o the surface of the water, can never warm it, except so
far as it is conducted downward by some other medium than the water
itself. Count Rumford confined cakes of ice in the bottom of glass jars,
and, covering it with one thickness of paper, poured boiling-hot water
on the top of it, and there it remained for hours without melting the
ice. The paper was placed over the ice, so that the hot water could not
be poured on it, which would have thawed it at once. Every man who has
poured hot water into a frozen pump, hoping to thaw out the ice by this
means, has arrived at the fact, if not at the theory, that ice will not
melt by hot water on the top of it. If, however, a piece of lead pipe be
placed in the pump, resting on the ice, and hot water be poured through
it, the ice will melt at once. In the first instance, the hot water in
contact with the ice becomes cold; and there it remains, because cold
water is heavier than warm, and there it will remain, though the top be
boiling. But when hot water is poured through the pipe, the downward
current drives away the cold water, and brings heated particles in
succession to the ice.

Heat is propagated in water, then, only by circulation; that is, by the
upward movement of the heated particles, and the downward movement of
the colder ones to take their place. Anything which obstructs
circulation, prevents the passage of heat. Chocolate retains heat longer
than tea, because it is thicker, and the hot particles cannot so readily
rise to be cooled at the surface. Count Rumford illustrated this fact
satisfactorily, by putting eider-down into water, which was found to
obstruct the circulation, and to prevent the rapid heating and cooling
of it. The same is true of all viscous substances, as starch and glue;
and so of oil. They retain heat much longer than water or spirits.

In a soil saturated with water, or even in water thickened with mud,
there could then be but little circulation of the particles, even were
the heat applied at the bottom instead of the top. Probably the soil,
though saturated with water, does, to some extent, transmit heat from
one particle of earth to another, but it must be but very slowly.

In the chapter upon Temperature as affected by Drainage, farther
illustrations of this point may be found.

AERATION BY DRAINS.

Among the advantages of thorough-drainage, is reckoned by all, the
circulation of air through the soil. No drop of water can run from the