THE GULF STREAM. 191 
stantly changing, according to the seasons. At one time the colder 
water from Davis’s Straits spreads like a fan near the surface, driving 
the Gulf Stream to the east,* and at another, large masses of warm 
water extend towards the Faroe Islands, with branches toward Iceland 
and the coast of Portugal. 
An examination of an isothermal chart of the Atlantic clearly shows 
the effect of the isolation of the Northern Atlantic, the area of maxi- 
mum temperature (82°) extends over a far greater space in the North 
than in the South Alantic. The Gulf of Mexico and the Caribbean be- 
come greatly superheated in September (to above 86°), the effect of this 
superheating in conjunction with the westerly equatozial drift being 
seen clearly in the northerly extension of the isothermal lines. In the 
South Atlantie,t owing in part tothe greater regularity in the shape of 
the basin, the difference in the extension of the isothermal lines is but 
little marked. 
The temperature sections of the Challenger, trom Teneriffe to Som- 
brero, show remarkably well the great contrast in temperature between 
the eastern and western basins of the Atlantic, which are separated by 
the Dolphin Rise. In the eastern basin the cold water on the bottom 
is supplied by the indratt from the South Atlantic, while the warmer 
surface water of the western basin is due to the westerly equatoriai 
currents. We seem, therefore, to have masses of water of different 
temperatures accumulated at certain points by surface or bottom cur- 
rents, to be distributed again, either north or south, into the generat 
oceanic circulation, thus restoring the equilibrium disturbed by the un- 
equal distribution of heat and cold on the surface of the ocean. 
Another temperature section (Fig. 1), which I shall borrow from the 
Challenger soundings, to complement the work of the Blake in the same 
regions, is that which extends from Halifax to the Bermudas, and thence 
to St. Thomas. The temperatures observed by these vessels show 
plainly the path of the warm surface water, which flows outside of the 
West India Islands, and joins the Gulf Stream proper, whose waters 
when united are banked against the cold Labrador current in its course 
along the American coast. 
Undoubtedly, the early observations made upon the temperature of 
the ocean were defective, owing to the somewhat imperfect instruments 
at the disposal of the early explorers; yet they determined the general 
position of the cold and warm currents of the ocean along our shores. 
*The direction from which the currents come is plainly shown by the nature of the 
bottom specimens, made up in part of globigerinze brought by the warmer southerly 
surface currents, and in part of northern foraminifera and of volcanic sand derived 
from Jan Mayen and Spitzbergen. The dividing lines between these deposits may be 
considered as the boundaries of the aretic current where it passes under the Gulf 
Stream. 
+The parallelism of temperature is also very marked in the South Pacific, where 
there are no disturbing influences. (See J. J. Wild, Thalassa, (pl. xv.) and Chal- 
lenger Temperatures. ) 
