¹ Gallman, Robert, “Trends in the Size Distribution of Wealth in the Nineteenth Century: Some Speculations” in Lee, Soltow, editor, Six Papers on the Size Distribution ‘of Wealth and Income, Studies in Income and Wealth, Vol. XXXIII (New York: National Bureau of Economic Research, 1969Google Scholar); Lee Soltow, Patterns of Wealthholding in Wisconsin Since 1850 (to be published by the University of Wisconsin Press).

² A study was made of the relationship between personal estate and number of slaves for 1,105 counties in the South in 1860. An almost perfect relationship was found to exist between these two variables. Counties averaging one more slave per adult free male had $900 more personal estate per adult free male, as determined from the slope of a linear fit. Sources are Census Office, 8th Census, 1860, Population of the United States in 1860 (1864), pp. 2–592; and Statistics of the United States in 1860 (1866), Table 3, pp. 296–319:(this includes mortality, property, etc.).

³ A sample of size 13,696 was drawn by the author from the microfilm of the manuscripts of the 1860 Census. This sample is further described in Table 3.

⁴ The author has worked with slave totals in some Mississippi counties in 1860. It is difficult to combine the number of slaves of more than one plantation owned by one owner.

⁵ Gavin Wright has studied farms in the cotton South in 1850 and 1860. He found that the top 5 percent of farms had 39.0 percent of slaves in 1850 and 38.9 percent in 1860. His index of concentration of the distribution of slaves among farms was .723 in 1850 and .747 in 1860. See “‘Economic Democracy’ and the Concentration of Agricultural Wealth in the Cotton South, 1850–1860,” Agricultural History, XUV (Jan. 1970) 63–93Google Scholar.

⁶ The 14 states and the District of Columbia include Alabama, Arkansas, Delaware, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, Missouri, North Carolina, South Carolina, Tennessee, and Virginia. Distributions for 1850 and 1860 will include in the South the state of Texas.

⁷ The microfilm was placed on the reader and a mark was placed on the screen. A certain number of cranks or spins of the film was made, with the families being recorded if their names fell on the screen mark. Every second film was read only on backward spins. The number with fifteen or more slaves was twenty-two times as large in the special sampling of rich as it was in the regular sample.

⁸ The chi-square test, using 1850 frequencies as theoretical frequencies and 1830 sample frequencies as observed frequencies, gives a value from a test of goodness of fit which is at most twice that given in a chi-square table at the 5 percent level of significance.

⁹ Bureau of the Census, A Century of Population Growth from the First Census of the United States to the Twelfth, 1790–1900 (Washington, 1909), pp. 134Google Scholar, 136.

¹⁰ Census Office, 10th Census, 1880, Compendium of the Tenth Census (1883), p. 650. The average number of free persons per free family was 5.53 in 1790 and 5.47 in 1860.

¹¹ These figures are derived from samples I took of these censuses. See Tables 3 and 7 for more details,

¹² The sample for 1860 shows that only 14 percent of adult free males in the South were foreign born.