Plio-Pleistocene Eagle Predation on Fossil Cercopithecids from the Humpata Plateau, Southern Angola

AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 139:421–429 (2009)

Brief Communication: Plio-Pleistocene Eagle Predation on Fossil Cercopithecids From the Humpata Plateau, Southern Angola

Christopher C. Gilbert,1,2* W. Scott McGraw,3 and Eric Delson4,5,6

1Department of Anthropology, Yale University, New Haven, CT 06520-8277 2Yale Institute for Biospheric Studies, Yale University, New Haven, CT 06520-8277 3Department of Anthropology, The Ohio State University, Columbus, OH 43210-1106 4Department of Anthropology, Lehman College/CUNY, Bronx, NY 10468 5Department of Vertebrate Paleontology, American Museum of Natural History, New York, NY 10024 6New York Consortium in Evolutionary Primatology (NYCEP)

KEY WORDS Taung; Tai; primate evolution; South Africa; West Africa

ABSTRACT Recent studies suggest a large raptor compare the Angolan and Taung material to remains of such as the crowned eagle (Stephanaoetus coronatus) extant primates killed by crowned eagles in the Ivory was responsible for collecting at least a portion of the Coast’s Tai National Park. Our analysis indicates that primate fauna from the South African fossil site of the size distribution and composition of fauna from the Taung, including its lone hominin specimen. This tapho- localities is quite similar and that there are striking con- nomic signature at Taung is currently regarded as a sistencies in damage to the crania from each site. The unique and, most likely, isolated case in primate and absence of large bodied (>20 kg) primates and other human evolution. However, the activities of large, car- mammalian taxa at the Taung hominin locality and Tai, nivorous birds should also be detectable at other primate and their rarity in Angola, combined with the strong fossil localities in Africa if raptors have been a strong likelihood that raptor nests were positioned near ﬁssure selective force throughout primate evolution. Over the openings at both fossil localities, provides additional sup- last 60 years, a collection of extinct cercopithecids has port for eagle involvement. On the basis of this evidence, been assembled from several cave breccias on the Hum- we conclude that at least some of the Angolan cerco- pata Plateau in southern Angola. The material, dated pithecids were most likely raptor prey and hypothesize near the Plio-Pleistocene boundary, includes an assort- that raptor predation has been a strong and perhaps ment of craniodental and postcranial remains variably underappreciated selective force during the course of pri- assigned to Papio (Dinopithecus) cf. quadratirostris, mate evolution. Am J Phys Anthropol 139:421–429, Parapapio, Cercopithecoides, and Theropithecus.We 2009. VC 2009 Wiley-Liss, Inc.

Several recent studies suggest a large raptor was re- 1. The body sizes of animals in the Taung assemblage sponsible for collecting at least a portion of the prima- are homogeneous and dominated by individuals\20 kg. tes from the South African fossil site of Taung, includ- 2. Crania from Taung are relatively complete. These ing its lone hominin specimen, the juvenile holotype of specimens frequently have holes in the vault and occi- Australopithecus africanus (Berger and Clarke, 1995; pitals are commonly absent. The brain or endocast is McGraw et al., 2006; Berger, 2006; Berger and often exposed. There is no evidence of chewing. McGraw, 2007). Currently, this taphonomic signature 3. Nontypical carnivoran prey (e.g., tortoise shells) are at Taung appears to be a unique and isolated case in present and show no evidence of chewing. primate and human evolution. However, if large, car- nivorous birds have been a strong selective force throughout primate evolution, then their activities should also be detectable at other primate fossil local- Grant sponsors: Leakey Foundation, Stony Brook University, the ities in Africa. Here we present evidence of probable Yale Institute for Biospheric Studies, the National Science Founda- raptor predation in a collection of cercopithecid fossils tion, Yerkes National Primate Research Center, Ohio State Univer- from Angola’s Humpata Plateau. sity, PSC-CUNY, Lehman College, and NYCEP. *Correspondence to: Christopher C. Gilbert, Department of Anthropology and Yale Institute for Biospheric Studies, P.O. Box 208277, Yale University, New Haven, CT 06520-8277, USA. BACKGROUND E-mail: [email protected] The Raptor Hypothesis Received 17 July 2008; accepted 26 November 2008 The raptor hypothesis (Berger and Clarke, 1995) iden- tiﬁed six features of the Taung fauna consistent with the DOI 10.1002/ajpa.21004 implication of raptors (rather than carnivorans) as the Published online 24 February 2009 in Wiley InterScience main accumulating agents: (www.interscience.wiley.com).

