The Remedial Investigation/Feasibility Study conducted by the U.S. Environmental Protection Agency within the Spokane River Basin of northern Idaho and eastern Washington included extensive data-collection activities in numerous studies to determine the nature and extent of trace-element contamination within the basin. The objective of this particular study was to improve our understanding of the effects of different river discharges and lake levels of Coeur d'Alene Lake on the transport of cadmium, zinc, and lead within the main stem Coeur d'Alene River. In particular, water-quality data and loads during a broad range of hydrologic conditions were examined to determine if the river channel, flood plain, and associated ground water along the main stem Coeur d'Alene River acted as sources or sinks of trace elements. Water-quality samples were collected at six riverine stations and one lake station along a 35-mile reach during March, June, September, and October of 1999. Samples were analyzed for whole-water recoverable, filtered (0.45 micrometer), and dissolved (0.01 micrometer) concentrations of cadmium, zinc, and lead. Concentrations and loads of cadmium and zinc measured during the four sampling trips were predominately in the filtered and dissolved fraction ,rather than particulate. The smallest concentrations were measured during the June sampling trip when flows were high and snowmelt runoff diluted riverine concentrations. Conversely, the largest concentrations were measured during the latter two sampling trips when flows were low because a larger proportion of the river's discharge was contributed by ground-water inflow. During each sampling trip, cadmium and zinc concentrations generally decreased in a downstream directioeven as discharge increased in a downstream direction. Spatial and temporal trends exhibited by lead concentrations and loads during the four sampling trips were different from those of cadmium and zinc because of the propensity for lead to adsorb to sediment particles. Whole-water recoverable lead concentrations and loads during the four sampling trips were predominantly in the particulate fraction, with filtered and dissolved concentrations and loads composing a much smaller proportion of the recoverable fraction compared to cadmium and zinc. Filtered lead concentrations generally increased at a faster rate in the downstream direction than dissolved lead concentrations; thus, colloidallead either was being formed by complexation reactions or being added by sediment erosion in the downstream direction.