Understanding how biodiversity is maintained is a central issue in ecology. One long-standing theory is that species can coexist if they differ in their requirements for limiting resources. For plant species, the availability of soil resources such as water, nitrogen, and phosphorus may be critical in determining which species persist in a given area. This research will examine how a diverse group of palm species can coexist in tropical mountain forest in Panama. A series of seedling transplant experiments will be performed to test the hypothesis that differences in plant performance across a gradient of soil fertility results from differences in resource allocation patterns among species. Specifically, it is predicted that species found on the most nutrient-poor sites will allocate relatively more of their resources to root biomass and to anti-herbivore defenses. These allocation patterns should reflect the increased cost of extracting nutrients from the soil, and increased cost of herbivore damage when nutrients are in short supply. This study will be among the first experimental tests of a prominent hypothesis to explain how tropical tree species coexist. Furthermore, the proposed research will add new information on the relative importance of herbivore pressure and belowground resources in determining the distribution of forest understory palms, a group of species that have been heavily exploited for horticultural use.