At present, it is generally believed that crystal size has no direct influence on strain rate in the ice sheets and that the fraction of strain rate enhancement there which is not ascribable to c axis fabric is due to impurity content. Here we challenge this view because it is not consistent with recent results from analyses of deformation at Meserve Glacier and instead ascribe residual enhancement in the ice sheets to variations in crystal size. We resurrect the idea that variations of crystal size can be an important part of the total shear enhancement in the ice sheets, though agree with Paterson that this effect is generally dominated by variations of crystal fabric. We propose that the enhanced shear strain rate of ice age ice in southern Greenland, as inferred from tilt of the Dye 3 borehole, can be explained as a result of combined fabric variations and crystal size variations, with these two ice properties accounting for roughly 70% and 30% of the average enhancement, respectively. Permitting a grain size dependence of ice viscosity also resolves the quandary concerning closure and tilt of the Agassiz Ice Cap borehole. ¿ 2000 American Geophysical Union