358 KRETZ. 



formulae and table following, which I reproduce from Dr. Da- 

 vis' memoir : 



To Boss' value add the quantity 



»z+A'(r— 1883) when r< 1866 



or 



w + A'''(r— 1883) when r> 1866. 



Here 



m = K. G. C— Boss (good for 1883) 



and is obtained from the Berlin Jahrbuch, 1884, Appendix. 

 K and K' are the annual variations of in computed as shown 

 in the table, p. 359, in which we assume Ad^ = o. 



There still remained a number of cases to be treated, how- 

 ever, chiefly right ascensions, for Boss gives corrections to the 

 declinations only. For all of these I deduced corrections by 

 direct comparison with some suitable catalogue whose system 

 was well known. The labor was greatly simplified by the fact, 

 that no account had to be taken of change in right ascension or 

 in declination, as my stars are all situated within a few degrees of 

 each other. The rule laid down was to compare as many stars 

 as possible (usually about 12) within not more than one hour 

 in right ascension, and five degrees in declination on either side 

 of the center of my plate. Systematic corrections were thus 

 deduced for the following catalogues : 



Bessel [{^) of Sect. I), Zones 4.6^ and 50J. — Auwers. in his 

 zone of the A. G. C, gives corrections to all those of Bessel's 

 zones which fall within the limits of his catalogue. He shows 

 that they consist of two parts, a systematic one, depending on 

 the constants used in the reductions, and one due purely to ac- 

 cidental causes. Luther's tables fail to eliminate the latter class. 

 Without attempting to distinguish between them, I deduced the 

 total amount by direct comparison with the A. G. C., correcting 

 for proper motion whenever that was possible. I find thus : 



A. G. C.— Bessel, Zone 464 (in «) = + o M90 

 " " " " (in (5) = — 3^'. 60 



A. G. C. — Bessel, Zone 503 (in a) — -Y 0M22 

 " " (inc5) = +3^''.l5 



(18) 



