and Electrical Resistance of Iron. 231 



These values of -™- are of the same order of magnitude 



in all the sets of curves of both series taken within 50° of the 

 critical temperature, and they seem to represent an approxi- 

 mately constant property of the iron in this region. 



The area enclosed by each hysteresis loop has been summed 



up and divided by Air, giving the quantity — jHrfB, the 



work done in ergs per c.cm. of the iron by the magnetic 

 forces during the performance of one cycle. The results are 

 given beside the corresponding temperatures in the Tables 

 numbered X. {a and b), and are plotted in curves 14-16, 

 which show the way in which the hysteresis diminishes to zero 

 as the iron is heated up to the critical temperature. 



No evidence of hysteresis above this temperature could be 

 experimentally obtained, a result which one might be led to 

 expect from the fact that the permeability at these tempera- 

 tures is the same in all magnetic fields. 



A comparison of the two curves No. 15 shows the marked 

 diminution of hysteresis at all temperatures which resulted 

 from annealing at 1150° instead of at 840°. The curves ex- 

 pressing the hysteresis for cycles within definite limits of 

 induction (Curves 16 I. & n.) are not comparable, as the 

 limits during the later series were, owing to oxidation, greater 

 than during the former. In order to render them comparable 

 a third curve has been dotted in, for which the ordinates of I. 

 were multiplied by a constant I'll (the assumed ratio of the 

 hysteresis when B= +4260 to that when B= +4550, which, 

 according to results of many observers, cannot at ordinary 

 temperatures be far wrong). Comparing then curve I. 

 with curve ill. an idea is obtained of the extent to which the 

 hysteresis can be diminished by annealing at the high tem- 

 perature 1150°, instead of at the lower one 840°. The com- 

 parison is given on Table X. c. 



Table X. d. shows concisely the various stages of annealing 

 of specimen B. The hysteresis at ordinary temperatures for 

 B= +4550 was originally 1480 ergs per cub. centim. per 

 cycle. Prolonged annealing at 840° reduced this to about 

 800 ergs ; whilst annealing at 1150° brought the hysteresis 

 down to 612 ergs. 



In a paper read before the Institution of Civil Engineers 

 about a year ago*, Prof. Ewing described some tests which 



* Ewing, " Magnetic Testing of Iron and Steel," Journ. Inst. C. E. 

 May 1896. 



S2 



