In order to make more efficient use of mineral nitrogen fertilizers, the mineralization of organic N has to be fully understood and taken into account when meeting the nitrogen demand of crops. Aerobic long-term incubation experiments (147 days) based on themethod by Stanford and Smith (1972, Soil Sci Soc Am Proc 36: 465), modified by Nordmeyer and Richter (1985, Plant Soil 39: 433), were carried out with paddy soils (Typic Epiaquepts) from two different locations in the Taihu Region in Eastern China, each in three depth increments (0-30, 30-60, 60-90 cm). The two soils had pH (H2O) values between 6 and 7, silty clay loam texture, C-org contents (0-30 cm) of 8.9 and 16.6 g kg(-1) and N-tot contents of 1030 and 1650 mg kg(-1), respectively. Mineralization parameters were estimated from the cumulative amounts of leached nitrate-N via non-linear regression, using a two-pool approach based on first-order kinetics reactions. The results were compared to those from incubation experiments carried out with topsoils from the southern edge of the Chinese Loess Plateau (Roelcke et al., 1996, Trans. 9th Nitrogen Workshop, Braunschweig, p, 313). The optimized parameters were used to simulate the N regime in paddy soils with a deterministic model (Han et al., 2001, Proc. Int. Conf. China, p. 411). A winter wheat growing period (November 1995 to June 1996) in Wuxi County was simulated. A total of 220 kg ha(-1) mineral N fertilizer were applied as NPK and urea, split into four doses. Simulated results were compared to the measured mineral nitrogen (N-min) contents over the growing season. Cumulative aerobic N mineralization (0-1.0 m profile) yielded 84 kg N ha(-1). The modelled total cumulative N losses amounted to 76 kg N ha(-1) or 35% of the N applied.