- Bouskila, A. 2001: A habitat selection game of interactions between rodents and their predators. — Ann. Zool. Fennici 38: 55–70. I developed a game theoretic model for habitat selection of prey and a generalist predator. In the model, both prey and predator may choose between either a simple or a complex habitat. A second predator is restricted to hunting only in the simple habitat. The model is applied to a system of rodents and their predators: snakes (the generalist predator) and owls. The simplest version of the model predicts that snakes and rodents distribute them-selves among the two microhabitats according to the relative magnitude of risk for ro-dents. Under various conditions (moonlight, competition among rodents and dilution of the risk) the model predicts that snakes distribute themselves among habitats in a way that dampens rodent reactions to variation in owl predation risk and to effects of competition. When rodents are abundant the model predicts that snakes will show a weaker reaction to moonlight. The predictions of the model are qualitatively comparable to field data of microhabitat use of kangaroo rats and sidewinders from the Mojave Desert. The model can also describe a habitat selection game at a larger scale (e.g., movements of snakes into a rich desert oasis). Although a game between predators and prey may not be the cause for all their movements among habitats in the field, a game perspective may contribute expla-nations for what would otherwise be unintuitive habitat shifts.