We determined survival rates, causes of mortality, and documented impacts of harvest on ≥1.5-year-old male (hereafter, male) Columbian black-tailed deer (Odocoileus hemionus columbianus) in 2 Washington, USA, game management units (GMUs; Skookumchuck and Snoqualmie) characterized by different hunting-season structures. We monitored 66 males (n = 28 and 38 annually) in Skookumchuck and 58 males (n = 26 and 32 annually) in Snoqualmie, September 1999–September 2001. Annual survival rates were 0.498 (SE = 0.066) in Skookumchuck and 0.519 (SE = 0.067) in Snoqualmie. Survival rates derived from population age structure did not differ from rates derived from radiotelemetry. Harvest was the primary mortality factor for each population, accounting for 67% (SE = 7; Skookumchuck) to 44% (SE = 9; Snoqualmie) of total annual mortality. Annual harvest-specific mortality rates were 0.317 (SE = 0.032) in Skookumchuck and 0.211 (SE = 0.021) in Snoqualmie, likely due to longer hunting seasons and greater hunter effort in Skookumchuck. Following the elimination of a late buck season centered on the rut in Snoqualmie, male harvest declined 56% and annual survival increased 60%, indicating that male harvest was largely additive to other mortality. Our results indicated that harvest was the primary influence on male black-tailed deer populations in Washington, was additive, and that the effect of harvest varied with hunting-season structure and hunter effort. Managers should not assume that harvesting removes a constant proportion of the male population annually, and management models that assume compensatory mortality in adult harvest may result in overharvest of male populations.