Comparative locomotor behavior of chimpanzees and bonobos The influence of morphology on locomotion.код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 91 :83-98 (1993) Comparative Locomotor Behavior of Chimpanzees and Bonobos: The influence of Morphology on Locomotion DIANE M. DORAN Department of Anatomical Sciences, SUNY at Stony Brook, Stony Brook, New York 117948081 KEY WORDS Pygmy chimpanzees, Positional behavior, Pan paniscus, Pan troglodytes ABSTRACT Results from a 10 month study of adult male and female bonobos (Pan paniscus) in the Lomako Forest, Zaire, and those from a 7 month study of adult male and female chimpanzees in the Tai Forest, Ivory Coast (Pan troglodytes verus), were compared in order to determine whether there are species differences in locomotor behavior and substrate use and, if so, whether these differences support predictions made on the basis of interspecific morphological differences. Results indicate that bonobos are more arboreal than chimpanzees and that male bonobos are more suspensory than their chimpanzee counterpart. This would be predicted on the basis of male bonobo’s longer and more narrow scapula. This particular finding is contrary to the prediction that the bonobo is a “scaled reduced version of a chimpanzee” with little or no positional behavior difference as had been suggested. This study provides the behavioral data necessary to untangle contradictory interpretations of the morphological differences between chimpanzees and bonobos, and raises a previously discussed (Fleagle: Size and Scaling in Primate Biology, pp. 1-19, 1985) but frequently overlooked point-that isometry in allometric studies does not necessarily equate with behavioral equivalence. Several researchers have demonstrated that bonobos and chimpanzees follow the same scaling trends for many features, and are in some sense functionally equivalent, since they manage to feed and reproduce. However, as reflected in their morphologies, they do so through different types and frequencies of locomotor behaviors. o 1993 Wiley-Liss, Inc. Since the discovery of the bonobo (Panpaniscus; Schwarz, 1929), there has been considerable debate as to the nature and meaning of the morphological differences between it and the chimpanzee (P. troglodytes). This paper assesses how these morphological differences are correlated with differences in behavior, by reviewing the morphological differences between the species, considering several proposed explanations for these differences, and finally, examining whether these behavioral hypotheses can be supported by providing results from a comparative study of bonobo and chimpanzee locomotor behavior and substrate use. 0 1993 WILEY-LISS. INC INTERSPECIFIC MORPHOLOGICAL DIFFERENCES The degree of size difference between bonobos and chimpanzees has received particular attention. Coolidge (1933) described a small specimen and estimated that bonobos were one-half the size of chimpanzees. Ironically, as museum collections grew, it became apparent that the early specimen described by Coolidge (1933) was the smallest of all specimens collected. Received July 13, 1992; accepted October 30, 1992 84 D.M. DORAN More recently, with further study of additional specimens, some researchers have argued that the purported size difference is small or nonexistent (Schultz, 1969; Zihlman and Cramer, 1978; Horn, 1979; McHenry and Corruccini, 1981; Susman and Jungers, 19811, whereas others continue to see the bonobo as a smaller, more gracile animal (Pilbeam and Gould, 1974; Cousins, 1978; Johnson, 1981; Shea, 1983,1984). Much of this confusion has resulted from the failure t o distinguish among the three geographically distinct chimpanzee subspecies which, in themselves differ significantly in body size (Jungers and Susman, 1984). In comparison with the three subspecies of chimpanzees, the bonobo does not differ significantly in body weight from P. t. schweinfurthii, but is significantly lighter than P. t. troglodytes (Jungers and Susman, 1984). The sole available body weight for the third subspecies of chimpanzee (P. t. uerus) places it closer in size to P. t. schweiiafurthii than to P. t. troglodytes (Jungers and Susman, 1984). Beyond differences in body weight, bonobos and chimpanzees differ with respect to skull shape and relative size (Coolidge, 1933; Weidenreich, 1941; Fenart and Deblock, 1973; Cramer, 1977; Shea, 19821, teeth (Kmzey, 1971,1984; Johanson, 19741, and postcrania. Of particular interest with regard to potential differences in locomotor behavior are the postcranial differences. These bonobo differences include: a significantly lower intermembral index (Susman, 1979); a more gracile upper body with a shorter clavicle (Zihlman and Cramer, 19781, reduced chest girth (Coolidge and Shea, 19821, and, at least in males, a narrower scapula (Coolidge, 1933; Roberts, 1974; Jungers and Susman, 1984; Shea, 1986); a smaller pelvis (Zihlman and Cramer, 1978), with a shorter ilium than that of P. t. uerus or P. t. troglodytes, and a shorter pubis than any of the three subspecies of chimpanzees (Jungers and Susman, 1984); and hands with more curved proximal and middle phalanges, and metacarpals that display less well developed dorsal ridges and thicker cortices in their metacarpals and phalanges (Susman, 1979). Finally, in marked contrast to chimpanzees, bonobos are characterized by minimal sex- ual dimorphism in their postcranial skeletons, in spite of significant sexual dimorphism in body weight (Zihlman and Cramer, 1978; Jungers and Susman, 1984). In assessing how many of the morphological distinctions between the two species are related to body size differences, Shea (1984, 1986) and Jungers and Susman (1984) find, based on studies of ontogenetic allometry, that with regard to chest girth, arm span, scapula size and shape, and intralimb proportions of the two species, the ontogenetic scaling trends exhibit considerable overlap. As Shea (1984) notes, “the proportions observed in adult common chimpanzees are those predicted if the growth patterns of the pygmy chimpanzee are simply extended to larger terminal size.” This is reminiscent of Coolidge’s claim (1933, 1984) that the bonobo is a “true paedomorph,” that is, a n animal which retains juvenile characters in the adult form. PROPOSED INTERSPECIFIC BEHAVIORAL DIFFERENCES Several hypotheses, including the “suspensory” (Susman, 19791, the “best prototype of prehominid ancestors” (Zihlman et al., 19781, the “scaled reduced common chimpanzee” (McHenry and Corruccini, 19811, and the “true paedomorph (Coolidge, 1933) have been proposed to account for these postcranial morphological differences. Implicit in these hypotheses is the existence of behavioral differences between the two species which are associated with their morphological differences. The “suspensory” hypothesis suggests that these differences are specializations resulting from the bonobos’ isolation and adaptation to forest dwelling and arboreal existence (MacKinnon, 1978; Susman, 1979; Horn, 1979; Johnson, 1981; Latimer et al., 1981). Among the morphological evidence cited to support this is the bonobos’ possibly smaller body size, reduced postcranial sexual dimorphism, long and narrow scapula (in males), and curved phalanges (Coolidge, 1933; Frechkop, 1935; Roberts, 1974; Susman, 1987). Behavioral predictions include the bonobos more frequent use of arboreal behaviors, particularly more armswinging, in comparison with chimpanzees. LOCOMOTOR BEHAVIOR OF CHIMPANZEES AND BONOBOS A second, and mutually