On the Corona of the Sun. 



129 



measure about thirty additional lines, thus increasing the number of 

 lines considerably."* 



Captain Abney informs me as follows : " The spectrum of the corona 

 had fewer lines in the Caroline Island eclipse (1883) than in the 

 Egyptian eclipse (1882), and the corona was much brighter at one limb 

 than at the other in 1883. I think I can trace reversed Fraunhofer 

 lines beyond the bright lines H and K away from the moon's edge." 



It would seem probable that at the time of the eclipse of 1883, the 

 amount of light-emitting gas was smaller relatively to the number 

 of incandescent particles than at the time of the eclipse of the pre- 

 vious year. This supposition agrees with the fact that the scattered 

 solar light, showing the Fraunhofer lines, was strong in 1883. 



There may be another connexion between the corona and the pro- 

 minences besides that of a supply of gaseous matter, namely, one due 

 to an increase of electric potential of the ejected matter when the 

 prominences are numerous and large. 



The electric disturbances which accompany the formation of large 

 sum spots are well known to be of sufficient magnitude to be felt upon 

 the earth, by causing changes in the distribution of the terrestrial 

 magnetism sufficiently great to affect our instruments, f The Astro- 



* " Phil. Trans.," 1884, pp. 264 and 270. 



f We do not know the mode in which the sun acts upon our magnets. The solar 

 action may be a direct one due to changes in the sun's magnetism, or to an electro- 

 magnetic action due to electric currents, or to electrified matter in motion with a 

 high velocity. The views suggested in this lecture of the sun's electrified state, and 

 of the nature of the corona may possibly throw some light on this point. Two 

 distinct modes of the sun's action on the magnetic needle seem to be possible : — 



(a.) The sun being a charged conductor separated from the earth, also a conductor, 

 by an insulating vacuum, would affect the distribution of the earth's electricity 

 by its power of statical induction. As the earth rotates currents would be set up 

 about it to effect the redistribution of electricity required to satisfy the inducing 

 influence of the sun. May we not find in these earth- currents an explanation 

 of some of tbe phenomena of the earth's magnetism ? However this may be, 

 the changes in the sun's statical induction which would follow from the shooting 

 forth of the electrified matter of the corona, may well so affect the earth- currents 

 as through them to bring about the disturbances observed in the needle. The 

 electrified matter of the corona which leaves the sun will still go on, even when too 

 diffused to be visible, and will still continue to produce upon the earth the effect 

 due to its charge of electricity. The amount of this action will depend greatly upon 

 the direction of the projected matter relatively to the position of the earth. 



(b.) The other possible mode of action of the corona would be to suppose an 

 electromagnetic action upon the earth. The electrified coronal matter moving with 

 a high velocity would act similarly in this respect to electric currents. Among other 

 difficulties we must consider the rapid decrease of electromagnetic action at a 

 distance. 



If the sun is a charged body, then in consequence of continually parting with 

 matter charged with electricity of the same name as that of the sun's charge, the 

 sun's potential would be slowly decreasing. This consideration would be in support, 

 VOL. XXXIX. K 



