The objects in view were to test his perception of the spectrum colours, and then the extent of his retinal field for colour. This last is not recorded here. The spectrum colours were reduced to uniform luminosity between λ 4600 and λ 6600. Diaphragms containing holes of different sizes were placed in front of the last prism, and thus a round spot of monochromatic light of the same luminosity was produced upon the screen when a slit was passed through the spectrum. From the red end to λ 5270 he called the whole of the colours white, and from that point he began to see blue, called the colours bluish and blue. When the full illumination for all the colours was used, the same results were obtained. From this examination it would appear that he was totally deprived of the sensation of any colour except of blue. A subsequent examination of his perception of the luminosity of different rays, however, has to be taken into account, for in the first examination he had no light of pure white with which to compare the colours. In the next experiments, a strip of white light was placed in juxtaposition to the colour, and the results were slightly different. The table below gives his luminosity measures (Fig. 36). Col. I. is the empyric scale number, II. is the wave-length, III. the luminosity of the colour to the normal eye, IV. the luminosity to X., and V. the ratios of III. to IV. In the diagram, his luminosity curve X. is shown, its area being 1400 against 1650 for the normal eye. His central perception of light, as arrived at by the extinction method, was only two-thirds of that of the normal eye; hence his area of luminosity should be 1100. As it is 1400, the ordinates of the above curve should be multiplied by 0·8, to compare with that of the normal eye.