It will appear in Chapter XI. that some of the first endeavours to avoid burner troubles were based on the dilution of the acetylene with carbon dioxide or air before the gas reached the place of combustion; while the subsequent paragraphs will show that the same result is arrived at more satisfactorily by diluting the acetylene with air during its actual passage through the burner. It seems highly probable that the beneficial effect of the earliest methods was due simply or primarily to the dilution, the molecules of the acetylene being partially protected from the heat of the burner by the molecules of a gas which was not injured by the high temperature, and which attracted to itself part of the heat that would otherwise have been communicated to the hydrocarbon. The modern injector burner exhibits the same phenomenon of dilution, and is to the same extent efficacious in preventing polymerisation; but inasmuch as it permits a larger proportion of air to be introduced, and as the addition is made roughly half-way along the burner passage, the cold air is more effectual in keeping the former part of the tip cool, and in jacketing the acetylene during its travel through the latter part, the bore of which is larger than it otherwise would be. INJECTOR AND TWIN-FLAME BURNERS.--In practice it is neither possible to cool an acetylene burner systematically, nor is it desirable to construct it of such a large mass of some good heat conductor that its temperature always remains below the dissociation point of the gas. The earliest direct attempts to keep the burner cool were directed to an avoidance of contact between the flame of the burning acetylene and the body of the jet, this being effected by causing the current of acetylene to inject a small proportion of air through lateral apertures in the burner below the point of ignition. Such air naturally carries along with it some of the heat which, in spite of all precautions, still reaches the burner; but it also apparently forms a temporary annular jacket round the stream of gas, preventing it from catching fire until it has arrived at an appreciable distance from the jet. Other attempts were made by placing two non- injector jets in such mutual positions that the two streams of gas met at an angle, there to spread fan-fashion into a flat flame. This is really nothing but the old fish-tail coal-gas burner--which yields its flat flame by identical impingement of two gas streams--modified in detail so that the bulk of the flame should be at a considerable distance from the burner instead of resting directly upon it. In the fish-tail the two orifices are bored in the one piece of steatite, and virtually join at their external ends; in the acetylene burner, two separate pieces of steatite, three-quarters of an inch or more apart, carried by completely separate supports, are each drilled with one hole, and the flame stands vertically midway between them. The two streams of gas are in one