Brief communication Dynamic plantar pressure distribution during locomotion in Japanese macaques (Macaca fuscata).код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 142:149–156 (2010) Brief Communication: Dynamic Plantar Pressure Distribution during Locomotion in Japanese Macaques (Macaca fuscata) Eishi Hirasaki,* Yasuo Higurashi, and Hiroo Kumakura Laboratory of Biological Anthropology, Graduate School of Human Sciences, Osaka University, Suita, Osaka 565-0871, Japan KEY WORDS foot function; bipedal walking; quadrupedal walking; kinetics ABSTRACT To better place the form and motion of the human foot in an evolutionary context, understanding how foot motions change when quadrupeds walk bipedally can be informative. For this purpose, we compared the pressures beneath the foot during bipedal and quadrupedal walking in Japanese macaques (Macaca fuscata). The pressure at nine plantar regions was recorded using a pressure mat (120 Hz), while the animals walked on a level walkway at their preferred speeds. The results revealed substantial differences in foot use between the two modes of locomotion, and some features observed during bipedal walking resembled human gait, such as the medial transfer of the center of pressure (COP), abrupt declines in forefoot pressures, and the increased pressure beneath the hallux, all occurring during the late-stance phase. In particular, the medial transfer of the COP, which is also observed in bonobos (Vereecke et al.: Am J Phys Anthropol 120 (2003) 373–383), was due to a biomechanical requirement for a hind limb dominant gait, such as bipedal walking. Features shared by bipedal and quadrupedal locomotion that were quite different from human locomotion were also observed: the heel never contacted the ground, a foot longitudinal arch was absent, the hallux was widely abducted, and the functional axis was on the third digit, not the second. Am J Phys Anthropol 142:149–156, 2010. V 2009 Wiley-Liss, Inc. Obligate bipedal locomotion is a feature that separates humans from other primates. To better understand the mechanics of human bipedal locomotion, many experimental studies have examined bipedal and quadrupedal walking in nonhuman primates (e.g., Jenkins, 1972; Ishida et al., 1974; Kimura et al., 1979; Yamazaki et al., 1979; Stern and Susman, 1981; Shapiro and Jungers, 1988; Aerts et al., 2000; D’Août et al., 2004). The biomechanical conditions required for the evolution of this type of gait, however, are not fully understood, partly because quantitative data are lacking on the motion of the foot during locomotion. Because the foot performs a pivotal role in the interaction between the body and the substrate during locomotion (Sigmon and Farslow, 1986), studies have focused on the morphology and function of the foot, and a great deal of information has accumulated from studies on primate feet and fossil evidence from early hominids (e.g., Susman, 1983; Lewis, 1989; Gebo, 1993a; Vereecke et al., 2005a). Foot kinematics and kinetics during primate locomotion have also been studied, based mostly on qualitative observations (Elftman and Manter, 1935; Okada, 1985; Meldrum, 1991, 1993; Gebo, 1992, 1993a,b; Schmitt and Larson, 1995), and they have contributed to understand the evolution of human locomotor strategies (Wunderlich, 1999). Quantitative comparisons of foot motion during bipedal and quadrupedal walking, however, are relatively scarce, with the exceptions of the studies by Wunderlich (1999, 2003) and Vereecke et al. (2003). To better understand the form and motion of the human foot in an evolutionary context, knowing how foot motions change when quadrupeds walk bipedally can be informative, because that is what happened when our ancestors stood up on the ground or when early hominoids stood up on a branch (Crompton et al., 2003, 2008). For this study, we obtained quantitative data on the pressure beneath the foot during quadrupedal and bipedal walking in Japanese macaques, emphasizing the differences between the two modes of locomotion. Although African apes may be more appropriate subjects for studying the evolution of bipedalism (Vereecke et al., 2003, 2004, 2005b), we reduced the effects of synapomorphy by using macaques, which are phylogenetically farther from humans than African apes. Consequently, any humanlike features found in the bipedal gait of macaques can be more plausibly attributed to the biomechanical requirements of bipedal walking rather than to synapomorphy (Hirasaki et al., 2004). Therefore, we measured foot pressures in macaques during bipedal walking and examined how they differed from those during quadrupedal walking. C 2009 V WILEY-LISS, INC. C Grant sponsor: Japan Society for the Promotion of Science; Grant number: 14704005. *Correspondence to: Eishi Hirasaki, Laboratory of Biological Anthropology, Graduate School of Human Sciences, Osaka University 1-2 Yamadaoka, Suita, Osaka 565-0871, Japan. E-mail: email@example.com Received 19 February 2009; accepted 28 October 2009 DOI 10.1002/ajpa.21240 Published online 21 December 2009 in Wiley InterScience (www.interscience.wiley.com). 