226 STEPHEN DEM. GAGE 



Methods of Water Analysis, 6 except that it contains only i per cent of 

 agar, instead of i . 5 per cent as recommended by that committee. 

 Litmus-lactose agar as a substitute for gelatin or agar in determining 

 the numbers of bacteria is slowly gaining a foothold in many labora- 

 tories, its advantage being that it permits a distinction to be made 

 between the types of bacteria which do and do not produce acid fer- 

 mentation of lactose, the determinations of the presence and numbers 

 of the fermenting organisms being of considerable significance, as will 

 be shown further on. 



Temperature oj incubation. The determination of numbers of 

 bacteria on plates which have been incubated at 20 C. or at room 

 temperature is the usual practice. In many laboratories, where gela- 

 tin is employed as the routine medium, the plates are incubated at a 

 somewhat lower temperature, usually about 15 C., in order to prevent 

 the rapid growth of the liquefying bacteria and to minimize the errors 

 due to such liquefaction. In other laboratories, where agar is 

 employed, it is the custom to incubate the plates at temperatures of 

 23 to 26 C., in the endeavor to obtain higher counts with shorter 

 periods of incubation. The recommendation of the Committee on 

 Standard Methods, that a uniform temperature of 20 C. be employed, 

 has become quite generally adopted in American practice. It is, of 

 course, unnecessary to enter into a consideration of the significance of 

 the numbers of bacteria determined by this procedure at this time. 



The value of a determination of the numbers of bacteria which 

 are able to develop at body temperature was first advanced in 1892 

 by Wurtz. 7 In 1893 Matthews 8 quite clearly demonstrated the dis- 

 tinction in the bacterial content of different classes of polluted and 

 non-polluted waters by the use of lactose agar plates incubated at 

 body temperature, thus confirming the deductions of Wurtz. Routine 

 determinations of the number of bacteria producing acid fermenta- 

 tion of lactose (colon type) on plates incubated at 38 to 40 C. have 

 been made on certain classes of waters since 1896 at Lawrence, and 

 a similar procedure has been adopted by a large number of water bac- 

 teriologists. The value of counts of the total number of bacteria 

 developing at 38 to 40 C., however, appears to have been lost sight 

 of in the rush of bacteriologists to study the acid-fermenting types, 

 until Winslow and Niebecker 9 in 1903, after extensive investigations, 



