396 BOOK IX. 



the cadmia ; this sweeping is done twice a year. The soot mixed with 

 pompholyx and the cadmia, being chipped off, is thrown down through 

 a long chute made of four boards joined in the shape of a rectangle, 

 that they should not fly away. They fall on to the floor, and are sprinkled 

 with salt water, and are again smelted with ore and litharge, and become 

 an emolument to the proprietors. Such chambers, which catch the metallic 

 substances that rise with the fumes, are profitable for all metalliferous 

 ores ; but especially for the minute metallic particles collected by washing 

 crushed ores and rock, because these usually fly out with the fire of the 

 furnaces. 



I have explained the four general methods of smelting ores ; now I 

 will state how the ores of each metal are smelted, or how the metal is obtained 

 from the ore. I will begin with gold. Its sand, the concentrates from 

 washing, or the gold dust collected in any other manner, should very often 

 not be smelted, but should be mixed with quicksilver and washed with tepid 

 water, so that all the impurities may be eliminated. This method I ex- 

 plained in Book VII. Or they are placed in the aqua which separates 

 gold from silver, for this also separates its impurities. In this method we 

 see the gold sink in the glass ampulla, and after aU the aqua has been drained 

 from the particles, it frequently remains as a gold-coloured residue at the 

 bottom ; this powder, when it has been moistened with oil made from 

 argoP'', is then dried and placed in a crucible, where it is melted with borax 

 or with saltpetre and salt ; or the same very fine dust is thrown into molten 

 silver, which absorbs it, and from this it .is again parted by aqua valens-^. 



It is necessary to smelt gold ore either outside the blast furnace in a 

 crucible, or inside the blast furnace ; in the former case a small charge of ore 

 is used, in the latter a large charge of it. Rudis gold, of whatever colour 

 it is, is crushed with a libra each of sulphur and salt, a third of a libra of copper, 



" stuff flies up with the smoke to the upper chamber, and adheres to the walls of the roof. 

 " The substance which is thus formed has at first the appearance of bubbles on water, after- 

 " ward increasing in size, it looks like skeins of wool. The heaviest parts settle in the bottom, 

 " while some fall over and around the furnaces, and some lie on the floor of the building. 

 " This latter part is considered inferior, as it contains a lot of earth and becomes full of dirt." 



Pliny (xxxiv, 33) appears somewhat confused as to the difference between the 

 two species : " That which is colled pompholyx and spodos is found in the copper-smelting 

 " furnaces, the difference between them being that pompholyx is separated by washing, while 

 " spodos is not washed. Some have called that which is white and very light pompholyx, and 

 " it is the soot of copper and cadmia ; whereas spodos is darker and heavier. It is scraped 

 " from the walls of the furnace, and is mixed with particles of metal, and sometimes with 

 " charcoal." (xx.xiv, 34.) " The Cyprian spodos is the best. It is formed by fusing 

 " cadmia with copper ore. This being the hghtest part of the metal, it flies up in the fumes 

 " from the furnace, and adheres to the roof, being distinguished from the soot by its whiteness. 

 " That which is less white is immature from the furnace, and it is this which some call ' pom- 

 " pholyx.' " Agricola {De Natiira Fossilinm, p. 3=50) traverses much the same ground as the 

 authors previously quoted, and especially recommends the pompholyx produced when making 

 brass by melting alternate layers of copper and calamine (cadmia fossilis). 



^''Oleo, ex fcce vini sicca confecio. This oil, made from argol, is probablj' the 

 same substance mentioned a few lines further on as " wine," distilled by heating argol in a 

 retort. Still further on, salt made from argol is mentioned. It must be borne in mind that this 

 argol was crude tartrates from wine vats, and probably contained a good deal of organic 

 matter. Heating argol sufficiently would form potash, but that the distillation product could 

 be anything effective it is difficult to see. 



'^Aqua valens. No doubt mainly nitric acid, the preparation of which is explained 

 at length in Book X, p. 430)- 



