Disentangling effects of induced plant defenses and food quantity on herbivores by fitting nonlinear models.

Plants can respond to herbivore damage through both broad-scale (systemic) and localized induced responses. While many studies have quantified the impact of systemic responses on herbivores, measuring the impact of localized changes is difficult because plant tissues that have suffered direct damage may represent both a lower quality and a lower quantity of food. This article uses nonlinear models to disentangle the confounding effects of prior herbivory on food quantity and quality. The first (null) model assumes that herbivore performance is determined only by the quantity of food available to an average herbivore. Modified models allow two distinct effects of damage-induced defenses: an increase in the amount of food each herbivore is required to consume in order to achieve maximum performance and a reduction in the maximum performance even when herbivores are fed ad lib. Maximum likelihood methods were used to fit the models to data from field experiments in which Colorado potato beetle (Leptinotarsa decemlineata) larvae were reared on three varieties of potatoes that had been damaged to varying degrees by adult beetles. Prior damage reduced the mean mass of beetles at pupation, and this effect was due to both a decrease in food quantity and induced changes in food quality. In contrast, beetle survival was affected in some cases by reduced food quantity but showed no responses that could be attributed to induced defenses. I discuss this result in the context of previous studies of induced (mostly systemic) responses in the potato-potato beetle system, and I suggest that detailed studies of particular chemical responses and the proposed method of combining bioassays with quantitative models should be used as complementary approaches in future studies of herbivore-induced defenses in plants.