JWSS - Journal of Water and Soil Science

---

In order to determine the effects of number of irrigations on stages of development, yield, and yield components of cumin, an experiment was conducted in 1993 - 94 growing season with a randomized complete block design and four replications. The location of the experiment was Isfahan University of Technology Research Station in Shervedan Field. The irrigation treatments were: I1- Irrigation after planting I2- Irrigation after planting and at the stage of full standing of plants I3- Irrigation after planting, full standing of plants, and anthesis and I4- Irrigation after planting, full standing of plants, anthesis, and beginning of grain filling stage. The stages of development (from emergence to anthesis) occurred simultaneously in all treatments. However, the water deficit in I1 and I2 caused the plants to reach their physiological maturity 5 days sooner than the other two treatments. The time intervals between anthesis and physiological maturity in I4 and I3 treatments were 12 days and 5 days longer than those in the other two treatments, respectively. The grain yield of I1 and I2 were approximately 60% of I3 and no significant difference was observed between the grain yields of I1 and I2 treatments, nor between I3 and I4 treatments. The grain yields in I3 and I4 treatments were about 1742 and 1800 Kg/ha, respectively. The number of umbels per plant, the most important grain yield component, was greater in I3 and I4 than the other two irrigation regimes, but the number of seeds per umbel and 1000-seed weight of I3 and I4 were lower than that in I1 and I2 treatments.

---

Limited time for crop residue decomposition and complete seedbed preparation in double-cropping necessitates reduced tillage which is associated with its shortcomings. Information about seedbed preparation aspects in double-cropping of wheat and mungbean in Isfahan is lacking. For this reason, the effects of various seedbed preparation methods on growth, seed yield and yield components of mungbean (experimental line 1-61-16) were studied in a wheat-mungbean double-cropping system during 1996 at the Agricultural Research Station, Isfahan University of Technology. Two residue management treatments (burned and nonburned) along with four tillage systems: moldboard plow, disk, khischee and no-till were laid out in a strip plot design within a randomized complete block design with four replications. Improved establishment and plant growth conditions associated with burning residue treatment significantly increased plant density, above ground plant weight, number of nodes per branch, plant height, number of pods per main stem and per square meter, number of seeds per pod on main stem and on branches, number of seeds per main stem, seed yield, biological yield and harvest index. Higher plant density in this treatment resulted in significant reduction in the number of branches and number of pods per branch. Desirable conditions for establishment and growth in disk and moldboard plow treatments significantly increased plant density, above ground plant weight, number of pods per main stem and per square meter, number of seeds per pod on main stem and on branches, number of seeds per main stem, seed yield, biological yield and harvest index. No-till treatment produced the least results in respect to the above traits. The results indicated that burning residues with disk tillage system might be an appropriate seedbed preparation method under conditions similar to this experiment.

---

In order to study the genetic and environmental variability of morphological and phenological characteristics and also to gain a better understanding of the morphological basis of yield variation in soybean, an experiment was conducted at Research Farm, College of Agric., Isfahan University of Technology, in 1996. The experiment was arranged in three augmented designs with 285 lines and 5 control varieties. There were highly significant differences among genotypes for all the characteristics studied. The phenotypic coefficients of variability were greater than the genotypic ones for all the traits although the differences for most of them were small. The highest phenotypic and genotypic coefficients of variation, in order of magnitude, were obtained for number of pods per plant, days to flowering, plant height, height of the lowest pod, and number of lateral branches. Heritability estimates and the percent of genetic improvement for these traits and seed weight were high but for seed yield per plant were low. The lowest coefficients of variability were related to number of seeds per pod and days to germination. The results of stepwise regression analysis indicated that the maximum variation in seed yield could be attributed to the number of pods per plant, number of seeds per pod, and 100-seed weight. Factor analysis in addition to emphasizing the importance of yield components, resulted in four factors which determined 97.34% of yield variation. These factors which were indicators of physiological sink and source, with respect to the traits contained, were named as photosynthetic sink, plant architecture, fixed capital, and weight, respectively. In general, it was concluded that in order to improve soybean cultivars, selection should be in favor of plants with stronger structure, higher number of leaves, nods and pods per plant, and higher seed weights.

