TRANSACTIONS OF THE SECTIONS. 17 



where n must be found by the method of trial and error for each medium in parti- 

 cular, and e n , a» are certain known functions of wand of the observed indices. 



Thus the formula contains three arbitrary constants, which must be determined 

 from the results of observation. 



When these constants are properly determined for any medium, the formula, even 

 in the case of the most highly dispersive media which have been observed, is found 

 to represent very accurately the observations, the utmost error being only a few units 

 in the fourth place of decimals. 



Experiments and Conclusions on Binocular Vision. 

 By Professor William B. Rogers, Boston, U.S. 



The following experiments, intended to test the theory of the successive com- 

 bination of corresponding points in stereoscopic vision, are I believe in part new, 

 and are in part moditied repetitions of experiments already described by Professor 

 Wheatstone and Professor Dove. 



1. Let two slightly inclined luminous lines, formed by narrow slits in a strip of 

 black card-board, be combined into a perspective line, either with or without a 

 stereoscope. Looking at this for a few seconds, so as to induce the reverse ocular 

 spectrum, and then directing the eyes towards the opposite wall of the apartment, 

 a single spectrum will be observed having the attitude and relief of the original 

 binocular resultant. 



As a strong illumination of the lines is necessary to bring out the full effect, the 

 card-board should be held between the eyes and some brilliantly white surface, as 

 the globe of a solar lamp or a strongly illuminated cloud, care being taken to pre- 

 vent the entrance of extraneous light. 



2. Using the same arrangement, let the luminous lines be regarded in succession 

 each by the corresponding eye, the other eye being shaded so that no direct bino- 

 cular combination can be formed. On looking towards the wall, it will be seen that 

 the two subjective images unite to form a single spectral line, having the same relief <ix 

 if the lines had been directly combined by simultaneous vision, either with or without 

 a stereoscope. 



While the perspective image continues distinctly visible, let either eye be closed, 

 the other being still directed towards the wall. The image will instantly lose 

 its relief and take its position on the plane of the wall as an inclined line, corre- 

 sponding to the subjective image in the eye that has remained open. When 

 the subjective impressions have been sufficiently strong, it is easy to alternate 

 these effects, by projecting first the picture proper to the right eye, then that of 

 the left, on the plane of the wall, with their respective contrary inclinations ; and 

 then looking with both eyes, we see the resultant image start forth in its perspec- 

 tive attitude. 



It is hardly necessary to say that to obtain these effects satisfactorily the lines 

 should be very strongly illuminated, and the observer should have some practice in 

 experiments on subjective vision. Under these conditions I have found the results 

 to be perfectly certain and uniform. 



In these experiments, according to the theory of Sir David Brewster, the result- 

 ant spectrum, instead of being a single line in a perspective position, ought to pre- 

 sent the form of two lines inclined or crossing, situated in the plane of the wall 

 without projection or relief. The conditions of the experiments are such as to 

 exclude all opportunity of a shifting of the image on the retina, and such shifting is 

 obviously essential to the successive combination of pairs of points required by the 

 theory in the production of perspective effect 



In reference to the first experiment, it might perhaps be maintained that, as the 

 perspeetiveness of the original resultant on which the eyes were converged formed 

 part of the direct perception in first combining the lines, it would be likely through 

 association to be included also in the spectral or subjective perception. But this 

 consideration, which at best appears to me of little weight, is entirely inapplicable 

 to the conditions of the second experiment. For here the eyes are in the first 

 place impressed in succession with their respective images, and are not allowed to 

 see the resultant ; and yet when they are together directed to the wall, the percep- 

 tion of the single perspective resultant is at once originated. 



1860. 2 



