Recruitment Limitation and Population Density in the Harvester Ant, Pogonomyrmex Occidentalis

We evaluated the hypothesis that recruitment limitation is important in determining density in harvester ant populations. Combining field observations and experiments, we conclude that the population density of Pogonomyrmex occidentalis is largely determined by recruitment limitation, and we demonstrate that variation in population density is a function of the dispersal patterns of foundress queens. The density of established colonies, the number of new colonies, and the number of foundress queens are all highly correlated in this population. The density of foundress queens predicts the density of new colonies in the following year. The density of propagules (foundress queens), new colonies, and established colonies is significantly negatively correlated with the distance from the mating site that is the source of the propagules, and the distance from this source can be used to predict the local density of new and established colonies in this population. Finally, we show that the experimental addition of foundress queens significantly increases the number of new colonies in supplemented vs. control plots. At the landscape level, the population is a mixture of areas with low and high density, with no evidence of overall density dependence. Density dependence is detectable only in local areas that receive sufficient propagules and, thus, depends on the proximity to the source of propagules. Consistent dispersal from a specific geographic site results in variation in population density independent of any variation in habitat quality.