exclusive, explanation is that the bonobo is a “scaled reduced version” of a chimpanzee (Corruccini and McHenry, 1979; McHenry and Corruccini, 1981) with morphological differences between the two species due simply to body size differences and the effects of allometry. Behavioral predictions for this hypothesis are that there are no positional behavior differences. In fact, McHenry and Corruccini (1981) state that “in most respects the postcranium of Pan paniscus is an allometrically scaled version of P. troglodytes and they both retain essentially the same locomotor and postural adaptations.” It is known that animals with different body sizes face different constraints within the same habitat, and as a result are characterized by different locomotor and postural profiles (Cartmill and Milton, 1977; Fleagle and Mittermeier, 1980; Fleagle, 1985; Cant, 1989; Doran, 1992a, in press). In a given arboreal habitat, larger animals leap less and climb, bridge and use suspensory behavior more frequently than smaller animals. Therefore, in this “scaled down chimpanzee” hypothesis, the difference seen in the behavior of bonobos and chimpanzees should not be any greater than that between chimpanzees with a comparable body size difference. Finally, Coolidge (1933, 1984) noted that the adult bonobo is a “true paedomorph,” because some of its adult proportions and features are present in juvenile chimpanzees. Most important of these features in regards to predictable locomotor differences is the long and narrow scapula. Historically, a functional correlation has been drawn between narrow scapulae and arm-swinging locomotion (Coolidge, 1933; Susman et al., 1980). This has frequently been cited to predict that adult bonobos are more suspensory than their chimpanzee counterpart (Coolidge, 1933; Frechkop, 1935; Roberts, 1974; Susman et al., 1980). Shea (1984, 1986) and Jungers and Susman (1984) have demonstrated that the bonobo scapula scales ontogenetically with that of chimpanzees. Thus the same morphological difference that distinguishes bonobos from chimpanzees also serves to distinguish immature and mature chimpan- 85 zees (and bonobos). Since ontogenetic changes occur in primate locomotion (i.e., Goodall, 1968; Rose, 1977; Crompton, 1983; Doran, 1992a), one could predict that bonobos are more similar to immature chimpanzees in their behavior than they are to adult chimpanzees. Since immature chimpanzees are more suspensory than mature chimpanzees (Doran, 1992a), one could also predict that adult bonobos are more suspensory than adult chimpanzees. Predicted locomotor differences cannot be presently supported or rejected by existing locomotor data (but see Susman, 1984; Hunt, 1992). In order to accurately assess differences in arboreal behavior it is essential that bonobos and chimpanzees be studied in as similar habitat (i.e., lowland rain forest) as possible. This paper presents new data on the locomotor behavior and substrate use of bonobos in the Lomako Forest, Zaire and compares it to recently published data on chimpanzees from the Tai Forest, Ivory Coast (Doran, in press) in order to test whether there are predictable differences in the locomotor behavior of forest dwelling bonobos and chimpanzees. Specific questions to be addressed include- 1. Are bonobos more arboreal than chimpanzees, and if so, in what way? Do they engage in more frequent arm-swinging than chimpanzees? 2. If there are locomotor differences between the two species, do they follow predictions based on body size differences alone, or are there species differences in behavior that cannot be explained by a possible body size difference? 3. Are bonobos more similar to immature rather than mature chimpanzees in their locomotor behavior? METHODS Study site and sampling methods: Pan paniscus Pan paniscus was observed for a total of 220 hours during 10 consecutive months, from December 1986 through October 1987, in the Lomako Forest, Zaire. For a complete description of the study site see Badrian and 86 D.M. DORAN Badrian (1977), Badrian and Malenky (1984), and White (1986). Results of locomotor behavior and substrate use are based on continuous focal animal locomotor bout sampling and include the proportion of all bouts spent in each locomotor activity (% bouts-no distance), the mean distance travelled per bout, and the percent of bouts weighted by the distance travelled per bout (% of bouts with distance). The most accurate way to interpret locomotor bout data is from the percent of bouts (with distance) (Doran, 1992b), and therefore all analyses of bonobos in this study are based on bouts (with distance). The percentage of bouts (no distance) are included for comparison with previous studies. On each occasion that bonobos were contacted, they were followed for as long as possible. Arboreal andlor terrestrial travel was recorded in 72.1% of all sightings (n = 111). In 27.9% of all sightings, the animals were lost immediately upon their first descent to the ground, and thus, no travel was recorded. During follows of bonobos, 45% of all sightings with travel (n = 80) include sightings with terrestrial travel. I used two measures of terrestrial travel. Direct ground travel, which accounts for 14.3% of total ground distance followed, is when the chimpanzees are visible during travel, although often this includes only glimpses of moving animals. Indirect travel, which accounts for 85.7% of all ground follows, is when visual contact is interrupted during the follow, and resumed within 20 minutes. If the exact route the chimpanzee took was not known, the shortest distance between the starting and ending points was measured. Qualitative intersite comparisons of Pan paniscus are based on observations and discussions with Drs. Kuroda and Furuichi during a one week visit to the Wamba (Zaire) study site in October 1987. For a description of the study site see Kano (1980, 1992), Kuroda (1980) and Furuichi (1987). Study site and sampling methods: Pan troglodytes verus One nonprovisioned, habituated community of Pan troglodytes uerus was observed for 430 hours during seven consecutive months from March through September 1988 in the Tai National Park, Ivory Coast. The Tai Forest was considered the most appropriate study site for comparison with bonobos becuase it is a lowland rain forest which shares many tree species in common with the Lomako Forest. Since a primary consideration of this study is to determine whether there are differences in how arboreal the two species are, it was considered essential that they be studied in environments that are as similar a s possible. Since bonobos are restricted to lowland rain forest, it was determined that chimpanzees should be studied in a lowland rain forest. In addition, the chimpanzees of the Tai Forest have been studied continuously since 1979 by Dr. C. Boesch and H. Boesch, and so represent the only existing community of well habituated lowland rain forest dwelling chimpanzees. For a complete description of the study site and habituation process see Boesch and Boesch (Boesch, 1978, 1991; Boesch and Boesch, 1981,1983,1984,1989). Focal animal sampling (Altmann, 1974) was used and each focal animal was followed for a n entire day, or a s long as possible (mean length of focal animal sample = 259 min). Results reported here are from Doran (199213) and are from one-minute instantaneous sampling (1msps). Comparison of data from the two separate sites Several factors allow for direct comparisons of the results from the two sites. One observer made