This paper discusses the extent of rape in the U.S. today, including a critical analysis of federal and independent sources of incidence data. Rape incidence estimates derived from the National Crime Survey (NCS) are flawed because of measurement methods that undermine full disclosure of victimization. Conclusions include (a) the incidence of rape is much higher than federal statistics suggest, but data are insufficient to resolve whether an epidemic is underway, (b) acquaintance rape is far more common than documented in crime surveys, and (c) improvement in rape detection is promised by the proposed revisions in the NCS.