Background: Surgical-treatment outcomes for melanoma in African Americans are poorly characterized as a result of low incidence of melanoma among African Americans. We examined differences by race in overall and melanoma-specific survival, stratified by receipt of surgical treatment and by specific types of surgical treatment. Methods: Data from the 1973-2004 public-use Surveillance, Epidemiology and End Results Program (SEER) were analyzed by Cox proportional hazard models to compare the effects of surgical treatments on overall and melanoma-specific survival in blacks, whites, and other race, controlling for confounding demographic and tumor-related variables. Results: Of 151,154 patients with first primary melanoma (148,883 whites, 789 blacks and 1,532 other race), 142,653 (94.4%) received surgical treatment. Among patients who received surgical treatment, 10-year melanoma-specific survival was lower in blacks (73%) than in whites (88%) and other race (85%); black patients were at significantly higher risk of overall and melanoma-specific mortality when compared with white (hazard ratio [HR] = 1.64, 95% confidence interval [CI] 1.44-1.86, P < 0.0001 and HR = 1.50, 95% CI 1.25-1.79, P < 0.0001, respectively) and with other race (HR = 1.55, 95% CI 1.31-1.85, P < 0.0001 and HR = 1.49, 95% CI 1.16-1.91, P = 0.0017, respectively). Blacks who underwent biopsy, wide excision and surgery not otherwise specified were at higher risk of overall mortality compared with whites with the same treatment. Conclusion: Overall and melanoma-specific survival was lower in blacks undergoing surgical treatment for melanoma compared to both whites and other race. Reasons for these disparities remain poorly understood.