We examined how harvest affected natural mortality rates and how the dynamics of northern bobwhite (Colinus virginianus, hereafter bobwhite) covey membership might affect the additive nature of harvest. We conducted repeated searches for coveys and radiomarked all coveys found on 12 259-ha study areas in east-central Kansas, USA. We simulated a harvest in 6 randomly selected areas each year during November–January, 1997–2000, by trapping and removing 60% of bobwhites in each covey. We used flush counts and radiotelemetry to measure harvest effects on natural mortality rates, overall winter survival, covey size, and densities. Observed natural mortality rates of radiomarked individuals were similar between harvested (50.6 ± 4.3%) and unharvested areas (52.1 ± 4.7%). Estimated winter survival was 47.9% on the unharvested areas and 20.9% on the harvested areas. Harvest did not affect the number of coveys, average covey size, or density of bobwhites, indicating that coveys coalesced through local movement. Our results indicate that harvest is additive to natural mortality and suggest that local movement to maintain optimal group sizes can mask the true effect of harvest on observed densities in small areas.