166 cosmos. 



itself also variable, appears very problematical. Struve* 

 himself merely says, Suspicor minorem esse variabilem. 

 Variability is by no means a necessary concomitant of red- 

 ness. There are many red stars : some of them veiy red — « 

 as Arcturus and Aldebaran — in which, however, no variabil- 

 ity has as yet been discovered. And it is also more than 

 doubtful in the case of a star of Cepheus (No. 7582 of the 

 catalogue of the British Association), which, on account of 

 its extreme redness, has been called by William Herschel 

 the Garnet Star (1782). 



It would be difficult to indicate the number of periodically 

 variable stars for the reason that the periods already determ- 

 ined are all irregular and uncertain, even if there were no 

 other reasons. The two variable stars of Pegasus, as well 

 as a Hydrse, e Aurigse, and a Cassiopeise, have not the cer- 

 tainty that belongs to Mira Ceti, Algol, and 6 Cephei. In 

 inserting them, therefore, in a table, much will depend on 

 the degree of certainty we are disposed to be content with. 

 Argelander, as will be seen from the table at the close of 

 this investigation, reckons the number of satisfactorily de- 

 termined periods at only twenty-four. f 



The phenomenon of variability is found not only both in 

 fed and in some white stars, but also in stars of the most di- 

 versified magnitude ; as, for example, in a star of the first 

 magnitude, a Orionis ; by Mira Ceti, a Hydras, a Cassiopeia, 

 and (3 Pegasi, of the second magnitude ; (3 Persei, of the 2 # 3d 

 magnitude ; and in 7) Aquilse, and /3 Lyres, of the 3 -4th mag- 

 nitude. There are also variable stars, and, indeed, in far 

 greater numbers, of the sixth to the ninth magnitude, such 

 as the variabiles Coronse, Virginis, Cancri, et Aquarii. The 

 star x Cygni likewise presents very great fluctuations at its 

 maximum. 



* Compare Madler, Astr., s. 438, note 12, with Struve, Stellarum 

 compos. Mensurce Microm., p. 97 and 98, star 2140. "I believe," says 

 Argelander, "it is extremely difficult with a telescope having a great 

 power of illumination to estimate rightly the brightness of two such 

 different stars as the two components of a Herculis. My experience 

 is strongly against the variability of the companion ; or, during my 

 many observations in the daytime with the telescopes of the meridian 

 circles of Abo, Helsingfors, and Bonn, I have never seen a Herculis 

 single, which would assuredly have been the case if the companion at 

 its minimum were of the seventh magnitude. I believe the latter to 

 be constant, and of the fifth or 5*6th magnitude." 



t Madler's Table (Astron., s. 435) contains eighteen stars, with widely 

 differing numerical elements. Sir John Herschel enumerates more than 

 forty -five, including those mentioned in the notes. — Outlines, § 819-826 



