Architecture, size, and reproduction in plants : a pollination study of Mertensia ciliata (James) G. Don

Graduation date: 1982I experimentally test the hypothesis that the potential for selfing\ud increases with plant size in Mertensia ciliata (Boraginaceae), a self-compatible,\ud profusely-flowering perennial. This follows from the\ud premises that 1) pollen dispersal by pollinators between flowers is\ud limited, and 2) individual pollinators, in this study bumblebees, will\ud visit more flowers per visit on large than on small plants thereby\ud promoting intra- rather than interplant dispersal of pollen. I show\ud that while M. ciliata is self-compatible, outcrossing results in greater\ud reproductive output (seed numbers and seed weight) than selfing. Thus,\ud under the hypothesis above, the reproductive output of flowers should\ud decline with increasing plant size.\ud I demonstrate, through pollen carryover experiments, that pollen\ud transfer by bumblebees is extensive. Observations of pollinator\ud foraging behavior show that individual bumblebees visit only a few more\ud flowers and stems, and indeed encounter a smaller proportion of a\ud plant's flowers and stems on visits to large than to small plants.\ud Large plants attract more pollinators per minute. I use these results to predict that flowers on all plants should receive equal amounts of\ud outcrossed and total (self + outcrossed) pollen, and that selfing rates\ud should not differ among plants. This is supported by direct\ud measurements of pollen receipt by flowers, and of the reproductive\ud output of flowers on large and small plants in the field. No\ud differences were found among plants in outcrossed and total pollen\ud receipt, and in seed-set per flower and seed weight.\ud I examine the pattern of insect visitation in more detail to show\ud that individual bumblebees encounter only a small number and proportion\ud of flowers and stems per visit on all plants, and encounter a smaller\ud proportion on large than on small plants. Individual bees, then,\ud exploit large plants less intensely per visit than small plants. Bees\ud move predominantly between neighboring plants and fly randomly with\ud respect to direction. Many insect visitors are nectar robbers. I\ud propose four factors to explain the short visits of pollinators and the\ud less intense exploitation of large plants by individual bumblebees.\ud These include the complex architecture of the flowering display, the\ud circular geometry and density of stems in plants, the close proximity of\ud plants in the population, and the variance among flowers of plants in\ud nectar reward caused by visits of nectar robbers