This paper aims to show that the distinction between two types of body (organic and inorganic), which is sometimes appealed to in order to clear up the Aristotelian definition of soul, sheds some light on the ontological status of prostheses and (more broadly) on the issue of the "multiple readability" of life-functions. After discussing the strain that the definition of soul imposes upon the basic hylomorphic categories, the paper turns to the idea of a "functionally defined body", whose organs are defined by the power to perform their proper functions. This way of talking about the body suggests that Aristotle is committed to a principle of Functional Definition, which allows him to decide dubious cases of kind-membership solely on the basis of the presence or absence of the relevant powers-i.e., without drawing on any further considera-tions. The paper argues that even if (taken in isolation) that criterion seems unable to avoid the inclusion of artificial items under the same kinds as their biological counterparts, Aristotle's philosophy of nature provides independent reasons for rejecting that unwelcome result of Functional Definition. Given the Aristotelian recognition of strong natural unities, which are "one" not because of their material continuity but because of partaking in the same principle of change ((pUGit;), artificial limbs cannot be equated with their models. Accordingly, they perfom the same task, but in a sui generis way.