[ 237 ] 102 



another formula for the correction to be applied to the hour-angle, deter- 

 mined as before, in order to restore it to the exactitude of which the gen- 

 eral formula is susceptible. The rule for practice, then, is to observe the 

 greatest number of altitudes possible in a brief interval ; which ought not 

 to exceed, at most, four or five minutes of time. 



Finally, in order to render all the circumstances, as nearly as may be, 

 equal in the comparisons which have then to be instituted to assure us of 

 the daily rate of the chronometers, and the absolute time of the place, it 

 should be made a point to observe the same stars to the east and to the 

 west, as long as their presence above the horizon continues to offer favora- 

 ble positions for good hour-angles; and as the period approaches for giving 

 up some of them, before dispensing with these to iuiroduce others, it is 

 well to take in the latter, in common, for a night or two, and connect, in 

 this way, the comparisons of stars that are going out, with those of stars 

 that are coming in. 



Such are the principles and motives which have directed me in the con- 

 stant employment of the method of absolute altitudes for determining the 

 time of a place. 



111. The latitude of any place cannot be regarded as exactly determined 

 but by deduction from several series of observations, made to the north 

 and to the south of the zenith of the place. The object of this precept is 

 liere the same with that which, in determining time, obliges the observation 

 of stars to the east and to the west of the meridian : this object is to compen- 

 sate the constant errors of the instrument, or those which may arise from 

 a defective adjustment. 



Two good series taken in one night — one to the north, the other to the 

 south ol the zenith, and each of from 10 to 15 readings — can furnish, by 

 their mean result, a latitude sufficiently precise for the wants of geogra- 

 phy. Much more would be required if the station in question is destuied 

 to serve as a term of comparison in a great geodetic operation, or to be- 

 come some day the locality of an important town. An observer in this re- 

 gard, may anticipate after times, and estimate what he is allowed to do in 

 supererogation, as it were, without injury to the more special and tempo- 

 rary interests of his mission. 



In our hemisphere, the sky south of the zenith is always sufficiently full 

 of principal stars to furnish, in the course of one night, circum-meridian 

 series; and to the north, the constellations about the pole ofler to the ob- 

 server everything he needs. The North star, above all, by its proximity 

 to the pole, is cnpable of being observed at any point on its parallel, and 

 of being used to deduce the latitude of a place with as much precision as 

 if it had been taken at one or the other of its meridian passages. In the 

 short but fine season which may be annually devoted to the reconnoissance 

 of those temperate regions, such as 1 have had occasion to explore, there 

 is rarely a night in which a propitious sky does not ofler intervals favorable 

 for some observations, giving at least an approximate latitude, should it be 

 unsuitable to make a longcrhalt. 



By giving orders to the sentries of the camp, according to what he de- 

 sires to execute during the night, an active observer can always recon- 

 cile some hour of sleep with his duty, whatever little importance may 

 attach to the station, even if it be of the order of those that may be neg- 

 lected without losing much. Now that any method puts us in the way 