150 E. HIRASAKI ET AL. TABLE 1. Gait parameters during the quadrupedal and bipedal sequences, in which pressure data were collected Quadrupedal (n 5 13) Bipedal (n 5 14) Signiﬁcance Walking speed (km/h) Stride length (m) Step frequency (Hz) Contact duration (s) Duty factor (%) 2.79 (60.17) 2.91 (60.14) n.s 0.73 (60.03) 0.68 (60.03) P \ 0.001 2.11 (60.16) 2.39 (60.10) P \ 0.001 0.60 (60.04) 0.54 (60.05) P \ 0.01 63.7 (62.5) 64.1 (62.3) n.s. Values are the mean 6 SD. Student’s t test was used to compare the results between quadrupedal and bipedal walking. During bipedal walking, only two limbs support the body, which is reﬂected in many aspects of the plantar pressure proﬁles. We predicted the following. First, the pressures would be substantially higher during bipedal walking than during quadrupedal walking; the higher step frequency of bipedal walking (Kimura, 1985; Aerts et al., 2000; Hirasaki and Kumakura, 2004) would also increase plantar pressures. Second, the hallux would be abducted more during bipedal walking to provide more stability. Third, the center of pressure (COP) would mediolaterally deviate more during bipedal walking than during quadrupedal walking, because the upper part of the body is supported only by one hind limb during the single-limb support phase of bipedal walking. In particular, the trunk moves laterally toward the side of the swinging hind limb at the end of the stance phase. Therefore, our third prediction can be rephrased as follows: the COP would deviate more medially at the end of stance phase during bipedal walking than during quadrupedal walking. MATERIALS AND METHODS We studied two adult Japanese macaques (Macaca fuscata), a 10.6-kg, 16-year-old male, and a 4.7-kg, 11-year-old, female. During bipedal and quadrupedal walking on the ﬂoor, the foot-pressure distribution was recorded using a sensor sheet (Big-mat 1/4, 22 3 24 cm2, 5-mm resolution; distributed in Japan by Nitta in cooperation with Tekscan, Boston, MA) at a sampling frequency of 120 Hz. The sheet was sufﬁciently large (440 3 480 mm2) to accommodate the feet of the macaques (length, L: L 5 157 mm and L 5 138 mm). To examine steady-state locomotion, the sensor sheet was set at the center of the 7-m walkway. The macaques walked at their preferred walking speeds. Two video cameras were set perpendicular to the walkway: a distant one was used to measure walking speed and a close one to capture foot images. A third video camera was placed at the end of walkway to monitor the progression of the macaques along the long axis of the walkway. We analyzed only sequences in which at least one complete foot-pressure proﬁle was recorded, and the animal walked parallel to the long axis of the walkway at a steady speed. For each foot-pressure image, the COP was estimated in the same manner as for estimating the center of gravity (Goldstein, 1980). Functional foot length was deﬁned as the length from the proximal-most point of the contact area of the foot to the distal-most point of the third digit (Vereecke et al., 2003). The footprogression angle, which was deﬁned as the angle between the lines used to measure functional foot length and the direction of progress (i.e., the long axis of the walkway), was calculated to estimate out-toeing. Hallux abduction was deﬁned as the angle between the long axis of the hallux and the line representing the functional foot length deﬁned earlier. Nine foot regions were American Journal of Physical Anthropology deﬁned based on the functional morphology of the macaque foot: the tarsal area, medial (second and third) metatarsal shafts (hereafter, referred to as the ‘‘medial metatarsals’’), lateral (fourth and ﬁfth) metatarsal shafts (‘‘lateral metatarsals’’), ﬁrst metatarsophalangeal (MP) joint (MPJ I), medial (second and third) MP joints (MPJ II, III), lateral (fourth and ﬁfth) MP joints (MPJ IV, V), hallux, medial (second and third) digits (Digits II, III), and lateral (fourth and ﬁfth) digits (Digits IV, V). The second and third and the fourth and ﬁfth MP joints were analyzed as the ‘‘medial’’ and ‘‘lateral’’ MP joints, respectively, because the resolution of the system (5 mm) was insufﬁcient to deﬁne areas for each metatarsal head accurately. The same was true for the digits. In each of the nine regions, the 1-cm2 area with the highest pressure throughout the stance phase was chosen, and the changes in pressure in each regional peak pressure area were recorded. The palmar pressures were also recorded, but it will not be discussed here as they are beyond the scope of this study. Although our animals walked at their