The Black-White gap in completed schooling among Southern born men narrowed sharply between the World Wars after being stagnant from 1880 to 1910. We examine a large scale school construction project, the Rosenwald Rural Schools Initiative, which was designed to dramatically improve the educational opportunities for Southern rural Blacks. From 1914 to 1931, nearly 5,000 school buildings were constructed, serving approximately 36 percent of the Black rural school-aged Southern population. We use historical Census data and World War II enlistment records to analyze the effects of the program on school attendance, literacy, high school completion, years of schooling, earnings, hourly wages, and migration. We find that the Rosenwald program accounts for at least 30 percent of the sizable educational gains of Blacks during the 1910s and 1920s. We also use data from the Army General Classification Test (AGCT), a precursor to the AFQT, and find that access to Rosenwald schools increased average Black scores by about 0.25 standard deviations adding to the existing literature showing that interventions can reduce the racial gap in cognitive skill. In the longer run, exposure to the schools raised the wages of blacks that remained in the South relative to Southern whites by about 35 percent. For blacks the private rate of return to a year of additional schooling induced by Rosenwald was about 18 percent. Moreover, Rosenwald significantly increased Northern migration of young adult Blacks, with no corresponding impact on schoolage Blacks or young adult Whites, likely fueling further income gains. Across all outcomes, the improvements were highest in counties with the lowest levels of Black school attendance suggesting that schooling treatments can have a very large impact among those at the bottom of the skill distribution.