IX 



strong, medium or weak acid or alkali formers, will pro- 

 duce the colour change. The latter is not dependent only 

 upon a change of reaction. If this were so, the strong 

 acid former of the coli group would turn the colour 

 red or magenta before the weak acid former, which is 

 not found to be the case, as 51b produced the colour 

 change at a later date than KE X in spite of this (KEj) 

 forming in addition to the acid a considerable amount 

 of alkali; on the other hand it appeared that in the ty- 

 phoid group the strongest alkali producer was the first 

 to give the colour reaction. 



Yet by comparing this result with the changes cau- 

 sed by the coli variety KE : , which, as before stated, 

 produced alkali to the same extent as B. typhosus, but 

 nevertheless changed the colour to magenta, we cannot 

 regard the reaction as due to the quantity of alkali pro- 

 duced. Still further evidence is supplied by the fact, 

 that although in the final results, a red and a yellov 

 colour of the medium will signify an acid and an 

 alkaline reaction respectively, in the intermediate stages 

 of colour change produced by the typhoid group, a 

 varying shade of red may occur in the originally neu- 

 tral or alkaline media, but yet the reaction be alkaline. 

 Whereas with the coli group, in the originally neutral 

 and alkaline media the reaction is acid, though the co- 

 lour in the intermediate stages may be a varying shade 

 of yellow, as shown by 51b. (In the Tables + and + 

 signify acid and alkaline reaction respectively). 



It is at present quite impossible to know how to 

 account for the difference in the power possessed by 

 B. typhosus and B. coli to produce colour change in 

 3% lactose neutral red bouillon. The experiments we 

 have made in this direction not having yielded suffi- 

 ciently definite results, we can only say that the occur- 



- 36 



