We plan a functional approach to understand the structure and dynamics of the tree communities of the hyper diverse forests of Amazonian Colombia. Large numbers of extremely rare tree species characterize these forests. A functional approach provides insight into the biology of extremely rare species and enables analyses that include all species. We have already mapped, identified and measured the free-standing trees and shrubs larger than 1-cm in diameter at breast height (DBH) in a 25-ha plot in terra firme forest at the National Natural Park of Amacayacu. At Amacayacu, nearly 1.000 species have been identified so far among the 123.268 mapped individuals. More than 400 of these species are expected to be represented by just one or two individuals in 25 ha. A functional approach will allow us to incorporate these rare ecological ciphers into a comprehensive understanding of the Amacayacu forest. We will evaluate the relative roles of interspecific competition and ecological filters using the previously completed census. Interspecific competition is always most severe among species with similar ecological requirements. Thus, the ecologies of near neighbors are predicted to be different if competition structures the Amacayacu tree community. In contrast, ecological filters are imposed by the interaction between environmental gradients and species’ ecological requirements. The ecologies of near neighbors are predicted to be similar if ecological filters or species’ responses to environmental gradients structure the Amacayacu tree community. We will use functional traits to represent species’ ecological requirements and randomization tests to evaluate whether the observed distributions of functional traits of near neighbors are more similar or more different than expected by chance. We will repeat this analysis using phylogenetic similarity calculated under the current phylogenetic hypothesis of the Angiosperm Working Group. Divergent results for analyses of functional and phylogenetic similarity might distinguish direct versus apparent competition. The functional trait data to be collected under this proposal will enable a second generation of studies when the second census is completed in 2013. A fundamental premise of functional ecology is strong coordination between observed traits and unobserved performance (growth rate, survived or died). This premise has rarely been evaluated for perennial plants because the long-term records required to document performance are rarely available. Measurements of growth and mortality rates of 123.268 trees at Amacayacu will enable powerful tests of the relationships hypothesized between plant performance and the consensus functional traits. The second census of the Amacayacu plot will also enable evaluation of the roles of competition and ecological filters for spatially explicit, dynamic data. We will evaluate relationships between the observed performance of each tree and its functional and phylogenetic similarity to its nearest neighbors. Finally, the wood density data to be collected under this proposal will improve estimates of above-ground carbon sequestration in the Amacayacu forest and all of the functional trait data to be collected will be brought to bear on changes in the species composition that are likely to occur between 2008 and 2013. This proposal will provide insight into these global, regional and local processes.