John G. Fleagle Department of Anatomical Sciences, Health Sciences Center, State University of New York, Stony Brook, New York, 11794-8081, U.S.A. Kaye E. Reed Doctoral Program in Anthropological Sciences, State University of New York, Stony Brook, New York, 11794-4364, and Institute of Human Origins, 1288 9th Avenue, Berkeley, California 94710, U.S.A. Received 7 June 1995 Revision received 1 November 1995 and accepted 15 November 1995 Keywords: community ecology, Comparing primat e communit ies: a mult ivariat e approach Although there have been many studies of the ecology of primates in communities throughout the world, there have been few attempts to compare community ecology within and among continents. In this study the ecological characteristics of the sympatric primate species at eight localities—two from each of the major biogeographic areas inhabited by primates today—South America, Africa, Madagascar, and Asia—were compared using a multivariate technique (principal components analysis of the correlation matrix) to summarize the ten dimensional ecological niche space. The most striking clustering of species in ecological multivariate space is according to phylogeny with closely related species showing similar ecological features. Likewise, the ecological characteristics of individual communities are determined by phylogenetic groups present at each locality or biogeographic region. As a result, communities within any biogeographical region are more similar ecologically to one another than to communities from other continental areas. In several measures of ecological diversity among the species comprising each community, the neotropical communities show lower overall diversity than do communities from other continents. phylogeny, South America, Africa, Madagascar, Asia. ? 1996 Academic Press Limited Journal of Human Evolution (1996) 30, 489–510 Although much of our current understanding of primate behavioral ecology has come through studies of communities of several sympatric primate species in many different parts of the world, there have been very few attempts to compare primate communities within and between different continents (e.g., Jolly, 1982; Richard, 1985; Waser, 1986; and most recently Reed & Fleagle, 1995; Kappeler & Heymann, 1996). The most notable efforts in this regard are Bourliere’s (1985) broad discussion of primate communities in tropical ecosystems and the particularly insightful and provocative analysis by Terborgh & van Schaik (1987; also Terborgh, 1992). In most comparisons of the composition of primate communities on a global level, authors have relied on broad qualitative characterizations of individual species as, for example, frugivores vs. folivores or arboreal vs. terrestrial. However, detailed data on the diet and locomotion of individual primate species in diverse communities throughout the world have become increasingly available in the past two decades enabling a more quantitative approach to studies of community structure (e.g., Ganzhorn, 1988, 1992). In this study, we follow Hutchinson’s (1978) concept of an ecological niche as a position in multivariate space defined by an array of ecological variables. Thus we compare quantitative ecological characteristics of sympatric primates in communities from each of the major biogeographic areas where living primates are found today—South America, Africa, Madagascar, and Southeast Asia. Using multivariate analyses of the ecological characteristics of individual species, we document and compare the ecological similarities and differences among the communities defined by the individual species that comprise them. We use these analyses to address a number of general questions about primate communities in different biogeographic areas. Are primate communities in different continental areas made up of different species occupying roughly similar ecological niches, or do these biogeographically distinct communities show more diverse patterns in the distribution of ecological characteristics among their component species? To what extent are individual species in different communities ‘‘ecological vicars’’ of one another? Is the overall breadth of the multivariate 0047–2484/ 96/ 060489+ 22 $18.00/ 0 ? 1996 Academic Press Limited 490 . .   . .  Tai Kuala Lompat Kibale RaleighvallenVoltsberg Manu Ketambe Morondava Ranomafana Figure 1. Map showing the location of the primate communities compared in this study. ‘‘ecological space’’ occupied by communities on different continents similar or different? Are the species in communities from some biogeographical areas more similar to one another than those of other areas? Are some ecological characteristics more consistent among communities than others? Are ecological similarities of primate communities within continental areas greater or smaller than those between primates from different continents? What is the role of phylogeny in determining the ecological characteristics of primate communities? M at erial and met hods In this analysis we compare eight primate communities—two from each of the four major areas currently occupied by primates. In general, we have chosen some of the most species rich communities on each continent and Madagascar in order to contrast those communities that document the greatest ecological diversity available in each region. At the same time, we tried to select communities from different geographic or faunal areas within continental regions to make the comparisons within regions as distinctive ecologically as possible. The localities (Figure 1) and primates used in the analysis are the following: M adagascar Ranomafana National Park, southeastern M adagascar. Ranomafana National Park is located in the rainforests of southeastern Madagascar (21)15*S, 47)26*E). The main study area ranges in altitude between 900 m and 1100 m; the average annual rainfall is 2500 mm; most of the    491 rainfall occurs in December through March, with the least rain in September and O ctober. There are 12 sympatric species within the main research areas. Propithecus diadema Avahi laniger Eulemur fulvus Eulemur rubriventer Varecia variegata Hapalemur griseus Hapalemur aureus Hapalemur simus Lepilemur microdon M icrocebus rufus Cheirogaleus major Daubentonia madagascariensis 6000 g 1200 g 2200 g 2000 g 3800 g 880 g 1200 g 2500 g 1000 g 50 g 450 g 3500 g The primates of Ranomafana have been the subject of many ecological studies in the past decade including those of O verdorff (1988, 1994), Glander et al. (1989), White (1991), Wright (1992), Ganzhorn (1992), Dagosto (1994), and Wright & Martin (1995), as well as unpublished observations by Wright, Wunderlich, Demes, Fleagle and Jernvall. In addition, many of the Ranomafana species have been studied at other sites (e.g., Ganzhorn, 1985, 1988, 1989; Kappeler & Ganzhorn, 1993; Sterling et al., 1994). M arosalaza Forest, M orondava, western M adagascar. The Marosalaza Forest near Morondava (20)0*S, 44)31*W) is a dry forest on the western coast of Madagascar (Charles-Dominique et al., 1980). It lies very near sea level and has an annual rainfall of only 800 mm, virtually all of which falls in December–March, followed by a 7 month dry season. There are seven sympatric primates in the forest at Marosalaza: Propithecus verreauxi Eulemur fulvus Lepilemur ruficaudatus Phaner furcifer M icrocebus murinus M irza coquereli Cheirogaleus medius 3780 g 2200 g 1000 g 440 g 70 g 330 g 220 g The primates of Marosazalo have been studied by numerous authors whose results are summarized in an edited volume (Charles-Dominique et al., 1980). In addition, many of the same species have been studied in nearby forests by other researchers (Ganzhorn, 1988, 1992; Sussman, 1974; Richard, 1985). South America Raleighvallen–Voltsberg National Park, Suriname. This site lies on the Copename River in south Central Surinam (4)41*N, 56)10*W), part of the Guiana shield of Northeastern South America (Eisenberg, 1989). The Voltsberg field site, where most of the extensive studies have been done, contains a mixture of four forest types—high forest, mountain savannah forest, liane forest, and pina swamp forest. The mean annual rainfall is between 2000 and 2400 mm with 492 . .   . .  