236 COLOUR VISION 



complete (approximate dichromatism). Konig and Abney worked on 

 this theory and have brought forward much evidence in its favour. 

 There is, however, another possibility, viz., that there is a displacement 

 of the normal sensation curves. If, for example, the green sensation 

 curve is of the same form and magnitude as in the normal, but is dis- 

 placed towards the red end of the spectrum, the colour sensations of 

 the individual will be abnormal. It is clear that it is possible to have 

 a combination of the two characteristics, shift of one or more curves 

 combined with reduction of one or more fundamental sensations. Of 

 cases bearing out this conjecture we have at present no evidence, but 

 there is good evidence that both reduction cases and cases in which 

 one curve, otherwise normal, is shifted, actually occur. In this chapter 

 we shall consider the cases of approximate dichromatism only. 



If, for example, the red sensation is defective, the ordinate of the 

 red sensation curve for any given wave-length will be less than the 

 corresponding ordinate for the normal, e.g., one-half. Then for all 

 other wave-lengths the ordinates of the red sensation curve are also 

 half those of the normal curve. Hence it follows that the area of the 

 red sensation curve must be half the area of the corresponding normal 

 curve. 



We have no method of measuring in absolute units the sensation 

 produced when light of a given intensity stimulates the retina. Con- 

 sequently we are unable to determine whether the maximum ordinate 

 of the sensation curves is the same for all persons. What is actually 

 done is to compare the sensations produced by given amounts of light 

 of different colours for each observer. Thus we take some one kind 

 of light as a standard and compare the relative stimulation of the 

 components produced by other kinds of light with that produced by this 

 standard light. In determining a luminosity curve the standard is the 

 white light of the recombined spectrum. In examining the colour- 

 defective the complication arises that this standard white is not the 

 same for the normal and colour-blind observers. The defect influences 

 the sensation derived from stimulation with white light as well as the 

 sensations derived from colours. Watson has so lucidly explained the 

 bearing of this fact upon the deductions that I cannot do better than 

 quote his remarks 1 . 



" As long as we confine our attention to the part of the spectrum 

 between the extreme red and the blue, the effect of the blue sensation 

 on the luminosity may in general be neglected as being too small to 

 1 Proc. Boy. Soc. Lond. A. LXXXVIII., p. 410, 1013, 



