Earth-based radar has been one of the few, and one of the most important, sources of new information about Mercury during the three decades since the Mariner 10 encounters. The emphasis during the past 15 years has been on full-disk, dual-polarization imaging of the planet, an effort that has been facilitated by the development of novel radar techniques and by improvements in radar systems. Probably the most important result of the imaging work has been the discovery and mapping of radar-bright features at the poles. The radar scattering properties of these features, and their confinement to permanently shaded crater floors, is consistent with volume backscatter from a low-loss volatile such as clean water ice. Questions remain, however, regarding the source and long-term stability of the putative ice, which underscores the need for independent confirmation by other observational methods. Radar images of the non-polar regions have also revealed a plethora of bright features, most of which are associated with fresh craters and their ejecta. Several very large impact features, with rays and other bright ejecta spreading over distances of 1,000 km or more, have been traced to source craters with diameters of 80–125 km. Among these large rayed features are some whose relative faintness suggests that they are being observed in an intermediate stage of degradation. Less extended ray/ejecta features have been found for some of the freshest medium-size craters such as Kuiper and Degas. Much more common are smaller (<40 km diameter) fresh craters showing bright rim-rings but little or no ray structure. These smaller radar-bright craters are particularly common over the H-7 quadrangle. Diffuse areas of enhanced depolarized brightness have been found in the smooth plains, including the circum-Caloris planitiae and Tolstoj Basin. This is an interesting finding, as it is the reverse of the albedo contrast seen between the radar-dark maria and the radar-bright cratered highlands on the Moon.