This paper distinguishes the base domain of an economic theory (in which predictions are relatively unambiguous) from, respectively, the domains of intended application and of legitimate testing; it argues that the domain of legitimate testing is not generally restricted to that of intended...