P R O ME T H E US P R E S S / P A L A E O N T O L O G I C A L N E T W O R K F O UN D A T I O N

Journal of Taphonomy

(TERUEL) 2010

Available online at www.journaltaphonomy.com

de Ruiter et al.

VOLUME 8 (ISSUE 1-2)

Investigating the Role of Eagles as Accumulating Agents in the Dolomitic Cave Infills of South Africa Darryl J. de Ruiter* Department of Anthropology, Texas A&M University, College Station, TX 77843, USA

Sandi R. Copeland Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology, Deutscher Platz 6, Leipzig D-04103, Germany Department of Anthropology, University of Colorado at Boulder, Boulder CO 80309, USA

Julia Lee-Thorp Division of Archaeological, Geographical and Environmental Sciences University of Bradford, Bradford BD7 1DP, UK

Matt Sponheimer Department of Anthropology, University of Colorado at Boulder, Boulder CO 80309, USA Journal of Taphonomy 8 (1-2) (2010), 129-154. Manuscript received 15 March 2009, revised manuscript accepted 1 August 2009. The potential importance of large raptors as accumulators of early hominins was highlighted by the suggestion that the Taung Child was killed and deposited by an eagle (Berger & Clarke [1995] Journal of Human Evolution, 29: 275-299), and it has been hypothesized that eagles might have had a significant impact on the evolution of predator avoidance behaviors in early hominins (Berger & McGraw [2007] South African Journal of Science, 103: 496-498). In this study, we compare skeletal part representation of procaviid and cercopithecid fossils from the dolomitic cave infills of South Africa to a series of modern eagle-derived bone accumulations. We supplement skeletal part analysis with data on strontium isotope ratios (87Sr/86Sr) in the Bloubank Valley that allow us to source fossils to particular geological substrates. Of the fourteen discrete faunal assemblages examined, nine were inconsistent with eagles as accumulators of procaviids or cercopithecids, while five revealed possible, though not definitive, evidence of eagle involvement. A lack of support for eagles as collectors of the smaller mammals that Article JTa100. All rights reserved.

*E-mail: [email protected]

129

Eagles as accumulating agents

make up their typical prey weakens the hypothesis that eagles represented a significant threat to the larger, presumably more difficult to capture, juvenile hominins. The majority of the animals sampled for 87Sr/86Sr ratios at Swartkrans were consistent with being derived from local dolomites, including four Papio specimens, while we documented a non-local origin for a single procaviid and a single bovid from the Hanging Remnant of Member 1. In contrast, all of the procaviid specimens and a single bovid specimen from Sterkfontein Member 4 exhibited nonlocal strontium signals. Turning to the Taung Child, at present a clear link between it and the original Taung faunal assemblage examined by Raymond Dart cannot be established. In addition, preparation damage cannot be ruled out as the source of several marks on the Taung skull that have been putatively assigned to eagle talon damage. As a result, the hypothesized influence of large raptors such as eagles on the evolution of predator avoidance strategies in early hominins remains intriguing but unsubstantiated. Keywords: TAPHONOMY, EAGLE PREDATION, PROCAVIID, CERCOPITHECID, TAUNG

CHILD, STRONTIUM ISOTOPES, ACCUMULATING AGENT Introduction Determining the bone accumulating agent(s) responsible for producing a faunal assemblage is necessary in order to gauge the potential impact of prey selection and bone destruction biases on faunal assemblage composition. A variety of mammalian bone accumulators have been identified in the fossil cave infills of South Africa, including leopards (Panthera pardus) (Brain, 1969, 1981, 2004; de Ruiter, 2004; de Ruiter & Berger, 2000), hyaenas (Crocuta crocuta, Parahyaena brunnea) (Brain, 1981; de Ruiter et al., 2008a; Kuhn et al., 2008; Lacruz & Maude, 2005; Pickering, 2002; Pickering et al., 2004), and porcupines (Hystrix africaeaustralis) (Brain, 1981; Maguire et al., 1980). Extinct felids such as Dinofelis, Megantereon, and Homotherium, and extinct hyaenids such as Pachycrocuta and Chasmaporthetes, have also been put forward as potentially significant bone accumulating agents (Brain, 1981, 2004; Vrba, 1976), though the extent of their contributions is difficult to gauge (de Ruiter, 2001, 2004). Non-mammalian accumulators such as owls (e.g. Bubo africanus, B. capensis, B. lacteus,

Tyto alba) have also been implicated, though principally in the collection of microfaunal remains (Andrews, 1990; Avery, 2001; Brain, 1981). Brain (1981) initially raised the possibility that eagles might have accumulated many of the smaller mammals in the fossil cave infills, including rabbits (Leporidae) and hyraxes (Procavia capensis). Subsequently, Berger & Clarke (1995) hypothesized that some form of large eagle might have been responsible for the death and deposition of the Taung Child based on their analysis of the associated Taung fauna, though their hypothesis was challenged by some. Hedenstrom (1995) observed that if the Taung Child were carried by an eagle, based on its reconstructed body mass it would have had to have been dismembered first. Berger & Clarke (1996) responded by pointing out that the estimated body mass of the Taung Child might have been less than that cited by Hedenstrom (1995), that raptors often eviscerate their prey prior to transport (thus reducing the weight by as much as 30%), and that it is impossible to determine the distance an eagle might have had to transport the Taung Child. McKee (2001) argued that

130

de Ruiter et al.

the Taung assemblage was more consistent with leopard predation, and that there was little indication of raptor damage on the Taung Child itself. Notwithstanding, it remains possible that the Taung Child was collected by some form of large eagle. Several studies of the bone accumulating behaviors of large raptors in Africa have concluded that black eagles (Aquila verreauxii) and crowned eagles (Stephanoaetus coronatus) represent significant predation risks to both procaviids and monkeys (Cercopithecidae) (Cruz-Uribe & Klein, 1998; McGraw et al., 2006; Mitani et al., 2001; Sanders et al., 2003). These papers documented the taphonomic signatures that these large avians impart on procaviid and monkey skeletal remains. Using these data as a model, Berger (2006) and Berger & McGraw (2007) observed several marks in the orbits and on the face of the Taung Child possibly attributable to eagle talons, providing direct support for the hypothesis that the Taung Child was killed and deposited by a large eagle. Berger & McGraw (2007) further hypothesized that large birds of prey might have had a significant impact on the evolution of predator avoidance behavior in early hominins. The present study represents the first systematic investigation of the potential impact of eagles on faunal assemblage composition in the Plio-Pleistocene fossil accumulations of South Africa. Given that the association between the Taung Child and the Taung fauna is not secure (see below), the search for taphonomic traces attributable to eagles in other assemblages gains importance. The goal of this paper is to seek out indications of eagles as bone accumulating agents in the dolomitic cave infills of the Bloubank and Makapan Valleys of South Africa. Although it is clear that a variety of large mammals were involved in collecting the numerous macromammals recovered from

these caves (Brain, 1981), in this paper we will focus on two taxa that feature prominently in eagle-produced bone assemblages: procaviids and cercopithecids. The aim will be to determine if fossils belonging to these taxa evince taphonomic traces that would point to large eagles as their accumulators. If there is evidence favoring eagle contribution to any of the fossil assemblages, this would provide additional, albeit indirect, substantiation of eagles as potential predators of juvenile hominins. In contrast, if there is little or no support for eagles as collectors of the smaller mammals that make up their typical prey, this would weaken the hypothesis that eagles were transporting hominins to the fossil sites. In addition to taphonomic investigation of procaviid and cercopithecid fossils, we applied a strontium isotope (87Sr/86Sr) sourcing method to a series of procaviid, cercopithecid, equid and bovid teeth from the Bloubank Valley. Strontium (Sr) occasionally substitutes for calcium (Ca) during enamel deposition, and because the mass difference between the two isotopes of Sr is relatively small, it does so without significant biological fractionation. Since local rock 87Sr/86Sr ratios are based primarily on initial rubidium (Rb) concentrations and on the age of the local geology, rocks from different geological substrates will have different relative concentrations of these isotopes (Faure & Powell, 1972). Soluble or bioavailable strontium is taken up by plants growing on these substrates and then by animals eating the plants; hence animals resident on a particular geological substrate during the period of enamel mineralization will incorporate ratios of 87Sr/86Sr into their teeth that reflect the local concentrations of these isotopes. Examination of 87Sr/86Sr ratios therefore allows us to source individual fossils to specific geological substrates. From a taphonomic perspective, we are able to determine if a

