In revising this article, I benefited from the helpful comments of Gavin Wright, Willam N. Parker, Michael Edelstein, Gerald Jaynes, and an anonymous referee. I would also like to acknowledge the research assistance of John Caparusso.Google Scholar

¹ Brooks, Robert Preston, The Agrarian Revolution in Georgia, 1865–1912 (Madison, Wis., 1914), p. 7.Google Scholar

² Brooks, Robert Preston, The Agrarian Revolution in Georgia, 1865–1912 (Madison, Wis., 1914), pp. 69, 94.Google Scholar

³ Brooks, Robert Preston, The Agrarian Revolution in Georgia, 1865–1912 (Madison, Wis., 1914),, pp. 79–80, 89.Google Scholar To explain the predominance of share tenancy in the Upper Piedmont, Brooks offers the same hypothesis as Higgs, Robert, “Patterns of Farm Rental in the Georgia Cotton Belt, 1880–1900,” this JOURNAL, 34 (06 1974), pp.474–75, relating the cost of supervision to the size of landholdings.Google Scholar

⁴ Ransom, Roger and Sutch, Richard, One Kind of Freedom: The Economic Consequences of Emancipation (New York, 1977), pp. 151–56;Google ScholarWright, Gavin and Kunreuther, Howard, “Cotton, Corn and Risk in the Nineteenth Century,” this JOURNAL, 35 (09 1975), pp. 526–27;Google ScholarWright, Gavin, Political Economy of the Cotton South (New York, 1978), pp. 164–66;Google ScholarTemin, Peter, “Patterns of Cotton Agriculture in Postbellum Georgia,,” this JOURNAL, 43 (09 1983). pp. 661–74.Google Scholar

⁵ Alternative explanations of the trend in crop production can be found in DeCanio, Stephen, “Cotton ‘Overproduction’ in Late Nineteenth Century Southern Agriculture,” this JOURNAL, 33 (09 1973), pp. 608–33;Google ScholarRansom, Roger and Sutch, Richard, “The ‘Lock-In’ Mechanism and Overproduction of Cotton in the Postbellum Cotton South,” Agricultural History, 49 (04 1975), pp. 405–25;Google ScholarWright, Gavin and Kunreuther, Howard, “Cotton, Corn and Risk in the Nineteenth Century,” pp. 526–51;Google ScholarMandle, Jay R., The Roots of Black Poverty: The Southern Plantation Economy after the Civil War (Durham, N.C. 1978), chaps. 1–5;Google ScholarWiener, Jonathan M., “Class Stucture and Economic Development in the American South, 1865–1955,” American Historical Review, 84 (10 1979), pp. 970–92;CrossRefGoogle Scholar and McGuire, Robert A., “A Portfolio Analysis of Crop Diversification and Risk in the Cotton South,” Explorations in Economic History, 17 (10 1980), pp. 342–71.CrossRefGoogle Scholar

⁶ These regional divisions are from Gray, Lewis Cecil, History of Agriculture in the Southern United States to 1860, vol. 1 (Clifton, N. J., 1973), pp. 532–37;Google Scholarand Brooks, Agrarian Revolution in Georgia, pp. 69–72.Google Scholar

⁷ Emerson, F. V., “Geographic Influences in American Slavery,” Bulletin of the American Geographical Society of New York, 43 (01–03 1911), pp. 13–26, 106–18, 170–81,CrossRefGoogle Scholarprovides a general discussion of how ecological conditions influenced the development of the Southern plantation system. For a more detailed description of ecological conditions in the Georgia Cotton Belt, see Loughridge, R. H., “Report on the Cotton Production of the State of Georgia,” in U.S. Census Office, Tenth Census, 1880, vol. 6,Google ScholarHilgard, Eugene W., Special Agent, Report on Cotton Production in the United States, Part II (Washington, D.C., 1884), pp. 259–450;Google Scholar and Harper, Roland M., “The Natural Resources of Georgia,” Bulletin of the University of Georgia, 30 (02 1930), pp. 1–155.Google Scholar

⁸ Harper, Roland M., “The Natural Resources of Georgia,” Bulletin of the University of Georgia, 30 (02 1930), 1–155.Google Scholar

