August 28, 1890] 



NATURE 



419 



more condensed still, the flutings of carbon added, as in 

 Andromeda ; it does not seem unreasonable to suppose 

 that any increase of temperature brought about by the 

 increased number of collisions should increase the inten- 

 sity of the lines of hydrogen or carbon in the spectrum of 

 a nebula. 



The observations already accumulated show conclu- 

 sively that in the nebulae — even those so far condensed 

 as the one in Andromeda — the temperature is low ; in 

 other words, the meteorites are very far apart ; regular 

 variability, therefore, would for this reason be very 

 difficult to detect. It is probable, therefore, that in all 

 the cases previously recorded, we are not dealing with 

 the results of rhythmic action, but the interpenetration 

 of nebulous streams or sheets. When, however, we 

 come to the stars — that is, the more condensed nebulae 

 — in Group I. and Group II., the temperature is higher, I 

 the condensation is greater, and the interaction of double ' 

 or multiple nebute can be more easily traced. This i 

 fundamental difference of structure between these bodies j 

 and stars like the sun should be revealed in the pheno- ' 

 mena of variability ; that is to say, the variability of the 

 uncondensed swarms should be different in kind as well 

 as in degree from that observed in bodies like the sun or 

 a Lyrae, taken as representing highly condensed types. 



Since the stars with bright lines, as I have shown, 

 belong to the former group, and since, therefore, they 

 are very akin to nebula?, we might, reasoning by analogy, 

 suppose that any marked variability in their case also 

 would be accompanied by the coming out of the bright 

 hydrogen lines. This is really exactly what happens both 

 in /3 Lyras and in y Cassiopeiae. In ^ Lyr^ the ap- 

 pearance of the lines of hydrogen has a period of 

 between six and seven days, and in y Cassiopeiae they 

 appear from time to time, although the period has not 

 yet been determined. 



Another star of Group I., »? Argus, is also remarkable 

 from the fact that its light varies in the same sort of way. 

 This star is in the southern hemisphere, and during the 

 last twenty or thirty years a considerable discussion has 

 been going on among astronomers as to whether the 

 surrounding nebula is or is not changing its position 

 with regard to the star in question, which has a bright- 

 line spectrum like li Lyrae, and a period not of thirteen 

 days, but of seventy years. The light varies from the 

 sixth to the first magnitude. 



Leaving Group Land coming to Group II., there is 

 one star, Mira Ceti, whose variations in light-intensity 

 may be taken as characteristic. The history of the 

 discovery of this star's variability has already been 

 given. What happens to it in just a little less than a 

 year is this. First, it is of the second magnitude, and 

 then in about eighty days it descends to the tenth, and, 

 so far as observations with ordinary instruments go, it is 

 invisible. In about another hundred days it again be- 

 comes visible as a star of the tenth magnitude. It then 

 increases its light to the second magnitude, and begins 

 the story over again. But sometimes at the maximum its 

 brilliancy is not quite constant. That is to say, sometimes 

 it goes nearer the first magnitude than the second. What 

 happens to the light of the star below the tenth magnitude 

 it is not easy to say. What one knows is that to some 

 telescopes it remains invisible for about 140 days or 

 something like that, and then it begins its cycle over 

 again. 



I owe to the kindness of Mr. Knott the opportunity of 

 studying several light-curves of " stars " of this group, and 

 they seem to entirely justify the explanation which I have 

 put forward. It is necessary, however, that the curves 

 should be somewhat carefully considered, because in 

 some cases the period of the minimum is extremely small, 

 as if the secondary body scarcely left the atmosphere of 

 the primary one but was always at work. But when we 



NO. 1087, VOL. 42] 



come to examine the shape of the curves mor^ carefully^ 

 what we find is that the rise to maximum is extremely 

 rapid : in the case of U Geminorum, for instance, there is 

 a rise of five magnitudes in a day and a half; whereas 

 the fall to minimum is relatively slow. The possible 

 explanation of this is that the rise of the curve gives us the 

 first sudden luminosity due to the collisions of the swarms, 

 while the descent indicates to us the gradual toning down 

 of the disturbance. If it be considered fair to make the 

 descending curve from the maximum exactly symmetrical 

 with the ascending one on the assumption that the 

 immediate effect produced is absolutely instantaneous, 

 then we find in all cases that I have so far studied that 

 the star would continue for a considerable time at its 

 minimum. 



Broadly speaking, then, we may say that the variables 

 in this group are close doubles; the invisibility of the 

 companion being due to its nearness to the primary or to 

 its faintness. 



We now pass from Groups I. and II. to III., IV., and V. 

 These contain the hottest, and therefore brightest bodies 

 in the heavens. They are, moreover, more condensed 

 than those we have considered. On this ground, theif 

 normal light cannot be increased to any very great extent 

 by any constantly recurring action, but it may be reduced 

 by eclipses caused by the revolution of still further con- 

 densed secondary swarms. The nearer the primary, and 

 therefore the smaller the period of the secondary body, the 

 more likely is the eclipse to occur regularly. There are 

 several Group IV. stars of this class, notably Algol, to the 

 first observations of which we have already referred. 



This body, which is always visible in our latitudes, 

 well illustrates this class of variable. If we take the 

 beginning of a cycle, it is a star of the second magnitude ; 

 suddenly in three hours it goes down to the fourth, and 

 then it comes up in another three hours to the second, 

 and goes on again for very nearly three days ; and then 

 it goes down again, comes up again, and goes on again 

 for another three days, and so on. 



There is another star very like this— a star which is in 

 81° N. declination. No. 25 in Argelander's Catalogue. 

 The difference between Algol and this is that the rise 

 and fall are a little more rapid. Its light is feeble for 

 about the same time as the other one, but at the bottom 

 the curve is flat, by which I mean that, instead of going 

 suddenly down and coming suddenly up again, it stops at 

 its least luminosity for some little time. 



Prof. Pickering 1 has demonstrated by photographs of 

 the spectra of Algol that Goodricke's explanation of its 

 periodical variability is correct, the companion having 

 no light of its own. In the case of the star D. 81° N. 25 

 there must be luminosity from the star which eclipses the 

 other. And a very beautiful justification of this view has 

 recently been noted, because, although there is no change 

 in the spectrum of Algol, there is a considerable change 

 in the spectrum of the star, the bottom curve of which is 

 flat, showing that probably the companion has an absorb- 

 ing action of some kind on the light of the central star 

 passing through it or its surroundings. The light 

 practically changes very much as our sunlight would 

 change if it had to pass through the atmosphere of 

 another sun somewhat like itself coming between us. 



In Group VI. we again have a new condition. In 

 these stars the light is relatively faint, and the variation 

 is doubtless due to swarms of meteorites moving round a 

 dim or nearly dark body, the maximum occurring at 

 periastron when the tidal action in the swarm is greatest ; 

 hence the addition of the hght of what we with our solar 

 conditions should term a large comet would make a great 

 difference in the total radiation. 



J. Norman Lockyer. 



» Proc. Amer. Acad. Sci., vol. xvi. p. 17 



