Differences between the sexes are pervasive in both animals and plants. Such dimorphism can result from selection directly on a trait or it can be a result of genetic correlations between traits. The investigators will test the hypothesis that multiple characters exhibit sex differences in the dioecious plant Silene latifolia, because of genetic correlations between traits. This hypothesis will be investigated with two quantitative-genetic experiments. The first incorporates a set of controlled crosses and subsequent measures of progeny, to determine the pattern and magnitude of genetic covariances among traits. The second experiment will involve artificial selection to investigate how labile the genetic correlations are between flower traits and physiological traits. They will select on the combination of traits rather than on individual traits to alter the correlation between traits, the most powerful way of breaking correlations. These experiments are centered on measuring and altering genetic correlations. These correlations can occur because of pleiotropy, in which one gene affects more than one trait, or because of linkage, in which the genes affecting traits are inherited together. The first of these, pleiotropy, leads to long-term correlations between traits, whereas linkage can be easily broken by selection. If the investigators are unable to break the correlation between traits, then this will lend credence to the hypothesis that pleiotropy contributes to patterns of genetic correlation and differences between the sexes.
