A substorm surge on September 25, 1981 was observed by the imagers on the high-altitude DE-1 spacecraft and encountered by the coplanar, low-altitude DE-2 spacecraft. This case is unique among those that we examined in that DE-2 traversed the surge during the ~10--15 min period of its rapid poleward motion through the plasma sheet. Because of this, the DE-2 measurements provide a critical test for the recent proposal that the substorm expansion phase is due to an anti-sunward-propagating reduction in the large-scale magnetospheric electric field imparted to the magnetosphere from the solar wind [Lyons, 1995>. A specific prediction of this theory is that the electric field reduction in the equatorial magnetosphere maps to the ionosphere as a poleward-moving electric field reduction. In the ionosphere, large growth-phase electric fields are expected poleward of the active surge aurora, and significantly weaker electric fields are expected equatorward of the active aurora. This electric field pattern is expected to propagate poleward through the plasma sheet with the surge. The DE-2 measurements show significant southwestward electric fields within the portion of the plasma sheet poleward of the active auroral region, and greatly reduced electric fields within the heated central plasma sheet that was left behind by the narrow region of poleward-moving, active aurora. We also find weak electric fields equatorward of the region of active aurora in typical DE-2 passes over auroral surges, where the region of active aurora had moved poleward to very near the magnetic separatrix prior to the satellite pass. These observations agree strongly with the Lyons [1995> predictions. Consistent with previous observations, the DE-2 measurements also show strong and variable electric fields within the region of active surge; however these fields are not specifically addressed by the theory. ¿ American Geophysical Union 1996