rk are found, embracing nearly the whole trunk for some
feet. The outside of these tunnels, which are never quite straight,
but wander irregularly along stem and branch, resembles in texture a
coarse sandpaper; and the color, although this naturally varies with
the soil, is usually a reddish brown. The quantity of earth and mud
plastered over a single tree is often enormous; and when one thinks
that it is not only an isolated specimen here and there that is
frescoed in this way, but often all the trees of a forest, some idea
will be formed of the magnitude of the operations of these insects,
and the extent of their influence upon the soil which they are thus
ceaselessly transporting from underneath the ground.

In traveling through the great forests of the Rocky Mountains or of
the Western States, the broken branches and fallen trunks, strewing
the ground breast-high with all sorts of decaying litter, frequently
make locomotion impossible. To attempt to ride through these Western
forests, with their meshwork of interlocked branches and decaying
trunks, is often out of the question, and one has to dismount and drag
his horse after him as if he were clambering through a wood-yard. But
in an African forest not a fallen branch is seen. One is struck at
first at a certain clean look about the great forests of the interior,
a novel and unaccountable cleanness, as if the forest bed was
carefully swept and dusted daily by unseen elves. And so indeed it is.
Scavengers of a hundred kinds remove decaying animal matter, from the
carcass of a fallen elephant to the broken wing of a gnat; eating it,
or carrying it out of sight and burying it in the deodorizing earth.
And these countless millions of termites perform a similar function
for the vegetable world, making away with all plants and trees, all
stems, twigs, and tissues, the moment the finger of decay strikes the
signal. Constantly in these woods one comes across what appear to be
sticks and branches and bundles of fagots, but when closely examined
they are seen to be mere casts in mud. From these hollow tubes, which
preserve the original form of the branch down to the minutest knot or
fork, the ligneous tissue is often entirely removed, while others are
met with in all stages of demolition. There is the section of an
actual specimen, which is not yet completely destroyed, and from which
the mode of attack may be easily seen. The insects start apparently
from two centres. One c