THE SENSATION OF SIGHT. 251 



sensation. Just as the difference of sensation of light and 

 warmth depends demonstrably upon whether the rays of 

 the sun fall upon nerves of sight or nerves of feeling, so 

 it is supposed in Young's hypothesis that the difference 

 of sensation of colours depends simply upon whether one 

 or the other kind of nervous fibres are more strongly 

 affected. When all three kinds are equally excited, the 

 result is the sensation of white light. 



The phenomena that occur in red-blindness must be 

 referred to a condition in which the one kind of nerves, 

 which are sensitive to red rays, are incapable of excita- 

 tion. It is possible that this class of fibres are wanting, 

 or at least very sparingly distributed, along the edge of 

 the retina, even in the normal human e3^e. 



It must be confessed that both in men and in quadru- 

 peds we have at present no anatomical basis for this 

 theory of colours ; but Max Schultze has discovered a struc- 

 ture in birds and reptiles which manifestly corresponds 

 with what we should expect to find. In the eyes of many 

 of this group of animals there are found among the rods 

 of the retina a number which contain a red drop of oil in 

 their anterior end, that namely which is turned towards 

 the light ; while other rods contain a yellow drop, and 

 others none at all. Now there can be no doubt that red 

 light will reach the rods with a red drop much better 

 than light of any other colour, while yellow and green 

 light, on the contrary, will find easiest entrance to the 

 rods with the yellow drop. Blue light would be shut off 

 almost completely from both, but would affect the colour- 

 less rods all the more effectually. We may therefore with 

 great probability regard these rods as the terminal organs 

 of those nervous fibres which respectively convey impres- 

 sions of red, of yellow, and of blue light. 



I have myself subsequently found a similar hypothesis 

 very convenient and well fitted to explain in a most 

 12 