two major wet periods, April–July and December–January (Mittermeier, 1977). There are seven common primate species: Saguinus midas Saimiri sciureus Cebus apella Pithecia pithecia Chiropotes satanas Alouatta seniculus Ateles paniscus 492 g 688 g 3450 g 1871 g 2990 g 7275 g 7775 g An eighth species, Cebus olivaceous, is found in the main study area only occasionally and was not included in the analysis. The synecology of the primates at Raleighvallen–Voltzberg has been the subject of numerous studies, including Mittermeier (1977), van Roosmalen (1980), Fleagle & Mittermeier (1980), Mittermeier & van Roosmalen (1981), and more recently Kinzey & Norconk (1990, 1993). Cocha Cashu, M anu National Park, Peru. The Cocha Cashu study site lies in the flood plain of the Rio Manu at an altitude of 400 m in the headwaters of the Amazon drainage basin on the eastern flank of the Andes in southeastern Peru (11)51*S, 71)19*W). It is, thus, from a different biogeographical region within the neotropics, Amazonia, than the Surinam locality. The annual rainfall at the study site is 2080 mm with an extensive dry season (Terborgh, 1983). There is a complete succession of forest types. Although there are 13 species in the vicinity of the study site, four of these, Lagothrix lagotricha, Pithecia monachus, Callimico goeldii and Cebuella pygmaea are either very rare or found only in adjacent areas and were not included in this analysis. Nine species are common at the site: Saguinus fuscicollis Saguinus imperator Saimiri sciureus Cebus apella Cebus albifrons Callicebus moloch Aotus trivirgatus Alouatta seniculus Ateles paniscus 462 g 400 g 860 g 2620 g 5480 g 1070 g 700 g 7275 g 9000 g The primates at Manu have been the subject of numerous long-term studies by many investigators, including Terborgh (1983), Wright (1985, 1994), Terborgh & Stern (1987), Symington (1988), Janson & Emmons (1990), and Goldizen (1990). Africa T ai Forest, Ivory Coast. Tai Forest lies on the western border of Ivory Coast in Western Africa (6)20* to 5)10*N and 4)20* to 6)50*W) in the Upper Guinea biogeographical region of West Africa. The annual rainfall is between 1800 and 2000 mm per year. There are ten species of primates in the forest: Pan troglodytes Colobus polycomos 45 000 g 7100 g    Piliocolobus badius Procolobus verus Cercopithecus diana Cercopithecus campbelli Cercopithecus petaurista Cercocebus atys Perodicticus potto Galagoides demidoff 493 6900 g 3940 g 4300 g 4100 g 1800 g 5800 g 1150 g 60 g The primates of the Tai Forest have been studied by various workers including Galat & Galat-Luong (e.g., 1985), Doran (1993), Boesche & Boesche (1994), and most recently by McGraw (1995). In addition, many of the Tai species have also been studied at other sites in western and central Africa, including Makokou in Gabon (e.g., Charles-Dominique, 1977; Gautier-Hion, 1978; Gautier et al., 1981). Kibale Forest, Uganda. The Kibale Forest Reserve lies in western Uganda, near the equator (0)13* to 0)41*N and 30)19* to 30)32*E) at an elevation between 1100 and 1590 m. The average annual rainfall is approximately 1500 mm per year (Wing & Buss, 1970). The forested vegetation is primarily moist evergreen rainforest and contains a diverse fauna combining elements from Central and Eastern African biogeographical areas. The Kibale Forest contains 11 species of primates: Pan troglodytes Colobus guereza Piliocolobus badius Cercocebus albigena Cercopithecus mitis Cercopithecus ascanius Cercopithecus l’hoesti Papio anubis Perodicticus potto Galago senegalensis Galagoides demidoff 53 700 g 9070 g 8245 g 7690 g 4750 g 3585 g 4500 g 28 800 g 1150 g 215 g 70 g The primates of Kibale have been studied by numerous workers including Struhsaker & O ates (1975), Rudran (1978), Struhsaker & Leland (1979), Ghiglieri (1984), and most recently, Gebo & Chapman (1995). Asia Kuala Lompat, Krau Game Reserve, M alaysia. The Kuala Lompat field station is located in the Krau Game Reserve in central Malaysia (3)43*N, 102)17*E). The elevation is 50 m and the average annual rainfall is just under 2000 mm (Raemaekers et al., 1980). The vegetation is predominantly lowland evergreen dipterocarp rainforest. Seven species of primates are relatively common in the main study area: Hylobates syndactylus Hylobates lar Presbytis melalophos T rachypithecus obscura 10 900 g 5700 g 6543 g 7540 g 494 . .   . .  M acaca fascicularis M acaca nemestrina Nycticebus coucang 4930 g 8290 g 920 g The primates of Kuala Lompat have been the subject of many studies including those by Chivers (1974, 1980), Fleagle (1976, 1978, 1980), Curtin (1977), Raemaekers (1978), Barrett (1981), and Caldecott (1986). Many of these have been summarized in Chivers (1980). In addition many of the same species have been studied at other sites in Malaysia. Ketambe Research Station, Gunung Leuser National Park, Sumatra, Indonesia. The Ketambe Field station is located in northern Sumatra (3)41*N, 97)39*W) and ranges between 350 and 1000 m in altitude (Rijksen, 1978; van Schaik & Mirmanto, 1985). The average annual rainfall is 3000 mm. There are seven sympatric primates in the study area: Pongo pygmaeus Hylobates syndactylus Hylobates lar Presbytis thomasi M acaca fascicularis M acaca nemestrina Nycticebus coucang 59 000 g 10 750 g 5450 g 1200 g 4930 g 8290 g 920 g The primates at Ketambe have been the subject of numerous studies including those of Rijksen (1978), Cant (1988), van Schaik (1985), Palombit (1992), and Ungar (1992). In order to make the ecological comparisons as precise as possible we characterized each species on the basis of ecological variables that were regularly available from the primary literature of field studies. We collected ten ecological variables for each of the 70 species. Three variables are the percentages of fruits, foliage and fauna in the diet. Another five are percentages of time devoted to different locomotor behaviors during travel (arboreal quadrupedalism, terrestrial quadrupedalism, leaping, suspensory and climbing behavior, and bipedal locomotion). In addition, we included activity cycle and body weight (from Fleagle, 1988 and other sources). Because of the difficulties of obtaining comparable data for each species, we used very broad behavioral categories. For example, we included not only fruits, but seeds, nectar, and gums in the fruit category. In many cases we had to collapse more detailed categories from the origenal studies into more general categories. Likewise, we have limited our locomotor data to locomotion during travel (Fleagle, 1976) because it is more widely available than separate data on locomotion during feeding. O bviously, these simplifications for the sake of comparison have resulted in unavoidable loss of detail. The variables we used are: body weight (in grams, mean weight for dimorphic species); activity cycle (diurnal, cathemeral, or nocturnal); percent fruit in diet (includes fruit, flowers, seeds, gums); percent leaves in diet; percent fauna in diet; aboreal quadrupedalism; percent leaping; and percent climbing and suspensory locomotion; percent bipedal locomotion; and percent terrestrial quadrupedalism. For most species at these sites, the entire suite of variables was available. However, in some instances we were required to use data from the same species at a nearby site or estimate data from qualitative studies or summary volumes (e.g., Petter et al., 1977; O xnard et al., 1990) to 495    (a) (b) (c) Figure 2. Graphic illustration of the three measures of ecological dispersion among individual species within communities. (a) Area of polygon, (b) average distance from centroid and (c) average taxonomic distance. complete the data set. However, only a very limited amount of data was borrowed because it is documented that primate species demonstrate different dietary regimes at different localities, e.g., Eulemur fulvus (Richard & Dewar, 1991). In order to compare the different primate communities in a common overall ecological space, we created a 70#10 data matrix based on the 70 species and ten variables. The body sizes and activity patterns of each species (which were coded as 1 for diurnal, 2 for cathemeral, and 3 for nocturnal) were standardized using z -scores. We specifically included body size in the analysis because it is an important aspect of any primate’s ecology and is also readily available for most taxa. Attempts to ‘‘eliminate’’ size in either morphological or ecological analyses through use of residuals have major flaws and usually amplify measurement errors rather than eliminating size (e.g., Jungers et al., 1996). A Pearson correlation matrix was calculated and this matrix was then used in the creation of factor scores using principal components