BOOK IX.[1

Since I have written of the varied work of preparing the ores, I will now write of the various methods of smelting them. Although those who burn, roast and calcine[2] the ore, take from it something which is mixed or combined with the metals; and those who crush it with stamps take away much; and those who wash, screen and sort it, take away still more; yet they cannot remove all which conceals the metal from the eye and renders it crude and unformed. Wherefore smelting is necessary, for by this means earths, solidified juices, and stones are separated from the metals so that they obtain their proper colour and become pure, and may be of great use to mankind in many ways. When the ore is smelted, those things which were mixed with the metal before it was melted are driven forth, because the metal is perfected by fire in this manner. Since metalliferous ores differ greatly amongst themselves, first as to the metals which they contain, then as to the quantity of the metal which is in them, and then by the fact that some are rapidly melted by fire and others slowly, there are, therefore, many methods of smelting. Constant practice has taught the smelters by which of these methods they can obtain the most metal from any one ore. Moreover, while sometimes there are many methods of smelting the same ore, by which an equal weight of metal is melted out, yet one is done at a greater cost and labour than the others. Ore is either melted with a furnace or without one; if smelted with a furnace the tap-hole is either temporarily closed or always open, and if smelted without a furnace, it is done either in pots or in trenches. But in order to make this matter clearer, I will describe each in detail, beginning with the buildings and the furnaces.

A wall which will be called the "second wall" is constructed of brick or stone, two feet and as many palms thick, in order that it may be strong enough to bear the weight. It is built fifteen feet high, and its length depends on the number of furnaces which are put in the works; there are usually six furnaces, rarely more, and often less. There are three furnace walls, a back one which is against the "second" wall, and two side ones, of which I will speak later. These should be made of natural stone, as this is more serviceable than burnt bricks, because bricks soon become defective and crumble away, when the smelter or his deputy chips off the accretions which adhere to the walls when the ore is smelted. Natural stone resists injury by the fire and lasts a long time, especially that which is soft and devoid of cracks; but, on the contrary, that which is hard and has many cracks is burst asunder by the fire and destroyed. For this reason, furnaces which are made of the latter are easily weakened by the fire, and when the accretions are chipped off they crumble to pieces. The front furnace wall should be made of brick, and there should be in the lower part a mouth three palms wide and one and a half feet high, when the hearth is completed. A hole slanting upward, three palms long, is made through the back furnace wall, at the height of a cubit, before the hearth has been prepared; through this hole and a hole one foot long in the "second" wall--as the back of this wall has an arch--is inserted a pipe of iron or bronze, in which are fixed the nozzles of the bellows. The whole of the front furnace wall is not more than five feet high, so that the ore may be conveniently put into the furnace, together with those things which the master needs for his work of smelting. Both the side walls of the furnace are six feet high, and the back one seven feet, and they are three palms thick. The interior of the furnace is five palms wide, six palms and a digit long, the width being measured by the space which lies between the two side walls, and the length by the space between the front and the back walls; however, the upper part of the furnace widens out somewhat.

There are two doors in the second wall if there are six furnaces, one of the doors being between the second and third furnaces and the other between the fourth and fifth furnaces. They are a cubit wide and six feet high, in order that the smelters may not have mishaps in coming and going. It is necessary to have a door to the right of the first furnace, and similarly one to the left of the last, whether the wall is longer or not. The second wall is carried further when the rooms for the cupellation furnaces, or any other building, adjoin the rooms for the blast furnaces, these buildings being only divided by a partition. The smelter, and the ones who attend to the first and the last furnaces, if they wish to look at the bellows or to do anything else, go out through the doors at the end of the wall, and the other people go through the other doors, which are the common ones. The furnaces are placed at a distance of six feet from one another, in order that the smelters and their assistants may more easily sustain the fierceness of the heat. Inasmuch as the interior of each furnace is five palms wide and each is six feet distant from the other, and inasmuch as there is a space of four feet three palms at the right side of the first furnace and as much at the left side of the last furnace, and there are to be six furnaces in one building, then it is necessary to make the second wall fifty-two feet long; because the total of the widths of all of the furnaces is seven and a half feet, the total of the spaces between the furnaces is thirty feet, the space on the outer sides of the first and last furnaces is nine feet and two palms, and the thickness of the two transverse walls is five feet, which make a total measurement of fifty-two feet.[3]

Outside each furnace hearth there is a small pit full of powder which is compressed by ramming, and in this manner is made the forehearth which receives the metal flowing from the furnaces. Of this I will speak later.

Buried about a cubit under the forehearth and the hearth of the furnace is a transverse water-tank, three feet long, three palms wide and a cubit deep. It is made of stone or brick, with a stone cover, for if it were not covered, the heat would draw the moisture from below and the vapour might be blown into the hearth of the furnace as well as into the forehearth, and would dampen the blast. The moisture would vitiate the blast, and part of the metal would be absorbed and part would be mixed with the slags, and in this manner the melting would be greatly damaged. From each water-tank is built a walled vent, to the same depth as the tank, but six digits wide; this vent slopes upward, and sooner or later penetrates through to the other side of the wall, against which the furnace is built. At the end of this vent there is an opening where the steam, into which the water has been converted, is exhausted through a copper or iron tube or pipe. This method of making the tank and the vent is much the best. Another kind has a similar vent but a different tank, for it does not lie transversely under the forehearth, but lengthwise; it is two feet and a palm long, and a foot and three palms wide, and a foot and a palm deep. This method of making tanks is not condemned by us, as is the construction of those tanks without a vent; the latter, which have no opening into the air through which the vapour may discharge freely, are indeed to be condemned.

Fifteen feet behind the second wall is constructed the first wall, thirteen feet high. In both of these are fixed roof beams[4], which are a foot wide and thick, and nineteen feet and a palm long; these are placed three feet distant from one another. As the second wall is two feet higher than the first wall, recesses are cut in the back of it two feet high, one foot wide, and a palm deep, and in these recesses, as it were in mortises, are placed one end of each of the beams. Into these ends are mortised the bottoms of just as many posts; these posts are twenty-four feet high, three palms wide and thick, and from the tops of the posts the same number of rafters stretch downward to the ends of the beams superimposed on the first wall; the upper ends of the rafters are mortised into the posts and the lower ends are mortised into the ends of the beams laid on the first wall; the rafters support the roof, which consists of burnt tiles. Each separate rafter is propped up by a separate timber, which is a cross-beam, and is joined to its post. Planks close together are affixed to the posts above the furnaces; these planks are about two digits thick and a palm wide, and they, together with the wicker work interposed between the timbers, are covered with lute so that there may be no risk of fire to the timbers and wicker-work. In this practical manner is constructed the back part of the works, which contains the bellows, their frames, the mechanism for compressing the bellows, and the instrument for distending them, of all of which I will speak hereafter.

In front of the furnaces is constructed the third long wall and likewise the fourth. Both are nine feet high, but of the same length and thickness as the other two, the fourth being nine feet distant from the third; the third is twenty-one and a half feet from the second. At a distance of twelve feet from the second wall, four posts seven and a half feet high, a cubit wide and thick, are set upon rock laid underneath. Into the tops of the posts the roof beam is mortised; this roof beam is two feet and as many palms longer than the distance between the second and the fifth transverse walls, in order that its ends may rest on the transverse walls. If there should not be so long a beam at hand, two are substituted for it. As the length of the long beam is as above, and as the posts are equidistant, it is necessary that the posts should be a distance of nine feet, one palm, two and two-fifths digits from each other, and the end ones this distance from the transverse walls. On this longitudinal beam and to the third and fourth walls are fixed twelve secondary beams twenty-four feet long, one foot wide, three palms thick, and distant from each other three feet, one palm, and two digits. In these secondary beams, where they rest on the longitudinal beams, are mortised the ends of the same number of rafters as there are posts which stand on the second wall. The ends of the rafters do not reach to the tops of the posts, but are two feet away from them, that through this opening, which is like the open part of a forge, the furnaces can emit their fumes. In order that the rafters should not fall down, they are supported partly by iron rods, which extend from each rafter to the opposite post, and partly supported by a few tie-beams, which in the same manner extend from some rafters to the posts opposite, and give them stability. To these tie-beams, as well as to the rafters which face the posts, a number of boards, about two digits thick and a palm wide, are fixed at a distance of a palm from each other, and are covered with lute so that they do not catch fire. In the secondary beams, where they are laid on the fourth wall, are mortised the lower ends of the same number of rafters as those in a set of rafters[5] opposite them. From the third long wall these rafters are joined and tied to the ends of the opposite rafters, so that they may not slip, and besides they are strengthened with substructures which are made of cross and oblique timbers. The rafters support the roof.

In this manner the front part of the building is made, and is divided into three parts; the first part is twelve feet wide and is under the hood, which consists of two walls, one vertical and one inclined. The second part is the same number of feet wide and is for the reception of the ore to be smelted, the fluxes, the charcoal, and other things which are needed by the smelter. The third part is nine feet wide and contains two separate rooms of equal size, in one of which is the assay furnace, while the other contains the metal to be melted in the cupellation furnaces. It is thus necessary that in the building there should be, besides the four long walls, seven transverse walls, of which the first is constructed from the upper end of the first long wall to the upper end of the second long wall; the second proceeds from the end of this to the end of the third long wall; the third likewise from this end of the last extends to the end of the fourth long wall; the fourth leads from the lower end of the first long wall to the lower end of the second long wall; the fifth extends from the end of this to the end of the third long wall; the sixth extends from this last end to the end of the fourth long wall; the seventh divides into two parts the space between the third and fourth long walls.

To return to the back part of the building, in which, as I said, are the bellows[6], their frames, the machinery for compressing them, and the instrument for distending them. Each bellows consists of a body and a head. The body is composed of two "boards," two bows, and two hides. The upper board is a palm thick, five feet and three palms long, and two and a half feet wide at the back part, where each of the sides is a little curved, and it is a cubit wide at the front part near the head. The whole of the body of the bellows tapers toward the head. That which we now call the "board" consists of two pieces of pine, joined and glued together, and of two strips of linden wood which bind the edges of the board, these being seven digits wide at the back, and in front near the head of the bellows one and a half digits wide. These strips are glued to the boards, so that there shall be less damage from the iron nails driven through the hide. There are some people who do not surround the boards with strips, but use boards only, which are very thick. The upper board has an aperture and a handle; the aperture is in the middle of the board and is one foot three palms distant from where the board joins the head of the bellows, and is six digits long and four wide. The lid for this aperture is two palms and a digit long and wide, and three digits thick; toward the back of the lid is a little notch cut into the surface so that it may be caught by the hand; a groove is cut out of the top of the front and sides, so that it may engage in mouldings a palm wide and three digits thick, which are also cut out in a similar manner under the edges. Now, when the lid is drawn forward the hole is closed, and when drawn back it is opened; the smelter opens the aperture a little so that the air may escape from the bellows through it, if he fears the hides might be burst when the bellows are too vigorously and quickly inflated; he, however, closes the aperture if the hides are ruptured and the air escapes. Others perforate the upper board with two or three round holes in the same place as the rectangular one, and they insert plugs in them which they draw out when it is necessary. The wooden handle is seven palms long, or even longer, in order that it may extend outside; one-half of this handle, two palms wide and one thick, is glued to the end of the board and fastened with pegs covered with glue; the other half projects beyond the board, and is rounded and seven digits thick. Besides this, to the handle and to the board is fixed a cleat two feet long, as many palms wide and one palm thick, and to the under side of the same board, at a distance of three palms from the end, is fixed another cleat two feet long, in order that the board may sustain the force of distension and compression; these two cleats are glued to the board, and are fastened to it with pegs covered with glue.

The lower bellows-board, like the upper, is made of two pieces of pine and of two strips of linden wood, all glued together; it is of the same width and thickness as the upper board, but is a cubit longer, this extension being part of the head of which I have more to say a little later. This lower bellows-board has an air-hole and an iron ring. The air-hole is about a cubit distant from the posterior end, and it is midway between the sides of the bellows-board, and is a foot long and three palms wide; it is divided into equal parts by a small rib which forms part of the board, and is not cut from it; this rib is a palm long and one-third of a digit wide. The flap of the air-hole is a foot and three digits long, three palms and as many digits wide; it is a thin board covered with goat skin, the hairy part of which is turned toward the ground. There is fixed to one end of the flap, with small iron nails, one-half of a doubled piece of leather a palm wide and as long as the flap is wide; the other half of the leather, which is behind the flap, is twice perforated, as is also the bellows-board, and these perforations are seven digits apart. Passing through these a string is tied on the under side of the board; and thus the flap when tied to the board does not fall away. In this manner are made the flap and the air-hole, so when the bellows are distended the flap opens, when compressed it closes. At a distance of about a foot beyond the air-hole a slightly elliptical iron ring, two palms long and one wide, is fastened by means of an iron staple to the under part of the bellows-board; it is at a distance of three palms from the back of the bellows. In order that the lower bellows-board may remain stationary, a wooden bolt is driven into the ring, after it penetrates through the hole in the transverse supporting plank which forms part of the frame for the bellows. There are some who dispense with the ring and fasten the bellows-board to the frame with two iron screws something like nails.

The bows are placed between the two boards and are of the same length as the upper board. They are both made of four pieces of linden wood three digits thick, of which the two long ones are seven digits wide at the back and two and a half at the front; the third piece, which is at the back, is two palms wide. The ends of the bows are a little more than a digit thick, and are mortised to the long pieces, and both having been bored through, wooden pegs covered with glue are fixed in the holes; they are thus joined and glued to the long pieces. Each of the ends is bowed (_arcuatur_) to meet the end of the long part of the bow, whence its name "bow" originated. The fourth piece keeps the ends of the bow distended, and is placed a cubit distant from the head of the bellows; the ends of this piece are mortised into the ends of the bow and are joined and glued to them; its length without the tenons is a foot, and its width a palm and two digits. There are, besides, two other very small pieces glued to the head of the bellows and to the lower board, and fastened to them by wooden pegs covered with glue, and they are three palms and two digits long, one palm high, and a digit thick, one half being slightly cut away. These pieces keep the ends of the bow away from the hole in the bellows-head, for if they were not there, the ends, forced inward by the great and frequent movement, would be broken.

The leather is of ox-hide or horse-hide, but that of the ox is far preferable to that of the horse. Each of these hides, for there are two, is three and a half feet wide where they are joined at the back part of the bellows. A long leathern thong is laid along each of the bellows-boards and each of the bows, and fastened by T-shaped iron nails five digits long; each of the horns of the nails is two and a half digits long and half a digit wide. The hide is attached to the bellows-boards by means of these nails, so that a horn of one nail almost touches the horn of the next; but it is different with the bows, for the hide is fastened to the back piece of the bow by only two nails, and to the two long pieces by four nails. In this practical manner they put ten nails in one bow and the same number in the other. Sometimes when the smelter is afraid that the vigorous motion of the bellows may pull or tear the hide from the bows, he also fastens it with little strips of pine by means of another kind of nail, but these strips cannot be fastened to the back pieces of the bow, because these are somewhat bent. Some people do not fix the hide to the bellows-boards and bows by iron nails, but by iron screws, screwed at the same time through strips laid over the hide. This method of fastening the hide is less used than the other, although there is no doubt that it surpasses it in excellence.

Lastly, the head of the bellows, like the rest of the body, consists of two boards, and of a nozzle besides. The upper board is one cubit long, one and a half palms thick. The lower board is part of the whole of the lower bellows-board; it is of the same length as the upper piece, but a palm and a digit thick. From these two glued together is made the head, into which, when it has been perforated, the nozzle is fixed. The back part of the head, where it is attached to the rest of the bellows-body, is a cubit wide, but three palms forward it becomes two digits narrower. Afterward it is somewhat cut away so that the front end may be rounded, until it is two palms and as many digits in diameter, at which point it is bound with an iron ring three digits wide.

The nozzle is a pipe made of a thin plate of iron; the diameter in front is three digits, while at the back, where it is encased in the head of the bellows, it is a palm high and two palms wide. It thus gradually widens out, especially at the back, in order that a copious wind can penetrate into it; the whole nozzle is three feet long.

The upper bellows-board is joined to the head of the bellows in the following way. An iron plate[7], a palm wide and one and a half palms long, is first fastened to the head at a distance of three digits from the end; from this plate there projects a piece three digits long and two wide, curved in a small circle. The other side has a similar plate. Then in the same part of the upper board are fixed two other iron plates, distant two digits from the edge, each of which are six digits wide and seven long; in each of these plates the middle part is cut away for a little more than three digits in length and for two in depth, so that the curved part of the plates on the head corresponding to them may fit into this cut out part. From both sides of each plate there project pieces, three digits long and two digits wide, similarly curved into small circles. A little iron pin is passed through these curved pieces of the plates, like a little axle, so that the upper board of the bellows may turn upon it. The little axle is six digits long and a little more than a digit thick, and a small groove is cut out of the upper board, where the plates are fastened to it, in such a manner that the little axle when fixed to the plates may not fall out. Both plates fastened to the bellows-board are affixed by four iron nails, of which the heads are on the inner part of the board, whereas the points, clinched at the top, are transformed into heads, so to speak. Each of the other plates is fastened to the head of the bellows by means of a nail with a wide head, and by two other nails of which the heads are on the edge of the bellows-head. Midway between the two plates on the bellows-board there remains a space two palms wide, which is covered by an iron plate fastened to the board by little nails; and another plate corresponding to this is fastened to the head between the other two plates; they are two palms and the same number of digits wide.

The hide is common to the head as to all the other parts of the body; the plates are covered with it, as well as the front part of the upper bellows-board, and both the bows and the back of the head of the bellows, so that the wind may not escape from that part of the bellows. It is three palms and as many digits wide, and long enough to extend from one of the sides of the lower board over the back of the upper; it is fastened by many T-headed nails on one side to the upper board, and on the other side to the head of the bellows, and both ends are fastened to the lower bellows-board.

In the above manner the bellows is made. As two are required for each furnace, it is necessary to have twelve bellows, if there are to be six furnaces in one works.

Now it is time to describe their framework. First, two sills a little shorter than the furnace wall are placed on the ground. The front one of these is three palms wide and thick, and the back one three palms and two digits. The front one is two feet distant from the back wall of the furnace, and the back one is six feet three palms distant from the front one. They are set into the earth, that they may remain firm; there are some who accomplish this by means of pegs which, through several holes, penetrate deeply into the ground.

Then twelve short posts are erected, whose lower ends are mortised into the sill that is near the back of the furnace wall; these posts are two feet high, exclusive of the tenons, and are three palms and the same number of digits wide, and two palms thick. A slot one and a half palms wide is cut through them, beginning two palms from the bottom and extending for a height of three palms. All the posts are not placed at the same intervals, the first being at a distance of three feet five digits from the second, and likewise the third from the fourth, but the second is two feet one palm and three digits from the third; the intervals between the other posts are arranged in the same manner, equal and unequal, of which each four pertain to two furnaces. The upper ends of these posts are mortised into a transverse beam which is twelve feet, two palms, and three digits long, and projects five digits beyond the first post and to the same distance beyond the fourth; it is two palms and the same number of digits wide, and two palms thick. Since each separate transverse beam supports four bellows, it is necessary to have three of them.

Behind the twelve short posts the same number of higher posts are erected, of which each has the middle part of the lower end cut out, so that its two resulting lower ends are mortised into the back sill; these posts, exclusive of the tenons, are twelve feet and two palms high, and are five palms wide and two palms thick. They are cut out from the bottom upward, the slot being four feet and five digits high and six digits wide. The upper ends of these posts are mortised into a long beam imposed upon them; this long beam is placed close under the timbers which extend from the wall at the back of the furnace to the first long wall; the beam is three palms wide and two palms thick, and forty-three feet long. If such a long one is not at hand, two or three may be substituted for it, which when joined together make up that length. These higher posts are not placed at equal distances, but the first is at a distance of two feet three palms one digit from the second, and the third is at the same distance from the fourth; while the second is at a distance of one foot three palms and the same number of digits from the third, and in the same manner the rest of the posts are arranged at equal and unequal intervals. Moreover, there is in every post, where it faces the shorter post, a mortise at a foot and a digit above the slot; in these mortises of the four posts is tenoned a timber which itself has four mortises. Tenons are enclosed in mortises in order that they may be better joined, and they are transfixed with wooden pins. This timber is thirteen feet three palms one digit long, and it projects beyond the first post a distance of two palms and two digits, and to the same number of palms and digits beyond the fourth post. It is two palms and as many digits wide, and also two palms thick. As there are twelve posts it is necessary to have three timbers of this kind.

On each of these timbers, and on each of the cross-beams which are laid upon the shorter posts, are placed four planks, each nine feet long, two palms three digits wide, and two palms one digit thick. The first plank is five feet one palm one digit distant from the second, at the front as well as at the back, for each separate plank is placed outside of the posts. The third is at the same distance from the fourth, but the second is one foot and three digits distant from the third. In the same manner the rest of the eight planks are arranged at intervals, the fifth from the sixth and the seventh from the eighth are at the same distances as the first from the second and the third from the fourth; the sixth is at the same distance from the seventh as the second from the third.

Two planks support one transverse plank six feet long, one foot wide, one palm thick, placed at a distance of three feet and two palms from the back posts. When there are six of these supporting planks, on each separate one are placed two bellows; the lower bellows-boards project a palm beyond them. From each of the bellows-boards an iron ring descends through a hole in its supporting plank, and a wooden peg is driven into the ring, so that the bellows-board may remain stationary, as I stated above.

The two bellows communicate, each by its own plank, to the back of a copper pipe in which are set both of the nozzles, and their ends are tightly fastened in it. The pipe is made of a rolled copper or iron plate, a foot and two palms and the same number of digits long; the plate is half a digit thick, but a digit thick at the back. The interior of the pipe is three digits wide, and two and a half digits high in the front, for it is not absolutely round; and at the back it is a foot and two palms and three digits in diameter. The plate from which the pipe is made is not entirely joined up, but at the front there is left a crack half a digit wide, increasing at the back to three digits. This pipe is placed in the hole in the furnace, which, as I said, was in the middle of the wall and the arch. The nozzles of the bellows, placed in this pipe, are a distance of five digits from its front end.

The levers are of the same number as the bellows, and when depressed by the cams of the long axle they compress the bellows. These levers are eight feet three palms long, one palm wide and thick, and the ends are inserted in the slots of the posts; they project beyond the front posts to a distance of two palms, and the same distance beyond the back posts in order that each may have its end depressed by its two cams on the axle. The cams not only penetrate into the slots of the back posts, but project three digits beyond them. An iron pin is set in round holes made through both sides of the slot of each front post, at three palms and as many digits from the bottom; the pin penetrates the lever, which turns about it when depressed or raised. The back of the lever for the length of a cubit is a palm and a digit wider than the rest, and is perforated; in this hole is engaged a bar six feet and two palms long, three digits wide, and about one and one-half digits thick; it is somewhat hooked at the upper end, and approaches the handle of the bellows. Under the lever there is a nail, which penetrates through a hole in the bar, so that the lever and bar may move together. The bar is perforated in the upper end at a distance of six digits from the top; this hole is two palms long and a digit wide, and in it is engaged the hook of an iron implement which is a digit thick. At the upper part this implement has either a round or square opening, like a link, and at the lower end is hooked; the link is two digits high and wide and the hook is three digits long; the middle part between the link and the hook is three palms and two digits long. The link of this implement engages either the handle of the bellows, or else a large ring which does engage it. This iron ring is a digit thick, two palms wide on the inside of the upper part, and two digits in the lower part, and this iron ring, not unlike the first one, engages the handle of the bellows. The iron ring either has its narrower part turned upward, and in it is engaged the ring of another iron implement, similar to the first, whose hook, extending upward, grips the rope fastened to the iron ring holding the end of the second lever, of which I will speak presently; or else the iron ring grips this lever, and then in its hook is engaged the ring of the other implement whose ring engages the handle of the bellows, and in this case the rope is dispensed with.

