A venerable generalization about community resistance to invasions is that more diverse communities are more resistant to invasion. However, results of experimental and observational studies often conflict, leading to vigorous debate about the mechanistic importance of diversity in determining invasion success in the field, as well as other ecosystem properties, such as productivity and stability. In this study, we employed both field experiments and observational approaches to assess the effects of diversity on the invasion of a subtidal marine invertebrate community by three species of nonindigenous ascidians (sea squirts). In experimentally assembled communities, decreasing native diversity increased the survival and final percent cover of invaders, whereas the abundance of individual species had no effect on these measures of invasion success. Increasing native diversity also decreased the availability of open space, the limiting resource in this system, by buffering against fluctuations in the cover of individual species. This occurred because temporal patterns of abundance differed among species, so space was most consistently and completely occupied when more species were present. When we held diversity constant, but manipulated resource availability, we found that the settlement and recruitment of new invaders dramatically increased with increasing availability of open space. This suggests that the effect of diversity on invasion success is largely due to its effects on resource (space) availability. Apart from invasion resistance, the increased temporal stability found in more diverse communities may itself be considered an enhancement of ecosystem function.

In field surveys, we found a strong negative correlation between native-species richness and the number and frequency of nonnative invaders at the scale of both a single quadrat (25 × 25 cm), and an entire site (50 × 50 m). Such a pattern suggests that the means by which diversity affects invasion resistance in our experiments is important in determining the distribution of invasive species in the field. Further synthesis of mechanistic and observational approaches should be encouraged, as this will increase our understanding of the conditions under which diversity does (and does not) play an important role in determining the distribution of invaders in the field.