Memoir on Nickel. 65 



black from this solution by sulphurated hydrogen and by 

 water in white flakes, which the nitric acid could then dis- 

 solve, and with which they gave by evaporation crystals that 

 could readily be distinguished to be nitrate of bismuth. This 

 substance, therefore, was oxide of bismuth^ united to an 

 acid which rendered it proof against the Common sol- 

 vents. I prcsumc^d that it mio;ht be arsenic add ; and in- 

 deed mv conjectures were confirmed by ulterior trials. Ar- 

 seniate of bismuth, made in a direct manner, exhibits ex* 

 actly the same phoenomena. 



Experiment II. 



The liquor of the first experiment, arising front the ac- 

 tion of the nitric acid on the roasted ore, and which was 

 of a beautiful green colour, was evaporated, and then diluted 

 with water to precipitate the bisinuth \Vhich might be found 

 in it. There was no appearance of its being turbid. The 

 whole bismuth, therefore, had been separated by the arsenic 

 acid. It contained copper. Iron immersed in it imme- 

 diately detected its presence. Having added sulphurated^ hy* 

 drogen, thiv. metal alone was precipitated in the state of hy- 

 drogenalcd sulphuret, under the form offtakes, of a chest- 

 nut-brown colour. It was but little in quantity in the ore, 

 and made at most the l-50th part of it. 



Experiment IIL 

 The copper having by these means been totally precipi- 

 tated from the nitric solution, since it no longer coloured 

 iron, I supersaturated it with caustic potash in order to 

 obtain the oxides pure, and to take from them the arsenic 

 acid which they might retain ; but as this method was not 

 attended with complete success, I was obliged to have re- 

 roiirsc to another. The one I preierred was as follows : — 

 1 again effected a solution of the oxides in nitric acid, and 

 poured into ihmx an excess of hydro-sulphuret of potash. 

 Tlie decoinposiiion was complete : the arsenic acid remained 

 in the liquor combinx-d with the potash, while the oxides 

 were deposited in combination with sulphur and sulphurated 

 hydrogen in the state of hydrogenated sulpiiurct, and under 

 tile form of black flakes.' This method of separating the 

 arsenic is so sure, that it induced me to propose it for de- 

 termining the quantity of that metal in any ore whatever. 

 I employed, it with much success in the trexitment of arseni- 

 cal ore of cobalt. I separated from it all the cobalt and ar- 

 senic : by treating only with nitric acid and alkalies, the 

 greater part <jf these tv/o metals rtMnain united together in 

 Vol. 20. No. 77. Oct. iHUi. K " the 



