Magnetic Properties of Iron and Nickel. 359 



The temperature was determined by a thermo- junction of 

 platinum and platinum-iridium, one millimetre on the 

 galvanometer scale corresponding to about 2° centigrade o£ 

 temperature. 



(4) The first step in the experimental work was the choice 

 of a suitable strength of alternating current. Curves of I 

 and H were constructed for alternating currents of gradually 

 increasing strength and one virtual ampere was found to be 

 sufficient to reduce the hysteresis, in iron, to a negligible 

 quantity without appreciably lowering the intensity of 

 magnetization at ordinary temperatures. This strength of 

 current gave a maximum alternating- field at the circum- 

 ference of the iron wire of about 9 c.G.s. units, the field, of 

 course, becoming less towards the centre. The effect of 

 variation of frequency was tried and was found to be so small 

 for an increase or decrease of 50 per cent, that any small 

 change in the speed of the alternator was quite unimportant. 

 Throughout the experiments a frequency of 33 alternations 

 per second was maintained nearly constantly. 



The suppression of hysteresis is more difficult in nickel than 

 in iron, and a current of 2 virtual amperes only partially 

 brought the magnetism to the anhysteretic state, and even 

 currents twice as strong as this did not suppress hysteresis 

 altogether. 



(5) In connexion with this part of the investigation a 

 series of experiments was made to trace the effect of in- 

 crease in the strength of the alternating current on the 

 intensity of magnetization of iron. The results are exhibited 

 in the first diagram, I being plotted vertically against the 

 strength of the alternating current i a which was carried as 

 far as 10 virtual amperes. Under the action of strong- 

 currents the wire had to be vigorously cooled with a current 

 of cold water in the way referred to above. Constant fields 

 of 0*55 to 16*6 c.G.s. units were applied in steps, while the 

 alternating current was varied in strength at each step. The 

 curves are of the hyperbolic type, and apparently an infinite 

 current would be required to reduce the magnetic intensity to 

 zero. At the beginning when no alternating current is running 

 hysteresis is in unimpaired action and the intensity is low for 

 a given very small force, but as the alternating current grows 

 the freedom conferred upon the molecular magnets allows 

 them to obey the applied field, and the intensity increases and 

 reaches a maximum when i a is about 0*5 ampere; for greater 

 values of i a the magnetic intensity smoothly declines at a 

 diminishing rate. If the alternating current is now steadily 



