Testing assumptions of the enemy release hypothesis: generalist versus specialist enemies of the grass Brachypodium sylvaticum.

The enemy release hypothesis (ERH) suggests greater success of species in an invaded range due to release from natural enemies. The ERH assumes there will be more specialist enemies in the native range and that generalists will have an equal effect in both ranges. We tested these assumptions with the grass Brachypodium sylvaticum in the native range (Switzerland) and invaded range (Oregon, USA). We assessed all the kinds of damage present (caused by fungi, insects, mollusk and deer) on both leaves and seeds at 10 sites in each range and correlated damage with host fitness. Only two of the 20 fungi found on leaves were specialist pathogens, and these were more frequent in the native range. Conversely there was more insect herbivory on leaves in the invaded range. All fungi and insects found on seeds were generalists. More species of fungi were found on seeds in the native range, and a higher proportion of them were pathogenic than in the invaded range. There were more kinds of enemies in the native range, where the plants had lower fitness, in accordance with the ERH. However, contrary to assumptions of the ERH, generalists appear to be equally or more important than specialists in reducing host fitness.