I must admit that these results far exceeded my expectations, although they only confirmed the figures claimed by the patentee; and there are not wanting indications that, when worked on a large scale and continuously, they might be even still further lowered, as it is impossible to obtain the most economical results when making less than 10,000 cubic feet of the gas, as the proper temperature of the walls of the generator are not obtained until after several makes; and it is only after about 8,000 cubic feet of gas has been made that the best conditions are fulfilled. It will enable a sounder judgment to be formed of the working of the process if the complete experimental figures for a make of gas be taken. COMPOSITION OF THE GAS. Hydrogen. 46.75 Olefines. 7.59 Ethane. 6.82 Methane. 11.27 Carbon monoxide. 11.65 Carbon dioxide. 0.50 Oxygen. 0.17 Nitrogen. 8.25 ------- 100.00 UNPURIFIED GAS CONTAINED Carbon dioxide. 2.32 per cent. Sulphureted hydrogen. 2.84 " Total sulphur per 100 cu. ft. = 6.67 Ammonia. nil Bisulphide of carbon. nil Gas produced Naphtha used Gals. Pts. 1st. Make. 3,600 cu. ft. 10 7 2d. " 2,800 " 7 6 3d. " 2,300 " 5 3 ------ --- -- 8,700 24 0 The last portion of the table shows the economy which arises as the whole apparatus gets properly heated. Thus the first make used 3 gallons naphtha per 1,000 cubic feet, the second 2 gallons 6 pints per 1,000 cubic feet, and the third 2 gallons 4 pints per 1,000 cubic feet, and it is, therefore, not unreasonable to suppose that in a continuous make these figures could be kept up, if not actually reduced still lower. In introducing the oil it is not injected, but is simply allowed to flow in by gravity, at a point about half way up the column of fuel, the taps for its admission being placed at intervals around the circumference of the generator, and oil at first begins to flow down the inside wall of the generator, but being vaporized by the heat, the vapor is borne up by the rush of steam and water gas, and is cracked to a permanent gas in the upper layer of fuel. This I think is the secret of not being able to use heavier grades of oil, these being sufficiently non-volatile to trickle down the side into the fire box at the bottom, and so to escape volatilization. I have tried to steam-inject the oil, but have not found that it yields any better results.