We determine the direct radiative forcing of Saharan dust aerosols by combining aerosol information derived from Nimbus-7 TOMS with radiation measurements observed at the top of atmosphere (TOA) by NOAA-9 ERBE made during February-July 1985. Cloud parameters and precipitable water derived from NOAA-9 HIRS2 were used to aid in screening for clouds and water vapor in the analyses. Our results indicate that under cloud-free and dry conditions there is a good correlation between the ERBE TOA outgoing longwave fluxes and the TOMS aerosol index measurements over both land and ocean in areas under the influence of airborne Saharan dust. The ERBE TOA outgoing shortwave fluxes were also found to correlate well with the dust loading derived from TOMS over ocean. However, the calculated shortwave forcing of Saharan dust aerosols is very weak and noisy over land for the range of solar zenith angle viewed by the NOAA-9 ERBE in 1985. Sensitivity factors of the TOA outgoing fluxes to changes in aerosol index were estimated using a linear regression fit to the ERBE and TOMS measurements. The ratio of the shortwave-to-longwave response to changes in dust loading over the ocean is found to be roughly 2 to 3 but opposite in sign. The monthly averaged clear-sky TOA direct forcing of airborne Saharan dust was also calculated by multiplying these sensitivity factors by the TOMS monthly averaged clear-sky aerosol index. Both the observational and theoretical analyses indicate that the underlying surface properties, dust layer height, ambient moisture content, and the presence of cloud all play important roles in determining the TOA direct radiative forcing due to mineral aerosols. ¿ 2000 American Geophysical Union