We have separated the last two from the rest because the methods used for their preparation are more of the character of laboratory operations, and because we do not mean to include these in our general exposition of metallurgic principles. The history of metallurgy, up to the most recent times, is obscure. It is only since about the beginning of this century that the art has come to be at all scientifically criticized; and in the case of the most important processes all that science has been able to do has been merely to put her stamp upon what experience has long found to be right. Great and brilliantly successful scientific efforts in the synthetic line are not wanting, but they all belong to recent times. Science, by its very nature, aims at publicity; empiricism at all times has done the reverse; hence a history of the development of the art of metallurgy does not and cannot exist. A few historical notes on the discovery of certain of the useful metals are given in the introduction to Metals (q.υ.). General Sequence of Operations.— Occasionally metallic ores present themselves in the shape of practically pure compact masses, from which the accompanying matrix or “gangue” can be detached by hand and hammer. But this is a rare exception. In most cases the “ore,” as it comes out of the mine, is simply a mixture of ore proper and gangue, in which the latter not unfrequently predominates so much that it is not the gangue but the ore that really occupies the position of what the chemist would call the impurity. Hence, in general, it is necessary, or at least expedient, to purify the ore as such before the liberation of the metal is attempted. Most metallic ores are specifically heavier than the impurities accompanying them, and their purification may be (and generally is) effected by reducing the crude ore to a fine enough powder to detach the metallic from the earthy part, and then washing away the latter by a current of water, as far as possible. In the case of a “reguline” ore, such as auriferous quartz, for instance, the ore thus concentrated may consist substantially of the metal itself, and require only to be melted down and cast into ingots to be ready for the market. This, however, is a rare case, the vast majority of ores being chemical compounds, which for the extraction of their metals demand chemical treatment. The chemical operations involved may be classified as follows :—

1. Fiery Operations.— The ore, along in general with some kind of “flux,” is exposed to the direct action of a powerful fire. The fire in most cases has a chemical, in addition to its obvious physical function. It is intended either to burn away certain components of the ore—in which case it must be so regulated as to contain a sufficient excess of unburned oxygen; or it is meant to deoxidize (“reduce ”) the ore, when the draught must be restricted so as to keep the ore constantly wrapped up in combustible flame gases (carbonic oxide, hydrogen, marsh-gas, &c.). The vast majority of the chemical operations of metallurgy fall into this category, and in these processes other metalreducing agents than those naturally contained in the fire (or wind) are only exceptionally employed.

2. Amalgamation.— The ore by itself (if it happens to be a reguline one), or the ore plus certain reagents (if it does not), is worked up with mercury so that the metal is obtained ultimately as an amalgam, which can be separated mechanically from the dross. The purified amalgam is subjected to distillation, when the mercury is recovered as a distillate while the metal remains.

3. Wet Processes.— Strictly speaking, certain amalgamation methods fall under this head; but, in its ordinary acceptance, the term refers to processes in which the metal is extracted either from the natural ore, or from the ore as it is after roasting or some other preliminary treatment, by means of an aqueous acid or salt solution, and from this solution precipitated—generally in the reguline form—by some suitable reagent.

Few methods of metal extraction at once yield a pure product. What as a rule is obtained is a more or less impure metal, which requires to be “refined” to become fit for the market. We now pass to the individual consideration of the several steps referred to.

