1910.] Light Perception to Colour Perception. 467 



X 5877. Nos. 2 and 3 have a spectrum of normal length and of exactly the 

 same size, whilst 1 has a spectrum which is shortened at both ends but more 

 shortened on the red side. The centre of the smallest patch of 1 is A 5543, 

 and the centres of 2 and 3 X5306 and A, 5174 respectively. It would appear 

 as if these were three similar cases with different absorption. 



We know that the portions of the spectrum which differ are those which 

 are most influenced by the varying ])igmentation of fclie yellow spot. It will 

 be noticed that the total ehect is the same, the spectrum being divided into 

 10 patches in each case, so that what is gained on one side is lost on the 

 other. No. 2 is the spectrum of Sir J. J. Thomson, further details of wliose 

 colour perception I have given in the ' Proceedings of the lioyal Society,' B, 

 vol. 76, 1905, p. 194. Sir J. J. Thomson, when making the match previously 

 mentioned with Eayleigli's apparatus, put more green than the normal in the 

 mixed colour, but the other two trichroniics made the normal match. It is 

 obvious that this alteration in light perception could not produce the defect 

 in colour perception which was found. 



By the use of the spectrometer described above, it is easy to detect cases of 

 defective colour perception of a smaller degree than the two classes I have 

 just described. In fact, if we regard the normal individual as hexachromicj 

 we can differentiate classes possessing five and four colour sensations 

 respectively, which may be called pentachromic and tetrachromic. In these 

 cases, the only loss is the power of differentiation of hue. The sensibility to 

 light, i.e. to luminosity, may be normal throughout the whole spectrum. 

 These minor cases will, as a rule, match wools perfectly, bixt may make slight 

 mistakes when asked to name the colours of single skeins. The pentachromic 

 may be regarded as differentiating the colours red, yellow, green, blue, violet, 

 while the tetrachromic would name only the colours red, yellow, green, violet. 

 It is evident that none of these classes can be explained by assuming the 

 absence of any substance or set of percipient elements sensitive to light over 

 a consideralde area of the spectrum. Any theory of colour vision must 

 explain the position of the colours and the varying size of the monochromatic 

 j)atches in each case. 



