PALAIOS, 2010, v. 25, p. 439448 Research Article DOI: 10.2110/palo.2009.

p09-116r

RECONSTRUCTING THE PALEOECOLOGY OF TAUNG, SOUTH AFRICA FROM LOW MAGNIFICATION OF DENTAL MICROWEAR FEATURES IN FOSSIL PRIMATES
FRANK L9ENGLE WILLIAMS* and JAMES W. PATTERSON
Department of Anthropology, Georgia State University, Atlanta, Georgia 30303, USA e-mail: frankwilliams@gsu.edu

ABSTRACT Taung, South Africa yielded the first Pliocene Hominini fossil, Australopithecus africanus, recovered from a lime quarry in 1924. To identify whether the habitat of the site differed from present-day conditions, dietary preferences of fossil papionins from Taung, including Parapapio antiquus (n 5 8), Papio izodi (n 5 12), and indeterminate specimens (n 5 10) were examined under low magnification to discern patterns of dental microwear. The comparative fossil sample from Sterkfontein Member 4 includes Parapapio broomi (n 5 10) and Parapapio jonesi (n 5 5). Extant Papio ursinus (n 5 20), a savannadwelling baboon from South Africa, provides a modern analogue. Six dental use-wear scars on the paracone of the second molar (M2) were recorded and the data analyzed using ANOVA with Tukeys test to detect whether group differences were present for each feature; linear regression identified significant covariation of microwear features. Principal components analysis and discriminant function analysis were utilized to identify species-specific dietary signals. Extant Papio ursinus is separated from the extinct taxa solely by a relatively greater number of fine scratches with respect to the other microwear features. Papio izodi overlaps primarily with extant Papio and secondarily with Parapapio, which forms a more discrete grouping that includes Parapapio antiquus from Taung. A wetter, more closed environment is suggested for Taung and Sterkfontein Member 4 compared to the habitat of present-day central South Africa. INTRODUCTION The limestone cave deposits of the Transvaal and adjoining regions of South Africa preserve a wide variety of fauna dating to the Plio Pleistocene (, 2.51.5 Ma). Foremost among these are the australopiths, which provide evidence of early hominin evolution. Unique among these specimens, the famous Taung child is the first Australopithecus africanus specimen discovered and the only hominin specimen found at Taung. Like other South African sites, Taung is difficult to date chronometrically. Nevertheless it has been assigned a late Pliocene age on the basis of first and last appearance dates of extinct taxa correlated with East African controls. Fossil papionin monkeys, well represented across East and South African sites, have factored heavily in these analyses, yielding a date for Taung of ,2.3 Ma (Delson, 1984). As they evolved concurrently and exploited similar environments, papionins have proven to be useful hominin analogues (Elton, 2007; Hughes et al., 2008). Taung fossil papionins Parapapio antiquus (herein Pp.) and Papio izodi (herein P.) provide insight into the ecological conditions experienced by the four-year-old Taung child. Identifying the diets of these papionin specimens, through the examination of dental microwear using low-magnification stereomicroscopy, presents a novel approach to reconstructing the paleoecology of Taung.

Previous Habitat Reconstructions for South Africa The discovery of the Taung child led Dart (1926) to develop the savanna hypothesis, i.e., that inhabiting an open, savanna environment was the primary factor in the origin and evolution of hominins, assuming that the environment of the Taung child was similar to the modern dry and open scrub brush typical of central South Africa. There is abundant research, however, that indicates substantial variability in regional and global climate during this period. While the savanna hypothesis does not reflect the variability observed in the paleoclimate, the general trend toward drier environments during the Plio Pleistocene in southern Africa has undoubtedly influenced hominin evolution, and this appears to be the case with a variety of taxa (deMenocal, 1995). Yet southern Africa presents some difficulties in paleoecology due to the irregular and complex depositional history of the cave sites. Furthermore, at least four global climatic processes had some bearing on regional climatic variation in Africa, including (1) orbital forcing; (2) the emergence of C4-dominated biomes, a gradual process beginning ,8 Ma with substantial regional variation; (3) intensification of Northern Hemisphere glaciation (INGH), occurring ,2.5 Ma; and (4) the development of Walker Circulation, easterly trade winds across the Pacific Ocean that maintain sea-surface temperatures (Maslin and Christensen, 2007). The interaction of these processes appears to account for much of the observed fluctuations in the climate of PlioPleistocene southern Africa. These global climatic processes directly contributed to the regional fluctuations and the general drying trend observed in southern Africa (Fig. 1). Analyses of stable carbon isotopes have provided evidence that variation in biome composition is related to changes in global climate affecting patterns of temperature and rainfall in southern Africa. This transition has been highly variable across the region with contractions and expansions of C4 and C3 biomes. A general trend toward drier conditions, characterized by the expansion of C4 vegetation, is observed in the region. Isotopic studies of two speleothems at Makapansgat provide evidence of the initial shift from C3 to C4 vegetation (Hopley et al., 2007). A late Mioceneearly Pliocene speleothem (Collapsed Cave) indicates large quantities of C3 vegetation, while a 1.7 Ma speleothem indicates a marked increase in C4 vegetation with variability of the proportions of C3/C4 vegetation throughout its formation. More isotopic evidence from Makapansgat, derived from the tooth enamel of A. africanus, indicates that 25%50% of its diet consisted of C4 food sources, either directly through the consumption of C4 plants or indirectly through the exploitation of animals with a diet of C4 plants (Sponheimer and Lee-Thorp, 1999). This also appears to be the case for A. africanus from Sterkfontein Member 4 as well, with ranges of 30% 60% reported (van der Merwe et al., 2003). Dietary reconstructions based on dental morphology suggest Australopithecus subsisted on hard foods requiring little incisal preparation, possibly fruits with hard husks or seed pods (Kay, 1985). It appears that early hominins, both australopiths and early Homo, exploited C4 resources whether their environments were closed or open (Sponheimer and Lee-Thorp, 2003; Lee-Thorp et al., 2007).
0883-1351/10/0025-0439/$3.00

* Corresponding author.

Copyright

2010, SEPM (Society for Sedimentary Geology)

440

WILLIAMS AND PATTERSON

PALAIOS

FIGURE 1Summary diagram of the chronological orientation of Taung and Sterkfontein Member 4 using Delsons African cercopithecid zones (Delson, 1984). Climatic data is also shown, from left to right: (a) Pliocene to recent geomagnetic polarity; (b) African climate variability, representing the influence of precessional insolation forcing of monsoonal climate and glacial cycles (beginning after 2.8 Ma and intensifying after 1 Ma) on the increasing aridity of African climates (differences along the horizontal serve to visually distinguish major time intervals); (c) East African soil carbonate; (d) d13C illustrating the shift from woodland to grassland in this area [the solid boxes represent data from Cerling and Hay (1988); the solid triangles are from Cerling (1992); and the open circles are from Wynn (2000)], d18O showing glacial cover; (e) African cercopithecid zones (where U indicates the upper and L indicates the lower levels of each zone), and likely geochronological position of (f) Taung and (g) Sterkfontein. Columns a, e, f, and g from Ciochon (1993; used with permission from the University of California Press); columns b, c, and d from deMenocal (2004; reprinted with permission from Elsevier).

