216 ANTI-BACTERIAL PROPERTIES OF SERUM 



different animals, the amount which has to be taken up 

 in order to produce lysis varying in different instances. 

 The striking difference in the relative toxicity of guinea- 

 pig's and rabbit's complement, when tested both on rabbit's 

 and guinea-pig's corpuscles, is a striking example of this 

 (p. 74). Even more striking, however, is the relative 

 toxicity of complement when tested on different bacteria ; 

 many organisms, as already pointed out, combining with 

 bactericidal complement, without any bactericidal action 

 resulting. In the case of an anti-bacterial serum it is accor- 

 dingly not only essential that the complement be brought 

 into combination, or, as it is often expressed, that the com- 

 plement should suit the immune-body, but also that it 

 should have the power to produce the necessary destructive 

 effect. We have repeatedly insisted on the importance of 

 studying the combination of the three elements concerned, 

 viz. receptors, immune-bodies and complements by the 

 absorption method. The mere occurrence of haemolysis, for 

 example, gives no indication of the total combining affinities. 

 Many times the amount both of immune-body and of com- 

 plement necessary for lysis may be taken up, and, on the 

 other hand, the non-occurrence of bactericidal effect, or 

 even of haemolysis, is no indication that complement 

 has not entered into combination. It is only by study- 

 ing the alteration in the properties of the fluid in which 

 the particular reaction has taken place that we can judge of 

 the substances which have been absorbed, and thus be able 

 to understand the behaviour of the complement molecules. 



