252 On the Errors in Lo7igUude 



The rates from August 12 to 26, were determined by the smi's 

 transit, and those from August 9th (the day on which the chrono- 

 meters were landed) to the 12th, by a comparison with the clock, 

 supposing its mean rate +69*"5, as no observation occurred 

 during this interval ; and by this table it appears, that the chro- 

 nometers when landed were rapidly diminishing their gaining 

 rates, and increasing their losing ones. In the others the effect 

 was almost immediate. 



The clock and chronometers were likewise landed upon a small 

 island in Fair Haven, on the north coast of Spitzbergen, on the 

 30th June ; and, as the same circumstances occurred, it will be 

 needless to detail them. 



The following table is intended to show the difference between 

 the rates on board the ships and what they would have been had 

 they been on shore. 



The errors of the chronometers in April, were those obtained 

 in London l)eforc the ships sailed ; those on the 25th of August 

 were determined at the observatory on Dane's Island, Spitzber- 



aen. 



