PLURALITY OF COMPLEMENTS OF THE SERUM. 197 



i 



plementophile groups. To one, therefore, conversant with this con- 

 ception, Bordet's conclusion cannot appear otherwise than forced. 

 The unity of the complement would only then be demonstrated 

 by Bordet's experiment if in the immune serum employed for absorp- 

 tion but a single complementophile group came into action, and 

 not a plurality of groups. 



Despite these objections raised against Bordet's evidence, and in 

 spite of Ehrlich and Morgenroth's previous positive demonstration 

 of the plurality of the complements, it seemed advisable, owing 

 to the importance of the question, to enter once more upon a thorough 

 investigation of the subject. We at first confined ourselves to the 

 complements which effect the hsemolytic actions, and have been 

 able to bring forward a large number of new and more conclusive 

 proofs for the diversity of these complements in the same serum. 

 These investigations have in part already been mentioned by Ehrlich 

 at the Congress of Naturalists in Hamburg. 



The method of the experiments was guided by the following 

 considerations. If only a single complement is present in a cer- 

 tain serum, it follows that all the complement actions of this serum 

 would be weakened equally by any given influence, chemical, physi- 

 cal, or thermic. If, on the contrary, our view of the plurality of 

 complements is correct, it should be possible through appropriate 

 experimental conditions to influence the serum in such a way that 

 only a part of the complements will be destroyed, while others remain 

 intact. Not only the absolute inhibition of the action of a few com- 

 plements, but also marked quantitative differences in the impair- 

 ment of the individual completions can only be satisfactorily explained 

 by the assumption of different substances as carriers of these prop- 

 erties. A single complement would have all its functions impaired 

 equally. 



We have especially subjected the complementing property of 

 goat serum to a thorough analysis, using for this purpose five different 

 combinations which can be activated by goat serum. For simplicity's 

 sake, we shall designate them by the following numbers : 



Case I. Guinea-pig blood inactive normal goat serum. 



Case II. Rabbit blood inactive normal goat serum. 



Case III. Rabbit blood inactive serum of goats immunized with 

 rabbit blood. 



Case IV. Ox blood inactive serum of goats immunized with 

 ox blood. 



