45 Anahfis of the Spinel Ruby. 



of jaflUming different colours according to the quantities- of 

 oxygen which it ahforbs ; that when it is faturated with this 

 principle, it is red and acid j and when it contains lets, it is 

 green, and in the ftdte of an oxyd, we mall readily conceive 

 how this oxydated metal may colour the emerald and the 

 ruby. It follows then from thefe considerations, that chrome 

 exifts in the emerald in the ftate of an oxyd, and in the ruby 

 in the ftate of an acid; and that this acid is there, doubtlefs, 

 in a falinc combination with alumine or magnelia, and per- 

 haps with both at the fame time. With regard to the pro- 

 portion of the chromic acid in the ruby, I have not been able 

 to determine it very accurately, on account of the affinity 

 which it has with alumine, from which it is difficult to fe- 

 parate it completely. I am however of opinion that it may 

 be eftimated, without committing any very fenfible error, at 

 between 5 and 6 parts in 100 of ruby. 



XII. I now return to the folution of alumine in cauftic 

 potaih (Exp. VII,). I fuper-faturated it with the muriatic 

 acid, and precipitated it afterwards by the carbonat of am- 

 monia. The depofit warned and kept at a red heat for a con- 

 fiderable time weighed 85 parts. This matter had all the 

 propert : t. of alumine. As I had however fcarcely found any 

 filiceous earth during the courfeof this analyfis, and as Klap- 

 roth fays that he.obtained 16 in 100 parts of this (tone, I 

 wifhed to allure m\ fell" whether there might not be fome of 

 it remaining in the alumine. For that purpofe I dilTolved 

 the 82 parts above-mentioned in the fulphuric acid, and I 

 indeed obtained an infoluble rcfuluam which weighed three 

 parts, and which was filiceous earth. This, with the i~ 

 part obtained (2xp. X.), makes 4) parts. But it mull be 

 recollected, that during the pulverifation of the 100 parts of 

 rubv, the details of the analyfis of which I have explained, 

 5 parts were taken from the mortar of filex ; from which it 

 follows, that the 4^ parts of this fubftance found in the courfe 

 of thefe operations did not belong -to the ruby. Whatever 



method/ 



