i6o 



HISTORY OF SCIENCE. 



band like the rainbow may be thrown on white paper by means of a 

 glass prism. 



FIG. 66. 



How well Descartes could employ the Baconian method may be 

 seen by the way in which he arrived at the conclusion that the colours 

 of the rainbow are due to refraction and to no other cause. He sought, 

 he says, whether there were any other thing in which these colours 

 appear in the same manner, in order that, by comparing one with the 

 other, he might better judge of the cause of the colours. " Then, re- 

 collecting that a prism or triangle exhibits similar colours, I examined 

 one like MNP (Fig. 67), where two surfaces, N M and N p, are quite flat, 

 and incline one to another at an angle of 

 about thirty or forty degrees , so that, if the 

 rays of the sun pass through N M at right 

 angles, and thus undergo no sensible refrac- 

 tion, they must undergo a considerable re- 

 fraction in passing out through N p. Cover- 

 ing one of these two surfaces with a dark 

 body, in which there was a rather narrow 

 opening, D E, I noticed that the rays, pass- 

 ing through this opening and thence falling 

 upon a white cloth or paper F G H, depicted 

 there all the colours of the rainbow, and the 

 red was always depicted towards F and the 

 blue or violet towards H. From this I 

 learnt, first, that the curvature of the surfaces 

 of the drops of water is not necessary for 

 F IG 6 7> the production of these colours, for the 



surfaces of the prism are quite flat; nor 

 is a particular magnitude of the angles of refraction necessary, for 



