286 Proceedings of the Royal Society 
of density, we see that the influence of atmospheric conditions does 
not cease when we leave the surface. In the trade-wind districts 
we find the excessive concentration extending to greater or less 
depths, according to the ocean which we are considering, and, 
as might have been expected, this effect is much more sensible in 
the Atlantic than in the Pacific. In the Atlantic, again, the 
amount of concentration is much greater in the northern than in 
the southern portion, while in the Pacific the reverse is the case. 
The great amount of concentration in the North Atlantic is doubt- 
less due chiefly to the form of its basin. The south-east trade 
which blows with a preponderating force drives constantly fresh 
supplies of water in to the northern basin, where a large portion of 
it is kept constantly circulating by the prevailing winds in a region 
where the atmosphere is ever greedy of moisture. This circulation, 
assisted by the yearly oscillations of temperature, bring about a 
constant concentration of the deep waters, and at the same time, 
as I have pointed out, a heating of the deeper waters by convec- 
tion. The effect of this concentration of the water is to render it 
heavier than an equal bulk of water further to the south, and con- 
sequently it forces its way southwards along the bottom. This 
under outflow from the North Atlantic was conjectured by Cap- 
tain Tizard from a consideration of the temperature of the deep 
water of the western basin of the South Atlantic, and a glance 
at the vertical section of the Atlantic is sufficient to show the 
great likelihood of its existence. From a consideration of this 
section it would appear probable that at some position about 40° 
north latitude a tolerably uniform density of from 1*0265 to 1*0270 
will be found, and, indeed, in the neighbourhood of the Azores 
the following bottom specific gravities were observed : — 
Depth in fathoms, . . 1675 1000 900 750 
Specific gravity at 15*56, . 1*02670 1*02693 1*02691 1*02679 
Although we find that in all tropical and temperate regions the 
temperature of the water is highest at the surface, and decreases 
as the depth increases, we do not by any means invariably find the 
highest specific gravity at the surface, nor does it decrease regu- 
larly with increasing depth. The water is usually least salt 
at between 500 and 1000 fathoms from the surface. The maximum 
specific gravity at any locality is .frequently observed at depths of 
