loS 



NATURE 



[December i, 189? 



standing difficulty in the way of believing the sun to be the 

 -direct cause of magnetic storms in the earth, though hitherto 

 every effort in this direction has been disappointing. This diffi- 

 culty is clearly stated by Prof, W. G. Adams, in the following 

 sentences, which I quote from his Report to the British Asso- 

 ciation of 1881 (p. 469) "On Magnetic Disturbances and Earth 

 •Currents" : — " Thus we see that the magnetic changes which 

 take place at various points of the earth's surface at the same 

 instant are so large as to be quite comparable with the earth's 

 total magnetic force ; and in order that any cause may be a true 

 and sufficient one, it must be capable of producing these changes 

 rapidly." 



The primary difficulty, in fact, is to imagine the sun a vari- 

 able magnet or electro-magnet, powerful enough to produce at 

 the earth's distance changes of magnetic force amounting, in 

 extreme cases, to as much as 1/20 or 1/30, and frequently, in 

 ordinary magnetic storms, to as much as 1/400 of the undis- 

 turbed terrestrial magnetic force. 



The earth's distance from the sun is 228 times the sun's 

 radius, and the cube of this number is about 12,000,000. 

 Hence, if the sun were, as Gilbert found the earth to be, a 

 globular magnet, and if it were of the same average intensity of 

 magnetization as the earth, we see, according to the known law 

 of magnetic force at a distance, that the magnetic force due to 

 the sun at the earth's distance from it, in any direction, would 

 be only a twelve-millionth of the actual force of terrestrial mag- 

 netization at any point of the earth's surface in a corresponding 

 position relatively to the magnetic axis. Hence the sun must 

 be a magnet^ of not much short of 12,000 times the average 

 intensity of the terrestrial magnet (a not absolutely inconceiv- 

 able supposition, as we shall presently see) to produce, by 

 direct action simply as a magnet, any disturbance of terrestrial 

 magnetic force sensible to the instruments of our magnetic 

 observatories. 



Considering probabilities and possibilities as to the history of 

 the earth from its beginning to the present time, I find it un- 

 imaginable but that terrestrial magnetism is due to the great- 

 ness and the rotation of the earth. If it is true that terrestrial 

 magnetism is a necessary consequence of the magnitude and the 

 rotation of the earth, other bodies comparable in these qualities 

 with the earth, and comparable also with the earth in respect 

 to material and temperature, such as Venus and Mars, must be 

 magnets comparable in strength with the terrestrial magnet, and 

 they must have poles similar to the earth's north and south 

 poles on the north and south sides of their equators, because 

 their directions of rotation, as seen from the north side of the 

 ecliptic, are the same as that of the earth. It seems probable, 

 also, that the sun, because of its great mass and its rotation in 

 the same direction as the earth's rotation, is a magnet with 

 polarities on the north and south sides of its equator, similar to 

 the terrestrial northern and southern magnetic polarities. As 

 the sun's equatorial surface-velocity is nearly four and a half 

 times the earth's, it seems probable that the average solar 

 magnetic moment exceeds the terrestrial considerably more 

 than according to the proportion of bulk. Absolutely 

 ignorant as we are regarding the effect of cold solid rotating 

 bodies such as the earth, or Mars, or Venus, or of hot fluid 

 rotating bodies such as the sun, in straining the circumambient 

 ether, we cannot say that the sun might not be looo, or 10,000, 

 or 100,000 times as intense a magnet as the earth. It is, there- 

 fore, a perfectly proper object for investigation to find whether 

 there is, or is not, any disturbance of terrestrial magnetism, 

 such as might be produced by a constant magnet in the sun's 

 place with its magnetic axis coincident with the sun's axis of 

 rotation. Neglecting for the present the seven degrees of 

 obliquity of the sun's equator, and supposing the axis to be 

 exactly perpendicular to the ecliptic, we have an exceedingly 

 simple case of magnetic action to be considered : a magnetic 

 force perpendicular to the ecliptic at every part of the earth's 

 orbit and varying inversely as the cube of the earth's dis- 

 tance from the sun. The components of this force parallel 

 and perpendicular to the earth's axis are, respectively, 0*92 and 

 o'4 of the whole ; of which the former could only be perceived 

 in virtue of the varying distance of the earth from the sun 



' The moon's apparent diameter being always nearly the same as thesun's, 

 the statements of the last four sentences are applicable to the moon as well 

 as to the sun, and are important in connection with speculation as to the 

 ^;ause of the lunar disturbance of terrestrial magnetism, discovered nearly 

 ^fty years ago by Kreil and Sabine. 