all of the observations (of both species) used in this study. Thus, all categories of locomotor activity, substrate, and social and maintenance activity are identical a t the two study sites. Data obtained from concurrent instantaneous and continuous locomotor bout (with distance) sampling of chimpanzee locomotor behavior do not yield different results (Doran, 1992b). Thus, results obtained from instantaneous sampling of chimpanzees and locomotor bout sampling of bonobos are directly comparable. However, there are differences a t the two study sites that make direct comparisons difficult. The primary difficulty was the uneven level of habituation a t the two study 87 LOCOMOTOR BEHAVIOR OF CHIMPANZEES AND BONOBOS TABLE 1. A comparison of data from the Lomako Forest pygmy chimpanzee study with data from the Tai Forest chimpanzee study 1. Dates of studyitotal time spent at each site 2. No. of sightings' 3. Total time of sightings 4. Mean length of sighting 5. Mean length of focal animal sighting 6. Distribution of sighting throughout day (9% total time) P6 6-8 aio 10-12 12-14 14-16 16-18 7. Distributiodsex Males Females I Jnknown IA Lomako Tai 12/1/88-10/2/88 10 months 83 230 hr 168 rnin 18 rnin (range = 2-102) 03/01/88-09/25/88 07 months 74 438 hr 355 min 259 min (range = 16-660) 08 27 28 19 12 05 02 1,797 bouts 2,653 bouts 465 bouts 00 03 16 26 21 21 12 10,290 min 11,098 min 97 min sighting is defined as any uninterrupted follow of the group in which visual contact is not lost for more than 10 minutes. sites, which is clearly evident in a comparison of the data collected at the two sites (Table 1).The study of chimpanzees in the Tai Forest yielded a greater number of total hours of observation (in spite of the shorter duration of the study), longer mean duration per sighting and per focal animal follow, and a more even distribution of sightings by sex and throughout all hours of the day compared with the data collected in the bonobo study (Table 1). In addition, although Tai chimpanzees could be followed during terrestrial travel and accurate estimates of time spent on the ground versus time spent arboreally could be made, this was not the case with Lomako bonobos. Since there were fewer difficulties in observing the arboreal locomotion of bonobos, all quantitative comparisons made between bonobos and chimpanzees in this study are based on arboreal behavior only. Statistical analyses G Tests of Independence (RXC contingency tables) were used to determine whether the frequency of locomotor behavior of one species is independent of the frequency of locomotor behavior used by the other. The various statistics used are described in Sokal and Rohlf (1981). Significance values are from Rohlf and Sokal (1981). Symbols for significance used include * = P < 0.05; ** = P < 0.01; *** = P < 0.001; ns = not significant. All G values have been corrected by Williams correction factor in order to obtain a better approximation of the chi-square distribution. All tests are two-tailed and performed on raw data. RESULTS Arboreal locomotor behavior of Pan paniscus There are no significant differences in the frequency of locomotor bouts (with distance) spent in each locomotor activity by adult mothers with infants and adult females without infants (Table 2; G = 9.0, ns). There are, however, significant differences in the arboreal locomotor behavior of adult male and female bonobos (Table 2; G [males vs. mothers] = 51.7***; G [males vs. females] = 52.0""").Males are less quadrupedal and use more quadrumanous climbing and scrambling than females during arboreal locomotion. There is no significant sex difference in the frequencies of suspensory behavior, bipedalism, or leaping. Both sexes show significant differences in the frequencies of locomotor behaviors used during feeding and travel (Table 3; G [males] = 50.4"**;G [mothers] = 16.1**). For both sexes, quadrupedalism is used more frequently during feeding than during arboreal travel. 88 D.M. DORAN TABLE 2. Sex differences in the arboreal locomotor actiuitv 0uad Adult males 26.1 % of bouts (with distance) 17.7 % of bouts (no distance) 4.3 Mean distance b o u t (steps) n = 993 bouts Mothers with infants 44.4 % of bouts (with distance) 26.7 % of bouts (no distance) 5.7 Mean distance b o u t (steps) n = 468 bouts Adult females (no infants) %, of bouts (with distance) 39.2 % of bouts (no distance) 25.0 Mean distancehout (steps) 5.0 n = 1147 bouts G ualues Mothers with infants vs. adult females (no infants) Adult males vs. mothers with infants Adult males vs. adult females (no infants) G G G = = = of Pan uaniscus Categories of locomotor activity’ Climb Susu Bioed Leao 57.9 52.8 3.2 10.0 15.3 1.9 1.1 1.4 2.4 4.8 12.8 1.1 42.8 44.4 3.3 7.8 16.7 1.6 1.9 1.5 4.3 3.1 10.7 1.0 44.6 45.9 3.1 11.8 18.8 2.0 1.2 1.2 3.2 3.1 9.1 1.1 9.0 ns2 51.7*** 52.0*** ’Locomotor categories: quadrupedalism (quad),quadrumanous climbing and scrambling (climb),suspensory (susp), aided and unaided bipedalism (biped), leaping (leap). ’Significance levels: * = P < 0.05, ** = P < 0 01, *** = P < 0.01. ns = not significant. nents, including travellascend (ascend), defined as travel in a vertical direction from ground; travelhorizontal (travel), defined as arboreal travel in a horizontal direction Travel Feed along “arboreal highways”; traveVexit (exit), Males Females Males Females defined as exiting one tree and entering Quadrupedalism 20.0 37.1 47.1 49.7 an adjacent one; travelfdescend (descend), Climbkcram 64.3 43.3 40.8 41.7 defined as travel in a vertical direction Suspensory 9.9 14.9 7.0 5.7 Bipedal 0.6 1.5 2.8 0.8 towards the ground; traveVto ground Leaping 5.1 3.2 2.2 2.1 (ground), defined as the final 5-7 m during n 691 850 152 156 G (male vs. female) 84.6*** 2.2 ns traveudescend during which the animal 50.4*** G (male feed vs. male travel) moves from an arboreal to a terrestrial sub16.1** G (female feed vs. female travel) strate. Significance levels same as in Table 2. In addition to subdivisions in the category of travel, the locomotor category of quadruSex differences in arboreal locomotion oc- manous climbing and scrambling has been cur during travel rather than during feeding divided into its 4 separate components. (Table 3; G [feed] = 2.2, ns; G [travel] = These activities include: climbing (climb), 84.6***). Both males and females use large scrambling (scram.),tree-swaying (t. sway), amounts of quadrupedalism and quadru- and bridging (bridge). There are no significant sex differences in manous climbing and scrambling during arboreal feeding. However, during arboreal locomotor activities used in ascending from travel, males use considerably more quadru- the ground, descending during travel, or demanous climbing and scrambling and less scending to the ground (Table 4; G [asquadrupedalism than females. Males also cend] = 0.01, ns; G [descend] = 9.2, ns; G use slightly more leaping and less suspen- [ground] = 0.03, ns). Both males and females use vertical climbing as the primary sory behavior than females. In order to pinpoint exactly what differ- activity during ascent from the ground. The ences occur during arboreal travel locomo- most frequent means of descending to the tion, the category of locomotion during ground is by climbing or sliding down a vertravel is further subdivided into five compo- tical trunk, however transferring from a TABLE 3. A comparison of Pan paniscus arboreal locomotron used during trauel uersus feeding, expressed as percentages of distance traveled during bout sampling of arboreal locomotor behauior