Locomotion in captive Leontopithecus and Callimico: a multimedia study.

Video studies, gait analysis, footprint tracks, and observational scan sampling show that, in comparably furnished enclosures, Leontopithecus rosalia and Callimico goeldii are superficially similar in their use of predefined locomotor patterns but differ profoundly in many underlying details which reflect differences in postcranial morphology. Each uses pronograde arboreal quadrupedal walking, quadrupedal bounding, and vertical climbing with comparable frequency, and both shift to bounding while moving quadrupedally at high speeds. In walking, both species use a diagonal sequence gait. However, in Callimico the distance per bout traveled while walking or running is shorter than in L. rosalia and there is an emphasis on leaping (from a stationary position) and bounding-leaps (saltational extensions of pronograde quadrupedalism), in contrast with the basically quadrupedal style of L. rosalia. This dichotomy is consistent with anatomical specializations, such as forelimb elongation in Leontopithecus and hindlimb elongation in Callimico. In vivo hand- and footprint studies demonstrate grasping halluces in both species while walking. Limb stances in L. rosalia during "transaxial bounding" involve an overstriding hindlimb, a predominance of oblique rather than in-line travel, and unique hand and foot positions. Anatomically, this locomotor style may be associated with reduced dexterity of the elongate hands and a relatively short hallux. The captive locomotor profiles for both species probably reflect biased samples of the locomotor repertoire of their wild counterparts. Nevertheless, these data reflect species-specific integrations of locomotor behavior and morphology, and corroborate expectations of locomotor diversity among callitrichine primates, even those of similar body size. It is suggested, however, that conventional quantitative studies of locomotor profiles may prove inadequate for resolving subtle aspects of locomotor morphology and behavior.