Biodiversity consequences of predation and host plant hybridization on an aphid-ant mutualism

We examined the hypothesis that mutualists, predators, and host plant quality act in concert to determine the distribution and abundance of a common herbivore. The aphid, Chaitophorus populicola, is found only in association with ants, which provide tending services and protection from predators. As a consequence, aphid abundance declined by 88% on host plants located greater than or equal to6 m from an ant mound. Differences in host plant quality resulted in aphid fecundity being greatest on narrowleaf cottonwoods, 7-22% less on backcross hybrids, 37-46% less on F-1 hybrids, and 57-61% less on Fremont cottonwoods. Due to the combined effects of these factors, we found that the realized aphid habitat was only 21% of their potential habitat. On trees where aphids and tending ants are present, aphids and ants greatly outnumber any other arthropod species; therefore, we examined the hypothesis that the factors affecting aphid density and distribution are also determinants of arthropod community structure. On a per-tree basis, observational data showed that arthropod species richness was 51% greater and abundance was 67% greater on trees where aphid-ant mutualists were absent relative to trees where they were present. When aphids were experimentally removed and ants abandoned the tree, we found the same pattern. On a per-tree basis, arthropod species richness increased by 57%, and abundance increased by 80% where aphid-ant mutualists were removed, relative to control trees. Overall, a total of 90 arthropod species were examined in this study: 56% were found only on trees without aphid-ant mutualists, 12% were found only on trees with aphid-ant mutualists, and 32% were common to both. Specific guilds were also affected; the aphid-ant mutualism had a negative effect on herbivores, generalist predators, and other species of tending ants, and a positive effect on specialist enemies of aphids. These results suggest that, by examining the factors responsible for the population dynamics of a common herbivore, we may also uncover mechanisms that determine arthropod community structure. Furthermore, studies that address the role of mutualism in shaping communities are underrepresented in the literature; these results emphasize the importance of mutualism in ecological communities.