Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Geography and Anthropology

First Advisor

Carville Earle


The United States has undergone massive changes in its population geography since 1790. A major component of these population shifts has been the rapid movement of the frontier westward. This dispersal of population through migration from relatively high densities along the eastern seaboard to the sparsely populated west was followed by concentration across most of the country after the general "closure" of the frontier. The rates of settlement growth and population concentration trends were not steady, but cyclical in nature, and they varied among the different regions of the country. This variation leads to a comparison of these phenomena with long-wave trends. Settlement growth rates are measured for each decade. Settled area includes all counties with at least two persons per square mile at the time of the census. Two fundamentally different methods are employed to measure population and manufacturing concentration. These are the Hoover index and spatial autocorrelation coefficients (Moran's l, Geary's c, and local measures). Both employ county-level population, manufacturing employment, and areal data, while spatial autocorrelation calculations also use digital county maps. The Hoover index of population concentration is figured at the city-system, regional, and national scales for all decades from 1790 to 1990. The results of the frontier settlement analysis show a marked fall of frontier settlement after 1910. The population concentration section indicate declines in national population concentration from 1790 to 1910 (using constant area), followed by marked concentration. These two findings support 1910 as the year of frontier "closure.". Additionally, the relationship of manufacturing employment concentration to population concentration is explored from 1840 to 1990. Estimations of manufacturing employment for counties are made for 1910, because these data were not included in the 1909 Census of Manufactures. The absolute differences between population concentration and manufacturing employment concentration measures are calculated at the national, regional, and city-system levels. Manufacturing employment generally has been more concentrated than population, using the Hoover index, with the greatest differences occurring between 1900 and 1910. Numerous maps and figures are included to illustrate the westward expansion, as well as population and manufacturing employment concentration trends.