LOCOMOTOR BEHAVIOR OF CHIMPANZEES AND BONOBOS 89 TABLE 5. Frequency of subcategorces of quadrupedalcsm and of quadrumanous climbing and scrambling in Pan uaniscus Percent _ _ _ time _ _ . Males Mothers Quadrupedalism Knuckle-walking 17.0 Palmigrade 83.0 n 176 G (males vs. mothers) Quadrumanous climbinghcrambling Quadrumanous climbing 86.3 Scrambling 13.7 n 513 G (males vs. mothers) ns = 13.0 87.0 121 0.77 n s 86.3 14.1 208 0.01 ns not significant. trunk to some other substrate and using body weight to ridelcrash it to the ground (treesway) is also a common means of descent. There are significant sex differences in how bonobos travel horizontally and exit from one tree to another (Table 4; G [exit] = 20.9**; G [travel] = 19.3**).During horizontal arboreal travel, males use less quadrupedalism and more quadrumanous climbing, tree-swaying, and leaping than females. When crossing a discontinuity in arboreal substrate (exit), although both males and females use bridging most frequently, females use this conservative means of crossing more frequently than males. Females also use more suspensory behavior when crossing a discontinuity, whereas males use much more leaping than females. There are no sex differences in the type of quadrupedalism or quadrumanous climbing/scrambling used during arboreal locomotor activity (Table 5; G [quadrupedalism] = 0.77, ns; G [climbk.cramblel = 0.01, ns). Eighty-five percent of all arboreal quadrupedal activity for both males and females is palmigrade rather than knuckle-walking. Of all quadrumanous climbing and scrambling locomotor bouts, both males and females spend 86% of bouts (with distance) in climbing, and only 14%in scrambling activities. Arboreal substrate use of Pan paniscus There are significant differences in the substrate use of bonobo males, mothers, and adult females without infants (Table 6; G [males vs. mothers] = 33.9***; G [males vs. ~ 90 D.M. DORAN TABLE 6. Pan paniscus substrate use during arboreal locomotion Adult males % of bouts (with distance) 24.7 12.4 % of bouts (no distance) 15.6 8.0 Mean distancelsubstrate 4.6 4.5 n = 963 bouts Mothers with infants I of bouts (with distance) 15.5 19.6 % of bouts (no distance) 10.3 12.1 Mean distancdsubstrate 5.2 5.6 n = 453 bouts Adult females (no infants) I of bouts (with distance) 13.9 20.3 % of bouts (no distance) 9.0 12.2 Mean distancehbstrate 5.2 5.6 n = 1,116 G values (% of bouts [with distance] converted to raw nos.) Mothers with infants vs. adult females (no infants) G = 19.3*** Adult males vs. mothers with infants G = 33.9*** Adult males vs. adult females (no infants) G = 117.1*** 32.3 33.6 2.8 14.1 31.6 1.3 16.5 11.2 4.3 40.0 34.6 4.0 9.8 30.7 1.1 15.0 12.1 4.3 47.1 39.7 4.0 10.7 32.7 8.0 6.4 4.3 1.1 'Substrates are as follows: trunk ( T R b s t o u t , primary members of the tree; bough (BOLsecondary elements that range 15-20 cm in diameter; branch ( B R t t e r t i a r y supports that range from 2-15 cm in diameter; Lane (LI). Significance levels are the same as in Table 2 females] = 117.1"""; G [mothers vs. females] = 19.3"""). Adult females without infants differ from adult mothers in using branches more frequently, and lianes less frequently. Both adult mothers and females differ in substrate use from males in using trunks less frequently and boughs and branches more frequently. Foliage use is approximately equal to or slightly higher in males. Relationship between arboreal substrate use and locomotor activity of Pan paniscus For all substrates, except branches, there are no sex differences in the type of locomotor activity performed on a given substrate (Table 7; G [trunk] = 1.1 ns; G [bough] = 2.3, ns; G [branch] = 12.9"; G [foliage] = 7.6 ns; G [lianel = 3.2, ns). Quadrumanous climbing is the most frequently occurring activity on trunks and lianes. Palmigrade quadrupedalism is the predominant activity on boughs. Quadrumanous climbing and scrambling is used most frequently in foliage. Adult male and female bonobos do however, differ in the type of activities performed on branches (Table 7; G = 12.9"). Females are more frequently quadrupedal on branches than are males. Males climb and scramble, leap, arm-swing, and walk bipedally (aided) more frequently on branches than females do. Ranging behavior of Pan paniscus: Arboreal and terrestrial travel Arboreal travel, defined here as above ground travel between feeding or resting sites, contributes a significant portion of Lomako Forest bonobo ranging. The exact proportion of overall distance travelled arboreally versus terrestrially cannot be ascertained at this time for reasons cited above. However, 86% of all sightings that included travel of any kind (n = 80) included some arboreal travel. The mean arboreal distance travelled per arboreal sighting is 234 m, ranging from 6 to 1,169 m (n = 69). Fifty-two percent of arboreal travel sightings were of a distance of <ZOO m. Twentynine percent of these sightings were of distances ranging between 200 and 400 m, and 19% of these sightings were for distances of >400 m. Two sightings included group arboreal travel of >1km. The mean distance of terrestrial travel per sighting, based on combined direct and indirect sightings, is 359.3 m, ranging from 01 to 1,700 m (n = 45). Nearly all observed terrestrial travel (99.7% of bouts with dis- LOCOMOTOR BEHAVIOR OF CHIMPANZEES AND BONOBOS 91 TABLE 7. Percentage of male and female locomotor bouts (with distance) spent in each locomotor activity in Pan paniscus, related to locomotor types Substrate Quad Categories of locomotor activities susp Biped Climb Leap n ~ Trunk Males Females G = 1.1ns Bough Males Females G = 2.3 ns Branch Males Females G = 12.9* Foliage Males Females G = 7.6 Liane Males Females G = 3.2 ns - 97.3 98.9 2.0 1.2 150 101 78.0 85.5 12.7 7.1 3.8 4.7 1.4 0.5 4.1 2.2 77 136 49.0 57.0 27.0 24.7 16.9 15.8 2.9 0.4 4.2 2.1 324 443 0.2 0.2 68.9 61.1 17.8 27.1 - 13.1 11.5 304 365 3.7 7.9 88.2 84.1 6.1 5.7 2.0 1.9 0.3 108 71 Significance levels are the same as in Table 2 tance) is quadrupedal knuckle-walking. Unaided bipedalism accounts for the remaining 0.3% of bouts. Ranging behavior of Pan troglodytes verus: Arboreal and terrestrial travel Arboreal andor terrestrial travel was recorded during 89.9% of sightings (n = 79). Although adult Pan troglodytes uerus spent considerable time arboreally (Doran, 1992b), travel between feeding and resting sites was exclusively terrestrial. During the 7 month study, no adult arboreal travel comparable to that seen in bonobos was recorded. Eighty-eight percent of the arboreal locomotion during travel recorded for Tai chimpanzees was strictly ascending into feeding or resting trees, or descending from them to the ground (n = 6 5 , l msps). The remainder of adult arboreal travel (n = 7 , l msps) was spent exiting one feeding or resting tree, and entering a smaller adjacent tree, either to rest, or to use in descending to the ground. The mean horizontal distance travelled during these 8 episodes of arboreal travel was 53.0 m, ranging from 25 to 96 m. The only incident of arboreal travel between feeding or resting sites at Tai involved a juvenile male (Sartre). On August 16, 1988, he and his mother (Salome) and her infant were grooming in a tree. The mother (with infant) descended to the ground, and slowly knuckle-walked 120 m to another tree, and then ascended into it. The juvenile did not descend to the ground, but, instead travelled along an arboreal route to join his mother. In 430 hrs of observation, this was the single episode of arboreal travel