350 Dr. L. T. More on the Changes in Length 



if the magnet is cut in two the ends are held together by a 

 force W/Sir per square centimetre, showing that this force 

 must always be present when a rod is magnetized. The 

 agreement between the theoretical value B 2 /87r and the expe- 

 rimental law for the lifting-power has been recently shown 

 by E. Taylor Jones*. This effect for high intensities of magne- 

 tization is a large one, and becomes one of the most important 

 factors in the observed changes in length. The second of 

 these mechanical stresses is the effect due to the change in 

 Young's Modulus when the rod is magnetized. That the 

 elasticity is influenced by magnetization was shown theore- 

 tically by J. J. Thomson!, and the phenomenon was observed 

 last year in some experiments made by the writer ; but no 

 quantitative results could be obtained beyond the fact that 

 the elasticity for soft iron was slightly diminished. A. Bock J 

 concludes from his work that the decrease in elasticity must 

 be less than one-half per cent. If the wire is not stretched 

 by weights, this decrease in the elasticity will affect only the 

 contraction due to the B 2 /87r force. This decrease in the 

 elasticity, according to Bock, is very small and less than one- 

 half per cent. On the other hand, if the wire is loaded with 

 weights this effect becomes very marked since a large quan- 

 tity, the stretch of the wire by the great weight, is altered. 

 This question will be more fully discussed later in the paper. 

 If these two indirect actions were allowed for, there would 

 remain only the direct action of the magnetism upon the 

 metal under a constant tension. This latter relation would 

 evidently furnish comparable results, and may in the future 

 throw some light upon the action of magnetism on matter. 



Apparatus. 



Since it was necessary to obtain the modulus of elasticity 

 for the specimen experimented on, in order to make a proper 

 correction for the electromagnetic stress, and as I wished to 

 observe the effects of mechanical stress, it was convenient to 

 experiment on thin long wires of the metal placed in a vertical 

 position. To magnify the phenomena, a system of levers was 

 used involving the tilting of a mirror mounted on three legs 

 very close together ; a method first invented by Professor 

 Rowland. The general plan of using levers and a jacket 

 cylinder was suggested to me by Dr. Ames. 



The wire to be tested was suspended from a tall tripod 



* Jones, loccit. 



f ' Application of Dynamics to Physics and Chemistry,' p. 58. 



t Bock, loc. cit. 



