672 STREPTOCOCCUS 



The behaviour of rough variants appears to be inconstant. Many, probably most, rough 

 strains are bile-sohible (Griffith 1923, Reimann 1927, Downie et al. 1931) ; but some are 

 less readily lysed than normal, smooth strains, and a few are apparently quite insoluble. 

 It seems clear that this test, properly performed, is one of the most reliable at our disposal. 

 A smooth strain that proves insoluble in sodium deoxycholate should not be given the 

 title Sir. pneumonioB without a clear statement that, because of this important divergence 

 from the specific characters, the diagnosis must be regarded as uncertain. 



The use of bile has provided another diflferential test within this group. Weissenbach 

 (1918) noted that enterococci grew well in a medium containing 10 per cent, of bile, while 

 Str. fyngenes and other streptococci tested by him failed to grow in this medium. The 

 observation that enterococci grow freely in high concentrations of bile has been confirmed 

 by many subsequent observers ; but it would appear that this character is not peculiar 

 to this species, or group. The abiUty of other species and types to grow in bile-containing 

 media (10 per cent, and 40 per cent.) has been studied by several workers (see Belenky 

 and Popowa 1929, Minett and Stableforth 1931, Lancefield 1933, Hare and Colebrook 1934, 

 Hare 1935, Hare and Maxted 1935, Colebrook, Maxted and Johns 1935). The possible 

 significance of this character in the classification of streptococci other than the enterococci 

 will be more conveniently discussed in relation to the results obtained by antigenic analysis. 



Another test that has been apphed in the classification of this group is the ability 

 to reduce, and thus decolorize, methylene hlue added to milk, usually in a concentration of 

 1 : 5,000 (Avery, R. C. 1929ff, h). This test would appear to depend in part on the abihty 

 of the organism to multiply in the presence of methylene blue, in part on the Eh that the 

 growing organism induces. Many of the strains that reduce methylene blue have been 

 isolated from animal sources or from milk. There is general agreement that strains of 

 hsemolytic streptococci isolated from severe human infections fail to do so. Enterococci, 

 however, reduce methylene blue actively (Kleckner 1935), even in a concentration of 

 1 : 1,000. Here again, the significance of this test must be considered in relation to the 

 antigenic structure of the various species or types. 



A fourth test that is of considerable value from the point of view of classification is 

 that of heat resistance. It has long been known that heat-resistant streptococci occur in 

 milk (see Ayers and Johnson 1910, 1913, Ayers, Johnson and Davis 1918) ; and Logan 

 (1914) recorded the presence of heat-resistant streptococci in the human faeces. Houston 

 and McCloy (1916) noted that heat resistance was characteristic of enterococci, and this 

 observation has been amply confirmed by Dible (1921) and by many subsequent observers. 

 It is usual to employ exposure to a temperature of 60° C. for 20 to 30 minutes as an arbitrary 

 test of heat resistance. The problem that arises in this connection is the identity or non- 

 identity of the milk and fsecal strains of heat-resistant streptococci ; and this will be 

 considered in a later section. Other tests, which will be considered under the classifica- 

 tion of the enterococci, comprise ability to grow at 45° C, at pH 9-6, and in the presence 

 of 6-5 per cent. NaCl. 



Antigenic Structure. — It will be convenient to begin our discussion of anti- 

 genic structure with the pneumococcus, in part, because this is a well-defined 

 species in which the complications referred to above do not arise, in part, because 

 it was in fact one of the first species in which a division into immunologically diff- 

 erent races, or types, was clearly demonstrated, and, in part, because the chemical 

 study of antigenic structure was initiated with this organism, and we still know far 

 more about the chemical differences that determine the antigenic specificity of the 

 various types of pneumococci than we do about similar differences in any other 

 bacterial species. 



The Antigenic Structure of the Pneumococci. — Neufeld and Handel (1909) 

 first demonstrated the existence of antigenically different types of pneumococci. 

 They studied the protective effect of different antipneumococcal sera in mice, and 



