COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 251 



accompanied by aurora visible even in the south of England, whereas the disturbances treated by 

 BIRKELAND attained the development of a first- or a second-class storm if anywhere only in a portion of 

 the polar regions. In the stations in temperate latitudes they wore mostly of the size one meets with every 

 other day, and if there were any auroral discharge accompanying them, it was not visible outside the 

 region where auroral frequency is high. 



7. Returning to the table particularising BIRKELAND'S plates, wo have now to consider the significance 

 attached to the descriptive terms applied to the disturbances. 



Explaining the term " equatorial," on p. 62, BIRKELAND says : 



" . . . it is not unusual to find perturbations that are best developed and most powerful at the Equator. 

 It has even been found that these perturbations in the regions about the Equator act principally upon the 

 horizontal intensity . . . Such perturbations we ... call equatorial ... Of these there are . . . two 

 kinds . . . , such as produce an increase in the horizontal intensity, and such as produce a diminution 

 . . . The first ... wo have called positive . . . , the second . . . negative equatorial perturbations." 



Of polar elementary storms BIRKELAND says, pp. 84-85 : 



(1) "They are comparatively strong at the poles (meaning the north polar regions). The simul- 



taneously perturbing forces, even as far north as the 60th parallel, have already sunk to about 

 a tenth of their strength in the auroral zone. 



(2) " They are of short duration, frequently lasting not more than 2 or 3 hours. 



(3) " The conditions before and after are comparatively quiet. 



(4) " The oscillations at the (north) polar stations, especially the more southern ones, run a simple 



course. At the poles, they are often characterised by a simple increase to a maximum, and 

 decrease to zero. We may sometimes, even at the northern stations, have to some extent an 

 undulating form, answering to a slow turning of the perturbing force." 



Of compound perturbations no general definition seems to be given. Judging by individual cases they 

 are a combination of phenomena, " equatorial perturbations " predominating at one stage, and " polar 

 elementary storms " at another. 



BIRKELAND'S discussion of " cyclo-median " storms on p. 144 is somewhat lacking in clearness. After 

 expressing his belief that " electric cyclones, wandering over the Earth's surface," according to a suggestion 

 of Dr. AD. SCHMIDT, are a very rare phenomenon at least in large storms, he adds : " It appears, however, 

 that there is a class of perturbations that are due to current-systems which appear in lower latitudes at a 

 height above the Earth that is small in proportion to the Earth's dimensions ... In the whole of our 

 material, we have not found more than one considerable perturbation that in its entirety must be due to 

 systems that come near to the Earth in lower latitudes." 



It is this single occurrence (October 6, 1902) that is characterised as "cyclo-median." Judging by a 

 remark op p. 150, the term was intended to signify that the disturbance was "as great in medium as in 

 high latitudes," and that the electrical currents to which it was due were " vortical in form." 



8. It is difficult to discriminate between Prof. BIRKELAND'S observations and his theories, as the two 

 are so interwoven in his pages. Thus some reference to his theoretical views may tend to clearness. He 

 believes that the "equatorial" perturbations are due to electric currents encircling the Earth near the plane 

 of the magnetic Equator, at a distance above the Earth which is similar to, possibly greater than, the 

 Earth's radius. A current thus situated in a plane perpendicular to the Earth's magnetic axis would 

 naturally give a force which in the equatorial regions would be roughly in the magnetic Meridian thus 

 affecting H almost exclusively and the intensity would be greater in the Equator. Unless the height of 

 the current were large, the disturbance would fall off rapidly as we departed from the Equator, and in 

 even low latitudes there would be a large vertical component. 



There does not seem to be in the volume any close comparison of the amplitude of the " equatorial " 

 perturbations experienced on the same occasion at different stations ; but the numerical data as to the 

 disturbances during individual equatorial perturbations do not show such predominance in the equatorial 

 regions as the definition leads one to expect. 



So far as I can judge, a considerable number of the movements discussed in our Chapter IX, including 



2 K 2 



