1. The artificial limb. 2. The attachment of this limb to the trunk. The artificial limb itself is divided into two parts: 1. A conical socket. 2. A part which replaces the missing limb and is in fact a terminal functional appliance. Two conditions must be considered, whether or not there remains attached to the trunk a segment of the limb capable of being fitted into the base of the artificial limb, to which it gives support, and to which, in addition, it can communicate movement. Accordingly the artificial limb differs essentially for: 1. Disarticulation of the shoulder and of the hip. 2. Amputation of the arm and of the thigh. In the first case we attach to the trunk an instrument which is entirely passive. In the second we attempt to turn to account the active movements of the stump. These various parts do not lend themselves to a general description applicable at once to the upper and lower limbs. Not only are the modes of attachment and the functional artificial limb quite different, but the bucket does not serve the same purposes. _The position of the scar._--The stump, which fits the bucket exactly, transmits to it two kinds of force: 1. The force of vertical pressure. 2. Lateral force corresponding to the angular movements of the joint above. The lateral force is transmitted by the whole of one surface of the stump to the corresponding lateral surface of the bucket: by the anterior and posterior surfaces only in the case of hinge joints such as the elbow and the knee: by all surfaces in the case of joints with movements of circumduction such as the shoulder and the hip. Vertical pressure exercised upwards or downwards may cause the limb to press upon the bucket at two points: (1) on the summit of the cone, _i.e._ on the extremity of the stump; (2) on the base of the cone, _i.e._ on the bony prominences around the last remaining joint. The adjustment is never sufficiently accurate for the relief due to the fitting of the stump in the bucket to be of much importance. We should take it as a general rule that a scar cannot stand pressure or friction; and that in consequence, when we amputate under favourable conditions, we should arrange to place the scar in such a position that from our knowledge of the suitable prosthetic apparatus these two evils will be avoided. It should be added, however, that after perfect primary union, the narrow and mobile scar is very tolerant, but it must also be remembered--especially as will be seen in the lower limb--that this condition is rarely realised in war surgery. The length of the stump is often estimated by reference to that of the other limb; amputation at the upper, middle, or lower third of the thigh, of the leg, of the arm, or of the forearm. This is convenient, starting from a certain minimum length, but there is an _absolute minimum length_ below which the stump has insufficient leverage and tends moreover to escape from the bucket.