131

Eagles as accumulating agents

particular individual animal was a local resident, or if it was transported to a given site from a remote geological substrate. Cercopithecids are often highly mobile animals, i.e. they often move widely both before and after enamel deposition has ceased. Home ranges of baboons (Papio ursinus) in South Africa vary greatly depending on habitat and group size, though published accounts of ranges from 9.1 to 33.7 km2 demonstrate the mobility of these animals (Skinner & Smithers, 1990). Procaviids, on the other hand, tend not to travel as widely, occupying ranges of approximately 4.2 km2 (Estes, 1991); while this may seem like a large territory, such a range reduces to an area of about 65x65 m. Male procaviids disperse from their natal group at maturity, potentially confounding attempts to assess perimortem movement via Sr isotopes. However, since dental development will be largely complete by the time an animal is forced to disperse, we do not expect this to significantly impact our results. We use Sr data to test whether the fossil procaviids and cercopithecids were brought into the fossil cave assemblages from somewhere other than the local geological substrate, in this case the dolomites of the Malmani Subgroup of the Transvaal Supergroup. These Malmani dolomites traverse northeastwards in a 6-10 km broad swath through the World Heritage Area in which the Bloubank Valley is located (see Figure 1 in Copeland et al., 2010). We further compare these procaviid and cercopithecid data to 87Sr/86Sr ratios in a series of bovids and an equid to determine if these larger animals were also resident in the Malmani dolomites, or if they too might be traced to a more distant source. A general objection that might be raised concerns the nesting patterns of eagles relative to the fossil cave infills of the Bloubank and Makapan Valleys. Bones

accumulated by eagles are found in direct association with their nests, therefore if eagles do not nest in the vicinity of caves, investigation of eagle involvement might produce spurious results. However, we note that two of the eagles identified as potential killers of the Taung Child, the crowned eagle and the martial eagle (Polemaetus bellicosus), regularly construct their nests in trees (Steyn, 1982). In addition, although black eagles normally build their nests on cliff faces, they have also been recorded constructing nests in trees (Gargett, 1990; Steyn, 1982). Given that trees tend to be concentrated around the openings of caves in South Africa (Brain, 1981), and given that the paleoenvironments reconstructed for most of the fossil cave infills include a significant wooded component (Avery, 2001; Benefit & McCrossin, 1990; Brain et al., 1988; de Ruiter et al., 2008b; Reed, 1997), we see no reason to suggest that eagles could not have built their nests in the immediate vicinity of the caves. Little attention has been paid to the avian fossils recovered from the Bloubank and Makapan Valley caves, thus we have little information regarding the presence of raptors in the assemblages. At present, the only raptor fossils known have been assigned only as far as the family Accipitridae, which includes eagles, hawks and vultures (Watson, 2004). Eagles as Bone Accumulators Cruz-Uribe & Klein (1998) documented a distinct pattern of skeletal part representation in procaviids recovered beneath large eagle nests in South Africa. They noted that cranial remains tended to be well represented, while post-cranial elements were notably rare. Of these post-cranial remains, hindlimb elements

132

de Ruiter et al.

tended to outnumber forelimb elements, and Cruz-Uribe & Klein (1998) considered this relative overabundance of hindlimb elements to be as diagnostic of eagle predation on procaviids as the prevalence of cranial elements. Similarly, a recurrent pattern appears in skeletal remains recovered from crowned eagle nests documented in studies in Uganda (Mitani et al., 2001; Sanders et al., 2003) and Ivory Coast (McGraw et al., 2006): crowned eagles are significant predators of primates, producing bone accumulations in which cranial remains predominate over post-cranial remains, with hindlimb elements significantly outnumbering forelimb elements. These data are consistent with those of eagle nests in South Africa, although in the Taï Forest cranial remains do not predominate so overwhelmingly (McGraw et al., 2006). We use this pattern of skeletal part representation to document the probable involvement of large eagles in the faunal assemblages of the Bloubank Valley, as it appears consistent across raptor taxa and at a continent-wide scale (Cruz-Uribe & Klein, 1998; Gargett, 1990; McGraw et al., 2006; Mitani et al., 2001; Sanders et al., 2003). Alongside patterns of skeletal part representation, patterns of bone surface modification and fragmentation can also be used to identify eagles as accumulating agents (Cruz-Uribe & Klein, 1998; McGraw et al., 2006; Mitani et al., 2001; Sanders et al., 2003). However, Cruz-Uribe & Klein (1998) noted that bone surface modifications in the form of beak marks, talon marks, or gastric etching were rare in their large eagle samples, thus they relied principally on skeletal part representation and age profiles to assess the relative input of eagles in a series of fossil assemblages from the caves of the Western Cape. Likewise in Uganda, damage imparted to bones by crowned eagles tended to be

rare, and tended to be restricted to cranial elements (Mitani et al., 2001; Sanders et al., 2003). In contrast, in the Taï Forest sample crowned eagle damage was commonly seen in both cranial and post-cranial elements (McGraw et al., 2006). In the fossil cave infills of South Africa, factors such as bone fragmentation and cortical exfoliation from weathering, trampling, and geological abrasion can obscure taphonomic marks, thus eliminating traces that were originally present on the bones (Cruz-Uribe, 1991; Newman, 2004). In addition, adherent sediment or CaCO3 encrustation can further shroud such taphonomic traces (Brain, 1981; de Ruiter et al., 2008a). As a result, we rely minimally on the appearance of bone surface modifications and bone breakage patterns in diagnosing potential eagle involvement. Three other criteria have been suggested to be indicative of eagle predation (Cruz-Uribe & Klein, 1998; McGraw et al., 2006). First is the relatively homogeneous taxonomic composition seen in the bone assemblages, resulting in a consistent representation of similar body sized individuals. Typical eagle prey in the Taï Forest, for instance, tend to average approximately 4-6 kg (McGraw et al., 2006), though eagles have been documented transporting prey larger than 6kg (Berger & Clarke, 1996); in cases where prey exceeds the carrying capacity of the bird, they are known to disarticulate prey and transport it to nesting sites in pieces (Steyn, 1981). Given that the fossil cave infills of South Africa were likely accumulated by a variety of bone collecting agents over an indeterminate span of time (Brain, 1980), it is not possible to compare the patterns in body size representation across the sites with those seen in modern eagle nests. Rather, we will concentrate on the two groups of mammals that are most consistently represented in the eagle nests:

133

Eagles as accumulating agents

procaviids and cercopithecids. Second, in eagle derived assemblages there appears to be a high proportion of crania with attached mandibles. However, given the high levels of comminution in the fossil cave infills, crania with attached mandibles are exceedingly rare, a factor not necessarily related to eagle predation (or lack thereof). In addition, this pattern of attached mandibles is not evidenced in the modern Taï Forest sample (McGraw et al., 2006), thus limiting its potential as a diagnostic tool. The third criterion revolves around the relative abundance of juvenile versus adult individuals. In the modern South African eagle samples, very young procaviid individuals are poorly represented, likely because they are much more difficult for eagles to capture as a result of their propensity for staying near protective shelter (Cruz-Uribe & Klein, 1998). Given that the majority of procaviid remains in the fossil assemblages are comprised of isolated teeth, establishing reliable age profiles is not feasible. Regarding primates, McGraw et al. (2006) determined that there did not appear to be a bias towards any particular age group, regardless of the cercopithecid species involved. In addition, there did not appear to be any bias relating to skeletal part representation between adults and juveniles, and damage patterns were similar across age groupings. As a result, we do not rely on age profiles to test for the influence of eagle predation. The net effect of these data is that the most reliable line of evidence, at least in terms of assessing eagle contribution to the fossil assemblages of South Africa, involves an analysis of skeletal part representation. We supplement data on skeletal part representation by examining fluctuations in 87Sr/86Sr ratios in procaviid and cercopithecid fossils recovered from the cave infills, to document whether any of the animals sampled might

