Again, there are others in which the spectrum seems decidedly somewhat shortened at the red end compared with my own, and the luminosity curves point to them as being strictly colour deficient in the red and nothing else. As they see all colours, they have been classed as another form of normal vision. The deficiency in both these cases is so small that white is their neutral colour, but there is evidence that the hues are slightly changed. I do not wish any one to accept my deductions as being more correct than those who hold differently, but the results of examination by the luminosity methods appear to me difficult to reconcile with any other view. There are, however, a large number of cases in which, though complete red- or green-blindness is wanting, there is no doubt that more than slight colour deficiency exists. For instance, in Fig. 23 we have the curve of luminosity of the spectrum as measured by a very acute scientific observer, and it is compared with that of normal colour vision. He certainly is not completely blind to any sensation. An inspection and comparison of the two curves will show that he is defective in the green sensation, although it is present to a large extent. The deficiency is obvious enough. An endeavour to find his neutral point was most interesting. At 39 in the scale he saw a little colour, but at 39·5 all colour had vanished, and between the coloured patch and the white he saw no difference. This similarity he saw till 47·3 in the scale, when he began to see a faint trace of colour. There is a large piece of the spectrum, then, which to him is grey. It must be recollected that all three sensations were excited in this region, but some more than others. Now, experiment has shown that, with normal vision, two per cent. of any colour may be mixed with a pure colour without its being perceived. It is not surprising, therefore, that although the red, or the green, or the blue may be present in an intensity above that required to form white, yet the resulting sensation should pass for white. It may be remarked that red and white when mixed he never mistook for yellow, and he always recognised yellows and red; yellowish green, however, he called pale yellow. [Illustration: FIG. 23.] Another example of partial red-blindness is also instructive. Fig. 23 also shows it graphically. There is no doubt as to the nature of the defect. The spectrum is slightly shortened, and the luminosity of this part of the spectrum is less than that of normal vision. There was no difficulty in distinguishing every colour, though the positions of the colours from yellow to green seemed to be shifted; but no neutral point could be traced. Apparently, both this case and the former are about equally colour defective; but in this last the same reasons do not apply for the existence of a neutral point. (For measures see page 214.) CHAPTER VII.