In 1964, Seyfried reported a family case where the inheritance of ABO blood groups did not follow the one genetic locus-three allelic model proposed by Bernstein. The ABO blood types of the father and mother were O and AB, respectively, but those of their two children were AB. The blood type of the mother’s mother was O. The next year Yamaguchi and colleagues reported an A2B3 phenotype. The family study suggested that the A2B3 phenotype was inherited in the cis manner, and they named this allele the cis-AB allele. As opposed to the regular trans-type of AB where the A allele is derived from one parent and the B allele is derived from the other parent, the cis-AB behaves as one unit and transcends from one parent to the next generation. Yoshida and colleagues characterized A and B transferase activity in the sera of the cis-AB individuals, and proposed that there are two discrete molecular mechanisms of the occurrence of the cis-AB allele: either crossing over, resulting in 2 alleles on a chromosome, or structural mutations leading to the production of an enzyme with bifunctional activity. Our molecular study has identified structural mutations in several cis-AB alleles.