This Memorandum presents the findings of a pilot study that deals with the determinants of residential land values in an urban area. As part of the RAND Urban Transportation effort, one primary purpose of this study was to develop a land-value submodel for incorporation in a model designed to simulate the effects of changes in the transportation system of an urban community. The model relates land values to several different factors, particularly to a sites amenities and accessibility to various forms of economic activity. Since residential land values were of main concern, the study concentrates on a sample of single-family properties. As a supplement, and as a basis of comparison, included is a sample of all land-use properties for urban areas. There were important differences between the two samples, and, as might be expected, much more regular patterns occur in the homogeneous single-family grouping. The findings revealed a sufficiently high degree of stability to confirm the basic nature of the relationships, but random variations made the models too unstable to use for accurate predictions of individual parcel land values.