Magnetization of Soft Iron Bars. 453 



experiment was carried out in the Physical Laboratory of the 

 then Tokyo Daigaku. It was very similar to the present one 

 in its nature, conducted, however, in a different way, in so far 

 as the magnetizing field was kept constant while the mass of 

 iron magnetized was increased by steps. 



The arrangement of the magnetizing solenoid and the 

 magnetometer was the same as in the present experiment. 

 The dimensions of the solenoid were : — length 9*25 centim.; 

 external diameter 2 centim.; internal diameter 0*86 centim.; 

 number of turns of wire per centimetre 132*6. The length 

 of soft iron wire was 9 centim. and its diameter 1 millim. 



In the first experiment the magnetizing field of 46 C.G.S. 

 units was kept constant, and the soft iron wires were put into 

 the solenoid one by one. At first these wires were sucked 

 into the solenoid very vigorously, and the magnetometer 

 readings were very nearly proportional to the number of wdres 

 in the solenoid. But when the number of iron wires w T as 

 about 10, some of the wires began to throw themselves out of 

 the solenoid, and even when pushed in again with a finger 

 would jump out upon its removal.* They were, however, 

 capable of being kept in their normal position by a careful 

 adjustment. Obviously every one of the wires is under two 

 forces — one, that due to the solenoid, tending to keep the wire 

 in, and the other, due to the magnetism of its fellow wires, 

 tending to push it off: consequently when the latter becomes 

 greater than the former, the wires will be in an unstable state 

 (such a state might be imagined to be taking place between 

 parts of a solid bar). It is interesting to see that, after this 

 event, the magnetometer readings fall rapidly short of being 

 proportional to the number of wires in the solenoid, and after 

 the number was increased to about 25 the magnetometer 

 readings remained practically constant, even if the w r ires 

 were increased to as many as 41. 



Figs. 8 to 14, PI. V., represent the results of this experi- 

 ment. Fig. 8 shows the increase of magnetic moment due 

 to the increase of iron wires, and fig. 9 the mean intensity of 

 magnetization in the whole substance of the iron. 



Similar experiments were made for the constant mag- 

 netizing fields 50, 11, 5"5, 2'75 C.G.S. units. These are 

 drawn in arbitrary scales for the want of the original notes at 

 the present. By comparing these, however, with the curves 

 for field = 46, one can form a fair estimate of what they 



* After the number of wires had been increased to 38, they did not 

 jump out even if slightly disturbed from the normal position. But I was 

 unable to tell whether this was a magnetic effect, or simply a mechanical 

 one due to the increase of friction. 