⁹ Gray, History of Agriculture in the Southern United States, pp. 488–92;Google ScholarAllman, John M., III, “Yeoman Regions in the Antebellum Deep South: Settlement and Economy in Northern Alabama, 1815–1860” (Ph.D. dissertation, University of Maryland, 1979), chap. 2;Google ScholarHahn, Steven, The Roots of Southern Populism: Yeoman Farmers and the Transformation of the Georgia Upcountry, 1840– 1890 (New York, 1983), part I;Google Scholar and Weiman, David F., “Petty Commodity Production in the Cotton South: Upcountry Farmers in the Georgia Cotton Economy, 1840 to 1880” (Ph.D. dissertation, Stanford University, 1983), chap. 2.Google Scholar

¹⁰ Owsley, Frank L., Plain Folk of the Old South, (Chicago, 1965), pp. 23–50.Google Scholar

¹¹ In 1860 Upcountry households with fewer than ten slaves owned almost 40 percent of the region's slave population, while in the Cotton Belt, small slaveholders owned less than 20 percent of the slave population; U.S. Census Office, Eighth Census, 1860, vol. 3, Agriculture of the United States in 1860, pp. 224–27.Google Scholar

¹² Estimates of the seed and feed requirements on Southern farms and the nutritional value of food crops relative to corn are taken from Gallman, Robert E., “Self-Sufficiency in the Cotton Economy of the Antebellum South,” Agricultural History, 44 (01 1970), p. 7;Google ScholarRansom, Roger and Sutch, Richard, “Debt Peonage in the Cotton South after the Civil War,” this JOURNAL, 32 (09 1972), pp. 659–62; and Ransom and Sutch, One Kind of Freedom, appendix E. In calculating the residual food supply for the regions of Georgia in 1860 and 1880, I have followed the procedure used by Ransom and Sutch.Google Scholar

¹³ Cotton output is measured in 400-pound bales. Because the average weight of a bale of cotton in 1880 was 475 pounds, output levels in this year have been adjusted accordingly.Google Scholar

¹⁴ The average temperature in the Upper Piedmont was only two degrees (Farenheit) lower than in the southern section, and these colder temperatures were concentrated in the winter. Harper, “The Natural Resources of Georgia,” pp. 63, 65.Google Scholar In addition, Ransom and Sutch, “Economic Regions of the South in 1880,” Southern Economic History Project Working Paper Series, No. 3 (Berkeley, Calif., 1971), do not distinguish between the Upper and Lower Piedmont of Georgia.Google Scholar

¹⁵ Gray, History of Agriculture in the Southern United States, pp. 452–58;Google ScholarSmith, Alfred G., Economic Readjustment of an Old Cotton State (Columbia, S.C., 1958), p. 159;Google ScholarBonner, James C., History of Georgia Agriculture 1732–1860 (Athens, Ge., 1964), pp. 53–54. According to the ledgers and cash book of Thomas Morris, a merchant from Franklin county, Upcountry farmers received 2 to 2.25 cents a pound for unginned cotton (which is equal to 6 to 6.75 cents a pound for lint cotton) in the 1840s and 1850s, when the price of cotton in New Orleans was 8 to 9 cents a pound. The Ledgers of Thomas Moms and Morris and Freeman, Cash Books at the Georgia Department of Archives and History.Google Scholar

¹⁶ Phillips, Ulrich B., A History of Transportation in the Eastern Cotton Belt to 1860 (New York, 1908), chaps. 1, 7.Google Scholar

¹⁷ Weiman, “Petty Commodity Production in the Cotton South,” pp. 215–16, 221–22.Google Scholar The spread of cotton culture into the Cherokee Territory accounts for the large change in the crop mix in the Upper Piedmont in the 1850s. In the Cherokee Territory the ratio of cotton to corn production increased from 4.7 to 12.0 bales per (thousands of) bushels between 1850 and 1860, as compared to a change of only 8.2 to 10.6 bales per bushels in the eastern half of the Upper Piedmont. The regional concentration of growth in cotton production does not support Hahn's, Steven thesis, Roots of Southern Populism, pp. 48–49, that increasing specialization in cotton manifested the “emerging” contradictions in Upcountry society on the eve of the Civil War, but, as demonstrated below, anticipates the widespread economic changes in the region during Reconstruction.Google Scholar

