BACTERIOLOGY OF WATER 231 



respectively of the sample. When a water to which a 

 more stringent standard for B. coli can be applied is to be 

 examined, quantities of 10 c.c., 20 c.c., and 20 c.c. are, in 

 addition, added to double-strength tubes containing 

 similar quantities of the broth. This will give a total 

 quantity of water examined of over 100 c.c. The tubes 

 are then incubated at 37 C. or, better, at 42 C. 

 for twenty-four to forty-eight hours. The production of 

 acid and gas is only presumptive evidence of the presence 

 of B. coli. Other organisms are frequently responsible 

 for the change, hence the necessity for the isolation and 

 identification of the organism. Contrariwise, the colon 

 bacilli may not give the reaction for seventy- two hours, 

 or they may be inhibited or killed by the bile-salt, but 

 in these cases the bacilli may be regarded as attenuated 

 forms and their non- detection as of no serious consequence. 

 Negative results are very occasionally due to death or 

 inhibition of colon bacilli through overgrowth of strepto- 

 cocci. This only seems to occur with grossly polluted 

 waters, especially when large quantities of water (such 

 as 100 c.c.) are tested (Prescott and Winslow). 



The tubes which received least water, and which 

 show acid and gas, are used for making the secondary 

 cultures for which neutral red bile- salt agar or Conradi- 

 Drigalski agar are convenient. The use of Petri dishes 

 for these secondary cultures is not necessary. A loopful 

 of liquid from a primary tube showing acid and gas is 

 added to a test-tube containing 10 c.c. of sterile water. 

 This is agitated, and a loopful transferred to another 

 test-tube of sterile water. The latter is agitated, and a 

 loopful added to a third tube of sterile water. A loopful 

 of the first dilution is smeared over the surface of a sloped 

 tube of medium, and similar quantities of the other two 

 dilutions are smeared over the surfaces of two other sloped 

 tubes. One or other of these secondary cultures is pretty 

 certain to give discrete colonies. Characteristic colonies 

 are subcultured into tubes of nutrient broth, and when 

 growth is sufficient the broth cultures are examined for 

 the presence of motile organisms, and preparations are 

 examined by Gram's method. The attributes of a typical 

 colon bacillus are given on p. 90. From the broth culture 

 inoculations into the various media necessary to identify 

 the organism are made. The estimation of B. coli is 



