The amount of light developed by any coal-gas flame is directly proportional to the degree of intensity to which the temperature of these carbon particles is raised, and the length of time they remain in the flame before being finally consumed. It becomes, therefore, a matter of considerable importance to know the conditions which are most conducive to the early liberation in the flame of free carbon, and the attainment by it of an exalted temperature.

Looking at the flame (say) of a common slit burner, it is seen to be divided into two sharply defined and wholly distinct portions. First, there is—immediately surrounding the burner head, and extending to some distance from it—a dark, transparent area, which, on closer examination, is found to consist of unignited gas enclosed in a thin envelope of bright blue flame. Second, there is (beyond this central area) a zone, or belt, of brightly luminous flame, white and opaque; the latter property indicating the presence of solid matter at this What is a gas flame? part of the flame. That the dark central portion of the flame consists chiefly of unignited gas may be shown in various ways, in addition to the evidence afforded by its complete transparency. Thus, if a small glass tube be taken, and its lower end inserted in the flame at this point, the unburnt gas will pass up the tube, and may be lighted at its upper extremity. A splinter of wood thrust through this portion of the flame is charred first at the two edges of the flame; while, in like manner, a piece of platinum foil remains dull in the centre of the flame, and glows only at the points of contact with the outer air. The presence of solid carbon in the luminous portion of the flame may be shown by inserting therein any cold substance (such as a piece of metal or porcelain), which, reducing the temperature of the heated particles of carbon below the point at which they are consumed, becomes instantly coated on its under surface with a deposit of soot. Or, if the flame be suddenly cooled by gently blowing upon its surface, the same result is brought about; clouds of soot are given off, and the flame "smokes."[3]

Fig. 4.—Showing the Two Zones of the Flame, and the Method of Demonstrating the Presence of Unburnt Gas in the Flame.

The existence, in the midst of the flame, of an area of unconsumed gas is due to the cold gas, as it issues from the burner, cooling the interior of the flame below the temperature required for its ignition, as well as to its not at once meeting with sufficient air for complete combustion. The causes which affect the luminous zone of the flame are not so readily explained. It has been stated that the luminosity of the flame is due to the particles of carbon, which are separated out of the hydrocarbons in the gas, being raised to a white heat. To decompose the hydrocarbons, a very high temperature is required; and, on account of the cooling effect of the stream of cold gas, this is not attained except at some distance from the burner. The abstraction of heat by the burner itself is also a cause of the reduction of the temperature of How the flame is cooled. the flame; and, on this account, burners of porcelain, steatite, or similar composition, being bad conductors of heat, have an advantage over those made of metal. So considerable is the cooling influence of the gas stream, that, within certain limits, the distance, from the burner head, at which the luminosity of a flame commences, is proportionate to the velocity with which the gas issues; or, in other words, the pressure at which it is delivered from the burner. The effect is heightened by the tendency (which has been before remarked) of a stream of gas, issuing under pressure, to draw upon itself and mix with the surrounding air. Thus, with each increment of pressure the luminous zone of the flame is farther removed, until a point is reached at which the gas is so mixed with air before being consumed that the luminosity of the flame is completely destroyed.

But it must not be assumed, because of the foregoing remarks, that the pressure at which the gas issues from the burner is altogether an unmixed evil. In flat-flame burners it fulfils the important function of promoting intensity of combustion, by bringing the white-hot particles of carbon into intimate and rapid contact with the air that is necessary for complete combustion. In Argand burners this duty is discharged by the glass chimney; but with flat-flame burners it devolves entirely upon the pressure at which the gas issues from the Effects of pressure in the gas supply. burner. It will be seen, therefore, that the pressure of the gas is a factor of considerable importance in determining the amount of light afforded by a gas flame, as it is a matter requiring careful adjustment with each and every burner. On the one hand, with an excessive pressure the intensity of combustion is increased; but the separated carbon does not remain so long in the flame. The area of luminosity is thereby decreased, and the total light yielded is reduced. On the other hand, with insufficient pressure the combustion is not energetic enough to raise the particles of carbon to a white heat; consequently, the illuminating power of the flame is feeble, or else the carbon escapes unconsumed as smoke.

The thickness of the flame produced by any burner has also an important bearing upon the degree of light afforded; and this property of thickness, again, is dependent upon the width of slit, in the case of batswings (or, in the case of union-jets, upon the size of orifices), and the pressure at which the gas is supplied. The thickness of the flame yielded by any burner will obviously vary inversely with the pressure at which the gas is supplied to it. With a thin flame, all parts of the flame are so completely exposed to the air, that the particles of carbon are no sooner raised to the temperature required to enable them to give out light than they are entirely consumed. With a thicker flame the carbon separated in the midst of the flame exists for a sensibly longer period of time in the white-hot state before it reaches the outside of the flame, and meets with sufficient oxygen for its complete combustion. Thus we find that the best flat-flame burners have comparatively wide orifices; while the pressure at which the gas is delivered from the burner is carefully reduced to the lowest point at which a firm flame is obtained, without smoke. Similarly, in the best Argands the pressure is considerably diminished within the burner, and the gas allowed to issue gently through relatively large holes; while the chimney is carefully adapted to draw upon the surface of the flame just sufficient air to completely consume the quantity of gas which the burner is calculated to deliver.

IMPROVEMENTS IN FLAT-FLAME BURNERS.

Although, there is no doubt, they were made empirically, and in ignorance of the real effects of pressure upon the flame, the first steps towards increasing the efficiency of flat-flame burners were in the right direction of reducing the excessive pressure at which the gas was formerly allowed to burn. They consisted in the adoption of simple arrangements for obstructing the passage of the gas through the burner, and so retarding its flow. The crudeness of the means which were employed is sufficient evidence that the end aimed at was, at best, but dimly discerned. The body of the burner was stuffed with wool, or pieces of wire gauze; which impeded the progress of the gas;