi- Aung A.K. (1999) Revision of the stratigraphy and age of the pri- cates that postorbital closure was absent or incomplete. mate-bearing Pondaung Formation. In: Tun T. (ed.), Proceed- Therefore, if amphipithecids are anthropoids, postorbital ings of the Pondaung Fossils Expedition Team. Office of closure must have evolved independently twice (Beard and Strategic Studies, Ministry of Defense, Yangon, Myanmar, MacPhee, 1994), or evolved and then been lost in the anthro- pp. 131–151. poid lineage. Beard K.C. and MacPhee R.D.E. (1994) Cranial anatomy of Shos- The upper and lower teeth, mandibular structure, as well honius and the antiquity of Anthropoidea. In: Fleagle J.G. and Kay R.F. (eds.), Anthropoid Origins. Plenum Press, New as humeral, talar, and calcaneal fragments and the talus pro- York, pp. 55–98. vide detailed evidence for the adaptive profile of amphipith- Benammi M., Chaimanee Y., Jaeger J.-J., Suteethorn V., and ecids. The larger-bodied amphipithecids were as large as any Ducrocq R.-M. (2001) Eocene Krabi basin (southern Thai- known Eocene primates, and comparable in size to the larg- land): Paleontology and magnetostratigraphy. Geological est extant platyrrhines and strepsirrhines. The mandibular Society of America Bulletin, 113: 265–273. corpora of these large-bodied amphipithecids suggest an Berggren W.A., Kent D.V., Swisher C.C. III, and Aubry M.-P. ability to resist large chewing loads and to transfer muscle (1995) A revised Cenozoic geochronology and chronostratig- raphy. In: Berggren W.A., Kent D.V., and Aubry M.-P. (eds.), forces from the balancing side to the working side of the jaw, Geochronology, Time Scales, and Global Stratigraphic Corre- thus increasing the muscle force available for mastication. lation. Society for Sedimentary Geology Special Publication, The robust, spatulate upper central incisor and projecting 54, Tulsa, Oklahoma, pp. 129–212. robust upper and lower canines show that they used the ante- Cartmill M. (1980) Morphology, function and evolution of the rior teeth for powerful separation of food items, as occurs in anthropoid postorbital septum. In: Ciochon R.L. and Chiarelli fruit husking by living anthropoids. The molars have weak A.B. (eds.), Evolutionary Biology of the New World Mon- keys and Continental Drift. Plenum Press, New York, pp. shearing crests, indicating a low-fiber diet of fruits or seeds. 243–274. The presence of thick enamel in Pondaungia suggests a Chaimanee Y., Thein T., Ducrocq S., Soe A.N., Benammi M., Tun hard-object, low-fiber diet, possibly seed predation. The T., Lwin T., Wai S., and Jaeger J.-J. (2000) A lower jaw of smaller and more primitive amphipithecid Myanmarpithe- Pondaungia cotteri from the late middle Eocene Pondaung cus weighed 1–2 kg. Its cheek teeth suggest a frugivorous Formation (Myanmar) confirms its anthropoid status. Pro- diet. ceedings of the National Academy of Sciences of the United States of America, 97: 4102–4105. The humerus, ulna, and calcaneus of one large amphipith- Ciochon R. and Gunnell G.F. (2002) Chronology of primate dis- ecid species suggest that this animal was an above-branch coveries in Myanmar: influences on the anthropoid origins quadruped, and most likely a slow, cautious one. The debate. Yearbook of Physical Anthropology, 45: 2–35. humeral structure resembles extant slow-moving primate Ciochon R. and Gunnell G.F. (2004) Eocene large-bodied primates quadrupedal species like the living lorises and Alouatta. The of Myanmar and Thailand: morphological considerations and shoulder structure suggests a wide range of motion including phylogenetic affinities. In: Ross C.F. and Kay R.F. (eds.), overhead reaching. Midshaft bone strength was exceptional Anthropoid Origins: New Visions. Kluwer Academic/Plenum Publishers, New York, pp. 249–282. and elbow stability was