In a fertilization trial set up on calcareous humous sandy soil, soil samples were taken from rye fields from the ploughed layer at ten occassions during the vegetation period to examine nitrate desorption in a continuous aqueous flow. During the procedure a filter-paper disk and a porcelain plate were placed at the bottom of a Schachtschabel tube of 3 cm diameter, ending in a capillary, then followed by a mixture of soil sample and quartz sand, afterwards by a repeated quartz layer and filter paper the soil column was set up. In this column using a peristaltic pump, in a constant rate of liquid flow, where the volumes of liquid being introduced and removed were identical, the mixing became minimum due to the liquid flow of 2 ml/min, in each 5 minutes 10 ml fractions were removed. The results of the tests can be summarized as follows: The effect of N fertilization applied in the treatments was consistently effective in the quantities desorbed in the first 5-minute fraction. The nitrate desorption measured in the continuous aqueous flow reached its maximum during the first (5 minutes) fraction, then it decreased. The intensity of desorption was very high at the beginning, after five minute more than 70% of the total amount desorbed over 25-minute period was found already in the fraction. Also the N fertilization had an effect on the intensity of desorption, as in the first fraction in the non-fertilized treatment 50% of the total amount was found, while in the N fertilized treatments it exceeded the 70% of the total amount. The seasonal changes (Table 3) could be proved, but no uniform trend was found. In desorption tests on calcareous poorly homous sandy soil, the fractionation procedure with a continuous constant rate well characterized the effect of nitrogen fertilization, and the extent of nitrogen supply.