506 Professor H. C. H. Carpenter [March 7, 



may delay the change to well below this temperature. He then 

 applies these general considerations to the case of steel. 7 iron 

 normally inyerts to /? iron at about 900° C. The recrygtallization of 

 work-hardened a iron is very slow below 500° C. This gives an 

 interval of 4oO°. In the case of the 0*9 carbon steel, however, this 

 range is halved, since 7 iron does not invert till about 700° C. on 

 cooling. From this it is argued that with increasing carbon contents 

 there is more likelihood of some of the amorphous phase formed by 

 the breakdown of the 7 constituent being retained than in pure iron 

 itself. Another factor also enters in which helps to promote the 

 retention of the amorphous phase. Xofc only the iron, but also the 

 iron carbide has to be considered. When the austenite breaks down, 

 the carbide molecules are regarded as remaining " closely intermingled 

 with those of the a iron." Before the amorphous material can 

 recrystallize two different kinds of molecules have to segregate, those 

 of iron and its carbide ; this segregation must be a slow process in an 

 undercooled viscous mass, and rapid cooling is regarded as allowing 

 the minimum temperature of crystallization to be passed before it has 

 had time to take place. 



Benedicks has shown that pressure aids the retention of the 

 austenitic condition in steels. The change from the crystalline to the 

 amorphous state in metals is in practically all cases accompanied by 

 an increase in volume ; 7 iron is denser than a iron, and therefore 

 even denser than amorphous a iron. Accordingly Humfrey argues 

 that after the formation of a certain amount of amorphous material, 

 the pressure set up will be sufiBcient to retain some of the austenite in 

 the unchanged condition. According to him, then, the particular 

 structure known as martensite is due to two constituents, (1) 7 iroQ 

 (austenite) ; (2) " an amorphous solution of a iron and carbide " ; 

 the form in which the two constituents occur being due to the 

 tendency of the austenite crystals to break up along their gliding 

 planes. It will be seen therefore that the conclusion reached by 

 Humfrey is almost identical with that arrived at by Professor Edwards 

 and myself, bnt that the actual mechanism of the change is different. 



As to whether on passing from one allotropic form to another it 

 is necessary to assume the existence of an intermediate amorphous 

 state as postulated by Humfrey, it is not necessary for my present 

 purpose to discuss. It may be pointed out, however, that the passage 

 from the crystalline to the amorphous condition involves the absorp- 

 tion of energy, and he has not explained how that energy can be 

 supplied from the interior of a cooling mass of material, and to render 

 the theory complete in this respect an explanation must be forth- 

 coming. On the other hand, when, as Professor Edwards and I have 

 supposed, both slipping and perhaps twinning of the austenite crystals 

 occur as a result of the quenching stresses, energy is absorbed and 

 may be stored up in the amorphous layers. 



Within the limits of this lecture it has not been possible to con- 



