Magnetisation in the Length of Metal Rods. 121 



the retraction after the maximum elongation increases with the 

 external magnetising force. 



No certain indication of an approach to a limit of retraction is ob- 

 servable in the curves. Stronger magnetising forces would of course 

 show one, and I hope to be able to repeat the experiments with greater 

 battery power. 



The maximum elongation is reached by the three rods with magne- 

 tising currents which are the same in order of magnitude as the 

 diameters of the rods. 



Lastly, it appears from the curves that the amount of maximum 

 elongation is smaller when the diameter of the wire is greater. The 

 successive maxima are 7, 6, and 4'25, and if an error of a quarter of 

 a scale division be allowed, these maxima will be found to be in- 

 versely proportional to the square roots of the diameters of the re- 

 spective wires. 



7 x -/265 = 112 

 6 x v/365 = 114 

 4'25x </625=106* 



It seems to me difficult to account satisfactorily for this variation 

 of the maximum elongation. It is of course easy to understand why 

 a greater external magnetising force should be required to produce a 

 given intensity of magnetisation in a thick rod than in a thin one. 

 But it is not at first sight at all evident why, when the same mag- 

 netisation is produced, the elongation should not be the same in both 

 cases. Possibly my results may be due to a mere accident, such as a 

 difference in the qualities of the three specimens of iron ; but their 

 apparent regularity renders such an explanation somewhat improbable. 



It seemed extremely desirable that, if possible, a connexion should be 

 established between the point of maximum elongation and some defi- 

 nite phase of the magnetisation of the rod. Much time and labour 

 were spent in endeavours to investigate the magnetisation by a method 

 of induction ; but probably owing to the fact that the galvanometer 

 used — one of Elliott's Thomson galvanometers, with the usual astatic 

 system of magnets and an aluminium damping vane — was unsuited 

 for the purpose, no results of any value were obtained. I was more 

 successful in an attempt to measure by a deflection method the rela- 

 tive values of the temporary moments which various magnetising 

 currents produced in the three rods. The coil was placed in a 

 horizontal position, and one of the rods inserted in the tube, where 

 it was supported axially by means of corks at the two ends. A re- 

 flecting galvanometer was placed at a suitable distance from it, as 

 determined by preliminary trials, and the height and disposition of 

 the galvanometer were so adjusted that its magnet was on a level with 

 * If the last elongation liad been 4"5 the product "would have been LIS. 



