Most population genetics theory assumes that the alleles at a given locus segregate according to Mendel's first law and that there is no prezygotic selection. These assumptions are probably valid in the majority of cases. Very little prezygotic selection has been found to occur in animals. In fact, the results of experiments in Drosophila which seem to indicate a lack of gene activity in mature sperms (Muller and Settles, 1926; McCloskey, 1964) have so pervaded animal genetics that only as a last resort would anyone propose prezygotic selection as the explanation of a non-Mendelian recovery ratio. Events formally equivalent to gametic selection can occur in the absence of gene activity in sperms, however (Lindsley and Grell, 1969), provided a mechanism exists whereby gene products can be differentially partitioned to spermatids during meiosis. Gametic selection may be an important phenomenon in plant evolution (Jones, 1920; Beale, 1939; Afzal and Khan, 1950; Bateman, 1956; Ricelli-Mattei, 1968; Harding and Tucker, 1969). The extended haploid phase in the life cycles of many plants certainly presents the opportunities for selection. One striking example of prezygotic selection in barley has been presented by Doll (1968) who found that about onefourth of the newly induced chlorophyll mutants have unequal gametic ratios. As for meiotic drive, it is by no means a common phenomenon. On the other hand, a considerable number of isolated cases have now been described in both plants and animals (examples: Gershenson, 1928; Sturtevant and Dobzhansky, 1936; Rhoades, 1942; Dunn, 1956; Cameron and Moav, 1957; Sandler et al., 1959; Loegering and Sears, 1963; Redei, 1965; Hickey and Craig, 1966; Hanks, 1969). The extent to which these loci are important in evolution is still generally unknown although several cases have been extensively investigated (sex-ratio in the Drosophila obscura species group, Sturtevant and Dobzhansky, 1936; Stalker, 1961; t-alleles in the house mouse, Lewontin and Dunn, 1960; Lewontin, 1962, 1968; SD in Drosophila melanogaster, Hiraizumi et al., 1960; Katoaka, 1967; Hartl, 1970a). The possibility that non-Mendelian segregation may at times be important in evolution makes it seem worthwhile to develop a general theory of evolution which can accommodate prezygotic selection and meiotic drive. Several simple cases involving two alleles have been worked out (Haldane, 1924; Prout, 1953; Bruck, 1957; Lewontin, 1958, 1968; Hiraizumi et al., 1960; Scudo, 1967; Hartl, 1970b, 1970c). The present paper extends the theory to multiple alleles at one locus. The aim is to present a general formulation of the problem and to apply it to a few special cases. The fundamental theorem of natural selection is also extended to include nonMendelian segregation.