(52 Art. 9.— T. Terada : 



we obtain 0'25. The corresponding value of x in Fig. 21 is 

 found to be nearly 17. Hence, we must liave, assuming for 

 simplicity's sake /^=1, 



2H-^17x 180 = 3000; 



, -. 1530 2-10 • . 1 1 AT • +1 



lience ^''= , or — - — \\\ round number. JNow assunnng the 



k Ic 



conductivity of tlie earth to l)e 10"'^ as is suggested by the result 

 of Schuster's investigation, we ol)tain 



;. = 1-2 X 10^ cm. or 1200 km., say. 

 Nearly the same value may l)e oljtained, if we take 10'" waves 

 instead. We have no theoretical ground at liand for assuming 

 tliat the wave length varies with the periods; still, judging from 

 the remarkable resemblance of the two diagrams above compared, 

 it seems plausible to assume provisionally that the wave length 

 is at least of the same order of magnitude, and the different 

 periods are determined not so much by the difference of the 

 wave length as of the velocity of propagation. In the above 

 example, where i'.= r2x]0' km. for T=180' the velocity of 

 propagation v will be about 7 ^^^ The above estimation was 



solely based on the assumption that the equations (11), (12) etc. 

 hold rigorously. Judging from the analogy of the case in- 

 vestigated hy Schuster, it is probable that the actual reduction, 

 of the vertical component will be decidedly more remarkable 

 than the calculated value. If such be the case, the value of x 

 and accordingly the values of X and v will become less than those 

 above estimated. 



A closer examination of Fig. 12 sliows that the asymptotic 

 value of AZjàX,n is not 1, but somewhat near 0*8. If we compare 

 Fig. 12 with a diagram obtained b}^ multiplying the ordinales 

 of Fig. 21 by 0*8, tlie acceptable value of x become about 10 

 instead of 17. But this does not alter the order of magnitude 

 of the results. 



1) According to Birkeland, the lateral velocity of the current producino- some dis- 

 turbances was of the order of 100 km. per minvite, while that of some auroral bands observed 

 in the polar region was about 300 m. per sec, which is very small in comparison with the 

 above calculated velocity of the hypothetical current. 



