Bacteria Developing at Different Temperatures 227 



arrived at exactly the same conclusions as had Matthews ten years 

 earlier. The peculiar advantage which such determinations have, is 

 that results may be obtained in 18 to 24 hours, or one to three days 

 earlier than is possible with plates incubated at room temperature. 



The significance of the numbers of bacteria developing on plates 

 incubated at temperatures above 40° C. has received very little 

 study, although the fact that such bacteria are more or less com- 

 mon has been known for some time. In 1888 Gl()big'° first dem- 

 onstrated that certain types of bacteria capable of growing at tem- 

 peratures of 50° C. or higher were common in the soil, although 

 Miquel had shown the existence of such bacteria some five years 

 before. In 1889-91 Miquel" found organisms of this class, which 

 he named thermophylic bacteria, to be prevalent in polluted waters, 

 but failed to detect them in spring waters. Macfadyen and Blaxall'^ 

 in 1894 demonstrated the presence of thermophylic bacteria in hu- 

 man feces, and Rabinowitschi'^ in the following year found this 

 type of bacteria to be present in the excrement from the majority 

 of domestic animals. More recently, Houston, ''^ 1898, has isolated 

 similar types from London sewage and from Thames water, and 

 has called attention to their significance in determining the extent 

 of pollution of a water. Tests for bacteria of this type included 

 in this paper were made on plates incubated at a uniform temper- 

 ature of 50° C. 



So far as has come to the knowledge of the writer, no comparative 

 study has ever been made of the numbers of bacteria in different 

 classes of waters which will develop on plates incubated at 30° C, 

 although it seemed reasonable to believe that some intermediate 

 temperature between 20° and 40° would prove favorable for the 

 growth of a large number of bacteria during a very short period 

 of incubation. 



Time of incubation. — The time allowed to elapse betw'een plating 

 and counting varies widely in different laboratories, depending upon 

 local conditions and the medium employed. Owing to the slow 

 development of bacteria at 20° C, it is impossible to obtain counts 

 in less than 48 hours on plates incubated at that temperature. 

 When gelatin is employed, it is usual to count plates on the second 

 or third day, the value of the determinations being obscured by 



