ASTRONOMICAL DETERMINATIONS. 21 



In the selection of a station, care must be taken to avoid a locality 

 where, for any cause, the ground is liable to be seriously jarred, as, for in- 

 stance, proximity to a railroad track or to a street over which heavy 

 wagons pass. It should have a clear view from the southern horizon thr< nigh 

 the zenith to the northern horizon. It is desirable to locate at a convenient 

 distance from a telegraph station, as it is necessary to bring a wire in from 

 such station for the purpose of comparing chronometers. If possible, the 

 station should be selected upon a public reservation, in order that the per- 

 manence of the monument marking the spot, which is to be erected, may 

 be assured. But, in any event, one should avoid a locality in which such a 

 monument is likely to be disturbed. 



The support of the instrument should consist of a brick pier sunken 

 fully three feet in the ground and rising above it to the requisite height. 

 Upon this should be placed for the immediate support of the instrument, a 

 block of stone well set in mortar. The chronograph may be set up on an 

 ordinary table. < her all should be erected a wall tent with a slit closed 

 by flaps in the roof, which can lie opened when observing. The instrument 

 is set up upon the pier, collimated, leveled, and the vertically of the 

 threads tested as accurately as possible, and is then pointed upon the pole 

 star. This places it somewhere near the meridian. Then taking the time 

 of transit of a star which culminates close to the zenith, and comparing this 

 time with the right ascension of the star, a sufficiently close approximation 

 to the clock error is obtained for use in placing the instrument in the meri- 

 dian. The instrument is then turned in azimuth to point upon some close 

 circum-polar star approaching upper or lower culmination, moving the in- 

 strument in azimuth with the tangent screw so as to keep the star under 

 the middle wire up to the instant of culmination. If this is done accurately 

 at the first attempt, the instrument is placed nearly in the meridian and is 

 ready for work, but it commonly happens that more than one trial is 

 required before the meridian is reached. In any case, the result should be 

 verified by a second star, before proceeding with the observations. 



OBSERVATIONS FOR LATITUDE. 



As preliminary to this work it is necessary to prepare a list of pairs of 

 stars, the two stars of each pair having such zenith distances that they will 



