24 AMERICAN MINERAL WATERS. 



devised by the writer. A rectangular metal case, sufficiently large 

 to accommodate six bottles, which are held in an upright position 

 by a removable perforated metal support, is provided with a lock 

 and key, and is suspended within a larger wooden box having a 

 surrounding air space of 3 inches at the top, bottom, and sides for 

 cracked ice. The hinged lid of this box may be left unlocked for the 

 purpose of reicing when shipments are made over long distances. 

 The recorded temperatures of a large number of water samples, as 

 received at the laboratory, varied from 5 to 15 C. 



The following table from the committee on standard methods of 

 water analysis * shows the maximum limits of time allowable after 

 collection before making the examination: 



Hours. 



Ground waters 6 



Fairly pure surface water 6 



Polluted surface water 6 



Sewage affluents Immediate plating. 



Raw sewage Immediate plating. 



In most instances the samples herein reported were examined, 

 either at source, or after longer intervals of time had elapsed than are 

 allowed by the Committee, owing to the distance between the springs 

 and the laboratory; they were, however, properly iced during this 

 time. The temperature of the water at the time of sampling was 

 taken in most instances. 



BACTERIOLOGICAL METHODS. 



In the present status of bacteriology, there is no known method by 

 which the total number of living bacteria can be determined. The 

 results obtained from any examination simply show the number and 

 kind of bacteria which will develop under absolute conditions em- 

 ployed in making that particular analysis; but for practical pur- 

 poses, the methods outlined by the committee permit the analyst to 

 arrive at certain more or less definite results, and enable him to draw 

 certain conclusions as to the probable purity of the water. 



MEDIA EMPLOYED. 



The standard medium for determining the number of bacteria in 

 water is nutrient gelatin or agar. Owing to the amount of field 

 work done and the polluted waters encountered, the determinations 

 were all based on agar counts. The agar medium used in these 

 water investigations is made from nutrient broth prepared from 

 fresh, lean meat, Witte's peptone, and 15 grams per liter of the 

 highest grade thread agar obtainable. Sodium chlorid is added to 



i This title will be subsequently designated and referred to as "Committee" without further qualifica- 

 tion. 



