All our practical methods for making light consist in heating a solid body in one way or another. In the case of the electric light we heat electrically a carbon rod or filament, or else, as in the Nernst lamp, a rod composed of magnesia and the rare earths. In the case of the lime-light we heat a cylinder of lime. In an ordinary gas or candle flame we heat small particles of carbon, and the same is the case even in the sun itself. But this process manufactures not only the single octave of radiation which can affect our eyes as light, but a dozen other octaves of radiation to which they are insensible. Hence it follows that of the whole radiation from a gas flame, only about 3 per cent. is eye-affecting light, the remainder is dark heat. In the case of an incandescent electric lamp, this _luminous efficiency_ may amount to 5 per cent., and in the electric arc to 10 or 15 per cent. There is, however, always a great dilution of the useful light by useless dark heat. The proportion of the light or eye-affecting radiation to the dark heat in the total radiation from any source of light increases with the temperature, but it is not always merely a question of temperature. Thus the electric arc is hotter than a candle flame, and the sun is hotter than the electric arc. Hence, whilst the luminous rays only form 3 parts out of 100, or 3 per cent. of the radiation of a candle, they constitute 10 to 15 per cent. of those of the electric arc, and more than 30 per cent. of those of the sun. On the other hand, the glow-worm and the fire-fly seem to have possession of a knowledge and an art which is as yet denied to man. It has been shown by Professor Langley and Mr. Very that nearly the whole of the radiation from the natural torch of the fire-fly is useful light, and none of it is useless dark heat. Hence these photogenic (light-producing) insects have the art, which we have not, of creating _cold_ light, or unadulterated luminous radiation. At the present moment in ordinary incandescent or glow-lamp electric lighting we require to expend an amount of power, called one horse-power, to produce illumination equal to that of 600 candles. Supposing, however, that all our power could be utilized in generating merely the rays useful for vision, or which can impress our eyes, we might be able to create by the expenditure of one horse-power more than twenty times as much illumination, that is, a light equal to 12,000 candles. These figures show us what rewards await the inventor who can discover a means of generating æther waves having wave-lengths strictly limited to the range lying between the limits 0·4μ and 0·7μ without, at the same time, being obliged to create radiation comprising longer waves which are not useful for the purpose of rendering objects visible to us. For the purposes of artificial illumination we require only the æther waves in this one particular octave, and nothing else.