VC 2009 WILEY-LISS, INC. 422 C.C. GILBERT ET AL. 4. Several large bird eggshells are present. Berger and Clarke (1995) and Hrdlicka (1925) suggest that eggs similar in size to goose eggs were recovered during excavation. 5. Except for the Taung Australopithecus child, no hominins are present in the assemblage. While adult hominins appear to have been prey items in carnivoran assemblages such as Swartkrans, they were probably difﬁcult prey items for raptors and therefore unlikely to be present in raptor-accumulated deposits. 6. The Taung site was likely positioned in a locality consistent with nesting behavior of extant eagles.

Eagle predation at Tai The Taung raptor hypothesis was bolstered by studies of the predatory behavior of crowned eagles (Stepha- noaetus coronatus) in the Ivory Coast’s Tai Forest (Shultz, 2001; Schultz et al., 2004; McGraw et al., 2006). African crowned eagles are noted hunters of small and medium-sized mammals up to 20 kg, with a strong preference (up to 80% of their diet) for primates (Struhsaker and Leakey, 1992; Berger and Clarke, 1995; Sanders et al., 2003; McGraw et al., 2006). The Tai forest is home to 11 primate species, including eight cercopithecid monkeys that have been under study since 1989. Data on crowned eagles were collected at Tai between 1998 and 2001 (Shultz, 2001; Schultz et al., 2004; McGraw et al., 2006). Over the course of 34 months, 1,200 bones from feeding remains were recovered from 16 nests including 669 primate elements (see Fig. 1). Of these, 635 could be assigned to at least family level. The minimum number of primate individuals for the entire sample was 204. Key results from this study include:

Crowned eagles are capable of obtaining large prey including adult male sooty mangabeys (Cercocebus atys) weighing at least 12 kg. In fact, these large ter- Fig. 1. Representative sample of primate remains found in a restrial papionins are hunted preferentially and study of African crowned eagle predation in the Tai Forest. (a) appear in the faunal assemblage at the highest rate Sample of cercopithecid postcranial and cranial remains; (b) relative to their abundance on the landscape of any close-up of cercopithecid cranial remains, illustrating typical cercopithecid taxon at the site. damage to the occipital and basicranium. Crowned eagles leave a distinct taphonomic signature on their prey remains, particularly the crania. The dis- prey item at Taung (Gilbert, 2007). A small species of tinctive damage is largely because of eagles’ preference Papio, P. izodi, is also common in the Taung cerco- for exposing and consuming the cranial contents, partic- pithecid assemblage. There is still some uncertainty ularly the eyes and brain. Consequently, the crania dis- whether all the monkey fossils derive from the same play puncture marks behind and around the orbits, and ‘‘horizon’’ as the unique Australopithecus africanus, orbital floors are routinely broken. V-shaped breaks and but at least some were collected at the same time and can-opener like punctures are common on the vault, presumably place in the 1920s. and the cranial base is missing in the majority of skulls. Beak and talon scratch marks are present around these The combined results of these previous studies firmly cranial regions in variable quantities. Similar to other establish that the primate fauna at Taung was collected, studies documenting distinctive taphonomic signatures at least in part, by a large raptor such as the African on bone surfaces (e.g., Blumenschine et al., 1996), the crowned eagle. However, these results beg the question: gross damage inflicted by crowned eagles is easily Is such an accumulation an isolated taphonomic inci- observed without the use of high-powered microscopy. dent, or part of a larger pattern over the course of pri- Given the behavior of extant crowned eagles at Tai, it mate evolution? is entirely plausible that the Taung child—estimated to have weighed between 10 and 12 kg (Berger and Fossil primate sample from the Clarke, 1995)—could have been the victim of an eagle Humpata Plateau, Angola attack. Moreover, damage on the Taung child and papionin monkeys from Taung is consistent with dam- The fossil monkeys discussed here derive from several age on primates known to be victims of avian carni- sites on the Humpata Plateau: Leba (cave and fissures), vores (McGraw et al., 2006). Similar to the situation at Tchiua, Malola Kiln, and Cangalongue (cave and three Tai, a relative of extant terrestrial mangabeys, Procer- fissures/quarries) (Figs. 2 and 3). The Humpata plateau cocebus antiquus, appears to have been the preferred is formed, in large part, of Late Precambrian or Cam-