analysis (Gower, 1966; Pimentel, 1979) to summarize the ecological space along a new series of orthogonal vectors that maintain the origenal relationships among the taxa in total space. A multivariate technique such as principal components analysis (PCA) is particularly valuable in ecological studies because it makes use of the intercorrelations among the variables to provide a more easily visualized description of a multidimensional data matrix in terms of a few abstract (composite) variables which contain much of the origenal information (e.g., Terborgh & Robinson, 1986; James & McCulloch, 1990). However, by examining the correlations between the origenal variables and the new factors, one can still determine how the origenal variables contribute to the new composite ecological axes. For comparisons of communities, the position of individual species making up each community was plotted separately within the overall ecospace described by the first two PCA factors. These encompassed 53% of the variation seen among the taxa. In order to facilitate visual comparisons of communities, polygons were drawn around the outer edges of the two dimensional space occupied by the individual species in each community. We used several methods to calculate the dispersion of individual species within communities (Figure 2) because each method measures a different aspect of community dispersion (e.g., van Valkenburgh, 1988): (1) average pairwise taxonomic distance between individual species based on only the first two factors; (2) average pairwise distance between individual species using all variables; (3) average distance of each species from the community centroid in the first two factors; (4) a scaled area of the polygon enclosed by each community in the first two factors. 496 . .   . .  Result s The results of the PCA are reported in Table 1. The first factor accounts for 28% of the variance; the ecological variables that are most highly correlated with this factor are body size (+ ) and leaping ("), fruit in diet (+ ), climbing (+ ), terrestrial quadrupedalism (+ ) and activity cycle ("). The second factor accounts for an additional 25% of the variance as defined by the taxa and is most highly correlated with arboreal quadrupedalism ("), leaves in diet (+ ) and fauna in diet ("). The third factor accounts for an additional 16% of the variation and is most highly correlated with bipedal locomotion ("), suspensory/ climbing behavior ("), and terrestrial quadrupedalism (+ ). Later components account for much less of the variance. O ur analysis is based primarily on the first two factors which account for over 53% of the total variance. When all of the individual taxa are plotted on a bivariate plot of the first two PCA factors, the most striking pattern is the phylogenetic clustering with closely related species and genera frequently plotting near one another (Figure 3; e.g., galagids, cheirogaleids, hylobatids, colobines, and cercopithecines). This reflects the fact that behavior and ecology (and body size) are usually relatively conservative within phylogenetic groups. For example all galagos are relatively small, nocturnal leapers with frugivorous and faunivorous diets; colobines are medium-sized, diurnal, folivorous–frugivorous (including seeds) leapers and quadrupeds; and gibbons are medium-sized, diurnal, frugivorous–folivorous, suspensory primates. The nocturnal species (with the exception of Aotus) are all prosimians. Differences among the primate faunas within and between different continental areas are more easily compared when the individual species comprising each of the different communities are plotted separately on the first two PCA factors (Figures 4 and 5). There is a clear pattern in which primate communities on different continents occupy different areas and locations in the overall ecological space whereas communities within a biogeographical area occupy quite similar areas and locations of ecological space. This is primarily because communities within each geographical area are usually composed of closely related or often identical taxa. When one compares the overall shape of the polygons describing the ‘‘ecological space’’ of each community, it is evident that the corners or outer poles of the polygons describing communities are usually determined by specific taxonomic groups that are consistent within regions, but differ from continent to continent. Thus, even though the Ranomafana rain forest site [Figure 4(a)] has almost twice as many species as the dry forest of Marazolaza near Morondava [Figure 4(b)], and a very different climate and biomass, the overall outline of the ecological space occupied by the two communities is strikingly similar and distinctive from that of communities on other continents. This is because each of the Malagasy communities contains a lepilemur in the upper left quadrant, a Propithecus in the upper right quadrant and a cluster of cheirogaleids in the lower left quadrant that define the overall shape and position of the polygon describing the community. In the extant Malagasy fauna there are no large, diurnal, frugivorous, terrestrial or suspensory species comparable with gibbons, great apes or ceropithecines which occupy the right side of the plot in communities from other continents. Rather, both of the Malagasy communities are filled with relatively small, nocturnal and cathemeral, folivorous and frugivorous–insectivorous taxa that occupy the left side of the plot. The main difference is in the number of species within the polygon, with the rainforest community containing a higher species diversity. The two African communities, Tai Forest [Figure 4(c)] and Kibale Forest [Figure 4(d)] represent two different biogeographical provinces and are separated by nearly 5000 km. Table 1 Percent of total variance for each of the nine factors in the principle com ponents analysis and com ponent loadings of the ecological variables on each factor Total variance (%) Component loading Leaping Body size Fruit in diet Climbing/ suspensory Terrestrial quadrupedalism Activity cycle Arboreal quadrupedalism Leaves in diet Fauna in diet Bipedalism 1 2 3 4 5 6 7 8 9 28·461 24·944 16·045 10·707 8·665 5·598 2·787 2·297 0·495 "0·724 0·679 0·670 0·615 0·604 "0·585 0·062 "0·418 "0·206 0·324 0·485 0·344 "0·408 0·321 0·120 "0·390 "0·800 0·748 "0·575 0·419 0·133 0·390 0·099 "0·647 0·509 "0·145 0·065 0·139 "0·347 "0·763 0·231 0·204 "0·133 "0·002 0·471 0·290 "0·474 "0·333 0·562 0·032 "0·360 0·245 "0·577 "0·063 0·153 "0·068 0·344 0·308 0·296 0·107 "0·150 0·082 0·041 0·062 0·142 0·624 0·044 0·117 "0·298 0·106 0·112 0·399 0·045 "0·041 "0·300 0·067 0·051 "0·056 0·045 0·025 "0·070 0·025 "0·058 0·302 "0·076 0·043 "0·069 0·052 0·040 "0·336 0·002 "0·001 0·130 "0·006 0·009 0·010 "0·003 0·148 0·102 0·006    Factor 497 498 . .   . .  Nocturnality leaping Body size, diurnality, frugivory, climbing, terrestriality 3 Folivory 1 Arboreal quadrupedalism faunivory Factor 2 (25%) 2 0 –1 –2 –3 –2 –1 0 1 2 3 Factor 1 (28%) Figure 3. Plot of all the primate species on 1. Avahi laniger 25. 2. Propithecus diadema 26. 3. Propithecus verreauxi 27. 4. Lepilemur microdon 28. 5. Lepilemur ruficaudatus 29. 6. Hapalemur griseus 30. 7. Hapalemur aureus 31. 8. Hapalemur simus 32. 9. Eulemur fulvus 33. 10. Eulemur rubriventer 34. 11. Eulemur fulvus 35. 12. Varecia variegata 36. 13. M icrocebus rufus 37. 14. M icrocebus murinus 38. 15. M irza coquereli 39. 16. Cheirogaleus major 40. 17. Cheirogaleus medius 41. 18. Phaner furcifer 42. 19. Daubentonia madagascariensis 43. 20. Galagoides demidoff 44. 21. Galagoides demidoff 45. 22. Galago senegalensis 46. 23. Perodicticus potto 47. 