The purpose of this note is to prove the following "boundedness" stated in [Ol]. Let X and Y be separated Deligne-Mumford stacks of finite presentation over an algebraic space S and define Hom S (X , Y) as in [Ol, 1.1]. Assume that X is flat and proper over S, and that locally in the fppf topology on S, there exists a finite flat surjection Z → X from an algebraic space Z. Let Y → W be a quasi-finite proper surjection over S to a separated algebraic space W over S of finite presentation. By [Ol, 1.1] we then have Deligne-Mumford stacks Hom S (X , Y) and Hom S (X , W ).is of finite type.Remark 1.2. For a simple example to illustrate this theorem, consider the case when X = X is a smooth proper scheme over S, Y = BG for some finite group G, and W = S. Then the stack Hom S (X , Y) classifies G-torsors on X , and Theorem 1.1 essentially amounts to [SGA4, XVI.2.2] (this special case is in fact used in the proof; see section 3).Note that in the case when S is the spectrum of a field, then X has a coarse moduli space π : X → X and the formation of this moduli space commutes with arbitrary base change (see [Ol, 2.11] for a discussion of this). In this case the right side of 1.1.1 is canonically isomorphic to Hom S (X, W ) by the universal property of coarse moduli spaces.Our interest in this theorem comes from the theory of moduli spaces for twisted stable maps. As explained in [Ol2] the above theorem combined with the existence of a universal twisted curve (constructed in loc. cit.) yields a very quick proof of boundedness for the Abramovich-Vistoli moduli space of twisted stable maps [A-V]. Considering the very general nature of 1.1 we also hope that it will have interesting applications elsewhere in proving boundedness for moduli spaces (already some other applications have been found in recent work of Lieblich and Kovács [LK]).The proof of 1.1 is a rather complicated devissage to the result [SGA4, XVI.2.2]. In section 2 we explain a construction (well-known to experts) called "rigidification" which enables one to"kill off" generic stabilizers of stacks. In section 3 we study a key special case of 1.1, from which the general case will be deduced. In section 4 we collect together various rather general results which will be used for the devissage. In