Differences in Terrestrial Velocity in Macaca and Presbytis THEODORE I. GRAND Oregon Regional Przmnte Rc7senrrh Center, 505 N.W. 1 8 5 th Avenue, Ben-oerton, Oregon 97005 KEY WORDS Pres bytis. Locomotor behavior . Ground speed . Macaca . ABSTRACT Over 150 sequences of terrestrial movement were timed and measured i n the toque macaque, the gray langur, and the purple-faced langur to determine the relation between velocity, gait, and psychosocial context. Species differences were found in mean velocity, “favored” gait, and surface preference. All three species used the walk and the gallop at the slowest and fastest speeds respectively. The macaques and gray langurs walked frequcntly, but the purple-faced langurs were never seen to do so. At intermediate speeds, the macaques used either the “fast” walk or the “slow” gallop, whereas the gray langurs used irregular patterns of walk-gallop-walk-gallop. The purple-faced langurs were faster (about 20 fps) and less variable than the other species, regardless of distance. These data suggest that motor expression varies among the Cercopithecoid monkeys; the correlation between locomotion and anatomy is not as close as it is among wholly terrestrial or arboreal forms; slowness i n the macaque is an expression of social confidence, not of biomechanical inability, high speed i n the purple-faced langur is due to psychosocial factors rather than to “terrestrial” adaptability. The subject of this paper is the use of one can establish the relation between supground speed to measure the behavior of port surface and body structure. An adethree monkeys, a macaque and two langurs. quate functional-anatomical paradigm reThis analysis is part of a broader exam- sults. Most primates, however, divide their time ination of the ecology and sociology of Ceylonese primates (Eisenberg et al., ’72; Rip- between canopy supports and the ground, and the clear-cut distinction seen i n the ley, ’67, ’70; Dittus, ’74). One focus of the overall study, habitat above animals is less sharp (Rose, ’73). preference, is a debated issue of primate Those South American forms which spend behavior. The confusion is partly due to the 100% of each day i n the trees cannot be clear terrestrial-arboreal dichotomy of other judged structurally “incapable” of moving lineages. Completely arboreal and terres- on the ground. The African and Asian species move trial forms have appeared over long time periods and represent extremes of some about i n a more flexible and diverse fashion. theoretical ground-canopy continuum. In Thus, they are somewhere within the centhe horse, for example, surface occupation ter of the ground-canopy continuum. That a and extreme anatomical adaptation are moderate degree of surface occupation imclearly correlated. The three-toed sloths, by plies structural adaptation must be proved contrast, spend almost all their time within in each case. What may look like a n adapthe canopy, descending awakwardly for tation to one surface may be the preferonly a few minutes once a week. This polarization of habitat is unequivocal, and the 1 Publication No. 823 of the Oregon Regional Primate anatomical extensions are unquestioned. Research Center, supported in part by Grants RR-00163 from the National Institutes of Health, USPHS. and by By simply measuring the time spent by a the Smithsonian Biological Program i n Ceylon (Primatr given species upon one or the other surface, Ecology Survey) AM. J. P H Y S A . N T H R O P . , ~101-108. ~: 101 102 THEODORE I. G R A N D ence of the local population, not a structural character of the species. Therefore, in this comparison a simpler question than what "percent time" is spent on one surface is posed: While a n individual is o n the ground, what does he do, how far does he move, how much time docs it take? In other words, how fast does he travel? The toque macaque, gray langur, and purplefaced langur, living within the same habitat, provide a three-way comparison and narrow the discussion to two fundamental questions: Does the toque macaque, which differs structurally more than the other two species, also differ more in speed? Are the two structurally similar larigurs also similar i n the way they travel over open ground? The conclusions from these data relate velocity and gait pattern, on the one hand, with the social context of individual movement. on the other. METHODOLOGY The three species - the toque macaque (Mncaca sinica), the gray langur (Presbytis en tellus), and the purple-faced larigur ( P , senex)-were observed in the open parkland habitat of Polonnaruwa, Ceylon. Tree crowns overlapped in some places but were discontinuous in others. This structural fact, as well as the availability (or lack) of food, brought each of the species to the ground for some part of each day, although on occasion P. scnex remained aloft for several days. Where the shrub layer was cleared, the short grassy cover made i t easy to observe and measure the animals. Over 150 movement sequences were timed to the nearest 0.1 second and tape measured to the nearest foot. Each point in figure 3 is one sequence in feetisecond (fps), the triangle representing the walk, the circle the gallop. In each case, the associated gait, age, and sex of the individual, as