ANIMAL AND VEGETABLE 223 



the choroid, that is, the rods and cones, are the elements 

 in which the visual impressions begin. 



46 It thus appears that the real end-organs of vision, 

 the rods and cones, must be in some way connected func- 

 tionally, if not structurally, with the nerve filaments that 

 pass to the optic nerve, and it is evident that these rods 

 and cones, being backwards from the light towards the 

 sclerotic, must receive the light waves after they have 

 passed through the internal layers of the retina, except at 

 the f ovea, where, all the other layers having thinned off, 

 the basal fibres of the cones themselves are directly exposed 

 to the light waves." (Thornton.) 



Before we accept the above conclusions as final it will 

 be well to ponder the matter carefully. 



There are several points which call for consideration. 

 Cones are absent in some animals and rods in others. 

 Light produces changes in pigment, but while the outer 

 limbs of the rods are tinged with a pigment termed 

 " visual purple," derived from the pigment cells of the 

 outer layer of the retina, it can hardly be essential to vision, 

 as it is " absent from the cones of the fovea and entirely 

 wanting in some animals that see well." 



I am not going to suggest that the epithelial pigment 

 cells of the retina contain selenium, but I do suggest that 

 they are composed of or contain some substance which 

 has the property of transforming waves of light into waves 

 of neuro-electricity, possibly by causing enormously rapid 

 alterations of resistance in the sensory nerve- circuits 

 connected functionally, if not structurally, with the cells^ 



" We do not know," says Thornton, " how the undula. 

 tions of light become converted into nervous impulses that 

 give rise to visual sensations." 



The three following diagrams (Figs. 120, 121, 122) may 

 with advantage be considered in their relation to the known 

 optical law that " ordinary light consists of vibrations 



