406 



Dr. E. W. A. Walker. 



[May 16 r 



from another given micro-organism, the two (so long as they remain 

 indistinguishable) can no longer be regarded as distinct varieties. Until; 

 some differentiating character has been discovered, they must be looked upon 

 as identical. It has been shown in my experiments that these considerations 

 apply to streptococci in respect of their reactions in Gordon's media. And it 

 has also been shown that, on sub-culture through the animal body, as well as 

 on sub-culture in artificial media, the test reactions of particular streptococci 1 

 tend to exhibit striking variations. 



The conclusion seems to follow unavoidably that Gordon's tests do not 

 afford an adequate basis for a classification of streptococci. This conclusion, 

 however, does not necessarily imply that the tests are valueless. Nor does it 

 suggest that fixed and definite distinct varieties do not exist among the 

 streptococci. But it maintains that the existence of such varieties has not 

 been proved by the application of the tests, nor even rendered in the least 

 degree probable. 



Even had the tests been found to be reasonably constant under ordinary 

 conditions, the results obtained by their use do not on close examination' 

 appear to supply so clear and decisive a differentiation of streptococci as 

 might be supposed from writings on this subject. Thus, among six variants- 

 of the S. equinus described by Andrewes and Horder (1906, 7) — a type 

 of streptococcus "which seems entitled to rank as a species" in their 

 opinion — no less than three are identical with variants of their S. pyogenes-,. 

 so far as Gordon's tests themselves afford any indication. The distinction 

 which they actually draw between the organisms is made to rest on 

 characters which form no part of Gordon's tests, such as pathogenicity,, 

 the presence or absence of growth in gelatin, and so forth. 



Again, if one looks into the reactions actually laid down by Gordon 

 himself in his latest communication (1911, 4) as typical of S. salivarius, 

 S. pyogenes, and S. fcecalis, one finds a range of admitted possible variation 

 such as brings S. pyogenes very near S. salivarius on the one hand, and 

 equally near S. fozcalis on the other. For it is to be noted that, in the 

 form in which the reactions are now given, we have two alternatives for 

 S. salivarius, and no less than eight alternatives for S. pyogenes (see Table IV); 

 Further, the only absolutely fixed distinction, as regards the test reactions 

 between what may pass as a S. pyogenes and what is to be classed as a 

 raffinose-negative S. salivarius, is the difference in the reaction to neutral 

 red. Similarly, the only distinction in the test reaction between another 

 form of S. pyogenes and the S. fcecalis is again the reaction in the neutral red 

 test — a test which Andrewes and Horder have noted (along with salicin) as- 

 " more liable to vary than most of the others." 



