The Laws Observatory. 493 
B. A. C. 7565 as given in the Coast Survey Catalogue, from 
which, as stated above, the pairs were selected. Prof. Ficklin 
actually observed on two nights an eighth magnitude star, Dm. 
49°,3597, which happens to be very near the place given for 
B. A. C. 7565 in the Coast Survey Catalogue. 
The mean declinations for 1875.0 of the stars observed, to- 
gether with other useful data pertaining to them, have been 
furnished me by Prof. T. H. Safford, Director of the Field 
Memorial Observatory of Williams College, and I desire in 
this place to make grateful acknowledgment of his kindness. 
The declinations for 1875.0 of the stars with proper motions, 
probable errors, etc., as furnished by Prof. Safford are given 
in Table II. These declinations have been rigorously 
reduced to the system of declinations of the Berliner Jahr- 
buch. All the stars, excepting Groombridge 2865 are con- 
tained in Prof. Safford’s Catalogue of 2018 Stars. But the 
declinations as given by Prof. Safford in Table II are deduced 
almost entirely from observations more modern than those 
available in 1876, when this catalogue was prepared. The 
average mean epoch of the observations used is 1875. 
The reductions to apparent place have been made with the 
Independent Star Numbers of the American Ephemeris and 
corrections for proper motion during the year have been 
applied. The results of the observations of each pair are 
collected in Table III. 
Weights. Of the results of 53 observations for latitude a 
few are so discordant that there can be no question as to the 
advisability or necessity rather of rejecting them. I have 
concluded to reject all observations whose results differ from 
the mean by more than three seconds. This excludes five of 
the 53 values of the latitude. They are marked with 
brackets in Table VI. The best observation rejected differs 
from the mean by 3”.4 and the most discordant observation 
retained differs from the mean by 2”.6. 
In discussing in the usual manner the results of the 
remaining 48 observations I find that the probable ied! of 
the latitude resulting from a single observation of a pair is 
+ 0”.70. This includes the probable error of the half sum of 
the declinations. The probable error of one observation 
