How many physicians are needed? Can they be obtained? A framework within which to explore these two questions is proposed together with a review of recent literature. Section I outlines the increase in planning. Section II defines the goal of the delivery system as an improvement in the health of the population and raises the difficulties of implementing the goal. Section III reviews methods for forecasting need: (1) professionally defined standards, (2) physician/population ratios, (3) economic models of demand, and (4) system models of the delivery system. The first two bear little relation to the expected utilization of physicians or to the goal of improving health. Economic models reflect utilization, but neglect access costs. System models suffer from estimation problems and the lack of outcome measures. Section IV focuses on the supply of physicians; they seem to respond to environmental and other amenities, with less emphasis on income; they seek proximity to colleagues and modern facilities. Six ways to determine if a shortage exists are reviewed in Sec. V. Section VI demonstrates that the supply of physicians has continually been underestimated. Locational inequity appears to be the major unsolved problem.