have been brought to the Bloubank Valley from a remote geological substrate. Materials and Methods We examined faunal assemblages from Sterkfontein, Swartkrans, Kromdraai, Coopers, Motsetse, Gladysvale, Plovers Lake, and Makapansgat (Table 1). These localities encompass a diverse array of late Pliocene to late Pleistocene deposits in South Africa, including all of the Bloubank and Makapan Valley sites for which good faunal assemblage data are available. Skeletal part data on procaviid and cercopithecid materials were collected firsthand from Sterkfontein, Swartkrans, Kromdraai, Coopers, Plovers Lake, and Makapansgat; where necessary we included data from published accounts for Sterkfontein (Brain, 1981; Pickering, 1999), Motsetse (Berger & Lacruz, 2003), Gladysvale (Lacruz et al., 2002) and Makapansgat (Reed, 1996). Procaviid remains are rare in some of the assemblages, and it is unclear whether this relates to a real paucity in the original deposition, or if another factor such as excavator bias was in operation. This applies especially to the assemblages recovered from the Hanging Remnant of Swartkrans Member 1, Kromdraai, and Sterkfontein Member 4, as these were all subjected to biased collecting procedures upon their initial discovery and excavation. Robert Broom, in particular, spent a great deal of time searching for the skulls and teeth of fossil hominins, primates, carnivores, and suids, discarding the majority of specimens that were of less interest; many of these discarded specimens were subsequently sold to, or carried away by, visitors to the various sites (Broom, 1951), thereby skewing faunal representation in the early collections. Sterkfontein Member 4 in this study refers

134

de Ruiter et al.

only to those specimens collected by Robert Broom and John Robinson between 1936 and 1953 which are housed in the Transvaal Museum. Cercopithecids are known exclusively from isolated teeth at Plovers Lake, and they are entirely unknown at Motsetse, therefore we discuss only procaviid data for these latter two sites. A single, isolated procaviid tooth is known from Makapansgat Member 3. Procaviids are known only from craniodental remains in Makapansgat Member 4, including four near complete crania assigned to Gigantohyrax maguirei. As a result, we rely principally on Makapansgat cercopithecid remains in this study. Skeletal part analysis Skeletal part data were recorded for all procaviid and cercopithecid taxa in the assemblages. There are three principal procaviid species that are variously represented in the fossil cave infills, two extinct (Procavia transvaalensis and Procavia antiqua) and one extant (Procavia capensis). Although these three species differ in body size, we consider them together to maintain sufficient sample sizes, and on the assumption that predatory eagles would not have preferentially selected any particular species. An additional extinct procaviid species, Gigantohyrax maguirei, is known from Member 4 of Makapansgat; this species is about 1.5x the size of a modern procaviid, but has not been recorded in any other faunal assemblage in South Africa. A number of cercopithecid taxa are represented in the faunal assemblages of the South African cave infills (Brain, 1981; Delson, 1984; Freedman, 1957). Post-cranial remains tend to be rare, and they are difficult to attribute taxonomically. Therefore, in order to maintain sufficient sample sizes, we

consider all cercopithecids together as an analytical unit. Because we are seeking an independent test of the potential impact of eagles on faunal assemblage composition in relation to hominins, we do not include the hominin fossils themselves in this study (this applies in particular to juvenile hominins). In their initial study, Cruz-Uribe & Klein (1998) used minimum numbers of individuals (MNI) per skeletal element as their analytical unit. However, owing to difficulties encountered in the calculation and interpretation of MNI (Grayson, 1984; Plug & Plug, 1990), we prefer to use the minimum numbers of elements (MNE) for our computation. Although it is possible that our usage of MNE might over-count individuals brought to the caves intact, we judge this to be unlikely owing to the high degree of bone comminution evident in the faunal assemblages (Brain, 1981, 2004). As a result, we assume that the use of the MNE as our unit of quantification will not significantly influence the results of this study. We construct a series of bone survival profiles for both procaviid and cercopithecid materials in the fossil assemblages of South Africa. We compare these bone survival profiles to those developed for modern correlates to test whether there is any evidence that eagles contributed significant quantities of animals to the fossil assemblages in question. If there is an indication of eagle contribution amongst procaviid and cercopithecid fossils, then the hypothesis that eagles might also have been accumulating juvenile hominins cannot be refuted. If, on the other hand, there is little indication that eagles contributed to the fossil assemblages, this would weaken an argument in favor of eagle predation on procaviids and cercopithecids, and in turn juvenile hominins, at least in the Bloubank Valley. Our reasoning is, if eagles are not

135

Eagles as accumulating agents

Member/deposit Member 1, Hanging Remnant Member 1, Lower Bank Member 2 Member 3 Kromdraai A Kromdraai B Coopers D Member 4 Member 5, Oldowan Infill Motsetse External deposit Internal deposit Member 3

Eagle involvement as determined by the present study possible no no no possible no no possible no no no no possible

Published source for fauna de Ruiter, 2001, 2004 de Ruiter, 2001, 2004 de Ruiter, 2001, 2004 de Ruiter, 2001, 2004 Brain, 1981 Brain, 1981 de Ruiter et al, 2009 Brain, 1981 Pickering , 1999 Berger & Lacruz, 2003 Lacruz et al, 2002 de Ruiter et al, 2008 Reed, 1996

Table 1. Faunal assemblages for which hyrax and cercopithecid skeletal element data were recorded.

Site Swartkrans

Kromdraai Coopers Sterkfontein Motsetse Gladysvale Plovers Lake Makapansgat

136

de Ruiter et al.

accumulating their more favored prey (procaviids and cercopithecids) in the fossil caves, then there is little rationale for suggesting that they were accumulating the larger, presumably more difficult to capture, juvenile hominins. This would cast substantial doubt on the hypothesis that eagles represented important threats to juvenile hominins, or that they significantly impacted the evolution of predator avoidance behaviors (Berger & McGraw, 2007). Sourcing fossils using 87Sr/86Sr isotope ratios We characterized biologically available 87 Sr/86Sr by analyzing plants from several different geological substrates in and around the Bloubank Valley and found that they were highly significantly different (see Copeland et al., 2010, for details and for collection and analytical procedures). The ranges and means of 87Sr/86Sr values of fossil fauna recovered from the sites of Sterkfontein, Swartkrans, and Gladysvale are similar to the ranges and means of values in modern fauna and flora, indicating that the fossils have not been significantly altered by diagenesis. Therefore, we can use the 87Sr/86Sr ratios of fossil specimens to determine if they were locally derived (i.e. from the Malmani dolomites of the Bloubank Valley and environs), or if they were brought into the cave from farther afield (see map in Figure 1, Copeland et al., 2010). For the present study, we sampled 87 Sr/86Sr in a series of fossil teeth from the sites of Swartkrans and Sterkfontein (Table 2). Because laser ablation of tooth enamel is less destructive than more traditional techniques, yet renders similar results (Copeland et al., 2008), we used this as our preferred technique. The teeth were measured using a New Wave 213 nm

laser coupled to a NuPlasma high resolution multi-collector inductively coupled plasma mass spectrometer (MC-ICP-MS) in the Africa Earth Observatory Network EarthLAB Facility at the University of Cape Town, South Africa. In addition to single strontium isotope samples recovered from a selection of fossil teeth, we also undertook incremental sampling wherever possible to document fluctuations in 87Sr/86Sr ratios over ontogenetic time in individuals. Specimens with 87Sr/86Sr ratios different from the local Malmani dolomites that host the fossil-bearing caves likely originated elsewhere, and either traveled to the vicinity of the caves after enamel deposition was complete, or were carried to the caves perimortem. The small size of procaviid teeth makes incremental sampling more difficult, therefore in many cases we relied on single laser ablation samples. Procaviids tend to be restricted in range to the rocky outcrops that provide their shelter, resulting in relatively small home ranges (Estes, 1991; Skinner & Smithers, 1990). Cercopithecids tend to have larger home ranges, and we include a sample of fossil cercopithecids to determine if they can be sourced to the local Malmani dolomites, or if they are derived from farther afield. Results Skeletal part representation As noted above, modern eagle predation results in distinctive skeletal part profiles for procaviids and cercopithecids (Cruz-Uribe & Klein, 1998; McGraw et al., 2006; Mitani et al., 2001; Sanders et al., 2003). In contrast to the typical eagle pattern, Cruz-Uribe & Klein (1998) noted that while fossil cave assemblages in the Western Cape of South Africa were

137

Eagles as accumulating agents

Table 2. Fossil specimens sampled for 87Sr/86Sr ratios. Accession # Taxon Swartkrans Member 1, Hanging Remnant SK 161 Procavia antiqua SK 182 Procavia antiqua SK 457 Papio hamadryas robinsoni SK 500 Papio hamadryas robinsoni SK 598 Papio hamadryas robinsoni SK 623 Papio hamadryas robinsoni SK 2304 Tragelaphus strepsiceros SK 2576 Tragelaphus strepsiceros SK 2626 Equus capensis SK 4076 Procavia antiqua SK 12363 Raphicerus sp. SKW 5946 Connochaetes sp.

