786 



BIRKELAND. THE NORWEGIAN AURORA POLARIS EXPEDITION, 1 QO2 1903. 



TABLE CXXVII. 



If by the aid of these numbers, we eliminate the effect of the earth-currents for instance, on tin 

 Z?-curve the 28 29th November, we find that the change in the earth-current takes place very nearlv ai 

 the time when the perturbing force attains its maximum. 



The same thing will be found in a number of other simple storms when we operate in the same 

 manner, e. g. in those of 



Dec. 89, Nov. 5, March i, Oct. n. 



From this it appears that the effect of the earth-current in the declination is considerable, so con- 

 siderable in fact, that the first distinct maximum in the earth-currents seems to produce the principal 

 maximum of the Z>-curve. 



In that time-interval the curve corresponding to the direct effect of the outer system vario 

 slightly as it approaches a maximum, whereas the variation in the earth-current curve is very marked. 



As the effect of the earth-current in this district brings about an increase of the deflection, i: 

 be easily understood that the two maxima may be very nearly simultaneous, a result at which Professor 

 Weinstein has long since arrived; but it does not follow, as he seems inclined to suppose, that tht 

 induction-phenomenon is out of the question. It is our opinion, on the contrary, that in these storms it 

 comes out very clearly and distinctly. 



I can also here point out a peculiarity about the deflection in D after the distinct maximum, which, 

 though it may seem unimportant, is yet very characteristic as regards both this storm and a number of 

 others. I refer to the slight bulging exhibited by the descending branch of the curve. 



This occurs at the times when the change takes place in the earth-current curves. Here their 

 effect is only slight, and the reason for the somewhat altered character that the curve has here acquired 

 is evidently that in this region the curve will mainly represent the variation in the outer current-system, 

 while before it was also influenced to a great extent by the earth-currents. This little peculiarity w 

 find again in most similar storms, the phenomenon being in some of them more distinct than here, in 

 others less so. 1 will only refer the reader to those storms mentioned above. 



A number of examples of this kind can also be shown in the material from 1902 03, as for 

 instance, on PI. XVIII, the course of the //-curve from about 23 h 2o m to 24'' at the western Central 

 European stations in connection with the simultaneous maximum at about 23'' 40; on PI. XIX, 

 course of the //-curve at Tiflis just before iy h in connection with the intermediate maximum at Matotcl 

 kin Schar, etc. 



On a comparison of the variations in the horizontal intensity with the earth-currents, we may t( 

 some extent make remarks similar to those we have just made regarding the declination. 



The agreement here, however, is not nearly so great; indeed, in the less powerful storms i 

 often impossible to demonstrate distinct synchronous serrations. In more powerful storms, the agreemi 

 is often somewhat better. Thus an elimination of the effect of the earth-current in the //-curve is attendee 

 with considerable difficulty, and probably cannot invariably be performed with the material at < 

 disposal. 



