Dr. Percy on the Composition of Beudantife. 163 



11. By Reinsch's test (electrotype copper being used), so- 

 lution (6.) gave ample evidence of the presence of arsenic. 



12. The mineral was fused with a mixture of carbonate of 

 potash and soda. The product was washed with boiling 

 water, when a clear colourless solution and a brown residue 

 were obtained. 



13. Reinsch's test applied to solution (12.) detected arsenic; 

 by heat in an inclined open tube, the gray coating was re- 

 moved, and a white, sparkling, crystalline sublimate obtained. 



14. Marsh's test applied to solution (12.) also gave unequi- 

 vocal evidence of the presence of arsenic in considerable 

 quantity. 



15. Solution (12.) was rendered acid by hydrochloric acid, 

 and treated with sulphuretted hydrogen. The precipitate was 

 digested with the strongest nitric acid ; the solution was filtered, 

 reduced by evaporation, and treated in Marsh's apparatus, as 

 modified by the Prussian government (Chem. Gaz. vol. iii. 

 p. 46), when an abundant deposit of metallic arsenic was ob- 

 tained. This was converted into white sublimate, which was 

 rendered yellow by sulphuretted hydrogen. 



16. To the solution (15.) left, after separation of sulphuret 

 and free sulphur, and reduced by evaporation, hydrochlorate 

 of ammonia, and sulphate of magnesia were added. Alter some 

 time numerous crystalline particles appeared on the surface 

 of the liquid, and on the sides and bottom of the glass vessel. 

 This crystalline precipitate, tested with molybdate of ammonia 

 (Svanberg's process), gave a yellow precipitate. 



17. The solution of the mineral in hydrochloric acid was 

 treated with excess of ammonia and then digested with hydro- 

 sulphate of ammonia. The precipitate was treated with strong 

 nitric acid. Sulphuretted hydrogen was passed through the 

 diluted acid solution, which was afterwards filtered and boiled 

 with nitric acid. Excess of ammonia was added. The pre- 

 cipitate was heated with carbonate of soda and nitrate of 

 potass on platinum before the blowpipe, but no green colo- 

 ration was produced. 



18. The precipitate by ammonia (17.) was boiled with 

 potash. The potash solution was rendered acid by hydro- 

 chloric acid, and afterwards alkaline by ammonia, but no 

 sensible precipitate followed. 



19. The brown residue (12.) was dissolved in hydrochloric 

 acid. Sulphuretted hydrogen was passed through the solu- 

 tion, which was afterwards filtered and boiled with chlorate 

 of potash. Ammonia was added. The filtrate was tested with 

 oxalate of ammonia, but no turbidity followed, 



20. The filtrate (19.) was tested with phosphate of soda, 



M 2 



