Rice plants (Oryza sativa L.) grown on soils containing low soil testextractable P frequently do not respond to fertilizer P application under reduced conditions. The lack of rice response to fertilizer P in soils with low extractable P has been attributed to increased solubility of Fe-associated P upon flooding. The increased solubility of Mn oxides and release of Mn-associated P in flooded soils may also increase P availability to rice plants. The effect of Mn oxides on P availability has not been adequately described or quantified. In this experiment we used rice soils with low soil test P and high or low Mn or Fe oxides to measure the effects of Mn oxides on P availability under field, greenhouse, and laboratory conditions. Soils were incubated under oxidized or reduced conditions for 35 days. At the end of 0, 2, 5, 8, 11, 15, 20, 25, 30, and 35 days, soil solution pH and Eh and soluble and extractable P, Fe, and Mn were determined. Greenhouse and field experiments showed that rice plants did not respond to P application in three of four soils despite low extractable P. A large increase in P concentration following reduction was observed and was correlated positively with reactive Fe and Mn oxides in soils. During incubation water-soluble P in soils increased from 0.06 to 0.37 mg P kg−1, Fe from 0.2 to 22.0 mg Fe kg−1, and Mn from 0.08 to 6.43 mg Mn kg−1. Linear stepwise regression analysis indicated that 65 to 91% of solution P variation was attributable to Fe oxides and about 14 to 28% was attributable to Mn oxides. The contribution of Fe reduction to increased P availability was highly significant and related to large amounts of poorly crystalline Fe oxides in soils extracted by oxalate. The manganese oxide contribution to P availability was important, especially in the early stages of soil reduction in soils with large amounts of reactive Mn. In soils dominated by Mn oxides, the reduction of Mn-associated P can be a significant source of P for plants.