22 A MANUAL OF TOPOGRAPHIC METHODS. 



culminate at nearly equal distances from the zenith, one to the north and 

 the other to the south of it. Such a list can be prepared from the Safford 

 Catalogue of the Wheeler Survey. For this it is necessary to know the 

 approximate latitude of the station, the right ascensions and the declina- 

 tions of the stars. The zenith distance of a star is equal to its declination, 

 minus the latitude of the place. The stars of each pair should culminate 

 within a few minutes of one another. They must be observed consecu- 

 tively, and, therefore, those stars should be selected which culminate as 

 nearly as possible together,- leaving only a sufficient interval of time 

 between them for setting the instrument. 



Before beginning to observe, the instrument should be closely collimated 

 and drawn into the meridian. 



Upon the approach of the first star of the pair to the meridian, the 

 instrument should lie set for it, using the vertical circle for that purpose, 

 and setting the spirit level upon the vertical circle as nearly level as possi- 

 ble. Then, as the star traverses the field of the telescope, keep the movable 

 thread in the reticule upon it by means of the micrometer screw until it 

 crosses the middle vertical thread. Then read and record the micrometer 

 and the two ends of the level bubble. Without disturbing in the slightest 

 degree the setting of the telescope, turn the entire instrument 180° upon its 

 bed plate, when it will point north of the zenith, at the same angle that it 

 formerly pointed south, or vice versa, as the case may be, and will be set 

 for the other star upon the opposite side of t'-e zenith. As this approaches 

 culmination, follow it with the micrometer as before, until it reaches the 

 middle thread; then record as before the readings of the micrometer and of 

 the level, whether it has changed or not. 



This constitutes the observations upon a single pair of stars. For the 

 determination of latitude twenty such pairs of stars should be observed 

 each evening, if possible, and the same pairs of stars should, also assuming 

 it to be possible, be observed upon other evenings. The following exam- 

 ple, taken from observations at Rapid, South Dakota, shows a portion of 

 the star list and the form of record: 



