THE TOXINS OF BACILLUS ANTHRACIS. 313 



and the alkaloid, pathogenic effects were produced in animals 

 closely similar to those produced by the bacilli themselves. 

 Martin adduced evidence to show that, of the symptoms of the 

 disease, the fever was mostly due to the albumoses, while the 

 oedema and congestion were mostly due to the alkaloid, which 

 acted as a local irritant. He showed that prolonged boiling 

 destroyed the activity of the albumoses, but not that of the 

 alkaloid. Further, from the body fluids of animals dead of 

 anthrax he isolated poisonous bodies similar to those produced 

 by the bacilli growing in this artificial medium. Hankin, in a 

 later research with Wesbrook, arrived at the conclusion that 

 the bacillus anthracis produces a ferment which, diffusing out 

 into the culture fluid, elaborates albumoses from the proteids 

 present in it. The bacilli also produce albumoses directly with- 

 out the intervention of a ferment. The albumoses produced in 

 the latter way, when injected in small doses, cause in susceptible 

 animals immunity against subsequent inoculation with virulent 

 bacilli, but are only toxic to animals not very susceptible to the 

 disease. Marmier, after cultivating the B. anthracis in peptone 

 solution containing certain salts, removed all the albumoses from 

 the resultant liquid, and from them, either by dialysis or extrac- 

 tion with glycerin, isolated a body which gave no reactions of 

 albuminoid matter, peptone, propeptone, or alkaloid. This he 

 considers the toxin. It killed animals susceptible to anthrax by 

 a sort of cachexia, and in suitably small doses could be used to 

 immunise them against subsequent inoculation with virulent 

 bacilli. It was chiefly retained within the bacilli when these 

 were growing in the most favourable conditions. Unlike the 

 toxins of tetanus and diphtheria, and unlike ferments, it was 

 not destroyed by heating to 110 C. The toxin produced by 

 the B. anthracis growing in a fluid medium remains intimately 

 associated with the bacterial protoplasm, as such cultures when 

 filtered are relatively non-toxic. 



From this account of the researches into the toxins of the 

 B. anthracis, it will be seen that our knowledge is far from com- 

 plete. It is difficult to say what interpretation is to be put 

 on the results of Hankin and Wesbrook. The researches of 

 Marmier rather indicate that, as is the case with the toxins of 

 other bacilli, the toxin of anthrax may belong to a group of non- 

 proteid bodies of whose chemical nature we are in complete 



