Ch.il south AMERICA. 311 



the eaftward of his reckoning. The captain of the 

 Louis Erafnie foiind the difference nearly the fame as 

 the captain of the Marquis d'Antin, who made ufe of 

 an equation. Both, as I have already obfcrved, found- 

 ed their corredions on the variations ; differing very 

 confiderably from that delineated on the charts. 



The great variety of currents met with in failing 

 round Cape Horn, being fometimes ilrong, fometimes 

 moderate, and at others fcarce perceivable, induces 

 me to think that they were not confiderable in corred- 

 -ing the account. For their velocity being uncertain, 

 it is in f^cl: only committing a voluntary error ; and 

 as the variations enable us to guefs at our longitud-e 

 within two or three degrees and as after making ufe 

 of the equation^ the place of the fhip cannot be known 

 nearer the truth, the corredion is entirely ufelefs ; and 

 the inference drawn from obferving the variation, is a- 

 bundantly fufficient for fecurity. 1 fay, that the place 

 of the fliip may be known fo within two or three de- 

 grees •, and a more exadl conformity between the cor- 

 re(5led reckoiiing and the time of making land would 

 be rather fortuitous than the effect of accuracy, in 

 making the correclion. The difference of one or two 

 degrees in the variations, an error unavoidable, may 

 produce in the longitude an error of three or four de- 

 grees, or even more, according to the place of the 

 Ihip. Every one on board the three frigates, found 

 then* reckoning to the Vv'cilward of thefhip's true place, 

 ;hough they had made an allowance for the currents, 

 and the difterence between fome of their accounts was 

 not fuiall, as I have already noticed. * This was ow- 

 ing to the like uncertainty in the journals of other 



* There is a fttopg currep.t to the W. in thefe Teas, and by the 

 |>eft obfervations I could make (at a mean) it is about one mile 

 per hour, bur in feme places much more, efpecially near the 

 IDOUth of tne river Amazf>n. 



. i>i.B. I gen -rally found the above allovyiince to anfwer from 

 lat. 4 6. 10 5 ;. S. A, 



X 4 ing 



