The classical community ecology approach assumes that a species is a homogenous, unstructured entity and that all individuals occupy the same distinct trophic position. While this simplification often improves the tractability of food-web models, their utility may be limited because natural populations are often comprised of individuals that differ in size and developmental stage. This size-structure creates size-structured interactions in which the type and strength of interactions among members of the community is determined by the current relative size of the involved individuals. As a consequence, the size structure within a population strongly determines species population dynamics, inter-specific interactions, and the structure and dynamics of communities. The importance of population size structure has been recognized in the past, but recent studies that have specifically considered the role of size-structured interactions have improved our understanding of competition, coexistence among intraguild predators, host-parasitoid or host-disease interactions, and population dynamics of cannibalistic species. This recent empirical and theoretical work represents an innovative step forward because it attempts to reconcile erroneous predictions made by simple models that do not account for size structured interactions with complex natural systems. However, to date there have been no coherent explanations of when it is imperative to include size structure to predict community dynamics and ecosystem processes. What are the specific mechanisms that lead to the observed deviations of dynamics between size structured and unstructured systems? How common are these mechanisms in natural systems and what is their general importance compared to other community processes? This session will focus on new original research addressing these specific questions and will provide a cohesive framework to evaluate the causes and implications of size structure at the population and community-level in a wide range of ecological contexts. For this purpose, the session will incorporate both theoretical and empirical perspectives on this issue and include talks from leading researchers working in diverse systems.