In this report we present evidence from the low-altitude DMSP F7 satellite that the poleward edge of auroral luminosity in the nightside auroral zone does not necessarily correspond to the boundary between plasma-filled flux tubes and flux tubes devoid of plasma. Assuming that the low-altitude boundary corresponds to the boundary between the lobe and the plasma sheet, this implies that the boundary between open and closed field lines may lie poleward of the most poleward auroral luminosity. Thus the assumption that the poleward boundary of auroral luminosity is a good indicator of the open-closed boundary may not always be correct. Furthermore, we show clear evidence that an auroral surge may also be located equatorward of the open-closed boundary. Therefore, tailward of the region of the plasma sheet to which the surge is connected there may exist undisturbed plasma sheet that has not yet been disconnected from the ionosphere. This means that substorm-associated reconnection does not necessarily begin to reconnect lobe field lines at the onset of a substorm. Moreover, available evidence strongly suggests that the arc that brightens at the onset of a substorm and that develops into a surge maps to the inner magnetotail, to that region at the inner edge of the plasma sheet where the magnetic field changes from a dipolar to a tail-like configuration. This would be consistent with recent studies that connect auroral breakup to the near-Earth (
