VARIETIES OF COLON BACILLI 179 



production of a greenish fluorescence in neutral red 

 broth as one of his tests of " typical " B. coli and ap- 

 parently found it valuable. 



Fermentation of various carbohydrate media has be- 

 come the most common method of subdividing the 

 bacilli of the colon group during the last few years, largely 

 as a result of the work of MacConkey. Sugar broths for 

 this test are generally put up in fermentation tubes of 

 some sort so that the gas formation may be observed 

 and jjerhaps measured, while acid production may be 

 indicated by the addition of litmus or accurately deter- 

 mined by titration. The old-fashioned fermentation 

 tube with a bulb and a stand has given way in most 

 water laboratories to a plain bent tube of even bore 

 and, more recently, to a still simpler device, a small vial 

 inverted in an ordinary test-tube of sugar broth. The 

 air in the top of the vial is driven out on sterilization 

 and the presence or absence of gas can be easily deter- 

 mined, although it is not possible to measure its quan- 

 tity with accuracy. If an ordinary bent tube is used 

 the amount of gas in the closed arm may be conven- 

 iently measured by the Frost gasometer (Frost, 1901). 

 If a measurement of the gas ratio is desired a few centi- 

 meters of strong sodium or potassium hydrate are added 

 and mixed with the broth by cautiously tipping the 

 tube; a second measurement determines the amount of 

 gas absorbed (assumed to be CO2). 



It has been pointed out in Chapter VI that the gas 

 ratio appears to be a reaction of slight importance as 

 thus determined. 



The list of fermentable substances used by various 



