186 DR. S. W. J. SMITH AND ME. J. GUILD: A THERMOMAGNETIC STUDY OF 



The contribution of the carbide to the magnetisability of the material need not be 

 considered at high temperatures because it is known that it is relatively very feeble 

 above 250 C.* 



6. Elimination of Hysteresis in Strong Fields. 



Another significant difference between the curves in weak and strong fields is 

 observable, particularly below 690 C. 



In the former, the falling temperature observations lie increasingly above those for 

 rising temperatures as the temperature is reduced ; in the latter, the corresponding 

 observations practically coincide. This difference is no doubt an effect of the same 

 cause as that to which ordinary magnetic hysteresis is due, namely, the existence of 

 various molecular groupings which make alignment difficult in weak fields but which 

 are broken up when intense fields are applied. 



Weak fields, although increasingly aided by thermal agitation as the temperature 

 rises, are unable during heating to break up all of these molecular groupings. But 

 the magnetism reappears during cooling in a medium in which, after heating to a high 

 temperature, the molecular groupings are less extensive than before. Hence a greater 

 intensity of magnetisation will now tend to arise under the same field as before, 

 because the force required to maintain a given degree of alignment is less than that 

 required to induce it against already existing groupings.! 



In strong fields, on the other hand, the resistance to alignment is determined 

 mainly by the thermal agitation. This is the same at a given temperature whether 

 that is approached from above or below. Consequently the coincidence of the heating 

 and cooling curves in the field of 200 C.G.S. units, except over the region where the 

 lag in the transformation of the eutectoid occurs, is an indication that the observations 

 taken continuously in this field are practically free from the effects of hysteresis. 

 This means that they are practically the same as they would have been if the 

 material had been demagnetised between successive applications, at different tempe- 

 ratures, of the field in question. 



It therefore appeared that it would be possible to make quantitative use of curves 

 obtained in strong fields in the way indicated below. 



7. A Method of Estimating the Percentage, Composition of the Eutectoid. 

 Considering fig. 5, we may suppose that the contribution of the iron in the 

 eutectoid, to the total magnetisation at 697 C. during heating, is represented by the 

 distance between the point P on the upper curve at that temperature and the 

 corresponding point Q on the lower curve at the same temperature. In the one case 

 the eutectoid has not yet begun to lose its magnetism ; in the other it is just about 

 to regain it. 



* Loc. cit. XXIV., p. 63, 1911. 



t Such considerations are accentuated by the fact that, although the solenoidal field remains constant, 

 the field within the specimen increases as its magnetisation falls. 



