Royal Society. 65 



takes place will, of course, be reversed by reversing either the 

 course of the ray or the poles of the magnet. Hence it follows that 

 the polarized ray is made to rotate in the same direction as the cur- 

 rents of positive electricity are circulating, both in the helices com- 

 posing the electro-magnet, and also in the same direction as the 

 hypothetical currents, which, according to Ampere's theory, circu- 

 late in the substance of a steel magnet. The rotatory action was 

 found to be always directly proportional to the intensity of the 

 magnetic force, but not to that of the electric current ; and also to 

 be proportional to the length of that portion of the ray which re- 

 ceives the influence. The interposition of substances which occa- 

 sion no disturbance of the magnetic forces, produce no change in 

 these effects. Magnets consisting only of electric helices act with 

 less power than when armed with iron, and in which magnetic ac- 

 tion is consequently more strongly developed. 



The author pursues the inquiry by varying in a great number of 

 ways the circumstances in which this newly-discovered influence is 

 exerted ; and finds that the modifications thus introduced in the re- 

 sults are all explicable by reference to the general law above stated. 

 Thus the effect is produced, though in a less degree, when the po- 

 larized ray is subjected to the action of an ordinary magnet, instead 

 of one that derives its power from a voltaic current ; and it is also 

 weaker when a single pole only is employed. It is, on the other 

 hand, increased by the addition of a hollow cylinder of iron, placed 

 within the helix, the polarized ray traversing its axis being then 

 acted upon with great energy. Helices act with equal power in 

 any part of the cylindric space which they enclose. The heavy glass 

 used in these experiments was found to possess in itself, no specific 

 magneto-inductive action. 



Different media differ extremely in the degree in which they are 

 capable of exerting the rotatory power over a polarized ray of light. 

 It is a power which has no apparent relation to the other physical 

 properties, whether chemical or mechanical, of these bodies. Yet, 

 however it may differ in its degree, it is always the same in kind ; 

 the rotation it effects is invariably in one direction, dependent, how- 

 ever, on the directions of the ray and of the magnetic force. In 

 this respect it differs essentially from the rotatory power naturally 

 possessed by many bodies, such as quartz, sugar, oil of turpentine, 

 &c, which exhibit the phenomena of circular polarization ; for in 

 some of these the rotation takes place to the right, and in others to 

 the left. When, therefore, such substances are employed as dimag- 

 netics, the natural and the superinduced powers tend to produce 

 either the same or opposite rotations ; and the resulting effects are 

 modified according as they are cumulative in the former case, and 

 differential in the latter. 



In the concluding section of the paper, the author enters into ge- 

 neral considerations on the nature of the newly-discovered influence 

 of electricity and magnetism over light, and remarks that all these 

 powers possess in common a duality of character which constitutes 

 them a peculiar class, and affords an opening which before was 



Phil. Mas. S. 3. Vol. 28. No. 1 84. Jan. 1 84b". F 



