Kroeber Anthropological Society Papers, Nos. 71-72, 1990 Diet, Species Diversity and Distribution of African Fossil Baboons Brenda R. Benefit and Monte L. McCrossin Based on measurements of molar features shown to be functionally correlated with the proportions of fruits and leaves in the diets of extant monkeys, Plio-Pleistocene papionin baboons from southern Africa are shown to have included more herbaceous resources in their diets and to have exploited more open country habitats than did the highly frugivorous forest dwelling eastern African species. The diets of all species offossil Theropithecus are reconstructed to have included more fruits than the diets of extant Theropithecus gelada. Theropithecus brumpti, T. quadratirostris and T. darti have greater capacitiesfor shearing, thinner enamel and less emphases on the transverse component of mastication than T. oswaldi, and are therefore interpreted to have consumed leaves rather than grass. Since these species are more ancient than the grass-eating, more open country dwelling T. oswaldi, the origin of the genus Thero- pithecus is attributed tofolivorous adaptations by large papionins inforest environments rather than to savannah adapted grass-eaters. Reconstructions of diet and habitat are used to explain differences in the relative abundance and diversity offossil baboons in eastern and southern Africa. INTRODUCTION Interpretations of the dietary habits of fossil Old World monkeys have been based largely on analogies to extant mammals with lophodont teeth (Jolly 1970; Napier 1970; Delson 1975; Andrews 1981; Andrews and Aiello 1984; Temerin and Cant 1983). With the exception of Jolly's (1972) study of Theropithecus oswaldi, such inter- pretations have focused on the origins of Cercopithecoidea with relatively litde attention given to more recent dietary diversification widtin the superfamily. In this study the dietary habits of Plio- Pleistocene cercopithecine monkeys from fossil sites in eastern and southern Africa are recon- structed on the basis of dental features shown to be functionally correlated to diet among extant cercopithecoids (Kay 1977, 1978, 1981, 1984; Kay and Covert 1984; Kay and Hylander 1978; Benefit 1987). A major concern of the study is to better understand the differences in the patterns of species diversity and the relative abundance of cercopithecine monkeys which existed in the eas- tern and southern regions of the African continent during the Plio-Pleistocene. In cave deposits of southern Africa Thero- pithecus is rare, comprising 7% of the total cercopithecoid fauna collected prior to 1976 (Freedman 1976). In contrast, 84% of the mon- keys collected at the Omo and 85% of those collected at Koobi Fora, both in eastern Africa, prior to the same date belong to the genus Thero- pithecus (Eck 1976; Leakey 1976). In addition, Theropithecus is more diverse in deposits from eastem Africa, with five species occurring in the fossil deposits, as opposed to only two in south- ern Africa. The opposite pattem of diversity and abundance between eastern and southern Africa is observed for members of the Papionina (Papio, Cercocebus, Parapapio, Gorgopithecus, and Dinopithecus). [We follow Szalay and Delson (1979) in recognizing two tribes of cercopithe- cines, Cercopithecini and Papionini, and three subtribes of the Papionini: Theropithecina (gela- das, fossil and modern), Macacina (macaques, fossil and modern) and Papionina (baboons, drills, mandrills and mangabeys, fossil and mod- ern)]. Eighty-four percent of the fossil monkeys in southern Africa are papioninans (Freedman 1976), while only 10% of those in eastern Africa are members of this subtribe. In southern Africa at least nine species of papioninan monkeys have been recovered, but only three species are known to have occurred in eastern Africa (although the actual number is presendy indeterminable due to the fragmentary and incomplete nature of the ma- terial). Reconstructions of the dietary habits of these extinct animals, in combination with infor- mation about the habitats in which they lived and studies of food consumption and habitat use by living mammals, are used to describe a new sce- nario about the evolution, diversity, distribution and relative abundance of fossil Theropithecus and other baboons during the Plio-Pleistocene. MATERIALS AND METHODS Fossil cercopithecine monkeys were sampled from Plio-Pleistocene deposits in southern Africa (Sterkfontein Member 4, Kromdraai Members A and B, Taung, and Swardcrans Member 1) and eastern Africa (Laetoli and Olduvai, Tanzania; Olorgesailie and Koobi Fora, Kenya -- Areas 1- 203; Omo, Ethiopia -- Usno Formation, Shun- 78 gura Formation Members B- H, and Kalam Area). Spe- cies sampled are listed by deposit in Table 1. [A complete list of specimens sampled is given in Benefit (1987).] Taxonomic identi- fications for the southern African fossils are largely based on Freedman (1957, 1961a, 1961b, 1965, 1976), Freedman and Brain (1972), Freedman and Stenhouse (1972), Maier (1971a, 1971b) and Eisen- hart (1974). Identification of fossil monkeys from eastern Africa are based on descriptions by R.E. Leakey (1969), Jolly (1970, 1972), M.G. Leakey (1976, 1982), Leakey and Leakey (1973a, 1973b, 1976), Eck (1977), Eck and Howell (1982), Eck and Jablonski (1984) and Leakey and Delson (1987). Biostratigraphic dating of the southern African cave deposits indicates that Ma- kapansgat is the oldest site at approximately 3.0 million years (my), followed by Sterkfontein (3.2-2.5 my and 1.75-1.4 my), Krom- draai (2.7-1.8 my), Taung (2.3-1.0 my) and Swart- krans (1.8-1.6 my and 1.25-0.9 my) (Vrba 1982). Eastern African deposits are more securely dated on ra- diometric grounds. Laetoli is considered to be 3.8-3.5 my (Leakey et al. 1976; Leakey and Hay 1982), the Usno and Shungura Forma- tions at the Omo range in age from 2.9-1.3 my (Shuey et al. 1974; Brown and Nash 1976; Brown et al. 1985), Koobi Fora and Ileret Formations range in age from 3.3-1.5 my (Feibel et al. 1989), Olduvai from 2.2-0.6 my (Leakey and Hay 1982) and Olorgesailie from 1.3-0.5 my (Potts 1989). Because of the long Table 1. Numbers of fossil cercopithecine molars measured. L = Lower, U = Upper. EAST AFRICA Cercocebus sp., KOOBI FORA Cercocebus ado, OLDUVAI Parapapio ado, LAETOLI Parapapio jonesi, KANAPOI Parapapio sp., KOOBI FORA Papionini indet, BARINGO Papio sp., OLDUVAI Theropithecus oswaldi, KOOBI FORA & ILERET (combined) AREA 1 (1.5-1.6 my) AREA 8 (1.5-1.6 my) AREA 103 (1.6-1.7 my) AREA 10 (1.7-1.9 my) AREA 123 (1.7-1.9 my) AREA 130 (1.8-1.9 my) AREA 104 (1.7-2.0 my) AREA 106 (2.0 my) AREA 116 (2.0 my) OMO SHUNGURA FORMATION Member G Member H Kalam area (Members J-L) OLDUVAI OLORGESAILIE Theropithecus brumpti KOOBI FORA/TULU BOR AREA 117 &204 (3.3 my) AREA 203 (3.3 my) OMO, SHUNGURA FORMATION Member C Member D Theropithecus quadratirostris OMO USNO FORMATION Member 11 LM1 5 1 5 1 2 13 61 3 8 5 3 3 3 8 2 4 1 1 2 4 1 1 2 11 39 35 LM2 LM3 1 1 8 10 1 1 3 6 1 2 3 40 1 5 5 3 12 23 UMI UM2 UM3 T7 1 T 5 4 8 5 2 2 2 1 2 17 1 1 2 1 1 8 1 4 12 44 1 2 8 1 1 1 1 1 4 1 1 1 8 25 2 3 4 35 4 2 6 1 2 6 6 1 2 1 SOUTHERN AFRICA Dinopithecus ingens, SWARTKRANS Gorgopithecus major, KROMDRAAI Papio angusticeps, KROMDRAAI TAUNG Papio robinsoni, SWARTKRANS Parapapio ionesi, SWARTKRANS STERKFONTEIN Parapapio whitei, STERKFONTEIN Parapapio broomi, STERKFONTEIN Theropithecus darti, SWART[RANS Ml LM2 LM3 3 5 7 1 3 4 2 2 2 3 1 6 10 10 2 1 2 3 7 3 2 4 4 3 4 3 2 4 UMi 4 3 1 5 1 1 2 2 2 UM2 7 4 2 2 15 2 4 2 5 3 UM3 5 2 