Human activities associated with agricultural, residential, commercial, and industrial land uses have affected the quality of water in the four stratified-drift aquifers examined in Connecticut. A study to evaluate quantitatively the effects of human activities, expressed as land use, on regional ground-water quality was initiated in 1984 as part of the U.S. Geological Survey's Toxic Waste-round-Water Contamination Program. Water-quality data were collected from 116 shallow stainless-steel wells installed beneath or immediately downgradient from seven types of land use areas within the Pootatuck, Pomperaug, Farmington, and Hockanum River valleys in Connecticut. Analysis of variance on the ranked concentrations of 21 largely uncensored or slightly censored constituents, and contingency-table analysis of the frequency of detection of 49 moderately to highly censored constituents indicate that 27 water-quality variables differ at the 0.05 level of significance for samples from at least one land use area. For most constituents, concentrations or detection frequencies are lowest in samples from the undeveloped areas, which characterize background water-quality conditions. The effect of agricultural land use on groundwater quality reflects tillage practices; tilled areas affect the water quality to a greater degree than do untilled areas. Twenty percent of the wells in the tilled agricultural areas yielded water with concentrations of nitrate plus nitrite-nitrogen exceeding 10 milligrams per liter. Atrazine detections in one-third of the wells in areas of tilled agricultural land use were significantly more common than in the undeveloped areas. Ground-water quality beneath sewered residential areas is more severely affected by inorganic and organic nonpoint-source contaminants than is water quality beneath unsewered residential areas. Median concentrations or detection frequencies of most physical properties and inorganic constituents of ground water are higher in sewered than in unsewered residential areas. Generally low concentrations (less than 1.0 microgram per liter) of one or more of 17 volatile organic compounds were detected in samples from 62 percent of the wells in the unsewered residential areas. Most of these compounds were detected in less than 10 percent of the ground-water samples from the unsewered residential areas, however, and consequently, their frequency of detections was not significantly different than in samples from other land use areas. The detection of chloroform in ground-water samples from 47 percent of the wells in the sewered residential areas is significantly higher than the frequency of detection of chloroform in samples from the undeveloped, tilled agricultural, and unsewered residential areas. The quality of ground water is adversely affected beneath commercial areas more so than beneath all other land use areas. Median concentrations of sodium (22.5 milligrams per liter), chloride (36 milligrams per liter), and dissolved solids (286 milligrams per liter) are highest in ground-water samples in commercial areas. Detections of tetrachloroethylene, trichloroethylene, and 1,2-transdichloroethylene were significantly more common in ground-water samples from the commercial areas than in samples from one or more of the other land use areas. Tetrachloroethylene was detected in water samples from 50 percent of the observation wells in the commercial areas at concentrations of up to 1,300 micrograms per liter. Trichloroethylene and 1,2-transdichloroethylene were found at concentrations of up to 20 and 55 micrograms per liter, respectively, in samples from more than 40 percent of the wells in the commercial areas. Although industrial areas occupy only a small part of each of the study areas, they have a disproportionately large effect on ground-water quality. One or more of 12 volatile organic compounds were detected in water samples from 91 percent of the observation wells in the industrial areas