176 Trowbridge and Adams — Circular Magnetization, 



currents was a constant. The frequency used by Klemencic 

 was about 9 X 10 7 oscillations per second. 



The experiments described below are with oscillatory cur- 

 rents of frequencies ranging from 600 to 3000, with maximum 

 field strengths of higher values than those used by Klemencic. 

 It appears from the results given below that with oscilla- 

 tory currents of as low frequencies as this, the permeability is 

 not a constant, but depends upon the strength of the magnetic 

 field ; that is, the iron behaves toward oscillatory currents in 

 much the same manner as it does toward steady currents. 



Similar results were brought out recently in an article by 

 Marchant.* He showed that the same dependence of per- 

 meability to field strength holds in the case of oscillatory dis- 

 charges as in the case of steady currents. No figures are given 

 in his article by means of which a quantitative idea of this 

 relationship may be obtained. 



The arrangement of apparatus used in our investigation was 

 as follows : B is a battery of 10,000 storage cells ; C a con- 



I'k 



wn 



denser consisting of 300 glass plates, coated on both sides with 

 tin-foil, the coated surface being 43X33 centimeters. The 

 battery circuit was controlled by a switch at E. The discharge 

 circuit, CADC, contained a spark-gap at D, of cadmium termi- 

 nals. The self-induction of this circuit could be changed by 

 placing coils of copper wire of different dimensions at A. 



When the switch at E was closed the condenser was charged ; 

 it then discharged itself through the circuit CADC. This 

 latter circuit was so proportioned that the discharge was 

 always of an oscillatory character, that is, approximately, by 

 the well-known theory, R, 2 was less than 4L/C, where R is the 

 resistance, L the self-induction, and C the capacity of the 

 circuit. The self-induction of A was large enough so that the 

 self-induction of the rest of the circuit could be neglected ; 

 and the electrostatic capacity of the circuit could also be neg- 

 lected in comparison with the capacity of the condenser. 



* Nature, Aug. 30, 1900, p. 413. 



