Determinants of polyphagy by a woolly bear caterpillar: a test of the physiological efficiency hypothesis

The physiological efficiency hypothesis argues that the physiological efficiency of food utilization determines feeding habits of herbivorous insects. Although relatively unsuccessful at explaining dietary specificity, it may explain the food‐mixing habit of individually polyphagous herbivores because they may opportunistically increase physiological efficiency by optimizing nutrient balance or diluting toxins in the course of feeding on multiple host‐plant species. This study tests two predictions of the physiological efficiency hypothesis with the woolly bear caterpillar, Grammia geneura (Strecker) (Lepidoptera: Arctiidae). Namely, both herbivore performance (survival, developmental rate, pupal mass) and growth efficiency should be better on mixed‐plant diets than on single‐plant diets. In a series of three laboratory experiments, I found that larval survival and developmental rate on mixed‐plant diets were superior to performance on single‐plant diets in only one of four cases. In all other cases, mixed‐plant diets were either inferior (female pupal mass) to single‐plant diets or not detectably different from them. Larval growth efficiency on mixed‐plant diets was never superior to that on single‐plant diets. In mixed‐plant treatments, caterpillars often selected a diet that included plant species of relatively low suitability alone, thereby suffering reduced performance and growth efficiency. These results contradict predictions of the physiological efficiency hypothesis, indicating the limitations of the conventional focus on the physiological constraints on food utilization as an explanation for both individual polyphagy and dietary specificity in herbivorous insects.