Experimental analysis of territory size in a population of the fire ant Solenopsis invicta

Alternative models of territoriality are based on contrasting assumptions about the behavioral processes determining territory size. In a series of controlled field experiments on the fire ant Solenopsis invicta, I tested whether territory size is affected by the availability of food, as predicted by most economic models, and whether territory size is affected by fighting ability, as predicted by models of competition among neighbors. Abundant food was offered for 30-35 days to selected colonies either immediately next to the nest (experiment 1) or at peripheral sites near the territory boundary (experiment 2). These food supplements had no detectable effect on territory size. Furthermore, food placed near the periphery of the territory did not significantly alter local boundary positions. During both experiments, large colonies lost more territory than did small colonies, reflecting temporary declines in worker number due to the seasonal production of reproductives. Such losses by large colonies during the summer months create opportunities for newly founded colonies to expand territories. In a third experiment, colonies from which workers were removed lost significantly more territory than did unmanipulated controls. These results show that territory areas in S. invicta are strongly affected by the relative fighting ability of neighboring colonies but provide no evidence that colonies adjust territory area in response to short-term changes in the availability of food.