ant-lion immediately begins twisting its head about, and shovelling up the sand with the greatest agility, jerking it about on each side and backwards, but never forwards, as misrepresented in some figures, until the hole is made so much deeper, and such a disturbance caused in the sides of the hole, that the insect is almost sure to be brought down to the bottom, when it is seized by the ant-lion, which immediately endeavours to draw it beneath the sand; and if it be very boisterous, the ant-lion beats it about, holding it firmly by the jaws until it is too weak for further resistance. Hence, as the head of the ant-lion is immersed in the sand, it is evident that the accounts given in popular works of the instinct by which it throws the sand in the direction of the escaping prey are not quite correct. The act of throwing up the sand, when an insect has fallen into the pit and attempts to escape, has evidently for its chief object that of making the pit deeper and more conical, and therefore more difficult of ascent." [Illustration: Ant-Lion's Pitfalls, in an experimenting-box.] It is by the action of the hinder pair of its legs that the ant-lion drags itself backwards, the other four pair being extended trailing after it, and leaving an impression on the surface of the fine sand over which it has passed; and when burrowing its way beneath the surface of the sand, it proceeds by short steps backwards. A portion of sand at each step is thrown on the head, owing to the hump-like form of the back: this is immediately jerked away, the body at the same time advancing another step in its backward and spiral motion. Where it rests, a little hillock of sand is raised by the body of the ant-lion underneath; while its jaws emerge and spread flat on the surface. It now probably commences its pitfall, the mode of excavating which we have given in detail. From the spiral course described by the ant-lion in its backward progress appears to have arisen the idea of its tracing out a circle as the outline of its pitfall--as would an architect or engineer; but whence sprang the often-repeated statement, that the ant-lion loads its head with sand by means of one of its legs, that nearest the centre of the circle, we cannot conjecture. Nor do we know how, as it works entirely buried with the exception of the head, the ant-lion can act when it meets with a stone or other obstacle, as M. Bonnet states he has repeatedly witnessed. He observes that if the stone be small, it can manage to jerk it out in the same manner as the sand; but when it is two or three times larger and heavier than its own body, it must have recourse to other means of removal. The larger stones it usually leaves till the last; and when it has removed all the sand which it intends, it then proceeds to try what it can do with the less manageable obstacles. For this purpose it crawls backwards to the place where a stone may be, and thrusting its tail under it,