American Journal of Physical Anthropology FOSSIL EVIDENCE FOR EAGLE PREDATION ON CERCOPITHECIDS 423 brian dolomitic limestones. As with some of the dolo- mites of South Africa, solution cavities developed and acted as traps for animal remains and perhaps for still- living individuals as well (see Brain, 1981). Fossils are rare but may be locally abundant in brown or reddish calcified breccias, which include fallen roof blocks, terra rossa and some chert. It has previously been established that monkeys are dominant members of the fauna at Tchiua and some of the other fissures found on the Humpata Plateau (Pick- ford et al., 1992). Over the past half-century, specimens have been referred to various taxa such as Parapapio, Cercopithecoides, Papio (Dinopithecus), and Theropithe- cus by different authors (e.g., Dart, 1950; Mouta, 1950; Arambourg and Mouta, 1954; Freedman, 1957; Antunes, 1962, 1965; Minkoff, 1972; Simons and Delson, 1978; Pickford et al., 1990, 1992; Delson and Dean, 1993; Jablonski, 1994, 2002; Delson et al., 2000). The taxo- nomic affiliation of some of these specimens is a continued source of debate, but these issues are of secondary importance for the present discussion. While most of the assigned taxa are large-bodied cercopithecids, many of Fig. 2. Map of Southern Africa showing the location of Plio- the recovered specimens are, in fact, subadult individu- Pleistocene cercopithecid sites discussed in this study. 1, Leba, als and therefore of significantly smaller body mass. Tchiua, and Malola Kiln; 2, Cangalongue; 3, Taung; 4, Swart- According to published faunal lists and research in krans; 5, Sterkfontein; 6, Gladysvale; 7, Kromdraai; and 8, progress, large mammals other than primates are Makapansgat.

Fig. 3. Representative sample of fossil cercopithecid remains from the Plio-Pleistocene of the Humpata Plateau in Angola. Num- bers correspond to DGUNL LEBA specimen numbers (see Delson and Dean, 1993).

American Journal of Physical Anthropology 424 C.C. GILBERT ET AL.

TABLE 1. Faunal lists of Humpata Plateau fossil localities compared with Taung and Swartkrans Order, Family (Subfamily) Taung ‘‘Leba’’ Tchiua Malola Cangalongue 3 Cangalongue 2 Swartkrans Large mammals Carnivora Felidae (Felinae) X X X Felidae (Machairodontinae) X Hyaenidae X X Canidae X Artiodactyla Bovidae (Antilopinae) X X Bovidae (Bovinae) X X X Bovidae (Hippotraginae) ? X X Bovidae (Alcelaphinae) XX Bovidae (Peleinae) X Bovidae (Reduncinae) X X Suidae X ? X Anthracotheriidae ? Hippopotamidae X Girafﬁdae X Perissodactyla Rhinocerotidae X Equidae XX Proboscidea Elephantidae X Hyracoidea Procaviidae X Medium-sized mammals Primates Hominidae X X Cercopithecidae (Cercopithecinae) X X X X X X X Cercopithecidae (Colobinae) X X Carnivora Canidae X X Mustelidae XX Artiodactyla Bovidae (Cephalophinae) X Bovidae (Antilopinae) X X Suidae XXX Pholidota Manidae X Hyracoidea Procaviidae XX Small mammals Hyracoidea Procaviidae X X X X Carnivora Herpestidae X Lagomorpha Family indet. X Rodentia Bathyergidae X X X X Nesomyidae (Mystromyinae) X X Nesomyidae (Dendromurinae) X X X X X Muridae (Gerbillinae) X X X X X Muridae (Deomyinae) X X X Muridae (Murinae) X X X X X X Muridae (Otomyinae) X X X X Cricetidae X Gliridae (Graphiurinae) X X X X Pedetidae X X Petromuridae X Myoxidae X Hystricidae X ? X Chiroptera Rhinolophidae X X X X Vespertilionidae X Nycteridae X Soricomorpha Soricidae (Myosoricinae) X Soricidae (Crocidurinae) X X X X X X Macroscelidea Macroscelididae X X X X X

