NITROGEN METABOLISM OF BACTERIA 393 



as in the preceding experiment. The incubation periods varied 

 from one to twenty-seven days, a constant temperature of 

 30°C. being employed. The results are not given here in tabu- 

 lated form,^ but may be summed up briefly as follows. 



The titratable acidity was at no time high. Towards the 

 end of the experiment the reaction in the glucose and lactose 

 tubes was practically the same as that of the uninoculated con- 

 trols, 10, as compared with 28 and 30, the highest points of 

 acidity. The Sorensen titration increased simultaneously in all 

 of the inoculated media, and in practically the same degree. 

 The biuret tests indicate the same metabolic changes as the 

 Sorensen titration. The peptone was completely utilized in each 

 of the five media before three weeks of incubation had elapsed. 

 The slow rate at which the sugars were fermented is noteworthy. 

 The first evidence of disappearance of the sugars, even in the 

 tubes originally containing only 0.2 per cent of the carbohy- 

 drates, occurred after two weeks, while twenty-five days were 

 required for their complete disappearance from the tubes origi- 

 nally containing 0.5 per cent. 



A similar study was made with Bacillus coli and Proteus vul- 

 garis. The details of the experiment were the same as in the 

 previous experiment, except that with these two organisms two 

 additional tests were conducted, ilamely the indol and methyl 

 red tests. The Salkowski method of determining the presence 

 of indol was employed. The test was made by adding 0.25 cc. 

 of concentrated sulphuric acid to a few cubic centimeters of the 

 fluid under observation, immediately cooling the mixture in cold 

 water, and slowly adding a 0.5 per cent solution of sodium nitrite 

 while the tube was held in an almost horizontal position. This 

 method permits of a minimum amount of diffusion, so that the 

 color is made all the more apparent in the case of a positive 

 test. Difficulty in procuring the Ehrlich aldehyde prevented 

 the use of this agent until near the close of the investigation. A 

 comparative study of the two methods has shown them to be 

 equally reliable in our own hands. 



1 The original tables may be consulted in the doctorate thesis (Berman), a 

 copy of which is in the possession of each of the writers and of the Yale Uni- 

 versity Graduate School. 



