214 ANTI-BACTERIAL PROPERTIES OF SERUM 



plement, and the complement fixed may be approximately 

 proportional to the amount of immune-body present. In 

 other cases, however, additional molecules of immune-body 

 do not fix additional complement. In the mode of union of 

 complement there is thus a wide difference from what 

 obtains in the case of enzyme action. 



The studies which we have carried out show that it is 

 quite impossible to consider the complement of the serum 

 as being made up of similar molecules. On the contrary, 

 variations both in the combining affinities and in toxic effects 

 are met with. In the case of the complements of the same 

 animal, Ehrlich and Morgenroth showed differences in 

 certain of their physical properties ; for example, resistance 

 to heat, capacity of passing through a porcelain filter, &c. 

 We have also demonstrated a striking case where a part of 

 the complement acts on sensitized guinea-pig's corpuscles, 

 but not on sensitized ox's corpuscles. Perhaps, however, the 

 most striking example is afforded in the case of absorption 

 of complement by bacteria, the bactericidal complement as 

 tested on one bacterium being absorbed by a great many 

 different bacteria before the haemolytic moiety is appre- 

 ciably affected. We have, however, pointed out that so 

 far as combining affinity is concerned the difference in this 

 last case is rather one of degree than of quality, as if a 

 sufficient amount of the bacteria be used the haemolytic 

 complement is absorbed also. It thus seems to be rather 

 an instance of preferential absorption, the molecules with 

 greatest affinity combining first. When powerful com- 

 plement absorbers are used, on the other hand, the com- 

 plements of the serum can be almost entirely removed. 



Whilst therefore certain differences in the complement 

 molecules of the same animal undoubtedly exist, these in 

 many instances merely represent gradations in combining 

 affinity. The haptophore group of complement does not at 

 all show the special affinity which exists in the case of 

 immune-body ; there is a certain community in its com- 



