PART I. ON MAGNETIC STORMS. CHAP. III. 



In the United States, the conditions are fairly uniform all the time. The current-arrows show a 

 great convergence of the perturbing force. 



Owing to the above-mentioned similarity between the form of the curve at Honolulu and that at 

 the three eastern American stations, we may conclude that this polar storm must have an effect in Hono- 

 lulu. It is impossible to take out any decided values; but a glance at the curve will show that the effect 

 consists in a perturbing force directed towards the north-east. The current-arrow, inasmuch as it is 

 dependent upon the polar system, thus comes to be directed towards the south-east. In this way the 

 force at Honolulu completes the area of convergence. 



In the above-mentioned equatorial district on the eastern hemisphere, the forces are directed along 

 the magnetic parallels. 



With regard to the wiew to be taken of this perturbation, it may in the first place be considered 

 probable that the conditions in the north of America are mainly determined by a polar elementary storm 

 at first not very far north-east of Sitka. The centre afterwards travels westwards. It may be remarked 

 that during the perturbation this district passes midnight. The current-arrow about the centre is pro- 

 bably directed westwards along the auroral zone. The storm is in the main of a character similar to 

 those that usually occur a little before midnight, with their centre near our Norwegian stations, and almost 

 always travelling eastwards. 



As regards the simultaneous perturbation over the district between Kew and Batavia, it seems 

 impossible, both on account of the form of the field and of the magnitude of the force, that this storm 

 can be a direct effect of the polar system. On the other hand, the field must immediately suggest the 

 thought of the current round the earth as the cause of the perturbation. Some doubt may be felt on 

 this hand owing to the disturbing influence occasioned by the polar storm in the western hemisphere. 

 We have previously mentioned conditions, however, especially as regards Honolulu, which indicate that 

 there two systems appear simultaneous in H, counteracting at one another. The polar system, from the 

 form of the curve, must be assumed to act in a northerly direction, when the other must act i a southerly 

 direction in order to compensate the former, in which case the conditions in Honolulu should be in 

 accordance with those in the eastern hemisphere. 



According to this, it is not improbable that simultaneously with the polar storm there is a pertur- 

 bation answering to a current round the earth from east to west, a perturbation of the type we have 

 called negative equatorial storms. Owing to the slight variation of the force from place to place, and 

 to the uniform course of the perturbation, this current may be assumed to lie at a distance from the 

 earth of at least a magnitude equal to the radius of the earth; and symmeiry would point to the 

 regions round the plane of the magnetic equator as its situation. 



The main features in the form of the field may thus be explained, as we have seen, fairly simply 

 in the above manner. If we look at the charts, however, we see, that the field bears an unmistakable 

 resemblance to those that we should expect to find during the cyclo-median storms. Under such an 

 assumption, the perturbing forces that appear at Sitka at about g h 15 also receive quite a simple ex- 

 planation. It is only necessary to refer to the photographs of the terrella, when, if we compare the 

 light-area in fig. 68, i with our field, we find the resemblance is striking, if we imagine Sitka as being 

 near the uppermost angle. If we then imagine the field moved westwards with the sun, we have more 

 or less the conditions of Charts II and III. The arrow at Christchurch on Chart II is worthy of notice. 

 It answers to that part of the light-area that falls upon the southern hemisphere; and the direction of the 

 arrow is also in accordance with what we should expect to find if the system on Chart I were moved 

 westwards. There may well be some doubt as to the view to be taken of the conditions. Perhaps the 

 most probable is that at first the perturbation partakes most of the nature of a cyclo-median storm, and 

 subsequently changes into a more purely polar one. 



