The remarkable change in the character of a gas-flame produced by admitting air to mix with the gas before combustion is, of course, easily explained. The chemical action takes place with much greater facility under these circumstances. The union of the carbon in the coal gas with the oxygen then takes place so thoroughly and instantaneously that the carbon never seems to have abandoned the gaseous form even for a moment in the course of the transformation. But in the case where air is not permitted to mingle with the gas, the supply of oxygen to unite with the incandescent gases can only be obtained from the exterior of the flame. The consequence is that the glowing gas charged with carbon vapour is chilled to some extent by contact with the cold air. It therefore seems as if the union of the hydrogen with the oxygen permitted the particles of carbon in the flame to resume their solid form for a moment. But in that solid form these particles, being at a high temperature, have a wonderful efficiency for radiation, and consequently brilliance is conferred upon the light. Most of the particles of carbon speedily unite with the surrounding oxygen, and re-enter the gaseous state in a different combination. Some of them, however, may escape this fate, in which case they assume the undesirable form of smoke. The Bunsen lamp can thus be made to give an illustration of the fact that when carbon vapours receive a chill, the immediate effect of the chill is to transform the carbon from the gaseous form to myriads of particles in the liquid, or more probably in the solid form. In the latter state the carbon possesses a power of radiation greatly in excess of that which it possessed in the gaseous state, even though the gas may have been at a much higher temperature than the white-hot solid particles.