preferred walking speed, which ranged from 1.3 to 4.3 km/h, we selected narrow and similar speed ranges (2.5–3.1 km/h for quadrupedalism and 2.6–3.1 km/h for bipedalism) for the analyses in this study to avoid the confounding effects of walking speed on the pressure proﬁles. However, we plan to address the effects of walking speed in future studies, as it has been reported that speed inﬂuences the plantar (and palmar) pressures (Rosenbaum et al., 1994; Vereecke et al., 2005b; Patel and Wunderlich, 2008). The experiment was approved by the Institutional Animal Care and Use Committee of Osaka University. RESULTS Although the subjects walked at similar speeds during quadrupedal (2.79 km/h) and bipedal (2.91 km/h) walking (Table 1), the step frequency and stride length differed between the two modes of locomotion. During bipedal walking, the strides were shorter and the step frequencies were higher than during quadrupedal walking. The absolute contact durations were shorter in bipedal walking than in quadrupedal walking, but no signiﬁcant difference was found in the duty factors (Table 1). These tendencies were consistent with previous studies of Japanese macaques (Kimura, 1985; Hirasaki and Kumakura, 2004) and bonobos (Aerts et al., 2000). Foot pressure proﬁle during quadrupedal walking The foot-pressure proﬁle analysis showed that the ﬁrst contact during quadrupedal walking was typically made by the tarsal area close to the calcaneocuboid joint (Fig. 1a). The heel never touched down throughout the stance phase, as indicated by the fact that the functional foot length was signiﬁcantly shorter (81.4%) than the actual foot length (Table 2). Because the toe tips were in 151 PLANTAR PRESSURE DISTRIBUTION IN MACACA FUSCATA Fig. 1. Plantar pressure proﬁles during quadrupedal and bipedal walking in Japanese macaques (Macaca fuscata). (a) Typical pressure proﬁles at ﬁve speciﬁc times of the stance phase during quadrupedal and bipedal walking, respectively. The highest pressure is in red and the lowest in dark blue. The periods expressed as the percentage of stance phase duration are shown below each panel. To help clarify which part of the foot was in contact at each instant, the contour of the entire contact area of the foot during the stance phase is superimposed. (b) The typical foot posture during quadrupedal walking. (c) Typical pressure proﬁles during bipedal walking. During bipedal walking, initial contact was made by the lateral forefoot close to the ﬁfth metatarsal head, whereas the tarsal part touched down ﬁrst during quadrupedal walking. (d) The typical foot posture during bipedal walking. (e) Typical trajectory of the center of pressure (COP) during quadrupedal walking. The trajectory was relatively straight. (f) A case in which the hallux supported more weight during quadrupedal walking. (g) A typical trajectory of the COP during bipedal walking. The COP shifted medially at the level of the metatarsal heads. TABLE 2. Functional foot length, foot-progression angle, and hallux abduction during locomotion Quadrupedal (n 5 13) Bipedal (n 5 14) Signiﬁcance Functional foot length (%) Foot-progression angle (8) Hallux abduction (8) 81.6 6 2.7 83.1 6 3.9 n.s. 5.4 6 6.4 9.0 6 7.5 n.s. 44.6 6 4.6 53.5 6 9.7 P \ 0.01 Values are the mean 6 SD. Student’s t test was used to compare the results between quadrupedal and bipedal walking. Fig. 2. Pressures on each area of the foot during (a) quadrupedal and (b) bipedal walking. An important difference between the two modes of locomotion was that the values were greater for the medial digits (broken red line) and metatarsal heads (solid red line) than for the lateral digits (broken green line) and metatarsal heads (solid green line) during bipedal walking, especially in the later stance phase, whereas no such tendency was observed during quadrupedal walking. Another difference was in the use of the hallux; the pressure at the hallux (broken blue line) and ﬁrst metatarsal head (solid blue line) peaked later in the stance phase during bipedal walking and in the mid-stance phase during quadrupedal walking. American Journal of Physical Anthropology 152 E. HIRASAKI ET AL. Fig. 4. The mediolateral ratio of plantar pressures. The ratio was calculated as the sum of the pressures for the medial (second and third) metatarsals, medial MP joints, and medial digits, divided by the sum of the pressures for the lateral (fourth and ﬁfth) metatarsals, lateral MP joints, and lateral digits. The solid and broken lines represent bipedal and quadrupedal gait, respectively. Fig. 3. The relative contact timing and duration for the nine foot regions expressed as a percent of the total stance phase duration during (a) quadrupedal and (b) bipedal locomotion. M, metatarsals 5 medial metatarsal shafts; L, metatarsals 5 lateral metatarsal shafts; MPJ I 5 ﬁrst metatarsophalangeal (MP) joints; MPJ II, III 5 second and third MP joints; MPJ IV, V 5 fourth and ﬁfth MP joints; Digits II, III 5 second and third digits; Digits IV, V 5 fourth and ﬁfth digits. Blackened areas indicate that the foot region was in contact with the ground in more than 75% of the measured steps; dark gray areas indicate frequent, but less predictable contact (50–75%); light gray areas indicate contact of the foot region occurring in between 25 and 50% of the steps. contact with the ground, and digits were not curled very much (Fig. 1b), unlike in the case of bonobos (Vereecke et al., 2003), the shorter functional foot length indicates that the heel was elevated. The greatest pressure during the early stance phase was observed in the tarsal region close to the calcaneocuboid joint (Figs. 1a,e and 2a, black line). Then, the COP progressed toward the third metatarsal head at the early stance phase in a relatively straight trajectory (Fig. 1e). At the mid-stance phase, the tarsal and metatarsal regions of the foot were elevated (Figs. 1a and 3a), and the pressure was greatest near the second to ﬁfth metatarsal heads (Figs. 1a,e and 2a, solid green and red lines), although the values were slightly larger at the lateral metatarsal heads (green line) than the medial ones (red line). Next, the COP shifted medially slightly, but the pressures under the lateral regions of the foot were larger than those under the medial regions throughout the stance phase (Fig. 4, broken line). The COP shifted farther to the third digit, which left the substrate last, suggesting that the functional axis of the macaque foot is aligned with the third digit. The hallux, which left the ground earlier than the other digits, and the ﬁrst metatarsal head had their peak pressures in the early stance phase (Fig. 2a, solid and broken blue lines, respectively). American Journal of Physical Anthropology A different pattern in which the hallux touched down ﬁrst, as in the case of bonobo bipedal and quadrupedal walking and gibbon bipedal walking (Vereecke et al., 2003, 2005b), was also found in 30–40% of the walking sequences collected. In most cases, the initial touch was light, but occasionally the hallux supported more weight during the early stance phase, resulting in a V-shaped trajectory and medial deviation of the COP (Fig. 1f). This was observed in both of the study subjects. Foot-pressure proﬁle during bipedal walking During bipedal walking, the initial contact was almost always made by the lateral forefoot close to the ﬁfth metatarsal head (Figs. 1c and 3b), suggesting that the foot was in an inverted position. This was consistent with Okada’s (1985) observation in the same species. Then, the foot was everted gradually. The COP ﬁrst traveled backward to the tarsal area, although the heel never touched the ﬂoor (Fig. 1d), as indicated by a functional foot length that was shorter than the actual foot length (83.1%, Table 2). This value was similar to that during quadrupedal walking (81.4%), despite the foot supporting substantially more weight during bipedal walking than during quadrupedal walking. The greater weight supported resulted in higher peak pressures; the peak pressure for each foot region during bipedal walking was more than twice that during quadrupedal walking (Fig. 2, Table 3). The greatest pressure during the early stance phase was observed near the calcaneocuboid joint (Figs. 1c, 1g, and 2b, black line), as in quadrupedal walking. Unlike the roll-off pattern in a quadrupedal gait, however, the tarsal and metatarsal regions of the foot remained in contact with the ground until later in the stance phase (Fig. 3). This is seen clearly in the COP fore-aft movements; the COP during bipedal walking stayed proximally much longer (until ca. 30% of the stance phase, Fig. 5) and then progressed anteriorly during the ﬂat foot and mid-stance phases (Fig. 1g). This may have resulted from the difference in trunk posture. Because the center of the PLANTAR PRESSURE DISTRIBUTION IN MACACA FUSCATA 153 TABLE 3. The maximum pressure for each foot region Quadrupedal (kPa) (n 5 13) Tarsal area M. metatarsals L. metatarsals MPJ I MPJ II, III MPJ IV, V Hallux Digits II, III Digits IV, V 49.3 21.3 29.0 52.5 48.5 56.0 34.8 33.2 32.6 6 6 6 6 6 6 6 6 6 11.5 3.5 12.2 27.7 14.9 19.8 19.4 13.3 18.2 Bipedal (kPa) (n 5 14) 102.5 52.2 55.0 85.1 98.1 69.9 75.2 78.8 48.2 6 6 6 6 6 6 6 6 6 16.4 25.6 7.5 53.5 32.4 21.7 49.9 5.4 9.3 See text and Fig. 3 for the abbreviations. body mass is located in the lumbar region of the trunk, its position relative to the feet is more forward in pronograde postures than in orthograde postures, inducing rapid forward transfer of the COP during quadrupedal walking. In contrast, when walking bipedally, the center of mass is at the rear of the foot during the early half of the stance phase, keeping the COP in the