Metacommunity theory predicts that intermediate dispersal rates among local habitat patches increases the persistence of prey populations in the presence of a predator. There are few field experimental tests of these predictions, however, because of the difficulty in controlling dispersal rates of predators and prey. Using experimental mesocosms, we determined the predation rate of a generalist predator (damsel bugs, Nabis spp.), on two leafhopper species (Agallia constricta and Ceratagallia agricola), which are pests on red clover (Trifolium pratense). Second, we examined if intraguild predation was present in damsel bugs in the presence of C. agricola in simple and complex habitats. Finally, we tested the effects of intermediate dispersal rates and host-plant specificity on the persistence of leafhopper (A. constricta) populations by varying the dispersal of leafhoppers and by providing orchard grass (Dactylis glomerata) as an alternate host plant to red clover.

Results/Conclusions The damsel bug showed a sigmoidal functional response on A. constricta with a high saturation density and a linear functional response on C. agricola. There was no evidence of intraguild predation in damsel bugs, and increased habitat complexity decreased the per capita predation rate of the damsel bug on C. agricola. Dispersal decreased the predation rate on A. constricta compared to the no-dispersal treatment. Varying the host plants available to leafhoppers had no significant effect on leafhopper survival. Our findings that intermediate dispersal rates decreased the predation rate compared to the no-dispersal treatment are consistent with predictions from theory. The absence of intraguild predation is contrary to what has been observed in other studies of damsel bugs, however, which may be a result of abundant leafhopper prey. Finally, the number of leafhoppers eaten by damsel bugs was larger than is reported elsewhere, which has implications for the role of damsel bugs as natural enemies of leafhoppers in forage-crop systems.