American Journal of Physical Anthropology FOSSIL EVIDENCE FOR EAGLE PREDATION ON CERCOPITHECIDS 425

TABLE 1. (Continued) Order, Family (Subfamily) Taung ‘‘Leba’’ Tchiua Malola Cangalongue 3 Cangalongue 2 Swartkrans Afrosoricida Chrysochloridae (Amblysominae) X Chrysochloridae (Chrysochlorinae) X Reptiles Order indet. X Testudines Cheloniidae X X X Squamata Lacertidae X X Ophidea, Family indet. X Birds Order indet.XXXX Eggshells X Strigiformes Strigidae X Amphibians Order indet.X X

X, taxon is present; ?, taxon may be present, but assignment is uncertain. Taung faunal list from Berger and Clarke (1995). ‘‘Leba’’ faunal list from ongoing study of material originally mentioned by Arambourg and Mouta (1954) and Antunes (1962, 1965); other Angolan sites from Pickford et al (1992). The Tchiua fauna refers to the monkey-bearing gray breccias (Pickford et al., 1992). Com- bined Swartkrans faunal list for Members 1-3 from Avery (2001) and Watson (2004).

TABLE 2. Size distribution of mammalian taxa across Plio-Pleistocene sites examined here Site Taung Tchiua Malola Cangalongue 3 Cangalongue 2 Swartkrans Large mammals (%) 16.67 0.00 15.38 0.00 25.00 35.71% Medium mammals (%) 20.83 10.00 7.69 22.22 25.00 19.05% Small mammals (%) 62.50 90.00 76.92 77.78 50.00 45.24% Large 1 medium mammals (%) 37.50 10.00 23.08 22.22 50.00 54.76%

Data taken from Table 1 without reference to frequency of taxon occurrences. For comparison, Swartkrans represents a carnivoran- based accumulation whereas Taung represents a raptor-based accumulation. Note that the Angolan faunas are generally much more similar to Taung rather than Swartkrans, particularly in the low percentage of large mammal remains and the high percentage of small mammal remains ([60%). The similarity of Tchiua, Malola Kiln, and Cangalongue 3 to Taung indicates an increased likelihood of raptors rather than carnivorans as a major taphonomic agent. Other South African sites such as Sterkfontein Mem- bers 4-5, Makapansgat Member 3, Kromdraai A and B, and Gladysvale show similar size-distributions as Swartkrans (Berger and Clarke, 1995).

