434 An Essay on the Rejiecliun, Refraction, and Injlection 



if the thickness of the plate be extended to GH, these rays will 

 have separated very considerably bv the time they arrive at P 

 and Q ; and supposing the eye to be at E, the ray QQ will not 

 enter it at all, but the rays will be refracted into it from T, and 

 the two images will in consequence appear at P and T. Now, 

 whether we consider the difference in the angles PET and 

 peq^ or the quantity of space between the images on the two 

 surfaces, it is evident from this, that the distance between the 

 images must always increase in proportion as the thickness of 

 the plate increases. 



4, In order to account for there being six images, it is only 

 necessary to observe, that in this case there are three different 

 surfaces for the rays to pass through ; that is, some of them pass 

 through the two parallel surfaces KM and LN (fig. 3.) others 

 through KM and MN, and others again through KL and LN; 

 and as there are two images for each surface, it follows of course 

 tbat there must be six images altogether. Every one knows that 

 the number of images in a muitiplving glass is always in propor- 

 tion to the luimber of its surfaces ; and if I can make it appear 

 that in common glass of this figure and in the same situation 

 we should see three images, there can be no difficulty in under- 

 standing why there should be six images in Iceland crystal or 

 any other double refracting substance of the same figure. 



For the convenience of measurement, I have made the angles 

 of refraction in this figure to be half the angle of incidence, which 

 I believe is rather more than the ordinary refraction in glass; but 

 if the angles had been less, the images would have been shown 

 ju'st as well, — with only this difference, that the eye must have 

 been removed further from the plate. Let K L M N then represent 

 a plate of common glass, and rays of light from A, striking upon 

 O P and Q, refracted into the eye at E by the common law of 

 refraction, and three images of the same object will appear at 

 RS and T. Now if there was some other substance in the plate 

 that refracted the rays in a greater or less degree, every one of 

 these images would of course be doubled, as I have shown in 

 fig. 1, and the jjhaeuomenon consequently is explained. 



5. From not considering that the multiplication of the images 

 in Iceland crystal, as in glass, depends upon the number of sur- 

 faces the rays have to pass through, and not from any new i^odi- 

 fication either in the crystal or tlie rays, it has surprised at 

 least the early writers on optics not a little, that the two images 

 of any object, viewed through Iceland crystal, should not be 

 doui)led when these rays have to pass through a second crystal 

 before they enter the eye; and I believe it has never yet been sa- 

 tisfactorily explained, why sometimes there should be four images 

 exhibited by the two crystals and at others only two. " It is 



wonderful," 