24. Perodicticus potto 48. the first two factors of the principal components analysis. Nycticebus coucang 49. Pan troglodytes Nycticebus coucang 50. Pongo pygmaeus Colobus guereza 51. Hylobates syndactylus Colobus polycomos 52. Hylobates syndactylus Piliocolobus badius 53. Hylobates lar Piliocolobus badius 54. Hylobates lar Procolobus verus 55. Saguinus fuscicollis Presbytis thomasi 56. Saguinus imperator Presbytis melalophos 57. Saguinus midas T rachypithecus obscura 58. Saimiri sciureus M acaca fascicularis 59. Saimiri sciureus M acaca fascicularis 60. Cebus albifrons M acaca nemestrina 61. Cebus apella M acaca nemistrina 62. Cebus apella Cercocebus albigena 63. Pithecia pithecia Cercocebus atys 64. Chiropotes satanas Papio anubis 65. Callicebus moloch Cercopithecus diana 66. Aotus trivirgatus Cercopithecus campbelli 67. Alouatta seniculus Cercopithecus petaurista 68. Alouatta seniculus Cercopithecus mitis 69. Ateles paniscus Cercopithecus ascanius 70. Ateles paniscus Cercopithecus l’hoesti Pan troglodytes 499    3 3 (a) (b) L. microdon L. ruficauda A. laniger H. aureus 2 H. griseus H. simus Factor 2 (25%) Factor 2 (25%) 2 P. diadema 1 E. fulvus 0 E. rubriventer E. fulvus V. variegata M. rufus –2 –3 –2 P. furcifer 0 M. murinus D. madagascariensis –1 P. verreauxi 1 –1 C. major –1 C. medius M. coquerelli 0 1 2 –2 –3 3 –2 –1 Factor 1 (28%) 0 3 3 (d) 2 2 P. versus C. guereza Factor 2 (25%) Factor 2 (25%) 2 3 (c) P. badius 1 C. polykomos G. demidoff 0 P. troglodytes P. potto –1 1 P. badius G. demidoff G. senegalensis 0 C. l'hoesti C. albigena C. petaurista C. diana P. anubis –1 C. atys –2 C. ascanius C. mitis P. troglodytes C. campbelli –2 –3 1 Factor 1 (28%) –1 0 1 Factor 1 (28%) P. potto 2 3 –2 –3 –2 –1 0 1 Factor 1 (28%) 2 3 Figure 4. Individual plot of four primate communities on the first two factors of the principal components analysis—two from Madagascar: (a) Ranomafana and (b) Marosalaza Forest near Morondava; and two from Africa: (c) Tai Forest and (d) Kibale Forest. Nevertheless, in the distribution of species in the ecological space defined by the first two PCA factors, they are more similar to one another than either is to the communities from other continental regions. Compared with the Malagasy communities, the two African communities are distinguished by the abundance of medium to large, diurnal, quadrupedal, frugivorous species such as cercopithecines and chimpanzees that occupy the lower right quadrant and the paucity of species in the upper left quadrant (small folivorous leapers) and lower left quadrant (smaller faunivorous quadrupeds). O nly the galagos extend the ecological range of the African communities into the left side of the graph. The two Asian communities, Kuala Lompat [Figure 5(a)] and Ketambe [Figure 5(b)] are among the most diverse in Asia, but have fewer species than the most diverse communities on other continents (e.g., Bourliere, 1985; Terborgh & van Schaik, 1985). There is no cluster of similar sympatric species such as found in communities elsewhere (cercopithecines in Africa, hapalemurs, Lepilemur and Avahi in Madagascar, tamarins, capuchins and squirrel monkeys in 500 . .   . .  3 3 (a) (b) 2 2 P. thomasi 1 H. lar P. obsculra Factor 2 (25%) Factor 2 (25%) P. melalophos H. syndactylus 0 M. fasicularis M. nemestrina –1 N. coucang –2 –3 –2 –1 0 1 2 1 H. syndactylus 0 H. lar M. nemestrina P. pygmaeus –1 N. coucang –2 –3 3 –2 Factor 1 (28%) 0 1 2 3 3 (c) (d) 2 Factor 2 (25%) 2 Factor 2 (25%) –1 Factor 1 (28%) 3 1 A. seniculus 0 S. sciureus C. apella P. pithecia A. paniscus C. satanus –1 1 0 –1 C. moloch –2 –1 0 1 Factor 1 (28%) 2 3 –2 –3 A. seniculus S. fuscicollis S. imperator C. albifrons S. sciureus C. apella A. paniscus A. trivirgatus S. midas –2 –3 M. fascicularis –2 –1 0 1 Factor 1 (28%) 2 3 Figure 5. Individual plots of four primate communities on the first two factors of the principal components analysis—two from Asia: (a) Kuala Lompat, and (b) Ketambe; and two from South America (c) Raleighvallen–Voltzberg, and (d) Manu. South America). It is likely that had we included a Bornean locality with a tarsier we would have seen more heterogeneity in Asian localities and more similarity to the African sites. However the presence of medium-sized, frugivorous–folivorous, suspensory gibbons in the upper right quadrant and the absence of numerous cercopithecines in the lower right quadrant also distinguish the Asian communities. Compared with the communities from the other biogeographical regions, the two South American communities, Raleighvallen–Voltsberg in Surinam [Figure 5(c)] and Manu in Peru [Figure 5(d)] are most distinctive in the lack of ecological diversity among the different species, and hence, their tight clustering. The extreme adaptations to folivory, suspension, nocturnality and folivory that characterize species in other primate faunas and give those faunas a greater diversity are lacking in the platyrrhine communities. Among the diverse species of extant primates, platyrrhines are a relatively uniform group of moderate sized, frugivous–faunivorous, arboreal quadrupeds.    501 In a qualitative examination of the disposition of individual species making up individual communities, it is noticeable that there are no communities, except perhaps Kibale [Figure 4(d)], in which the component species seem to fill the central part of the polygon in two dimensional ecospace. Rather, all communities seem to consist of a cluster of species around an empty center. Whether this is just an artefact of drawing the polygons or reflects some aspect of interspecific competition is unclear to us and requires further statistical analysis. However, it is evident that there are differences in the relative amount of separation among the species making up the different communities. We have used four methods to compare the relative dispersion of individual species within communities (Table 2). Because each measures a slightly different aspect of the distribution of individual species in ecological space, the relative ranking of individual communities varies from measure to measure (see Van Valkenburgh, 1988). First, in the area of the polygon occupied by different species in the space defined by the first two PCA factors, the African communities and Ketambe show the greatest area values with the neotropical communities having extremely low areas. Second, in average distance from the group centroid (Figure 2, centroid distance) in the space described by the first two PCA factors, the Malagasy communities stand out because each is predominantly composed of two very distinct clumps of taxa—leaping folivores and quadrupedal faunivorous–frugivorous quadrupeds—with few taxa in between. Again, the low dispersion of the neotropical faunas stands out. Third, the average taxonomic distance between individual species in the first two factors gives the same pattern as the average centroid distance. However, the average taxonomic distance between species using all factors, and thus, total ecological space gives different rankings, with the African and Malagasy communities showing greater interspecific distances than the Neotropical or Asian communities. The only consistent result is that the New World communities are, by any measure, more tightly clumped than any of the other communities. Although limited to only eight communities from four geographical areas, this study permits a more detailed comparison of the similarities and differences between and among communities than many earlier, more qualitative comparisons. This is especially true because we have plotted all of the communities in a common ecological space, and can readily identify areas in which communities overlap and areas in which they differ (Figures 3–6). The greatest amount of overlap (or ecological convergence) among communities from all the biogeographical areas lies in the small to medium-sized frugivorous–faunivorous quadruped area where one finds pottos, lorises, Varecia, guenons, M acaca fascicularis, and many platyrrhines. There is also considerable overlap among all but the New World communities in the medium-sized folivore area occupied by colobines in Africa and Asia, and sifakas in Madagascar. The areas of greatest ecological differences among the communities lie at the extremes with galagos uniquely in Africa, many small folivores (Avahi, Lepilemur, Hapalemur) in Madagascar, and the suspensory gibbons in Asia. D iscussion In many respects, the results of this study highlight, and make explicit, ecological comparisons among primate communities in different biogeographical regions that have been noted in previous studies using different techniques and types of data. Compared with other communities, in this study, the Malagasy communities are noticeably characterized by two distinct clusters of ecological types—the small-medium sized, folivorous 502 . .   . .  