PART I. ON MAGNETIC STORMS. CHAPT. I. 



47 



This is, in the main, what can in general be said with regard to the placing of the normal line. 

 In certain cases special circumstances may arise which may make it necessary to take other things 

 into consideration, our material being somewhat imperfect for these determinations, as we have only 

 magnetograms for separate days from the foreign observatories, and these separate days are just some 

 of the perturbed ones. Fortunately, in the case of several places, there are several curves upon one 

 magnetogram, so that in this way the neighbouring curves accompany them, a circumstance which has 

 been of great importance to us. 



On the Plates in which the magnetograms are reproduced, the normal line that has been employed 

 in the calculation is generally drawn. 



15. On the Separation of Simultaneous Perturbations. The perturbing force calculated according 

 to the above-mentioned method, will give us the resultant of all the perturbing forces that are present 

 at the moment. Now it will often happen that we at any rate have one system of perturbations which 

 is predominant, so that the total perturbing force gives us directly the effect of this system. But it may 

 also frequently happen that at the same time we have to do with several perturbations, that, in other 

 words, we have in the actual field the superposition of fields from several current-systems. It may then 

 be important to find the effect of each separate one in other words to decompose the total perturbing 

 force into several partial forces, each of which is the effect of an independent current-system, or is at 

 any rate due to relatively independent causes. 



A decided rule for the permissibility of such a decomposition can in general scarcely be given. 

 The reasons that favour the interpretation of the total perturbation as the resultant effect of several 

 simultaneously acting systems, must be apparent from the single case in question. 



We will here, however, draw particular attention to two circumstances, which will be of some 

 importance. 



(1) When the perturbing force during a protracted calm perturbation suddenly changes its direction 

 and strength, only to return once more, after some time, to its original value, it will be natural to 

 conclude that a change such as this is due to an independent system appearing at the same time. 



If this sudden change in P for all places on the earth is only a change in strength, there will, on 

 the other hand, be little reason for assuming the presence of an independent system. 



(2) Another thing which may lead to the settlement of this question is the examination of those 

 places on the earth in which the perturbing force is greatest. 



If, during a perturbation that is strongest at one particular place on the earth, a sudden change 

 takes place that is greatest at a spot situated at a great distance from the first-named place, this must 

 of necessity be regarded as two separate phenomena that work into one another. 



It will thus often happen that during a perturbation that is highly developed at the equator, there 

 appears a change, which increases towards the north pole. Here then, we have undoubtedly to do 

 with two different current-systems, one with its point of departure in the polar regions, and one 

 equatorial current-system. 



Frequently, however, the existence of independent systems may be recognised, although, with the 

 material at our disposal, we may not have the means wherewith to discriminate their magnetic effect. 



It will often be a matter of judgement, whether to undertake a decomposition of the total perturbing 

 force or not. 



It is very fortunate when a protracted perturbation is very quiet and uniform in direction, and 

 the intermediate one is relatively strong and not of very long duration. In such a case, it would be 

 natural to take out the effect of the intermediate storm by drawing a normal line that harmoniously 

 connects the curves before and after it. 



