THE COLON GROUP OF BACILLI 109 



specific cultures. It appeared that a rather definite 

 proportion of tubes showing a characteristic fermenta- 

 tion proved on further examination to contain bacilli 

 of the colon group; and it was therefore suggested 

 that the dextrose broth test alone might be used as a 

 rapid " presumptive " test. The underlying principle 

 of this method is that B. coli develops rapidly* in dex- 

 trose broth with gas formation of from 25 to 70 per 

 cent of the capacity of the closed arm of the fermenta- 

 tion tube. Of this gas approximately one-third is 

 carbon dioxide and two-thirds hydrogen, that is, as the 



gas formula is generally expressed, = . 



C(J2 I 



In testing a water by this method a series of samples, 

 in suitable dilution, .001, .01, .1, i.o, or 10 c.c., are added 

 directly to the dextrose-broth tubes and incubated for 

 24 hours at 37. 



On measurement of the gas, if the results above given 

 are obtained, the reaction is considered typical. If the 

 amount of gas is between 10 and 25 percent or more than 

 70 per cent, or the percentage of carbon dioxide is greater 

 than 40, the reaction is considered atypical. If no gas 

 forms, or less than 10 per cent, the test is called negative. 



In recent years. Irons (Irons, 1901) was perhaps the 

 first to call attention to the value of this method, 

 stating that " when the dextrose tube yields approx- 

 imately 33 per cent of CO2, Bacillus coli communis is 

 almost invariably present." In the next year the 

 reliability of the fermentation test as an indication of 

 B. coli was worked out by Gage (Gage, 1902) as given 

 in the table on p. no: 