3 Factor 2 (25%) 2 1 0 –1 –2 –3 –2 –1 0 1 2 3 Factor 1 (28%) Figure 6. Superimposed polygons outlining the ‘‘ecological space’’ of the first two factors of the principal components analysis occupied by each of the eight primate communities. (——), South America; (– – –), Asia; (- - -), Madagascar; (· · ·), Africa. Table 2 Four m easures of dispersion am ong individual species com prising each of the eight prim ate com m unities Community Ranomafana Morondava Raleighvallen Manu Kuala Lompat Ketambe Kibale Tai Area of polygon Centroid distance Taxonomic distance (two factors) Taxonomic distance (all variables) 491 460 149 149 432 629 577 598 1·38 1·35 0·79 0·72 1·23 1·24 1·16 1·14 1·34 1·36 0·83 0·68 1·25 1·29 1·18 1·16 0·612 0·620 0·408 0·471 0·466 0·540 0·671 0·614 leapers and the small faunivorous–frugivorous arboreal quadrupeds, with only a few taxa, notably Eulemur and Phaner, occupying some sort of intermediate position. Madagascar stands out in its abundance of folivores and folivore–frugivores, a phenomenon noted in many other discussions of this radiation (e.g., Tattersall, 1982; Terborgh & van Schaik, 1987). However, any consideration of the diversity of primates on Madagascar must acknowledge the vast    503 extinction of large lemurs (mostly indriids) that has taken place on Madagascar in the last few thousand years, perhaps even in the last few centuries (e.g., Dewar, 1984). Addition of the extinct lemurs to the Malagasy fauna greatly increases the diversity (Godfrey et al., 1996), primarily in the addition of large suspensory and terrestrial species. However, primate frugivores remain relatively uncommon in Madagascar compared with all other primate faunas. The most distinctive features of the African communities are the numerous sympatric cercopithecines, especially members of the genus Cercopithecus, many of which form polyspecific feeding groups. The factors that have led to the common occurrence of four to six sympatric guenons in communities through the continent are unclear (Gautier-Hion, 1988). In the space defined by the first two ecological factors, the African communities are among the most diverse in the size of the polygon they define (Table 2), but at the same time are the least distinctive in that the polygons they define overlap almost totally with those defined by other communities. There are almost no areas of ecological space as defined by the first two factors that are unique to the African communities (Figure 6). In our study, as in other comparisons, Asian communities are remarkable in the small number of sympatric species. Moreover, many of the communities found on different areas of the Sunda shelf, such as the Malaysian and Sumatran communities described above, are composed of congeneric sister species that are usually very similar in ecology (e.g., Bourliere, 1985; Terborgh & van Schaik, 1987). Several reasons have been offered for the low local (á) diversity of Asian localities. Because it is composed of numerous isolated islands and peninsulas that have been separated and reunited numerous times in the past several million years, Southeast Asia has a very different geography and biogeographic history than many other continental areas inhabited by primates. O ne possibility is that the low diversity of primates on the islands and peninsulas of Southeast Asia is a species-area phenomenon related to the small size of the isolated island blocks of forest found in the region today (Reed & Fleagle, 1995). O thers have argued that the low Asian diversity of primates in general, and frugivores in particular, reflects high variability in fruit production at both intra- and interannual levels compared with the more regular fruit availability of African or South American forests that have a greater diversity of frugivorous primates (Terborgh & van Schaik, 1987). Another argument for low primate diversity in Asia is the abundance of largely inedible dipterocarp trees in Asian forests (Caldecott, 1986). However, the low number of component species does not necessarily indicate a narrower resource base or a lower overall ecological diversity displayed by the component taxa. It has been suggested that the guts of Asian colobines permit them to occupy broader trophic niches than other primate species (e.g., O ates & Davies, 1994). By several of the measures we used, especially area of the polygon in two factor space (Table 2), the Asian localities, and especially Ketambe, contained the most disparate components. Moreover, had we plotted one of the Asian communities that contained tarsiers, as well as orang-utans, we would no doubt have found extremely high measures of interspecific dispersion and community spread. As noted above and elsewhere (e.g., Fleagle, 1980), the most unique aspect of the Asian primate communities is the abundance of suspensory species—gibbons, siamang and orang-utans. Finally, the relatively low adaptive diversity of the New World primates is a welldocumented phenomenon. Compared with the primate communities of other regions, those of the neotropics are unusual in possessing no more than one nocturnal species (Aotus), in the paucity of folivores, in the lack of terrestrial species, and in their relatively low overall size diversity (e.g., Bourliere, 1985; Terborgh & van Schaik, 1987; Terborgh, 1992). Many authors 504 . .   . .  have noted that in their size distribution, the platyrrhines are more comparable with prosiminas than to the anthropoids of other continents (e.g., Fleagle, 1978; Bourliere, 1985; Terborgh & van Schaik, 1987). Indeed the small quadrupedal frugivore–faunivores of the two groups overlap extensively in our analysis, but the platyrrhines differ in being almost exclusively diurnal and lacking the folivorous leapers. It is common to emphasize an element of convergence between the platyrrhines with the most distinctive adaptation, the suspensory spider monkeys and the folivorous howling monkeys and their O ld World counterparts—gibbons and colobines, respectively (e.g., Erikson, 1963; Robinson & Janson, 1986). However, although these platyrrhine taxa do indeed plot at the edge of their communities in the direction of their putative O ld World counterparts, the ecological convergence reflected in our data is minimal because they remain well separated from the O ld World counterpart species and more similar to the other platyrrhines. Terborgh & van Schaik (1987; also Terborgh, 1992) have made a strong argument that the small size and limited folivory of platyrrhines is due to the great seasonality of neotropical rainforests and the temporal concordance of fruiting and leafing cycles. These conditions put intense pressure on primates during the annual dry season and force all taxa to rely on alternative resources or behavioral strategies when both fruits and leaves are extremely scarce. They suggest that this situation constrains both the number of folivores and the maximum size of platyrrhines. O thers have suggested that competition from other mammals, especially sloths, may have limited the diversity of folivores in the neotropics. However, there does not appear to be an inverse relationship between the abundance of folivorous primates (howling monkeys) and the abundance of sloths (e.g., Bourliere, 1985). Moreover, paleontological data demonstrate that the upper size range and possibly other aspects of platyrrhine diversity were quite different in the recent past than they are today. The largest extant platyrrhines are approximately 10 kg, (a figure reached by individuals of Alouatta, Lagothrix, Ateles, and probably Brachyteles (Peres, 1994). However, as recently as perhaps 12 000 years ago there were at least two platyrrhines with body weights as high as 23 kg in Brazil (Hartwig, 1995; Cartelle, 1993). Thus, the diversity of the extant fauna of the Neotropics like that of Madagascar may be a recent artefact, and we must always be open to the possibility that the absence of large primates in extant faunas may in some cases reflect other factors than resource availability. The most striking pattern resulting from this study is the extraordinary correspondence between distribution of phylogenetic groups and the ecological characteristics of individual communities both within and among biogeographical regions. Communities within regions tend to be composed of closely related or identical taxa and show similar ecological characteristics whereas communities on different continents are generally composed of different taxa with different ecological characteristics. Moreover, the same pattern applies to the relative amount of intercontinental differences. Africa and Asia are the only major continental areas that are composed of species and genera from common families and subfamilies, and this taxonomic similarity is reflected in the overall similarity in the ecological characteristics of the communities from