﻿Waves of Magnetic Flux along Iron Wires. 435 



they were, to a first approximation, independent of the 

 frequency. 



More recently Perkins*, working at a frequency of 60, 

 has measured by means of a quadrant -electrometer the 

 amplitude and phase, at different points along an iron bar, 

 of the flux oscillation which had been started in the usual 

 way at the centre. The bar used was 1 metre long, 

 2*83 cm. 2 in cross-section, and observations were made over 

 a range; of 25 cm. from its centre. 



The phase was found to change less rapidly as the distance 

 from the magnetizing-coil increased, and the "velocitvof 

 magnetization " obtained varied from 



7500 cm./sec. at the centre 

 to 21000 cm./sec. at 21 cm. from the centre. 



The curve he obtained showing the relation between the lag- 

 in phase of the flux at any point and the distance of the point 

 from the magnetizing-coil agrees in general character with 

 the corresponding small portions of the curves obtained 

 by us. 



The same problem has been theoretically discussed by 

 Thomson f. He assumes that the permeability is constant 

 for all values of the induction, and takes no account of the 

 lag in phase of the induction behind the magnetizing force. 

 As the flux-density or induction diminishes with great 

 rapidity as we pass down the bar, and as both the permeabilitv 

 and the lag in phase of the induction vary within wide limits 

 for different values of the latter, the theoretical conclusions 

 arrived at on the assumption that one of these two quantities 

 is constant and the other negligible, can scarcely be expected 

 to be realized in practice. 



2. In all the experimental investigations hitherto carried 

 out on this subject, the range along the rod over which it was 

 possible to determine even roughly the amplitude and phase 

 of the flux oscillations at different points was limited, and 

 the apparent behaviour over a short range afforded a very 

 doubtful and misleading description of the whole pheno- 

 menon. No attempt, moreover, was made to determine the 

 change of form of the flux-waves as thev passed alono- 

 the bar. 



For this investigation the wave-tracer J designed by one 

 of us is peculiarly well suited, as it enables us to determine 



* H. A. Perkins, Amer. Jour. Sci. vol. xviii. p. 165 (1904). 

 t J. J. Thomson, ' Recent Researches,' p. 302. 



X T. R. Lyle. " Wave-Tracer and Analyser/' Phil. Mao - , vol. vi. p. 549 

 (1903). 



2F 2 



