FELDMAN and ESHEL (1982) have introduced a precise two locus genetical model to examine the spread of "selfish" alleles in a genetically "altruistic" population. They have shown that, with complete linkage between t h e gene affecting "altruism" and another determining parental "interference", the development of "selfishness" may be prevented. Furthermore, "interference", may allow the development of "altruism" in a "selfish" population. Small population results illustrate how a deterministic analysis may not provide a satisfactory description of how particular rare events occur. The usefulness of major gene models of this type is discussed.