This metal is widely distributed in some of the common rocks ;, e.g. as carbonate in dolomite and magnesite, as sulphate in kainite and kieserite, and in many saline springs; and as chloride in car-nallite and sea-water. The metal was first obtained in a compact form by fusing the anhydrous chloride with potassium. Later, Bunsen prepared it by electrolysing the fused chloride in a porcelain crucible, centrally divided by a vertical partition formed of the cover of a similar crucible ground to fit; hanging over this partition is a support carrying 2 poles of retort carbon, the negative one being notched for the reception of the deposited metal, which, being lighter than the fused chloride, would otherwise reach the surface and ignite. Matthiessen avoids the difficulty of preparing the pure chloride by using a mixture of 3 molecules of potassium chloride and 4 of magnesium chloride, together with some ammonium chloride, fusing the mass, and electrolysing without any notches in the negative pole, they being unnecessary here, as the metal is heavier than the fused mass.

For preparing magnesium on the large scale, Caron et Deville's method is mostly used. It is as follows:- A mixture is made of 60 parts fused magnesium chloride, 48 of finely powdered fluorspar, and 23 of. sodium. The sodium is cut into little fragments after being carefully freed from-oxide and any petroleum adhering to the surface. A porcelain crucible is heated to- redness in a furnace, and into it the mixture is thrown; the cover is replaced and held firmly down during the violent reaction which soon takes place. When the crepitations have ceased, the crucible is opened, and discovers a very fluid mass consisting of undecomposed magnesium chloride, of salt resulting from the- reaction of the sodium upon a part of the magnesium chloride used in excess, and of calcium- fluoride; no particle of mag-nesium is yet visible, it being disseminated in-microscopic globules in the slag. The portions which have attached themselves to the upper part of the crucible are stirred up with; the remainder by a clean iron rod. - The crucible is drawn from the fire, and left to cool for some minutes; then pure dry pulverized calcium fluoride is thrown in small portions on the surface of the red liquid, so as to hasten the cooling, at the same time that the mass is stirred till it becomes pasty.

The metallic globules are thus brought to agglomerate, and if properly performed, by the time the slag has consolidated, all or nearly all the magnesium will have run into one large mass occupying the bottom of the saline fluid. A little of the slag coats the metal and preserves it from oxidation by the air, and ignition; should the latter accident happen, combustion may be stopped by throwing a little calcium fluoride upon the incandescent metal. The crucible is broken when cold, and the metal is chipped out, amounting, with the few little globules scattered through the saline slag, to about 9 1/4 parts, or 3/4 of the quantity which the sodium used ought to furnish. The slag remaining should be re-melted and re-treated seve-ral times, to afford further small quantities of the metal, by which the yield may be brought as high as 45 oz. of crude magnesium for 100 oz. of sodium used.

A modification of this process, which gives good results, is as follows:-The magnesium chloride is prepared by the method just described; 6 parts of this are mixed with 1 of fused salt (or better, of Wohler's mixture of salt and potassium chloride), and 1 of dry pure calcium fluoride, the whole being previously reduced to powder; 1 part of sodium in fragments is well mingled with the chloride, and the whole is thrown into a red-hot crucible; fusion and stirring are repeated as in the process just described, and when the metal has been brought together in a single ball, the whole mass is poured out on a stone or iron slab; on breaking the •lag, globules of magnesium are discovered and removed, and the slag is re-melted to afford further small quantities. To make ingots of magnesium, the crude pieces are fused with a mixture of magnesium chloride, salt, and calcium fluoride; by somewhat increasing the proportion of this last-named substance, which is gradually added to the melting-pot, the slag is rendered less fusible than the magnesium, so that the metal can be poured into a mould at the moment when the slag sets. Perhaps a better plan is to melt together 60 parts common salt and 75 of potassium chloride (Wohler's flux), which is run upon a clean iron slab.

The cooled mass is broken, and introduced into a hot crucible at the same time as the globules of magnesium, which hare previously been carefully cleaned by means of nitric acid. The whole is fused, and soon the magnesium floats on the saline slag; but at the moment when the latter begins to solidify, the magnesium becomes denser than it, and falls to the bottom of the crucible, where it collects in one mass. Even after this operation, the magnesium is not pure, containing carbon, silicon, and nitrogen; to obtain it in a state of absolute purity, it must be distilled.

Wohler suggests using the fused double chloride of magnesium and sodium, to avoid the trouble of preparing the anhydrous magnesium chloride. Sonstadt proposes a mixture of potassium and magnesium chlorides, which may be obtained by fusing carnal lite, carefully avoiding the presence of kieserite, which may cause explosions.

The distillation of magnesium to effect its purity, as alluded to in the last paragraph but one, is thus effected: the crude metal is put into an iron crucible (with an iron tube passing through from the bottom to within 1/2 in. of the cover) until it is filled to the level of the mouth of the tube, when the lid is carefully luted and screwed down, and the air is displaced by a current of hydrogen or coal-gas. As the crucible becomes heated, the magnesium distils over, passing through the upright tube into a receptacle beneath, where it forms a compact mass that may be melted and moulded as desired.

For the manufacture of magnesium wire, use is made of an apparatus introduced by Mather and Piatt, Salford.

Metallic magnesium is employed chiefly as a source of .light for fireworks, signals, and photography: 2 1/2 oz. of burning magnesium will afford a light equal to that from 74 candles burning 10 hours and consuming 20 lb. of stearine, and the light is of an intensely white character. The metal has a sp. gr. of 1.75; in dry air it retains its silvery lustre, but in damp air it assumes a coat of oxide; it melts at a red heat, and boils at a little higher temperature than the volatilizing-point of zinc.