Se galen et al. (2007) provide isotopic evidence, from pedogenic and biomineral carbonate d13C, suggesting that C4 vegetation was present in southern Africa during the early Pliocene, only coming to dominate ecosystems during the early Pleistocene, with extensive regional variability (deMenocal, 2004). These climatic changes appear to be largely related to patterns of rainfall. Exemplifying this relationship, oceanic core samples from the western coast of South Africa (ODP leg 175 Site 1085), exhibit four peaks in the biogenic and terrigenous components of the sediment, reflecting increased preservation and higher productivity at 3.2, 3.0, 2.4, and 2.25 Ma (Christensen et al., 2002). Eolian and fluvial processes are implicated in the sedimentation while fluctuation of sediment composition is correlated with monsoonal activity, indicating a wetter climate around the peaks. Periods of increased wetness are also indicated by an examination of younger deposits at Gladysvale when compared to sedimentation at older sites (Pickering et al., 2007). Effective precipitation seems to be the primary factor affecting the type of sedimentation at these sites. Flowstone deposition is correlated with greater effective precipitation and clastic deposition with shorter arid periods. Isotopic evidence implies an increase in C3 vegetation outside of the caves during the periods of flowstone deposition suggesting that the South African deposits were episodic in nature, representing relatively short geological periods of time. Pollen recovered from hyena coprolites found at Equus Cave, Taung, provides further evidence for climate reconstruction (Scott, 1987). A shift from cool and moderately humid habitats supporting open grasslands with small shrubs and occasional trees occurred during the

late Pleistocene, followed by a warmer and drier environment supporting shrubby karoid (Karoo-type) vegetation with open Acacia savanna. A thornveld biome (open-mixed bush and tree savanna) is evidenced in the palynological record from the terminal Pleistocene to the present. Scott (1987) suggests denser woodland was present earlier in the Pleistocene; however, this portion of the deposit could not be accurately dated. Another indirect reconstruction of Taung during the PlioPleistocene suggests that a large raptor played a major role in the accumulation of fossil fauna at the site based upon taphonomic signals of raptor predation and the type of bones preserved at Taung (Berger and Clarke, 1995). A more recent study of the taphonomic aspects of crowned eagle predation reinforces this argument (Sanders et al., 2003). The involvement of a raptor in the accumulation at Taung implies a woodland or forest environment at the site. Taung and Sterkfontein Member 4 appear to have accumulated under somewhat similar climates. Faunal assemblages from Sterkfontein suggest open woodland with bushveld (mixed grassland and woodland typical of the Transvaal) and thicket areas (Reed, 1997), medium density woodland (Vrba, 1975), open woodland to forest (McKee, 1991), moderately open savanna (Vrba, 1985), open savanna (Benefit and McCrossin, 1990), and a succession of riverine grasslands to open savanna (Avery, 2001). Dating of Taung and Sterkfontein Member 4 Taung was once thought to be the oldest australopith site and at ka deposits are other times the youngest (McKee, 1993). The Hrdlic considered to be the youngest at Taung and McKee (1993) suggests that

PALAIOS

PALEOECOLOGICAL RECONSTRUCTION OF TAUNG, SOUTH AFRICA

441

FIGURE 2Maps of South Africa showing simulated vegetation cover for the mid-Pliocene, glacial and interglacial periods, and the present day, along with d18O data, illustrating paleoenvironmental variability; T 5 Taung site; S 5 Sterkfontein Member 4. Paleoclimate models modified after Hughes et al. (2008); present-day vegetation cover adapted from Cowling et al. (2008); d18O values adapted from deMenocal (2004).

all of the cercopithecid specimens except for Parapapio broomi (or Pp. antiquus) are associated with these deposits. Parapapio specimens are associated with the Dart (1926) deposits, which are seen as the likely origin of the australopith specimen and as being no younger, and likely ka deposits. In faunal comparisons between Taung older than the Hrdlic and the other sites, Makapansgat Member 34 and Sterkfontein Member 4 emerge as the most similar. McKee (1993) suggests that considerable time depth characterizes many of the deposits, but establishes a range of dates for each species represented at Taung, reaching a terminus of 2.62.4 Ma, roughly contemporaneous with Sterkfontein Member 4 (see also McKee, 1996; Berger et al., 2002). ka deposits Tobias et al. (1993) argue that parts of the Dart and Hrdlic are associated and accept Delsons (1984) estimate of 2.3 Ma for Taung. The climatic and chronometric information relating to Taung and Sterkfontein Member 4 are summarized in Figure 1. Simulations of climatic conditions and vegetation cover for the mid-Pliocene glacial conditions and interglacial conditions illustrate the effects of the INHG on the paleoclimate of South Africa (Fig. 2). These simulations were produced with HadCM3 and BIOME 4 (Hughes et al., 2008), and when compared to the d18O record, the frequency of vegetation shifts from glacial to interglacial periods is apparent.

Papionin Evolution in Southern Africa Gear (1926) was the first to provide preliminary descriptions of papionin specimens found at Taung and concluded that there were two forms present. The differentiation of Parapapio from Papio has relied on morphological differences including a shortened muzzle, lack of an anteorbital drop, a sloping face, and the reduction of the hypoconulid of M3, among other traits (Williams et al., 2007). Gilbert (2007) has suggested that specimens from Taung referred to Pp. antiquus should be reclassified as Procercocebus antiquus to support evidence of polyphyly in mangabeys. This change is of particular interest to this study as this species is only found at Taung. Although Pp. antiquus exhibits a microwear signal distinct from those species from Sterkfontein Member 4, substantial overlap across the genus does not support this new taxonomic attribution. Approaches to Examining Dental Microwear Various methods have been employed in order to understand how dental microwear preserves information concerning food processing and consumption; each has strengths and weaknesses that can be summarized using three criteria: (1) resolution, referring to the degree

