4 PROCEEDINGS OP THE AMERICAN ACADEMY 



I can depend upon the stability of the soil in the Observatory grounds. 

 This collimator has an object-glass of five inches in diameter mounted 

 upon a pier south of the circle ; on the same pier are Y's for the 

 other collimator, and the object-glass is between them, but can be 

 turned down so that the telescope of two and one half inches' aperture 

 can be placed in the Y's over it. The telescope, which has pivots 

 and a level, is usually placed upon the north collimator pier, which 

 has Y's, but no object-glass. 



In the ordinary position, then, we have two collimators which 

 can be set upon each other; the focal mark is one hundred Paris 

 feet south of the five-inch object-glass, and the north collimator has 

 an excellent micrometer. 



"When the north collimator telescope is transferred to the south Y's, 

 it can be used with its level for the determination of the horizontal 

 point, as well as in its usual position. 



The whole arrangement is ingenious, original, and economical, and 

 admirably adapted to an observatory whose means are insufficient for 

 strictly primary work, while, with the addition of a few not very ex- 

 pensive auxiliaries, it can be readily adapted to primary work. The 

 object-glass and eye-piece are interchangeable, a requirement which 

 I should always make for any fixed instrument, (I believe the Rep- 

 sokls always construct their large instruments so,) and there is no 

 difficulty whatever in observing the nadir whenever we please. 



But at present I have no intention of attempting a strictly primary 

 catalogue ; enough such catalogues are now in course of construction, 

 and I need only to take care that the casual errors of the primary 

 catalogue which I employ do not become systematic errors in my re- 

 sults. Of this there is some danger when the Jahrbuch in its present 

 condition is employed without discriminating between well and ill deter- 

 mined polars, — that is, without reading Professor Auwers's memoir. 



The general arrangement of the tables which follow is like that in 

 my previous paper, Vol. XIX. of the Proceedings, pages 324-352. 

 The mean right ascensions are given in two different ways, according 

 as their places are or are not given in one of the great ephemerides. 

 In the former case the datum of each observation is the correction to 

 the ephemeris ; and the mean right ascensions for 1884.0 are obtained 

 by applying to that given in the ephemeris the mean of these correc- 

 tions. For other stars no proper motions are here applied. 



As the observations reduced to 1885.0 will soon follow these, I 

 have thought it needless to form a catalogue of mean right ascensions 

 for this paper. 



