Much of spatial ecology since the late 1960s has been dominated by two theories, the dynamic theory of island biogeography and the classical metapopulation theory. The latter theory largely replaced the former one in the 1980s, especially in conservation applications. It is only recently that ecologists have fully realized that a relatively simple general theory can be readily constructed that makes some of the simplifying assumptions of the two earlier theories unnecessary. The spatially realistic metapopulation theory thereby provides a more unified framework for spatial ecology than the island theory or the classical metapopulation theory. This article describes the application of the spatially realistic metapopulation theory to real metapopulations living in highly fragmented landscapes. I discuss the principal messages for population ecology and conservation biology, and I also place this theory into a broader context of other approaches to spatial ecology.