---

Obtaining high seed and oil yields of sunflower requires coincidence of vegetative and reproductive stages of growth of the plant with suitable environmental conditions via selecting appropriate planting date. Since the suitable date of planting for sunflower cultivars under Isfahan environmental conditions was not determined, this experiment was conducted in 1996 at the Agricultural Research Station, Isfahan University of Technology. In this study, five dates of planting (April 27, May 12 and 27 and June 12 and 29) and three open pollinated sunflower cultivars (Record, Vnimik 8931 and Armavirec) were evaluated using a randomized complete block design with split-plot layout in three replications. Date of planting was considered as the main plot and cultivars were randomized in the sub-plots.

Number of seeds per head (SH), l000-seed weight (SW), seed oil percent (SOP) and, consequently, yields of oil (OY) and seed (SY) were significantly reduced as planting was delayed. Reduction in these traits were considered to be related to the coincidence of vegetative and reproductive growth stages with higher temperatures prevailing at later plantings. Vnimik 8931 had higher SH and SW, and thus produced higher SY. This cultivar had lower SOP than Record, but produced higher OY than Record due to its higher SY. Armavirec was ranked the least for the measured traits, except for its SH that was slightly higher than that of Record. SH was the most contributing trait to the increase in SY, and SY was the most determining trait for the increase in OY. Date of planting by cultivar interactions for SH, SY and OY were significant. Nevertheless, the highest amount of the measured traits were obtained with Vnimik 8931 at the first planting date.

---

To determine the relationships between yield and its components, and to find the direct and indirect effects of yield-related traits on soybean yield, 240 genotypes were grown in the Research Station, College of Agriculture, Karaj, Iran, in 1997-98 using an augmented design. The study of correlation analysis showed that number of seeds per plant and seed yield per plant had the highest significant correlation coefficient (r=0.92). Other significant correlation coefficients were found between biomass per plant and yield (r=0.86) and between number of pods per plant and yield (r=0.67). Results of stepwise regression analysis revealed that number of seeds per plant, seed weight, and number of seeds per pod were the three major traits affecting seed yield in soybean. However, path analysis showed that only two of the three above-mentioned traits, namely, number of seeds per plant and seed weight, are quite important for soybean selection programs. Stepwise regression analysis was used again, omitting number of seeds per plant, which showed that number of pods per plant, seed weight, plant height and number of days to 90% maturity, are important contributors to yield. Path analysis, this time, revealed that the correlation effects to plant height and number of days to 90% maturity on yield is due to the indirect effects exerted through other traits. It is concluded that three traits, namely, number of seeds per plant, seed weight and number of pods per plant have notable effects on soybean seed yield.

---

It is believed that various types of field bean, including pinto, white and red, differ in adaptability to high temperatures and may, thus, differ in response to delay in planting. In order to evaluate this response, an experiment was conducted during 1996 at the Agricultural Research Station, Isfahan University of Technology, using a randomized complete block design with split-plot layout. Main plots consisted of four planting dates (April 28, May 13 and 28 and June 13) and sub-plots included four genotypes of common bean (red bean, c.v. Naz pinto beans, experimental lines 11816 and 16157 and a white bean, experimental line 11805).

 Number of branches per plant, number of pods per branch and per unit area, number of seeds per pod of main stem and branch, number of seeds per main stem, per branch and per unit area, 100-seed weight and seed yield significantly reduced, while harvest index significantly increased by delay in planting and consequent increases in temperature and reduction in time for growth. The lower harvest index obtained with early planting was the result of the lower efficiency of the produced vegetative growth due to the coincidence of seed filling period with high temperatures. Pinto bean line 11816 ranked the highest for number of branches per plant and harvest index among the genotypes evaluated and produced the highest seed yield (3030 kg ha^-1). Although red bean Naz ranked the highest for number of pods and seed per main stem and per unit area, it had the lowest harvest index and 100-seed weight and, consequently, produced the least seed yield (2254 kg ha^-1). The results obtained indicate that delay in planting adversely affects bean seed yield. Pinto bean line 11816 may have higher yield potential among the genotypes studied at all planting dates under conditions similar to the present experiment. No specific relationship was observed between apparent seed characteristics and plant tolerance to heat.

