In continental comparisons, a global invader increases in density and experiences less herbivore damage in non-native ranges

Escaping consumers, or enemy release, is widely assumed to be a driving force behind non-native invasion, despite a complex body of literature that supports enemy release in some cases but not in others. However, few studies of enemy release integrate range comparisons, within-range geography, performance of the invasive species, consumer damage, and consumer abundance. To integrate these aspects of enemy release, we studied Conyza canadensis at 20 sites across its native range of North America and at 17 sites across its non-native range of Eurasia. Conyza population density was 7.0 ± 3.2 individuals/m2 in the native range vs. 18.3 ± 3.3 in the non-native range, where plants were more than 60% taller. Concomitantly, Conyza experienced five to six times more consumer damage in its native range than in its non-native range. Decision tree models indicated that low latitudes of the sampled sites, and the greater height and specific leaf area of Conyza plants best predicted herbivore damage, but these did not segregate between the native range and the non-native range. In contrast, only native and non-native ranges were clearly differentiated between high herbivore damage in the former and low damage in the latter. Our biogeographically explicit study is consistent with the enemy release hypothesis in that herbivore damage decreased in the non-native ranges and this corresponded with an increase in population density and plant size.