98 Prof. Faraday on some Points of Magnetic Philosophtj. 



times in another, but with the preservation of their full and 

 equivalent amount, then how are we to consider them disposed 

 of in the case of a cylinder or globular magnet, placed in air or 

 vacuo, so as to be entirely self-dependent ? — or in the case of a 

 magnetic sphere placed in an inverted position in the magnetic 

 field, so as to be entirely surrounded and enclosed by magnetic 

 forces having a contrary direction to its own (3321. 3238.) ? 



3332. If we say that the dualities of such a magnet are de- 

 pendent on each other (which is the third case (3327.), then 

 we have to consider how this can be, consistently with the 

 distant mutual action, either of magnetic fluids or electric 

 currents, acting in right lines only. Such action must then 

 be through the body of the magnet (3260.). If we confine our 

 attention to magnetic fluids, then the direction of their forces 

 towards each other through the magnet when it is alone, must be 

 of the like nature as their direction to approached iron, in which 

 they are supposed to induce collections of the contrary fluids, or 

 towards the fluids at the contrary poles of approached equal or 

 superior magnets; i. e. the two poles of the magnet must be 

 conceived of as centres of force, sometimes exerting their power 

 towards each other in a given direction through the body of the 

 magnet, and at other times exerting them outwardly to external 

 poles in a direction exactly the contrary. But the currents 

 which are evolved by the rotation of the magnet, or of discs of 

 metal combined with it (3119. 3163.), show that the direction 

 of the force (which is its polarity) is not thus reverse in the 

 two halves of the case, but is the same within the magnet as in 

 the prolongation of direction through and beyond the pole ; 

 and also, that whether the magnet be alone, and therefore sup- 

 posed to have the polar forces exerted on each other through it, 

 or be in relation to outer magnets, so as to have this exertion 

 of force entirely removed from its interior, still it is always the 

 same; having in both cases the same condition, direction, and 

 amount of power within it (3116.). 



3333. If the charged and polar state of the magnet be sup- 

 posed to depend upon molecular electric currents, held by some 

 internal condition in a position of parallelism, it is impossible 

 that these can act backwards upon each other through the mag- 

 net in straight lines, so as to put the northness and southness 

 of the pole in mutual dependence, as they are supposed to be in 

 relation to external poles, without the currents themselves being 

 displaced and turned, until the whole magnet is neutralized ; 

 falling back into the undeveloped state, just as a piece of soft 

 iron falls back. When this return of state happens in soft iron 

 or steel in any degree, a helix round these show the induced 

 currents consequent on such a change ; and a loop (3133. 321 7.) 



