Phosphorus (P) application and leaching of P in soils affect P distribution and can significantly influence P losses. The effect of P addition and long-term P leaching on P fractionation in 11 calcareous soils of varying properties was investigated. The soils were spiked with 200 mg P·kg^-1 and the soil columns were leached with distilled water over 40 days. In order to maximize P leaching and simulate the long-term leaching of P, we added more water to the columns than the mean annual rainfall of the study area. Phosphorus in native and amended soils were fractionated before and after leaching by sequential extraction procedure, in which the P fractions were experimentally defined as exchangeable (NaHCO₃-P), Fe- and Al-bound (NaOH-P), Ca-bound (HCl-P), and residual P (Res-P) fractions. The results showed that in native soils the major proportion of P was associated with Ca. Among the P fractions, the percentage of NaHCO₃-P and NaOH-P increased, whereas that of Res-P and HCl-P decreased with applied P, indicating former fractions served as a primary sink for P fertilizers added to soil. There were changes in the proportional distribution of P in all the soils, and P was redistributed among the various fractions during leaching of P. In general the proportions of P associated with the NaOH-P and Res-P tended to decrease, with corresponding increases in the HCl-P fractions after leaching, resulting in a minimal amount of P being leached from these calcareous soils. In contrast, there was little decrease in NaHCO₃-P fraction with leaching. The moderately labile P forms (NaOH-P and Res-P pools) seem to act as slow-release P sources contributing to long-term P release. The amounts of P extracted in the NaHCO₃ and NaOH fractions in native soils were good predictors of cumulative P leached during the 40 days of leaching in these calcareous soils.