71 



an appearance of relief as in nature. The following is the manner in 

 which I adapt to this purpose the refracting form of the stereoscope. 



Having procured eight stereoscopic pictures of an object of a 

 steam-engine for example in the successive positions it assumes in 

 completing a revolution, I affix them, in the order in which they were 

 taken, to an octagonal drum, which revolves on a horizontal axis 

 beneath an ordinary lenticular stereoscope and brings them one 

 after another into view. Immediately beneath the lenses, and with 

 its axis situated half an inch from the plane of sight, is fixed a solid 

 cylinder, 4 inches in diameter, capable of being moved freely on its 

 axis. This cylinder, which is called the eye-cylinder, is pierced 

 throughout its entire length (if we except a diaphragm in the centre 

 inserted for obvious reasons) by two apertures, of such a shape, and 

 so situated relatively to each other, that a transverse section of the 

 cylinder shows them as cones, with their apices pointing in opposite 

 directions, and with their axes parallel to, and distant half an inch 

 from, the diameter of the cylinder. Attached to the axis of the eye- 

 cylinder is a pulley, exactly one-fourth the size of a similar pulley 

 affixed to the axis of the picture-drum, with which it is connected 

 by means of an endless band. The eye-cylinder thus making four 

 revolutions to one of the picture-drum, it is evident that the axes of 

 its apertures will respectively coincide with the plane of sight four 

 times in one complete revolution of the instrument, and that, 

 consequently, vision will be permitted eight times, or once for each 

 picture. 



The cylinder is so placed that at the time of vision the large ends 

 of the apertures are next the eyes, the eifect of which is that when 

 the small ends pass the eyes, the axes of the apertures, by reason of 

 their eccentricity, do not coincide with the plane of sight, and vision 

 is therefore impossible. If, however, the position of the cylinder be 

 reversed end for end, vision will be possible only when the small ends 

 are next the eyes, and the angle of the aperture will be found to sub- 

 tend exactly the pencil of rays coming from a picture, which is so 

 placed as to be bisected at right angles by the plane of sight. Hence 

 it follows that, the former arrangement of the cylinder being re- 

 verted to, the observer looking along the upper side of the aperture 

 will see a narrow strip extending along the top of the picture ; then, 

 moving the cylinder on and looking along the lower side of the 



G 2 



