When the gas is lighted at the burner, and the glass closed, the burner begins to act at once, although some minutes are necessarily required to elapse before its full brilliancy is gained. The cold air passes in through the tubes provided for it, and when these are heated to the fullest extent on their outside, by the hot fumes from the burner, they so readily part with their heat to the air that a temperature of 1,000° to 1,200° Fahr. is easily obtained in the air when it arrives inside, and commences in turn to heat the burner-tubes. The air-tubes are placed so as to intercept the hot gases as completely as possible; and also, of course, obtain heat by conduction from the sides of the annular body. It is evident that the number and dimensions of these tubes might be increased so as to abstract almost all the heat from the escaping fumes, but for the limitations imposed, first, by a consideration of the actual quantity of air required to support combustion, and, secondly, by the obligation to let sufficient ascensional power remain in the gases which are left to pass out through the upper chimney. If the gases are cooled too much, they will either fall back into the lamp and extinguish the flame, or will be removable only by the draught of a long chimney. It will probably be the aim of the inventor to balance these requirements, and so to produce burners with very short or longer chimneys, according as appearance is to be consulted or the highest possible effect produced. The burner is a ring of brass tubes of considerable diameter, in proportion to the quantity of gas consumed, and thus provides for the delivery of gas expanded by heat. In connection with this device an explanation may be found of the failure of the British Association Committee on Gas Burners to find any advantage from previously heating the air and gas consumed. The Committee did not make the necessary provision for the increased bulk of the combustible and its air supply, caused by their heightened temperature; and the same quantity of gas measured cold (at the meter) could only be driven through the ordinary small burner holes at a velocity destructive of good results. Herr Frederick Siemens perceived this in his early experiments, and not only increased the orifices of his burners, but provided for the closer contact of the more rarefied gas and air by the use of notched deflectors, which are now an essential part of his apparatus. Mr. Grimston also uses separate tubes of large area for his hot gas, but dispenses with deflectors, save in so far as the same duty may be performed by the plain lower edge of the inner cylinder of the lamp body, and the indentation of the glass beneath, which, as will be noticed, is made to follow the shape of the flame. It only remains now to speak of the flame and its qualities. It is, in the first place, a flame of hot gas, burning at an extremly small velocity of flow, and