Natural selection on floral traits: variation in agents and targets

点击翻译按钮获取中文摘要

Flowers determine mating opportunities in animal-pollinated plants because pollinators transport male gametes (pollen) between flowers and plants. Pollinators form an essential "ecosystem service" in natural and agricultural ecosystems by ensuring fruit and seed production. As a result, natural selection by pollinators contributes to the tremendous diversity in wild flowers and inflorescences. Our understanding of floral evolution is still limited in several areas. These include: (1) selection through siring success, even though male parents contribute half the genes to each generation; (2) how selection affects floral traits when floral visitors are diverse; and (3) how often reproduction is limited by pollen receipt versus the availability of water, nutrients or other resources. My proposed research addresses these gaps.**Polemonium brandegeei is a perennial herb pollinated by hummingbirds and hawkmoths, and visited by pollen consuming flies (pollen thieves). Our work to date indicates that the two pollinators exert contrasting natural selection on sex-organ position and flower width, suggesting that flower shape may be a compromise between the phenotype that is best under each pollinator. My lab will first conduct controlled experiments measuring how selection by each pollinator is modified by differences in flower number, a trait that varies widely in natural populations. Second, we will determine how pollen consumers affect siring potential of different flower shapes, and whether they limit the pollen available for seed set. Third, we will initiate long-term studies to quantify selection in natural populations. We will include estimates of siring success (paternity) based on genetic fingerprints. Supplementation of natural pollen receipt will distinguish selection by pollinators from selection caused by other factors such as nutrient limitation or drought.**Lady's slipper orchids (Cypripedium) trap insect visitors in their flowers rather than producing floral rewards to retain pollinators. I hypothesize that the small flowers and narrow pollination routes of endangered C. candidum contribute to low fruit set in northern populations because few insects are both small and strong enough to escape past the sex organs. Higher fruit set in widespread C. parviflorum and hybrids may reflect larger flowers and wider pollination routes. These differences in fruit set may affect gene transfer between species. First, I propose to extend our surveys of flower size, fruit set, and habitat quality across C. candidum's range, compare pollinator communities in selected populations, and use genetic fingerprints to assess the direction of gene flow between species. Second, my lab will examine the direct effect of pollination route size on pollen movement; we will use plants from a C. parviflorum population, where flowers are comparable in size to either typical flowers of C. candidum (small) or C. parviflorum (larger) flowers, but do not differ in other traits such as flower colour or scent.**These studies will advance my long-term goals of understanding how diverse floral visitors affect selection on floral design and display, and how floral phenotypes affect reproduction and mating patterns when flowers do not reward pollinators. They will also contribute to general evolutionary questions regarding the importance of variation in selection for the maintenance of genetic variation, evolution of specialized vs. generalized phenotypes, and gender conflict in selection. Findings will increase our understanding of evolution and ecological genetics, and will inform conservation biologists about strategies for preserving endangered species, and members of the public interested in evolution and natural history.