¹ Lampard, Eric E., “American Historians and the Study of Urbanization,” American Historical Review, 67 (Oct. 1961), 52CrossRefGoogle Scholar.

² Lampard, Eric E., “The History of Cities in the Economically Advanced Areas,” Economic Development and Cultural Change, 3 (Jan. 1955), 83–84CrossRefGoogle Scholar.

³ For a more comprehensive review of this literature see Lindstrom, Diane and Sharpless, John, “Urban Growth and Economic Structure in Antebellum America,” in Uselding, Paul, ed., Research in Economic History, vol. 3 (Greenwich, Conn., 1978), 161–216Google Scholar.

⁴ North, Douglass C., The Economic Growth of the United States, 1790 to 1860 (Englewood Cliffs, N. J., 1961)Google Scholar; Rubin, Julius, Imitation by Canal or Innovation by Railroad: A Comparative Study of the Response to the Erie Canal in Boston, Philadelphia and Baltimore, unpublished Ph.D. dissertation, Columbia Univ., 1959Google Scholar; Taylor, George Rogers, The Transportation Revolution, 1815–1860 (New York, 1951)Google Scholar.

⁵ Williamson, Jeffrey G., “Antebellum Urbanization in the American Northeast,” this Journal, 25 (Dec. 1965), 592–608Google Scholar; Crowther, Simeon J., “Urban Growth in the Mid-Atlantic States, 1785–1850,” this Journal, 36 (Sept. 1976), 624–43Google Scholar; Lindstrom, Diane L., “Demand, Markets and Eastern Economic Development: Philadelphia, 1815–1840,” this Journal, 35 (Mar. 1975), 271–73Google Scholar; Lindstrom and Sharpless, “Urban Growth and Economic Structure.”

⁶ In this sense our paper is more in the tradition of Pred, Allan R., The Spatial Dynamics of Urban-Industrial Growth, 1800–1914 (Cambridge, Mass., 1966)Google Scholar than that of Williamson, Jeffrey G. and Swanson, Joseph A., “The Growth of Cities in the American Northeast, 1820–1870,” Explorations in Economic History, 2nd series, 4 (Supp. 1966), 3–101Google Scholar.

⁷ Heilbrun, James, Urban Economics and Public Policy (New York, 1974), Ch. 7.Google Scholar

⁸ See Pred, Allan R., Urban Growth and the Circulation of Information: The United States System of Cities 1790–1840 (Cambridge, Mass., 1973)CrossRefGoogle Scholar for a complementary approach emphasizing information flows.

⁹ Nourse, Hugh O., Regional Economics (New York, 1968), Ch. 3.Google Scholar

¹⁰ Taylor, Transportation Revolution, p. 246. See also Pred, Urban Growth, p. 4, and Sharpless, John, City Growth in the United States, England and Wales, 1820–1861 (Ann Arbor, 1975), p. 130.Google Scholar

¹¹ Executive Office of the President, Office of Management and Budget, Standard Metropolitan Statistical Areas, Revised (Washington, 1975)Google Scholar.

¹² A listing of the 103 SMS As is available on request.

¹³ Executive Office of the President, Standard Areas, p. iii.

¹⁴ Ibid., p. 1–4.

¹⁵ Note that Williamson and Swanson, “Growth of Cities,” Crowther, “Urban Growth” and Lindstrom and Sharpless, “Urban Growth and Economic Structure” use differing definitions of a city's hinterland. All these authors, however, use a broader geographical definition of hinterland than that applied in this paper.

¹⁶ In those cases where the SMSA contains more than one central city (for example, the Allentown-Bethlehem-Easton SMSA) the population and manufacturing employment reported for all components was added to derive the city (versus hinterland) statistic used below.

¹⁷ Bureau of Economic Analysis, U. S. Dept. of Commerce and Economic Research Service, U. S. Department of Agriculture, 1972 OBERS Projections of Regional Activity in the United States, vol. 2 (Washington D. C., 1974)Google Scholar.

