D emography is an inherently spatial science, since it almost always deals with human populations in a defined geographic region, but spatial analysis has thus far played only a small role in the development and testing of demographic theory. There are several reasons for this, including the recency of many of the more useful spatial statistical approaches, and the fact that most people practicing demographic science are not in geography and have not been encouraged to think spatially. Yet, even in geography, few population specialists adopt specifically spatial approaches to their research beyond the measurement of the movement of people from one region to another, or the comparison of demographic trends among different regions. In the past few decades, demographic research has focused particularly on the analysis of survey data drawn from interviews conducted at the household level, and as a consequence, theory has focused heavily on individual-level influences on demographic behavior. The development of surveys such as the National Survey of Family Growth in the United States, and the US-funded Demographic and Health Surveys in less developed nations represented an important step in demographic research because the previous heavy reliance on aggregated data, especially from censuses and vital statistics, left gaps in our knowledge about how individuals think and behave. Now, however, the confluence of powerful geographic information system technologies, advances in the design of spatial statistics, and the increasing availability of georeferenced databases has improved vastly the ability of demographers to think spatially. As a result, there is a reawakening of interest in models of human behavior that place individuals in the environmental context of space and time. Demography is not only spatial, but it is also by nature interdisciplinary. The demographic transition, which provides the organizing framework for most demographic research, is really a complex set of transitions, each of which draws upon expertise in