One example of the response of ionospheric convection and the polar cap boundary to a sudden change in the interplanetary magnetic field (IMF) orientation has been studied by using ground magnetometers, the Super Dual Auroral Radar Network (SuperDARN), and Defense Meteorological Satellite Program (DMSP) particle detectors when the IMF suddenly changed from northward (+ 6 nT) to strongly southward (-19 nT) at 1716 UT on 5 September 1995. The B-z component was fairly constant for similar to2 hours before and similar to25 min after the sudden IMF change. The convection flow changed almost simultaneously over a global extent. This initial change of the convection pattern can be characterized by a sudden formation of a large flow vortex in the afternoon sector. This agrees with the earlier findings by Ruohoniemi and Greenwald  and Ridley et al. . On the other hand, the response of the polar cap boundary (or its proxy) is more complicated. The Saskatoon radar, located in the late morning sector, observed an equatorward shift of the cusp scatter region simultaneously with the initial response of the convection flows. The DMSP particle data also showed a simultaneous equatorward expansion of the auroral oval in the 2100 magnetic local time (MLT) sector. The radar and particle data indicate the immediate equatorward expansion of the precipitation regions in the noon and premidnight sectors. About 10-20 min after the initial change, there were changes observed in the dusk region, namely, an equatorward expansion of the current reversal boundary observed by the Greenland magnetometer chain in the dusk sector between 1740 and 1750 UT and an equatorward expansion of the convection reversal boundary detected by the Stokkseyri, Halley, and Syowa radars. The delayed responses were observed 18-8 min before a substorm onset was recorded at midlatitude stations at 1756 UT. These observations indicate that there were two kinds of ionospheric responses to the southward turning of the IMF; the first response is the formation of the convection vortex and the equatorward shift of the polar cap boundary at noon and at similar to 2100 MLT, and the second response is the equatorward expansion of the convection reversal boundary in the dusk sector. We make the case that the first response is associated with the propagation of magnetosonic waves and that the second response is consistent with the Cowley and Lockwood  picture of the redistribution of the newly created open flux in the polar cap region.