On the Rotation of Metallic Spheres by Electricity. 107 



I am not aware that any other theory of light has accounted 

 for complete polarization by reflexion, when the angle of inci- 

 dence is such that tan ^=/x. 



It is evident from the above results that this theory gives the 

 following answer to the question under consideration : — If the 

 plane of reflexion of a ray completely polarized by reflexion be 

 called the plane of polarization, its transverse vibrations are per- 

 pendicular to the plane of polarization. 



Since the angle of incidence for complete polarization of a 

 reflected ray is found by measurement to have the above value, 

 which the theory gives on the hypothesis that the atoms of the 

 refracting medium occupy a small space compared to the vacant 

 space, the accordance of the theoretical result with fact is an in- 

 cidental confirmation of that hypothesis. 



Cambridge Observatory, 

 January 13, 1859. 



XIV. On the Rotation of Metallic Spheres by Electricity. By 

 George Gore, Esq. From a letter addressed to Dr. Bence 

 Jones*. 



IT has occurred to me that the phaenomenon of the rolling ball, 

 described in the Supplementary Number of the Philoso- 

 phical Magazine of July last, forms a good illustration of the in- 

 fluence of the element time in the jiroduction of physical phseno- 

 mena, which Mr. Faraday has so frequently insisted upon. 



By reference to the description of the experiment, it will be 

 found that the ball moves continuously in one uniform direction, 

 and that it will move in either direction with equal facility. The 

 explanation given of the motion in the article referred to, is that 

 it appears to be due to an intermittent electro-thermic action 

 taking place at the pointsof contact of the ball and rails at a minute 

 distance behind the line of centre of gravity of the ball, i. e. that 

 the passage of the electric current through the points of contact 

 produces heat, and the heat causes motion by expanding into 

 small protuberances the surfaces of the ball and rails at those 

 points. It is evident that if the passage of the current, the heat, 

 and the expansion, occurred simultaneously, without any period 

 of time (however small) elapsing between them, the expansions 

 would take place precisely under the centre of the ball, and the 

 ball would have as great a tendency to move in one dipcction as 

 in the other, and the progressive revolving motion would not be 

 sustained. But if wc suppose that a minute period of time is 

 occupied in the production of the heat by the electricity, or of 



* Communicated by Dr. Bence Jones. 



