115. It sometimes happens that the precipitated gold is contaminated by lead sulphate, which cannot readily be removed in the washing, and which, does not mix with the slag proper. This cannot be skimmed off in the manner described, but is removed by means of a red-hot scorifier held in the tongs.

116. If the melted metal is pure, it appears greenish and motionless, but if any base metal remains, colored rings, or spots, are seen, moving from the center outwards. These rings are base metal oxides, and if they are supposed to be caused by iron, or copper, a few pieces of borax are thrown in, and allowed to melt while the pot is left uncovered. After a time the new slag thus formed is skimmed off, and more borax put in, and, if the metal is very impure, a little nitre is added to help the oxidation of the base metal, for if the operation takes too long a time, the pot will be cut immediately above the gold. If the impurity consists of lead, a little bone ash is better than borax to absorb the oxide, as it is formed, especially when a black lead pot is used.

117. When all is ready, the pot is seized by the tongs and the metal is poured into a cast-iron ingot mould, which has previously been warmed and smoked, or oiled. A little oil is poured on the top of the bar before it solidifies. Unless the gold is to be assayed and stamped at the works, it is not necessary to be careful about the shape or appearance of the bar. It is then as well, and saves some trouble, to allow the metal to cool in the pot, which is then broken; in any case it can be used but once. But this cannot be done if a black lead pot is used, because it will serve many times. The slag and pots are preserved, and either sold, or crushed and treated with quicksilver for the gold which they contain.

118. B. Concentrated Pyrites Containing Gold and Silver. - This material is obtained in the same manner as the concentrations containing gold only. The silver is usually of secondary importance as to its value.

Roasting

119. Roasting. This is conducted in the same manner as that for gold, but as the chlorination of the silver in the vats, by means of cold chlorine, does not give satisfactory results, it is better to form the silver chloride in the furnace, for which purpose an addition of salt is necessary. One per cent., or 20 pounds to the ton of ore, is generally enough for fifty ounces of silver. In some cases the salt is mixed with the ore when charged, but this is not always safe, as a very large loss of gold sometimes results.

120. If it is found, by means of assays, that a loss results from charging the salt with the ore, the course adopted is, to roast as directed for gold, then to let the ore cool somewhat in the furnace, and throw in the salt through one of the doors, scattering it as much as possible over the ore. The doors are then closed until the salt ceases to crackle, after which it is rapidly and vigorously mixed with the ore by means of the hoe, and the charge is drawn out within 20 minutes after putting in the salt. The heat must be high enough to cause the reaction between salt the and the metal sulphates, yet not so high as to produce heavy fumes, or blue flames; nor should a yellow substance be seen on the side of the furnace, where the ore is drawn out. The hot ore evolves copious fumes of volatile chlorides and free chlorine. It is not spread at once, but is left in the heap to be acted upon by the gases for an hour, after which it is spread on the cooling floor.

121. The best results are produced by adding to the ore, before roasting, a little more salt than the quantity required by theory to chloridise the silver present; that is, for fifty ounces of silver not more than two or three pounds of salt, which should be finely ground. The one per cent. should be added at the finish as above. The use of a little salt at the beginning in this way caused no blue flames of copper chlorides, heavy fumes, nor loss of gold, while it caused an increase in the yield of silver amounting to three or four ounces to the ton. It was not proved whether the result would have been the same if the final addition of one per cent of salt had been omitted, as it could do no harm, and cost almost nothing. But when no salt whatever was used, the result was unsatisfactory, and considerably more chlorine was required in the vats.

Ores differ much in their behavior in working, and without the cause of the difference being manifest, even from an analysis. Thus there are cases in which ten per cent of salt has been mixed with the ore before roasting, without causing a loss of gold. It therefore behooves the operator not to rely solely on the instructions given in any book, but to watch his results carefully and constantly, until he has established a satisfactory mode of procedure.

122. After roasting, the work is carried on precisely as directed for ore containing no silver, up to the point at which the gold leaching is finished, except that it is not necessary to extract the last trace of soluble gold with water, because, if a little remains, it wi!l be obtained with the silver. It is, however, proper to remark that, if the ore is of such a character that much salt must not be used at the commencement of the roasting, the lumps from the sifting should be washed, to remove soluble chlorides, before being re-roasted.