in such
a science as Meteorology. Hence Mr. Milne Home's proposal for the
plantation of Malta seemed to offer an exceptional opportunity for
progress. Many of the conditions are favourable to the simplicity of the
result; and it seemed natural that, if a searching and systematic series
of observations were to be immediately set afoot, and continued during
the course of the plantation and the growth of the wood, some light
would be thrown on the still doubtful question of the climatic influence
of forests.

Mr. Milne Home expects, as I gather, a threefold result:--1st, an
increased and better regulated supply of available water; 2nd, an
increased rainfall; and, 3rd, a more equable climate, with more
temperate summer heat and winter cold.[46] As to the first of these
expectations, I suppose there can be no doubt that it is justified by
facts; but it may not be unnecessary to guard against any confusion of
the first with the second. Not only does the presence of growing timber
increase and regulate the supply of running and spring water
independently of any change in the amount of rainfall, but as
Boussingault found at Marmato,[47] denudation of forest is sufficient to
decrease that supply, even when the rainfall has increased instead of
diminished in amount. The second and third effects stand apart,
therefore, from any question as to the utility of Mr. Milne Home's
important proposal; they are both, perhaps, worthy of discussion at the
present time, but I wish to confine myself in the present paper to the
examination of the third alone.

A wood, then, may be regarded either as a _superficies_ or as a _solid_;
that is, either as a part of the earth's surface slightly elevated above
the rest, or as a diffused and heterogeneous body displacing a certain
portion of free and mobile atmosphere. It is primarily in the first
character that it attracts our attention, as a radiating and absorbing
surface, exposed to the sun and the currents of the air; such that, if
we imagine a plateau of meadow-land or bare earth raised to the mean
level of the forest's exposed leaf-surface, we shall have an agent
entirely similar in kind, although perhaps widely differing in the
amount of action. Now, by comparing a tract of wood with such a plateau
as we have just supposed, we shall arrive at a clear idea of the
specialities of the former. In the first place, then, the mass of
foliage may be expected to increase the radiating power of each t