THEORETICAL COMPARISONS OF INDIVIDUAL SUCCESS BETWEEN PHENOTYPICALLY PURE AND MIXED GENERALIST PREDATOR POPULATIONS

We investigated by individual-based simulations the role of individual variation in spider predatory foraging strategies on prey captures an d spider fecundity. We created 18 generalist intraspecific spider phen otypes by crossing 3 levels of aggressiveness with 6 strategies of pre y species selection. These phenotypes were tested in 1 and 4 patches i n a fixed area habitat and with 2 and 5 prey species. In a first group of simulations, spiders of all phenotypes competed for prey in a mixe d-phenotype and stochastic environment. Prey selection strategy had a highly significant effect on predator success. Aggressiveness had no d irect effect. Individual variation within phenotypes was high. Habitat fragmentation caused the decline of prey numbers and, thus, prey capt ures and spider fecundity. Increased prey richness enhanced spider fec undity. The phenotype that achieved the highest fecundity and captured the most prey exhibited high aggressiveness while specializing on the prey type which was statistically predicted to be numerically promine nt in the next time interval. The phenotype that did the least well in the system was not aggressive and specialized on a randomly chosen pr ey type per time interval. In the second set of simulations, we comple ted separate trial runs of the ''best'' and ''worst'' phenotypes for s pider populations composed of a single phenotype. Here, the aggressive and statistical strategy captured less prey and was less fecund than the non-aggressive, random phenotype. Thus, predation results from pur e phenotype populations do not apply to mixed-phenotype ones.