52 Art. 9.-T. Terada: 



either vertically, horizontally or in a rotatory manner. Secondly, 

 it is also possible that tlie cnrrent itself undergoes a periodic 

 fluctuation, either in the total intensity or in the distribution 

 in its different parts. IMore probable is the combination of the 

 above two modes. Tliirdly, we may suppose a system of parallel 

 currents arranged at nearly equal intervals propagated per- 

 pendicular to itself, over the point of observation with a finite 

 velocity, in which case the waves must show a time difference in 

 difference stations. If the results of observations confirm the 

 exact simultaneity of waves in widely distant stations beyond 

 all doubt, the last hypothesis will naturally fall out. Birkeland, 

 indeed, observed the approximate simultaneity of some waves 

 with short periods in two stations so widely apart as Potsdam 

 and Bossekop. The two examples reproduced in his report, 

 however, refer only to nearl}^ the same midnight hour, where 

 the direction of the atmospheric current might well have been 

 approximately parallel to the line joining the two stations. 

 Again, according to the result of the simultaneous observations 

 made at Kyoto, Misaki and Sendai. in April, 1909, by Dr. 

 Kadooka and Prof. Tanakadate, a similar simultaneity is observed 

 within the limit of experimental error; but in this case the 

 distances were not very great. Though these two observations 

 strongly speak against the progressive nature of the periodic 

 disturbances, a further accumulation of evidence Avill not be 

 considered superfluous for deciding the point beyond all doubt. 

 On the other hand, some disturbances of a longer duration 

 investigated by Ad. Schmidt^^ were actually progressive. 

 Birkeland also attributed a velocity of translation at the rate of 

 10Ü km. per minute to some class of perturbations. A possibility 

 of occasional occurrence of progressive waves, if not of regular 

 phenomena, seems to be not yet dispproved. On this view, it 

 will be of some theoretical interest to include the case of 

 progressive waves among tlie possible cases and see how far the 

 hypothesis is favourable or unfavourable for explaining the 

 different peculiarities of the observed phenomena. 



li Ad. Schmidt, Met. Zs., 16 1899, p. 385. 