¹⁸ Moses, Leon and Williamson, Harold F. Jr., “The Location of Economic Activity in Cities,” The American Economic Review, 57 (May, 1967), 212Google Scholar. A final justification for adopting the SMSA concept rests on the rather robust results, reported below, examining the relationship between 1860 and 1900 city and hinterland populations.

¹⁹ Heilbrun, Urban Economics, pp. 142–44. Also see Sharpless, City Growth Ch. 8, and Lindstrom and Sharpless, “Urban Growth and Economic Structure,” pp. 165–68, on the use of location quotients.

²⁰ Executive Office of the President, Standard Areas.

²¹ U. S. Bureau of the Census, Historical Statistics of the United States, Colonial Times to 1970 (Washington, 1975), Series A5–72, pp. 11–12Google Scholar.

²² The residuals, or more precisely the differences in city versus hinterland ranking, are interesting. Several New England cities (for example, Lowell, Fall River) rank much higher than their respective hinterlands. On the other hand smaller (newer) cities in the midwest exhibit the opposite tendency.

²³ This conclusion, although based on a more restrictive geographical definition of an urban hinterland, is consistent with that advanced by Crowther, “Urban Growth,” Lindstrom, “Eastern Economic Development,” Lindstrom and Sharpless, “Urban Growth and Economic Structure,” and Williamson and Swanson, “Growth of Cities.”

²⁴ A Spearman rank correlation of city and hinterland population growth from 1800 to 1830 (n=30) yields a coefficient of +0.329. This is significant at the 90 percent confidence interval indicating the importance of intra-area trade to urbanization patterns during this earlier period.

²⁵ Since New England manufacturing employment with one exception was reported on a county basis whereas SMS As are composed of townships (or parts of counties), it was necessary to estimate the SMSAs' manufacturing components. This was accomplished by allocating county manufacturing employment to or between SMSAs on the basis of population data reported at the township level. Given the labor market homogeneity of the SMSA definition, this technique should adequately approximate true SMSA manufacturing employment. For the Boston SMSA, where a crosscheck was possible, this method resulted in an estimate of 80,507 versus the census reported 80,614.

²⁶ If we restrict the analysis to the 29 SMSAs for which 1800–60 population statistics are available (thus eliminating “new” cities), and those that are specialized in manufacturing, measured by a location quotient greater than one, the rank correlation coefficient is +0.410. This is significant at the 95 percent confidence interval. These results further support the analysis advanced by Lindstrom and Sharpless, “Urban Growth and Economic Structure,” 165–69.

²⁷ Eighty-six of the ninety SMSAs contained central cities with a population of at least 2,500, the usual urban criteria, by 1860.

²⁸ Taylor, Transportation Revolution, p. 389.

²⁹ The similarity of this result to that obtained by Deane and Cole is intriguing. See Deane, and Cole, , British Economic Growth, 1688–1959 (Cambridge, 1964), pp. 106–22Google Scholar. Also see Williamson and Swanson, “Growth of Cities,” 44–58, and Lindstrom and Sharpless, “Urban Growth and Economic Structure,” 184–85.

³⁰ Neither restricting our analysis to those SMSAs containing cities of at least 2,500 inhabitants in 1830 (n = 36, R_r = +0.174) nor examining those SMSAs with an 1860 location quotient greater than one and city population of at least 2,500 (n = 34, R_r = -0.099) changes our conclusion for the 1830–60 period. In these two tests the correlation coefficients are insignificantly different from zero.

³¹ U. S. Bureau of the Census, Historical Statistics, pp. 11–12.

³² Although our methods do not allow for identification of the differential factors at work in the post- versus antebellum milieu, it appears reasonable to hypothesize that a combination of (1) integration of an SMSA-oriented national rail system, (2) increased agglomeration economies, and (3) rapid growth in intermediate and final (as opposed to resource-oriented initial) stages in the manufacturing sequence explain this discontinuity with the past.

³³ Our statistical findings on the urbanization forces at work in the United States system of cities during the nineteenth century thus are broadly congruent to the conceptual framework and empirical evidence advanced in Pred, Spatial Dynamics.