282 INFECTION AND IMMUNITY. 



but which contained no precipitin. Moreschi pro- 

 duced in fowls antiserum which was high in 

 precipitin concentration but did not produce com- 

 plement fixation. In like manner the presence of 

 antibody complement deviation in the absence of 

 agglutinating properties has been noted. In gen- 

 eral, complement deviation antibody is destroyed 

 at higher temperatures than agglutinins. 



It is in lytic amboceptors that complement fixa- 

 tion antibody has its closest analogy and opinions 

 are divided as to the identity of the two. Neufeld 

 and Haendel regard amboceptor and complement 

 fixation antibody as distinct from each other and 

 apply the name "Bordet's antibody" to the latter. 

 They reach this conclusion from the fact that if 

 cholera bacilli and their antiserum are mixed with 

 complement and allowed to act at C., hemolytic 

 complement is absorbed but not that necessary for 

 bacteriolysis. The same mixture allowed to act at 

 37 C. results in an inhibition of action of both 

 complements. The objection to such a proof lies 

 in the fact that the difference may be due to a 

 difference in the complements according to Ehrlich 

 and his school, who believe in the multiplicity of 

 complements. Neufeld and Haendel have also 

 immunized animals with a certain water vibrio 

 and obtained a serum which was bacteriolytic 

 against this vibrio, but not against cholera vibrios. 

 Complement deviation, on the other hand, was 

 effected by using either of the two micro-organisms 

 as antigen. 



If Ehrlich's definition of amboceptor as a sub- 

 stance which unites antigen with complement is 

 accepted, it is apparent that the term will apply 



