

1824.] Mr. Children's Chemical Examination of Roselite. 441 



Heuland. Its great resemblance with the arseniate cobalt from 

 the same locality had hitherto caused its being placed with it. 



Chemical Examination of Roselite. By J. G. Children, FRS. 



In glass matrass, decrepitates and gives off water ; the fine 

 deep rose colour changes to black. 



With borax, on the platina wire, and in the oxidating flame, 

 the assay dissolves readily, and gives an intensely deep blue 

 glass. In the reducing flame, the colour becomes lighter ; no 

 appearance of reduced copper. 



With salt of phosphorus on the platina wire, the assay dissolves 

 readily and completely, and gives results similar in both flames 

 to those with borax. 



The assay dissolves with facility in muriatic acid, and, after 

 evaporation to dryness, the residuum is wholly soluble in water. 



A minute fragment digested in a solution of caustic potash, 

 on a slip of glass, evaporated to dryness, redissolved, and the 

 alkali neutralized with nitric acid, gave with nitrate of silver 

 and ammonia, a brown red precipitate of arseniate of silver. 



Another minute fragment gave with a drop of muriatic acid a 

 fine blue solution ; by dilution with water, the colour disappeared. 

 A drop of the diluted solution gave an abundant precipitate with 

 oxalate of ammonia. 



Another drop, evaporated to dryness on a polished steel 

 blade, left no trace of copper. 



Another drop gave with prussiate of potash a yellowish green 

 tint, without any indication of copper. 



Another drop, treated with bicarbonate of ammonia and 

 phosphate of soda, gave decided evidence of the presence of 

 magnesia. 



These experiments are sufficient to show, that the composition 

 of Roselite consists of arsenic acid, united to oxide of cobalt, 

 lime and magnesia, elements which, according to Phillips 

 (Mineralogy, p. 178), constitute the Picropharmacolite of Stro- 

 meyer, who found their proportions to be : 



Lime 24-64 



Magnesia 3*2 1 



Arsenic acid 46-97 



Oxide of cobalt 0-99 



Water 23-97 



99-78 



The whole quantity of Roselite that M. Levy could afford me 

 for my experiments, consisted of three or four minute crystals, 

 about the size of a small pin's head, so that any attempt to ascer- 

 tain the relative quantities of the ingredients would have been 



