28 report — 1842. 



me that I was able to explain this singular pheenomenon. The three following pro- 

 positions, inferred from the said studies, may suffice for my present purpose : — 



1st. When an electric current is made to circulate round an iron endowed with 

 magnetic polarity, a second is generated, which, together with that pre-existing, forms 

 a system of two polarities, more or less strong than the first, or even entirely 

 neuter, according to the force and the direction of the current employed. 



2nd. The magnetic strengthening or weakening which is observed (when work- 

 ing as above mentioned), is proportionably less when the pre-existing polarity is 

 stronger. 



3rd. When magnetism is communicated to iron not magnetized, and not altered 

 in its magnetic susceptibility, and a transitory current is then caused to circulate 

 round it, a greater alteration is caused in the iron when it tends to magnetize it in 

 a contrary direction, that is, to produce the north pole in the extremity in which the 

 south pole is found. 



Let us endeavour to demonstrate these three propositions. The co-existence of 

 two conspiring polarities in one piece of iron I should not be able now to prove ; 

 but fortunately that is not necessary ; and to explain the phaenomenon of which we 

 treat, it suffices that the co-existence in a piece of iron of such polarities cannot be 

 doubted when they are contrary. 



That such co-existence of opposite polarities is possible, is proved by the known 

 fact, that two magnetic needles connected by poles of different name in contact, pre- 

 sent a whole sometimes without any polarity, although, after whatever time they 

 have been disunited they are both magnetized, as they were before they were con- 

 nected. Still the consecutive points which are sometimes observed in magnetized 

 needles, offer a proof that such magnetic systems may exist in the same iron. But 

 here it is necessary to prove their existence in iron wires, in which not only are no 

 consecutive points, but where there is no other trace of magnetism. 



To succeed in this separation, I set to work to study those operations which serve 

 to destroy magnetism, but not to generate it, at least in small masses of iron. Such 

 are heating, blows, shocks, friction, flexure, &c. Wires or cylinders of iron or steel, 

 from four to nine centimetres in length, and weighing from four or five grammes to 

 almost a hundred grammes, I have never succeeded in magnetizing by the said 

 means, especially if the irons had never been magnetized, or, had they been so, if 

 the magnetic state had been taken from them by heating, or by any other of the 

 said operations. Thus, as often as I destroyed by such means the magnetism pro- 

 duced in the iron by the electric currents, or by any other magnetizing agent, I never 

 saw any variation in the magnetic susceptibility. The propositions deduced from the 

 numerous experiments made upon the de-magnetizing action of the above-men- 

 tioned operations, cannot be inserted in this abstract. It may suffice to record the 

 following : — 



If an iron deprived of magnetism, and whose susceptibility has not been altered, 

 be magnetized, and afterwards subjected to one of the said de-magnetizing opera- 

 tions, the portion of magnetism it will lose will be greater when the magnetization 

 communicated to it has been weak. The following is a proof : — 



A large iron wire, eight centimetres in length, and weighing twenty-seven grammes, 

 was magnetized, so that the needle of the magnetometer deviated eight degrees*. 

 Being let fall upon a table from the height of two decimetres, it lost so much force 

 that the needle deviated but one degree. But having then magnetized it so that the 

 needle deviated to 25°, and afterwards subjected it to a shock equal to the preceding, 

 it yet did not lose the half of the magnetism which it had, since it caused the needle 

 to deviate 14°. 



* This apparatus consists of a needle, from the extremity of the cap of which rises 

 vertically a stick of brass, to which, by means of a running ring furnished with screws , 

 in order to stop it at a convenient height, is attached a small horizontal rod, which carries 

 a tube of glass surrounded by a coil of copper wire covered with silk. The tube is parallel 

 to the magnetic needle, and its central point is in the perpendicular which passes through 

 the centre of the needle itself. The wire is so placed that its axis coincides with that of the 

 tube, and from the firm deviation of the under needle we may judge of the magnetic force 

 which the iron already possesses, or of that which is acquired by it in discharging the 

 Leyden jar upon the copper coil which surrounds it. 



