The largest item in this excessive cost of extraction is for the vessels in which the Na2CO, is heated in admixture with powdered coal. It has not hitherto been found possible to heat the mixture of Na2CO3 and powdered coal to the necessary temperature except in cylindrical wrought-iron vessels of very small diameter, and these small wrought-iron cylinders are so rapidly destroyed that their cost stands for fully 1/2 of the present total cost of sodium, and for nearly 1/2 of the present total cost of aluminium.

There is surely room here for improvement, and the cost of sodium for this item will doubtless be diminished. Still, one can hardly hope that sodium will ever become cheap enough to permit of aluminium reduced by sodium being largely applied to the many practical uses of which aluminium is capable. A cheaper reducing agent than sodium, and a cheaper artificial are of aluminium than Al2Cl6,2NaCl: these, Wel-don thinks, are essential to aluminium becoming commonly and extensively employed.

J. Morris, of Uddington, near Glasgow, claims to obtain aluminium by treating an intimate mixture of alumina and charcoal with carbon dioxide. For this purpose a solution of aluminium chloride is mixed with powdered wood-charcoal and lamp-black, then evaporated until it forms a viscous mass, which is shaped into balls. During the evaporation, hydrochloric acid is given off. The residue consists of alumina intimately mixed with charcoal. The balls are dried, then treated with steam in appropriate vessels for the purpose of driving off all the chlorine, care being taken to keep the temperature so high that the steam is not condensed. Now the temperature is raised, so that in the dark the tubes are seen to be at low-red heat, and dry carbon dioxide is passed through. This is said to be reduced by the charcoal to carbon monoxide, which now, as affirmed by Morris, reduces the alumina to aluminium. Although the quantity of the escaping carbon monoxide is in general a good indication of the progress of the reduction, it' is nevertheless advisable not to continue the heating of the tubes or vessels until the evolution of this gas has ceased or even nearly ceased, as in consequence of slight differences in the consistence of the balls, some of them give up all their carbon sooner than the others.

The treatment of the balls with carbon dioxide for the purpose of the reduction lasts about 30 hours, when the substances are mixed in the proportions of 5 parts carbon to 4 alumina. The metal appears as a porous, spongy mass. It is. freed from the residual alumina and, particles of charcoal by fusion and mechanical treatment, and then poured into moulds.

Niewerth, of Hanover, suggests a novel process for reducing aluminium First, ferro-silicon is mixed with alu--minium fluoride in proper proportions) and the mixture is submitted to red or melting heat, when the charge is decomposed into volatile silicon fluoride, iron, and aluminium, the 2 latter forming an alloy. To obtain copper-aluminium alloy from this iron alloy, the latter is melted with metallic copper; the copper by reason of its greater affinity, unites with the aluminium, while the iron retains but an insignificant amount of aluminium. On the mass cooling, the copper-bronze and iron separate out in such a manner that both bodies can be readily isolated. In place of the pure aluminium fluoride, cryolite, which occurs in nature, may be advantageously employed, or aluminium chloride may. be used. In the latter case, silicon chloride and iron-aluminium alloy are formed. Or, pure silicon and aluminium fluoride, or cryolite, or aluminium chloride may be used, when pure aluminium is obtained.

In the second process, the compound of aluminium with chlorine or fluorine decomposed by any means into a volatile state is brought, strongly heated, into contact with a mixture of 62 parts soda carbonate, 28 of coal, and 10 of chalk, also in a highly heated condition. From this mixture, sodium is disengaged, and this reduces the gaseous aluminium fluoride or chloride in such a manner that the nascent sodium generates tree aluminium from the fluoride or chloride as the case may be. In place of the above mixture, others which produce sodium may be employed, or those from which potassium is formed. In another process, Niewerth's newly invented furnace may be employed, but the process may also be carried on in a crucible or other furnace. The furnace in question consists of 3 shaft furnaces made of fire-resisting material, 2 of which shut by means of some contrivance, e.g.t a convex cast-iron cover. These furnaces communicate with the other furnace by channels, which can be closed by slides. Blast pipes are fitted to 2 of the furnaces, and all 3 are fitted with discharge apertures. Short tubes connected with a steam reservoir admit steam to the 2 furnaces, which are filled with some suitable fuel, e. g., coke, and 'by the admission of the blast are blown very hot.

The covers ore meanwhile lifted up. The middle furnace is then charged with 3 charges in proper order. The first charge consists of a mixture of soda carbonate (NaO,C02) + Carbon (C) + Sulphur (S) + Alumina (A1203). The second charge is alumina sulphate, the third charge a flux, preferably a mixture of soda and potash chlorides. This furnace must be strongly heated at the beginning of the operation; it is best to fill with coke first, and as soon as that is warm to put the charge on the coke so that it comes to the bottom with the burning coke. A mixing of the charges with coal is usually not necessary.

The process then continues as follows:- The cover of one furnace is shut down, the slide drawn up, and the blast is cut off. A suitable quantity of steam is now admitted, which spreads itself over the glowing coke and penetrates downwards through it, breaking up into its constituent parts oxygen and hydrogen. The oxygen forms with the carbon of the coke carbonic oxide gas, while the hydrogen remains uncombined. The gases thus formed during their passage through the extremely hot coke themselves acquire a very high temperature, and at length pass by a channel into the third furnace where the charge lies. The highly heated gases, carbonic oxide and hydrogen, act upon the charge, so that the first charge breaks up into a. combination of sodium sulphide and aluminium sulphide, from which, by means of the second charge (alumina sulphate), free metallic aluminium is formed. Passing into the melting zone, the aluminium will melt, and, if it be drawn off, the flux added thereto will assist the fusion of the aluminium together; it is not, however, absolutely necessary.

When the gases generated in the first furnace are too cool, the second furnace is closed by the cover, its coke having been kept hot by the blast, and the processes are repeated.

Instead of the bimetallic sulphide, pure aluminium sulphide may be employed, or a mixture from which it is generated, or, again, pure soda sulphide, potash sulphide, copper sulphide, or other metallic sulphides or analogous compounds producing the same effect alone. In the latter case, aluminium is obtained in combination with the metal of the sulphide. Instead of the one charge alumina + soda carbonate + sulphur + coal, a mixture of alumina, sulphur, and coal only may be introduced; or further, the alumina sulphate may be replaced by alumina. The one charge may also be formed out of sodium sulphide, potassium sulphide, and other metallic sulphides, and the others may be produced from alumina or alumina sulphate.