in the Pyrites, 207 



bined with fulphur 5 and in copper-ore, with a cop- 

 per-iron earth: and this gives a hint to roaft with 

 the greateft caution, not with a ftrong, brifk fire, 

 both the copper-ores for copper- proofs, as thefe 

 never want arfenic, and the cobaldifh filver-ores, 

 for filver : alfo to burn well and with the greateft 

 attention the copper-ftone > as in the one a caking, 

 or an impure filver-grain, may be expected ; in the 

 other, a great deal of arfenical impurity in the 

 black- copper, with a large fhaje of copper fpeifi, 

 or leg. Why arfenic adheres more obftinately than 

 fulphur, is, doubtlefs, owing to its being a metal, 

 at lead a femi- metal -, whereby, in its entire nature, 

 texture, and weight, it mud approach nearer to an 

 actual metal-earth, either iron or copper, than the 

 tender, light, faline, and inflammable body of ful- 

 phur. Nay, all arfenical ores let go their fugitive, 

 poifonous portion •, for inftance, cobald, whether 

 or no it yields fmalt, alfo kupffer- nickel, red-goldifh 

 ore and zwitter. Yet here we muft not omit ob- 

 ferving, as fomething very remarkable, that the 

 fulphur --pyrites, when not at all, or but little arfe- 

 nical, yields its fulphur very readily, and without 

 leaving any refidue \ though the fulphur, which is 

 along with the lead in the glitter or galena, takes a 

 longer time for feparation ; flicks ciofer (till to the 

 regulus, as is but too well known from its incine- 

 ration for making the glafs, and rather carries off 

 quickfilver along with itfelf than lets it go. 



It may be afked, whether it be not poffible fo to 

 unite a metal with arfenic, as to fublime it therewith, 

 in the fame manner as we do fulphur and quickfilver. 



And here (1.) the queflion is not concerning a 

 fmall quantity of metal, as all arfenic s may thus 

 afreet all forts of metals ? for inftance, the lead and 



the 



