52 E. M. NELSON ON BINOCULARS. 



the left-hand half of the objective to the right eye direct, the 

 right half being deflected by a prism to the left eye. It would, 

 were it not for the transposing effect of the eye-pieces, give 

 pseudostereoscopic vision. This is by far the most practical 

 a,nd best of all orthostereoscopic binoculars. There is plenty of 

 light, the images in both tubes, if the prism is well made, are 

 excellent, and the great ease with which the instrument can be 

 converted to a monocular, together with its simplicity of con- 

 struction, will always cause it to hold the first place. 



The Abbe is an indifferent instrument ; it yields a double 

 image in the side tube, and while contracted tubes are found 

 more agreeable to use than the semicircular eye-caps, neither 

 can be called pleasant to work with. 



The Beck is a very good plan for a simple microscope. It is 

 better to make it with a single prism on either side, instead of 

 four prisms, in which case it must be adjusted for the observer's 

 own use, as it will not be possible to alter the adjustment to 

 suit different distances of eye centres. 



The Cherubin d'Orleans yields excellent results, but the con- 

 struction is only possible with low powers ; nevertheless, the 

 stereoscopic effect is more perfect than in any other form. 



Holmes can hardly be called a practical construction. 



ISTachet II. — I have not tried this form, but should think that 

 if the two prisms were joined into one it would be an improve- 

 ment, and would make a very efficient orthostereoscope. 



Powell is a very good non- stereoscopic binocular. 



Stephenson is the best erecting binocular. The prisms re- 

 quire to be very carefully worked. 



Tolles is the best binocular eye-piece, but the prisms require 

 most careful working. 



The Wenham we have already discussed. Mine, with a prism 

 by Powell, gives excellent results, e.q.^ the secondaries of a tri- 

 ceratiumin balsam are shown in both tubes by a | of SS*^ and a 

 power of 80 diameters. 



An examination of the table shows the identity of Abbe's 

 theory, as far as it goes, with that of Carpenter. Abbe's 

 theory is shown by the cC being in the middle for orthostereo- 

 scopism, and the mM in the middle for pseudostereoscopism. 



Abbe's conditions of Q D and D Q do not indicate the pre- 

 sence or absence of a " cross-over," which my table does. 