largely absent at many of the Humpata monkey-bearing RESULTS sites (see Table 1). Where micromammals exist among the monkey-bearing fissures, they are typically domi- Gross damage comparisons of cercopithecid specimens nated by nocturnal species. These facts led Pickford et from the Humpata Plateau to those from Taung and Tai al. (1990 1994) to suggest that these micromammals, are presented in Figures 4–8. In addition to gross simi- along with small reptiles and other birds, were carried larities of damage among the Angolan, Taung, and Tai into the caves or dropped into the fissures by ‘‘owls and specimens, we also note that the Angolan faunas, partic- other raptors.’’ ularly Tchiua, display 5 out of the 6 hallmarks of eagle The suggestion of raptors as accumulating agents for predation (see above). the micromammals led us to re-examine the Angolan cercopithecid fauna for evidence of raptor predation as 1. The body size of individuals in the Angolan assem- well. In addition to gross damage comparisons, we calcu- blage are relatively homogeneous and dominated by lated simple size-distribution profiles using the faunal individuals \20 kg (see Tables 1–2 for comparison). lists provided in Table 1. Excluding the incomplete list The small-to-medium body size distribution of most for the ‘‘Leba’’ assemblage, the values for the Angolan cercopithecid individuals is due, in large part, to a sites compared to Taung, a proposed raptor-based accu- high percentage of subadult specimens (see Table 3). mulation (Berger and Clarke, 1995; McGraw et al. 2006; 2. Crania from Angola, particularly cercopithecid crania, Berger 2006; Berger and McGraw, 2007), and Swart- are relatively complete. These specimens frequently krans, a proposed carnivoran-based accumulation (e.g., have holes in the vault and orbits, occipitals are com- Brain, 1981; Watson, 2004), are provided in Table 2. As monly absent, and the brain is commonly exposed a final comparison, we also compiled age-distribution (see Figs. 4–7). There is no documented evidence of profiles for the cercopithecids found at extant crowned chewing. eagle sites of Tai and Kibale and the Plio-Pleistocene 3. Nontypical carnivore prey (e.g., snake skeleton) are Angolan sites on the Humpata Plateau (Table 3). present and show no evidence of chewing. In addition,

American Journal of Physical Anthropology 426 C.C. GILBERT ET AL.

TABLE 3. Percentage of subadult cercopithecid specimens taken by crowned eagles at Tai and Kibale compared with the percentage of subadult cercopithecid specimens found at Plio-Pleistocene sites on the Humpata Plateau Site Tai Kibale Leba Tchiua Adults (%) 62.37 48.54 40.00 37.50 Subadults (%) 37.63 51.46 50.00 50.00 Uncertain (%) 0.00 0.00 10.00 12.50 Sample size 194 103 20 8

Data for Tai taken from McGraw et al. (2006) and data for Kibale taken from Trapani et al. (2006). Data for Humpata sites taken from our own observations and from previously published descriptions. The collections from Cangalongue 3 and Malola Kiln include only 1 cercopithecid specimen each; therefore these sites are excluded. Following Trapani et al. (2006), adults are deﬁned as individuals with all molars fully erupted and all other specimens are subadult (including juveniles and infants). Note the similarity in the high percentage of subadult cercopithecid specimens found between the crowned eagle sites of Tai and Kibale and the Plio-Pleistocene Angolan sites Leba and Tchiua.

Fig. 5. Examples of cranial V-shaped breaks and depressions characteristic of eagle predation found on cercopithecid specimens from Tai and the Humpata Plateau. Humpata specimens top row, left to right: MCZ 19870 (subadult), TCH 38 (subadult). Humpata specimens bottom row, left to right: LEBA 10 (subadult), LEBA 12 (subadult). Note the V-breakage pattern on MCZ 19870 and the depressions present on TCH 38, LEBA 10, and LEBA 12.

the size-distribution proﬁles indicate that the faunas present at Tchiua, Malola Kiln, and Cangalongue 3 are dissimilar to classic carnivoran assemblages such as Swartkrans. While the percentage of large-sized and medium-sized mammal remains is typically over 50% for carnivoran assemblages such as Swartkrans (Table 2; see also Berger and Clarke, 1995), the assemblages at Tchiua, Malola, and Cangalongue 3 are similar to Taung in that they possess a low percentage of large-sized and medium-sized mammal remains and instead display a high percentage of small mammal remains ([60%; see Table 2; see also Berger and Clarke, 1995). The size-distribution for Cangalongue 2 is more similar to Swartkrans than Taung or the other Angolan sites, and this site will Fig. 4. Examples of orbital damage characteristic of eagle not be discussed further. predation found on primate specimens from Tai, Taung, and the Humpata Plateau. Note the consistent damage to the orbital 4. No adult hominins are present in the assemblages. ﬂoor in specimens from Tai, Taung, and the Humpata Plateau. 5. The Angolan monkey-bearing sites were likely posi- Humpata specimens top row, left to right: MCZ 19870 (sub- tioned in localities consistent with nesting behavior of adult), TCH 38 (subadult). Humpata specimens bottom row, left extant African crowned eagles. Pickford et al. (1990 to right: LEBA 10 (subadult), LEBA 13 (subadult). 1994) suggest that owls and other raptors probably

