Pollen donor performance can be consistent across maternal plants in wild radish (Raphanus sativus, Brassicaceae): a necessary condition for the action of sexual selection

The possibility that sexual selection has played a role in the evolution of plant reproductive characters remains interesting, but controversial. One reason is that clear demonstration of the necessary conditions for sexual selection is lacking. For sexual selection to occur, differences in pollen donor performance must be consistent across maternal plants and not be due to mechanisms such as self-incompatibility that produce interactions between maternal plants and pollen donors. Here, I performed two experiments with wild radish to test whether differences in pollen donor performance are consistent across maternal plants and not due to subtle effects of the incompatibility system. In the first, all maternal and paternal lineages had different S-alleles. There were 16 maternal plants, four in each of four lineages and four pollen donors, one in each of four lineages. Rank order of pollen donor performance, in terms of number of seeds sired after mixed pollination, was highly consistent across maternal plants and maternal lineages. In addition, maternal stress treatment had an effect on mating success of pollen donors, but the effect was subtle and did not affect the rank order of seeds sired by the four pollen donors. In the second experiment, pollinations were performed on both mature and immature stigmas. Immature stigmas allowed some self seed set, so the incompatibility system was compromised. There was some nonrandom seed paternity on both mature and immature stigmas. However, the amount of nonrandom mating was less on immature stigmas. Taken together, these experiments show that the kind of consistent nonrandom mating necessary for sexual selection occurs in wild radish, but that the incompatibility system and the mechanisms for sorting among compatible mates may overlap in time of development or in some pathways.