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



Investigations upon this subject have also been made by 

 Berget*, Nagaokaf, LochnerJ, Jones §, and Bock||. 



For convenience I have collected in a summary the results 

 obtained by the different observers. 



Soft Iron. 



Soft iron elongates when magnetized. The elongation 

 attains a maximum, and then diminishes with increasing 

 strength of field until a state is reached when the rod returns 

 to its original length. Further increase of field causes the 

 rod to contract. 



There is no minimum length, the rod approaching asymp- 

 totically a limiting value. 



For a given strength of field, both hardening and annealing 

 diminish the elongation and increase the contraction shown by 

 the rod before it was subjected to these operations (Bidwell). 



Tension also diminishes the elongation and increases the 

 contraction of the rod. 



For a sufficiently great tension no elongation occurs, the 

 rod contracting upon the application of the smallest mag- 

 netizing forces. 



For a given length, the effects both of elongation and of 

 contraction are greater for thin than for thick rods (Bidwell) . 



S. J. Lochner^F comes to the conclusion from his own 

 experiments that the converse is true — that thick bars give 

 greater expansion than thin ones. 



Very little reliance can be put in these last experiments, 

 and the dependence of the change of length upon the ratio of 

 the length to the diameter cannot be inferred from them. 

 Bidwell used three rods, 10 centimetres long and 2*65, 3*65, 

 and 6*25 millimetres in diameter, and assumed them to be of 

 iron of similar composition. It is well known that different 

 specimens of iron, apparently similar in structure, give results 

 that vary 25 per cent, and more. So that the small variations 

 in the change of length noted by him cannot safely be said to 

 be due to the differences in their diameters, especially as he 

 made no determination of the permeability. Lochner avoided 

 this error by testing an iron rod, and then, after having cut 

 off a portion, testing it again. He, however, took no pre- 

 cautions to have the field uniform. His solenoid was nearly 



* Compt. Rend. torn. cxv. p. 722. 



t Wied. Ann. liii. pp. 481 & 487 (1894). 



t Phil. Mag. vol. xxxvi. p. 504 (1893). 



§ Phil. Mag. vol. xxxix. p. 254 (1895). 



|| Wied. Ann. 1895. 



11 Lochner, loc cit. 