American Journal of Physical Anthropology FOSSIL EVIDENCE FOR EAGLE PREDATION ON CERCOPITHECIDS 427

Fig. 6. Examples of missing and damaged occipitals characteristic of eagle predation found on cercopithecid specimens from Tai, Taung, and the Humpata Plateau. Taung specimens, from left to right: M3078, UCMP 56694, TP8, SAM 5364. Hum- Fig. 7. Examples of talon scratches characteristic of eagle pata specimens, from left to right: MCZ 19870 (subadult), TCH predation found on cercopithecid specimens from Tai, Taung, 25, TCH 38 (subadult), LEBA 13 (subadult). and the Humpata Plateau. Taung specimen, SAM 5364. Hum- pata specimens, Top: LEBA 05; Bottom: TCH 38 (subadult). Note the similar pattern of scratches found on all of the carried small-prey items into the ﬁssures and caves; specimens. we also believe that the primates were either carried into the caves and ﬁssures or dropped into them from nests above. While extant crowned eagles prefer for- there are other points regarding the Angolan assemb- ests and do not currently range as far south as south- lages suggestive of eagle behavior. For example, subadult ern Angola, the reconstructed paleoenvironment of cercopithecids are collected at high rates by crowned the Humpata Plateau during the Plio-Pleistocene eagles at modern African sites such as Tai and Kibale includes more forested environments similar to (e.g., McGraw et al., 2006; Trapani et al., 2006). Further- crowned eagle preferred habitat (Pickford et al., more, crania and cranial fragments are often the most 1994). These trees would have provided ample oppor- likely elements to be preserved in and around eagle tunities for eagle nesting sites, making it most likely nests (McGraw et al., 2006; Trapani et al., 2006), partic- that the monkey remains were dropped into cavities ularly subadult cranial material (Trapani et al., 2006). from nests above. There is no published evidence of The dominance of subadult individuals as seen from water transportation as an accumulating agent (Pick- numerous crania in the Angolan assemblages suggest a ford et al., 1990). similar bias during the African Plio-Pleistocene (Table 3). The bias in the case of the Humpata Plateau may also be generally because of a preference for individuals DISCUSSION AND CONCLUSIONS under 20 kg, as some of the adult cercopithecids from this region weighed around or over 20 kg (Delson et al., In addition to the gross comparisons as well as the 2000). There is also evidence, however, of eagle preda- consistency of the taphonomic signature with Taung, tion on large adult cercopithecids as well. Specimen

American Journal of Physical Anthropology 428 C.C. GILBERT ET AL. course of primate evolution and may have had a role in shaping the socioecology of African cercopithecids since the Plio-Pleistocene. Additional primate localities and their various taphonomic signatures in the African fossil record will determine exactly how strong a selection pressure raptors historically represent, particularly relative to terrestrial predators.

ACKNOWLEDGMENTS The authors thank L. Berger, S. Shultz, S. Thomsett, and the assistants of the Tai Monkey Project. The authors also thank M. T. Antunes and B. Senut for per- mission to study the Leba fossils, and M. Moniz for assistance in identifying the Leba bovids. Finally, we thank the associate editor and three anonymous reviewers who greatly improved this manuscript.

