The present study returned to two of the same northern California locations in the winter of 2003-2004 to further examine rain characteristics, again using S-band profilers, but this time also employing collocated raindrop disdrometers to directly measure the DSD at the surface. These observations were obtained as part of NOAA's Hydrometeorology Testbed (HMT) project. One S-band/disdrometer/gauge site was located on the coastline very near sea level and the other was approximately 10 km inland at 475 m MSL in the Coastal Mountain Range. Using the S-band profiler data, the precipitating periods were again objectively categorized as brightband, non-brightband, or convective. The brightband rain spectra were very similar to the classical DSD that Marshall and Palmer observed in stratiform winter rain. Drop concentrations are nearly equal for BB and NBB periods at D = 1 mm. However, the non-brightband periods contain nearly an order of magnitude fewer large drops (D > 3 mm) and roughly a factor of 3 more small drops (D<0.6). Z-R relations (Z = aRb) derived from the disdrometer data and from the S-band/gauge data at both sites show much smaller coefficients (a) for the non-brightband periods. These findings are consistent with the earlier work, in which the DSD were inferred by remote sensing, and lend further evidence that the NBB spectra are produced by a shallow, but orographically-enhanced, condensation and coalescence process with no involvement from large ice particles. In contrast, the BB spectra are produced at least partly from melted large snowflakes originating in deeper, colder regions of the storms.