Resting on beams fixed in the two walls is a longitudinal beam, at a distance of four and a half feet from the back posts; it is two palms wide, one and a half palms thick. There are mortised into this longitudinal beam the lower ends of upper posts three palms wide and two thick, which are six feet two palms high, exclusive of their tenons. The upper ends of these posts are mortised into an upper longitudinal beam, which lies close under the rafters of the building; this upper longitudinal beam is two palms wide and one thick. The upper posts have a slot cut out upward from a point two feet from the bottom, and the slot is two feet high and six digits wide. Through these upper posts a round hole is bored from one side to the other at a point three feet one palm from the bottom, and a small iron axle penetrates through the hole and is fastened there. Around this small iron axle turns the second lever when it is depressed and raised. This lever is eight feet long, and its other end is three digits wider than the rest of the lever; at this widest point is a hole two digits wide and three high, in which is fixed an iron ring, to which is tied the rope I have mentioned; it is five palms long, its upper loop is two palms and as many digits wide, and the lower one is one palm one digit wide. This half of the second lever, the end of which I have just mentioned, is three palms high and one wide; it projects three feet beyond the slot of the post on which it turns; the other end, which faces the back wall of the furnaces, is one foot and a palm high and a foot wide.

On this part of the lever stands and is fixed a box three and a half feet long, one foot and one palm wide, and half a foot deep; but these measurements vary; sometimes the bottom of this box is narrower, sometimes equal in width to the top. In either case, it is filled with stones and earth to make it heavy, but the smelters have to be on their guard and make provision against the stones falling out, owing to the constant motion; this is prevented by means of an iron band which is placed over the top, both ends being wedge-shaped and driven into the lever so that the stones can be held in. Some people, in place of the box, drive four or more pegs into the lever and put mud between them, the required amount being added to the weight or taken away from it.

There remains to be considered the method of using this machine. The lower lever, being depressed by the cams, compresses the bellows, and the compression drives the air through the nozzle. Then the weight of the box on the other end of the upper lever raises the upper bellows-board, and the air is drawn in, entering through the air-hole.

The machine whose cams depress the lower lever is made as follows. First there is an axle, on whose end outside the building is a water-wheel; at the other end, which is inside the building, is a drum made of rundles. This drum is composed of two double hubs, a foot apart, which are five digits thick, the radius all round being a foot and two digits; but they are double, because each hub is composed of two discs, equally thick, fastened together with wooden pegs glued in. These hubs are sometimes covered above and around by iron plates. The rundles are thirty in number, a foot and two palms and the same number of digits long, with each end fastened into a hub; they are rounded, three digits in diameter, and the same number of digits apart. In this practical manner is made the drum composed of rundles.

There is a toothed wheel, two palms and a digit thick, on the end of another axle; this wheel is composed of a double disc[8]. The inner disc is composed of four segments a palm thick, everywhere two palms and a digit wide. The outer disc, like the inner, is made of four segments, and is a palm and a digit thick; it is not equally wide, but where the head of the spokes are inserted it is a foot and a palm and digit wide, while on each side of the spokes it becomes a little narrower, until the narrowest part is only two palms and the same number of digits wide. The outer segments are joined to the inner ones in such a manner that, on the one hand, an outer segment ends in the middle of an inner one, and, on the other hand, the ends of the inner segments are joined in the middle of the outer ones; there is no doubt that by this kind of joining the wheel is made stronger. The outer segments are fastened to the inner by means of a large number of wooden pegs. Each segment, measured over its round back, is four feet and three palms long. There are four spokes, each two palms wide and a palm and a digit thick; their length, excluding the tenons, being two feet and three digits. One end of the spoke is mortised into the axle, where it is firmly fastened with pegs; the wide part of the other end, in the shape of a triangle, is mortised into the outer segment opposite it, keeping the shape of the same as far as the segment ascends. They also are joined together with wooden pegs glued in, and these pegs are driven into the spokes under the inner disc. The parts of the spokes in the shape of the triangle are on the inside; the outer part is simple. This triangle has two sides equal, the erect ones as is evident, which are a palm long; the lower side is not of the same length, but is five digits long, and a mortise of the same shape is cut out of the segments. The wheel has sixty teeth, since it is necessary that the rundle drum should revolve twice while the toothed wheel revolves once. The teeth are a foot long, and project one palm from the inner disc of the wheel, and three digits from the outer disc; they are a palm wide and two and a half digits thick, and it is necessary that they should be three digits apart, as were the rundles.

The axle should have a thickness in proportion to the spokes and the segments. As it has two cams to depress each of the levers, it is necessary that it should have twenty-four cams, which project beyond it a foot and a palm and a digit. The cams are of almost semicircular shape, of which the widest part is three palms and a digit wide, and they are a palm thick; they are distributed according to the four sides of the axle, on the upper, the lower and the two lateral sides. The axle has twelve holes, of which the first penetrates through from the upper side to the lower, the second from one lateral side to the other; the first hole is four feet two palms distant from the second; each alternate one of these holes is made in the same direction, and they are arranged at equal intervals. Each single cam must be opposite another; the first is inserted into the upper part of the first hole, the second into the lower part of the same hole, and so fixed by pegs that they do not fall out; the third cam is inserted into that part of the second hole which is on the right side, and the fourth into that part on the left. In like manner all the cams are inserted into the consecutive holes, for which reason it happens that the cams depress the levers of the bellows in rotation. Finally we must not omit to state that this is only one of many such axles having cams and a water-wheel.

I have arrived thus far with many words, and yet it is not unreasonable that I have in this place pursued the subject minutely, since the smelting of all the metals, to which I am about to proceed, could not be undertaken without it.

The ores of gold, silver, copper, and lead, are smelted in a furnace by four different methods. The first method is for the rich ores of gold or silver, the second for the mediocre ores, the third for the poor ores, and the fourth method is for those ores which contain copper or lead, whether they contain precious metals or are wanting in them. The smelting of the first ores is performed in the furnace of which the tap-hole is intermittently closed; the other three ores are melted in furnaces of which the tap-holes are always open.

First, I will speak of the manner in which the furnaces are prepared for the smelting of the ores, and of the first method of smelting. The powder from which the hearth and forehearth should be made is composed of charcoal and earth (clay?). The charcoal is crushed by the stamps in a mortar-box, the front of which is closed by a board at the top, while the charcoal, crushed to powder, is removed through the open part below; the stamps are not shod with iron, but are made entirely of wood, although at the lower part they are bound round at the wide part by an iron band.

The powder into which the charcoal is crushed is thrown on to a sieve whose bottom consists of interwoven withes of wood. The sieve is drawn backward and forward over two wooden or iron rods placed in a triangular position on a tub, or over a bench-frame set on the floor of the building; the powder which falls into the tub or on to the floor is of suitable size, but the pieces of small charcoal which remain in the sieve are emptied out and thrown back under the stamps.

When the earth is dug up it is first exposed to the sun that it may dry. Later on it is thrown with a shovel on to a screen--set up obliquely and supported by poles,--made of thick, loosely woven hazel withes, and in this way the fine earth and its small lumps pass through the holes of the screen, but the clods and stones do not pass through, but run down to the ground. The earth which passes through the screen is conveyed in a two-wheeled cart to the works and there sifted. This sieve, which is not dissimilar to the one described above, is drawn backward and forward upon narrow boards of equal length placed over a long box; the powder which falls through the sieve into the box is suitable for the mixture; the lumps that remain in the sieve are thrown away by some people, but by others they are placed under the stamps. This powdered earth is mixed with powdered charcoal, moistened, and thrown into a pit, and in order that it may remain good for a long time, the pit is covered up with boards so that the mixture may not become contaminated.

They take two parts of pulverised charcoal and one part of powdered earth, and mix them well together with a rake; the mixture is moistened by pouring water over it so that it may easily be made into shapes resembling snowballs; if the powder be light it is moistened with more water, if heavy with less. The interior of the new furnace is lined with lute, so that the cracks in the walls, if there are any, may be filled up, but especially in order to preserve the rock from injury by fire. In old furnaces in which ore has been melted, as soon as the rocks have cooled the assistant chips away, with a spatula, the accretions which adhere to the walls, and then breaks them up with an iron hoe or a rake with five teeth. The cracks of the furnace are first filled in with fragments of rock or brick, which he does by passing his hand into the furnace through its mouth, or else, having placed a ladder against it, he mounts by the rungs to the upper open part of the furnace. To the upper part of the ladder a board is fastened that he may lean and recline against it. Then standing on the same ladder, with a wooden spatula, he smears the furnace walls over with lute; this spatula is four feet long, a digit thick, and for a foot upward from the bottom it is a palm wide, or even wider, generally two and a half digits. He spreads the lute equally over the inner walls of the furnace. The mouth of the copper pipe[9] should not protrude from the lute, lest sows[10] form round about it and thus impede the melting, for the furnace bellows could not force a blast through them. Then the same assistant throws a little powdered charcoal into the pit of the forehearth and sprinkles it with pulverised earth. Afterward, with a bucket he pours water into it and sweeps this all over the forehearth pit, and with the broom drives the turbid water into the furnace hearth and likewise sweeps it out. Next he throws the mixed and moistened powder into the furnace, and then a second time mounting the steps of the ladder, he introduces the rammer into the furnace and pounds the powder so that the hearth is made solid. The rammer is rounded and three palms long; at the bottom it is five digits in diameter, at the top three and a half, therefore it is made in the form of a truncated cone; the handle of the rammer is round and five feet long and two and a half digits thick; the upper part of the rammer, where the handle is inserted, is bound with an iron band two digits wide. There are some who, instead, use two rounded rammers three and a half digits in diameter, the same at the bottom as at the top. Some people prefer two wooden spatulas, or a rammer spatula.

In a similar manner, mixed and moistened powder is thrown and pounded with a rammer in the forehearth pit, which is outside the furnace. When this is nearly completed, powder is again put in, and pushed with the rammer up toward the protruding copper pipe, so that from a point a digit under the mouth of the copper pipe the hearth slopes down into the crucible of the forehearth,[11] and the metal can run down. The same is repeated until the forehearth pit is full, then afterward this is hollowed out with a curved blade; this blade is of iron, two palms and as many digits long, three digits wide, blunt at the top and sharp at the bottom. The crucible of the forehearth must be round, a foot in diameter and two palms deep if it has to contain a _centumpondium_ of lead, or if only seventy _librae_, then three palms in diameter and two palms deep like the other. When the forehearth has been hollowed out it is pounded with a round bronze rammer. This is five digits high and the same in diameter, having a curved round handle one and a half digits thick; or else another bronze rammer is used, which is fashioned in the shape of a cone, truncated at the top, on which is imposed another cut away at the bottom, so that the middle part of the rammer may be grasped by the hand; this is six digits high, and five digits in diameter at the lower end and four at the top. Some use in its place a wooden spatula two and a half palms wide at the lower end and one palm thick.

The assistant, having prepared the forehearth, returns to the furnace and besmears both sides as well as the top of the mouth with simple lute. In the lower part of the mouth he places lute that has been dipped in charcoal dust, to guard against the risk of the lute attracting to itself the powder of the hearth and vitiating it. Next he lays in the mouth of the furnace a straight round rod three quarters of a foot long and three digits in diameter. Afterward he places a piece of charcoal on the lute, of the same length and width as the mouth, so that it is entirely closed up; if there be not at hand one piece of charcoal so large, he takes two instead. When the mouth is thus closed up, he throws into the furnace a wicker basket full of charcoal, and in order that the piece of charcoal with which the mouth of the furnace is closed should not then fall out, the master holds it in with his hand. The pieces of charcoal which are thrown into the furnace should be of medium size, for if they are large they impede the blast of the bellows and prevent it from blowing through the tap-hole of the furnace into the forehearth to heat it. Then the master covers over the charcoal, placed at the mouth of the furnace, with lute and extracts the wooden rod, and thus the furnace is prepared. Afterward the assistant throws four or five larger baskets full of charcoal into the furnace, filling it right up; he also throws a little charcoal into the forehearth, and places glowing coals upon it in order that it may be kindled, but in order that the flames of this fire should not enter through the tap-hole of the furnace and fire the charcoal inside, he covers the tap-hole with lute or closes it with fragments of pottery. Some do not warm the forehearth the same evening, but place large charcoals round the edge of it, one leaning on the other; those who follow the first method sweep out the forehearth in the morning, and clean out the little pieces of charcoal and cinders, while those who follow the latter method take, early in the morning, burning firebrands, which have been prepared by the watchman of the works, and place them on the charcoal.

At the fourth hour the master begins his work. He first inserts a small piece of glowing coal into the furnace, through the bronze nozzle-pipe of the bellows, and blows up the fire with the bellows; thus within the space of half an hour the forehearth, as well as the hearth, becomes warmed, and of course more quickly if on the preceding day ores have been smelted in the same furnace, but if not then it warms more slowly. If the hearth and forehearth are not warmed before the ore to be smelted is thrown in, the furnace is injured and the metals lost; or if the powder from which both are made is damp in summer or frozen in winter, they will be cracked, and, giving out a sound like thunder, they will blow out the metals and other substances with great peril to the workmen. After the furnace has been warmed, the master throws in slags, and these, when melted, flow out through the tap-hole into the forehearth. Then he closes up the tap-hole at once with mixed lute and charcoal dust; this plug he fastens with his hand to a round wooden rammer that is five digits thick, two palms high, with a handle three feet long. The smelter extracts the slags from the forehearth with a hooked bar; if the ore to be smelted is rich in gold or silver he puts into the forehearth a _centumpondium_ of lead, or half as much if the ore is poor, because the former requires much lead, the latter little; he immediately throws burning firebrands on to the lead so that it melts. Afterward he performs everything according to the usual manner and order, whereby he first throws into the furnace as many cakes melted from pyrites[12], as he requires to smelt the ore; then he puts in two wicker baskets full of ore with litharge and hearth-lead[13], and stones which fuse easily by fire of the second order, all mixed together; then one wicker basket full of charcoal, and lastly the slags. The furnace now being filled with all the things I have mentioned, the ore is slowly smelted; he does not put too much of it against the back wall of the furnace, lest sows should form around the nozzles of the bellows and the blast be impeded and the fire burn less fiercely.

This, indeed, is the custom of many most excellent smelters, who know how to govern the four elements[14]. They combine in right proportion the ores, which are part earth, placing no more than is suitable in the furnaces; they pour in the needful quantity of water; they moderate with skill the air from the bellows; they throw the ore into that part of the fire which burns fiercely. The master sprinkles water into each part of the furnace to dampen the charcoal slightly, so that the minute parts of ore may adhere to it, which otherwise the blast of the bellows and the force of the fire would agitate and blow away with the fumes. But as the nature of the ores to be smelted varies, the smelters have to arrange the hearth now high, now low, and to place the pipe in which the nozzles of the bellows are inserted sometimes on a great and sometimes at a slight angle, so that the blast of the bellows may blow into the furnace in either a mild or a vigorous manner. For those ores which heat and fuse easily, a low hearth is necessary for the work of the smelters, and the pipe must be placed at a gentle angle to produce a mild blast from the bellows. On the contrary, those ores that heat and fuse slowly must have a high hearth, and the pipe must be placed at a steep incline in order to blow a strong blast of the bellows, and it is necessary, for this kind of ore, to have a very hot furnace in which slags, or cakes melted from pyrites, or stones which melt easily in the fire[15], are first melted, so that the ore should not settle in the hearth of the furnace and obstruct and choke up the tap-hole, as the minute metallic particles that have been washed from the ores are wont to do. Large bellows have wide nozzles, for if they were narrow the copious and strong blast would be too much compressed and too acutely blown into the furnace, and then the melted material would be chilled, and would form sows around the nozzle, and thus obstruct the opening into the furnace, which would cause great damage to the proprietors' property. If the ores agglomerate and do not fuse, the smelter, mounting on the ladder placed against the side of the furnace, divides the charge with a pointed or hooked bar, which he also pushes down into the pipe in which the nozzle of the bellows is placed, and by a downward movement dislodges the ore and the sows from around it.

After a quarter of an hour, when the lead which the assistant has placed in the forehearth is melted, the master opens the tap-hole of the furnace with a tapping-bar. This bar is made of iron, is three and a half feet long, the forward end pointed and a little curved, and the back end hollow so that into it may be inserted a wooden handle, which is three feet long and thick enough to be well grasped by the hand. The slag first flows from the furnace into the forehearth, and in it are stones mixed with metal or with the metal adhering to them partly altered, the slag also containing earth and solidified juices. After this the material from the melted pyrites flows out, and then the molten lead contained in the forehearth absorbs the gold and silver. When that which has run out has stood for some time in the forehearth, in order to be able to separate one from the other, the master first either skims off the slags with the hooked bar or else lifts them off with an iron fork; the slags, as they are very light, float on the top. He next draws off the cakes of melted pyrites, which as they are of medium weight hold the middle place; he leaves in the forehearth the alloy of gold or silver with the lead, for these being the heaviest, sink to the bottom. As, however, there is a difference in slags, the uppermost containing little metal, the middle more, and the lowest much, he puts these away separately, each in its own place, in order that to each heap, when it is re-smelted, he may add the proper fluxes, and can put in as much lead as is demanded for the metal in the slag; when the slag is re-melted, if it emits much odour, there is some metal in it; if it emits no odour, then it contains none. He puts the cakes of melted pyrites away separately, as they were nearest in the forehearth to the metal, and contain a little more of it than the slags; from all these cakes a conical mound is built up, by always placing the widest of them at the bottom. The hooked bar has a hook on the end, hence its name; otherwise it is similar to other bars.

Afterward the master closes up the tap-hole and fills the furnace with the same materials I described above, and again, the ores having been melted, he opens the tap-hole, and with a hooked bar extracts the slags and the cakes melted from pyrites, which have run down into the forehearth. He repeats the same operation until a certain and definite part of the ore has been smelted, and the day's work is at an end; if the ore was rich the work is finished in eight hours; if poor, it takes a longer time. But if the ore was so rich as to be smelted in less than eight hours, another operation is in the meanwhile combined with the first, and both are performed in the space of ten hours. When all the ore has been smelted, he throws into the furnace a basket full of litharge or hearth-lead, so that the metal which has remained in the accretions may run out with these when melted. When he has finally drawn out of the forehearth the slags and the cakes melted from pyrites, he takes out, with a ladle, the lead alloyed with gold or silver and pours it into little iron or copper pans, three palms wide and as many digits deep, but first lined on the inside with lute and dried by warming, lest the glowing molten substances should break through. The iron ladle is two palms wide, and in other respects it is similar to the others, all of which have a sufficiently long iron shaft, so that the fire should not burn the wooden part of the handle. When the alloy has been poured out of the forehearth, the smelter foreman and the mine captain weigh the cakes.

Then the master breaks out the whole of the mouth of the furnace with a crowbar, and with that other hooked bar, the rabble and the five-toothed rake, he extracts the accretions and the charcoal. This crowbar is not unlike the other hooked one, but larger and wider; the handle of the rabble is six feet long and is half of iron and half of wood. The furnace having cooled, the master chips off the accretions clinging to the walls with a rectangular spatula six digits long, a palm broad, and sharp on the front edge; it has a round handle four feet long, half of it being of iron and half of wood. This is the first method of smelting ores.

Because they generally consist of unequal constituents, some of which melt rapidly and others slowly, the ores rich in gold and silver cannot be smelted as rapidly or as easily by the other methods as they can by the first method, for three important reasons. The first reason is that, as often as the closed tap-hole of the furnace is opened with a tapping-bar, so often can the smelter observe whether the ore is melting too quickly or too slowly, or whether it is flaming in scattered bits, and not uniting in one mass; in the first case the ore is smelting too slowly and not without great expense; in the second case the metal mixes with the slag which flows out of the furnace into the forehearth, wherefore there is the expense of melting it again; in the third case, the metal is consumed by the violence of the fire. Each of these evils has its remedy; if the ore melts slowly or does not come together, it is necessary to add some amount of fluxes which melt the ore; or if they melt too readily, to decrease the amount.

The second reason is that each time that the furnace is opened with a tapping-bar, it flows out into the forehearth, and the smelter is able to test the alloy of gold and lead or of silver with lead, which is called _stannum_.[16] When the tap-hole is opened the second or third time, this test shows us whether the alloy of gold or silver has become richer, or whether the lead is too debilitated and wanting in strength to absorb any more gold or silver. If it has become richer, some portion of lead added to it should renew its strength; if it has not become richer, it is poured out of the forehearth that it may be replaced with fresh lead.

The third reason is that if the tap-hole of the furnace is always open when the ore and other things are being smelted, the fluxes, which are easily melted, run out of the furnace before the rich gold and silver ores, for these are sometimes of a kind that oppose and resist melting by the fire for a longer period. It follows in this case, that some part of the ore is either consumed or is mixed with the accretions, and as a result little lumps of ore not yet melted are now and then found in the accretions. Therefore when these ores are being smelted, the tap-hole of the furnace should be closed for a time, as it is necessary to heat and mix the ore and the fluxes at the same time; since the fluxes fuse more rapidly than the ore, when the molten fluxes are held in the furnace, they thus melt the ore which does not readily fuse or mix with the lead. The lead absorbs the gold or silver, just as tin or lead when melted in the forehearth absorbs the other unmelted metal which has been thrown into it. But if the molten matter is poured upon that which is not molten, it runs off on all sides and consequently does not melt it. It follows from all this that ores rich in gold or silver, when put into a furnace with its tap-hole always open, cannot for that reason be smelted so successfully as in one where the tap-hole is closed for a time, so that during this time the ore may be melted by the molten fluxes. Afterward, when the tap-hole has been opened, they flow into the forehearth and mix there with the molten lead. This method of smelting the ores is used by us and by the Bohemians.

The three remaining methods of smelting ores are similar to each other in that the tap-holes of the furnaces always remain open, so that the molten metals may continually run out. They differ greatly from each other, however, for the tap-hole of the first of this kind is deeper in the furnace and narrower than that of the third, and besides it is invisible and concealed. It easily discharges into the forehearth, which is one and a half feet higher than the floor of the building, in order that below it to the left a dipping-pot can be made. When the forehearth is nearly full of the slags, which well up from the invisible tap-hole of the furnace, they are skimmed off from the top with a hooked bar; then the alloy of gold or silver with lead and the melted pyrites, being uncovered, flow into the dipping-pot, and the latter are made into cakes; these cakes are broken and thrown back into the furnace so that all their metal may be smelted out. The alloy is poured into little iron moulds.

The smelter, besides lead and cognate things, uses fluxes which combine with the ore, of which I gave a sufficient account in Book VII. The metals which are melted from ores that fuse readily in the fire, are profitable because they are smelted in a short time, while those which are difficult to fuse are not as profitable, because they take a long time. When fluxes remain in the furnace and do not melt, they are not suitable; for this reason, accretions and slags are the most convenient for smelting, because they melt quickly. It is necessary to have an industrious and experienced smelter, who in the first place takes care not to put into the furnace more ores mixed with fluxes than it can accommodate.

