Notice of Topaz and Phenakite. 331 



It is scratched by beryl. Sp. gr.=2.S0 . . . 2.97 — ^varying in dif- 

 ferent specimens, chiefly on account of shght intermixtures of 

 grains of quartz, or albite. 



Before the blowpipe, in very thin fragments, it phosphoresces, 

 turns white, and fuses with difficulty into a milk white, semi- 

 blebby globule. With borax, it forms a transparent glass ; and with 

 the microcosmic salt, fuses into an opaque, white globule. Thir- 

 ty-three centigrammes, in the condition of an impalpable powder, 

 •after heating for fifteen minutes in a platina crucible, lost 0.75 

 centigramme. It was then mingled with three times its weight 

 of anhydrous carbonate of soda, and ignited for one hour. The 

 mixture shrunk into a somewhat firm, white, porous mass. It 

 was crushed in a mortar and treated with hydrochloric acid. A 

 perfect solution was thereby obtained, with the exception of cer- 

 tain flocculi of silica. The silica was separated and ignited: it 

 weighed 18 centigrammes, or 55.81 p. e. The solution from 

 which the silica had been separated, was precipitated with ammo- 

 nia. This clear supernatant fluid was tested for lime and mag- 

 nesia without detecting either of those earths. The precipitate, 

 while moist, was largely treated with a solution of carbonate of 

 ammonia, by which it was almost wholly taken up after a period 

 18 hours, — frequent agitation having been employed to favor the 

 solution. The undissolved residuum, which was trifling in quan- 

 tity, compared with the portion taken up, was dissolved in hydro- 

 chloric acid, and to it were added sulphuric acid and potassa. No 

 crystals of alum were formed ; from whence it was concluded 

 that the undissolved matter was chiefly glucina. Had not other 

 engagements intervened, it would have been easy to have com- 

 pleted the analysis on the portion of mineral first pulverized ; but 

 sufficient information has perhaps been gained, to aflbrd a proba- 

 ble corroboration of the opinion originally made up from its natu- 

 ral properties. 



I suspect the existence of Phenakite also at Paris, in Maine, 

 where I obtained many years ago, small, white, hexagonal prisms 

 in granite, associated with tourmaline ; and which I then sup- 

 posed to be beryl. I regret that my collection in this city does 

 not furnish me with the means of putting this conjecture to the 

 test. 



New Haven, May 29, 1838. 



