¹ Fogel, Robert W. reviews much of the debate in “Nutrition and the Decline in Mortality since 1700: Some Preliminary Findings,” in Engerman, Stanley L. and Galiman, Robert E., eds., Long-Term Factors in American Economic Growth (Chicago, 1987), pp. 439–555.Google Scholar

² For a discussion of procedures and characteristics of the sample see Steckel, Richard H., “Census Matching and Migration: A Research Strategy,” Historical Methods, 21 (forthcoming, 1988). Note that sampling procedures excluded families in which the head or all children died. Plausible conjectures about mortality rates for heads, mortality rates for children whose father died (based on implications of nonsurvival of the wife reported in fn. 15), the distribution of the number of children ever born, and mortality rates of all children suggest that exclusion of these families probably biased children's nonsurvival rates downward by less than 9 percent.CrossRefGoogle Scholar

³ According to census data and divorce figures in Wright, Carroll D., Marriage and Divorce in the United States, 1867 to 1886 (Washington, D.C., 1897) the ratio of divorces reported from 1870 to 1879 to the average female population aged 20 and above from 1870 to 1880 was about 1.3 percent. Of course, not everyone aged 20 and above was married, but even a doubling of the figure of 1.3 percent indicates that divorce probably comprised a small part of the total separations depicted for women in Table 1. Initials in place of given names and garbled names created ambiguities in determining survival of about 6 percent of the women and 11 percent of the children. I exclude these ambiguous cases, but the results are unaffected by their inclusion in the data base.Google Scholar

⁴ The conditions under which children left home are under investigation.Google Scholar

⁵ Haines, Michael R. and Avery, Roger C., “The American Life Table of 1830–1860: An Evaluation,” Journal of Interdisciplinary History, 11 (Summer 1980), p. 88.CrossRefGoogle ScholarPubMed

⁶ Dublin, Louis I., Health Progress, 1936–1945 (New York, 1948), p. 26, p. 78.Google Scholar

⁷ Antiseptic precautions during delivery substantially improved maternal health. See Dublin, Louis I., Lotka, Alfred J., and Spiegelman, Mortimer, Length of Life: A Study of the Life Table (New York, 1949), pp. 154–55.Google Scholar

⁸ Calculations proceeded from assumptions that deaths were evenly distributed across ages beyond infancy and that one-half of infant deaths occurred during the first month after birth.Google Scholar

⁹ See, for example, Haines and Avery, “The American Life Table”.Google Scholar

¹⁰ The categories follow those in appendix B of Thernstrom, Stephan, The Other Bostonians (Cambridge, Mass., 1973). Given the imprecision of occupational listings and the debate over appropriate groupings, the results give only a general impression of survival by occupation. See Steckel, “Census Matching,” for additional discussion of these categories.CrossRefGoogle Scholar

¹¹ Unfortunately, the number of blacks in the sample was too small for statistical analysis.Google Scholar

¹² In 1870 the South had 14.2 percent of the reported divorces (according to Wright, Marriage and Divorce) but 30.7 percent of the total population, which discredits the idea that divorce explained the high losses for the South. If the sample is divided by age into groups below age 45 and ages 45 and above, systematic regional differences emerged among the younger but not the older group, which may reflect contamination from sample selectivity at older ages.Google Scholar

¹³ See, for example, Buley, R. Carlyle, The Old Northwest Pioneer Period, 1815–1840, 2 vols. (Bloomington, 1950), chaps. 4, 5.Google Scholar

¹⁴ Following this line of thought, experiments were conducted with variables for subregions within the South, using as categories the coastal states from North Carolina to Louisiana and all states and marginally significant in the coastal states (expected loss = 18.7 percent, t = 1.94) compared with the Northeast (expected loss = 13.3 percent). Among women expected losses were 15.2 percent in the coastal states(t = 2.61), 13.1 percent in the other southern states(t = 1.77), and 8.7 percent in the Northeast. Lower mortality rates in the interior states of the South are consistent with relatively greater stature found among recruits from this region; see Margo, Robert A. and Steckel, Richard H., “The Heights of American Slaves: New Evidence on Slave Nutrition and Health,” Social Science History, 6 (Fall 1982), pp. 516–38.CrossRefGoogle ScholarPubMed