well as the psychosocial bases of the movement, were observed. Each sequence was judged for linear directness, evenness (lack of overall acceleration and deceleration), and gait regularity. Irregular sequences like those among the gray langurs (i.e., walk-gallop-walk-gallop) are described separately and eliminated from the calculations of mean velocity. Film records from which tracings were made are included. Sampling differences occurred which reflect presumptive behavioral differences. The gray langurs and macaques were easy to approach and measure, but the macaques ran at higher speeds only in complicated chase sequences. The purple-faced larigurs were difficult to approach and I never saw them walking over open ground; they did, however, walk frequently within the canopy. Because of the stereotyped and isolated movements of these langurs, I was able to anticipate them. While at some distance removed, I could time accurately each in&vidual and identih take-off and target points: after the troop had relocated, I could approach and tape measure the distance. OBSERVATIONS i " u c n sinica. The biomechanics of the two gaits should he characterized. The walk is used at the slowest speeds (fig. 1A). At least one hand or foot is in contact with the ground at any instant. Most commonly, the limbs move in diagonal pairs, that is, one hindlimb and the diagonally opposite forelimb. Occasionally, the hindlimb and forelimb on the same side are moved together, a lateral sequence gait. The body mass has low vertical dlsplacement and little acceleration. A complete cycle takes about one second. The run or gallop is used for higher speeds (fig. 1B). The forelimbs and hindlimbs move in pairs, and right and left palms (and soles) strike closely in time. In the earlier phase, the second foot contact is almost completed, the palmas contact just begins. In the later phase, the body is off the ground but for the right palm. Within each cycle, one or two periods occur when the body is off the ground. The trajectory of truncal mass moves through an arc with greater vertical displacement and greater change i n acceleration. Two cycles may be completed within one second. (For detailed analyses of gait formulas, see Muybridge, '57; Hildebrand, '67; Prost and Sussman, '69.) Along the top of the graph (fig. 3A), the slowest speeds are associated with the walk; along the bottom, the faster movements are correlated with the run or gallop. The walk averages 3 fps; the run 10 fps. Occasionally these animals move at very high speeds. Circular and irregular chase sequences at 15 f p s were difficult to measure. At intermediate speeds, the gaits were SPEED I N MACACA AND P R E S B Y T I S / \ a j i 104 THEODORE I. GRAND intergraded. Sometimes a “fast” walk or a “slow” gallop was employed at the same speed. At these intermediate speeds, the animal had a choice of gait in moving from place to place, the choice reflecting the social circumstances and their impact upon the individual. This relation of speed and gait to state of mind are elaborated upon within the DISCUSSION. Troop movement is a combination of slow and fast gaits of all members. Walking and moderate running were frequent and equally easy to measure. Some individuals appeared to be leading the troop, others trying to catch up. Chase and pursuit within the troop’s flow were irregular, occasional incidents of social life. Together the slowest and fastest sequences averaged 5 fps. This approximation provides some idea of the slow rate of macaque travel. Presbytis entellus. Out of 73 observations, only ten walking sequences could be measured. The animals had the peculiar habit of walking a few steps, galloping a few more, walking a few, and then abruptly sitting. Eliminating this irregular pattern, used commonly at intermediate speeds, caused the striking separation of low-speed walk froni high-speed gallop. The gallop varied both in distance covered and in time elapsed (6 to 20 fps) : both short and long distances were covered slowl y and quickly; the average was 11.5 fps or 7.8 mph. One structural characteristic of the galloping langur diverges significantly from that of the macaque. Whereas the latter moves with a relatively straight back, the langur flexes (or gathers) its back to about two-thirds of its length (fig. 2 ) . Change of lead, designated by the forward striking hand or foot of a pair, was common during the gallop. In three successive cycles, a n animal might strike with the hind feet leftlright, leftlright, leftlright, while the forefeet alternated leftlright, rightoeft, left/right. Change also occurred between hindlimb and forelimb strike. Analyses of six long film sequences showed no regular or favored patterns. Most likely, i f the animal did not alternate leads, he Fig. 3 Velocity and gait pattern graphs (Triangle represents walking sequences; circle represents galloping sequences). A , M. sinicn. B, P. cntcllus. C, P. senex. 105 SPEED I N M A C A C A AND PRESBYTIS 1 w z . j:=;: 7 + . 31t15 f p s (n.27) .. . ..>. 3 . . s : ' . ' 2 . 102233fps (n.27) * 40 20 60 I 100 80 I 120 I 140 .. . i . E .. 0 Pre&y?is entellus 7 . 2 7 t 0 4 f p s (n=10) I- " *: B C #..