A theoretical model for deformation of a thrust sheet at a ramp which takes into account (1) viscous deformation within the upper plate, (2) plastic deformation within the fault zone, and (3) gravity shows that the upper plate's resistance to deformation can be an important impediment to thrust sheet movement. Using representative parameters for the Pine Mountain block of Kentucky, Virginia, and Tennessee, upper plate bending resistance is shown to be of the same order as fault zone drag. Gravity can be either an important resisting force or a moderate aid in the motion of the thrust sheet depending on how the paleotopography at the time of thrust sheet motion is reconstructed. Two end-member cases are considered. (1) In the 'no erosion' case the surface of the block mimics the shape of the fault zone. In this case, gravity is the largest resistance to movement, greater than either internal deformation or fault zone drag. (2) In the 'stacking' case, erosion is assumed to alter the topographic profile of the block and overlying sheets in such a way as to provide a surface slope in the direction of movement. Gravity in this case aids movement of the Pine Mountain block but cannot provide all the work necessary to move the upper plate at a ramp. A tectonic push is also required. The bending stresses are a significant portion of the total stress field and are responsible for the existence of zones of both potential normal and potential reverse faults in the model. An element of the sheet situated on the lower or upper surface of the block will, in traveling from the bottom to the top of the ramp, experience both normal and reverse faulting regimes. This is perhaps an explanation for the complex association of normal and reverse faults found in some major thrust sheets.