Studies in Visual Sensation. 



467 



-afford a mid-sensation between the black and the reduced value of the 

 white. It is here assumed that the reduced illumination makes so 

 small a difference in the black as to be inappreciable and practically 

 negligible. 



Fortunately for experimental work a slight reduction of the illumina- 

 tion makes but little difference in the mid-point for sensation. A 

 reduction of the physical luminosity of the white paper by 15 per cent, 

 only reduces the sensation it affords by 4 per cent., and the additional 

 stimulus to be added to give the new mid-point is only 0'74 per cent. 



It may be pointed out that the general fact of the alteration of 

 sensation values by changes in the illumination is quite familiar. An 

 ill-lit engraving not only looks duller, but the relative intensities of the 

 shading are not preserved. And the fact would probably be more 

 noticeable were it not that we are daily accustomed to changes of 

 illumination of the same scene as the sun declines and sinks below the 

 horizon. 



I shall return presently to the question of illumination so as to bring 

 these facts into relation with the results of the further experimental 

 work to be ere long described. 



If the provisional scale represented by the graphic curve gives an 

 approximation to the relative values of stimulus and sensation, that is 

 to say, of physical luminosity and apparent brightness to the eye, we 

 may use it to interpret the facts which I mentioned at the outset with 

 regard to the physical illumination of a surface of white paper and its 

 apparent brightness. Let us suppose that with standard illumination 

 the luminosity of the surface is 64, the corresponding value for sensa- 

 tion in terms of brightness is 89. If now the physical luminosity is 

 reduced to one-fourth, it will have the value 64 ^ 4 = 16, the 

 corresponding value of which is, for sensation, 56. One-quarter the 

 illumination thus affords about two-thirds the brightness, which is 

 pretty well in accordance with the testimony of sensation. The 9 per 

 cent, luminosity of blue gives a sensation-luminosity or brightness of 

 44 per cent., and the 18 per cent, luminosity of red a brightness of 

 59 per cent. These again accord very fairly with the verdict of 

 the eye. 



Having now obtained a fairly even shading from white into black, 

 colours were next dealt with. Coloured papers were employed, and no 

 attempt was made to obtain colours with any approach to spectral 

 purity. Continuous shading will alone be considered for comparison 

 with that of black into white. The curves for five colours on black 

 were experimentally determined and plotted. The early work was 

 purely empirical. Plus and minus alterations at different parts of the 

 extent of each curve were introduced until the eye was satisfied that 

 there was an approximately even shading from black into the colour 

 under investigation. But when it was found that in each case for 



