Evolution of Dioecy from Gynodioecy: Artificial Selection for Sex Allocation Patterns

点击翻译按钮获取中文摘要

Project Summary. Knowledge of plant reproductive systems is critical for agriculture and management of endangered species as well as for an understanding of basic evolutionary processes. Plants are usually hermaphroditic, with both male and female function in the same flowers. In many different plant groups, however, evolution has led to sexually dimorphic breeding systems including gynodioecy (females and hermaphrodites in populations) and dioecy (populations with separate male and female plants). This separation has been associated with changes in allocation of resources to male and female function. Few studies examine the genetic potential for shifts in allocation patterns to male and female function, and have taken into account known evolutionary relationships of species with various reproductive systems. The proposed study will examine the genetic potential for sex allocation shifts with artificial selection on allocation to male and female function in two closely related Hawaiian species of Schiedea (Caryophyllaceae). Given the range of allocation patterns within the genus, an artificial selection program for greater allocation to male function in hermaphrodites and to female function in females should demonstrate significant genetic potential for change. In the first experiment, additive genetic variances of sex allocation traits and their genetic covariances with other reproductive characters will be estimated. In the second experiment, artificial selection will be performed for increased biomass of stamens in hermaphrodites and increased carpel and fruit biomass in females. Both direct responses of these traits and indirect responses of correlated traits will be measured, as trait correlations may enhance or impede selection for specialization in male and female function. S. salicaria is predicted to become more male-like and therefore more similar to the sexually dimorphic S. adamantis hermaphrodites. Comparative studies have often been used to predict the course of evolution. The proposed research takes this approach forward, by determining whether evolutionary trends hypothesized on the basis of comparative studies can be produced using artificial selection.