making Experiments upon Elliptic Polarization 235 



The light used in these observations was that of a candle 

 placed at a short distance, and was admitted through small 

 apertures at the ends of the tubes.* The Nicol's prism in the 

 first tube having been secured in a position in which its princi- 

 pal plane was inclined 45 to the plane of incidence, and the 

 two arms having been set at the proper angle with the surface 

 of the metal, the Fresnel's rhomb and the Nicol's prism in the 

 second tube were moved simultaneously, until the image of the 

 candle became as faint as possible. Had light perfectly homo- 

 geneous been employed, the image could have been made to 

 vanish altogether ; but instead of vanishing, it became highly 

 coloured ; and our rule in observing was to make the blue at 

 one side of it, and the red at the other, equally vivid, so as to get 

 results which should belong, as nearly as possible, to the mean 

 ray of the spectrum. When this was done, the angles and |3 

 (subject, however, to certain corrections which will be hereafter 

 explained) were respectively read off from the divided circles 

 belonging to the rhomb and the prism. The observations were 

 made at large incidences, because it is within the last thirty de- 

 grees of incidence that the phenomena go through their most 

 rapid changes. 



If we now cast our eyes on the above Table, making due 

 allowance for the uncertainty arising from the dispersion of the 

 metal, we shall be struck with the agreement between the cal- 

 culated and observed numbers. The differences are greatest in 



* See the description of the instrument in the Proceedings, VOL. i. p. 159 

 p. 138). 



