THE idea that there exists an intimate relationship between the forms of incompatibility revealed by blood transfusion and tissue transplantation has been more widely accepted than any concrete evidence for it allows.(Arguments for and against thetheory have been summarized by Medawar, 1943.) Immunity to" transplanted" blood appears to be of two sorts: innate (as with the ABO antigen system of the blood of human beings and anthropoid apes), or newly acquired in response to an antigenic stimulus. The latter represents the more usual state of affairs, and it is to this category that skin transplantation immunity belongs (Gibson and Medawar, 1943; Medawar, 1944, 1945a). The problem dealt with in this paper can be expressed in the following restricted but nevertheless workable form: do the cellular constituents of blood share antigens in common with the antigens of skin epithelium? The case of red cells may be considered first. If red cells and skin epithelium share antigens in common, then (a) the grafting of foreign homologous skin should in a proportion of cases elicit pari pasW the formation of red-cell isoagglutinins or isolysins; and (b) the transfusion or grafting of red cells should confer. an appreciable degree of immunity towards skin latergrafted from their, donor. In theory, these corollaries are equally applicable to the case of leucocytes; but the former,(a), is not workable because immutiity to leucocytes cannot bedemonstrated by an in vitro reaction. Immunity to homologous grafted skin may be recognized, however, by the suppression of cell-division in skin transplanted to a specifically immunized animal and by its greatly accelerated breakdown (GibsonandMedawar, 1943; Medawar, 1944, 1945a, 1946).