A network of ionosondes in the Australian-Japanese region recorded the ionospheric response to a major magnetic storm that occurred near 2400 UT on November 3, 1993, when Kp reached 7. Most stations recorded a positive phase (increase) in NmF2 on November 4 which was well produced by the field line interhemispheric plasma (FLIP) model despite a large relative increase in the model molecular neutral densities. The positive phase appears to be related to a storm-induced large abrupt uplift of the ionospheric peak height (hmF2) that may have been caused by a neutral wind surge or by an electric field. The storm caused large winds to blow from the pole to the equator, raising the measured hmF, by 50--100 km over quiet time values. There was a negative phase (decrease) on November 5 at the higher midlatitude stations but not at the low-latitude stations. The negative phase was not well reproduced by the model. At low latitudes, a short-period wave-like structure was seen in hmF2, NmF2, and equivalent winds. The quiet time equivalent winds in the eastern Australian sector show a pronounced semidiurnal tide signature at all latitudes with a phase shift of about 1 hour for every 15 deg increase in latitude. An algorithm which adjusts the atomic to molecular neutral density ratio by changing the neutral temperature and atomic oxygen density was used to help reproduce the observed electron density. The changes to the standard neutral density ratio were small at Hobart except during the stormy period. At Townsville, the needed changes to the neutral density ratio were generally small at daytime, but large increases were needed at night during quiet periods. The algorithm is unable to uniquely determine whether the changes in neutral density ratio come from changes in temperature or changes in atomic oxygen density. When averaged over the full 14-day campaign, there was good agreement between the modified and standard model neutral temperatures except at the low-latitude stations at night. In this study the ionospheric model reproduced the nighttime ionosphere well at the higher latitudes, but the lower the latitude, the more the model underestimated the nighttime density. Because the low-latitude stations do not have a significant plasmasphere to support the ionosphere, this study provides some evidence that problems in modeling the nighttime ionosphere may not be due to uncertainties in the plasmaspheric ion flux. ¿ 1998 American Geophysical Union