BACTERIOLOGY OF WATER 233 



Gram-stained preparations are made from the bottom 

 liquid. Or the actual tubes used for the primary examina- 

 tion for B. coli may be examined. 



Not all streptococci will grow in the presence of bile 

 salt, but as the group that is important in the sanitary 

 analysis of water i.e., those from human excrement 

 does, omission to find others is of little consequence. 



For the isolation of streptococci, Conradi-Drigalski agar 

 is convenient. Houston showed that fsecal streptococci 

 from human sources produce acid from lactose and 

 saccharose media, and generally from salicin too. They 

 produce acid and clot in milk, produce a uniform turbidity 

 in broth, and reduce neutral red that is, they correspond 

 to the S. fa'calis of Andrewes and Horder (p. 134). Hous- 

 ton found no reliable distinction between the streptococci 

 of man and animals. With the horse, sheep, rabbit, 

 and cow, excrement is deficient in streptococci, and 

 especially in lactose-positive streptococci as compared 

 with human excrement. 



It is important to note that the human excrement 

 group of streptococci is of a brevis type. If, in the examina- 

 tion of a Gram-stained slide, Gram-positive diplococci 

 are noticed, the search should be kept up, since perhaps 

 these diplococci are streptococci that have not grown 

 out. In most of such cases definite chains will be found 

 sooner or later. 



In contradistinction to B. coli, streptococci do not 

 multiply in water after sampling (non-multiplying factor). 

 Previously regarded as indicators of recent pollution, 

 owing to an idea that they soon died out in water, they 

 are now regarded by Horrocks and most other workers 

 as living as long as, if not longer than, B. coli. Their 

 number should not exceed that permissible for B. coli. 



Enumeration of B. Welchii. Quantities amounting to 

 250 c.c. for a surface water, and up to a litre for a deep- 

 well water, are examined as described on p. 87. This 

 organism is a ' non-multiplying factor.' Savage considers 

 that it should be absent from 100 c.c. of a surface water 

 and from a litre of a deep-well water. The estimation 

 is not usually included in routine examinations, owing 

 to the uncertainty as to its interpretation. Thresh has, 

 however, suggested a number of standards (see * Applied 

 Bacteriology '). 



