The fuel for the brass furnace is always hard coke, which is prepared in ovens and broken into lumps about the size of hens' eggs: in lighting the fire, a bundle of shavings, chips of cork, or any similar combustible, is first thrown in and ignited, and then some coke or charcoal is added. It is also usual to put the pot in the fire at an early stage, and with its mouth downwards; by this means the thin edge which admits the most easily of expansion gets hot first, and the heat plays within the crucible, so as to warm it gradually, it is not reversed until the whole is red hot; putting it in bottom downwards would be almost certain to cause it to crack.

The pot is now bedded upon the fuel, and the brass-founder, whilst making up the fire, puts an iron cover with a long central handle over the mouth of the pot, to prevent the small cokes which are now thrown on from entering the same. Next, the chargeof metal is put in the crucible, and three or four large picces of coke are placed across the mouth of the pot; the tile is put on the furnace, the damper is thru adjusted to heat the crucible quickly, and the whole is left to itself until the metal is run down.

The gold and silver refiners and jewellers manage their furnaces much in the same way, except that they support the crucible upon a hollow earthen stand placed on the fire-bars to catch any leakage, and also put an earthen cover over its mouth. They generally use coke, although charcoal is a purer fuel, and is laid upon the fluid metals to prevent oxidation.*

The generality of the metals are far more disposed to oxidation when in the melted condition than when solid, it is therefore usual, whilst they are in the crucible, to protect their surfaces the air with some flux, to lessen their disposition to oxidize.

In the iron furnace, the slag from the lime floats on the metal and fulfill this end; many brass-founders always throw broken glass, charcoal dust, sandiver, or sal-enixon, into the melting-pot, by others these precautionary measures are altogether neglected. The black and white fluxes, borax and saltpetre, are also used for the precious metals, and oil or resin for the more fusible, as lead or tin, but excess of heat should be at all times avoided.

The generality of the fusible metals may be mixed in all proportions. Those in which the melting points are tolerably similar may be easily combined, such as lead with tin, or gold with silver, or copper, these appear to call for no instructions beyond moderation in the heat employed, but the difficulty of making definite and uniform alloys increases, when the melting points of the metals, or their qualities or quantities, are widely dissimilar.

* Descriptions of the methods of constructing the white melting pots used for brass, iron and steel, will be found in the Transactions of the Society of Arts: namely, Mr. Marshall's method, vol. 41, p. 52; Mr. Anstey's, vol. 43, p. 32, and Mr. Charles Sidney Smith's, vol. 47, p. 54. The above, and the so called blue pots, or black-lead pots, are not burned until they are put into the fire for use; but the Hessian pots, the English brown or clay pots, the Cornish and the Wedgwood crucibles, are all burned before use. The account of some comparative trials on all of these kinds will be found attached to Mr. Anstey's paper.

It may bo further observed, that the pots for brass are too porous for gold and silver, as they suck up too much of the same: the black-lead pots are closer and better for the precious metals, and they withstand change of temperature beat of any kind; they are however the most expensive, but cannot be safely used with fluxes. The Hessian crucibles resist the fluxes, and serve with care for several consecutive meltings; the English clay pots, which resemble the Hessian, are safe for one or sometimes for more meltings, and their cost is trifling. The pots for gold and silver are occasionally coated or hated externally with clay as a protection.

In mixing alloys with new metals, it is usual to melt the less fusible first, and subsequently to add the more fusible; the mixture is then stirred well together, and common opinion seems to be in favour of running the metal into an ingot mould, as the second fusion is considered more thoroughly to incorporate the mixture. Sometimes, with the same view, the alloy is granulated, by pouring it from the crucible into water, either from a considerable height through a colander, or over a bundle of birch twigs, which subdivide it into small pieces; others condemn such practices, and greatly prefer the first fusion, in order to avoid oxidation, and departure from the intended proportions.

But in many, and perhaps in most cases, it is the practice to fill the melting pan, or the crucible, in part with old alloy, consisting of fragments of spoiled or worn-out work; and to which is added, partly by calculation but principally by trial, a certain quantity of new metals. This is not always done from motives of economy alone, but from the opinion that such mixtures cast and work better than those made entirely of new metals.

When small quantities of a metal of difficult fusion, are added to large proportions of others which are much more fusible, the whole quantities are not mixed at once. Thus, in pewter, it would be scarcely possible to throw into the melted tin the half per cent, or the one per cent. of melted copper with any certainty of the two combining properly, and it is therefore usual to melt the copper in a crucible, and to add to it two or three times its weight of melted tin: this as it were, dilutes the copper, and makes the alloy known as temper, which may be fused in a ladle, and added in small quantities to the fluid pewter or to the tin, as the case may be, until on trying the mixture by the assay (see page 284) its proportions are considered suitable.

The metal for printer's type is often mixed nearly in the same manner; the copper is first melted alone in a crucible, the antimony is melted in another crucible, and is poured into the copper, sometimes a little lead is also added. The hard alloy and the tin are then introduced to the mass of type-metal or lead, also in great measure by trial, as old metal mostly enters into the mixture.