March 2, 191 1] 



NATURE 



1 1 



paper, had Dr. Chree chosen, he could have had at his 

 command data from Europe and Asia which, combined 

 with his own, would have served admirably to have tested 

 the main contentions. He might thus have been credited 

 with a really helpful contribution to the subject ; but no 

 such attempt has been made. 



In view of the discussions which have arisen with regard 

 to time data from present magnetograms, I have made 

 request of each observatory for a statement of the method 

 employed. From the reports thus far received, it is found 

 that no institution has made a more earnest attempt to 

 allow for all sources of error than is the case at the 

 Coast and Geodetic Survey observatories. It is evident 

 that in several instances equal care, for one reason or 

 another, is not given by others, but, judging from the fresh 

 interest aroused by the present investigations, there is 

 t very reason to expect considerable improvement here- 

 after. This may be a sufficiently useful end to have 

 ;ichieved, even if nothing else had resulted from the 

 ■■'Searches. 



Theory. 

 The hypothesis of ionic currents which I have employed 

 in the study of magnetic disturbances thus far treated is 

 based on the existence of a primary set of electric currents 

 circulating around the earth overhead. Quoting from my 

 paper No. i (Terr. Mag., vol. xv., pp. 122-3) : — 



" Since magnetic observations made at various points on 

 the Earth's surface have revealed the existence of a definite 

 system of atmospheric electric currents, it follows at once 

 that if the atmosphere is made more conducting at any 

 point, an extra current will be started and set in motion 

 by the pre-existent electromotive force or its equivalent. 

 The direction followed by the new current depends upon 

 its origin, upon the direction of the electromotive force at 

 that point, and upon the deflecting effect of the Earth's 

 magnetic field and of the Earth's rotation on the electric 

 carriers. In other words, while we shall look chiefly to 



.extra-terrestrial agencies for ionising the air and thus 

 splitting it up into carriers of positive and of negative 

 charges, we look to the atmospheric electric field and to 

 the Earth's rotation for furnishing the energy necessary to 

 drive the ions over the Earth and by their motion produce 

 the effects observed during a magnetic storm." 



In No. 3 (idem, vol. xvi., p. 34) I summarise the 

 evidence available regarding the outside electric field as 

 based upon the harmonic analyses of the earth's magnetic 

 condition by Adams, Schmidt, and Fritsche. I show that 

 its general characteristics are very similar to the supposedly 

 internal magnetisation of the earth. The outside currents, 

 if negative ones, would have to circulate around the earth 

 from west to east, hence in the same direction as the 

 rotation of the earth. Starting with these currents, I find 

 it possible to account for the earth's own magnetisation if 

 the earth's average magnetic permeability is on the order 

 of 135 as referred to air and for a magnetising force of 

 about 0-0024 C.G.S. This value, while seemingly large, is 

 not impossible, judging from the experiments of Lord 

 Rayleigh and of C. Baur on iron, using small magnetising 

 forces. Furthermore, it must be borne in mind that we 

 are absolutely ignorant as to what effect the great pressures 

 existing at but a few kilometres below the surface may 

 have on the permeability of magnetisable substances — 

 possibly the effect of increased temperature with depth may 

 be completely annulled by the comparatively more rapid 

 increase in pressure. 



In brief, I have set up the hypothesis that the earth is 

 chiefly an electromagnet, the magnetising currents being 

 outside, and consisting of negative electric currents circu- 

 lating overhead in the same general direction as that of 

 the earth's rotation. I follow out the consequences, and 

 show that this hypothesis harmonises with the Gaussian 

 analysis, from which it had been hitherto almost 



. universally concluded that the magnetising causes must be 

 contained chiefly inside the earth» 



By thus putting the magnetic state of our planet 

 primarily in the control of outside electric currents, many 

 of the outstanding problems of terrestrial magnetism are 

 greatly simplified. Any variation, periodic or spasmodic, 

 in the intensity and direction of the magnetising currents 

 must, of course, be followed almost immediately by corre- 

 sponding changes in the earth's magnetisation. It thus 



NO. 2157, VOL. 86] 



becomes clearer now how, in the space of but a few 

 minutes, such great changes can occur in the magnetic 

 condition of our planet as are experienced during magnetic 

 storms. Anything that will cause a change of conductivity 

 in the strata containing the outside currents will evoke 

 changes in the latter, which in turn are revealed in effects 

 on our magnetic needles. There may be many such causes 

 — the theory need not restrict itself to any one, as, for 

 example, kathode rays. 