2 10 2 5 2 3 79 time ranges represented at most samples of monkeys were examii their stratigraphic unit of provei possible. Dietary estimates for the f based on the relationship betw phology and diet in nine extar have been studied extensively in which the proportions of leaves annual diet are known (Table considered were Colobus badit bus guereza (n=22), Colobus Presbytis melalophos (Kay an Presbytis obscura (Kay and Cc cocebus galeritus (n=20), Cerc (n=25), Macaca nemestrina (n fascicularis (n=34). Additional were taken from Kay (1978, Covert 1984) and Benefit (198' nology and measurements ust follow Jolly (1972), Delson (1' 1981) and Benefit (1987). Folivorous monkeys use puncture and shear leaves wi monkeys crush and grind hard fruits and seeds (Walker and Mi livores emphasize the shearing molar occlusion as the mandibl and medially, while frugivoro Table 2. Average percentages of in the annual diets of the extan species sampled. Colobus badius Colobus guereza Colobus satanas Presbytis melalophos Presbytis obscura Cercocebus galeritus Cercocebus albigena Macaca nemestrina Macaca fascicularis Data taken from Struhsaker (1975, 1! Clutton-Brock (1975), Gatinot (1975), tier-Hion (1978,1980), McKey (1978), Waser (1977, 1984), Quris (1975) Caldecott (1986), Aldrich-Blake (198 MacKinnon (1978), Mah (1980), Whe makers and Chivers (1980). t of the deposits, grinding takes place during the second phase of ined according to occlusion as the jaw moves lingually and mesi- nience whenever ally, with the protoconid and hypoconid coming into direct contact with the protocone and hypo- ossil species are cone as a result (Crompton and Hiiemae 1970; reen molar mor- Kay and Hiiemae 1974; Kay 1975, 1978). The it species which tendency for monkeys that include large amounts the wild and for of leaves in their diets to have longer shear crests and fruits in the than monkeys which eat greater amounts of fruits 2). The species and seeds has been demonstrated by Kay (1975, us (n=33), Colo- 1978, 1981; Kay and Covert 1984). satanas (n=2), In this study eight shear crest lengths for Id Covert 1984), each upper and lower molar were measured using vert 1984), Cer- an ocular micrometer mounted in a stereoscopic ocebus albigena microscope (Figure IA). Shearing crest develop- ,=6) and Macaca ment was then appraised in four ways: 1) from comparative data calculation of Kay's (1984) shear quotient (SQ) 1981; Kay and for lower second molars; 2) from the sum of 7). Dental termi- shear crests gauged against molar length (abbre- ed in this paper viated PERS); 3) from the sum of lingual (for )73), Kay (1978, lower molars) or buccal (for upper molars) shear crests relative to molar length (PERLS); and 4) their molars to from the sum of shear crests bordering the lingual hile frugivorous and buccal notches relative to molar length ler, more fibrous (PERMS). The shear quotient is calculated using urray 1975). Fo- the formula SQ = 100(So-Se/Se), with So (ob- or first phase of served shear) equal to the sum of the lengths of le moves upward the eight shear crests, and Se (expected shear) ius crushing and equal to 2.79 (lower second molar length) to the 0.982 exponent. Because the shear quotient ex- aggerates variation in the observed sum of shear, fruits and leaves the simple indices (PERS, PERLS, PERMS) it cercopithecoid were also used. Unworn molars were measured for calculations of SQ, PERS and PERMS. FRUIT LEAVES Since the lingual cusps of lower molars and the 7.6 74.7 buccal cusps of upper molars wear less rapidly 13.6 68.3 than cusps on the opposite side of the tooth, PERLS has the advantage of being measurable on 58.0 37.0 both unworn and moderately worn molars. PERMS is especially useful for the measurement 58.0 39.0 of lower molars with hypoconulids or accessory cuspules on the posthypocristid, such as the third 44.0 36.0 molars of all species and the first and second molars of Theropithecus. 77.0o 13.0 oFor each of the extant species listed above, the average shear quotient was plotted against the 71.0 2.6 average proportion of fruits and leaves consumed 74.2 13.0 annually (Figure 2). Shear quotient was found to *2 13 be significantly correlated to diet (Table 3). The 62.5 20.0 same procedure was used for the shear indices of each upper and lower molar (Benefit 1987). For 978), Marsh (1981), those indices found to be significantly correlated ,Oates (1977), Gau- to diet, regression equations best describing the ,Homewood (1978), relationship (Table 4) were used to estimate the ,Freeland (1979), proportions of fruits and leaves eaten by the ex- 0), MacKinnon and tinct monkeys. ady (1980) and Rae- Greater shear crest length on the molars of folivorous colobine monkeys is accomplished by 80 Figure 1. Measurements of molar features functionally correlated to diet among extant cercopithecoids. A. Lengths of shear crests, numbers 1-8; END = entoconid, MED = metaconid, HYD = hypoconid, PRD = protoconid. B. Lingual notch height; NH = vertical height from the notch to the base of the lin- gual notch, NR = vertical height from the base of the notch to the cervix below. C. Proximity between mesial pairs of cusps; MW = mesial width, MCP = distance between mesial cusp tips. D. Enamel thick- ness. a decrease in the heights of the central basin, me- sial fovea and distal fovea above the cervix, rather than through an increase in absolute cusp height (Benefit 1987). As a result the lingual notch on lower molars (buccal notch on upper molars) is significantly taller and the height of the crown from the base of the notch to the cervix is significantly lower on colobine than on frugivo- rous cercopithecine molars (Delson 1973; Benefit and Pickford 1986; Benefit 1987) (Figure 1B). As for the shear indices, index NHNR (height of the crown above the notch/height of the crown below the notch x 100) was plotted against the proportions of fruits and leaves eaten annually by the extant species and found to be significantly correlated to diet for the lower second and third molars as well as for the upper second molar (Ta- ble 3). Regression equations based on NHNR (Table 4) are used in additon to the shear indices to reconstruct the diets of the fossils papionins. A NCP NCP CENCOPITHECINAA COLOBINAE c ENAMEL ENAMEL ..A T NlI C K N I 9 8 Nzl R It a a NHY D P AR D E X PO 0 e D THICKNESS THICNES D 81 Figure 2. Bivariate plots of shear quotient values against percentages of food items in the annual diets of extant monkeys. 24. 7+ 22, 54. 20, i, 12. 10. a .58 S S 4 so 0 32 -2 44. -4 2. -S -S 10 20 20 40 LEAVES 50 s0 T0 so S14 St 9 0 I' 3+ 4t 24. 14. 