[ 507 ] 



LXVI. Note on Stereoscopic Vision. 

 By Professor Helmholtz*. 



THE motives by which we judge the distance of the objects 

 before us with one eye are the following :— 



1. The outlines of the more distant objects are covered by 

 those of the nearer, where they meet. It is this circumstance 

 which causes the difficulty we have in recognizing the fact 

 that the image projected by a convex lens or a concave mirror 

 is nearer to the observer than the lens or the mirror. 



2. The object which throws a projected shadow upon any 

 surface is situated always before that surface. These first 

 two motives are very rarely overpowered by any other ones 

 — as, for instance, by stereoscopic combination. This is easily 

 demonstrated by Dove's pseudoscope, an instrument contain- 

 ing two rectangular prisms, and showing to each eye a reflected 

 image inverted from right to left. It produces an inverted 

 stereoscopic effect if there are no projected shadows, and no 

 outline covered by the outline of a nearer body. 



3. If the head is moved to the right or the left, upwards or 

 downwards, the direction of the eyes remains steady if the 

 object observed is infinitely distant, but is altered the more the 

 nearer the object is. If the head is brought nearer to the 

 object, the convergence of the eyes increases ; if it is moved 

 backwards, the convergence diminishes. 



We may call the peculiarity of the eye producing these 

 phenomena the parallax for motion of the head. At the same 

 time the relative situation of objects of different distance in 

 the field of vision is altered. Distant objects apparently go 

 with the observer, nearer objects in opposite directions. Also 

 this overpowers stereoscopic combination. In Brewster's 

 stereoscope the two images are brought together with a mo- 

 derate convergence of the eyes, but are nearly in the focus 

 of the lenses, so that these project images at an infinite dis- 

 tance. The instrument is adapted to presbyopic vision. But 

 if it is fastened to a table and the head of the observer is moved, 

 the objects appear to be far more distant than the point of 

 convergence. 



I concluded from this and similar observations that the per- 

 ception of the absolute convergence of our eyes is very in- 

 distinct, and that only differences of convergence, related to 

 apparently near or distant objects, produce the stereoscopic 

 effect. But lately I have observed that certain apparent mo- 

 tions of binocular objects may bo observed, which prove that 



* Communicated by the Physical Society, having been read at the 

 Meeting on April 9. 



