THE GULF STEEAM. 413 



the coasts of Georgia and part of South Carolina. In the first instance, 

 the Southerly winds forced the Current to the edge of soundings, where 

 it then ran from 1^ to 2 knots ; and in the latter instance, the Easterly 

 wind forced the Current upon soundings. With West and N.W. winds, 

 the Stream would be removed some miles farther off. 



The Survey steamer Blake, in July, 1880, when South-Eastward of 

 Cape Eomain, for 15 miles or more from the 100-fathoms Hne, experienced 

 a strong current running to the S.W., dragging the vessel, with her trawl 

 down in 142 fathoms, at the rate of 2 knots an hour. The same thing 

 occurred in June, 1886. 



Erom what has been said, it is clear that the eddies about the edges 

 of the Stream must vary according to the circumstances above explained. 

 Along these edges, but more particularly along the outer edge, there is 

 generally a current in a different direction, which is accelerated by the 

 wind in proportion to its strength, when blowing contrary to the Stream, 

 and retarded, and perhaps altogether obstructed, by the wind blowing in 

 the direction of the Stream. In the latter case the limits of the Stream 

 will be extended. 



In December, 1887, the barque Grenada, bound from Laguna to New 

 York, found no current whatever in the Gulf Stream, experiencing strong 

 head winds. Between February 14th — 18th, 1888, the schooner Elhridge 

 Souther, crossed the Gulf Stream area several times between lat. 28° and 

 33° N., and experienced no current, with N.W. and N.E. winds. 



(406.) Conclusion. — We have thus described this famous Stream, per- 

 haps at greater length than is needed, from its origin in the Caribbean Sea 

 and Mexican Gulf to the offings of Cape Cod and Nantucket, a portion of 

 its course much better known than any other, and more easy of definition, 

 which is not the case with its farther progress. 



The width of the Stream, between the Tortugas and Havana, has been 

 estimated to average only about 40 miles, and not 1,200 feet deep, 

 travelling with a mean annual velocity on the surface of 32-7 miles per 

 day, as previously estimated in this work, or of 30-7 miles as calculated 

 from the Meteorological Office charts. 



Passing onwards to the Narrows of the Gulf, between the Bemini 

 Islands and Cape Florida, distant 215 miles from the first section, we 

 come to that part which, beyond contradiction, decides the whole cha- 

 racter of the Gulf Stream, as the whole of it passes over this outfall. 

 The Stream is here 39^ miles wide, and the water above the temperature 

 of 60° is about 1,200 feet deep in the centre of the Stream. The sectional 

 area of the Stream in this part may therefore be taken at about 6-64 

 square miles. 



Its Velocity here was calculated, in former editions of this work, to be 

 about 65-4 miles per day on an annual mean ; but the data of the Meteor- 

 ological Department (1872) gave an annual mean of 48-0 miles per day (in 

 Spring, 43-6 miles ; Summer, 45-1 miles; Autumn, 46-8 miles; and Winter, 

 63 "4 miles per day) ; which is contrary to previous calculations. The 

 later American investigations make it about 72 miles a day in the axis. 



Now, as the inference is that the colder substratum of the Gulf Stream 

 is moving in a direction opposite to its course, that is, Southwardly, there 