proximal regions of the foot. Occasionally, the hallux touched down ﬁrst, as in the case of quadrupedal walking. However, this was less frequent than during quadrupedal walking (\15%), and the touch appeared to be light; this was followed immediately by substantial touchdown of the tarsal region close to the calcaneocuboid joint. The V-shaped trajectory of the COP was found in one walking sequence only. In the other case, the initial contact by the hallux was not reﬂected in the trajectory of the COP because of the small values and short duration. At 60% of the stance phase, the COP started to shift medially, as clearly shown in Figure 4. Because no signiﬁcant difference was seen in the pressures between the medial and lateral metatarsal regions (Fig. 2, brown and orange lines, respectively), the medial transfer of the COP occurred at the level of the metatarsal heads (Fig. 2). At toe-off, the COP shifted laterally (Fig. 4) and the third digit left the substrate last (Fig. 1g). Abrupt declines occurred in the digit pressures at the end of the stance phase, which resembled the human gait (Fig. 2b, green and red-broken lines). The pressures at the ﬁrst MP joint (Fig. 2b, solid blue line) and hallux (broken blue line) peaked late in the stance phase, and their magnitudes were similar to that of the lateral MP joints, suggesting that the big toe played a more important role during bipedal walking than during quadrupedal walking. Nevertheless, as in the case of quadrupedal walking, the hallux left the ground before the other digits due to its abducted position. The hallux was abducted substantially during both bipedal and quadrupedal walking, and abduction was more prominent during bipedal walking than during quadrupedal walking (Table 2, P \ 0.05). The foot-progression angle measurements revealed that the foot was out-toed slightly more during bipedal walking than during quadrupedal walking, although the difference was not signiﬁcant due to the large variation (Table 2). In addition, the direction of walking was deﬁned as the long axis of the walkway. Although we removed the sequences in which the animals did not walk parallel to the walkway, this deﬁnition might have reduced the reliability of the foot-progression angle values. Fig. 5. Fore-aft movement of the COP is represented as the relative distance from the tip of the third digit, and is plotted against time expressed as a percentage of the stance phase. The solid and broken lines represent bipedal and quadrupedal gaits, respectively. The COP in bipedal walking traveled backward ﬁrst and stayed close to the tarsal area for much longer than during quadrupedal walking. In all the analyzed sequences of both bipedal and quadrupedal walking, the medial metatarsal area was in contact with the ground (Figs. 1 and 2), suggesting that no longitudinal foot arch exists in the macaque foot. DISCUSSION The main ﬁnding of this study was that Japanese macaques use their feet very differently in bipedal and quadrupedal locomotion. As expected, the pressures were much greater during bipedal walking than during quadrupedal walking (Fig. 2 and Table 3). This was not surprising, as only two hind limbs support the body weight during bipedal walking, whereas all four limbs do so during quadrupedal walking. However, because the hind limbs of the Japanese macaque support more weight than the forelimbs during quadrupedal walking (Kimura, 1985; Hirasaki and Matano, 1996), the observed peak values of bipedal walking, which were more than twice larger than those of quadrupedal walking, seemed to be too great. Shorter contact durations (Table 1) during bipedal walking likely increased the peak values. The trajectory of the COP also differed between the two modes of locomotion. During quadrupedal walking, the COP traveled in a relatively straight line forward from the tarsal area to the third or fourth digit, whereas during bipedal walking, it traveled backward from the lateral metatarsal to the tarsal area and then moved forward and medially across the metatarsal heads. The later part of this motion (i.e., the medial transfer) resembled the human COP trajectory. In humans, the COP transfers medially to the second MP joint at the end of the stance phase (e.g., Barnett, 1956). Because this pattern was also observed in Japanese macaques, which are phylogenetically distant from humans, it sugAmerican Journal of Physical Anthropology 154 E. HIRASAKI ET AL. gests that the medial transfer at the end of the stance phase results from the biomechanical requirements of bipedal walking; in bipedal walking, the upper part of the body is supported only by one hind limb during single-limb support phase and moves laterally onto the other hind limb at the end of the stance phase. With this lateral movement of body mass and probably because of the slightly more out-toed feet, the