Bats, birds, and Burmeistera: The evolution of specialized pollination in the neotropics

Specialization in pollination systems has been a central process in the evolution and diversification of angiosperms. However, we still lack an understanding of why plants specialize or switch pollination modes. I studied various aspects of the pollination and floral evolution of the neotropical subshrub, Burmeistera (Campanulaceae). Videotaping flowers and quantifying pollen transfer demonstrated that most species are pollinated by bats, with hummingbirds occasionally serving as secondary pollinators. One species (Burmeistera rubrosepala) was exclusively pollinated by hummingbirds. Floral traits matched the traditional bat and hummingbird pollination syndromes. The degree of exsertion of the floral reproductive parts varied extensively across the genus. To explore the adaptive significance of this trait, I analyzed pollen deposition on Burmeistera flowers, and captured bats to examine pollen loads on their fur. Results show that variation in exsertion length does not correspond to specialization on different pollinators; hummingbirds are not more effective pollinators of short-exserted flowers, and different bat species are just as effective pollinators of all flower types. However, pollen from differently-exserted flowers was found on different regions of bats\u27 heads, suggesting that variation in exsertion may serve to reduce competition for pollination by spatially partitioning bats\u27 bodies. Flight cage experiments with bats and pairs of Burmeistera provide further support for this hypothesis: the greater the difference in exsertion length, the less pollen transferred interspecifically. Additionally, null model analyses of 18 sites (each with 2-4 Burmeistera species) demonstrated that exsertion lengths of coexisting species were significantly overdispersed relative to what would be expected by chance. Thus results provide strong evidence of floral character displacement: coexisting Burmeistera evolved different exsertion lengths to reduce competition for pollination. For the final part of this project, I wanted to know why Burmeistera specialize on pollination by either bats or hummingbirds rather than using both. Flight cage experiments with artificial flowers demonstrated that flower-pollinator fit is critical; wide corollas guided bat snouts better, while narrow corollas guided hummingbird bills better. Poor fit resulted in variable entry angles and decreased pollen transfer. This adaptive tradeoff was strong enough to favor specialization rather than generalization on both bats and hummingbirds