Five experiments were conducted to investigate the ability of different phytase products to improve P digestibility in finishing pigs. A corn-soybean meal basal diet containing 0.50% Ca, 0.32% P, and 0.40% Cr(2)O(3) was used to calculate apparent P and GE digestibility. Pigs were individually penned and fed their respective diet for ad libitum intake for 12 d before fecal sampling on d 13 and 14 and blood collection on d 14 for plasma P determination. Experiments 1 through 4 used gilts with across-trial average initial and final BW of 84 and 97 kg, respectively. Pigs were fed Natuphos (Exp. 1), OptiPhos ( Exp. 2),Phyzyme ( Exp. 3), or RonozymeP ( Exp. 4) at 0, 200, 400, 600, 800, or 1,000 phytase units (FTU)/kg ( where 1 FTU is defined as the quantity of enzyme required to liberate 1 mu mol of inorganic P per min, at pH 5.5, from an excess of 15 mu mol/L of sodium phytate at 37 degrees C). Experiment 5 used barrows with initial and final BW of 98 and 111 kg, respectively, and were fed diets containing 0, 500, or 1,000 FTU/kg of Natuphos, OptiPhos, Phyzyme, or RonozymeP. Pigs fed Natuphos ( Exp. 1) and OptiPhos ( Exp. 2) exhibited a linear and quadratic ( P < 0.01) improvement in P digestibility with increasing levels of dietary phytase, whereas pigs fed Phyzyme ( Exp. 3) and RonozymeP ( Exp. 4) exhibited a linear ( P < 0.01) improvement in apparent P digestibility with increasing levels of dietary phytase. In Exp. 5, the improvement in apparent P digestibility with increasing levels of dietary phytase was linear ( P < 0.01) for Natuphos, Phyzyme, and RonozymeP, but was linear and quadratic ( P < 0.01) for OptiPhos. Based on regression analysis, inorganic P release at 500 FTU/kg was predicted to be 0.070, 0.099, 0.038, and 0.030% for Natuphos, OptiPhos, Phyzyme, and RonozymeP, respectively. These estimates are comparable with those of pigs in Exp. 5, for which the estimated inorganic P release at 500 FTU/kg was 0.102, 0.039, and 0.028% for OptiPhos, Phyzyme, and RonozymeP, respectively, but not for the 0.034% value determined for Natuphos. The effect of dietary phytase on GE digestibility was inconsistent with a linear ( P < 0.01) improvement in GE digestibility noted for OptiPhos ( Exp. 2 and 5) and RonozymeP ( Exp. 4), but the quadratic ( P < 0.01) improvement for Natuphos. There was no effect of dietary phytase on plasma inorganic P. The data presented show clear improvements in P digestibility, with the estimated level of inorganic P release being dependent on phytase source and level.