A Study of the B. coli Group. 
321 
Criteria Employed for Differentiation of Types. 
In classifying coliforin bacilli into particular types MacConkey's 
criteria were for the most part adopted, and for all practical purposes 
proved sufficiently complete. Kaffinose and salicin fermentations were, 
however, found to yield further differentiation, and these tests were therefore 
included among the criteria adopted. 
Certain other fermentable substances which have been employed by 
different workers proved of little or no differential value ; thus maltose was 
found to be fermented by all the lactose-fermenting strains ; glucose- 
fermenting, non-lactose-fermenting strains, however, vary m their action on 
this sugar. The same was found to be true for mannite. Laevulose and 
galactose were with few exceptions fermented by all the glucose-fermenting 
strains and glycerin by all the lactose-fermenters. Of course, as is well 
known, all lactose fermenters are also glucose fermenters. The behaviour of 
these bacilli in raffinose in the majority of cases corresponded to their effect 
on saccharose (as shown also by Winslow and Walker) , but a certain pro- 
portion exhibited differences in their effects on these sugars. 
The action of the lactose fermenters on dextrine differs from the 
fermentation produced in other carbohydrates ; in the case of simple peptone 
water media containing dextrine and neutral red (as indicator), it w^as found 
that after twenty -four hours' incubation the neutral red became of a bright 
yellow colour and there was some degree of gas production ; the gas formed 
w^as analysed and found to consist mainly of hydrogen. The medium 
remained neutral. This effect was common to all the lactose-fermenting types 
and the test afforded no information as regards differentiation of v^arious 
strains. 
With one exception all the strains that fermented inosite also fermented 
adonite. In all cases the Voges and Proskauer reaction was only given by 
inosite-fermenting strains. 
It was found on repeated testing of various strains that the presence (or 
absence) of motility was a definite and constant character. 
Similarly the presence (or absence) of indol in peptone water cultures 
(after ten days at 37° C.) proved to be a stable property of these organisms. 
It was concluded therefore that of the various tests used by different 
workers, the most complete set of criteria for the differentiation of B. coli 
types were : (1) Presence or absence of motility ; (2) production of indol ; (3) 
the Voges and Proskauer reaction ; the fermentation of (4) glucose, (5) 
lactose, (6) dulcite, (7) saccharose, (8) adonite, (9) inulin, (10) inosite, (11) 
raffinose, (12) salicin. 
The majority of the strains to be described were re-tested after about two 
or three months, and apart from the variations noted (p. 360) their characters 
proved markedly stable. 