¹⁸ Ransom and Sutch, One Kind of Freedom, pp. 151–56; Wright and Kunreuther, “Cotton, Corn and Risk in the Nineteenth Century,” pp. 526–27; Wright, Political Economy of the Cotton South, pp. 164–66. Temin, Peter, “Patterns of Cotton Agriculture in Postbellum Georgia,” pp. 661–74;Google Scholar and McGuire, Robert and Higgs, Robert, “Cotton, Corn, and Risk in the Nineteenth Century: Another View,” Explorations in Economic History, 14 (04 1977), pp. 167–82, have questioned the extent of the shift to cotton on Southern farms after the Civil War.CrossRefGoogle Scholar

¹⁹ According to the state census, Upcountry farmers cultivated 1.13 million acres of land in 1875, while the federal census reports 1.86 million acres in crops in 1880. This difference is largely due to the deficiencies of the state census of agriculture which was suspened after 1875. See Table 5.Google Scholar

²⁰ Compare the derived estimates of the crop mix in 1875 with the crop mix in 1860 and 1880.Google Scholar

²¹ As many economic historians have noted, farm households in the Cotton Belt cultivated smaller crops of grains after the Civil War, because they had fewer stocks of swine and cattle to maintain. See for example, Reid, Joseph D. Jr, “White Land, Black Labor and Agricultural Stagnation: The Causes and Effects of Sharecropping in the Postbellum South,” Explorations in Economic History, 16 (01 1979), pp. 46–48.CrossRefGoogle Scholar But, without adequate holdings of livestock, they had to purchase meat and animal products from local country stores. Interregional trade patterns, based on rail shipments from the Midwest and upper South to Atlanta in 1876, support this conclusion, for they demonstrate that animal products comprised over half of the value of foodstuffs shipped to stations in the Cotton Belt through Atlanta, while corn and other grains accounted for only 20 percent of these shipments. See Georgia, , Department of Agriculture, Annual Report of the Commissioner (Atlanta, 1876), pp. 49–54;Google Scholar and Western and Atlantic Railroad, Annual Report of the Officers (Atlanta, 1867, 1869).Google Scholar

²² Ransom and Sutch, One Kind of Freedom, appendix E, estimate that the minimum subsistence needs of the rural population was equal to 15 bushels of corn per capita. The regional difference in farm self-sufficiency is also evident in the analysis of food production in Ransom and Sutch, “Debt Peonage in the Cotton South,” pp. 662–65.Google Scholar

²³ Temin, “Patterns of Cotton Agriculture in Postbellum Georgia,” pp. 664–65.Google Scholar

²⁴ The degree of specialization in cotton increased in all regions of the five cotton states, except in the Tennessee Valley of Alabama and in the northeast prairie of Mississippi. The largest change in the crop mix occurred in South Carolina, Louisiana and the delta regions of Mississippi. The results of this analysis are available from the author upon request. I am grateful to Gavin Wright for providing me with the data set used to make these calculations.

Because the 1859 cotton harvest was unusually large in the New South, the results presented here underestimate the change in the crop mix between 1860 and 1880 for the regions of Alabama, Mississippi and Louisiana. See Schaefer, Donald F., “The Effect of the 1859 Crop Year Upon Relative Productivity in the Antebellum South,” this JOURNAL, 43 (12 1983), pp. 859–63.Google Scholar

²⁵ Wright and Kunreuther, “Cotton, Corn and Risk in the Nineteenth Century;” Reid, “White Land, Black Labor and Agricultural Stagnation;” and Temin, “Patterns of Cotton Agriculture in Postbellum Georgia”.Google Scholar

