Impulses of Light upon the Retina. 243 



younger eyes than mine, and who shall have the courage to 

 repeat the experiments with the direct light of the sun, and 

 with a disc having narrow slits, and revolving upon a fixed 

 axis so as to have its velocity uniform, will be able to obtain 

 an accurate representation of the pattern in question. 



Within and around the central lozenge, is seen with great 

 distinctness the dark-lined network pattern already men- 

 tioned, and apparently unconnected with the larger pattern. 

 As the spaces, however, or patches, which compose the larger 

 pattern diminish in size towards the centre, it is possible that 

 the dark-lined network, with dark specks in the centre of the 

 figures, and having all the regularity of a geometrical figure 

 drawn with ink, may be the central part of the larger pat- 

 terns seen more distinctly by direct vision ; but I cannot ad- 

 mit this notion, because under favourable circumstances a 

 similar dark-lined pattern, with extremely small spaces be- 

 tween the meshes, appears throughout the rest of the field of 

 view, especially in the external part of it where it first begins 

 to show itself. 



The colours which appear in the principal pattern are 

 chiefly bluish purple, and its complementary colour lemon-yel- 

 low, but as the former increases in depth or approaches to 

 blackness, the latter becomes more white. These different 

 colours sometimes appear in the different patches of the pat- 

 tern, and sometimes they appear in succession over a consi- 

 derable part of the field. They are, however, most distinctly 

 seen in the central lozenge, the inner part being sometimes 

 purple and then yellow, while the outer part of it is first yel- 

 low and then purple. The central lozenge is sometimes mark- 

 ed out by whitish, and sometimes by greenish light, and 

 I have frequently seen in its centre a blush-red of a very rare 

 tint. The succession of colours in the lozenge is very beauti- 

 ful, each colour advancing to the centre, replacing that which 

 preceded it, and then disappearing. 



The cause of the colours themselves is obvious. The action 

 of white light on the retina renders it first insensible to the 

 red rays*, and consequently a white object or ground :ip- 

 pears bluish purple or blue in solar light, and green in candle 

 li^ht, the colour varying with the intensity of the exciting 

 light, and with the distance of the image of the white object 

 from the excited point. The other colour which appears in 

 the preceding experiments is a faint lemon-yellow, which is the 

 Complementary colour of the bluish purple. It deserves also 

 to be noticed that lliuse colours are the very same as those 



* Sec Lond, and Edinb. Phil. Mag., vol. iii. p. 109. 

 2 I 2 



