A meta‐analysis of insularity effects on herbivory and plant defences

Abstract

Aim Plants on islands are often subjected to lower levels of herbivory relative to those found at mainland sites. As a consequence, island plants are predicted to exhibit lower levels of physical or chemical defences, which renders them more susceptible to introduced herbivores. Yet, instances of high pressure by superabundant herbivores native to islands have been reported in many insular systems, which presumably would result in heightened plant defences. To date, no quantitative review has been conducted to determine how common these contrasting patterns are and their implications for the evolution of plant‐herbivore interactions. Location Islands worldwide. Taxon Plants, insects, molluscs, mammals. Methods We conducted a meta‐analysis of insularity effects on herbivory and plant defences by including studies that involved island‐mainland comparisons of the same plant species in both environments (90% of cases), or insular endemics versus mainland congeners (10% of cases). We tested for differences between mammalian and invertebrate (mollusc or insect) herbivory as well as between plant chemical and physical defences by specifying comparisons based on the type of herbivore (vertebrate or invertebrate) or plant trait included in the study. Results Mammalian herbivory was significantly higher on islands than on mainlands. In contrast, no significant effect was observed on invertebrate herbivory. In addition, we found no significant difference in either plant physical or chemical defences between insular and mainland plants, though physical defences tended to be higher for plants on islands. Main conclusions All analysed mammal studies focused on species introduced to islands, suggesting greater susceptibility of insular plants to exotic mammals, whereas the lack of effects in the case of invertebrate herbivory suggests no difference in susceptibility to molluscs or insects between insular and mainland plants. Interestingly, plant trait patterns suggest a trend for increased physical defences by insular plants, possibly due to heightened pressure by exotic mammalian herbivores on islands, whereas chemical defences appear uncorrelated to differences in herbivory. These findings call for further experimental and observational studies measuring defences and herbivory for multiple sympatric plant species occurring at both mainland and island sites within a system, or comparing insular endemics to congeneric mainland species.

A roadmap for future research on insularity effects on plant–herbivore interactions

“…As a result, many insular plant species exhibit defensive (or seed dispersal) anachronisms, such as in New Zealand forests (Burns, 2016(Burns, , 2019 or mediterranean (e.g., Balearic and Canary Islands) shrublands (Capó et al, 2021;Cubas et al, 2019;Irl et al, 2012) shaped by extinct species of birds or ungulates, respectively. Likewise, introductions of non-native mammals in the last few centuries could also affect patterns of plant defence investment (Moreira et al, 2021).…”

“…Further, a recent meta‐analysis of 21 studies comparing mainland versus insular plant populations found that vertebrate herbivory (mainly by introduced mammals) was greater on islands. Invertebrate herbivory and physical and chemical plant defences (see Glossary) did not differ between islands and the mainland (although physical defences tended to be higher on islands) (Moreira et al., 2021). Thus, although studies remain limited, the patterns obtained so far, based on available data, do not provide a consistent picture and, in many cases, are unsupportive of predictions.…”

“…Most studies to date have been restricted to specific systems or regions (80% in temperate areas; Figure 1) and often to comparisons using a single plant species and one island versus mainland site (Table 1; Moreira et al., 2021). Although meta‐analytical tools can be used to compare patterns across multiple systems, there are intrinsic limitations in doing so, and reaching strong conclusions is frequently limited by heterogeneity among studies.…”

“…Map showing the main insular systems of the 21 studies used in a recent meta‐analysis comparing plant defences and herbivory in mainland versus insular plant populations (Moreira et al., 2021). The numbers in circles represent the study identities included in Table 1.…”

“…Insularity research to date has been strongly biased toward vertebrate herbivory (close to 70% of the studies; Table 1; Moreira et al., 2021). However, invertebrate herbivory, which can be equally or even more ecologically important, may exhibit patterns that do not necessarily match those for mammals or other vertebrates (e.g., birds, reptiles) owing to differences in animal dispersal ability, abiotic tolerance and dietary breadth, in addition to community‐level differences (i.e., species composition and diversity) (Burns, 2019).…”