Compliant walking in primates Elbow and knee yield in primates compared to other mammals.код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 125:42–50 (2004) Compliant Walking in Primates: Elbow and Knee Yield in Primates Compared to Other Mammals Eileen Larney1 and Susan G. Larson2* 1 Interdepartmental Doctoral Program in Anthropological Sciences, Stony Brook University, Stony Brook, New York 11794-4364 2 Department of Anatomical Sciences, Stony Brook University School of Medicine, Stony Brook, New York 11794-8081 KEY WORDS arborealism limb compliance; limb yield; early primate evolution; marsupials; ABSTRACT It has been suggested that primates utilize a compliant gait to help reduce peak locomotor stresses on their limbs (Schmitt  J. Hum. Evol. 26:441– 458; Schmitt  Primate Locomotion, p. 175– 200; Schmitt  J. Zool. Lond. 248:149 –160). However, the components of such a gait, i.e., increased step length, prolonged contact time, and substantial limb yield, have only been documented on a handful of primate species. In order to explore the generality of this claim, elbow and knee angles during walking were documented at touchdown, midstance, and liftoff in a sample of primates, carnivores, marsupials, rodents, and artiodactyls, all under 25 kg. Limb yield was calculated as the change in angle from touchdown to midstance, and re-extension as the change in angle from midstance to liftoff for both forelimbs and hind limbs. Use of a compliant gait (as reﬂected in signiﬁcant limb yield) in primates was conﬁrmed for both forelimbs and hind limbs. However, there was variability within primates in the degree of either elbow or knee yield. Surprisingly, marsupials were found to exhibit almost as much elbow yield and even greater knee yield than primates. Carnivores and rodents display a modest amount of limb yield during walking, while artiodactyls appear to display a relatively stiff gait. These data are consistent with the suggestion that the use of a compliant gait to attenuate peak substrate reaction forces may have facilitated the primate invasion of a small-branch niche. However, limb compliance (as reﬂected by elbow or knee yield) does not appear to be exclusive to the primate order. Am J Phys Anthropol 125:42–50, 2004. Use of clawless, grasping extremities has long been recognized as a distinctive characteristic of primates (e.g., LeGros Clark, 1959). According to current views on primate origins, this as well as other early primate specializations can be related to adaptations for foraging and feeding on small terminal branches (Cartmill, 1972, 1974, 1992; Rasmussan, 1990; Sussman, 1991; Hamrick, 1998; Fleagle, 1999). The use of clawless, grasping hands and feet to forage and feed on small terminal branches has, in turn, been either directly or indirectly related to a variety of other morphological or behavioral specializations in primates. As Schmitt (1999) argued, traveling on small branches presents conﬂicting demands for long and mobile limbs to reach and grasp discontinuous supports, and the necessity for keeping a low center of mass for balance. A crouching posture is a means of resolving this conﬂict, but raises a new problem of potentially high locomotor stresses on the limbs due to large joint moments created by a substrate reaction force vector that is far removed from joint centers (Biewener, 1982, 1983, 1989; Schmitt, 1999). Primates, therefore, need ways to improve stability on branches and to attenuate high locomotor stresses on mobile, unstable limbs, especially the forelimbs. In order to achieve one or both of these goals, primates have evolved a suite of unusual locomotor characteristics, including greater reliance on hind limbs for both support and propulsion (Kimura et al., 1979; Reynolds, 1985; Kimura, 1985, 1992; Demes et al., 1994), use of a diagonal sequence/ diagonal couplets walking gait (Howell, 1944; Prost, 1965, 1969; Hildebrand, 1967; Rollinson and Martin, 1981; Vilensky, 1989; Vilensky and Larson, 1989), changes in muscle recruitment patterns (Lar- © 2004 WILEY-LISS, INC. © 2004 Wiley-Liss, Inc. Grant sponsor: NSF; Grant number: BCS-0109331. *Correspondence to: Susan G. Larson, Department of Anatomical Sciences, Stony Brook University School of Medicine, Stony Brook, NY 11794-8081. E-mail: firstname.lastname@example.org Received 26 March 2003; accepted 18 June 2003. DOI 10.1002/ajpa.10366 Published online 19 November 2003 in Wiley InterScience (www. interscience.wiley.com). 