Analysing the type of disturbance of which the one of 

 May 8, 1902, may be taken as typical, it is seen that we 

 have before us but a miniature reproduction of the earth's 

 own magnetisation. In brief, a system has been found 

 which will produce magnetic disturbance effects precisely 

 similar to the permanent magnetic effects referred to the 

 earth (idem, vol. xv., pp. 25-30, 117, and vol. xvi., pp. 

 33-48). 



Ihe type of disturbances which Prof. Birkeland refers 

 to " equatorial " currents is thus found to be merely a 

 general disturbance of the entire magnetic condition of the 

 earth, of such a simple character that the first harmonic 

 may give a sufficiently complete representation of the 

 observed perturbations. The theory advanced in my papers 

 is that the same electric-current system which may have 

 to be held accountable for the production of the earth's 

 magnetisation will also suffice for the production of the 

 disturbances considered. I propose the name, therefore, 

 of "simple magnetic perturbation," in place of Prof. 

 Birkeland's " equatorial magnetic perturbation," adding 

 the words " positive " or " negative," just as he does, 

 according to whether the general effect is to increase 

 momentarily the earth's magnetisation or to decrease it. 

 Prof. Birkeland and I are not in agreement as to the 

 direction in which the outside currents must go to produce 

 the observed magnetic effects (idetn, vol. xvi., pp. 33-48). 



The careful reader will not fail to observe that the 

 theory, as above briefly outlined, is considerably different 

 from that which Dr. Chree imputed to me on p. 51 of his 

 paper cited above. Furthermore, the calculation which he 

 says he is unable to follow, although others have done so, 

 was merely a preliminary attempt to account for the 

 observed progression of sudden disturbances and to get 

 some idea as to the order of the altitude at which the 

 supposed currents would have to circulate. If Dr. Chree 

 has something better to offer I shall be glad .to know it. 

 Such interest is being manifested now on all sides, that it 

 will doubtless not be long before a fairly satisfactory theory 

 will be forthcoming. Someone must make the attempt, 

 however, to rear a structure; not all of us are willing to 

 rest contented with merely pulling down. One corre- 

 spondent has hit upon another promising clue, which is at 

 present being tested. In the meanwhile, I believe our 

 hypotheses have amply justified themselves by the many 

 new questions raised and the fresh incentives given to 

 investigation. L. A. Baijer. 



Washington, D.C., February 10. 



Postscript. — From Nature of February 2, just received, 

 it is noticed on p. 461 that Prof. Schuster presented a 

 paper before the Royal Society on January 26 entitled 

 "The Origin of Magnetic Storms," in which a critical 

 examination is made of the theory that magnetic storms 

 are caused by streams of electrified corpuscles ejected from 

 the sun. Prof. Schuster, after various calculations, 

 reaches the following conclusions : — 



"If magnetic disturbances are produced by rays 

 emanating from the sun, it can therefore only be in an 

 indirect manner. We may imagine that the injection of 

 corpuscles ionises the upper portions of the earth's atmo- 

 sphere, and consequently renders the already existing 

 electromotive forces more effective, or we may imagine 

 that the approach towards the earth's magnetic field of 

 highly conducting material containing ions of both kinds 

 acts by induction. The effect of such induction would 

 primarily be an increase in the horizontal and a diminu- 

 tion of the vertical forces, while the currents induced in 

 the earth, tending to diminish the horizontal forces, would, 

 owing to the inertia, die out more slowly, so that a semi- 

 permanent effect would be left after the storm." 



It will be seen that the theory as set forth in my paper 

 above is entirely in harmony with Prof. Schuster's con- 

 clusions ; it is, in fact, largely based on his previous re- 



