A procedure for recording the electroretinogram (ERG) in mice with a coiled stainless steel-type electrode was developed in order to examine retinal toxicity. Mice received a single i.v. of sodium iodate (SI), a retinotoxic compound, via the tail vein at a dose of 12.5, 25, or 50 mg/kg, and the ERG was recorded periodically for 28 days after dosing. In addition, the retina was examined histopathologically on day 30 after dosing. 1. The mice were anesthetized with mixed anesthetics of urethane, xylazine and ketamine after 30 to 60 min of dark-adaptation. Sixteen responses to repetitive 1.2 J light stimuli at a frequency of 0.2 or 0.1 Hz were averaged by a microcomputer. Body temperature of the mice was kept constant at 37 to 38 degrees C using a thermostatically controlled heating mat. Under these conditions, stable ERG a-wave, b-wave, oscillatory potentials and c-wave could be recorded for 28 days. 2. SI at doses of 25 mg/kg or more caused depression of the amplitudes of the oscillatory potentials, and enhancement of the a- and b-wave amplitudes, while the c-wave was already extinguished on day 1 after dosing. Following these changes, the amplitudes of the a- and b-wave decreased from day 3 or 7 after dosing. These changes did not recover until day 28 after dosing. 3. Upon histopathologic examination of the retina, folding of the outer nuclear layer, disarrangement of the rods and cones, decrease of the visual cells and swelling and decrease of the pigment epithelial cells were observed with SI at 25 mg/kg or more. 4. Using this recording technique, it was confirmed that a stable ERG was recorded repeatedly for 28 days in mice, and the effects of SI on the ERG could be detected. Histopathologic findings in the retina revealed the abnormal portions were correlated well with the changes in the ERG. These results indicate that the ERG recording procedure developed in this study is useful for evaluating retinal toxicity in mice.