Abstract
In order to determinate the optimum nitrogen rate and time for split N application based on phonological stages of Bahar hybrid rice, an experiment was conducted over two years (2007-2008) using a randomized complete block design with three replications at Rice Research Institute of Iran (RRII) farm, Rasht. Factorial arrangement of the following factors was used: three N rates (90, 120 and 150 kg pure N in hectare) and six different phonological stages (T1: using leaf color chart with number 4; T2; 1/3 base fertilizer, 1/3 in rise primary bud stage, 1/3 in booting stage; T3; 1/3 base fertilizer, 1/3 in tillering stage, 1/3 in rise primary bud stage; T4: 1/2 base fertilizer, 1/4 in tillering stage, 1/4 in booting stage; T5: 1/3base fertilizer, 1/3in tillering stage, 1/3 in booting stage; T6; 2/3 base fertilizer, 1/3 in rise primary bud stage) with a control plot in each replication. After sowing in bed nursery the twenty days old seedlings were transplanted in the experimental plots. Results of analysis of variance clearly showed significant differences between different phonological stages on many characteristics. Means differences were highly significant for grain yield. The T6 treatment produced the highest yield of 8.373 and 7.920 t ha-1 in two years, respectively. However, 120 kg N ha-1 when applied as T5, produced the highest yield (8.760 t ha-1). Combined analysis of variance showed that yield and yield components.were affected by time of N application. Combined comparison of means also showed that the T5 treatment and 120 kg N ha-1 produced the highest yield of 7.925 and 7.514 t ha-1 in the first and second year, respectively. According to the results, it can be recommended that for Bahar hybrid rice maximum yield could be achieved by application of 120 kg N ha-1 when split as defined in T5 (1/3 based fertilizer + 1/3 first tillering stage + 1/3 booting stage).

Keywords: Hybrid rice, Leaf Color Chart (LCC), Nitrogen fertilizer