Mycorrhizal disruption as an adaptive strategy in plant invasions: a critique

Some plants might become invasive because they produce toxic allelochemicals that disrupt mycorrhizal connections with native competitors, thereby increasing availability of limiting nutrient. In such a scenario, eco-evolutionary feedbacks during invasion could evolve if these allelochemicals are costly and mycorrhizae quickly adapt. On the other hand, investment into allelochemical production usually requires significant energy (i.e., carbohydrates) and limiting nutrients (e.g., N, P, S), and involves significant costs, including loss to metabolism of soil microbiota and opportunity costs to growth, reproduction, and mitigating biotic and abiotic stress. Using field surveys and manipulative experiments with Alliaria petiolata to directly for eco-evolutionary feedbacks, we find instead that mycorrhizal disruptions are virtually undetectable. Instead, phenotypic plasticity, herbivory, and maternally inherited defences may create the appearance of eco-evolutionary dynamics in allelopathy for glucosinolates that serve other important ecological roles. Using simulations and common garden studies with Lythrum salicaria (purple loosestrife) and other invasive plants, we find that phenotypic differences can evolve rapidly along abiotic gradients within the introduced range through both adaptive and stochastic processes. This kind of spatial autocorrelation in quantitative traits can be misinterpreted as evidence for eco-evolutionary dynamics, and it can lead to incorrect rejection of the null hypothesis that native and introduced populations are the same, on average. Notwithstanding the bold claims of many reviews and original research papers, hard evidence that mycorrhizal disruption plays an important ecological role in plant invasion can be largely explained as either (i) side effects of active chemicals that evolved for other purposes (e.g., pathogen or herbivore defence, phenological timing), or (ii) provocative and highly significant results that have been retracted or have yet to be independently validated.