Population and community ecology need a large-scale perspective because local patterns (of biodiversity) and processes (trophic interactions) are influenced by the regional setting. The ratio of the foraging range and/or dispersal ability to the distance between landscape elements influences local population dynamics. The spatial scale experienced by a species may be linked to its trophic level and also to traits such as body size, resource specialization, rarity, and population size variability. Hence, communities are assemblages of species with different spatial strategies. Effects of habitat loss and habitat fragmentation on plant-herbivore, herbivore-enemy, as well as plant-pollinator interactions are contingent on species and landscape. Metapopulation theory provides a unifying frame to approach plant-insect systems across fragmented landscape, although the landscape context is often ignored. In some cases theory is far ahead of empirical research. We call for more population data on large spatial and temporal scales to better understand plant-insect populations across fragmented landscapes.