268 BKCENT PROGRESS OF THE THEORY OF VISION. 



that differences in sensation which are easily appre- 

 hended appear to us greater than those less obvious. 

 Differences of colour which are actually before our eyes 

 are more easily apprehended than those which we only 

 keep in memory, and contrasts between objects which are 

 close to one another in the field of vision are more easily 

 recognised than when they are at a distance. All this 

 contributes to the effect. Indeed, there are a number of 

 subordinate circumstances affecting the result which it 

 would be very interesting to follow out in detail, for they 

 throw great light upon the way in which we judge of 

 local colour : but we must not pursue the inquiry further 

 here. I will only remark that all these effects of contrast 

 are not less interesting for the scientific painter than for 

 the physiologist, since he must often exaggerate the 

 natural phenomena of contrast, in order to produce the 

 impression of greater varieties of light and greater fulness 

 of colour than can be actually produced by artificial 

 pigments. 



Here we must leave the theory of the Sensations of 

 Sight. This part of our inqiiiry has shown us that tl>e 

 qualities of these sensations can only be regarded as signs 

 of certain different qualities, which belong sometimes to 

 light itself, sometimes to the bodies it illuminates, but 

 that there is not a single actual quality of the objects 

 seen which precisely corresponds to our sensations of sight. 

 Nay, we have seen that, even regarded as signs of real 

 phenomena in the outer world, they do not possess the 

 one essential requisite of a complete system of signs — 

 namely, constancy — with anything like completeness ; so 

 that all that we can say of our sensations of sight is, 

 that ' under similar conditions, the qualities of this sen- 

 sation appear in the same way for the same objects.' 



And yet, in spite of all this imperfection, we have also 