10 20 20 40 50 *0 70 s0 FRUITS 1 = Cercocebus albigena; 2 = Cercocebus galeritus; 3 = Macaca nemestrina; 4 = Macaca fascicularis; 5 = Colobus satanas; 6 = Colobus guereza; 7 = Colobus badius; 8 = Presbytis melalophos; 9 = Presbytis obscura. S Kay 1984; + Benefit 1987 The emphasis colobine and cercopithecine monkeys place on shearing and grinding is appar- ent not only from the lengths of the shear crests and notch heights but also from the manner and rate at which the teeth wear. Although the enamel is thick, the molars of cercopithecine monkeys are adapted to wear flat rapidly, while the cusps of thin enamelled colobine monkeys maintain their height and integrity even when large areas of dentine are exposed (Figure 3). The rapid rate at which cercopithecine teeth wear is probably rela- ted to the significantly closer proximity of their cusp tips and loph(id)s than is observed for colobines (Benefit 1987), and the resulting con- striction of the central basin. The space in which occluding cusps and basins can interdigitate is ex- tremely limited on cercopithecine molars. Con- sequently, the chance that opposing cusps will rub against each other is enhanced, as is the rate at which the crown wears. The cusp tips of colo- bine molars are set much further apart than those of cercopithecines (Benefit 1987). The wide dis- tance between the cusps and the subsequently large size of the central basin allows the occlu- ding cusps to interdigitate more freely, leading to a decrease in the rate at which the molars wear. The combination of low cusp relief and close cusp proximity causes cercopithecine molars to wear flat rapidly. The shearing capacity of the molar is lost as wide enamel rims, created by the merging together of worn grinding facets, form around circular and concave patches of dentine on 14 12 0 -2. -4. -S, -S. I. -10 X 7 Table 3. Pearson correlation coefficients for den- tal indices significantly correlated to the average proportions of fruits and leaves included in the annual diets of nine extant monkey species. L = Lower, U = Upper. INDEX LM2 SQ LM2 PERS LM2 PERLS LM2 PERMS UM2 PERS UM2 PERLS UM2 PERMS LM2 FL LM3 FL LM2 NHNR LM3 NHNR UM2 NHNR FRUIT LEAVES -0.9450 0.9710 -0.9642 0.9766 -0.9738 0.9855 -0.9599 0.9773 -0.9882 0.9806 -0.9815 0.9743 -0.9744 0.9751 -0.9302 0.9149 -0.9087 -0.9648 0.9682 -0.9828 0.9695 -0.9506 0.9593 the flat cusp tips (Figure 3). Because cercopithe- cine crowns are flared, with greater width at the cervix than between mesial and distal pairs of cusps, the perimeter of the enamel rim and the surface area that can be devoted to grinding in- creases as the crown wears. On colobine molars, which experience little change in cusp relief as they wear, elongated areas of dentine surrounded by thin enamel rims occur along the crests of lophs, providing little surface area for crushing and grinding. For the extant species sampled, significant correlations were found between proportions of fruits and leaves consumed and the degree to which the lower second and third molars and upper second molars were flared (mesial crown width at the apex of cusps/mesial width x 100) (Figure 1C; Table 3). As for cusp relief and shear indices, regression equations based on these measurements were used to estimate the diets of fossil species. In addition to the dental indices mentioned above, enamel thickness was also measured on the extant and fossil teeth (Figure ID). Because there was little variation in enamel thickness within each subfamily for the extant sample, cor- relations between a simple enamel index relating enamel thickness to molar length and diet was not found to be significant, and no regression equa- tion was computed. However, the relationship between enamel thickness and diet was demon- strated by Kay (1981; Kay and Covert 1984) on the basis of logarithm-transformed measure- ments. Enamel thickness is therefore referred to in this paper as a general indicator of frugivory and folivory. One of the inherent problems in reconstruc- ting the diets of extinct monkeys based on dental morphology is that the dental indices alone do not differentiate grass-eating from leaf-eating. The molars of extant grass-eating Theropithecus gel- ada are unique among cercopithecids in that they combine characteristics of both folivorous col- obine and frugivorous cercopithecine molars. From the lingual perspective, the lower molars of T. gelada resemble colobine monkeys with high occlusal relief, low crown height below the lin- gual notch and long shear crests. From buccal and occlusal perspectives, Theropithecus lower molars resemble cercopithecines with close cusp proximity, small central basins, short shear crests and a high position of the lowest point of the lo- phids above the cervix. As might be expected from consideration of the cercopithecine charac- teristics of the molars, the cusp tips wear flat rapidly with flat enamel rims quickly forming on the molars. As noted by Jolly (1972), the pattern of enamel ridges is more elaborate on Theropithe- cus than other monkeys due to the presence of additional ridges, clefts and infoldings of enamel along the mesial and distal shelves. Grasses are milled between the teeth as occluding molars scrape transversely against each other in a manner similar to that of grazing ungulates. The enamel rims differ from those of other Cercopithecinae in being positioned well above the occlusal basin. The height of the wear surface is thought to main- tain the lifetime of the crown as the teeth are subjected to rapid wear resulting from abrasion by the grasses and soil particles that adhere to plants (Jolly 1972). From consideration of Theropithecus molars it is surmised that either the consumption of abra- sive silicaceous grasses and/or the cercopithecine characteristic of close cusp prximity are respon- sible for the rapid rate of cusp deformation. We conclude that if a species has long shear crests, colobine-like cusp proximity and thin enamel, but a rapid rate of cusp deformation, it may have eaten grass rather than leaves. Patterns of wear are therefore considered together with dental in- dices in reconstructing the diets of fossil species. However, differentiating grass-eating from leaf- 82 83 eating molars, given a combination of colobine and cercopithecine characteristics (as in Thero- pithecus), presents an extremely difficult and perhaps unresolvable situation. RESULTS Papionina Dietary estimates for fossil Cercopithecinae are presented in Table 5. In general, southern African papioninans were found to be less com- mitted to frugivory than their eastern African counterparts, the average diet of the former con- sisting of 56% fruits and 27% leaves (range: 50- 66.5% fruits, 16-33% leaves) and the latter 61% fruits and 22.5% leaves (range: 52-83% fruits, 0-31.5% leaves). Among the southern African species only the diet of Dinopithecus ingens (66.5% fruits) was found to be highly frugivo- rous. Otherwise, three of the southern species were reconstructed as having consumed 58-60% fruits and five as 50-55% fruits. In eastern Africa, fossil Parapapiojonesi (83% fruits), Cer- cocebus (74% fruits) and Papio (67% fruits) were found to be the most committed to frugivory (74% fruits, on average), while Parapapio ado and Parapapio spp. were found to be the least frugivorous (52-53% fruits). Theropithecus The dietary habits of fossil Theropithecus are more difficult to assess than those of other ba- boons due to their unusual molar morphology which includes the presence of accessory cus- pules along shear crests bordering mesial and distal shelves. The accessory cuspules make it Table 4. Linear regression equations used to reconstruct the proportions of fruits and leaves consumed by fossil species. L = Lower, U = Upper. St d, Err. Fruit=62.53 - (2.37139 * LM2SQ) Leaves=24.67 + (2.48568 * LM2SQ) Fruit=57.77 - (-0.97320 * LM2SQW) Leaves=25.25 + (0.97424 * LM2SQW) Fruit=236.59 - (3.0471 * LM2FL) Leaves=-145.70 + (2.94397 * LM2FL) Fruit=95.24 - (0.55408 * LM2NHNR) Leaves=-145.70 + (0.5472 * LM2NHNR) Fruit=301.29 - (0.8998 * LM2PERS) Leaves=-217.57 + (0.89692 * LM2PERS) Fruit=203.04 - (1.20434 * LM2PERMS) Leaves=-159.24 + (1.50747 * LM2PERMS) Fruit=102.79 - (0.65061 * LM3NHNR) Leaves=-18.37 + (0.63173 * LM3NHNR) Fruit=119.11 - (1.22808 * UM2NHNR) Leaves=-35.72 + (1.21971 * UM2NHNR) Fruit=318.61 - (0.97324 * UM2PERS) Leaves=-229.35 + (0.95017 * UM2PERS) Fruit=337.32 - (2.0863 * UM2PERLS) Leaves=-247.89 + (2.0353 * UM2PERLS) Fruit=265.76 - (1.77564 * UM2PERMS) Leaves=-179.61 + (1.7489 * UM2PERMS) R-squared 0.89304 0.94290 Estimate 8.887 6.625 0.94712 8.1536 0.94914 7.8701 0.86529 11.6888 0.86370 12.6289 0.93086 9.3229 0.93739 8.7323 0.92977 9.3967 0.95370 7.5092 0.92131 9.9461 0.95507 7.3971 0.92580 6.5573 0.940 8.5481 0.90372 11.0018 0.92026 9.8542 0.97732 5.3396 0.96167 6.8321 0.96337 6.7857 0.94983 7.8163 0.94946 7.9708 0.95089 7.7333 84 Figure 3. Comparison of worn colobine and cercopithecine molars. LOWER MOLARS M t ~~~~D DX - CERCOPITHECINAE COLOBINAE a D difficult to measure mesial and distal shear crests accurately. As a result the shear quotient, which incorporates the lengths of all crests, is highly va- riable. The length of shear crests bordering the central basin provide the least variable and therefore most reliable measure of shear for Theropithecus. Estimated proportions of fruits and leaves consumed by fossil Theropithecus based on SQ, PERMS and the average of predictions of all indi- ces are summarized in Table 6. Predictions based on index PERMS for lower second molars were found to more accurately reflect the diet of extant T. gelada than those based on shear quotients (Table 6). It is therefore reasonable to assume that predictions based on PERMS for fossil spe- cies are also more accurate, and greater weight is placed on these results. All dietary predictions indicate that the south- ern African T. darti and the more ancient of the eastern African Theropithecus, T. brwnpti (3.2- 2.0 my) and T. quadratirostris (3.4-3.2 my), have had longer shear crests and a higher poten- tial for folivory than the more recent eastern African species T. oswaldi (2.5-0.5 my). These results are partly corroborated by the presence of thicker enamel on the molars of T. oswaldi from the Omo [enamel thickness (1.5 mm)/crown length (17.3 mm) = 8.7%] than on molars of T. brumpti from the same site [enamel thickness (1.16 mm)/crown length (15.6 mm) = 7.4%], indicating a greater potential for frugivory for the former. According to index PERMS all fossil Theropithecus included more fruits in their diets than do extant geladas, which are rarely frugivo- rous in the wild (Dunbar 1983). It is impossible to know whether fossil Theropithecus consumed leaves or grasses. Examination of wear striations under a light mi- croscope revealed deep bucco-lingually oriented parallel striations and the relative absence of pits on the molars of T. oswaldi from Koobi Fora (Benefit 1987). These deep striations were pro- bably caused by the inclusion of grit in the diet of T. oswaldi. They are also indicative of a heavy reliance on the transverse component of mastica- tion, such as is associated with grass-eating in modern T. gelada. It is plausible that the non- fruit component of the diet of T. oswaldi consis- ted of the blades, seeds and rhizomes of grasses as suggested by Jolly (1972), rather than leaves. However, the diet of T. oswaldi is reconstructed here as having been more eclectic than that of T. gelada, counter to Jolly's (1972) suggestion that the species predominantly ate grass. The molars of T. brunpti are generally more gracile than those of T. oswaldi with fewer acces- sory cuspules and infoldings of enamel. Deep transverse striations are not apparent on the worn molars of T. brwnpti, indicating that little grit ad- hered to its food and that possibly the transverse component of mastication was not emphasized by the species. Theropithecus brumpti may have been a true papionin folivore rather than a grass- eater. A similar diet is suggested for T. quadra- tirostris and T. darti. The rapid rate of cusp deformation observed for the molars of T. brumpti can be attributed to the close proximity of the molar cusps and the consumption of fruits, rather than to a diet of abrasive grass. 85 Table 5. Estimated proportions of fruits and leaves con- Ther sumed by fossil cercopithecines. (For Omo Theropithecus R0 and Theropithecus gelada, s = shear crests measured in a straight line without inclusion of accessory crests and cuspules and a = additive shear crest measurements with inclusion of all features.) EAST AFRICA FRUIT LEAVES Cercocebus sp. [OOBI FORA LM2SQ 79.6 6.8 LM2SQW 66.6 16.4 LM2NHNR 65.9 17.2 LM2PERS 84.2 0 LM2PERMS 72.5 4.2 LM2FL 92.4 0 LM3NHNR 67.9 18.4 UM2NHNR 52.3 30.6 UM2PERS 74.9 8.5 UM2PERLS 70.7 12.2 UM2PERMS 71.3 11.95 mean 74 11.5 Cercocebus ado OLDUVAI LM3NHNR 74.6 8.9 Parapapio ado LAETOLIL LM2SQ 60.0 27.3 LM2SQW 56.1 26.9 LM2NHNR 50.8 32.1 LM2PERS 58.9 24.0 LM2PERMS 41.4 43.1 LM2FL 0 83.5 UM2NHNR 82.0 1.1 UM2PERMS 68.9 14.2 mean 52 31.5 Parapapio jonesi KANAPOI LM2SQW 83.1 0 Paraoapio sp. OOBI -?ORA LM2SQW 58.4 24.6 LM3NHNR 69.7 12.7 UM2NHNR 32.3 50.0 mean 53 29 Papionini indet. BAR-NGO LM3NHXR 63.3 19.9 Papio sp. OLDUVAI LM2FL 39.0 45.2 LM2NHNR 61.9 21.2 LM3NHNR 60.0 23.1 UM2NHNR 21.0 61.7 UM2PERS 99.5 0 UM2PERLS 96.3 0 UM2PER!S 90.5 0 mean 67 13 opithecus oswaldi OBI FORA & ILERET LM2FL 61.7 23.2 LM2PERMS 49.8 32.5 UM2PERS 72.7 10.7 UM2PERLS 81.6 1.5 UM2PERMS 58.6 24.4 mean 53 42 AREA 1 (1.5-1.6 my) LM2SQW 51.4 31.9 LM2NHNR 15.4 67.3 UM2PERS 100.0 0 UM2PERLS 88.2 0 UM2PERMS 81.6 1.8 mean 67 20 AREA 8 (1.5-1.6 my) LM2SQW 49.8 33.2 LM2FL 48.2 36.3 LM2NHNR 54.3 28.7 mean 51 33 AREA 103 (1.6-1.7 my) LM2FL 100.0 0 LM2NHNR 43.0 39.8 LM2PERMS 41.25 43.2 mean 61 28 AREA 10 (1.7-1.9 my) UM2NHNR 9.0 73.6 UM2PERMS 33.9 48.7 mean 21.5 61 AREA 123 (1.7-1.9 my) LM3NHNR 26.3 55.9 AREA 130 (1.8-1.9 my) LM2SQ 63.1 24.0 LM2SQW 58.6 24.4 LM2NHNR 45.6 37.3 LM2PERS 65.0 18.0 LM2FL 80.4 5.2 LM3NHNR 34.3 48.1 UM2NHNR 34.9 47.9 UM2PERS 63.0 15.1 UM2PERLS 71.0 11.9 UM2PERMS 64.8 18.3 mean 58 25 AREA 104 (1.7-2.0 my) LM2SQ 36.1 52.4 LM2SQW 46.9 36.1 LM2NHNR 17.1 65.3 LM2PERS 36.1 46.7 LM2PERMS 40.1 44.7 LM2FL 30.5 53.5 mean 27.5 50 AREA 106 (2.0 my) LM2NHNR 47.9 35.0 Theropithecus brumpti KOOBI FORA/TULU BOR AREA 117 (3.2-3.3 my) LM2SQ 46.1 41.7 LM2SQW 51.0 32.0 LM2NHNR 26.6 56.0 LM2PERS 45.4 37.5 LM2PERMS 26.3 62.0 LM3NHNR 28.3 54.0 UM2PERS 70.3 13.0 UM2PERLS 72.2 10.7 UM2PERMS 65.0 18.1 mean 48 36 AREA 203 (3.3 my) LM3NHNR 26.35 55.9 OMO SHUNGURA FORMATION Member unknown Members C-G Member C Member D LM2SQW a 61.4 21.6 LM2FL a 13.2 70.1 mean 37 46 SQ 28.8 60.1 LM2SQ a 55.4 32.1 a 11.4 78.1 LM2PERS s 57.0 25.9 a 17.0 66.6 LM2PERMS 36.8 50.1 mean a 50 36 mean a 22 65 LM2SQ a 0.6 89.5 LM2PERS a 0.8 82.0 LM2PERMS 28.4 60.4 mean 10 77 AREA 116 (2.0 my) OLDUVAI OLORGESAILIE UM2PERS 82.9 0.8 UM2PERLS 97.8 0 UM2PERMS 38.8 43.2 mean 73 15 LM2FL 100.0 0 LM2PERS 72.9 10.1 LM2PERMS 54.0 27.4 mean 76 12 LM2SQW 49.2 33.8 LM2FL 92.8 0 LM2NHNR 22.7 59.8 LM2PERS 55.4 27.5 LM2PERMS 47.3 35.7 LM3NHNR 0 99.6 UM2PERS 87.4 0 UM2PERLS 100.0 0 UM2PERMS 64.9 18.2 mean 53 30 Theropithecus quadratirostris OMO LM2PERS a 0 83.8 LM2PERMS 28.4 60.4 mean 14 72 Theropithecus aelada EXTANT LM2SQ m 60.2 27.1 a 30.7 58.1 LM2PERS m 43.0 39.8 a 0 90.9 LM2PERMS 0 86.5 mean a 34 51 mean a 10 78.5 86 Table 5, continued. Dinopithecus ingens SWARTKRANS Gorgopithecus major KROMDRAAI Papio angusticeps KROMDRAAI Papio angusticeps TAUNG Papio robinsoni SWARTKRANS SOUTHERN AFRICA FRUIT LEAVES LM2SQ 68.9 14.3 LM2SQW 62.1 20.9 LM2NHXR 63.8 19.3 LM2PERS 73.1 9.9 LM2PERMS 62.4 16.8 LM2FL 66.0 19.2 LM3NHNR 60.8 22.4 UM2PERS 75.0 8.4 mean 66.5 16 LM2SQ 36.7 51.8 LM2SQW 47.5 35.5 LM2PERS 36.9 45.9 LM2PERMS 31.4 55.6 LM2FL 66.3 18.9 LM2NHNR 50.2 32.7 LM3NHNR 28.5 27.6 UM2PERS 100.0 0 UM2PERLS 100.0 0 UM2PERMS 100.0 0 mean 60 27 LM2SQ 58.3 29.1 LM2SQW 58.0 25.0 LM2NHSR 61.8 21.2 LM2PERS 61.9 