The powder out of which this furnace hearth and the adjoining forehearth and the dipping-pot are usually made, consists mostly of equal proportions of charcoal dust and of earth, or of equal parts of the same and of ashes. When the hearth of the furnace is prepared, a rod that will reach to the forehearth is put into it, higher up if the ore to be smelted readily fuses, and lower down if it fuses with difficulty. When the dipping-pot and forehearth are finished, the rod is drawn out of the furnace so that the tap-hole is open, and through it the molten material flows continuously into the forehearth, which should be very near the furnace in order that it may keep very hot and the alloy thus be made purer. If the ore to be smelted does not melt easily, the hearth of the furnace must not be made too sloping, lest the molten fluxes should run down into the forehearth before the ore is smelted, and the metal thus remain in the accretions on the sides of the furnace. The smelter must not ram the hearth so much that it becomes too hard, nor make the mistake of ramming the lower part of the mouth to make it hard, for it could not breathe[17], nor could the molten matter flow freely out of the furnace. The ore which does not readily melt is thrown as much as possible to the back of the furnace, and toward that part where the fire burns very fiercely, so that it may be smelted longer. In this way the smelter may direct it whither he wills. Only when it glows at the part near the bellows' nozzle does it signify that all the ore is smelted which has been thrown to the side of the furnace in which the nozzles are placed. If the ore is easily melted, one or two wicker baskets full are thrown into the front part of the furnace so that the fire, being driven back by it, may also smelt the ore and the sows that form round about the nozzles of the bellows. This process of smelting is very ancient among the Tyrolese[18], but not so old among the Bohemians.

The second method of smelting ores stands in a measure midway between that one performed in a furnace of which the tap-hole is closed intermittently, and the first of the methods performed in a furnace where the tap-hole is always open. In this manner are smelted the ores of gold and silver that are neither very rich nor very poor, but mediocre, which fuse easily and are readily absorbed by the lead. It was found that in this way a large quantity of ore could be smelted at one operation without much labour or great expense, and could thus be alloyed with lead. This furnace has two crucibles, one of which is half inside the furnace and half outside, so that the lead being put into this crucible, the part of the lead which is in the furnace absorbs the metals of the ores which easily fuse; the other crucible is lower, and the alloy and the molten pyrites run into it. Those who make use of this method of smelting, tap the alloy of gold or silver with lead from the upper crucible once or twice if need be, and throw in other lead or litharge, and each absorbs that flux which is nearest. This method of smelting is in use in Styria[19].

The furnace in the third method of smelting ores has the tap-hole likewise open, but the furnace is higher and wider than the others, and its bellows are larger; for these reasons a larger charge of the ore can be thrown into it. When the mines yield a great abundance of ore for the smelter, they smelt in the same furnace continuously for three days and three nights, providing there be no defect either in the hearth or in the forehearth. In this kind of a furnace almost every kind of accretion will be found. The forehearth of the furnace is not unlike the forehearth of the first furnace of all, except that it has a tap-hole. However, because large charges of ore are smelted uninterruptedly, and the melted material runs out and the slags are skimmed off, there is need for a second forehearth crucible, into which the molten material runs through an opened tap-hole when the first is full. When a smelter has spent twelve hours' labour on this work, another always takes his place. The ores of copper and lead and the poorest ores of gold and silver are smelted by this method, because they cannot be smelted by the other three methods on account of the greater expense occasioned. Yet by this method a _centumpondium_ of ore containing only one or two _drachmae_ of gold, or only a half to one _uncia_, of silver,[20] can be smelted; because there is a large amount of ore in each charge, smelting is continuous, and without expensive fluxes such as lead, litharge, and hearth-lead. In this method of smelting we must use only cupriferous pyrites which easily melt in the fire, in truth the cakes melted out from this, if they no longer absorb much gold or silver, are replenished again from crude pyrites alone. If from this poor ore, with melted pyrites alone, material for cakes cannot be made, there are added other fluxes which have not previously been melted. These fluxes are, namely, lead ore, stones easily fused by fire of the second order and sand made from them, limestone, _tophus_, white schist, and iron stone[21].

Although this method of smelting ores is rough and might not seem to be of great use, yet it is clever and useful; for a great weight of ores, in which the gold, silver, or copper are in small quantities, may be reduced into a few cakes containing all the metal. If on being first melted they are too crude to be suitable for the second melting, in which the lead absorbs the precious metals that are in the cakes, or in which the copper is melted out of them, yet they can be made suitable if they are repeatedly roasted, sometimes as often as seven or eight times, as I have explained in the last book. Smelters of this kind are so clever and expert, that in smelting they take out all the gold and silver which the assayer in assaying the ores has stated to be contained in them, because if during the first operation, when he makes the cakes, there is a _drachma_ of gold or half an _uncia_ of silver lost from the ores, the smelter obtains it from the slags by the second smelting. This method of smelting ores is old and very common to most of those who use other methods.

Although lead ores are usually smelted in the third furnace--whose tap-hole is always open,--yet not a few people melt them in special furnaces by a method which I will briefly explain. The _Carni_[22] first burn such lead ores, and afterward break and crush them with large round mallets. Between the two low walls of a hearth, which is inside a furnace made of and vaulted with a rock that resists injury by the fire and does not burn into chalk, they place green wood with a layer of dry wood on the top of it; then they throw the ore on to this, and when the wood is kindled the lead drips down and runs on to the underlying sloping hearth[23]. This hearth is made of pulverised charcoal and earth, as is also a large crucible, one-half of which lies under the furnace and the other half outside it, into which runs the lead. The smelter, having first skimmed off the slags and other things with a hoe, pours the lead with a ladle into moulds, taking out the cakes after they have cooled. At the back of the furnace is a rectangular hole, so that the fire may be allowed more draught, and so that the smelter can crawl through it into the furnace if necessity demands.

The Saxons who inhabit Gittelde, when smelting lead ore in a furnace not unlike a baking oven, put the wood in through a hole at the back of the furnace, and when it begins to burn vigorously the lead trickles out of the ore into a forehearth. When this is full, the smelting being accomplished, the tap-hole is opened with a bar, and in this way the lead, together with the slags, runs into the dipping-pots below. Afterward the cakes of lead, when they are cold, are taken from the moulds.

In Westphalia they heap up ten wagon-loads of charcoal on some hillside which adjoins a level place, and the top of the heap being made flat, straw is thrown upon it to the thickness of three or four digits. On the top of this is laid as much pure lead ore as the heap can bear; then the charcoal is kindled, and when the wind blows, it fans the fire so that the ore is smelted. In this wise the lead, trickling down from the heap, flows on to the level and forms broad thin slabs. A few hundred pounds of lead ore are kept at hand, which, if things go well, are scattered over the heap. These broad slabs are impure and are laid upon dry wood which in turn is placed on green wood laid over a large crucible, and the former having been kindled, the lead is re-melted.

The Poles use a hearth of bricks four feet high, sloping on both sides and plastered with lute. On the upper level part of the hearth large pieces of wood are piled, and on these is placed small wood with lute put in between; over the top are laid wood shavings, and upon these again pure lead ore covered with large pieces of wood. When these are kindled, the ore melts and runs down on to the lower layer of wood; and when this is consumed by the fire, the metal is collected. If necessity demand, it is melted over and over again in the same manner, but it is finally melted by means of wood laid over the large crucible, the slabs of lead being placed upon it.

The concentrates from washing are smelted together with slags (fluxes?) in a third furnace, of which the tap-hole is always open.

It is worth while to build vaulted dust-chambers over the furnaces, especially over those in which the precious ores are to be smelted, in order that the thicker part of the fumes, in which metals are not wanting, may be caught and saved. In this way two or more furnaces are combined under the same vaulted ceiling, which is supported by the wall, against which the furnaces are built, and by four columns. Under this the smelters of the ore perform their work. There are two openings through which the fumes rise from the furnaces into the wide vaulted chamber, and the wider this is the more fumes it collects; in the middle of this chamber over the arch is an opening three palms high and two wide. This catches the fumes of both furnaces, which have risen up from both sides of the vaulted chamber to its arch, and have fallen again because they could not force their way out; and they thus pass out through the opening mentioned, into the chimney which the Greeks call [Greek: kapnodochê], the name being taken from the object. The chimney has thin iron plates fastened into the walls, to which the thinner metallic substances adhere when ascending with the fumes. The thicker metallic substances, or _cadmia_,[25] adhere to the vaulted chamber, and often harden into stalactites. On one side of the chamber is a window in which are set panes of glass, so that the light may be transmitted, but the fumes kept in; on the other side is a door, which is kept entirely closed while the ores are being smelted in the furnaces, so that none of the fumes may escape. It is opened in order that the workman, passing through it, may be enabled to enter the chamber and remove the soot and _pompholyx_[26] and chip off the _cadmia_; this sweeping is done twice a year. The soot mixed with _pompholyx_ and the _cadmia_, being chipped off, is thrown down through a long chute made of four boards joined in the shape of a rectangle, that they should not fly away. They fall on to the floor, and are sprinkled with salt water, and are again smelted with ore and litharge, and become an emolument to the proprietors. Such chambers, which catch the metallic substances that rise with the fumes, are profitable for all metalliferous ores; but especially for the minute metallic particles collected by washing crushed ores and rock, because these usually fly out with the fire of the furnaces.

I have explained the four general methods of smelting ores; now I will state how the ores of each metal are smelted, or how the metal is obtained from the ore. I will begin with gold. Its sand, the concentrates from washing, or the gold dust collected in any other manner, should very often not be smelted, but should be mixed with quicksilver and washed with tepid water, so that all the impurities may be eliminated. This method I explained in Book VII. Or they are placed in the _aqua_ which separates gold from silver, for this also separates its impurities. In this method we see the gold sink in the glass ampulla, and after all the _aqua_ has been drained from the particles, it frequently remains as a gold-coloured residue at the bottom; this powder, when it has been moistened with oil made from argol[27], is then dried and placed in a crucible, where it is melted with borax or with saltpetre and salt; or the same very fine dust is thrown into molten silver, which absorbs it, and from this it is again parted by _aqua valens_[28].

It is necessary to smelt gold ore either outside the blast furnace in a crucible, or inside the blast furnace; in the former case a small charge of ore is used, in the latter a large charge of it. _Rudis_ gold, of whatever colour it is, is crushed with a _libra_ each of sulphur and salt, a third of a _libra_ of copper, and a quarter of a _libra_ of argol; they should be melted in a crucible on a slow fire for three hours, then the alloy is put into molten silver that it may melt more rapidly. Or a _libra_ of the same crude gold, crushed up, is mixed together with half a _libra_ of _stibium_ likewise crushed, and put into a crucible with half an _uncia_ of copper filings, and heated until they melt, then a sixth part of granulated lead is thrown into the same crucible. As soon as the mixture emits an odour, iron-filings are added to it, or if these are not at hand, iron hammer-scales, for both of these break the strength of the _stibium_. When the fire consumes it, not alone with it is some strength of the _stibium_ consumed, but some particles of gold and also of silver, if it be mixed with the gold[29]. When the button has been taken out of the crucible and cooled, it is melted in a cupel, first until the antimony is exhaled, and thereafter until the lead is separated from it.

Crushed pyrites which contains gold is smelted in the same way; it and the _stibium_ should be of equal weight and in truth the gold may be made from them in a number of different ways[30]. One part of crushed material is mixed with six parts of copper, one part of sulphur, half a part of salt, and they are all placed in a pot and over them is poured wine distilled by heating liquid argol in an ampulla. The pot is covered and smeared over with lute and is put in a hot place, so that the mixture moistened with wine may dry for the space of six days, then it is heated for three hours over a gentle fire that it may combine more rapidly with the lead. Finally it is put into a cupel and the gold is separated from the lead[31].

Or else one _libra_ of the concentrates from washing pyrites, or other stones to which gold adheres, is mixed with half a _libra_ of salt, half a _libra_ of argol, a third of a _libra_ of glass-galls, a sixth of a _libra_ of gold or silver slags, and a _sicilicus_ of copper. The crucible into which these are put, after it has been covered with a lid, is sealed with lute and placed in a small furnace that is provided with small holes through which the air is drawn in, and then it is heated until it turns red and the substances put in have alloyed; this should take place within four or five hours. The alloy having cooled, it is again crushed to powder and a pound of litharge is added to it; then it is heated again in another crucible until it melts. The button is taken out, purged of slag, and placed in a cupel, where the gold is separated from the lead.

Or to a _libra_ of the powder prepared from such metalliferous concentrates, is added a _libra_ each of salt, of saltpetre, of argol, and of glass-galls, and it is heated until it melts. When cooled and crushed, it is washed, then to it is added a _libra_ of silver, a third of copper filings, a sixth of litharge, and it is likewise heated again until it melts. After the button has been purged of slag, it is put into the cupel, and the gold and silver are separated from the lead; the gold is parted from the silver with _aqua valens_. Or else a _libra_ of the powder prepared from such metalliferous concentrates, a quarter of a _libra_ of copper filings, and two _librae_ of that second powder[32] which fuses ores, are heated until they melt. The mixture when cooled is again reduced to powder, roasted and washed, and in this manner a blue powder is obtained. Of this, and silver, and that second powder which fuses ores, a _libra_ each are taken, together with three _librae_ of lead, and a quarter of a _libra_ of copper, and they are heated together until they melt; then the button is treated as before. Or else a _libra_ of the powder prepared from such metalliferous concentrates, half a _libra_ of saltpetre, and a quarter of a _libra_ of salt are heated until they melt. The alloy when cooled is again crushed to powder, one _libra_ of which is absorbed by four pounds of molten silver. Or else a _libra_ of the powder made from that kind of concentrates, together with a _libra_ of sulphur, a _libra_ and a half of salt, a third of a _libra_ of salt made from argol, and a third of a _libra_ of copper resolved into powder with sulphur, are heated until they melt. Afterward the lead is re-melted, and the gold is separated from the other metals. Or else a _libra_ of the powder of this kind of concentrates, together with two _librae_ of salt, half a _libra_ of sulphur, and one _libra_ of litharge, are heated, and from these the gold is melted out. By these and similar methods concentrates containing gold, if there be a small quantity of them or if they are very rich, can be smelted outside the blast furnace.

If there be much of them and they are poor, then they are smelted in the blast furnace, especially the ore which is not crushed to powder, and particularly when the gold mines yield an abundance of it[33]. The gold concentrates mixed with litharge and hearth-lead, to which are added iron-scales, are smelted in the blast furnace whose tap-hole is intermittently closed, or else in the first or the second furnaces in which the tap-hole is always open. In this manner an alloy of gold and lead is obtained which is put into the cupellation furnace. Two parts of roasted pyrites or _cadmia_ which contain gold, are put with one part of unroasted, and are smelted together in the third furnace whose tap-hole is always open, and are made into cakes. When these cakes have been repeatedly roasted, they are re-smelted in the furnace whose tap-hole is temporarily closed, or in one of the two others whose tap-holes are always open. In this manner the lead absorbs the gold, whether pure or argentiferous or cupriferous, and the alloy is taken to the cupellation furnace. Pyrites, or other gold ore which is mixed with much material that is consumed by fire and flies out of the furnace, is melted with stone from which iron is melted, if this is at hand. Six parts of such pyrites, or of gold ore reduced to powder and sifted, four of stone from which iron is made, likewise crushed, and three of slaked lime, are mixed together and moistened with water; to these are added two and a half parts of the cakes which contain some copper, together with one and a half parts of slag. A basketful of fragments of the cakes is thrown into the furnace, then the mixture of other things, and then the slag. Now when the middle part of the forehearth is filled with the molten material which runs down from the furnace, the slags are first skimmed off, and then the cakes made of pyrites; afterward the alloy of copper, gold and silver, which settles at the bottom, is taken out. The cakes are gently roasted and re-smelted with lead, and made into cakes, which are carried to other works. The alloy of copper, gold, and silver is not roasted, but is re-melted again in a crucible with an equal portion of lead. Cakes are also made much richer in copper and gold than those I spoke of. In order that the alloy of gold and silver may be made richer, to eighteen _librae_ of it are added forty-eight _librae_ of crude ore, three _librae_ of the stone from which iron is made, and three-quarters of a _libra_ of the cakes made from pyrites, and mixed with lead, all are heated together in the crucible until they melt. When the slag and the cakes melted from pyrites have been skimmed off, the alloy is carried to other furnaces.

There now follows silver, of which the native silver or the lumps of _rudis_ silver[34] obtained from the mines are not smelted in the blast furnaces, but in small iron pans, of which I will speak at the proper place; these lumps are heated and thrown into molten silver-lead alloy in the cupellation furnace when the silver is being separated from the lead, and refined. The tiny flakes or tiny lumps of silver adhering to stones or marble or rocks, or again the same little lumps mixed with earth, or silver not pure enough, should be smelted in the furnace of which the tap-hole is only closed for a short time, together with cakes melted from pyrites, with silver slags, and with stones which easily fuse in fire of the second order.

In order that particles of silver should not fly away[35] from the lumps of ore consisting of minute threads of pure silver and twigs of native silver, they are enclosed in a pot, and are placed in the same furnace where the rest of the silver ores are being smelted. Some people smelt lumps of native silver not sufficiently pure, in pots or triangular crucibles, whose lids are sealed with lute. They do not place these pots in the blast furnace, but arrange them in the assay furnace into which the draught of the air blows through small holes. To one part of the native silver they add three parts of powdered litharge, as many parts of hearth-lead, half a part of galena[36], and a small quantity of salt and iron-scales. The alloy which settles at the bottom of the other substances in the pot is carried to the cupellation furnace, and the slags are re-melted with the other silver slags. They crush under the stamps and wash the pots or crucibles to which silver-lead alloy or slags adhere, and having collected the concentrates they smelt them together with the slags. This method of smelting _rudis_ silver, if there is a small quantity of it, is the best, because the smallest portion of silver does not fly out of the pot or the crucible, and get lost.

If bismuth ore or antimony ore or lead ore[37] contains silver, it is smelted with the other ores of silver; likewise galena or pyrites, if there is a small amount of it. If there be much galena, whether it contain a large or a small amount of silver, it is smelted separately from the others; which process I will explain a little further on.

Because lead and copper ores and their metals have much in common with silver ores, it is fitting that I should say a great deal concerning them, both now and later on. Also in the same manner, pyrites are smelted separately if there be much of them. To three parts of roasted lead or copper ore and one part of crude ore, are added concentrates if they were made by washing the same ore, together with slags, and all are put in the third furnace whose tap-hole is always open. Cakes are made from this charge, which, when they have been quenched with water, are roasted. Of these roasted cakes generally four parts are again mixed with one part of crude pyrites and re-melted in the same furnace. Cakes are again made from this charge, and if there is a large amount of copper in these cakes, copper is made immediately after they have been roasted and re-melted; if there is little copper in the cakes they are also roasted, but they are re-smelted with a little soft slag. In this method the molten lead in the forehearth absorbs the silver. From the pyritic material which floats on the top of the forehearth are made cakes for the third time, and from them when they have been roasted and re-smelted is made copper. Similarly, three parts of roasted _cadmia_[38] in which there is silver, are mixed with one part of crude pyrites, together with slag, and this charge is smelted and cakes are made from it; these cakes having been roasted are re-smelted in the same furnace. By this method the lead contained in the forehearth absorbs the silver, and the silver-lead is taken to the cupellation furnace. Crude quartz and stones which easily fuse in fire of the third order, together with other ores in which there is a small amount of silver, ought to be mixed with crude roasted pyrites or _cadmia_, because the roasted cakes of pyrites or _cadmia_ cannot be profitably smelted separately. In a similar manner earths which contain little silver are mixed with the same; but if pyrites and _cadmia_ are not available to the smelter, he smelts such silver ores and earths with litharge, hearth-lead, slags, and stones which easily melt in the fire. The concentrates[39] originating from the washing of _rudis_ silver, after first being roasted[40] until they melt, are smelted with mixed litharge and hearth-lead, or else, after being moistened with water, they are smelted with cakes made from pyrites and _cadmia_. By neither of these methods do (the concentrates) fall back in the furnace, or fly out of it, driven by the blast of the bellows and the agitation of the fire. If the concentrates originated from galena they are smelted with it after having been roasted; and if from pyrites, then with pyrites.

Pure copper ore, whether it is its own colour or is tinged with chrysocolla or azure, and copper glance, or grey or black _rudis_ copper, is smelted in a furnace of which the tap-hole is closed for a very short time, or else is always open[41]. If there is a large amount of silver in the ore it is run into the forehearth, and the greater part of the silver is absorbed by the molten lead, and the remainder is sold with the copper to the proprietor of the works in which silver is parted from copper[42]. If there is a small amount of silver in the ore, no lead is put into the forehearth to absorb the silver, and the above-mentioned proprietors buy it in with the copper; if there be no silver, copper is made direct. If such copper ore contains some minerals which do not easily melt, as pyrites or _cadmia metallica fossilis_[43], or stone from which iron is melted, then crude pyrites which easily fuse are added to it, together with slag. From this charge, when smelted, they make cakes; and from these, when they have been roasted as much as is necessary and re-smelted, the copper is made. But if there be some silver in the cakes, for which an outlay of lead has to be made, then it is first run into the forehearth, and the molten lead absorbs the silver.

Indeed, _rudis_ copper ore of inferior quality, whether ash-coloured or purple, blackish and occasionally in parts blue, is smelted in the first furnace whose tap-hole is always open. This is the method of the Tyrolese. To as much _rudis_ copper ore as will fill eighteen vessels, each of which holds almost as much as seven Roman _moduli_[44], the first smelter--for there are three--adds three cartloads of lead slags, one cartload of schist, one fifth of a _centumpondium_ of stones which easily fuse in the fire, besides a small quantity of concentrates collected from copper slag and accretions, all of which he smelts for the space of twelve hours, and from which he makes six _centumpondia_ of primary cakes and one-half of a _centumpondium_ of alloy. One half of the latter consists of copper and silver, and it settles to the bottom of the forehearth. In every _centumpondium_ of the cakes there is half a _libra_ of silver and sometimes half an _uncia_ besides; in the half of a _centumpondium_ of the alloy there is a _bes_ or three-quarters of silver. In this way every week, if the work is for six days, thirty-six _centumpondia_ of cakes are made and three _centumpondia_ of alloy, in all of which there is often almost twenty-four _librae_ of silver. The second smelter separates from the primary cakes the greater part of the silver by absorbing it in lead. To eighteen _centumpondia_ of cakes made from crude copper ore, he adds twelve _centumpondia_ of hearth-lead and litharge, three _centumpondia_ of stones from which lead is smelted, five _centumpondia_ of hard cakes rich in silver, and two _centumpondia_ of exhausted liquation cakes[45]; he adds besides, some of the slags resulting from smelting crude copper, together with a small quantity of concentrates made from accretions, all of which he melts for the space of twelve hours, and makes eighteen _centumpondia_ of secondary cakes, and twelve _centumpondia_ of copper-lead-silver alloy; in each _centumpondium_ of the latter there is half a _libra_ of silver. After he has taken off the cakes with a hooked bar, he pours the alloy out into copper or iron moulds; by this method they make four cakes of alloy, which are carried to the works in which silver is parted from copper. On the following day, the same smelter, taking eighteen _centumpondia_ of the secondary cakes, again adds twelve _centumpondia_ of hearth-lead and litharge, three _centumpondia_ of stones from which lead is smelted, five _centumpondia_ of hard cakes rich in silver, together with slags from the smelting of the primary cakes, and with concentrates washed from the accretions which are usually made at that time. This charge is likewise smelted for the space of twelve hours, and he makes as many as thirteen _centumpondia_ of tertiary cakes and eleven _centumpondia_ of copper-lead-silver alloy, each _centumpondium_ of which contains one-third of a _libra_ and half an _uncia_ of silver. When he has skimmed off the tertiary cakes with a hooked bar, the alloy is poured into copper moulds, and by this method four cakes of alloy are made, which, like the preceding four cakes of alloy, are carried to the works in which silver is parted from copper. By this method the second smelter makes primary cakes on alternate days and secondary cakes on the intermediate days. The third smelter takes eleven cartloads of the tertiary cakes and adds to them three cartloads of hard cakes poor in silver, together with the slag from smelting the secondary cakes, and the concentrates from the accretions which are usually made at that time. From this charge when smelted, he makes twenty _centumpondia_ of quaternary cakes, which are called "hard cakes," and also fifteen _centumpondia_ of those "hard cakes rich in silver," each _centumpondium_ of which contains a third of a _libra_ of silver. These latter cakes the second smelter, as I said before, adds to the primary and secondary cakes when he re-melts them. In the same way, from eleven cartloads of quaternary cakes thrice roasted, he makes the "final" cakes, of which one _centumpondium_ contains only half an _uncia_ of silver. In this operation he also makes fifteen _centumpondia_ of "hard cakes poor in silver," in each _centumpondium_ of which is a sixth of a _libra_ of silver. These hard cakes the third smelter, as I have said, adds to the tertiary cakes when he re-smelts them, while from the "final" cakes, thrice roasted and re-smelted, is made black copper[46].

