Field experiments were conducted in wet seasons (June-December) for 4 years on a clay loam Haplaquept, to study the effects of different N management practices on yield, urea and ammonium-N in flood water and N nutrition of an clite rice cv. CR 1009, grown in rainfed lowlands. During the first 3 years of the experiment, fertilizer management practices like band placement of neem-cake-coated urea (NCU), broadcast application of sulphur-coated urea (SCU) at sowing, or point placement of urea supergranules (USG) 3 weeks after germination at 40 kg N/ha gave grain yields of 3·1–3·4 t·ha, which were almost equal to that of split application of prilled urea (PU). In the 4th year of the experiment, besides NCU and USG, single dose applications of PU as band placement, incorporation in the soil at sowing or broadcast incorporation of soil-treated urea at early tillering was also found to have similar effect on grain yield and N uptake as split application of PU. The flood water of the treatment receiving broadcast application of PU at tillering contained some urea and ammonium N, which rapidly decreased to negligible amounts in 3·4 days.
The results suggest that, depending upon the feasibility, any one of the single dose application methods at sowing time or 3 weeks after germination may be adopted in this system of rice culture, which avoids top-dressing of PU to surface flowing flood water of greater depths at later stages of crop growth.