On Eapid Periodic Variations of Terrestrial Magnetism. 

 àZ^ a 



which may be written = 4 , , 



_ 4:7r/j.kn _ ^/ik?-^ 



if we put 



27r 



61 



denoting the period by T^ i.e. putting ''^—-rp 



The variation of the calculated ratio AZjaX,,, witli x as altscissa, 



may be seen from the annexed figure (Fig. 21). Referring to 



Fis. 21. 



'Sx, 



fO 



zo 



30 



40 



X-^' 





Figs. 11 and 12, and comparing the present theoretical curve 

 with the curve of the observed ratio with IjT as abscissa, we 

 may observe a striking resemblance, thougli tlie ratio of the 

 scales for the two diagrams is not yet hnoAvn. If we take, for 

 example, the 3'" wave for the purpose of comparison, the 

 observed value of dZ.JdX,,, is nearly 0*28. jMaking a discount 

 of 10 % on account of tlie reason discussed at the end of § 13, 