Comminution of Ores.— Assuming the ore to be given in the shape of large lumps, these must first be broken up into small stones (of about the size of those used for macadamizing a road) before they can go to the grinding-mill. This formerly used to be done by handwork; nowadays it is preferably effected by means of an American invention called the stone-breaker (fig. 1). This consists essentially of two substantial vertical iron plates; one is fixed, the other is connected withan excentric worked by an engine so as to alternately dash against and recede from the former. The lumps of ore, in passing through this jaw-like contrivance, are broken up into smaller fragments fit for the mill. For the production of a coarse powder revolving cylinders are often employed. Two cylinders of equal diameter and length, made of iron, steel, or stone, are suspended by parallel axes in close Proximity to each other. The width of the slit between them can e made to vary according to the requirements of the case. The cylinders are made to revolve in opposite directions, so that the stones when run into the groove formed by their lipper halves are drawn between them and are crushed into bits of a size depending on the least distance between the two surfaces. Exceptionally hard stones might bring the machine to a standstill or cause breakages; hence only one of the two axes of rotation is absolutely fixed; the cushions of the other are only held in relatively fixed positions, each between a couple of guiding rails, by means of powerful springs at their backs. The springs are made of alternate disks of indiarubber and sheet-iron, and yield appreciably only to very strong pressures. When an exceptionally hard stone comes on, they yield and allow it to pass through uncrushed. Sometimes two sets of cylinders are arranged one above the other, so that the grit from the upper falls into the jaws of the lower set to receive further comminution. The diameter of the cylinders is from a foot to a yard, their length from 9 inches to a yard, the velocity of a point on the periphery a foot to a yard per second. The quantity of ore reduced per hour per horse-power is about 5 cubic feet for quartz or other hard minerals, and about 14 cubic feet for minerals of moderate hardness.

For the production of a relatively fine powder the pounding-mill is frequently used, which, in its action, is analogous to a mortar and pestle. The mortar is a rectangular trough, while the pestle is replaced by a parallel set of heavy metal or metal-shod beams, which (by means of a revolving cylinder with cogs catching projections on the beams) are lifted up in succession and tlιen let fall by their own weight so as to pound up the ore in the trough. The ore is supplied from a prismatic reservoir with a sloping bottom leading into a canal through which the stones slide into the trough. A current of water, which constantly flows into the trough from below, lifts up the finer particles and carries them away over the edge of the trough into a settling tank.

The object pursued in powdering an ore is to prepare it for being purified by washing. But the velocity with which a solid particle falls through water depends on its size as well as on its specific gravity—an increase in either accelerating the fall; hence, where the difference in specific gravity between the things to be separated is small, the washing must be preceded by a separation of the ore-powder into portions of approximately equal fineness. This is often effected by passing the ore through a system of sieves of different width of mesh superposed over one another, the coarser sieve always occupying the higher position. Sometimes the sieves are made to “go dry,” sometimes they are aided in their action by a current of water which, more effectually than mere shaking, prevents adherence of dust to coarser parts.

Another contrivance is the “Drum” (fig. 2). A long perforated circular cylinder made of sheet-iron, open at both ends, is suspended, in a sloping position, by a revolving shaft passing through its axis. The size of the perforations is generally made to increase in passing from the upper to the lower belts of the cylinder. While the drum is revolving, the ore, suspended in water, flows in at the upper end, and in travelling down it casts off first its finest and then its coarser parts, the coarsest only reaching the exit at the lower end. The several grades of powder produced fall each into a separate division of the collecting tank.

The drum, or course, is subject to endless modifications. A very ingenious combination is H. E. Taylor’s “Drum Dressing Machine’’ (fig. 3). It consists of three truncated cone-shaped drums D, fixed co-axially to the same horizontal revolving shaft, so that the narrow end of No. 1 projects into the wider end of No. 2, and No. 2 similarly into No. 3. The drums are not perforated, but are armed inside with screw-threads formed of strips of sheet metal fixed edgeways to the drum. The ore grit to be dressed is placed in a hopper A, and from it, by a worm B fixed to the revolving shaft, is being screwed forward into a short fixed truncated cone C projecting into the revolving drum No. 1, into which it flows in a constant current. The rotary motion of the drum tends to convey the ore along the spiral path prescribed by the screw-thread towards the other end, and from it into drum No. 2, and so on. But the ore in each drum meets with a jet of water E impelling it the opposite way, and the effect is that, in each drum, the lighter parts follow the water, and with it run off over the entrance edge to be collected in a special tank, while the coarser parts roll down the spiral path toward the next drum to undergo further parting. The tank or pit for drum 1 receives the finest and lightest parts, that of drum 2 a heavier, that of drum 3 a still heavier portion, while only the very heaviest matter finds its way out of the exit end of No. 3 into a fourth receptacle.