442

WILLIAMS AND PATTERSON

PALAIOS

of detail detected and the type of information collected; (2) expense, in terms of both resources and time; and (3) reliability. Methods for observing dental microwear at different scales include low-magnification stereomicroscopy (LMS), scanning electron microscopy (SEM), and confocal microscopy (CM). LMS was one of the earliest methods employed to examine dental microwear. While LMS provides a relatively low resolution (353), it features high repeatability and low observer error (Semprebon et al., 2004). SEM provides the highest resolution (e.g., Teaford et al., 1996) and micrographs can be analyzed with various computer programs (Ungar, 2002) to compare dimensions and frequencies (Walker et al., 1978; Gordon, 1984; Teaford et al., 1996; Grine et al., 2002). CM is a more recent alternative and has proven effective in differentiating the fracture properties of foods ingested by extinct hominins (Scott et al., 2005, 2006). All of these methods are capable of covering a similar area of the tooth and have been demonstrated to be effective in determining some of the physical properties of the food resources exploited by an individual. Further complicating the study of dental microwear are taphonomic alterations, which can render dental microwear features unreadable; however, King et al. (1999) examined the effect of taphonomic processes on dental microwear and found that microwear was either obliterated by the taphonomic process or survived unchanged. The resulting obliteration is easily distinguishable and is the primary determinant of the appropriateness of a tooth for dental microwear examination. Additionally, mandibular corpus width and depth and mandibular length have been demonstrated to have predictable effects on molar microwear by variably affecting the amount of compression and shear and therefore the size of microwear features in humans (Mahoney, 2006). The unintended ingestion of materials during feeding also appears to affect dental microwear signals. Mainland (2003) found that the ingestion of soil during grazing and browsing could affect the dental microwear signals of Gotland sheep (Ovis aries). Similar effects have been noted in Papio ursinus, where the ingestion of large amounts of exogenous grit from the exploitation of hypogeous foods during the dry season resulted in larger microwear features and higher numbers of pits (Daegling and Grine, 1999). Dental microwear evidence can provide information about the dietary differences between specimens and the types of resources that are available in a given environment. To the degree that there is a phylogenetic component to similarity in diet, dental use-wear scars may reflect phylogenetic relationships among the fossil primates. Comparing the dental microwear of fossil Parapapio and Papio to extant Papio ursinus can help to clarify differences between the types of foods utilized by each group. These potential differences are employed here to address whether the habitat of the Taung child, A. africanus, differed from the fairly xeric ecology characterizing the Cape region of present-day South Africa. MATERIALS AND METHODS Resin dental casts were obtained from dental impressions molded directly on the molars of specimens using polyvinylsiloxane (President Plus Jet Regular Body-Surface Activated 4605, Coltene/Whaledent Inc.). When necessary, previously applied shellac was removed from the specimens using Zip Strip (Star Bronze Co.), followed by the application and drying of 95% isopropyl alcohol to remove the Zip Strip (Semprebon et al., 2004). Dental microwear was observed on epoxy resin casts of the dental molds. These casts were created by mixing resin and hardener (EpoKwick, made by Buehler Ltd.). The catalyzed casting material was centrifuged to remove air pockets. The casting resin was then poured into putty crucibles that were stabilized using putty hardener (Coltene/Whaledent Inc.). A total of 65 specimens were examined, consisting of Pp. antiquus (n 5 8), P. izodi (n 5 12), and indeterminates (n 5 10) from Taung; Pp. broomi (n 5 10) and Pp. jonesi (n 5 5) from Sterkfontein Member 4;

and the only living papionin native to the Cape region of southern Africa, P. ursinus (n 5 20). Papio ursinus from the Western Cape, Orange Free State, and Transvaal region were examined from the collections of the South African Museum (Cape Town). All Pp. broomi and Pp. jonesi specimens were molded at the Transvaal Museum (Pretoria). Fossils attributed to Pp. antiquus, P. izodi, and indeterminate Taung specimens were molded at the University of the Witwatersrand (Johannesburg), the Transvaal Museum, and the American Museum of Natural History (AMNH, New York). The specimens from the AMNH were on loan from the University of California (Berkeley) to Eric Delson at the time. All of the specimen numbers with associated repositories are listed in Supplementary Data 11. To safeguard against misinterpreting postmortem taphonomic features as resulting from diet, El-Zaatari et al. (2005) examined resin dental casts under light microscopy before preparing specimens for SEM, and a similar procedure was followed here. Evidence of postmortem taphonomy included shiny or polished surfaces or an overabundance of refractive (small) pits indicative of highly crenulated surfaces which are atypical of dietary scars. Extant Papio ursinus was observed to identify patterns of dental microwear that are relatively free of postmortem taphonomy for comparison with the fossil specimens. Areas of the occlusal surface that were obviously damaged from casting, molding, or taphonomic processes were excluded from the study. Observations of Dental Microwear Using LMS, microwear features were distinguished by their light refractive properties through the use of external illumination with a fiber-optic light source positioned obliquely to the occlusal plane of a resin dental cast. Sampling was limited to a 0.4 mm2 area on the paracone of the second molar and six variables were counted per specimen: (1) small pits, (2) large pits, (3) fine scratches, (4) coarse scratches, (5) hypercoarse scratches, and (6) puncture pits (Godfrey et al., 2004; Semprebon et al., 2004). Two trials were conducted per specimen and the average value for each feature was included in subsequent analyses (see Supplementary Data 11 for these data and Supplementary Data 21 for descriptive statistics by taxon). Small pits are shallow, easily refract light, and the entire range of them can be observed when an external light source is maneuvered obliquely. Large pits are deeper, less refractive, and at least two times larger than small pits. Excessive pitting has been associated with hardobject exploitation in both ungulates and primates (Semprebon et al., 2004). Fine scratches, possibly resulting from the silicates in grasses, are shallow, refractive, and associated with graminivory (Godfrey et al., 2004). Coarse scratches are also somewhat refractive, but they are deeper and reflect coarser dietary items than those typical of grasses. Hypercoarse scratches and puncture pits are much heavier, nonrefractive features. Hypercoarse scratches are steep-sided furrows etched onto the occlusal surface, possibly from a particle being dragged along the molar cusps during mastication. Particles might include grit adhering to rhizomes, tubers, corms, or bulbs as well as particles from the hard endocarp of seed coats. Puncture pits (and smaller pits) may also result from seed predation, hard-object feeding, or possibly from the inadvertent mastication of exogenous grit and small rocks during the consumption of underground storage organs (USOs); these scars have irregular edges and can be described as perforations of the occlusal surface. A large puncture pit can be 25% of a 0.4 mm2 ocular reticle. The paracone is a large mesial-buccal cusp on the maxillary second molar (M2). It exhibits a broad surface that is only moderately steep and does not extend deeply into the talon basin. Therefore, the paracone is relatively flat and thus captures the dietary signal of foods that are crushed and pulverized rather than sheared. Use-wear scars were counted along the broad flat face of the paracone unless
1

www.paleo.ku.edu/palaios

PALAIOS

PALEOECOLOGICAL RECONSTRUCTION OF TAUNG, SOUTH AFRICA

443

FIGURE 3Bivariate comparison of number of small pits versus number of fine scratches. For each taxon, 95% confidence ellipses around group centroids are shown. Star 5 P. ursinus (P. u.); X 5 P. izodi (P. i.); cross 5 Pp. antiquus (Pp. a.); circle 5 Pp. jonesi (Pp. j.); triangle 5 Pp. broomi (Pp. b.); rectangle 5 indeterminate (indet.).

