PART I. ON MAGNETIC STORMS. CHAPT. I. 43 



In order, therefore, to obtain a clear conception of the perturbation, such as it actually appears on 

 the earth, there are in particular two important points upon which enlightenment is to be sought, namely, 



(1) How is the force distributed upon the earth at a definite point of time during the perturbation ? 



(2) How does the distribution of force change with time ? 



The investigation of these two points has formed one of our principal tasks. 



Our investigations were thus in the first place directed towards finding out how an individual 

 perturbation developes, and what course it takes. We find that for the solution of this problem it 

 has been particularly important to study with special exactitude the simplest phenomena, those in which 

 the course is simple and with no great, sudden changes, as at the outset it seems probable that we 

 are here face to face with elementary phenomena, which together may form the multiplicity of mag- 

 netic storms. 



As, however, there will, as a rule notwithstanding the many great similarities always be many 

 individual peculiarities in each perturbation, which should be specially mentioned, we have decided to 

 treat each perturbation separately, each accompanied by a description. We have, however, tried to 

 arrange them together in groups according to their special character, in such a way that the various 

 elementary types come first, after which the more compound perturbations will be treated. 



There may also be a question of finding average characteristics of a large number of perturbations 

 at one particular place on the earth. It appears, however, that there are several kinds of perturbations, 

 and in order to pick out the average characteristics, it is necessary to keep to one particular kind. 

 Moreover, the course of the perturbations in one place will be greatly dependent upon the time of day. 

 It will thus also be necessary, starting from this point of view, first to proceed to a close investigation 

 of the distribution and course of the perturbations. 



In the treatment of the separate perturbations, we have, in accordance with the above remarks, 

 employed the following mode of procedure. 



The horizontal and vertical components of the perturbing force are calculated for all the observa- 

 tories for a series of points of time within the period in which the perturbation appears, and the result 

 is given in tables. 



In order to obtain a clear idea of the distribution of force, we have employed a synoptic repre- 

 sentation on charts. The direction of the horizontal component of the perturbing force, which was 

 originally determined in relation to the magnetic meridian, is fixed in relation to the astronomical, by 

 the aid of declination. 



Now it might seem reasonable to pick out the perturbing forces themselves, and place them, with 

 their particular direction and magnitude, on the charts. We have, however, instead of the perturbing 

 forces themselves, to mark so-called "current-arrows". These would give the direction of the horizontal 

 current that would produce, above the place, a magnetic force in the direction of the perturbing force. 

 The size of the current-arrows is proportional in every case to the perturbing force, and gives the 

 force in magnetic units. 



This mode of representation is specially chosen out of regard to the Norwegian stations; for there, 

 during a whole series of the greatest polar perturbations, the force will undoubtedly be produced by 

 currents that flow almost horizontally; and the current-arrow then nearly gives the direction of the 

 horizontal current. We have, moreover, other groups of perturbations, e. g. those which we have called 

 equatorial perturbations and cyclo-median perturbations, which are also best represented by current arrows. 



This mode of marking also presents advantages with regard to the geometrical representation of 

 the vertical component of the perturbing force. 



It must not, however, be assumed that the current-arrow indicates that a current is actually flowing 

 in the direction staled, all over the place. The perturbing force may, in the first place, be generated 



