[ 237 ] 104 



night observations. Yet, as these few have been useful to me in the con- 

 struction of the map, they furnish me with grounds of congratuhuion upon 

 my extra caution. They will be recognised, in the accompanying Tuble 

 of Geographical Position?, as latitudes expressed only in round minutes, 

 without seconds. 



IV. The geographer, if he be provided with a telescope, may determine 

 longitudes by observations of the eclipses of the satellites of Jupiier, of 

 (what is much more important) eclipses of the sun, of occultations of the 

 planets and the iSxed stars by the moon ; and, finally, of the transits of the 

 inferior planets over the sun's disc. The " Epbemerides" apprize him be- 

 forehand of the periods and of the principal places where these phenomena 

 will be visible, and furnish, also, the elements of calculation necessary as 

 well to prepare for the observations themselves, as to deduce from them the 

 longitude of the station where they shall have been made. But these phe- 

 nonjena are of rare occurrence ; and when tliey do present themselves, the 

 condition of the atmosphere does not always favor the observation. Mt ans, 

 therefore, more prompi. and even more precise, were a great desideratum 

 in science — and science has supplied them. 



Such means are found in the transit instrument. By observing with that 

 instrument the right ascension of the moon's bright limb, and the right 

 ascension of some .'tar near the moon's parallel, and not differing much 

 from her in meridian transit, it will enable the observer to determine the 

 difference between the meridian of his place and the primary meridian of 

 the "Epbemerides" in his hands, or any other meridian where he may 

 suppose that similar and corresponding observations have been made. 



But both of these methods are yet ol very rare resort for geography, 

 inasmuch as the first cost of the instruments, the difficulty and expense 

 of transportation, and even the time they require for setting up and adjust- 

 ment, are far from being always at the disposal of the exploring traveller. 

 He finds himself reduced, then, like the navigator, to the method of 

 lunar distances, both for obtaining absolute longitudes, and for control- 

 ling the differences of successive meridians indicated to him by the chro- 

 nometer. 



This method of lunar distances is laborious in the observation, as well 

 as in the calculation ; but it is certain in its results, when applied with, 

 discernment. It demands a great number of individual observations, in 

 order to attenuate the errors which may arise from defective adjustment 

 of the instrument, or from the habit of the observer in making the con- 

 tact — some being liable to have such contact too strong; others too weak. 

 A small error in the measured distance has a great effect upon the longi- 

 tude calculated; and the mode to divide these errors consists in observing 

 several series of^ distances to stars, some of which are to the east, and others 

 to the west of the moon. These series must be combined, as much as 

 possible, by couples or pairs ; — that is to say, after having taken the first 

 series on one side of the moon, the second should be then immediately 

 taken on the opposite side. It is in this manner, at least, that I have the 

 habit to do ; and it seems to me, from my experience, that more confi- 

 dence is to be placed in a longitude that is the mean result of two 

 series only, managed in the manner I have indicated, than if it were the 

 mean result of a greater number of scries taken throughout on the same 

 side of the moon. When the distances to be measured are between the 



