450 



upon the internal medium, and let F t be proportional to the resultant 

 attraction of the cubical mass towards the magnet. Let Po, W 2 

 and F 2 have similar significations when the force acts at right angles 

 to the line of compression. Then we may put 



Now, in a compressed magnetic cube, experiment proves that 



or 



i.e. Pj-P 2 ^-(ly^-M,,). 



As long as we are ignorant of the properties of the medium within 

 the body, we will, for the sake of completeness, consider the follow- 

 ing three distinct cases. 



I. The attracting force upon the medium within the cube is the 

 same when the force acts in either the one or the other of the two 

 directions, with respect to the line of compression. Here 



M 1= M 2 , 

 hence P l >P 2 . 



II. In whichever of the two directions of the force the attraction 

 of the particles may be greatest, the attraction of the internal me- 

 dium is also greatest in the same direction. Here, according as Pj 

 is greater or less than P 2 , M } is greater or less than M 2 ; hence, 

 inasmuch as 



andMj>M 2 . 



III. In whichever of the two directions of the force the attraction 

 of the particles may be greatest, the attraction of the internal me- 

 dium is greatest in the direction perpendicular to the same. Here, 

 according as Pj is greater or less than P 2 , Mj is less or greater than 

 M 2> so that the two hypotheses 



(1) Pj>P 2 and M^Mj, and 



(2) P J <P 2 andM 1 >M 2 



are both compatible with the sole condition, 