Sampling method single sample single sample incremental sample incremental sample incremental sample incremental sample incremental sample incremental sample incremental sample incremental sample incremental sample incremental sample

Swartkrans Member 1, Lower Bank SKX 5285 Raphicerus sp. SKX 8494a Raphicerus sp. SKX 8535a Raphicerus sp. SKX 14150 Raphicerus sp. SKX 14769a Connochaetes sp.

single sample incremental sample incremental sample incremental sample incremental sample

Swartkrans Member 2 SK 199 SKX 653 SKX 671 SKX 1373 SKX 3170

single sample incremental sample incremental sample single sample single sample

Procavia transvaalensis Procavia transvaalensis Connochaetes sp. Procavia transvaalensis Procavia transvaalensis

Sterkfontein Member 4 Sts 107 Procavia antiqua single sample Sts 108 Procavia antiqua single sample Sts 1996 Antidorcas recki incremental sample Sts 2580 Damaliscus/Parmularius incremental sample Sts b Procavia sp. single sample Sts g Procavia sp. single sample Note: incrementally sampled specimen values were averaged and plotted in Figure 3.

characterized by a predominance of craniodental remains, they were also characterized by a relatively high proportion of post-cranial remains, with forelimb elements predominating over hindlimb elements. They therefore considered both a high proportion of postcranial remains and/or a predominance of

forelimb over hindlimb elements to be contra-indications of eagle involvement in a fossil assemblage. Figure 1 presents a series of bone survival profiles for procaviids recovered from the fossil assemblages examined in this study. Figure 2 presents bone survival profiles for cercopithecids, and

138

de Ruiter et al.

in most cases the skeletal part representations of procaviids and cercopithecids are in agreement. With a few notable exceptions, mandibles predominate, though post-cranial remains are often well represented. Within the Swartkrans assemblages, the Lower Bank of Member 1, Member 2 and Member 3 all display relatively elevated proportions of procaviid post-cranial remains, and in all cases forelimbs outnumber hindlimbs. Turning to the cercopithecids, the pattern is clearer still: post-cranial remains are well represented, and forelimb elements again outnumber hindlimb elements. In all cases, the procaviid and cercopithecid skeletal part profiles are statistically significantly different from modern eagle assemblages (Kolmogorov-Smirnoff two-sample test; Table 3). These combined skeletal part data are inconsistent with eagles as significant bone accumulating agents in these assemblages. In contrast, the Hanging Remnant of Member 1 of Swartkrans differs in that post-cranial remains of both procaviids and cercopithecids are rare, raising the possibility that they might have been collected by eagles. However, this paucity of post-cranial remains could be a result of the biased fossil collecting procedures employed in the earliest excavations at this site, where a cranial-centric search pattern dominated (Broom, 1951). In both the procaviid and cercopithecid assemblages of the Hanging Remnant, forelimb elements surpass hindlimb elements, particularly within the procaviids, a pattern at odds with eagle accumulations. In addition, the skeletal part profiles of procaviids and cercopithecids are significantly different from modern eagle accumulations (Table 3), though in the case of the cercopithecids the high proportion of mandibles in the fossil assemblage (an indicator of eagle contribution) represents the most notable departure between the fossil and

modern assemblages. It is therefore possible that eagles were significant contributory agents in the Hanging Remnant, though this requires additional testing, preferably with a taphonomically representative fossil sample (e.g. Sutton et al., 2009). Kromdraai A shows a clear preponderance of procaviid cranio-dental elements, and a marked scarcity of post-cranial elements. And, while the cercopithecid skeletal part data indicate a slightly elevated proportion of post-cranial remains in Kromdraai A, including a predominance of forelimb elements, they nonetheless remain poorly represented. The skeletal part profile of procaviids is statistically indistinguishable from that of modern eagle accumulations, though there is a significant difference between the primate skeletal part profile and modern eagle collections (Table 3). In this latter case, the very high proportion of cercopithecid cranio-dental remains in KA is apparently influencing the statistical test. Since this high proportion of craniodental remains is a characteristic of eagle accumulations, eagles cannot be ruled out as accumulating agents in KA. Although there is again a possibility that Broom’s biased collecting procedures might be at fault, these data are nonetheless suggestive that the procaviids and cercopithecids recovered from Kromdraai A were collected by large eagles. Kromdraai B revealed a single procaviid humerus fragment and no cranio-dental remains, hinting that eagles were not significant contributors; however, we are not inclined to draw inferences from a single data point. Turning to cercopithecids, we see a distinct prevalence of post-cranial remains, with a preponderance of forelimb over hindlimb elements. The skeletal part profile of fossil primates is significantly different from that of modern eagle accumulations, a pattern which is inconsistent with eagle predation.

139

Eagles as accumulating agents

Figure 1. Bone survival profiles for procaviids recorded in the fossil assemblages examined in this study. Percentages represent proportions of skeletal elements relative to the MNE of the most common element.

140

de Ruiter et al.

Figure 2. Bone survival profiles for cercopithecids recorded in the fossil assemblages examined in this study. Percentages represent proportions of skeletal elements relative to the MNE of the most common element.

141

Eagles as accumulating agents

Thus it does not appear that eagles played a significant role in the accumulation of the Kromdraai B assemblage. Procaviid post-cranial remains are well represented in the Cooper’s D assemblage, with a predominance of forelimb over hindlimb elements. The pattern of cercopithecid skeletal part representation is equally clear, with a relative abundance of post-cranial over craniodental remains, and an equal representation of forelimb and hindlimb elements. In both the procaviid and cercopithecid skeletal part profiles we see a statistically significant difference between the fossil assemblage and modern eagle collections (Table 3). The combined procaviid and cercopithecid data provide no support for eagles as accumulating agents in the Cooper’s D assemblage. Procaviids are rare in both Members 4 and 5 (Oldowan Infill) of Sterkfontein, and in both cases only cranio-dental remains are known. Biased collection procedures focusing on cranial remains (Broom, 1951) in the Member 4 assemblage (1936-1953) might be responsible for the absence of procaviid post-cranial remains, but the same cannot be said for the Oldowan Infill assemblage of Member 5, which was excavated using a total collection strategy (Kuman & Clarke, 2000). However, the single procaviid mandible fragment recovered from the Oldowan Infill does not provide strong support for eagles as accumulating agents, as we are again unwilling to draw inferences from a single data point. The cercopithecid evidence is clearer, with much larger sample sizes available for both Members. Member 4 shows a distinct overabundance of cercopithecid cranio-dental remains, and within the rare post-cranial elements, a relative surplus of hindlimb over forelimb elements, both consistent with eagles. The skeletal part profile of procaviids in Member

4 is statistically indistinguishable from that of modern eagle accumulations, though the cercopithecid skeletal part profile deviates significantly from the modern (Table 3). In the latter case, the high proportion of craniodental remains is clearly influencing the outcome of the test, thus eagles remain a potential contributor. These data, therefore, do not discount eagles as contributing agents in the Member 4 assemblage. The Member 5 Oldowan Infill cercopithecid assemblage, on the other hand, reveals a preponderance of post-cranial elements, with a prevalence of forelimb over hindlimb elements. Although there is a statistically significant difference between skeletal part profiles of fossil and modern procaviids, this is likely a result of the very small sample size, and thus cannot be considered meaningful. That being said, the larger primate skeletal part profile from the Oldowan Infill is also statistically significantly different from that of modern eagles. As a result, there is little support for eagles as accumulating agents in the Oldowan Infill of Member 5. The external deposits of Gladysvale show a relative abundance of procaviid postcranial over cranio-dental remains, alongside a preponderance of forelimb over hindlimb elements. A similar pattern is seen in the cercopithecid assemblage, where there are no cranio-dental fossils, and a predominance of forelimb over hindlimb elements. In both cases, the Gladysvale External Deposit skeletal part profiles are statistically significantly different from modern eagle accumulations (Table 3). These combined data provide no evidence of eagle contribution to the Gladysvale External Deposit assemblage. Motsetse displays a relative over-abundance of post-cranial remains of procaviids, contrasted with a slight predominance of hindlimb over forelimb elements. The procaviid skeletal

