BACTERICIDAL TEST-TUBE EXPERIMENTS. 



133 



In the following experiments the method is the same as that of 

 the above authors, to whose work I refer for these details. The 

 phenomenon of the deflection of complement can be exhibited in 

 two ways: First by employing as a source of complement an active, 

 in itself not bactericidal serum and showing that when decreasing 

 amounts of inactive immune serum are added, only the medium 

 amounts of the same exert a bactericidal effect, whereas both the 

 larger and the smaller amounts are ineffective. The results shown 

 in Table I will serve as an example of this. 



TABLE I. 1 



Control I. -glis cc. bouillon culture + 2 cc. 0.85% salt solution. Result, oo. 

 II. 0.4 cc. active pigeon serum +-^ cc. bouillon culture. Result, oo 

 " III. Sterility of all the sera, 0. 



The second method consists in employing a serum or serum mix- 

 ture which will kill the amount of bacteria employed. By adding 

 to this decreasing amounts of inactive immune serum (or, as a con- 

 trol, inactive normal serum) it is found that the immune serum, 

 in proportion to the amount added, exerts an antibactericidol effect, 

 whereas the normal serum fails entirely to do this or does so in a 

 very much less degree. This is illustrated by the results in Tables 

 III and IV, columns 1 and 2. 



In opposition to the explanation furnished by Neisser and 

 Wechsberg, according to which the deflection of complements is 

 caused by an excess of amboceptors, the following points have been 

 raised : 



A. The phenomenon is due to agglutination of the bacterial 



culture ; 



B. It is due to normal anticomplements (Metchnikoff). 



1 Each tube also receives three drops of buillon as in Xeisser and Wechs- 

 berg's exepriments. This applies also to the rest of our experiments. 



