The effect of different NPK supply levels on grass hay yield, NPK content and uptake was investigated on a calcareous chernozem soil with light loam texture 3–5% CaCO3 content originally very well supplied with Mn, well supplied with Mg and Cu, moderately supplied with N and K and poorly supplied with P and Zn. The factorial NPK fertilizing long-term field trial was set up in the autumn of 1973. In the autumn of 2000, an eightcomponent grass mixture was sown. Among the components, there was now leguminous crop. Grass hay yields, hay NPK contents and –uptakes are presented in this paper for the period of 2001 to 2014.

With applying yearly 0, 100, 200 and 300 kg N ha-1, and several 0, 500, 1000 and 1500 kg P2O5 ha-1, or kg K2O ha-1 dose build up PK fertilization, last time in the autumn of 1999, poor, moderate, good, and excess soil NPK levels were established in the trial.

Grass hay yields and NPK contents varied according to the time passed since grass mixture sowing as well as according to the different soil NPK supplies. From the results, we got information about the crop year impacts, as well as the residual PK effects. Hay yields increased until N2–N3, P1–P2 and K1 levels. In the trial, grass hay NPK contents could be affected by the time elapsed since grass mixture sowing, as well as by the changing grass composition, the latter one depending on soil NPK supply as well.

The vegetative grass hay showed significant NK luxury uptake, the amount of NK uptake increased as far as N3K3 level. P “behaved” completely differently: similarly to hay yields, P contents increased only until P1–P2 levels.

From the difference between applied NPK amounts and grass NPK uptakes, we can get information about the specialities of NPK turnover in the different NPK treatments. N and P turnover showed similar characteristics, while K turnover was completely different from the other two macro nutrients.

Keywords: established grassland, NPK fertilization, long-term field trial, hay yields, NPK contents and uptakes