and Coloured Shadows. 436 



yellow, and green, or, when the shadows are not separated, 

 they are fringed with these colours : the green is obscure, 

 but the yellow reflected from the opposite, and the com- 

 plementary violet reflected by the Atmosphere are remark- 

 ably distinct. 



The following experiment, which will lead to the ex- 

 planation of another class of these phenomena, will throw 

 further light upon the subject. If we look through the 

 green glass with one eye, as before, and after observing the 

 complementary colour upon the removal of the glass, open 

 the other eye, we shall find, that the paper seen with both 

 eyes approaches nearer to its proper colour, than when seen 

 with the impressed eye only ; if we then close the impressed 

 eye, and view the paper with the eye which was previously 

 closed, we shall observe that its colour is a pale green ; 

 and by opening and closing the different eyes in succession, 

 it will b* discovered, that the alternations produce colours 

 complementary to each other. In this experiment, the 

 standard for white is formed by the mixed sensation pro- 

 duced by the impressed and the unimpressed eye, as may 

 be readily perceived by viewing the paper with both eyes 

 open; and, of course, the paper is seen in complementary 

 colours with reference to this standard. 



If we impress the eye, by looking through a crimson 

 instead of a green glass, the effect will be reversed, and 

 the accidental colour will be green. If, in this state of the 

 eye, we hold a narrow strip of white paper vertically, 

 about a foot from the eye, and fix both eyes upon an object 

 at some distance beyond it, the impressed eye will see the 

 paper of a pink colour, while to the other eye it will appear 

 green. If the two images are made to overlap each other, 

 the overlapping parts will appear white. This experiment, 

 which varies only in form from that which precedes it, is 

 similar to Mr. Smith's, as stated in Brewster's Optics, 

 page 310, with only this difference, that in Mr. Smith's 

 experiment, one of the eyes is strongly impressed by the 

 light of a candle placed near it, from which the other eye 

 is protected, instead of being impressed by the light trans- 

 mitted by the crimson glass. 



It appears from this investigation, that these phenomena 

 may be divided into three classes; in the first, the eye 



2 F 2 



