64 PHENOMENA OF LIGHT AND COLOUR. 



the images, therefore, appear to have rounded corners.* 

 This explanation is fanciful, but the one referring to the 

 propagation of light in conical form, although of little value, 

 is suggestive, for the true explanation can be obtained by 

 drawing a series of co-axial double cones with their apices 

 at various points on the sides of the aperture, the Sun's disc 

 and its image being the bases of each double cone. The 

 overlapping of the separate images of the Sun's disc, thus 

 drawn, causes the images formed by the aperture to be 

 circular, if the aperture is small, or to have rounded corners, 

 if the aperture is large. 



Aristotle was fully aware that reflection takes place from 

 mirrors and other smooth surfaces. He often uses the word 

 AnaJclasis, a breaking back or aside, to denote this pheno- 

 menon, especially in his descriptions of halos and rainbows. 

 There does not appear to be any passage in his works, 

 however, showing that he was aware of the equality of the 

 angles of incidence and reflection. This seems to have been 

 stated for the first time in Euclid's Catoptrics, Prop, i., 

 where the law is enunciated and proved for plane, convex, 

 and concave mirrors. 



He does not use the word Anahlasis or any other word 

 in such a way as to show that he was acquainted with the 

 phenomenon of refraction, but in Meteorol. iii. c. 4 there 

 are some passages which deserve special notice in connection 

 with this question. After speaking about the strange optical 

 illusion in the case of a man whose sight was very weak and 

 who saw an image of himself in consequence of the adjacent 

 air acting like a mirror, Aristotle says : — " Wherefore head- 

 lands appear inverted in the sea, everything appears larger 

 when the easterly winds (eupoi) blow, and also objects seen 

 through mists, e.g., the Sun and stars seem to be larger 

 when rising or setting than when they are high in the 

 heavens." t 



Ideler says that these examples, given by Aristotle, 

 pertain not so much to reflection of light as to refraction.! 

 This is not so. They pertain mainly to reflection and 

 absorption. The phenomena of absorption were only im- 

 perfectly understood by Aristotle, but many statements he 

 makes about light and colour show that he never lost sight 

 of what appeared to be the effects of the medium between 



'■'• Problems, xv. 6. f Meteorol. iii. c. 4, s. 4. 



I Aristot. Meteorol. Leipzig, 1836, vol. ii., p. 20. 



