42 THE PHYSICAL GEOGRAPHY OF THE SEA. 
Frozen Ocean (Plate IX.). Accordingly, they take 47), in obe- 
dience to this physical law, the most direct course by which na- 
ture will permit them to reach their destination. And this course, 
as already remarked 49), is nearly that of the great circle, and 
exactly that of the supposed cannon ball. 
51. Many philosophers have expressed the opinion— indeed, the 
belief is common among mariners (§ 46) — that the coasts of the 
United States and the Shoals of Nantucket turn the Gulf Stream 
toward the east ; but if the view I have been endeavoring to make 
clear be correct — and I think it is — it appears that the course of 
the Gulf Stream is fixed and prescribed by exactly the same laws 
that require the planets to revolve in orbits, the planes of which 
shall pass through the centre of the sun ; and that, were the Nan- 
tucket Shoals not in existence, the course of the Gulf Stream, in 
the main, would be exactly as it is and where it is. The Gulf 
Stream is bound over to the North Sea and Bay of Biscay partly 
for the reason, perhaps, that the waters there are lighter than 
those of the Mexican Gulf 36);* and if the Shoals of Nantuck- 
et were not in existence, it could not pursue a more direct route. 
The Grand Banks, however, are encroaching, and cold currents 
from the north come down upon it : they may, and probably do, 
assist now and then to turn it aside. 
52. Now if this explanation as to the course of the Gulf Stream 
and its eastward tendency hold good^ a current setting from the 
north toward the south should have a westward tendency. It 
should also move in a great circle (§ 49), or rather in the circle of 
trajection, calling thus the circle traced upon the earth which 
would be described by a trajectile moving through the air without 
resistance and for a great distance. Accordingly, and in obedi- 
ence to the propelling powers, derived from the rate at which dif- 
ferent parallels are whirled around in diurnal motion, we find the 
current from the north, which meets the Gulf Stream on the Grand 
Banks (Plate IX.), taking a sowthw estwardly direction, as already 
described (^ 45). It runs down to the tropics by the side of the 
Gulf Stream, and stretches as far to the west as our own shores 
will allow. Yet, in the face of these facts, and in spite of this 
force, both Major Rennell and M. Arago make the coasts of the 
* The waters of the Atlantic generally contain 5i per cent, more of saline matter 
than those of the English Channel. — M. Bouillon la Grange. 
