ANALYSIS OF A MltfKRAT/ WATE& <27£ 



52. This precipitate was treated with distilled vinegar, 

 •until the vinegar no farther acted upon it. The residuum, 



when washed and dried, weighed gr. 3*2. Consequently gr. Ca-bonate of 

 16*55 of carbonate of magnesia were dissolved by the vine- a n <?subWftJo- 



gar, and gr. 3*2 of subcarbonate of alumina were left. n.ite of aln- 



mine. 



53. The concentrated mineral water, which was separated 



from the earthy carbonates &c. in 40, was diluted with dis- 

 tilled water, and oxalate of ammonia added, until the whole 

 of the lime fell down. The oxalate of lime that was thus 

 obtained was well washed and dried. It weighed gr. 482*25 Muriate of 

 zr to gr. 241 of pure lime ~ to gr. 482-25 of dry muriate ime# 

 of lime, taking Kirwan's estimate for the composition of the 

 latter. 



54. Pure potash was now added to the filtered solution 

 left in 53. The precipitate was separated, redissolved in 

 muriatic acid, and precipitated by subcarbonate of soda. 

 The precipitate washed and dried weighed gr. 118*3. 



55. This precipitate was treated with distilled vinegar un- 

 til no more dissolved. The subcarbonate of magnesia thus 

 dissolved weighed 99'05, equal to gr. 143*45 of muriate ^^ Muriate* of 

 magnesia according to Kirwan. The alumina when heated magaemand 

 red hot washed and dried weighed gr. 17*25 ~ to nearly gr. a umme * 

 57*3 of muriate of alumina. 



5u\ 88180 grains of the fresh water were evaporated to 

 drynesV. A quantity of muriatic gas escaped towards the 

 end of the operation from the partial decomposition of the 

 muriates of magnesia and alumina. Water and a small 

 quantity of muriatic acid were now added to dissolve the so- 

 luble parts of the mass; the insoluble powder was then 

 treated with strong muriatic acid, washed, dried, and weigh- 

 ed. It was now thrown into a hot solution of subcarbonate 

 of soda, from which it was separated, and treated with di- 

 luted muriatic acid, which dissolved a minute proportion 

 of carbonate of lime. This was precipitated and separated 

 by a filter, but did not exceed 0*25 of a grain by computa- 

 tion. --jg[ . 



57. The insoluble powder, whfch could consist only dfsilex. 

 »iliea and alumina, was boiled in a silver crucible with a so- 

 lution of pure potash, the solution was decanted off, and the 



T 2 white 



