108 PROCEEDINGS OF THE AMERICAN ACADEMY 



powerful battery flows in the helix. Two helices attract and repel as 

 two magnets ; a single helix and a magnet attract and repel as two 

 magnets. In both cases, particularly in the first, the action is much 

 weaker than when we experiment on two magnets. In all the electro- 

 dynamical motions, it is well known that those are the weakest which 

 are produced by the reciprocal action of currents alone, and that a 

 great gain is effected when we substitute for one or bo^ of the currents 

 some kind of iron or steel magnet. Moreover, in electro-dynamical 

 induction, the same superiority appears on the side of the currents of 

 the steel magnets. The currents induced by such magnets are much 

 stronger than those induced by a battery current. When the battery 

 current flows around a piece of soft iron, making it an electro-magnet, 

 then we have the best possible source for induced currents. From 

 all these facts, many of which are familiar, I infer that the battery cur- 

 rents, although possessing a greater magnetizing power than those at- 

 tached to a steel magnet, are, nevertheless, of less intrinsic energy. 



" Whence, then, the question recurs, does this superior efficiency of 

 the weaker currents in imparting magnetism to iron proceed ? One 

 cause, without doubt, is the favorable position in which the inducing 

 currents act upon the dispersed currents of the unmagnetic iron. The 

 superficial action of one piece of iron or steel upon another, to which 

 we have already referred, when interpreted by the light of Ampere's 

 theory, amounts to this. The inducing currents of the original mag- 

 net and the induced or directed currents of ,the other bar touch, like 

 any two circles external to one another, only at a single point. If the 

 circulation was around the whole mass of each bar, these circles must 

 still rapidly separate from one another. When we add to this that the 

 flow is about each single particle in each bar, it is obvious that the cur- 

 rents which direct and those which are directed are, for the most part, 

 so remote from one another, and so oblique, as to act at a very great 

 mechanical disadvantage. Moreover, the portions of the circuit which 

 are opposite to the adjacent portions exert a contrary action to that of 

 the latter, and diminish the small result which otherwise might be pro- 

 duced. In electro-magnetizing, the battery current flows wholly round 

 the piece of iron to be magnetized. Throughout the whole circula- 

 tion, every portion of it is near to at least some part of the iron, so as 

 to act favorably upon it. When we magnetize iron by touch, we can, 

 it is true, turn the different sides up and touch it on all of them. We 

 might even make the original magnet hollow, and insert the bar to be 



