Fe-P, considered a less available fraction, had a significant negative correlation with Olsen-P in both treatments in 2010. This might imply that Fe-P is averse to bioavailability of P in wet years in this area (Ma et al., 2009). But there was a significant positive correlation between Fe-P and Olsen-P in PFM plots in 2009. This suggests that the effectiveness of Fe-P for crop growth may be enhanced by adding PFM to soil in dry years. The Ca10-P content was significantly higher in both treatments than the concentrations of other fractions, making it a dominant fraction among inorganic P fractions. Ca10-P is generally considered unavailable to most plants (Samadi and Gilkes, 1999; Zhang et al., 2004a).