Gold may also be separated from solutions which contain no cyanide, by treatment with an excess of protochloride of tin, which produces a precipitate of purple of Cassius (stannate of gold), from which the gold is easily reduced by heat to the metallic state. Sulphurous acid will also reduce the gold. But in the latter case, the liquid should be heated.

Granulated gold is obtained by running the molten metal in a thin stream, from a certain height, into a large quantity of cold water.

Recovery of Silver. (Roseleur's Methods.)

Old solutions which contain silver in the form of a simple salt are very easily treated. It is sufficient to add to them, in excess, a solution of common salt or hydrochloric acid, when all the silver will be precipitated in the state of chloride of silver, which, after washing, may be em ployed for the preparation of new baths, or reduced to the metallic form by one of the processes about to be de scribed. Solutions of nitrate of silver, desilverizing acids, etc., fall in this category.

Common salt, on the other hand, is without action upon solutions containing silver in the state of a double salt, and will rather facilitate solution than precipitation. The double tartrate of potassium and silver (whitening bath), the double sulphite of sodium and silver (immersion bath), etc., belong to this category. Before employing common salt with such solutions, the liquids should be strongly acidulated with sulphuric acid, which restores the silver to the state of a simple salt, readily precipitated by common salt. Hydrochloric acid, however, precipitates silver well from these solutions.

Solutions which contain silver, as cyanide, form another exception; and in order to recover all the metal, we must

again have recourse to the process employed for similar combinations of gold; that is to say, evaporation to dryness, and reduction of the mass in a crucible, with an addition of carbonate of sodium and powdered charcoal. The silver is then obtained as a fused button at the bottom of the crucible. (See Comments at close of this chapter.) When the silver has been obtained in the state of the chloride of silver, it may be reduced to the metallic state by many processes, the principal of which are:

1. Well-washed chloride of silver (water does not dissolve a trace of it) is placed in a stoneware dish with two or three times its weight of zinc, and the whole is covered with water strongly acidulated with sulphuric acid. As soon as they are in contact, these substances react upon each other: the sulphuric acid and the zinc decompose the water into its elements, the oxygen of which oxidizes the zinc, which then combines with the acid to form sulphate of zinc (white vitriol or white copperas), a very soluble salt. On the other hand, the hydrogen seizes upon the chlorine of the silver, transforming it into hydrochloric acid, also very soluble in water. The silver is left behind in the form of an impalpable powder, and when all the zinc is dissolved the contents of the dish may be filtered. The silver, which remains upon the filter, must be thoroughly washed with pure water, and may then be dissolved in pure nitric acid to form a pure nitrate of silver.

2. The chloride of silver, freed from foreign metallic salts by washing, is mixed with four times its weight of crystallized carbonate of sodium, and half its weight of pulverized charcoal. The whole is made into a homogeneous paste, which is thoroughly dried, and then introduced into a strongly heated crucible. When all the material has been introduced the heat is raised, to promote

complete fusion, and to facilitate the collection of the separate globules of silver into a single button at the bottom of the crucible, where it will be found after cooling.

If granulated silver is wanted, pour the metal in a thin stream, and from a certain height, into a large volume of water, as in the case of gold.

This method is almost exclusively employed in electrogilding and silvering works.

During this operation, the carbonic acid of the carbonate of sodium having disappeared, the oxygen of the soda unites with the carbon of the charcoal forming gaseous products, which escape. In its turn, the sodium combines with the chlorine of the chloride of silver, forming chloride of sodium, while the metallic silver is set free. The fact is worthy of remark, that the chloride of sodium, which is decomposed by a silver salt in aqueous solution, is reconstructed by igneous fusion in the presence of

chloride of silver.

3. Lastly, we may employ in many cases a plate or sheet of copper for separating silver from its salts; but it is quite difficult to free the deposited metal entirely from traces of copper.

Recovery of Nickel.

To recover nickel from old solutions, Urquhart proposes. to (Electro-Plating, 165) take advantage of the observation of Unwin (see Chapter XLV., p. 376), that the double sulphate of nickel and ammonium is almost insoluble in a concentrated solution of sulphate of ammonium.

He makes up a saturated solution of ammonium sulphate in warm water, and adds it to the old nickel-plating solu tion, with constant stirring, and after the lapse of a few minutes, a granular precipitate of the double sulphate will begin to separate. The addition of ammonium sulphate

should be continued, from time to time, until the liquid is colorless. The precipitated salt is very pure, and may be used directly in making a new bath.

Recovery of Platinum. (Roseleur.)

This process is very simple, and consists in first acidulating with hydrochloric acid any solution of platinum, unless it be already acid, and then immersing in it for some hours a well cleansed bar of iron. The platinum is reduced in the form of a black powder, which is washed, and then calcined to a white heat. By dissolving it in aqua regia we reconstitute the chloride of platinum necessary for the preparation of the platinizing baths.

Platinum solutions may also be reduced by evaporating them to dryness, and strongly calcining the residue, which is then washed upon a filter in order to remove the soluble salts, and again submitted to a white heat. The platinum thus obtained is soluble in aqua regia.

Recovery of Copper. (Roseleur.)

In works where great quantities of copper are operated upon, it is advantageous to recover the metal dissolved in the cleansing baths, which are allowed to go to waste with the rinsing water, by the majority of gilders, silver electro-platers, and galvanoplastic operators. The recovery of such copper is an easy and inexpensive process. All the liquids holding copper are collected in a large cask filled with wrought or cast-iron scraps. By the contact of the copper solution with the iron, a chemical reaction immediately takes place, by which the iron is substituted for the copper to make a soluble salt, while the copper falls to the bottom of the cask as a brown powder. The cask should be sufficiently large to hold all the

liquids employed in a day's work. The liquids are decanted every morning.

The old iron scrap is generally suspended in a willow basket near the top of the liquid, and, by occasionally moving it about in the liquid, the metallic powder of copper alone falls to the bottom of the cask.

The same method is employed for recovering the copper from old cleansing acids, or from spent galvanoplastic baths. The copper thus obtained is quite pure, and, by calcining it in contact with the air, a black oxide of cop per is obtained which is serviceable for enriching and neu tralizing galvanoplastic baths too strongly acidified.

Utilization of Other Wastes. (Roseleur.)

Nothing whatever should be allowed to go to waste in well-conducted works. Sweepings, sawdust, residues from the bottoms of scratch-brushing tubs, filters, papers, rags, etc., should be carefully collected, mixed, and burned in a furnace adapted for the purpose. The resulting ashes are finely pulverized, sifted, and sold to certain operators called "sweep smelters," who extract the precious metals from them.

The method in vogue in France, as described by Roseleur, and which is a very equitable one, is as follows:

The purchaser is called in, and, after having thoroughly mixed the heaps of ashes, several samples are taken with a sampling-gauge, which is a rod hollowed out laterally. The various samples are then mixed, and divided into two parts, one for the buyer and the other for the seller, each of whom has one assayed for gold and silver. From the comparison of the two assay certificates, the buyer takes the lot at its bullion value, after deducting a certain sum for the expense of the operation and his profit.

The sweep-smelter, in turn, pulverizes and sifts the