1919] on The Hardening of Steel 503 



established. Accordiiiu" to the view put forward by us in 1914, mar- 

 tensite is simply twinned austenite produced by this internal movement, 

 and further along the twinning planes amorphous material is produced 

 by the distortion of the crystalline units, which material we regard as 

 corresponding to Beilby's hard vitreous-amorphous phase produced 

 by mechanical work. In order to establish this theory therefore 

 evidence should be produced that these actions take place. For such 

 evidence in full the original paper must be consulted, and only the 

 briefest resume can be given here. The view that internal straining 

 occurs during the quenching of steel has only been challenged by one 

 authority, Dr. Eosenhain, who says : "This view meets with the insuper- 

 able difficulty that although quenching does set up severe stresses in 

 steels, it does not cause any serious flow or movement ; the strain 

 hardening of metals only becomes marked when severe plastic flow 

 has occurred." This objection however breaks down on examination. 

 It is true that in cold working a metal a large degree of strain is 

 needed to produce great hardness, but in this case the direction of 

 strain is constant throughout the mass, so that the strains are additive, 

 and these result in a great change of outer dimensions. But the 

 strains in the slipping planes set up by hardening from quenching are 

 quite different. They are not in constant but in opposite directions 

 along different internal planes, and they therefore offset each other 

 and do not deform the mass as a whole. That severe straining does 

 occur therefore may be accepted as established. The microscopic 

 evidence for this straining was regarded by us in our original paper 

 as being due to a great development of twins in the steel. It 

 v>-as pointed out, however, in the discussion that although twinning 

 might have occurred, this would not necessarily cause any increase in 

 hardness, since the two halves of every twin should dovetail into each 

 other, and therefore amorphous material need not necessarily be pro- 

 duced in the twinning planes. This may be so, but for the purpose 

 of this theory it does not matter whether the movement is regarded 

 as taking place along twinning or along slip planes ; in the latter case 

 it is generally agreed that amorphous material must result by the dis- 

 tortion of the crystalline units. According to this view, then, 

 martensite is austenite which has been distorted by the formation of 

 amorphous films along the slip planes of y iron. It will be observed 

 that I have already begun to use Sir George Beilby's theory of the 

 hard amorphous state in order to explain the hardening of steel by 

 cjuenching, by suggesting that amorphous material is formed along 

 the slip planes of y iron. What is now required is to show, if 

 possible, that these layers persist on quick cooling. Beilby has shown 

 that when strain-hardened metals are heated, the amorphous layers 

 re-crystallize, and the actual temperature at which this occurs varies 

 with the particular metai and with the degree of strain hardening. 

 Accordingly, if in quenching the rate of cooling is sufficiently quick, 

 the temperature of the mass will be lowered below that at which re- 



YoL. XXII. (No. 113) 2 31 



