X'ZO Ajialyjis of the Arjen'iates 



delicate operations. If the acid is too ftrong, a great par' of 

 the fiitphiir is converted into fulphuric acid ; fo that, in either 

 Cafe, there is room for error. I thought that, to avoid this, 

 (except ill cafes where any metal which, with fulphuric acid, 

 forms an infoluble fait, was prefent,) ttrong nitric acid might 

 bo ufed, and all the lulphur ( onverted into acid. If potafli, 

 foda *, or ammonia, are ufed a? precipilants of the different 

 metals, the quantity of uilphur may be eafily afcertained. 

 I took a given weight of fulphur, and converted it into ful- 

 phuric acid by means of nitric acid. I then neutralized and 

 evaporated it. Nitrate of barytes, poured in, gave a precipi- 

 tate w?hich, in one experiment, indicated a proportion of 

 fulphur equal to 14,4, and in another to 14,6, contained in 

 one hundred of fulphatc of barvtcs. A difference fo trifling 

 need not be regarded. According to M. Lavoifier, fulphuric 

 acid contains 71 of fulphur, and 29 of oxygen; and, accord- 

 ing: to the fvnoptic tables of M. Fourcroy, fulphate of barytes 

 contains '^3 per cent, of fulphuric acid : therefore, by this 

 calculation, one hundred of fuiphaie of barytes contain 23,43 

 of fulphur, inftcad of 14,4, or 14,6. I do not pretend to 

 account for fo jvreat a diflerence in thefe refults t ; but that 

 very diftevence led me (bv exciting me to doubt thofe which 

 i had obtained, and inducing me to frequent repetition,) to 

 a more pofitive convi6lion of the proportions I have men- 

 tioned. JVI. Lavoifier obtained his proportions by combuf- 

 tion ; and, admitting even that nothing was lo(l^, it muft 

 have been rather difiicult to obtain the fulphuric acid in a 

 Itate proper to value the quantity. Indeed I do not know 

 of any direft experiments which prove, in a fatisftifitoiy 

 manner, that we have ever obtained that acid perfeftly free 

 from water; unlefs when combined with an earth, or an 

 rdkali, in fome fait, and that lalt calcined in a very ftrong 

 red heat. 



To afcertaln the quantity of metallic arfenic in mlfplckel, 

 arfenical pyrites, <kc. the moft advantageous method is, to 

 acidify it by nitric acid, and then to combine it with oxide 

 of lead. This arfeniate of lead (containing, as was before 

 faid, 33,2 per cert, of acid,) may be eftimated to contain 

 22 per cent, of metallic arlenic. If both fulphur and arlenic 

 ;:rc prefent, lead mav be equally ferviceable, after both have 

 been acidified ; for fulphate of lead is not materiaiiy folubic 

 in any acid ; wherca-<, on the contrary, arfeniate of lead is 

 very much lo. 



•■ Sec note in p-.ge 144. 



f 1 was particularly CRiitiou'; in afcertaining iliat, during tlint cxperi- 

 nifiii, no luiphurcous acid !i ad been producd, tile formation of which 

 would Uave esliiv rctounted for an\- diiVereiKe. 



When 