The _rudis_ copper from which pure copper is made, if it contains little silver or if it does not easily melt, is first smelted in the third furnace of which the tap-hole is always open; and from this are made cakes, which after being seven times roasted are re-smelted, and from these copper is melted out; the cakes of copper are carried to a furnace of another kind, in which they are melted for the third time, in order that in the copper "bottoms" there may be more silver, while in the "tops" there may be less, which process is explained in Book XI.

Pyrites, when they contain not only copper, but also silver, are smelted in the manner I described when I treated of ores of silver. But if they are poor in silver, and if the copper which is melted out of them cannot easily be treated, they are smelted according to the method which I last explained.

Finally, the copper schists containing bitumen or sulphur are roasted, and then smelted with stones which easily fuse in a fire of the second order, and are made into cakes, on the top of which the slags float. From these cakes, usually roasted seven times and re-melted, are melted out slags and two kinds of cakes; one kind is of copper and occupies the bottom of the crucible, and these are sold to the proprietors of the works in which silver is parted from copper; the other kind of cakes are usually re-melted with primary cakes. If the schist contains but a small amount of copper, it is burned, crushed under the stamps, washed and sieved, and the concentrates obtained from it are melted down; from this are made cakes from which, when roasted, copper is made. If either chrysocolla or azure, or yellow or black earth containing copper and silver, adheres to the schist, it is not washed, but is crushed and smelted with stones which easily fuse in fire of the second order.

Lead ore, whether it be _molybdaena_[47], pyrites, (galena?) or stone from which it is melted, is often smelted in a special furnace, of which I have spoken above, but no less often in the third furnace of which the tap-hole is always open. The hearth and forehearth are made from powder containing a small portion of iron hammer-scales; iron slag forms the principal flux for such ores; both of these the expert smelters consider useful and to the owner's advantage, because it is the nature of iron to attract lead. If it is _molybdaena_ or the stone from which lead is smelted, then the lead runs down from the furnace into the forehearth, and when the slags have been skimmed off, the lead is poured out with a ladle. If pyrites are smelted, the first to flow from the furnace into the forehearth, as may be seen at Goslar, is a white molten substance, injurious and noxious to silver, for it consumes it. For this reason the slags which float on the top having been skimmed off, this substance is poured out; or if it hardens, then it is taken out with a hooked bar; and the walls of the furnace exude the same substance[48]. Then the _stannum_ runs out of the furnace into the forehearth; this is an alloy of lead and silver. From the silver-lead alloy they first skim off the slags, not rarely white, as some pyrites[49] are, and afterward they skim off the cakes of pyrites, if there are any. In these cakes there is usually some copper; but since there is usually but a very small quantity, and as the forest charcoal is not abundant, no copper is made from them. From the silver-lead poured into iron moulds they likewise make cakes; when these cakes have been melted in the cupellation furnace, the silver is parted from the lead, because part of the lead is transformed into litharge and part into hearth-lead, from which in the blast furnace on re-melting they make de-silverized lead, for in this lead each _centumpondium_ contains only a _drachma_ of silver, when before the silver was parted from it each _centumpondium_ contained more or less than three _unciae_ of silver[50].

The little black stones[51] and others from which tin is made, are smelted in their own kind of furnace, which should be narrower than the other furnaces, that there may be only the small fire which is necessary for this ore. These furnaces are higher, that the height may compensate for the narrowness and make them of almost the same capacity as the other furnaces. At the top, in front, they are closed and on the other side they are open, where there are steps, because they cannot have the steps in front on account of the forehearth; the smelters ascend by these steps to put the tin-stone into the furnace. The hearth of the furnace is not made of powdered earth and charcoal, but on the floor of the works are placed sandstones which are not too hard; these are set on a slight slope, and are two and three-quarters feet long, the same number of feet wide, and two feet thick, for the thicker they are the longer they last in the fire. Around them is constructed a rectangular furnace eight or nine feet high, of broad sandstones, or of those common substances which by nature are composed of diverse materials[52]. On the inside the furnace is everywhere evenly covered with lute. The upper part of the interior is two feet long and one foot wide, but below it is not so long and wide. Above it are two hood-walls, between which the fumes ascend from the furnace into the dust chamber, and through this they escape by a narrow opening in the roof. The sandstones are sloped at the bed of the furnace, so that the tin melted from the tin-stone may flow through the tap-hole of the furnace into the forehearth.[53]

As there is no need for the smelters to have a fierce fire, it is not necessary to place the nozzles of the bellows in bronze or iron pipes, but only through a hole in the furnace wall. They place the bellows higher at the back so that the blast from the nozzles may blow straight toward the tap-hole of the furnace. That it may not be too fierce, the nozzles are wide, for if the fire were fiercer, tin could not be melted out from the tin-stone, as it would be consumed and turned into ashes. Near the steps is a hollowed stone, in which is placed the tin-stone to be smelted; as often as the smelter throws into the furnace an iron shovel-ful of this tin-stone, he puts on charcoal that was first put into a vat and washed with water to be cleansed from the grit and small stones which adhere to it, lest they melt at the same time as the tin-stone and obstruct the tap-hole and impede the flow of tin from the furnace. The tap-hole of the furnace is always open; in front of it is a forehearth a little more than half a foot deep, three-quarters of two feet long and one foot wide; this is lined with lute, and the tin from the tap-hole flows into it. On one side of the forehearth is a low wall, three-quarters of a foot wider and one foot longer than the forehearth, on which lies charcoal powder. On the other side the floor of the building slopes, so that the slags may conveniently run down and be carried away. As soon as the tin begins to run from the tap-hole of the furnace into the forehearth, the smelter scrapes down some of the powdered charcoal into it from the wall, so that the slags may be separated from the hot metal, and so that it may be covered, lest any part of it, being very hot, should fly away with the fumes. If after the slag has been skimmed off, the powder does not cover up the whole of the tin, the smelter draws a little more charcoal off the wall with a scraper. After he has opened the tap-hole of the forehearth with a tapping-bar, in order that the tin can flow into the tapping-pot, likewise smeared with lute, he again closes the tap-hole with pure lute or lute mixed with powdered charcoal. The smelter, if he be diligent and experienced, has brooms at hand with which he sweeps down the walls above the furnace; to these walls and to the dust chamber minute tin-stones sometimes adhere with part of the fumes. If he be not sufficiently experienced in these matters and has melted at the same time all of the tin-stone,--which is commonly of three sizes, large, medium, and very small,--not a little waste of the proprietor's tin results; because, before the large or the medium sizes have melted, the small have either been burnt up in the furnace, or else, flying up from it, they not only adhere to the walls but also fall in the dust chamber. The owner of the works has the sweepings by right from the owner of the ore. For the above reasons the most experienced smelter melts them down separately; indeed, he melts the very small size in a wider furnace, the medium in a medium-sized furnace, and the largest size in the narrowest furnace. When he melts down the small size he uses a gentle blast from the bellows, with the medium-sized a moderate one, with the large size a violent blast; and when he smelts the first size he needs a slow fire, for the second a medium one, and for the third a fierce one; yet he uses a much less fierce fire than when he smelts the ores of gold, silver, or copper. When the workmen have spent three consecutive days and nights in this work, as is usual, they have finished their labours; in this time they are able to melt out a large weight of small sized tin-stone which melts quickly, but less of the large ones which melt slowly, and a moderate quantity of the medium-sized which holds the middle course. Those who do not smelt the tin-stone in furnaces made sometimes wide, sometimes medium, or sometimes narrow, in order that great loss should not be occasioned, throw in first the smallest size, then the medium, then the large size, and finally those which are not quite pure; and the blast of the bellows is altered as required. In order that the tin-stone thrown into the furnace should not roll off from the large charcoal into the forehearth before the tin is melted out of it, the smelter uses small charcoal; first some of this moistened with water is placed in the furnace, and then he frequently repeats this succession of charcoal and tin-stone.

The tin-stone, collected from material which during the summer was washed in a ditch through which a stream was diverted, and during the winter was screened on a perforated iron plate, is smelted in a furnace a palm wider than that in which the fine tin-stone dug out of the earth is smelted. For the smelting of these, a more vigorous blast of the bellows and a fiercer fire is needed than for the smelting of the large tin-stone. Whichever kind of tin-stone is being smelted, if the tin first flows from the furnace, much of it is made, and if slags first flow from the furnace, then only a little. It happens that the tin-stone is mixed with the slags when it is either less pure or ferruginous--that is, not enough roasted--and is imperfect when put into the furnace, or when it has been put in in a larger quantity than was necessary; then, although it may be pure and melt easily, the ore either runs out of the furnace at the same time, mixed with the slags, or else it settles so firmly at the bottom of the furnace that the operation of smelting being necessarily interrupted, the furnace freezes up.

The tap-hole of the forehearth is opened and the tin is diverted into the dipping-pot, and as often as the slags flow down the sloping floor of the building they are skimmed off with a rabble; as soon as the tin has run out of the forehearth, the tap-hole is again closed up with lute mixed with powdered charcoal. Glowing coals are put in the dipping-pot so that the tin, after it has run out, should not get chilled. If the metal is so impure that nothing can be made from it, the material which has run out is made into cakes to be re-smelted in the hearth, of which I shall have something to say later; if the metal is pure, it is poured immediately upon thick copper plates, at first in straight lines and then transversely over these to make a lattice. Each of these lattice bars is impressed with an iron die; if the tin was melted out of ore excavated from mines, then one stamp only, namely, that of the Magistrate, is usually imprinted, but if it is made from tin-stone collected on the ground after washing, then it is impressed with two seals, one the Magistrate's and the other a fork which the washers use. Generally, three of this kind of lattice bars are beaten and amalgamated into one mass with a wooden mallet.

The slags that are skimmed off are afterward thrown with an iron shovel into a small trough hollowed from a tree, and are cleansed from charcoal by agitation; when taken out they are broken up with a square iron mallet, and then they are re-melted with the fine tin-stone next smelted. There are some who crush the slags three times under wet stamps and re-melt them three times; if a large quantity of this be smelted while still wet, little tin is melted from it, because the slag, soon melted again, flows from the furnace into the forehearth. Under the wet stamps are also crushed the lute and broken rock with which such furnaces are lined, and also the accretions, which often contain fine tin-stone, either not melted or half-melted, and also prills of tin. The tin-stone not yet melted runs out through the screen into a trough, and is washed in the same way as tin-stone, while the partly melted and the prills of tin are taken from the mortar-box and washed in the sieve on which not very minute particles remain, and thence to the canvas strake. The soot which adheres to that part of the chimney which emits the smoke, also often contains very fine tin-stone which flies from the furnace with the fumes, and this is washed in the strake which I have just mentioned, and in other sluices. The prills of tin and the partly melted tin-stone that are contained in the lute and broken rock with which the furnace is lined, and in the remnants of the tin from the forehearth and the dipping-pot, are smelted together with the tin-stone.

When tin-stone has been smelted for three days and as many nights in a furnace prepared as I have said above, some little particles of the rock from which the furnace is constructed become loosened by the fire and fall down; and then the bellows being taken away, the furnace is broken through at the back, and the accretions are first chipped off with hammers, and afterward the whole of the interior of the furnace is re-fitted with the prepared sandstone, and again evenly lined with lute. The sandstone placed on the bed of the furnace, if it has become faulty, is taken out, and another is laid down in its place; those rocks which are too large the smelter chips off and fits with a sharp pick.

Some build two furnaces against the wall just like those I have described, and above them build a vaulted ceiling supported by the wall and by four pillars. Through holes in the vaulted ceiling the fumes from the furnaces ascend into a dust chamber, similar to the one described before, except that there is a window on each side and there is no door. The smelters, when they have to clear away the flue-dust, mount by the steps at the side of the furnaces, and climb by ladders into the dust chamber through the apertures in the vaulted ceilings over the furnaces. They then remove the flue-dust from everywhere and collect it in baskets, which are passed from one to the other and emptied. This dust chamber differs from the other described, in the fact that the chimneys, of which it has two, are not dissimilar to those of a house; they receive the fumes which, being unable to escape through the upper part of the chamber, are turned back and re-ascend and release the tin; thus the tin set free by the fire and turned to ash, and the little tin-stones which fly up with the fumes, remain in the dust chamber or else adhere to copper plates in the chimney.

If the tin is so impure that it cracks when struck with the hammer, it is not immediately made into lattice-like bars, but into the cakes which I have spoken of before, and these are refined by melting again on a hearth. This hearth consists of sandstones, which slope toward the centre and a little toward a dipping-pot; at their joints they are covered with lute. Dry logs are arranged on each side, alternately upright and lengthwise, and more closely in the middle; on this wood are placed five or six cakes of tin which all together weigh about six _centumpondia_; the wood having been kindled, the tin drips down and flows continuously into the dipping-pot which is on the floor. The impure tin sinks to the bottom of this dipping-pot and the pure tin floats on the top; then both are ladled out by the master, who first takes out the pure tin, and by pouring it over thick plates of copper makes lattice-like bars. Afterward he takes out the impure tin from which he makes cakes; he discriminates between them, when he ladles and pours, by the ease or difficulty of the flow. One _centumpondium_ of the lattice-like bars sells for more than a _centumpondium_ of cakes, for the price of the former exceeds the price of the latter by a gold coin[54]. These lattice-like bars are lighter than the others, and when five of them are pounded and amalgamated with a wooden mallet, a mass is made which is stamped with an iron die. There are some who do not make a dipping-pot on the floor for the tin to run into, but in the hearth itself; out of this the master, having removed the charcoal, ladles the tin and pours it over the copper-plate. The dross which adheres to the wood and the charcoal, having been collected, is re-smelted in the furnace.

Some of the Lusitanians melt tin from tin-stone in small furnaces. They use round bellows made of leather, of which the fore end is a round iron disc and the rear end a disc of wood; in a hole in the former is fixed the nozzle, in the middle of the latter the blow-hole. Above this is the handle or haft, which draws open the round bellows and lets in the air, or compresses it and drives the air out. Between the discs are several iron hoops to which the leather is fastened, making such folds as are to be seen in paper lanterns that are folded together. Since this kind of bellows does not give a vigorous blast, because they are drawn apart and compressed slowly, the smelter is not able during a whole day to smelt much more than half a _centumpondium_ of tin.

Very good iron ore is smelted[55] in a furnace almost like the cupellation furnace. The hearth is three and a half feet high, and five feet long and wide; in the centre of it is a crucible a foot deep and one and a half feet wide, but it may be deeper or shallower, wider or narrower, according to whether more or less ore is to be made into iron. A certain quantity of iron ore is given to the master, out of which he may smelt either much or little iron. He being about to expend his skill and labour on this matter, first throws charcoal into the crucible, and sprinkles over it an iron shovel-ful of crushed iron ore mixed with unslaked lime. Then he repeatedly throws on charcoal and sprinkles it with ore, and continues this until he has slowly built up a heap; it melts when the charcoal has been kindled and the fire violently stimulated by the blast of the bellows, which are skilfully fixed in a pipe. He is able to complete this work sometimes in eight hours, sometimes in ten; and again sometimes in twelve. In order that the heat of the fire should not burn his face, he covers it entirely with a cap, in which, however, there are holes through which he may see and breathe. At the side of the hearth is a bar which he raises as often as is necessary, when the bellows blow too violent a blast, or when he adds more ore and charcoal. He also uses the bar to draw off the slags, or to open or close the gates of the sluice, through which the waters flow down on to the wheel which turns the axle that compresses the bellows. In this sensible way, iron is melted out and a mass weighing two or three _centumpondia_ may be made, providing the iron ore was rich. When this is done the master opens the slag-vent with the tapping-bar, and when all has run out he allows the iron mass to cool. Afterward he and his assistant stir the iron with the bar, and then in order to chip off the slags which had until then adhered to it, and to condense and flatten it, they take it down from the furnace to the floor, and beat it with large wooden mallets having slender handles five feet long. Thereupon it is immediately placed on the anvil, and repeatedly beaten by the large iron hammer that is raised by the cams of an axle turned by a water-wheel. Not long afterward it is taken up with tongs and placed under the same hammer, and cut up with a sharp iron into four, five, or six pieces, according to whether it is large or small. These pieces, after they have been re-heated in the blacksmith's forge and again placed on the anvil, are shaped by the smith into square bars or into ploughshares or tyres, but mainly into bars. Four, six, or eight of these bars weigh one-fifth of a _centumpondium_, and from these they make various implements. During the blows from the hammer by which it is shaped by the smith, a youth pours water with a ladle on to the glowing iron, and this is why the blows make such a loud sound that they may be heard a long distance from the works. The masses, if they remain and settle in the crucible of the furnace in which the iron is smelted, become hard iron which can only be hammered with difficulty, and from these they make the iron-shod heads for the stamps, and such-like very hard articles.

But to iron ore which is cupriferous, or which when heated[56] melts with difficulty, it is necessary for us to give a fiercer fire and more labour; because not only must we separate the parts of it in which there is metal from those in which there is no metal, and break it up by dry stamps, but we must also roast it, so that the other metals and noxious juices may be exhaled; and we must wash it, so that the lighter parts may be separated from it. Such ores are smelted in a furnace similar to the blast furnace, but much wider and higher, so that it may hold a great quantity of ore and much charcoal; mounting the stairs at the side of the furnace, the smelters fill it partly with fragments of ore not larger than nuts, and partly with charcoal; and from this kind of ore once or twice smelted they make iron which is suitable for re-heating in the blacksmith's forge, after it is flattened out with the large iron hammer and cut into pieces with the sharp iron.

By skill with fire and fluxes is made that kind of iron from which steel is made, which the Greeks call [Greek: stomôma]. Iron should be selected which is easy to melt, is hard and malleable. Now although iron may be smelted from ore which contains other metals, yet it is then either soft or brittle; such (iron) must be broken up into small pieces when it is hot, and then mixed with crushed stone which melts. Then a crucible is made in the hearth of the smith's furnace, from the same moistened powder from which are made the forehearths in front of the furnaces in which ores of gold or silver are smelted; the width of this crucible is about one and a half feet and the depth one foot. The bellows are so placed that the blast may be blown through the nozzle into the middle of the crucible. Then the whole of the crucible is filled with the best charcoal, and it is surrounded by fragments of rock to hold in place the pieces of iron and the superimposed charcoal. As soon as all the charcoal is kindled and the crucible is glowing, a blast is blown from the bellows and the master pours in gradually as much of the mixture of iron and flux as he wishes. Into the middle of this, when it is melted, he puts four iron masses each weighing thirty pounds, and heats them for five or six hours in a fierce fire; he frequently stirs the melted iron with a bar, so that the small pores in each mass absorb the minute particles, and these particles by their own strength consume and expand the thick particles of the masses, which they render soft and similar to dough. Afterward the master, aided by his assistant, takes out a mass with the tongs and places it on the anvil, where it is pounded by the hammer which is alternately raised and dropped by means of the water-wheel; then, without delay, while it is still hot, he throws it into water and tempers it; when it is tempered, he places it again on the anvil, and breaks it with a blow from the same hammer. Then at once examining the fragments, he decides whether the iron in some part or other, or as a whole, appears to be dense and changed into steel; if so, he seizes one mass after another with the tongs, and taking them out he breaks them into pieces. Afterward he heats the mixture up again, and adds a portion afresh to take the place of that which has been absorbed by the masses. This restores the energy of that which is left, and the pieces of the masses are again put back into the crucible and made purer. Each of these, after having been heated, is seized with the tongs, put under the hammer and shaped into a bar. While they are still glowing, he at once throws them into the very coldest nearby running water, and in this manner, being suddenly condensed, they are changed into pure steel, which is much harder and whiter than iron.

The ores of the other metals are not smelted in furnaces. Quicksilver ores and also antimony are melted in pots, and bismuth in troughs.

I will first speak of quicksilver. This is collected when found in pools formed from the outpourings of the veins and stringers; it is cleansed with vinegar and salt, and then it is poured into canvas or soft leather, through which, when squeezed and compressed, the quicksilver runs out into a pot or pan. The ore of quicksilver is reduced in double or single pots. If in double pots, then the upper one is of a shape not very dissimilar to the glass ampullas used by doctors, but they taper downward toward the bottom, and the lower ones are little pots similar to those in which men and women make cheese, but both are larger than these; it is necessary to sink the lower pots up to the rims in earth, sand, or ashes. The ore, broken up into small pieces is put into the upper pots; these having been entirely closed up with moss, are placed upside down in the openings of the lower pots, where they are joined with lute, lest the quicksilver which takes refuge in them should be exhaled. There are some who, after the pots have been buried, do not fear to leave them uncemented, and who boast that they are able to produce no less weight of quicksilver than those who do cement them, but nevertheless cementing with lute is the greatest protection against exhalation. In this manner seven hundred pairs of pots are set together in the ground or on a hearth. They must be surrounded on all sides with a mixture consisting of crushed earth and charcoal, in such a way that the upper pots protrude to a height of a palm above it. On both sides of the hearth rocks are first laid, and upon them poles, across which the workmen place other poles transversely; these poles do not touch the pots, nevertheless the fire heats the quicksilver, which fleeing from the heat is forced to run down through the moss into the lower pots. If the ore is being reduced in the upper pots, it flees from them, wherever there is an exit, into the lower pots, but if the ore on the contrary is put in the lower pots the quicksilver rises into the upper pot or into the operculum, which, together with the gourd-shaped vessels, are cemented to the upper pots.

The pots, lest they should become defective, are moulded from the best potters' clay, for if there are defects the quicksilver flies out in the fumes. If the fumes give out a very sweet odour it indicates that the quicksilver is being lost, and since this loosens the teeth, the smelters and others standing by, warned of the evil, turn their backs to the wind, which drives the fumes in the opposite direction; for this reason, the building should be open around the front and the sides, and exposed to the wind. If these pots are made of cast copper they last a long time in the fire. This process for reducing the ores of quicksilver is used by most people.

In a similar manner the antimony ore,[57] if free from other metals, is reduced in upper pots which are twice as large as the lower ones. Their size, however, depends on the cakes, which have not the same weight everywhere; for in some places they are made to weigh six _librae_, in other places ten, and elsewhere twenty. When the smelter has concluded his operation, he extinguishes the fire with water, removes the lids from the pots, throws earth mixed with ash around and over them, and when they have cooled, takes out the cakes from the pots.

Other methods for reducing quicksilver are given below. Big-bellied pots, having been placed in the upper rectangular open part of a furnace, are filled with the crushed ore. Each of these pots is covered with a lid with a long nozzle--commonly called a _campana_--in the shape of a bell, and they are cemented. Each of the small earthenware vessels shaped like a gourd receives two of these nozzles, and these are likewise cemented. Dried wood having been placed in the lower part of the furnace and kindled, the ore is heated until all the quicksilver has risen into the operculum which is over the pot; it then flows from the nozzle and is caught in the earthenware gourd-shaped vessel.