142

Procaviids, Procaviids, Cercopithecids, black eagle nests crowned eagle nests crowned eagle nest Assemblage (South Africa) (South Africa) (Ivory Coast) Swartkrans Member 1, Hanging Remnant 0.23 0.23 0.77 Swartkrans Member 1, Lower Bank 0.39 0.42 0.27 Swartkrans Member 2 0.27 0.27 0.28 Swartkrans Member 3 0.29 0.29 0.45 Kromdraai A 0.17 0.18 0.72 Kromdraai B 0.82 0.85 0.24 Cooper’s D 0.40 0.43 0.22 Sterkfontein Member 4 0.20 0.20 0.83 Sterkfontein Member 5, Oldowan Infill 0.44 0.45 0.27 Motsetse 0.59 0.58 Gladysvale External Deposit 0.72 0.75 0.50 Plovers Lake Internal Deposit 0.28 0.30 Makapansgat Member 3 0.58 Makapansgat Member 4 0.79 Note: significant differences are evident between several samples; values in italics are statistically significant to the 0.05 level, and values in bold are significant to the 0.01 level. Procaviid skeletal part data from South Africa from Cruz-Uribe & Klein (1998); cercopithecid skeletal part data from Ivory Coast from McGraw et al. (2006).

Table 3. Critical values (D) for the Kolmogorov-Smirnoff two-sample test comparing body part frequencies presented in Figures 1 and 2 against body part frequencies from modern eagle accumulations.

de Ruiter et al.

143

Eagles as accumulating agents

part profile is significantly different from modern eagle assemblages. Although there are no cercopithecids known from Motsetse, the high proportion of procaviid post-cranial remains is sufficient to rule out eagles as accumulating agents. The Plovers Lake internal deposit likewise has no cercopithecid remains (apart from isolated teeth), though the procaviid post-cranial remains are well represented, with a relative abundance of forelimb over hindlimb elements. The procaviid skeletal part profile is significantly different from that of modern eagle accumulations, therefore we do not see a clear indication of eagles depositing materials at Plovers Lake. Member 3 of Makapansgat exhibits an abundance of cercopithecid cranio-dental over post-cranial remains, which is suggestive of eagle contribution. However, among the post-cranial materials, there is a clear overrepresentation of forelimb elements compared to hindlimb, arguing against eagle contribution. The skeletal part profile of Member 3 cercopithecids is statistically significantly different from that of modern eagle accumulations (Table 3), though the high proportion of cranio-dental remains is clearly influencing the outcome of the statistical testing. As a result, it is possible that eagles contributed to the Member 3 assemblage, though the data are equivocal. Makapansgat Member 4 shows a clear preponderance of cranio-dental remains, and an even representation of forelimb and hindlimb elements. In addition, there are a series of four nearly intact crania attributed to a species unique to this deposit, Gigantohyrax maguirei. These data combine to indicate that eagles might have been significant contributors to the Member 4 accumulation. However, the skeletal part profile of cercopithecids is statistically significantly different from modern eagle accumulations (Table 3), though this difference is likely being influenced by the

high proportions of cranio-dental remains. Based on the preponderance of cercopithecid cranio-dental remains, we cannot rule eagles out as potential contributors to the Member 4 assemblage. 87

Sr/86Sr ratios

Biologically available strontium isotope ratios are highly differentiated between geological substrates in the Bloubank Valley and environs area, allowing us to distinguish fossils brought to the caves from non-dolomite locations. Figure 3 presents range plots of biologically available 87Sr/86Sr ratios from a series of geological substrates as determined by plants, along with 87Sr/86Sr isotope ratio values for a selection of fossils from the sites of Swartkrans and Sterkfontein. The geological substrates that we document include the Malmani dolomites, Timeball Hill shales, and the Hekpoort andesite/basalts (see Figure 1 in Copeland et al., 2010). Strontium isotope ratios in the geological substrates to the south of the Bloubank Valley are currently under investigation. As noted above, some of our fossil specimens were incrementally sampled, while others provided only single samples (Table 2). Incrementally sampled specimens were averaged to provide single data points in Figure 3. All of the fossil caves are housed in dolomitic bedrock, therefore the values for the dolomites represent the signal we would expect if fossils were locally derived; i.e. from the Malmani dolomites. Most of the Swartkrans steenbok (Raphicerus sp.) fossils are consistent with a source on the surrounding dolomites, with the exception a single individual (SKX 12363) that appears to be derived from a non-dolomite substrate. This specimen was recovered from Member 2 of Swartkrans, in

144

de Ruiter et al.

which there is no other indication of eagle contribution (Table 1). Given that the nearest non-dolomite substrate is less than 5 km away from Swartkrans (see Figure 1 in Copeland et al., 2010), it is possible that this individual moved of its own volition, rather than being carried by a large eagle. The majority of the Swartkrans procaviids fall within the range of the dolomites. Only one individual (SK 4076) from the Hanging Remnant of Member 1 was apparently externally derived, possibly from the Hekpoort andesite/basalt substrate (Figure 3). These data raise the possibility that eagles contributed to the Hanging Remnant assemblage, as reflected in the skeletal part profiles from this assemblage, despite the lack of indication of eagle involvement in the other Swartkrans assemblages. In contrast, all of the cercopithecids (Papio) that we sampled from the Hanging Remnant of Member 1 of Swartkrans can be sourced to the Malmani dolomites in which the cave is located. The combined Swartkrans Sr data therefore suggest that the majority of the animals in the cave are of local origin, thereby arguing against the necessity of their having been brought to the cave from a long distance away, presumably by eagles. The procaviids from Sterkfontein all plot outside the range of the local dolomites (Figure 3). We have not yet sampled a geological substrate with equivalent 87Sr/86Sr values, though the possibility exists that such a signal might be expected in an animal living in a riverine context (e.g. Sillen et al., 1998). However, such a conclusion is inconsistent with what is known of procaviid habitat and ecological preferences (Skinner & Smithers, 1990). Extensive human modification of the area to the south of the Bloubank Valley may have obscured some of the original soil profiles and associated biologically available strontium

isotope ratios; we are still investigating this point. Notwithstanding, the exogenous 87Sr/86Sr values provide support for the suggestion that the procaviids of Member 4 of Sterkfontein might have been accumulated by eagles. We currently have no data for Sterkfontein cercopithecids. On present data, we cannot rule out eagles as contributors to the Sterkfontein Member 4 assemblage, therefore we cannot yet rule out the possibility that they might also be responsible for some of the hominin juveniles known from this deposit. Incrementally sampled specimens represent an additional test of the sourcing of fossil individuals relative to geological substrates. Figure 4 provides a plot of these specimens alongside the 87Sr/86Sr values we would expect to see among teeth that mineralized on the dolomites. Apart from the procaviid (SK 4076) and steenbok (SKX 12363) specimens mentioned above, all but one of the incrementally sampled individuals from Swartkrans are consistent with local dolomites. The one Connochaetes individual with some exogenous values (SKW 5946) might have spent a portion of its life on the dolomites, and a portion beyond them; such a pattern is not surprising in such a migratory taxon. All of the Swartkrans cercopithecids that we sampled give values consistent with spending most, if not all, of their lives on the same dolomite substrate that houses Swartkrans. A single exogenous Sterkfontein specimen (Sts 2580, cf. Damaliscus) is clearly not local, but cannot at present be linked to a particular geological substrate that we have sampled. Given the possible migratory behavior of this small alcelaphine, it is conceivable that the individual moved into the Bloubank Valley area after reaching maturity (i.e. after enamel deposition ceased). Of perhaps greater interest is the observation that so few sampled individuals appear to

145

Eagles as accumulating agents

Figure 3. Range plots of biologically available 87Sr/86Sr ratios as determined by plants from a series of geological substrates in the vicinity of the Bloubank Valley, with individually plotted fossil specimen values. Incrementally sampled specimens averaged to derive individual points. Only two Swartkrans specimens are inconsistent with the dolomites, while all of the Sterkfontein samples suggest that they were sourced from a remote geological substrate.

