The study obtained data on relevant variables for 122 cities over the 14-year period. Homicide data were obtained from the Supplementary Homicide Report data file compiled by Fox (2000). Data on illicit drug market activity for the same period were taken from a Uniform Crime Report dataset compiled by Chilton and Weber (1999); and data on the social, economic, and demographic characteristics of the cities were drawn from the 1980 Summary Tape File 3C of the U.S. Census Bureau. Race-specific homicide rates were calculated for each city in each year. The key predictor variable for race-specific homicide was a race-specific measure of drug-market activity, i.e., the sale of cocaine/opiate arrest rate for Blacks and Whites. The study found that change in drug-market indicators was positively associated with change in both Black and White homicide rates in large U.S. cities over the period studied. This relationship was substantially stronger for Blacks than for Whites. Also, the socioeconomic moderators of the drug-market/violence link varied by race, with racial inequality being particularly important for Blacks and resource deprivation being especially important for Whites. The findings indicate that the drug-market-homicide connection is much more complex than expected, such that simplistic theories on the drug market-violence nexus require additional study. 2 tables, 24 notes, and 59 references