have examined the factors of the ABO and M N blood-group systems in 107 families consisting of father, mother and children, and in 218 unrelated subjects, resident in England. Of the 214 parents ten were blood relations of one or other of the remaining 204, who with the 218 unrelated persons above-mentioned made a random sample of 422 subjects. The distribution of these 422 amongst the various blood groups has been reported and the results analysed statistically (19386).The present paper is concerned with the heredity of the blood groups, as exemplified by the results of the examination of parents and their children.THE ABO SYSTEM I n the second of the above-mentioned papers the theory of Bernstein (1924), which is generally accepted as explaining the heredity of the ABO factors, was outlined, andit was shown by a statistical test that the distribution of the ABO factors among the unrelated subjects was consistent with this theory. The types of children which may result from the different kinds of mating are shown in Table I. A necessary consequence of this theory Table I A
x 3 B x B O x A B A x A B B x A B A B x AB
A , B A , B, A S A , B, AB A , B, ABis that no child can have a factor which is present in neither of its parents, e.g. an A child cannot be produced by non-A parents. Further, a parent of group 0 cannot have a child o f group AB, nor can a group AB parent have a child of group 0.As mentioned in our second paper (19386), the factor A can be divided into two subfactors, A, and A,, and so the original four groups are increased to six: 0, A,, A,, B, A,B and A,B. The four-gene theory of the method of inheritance of this extended ABO system was put forward by Thomsen, Friedenreich & Worsaae ( 1 930 a , 6 ) . It is an extension of the three-gene theory of Bernstein which it in no way invalidates. The types of matings and their possible offspring are given in Table 11.Of the 107 families detailed in Table I V of our first paper (1938a) the first six had been * Supported by the Rockefeller Foundation.