The Triangle Area Water Supply Monitoring Project was formed by a consortium of local governments and governmental agencies in cooperation with the U.S. Geological Survey to supplement existing data on conventional pollutants, nutrients, and metals to enable eventual determination of long-term trends; to examine spatial differences among water supplies within the region, especially differences between smaller upland sources, large multipurpose reservoirs, and run-of-river supplies; to provide tributary loading inlake data for predictive modeling of Falls of the Neuse and B. Everett Jordan reservoirs; and to establish a database for synthetic organic compounds. Water-quality sampling began in October 1988 at 35 sites located on area run-of-river and reservoir water supplies and their tributaries. Sampling has continued through 1994. Samples were analyzed for major ions, nutrients, trace metals, pesticides, and semivolatile and volatile organic compounds. Monthly concentration data, high-flow concentration data, and data on daily mean streamflow at most stream sites were used to calculate loadings of nitrogen, phosphorus, suspended sediment, and trace metals to reservoirs. Stream and lake sites were assigned to one of five site categories-- (1) rivers, (2) large multipurpose reservoirs, (3) small water-supply reservoirs, (4) streams below urban areas and wastewater-treatment plants, and (5) headwater streams--according to general site characteristics. Concentrations of nitrogen species, phosphorus species, and selected trace metals were compared by site category using nonparametric analysis of variance techniques and qualitatively (trace metals). Wastewater-treatment plant effluents and urban runoff had a significant impact on water quality compared to reservoirs and headwater streams. Streams draining these areas had more mineralized water than streams draining undeveloped areas. Moreover, median nitrogen and nitrite plus nitrate concentrations were significantly greater than all other site categories. Phosphorus was significantly greater than for reservoir sites or headwater streams. Few concentrations of trace metals were greater than the minimum reporting limit, and U.S. Environmental Protection Agency drinking-water standards were rarely exceeded. Detections, when they occurred, were most frequent for sites below urban areas and wastewater-treatment plant effluents. A small number of samples for analysis of acetanilide, triazine, carbamate, and chlorophenoxy acid pesticides indicate that some of these compounds are generally present in area waters in small concentrations. Organochlorine and organophosphorus pesticides are ubiquitous in the study area in very small concentrations. Trihalomethanes were detected at sites below urban areas and wastewater-treatment plants. Otherwise, volatile organic compounds and semivolatile compounds were generally not detected. Suspended-sediment, nitrogen, phosphorus, lead, and zinc loads into Falls Lake, Jordan Lake, University Lake, Cane Creek Reservoir, Little River Reservoir, and Lake Michie were calculated. In general, reservoirs act as traps for suspended sediment and constituents associated with suspended sediments. During 1989-94, annual suspended-sediment load to Falls Lake ranged from 29,500 to 88,200 tons. Because Lake Michie trapped from 83 to 93 percent of the suspended sediment delivered by Flat River, Flat River is a minor contributor of suspended sediment to Falls Lake. Yields of suspended sediment from Little River, Little Lick Creek, and Flat River Basins were between 184 and 223 tons per square mile and appear to have increased increased slightly from yields reported in a study for the period 1970-79. Annual suspended-sediment load to Jordan Lake ranged from 271,000 to 622,000 tons from 1989 through 1994 water years. The Haw River contributed more than 75 percent of the tota load to Jordan Lake. The suspended-sediment yields for Haw River and Northeast Cree