940 A MANUAL OF PHYSIOLOGY 



all these wave-lengths fall together, in the proportion in which 

 they are present in sunlight, upon the same part of the retina, 

 the resultant physiological effect is very different ; we are no 

 longer able to distinguish red, blue, green, etc. ; we receive 

 the single sensation of white light. The sensation is a simple 

 one ; in consciousness we have no hint that it has a multiple 

 physical cause. 



But we find further that it is not necessary for the sensation 

 of white light that waves of every length present in the solar 

 spectrum should be mixed. If rays of wave-lengths 675 JU/A 

 (which acting alone produce the sensation of red) be mixed in 

 certain proportions i.e., be allowed to fall on the same part of 

 the retina with rays of wave-length 496 ////, (which give the 

 sensation of bluish-green), the resultant sensation is also that 

 of white light. And an indefinite number of sets can be com- 

 bined, two and two, so as to give the same sensation of white. 

 Such colours are called complementary. The following are 

 pairs of complementary colours : 



Red and bluish-green. Yellow and ultramarine-blue. 



Orange and cyan-blue.* Greenish- yellow and violet. 



The green of the spectrum has no simple complementary 

 colour ; purple, a colour not present in the spectrum, but obtained 

 by mixing light from the two spectral extremes i.e., by mixing 

 red and violet may be considered complementary to it. Sup- 

 pose now that one of a pair of complementary colours is added 

 to the other in greater intensity than is required to give white, 

 the resultant sensation is a colour which has a certain amount 

 of resemblance both to white and to the colour present in excess. 

 Thus, if the two colours are orange and blue, and the blue is 

 present in greater intensity than is necessary to give white, the 

 resultant colour is a whitish or pale blue, or, to use the technical 

 phrase, an unsaturated blue. The more nearly the intensity of 

 the blue rays in the mixed light approaches the proportion 

 necessary to give white, the less saturated is the resultant colour ; 

 the greater the excess of blue, the more nearly does the resultant 

 sensation approach that of the saturated blue of the spectrum. 

 But any non-saturated spectral colour produced by the mixture 

 of two complementary colours may be equally well produced by 

 the mixture of the corresponding spectral colour with a certain 

 quantity of ordinary white light. And it is found that when 

 two spectral colours which are not complementary are mixed 

 together the resultant is not white, but a colour which may be 

 matched by some spectral colour lying between the two (or by 

 purple), either without addition or plus a larger or smaller 

 * Cyan-blue is a greenish-blue. 



