542 VISION 



B. SUCCESSIVE COLOR INDUCTION 



When one looks fixedly for some seconds at a red object on a white ground 

 (Fig. 236), and then turns the eye toward the white ground, he sees on the 

 latter a distinct after-image which reproduces the object exactly in all re- 

 spects but one instead of being red it is greenish blue. If the object were 

 greenish blue the after-image would be red. 



For every color tone in the spectrum there is another, which in exactly 

 the same way as in this example evokes and in turn is evoked by the first 

 as an after-image. Pure green, however, forms an apparent exception to the 

 rule. Its after-image is purple but purple is a color 

 which does not occur in the spectrum ; its relations thereto 

 will be mentioned presently. Other pairs of colors which 

 are related to each other in this way are : orange and blue, 

 golden yellow and blue, yellow and indigo, etc. 



These phenomena show that objective light can be 

 subjectively destroyed, also that a certain intimate rela- 

 tion exists between the different colors. In order to inquire further into these 

 facts which necessarily form the foundation stones for every theory of color, 

 we must see what results from the mixture of colors. 



C. COLOR MIXTURE 



By a mixture of two or more colors we mean that color which is experi- 

 enced when a given point on the retina is struck simultaneously by rays of 

 different wave lengths. Every color mixture is therefore a summation effect 

 of different light rays. 



The best and for many purposes the only possible mode of procedure is 

 to mix pure spectral colors. This may be done by complicated apparatus 

 which enables one to isolate two spectral rays of different wave lengths and 

 to throw them on the same spot of a screen where the mixture can be com- 

 pared with a reference color. 



In this way the different rays are brought into the eye at the same time. 

 But the experiment can be so arranged also that the rays to be tested will 

 fall successively on the same spot of the retina. Then if the sequence is rapid 

 enough a mixture of the two will take place just as in the experiment with 

 white and black sectors (page 539). 



Very convenient for this purpose are Maxwell's disks. They are circular, 

 colored disks having a radial slit, so that two or more of them can be over- 

 lapped and varying portions of each be exposed to view. If the complex disk 

 thus composed of two or more colored sectors is rotated rapidly by mer.ns of a 

 clockwork, the resulting mixture will depend upon the saturation of the indi- 

 vidual colors employed and the relative sizes of the different sectors. 



The results of color mixture which interest us most are those obtained with 

 the above-mentioned pairs of color red and greenish blue, yellow and indigo, 

 etc. Experiment has shown that each of these pairs when its components 

 are mixed at a certain relative intensity, produces the sensation of white or 

 gray (which is but a feebly illuminated white). Since each of these colors 