=. V 11.3t2.8 fD.5. ln.63) I I I I I 20 40 60 80 I 100 I 120 I 100 120 I 140 I 160 I 140 I60 I 180 I 200 I 180 I 200 I I I I 40 60 80 I DISTANCE ( feet 1 Figure 3 I I I 220 240 I 260 106 THEODORE I. GRAND would drift toward the right or left. Why the gray langurs, but neither the macaques nor the purple-face langurs, use frequent lead changes is not clear. The latter ccrtainly do not drift away from their intended lines of travel. Older infants and juveniles, 5 to 15 months of age, tended to move at higher speeds than adults. About 20 sequences were measured. The young appeared more wary as they approached the ground, looking about from the base of the tree trunk before going “all out” to reach the target tree. Sometimes a young animal lagged behind mother and troop, then ran to catch up with her. Females moved in a slower, more irregular fashion. Thus both age and sex differences in movement are suggested. Like the macaques, the langur troop moved in amoeboid fashion, spread out i n time and space. E. g., one animal covered a distance and stopped, a few minutes later a second animal moved in the same direction, over a roughly parallel route, but not on a n identical pathway. Over the next half hour, other animals moved in the same direction. During this shift, feeding sometimes took place. When the animals approached the target trees, individuals chose to climb different trees. In this watchful, but not anxious manner. the entire group moved slowly forward. Presbytis senex. These animals were never timed walkmg on the ground (fig. 3C). All 31 sequences show direct, almost “flat out” movement. In a personal communication, Manley says. “I suspect in all your measured sequences you have animals ‘making a dash for it’; i.e., moving at maximum speed.” As the graph shows, the rate of travel was a regular 18 f p s f 2.1 from 20 up to 200 feet. Since this 12.3 mph appeared to be their maximum performance, the 2 3 mph estimate by Eriders (’46) is difficult to explain. To move between trees whose crowns were not connected, the troop had to choose circuitous routes or come to the ground. When the latter happened, the first animal descended to the base of the tree, looked or scanned about, then in a continuous motion, dropped to the ground, galloped directly to the target tree, and hopped into it. Then the animal looked around once again, and in less tense fashion, climbed into the higher branches. Each troop member went through the same stereotyped performance. The peculiar tension associated with this point-to-point movement has no counterpart among the adult gray langurs or the toque macaques. Because they traveled on the ground one-by-one in a straight line, they were easily trapped by dogs (Ripley, personal communication). DISCUSSION Obvious differences in average ground speed clearly distinguish the three species. The macaques are the slowest and most variable. They walk and run both short and long distances and achieve high speeds in short, irregular bursts. The gray langurs have about the same average speed arid they also vary their mode of travel. By contrast, the purple-faced langur is the fastest and most invariant. Whether traveling 20 or 200 feet, they never varied their speed. These facts suggest three problems about motor variability: (1) the anatomical bases of speed differences, (2) speed on the ground as a terrestrial skill, (3) the psychosocial stimuli of movement. The first aspect is the structural bases of movement. The difference in average running speed between macaque and purplefaced langur is about 9 fps. The macaques, however, are capable of short spurts of 1517 f p s , close to the maximum speed of the purple-faced langur. The extra speed of the latter can be explained by greater trunk and limb length and greater truncal flexibility. To relate structure to performance and to distinguish between species, it is not necessary to use any more precise osteometrics. Performance differences between the gray and purple-faced langurs cannot be explained by body structure. The average speed of both differs by 7 f p s , but the gray langurs were timed at 18-19 fps. Thus, they were capable of speeds equal to those of the purple-faced langurs but seldom had recourse to them. The second aspect of species variability is whether speed represents a significant terrestrial skill for these animals. Flight by antelope and horse and pursuit by cheetah are described in mileshour, and the achievement of high ground speed is a significant feature of terrestrial adaptation. Because of these specialized cases, mph has become the essence of speed. Absolute speed measures also suggest comparisons SPEED IN MACACA AND PRESBYTIS between smaller and larger animals: the squirrel and elephant travel at 17 and 15 mph, respectively (Layne and Benton, ’54; Howell, ’65). Body lengths per unit time would be a more appropriate scale here; in fact, speed in mph is irrelevant in smaller forms and is inadequate in comparing smaller and larger ones. Among these primate species, even feet/ second may not be a n appropriate measure of movement. The purple-faced langur can run 18 fps, but this performance explains neither where nor under what selective pressures the skill evolved. As a matter of fact, speed is only one facet of survival; rapid acceleration and the ricochet, both used