A field investigation into potential reproductive and resistance advantages of purple-petalled plants in polymorphic populations of horsenettle (Solanum carolinense)

Background and aims – Variation in flower color among individuals within plant populations has long fascinated horticulturalists and evolutionary biologists alike. Nevertheless, the diversity of genetic factors and ecological interactions that act to maintain flower-color polymorphisms over time are understood for very few species. Here, we suggest that Solanum carolinense (horsenettle) makes an excellent model system for studying the evolutionary ecology of a flower-color polymorphism, and we begin to shed insight into factors that might affect the stability of the polymorphism. Material and methods – We transplanted 24 horsenettle ramets of each of 40 field-collected genets (10 genets per each of four source fields) into a common-garden plot (in an oldfield with an existing horsenettle population). For each ramet, we visually characterized petal color and quantified production of flowers, fruits, seeds, and damage by 11 species of herbivores. Key results – Rather than varying continuously within ramets and among genets, petal color fell into three relatively discrete categories that are common in most horsenettle populations: purple, white, and mauve. In the common-garden experiment, purple-petalled morphs initiated significantly more flower buds, opened more flowers, and produced more fruits and seeds than did the mauve-petalled or white-petalled morphs. Petal-color morphs differed only slightly in resistance to a minority of the herbivore community, with no consistent advantage for color morphs across species of herbivores or types of tissues fed upon. Conclusion – Horsenettle has numerous characteristics that make it an excellent system for studying the genetics and ecology that drive the maintenance of flower-color polymorphisms. Although pleiotropic effects of flower-color on resistance to herbivory appear to be minimal in horsenettle, potential pleiotropies that differentially affect maternal and paternal reproduction, or sexual and vegetative reproduction, are worthy of further study.