Rausher 9800876 Mutualisms are ecological interactions between species in which both species benefit from the interaction. This project will evaluate the assumptions and predictions of the cost-benefit framework for the extrafloral nectar-mediated mutualism between ants and Chamaecrista fasciculata, the partridge pea. A field experiment will evaluate four assumptions: 1) the mutualistic trait (extrafloral nectar) is costly to provide, 2) the extrafloral nectar provides a benefit through reduction of herbivore damage by ants attracted to the nectar, 3) the balance of costs and benefits produces stabilizing selection favoring an intermediate optimum, and 4) manipulation of costs and benefits should change the value of the selective optimum. In addition, this project will assess whether the cost-benefit framework can explain geographic variation in extrafloral nectar traits in C. fasciculata. Geographic variation in the trait will be surveyed, and the hypothesis that variation between populations is consistent with the presence of different selective optima will be tested using a reciprocal transplant experiment. Mutualisms are ubiquitous, but theoretical and mechanistic understanding of the processes that underlie them has progressed slowly. Of the models that have been used to explain the evolution of mutualisms, the cost-benefit approach is the most general and the most appropriate framework for understanding the evolution of adaptive traits involved in mutualisms. This project will evaluate the assumptions and predictions of the cost-benefit approach in regard to a mutualism between ants and a common annual plant of the eastern United States, Chamaecrista fasciculata (the partridge pea). Data from this project should provide an important empirical basis for future discussion of the pattern of selection on mutualistic traits.