We derive profiles of the aerosol extinction-to-backscatter ratio, Sa, at 355 nm using aerosol extinction and backscatter profiles measured during 1998 and 1999 by the operational Raman lidar at the Department of Energy Atmospheric Radiation Measurement Program Southern Great Plains site in north central Oklahoma. Data from this Raman/Rayleigh-Mie lidar, which measures Raman scattering from nitrogen as well as the combined molecular (Rayleigh) and aerosol (Mie) scattering at the laser wavelength, are used to derive aerosol extinction and backscattering independently as a function of altitude. Because this lidar operates at 355 nm, where molecular backscattering is comparable to aerosol backscattering, Sa retrievals are generally limited to conditions where aerosol extinction at 355 nm is >0.03 km-1. The mean value of Sa at 355 nm derived for this period was 68 sr with a standard deviation of 12 sr. Sa was generally about 5--10 sr higher during high aerosol optical thickness (AOT) (>0.3) conditions than during low AOT (15%) variations in the vertical profile of Sa occurred about 30% of the time, which implies that significant variability in the vertical distribution of the aerosol size distribution, shape, and/or composition often occurs. The Raman lidar measurements of Sa were compared with estimates of particle size and refractive index derived from an algorithm that uses ground-based Sun photometer measurements of Sun and sky radiance. For 17 cases of coincident Raman lidar and Sun and sky radiance measurements, Sa was linearly correlated with the aerosol fine mode effective radius and the volume ratio of fine/coarse particles. ¿ 2001 American Geophysical Union