348 M. Arfwedson's Analysis of some Minerals. [May, 



In order to free the boracite as completely as possible from 

 all admixture of the matrix, consisting of gypsum, a portion of it 

 reduced to a fine powder was boiled repeatedly with water. It 

 was afterwards collected on the filter, washed, and dried. 



0*849 oramme of it were mixed in a platinum crucible with 

 3 o-rammes of Derbyshire fluor spar in fine powder, made into a 

 paste with concentrated sulphuric acid, and then heated till it 

 was reduced to dryness, taking the requisite precautions to 

 avoid the dissipation of any portion from the effervescence, and 

 the evolution of the gas which took place. The dry mass was 

 then exposed to a red heat. For the greater security, the process 

 was repeated, but no fluoboric gas was disengaged ; showing 

 that the mineral had been completely decomposed by the first 



process. 



The sulphate of magnesia was then extracted by water, and 

 the undissolved portion was washed so long upon the filter that 

 I was sure none of the magnesia remained mixed with the gyp- 

 sum. The filtered water was then freed from the gypsum which 

 it contained in solution by oxalate of ammonia. It was then 

 evaporated to dryness, and exposed to a red heat. The salt thus 

 obtained weighed 0*758 gr. and possessed the characters of pure 

 sulphate of magnesia. The quantity of magnesia in it amounted 

 to 0-257, and consequently the remaining 0*592 gr. necessary to 

 make up the weight of the boracite must have been boracic 

 acid. 



100 parts of boracite then contain 



Boracic acid. 69*7 



Magnesia 30*3 



100*0 



Gay-Lussac and Thenard found that boracic acid contains 33 

 per cent of oxygen. If this be the case, the oxygen in 69*7 is 

 23. 30*3 parts of magnesia contain, on the other hand, 11*73 

 parts of oxygen; but 11*73 x 2 = 23*26; so that boracic acid 

 contains twice as much oxygen as magnesia. As long as the 

 constitution of boracic acid remains disputed, I will not allege 

 this coincidence as a proof of the accuracy of my analysis ; but 

 merely as a circumstance from which Gay-Lussac and Thenard's 

 conclusions may receive some support. 



