NATURE OF THE ACTIVE PRINCIPLES 313 



It ic(]iiirc.s apparently a smaller amount ot serum in order to 

 neutralize the lethal lactors than the active principles of the 

 phenomenon. This is based on experiments described elsewhere 

 (p. 251). 



Stolyh^vo agrees ^vith my previous observations that there is a 

 suggestive parallelism between the ability of the filtrates to pro- 

 duce the phenomenon and their general toxicity. Thus, in his 

 experiments the "agar washings" filtrates, highly active in the 

 elicitation of the phenomenon, -were able to produce symptoms 

 of general intoxication in very small doses. Filtrates ^\■hich lost 

 their phenomenon-producing potency after prolonged storage 

 ^vere also deprived of lethal effect. It is interesting that on the 

 other hand, the refractoriness of rabbits to the phenomenon did 

 not parallel their resistance to the general intoxication. 



The materials employed by Kielanowski and Selzer (filtrates 

 of seven day old broth cultures of B. coli) possessed on intra- 

 venous injection a general toxicity similar to that described 

 abo\e. The animals sho^ved a strong dyspnea, unrest, accompanied 

 by profuse urination and diarrhea. The symptoms disappeared 

 spontaneously and did not lead to the death of the animals. As 

 ])ointed out, broth cultures are of considerably lower general 

 toxicity. These authors found the same symptomatology in nor- 

 mal animals as ^vell as in those prepared by an intradermal injec- 

 tion of an active bacterial filtrate. 



In many unpublished observations I gained the impression that 

 the de\elopment of a strong skin reaction protected the rabbits 

 to a slight degree against the lethal effect of highly toxic prepara- 

 tions. The obvious interpretation is that the prepared skin site 

 removed some of the lethal factors from the general circulation. 



P. Bordet (1933) concentrated B. coli culture filtrates by the 

 folloAving" method: 



T^vo or three per cent of acetic acid was added to filtrates of 

 eight day old broth cultures of B. coli. The acidulated filtrates 

 \vere incubated for one to two days at 37- c. The precipitate 

 Avhich ^vas easily separable by centrifugalization \vas alkalinized 

 and made up to 0.1 of the original \olume in normal saline, 

 washed several times and then suspended in the desired volume 

 of normal saline. The intravenous injection of 1 to 2 c.c. of a 

 filtrate concentrated in this manner produced lesions in the respir- 

 atory system even without the preparatory injection. TAventy- 

 four hours after the injec tion there were observed deep breathing, 



