Fracture-permeability relations of shale are needed in hydraulic fracturing, CO2 sequestration, and nuclear waste disposal. In this study, we use triaxial coreflood experiments in combination with x-ray tomography to study fracture generation and permeability during injection of water. The experiments were conducted on Utica shale at temperatures from 25-50 °C and confining pressures from 3.4 to 13.8 MPa. Fractures were generated using direct shear methods with semi-circular anvils. Permeability was monitored following fracture events as a function of continuing deformation and hydrostatic pressure. Fractures propagating parallel to bedding were an order of magnitude more transmissive than fractures propagating perpendicular to layers. Maximum whole rock permeability ranged from 70-900 mD. Shale strength when fractures propagated parallel to bedding was a linear function of the angle between the fracture and the direct-shear plane and was Strength (MPa) = 0.8034 x Angle (°) + 26.259. This behavior appeared to be related to the amount of shear-plane deformation accommodated by bedding planes. Permeability was not a clear function of this angle but peaked sharply for fractures that crossed bedding planes at angle near 45°. Significant deformation (>1%The flow of fluids through fractured shale is important in hydraulic fracturing, CO2 sequestration, and nuclear waste disposal. In hydraulic fracturing, the goal is to produce hydrocarbon by generating (or reactivating) pervasive, permeable fracture systems. In CO2 sequestration, risk assessment studies must consider the possibility that pressurization of the storage reservoir could induce fractures in caprock (which may be shale) where the consequences are governed by the permeability of the damage zone. In nuclear waste disposal located in shale formations, risk assessment studies must consider the permeability of potential flow paths generated during excavation of the repository, within pre-existing fracture networks, and within any subsequent fracture events induced by seismicity.