The objectives of this paper are to (1) review existing diversity models, (2) identify principles that explain patterns of plant species diversity, (3) discuss implications for forest management, and (4) identify research needs. Many current theories cast disturbance as the key player in maintaining species diversity by preventing competitive dominance of one or a few species. Equilibrium and nonequilibrium theories alike agree that maximum diversity should occur at intermediate size, frequency, and intensity of disturbance. These models do not adequately predict patterns at all spatial scales or across community types. A mechanistic theory is needed to explain diversity patterns at the patch, stand, and landscape scales, as well as across site quality and successional gradients. Such a theory should be based upon the interaction between species' life‐history characteristics and the nature of disturbance. New research is needed in the following areas: identifying disturbance‐life‐history interactions, particularly with respect to the short‐term and long‐term effects of disturbance; quantifying patch diversity and determining its relationship with species diversity; determining relationships between species diversity and structural and functional diversity; and identifying appropriate standards of comparison for managed stands. Comparative studies in different ecosystem types, such as boreal and tropical forests, should be encouraged to help clarify the relative importance of processes that influence diversity.