Drift Ice and Currents of the North Atlantic. 387 



It is well known that vessels in the northern part of the Gulf 

 Stream, while steering parallel to its general course, have met 

 with successive and striking changes in the temperature of the 

 water and sometimes with ice, to the southward of Nova Scotia 

 and Newfoundland, and in the proper line of the polar current. 

 This is well shown in the journals of the ships Eliza and Grand 

 Turk, as published in some former editions of the Coast Pilot. 

 In latitude 41° 53', longitude 56° 52', the Eliza found the water 

 at the depth of 70 fathoms two degrees warmer than that at the 

 surface, the temperature of the latter being 40°, and an ice island 

 bearing S. S. E., distant seven miles. S. S. W. and S. of the 

 Grand Bank, and in nearly the above latitude, the Eliza again 

 met with cold water and passed several ice islands. Rennell has 

 also recognized these cold veins or bodies of water in the Gulf 

 Stream. It appears, therefore, that in this portion of the Gulf 

 Stream, neither its presence nor its actual limits can be determined 

 with certainty by the thermometer, during the ice season. 



It appears in the pages of the Coast Pilot, that Capt. Billings, 

 in June, 1791, found the temperature of the water in the Gulf 

 Stream to have fallen ten degrees, in latitude 39°, southward of 

 the Bank, and that the like had been observed by Dr. Franklin 

 and Col. Williams, in the same region. But, judging from the 

 latitude, it is not improbable that these observations were made 

 to the southward of the true border of the Gulf Stream. If this 

 be the true solution, it is indicative of the partial re-appearance of 

 the polar current, after passing beneath the Gulf Stream ; and 

 there is evidence of its further course to the S. W. and W. S. W., 

 near the border of the Stream. 



This leads us to notice a probable, if not a principal cause of 

 the great variations, which have been reported, in the position 

 and limits of the Gulf Stream, in its eastward progress. Rennell, 

 we conceive, rightly supposes an overflow or outspreading of the 

 Gulf Stream upon the ocean waters, as it proceeds in its course. 

 Now we know, from well established cases, that overflowing 

 streams, upon denser waters, are often very shallow ; and Capt. 

 Bayfield has shown, in the case of the estuary and Gulf of St. 

 Lawrence, that the effect of a storm is to break up, for the time, 

 this superficial current and amalgamate it with the deeper and 

 colder waters. Hence we may infer, that in good weather and 

 a smooth sea, the thermometric breadth of the Gulf Stream may 



