258 



If the magnet be strong enough, a phenomenon of the same kind is 

 obtained when in the above-mentioned experiment the larger tube 

 is put equatorially on the iron pieces, and the current concentrated 

 in the lowest part of the tube. Then, where the magnetic action is 

 greatest, the illuminated epibolic straight line is interrupted, termi- 

 nating in two cusps, which in the case of sulphurous acid show a green 

 colour. Waving light goes from one cusp to the other. In a room 

 quite dark, highly diffused light is seen moving from one pole to the 

 other, along lines of magnetic force. 



When the sphere previously mentioned is placed on the iron pieces 

 so that the two opposite apertures fall within the equatorial plane, 

 the transmitted current will be concentrated along an epibolic curve, 

 which in this case is one of the two arcs of the great circle within 

 the equatorial plane, supplementary to each other, and joining the 

 two apertures. If it be the lower arc, turned towards the magnet, 

 the epibolic current starting from the positive wire terminates in a 

 cusp, where the action of the magnetic force, supposed generally to 

 be strong, is greatest, while from the opposite side waving light 

 enters the sphere. In this case there is only one cusp. 



II. The light of the negative wire bent by the magnet into curves 

 and surfaces. 



In order to render the new class of phenomena most brilliant and 

 well denned, larger tubes of a cylindrical shape, into which long 

 wires enter at both ends, are to be selected. The light surrounding 

 the negative wire must be as bright as possible, and the well-known 

 dark space by which it is bounded must be very broad. The 

 magnet acts on this light in a peculiar way, having no analogy with 

 phenomena hitherto observed. I easily discovered the law giving 

 in all cases the exact description of the phenomenon. 



The light emanating from any not isolated point of the negative 

 wire, and diverging in all directions towards the interior surface of the 

 surrounding tube, is bent by the action of the magnet into the mag- 

 netic curve, ivhich passes through this point. According to the law 

 already mentioned, such a curve is the only one along which an elec- 

 tric current can move without being disturbed by the magnet. It 

 equally represents the form which a chain of infinitely small iron 

 needles, absolutely flexible and not subjected to gravity, would 



