We studied the short-term response of wildlife to clear-cutting in four blocks (52-114 km(2)) that were logged in 100- to 250-ha clustered patches. Surveys were conducted, 2 years before and 2 years after logging, to determine the relative abundance of 12 wildlife species, and telemetry data were also gathered on four species. Small mammals, species with the smallest home ranges (less than or equal to 1 ha), either remained in the clearcut patches or had replacement habitat in the buffer strips. Most species with home ranges up to 25 ha (spruce grouse, Falcipennis canadensis Linne; snowshoe hare, Lepus americanus Erxleben) were excluded from clearcut patches. Species with home ranges greater than or equal to 5 km(2) (marten, Martes americana Turton; moose, Alces alces Linne) remained in some residual forest patches scattered throughout clearcuts and in the adjacent uncut forest. In their home range, these two species avoided clearcut patches where the shrub layer and coniferous regeneration were scattered. Because many wildlife species depend on residual forest, the important issue is not the size of clearcut patches but the extent and configuration of the remaining forest. Instead of a clustered distribution of clearcut patches, we propose two harvest scenarios more compatible with integrated wildlife-forest management objectives on a local scale.