24 



A POPULAR ACCOUNT 



This \vill be understood by attending to 

 the effects of the prism on the circular 

 spot. 



Let S, fig> 26, be the circular illumi- 

 nated spot cast upon the screen 

 by the light proceeding directly Fig. 26. 

 from the aperture without be- 

 ing intercepted by the prism. 

 Let S A be the distance on the 

 screen to which the least re- 

 frangible rays are deflected. 

 These rays, which before fell 

 upon the circle S, will now 

 fall upon the circle A. Let 

 S Z be the distance on the 

 screen to which the most re- 

 frangible of the rays incident 

 on S are deflected by the 

 prism. These rays, which be- 

 fore fell upon the circle S, 

 now fall upon the circle Z. 

 The rays of all the interme- 

 diate degrees of refrangibility 

 are deflected by the prism, 

 so as to fall upon circles whose /" \ 

 centres occupy the entire space ( s J 

 from A to Z. From A take \^^s 

 a distance A C, equal to 

 twice the radius of the circle A, or, 

 what is the same, to the breadth of the 

 spectrum. A circle described round the 

 centre C, with a diameter equal to A C, 

 will evidently touch the circle A, but 

 neither circle will be within or upon the 

 other. It is evident, however, that every 

 circle of the same diameter, whose cen- 

 tre lies between A and C, must lie partly 

 upon the circle A, and partly upon the 

 circle C. Hence it is evident, that the 

 rays which, deflected from S, illuminate 

 the circle A, and those which illuminate 

 the circle C, are not intermixed. But 

 between the points C and A is a consi- 

 derable space, and between the degrees 



of refrangibility which would cause rays 

 to be deflected to these points, are many 

 intermediate degrees, in virtue of which 

 rays would be deflected upon innume- 

 rable circles, whose centres lie between 

 C and A. It follows, therefore, that all 

 these lights of intermediate refrangibi- 

 lities are intermixed with lights upon 

 the circles A and C, neither of which, 

 therefore, shine with pure homogeneous 

 light. Since the lights of different re- 

 frangibilities which are thus intermingled 

 are diffused over circles whose centres 

 occupy the space A C, the number of 

 such circles must be proportional to 

 the space A C, or to the breadth of the 

 illuminated circle S, increasing and di- 

 minishing as that breadth is increased 

 or diminished. In this proportion, then, 

 lights of different refrangibilities are 

 mixed in the spectrum ; and, therefore, 

 every means which can diminish the 

 breadth of the spectrum will propor- 

 tionally increase the purity of the lights. 

 The first method which occurred to 

 Newton to accomplish this, and, at the 

 same time, to remove the penumbral 

 fringe already mentioned, was to per- 

 forate the screen in the space L L' (fig. 

 25), and to receive the light trans- 

 mitted through the perforation on a 

 second screen behind the first. By this 

 means the penumbral ring would be 

 received upon the first screen at L' P, 

 L P', and the uniform light of L L' would 

 be admitted to the second screen through 

 the perforation. In this case, the breadth 

 of the illuminated circle on the second 

 screen, and therefore that of the spec- 

 trum, would be nearly equal to that of 

 the perforation. In proportion as the 

 breadth of the spectrum would be thus 

 reduced, the intermixture of heteroge- 

 neous lights would be diminished, and, as 

 27. 



we have just explained, the penumbral 

 light would be altogether intercepted. 



He, however, accomplished what he 

 aimed at more effectually by the follow- 



ing method. At a distance of ten or 

 twelve feet from the hole O, Jig. 27, he 

 placed a lens L, which formed an image 

 of the hole at O'. If the lens were so 



