Search for a command to run...
According to the habitat—matrix model, arboreal microhabitat specialists are adapted behaviorally and morphologically for locomotion in different subsets of the vegetation, each characterized by its three—dimensional structure, and these adaptations help explain morphological patterns among coexisting species. I tested this model's predictions for Anolis lizards near Monteverde, Costa Rica. Anolis humilis, A. tropidolepis, and A. woodi were active at different heights in the shaded forest understory, and A. insignis inhabited the canopy. Gap specialists, A. altae and A. intermedius, resembled one another in microhabitat use but were largely separated by elevation. Adult males of forest—understory and gap species were active higher above ground than adult females, which averaged higher than juveniles. In their different structural environments, species, sexes, and age classes differed in proportional use of locomotor modes: running, jumping, and crawling. During field observations of forest—understory and gap species, frequency of crawling was highest for anoles that used slender, widely spaced supports, though much variation in crawling frequency was unexplained. Frequency of jumping increased as mean support size and average distance between supports decreased; the latter variable, expressed relative to body length, accounted for most of the variation in jump frequency. With enclosure experiments, I assessed proximate effects of microhabitat features on locomotor behavior and removed these effects to test for interspecific differences in intrinsic locomotor tendencies. Jump frequency of A. altae consistently increased with decreasing distance between supports, whereas the effects of support diameter were more complex, and varied with spacing of supports. Support diameter exerted both a surface—area effect (a lower tendency to jump from larger supports compared with smaller ones) and, where supports were widely spaced, a target—size effect (a higher tendency to jump to larger supports compared with smaller ones). Compared in the same array of supports, A. tropidolepis, A. altae, and A. intermedius, although similar in body size, differed in frequency of jumping. Thus, differences in locomotor behavior among species reflected not only proximate influences of vegetation structure but also intrinsic tendencies. Morphological traits were strongly associated with locomotor behavior and microhabitat specialty. Differences in limb proportions of primarily running anoles (A. altae and A. intermedius), crawling anoles (A. insignis), and anoles that jumped frequently (A. humilis, A. tropidolepis, and A. woodi) accorded with predictions from biomechanics. Body size was also functionally related to locomotion and was correlated with microhabitat structure. Thus, ability to exploit various structural environments may depend not only on body shape but also size. Nonrandom patterns of interspecific differences in morphology suggested limiting similarity without revealing the underlying ecological process(es). The mechanistic basis for these patterns,however, appears to lie, at least in part, in the functional relationships between morphology and habitat structure.