The effects of pollination and resources on the reproduction and establishment of Lespedeza virginica (Fabaceae).

Many flowering plants require animal pollinators for reproduction, and plants may compete for shared pollinators. Introduced plants may be new competitors which reduce the pollination and fruit set of native species. Few studies have examined pollination competition, used field-wide competitor removals, or examined the potential effects of introduced plants on pollination of natives. I investigated pollination competition between the native Lespedeza virginica (Fabaceae) and an introduced plant, Centaurea maculosa (Asteraceae), in old-fields of the Edwin S. George Reserve, Livingston County, Michigan, USA. I did field-wide removals of C. maculosa and determined if fruit set of Lespedeza was pollen-limited. Lespedeza is self-compatible, but it can not auto-pollinate and can be pollen-limited. I also added seeds to determine if populations were seed-limited. In addition, I experimentally manipulated resource availability to determine if Lespedeza's production of cleistogamous and chasmogamous fruits was resource-limited. Finally, I review experimental tests of competition for pollination and propose a model based on indirect interactions. In 1995 when I reduced Centaurea flower densities, visits per flower to Lespedeza significantly increased. However, fruit set of Lespedeza was not affected, suggesting that visitation in the presence of Centaurea was above a threshold required for fruit set. In 1996 neither visitation nor fruit set were affected by Centaurea removal. Pollinator visitation was lower in 1996, and there was a stronger correlation between visitation and fruit set, suggesting that visitation was below the threshold required for maximum fruit set. Therefore, in one of only a few studies of effects of introduced plants on natives as a result of pollinator-sharing, my results indicate no effect on fruit set. When I added seeds, I found a significant increase in seedling number, and subsequent survivorship data indicate that pollen limitation of seed set could affect subsequent populations. When I increased resource availability, the number of both cleistogamous and chasmogamous fruits increased, but chasmogamous fruits increased proportionately more. I propose a new evolutionary explanation for this pattern, based on how the cost:benefit ratio of chasmogamous flowers may change as flower number increases, due to increasing geitonogamy and decreasing pollen donation.PhDBiological SciencesBotanyEcologyUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/132001/2/9938535.pd