Does the built environment affect how often and how far people drive or walk or when they will take the bus or the train? If so, how? A lively, expanding literature continues to investigate the potential for causal links between urban design and travel behavior, yet there remain many gaps and considerable disagreement. Our purpose here is mainly to identify what past research has to say on these questions. We also try to explain why these studies reach different conclusions and how and where this work might be usefully improved. The first, and perhaps best-known, group of studies on this topic investigates how travel behavior and travel investment affect land use. There is also a long if more recent practice of viewing these links from the opposite direction; that is, how does land use influence urban travel? We consider this second question in more detail following a brief review of the first. Though not our focus, most questions about land-use/transportation links over the past century concern the influence of transportation infrastructure on development patterns. Analysts ask how highways and mass transit contribute to decentralization trends, how they affect the local balance of jobs and housing, or how they affect the pattern of commercial investment (see, e.g., the reviews in Gómez-Ibáñez, 1985b; Giuliano, 1989, 1991, 1995a, 1995b; Cervero and Landis, 1995). The basic idea is this: People choose their homes and locate their businesses based in part on their proximity to work, other potential destinations, and the markets for their products and labor generally (see, e.g., Von Thunen, 1826; Weber, 1928; Losch, 1954; Alonso, 1964; Muth, 1969; Mills, 1972; Solow, 1973; Fujita, 1989; Anas, Arnott, and Small, 1997). That is, the cost of transporting people and things over space depends on the distances and resources required. Once these costs are fixed, perhaps by the establishment of a central downtown or transshipment point, the price of land at each location is determined by demand. This in turn is determined, again in part, by how much money one has left after accounting for the transportation costs associated with that location.