Violle and Féry, also Erdmann, have proposed the use of acetylene as a standard of light. As a standard burner Féry employed a piece of thermometer tube, cut off smoothly at the end and having a diameter of 0.5 millimetre, a variation in the diameter up to 10 per cent. being of no consequence. When the height of the flame ranged from 10 to 25 millimetres the burner passed from 2.02 to 4.28 litres per hour, and the illuminating power of the light remained sensibly proportional to the height of the jet, with maximum variations from the calculated value of ±0.008. It is clear that for such a purpose as this the acetylene must be prepared from very pure carbide and at the lowest possible temperature in the generator. Further investigations in this direction should be welcome, because it is now fairly easy to obtain a carbide of standard quality and to purify the gas until it is essentially pure acetylene from a chemical point of view. L. W. Hartmann has studied the flame of a mixture of acetylene with hydrogen. He finds that the flame of the mixture is richer in light of short wave-lengths than that of pure acetylene, but that the colour of the light does not appear to vary with the proportion of hydrogen present. Numerous investigators have studied the optical or radiant efficiency of artificial lights, _i.e._, the proportion of the total heat plus light energy emitted by the flame which is produced in the form of visible light. Some results are shown in the table on the previous page. Figures showing the ratio of the visible light emitted by various illuminants to the amount of energy expended in producing the light and also the energy equivalent of each spherical Hefner unit evolved have been published by H. Lux, whose results follow: