MODEL AND FIELD-TEST OF PREY CONTROL EFFECTS BY SPIDER ASSEMBLAGES

We investigated by computer simulations and a field experiment the hyp othesis that a species assemblage of generalist predators imposes more mortality on prey populations than a single species does. The simulat ion tracked the energetics of individual spiders feeding on prey in a patchy and variable habitat. Simulation tests showed that increasing t he number of spider species and the variability of prey body sizes con tributed significantly to greater prey limitation and spider survival. Higher variability in spider body sizes had the reverse effect and in creasing the number of prey species had mixed effects. Large spiders, although less abundant than smaller ones, cropped disproportionally mo re prey biomass by capturing large prey and should have injured, thus reduced the capture rate, of smaller spiders. Simulations also showed that as model habitats were fragmented prey biomass dropped, causing a reduction in spider success variables. Increasing fragmentation enhan ced the contribution of built-in negative environmental effects on the growth rate of prey. We showed with a factorial field experiment that lower prey biomass resulted from greater spider species richness, as predicted, higher prey richness increased prey biomass only if spiders were present, and habitat fragmentation only decreased prey biomass i n the absence of spiders. At equal spider biomass, larger spiders, whi ch were present only in the highest richness treatment, probably fed m ore than smaller ones. Only larger spiders could capture the largest p rey, which represented a greater fraction of the total biomass. Spider predation seemed indifferent to fragmentation.