CROSSING DISTANCE EFFECTS ON PREZYGOTIC PERFORMANCE IN PLANTS - AN ARGUMENT FOR FEMALE CHOICE

Seed set in the perennial larkspur Delphinium nelsonii is greater in c rosses between plants growing an intermediate distance apart than in s horter and longer crosses. Since crossing distance is an attribute of a specific combination of parents, its effect on seed set represents a n interaction of these parents, and since genetic similarity declines with physical distance in D. nelsonii populations, the interaction ref lects parental genetic similarity. Understanding processes responsible for the effect is simplified if seed set differences result from prez ygotic events, because in this case no inbreeding or outbreeding depre ssion are involved. In 1990 we observed pollen tubes after performing crosses between plants separated by 1 m, 10 m, or 100 m, and found tha t the intermediate, 10 m crosses delivered the most tubes to the ovary . A combined analysis of the 1990 experiment and three earlier experim ents showed that this prezygotic difference in performance is signific ant and of the same magnitude as seed set differences. Disproportionat e failure of pollen tubes in 1 and 100 m crosses seems likely to refle ct a trait of the maternal parent rather than of pollen, because the k inship of haploid pollen to a zygote it has fertilized always exceeds maternal kinship to that zygote. Thus the conditions for ''pollen suic ide'' to evolve by natural selection are more restrictive than those f or ''female choice'' to evolve, and the latter is more likely to contr ol the outcome of the parental interaction.