The quality of water along flowpaths in a surficial aquifer system in Manchester, Connecticut, was studied during 1993-95 as part of the National Water Quality Assessment program. The flowpath study examined the relations among hydrogeology, land-use patterns, and the presence of contaminants in a surficial aquifer in an urban area, and evaluated ground water as a source of contamination to surface water. A two-dimensional, finite-difference groundwater- flow model was used to estimate travel distance, which ranged from about 50 to 11,000 feet, from the source areas to the sampled observation wells. Land use, land cover, and population density were determined in the source areas delineated by the ground-water-flow simulation. Source areas to the wells contained either high- or medium-density residential areas, and population density ranged from 629 to 8,895 people per square mile. Concentrations of selected inorganic constituents, including sodium, chloride, and nitrite plus nitrate nitrogen, were higher in the flowpath study wells than in wells in undeveloped areas with similar aquifer materials. One or more of 9 volatile organic compounds were detected at 12 of 14 wells. The three most commonly detected volatile organic compounds were chloroform, methyl-tert-butyl ether, and trichloroethene. Trichloroethene was detected at concentrations greater than the maximum contaminant level for drinking water (5 micrograms per liter) in samples from one well. Four pesticides, including dichloro diphenyl dichloroethylene, dieldrin, dichloroprop, and simazine were detected at low concentrations. Concentrations of sodium and chloride were higher in samples collected from wells screened in the top of the saturated zone than in samples collected from deeper zones. Volatile organic compounds and elevated concentrations of nitrite plus nitrate as nitrogen were detected at depths of as much as 60 feet below the water table, indicating that the effects of human activities on the ground-water quality extends to the bottom of the surficial aquifer. The age of ground water, as determined by tritium and 3helium concentrations, was 0.9 to 22.6 years. pH, alkalinity, and calcium were higher and concentrations of dissolved oxygen were lower in ground-water samples with ages of 10 years or more than in samples younger than 10 years. In addition, concentrations of sodium, chloride, and nitrite plus nitrate nitrogen were low in ground-water samples with ages of 10 years or more, indicating that concentrations of these compounds may be increasing with time or that the recharge areas to these wells may have had less intensive urban land use. Methyl-tert-butyl ether was detected only in wells with ground water ages of less than 11 years, which is consistent with the date of introduction of this compound as a gasoline additive in Connecticut. Analysis of additional samples collected for analysis of stable nitrogen isotopes indicated that the most likely source of elevated concentrations of nitrate nitrogen was lawn and garden fertilizers, but other sources, including wastewater effluents, soil organic nitrogen, and atmospheric deposition, may contribute to the total. Population density was positively correlated (at the 97 percent confidence level) to concentrations of nitrite plus nitrate as nitrogen. Water quality in the Hockanum River aquifer has been degraded by human activities, and, after discharge to surface water, affects the water quality in the Hockanum River. On an annual basis, ground-water discharge from the study area to the river (as measured at a downstream continuous-record gaging station) contributes about 5 percent of the annual load of nitrite plus nitrate nitrogen, but, during low flow, contributes 11 percent of the nitrite plus nitrate nitrogen, 32 percent of the calcium, and 16 percent of the chloride to the river.