Understanding how changes to the quality of habitat patches affect the distribution of species across the whole landscape is critical in our human-dominated world and changing climate. Although patterns of species' abundances across a landscape are clearly influenced by dispersal among habitats and local species interactions, little is known about how the identity and origin of dispersers affect these patterns. Because traits of individuals are altered by experiences in their natal habitat, differences in the natal habitat of dispersers can carry over when individuals disperse to new habitats and alter their fitness and interactions with other species. We manipulated the presence or absence of such carried-over natal habitat effects for up to eight generations to examine their influence on two interacting species across multiple dispersal rates and different habitat compositions. We found that experimentally accounting for the natal habitat of dispersers significantly influenced competitive outcomes at all spatial scales and increased total community biomass within a landscape. However, the direction and magnitude of the impact of natal habitat effects was dependent upon landscape type and dispersal rate. Interestingly, effects of natal habitats increased the difference between species performance across the landscape, suggesting that natal habitat effects could alter competitive interactions to promote spatial coexistence. Given that heterogeneity in habitat quality is ubiquitous in nature, natal habitat effects are likely important drivers of spatial community structure and could promote variation in species performance, which may help facilitate spatial coexistence. The results have important implications for conservation and invasive species management.carryover effects | natal habitat effect | competition | metacommunity | dispersal C ommunities do not exist in a vacuum; instead, they are connected to each other through dispersal of interacting species. Consequently, dispersal among different kinds of habitat patches is increasingly recognized as a key factor driving the dynamics and structure of communities from local to regional spatial scales (1-5). In classic models, the persistence and dynamics of populations within a patch are determined by two factors: the rate of dispersal between patches of habitat and the species fitness within each local patch (6-8). This view has persisted into metacommunity theory (multiple communities connected by dispersal of individuals between the patches of habitat they occur in) as well, where the influence of dispersers on community dynamics is generally considered only in terms of their numbers (i.e., dispersal rate) and habitat-specific performance (1, 9-11). Implicit in this case, and for most of spatial ecology, is that the interactions and population dynamics within a habitat are solely determined by the quality of that habitat. However, this approach ignores the often substantial variation in individual traits and fitness within and across habitats in a natural sys...