The objective of this study is to compare the laboratory behavior of natural (Mode II) and induced (Mode I and II) fractures during stress-dependent permeability tests (SDk) to verify under which conditions the hydraulic behavior of induced fractures may be assumed to be representative of that of in-situ natural fractures. Fracture modes are identified by the way the force that enables the crack to propagate is applied and may be of three different types. Mode I fractures (propagated by a tensile stress normal to the plane of the crack) and Mode II fractures (propagated by a shear stress acting parallel to the plane of the crack and perpendicular to the crack front) are of import in the present study. To test under which conditions this assumption may hold, cylindrical specimens of different rock types including plugs with an intact matrix and plugs with a natural fracture propagating through their body were selected. When possible, the plugs were grouped in pairs so as to form homogeneous sets characterized by a coherence in specific matrix properties (porosity, density, permeability), with each pair including a plug containing a natural fracture. Induced Mode I and Mode II fractures were then propagated axially through the intact specimens and each plug set was tested for SDk. All of the tests were run under similar triaxial test conditions.