1919] on The Hardening of Steel 487 



band of devoted research workers who have been the pioneers in 

 founding the modern science of metallography. 



The application of the microscope to the study of the constitution 

 of iron and steel, regarding these materials as igneous rocks, is due to 

 Henry Clifton Sorby, of Sheffield, who died only a few years ago, 

 while the use of the thermo-electric pyrometer as a means of studying 

 the changes of internal energy in the same materials is the work of a 

 Frenchman who is still living — M. Henri le Chatelier. The methods 

 opened up by these pioneers have proved extremely fruitful, and the 

 results achieved by their systematic exploitation have brought certainty 

 into a field where previously only speculation reigned. With these 

 results it is necessary to deal briefly. 



It will be convenient first to consider the evidence afforded by 

 accurate pyrometry, although historically the microscope led the way. 

 It is now known that the purest iron which can be obtained, containing 

 as it does not more than three parts of impurity per 10,000, can exist 

 in at least three different modifications, and there is every reason to 

 believe that these modifications are a property of the iron and not in 

 any way due to the small amount of impurities mentioned. If the 

 cooling of a sample of such iron from the melting point (1505° C.) is 

 followed by means of a delicate pyrometer and potentiometer, the 

 cooling is found to be normal for about 600°. At about 900° C, 

 however (Fig. 3), there is a sudden and large evolution of heat which 

 completes itself within a few degrees and is sufficient to arrest the 

 fall of temperature completely for a time ; after this the cooling 

 proceeds for an interval, and then another evolution of heat occurs. 

 This is different in character and amount from the previous one ; it 

 is markedly smaller in quantity and is spread over a considerable 

 temperature interval, which ranges approximately 78<»°-740° C. In 

 the latter case the evolution of heat is never sufficient to arrest the 

 fall of temperature completely. Below this point there is no further 

 abnormality in the rate of cooling to the ordinary temperature. 

 These two heat evolutions were rightly regarded by the early workers 

 as marking critical points in the history of the iron ore cooling, and 

 were called by their discoverer, the late M. Osmond, Arg and Aro. 

 From the freezing point down to Arg the iron is known as y, 

 between Arg and Aro as ^, and below Ar2 as a. These three varieties 

 of iron have characteristic properties, of which, from the point of 

 view of the argument of this lecture, the following should be noted : 

 y iron is completely non-magnetic and has a considerable power of 

 dissolving carbon, the amount of which varies with the temperature ; 

 it has also a characteristic crystalline form. yS iron is also non- 

 magnetic, but its capacity for dissolving carbon is almost nil, and it 

 also has a characteristic crystalline form, a iron, on the other hand, 

 is magnetic, but in all other respects it resembles ^ iron very closely, 

 and no difference in crystalline symmetry between these two varieties 

 has yet been discovered. A considerable controversy has raged for 



YOL. XXII. (No. 113) 2 L 