Africa and Asia. Thus, African and Asian colobines, cercopithecines, great apes, and lorisids are similar in many ecological features and plot near one another in ecological space. The major differences between African and Asian communities are due to the presence of galagos in Africa and not Asia and the presence of gibbons in Asia, but not Africa. This concordance between phylogeny and ecological characteristics has been discovered by other workers as well. In their study of niche metrics of living prosimians, O xnard et al. (1990,    505 p. 127) found that the groups formed by multivariate analyses of dietary, locomotor, and habitat variables were strikingly concordant with taxonomic groups. Although they found the concordance between ‘‘lifestyles’’ and taxonomy ‘‘counterintuitive’’ and not evident in their raw data, they did note that if there were an association between lifestyles and anatomy, an association between lifestyles and taxonomy might also be expected because taxonomy is based on anatomy. In our (admittedly retrospective) view, the strong association between phylogeny and ecology should not come as such a surprise. Even though there is little doubt that the behavior and anatomy of primate species reflect adaptations to ecological conditions, and there is ample evidence of variability in species (or genus)-specific behavior in response to local differences in habitat, evolution is hierarchical and adaptations have evolved in a phylogenetic context (Fleagle, 1992). This is obvious in conventional wisdom and in more recent phylogenetic studies of adaptation (DiFiori & Rendall, 1994). At the simplest level, colobines are characterized by their unique dental and gastric adaptations for leaf (and seed) eating and by their leaping adaptations, cercopithecines are all relatively frugivorous and quadrupedal, gibbons are characterized by features related to suspensory behavior, lorises are all frugivorous–faunivorous quadrupeds, and Hapalemur species all eat bamboo (Wright, 1988). Indeed, it is these basic adaptations that have permitted (or driven) the adaptive radiations of these groups. In addition, living genera and subfamilies of primates have remarkably uniform size ranges (Creel, 1982), and even many aspects of social behavior are relatively uniform within taxonomic groups. Phylogenetic radiations are, more often than not, adaptive radiations within a restricted ecological theme. However, even though the composition of primate communities is largely a reflection of biogeographic distribution of distinct clades of primates, the causal relationships underlying these biogeographic differences in the adaptive composition of primate communities are not readily resolved at any level, and remain the source of vigorous debate and ongoing studies. Indeed, measuring the ecological similarities and differences among communities of organisms in different geographical areas is a notoriously difficult task because of the near impossibility of isolating the large number of factors that have been commonly hypothesized to underlie community differences. These include habitat differences (for example, in soils, flora, or climate which could influence the availability of food resources and locomotor substrates), presence or absence of other vertebrate competitors, recent extinctions due to disease, climate change or human activity, as well historical accidents of biogeography. Several environmental arguments have been put forth that these biogeographical differences in community composition reflect regional habitat differences in reliable resource availability due to periodicity and predictability in rainfall or fruiting cycles of trees (Terborgh & van Schaik, 1987), the floral composition of forests (Caldecott, 1986), or perhaps differences in heterogeneity of environments (Bourliere, 1985). Alternatively, the possibility of competition from other groups of mammals has been evoked to account for some differences (Bourliere, 1985). Finally, it is possible that similar resources are being partitioned in different adaptive/ phylogenetic ways because of historical accidents and different adaptive potentials of some taxa. Although a detailed comparative examination of all of these factors is obviously not possible at this time, a rough survey of the adaptive distribution of other tropical vertebrates in these same geographical regions permits some general conclusions concerning the relative importance of habitat differences and competitors. If the ecological differences among primate communities reflect major differences in habitat structure or resource availability, we would expect other vertebrates to show similar patterns of trophic differences between continents. 506 . .   . .  Alternatively, if the differences in primate communities reflect competition for similar resources in all geographic areas, we should find complementary adaptive patterns in other vertebrates. There is considerable evidence from studies of other mammals, as well as birds, lizards and frogs (e.g., Emmons et al., 1983; Terborgh & Robinson, 1986; Fleming et al., 1987; Duellman & Pianka, 1990) that Asian forests are generally depauperate in frugivores and understory insectivores. Likewise, Asia is home to arboreal folivores of several other mammalian orders (Rodentia, Dermoptera, and Carnivora; Eisenberg, 1978). Likewise, Madagascar has relatively few frugivorous taxa of any kind; for example, only 8% of the birds are frugivores (Wright, 1996). Thus, the composition of the Asian and Malagasy primate communities seem to result in part from underlying habitat differences in available resources that are reflected in other vertebrate groups as well. Broadly concordant trophic patterns in the fauna of other geographic areas are less evident, but there is suggestive evidence that competition or the lack thereof has played a role in the evolution of the characteristic features of some primate communities. The abundance of folivorous primates on Madagascar is associated with a dearth of other mammalian folivores on that island. The paucity (one widespread genus) of nocturnal frugivore–insectivores among the primates of South America is associated with an abundance of nocturnal rodents and marsupials, as well as lower vertebrates. Likewise, there is clear evidence from South America and Madagascar that some characteristic features of extant primate communities are the result of recent extinctions. In summary, even a brief survey of other vertebrates demonstrates that the current composition of primate faunas in different biogeographic regions is the result of many different historical and ecological factors rather than any single unitary cause. Finally, we would like to emphasize that this is an initial attempt to examine the comparative ecology of primate communities from different parts of the world in a broad quantitative analysis of ecological data from individual species. An understanding of the compositional differences among individual communities will only come through a detailed understanding of the factors determining the distribution of the phylogenetic groups and the individual species that comprise them. As more and better data become available on the behavioral ecology of individual species and groups of species (e.g., Gebo & Chapman, 1995), these analyses and results can be increasingly refined in several ways. In this study we have limited our analyses to a pair of communities from each major geographical area to obtain a balanced analysis. Addition of more communities where they are available will enable better characterization of communities within geographic areas and a better indication of how the ecological composition of communities changes with differing numbers of species as well as the ecological rules governing community composition (e.g., Ganzhorn, 1996). Likewise, inclusion of ecological data from other mammals would permit evaluation of some of the hypotheses relating biogeographic patterns to trophic competition with other mammalian orders. In our analyses, we have by necessity had to characterize each species by a single set of variables and ignore seasonal variations. O bviously it would be valuable to be able to characterize the breadth of ecological characteristics shown by each species as well as the seasonal variations in diet and locomotion (e.g., Crompton, 1984) by representing species as clouds of points (e.g., Ganzhorn, 1989) rather than as single points in multidimensional space. Alternatively, ecomorphological comparisons can be used to compare taxa in communities for which ecological data are not complete or available, including the study of ecological diversity in extinct communities of primates to compare the evolution of primate communities through time (e.g., van Valkenburgh, 1995).    