894 THE HUMAN BLOOD GROUPS 



Another method of identification consists in the use of immune sera against Forss- 

 man's antigen, some of which agglutinate intensely human blood of group A, as was 

 discovered by Schiff and Adelsberger. 



The assumption of two agglutinins and two agglutinogens as underlying the 

 typical iso-reaction of human blood is, in a general way, supported by the results of 

 absorption experiments. Thus, a serum O treated with corpuscles A still contains 

 agglutinins for cells B and vice versa, and cells AB absorb from any serum the aggluti- 

 nins for cells A and B, The conception of three agglutinins and three agglutinogens, 

 as advanced by Moss,' would require that cells A or B absorb the agglutinins from sera 

 O just as well for cells A as for B, which is not borne out by the experimental results. 



While the hypothesis of two pairs of agglutination elements is practically ade- 

 quate it cannot be considered as strictly accurate. It was found that by absorb- 

 ing sera O with cells A or B not infrequently a part of the agglutinins is removed 

 for cells B or A, indicating that in sera a part of the agglutinins a and may 

 exist in combination.^ A certain relationship of the agglutinogens A and B is 

 indicated by the experiments of Hooker and Anderson,^ Landsteiner and van der 

 Scheer,4 and Doelter.s 



In regard to corpuscles a new point was raised by the genetic theory of Bern- 

 stein,^ While von Dungern and Hirschfeld explain the inheritance of the blood groups 

 by postulating two dominant Mendelian genes A and B, the absence of which would 

 determine the appearance of group O, Bernstein, by a calculation based on the fre- 

 quency of the group factors, was led to the assumption of three allelomorph genetic 

 factors^ — A, B, R. The genetic formulas for the groups are, according to Bernstein's 

 hypothesis, the following: 



Groups O A B AB 



Genetic formulas RR AA or BB or AB 



of Bernstein . . AR BR 



Since R is a genetic factor, the formula RR for corpuscles does not imply the exist- 

 ence in these corpuscles of a special agglutinogen demonstrable by serological reac- 

 tions, but in a recent paper Schiff^ reports that he obtained fluids reacting by prefer- 

 ence on erythrocytes O by absorbing certain normal beef sera with red cells AB. The 

 corpuscles would then seem to be characterized not merely by the absence of the 



' Moss, W. L.: Bull. Johns Hopkins Hasp., 21, 63. 1910. 



2 Landsteiner, K., and Witt, D. H.: /. Immunol., 11, 221. 1926. This phenomenon is possibly 

 of significance for explaining some observations of Dyke, S. C: Brit. J. Exper. Path., 7, 294. 1926. 



3 Hooker, S. B., and Anderson, L. M.: /. Immunol., 6, 419. 1921. 



-•Landsteiner, K., and van der Scheer, J.: /. Exper. Med., 42, 123. 1925; Proc. See. Exper. Dial. 

 &• Med., 22, 289. 1925. 



s Doelter, W. : Ztschr.f. Immunitatsforsch. u. exper. Therap., 43, 95. 1925. 



* Bernstein, F.: Ztschr.f. indukt. Abstamm. u. Vererb., 37, 237. 1925; Bernstein's theory is 

 strongly supported by the recent studies of SchifE, F.: Deutsche Ztschr. d. ges. gerichtl. Med., 9, 369. 

 1927; and Thomsen, O., ibid., 10, i. 1927. 



7 For a full presentation of the genetic data see chapter Ixix in this volume by Ottenberg and 

 Beres. 



* Schiff, F.: Klin. Wchnschr., 6, 303. 1927. 



I 



