Abstract The static fertilisation experiment conducted in Skierniewice (Central Poland) since 1923 investigates the effect of mineral fertilisation with lime (CaNPK) or without lime (NPK) on the accumulation and release of phosphorus in reference to phosphorus sorption properties in the sandy soil profile. In the case of application of same doses of mineral fertilisers, the content of total phosphorus was higher in NPK than CaNPK soil. Parameters related to sorption capacity and bonding energy from Langmuir and Freundlich model of P sorption were significantly lower in CaNPK than NPK soil profile. This was particularly caused by a lower content of poorly crystallised hydro(oxide) aluminium and iron forms in CaNPK than NPK soil. Higher content of oxide-extractable and bioavailable phosphorus extracted with double lactate solution, dissolved reactive phosphorus in water solution as well as degree of phosphorus saturation in the CaNPK soil profile suggests higher mobility and possibility of occurrence of losses of phosphorus from the profile of limed soil than from acidified soil. Therefore, management of phosphate fertilizers on permanently limed sandy soils requires the optimisation of phosphorus doses to a greater degree corresponding to the actual take-off of the element with crop. An additional finding of the study was evidence of the possibility of re-estimating contents of bioavailable phosphorus and, as a consequence, the degree of phosphorus saturation with Mehlich3 method in strongly acid soil receiving P mineral fertilisers, which can make it difficult to use the test for fertiliser recommendation.