between (non-adjacent) feeding sites. lnterspecific differences in arboreal locomotor behavior and substrate use Both male and female bonobos use more quadrupedalism and less quadrumanous climbing and scrambling than their chimpanzee counterparts (Table 8; G [males] = 33.2***; G [mothers] = 11.8"). The species difference is more pronounced between males, with P. paniscus males additionally using more suspensory and leaping behavior than male P. troglodytes. For both sexes, there is a significant species difference in the type of arboreal quadrupedalism used (Table 9; G [males] = 14.1***;G [mothers] = 25.6***). Both male and female bonobos use far more palmigrade quadrupedalism and less knucklewalking quadrupedalism than Tai chimpanzees. In fact, approximately 85% of bonobo arboreal quadrupedalism is palmigrade, whereas only 3 0 4 0 % of chimpanzee arbo- 92 D.M. DOFUN TABLE 8. Interspecific differences in chimpanzee arboreal locomotor behavior Pan paniscus Males Mothers' Pan troglodytes Males Mothers' 44.4 11.6 30.3 26.1 Quadrupedalisrn 59.8 42.8 76.7 Quad. climb 57.9 Suspensory 10.0 7.8 5.8 7.4 1.9 5.8 0.8 Biped a1 1.1 1.6 Leap 4.8 3.1 n 993 468 103 122 G Values Pan paniscus males vs. mothers 51.7""" 17.7** P. troglodytes males vs. mothers Males: P. paniscus vs. P. troglodytes 33.2*"* Mothers: P. paniscus vs. P. troglodytes 11.8" 'Since there were no adult females without infants in the Tai community studied, interspecific comparisons between females are restricted to mothers Significance levels are the same a s in Table 2. TABLE 9. A comparison of Pan paniscus and Pan troelodvtes arboreal auadruoedalism Arboreal quadrupedalism i%) Pan paniscus Pan troglodytes Males Mothers Males Mothers 72.7 57.1 17.0 13.0 Knuckle-walking 42.9 87.0 27.3 Palmigrade 83.0 quadrupedalism n 176 121 11 35 G (males vs. mothers) 0.77 ns 0.88 ns G (Interspecific difference: males) 14.1*"" G (Interspecific difference: females) 25.6*** Significance levels are the same as in Table 2. real quadrupedalism is palmigrade. In addition, within each species, females are more palmigrade in their arboreal quadrupedalism than the males are. There are neither sex differences within each species nor interspecific differences in the type of quadrumanous climbing and scrambling used (G [males] = 0.01, ns; G [mothers] = 2.14, ns). For both sexes, quadrumanous climbing is a far more frequent activity than scrambling. There are significant interspecific differences in substrate use for both males and females (Table 10; G [males] = 11.8"; G Lmothers] = 38.6"""). Male and female Pan paniscus both use trunks less often and boughs and branches more often than male and female Tai chimpanzees. Foliage use is approximately equal in the two species. Lianes are used less frequently by bonobo mothers than chimpanzee mothers. TABLE 10. Interspecific differences in chimpanzee arboreal substrate use Pan paniscus Males Mothers Pan troglodytes Males Mothers Substrate Trunk 24.7 15.5 34.9 34.4 Bough 12.4 9.7 7.4 19.6 Branch 32.3 40.0 18.4 22.9 9.8 18.4 10.7 Foliage 14.1 15.0 18.4 24.6 Liane 16.5 n 963 453 103 122 G values Pan paniscus males vs. mothers G = 51.7%"" P. troglodytes males vs. mothers G = 4.2 ns Males: P. paniscus vs. P. troglodytes G = 11.8" Mothers: P. paniscus vs. P. troglodytes G = 38.6""" Significance levels are the same as in Table 2 There are no interspecific differences in the frequency of activity performed on trunks, branches, foliage, or lianes (Table 11;G [trunk] = 0.34, ns; G [branch] = 7.24, ns; G [foliage] = 4.61, ns; G [lianel = 4.50, ns). For both species, quadrumanous climbing is the most frequently used activity on vertical trunks and lianes. Quadrupedalism accounts for 26% of all locomotor activities occurring on trunks at Tai. However, this is because horizontal (treefalls) and vertical trunks are included in the same category. Quadrupedalism occurs on horizontal trunks. The results of trunk use are not statistically different at the two sites. Quadrumanous climbing and scrambling is used most frequently on branches, although both bonobos and chimpanzees use climbing, scrambling, and suspensory behavior as well on branches. In foliage, quadrumanous climbing and scrambling and suspensory behavior are frequent activities for both species. However, there is an interspecific difference in bough use (Table 11; G = 21.5"""). Bonobos are most frequently quadrupedal (and particularly palmigrade quadrupedal) when on boughs, whereas chimpanzees use little quadrupedalism and far more quadrumanous climbing, scrambling and aided bipedalism. DISCUSSION Pan paniscus locomotor and ranging behavior Data are presently insufficient to determine exactly what proportion of bonobo LOCOMOTOR BEHAVIOR OF CHIMPANZEES AND BONOBOS 93 TABLE 11. Interspecific differences in the relationship between arboreal substrate and locomotor activity performed % of combined adult male and female locomotor bouts spent in each locomotor activity Substrate Trunk P. paniscus P. troglodytes G = 0.34, ns Bough P. paniscus P. troglodytes G = 21.5*** Branch P. paniscus P. troglodytes G = 7.24, ns Foliage P . paniscus P. troglodytes G = 4.61, n s Liane P. paniscus P. troglodytes G = 4.50. n s . Climb s us p Biped 3.5 26.0 94.6 74.0 - 0.4 - 1.5 - - 260 77 82.6 31.2 9.4 56.2 - 0.9 12.5 2.8 213 - 53.6 46.8 25.6 21.3 16.3 19.1 1.4 8.5 3.0 4.2 766 47 0.3 64.6 75.0 22.9 20.8 - 12.3 4.2 - 669 24 93.9 100.0 - 2.1 - - 2.9 - 4.2 Leap __ Quad 1.1 n 16 145 47 Significance levels same as in Table 2. TABLE 12. Comparison of this study of Pan paniscus arboreal locomotor behavior with Susman’s study at the same site: percentages of arboreal locomotor bouts snent in each activity __... Present study (combined male and female adults) 9% bouts %bouts (with (no distance) distance) Susman (1984) (all age classes) %bouts (no distance) Quadrupedalism 35.2 22.5 Climb/scram 49.4 48.2 Suspensory 10.3 17.1 Bipedal 1.3 1.2 Leap 3.8 10.8 n 2,608 2,608 G values G (bouts with distance vs. pervious work) G (bouts no distance vs. urevious work) 31 31 21 10 6 1,700 65.0*** 290.9*** Significancelevels same as in Table 2 travel is made arboreally versus terrestrially. It is most likely that, like chimpanzees, bonobos’ primary means of travel between feeding and resting sites is knuckle-walking quadrupedalism. However, data presented here indicate that bonobos can, and do, travel substantial distances arboreally. There are significant sex differences in the frequencies of locomotor behaviors used (particularly on branches) during arboreal horizontal travel and in exiting from one tree to an adjacent one. Females use more quadrupedalism and less climbing, treeswaying and leaping than males during horizontal arboreal travel. Results of males decreased quadrupedalism and increased climbing and scrambling are in accordance with predictions based on body size. As males and females travel through the same “arboreal highways”, the smaller females encounter more substrates (relative to males) that are sturdy enough to support them during quadrupedalism. The larger males find fewer substrates that support them during quadrupedalism, and this probably accounts for their greater incidence of climbing and scrambling in comparison with females. However, contrary to predictions, when exiting trees, males leap more frequently and bridge less frequently than females. Cartmill and Milton (1977) argue that larger animals should be more cautious (and thus leap less) than smaller animals. This was not the case in this study. One confounding factor here is that