---

Edible-oil flax (Linum usitatissimum L.) can be an alternate oil-seed crop in Isfahan region. Seeding date is very important in obtaining higher grain yields and a successful flax production. The present study was conducted at the research farm, Isfahan University of Technology in the year 2000 to determine the effect of seeding date and genotype on yield, yield components and maturity of edible-oil flax. In this study, a RCBD with three replications, in which the treatments were organized as a split-plot experiment, was used. Seeding dates (Oct. 17, Nov. 16, March 15, April 13, May 14, June 13, July 15) and genotypes (four breeding lines of edible-oil flax) were considered as the main and the subfactor, respectively. The number of capsules per plant, seeds per capsule, 100-seed weight and seed yield were highest for all genotypes in the first seeding date. Average seed yield in the first seeding date was almost twice, triple and eight times as much as the second, third and the last three seeding dates, respectively. In the second seeding date there was no emergence because of low temperature. The genotype and seeding date interaction on seed yield and maturity was significant. In general, delayed seeding was accompanied by reduction of emergence, number of days to maturity, yield and yield components in all genotypes, with the exception that the last seeding date led to increased number of days to maturity and seeds per capsule. Based on regression analysis, seed weight, number of seeds per capsule and number of seedlings/m2 were the most important components that contributed in seed yield variation, respectively. But, most of the variations in seed yield per plant was affected by number of copsules per plant and number of seeds per capsule, and in that order.

---

Genetic variation of 300 genotypes of durum wheat comprising CIMMYT/ICARDA and Iranian germplasm was evaluated in 1999-2000 at Researh Farm of College of Agriculture, Isfahan University of Technology located at Lavark, NaJaf-Abad. Days to heading, days to maturity, plant height, spike length, grain weight per spike, number of grain per spike, number of spikes per m², 1000-grain weight, test weight, grain yield, biological yield, and harvest index were recorded. Coefficients of correlation, step-wise regression, factor analysis and cluster analysis of characters and genotypes were conducted using the studied traits. Results indicated that considerable genetic variations exist for the traits and, in particular for grain yield, harvest index, number of spikes per unit area and number of grain per spike. Grain yield had a positive and significant correlation with days to heading, days to maturity, number of grain per spike and grain weight per spike. Factor analysis for the genotypes detected 6 factors, which explained 76.7 percent of the total variation among data. These factors basically involved potential of assimilate distribution, different aspects of plant storage, source-sink relationships, plant height and tillering potential. Cluster analysis was similar to factor analysis in grouping the characters. According to cluster analysis, the genotypes were calssified into 6 clusters with significant differences among all groups. Mean comparisons of traits in these groups showed that genotypes of groups 5 and 6 were superior in grain yield and harvest index and are beneficial to the local durum breeding objectives.

---

Flax (Linum usitatissimum L.), an oilseed crop, is widely adapted and grown in many regions of the world. Oil from regular flaxseed is used as an industrial drying oil because of the high level of linolenic acid (>50 %). However, the oils from new mutant genotypes of flax with a very low linolenic acid concentration (<2 %) are edible. Yellow seed colour can be used as a visual marker to distinguish edible-oil genotypes of flax from those of industrial type that are usually brown-seeded. In this study, different lines of flax with two seed colours (yellow and brown) in combination with two levels of linolenic acid (high and low) were evaluated in a randomized complete block design for agronomic traits, especially seed yield and its components. The results indicated that lines with high linolenic acid concentration had significantly higher seed yield than those with low linolenic acid. However, other characteristics including those of seed yield components were not siginficantly affected by linolenic acid concentration. Seed colour had a significant effect on number of seedling/m², basal branches, capsules per plant and seed yield per plant. Although seedling emergence was lower in yellow-seeded lines, they had more basal branches, capsules per plant and seed yield per plant. Higher seed yield per plant in yellow-seeded lines can be attributed to higher number of capsules per plant as a result of lower seedling emergence and plant density. Seed yield was not significantly different between brown and yellow-seeded lines. Thus, the effect of lower plant density in yellow-seeded lines was compensated by their higher basal branches and number of capsules per plant.

