Predation, apparent competition, and the structure of prey communities

It is argued that alternate prey species in the diet of a food-limited generalist predator should reduce each other's equilibrial abundances, whether or not they directly compete. Such indirect, interspecific interactions are labeled apparent competition. Two examples are discussed in which an observed pattern of habitat segregation was at first interpreted as evidence for direct competition, but later interpreted as apparent competition resulting from shared predation. In order to study the consequences of predator-mediated apparent competition in isolation from other complicating factors, a model community is analyzed in which there is no direct interspecific competition among the prey. An explicit necessary condition for prey species coexistence is derived for the case of one predator feeding on many prey species. This model community has several interesting properties: (1) Prey species with high relative values for a parameter r a are "keystone" species in the community; (2) prey species can be excluded from the community by "diffuse" apparent competition; (3) large changes in the niche breadth of the predator need not correspond to large changes in predator density; (4) the prey trophic level as a whole is regulated by the predator, yet each of its constituent species is regulated by both the predator and available resources; (5) increased productivity may either increase, decrease, or leave unchanged the number of species in the community; (6) a decrease in density-independent mortality may decrease species diversity. These conclusions seem to be robust to changes in the prey growth equations and to the incorporation of predator satiation. By contrast, adding prey refugia or predator switching to the model weakens these conclusions. If the predator can be satiated or switched, the elements a _ij comprising the community matrix may have signs opposite the long-term effect of j upon i. The effect of natural selection upon prey species coexistence is discussed. Unless r _i , K _i , and a _i are tightly coupled, natural selection within prey species i will tend to decrease the equilibrial abundance of species j.