A pot experiment was conducted at glasshouse of Bangladesh Institute of Nuclear Agriculture (BINA), Mymensingh, with 18 extensively cultivated soils of Bangladesh to evaluate extractable P and to determine its critical limit for mustard. The available P contents in soils were extracted by four methods viz. Olsen, Hunter, Bray II and Nelson. The mean extractable P in soils was found to be in the order of Nelson > Bray 11 > Hunter > Olsen. Considerable variation was found in the amount of extractable P depending on the soils and extractants used. Dry matter yields were remarkably increased with addition of increasing rate of phosphatic fertilizer and the soil having low extractable P responded better to applied P. All extractable P except Nelson correlated positively and significantly with relative yield. Extractable P of all extractants except Nelson was positively and significantly correlated with P content " of mustard in all and low pH soils. Phosphorus uptake was also positively correlated with all extractants in all and low pH soils. The critical limits of soil extractable P for mustard for Olsen, Hunter, Bray II and Nelson method were estimated as 8.6, 9.5, 14.5 and 20.0 ppm (Graphical method) and 10.5, 12.5, 16.5, and 16.5 ppm (Statistical method) respectively. Based on B2 value it may be-concluded that Nelson method of P extraction was found to be the best method for estimating P response of mustard.