6i4 



NATURE 



{Oct. 27, 1 88 1 



and besides that there were from time to time irregular, or 

 apparently irregular, disturbances following no observaljle law. 

 It was known, too, that the diurnal fluctuation, which ha^ now- 

 been know n considerably more than a century, was greater in 

 summer than in winter. It had been also observed that the->e 

 apparently irregular fluctuations in the direction of the magnet 

 are observed when an aurora is seen.' If there is an aurora 

 there are sm-e to be these fluctuations, and if there are these 

 fluctuations the probability is, if other circumstances permit, that 

 we shall see an aurora. The connection between the two was 

 made out about the year 1750, so that it is by no means new. 

 Of course we cannot expect that for every magnetic disturbance 

 we shall have a visible aurora ; for in the first place the disturb- 

 ance may take place in the day-time, in which case of course no 

 aurora can be seen ; then, supposing even it takes place at night, 

 it may be that at the time the whole sky is covered with cloud;, 

 which prevent the aurora being seen ; or again, it may be a 

 bright night with the moon shining brightly, not far from the 

 full, and then a faint aurora would not attract much notics. In 

 fact I haN-e often felt in doubt when I saw a luminous streak in 

 the sky on a moonlight night whether it was an auroral streamer 

 or merely a mare's-tail cloud illuminated by the light of the 

 moon. After watching some time one can generally determine 

 which it is, because if it be an auroral streamer it is pretty sure 

 to be unsteady ; but if the observer happens to have a small 

 spectroscope in his pjcket, or even a prism and a slit, the dis- 

 tinction can be made out at once, on account of the peculiar 

 spectrum of auroral light. 



There appears then to be evidently some intimate connection 

 between magnetic disturbance and the aurora. The recent pro- 

 gress of telegraphy has caused us to be familiarly acquainted 

 with another electrical phenomenon, or, if you like, magneto- 

 electrical. (I assume here tliat the aurora is an electrical pheno- 

 menon — that, in fact, has long been admitted — for considerably 

 more than a century.) I allude to the earth-currents. In tele- 

 graphy we have occasion to use insulated wire, the ends of 

 which are placed, or may be placed, in connection with the earth. 

 Now when that is done it frequently happens that, without sending 

 any current from the battery at all through the line, there is a 

 niDre or less powerful current trans.nitted along the wire, which 

 is made evident by the deflection of the galvanometer. In fact, 

 in certain case, these currents are so strong that they interfere 

 with the working of the lines. 



At the failure of the first Atlantic cable in 1865, Sir William 

 Thomson (whour I am happy to see before me) made some experi- 

 ments with these earth-currents, as they are called, transmitted 

 through this cable. The failure was of such a nature as would 

 have been caused if there wa> a breakage in the cable something 

 like 300 milei off, and currents were transmitted through the 

 cable, indicating an electromotive force, as it is called, amount- 

 ing to one or two Daniell's cells; on one occasion to five or 

 six; and currents more powerful even than those are observed 

 from time to time. 



Now it is well known that at times of magnetic disturbance 

 ■we have these earti-cm-rents powerful ; and as I mentioned that 

 magnetic disturbances and aurora; come together, we have here a 

 third phenomenon, that of earth-currents, which accompanies 

 the t .vo former. These three are evidently intimately connected 

 with one another, whatever be the cause of that connection. 

 But at present I have said nothing whatever of the relation of 

 these three phenomena to the sun. Of course any one would say 

 there is the remote relation, that it is to the radiation from the 

 sun that all the great changes that take place on the surface of 

 the earth are due : the evaporation of moisture, the heating of 

 the air, and consequent production of winds ; and in a remote 

 sense, therefore, there would in all probability be a relation 

 between these three phenomena and the sun But that relation 

 is very far from remote. I forgot at the proper time to mention 

 one circumstance connected with these earth-currents before I 

 came to the sun, which, if you will allow me, I will do now. I 

 mean the magnetic disturbances. 



