BURGESS AND KELLBERG: CRITICAL RANGES OF IRON 439 



In figure 2, the observations on heating are represented by- 

 circles and on cooling by crosses. The shift of the heating curve 

 with respect to the cooling curve appears to be real as shown, 

 since the iron returns exactly to the same resistance at 0°C. after 

 heating. This non-coincidence of heating and cooling curves is 

 probably caused by the different rates of heating and cooling, 

 the former being about 0.10 deg. / sec. and the latter 0.06 deg. / 

 sec. at 900°C. 



It is seen from figures 2 and 3 that the resistance of iron in- 

 creases from 0°C. without any anomalies — except possibly a 

 minute one at 730°C., due to less than 0.01 per cent of carbon — 



Te/njberafure Co^ficisnt 

 of Xmn. 



JT. 



ZOO' 



Jl»* 



WX)' 



sotr 





Temperature Centiyrode 



Fig. 3. Temperature coefficient of pure iron. 



with a gradually increasing temperature coefficient to above 

 650° or until the neighborhood of A^ is reached. As At is ap- 

 proached the resistance rises rapidly, and at Ai there is an in- 

 flexion in the resistance-temperature curve shown as a cusp 

 at 757°C. in the temperature-coefficient curve. At Ac,, the resist- 

 ance of iron falls abruptly by some 0.005 of its value, which is 

 recovered within a 25° interval, and above Ac^ increases gradually 

 again. On cooling the reverse phenomenon is observed at 

 Ar3, which is accompanied by a slight increase in resistance with 

 falling temperature, preceded by an interval of relatively slight 

 changes in resistance. These effects are shown best in the 

 open scale plot in figure 2 of the A3 region and in figure 3. 



