j motte ¥ (p41 
On Variation and Adaptation in Bacteria, Illustrated by 
Observations upon Streptococci, with Special Reference to 
the Value of Fermentation Tests as Applied to these 
— Organisms. | 
By E. W. Arntey Waker, M.A., D.M., Fellow and Tutor of University 
College, and Lecturer in Pathology in the University of Oxford. 
(Communicated by Prof. Francis Gotch, F.R.S. Received February 18,—Read 
March 30, 1911.) 
(From the Department of Pathology, University of Oxford.) 
The fermentation tests introduced by M. H. Gordon(1) in his study of 
streptococci isolated from saliva and some other sources were proposed by 
him as a means of differentiating among these organisms and separating 
them into definite varieties. 
Gordon’s methods were applied by A. C. Houston (2) in observations upon 
‘streptococci obtained from normal human feces and from milk, and they 
were largely relied on by Andrewes and Horder (3) in their investigation of 
pathogenetic streptococci from disease in man. 
The tests in question concern the reaction (acid or alkaline) produced by 
any given streptococcus during three days’ growth at 37° C. in a series of 
tubes of alkaline sugar-free litmus broth to which 1 per cent. of saccharose, 
lactose, raffinose, inulin, salicin, coniferin, and mannite respectively have been 
added. The production of acid in sufficient amount to change the colour of 
the medium from blue to red constitutes a positive reaction in each case, 
and is believed to be due to the fermentation of the added carbohydrate 
substance by the streptococci. 
To these culture tests in the seven different carbohydrates two others were 
added by Gordon, namely, the presence or absence of clotting in litmus milk 
within three days at 37° C., and the reduction or non-reduction of neutral 
red broth in anaérobic culture at 37° C. within two days. 
In Houston’s observations the coniferin test was omitted, and so far as 
I am aware this glucoside can now no longer be obtained commercially. 
It was, however, employed by Andrewes and Horder, who, as the result of 
their investigations, maintained their ability to distinguish well-marked 
differences of type among their streptococci. These observers elaborated a 
method of classification based on Gordon’s tests, but taking into consideration 
in addition the length of the streptococcal chains produced in ordinary broth 
VOL. LXXXIII.—B. 2k 
