If a Lobster be caught by one of its claws or by a leg, it very readily parts with the limb in its struggles to escape; and if one of the limbs be crushed or otherwise injured, it is often cast off by the animal. The separation always takes place at the same point, near the base of the limb, and is not simply due to the limb breaking at its weakest part. It is a reflex act, brought about by a spasmodic contraction of some of the leg muscles. At the place of separation, corresponding to the junction of the second and third segments of the limb, which, as already mentioned, are soldered together, the internal cavity is crossed by a transverse partition, having only a small aperture in the centre through which the nerves and bloodvessels pass. When the limb is cast off, this small opening quickly becomes closed by a clot of blood, and further bleeding is stopped. If, as sometimes happens, a limb which has been seriously injured is not cast off, the animal not infrequently bleeds to death. This power of self-mutilation or _autotomy_, as it is called, is frequently used by Crustacea as a means of escaping from their enemies, and is closely connected with the power of _regeneration_ of lost appendages. Beneath the scar which forms on the stump of a separated limb a sort of bud grows, and gradually assumes the form of the lost segments. At the next moult this straightens out, and, increasing in size at succeeding moults, it ultimately provides, in normal cases, a new member similar in every detail to that which had been lost. Occasionally it happens, under circumstances not yet altogether understood, that the process of regeneration may, so to speak, go wrong, and in this way various malformations and abnormalities result. For instance, it has been found that, if the larger, crushing claw of a very young Lobster be removed by operation or by accident, the limb which grows in its place may assume the form of the smaller, toothed claw. Further, in some other Crustacea (but not in the Lobster, except in the very youngest stages), it is found in such cases that, after removal of the large claw, the claw of the other side assumes at the next moult the form of a crushing claw, so that there is a "reversal of asymmetry." A still more remarkable change sometimes occurs when one of the eye-stalks is injured. If only the tip of the eye-stalk be cut off, so that the nerve-ganglion which lies in the basal part of the stalk remains uninjured, it will be found that a normal eye is in course of time regenerated. If, however, the whole eye-stalk be amputated, and with it the optic ganglion, there grows in its place, not a new eye-stalk, but a segmented appendage similar to one of the flagella of the antennules. This fact is considered by some zoologists to indicate that the eye-stalks are, like the antennules, true appendages, homologous with the mouth parts and limbs, but this is a much-disputed question into which we cannot enter further here.