43 LIMB YIELD IN PRIMATES Fig. 1. Compliant gait model from Schmitt (1999). Schematic primate forelimb is shown undergoing a step using a stiff gait (solid line and open segments), and a compliant gait (dashed line and hatched segments). Compliant shoulder follows a ﬂatter path during contact, maintains contact for a longer time, covers a greater excursion angle, and displays signiﬁcant yield at elbow. A vertical force curve (solid line) for stiff limb has a short duration and high peak, while that for compliant limb (dashed line) has a longer time over which to develop force and thus yields lower peak vertical force. son and Stern, 1989; Larson, 1998), and use of relatively long stride lengths (Vilensky, 1980; Alexander and Maloiy, 1984; Reynolds, 1987) at relatively low stride frequencies (Alexander and Maloiy, 1984; Demes et al., 1990), brought about by relatively long limb bones (Alexander et al., 1979) and large limb angular excursions (Reynolds, 1987; Larson et al., 2000, 2001). Demes et al. (1990) proposed that increased step length and stride length allow a primate to achieve higher walking speeds while avoiding the potential for branch sway caused by a high-frequency gait. Increased step length can also lead to prolonged contact time on a substrate, which improves security and stability when traveling in a terminal branch habitat (Cartmill, 1985; Larson et al., 2001). In addition, increased step length and prolonged contact time, coupled with substantial limb yield, are components of a compliant gait (McMahon, 1985; McMahon et al., 1987), which Schmitt (1994, 1998, 1999) showed can reduce peak locomotor forces on limbs in primates (Fig. 1). He suggested that the use of a compliant gait to attenuate peak forces may have facilitated the invasion of the small-branch niche by the earliest members of the primate order by allowing them to use a crouched posture while having relatively long, mobile limbs. Schmitt (1999), however, examined only the forelimb, and although he suggested that the use of a compliant walking gait is rare among nonprimate mammals, relatively few comparative data exist. Indeed, some partially arboreal nonprimate mamma- lian species were found to display large limb angular excursions (Larson et al., 2001), which could contribute to longer step lengths, a component of a compliant gait. Unfortunately, it has yet to be demonstrated whether any of these partially arboreal species display the limb yield that is essential for a gait to be truly compliant (Larson et al., 2001). The goal of the present study is to test the assertion by Schmitt (1999) of the general use of a compliant gait in primates in contrast to a more stiff gait in nonprimate mammals, through documentation of elbow and knee yield in a variety of mammalian species. In the course of this study, the following questions will be addressed: 1) Is forelimb compliance (as reﬂected in elbow yield) universal in primates? 2) Is the primate hind limb also compliant? 3) Are primates the only order to exhibit signiﬁcant limb yield indicative of a compliant walking gait? 4) Within primates (and other mammalian groups that may be found to exhibit limb compliance), is there variability in limb yield, and if so, is greater limb yield associated with arboreality, supporting the proposal that limb compliance evolved as an accommodation to terminal branch use? METHODS Limb positional data were derived from videotapes of captive animals taken at the following zoos and research centers: Bronx Zoo, New York, NY; Cleveland Zoo, Cleveland, OH; Pittsburgh Zoo, Pittsburgh, PA; San Diego Zoo, San Diego, CA; Duke Primate Center, Durham, NC; Primate Locomotion Laboratory, Stony Brook, NY; and Laboratoire d’Ecologie Générale, CNRS/Unité de Recherche 1183, Brunoy, France. Most of the videotapes were recorded by one of the authors (S.G.L.) with a Panasonic AG 195 camcorder equipped with an electronic shutter (1/500 sec) at 60 ﬁelds/sec. Additional video was recorded by Pierre Lemelin using a Sony HandyCam Super 8-mm camcorder, and by Mark Hamrick using a Panasonic PV 900 camcorder. In most cases, subjects were 5 or more meters from the camera, minimizing parallax problems (Plagenhoef, 1979; Winter, 1990). Video analysis was performed on a Panasonic AG-7500 VCR, using frame-by-frame playback (60 frames/sec) to identify steps in which the subject was traveling in a relatively straight path, approximately perpendicular to the line of sight of the camera. Only walking gaits were analyzed of steps on substrates inclined no more than 20° to the horizontal. As described in Larson et al. (2000, 2001), the nature of the data source makes it impossible to control for speed or substrate type, or to correct for any out-of-plane motion. The latter was especially problematic in regard to documenting knee angles in lorisids, which display considerable hind limb rotation during a step. We were therefore unable to collect knee angle data for Perodicticus potto or Nycticebus pygmaeus, but were able to substitute three-dimensional (3D) elbow and knee angles for Loris tardigradus and N. coucang, gener- 44 E. LARNEY AND S.G. LARSON Fig. 2. Drawing of brown lemur Eulemur fulvus walking on a pole, taken from video records to illustrate limb angles reported here. Upper row shows elbow angles taken at forelimb touchdown (TD), midsupport (Mid), and liftoff (LO). Lower row shows knee angles taken at corresponding points for hind limb. Limb yield is deﬁned as change in angle from TD to Mid; limb re-extension is change from Mid to LO. ously given to us by Nancy Stevens who collected biplanar kinematic data on these species (for a description of her methodology, see Stevens, 2003). Video images were imported into a Pentium microcomputer with a MiroVIDO DC30 Plus (Pinnacle Systems) frame grabber board, and digitized for quantitative analysis using Peak Motus 2000 Software (Peak Performance Technologies, Inc.). Elbow angle was deﬁned as the anterior angle between a midline axis of the arm (approximation of line between shoulder joint center and lateral epicondyle of humerus) and forearm (approximation of line between lateral epicondyle of humerus to center of wrist) of the forelimb, and knee angle was the posterior angle between a midline axis of the thigh (approximation of line between hip joint center and lateral epicondyle of femur) and leg (approximation of line between lateral epicondyle of femur to center of ankle) of the hind limb (Fig. 2). Elbow and knee angles were documented at touchdown (TD), the ﬁrst frame in which the limb is in contact with a support; midsupport (Mid), the point when wrist or ankle was immediately below the shoulder or hip;1 and liftoff (LO), the termination of elbow or knee extension near the end of stance phase, prior to the limb ﬂexion associated with lifting the hand or foot off the ground for swing phase (Fig. 2). We elected to document elbow and knee angles at the latter point rather than when the hand or foot actually lost contact with the substrate, since our goal was to document the magnitude of change in limb ﬂexion through stance phase, and often hand or foot release involves renewed ﬂexion of the limb in preparation for swing phase. Only quadrupedal mammalian species with an average body weight under 25 kg were included in the study, since this encompasses the 1 According to data published by Schmitt (1994, 1998), midsupport as deﬁned here is the point of maximum elbow or knee ﬂexion in primates. We are making the assumption that this is also true for nonprimates. range of most primate species (Table 1). Efforts were made to quantify at least 10 steps of both the forelimb and hind limb for each animal subject. However, this was not always possible (Table 1). Limb yield was calculated as the change in angle from TD to Mid, and re-extension from Mid to LO for both forelimbs and hind limbs. Data were analyzed in SPSS (version 11 for Windows). To avoid bias due to unequal numbers of steps for different species, species means were ﬁrst derived, and group means for each mammalian order were computed from species means. The nonparametric Kruskal-Wallis Test for independence of samples was used to compare group means (Conover, 1999). We chose to run this nonparametric statistical test because our small sample size did not meet certain criteria of parametric statistics. Correlation analysis using the nonparametric Spearman’s rho statistic on ranked data was used to explore for relations to overall body size and between different variables. RESULTS Mean elbow and knee angles at TD, Mid, and LO for each group are reported in Table 2, and are displayed along with error bars (mean ⫾ 2 SE) in Figure 3a,b. Interestingly, primate forelimbs and hind limbs are more extended at nearly every stage of the step cycle than in other mammalian groups. In regard to limb yield, all groups display some decrease in elbow or knee angle at midphase; however, this decrease is most marked in primates and marsupials. To better illustrate these changes in limb angles, mean yield (TD[3:Directions MT/86]Mid) and reextension (Mid[3:Directions MT/86]LO) angles for the ﬁve groups are displayed in Figure 3c,d. The Kruskal-Wallis tests for independence of samples for forelimb and hind limb yield were signiﬁcant at P ⬍ 0.01. Primates display the greatest degree of elbow yield, but are followed closely by marsupials. Carnivores and rodents and others exhibit a modest amount of elbow yield, while artiodactyls show the least elbow ﬂexion. Post hoc tests indicate that primates are signiﬁcantly different from all groups except marsupials. For the hind limb, marsupials actually exceed primates in knee yield, although the value for primates is also quite high. As with the forelimb, carnivores, rodents, and others display a modest amount of knee yield, and artiodactyls again exhibit the smallest amount of knee ﬂexion. Post hoc tests again indicate that primates are signiﬁcantly different from all groups except marsupials. Comparing forelimbs to hind limbs, carnivores, marsupials, and artiodactyls all display slightly greater knee yield than elbow yield. For primates and rodents and others, the reverse situation obtains, but in all cases, the differences are fairly small. Several individual species outside the primate and marsupial groups exhibit substantial limb yield that is