Multitasking in a plant-ant interaction: how does Acacia myrtifolia manage both ants and pollinators?

Plant associations with protective ants are widespread among angiosperms, but carry the risk that ants will deter pollinators as well as herbivores. Such conflict, and adaptations to ameliorate or prevent the conflict, have been documented in African and neotropical acacias. Ant-acacia associations occur in Australia, but little is known of their ecology. Moreover, recent phylogenetic evidence indicates that Australian acacias are only distantly related to African and American acacias, providing an intercontinental natural experiment in the management of ant-pollinator conflict. We examined four populations of Acacia myrtifolia over a 400-km environmental gradient in southeastern Australia using ant and pollinator exclusion as well as direct observation of ants and pollinators to assess the potential for ant-pollinator conflict to affect seed set. Native bees were the only group of floral visitors whose visitation rates were a significant predictor of fruiting success, although beetles and wasps may play an important role as "insurance" pollinators. We found no increase in pollinator visitation or fruiting success following ant exclusion, even with large sample sizes and effective exclusion. Because ants are facultative visitors to A. myrtifolia plants, their presence may be insufficient to interfere greatly with floral visitors. It is also likely that the morphological location of extrafloral nectaries tends to draw ants away from reproductive parts, although we commonly observed ants on inflorescences, so the spatial separation is not strict. A. myrtifolia appears to maintain a generalized mutualism over a wide geographic range without the need for elaborate adaptations to resolve ant-pollinator conflict.

Insect community structure covaries with host plant chemistry but is not affected by prior herbivory.

By feeding on plant tissue, insect herbivores can change several characteristics of their hosts. These changes have the potential to alter the quality of the plant for other herbivore species, potentially altering the structure of the community of species attacking the plant at a later point in time. We tested whether herbivory early in the season changes host plant performance, polyphenol chemistry, and the community structure of sessile herbivores later in the season. We experimentally manipulated densities of early-season moth caterpillars on a set of young oak trees and measured tree growth, reproduction, leaf chemistry, and the abundance and community composition of leafmining and galling species later in the season. The experimental manipulations of early-season herbivores did not affect late-season leaf chemistry or tree performance. Early-season herbivores had a weak negative effect on the abundance of gallers and a positive, tree-dependent effect on the overall diversity of late-season sessile herbivores. The chemical composition of leaves covaried with the species composition of the late-season leafmining and galling community. Both the chemical composition of the host tree and the late-season insect community structure were strongly affected by the growth location of the tree. Our results suggest that plant-mediated indirect effects between herbivores might play a limited role in this system, whereas the underlying variation in plant chemistry is an important factor structuring the associated insect community. Our results emphasize that factors other than prior herbivory can be important determinants of plant chemistry and the community composition of herbivores.