It is pointed out that the boundary layer plasma must somehow leave the magnetospheric system. We propose that the boundary layers on the dawn and dusk flanks are continued into the far magnetotail, still on closed magnetic field lines. Within the combined layer there would be a dusk to dawn electric field for antisunward convection (as is the case for the boundary layers on the dawn and dusk flanks nearer to the earth). It is suggested that the boundary layer flow is so massive that the flow itself can generate the right electric field for continued flow. Most of the plasma in the mantle over the polar caps would also be convected out in a similar manner, rather feeding the plasma sheet. There would be a stagnation point inside the magnetotail, with any boundary or mantle plasma diffusing earthward of this point becoming the plasma sheet. It is deduced that no steady state solution is possible.