masked in the ordinal averages. The fossa 45 LIMB YIELD IN PRIMATES TABLE 1. List of species studied Taxon Carnivores (n ⫽ 10) Ailurus fulgens Arctictis binturong Cryptoprocta ferox Eira barbara Felis caracal Leopardus wiedi Nasua nasua Neofelis nebulosa Suricata suricatta Vulpes corsac Marsupials (n ⫽ 5) Caluromys philander Dasyurus maculatus Dendrolagus matschiei Phascolarctos cinereus Primates Anthropoids Cercopithecoids (n ⫽ 15) Cercocebus galeritus Cercopithecus ascanius Cercopithecus diana Cercopithecus lhoesti Cercopithecus neglectus Chlorocebus aethiops Colobus angolens Erythrocebus patas Nasalis lavartus Macaca mulatta Mandrillus sphinx Papio anubis Trachypithecus auratus Trachypithecus cristata Trachypithecus francoisi Ceboids (n ⫽ 7) Alouatta seniculus Aotus trivirgatus Cebuella pygmaea Lagothrix lagothricha Pithecia pithecia Saguinus geoffroyi Saimiri sciureus Prosimians Lemuroids (n ⫽ 5) Daubentonia madagascariensis Eulemur fulvus Hapalemur griseus Lemur catta Lorisoids (n ⫽ 4) Loris tardigradus Nycticebus coucang Nycticebus pygmaeus Perodicticus potto Rodents and others (n ⫽ 6) Dasyprocta aguti Dinomys branickii Dolichotis patagonum Myoprocta acouchy Tupaia glis Artiodactyls (n ⫽ 5) Gazella dorcas Madoqua guentheri Muntiacus muntjac Neotragus moschatus Sylvicapra grimmia 1 Body size (kg) No. of steps FL No. of steps HL 4.00 11.50 9.50 4.50 16.00 2.501 4.00 15.201 0.79 5.50 3 6 5 8 9 2 3 6 7 13 0 0 3 4 6 0 3 0 7 10 0.30 2.50 7.20 9.30 6 8 6 16 6 4 0 12 7.40 3.31 4.55 5.901 5.501 3.62 8.62 9.45 20.40 9.90 22.25 19.20 7.102 6.18 7.52 13 3 15 1 2 11 4 11 7 12 18 14 9 6 4 2 2 9 0 0 11 3 12 4 14 13 14 7 5 3 Red howler monkey Owl monkey Pygmy marmoset Woolly monkey White-faced saki Geoffroy’s tamarin Squirrel monkey 5.95 0.77 0.12 7.15 1.76 0.49 0.72 17 11 2 10 5 6 8 10 10 2 11 9 5 8 Aye aye Brown lemur Gentle lemur Ring-tailed lemur 3.00 2.21 0.94 2.21 13 14 13 14 0 9 10 9 Slender loris Slow loris Pygmy slow loris Potto 0.19 0.65 0.42 1.10 304 304 9 9 304 304 0 0 Orange-rumped agouti Pacarana Patagonian cavy Acouchy Tree shrew 2.65 12.50 12.50 0.95 0.17 16 3 12 3 2 13 2 10 4 2 Dorcas gazelle Guenther’s dik dik Muntjac Suni Gray duiker 17.50 4.60 16.00 6.50 18.50 11 2 2 6 2 9 0 2 7 3 Common name Red panda Binturong Fossa Grey headed tayra Caracal Margay Coati Clouded leopard Meerkat Corsac fox Woolly opossum Quoll Tree kangaroo Koala Tana River mangabey Red-tailed monkey Diana monkey L’Hoest’s guenon DeBrazza’s monkey Vervet monkey Angolian colobus monkey Patas monkey Proboscis monkey Rhesus macaque Mandrill Olive baboon Ebony langur Silvered langur Francois langur Body size estimate from Nowak (1999). Body size estimate from Rowe (1996); otherwise from Larson et al. (2001). 3 Data from Krakauer et al. (2002). 4 Data from Stevens (2003). 2 46 128.90 ⫾ 2.20 126.63 ⫾ 4.27 122.74 ⫾ 2.71 117.51 ⫾ 4.72 113.60 ⫾ 4.14 10.86 ⫾ 1.82 7.32 ⫾ 4.55 17.97 ⫾ 7.86 6.91 ⫾ 3.89 1.87 ⫾ 4.83 The goal of this project was to document change in elbow and knee angles through a walking step in a variety of primates and nonprimate mammals. This was done by documenting angles at three discrete points in the step cycle, and calculating the changes Standard error. 121.93 ⫾ 3.36 117.61 ⫾ 3.07 101.61 ⫾ 1.55 107.57 ⫾ 3.84 103.01 ⫾ 2.80 22.60 ⫾ 1.98 9.66 ⫾ 2.89 19.94 ⫾ 6.98 11.64 ⫾ 2.96 2.54 ⫾ 3.93 144.53 ⫾ 1.991 127.27 ⫾ 3.59 121.55 ⫾ 8.26 119.20 ⫾ 3.66 105.54 ⫾ 4.75 Primates Carnivores Marsupials Rodents and others Artiodactyls DISCUSSION 1 Re-extension Mid 118.03 ⫾ 3.19 119.31 ⫾ 1.95 104.77 ⫾ 5.15 110.60 ⫾ 5.33 111.74 ⫾ 6.24 22.02 ⫾ 1.98 12.03 ⫾ 3.36 22.56 ⫾ 3.67 8.77 ⫾ 2.68 6.16 ⫾ 2.37 140.05 ⫾ 3.17 131.35 ⫾ 4.57 127.33 ⫾ 3.79 117.90 ⫾ 6.65 117.90 ⫾ 6.65 142.22 ⫾ 2.27 127.89 ⫾ 2.57 109.27 ⫾ 3.80 123.38 ⫾ 5.22 118.73 ⫾ 7.43 Hind limb Yield TD LO Re-extension Mid Forelimb Yield TD Group TABLE 2. Group mean elbow and knee angles (in degrees) at touchdown, midstance, and liftoff (Cryptoprocta ferox) and tree shrew (Tupaia glis) both display elbow yield around 20°, comparable to that of most primates. The margay (Felis wiedi) also displays substantial elbow yield. For the hind limb, only the tayra (Eira barbara) displays an unusual level of knee yield. Mean touchdown, midsupport, and liftoff angles for primates grouped into superfamilies are reported in Table 3 and displayed in Figure 4a,b; Figure 4c,d shows the mean changes in elbow and knee angles for each of these groups. Elbow yield is most extreme in lorisoids, followed by ceboids and lemuroids, and is least in cercopithecoids (test for independence of samples was signiﬁcant at P ⫽ 0.01). Conversely, the amount of knee yield is highest in ceboids and cercopithecoids, and least in lorisoids (including data only for L. tardigradus and N. coucang; see Methods). However, the test for independence of samples among primate superfamilies was not signiﬁcant for knee yield. All primates but cercopithecoids ﬂex their elbows more than marsupials on