Published online by Cambridge University Press: 04 January 2016
Despite the importance of marriage for the economic and demographic history ofthe nineteenth-century United States, there are few published estimates of thetiming and incidence of marriage and no published studies of its correlatesbefore 1890, when the Census Office first tabulated marital status by age, sex,and nativity. In this article I rely on the 1860 Integrated Public Use MicrodataSeries census sample to construct national and regional estimates of whitenuptiality by nativity and sex and to test theories of marriage timing. Isupplement this analysis with two new public use samples of Civil War soldiers.The Gould sample, collected by the U.S. Sanitary Commission between 1863 and1865, allows me to test whether height and body mass influenced white men'spropensity to marry. Additionally, a sample of Union Army recruits linked to the1860 census, created as part of the Early Indicators of Later Work Levels,Disease, and Death project, allows me to combine suspected economic,demographic, and anthropometric correlates of marriage into a multivariate modelof never-married white men's entrance into first marriage. The results indicatethat nuptiality was moderately higher in 1860 than it was in 1890. In contrastto previous studies that emphasize the primary importance of land availabilityand farm prices, I find that single women's opportunity to participate in thepaid labor force was the most important determinant of marriage timing. I alsofind modest support for the hypothesis that height affected men's propensity tomarry, consistent with the theory that body size was a sign to potentialmarriage partners of future earnings capacity and health.