256 Mr. J. D. Smith's Experiments on 



chardson's description of sulphuret of donium, in the Records 

 of General Science of June last. This precipitate on a further 

 examination proved to be alumina, with a trace of oxide of 

 iron. This experiment, as I have before mentioned, suggested 

 the suspicion that the oxide of donium of Mr. Richardson 

 was in reality alumina and peroxide of iron. 



H. J. Brooke, Esq., having kindl}' presented me with a 

 small portion of the mineral (Davidsonite) from which the 

 oxide of donium was stated to have been obtained, I re- 

 duced about 10 grs. of it to a very fine powder: this, which 

 was of a light flesh-colour, was fused with thrice its weight 

 of a mixture of carbonate of soda and potash in a platinum 

 crucible. The fused mass was then treated with dilute hydro- 

 chloric acid, the solution evaporated to dryness, the residuum 

 dissolved in distilled water acidulated with hydrochloric acid, 

 and the silica separated by a filter, when about four fluid 

 ounces of solution were obtained. 



I then dissolved O't gr. of peroxide of iron in hydrochloric 

 acid, added it to a solution of 200 grs. of alum, and precipi- 

 tated this mixture by ammonia; this precipitate when well 

 washed was redissolved in dilute hydrochloric acid, and die 

 solution evaporated to the consistence of a syrup: a few small 

 prismatic crj'stals of a yellow colour, which deliquesced when 

 exposed to the air, were deposited : and when evaporated to 

 dryness a residue of a light yellow colour and of an astrin- 

 gent taste, at first sweetish and much resembling that of alum, 

 was obtained : this was then redissolved in a pint of distilled 

 water. H fluid ounce of this solution was diluted with about 

 •S ounces of water, and excess of ammonia added to it ; the 

 precipitate at first formed was immediately redissolved ; but 

 all further attempts again to procure this entire re-solution, 

 with another portion of the solution of alumina and iron, either 

 by pouring it into the ammonia or the ammonia into it, proved 

 abortive. The solution after standing for 24 hours wasdecanted, 

 leaving a very small quantity of a reddish brown flocculent 

 precipitate, evidently the peroxide of iron. This solution 

 having been carefully evaporated to dryness, and heated to 

 redness in a platinum capsule, a light white friable mass re- 

 mained, which, treated with hydrochloric acid, partially dis- 

 solved, and the solution when tested with hydrosulphate of 

 ammonia gave a light green precipitate. 



The solution obtained from the Davidsonite having been 

 poured into a large excess of ammonia, gave a gelatinous pre- 

 cipitate; this was frequently shaken, and at the expiration of 

 two or three days the supernatant liquor was carefully de- 

 canted. Afresh portion of ammonia was then added, and the 