21.1 LM2PERMS 56.1 24.7 LM2FL 53.3 31.4 LM3NHNR 45.9 36.9 UM2PERS 54.0 29.0 UM2PERLS 60.8 21.9 UM2PERMS 65.5 17.6 mean 58 24 LM3NHNR 63.3 20.0 UM2PERS 46.8 36.0 UM2PERLS 44.7 37.6 UM2PERMS 45.1 37.7 mean 50 33 LM2SQ 58.4 29.0 LM2SQW 58.3 24.7 LM2PERS 57.8 25.2 LM2PERMS 56.1 24.7 LM2NHNR 54.1 28.9 LM3NHNR 55.1 28.0 UM2PERS 67.1 16.1 UM2PERLS 68.0 14.8 UM2PERMS 61.5 21.6 mean 60 23 Parapapio Jonesi SWARTKRANS STERKFONTEIN Parapapio whitei SWARTKRANS STERKFONTEIN Parapapio broomi STERKFONTEIN Theropithecus darti SWARTKRANS LM2SQW 15.9 67.1 LM2FL 95.1 0 LM2NHNR 53.4 29.5 mean 55 32 LM2SQ 48.4 39.4 LM2SQW 53.3 29.8 LM2NHNR 62.2 20.8 LM2PERS 49.8 33.2 LM2PERMS 51.4 30.6 LM2FL 60.3 24.6 LM3NHXR 55.6 27.5 UM2PERS 48.5 34.4 UM2PERLS 40.8 41.4 UM2PERMS 61.0 22.1 mean 53 30 UM2PERS 67.9 15.4 UM2PERLS 61.1 21.6 UM2PERMS 27.5 55.0 mean 52 31 LM2SQ 46.8 41.1 LM2SQ;; 51.3 31.7 LM2NHNR 48.2 34.6 LM2PERS 45.7 37.2 LM2PERMS 46.6 36.6 LM2FL 36.6 47.5 LM3NHNR 45.2 37.5 UM2PERS 83.5 0.2 UM2PERLS 97.2 0 UM2PERMS 81.6 1.7 mean 54 27 LM2SQ 60.4 22.7 LM2FL 59.6 25.3 LM2NHNR 56.3 26.6 LM2PERS 67.3 15.2 LM2PERMS 45.3 38.2 LM3NHNR 43.9 38.8 mean 55.5 28 LM2SQ 23.6 65.4 LM2SQW 45.6 37.4 LM2NHNR 12.1 70.3 LM2PERS 33.1 41.7 LM2PERMS 35.8 50.1 LM2FL 95.4 0 LM3FL 5.9 77.2 LM3NHNR 0 84.9 mean 31 53 DISCUSSION The relative proportions of fruits, leaves and grasses in the diets of fossil baboons is probably related to whether or not they occupied forest or woodland habitats versus open or treeless savan- nahs. African grasslands are characterized by low and seasonal fruit productivity. Grasses are an important supplement to the diets of baboons living near or on the savannah during times of the year when fruits are unavailable (Dunbar 1983). Extant baboons exploiting forested habitats tend to eat higher quantities of fruits than grasses or herbs, while the opposite is true of baboons li- ving in scrub savannah (Dunbar 1983). The diet of Papio baboons living in unforested (desertic) areas of Namibia consisted of 80% grasses (Hamilton et al. 1978). In a similar environment in Ethiopia the baboon diet consisted of 40% grasses (Dunbar and Dunbar 1974). In contrast, grass composed only 10-20% of the diet of baboons living in heavily forested areas, and 20- 50% of the diet of baboons living in wooded but not heavily forested regions (Post 1978; Popp 1978; Kummer 1968; Sharman 1980; Dunbar and Dunbar 1974). Based on the principle of uni- formitarianism, it is assumed that the same relationship between diet and habitat observed for extant baboons would have existed for fossil papionins. 87 The idea that the diets and habitats of extant species can be used to infer that highly frugivo- rous fossil baboons of eastern Africa occupied forest or woodland habitats has been suggested independently by Leakey (1982). It is probably more than coincidence that the least frugivorous of the eastern African baboons, Parapapio ado, comes from the site of Laetoli which has been reconstructed as a dry savannah environment (Leakey 1982). Indications that the southern African papioninans were less frugivorous than those from eastern Africa (Table 5) is similarly consistent with reconstructions of the southern cave deposits as representing drier, more open savannah habitats than was typical of eastern African sites such as Koobi Fora and the Omo (Boaz 1977). In general, baboons from Sterkfontein and Taung appear to have included larger quantities of grasses or herbaceous foods in their diets than the baboons from Kromdraai and Swartkrans. The presence of the most frugivorous of the southern baboons at Swartkrans, Dinopithecus ingens, indicates that the site may have been associated with a higher degree of tree cover than Taung or Sterkfontein. This evidence contradicts inter- pretations of Swartkrans as representing a drier environment than was represented at Sterkfontein (Cartmill 1967; Vrba 1982). A situation analogous to that of the baboons was discerned for australopithecines based on the study of deciduous teeth (Grine 1981). Similar to the baboons, Grine (1981) observed that Aus- tralopithecus robustus at Swatans had shorter Table 6. Summary of dietary predictions for Theropithecus. SHEAR FRUITS Mean Range T. gelada 45.5 31-60 T. oswaldi 50 36-63 QUOTIENT LEAVES/GRASSES Mean Range 42.5 27-58 38 24-52 T. brumpti T. darti 28.5 0.5-46 24 60 32-89.5 65 PERMS FRUITS Mean Range T. gelada 0 T. oswaldi T. brumpti T. darti T. quadratirostris 47 30.5 40-54 26-3 7 36 28 LEAVES/GRASSES Mean Range 86.5 36 27-45 56 50-62 50 60 T. gelada T. oswaldi MEAN OF ALL FRUITS Mean Range 22 10-34 65.5 21.5-76 INDICES LEAVES/GRASSES Mean Range 65 51-78.5 36 12.5-61 T. brumpti T. darti 33 10-51 31 53 36-77 53 14 72 T. quadratirostris 88 shear crests and probably ate more hard fruits than Australopithecus africanus at Taung and Sterkfontein. However, Grine (1981) attributed the shorter shear crests of A. robustus to the con- sumption of small and hard, dry-adapted fruits and reconstructed Swartkrans as more xeric than Taung and Sterkfontein. It is equally possible that, as is the case for the baboons at Taung and Sterkfontein, A. africanus had longer shear crests than A. robustus because it supplemented its diet with grasses and herbs. According to Dunbar (1983), hard and small legumes are extremely scarce on the African savannah and would have proved an unlikely diet for any primate, including australopithecines. Based on this evidence, Grine's (1981) data corroborate the reconstruc- tion of Swartkrans as having been more mesic than Taung or Sterkfontein. The inclusion of more fruits in the diet of fossil Theropithecus indicates that its habitat may have been characterized by a higher degree of tree cover than that occupied by the extant species T. gelada. Theropithecus oswaldi, which probably consumed grasses rather than leaves, is likely to have occupied open country habitats adjacent to more wooded areas, such as grasslands growing along the margins of shallow lakes where sea- sonal flooding inhibited the growth of trees, as suggested by Jolly (1972). If the geologically more ancient of the eastern African species, T. brumpti and T. quadratirostris, consumed leaves rather than grasses as has been suggested here, it is possible that they inhabited forested environ- ments. This reconstruction is consistent with postcranial studies by Ciochon (1986) indicating that T. brumpti may have been less cursorial than T. oswaldi and that it may have occupied a forest habitat similar to that of the extant mandrill. The occurrence of T. darti, which is also reconstruc- ted as having eaten more leaves than grass, at Swartkrans provides further evidence that the site may have been more mesic than other studies have indicated. Since leaf-eating, forest adapted Theropithecus are more ancient than savannah dwelling species, we suggest that the origin of the genus is linked to the beginnings of leaf- eating in forest dwelling baboons, as opposed to grass-eating in open country environments as suggested by Jolly (1972). The dietary and habitat preferences of the fossil cercopithecines almost certainly influenced the patterns of species diversity and relative abundance observed for Plio-Pleistocene Thero- pithecus and papioninans in eastern and southern Africa. If the origin of the genus Theropithecus is linked to the beginnings of leaf-eating in forest dwelling baboons, as suggested here, the paucity of Theropithecus in southern Africa may be ex- plained by the absence of forested environments in that region. This