Previous analyses of subsurface flow down sloping, shallow soil hillsides are further simplified. Surprisingly, whether the starting point is Richards' or Boussinesq's equation, the derived analytical expressions for cumulative outflow are identical, and they agree very well with experimental results from the Coweeta Hydrological Laboratory <Hewlett and Hibbert, 1963>. We show that Boussinesq's model can be improved from a theoretical point of view at a very small cost. We also solved Boussinesq's equation numerically. However, this improved model did not fit the experimental data any better, while the numerical solution only showed a marginal agreement with the observed data. ¿ 1999 American Geophysical Union