688 LECTURE XXX. 



extent of the investigation in this field and to have shown how closely 

 related it is to physiological conceptions. We shall now turn our attention 

 to the results of Ehrlich's hypothesis in the study of hemolysis, which is 

 more closely related to our subject. 



It is a well-known fact that the sera of many varieties of blood are able 

 to dissolve the red corpuscles of other species of animals. This fact is the 

 cause of the bad results which have resulted from the attempts at the trans- 

 fusion of blood into human beings. For a long time little was known 

 concerning the nature of hemolysis. Recently Belfanti and Carbone 1 

 have established the fact that the serum of a horse into which the red cor- 

 puscles from rabbits have been injected, has a much more poisonous effect 

 upon rabbits than does normal horse serum. Bordet, 2 whom we have to 

 thank for developing our conception of hemolysis, showed that the serum 

 of guinea pigs into which there had been repeated intraperitoneal injections 

 of from three to five cubic centimeters of defibrinated rabbit's blood, would, 

 when placed in a test-tube, rapidly dissolve the red corpuscles of the rabbit, 

 whereas the normal serum of the guinea pig either did not have this property 

 at all, or showed but slight evidence of it. Here again we are dealing with 

 quite specific effects, and there is really a formation of anti-bodies here. 

 This discovery is of especial interest because Bordet has shown that even 

 cells to which we are not accustomed to ascribe toxic effects have a quite 

 similar effect to that of the bacteria. The phenomenon is not remarkable. 

 It merely shows us that every species of animal has its own peculiarly 

 constituted cells and thereby its own specific metabolism. 



We must, first of all, consider the explanation of how the hemoglobin 

 is removed from the red corpuscles under the action of the serum which 

 is employed. It cannot be a question of variations in the osmotic pres- 

 sure, for even a fraction of a milligram of the serum exerts this effect, and 

 again the specific action also contradicts any such assumption. All of our 

 knowledge points to a poisonous effect upon the red corpuscles them- 

 selves. We shall now try to apply Ehrlich's side-chain theory, as far as 

 possible, to the phenomenon of hemolysis. Bordet succeeded in proving 

 that a hemolytic serum loses its effect if it is heated to 55 C. The 

 serum is then designated as inactive serum. In order to avoid any mis- 

 conceptions, we had best take up a specific example. Let us assume that 

 we have a guinea pig which has previously been treated with rabbit's 

 blood. If the serum of this animal is allowed to fall drop by drop upon 

 an opaque solution of red corpuscles from a rabbit, contained in isotonic 



1 S. Belfanti and P. Carbone: Giorn. della R. Accad. di med. di Torino, 1898. 



2 J. Bordet: Ann. inst. Pasteur, 12, 688 (1898) ; 13, 273 (1899) ; 14, 257 (1900) ; 15, 303 

 (1901). Cf. Von Dungern: Miinchener med. Wochschr. Nos. 13 and 14, pp. 405, 449 

 (1899). K. Landsteiner: Zentr. Bacteriol. 25 (1899). P. Ehrlich and J. Morgenroth: 

 Berliner klin. Wochschr. 1899, 1900, 1901. P. Ehrlich and H. Sachs: 1902. 



