REDUCTION OF OBSERVED TRANSITS. 15 



evening and morning observations of the mark, took place with uniformity 

 between the evening reading and ll h . 30 m mean time. If the mark has not 

 been observed in the morning, the mean change 2" - 5 has been used, and a 

 proportional part applied to the evening observation of the mark. This 

 azimuth-error has then been combined with those obtained from circumpolar 

 stars. For example, on November 13, Table III., the difference of azimuth 

 by the evening and morning readings of the mark is 3"-37. Clock stars were 

 observed at about 10 h , that is, four hours after the observation of the mark. 

 It is assumed that a change of 2"-45 took place in those four hours, and that, 

 therefore, the azimuth-error by the mark was 0"'92. The error used for 

 the group is the mean between this and 0"'72 found by Polaris at 9 h . 40 m . 



Table III. contains the whole of the determinations of azimuth-error and 

 the observations of the meridian mark. 



TRANSITS OF STARS AND OF THE MOON OBSERVED AT HONOLULU, AND DEDUCED 

 ERROR OF THE TRANSIT-CLOCK DENT 1916. 



37. The stars observed for clock-error were taken from the catalogue of 

 clock stars in use at the Koyal Observatory, Greenwich. The Mean Right 

 Ascensions were brought up to the beginning of the year from the Greenwich 

 Catalogue of 2760 Stars, Epoch 1864, and are identical with those adopted at 

 the Royal Observatory. For those stars found in the Nautical Almanac the 

 apparent places have been taken from that work, a small constant correction 

 being applied to reduce them to the standard of the 1864 Catalogue. The 

 Moon Culminating Stars, for which good places have not been recently 

 obtained, have been rejected. The Apparent Right Ascensions of such stars 

 of the Greenwich list as were not in the Nautical Almanac were brought up, 

 during the outward voyage, for every ten days of the period during which 

 they were likely to be required. 



38. In reducing the observed transits of clock stars the mean of the five 

 wires has been, for convenience, reduced to the center wire by the correction 

 O-02, additive when the micrometer was east. The diurnal aberration has 

 been taken as follows : 



a 



For Polaris o 80 



, , 8 Ursas Minoris o'oz 



, , Cephei 5i '4 



, , X UrsJB Minoris i o3 



, , all the Clock Stars o 02 



These quantities have been applied to the Apparent Right Ascensions 

 throughout. 



