Although channeled lava flows are common in basaltic volcanism, relationships between channel morphology, eruption and emplacement parameters, and lava properties are not well understood. Several models have commonly been used to constrain these relations, but they have not been well tested on natural or simulated lava flows over a wide range of parameter space. Here, we test the accuracy and assumptions of a moderately simple analytic rectangular channel solution by comparing the behavior of well-controlled laboratory polyethylene glycol (PEG) channeled flows to the analytic solution for isothermal, steady Newtonian flow in a rectangular channel with constant dimensions. This analytic solution agrees well with laboratory measurements. Volumetric effusion rates (Q; m3 s-1) calculated from the analytical model using measured PEG flows as input yield ratios of Qcalculated/Qpumped of ~0.2 to 3.6, and flow rates calculated from a best fit surface velocity profile to measured velocities give more accurate ratios of ~0.8 to 1.2. We find a very weak dependence of solution accuracy on slope, attributable to flow front effects within the laboratory flows. We subsequently apply the solution to several subaerial and submarine terrestrial flows as well as extraterrestrial channeled flows over a wide range of flow parameters. Viscosity ranges and flow rates obtained using measured channel dimensions and assumed lava properties are plausible. Interestingly, the resulting extraterrestrial estimates of viscosities and flow rates tend to fall closer to known terrestrial measurements and estimates of channel flow than to previous planetary estimates. We therefore suggest that the analytic Newtonian rectangular channel flow model is a more appropriate physical model for many channeled terrestrial and planetary flows than the Newtonian infinite sheet flow and approximation to Bingham channel flow widely used previously. ¿ 2001 American Geophysical Union