Two experiments were conducted to determine effective P fertilization strategy in fish ponds in relation to sediment P saturation level. Experiment 1 was conducted in cement tanks with five levels of P saturation in the sediments (5, 24, 44, 60 and 79%) and with three P fertilization rates (0.2, 0.1 and 0.05 g m - 3 day - 1 , N:P ratios of 2:1, 4:1 and 8:1, respectively). Nile tilapia (Oreochromis niloticus) were cultured in the tanks for 57 days. Results showed that the mean concentration of soluble reactive phosphorus (SRP) in the water column increased with increasing sediment P saturation and P fertilization rate. The maximum net fish yield (NFY), 4.2 ± 0.3 g m - 3 day - 1 , was obtained at an SRP concentration of 0.3 mg l - 1 ; higher concentrations did not increase fish yield. These levels of SRP and NFY were attainable with P fertilization rates of 0.2 g m - 3 day - 1 , 0.1 g m - 3 day - 1 and 0.05 g m - 3 day - 1 and N:P ratios of 2:1, 4:1 and 8:1 in ponds where the level of sediment P saturation was below 10%, above 45% and above 60%, respectively. Experiment 2 was conducted in earthen ponds to test and verify the P fertilization rate based on the cement tank results. Three new and three old ponds with 8 ± 1.7% and 88 ± 7.3% sediment P saturated in the top 5 cm of mud were fertilized at a rate of 0.2 g m - 3 day - 1 and 0.05 g m - 3 day - 1 and N:P ratio of 2:1 and 8:1, respectively. Nile tilapia were cultured at a density of two fish m - 3 for 85 days. The mean NFY obtained in new and old ponds were 1.73 ± 0.08 g m - 3 day - 1 and 2.24 ± 0.32 g m - 3 day - 1 , respectively, which were not significantly different (P > 0.05). We conclude that P fertilization rate should be based on P saturation level in mud to overcome the problem of under or over supply of P in fish ponds.