---

Planting pattern through changing vegetative growth and utilization of environmental resourses affects yield components and seed yield. These effects were studied in the spring of 2000 at the Agricultural Research Station, Isfahan University of Technology, using a randomized complete block design with a split plot layout and three replications. Main plots consisted of three row distances (30 cm flat, and 45 and 60 cm on bed), and sub-plots included three planting densities (30, 40, and 50 plants m^-2). The experiment was planted on March 13 using local safflower variety of Isfahan, named Koseh. Increase in row distance and plant density enhanced most developmental stages of safflower. Leaf area index was not significantly affected by row distance but increased as planting density increased. Up to the 50% flowering stage, 30 cm row distance produced highest plant dry weight, but had the least dry weight at the end of the growing season, probably due to the strong shedding of leaves. Up to the end of flowering stage, 50 plants m^-2 treatment produced the highest plant dry weight, but ranked lower as compared to 30 plants m^-2 treatment at the physiological maturity, probably due to the shedding of leaves. Row distance had no significant effect on number of branches per plant and per square meter, number of heads per branch, number of seeds per head, 1000-seed weight and harvest index. But number of heads per plant and per square meter as well as seed yield per plant and per square meter significantly decreased as planting distance increased. Plant density had no significant effect on number of branches per plant, number of heads per square meter, number of seeds per head, 1000-seed weight and seed yield. Increase in planting density increased number of branches per square meter, but reduced number of heads per branch and per plant, seed yield per plant and harvest index. The highest seed yield (4769 kg ha^-1) was obtained with 30 cm row distance and 40 plants m^-2 treatment. On average, 397 kg ha^-1 petal was harvested, which has a large economic value. However, petal clipping, over all treatments, reduced seed yield by about 7.4%. Considering the advantages of uniform distribution of plants and adaptation of safflower to flat planting, 30 cm row distance with 40 plants m^-2 might be appropriate for planting safflower under conditions similar to this experiment.

---

In order to investigate agronomic traits and yield potential of edible-oil flax (Linum usitatissimum L.) as a second crop in Isfahan region, different genotypes were evaluated in separate experiments in early spring (April, 7) and summer (July, 16) planting dates, using a Randomized Complete Block Design with three replications. The experiments were conducted in agricultural research farm, Isfahan University of Technology in the year 2000. The results showed that summer planting considerably and significantly reduced number of seedlings per unit area, days to 50% flowering, and seed yield. However, maturity of the plants was delayed because of summer planting. According to overall average obtained for genotypes, seed yield was 1472 and 213 kg/ha in the first and second planting dates, respectively. There was a significant difference between genotypes for number of seedlings per unit area, days to 50% flowering and maturity in both planting dates. However, genotypes were significantly different for seed yield in the first planting date. Summer planting also non-significantly reduced yield/plant, capsules/plant and 100-seed weight and increased seeds/capsule. The differences between genotypes for these traits in both planting dates and for seeds/capsule in the first planting date were significant. The significant interaction between genotypes and planting dates on seeds/capsule and 100-seed weight was because of increasing or decreasing levels of these traits in some genotypes when planting date was delayed. The results of regression analysis, correlation coefficients and path analysis showed that in both planting dates, capsules/plant followed by seeds/capsule and 100-seed weight were the major components of yield/plant. Capsules/plant had the most (approximately 80%) contribution in variation of yield/plant in both planting dates. The number of plants per unit area affected yield/plant via indirect and negative effect of capsules/plant.

