300 THREE CRUISES OF THE “ BLAKE.” 
North Atlantic, in the heart of the region of the tradewinds, 
and in the South Atlantic in the corresponding regions to the 
east of the coast of Brazil. To determine from its density the 
salinity of the water, the water specimens were reduced to its 
value at a standard temperature, from the tables prepared by 
Professor Hubbard of Washington, and slightly corrected by 
the physicists of the “ Challenger.” 
The salinity is affected by the fixing of the salts held in so- 
lution, which, like lime and silica, are precipitated, as it were, 
by organic substances. 
In the regions of great rainfall near the equator, the surface 
water is frequently quite fresh ; and in the temperate zone there 
is nearly an equilibrium between the rain and evaporation. To 
these subdivisions of concentration and precipitation Thomson 
assigned an important part in the formation of oceanie currents, 
for these result from the constant interchange of vapor and 
water between the air and the sea. 
It is interesting to compare the density of the Caribbean 
and that of the Gulf of Mexico (below 1 0270), both enclosed 
seas, but supplied with water from a region of low specific 
gravity (between 1.0260 and 1.0270), with the high density of 
seas, like the Red Sea and the Mediterranean (above 1.0280), 
where evaporation goes on with immense activity, although they 
are both in communication with oceanic basins. In connection 
with this we might also compare the density of Hudson's Day 
with that of the Baltie, and that of the Great Salt Lake with 
that of the Dead Sea and the Caspian, representing climatic 
conditions totally unlike. 
As has been well said by Moseley, we live in the depths of 
the atmosphere, just as deep-sea creatures live in those of the 
sea; but our range, measured by the difference of temperature, 
is far wider than that to which deep-sea animals, even when 
subjected to the greatest extremes, are exposed. Terrestrial ani- 
mals can withstand differences of heat and cold far better than 
marine animals. The difference of temperature between an 
arctic winter and a tropical summer, for instance, is from three 
to four times as large as the widest range of temperature to 
which any marine animal can be subjected. The lowest tem- 
