We measured herbivore abundance on two species of willows (Salix sericea and S. eriocephala) and their interspecific hybrids in two common garden experiments to test alternative hypotheses concerning the genetic effects of hybridization on plant resistance. The first experiment used potted clones of naturally occurring parental and hybrid plants. Purity of parental plants was determined using 20 RAPD markers (pure S. eriocephala had a score of 1.0 and pure S. sericea had a score of 0). The cloned hybrid plants had hybrid scores between 0.40 and 0.60, indicating they were intermediate hybrids and probably had recombinant (F2-type) genotypes. The second experiment used potted one-year-old interspecific F1 progeny and intraspecific progeny from crosses between genetically pure parents. We counted numbers of herbivores on plants after exposure to attack for about 2 months. Herbivore abundance data were analyzed with ANOVA and a priori contrasts were used to test fit of the data to 5 hypotheses (No Difference, Additive, Dominance, Hybrid Susceptibility, or Hybrid Resistance). We found genetic effects of hybridization on susceptibility to some herbivores in both experiments, with the additive, dominance, and hybrid susceptibility hypotheses supported by different herbivore species. There was also significant genetic variation among clones and genotypes within hybrid and parental taxa for herbivore resistance. For several herbivore species we found no difference in the susceptibility among hybrids and parents, which indicates genetic effects of hybridization on resistance were absent for these species. Differences between the hypotheses supported from this study and data from four years of prior field studies suggest that environmental variation is an important factor affecting relative hybrid and parental susceptibility in the field.