PEIRCE. MAGNETIZATION IN IRON. 129 



while the primary circuit is attached to the alternate current mains, 

 the secondary coil of the transformer be slowly drawn off the core 

 and the primary coil, by help of some mechanical device, it is possible 

 to send through the solenoid a long series of cm-rents alternating in 

 direction and gradually decreasing in intensity and thus to demagne- 

 tize the iron rod very well. We had an apparatus of this kind perma- 

 nently connected with our apparatus, but it was not shown in Figure 

 1 lest the diagram be too complex. 



When the direction of a strong electrical current in the circuit of 

 the large solenoid (S) in which the iron rods to be tested were magnet- 

 ized, was suddenly reversed, some time was needed to establish the 

 new current in its full value, and the change in the magnetic flux 

 through the test coils wound upon the rods was not complete until 

 after several seconds. This fact, due to the large inductance in the 

 circuit, made it unsafe to employ a ballistic galvanometer of ordinary 

 type for measuring this flux change, and we had recourse to a long 

 period instrument of a kind which has been used for a number of years 

 in the Jefferson Laboratory. The particular galvanometer (G) we 

 chose, had a period of 156 seconds which was quite long enough for 

 our purposes, but we had a much more slowly moving instrument at 

 hand in case of need. Any fairly long throw of G could be determined 

 with an error of less than one tenth of one per cent, and we could do 

 better than this by careful repetition. G is shown in Figure 4, Plate 1. 



The main currents in the solenoid circuit were measured with the 

 help of a series of Weston Ampei*emeters (two of which are shown 

 diagrammatically as U and V in Figure 1) properly arranged for the 

 special intensity ranges we needed, but the accurate determination of 

 large currents was made by aid of a potentiometer (Figure 5) with 

 standard cadmium cells, which measured the potential drop across a 

 standard one hundredth of an ohm resistance (R) by Crompton, which 

 had been tested against another standard by Wolff. The largest 

 currents we used could not very well be allowed to run very long 

 through the coils because the amount of heat set free in the circuit 

 was enormous. Indeed, with an energy expenditure of more than 

 fifty kilowatts, the heating problem, in spite of running water in the 

 core of the solenoid needed careful consideration. As a matter of 

 fact, the only difficulty we finally encountered was a slight falling off 

 of our largest currents with repeated throws, owing to a little increase 

 in the resistance of the circuit, and this came at a place where the flow 

 of inductance through the test coil changed very slowly with H. 

 To save time we arranged a standard condenser (Elliot Brothers, 



