AbstractAn examination has been made of the results obtained by the Marshall test to determine the value of the test for the design of dense bituminous mixtures for road surfacings and for comparing their resistance to deformation.It is shown that the optimum binder contents obtained from the density and stability curves are similar to those specified in B.S.594 for the driest schedules having the highest resistance to deformation and that the test would exclude those materials in the other schedules. In the absence of information on the relative deformation on roads of surfacings made with different types of bituminous material, simulative laboratory tests have been used to place compositions in order of resistance to deformation. Comparison of the results of the Marshall test with the Road Research Laboratory wheel‐tracking tests shows that neither the stability nor the flow value provides a valid measure of the resistance to deformation of dense bituminous materials in common use. However, for rolled asphalts the ratio stability/flow correlates moderately well with the logarithm of the rate of tracking measured in the wheel‐tracking machine; the correlation for dense tar surfacing compositions is poor. Comparison with the Hveem Stabilometer suggests that setting an upper limit on flow value is approximately equivalent to setting a lower limit on the internal friction. As with many mechanical tests, high Marshall stability values are promoted by high stone content, high viscosity of binder, rough‐textured aggregate, not too low a filler content and good compaction.