as 20 to 40 cents per ton. A part of this effluent solution, just sufficient in quantity to provide barren washes for the ore residues was diverted and passed through one or more boxes of zinc shavings to strip the remaining assay values and produce a strictly barren solution. By this means an unusually low consumption of zinc dust was produced and the resulting precipitate was considerably higher in value than it would otherwise have been. The cost of maintaining the box of zinc shavings was small as only part of the daily solution tonnage went through it and it needed very little attention on account of the small quantity of precious metal deposited therein. In recent years, however, the zinc dust process has been so developed and improved that it is now hardly worth while going to the additional trouble of the double system.

There are several advantages in the zinc dust method. (1) Zinc dust costs less per pound than zinc shavings. (2) Less zinc is usually consumed per unit of fine bullion than in the case of zinc shavings. (3) It is superior to shavings for precipitating coppery solutions because it does not afford the same opportunity for the copper to form an impervious skin over the zinc surface and stop further action by the solution. (4) The laborious and disagreeable task of dressing and cleaning up the zinc boxes is avoided, and (5) greater security from pilfering the precipitate is effected.

With clean solutions high in silver it is possible, even when the El Rayo system is not used, to produce bullion which by direct melting assays over 900 fine with a consumption of zinc dust not over one part to 1 of fine bullion, by weight; indeed in some of the Pachuca plants the zinc dust consumption has been less than 0.8 kilo per kilo of fine bullion.

Electrolytic Precipitation. This was introduced on a commercial scale as an adjunct to the cyanide process on the Rand for precipitating the weak and low-grade solutions resulting from the treatment of slime by decantation, and was known as the Siemens Halske process. The anodes were of sheet iron enclosed in burlap to prevent short circuiting and the cathodes were formed by

hanging four or five sheets of lead foil cut into strips upon a stiff wire, and shaking them till the strips separated and formed a bunchy mass. The anodes were placed about four inches apart and the cathodes occupied the entire space between. It will be noted that the cathode area was four or five times greater than that of the anode. The current density commonly used was 0.04 ampere per square foot of anode surface and about 16 cu. ft. of box capacity were required per ton of solution per 24 hours. The gold was deposited as a firmly adherent film or plating and was recovered each month by melting down the cathodes and cupelling the resulting pigs of lead. The litharge derived from this operation was reduced in a reverberatory furnace at the mine and sold to the Rand Central Ore Reduction Company who treated it by Park's process and ther rolled the softened and barren lead into foil which was sold back to the mine companies.

The anodes slowly dissolved with formation of ferrous hydroxide and a little prussian blue, forming a sludge which often assayed high in gold. The sludge was either sold to the Rand Central Company or roasted and treated by cyanide at the mine. The process had its drawbacks, the chief of which were an uncertainty and variability in the results obtained, a lack of simplicity, and the loss and expense incidental to the handling of various byproducts, and it was gradually supplanted by the development of the zinc lead couple.

At the Minas Prietas plant of Charles Butters and Company for the treatment of pan amalgamation tailing containing copper, H. T. Durant introduced an important modification of the South African practice. Owing to the large amount of copper that was deposited with the gold and silver the cupellation of the lead foil cathodes was impossible, so he conceived the idea of raising the current density sufficiently to cause the whole of the precipitated gold, silver and copper to be deposited as a nonadherent sludge which dropped off and fell to the bottom of the box from which it was removed each month in a manner similar to that used in cleaning up a zinc box. The detachment of the deposit from the cathode was assisted by periodically wiping

down the plates with rubber-edged brushes without removing them from the box.

The anodes were made of lead sheets 3/16 in. thick, peroxidized by being placed in a bath of 1% potassium permanganate and subjected to the action of the current for one hour at a current density of about one ampere per square foot. Owing to the use of graphite in the rolling of the lead sheets it was found necessary to scour off the surface film so as to expose a clean metallic surface to the action of the solution. These plates had a life of from twelve to eighteen months under the conditions to which they were subjected. They were placed in the compartments parallel with the flow of solution and about seven inches apart from centre to centre. The cathodes which were made of commercial tin plate were set in the centres of the spaces between the anodes so that the distance between anode and cathode was from 3 to 31⁄2 inches. Both electrodes were soldered to their respective copper leads and remained permanently in position. The current density varied in the different boxes from 0.25 to 0.56 amperes per square foot and the voltage across the electrodes was from 2.5 to 3.

