258 ON LIGHT. 



(42.) The three primary colours assumed in the 

 above figure are red, green, and blue, each in its highest 

 degree of purity and imdilufio??, for it will be readily 

 apprehended that while the admixture of any one, in 

 however small a proportion, will produce a rich though 

 a mixed tint, that of both the others tends to dilution. 

 The only three colours which answer all the experi- 

 mental conditions, are these three. This may seem 

 contrary to the experience of the artist, who from his 

 habitual practice in mixing the colours he uses (all of 

 them without exception compound tints), would name 

 yellow, in place of green, as the intermediate primary. 

 The reason is obvious. In all the yellows which he 

 uses there is a large admixture of red with green, and 

 in all his blues more or less green. When mixed, then, 

 there is sure to be a preponderance of green, while the 

 red goes to neutralize a portion of the other two, and 

 so to dilute the outstanding green. On the other hand, 

 the direct mixture of the prismatic yellow mid blue, in 

 whatever proportions, can no-how be made to produce green, 

 as Professor Maxwell's, M. Helmholz's, and my own 

 experiments* have distinctly proved ; while that of the 

 prismatic green and red does produce yellow. This 

 will be better understood when we come to speak of 

 the absorption of coloured light. 



(43.) Since at each point of a compound spectrum 

 so constituted, all the three primary elements, in what- 

 ever proportion mixed, have one and the same degree 

 of reh-angibility, it is evident that the compound tint 

 * See "Notices of the Royal Society," vol. x. p. 52. 



