( 376 ) 



period of the empirical term \\iiicii Newcomb fixed on 273 years. 

 Hoping to find some indication abont tlie empirical law which wonld 

 represent tlie outstanding differences, I put togetlier for the wliole 

 period 1847 — 1902 the vahies of the mean annnal errors in h)ngitiuie 

 or in R. A. according to the observations at Green wicli. Those as 

 far as 1882 ^) were borrowed from Stone's papers in the Month/. 

 JSfot., applying to liis results Newcomb's corrections, ^yhile those 

 for the snbseqnent years were taken from the Annnal Repoi'ts of 

 the Astronomical Society. To all tiiese resnlts the small corrections 

 were apphed for the reducliou of the observations to the same standard 

 time-star catalogue. As such I adopted the 2"^^ 10 year Catalogue. 

 I added to the Greenwich results: for the years 1862 — 1874 

 Newcomb's residts which partly depend on the Washington ol)serva- 

 tions, for the years 1880—1892 the results of the ol)servations at 

 Gxford as given by Stone, ap})lying to both Newcomb's corrections 

 and for the years 1895 — 1902 the resnlts derived from the Green- 

 wich observations by myself {\ (1 -|- H) in the first table of section 

 5). From a compai'ison of the resnlts of different observatories for 

 the same year we may infer that the}' are tolerably accurate. The 

 differences between my results and those computed at Greenwich 

 range from 0."00 to 0."36. 



I do not, however, give these annual mean errors here, as I did 

 not succeed in deriving anything from them with certainty, l^y 

 assuming the existence of a new inequality with a period of about 

 50 years with maxima about 1862 and 1887 and a coefficient of 

 about 3" we should attain a somewhat better, but even then not 

 an absolute agreement. 



So the only thing I could do to obtain the mean corrections 

 required for ni}' purpose, was to represent the annual mean errors 

 from 1886 — 1902 by a smooth curve. The following values were 

 derived from it. 



1895.0 (fA= + 0"53 



1896.0 1.06 



1897.0 1.44 



1898.0 1.72 



1899.0 1.93 



1900.0 2.09 



1901.0 2.21 



1902.0 2.28 



1903.0 2.30 



1) For the years 1847—1861 the new reduction of the Greenwich observations 

 of the moon {Monthl. Not. Vol. 50) was used. 



