CHAPTER III.

8 Cephei.. B Lyra.

Variable Stars.-o Ceti. Algol. R Corona Borealis. n Argus. - Miscellaneous remarks.—Temporary Stars.-Notices of Stars which have disappeared.-Proper motion. -Motion of the System through space.Summary by W. Struve.-Proper motion first suspected by Halley.- Wright's hypothesis of a Central Sun.-Revived by Mädler.-Stars which are probably Centres of Systems.

HERE are many stars which exhibit periodical changes of

are f

:

brilliancy these are termed variable stars. Upwards of a hundred stars are now known to belong to this class, and many more are put down as 'suspected.'

One of the most interesting, as also the first that was recognised, of these curious objects, is o Ceti, or Mira [sc. stella]. It appears about 12 times in 11 years; remains at its greatest brightness for about a fortnight, when it sometimes equals in brilliancy a star of the 2nd magnitude; decreases during about 3 months, till it becomes totally invisible; it remains so for about 5 months, and then gradually recovers its brilliancy during the remaining 3 months of its period. Its maximum brightness is not always the same, nor does it always increase or diminish by the same graduations; neither are the successive intervals of its maxima equal. The mean period is 331d 8h, but it would appear from the researches of Argelander a that this period is subject to a cyclical variation embracing 88 such periods, which has the

a Ast. Nach., 624.

effect of gradually lengthening and shortening alternately these periods to the extent of 25 days one way and the other. It is not improbable too that the irregularities of its maximum brilliancy are also periodical. On Oct. 5, 1839 (the epoch of maximum for that year, according to Argelander) Mira was unusually bright, excelling a Ceti and equalling ẞ Aurige. On the other hand, according to the testimony of Hevelius, between Oct. 1672 and Dec. 1676 it did not appear at all b.

I append a few details connected with the history of this

star:

On Aug. 13, 1596 David Fabricius noted a star in Cetus to be of the 3rd magnitude, and that in October of the same yeare it disappeared. 7 years later, or in 1603, Bayer, the author of the stellar nomenclature still in use, affixed the letter omicron (o) to a star in Cetus placed exactly where the star of Fabricius had disappeared. He observed it to be of the 4th magnitude, but not comparing with his own the former observations of Fabricius he failed to make the discovery which was within his grasp.

In the beginning of Dec. 1638 Phocylides Holwarda of Franecker saw this star shining brighter than one of the 3rd magnitude. In the summer of the following year he was unable to find any trace of it, but on Oct. 7 he again perceived it; and to him may be assigned the honour of having first discovered the existence of variable stars.

In 1648 Hevelius commenced a careful series of observations, which were carried on till 1662, during which period the certainty of the discovery was placed beyond a doubt, and a first approximation was made to a knowledge of the attendant circumstances. In the following century, between the years 1779 and 1790, Sir W. Herschel observed this star with his wonted diligence, and materially added to our knowledge of it e. In more recent times the name of Argelander may be singled out as specially associated with o Ceti.

b Lalande, Astronomie. Art. 794. Kepler, De Stella Nová, cap. xxiii. p. 115.

a Historiola Mire Stellæ. Fol. Gedan, 1662.

e Phil. Trans., vol. lxx. p. 338.

Algol, or B Persei, is a variable star of short period, which from its position may often be brought under notice. It is commonly of the 2nd magnitude: from that it descends to the 4th magnitude in a period of about 31, and at this it remains for about 20m. Another period of 31 then brings the star up to the 2nd magnitude, at which it remains for another period of 2d 13h, when similar changes recur. Near the epoch of maximum and minimum the variations of brilliancy proceed slowly, but at the intermediate stages they are much more rapid, and therefore noticeable. The exact period in which all these changes take place is 2d 20h 48m 55".

The observations of Argelander, Heis, and Schmidt tend to shew that the period of Algol is less than what it was in former years, but that this diminution is not uniformly progressive, inasmuch as an augmentation has now set in; and it may be inferred that future and long-continued observations will result in this change of period itself being found to be periodical.

The variability of Algol was discovered by Montanari in 1669 and by Maraldi in 1694: its period was determined by Goodricke in 1782, who also may be said to have re-discovered its variability.

Cephei is another variable star which derives additional interesti from the fact that its position in the heavens permits frequent observation of it in these latitudes. Its period is 5d 8h 47m, counting from minimum to minimum, and its range from between the 3rd and 4th to the 5th magnitude. The interval between the maximum and minimum is greater than that between the minimum and maximum, the former being 3d 19h, the latter only 1a 14". The variability of this star was discovered by Goodricke in 1784.

