It is clear that this complete liberty of action has its disadvantages; the elbow joint can never be made to assume a fixed position against any passive resistance; moreover, in actual practice the useful range of either of these movements is small. For this reason attempts have been made to devise methods by which they can be limited in the various worker's arms which have been designed since the beginning of the war. In all these arms the mechanism of the elbow joint is derived from that of the ploughman's arm of Gripouilleau. Unfortunately none of these mechanisms in which a pressure screw is used for fixation possess any strength. At first sight, in a new appliance they appear attractive and work well, but it is well known to all mechanics that the thread of a screw which is in constant use quickly wears and then it is impossible to tighten it. At the extremity of an artificial arm, whether it be an arm of natural shape or a worker's arm simplified to the form of a jointed rod, any of the appliances already described for forearm amputations can be screwed on as required. It is by the use of these appliances that Gripouilleau's old ploughman's arm, which ended in an interchangeable hook and ring, has been improved. Apart from their actual economic value, results have been obtained by use of these terminal appliances, in many different skilled trades, which are of the greatest possible interest. For reasons that we have indicated in describing the attachment of the elbow, the various attempts that have been made to give to the wrist a mobility that is under control have not so far led to the invention of an appliance that is both strong and durable. For this reason we consider that until something new is designed it is better to make the terminal appliance a fixed one. _CHAPTER IX_ ARTIFICIAL LIMBS FOR DISARTICULATION THROUGH THE SHOULDER JOINT AND AMPUTATION THROUGH THE DELTOID MUSCLE So far as function is concerned these operations are identical; a short arm stump is incapable of transmitting movements to the socket of the artificial limb. That is to say, our appliance will be a purely passive one, and at the present time it is useless to attempt to make any sort of worker's arm. We must aim simply at supplying a limb which imitates the external shape of the arm, with an elbow joint which can be locked with a ratchet. At most it is possible by a movement of the opposite shoulder to work an automatic thumb by means of a cord, as a rule, however, a simple spring thumb is preferred. Nothing need be added to what has been said in the previous chapter about the elbow joint or the hand. In the fitting of an attachment over the shoulder the amputation through the deltoid presents an actual advantage. For in this case the shape of the point of the shoulder is preserved, and the attachment carried out as described on pages 87 and 130 fixes the appliance very securely.