EYE AS A CAMERA OBSCURA. 343 



condition in which that of a cat, or dog, or ox is, when subdued light causes 

 the iris to expand, and allows the reflecting tapetum to come into play, so that 

 the considerations which I have to urge apply to the mammal as much as to the 

 fish, provided they are taken with pupils equally dilated ; but as the tapetum in 

 the shark is very large, very brilliant, and always in action, I shall restrict myself 

 for the present to it. 



The light, which penetrates to the bottom of a shark's eye, will, in part, be re- 

 flected from the retina (a phenomenon which for the present I disregard), in part 

 traverse it, and reach the tapetum, where a portion will be lost by absorption and 

 irregular reflection or dispersion, and (what alone concerns us here) in part un- 

 dergo direct reflection, return through the retina, and escape by the pupil. This 

 returned light will impress the retina in traversing it, and illuminate external 

 objects on leaving the eye. 



The first question, then, is, " How will this light impress the retina ?" Ac- 

 cording to J. Muller and W. Mackenzie, as we have already seen, only inju- 

 riously, so far as freedom from the sensation of dazzling, or distinctness of 

 visual perception, are concerned ; according to Todd and Bowman "probably" by 

 " increasing the visual power, particularly when the quantity of light admitted 

 into the eye is small." * I have urged elsewhere that " what is equivalent to 

 two rays of light falling upon the retina will produce two impressions. We send 

 a capillary sunbeam through the retina in one direction, and instantly return it 

 through that membrane, a little diminished in intensity, in the opposite direction ; 

 if it determined a sensation in its first passage, what is there to prevent its doing 

 so in its second ? If, for simplicity's sake, we suppose exactly the same points 

 of the retina to be traversed by the incident and the reflected ray, then (unless 

 the luminous intensity of the incident ray was so great as by its passage to ex- 

 haust the sensibility of the retina), the reflected ray will repeat somewhat less 

 powerfully the impression made by the incident one. The difference will be as 

 great as there is between a sound and its echo, but not greater. 



" On this view of matters, the tapetum, especially in twilight, will serve the 

 important purpose of making every perceived ray of light tell twice upon the 

 retina, so that the sensation it produces will either be increased in distinctness or 

 in duration, and probably in both." f 



I will not deny that we are not entitled at once to infer that because a molecular 

 change (modulation, vibration, polarization?) transmitted through a special struc- 

 ture in one direction produces a peculiar sensation, it will certainly produce the 

 same sensation on being transmitted through that structure in the opposite direc- 

 tion ; but there are strong analogies in favour of such a view, and it is entitled to 

 be regarded as a likely hypothesis. 



* Physiology of Man, chap, xvii., p. 23. \ Researches on Colour-Blindness, p. 99. 



VOL. XXI. PART. II. 4 Z 



