INTRODUCTION

Rice plays the most important role for feeding of millions of Asian people, where more than 90% of the world’s rice is grown and eaten (IRRI, 1989). Also, north of Iran as one of the main areas for producing rice inside the country, plays a critical role for feeding Iranian consumers. Rice is the second important crop both in crop production and in human nutrition worldwide. One of the main problems of rice production in about 500,000 ha of northern riceland of Iran is soil deficiency for most of macronutrients. Two of these elements are nitrogen and potassium. FAO (2000) has predicted that the demands for rice will outstrip its production. For overcoming this problem breeders in Asia tries severely to release higher yielding cultivars and for further increasing rice production, Chinese rice breeders developed rice hybrid cultivars, which claims that they gave a 20-30% higher yield in comparison to the best high yielding varieties (Lin and Yuan, 1980).

For higher production of yield using new hybrid cultivars, nitrogen is one of the main parts of rice production technology and also potassium is an important component for some of quality characteristics of rice (Prasad and De Datta, 1979; De Datta and Craswell, 1982; Kropff et al., 1993). The main reasons of nitrogen deficiency which is widely reported in lowland rice growing soils worldwide (Kundu et al., 1996; Fageria and Baligar, 1996), are loss of nitrogen through leaching, volatilization, surface runoff and denitrification (Prakasa Rao and Prasad, 1980; Fillery et al., 1989; Buresh and De Datta, 1990; Prasad, 1998; Prasad et al., 1999; Fageria and Barbosa Filho, 2001; Kumar and Prasad, 2004).

Generally, forms and amounts of applied fertilizers influence to a great extent on the yield and quality of rice (Wery et al., 1988; Gunes et al., 1995; Sidiras et al., 1999). Also, use of nitrogen and potassium efficient genotypes is one of the critical complementary strategies in improving rice yield and reducing cost of production. Nitrogen and potassium efficiency should be taken into account for improving new cultivars. Increasing high quality rice yield per unit area through use of appropriate nitrogen and potassium management practices has become an essential component of modern rice production technology (Fageria and Barbosa Filho, 2001). Practice of proper management strategies like adequate rate and timing of application and use of efficient crop genotypes, may increase rice yield and influence cost of production, simultaneously. The objective of present study was to evaluate yield components of rice cultivars when receiving different amounts of nitrogen and potassium fertilizers.

MATERIALS AND METHODS

This experiment has been undertaken on field trial, on rice cultivars Neda and Tarrom (two Iranian improved and landrace rice cultivars, respectively) in 2004 and 2005. The farm soil contained 515 g kg–¹ clay, 295 g kg–¹ silt, 90 g kg–¹ sand, 23.4 g kg–¹ total nitrogen, 213 mg kg–¹ available potassium and 23.6 percent lime. Four levels of potassium fertilizer (0, 75, 150 and 225 kg k₂ O ha–¹ from potassium sulfate source) and four levels of nitrogen fertilizer (0, 50, 100 and 150 kg N ha–¹ from urea source) have been applied. Half of the amount of potassium fertilizer was added in transplanting date and the rest was added in shooting stage and of the amount of nitrogen fertilizer was used in transplanting time, was added in tillering stage and the rest was added in flowering initiation stage. A split-factorial design with 3 replications based on a randomized complete block design with sub-plot size of 3x4 m and plant space of 25 cm have been used and 3 seedlings were transplanted per a hill. Samples were taken from flag leaf in heading stage to determine the amount of leaf mineral elements. Amount of leaf potassium and nitrogen were measured and plant height, panicle length, length and width of flag leaf, number of tillers, number of grains per panicle, number of hollow grains, 1000 grain weight, grain yield, biological dry matter and shoot dry matter were also determined.

To calculate grain harvest index, the grain yield at 14 humidity percent were divided on biological dry matter which is summation of grain yield and shoot dry matter. Nitrogen and potassium efficiency for plant height, number of tiller, length of flag leaf, number of grain per panicle, grain yield, shoot dry matter, biological dry matter and harvest index were calculated using the following formulas (Fageria, 1998; Fageria and Barbosa Filho, 2001):

where, N_f is the measure of corresponded trait of high level of nitrogen fertilized plot, N_u is the measure of corresponded trait of low level of unfertilized plot and N_a is the quantity of N applied (kg).

where, NE_Y is the nitrogen efficiency of grain yield and NE_TDM is the nitrogen efficiency of biological dry matter.

where, K_f is the measure of corresponded trait of high level of potassium fertilized plot, K_u is the measure of corresponded trait of low level of unfertilized plot and K_a is the quantity of K₂O applied (kg).

where, KE_Y is the nitrogen efficiency of grain yield and KE_TDM is the potassium efficiency of biological dry matter.

