128 THE CANADIAN NATURALIST. [Yol. vii. 



The well water contained nitrate of potash in sufficient quantity 

 to admit of its separation by crystallization ; and the predomi- 

 nant constituent in the other specimens was sulphate of lime. 



There is now every reason to suppose that the water of Winter 

 River will be brought into Charlottetown before long, through a 

 proper system of supply pipes, although the same partisan feeling 

 on the ' water question' finds expression there as elsewhere. 



EsTi3iATioN or Phosphoric Acid. — As our deposits of 

 apatite continue to be opened up, a rapid method for the deter- 

 mination of the percentage of phosphoric acid will become 

 more and more a desideratum. Several articles have recently 

 appeared on this subject in scientific journals, and the following 

 abstract of one of them is given in the Journal of the Chemical 

 Society for March* : 



" Two grams of the phosphate to be examined are treated at 

 the ordinary temperature with 50 c.c. of dilute hydrochloric or 

 nitric acid, the solution is filtered, and the filtrate treated with 

 citric acid, and then with excess of ammonia ; the phosphoric 

 acid is then precipitated with solution of magnesium chloride in 

 excess, whereby the precipitated ammonio-magnesium phosphate 

 is made to subside more rapidly than it would otherwise do The 

 supernatant liquid is now separated from the precipitate by means 

 of an aspiration-filter, and the precipitate is washed with am- 

 moniacal water, which is afterwards removed by the same means. 

 The precipitate is next dissolved by means of a few drops of nitric 

 acid and the phosphoric acid estimated with uranium acetate 

 solution, according to a modification of Leconte's method. 



Boussingault has stated that an excess of ammonium citrate 

 holds in solution a considerable portion of the ammonio-magne- 

 sium phosphate ; but the author finds that by using not more than 

 80 to 100 parts of citric acid to one of phosphoric acid contained 

 in the substance, no loss is experienced. 



He also finds that by adding excess of magnesium chloride, 

 keeping the proportion of citric acid within proper limits, adding 

 the right quantity of ammonia in excess, and not allowing the 

 total volume of the solution to exceed a certain amount, accurate 

 results can be readily obtained in presence of lime, iron and 

 alumina. " 



* Any one interested in this matter should consult Joulin's paper 

 in the ' Chemical News ' for May 9th. 