NO. 1205, VOL. 47] 



in the course of a year ; while the latter would give rise 

 to a daily variation, the same as would be observed if the 

 red ends of terrestrial magnetic needles were attracted to- 

 wards an ideal star of declination 0° and right ascension 

 270°. Hence, to discover the disturbances of terrestrial 

 magnetism, if any there are, which are due to direct action 

 of the sun as a magnet, the photographic curves of the 

 three magnetic elements given by each observatory should be 

 analysed for the simple harmonic constituent of annual period 

 and the simple harmonic constituent of period equal to the 

 sidereal day. We thus have two very simple problems, each of 

 v^hich may be treated with great ease separately by a much 

 simplified application of the principles on which Schuster has 

 treated his much more complex subject, according to Gauss' 

 theory as to the external or internal origin of the disturbance, 

 and Prof. Horace Lamb's investigation of electric currents in- 

 duced in the interior of a globe by a varying external magnet. 

 The sidereal diurnal constituent which farms the subject of the 

 second of these simplified problems is smaller, but not much 

 smaller, than the solar diurnal term which, with the solar semi- 

 diurnal, the solar ter-diurnal, the solar quarter-diurnal con- 

 stituents form the subjects of Schuster's paper. The conclusion 

 at which he has arrived, that the source of the disturbance is 

 external, is surely an ample reward for the great labour he 

 has bestowed on the investigation hitherto ; and I hope 

 he may be induced to undertake the comparatively slight 

 extension of his work which will be required for the 

 separate treatment of the two problems of the sidereal 

 diurnal and the solar annual constituents, and to answer 

 for each the question : — Is the source external or internal ? 



But even though external be the answer found in each case, 

 we must not from this alone assume that the cause is direct 

 action of the sun as a magnet. The largeness of the solar semi- 

 diurnal, ter-diurnal, and quarter-diurnal constituents found by 

 the harmonic analysis, none of which could be explained by the 

 direct action of the sun as a magnet, demonstrate relatively large 

 action of some other external influence, possibly the electric 

 currents in our atmosphere, which Schuster suggested as a pro- 

 bable cause. The cause, whatever it may be, for the semi- 

 diurnal and higher constituents would also probably have a 

 variation in the solar diurnal period on account of the difference 

 of temperature of night and day, and a sidereal and annual 

 period on account of the difference of temperature between 

 winter and summer. 



Even if, what does not seem very probable, we are to be led 

 by the analysis to believe that magnetic force of the sun is 

 directly perceptible here on the earth, we are quite certain that 

 this steady force is vastly less in amount than the abruptly vary- 

 ing force which, from the time of my ancestor in the Presidential 

 Chair, Sir Edward Sabine's discovery, ^ forty years ago, of an 

 apparent connection between sunspots and terrestrial magnetic 

 storms, we have been almost compelled to attribute to disturbing 

 action of some kind at the sun's surface. 



As one of the first evidences of this belief, I may quote the 

 following remarkable sentences from Lord Armstrong's 

 Presidential Address to the British Association at Newcastle, 

 in 1863 : — 



" The sympathy also which appears to exist between forces 

 operating in the sun and magnetic forces belonging to the earth 

 merits a continuance of that close attention which it has already 

 received from the British Association, and of labours such as 

 General Sabine has, with so much ability and effect, devoted to 

 the elucidation of the subject. I may here notice that most re- 

 markable phenomenon which was seen by independent observers 

 at two different places, on September i, 1859. A sudden out- 

 burst of light, far exceeding the brightness of the sun's surface, 

 was seen to take place, and sweep like a drifting cloud over a 

 portion of the solar face. This was attended with magnetic 

 disturbances of unusual intensity, and with exhibitions of aurora 

 of extraordinary brilliancy. The identical instant at which the 

 effusion of light was observed was recorded by an abrupt and 

 strongly-marked deflection in the self-registering instruments at 

 Kew. The phenomenon as seen was probably only part of what 

 actually took place, for the magnetic storm in the midst of which 

 it occurred commenced before, and continued after the event. 

 If conjecture be allowable in such a case, we may suppose that 

 this remarkable event had some connection with the means by 



' Communication to the Royal Society, March 18, 1852 {Phil. Trans., 

 vol. clxii. p. 143). 



