SPECIAL EXAMINATIONS 359 



incubate at 37 C. for twenty-four hours. Examine a 

 hanging drop, and if cocci are seen, make a subculture by 

 smearing a platinum loopful of culture over a plate of 

 Drigalski and Conradi's medium (nutrose, -lactose, -agar). 

 Incubate at 37 C. for twenty-four hours, and then sub- 

 culture all the minute colonies into tubes of Houston's 

 Lemco medium (see p. 222), containing 0-5 per cent of 

 lactose, mannite, raffinose, saccharose, and salicin respect- 

 ively, and also into milk. Incubate for two days at 37 C, 

 and observe results. 



From various researches, extending over ten years, into 

 the characteristics of faecal streptococci (see Fifth Research 

 Report, and other references there, p. 13), Houston has 

 found that of 100 " sewage works " streptococci 



All produced acid in lactose and raffinose media. 



All clotted milk. 



None reduced nitrates to nitrites. 



All but three produced acid in salicin medium. 



Forty-nine produced acid in saccharose medium. 



Only four produced acid in mannite medium. 



These reactions differ from those given by Andrewes and 

 Horder for streptococcus faecalis (see p. 224). 



In 1909, out of 156 samples of raw river water (Thames, 

 Lee, and New River, 52 from each), 28 were found to contain 

 streptococci in 1 c.c. ; 1908 subcultures were made, and from 

 these 71 streptococci were isolated, plus 3 from o*i c.c. of 

 water, making 71 -f- 30, or 101 in all, or say 1 11, so as to over- 

 state any error. That is, in lactose-positive streptococci 

 were found in 156 c.c. of raw river water, which amount 

 would yield of all bacteria on gelatin plate over two million 

 colonies ; on agar plate about sixty thousand ; and on bile- 

 salt agar about six and a half thousand. The ratio, therefore, 

 of streptococci to total bacteria (growing on gelatin plate) in 

 the raw river water works out at one to twenty thousand. 

 The type of streptococcus usually met with in these 156 

 samples, Houston describes as the la-mi-ra-sac-sal variety 

 (i.e. acid in lactose, clot in milk, acid in raffinose, saccharose, 

 and salicin media). As noted above, of the faecal streptococci 

 (100) examined by him, 97 clotted milk, and produced acid in 

 lactose, raffinose, and salicin; and of these 97, 48 produced 

 acid and 49 no appreciable change in a saccharose medium. 



In other researches he found that in human faeces of the 



