Denitrification causes important losses of N-fertilizer in rice-fields, where high temperature and high production of organic matter favour denitrification losses. Two techniques have been used to quantify the denitrification losses: the 15 N technique, which can be used to quantify the amount finally incorporated, and the acetylene inhibition technique which is a direct measure of the quantities lost.
Both techniques were applied in enclosures (diameter = 44 cm) in the field while moreover bio-assays in 3 1 glass beakers were carried out. In all experiments where nitrate was added we found a rapid decrease of nitrate; usually about 30-50ยฐo of the nitrate that disappeared was recovered as N 2 0. As in one experiment, in which we measured the N 2 0 disappearance rate as well, the N 2 0 itself decreased at a rather constant rate of 20% per day, a correction must be made for this N 2 0 decrease in the calculations of the nitrate disappearance rate. Although we have only one series in which the decrease of N 2 0 was measured, the mathematical analysis indicates that as much as 80ยฐo of the N-fertilizer is actually lost. This figure is in full agreement with the 15 N experiments; if the 'SN was applied early only about 7% was recovered in soil and plants, while if it was applied later (after 7 weeks) about 20% was incorporated.
Denitrification rate could be fitted on an negative exponential regression line; the rate constant increased during the summer. It is suggested that organic matter caused this increase.
During denitrification considerable quantities of nitrite appear, which later on disappear again by processes still unknown; the nature of the available organic matter may be important for this nitrite production.
With N-serve we tried to inhibit NH 3 oxidation. In this way we tried to prevent the considerable N losses and to demonstrate that the nitrite produced in our experiments was not derived from NH 3 oxidation. Nserve, however, had very little influence. It is probably inactivated by absorption onto the sediments.
From these results it is suggested that the efficiency of N-application may be considerably increased by using low doses of N-fertilizer, but applied late in the growing season, e.g. 7 weeks after sowing. This favours environmental protection as well.