In forest soils where a large fraction of total phosphorus (P) is in organic forms, soil micro-organisms play a major role in the P cycle and plant availability since they mediate organic P transformations. However, the correct assessment of organic P mineralization is usually a challenging task because mineralized P is rapidly sorbed and most mineralization fluxes are very weak. The objectives of the present work were to quantify in five forest Spodosols at soil depths of 0-15 cm net mineralization of total organic P and the resulting increase in plant available inorganic P and to verify whether net or gross P mineralization could be estimated using the Cor N mineralization rates. Net mineralization of total organic P was derived from the net changes in microbial P and gross mineralization of P in dead soil organic matter. We studied very low P-sorbing soils enabling us to use lower extractants to assess the change in total inorganic P as a result of gross mineralization of Pin dead soil organic matter. In addition, to enable detection of gross mineralization of Pin dead soil organic matter, a long-term incubation (517 days) experiment was carried out. At the beginning of the experiment, total P contents of the soils were very low (19-51 mu g g(-1)) and were essentially present as organic P (17-441.mu g g(-1), 85-91%) or microbial P (6-14 mu g g(-1); 24-39%). Conversely, the initial contents of inorganic P were low (2-7 mu g g(-1); 9-15%). The net changes in the pool size of microbial P during the 517 days of incubation (4-8 mu g g(-1)) and the amounts of P resulting from gross mineralization of dead soil organic matter (0.001-0.018 mu g g(-1) day-1; 0.4-9.5 mu g g(-1) for the entire incubation period) were considerable compared to the initial amounts of organic P and also when compared to the initial diffusive iP fraction (<0.3 mu g g(-1)). Diffusive iP corresponds to the phosphate ions that can be transferred from the solid constituents to the soil solution under a gradient of concentration. Net mineralization of organic P induced an important increase in iP in soil solution (0.6-10 mu g g(-1); 600-5000% increase) and lower increases in diffusive iP fractions (0.3-5 mu g g(-1); 300-2000% increase), soil solid constituents having an extremely low reactivity relative to iR Therefore, soil micro-organisms and organic P transformations play a major role in the bioavailability of P in these forest soils. In our study, the dead soil organic matter was defined as a recalcitrant organic fraction. Probably because gross mineralization of P from this recalcitrant organic fraction was mainly driven by the micro-organisms' needs for energy, the rates of gross mineralization of C, N and P in the recalcitrant organic fraction were similar. Indirect estimation of gross mineralization of P in dead soil organic matter using the gross C mineralization rate seems thus an alternative method for the studied soils. However, additional studies are needed to verify this alternative method in other soils. No relationships were found between microbial P release and microbial C and N releases.