These general principles govern all burners, and we will now take the ordinary forms in detail. In the ordinary flat flame burner, given a good non-conducting material, and a well regulated gas supply, little more can be done, while burning it in the ordinary way, to increase its luminosity; and it is the large surface of flame exposed to the cooling action of the air which causes this form of burner to give the lowest service of any per cubic foot of gas consumed. Much is done, moreover, by faulty fittings and shades, to reduce the already poor light given out, because the light-yielding power of the flame largely depends upon its having a well rounded base and broad, luminous zone; and when a globe with a narrow opening is used with such a flame—as is done in 99 out of 100 cases—the updraught drags the flame out of shape, and seriously impairs its light-giving powers, a trouble which can be got over by having the globe with an opening at the bottom not less than 4 inches in diameter, and having small shoulders fixed to the burner, which draw out the flame and protect the base from the disturbing influence of draughts.
The Argand burner differs from the flat flame burners in that a circular flame is employed. The air supply is regulated by a cylindrical glass, and this form of burner gives a better service than the flat flame burner, as not only can the supply of gas and air be better adjusted, but the air being slightly warmed by the hot glass adds to the temperature of the flame, which is also increased by radiation from the opposite side of the flame itself.
The chief loss of light in such a burner depends upon the fact that, being circular, the light from the inner surface has to pass through the wall of flame, and careful photometric experiments show that the solid particles present in the flame so reduce its transparency that a loss amounting to about 25 per cent. of light takes place during its transmission.
The height of the flame also must be carefully adjusted to the size of the flame, as too long a chimney, by increasing the air supply unduly, cools, and so lowers the illuminating power of the flame. Experiments with carbureted water gas gave the following results, with a consumption of 5 cubic feet per hour:
| Size of Chimney. | Height of Flame. | Candle Power. | |
| 6 X 1-7/8 | 2-1/2 | 21 | |
| 7 X 1-7/8 | 2-1/4 | 21.3 | |
| 8 X 1-7/8 | 2-1/8 | 20.8 | |
| 9 X 1-7/8 | 1-7/8 | 18.2 |
For many years no advance was made upon these forms of burner, but when, ten years ago, it was recognized that anything which cools the flame reduces its value, while anything which increases its temperature raises its illuminating power, then a change took place in the forms of burner in use, and the regenerative burners, introduced by such men as Siemens, Grimston, and Bower, commenced what was really a revolution in gas lighting.
By utilizing the heat contained in the escaping products of combustion to raise the temperature of the gas and air which are to enter into combination in the flame, an enormous increase in the temperature of the solid particles of carbon in the flame is obtained, and a far greater and whiter light is the result.
The Bower lamp, in which (at any rate in the later forms) the flame burns between a downward and an upward current of air, was one of the first produced, and so well has it been kept up to date that it still holds its own; while as types of the "inverted cone" regenerative burner, we may also take the Cromarty and Wenham lights, which have been followed by a host of imitators, and so closely are the original types adhered to that one begins seriously to wonder what the use of the Patent Office really is.
The Schulke, and the last form of Siemens regenerative burner, however, stand apart from all the others by dealing with flat and not conical flames, and in both regeneration is carried on to a high degree. The only drawback to the regenerative burner is that it is by far the best form of gas stove as well as burner, and that the amount of heat thrown out by the radiant solid matter in the flame is, under some circumstances, an annoyance. But, on the other hand, we must not forget that this is the form best adapted for overhead burners, and that nearly every form of regenerative lamp can be adapted as a ventilating agent, and that with the withdrawal of the products of combustion from the air of the room, the great and only serious objection to gas as an illuminant disappears.
When coal gas is burned, the hydrogen is supposed to be entirely converted into water vapor, and the carbon to finally escape into the air as carbon dioxide; and if this were so, every cubic foot of gas consumed would produce approximately 0.52 cubic foot of carbon dioxide and 1.34 cubic feet of water vapor, while the illuminating power yielded by the cubic foot of gas will, of course, vary with the kind of burner used.