average, but knee yield in marsupials is comparable to that of the anthropoid species. Correlation analysis across the primate sample reveals a tendency for larger species to have more extended elbows and knees throughout a step, as predicted by Biewener (1982, 1983, 1989; see also Polk, 2002), and to display smaller changes in elbow or knee angles (with the exception of knee yield).2 There are also signiﬁcant correlations between the degree of elbow yield and re-extension, although not between knee yield and re-extension. Since many of the variables display some correlation to body size, partial correlation analysis (on ranked data) was used to explore for intervariable relationships, controlling for body size. Most of the intervariable correlations do in fact appear to be due to their mutual correlations to overall size. As might be expected, there are signiﬁcant negative partial correlations between midsupport elbow angle and forelimb yield (r ⫽ ⫺0.59, P ⬍ 0.01), indicating that at any body size, the less extended the elbow is at midsupport, the greater the degree of yield. Midsupport knee angle and hind limb yield did not display a signiﬁcant partial correlation, although hind limb yield and forelimb yield are correlated to each other (r ⫽ 0.49, P ⫽ 0.02). 20.29 ⫾ 1.67 10.28 ⫾ 2.44 7.66 ⫾ 3.44 15.81 ⫾ 3.89 15.73 ⫾ 6.02 LO E. LARNEY AND S.G. LARSON 2 Spearman’s rank correlation coefﬁcients to body weight: FL TD, r ⫽ 0.48; FL Mid, r ⫽ 0.75; FL LO, r ⫽ 0.63; FL Yield, r ⫽ ⫺0.77; FL Re-Ext, r ⫽ ⫺0.72; HL TD, r ⫽ 0.75; HL Mid, r ⫽ 0.75; HL LO, r ⫽ 0.51; HL Yield, r ⫽ 0.05; HL Re-Ext, r ⫽ ⫺0.65; all except HL Yield are signiﬁcant at P ⬍ 0.01. LIMB YIELD IN PRIMATES 47 Fig. 3. Error bars represent group means ⫾ 2 standard errors of mean for each order. a: Error bars and means for elbow angles at TD (light gray), Mid (black), and LO (dark gray). b: Error bars and means for knee angles at TD (light gray), Mid (black), and LO (dark gray). On average, primates display more extended limbs at nearly every stage of a step cycle than other mammals. c: Error bars and means for elbow yield (black) and re-extension (gray). d: Error bars and means for knee yield (black) and re-extension (gray). Average limb yield is high in primates and marsupials, modest in carnivores and rodents and others, and low in artiodactyls. in angle from touchdown to midsupport, and from midsupport to liftoff. Our results demonstrate that substantial yield at the elbow is universal among primates, and that most primates, except lorisoids, also display marked knee yield. To the degree that limb yield is indicative of the use of a compliant gait, these results support the assertion of Schmitt (1994, 1998, 1999) that primates as a group are characterized by use of a compliant gait. However, primates are not unique in displaying limb yield: marsupials also exhibit signiﬁcant yield at both the elbow and knee, and carnivores and rodents display modest limb yield. Only artiodactyls appear to use a genuinely stiff (absence of yield) walking gait. They begin a step with both their elbows and knees in ﬂexed postures, and while the elbow undergoes re-extension in the second half of support phase, the knee angle remains largely unchanged through a step. It is possible, however, that artiodactyls may exhibit more yield and/or re-extension at their ankle joint (hock) rather than knee. Within the primate order, there was variation in limb yield and re-extension, especially for the hind limb. Although larger species tend to display more extended limbs and less elbow yield and limb reextension, most larger species are cercopithecoids and smaller species are prosimians, making it difﬁcult to separate size from phylogenetic inﬂuences. In addition, some of the largest species in the sample are terrestrial quadrupeds, adding the confounding effect of substrate differences. The fact that all primate groups except cercopithecoids typically display more forelimb yield than hind limb yield supports suggestions that the dual role of the forelimb for both locomotion and foraging/manipulation makes it especially vulnerable to high locomotor stresses, and therefore, it is especially advantageous if peak substrate reaction forces can be attenuated to some degree, such as through the use of a compliant gait (Schmitt, 1994, 1998, 1999). However, this raises the question as to why cercopithecoids do not follow the same pattern. Of the taxa examined here, cercopithecoids are unusual in that they comprise the only group that includes terrestrial and semiterrestrial species. In their display of greater hind limb than forelimb yield, cercopithecoids are more like terrestrial nonprimate quadrupeds. Current views on cercopithecoid origins suggest that Victoriapithecus, a potential common ancestor of colobines and cercopithecines, may have indeed been semiterrestrial (von Konenigswald, 1969; Simons, 1972; McCrossin and Beneﬁt, 1992, 1994; McCrossin et al., 1998; Beneﬁt and McCrossin, 2002). Therefore, modiﬁcations for some measure of terrestriality, including emphasis on stability at the shoulder and rapid ﬂexion/extension at the elbow (McCrossin et al., 1998), may have been maintained to varying degrees among extant cercopithecoids, despite the fact that many are now largely or exclusively arboreal. A modest degree of elbow yield during walking may be a similar retention. Finally, marsupials exhibit comparable levels of limb yield to primates, and they were shown to be second only to primates in limb excursion angles 48 6.96 ⫾ 2.19 13.44 ⫾ 4.08 16.31 ⫾ 2.60 19.07 ⫾ 3.92 Standard Error. 