EXAMINATION OF WATER 583 



water known to be polluted, and of which an estimate of the number 

 of organisms present is desired. In order to count the colonies if 

 very numerous, ink lines may be drawn across the bottom of the 

 Petri dishes so as to divide them into sectors. Ruled paper discs 

 (Pakes's discs) upon which the dishes are placed can also be obtained. 

 The colonies in the sectors are then much more easily counted : 

 or if the colonies be very numerous and evenly distributed, the 

 number in two or three of the sectors may be counted, and the 

 total number on the plate estimated by calculation. 



SEARCH FOR BACILLUS COLI, ETC. Various media may 

 be employed for the detection, isolation, and enumeration 

 of B. coli. The writer generally employs as a preliminary, 

 glucose bile-salt peptone-water, but many other media 

 may be employed, e.g. formate or neutral-red broth, or 

 if the organism is abundant, neutral-red bile-salt agar. 



As a routine, 50 c.c. should be the minimal quantity 

 examined for the presence of the Bacillus coli, quantities 

 from a minimum of 0*1 c.c. to a maximum of 50 c.c. being 

 added to the tubes of culture media. 



It is preferable to add the water directly to the tubes 

 of culture medium, even with the larger amounts, and 

 not to concentrate the bacteria by any method. The 

 culture media may be diluted with at least an equal volume 

 of the water without interfering with their cultural pro- 

 perties, and large tubes or small flasks are used for the 

 larger amounts. 



In the case of glucose or lactose bile-salt peptone- water, 

 the medium may for the larger amounts be prepared of 

 double strength. The glucose or lactose bile-salt peptone 

 water should be incubated at 42 C. for not less than 

 forty-eight hours. 



For composition of glucose formate broth, glucose and lactose 

 bile-salt media, and neutral-red broth, see p. 590, et seq. While a 

 lactose medium has the advantage of excluding a number of forms 

 which, though fermenting glucose, do not ferment lactose, and are 

 therefore not typical B coli, Houston has found that a glucose 