Others build a hollow vaulted chamber, of which the paved floor is made concave toward the centre. Inside the thick walls of the chamber are the furnaces. The doors through which the wood is put are in the outer part of the same wall. They place the pots in the furnaces and fill them with crushed ore, then they cement the pots and the furnaces on all sides with lute, so that none of the vapour may escape from them, and there is no entrance to the furnaces except through their mouths. Between the dome and the paved floor they arrange green trees, then they close the door and the little windows, and cover them on all sides with moss and lute, so that none of the quicksilver can exhale from the chamber. After the wood has been kindled the ore is heated, and exudes the quicksilver; whereupon, impatient with the heat, and liking the cold, it escapes to the leaves of the trees, which have a cooling power. When the operation is completed the smelter extinguishes the fire, and when all gets cool he opens the door and the windows, and collects the quicksilver, most of which, being heavy, falls of its own accord from the trees, and flows into the concave part of the floor; if all should not have fallen from the trees, they are shaken to make it fall.

The following is the fourth method of reducing ores of quicksilver. A larger pot standing on a tripod is filled with crushed ore, and over the ore is put sand or ashes to a thickness of two digits, and tamped; then in the mouth of this pot is inserted the mouth of another smaller pot and cemented with lute, lest the vapours are emitted. The ore heated by the fire exhales the quicksilver, which, penetrating through the sand or the ashes, takes refuge in the upper pot, where condensing into drops it falls back into the sand or the ashes, from which the quicksilver is washed and collected.

The fifth method is not very unlike the fourth. In the place of these pots are set other pots, likewise of earthenware, having a narrow bottom and a wide mouth. These are nearly filled with crushed ore, which is likewise covered with ashes to a depth of two digits and tamped in. The pots are covered with lids a digit thick, and they are smeared over on the inside with liquid litharge, and on the lid are placed heavy stones. The pots are set on the furnace, and the ore is heated and similarly exhales quicksilver, which fleeing from the heat takes refuge in the lid; on congealing there, it falls back into the ashes, from which, when washed, the quicksilver is collected.

By these five methods quicksilver may be made, and of these not one is to be despised or repudiated; nevertheless, if the mine supplies a great abundance of ore, the first is the most expeditious and practical, because a large quantity of ore can be reduced at the same time without great expense.[58]

Bismuth[59] ore, free from every kind of silver, is smelted by various methods. First a small pit is dug in the dry ground; into this pulverised charcoal is thrown and tamped in, and then it is dried with burning charcoal. Afterward, thick dry pieces of beech wood are placed over the pit, and the bismuth ore is thrown on it. As soon as the kindled wood burns, the heated ore drips with bismuth, which runs down into the pit, from which when cooled the cakes are removed. Because pieces of burnt wood, or often charcoal and occasionally slag, drop into the bismuth which collects in the pit, and make it impure, it is put back into another kind of crucible to be melted, so that pure cakes may be made. There are some who, bearing these things in mind, dig a pit on a sloping place and below it put a forehearth, into which the bismuth continually flows, and thus remains clean; then they take it out with ladles and pour it into iron pans lined inside with lute, and make cakes of it. They cover such pits with flat stones, whose joints are besmeared with a lute of mixed dust and crushed charcoal, lest the joints should absorb the molten bismuth. Another method is to put the ore in troughs made of fir-wood and placed on sloping ground; they place small firewood over it, kindling it when a gentle wind blows, and thus the ore is heated. In this manner the bismuth melts and runs down from the troughs into a pit below, while there remains slag, or stones, which are of a yellow colour, as is also the wood laid across the pit. These are also sold.

Others reduce the ore in iron pans as next described. They lay small pieces of dry wood alternately straight and transversely upon bricks, one and a half feet apart, and set fire to it. Near it they put small iron pans lined on the inside with lute, and full of broken ore; then when the wind blows the flame of the fierce fire over the pans, the bismuth drips out of the ore; wherefore, in order that it may run, the ore is stirred with the tongs; but when they decide that all the bismuth is exuded, they seize the pans with the tongs and remove them, and pour out the bismuth into empty pans, and by turning many into one they make cakes. Others reduce the ore, when it is not mixed with _cadmia_,[60] in a furnace similar to the iron furnace. In this case they make a pit and a crucible of crushed earth mixed with pulverised charcoal, and into it they put the broken ore, or the concentrates from washing, from which they make more bismuth. If they put in ore, they reduce it with charcoal and small dried wood mixed, and if concentrates, they use charcoal only; they blow both materials with a gentle blast from a bellows. From the crucible is a small pipe through which the molten bismuth runs down into a dipping-pot, and from this cakes are made.

On a dump thrown up from the mines, other people construct a hearth exposed to the wind, a foot high, three feet wide, and four and a half feet long. It is held together by four boards, and the whole is thickly coated at the top with lute. On this hearth they first put small dried sticks of fir wood, then over them they throw broken ore; then they lay more wood over it, and when the wind blows they kindle it. In this manner the bismuth drips out of the ore, and afterward the ashes of the wood consumed by the fire and the charcoals are swept away. The drops of bismuth which fall down into the hearth are congealed by the cold, and they are taken away with the tongs and thrown into a basket. From the melted bismuth they make cakes in iron pans.

Others again make a box eight feet long, four feet wide, and two feet high, which they fill almost full of sand and cover with bricks, thus making the hearth. The box has in the centre a wooden pivot, which turns in a hole in two beams laid transversely one upon the other; these beams are hard and thick, are sunk into the ground, both ends are perforated, and through these holes wedge-shaped pegs are driven, in order that the beams may remain fixed, and that the box may turn round, and may be turned toward the wind from whichever quarter of the sky in may blow. In such a hearth they put an iron grate, as long and wide as the box and three-quarters of a foot high; it has six feet, and there are so many transverse bars that they almost touch one another. On the grate they lay pine-wood and over it broken ore, and over this they again lay pine-wood. When it has been kindled the ore melts, out of which the bismuth drips down; since very little wood is burned, this is the most profitable method of smelting the bismuth. The bismuth drips through the grate on to the hearth, while the other things remain upon the grate with the charcoal. When the work is finished, the workman takes a stick from the hearth and overturns the grate, and the things which have been accumulated on it; with the brush he sweeps up the bismuth and collects it in a basket, and then he melts it in an iron pan and makes cakes. As soon as possible after it is cool, he turns the pans over, so that the cakes may fall out, using for this purpose a two-pronged fork of which one prong is again forked. And immediately afterward he returns to his labours.

END OF BOOK IX.

FOOTNOTES:

[1] The history of the fusion of ores and of metals is the history of individual processes, and such information as we have been able to discover upon the individual methods previous to Agricola we give on the pages where such processes are discussed. In general the records of the beginnings of metallurgy are so nebular that, if one wishes to shirk the task, he can adopt the explanation of William Pryce one hundred and fifty years ago: "It is very probable that the nature and use of Metals were not revealed to Adam in his state of innocence: the toil and labour necessary to procure and use those implements of the iron age could not be known, till they made part of the curse incurred by his fall: 'In the sweat of thy face shalt thou eat bread, till thou return unto the ground; in sorrow shalt thou eat of it all the days of thy life' (Genesis). That they were very early discovered, however, is manifest from the Mosaick account of Tubal Cain, who was the first instructor of every artificer in Brass [_sic_] and Iron" (_Mineralogia Cornubiensis_, p. 2).

It is conceivable that gold could be found in large enough pieces to have had general use in pre-historic times, without fusion; but copper, which was also in use, must have been smelted, and therefore we must assume a considerable development of human knowledge on the subject prior to any human record. Such incidental mention as exists after record begins does not, of course, extend to the beginning of any particular branch of the art--in fact, special arts obviously existed long before such mention, and down to the complete survey of the state of the art by Agricola our dates are necessarily "prior to" some first mention in literature, or "prior to" the known period of existing remains of metallurgical operations. The scant Egyptian records, the Scriptures, and the Shoo King give a little insight prior to 1000 B.C. The more extensive Greek literature of about the 5th to the 3rd centuries B.C., together with the remains of Greek mines, furnish another datum point of view, and the Roman and Greek writers at the beginning of the Christian era give a still larger view. After them our next step is to the Monk Theophilus and the Alchemists, from the 12th to the 14th centuries. Finally, the awakening of learning at the end of the 15th and the beginning of the 16th centuries, enables us for the first time to see practically all that was known. The wealth of literature which exists subsequent to this latter time makes history thereafter a matter of some precision, but it is not included in this undertaking. Considering the great part that the metals have played in civilization, it is astonishing what a minute amount of information is available on metallurgy. Either the ancient metallurgists were secretive as to their art, or the ancient authors despised such common things, or, as is equally probable, the very partial preservation of ancient literature, by painful transcription over a score of centuries, served only for those works of more general interest. In any event, if all the direct or indirect material on metallurgy prior to the 15th century were compiled, it would not fill 40 pages such as these.

It may be of service to give a tabular summary indicating approximately the time when evidence of particular operations appear on the historical horizon:

Gold washed from alluvial Prior to recorded civilization

Copper reduced from ores by smelting Prior to recorded civilization

Bitumen mined and used Prior to recorded civilization

Tin reduced from ores by smelting Prior to 3500 B.C.

Bronze made Prior to 3500 B.C.

Iron reduced from ores by smelting Prior to 3500 B.C.

Soda mined and used Prior to 3500 B.C.

Gold reduced from ores by concentration Prior to 2500 B.C.

Silver reduced from ores by smelting Prior to 2000 B.C.

Lead reduced from ores by smelting Prior to 2000 B.C. (perhaps prior to 3500 B.C.)

Silver parted from lead by cupellation Prior to 2000 B.C.

Bellows used in furnaces Prior to 1500 B.C.

Steel produced Prior to 1000 B.C.

Base metals separated from ores by water Prior to 500 B.C. concentration

Gold refined by cupellation Prior to 500 B.C.

Sulphide ores smelted for lead Prior to 500 B.C.

Mercury reduced from ores by (?) Prior to 400 B.C.

White-lead made with vinegar Prior to 300 B.C.

Touchstone known for determining gold and silver Prior to 300 B.C. fineness

Quicksilver reduced from ore by distillation Prior to Christian Era

Silver parted from gold by cementation with salt Prior to " "

Brass made by cementation of copper and calamine Prior to " "

Zinc oxides obtained from furnace fumes by Prior to " " construction of dust chambers

Antimony reduced from ores by smelting (accidental) Prior to " "

Gold recovered by amalgamation Prior to " "

Refining of copper by repeated fusion Prior to " "

Sulphide ores smelted for copper Prior to " "

Vitriol (blue and green) made Prior to " "

Alum made Prior to " "

Copper refined by oxidation and poling Prior to 1200 A.D.

Gold parted from copper by cupelling with lead Prior to 1200 A.D.

Gold parted from silver by fusion with sulphur Prior to 1200 A.D.

Manufacture of nitric acid and _aqua regia_ Prior to 1400 A.D.

Gold parted from silver by nitric acid Prior to 1400 A.D.

Gold parted from silver with antimony sulphide Prior to 1500 A.D.

Gold parted from copper with sulphur Prior to 1500 A.D.

Silver parted from iron with antimony sulphide Prior to 1500 A.D.

First text book on assaying Prior to 1500 A.D.

Silver recovered from ores by amalgamation Prior to 1500 A.D.

Separation of silver from copper by liquation Prior to 1540 A.D.

Cobalt and manganese used for pigments Prior to 1540 A.D.

Roasting copper ores prior to smelting Prior to 1550 A.D.

Stamp-mill used Prior to 1550 A.D.

Bismuth reduced from ore Prior to 1550 A.D.

Zinc reduced from ore (accidental) Prior to 1550 A.D.

Further, we believe it desirable to sketch at the outset the development of metallurgical appliances as a whole, leaving the details to special footnotes; otherwise a comprehensive view of the development of such devices is difficult to grasp.

We can outline the character of metallurgical appliances at various periods in a few words. It is possible to set up a description of the imaginary beginning of the "bronze age" prior to recorded civilization, starting with the savage who accidentally built a fire on top of some easily reducible ore, and discovered metal in the ashes, etc.; but as this method has been pursued times out of number to no particular purpose, we will confine ourselves to a summary of such facts as we can assemble. "Founders' hoards" of the bronze age are scattered over Western Europe, and indicate that smelting was done in shallow pits with charcoal. With the Egyptians we find occasional inscriptions showing small furnaces with forced draught, in early cases with a blow-pipe, but later--about 1500 B.C.--with bellows also. The crucible was apparently used by the Egyptians in secondary melting, such remains at Mt. Sinai probably dating before 2000 B.C. With the advent of the Prophets, and the first Greek literature--9th to 7th century B.C.--we find frequent references to bellows. The remains of smelting appliances at Mt. Laurion (500-300 B.C.) do not indicate much advance over the primitive hearth; however, at this locality we do find evidence of the ability to separate minerals by specific gravity, by washing crushed ore over inclined surfaces with a sort of buddle attachment. Stone grinding-mills were used to crush ore from the earliest times of Mt. Laurion down to the Middle Ages. About the beginning of the Christian era the writings of Diodorus, Strabo, Dioscorides, and Pliny indicate considerable advance in appliances. Strabo describes high stacks to carry off lead fumes; Dioscorides explains a furnace with a dust-chamber to catch _pompholyx_ (zinc oxide); Pliny refers to the upper and lower crucibles (a forehearth) and to the pillars and arches of the furnaces. From all of their descriptions we may conclude that the furnaces had then reached some size, and were, of course, equipped with bellows. At this time sulphide copper and lead ores were smelted; but as to fluxes, except lead for silver, and lead and soda for gold, we have practically no mention. Charcoal was the universal fuel for smelting down to the 18th century. Both Dioscorides and Pliny describe a distillation apparatus used to recover quicksilver. A formidable list of mineral products and metal alloys in use, indicate in themselves considerable apparatus, of the details of which we have no indication; in the main these products were lead sulphide, sulphate, and oxide (red-lead and litharge); zinc oxide; iron sulphide, oxide and sulphate; arsenic and antimony sulphides; mercury sulphide, sulphur, bitumen, soda, alum and potash; and of the alloys, bronze, brass, pewter, electrum and steel.

From this period to the period of the awakening of learning our only light is an occasional gleam from Theophilus and the Alchemists. The former gave a more detailed description of metallurgical appliances than had been done before, but there is little vital change apparent from the apparatus of Roman times. The Alchemists gave a great stimulus to industrial chemistry in the discovery of the mineral acids, and described distillation apparatus of approximately modern form.

The next period--the Renaissance--is one in which our descriptions are for the first time satisfactory, and a discussion would be but a review of _De Re Metallica_.

[2] See footnote 2, p. 267, on verbs used for roasting.

[3] Agricola has here either forgotten to take into account his three-palm-thick furnace walls, which will make the length of this long wall sixty-one feet, or else he has included this foot and a half in each case in the six-foot distance between the furnaces, so that the actual clear space is only four and a half feet between the furnace with four feet on the ends.

[4] The paucity of terms in Latin for describing structural members, and the consequent repetition of "beam" (_trabs_), "timber" (_tignum_), "billet" (_tigillum_), "pole" (_asser_), with such modifications as small, large, and transverse, and with long explanatory clauses showing their location, renders the original very difficult to follow. We have, therefore, introduced such terms as "posts," "tie-beams," "sweeps," "levers," "rafters," "sills," "moulding," "braces," "cleats," "supports," etc., as the context demands.

[5] This set of rafters appears to start from the longitudinal beam.

[6] Devices for creating an air current must be of very old invention, for it is impossible to conceive of anything but the crudest melting of a few simple ores without some forced draft. Wilkinson (The Ancient Egyptians, II, p. 316) gives a copy of an illustration of a foot-bellows from a tomb of the time of Thotmes III. (1500 B.C.). The rest of the world therefore, probably obtained them from the Egyptians. They are mentioned frequently in the Bible, the most pointed reference to metallurgical purposes being Jeremiah (VI, 29): "The bellows are burned, the lead is consumed in the fire; the founder melteth in vain; for the wicked are not plucked away." Strabo (VII, 3) states that Ephorus ascribed the invention of bellows to Anacharsis--a Thracian prince of about 600 B.C.

[7] This whole arrangement could be summarized by the word "hinge."

[8] The rim of this wheel is obviously made of segments fixed in two layers; the "disc" meaning the aggregate of segments on either side of the wheel.

[9] It has not been considered necessary to introduce the modern term _twyer_ in these descriptions, as the literal rendering is sufficiently clear.

[10] _Ferruminata_. These accretions are practically always near the hearth, and would correspond to English "sows," and therefore that term has been adopted. It will be noted that, like most modern metallurgists, Agricola offers no method for treating them. Pliny (XXXIV, 37) describes a "sow," and uses the verb _ferruminare_ (to weld or solder): "Some say that in the furnace there are certain masses of stone which become soldered together, and that the copper fuses around it, the mass not becoming liquid unless it is transferred to another furnace; it thus forms a sort of knot, as it were, of the metal."

[11] What are known in English as "crucible," "furnace well," "forehearth," "dipping-pot," "tapping-pot," "receiving-pot," etc., are in the text all _catinus_, _i.e._, crucible. For easier reading, however, we have assigned the names indicated in the context.

[12] _Panes ex pyrite conflati_. While the term _matte_ would cover most cases where this expression appears, and in many cases would be more expressive to the modern reader, yet there are instances where the expression as it stands indicates its particular origin, and it has been, therefore, considered advisable to adhere to the literal rendering.

[13] _Molybdaena_. See note 37, p. 476. It was the saturated furnace bottoms from cupellation.

[14] The four elements were earth, air, fire, and water.

[15] "Stones which easily melt in the fire." Nowhere in _De Re Metallica_ does the author explain these substances. However in the _Interpretatio_ (p. 465) he gives three genera or orders with their German equivalents, as follows:--"_Lapides qui igni liquescunt primi generis,--Schöne flüsse; secundi,--flüsse zum schmeltzen flock quertze; tertii,--quertze oder kiselstein."_ We confess our inability to make certain of most of the substances comprised in the first and second orders. We consider they were in part fluor-spar, and in any event the third order embraced varieties of quartz, flint, and silicious material generally. As the matter is of importance from a metallurgical point of view, we reproduce at some length Agricola's own statements on the subject from _Bermannus_ and _De Natura Fossilium_. In the latter (p. 268) he states: "Finally there now remain those stones which I call 'stones which easily melt in the fire,' because when thrown into hot furnaces they flow (_fluunt_). There are three orders (_genera_) of these. The first resembles the transparent gems; the second is not similar, and is generally not translucent; it is translucent in some part, and in rare instances altogether translucent. The first is sparingly found in silver and other mines; the second abounds in veins of its own. The third genus is the material from which glass is made, although it can also be made out of the other two. The stones of the first order are not only transparent, but are also resplendent, and have the colours of gems, for some resemble crystal, others emerald, heliotrope, lapis lazuli, amethyst, sapphire, ruby, _chrysolithus_, _morion_ (cairngorm?), and other gems, but they differ from them in hardness.... To the first genus belongs the _lapis alabandicus_ (modern albandite?), if indeed it was different from the alabandic carbuncle. It can be melted, according to Pliny, in the fire, and fused for the preparation of glass. It is black, but verging upon purple. It comes from Caria, near Alabanda, and from Miletus in the same province. The second order of stones does not show a great variety of colours, and seldom beautiful ones, for it is generally white, whitish, greyish, or yellowish. Because these (stones) very readily melt in the fire, they are added to the ores from which the metals are smelted. The small stones found in veins, veinlets, and the spaces between the veins, of the highest peaks of the Sudetic range (_Suditorum montium_), belong partly to this genus and partly to the first. They differ in size, being large and small; and in shape, some being round or angular or pointed; in colour they are black or ash-grey, or yellow, or purple, or violet, or iron colour. All of these are lacking in metals. Neither do the little stones contain any metals which are usually found in the streams where gold dust is collected by washing.... In the rivers where are collected the small stones from which tin is smelted, there are three genera of small stones to be found, all somewhat rounded and of very light weight, and devoid of all metals. The largest are black, both on the outside and inside, smooth and brilliant like a mirror; the medium-sized are either bluish black or ash-grey; the smallest are of a yellowish colour, somewhat like a silkworm. But because both the former and the latter stones are devoid of metals, and fly to pieces under the blows of the hammer, we classify them as sand or gravel. Glass is made from the stones of the third order, and particularly from sand. For when this is thrown into the heated furnace it is melted by the fire.... This kind of stone is either found in its own veins, which are occasionally very wide, or else scattered through the mines. It is less hard than flint, on account of which no fire can be struck from it. It is not transparent, but it is of many colours--that is to say, white, yellowish, ash-grey, brown, black, green, blue, reddish or red. This genus of stones occurs here and there in mountainous regions, on banks of rivers, and in the fields. Those which are black right through to the interior, and not merely on the surface, are more rare; and very frequently one coloured vein is intersected by another of a different colour--for instance, a white one by a red one; the green is often spotted with white, the ash-grey with black, the white with crimson. Fragments of these stones are frequently found on the surface of the earth, and in the running water they become polished by rubbing against stones of their own or of another genus. In this way, likewise, fragments of rocks are not infrequently shaped into spherical forms.... This stone is put to many uses; the streets are paved with it, whatever its colour; the blue variety is added to the ash of pines for making those other ashes which are used by wool-dyers. The white variety is burned, ground, and sifted, and from this they make the sand out of which glass is made. The whiter the sand is, the more useful it is."

Perusal of the following from _Bermannus_ (p. 458) can leave little doubt as to the first or second order being in part fluor-spar. Agricola derived the name _fluores_ from _fluo_ "to flow," and we in turn obtain "fluorite," or "fluorspar," from Agricola. "_Bermannus_.--These stones are similar to gems, but less hard. Allow me to explain word for word. Our miners call them _fluores_, not inappropriately to my mind, for by the heat of fire, like ice in the sun, they liquefy and flow away. They are of varied and bright colours. _Naevius_.--Theophrastus says of them that they are made by a conflux in the earth. These red _fluores_, to employ the words just used by you, are the ruby silver which you showed us before. _Bermannus_.--At the first glance it appears so, although it is not infrequently translucent. _Naevius_.--Then they are rubies? _Bermannus_.--Not that either. _Naevius_.--In what way, then, can they be distinguished from rubies? _Bermannus_.--Chiefly by this sign, that they glitter more feebly when translucent. Those which are not translucent may be distinguished from rubies. Moreover, _fluores_ of all kinds melt when they are subject to the first fire; rubies do not melt in fire. _Naevius_.--You distinguish well. _Bermannus_.--You see the other kind, of a paler purple colour? _Naevius_.--They appear to be an inferior kind of amethyst, such as are found in many places in Bohemia. _Bermannus_.--Indeed, they are not very dissimilar, therefore the common people who do not know amethysts well, set them in rings for gems, and they are easily sold. The third kind, as you see here, is white. _Naevius_.--I should have thought it a crystal. _Bermannus_.--A fourth is a yellow colour, a fifth ash colour, a sixth blackish. Some are violet, some green, others gold-coloured. _Anton_.--What is the use of _fluores_? _Bermannus_.--They are wont to be made use of when metals are smelted, as they cause the material in the fire to be much more fluid, exactly like a kind of stone which we said is made from pyrites (matte); it is, indeed, made not far from here, at Breitenbrunn, which is near Schwarzenberg. Moreover, from _fluores_ they can make colours which artists use."

