If habitat selection is adaptive, habitat choice by individuals should be directly associated with reproductive success. With that expectation in mind, we examined consequences of habitat choice in Sage Sparrows (Amphispiza belli) in southern California. Habitat selection in Sage Sparrows was pronounced, with vegetation measurements within territories differing significantly from those of interspersed unoccupied areas. Likewise, sites where sparrows placed their nests were significantly different from random sites within territories. Both successful territories (those containing a nest that fledged at least one young) and successful nest sites differed significantly in habitat from those that were unsuccessful. However, there was an inverse relationship between choice and its consequences; the mean score of unsuccessful territories on a discriminant function describing the habitat gradient from unoccupied to occupied areas (i.e., choice) was significantly higher than the mean for successful ones. In other words, birds preferred to settle in areas in which they did worse. Thus, attractiveness of habitat has become inversely related to its suitability, and the study site now serves as an ''ecological trap.'' Because predation accounted for 80-90% of nest loss, we speculate that decoupling of habitat attractiveness from suitability has been produced by redistribution of predators (mainly snakes) due to anthropogenic landscape-level changes in the region.

If habitat selection is adaptive, habitat choice by individuals should be directly associated with reproductive success. With that expectation in mind, we examined consequences of habitat choice in Sage Sparrows (Amphispiza belli) in southern California. Habitat selection in Sage Sparrows was pronounced, with vegetation measurements within territories differing significantly from those of interspersed unoccupied areas. Likewise, sites where sparrows placed their nests were significantly different from random sites within territories. Both successful territories (those containing a nest that fledged at least one young) and successful nest sites differed significantly in habitat from those that were unsuccessful. However, there was an inverse relationship between choice and its consequences; the mean score of unsuccessful territories on a discriminant function describing the habitat gradient from unoccupied to occupied areas (i.e., choice) was significantly higher than the mean for successful ones. In other words, birds preferred to settle in areas in which they did worse. Thus, attractiveness of habitat has become inversely related to its suitability, and the study site now serves as an “ecological trap.” Because predation accounted for 80–90% of nest loss, we speculate that decoupling of habitat attractiveness from suitability has been produced by redistribution of predators (mainly snakes) due to anthropogenic landscape‐level changes in the region.