202 PROCEEDINGS OF THE AMERICAN ACADEMY 



effect is induced in the magnetized mass which can only be compared 

 to a vibration or wave. The reversals of the inducing current cause 

 corresponding reversals of polarity in the body acted on, and as these 

 reversals are continuous and very rapid (5,000 to 6,000 per minute, 

 for example), a molecular vibration probably arises. It is probable 

 that a condition of strain or set is one of the jjhenomeua of mag- 

 netism. 



The particles have been made to assume a certain definite or polar 

 relation or position. When, however, a powerful movement or vibra- 

 tion is caused, it is evident that when this vibration has become com- 

 plete, — that is, involving the whole mass, — all previous conditions 

 of strain or "permanent set" will be overcome. It must be remarked 

 that, in order to perfectly attain this result in all cases, the exciting 

 force must be sufficient. 



When the alternating current ceases, the body acted on is left per- 

 fectly free from polarity. It is, however, in a state of extreme sensi- 

 tiveness, and must be allowed to remain at rest for a short time. If 

 it is placed north and south, it will assume polarity, and very strongly, 

 if struck with a hammer when held in the position of the dip. 



Demagnetization requires but a short time in most cases, — from 

 one to three minutes if the current is properly adjusted. There are 

 several ways of performing the experiment, but it will be sufficient at 

 this time to refer to a few of them. The most effective method is to 

 enclose the mass to be demagnetized in a coil of such a length that 

 the whole body will occupy an approximately central position. The 

 coil may be a simple one, in which case it must stand east and west, 

 and before removing the object the electric machine must be stopped, 

 and the current allowed to die away. Also, when the object is taken 

 out of the coil, it must be carefully shielded from the earth's induc- 

 tion. Or the coil may be so constructed that it can be opened or 

 divided at the centre without breaking circuit, and then the object 

 can be taken out without stopping the alternating current. One of 

 the coils we used was made of No. 12 copper wire, wound as one coil, 

 but in halves, with an elastic connection. It is well known that, with 

 an alternating current, self-induction in the coil materially reduces 

 the current, and therefore the coil should be one of a comparatively 

 small number of turns. 



Demagnetization of small masses, not too retentive of magnetism, 

 may be performed by placing them on the end of a bar contained in 

 the coil, which is a part of the alternating circuit. A bar of low steel, 

 somewhat longer than the coil, was used, and the small objects placed 

 on its projecting end. 



