310 Mr. W. Le Conte Stevens on 



RL be parallel to the wall, it follows that the phantom 

 surface must be a plane S S' parallel to the wall. This expla- 

 nation fails to account for the apparent curvature S" S"' ; 

 and on actually trying the experiment it will be found that 

 the convexity is very slight unless the optic convergence be 

 strong. 



2. The Binocular Eye. 

 In examining this phenomenon it is indispensable that the 

 observer distinguish between subjective effects and objective 

 realities, between a sensation and its exciting cause. In 

 seeing with two eyes, the sensation is the same as if both 

 were fused into a single binocular eye. midway between them, 

 whose optic centre is the point of origin in all estimates of 

 direction and distance. The two visual lines are subjectively 

 combined into a single line, extending from the yellow spot 

 of the binocular eye as far as the external point to which the 

 observer mentally refers the combined retinal picture. There 

 can hence be no perceived intersection of visual lines as 

 assumed by Brewster. The recognition of optic convergence 

 or divergence is through the muscular sense ; and the locali- 

 zation of objects in the field of view is the interpretation of a 

 complex sensation, not a simple geometric determination. I 

 have recently devised a refracting stereoscope *, by means of 

 which the same stereograph, within a few seconds of time, 

 may be viewed successively by optic divergence, with natural 

 relief, and by slightly greater convergence with reversion of 

 relief. The binocular image is seen either alone or accom- 

 panied by a pair of monocular images, as may be preferred, 

 and varies in apparent distance and size according to the 

 degree of muscular strain attendant upon binocular vision in 

 each case. The appearance of monocular images can be easily 

 secured with the ordinary lenticular stereoscope, by removing 

 the longitudinal screen while the relation between the two 

 visual lines is kept unchanged. The subjective effect is that 

 the binocular image, in full relief, remains apparently single 

 and directly in front of the binocular eye, while on each side 

 of it a separate monocular image is perceived by indirect 

 vision. This is true for optic divergence as well as conver- 

 gence; and the recognition of the subjective union of the 

 eyes must underlie any attempt to explain the phenomena of 

 binocular vision. 



3. Gradation in Retinal Fusion. 

 However mistaken Brewster may have been in his theory 



* American Journal of Science, March 1882. 



