with the vapour of the salt, is converted into protochloride of iron, sulphuret of sodium being simultaneously formed. The protochloride is then said to be decomposed into subperchloride with the evolution of chlorine, which, acting on the sulphuret of sodium, produces again chloride of sodium and chloride of sulphur; which latter is disengaged, leaving a much less quantity of sulphur in the residual coke, and that being in the harmless form of sulphuret of sodium.

Coke thus prepared is said to yield very superior iron in the cupola.

The preparation of coke has become a very extensive trade in this country within the last few years, in consequence of the increasing demand for railways and smelting operations; and the small coal, which was formerly of comparatively little value, has, in consequence of its peculiar adaptation to the production of coke, become a most important source of profit. So superior indeed is the coke made from small coal, both in density and strength, that it has even been proposed, in some cases, to grind the large coal for the purpose of making coke. There is some difference as respects the management of different qualities of coal in converting them into coke. The great object of the coke-burner is to produce a dense, compact coke, in such large masses, that in cases of long carriage, either by sea or land, these lumps may arrive at their destination without being broken up into such small pieces as to be unfit for smelting or locomotive purposes. This object is most effectually accomplished by burning large quantities of coal of sufficient depth to ensure long upright masses of coke in the oven at the end of the operation. But some coals, which contain a large percentage of ash, form a coke which melts easily, and when the ovens are charged with too deep a load of such coal, the coke melts, and covers the upper layers in such a manner that the air cannot penetrate to the unburnt coal at the bottom of the oven, and the produce is a mixture of charred coal and coke. On the other hand, when coal contains a large proportion of ash, which does not melt in coking, the same result is observed. This arises, however, from a different cause, the mass of coke at the top which is first formed continues to burn, leaving its ash on the surface, and this accumulating, at length falls down into the crevices, and chokes the fire before it has time to reach the bottom.

* One company near Newcastle-on-Tyne producing upwards of 200,000 tons per

annum.

layer of coal; thus, in Fig. 51, a is the unburnt coal, b that

partially coked, c the coke already formed, and d the ashes which have fallen down, and prevented the access of air to support the combustion. In such cases, the difficulty in the coking process is obviated by building smaller ovens, and loading them lightly.

When suitable coal exists, the advantages of heavy loading are very obvious. The whole mass of coke is longer exposed to

the hardening effect of the fire, and the heavy products of distillation, ascending from the bottom layers of coal, deposit large portions of their carbon on the side of the already formed coke, thus increasing the produce from a given quantity of coal. The ovens about to be described are now generally constructed to contain from 7 to 10 tons of coal, and produce a most superior coke, both as respects quality and quantity.

FIG. 52.

After the oven has ceased to burn, the whole is made perfectly close, and the coke kept in the heated state for some time, contracts in bulk, and becomes much harder and more compact. In coke intended for iron and other smelting operations, where it is exposed to the action of a blast, long before it really does any efficient duty, the above qualities are most valuable. Figs. 52, 53 and 54, are drawings of the ovens, showing them in different positions, about 10 feet in diameter, and 4 feet from the floor to the springing of the dome; and in favourable localities each oven will cost about £25: a, Fig. 52, indicates a series of flues under the floor of the ovens, ending in a chimney b, the object of which is to assist in cooling the floor as rapidly as possible after the coke has

been drawn out, so that there may be no distillation of the coal

of the next charge, which would tend to produce an inferior coke at the bottom. It may also be noticed here, that the dome ought to be made as flat as possible consistent with durability, in order to reflect the heat as much as possible down upon the coal, and preserve the heat of the oven generally.

Figs. 55, 56, and 57, show a similar form of oven, in which the smoke is, however, carried along the flues a to a high chimney, thus avoiding the inconvenience of the large volumes of smoke which are poured forth by a range of coke ovens; b, Fig. 55, is a cast-iron pillar sup

porting a railway for conveying the coal waggons to any particular oven, so as to be charged in front, it having been found that when the coal is more equally spread in the ovens by manual labour, the charge is burnt off in less time than when the waggon discharges the coal into the oven all at once through the opening in the dome, independent of saving in the wear and tear of the dome occasioned by the passage of heavy coal waggons.

Fig. 58 represents the arrangement now generally adopted

where there are a number of coke-ovens, both for the purpose of economy in the construction and to carry off the products of combustion by one chimney.

A patent has been secured for drawing the whole of the coke at once from the oven, so as to save labour and time, but it has not come into general use. Indeed, the great point to which attention ought to be directed is, especially to increase the yield and the rapidity of the coking process, so as to save the great waste from the escaping gases and small residual coke, technically called brees. The large ovens have in some measure remedied this latter point, and an attempt to convey the waste heat between the lining of the ovens and the outer wall to facilitate the coking operation, did not answer in an economical point of view, the first and subsequent expenses amounting to more than the gain.

M. Ebelmen has made an interesting series of experiments to determine the composition of the gas which escapes from coke-ovens. The coke-ovens of Seraing, constructed as shown in the accompanying drawings, were the subject of the experi

ments.

Fig. 59 shows a vertical section through the ovens, and the boiler which is set above them, and a longitudinal elevation of a part of the range.

VOL. I.

9