VANISHED STARS. 163 



bricius as suddenly shining in the neck of Cetus in the year 

 1596 ; and as disappearing in October of the same year, like- 

 wise advances this position as a proof to the contrary. (Kep- 

 ler, De Stella Nova Serp., p. 112.) Is it allowable to in- 

 fer, from the frequent lighting up of such stars in the same 

 constellations, that in certain regions of space those, name- 

 ly, where Cassiopeia and Scorpio are to be seen the condi- 

 tions of their illuminations are favored by certain local re- 

 lations ? Do such stars as are peculiarly fitted for the ex- 

 plosive temporary processes of light especially lie in those 

 directions ? 



The stars whose luminosity was of the shortest duration 

 were those of 389, 827, and 1012. In the first of the above- 

 named years, the luminosity continued only for three weeks ; 

 in the second, four months ; in the third, three. On the 

 other hand, Tycho Brahe's star in Cassiopeia continued to 

 shine for seventeen months ; while Kepler's star in Cygnus 

 (1600) was visible fully twenty-one years before it totally 

 disappeared. It was again seen in 1655, and still of the 

 third magnitude, as at its first appearance, and afterward 

 dwindled down to the sixth magnitude, without, however 

 (according to Argelander's observations), being entitled to 

 rank among periodically variable stars. 



Stars that have disappeared. The observation and 

 enumeration of stars that have disappeared is of importance 

 for discovering the great number of small planets which prob- 

 ably belong to our solar system. Notwithstanding, however, 

 the great accuracy of the catalogued positions of telescopic 

 fixed stars and of modern star-maps, the certainty of convic- 

 tion that a star in the heavens has actually disappeared since 

 a certain epoch can only be arrived at with great caution. 

 Errors of actual observation, of reduction, and of the press,* 



* On instances of stars which have not disappeared, see Argelander, 

 in Schumacher's Astronom. Nachr., No. 624, s. 371. To adduce an ex- 

 ample from antiquity, I may point to the fact that the carelessness with 

 which Aratus compiled his poetical catalogue of the stars has led to the 

 often-renewed question whether Vega Lyra? is a new star, or one which 

 varies in long periods. For instance, Aratus asserts that the constella- 

 tion of Lyra consists wholly of small stars. It is singular that Hippar- 

 chus, in his Commentary, does not notice this mistake, especially as he 

 censures Aratus for his statements as to the relative intensity of light in 

 the stars of Cassiopeia and Ophiuchus. All this, however, is only ac- 

 cidental and not demonstrative ; for when Aratus also ascribes to Cyg- 

 nus none but stars " of moderate brilliancy," Hipparchus expressly re- 

 futes this error, and adds the remark that the bright star in the Swan 



