The debate over the self-regulating capacity of ecosystems, and especially the role of insects, is somewhat reminiscent of the debate on the now-recognized importance of density-dependent feedback regulation of population size, and is a useful example of how science develops. Although this hypothesis is not contingent on natural selection at the supraorganismal level, concepts of group selection have developed from and contributed to this hypothesis. Individual attributes affect other organisms and environmental conditions and generate feedback on individual fitness. Such feedback selection contributes to the inclusive fitness of an individual. The intensity of this feedback is proportional to the relatedness of the interacting individuals. The greatest feedback selection is between the near kin. The frequency of interaction and the intensity of feedback selection decline as interacting individuals become less related. Homeostasis at the supraorganismal levels depends only in part on the selection for attributes that benefit assemblages of organisms. The major challenges for ecologists include defining stability and evaluating the effect of mechanisms such as biodiversity and herbivory that contribute to stability. Some species are known to control ecosystem properties, and their loss or gain can severely affect an ecosystem structure or function. The effects of different species are often complementary, such that diverse assemblages should be better buffered against changes in ecosystem properties in heterogenous environments.