324 Dr. G. J. Stoney on the Limits of Vision i 



vividly what seemed to be a tiny horseman galloping violently 

 from right to left, or from left to right, as the case might be. 

 All the movements of the horse were reproduced, the dashing 

 about of the sabre-tasche, the coloured uniform, the move- 

 ments of the horseman. It cannot have been in the retina 

 that this revival took place. It must have been in a much 

 more deep-seated part of the brain. 



It would, I think, be of very great interest to ascertain 

 from the inhabitants of a blind asylum, whether those who 

 have recently had their retinas extirpated, or rendered func- 

 tionless, continue to dream of scenery, so long as the memory 

 of visual perceptions is recent. I should expect they would, 

 as the structures which they have lost do not seem to be con- 

 cerned in either memory or dreaming. 



From a review of all the evidence it appears clear that the 

 retinal image is only one of the stepping-stones in a rather 

 long progress from the object in nature to the event in the 

 brain which is the direct adjunct of visual perception. Why, 

 then, it may be asked, is an image necessary ? Why is it 

 never absent ? Why is not something quite different some- 

 times substituted for it ? The following is, I think, a sufficient 

 answer. There must be some difference in the events occurring 

 in the occipital lobe in order that two points of an object may 

 be seen distinct from one another. To bring this about either 

 a different nervelet must have been acted upon in the organ 

 of sight, or the same nervelet must have been differently 

 acted on. In the case that actually occurs, it would appear 

 that a different nervelet is acted upon when the points of the 

 object are sufficiently separated to be seen as two, and that a 

 difference of action on the same nervelet is reserved for ex- 

 hibiting to us variations of brightness and colour, but not of 

 position. Now this can manifestly be effected by distributing 

 the points of an image of the object over an apparatus such as 

 the layer of rods and cones, consisting] of closely packed 

 individuals, each of which is capable of acting on its own 

 nervelet ; or through an intermediate apparatus, which con- 

 sists of channels for transmitting light as numerous as the 

 rods and cones, each of which conducts the light from a spe- 

 cific point of the image to its own rod or cone, which latter 

 may, in this case, be situated at a distance from the place 

 where the image is formed. The first of these is the arrange- 

 ment which we find in our own eyes ; the other seems to be 

 that which we find in the compound eyes of insects. Now it 

 is doubtful whether any other machinery for bringing about 

 the result than one or other of these two can be devised. 

 These, at all events, are the ways in which nature attains the 

 end ; so that neither man nor nature seems to have found out 



