The kind of explanation involved in the "CDEF" symbolism of the Rh blood groups is examined both at the level of the antigen and at the level of the gene. The B system of cattle is used as a model throughout the discussion and parallels are drawn between that system and Rh. For example, in the discussion of compatibilities and incompatibilities among blood factors of a system, attention is called to those examples where one blood factor, as K of cattle, appears to be completely dependent on one or more others (B and G) which occur independently of the dependent factor. The parallel in Rh is the "cef" relationship in which f has not been observed independently of the set cef. Relationships of this order not only defy interpretation on the basis of linked genes but also on the basis of independently mutating elements within a single gene. They are, however, reaily resolved in serological terms, namely by cross or over- lapping reactions. Such reactions are precisely what might be expected among the products of a single set of allelic genes. Consequently, the current debate is returned to the level of the antigen where it more properly belongs. It is proposed that incompatibilities among blood factors relate mainly to differences in isomerism or surface configuration in antigens or haptens of a homologous series whereas differences in sub-stituent groups at a number of intramolecular antigenic sites account for compatibilities. Steric hindrance would then play the predominant role in eliminating whole blocks of serological reactions. Finally, it is pointed out that in immunogenetic studies, as in other genetic studies, the most immediate approach to an understanding of the gene is through the study of the effects of gene mutations. Two possible orders of mutational effects among B alleles of cattle are discussed. Since the view is adopted that crossing over may be one of the many forces capable of producing mutations, the pseudoallelic effect is not excluded.