hypothesis is consistent with the absence of leaf-eating colobine monkeys in the southern African cave deposits. The only fossil colobine found in southern Africa, Cerco- pithecoides, exhibits open country cursonral postcranial adaptations (Birchette 1981) and an unusual pattern of tooth wear that can only have resulted from the consumption of grasses (Benefit 1987). Alternatively, Theropithecus in southern Africa may have suffered from com- petition with grass-eating savannah adapted papioninan baboons. Papioninans were present in southern Africa during the late Miocene at the site of Langebaanweg (Grine and Hendey 1981), but Theropithecus did not occur in the area until the Middle Pliocene at the site of Makapansgat. If Theropithecus were endemic to eastern Africa, as seems likely, they may have arrived in south- ern Africa after the papioninan baboons had successfully fllled the grass-eating niches avail- able to monkeys, inhibiting Theropithecus from "swamping" the southern grasslands with its high population numbers, as T. oswaldi did in eastern Africa. The greater diversity of Theropithecus in the eastern region is in part attributable to reduced competition between species of the genus as a re- sult of differing preferences for fruits, leaves and grasses. The overwhelming abundance of T. os- waldi fossils in collections from eastern Africa between 2.5 my and 0.5 my may be due to its having lived closer to fluvial and lacustrine depo- sitional environments than the more forest adapted monkeys, resulting in a higher frequency of fossilization. It is also likely that population numbers of T. oswaldi were absolutely greater than those of the forest cercopithecines. Popula- tion densities of extant T. gelada are considerably higher than those of any known population of Papio, presumably because dense and evenly dis- tributed grasses can support larger numbers of animals than forest resources which are more sparsely and patchily distributed (Dunbar 1983). The lower species diversity and rarity of papioninan fossils in eastern Africa may have re- sulted from their preference for forest habitats. The eastern mangabeys and baboons would have competed for forest resources with T. brumpti and large-bodied colobine monkeys. Plio-Pleis- tocene colobine monkeys seem to have been less specialized for folivory than their extant relatives. Paracolobus and Rhinocolobus have been recon- structed as including almost equal portions of fruits and leaves in their diets (Benefit 1987, 1990). Competition between the forest baboons and colobines would have been more intense during the Plio-Pleistocene than it is today. As a 89 consequence, the diversity of forest dwelling members of both Colobinae and Cercopithecinae seems to have been effected. Since Papio ba- boons did not become the dominant savannah monkey in eastern Africa until after the demise of T. oswaldi, when it presumably began to exploit the grassland habitats for the first time, it is pos- sible that competition with T. oswaldi prevented the papioninans from taking advantage of grass- land resources at an earlier time. Competition with other forest monkeys and lack of grass-eat- ing savannah adaptations, combined with lower population densities in forest habitats, are pro- bably responsible for the low numbers and diversity of papioninan fossils at deposits in east- ern Africa. Thus, the greater abundance and diversity of the southern papioninan baboons is attributed to the absence of, and lack of com- petition with, T. oswaldi and forest dwelling colobines, to the general tendency for grasslands to support large numbers of animals, and to a di- versity of dietary preferences among the baboons themselves. SUMMARY Measurements of dental features shown to be functionally correlated to diet among extant mon- keys were used to establish criteria from which to assess the relative proportions of fruits and leaves/grasses consumed annually by extinct ba- boons from Plio-Pleistocene deposits in eastern and southern Africa. The reconstructed diets of fossil papioninans from southern Africa include a generally higher percentage of herbaceous mate- rials than do the diets of their eastern African counterparts. It is suggested that the southern baboons were savannah adapted, supplementing their diets with grasses dunng periods when fruits were seasonally unavailable, in a manner similar to extant Papio baboons (Dunbar 1983). The more frugivorous mangabeys and baboons of the eastern region probably occupied habitats with higher tree cover such as forests and wood- lands. Competition from large-bodied colobine monkeys and both grass- and leaf-eating Thero- pithecus, in addition to lower population densities, may be responsible for the rarity and low diversity of eastern African papioninan ba- boons. The greater abundance and diversity of the southern savannah adapted baboons is attributed to the late ardval and scarcity of Thero- pithecus, as well as to the capacity for grasslands to support large numbers of animals. Extinct species of Theropithecus are recon- structed to have included more fruits in their diets than do extant gelada baboons. Theropithecus brumpt was observed to have a greater potential for shearing and thinner enamel than T. oswaldi. It also placed less emphasis on the transverse component of mastication than T. oswaldi. This evidence indicated that T. brumpti may have con- sumed leaves rather than grasses. It is postulated that T. brwnpti inhabited forested habitats, unlike the more recent species T. oswaldi which consumed grasses and occupied open country habitats such as grasslands along seasonally flooded lakes and rivers. Since the folivorous Theropithecus species (T. brumpti, T. quadratir- ostris and T. darti) are geologically older than the grass-eating species (T. oswaldi and T. gelada), the origin of the genus is attributed here to the beginnings of folivory in a large-bodied forest dwelling papionin. The paucity of Theropithecus in southern African deposits can be explained by the absence of forested habitats in the region, and/or by the inability of T. darti to successfully compete with the savannah adapted papioninan baboons. The abundance of Theropithecus, es- pecially T. oswaldi, in eastern African deposits may be due to their occupation of habitats that were more prone to fossilization, such as the shores of lakes and rivers, as well as to their greater population numbers. ACKNOWLEDGMENTS We are grateful to the Office of the Pres- ident, Republic of Kenya for permission to conduct research in Kenya. We are also grateful to the Director (Richard Leakey) of the National Museums of Kenya, as well as to the curators (Meave Leakey and Lou Jacobs) and staff (Alice Maundu, Aifreda Ibui and Mary Muungu) of the Department of Palaeontology, for permission to study materials in their care and for the generous use of their resources. Access to southern Afri- can fossils and resources was granted by Philip Tobias of the University of the Witswatersrand Medical School, Department of Anatomy, and C.K. Brain, Elizabeth Vrba, Alan Turner and Da- vid Panagos of the Transvaal Museum. Prof F. Clark Howell generously granted access to speci- mens collected by the Omo expedition which are temporarily housed in the Laboratory for Human Evolutionary Studies at U.C. Berkeley. Grati- tude is also expressed to Guy Musser and Wolfgang Fuchs of the American Museum of Natural History where the comparative extant monkeys were measured. We thank Richard F. Kay for kindly teaching