1 LO Re-extension Mid 127.09 ⫾ 3.71 109.29 ⫾ 6.16 110.31 ⫾ 1.63 101.36 ⫾ 2.47 23.43 ⫾ 2.99 25.34 ⫾ 2.63 18.08 ⫾ 3.34 7.09 ⫾ 2.20 150.52 ⫾ 1.98 134.63 ⫾ 5.68 128.40 ⫾ 1.87 108.45 ⫾ 0.28 151.12 ⫾ 2.20 133.36 ⫾ 3.72 132.86 ⫾ 4.74 133.76 ⫾ 4.56 15.82 ⫾ 1.56 21.91 ⫾ 2.10 22.54 ⫾ 3.72 31.99 ⫾ 7.54 16.21 ⫾ 2.11 27.53 ⫾ 4.04 26.90 ⫾ 3.88 33.62 ⫾ 4.64 151.51 ⫾ 1.371 138.98 ⫾ 2.64 137.22 ⫾ 5.79 135.38 ⫾ 8.44 Cercopithecoids Ceboids Lemuroids Lorisoids 135.30 ⫾ 2.57 111.45 ⫾ 5.17 110.31 ⫾ 7.92 101.77 ⫾ 8.71 Hind limb Yield TD LO Re-extension Forelimb Mid Yield TD Superfamily TABLE 3. Primate mean elbow and knee angles (in degrees) at touchdown, midstance, and liftoff 134.05 ⫾ 2.60 122.73 ⫾ 5.01 126.62 ⫾ 1.05 120.44 ⫾ 1.45 E. LARNEY AND S.G. LARSON (Larson et al., 2000, 2001), which could indicate increased step length, both important aspects of a compliant gait. However, conclusions about marsupial gait based on the data presented here should be tempered by the fact that our sample size is small and includes two highly arboreal species, the woolly opossum (Caluromys philander), which was already shown to display morphological and behavioral convergences with primates (Lemelin, 1996, 1999; Lemelin and Schmitt, 2001; Lemelin et al., 2002; Schmitt and Lemelin, 2002), and the koala (Phascolarctos cinereus). With this said, the dramatic yield seen in these marsupial species and primates does tend to support a link between the use of a compliant gait and arboreal habits (Schmitt, 1994, 1998, 1999). This link is further supported by the individual species outside of primates and marsupials that display unusual changes in limb angles, such as the fossa (Cryptoprocta ferox), which was described as extremely agile in arboreal environments (Garbutt, 1999), as well as the margay (Felis wiedi) (Konecny, 1989), tayra (Eira barbara) (Nowak, 1999), and tree shrew (Tupaia glis) (LeGros Clark, 1959), which are all regular tree dwellers. Schmitt (1994, 1998, 1999) suggested that the use of a compliant gait is a derived character for primates related to their arboreal origins. If this is correct, then the high limb yield reported here for marsupials would be a case of convergence. Alternatively, limb yield and use of a compliant gait could be primitive retentions in primates as well as marsupials. Whether the origins of mammalian posture and locomotion are related to terrestriality or arboreality has long been a topic of debate (for an overview, see Jenkins, 1974), and which view is adopted impacts upon whether the arboreal habits of primates are viewed as a departure from the primitive mammalian condition or an elaboration of it. Theories on primitive mammalian locomotor adaptation have suffered from a lack of sufﬁciently old fossils representing common ancestral forms (Jenkins, 1974). However, a recently discovered fossil Eomaia scansoria from the early Cretaceous period in China is believed to be the oldest known Eutherian mammal (Ji et al., 2002). Its elongated ﬁngers and toes have been interpreted to reﬂect climbing abilities and arboreal habits (Ji et al., 2002), supporting an arboreal origin for mammalian postural and locomotor characteristics. In addition, primitive members of the marsupial lineage are also believed to have been climbers (Kielan-Jaworowska et al., 1979). Therefore, if the use of a compliant gait is indeed an adaptation to arboreal habits, it may have arisen early in mammalian evolution, and limb yield could be a primitive retention in primates as well as marsupials. However, this would not alter the fact that as a mechanism for reducing peak forces on the limbs, a compliant gait would have facilitated the invasion and subsequent radiation of basal primates into a terminal branch niche. LIMB YIELD IN PRIMATES 49 Fig. 4. Error bars represent group means ⫾ 2 standard errors of mean for different primate superfamilies (marsupials included for comparison). a: Error bars and means for elbow angles at TD (light gray), Mid (black), and LO (dark gray). b: Error bars and means for knee angles at TD (light gray), Mid (black), and LO (dark gray). On average, cercopithecoids display more extended limbs at every stage of a step cycle than other primate groups. c: Error bars and means for elbow yield (black) and re-extension (gray). d: Error bars and