Understanding the factors that drive the maintenance of polymorphisms has long been of interest to evolutionary biologists. Theory predicts that polymorphism should be rare, as selection should remove the less fit variant; nevertheless numerous examples of polymorphism exist globally. Polymorphism in floral traits is of particular interest since it can promote reproductive isolation. Previous studies have shown floral color polymorphism can be maintained through frequency-dependent selection, species reinforcement, and temporal shifts in fitness associated with precipitation or herbivory. The floral bract of Castilleja coccinea, the scarlet paintbrush, ranges from orange-red to yellow. Little is known about how this variation is maintained. In this study, we set out to describe the color polymorphism in C. coccinea across its range, and to identify the potential drivers maintaining this polymorphism. We had two objectives: 1) to characterize the spatial distribution of color morphs of C. coccinea and 2) to identify potential abiotic or biotic agents. To address the first objective, we used herbarium specimen data to determine if color morphs were evenly distributed throughout the range. To address the second objective, we focused on a contact zone between morphs and compared fitness traits (seed set and herbivory), morphological variation, and community composition between populations of different morphs in the field and using common gardens. .
Results/Conclusions
The results of this study indirectly suggest both pollinators and edaphic conditions as potential selection pressures driving color variation in Castilleja coccinea. We established that C. coccinea morphs are non-randomly distributed throughout the range of the species, and that most populations are comprised of a single morph. This pattern is consistent with selection as the primary driver of polymorphism in this species. However, the transition between color morphs was patchy rather than clinal, with both red and yellow populations within the flight range of most pollinators, suggesting that differences in pollinator communities may not have been the initial selection pressure driving variation. Instead, we found evidence that red morphs are more common in sites that are drier and have more canopy cover, both range-wide and within populations. This is consistent with diversification associated with a shift in ecological niche. However, the shift in floral color was also associated with differences in floral structures and slightly higher female (fruit to flower ratio) and male fitness (more yellow progeny), suggesting that these floral trait shifts are also associated with adaptation to the pollinator environment.