ABO groups of infants and children dying in the west of Scotland (1949-1951).

Hirszfeld and Zborowski (1925) and Levine (1943) found fewer group A children in AO families (father group A, mother group O) than in OA families (father group O, mother group A). The statistical significance of this deficiency of group A children was firmly established by Waterhouse and Hogben (1947) who reckoned that 25 per cent, of the expected number of group A children were missing from the AO families of the general population. By Bernstein's hypothesis, however, group A children are conceived with equal frequency in AO and in OA families, and it therefore follows that the missing children must have died some time between conception and childhood. Apart from ABO iso-immunization, or some similar mechanism involving the natural alpha-antibodies carried by the mothers of AO families, all factors respon sible for the loss of group A children should operate alike in AO and OA families. Thus Waterhouse and Hogben concluded that ABO iso-immunization was probably responsible for the deficiency of group A children observed in AO families (i.e. for the loss of about 3 per cent, of all conceptions in the general population), the total loss of life from haemolytic disease of the newborn due to Rh-sensitization being less than 0-5 per cent, of all conceptions. The fate of the missing children has yet to be ascertained. Because ABO iso-immunization does not contribute materially to the incidence of haemolytic disease, and because the agglutinins concerned are normally present in the maternal blood early in pregnancy, Waterhouse and Hogben suggested that ABO iso immunization acts early in foetal life, causing abortions and miscarriages. It is undoubtedly true that very few of the missing children die of haemolytic disease, but it may be argued that immaturity of the agglutinogens makes it impossible for ABO iso-immunization to act early in foetal life. Furthermore, the absence of a significant deficiency of group A among the newborn of group O mothers in a series of 2,000 births reported by Boorman (1950), and in a series of 7,856 births reported by Bryce and others (1950), suggests that some of the observed deficiency of group A children may have arisen after rather than before birth, an excess of group A children born of group O mothers having died at an early age from conditions other than haemolytic disease.