62 Messrs. K. Honda and T. Terada on the Change of 

 Table XXIV. (continued). 



Tungsten- Steel. 



T=439 gr. 



T=ioae g r. 



T = 1810gr. 



II. 





ii. 



BE 



» xio-. 



14-0 



-001. 



lfrl 



-001 



18-2 



-0-04 



194 



-0-04 



18-2 



-0-03 



20-5 



-008 



23-1 



-0-17 



224 



-0-07 



23-0 



-0 07 



25-4 



- 01)8 



25«6 



-0-03 



384 



4-1-24 



36-6 



-fl'6l 



302 



+0-80 



->(■>•;> 



4-148 



tt8-l 



4-1-89 



4L-1 



+ 1-34 



75 7 



4-1-58 



91-9 



+ 2-02 



68-5 



+ 1*7 



144-3 



4-1-61 



270 



+ 2-12 



224 



4-1-88 



230 



4-1-66 



387 



4-1-99 



348 



4-1-81 



361 



4-1-69 



486 



+ 210 



472 



4-1-78 



460 



+ 1-69 



In the former results, the value of -„ depended consider- 

 ably on the suspended weight ; but in our recalculated results 

 it was only slightly affected by the weight, except in the 

 case of nickel in weak fields, where its initial decrease was 

 considerably reduced. 



For Swedish iron and tungsten-steel, the results somewhat 

 resemble those of the present experiments deduced from the 

 magnetic elongation. But they are several times greater as 

 compared with the results of the direct experiment. In 

 nickel, the course of the curve is considerably different, from 

 that of the present experiments ; its form is, however, very 

 interesting, if we consider it in connexion with that of the 



curve K to H. It is here to be noticed that as regards 



magnetic quality, the specimen in the present experiments 

 was very different from that used in the former experiments, 

 so that the quantitative comparison of the two results is of 

 little value. 



§ 3. Experiments on the Change of Rigidity 

 by Magnetization. 



(a) Measurement of the Change of Rigidity by the 

 Oscillation Method. 



One of our methods consisted in giving torsional oscillation 

 to the wires magnetized with different fields and calculating 



