as an undercurrent to the equator, where it again rises and moves
toward the poles.

Now the air of the equator, moving with the earth's rotary motion,
has a greater velocity than the earth itself at high northern or
southern latitudes, and consequently appears to gain an eastward
motion in its progress toward the poles. Without friction, this
relative eastward motion would increase as the air moves toward the
poles, and diminish at the same rate as the air returns, till at the
equator the velocity of the earth and of the air would again be equal;
but friction reduces the motion of the returning air to that of the
earth, at or near the calms of the tropics; so that the air, passing
the tropics, gains a relative westward motion in its further
progress through the torrid zone. The southwestward motion thus
produced between the tropic of Cancer and the equator is the
well-known trade-wind.

Now, according to this theory, the prevailing winds of our temperate
latitudes ought to have a southeastward motion as far as the calms
of Cancer or "the horse latitudes." Moreover, instead of these calms,
there should still be a southward motion. But observation has shown,
that though the prevailing lower winds of our latitude move eastward,
still their motion is toward the north rather than the south; so
that they appear to contradict the theory by which the trade-winds
are explained.

To account for these anomalies, Lieut. Maury has invented a very
ingenious hypothesis, which is published in his "Physical Geography
of the Sea." He supposes that the air, which passes from the equator
toward the poles in the upper regions of the atmosphere, is brought