Does fecundity drive the evolution of insect diet?

We investigate whether egg load (a surrogate for fecundity) drives host specificity in a herbivorous insect. In many insects, including our study organism (Edith's checkerspot butterfly), both egg load and tendency to accept low-ranked hosts increase during each search for an oviposition site. Effects on host acceptance of egg load and passage of time are thereby potentially confounded. We conducted two experiments designed to disentangle these effects. In both experiments, we estimated the times of first acceptance of both a high-ranked and a low-ranked host, without allowing the insects to oviposit. In the first experiment, we measured egg load at the time of first acceptance of the low-ranked host. The later the time of first acceptance, the higher was the fecundity. We therefore reject the hypothesis that all insects accepted the low-ranked host at the same predetermined egg load. In the second experiment, we measured egg load 48 h after the high-ranked host was first accepted. We found no relationship between egg load and timing of acceptance of the low-ranked host. Insects with higher rates of egg accumulation did not accept the low-ranked host sooner. Taken together, these results suggest that acceptance of the low-ranked host is not driven directly by egg load. Rather, this acceptance results from some other process that is influenced by time since last oviposition. We conclude that there is no evidence to support the assumption that females with high rates of egg accumulation are more likely to accept low-ranked hosts.