Pollinator-mediated competition and sources of pollen limitation in Limnanthes douglasii rosea and Limnanthes alba

Pollinator visits are essential for reproduction in many plants, yet interspecific movements of pollinators can also lead to competitive interactions between co-flowering species. Pollination-mediated reductions in fertility could potentially lead to exclusion of competing plant species, and may generate spatial variation in the associations among co-flowering species across a landscape. This would be especially true if a new plant species immigrating into a community experiences a rarity disadvantage in competition for the services of pollinators it shares with resident species. Resulting decreases in seed production could reduce the chances for establishment in the new site. Pollinator-mediated mechanisms can function through direct interference competition or through forms of pollen limitation. Species that experience severe pollen limitation in seeds set may be those that are least likely to establish. Understanding how plant and pollinator community factors can potentially interact to influence the level of pollen limitation in sites may eventually help us predict which factors will be most important for establishment and maintenance of species.;In chapter one, I document the potential for heterospecific pollen transfer to cause competitive interactions between two annual species with locally disjunct distributions. Pollinators move readily between Limnanthes douglasii rosea and L. alba when they are presented together, but greenhouse deposition of heterospecific pollen decreased fertility in both species. The decrease in seeds produced was much more pronounced in L. d. rosea (91%) than in L. alba (41%) so subsequent experiments focused on L. d. rosea. In field arrays, L. d. rosea plants receiving pollen from heterospecific neighbors first showed >60% reduction in per-flower fertility. Under natural pollination conditions, heterospecific pollen transfer has the ability to decrease the fertility of L. d. rosea and accordingly, may contribute to the fine-scale exclusion of these two species.;In chapter two, we test for frequency dependence in pollinator-mediated competition between two vernal pool annuals native to California, Limnanthes alba and Limnanthes douglasii rosea. These species exhibit a fine-scale spatial pattern of exclusion, and have been shown to compete asymmetrically through heterospecific pollen transfer, with L.d.rosea being more dramatically affected. We tested for frequency-dependence in pollinator-mediated competition using pure- and mixed-species experimental arrays at varying densities and frequencies in populations of both species. Above-ground competition at high arrays densities was indicated by reduced flowers per plant in both species. Plants of both species that were rare within an array showed a marked decrease in seeds produced per flower, evidently caused primarily by pollinator-mediated competition. The impact of competition was asymmetrical, with the per-flower and whole-plant fertility of L.d.rosea being more dramatically reduced when present at low frequency within arrays. Rarity disadvantage occurred within arrays at both the L. d. rosea and the L. alba home sites.;In chapter three, I test for pollen limitation of Limnanthes douglasii rosea in its home-site and if the level of pollen limitation is related to the phenology of co-flowering plant and pollinator species. I did not find any significant correlates with pollen limitation during the year that I conducted the test, but I did find significant associations between co-flowering community phenology and important aspects of pollinator efficiency.