[16] _Stannum_. (_Interpretatio_,--_werck_, modern _werk_). This term has been rendered throughout as "silver-lead" or "silver-lead alloy." It was the argentiferous lead suitable for cupellation. Agricola, in using it in this sense, was no doubt following his interpretation of its use by Pliny. Further remarks upon this subject will be found in note 33, p. 473.

[17] _Expirare_,--to exhale or blow out.

[18] _Rhetos_. The ancient Rhaetia comprised not only the greater part of Tyrol, but also parts of Switzerland and Lombardy. The mining section was, however, in Tyrol.

[19] _Noricum_ was a region south of the Danube, embracing not only modern Styria, but also parts of Austria, Salzberg, and Carinthia.

[20] One _drachma_ of gold to a _centumpondium_ would be (if we assume these were Roman weights) 3 ozs. 1 dwt. Troy per short ton. One-half _uncia_ of silver would be 12 ozs. 3 dwts. per short ton.

[21] For discussion of these fluxes see note page 232.

[22] _Carni_. Probably the people of modern Austrian Carniola, which lies south of Styria and west of Croatia.

[23] HISTORICAL NOTE ON SMELTING LEAD AND SILVER.--The history of lead and silver smelting is by no means a sequent array of exact facts. With one possible exception, lead does not appear upon the historical horizon until long after silver, and yet their metallurgy is so inextricably mixed that neither can be considered wholly by itself. As silver does not occur native in any such quantities as would have supplied the amounts possessed by the Ancients, we must, therefore, assume its reduction by either (1) intricate chemical processes, (2) amalgamation, (3) reduction with copper, (4) reduction with lead. It is impossible to conceive of the first with the ancient knowledge of chemistry; the second (see note 12, p. 297) does not appear to have been known until after Roman times; in any event, quicksilver appears only at about 400 B.C. The third was impossible, as the parting of silver from copper without lead involves metallurgy only possible during the last century. Therefore, one is driven to the conclusion that the fourth case obtained, and that the lead must have been known practically contemporaneously with silver. There is a leaden figure exhibited in the British Museum among the articles recovered from the Temple of Osiris at Abydos, and considered to be of the Archaic period--prior to 3800 B.C. The earliest known Egyptian silver appears to be a necklace of beads, supposed to be of the XII. Dynasty (2400 B.C.), which is described in the 17th Memoir, Egyptian Exploration Fund (London, 1898, p. 22). With this exception of the above-mentioned lead specimen, silver articles antedate positive evidence of lead by nearly a millennium, and if we assume lead as a necessary factor in silver production, we must conclude it was known long prior to any direct (except the above solitary possibility) evidence of lead itself. Further, if we are to conclude its necessary association with silver, we must assume a knowledge of cupellation for the parting of the two metals. Lead is mentioned in 1500 B.C. among the spoil captured by Thotmes III. Leaden objects have frequently been found in Egyptian tombs as early as Rameses III. (1200 B.C.). The statement is made by Pulsifer (Notes for a History of Lead, New York 1888, p. 146) that Egyptian pottery was glazed with lead. We have been unable to find any confirmation of this. It may be noted, incidentally, that lead is not included in the metals of the "Tribute of Yü" in the Shoo King (The Chinese Classics, 2500 B.C.?), although silver is so included.

After 1200 or 1300 B.C. evidences of the use of lead become frequent. Moses (Numbers XXXI, 22-23) directs the Israelites with regard to their plunder from the Midianites (1300 B.C.): "Only the gold and the silver, the brass [_sic_], the iron, the tin, and the lead. Everything that may abide the fire, ye shall make it go through the fire, and it shall be clean; nevertheless, it shall be purified with the water of separation, and all that abideth not the fire ye shall make go through the water." Numerous other references occur in the Scriptures (Psalms XII, 6; Proverbs XVII, 3; XXV, 4; etc.), one of the most pointed from a metallurgical point of view being that of Jeremiah (600 B.C.), who says (VI, 29-30): "The bellows are burned, the lead is consumed of the fire; the founder melteth in vain; for the wicked are not plucked away. Reprobate silver shall men call them because the Lord hath rejected them." From the number of his metaphors in metallurgical terms we may well conclude that Jeremiah was of considerable metallurgical experience, which may account for his critical tenor of mind. These Biblical references all point to a knowledge of separating silver and lead. Homer mentions lead (Iliad XXIV, 109), and it has been found in the remains of ancient Troy and Mycenae (H. Schliemann, "Troy and its Remains," London, 1875, and "Mycenae," New York, 1877). Both Herodotus (I, 186) and Diodorus (II, 1) speak of the lead used to fix iron clamps in the stone bridge of Nitocris (600 B.C.) at Babylon.

Our best evidence of ancient lead-silver metallurgy is the result of the studies at Mt. Laurion by Edouard Ardaillon (_Mines du Laurion dans l'Antiquité_, Paris, 1897). Here the very extensive old workings and the slag heaps testify to the greatest activity. The re-opening of the mines in recent years by a French Company has well demonstrated their technical character, and the frequent mention in Greek History easily determines their date. These deposits of argentiferous galena were extensively worked before 500 B.C. and while the evidence of concentration methods is ample, there is but little remaining of the ancient smelters. Enough, however, remains to demonstrate that the galena was smelted in small furnaces at low heat, with forced draught, and that it was subsequently cupelled. In order to reduce the sulphides the ancient smelters apparently depended upon partial roasting in the furnace at a preliminary period in reduction, or else upon the ferruginous character of the ore, or upon both. See notes p. 27 and p. 265. Theognis (6th century B.C.) and Hippocrates (5th century B.C.) are frequently referred to as mentioning the refining of gold with lead; an inspection of the passages fails to corroborate the importance which has been laid upon them. Among literary evidences upon lead metallurgy of later date, Theophrastus (300 B.C.) describes the making of white-lead with lead plates and vinegar. Diodorus Siculus (1st century B.C.), in his well-known quotation from Agatharchides (2nd century B.C.) with regard to gold mining and treatment in Egypt, describes the refining of gold with lead. (See note 8, p. 279.) Strabo (63 B.C.-24 A.D.) says (III, 2, 8): "The furnaces for silver are constructed lofty in order that the vapour, which is dense and pestilent, may be raised and carried off." And again (III, 2, 10), in quoting from Polybius (204-125 B.C.): "Polybius, speaking of the silver mines of New Carthage, tells us that they are extremely large, distant from the city about 20 stadia, and occupy a circuit of 400 stadia; that there are 40,000 men regularly engaged in them, and that they yield daily to the Roman people (a revenue of) 25,000 drachmae. The rest of the process I pass over, as it is too long; but as for the silver ore collected, he tells us that it is broken up and sifted through sieves over water; that what remains is to be again broken, and the water having been strained off it is to be sifted and broken a third time. The dregs which remain after the fifth time are to be melted, and the lead being poured off, the silver is obtained pure. These silver mines still exist; however, they are no longer the property of the State, neither these nor those elsewhere, but are possessed by private individuals. The gold mines, on the contrary, nearly all belong to the State. Both at Castlon and other places there are singular lead mines worked. They contain a small proportion of silver, but not sufficient to pay for the expense of refining" (Hamilton's Trans.). Dioscorides (1st century A.D.), among his medicines, describes several varieties of litharge, their origin, and the manner of making white-lead (see on pp. 465, 440), but he gives no very tangible information on lead smelting. Pliny, at the same period in speaking of silver, (XXXIII, 31), says: "After this we speak of silver, the next folly. Silver is only found in shafts, there being no indications like shining particles as in the case of gold. This earth is sometimes red, sometimes of an ashy colour. It is impossible to melt it except with lead ore (_vena plumbi_), called _galena_, which is generally found next to silver veins. And this the same agency of fire separates part into lead, which floats on the silver like oil on water." (We have transferred lead and silver in this last sentence, otherwise it means nothing.) Also (XXXIV, 47) he says: "There are two different sources of lead, it being smelted from its own ore, whence it comes without the admixture of any other substance, or else from an ore which contains it in common with silver. The metal, which flows liquid at the first melting in the furnace, is called _stannum_ that at the second melting is silver; that which remains in the furnace is _galena_, which is added to a third part of the ore. This being again melted, produces lead with a deduction of two-ninths." We have, despite some grammatical objections, rendered this passage quite differently from other translators, none of whom have apparently had any knowledge of metallurgy; and we will not, therefore, take the several pages of space necessary to refute their extraordinary and unnecessary hypotheses. From a metallurgical point of view, two facts must be kept in mind,--first, that _galena_ in this instance was the same substance as _molybdaena_, and they were both either a variety of litharge or of lead carbonates; second, that the _stannum_ of the Ancients was silver-lead alloy. Therefore, the metallurgy of this paragraph becomes a simple melting of an argentiferous lead ore, its subsequent cupellation, with a return of the litharge to the furnace. Pliny goes into considerable detail as to varieties of litharge, for further notes upon which see p. 466. The Romans were most active lead-silver miners, not only in Spain, but also in Britain. There are scores of lead pigs of the Roman era in various English museums, many marked "_ex argent_." Bruce (The Roman Wall, London, 1852, p. 432) describes some Roman lead furnaces in Cumberland where the draught was secured by driving a tapering tunnel into the hills. The Roman lead slag ran high in metal, and formed a basis for quite an industry in England in the early 18th century (Hunt, British Mining, London, 1887, p. 26, etc.). There is nothing in mediæval literature which carries us further with lead metallurgy than the knowledge displayed by Pliny, until we arrive at Agricola's period. The history of cupellation is specially dealt with in note on p. 465.

[25] _Cadmia_. In the German Translation this is given as _kobelt_. It would be of uncertain character, but no doubt partially furnace calamine. (See note on p. 112.)

[26] _Pompholyx_. (_Interpretatio_ gives the German as _Weisser hütten rauch als ober dem garherde und ober dem kupfer ofen_). This was the impure protoxide of zinc deposited in the furnace outlets, and is modern "tutty." The ancient products, no doubt, contained arsenical oxides as well. It was well known to the Ancients, and used extensively for medicinal purposes, they dividing it into two species--_pompholyx_ and _spodos_. The first adequate description is by Dioscorides (V, 46): "_Pompholyx_ differs from _spodos_ in species, not in genus. For _spodos_ is blacker, and is often heavier, full of straws and hairs, like the refuse that is swept from the floors of copper smelters. But _pompholyx_ is fatty, unctuous, white and light enough to fly in the air. Of this there are two kinds--the one inclines to sky blue and is unctuous; the other is exceedingly white, and is extremely light. White _pompholyx_ is made every time that the artificer, in the preparation and perfecting of copper (brass?) sprinkles powdered _cadmia_ upon it to make it more perfect, for the soot which rises being very fine becomes _pompholyx_. Other _pompholyx_ is made, not only in working copper (brass?), but is also made from _cadmia_ by continually blowing with bellows. The manner of doing it is as follows:--The furnace is constructed in a two-storied building, and there is a medium-sized aperture opening to the upper chamber; the building wall nearest the furnace is pierced with a small opening to admit the nozzle of the bellows. The building must have a fair-sized door for the artificer to pass in and out. Another small building must adjoin this, in which are the bellows and the man who works them. Then the charcoal in the furnace is lighted, and the artificer continually throws broken bits of _cadmia_ from the place above the furnace, whilst his assistant, who is below, throws in charcoals, until all of the _cadmia_ inside is consumed. By this means the finest and lightest part of the stuff flies up with the smoke to the upper chamber, and adheres to the walls of the roof. The substance which is thus formed has at first the appearance of bubbles on water, afterward increasing in size, it looks like skeins of wool. The heaviest parts settle in the bottom, while some fall over and around the furnaces, and some lie on the floor of the building. This latter part is considered inferior, as it contains a lot of earth and becomes full of dirt."

Pliny (XXXIV, 33) appears somewhat confused as to the difference between the two species: "That which is called _pompholyx_ and _spodos_ is found in the copper-smelting furnaces, the difference between them being that _pompholyx_ is separated by washing, while _spodos_ is not washed. Some have called that which is white and very light _pompholyx_, and it is the soot of copper and _cadmia_; whereas _spodos_ is darker and heavier. It is scraped from the walls of the furnace, and is mixed with particles of metal, and sometimes with charcoal." (XXXIV, 34.) "The Cyprian _spodos_ is the best. It is formed by fusing _cadmia_ with copper ore. This being the lightest part of the metal, it flies up in the fumes from the furnace, and adheres to the roof, being distinguished from the soot by its whiteness. That which is less white is immature from the furnace, and it is this which some call '_pompholyx_.'" Agricola (_De Natura Fossilium_, p. 350) traverses much the same ground as the authors previously quoted, and especially recommends the _pompholyx_ produced when making brass by melting alternate layers of copper and calamine (_cadmia fossilis_).

[27] _Oleo, ex fece vini sicca confecto_. This oil, made from argol, is probably the same substance mentioned a few lines further on as "wine," distilled by heating argol in a retort. Still further on, salt made from argol is mentioned. It must be borne in mind that this argol was crude tartrates from wine vats, and probably contained a good deal of organic matter. Heating argol sufficiently would form potash, but that the distillation product could be anything effective it is difficult to see.

[28] _Aqua valens_. No doubt mainly nitric acid, the preparation of which is explained at length in Book X, p. 439.

[29] _Quod cum ignis consumit non modo una cum eo, quae ipsius stibii vis est, aliqua auri particula, sed etiam argenti, si cum auro fuerit permistum, consumitur._ The meaning is by no means clear. On p. 451 is set out the old method of parting silver from gold with antimony sulphide, of which this may be a variation. The silver combines with sulphur, and the reduced antimony forms an alloy with the gold. The added iron and copper would also combine with the sulphur from the antimony sulphide, and no doubt assist by increasing the amount of free collecting agent and by increasing the volume of the matte. (See note 17, p. 451.)

[30] There follow eight different methods of treating crude bullion or rich concentrates. In a general way three methods are involved,--1st, reduction with lead or antimony, and cupellation; 2nd, reduction with silver, and separation with nitric acid; 3rd, reduction with lead and silver, followed by cupellation and parting with nitric acid. The use of sulphur or antimony sulphide would tend to part out a certain amount of silver, and thus obtain fairly pure bullion upon cupellation. But the introduction of copper could only result deleteriously, except that it is usually accompanied by sulphur in some form, and would thus probably pass off harmlessly as a matte carrying silver. (See note 33 below.)

[31] It is not very clear where this lead comes from. Should it be antimony? The German translation gives this as "silver."

[32] These powders are described in Book VII., p. 236. It is difficult to say which the second really is. There are numbers of such recipes in the _Probierbüchlein_ (see Appendix B), with which a portion of these are identical.

[33] A variety of methods are involved in this paragraph: 1st, crude gold ore is smelted direct; 2nd, gold concentrates are smelted in a lead bath with some addition of iron--which would simply matte off--the lead bullion being cupelled; 3rd, roasted and unroasted pyrites and _cadmia_ (probably blende, cobalt, arsenic, etc.) are melted into a matte; this matte is repeatedly roasted, and then re-melted in a lead bath; 4th, if the material "flies out of the furnace" it is briquetted with iron ore and lime, and the briquettes smelted with copper matte. Three products result: (_a_) slag; (_b_) matte; (_c_) copper-gold-silver alloy. The matte is roasted, re-smelted with lead, and no doubt a button obtained, and further matte. The process from this point is not clear. It appears that the copper bullion is melted with lead, and normally this product would be taken to the liquation furnace, but from the text it would appear that the lead-copper bullion was melted again with iron ore and pyrites, in which case some of the copper would be turned into the matte, and the lead alloy would be richer in gold and silver.

HISTORICAL NOTE ON GOLD.--There is ample evidence of gold being used for ornamental purposes prior to any human record. The occurrence of large quantities of gold in native form, and the possibility of working it cold, did not necessitate any particular metallurgical ingenuity. The earliest indications of metallurgical work are, of course, among the Egyptians, the method of washing being figured as early as the monuments of the IV Dynasty (prior to 3800 B.C.). There are in the British Museum two stelae of the XII Dynasty (2400 B.C.) (144 Bay 1 and 145 Bay 6) relating to officers who had to do with gold mining in Nubia, and upon one there are references to working what appears to be ore. If this be true, it is the earliest reference to this subject. The Papyrus map (1500 B.C.) of a gold mine, in the Turin Museum (see note 16, p. 129), probably refers to a quartz mine. Of literary evidences there is frequent mention of refining gold and passing it through the fire in the Books of Moses, arts no doubt learned from the Egyptians. As to working gold, ore as distinguished from alluvial, we have nothing very tangible, unless it be the stelae above, until the description of Egyptian gold mining by Agatharchides (see note 8, p. 279). This geographer, of about the 2nd century B.C., describes very clearly indeed the mining, crushing, and concentration of ore and the refining of the concentrates in crucibles with lead, salt, and barley bran. We may mention in passing that Theognis (6th Century B.C.) is often quoted as mentioning the refining of gold with lead, but we do not believe that the passage in question (1101): "But having been put to the test and being rubbed beside (or against) lead as being refined gold, you will be fair," etc.; or much the same statement again (418) will stand much metallurgical interpretation. In any event, the myriads of metaphorical references to fining and purity of gold in the earliest shreds of literature do not carry us much further than do those of Shakespeare or Milton. Vitruvius and Pliny mention the recovery or refining of gold with mercury (see note 12, p. 297 on Amalgamation); and it appears to us that gold was parted from silver by cementation with salt prior to the Christian era. We first find mention of parting with sulphur in the 12th century, with nitric acid prior to the 14th century, by antimony sulphide prior to the 15th century, and by cementation with nitre by Agricola. (See historical note on parting gold and silver, p. 458.) The first mention of parting gold from copper occurs in the early 16th century (see note 24, p. 462). The first comprehensive description of gold metallurgy in all its branches is in _De Re Metallica_.

[34] _Rudis_ silver comprised all fairly pure silver ores, such as silver sulphides, chlorides, arsenides, etc. This is more fully discussed in note 6, p. 108.

[35] _Evolent_,--volatilize?

[36] _Lapidis plumbarii facile liquescentis_. The German Translation gives _glantz_, _i.e._, Galena, and the _Interpretatio_ also gives _glantz_ for _lapis plumbarius_. We are, however, uncertain whether this "easily melting" material is galena or some other lead ore.

[37] _Molybdaena_ is usually hearth-lead in _De Re Metallica_, but the German translation in this instance uses _pleyertz_, lead ore. From the context it would not appear to mean hearth-lead--saturated bottoms of cupellation furnaces--for such material would not contain appreciable silver. Agricola does confuse what are obviously lead carbonates with his other _molybdaena_ (see note 37, p. 476).

[38] The term _cadmia_ is used in this paragraph without the usual definition. Whether it was _cadmia fornacis_ (furnace accretions) or _cadmia metallica_ (cobalt-arsenic-blende mixture) is uncertain. We believe it to be the former.

[39] _Ramentum si lotura ex argento rudi_. This expression is generally used by the author to indicate concentrates, but it is possible that in this sentence it means the tailings after washing rich silver minerals, because the treatment of the _rudis_ silver has been already discussed above.

[40] _Ustum_. This might be rendered "burnt." In any event, it seems that the material is sintered.

[41] _Aes purum sive proprius ei color insederit, sive chrysocolla vel caeruleo fuerit tinctum, et rude plumbei coloris, aut fusci, aut nigri._ There are six copper minerals mentioned in this sentence, and from our study of Agricola's _De Natura Fossilium_ we hazard the following:--_Proprius ei color insederit_,--"its own colour,"--probably cuprite or "ruby copper." _Tinctum chrysocolla_--partly the modern mineral of that name and partly malachite. _Tinctum caeruleo_, partly azurite and partly other blue copper minerals. _Rude plumbei coloris_,--"lead coloured,"--was certainly chalcocite (copper glance). We are uncertain of _fusci aut nigri_, but they were probably alteration products. For further discussion see note on p. 109.

[42] HISTORICAL NOTE ON COPPER SMELTING.--The discoverer of the reduction of copper by fusion, and his method, like the discoverer of tin and iron, will never be known, because he lived long before humanity began to make records of its discoveries and doings. Moreover, as different races passed independently and at different times through the so-called "Bronze Age," there may have been several independent discoverers. Upon the metallurgy of pre-historic man we have some evidence in the many "founders' hoards" or "smelters' hoards" of the Bronze Age which have been found, and they indicate a simple shallow pit in the ground into which the ore was placed, underlaid with charcoal. Rude round copper cakes eight to ten inches in diameter resulted from the cooling of the metal in the bottom of the pit. Analyses of such Bronze Age copper by Professor Gowland and others show a small percentage of sulphur, and this is possible only by smelting oxidized ores. Copper objects appear in the pre-historic remains in Egypt, are common throughout the first three dynasties, and bronze articles have been found as early as the IV Dynasty (from 3800 to 4700 B.C., according to the authority adopted). The question of the origin of this bronze, whether from ores containing copper and tin or by alloying the two metals, is one of wide difference of opinion, and we further discuss the question in note 53, p. 411, under Tin. It is also interesting to note that the crucible is the emblem of copper in the hieroglyphics. The earliest source of Egyptian copper was probably the Sinai Peninsula, where there are reliefs as early as Seneferu (about 3700 B.C.), indicating that he worked the copper mines. Various other evidences exist of active copper mining prior to 2500 B.C. (Petrie, Researches in Sinai, London, 1906, p. 51, etc.). The finding of crucibles here would indicate some form of refining. Our knowledge of Egyptian copper metallurgy is limited to deductions from their products, to a few pictures of crude furnaces and bellows, and to the minor remains on the Sinai Peninsula; none of the pictures were, so far as we are aware, prior to 2300 B.C., but they indicate a considerable advance over the crude hearth, for they depict small furnaces with forced draught--first a blow-pipe, and in the XVIII Dynasty (about 1500 B.C.) the bellows appear. Many copper articles have been found scattered over the Eastern Mediterranean and Asia Minor of pre-Mycenaean Age, some probably as early as 3000 B.C. This metal is mentioned in the "Tribute of Yü" in the Shoo King (2500 B.C.?); but even less is known of early Chinese metallurgy than of the Egyptian. The remains of Mycenaean, Phoenician, Babylonian, and Assyrian civilizations, stretching over the period from 1800 to 500 B.C., have yielded endless copper and bronze objects, the former of considerable purity, and the latter a fairly constant proportion of from 10% to 14% tin. The copper supply of the pre-Roman world seems to have come largely, first from Sinai, and later from Cyprus, and from the latter comes our word copper, by way of the Romans shortening _aes cyprium_ (Cyprian copper) to _cuprum_. Research in this island shows that it produced copper from 3000 B.C., and largely because of its copper it passed successively under the domination of the Egyptians, Assyrians, Phoenicians, Greeks, Persians, and Romans. The bronze objects found in Cyprus show 2% to 10% of tin, although tin does not, so far as modern research goes, occur on that island. There can be no doubt that the Greeks obtained their metallurgy from the Egyptians, either direct or second-hand--possibly through Mycenae or Phoenicia. Their metallurgical gods and the tradition of Cadmus indicate this much.