²⁶ The regression equation tries to determine the effect of farm size and, in particular, of small farms on specialization in cotton. Using county data, the equation relates the share of tilled acreage in cotton to average farm size and the percentage of small-sized and medium-sized farms, controlling for differences in population density and the rate of tenancy. The estimated regression equations are presented in the Appendix. The results of the analysis differ substantially from those obtained by Wright and Kunreuther. Specialization in cotton is strongly correlated with the rate of tenancy only in Cotton Belt counties and with the percentage of small farms only in Upcountry counties. More generally, a Chow test rejects the hypothesis that the equations for the Cotton Belt and Upcountry were derived from the same underlying sample at a 1 percent confidence level.Google Scholar

²⁷ Temin, “Patterns of Cotton Agriculture in Postbellum Georgia” pp. 669, 774. The relationship between specialization in cotton and race cannot be interpreted causally, as Temin seems to imply in his analysis: “Specialization was a Piedmont rather than a southern phenomenon. It seems likely, moreover, that this geographical diversity can be explained by the racial composition of the regional inhabitants, rather than by farm size and tenure variables (p. 674). ” Because the racial composition of the population effectively discriminates between counties in the Cotton Belt and the Upcountry, the density of blacks and whites per acre are negatively correlated with each other and are strongly correlated with other possible determinants of the crop mix, such as the rate of tenancy and average farm size. The pattern of intercorrelations explains why “neither the farm size nor the tenure variables has a significant effect” on the crop mix in his regression equations (p. 669).Google ScholarDeCanio, Stephen derived a similar relationship between cotton production and race in his study, Agriculture in the Postbellum South: The Economics of Production and Supply (Cambridge, Mass., 1974), pp. 180–84, chap. 6.Google Scholar

²⁸ Excellent surveys of the issues discussed in this paragraph from different theoretical perspectives can be found in Woodman, Harold D., “Sequel to Slavery: The New History Views the Postbellum South, ” Journal of Southern History, 43 (11 1977), pp. 523–54;CrossRefGoogle ScholarParker, William N., “The South in the National Economy, 1865–1970,” Southern Economic Journal, 46 (04 1980), pp. 1019–48;CrossRefGoogle Scholar Jonathan Wiener, “Class Structure and Economic Development in the American South,” and Wright, Gavin, “The Strange Career of New Southern History,” Reviews in American History, 10 (12 1982), pp. 164–80.CrossRefGoogle ScholarSee also the symposium on One Kind of Freedom in Explorations in Economic History, 16 (Jan. and Apr. 1979).Google Scholar

²⁹ This explanation is advanced by Schwartz, Michael, Radical Protest and Social Structure: The Southern Farmers' Alliance and Cotton Tenancy, 1880–1890 (New York, 1976), pp. 63–67, 73–88;Google ScholarWright, Political Economy of the Cotton South, pp. 172–74; and Hahn, The Roots of Southern Populism, part II.Google Scholar

³⁰ This explanation is also offered by Flynn, Charles L. Jr, White Land, Black Labor Caste and Class in Late Nineteenth-Century Georgia (Baton Rouge, La., 1983), pp. 140–44.Google Scholar

³¹ Poor's Manual of the Railroads of the United States, vol. 1 (New York, 1868–1869).Google Scholar

³² Between 1860 and 1870, track mileage in Upcountry Georgia increased by only 36 miles, from 176.5 to 212.2 miles, while in the next decade it increased by 368 miles to a total of 580 miles. Poor's Manual of the Railroads, vols. 2, 12 (New York, 1869–1870, 1879–1880).Google Scholar

³³ Stover, John F., The Railroads of the South, 1865–1900 (Chapel Hill, N. C., 1951), pp. 110–14,Google Scholar discusses the importance of the Atlanta and Charlotte Air Line in completing a through line from Richmond to Atlanta; see also, Klein, Maury, “The Strategy of Southern Railroads,” American Historical Review, 73 (04 1968), p. 1059.CrossRefGoogle Scholar

³⁴ Of the 67 railroad stations in the Upcountry, 33 had telegraph stations by 1880; see Dun, R. G. and Company, Mercantile Agency Reference Books (Jan. 1880). I am grateful to the Dun and Bradstreet Company for permission to use these materials.Google Scholar

