On Rapid Periodic Variations of Terrestrial Magnetism. y]^ 



iVgain, cases are quite common where remarkable trains 

 of waves occur without any general swelling of the horizontal 

 component. Among such trains, we may distinguish two types, 

 viz., those with the al)rupt beginning and those gradually 

 increasing in amplitudes. The former, which is most frequent 

 during night, may prol)ably be explained b}^ a transient increase 

 of the atmospheric current with a duration comparable with, or 

 shorter than the natural period of the atmospheric oscillation. 

 Indeed, the first wave of the train of this type is often sensibly 

 shorter than the following regular waves. The latter class, 

 whicli is generally the case with the shorter day- waves and also 

 sometime with the longer night-waves, may be caused eitlier 

 by the mechanical disturbance of the atmosphere propagated 

 from a remote region, or by some different agents. Among 

 other conceivable causes of the vertical vibration of the upper 

 atmosphere, we may cite the instability of the discontinuous 

 horizontal motion of the higher layers. That the upper layer 

 has a considerable angular velocity with respect to the earth's 

 surface is well known from the observation of the "illuminated 

 night cloud." It is then probable that a favorable vertical 

 distribution of density may cause a reinarkaljle wave motion of 

 different types. The gravitational wave, as occurs in the case 

 of an incompressible fluid, will not answer our purpose. P)Ut 

 it is more than probal)le that a radial or vertical expansional 

 wave whicli is the most persistent type, is excited in this way — a 

 case somewhat analogous to the excitation of the organ pipe 

 by the stream of air running along the loop of the vibrating air 

 column. If the upper atmosphere be arranged in different 

 layers with different temperatures and different general wind 

 velocities, as is the case in the troposphere, different periods 

 may occur at the same time. Tlie irregularities of the pulsations 

 during day time might well have been produced in this way, 

 if we consider that the laminar structure may I)e enhanced in 

 some way or other, by the influence of solar radiation. Beside 

 the thermal efïect, light pressure may also play some sensible 



