The role of relative sea level in controlling aggradation and degradation in alluvial rivers has been a central subject in geology. The conventional wisdom suggests that (1) rivers aggrade in response to sea level rise and degrade in response to sea level fall, and (2) as long as sea level remains stationary, rivers finally attain the equilibrium state ("grade"), after which no net deposition and no net erosion take place despite a steady sediment supply on a geological time scale. Using a moving-boundary morphodynamic model, Swenson & Muto (this volume) analyzed the basic controls on alluvial-river response to falling sea level and the conditions necessary for alluvial grade. The argument here is an experimental corroboration of model predictions. Different runs of experiments to build miniature fluviodeltaic systems with specified sea level fall were conducted with flume-tank facilities. Upstream sediment and water discharges were kept constant during the entire period of each run. Our experimental observations are in agreement with the model predictions that in the high-gradient, uniformly-sloping shelf setting, (1) sea level fall decelerating in a particular pattern allows rivers to be graded, and (2) with steady sea level fall, the river inevitably is aggradational in the early stage and then, without attainment of grade, becomes degradational.A geometrical model inspired from the experimental results predicts that where a fluviodeltaic system progrades on a drowned antecedent alluvial plain, rivers intrinsically tend to approach and finally attain grade at any constant rate of sea level fall if the length of the drowned antecedent reach is sufficient.