338 IVAN C. HALL AND LILLIAN J. ELLEFSON 



and acid-forming colonies isolated upon the unused half of the 

 plate. After all but 22 samples had been examined we began 

 to realize that there are possibly two conditions in which the 

 positive presumptive test may truly be due to B. coli and yet 

 great difficulty be encountered in the isolation of the proper 

 organisms due to the fact that no red colonies appear on the 

 litmus lactose agar plate. 



The first of these is the occurrence of so-called "attenuated" 

 coliform bacilli which may be considered to have temporarily or 

 partially lost their property of fermenting lactose under aerobic 

 conditions. Many water analysts are well aware of this occur- 

 rence and so the new standard methods for 1917 emphasize the 

 necessity of testing suspicious looking surface colonies of true 

 morphology in lactose broth for gas formation, even though they 

 do not produce acid upon the litmus lactose agar plate. But 

 it is to a degree only accidental that this procedure sometimes 

 results in identification of B. coli; in our experience blue surface 

 colonies usually, but not invariably, fail to form gas in the fer- 

 mentation tube. Our work holds no suggestion as to the elimi- 

 nation of this difficulty. 



The second condition is that in which strongly proteolytic 

 organisms, e.g., certain of the hay bacillus group, not ferment- 

 ing lactose, liberate sufficient alkali to neutralize any acid due 

 to B. coli in the plates, thus suppressing the appearance of 

 red colonies except when widely separated; a further aggrava- 

 tion is that these organisms sometimes have marked spreading 

 proclivities. Fortunately the use of gentian violet is quite 

 efficacious in preventing this particular source of trouble, since 

 most aerobic sporulating bacteria are inhibited as well as the 

 Gram-positive sporulating anaerobes. It was quickly deter- 

 mined that the addition of 1-100,000 gentian violet to litmus 

 lactose agar served admirably to inhibit the undesirable organ- 

 isms, not always completely, but at any rate sufficiently to 

 prevent their spreading and interfering with the display of acid 

 by B. coli. The color imparted by this amount of dye is quite 

 insufficient to mask the desired color changes in the litmus. 

 In advocating the use of gentian violet in this manner for the 



