104 ON COLOUR. Pakt I. 



affected by the accidental colour than others. These, how- 

 ever, are the exceptions, not the rule ; and in ordinary cases, 

 unless the eye is closed, after looking intensely at a coloured 

 object, no such change takes place. Besides, this would be 

 obviated in looking at a complicated polychrome design, 

 where more than one colour is seen at once, and each would 

 of course appear without being confounded with its acci- 

 dental image. Some colours again are modified more or less 

 by the juxtaposition of their immediate neighbours ; and this 

 modification is an additional reason why the accidental image 

 of a colour would not appear, and would prevent any one of 

 the colours being altered by it. What the effect of a colour 

 may become after the eye has looked at it until fatigued is 

 unimportant, the object not being to judge of a colour when 

 the eye is in that condition ; and the retina is little liable to 

 be so affected in looking at a combination of many colours. 

 When the eyes have dwelt so long on one colour that this 

 effect is produced, the remedy is (as Mr. Field very properly 

 recommends) that " they should be gradually passed to its 

 opposite colour, and refreshed amid compound or neutral 

 tints, or washed in the clear light of day." Thus if the eye 

 is fatigued by looking long at red, and this has lost much of 

 its real hue, and ceased to appear of a true red colour, the eye 

 should be relieved by turning to green (its accidental colour) ; 

 when on looking again at the red it sees it, as at first, in its 

 full force. But though the eye might be so deceived by long 

 dwelling on one colour, it would not be so affected by a large 

 number. 



VIII. I have already stated that each colour should be of 

 the same tone, or intensity*; but that there are certain cases 

 in which a lighter tone may be introduced to brighten up a 

 design f; and in some again the quantity of one or more must 

 be regulated by the effect required, as when it is desirable to 

 * T. 75. t PP- 97, 101, and Sect. XVIII. 



