346 



IVAN C. HALL AND LILLIAN J. ELLEFSON 



the fourth and 1 on the fifth. In no case were acid colonies 

 developed on the plates. All were due to sporulating anaerobes 

 as shown by the corresponding tests with the heated fractions 

 of these samples. All these samples had originally contained 

 B. coli according to the Bureau of Sanitary Engineering though 

 in no case in less than 10 cc. 



At this stage in the investigation we were ready to adopt 

 the use of 1-100,000 gentian violet in all of the litmus lactose 

 agar plates as a routine procedure in the isolation of B. coli 

 from positive presumptive tests, upon the basis of experimental 

 trials with pure cultures of B. coli and the hay bacillus. The 



TABLE 5 



Tests for B. coli and sporulating gas-forming anaerobes in 23 samples of water — 



using 1-20,000 gentian violet in the presumptive test and 1^-100,000 



gentian violet in the litmus lactose agar plates 



* Not the same samnle. 



considerations involved in this change of method have already 

 been discussed fully under the heading of ''technic." Twenty- 

 two new samples were tested according to this plan as shown in 

 table 5. 



At least 18, or over 81 per cent, of these samples contained 

 anaerobic sporulating gas formers, as judged by the positive 

 tests in the standard presumptive inoculated with heated 

 samples. In 1 case B. coli escaped the heating and was isolated 

 and identified as the gas former; in 3 others. Gram-positive- 

 aerobic sporulating gas formers belonging to the hay bacillus 

 group were isolated. The recognition of members of this group 