enhanced in habitually flexed posi- Ciochon R., Gingerich P.D., Gunnell G.F., and Simons E.L. (2001) tions. No features suggest rapid above-branch locomotion Primate postcrania from the late middle Eocene of Myanmar. nor a vertical clinging and leaping or simple leaping loco- Proceedings of the National Academy of Sciences of the motion. A talus of another large amphipithecid of uncertain United States of America, 98: 7672–7677. taxonomic assignment lacks certain features associated with Conroy G.C. (1987) Problems of body-weight estimation in fossil vertical postures or springing. However, it was tightly artic- primates. International Journal of Primatology, 8: 115–137. Currey J.D. and Alexander R.M. (1985) The thickness of the walls ulated with the tibia making a stable ankle joint, features of tubular bones. Journal of Zoology (London), 206: 453– consistent with rapid arboreal quadrupedalism (Marivaux et 468. al., 2003). Dagosto M. (1983) Postcranium of Adapis parisiensis and Lept- adapis magnus (Adapiformes, Primates). Folia Primatolog- ica, 41: 49–101. Acknowledgments Dagosto M. (1993) Postcranial anatomy and locomotor behavior in Eocene primates. In: Gebo, D.L. (ed.), Postcranial Adaptation This paper draws from and summarizes the results of two in Nonhuman Primates. Northern Illinois University Press, investigations as they relate to the Amphipithecidae—that of De Kalb, Illinois, pp. 199–219. Kay et al. (2004c) on the phylogeny of Anthropoidea and of Demes B. and Jungers W.L. (1993) Long bone cross-sectional Vol. 113, 2005 PHYLOGENY AND PALEOBIOLOGY OF AMPHIPITHECIDAE 41

dimensions, locomotor adaptations and body size in prosim- origins. In: Fleagle J.G. and Kay R.F. (eds.), Anthropoid Ori- ian primates. Journal of Human Evolution, 25: 57–74. gins. Plenum Press, New York, pp. 361–445. Ducrocq S. (1998) Eocene primates from Thailand: are Asian Kay R.F., Rossie J.B., Colbert M.W., and Rowe T. (2004a) Obser- anthropoideans related to African ones? Evolutionary Anthro- vations on the olfactory system of Tremacebus harringtoni pology, 7: 97–104. (Platyrrhini, early Miocene, Sacanana, Argentina) based on Ducrocq S. (1999) Siamopithecus eocaenus, a late Eocene anthro- high resolution X-ray CT scans. American Journal of Physi- poid primate from Thailand: its contribution to the evolution cal Anthropology, Supplement 35: 123–124. of anthropoids in Southeast Asia. Journal of Human Evolu- Kay R.F., Schmitt D., Vinyard C., Perry J. M.G., Takai M., Shige- tion, 36: 613–636. hara N., and Egi N. (2004b) The paleobiology of Amphipithe- Egi N., Tun S.T., Takai M., Shigehara N., and Tsubamoto T. (2004) cidae, South Asian late Eocene primates. Journal of Human Relative geographical and body size distributions of the Evolution, 46: 3–25. Pondaung primates with a comment on the taxonomic assign- Kay R.F., Williams B.A., Ross C., Takai M., and Shigehara N. ment of NMMP 20, a postcranium of Amphipithecidae. (2004c) Anthropoid origins: a phylogenetic analysis. In: Ross Anthropological Science, 112: 67–74. C.F. and Kay R.F. (eds.), Anthropoid Origins: New Visions. Gebo D. (1988) A hominoid proximal humerus from the Early Kluwer Academic/Plenum Publishers, New York, pp. 91– Miocene of Rusinga Island, Kenya. Journal of Human Evolu- 135. tion, 17: 393–401. Kay R.F., Campbell V.M., Rossie J.B., Colbert M.W., and Rowe T. Gingerich P. (1980) Eocene Adapidae, paleobiology, and the origin (2004d) The olfactory fossa of Tremacebus harringtoni of South American Platyrrhini. In: Ciochon R.L. and Chiarelli (Platyrrhini, early Miocene, Sacanana, Argentina): implica- A.B. (eds.), Evolutionary Biology of the New World Mon- tions for activity pattern. Anatomical Record (in press). keys and Continental Drift. Plenum Press, New York, pp. Kirk E.C. and Kay R.F. (2004) The evolution of high visual