466 



Mr. C. Lloyd Morgan. 



hypothesis is, however, needed. The fact is a necessary corollary from 

 the nature of the curve which brings stimulus and sensation into rela- 

 tion with each other. This may best be illustrated by taking a some- 

 what extreme case, and dealing only with the value of mid-grey. Let 

 us suppose that 12 per cent, of white stimulus gives, under a given 

 illumination, a sensation of approximately 50 per cent, on the arbitrary 

 scale — that is to say, a sensation half-way between black and white. 

 And let us further suppose that the illumination is reduced to one-half. 

 What will be the effects in sensation 1 It might at first sight be 

 supposed that since the full- white was reduced by half, and the 12 per 

 cent, for mid-grey also reduced by half, the sensations underwent a 

 similar reduction. But further consideration shows that the two scales 

 (that for stimulus and that for sensation) being unequally reduced, the 

 position of the mid-point for sensation is necessarily shifted. 



Fig-. 2. 



V 



a 







4 







6 







a 















*.? 











80 



















































60 





/Orii 





mid-p 



lint:. 















7 ° 

 Reduc 



ed mi 



d-pob 













40 









































20 























\o 



















X 



Beference to fig. 2 shows that 50 per cent, stimulus affords 83 per 

 cent, sensation, and that 6 per cent, stimulus affords 36 per cent, sensa- 

 tion. But 36 per cent, sensation is not the mid-point between per 

 cent, and 83 per cent. The mid-sensation will be 41 '5 per cent., and 

 this requires 8 per cent, of stimulus. Hence, for the given reduction 

 of illumination an additional 2 per cent, of stimulus is required to 



