904 THE HUMAN BLOOD GROUPS 



substances with alcohol were made with a number of different erythrocytes and with 

 the human isoagglutinogens.^ In the latter case the extracts were found to give 

 flocculation reactions and complement fixation with group-specific immune sera and 

 to stimulate the production of antibodies when injected in mixture with proteins. 

 These facts support the view that the individual differences which have been brought 

 out by serological tests are due to variations of specific substances not belonging to 

 the class of proteins. The reports on isoprecipitins, perhaps indicating individual pro- 

 tein differences, are few and not quite convincing.^ 



One may hope that further work will ultimately succeed in isolating the agglu- 

 tinogens A and B in a pure state and will bring to light a characteristic chemical dif- 

 ference between them. It is still doubtful, however, whether a special chemical sub- 

 stance corresponds to each of the serological factors which are found when several 

 different antigens are tested with a number of serological reagents. 



COMPATIBILITY TESTS PRELIMINARY TO TRANSFUSION^ 



The use of the isoagglutinin reaction for the selection of proper blood donors was 

 suggested by the present writer in the first publication on the blood groups in 1901, 

 and this application has been emphasized by Hektoen and others. At the present time 

 the test has become a routine procedure. It can be taken as established that incom- 

 patibility as regards isoagglutination is the main cause of untoward events following 

 transfusion, and a number of accidents could be traced directly to this circumstance 

 (Plehn, Brem, Jervell, De Pemberton;^ cf. the experiments of Schulz). An experimental 

 confirmation of this view was offered by the work of Ottenberg and his collaborators, 

 on transfusions in dogs and cats.^ 



Ottenberg'' called attention to the fact that a difference exists between the trans- 

 fusion of a blood, the serum of which agglutinates or hemolyzes the corpuscles of the 

 recipient and the reverse case, i.e., the injection of blood cells which are agglutinated 

 by the recipient's serum. In the latter event the erythrocytes come in contact with a 

 large amount of agglutinating serum, while in the former the serum of the donor is in 

 most cases mixed with such an excess of the patient's blood as not to affect it to any 

 appreciable extent. 



According to this principle, group O donors (universal donors) are frequently em- 

 ployed for patients of any group. The usefulness of this procedure becomes evident 

 in emergency cases. Similarly, AB patients are transfused with the blood of all other 



' Landsteiner, K., and van der Scheer, J.: Joe. cil.; Doelter, W.: loc. cit.; Brahn, B., and SchiflF, 

 F.: Klin. Wchnschr., 5, 1455. 1926; Witebsky, E.: Ztschr.f. Immunitdtsforsch. u. exper. Therap., 48, 

 369; 49, I, 517. 1926-27; Halber, VV., and Hirszfeld, L.: loc. cit.; Bordet, J., and Renault: Soc. Biol., 

 95, 888. 1926. 



2 Cf. Landsteiner, K., and Levine, P.: loc. cil. Brahn and Schiff consider the existence of group- 

 specific proteins in the erytlirocytes. 



3 For clinical aspects of transfusions see Jagic, N.: Wien. klin. Wchnschr., 40, 1565. 1927; 

 Breitner, B.: ibid., 41, 77. 1928; Rolleston, H.: Bril. M. J., 2, 969. 1926. 



4 de Pemberton, J.: J. Iowa Slate Med. Soc, 10, 170. 1920. 



5 Ottenberg, R., Kaliski, A. J., and Friedman, S.: /. Med. Research, 28, 141. 1913; Ottenberg, 

 R., and Thalhimer, W.: ibid., 33, 213. 1915. 



* Ottenberg, R.: /. E.vper. Med., 13, 425. 1911. 