acuity 123–138. in the Anthropoidea. In: Ross C.F. and Kay R.F. (eds.), Gunnell G.F., Ciochon R., Gingerich P.D., and Holroyd P.A. (2002) Anthropoid Origins: New Visions. Kluwer Academic/Plenum New assessment of Pondaungia and Amphipithecus (Pri- Publishers, New York, pp. 539–602. mates) from the late middle Eocene of Myanmar, with a com- Kirk E.C. and Simons E.L. (2000) Diet of fossil primates from the ment on ‘Amphipithecidae’. Contributions of the Museum of Fayum Depression of Egypt: a quantitative analysis of molar Paleontology, the University of Michigan, Volume 30, pp. shearing. Journal of Human Evolution, 40: 203–229. 337–372. Marivaux L., Chaimanee Y., Ducrocq S., Marandat B., Sudre J., Harrison T. (1989) New postcranial remains of Victoriapithecus Soe A.N., Tun S.T., Htoon W., and Jaeger J.-J. (2003) The from the middle Miocene of Kenya. Journal of Human Evolu- anthropoid status of a primate from the late middle Eocene tion, 18: 3–54. Pondaung Formation (Central Myanmar): tarsal evidence. Jaeger J.-J., Chaimanee Y., and Ducrocq S. (1998a) Origin and Proceedings of the National Academy of Sciences of the evolution of Asian hominoid primates. Paleontological data United States of America, 100: 13173–13178. versus molecular data. Comptes Rendus de l’Académie des Martin R.D. (1990). Primate Origins and Evolution: A Phyloge- Sciences, Série 3, 321: 73–78. netic Reconstruction. Chapman and Hall, London. Jaeger J.-J., Soe U.A.N., Aung U.A.K., Benammi M., Chaimanee Mon H. (1999) Nannopaleontological analysis of the rock samples Y., Ducrocq R.-M., Tun T., Thein U.T., and Ducrocq S. collected by the Pondaung Fossils Expedition Team. In: Tun (1998b) New Myanmar middle Eocene anthropoids. An T. (ed.), Proceedings of the Pondaung Fossil Expedition Asian origin for catarrhines? Comptes Rendus de l’Académie Team. Office of Strategic Studies, Ministry of Defense, des Sciences, Série 3, 321: 953–959. Yangon, pp. 94–121. Jolly C.J. (1967) The evolution of the baboons. In: Vartborg, H. Radinsky L. (1977) Early primate brains: fact and fiction. Journal (ed.), The Baboon in Medical Research: Proceedings of the of Human Evolution, 6: 79–86. First International Symposium on the Baboon and Its Use as Rasmussen D.T. and Simons E.L. (1992) Paleobiology of the an Experimental Animal, 2. University of Texas Press, Aus- oligopithecines, the earliest known anthropoid primates. tin, pp. 23–50. International Journal of Primatology, 13: 1–32. Kay R.F. (1975) The functional adaptations of primate molar teeth. Rose M.D. (1988) Another look at the anthropoid elbow. Journal American Journal of Physical Anthropology, 43: 195–216. of Human Evolution, 17: 193–224. Kay R.F. (1978) Molar structure and diet in extant Cercopithe- Rose M.D. (1989) New postcranial specimens of catarrhines from coidea. In: Butler P.M. and Joysey K.A. (eds.), Development, the middle Miocene Chinji formation, Pakistan: descriptions Function and Evolution of Teeth. Academic Press, London, and a discussion of the proximal humeral functional morphol- pp. 309–339. ogy in anthropoids. Journal of Human Evolution, 18: 131– Kay R.F. (1981) The nut-crackers: a new theory of the adaptations 162. of the Ramapithecinae. American Journal of Physical Anthro- Rose M.D. (1993) Functional anatomy of the elbow and forearm in pology, 55: 141–152. primates. In: Gebo, D.L. (ed.), Postcranial Adaptation in Non- Kay R.F. and Cartmill M. (1977) Cranial morphology and adapta- human Primates. Northern Illinois University Press, De Kalb, tions of Palaechthon nacimienti and other Paromomyidae Illinois, pp. 70–95. (Plesiadapoidea? Primates), with a description of a new genus Ross C.F. (1994) The craniofacial evidence for anthropoid and tar- and species. Journal of Human Evolution, 6: 19–35. sier relationships. In: Fleagle J.G. and Kay R.F. (eds.), Kay R.F. and Covert H.H. (1984) Anatomy and behaviour of Anthropoid Origins. Plenum Press, New York, pp. 469–547. extinct primates. In: Chivers D.J., Wood B.A., and Alan Bils- Ross C.F. (1996) An adaptive explanation for the origins of the borough, (ed.), Food Acquisition and Processing in Primates. Anthropoidea (Primates). American Journal of Primatology, Plenum Press, New York, pp. 467–508. 