In 1963 the Beectling Report on Britain's railways proposed the closure of a substantial part of the railway network (British Railways Board, 1963). Although many services were withdrawn, disquiet at the Report's failure to analyse the social benefits of rail passenger services eventually led to the application of the technique of cost-benefit analysis to the problem of unremunerative rail services. However, although such studies revealed that the retention of many rural rail services could not be justified on social grounds, these services were not withdrawn. This paper outlines briefly the history of attempts to deal with the question of the "optimal size" of the rail passenger network. Secondly, it considers the social cost-benefit case for reducing the present size of the network in Britain and the quantifiable benefits from such a reduction. The problem of joint costs is discussed. Finally, it attempts to explain the failure of successive Governments to apply the results of cost-benefit studies in practice, and suggests how the decision-making process might be altered to facilitate a more rational approach towards public transport problems in Britain.