Descending Locomotion in Primates

Primates are an order of mammals that lack claws. Therefore, arboreal primates must apply opposing pressures with their digits to grasp supports and move through their habitats. This requirement may affect the mechanics of specific aspects of arboreal travel, such as descent, in the locomotion of primates compared to clawed non-primates, and may have influenced the evolutionary selective pressures that primates experienced over time. It has been hypothesized that larger primates are less likely to descend supports headfirst than smaller primates and clawed non-primates, however, this phenomenon has never been considered in a comparative context. Knowing how body size, anatomical proportions, and environment interact to affect locomotor behaviors is central to linking morphology with behavior, such as when evaluating hypotheses of primate origins.

This thesis analyzed descending locomotion in nine species of strepsirrhine primates that occupy four locomotor categories: large arboreal generalists, representing above branch quadrupeds weighting over 1 kg; small arboreal generalists, representing above branch quadrupeds under 1 kg; slow climbers; and vertical clingers and leapers. Primates were video recorded moving on supports ranging from horizontal to vertical in 15° increments. I tested specific hypotheses about gait and kinematic changes in response to declines that have been observed in primates moving down supports as steep as approximately 30° to see if these patterns would be replicated in primates moving on steeper support orientations.

I found that primates under 1 kg always used headfirst descent on all supports. For primates above 1 kg, body size appeared to be an important factor in determining behavior, but it also appeared that anatomical differences might have enabled one of the largest species in the sample, Varecia variegata, to perform vertical headfirst descent, while relatively smaller species like Lemur catta were not observed to use this behavior on supports greater than 45°. Within these large arboreal generalists, increases to individual age also seemed to drive behaviors away from headfirst descent in favor of tail first descent. Frequencies of headfirst descent were compared to other mechanisms of descent, such as tail first descent, were scored for a total of 3139 observed descents. These observations were incorporated into a Bayesian multilevel model that included information on the support condition (including orientation and diameter), morphological information for each species (average intermembral index and foot proportions), as well as individual mass and age. The model was then used to predict the probability of headfirst descent on various supports in simulated ancestral primates that exemplified different hypotheses of primate origins. It was found that features including body mass, support orientation, and foot and limb proportions greatly affected the predicted probabilities of headfirst descent. Large primates with lower intermembral indices and smaller feet were least likely to use headfirst, especially as supports became more steeply angled. Species that were smaller, with relatively longer arms or larger feet tend to use headfirst descent most frequently, even on vertical supports. The model predicted less headfirst descent in the very smallest primates on near horizontal supports, driven by observations of leaping in the smallest species in this sample.

Headfirst descents were analyzed for footfall patterns to evaluate temporal aspects of gait, and to test the hypothesis that limb phase should decrease as supports become steeper and that contact period with the support should increase, and relatively more so for the forelimbs than hind limbs. It was found that limb phase did significantly decrease across the sample as support orientation became steeper, and that both forelimb and hind limb contact times increased as proportion of the total stride period, although the forelimb did not increase relatively more than the hind limb in many species.

Headfirst descents were then analyzed for changes to kinematic aspects of gait including effective limb length, joint angles at key points during the stride, limb excursions, and velocity. I found that as supports became steeper primates across the sample reduced trunk inclination bringing the body parallel to the support and reducing the distance of the center of mass from the support, consistent with pitch-avoidance strategies. The forelimb remained compliant and highly protracted with increasing support orientation; the hind limb did not remain complaint and instead became significantly retracted in primates travelling on supports of 60° and steeper. Speed was generally reduced as support orientation increased.

Overall this study demonstrated that strepsirrhine primates capable of headfirst descent span a range of body masses up to approximately 4 kg in this sample. Across this range of body masses common strategies for traveling headfirst on supports included adopting slow trotting gaits with extended periods of hind limb retraction and forelimb protraction. Species that did not perform headfirst descent may have been limited in their ability to perform this behavior by aspects of their anatomy, such as having relatively short arms or less developed muscles for pedal grasping compared to species that were more adept at headfirst descent.

Placing these results into an evolutionary context, a small primate ancestor would be least impacted in its ability to navigate using headfirst descent on supports of all angles, whereas a larger ancestral primate might have been limited in the arboreal supports it could have navigated headfirst. Leaping may have been an alternative to grasping mediated headfirst descent in very small early primates, while alternatives to headfirst descent, such as tail first decent, that were only observed in larger species might have emerged later as various primate lineages increased in body mass but retained the characters of primate origins, grasping feet and nails.