REACTING POTENCY 285 



seniin exceeded twenty-lour lioiirs.^ The followino two explana- 

 tions of the results of these experiments suggest themsehes. 



A. — A local state of reactivity may l^e elicited by the interac- 

 tion of the locally injected horse serum with the circulating anti- 

 horse antibodies of the sensitized rabbits. The reactive site thus 

 obtained may then react ^vith the bacterial filtrate injected in- 

 travenously after a suitable period of time. 



B. — It is also possible, however, that the preparatory factors 

 of the bacterial filtrate injected intravenotisly localize at the site 

 of the intradermal injection of horse serum in the sensitized 

 rabbits and then, in turn, induce at this site a state of reactivity 

 to the interaction of locally injected horse sertmi with actively 

 accpnred anti-horse antibodies. As mentioned above, potent bac- 

 terial filtrates may produce a preparatory effect by ^vay of the 

 vascular system provided some local change raises the perme- 

 ability of the blood vessels, thus allowing" their passage into the 

 j:)erivascidar tissue. It is significant in this connection that the 

 incubation period necessary to induce the state of reactivity in 

 this manner may be less than tAvo hotns. It is possible, there- 

 fore, that the inflammatory reaction produced by the intrader- 

 mal injection of the horse serum into sensitized rabbits may per- 

 mit a localization of the sort with a subsequent elicitation of the 

 state of reactivity to the reacting factors residting from the anti- 

 gen-[-antibody interaction. In order to elucidate the mechanism 

 of the reactions obtained in experiments just described, advan- 

 tage was taken of the previous observations that the skin-pre- 

 j)aratory potency of certain bacterial filtrates is not necessarily 

 j)arallel to its reacting" potency. Thtis, quite frequently menin- 

 gococctis cidttne filtrates may possess a high reacting potency Avith 

 a low preparatory potency, whilst B. typhosus cidtiue filtrates 

 may have a skin-preparatory potency considerably higher than 

 that of meningococctis with the reacting potency lower than that 

 of meningococcus. 



With this consideration in mind, rabbits Avere sensitized by an 

 intravenous injection of 1 c.c. of horse serum. The period of 



' rhcie was no eftect of baclfiial liliialcs injected intravasculaiiv upon reactions 

 to passively acquired antigen+antibody complexes. 



The subcutaneous treatment of rabbits with bacterial filtrates one hour prior 

 to the intradermal test with horse serum in rabbits sensitized to horse serum 

 remained without effect: and also the period of sensitization with horse serum 

 shorter than (> (ia\s pro\cd insnliicient. 



