We use a life course framework to analyze lifetime patterns of mortality among black Americans. Using this framework directs attention to specific questions regarding the potential causes of racial group differentials in mortality, and we hope moves the field toward more comprehensive and testable explanations. The work on aging, the life course, and health has long highlighted the racial crossover effect in late-life mortality (e.g., Johnson 2000). While there are heated debates about the causes of this racial crossover in the United States (e.g., Johnson 2000; Preston et al. 1996), demographers have noted its existence in both cross-sectional population-level data, and in longitudinal panel studies (Johnson 2000). Gibson (Gibson 1991, 1994; Gibson and Jackson 1987) speculated that the racial crossover is based upon a series of mortality sweeps beginning in the black population in midlife, thereby leaving a hardier group of blacks in very older ages whose probability of survival in comparison to whites' reverses and becomes more favorable.