Sibling Competition and Evolution of Pollen Unit, Ovule Number, and Pollen Vector in Angiosperms

Abstract

The pollination biology of angiosperms affects the quantity and qual- ity of seeds produced by an individual. The amount of genetic variability present in the seeds of a single fruit is dependent on the number of male parents involved in fertilization. The variation in genetic relatedness of developing siblings in an ovary may lead to varying degrees of competition among young seeds for maternal nu- trients. Kin selection theory predicts that closely related seeds resulting from fertil- ization by a single paternal parent will compete less with each other than seeds fertilized by different fathers. Ramdom dispersal of single pollen grains character- istic of wind pollination tends to produce less closely related seeds in multi-ovulate ovaries. Pollen transferred between flowers in large aggregations by specialized an- imal pollinators gives rise to seeds that are full siblings. The evolution of animal pollination, large pollen unit size, and multi-ovulate ovaries in some angiosperms may be understood in terms of competitive interactions among the developing seeds. The multi-seeded angiosperm fruit contains a unique collection of in- dividuals. These individuals are young sporophytes that share a common maternal parent. The group is special because of the relatively high ge- netic relatedness of its members. The high genetic similarity is important in two ways. First, within a single fruit it sets an upper limit to the amount of genetic variation controlling the rate and quality of seed development. Second, the large number of genes by descent shared by these sporophytes will augment the inclusive fitness component of offspring reproductive success and influence their competitive interactions. Inclu- sive fitness is defined here as the total reproductive value of an individual including personal production of offspring plus the individual's effects on the reproduction of relatives (Hamilton 1964; West Eberhard 1975). Haldane (1923) originally noted that the pollination system in angio- sperms determines the number of paternal parents involved in fertiliza- tion. Later Hamilton (1964) suggested that seeds in the flowering plant ovary can be half-sibs, full-sibs, or genetically identical individuals de- pending on the manner in which pollen is transferred between plants. He also pointed out that the variation in genetic relatedness of sibs might influence the degree of competition and cooperation among them. To date there has been no experimental confirmation of the hypothesis that kin selection has led to the evolution of cooperative interactions among fertilized ovules in the angiosperm ovary. The difficulties encountered in studying kin selection as an evolutionary process in the laboratory have been pointed out by Wade (1980). In the present paper I attempt