The object of leaving the southern side of the square encampment open is as follows. While some of the black workers are engaged on duty as guard, another and larger division are engaged on duty as purveyors. These enter and leave the quadrangle by its open or southern side along the dotted line marked _a_ to the central point _c_. The incoming line is composed of individuals each bearing a burden of fragments of flowers or aromatic leaves. These are all deposited in the centre of the quadrangle _c_. Along the other diagonal _e_ there is a no less incessantly moving double line of yellow workers (I.), whose office it is to convey the supplies deposited by the black workers at _c_ to _b_, which is the gateway of the fortress. It is remarkable that no black ant is ever seen upon the line _e_, and no yellow one upon the line _a_; each keeps his own separate station, and follows his own particular duty with a steadfastness and apparent adherence to discipline that are most astonishing. The hole at _d_ seems to be a ventilating shaft; it is never used as a gateway. [Illustration: Fig. 7.] Section of the nest reveals, besides galleries, a small chamber about three feet below the surface, across which is spread, like a spider's web, a network of squares spun by the insects, the squares being about 1/4 inch across, and the ends of the whole net being fastened to the earthen walls of the chamber. In each one of the squares, supported by the web, sits one of the honey-making ants (II.). Here these honey makers live in perpetual confinement, and receive a constant supply of flowers, pollen, &c., which is continually being brought them by (I.), and which, by a process analogous to that performed by the bee, they convert into honey. Such is an epitome of the only account that the world has yet received of the habits and economy of this wonderful insect, whose instincts of military organization seem to be not less wonderful than those of the Ecitons, though in this case they are developed with reference to defence, and not to aggression. It is especially noteworthy that the black and yellow workers are believed to belong to 'two separate genera;' for if this is the case, it is the only one I can recall of two distinct species co-operating for a common end; for even the nearest parallel which we find supplied in other species of ants maintaining aphides, is not quite the same thing, seeing that the aphides are merely passive agents, like Class II., of the honey-making ant, and not actively co-operating members of the community, like Class I. _Ecitons._--We have next to consider the habits of the wonderful 'foraging,' or, as it might be more appropriately called, the military ant of the Amazon. These insects, which belong to several species of the same genus, have been carefully watched by Belt, Bates, and other naturalists. The following facts must therefore be accepted as fully established.