The center of the circulation of the remains of Hurricane Alicia passed to the west of the National Severe Storms Laboratory's Doppler radar in Norman, Oklahoma on 19 August 1983. Three types of mesoscale precipitation were noted: the central area, principal rainband, and outer band. The central area and principal rainband were relatively stratiform, and the outer band was cellular. The maximum radial wind speed in the storm of 25–30 m s−1 was found in the principal band, and was probably in part a consequence of latent-heat release. The principal rainband, located to the cast and southeast of the storm's center, was responsible for much of the heavy rainfall over central Oklahoma, especially during the evening and nighttime hours, in response to low-level warm advection south of the cyclone late in the afternoon. We obtained a mesoscale, dual-Doppler analysis of the wind field at low levels from two single-Doppler radar volume scans recorded two hours apart. The basis of the analysis technique is that the bearing of the storm center from the radar changes appreciably over the period of observation. We refer to this technique, in which it is assumed that the mesoscale aspects of the wind field are quasi-steady between data collection periods, and in which the geometry of the baseline is determined by the motion of the feature being analyzed, as synthetic dual-Doppler analysis. The sensitivity of the synthetic dual-Doppler analysis to objective-analysis parameters and to estimates of the system's velocity is discussed. The analyzed wind field compared well with conventional surface and upper-air data. The mesoscale, cyclonic vortex associated with the storm was elliptically shaped, and elongated slightly along its direction of motion. Vertical cross sections of azimuthally averaged radial and azimuthal wind components were compared to similar analyses based upon data recorded aboard a NOAA aircraft prior to Alicia's landfall. A core region of approximate solid-body rotation had increased in radial extent significantly, while the highest azimuthal wind speed had decreased by a factor of two. The cyclone decayed much more rapidly than a barotropic cyclone in an Ekmann boundary layer, but less rapidly than a warm-core baroclinic cyclone whose heating source had been turned off. It is suggested that synthetic dual-Doppler analysis is a viable technique for studying the mesoscaleaspects of mobile cyclones and fronts with Doppler radars.