FIGURE 4Bivariate comparison of small pit and coarse scratch counts, showing 95% confidence ellipses around group centroids. Star 5 P. ursinus (P. u.); X 5 P. izodi (P. i.); cross 5 Pp. antiquus (Pp. a.); circle 5 Pp. jonesi (Pp. j.); triangle 5 Pp. broomi (Pp. b.); rectangle 5 indeterminate (indet.).

taphonomic evidence obscured this surface. The cuspal edge was avoided to standardize the observations. All of the data were collected by one of the authors (JWP) to eliminate possible intra-observer error. An error study was performed on microwear counts of 10 specimens. The microwear features of each specimen were counted on two separate occasions and the resulting counts were examined by a t-test and a Mann-Whitney U test to measure the intra-observer error, yielding insignificant differences between the trials (p values 5 0.8570.264 for t-test and 0.300.875 for the Mann-Whitney U). We acknowledge that one can have high error and no significant difference between two trials. To address this issue, JWP counted microwear features on eight teeth twice in succession and the mean percent difference for each microwear feature was calculated (100 3 absolute difference between trial1 and trial2/mean of two observations). The smallest mean percent difference for the two trials was lowest for small pits (4.8%) and coarse scratches (6.7%) and somewhat higher for fine scratches (11%) and large pits (11.7%). Both puncture pits and hypercoarse scratches had mean percent differences of 0% owing to the scarcity of these heavy use-wear scars in the specimens. These absolute error rates are consistent with other studies of dental microwear (Grine et al., 2002; Semprebon et al., 2004). As Semprebon et al. (2004) found in their assessment of the validity of LMS in determining primate diets, graminivory is associated with high frequencies of scratches. In their study, graminivores made up the lower right angle of what they called the use-wear trophic triangle, while browsers occupied the left corner and mixedhard-object feeders were at the apex. In general their results indicate that primate graminivores, hard-object specialists, and leaf browsers are distinguished by relative frequencies of pits and scratches, i.e., high numbers of scratches and low numbers of pits for graminivores, low numbers of scratches and pits for folivores; and high numbers of scratches and pits for seed predators and extractive foragers. It is important to note that while primates often concentrate on certain resources, many of their diets can be quite flexible and difficult to categorize. One-way analysis of variance (ANOVA), with posthoc Tukeys Honestly Significant Differences (HSD), was employed to compare differences between the four taxa separately for each microwear feature. Linear regressions were performed to assess the significance of the

relationship between each pair of microwear features. For significant associations between traits, a bivariate plot was used to show the overall differences between the taxa with respect to these features, and linear regressions with 95% confidence ellipses around group centroids were used to assess differences (Figs. 34). Since the heavier microwear features, such as hypercoarse scratches and puncture pits, were only rarely observed in these taxa, ANOVA and linear regression for these traits were impractical. In this case we utilized a bar graph to show the mean counts of these features within two standard deviations. Principal components analysis (PCA) of four common microwear features shows the patterns of variance and covariance within the dataset. Canonical scores generated from a discriminant function analysis show the placement of indeterminate specimens from Taung. Finally, total pit and scratch counts were compared to an approximation of the use-wear trophic triangle of Semprebon et al. (2004). RESULTS Analysis of Variance The ANOVA results (Table 1) demonstrate significant differences exist across taxa for each dental microwear feature. The F values show that substantial between-group differences are evident; however, the F value for fine scratches is an order of magnitude larger than those associated with the other use-wear features. Tukeys Honestly Significant Differences Table 2 shows that indeterminate specimens exhibit significant differences in small pits compared to Pp. antiquus and Pp. broomi. The P. ursinus sample is significantly different in counts of small pits from all Parapapio species, including Pp. antiquus from Taung. The fossils attributed to Pp. broomi show a significant difference in presence of large pits compared to P. ursinus and indeterminate Taung specimens. Indeterminate specimens are also significantly different in the presence of fine scratches when compared to Pp. broomi and P. ursinus (Table 3); however, it is P. ursinus that is significantly different from all of the fossil

444

WILLIAMS AND PATTERSON

PALAIOS

TABLE 1ANOVA results for each dental microwear feature compared to taxon (significant results in bold). F value Small pits Large pits Fine scratches Coarse scratches 7.736 2.651 28.204 5.159 p value 0 0.031 0 0.001

TABLE 3Tukeys Honestly Significant Differences Post-Hoc ANOVA for fine scratches (below) and coarse scratches (above) by taxon (significant results in bold). Pp. 5 Parapapio; P. 5 Papio. P. izodi 0.574 0.872 0.999 0.001 0.853 0.001 0.703 0.004 1 0.001 Pp. antiquus 0.242 0.004 0 0.71 0.001 Pp. broomi 1 0.773 0.136 0.083 0.015 Pp. jonesi 0.075 0.001 0.963 0.04 0.001 P. ursinus 1 0.248 0.221 0.989 0.065

Indeterminate Indeterminate P. izodi Pp. antiquus Pp. broomi Pp. jonesi P. ursinus

taxa for the presence of fine scratches. The sample of Pp. broomi also shows a significant difference in fine scratches compared to Pp. antiquus and P. izodi, suggesting that distinctions between the signals from Sterkfontein and Taung also exist. Reinforcing this distinction is the significantly great number of coarse scratches in P. izodi from Taung compared to Pp. jonesi from Sterkfontein. Sterkfontein specimens are not uniform, however, as significant differences exist between Pp. jonesi and Pp. broomi in the presence of coarse scratches. Linear Regression Linear regression of each pair of microwear features yielded significant values only for the comparison between small pits and fine scratches (p 5 0.024), and between small pits and coarse scratches (p 5 0.036; Table 4). These significant differences are emphasized in bivariate plots to show the extent to which differences among taxa can be demonstrated with these pairs of use-wear features. The 95% confidence ellipses around group centroids show that P. ursinus is distinct with its higher number of fine scratches, while Pp. broomi is divergent in its presence of small pits (Fig. 3). The other taxa show a large degree of overlap. For the comparison between small pits and coarse scratches, Pp. jonesi is distinct from other taxa and only overlaps with Pp. antiquus (Fig. 4). The range for indeterminate specimens completely surrounds the relatively small confidence ellipse of P. ursinus and extends into that of P. izodi and Pp. broomi, suggesting that some indeterminates may be classified as either Papio or Parapapio. Mean Counts of Hypercoarse Scratches and Puncture Pits The heaviest microwear features are relatively rare in the taxa examined and mean counts with two standard deviations are shown in Figure 5. The largest number of heavy microwear features is exhibited by P. izodi, followed by Pp. broomi. The Taung indeterminate specimens and Pp. antiquus resemble one another in the number of hypercoarse scratches and the lack of puncture pits, whereas P. ursinus exhibits the opposite configuration with some puncture pits but no hypercoarse scratches. Principal Components Analysis In Figure 6, a PCA is shown and the associated component loadings are listed in Table 5. PC Axis 1 explains 37.5% of the variation, while PC Axes 2 and 3 each explain 26.0% of the variation (PC Axis 3 not shown).
TABLE 2Tukeys Honestly Significant Differences Post-Hoc ANOVA for small pits (below) and large pits (above) by taxon (significant results in bold). Pp. 5 Parapapio; P. 5 Papio. P. izodi 1 0.636 0.008 0.006 0.096 0.999 0.219 0.209 0.662 0.261 Pp. antiquus 1 1 1 0.998 0.001 Pp. broomi 0.05 0.07 0.083 0.999 0.001 Pp. jonesi 0.957 0.957 0.97 0.612 0.026 P. ursinus 1 1 1 0.029 0.977