The statistical analyses of data were conducted by analysis of variance and the F-test was used to determine treatment significance using MSTATC statistical software. Duncan's multiple range test was used to compare treatment means at 5 and 1% probability levels.

RESULTS AND DISCUSSION

Table 1 demonstrates significant and non significant effects of N and K applications for tested rice cultivars within two years on 14 agronomic important traits. Cultivars, nitrogen and potassium affect significantly on all of considered traits including plant height, number of tillers, leaf length, leaf width, panicle length, number of seeds per panicle, grain yield, shoot dry matter, straw plus grain dry matter, harvest index, leaf potassium contents, 1000 grain weight, leaf nitrogen contents and percentage of hallow grain. Although the effects of these three factors were highly significant, there was seldom significant effects for their interaction with year or with each other. The differences among interaction of cultivar and nitrogen were significant for all traits except K uptake in leaf (Table 2). However, only plant height, K uptake in leaf, hollow grain and number of tillers showed significant differences in interaction of cultivar and potassium. Interaction of nitrogen and potassium illustrated non significant effects for most of measured traits except for plant height, number of grain per panicle and biological dry matter. Finally, significant results have been achieved only for number of tillers, shoot and biological dry matter and harvest index due to effects of three way interactions among nitrogen, potassium and cultivar.

Table 1:	Significance of F values derived from analysis of variance of yield and yield components of two cultivars, four N and four K levels
Y = Year C = Cultivar N = Nitrogen K = Potassium ** (p<0.01) * (p<0.05) ns = not significant

Table 2:	Reaction of yield and yield components of rice cultivars to different levels of N and K application
Mean values with the different letter(s) are significantly different, C = Cultivar, T = Tarrom, N = Neda N0 = without N N1 = 50 kg N ha–1 N2 = 100 kg N ha–1 N3 = 150 kg N ha–1 K0 = without K K1 = 75 kg K2O ha–1 K2 = 150 kg K2O ha–1 K3 = 225 kg K2O ha–1

Nitrogen and potassium efficiency for 10 considered traits showed that there are differences between Neda and Tarrom cultivars for both of macronutrients (Table 3). Nitrogen efficiencies were higher than potassium efficiencies in all of studied traits, except in width of flag leaf. These efficiencies were not high enough for height, number of tillers, length of flag leaf, length of panicle, number of grain per panicle and harvest index, however, for grain yield, shoot dry matter and biological dry matter they were high enough. Nitrogen plays an efficient role with 18.77 and 18.47 for Tarrom and Neda, respectively. However, role of potassium with 2.378 and 2.919, respectively for Tarrom and Neda seems not to be so efficient. Similar trends have governed on N- and K-efficiencies of Tarrom and Neda for traits shoot and biological dry matter (Table 3).

Plant height and number of tiller: Plant height and number of tiller per plant in both cultivars were different and nitrogen and potassium have significantly affected on these two characters. Number of tiller per plant in Neda was greater than Tarrom. Also, interactions of cultivar and nitrogen and cultivar and potassium on these two traits were significant. Three way interactions of cultivar, nitrogen and potassium were significant only on number of tiller per plant but not on plant height (Table 1).

Table 3:	Efficiency of nitrogen and potassium application in different rice cultivars for measured traits

Cultivars Tarrom and Neda showed different trends for various levels of nitrogen and potassium applications for the measured traits. By increasing nitrogen level plant height increased from 67.42 to 79.38 cm and from 92.75 to 108.40 cm in Tarrom and Neda, however, increasing potassium level increased plant height from 71.90 to 76.21 cm and from 94.67 to 104.30 cm in Tarrom and Neda, respectively (Table 2).

The efficiency of nitrogen application for plant height in Neda (0.10) was greater than Tarrom (0.08). The same trend has governed on the efficiency of potassium application on plant height, means that Neda with 0.050 has used potassium fertilizer more efficient than Tarrom with 0.023 (Table 3).

Due to nitrogen application, number of tiller per plant in Tarrom and Neda changed from 26.88 to 31.88 and from 31.08 to 33.04 tillers per plant, respectively (Table 2). A number of researchers have reached into a result that application of nitrogen can increase number of tillers in rice cultivars (Imam and Nicknejad, 1994; Fageria and Barbosa Filho, 2001; Shen et al., 2003; Qian et al., 2004). Nevertheless, potassium application also increased number of tillers per plant from 23.83 to 26.00 and from 26.88 to 31.17 in Tarrom and Neda, respectively (Table 2). The result of increasing number of tillers due to potassium application has also been supported by some other considerations (Bansal et al., 1993; Pandey et al., 1993; Kalita et al., 1995; Ojha et al., 2000).

