Many details of the coupling between mechanical deformation and fluid remain unclear. Although a number of mechanisms have been proposed by which development of pore pressure excess leads to mechanical weakness and instability, the model predictions are sensitively dependent on the effect of stress on fluid transport properties. Permeability is a second rank tensor, and significant anisotropy will develop in response to a deviatoric stress field. Little experimental data exist on this very important aspect. Under hydrothermal conditions, diffusive mass transfer and solution-precipitation process may also play important roles in controlling the coupling between fluid flow and deformation. To extrapolate experimental data to crustal dynamics, one needs to address the difficult question of (spatial and temporal) scaling. In the context of rock friction and earthquake mechanics, the major difficulties are related to the characteristics slip distances and velocity dependence. Major advances have to be made in the fundamental understanding of the micro mechanisms of frictional sliding before one can with some confidence scale up the characteristic distances which appear in friction constitutive equations. The velocities attained in most laboratory measurements on frictional sliding are still relatively slow compared with the coseismic slip velocities. Recent theoretical analyses suggest that the dynamics of faulting is very sensitive to frictional behavior at the high-velocity end, and therefore it is desirable to have systematic measurements over a broad range of slip velocities.