The first axis largely separates extant baboons (P. ursinus) with positive scores and fossil taxa with negative scores. The primary exception is the placement of the fossil P. izodi, which overlaps slightly within the 95% confidence ellipse of the group centroid for P. ursinus. Some individuals with scores close to zero were difficult to classify and included primarily the indeterminate specimens, suggesting that dental microwear mimics the morphological ambiguities of these specimens. Parapapio antiquus overlaps extensively with Pp. jonesi and the two exhibit a more eclectic dental microwear signal on PC Axis 1 compared to Pp. broomi from Sterkfontein and P. izodi from Taung. Some indeterminate specimens fall within the range of Pp. antiquus (T 125958, T 56658, and T 13), while others share more similarities with P. izodi. These observations are reflected in the high negative loading of small pits and high positive loadings of fine and coarse scratches on PC Axis 1 (Table 5). Along PC Axis 2, Pp. broomi from Sterkfontein is distinct from indeterminate individuals and P. izodi from Taung due to the greater proportion of large pits (Table 5). The Taung fossils, including Pp. antiquus, P. izodi, and indeterminate specimens, are not well separated from one another. The Pp. antiquus sample shares substantial overlap with Pp. broomi and Pp. jonesi from Sterkfontein, but is not distinct from fossil and extant Papio. The sample of Pp. antiquus has a larger range of values than P. izodi on PC Axis 2, overlapping somewhat more with Pp. broomi from Sterkfontein than other fossil primates from Taung. Discriminant Function Analysis Canonical scores axes derived from discriminant function analysis, plotted with 95% confidence ellipses around group centroids, show clear distinctions between the taxa (Fig. 7). The fossils are largely distinct from extant P. ursinus along the first axis, based on the relatively large number of fine scratches compared to other microwear features for this taxon (Table 5). Parapapio broomi is isolated from Pp. jonesi on canonical scores axis 1; these two Sterkfontein taxa are also distinct from the Taung fossils which show extensive overlap. The relative number of coarse scratches and small pits with respect to the other features helps to explain the positive projection of the taxa from Taung (Pp. antiquus, P. izodi, and indeterminates). Along the second axis, P. ursinus is largely distinct from Parapapio, particularly Pp. broomi and Pp. jonesi, by the relatively large number of small pits in the latter two taxa. Also along the second axis, P. ursinus and P. izodi are similar, along with most indeterminate specimens from Taung, primarily due to the relatively large number of coarse scratches (Table 5).
TABLE 4Linear regression results for each pair-wise comparison of dental microwear features (significant results in bold). F value Small pits vs. large pits Small pits vs. fine scratches Small pits vs. coarse scratches Large pits vs. coarse scratches Coarse scratches vs. fine scratches 1.476 5.312 4.592 0.01 2.387 p value 0.229 0.024 0.036 0.921 0.127

Indeterminate Indeterminate P. izodi Pp. antiquus Pp. broomi Pp. jonesi P. ursinus

PALAIOS

PALEOECOLOGICAL RECONSTRUCTION OF TAUNG, SOUTH AFRICA

445

TABLE 5Component loadings for principal components axes and canonical discriminant functions standardized by within variance for canonical scores axes. Principal Principal Canonical components axis 1 components axis 2 scores axis 1 Small pits Large pits Fine scratches Coarse scratches 20.769 20.192 0.67 0.651 0.193 0.937 0.252 0.245 0.15 20.157 21.021 0.278 Canonical scores axis 2 0.854 0.488 0.17 21.161

range of primate grazers characterized by a high scratch and low pit count (Semprebon et al., 2004). Additionally, none of the specimens examined exhibit relatively large numbers of both pits and scratches typical of seed predators and hard-object specialists. All of the fossil specimens exhibit a greater number of pits than observed in Semprebon et al.s primate leaf browsers (Fig. 8). DISCUSSION It is generally possible to distinguish fossil from extant taxa and even Parapapio from fossil Papio based on dental microwear signals. This implies a difference in the exploitation or availability of resources between the extinct and extant taxa. The results offer no support for the attribution of Pp. antiquus at Taung to Procercocebus as proposed by Gilbert (2007). Paleoenvironment Comparisons to the Use-Wear Trophic Triangle All of the taxa examined fall into the upper and lower left areas of Semprebon et al.s (2004, fig. 3) use-wear trophic triangle. Many specimens display a relatively high degree of pitting and a range of scratches from very few to moderate amounts (Fig. 8). Some P. ursinus specimens correspond to primate leaf browsers while others exhibit relatively greater numbers of pits. None of the fossils falls within the The initial appearance of C4 grasses during the late Mioceneearly Pliocene and the subsequent expansion of this biome around 1.7 Ma (Hopley et al., 2007) seem to be related to regional climatic variation driven by global climatic changes. Taung and Sterkfontein appear to be roughly contemporaneous, dating to a little before 2 Ma, and likely shared similar although not identical environments. Fine scratches and their relationship to small pits emerge as one of the

FIGURE 5Histogram showing average counts of hypercoarse scratches and puncture pits per taxon with two standard deviations; star 5 P. ursinus; X 5 P. izodi; cross 5 Pp. antiquus; circle 5 Pp. jonesi; triangle 5 Pp. broomi; rectangle 5 indeterminate (indet.); X 5 P. izodi; star 5 P. ursinus; cross 5 Pp. antiquus; triangle 5 Pp. broomi; circle 5 Pp. jonesi.

FIGURE 6Principal components axes 1 and 2 shown with 95% confidence ellipses around group centroids. Star 5 P. ursinus (P. u.); X 5 P. izodi (P. i.); cross 5 Pp. antiquus (Pp. a.); circle 5 Pp. jonesi (Pp. j.); triangle 5 Pp. broomi (Pp. b.); rectangle 5 indeterminate (indet.).

FIGURE 7Canonical scores axes 1 and 2 shown with 95% confidence ellipses around group centroids. Star 5 P. ursinus (P. u.); X 5 P. izodi (P. i.); cross 5 Pp. antiquus (Pp. a.); circle 5 Pp. jonesi (Pp. j.); triangle 5 Pp. broomi (Pp. b.); and rectangle 5 indeterminate (indet.).

446

WILLIAMS AND PATTERSON

PALAIOS

FIGURE 8Total pits and total scratches in sample compared to an approximated use-wear trophic triangle (Semprebon et al., 2004; fig. 3). The three points of the triangle consist of average counts of pits and scratches for (1) primate leaf browsers with relatively low numbers of pits and scratches (including Lepilemur, Avahi, Alouatta, Trachypithecus, Nasalis, Colobus, and Symphalangus), (2) primate graminivores (e.g., Hapalemur and Theropithecus) characterized by high scratch low pit counts, and (3) primate hard-object specialists (including seed predators such as Cacajao, Chiropotes and Pithecia, as well as extractive foragers such as Cebus and Daubentonia), which exhibit relatively large numbers of both pits and scratches. Taxa from Semprebon et al. (2004). Star 5 P. ursinus (P. u.); X 5 P. izodi (P. i.); cross 5 Pp. antiquus (Pp. a.); circle 5 Pp. jonesi (Pp. j.); triangle 5 Pp. broomi (Pp. b.); and rectangle 5 indeterminate (indet.).

