MODEBJS OPTICS AND PAINTING. 4 i 9 



pose, the superposition of the blue and yellow tints furnished by po- 

 larized light has lately sometimes been used, but, though the result 

 obtained is quite correct, it may be objected that this experiment was 

 perfectly well known to Sir David Brewster, the great modern de- 

 fender of the old theory, as well as to all the physicists who were his 

 followers ; and this knowledge does not seem in the least, for more 

 than a quarter of a century, to have weakened their confidence. Nor 



Fig. 2. 



L 



YELLOW 



would it be perfectly satisfactory if I should bring about the union of 

 blue and yellow light by the method of revolving colored disks, as is 

 so often done ; for, when we come to analyze this latter plan, we find 

 that it consists, essentially, in presenting yellow and blue light to the 

 eye, not simultaneously, but by a distinct succession of alternate acts. 

 It is true that in this convenient mode of experimenting the results 

 are the same as in that of simultaneous presentation, but just this 

 point again would require proof, and, in a fundamental experiment like 

 the present, ought not to be passed over in silence. To avoid these 

 difficulties, I have contrived another plan, which will admit of our 

 readily grasping the whole process, and inspecting its quite simple de- 

 tails. We have now upon the screen two large squares of blue light, 

 and near them are two corresponding squares of yellow light (Fig. 3), 

 and I can readily contrive matters so that the portion of the screen 

 which is illuminated by one of the yellow squares shall also receive 

 the light of a blue square. This we now have, and the result, as you 

 see, is not the production of green light, or of light whose hue at all 



