Vegetational diversity is generally thought to be associated with ecosystem stability and resilience to perturbations such as insect outbreaks. The enemies' hypothesis states that vegetational diversity contributes to greater top–down control of insect pests, by providing further resources to natural enemies than homogeneous environments. However, direct evaluation of this hypothesis is difficult because different species of natural enemies can respond to vegetational diversity in dissimilar manners and at different spatial scales depending on functional traits such as prey/host specificity and dispersal. In this study, we specifically test the enemies' hypothesis at the landscape level in a continuous forest environment. We investigated how parasitism of spruce budworm larvae by the common parasitoids Apanteles fumiferanae and Glypta fumiferanae vary with forest diversity and host larval density at different spatial scales in the province of Quebec (Canada). We found that parasitism rates of the two parasitoid species we examined respond in opposite ways to forest diversity. Parasitism by A. fumiferanae was positively related to forest diversity, whereas parasitism by G. fumiferanae was negatively related to forest diversity. In agreement with the enemies' hypothesis, we also found that spruce budworm larval density decreased with forest diversity. We discuss these results with respect to the enemies' hypothesis and the presumed host range of the parasitoids species we examined, as well as their body size. Because A. fumiferanae kills its host earlier than G. fumiferanae, we conclude that northern forest landscapes could be more affected by spruce budworm defoliation than southern forests during the present and future outbreaks.