558 REUBEN L. KAHN 



taining Endo medium were kept on the working table exposed to 

 direct light for close to two weeks without any appreciable 

 change in the butt; the slant, however, became red after twenty- 

 four hours. This is explained by the fact that air can not come 

 in direct contact with the medium in the butt; no oxidation of 

 the sulfite takes place, and the medium remains unchanged. 

 Endo tubes have been kept in this laboratory for four weeks in 

 the ice-box without any apparent deterioration. 



It might also be added that unless a sufficient number of 

 previous examinations of a given water have established the 

 absence of B. coli in it, we do not wait for gas production in 

 lactose broth before inoculating in Endo tubes. It was observed 

 again and again that if waters showing a high bacterial count 

 be inoculated in lactose broth fermentation tubes and after 

 about fifteen hours incubation — when there is no sign of gas 

 production — inoculations made in Endo tubes, typically metallic 

 colonies on the Endo slant with gas production in the butt often 

 develop simultaneously with gas in the original lactose fermen- 

 tation tubes. Thus, at the appearance of the "presumptive 

 test" we would have the "completed test" also. These findings 

 have led to the application of this procedure to all waters com- 

 ing into this laboratory in which the absence of B. coli has not 

 been established by previous tests. 



The question of the strength of agar to be employed in the 

 Endo tube media is of some importance. In a comparative 

 study of Endo media containing 1.5 per cent and 3 per cent 

 agar, no marked variations were observed in the amount of gas 

 produced in the butt. A strong gas producing organism would 

 usually produce much gas, while a weak one, would produce 

 only a few bubbles, in both cases. 



It was observed also that colonies of B. coli have, as a rule, 

 a stronger metallic film, when grown on Endo medium contain- 

 ing 1.5 per cent agar than 3 per cent agar. The difficulty with 

 the former, however, is the fact that the surface of the slant is 

 apt to be moist. This moisture may prevent the formation of 

 isolated colonies of B. coli — and such colonies are desirable for 

 the final isolation of these organisms. 



