A hydroelectric lake on the Oconee River in central Georgia was sampled on a quarterly basis for four years to determine the impact of three insecticides on the pesticide loading of fish. The insecticides carbofuran (2,3,‐dihydro‐2,2‐dimethyl‐7‐benzofuranyl methyl‐carbamate), fenvalerate [cyano(3‐phenoxyphenyl)methyl 4‐chloro alpha (1‐methylethyl)benzeneacetate], and azinphosmethyl [0,0‐Dimethyl S‐(4‐oxo‐1,2,3‐benzotriazin‐3‐(4H)‐ylmethyl phosphorodithioate] were used to control cone and seed insects on a pine seed orchard that is adjacent to Lake Oconee. Fish populations in areas of the lake that receive runoff from treated and untreated portions of the seed orchard were sampled using gill nets. Water and fish samples were collected monthly in 1981–1983 and quarterly in 1984. Species collected included carp (Cyprinus carpio), sucker (Catostomus spp., Erimyzon spp., Moxostoma spp., Minytrema spp.), bass (Micropterus spp.), channel catfish (Ictalurus punctatus), spotted gar (Lepisosteus spp.), and bluegill sunfish (Lepomis macrochirus). Few carbofuran residues (< 7 percent of fish samples contained detectable residues) and no azinphosmethyl or fenvalerate residues were detected despite measured inputs into the lake after insecticide application. Persistent agricultural insecticides including dieldrin, lindane, DDT, and toxaphene were detected in 32, 36, 58, and 83 percent of sampled fish, respectively. Screening for 20 other insecticides resulted in no detection of residues at a detection limit of 30–100 μg kg‐1.