³⁵ U.S. Census Office, Eighth Census, 1860, vol. 1, Population of the United States in 1860, p. 75; U.S. Census Office, Ninth Census, 1870, vol. 1,Google ScholarThe Statistics of the Population of the United States (Washington, D.C., 1872), pp. 99–107;Google Scholarand U.S. Census Office, Tenth Census, 1880, vol. 1, Statistics of the Population of the United States, pp. 120–30. Only four cities and towns in the Upcountry, excluding Atlanta, had more than 750 people in 1860 and 1870, and their total population in 1870 was 8,677. Between 1870 and 1880, the number of cities and towns in the Upcountry, excluding Atlanta, increased to 14, and their total population grew to 21,308.Google Scholar

³⁶ Georgia, , Comptroller General's Office, Annual Report of the Comptroller General (Atlanta, 1860, 1870, 1880), pp. 44–49, 30–35, 133–36.Google Scholar The value of merchandise and holdings of money and liquid assets have been deflated by an index of retail prices derived by Hoover, Ethel D., “Retail Prices after 1850,” in Parker, William N., ed., Trends in the American Economy in the Nineteenth Century, vol. 24 (Princeton, 1960), p. 142. Deflating the nominal values by the wholesale price index for the period does not after the results substantially.Google Scholar

³⁷ Sylla, Richard, “Federal Policy, Banking Market Structure, and Capital Mobilization in the United States, 1863–1913,” this JOURNAL, 29 (12 1969), pp. 657–86;Google ScholarRansom and Sutch, “Debt Peonage in the Cotton South,” pp. 643–51 and One Kind of Freedom, pp. 106–14;Google Scholar and James, John, “Financial Underdevelopment in the Postbellum Cotton South,” Journal of Interdisciplinary History, 11 (Winter 1981), pp. 443–54.CrossRefGoogle Scholar

³⁸ Dun, R. G. and Company, Mercantile Agency Reference Books (Jan. 1870, 1880).Google Scholar

³⁹ Interior cotton centers are defined in Ransom and Sutch, One Kind of Freedom, pp. 118–19, 300–305.Google Scholar

⁴⁰ As an alternative test of this hypothesis, an analysis of variance, controlling for the population of the store location and the presence of a country courthouse and a post office, demonstrates locations on a rail line had two more stores than those without rail transportation.Google Scholar

⁴¹ The effect of improved terms of trade on self-sufficient households is discussed in Hymer, Stephen and Resnick, Stephen, “A Model of an Agrarian Economy with Non-Agricultural Activities.” American Economic Review, 59 (09 1969), pp. 493–506.Google Scholar

⁴² Rome, [Georgia, ], The Daily Tribune, vol. 3, no. 193 (02 3, 1880).Google Scholar

⁴³ Woodman, Harold D., King Cotton and His Retainers (Lexington, Ky., 1968), pp. 271–72;Google Scholar and Killich, J. R., “The Transformation of Cotton Marketing in the Late Nineteenth Century,” Business History Review, 55 (Summer 1981), pp. 143–69.CrossRefGoogle Scholar

⁴⁴ The Beazley Family, for example, which operated a store in Bartow country during this period, paid their customers approximately 9.25 cents a pound for their cotton between 1887 and 1890; see the Beazley Family, Farm Account Books, 1887–1903 at the Georgia Department of Archives and History. The price of cotton in New Orleans during the same period varied between 9.25 and 11.88 cents a pound.Google Scholar

⁴⁵ Georgia, , Department of Agriculture, Annual Report of the Commissioner (Atlanta, 1877), pp. 10–25, reports the results of numerous experiments made throughout the state to determine the effectiveness of fertilizers in the cultivation of cotton. In a typical example, the application of 200 pounds of phosphate fertilizers per acre increased yields by 60 percent, and over three-quarters of the cotton output on fertilized lands was picked between September 17 and October 26, as compared to only half of the cotton grown on unfertilized land.Google Scholar

