748 BACTERIAL FLORA OF THE SYDNEY WATER SUPPLY, 



Pakes* adds the suspected water to Kitasato's sodium formate- 

 glucose-bouillon. This is prepared by adding 2% glucose and 

 0-4% sodium formate to ordinary bouillon, and after neutralising 

 adding 2 c.c. normal sodium hydrate (presumably to every 

 100 c.c. medium). The medium is boiled, filtered, placed in test 

 tubes and sterilised. After adding a certain quantity of water 

 or a bacterial suspension representing a quantity of the water, 

 the bouillon tubes are placed in Buchner's tubes and incubated 

 under anaerobic conditions at 42° C. until the medium becomes 

 turbid. If no growth is apparent after 72 hours the tubes are 

 thrown away. Those which show a growth are used for making 

 further, cultures on ordinary gelatine and agar. By this method 

 Bad. coli comm. is generally found in nearly pure culture. 

 With regard to the value of the separation he considers that 

 when coli is found in quantities of water up to 20 c.c. the water 

 should be condemned ; in from 20-50 c.c. the water is suspicious; 

 between 50 and 100 c.c. it is slightly suspicious ; and in quantities 

 greater than 100 c.c. it is probably safe. When none can be 

 discovered in two litres the water is absolutely safe. The 

 presence of coli in water from a deep well is enough to condemn. 



Although most of the methods suggested for the selective 

 examination of water have for their object the separation of 

 Bad. typhi, yet it is practically agreed that they fail in their 

 object unless these organisms are present in overwhelming 

 quantities. Most of the methods, however, are, to a certain 

 extent, successful in aiding the separation of Bad. coli commune, 

 which, when found in certain quantities of water, indicate pollu- 

 tion, and therefore the possible presence of Bad. typhi, especially 

 when cases of typhoid fever appear among those who have used 

 the water. If typhoid bacteria are suspected to be in a water, 

 and Bad. coli commune is found in a certain volume, it is probable 

 that Bad. typhi is or has been present. As an instance, Clark 

 and McGagef could not find Bad. typhi in the Merrimack River 



* Pakes, Brit. Med. Journal, 1900, 188. 

 t Clark and McGage, Centralblatt fur Bakt. 1 Abt. xxvii., 678, Pief. 



