1852.] 123 



III. 2'4806 grammes gave: 



Silicic acid . . . . . . 0-8288 grammes. 



Sesquioxide of iron i 



chromium V . . . 0-4502 '< 



Alumina \ 



Pyrophosphate of magnesia . . . 2-3499 " 



<c 



Magnesia 0-0451 



IV. 2.6148 grammes gave : 



Water 0-3344 grammes. 



In analysis No. I. the fine powder was fused in a platinum crucible with car- 

 bonate of soda and nitrate of potash ; the fused mass was decomposed by hydro- 

 chloric acid, and evaporated to dryness ; the dry mass moistened with hydro- 

 chloric acid, heated and filtered. The silicic acid, thus separated, containing 

 still a small quantity of chromium, etc., was fused a second time by carbonate of 

 soda, and treated as above; the solution containing the small quantity of chro- 

 mium was added to the other part. The acid liquid containing in solution the 

 sesquioxides of chromium and iron, alumina and magnesia, was very nearly 

 neutralized by carbonate of soda, and precipitated by carbonate of baryta. The 

 mixture was often stirred, and after two days filtered. The weaker bases were 

 completely precipitated, and did not contain a trace of magnesia. The precipi- 

 tate of alumina, the sesquioxides of iron and chromium, and the excess of car- 

 bonate of baryta, were treated with sulphuric acid and the soluble sulphates 

 filtered from the insoluble sulphate of baryta. The filtrate was precipitated by 

 ammonia and boiled, to prevent the dissolution of sesquioxide of chromium. This 

 precipitate, after having been filtered and dried, was finely powdered, and fused 

 with carbonate of soda and nitrate of potash, in order to oxidize the sesquioxide 

 of chromium into chromic acid. From the solution in water alumina was preci- 

 pitated by carbonate of ammonia. The filtrate was then acidulated with hydro- 

 chloric acid, the chromic acid reduced by alcohol, and the sesquioxide of chro- 

 mium precipitated from the boiling solution by ammonia. 



After baryta had been separated by sulphuric acid from the filtrate of the weak 

 bases, magnesia was precipitated from the ammoniacal solution by phosphate of 

 ammonia, with the requisite precautions. In analysis No. III. I decomposed 

 the finely powdered mineral (previously heated to redness) by evaporation with 

 sulphuric acid. The silicic acid not having a white color, was fused with car- 

 bonate of soda and nitrate of potash as above, and the solution containing 

 alumina, sesquioxides of chromium and iron, and magnesia added to the other 

 parts. The sesquioxides were separated from magnesia by carbonate of baryta. 



The solution containing the magnesia, from which baryta was separated by 

 sulphuric acid, and to which an excess of chloride of ammonium had been added, 

 gave with ammonia a white gelatinous precipitate, which was insoluble in caustic 

 potash, but it gave the other reactions of magnesia. In analyzing this mineral 

 1 tried at first to separate magnesia from the sesquioxides by chloride of ammo- 

 nium and ammonia, but although I had ounces of ammoniacal salts in solution, 

 only half of the magnesia (17 per cent.) remained in solution. I dissolved and 

 precipitated thus three times, but at last I had to give it up, and separated the 

 balance which was remaining with the weak bases by carbonate of baryta. I 

 never before had such difficulties in separating alumina from magnesia, and H. 

 Rose, in his newest edition, is still in favor of this method, and separated the 

 small quantity of magnesia, which always falls down With the alumina, by dis- 

 solving the latter in caustic potash. I shall make further experiments with the 

 magnesia separated from Rhodophyllite, and intend to make communication of 

 my results to the Academy, if I find them interesting enough. 



For the estimation of the alkalies, the mineral, previously heated to redness, 

 was decomposed by sulphuric acid, the soluble sulphates were boiled with car- 

 bonate of baryta, and the carbonates of the alkalies with traces of magnesia 

 extracted by water. The solution v/as evaporated to dryness, dissolved and 

 filtered from som.e magnesia which remained ; but the last trace of it was sepa= 

 rated by oxide of mercury after the carbonates were converted into chlorides. 



