Studies in Visual Sensation. 



471 



If these results be accepted as giving a sufficiently close agreement 

 it follows, first, that for colour shading the percentages of stimulus 

 required are dependent on the physical luminosity of the colours 

 employed, and secondly that all the data obtained by the method 

 of shading can be plotted on a single curve which exhibits the relation 

 of stimulus to sensation in visual impressions. It also follows that 

 if the intensity of illumination of a disc for white-black shading be 

 so reduced as to lower its luminosity to that, say, of orange under 

 full illumination, the mid-point value will be the same as that for 

 orange on black. I am instituting experiments to test the accuracy 

 of this result ; but they are at present incomplete. 



Incomplete too are experiments on the method of least perceivable 

 difference. 



I find that under certain conditions of illumination and at a given 

 distance from the eye, the amount of white necessary to give a just 

 observable grey ring on a black disc is approximately 0*1 per cent., 

 while under the same conditions the amount of black necessary to 

 give a just observable grey ring on a white disc is approximately 

 1*1 per cent. I believe, though I cannot assert with confidence, that 

 the least perceivable amounts of white on an intervening series of 

 greys are such as to give a geometrical series. But I find this method 

 of least perceivable increments of sensation — lying though it does at 

 the very basis of so much psychophysical work in the past — far from 

 easy of application, since the required increments are small, and since 

 it is difficult to say what is just perceivable. The extremes I have 

 quoted indicate a geometrical series of 240 stages, with a mid-point of 

 nearly 23 per cent, of white — which is nearer the results with the ring- 

 discs than those obtained by continuous shading. 



I have also attempted to check the foregoing luminosity determina- 

 tions by finding the least perceivable amount of coloured paper on a 

 black disc. On the assumption that the amount required is inversely 

 proportional to the luminosity, the results obtained are not very 

 different from those above given. But since I do not regard these 

 results as comparable in accuracy to those obtained on Sir Wm. 

 Abney's method, I do not think it necessary to quote them here. 



I have now described the experimental work on which a purely 

 arbitrary scale of visual sensation in relation to the exciting stimuli is 

 based. It is mainly founded on an appeal to my own eye, which is 

 fairly normal with regard to colour sensation. Unquestionably it 

 depends on the personal equation, But I have now only to determine 

 the luminosities of any coloured surfaces, and I can by reference to the 

 scale construct a disc which shall give without further experimental 

 work an even shading of the one into the other. For example, I had 

 not experimentally determined the mid-point for the shading of light 

 blue and orange. By calculation the mid-point should be 48"9 per 



