This dissertation explores plant adaptations to flooding and drought stress. In the first study, I assessed flooding tolerance as a function of life history stage in Itea virginica, a species of shrub from cypress tupelo forests of the United States. Results from this study indicate that limited flooding tolerance of juveniles restricts the distribution patterns of adults (Chapter 1). In a complementary study, I assessed adaptive evolution in Elliott's blueberry (Vaccinium elliottii). In heterogeneous landscapes, natural selection can result in the evolution of locally adapted ecotypes. However, if habitats differ in size or quality, demographic source-sink dynamics can shape the evolutionary trajectory of species. I conducted a multiyear reciprocal transplant experiment to test whether V. elliottii is locally adapted to upland and floodplain forests in South Carolina. These contrasting habitats vary tremendously in water table depth, light levels and edaphic conditions. In the greenhouse, I exposed individuals to drought and flooding to assess selection on traits in response to disparate abiotic stresses. Finally, I quantified population differentiation and gene flow via microsatellite markers. V.elliottii families exhibited significantly higher fitness in upland relative to floodplain forests, regardless of the habitat of origin. Similar results from the greenhouse show that V. elliottii is better adapted to drought than flooding. The population density of this species is higher in upland than floodplain forests and upland populations harbor significantly greater genetic diversity. This disparity in population size produces asymmetrical gene flow from upland to floodplain populations. These patterns are consistent with genetic source-sink dynamics, in which adaptation to a marginal habitat is constrained by immigration from a benign habitat (Chapter 2). V. elliottii exhibits significant phenotypic plasticity in foliar, ecophysiological and root-based traits. Theoretical models predict that under source-sink dynamics, species evolve traits that maximize fitness in the source habitat at the expense of fitness in the sink habitat. The phenotypic plasticity expressed by this species contradicts this expectation; in Chapter 3, I discuss three hypotheses that could resolve this paradox: phenotypic plasticity could be a phylogenetic legacy, this species could be undergoing niche expansion, or plasticity could be adaptive within upland forests.