Two contributions are provided to the economic and transportation literature on rail transit ridership fare elasticities. First, several fare variables are constructed with controls for the decentralized nature of metropolitan employment; this relaxes the traditional assumption that rail transit patrons commute to a central city station. Second, rail transit demand elasticities are estimated using data from the Washington, D.C., metropolitan area rail transit system; research on this system has been absent from the literature on demand elasticities. Analysis of the extensive data shows that restricting rail patron travel patterns according to the monocentric urban model underestimates by 35% to 68% the responsiveness of ridership to fare increases.