B Lyræ is a variable star, remarkable as having a double maximum and minimum within its simple period. Goodricke, the discoverer, assigned to it a period of about 634, but the more recent observations of Argelander shew that the true period is double this; or, more exactly, 12d 21h 53m-thus set forth:

f The Saxon farmer Palitzch, noted in 1758, is stated by Sir J. Herschel to for his early detection of Halley's comet have done the same thing.

8 Argelander, Ast. Nach., 624.

Starting from a maximum when the star is of mag. 34 it reaches the first minimum of mag. 43; then follows a second maximum, and after that a second minimum, but at this second period of least light the star is fainter than before, being only equal to a 45 magnitude. Argelander has further ascertained that, as in the case of o Ceti, the period of B Lyræ is itself variable; that down to the year 1840 it was increasing, but that then it began to decrease, which state of things still continues. The annual amount of the increment gradually grew less till the stationary epoch, whence we may anticipate by analogy that now the decrement will gradually become more rapid.

The variable star R Coronæ Borealis is noticeable from the fact that on some occasions the fluctuations in brightness between the maximum and minimum epochs are so inconsiderable as to be scarcely perceptible, but that after some years of these scarcely sensible variations, the fluctuations become so considerable that at its minimum the star entirely disappears. To Argelander also we owe the knowledge of this fact. The period of this star is 323 days. At its maximum its brilliancy is that of a star of the 6th magnitude. Its variability was discovered by Pigott in 1795.

η

Perhaps the most remarkable variable star with which we are acquainted is 7 Argûs-an object unfortunately not visible in this hemisphere, at least in our latitude of it. The following historical notes, down to 1850, have been strung together by Humboldt 1 :—

As early as the year 1677, Halley, on his return from St. Helena, frequently expressed a doubt respecting the constancy of the brightness of the stars in the constellation Argo; he had especially in his mind those belonging to the prow and the deck, the magnitudes of which had been indicated by Ptolemy. But the uncertainty of the ancient designations, the numerous variations of the manuscript of the Almagest, and especially the difficulty of obtaining exact evaluations of the brightness of the stars, did not permit him to transform his suspicions into a certainty. In 1677 he classed 7 Argûs among the stars of the 4th mag

η

h Quoted in Arago's Pop. Ast., vol. i. pp. 258, Eng, ed.

nitude; in 1751 La Caille found it to be of the 2nd magnitude. Subsequently it resumed its original appearance, for Burchell, during his residence in South Africa from 1811 to 1815, noted it to be of mag. 4. From 1822 to 1826 it appeared to be of mag. 2 to Brisbane in New South Wales, and Fallows at the Cape. In 1827 Burchell, then residing at St. Paul in Brazil, found it to be of mag. 1, and almost as bright as a Crucis. A year afterwards it had decreased to the 2nd magnitude. To this class it still belonged on Feb. 29, 1828, when Burchell observed it at Goyaz, and it is under this magnitude that Johnson and Taylor have entered it in their catalogues, 1829-1833. When Sir J. Herschel was at the Cape between 1834 and 1837 he placed it constantly between mags. 2 and 1: but on Dec. 16 in the latter year, whilst scrutinising the stars lying around the great nebula in Argo, his attention was attracted towards a strange phenomenon7 Argûs, which he had so frequently observed on former occasions, had so rapidly increased in brightness as to equal a Centauri, surpassing every other star in the heavens except Canopus and Sirius. Its maximum brilliancy occurred on or about Jan. 2, 1838. Thenceforward it began to fade away; in April, however, it was still as bright as Aldebaran. This diminution went on till April 1843 without the star falling at any time below the 1st magnitude. In April a rapid augmentation set in, and according to the observations of Mackay at Calcutta and Maclear at the Cape, ʼn Argûs surpassed Canopus and scarcely fell short of Sirius in brilliancy. Under date of Feb. 1850 Lieut. Gilliss, then in Chili, reported 7 Argûs to be of a reddish yellow colour, somewhat darker than that of Mars, and very nearly as bright as Canopus.

n

Some remarkable circumstances connected with 7 Argûs and the nebula surrounding it will more appropriately be adverted to in the next chapter.

Several explanations have been offered to account for the phenomenon of a variable star, but none are satisfactory on account of the irregularities of the periods offering a bar to any hypothesis supposing a` regular series of changes. Bouilliaud, in the case of o Ceti, ascribed its variability to its being a globe rotating on