206 G. H. KNIBBS. 



In (6), Kapteyn attempted to determine the value of the unit 

 V: using the 51 stars of Kempf's 1892 determination and the 

 value above mentioned for the direction of the apex, he obtained 

 instead of Kempf's result previously quoted, 



V= 12*3 ± 3-0 kilometres per second: 

 varying the weights however, and solving by the method of least 

 squares, this was altered to 



F=10*7 ± 3*1 kilometres per second. 

 After a further discussion, as to the inclusion and treatment of 

 certain star-groups, and also upon the weight of the results, 

 Kapteyn submitted as the most probable values — 

 Solar velocity 7=167 kilometres per second = 3*53 E annually 

 Meanstellarvel.[s] = 3M „ „ =6-57 E „ 



E denoting the mean distance from the sun from the earth. 



With respect to (c), Kapteyn's investigation appeared to shew 

 that the average parallax (x) of a star of the photometric magni- 

 tude (m) can be expressed by the equation — 



7T m = & m 7T , 



and if k be taken as equal to l/v2, or '7071, then tt will have the 

 following numerical values, viz., 



For spectral type I. tt o = 0-"063 

 „ II. „ 0143 



Or for all stars ,, 0-106. 



In regard to (d) Kapteyn concluded that any error resulting 

 from defect in the precession constant will not prejudice the 

 results obtained more than about two per cent. 



(123) Newcomb, 1899 (March) — Newcomb further examined 

 the question of the quantity and direction of the solar motion, in 

 the March number of the Astronomical Journal for 1899. 1 In the 

 iirst part of his paper he considered "the absolute speed of the 

 solar motion derived from the observed parallax of stars." Using 

 for the coordinates of the solar apex the values 



1 Some points relating to the Solar motion and the mean parallax of 

 stars of different orders of magnitude. — Astr. Journ., Vol. xx., (No. 457) 

 pp. 1-6, 1899. 



