2 28 Stephen DeM. Gage 



liquefaction when longer periods of incubation are followed. It 

 is customary to incubate agar plates three to four days, although 

 in some laboratories even longer periods are allowed to elapse before 

 the plates are counted. The recommendation of the Committee 

 on Standard Methods, that a uniform period of incubation of 48 

 hours be employed, has been quite generally adopted by users of 

 gelatin, but has found less favor among users of agar. At Lawrence 

 it has always been the custom to allow four days to elapse between 

 planting and counting. 



With the use of higher temperatures, and the more rapid growth 

 of bacteria at such temperatures, it becomes possible to adopt a 

 shorter period of incubation. Since it is extremely desirable that 

 the results of bacterial determinations be available at the earliest 

 possible moment, the writer has adopted a uniform time of incu- 

 bation of 24 hours on all plates grown at temperatures higher than 

 20° C. ; consequently the numbers of bacteria included in the dis- 

 cussion and tables beyond, unless otherwise stated, were obtained 

 on plates incubated four days at 20° C, and 24 hours at 30°, 40°, 

 and 50° C, respectively. 



The use of ratios between different bacterial counts. — In comparing 

 the analytical results obtained from dififerent waters, or from different 

 samples of the same water, it is quite usual to express the results 

 of such comparison as the ratio, or per cent, which one is of another. 

 The mathematical expression of the ratios between the results of 

 different determinations is less common, and has hitherto been 

 confined to the chemical side of the analysis, so far as the writer 

 knows, no use ever having been made of the ratios between the 

 results of different bacteriological determinations on the same sam- 

 ples. 



In the present study we have the results of bacterial counts on 

 agar and on litmus-lactose agar plates which have been grown at 

 four different temperatures, and upon which the number of bacterial 

 colonies have been counted daily until the maximum number of 

 colonies has developed. In addition to the determined numbers of 

 bacteria on each of the plates, and the numbers of bacteria which 

 are able to cause acid fermentation of lactose, both of which determi- 

 nations may have a place in the interpretation of the character of a 



