A controversy of relevance to the study of biological form involves the concept of adaptation. This controversy is illustrated by the structure and function of the human hand. A review of the principal definitions of adaptation points to two main problems: (1) they are qualitative and make reference to the whole structure (or substructural feature) and (2) they are based on the idea of natural selection as a moulding factor. The first problem would be solved by a definition that encompasses quantitative measures of the effects of selection, drawing on new advances in the comparative method. The second problem is deeper and presents greater conceptual difficulties. I will argue that the idea of natural selection as a moulding factor depends on the notion of a genetic program for development. But regarding the hand, experimental evidence on limb development challenges the idea of a genetic program for skeletal pattern formation, undermining a simple application of standard adaptationist concepts. These considerations lead to a revised definition of adaptation and interpretation of the evolutionary determinants of the hand's form.