one of us (B.R.B.) his method of measuring shear crests, John Fleagle for providing much appreciated encouragement and advice, and Clifford Jolly, Eric Delson and 90 Terry Harrison for reading earlier versions of this paper. Funding for the project was provided by an I.T.T. International Fellowship to Kenya (1982-1983) and a New York University Dean's Dissertation Fellowship to B.R.B. REFERENCES CITED Aldrich-Blake, F.P.G. (1980) Long-tailed ma- caques. In D.J. Chivers (ed.), Malayan Forest Primates. New York: Plenum Press. Pp.147- 166. Andrews, Peter (1981) Species diversity and diet in monkeys and apes during the Miocene. In Christopher B. Stringer (ed.), Aspects of Human Evolution. London: Taylor and Fran- cis. Pp.25-61. Andrews, Peter and Leslie Aiello (1984) An evolutionary model for feeding and positional behavior. In D.J. Chivers, Bernard A. Wood and A. Bilsborough (eds.), Food Acquisition and Processing in Primates. New York: Ple- num Press. Pp.422-460. Benefit, Brenda R. (1987) The Molar Mor- phology, Natural History, and Phylogenetic Position of the Middle Miocene Monkey Victoriapithecus. New York University, un- published Ph.D. dissertation. Benefit, Brenda R. (1990) Fossil evidence for the dietary evolution of Old World Monkeys. American Journal of Physical Anthropology 81:193. Benefit, Brenda R. and Martin Pickford (1986) Miocene fossil cercopithecoids from Kenya. American Journal of Physical Anthropology 69:441-464. Birchette, M.G. (1981) Postcranial remains of Cercopithecoides. American Journal of Physi- cal Anthropology 54:201. Boaz, Noel T. (1977) Paleoecology of early Hominidae in Africa. Kroeber Anthropologi- cal Society Papers 50:37-62. Brown, Francis H. and W.P. Nash (1976) Ra- diometric dating and tuff mineralogy of Omo Group deposits. In Yves Coppens, F. Clark Howell, Glynn Ll. Isaac and Richard E.F. Leakey (eds.), Earliest Man and Environments in the Lake Rudolf Basin. Chicago: Univer- sity of Chicago Press. Pp.50-63. Brown, Francis H., I. McDougall, T. Davies and R. Maier (1985) An integrated Plio-Pleisto- cene chronology for the Lake Turkana Basin. In Eric Delson (ed.), Ancestors: The Hard Evidence. New York: Alan R. Liss, Inc. Pp. 82-90. Caldecott, J.O. (1986) An ecological and be- havioral study of the pig-tailed macaque. Contributions to Primatology 21:1-93. Cartmill, Matthew (1967) The early Pleistocene mammalian microfaunas of sub-Saharan Africa and their ecological significance. Quaternaria 9:169-198. Ciochon, Russell L. (1986) The Cercopithecoid Forelimb: Anatomical Implications for the Evolution of African Plio-Pleistocene Species. University of California, Berkeley, unpub- lished Ph.D. dissertation. Clutton-Brock, T.H. (1975) Feeding behavior of red colobus and black and white colobus in East Africa. Folia Primatologia 23:165-207. Crompton, A.W. and K. Hiiemae (1970) Molar occlusion and mandibular movement during occlusion in the American opossum, Didelphis marsupialis. Zoological Journal of the Lin- naean Society 49:21-47. Delson, Eric (1973) Fossil Colobine Monkeys of the Circum-Mediterranean Region and the Evolutionary History of the Cercopithecidae (Primates, Mammnalia). Columbia University, unpublished Ph.D. dissertation. Delson, Eric (1975) Evolutionary history of the Cercopithecidae. In Frederick S. Szalay (ed.), Approaches to Primate Paleobiology. Con- tributions to Primatology, volume 5. Basel: Karger. Pp.167-217. Dunbar, R.I.M. (1983) Theropithecines and hominids: Contrasting solutions to the same ecological problem. Journal of Human Evolu- tion 12:647-658. Dunbar, R.I.M. and P. Dunbar (1974) Ecolo- gical relations and niche separation among sympatric terrestrial primates in Ethiopia. Folia Primatologia 21:36-60. Eck, Gerald G. (1976) Cercopithecoidea from Omo Group deposits. In Yves Coppens, F. Clark Howell, Glynn Ll. Isaac and Richard E.F. Leakey (eds.), Earliest Man and Environ- ments in the Lake Rudolf Basin. Chicago: University of Chicago Press. Pp.332-344. Eck, Gerald G. (1977) Diversity and frequency distribution of Omo Group Cercopithecoidea. Journal of Human Evolution 6:55-63. Eck, Gerald G. and F. Clark Howell (1982) Un primate des formations plio-pleistocenes d'Afrique oriental: Theropithecus brumpti (Arambourg), (Primates, Cercopithecidae). Comptes Rendu, Seances de l'Academie des Sciences. Serie 1 295:397-400. Eck, Gerald G. and Nina G. Jablonski (1984) A reassessment of the taxonomic status and phy- letic relationships of Papio baringensis and Papio quadratirostris (Primates: Cercopitheci- dae). American Journal of Physical Anthro- pology 65:109-134. Eisenhart, W.L. (1974) The Fossil Cercopithe- 91 coids of Makapansgat and Sterkfontein. Har- vard University, unpublished A.B. thesis. Feibel, Craig S., Francis H. Brown and I. McDougall (1989) Stratigraphic context of fossil hominids from the Omo Group deposits: Northern Turkana Basin, Kenya and Ethiopia. American Journal of Physical Anthropology 78:595-622. Freedman, L. (1957) The fossil Cercopithecoi- dea of South Africa. Annals of the Transvaal Museum 23:121-262. Freedman, L. (1961a) New cercopithecoid fos- sils, including a new species from Taung, Cape Province, South Africa. Annals of the South Africa Museum 46:1-14. Freedman, L. (1961b) Some new fossil cerco- pithecoid specimens from Makapansgat, South Africa. Palaeontologia Africana 7:743. Freedman, L. (1965) Fossil and subfossil pri- mates from the limestone deposits at Taung, Bolt's Farm and Witkrans, South Africa. Palaeontologia Africana 9:18-48. Freedman, L. (1976) South African fossil Cercopithecoidea: A reassessment including a description of new material from Maka- pansgat, Sterkfontein, and Taung. Journal of Human Evolution 5:297-315. Freedman, L. and C.K. Brain (1972) Fossil cercopithecoid remains from the Kromdraai australopithecine site (Mammalia, Primates). Annals of the Transvaal Museum 28:1-16. Freedman, L. and N.S. Stenhouse (1972) The Parapapio species of Sterkfontein, Transvaal, South Africa. Palaeontologia Africana 14:93- 111. Freeland, W.J. (1979) Mangabey (Cercocebus albigena) social organization and population density in relation to food use and availability. Folia Primatologia 32:108-124. Gatinot, B.L. (1978) Characteristics of the diet of west African red colobus. Recent Advan- ces in Primatology 1:253-255. Gautier-Hion, A. (1978) Food niches and co- existence in sympatric primates in Gabon. Recent Advances in Primatology 1:269-286. Gautier-Hion, A. (1980) Seasonal variations in diet related to species and sex in a community of Cercopithecus monkeys. Journal of Animal Ecology 49:237-269. Grine, F.E. (1981) Trophic differences between 'gracile' and 'robust' australopithecines: A scanning electron microscope analysis of occlusal events. South African Journal of Sci- ence 77:203-230. Grine, F.E. and Q.B. Hendey (1981) Earliest primate remains from South Africa. South African Journal of Science 77:374-376. Hamilton, W.J., R.E. Buskirk and W.H. Bus- kirk (1978) Omnivory and utilization of food resources by chacma baboons, Papio ursinus. American Naturalist 112:110-120. Homewood, K.M. (1978) Feeding strategy of the Tana mangabey (Cercocebus galeritus galeritus) (Mammalia: Primates). Journal of Zoology 186:375-391. Jolly, Clifford J. (1970) The large African mon- keys as an adaptive array. In J.R. Napier and P.H. Napier (eds.), Old World Monkeys. New York: Academic Press. Pp.141-174. Jolly, Clifford J. (1972) The classification and natural history of Theropithecus (Simopithe- cus) (Andrews, 1916) baboons of the African Plio-Pleistocene. Bulletin