The usefulness of biogeographic comparisons for prioritizing experimental work on plant invasions: an example with Verbascum thapsus (common mullein)

Background/Question/Methods

Over the last decade, hundreds of controlled experiments have been conducted to determine the mechanisms that underlie plant invasions. A majority of these studies explore the role of enemy release (ER) or the evolution of increased competitive ability (EICA) in facilitating invasion. However in many cases, the applicability of the ER and EICA hypotheses to a given system is not grounded by patterns observed in the field, and yet they are tested to the exclusion of other hypotheses, for example those exploring the role of altered competitive regimes. Even more fundamentally, mechanistic studies are often performed without first confirming that the species in question performs demonstrably better in the introduced range. To address this gap, we assessed plant performance, herbivore enemy prevalence and attack, and competitive regime in more than 50 native and introduced populations of Verbascum thapsus (common mullein), an introduced weed in North America.

Results/Conclusions

We found that introduced mullein performs better than its native counterparts at both the population and individual scales. However, data describing enemy prevalence and attack do not clearly demonstrate enemy release. For example, introduced mullein had less chewing damage than natives, but exhibited higher levels of a common wounding response. Additionally, although introduced plants experience complete release from attack by one guild (snails) and significant reduction in attack by others (aphids, leafhoppers, and caterpillars), members of all the major aboveground feeding guilds that are present in the native range are also represented in the introduced range. In terms of competitive regime, we found that the community of plants co-occurring with mullein significantly differs between ranges, with forbs dominating native communities and grasses dominating introduced communities. Additionally, while native mullein density is negatively correlated with neighboring plant cover (indicating that competition reduces its performance), introduced mullein appears unaffected by co-occurring plants, possibly because the system is water-limited. These data suggest and equivocal role for enemy release in facilitating mullein’s invasion, while at the same time suggesting that alternative hypotheses testing the role of competition deserve investigation.