Extrafloral nectaries have a limited effect on the structure of arboreal ant communities in a Neotropical savanna.

How environmental contexts shape the strength of species interactions, and their influence on community structure, remains a key focus for the field of community ecology. In particular, the extent to which local competitive interactions impact community structure, and whether this differs across contexts, persists as a general issue that is unresolved across a broad range of animal systems. Studies of arboreal ants have shown that competitive interactions over carbon-rich exudates from extrafloral nectaries (EFNs) and homopteran aggregations can have positive and negative effects on the local abundances of individual species. Nevertheless, it is still unclear the extent to which these local effects scale to community-level effects. Here we address the role of food from extrafloral nectaries on the structure of arboreal ant communities in a savanna of central Brazil. We did this with a combination of a diversity survey across tree species with and without EFNs, a repeated survey at times of peak EFN activity, and testing of our survey findings with two experimental manipulations of nectar availability that also provided supplementary nesting cavities. Species richness, but not composition, differed significantly between trees with and without EFNs. However, trees with EFNs had, on average, only 9% more species than those without EFNs. Furthermore, ant species richness did not differ significantly between periods of high and low EFN activity. Although nectar supplementation significantly affected nest occupation rates, this difference was seen solely in. the experiment with a massive supply of nectar and there was no effect on total ant richness or identity of the focal assemblages. Our findings suggest that the effects of extrafloral nectar on the abundances of arboreal ants at local scales do not scale to a strong structuring force at the community level. We suggest that this is most likely due to a lack of specificity of community members for EFN tree species, and the diffuse temporal and spatial nature of the availability of active EFNs. These properties mean that observable short-lived activity and competition over particular EFNs does not ultimately drive lasting changes in the associated assemblage of species, and therefore, the community as a whole.