Abstract Detail

Reproductive Processes

Harrington, Aidan [1], Briscoe Runquist, Ryan [2], Moeller, David [1].

Pollinator-mediated selection via male fitness drives floral variation across the range of Clarkia xantiana.

Evolutionary biologists have long interpreted variation in floral morphology as a consequence of selection by pollinators. However, the extent to which population variation in floral form is determined by pollinator selection versus other biotic and abiotic factors remains unresolved. Moreover, plants with bisexual flowers experience selection through both male and female fitness simultaneously but most studies of natural selection focus on female fitness alone. In Clarkia xantiana, a common garden experiment shows a pronounced gradient in flower size, color, and herkogamy from the western to eastern range margins. We tested if and how pollinator-mediated selection contributes to observed spatial patterns of floral differentiation using a novel approach. We reciprocally translocated genotypes from opposite ends of the gradient into experimental arrays in order to create phenotypically variable experimental populations, which improves the power to detect selection. We then placed experimental arrays in 16 populations spanning the entire west-east gradient to ask whether patterns of selection recapitulate patterns of population differentiation. We quantified selection via female fitness (pollen deposition) and male fitness (pollinator visitation, which closely correlates with siring success in prior research). We found that selection via female function was insufficient to explain variation in any floral trait across the range. Conversely, selection via male function favored larger flowers in the east compared to west, paralleling patterns of differentiation. Surprisingly, the trend was opposite for female fitness (albeit marginally insignificant) where smaller flowers were favored in the east compared to west. There was also some evidence that flower color was selected in the direction opposite to the geographic pattern of natural variation via male fitness (and unrelated to female fitness). Our results suggest that non-pollinator agents of selection may be important for explaining spatial variation in floral traits that are often thought to be involved in pollinator attraction (e.g. color). In addition, male fitness appears to be a critical factor that can drive the evolution of flower size across populations, consistent with past studies of individual populations.

1 - University of Minnesota, Twin Cities, Plant and Microbial Biology, St. Paul, MN
2 - Evolution And Ecology, 250 Biological Sciences, 1445 Gortner Ave, St. Paul, MN, 55108, United States

none specified

Presentation Type: Oral Paper
Number: RP4002
Abstract ID:637
Candidate for Awards:None

Copyright © 2000-2022, Botanical Society of America. All rights reserved