The evolution of two gene systems and of associated hybrid substances in columbidae.

F E R T I L E hybrids have been obtained from mating to Streptopelia risoria representatives of each of four other species of the genus-chinensis, humilis (or tranquebarica) , orientalis and senegalensis. Following backcrosses of the species hybrids and selected backcross hybrids to risoria, antigenic characters of the erythrocytes peculiar to each of these four species, in contrast to risoria, have been transferred to and isolated as units in risoria (IRWIN 1939; IRWIN and CUMLEY 1947; unpublished data). Such backcrolss birds are indistinguishable from risoria except for the presence of a cellular antigen derived from one of the four species. Antigenic factors (specificities or groupings) not demonstrable in the parental species have been observed in all the species hybrids, and in certain backcross individuals always in association with the antigens of one or the other o f two antigenic systems. These systems are called group-4 (composed of ch-4 from chinensis, hu-4 from humilis, or-4 from orientalis and se-4 from senegalensis) and group-8 (composed of ch-8, hu-8, or-8 and se-8 from the respective species and ri-8 from risoria). These new antigenic specificities are termed “hybrid substances”, and are detectable on the cells of all four kinds of species hybrids. All backcross hybrids with any member of group 4 or group 8 carry the hybrid substance associated with that antigenic character (PALM and IRWIN 1962; UNDERKOFLER and IRWIN 1965). Although the hybrid substances associated with the antigens of group 8 appear to be the products of interaction between alleles, either effecting the respective antigens of group 8 or linked to these alleles (PALM 1955; UNDERKOFLER and IRWIN 1965), recent evidence (IRWIN 1966a, 1966b) has shown that the specificities of the hybrid substance associated with the group-4 antigens in the backcross hybrids are a function of the interaction between the gene (or gene complex) affecting an antigen of group 4 and that for one of the genetically independent antigens of group 8. Further, all available evidence indicates strongly that certain of the antigenic specificities of hu-4 in humilis, or-4 in orientalis and se-4 in senegalensis are actually the result of interaction between a gene, or subgene, for the antigen of group 4 and the gene for group 8 within these species. That is, the antigenic specificities which previously