110 



*arus — Apparent Hysteresis. 



respectively about half as large for L or twice as large for C as 

 the second deflections due to the making of the field. Figure 

 21 also shows that if the residual strain or slip is wiped out for 

 a longitudinal field it is also wiped out for a circular field. In 

 cases L there is increase of rigidity and decrease of viscosity ; 

 in cases C decrease of both rigidity and viscosity. The dif- 

 ferential and summational results are well given in the figure. 

 In cases L' and C there is increase and decrease of rigidity 

 only with no viscous effect. 



To sum up the argument it is briefly this : Magnetization is 

 regarded as a means of shaking up the molecular mechanism 

 and thus to produce temporary molecular instability or momen- 

 tarily very low viscosity. Hence if mechanical strain has 

 been stored up in the metal it will be instantly released and 

 the metal will correspondingly yield to external stress. This 

 view seems to give a good account of all the hysteresis-like 

 phenomena met in torsional magnetostriction and makes it 

 possible to describe the behavior of all paramagnetic metals in 

 a single explanation. 



The one quantity by which the experiments seem to be 

 characterized is the cyclic slope, or the mean slope determined 

 by the temporary (viscous) and the permanent (elastic) deflec- 

 tions, which increases (numerically) with the strength of the 

 field and is nearly independent of the twist imparted, remem- 

 bering that the remarks have no meaning beyond the elastic 

 limits. It would seem, therefore, that in terms of this parameter 

 the phenomena as a whole for any field or metal, may be sim- 

 plified. Whether the slip constitutes a kind of molecular 

 " back lash," and whether there is any meaning to its obvious 

 analogy to the action of a "coherer" cannot here be answered. 



Brown University, Providence, U. S. A. 



