Abstract Detail

Nº613/287 - Andromonoecy as a convergent resolution to intra-locus sexual conflict in bisexual flowers
Format: ORAL
Authors
Kai-Hsiu Chen and John R. Pannell
Affiliations
University of Lausanne, Department of Ecology and Evolution, Lausanne, Switzerland
Abstract
Backgrounds: Andromonoecy, the production of male and bisexual flowers by the same hermaphroditic individual, has evolved numerous times independently across flowering plants and is found in 7% of species. Despite its repeated evolution, its function remains poorly studied. Here, we consider the convergent evolution of andromonoecy as a resolution to sexual antagonism, drawing on data from the alpine plant Pulsatilla alpina that shares common features with many andromonoecious species. Hypothesis: We expected to detect evidence for strong sexual conflict on female allocation but not male allocation traits in bisexual flowers, i.e., female and male functions have different fitness optima only in the female allocation trait, thus favoring the production of unisexual male and bisexual flowers. Methods: We analyzed and compared selection gradients via female and male functions for five floral traits, including sex allocation, phenology, and display traits, in bisexual flowers of P. alpina using a paternity analysis. Results: Our results indicate strong sexual conflict in the number of pistils and flowering date but not in the number of stamens in bisexual flowers. Specifically, we found evidence for disruptive selection on pistil number within flowers, such that flowers with many or no pistils contributed more to fitness than those with an intermediate number of pistils. By abandoning the female function, unisexual male flowers possess phenotypes that approach the male optimum of different floral traits. Conclusion: We show that the unisexual male flowers of andromonoecious species may allow hermaphrodites to resolve sexual conflict because modules with different genders can promote fitness through their specialized sexual functions. We discuss the extent to which modular variation in sex allocation in species with other sexual systems involving unisexuality and bisexuality (e.g., monoecy and gynomonoecy) may be explained similarly.