distinguishing factors between extant P. ursinus and fossil specimens (Figs. 67). There is a significant amount of overlap between P. ursinus and P. izodi, suggesting that increased dietary flexibility may have played a pivotal role in the success of this genus. Fecal isotope studies of P. ursinus with distinctly different diets in two separate South African savanna habitats demonstrate this flexibility (Codron et al., 2006). Diets in the two environments were shown to consist of ,10%20% and ,30%50% C4 grasses and CAM-photosynthesizing succulents. Based on this isotopic data, it is likely that some portion of the observed scratches on the teeth of P. ursinus resulted from the utilization of grasses. In addition, P. ursinus has been observed to exploit USOs, which are typically found in non-graminaceous C4 plants (Hill and Dunbar, 2003). Papio ursinus, however, generally lacks the large number of heavy use-wear scars found in fossil P. izodi (Fig. 5) and lacks the heavy pitting associated with fossil Parapapio and Papio (Fig. 3). The practice of washing USOs prior to consumption in P. ursinus may significantly reduce the exogenous grit ingested. The fossil specimens display higher degrees of pitting than are typical of primate leaf browsers (Fig. 8). This suggests the possibility that these primates were incorporating foods with more resistant properties such as USOs. Other fossil primates with extremes in pit frequencies, such as Hadropithecus stenognathus, have been postulated as USO exploiters (Catlett et al., 2010; L.R. Godfrey, personal communication, 2010). Plants with USOs typically occur in arid areas but are often found in riverine habitats as well. Dominy et al. (2008) found that the USOs are hard and tough, consistent with higher incidences of pitting. The heavy pitting observed in fossil Parapapio may stem from the consumption of exogenous grit adhering to USOs or the USOs themselves. The two primate taxa from Taung, P. izodi and Pp. antiquus, exhibit some overlap in the hard components of their diets. Although fossil P. izodi

exhibits larger average numbers of hypercoarse scratches and puncture pits (Fig. 5), Pp. antiquus may have generally incorporated greater amounts of gritty foods as evidenced by their higher small and large pit frequencies. It is important to emphasize that microwear can provide information about the physical properties of the foods that were consumed. While lighter microwear features have become associated with softer foods (often graminaceous C4 plants) and heavier features with harder foods (often C3 plants), the distinction between C3 and C4 plants based on microwear is not straightforward. In fact, discrepancies between analyses of dental microwear that indicate hard objects in the diet and associated muscularity (and therefore C3 resources) and stable carbon isotopes, indicating high proportions of C4 resources, have been noted among the australopiths leading to what has been termed the C4 conundrum (Sponheimer and Lee-Thorp, 2003). Isotopic analysis of Parapapio specimens from Sterkfontein Member 4 indicates that #40% of their diet consisted of C4 resources (Codron et al., 2005). USOs, however, occur in both C3 and C4 plants and the exploitation of USOs by mole rats has been shown to produce variable isotopic signatures (Yeakel et al., 2007). The dental microwear signals of the extinct specimens from both Taung and Sterkfontein Member 4 may indicate that gritty foods were available and exploited at these sites. Nevertheless, substantial scratch frequencies in both Pp. broomi and P. izodi (although much less than in P. ursinus) indicate the possibility of grasses contributing to the diet at each of these sites. The two taxa from Sterkfontein Member 4 exhibit some degree of overlap. While both groups appear to have largely exploited harder foods as indicated by their high pit frequencies, Pp. broomi stands apart from Pp. jonesi with respect to scratch and puncture-pit frequency. These microwear features appear to support the conclusions of Fourie et al. (2008), who found no craniometric evidence to support separate Parapapio species at Makapansgat but found that two dietary groups existed thereone with a mixed C3 diet, including significant amounts of rootstocks (loosely correlated with Pp. broomi), and another with a mixed C4 diet (Pp. whitei and Pp. jonesi). The likelihood that USOs were exploited to some degree by the taxa examined in the present study agrees with isotopic evidence of C4 utilization in Parapapio at Sterkfontein Member 4 (Codron et al., 2008). It appears that in some cases, such as in P. izodi, grassland components were more heavily incorporated into the dietary repertoire. The microwear signals of the specimens seem to reflect the regional variability of climate, including the general trend toward increasingly drier conditions associated with the expansion of C4 grasses, which must have been experienced at some level by the Taung child. Parapapio antiquus from Taung Gilbert (2007) suggests that Pp. antiquus from Taung should be reassigned to a new taxon, Procercocebus antiquus, based on the observed similarities in premolar dimensions and craniomandibular morphology between these specimens and living mangabeys of the genus Cercocebus. Gilbert (2007) also points to genetic data to support a taxonomic shift in African papionins by 6.96.1 Ma into two groups, Theropithecus-Lophocebus-Papio and Mandrillus-Cercocebus, with each group sharing a common ancestor 3.43.0 Ma and 4.13.6 Ma, respectively (Tosi et al., 2003). It is also suggested that based on the divergence dates, the Cercocebus lineage was present in Africa by 2.0 1.5 Ma. While this is a reasonable assumption, it does not necessitate the presence of Procercocebus at Taung. To the extent that diet approximates phylogeny, the use-wear patterns observed at Taung can also be considered in the discussion of species attribution. As shown here, the Parapapio specimens display a considerable degree of overlap in their microwear signals, and the genus can generally be distinguished from fossil and particularly from extant Papio. The sample of Pp. antiquus is more similar to Sterkfontein Parapapio than its purported