---

Despite its economic importance, not much information exists on growth and productivity of durum wheat in Iran and Isfahan. In order to evaluate growth, grain yield, and attributes in three durum wheat genotypes, Osta/Gata, Dipper-6, and PI40100, with five planting densities of 200, 275, 350, 425 and 500 seeds/m², a four-replicate RCBD field study with a spilt-plot layout was carried out in winter 2000 to spring 2001 at Lavark Research Farm, College of Agriculture, Isfahan University of Technology. Genotypes were the main-plots and planting densities, the sub-plots. Each plot contained six rows, spaced 25cm apart, each 6m in length. While Osta/Gata produced a greater spikes/m², 1000-grain weight and grain yield, PI40100 indicated a greater height, leaf area index (LAI) and biological yield, and Dipper-6 produced a larger number of grains per spike and harvest index, compared to the other two genotypes. Increasing plant density led to an increase in LAI, spikes/m², grain yield, biological yield and harvest index but a decrease in grains/spike and 1000-grain weight. Grain yield was positively correlated with spikes/m², grains/spike and 1000-grain weight, despite negative correlations with LAI and plant height. It may be concluded that with environmental conditions such as those encountered in Isfahan in 2000-2001, Osta/Gata can be planted at approximately 425 seeds/m², if an acceptable grain yield is to be achieved.

---

Although safflower is known to be a cool-season crop, it is usually planted as a summer crop in Isfahan. Thus, an experiment was conducted in 2000 at the Agricultural Research Station, Isfahan University of Technology, to study the effects of date of planting on growth, yield components, and seed yield of safflower. Five planting dates (March 12, April 12, May 10, June 8, and July 12) and four safflower genotypes (Arak 2811, local variety Koseh, Nebraska 10, and Veramin 295) were evaluated using a randomized complete block design with split-plot layout in three replications. Date of planting was considered as the main plot and cultivars were randomized in sub-plots. Delay in planting from March 12 to may 10 reduced plant dry weight per unit area, number of heads per plant, number of seeds per head, seed yield per unit area, harvest index and petal yield. The above traits increased as planting was further delayed from May 10 to June 8. Highest seed oil and lowest seed protein contents were also obtained for this planting date. Plants of July 12 planting date did not reach physiological maturity. Among the genotypes evaluated and over planting dates, the highest and lowest number of heads per plant, 1000-seed weight, and seed yield were produced byArak 2811 and Veramin 295 (mean of the first and second planting dates), respectively. Highest seed yield (4512 kg ha

^-1) was produced by local variety Koseh in June 8 planting date. It might be concluded that this variety has adapted to the summer planting conditions of Isfahan by natural selection.

---

An experiment was conducted at Rasht Rice Research Institute, in 2001 to study trait relationships and to determine the direct and indirect effects of different characteristics on grain yield of rice (Orgza sativa L.). A 9x2 factorial experiment in complete randomized block design with 3 replications was used. The two factors were 9 cultivars, namely: Hashemi, Ali Kazemi, Binam, Sepeedroud, Nemat, Khazar, Taichung, Chanto, and Usen, and 2 planting patterns (15x15 cm and 30x30 cm). Significant differences were detected among cultivars, between planting patterns, and their interaction effects. The direct effect of the number of panicles/m² on grain yield was positive and significant in both planting patterns and their combined analysis. Heading date had a positive and significant direct effect on the number of panicles/m² in both planting patterns and their combined analysis. Only in 30x30 cm planting pattern, a positive correlation was found between biomass at heading date and the number of panicles/m². Grain weight per plant had a direct and positive effect on panicle weight in the 3 cases studied. Also, in all the cases studied, flag leaf area had positive direct effects on grain weight per plant. Grain filling rate and effective grain filling period increased in 30x30 cm planting pattern. Native cultivars demonstrated longer latent period. Grain filling rate and effective grain filling period showed positive direct effects on grain yield in both plantings. According to the results obtained, an increase in the number of panicles/m², rate of grain filling, and effective grain filling period would enhance grain yield. Selection for heading date, grain yield per plant, and flag leaf area would also indirectly increase grain yield.

