BOOK VII. 245 



is first roasted in a hot fire for about six or eight hours ; next, when it has 

 cooled, it is crushed and washed ; then the concentrates made by washing 

 are again roasted, crushed, washed, dried, and weighed. The portion which 

 it has lost whilst it is being roasted and washed is taken into account, and 

 these concentrates by washing represent the cake which will be melted out 

 of the copper ore. Place three centumpondia (lesser weights) of this, mixed 

 with three centumpondia (lesser weights) each of copper scales^", saltpetre, 

 and Venetian glass, mixed, into the triangular crucible, and place it in the iron 

 hoop which is set on the hearth in front of the double bellows. Cover the crucible 

 with charcoal in such a way that nothing may fall into the ore which is to be 

 melted, and so that it may melt more quickly. At first blow a gentle blast with 

 the bellows in order that the ore may be heated gradually in the fire ; then 

 blow strongly till it melts, and the fire consumes that which has been added to 

 it, and the ore itself exudes whatever slag it possesses. Next, cool 

 the crucible which has been taken out, and when this is broken you will find 

 the copper ; weigh this, in order to ascertain how great a portion of the ore 

 the fire has consumed. Some ore is only once roasted, crushed, and washed ; 

 and of this kind of concentrates, three centumpondia (lesser weights) are 

 taken with one centumpondium each of common salt, argol and glass- 

 galls. Heat them in the triangular crucible, and when the mixture has 

 cooled a button of pure copper will be found, if the ore is rich in this metal. 

 If, however, it is less rich, a stony lump results, with which the copper is 

 intermixed ; this lump is again roasted, crushed, and, after adding stones 

 which easily melt and saltpetre, it is again melted in another crucible, and 

 there settles in the bottom of the crucible a button of pure copper. If you 

 wish to know what proportion of silver is in this copper button, melt it in a 

 cupel after adding lead. With regard to this test I will speak later. 



Those who wish to know quickly what portion of silver the copper ore 

 contains, roast the ore, crush and wash it, then mix a little yellow litharge 

 with one centumpondium (lesser weights) of the concentrates, and put the 

 mixture into a scorifier, which they place under the muffle in a hot furnace for 

 the space of half an hour. When the slag exudes, by reason of the melting force 

 which is in the litharge, they take the scorifier out ; when it has cooled, they 

 cleanse it of slag and again crush it, and with one centumpondium of it they 

 mix one and a half unciae of lead granules. They then put it into another 

 scorifier, which they place under the muffle in a hot furnace, adding to the 

 mixture a little of the powder of some one of the fluxes which cause ore to 

 melt ; when it has melted they take it out, and after it has cooled, cleanse 

 it of slag ; lastly, they heat it in the cupel till it has exhaled all of the lead, 

 and only silver remains. 



Lead ore may be assayed by this method : crush half an uncia of 

 pure lead-stone and the same quantity of the chrysocolla which they call 

 borax, mix them together, place them in a crucible, and put a glowing coal 



'"It is difficult to see why copper scales (squamae aeris — copper oxide ?) are added, 

 unless it be to collect a small ratio of copper in the ore. This additional copper is not 

 mentioned again, however. The whole of this statement is very confused. 



