A field experiment was conducted during the rainy seasons of 1986 and 1987 at New Delhi, to study the response of 'Pusa 33' and 'Pusa 312' direct-seeded rice (Oryza sativa L.) to level and time of N application under irrigated upland conditions. 'Pusa 312' rice was taller and showed more tillers, heavier panicles, grain yield (37.62-41.75 q/ha) and N uptake than 'Pusa 33' in both the years. Increased levels of N increased the growth, yield components, yield and N uptake up to 80 kg/ha. Application of N in 4 equal spits, viz. as basal or at seedling stage, at tillering. panicle-initiation and heading stages, could not exert any significant influence on plant height, tiller and panicle production compared with those in 3, omitting the last one at the heading stage. However. panicle weight, grain yield and N uptake were significantly higher in 4 splits, irrespective of the time of application the first split. The time of application of the last split at heading, had marked effect, whereas the time of the first split either basal or at seedling stage, did not show varied effect.