MUNICIPAL WATER SUPPLIES. II3 



logical reactions above noted is an important one. At the 

 outset it must be noted that neither chemical nor bacteriolog- 

 ical examinations alone are sufficient to safeguard a water 

 supply. The two classes of tests must go hand in hand, 

 each having various advantages over the other. It may be 

 of interest to consider these points a little in detail. 



In the first place, the chemical analj'sis shows the true 

 chemical nature of the water, something which the bacterio- 

 logical tests do not do. Again, the chemical tests are usually 

 applied to much larger quantities of the water and, therefore, 

 give a truer "average." As against this, however, the 

 examination is more complicated and, therefore, except in 

 large laboratories, would not be made so often. 



As regards the tests more especially indicative of sewage 

 contamination, we have on one hand the chemical test for 

 chlorine and on the other the bacteriological tests for Bacillus 

 coll communis. The former is quite simple and very certain. 

 The latter, although not so simple, are probably just as certain 

 by reason of their directness. 



xAs concerns the nitrogenous matter of the water, we may 

 contrast the chemical and bacteriological tests in manner fol- 

 lowing : The two ammonias, nitrites and nitrates give a 

 measure of the organic matter in the water. This organic 

 matter must serve at one or another time as a good culture 

 medium for whatever bacteria there may be in the water. 

 The relative proportion of these items tells us whether the 

 medium is in good, active condition or becoming exhausted. 

 It is reasonable to assume the bacterial activity greater when 

 the chemical activity is greater, as occurs in the early stages 

 of decomposition ; that is, during the free ammonia and 

 nitrite stages especially. This is an assumption based on the 

 chemical determinations. On the other hand, the gelatine 

 plate cultures and counting of colonies give a direct count of 

 the number of bacteria, and one determination should confirm 

 the other. 



Again, the chemical examination is much more accurate 



