I have tried to describe the functional attainments of amputees using conventional prostheses. I have outlined the work we have been doing on pneumatic arms for infants and children. I have discussed some of the practical problems in applying pneumatic and myo-electric systems to adults. The inescapable conclusion that one must reach is that two basic clinical requirements must be met before really significant advance is made in this field. Firstly, we must find a method of providing sensory information from the prosthesis. Secondly, we must find a way of utilising more control sites. So far we have only been able to make use of direct movements or myo-electric signals. In the high level bilateral subjects more controls are required than appear to be available. Capener has suggested, in more than one discussion that I can recall, the possibility of using the voice and I have no doubt that a subject could be trained to reserve certain frequencies for prosthetic controls. It may be that we will have to think along some such lines as this.

1. Congenital defects of the extremities are described. Although the detailed anatomy is infinitely variable, a broad classification in relation to prosthetic management has been suggested.

2. Most patients with these deformities can be fitted with a prosthesis without major surgical intervention. With this they will have at least as good function as they would have after amputation. A plea is made for a conservative attitude in this respect. It is suggested that recourse to amputation should be confined to cases in which prosthetic equipment falls short of functional and cosmetic requirements, and that, when possible, it should be deferred until the child is old enough to share in the decision.

3. The prostheses applicable to the various types of deformity are briefly described.

4. The application of similar techniques to cases of acquired shortening is mentioned.

5. The incorporation of certain features of artificial arms in flail arm splints is discussed.