This dissertation investigated the evolutionary ecology of a type of obligate mutualism, the form of agriculture found in ants. It presents the results of two laboratory experiments with the fungus-gardening ant species, Trachymyrmex septentrionalis, a species occurring in the Apalachicola National Forest in Florida that is closely related to mostly tropical leaf-cutting ants in the genera Atta and Acromyrmex. Fungus-gardening ants supply substrates to their fungus garden and feed the products of the fungus to their larvae. Because ants make choices in the substrates they collect, an interesting topic is the relationship between the choice of substrates and the quality and amounts of fungus and ants produced. No previous studies have measured the consequences of these choices on the relative sizes of fungus gardens or the quality and quantity of ants produced.
The first experiment indicated a positive relationship between subtrate preference and both ant and fungal performance but it was not perfect. The ants preferred bluejack oak (Quercus incana) catkins and caterpillar frass over oak leaves or flowers (Gaylussacia dumosa). Colonies on the high catkin diet produced more fungal biomass than could be subsequently used, indicating only indirect feedback between forager activity and the colony’s nutritional demands. The biomass of ant produced was similarin all groups, with the exception of being low in colonies that received only flowers. Colonies that had rejected tussock caterpillar frass were later able to produce as much ant biomass on a low catkin diet as those on the high catkin diet.
In the second experiment, the successful replacement of T. septentrionalis’s fungal cultivar, with a foreign fungus obtained in Louisiana from colonies of A. texana, did not change the substrate preference of the ants or the amount or quality of the ants produced. However, T. septentrionalis colonies with A. texana fungus produced more fungal biomass. Under the conditions of these experiments, the relationship between T. septentrionalis and its native fungal cultivar appear to be mutually adaptive, but that mutualism does not seem to be the product of a tight feedback mechanism on an ecological scale.
