Next taking the emerald green disc, if we adopt the same procedure it will be found that for some shades of the ring there are two places in the spectrum from which the colours reflected give the same brightness. By plotting curves in exactly the same way as that shown for the curve of luminosity at page 78, substituting for the open aperture of the sector the angular value of the white used, we can show graphically the correct reflection for each part of the spectrum. Sometimes three places in the spectrum will be read, as giving equal reflections from the coloured disc and the grey ring. The accompanying figures show the results obtained for reflection from vermilion, emerald green, and French blue, after having made a correction for the white by adding the amount which the black reflects. The scale is that of the prismatic spectrum employed. On page 46 we stated that a white surface could be made to appear darker than a black surface, by illuminating the latter and cutting off the light from the former. By placing the black surface in place of one of the coloured ones, as shown in page 82, the luminosity of the black surface can be ascertained. In this case it was found that almost exactly 5% of the white light from the crater of the positive pole was reflected. In the table the original measures are shown, and also the corrected measures, and for convenience sake the intensity of every ray throughout the length of the spectrum reflected from white, has been taken as 100. The position of the reference lines on the scale (Fig. 15) are as follows-- Fig. 15.--Intensity of Rays reflected from Vermilion, Emerald Green, and French Ultramarine. B=101, C=96·25, D=89, E=79·9, F=71·5, G=53·5. VERMILION.