Of the large number of other ore-dressers, only two need be mentioned here.

The “Clausthal Turn-Table” consists of a circular table, the surface of which rises from the periphery towards the centre so as to form a very flat cone of about 170°, which is fixed co-axially to a vertical rotary shaft. At the apex of the table, surrounding the shaft, but independent of its motion, there is a circular trough of sheet zinc, divided into two compartments; one receives a stream of water carrying the ore, the other a supply of pure water. A large annular trough of sheet zinc is placed below the periphery of the table, so as to receive whatever may fall over the edge. It also is divided into compartments, as shall be explained further on. Supposing the table to be at rest, a sector of about 60° of it would be constantly run over by the ore-mud out of the first compartment of the upper trough. This mud current would suffer partial separation into heavier and lighter parts,—rich ore resting in the higher and poorer in the lower latitudes, and a still poorer ore falling over the periphery into the lower trough. The same happens with the moving table; only each sector of such partially analysed ore undergoes further purification by passing through about 90° of watershower. After passing this, it meets with a perforated fixed waterpipe going up radially to about half the radius of the table. This pipe also carries sweeping brushes, so that the belt of ore from the lower latitudes of the table is swept off into the corresponding section of the receiving trough. What of ore remains on the higher latitudes subsequently meets with a similar arrangement which sweeps it off into its compartment. If the table turns from the left to the right, and we follow the process, beginning at the left edge of the ore-mud compartment, it will be seen that a first sector of the receiving trough gathers the light dross, a succeeding one an intermediate product, a third the most highly purified ore. The “intermediate” is generally run into the ore-mud trough of a second table to be further analysed.

In the “Continuous Wash-Pumps” (Continuirliche Setzpumpe) of the Harz, three funnel-shaped vessels (one of which is shown in fig. 4) are set in a frame beside one another, but at different levels, so that any overfloλv from No. 1 runs into No. 2 and thence into No. 3. Each funnel communicates below with its own compartment of a common cistern.

Into each funnel a riddle with narrow meshes is inserted somewhere near the upper end, while, beside the riddle, there is a pump of short range, which, by means of an excentric, is worked so that the piston alternately goes rapidly down and slowly up. The mode of working is best explained by an example. At Breinigerberg in Rhenish Prussia the apparatus serves to separate a complex ore into the following four parts, which we enumerate in the order of their specific gravities—(1) galena (the heaviest), (2) pyrites, (3) blende, (4) dross. Sieve No. 1 is charged vrith granules of galena, just large enough not t® slip through the meshes, No. 2 similarly with granules of .

pyrites, No. 3 with those of blende. The crude ore-mud goes into sieve 1; the jerking action of the pump alternately tosses the particles up into the water and allows them to fall; the heaviest naturally come down first, but what is most striking is that nothing will pass through the bed of galena but what is at least as heavy as galena itself. In a similar manner No. 2 and No. 3 funnels sift out the pyrites and the blende respectively, so that almost nothing but dross runs off ultimately. The apparatus is said to do its work with a wonderful degree of precision, and of course is susceptible of wider application, but it ceases to work when the raw material is a slime so fine that the particles fall too slowly.

Modes of Producing High Temperatures.— Most of what is to be said on this topic has already been anticipated in the articles Fuel, Furnace, and Bellows; but a few notes may be added on specially metallurgic points.

Furnace Materials. — In a metallurgic furnace the working parts at least must be made of special materials capable of withstanding the very high temperatures to which they are exposed and the action of the fluxes which may be used. No practically available material fully meets both requirements, but there is no lack of merely fire-proof substances.

