Recently there has been considerable attention given to the possible deleterious effects of residual hydrogen peroxide on both the cornea and the crystalline lens of the eye. We measured residual hydrogen peroxide levels of the AOSept disinfection system at regular intervals over a 4 month period. By 1 month the residual hydrogen peroxide level was 21 ppm +/- 9.4 (mean +/- standard deviation); at the end of 2 months it was 36 ppm +/- 17.6. At 3 months, when disc replacement is recommended, the residual hydrogen peroxide level was 43 ppm +/- 4.7 (range: 17-98 ppm). At 4 months the residual hydrogen peroxide level rose to a mean of 48 ppm +/- 18.2. The increase in measured residual hydrogen peroxide between month 1 and each successive month was found to be statistically significant. However, these levels failed to induce an increase in symptomatology or slit lamp findings. As determined by scanning electron microscopy there was an erosion of the platinum coating with time that appeared to coincide with the measured residual peroxide level. Cultures of randomly selected platinum discs at the end of 4 months failed to reveal more than isolated positive findings of Staphylococcus epidermidis. While an increase in residual peroxide with time may lead to greater disinfecting capabilities of this system, the possibility of corneal damage as a direct result of high peroxide levels is of concern.