THERMOMETER in NAVIGATION. gi 



the other from the north-weft. That this is a branch of 

 the gulf-ftream is rendered probable by the appearance of 

 great quantities of gulf-weed on the i 8th A. M. and 

 the circumftance of the flying-hili appearing on the 19th 

 which probably had followed the warm ftreani into an 

 higher latitude than I can, after looking into many voy- 

 ages, find them to have been previouily noticed. It will 

 appear alfo from the homeward bound track, that on the 

 I 8th of Aug. A. M. we itruck a warm current and con- 

 tinued in it feveral days, which from the longitude could 

 be no other than the current before noticed in 1794, as 

 after quitting the gulf-ftream, we had been for feveral 

 days in the feas cooled by the proximity of the banks of 

 Newfoundland. I have dwelt longer than at firft fight 

 may appear neceffary on this current, becaufe, though it 

 has been fuppofed to exifl: to the fouth-eaft of the banks 

 of Newfoundland, it has not been traced fo far north as 

 the latitude of the fuppofed Jacquet-Ifle, that is to lat. 

 47, long. 39. It is probably continued in about a noith- 

 eaft direction , and extends entirely acrofs the Atlantic, 

 till it ultimately firikcs the coafts of Ireland and the He- 

 brides, after having loft in its long courfe in thofe north- 

 ern latitudes much of its heat and at laft being reduced 

 to the temperature of the feas, through which it flow^s. 

 That fuch a current really exifts through the whole of 

 this extent is rendered highly probable from various pro- 

 dudtions of the tropical regions being frequently thrown 

 on thofe fhores, hitherto fuppofed to be the accidental 

 effects of ftorms and not of the unvarying courfe of na- 

 ture. The firft notice of fuch fubftances being caft on 

 thofe Iflands will be found in Vol. III. p. 54c, of the 

 Abridgement of the Philofophical Tranfadtions, which 

 abridgement' was publiflied in 1 749 ; but the papers 

 abridged many years before.* We here find the fadls 



M 2 ftated 



• Phil. Tranf. Vol. X. p. 396. and Vol. XIX. p. 298. 



