magnetic influence of the earth, and producing a state of the trade
not unlike that which induces the formation of cumuli, and which attracts
the surface atmosphere from the Sound in over the land: for the _tendency
to cumulus condensation precedes the breeze_, and the breeze is often
wanting in the hottest days where no such tendency to the formation of
cumuli exists. The same is true of sea breezes elsewhere. They do not blow
in upon some of the hottest surfaces. Where they do exist, they do not
always blow, but are wanting during the hottest days; and careful
observers have identified their appearance with the formation of cumuli,
or other condensation, upon the hills inland. They are not, therefore, the
result of ascending currents of heated air.

The received theory regarding sea and land breezes is a mistaken one in
another respect. There is no such thing as a land wind corresponding in
force to, and the opposite of, the sea breeze--occasioned by the
comparative warmth of the ocean. These breezes blow mainly within the
trade-wind region. Of course they are either beneath the belt of rains or
the adjoining trades. They are said to be, and doubtless are, most active
and strongly marked on lines of coast, particularly the Malabar coast, and
where the trade-winds are drawing usually from them. In the day-time, when
the action of the sun increases the action of the magnetic currents upon
the land, or there are _elevations inland_ which approach the
counter-trade, and especially if it is elevated near the coast, as the
Malabar coast is by the Ghauts, the attraction of this atmosphere over it
_reverses the trade_, or inclines it in upon the land, and it blows in
obliquely or perpendicularly, according to the relative trending of the
coast and the direction of the surface-trade. Thus, where islands are
situated within the range of the trades, the latter will be _reversed_
during the day on the _leeward_ side, but continue to blow as land winds
during the night. So they are sometimes deflected in upon the land on the
sides, during the day, and in like manner return to their course in the
night. So, too, the north-east trades of Northern Africa, are occasionally
(though feebly where the coast is flat) deflected during the day-time, and
blow in as N. W. winds. Upon the southern coast of Africa the S. E. trade
is deflected, and blows in as a S. W. wind. Upon the south-western coast
of North America, the N. E. trades are deflec