The Sanitary Value of the 



and lakes withdraw them from water. Also that nitrates once formed 

 may be again reduced to ammonia compounds or even to free nitrogen. 

 If so, then an undue importance has been attached to the determina- 

 tion of nitrogen in this form, by Frankland and others, who make use 

 of the term " previous sewage contamination " to denote the amount 

 of sewage which has been discharged into a water and undergone oxi- 

 dation, estimating the same by determining the total combined inor- 

 ganic nitrogen present, as ammonia, nitrates and nitrites, and after 

 deducting the amount of nitrogen present in these forms in rain 

 water, calculating how much " average London sewage" would be 

 required to account for it. It will therefore be seen that much differ- 

 ence of opinion exists as to the degree of importance to be attached 

 to the presence of these compounds in water, and that the questions 

 involved are exceedingly intricate and a3 yet very far from being set- 

 tled. Various methods are employed in their determination, none of 

 which are perfectly satisfactory. 



We may now dismiss from further consideration the inorganic con- 

 stituents of potable waters, and proceed to describe those methods of 

 analysis developed during late years, which have for their object the 

 determination of the amount and so far as possible the nature of the 

 organic constituents, so much more important than the inorganic, 

 from a sanitary point of view. The following are the most im- 

 portant of these processes : 



1. Permanganate methods. — In 1819 Forchammer proposed the use 

 of permanganate of potassium for the determination of the oxidizable 

 matter in water. This salt imparts, even in very small quantities, a 

 decided pink color to water, and as it contains a large amount of 

 oxygen which it readily gives up to reducing bodies, thereby losing its 

 characteristic color, its use affords a ready method for the determina- 

 tion of the amount of many such substances in a state of solution. 

 As applied to the determination of organic matter however, the results 

 can have no absolute value, since different kinds of organic matter 

 require for oxidation different amounts of oxygen, and moreover differ 

 in the completeness with which they are oxidized and the length of 

 time required. Much also depends upon the exact method in which 

 the test is applied, and various inorganic compounds which may be pres- 

 ent, reduce the permanganate and become a source of error. Neverthe- 

 less this test has a certain value, and it was at one time even thought that 

 the amounts of putrescible matter might be estimated by the amount of 

 permanganate decolorized during the first few minutes, since such 

 matter probably first undergoes oxidation. Frankland held this view, 

 as did Angus Smith, but the former has of late entertained a different 

 opinion. W. A. Miller, Kubel, Tidy, Letheby, Schultze and others 