By way of literary evidences, the following lines from Homer (Iliad, XVIII.) have interest as being the first preserved description in any language of a metallurgical work. Hephaestus was much interrupted by Thetis, who came to secure a shield for Achilles, and whose general conversation we therefore largely omit. We adopt Pope's translation:--

There the lame architect the goddess found Obscure in smoke, his forges flaming round, While bathed in sweat from fire to fire he flew; And puffing loud the roaring bellows blew. * * * In moulds prepared, the glowing ore (metal?) he pours. * * * "Vouchsafe, oh Thetis! at our board to share The genial rites and hospitable fare; While I the labours of the forge forego, And bid the roaring bellows cease to blow." Then from his anvil the lame artist rose; Wide with distorted legs oblique he goes, And stills the bellows, and (in order laid) Locks in their chests his instruments of trade; Then with a sponge, the sooty workman dress'd His brawny arms embrown'd and hairy breast. * * * Thus having said, the father of the fires To the black labours of his forge retires. Soon as he bade them blow the bellows turn'd Their iron mouths; and where the furnace burn'd Resounding breathed: at once the blast expires, And twenty forges catch at once the fires; Just as the God directs, now loud, now low, They raise a tempest, or they gently blow; In hissing flames huge silver bars are roll'd, And stubborn brass (copper?) and tin, and solid gold; Before, deep fixed, the eternal anvils stand. The ponderous hammer loads his better hand; His left with tongs turns the vex'd metal round. And thick, strong strokes, the doubling vaults rebound Then first he formed the immense and solid shield;

Even if we place the siege of Troy at any of the various dates from 1350 to 1100 B.C., it does not follow that the epic received its final form for many centuries later, probably 900-800 B.C.; and the experience of the race in metallurgy at a much later period than Troy may have been drawn upon to fill in details. It is possible to fill a volume with indirect allusion to metallurgical facts and to the origins of the art, from Greek mythology, from Greek poetry, from the works of the grammarians, and from the Bible. But they are of no more technical value than the metaphors from our own tongue. Greek literature in general is singularly lacking in metallurgical description of technical value, and it is not until Dioscorides (1st Century A.D.) that anything of much importance can be adduced. Aristotle, however, does make an interesting reference to what may be brass (see note on p. 410), and there can be no doubt that if we had the lost work of Aristotle's successor, Theophrastus (372-288 B.C.), on metals we should be in possession of the first adequate work on metallurgy. As it is, we find the green and blue copper minerals from Cyprus mentioned in his "Stones." And this is the first mention of any particular copper ore. He also mentions (XIX.) pyrites "which melt," but whether it was a copper variety cannot be determined. Theophrastus further describes the making of verdigris (see note 4, p. 440). From Dioscorides we get a good deal of light on copper treatment, but as his objective was to describe medicinal preparations, the information is very indirect. He states (V, 100) that "pyrites is a stone from which copper is made." He mentions _chalcitis_ (copper sulphide, see note on, p. 573); while his _misy_, _sory_, _melanteria_, _caeruleum_, and _chrysocolla_ were all oxidation copper or iron minerals. (See notes on p. 573.) In giving a method of securing _pompholyx_ (zinc oxide), "the soot flies up when the copper refiners sprinkle powdered _cadmia_ over the molten metal" (see note 26, p. 394); he indirectly gives us the first definite indication of making brass, and further gives some details as to the furnaces there employed, which embraced bellows and dust chambers. In describing the making of flowers of copper (see note 26, p. 538) he states that in refining copper, when the "molten metal flows through its tube into a receptacle, the workmen pour cold water on it, the copper spits and throws off the flowers." He gives the first description of vitriol (see note 11, p. 572), and describes the pieces as "shaped like dice which stick together in bunches like grapes." Altogether, from Dioscorides we learn for the first time of copper made from sulphide ores, and of the recovery of zinc oxides from furnace fumes; and he gives us the first certain description of making brass, and finally the first notice of blue vitriol.

The next author we have who gives any technical detail of copper work is Pliny (23-79 A.D.), and while his statements carry us a little further than Dioscorides, they are not as complete as the same number of words could have afforded had he ever had practical contact with the subject, and one is driven to the conclusion that he was not himself much of a metallurgist. Pliny indicates that copper ores were obtained from veins by underground mining. He gives the same minerals as Dioscorides, but is a good deal confused over _chrysocolla_ and _chalcitis_. He gives no description of the shapes of furnaces, but frequently mentions the bellows, and speaks of the _cadmia_ and _pompholyx_ which adhered to the walls and arches of the furnaces. He has nothing to say as to whether fluxes are used or not. As to fuel, he says (XXXIII, 30) that "for smelting copper and iron pine wood is the best." The following (XXXIV, 20) is of the greatest interest on the subject:--"Cyprian copper is known as _coronarium_ and _regulare_; both are ductile.... In other mines are made that known as _regulare_ and _caldarium_. These differ, because the _caldarium_ is only melted, and is brittle to the hammer; whereas the _regulare_ is malleable or ductile. All Cyprian copper is this latter kind. But in other mines with care the difference can be eliminated from _caldarium_, the impurities being carefully purged away by smelting with fire, it is made into _regulare_. Among the remaining kinds of copper the best is that of Campania, which is most esteemed for vessels and utensils. This kind is made in several ways. At Capua it is melted with wood, not with charcoal, after which it is sprinkled with water and washed through an oak sieve. After it is melted a number of times Spanish _plumbum argentum_ (probably pewter) is added to it in proportion of ten pounds of the lead to one hundred pounds of copper, and thereby it is made pliable and assumes that pleasing colour which in other kinds of copper is effected by oil and the sun. In many parts of the Italian provinces they make a similar kind of metal; but there they add eight pounds of lead, and it is re-melted over charcoal because of the scarcity of wood. Very different is the method carried on in Gaul, particularly where the ore is smelted between red hot stones, for this burns the metal and renders it black and brittle. Moreover, it is re-melted only a single time, whereas the oftener this operation is repeated the better the quality becomes. It is well to remark that all copper fuses best when the weather is intensely cold." The red hot stones in Gaul were probably as much figments of imagination as was the assumption of one commentator that they were a reverberatory furnace. Apart from the above, Pliny says nothing very direct on refining copper. It is obvious that more than one melting was practised, but that anything was known of the nature of oxidation by a blast and reduction by poling is uncertain. We produce the three following statements in connection with some bye-products used for medicinal purposes, which at least indicate operations subsequent to the original melting. As to whether they represent this species of refining or not, we leave it to the metallurgical profession (XXXIV, 24):--"The flowers of copper are used in medicine; they are made by fusing copper and moving it to another furnace, where the rapid blast separates it into a thousand particles, which are called flowers. These scales are also made when the copper cakes are cooled in water (XXXIV, 35). _Smega_ is prepared in the copper works; when the metal is melted and thoroughly smelted charcoal is added to it and gradually kindled; after this, being blown upon by a powerful bellows, it spits out, as it were, copper chaff (XXXIV, 37). There is another product of these works easily distinguished from _smega_, which the Greeks call _diphrygum_. This substance has three different origins.... A third way of making it is from the residues which fall to the bottom in copper furnaces. The difference between the different substances (in the furnace) is that the copper itself flows into a receiver; the slag makes its escape from the furnace; the flowers float on the top (of the copper?), and the _diphrygum_ remains behind. Some say that in the furnace there are certain masses of stone which, being smelted, become soldered together, and that the copper fuses around it, the mass not becoming liquid unless it is transferred to another furnace. It thus forms a sort of knot, as it were, in the metal."

Pliny is a good deal confused over the copper alloys, failing to recognise _aurichalcum_ as the same product as that made by mixing _cadmia_ and molten copper. Further, there is always the difficulty in translation arising from the fact that the Latin _aes_ was indiscriminately copper, brass, and bronze. He does not, except in one instance (XXXIV., 2), directly describe the mixture of _cadmia_ and copper. "Next to Livian (copper) this kind (_corduban_, from Spain) most readily absorbs _cadmia_, and becomes almost as excellent as _aurichalcum_ for making _sesterces_." As to bronze, there is no very definite statement; but the _argentatium_ given in the quotation above from XXXIV, 20, is stated in XXXIV, 48, to be a mixture of tin and lead. The Romans carried on most extensive copper mining in various parts of their empire; these activities extended from Egypt through Cyprus, Central Europe, the Spanish Peninsula, and Britain. The activity of such works is abundantly evidenced in the mines, but very little remains upon the surface to indicate the equipment; thus, while mining methods are clear enough, the metallurgy receives little help from these sources. At Rio Tinto there still remain enormous slag heaps from the Romans, and the Phoenician miners before them. Professor W. A. Carlyle informs us that the ore worked must have been almost exclusively sulphides, as only negligible quantities of carbonates exist in the deposits; they probably mixed basic and siliceous ores. There is some evidence of roasting, and the slags run from .2 to .6%. They must have run down mattes, but as to how they ultimately arrived at metallic copper there is no evidence to show.

The special processes for separating other metals from copper by liquation and matting, or of refining by poling, etc., are none of them clearly indicated in records or remains until we reach the 12th century. Here we find very adequate descriptions of copper smelting and refining by the Monk Theophilus (see Appendix B). We reproduce two paragraphs of interest from Hendrie's excellent translation (p. 305 and 313): "Copper is engendered in the earth. When a vein of which is found, it is acquired with the greatest labour by digging and breaking. It is a stone of a green colour and most hard, and naturally mixed with lead. This stone, dug up in abundance, is placed upon a pile and burned after the manner of chalk, nor does it change colour, but yet loses its hardness, so that it can be broken up. Then, being bruised small, it is placed in the furnace; coals and the bellows being applied, it is incessantly forged by day and night. This should be done carefully and with caution; that is, at first coals are placed in, then small pieces of stone are distributed over them, and again coals, and then stone anew, and it is thus arranged until it is sufficient for the size of the furnace. And when the stone has commenced to liquefy, the lead flows out through some small cavities, and the copper remains within. (313) Of the purification of copper. Take an iron dish of the size you wish, and line it inside and out with clay strongly beaten and mixed, and it is carefully dried. Then place it before a forge upon the coals, so that when the bellows act upon it the wind may issue partly within and partly above it, and not below it. And very small coals being placed round it, place copper in it equally, and add over it a heap of coals. When, by blowing a long time, this has become melted, uncover it and cast immediately fine ashes of coals over it, and stir it with a thin and dry piece of wood as if mixing it, and you will directly see the burnt lead adhere to these ashes like a glue. Which being cast out again superpose coals, and blowing for a long time, as at first, again uncover it, and then do as you did before. You do this until at length, by cooking it, you can withdraw the lead entirely. Then pour it over the mould which you have prepared for this, and you will thus prove if it be pure. Hold it with pincers, glowing as it is, before it has become cold, and strike it with a large hammer strongly over the anvil, and if it be broken or split you must liquefy it anew as before."

The next writer of importance was Biringuccio, who was contemporaneous with Agricola, but whose book precedes _De Re Metallica_ by 15 years. That author (III, 2) is the first to describe particularly the furnace used in Saxony and the roasting prior to smelting, and the first to mention fluxes in detail. He, however, describes nothing of matte smelting; in copper refining he gives the whole process of poling, but omits the pole. It is not until we reach _De Re Metallica_ that we find adequate descriptions of the copper minerals, roasting, matte smelting, liquation, and refining, with a wealth of detail which eliminates the necessity for a large amount of conjecture regarding technical methods of the time.

[43] _Cadmia metallica fossilis_ (see note on p. 112). This was undoubtedly the complex cobalt-arsenic-zinc minerals found in Saxony. In the German translation, however, this is given as _Kalmey_, calamine, which is unlikely from the association with pyrites.

[44] The Roman _modius_ (_modulus_?) held about 550 cubic inches, the English peck holding 535 cubic inches. Then, perhaps, his seven _moduli_ would be roughly, 1 bushel 3 pecks, and 18 vessels full would be about 31 bushels--say, roughly, 5,400 lbs. of ore.

[45] Exhausted liquation cakes (_panes aerei fathiscentes_). This is the copper sponge resulting from the first liquation of lead, and still contains a considerable amount of lead. The liquation process is discussed in great detail in Book XI.

[46] The method of this paragraph involves two main objectives--first, the gradual enrichment of matte to blister copper; and, second, the creation of large cakes of copper-lead-silver alloy of suitable size and ratio of metals for liquation. This latter process is described in detail in Book XI. The following groupings show the circuit of the various products, the "lbs." being Roman _librae_:--

CHARGE. PRODUCTS.

{ Crude ore 5,400 lbs. } Primary matte (1) 600 lbs. { Lead slags 3 cartloads } 1st { Schist 1 cartload } Silver-copper alloy (A) 50 " { Flux 20 lbs. } { Concentrates from } Slags (B) { slags & accretions Small quantity }

{ Primary matte (1) 1,800 lbs. } Secondary matte (2) 1,800 lbs. { Hearth-lead & litharge 1,200 " } { Lead ore 300 " } Silver-copper-lead 2nd { Rich hard cakes (A_{4}) 500 " } alloy (liquation { Liquated cakes 200 " } cakes) (A_{2}) 1,200 " { Slags (B) } { Concentrates from } Slags (B_{2}) { accretions }

{ Secondary matte (2) 1,800 lbs. } Tertiary matte (3) 1,300 lbs. { Hearth-lead & litharge 1,200 " } Silver-copper-lead { Lead ore 300 " } alloy (liquation 3rd { Rich hard cakes (A_{4}) 500 " } cakes) (A_{3}) 1,100 " { Slags (B_{2}) } Slags (B_{3}) { Concentrates from } { accretions }

{ Tertiary matte (3) 11 cartloads } Quaternary hard cakes { Poor hard cakes (A_{5}) 3 " } matte (4) 2,000 lbs. 4th { Slags (B_{3}) } Rich hard cakes of { Concentrates from } matte (A_{4}) 1,500 " { accretions }

{ Roasted quartz } Poor hard cakes of 5th { Matte (4) (three } matte (A_{5}) 1,500 lbs. { times roasted) 11 cartloads } Final cakes of matte (5)

6th Final matte three times roasted is smelted to blister copper.

The following would be a rough approximation of the value of the various products:--

(1.) Primary matte = 158 ounces troy per short ton. (2.) Secondary matte = 85 " " " (3.) Tertiary matte = 60 " " " (4.) Quaternary matte = Indeterminate. A. Copper-silver alloy = 388 ounces Troy per short ton. A_{2} Copper-silver-lead alloy = 145 " " " A_{3} " " " = 109 " " " A_{4} Rich hard cakes = 97 " " " A_{5} Poor hard cakes = Indeterminate. Final blister copper = 12 ozs. Troy per short ton.

[47] This expression is usually used for hearth-lead, but in this case the author is apparently confining himself to lead ore, and apparently refers to lead carbonates. The German Translation gives _pleyschweiss_. The pyrites mentioned in this paragraph may mean galena, as pyrites was to Agricola a sort of genera.

[48] (_Excoquitur_) ... "_si verò pyrites, primò è fornace, ut Goselariae videre licet, in catinum defluit liquor quidam candidus, argento inimicus et nocivus; id enim comburit: quo circa recrementis, quae supernatant, detractis effunditur: vel induratus conto uncinato extrahitur: eundem liquorem parietes fornacis exudant._" In the Glossary the following statement appears: "_Liquor candidus primo è fornace defluens cum Goselariae excoquitur pyrites,--kobelt; quem parietes fornacis exudant,--conterfei._" In this latter statement Agricola apparently recognised that there were two different substances, _i.e._, that the substance found in the furnace walls--_conterfei_--was not the same substance as that which first flowed from the furnace--_kobelt_. We are at no difficulty in recognizing _conterfei_ as metallic zinc; it was long known by that term, and this accidental occurrence is repeatedly mentioned by other authors after Agricola. The substance which first flowed into the forehearth presents greater difficulties; it certainly was not zinc. In _De Natura Fossilium_ (p. 347), Agricola says that at Goslar the lead has a certain white slag floating upon it, the "colour derived from the pyrites (_pyriten argenteum_) from which it was produced." _Pyriten argenteum_ was either marcasite or mispickel, neither of which offers much suggestion; nor are we able to hazard an explanation of value.

HISTORICAL NOTE ON ZINC. The history of zinc metallurgy falls into two distinct lines--first, that of the metal, and second, that of zinc ore, for the latter was known and used to make brass by cementation with copper and to yield oxides by sublimation for medicinal purposes, nearly 2,000 years before the metal became generally known and used in Europe.

There is some reason to believe that metallic zinc was known to the Ancients, for bracelets made of it, found in the ruins of Cameros (prior to 500 B.C.), may have been of that age (Raoul Jagnaux, _Traité de Chimie Générale_, 1887, II, 385); and further, a passage in Strabo (63 B.C.-24 A.D.) is of much interest. He states: (XIII, 1, 56) "There is found at Andeira a stone which when burnt becomes iron. It is then put into a furnace, together with some kind of earth, when it distils a mock silver (_pseudargyrum_), or with the addition of copper it becomes the compound called _orichalcum_. There is found a mock silver near Tismolu also." (Hamilton's Trans., II, p. 381). About the Christian era the terms _orichalcum_ or _aurichalcum_ undoubtedly refer to brass, but whether these terms as used by earlier Greek writers do not refer to bronze only, is a matter of considerable doubt. Beyond these slight references we are without information until the 16th Century. If the metal was known to the Ancients it must have been locally, for by its greater adaptability to brass-making it would probably have supplanted the crude melting of copper with zinc minerals.

It appears that the metal may have been known in the Far East prior to such knowledge in Europe; metallic zinc was imported in considerable quantities from the East as early as the 16th and 17th centuries under such terms as _tuteneque_, _tuttanego_, _calaëm_, and _spiauter_--the latter, of course, being the progenitor of our term spelter. The localities of Eastern production have never been adequately investigated. W. Hommel (Engineering and Mining Journal, June 15, 1912) gives a very satisfactory review of the Eastern literature upon the subject, and considers that the origin of manufacture was in India, although the most of the 16th and 17th Century product came from China. The earliest certain description seems to be some recipes for manufacture quoted by Praphulla Chandra Ray (A History of Hindu Chemistry, London, 1902, p. 39) dating from the 11th to the 14th Centuries. There does not appear to be any satisfactory description of the Chinese method until that of Sir George Staunton (Journal Asiatique Paris, 1835, p. 141.) We may add that spelter was produced in India by crude distillation of calamine in clay pots in the early part of the 19th Century (Brooke, Jour. Asiatic Soc. of Bengal, vol. XIX, 1850, p. 212), and the remains of such smelting in Rajputana are supposed to be very ancient.

The discovery of zinc in Europe seems to have been quite independent of the East, but precisely where and when is clouded with much uncertainty. The _marchasita aurea_ of Albertus Magnus has been called upon to serve as metallic zinc, but such belief requires a hypothesis based upon a great deal of assumption. Further, the statement is frequently made that zinc is mentioned in Basil Valentine's Triumphant Chariot of Antimony (the only one of the works attributed to this author which may date prior to the 17th Century), but we have been unable to find any such reference. The first certain mention of metallic zinc is generally accredited to Paracelsus (1493-1541), who states (_Liber Mineralium_ II.): "Moreover there is another metal generally unknown called _zinken_. It is of peculiar nature and origin; many other metals adulterate it. It can be melted, for it is generated from three fluid principles; it is not malleable. Its colour is different from other metals and does not resemble others in its growth. Its ultimate matter (_ultima materia_) is not to me yet fully known. It admits of no mixture and does not permit of the _fabricationes_ of other metals. It stands alone entirely to itself." We do not believe that this book was published until after Agricola's works. Agricola introduced the following statements into his revised edition of _Bermannus_ (p. 431), published in 1558: "It (a variety of pyrites) is almost the colour of galena, but of entirely different components. From it there is made gold and silver, and a great quantity is dug in Reichenstein, which is in Silesia, as was recently reported to me. Much more is found at Raurici, which they call _zincum_, which species differs from pyrites, for the latter contains more silver than gold, the former only gold or hardly any silver." In _De Natura Fossilium_ (p. 368): "For this _cadmia_ is put, in the same way as quicksilver, in a suitable vessel so that the heat of the fire will cause it to sublime, and from it is made a black or brown or grey body which the Alchemists call _cadmia sublimata_. This possesses corrosive properties to the highest degree. Cognate with this _cadmia_ and pyrites is a compound which the Noricans and Rhetians call _zincum_." We leave it to readers to decide how near this comes to metallic zinc; in any event, he apparently did not recognise his _conterfei_ from the furnaces as the same substance as the _zincum_ from Silesia. The first correlation of these substances was apparently by Lohneys, in 1617, who says (_Vom Bergwerk_, p. 83-4): "When the people in the smelting works are smelting, there is made under the furnace and in the cracks in the walls among the badly plastered stones, a metal which is called _zinc_ or _counterfeht_, and when the wall is scraped it falls into a vessel placed to receive it. This metal greatly resembles tin, but it is harder and less malleable.... The Alchemists have a great desire for this _zinc_ or bismuth." That this metal originated from blende or calamine was not recognised until long after, and Libavis (_Alchymia_, Frankfort, 1606), in describing specimens which came from the East, did not so identify it, this office being performed by Glauber, who says (_De Prosperitate Germanias_, Amsterdam, 1656): "Zink is a volatile mineral or half-ripe metal when it is extracted from its ore. It is more brilliant than tin and not so fusible or malleable ... it turns (copper) into brass, as does _lapis calaminaris_, for indeed this stone is nothing but infusible zinc, and this zinc might be called a fusible _lapis calaminaris_, inasmuch as both of them partake of the same nature.... It sublimates itself up into the cracks of the furnace, whereupon the smelters frequently break it out." The systematic distillation of zinc from calamine was not discovered in Europe until the 18th Century. Henkel is generally accredited with the first statement to that effect. In a contribution published as an Appendix to his other works, of which we have had access only to a French translation (_Pyritologie_, Paris, 1760, p. 494), he concludes that zinc is a half-metal of which the best ore is calamine, but believes it is always associated with lead, and mentions that an Englishman lately arrived from Bristol had seen it being obtained from calamine in his own country. He further mentions that it can be obtained by heating calamine and lead ore mixed with coal in a thick earthen vessel. The Bristol works were apparently those of John Champion, established about 1740. The art of distillation was probably learned in the East.

Definite information as to the zinc minerals goes back to but a little before the Christian Era, unless we accept nebular references to _aurichalcum_ by the poets, or what is possibly zinc ore in the "earth" mentioned by Aristotle (_De Mirabilibus_, 62): "Men say that the copper of the Mossynoeci is very brilliant and white, no tin being mixed with it; but there is a kind of earth there which is melted with it." This might quite well be an arsenical mineral. But whether we can accept the poets or Aristotle or the remark of Strabo given above, as sufficient evidence or not, there is no difficulty with the description of _cadmia_ and _pompholyx_ and _spodos_ of Dioscorides (1st Century), parts of which we reproduce in note 26, p. 394. His _cadmia_ is described as rising from the copper furnaces and clinging to the iron bars, but he continues: "_Cadmia_ is also prepared by burning the stone called pyrites, which is found near Mt. Soloi in Cyprus.... Some say that _cadmia_ may also be found in stone quarries, but they are deceived by stones having a resemblance to _cadmia_." _Pompholyx_ and _spodos_ are evidently furnace calamine. From reading the quotation given on p. 394, there can be no doubt that these materials, natural or artificial, were used to make brass, for he states (V, 46): "White _pompholyx_ is made every time that the artificer in the working and perfecting of the copper sprinkles powdered _cadmia_ upon it to make it more perfect, the soot arising from this ... is _pompholyx_." Pliny is confused between the mineral _cadmia_ and furnace _calamine_, and none of his statements are very direct on the subject of brass making. His most pointed statement is (XXXIV, 2): "... Next to Livian (copper) this kind best absorbs _cadmia_, and is almost as good as _aurichalcum_ for making sesterces and double asses." As stated above, there can be little doubt that the _aurichalcum_ of the Christian Era was brass, and further, we do know of brass sesterces of this period. Other Roman writers of this and later periods refer to earth used with copper for making brass. Apart from these evidences, however, there is the evidence of analyses of coins and objects, the earliest of which appears to be a large brass of the Cassia family of 20 B.C., analyzed by Phillips, who found 17.3% zinc (Records of Mining and Metallurgy, London, 1857, p. 13). Numerous analyses of coins and other objects dating during the following century corroborate the general use of brass. Professor Gowland (Presidential Address, Inst. of Metals, 1912) rightly considers the Romans were the first to make brass, and at about the above period, for there appears to be no certainty of any earlier production. The first adequate technical description of brass making is in about 1200 A.D. being that of Theophilus, who describes (Hendrie's Trans., p. 307) calcining _calamina_ and mixing it with finely divided copper in glowing crucibles. The process was repeated by adding more calamine and copper until the pots were full of molten metal. This method is repeatedly described with minor variations by Biringuccio, Agricola (_De Nat. Fos._), and others, down to the 18th Century. For discussion of the zinc minerals see note on p. 112.

