Most discussions in epistemology assume that believing that p is a necessary condition for knowing that p. In this paper, I will present some considerations that put this view into doubt. The candidate cases for knowledge without belief are the kind of cases that are usually used to argue for the so-called 'extended mind' thesis.