ABSTRACT
The landscape is essentially heterogeneous and dynamic, as we have seen in the introductory part of this work. Whether under the influence of human activities or natural disturbances, it is a mosaic of habitats, and a large number of species use several elements of this mosaic during their life cycle. The complexity and heterogeneity of structures and their dynamics are often recognized but they have been simplified by empirical ecologists or theoreticians. Until the 1970s, researchers restricted their domain of investigation to zones that were homogeneous or considered to be so (Lefeuvre and Bamaud, 1988). Since then, the development of patch theory (Pickett and White, 1985; Chapter 4 of this book) has offered theoreticians a simplified view of heterogeneity defining space as a set of patches arranged in an ecologically neutral matrix (Wiens, 1995). Such simplification of spatial heterogeneity has led to many developments in the ecology of fragmented populations, based mostly on models of metapopulations (Gilpin and Hanski, 1991; Hanski and Gilpin, 1997). In this context, the specialist species, that is, those using only one type of landscape element, have been abundantly studied because they are sensitive to fragmentation (Farina, 1998).
