Resource Allocation, Coexistence, and the Niche Structure of a Streambank Salamander Community

Abstract

A streambank salamander community has proved to be an ideal system for investigating resource allocation and the evolution of community organization. Four aspects of microhabitat were investigated: substrate, substrate moisture, proximity to surface water, and size/type of cover utilized. Selection of substrate particle size was strongly positively correlated intra— and interspecifically (particularly the latter) with salamander body size, appearing to be a highly adaptive feature. The breadth or diversity of utilization of microhabitat components was interspecifically inversely related to body size. Definite shifts in microhabitat utilization and narrowing of niche (resource) breadths were observed in the presence of larger congeners. The mechanism of the shifts as well as the selective force that molded the organization of this community appear to be interspecific interference competition. Therefore, body sizes and species compositions in streambank salamander communities are determined by substrate size, its patchiness, and the presence of interference competitors. Observed macrohabitat distribution patterns have been analytically described and can be predicted by a knowledge of microhabitat preferences and niche breadths. The ratios of snout—vent lengths or head widths between adjacent species on a body size gradient for both adults and juveniles were remarkably constant, generally between 1.2 and 1.3. Mean prey size (width, length, and volume) also formed a geometric spacing which paralleled body sizes. The coefficient of variation (standard deviation/mean) was constant within a given measure of prey size. Morphological ratios for sympatric competitors in the range of 1.2 to 1.3 have been interpreted by investigators as character displacement to allocate prey sizes and therefore minimize competition. Such interpretations must be made carefully because in this study observed body size differences primarily reflect microhabitat allocation (the principal niche dimension of this study) and only secondarily allocation of prey sizes. Also, species packing (d/w) along the prey size resource gradient is much greater than anticipated by the theoretical May—MacArthur model and is demonstrated to be high in a community of opportunistic generalist insectivores. Species packing also requires careful interpretation because in reality spacings and widths of utilization functions are not uniform and separation of community members along a single niche dimension is rarely if ever encountered. Total niche overlap among community members (species and age classes) can accurately be determined because the relationships among the four microhabitat resources and the prey size dimension are known. The use of prey taxa as a niche dimension for opportunistic generalistic insectivores either overestimates or underestimates total niche overlap, since the utilization of prey items is a function of their size, the microhabitat preference of the predator, and stochastic events of prey occurrences. The community appears to be well coadapted since the following have been demonstrated: (1) total niche overlap is low, therefore minimizing competitive interactions, (2) interspecific predation could not be documented and must be minimal, and (3) demographic strategies are consistent in all macrohabitats regardless of congeneric composition.