]4 



AKT. 12.-K. HONDA. S. SHIMIZU. AND ?. KUSAKAbE 



The abovt' ivsnlt.s for soft iron approximately agree with those 

 given by the method of elongation both ([iialitatively and quantitatively. 



Hilt, the specimens in these two cases are not the same, and moreover 



oE .... 

 — p— IS a function of a stress ; hence we can not, in a strict sense, 



conform these two results. Our results, when compared with those of 



j)i-inious experimenters, are markedly large, especially with small 



It^adinirs. 



Wolfram ntfcl. The change of elasticity by magnetization for 



wolfram steel is (juite similar to that of soft iron. The curves of 



depression is less asymptotic to the axis of the magnetizing force. 



The initial small decrease of elasticity is more marked than in the case 



of soft iron and occurs even with the smallest load. The followinij 



table and Fig. 4 show the character of the change of depression. 



The values of -w- are given in the following table : — 



