Date of Completion


Embargo Period



merosity, pentamery, petal number, meristic variation, Polemoniaceae, Phlox, floral evolution, floral selection, floral modularity

Major Advisor

Carl Schlichting

Associate Advisor

Elizabeth Jockusch

Associate Advisor

Gregory Anderson

Associate Advisor

Pamela Diggle

Field of Study

Ecology and Evolutionary Biology


Doctor of Philosophy

Open Access

Open Access


Petal number is diverse among angiosperms, yet many clades are fixed for five-petaled flowers, with widespread stasis within and among species. This stasis has been presumed adaptive, maintained by pollinator-mediated stabilizing selection. Despite broad patterns of stasis, some species exhibit within-individual variation that can be used to test mechanisms of stabilizing selection that promote stasis on broader scales. By testing for a genetic basis to variation in petal number, and for pollinator-mediated or pleiotropy-mediated stabilizing selection on this variation, the sufficiency of adaptive stabilizing selection as a driver of petal number stasis can be evaluated.

I conduct a selection experiment and show that within-individual variation for petal number is genetic. Therefore, stabilizing selection can act to this variation to enforce canalization. Correlated responses in other floral organs to selection for more, but not fewer, petals show that pleiotropic constraints with other floral organ numbers cannot entirely explain petal number stasis. Comparing levels of petal number variation in natural populations between autogamous and outcrossing congenerics show no relaxation of pollinator-mediated selection under autogamy, suggesting that pollinators are indiscriminate. Further evidence from a pollinator visitation experiment shows that pollinators exhibit no preference for five-petaled flowers, failing to explain stasis. My work demonstrates that while variation in petal number is under genetic control, mechanisms controlling stasis remain elusive. Focusing on pollinator preference as the driver of stasis may be misguided, and I suggest that environmental factors, developmental constraints, and pleiotropy are more important in explaining the adaptive nature of petal number.