84 



Mr. E. C. Grey. Decomposition of 



of fatty material in the bacterial cells appears to be used up during the- 

 fermentation. 



It is to the attempt to understand the cause of this synthesis of fat or wax 

 (which was accompanied by a high yield of lactic acid) that the series of 

 experiments described in Part II owes its length. Attempts were made 

 to obtain a similar yield of the substance by change of the temperature 

 in various ways, and by alteration in the concentration of the salt. The 

 attempts were never as successful as in Experiment .3, but the variations 

 in the resulting proportions of the products led to the belief that the 

 fermentation by B. coli comnmnis really represented the sum of several 

 independent fermentations, which belief proved to be amply justified by the 

 results of the experiment which was devised to settle this question, and whiclr 

 is described in Part III. 



In Table I the results are arranged in order of decreasing yield of alcohoL 

 With the method of Martin, the estimation of alcohol is of a high order of 

 accuracy. In fig. 2 these results are recorded graphically. In fig. 3 the 

 same results are arranged from left to right, in order of increasing con- 

 centration of the potassium sulphate used in the experiments. It will be 

 seen that, with the exception of the relative positions of No. 3 and No. 5,. 

 the results grouped in this way give rise to much the same curves as when 

 arranged according to the yield of alcohol, from which it may be inferred 

 that the cause responsible for the variations in the proportions of the 

 products is the change in the concentration of the salt solution. The 

 manner in which the change in the concentration of the potassium sulphate 

 solution influences the proportion in which the products of the fermentation 

 appear at the end of the experiment was not clear until the experiments 

 described in Part III had been carried out, for, unfortunately, the bacteria 

 were not counted in the earlier experiments. It is now clear, however, 

 that in dilute solution there is a greater diminution in the number of 

 living bacteria, or a greater action of the dying or dead cells, than in 

 the cases where the concentration of potassium sulphate is greater, and, 

 as will be seen later, the death of the cells is accompanied by the 

 production of alcohol and those other products which are formed in 

 conjunction with it, while the production of lactic acid is more closely 

 associated with tlie rapid multiplication of the cells. Thus, at the one 

 end of the figures the results are to be correlated with rapid death' 

 of the bacteria introduced, and at the other with less death and more 

 inultiplication. 



JVote. — In fig. 3 Experiment 3 has been placed at the end of the series. 

 There was uncertainty about the conditions of this experiment. It, however. 