One of the first fruits of the establishment of regular magnetic 

 observatories was the remarkable discovery that these magnetic 

 storms occurred simultaneously over large tracts of the earth's 

 surface ; so that even the sudden and a]>parently capricious 

 variations of say the direction of the declination needle would 

 be observed simultaneously at the same moment of absolute, 

 not local time, at places far separated from one another, such as 



^ At least of the dancing kind ; faint, steady aurorEe do 

 accompanied by sensible magnetic disturbances. 



to be 



London and Paris, and London and Lisbon even.' The cause 

 of this magnetic disturbance, whatever it may be, must be ons 

 very widely spread. In discussing the results which have besn 

 obtained at the colonial magnetic observatories. Sir Edward 

 Sabine made a remariiable discovery, namely, that whether you 

 take the range of ordinary diurnal fluctuati jiis of the magnet, or 

 whether you take the frequency and magnitude of the.e mag- 

 netic disturlsances that I spoke o^, in both cases there appeared 

 to be a decennial period, or a period nearly decennial, and that 

 corresponded to the period of solar spots, corresponded not merely 

 as to the duration of the period, but also as to the time of the 

 maximum ; so that in those years when the sun showeJ an 

 unusual number of spots of unusual magnitude, both the regular 

 diurnal variation of the magnet was greater than the average, 

 and there were more numeroas and more violent magnetic 

 storms ; on the other hand, when the sun was comparatively 

 free from spots, the magnetic elements were, comparatively 

 speaking, in a tranquil state. In the older observations the 

 declination was the only one of the magnetic elements which had 

 been observed, but all three components of the mrgnetic force 

 were observed in the-e ob ervatories, and accordingly the phe- 

 nomena C3ukl be more searchingly investigated. Further re- 

 search has fully confirmed this connection, so that there can be 

 no doubt now that there is some intimate connection, whatever 

 be its nature, between soUr s;:ots and magnetic disturljances. 



I will mention one circumstance which is a remarkable corro- 

 boratijn of this observation. The late Mr. Carrington for many 

 years was engaged in a seriv:s of m^st careful and elaborate 

 measurements of the positions and magnitudes of the solar 

 spots. The way he worked was by throwing a large image 

 of the sun by means of an equatorially-mounted telescope, 

 with its eyepiece suitably focussed, on to a fixed screen. One 1 

 day he was engaged at this work when he saw two bright 

 spots on the screen. His first impression was that the screen 

 which was used to shut off the light of tlie sun, which otherwise 

 would have passed down outside of the object-glass of the tele- 

 scope, outside the tube, had got disarranged somehow or other, 

 and that it was merely the sun shining through the holes, and 

 coming oil the screen. He moved the telescjpe a little, and 

 these spots moved with the image of the sun, proving that it was 

 not merely the sun shining through holes in the shading screen, 

 but that they really belonged to the sun. They remained visible 

 some minutes, during which they moved over a very sensible 

 portio 1 of the sun's disk at such a rate that the actual lineal mo- 

 tion of them mast have been — I forget the figures, but I think it 

 was something like 100 or 150 miles per second. Moreover one 

 of them passed over a dark spot, which is confirmatory of the old 

 observation of Wilson, that the spots are at a lower level than 

 the general surface of the sun. 



Now it so happened that on examining the records of the mag- 

 netic n;edle, which were kept automatically by a photographic 

 process at Kew, just at the moment when these spots were seen, 

 there was an unusually great magnetic disurbance. Well tlien, 

 what can be the connection between these apparently so dis- 

 similar, ajiparently so disconnected phenomeni, and what is the 

 cause in the first instance of the three terrestrial phenomena I 

 fir-t me itioned — magnetic disturbances, aurora, and earth 

 currents ? 



Different theories have been started as to this connection. 

 Some have supposed that the disturbance of the magnetic needle 

 was an electro-magnetic effect due to the earth currents ; others 

 have supposed, on the contrary, that the earth currents were due 

 to the electro-magnetic induction produced by a change in the 

 magnetism of the earth. But what of the aurora; ? It has long 

 been recognised that the aurora is an electrical phenomenon. It 

 has been supposed to be imitate! — and there can be no reason- 

 able doubt that the supposition is a correct one — by sending an 

 ordinary electric discharge through a highly-exhausted tube. But 

 whence comes the electromotive force requisite to effect that 

 discharge ? My colleagues are not in any way responsible for 

 what I am going to advance. I an going to suggest a cause for 

 this phenomenon which, so far as I know, has not hitherto been 

 broached," and of course you must take it for what it is worth. 

 It has not seen the light, and therefore has not had the oppor- 

 tunity of being subjected to the criticism of men of science. If 

 laboratory experiments are to be any guide to us it requires no 



' A number of photographic records from various magnetic observat:)ries 

 have recently been compared and discussed by Prof. W G. Adams. 



^ This refers to the theory as a whole : the individual parts of it had 

 mostly formed limbs, so to speak, of one or other of a set of theories which, 

 taken in their entirety, must be regarded as quite different. 



