117 



is induced in a steel bar at right angles. This bar may again induce 

 magneto-electricity in a wire at right angles to it, by which another 

 bar may become magnetic; and so on, showing a repetition of simi- 

 lar powers successively brought into action, but their efficiency at 

 each step greatly diminished. 



The effects hitherto described were due to a momentary action : 

 in order to obtain continuous action the author applied the principle 

 of circular motion. For this purpose a thick copper disc was made 

 to revolve near the magnet, so that a portion near its edge passed 

 between the ends of two bars of iron which concentrated and ap- 

 proximated the poles. The edge and a portion round the centre of 

 the disc were well amalgamated: an amalgamated conductor was ap- 

 plied to the edge of the disc near the poles, and with this, one end 

 of the wire of the galvanometer was connected, the other end being 

 connected with the centre of the disc. While the disc revolved, the 

 needle of the galvanometer was permanently deflected at least 45° 

 in one direction; and when the motion of the disc was reversed, the 

 permanent deflection was in the opposite direction. 



When the disc revolved horizontally in the direction of the sun's 

 daily motion, the unmarked pole being beneath the disc and the 

 marked pole above, it appeared, by the indications of the galvano- 

 meter, that positive electricity was collected at the edge of the disc 

 nearest to the poles: if the marked pole was below and the un- 

 marked pole above, then negative electricity was collected at that 

 part of the disc : and if in either case the direction of the motion 

 was reversed, the nature of the electricity collected at the same 

 place was also reversed. 



The experiment being made in a still more simple form, by pass- 

 ing a plate of copper longitudinally between the poles of the mag- 

 net, it appeared that positive electricity was collected on one edge 

 of the plate, and negative on the opposite; and if the plate was 

 passed in the contrary direction, then the electricities on the edges 

 were reversed. 



When a wire was passed laterally between the poles, similar results 

 were obtained. 



The law according to which the electricity excited depends upon 

 the pole of the magnet near which a wire moves, and the direction 

 of its motion, although not so expressed by the author, appears to 

 be this : Let the wire revolve parallel to itself about a bar magnet, so 

 that its centre coincides with any curve ; — for example, (in order to 

 mark more readily the points where the direction of the current of 

 electricity changes,) with an ellipse, the major axis of which coincides 

 with the axis of the magnet, and the minor axis passes through its cen- 

 tre; let the wire be inclined at any angle to the plane of the ellipse, 

 which in the first instance we will suppose to be horizontal, and that 

 the marked end of the magnet is pointing north; and let the wire 

 move parallel to itself in the direction of the sun's daily motion ; then 

 while the wire revolves from the western extremity of the axis minor 

 round the marked pole to the eastern extremity, the electric current 

 will be from the end of the wire below to the end above the orbit : 



