Oct. 27, 1 881] 



NATURE 



61: 



discharge through it than they did before. In this way we may 

 conceive that in a great outbreak like that observed by Mr. 

 Carrington, where the hot interior of the sun is turned 

 up, as it were, and radiates towards the earth, the facility 

 fjr the passage of the electric dicbarge is increased, and 

 it may be very rapidly increased. So that according to 

 thii theory the foundation of these three phenome la lies 

 in atmospheric electricity, which forms as it were the maga- 

 zine, and the solar radiation, as it were, supplies the match, 

 and allows it to be discharged. Of course, over and above that, 

 when solar radiation is active, all th_- phenomena which de,iend 

 on solar radiation may be expected to be active too ; and there- 

 fore beyond its inflaence in firing the match, to speak meta- 

 phorically, this s )lar radiation, w hen more active than usual, and 

 la-iting, will also produce a more rapid development of all those 

 proce^ses at the surface of the earth which depend on solar 

 radiation, among others, no doubt, the generation of atmo- 

 spheric electricity, although we are not at present able to explain 

 with certain'y the manner in which it is produced. So that in 

 two ^^■a^■s, by applying the match to tlie train already laid and 

 by grtiduttlly inaiuifacturing the powder, the increased solar radia- 

 ti 'n may cause an increase in those electric discharges and earth- 

 currents as the result of the redistribution of the induced elec- 

 tricity at the surface of the earth, and thereby a disturbance of 

 the magnetic elements. I do not know of any other theory 

 than tha': of atmospheric electricity which furnishes anythi:ig 

 like sufficient electromotive force to account for these auroral 

 discliarges, if they are really electric discharge-; analogous to 

 those which take place in exhausted tnbes. As I said, it has been 

 supposed by some that the magnetic disturbances are due to 

 earth-cuirent-; ; according to the the :>ry which I have advanced, 

 they are due rather to a vast assemblage of currents, partly atmo- 

 spheric and partly terrestrial. An objection has sometimes been 

 taken to the supposition that the magnetic disturbances are 

 due to e.arth -currents arising from the con^ideration of the 

 electro-magnetic effect which the eai'th-current actually observed 

 would have upon the needle. Bat this, 1 think, is obviated when 

 y jU re jember that an earth-current actually observed is merely 

 what results from the examination of a very small portion of 

 this vast electric system, stretching it may be over hundreds of 

 miles of country. At Greenwich, for instance, there are now 

 wires by which earth-currents are regularly observed. The co- 

 incidence between photographic traces left by the earth-currents 

 and those left by the magnetic storms is most re narkable. 

 Every jjeak of the one, you may say, answers to a peak of the 

 other. It has been noticed, however, that there appears to be 

 a slight difference in the time of the occurrence. It would 

 appear as if the disturbances preceled the earth-currents. Well, 

 that may very well be, because, according to the theory which I 

 have advanced, the effect on the mignet is the resultant effect of 

 a vast series of currents, partly terrestrial and partly atmospheric, 

 stretching over a very large region of country, whereas the 

 earth -currents observed are mert;ly obtained by tapping the 

 earth at a couple of places at no great di stance, so that the two 

 do not by any means necessarily correspond exactly. 



I forgot to mention at the proper time a diagram which 

 Capt. Abney has kindly prepared for me. This is a copy of the 

 diagram made by Mr. Ellis of the Royal Observatory, giving 

 the result of his discussion of the Greenwich observations on 

 two out of the three magnetic elements, namely, the declinati m 

 and horizontal force, as compared with sun-spjt frequency. [Mr. 

 Ellis's diagram in Part ii. of the Phi'osophical Transictions for 

 1S80 was then referred to.] 



You see that an examination of the pheiiDmena going on 

 a: the solar surface itself leads us to the conclusion that there 

 are vast currents up and down, by means of which the compara- 

 tively speaking cool upper portions are continually replaced by 

 hotter matter from b;nealh. Mr. Lockyer, in the lectures he 

 is about to give, I have no doubt will have a great deal of very 

 interesting evidence derived from spectroscopic study of the 

 phenomena to lay before you, bearing out that same conclusion 

 We have seen that the supposition that there is extra radiation 

 when the interior portions of the sun are ejected and come to 

 the surface, falls in very well with the known relationship 

 betvieen the occurrence of sun-spots and the three terrestrial 

 phenomena I have mentioned — magnetic disturbances, aurorre, 

 and earth-cur ents. I say between the sun-spots, although it 

 is not, strictly speaking, the sun-spots themselve;, but the 

 tremendous disturbances which are their precursors, and of 

 which they form the most easily-observed manifestation. 



