344 ON LIGHT. 



can be drawn across it, and to which it is symmetrical, 

 as shown in the preceding figure). 



(122.) If such a rhomboid be laid down on an ink- 

 spot on white paper, or, still better, on a small pinhole 

 in a plate of metal, and held up to the light, the ink-spot, 

 or the dot of light, will appear through it doubled : the 

 two images being separated, in the direc- 

 tion of the shorter diagonal of the face 

 through which they are seen, by an inter- 

 '^* " val of about one-ninth part of the thick- 

 ness seen through. Thus, if over the first rhomboid 

 another of equal thickness be laid conformably [i.e., so 

 that all the faces of the second shall be parallel to the 

 corresponding ones of the first), the only effect will be 

 that the apparent separation of the two images will be 

 doubled, just as if a single crystal of double thickness 

 had been used. But if from this position the upper 

 crystal be turned slowly round in its own plane upon 

 the lower, kept firm ; two other images will make their 

 appearance between the former, at first very faint and 

 almost close together, but, as the rotation of the upper 

 crystal continues, gaining strength (while the others grow 

 fainter) and opening out from each other in a direction 

 transverse to the line of junction of the first. When the 

 angle of rotation attains 45, four images are seen of 

 equal intensity, after which the two first grow fainter, 

 and at 90 vanish, the whole of their light having 

 passed into the other two and so on alternately. When 

 the upi^er rhomboid has made an exact semi-revolution 

 on the other, the image is single, and contains the whole 



