3/0 JJEHAVIOUR OF nACTKKlA. 



examination. At dilutions of i ]>er 100,000 the colonies on 

 ordinary agar were too numerous to be counted; on o. 1 per cent, 

 arsenite agar 44 colonies developed, and on 0.3 per cent, arsenite 

 agar (j colonies appeared. At dilutions of i per 10,000,000, 20 

 colonies appeared on ordinary agar and 2 colonies on o.i per 

 cent, arsenite agar, while the 0.3 ])er cent, arsenite plate remained 

 clean. Averaging the number for nine plates, it was a])parent 

 that about 90 per cent, of the total flora capable of growing on 

 agar were suppressed by arsenite in low concentration. About 

 10 ])er cent. ])roved resistant to o.i ])er cent. As^Og as arsenite, 

 and about 1 ]ier cent, were capable of tolerating 0.3 i)er cent, 

 or more. The intolerant forms included members of the coli 

 grou]). while the most highly resistant form belonged to the 

 putidiun family. 



Cattle Dip. — Seven-day "laboratory strength," containing 

 o.t6 per cent. As^Og as arsenite. Two series of platings were 

 carried out, one about a week after making uj) the tank afresh, 

 and a few days after the first cattle di])])ing, the other about a 

 month later, when the arsenic resistant flora had established 

 itself. In the first series the bacterial count was naturally k^w, 

 and amounted only to about 4,000,000 per cubic centimetre. 

 The count on 0.2 \wr cent, arsenite agar was still lower, thus 

 suggesting that certain organisms were capable of surviving in 

 presence of arsenite without being able to nniltiply freely. In 

 the second ])lating of older di]), the bacterial numbers had risen 

 to about 100,000.000 per cubic centimetre, and all forms were 

 arsenic resistant. Apparently the non-resistant forms, introduced 

 in the faeces of the dipped stock, had died out, leaving the tolerant 

 forms to develo]:) and dominate the tank. Seven different s])ecies 

 were isolated, of which the most frec(uent belonged to the 

 putidum group, and the least frequent to the group represented 

 by the earlier described B. arsenrcduccus. A few organisms 

 isolated in the course of the work were then tested in more 

 detail, adding four pathogenic members of the coli group, pro- 

 ciu-ed from the Director of Medical Research in Johannesburg. 



( I ) Sithtilis (iroiif^: B. subtilis, B. mcsentericiis, B. inycoides, 

 and B. aiithracls. were examined. ;\.ll proved sensitive to 

 arsenite. ])eing inhibited by concentrations of less than 0.05 per 

 cent, in broth media', and altogether suppressed by concentrations 

 above this. Somewhat more tolerant of arsenate, although inhi- 

 bited by 0.2 per cent. 



(2) Colon-typhoid group: The members tested included 

 B. coli, B. cntcriditis, B. paratyphosus A, B. Paratyphosus B, 

 B. typhosus, and B. alcaligenes, i.e., representatives from the 

 species capable of fermenting several sugars, through those which 

 produce acid without fermentation, to those which produce 

 neither acid nor gas. 



The limit of tolerance was found to be very low, ranging 

 only from 0.05 per cent, to 0.1 per cent AsgO. as arsenite 

 bouillon, although there was the suggestion that B. alcaligenes 