the smaller animals (females) are accompanied by their offspring and, not unreasonably, choose the more cautious method of bridging (reaching out and pulling foliage from the adjacent tree to form a bridge of foliage for her offspring t o cross on). She then swings across on the foliage, which accounts for her 94 D.M. DORAN higher incidence of suspensory behavior than males. This sex difference in exiting trees partially supports Kano's (1983) observation that males engage more frequently than females in high risk activities such as annswinging and diving. In this study a higher incidence of suspensory behavior was not noted. There is a sex difference in the frequency of substrate use as well. Males use trunks (and climb) more frequently than females. On two separate occasions during follows of adult female bonobos, a male exited the feeding tree and descended to the ground. The females travelled arboreally to a feeding tree. A few minutes after their arrival, the same lone male arrived on the ground, climbed up the trunk, and joined the females in the feeding tree. It seems possible that males may travel terrestrially relatively more frequently than females. This sex difference in travel would not be unprecedented since there is evidence that orangutan males travel more frequently on the ground than females (Galdikas and Teleki, 1981; Sugardjito and van Hoof, 1986). Further study is necessary to determine if it occurs on any regular basis, and if so, to what extent. Pan paniscus locomotor behavior: Comparisons with previously published work Table 12 compares the results of this study with those from earlier work of Susman (1984). The two studies are similar in that both were conducted at the same study site on the same communities of animals. This study was conducted 4 years after the completion of Susman's study. The categories of behavior and substrate are identical. There are two major differences in the sampling methods used in the two studies. The first is that although locomotor bout sampling was used in both studies, this study reports all results based on locomotor bouts (with distance), whereas Susman's results are based on the percentage of bouts (no distance) spent in a given activity. The second difference in the two studies is that this study reports the results of adults only, whereas Susman (1984) combines all age and sex classes. For purposes of comparison, male and female data from this study are combined into one category. In addition, results from both locomotor bouts (with distance) and bouts (no distance) are included in order to more accurately assess the difference in the results of the two studies. There are significant differences in the result of this study with those of Susman (Table 12; G [bouts with distance] = 65.0"""; G [bouts no distance] = 290.9"""). The results of the present study indicate less suspensory behavior, bipedalism and leaping, and a considerable increase in the amount of climbing and scrambling. When the results of this study are considered without weighting bouts with distance, the differences in the two studies, at least in regard to the frequency of suspensory behavior, are less marked. However, it has since been demonstrated that failure to weight bouts with distance distorts results by overestimating activities that occur relatively frequently, but have a short mean distance per bout (Doran, 1992b).This suggests that the higher frequency of suspensory behavior reported in earlier studies is due, in part, to the sampling method used. In addition, immature bonobos engage in more frequent suspensory and less frequent quadrupedal behavior than their adult counterparts (Doran, 1992a). The higher incidence of suspensory behavior reported by Susman may also be explained, in part, by his combining immature and adult data. However, all dissimilarities in the results of the two studies cannot be attributed solely to differences in sampling methods. Susman (1984) noted a difference in the locomotor behavior of the bonobos of the Lomako Forest through time, as a result of their increased habituation to observers. He reported t,hhat the frequency of bipedalism and leaping and diving decreased, and that of quadrumanous climbing and scrambling increased through time (Susman et al., 1980; Susman, 1984). These same trends are evident in a comparison of this study with his earlier results, indicating that further habituation has occurred. LOCOMOTOR BEHAVIOR OF CHIMPANZEES AND BONOBOS No quantitative description of the locomotor behavior of bonobos has been made at other sites. However, from qualitative descriptions (Horn, 1976; Kano, 1983; pers. obs.), it appears that there are some intersite behavioral differences. The most striking intersite difference is that at the Wamba (Zaire) study site, bonobos are reported to travel exclusively on the ground (per. com. Kuroda; pers. obs.). This difference is probably related to 3 factors: habitat differences, differences in the degree of habituation, and the possible effects of artificial provisioning. Although the Lomako and Wamba study sites are only separated geographically by a distance of approximately 300 km, there are major habitat differences at the two sites. The Lomako site is situated at a distance of 35 km from the nearest road or village, and consists largely of undisturbed primary forest. The Wamba study site is comprised of approximately 100 square km, which surrounds and includes the 5 hamlets of the village of Wamba (Kano and Mulavwa, 1984). There is more secondary forest at Wamba than in the Lomako, and the area has been influenced greatly by human inhabitants, as is evidenced by the relative paucity of wildlife in comparison with the Lomako. Thus, more primary forest with resultant different forest structure (more continuous arboreal routes) and possibly higher density of predators (due to undisturbed nature of forest with less human intervention) in the Lomako Forest may have some influence on the greater incidence of arboreal travel in the Lomako Forest compared with Wamba. The bonobos of the Lomako Forest have never been artificially provisioned. Neither are they fully habituated to the presence of human observers, whereas the bonobos at Wamba are fully habituated to human observers. Kuroda reported (pers. com.) that before habituation at Wamba, bonobos fled human observers by travelling in the trees. This suggests that the distances (of up to 1,200 m) travelled arboreally by Lomako bonobos are an artifact of the lack of habituation. I suggest that the length of the distances travelled refute that argument to some extent, since although travelling arbo- 95 really for short distances (<200 m) might be a means for the animals to distance themselves from the observer before descending to the ground, there were several sightings in which the animals travelled arboreally greater distances. Travel frequently occurred at a leisurely pace, after animals fed and rested in one tree, and then continued on to another tree. In one instance, this continued for 12 hrs, during which the animals fed and travelled arboreally (1,200 m) without descending to the ground. Provisioning may also have an effect on terrestriality. On the morning that I observed at Wamba, bundles of sugar cane were distributed under the night nests and the bonobos descended immediately to the ground and fed on the sugar cane, rested at the same site, and eventually moved off on the ground. It is not surprising that with food available on the ground the bonobos descended