Ants are ubiquitous in tropical and temperate ecosystems, making up as much as 10% of the total animal biomass (Holldobler and Wilson, 1991). Not surprisingly, ants greatly influence community and ecosystem processes; ants can profoundly affect habitat characteristics (Folgarait et al. 2002), the availability of resources for other species (Dostal et al. 2005), soil structure (Dostál et al. 2005) and their activities can be viewed as those of ecosystem engineers (Folgarait et al. 2002, Jones et al. 1994). A key community process in which ants participate is seed dispersal. Using ants as model systems, I was interested in assessing the roles ants play as seed dispersers and determining the consequences of losing a keystone ant mutualist for plant communities and ecosystem processes.

Within this context, I examined how habitat alteration in a tropical dry forest in Costa Rica affected ant communities and altered ant-mediated seed dispersal processes. I found that habitat disturbance altered ant community composition and the relative abundance of key seed-dispersing ant species. In all sites combined, ten species of ant dispersed seeds, with two species, Pheidole fallax and Ectatomma ruidum, in combination being responsible for 92% of all observed seed removals. Differences between these two species in nestmate recruitment behaviour lead to differences in seed dispersal distances, with the recruiting species, P. fallax, achieving much greater dispersal distances than the non-recruiting E. ruidum.

Next, I examined the variation in seed dispersal by ants along an extensive elevational gradient in Great Smoky Mountains National Park, USA, where I asked whether variation in ant community composition affected seed removals and seed dispersal distance of Trillium undulatum and Hexastylis arifolia, two common understory herbs found throughout the southern Appalachian Mountains. I found that measures of ant community and climate variables strongly covaried with elevation, while Trillium species richness and abundance did not. The most important variables predicting seed removals were average annual temperature and the abundance of Aphaenogaster rudis ants, both of which varied along the elevational gradient. Contrary to my expectations, seed dispersal by ants did not depend on ant community composition, but was dominated by one species, A. rudis, which occurred at every site and removed the vast majority of all observed seeds in this study. Though the ant fauna in the Great Smoky Mountains National Park is diverse, dispersal by ants of T. undulatum and H. arifolia, and likely other myrmecochores, is driven by A. rudis, a keystone mutualist in this system.

In a complimentary study, I was interested in experimentally testing the effects of A. rudis ants on plant population dynamics and ecosystem processes. Few studies to date have examined how a single ant species can affect both above and belowground processes over several years. In order to address this ecological question, I experimentally removed A. rudis colonies from deciduous forest plots to assess the population consequences for Hexastlylis arifolia, an understory herb dispersed by these ants, for soil nutrient dynamics, and for soil mesofauna. I found that the removal of A. rudis ants dramatically altered the spatial distribution of H. arifolia seedlings and resulted in a large increase in ammonium concentrations in ant removal plots in the first year of the study. This increase in ammonium concentrations in ant removal plots did not affect H. arifolia seed germination and seedling mortality rates. The effects of ant removal on the soil mesofauna were mixed, with no measurable effects on Collembola abundance and a large increase in oribatid mite abundance in the second year of the study. Taken together, these studies suggest that ant species play profound roles in community and ecosystem processes. From the perspective of plants, dispersal by different ant species can lead to different population trajectories, as behavioral and ecological characteristics shape the initial spatial distribution of plants within populations and define the context for future ecological, demographic, and genetic interactions among emerging seedlings. Furthermore, ants can directly influence ecosystem processes through their direct and indirect effects on belowground mesofauna and soil nutrient processes.