British Mineralogy. 183 



in nitrohydrochloric acid, a quantity of iron added and dissolved 

 in the solution, which then was precipitated in the cold by car- 

 bonate of barytes ; after nitration the excess of baryta was re- 

 moved from the solution by the addition of sulphuric acid. The 

 nitrate was then precipitated by colourless sulphide of ammonia, 

 and the manganese separated from the sulphides of nickel and 

 cobalt by acetic acid, and, after precipitation from the acid solu- 

 tion by sulphide of ammonium, was determined as usual. After 

 boiling the nitrate and separating some free sulphur by nitration, 

 the magnesia was precipitated as ammoniacal magnesian phos- 

 phate and determined as pyrophosphate of magnesia. The sul- 

 phides of nickel and cobalt were dissolved in nitrohydrochloric 

 acid and precipitated as oxides by caustic potash ; the amount of 

 cobalt in the same was then determined by Plattner's method. 



II. For the sake of confirming the results, a second analysis 

 was made by" a different method, namely : — A weighed portion 

 was dissolved in nitrohydrochloric acid; a small amount of silica 

 and free sulphur remained undissolved, which were determined, 

 and afterwards the amount of free sulphur determined as ignition 

 loss. To the filtrate was added chloride of barium in order to 

 throw down the sulphuric acid, which was then determined as 

 sulphate of baryta. The excess of baryta was now removed from 

 the solution by adding sulphuric acid and filtering ; the solution 

 was treated with a current of sulphuretted hydrogen passed 

 through it for several hours, during which it was kept on a warm 

 sand-bath ; it was then allowed to stand some eight hours, and, 

 after again passing a current of hydrogen through it, was fil- 

 tered, and the sulphide of arsenic oxidized by nitrohydrochloric 

 acid, filtered from free sulphur, and, after adding the solution of 

 a weighed amount of iron, was precipitated by ammonia. The 

 other constituents were determined as usual, except that the se- 

 paration of the iron from the nickel and cobalt was effected by 

 treating the sulphides with weak hydrochloric acid. The amounts 

 employed and obtained in these two analyses were as follows : — 



I. 18*29 grs. mineral employed; 12*46 grs. iron wire; sul- 

 phate of baryta obtained, 26*78 grs., equal to 3*67 sulphur, or 

 20*01 per cent.; 27*50 grs. sesquioxide of iron with arsenic acid; 

 and this, less 17*84 grs. sesquioxide of iron, left 9*66 grs. arsenic 

 acid, equivalent to 6*30 metallic arsenic, or 34*45 per cent. 

 20*04 grs. mineral afforded oxides of nickel and cobalt 7*11 

 grs., protoperoxide of manganese 0*098 gr., pyrophosphate of 

 magnesia 0*37 gr. 



II. 20*01 grs. mineral gave 0*17 silica, 0*68 free sulphur, 

 22*81 sulphate of barytes, 55*19 sesquioxide of iron with arsenic 

 acid, less 44*19 sesquioxide of iron from iron wire, sesquioxide 

 of iron 3*16 grs. 



