122 Mr. G. Gore on the Rotations [Dec. 9, 



axis by Ampere, Faraday, Sturgeon, and others, the magnet or wire has 

 either been immersed a large portion of its depth in mercury, or its 

 middle part has been connected by a wire with a surrounding annular 

 channel filled with mercury, and the electric current passed into or out of 

 the magnet or wire by means of that liquid, and the mercury has 

 formed an essential part of the arrangement. 



2. In different treatises on electro-magnetism, either no explanation 

 or different ones are given of the cause of the rotation of the magnet or 

 wire. The most usual cause assigned is the action of the portion of 

 current in the mercury and the fixed conductor immersed in it upon the 

 current in the rotating wire or magnet ; and the correctness of this ex- 

 planation is proved by the simultaneous movement of the mercury in the 

 opposite direction. 



De la Eive, in his Treatise on Electricity (English edition, 1853, vol. 

 i. page 259), very correctly remarks, " It must not be supposed that the 

 phenomenon of rotation is due, as has been erroneously stated, to the 

 action upon the magnet of the portions of the current that traverse it, 

 or that traverse the conducting-wires attached to the magnet and move 

 with it. In fact, how could a solid system be set in motion by a force 

 emanating from a portion of the very system itself, and connected with 

 it in an indissoluble manner? The action can only arise from a part of 

 the current which is independent of the system that moves. This part is 

 the portion of the circuit which is not connected with the magnet that is 

 set in rotation, and which consequently is situated independent of the 

 system in motion. This kind of action has been calculated by M. Ampere 

 in a perfectly rigorous manner." 



3. Although the action of the portion of the current in the mercury and 

 the fixed conductor in contact with it upon the movable current in the 

 rotating wire or magnet is a cause of the movement, there are other cir- 

 cumstances which cooperate to produce the rotation, and which are 

 themselves capable, without assistance, of producing it. Wiedemann* 

 and myself f have shown that an electric current passing from one end to 

 the other of a magnetized iron rod or wire having dissimilar poles at its 

 two extremities produces torsion of the wire (the two ends of the wire 

 moving of course in opposite directions), and that reversing the direction 

 either of the current or of the magnetic polarity, reverses the direction of 

 the twist. In all published cases of rotation of bar-magnets on their 

 axes by the influence of electric currents, the two ends of the magnet have 

 had dissimilar poles. By meditating upon certain facts connected with 

 this subject, I concluded that, by passing a current from one end to 

 the other of a magnetized rod or wire having similar poles at its two 

 ends, the magnet would probably rotate ; and experiment has demonstrated 

 that conclusion. 



* " Magnetischen Untersuchungen," Pogg. Ann. 1862, vol. cxvii. p. 193. 

 t "On Electro-torsion," Phil. Trans. 1874. 



