Understanding the forces that structure communities and dictate changes in community dynamics is one of the central goals in the field of ecology. Ecosystems are influenced by natural variation in abiotic and biotic conditions, as well as external forces such as human stressors. These factors can result in changes in communities across both short and long-time intervals. In addition, the strength of these variables can vary across a landscape, such that each location is shaped by a unique set of conditions. Furthermore, many species adapt to their surrounding environmental conditions which can further structure ecological patterns. Selection can act over relatively fine spatial scales (local adaptation) and over short temporal scales (rapid evolution). However, due to the open nature of marine ecosystems, marine ecologists historically neglected the importance of evolutionary processes in structuring communities.In this dissertation, I address three questions: (1) How have rocky shore ecosystems in northern California changed over long-time intervals spanning multiple decades? (2) What forces generate and maintain adaptive variation among marine populations? (3) What are the ecological consequences of this genetic and phenotypic variation; specifically, how does spatial and temporal variation in selection affect population dynamics, species interactions, and community processes? In my first chapter, I used a historical approach to understand long-term changes in a rocky shore community at Dillon Beach in northern California. Using an unpublished report from 1941, I resurveyed a mussel bed community to analyze community changes over time. In contrast to previous research that had shown massive declines in the diversity of mussel bed communities in southern California, I found little change in species richness and diversity over the ~80-year interval. However, I documented changes in the abundances of species related to their biogeographic affinities with increases in warm-adapted southern species and declines in cool-adapted northern species. Predator-prey interaction often evolve over broad geographic ranges within a context of substantial abiotic and biotic variation. However, few empirical studies have tested whether environmentally-driven variation in prey traits can shape selection on predator phenotypes across a landscape. In my second chapter, I analyzed the importance of spatial mosaics of selection in structuring the evolution of the drilling capacity of the Channeled Dogwhelk, Nucella canaliculata. I showed that dogwhelks from California populations were able to drill much larger and thicker mussels (Mytilus californianus), than those from Oregon populations. To quantify the spatial mosaic of mussel shell thickness along the coast, I analyzed M. californianus shells from the same study sites during three time periods (2000-2001, 2008-2009 and 2019). Mussel shells from Oregon were consistently thicker than those from California in the first two sampling periods. However, mussel shells appear to be thinning, particularly on the central Oregon coast. These changes may suggest that the selective landscape that shaped the evolution of this dogwhelk-mussel interaction may be shifting. In my third chapter, I studied the effects of short-term temporal variation in selection on a predator-prey interaction, as well as the cascading effects on the surrounding rocky shore community. By utilizing a population of predators known to vary in drilling traits (N. canaliculata on Bodega Head), I selected for predator phenotypes by rearing newly-hatched dogwhelks on four diet treatments encompassing different prey types and shell thicknesses. Dogwhelks that survived these treatments were tested in the laboratory for their ability to drill thick-shelled mussels. To assess the community consequences of selection in the field, I outplanted the lab-reared dogwhelks to field cages and quantified their effects on mussel bed succession over the course of one year. Despite the laboratory results indicating the potential for rapid adaptation and divergence in predatory traits, this eco-evolutionary feedback was not strong enough to result in clear community effects in the field. Although eco-evolutionary feedbacks have received considerable recent attention from ecologists, my findings suggest that a variety of processes may dampen the potential for strong eco-evolutionary dynamics in many natural communities. Collectively, my dissertation research advances our understanding of what forces drive rocky shore community dynamics across large spatial scales and across both short and long temporal scales. Over decadal timescales, I have shown that while a mussel bed community in northern California has had little change in species richness and diversity, there have been changes in the relative abundances of southern vs. northern species consistent with warming temperatures, and a thinning of mussel shells consistent with ocean acidification. Across ~1,000km of coastline, I have highlighted the importance of spatial mosaics of prey traits in shaping predator phenotypes. And lastly, over short time intervals, my results indicate that a marine predator can rapidly adapt to changes in prey resulting in divergent phenotypes; however, these feedbacks were not strong enough to lead to clear community effects in the field. Overall, my results emphasize the importance of studying species interactions and community dynamics within a broader ecological and evolutionary context of spatial and temporal variation and change. This perspective is likely to be increasingly necessary and insightful in an era where natural communities are faced with an accelerating pace of global change.