28 ASSAYING when taken from the cupel proves to be pure silver, it shows at once the value of the sam- ple of ore or bullion ; but if it contains gold, as in the gold assay, the amount of gold must be found out and subtracted from the weight of the button, and the amount of each metal will then be known. To this end the alloy of these metals is separated by the process called parting, or quartation, as it is usually conducted upon an alloy made to contain at least three parts of silver to one of gold. If the silver is in larger proportion, the gold cor- net will crumble; but when of small amount compared with the gold, it is shielded by the gold from the action of the dilute nitric acid which is used to dissolve out the silver. To insure a perfect union of the gold and silver added to it, it is well to melt them with lead, and then separate the lead by cupelling. More heat may be safely applied than when silver is cupelled without gold, as the alloy of these cannot waste by volatilization. The button is hammered out, heated red-hot, and annealed, and then rolled into a thin plate, which is coiled up of the size of a quill, and called a cornet. This is put into a parting glass, and two or three times its weight of pure nitric acid is poured upon it. Some heat is applied, when red fumes of hyponitric acid are given off, and in a short time the silver is dissolved, and the gold is left, still retaining the form of the coil, but forming a brittle, spongy, brown mass. The solution of silver is poured off, and a strong acid is added to the gold, and heated to dissolve out the last traces of silver. This is poured off, and the gold is washed with hot distilled water. It is carefully taken out, put in a crucible, and heated, when it shrinks to- gether and regains its metallic lustre and the fine color of gold, with its softness and flexi- bility. Being now weighed, the process is fin- ished by the calculation of the quantity lost. The silver is recovered by precipitating it from the solution by the introduction of bright sheets of copper, for which metal the acid has a greater affinity than for the silver. It is ascer- tained that in this process the silver is never entirely taken up by the nitric acid, and that some gold is dissolved by the strong acid, as is found by preserving for years the same acid to extract the last traces of silver. The inside of the bottle containing it becomes at last coated with fine gold. This has been no- ticed in the British mint, and full 30 grains of gold have been collected from bottles thus used. Very small errors are thus involved in estimat- ing the quantities of silver and gold by this process. Assayers and metallurgists at the present time prefer what is termed the wet method, performed by the aid of acids and so- lutions, and called wet in contradistinction to the dry or furnace assay, for the determination of the amount of iron, zinc, copper, and anti- mony in the ores of these metals. The esti- mation of the amount of iron in an ore is per- formed by the aid of a solution of perman- ganate of potassium. When a solution of this salt, which is of a beautiful violet color, is added to a solution of protoxide of iron, the protoxide is immediately converted into the peroxide, and the solution loses its color. If, however, the permanganate of potassium is added with constant stirring until all the pro- toxide is converted into peroxide, and one drop too much added, that one drop will color the whole iron solution very distinctly. It is found that the same amount of iron always requires the same amount of permanganate of potassium to give the first color. The per- manganate of potassium is termed a standard solution. If then 0'2 grm. of iron is dissolved in acid (muriatic), and the standard solution added from a measuring tube, we can deter- mine the amount of solution needed for 0'2 grm. iron ; and when an ore is dissolved, and changed to protoxide by dissolving zinc in it, and the standard solution added, we obtain the amount of the solution needed for the amount of iron in the ore. And the problem is solved by this proportion : as first amount of standard is to second amount of standard, so is 0'2 grm. of iron to the amount of iron in the ore. The dry method of assaying iron ores is still used to assist the masters of iron furnaces in plan- ning the proportions of ingredients to be used in the blast furnace for the production of iron. It is based upon the same principles as the re- ducing them in the blast furnace. The oxygen with which the metal is combined must be taken up by presenting to it some substance for which it has stronger attractions than for iron, and the earthy impurities must have such substances added to them that the product of their union will be a glassy fluid, through which the globules of metallic iron can easily sink and collect together in a button. Charcoal is the substance for deoxidizing the ore in the blast furnace and in the crucible. The matters for aiding the fusion, called the flux, vary ac- cording to the earthy ingredients of the ore. The desired glassy fluid is a silicate of lime and alumina, and it may be of magnesia. If the ores already contain much silica, carbonate of lime, with the addition of some alumina or common clay, constitutes the proper flux. Ores deficient in silica require an addition of it. Some ores contain such a mixture of proper fluxing ingredients, that they melt easily without any addition of these matters. In the crucible, a little borax increases very much the fusibility of the mixture. The ore and fluxes should be thoroughly ground and mixed together, and placed in a brasqued crucible, that is, one care- fully filled and rammed with fine charcoal, moist- ened with water to a paste, and out of the top of J which a cavity is excavated for holding the as-. I say sample. The crucible is to be placed in a j wind furnace, and gradually heated for half an | hour, when the whole force of the blast is | to be applied for half an hour longer. A but- | ton of cast iron will be found in the bottom of i the crucible when it has cooled. The wet as-