[49] "_... non raro, ut nonnulli pyritae sunt, candida...._" This is apparently the unknown substance mentioned above.

[50] One _drachma_ is about 3 ounces Troy per short ton. Three _unciae_ are about 72 ounces 6 dwts. Troy per short ton.

[51] In this section, which treats of the metallurgy of _plumbum candidum_, "tin," the word _candidum_ is very often omitted in the Latin, leaving only _plumbum_, which is confusing at times with lead. The black tin-stone, _lapilli nigri_ has been treated in a similar manner, _lapilli_ (small stones) constantly occurring alone in the Latin. This has been rendered as "tin-stone" throughout, and the material prior to extraction of the _lapilli nigri_ has been rendered "tin-stuff," after the Cornish.

[52] "_... ex saxis vilibus, quae natura de diversa materia composuit._" The Glossary gives _grindstein_. Granite (?).

[53] HISTORICAL NOTES ON TIN METALLURGY. The first appearance of tin lies in the ancient bronzes. And while much is written upon the "Bronze Age" by archæologists, we seriously doubt whether or not a large part of so-called bronze is not copper. In any event, this period varied with each race, and for instance, in Britain may have been much later than Egyptian historic times. The bronze articles of the IV Dynasty (from 3800 to 4700 B.C. depending on the authority) place us on certain ground of antiquity. Professor Gowland (Presidential Address, Inst. of Metals, London, 1912) maintains that the early bronzes were the result of direct smelting of stanniferous copper ores, and while this may be partially true for Western Europe, the distribution and nature of the copper deposits do not warrant this assumption for the earlier scenes of human activity--Asia Minor, Egypt, and India. Further, the lumps of rough tin and also of copper found by Borlase (Tin Mining in Spain, Past and Present, London, 1897, p. 25) in Cornwall, mixed with bronze celts under conditions certainly indicating the Bronze Age, is in itself of considerable evidence of independent melting. To our mind the vast majority of ancient bronzes must have been made from copper and tin mined and smelted independently. As to the source of supply of ancient tin, we are on clear ground only with the advent of the Phoenicians, 1500-1000 B.C., who, as is well known, distributed to the ancient world a supply from Spain and Britain. What the source may have been prior to this time has been subject to much discussion, and while some slender threads indicate the East, we believe that a more local supply to Egypt, etc., is not impossible. The discovery of large tin fields in Central Africa and the native-made tin ornaments in circulation among the negroes, made possible the entrance of the metal into Egypt along the trade routes. Further, we see no reason why alluvial tin may not have existed within easy reach and have become exhausted. How quickly such a source of metal supply can be forgotten and no evidence remain, is indicated by the seldom remembered alluvial gold supply from Ireland. However, be these conjectures as they may, the East has long been the scene of tin production and of transportation activity. Among the slender evidences that point in this direction is that the Sanskrit term for tin is _kastira_, a term also employed by the Chaldeans, and represented in Arabic by _kasdir_, and it may have been the progenitor of the Greek _cassiteros_. There can be no doubt that the Phoenicians also traded with Malacca, etc., but beyond these threads there is little to prove the pre-western source. The strained argument of Beckmann (Hist. of Inventions, vol. II., p. 207) that the _cassiteros_ of Homer and the _bedil_ of the Hebrews was possibly not tin, and that tin was unknown at this time, falls to the ground in the face of the vast amount of tin which must have been in circulation to account for the bronze used over a period 2,000 years prior to those peoples. Tin is early mentioned in the Scriptures (Numbers XXXI, 22), being enumerated among the spoil of the Midianites (1200 B.C.?), also Ezekiel (600 B.C., XXVII, 12) speaks of tin from Tarshish (the Phoenician settlement on the coast of Spain). According to Homer tin played considerable part in Vulcan's metallurgical stores. Even approximately at what period the Phoenicians began their distribution from Spain and Britain cannot be determined. They apparently established their settlements at Gades (Cadiz) in Tarshish, beyond Gibraltar, about 1100 B.C. The remains of tin mining in the Spanish peninsula prior to the Christian Era indicate most extensive production by the Phoenicians, but there is little evidence as to either mining or smelting methods. Generally as to the technical methods of mining and smelting tin, we are practically without any satisfactory statement down to Agricola. However, such scraps of information as are available are those in Homer (see note on p. 402), Diodorus, and Pliny.

Diodorus says (V, 2) regarding tin in Spain: "They dig it up, and melt it down in the same way as they do gold and silver;" and again, speaking of the tin in Britain, he says: "These people make tin, which they dig up with a great deal of care and labour; being rocky, the metal is mixed with earth, out of which they melt the metal, and then refine it." Pliny (XXXIV, 47), in the well-known and much-disputed passage: "Next to be considered are the characteristics of lead, which is of two kinds, black and white. The most valuable is the white; the Greeks called it _cassiteros_, and there is a fabulous story of its being searched for and carried from the islands of Atlantis in barks covered with hides. Certainly it is obtained in Lusitania and Gallaecia on the surface of the earth from black-coloured sand. It is discovered by its great weight, and it is mixed with small pebbles in the dried beds of torrents. The miners wash these sands, and that which settles they heat in the furnace. It is also found in gold mines, which are called _alutiae_. A stream of water passing through detaches small black pebbles variegated with white spots, the weight of which is the same as gold. Hence it is that they remain in the baskets of the gold collectors with the gold; afterward, they are separated in a _camillum_ and when melted become white lead."

There is practically no reference to the methods of Cornish tin-working over the whole period of 2,000 years that mining operations were carried on there prior to the Norman occupation. From then until Agricola's time, a period of some four centuries, there are occasional references in Stannary Court proceedings, Charters, and such-like official documents which give little metallurgical insight. From a letter of William de Wrotham, Lord Warden of the Stannaries, in 1198, setting out the regulations for the impost on tin, it is evident that the black tin was smelted once at the mines and that a second smelting or refining was carried out in specified towns under the observation of the Crown Officials. In many other official documents there are repeated references to the right to dig turfs and cut wood for smelting the tin. Under note 8, p. 282, we give some further information on tin concentration, and the relation of Cornish and German tin miners. Biringuccio (1540) gives very little information on tin metallurgy, and we are brought to _De Re Metallica_ for the first clear exposition.

As to the description on these pages it must be remembered that the tin-stone has been already roasted, thus removing some volatile impurities and oxidizing others, as described on page 348. The furnaces and the methods of working the tin, here described, are almost identical with those in use in Saxony to-day. In general, since Agricola's time tin has not seen the mechanical and metallurgical development of the other metals. The comparatively small quantities to be dealt with; the necessity of maintaining a strong reducing atmosphere, and consequently a mild cold blast; and the comparatively low temperature demanded, gave little impetus to other than crude appliances until very modern times.

[54] _Aureo nummo_. German Translation gives _reinschen gülden_, which was the equivalent of about $1.66, or 6.9 shillings. The purchasing power of money was, however, several times as great as at present.

[55] In the following descriptions of iron-smelting, we have three processes described; the first being the direct reduction of malleable iron from ore, the second the transition stage then in progress from the direct to indirect method by way of cast-iron; and the third a method of making steel by cementation. The first method is that of primitive iron-workers of all times and all races, and requires little comment. A pasty mass was produced, which was subsequently hammered to make it exude the slag, the hammered mass being the ancient "bloom." The second process is of considerable interest, for it marks one of the earliest descriptions of working iron in "a furnace similar to a blast furnace, but much wider and higher." This original German _Stückofen_ or high bloomery furnace was used for making "masses" of wrought-iron under essentially the same conditions as its progenitor the forge--only upon a larger scale. With high temperatures, however, such a furnace would, if desired, yield molten metal, and thus the step to cast-iron as a preliminary to wrought-iron became very easy and natural, in fact Agricola mentions above that if the iron is left to settle in the furnace it becomes hard. The making of malleable iron by subsequent treatment of the cast-iron--the indirect method--originated in about Agricola's time, and marks the beginning of one of those subtle economic currents destined to have the widest bearing upon civilization. It is to us uncertain whether he really understood the double treatment or not. In the above paragraph he says from ore "once or twice smelted they make iron," etc., and in _De Natura Fossilium_ (p. 339) some reference is made to pouring melted iron, all of which would appear to be cast-iron. He does not, however, describe the 16th Century method of converting cast into wrought iron by way of in effect roasting the pig iron to eliminate carbon by oxidation, with subsequent melting into a "ball" or "mass." It must be borne in mind that puddling for this purpose did not come into use until the end of the 18th Century. A great deal of discussion has arisen as to where and at what time cast-iron was made systematically, but without satisfactory answer; in any event, it seems to have been in about the end of the 14th Century, as cast cannon began to appear about that time. It is our impression that the whole of this discussion on iron in _De Re Metallica_ is an abstract from Biringuccio, who wrote 15 years earlier, as it is in so nearly identical terms. Those interested will find a translation of Biringuccio's statement with regard to steel in Percy's Metallurgy of Iron and Steel, London, 1864, p. 807.

HISTORICAL NOTE ON IRON SMELTING. The archæologists' division of the history of racial development into the Stone, Bronze, and Iron Ages, based upon objects found in tumuli, burial places, etc., would on the face of it indicate the prior discovery of copper metallurgy over iron, and it is generally so maintained by those scientists. The metallurgists have not hesitated to protest that while this distinction of "Ages" may serve the archæologists, and no doubt represents the sequence in which the metal objects are found, yet it by no means follows that this was the order of their discovery or use, but that iron by its rapidity of oxidation has simply not been preserved. The arguments which may be advanced from our side are in the main these. Iron ore is of more frequent occurrence than copper ores, and the necessary reduction of copper oxides (as most surface ores must have been) to fluid metal requires a temperature very much higher than does the reduction of iron oxides to wrought-iron blooms, which do not necessitate fusion. The comparatively greater simplicity of iron metallurgy under primitive conditions is well exemplified by the hill tribes of Northern Nigeria, where in village forges the negroes reduce iron sufficient for their needs, from hematite. Copper alone would not be a very serviceable metal to primitive man, and he early made the advance to bronze; this latter metal requires three metallurgical operations, and presents immeasurably greater difficulties than iron. It is, as Professor Gowland has demonstrated (Presidential Address, Inst. of Metals, London, 1912) quite possible to make bronze from melting stanniferous copper ores, yet such combined occurrence at the surface is rare, and, so far as known, the copper sources from which Asia Minor and Egypt obtained their supply do not contain tin. It seems to us, therefore, that in most cases the separate fusions of different ores and their subsequent re-melting were required to make bronze. The arguments advanced by the archæologists bear mostly upon the fact that, had iron been known, its superiority would have caused the primitive races to adopt it, and we should not find such an abundance of bronze tools. As to this, it may be said that bronze weapons and tools are plentiful enough in Egyptian, Mycenæan, and early Greek remains, long after iron was demonstrably well known. There has been a good deal pronounced by etymologists on the history of iron and copper, for instance, by Max Müller, (Lectures on the Science of Language, Vol. II, p. 255, London, 1864), and many others, but the amazing lack of metallurgical knowledge nullifies practically all their conclusions. The oldest Egyptian texts extant, dating 3500 B.C., refer to iron, and there is in the British Museum a piece of iron found in the Pyramid of Kephron (3700 B.C.) under conditions indicating its co-incident origin. There is exhibited also a fragment of oxidized iron lately found by Professor Petrie and placed as of the VI Dynasty (B.C. 3200). Despite this evidence of an early knowledge of iron, there is almost a total absence of Egyptian iron objects for a long period subsequent to that time, which in a measure confirms the view of its disappearance rather than that of ignorance of it. Many writers have assumed that the Ancients must have had some superior art of hardening copper or bronze, because the cutting of the gigantic stonework of the time could not have been done with that alloy as we know it; no such hardening appears among the bronze tools found, and it seems to us that the argument is stronger that the oldest Egyptian stoneworkers employed mostly iron tools, and that these have oxidized out of existence. The reasons for preferring copper alloys to iron for decorative objects were equally strong in ancient times as in the present day, and accounts sufficiently for these articles, and, therefore, iron would be devoted to more humble objects less likely to be preserved. Further, the Egyptians at a later date had some prejudices against iron for sacred purposes, and the media of preservation of most metal objects were not open to iron. We know practically nothing of very early Egyptian metallurgy, but in the time of Thotmes III. (1500 B.C.) bellows were used upon the forge.

Of literary evidences the earliest is in the Shoo King among the Tribute of Yü (2500 B.C.?). Iron is frequently mentioned in the Bible, but it is doubtful if any of the early references apply to steel. There is scarcely a Greek or Latin author who does not mention iron in some connection, and of the earliest, none are so suggestive from a metallurgical point of view as Homer, by whom "laboured" mass (wrought-iron?) is often referred to. As, for instance, in the Odyssey (I., 234) Pallas in the guise of Mentes, says according to Pope:

"Freighted with iron from my native land I steer my voyage to the Brutian strand, To gain by commerce for the laboured mass A just proportion of refulgent brass."

(Brass is modern poetic licence for copper or bronze). Also, in the Odyssey (IX, 465) when Homer describes how Ulysses plunged the stake into Cyclop's eye, we have the first positive evidence of steel, although hard iron mentioned in the Tribute of Yü, above referred to, is sometimes given as steel:

"And as when armourers temper in the ford The keen-edg'd pole-axe, or the shining sword, The red-hot metal hisses in the lake."

No doubt early wrought-iron was made in the same manner as Agricola describes. We are, however, not so clear as to the methods of making steel. Under primitive methods of making wrought-iron it is quite possible to carburize the iron sufficiently to make steel direct from ore. The primitive method of India and Japan was to enclose lumps of wrought-iron in sealed crucibles with charcoal and sawdust, and heat them over a long period. Neither Pliny nor any of the other authors of the period previous to the Christian Era give us much help on steel metallurgy, although certain obscure expressions of Aristotle have been called upon (for instance, St. John V. Day, Prehistoric Use of Iron and Steel, London, 1877, p. 134) to prove its manufacture by immersing wrought-iron in molten cast-iron.

[56] _Quae vel aerosa est, vel cocta_. It is by no means certain that _cocta_, "cooked" is rightly translated, for the author has not hitherto used this expression for heated. This may be residues from roasting and leaching pyrites for vitriol, etc.

[57] Agricola draws no sharp line of distinction between antimony the metal, and its sulphide. He uses the Roman term _stibi_ or _stibium_ (_Interpretatio_,--_Spiesglas_) throughout this book, and evidently in most cases means the sulphide, but in others, particularly in parting gold and silver, metallic antimony would be reduced out. We have been in much doubt as to the term to introduce into the text, as the English "stibnite" carries too much precision of meaning. Originally the "antimony" of trade was the sulphide. Later, with the application of that term to the metal, the sulphide was termed "grey antimony," and we have either used _stibium_ for lack of better alternative, or adopted "grey antimony." The method described by Agricola for treating antimony sulphide is still used in the Harz, in Bohemia, and elsewhere. The stibnite is liquated out at a low heat and drips from the upper to the lower pot. The resulting purified antimony sulphide is the modern commercial "crude antimony" or "grey antimony."

HISTORICAL NOTE ON THE METALLURGY OF ANTIMONY. The Egyptologists have adopted the term "antimony" for certain cosmetics found in Egyptian tombs from a very early period. We have, however, failed to find any reliable analyses which warrant this assumption, and we believe that it is based on the knowledge that antimony was used as a base for eye ointments in Greek and Roman times, and not upon proper chemical investigation. It may be that the ideograph which is interpreted as antimony may really mean that substance, but we only protest that the chemist should have been called in long since. In St. Jerome's translation of the Bible, the cosmetic used by Jezebel (II. Kings IX, 30) and by the lady mentioned by Ezekiel (XXIII, 40), "who didst wash thyself and paintedst thine eyes" is specifically given as _stibio_. Our modern translation carries no hint of the composition of the cosmetic, and whether some of the Greek or Hebrew MSS. do furnish a basis for such translation we cannot say. The Hebrew term for this mineral was _kohl_, which subsequently passed into "alcool" and "alkohol" in other languages, and appears in the Spanish Bible in the above passage in Ezekiel as _alcoholaste_. The term _antimonium_ seems to have been first used in Latin editions of Geber published in the latter part of the 15th Century. In any event, the metal is clearly mentioned by Dioscorides (1st Century), who calls it _stimmi_, and Pliny, who termed it _stibium_, and they leave no doubt that it was used as a cosmetic for painting the eyebrows and dilating the eyes. Dioscorides (V, 59) says: "The best _stimmi_ is very brilliant and radiant. When broken it divides into layers with no part earthy or dirty; it is brittle. Some call it _stimmi_, others _platyophthalmon_ (wide eyed); others _larbason_, others _gynaekeion_ (feminine).... It is roasted in a ball of dough with charcoal until it becomes a cinder.... It is also roasted by putting it on live charcoal and blowing it. If it is roasted too much it becomes lead." Pliny states (XXXIII, 33 and 34): "In the same mines in which silver is found, properly speaking there is a stone froth. It is white and shining, not transparent; is called _stimmi_, or _stibi_, or _alabastrum_, and _larbasis_. There are two kinds of it, the male and the female. The most approved is the female, the male being more uneven, rougher, less heavy, not so radiant, and more gritty. The female kind is bright and friable, laminar and not globular. It is astringent and refrigerative, and its principal use is for the eyes.... It is burned in manure in a furnace, is quenched with milk, ground with rain water in a mortar, and while thus turbid it is poured into a copper vessel and purified with nitrum ... above all in roasting it care should be taken that it does not turn to lead." There can be little doubt from Dioscorides' statement of its turning to lead that he had seen the metal antimony, although he thought it a species of lead. Of further interest in connection with the ancient knowledge of the metal is the Chaldean vase made of antimony described by Berthelot (_Comptes Rendus_, 1887, CIV, 265). It is possible that Agricola knew the metal, although he gives no details as to de-sulphurizing it or for recovering the metal itself. In _De Natura Fossilium_ (p. 181) he makes a statement which would indicate the metal, "_Stibium_ when melted in the crucible and refined has as much right to be regarded as a metal as is accorded to lead by most writers. If when smelted a certain portion be added to tin, a printer's alloy is made from which type is cast that is used by those who print books." Basil Valentine, in his "Triumphal Chariot of Antimony," gives a great deal that is new with regard to this metal, even if we can accredit the work with no earlier origin than its publication--about 1600; it seems possible however, that it was written late in the 15th Century (see Appendix B). He describes the preparation of the metal from the crude ore, both by roasting and reduction from the oxide with argol and saltpetre, and also by fusing with metallic iron. While the first description of these methods is usually attributed to Valentine, it may be pointed out that in the _Probierbüchlein_ (1500) as well as in Agricola the separation of silver from iron by antimony sulphide implies the same reaction, and the separation of silver and gold with antimony sulphide, often attributed to Valentine, is repeatedly set out in the _Probierbüchlein_ and in _De Re Metallica_. Biringuccio (1540) has nothing of importance to say as to the treatment of antimonial ores, but mentions it as an alloy for bell-metal, which would imply the metal.

[58] HISTORICAL NOTE ON THE METALLURGY OF QUICKSILVER. The earliest mention of quicksilver appears to have been by Aristotle (_Meteorologica_ IV, 8, 11), who speaks of it as fluid silver (_argyros chytos_). Theophrastus (105) states: "Such is the production of quicksilver, which has its uses. This is obtained from cinnabar rubbed with vinegar in a brass mortar with a brass pestle." (Hill's Trans., p. 139). Theophrastus also (103) mentions cinnabar from Spain and elsewhere. Dioscorides (V, 70) appears to be the first to describe the recovery of quicksilver by distillation: "Quicksilver (_hydrargyros_, _i.e._, liquid silver) is made from _ammion_, which is called _cinnabari_. An iron bowl containing _cinnabari_ is put into an earthen vessel and covered over with a cup-shaped lid smeared with clay. Then it is set on a fire of coals and the soot which sticks to the cover when wiped off and cooled is quicksilver. Quicksilver is also found in drops falling from the walls of the silver mines. Some say there are quicksilver mines. It can be kept only in vessels of glass, lead, tin (?), or silver, for if put in vessels of any other substances it consumes them and flows through." Pliny (XXXIII, 41): "There has been discovered a way of extracting _hydrargyros_ from the inferior _minium_ as a substitute for quicksilver, as mentioned. There are two methods: either by pounding _minium_ and vinegar in a brass mortar with a brass pestle, or else by putting _minium_ into a flat earthen dish covered with a lid, well luted with potter's clay. This is set in an iron pan and a fire is then lighted under the pan, and continually blown by a bellows. The perspiration collects on the lid and is wiped off and is like silver in colour and as liquid as water." Pliny is somewhat confused over the _minium_--or the text is corrupt, for this should be the genuine _minium_ of Roman times. The methods of condensation on the leaves of branches placed in a chamber, of condensing in ashes placed over the mouth of the lower pot, and of distilling in a retort, are referred to by Biringuccio (A.D. 1540), but with no detail.

[59] Most of these methods depend upon simple liquation of native bismuth. The sulphides, oxides, etc., could not be obtained without fusing in a furnace with appropriate de-sulphurizing or reducing agents, to which Agricola dimly refers. In _Bermannus_ (p. 439), he says: "_Bermannus_.--I will show you another kind of mineral which is numbered amongst metals, but appears to me to have been unknown to the Ancients; we call it _bisemutum_. _Naevius_.--Then in your opinion there are more kinds of metals than the seven commonly believed? _Bermannus_.--More, I consider; for this which just now I said we called _bisemutum_, cannot correctly be called _plumbum candidum_ (tin) nor _nigrum_ (lead), but is different from both, and is a third one. _Plumbum candidum_ is whiter and _plumbum nigrum_ is darker, as you see. _Naevius_.--We see that this is of the colour of _galena_. _Ancon_.--How then can _bisemutum_, as you call it, be distinguished from _galena_? _Bermannus_.--Easily; when you take it in your hands it stains them with black unless it is quite hard. The hard kind is not friable like _galena_, but can be cut. It is blacker than the kind of crude silver which we say is almost the colour of lead, and thus is different from both. Indeed, it not rarely contains some silver. It generally shows that there is silver beneath the place where it is found, and because of this our miners are accustomed to call it the 'roof of silver.' They are wont to roast this mineral, and from the better part they make metal; from the poorer part they make a pigment of a kind not to be despised." This pigment was cobalt blue (see note on p. 112), indicating a considerable confusion of these minerals. This quotation is the first description of bismuth, and the above text the first description of bismuth treatment. There is, however, bare mention of the mineral earlier, in the following single line from the _Probierbüchlein_ (p. 1): "Jupiter (controls) the ores of tin and _wismundt_." And it is noted in the _Nützliche Bergbüchlein_ in association with silver (see Appendix B).

[60] This _cadmia_ is given in the German translation as _kobelt_. It is probably the cobalt-arsenic-bismuth minerals common in Saxony. A large portion of the world's supply of bismuth to-day comes from the cobalt treatment works near Schneeberg. For further discussion of _cadmia_ see note on p. 112.