Dreyer — New Determination of Comtant of Precession. 621 



of Baily. But unfortunately Schumacher's tables depend too much on 

 the values given by Lalande for the auxiliary quantities necessary for 

 the reduction, Tvhich are far from exact, and Baily' s Catalogue is, 

 therefore, now of little or no use except as an index to the stars ob- 

 served by Lalande, In 1868 the late Dr. von Asten published new 

 tables of reduction, founded on a thorough discussion of Lalande's ob- 

 servations of such stars as also occur in Piazzi's Catalogue, Prom 

 these tables it became necessary to reduce anew about 3600 Bight 

 Ascensions of about 3300 stars; but although this involved a consider- 

 able amount of work, the comparison of the Eight Ascensions thus 

 derived with those of Baily showed how absolutely necessary this new 

 reduction was. I have to thank the Eoyal Irish Academy for placing 

 at my disposal a grant which enabled me to procure the assistance of 

 an experienced computer for this part of the work. It is scarcely 

 necessary to mention that I have taken into account the valuable col- 

 lection of corrections to Lalande's Observations which Argelander has 

 published in vol. vn. of the Bonn Observations. 



When the lEean Eight Ascensions for the year 1800 of all the stars 

 common to Lalande and Schjellerup had been computed, I proceeded 

 to reduce them to 1865-0, the epoch of Schjellerup's Catalogue. As 

 von Asten had used Bessel's constant for the construction of his tables, 

 and Schjellerup had used Struve's, I thought it best to adopt the 

 mean of the two constants, and deduce a correction to this mean as the 

 final result of the investigation. By taking the mean of the preces- 

 sions given for each star in Baily's and Schjellerup's Catalogue, I also 

 took the secular variation of the precession into account without any 

 trouble. I may add that I took every possible care to detect errors in 

 Baily's and Schjellerup's precessions; and whenever Argelander's cor- 

 rections to the original observations altered the place of the star for 

 1800 sufficiently to influence the third decimal of the precession, I 

 recomputed the latter quantity for the epoch 1800 with Bessel's 

 constant. 



The Eight Ascensions thus reduced to 1865 were next compared 

 with Schjellerup's results. "Wherever the difference Sch.-Lal. ex- 

 ceeded about 0^ '7, I examined the case carefully to find out whether 

 there was a case of proper motion, or whether one of the two observers 

 had failed by exactly one second (an error occasionally occurring in 

 zone observations). If this latter was the case it was generally easy 

 to detect the error by a reference to other star-catalogues, but if not, 

 the amount of proper motion was determined in a similar way, unless 



indeed it could be taken from Madler's Bradley, the Abo Catalogue, or 

 Argelander's list of 250 Proper llotions. As the majority of large 

 proper motions are independent of the motion of the sun in space, and 

 many adjacent stars are known to move in parallel directions, I 

 thought it better not to trust to the large number of stars employed 

 to do away with the effect of proper motion, but to correct for the 

 latter wherever I could detect it. 