to dodge a pursuer or to change direction or place, are other skills related to speed. Neither is speed of direct value for life within the canopy. On the one hand, distance traveled in feet is not the key because unobstructed linear routes do not exist; on the other hand, acceleration and jumping between branches may be the skill transmuted into high speed on the ground. High speed could derive from arboreal leaping. In any case, ground speed cannot be explained apart from the other aspects of the life of the purple-faced langur. We are left with the paradox that this animal, capable of the fastest, most regular performance, spends the least time on the ground. Within the psychosocial realm we may find the third potential explanation for movement differences. The anomaly of skilled performance upon the ground and (literal) avoidance of the ground is much more clearly demonstrated by measures of velocity and gait of individuals than by “percent time spent” upon the surface by the troop. It is more sensitive to the behavior of the individual and leads directly to the question: What limits the purple-faced langur to occasional use of the ground, on the one hand, and gives shared occupancy to the gray langurs and macaques, on the other? The particularities of speed lead us to the same answer: Macaques may choose to move at a “fast” walk or a “slow” gallop a t intermediate speeds, depending upon social circumstances; gray langurs, which spend a great deal of time on the ground, move at all speeds; and purple-faced langurs, which walk routinely over canopy supports, do not walk upon the ground. How- 107 ever, in the extreme south of Ceylon, near Calk, Manley found revealing exceptions to purple-faced langur behavior at Polonnaruwa. In some southern troops the a d mals spent much of their time on “rocky, virtually treeless coastal slopes.” These rocky slopes were not “quick-visit places” for the animals who spent much of the day on them, resting and feeding on creepers, herbs, and woody plants growing between the crevices. “All the animals seemed perfectly at home on these rocky surfaces and outcrops, sitting, walking leisurely across them, leaping u p or down from rock to rock.” In other words the animals were completely at ease and did not treat the rocky terrain as a n unusual part of their habitat. Despite their predilection for trees, P. senex appear to adapt easily to the ground when they have to. At Polonnaruwa all three species are capable of higher speeds, but only young gray langurs and purple-faced langurs do so with any regularity. Among these langurs, tension, apprehension, and wariness are characteristic features of each animal. The striking invariance of speed, the feeling that the individuals are moving “flatout,” supports this. Social confidence, state of mind, and perhaps occasional, as opposed to habitual use of the ground are the major forces upon the individual. Among chimpanzees rate of locomotion may communicate information about preferred objects (Menzel and Halperin, ’75); no doubt these other states are as easily transmitted among the monkeys. We conclude that the bases of high-speed i n the purple-faced langur are sociological and psychological rather than anatomical. Thus, it is not the osteometric differences among the species but the social organization and attitudes of the local population which are the bases of terrestrial occupation. ACKNOWLEDGMENTS A version of this paper was presented at the American Association of Physical Anthropologists meetings, held i n Denver, April, 1975. I gratefully acknowledge the cooperation of the Smithsonian personnel, including Drs. Eisenberg, Ripley, Dittus, Muckenhirn, and Manley and M. Lockhart. F. Shininger took film records for me. The ORPRC staff has been of assistance; M. 108 THEODORE I. GRAND B a s s offered editorial comments once more. In addition, A. Zihlman and S. Ripley read and commented upon the analysis. LITERATURE CITED Dittus, W. P. J . 1974 The ecology a n d behavior o f t h e toque monkey: Mnccicci sinicn. Unpublished Ph.D. thesis, U. Maryland. Eisenberg, J. F., N. A. Muckenhirn a n d R. Rudran 1972 The relation between ecology a n d social structure i n primates. Science, 176: 863-874. Enders, R. K. 1946 Speed i n the langur. J. Mammal., 27: 174-1 75. Ilildebrand, M. 1967 Symmetrical gaits of primates. Am, J. Phys. Anthrop., 26: 119-730. Howell. A. B. 1965 Speed i n Animals. Hafner, New York. Layne, J. N., and A. H. Benton 1954 Some speeds of small mammals. J. Mammal, 3 5 : 103104. Menzel, E. W., a n d S. Halperin 1975 Purposive behavior a s a basis for objective communication between chimpanzees. Science. 189: 6 5 2 4 5 4 . Muybridge, E. 1957 Animals i n Motion. Dover, New York. Prost, J. H., a n d R. W. Sussman 1969 Monkey locomotion o n inclined surfaces. Am. J . Phys. Anthrop.. 31: 53-58. Ripley, S. 1967 Leaping of langurs: A problem i n the study of locomotor adaptation. Am. J. Phys. Anthrop.. 26: 149-170. 1970 Leaves a n d leaf-monkeys. The social organization of foraging i n gray langurs. I n : Old World Monkeys. Evolution, Systematics. and Behavior, J. R. Napier a n d P. H. Napier. eds. Academic Press, New York, pp. 481-509. Rose, M. 1973 Quadrupedalism i n primates. Primates, 1 4 : 337-357.