460 On the Influence of Magnetisation on Iron Bars, 



the distances between the sets or groups being a little greater in 

 the latter than in the former line. 



In magnetization, the nearest molecules will become still 

 nearer; and as the entire groups of molecules cannot have an 

 arbitrary free motion, in some places there will be greater breaks 

 even when the entire wire is shortened, as in the following scheme : 



The resistance must thereby be increased ; but to what extent, 

 depends on the nature of the iron. If the groups of molecules 

 had free motion, the magnetization would necessarily induce a 

 decrease in resistance. And in fact if a glass tube be filled with 

 fresh iron filings and the two ends of a battery placed in the ends, 

 scarcely any current passes through the filings. In an experi- 

 ment of this kind the current of two DanielPs cells only deflected 

 the mirror of my reflecting galvanometer about 45 divisions : 

 when the tube was magnetized in a spiral, the deflection increased 

 to 120; on shaking the tube, to 450; on interrupting the mag- 

 netizing current, it sank to 170 ; and by a second shaking, to 44. 



The decrease of resistance in the direction at right augles to 

 the magnetic axis still requires explanation. The experiments 

 of Thomson are, indeed, such that a mere mechanical pull 

 must necessarily be succeeded by such a decrease; but they 

 prove nothing as to the specific action of magnetization. I folded 

 a covered wire of 47500 millims. in length and 0*36 millim. 

 thickness in 332 coils about a rectangular glass plate, and placed 

 this in a magnetizing spiral. To prevent the excitation of a cur- 

 rent of induction in the iron wire, 160 coils were passed in one, 

 and 166 in the other direction about the glass plate. The wire 

 interposed in the bridge offered the resistance 99*195 : when the 

 magnetizing current was closed, there was not the least mo- 

 tion in the mirror of the galvanometer, although the wire was 

 now magnetized at right angles to its length. An alteration of 

 0'0055 in the resistance would have displaced the scale by a 

 whole division ; hence if for these small values the displacements 

 are assumed to be proportional to the alterations in resistance, 

 a decrease of resistance in the proportion of 1 : 3000 would have 

 effected a displacement of six divisions. As a displacement of 

 O'l division is very distinct, a decrease in the resistance in the 

 ratio of 1 : 180000 could not have been overlooked; so that I 

 think I can fairly draw the conclusion that a decrease of resistance 

 in a direction at right angles to the magnetic axis has not been 

 demonstrated. 



Erlangen, March 1866. 