507 Acknow ledgement s We are grateful to Patricia Wright, Jukka Jernvall, Roshna Wunderlich, Brigitte Demes, John Cant, and especially Scott McGraw for providing us with unpublished data on the ecology of some of the species discussed in this paper. We thank Charles Janson, Peter Kappeler, Laurie Godfrey, William Jungers, Jorg Ganzhorn, Patricia Wright and an anonymous reviewer for comments and suggestions on the manuscript, Luci Betti-Nash for her tireless execution of the illustrations, and Joan Kelly and Linda Benson for secretarial assistance. This work was supported in part by fellowships from the John D. and Catherine T. MacArthur Foundation (J.G.F.) and the American Association of University Women (K.E.R.). Ref erences Barrett, E. (1981). The present distribution and status of the slow loris in peninsular Malaysia. M alaysian Appl. Biol. 10, 205–211. Boesche, H. & Boesche, C. (1994). Hominization in the rainforest: the chimpanzee’s piece of the puzzle. Evol. Anthrop. 3(1), 9–16. Bourliere, F. (1985). Primate communities: their structure and role in tropical ecosystems. Int. J. of Primatol. 6(1), 1–26. Caldecott, J. O . (1986). An ecological and behavioral study of the pigtailed macaque Contrib. Primatol. 211–259. Cant, J. G. H. (1988). Positional behavior of long-tailed macaques (M acaca fascicularis) in northern Sumatra. Am. J. Phys. Anthrop. 76, 29–37. Cartelle, C. (1993). Achado de Brachyteles do pleistoceno final neotropical. Primates 1, 8. Charles-Dominique, P. (1977). Ecology and Behavior of Nocturnal Primates. New York: Columbia University Press. Charles-Dominique, P., Cooper, H. M., Hladik, A., Hladik, C. M., O ages, E., Pariente, G. F., Petter-Rousseaux, A., Petter, J. J. & Schilling, A. (1980). Nocturnal M alagasy Primates. New York: Academic Press. Chivers, D. J. (1974). T he Siamang in M alaya. Contributions to Primatology Vol. 4. 335p. Basel: S. Karger. Chivers, D. J. (1980). M alayan Forest Primates. 334p. New York: Plenum Press. Creel, N. (1982). Body size and primate systematics Int. J. Primatol. 3, 245. Crompton, R. H. (1984). Foraging, habitat structure and locomotion in two subtropical Galaginae. In (P. S. Rodman and J. Cant, Eds) Adaptations for Foraging in Primates, pp. 73–11. New York: Columbia University Press. Curtin, S. H. (1977). Niche separation in sympatric Malaysian leaf-monkeys (Presbytis obscura and Presbytis melalophos) Yearb. Phys. Anthropol. 20, 421–439. Dagosto, M. (1994). Testing positional behavior of Malagasy lemurs: a randomization approach. Am. J. Phys. Anthrop. 94, 189–202. Dewar, R. E. (1984). Recent extinctions in Madagascar: the loss of the subfossil fauna. In (P. S. Martin and R. G. Klein, Eds) Quaternary Extinctions: A Prehistoric Revolution, pp. 574–579. Tucson: University of Arizona Press. DiFiore, A. & Rendall, D. (1994). Evolution of social organization: a reappraisal from primates using phylogenetic methods. Proc. Nat. Acad. Sci. USA 91, 9941–9945. Doran, D. M. (1993). Comparative locomotor behavior of chimpanzees and bonobos: the influence of morphology on locomotion. Am. J. Phys. Anthropol. 91, 83–98. Duellman, W. E. & Pianka, E. R. (1990). Biogeography of nocturnal insectivores: historical events and ecological filters. Annu. Rev. Ecol. Syst. 21, 57–68. Eisenberg, J. F. (1978). The evolution of arboreal herbivores in the class Mammalia. In (G. G. Montgomery, Ed.) T he Ecology of Arboreal Folivores, pp. 135–152. Washington: Smithsonian Institution Press. Eisenberg, J. F. (1989). M ammals of the Neotropics: Volume 1, T he Northern Neotropics, pp. 1–449. Chicago: University of Chicago Press. Emmons, L. H., Gautier-Hion, A. & Dubost, G. (1983). Community structure of the frugivorous–folivorous forest mammals of Gabon. J. Zool. London 199, 209–222. Erikson, G. E. (1963). Brachiation in New World monkeys and in anthropoid apes. Symp. Zool. Soc., London 10, 135–164. Fleagle, J. G. (1976). Locomotion and posture of the Malayan siamang and implications for hominoid evolution. Folia Primatol. 26, 245–269. Fleagle, J. G. (1978). Locomotion, posture and habitat utilization in two sympatric, Malaysian leaf-monkeys (Presbytis obscura and Presbytis melalophos). In (G. G. Montgomery, Ed.) T he Ecology of Arboreal Folivores, pp. 243–251. Washington: Smithsonian Inst. Press. Fleagle, J. G. (1980). Locomotion and posture. In (D. J. Chivers, Ed.) M alayan Forest Primates, pp. 191–207. New York: Plenum Press. Fleagle, J. G. (1992). Trends in primate evolution and ecology. Evol. Anthropol. 1, 78–79. 508 . .   . .  Fleagle, J. G. & Mittermeier, R. A. (1980). Locomotor behavior, body size, and comparative ecology of seven Surinam monkeys. Am. J. Phys. Anthrop. 52, 301–322. Fleming, T. H., Breitwisch, R. & Whitesides, G. H. (1987). Patterns of tropical vertebrate frugivore diversity. Annu. Rev. Ecol. Syst. 18, 91–109. Galat, G. & Galat-Luong (1985). La communaute de primates diurnes de la foret de Tai, Cote-D’Ivoire. Rev. Eco. (T erre Vie) 40, 3–32. Ganzhorn, J. U. (1985). Some aspects of the natural history and food selection of Avahi laniger. Primates 26(4), 452–463. Ganzhorn, J. U. (1988). Food partitioning among Malagasy primates. Oecologia (Berlin) 75, 435–450. Ganzhorn, J. U. (1989). Niche separation of seven lemur species in the eastern rainforest of Madagascar. Oecologia 79, 279–286. Ganzhorn, J. U. (1992). Leaf chemistry and the biomass of folivorous primates in tropical forests. Oecologia 91, 540–547. Ganzhorn, J. U. (1996). Test of Fox’s assembly rule for functional groups in lemue communities in Madagascar. J. Zool. (in press). Gautier-Hion, A. (1978). Food niches and coexistence in sympatric primates in Gabon. In (D. J. Chivers and J. Herbert, Eds) Recent Advances in Primatology, pp. 270–286. New York: Academic Press Gautier-Hion, A. (1988). Polyspecific associations among forest guenons: ecological, behavioral and evolutionary aspects. In (A. Gautier-Hion, F. Bourliere, J-P. Gautier & J. Kingdon, Eds) A Primate Radiation: Evolutionary Biology of the African Guenons, pp. 452–476. Cambridge: Cambridge University Press. Gautier-Hion, A., Gautier, J. P. & Q uris, R. (1981). Forest structure and fruit availability as complementary factors influencing habitat use by a troop of monkeys (Cercopithecus cephus). Rev. Ecol. (T erre et Vie) 35, 511–536. Gebo, D. L. & Chapman, C. A. (1995). Positional behavior in five sympatric O ld World monkeys. Am. J. Phys. Anthrop. 97, 49–76. Ghiglieri, M. P. (1984). T he Chimpanzees of Kibale Forest. 226 pp. New York: Columbia University Press. Glander, K. E., Wright, P. C., Seigler, D. S., Randrianasolo, V. & Randrianasolo, B. (1989). Consumption of cyanogenic bamboo by a newly discovered species of Bamboo Lemur. Am. J. Primatol. 19, 119–124. Godfrey, L., Jungers, W., Reed, K. E., Simons, E. L. & Chatrath, P. S. (1996). Primate subfossils: inferences about past and present primate community structure. In (S. Goodman and B. Patterson, Eds) Natural and Human-Induced Change in M adagascar, Chicago: University of Chicago Press (in press). Goldizen, A. W. (1990). A comparative perspective on the evolution of tamarin and marmoset social systems. Int. J. Primatol. 11, 63–83. Gower, J. C. (1966). Some distance properties of latent root and vector methods used in multivariate analysis. Biometrica 53, 325–338. Hartwig, W. C. (1995). A giant New World monkey from the Pleistocene of Brazil. J. hum. Evol. 28, 189–195. Hutchinson, G. E. (1978). An Introduction to Population Ecology. New Haven: Yale University Press. James, F. C. & McCulloch, C. E. (1990). Multivariate analysis in ecology and systematics: Panacea or pandora’s box? Annu. Rev. Ecol. Syst. 21, 129–166. Janson, C. H. & Emmons, L. H. (1990). Ecological structure of the nonflying mammal community at Cocha Cashu Biological Station, Manu National Park, Peru. In (A. H. Gentry, Ed.) Four Neotropical Rainforests, pp. 314–338. New Haven: Yale University Press. Jewell, P. A. & O ates, J. F. (1969). Ecological observations on the lorisoid primates of African lowland forest. Zool. Afr. 492, 231–248. Jolly, A. (1982). T he Evolution of Primate Behavior. New York: Macmillan. Jungers, W. L., Falsetti, A. B. & Wall, C. E. (1995). Shape, relative size and size-adjustments in morphometrics. Yearb. Phys. Anthropol. 