JULIUS BHBINBERG OX STEREOSCOPIC EFFECT AND A 



. i up the right and left halves of the objective, an<l the 

 images, after passing an erecting prism, are directed towards the- 

 right and left eye respectively. Now, strange to say, the un- 

 doubted efficacy of these arrangements depends upon causes 

 which, as the literature of the subject shows, have been very 

 imperfectly understood, and I believe are much misunderstood 

 even to this day. It is generally assumed that the two halves- 

 of the objective act as two separate lenses, viewing the object 

 from two different station points. This, however, is far from 

 the case. They could only do this if their optic axes were 

 inclined towards each other, as in the Greenough microscope. 

 But the single objective has only one optic axis, and all the 

 parts of that objective combine to form an image projection, 

 en from that one direction. Put shortly, the " perspective of 

 inclination " is the same for all parts of the objective. We will 

 make the point clearer by an illustration. 



Suppose a square in the plane of the object-stage to be the 

 object viewed, and to be in sharp focus on the image plane when 

 the whole objective is used. Then the image is also a square 

 when half of the objective L is used by itself, for all the rays 

 proceeding from any part of the object are reunited in the same 

 point of the image, whichever portion of the objective they have 

 traversed. Next, suppose the square to lie in a plane above the 

 one in true focus. Then, as we have seen in the earlier part of 

 our paper, all points of the square are represented in the image 

 plane by diffusion discs, and if we use the right half of the 

 objective only we have all the right halves of the diffusion discs 

 only. All the points are therefore shifted to the right, but the 

 whole image still remains a square. And if the object square 

 lies in a plane below the one in true focus the same holds good, 

 lepting that, the rays reaching the image plane having passed 

 their plane of true focus, the right halves of the diffusion discs- 

 have crossed over to the left, and the whole image is therefore' 

 bodily shifted to the left. 



On this very simple property, that parts of an objective used 

 by themselves bodily shift the image of any area lying in a plane 

 at righi angles to the optic axis of the whole lens, loithout any 

 change of actual shape, depends the stereoscopic effect of the 

 ivances we are considering. If we may talk of the different 

 parts of an objective as " looking at " an object, we might say 



