In this paper we investigate the possibility that the recently observed plasma mantle may be the most likely source of the plasma sheet. We do not discuss the origin question of the plasma mantle itself, although we present two rough models which we use mainly for the description of particle transport in the tail lobes and for a semiquantitative calculation of plasma mantle parameters (density, bulk velocity, and temperature distribution) at the plasma sheet boundary. Conservative estimates of the particle supply show that the source is adequate to replensish known losses from the plasma sheet. In addition, it is argued that at least the proton energies expected in the plasma sheet from such a model are in agreement with observations.