72 
THE PHYSICAL GEOGRAPHY OF THE SEA. 
Such is the process which is actually going on in nature ; and if 
we take the motions of these two particles as the type of the mo- 
tion of all, we shall have an illustration of the great currents in 
the air, the equator being near one of the nodes, and there being 
two systems of currents, an upper and an under, between it and 
each pole. 
Halley, in his theory of the trade-winds, pointed out the key to 
the explanation so far, of the atmospherical circulation ; but, were 
the explanation to rest here, a northeast trade-wind extending 
from the pole to the equator would satisfy it ; and were this so, 
we should have, on the surface, no winds but the northeast trade- 
winds on this side, and none but southeast trade-winds on the 
other side, of the equator., 
99. Let us return nowt© our northern particle (Plate I., p. 70), 
and follow it in a round from the north pole across the equator to 
the south pole, and back again. Setting off from the polar re- 
gions, this particle of air, for some reason which does not appear 
to have been very satisfactorily explained by philosophers, instead 
of trayeling 98) on the surface all the way from the pole to the 
equator, travels in the upper regions of the atmosphere until it 
gets hear the parallel of 30°. Here it meets, also in the clouds, 
the hypothetical particle that is coming from the south, and going 
north to take its place. 
100. About this parallel of 30° north, then, these two particles 
press against each other with the whole amount of their motive 
power, and produce a calm and an accumulation of atmosphere : 
this accumulation is sufficient to balance thq pressure of the two 
winds from the north and south. 
101. From under this bank of calms, which seamen call the 
"horse latitudes" (I have called them the calms of Cancer), two 
surface currents of wind are ejected; one toward the equator, as 
the northeast trades, the other toward the pole, as the southwest 
passage-winds. 
These winds come out at the lower surface of the calm region, 
and consequently the place of the air borne away in this manner 
must be suppHed, we may infer, by downward currents from the 
superincumbent air of the calm region. Like the case of a vessel 
of water which has two streams from opposite directions running 
in at the top, and two of equal capacity discharging in opposite 
