ELIMINATION OF SPURIOUS TESTS FOR B. COLI 345 



positive and thin Gram negative sporulating rods. An attempt 

 at repeated demonstration of the non resistance of this particu- 

 lar culture to gentian violet failed, but this and other observa- 

 tions show that exceptional strains of hay bacillus may be found 

 which are not inhibited completely by gentian violet. 



In the unheated samples there were 11 in which B. coli was 

 isolated from both the dye and the non-dye tests. In the case 

 of one of these in gentian violet lactose broth 2 cultures, appar- 

 ently B. cloacae, were isolated from well separated acid colonies 

 on litmus lactose agar. One of these was subsequently sepa- 

 rated into 2 subcultures having the properties of B. fluorescens- 

 liquefaciens and B. coli respectively; the other after a delayed 

 period was found to have lost the property of B. coli, i.e., lac- 

 tose fermentation, formerly possessed, and retained only the 

 characteristics of B. fluorescens-liquefacieris. Our interpretation 

 of this phenomenon has already been discussed. 



Four unheated samples yielded B. coli from the dye test and 

 not from the standard test. In 2 of these the plates were clearly 

 overgrown by spores which we feel so masked the colonies and 

 acid produced by B. coli that we could not pick out the col- 

 onies; in the other 2 no acid was displayed in either the first or 

 second tests, and the picked colonies proved not to be jthe colon 

 bacillus. 



Two unheated samples yielded B. coli from the standard test 

 and not from the dye test. In one of them the presumptive 

 test was positive only after three days, and the non-acid col- 

 onies picked from the plate failed to ferment lactose; the second 

 presumptive test was negative for five days. In the other case 

 a culture tentatively called B. cloacae was isolated; subsequently 

 examined with a view to the separation of the unheated B. coli 

 from the gelain liquefier, it was found that the culture pre- 

 sented only the characteristics of B. proteus. 



Five unheated samples failed to yield B. coli from either the 

 standard or the dye test. In 4 of these the presumptive test 

 with dye was negative during five day's incubation; the fifth 

 showed a bare trace of gas on the fifth day only. In the stand- 

 ard test all showed gas, 1 on the second day, 2 on the third, 1 on 



