ot stop here: the roots must be exterminated; so several dig around
the plant, throwing the earth backward, and after making it bare they
cut and girdle the roots until the plant is killed.

[Illustration: 1. With well-formed teeth. 2. Partially developed. 3.
Entirely obsolete.

MANDIBLES OF THE HARVESTING-ANT.]

Early in March the ants in the jar seemed to have completed their
domicile. At first, several chambers were visible through the glass, and
the galleries leading to them, but gradually the light was all shut out
by placing little particles of earth against the glass, thus depriving
me of the opportunity of watching their movements within the nest. So I
now took the jar to the barrens, and set it by the side of a nest which
was about a mile distant from where most of the ants were obtained. Here
I carefully broke it, and took the thin shell of glass from around the
nest, which did not fall, but stood six inches in height and eighteen
inches in circumference. With a large knife I removed a thin layer of
earth, which revealed three admirable chambers with galleries leading
from one to the other. Immediately below there were five chambers well
filled with ants, and below these other chambers were scattered
irregularly throughout, with only thin partitions between.

At various times I had given the ants moistened sugar on the thick
curved leaves of the live-oak, and several of these had been covered
while the ants were making their excavations. Two of the leaves were
three inches below the surface, and the ants had utilized them by making
the inner curved surface answer for the floor and sides of fine
chambers; and here a large number of ants, both soldiers and workers,
were crowded together. In other chambers I found the larvae, which were
greatly increased in size since I had placed them in the jar; and the
larvae of the carpenter-ant were being reared, as I found some smaller
than any I had introduced belonging to the harvester.

Very soon a great crowd of excited ants came from the hill near which I
had broken the jar, and began to transport the larvae, and also the
mature ants, to their own dominions. There was no fighting: the ants
from the jar submitted to being carried, not offering the least
resistance. A small worker would often take hold of a large soldier,
sometimes pushing, sometimes dragging, her through the sand, and she
would be as quiet as if dead or dying; but if we touch the little worker
she