To measure the radar backscatter from the ocean with enough accuracy to allow determination of the wind vector, the effect of attenuation in precipitating regions over the sea must be considered. The attenuation can be determined reasonably well from a knowledge of the brightness temperature observed by a microwave radiometer. For the SEASAT this radiometer was the SMMR. Two algorighms were developed to relate radiometric measurements and attenuation and hence to correct the measured values of scattering coefficient. The two algorithms were compared with the best available information on precipitation, but no direct information on attenuation was available. Hence, the basic comparisons reported here are between the wind vectors reported by surface observers and those determined by the SASS measurements with and without correction for atmospheric attenuation. In general, the attenuation correction appears to improve the results obtained. However, the 'correction' is not as good as it should be because of the differing size of the 18 GHz radiometer footprint and the SASS footprint, and because the scan patterns are different so that averages from several radiometer footprints may be involved in a single SASS correction.