The validity of the cubic law for flow along closed rock joint has been debated in the literature with mixed conclusions drawn by different authors. An effort was undertaken to reevaluate published experiments in a consistent fashion by summarizing available data in terms of hydraulic aperture vs. joint closure. It was determined that those relations were generally linear at low effective normal stresses (<25 MPa) and in some studies departed from a straight line as an irreducible flow rate was approached at higher stresses. A residual hydraulic aperture was used to account for this phenomenon. The slope (f) of the line, represents a geometric correction factor and is controlled by the surface roughness of the joints as inferred from the maximum joint closure. In all studies the flow rate is well predicted by a cubic function of the aperture.