Interactions of nitrogen and potassium on plant height and number of tillers per plant showed no significant effect with the exception of significant outcome for number of tiller per plant in Neda (Table 4).

Length and width of flag leaf and length of panicle: There were significant variations in two studied cultivars for length and width of flag leaf and length of panicle so that, these characteristics have been affected by nitrogen and potassium application. Also, interaction effects of cultivar and nitrogen on length and width of flag leaf and length of panicle were significant (Table 1). Length of flag leaf has significantly increased by nitrogen application from 19.63 to 29.25 cm and from 24.83 to 28.13 cm in Tarrom and Neda, respectively (Table 2). Potassium application has also changed length of flag leaf from 22.79 to 26.29 and from 25.54 to 27.88 cm in Tarrom and Neda, respectively (Table 2). Length of panicle has been influenced by nitrogen and potassium applications, so that; it changed from 17.88 to 27.42 and from 22.71 to 26.50 and also from 20.96 to 23.92 and from 24.08 to 26.38 cm in Tarrom and Neda by nitrogen and potassium applications, respectively (Table 2). Bansal et al. (1993) have also measured an increase in length of panicle because of nitrogen application.

The efficiency of nitrogen application for length of flag leaf in Neda (0.02) was smaller than Tarrom (0.06). The same trend has governed on the efficiency of potassium application on length of flag leaf, means that Neda with 0.013 has used potassium fertilizer less efficient than Tarrom with 0.019; however, these efficiencies were not important in width of flag leaf in both genotypes (Table 3).

Grain number per panicle and grain yield: There were significant differences in traits grain number per panicle and grain yield between both of the studied cultivars. These two characters were under influence of nitrogen and potassium applications. Interaction effects of cultivar and nitrogen on both traits and interaction of nitrogen and potassium on grain number per panicle were significant (Table 1). Application of nitrogen causes an increase in grain number per panicle from 94.23 to 130.46 in Tarrom and from 109.20 to 143.30 in Neda and in grain yield from 3099 to 5915 kg ha–¹ in Tarrom and from 3193 to 6104 kg ha–¹ in Neda, respectively (Table 2). These results are in accordance with the results of Bansal et al. (1993), Fageria and Barbosa Filho (2001), Fageria and Baligar (2001) and Shen et al. (2003).

Application of potassium cause also an increase in grain number per panicle from 106.29 to 116.17 in Tarrom and from 118.00 to 130.90 in Neda and in grain yield from 4527 to 4971 kg ha–¹ in Tarrom and from 4651 to 5196 kg ha–¹ in Neda, respectively (Table 2). Bansal et al. (1993), Sreemannarayana and Sairam (1993), Pandy et al. (1993), Kalita et al. (1995), Brouhi et al. (2000) and Ojha and Talukdar (2000) have measured an increase in both traits using potassium fertilizer.

Nitrogen and potassium interactions on grain number per panicle and grain yield were significant (p<0.05). Treatment N3K3 produced greatest number of grain and grain yield in both cultivars with 136 and 137.5 grains and 6136 and 6293 kg ha–¹, respectively (Table 4). Increasing in grain yield because of simultaneous application of nitrogen and potassium has been reported by Zhan and Wang (2005).

Table 4:	Interactive effect of N and K on yield and yield components
Mean values with the different letter(s) are significantly different, T = Tarrom, N = Neda

The efficiency of nitrogen application for length of panicle and grain yield in Neda (0.03 and 18.47) was smaller than Tarrom (0.06 and 18.77), respectively. The same trend has governed on the efficiency of potassium application on grain number per panicle, means that Neda with 0.016 has used potassium fertilizer more efficient than Tarrom with 0.012. However, the trend in grain yield is vise versa with 2.919 and 2.378 in Neda and Tarrom, respectively (Table 3).

Shoot and biological dry matter: There were significant differences among cultivars for amount of shoot and biological dry matter (Straw + Grain). Nitrogen and potassium application and interactions between cultivar and nitrogen and among cultivar, nitrogen and potassium were significant (p<0.01) in production of amount of shoot and biological dry matter (Table 1). Application of nitrogen in Tarrom causes an increase in the amount of shoot dry matter from 5035 to 6089 kg ha–¹ and biological dry matter from 8130 to 12000 kg ha–¹ whereas, in Neda shoot dry matter increased from 4633 to 6857 kg ha–¹ and biological dry matter increased from 7908 to 12920 kg ha–¹ (p<0.01). The same results were obtained from Bansal et al. (1993), Fageria and Baligar (2001), Fageria and Barbosa Filho (2001) and Shen et al. (2003). However, potassium increased shoot dry matter from 5386 to 5624 kg ha–¹ and biological dry matter from 9926 to 10580 kg ha–¹ in Tarrom. Shoot dry matter have been increased from 5807 to 6216 kg ha–¹ and biological dry matter from 10500 to 11420 kg ha–¹ (Table 2). The outcomes were in accordance with the achievements of Brouhi et al. (2000).

