16-candle coal gas. A cubic foot of methane requires 2 cubic feet of oxygen, or 10 cubic feet of air, for its complete combustion, when it yields 1 cubic foot of CO2 and 2 cubic feet of water vapour. Methane is present in coal-gas to the extent of from 30 to 40 per cent.

Ethylene has an illuminating power of 70 candles per 5 cubic feet, and requires for its combustion a larger proportion of oxygen than is the case with methane, 1 cubic foot of ethylene combining with 3 cubic feet of oxygen to form 2 cubic feet of CO2 and 2 cubic feet of water vapour, and in order to obtain this quantity of oxygen, the combustion of a cubic foot of the gas will require 15 cubic feet of air. Benzene is an extremely volatile liquid, its vapour burning with a heavy red smoky flame. In arriving at its illuminating power, we have to see the effect produced by known quantities of benzene vapour diffused in combustible gases, such as hydrogen, as it is not possible to burn the pure liquid for the development of light even in lamps specially constructed for the purpose, as it is not possible to obtain a flame free from smoke. Drs. Frankland and Thorne determined its value in this manner, and found the illuminating power per 5 cubic feet of its vapour to be 420 candles. Knublauch gives its illuminating value as 820 candles.

A cubic foot of benzene vapour requires for its complete combustion 7 cubic feet of oxygen, or 37 cubic feet of air, yielding as the products of combustion 6 cubic feet of CO, and 3 cubic feet of water vapour.

"Another hydrocarbon which is present in only minute quantities in the coal-gas itself, but is developed in the changes taking place in the flame, and plays a very important part in causing luminosity, is acetylene (C2H2).” Water at ordinary temperatures dissolves its own volume of the gas, and this is why it is present in such small quantities in coal-gas, as scrubbing and storage practically

remove it. One cubic foot of this gas requires 2 cubic feet of oxygen, or 12.5 cubic feet of air for its complete combustion, yielding 2 cubic feet of CO, and 1 cubic foot of water vapour. Its illuminating value has not yet been determined, owing to its solubility, but it has been observed to burn with a rich smoky flame.

The illuminating power of coal-gas is without doubt entirely due to the hydrocarbons just mentioned, and the prevalent idea entertained up to sixteen years ago was that the illuminating value depended upon the amount of ethylene, but this could not be the case, for, as shown by Professor Lewes, if it were true, 4 per cent. of ethylene, mixed with 96 per cent. of a combustible diluent, such as hydrogen, should give a 16 to 17-candle gas, whereas a mixture of 10 per cent. of ethylene and 90 per cent. of hydrogen has no luminosity. Barthelot concluded that the illuminating value of coal-gas was due to benzene vapour, but the opinion of Professor Lewes is (and anything originating from him is entitled to the greatest possible respect), that methane is one of the most important of the hydrocarbons present when the gas is burnt in such a way as to evolve from it its proper illuminating power, while the benzene vapour, small as the quantity is, comes next in importance, and the ethylene last. In the words of Professor Lewes, "it is the combined action of the hydrocarbons present which gives the effect, not any one of them acting alone."

We must now briefly discuss the properties of the “ combustible diluents" which give volume to the gas, and produce a negative effect upon the light emitted by the flame. The most important of these is the hydrogen which constitutes about half the volume of ordinary coal-gas. Hydrogen burns in contact with oxygen, with a non-luminous but intensely hot flame. In burning it produces a greater heat than the same weight of any other substance known, 1 lb.

of hydrogen on combining with 8 lb. of oxygen giving out as much heat as will raise 34,462 lb. of water through 1 degree centigrade. A cubic foot of hydrogen when burnt combines with half a cubic foot of oxygen to form a cubic foot of aqueous vapour. Carbonic oxide, the other combustible diluent, exists in coal-gas to the extent of from three to eight per cent. The action of the diluents in coalgas upon its illuminating power has been determined by taking ethylene with an illuminating power of 68.5 candles as a representative of the hydrocarbons, and diluting with the various constituents of coal-gas. Working in this way, Dr. Percy F. Frankland obtained the following results.

"These results show that with the combustible constituents, hydrogen reduces the illuminating power least with large quantities of the hydrocarbons, but that methane is preferable when in excess, as with low percentages of the illuminant, especially when burnt at a high temperature, methane itself becomes an illuminating agent. This is due to the fact that although when methane burns at ordinary temperature it is non-luminous, at a high temperature some of it is broken up into acetylene, which gives it luminosity.

"Carbonic oxide is the most injurious of the combustible diluents, 80 per cent. mixed with ethylene rendering it nonluminous, a result which would require 90 per cent. of hydrogen.

"The constituents of an illuminating gas have a very great influence upon the size of a coal-gas flame. Both hydrogen and carbonic oxide burn with an extremely short flame, as they only require one-half their own volume of oxygen for their complete combustion, and are able to obtain this with a very short travel from the burner, while methane, requiring twice its volume of oxygen, burns with a flame nearly four times the height, if consumed at the same rate of flow. The percentage of ethylene, ethane, and benzene, in coal-gas is so small, that they can do but little towards increasing the size of the flame, although when pure ethylene or ethane is burnt alone it would not be possible to consume it in an ordinary burner at the rate of 5 cubic feet per hour. Using the London Argand as employed for gas testing, it would not be possible to burn much more than 2 cubic feet of pure ethylene or ethane per hour without the flame smoking. Using ordinary 16-candle London coal-gas enriched by cannel, it may be roughly stated to yield a 3-inch flame in the London Argand, the height of the flame being due to methane, of which such gas contains about 35 to 40 per cent. If, however, car

buretted water-gas is taken, I have frequently seen a 2-inch flame emitting light equal to 22 candles value, this being (Lewes) due to its only containing 16 to 20 per cent. of methane. The influence of incombustible diluents on the illuminating power of ethylene has also been determined by Dr. P. F. Frankland, with the following results :