The Evolution of Finch Communities on Islands and Continents: Kenya vs. Galapagos

I describe the nonbreeding finch community of several habitats in continental Kenya, East Africa, and compare it with previously described communities of granivorous finches on Galapagos islands. The purpose of the comparison is to explore differences in structure between communities that have evolved on a continent and on an isolated archipelago, and to suggest reasons for the differences. The ultimate goal is to infer the factors important in the evolution of finch communities, of which the majority have a continental origin. Four factors that might be expected to influence finch communities differently in Kenya than Galapagos are: a greater diversity of possible competitors, an abundance of predators, movement between localities, and a greater diversity of resources. Differences predicted on the basis of these four factors are compared with actual differences between Kenya and Galapagos, in order to estimate their relative importance. The communities of Kenya and Galapagos were surprisingly different. Despite the greater age of the Kenya fauna, the diversity of beak sizes was less, the range of seed sizes exploited was less, and the overall utilization of seed resources by finches was incomplete. Species in Kenya were more differentiated by habitat, microhabitat, and seed species, and less differentiated by seed size than finches in Galapagos. Diets were on average more taxonomically specialized, and associations between diet and beak and body dimensions were correspondingly weaker. Kenya finches were sensitive to the proximity of trees and shrubs (cover) while foraging. These results, together with observations from a limited number of other studies, suggest that a dichotomy may exist between the finch communities of continents and isolated archipelagos. However, like Galapagos, local finch density in Kenya was related to food abundance, and diet overlaps between species coexisting in the same locality were low. Comparison of results with predictions from hypotheses based on the four factors suggests that competition among the large diversity of granivorous species in Kenya (i.e., finches and other granivorous birds, rodents, and ants) is a major reason for the differences between the finch communities of Kenya and Galapagos. The narrow range of beak and seed sizes used by finch species in Kenya is consistent with competition from other taxa, and competition between the many finch species constrained to a narrow range of seed sizes may explain the more pronounced habitat, microhabitat, and diet specializations. Thus, while the finch communities of continents and isolated archipelagos may differ in structure, there is an indication that in large part they represent alternative outcomes of the same process. Predation risk is the most likely cause of a preference for feeding near cover in Kenya, and it may explain why certain seed resources are not exploited by finches, such as ground seeds in areas of poor visibility (dense grasses). Most species appeared to be similarly sensitive to risk of predation; for example, finch species showed little partitioning along an axis of distance to cover. Thus, the main effect of predation on community evolution in Kenya may have been to further confine the resource spectrum available to finches, a role similar to that of competition from other taxa. There was no evidence that the effects of competition within localities were swamped by an influx of immigrant individuals and species. However, lesser indirect effects of movement may be present; movement may elevate local species diversity and thereby influence community structure by increasing the diversity of potential competitors and predators. The prediction that a greater diversity of resources in Kenya would lead to a greater diversity of morphological forms was not upheld, suggesting that resource differences are less influential than other factors.