Sexual conflict and selection on pistil and pollen traits

The incidence of sexual selection in plants is today acknowledged, however, just as in animals, evolution and maintenance of mate choice is gravely underexplored. Moreover, the potential for sexual conflicts to occur in plants has only been assessed empirically to a very limited degree. Using the hermaphroditic annual herb Collinsia heterophylla, I looked deeper into the role of pollen-pollen competition and female mate choice by assessing effects of indirect benefits and sexual antagonism on measures of male, female and progeny fitness. A distinctive feature is that I in addition to effects of parental identities also evaluated effects of multiple pistil and pollen traits. The main theme of this thesis concerns the delayed stigma receptivity previously documented in flowers of this species, a female trait assumed to enhance pollen competition, allowing a “fair start” of all pollen in the race towards the ovules, consequently allowing for sorting in favour of pollen grains or donors of superior quality. Within this context, a special focus has been devoted to the possibility of a sexual conflict over timing of stigma receptivity; certain plants produce pollen that are able to fertilize ovules when applied at early stages of floral development, at a maternal cost of reduced seed set. The results point to that the sexual conflict over timing of stigma receptivity is general, as in four different populations, some donors were able to fertilize ovules earlier than others at a cost of maternal reduced seed set. Because this ability of donors depended on donor-recipient population combination, which in turn co-varied with the magnitude of cost inflicted, our results suggest that sexually antagonistic co-evolution may occur within populations of C. heterophylla. Moreover, the maternal cost of reduced seed set was shown to be dose-dependent, such that a high pollen load in the first out of two pollinations resulted in a lower seed set compared to a low pollen load, following pollinations at early floral development. The pistil trait, delayed stigma receptivity, was related to higher paternal diversity within a capsule, which in turn was associated to a higher maternal seed set, demonstrating a positive effect of delaying stigma receptivity. Importantly, maternal parents with a late onset of stigma receptivity sired superior-quality offspring, directly demonstrating an indirect benefit of mate choice. As the ability to delay stigma receptivity was found to show heritable genetic variation, the results suggest that this trait has evolutionary potential. It is further likely that mate choice in C. heterophylla partly may be directly selected: given that maternal parents favoured late-fertilizing pollen donors, and the importance of delaying stigma receptivity for gaining indirect benefits, it should be advantageous to ameliorate costs of early fertilization. However, because delayed stigma receptivity did not influence siring success of pollen donors, it is probable that direct benefits of mate choice are accounted for by other preference traits than delayed stigma receptivity. In sum, this thesis highlights the multitude of selective forces that together decide male-female co-evolution, and points at the potential for sexual conflicts in plants. Indeed, taking sexual conflicts into account in plants could be ground breaking for our understanding of plant evolution, as it opens up an entirely new approach to studying important questions, such as how differentiation and speciation occurs and the related question of how genetic variability is maintained within populations