On lime coated loamy chemozem soil (mean temperature 10,9 degrees C, average of many years' precipitation 576 mm) in 1990, in the 17th year of a long-term NPK fertilization trial set up on the Experimental Farm of the Research Institute at Nagyhoresok, the distribution of NO3-N in the depth of the soil-profile was examined. Following the harvest of peas, soil samples were taken from each 20 cm soil-layer till 6 m depth, in the control plot and the 100, 200, 300 kg N/ha/year treatments. On the plots treated the sample drawing made by boring was carried out in 2 replications, so totally 7x6x5=210 samples were analysed. The main conclusions can be summed up as follows: 1. 35-63% of fertilizer-N exceeding the plant uptake was provable in the soil, in the form of NO3-N. Parallel with the rate of over-fertilization the amount of N found in the soil increased. In plots with negative N-balance, lower than 1 m depth there was hardly any leaching worth mentioning. 2. The maximum of NO3-N leached can be found in the upper layer of 1 m. The border of leaching reached the 17th year of the trial a depth of 5-5,5 m. Five years earlier this border was in a depth of 3,5-4 m, so the speed of leaching may correspond to 20-30 cm/year. If this movement downwards remains, the nitrate may reach in further 20-30 years the level of soil water, which can be found in 13-15 m depth. 3. On the control plot a supply of 42 kg/ha NO3-N was found in the upper 1 m layer, while in the treatment of 300 kg N/ha/year 827 kg/ha NO3-N; what could cover the N-demand of about 5 years' average wheat yield. The NO3-N supply of 2000 kg N/ha of the deeper than 1 m layers cannot be regarded as nutrient on plots with maximum N-doses. This N may pollute the soil, and after the lapse of 2-3 decades the soil-water. 4. Knowing the available N-supply of the soil, and the N-demand of the crop, the over-fertilization can be avoided. The pollution of the soil and soil-water by nitrate is not inevitably linked with fertilization.