of the British Mu- seum. Natural History, Geology Series 22(1): 1-123. Kay, Richard F. (1975) The functional adapta- tions of primate molar teeth. American Journal of Physical Anthropology 43(2):195-215. Kay, Richard F. (1977) The evolution of molar occlusion in the Cercopithecidae and early catarrhines. American Journal of Physical An- thropology 46:327-352. Kay, Richard F. (1978) Molar structure and diet in extant Cercopithecidae. In P.M. Butler and K. Joysey (eds.), Studies in the Development, Function, and Evolution of Teeth. London: Academic Press. Pp.309-339. Kay, Richard F. (1981) The nut-crackers -- A new theory of the adaptations of the Rama- pithecinae. American Journal of Physical Anthropology 55:141-152. Kay, Richard F. (1984) On the use of ana- tomical features to infer foraging behavior in extinct primates. In P.S. Rodman and J.G.H. Cant (eds.), Adaptations for Foraging in Nonhuman Primates. New York: Columbia University Press. Pp.21-53. Kay, Richard F. and Herbert H. Covert (1984) Anatomy and behavior of extinct primates. In D.J. Chivers, Bernard A. Wood and A. Bilsborough (eds.), Food Acquisition and Processing in Primates. New York: Plenum Press. Pp.467-508. Kay, Richard F. and K. Hiiemae (1974) Jaw movement and tooth use in recent and fossil primates. American Journal of Physical An- thropology 40:227-256. Kay, Richard F. and William L. Hylander (1978) The dental structure of mammalian folivores with special reference to primates and phalangeroids (Marsupialia). In G. Gene Montgomery (ed.), The Ecology of Arboreal Folivores. Washington, D.C.: Smithsonian Institution Press. Pp.173-192. 92 Kummer, Hans (1968) Social Organization of Hamadryas Baboons. Basel: Karger. Leakey, Mary D. and Richard L. Hay (1982) Fossil hominids from the Laetoli Beds. In Henry de Lumley and Marie-Antoinette de Lumley (eds.), Pretirage, Premier Congres International de Paleontologie Humaine, II. Nice: Centre National de Recherche Scienti- fique. Pp.753-765. Leakey, Mary D., Richard L. Hay, Garniss H. Curtis, Robert E. Drake, M.K. Jackes and Timothy D. White (1976) Fossil hominids from the Laetoli Beds. Nature 262:460466. Leakey, Meave G. (1976) Cercopithecoidea of the East Rudolf succession. In Yves Cop- pens, F. Clark Howell, Glynn LI. Isaac and R.E.F. Leakey (eds.), Earliest Man and Envi- ronments in the Lake Rudolf Basin. Chicago: University of Chicago Press. Pp.345-350. Leakey, Meave G. (1982) Extinct large colo- bines from the Plio-Pleistocene of Africa. American Journal of Physical Anthropology 58:153-172. Leakey, Meave G. and Eric Delson (1987) Fos- sil Cercopithecidae from the Laetolil Beds. In Mary D. Leakey and John M. Harris (eds.), Laetoli: A Pliocene Site in Northern Tanzania. Oxford: Clarendon Press. Pp.91-107. Leakey, Meave G. and Richard E.F. Leakey (1973a) Further evidence of Simopithecus (Mammalia, Primates) from Olduvai and Olor- gesailie. Fossil Vertebrates of Africa 3:101- 120. Leakey, Meave G. and Richard E.F. Leakey (1973b) New large Pleistocene Colobinae (Mammalia, Primates) from East Africa. Fos- sil Vertebrates of Africa 3:121-138. Leakey, Meave G. and Richard E.F. Leakey (1976) Further Cercopithecinae (Mammalia, Primates) from the Plio-Pleistocene of East Africa. Fossil Vertebrates of Africa 4:121- 146. Leakey, Richard E.F. (1969) New Cercopithe- cidae from the Chemeron Beds, Lake Baringo, Kenya. Fossil Vertebrates of Africa 1:53-69. MacKinnon, J.R. and K.S. MacKinnon (1978) Comparative feeding ecology of six sympatric primates in West Malaysia. Recent Advances in Primatology 1:305-321. Mah, Y.L. (1980) The Ecology and Behavior of Macaca fascicularis. University of Malay, Ku- ala Lumpur, unpublished Ph.D. dissertation. Maier, W. (197 la) New fossil Cercopithecoidea from the Lower Pleistocene cave deposits of the Makapansgat Limeworks, South Africa. Palaeontologia Africana 13:69-108. Maier, W. (1971b) Two new skulls of Para- papio antiquus from Taung and a suggested phylogenetic arrangement of the genus Para- papio. Annals of the South Africa Museum 59:1-16. Marsh, C. (1981) Diet choice among red colobus (Colobus badius rufomitratus) on the Tana River, Kenya. Folia Primatologia 35: 147-178. McKey, P. (1978) Soils, vegetation, and seed- eating by black colobus monkeys. In G. Gene Montgomery (ed.), The Ecology of Arboreal Folivores. Washington, D.C.: Smithsonian Institution Press. Pp.423-437. Napier, J.R. (1970) Paleoecology and catar- rhine evolution. In J.R. Napier and P.H. Napier (eds.), Old World Monkeys. New York: Academic Press. Pp.53-96. Oates, J. (1977) The guereza and its food. In T.H. Clutton-Brock (ed.), Primate Ecology. New York: Academic Press. Pp.276-321. Popp, J. (1978) Male Baboons and Evo- lutionary Principles. Harvard University, unpublished Ph.D. dissertation. Post, D. (1978) Feeding and Ranging Behavior of the Yellow Baboon (Papio cynocephalus). Yale University, unpublished Ph.D. disserta- tion. Potts, Richard (1989) Olorgesailie: New exca- vations and findings in early and middle Pleistocene contexts, southern Kenya rift val- ley. Journal of Human Evolution 18:477-484. Quris, R. (1975) Ecologie et organisation soci- ale de Cercocebus galeritus agilis dans le nord- est du Gabon. Terre et Vie 29:337-398. Raemakers, J.J. and D.J. Chivers (1980) So- cioecology of Malayan forest primates. In D.J. Chivers (ed.), Malayan Forest Primates. New York: Plenum. Pp.279-316. Sharman, M. (1980) Feeding, Ranging and Social Organization of the Guinea Baboon, Papio papio. University of St. Andrews, un- published Ph.D. dissertation. Shuey, R.T., Francis H. Brown and M.K. Croes (1974) Magnetostratigraphy of the Shungura Formation, Southwestern Ethiopia: Fine structure of the lower Matuyama polarity epoch. Earth and Planetary Science Letters 23: 249-260. Struhsaker, T.T. (1975) The Red Colobus Monkey. Chicago: University of Chicago Press. Struhsaker, T.T. (1978) Food habits of five monkey species in the Kibale Forest, Uganda. Recent Advances in Primatology 1:225-248. Szalay, Frederick S. and Eric Delson (1979) Evolutionary History of the Primates. New York: Academic Press. Temerin, L.A. and J.G.H. Cant (1983) The evolutionary divergence of Old World mon- 93 keys and apes. American Naturalist 122:335- 351. Vrba, E. (1982) Biostratigraphy and chronol- ogy, based particularly on Bovidae, of southern hominid-associated assemblages: Makapansgat, Sterkfontein, Taung, Krom- draai, Swartkrans; also Elandsfontein (Saldanha), Broken Hill (now Kabwe) and Cave of Hearths. In Henry de Lumley and Marie-Antoinette de Lumley (eds.), Pretirage, Premier Congres International de Paleontolo- gie Humaine, II. Nice: Centre National de Recherche Scientifique. Pp.707-752. Walker, P. and P. Murray (1975) An assess- ment of masticatory efficiency in a series of anthropoid primates with special reference to the Colobinae and Cercopithecinae. In Russell L. Tutde (ed.), Primate Functional Morphol- ogy and Evolution. The Hague: Mouton. Pp. 151-194. Waser, P.M. (1977) Feeding, ranging and group size in the mangabey Cercocebus albi- gena. In T.H. Clutton-Brock (ed.), Primate Ecology. New York: Academic Press. Pp. 183-222. Waser, P.M. (1984) Ecological differences and behavioral contrasts between two mangabey species. In P.S. Rodman and J.G.H. Cant (eds.), Adaptations for Foraging in Nonhwnan Primates. New York: Columbia University Press. Pp.195-216. Wheatley, B.P. (1980) Feeding and ranging of east Bornean Macacafascicularis. In D. Lind- burg (ed.), The Macaques. New York: Van Nostrand Reinhold Company. Pp.215-246. 94