LITERATURE CITED Antunes MT. 1962. Note sommaire sur quelques faunes de mammife`res Quaternaires de l’Angola. Proble`mes Actuels de Paleóntologie. Coll Internat Centre Nat Rech Sci. No. 104. Paris: Centre Nat. Rech. Sci. p 377–379. Antunes MT. 1965. Sur la faune de verte´bre´s du Pleístoce`ne de Leba, Humpata (Angola). Actes V Congr Panafricain de Pre´- histoire et de l’e´tude du Quaternaire, Tenerife, 1963. p 127– 128. Arambourg C, Mouta F. 1954. Les grottes et fentes a` ossements du sud de l’Angola. In: Actes II Pan-African Congress on Prehistory and Quaternary Studies, Algiers 1952. Paris: Arts et Me´tiers Graphiques. p 301–304. Fig. 8. Comparison of LEBA 05 frontal from the Humpata Avery DM. 2001. The Plio-Pleistocene vegetation and climate of Plateau with a red colobus frontal characteristic of eagle preda- Sterkfontein and Swartkrans, South Africa, based on micro- tion from Tai. mammals. J Hum Evol 41:113–132. Berger LR. 2006. Brief communication: Predatory bird damage to the Taung type-skull of Australopithecus africanus Dart LEBA 05, a [20 kg probable male, shows damage to the 1925. Am J Phys Anthropol 131:166–168. frontal that is consistent with African crowned eagle pre- Berger LR, Clarke RJ. 1995. Eagle involvement in accumulation dation (see Fig. 8); it has also been demonstrated that of the Taung child fauna. J Hum Evol 29:275–299. extant crowned eagles can successfully kill animals up to Berger L, McGraw WS. 2007. Further evidence for eagle predation of, and feeding damage on, the Taung child. S Afr J Sci 30 kg as well as lift animals heavier than themselves 103:496–498. (Brown, 1971; Steyn, 1973; Brown et al., 1982; Steyn, Blumenschine RJ, Marean CW, Capaldo SD. 1996. Blind tests 1983; Maclean, 1993; Trapani et al., 2006). Finally, ter- on inter-analyst correspondence and accuracy in the identifi- restrial mangabeys and their close relatives have been cation of cut marks, percussion marks, and carnivore tooth demonstrated to be preferred eagle food items over the marks on bone surfaces. J Arch Sci 23:493–507. past approximately two million years (McGraw et al., Brain CK. 1981. The hunters or the hunted? An introduction to 2006; Gilbert, 2007). A recent phylogenetic analysis sug- African cave taphonomy. Chicago: University of Chicago gests that a close relative of living terrestrial mangabeys Press. (Cercocebus) and mandrills (Mandrillus) is probably Brown LH. 1971. The relations of the crowned eagle Stepha- present in the Angolan faunas as well (Gilbert, 2008). noaetus coronatus and some of its prey animals. Ibis 113:240– Given the striking similarities among the Tai, Taung, 243. Brown LH, Urban E, Newmann K. 1982. The birds of Africa, and Angolan faunas, combined with the comparative dis- Vol. 1. London: Academic Press. similarity of the Swartkrans and Angolan faunas (see Dart RA. 1950. A note on the limestone caverns of Leba, near Table 2), we believe that a large raptor such as the Afri- Humpata, Angola. S Afr Archaeol Bull 5:149–151. can crowned eagle or a close relative, with a particular Delson E, Dean D. 1993. Are Papio baringensis R. Leakey, 1969, preference for cercopithecid monkeys, is most likely re- and P. quadratirostris Iwamoto, 1982, species of Papio or sponsible for the accumulation of at least some portion Theropithecus? In: Jablonski NG, editor. Theropithecus: the of the monkey-bearing faunas on the Humpata Plateau. rise and fall of a primate genus. Cambridge: Cambridge Uni- These Angolan faunas thus represent the second docu- versity Press. p 125–156. mented case of raptor predation in the cercopithecid fos- Delson E, Terranova CJ, Jungers WL, Sargis EJ, Jablonski NG, sil record, which suggests that others may not yet be rec- Dechow PC. 2000. Body mass in Cercopithecidae (Primates, Mammalia): estimation and scaling in extinct and extant ognized. The Angolan situation demonstrates that Taung taxa. Anthropol Pap Am Mus Nat Hist 83:1–159. was not unique, and we urge colleagues to consider simi- Freedman L. 1957. The fossil cercopithecoidea of South Africa. lar examples of raptor predation in published and newly Ann Transv Mus 23:121–262. recovered assemblages. Given the current evidence, we Gilbert CC. 2007. Craniomandibular morphology supporting the hypothesize that raptor predation has been a strong and diphyletic origin of mangabeys and a new genus of the Cercoce- perhaps underappreciated selective force during the bus/Mandrillus clade, Procercocebus. J Hum Evol 53:69–102.

