164 SCIENTIFIC EECORD FOR 1884. 



den) which have also been observed at Leydeu. A discussion of these 

 declinations is now in progress. 



For eliminating personal equation depending on the direction of the 

 star's motion, each star has been observed over half the wires in the 

 usual way, and over the other half through a reversing prism. 



All the observers make the clock sloicer when the stars move across 

 the field from left to right. 



Leyden Observatory. — Since 1881, 427 observations have been made, 

 and 1,124 in all ; 165 of the 303 stars have been observed once or oftener. 



Melbourne Observatory. — The observations will be begun in 1883. 



Sydney Observatory. — Sydney also proposes to observe these stars. 



Naples Observatory. — The observations will be begun in January, 1884. 



Washburn Observatory. — The observations were begun in May, 1884, 

 and up to January 1, 18S5, 1,800 observations have been made and are 

 completely reduced. The probable error of a single decl. is 0".41. The 

 constant of refraction of the Pulkowa Eefraction.tables requires a cor- 

 rection of + 0".30. 



A uniform ephemeris of the clock-stars,* — A noteworthy feature in the 

 progress of our astronomical annuals is the rapid increase that has taken 

 place within a few years in the number of the clock-stars. This increase 

 seems to have been brought about chiefly b^' the influence of the Berlin 

 list of 539 stars, which was published for the use of asti'ouomers who 

 were engaged in observing the zones of the northern heavens. The 

 convenience of this list for the use of observing parties in the field seems 

 to have led to the large increase that we find in the annual publications. 

 For the year 1886 the Nautical Almanac gives the mean positions of 198 

 stars, the Coniiaissance des Temps of 316, the Berliner Jahrbuch of 622, 

 and the American Ephemeris of 383 stars. 



That this great increase in the number of clock-stars has some ad- 

 vantages will not be denied, but, on the other hand, there seem to be 

 some disadvantages that ought to be considered. In the first place the 

 mean positions of the stars given in the different publications do not 

 agree as well as might be expected when the great number of observa- 

 tions of these stars is taken into account and the number of years over 

 which the observations are extended. These differences of ijosition 

 are indeed small, and their influence on the observed positions of other 

 stars and planets cannot be great ; but we have already examples in 

 astronomy of the extension of such small errors into large catalogues of 

 stars and of their development into periodic errors. 



The constants for the reduction of the clock-stars from one epoch to 

 another, and to apparent position, are now so well known, and the 

 adopted values are so nearly the same, that the differences produced by 

 the various reductions are frequently neglected ; but even here it f^eems 

 there is room for improvement. At the present time the values of the 



*Astr.Nacbr.,No.260:>. 



