May 1, 1895.] 



KNOWLEDGE. 



109 



and by five a.m. on March 31st, the disturbance was prac- 

 tically over. A second, but much slighter, set of tremors 

 began the following evening, but ceased at midnight. A 

 comparison with the traces of the magnetic storm which 

 accompanied the great spot of February, 1892 (Knowledge, 



being hidden by the lip of the spot on the side nearer the 

 centre of the sun. Further foreshortening subjected the 

 umbra to a like encroachment, until only the penumbra 

 on the side nearest the limb remained in sight. 



To a certain degree every solar observer will support 



.1 I I I 



"1 — I r 



1 — I — r 



1 — I — r- 



"I — I — r 



"I I I r~i I r 



1 



:^^-^.A^^A 



\'y^.-v.^ 



J L 



J L 



J I L 



_X L 



I I I I 



FlO. 2. — Photographic Trace, showing the movements of the Declination Magnet at Greenwich, March 30th and 31st, 189-1. 



May, 1892, p. 92) will show that there was nothing on the 

 present occasion to rival the short, swift vibrations of the 

 earlier disturbance, nor did this storm begin by that sharp 

 instantaneous twitch which was seen in that of 1892 — a 

 twitch which is a well-marked and most truly typical 

 feature of tirst-rate magnetic disturbances. 



The coincidence, therefore, in the present case is of a 

 much less convincing character than in those of the giant 

 spots, and if it stood alone but little could be made of it ; 

 but taken in connection with the evidence which the 

 greatest spots and storms supply, it tends to confirm the 

 view that the solar magnetic influence, whether direct or 

 indirect, is not associated with any precise position of a 

 spot on the disc, but is greatest when the spot is near the 

 central meridian. For the instances given in my paper, 

 alluded to above, referred mostly to magnetic storms which 

 had taken place when a great spot had just passed the 

 central meridian. The present is a case in which the 

 storm took place just before the most important visible 

 group reached it. 



Passing from the sunspots, the next remarkable feature 

 of our photograph is the contrast in brilliancy between 

 the centre of the field and its edges. Round the northern 

 group, and stretching far to the south in one vast sheet, is 

 a mass of intensely luminous matter, compared to which 

 the ordinary photosphere as seen towards the edges of the 

 photograph looks dim and dark. It cannot be questioned 

 that this matter is at once brighter and higher than the 

 general photosphere. The way in which it invades, and 

 overhangs, and in some cases apparently hides the 

 members of the western and northern spot-groups is 

 irresistible proof of its greater height. 



This fact is of some importance in the light of a 

 discussion which has recently been revived by the Rev. 

 F. Howlett, F.R.A.S., as to whether or no sunspots are 

 cavities. It is well known that Prof. Wilson, nearly one 

 hundred and twenty years ago, was led to conclude, from 

 watching the behaviour of a large, well-defined spot, that 

 this was their true form. As the spot approached the 

 limb, and was therefore seen strongly foreshortened, 

 the penumbra on the side remote from the limb narrowed 

 faster than that on the other side, and finally disappeared, 



Mr. Hewlett's attack on the theory which Prof. Wilson 

 based on this observation. There is no question nowadays 

 of the body of the sun being dark ; neither can we suppose 

 that the depth of even the deepest spot bears any 

 appreciable ratio to the solar radius, or even to the 

 diameter of the spot opening itself. If sunspots are 

 cavities, then, relatively to their area, they are exceedingly 

 shallow cavities, and we have uo right to expect that the 

 foreshortening of a spot near the limb will do much more 

 than apparently slightly misplace the central nucleus. 



The examination of a photograph like the present appears 

 to me to clearly demonstrate that, however relatively 

 shallow, the spots are still actual depressions. This 

 bright faculous matter so clearly overrides both penumbra 

 and nucleus, both in the central spots of the northern 

 group and in the following spots of the western, that its 

 greater elevation cannot be challenged. Seeing, then, that 

 we have four distinct regions of different intensity, this 

 bright faculous matter, the ordinary photosphere, the 

 penumbra of the spots and their nuclei, of which the 

 brightest is manifestly the highest, so that, to take the 

 northern group for example, the spots which it surrounds 

 are evidently depressions below its level, is it not reason- 

 able to conclude that the dark nucleus is a depression 

 below the less dark penumbra that surrounds it, and the 

 penumbra a depression below the brighter photosphere 

 around it '? 



The difficulty of arriving at a conclusion as to the true 

 form of sunspots from observation of the Wilsonian 

 phenomenon seems to me to have been scarcely grasped. 

 In the first place, on any reasonable hypothesis as to the 

 average depth of a spot, the Wilsonian effect must always 

 be but slight. Next, the instability of the vast majority 

 of spots is so great, that the fact of the nucleus being 

 central when the spot is in the centre of the disc, scarcely 

 even furnishes a presumption that it will be so five or six 

 days later. Then the effect of refraction in the solar 

 atmosphere, concerning the amount of whic^ we are 

 necessarily total ignorant, would tend to mask the 

 Wilsonian phenomenon. But above all, if the darkness of 

 a sunspot be due to absorption, and if the spot be a 

 depression, absorption will be greatly increased as the 