40: 205–230. Kay R.F. and Kirk E.C. (2000) Osteological evidence for the evo- Ross C.F. and Hylander W.L. (1996) In vivo and in vitro bone lution of activity pattern and visual acuity in primates. Ameri- strain in owl monkey circumorbital region and the function of can Journal of Physical Anthropology, 113: 235–262. the postorbital septum. American Journal of Physical Anthro- Kay R.F. and Simons E.L. (1980) The ecology of Oligocene Afri- pology, 101: 183–215. can Anthropoidea. International Journal of Primatology, 1: Runestad J.A. (1994). Humeral and femoral diaphyseal cross-sec- 21–37. tional geometry and articular dimensions in Prosimii and Kay R.F. and Williams B.A. (1994) Dental evidence for anthropoid Platyrrhini (Primates) with application for reconstruction of 42 R.F. KAY ANTHROPOLOGICAL SCIENCE

body mass and locomotor behavior in Adapidae (Primates; Simons E.L. and Rasmussen D.T. (1996) Skull of Catopithecus Eocene). Ph.D. dissertation, Johns Hopkins University School browni, an early Tertiary Catarrhine. American Journal of of Medicine, Baltimore. Physical Anthropology, 100: 261–292. Runestad J.A. (1997) Postcranial adaptations for climbing in Szalay F.S. and Dagosto M. (1980) Locomotor adaptations as Lorisidae (Primates). Journal of Zoology (London), 242: reflected on the humerus of Paleogene Primates. Folia Prima- 261–290. tologica, 34: 1–45. Schmitt D. (1996) Humeral head shape as an indicator of locomo- Takai M. and Shigehara N. (2004) New specimens of the tor behavior in extant strepsirrhines and Eocene adapids. Pondaung primates (latest middle Eocene, central Myanmar). Folia Primatologica, 67: 137–151. In: Ross C.F. and Kay R.F. (eds.), Anthropoid Origins: New Schon-Ybarra M.A. (1998) Arboreal quadrupedalism and forelimb Visions. Kluwer Academic/Plenum Press, New York, pp. articular anatomy of red howlers. International Journal of Pri- 323–340. matology, 19: 599–613. Takai M., Shigehara N., Aung A.K., Soe A.N., Tun S.T., Tsubam- Seiffert E.R., Simons E.L., and Simons C.V.M. (2004) Phyloge- oto T., and Thein T. (2001) A new anthropoid from the latest netic, biogeographic, and adaptive implications of new fossil middle Eocene of Pondaung, central Myanmar. Journal of evidence bearing on crown anthropoid origins and early stem Human Evolution, 40: 393–409. catarrhine evolution. In: Ross C.F. and Kay R.F. (eds.), Takai M., Shigehara N., Egi N., and Tsubamoto T. (2003) Endocra- Anthropoid Origins: New Visions. Kluwer Academic/Plenum nial cast and morphology of the olfactory bulb of Amphipithe- Publishers, New York, pp. 157–182. cus mogaungensis (latest middle Eocene of Myanmar). Shigehara N. and Takai M. (2004) Facial fragments of Amphipith- Primates, 44: 137–144. ecus. In: Ross C.F. and Kay R.F. (eds.), Anthropoid Origins: Tsubamoto T., Takai M., Shigehara N., Egi N., Aung A.K., Maung New Visions. Kluwer Academic/Plenum Publishers, New M., Danhara T., and Suzuki H. (2002) Fission-track zircon York, pp. 323–340. age of the Eocene Pondaung Formation, Myanmar. Journal of Shigehara N., Takai M., Kay R.F., Aung A.K., Soe A.N., Tsubam- Human Evolution, 42: 361–369. oto T., and Thein T. (2002) The upper dentition and face of Walker A. (1974) Locomotor adaptations in past and present pros- Pondaungia cotteri from central Myanmar. Journal of Human imian primates. In: Jenkins, F.A. (ed.), Primate Locomotion. Evolution, 42: 143–166. Academic Press, New York, pp. 349–381.