illation _heat_ plays quite as
necessary a part as _cold_, and before Bishop Heber could speak of
"Greenland's icy mountains," the equatorial ocean had to be warmed by
the sun. We shall have more to say upon this question afterwards.

The heating of the tropical air by the sun is _indirect_. The solar
beams have scarcely any power to heat the air through which they pass;
but they heat the land and ocean, and these communicate their heat to
the air in contact with them. The air and vapor start upwards charged
with the heat thus communicated.

Tropical Rains.

But long before the air and vapor from the equator reach the poles,
precipitation occurs. Wherever a humid warm wind mixes with a cold dry
one, rain falls. Indeed the heaviest rains occur at those places where
the sun is vertically overhead. We must enquire a little more closely
into their origin.

Fill a bladder about two-thirds full of air at the sea level, and take
it to the summit of Mount Blanc. As you ascend, the bladder becomes
more and more distended; at the top of the mountain it is fully
distended, and has evidently to bear a pressure from within. Returning
to the sea level you find that the tightness disappears, the bladder
finally appearing as flaccid as at first.

The reason is plain. At the sea level the air within the bladder has
to bear the pressure of the whole atmosphere, being thereby squeezed
into a comparatively small volume. In ascending the mountain, you
leave more and more of the atmosphere behind; the pressure becomes
less and less, and by its expansive force the air within the bladder
swells as the outside pressure is diminished. At the top of the
mountain the expansion is quite sufficient to render the bladder
tight, the pressure within being then actually greater than the
pressure without. By means of an air-pump we can show the expansion of
a balloon partly filled with air, when the external pressure has been
in part removed.

But why do I dwell upon this? Simply to make plain to you that the
_unconfined air_, heated at the earth's surface, and ascending by its
lightness, must expand more and more the higher it rises in the
atmosphere.

And now I have to introduce to you a new fact, towards the statement
of which I have been working for some time. It is this: _The ascending
air is chilled by its expansion_. Indeed this chilling is one source
of the coldness of the higher atmospheric regions. And now fix your
eye upon those mi