As a women's and an economic historian of seventeenth- and eighteenthcentury Britain, I welcome further analysis and reassessment, such as Joyce Burnette undertakes, of women, work, and wages during the industrial transition. This topic has fallen out of fashion, and yet there is still much that we do not know. I enjoyed reading Burnette's (2008) book, which challenged and engaged me and made me reflect on the approaches, methods, and evidence that I, as a historian, am persuaded by when it comes to making claims about women and work in the past.

Explanations of the gender pay gap and related labor market segregation remain fiercely debated. On the one side are those economic historians who take a primarily neoclassical view, in which competition among workers and employers eliminates wage differences that do not reflect productivity and occupational segregation that is not the outcome of choice. Persistent discrimination must reflect anticompetitive institutions, for instance, trade unions. A corollary of the neoclassical perspective is that markets are liberating, freeing agents, including women, from cultural stereotypes and ensuring that they get paid what they are worth, although of course this need not imply wage equality if there are gender differences in productivity. On the other side are those cultural historians who interpret wage differences as reflecting custom and, as far as women are concerned, the cultural deprecation of women's work, while occupational segregation represents gender stereotypes of fit work for women. In this view, socially and culturally constructed gender identities can influence market outcomes, producing discrimination in wages and work that persists even in the face of competitive forces.

This article investigates patterns of socioeconomic difference in the wartime morbidity and mortality of black Union soldiers during the American Civil War. Among the factors that contributed to lower probabilities of contracting and dying of disease were lighter skin color, a nonfield occupation, residence on a large plantation, and residence in a rural area prior to enlistment. Patterns of disease-specific mortality and timing of death suggest that the differences in the development of immunity to disease and in nutritional status prior to enlistment were responsible for the observed socioeconomic differences in wartime health. For example, the advantages of light-skinned soldiers over dark-skinned and of enlisted men formerly engaged in nonfield occupations over field hands resulted from differences in nutritional status. The lower wartime mortality of ex-slaves from large plantations can be explained by their better-developed immunity and superior nutritional status. The results of this article suggest that there were substantial disparities in the health of the slave population on the eve of the Civil War.