IX 



rencc of the reaction seems to depend partly upon a 

 formation of alkali or of acid and partly upon a direct 

 decomposition of the neutral red itself, since as a rule 

 it is not possible to regain the original red or yellow 

 colour either by the addition of acid or alkali or by 

 oxidation. 



Conclusions. 



(1). The difference in the power of IB. typhosus and IB. 

 coli to produce the respective colour changes in 

 neutral red is a quantitative and not a qualitative 

 one. 



(2). If any reliance at all can be placed upon the neu- 

 tral red method, as proposed by former investigators 

 the following points must be borne in mind. 



(a) IB. typhosus can within certain limits produce 

 similar colour changes as B. coli. 



(b) The results obtained must be supplemented with 

 other tests. 



(c) Glucose free bouillon is preferable to agar. The 

 media must be exactly titrated, as a certain 

 quantity of free acid will prevent the occurrence 

 of the reaction. 



(d) Media containing Va/o glucose or more, must 

 be avoided. 



(e) Even under the above conditions, it is not pos- 

 sible to use the method for distinguishing mem- 

 bers of the typhoid group from those of the 

 coli group. 



(3). For the differentiation of IB. typhosus and B. coli 

 by means of neutral red, the best medium to employ 

 is 3% lactose bouillon to which 0.5% of 1 % wa- 

 tery neutral red (Griibler) has been added. 

 Within 4 6 days B. typhosus will produce a decided 



37 - 



