182 



KNOWLEDGE. 



[August 1, 1898. 



would form an outer ring ; and as long as any portion of 

 the inside of one ring was visible, the same amount would 

 be visible of the other. It follows, then, from this view 

 that there is no necessity to fish up the umbra by refraction 

 from the bottom of a spot. As to its possibility, I am in 

 no way competent to form an opinion ; but as regards the 

 illustration given in support of the theory, I fancy that a 

 pair of bellows brought to bear upon the surface of the 

 water in the bowl would obliterate the image of the penny 

 at the bottom of it, just in the same way as a very slight 

 ripple coming upon the surface of calm water instantly 

 renders all objects invisible at the bottom : and, considering 

 what the eruptive force must be which produces a sunspot, 

 and the violence of the movements which must be going 

 on among the gases and vapours within its area, the con- 

 ditions do not seem favourable to the transmission of an 

 image by refraction. 



There is, however, another point which has always 

 seemed to me to require some explanation, viz., the rapidity 

 with which a spot opens out when still but a short distance 

 from the limb — more observable in some cases than in 

 others. I have often been surprised to find that a spot 

 which was little more than a mere line on one day had 

 opened out within twenty-four hours to a degree that I did 



Fig. 4.— a Group of Sunspote on the Second Day after passing the 

 East Limb. May 7th. 



not know how to account for. It has sometimes occurred 

 to me whether refraction could have anything to do with 

 it ; not refraction, however, within the spot itself, but that 

 of the sun's atmosphere, as we are, in this case, looking 

 along the surface of the sun for an immense distance, and, 

 consequently, through the greatest possible amount of his 

 atmosphere. The two drawings are very rough copies of 

 drawings of spots showing the extent to which a spot 

 opened out from one day to another after its first appear- 

 ance on the east limb. J. H. Jenkinson. 

 Ocklye, Crowborough, May 23rd, 1898. 



[The context of the passage from Proctor's " Old and New 

 Astronomy," quoted by Mr. Shackleton, fully supports Mr. 

 Jenkinson's suggestion that Proctor's reference to refraction 

 in sunspots was a mere casual incident in his argument. 

 The point he is really discussing is whether sunspots have 



their origin from below or above, and he supports the 

 former view with all his usual insight and masterly grasp 

 of principles. On the other hand, as he works out his 

 idea, he shows how slight was his personal observational 

 experience. The sharp well-detined circular outline is far 

 more characteristic of the western than of the eastern 

 border of spots, and decay in spot groups, in the great 

 majority of instances, begins to the eastward and works to 

 the front. To my own mind a sunspot appears a region of 

 upheaval — an upheaval which results in a breach in the 

 glowing shell we call the photosphere, and a smaller breach 

 in the less brilliant shell below it which forms the 

 penumbriE of spots. I regard the photosphere in the 

 neighbourhood of a spot, the penumbra, and the umbra, as 

 probably all convex to the general level of the sun's 

 surface ; the amount of convexity probably varies immensely 

 in dififerent spots. — E. Walter Maunder.] 



THE ECLIP.SE THEORY OF VARIABLE STARS. 

 To the Editors of Knowledge. 



SiBs, — The title of Colonel Markwick's paper seems to 

 me somewhat misleading. The theory referred to does 

 not embrace all variable stars, but a class only. 



My object in writing is, however, a dififerent one. Your 

 able contributor seems to assume that the surface-bright- 

 ness of stars is uniform. In the case of the sun we know 

 that this is not true. Near the edge the brightness is 

 not more than one-seventh of what it is near the centre. 

 This fact seems to be satisfactorily explained by the 

 assumption of a solar atmosphere, which there are other 

 grounds for believing. Assuming that Algol is similarly 

 constituted and that the satellite (Bi was quite dark, still 

 the light would not be constant from the time that the 

 whole of the satellite got in front of the bright star (A) 

 until it began to move oflf again. The minimum would 

 occur when the satellite occupied the most central position 

 in front of the bright star. If the eclipse was central it 

 would occur when the centres of both stars were in the 

 line of sight. 



I think it very probable that in these cases the eclipsing 

 body is not quite dark. If the satellite of Sirius crossed 

 the face of the bright star, the phenomena would probably 

 be undistinguishable from those caused by the passage of a 

 dark satellite. One thing, at all events, seems certain. 

 "\\'e know of no instance in which the eclipse is total ; but 

 on the assumption of a dark companion, it would be natural 

 to expect that its surface would be smaller than that of 

 the bright star, and that therefore the eclipse would not 

 in any event be total. 



Further, the eclipsing body may not be a star, but a 

 dense cloud of meteors. In many cases we are driven to 

 the conclusion that, if it be a star, its density is very small 

 in spite of its opacity and comparatively low temperature. 

 The sun's motion in space must ultimately afifect these 

 eclipse stars, though our periods of observation may not 

 hitherto have been long enough to detect the change. In 

 almost every instance the eclipse must be becoming either 

 greater or less. In the former case the duration of the 

 change will become greater and the difference between the 

 maximum and minimum wiU become greater also. In 

 the latter case the eclipse will become less and less until 

 it disappears altogether. Of course, where the eclipse is 

 now increasing it will ultimately decrease again after 

 becoming central ; and no doubt eclipses will hereafter 

 appear in the case of. stars whose light is at present con- 

 stant. Spica Virginis is very possibly such a star. Either 

 it has been an eclipse variable in the past, or else it will 

 become so in the future. WHS Monck 



