Population dynamics of a gynodioecious species.

Plants display remarkable sexual diversity. Regardless of this variation, most flowering plants are hermaphrodites and contain flowers with both male and female sex organs. However, various sexual systems have evolved in which male and female organs have been separated between flowers and between individuals.

In gynodioecious species, hermaphrodites co-exist with females, plants that lack pollen production (male-sterile individuals). Since females have lost their ability to produce pollen, they are dependent on hermaphrodites for fertilization. One necessary condition for the maintenance of this sexual polymorphism is the occurrence of a reproductive ‘female advantage’. The female advantage implies that females should produce more seeds or higher quality seedlings in comparison to hermaphrodites and the magnitude of the female advantage is dependent on the sex determination system. In the most common form of gynodioecy, nuclear-cytoplasmic gynodioecy, mutations that lead to male-sterility are located in the cytoplasm and the effect of these male-sterility mutations can be counteracted by nuclear alleles (male- fertility restorers). In species with nuclear-cytoplasmic gynodioecy, females require only a slight reproductive advantage over hermaphrodites to be maintained within a population (female advantage > 1). The female advantage is generally attributed to resource reallocation from pollen to seed production and avoidance of inbreeding depression since females are obligate outcrossers.

In most gynodioecious species that have been studied over the years, females had a clear reproductive advantage over hermaphrodites, although the magnitude of the female advantage generally differed among populations. Variation in the magnitude of the female advantage is not surprising since seed production of females is strongly dependent on pollen availability. In populations or patches with a low number of hermaphrodites, females may experience low seed set due to pollen limitation, while in populations or patches with a high number of hermaphrodites, females may have a reproductive advantage. When females produce more seeds than hermaphrodites the frequency of females may locally increase, which in turn may contribute to variation in sex ratios among populations. Population female frequencies are highly variable and generally range from a low percentage to more than 50% females. Many gynodioecious species also have populations that consist purely of hermaphrodites. When the frequency of females within a population is low, floral sex ratios can still be highly variable, since sex expression can also vary within individuals. Hermaphrodites of many gynodioecious species are dichogamous, indicating that they separate the maturation of male and female organs in time. Differences in timing of male and female function of a flower, typically lead to a shift in floral sex ratio during a population’s flowering season.

To gain insights into the various aspects that may affect seed production and genetic variation in gynodioecious plant species, I quantified seed production and offspring fitness across a wide range of populations that differed in population size and sex ratio to understand how these variables affected the female advantage and sex ratio evolution. I assessed genetic diversity in a number of spatially separated populations to understand how genetic variation is partitioned among populations and between sex morphs. Moreover, I studied temporal variation in female reproductive success within and among individuals across an entire flowering season to get a better understanding of the impact of within-population sex ratio variation on seed set. To address these research objectives, two gynodioecious species, Saxifraga granulata (meadow saxifrage) and Plantago coronopus (buck’s- horn plantain), were studied.

In this thesis, I show that variation in sex expression within inflorescences, among individuals and among populations highly affected seed production of my study species. In populations of P . coronopus, seed production was highly dependent on population sex ratio, in populations with a low frequency of females both females and hermaphrodites produced more seeds than in populations with a high female frequency. Females produced fewer seeds than hermaphrodites on average, however, offspring fitness of females was higher than offspring fitness of hermaphrodites. Furthermore, genetic analyses showed that both females and hermaphrodites were highly genetically diverse, but adult hermaphrodites showed increased levels of homozygosity as well as severe inbreeding depression after selfing. In S. granulata, the occurrence of female individuals was low, but flowers of hermaphrodites differentially invested in male and female function; the first-opening central flower generally had a long female phase and produced significantly more seeds than early and late lateral flowers. Furthermore, early flowering plants produced more seeds than late flowering plants. Populations of S. granulata were highly genetically diverse, which was most likely related to its high ploidy level (octoploid) and the ability to reproduce clonally.

Despite the differences between Saxifraga granulata and Plantago coronopus in terms of reproduction, floral biology, pollination mode, and ploidy level, I have shown that the large variation in sex expression within inflorescences, among individuals and among populations highly affected seed production. My results further showed that plant mating system and the extent of inbreeding depression affect the distribution of genetic variation among populations and sex morphs. Although not explicitly investigated, they also point to the importance of local spatial structure of gynodioecious populations for regulating the evolutionary maintenance of gynodioecy.