American Journal of Physical Anthropology FOSSIL EVIDENCE FOR EAGLE PREDATION ON CERCOPITHECIDS 429

Gilbert CC. 2008. African papionin phylogenetic history and Sanders WJ, Trapani J, Mitani JC. 2003. Taphonomic aspects of Plio-Pleistocene biogeography. Ph.D. dissertation, Stony Brook crowned hawk-eagle predation on monkeys. J Hum Evol University, Stony Brook, NY. 44:87–105. Hrdlicka A. 1925. The Taungs ape. Am J Phys Anthropol 8:379–392. Shultz S. 2001. Notes on interactions between monkeys and Jablonski NG. 1994. New fossil cercopithecid remains from the African crowned eagles in Tai National Park, Ivory Coast. Humpata Plateau, Southern Angola. Am J Phys Anthropol Folia Primatol 72:248–250. 94:435–464. Schultz S, Noe R, McGraw WS, Dunbar RIM. 2004. A commu- Jablonski NG. 2002. Fossil Old World monkeys: the late Neo- nity-level evaluation of the impact of prey behavioral and eco- gene radiation. In: Hartwig WC, editor. The primate fossil re- logical characteristics on predator diet composition. Proc R cord. Cambridge: Cambridge University Press. p 255–299. Soc Lond [Biol] 271:725–732. Maclean GL. 1993. Robert’s birds of southern Africa. Cape Simons EL, Delson E. 1978. Cercopithecidae and Parapithecidae. Town: John Voelcker Bird Book Fund. In: Cooke HBS, Maglio VJ, editors. Mammalian evolution in McGraw WS, Cooke C, Shultz S. 2006. Primate remains from Africa. Cambridge: Harvard University Press. p 100–119. crowned hawk-eagle (Stephanoaetus coronatus) nests in Ivory Steyn P. 1973. Eagle days: a study of African eagles at the nest. Coast’s Tai forest: implications for South African cave taphon- Johannesburg: Purnell. omy. Am J Phys Anthropol 131:151–165. Minkoff EC. 1972. A fossil baboon from Angola, with a note on Steyn P. 1983. Birds of prey of Southern Africa. Dover: Tanager. Australopithecus. J Paleontol 46:836–844. Struhsaker TT, Leakey M. 1992. Prey selectivity by crowned Mouta F. 1950. Sur la presence du Quaternaire ancien dans les hawk-eagles on monkeys in the Kibale Forest, Uganda. Behav hauts plateaux du Sud de l’Angola (Humpata, Leba). C R Ecol Sociobiol 26:435–443. Somm Soc Geol Fr 14:261–262. Trapani J, Sanders WJ, Mitani JC, Heard A. 2006. Precision Pickford M, Fernandes T, Aco S. 1990. Nouvelles decouvertes de and consistency of the taphonomic signature of predation by remplissages de ﬁssures a` primates dans le Planalto da Humpata, crowned hawk-eagles (Stephanoaetus coronatus) in Kibale Huila, Sud de l’Angola. C R Acad Sci Paris Serie II 310:843–848. National Park, Uganda. Palaios 21:114–131. Pickford M, Mein P, Senut B. 1992. Primate bearing Plio-Pleistocene Watson V. 2004. Composition of Swartkrans bone accumula- cave deposits of Humpata, Southern Angola. Hum Evol 7:17–33. tions, in terms of skeletal parts and animals represented. In: Pickford M, Mein P, Senut B. 1994. Fossiliferous neogene karst Brain CK, editor. Swartkrans: a cave’s chronicle of early man, ﬁllings in Angola, Botswana, and Namibia. S Afr J Sci 90: 2nd ed. Pretoria: Transvaal Museum Mongraph No. 8. p 35– 227–230. 73.

American Journal of Physical Anthropology