Mating Systems in Homogeneous Habitats: The Effects of Female Uncertainty, Knowledge Costs, and Random Settlement

The mating system of bobolinks was examined in a relatively homogeneous field. The population exhibited polygyny apparently resulting from random female dispersion. We present four mechanisms by which females choose breeding situations that may explain this situation. The first model suggests that female bobolinks cannot or do not choose between different breeding situations. We reject this hypothesis because data from other studies of bobolinks indicate a nonrandom nest distribution and apparent preferences of females for different vegetation cover. The second model presented is an extension of the Orians-Verner-Willson polygyny-threshold hypothesis, a perfect-knowledge model. Discrimination errors of females assessing breeding situations and costs of obtaining information on the quality of breeding situations lead to random dispersal within the population over a certain quality interval. The discrimination error is maintained in the population because of errors in predicting realized fitness and the risk that an originally monogamous female will become associated in a polygynous situation. The experience-based-choice and site-fidelity hypotheses suggest that females choose territories on the basis of past experience. Unlike the second hypothesis, they imply that the knowledge of different breeding situations and how to choose between them is not innate, but that territory choices are made using either of two much simpler behaviors, highly correlated with fitness and only indirectly related to differences in breeding situations. A random distribution can arise in the experience-based-choice hypothesis because of only slight differences in realized fitness between monogamous and polygynous strategies and because of the swamping effects of random mortality and discrete fitness products. As in the second hypothesis, a female territory-discrimination error and the cost of obtaining territory quality information promote randomness in homogeneous situations. In the site-fidelity hypothesis, experienced females choose territories without regard to other females or changes in territory quality, resulting in a quasi-random nest distribution. Our results suggest that heterogeneity in breeding-situation quality is not required for polygyny and that conditions more restrictive than habitat homogeneity are required to explain the evolution of monogamy.