248 THREE CRUISES OF THE “ BLAKE.” 
movements of our atmosphere, be dependent upon cosmic phe- 
nomena, practically independent of any secondary causes, and 
modified by them within very narrow limits ? 
The difference in salinity of certain oceanic districts is in it- 
self insufficient to explain oceanic circulation ; so that while the 
secondary causes referred to above are undoubtedly active as 
producing more or less extensive local circulation, we seem jus- 
tified in looking upon the differences of temperature of the zones 
of the ocean as the principal cause of the general oceanic cir- 
culation. We may state, in the main, that the density of the 
ocean water is least at the equator, gradually rises towards the 
poles, and attains its maximum at 60° of latitude. For the sake 
of convenience we may call the density of the ocean as one at a 
depth of 500 fathoms, and consider the strata of water above 
and below as having a less and a greater density,’ within very 
narrow limits; thus the watery envelope is not in a state of 
equilibrium. 
The most important disturbing factors of a uniform distribu- 
tion of oceanie temperature are the continental masses which lie 
in the path of the equatorial currents. A comparison of the 
position of the oceanic isotherms of the North and South At- 
lantic shows a striking contrast in their course north and south 
of the equator. A similar comparison between the Atlantic 
and Pacific brings out plainly the contrast in the course of the 
isotherms of two oceans, in which the disturbing effect is due 
in the one to continental masses, and in the other to large 
groups of oceanic islands. 
Perhaps the best example of the unstable equilibrium existing 
between adjoining oceanic areas is furnished by the heaping up 
of the waters driven by the tradewinds into the Gulf of Mexico 
from the Caribbean. The amount of this accumulation has 
actually been measured by officers of the United States Coast 
Survey. It gives an additional force at work to keep up the 
efficiency of the Gulf Stream. The Gulf of Mexico is consid- 
1 Ocean water, at depths exceeding either colder or warmer, it must expand ; 
1,000 fathoms, has a temperature of this it cannot do, on account of the pres- 
nearly 35° F., the temperature of great- sure. 
est density. Should the water become 
