VARIETIES OF STREPTOCOCCI. 191 



for example, found that the virulence of a streptococcus can be 

 enormously increased by growing it alternately (a) in a mixture 

 of human blood serum and bouillon {vide p. 46), and {U) in 

 the body of a rabbit ; ultimately, after several passages it pos- 

 sesses a super-virulent character, so that even an extremely minute 

 dose introduced into the tissues of a rabbit produces rapid septi- 

 caemia, with death in a few hours. It has been proved by Mar- 

 morek's experiments, and those of others, that the same species 

 of streptococcus may produce at one time merely a passing local 

 redness, at another a local suppuration, at another a spreading 

 erysipelatous condition, or again a general septicaemic infection, 

 according as its virulence is artificially increased. Such experi- 

 ments are of extreme importance as explaining to some extent 

 the great diversity of lesions in the human subject with which 

 streptococci are associated. 



Varieties of Streptococci. — Formerly the streptococcus pyo- 

 genes , and the streptococcus erysipelatis were regarded as two 

 distinct species, and various points of difference between them 

 were given. Further study, and especially the results obtained 

 by modifying the virulence, have shown that these distinctions 

 cannot be maintained, and now nearly all authorities are agreed 

 that the two organisms are one and the same, erysipelas being 

 produced when the streptococcus pyogenes of a certain standard 

 of virulence gains entrance to the lymphatics of the skin. 

 Petruschky, moreover, has shown conclusively that a strepto- 

 coccus cultivated from pus may cause erysipelas in the human 

 subject. He obtained a pure culture of a streptococcus from a 

 case of purulent peritonitis secondary to parametritis, the patient 

 having never suffered from erysipelas. By inoculations with 

 this culture he produced typical erysipelas in two women suf- 

 fering from cancer. 



More recently a distinction has been drawn between a streptococcus longus, 

 which forms long chains, and is pathogenic to rabbits or mice, and a strepto- 

 coccus brevis, which occurs in the mouth in normal conditions, and is without 

 pathogenic properties when tested experimentally. The growth of the former 

 in bouillon forms a somewhat granular deposit, that of the latter a more 

 abundant and flocculent deposit. Marmorek has, however, found that the 

 same streptococcus may at one time grow in short, at another in long chains, 

 and Kolle has shown that a streptococcus, which originally grew in long 

 chains, formed only short chains after being repeatedly passed through the 

 body of the mouse, the appearance of the growth in bouillon being corre- 