COP is transferred medially at the end of the stance phase. In contrast, during quadrupedal walking, the lateral movement of the body due to hind limb alteration is likely attenuated by forelimb support, resulting in the relatively straight trajectory of the COP. This is consistent with Kimura’s (1985) ﬁnding that the lateral component of the ﬂoor reaction force is greater during bipedal walking than during quadrupedal walking. In bonobos, unlike in Japanese macaques, quadrupedal walking is most often associated with a curved COP trajectory (Vereecke et al., 2003); this results in the functional foot axis running through the second digit. We attribute the difference between Japanese macaques and bonobos to the difference between a semipalmigrade quadrupedal gait and knuckle walking; during knuckle walking, the forelimbs support a much smaller proportion of the body weight compared to the semipalmigrade/ palmigrade quadrupedal gait (Kimura et al., 1979; Kimura, 1985; Demes et al., 1994). The forelimb support in knuckle walking may not be sufﬁcient to attenuate the lateral movement of the body, which results in the medial shift of the COP trajectory at the late-stance phase. The difference in the initial contact between the two modes of locomotion suggests that the foot was more outtoed, inverted, and down-toed (i.e., the toes are lower than the heel) when touching down in bipedal walking than in quadrupedal walking. As shown in Table 1, the foot was slightly more out-toed during bipedal walking, although the difference was not signiﬁcant due to the large standard deviations and a small sample size. The out-toed, inverted foot at touchdown may result from the externally rotated thigh during bipedal walking; this needs to be conﬁrmed by kinematic data, which is outside the scope of this paper. The down-toed foot postures may be partly attributable to the shorter stride length during bipedal walking compared to quadrupedal walking (Table 1); with a shorter stride, the hind limbs are not protracted as much, resulting in relatively downward foot postures, given a constant talocrural joint angle. Another difference between the two modes of locomotion was in the use of the hallux. The maximum pressures at the hallux and ﬁrst MP joint were observed at the late-stance phase during bipedal walking and the mid-stance phase of quadrupedal walking (Fig. 2). This seems unusual, because the hallux was abducted more during bipedal walking. This may have occurred partly because the COP progressed more slowly during bipedal walking than during quadrupedal walking (Figs. 3 and 5). In addition, out-toeing may increase the importance of the hallux for generating propulsive force in bipedal walking. The important role of the hallux in the latestance phase during bipedal walking again recalls human foot motions. The hallux of the Japanese macaques often touched down ﬁrst during both modes of locomotion, although this was less frequent during bipedal walking. Initial hallux contact is also observed in sifakas (Wunderlich and Schaum, 2007) and gibbons (Vereecke et al., 2005b) American Journal of Physical Anthropology and occasionally in bonobos (Vereecke et al., 2003). Vereecke et al. (2005b) postulated that this pattern was due to the arboreally adapted foot anatomy of apes (widely abducted hallux, especially in gibbons). This concurs with Okada’s (1985) report that hallucial plantar-ﬂexion before touchdown is frequently observed in arboreal spider monkeys, whereas terrestrial baboons do not show it. In contrast, in humans, the hallux is usually dorsiﬂexed before touchdown. Okada (1985) argued that the hallucial dorsiﬂexion in humans evolved to enhance the foot arch and to strain the sole before touchdown. If this is true, bipedal walking of the Japanese macaque, in which initial hallucial contact is infrequent, could be an adaptation to terrestrial locomotion, although this is speculative and more information is necessary. Another important point is the abrupt decline in pressures during the late-stance phase in bipedal walking (Fig. 2b). This was not observed during quadrupedal walking by Japanese macaques (Fig. 2a) and is also absent during ape locomotion (Vereecke et al., 2003, 2005b). The smoother declines during quadrupedal walking are thought to result from the forelimb support, which reduces the hind limb load at the end of the stance phase. The abrupt decline in the pressures observed in the bipedal walking of Japanese macaques resembles the human gait pattern. Nevertheless, although the shape is similar, the magnitude may be smaller, because the ﬂoor reaction force data (Ishida et al., 1974) do not suggest a strong push-off by the toes at the end of the stance phase. More detailed kinematic data need to be collected simultaneously with pressure data in future studies. The two modes of locomotion