PALAIOS

PALEOECOLOGICAL RECONSTRUCTION OF TAUNG, SOUTH AFRICA

447

contemporaries at Taung and lacks any discernable evidence for seed predation or extractive foraging (Fig. 8) that might link it to extant Cercocebus. Therefore, the reassignment of Parapapio antiquus to a novel primate genus, Procercocebus, lacks support from analyses of dental microwear. Furthermore, a few indeterminate specimens fall within the range of Pp. antiquus (e.g., T 13, T 125958, and T 56658). Most other indeterminate specimens may be affiliated with P. izodi. CONCLUSIONS The dental microwear evidence presented in this study appears to fit well with previous paleoecological reconstructions. It also agrees with suggestions that dietary flexibility played an important role in the success of papionin and hominin taxa in adapting to the shift to more arid and open environments across southern Africa beginning in the late Miocene (Codron et al. 2008). Based on these results, extinct P. izodi stands apart from Parapapio sp. and resembles extant P. ursinus more than the other fossil taxa. While diet represents only one aspect of the evolution of species, the increased dietary flexibility of Papio undoubtedly contributed to the success of this genus in southern Africa. These findings also help to infer the ecological habitat experienced by the Taung child. It is likely that the open woodland environment of Taung, with significant proportions of C4 resources, may have constrained the dietary resources available to both australopiths and papionins, although the isotopic signatures of australopiths appear to be more variable than both extinct and extant papionins (Codron et al., 2008). The incorporation of C4 resources into the diets of papionins and hominins may be related to this dietary shift. About 2 Ma, however, the habitat at Taung (and Sterkfontein Member 4) probably contained fewer open grassland resources than is typical of much of contemporary central South Africa. ACKNOWLEDGMENTS We thank the curators who generously allowed FLW to make dental molds, including Francis Thackeray at the Transvaal Museum (Pretoria), Mike Raath at the University of the Witwatersrand Medical School (Johannesburg), Denise Hammer at the South African Museum (Cape Town), and Eric Delson at the American Museum of Natural History (New York City). Brian Carter, Gretchen Clymer, Kathryn Hudson, Sean Kiskel, Robin McLauchlin, Michael McPherson, Darby Proctor, Jessica Raines, and Lindsay Slowiczek kindly assisted in preparing the resin casts. We thank Laurie Godfrey, Gina Semprebon, and the Associate Editor for their suggested improvements to the paper. This research was supported by a Research Team Grant, a Research Professional Enhancement Grant, and a Tech Fee Grant from the Vice President for Research, Georgia State University. REFERENCES
AVERY, D.M., 2001, The PlioPleistocene vegetation and climate of Sterkfontein and Swartkrans, South Africa, based on micromammals: Journal of Human Evolution, v. 41, p. 113132. BENEFIT, B.R., and MCCROSSIN, M.L., 1990, Diet, species diversity, and distribution of African fossil baboons: Kroeber Anthropological Society Papers, v. 7172, p. 7993. BERGER, L.R., and CLARKE, R.J., 1995, Eagle involvement in accumulation of the Taung child fauna: Journal of Human Evolution, v. 29, p. 275299. BERGER, L.R., LACRUZ, R., and DE RUITER, D.J., 2002, Revised age estimates of Australopithecus-bearing deposits at Sterkfontein, South Africa: American Journal of Physical Anthropology, v. 119, p. 192197. CATLETT, K.K., SCHWARTZ, G.T., GODFREY, L.R., and JUNGERS, W.L., 2010, Life history space: A multivariate analysis of life history variation in extant and extinct Malagasy lemurs: American Journal of Physical Anthropology, doi: 10.1002/ajpa.21236. CERLING, T.E., 1992, Development of grasslands and savannas in East Africa during the Neogene: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 97, p. 241247. CERLING, T.E., and HAY, R.L., 1988, An isotopic study of paleosol carbonates from Olduvai Gorge: Quaternary Research, v. 25, p. 6378.

CHRISTENSEN, B.A., KALBAS, J.L., MASLIN, M., and MURRAY, R.W., 2002, Paleoclimatic changes in southern Africa during the intensification of Northern Hemisphere Glaciation: Evidence from ODP Leg 175 Site 1085: Marine Geology, v. 180, p. 117131. CIOCHON, R.L., 1993, Evolution of the cercopithecoid forelimb: Phylogenetic and functional implications from morphometric analyses: University of California Publications in Geological Sciences, v. 138, 272 p. CODRON, D., LEE-THORP, J.A., SPONHEIMER, M., DE RUITER, D., and CODRON, J., 2006, Inter- and intrahabitat dietary variability of chacma baboons (Papio ursinus) in South African savannas based on fecal d13C, d15N, and %N: American Journal of Physical Anthropology, v. 129, p. 204214. CODRON, D., LEE-THORP, J.A., SPONHEIMER, M., DE RUITER, D., and CODRON, J., 2008, What insights can baboon feeding ecology provide for early hominin niche differentiation?: International Journal of Primatology, v. 29, p. 757772. CODRON, D., LUYT, J., LEE-THORP, J.A., SPONHEIMER, M., DE RUITER, D., and CODRON, J., 2005, Utilization of savanna-based resources by Plio-Pleistocene baboons: South African Journal of Science, v. 101, p. 245249. COWLING, S.A., COX, P.M., JONES, C.D., MASLIN, M.A., and PEROS, M., 2008, Simulated glacial and interglacial vegetation across Africa: Implications for species phylogenies and trans-African migration of plants and animals: Global Change Biology, v. 14, p. 827840. DAEGLING, D.J., and GRINE, F.E., 1999, Terrestrial foraging and dental microwear in Papio ursinus: Primates, v. 40, p. 559572. DART, R., 1926, Taung and its significance: Natural History, v. 26, p. 315327. DELSON, E., 1984, Cercopithecid biochronology of the African PlioPleistocene: Correlations among eastern and southern hominid-bearing localities: Courier Forschungsinstitut Senckenberg, v. 69, p. 199218. DEMENOCAL, P.B., 1995, PlioPleistocene African climate: Science, v. 270, p. 5359. DEMENOCAL, P.B., 2004, African climate change and faunal evolution during the PliocenePleistocene: Earth and Planetary Science Letters, v. 220, p. 324. DOMINY, N.J., VOGEL, E.R., YEAKEL, J.D., CONSTANTINO, P., and LUCAS, P.W., 2008, Mechanical properties of plant underground storage organs and implications for dietary models of early hominins: Evolutionary Biology, v. 35, p. 159175. ELTON, S., 2007, Environmental correlates of the cercopithecoid radiations: Folia Primatologica, v. 78, p. 344364. EL-ZAATARI, S., GRINE, F.E., TEAFORD, M.F., and SMITH, H.F., 2005, Molar microwear and dietary reconstructions of fossil Cercopithecoidea from the Plio Pleistocene deposits of South Africa: Journal of Human Evolution, v. 49, p. 180 205. FOURIE, N.H., LEE-THORP, J.A., and ACKERMANN, R.R., 2008, Biogeochemical and craniometric investigation of dietary ecology, niche separation, and taxonomy of PlioPleistocene cercopithecoids from the Makapansgat limeworks: American Journal of Physical Anthropology, v. 135, p. 121135. GEAR, J.H.S., 1926, A preliminary account of the baboon remains from Taungs: South African Journal of Science, v. 28, p. 731747. GILBERT, C.C., 2007, Craniomandibular morphology supporting the diphyletic origin of mangabeys and a new genus of the Cercocebus/Mandrillus clade, Procercocebus: Journal of Human Evolution, v. 53, p. 69102. GODFREY, L.R., SEMPREBON, G.M., JUNGERS, W.L., SUTHERLAND, M.R., SIMONS, E.L., and SOLOUNIAS, N., 2004, Dental use-wear in extinct lemurs: Evidence of diet and niche differentiation: Journal of Human Evolution, v. 47, p. 145169. GORDON, K.D., 1984, Hominoid dental microwear: Complications in the use of microwear analysis to detect diet: Journal of Dental Research, v. 63, p. 10341046. GRINE, F.E., UNGAR, P.S., and TEAFORD, M.F., 2002, Error rates in dental microwear quantification using scanning electron microscopy: Scanning, v. 24, p. 144153. HILL, R.A., and DUNBAR, R.I.M., 2003, Climatic determinants of diet and foraging behaviour in baboons: Evolutionary Ecology, v. 16, p. 579593. HOPLEY, P.J., MARSHALL, J.D., WEEDON, G.P., LATHAM, A.G., HERRIES, A.I.R., and KUYKENDALL, K.L., 2007, Orbital forcing and the spread of C4 grasses in the late Neogene: Stable isotope evidence from South African speleothems: Journal of Human Evolution, v. 53, p. 620643. HUGHES, J.K., ELTON, S., and OREGAN, H.J., 2008, Theropithecus and Out of Africa dispersal in the PlioPleistocene: Journal of Human Evolution, v. 54, p. 4377. KAY, R.F., 1985, Dental evidence for the diet of Australopithecus: Annual Reviews in Anthropology, v. 14, p. 315341. KING, T., ANDREWS, P., and BOZ, B., 1999, Effect of taphonomic processes on dental microwear: American Journal of Physical Anthropology, v. 180, p. 359373. LEE-THORP, J.A., SPONHEIMER, M., and LUYT, J., 2007, Tracking changing environments using stable carbon isotopes in fossil tooth enamel: An example from the South African hominin sites: Journal of Human Evolution, v. 53, p. 595 601. MAHONEY, P., 2006, Microwear and morphology: Functional relationships between human dental microwear and the mandible: Journal of Human Evolution, v. 50, p. 452459.

