324 On an Adjustable and Reversible Stereoscope. 



picture is hidden by them from either eye. There is a second 

 sliding cross bar, to which is attached by . hinges a light 

 wooden screen, b b, which in the same experiments is pressed 

 down, as in fig. 6. By lifting this, as in fig. 7, the opening at 

 the middle leaves the left picture risible to the right eye while 

 the right picture is hidden from it ; and vice versa for the left 

 eye. The semi-lenses may now be taken out ; and for them 

 are substituted a pair of prisms, p,p (fig. 7), whose refracting- 

 angle is 12° or 15°. The bases of these are pressed against 

 the springs. The cross bar on which the stereograph rests (a) 

 is pushed back to the end of the longitudinal bar ; and on look- 

 ing through the prisms the binocular image is seen in pseudo- 

 scopic relief. This effect is very striking if the stereograph 

 selected be one in which ordinary perspective is not strong ; 

 that of the full moon has been found best. The cross bar may 

 now be pulled as near as convenient, the visual lines becoming 

 more convergent, but not uncomfortably so. Folding down 

 the Avooden screen (bb), a pair of monocular images of the 

 moon are seen, one on each side of the concave binocular 

 image, but themselves presenting no relief. They appear 

 very perceptibly larger than the binocular image, but slightly 

 smaller than either picture when viewed with normal vision 

 over the stereoscope. Since the prisms have no magnifying- 

 or diminishing-power, this illusion cannot be referred to the 

 effect of refraction, but is explicable only on physiological 

 grounds. 



By removing both prisms and semi-lenses the instrument 

 becomes a direct-vision stereoscope, for use with the unaided 

 eyes. To secure natural perspective, arrange the screens as in 

 fig. 6, but push the cross bar as far off as possible. By now 

 gazing, as if through the stereograph at a very distant point, 

 with the muscles of the eyes* perfectly relaxed, double vision 

 results, and the two interior images are seen to overlap and 

 then coalesce. For most persons this experiment is difficult 

 if the stereographic interval much exceeds the observer's in- 

 terocular distance, and it is not at first easy to secure enough 

 muscular relaxation. Some have found it belter to draw the 

 cross bar up as near as possible, and then push it off after 

 coalescence of images has been attained, the binocular image 

 being at first very dim. 



To secure the pseudoscopic effect by direct vision, arrange 

 the screens as in fig. 7, but push each of the cross bars out as 

 far as possible. Fix the gaze upon the projection (a) at the 

 top of the opening in the wooden screen, and keep it thus 

 fixed while the latter is pulled nearer. When this almost 

 reaches the position b b, in fig. 7, the interior images beyond 



