300 PROCEEDINGS OF THE AMERICAN ACADEMY 



In tlie case of two similar poles being brought together, there is no 

 perceptible change of distribution : the most delicate experiments 

 failed to show any. Probably a change does, however, take place ; 

 but the resulting curve is so like the former curve, that the difference 

 is imperceptible. 



On the theory of the analogy of the magnetic lines of force to elec- 

 tric currents, these laws will, I think, be nearly, though not com- 

 pletely, fulfilled. 



These laws, if true, and, if not, in so far as they are true, enable one 

 to determine the shape which the curve would assume were the 

 armature and magnet of one jiiece of metal ; i.e., if the juncture was 

 absolutely perfect. 



Moreover, the same method enables us to draw the limiting values 

 for different lengths of armature : for we have only to increase each 

 abscissa of the curve of the magnet as the total length of magnet and 

 armature is to that of the magnet alone, and to take the half-sum and 

 half-difference as before, to find the curves on the magnet ; then, draw- 

 ing a curve in the same manner to represent another magnet of length 

 of armature joined by a similar pole, the curves on the armature can 

 thus be determined. We shall see, then, that a soft-iron armature 

 acts like a steel armature of about double the length. Whether this 

 is due to contra-distribution, or to a superior magnetic capacity, I leave 

 to be determined. 



The following experiments were devised to show the change in 

 magnetic moment due to armatures. The experiments are rough ; but 

 they give far the best idea of the changes that take place in magnetic 

 distribution. A magnet was placed upon the floor at a fixed distance 

 from the galvanometer-needle, to the east of it, and pointing east and 

 west. Then an armature was added, and the position of the centre of 

 the magnet was marked upon the floor. Then the whole combination 

 was removed to a distance, and the deflection was noted. Then the 

 combination was brought back, with poles reversed, and was moved 

 toward the galvanometer till the deflection was equal to the previous 

 one. 



This at once showed the increase of distance between the poles ; for 

 it must have been equal to the distance of the centre of the magnet 

 from its former position, and the central point must have advanced 

 half this distance. Having determined the jjosition of the central 

 point, the magnet alone was now turned ujion it ; and the ratio of the 

 deflection of the magnet and ai-mature to that of the masnet alone 

 showed the change of magnetic moment. 



