Up- and down-link attenuations were simultaneously measured at 30/20 and 14/12 GHz for more than 10 rainfall events from 1979 to 1981 with two Japanese geostationary satellites, Communication Satellite (CS) and Broadcasting Satellite for Experimental Purposes (BSE). Statistical analyses of the ratio of up-link attenuation to down-link attenuation, the “attenuation ratio,” show that attenuation ratios of both CS and BSE do not necessarily agree with theoretical values assuming the Marshall and Palmer raindrop size distribution and that attenuation ratios vary widely from one event to another. Raindrop size distributions measured by a disdrometer are employed for the analysis of measured attenuation ratios. It is found that disdrometer-derived attenuation ratios agree well with measured ones for BSE but that the agreements are not so good for CS. A model calculation of bright band attenuation and a data analysis with rain type classification suggest that the above discrepancy between measured and disdrometer-derived attenuation ratios is caused by bright band attenuation. It is concluded that disdrometer data are useful for the estimation of slant path attenuation ratio for the heavy rainfall rate range.