oxygen in carbonic oxide. Equal volumes of the two gases were passed for two hours through pumice-stone heated to a dull red. The water collected weighed 1-2 grains and the carbonic acid 0.1 grain. A similar experiment was performed with the pumice-stone heated to full red. The water collected weighed 4.6 grains and the carbonic acid 1-2 grains. In this case the pumice-stone was blackened; and when it was heated in oxygen 1.59 grains of carbonic acid were obtained, equal to 43 grains of carbon; which must have been deposited by the dissociation of the carbonic oxide. Enough has been advanced in the present section to prove the difficulty, perhaps the impossibility, of demonstrating the precise nature of the part played by any hydrogen which may be present in the blast furnace. So far, however, as the promotion of reduction is concerned, it is probably immaterial whether this gas deoxidizes the ore direct, or whether it decomposes carbonic acid with the liberation. of carbonic oxide. In either case the presence of hydrogen must, it may be supposed, help to maintain the reducing energy of the gases. All my experiments tend to prove not only that this energy is intensified by the co-operation of hydrogen with carbonic oxide; but that a mixture of the two oxides of carbon (CO and CO2) had its power of deoxidizing iron ore increased by the presence of this third gas, perhaps beyond that point which might be expected from its relative volume. Experiments were first made to ascertain the mode of action of mixtures of the two oxides of carbon on the ore, with the following It A mixture was then made resembling somewhat in composition blast-furnace gas; and into this some hydrogen was introduced. then contained CO, 15·9, CO 33·5, H 4·2, N 464 100 volumes. In this, 100 volumes of reducing gases consist of CO 88-86, H 11-14; and the proportion of CO, to CO is 47 volumes of former to 100 of the latter, or taking the CO and H conjointly (57·7 volumes) we have 42 volumes of CO2 to 100 of reducing gases. After an exposure of 10 hours to a temperature of say 842° F. (450° C.), antimony melting, the loss of oxygen amounted to 11.7 per cent. of that originally combined with the iron in the ore. At a later period in the present section when the use of raw coal in the blast furnace is being considered we shall be better able to discuss the practical value of hydrogen as a reducing agent. Messrs. Baird & Co. of the Gartsherrie works in Lanarkshire have recently introduced a new industry at this well known establishment, which promises to become one of immense importance even in a national point of view. The gases from two of their blast furnaces fed with raw coal are aspirated by means of a large ventilating fan through huge chambers of sheet iron, in which they are cooled by suitable means, and the tar and ammonical water are thereby condensed. From the former the usual oils and pitch are obtained, and the latter, being used for the manufacture of sulphate of ammonia, will be able to render valuable service to agriculture. The average quantity of oxygen in the coals used at Gartsherrie is stated to be 11.06 per cent., and of hygrometric moisture 7.77 per cent.-or per ton of coal 2.212 cwts. of oxygen and 1·554 cwts. of moisture. Now this quantity of oxygen if united with hydrogen and added to the moisture would be equal to 451 lbs. or 45°1 gallons of water. I am informed by my friend Alderman Thomas Hedley, chairman of the gas works at Newcastle-on-Tyne, that 30 gallons of water are condensed per ton of coal distilled. A similar result is obtained at the coke works of Messrs. Pease & Partners, in which the coal is coked in closed ovens. The Durham coal contains on an average under 1 per cent. of water with from 10 to 11 per cent. of oxygen, which latter, by combining with hydrogen in the coal, will afford just about the 30 gallons of water referred to. According to Mr. Cosh, of the Gartsherrie works, 30 gallons also is the amount of water condensed in the apparatus referred to above; but the Gartsherrie coal, contain ing 7.77 per cent. of moisture and fully as much oxygen as the Durham coal, it follows, irrespective of any water formed by the action of hydrogen as a reducing agent, that a good deal must escape condensation in the apparatus used for collecting the tar and ammonia. In the Gartsherrie coal hydrogen exists to the extent of 5 per cent. of the weight of the coal; equal therefore to 2 cwts. for every ton of pig iron made. Assuming each ton of metal to contain 18.6 cwts. of iron (Fe), there will be 7.97, say 8 cwts. of oxygen, to be removed in the process of its reduction. For this 1 cwt. of hydrogen would suffice; so that there is twice as much of this element present as is required to withdraw all the oxygen from the oxide of iron. The want of any authentic information on the office performed by hydrogen in the blast furnace induced me to request permission, from a friend who is a furnace owner in Scotland, to extend my observations on smelting iron to cases when, instead of coke, raw coal is the fuel employed. Every possible facility was extended to my assistant, Mr. Rocholl, and I now propose to examine the results of his analyses and observations. The fuel employed at the furnaces in question is the well known splint of the Lanarkshire coal-field. In lumps it burns without any approach to fusion, or one piece adhering to another, but when powdered and heated in a crucible it coheres and forms a dense coke without any apparent change of volume. The coal appears perfectly dry to the eye and touch, nevertheless when dried at 100° C. (212° F.) it parts with more than 11 per cent. of water. Its constituents were found to be: The ironstone employed was partly calcined black-band, containing 1.34 per cent. of iron as protoxide and 48.63 per cent. as peroxide, together 49.97 per cent.; partly clay-band with 48 per cent. of iron as protoxide and 40.89 as peroxide, together 41.37 per cent.; and partly Spanish hematite ore containing 14.6 per cent. of water. The pig iron on analysis gave as follows: The particulars of the two furnaces which were examined are contained in the following Table: Average (12 observa- 200 392 Average (14 observa tions over 4 hours)... C. .F. 412 773 472 881 427 800 138 280 254 490 observa-} 190 374 42.39 37.46 10.93 |