means for knee yield (black) and re-extension (gray). Average elbow yield is higher than knee yield in all primate groups except cercopithecoids. ACKNOWLEDGMENTS This study is based on videotapes of animals at the following institutions: Bronx Zoo, New York, NY; Cleveland Zoo, Cleveland, OH; Pittsburgh Zoo, Pittsburgh, PA; San Diego Zoo, San Diego, CA; Duke Primate Center, Durham, NC; Primate Locomotion Laboratory, Stony Brook, NY; and Laboratoire d’Ecologie Générale, CNRS/Unité de Recherche 1183, Brunoy, France. Thanks go to Pierre Lemelin and Mark Hamrick, who collected some of the video, to Luci Betti-Nash for the preparation of Figure 2, and to Anthony Olejniczak for the development of a post hoc statistical test for Kruskal-Wallis results. LITERATURE CITED Alexander RM, Jayes AS, Maloiy GMO, Wathuta EM. 1979. Allometry of the limb bones of mammals from shrews (Sorex) to elephant (Loxodonta). J Zool Lond 189:305–314. Alexander RM, Maloiy GMO. 1984. Stride lengths and stride frequencies of primates. J Zool Lond 202:577–582. Beneﬁt BR, McCrossin ML. 2002. The Victoriapithecidae, Cercopithecoidea. In: Hartwig WC, editor. The primate fossil record. Cambridge: Cambridge University Press. p 241–253. Biewener AA. 1982. Bone strength in small mammals and bipedal birds: do safety factors change with body size? J Exp Biol 98:289 –301. Biewener AA. 1983. Allometry of quadrupedal locomotion: the scaling of duty factor, bone curvature and limb orientation to body size. J Exp Biol 105:147–171. Biewener AA. 1989. Scaling body support in mammals: limb posture and muscle mechanics. Science 245:45– 48. Cartmill M. 1972. Arboreal adaptations and the origin of the order Primates. In: Tuttle RH, editor. The functional and evolutionary biology of primates. Chicago: Aldine-Atheton. p 3–35. Cartmill M. 1974. Rethinking primate origins. Science 184:436 – 443. Cartmill M. 1985. Climbing. In: Hildebrand M, Bramble DM, Liem KF, Wake DB, editors. Functional vertebrate morphology. Cambridge: Belknap Press. p 73– 88. Cartmill M. 1992. New views on primate origins. Evol Anthropol 1:105–111. Conover WJ. 1999. Practical nonparametric statistics, 3rd ed. New York: John Wiley and Sons. Demes B, Jungers WL, Nieschalk U. 1990. Size- and speedrelated aspects of quadrupedal walking in slender and slow lorises. In: Jouffroy FK, Stack MH, Niemitz C, editors. Gravity, posture and locomotion in primates. Florence: Il Sedicesimo. p 175–197. Demes B, Larson SG, Stern JT Jr, Jungers WL, Biknevicius AR, Schmitt D. 1994. The kinetics of primate quadrupedalism: “hindlimb drive” reconsidered. J Hum Evol 26:353–374. Fleagle JG. 1999. Primate adaptation and evolution. 2nd ed. London: Academic Press. Garbutt N. 1999. Mammals of Madagascar. New Haven: Yale University Press. Hamrick MW. 1998. Functional and adaptive signiﬁcance of primate pads and claws: evidence from New World anthropoids. Am J Phys Anthropol 106:113–127. Hildebrand M. 1967. Symmetrical gaits of primates. Am J Phys Anthropol 26:119 –130. Howell AB. 1944. Speed in animals. Chicago: University of Chicago Press. Jenkins FA Jr. 1974. Tree shrew locomotion and the origins of primate arborealism. In: Jenkins FA Jr, editor. Primate locomotion. New York: Academic Press. p 85–115. Ji Q, Luo Z, Yuan C, Zhang J, Georgi J. 2002. The earliest known eutherian mammal. Science 416:816 – 822. Kielan-Jaworowska Z, Bown TM, Lillegraven JA. 1979. Eutheria. In: Lillegraven JA, Kielan-Jaworowska Z, Clemens WA, editors. Mesozoic mammals. Berkeley: University of California Press. p 221–258. Kimura T. 1985. Bipedal and quadrupedal walking of primates: comparative dynamics. In: Kondo S, editor. Primate morphophysiology, locomotor analyses and human bipedalism. Tokyo: University of Tokyo Press. p 81–104. 50 E. LARNEY AND S.G. LARSON Kimura T. 1992. Hind limb dominance during primate high-speed locomotion. Primates 33:465– 476. Kimura T, Okada M, Ishida H. 1979. Kinesiological characteristics of primate walking: its signiﬁcance in nonhuman walking. In: Morbeck ME, Preuschoft H, Gomberg N, editors. Environment, behavior, and morphology: dynamic interactions in primates. New York: Gustav Fischer. p 297–311. Konecny MJ. 1989. Movement patterns and food habits of four sympatric carnivore species in Belize, Central America. In: Redford KH, Eisenberg JF, editors. Advances in neotropical mammalogy. Gainesville: Sandhill Crane Press. p 234 –264. Krakauer E, Lemelin P, Schmitt D. 2002. Hand and body position during locomotor behavior in the aye-aye (Daubentonia madagascariensis). Am J Primatol 57:105–118. Larson SG. 1998. Unique aspects of quadrupedal locomotion in nonhuman primates. In: Strasser E, Fleagle J, Rosenberger A, McHenry H, editors. Primate locomotion: recent advances. New York: Plenum Press. p 157–173. Larson SG, Stern JT Jr. 1989. The role of propulsive muscles of the shoulder during quadrupedalism in vervet monkeys (Cercopithecus aethiops): implications for neural control of