The construction and improvement of farm buildings in the United States during the latter half of the nineteenth century was a task of farmcapital formation exceeded only by the effort to clear the land itself. As uses of the farmer's time and capital, the erection of the first crude shelters and the steady additions to the dwelling and to other farm buildings are of major interest and importance in the analysis of American agricultural development. The statistics available to measure this work are adequate to command attention; like all statistics of this period, they must, of course, be employed with caution.

American and European economists who work in the least developed countries of Africa, Asia, or the Middle East sometimes come away with the feeling of having learned more than they imparted. Nor is this surprising: the minds of economists are often more receptive to development than are the exotic economies in which they now work. In considering problems of underdevelopment and processes of development we learn—inadvertently, as it were—new things about conventional fields of economics and about the developed economies of Europe and America. These feedbacks have been particularly valuable to economic historians who have given us fresh insights into European, Russian, Japanese, and American development as a direct consequence of the present concern with developing the backward countries. Economic history is now wedded to economic development.

No one seriously questions the familiar association either between levels of urban development and degrees of industrial maturity or between rates of change in these two indices. It has even become commonplace in macroeconomic growth theory to simplify the complexities of structural change into some variation of the urban-rural two-sector model although, in the real world, shifts from commercial to industrial urban employment, let alone more complex intersectoral shifts, are of prime importance. Indeed, many of these growth models place great emphasis not upon changes in sector productivity but upon resource shifts between low- and high-productivity employment, while in empirical studies urban and rural population data very often appear as explicit substitutes for sectoral employment. But given the paucity of macroeconomic data for the antebellum period, especially prior to 1839, one is left puzzled by our relative inattention to the wealth of population census data by residence. American economic history textbooks are stuffed with quantitative information on the number of cities, their spectacular growth, and the percentage of population urbanized, but these may not be the most effective uses of this great data pool. The plethora of urban histories and the abundant attention to urban rivalry may be useful complements, but they are not very effective substitutes for quantitative analysis of overall American urbanization and experience with city size distribution. The very attention which the topic is currently receiving by economists, geographers, and historians suggests that much remains undone in the study of early American urbanization.