234 KECENT PROGRESS OF THE THEORY OF VISION. 



Whether by the irritation of a nerve we produce a 

 muscular movement, a secretion or a sensation depends 

 upon whether we are handling a motor, a glandular, or a 

 sensitive nerve, and not at all upon what means of irrita- 

 tion we may use. It may be an electrical shock, or tearing 

 the nerve, or cutting it through, or moistening it with a 

 solution of salt, or touching it with a hot wire. In the 

 same way (and this great step in advance was due to 

 Johannes Miiller) the land of sensation which will ensue 

 when we irritate a sensitive nerve, whether an impression 

 of light, or of sound, or of feeling, or of smell, or of taste, 

 will be produced, depends entirely upon which sense the 

 excited nerve subserves, and not at all upon the method 

 of excitation we adopt. 



Let us now apply this to the optic nerve, which is the 

 object of our present enquiry. In the first place, we 

 know that no kind of action upon any part of the body 

 except the eye and the nerve which belongs to it, can 

 ever produce the sensation of light. The stories of som- 

 nambulists, which are the only arguments that can be 

 adduced against this belief, we may be allowed to dis- 

 believe. But, on the other hand, it is not light alone 

 which can produce the sensation of light upon the eye, 

 but also any other power which can excite the optic 

 nerve. If the weakest electrical currents are passed 

 through the eye they produce flashes of light. A blow, 

 or even a slight pressure made upon the side of the eye- 

 ball with the finger, makes an impression of light in the 

 darkest room, and, under favourable circumstances, this 

 may become intense. In these cases it is important to 

 remember that there is no objective light produced in 

 the retina, as some of the older physiologists assumed, 

 for the sensation of light may be so strong that a se- 

 cond observer could not fail to see through the pupil the 

 illumination of the retina which would follow, if the 



