COS 74-10
Food quality of host-plant species alters ant-caterpillar predation strength in a temperate forest community

Wednesday, August 12, 2015: 11:10 AM
344, Baltimore Convention Center
Michael S. Singer, Biology Department, Wesleyan University, Middletown, CT
Emily R. Johnson, Biology Department, Wesleyan University, Middletown, CT
Robert E. Clark, Biology Department, Wesleyan University, Middletown, CT
Background/Question/Methods Variation in plant traits can alter the suppression of herbivores by carnivores via two major mechanisms. Plants can change the density of herbivores or carnivores, thus altering their ratio and the outcome of their interaction. Alternatively, plants can change the traits of herbivores or carnivores, thus altering their per capita interaction strength. Here we study how variation among tree species alters herbivore suppression by ants in a temperate forest community via density-related and trait-related mechanisms. One trait-related mechanism of particular interest is the slow-growth/high-mortality hypothesis, which proposes that poor food quality of plants can increase herbivore predation risk. We tested predictions of these mechanisms by quantifying effects of ant exclusion on caterpillar density across eight common tree species in a 2-year field experiment. We also measured ant density and calculated the per capita predation strength between ant and caterpillar communities on each tree species. To specifically address the slow-growth/high-mortality hypothesis, we used a caterpillar growth performance metric to estimate each tree species' food quality for caterpillars.

Results/Conclusions We found that tree species variation alters ant-caterpillar predation strength via the trait-related mechanism. The effect of ant exclusion on caterpillar abundance varied among the eight tree species in each year. Although the densities of caterpillars and ants varied among tree species, the ratio of ant density to caterpillar density did not because ant density tracked caterpillar density. Predation strength per tree species did not depend on caterpillar density, ant density, or their ratio. However, predation strength per tree species did depend on the trees' food quality for caterpillars, consistent with the slow-growth/high-mortality hypothesis. These results contrast with the mechanism by which food quality of host-plant species alters the strength of bird predation of caterpillars in the same community: overcompensating density-dependent predation.