556 On an Investigation of Young's Modulus for Wires, 



current makes the wire yield to the stress. (The apparatus 

 is now being modified to avoid heating-effects, and it is hoped 

 that any irregularities due to that cause will be removed.) 



It seems reasonable to suppose that if the act of magneti- 

 zation and demagnetization causes the wire to yield to the 

 stress, then on applying an alternating current to the solenoids 

 a considerable elongation would result ; this was accordingly 

 tried, and the immediate effect was an elongation, but as 

 soon as the current was stopped the wire steadily contracted 

 to its original state. The elongation began rapidly and then 

 slowed down, and finally stopped, while the current was still 

 on. This elongation seems to have been due to the heating 

 effects of hysteresis, for on substituting copper wires for the 

 iron no such effect occurred. 



This suggested an interesting method of getting the area 

 of the hysteresis-loop. The alternating current is put on and 

 the elongation measured by counting the bands which pass 

 until the latter have come to rest. Then a current (C) is 

 passed down the wires themselves until a steady elongation 

 of the same amount is produced. Then if we assume that 

 the energy supplied in the two cases to the wires is the same, 

 and we know the number of alternations per second, n suppose, 

 of the current in the solenoids, we can put 



kJHdI = C 2 R=CE, 



where J Hcil is the energy dissipated at each cycle of mag- 

 netization, I the length of the wire, and R is the resistance of 

 the wires. 



The following is an actual example for soft-iron wire: — 



= 1*2 amp., 



71=80, 



E = 2'5 volts, 



/ =75 cm., 



*KdI= \l Xl °* X ?' 5 =5000 ergs. 

 10x80x75 ft 



This neglects the heating due to Foucault currents, which 

 could be re luced by lamination, though the absence of such 

 elongation in copper wires seems to show that the heating due 

 to this cause is not very considerable compared with that due 

 to hysteresis. (But of course in copper the Foucault heating- 

 would be only - of that in iron wire.) 



In conclusion I wish to thank Prof. J. J. Thomson for 

 valuable criticism and kindly interest, and Prof. Poynting 

 for encouragement and suggestions ; also Mr. A. Goodliffe, 

 of King's College, for kind assistance in some of the obser- 

 vations, and Mr. W. G. Pye, of the Cavendish Laboratory, 

 for help in construction of the apparatus. 