448

WILLIAMS AND PATTERSON

PALAIOS

MAINLAND, I.L., 2003, Dental microwear in grazing and browsing Gotland sheep (Ovis aries) and its implications for dietary reconstruction: Journal of Archaeological Science, v. 30, p. 15131527. MASLIN, M.A., and CHRISTENSEN, B.A., 2007, Tectonics, orbital forcing, global climate change, and human evolution in Africa: Introduction to the African paleoclimate special volume: Journal of Human Evolution, v. 53, p. 443464. MCKEE, J.K., 1991, Paleo-ecology of the Sterkfontein hominids: A review and synthesis: Palaeontologia Africana, v. 28, p. 4151. MCKEE, J.K., 1993, Faunal dating of the Taung hominid fossil deposit: Journal of Human Evolution, v. 25, p. 363376. MCKEE, J.K., 1996, Faunal evidence and Sterkfontein Member 2 foot bones of an early hominid: Science, v. 271, p. 1301. PICKERING, R., HANCOX, P.J., LEE-THORP, J.A., GRUN, R., MORTIMER, G.E., MCCULLOUGH, M., and BERGER, L.R., 2007, Stratigraphy, U-Th chronology, and paleoenvironments at Gladysvale Cave: Insights into the climatic control of South African hominin-bearing cave deposits: Journal of Human Evolution, v. 53, p. 602619. REED, K.E., 1997, Early hominid evolution and ecological change through the African PlioPleistocene: Journal of Human Evolution, v. 32, p. 289322. SANDERS, W.J., TRAPANI, J., and MITANI, J.C., 2003, Taphonomic aspects of crowned hawk-eagle predation on monkeys: Journal of Human Evolution, v. 44, p. 87105. SCOTT, L., 1987, Pollen analysis of hyena coprolites and sediments from Equus Cave, Taung, southern Kalahari (South Africa): Quaternary Research, v. 28, p. 144156. SCOTT, R.S., UNGAR, P.S., BERGSTROM, T.S., BROWN, C.A., CHILDS, B.E., TEAFORD, M.F., and WALKER, A., 2006, Dental microwear texture analysis: Technical considerations: Journal of Human Evolution, v. 51, p. 339349. SCOTT, R.S., UNGAR, P.S., BERGSTROM, T.S., BROWN, C.A., GRINE, F.E., TEAFORD, M.F., and WALKER, A., 2005, Dental microwear texture analysis shows withinspecies diet variability in fossil hominins: Nature, v. 436, p. 693695. GALEN, L., LEE-THORP, J.A., and CERLING, T., 2007, Timing of C4 grass expansion SE across sub-Saharan Africa: Journal of Human Evolution, v. 53, p. 549559. SEMPREBON, G.M., GODFREY, L.R., SOLOUNIAS, M.R., and JUNGERS, W., 2004, Can low-magnification stereomicroscopy reveal diet?: Journal of Human Evolution, v. 47, p. 115144. SPONHEIMER, M., and LEE-THORP, J.A., 1999, Isotopic evidence for the diet of an early hominid, Australopithecus africanus: Science, v. 283, p. 368370. SPONHEIMER, M., and LEE-THORP, J.A., 2003, Differential resource utilization by extant great apes and australopithecines: Towards solving the C4 conundrum: Comparative Biochemistry and Physiology, v. 136, p. 2734.

TEAFORD, M.F., MAAS, M.C., and SIMONS, E.L., 1996, Dental microwear and microstructure in early Oligocene primates from the Fayum, Egypt: Implications for diet: American Journal of Physical Anthropology, v. 101, p. 527543. TOBIAS, P.V., VOGEL, J.C., OSCHADLEUS, H.D., PARTRIDGE, T.C., and MCKEE, J.K., 1993, New isotopic and sedimentological measurements of the Thabaseek deposits (South Africa) and the dating of the Taung hominid: Quaternary Research, v. 40, p. 360367. TOSI, A.J., DISOTELL, T.R., MORALES, J.C., and MELNICK, D.J., 2003, Cercopithecine Y-chromosome data provide a test of competing morphological evolutionary hypotheses: Molecular Phylogenetics and Evolution, v. 27, p. 510521. UNGAR, P.S., 2002, Reconstructing the diets of fossil primates, in Plavcan, J.M., Kay, R.F., Jungars, W.L., and van Schaik, C.P., eds., Reconstructing Behavior in the Primate Fossil Record: Kluwer Academic/Plenum Publishers, New York, p. 261 296. VAN DER MERWE, N.J., THACKERAY, J.F., LEE-THORP, J.A., and LUYT, J., 2003, The carbon isotope ecology and diet of Australopithecus africanus at Sterkfontein, South Africa: Journal of Human Evolution, v. 44, p. 581597. VRBA, E.S., 1975, Some evidence of chronology and palaeocology of Sterkfontein, Swartkrans, and Kromdraai from the fossil Bovidae: Nature, v. 254, p. 301304. VRBA, E.S., 1985, Environment and evolution: Alternative causes of the temporal distribution of evolutionary events: South African Journal of Science, v. 81, p. 229 236. WALKER, J., CLIFF, R.A., and LATHAM, A.G., 2006, U-Pb isotopic age of the StW 573 hominid from Sterkfontein, South Africa: Science, v. 314, p. 15921594. WALKER, A., HOECK, H.N., and PEREZ, L.M., 1978, Microwear of mammalian teeth as an indicator of diet: Science, v. 201, p. 908910. WILLIAMS, F.L., ACKERMANN, R.R., and LEIGH, S.R., 2007, Inferring PlioPleistocene southern African biochronology from facial affinities in Parapapio and other fossil papionins: American Journal of Physical Anthropology, v. 132, p. 163174. WYNN, J.G., 2000, Paleosols, stable carbon isotopes, and paleoenvironmental interpretation of Kanapoi, Northern Kenya: Journal of Human Evolution, v. 39, p. 411432. YEAKEL, J.D., BENNETT, N.C., KOCH, P.L., and DOMINY, N.J., 2007, The isotopic ecology of African mole rats informs hypotheses on the evolution of human diet: Proceedings of the Royal Society, Series B, Biological Sciences, v. 274, p. 1723 1730.

ACCEPTED APRIL 2, 2010