¹⁵ Although it is desirable to use values of the independent variables that prevailed in 1850 (at the outset of the period of observation on survival), because real estate is an incomplete measure of wealth and because a reasonably high positive correlation may have existed for total wealth in the two census years, I conducted experiments using the total wealth measure available for 1860 (real and personal estate) as an explanatory variable. The results, however, were similar to those using the value of real estate in 1850. In addition, the findings were essentially unchanged after adjusting wealth values for regional differences in prices as reported in Coelho, Philip and Shepherd, James, “Differences in Regional Prices: The United States, 1851–1880,” this JOURNAL, 34 (09 1974), 551–91;Google Scholar and David, Paul A. and Solar, Peter, “A Bicentenary Contribution to the History of the Cost of Living in America,” Research in Economic History, 2 (1977), pp. 1–80. Unfortunately, the dependent variable does not distinguish infant mortality from other childhood losses. The finding that occupation may have influenced the survival of those aged 0 but not children who began the decade of the 1850s at older ages suggests that infant mortality alone may have been sensitive to economic conditions. If a dummy variable for the wife's nonsurvival is included in the regressions given in Table 2, then her departure increased the expected loss by approximately 43 percent (from. 189 to. 270 at age 0, t = 1.13, and from. 143 to. 205 at ages 1–4, t = 2.27). The excess loss overstates mortality alone, however, because divorce caused some departures and some children were placed in foster homes if the mother died.Google Scholar

¹⁶ Although the real estate values reported by the census exclude debt, mortgages were typically 1 to 3 years and had downpayments of 50 to 75 percent. Thus the census data approximate net worth in real estate. For additional discussion of the wealth data see Richard H. Steckel, “Poverty and Prosperity: A Longitudinal Study of Wealth Accumulation, 1850–1860” (unpublished manuscript, Ohio State University, 1988).Google Scholar

¹⁷ Higgs, Robert, “Cycles and Trends of Mortality in 18 Large American Cities, 1871–1900,” Explorations in Economic History, 16 (10 1979), pp. 381–408.CrossRefGoogle ScholarPubMed

¹⁸ Long-distance migrants of this era sometimes mentioned illness and mortality that accompanied travel and adjustment to a new environment. See Cassedy, James H., Medicine and American Growth, 1800–1860 (Madison, 1986), pp. 72–77.Google Scholar

¹⁹ United Nations, The Determinants and Consequences of Population Trends (New York, 1973), p. 115.Google Scholar

²⁰ Lindert, Peter H. discusses the debate in “English Living Standards, Population Growth, and Wrigley-Schofield,” Explorations in Economic History, 20 (04 1983), pp. 131–55.CrossRefGoogle Scholar

²¹ Sokoloff, Kenneth L. and Villafior, Georgia C., “The Early Achievement of Modern Stature in America,” Social Science History, 6 (Fall 1982), pp. 453–81;CrossRefGoogle ScholarPubMedMargo, Robert A. and Steckel, Richard H., “Heights of Native-Born Whites during the Antebellum Period,” this JOURNAL, 43 (03 1983), pp. 167–74;Google ScholarPubMedSteckel, Richard H., “Growth Depression and Recovery: The Remarkable Case of American Slaves,” Annals of Human Biology, 14 (03–04 1987), pp. 111–32.CrossRefGoogle ScholarPubMed

²² See, for example, Vinovskis, Mans A., “Mortality Rates and Trends in Massachusetts before 1860,” this JOURNAL, 32 (03 1972), pp. 184–213.Google ScholarPubMed

²³ Steckel, Richard H., “A Dreadful Childhood: The Excess Mortality of American Slaves,” Social Science History, 10 (Winter 1986), pp. 427–65.CrossRefGoogle ScholarPubMed

²⁴ See, for example, Yasuba, Yasukichi, Birth Rates of the White Population in the United States, 1800–1860 (Baltimore, 1962).Google Scholar

²⁵ Assuming no regional differences in infant survival, the expected probabilities of nonsurvival calculated from the regressions suggest that the ratio of children under 5 to women aged 15 to 49 was biased upward by approximately 5 percent. This bias is relatively unimportant, however, because the child-woman ratio of the West North Central and the West South Central states in 1850 exceeded the national average by more than 60 percent and that of New England by more than 80 percent.Google Scholar

²⁶ McClelland, Peter D. and Zeckhauser, Richard J., Demographic Dimensions of the New Republic (Cambridge, 1982).Google Scholar

²⁷ See, for example, Jones, Alice Hanson, Wealth of a Nation to Be (New York, 1980).Google Scholar