1880.] 



Water supplies of Calcutta. 



91 



amount of work and time which is required for it, as a satisfactory analysis 

 by it cannot be performed in less than 4 or 5 days. On the other hand the 

 results obtained by Frankland's process are eminently trustworthy, and the 

 character of a water is determined by it with great precision. 



As I have been obliged to perform the work of analysis of the tank and 

 well waters of Calcutta during the spare time from my current duties, and 

 as some two hundred analyses had to be made by my own hands, it was 

 clearly impossible for me to use Frankland's more accurate process,and I was 

 compelled rather against my own notions of scientific accuracy to work with 

 Wanklyn's process, which as I have pointed out is not so trustworthy as the 

 other. In addition to this reason, I found that my predecessor in the office of 

 Analyst to the Corporation had been in the habit of testing the Calcutta 

 hydrant water by Wanklyn's process. As I had to carry on this method of 

 analysis on behalf of the Corporation, this therefore formed a very intelli- 

 gible standard of comparison for my work with the former water supply 

 of Calcutta. In addition however to these analyses of the hydrant water, 

 as will be seen subsequently, I have carried out for the last four years 

 monthly analyses of the hydrant water by Frankland's process, and it is 

 vipon these numbers that I shall base my conclusions as to the character 

 and quality of the present water supply. 



In Wanklyn's process there are two principal determinations. The first is 

 the estimation of the free ammonia present in the water, and of the albumi- 

 noid ammonia obtained by distillation with alkaline potassium permanganate. 

 In India, I have frequently combined these two jDrocesses, and the ammonia 

 from both is called the " Total Ammonia." The reason why these two pro- 

 cesses have been combined is, that in almost every case when I have tested 

 the potalle waters of India for free ammonia, I have found it to be almost 

 entirely absent. The fact appears to be, that at the very high temperature 

 which here obtains, the ammonia oxidizes with such extreme rapidity, that if 

 any free ammonia were present at the collection of the water, it would become 

 partially or wholly converted into inorganic nitrogenous matters before the 

 analysis could be performed, or, if the whole of the free ammonia were not 

 thus oxidized, the changes which go on from day to day are so great, that 

 for any true comparison in respect of this constituent between the sam- 

 ples of water analyzed, it would be necessary to analyze them at definite 

 intervals after collection. The " total ammonia" then, which is spoken of 

 subsequently, is the free ammonia present, if any, added to the ammonia 

 produced from the nitrogenous organic matter by the oxidizing action of 

 alkaline potassium permanganate. As pointed out before, it frequently hap- 

 pens that the whole of the nitrogenous organic matter present in the water 

 is not decomposed, and therefore the numbers obtained always represent the 

 minimum amount of impurity which can be present in the water. 