emanate from outside of the Malmani dolomites. Although some animals fall near the extremes of the dolomite values, their numbers are still consistent with spending the majority of their lives in the vicinity of the Bloubank Valley. Otherwise, the majority of taxa are notably homogeneous in their 87Sr/86Sr ratio distributions, consistent with spending their lives in relatively close proximity to the fossiliferous caves. Discussion Skeletal part data and 87Sr/86Sr ratios reveal that the majority of the fossil assemblages discussed here provide little direct evidence of eagle contribution. At Swartkrans, only

the Hanging Remnant shows possible eagle involvement in terms of skeletal part abundance, though Sr isotope data provide only limited support for such a conclusion. Notwithstanding, this assemblage has produced a significant number of juvenile hominins, and it is therefore possible that they might have been captured by eagles. Unfortunately, many of these juveniles are too badly damaged or incomplete to allow for direct investigation of taphonomic traces in the form of bone surface modifications, highlighting the need for the extraction of a taphonomically representative sample of faunal materials. The remaining Swartkrans assemblages do not provide support for substantial eagle involvement. Kromdraai A reveals a preponderance of both procaviid and cercopithecid cranio-

146

de Ruiter et al.

dental remains, consistent with eagle predation. Such a conclusion might explain one of the more curious features of the Kromdraai A assemblage, that being the notable prevalence of cercopithecids. It does little, however, to explain the anomalously high number of equids known from this deposit, and the absence of hominins, despite the large size of the assemblage (MNI>200). As a result, Kromdraai A remains a singularly intriguing deposit that warrants additional investigation. Although the selective collecting procedures of Robert Broom that favored cranial remains might be responsible, it seems unlikely that such a systematic bias against post-cranial remains, in particular of primates, characterized Broom’s work. This is because Broom was very careful about recovering primate fossils (Broom, 1951; Broom & Schepers, 1946), especially as primate post-cranial remains can appear deceptively like hominin fossils, in particular while still encased in breccia. In the final analysis, although eagles might have contributed to the assemblage, they did not capture any juvenile hominins, thus providing no additional support for eagles as significant accumulators of hominins. Kromdraai B, on the other hand, presents little evidence favoring eagle contribution, in particular relating to the cercopithecids. As a result, eagle predation does not appear to explain the anomalously high proportions of cercopithecids recorded in this relatively unique faunal assemblage. The assemblages recovered from Cooper’s D, Sterkfontein Member 5 (Oldowan Infill), Motsetse, Gladysvale, and Plovers Lake reveal little support for eagles as accumulating agents. Member 4 of Sterkfontein, on the other hand, does provide some substantiation of eagle involvement, despite the fact that this assemblage might have been biased as a result of selective early

excavation procedures. While data from Member 3 of Makapansgat are equivocal on the issue of eagle contribution, there is at least one juvenile hominin (MLD 5) from this assemblage that might have fallen within the body size range eagles could transport. Although intriguing, this single specimen from Makapansgat cannot be considered to represent a pattern, as numerous other potential agents are equally likely in the context of this cave. Skeletal part analysis of the cercopithecids from Makapansgat Member 4 is consistent with eagle predation, supporting an earlier assertion by Reed (1996) that eagles might have been responsible for accumulating the smaller mammals in this assemblage. In addition, Reed (1996) noted crush marks on 4 cercopithecid crania that were consistent with eagle talons. It is therefore of great interest to note that all of the A. africanus bearing assemblages in South Africa (Taung, Sterkfontein Member 4, Makapansgat Members 3 and 4) reveal at least the possibility of eagle contribution, begging the question of whether juveniles of this taxon might have been particularly susceptible to eagle predation. Unfortunately, few of the juveniles of A. africanus are complete or undamaged enough to allow close inspection for eagle derived damage. As a result, we return to the evidence initially provided in favor of the Taung Child being accumulated by a large eagle, as this is presently the only A. africanus specimen putatively displaying direct evidence of eagle predation (Berger & Clarke, 1995; Berger & McGraw, 2007). The Taung Child In their initial discussion of eagles as potential predatory agents, Berger & Clarke (1995) relied on 6 principal lines of

147

Eagles as accumulating agents

Figure 4. Plots of 87Sr/86Sr ratios from incrementally sampled specimens relative to the 87Sr/86Sr ratios of the Malmani dolomites; A) bovid and equid data and B) procaviid and cercopithecid data. Note that most specimens are consistent with having lived in the vicinity of the dolomites throughout the period of tooth mineralization. Given different enamel mineralization rates across the taxa studied here, the intervals represented by the ablation tracks do not correspond to consistent lengths of time for the various animals.

evidence to highlight the unique nature of the Taung assemblage: 1) the homogeneous body size of associated animals; 2) the common occurrence of complete crania with attached mandibles; 3) the presence of numerous tortoise carapaces; 4) the presence of large bird eggshells; 5) the lack of additional hominin finds; and, 6) the recovery of the Taung Child from a tufaceous rather than dolomitic cave, perhaps near cliffs that might have provided favorable eagle nesting sites. However, the lack of data regarding the

precise source of the Taung fossils renders several of these criteria uncertain. The Taung Child was initially blasted out of a breccia deposit housed within a tufaceous cave in late 1924, during the course of active mining operations (Dart, 1925; Dart & Craig, 1959). A sample of fossil materials collected by local miners accompanied the Taung Child to Dart’s home in Johannesburg in 1924 (Dart & Craig, 1959; Young, 1925), but it is unclear whether they were derived from the same deposit as the Taung Child, or if they were merely found

148

de Ruiter et al.

in the vicinity of the Taung Child’s blast site (or perhaps even from an entirely remote locale). To quote Broom (1934:480) when discussing the fauna associated with the Taung Child, “The pieces of bone breccia were obtained after the [Taung] skull had been found, and were of course not directly associated with it, but they were from the floor of the same cave.” However, even this latter conclusion of Broom’s is not necessarily the case (McKee & Tobias, 1994). The currently hypothesized locality of the Taung Child was reconstructed from historical records and from photographs taken nearly three years after the Taung Child was initially recovered from an active mine face (McKee, 1993a,b; McKee & Tobias, 1994). As a result, a clear link between the original Taung fauna and the Taung Child cannot be reliably established. Subsequent collections by Hrdlička (1925) likely sampled a nearby deposit, though there does not appear to be a direct link between the presumed sources of the Taung Child and Hrdlička fossils (McKee, 1993a, 1993b; McKee & Tobias, 1994). Further complicating the issue is the fact that most of the original fossils that accompanied the Taung Child to Dart’s home have subsequently been lost. As a result, the first four criteria of Berger & Clarke (1995) do not necessarily relate to the Taung Child itself, as none of the original Taung fauna can be directly associated with the hominin skull. The 5th criterion relies on an absence of evidence, and is thus provides no additional support for eagles as accumulators. The 6th criterion is more speculative, as McKee (1993) has demonstrated that tufa caves can form in the absence of cliffs, therefore there is no direct evidence that suitable cliffs for eagle nests necessarily existed in the immediate vicinity of the Taung Child deposit in the past. Because none of the original Taung fauna can be firmly associated with the

Taung Child, we must rely on the hominin fossil itself to reveal evidence of eagle involvement in its death and deposition. Berger (2006) and Berger & McGraw (2007) compared the patterns of bone surface modification that are characteristically produced in modern eagle derived assemblages to damage seen on the Taung Child. In particular, Berger & McGraw (2007) recognized a series of scratches and perforations on the frontal, face, and orbits of the Taung Child that they considered to be consistent with eagle talon damage. They discounted the likelihood that the scratches on the Taung cranium were the result of Dart’s preparation techniques, in part because of Dart’s (1959) statement that during his preparation of the specimen, “the rock parted”, implying that the breccia covering the left half of the face came away in a solid chunk without removal by chipping with a hammer and needle. However, we recognize damage to the left side of the Taung mandible that is most likely the result of preparation damage (Figure 5). Dart did not detach the mandible from the cranium until July of 1929, thus it was fully articulated at the time of cleaning in 1924. This series of 6 parallel striations therefore raises the possibility that the breccia did not come away from the Taung skull as cleanly as we might suppose based on Dart’s account. As a result, we are still left with the possibility that the damage to the Taung Child could be the result of Dart’s preparation of the specimen. Therefore the evidence supporting an eagle as the accumulator of the Taung Child is currently not conclusive. Conclusions For the majority of the assemblages investigated in this study, evidence in favor of eagles as

