OBSERVATIONS ON THE- TR?MOLI^E. £Q$ 



liad only an argillG-qnartial fubftance for their gangue, and ; re- 

 queued Mr. Chcnevix, vvlio has already rendered fuch ufeful 

 ier vices to mineralogy, to examine them by analyfis. I like- 

 wife gave him fome cryftals from amongft thofe phofphoref- 

 cent ones which I had broken,, and afterwards deprived of 

 their phofphorefcence, by digefting them for fome time in 

 nitric acid. 



My fufpicions were verified: Mr. Chenevix found only -y^ Probability that 

 of lime in each of thefe two analyfes. 3u.t what at the fame gj^^^ 

 time ftruck me, was that the tremolite taken from the phoi- lime or argil, is 

 phorefcent variety, having for its gangue a carbonate of limti tr.urfferfed m the 

 of the fpecies called the dolomi.e, which is likewife phofpho- 

 refcent, but whofe calcareous part had been taken away by 

 the nitric acid, gave by analyfis only T -J^- of argil, whilft that 

 which was taken from an abfolutely argillaceous gangue gave 

 tVz? °f ^ e ^ amc earth. Mr. Klaproth having found no argil 

 at all in the analyfis which he had before made of this fub- 

 ftance, it is probable that its exiftence in it, like that of the 

 calcareous carbonate, proceeds merely from its fimple inter-, 

 pofition. 



Thefe two observations appeared to me to be very intereft- Hence it is of 

 ing to the ftudy of mineralogy, efpecially to that part of it chemicaUna-* 

 which relates to analyfis ; fince it fliews with what care the lyfes of any mi- 

 chemift ought to avoid confounding, with the true conftituent neral Aou,d be 



o • • ... . repeated upon 



parts of a fubftance, thofe which are foreign to it, and only fpecimens taken 



interpofed between its parts. It is extremely common to find from dlfferent 

 the parts of a mineral, even in the ftate of cryftallization, en- 

 velope more or lefs of the portions of that fubftance which 

 conftitutes its gangue ; and what may likewife contribute 

 much to miftakes in this refpect, is the kind of conftancy with 

 which, (whatever may be its caufe, has hitherto been little 

 attended to) the fame fubftance, placed in fimilar circum- 

 ftances, admits this interpofed extraneous fubftance, in equal, 

 or nearly equal proportions. It is not therefore fufficient that 

 the chemift mould felecl for his analyfis, amongft the cryftals 

 of a fubftance, thofe which appear to him to be the moft pure A 

 (and the perfection of their form and tranfparency is the 

 ftrongeft prefumption which he can have in this refpect ;) but 

 he fhould alfo repeat the fame analytical procefs upon the 

 fame fubftance taken from totally different gangues. ' 



When 



