312 Mr C.V. Burton, Experiments on Colour-Perception; [Nov. 26, 
violet rays producing little if any effect. Dewar and M‘Kendrick 
also find that the visible rays are those which affect the current 
obtained from the eye; and in each case it is the yellow and 
green rays which are the most effective. 
They appear to ascribe the variations of current which they 
obtained to E. M. F.’s arising from a chemical action which is 
produced in the retina by the impact of light. They further 
suppose that when the eye is in darkness there is a certain current 
in the optic nerve, which changes in strength when light falling 
on the retina produces an additional EM. F. If, however, as 
would appear from the experiments of Helmholtz and others, 
the currents in the optic nerve really determine the sensations 
of vision, it seems probable that the current corresponding to 
any position on the retina travels to and from the brain by 
nearly coincident paths; (for if the currents from different parts 
of the retina followed a common path for any part of their 
course they would mutually interfere with one another). In 
that case, the E. M. F. along the optic nerve will not produce 
any current in the circuit, and the currents producing visual 
impressions will be entirely due to E. M. F.’s set up by the action 
of light on the retina. 
Now we have seen (2, b) that in the case of selenium the 
photo-voltaic £. M. F. follows changes of illumination almost 
instantaneously, just as visual impressions do; and again light 
falling on selenium produces fatigue, just as it does on the 
retina. In both cases too the fatigue is selective, though in the 
eye this is accounted for by the independence of the colour- 
sensations. 
As regards persistence of impressions, we know that if light 
falling on selenium produces a current of sufficient strength, this 
current will set up a polarizing E.M.F., so that when the light 
is removed there is a reversed current. From the memoir of 
Adams and Day, it does not appear that they observe any 
further persistent E.M.F. due to the action of light. If we 
suppose the effect of light on the retina to be of a photo-chemical 
nature, it is difficult to see how sensitiveness is maintained. 
In order always to keep fresh liquid in contact with the retina, 
there would have to be enormously rapid diffusion; and then 
too (as appears from the laws of colour-perception) there must 
be three sensitive compounds in the eye, each affected by a 
different range of colour, and each when acted on by light 
affecting a different portion of the retina. If the sensitive 
compounds are supposed to be contained in the retina itself, 
it seems impossible that they should be renewed with sufficient 
rapidity to maintain sensitiveness. 
Following the varying effects which Dewar and M*Kendrick 
obtained by the action of light on the eyes of different animals, 
I find they can all be explained on the supposition that the 
