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



4. Comparison of Cooling Curves. 



Turning now to the results obtained during continuous cooling, collected in fig. 4, 

 it will be seen that the temperature of rapid return of magnetism in the eutectoid 

 is no longer constant except perhaps in the hyper-eutectoid steels. In the hypo- 

 eutectoid steels, the temperature of rapid return appears to become continuously 

 lower as the percentage of carbon falls. Although ^his is true in general, it would 

 appear (from the present and other observations) that the behaviour of the different 

 steels is not quite as regular during cooling as during heating. 



The conditions under which the eutectoid forms during cooling are therefore 

 apparently less simple than those under which it dissolves during heating. 



Apart from surface phenomena, which probably exert an appreciable retarding 

 influence during heating as well as during cooling, it is important to remember that 

 whereas the transformation of the eutectoid precedes that of the excess iron during 

 heating the opposite is true during cooling. 



Since the solution expands when it changes into the eutectoid (with evolution of 

 heat) it is easy to see that, especially in the alloys weak in carbon, the pressure 

 exerted during the transformation, by the enveloping excess iron, may be a cause of 

 retardation which is present during cooling but is absent during heating. 



Experimental evidence concerning the retarding forces operating during cooling is 

 given below. 



5. Comparison of Results Obtained in Different Fields ivith the 0' 15 per cent. 



Carbon Steel. 



In the experiments which have been described the specimens were submitted 

 continuously to a constant field of 50 C.G.S. units. 



Fig. 5* shows the behaviour of the 0'15 per cent, carbon steel in similarly applied, 

 but weaker and stronger, fields of about 25 C.G.S. and 200 C.G.S. respectively. 



The ordinates represent intensities of magnetisation, in arbitrary units, as before. 

 The scales for the two fields are, however, very different. Thus the intensity of 

 magnetisation in C.G.S. units at 680 C. was actually about 13 times greater in the 

 stronger field than in the weaker. 



Observations taken during cooling are represented by crosses in each case. 



The effects, upon the magnetisation-temperature curve, of loss and gain of 

 magnetisability of the eutectoid component of the steel are much less pronounced in 

 the field of 25 units than they were in the field of 50 units. For example, the return 

 of magnetisation near 700 C. is now only just perceptible. 



* In order to exhibit the parallelism between the time-temperature (" inverse rate ") method of 

 determining the eutectoid point and the thermomagnetic method, fig. 5 includes curves showing how the 

 rate of variation of the intensity of magnetisation c/ with respect to the temperature 6 depends upon the 

 value of 6. The ordinates represent differences between values of ^/ (in arbitrary units) measured at 

 intervals of 4 5 C. 



