Abstract Experiments were conducted on the photoreductive dissolution of 54Mn-labeled synthetic oxides, prepared from MnO 4 2− oxidation of 54Mn(II), and natural labeled oxides formed in seawater from microbial oxidation of 54Mn(II). Sunlight increased the dissolution rate of synthetic oxides in seawater, an effect that increased with the duration of light exposure. The photodissolution of these oxides was found to result primarily from Mn reduction by H 2O 2, produced in seawater from the photoreduction of O 2 by dissolved organic matter. This conclusion was based on the previously observed marked stimulation of photodissolution by added humic compounds, the observed reductive dissolution of the oxides by added H 2O 2 and on the almost complete reversal of photodissolution by enzymatic (catalase) removal of H 2O 2. Sunlight had an even larger stimulatory effect on the reductive dissolution of 54Mn-labeled natural oxides. It increased specific dissolution rates to values of 6–13% h −1, 6–70 times higher than rates in the dark. In contrast to synthetic oxides, rates for natural oxides did not increase measurably with the duration of light exposure, were not appreciably altered by humic acid addition or by photolytic removal of natural organic matter, and were not substantially reduced by catalase addition. Furthermore, rates for reductive dissolution of natural oxides by H 2O 2 were only about 1/6th of those for synthetic oxides. These results indicate that the photoreductive dissolution of natural oxides in seawater is not primarily related to the photoproduction of H 2O 2, although such production appears to account for a small portion (ca. 10–20%) of the overall effect. Instead, both the chromophore and the reductant(s) involved in the reaction appear to reside with the bacterial/Mn oxide aggregates themselves. Although several possibilities can be postulated, the exact mechanism of the photochemical reaction remains obscure.