46 



K. TSURUTA. 



that, in the latter case, agitation has had its effects before the loading 

 or unloading takes place, while in the former case the conditions so 

 altered remain undisturbed during the loading or unloading. The 

 present case may, therefore, be regarded as equivalent to the preceding 



case with the effects of the hysteresis added. Now pulling or 

 relaxing the wire may probably produce to a certain extent, though 

 not quite to the same degree, similar effects as agitation itself in 

 shaking out the hysteresis, and moreover the effect of strengthening 

 or loosening the molecular ties holding the molecules of the iron in a 

 stable equilibrium under the load present. The former of these effects 

 will carry the representative point on the " on " branch downwards 

 (a and c), and on the "off" branch upwards (/) and d). The latter 

 effect will carry the representative point in case of pulling upwards 

 (a and />), and in case of relaxing downwards (c and d). Therefore, 

 in (a) and (d) the two effects conflict with each other, while in (c) 

 and (b) they are concurrent. If we assume the former effect to be the 



