INTRAGUILD PREDATION AMONG BIOLOGICAL-CONTROL AGENTS - THEORY AND EVIDENCE

Theoretical and empirical evidence developed in four subdisciplines of biological control (biocontrol of plant pathogens, weeds, nematodes, and arthropods) is brought to bear upon a shared question: the signifi cance of intraguild predation. Intraguild predation (''IGP'') occurs w hen two species that share a host or prey (and therefore may compete) also engage in a trophic inter-action with each other (parasitism or p redation). We describe the prevalence of IGP and its role in the popul ation dynamics of biological-control agents and target pests. IGP is a widespread interaction within many, but not all, communities of biolo gical-control agents. IGP appears to be pervasive among communities of control agents associated with nematode or arthropod pests. Common fo rms of IGP include pathogens that infect both herbivores and parasitoi ds of the herbivore; facultative hyperparasitoids, which can parasitiz e either an herbivore or a primary parasitoid of the herbivore; predat ors that attack herbivores that harbor a developing parasitoid; and pr edators that attach each other. In contrast, IGP appears to be relativ ely uncommon among biological-control agents of plant pathogens becaus e trophic interactions are less important than competition or antibios is. Likewise, biological-control agents of weeds interact primarily th rough competition alone because host ranges are mostly restricted to p lant taxa. Empirically based simulation models and general analytical models of interactions involving arthropod pathogens or facultative hy perparasitoids yield variable and often conflicting predictions for th e influence of IGP on the success of biological control. Models for pr edator-predator interactions, however, consistently predict that IGP d isrupts biological control. All the field-documented cases of IGP lead ing to disruption of biological control stem from studies of predators , including mites, insects, and predatory fishes. IGP between two pred ators or between a predator anti an adult parasitoid does not require mortality of the shared prey/host (i.e., the target pest); thus, IGP c an be intense, resulting in high levels of mortality for one or both o f the natural enemies, while the total mortality imposed on the target pest population is minimal. For this reason, we hypothesize that IGP by predators is particularly likely to influence the efficacy of biolo gical control. Our ability to develop successful programs of biologica l control will be enhanced by field studies that address the complexit y of trophic interactions occurring in agroecosystems. There is a crit ical need for additional manipulative experiments conducted in the fie ld that test not only population ecology theory for two-species intera ctions, but also community ecology theory for multispecies interaction s. (C) 1995 Academic Press, Inc.