22 



A POPULAR ACCOUNT 



emerged from the surface C D parallel 

 to ifslncidence ; for since the refracting 

 angles of the two prisms, A C B and 

 C B D, are turned in opposite directions, 

 and are equal, they neutralise each 

 other's effects, and the light emerges as 

 it entered. The ray M O, emerging from 

 the compound prism, was received upon 

 another prism I H K, and was dilated in 

 the same manner as if it had been re- 

 ceived directly from the aperture F, and 

 the spectrum was received in the usual 

 way upon the screen at P. 



Another prism XV Y was placed 

 above the compound one, in such a posi- 

 tion as to receive a ray reflected from 

 the surface B C ; and above it a paper 

 screen, placed to receive the light re- 

 fracted by it. 



Let us suppose that in the position 

 thus given to all the prisms, a vivid 

 spectrum appears on P, and no appear- 

 ance of light is exhibited on p. The 

 compound prism is now slowly turned 

 round its axis in the direction A C D B, 

 so as gradually to increase the obliquity 

 of the ray F M to the surface B C. At 

 a certain position, a strong violet light 

 will appear upon the screen p at v. 

 Maintaining the prisms, for the present, 

 in this position, let the circumstances of 

 the experiment be examined. The violet 

 part of the spectrum on P will be found 

 to have disappeared. If a screen be in- 

 terposed between the prism I H K, and 

 the compound prism, it will be found 

 that the light which falls upon it will be 

 of a colour which would result from the 

 mixture of all the colours of the spec- 

 trum, except the violet. If a screen be 

 interposed between the prism X V Y and 

 the compound prism, it will be found 

 that the light which falls upon it is the 

 violet. The inference is most obvious. 

 The incident ray F M has obtained the 

 limiting obliquity, corresponding to the 

 most refrangible or violet light. This 

 light is accordingly reflected by the sur- 

 face B C, and is transmitted through the 

 prism X V Y to the screen p, where it 

 appears. The violet light not passing 

 with the other parts of the sun-beam 

 through the prism BCD, has disap- 

 peared from the spectrum on P, which, 

 therefore, now terminates with the in- 

 digo light. The beam M O, before it is 

 dilated by the prism I H K, is in fact the 

 sun-beam FM deprived of the violet 

 light, which has been reflected at M, 

 and, therefore, it is composed of all the 

 colours of the spectrum except violet, as 

 appears by the spectrum, which is now 



exhibited on P. Thus it follows that 

 violet light is totally reflexible at an ob- 

 liquity, which is insufficient to prevent 

 the refraction of lights of other colours. 

 The prism A C D B was now slowly 

 turned round a little more in the same 

 direction, and the effects observed. The 

 screen JP, in addition to the violet lights, 

 was now illuminated by the indigo i % of 

 which the spectrum on P was observed 

 to be deprived. The place of the indigo 

 light on p was also next to v, but so as 

 to be less refracted. On interposing the 

 screen between the prism I H K and the 

 compound prism, it was found to be 

 illuminated with a colour, which would 

 result from the mixture of all the colours 

 of the spectrum, except violet and in- 

 digo ; and on the other hand, on inter- 

 ?osing it between the prism X V Y and 

 H K, it was illuminated with a colour 

 which would be produced^ by the mix- 

 ture of violet and indigo. 



The inference from these effects is 

 consistent with the former one. The 

 prism A C D B had now attained that 

 position which gave the incident ray 

 F M the limiting obliquity of the indigo 

 light. It was accordingly reflected, to- 

 gether with the more reflexible violet 

 light. The lights of other colours were 

 transmitted, and produced the ett'ects 

 observed on P, and between I H K and 

 ACDB. 



This process was continued, the prism 

 ACDB being slowly and gradually 

 turned on its axis in the same direction. 

 The lights, blue, green, yellow, orange, 

 and red, successively disappeared from 

 the spectrum on P, and at the same 

 times appeared in succession in their 

 proper places on p. In each case the 

 colour of the light in the beam M O was 

 such as would result from the mixture 

 of the colours on P, and the colour of 

 the light M N was such as would result 

 from the mixture of the colours on p. 



All these effects are obvious and 

 beautiful consequences, and therefore 

 confirmations of Newton's doctrine. As 

 the obliquity of the incident ray F M to 

 the surface B C is gradually increased, 

 it becomes successively equal to the 

 limiting obliquities of the several co- 

 loured lights, and in the same succes- 

 sion reflects them to the screen p ; the 

 other screen P being in the same order, 

 and at the same instants of time, de- 

 prived of them. Thus it appears, not 

 only that lights of different colours 

 are differently refrangible, but also that 

 they are differently reflexible ; and that 