---

A diallel analysis of eight bread wheat (Ttriticum aestivum L.) cultivars was conducted to determine genetic parameters and the type of genetic control for yield and yield components. The parents and their F1 hybrids were evaluated in a randomized complete block design with 3 replications. General and specific combining ability effects were estimated by the method 2 of the Griffings model I, and the genetic parameters were estimated by the Jinks-Hayman method. According to the analysis of variance, the variances of parents and crosses were significant for all of the traits, except for the fertile tillers per plant in parents and harvest index in crosses. The variance of GCA was significant for almost all of the traits. GCA to SCA ratio indicated a large additive effects for all the traits, except for the number of fertile tillers, grain yield per plant and biological yield. Darab and Chamran cultivars were the excellent general combiners for days to heading. However, Falat for plant height, Arvand for number of fertile tillers, spike length and grain weight per spike, Qods and Arvand for spikeletes per spike, Falat and Arvand for grain yield per plant and Qods and Arvand cultivars for biological yield were the best general combiners. Based upon the Jinks-Hayman method, the average degree of dominance for grain weight per spike, 1000- grain weight, grain yield per plant and biological yield indicated that these traits might be controlled by over dominance effects and other traits by partial dominance. The correlation between Yr and (Wr + Vr) for the flag leaf length, spikeletes per spike, grains per spike and 1000-grain weight indicated that recessive allels enhanced these traits. But dominant allels had their contribution to enhance other traits. The narrow-sense heritability for grain yield per plant (4%) and biological yield (14%) and harvest index (37%) was lower. However, for other traits it was more than 50%. Therefore, it can be inferred that indirect selection of grain yield through selection for yield components such as spikeletes per spike, grains per spike and grain weight per spike with high heritability and correlation with grain yield can be more effective.

---

In order to evaluate the effects of plant densities (10, 13, 20 and 40 plant/m² ) on the yield and yield components of two cultivars (Partow, Gohar) and a line (VC-1973A) of mungbean [Vigna radiata (L.) Wilczek] a field experiment was conducted at the Seed and Plant Improvement Institute (SPII) of Karaj in Summer 1998. This research carried out in a factorial experiment based on Randomized Complete Block Design (RCBD) with 4 replications. The results indicated that VC-1973A line had the highest grain yield. This line has superiority to other cultivars, due to the early and uniformity of seed maturity and easy mechanized harvest. Plant density has a significant effect (p<0.01) on the grain yield, in such a way that the densities of 20 and 10 plant/m² outyielded the highest (2221 kg/ha) and lowest (1650 kg/ha) grain yield respectively. Among the yield components, only the number of pods per plant indicated a significant difference due to the plant density. Correlation coefficient between attributes indicated that the plant height and height of first pod internode had a positive correlation with the plant density, while the correlation of the plant density with harvest index (HI), branch number and pod number per plant was negative. More over it seems that the pod number per plant is the most important components of mungbean’s yield, because it has a high (r = 0.88) and significant (p<0.01) correlation with the grain yield.

---

Under irrigation and in double-cropping system, a large amount of plant residue remains after harvest that along with the limited time for residue decomposition and complete seedbed preparation, necessitates reduced tillage and special residue management. In the present study, the effects of various seedbed preparation methods on vegetative growth, yield components and seed and oil yields of sunflower (Euroflor hybrid) were studied in a barley-sunflower cropping system during 2001 at the Agricultural Research Station, Isfahan University of Technology. Three residue management treatments (standing, partly removed and burned) along with five tillage systems (moldboard + disk chisel + disk disk moldboard and furrower as the minimum tillage) were laid out in a split-block design with three replications. Burning residue treatment significantly increased plant dry weight at various developmental stages and also head diameter. Number of seeds per head, 1000 seed weight, harvest index, and oil yield were non-significantly higher in the burned residue treatment. Seed yield was significantly higher in the burned and partly removed residue treatments. Moldboard + disk and chisel + disk treatments significantly produced higher plant dry weight at various developmental stages, head diameter and seed yield. Number of seeds per head, 1000 seed weight, harvest index, and oil yield were non-significantly higher in these treatments. Minimum tillage ranked the lowest for these traits. The results of this experiment indicate that chisel + disk treatment with the partly removed residues might be an appropriate seedbed preparation method in a barley-sunflower double planting under conditions similar to this experiment.

