The effects of mirex on two species of warm-water fishes were studied in three experiments in which the fish were exposed either by feeding a mirex-treated diet, or by treating the holding ponds with a mirex formulation. Bluegills were used in the feeding experiment, where three different levels of mirex were incorporated into the diet and fed to fish held in plastic pools, and in the first pond-exposure experiment, in which the fish were held in earthen ponds treated once with a mirex-corncob grit formulation. The third experiment used goldfish held in earthen ponds which also were treated once with the mirex-corncob grit formulation. In general, higher rates of exposure produced higher whole-body residues of mirex in the fish, and, except in bluegills from the lower-treatment ponds of the first contact experiment and in control fish, whole-body residues increased throughout the terms of these experiments. Soil, water, and vegetation samples from the two contact experiments, although subject to large differences between individual samples, contained relatively unchanging mirex concentrations, illustrating that this chemical is highly resistant to degradation or removal. No mortality or tissue pathology in the bluegills could be ascribed to mirex exposure, but the gills and kidneys of mirex-exposed goldfish showed reactions beginning with the samples taken 56 days after treatment, and the numbers of these fish surviving until termination of this experiment were inversely related to treatment level. Total serum protein and hematocrit percentages had no apparent relationship to treatment levels, nor was mirex exposure a demonstrable factor in growth rates in the contact experiments. However, growth of the bluegills in the highest treatment groups of the feeding experiment was adversely affected. Invertebrate populations seemed not to be affected by the mirex treatment in either of the earthen pond experiments.