Data on Colour- Blindness. Ill 



In order, now, to show the connexion between these 

 appearances and those of the pigmental colours, it is necessary 

 to state how the spectrum appearances are explained. For 

 this purpose I have added the diagram, .fig. 5, of the two 

 colour sensations in the manner shown by Clerk Maxwell in 

 his paper of 1860. The ordinates of the curves * give a 

 general idea of the variations of the luminous strength of the 

 sensations at the various points of the wave-length. Begin- 

 ning at the left-hand end of the figure, the yellow curve near 

 its origin has a small ordinate, and consequently the colour- 

 impression is dark. But further on the luminosity rapidly 

 increases till it arrives at its maximum. This series of colour- 

 impressions, therefore, exactly corresponds to the pigmental 

 class No. 6, i.e. yellow fully saturated, but darkened by 

 varying gradations of illumination. 



A little beyond the maximum, where the yellow curve 

 begins to descend, a new element comes in, as we enter on 

 the wave-lengths which excite the other or blue curve; and 

 beyond this we have both the yellow and blue acting together. 

 The effect of this mixture is to make white, and so to dilute 

 the yellow colour, and to diminish its saturation. This 

 accounts for the paler hue of the yellow as it approaches the 

 neutral, the dilution constantly increasing till the power of 

 the two colours becomes equilibrated, when the yellow dis- 

 appears and we get the neutral hue. 



In this part of the spectrum, therefore, we have two causes 

 of diminution of chromic strength : first the gradual loss 

 of luminosity of the yellow (indicated by the reducing 

 ordinate), and, secondly, the gradual loss of its saturation. 

 This combination will at once identify the colour-impression 

 with the pigmental class 7, where the yellow is mixed with 

 both white and black. 



The blue division of the spectrum is subject to a precisely 

 similar explanation. 



The position of the neutral point, in the powerful blue- 

 green, shows that the particular normal green hue at that 

 place is invisible as a separate colour to the dichromic eye. 

 But it has been explained in my paper that there is also a hue 

 of red, which is invisible. This does not exist in the spectrum, 

 but by mixing a spectral red with a spectral blue or violet in 

 proper proportions, the invisible hue is obtained. Helmholtz 

 calls it purple, and it is, to the normal eye, the complement 

 of the invisible green. 



* These curves do not express any definite magnitudes or quantitative 

 data. Prof. Maxwell did, I know, get out the exact curv r es for my 

 vision, and promised them to me, but I never had them. They will, 

 however, [ think, not differ much from those here shown. 



