396 L. D. BUSHNELL 



in which E = probable error, S = sum of residuals, N = number 

 of parallel determinations. The probable error as determined 

 for the first series is ±4.78, and for the second series is ±3.07. 



The probable error was not determined for ammonia, amino- 

 acid, or volatile-acid production, but dupUcate determinations 

 usually varied about as much as in the gas determinations. 



Nature of fermentation. It will be noted that this organism 

 produced considerable amounts of gas from peptone solution. 

 This probably accounted for the presence of bubbles of gas in 

 fermentation tubes containing non-fermentable carbohydrates. 

 The higher negative tension and the larger amounts of media in 

 the bottles tended to magnify the gas production from this 

 medium in bottles, as compared to that in tubes in the anaerobic 

 jar. (The latter usually showed a tension of about 350 mm. of 

 mercury.) 



When we mention the fact that an organism ferments a certain 

 carbohydrate we usually refer to the fact that the change results 

 in the production of gas or acid. These are products easilj' 

 determined and are evidently due to deep-seated changes in the 

 product. It is quite probable, however, that there may be or- 

 ganisms which bring about but shght changes in a carbohydrate, 

 not extensive enough to lead to the formation of acid or even gas. 

 It is also possible that these products may be produced and again 

 utilized in the synthesis of cell protoplasm, or changed to other 

 products of fermentation . It is a well known fact that the nitrif j'- 

 ing bacteria utilize CO2 as a source of carbon in the oxidation of 

 ammonia and nitrites to nitrates. Nikitinsky (1907) showed that 

 there were certain anaerobes which were able to utilize hydrogen. 

 For this purpose he used "konzentrierten kanalisationsflussig- 

 keit" and "Schlamm aus einem Absitzbecken" placed in an atmos- 

 phere of hydrogen. In one case he found that 500 grams of 

 "Schlamm" was able to combine with an average of 30 cc. or a 

 maximum of 70 cc. of hydrogen per day. 



It is a well known fact that many bacteria can utilize acids and 

 alcohols as a source of energy. It may also be possible that 

 bacteria utilize only a minimum amount of the carbohj'drate, 

 enough to establish their initial development, but do not produce 



