261 



a fine spiral, separated by a dark space from the positive wire, taking 

 its origin in a cusp, and expanding more and more when it approaches 

 to the negative wire. In changing the polarity of the magnet, 

 instead of the single spiral starting from the middle of the wire, we 

 obtain two such spirals starting from the two extremities of the wire, 

 impelled against the glass of the tube and revolving in the opposite 

 sense. 



The phenomena just described and all the various phenomena of 

 the same class may be fully explained by the laws of electro-magnetic 

 action. For this purpose, we admit that the first element of each 

 elementary current starting from any point of the positive wire is 

 directed towards the negative wire. This supposition follows from 

 the observed fact, that the positive wire, if parallel to the negative, 

 becomes luminous along its whole length on that side which is turned 

 towards the negative wire. When acted upon by the magnet, these 

 first elements, bound to the positive wire, are allowed to move freely 

 along this wire. The single point, or the system of points, where all 

 the first elements meet before leaving the wire, may easily be deter- 

 mined. The following elements, subjected to the same force, are 

 entirely at liberty to obey it. This action may generally be defined 

 thus : Imagine an element of the elementary electric current starting 

 from any point of a magnetic curve, which connects both poles ; 

 imagine also the molecular currents (as assumed by Ampere) within 

 the magnet continued round this magnetic curve. Then, if the 

 element be perpendicular to the magnetic curve, the full action of 

 the magnet turns it round the curve in a direction opposite to the 

 direction of Ampere's molecular currents. If it be not perpendicular, 

 this action is to be decomposed along the normal to the magnetic 

 curve within the osculating plane. 



I think it worthy of notice, that the same electro-magnetic laws, 

 applied to magnetic action on an already formed electric current, 

 either bound to the conducting wire or free to move in space, 

 equally hold in determining the path of the current when acted upon 

 by the magnet during its formation between the two electrodes. 

 Thus a moving electric particle, under magnetic action and tending 

 towards the negative electrode, may be regarded as describing a 

 curve, in an analogous way as a projected material point does acted 

 upon by gravity. 



