126 COLLECTED STUDIES IN IMMUNITY. 



Typhoid + inactive immune serum (dog) + normal active guinea-pig serum. 

 Vibrio Nordhafen + inactive immune serum (rabbit) + normal active horse 



serum ; 



it " " " + normal active goat 



ft " " " serum; 



it " " + normal active sheep 



serum; 



tt it ft " " + normal active guinea- 



pig serum. 



In order to meet the objection that the agglutinins may possibly 

 have interfered in the experiments we have devised the following 

 method of demonstrating the phenomenon in question : 



Typhoid bacilli were subjected for one hour at 37 C to the action 

 of inactive immune serum derived from a dog. As we know from the 

 haemolytic experiments of Ehrlich and Morgenroth, this results in 

 anchoring the interbody present in the immune serum to the bacteria. 

 The mixture was then centrifuged and the fluid poured off. After care- 

 fully shaking the sediment with a little fluid it was divided into two 

 equal parts, to one of which inactive immune serum (dog) was added, 

 while the other received some normal inactive dog-serum. Finally 

 there was added to both portions the same amount of a complement- 

 ing serum (normal active guinea-pig serum) which by itself was able 

 to kill the bacteria. At the end of three hours plate cultures were 

 made in the usual manner. The results showed that no destruction 

 had occurred in the tube containing the excess of immune serum, 

 whereas the culture in the other tube had been killed. 



reactivated for Vibrio Nordhafen by a complement derived from rabbits. In 

 this combination we again observed the phenomenon of deflection of com- 

 plement by an excess of immune body. But even normal inactivated goat 

 serum (which contains interbody) when used in exactly the same amounts 

 manifested deflection of complement. Since no quantitative difference could 

 be discovered between the immune serum and the normal serum,' we assume 

 that in this case the deflection of complement has been effected by a substance 

 in normal goat serum, as, for instance, another interbody of special affinity 

 or perhaps a normal anticomplement. Not every complement can be used 

 to reactivate a serum, for the complement may be deflected from the place of 

 its intended action by any interbody, provided merely that this possesses 

 sufficient affinity for the complement. It will be necessary to seek experi- 

 mentally for combinations in which such disturbing deflections are absent and 

 in which the difference in the affinity of the interbody which may be normally 

 present and of that produced artificially in large quantities becomes very mani- 

 fest. 



