394 M. R. SMIRNOW 



disappearance of the indol tests in from forty-eight to seventy- 

 two hours in all the strains. 



The tests in these latter experiments were made in the cul- 

 ture tubes themselves, not transferring to the peptone solution, 

 after seven days of growth. In order to exclude the possibility 

 of interference with the indol test by the presence of the glucose, 

 several cultures in both plain broth, and peptone, were made, 

 and grown at 37°C. for seven days. Glucose was added to 

 each of the cultures and they were then tested for indol. Posi- 

 tive tests were obtained in all cases, excluding any possibility 

 of such interference by the presence of the carbohydrate. An 

 interesting observation may also be mentioned at this junc- 

 ture. Cultures of the organisms in plain broth of seven days 

 growth to which phenol or sodium chloride were added showed 

 a decided increase in the indol reaction in case of the phenol 

 and a diminished reaction in the tubes to which the sodium 

 chloride was added. In the interpretation of these tests compari- 

 son was made with controls. It may be possible that the 

 presence of these substances intensifies or diminishes the color 

 produced, the differences not being due to actual variations in 

 the amount of indol formed. The different culture media them- 

 selves were tested for indol, after incubating for seven days, 

 for the purpose of control and they were found negative. 



Experiments were then carried out to determine the per- 

 manency of this change. The cultures in glucose broth after 

 the 35th transfer were grown in plain broth, transplanting every 

 day and tested on the seventh day of incubation. Four of the 

 strains of B. coli; nos. 44, 45, 46 and 52 gave sHght indol reactions 

 on the third transfer, no. 46 gave a good positive on the fifth 

 transfer, but the others took from five to ten more transfers 

 before they could be called " + " or " + +" positive. Nos. 57 

 and 95 took six transfers before a trace of indol appeared. No. 

 19, a very feeble indol producer in the control, remained nega- 

 tive up to the fifteenth transfer at which time the experiment 

 was discontinued. 



Investigations as to the agglutinabihty of these altered strains 

 of B. coli were also made, but the work is too meagre and the 