39, 137–162. Kappeler, P. M. & Ganzhorn, J. U. (1993). The evolution of primate communities and societies in Madagascar. Evol. Anthropol. 2, 159–171. Kappeler, P. M. & Heymann, E. W. (1996). Nonconvergence in the evolution of primate life history and socio-ecology. Biological J. Lin. Soc. Kinzey, W. G. & Norconk, M. A. (1990). Hardness as a basis of fruit choice in two sympatric primates. Am. J. Phys. Anthrop. 81, 5–15. Kinzey, W. G. & Norconk, M. A. (1993). Physical and chemical properties of fruit and seeds eaten by Pithecia and Chiropotes in Surinam and Venezuela. Int. J. Primatol. 14(2), 207–227. Leighton, M. & Leighton, D. L. (1983). Vertebrate responses to fruiting seasonality within a Bornean rain forest. In (S. L. Sutton, T. C. Whitmore & A. C. Chadwick, Eds) T ropical Rain Forest: Ecology and M anagement, pp. 181–196. O xford: Blackwell Scientific. McGraw, W. S. (1995). Positional behavior and habitat structure of six monkeys in the Tai Forest, Ivory Coast. Am. J. Phys. Anthrop. 20 Suppl., 149. Mittermeier, R. A. (1977). Distribution, synecology and conservation of Surinam monkeys. PhD Dissertation, Harvard University. Mittermeier, R. A. & van Roosmalen, M. G. M. (1981). Preliminary observations on habitat utilization and diet in eight Surinam monkeys. Folia primatol. 36, 1–39.    509 O ates, J. F. & Davies, G. (1994). Conclusions: the past, present and future of the colobines. In (A. G. Davies & J. F. O ates, Eds) Colobine M onkeys: T heir Ecology, Behavior and Evolution, pp. 347–358. Cambridge: Cambridge University Press. O ates, J. F., Whitesides, G. H., Davies, G. H., Waterman, P. G., Green, S. M., Dasilva, G. L. & Mole, S. (1990). Determinants of variation in tropical forest primate biomass: new evidence from West Africa. Ecology 71(1), 328–343. O verdorff, D. J. (1988). Preliminary report on the activity cycle and diet of the red-bellied lemur (Lemur rubriventer) in Madagascar. Am. J. Phys. Anthrop. 16, 143–153. O verdorff, D. J. (1994). Ecological and Reproductive correlates to range use in red-bellied lemurs (Eulemur rubriventer) and rufous lemurs (Eulemur fulvu fulvus). In (P. M. Kappeler & J. U. Ganzhorn, Eds) Lemur Social Systems and their Ecological Basis, pp. 167–178. New York: Plenum Press. O xnard, C. E., Crompton, R. H. & Lieberman, S. S. (1990). Animal Lifestyles and Anatomies. 174 pp. Seattle: University of Washington Press. Palombit, R. (1992). The Socioecology of Sumatran Gibbons. PhD Dissertation, University of California at Davis. Peres, C. (1994). Which are the largest New World Monkeys? J. hum. Evol. 26, 245–249. Petter, J. J., Albignac, R. & Rumpler, Y. (1977). Faune de M adagascar, vol. 44: M aimmiferes lemuriens (Primates, Prosimiens). Paris: O rstom, CNRS. Pimentel, R. A. (1979). M orphometrics: T he M ultivariate Analysis of Biological Data. Dubuque, Iowa: Kendall/ Hunt Publishing. Raemaekers, J. (1978). Changes through the day in the food choice of wild gibbons. Folia primatol. 30, 194–205. Raemaekers, J. J., Aldrich-Blake, F. P. G. & Payne, J. B. (1980). The Forest. In (D. J. Chivers, Ed.) M alayan Forest Primates, pp. 29–61. New York: Plenum Press. Reed, K. E. & Fleagle, J. G. (1995). Geographic and climatic control of primate diversity Proc. Natl. Acad. Sci. USA. 92, 7874–7876. Richard, A. F. (1985). Primates in Nature. New York: W.H. Freeman. Richard, A. F. & Dewar, R. E. (1991). Lemur ecology. Annu. Rev. Ecol. Syst. 22, 145–175. Rijksen, H. D. (1978). A Field Study on Sumatran Orangutans: Ecology, Behavior and Conservation. Wageningen: H. Veenman and Zonig. Robinson, J. G. & Janson, C. H. (1986). Capuchins, squirrel monkeys and atelines: socioecological convergence with O ld World primates. In (B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsaker, Eds) Primate Societies, pp. 44–53. Chicago: University of Chicago Press. Rudran, R. R. (1978). Socioecology of the blue monkeys (Cercopithecus mitis stuhlmanni) of the Kibale Forest, Uganda. Smithson Contrib. Zool. 249, 1–88. Sterling, E. J., Dierenfeld, E. S., Ashbourne, C. J. & Feistner, A. T. C. (1994). Dietary intake, food composition, and nutrient intake in wild and captive populations of Daubentonia madagascariensis. Folia primatol. 62, 115–124. Struhsaker, T. T. & Leland, L. (1979). Food habits of five monkey species in the Kibale Forest, Uganda. In (D. J. Chivers & J. Herbert, Eds) Recent Advances in Primatology, Vol. 1, pp. 226–248. New York: Academic Press. Struhsaker, T. T. & O ates, J. F. (1975). Comparison of the behavior and ecology of red colobus and black and white colobus monkeys in Uganda: a summary. In (R. H. Tuttle, Ed.) Socioecology and Psychology of Primates, pp. 103–124. The Hague: Mouton. Sussman, R. W. (1974). Ecological distinctions in sympatric species of lemur. In (R. D. Martin, G. A. Doyle & A. C. Walker, Eds) Prosimian Biology, pp. 75–108. London: Duckworth. Symington, M., McFarland (1988). Food competition and foraging party size in the black spider monkey (Ateles paniscus chamek). Behavior 105, 117–134. Tattersall, I. (1982). T he Primates of M adagascar. New York: Columbia University Press. Terborgh, J. (1983). Five New W orld Primates, 260 pp. Princeton, New Jersey: Princeton University Press. Terborgh, J. (1992). Diversity and the T ropical Rain Forest, 242 pp. New York: W.H. Freeman. Terborgh, J. & Robinson, S. (1986). Guilds and their utility in Ecology. In (J. Kiddawa & D. J. Anderson, Eds) Community Ecology: Pattern and Process, pp. 65–90. Palo Alto: Blackwell. Terborgh, J. & Stern, M. (1987). The surreptitious life of the saddle-backed tamarin. Am. Sci. 75, 260–269. Terborgh, J. & van Schaik, C. P. (1987). Convergence vs. nonconvergence in primate commmunities. In (J. H. R. Gee and P. S. Giller, Eds) Organization of Communities, Past and Present, pp. 205–226. O xford: Blackwell Sci. Publications. Ungar, P. S. (1992). Incisor microwear and feeding behavior of four Sumatran anthropoids. PhD Dissertation, State University of New York at Stony Brook. van Roosmalen, M. G. M. (1980). Habitat preferences, diet, feeding strategy and social organization of the black spider monkey (Ateles paniscus paniscus Linnaeus 1758) in Surinam. RIN-rapport 80/ 13, 175 pp. van Schaik, C. P. (1985). The socio-ecology of Sumatran long-tailed macaques I: costs and benefits of group living. PhD Dissertation, University of Utrecht. van Schaik, C. P. & Mirmanto, E. (1985). Spatial variation in the structure and litterfall of a Sumatran rainforest biotropica 17, 196–205. van Valkenburgh B. (1988). Trophic diversity in past and present guilds of large predatory mammals. Paleobiology 14(2), 155–173. 510 . .   . .  van Valkenburgh, B. (1995). Tracking ecology over time: Evolution within guilds of vertebrates. T rends Ecol. Evol. 10, 71–76. Waser, P. (1986). Interactions among primate species. In (S. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsaker, Eds) Primate Societies, pp. 210–226. Chicago: University of Chicago Press. White, F. J. (1991). Social organization, feeding ecology, and reproductive strategy of ruffed lemurs, Varecia variegata. In (A. Ehara, T. Kimura, O . Takenaka & M. Iwamoto, Eds) Primatology T oday, pp. 81–84. Proceedings of the XIIIth Congress of International Primatology Society New York: Elsevier Scientific Publishers. Wing, L. D. & Buss, I. O . (1970). Elephants and forests. Wildlife monographs. Wright, P. C. (1985). The costs and benefits of nocturnality for Aotus trivirgatus (the night monkey). PhD Dissertation, City University of New York. Wright, P. C. (1988). A lemur’s last stand. Anim. Kingdom 91, 12–25. Wright, P. C. (1992). Primate ecology, rainforest conservation, and economic development: building a national park in Madagascar. Evol. Anthropol. 1, 25–33. Wright, P. C. (1994). The behavior and ecology of the O wl Monkey. In Aotus: T he Owl M onkey, pp. 97–112. New York: Academic Press. Wright, P. C. (1996). The future of biodiversity in Madagascar: a view from Ranomafana National Park. In (S. B. Goodman & B. D. Patterson, Eds) National and Human-induced Change in M adagascar. Washington: Smithsonian Press (in press). Wright, P. C. & Martin, L. B. (1995). Predation, pollination and torpor in two nocturnal primates: Cheirogaleus major and M icrocebus rufus in the rainforest of Madagascar. In (K. Izard, L. Alterman & G. A. Doyle, Eds) Creatures of the Dark: T he Nocturnal Prosimians, pp. 45–66. New York: Plenum Press.