The independent evolution of floral phenotype is an important part of the process of gender specialization during the evolution of dioecy from hermaphroditism. However, we have little information on the genetic variation of floral traits in species with separate genders. Gynodioecious species (co-occurrence of females and hermaphrodites) have a breeding system intermediate between hermaphroditism and complete separation of the sexes (dioecy) and thus can provide insight into the genetic architecture underlying floral phenotype with respect to both primary (stamens and carpels) and secondary (petals) sexual traits. I used a nested breeding design to examine the potential for response to selection on floral traits and to examine whether this response would be similar in the two sex morphs of gynodioecious Fragaria virginiana. There was significant genetic variation underlying all floral traits, although narrow-sense heritabilities (ranging from -0.25 to 0.44) were, in most cases, much lower than broad-sense ones (ranging from 0.28 to 1.53). Moreover, the sex morphs differed significantly in their heritabilities for shared traits, such as stamen length, and showed a tendency towards differing significantly in others, like carpel number and petal length. In addition, correlations between the sex morphs for these traits (ranging from 0.41 to 0.58) were significantly greater than 0, but less than 1. These results indicate that greater sexual dimorphism could evolve in this population of F. virginiana, even if selection on these traits is not divergent. However, strong developmental integration of floral traits (e.g. stamen length and petal length) and high levels of nonadditive genetic variance may represent barriers to the evolution of complete sexual dimorphism.
