560 STREPTOCOCCUS 



The many attempts that have been made to evolve a satisfactory classification 

 of the streptococci provide an admirable example of the difficulties with which the 

 systematic bacteriologist is faced. As will be seen, there is no single criterion on 

 which reliance can be placed, even in making a primary division into sub-groups 

 that are themselves to be further divided by the application of other tests. It 

 happens that, in this particular genus, one of the most useful differential criteria 

 is provided by the changes induced by the growing organisms in media containing 

 blood ; but we shall note that this test, valuable as it is, cannot be too rigidly 

 applied. Here, as elsewhere, we have to apply a variety of criteria selected, on 

 the usual basis of statistical empiricism, as differentiating between groups each of 

 which possesses several highly correlated characters. Here, as elsewhere, we find 



that the method of antigenic 

 analysis is playing an in- 

 creasingly important part in 

 the differentiation and iden- 

 tification of those ultimate 

 >''"*.^ types, or varieties, for which 



> f ,•••"' *\ we need distinguishing names 



j'' f or labels. 



' *^ ; / The most convenient 



\ ^ \ \ method of discussing this 



* / \ J " problem will be to take 



.' .^-"--•^Bk,^ vZ ••'••>. various characters in turn, 



s J ^^^iv-"^ • ^^^ ^^® ^^^ ^^^ they enable 



^ / .^N \ us to differentiate between 



: \ \ one species, or type, and 



*\ / another. 



'^ ■ Morphology. — Taking the 



genus as a whole, the charac- 

 ters that might be regarded 

 as supplying possible differ- 

 FiG. 117. — 8tr. pyogenes. ential criteria are (1) the 



From 24-hours' culture on agar, showing long chains length of the chains formed, 



( X 1000) . (2) the shape of the individual 



cells forming them, (3) the 

 presence or absence of capsules, and (4) in the light of certain recent observations, 

 the very occasional presence of flagella. 



In the earlier days of bacteriology much attention was paid to chain formation 

 as a differential criterion, and such names as Streptococcus brevis, Streptococcus 

 longus, Streptococcus longissimus, and Streptococcus conglomeratus were coined to 

 denote strains with the corresponding tendencies to grow in short chains, long 

 chains, or chains which were very long or tangled. It has, however, come to be 

 generally recognized (Thalmann 1912, Brown 1919) that these characters are often 

 variable within a single strain ; so that, while the modal chain length of any species 

 may be sufficiently characteristic to deserve inclusion in a description of the specific 

 characters, it is quite useless for purposes of classification, and misleading when 

 employed for purposes of nomenclature. Some species or groups, such as Str. 

 pneumonice and the enterococcus, usually occur in pairs or very short chains and 

 never form chains of any considerable length. Others, such as Str. pyogenes, 