showed several shared features. During both bipedal and quadrupedal walking, the Japanese macaques never touched their heel down. They appeared to walk with terrestrial semiplantigrade foot postures (Gebo, 1993a, b). The ﬂexibility at the tarsometatarsal or calcaneocuboid joint (the midtarsal break) likely makes this foot posture possible (Hirasaki and Kumakura, 2003). Vereecke et al. (2003, 2004, 2005b) reported that the heel contacts the ground during both bipedal and quadrupedal walking in bonobos and during bipedal walking in gibbons, although a heelstrike comparable to human gait was observed only in bonobos. In gibbons, the heel contacts the substrate at mid-stance. On the basis of their results, Vereecke et al. (2005b) suggested that heel-strike is a synapomorphy of the great apes. In our experiments, the heel of the macaques was elevated throughout the stance phase. This concurs with the well-accepted idea that heel contact is unique to humans, apes, and atelines, although heelstrike is restricted to humans and great apes (Meldrum, 1991, 1993; Gebo, 1002, 1993a,b; Schmitt and Larson, 1995). Future additional evidence from other species would increase the robustness of the idea that heel contact is one of the features distinguishing apes from monkeys. Bipedal and quadrupedal walking were also similar in that the third digit left the ground last. This indicates that the functional axis of the foot was aligned with the third digit in Japanese macaques and is consistent with our morphological observations (Hirasaki and Kumakura, in press) that the interosseous muscles of the Japanese macaque are arranged around the third digit. This is different from chimpanzees and bonobos, in which the second digit leaves the ground last, showing that the functional axis is on the second metatarsal and digit (Wunderlich, 1999; Vereecke et al., 2003). PLANTAR PRESSURE DISTRIBUTION IN MACACA FUSCATA In summary, we found some humanlike characteristics in foot motion during bipedal walking— the medial transfer of the COP, the increased importance of the hallux during late-stance phase, and an abrupt decline in pressures during late stance phase—in Japanese macaques and attributed them to the biomechanical requirements of bipedal walking. Nevertheless, the other features in the foot motion of the macaque during bipedal walking differed greatly from those of humans: the heel did not contact the ground, no foot longitudinal arch was observed, the hallux was not aligned with the other digits, and the functional axis was not aligned with the second digit. Future studies should address how these features evolved in humans. ACKNOWLEDGMENTS We express our gratitude to Dr. Evie E. Vereecke (University of Liverpool), anonymous reviewers, Associate Editor, and Editor-in-Chief for their invaluable comments and suggestions on this manuscript. We are also grateful to Dr. Roshna Wunderlich (James Madison University) and Dr. Kristiaan D’Août (University of Antwerp) for their thoughtful and constructive comments on the earlier versions of this manuscript. LITERATURE CITED Aerts P, Van Damme R, Van Elsacker L, Duchêne V. 2000. Spatio-temporal gait characteristics of the hind-limb cycles during voluntary bipedal and quadrupedal walking in bonobos (Pan paniscus). Am J Phys Anthropol 111:503–517. Barnett CH. 1956. The phases of human gait. Lancet 2:617– 621. Crompton RH, Thorpe S, Wang WJ, Li Y, Payne R, Savage R, Carey T, Aerts P, Van Elsacker L, Hofstetter A, Günther MM, Richardson J. 2003. The biomechanical evolution of erect bipedality. Cour Forsch-Inst Senckenberg 243:135–146. Crompton RH, Vereecke EE, Thorpe SKS. 2008. Locomotion and posture from the common hominoid ancestor to fully modern hominins, with special reference to the last common panin/ hominin ancestor. J Anat 212:501–543. D’Août K, Vereecke E, Schoonaert K, De Clercq D, Van Elsacker L, Aerts P. 2004. Locomotion in bonobos (Pan paniscus): differences and similarities between bipedal and quadrupedal terrestrial walking, and a comparison with other locomotor modes. J Anat 204:353–61. Demes B, Larson SG, Stern JTJ, Jungers WL, Biknevicius AR, Schmitt D. 1994. The kinetics of primate quadrupedalism: ‘‘hindlimb drive’’ reconsidered. J Hum Evol 26:353– 374. Elftman H, Manter J. 1935. Chimpanzee and human feet in bipedal walking. Am J Phys Anthropol 20:69–79. Gebo DL. 1992. Plantigrady and foot adaptation in African apes: implications for hominid origins. Am J Phys Anthropol 89:29–58. Gebo DL. 1993a. Functional morphology of the foot in primates. In: Gebo DL, editor. Postcranial adaptation in nonhuman primates. DeKalb: Northern Illinois University Press. p 175–196. Gebo DL. 1993b. Reply to Dr. Meldrum. Am J Phys Anthropol 91:382–385. Goldstein H. 1980. Classical mechanics, 2nd ed. Reading, MA: Addison-Wesley. Hirasaki E, Kumakura H. 2003. Foot kinematics of Hylobates lar, Ateles geoffroyi, and Macaca fuscata during locomotion on arboreal and terrestrial substrates. Am J Phys Anthropol [Suppl] 36:114–115. 