PART II. POLAR MAGNETIC PHENOMENA AND TERRELLA EXPERIMENTS. CHAP. I. 343 



stations in question are in the western part of an area of convergence; and it therefore seems as if the 

 influence exerted by the negative system were also predominant in these southern latitudes. 



The forces here are of smaller strength, but in Charts VI and VII we see this condition developed 

 to a very much greater degree. The form of the field has undergone no special change, but the per- 

 turbing forces have now increased considerably in strength at the great majority of the stations. This 

 is especially the case on the night-side of the globe. At Cape Thordsen the forces have greatly increa- 

 sed, the most powerful being now found there, although at Ssagastyr the perturbing forces are almost 

 of the same magnitude. We now evidently have a powerful negative system of precipitation on the night 

 side of the globe, which also has a distinct effect in Jan Mayen. At both Little Karmakul and Godthaab, 

 on the contrary, there is, as Chart VI shows, a positive storm at i7 h 4O m ; while Chart VII, for i8 h 2o m , 

 shows a distinct negative polar storm at those stations. The effects of the positive storm, however, do 

 not come out distinctly on these two charts, as the negative storm, owing to its strength, seems to 

 dominate the whole area; but as we have no observations from the districts south of the auroral zone 

 on the afternoon-side of the earth, it is not possible to determine with any certainty the manner in which 

 the conditions actually develope. We have already seen from the curves that this is in all probability a 

 positive storm, and probably also the one that asserts itself to some extent at Bossekop and Sodankyla, 

 and is the cause of the current-arrow having so marked an easterly direction. Finally, if we look at the 

 conditions in the north of Europe and Asia on Chart VII, the discontinuity apparent on a comparison 

 of the conditions at Bossekop and Sodankyla with those at the other stations, would be difficult to ex- 

 plain, if we do not assume that a system of precipitation actually exists there, which counteracts the 

 strong negative system, of which the effects are so apparent everywhere else. 



Lastly, there is another circumstance which should be taken into consideration, namely, the condi- 

 tions in the vertical intensity. If we look at Chart VI, we see that at Bossekop there is a perturbing 

 force, of which the vertical component is directed downwards. A circumstance such as this cannot be 

 explained if we only assume the negative system, of which the area of precipitation falls north of the 

 place; for this would here act in the opposite direction. On the other hand, a positive storm north of 

 the place will actually produce positive values of P w and as already remarked in the account of the 

 preceding perturbation, the positive systems will as a rule have their area of precipitation somewhat to 

 the north of this place. 



The vertical intensity at Sodankyla, however, exhibits just the opposite condition. We have already 

 pointed out once or twice the abnormal condition appearing in the direction of the deflections in the 

 vertical intensity at Sodankyla, and we will therefore merely refer here to what has been previously 

 mentioned respecting the probable cause of this. 



At the three southern stations, the conditions appear to be mainly affected by the negative storm, 

 as the current-arrows indicate that this district is in the western part of an area of convergence; but it 

 is not, of course, on this account impossible that there may be positive precipitation in the district along 

 the southern part of the auroral zone from Norway westwards. 



If we assume that such a system exists, then Christiania and Gottingen would be situated in the 

 eastern portion of its area of divergence; here, however, the current-arrows are directed southwards. 

 Whether there is a negative storm-centre in the district east of this, or a positive storm-centre to the 

 west, the direction of the current-arrows at these stations will be very much the same. It may therefore 

 be very reasonably supposed that these two systems actually existed simultaneously; the conditions at 

 the more southerly stations would also be very much satisfactorily explained on the basis of such an 

 assumption. 



On Chart VIII, for I9 h 20, the powerful storms are over, at any rate at those stations from which 

 we have observations. Simultaneously with the decrease in the strength of the negative storm from the 



