134 AIDS TO BACTERIOLOGY 



than the latter. But virulent short-chain streptococci 

 are sometimes met with, and it is possible to transform one 

 variety into the other by culture. Penfold describes 

 the variations obtained as ' haphazard.' Strains that 

 correspond to haemolytic streptococci have been converted 

 into typical pneumococci and vice versa. 



Gordon divides the streptococci into four groups: 

 (i.) S. longus (from the mouth). Very long and compara- 

 tively straight chains. Broth remains clear with floccu- 

 lent deposit. Acid but no curd in milk, (ii.) S. medius 

 includes most pyogenic streptococci, which as a rule form 

 fair-sized curling chains; corresponds to Lingelsheim's 

 longus type. Broth acquires a flocculent deposit but 

 supernatant liquid remains clear. Milk becomes slightly 

 acid without curd, (iii.) 8. brevis. Short chains. Broth 

 uniformly turbid. Slight acid and usually curd in milk. 

 Includes pneumococcus. (iv.) S. scarlatina 1 or conglo- 

 meratus. Masses of chains. Deposit in broth but upper 

 liquid keeps clear. Acid and curd in milk. Later, 

 Gordon introduced fermentation tests as a basis of differ- 

 entiation and using these Andre wes and Horder dis- 

 .tinguish (1) S. pyogenes, already described. (2) S. 

 salivarius (found in saliva), a brevis type that clots milk, 

 is not haemolytic nor pathogenic for mice. (3) S. angino- 

 sus (inflamed fauces, scarlatinal throats, and rheumatism), 

 a longus type, with no action on salicin, is haemolytic 

 and pathogenic for mice. (4) S. fcecalis, a faecal organism 

 sometimes found in cystitis, meningitis, and pus. A brevis 

 type that clots milk, is not hsemolytic nor pathogenic 

 for mice, the only class to ferment mannitol. Perhaps 

 the Diplococcus rheumaticus is identical with this organism. 

 (5) The pneumococcus (q.v.). (6) S. equinus (from 

 horse dung), a brevis type, not clotting milk and the only 

 class that does not ferment lactose. 



Too much stress must not be placed on these differ- 

 ences, experience showing the characters mentioned are 

 anything but absolute. 



Salivary streptococci do not usually ferment salicin, 

 and Savage thinks the discovery of salicin-fermenting 

 streptococci in throats during an epidemic of sore throat 

 would point to a milk infection. 



Streptococcus mastitidis, found in streptococcic in- 

 flammation of the mammae of cows (mastitis), and S. an- 



