High-speed flow of light ions along magnetic field lines is not simply a consequence of those field lines having an open topology. Here we point out that the magnitude of the H+ flow velocity in the polar wind is not predicted by theoretical models but rather it is imposed as a boundary condition. Thermal diffusion, the electron temperature profile, and a diverging magnetic field can be shown to have important effects on the H+ flow velocity which need not be supersonic even when the pressure at large distances from the earth is low.