The effects of forest-management activities (timber cutting and herbicide application) on runoff components (total streamflow, direct runoff, and base flow) and on ground-water recharge per unit area were evaluated for two separate paired drainage basins of Quabbin Reservoir in central Massachusetts. The Cadwell Creek study area, studied from 1962-1973, included an experimental basin (Upper Cadwell Creek) and a control basin (Lower Cadwell Creek). In the experimental basin, herbicide was applied to mixed oaks, northern hardwoods, and understory vegetation in different riparian zones during the summers of 1967 and 1968, and some pine plantations were thinned or clear-cut during the winter of 1967-1968. These forest-management activities decreased the total basal area by about 34%. The decrease in total basal area resulted in an increase in total streamflow, direct runoff (total streamflow minus base flow), and ground-water recharge for six dormant seasons (October-April) and six growing seasons (May-September) during 1968-1973. Base flow increased for three dormant seasons and two growing seasons during 1968-1970 and the dormant seasons of 1971 and 1973. Base flow accounted for 34% and direct runoff accounted for 66% of the 94 mm (15%) increase in total streamflow during water years 1968-1973. Sixty-one percent of this increase in total streamflow occurred in the dormant seasons. The Dickey Brook study area, studied from 1985-1989, included an experimental basin (Dickey Brook) and a control basin (Dickey Brook Tributary). Some pine plantations were thinned or clear-cut in the headwaters of the experimental basin from October 1986 to March 1987 and October to December 1988. These forest-management activities decreased the total basal area by 24% during 1986-1987 and an additional 8% during 1988. The decrease in total basal area resulted in an increase in total streamflow, base flow, and ground-water recharge for only one dormant season and one growing season in 1987. Direct runoff only increased slightly during the 1987 dormant season. Base flow accounted for 91% and direct runoff accounted for 9% of the 92 mm (21%) increase in total streamflow during water year 1987. Seventy-seven percent of this increase in total streamflow occurred in the dormant season.