149

Eagles as accumulating agents

accumulating agents is not compelling. We would caution that other post-depositional processes might have obscured indications of eagle involvement (e.g. differential preservation or destruction, winnowing, etc.). In addition, the fossil assemblages examined here likely represent palimpsests of time-averaged materials, potentially influencing the bone survival profiles developed in this study. While these biasing factors are beyond the scope of this paper, recent research has demonstrated that taphonomic alterations do not consistently influence the taxonomic composition of the faunal assemblages (de Ruiter et al., 2008b). The consistency of the skeletal part patterns and strontium isotope compositions outlined above also argues against the suggestion that taphonomic biasing has irretrievably obscured potential evidence of eagle contribution. The strongest cases to be made for eagle involvement are in Kromdraai A, Sterkfontein Member 4, and Makapansgat Member 4. Kromdraai A has yet to reveal hominins, thus it provides no additional support for the hypothesis that eagles were significant threats to early hominin juveniles. Sterkfontein Member 4 has potentially been influenced by biased fossil collecting procedures. Combined with the uncertainty over direct evidence of eagle talon damage to the Taung Child itself, and the absence of direct evidence from the putative Taung faunal assemblage, the hypothesized influence of large raptors such as eagles on the evolution of predator avoidance strategies in early hominins remains unsubstantiated. However, the fact that all of the A. africanus-bearing assemblages reveal at least some evidence supporting eagles as accumulators is of significance. The possibility that eagles were significant predators of juvenile hominins remains germane, though the evidence

presented in favor of such a hypothesis is not yet sufficient. We suggest the following lines of evidence be pursued to resolve the issue of the accumulating agent of the Taung Child and perhaps other juvenile hominins: 1) a scanning electron micrograph study of the damage marks on the Taung Child itself to verify talons marks versus preparation damage marks; 2) detailed investigation of taphonomically representative samples from Makapansgat Member 3, Sterkfontein Member 4 and the Hanging Remnant of Swartkrans Member 1; 3) continued preparation of available breccia samples from Kromdraai A and B; 4) strontium isotope sampling of additional procaviid and cercopithecid materials, in particular from Kromdraai A and Sterkfontein Member 4; and, 5) taphonomic investigation of all juvenile hominins currently known from the Bloubank Valley fossil cave infills, including microscopic examination of bone surface modifications. Given the possibility that eagles might have represented a significant factor in the evolutionary development of early hominin behavioral patterns, determining whether they indeed were responsible for any of the early hominin fossils in the South African cave infills is especial importance, and clearly warrants further research. Acknowledgements We would like to thank Stephany Potze, Francis Thackeray, and Teresa Kearny of the Transvaal Museum for allowing access to fossils curated by the Transvaal Museum, Pretoria. Stephany Potze was especially instrumental in facilitating our isotopic sampling of the Swartkrans and Sterkfontein fossils. We thank Mike Raath, Bruce Rubidge, Lee Berger, Christine Steininger, and Bernhard Zipfel for access to fossils housed in the

150

de Ruiter et al.

Figure 5. Probable preparation damage to the left side of the Taung Child’s mandible.

Bernard Price Institute of the University of the Witwatersrand, Johannesburg. Mike Raath and Bernhard Zipfel in particular aided in photographing the Taung Child specimen. Jeff McKee and Lee Berger both presented interesting discussions regarding the possibility of an eagle accumulator at Taung, providing the impetus for writing this paper. Tom Stidham provided information on the large raptors known from the South African cave infills. We would like to thank Petrus le Roux,

Andy Duncan and Maartin de Wit at the Africa Earth Observatory Network (AEON), University of Cape Town, for enabling the laser Sr analysis of the fossils. We thank Daryl Codron, Jacqui Codron, Vaughn Grimes, and Mike Richards for assistance with the plant Sr analysis. We thank Travis Pickering and Amy Egeland for inviting us to participate in this special volume of the Journal of Taphonomy, and for their thoughtful comments on an earlier draft of the paper; we also

151

Eagles as accumulating agents

thank our anonymous reviewer for providing an insightful analysis of this paper. This research was funded by the Wenner-Gren Foundation, the National Science Foundation (USA), the National Research Foundation (RSA), the Palaeontological Scientific Trust, the International Research Travel Assistance Grant programs of Texas A&M University, and the Max Planck Society. This is AEON publication number 79. References Andrews, P. (1990). Owls, caves and fossils. University of Chicago Press, Chicago. Avery, D.M. (2001). The Plio-Pleistocene vegetation and climate of Sterkfontein and Swartkrans, South Africa, based on micromammals. Journal of Human Evolution, 41: 113-132. Benefit, B.R. & McCrossin, M.L. (1990). Diet, species diversity and distribution of African fossil baboons. Kroeber Anthropological Society Papers, 71-72: 77-93. Berger, L.R. (2006). Brief communication: Predatory bird damage to the Taung type-skull of Australopithecus africanus Dart 1925. American Journal of Physical Anthropology, 131: 166-168. Berger, L.R. & Clarke, R.J. (1995). Eagle involvement in the accumulation of the Taung child fauna. Journal of Human Evolution, 29: 275-299. Berger, L.R. & Clarke, R.J. (1996). The load of the Taung child. Nature, 379: 778-779. Berger, L.R. & Lacruz, R. (2003). Preliminary Report on the First Excavations at the New Fossil Site of Motsetse, Gauteng, South Africa. South African Journal of Science, 99: 279-281. Berger, L.R. & McGraw, W.S. (2007). Further evidence for eagle predation of, and feeding damage on, the Taung Child. South African Journal of Science, 103: 496-498. Brain, C.K. (1969). The probable role of leopards as predators of the Swartkrans australopithecines. South African Archaeological Bulletin, 24: 52-55. Brain, C.K. (1980). Some criteria for the recognition of bone-collecting agencies in African caves. In (Behrensmeyer, A.K., & Hill, A.P., eds.) Fossils in the Making. University of Chicago Press, Chicago, pp. 107-130. Brain, C.K. (1981). The Hunters or the Hunted? University of Chicago Press, Chicago.

Brain, C.K. (2004). A taphonomic overview of the Swartkrans fossil assemblages. In (Brain, C.K., ed.) Swartkrans: a cave’s chronicle of early man, 2nd Ed. Transvaal Museum Monograph No. 8. p 257-264. Brain, C.K., Churcher, C.S., Clark, J.D., Grine, F.E., Shipman, P., Susman, R.L., Turner, A. & Watson, V. (1988). New evidence of early hominids, their culture and environment from the Swartkrans Cave, South Africa. South African Journal of Science, 84: 828-835. Broom, R. (1934). On the fossil remains associated with Australopithecus africanus. South African Journal of Science, 31: 471-480. Broom, R. (1951). Finding the missing link. Watts, London. Broom, R. & Schepers, G.W.H. (1946). The South African Fossil Ape-Men, The Australopithecinae. Pretoria, Transvaal Museum Memoir, No. 2. Copeland, S.R., Sponheimer, M., le Roux, P.J., Grimes, V., Lee-Thorp, J.A., de Ruiter, D.J. & Richards, M.P. (2008). Strontium isotope ratios (87Sr/86Sr) of tooth enamel: a comparison of solution and laser ablation multicollector inductively coupled plasma mass spectrometry methods. Rapid Communication in Mass Spectrometry, 22: 3187-3194. Copeland, S.R., Sponheimer, M., Lee-Thorp, J.A., de Ruiter, D.J., le Roux, P.J., Grimes, V., Codron, D., Berger, L.R. & Richards, M.P. (2010). Using Strontium Isotopes to Study Site Accumulation Processes. Journal of Taphonomy, 8 (1-2): 115-127. Cruz-Uribe, K. (1991). Distinguishing hyena from hominid bone accumulations. Journal of Field Archaeology, 18: 467-486. Cruz-Uribe, K. & Klein, R.G. (1999). Hyrax and hare bones from modern South African eagle roosts and the detection of eagle involvement in fossil bone assemblages. Journal of Archaeological Science, 25: 135-147. Dart, R.A. (1925). Australopithecus africanus: the man-ape of South Africa. Nature, 115: 195-199. Dart, R.A. & Craig, D. (1959). Adventures with the Missing Link. Harper and Brothers, New York. Delson, E. (1984). Cercopithecid biochronology of the African Plio-Pleistocene: correlation among eastern and southern hominid-bearing localities. Courier Forschung Institut Senckenberg, 69: 199-218. de Ruiter, D.J. (2001). A methodological analysis of the relative abundance of hominids and other macromammals from the site of Swartkrans, South Africa. Ph.D. Dissertation, University of the Witwatersrand, Johannesburg. de Ruiter, D.J. (2004). Relative abundance, skeletal part representation and accumulating agents of macromammals at Swartkrans. In (Brain, C.K., ed.) Swartkrans: a cave’s chronicle of early man , 2nd Ed. Transvaal Museum Monograph No. 8. pp. 265-278.

