Oil the Magnetic Deformation of Nickel. 49 



A comparison of Tables II and IV, or III and V, shows again the 

 nature of the influence of temperature on magnetisation and contrac- 

 tion, but as the measurements at the two temperatures were not 

 made quite at the same time, the results probably do not accurately 

 represent this influence. 



If the results given here are compared with those given in the 

 former paper, it will be seen that the magnetic contraction at any 

 field observed in December, 1897 (Table III), is practically identical 

 with the former value, but that the magnetisation curve is very 

 different from the former one. The present magnetisation is much 

 greater at low and medium fields, but about the same as before at 

 higher fields. Further, the calculated value of the contraction is, 

 especially at medium and high fields, considerably less than before. 

 Hence the "corrected" contraction (a Z/Z) cannot now be the 

 same function of the magnetisation as before ; it is, in fact, now much 

 more nearly proportional to I 6 , as the last two columns of Tables III, 

 V, and VI show. 



It is impossible to say how much this discrepancy is due to the 

 slow change which appears to be always going on in the magnetic 

 properties of the nickel wire. In the former experiments the mag- 

 netisation was determined first, and a short time after the elongation 

 was observed, but no allowance was made for any change of magnet- 

 isation which might have taken place in the meantime. Still, it is 

 improbable that that would entirely account for the discrepancy. 



Some of the above results are shown graphically in figs. 1, 2, 3. 

 In fig. 1 the difference of ordinates of the highest and lowest curves 

 for any value of the field H represents the effect on the magnetisa- 

 tion of changing the load from 1'4 kg. to 8'4 kg., after the additional 

 load of 7 kg. has been applied and removed, and the field reversed, 

 several times. The intermediate curve is the magnetisation curve of 

 increasing reversals for the mean load 4'9 kg. These curves were 

 determined in January, 1898. 



In fig. 2 the curves represent, as functions of the field, the observed 

 contraction (December, 1897), the calculated contraction (mean of 

 November, 1897, and January, 1898), and the corrected contraction, 

 i.e., the difference between the observed and calculated contractions, 

 The calculated contractions of November and January are repre- 

 sented by the points -f + + and Q O respectively. The 

 observed contraction curve is practically the same as that given in 

 the former paper, but the calculated and corrected curves show 

 considerable differences. 



In fig. 3 the points marked + -f + . . represent the corrected con- 

 traction as a function of I 6 (Table III), and these points lie nearly 

 on a straight line through the origin. The temperature during all 

 the experiments represented by these curves was 10 C. 



