Abstract
Various experiments, especially recent experiments by Keller, clearly demonstrate that the thickness of a helium film is independent of the flow rate Q. This is consistent with our current picture of the hydrodynamics of a helium film if there are temperature gradients in the film which produce a thermodynamic potential ϕT=SdT that cancels the ρSvS22ρ term in the hydrodynamic potential μ. If such temperature gradients exist, they would produce unacceptably large evaporation rates unless the saturated vapor pressure pvap is a function of μ rather than T. If pvap=pvap(μ), then there should be no thinning of the film with changes in Q, and the theory is in complete agreement with Keller's results.