---

In order to investigate path analysis and study correlations among 16 agronomic traits, 49 Iranian and foreign rice entries were evaluated in a simple lattice square as experimental design with 2 replications at research farm of Tonekabon agricultural college in 2001. Analysis of variance showed significant difference among genotypes for all the studied traits (p<0.05). Phenotypic and genotypic correlation analysis showed positive and significant correlation between grain yield and the number of productive tiller, total tillers and number of grain per panicle. Based on path analysis of the traits, the number of productive tillers had the highest direct effect on the grain yield. Furthermore, the number of grain per panicle and 100-grain weight had a high direct effect on the yield. Overall, the results are suggestive of the direct selection for grain yield in these genotypes through above - mentioned traits.

---

In order to investigate the effect of different irrigation regimes on some agronomic and physiological characters of three chickpea cultivars (Jam, 301 and Pirooz), a field experiment was conducted in 1998 at the Agricultural Research Farm of Tabriz University, Tabriz, Iran. A split plot experiment based on a randomized complete block design with three replications was used, in which irrigation treatments (full irrigation, irrigation at only branching or flowering and or pod formation stage) were in the main plots and chickpea cultivars were in the sub plots. Under limited irrigation conditions there were no significant differences among irrigation regimes for green cover percentage and the number of pods per plant while rate and duration of grain filling, grain weight and grain yield were significantly higher for irrigation at pod formation than for irrigation at branching or flowering stage. The mean of all the traits, except for the number of seeds per pod were significantly higher for full irrigation than for limited irrigation treatments. Rate of grain filling, maximum grain weight and grain yield were significantly higher for irrigation at flowering than for irrigation at branching stage. While, other characters were not significantly different between these two treatments. Although, green cover percentage and seeds per pod in 301 were higher than those in Jam, but in other cases their response to water limitation was almost similar. All traits, except the number pods per plant, were noticeably lower for Pirooz compared to Jam and 301 cultivars. Green cover percentage showed the highest correlation with the grain yield. This study revealed that among phenological stages of chickpea, pod formation is the most sensitive to water deficit, and that under water limitation conditions chickpea yield could be improved by irrigation at this stage.

---

Effect of nitrogen (N) fertilizer levels and its split applications on yield and yield components of rice (Oryza sativa L.) Cv. Khazar was investigated in a completely randomized block design with 3 replications in a paddy light soil at Guilan province, Iran, 2003. In this experiment, six treatments including: T_1-control (no N fertilizer) T_2- 40 kg/ha N (at transplanting time) T_3- 80 kg/ha N (at transplanting, and tillering times) T_4- 80 kg/ha N (at transplanting, tillering, and panicle initiation times) T_5- 120 kg/ha N (at transplanting, and tillering times) and T_6- 120 kg/ha N (at transplanting, tillering, and panicle initiation times) were compared. Results showed that the highest fertile tiller number was obtained in the fifth and sixth treatments with double and triple split applications of 120 kg/h N (236 and 248 m^-2). The highest fertile filled spikelets percentage (84.8%), 1000-grain weight (26.1 g) and grain yield (4.83 t/ha) belonged to the sixth treatment, but grain yield and 1000-grain weight were not significantly differerent in the fourth and sixth treatments with three fertilizing times. This finding may have resulted from the third topdressing application of nitrogen fertilizer in panicle initiation and higher leaf area (44.8 and 45.5 Cm²), leaf greenness (39.4 and 39.9) and leaf nitrogen concentration (31.2 and 33.6 g/kg) during grain filling in the fourth and sixth treatments. Regression analysis also showed that flag leaf greenness (SPAD values at 5 days after flowering) and flag leaf area accounted for about 75% and 78% changes in yield, respectively. In conclusion, triple split application of 80 kgN/ha could be suggested for rice Cv. Khazar in these regions since the yield would be the same as the application of 120Kg/ha N.