locomotion in primates. J Motor Behav 21:457– 472. Larson SG, Schmitt D, Lemelin P, Hamrick MW. 2000. Uniqueness of primate forelimb posture during quadrupedal locomotion. Am J Phys Anthropol 112:87–101. Larson SG, Schmitt D, Lemelin P, Hamrick M. 2001. Limb excursion during quadrupedal walking: how do primates compare to other mammals? J Zool Lond 255:353–365. LeGros Clark WE. 1959. The antecedents of man. Edinburgh: Edinburgh University Press. Lemelin P. 1996. The evolution of manual prehensility in primates: a comparative study of prosimians and didelphid marsupials. Ph.D. thesis, Stony Brook University. Lemelin P. 1999. Morphological correlates of substrate use in didelphid marsupials: implications for primate origins. J Zool Lond 247:165–175. Lemelin P, Schmitt D. 2001. Convergence in gait mechanics between primates and woolly opossums: the origins of primate locomotion can be traced back to a ﬁne branch niche. Am J Phys Anthropol [Suppl] 32:97. Lemelin P, Schmitt D, Cartmill M. 2002. The origins of diagonalsequence walking gaits in primates: an experimental test involving two didelphid marsupials. Am J Phys Anthropol [Suppl] 34:101. McCrossin ML, Beneﬁt BR. 1992. A comparative assessment of the ischial morphology of Victoriapithecus macinnesi. Am J Phys Anthropol 87:277–290. McCrossin ML, Beneﬁt BR. 1994. Maboko Island and the evolutionary history of Old World monkeys and apes. In: Corruccini RS, Ciochon RL, editors. Integrative paths to the past: paleoanthropological advances in honor of F.C. Howell. New York: Prentice-Hall. p 95–112. McCrossin ML, Beneﬁt BR, Gitau SN, Palmer AK, Blue KT. 1998. Fossil evidence for the origins of terrestriality among Old World higher primates. In: Strasser E, Fleagle J, Rosenberger A, McHenry H, editors. Primate locomotion: recent advances. New York: Plenum Press. p 53–396. McMahon TA. 1985. The role of compliance in mammalian running gaits. J Exp Biol 115:263–282. McMahon TA, Valiant G, Frederick EC. 1987. Groucho running. J Appl Physiol 62:2326 –2337. Nowak RM. 1999. Walker’s mammals of the world, 6th ed. Volume I. Baltimore: Johns Hopkins University Press. Plagenhoef S. 1979. Dynamics of human and animal motion. In: Morbeck M, Preuschoft H, Gomberg N, editors. Environment, behavior, and morphology: dynamic interactions in primates. New York: Gustav Fisher. p 95–118. Polk JD. 2002. Adaptive and phylogentic inﬂuences on musculoskeletal design in cercopithecine primates. J Exp Biol 205: 3399 –3412. Prost JH. 1965. The methodology of gait analysis and gaits in monkeys. Am J Phys Anthropol 23:215–240. Prost JH. 1969. A replication study on monkey gaits. Am J Phys Anthropol 30:203–208. Rasmussan DT. 1990. Primate origins: lessons from a neotropical marsupial. Am J Primatol 23:209 –223. Reynolds TR. 1985. Stresses on the limbs of quadrupedal primates. Am J Phys Anthropol 67:351–362. Reynolds TR. 1987. Stride length and its determinants in humans, early hominids, primates, and mammals. Am J Phys Anthropol 67:351–362. Rollinson J, Martin RD. 1981. Comparative aspects of primate locomotion, with special reference to arboreal cercopithecines. In: Day MH, editor. Vertebrate locomotion. Symposia of the Zoological Society of London, no. 48. London: Academic Press. p 377– 427. Rowe N. 1996. The pictorial guide to the living primates. East Hampton: Poganias Press. Schmitt D. 1994. Forelimb mechanics as a function of substrate type during quadrupedalism in two anthropoid primates. J Hum Evol 26:441– 458. Schmitt D. 1998. Forelimb mechanics during arboreal and terrestrial quadrupedalism in Old World monkeys. In: Strasser E, Fleagle J, Rosenberger A, McHenry H, editors. Primate locomotion: recent advances. New York: Plenum Press. p 175–200. Schmitt D. 1999. Compliant walking in primates. J Zool Lond 248:149 –160. Schmitt D, Lemelin P. 2002. Origins of primate locomotion: gait mechanics of the woolly opossum. Am J Phys Anthropol 118: 231–238. Simons EL. 1972. Primate evolution: an introduction to man’s place in nature. New York: Macmillan Publishing Co., Inc. Stevens N. 2003. The effects of substrate size, orientation and compliance upon prosimian arboreal quadrupedalism. Doctoral dissertation, Stony Brook University. Sussman RW. 1991. Primate origins and the evolution of angiosperms. Am J Primatol 23:209 –223. Vilensky JA. 1980. Trot-gallop transition in a macaque. Am J Phys Anthropol 53:347–348. Vilensky JA. 1989. Primate quadrupedalism: how and why does it differ from that of typical quadrupeds? Brain Behav Evol 34: 357–364. Vilensky JA, Larson SG. 1989. Primate locomotion: utilization and control of symmetrical gaits. Annu Rev Anthropol 18:17– 35. von Konenigswald GHR. 1969. Miocene Cercopithecoidea and Oreopithecoidea from the Miocene of East Africa. In: Leakey LSB, editor. Fossil vertebrates of Africa, volume 1. London: Academic Press. p 39 –51. Winter DA. 1990. Biomechanics and motor control of human movement. New York: John Wiley and Sons.