upon leaving their night nests. These differences in forest structure, degree of habituation, and effects of provisioning make it difficult to interpret the differences in ranging at the two sites. Clearly, these intersite differences cannot be resolved without further study. However, it clearly points out a sharp contrast in the behavior of bonobos and chimpanzees. Whether habituated or not, no equivalent behavior for forest dwelling chimpanzees has been reported. lnterspecific differences in arboreal locomotor behavior There are species differences in chimpanzee and bonobo ranging behavior, arboreal locomotion, substrate choice, and type of locomotor activity performed on boughs. Both species (most likely) range between feeding and resting sites primarily by knuckle-walking quadrupedalism. However, arboreal travel is a more significant component of bonobo travel than of chimpanzee travel. This is as predicted on the basis of species differences in scapular shape. In comparisons of arboreal locomotor behavior, bonobos (both males and females) use more quadrupedalism and less quadrumanous climbing and scrambling than their chimpanzee counterparts. This trend is seen 96 D.M. DORAN for both species in comparisons of male and female behavior, with the smaller females using more quadrupedalism and less climbing and scrambling than the larger males. At first glance this supports the idea that this interspecific difference is a result of body-size differences between the two species. However, there is also a striking qualitative difference in quadrupedalism between species that is not present between the sexes. Bonobos not only use more arboreal quadrupedalism than chimpanzees do, but it is primarily palmigrade in nature, whereas chimpanzees almost always knuckle-walk, when walking quadrupedally. This difference is most noticeable on boughs, where bonobos use palmigrade quadrupedalism and chimpanzees use climbing, scrambling and aided bipedalism. This “reluctance” to use palmigrade quadrupedalism may influence the overall reduced usage of arboreal quadrupedalism by chimpanzees since knuckle-walking would, on average, require a larger substrate than palmigrade quadrupedalism does. No clear explanation presents itself to explain why this species difference should occur, although it suggests potential anatomical differences in the wrist and hand of the two species. An additional, species difference in arboreal locomotor behavior that cannot be explained on the basis of possible body size difference is the increased suspensory behavior of male bonobos in comparison with male chimpanzees. Based on body-size predictions the opposite should be true. Instead, as predicted by Susman (19791, increased suspensory behavior is probably correlated with the bonobo’s longer and more narrow scapula and more curved phalanges. In addition to differences in overall frequencies of locomotor activities, there are interspecific differences in substrate choice: chimpanzees use trunks more frequently; bonobos use boughs and branches more frequently. Since chimpanzee arboreal travel consists primarily of ascending into or descending from feeding or resting trees, it is not surprising that both male and female chimpanzees use trunks relatively more frequently (in comparison with other substrates) than bonobos. CONCLUSIONS My data support the prediction, based on scapular shape and degree of curvature of the phalanges, that bonobos are more suspensory than chimpanzees. It is clear that arboreal travel is a significant component of bonobo overall travel, and that no equivalent chimpanzee behavior has yet been reported. In addition, during arboreal locomotion, male bonobos arm-swing more frequently than male chimpanzees. This could not have been predicted on the basis of body size, since, if there is a body-size difference between P. t. uerus and P. paniscus (which has not yet been determined), then one would predict that (larger) chimpanzees would be more suspensory during arboreal locomotion than (smaller) bonobos. Clearly, the suggestion that the bonobo is a “scaled reduced version” of a chimpanzee with little or no positional behavior difference can be rejected. Although I presented no data on the ontogeny of locomotor behavior in this study, previously reported results indicate that the arboreal locomotor behavior of adult bonobos resembles that of immature rather than mature chimpanzees (Doran, 1992a),which thus supports Coolidge (1933) in his description of paedomorphism. The results of this study also support earlier predictions of Coolidge (19331, Roberts (19741, and Susman (1987)that associated with the bonobo’s longer and more narrow scapula are predictable behavioral differences of increased arboreality and suspensory behavior. ACKNOWLEDGMENTS I would like to thank Drs. John Fleagle, Charles Janson, William Jungers, Michael Rose, Randall Susman, and Russell Tuttle for their insightful comments on this manuscript. In addition, I would especially like to acknowledge the debt I owe Drs. John Fleagle and Dr. Randall Susman for their guidance during my graduate training. I am grateful to the Ministry of Scientific Research and the Station d’Ecologie Tropicale and its director, Dr. Henri Dosso, for permission to conduct research in the Ivory Coast. I thank the Centre Suisse de Recherche Scientifique and its director, M. Peter Lehman, LOCOMOTOR BEHAVIOR OF CHIMPANZEES AND BONOBOS for logistic support during my stay in the Ivory Coast. Special thanks to Dr. Christophe and Hedwige Boesch for sharing their friendship, guidance, and the chimpanzees of Tai with me. My field work at Tai would have been less profitable and far less enjoyable without the assistance and friendship of my field assistant, M. Gregoire Nohon. 1 a m grateful to the government of Zaire and its Institut de Recherche Scientifique for permission to study in the Lomako Forest. Special thanks go to Drs. Nancy ThompsonHandler, Richard Malenky, and Annette Lanjouw for my introduction to field work and to the pygmy chimpanzees. In addition, I am most grateful to Dr. Takayoshi Kano for permission to visit Wamba, and to Drs. Suehisa Kuroda and Takeshi Furuichi for gracious hospitality and lively discussions during my stay. I gratefully acknowledge the Louis B. Leakey Foundation, the Wenner Gren Foundation, and the National Science Foundation (to R. L. Susman) for their generous financial support. LITERATURE CITED Altmann J (1974) Observational study of behavior: sampling methods. Behaviour 49:227-267. Badrian A and Badrian N (1977) Pygmy chimpanzees. Oryx 13:463-468. Badrian A and Malenky R (1984) Feeding ecology ofPan paniscus in the Lomako Forest, Zaire. In RL Susman (ed.): The Pygmy Chimpanzee: Evolutionary Biology and Behavior. New York: Plenum Press, pp. 325-344. Boesch C (1978) Nouvelles observations sur les chimpanzes de la foret de Tai (Cote DIvoire). La Terre et la Vie 32:195-201. Boesch C (1991) The effects of leopard predation on grouping patterns in forest chimpanzees. Behaviour 117(3/):265-286. Boesch C and Boesch H (1981) Sex differences in the use of natural hammers by wild chimpanzees: a preliminary report. J. Hum. Evol. 10.585-593. Boesch C and Boesch H (1983) Optimisation of nutcracking with natural hammers by wild chimpanzees. Behaviour 3/4: 265-286. Boesch C and Boesch H (1984) Mental maps in wild chimpanzees: a n analysis of hammer transports for nut cracking. Primates 25:160-170. Boesch C and Boesch H (1989) Hunting behavior of wild chimpanzees in the Tai National Park. Am. J . Phys. Anthropol. 