982 



THE BELL SYSTEM TECHNICAL JOURNAL, OCTOBER 1951 



ing the nickel rod. Saturation is not quite obtained so that the A£ effect 

 measured is shghtly lower than the true value, but for relative frequency 

 comparisons this is not important. 



The first rod measured was 0.320 cm in diameter and 10.16 cm long and 

 was annealed at 1100°C. Five frequencies ranging from 22.5 kilocycles were 



0.14 



12 



2 0.10 

 2 o 

 Si^O.08 



o z 



/) *^ 0.04 



0.02 



25 



175 



50 75 100 125 150 



MAGNETIC FIELD STRENGTH, H, IN OERSTEDS 



Fig. 10 — Typical change in velocity and decrement of a polycrystalline rod as a func- 

 tion of the magnetizing field. 



20 



40 60 80 100 120 140 



FREQUENCY IN KILOCYCLES PER SECOND 



160 



Fig. 11 — Fractional change in Young's modulus, and the decrement, plotted as a func- 

 tion of frequency for rod No. 1 . 



used and the ratio of the change in Young's modulus to the value of Young's 

 modulus for the demagnetized rod is shown plotted in Fig. 11. This figure 

 shows also the decrement 5 = -k/Q. It is obvious that the decrement even- 

 tually decreases as the frequency rises, and this is contrary to the simple 

 theory of the micro-eddy-current effect,^ which indicates that the decrement 



