[en] The Reactor Safety Study (WASH-1400), originally commissioned by the United States Atomic Energy Commission and published in final form under the auspices of the Nuclear Regulatory Commission, is reviewed. In order to be useful for policy purposes, a study of the nuclear safety issues must necessarily deal with questions directly relevant to real choices between different energy technologies and between alternative power plant sites. It is pointed out here that while the Reactor Safety Study contains excellent summaries of background information for assessing nuclear risks, such policy relevant questions have not been properly addressed. Furthermore, certain large ambiquities in the estimation of nuclear accident probabilities and consequences have not been incorporated in the WASH-1400 calculations. To place these difficulties in context, the probabilities of several particularly important modes of reactor failure are discussed, and independent estimates of the consequences of a major accident at the Indian Point, Westchester County, New York site are presented. (34 references)