An experiment was conducted to determine the standardized total-tract digestibility (STTD) of phosphorus (P) in 5 sources of inorganic phosphate fed to pigs. The 5 sources of inorganic phosphate were dicalcium phosphate (DCP, 19.3% P), monocalcium phosphate (MCP, 22.7% P), monodicalcium phosphate (MDCP, 22.0% P) which is the mixture of MCP and DCP at a ratio of 3:1, monosodium phosphate (MSP, 24.5% P), and tricalcium phosphate (TCP, 19.4% P). Six barrows with an initial body weight of 42.4 ± 1.1 kg were individually housed in metabolism cages equipped with a feeder. Pigs were randomly allotted to a 6 × 6 Latin square design with 6 dietary treatments and 6 periods. Each experimental period consisted of a 4-d adaptation and a 5-d collection period. Five experimental diets contained 0.25% P from each inorganic phosphate as a sole source of P, and were formulated to contain calcium in amounts that were 1.3 times greater than the concentration of P except the TCP diet. A P-free diet mainly based on cornstarch, gelatin, and sucrose was also prepared to estimate the basal endogenous loss of P. Feces were collected using the marker-to-marker procedure. Data were analyzed using the MIXED procedure of SAS with a model including dietary treatment as a fixed variable and animal and period as random variables. Values for the apparent total-tract digestibility of P in DCP, MCP, MDCP, and MSP (83.0, 86.5, 81.4, and 88.7% respectively) were greater (P < 0.05) than in TCP (67.0%). Values for the STTD of P in DCP, MCP, MDCP, and MSP (89.8, 93.3, 88.2, and 95.5%) were also greater (P < 0.05) than in TCP (72.8%). In conclusion, digestibility of P in TCP was less than other inorganic P sources.