Phosphorus (P) is essential to plants and many crops require P fertilizers to optimize yields. However, there are concerns about future scarcity of the rock phosphates used to produce P chemical fertilizers. Soil organic phosphorus (SOP) might play a significant role in plant nutrition and soil P availability, reducing fertilizer re-quirements, but SOP mineralization is poorly understood. This study assessed rates of SOP mineralization and SOP speciation using a long-term field experiment in France, continuously cropped (1972-2000) and fertilized at three rates (triple superphosphate). The SOP concentration was determined every 2-3 years using the Saunders and Williams ignition method and converted to stock considering the soil mass per hectare; P speciation was determined using solution P nuclear magnetic resonance spectroscopy. The SOP stocks slightly increased with time, irrespective of P fertilization rates. Inorganic P stocks decreased without fertilization, but increased at the high P fertilization rate. By adapting a model describing soil organic carbon dynamics, the gross mineralization coefficient (k) was calculated with initial SOP stock (368 kg P ha- 1), P in maize residues (average of 17.0 kg P ha- 1 yr-1, differences across yields, plant parts and P fertilization rates) and their 17% incorporation rates into SOP (i.e. 2.9 kg P ha- 1 yr -1). Irrespective of P fertilization rates, the model calibration using SOP time series gave a k-value of 0.0047 yr- 1 (212 yr residence time) and a mineralization rate of 1.7 +/- 1.2 kg P ha- 1yr- 1. Speciation of SOP was 77% of orthophosphate monoesters, and 20% of orthophosphate diesters regardless of the P fertil-ization rates. The predominance of inositols hexakisphosphates, known to be strongly stabilized through sorp-tion, might explain the low SOP mineralization rate. The results for these soils suggest that SOP may only play a minor role in soil P availability and crop nutrition.