The efficiency of nitrogen application for shoot and biological dry matter in Neda (14.83 and 32.75) was greater than Tarrom (7.03 and 25.80), respectively. The same trend has governed on the efficiency of potassium application on shoot and biological dry matter, means that Neda with 1.976 and 4.553 has used potassium fertilizer more efficient than Tarrom with 1.275 and 3.503, respectively (Table 3).

Simultaneous application of nitrogen and potassium had significant effects on shoot dry matter and biological dry matter. In Tarrom, the highest amount of shoot dry matter was obtained from treatment N2K3 (6458 kg ha–¹) and the highest amount of biological dry matter was measured by 12340 kg ha–¹ in treatment N3K3. Meanwhile, in Neda, both of the highest amount of shoot and biological dry matter were obtained from treatment N3K2 (7022 and 13310 kg ha–¹, respectively) (Table 4). This finding was emphasized by Zhang and Wang (2005) that interactions between nitrogen and potassium could increase the amount of dry matter.

Harvest Index, 1000 grain weight and percentage of hollow grain: Harvest index, 1000 grain weight and percentage of hollow grain in Tarrom and Neda were significantly different in present experiment (Table 1). Nitrogen and potassium application and interaction of cultivar and nitrogen had a significant effect on these traits. Harvest index was significantly influenced by interaction of cultivar, nitrogen and potassium and percentage of hollow grain was significantly affected by nitrogen and potassium application, respectively. In Tarrom cultivar application of nitrogen increases harvest index from 0.381 to 0.493 and reduces percentage of hollow grain from 23.4 to 10.15%. Also, in Neda cultivar, harvest index has increased from 0.409 to 0.474 and percentage of hollow grain reduced from 7.81 to 5.98%. However, 1000 grain weight which was not influenced by nitrogen application in Tarrom has increased from 27.65 to 28.08 g in Neda (Table 2). Application of potassium have increased harvest index from 0.451 to 0.464 and from 0.439 to 0.455 and reduced percentage of hollow grain from 20.95 to 12.28% and from 8.61 to 5.35% in both Tarrom and Neda cultivars, respectively (Table 2). Simultaneous application of nitrogen and potassium has significantly affected on harvest index, 1000 grain weight and percentage of hollow grain. Highest amount of harvest index was obtained from treatment N2K2 with 0.499, highest 1000 grain weight was obtained from N3K0 treatment with 24.49 g and lowest percentage of hollow grain was obtained from treatment N3K3 (4.62 g) in Neda cultivar (Table 2).

The efficiency of nitrogen application for harvest index in Neda (0.56) was smaller than Tarrom (0.73). The same trend has governed on the efficiency of potassium application on harvest index, means that Neda with 0.64 has used potassium fertilizer less efficient than Tarrom with 0.68 (Table 3). The efficiencies of nitrogen and potassium were ignorable for 1000 grain weight and percentage of hollow grain in both genotypes.

Leaf nitrogen and potassium: The amount of leaf nitrogen and leaf potassium was different in both of the studied cultivars. Meanwhile, cultivar by nitrogen and cultivar by potassium interactions showed significant differences (Table 1). Increasing the potassium application increased the leaf potassium contents from 1.852 to 2.083% and from 1.745 to 2.044% in Tarrom and Neda, respectively (Table 2). Although there have been shown increasingly changes in leaf nitrogen contents by nitrogen application, none of cultivars showed significant variation. Also, non significant effects have been exhibited when nitrogen and potassium were simultaneously applied (Table 2).

CONCLUSIONS

Nitrogen and potassium application have significantly affected on economic, agronomic and physiological plant characteristics. Grain yield, shoot and biological dry matter, plant height and other components of yield are positively under influence of nitrogen or potassium application; however, hollow grain is an exception. Efficiencies of nitrogen or potassium application play different roles in various traits. Nitrogen or potassium may increase grain yield, shoot and biological dry matter, number of grain per panicle, length of panicle, length and width of flag leaf in both tested genotypes. Neda as an improved cultivar which carries a number of agronomical important genes showed better performance than Tarrom as a landrace cultivar. Application of 150 kg N ha–¹ with 225 kg K₂O ha–¹ (treatment N3K3) showed higher results in most of studied traits. Doing experiment with more genotypes and precise levels of N and K applications may result in more clear achievements.

ACKNOWLEDGMENTS

Thanks from the University of Mazandaran for funding, from staffs of Sari Agricultural Campus, Soil Science Department for their helps.