250 On the Errors m Longiludtt 



observations may, in any way, contribute to the moie accurate 

 determination of the longitude by this method. 



The sudden alteration in the rates of chronometers when taken 

 on board of ships, has been frequently observed by intelligent sea- 

 men ; and is gencrallv ascribed to the motion of the vessels. 

 Before, however, I attempt to account for this alteration, I shall 

 first prove that it actually takes place; and, in order to do this, 

 shall relate the circumstances connected with the chronometers 

 on board the Dorothea and Trent, commanded by Capt Buchan, 

 which occurred during the late voyage to the north pole. 



Soon after the arrival of the ships on the coast of !:ipitzbergen, 

 the chronometers on board the Dorothea (five in number) were 

 found to be rapidlv gaining on their former rates as determined in 

 London previous to the ships sailing; in consequence of which 

 the land appeared considerably to the westward of its true posi- 

 tion as determined by lunar observation, and they were found to 

 be still gaining daily, which appeared not only from each sub- 

 sequent set of Ulnars, but also by comparing the longitude of dif- 

 ferent points of land determined by the chronometers, with the 

 longitude of the same points ascertained in the same way some 

 time afterwards. 



For instance : The longitude of a remarkable point of land on 

 the north-west coast of Spitzbergen, called Cloven Cliflf, was 

 found bv a mean of the observations taken with the chronometers 

 on .lune 21, 1818, to be 10" 35' 27" E.; but the longitude of the 

 same point of land on July 31, was 10" 15' 37" E., making a dif- 

 ference of no less than about 20" of longitude in five weeks ; that 

 is, estimating the longitude with the same rates and errors as de- 

 termined in London before their departure. From this, therefore, 

 it appears, thev had been gaining on their former rates, or had 

 been increasing their gaining rates, and diminishing their losing 

 ones. 



An opportunity soon afterwards occurred of observing the ef- 

 fect produced upon the chronometers by removing them on shore. 

 On tl;? 0th of August, the chronometers, nine in number, were 

 landed on an island, where a tem])orary observatory had been 

 erected for the purpose, and the latitude of which had been ac- 

 curately detern.ined with a repeating circle made byTroughton, 

 when it was found that the acceleration immediately ceased ; for 

 the longitude of the place by chronometers, August 12, was 9' 42' 

 3b" E.i but on the 27th, it was 10° 1' 0" E., making a diflferencc 

 of IS' 24" of longitude in fifteen days, using the former rates. 



Since, therefore, the chronometers were getting enstcrhj by 

 their removal on shore, the acceleration must have ceased ; which 

 will appear upon consideration. 



A similar circumstance was observed bv Lieut. Franklin to 



take 



