The microwave brightness temperature measurements from the electrically scanned microwave radiometer (frequency=37 GHz) are compared with oceanic wind measurements from data buoys. It is shown that the brightness temperature can be manipulated to yield a measure of the surface roughening, which can be very well accounted for by a simple geometric optics model. The data of Hollinger (1971) at 1.4, 8.36, and 19.34 GHz were similarly manipulated and shown to require a surface with less slope variance than that predicted by the Cox and Munk (1955) optical measurements. It is also shown that the surface may be treated as isotropic to an accuracy equivalent to the roughening produced by a 2-m/s wind speed increment.