152

de Ruiter et al. de Ruiter, D.J. & Berger, L.R. (2000). Leopards as taphonomic agents in dolomitic caves – implications for bone accumulations in the hominid-bearing deposits of South Africa. Journal of Archaeological Science, 27: 665-684. de Ruiter, D.J., Brophy, J.K., Lewis, P.J., Churchill, S.E. & Berger, L.R. (2008a). Faunal assemblage composition and paleoenvironment of Plovers Lake, a Middle Stone Age locality in Gauteng Province, South Africa. Journal of Human Evolution, 55: 1102-1117. de Ruiter, D.J., Sponheimer, M., & Lee-Thorp, J.A. (2008b). Indications of habitat association of Australopithecus robustus in the Bloubank Valley, South Africa. Journal of Human Evolution, 55: 1015-1030. de Ruiter, D.J., Pickering, R., Steininger, C.M., Kramers, J.D., Hancox, P.J., Churchill, S.E., Berger, L.R. & Backwell, L. (2009). New Australopithecus robustus fossils and associated U-Pb dates from Cooper’s Cave (Gauteng, South Africa). Journal of Human Evolution, 56: 497-513. Faure G. & Powell T. (1972). Strontium Isotope Geology. Springer-Verlag: New York. Freedman, L. (1957). The fossil Cercopithecoidea of South Africa. Annals of the Transvaal Museum, 23: 121-262. Gargett, V. (1990). The black eagle: a study. John Voelcker Bird Book Fund, Johannesburg. Grayson, D.K. (1984). Quantitative Zooarchaeology: Topics in the Analysis of Archaeological Faunas. Academic Press, Orlando. Hedenstrom, A. (1995). Lifting the Taung Child. Nature, 378: 670. Hrdlička, A. (1925). The Taungs ape. American Journal of Physical Anthropology, 8: 379-392. Kuhn, B.F., Berger, L.R. & Skinner, J.D. (2009). Examining criteria for identifying and differentiating fossil faunal assemblages accumulated by hyenas and hominins using extant hyenid accumulations. Int. J. Osteoarchaeol. doi: 10.1002/oa.996. Kuman, K. & Clarke, R.J. (2000). Stratigraphy, artifact industries and hominid associations for Sterkfontein, Member 5. Journal of Human Evolution, 38: 827847. Lacruz, R.S. & Maude, G. (2005). Bone accumulations at brown hyaena (Parahyaena brunnea) den sites in the Makgadikgadi Pans, northern Botswana: taphonomic, behavioral and palaeoecological implications. Journal of Taphonomy, 3: 43-54. Lacruz, R.S., Brink, J.S., Hancox, P.J., Skinner, A.R., Herries, A., Schmid, P. & Berger, L.R. (2002). Palaeontology and geological context of a middle Pleistocene faunal assemblage from the Gladysvale Cave, South Africa. Palaeontologia Africana, 38: 99-114.

Maguire, J.M., Pemberton, D. & Collett, M.H. (1980). The Makapansgat Limeworks grey breccia: hominids, hyaenas, hystricids or hillwash? Palaeontologia Africana, 23: 75-98. McGraw, W.S., Cooke, C. & Schultz, S. (2006). Primate Remains from African Crowned Eagle (Stephanoaetus coronatus) Nests in Ivory Coast’s Tai Forest: Implications for Primate Predation and Early Hominid Taphonomy in South Africa. American Journal of Physical Anthropology, 131: 151-165. McKee, J.K. (1993a). Formation and geomorphology of caves in calcareous tufas and implications for the study of the Taung fossil deposits. Transactions of the Royal Society of South Africa, 48: 307-322. McKee, J.K. (1993b). Faunal dating of the Taung hominid fossil deposit. Journal of Human Evolution, 25: 363-376. McKee, J.K. (2001). The Taung raptor hypothesis: caveats and new evidence. American Journal of Physical Anthropology, Supplement 32: 107. McKee, J.K. & Tobias, P.V. (1994). Taung stratigraphy and taphonomy: preliminary results based on the 1988-1993 excavations. South African Journal of Science, 90: 233-235. Mitani, J.C., Sanders, W.J., Lwanga, J.S. & Windfelder, T.L. (2001). Predatory behavior of crowned hawkeagles (Stephanoaetus coronatus) in Kibale National Park, Uganda. Behavioral Ecology and Sociobiology, 49: 187-195. Newman, R. (2004). The incidence of damage marks on Swartkrans fossil bones from the 1979-1986 excavations. In (Brain, C.K. ed.) Swartkrans: a cave’s chronicle of early man, 2nd Ed. Transvaal Museum Monograph No. 8. p 217-228. Pickering, T.R. (1999). Taphonomic interpretations of the Sterkfontein early hominid site. Ph.D. Dissertation, University of Wisconsin, Madison. Pickering, T.R. (2002). Reconsideration of criteria for differentiating faunal assemblages accumulated by hyaenas and hominids. International Journal of Osteoarchaeology, 12: 127-141. Pickering, T.R., Dominguez-Rodrigo, M., Egeland, C.P. & Brain, C.K. (2004). Beyond leopards: tooth marks and the contribution of multiple carnivore taxa to the accumulation of the Swartkrans Member 3 fossil assemblage. Journal of Human Evolution, 46: 595604. Plug, C. & Plug, I. (1990). MNI counts as estimates of species abundance. South African Archaeological Bulletin, 45: 53-57. Reed, K.E. (1996). The paleoecology of Makapansgat and other African Plio-Pleistocene hominid localities. PhD Thesis, State University of New York at Stony Brook.

153

Eagles as accumulating agents Reed, K.E. (1997). Early hominid evolution and ecological change through the African PlioPleistocene. Journal of Human Evolution, 32: 289322. Sanders, W.J., Trapani, J. & Mitani, J.C. (2003). Taphonomic aspects of Crowned Hawk-Eagle predation on monkeys. Journal of Human Evolution, 44: 87-105. Sillen, A., Hall, G., Richardson, S. & Armstrong, R. (1998). 87Sr/86Sr ratios in modern and fossil foodwebs of the Sterkfontein Valley: Implications for early hominid habitat preference. Geochimica et Cosmochimica Acta, 62: 2463-2473. Skinner, J.D. & Smithers, R.H.N. (1990). The mammals of the southern African subregion. University of Pretoria Press, Pretoria. Steyn, P. (1982). Birds of prey of southern Africa. David Phillip, Cape Town.

Sutton, M., Pickering, T.R., Pickering, R., Brain, C.K., Clarke, R.J., Heaton, J. & Kuman, K. (2008). Newly discovered fossil- and artifact-bearing deposits, uranium-series ages, and Plio-Pleistocene hominids at Swartkrans Cave, South Africa. Journal of Human Evolution, 57: 688-696. Vrba, E.S. (1976). The fossil Bovidae of Sterkfontein, Swartkrans and Kromdraai. Transvaal Museum Memoir No. 21, Pretoria. Watson, V. (2004). Composition of the Swartkrans bone accumulations, in terms of skeletal parts and animals represented. In (Brain, C.K. ed.) Swartkrans: a cave’s chronicle of early man, 2nd Ed. Transvaal Museum Monograph No. 8. p 35-73. Young, R.B. (1925). The calcareous tufa deposits of the Campbell Rand, from Boetsap to Taungs Native Reserve. Transactions of the Geological Society of South Africa, 28: 55-67.

154