BACKGROUND Morning blood pressure surge (MS) is accepted to increase cardiovascular risk, but it is not clear how it should be defined. Because dipping on 24-h ambulatory blood pressure monitoring (ABPM) associates with improved outcome we hypothesized that MS will not predict mortality independent of dipping. METHODS We investigated a cohort of 2,627 patients referred for ABPM with available values for at least 1-h after awakening, and related them to all-cause mortality. RESULTS During 22,353 person-years of follow-up, 246 patients died. We used Cox proportional hazards models to explore mortality associated with different definitions of the MS. Only the "1-h" MS (difference between average blood pressure (BP) 1-h before and after awakening) was related to mortality: after multiple adjustments including 24-h systolic BP, those whose "1-h" MS was above median (> 12 mm Hg) had mortality hazard ratio (HR) of 0.61, 95% CI: 0.47-0.79, P < 0.001. In predetermined subgroup analyses, nondippers (n = 1,039), had a highly significant MS-related decrease in mortality: HR 0.49, 95% CI: 0.34-0.73, P < 0.001, unlike dippers (n = 1,588), HR = 0.90,95% CI: 0.60-1.34. CONCLUSION Among nondipping subjects referred for ABPM, the MS is associated with decreased mortality.