78:547-573. Cant J G (1989) Effects of sexual dimorphism in body size on feeding postural behavior of Sumatran Orangutans. Am. J . Phys. Anthropol. 74:143-148. 97 Cartmill M and Milton K (1977) The lorisiform wrist joint and the evolution of brachiating adaptations in the Hominoidea. Am. J . Phys. Anthropol. 47,249272. Coolidge H J (1933) Pan paniscus: pygmy chimpanzee from the south of the Congo River. Am. J. Phys. Anthropol. 18:l-59. Coolidge H J (1984) Historical remarks bearing on the discovery of Pan punzscus. In RL Susman (ed.):The Pygmy Chimpanzee: Evolutionary Biology and Behavior. New York: Plenum Press, pp. ix-xiii. Coolidge H J and Shea BT (1982) External body dimensions of Pan paiiiscus and Pan troglodytes chimpanzees. Primates 23t245-251. Corruccini RS and McHenry HM (1979) Morphological affinities of Pun puniscus and human evolution. Science 204:1341-1343. Cousins D (1978) The diminutive Pan. Int. Zoo. News 25:5-11. Cramer DL (1977) Craniofacial morphology of Pan paniscus. Contrib. Primatol. 10:l-64. Crompton RH (1983) Age differences in locomotion in two subtropical Galaginae. Primates 24r2P59. Doran DM (1992, a) The ontogeny of chimpanzees and pygmy chimpanzee locomotor behavior: a case study of paedomorphism and its behavioral correlates. J. Hum. Evol. 23:139-157. Doran DM (1992, b) A comparison of instantaneous and locomotor bout sampling methods: a case study of adult male chimpanzee locomotor behavior and substrate use. Am. J. Phys. Anthropol. 89:85-99. Doran DM (in press) Sex differences in adult chimpanzee positional behavior: the influence of body size on locomotion and posture. Am. J. Phys. Anthropol. Fenart R and Deblock R (1973) Pun paniscus et Pun troglodytes craniometrie: Etude comparative et ontogenique selon les methodes classiques et vestibulaires. Tome I, Mus. R. Afr. Centre. Tervuren Belg. Ann. Octavo Ser. Sci. Zool. 204r1473. Fleagle J G (1985) Size and adaptation in primates. In WL Jungers (ed.): Size and Scaling in Primate Biology. New York: Plenum Press, pp. 1-19. Fleagle J G and Mittermeier RA (1980) Locomotor behavior, body size and comparative ecology of seven Surinam monkeys. Am. J. Phys. Anthropol. 53:201314. Frechkop S (1935) Notes sur les mammiferes. XVII: A Propos du chimpanze de la rive gauche du Congo. Mus. R. Hist. Nat. Belg. Bull. 11:141. Furuichi T (1987) Sexual swelling, receptivity, and grouping of wild pygmy chimpanzee females at Wamba, Zaire. Primates 28(31:309-318. Galdikas BM and Teleki G (1981) Variations in subsistence activities of female and male Pongids: New perspectives on the origins of hominid labor division. Curr. Anthropol. 22:241-253. Goodall J (1968) The behavior of free-living chimpanzees in the Gombe Stream Reserve. h i m . Behav. Monogr. It161-311. Horn AD (1976) A Preliminary Report on the Ecology and Behavior of the Bonobo Chimpanzee (Pun paniscus, Schwarz, 1929)and a Reconsideration ofthe Evo- 98 D.M. DORAN lution of the Chimpanzee. Ph.D. Dissertation, Yale University, New Haven. Horn AD (1979) The taxonomic status of the bonobo chimpanzee. Am. J. Phys. Anthropol. 51:273-282. Hunt KD (1992) Positional behavior of Pan troglodytes in the Mahale Mountains and Gombe Stream National Parks, Tanzania. Am. J . Phys. Anthropol. 87t83-105. Johanson DC (1974) Some metric aspects of the permanent and deciduous dentition of the pygmy chimpanzee (Panpaniscus). Am. J. Phys. Anthropol. 41:3948. Johnson SC (1981)Bonobos: Generalized hominid prototypes or specialized insular dwarfs? Curr. Anthropol. 22:363-375, Jungers WL and Susman RL (1984) Body size and skeletal allometry in African apes. In RL Susman (ed.): The Pygmy Chimpanzee. New York: Plenum Press. Kano T (1980) Social behavior of wild pygmy chimpanzees (Pan paniscusi of Wamba: A preliminary report. J. Hum. Evol. 9:243-260. Kano T (1983) An ecological study of the pygmy chimpanzee (Pan paniscus) of Yalosidi, Republic of Zaire. Int. J. Primatol. 4:l-31. Kano T (1992) The Last Ape: Pygmy Chimpanzee Behauior and Ecology. Stanford, CA: Stanford University Press. Kano T and M Mulavawa (19841Feeding Ecology of the Pygmy Chimpanzees (Panpaniscus)of Wamba. In RL Susman (ed.): The Pygmy Chimpanzee: Evolutionary Biology and Behavior. New York: Plenum Press. pp. 233-274. Kinzey WG (1971) Evolution ofthe human canine tooth. Am. Anthropol. 73:680-694. Kinzey WG (1984)The dentition of the pygmy chimpanzee. In RL Susman (ed.): The Pygmy Chimpanzee: Evolutionary Biology and Behavior. New York: Plenum Press, 65-88. Kuroda S (1980) Social behavior of the pygmy chimpanzees. Primates 21(2):181-197. Mackinnon J (1978)The Ape Within Us. New York: Holt Rinehart and Winston. McHenry HM and Corruccini RS (1981) Pan paniscus and human evolution. Am. J. Phys. Anthropol. 54:355-367. Pilbeam DR and Gould SJ (1974) Size and scaling in human evolution. Science 186t892-901. Roberts D (1974) Structure and function of the primate scapula. In FA Jenkins Jr. (ed.): Primate Locomotion. New York: Academic Press. Rohlf FJ and Sokal RR (1981) Statistical Tables. New York: WH Freeman and Co. Rose MD (1977) Positional behavior of olive baboons (Papio anubus) and its relationship to maintenance and social activities. Primates 18t59-116. Schultz AH (1969) The skeleton of the chimpanzee. In GH Bourne (ed.): The Chimpanzee. Vol. 1. Basel: Karger. Schwarz E (1929) Das Vorkommen des Schimpansen auf den linken Kongo-Ufer. Rev. Zool. Bot. Afr. XVI(4):425426. Shea BT (1982) Growth and size allometry in the African Pongidae: cranial and postcranial analyses. Ph.D. thesis. Duke University. Shea BT (1983) Allometry and heterochrony in the african apes. Am. J. Phys. Anthropol. 62:275-290. Shea BT (1984) Allometry in the pygmy and common chimpanzees. In R.L. Susman (ed.): The Pygmy Chimpanzee: Evolutionary Biology and Behavior. New York: Plenum Press. Shea BT (1986) Scapula form and locomotion in chimpanzee evolution. Am. J. Phys. Evol. 70:475488. Sokal R and Rohlf FJ (1981) Biometry. New York: WH Freeman and Co. Sugardjito J and van Hooff J (1986) Sex-age class differences in positional behavior of Sumatran orangutan (Pongo pygmaeus abelii) in the Gunung Leuser National Park, Indonesia. Folia primatol. 47:14-25. Susman RL (1979) Comparative and functional morphology of hominoid fingers. Am. J . Phys. Anthropol. 50: 2 15-236. Susman RL (1984)The locomotor behavior of Pan panzscus in the Lomako Forest. In RL Susman (ed.): The Pygmy Chimpanzee: Evolutionary Biology and Behavior. New York: Plenum Press. Susman RL (1987) Pygmy chimpanzees and common chimpanzees: models for the behavioral ecology of the earliest hominids. In WG Kinzey (ed.): Primate Models for the Evolution of Human Behavior. New York: SUNY press. Susman RL, Badrian NL, and Badrian AJ (1980) Locomotor behavior of Pan paniscus in Zaire. Am. J. Phys. Anthropol. 53r69-80. Susman RL and Jungers WL (1981) Comment on: Bonobos: generalized hominoid prototype or specialized insular dwarf? by Johnson, S.C., Curr. Anthropol. 22: 369-3 70. Weidenreich F (1941) The brain and its role in the phylogenetic transformation of the human skull. Trans. Am. Philos. SOC. 31:321442. White FJ (1986) Behavioral Ecology of the Pygmy Chimpanzees. Ph.D. Dissertation. SUNY at Stony Brook, Stony Brook. Zihlman A and Cramer D (1978) Skeletal differences between pygmy (Punpaniscusi and common chimpanzees (Pan troglodytes). Folia Primatol. 29t86-94. Zihlman AL, Cronin JE, Cramer DL and Sarich VM (1978) Pygmy chimpanzee as a possible prototype for the common ancestor of humans, chimpanzees, and gorillas. Nature 275t744-746.