A common hypothesis to explain plants' invasive success is that release from natural enemies in the introduced range selects for reduced allocation to resistance traits and a subsequent increase in resources available for growth and competitive ability (evolution of increased competitive ability, EICA). However, studies have rarely studied both the genetic underpinning of the response to enemy release, or they either did not test for all aspects of competitive ability or did not select appropriate competitors.
In a series of experiments we used several populations of Lythrum salicaria (purple loosestrife) from its native (Europe) and invasive range (North America).
Firstly we tested enemy release:
Methods: Populations were exposed to all above-ground herbivores in replicated natural populations in the native range. The experiment was performed both with plants raised from field-collected seeds as well as with offspring of these where maternal effects were removed.
Key results: Absolute and relative leaf damage was higher for introduced than for native plants. Despite having smaller height growth rate, invasive plants attained a much larger final size than natives irrespective of damage, indicating large tolerance rather than effective defence. Origin effects on response to herbivory and growth were stronger in second-generation plants, suggesting that invasive potential through enemy release has a genetic basis.
Conclusions The findings support two predictions of the EICA hypothesis – a genetically determined difference between native and invasive plants in plant vigour and response to enemies – and point to the importance of experiments that control for maternal effects and include the entire spectrum of native range enemies.
Secondly we tested inter- and intra- specific competition:
Methods: In a set of common-garden experiments we both distinguished between competitive effect and response of invasive and native plants, and also by using both intraspecific and interspecific competition settings with a highly vigorous neighbour, Urtica dioica (stinging nettle), which occurs in both ranges.
Key Results While the intraspecific competition results showed no differences in competitive effect or response between native and invasive plants, the interspecific competition experiment revealed greater competitive response and effect of invasive plants in both biomass and seed production.
Conclusions The use of both intra- and interspecific competition experiments in this study revealed opposing results. While the first experiment refutes the EICA hypothesis, the second shows strong support for it, suggesting evolutionarily increased competitive ability in invasive populations of L. salicaria. It is suggested that the use of naturally co-occurring heterospecifics, rather than conspecifics, may provide a better evaluation of the possible evolutionary shift towards greater competitive ability.