Direct and Interactive Effects of Enemies and Mutualists on Plant Performance: A Meta-Analysis

Abstract

Abstract. Plants engage in multiple, simultaneous interactions with other species; some (enemies) reduce and others (mutualists) enhance plant performance. Moreover, effects of different species may not be independent of one another; for example, enemies may compete, reducing their negative impact on a plant. The magnitudes of positive and negative effects, as well as the frequency of interactive effects and whether they tend to enhance or depress plant performance, have never been comprehensively assessed across the many published studies on plant-enemy and plant-mutualist interactions. We performed a meta-analysis of experiments in which two enemies, two mutualists, or an enemy and a mutualist were manipulated factorially. Specifically, we performed a factorial meta-analysis using the log response ratio. We found that the magnitude of (negative) enemy effects was greater than that of (positive) mutualist effects in isolation, but in the presence of other species, the two effects were of comparable magnitude. Hence studies evaluating single-species effects of mutualists may underestimate the true effects found in natural settings, where multiple interactions are the norm and indirect effects are possible. Enemies did not on average influence the effects on plant performance of other enemies, nor did mutualists influence the effects of mutualists. However, these averages mask significant and large, but positive or negative, interactions in individual studies. In contrast, mutualists ameliorated the negative effects of enemies in a manner that benefited plants; this overall effect was driven by interactions between pathogens and belowground mutualists (bacteria and mycorrhizal fungi). The high frequency of significant interactive effects suggests a widespread potential for diffuse rather than pairwise coevolutionary interactions between plants and their enemies and mutualists. Pollinators and mycorrhizal fungi enhanced plant performance more than did bacterial mutualists. In the greenhouse (but not the field), pathogens reduced plant performance more than did herbivores, pathogens were more damaging to herbaceous than to woody plants, and herbivores were more damaging to crop than to non-crop plants (suggesting evolutionary change in plants or herbivores following crop domestication). We discuss how observed differences in effect size might be confounded with methodological differences among studies.

Third-party mutualists have contrasting effects on host invasion under the enemy-release and biotic-resistance hypotheses

“…For the ERH, not only were initial similarity of the herbivore and the microbe to the plant significant factors in determining plant invasion success, but the interaction between microbe and herbivore was as well. This mirrors empirical studies that have found non-additive affects when investigating PMI interactions (Morris et al 2007;Bennett 2013). This is most likely related to the plant having a harder time invading because benefits gained from being similar to the microbe are canceled by the increased similarity to the herbivore under the BRH.…”

Herbivory reduces plant interactions with above‐ and belowground antagonists and mutualists

“…Damage treatments, which mimicked the natural timing of A. vittatum feeding in this system, may have occurred so early in the plants' development that the resulting resource loss was more than the plants could tolerate. In some cases, mutualists can mitigate the negative impacts of plant enemies such as herbivores (Morris et al 2007), although this did not occur with C. sativus pollinators. In agricultural systems, cultivating mutualisms has been proposed as a method to increase yields and counteract the negative effects of plant damage (Kremen et al 2002, Strauss andMurch 2004).…”

“…Additional interactions between other groups (e.g., between above-and belowground pollinators; Barber et al 2011) are possible but beyond the scope of this study, as they would require additional factorial manipulations. Nonetheless, such experiments manipulating multiple interactors will be necessary in the future to quantify the contributions of indirect feedbacks to net plant effects as well as to address the possibility of nonadditive effects among community members (Morris et al 2007). …”

Interactive effects of multiple climate change variables on trophic interactions: a meta-analysis

“…We based our analyses of multiple interacting climate change variables on an underlying multiplicative model, as opposed to a more conservative additive model [74], because multiplicative models are more ecologically realistic representations of interacting effects [75] and the log response ratio itself is based on a multiplicative model [76]. Interaction effects were classified as multiplicative (the null hypothesis) if the 95% CI around the interaction effect size (L ab ) contained zero.…”