Declining global P reserves require a better understanding of P cycling in soil and related plant uptake. On managed grasslands, application of lime and fertilizer affects not only soil nutrient status, but also plant-species composition of the sward. We examined the P fractionation in the Rengen Grassland Experiment (RGE) on a naturally acid Stagnic Cambisol in the Eifel Mts. (Germany) 69 y after the setup of the experiment. A modified sequential Hedley fractionation was carried out for samples from 30 plots at 0–10 cm depth. Application of inorganic phosphorus fertilizer had diverse effects on inorganic (Pi) and organic P (Po) fractions. Resin-Pi, NaHCO3-Pi, NaHCO3-Po, NaOH-Pi, HCldil-Pi, HClconc-Pi, and HClconc-Po contents increased, while NaOH-Po significantly decreased and residual-P remained unaffected. Strongest enrichment occurred in the HCldil-Pi fraction, probably due to the chemical nature of the basic Thomas slag applied as P fertilizer. Without P fertilization, all fractions except residual-P were more or less depleted. Strong P limitation of the vegetation in the limed treatments without P led to lowered contents also for NaOH-Pi and NaOH-Po. However, NaOH-Po was largest in the Control and even exceeded the respective content in the treatments with P. It remained unclear why species adapted to a low soil P status did not access this P fraction though being P-limited. Published theory on the availability of Hedley P fractions does neither match P exploitation nor P nutritional status of the vegetation in the RGE. Regarding NaOH-Po as stable and HCldil-Pi as moderately labile led to a more realistic evaluation of plant P uptake. Evaluation of P availability on the basis of chemical extractions alone is questionable for conditions like in the RGE. On long-term grassland, plant-species composition has to be taken into account to estimate access of plants to soil P.