26 GOULD REDUCTION OF D'AGELET's OBSERVATIONS. 



Among the cases of this sort are the following : 



1783, April :i, KM Virginia. 1783, April 25, 449 Mayer. 



8, 34 Uor-s. 25, T = 11 A. 33m. 44s. 



9, a Persei. 30, T = 94. 17m. 20s. 

 13, T = 8A. 59m. I8s. July 27, 51 Piscium. 

 14,11 Leonis. 17*4, Sept. 30, 79 Pcgasi. 



18, 3 Lite. Oct. 15, 140 Tauri. 



19, T = 13A. 0m. 48s. 



In 23 instances'observations have been found so discordant, and apparently so erroneous, 

 that no plausible assumption could be made, so that it became necessary to reject the transit 

 over one thread, or one reading of the zenith-distance. For 10 stars, whose zenith-distance was 

 not observed, it was requisite in reducing to assume the declinations; but the close agreement 

 of the resultant right-ascensions has seemed in every case to justify the assumption. All of 

 th ?se cases are fully specified in the notes. In a few cases, not exceeding six or eight, the 

 discordance between the transits over different threads has left it uncertain which was errone- 

 ous ; and for these the mean has been used. 



Throughout the reductions it has become manifest that those observations for which only 

 the full minute was recorded are uncertain by two or three minutes, both in the transits and 

 the zenith-distances. The failure to detect this usage at first has occasioued much fruitless 

 labor. 



$ 1 1 . THE CATALOGUE. 



The equinox, to which the mean places of the catalogue should be referred, was fixed as 

 that of 1800.0, after much hesitation. The selection of this equinox was due to no want of 

 appreciation of the importance of referring the mean places to the mean equator of a date near 

 that at which the observations were made, but to the apparently greater advantages of the 

 epoch adopted. Excepting the circumpolar stars of Lalande, reducedby Fedorenko to the 

 mean equinox of 1790.0, all the observations of the last century, since the time of Bradley, 

 have been reduced to this fundamental epoch ; and the facility which the use of this epoch 

 affords for comparing d'Agelet's results with those of Piazzi and Lalande is too great to be 

 lightly disregarded. The mean places of the catalogue are, therefore, referred to the equinox 

 of 1800.0, but the actual date of every observation is given, and no correction for proper 

 motion has in any instance been applied; so that the place recorded is always that at which 

 the star was actually observed. 



For reducing to the mean equinox, day-numbers were computed from the A, B, C, D, E, 

 already given, (§ 7.) It may be well to record these also. 



(26) 



