BACTERICIDAL TEST-TUBE EXPERIMENTS. 



139 



IV. 



on a Plate on the Addition of the Inactive Goat Sera here mentioned. 



Control III. Active guinea-pig serum 0.1 cc. +0.01 cc. inactive goat immune 

 serum against vibrio Metchnikoff + ^-s cc. bouillon culture. 

 IV. Sterility of all the sera = 0. 



erysipelas of swine, hog cholera, dysentery. I am unable to say 

 what the cause of these contradictory results may be. One thing, 

 however, is shown thereby, namely, that there is no general law such 

 as Gruber assumes, and that, their toxicity taken for granted, his 

 experiments constitute rather a rare exception which must even be 

 regarded as an unfortunate coincidence. 



That immunization with any kind of bacteria does not cause the 

 formation of anticomplements with a general antibactericidal action, 

 in Gruber's conception is seen by glancing at Table IV, columns 2, 

 3, and 4. In these experiments only the Metchnikoff immune serum 

 exerted a complement-deflecting action, not, however, the immune 

 sera of two other goats immunized with vibrio Nordhafen, and with 

 staphylococcus pyogenes aureus. 



It is very easy to prove that in the Metchnikoff immune serum 

 the active factor which effects this anticomplementary action and 

 which develops as a result of the immunization is really an ambo- 

 ceptor; for by previously adding the corresponding dead bacteria 

 to the immune serum and later centrifuging, the amboceptor of 

 the serum is abstracted. It can then be shown that this amboceptor- 

 free immune serum has lost all its power to deflect complement, 

 provided, of course, a sufficient amount of bacteria was used. It 

 can further be shown in this way that the action proceeds quanti- 

 tatively. Thus if decreasing amounts of bacteria are employed 



