Answers are grounded in this paper and cite their sources.
‘… the study elegantly links, and in fact, unifies two of the most popular hypotheses attempting to explaining plant invasiveness, the EICA hypothesis and the novel weapons hypothesis …’ Uesugi & Kessler studied tall goldenrod (Solidago altissima), a perennial plant native to North America but highly invasive in Europe and other parts of the world. They studied goldenrod in its native range, in a 12-yr herbivore-exclosure experiment in an old field in eastern North America (Fig. 1). Since all study plots were initially established from the same goldenrod genotypes, any genetic differentiation observed between herbivore-exclusion and control plots can be causally linked to the experimental treatments, and it is thus evidence of rapid evolution in response to changes in herbivory. Uesugi & Kessler sampled offspring from each of the study plots and showed that in a common environment, goldenrod plants from herbivore-exclusion plots performed significantly better in the presence of interspecific competitors, and had substantially higher root concentrations of polyacetylenes, putative allelopathic agents of goldenrod, than plants from control plots. They also showed that the most abundant of the polyacetylenes strongly reduced germination and growth of an interspecific competitor in a bioassay. Taken together, this demonstrates that the release from natural enemies (created by insecticides) caused rapid evolution of increased competitive ability in goldenrod, and it strongly suggests that the underlying mechanism might be selection for increased production of polyacetylene, an allelochemical suppressing interspecific competitors. One intriguing aspect of the study of Uesugi & Kessler is that it elegantly links, and in fact, unifies two of the most popular hypotheses attempting to explaining plant invasiveness, the EICA hypothesis and the novel weapons hypothesis (Callaway & Ridenour, 2004), which proposes that some highly invasive plants may be dominant because they possess novel allelochemicals to which their new neighbours have not previously been exposed, and which therefore have particularly strong negative effects on these natives. Callaway & Ridenour (2004) already pointed out the possible link between novel weapons and EICA. They hypothesized that if allelopathy indeed confers a substantial fitness advantage, it might evolve rapidly after enemy release and could therefore, in some cases, be the key mechanism underlying EICA. However, at the time no empirical evidence existed for such an evolutionary change, and the EICA and Novel weapons hypothesis have largely been studied independently since then. Still, a link between enemy release, EICA and novel weapons is very likely, because both enemy release and novel weapons are expected to be strongly influenced by the phylogenetic distinctness of plant invaders (Fig. 2). Introduced plants that are phylogenetically distinct from their recipient communities are not only more likely to possess novel allelochemicals but also a chemical defence novel to the herbivores in the introduced range. As a consequence, phylogenetically distinct invaders experience greater enemy release (Cappuccino & Carpenter, 2005; Hill & Kotanen, 2009; Ness et al., 2011), and they should thus not only possess novel weapons but also the means for their rapid evolutionary increase. Because of this double advantage we should expect evolution of increased allelopathy to be common in phylogenetically distinct invaders, and it may well contribute to the frequently observed associations between phylogenetic novelty, chemical novelty and plant invasion success (Cappuccino & Arnason, 2006; Strauss et al., 2006). However, the study of Uesugi & Kessler has not been done in invasive populations, but in a native population of goldenrod, and it is now important to test whether evolution of increased allelopathy indeed also occurred in response to natural enemy release. Of course the interpretation of Uesugi & Kessler's results rests on the assumption of a trade-off between allocation to defence vs allelopathy, and their study did not provide any formal proof that such a trade-off exists. Providing these missing pieces of evidence will be crucial next steps towards understanding goldenrod invasion. Another important direction for future research should be to test for increased allelopathy of invasive populations across multiple invasive species, connect these data to phylogenetic distance, enemy release and invasion success, and thereby test the generality of Uesugi & Kessler's findings. Even though evolution experiments are the gold standard for testing eco-evolutionary hypotheses, and their usefulness for ecologists has long been recognized (Conner, 2003), evolution experiments in the field are still quite rare (Kawecki et al., 2012). The study of Uesugi & Kessler, together with other recent studies (Agrawal et al., 2012), clearly demonstrate the power of the experimental evolution approach. It also shows the strength of combining multiple approaches, in their case a long-term selection experiment in the field with common garden studies, bioassays and chemical analyses. Ultimately, it is such combinations of well-designed ecological and evolutionary experiments with thorough analyses of the underlying functional mechanisms – combining realism with precision, ‘why?’ and ‘how?’ questions – that most advance our understanding of ecological systems.
No takes yet. Share an insight, caveat, or question.
Oliver Bossdorf (2013) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: