ike Compositio7i of Peridot. 361 



The analysis was made with 1 '8985 gr., and gave the follow- 

 ing results : Oxygen. 



Silica 42-61 22-16 



Magnesia 4-8-36 . . . 18-911 _on.oi 



Protoxide of iron 8-86 ... l-90/ ~ 



Protoxide of manganese . . 0-15 



Lime 0-22 



Alumina 0-14- 



100-34 

 According to this result the presence of an alkali could not 

 be conjectured; yet as the presence of alkali often contributes 

 to the disintegration of minerals exposed to the weather, I 

 examined the olivine in respect to this particular, and took 

 a piece, of 1-901 gr., which was a little weathered. The 

 process consisted as usual in igniting it with carbonate of 

 barytes, separating the silica, and precipitating the solution 

 with sulphuric acid and caustic ammonia. The fluid sepa- 

 rated fi-om the precipitate was evaporated in a platina capsule 

 and heated to dryness. On dissolving the dry mass a little 

 mixture of a grayish earth remained, which was separated, 

 and the solution again evaporated. The remaining white 

 salt, weighing 0-155 gr., did not fuse with a low red heat: 

 it readily dissolved in water, and by spontaneous evapora- 

 tion it produced long acicular crystals which showed no ten- 

 dency to be altered by the weather. As it hence followed 

 that the greatest part of it consisted of sulphate of magnesia, I 

 again dissolved it in water : a few spiculae of gj'psum then 

 remained behind ; and I decomposed it with acetate of barytes, 

 then evaporated the fluid which had been freed from the pre- 

 cipitate, ignited the residuum, and deprived it of salt by 

 boiling water. After the evaporation of the filtered fluid and 

 its ignition in a suitable platina crucible, its weight had in- 

 creased only about 0-0005 gr. On closer examination the 

 cause of this appeared to be some earthy stains at the bottom, 

 which wei-e not changed by water, and which, without doubt, 

 proceeded from a little residuum of magnesia. According to 

 this examination no alkali could have existed in this olivine. 



In regard to the change which the olivine suffers from the 

 influence of the weather, from the mere appearance of it, it 

 may be concluded that the protoxide of iron passes thereby 

 into peroxide. Accoiding to the analysis also, the quantity 

 of iron in it is less than in the other kinds of peridot; and on 

 the other hand there is an excess of about 2 per cent of silica, 

 which seems to proceed from a diminished proportion of mag- 

 nesia. From this it might be concluded, that after the che- 

 mical connexion of the constituent parts had been destroyed 

 Vol. 66. No. 331. Nov. 1825. Z z by 