Now if there is reason to believe that, when the sun is in a 

 state of activity in this manner, there is increased radiation from 

 it, it may well be that the meteorology of the earth is affected 

 by the changes which take place at the surface of the sun ; but 

 the meteorology of the earth forms an exceedingly complicated 

 problem. We have, so to speak, to deal here with a very compli- 

 cated in'egral of a differential equation. I am speaking some- 

 what metaph jrically, but my words will be understood by the 

 matheoiaticians who happen to be present. We cannot very 

 directly connect that integral with the disturbing forces. One 

 thing, however, we may say : supposing thit there i- a system 

 of any kind subject to periodic disturbing forces — and we have 

 seen reason to believe that these great eruptions which take 

 place on the surface of the sun are, perhaps somewhat roughly, 

 periodic — if, I say, we hive a periodic system of disturbing 

 forces, then the system which is acted upon by these forces will 

 show a peri die disturbance which may be more or less con- 

 cealel by apparently capricious disturbances, but which yet may 

 be exjoected to come out in the long run ; and it has been sup- 

 posed by those who have studied meteorological phenomena that 

 there are indications of a decennial, or nearly decennial period 

 in some of the meteorological elements, for instance, the mean 

 temperature of the air and the fall of rain. Again, in some 

 observatories thermometers have been sunk to a considerable 

 depth in the earth, and observations of such thermometers were 

 carried out for a gre.it number of years by the Astrononer Royal 

 of Scotlmd, Prof. Piazzi Smyth, and they are regularly carried 

 out now at Greenwich. Connected with the annual variation of 

 te.nperalure between summer and winter there are, so to speak, 

 waves of heat and cold slowly propagated do vn from the surface 

 of the errth to the interior, rapidly decreasing in amplitude as 

 they descend, and by going a suitably moderate depth you get 

 these fluctuitions, indicating the annual fluctuation of the atmo- 

 spheric temperature, and free in a great measure from the 

 fluctuations which take place at much slio.ter periods. When 

 yon go a little way down the results given by these thermometers 

 seem to indicate something of a decennial or nearly decennial 

 period. Conflicting statemenl.s, however, have been made by 

 different ob erver^ as to the time of maximum of the meteoro- 

 1 jgii.-al elements which were supposed to have such a period, 

 and some have argued from the results that when the sun was io 

 a highly spotted condition we had a higher tenijierature than 

 usual, and some the reverse. Now this is an important matter 

 to attend to. Suppose we had such a system acted on by 

 periodic disturbing forces ; it will show at least in the mean a 

 corresponding periodic fluctuation, corresponding however only 

 as regards the length of the period. The epoch of maximum of 

 the element observed, whatever it may be, has no necessary 

 relation to the epoch of maximum of the disturbing forces, 

 excepting thit they are separated by a constant interval, and the 

 ep ich of maximum of the element observed may be different at 

 one locality from what it is at another. So that it is only the 

 period and not the epoch of maximum which you can expect to 

 arrive at possibly by an observation of such elements as I have 

 spoken of. It is very difficult indeed to say, even if a ten yearly 

 period be observed, what ought to be the year of greatest solar 

 radiati in if we have given the observed results. 



Is there any way in which we miy hope to attack that 

 problem? I think there is. It is by no means hipeless to 

 attempt to measure by a direct process the solar radiati >n. In- 

 struments hove been devised for the purp ise, called actiiio neters. 

 One was devised by the late Sir John Herschel, and goes by his 

 name, and it is a very beautiful instru nent ; but unfortunately it 

 is excessively fragile, and if the instrument has got rough travel 

 or rough work at all to go through, it is pretty sure to '.oe broke 1. 

 Other instruments have been devised for the purpose, and among 

 them I may mention one by Prof. Bsilfour Steivart. He has 

 lately devised a new actinometer, and one of his c 'ust u;tion has 

 recently been sent out to India, and is at present under trial. In 

 these cases heat is observed by a thermometer. Another has 

 been devised by Prof. Roscoe, depending on the chemical action 

 of radial ion. I think none 0! these have yet had a thoroughly 

 complete trial, because in such a climate as ous a fair trial can 

 hardly be made, since there are so many disturbing elements in 

 the lower atmosphere. An exceedingly slight cirru--haze makes 

 an enor nous difference in the am lunt of heot radiated from the 

 sun as received by u- without being deflected from its course. If 

 there be the slightest haze a good deal of the heat rays are de- 

 flected from their course, perhaps not much deflected, so that if 

 we take in the direction of the sun itself and the neighbouring 



