The main objective of this paper is to examine the frequency dependence of slant path rain attenuation at 15 and 35 GHz in order to determine how well the attenuation at one frequency can be predicted from that at another; this has application in the design of earth-to-satellite communications. Also, some inferences on the drop size distribution of rain can be drawn from simultaneous attenuation measurements at two frequencies. The theory of rain attenuation is reviewed, and it is noted that the attenuation is a complex function of drop size, shape, orientation, index of refraction, and rain intensity along the path. Attenuation ratios are computed on the basis of both drop size and rain rate for frequencies at which there are measured data. Results obtained by other investigators are reviewed, and it is found that frequently, attenuation ratios that would not have been predicted on the basis of a surface rain model such as a Laws and Parsons model have been reported; indications are that attenuation is often produced from a widely dispersed distribution of large raindrops. During eight rainy days in 1975–1976, more than 10,000 simultaneous measurements of 15- and 35-GHz attenuations were recorded in the Boston area. On the basis of these results it is concluded that it is not possible to represent the drop size distribution of rain along a slant path in the Boston area by a Laws and Parsons model, since the ratios of the 35-GHz attenuations to the 15-GHz attenuations are significantly below those that would have been predicted by that model. The ability to predict slant path rain attenuation at one frequency on the basis of that at another is shown to be a function of frequency separation, climatology, and the type of statistics which are desired.