155 Hirasaki E, Kumakura H. 2004. Head movements during locomotion in a gibbon and Japanese macaques. Neuroreport 15:643–647. Hirasaki E, Kumakura H. In press. Estimating the functional axis of the primate foot using the distribution of plantar muscles. Int J Primatol. Hirasaki E, Matano S. 1996. Comparison of locomotor patterns and the cerebellar complex in Ateles and Macaca. Folia Primatol 66:209–225. Hirasaki E, Ogihara N, Hamada Y, Kumakura H, Nakatsukasa M. 2004. Do highly trained monkeys walk like humans? A kinematic study of bipedal locomotion in bipedally trained Japanese macaques. J Hum Evol 46:739–750. Ishida H, Kimura T, Okada, M. 1974. Patterns of bipedal walking in anthropoid primates. In: Kondo S, Kawai M, Ehara A, Kawamura S, editors. Proceedings from the symposia of the 5th congress of the international primatological society. Tokyo: Japan Science Press. p 287–301. Jenkins FA Jr. 1972. Chimpanzee bipedalism: cineradiographic analysis and implications for the evolution of gait. Science 178:877–879. Kimura T. 1985. Bipedal and quadrupedal walking of primates: comparative dynamics. In: Kondo S, Ishida H, Kimura T, Okada M, Yamazaki N, Prost JH, editors. Primate morphophysiology, locomotor analysis and human bipedalism. Tokyo: University of Tokyo Press. p 81–104. Kimura T, Okada M, Ishida H. 1979. Kinesiological characteristics of primate walking: its signiﬁcance in human walking, In: Morbeck ME, Preuschoft H, Gomberg N, editors. Environment, behavior, and morphology: dynamic interactions in primates. New York: Gustav Fischer. p 297– 311. Lewis OJ. 1989. Functional morphology of the evolving hand and foot. Oxford: Clarendon Press. Meldrum DJ. 1991. Kinematics of the Cercopithecine foot on arboreal and terrestrial substrates with implications for the interpretation of hominid terrestrial adaptations. Am J Phys Anthropol 84:273–289. Meldrum DJ. 1993. On plantigrady and quadrupedalism. Am J Phys Anthropol 91:379–381. Okada M. 1985. Primate bipedal walking: comparative kinematics. In: Kondo S, Ishida H, Kimura T, Okada M, Yamazaki N, Prost JH, editors. Primate morphophysiology, locomotor analysis and human bipedalism. Tokyo: University of Tokyo Press. p 47–58. Patel BA, Wunderlich RE. 2008. Speed effects on palmar pressure in digitigrade baboons (Papio anubis). Am J Phys Anthropol [Suppl] 46:168–169. Rosenbaum D, Hautman S, Gold M, Claes L. 1994. Effects of walking speed on plantar pressure patterns and hindfoot angular motion. Gait Posture 2:91–197. Schmitt D, Larson SG. 1995. Heel contact as a function of substrate type and speed in primates. Am J Phys Anthropol 96:39–50. Shapiro LJ, Jungers WL. 1988. Back muscle function during bipedal walking in chimpanzee and gibbon: implications for the evolution of human locomotion. Am J Phys Anthropol 77:201–212. Sigmon BA, Farslow DL. 1986. The primate hindlimb. In: Swindler DR, Erwin J, editors. Comparative primate biology, Vol. 1: systematics, evolution, and anatomy. New York: Alan R Liss, Inc. p 671–718. Stern JT, Susman RL. 1981. Electromyography of the gluteal muscles in Hylobates, Pongo, and Pan: implications for the evolution of hominid bipedality. Am J Phys Anthropol 55:153– 166. Susman RL. 1983. Evolution of the human foot: evidence from Plio-Pleistocene hominids. Foot Ankle 3:365–376. Vereecke E, D’Août K, De Clercq D, Van Elsacker L, Aerts P. 2003. Dynamic plantar pressure distribution during terrestrial locomotion of bonobos (Pan paniscus). Am J Phys Anthropol 120:373–383. American Journal of Physical Anthropology 156 E. HIRASAKI ET AL. Vereecke E, D’Août, K, De Clercq D, Van Elsacker L, Aerts P. 2004. The relationship between speed, contact time and peak plantar pressure in terrestrial walking of bonobos. Folia Primatol 75:266–278. Vereecke EE, D’Août K, Payne R, Aerts P. 2005a. Functional analysis of the foot and ankle myology of gibbons and bonobos. J Anat 206:453–476. Vereecke E, D’Août, K, Van Elsacker L, De Clercq D, Aerts P. 2005b. Functional analysis of the gibbon foot during bipedal walking: plantar pressure distributions and three-dimensional ground reaction forces. Am J Phys Anthropol 128:659–669. American Journal of Physical Anthropology Wunderlich RE. 1999. Pedal form and plantar pressure distribution in anthropoid primates, Ph.D. dissertation, State University of New York, Stony Brook. Wunderlich RE. 2003. Dynamics of foot use during bipedal and quadrupedal walking in Pan paniscus. Am J Phys Anthropol [Suppl] 36:228. Wunderlich RE, Schaum JC. 2007. Kinematics of bipedalism in Propithecus verreauxi. J Zool 272:165–175. Yamazaki N, Ishida H, Kimura T, Okada M. 1979. Biomechanical analysis of primate bipedal walking by computer simulation. J Hum Evol 8:337–349.