⁴⁶ As early as 1847, the chief engineer of the Western and Atlantic Railroad wrote to the president of the Georgia Central: “I have advanced a letter through the papers to the Several Presidents of the Rail Roads between here and the Sea Board on the subject of transporting gypsum at a very low rate at a very low rate for agricultural purpose. I took this step because I thought it the most effectual way to raise the attention of the Farmers to the subject. When you see my letter in the Marietta papers you will see it addressed as much to the farmers as to the Rail Roads. I have long thought of the subject and I have no doubt you all concur with me in the propriety of at least making the experiment to introduce the use of it. I have seen it used myself and I am sure the return crop of the first year will repay us….” Correspondence of the Western and Atlantic Railroad, Letter Book of the Chief Engineer, pp. 130–31 at the Georgia Department of History and Archives. I am indebted to a fellow researcher at the Georgia Department of Archives and History for this reference.For a discussion of the use of fertilizers in antebellum Georgia, see Bonner, History of Georgia Agriculture, pp. 188–90.Google Scholar

⁴⁷ Taylor, Rosser H., “Commercial Fertilizers in South Carolina,” South Atlantic Quartarly, 29 (04 1930), pp. 179–89Google Scholarand “ The Sale and Application of Commercial Fertilizers in the South Atlantic States to 1900,” Agricultural History, 21 (Jan. 1947), pp. 46–52. In 1880, fertilizer manufacturers in Maryland and South Carolina owned 40 percent of the total capital invested in the industry in the United States and produced over a third of the total output of fertilizers;Google Scholar see Roland, William L., “Report on the Manufacture of Chemical Products and Salt,” in U.S. Census Office, Tenth Census, 1880, vol. 2, Report on the Manufactures of the United States (Washington, D. C., 1883), pp. 985–1028. This report also contains a brief history of the fertilizer industry in South Carolina.Google Scholar

⁴⁸ Merchants charged a 19 percent markup on fertilizers bought on credit, as compared to a 40 to 50 percent markup for foodstuffs purchased on time, Department of Agriculture, Annual Report of the Commissioner (Altanta, 1876), pp. 49–52.Google Scholar This differential, no doubt, further encouraged the diffusion of commercial fertilizers. Manufacturers of fertilizers also offered their products for sale at a “cotton option price;” see Georgia, , Department of Agriculture, Annual Report (Atlanta, 1877), appendix.Google Scholar

⁴⁹ The specification of the regression equation is discussed in fn. 26. Although the 1870 Census seriously underestimates the population, wealth and agricultural production of the Southern states, see Ransom and Sutch, One Kind of Freedom, pp. 53–54, the deficiencies of the census were most acute in the counties of the Cotton Belt. The census data, therefore, are assumed to provide more realiable estimates of the actual values for the Upcountry counties.Google Scholar

⁵⁰ As defined, the dependent variable varies with changes in the production of cotton relative to other field crops and with changes in the mix of other field crops, independent of cotton output. The latter effect, however, is numerically small, because corn was by far the most important food crop on Upcountry farms. Changes in the value of other field crops, therefore, reflect changes in corn output, and the dependent variable essentially measures the production of cotton relative to corn. The same results are obtained when the dependent variable is defined as the ratio of the value of cotton to the value of cotton and corn or in 1880 as the share of tilled acreage in cotton. (Compare equation 1 in 1880 in Table 8 with the equation for the Upper Piedmont in the Appendix.) This point was brought to my attention by an anonymous referee. Crop prices are from Ransom and Sutch, One Kind of Freedom, p. 263 (table F.4).Google Scholar

⁵¹ Before 1880, the Census Office reported the size distribution of farms according to improved acreage holdings, while in 1880, it is based on total acreage; U.S. Census Office, Eighth Census, 1860, vol. 3, Agriculture of the United States, p. 196; and U.S. Census Office, Tenth Census, 1880, vol. 3, Report on the Production of Agriculture in the United States, table V.Google Scholar

⁵² A Chow test rejects the hypothesis that the coefficients of the two equations are equal at a 1 percent confidence level.Google Scholar

⁵³ The poor fit of this equation may reflect the deficiencies of the 1870 Census discussed in fn. 49. Alternatively, it may be a consequence of the short-run effects of the war, which would be consistent with the statistical tests for structural change over the period 1860 to 1880.Google Scholar

⁵⁴ In the case of equation (1), the F-statistic with (6,34) degrees of freedom equals 1.06 and is not statistically significant. The same holds for equation (2).Google Scholar