There were six boxes, each 30 feet long and 10 feet wide, divided up into twelve compartments with baffle boards similar to those used for zinc precipitation. Each compartment contained 18 anodes and 17 cathodes. Four of the boxes were used for weak solution and two for strong.

WORKING RESULTS

Weak Solution.-Current density, 0.25 ampere per square foot; daily quantity of solution, 480 tons, total anode area 13,536 sq.

ft.

Strong Solution.-Current density 0.55 ampere per square

foot; daily quantity of solution, 216 tons.

Regeneration of Cyanide in Electrolytic Precipitation.Theoretically it should be possible to regenerate a part at least of the cyanide combined with the metals dissolved from the ore but under the usual conditions obtaining in practice there is also a tendency to oxidize the free cyanide present, so that the loss and gain about balance one another, and in electrolyzing solution containing only the usual amounts of gold and silver little or no net gain in cyanide is observed. When copper is also present, however, the case is for some reason different and a marked rise in cyanide strength is noticed at the tail of the boxes. At Prietas this gain averaged for strong solution 0.0146% KCN or 0.292 lb. per ton of solution and for the weak solution 0.01189% or 0.236 lb. per ton. The actual total cyanide consumption in the plant was from 2.5 to 3 lb. per ton of ore, and the amount regenerated in precipitation worked out at 0.58 lb. KCN per ton of ore treated.

This Minas Prietas method is likely to prove useful when treating ores containing copper soluble in cyanide solution, both on account of the recovery of some of the cyanide and also of the avoidance of difficulties caused by the copper in zinc precipitation. Its chief drawbacks are (1) the low efficiency of the current. The best efficiency shown was only about 14% and the worst was 5%. (2) The large cubic capacity and floor space required. (3) The failure to regenerate all the combined cyanide. In treating clean silver ores no regeneration whatever is apparent and even in the case of the coppery tailing at Minas Prietas it was only about 1/4 of the theoretically possible amount. (4) The disintegration of the anodes and cost of replacement. This perhaps may be overcome by the use of fused magnetite or passive iron.

Some interesting work in the development of electrolytic

precipitation is detailed in a paper by Hugh Rose (Bulletin A.I.M.E., Aug., 1916), the experiments being carried out by A. W. Hahn at the Santa Gertrudis Mill, Pachuca, Mexico. The first noteworthy point is that an anode composed of lead alloyed with 6 to 9% antimony was found to give better service than any material heretofore tried. And the second point is that in investigating the destruction of free cyanide by the action of the current such destruction was located at the anode and it was found that this could be largely obviated by maintaining a high concentration of protective alkali in the neighborhood of the anode. A porous diaphragm such as canvas separating the anode and cathode assisted in maintaining an alkaline anolyte, circulation of anolyte and catholyte being accomplished individually. In these experiments which were made principally on barren solution with a view to a regeneration of the cyanide, a substantial recovery of cyanide was shown, the cathode deposit consisting chiefly of zinc. It would seem, however, that if electrolysis is to be used its proper application would be in recovering the gold and silver from solution instead of in removing the by-products of a previous chemical precipitation. But however that may be the data obtained in this investigation are highly suggestive and might well be made the starting point for further research.

Regarding electrolytic precipitation in general, its application in the present stage of its development has a very limited utility in the cyanide process and it has hitherto been unable to compete with other methods. It would seem, however, to offer the most reasonable and scientific process for dealing with the metal in cyanide solutions and it is not unlikely that with further study and research ways may be found to surmount the present difficulties and drawbacks.

Aluminium Precipitation. The use of aluminium as a precipitant for gold out of cyanide solutions was patented by Moldenhauer in 1893 but he does not appear to have made any practical use of the process. H. F. Julian experimented with it, in the same year, but soon abandoned the idea. In 1910 S. F. Kirk