There have been many short-term field experiments in which the changes in Phosphorus (P) fractionations have studied. The fate of phosphorus derived from granular fertilizers in the present study has been examined in long-term field experiment on a calcareous sandy soil (Calcarosol) in a semi-arid environment (annual rainfall = 325 mm) at Walpeup, Victoria, Australia. The experimental plots were established in 1940 and had rotation of fallow-wheat-oat before 1956 and a fallow-wheat system since 1960. Five rates of granular single-superphosphate fertilizer was applied to each crop at rates of 0, 3, 6, 9 and 12 kg P/ha. Soil samples were taken from 0-10 cm before sowing in 2005. Soil P was sequentially extracted with distilled water, 0.5 M NaHCO3 (pH 8.5), 0.1 M NaOH, 0.5 M H2SO4¬, and 1 M HCl and H2SO4 /H2O2¬. Inorganic P in each fraction was determined colorimetrically, and total P using ICP-AES. Organic P fractions was the difference between total P in extract and inorganic P fractions. The soil samples were also analysed for resin-P and Olsen-P. A glasshouse experiment was conducted to examine the utilisation of the different pools of P in this soil by two different crops: wheat (Triticum aestivum L. cv. Yitpi) and chickpea (Cicer arietinum L.). Following 8 weeks growth, P in shoots and roots were analysed. The rhizosphere soil was sampled and analysed for P fractions to examine the changes of P fractions after cropping. Application of P fertilizer increased significantly all inorganic soil P (Pi) fractions (P<0.05) except HCl-Pi. By contrast, the proportion of organic P (Po) in the extracts tended to decrease with increasing rates of P fertilizer applied to the soil. Residual P, inorganic P and organic P fractions were accounted for an average of 66%, 27% and 8% of total soil P, respectively. The amount of resin-extractable P (11.8-41.9 mg/kg soil) was similar to the amount of water- plus bicarbonate-extractable Pi (12.4-43.4 mg/kg soil). Olsen P was strongly correlated with water extractable P fraction and resin-P (r2 = 0.98). Thus, the long-term application of P fertilizer resulted in an increase of residual P and inorganic P fractions. All Pi fractions in the rhizosphere soil sharply decreased following 8 weeks growth of either wheat or chickpea, indicating that both species were able to access all inorganic pools, and that the residual P fraction could be a buffer for more labile P pools. In contrast, all Po fractions decreased significantly (P<0.05) except water-extractable Po, which slightly increased after the first cropping. This indicates that organic P was mineralised to replenish solution Pi. The chickpea plant showed a better growth in comparison with wheat and took up more P where no P was added to the soil, which was probably due to greater root exudation and lower rhizosphere pH of chickpeas. Increasing P fertiliser rate increased total uptake by both species but increased more in chickpea. It appears that the long term application of fertilizer P to the calcareous sandy soil resulted in a build-up of residual P and Pi fractions. However, these fractions and high proportion of Po fractions could be a potential source for plant available P. Legume such as chickpea appear to have better access to “non-labile” P pools than wheat in this calcareous sandy soil.