20-1 Transactions of the Society. 



binocular system is that the semicircles in question should be 

 depicted according to the diagram 0, and for jiseudoscopic effect, 

 according to P; all other circumstances, except this one, heing 

 immaterial. 



An equivalent expression of the same rule (implying, however, 

 the tacit assumption that the observer's eyes are kept concentrically 

 to the axis of the oculars) would be : Orthoscopic vision is always 

 obtained, when the right half of the right pupil and the left half of 

 the left pupil ouly are employed — pseudoscopic vision in the 

 opposite case ; and whenever that condition is fulfilled, the same 

 kinds of visual impression will always result. It is quite indifferent, 

 whether the effect is obtained with crossing or non-crossing rays, 

 whether the image be erect, or inverted, or semi-inverted, and 

 whatever elements (lenses, prisms, mirrors, &c.) may be components 

 of the optical arrangement. 



This proposition — which is an outcome of a purely theo- 

 retical research into the circumstances of solid vision in the 

 Microscope — may be verified by various experiments. The author 

 has applied it, in the " stereoscopic ocular " described in the paper 

 quoted above, for obtaining at pleasure true orthoscopic or 

 pseudoscopic vision with a binocular arrangement which in itself is 

 non-stereoscopic. 



It has, however, been objected to this method in most decided 

 termSjt that orthoscopic vision is not obtained in binoculars with 

 non-erecting eye-pieces, unless the axes of the two halves of the 

 admitted cone cross each other ; and this view has been assumed 

 to be a sufficient basis for deciding a priori, without any experi- 

 mental trial, that the pretension of stereoscopic action of such an 

 arrangement as the author has described must arise from con- 

 founding imagination eftects with true stereoscopic impressions. 

 It will therefore be useful to give the demonstration in detail of 

 the principle pointed out above, which I did not think necessary 

 in the original paper. 



My consideration is based on a few suppositions, which are 

 generally adopted. These are : + — 



Stereoscopic effect (both in ordinary and in microscopic 

 vision) depends on difference of perspective in the two eyes of 

 the observer, i. e. on different parallactic displacement of con- 

 secutive layers of an object in their projections to the two 

 eyes. 



True stereoscopic (orthoscopic) effect is always obtamed 

 when the rigid eye sees the nearer planes of the object 



t This Journal, iii. (1880) p. 1088. 



X As to these premises, I refer the reader to Dr. Carpenter, ' The Micro- 

 scope,' 5th ed. § 28 et seq. 



