Prof. "Wheatstone on the Physiology of Vision. 515 



a totally different solid figure is perceived to that seen before the 

 transposition. I called this the converse figure, and showed that 

 it differs from the normal figure in the circumstance, that those 

 points which appear the most distant in the latter, appear the 

 nearest in the former. 



The pictures being, in the first place, presented directly to 

 then- corresponding eyes, as in the refracting stereoscope, and 

 exhibiting therefore the resultant image in its normal relief, the 

 conversion of the relief may be effected in three different ways, — 

 1st, by transposing the pictures from one eye to the other, as 

 mentioned above ; 2ndly, by reflecting the pictures, while they 

 remain presented to the same eye, as in the reflecting stereo- 

 scope ; and 3rdly, by inverting the position of the pictures with- 

 out transposing them. 



The following considerations will explain the cause of the con- 

 version of relief in the preceding cases. 



If two different objects, or parts of an object (fig. 5a), have a 

 greater lateral distance between them on the right-hand picture 

 than that which they have on the left-hand picture, the optic 

 axes must converge more to make the left-hand than to make 

 the right-hand objects coincide, and the left-hand object will 

 appear the neai'est. 



If the pictures be now transposed from one eye to the other 

 (fig. 5 a'), the greatest distance will be between the correspond- 

 ing points of the picture presented to the left eye ; the optic 

 axes must therefore converge less to make the left-hand objects 

 coincide, and the right-hand object will appear the nearest. 



If the pictures, remaining untransposed, be each separately 

 reflected (fig. 5 b), the relative distances of the corresponding 

 objects remain the same to each eye, and the left-hand object 

 will still appear nearest ; but in consequence of the lateral inver- 

 sion of the objects in each picture by reflexion, that which was 

 previously on the left will now be on the right, and therefore 

 the object which before appeared nearest will now appear furthest. 



When the pictures are turned upside down, still remaining 

 untransposed (fig. 5 c), the objects are reversed with respect to 

 the right and left, in the same manner as they are when reflected, 

 and the lateral distances between the objects remaining the same 

 to each eye, precisely the same conversion of relief is produced 

 ;i> in the preceding case, except that the resultant image is in- 

 verted. The diagram (fig. 5) represents all the possible changes 

 of the two binocular pictures; those marked N .show the normal 

 relief, and those marked C the converse relief. 



But it may be asked why, if the rcllcxion or inversion of the 

 binocular pictures of an object gives rise to the mental idea of 

 the converse relief, the same converse relief is not observed when 

 2L2 