Of native stones, a pure quartzose sandstone, free from marl, may be named as being well adapted for the generality of structures; but such sandstone, or indeed any kind of fire-proof stone, is not always at hand. What is more readily procured, and consequently more widely used, is refractory brick, made from “fire-clay.” The characteristic chemical feature of fire-clays is that in them the clay proper (always some kind of hydrated silicate of alumina) is associated with only small proportions of lime, magnesia, ferrous oxide, or other protoxides. If the percentage of these goes beyond certain limits, the bricks, when strongly heated, melt down into a slag. The presence of free silica, on the other hand, adds to their refractoriness. In fact the best fire-bricks in existence are the so-called Dinas bricks, which consist substantially of silica, contaminated only with just enough of bases to cause it to frit together on being baked. Dinas bricks, however, on account of their high price, are reserved for special cases involving exceptionally high temperatures. Amongst ordinary fire-bricks those from Stourbridge enjoy the highest reputation. It follows from what has just been said that, in a metallurgic furnace, lime-mortar cannot be used as a cement, but must be replaced by fire-clay paste.

In the construction of cupels, reverberatory furnaces, &c., only the general groundwork is, as a rule, made of built bricks, and this groundwork is coated over with some kind of special fire-proof and flux-proof material, such as bone-ash, a mixture of baked fire-clay and cokes or graphite, or of quartz and very highly silicated slags, &c. These beddings are put on in a loose powdery form, and then stamped fast. They offer the advantage that, when worn out, they are easily removed and renewed. The powerful draught which a metallurgic fire needs can be produced by a chimney, where the fuel forms a relatively shallow layer spread over a large grating; but, when closely-packed deep masses of fuel or fuel and ore have to be kept ablaze, a blast becomes indispensable.

Chimneys.— The efficiency of a chimney is measured by the velocity V with which the air ascends through it, multiplied by its section; and the former is in roughly approximate accordance with the formula

V≡⅛√2g⅛(T-T 0 )∕T 0 , where h stands for the height of the chimney, g for the acceleration of gravity (32 , 2 feet per second), and T and T o for the absolute temperatures (meaning the temperatures counted from - 273° C.) of the air within and the air without the chimney respectively, while ⅛ is a factor meant to account for the resistances which the air, in its progress through the furnace, &c., has to overcome. In practice T is taken as the mean temperature of the chimney gases, which theoretically is not unobjectionable; but the weakest point in the formula is the smallness and utter inconstancy of the factor h, which, according to Péclet, generally assumes some value of the power ⅜, &c. Yet the formula is of some use as enabling one to see the way in which V depends on h and (T - T 0 )∕T 0 conjointly,—to see, for instance, how deficient chimney height may be compensated for by an increase of temperature in the chimney gases, and vice versa. Blowing-Machines.— Of the several kinds of blowers described under Bellows (g.v.), the “fans” are the best means for producing large volumes of wind of relatively small but steady pressure; “bellows” are indicated in the case of work on a relatively small scale requiring moderate wind pressure; while the “cylinder blast” comes in where large masses of high-pressure wind are required. Two highly interesting blowing-machines, however, are omitted in that article, which may be shortly described here.

The “Water Blast” (Wassertrommelgebläse) is interesting historically, having been used metallurgically in Hungary for many centuries. A mass of water, stored up in a reservoir, is made to fall down continuously through a high narrow vertical shaft having air-holes at its upper end. The vertical column of water sucks in air through these holes and carries it down with it into a kind of inverted tub standing in a reservoir kept at a constant level. Air. and water there separate, the former flowing away through a pipe into a wind-box, from which it is led to its destination.

The “Cagniardelle” (figs. 5,6), so called fromits inventor Cagniard Latour, also utilizes water to carry air, but in quite another way. By means of a round shaft passing through its axis, a cylindrical drum of sheet-metal is suspended slantingly in a mass of water, so that the lower end is fully immersed,...