Quantitative Genetics, Selection, and the Adaptive Radiation of Darwin's Finches.

Abstract

Observations of selection in natural populations are a powerful, little used, method for investigating the adaptive significance of characters. In this thesis I measure and interpret sexual and natural selection on morphology in a small population of Darwin's Medium Ground finches (Geospiza fortis) in the Galapagos Isl and s. Natural selection favoring large body size regularly occurs in association with adult mortality. Beak depth and body weight are the targets of selection, and the selection can be related to changing ecological conditions. Selection favoring small body size associated with juvenile mortality appears to be a regular occurrence, but much of this selection has to be inferred because most of the mortality occurs before the juveniles are fully grown. Much of the variance in juvenile and adult mortality can be attributed to variance in body size, known to be a highly heritable trait. Therefore the expression of these two life history traits can be understood in terms of a negative genetic covariance between them, mediated by body size. Two other selection forces have been identified: sexual selection favoring large males, and fertility selection favoring small females: small females have been found to breed earlier than larger females. These two forces can be invoked in the evolution and maintenance of a sexual dimorphism in body size. I use measurements of genetic variation and covariation among characters to identify constraints on evolution, and in particular to gain an idea of the total net selection involved in the differentiation of species. The method is illustrated in the last two chapters. I present patterns of genetic covariation between juvenile and adult characters. The results show the presence of a large maternal effect on chick size at hatching, which rapidly disappears, and high genetic correlations between adult and chick size, from an early chick age. I use the genetic covariance measures to investigate the relationship between ontogeny and phylogeny in the Geospiza. The main conclusion is that ontogenetic differences between species can largely be explained as having arisen as correlated responses to selection confined to adults.