⁵⁵ In equation (1), the F-statistic with (6,34) degrees of freedom equals 6.2, which is significant a 1 percent confidence level; in equation (2), the same result holds.Google Scholar

⁵⁶ For a general discussion of the decline of the factorage system, see Woodman, King Cotton and His Retainers, pp. 254–94; and Ransom and Sutch, One Kind of Freedom, pp. 107–09.Google Scholar

⁵⁷ Wiener, Jonathan M., “Planter-Merchant Conflict in Reconstruction Alabama,” Past and Present, 68 (04 1975), pp. 73–94CrossRefGoogle Scholar and his Social Origins of the New South (Baton Rouge, 1978), part II.Google ScholarThe Georgia legislature made similar revisions in the state's lien law in the 1870s. See Brooks, Agrarian Revolution in Georgia, pp. 32–33; Hahn, Roots of Southern Populism, pp. 174–76; and Flynn, White Land, Black Labor, pp. 87–89.Google Scholar

⁵⁸ This interpretation of Southern railroad construction during the Reconstruction period is advanced by Klein, “The Strategy of Southern Railroads,” pp. 1052–68.Google Scholar For a more detailed examination of railroad construction in Georgia, see McGuire, Peter S., “The Railroads of Georgia, 1860–1880,” Georgia Historical Quarterly, 16, (09 1932), pp. 194–95.Google ScholarDoyle, Don Harrison, “Urban and Southern Culture: Economic Elites in Four New Southern Cities,” in Burton, Orville Vernon and McMath, Robert C., eds., Toward a New South? (Westport, Conn., 1982), pp. 11–36, discusses the efforts of elites in Atlanta to expand their commerce through the promotion of railroads.Google Scholar

⁵⁹ This point is emphasized by Woodman, King Cotton and His Retainers, pp. 269–94. According to Woodman (pp. 84–97), the decline of the factorage system actually began in the 1850s with the revival of itinerant merchants, who purchased cotton in the interior of the South from planters and farmers and consigned the crop to factors in coastal ports. Although these merchants were often financed by cotton factors, they competed with factors for control over the crop. The method of direct purchases and shipments employed after the Civil War depended on the growth of the cotton futures market in New York and New Orleans;Google Scholar see Dumbell, Stanley, “The Origins of Cotton Futures,” Economic History, 1 (1926–1929), pp. 259–67;Google ScholarChandler, Alfred D. Jr, The Visible Hand (Cambridge, Mass., 1977), pp. 209–15;Google Scholar and Lipartito, Kenneth J., “The New York Cotton Exchange and the Development of the Cotton Futures Market,” Business History Review, 57 (Spring 1983), pp. 50–72.CrossRefGoogle Scholar

⁶⁰ Clark, Thomas D., Pills, Petticoats and Plows (Norman, Okla., 1964), chaps. 1, 6;Google ScholarWoodman, King Cotton and His Retainers, pp. 285–87;Google Scholar and Woolfolk, George R., The Cotton Regency (New York, 1958), chap. 7.Google Scholar The growth of wholesale jobbers and the employment of sales agents is discussed more generally in Porter, Glenn and Livesay, Harold C., Merchants and Manufacturers (Baltimore, 1971), chap. 13;Google Scholarand Chandler, The Visible Hand, pp. 215–24.Google Scholar

⁶¹ Curry, Leonard P., Rail Routes South: Louisville's Fight for the Southern Market (Lexington, Ky., 1969);Google Scholar and Klein, Maury, History of the Louisville and Nashville Railroad (New York, 1972), pp. 102–04.Google Scholar

⁶² Woolfolk, The Cotton Regency, chaps. 6–7.Google Scholar

⁶³ Stover, The Railroads of the South, pp. 99–121;Google ScholarWoodward, C. Vann, Origins of the New South, 1877–1913 (Baton Rouge, La., 1971), pp. 292–97;Google Scholarand Klein, “The Strategy of Southern Railroads,” p. 1059.Google Scholar

⁶⁴ Chandler, The Visible Hand, chaps. 4, 5, 7.Google Scholar