Characterization of high-yield performance as affected by genotype and environment in rice

Characterization of high-yield performance as affected by genotype and environment in rice 1 first-author 1 2 1 corresp ‡Corresponding Author ‡Corresponding Author †E-mail:zhanggp@zju.edu.cn †E-mail:zhanggp@zju.edu.cn zhanggp@zju.edu.cn Agronomy Department, Zhejiang University, Hangzhou 310029, China Agronomy Department, Zhejiang University, Hangzhou 310029, China Shanghai Academy of Agricultural Sciences, Shanghai 201106, China Shanghai Academy of Agricultural Sciences, Shanghai 201106, China We characterized yield-relevant characters and their variations over genotypes and environments (locations and years) by examining two rice varieties (9746 and Jinfeng) with high yield potential. 9746 and Jinfeng were planted in two locations of Shanghai, China, during 2005 and 2006. The results show that there was a large variation in grain yield between locations and years. The realization of high yield potential for the two types of rice was closely related to the improved sink size, such as more panicles per square meter or grains per panicle. Stem and leaf biomasses were mainly accumulated from tillering stage to heading stage, and showed slow decline during grain filling. Meanwhile, some photosynthetic characters including net photosynthesis rate ( P n ), leaf area index (LAI), specific leaf area (SLA), fluorescence parameter (maximum quantum yield of PSII, F v / F m ), chlorophyll content (expressed as SPAD value), as well as nutrient (N, P, K) uptake were also measured to determine their variations over genotypes and environments and their relationships with grain yield. Although there were significant differences between years or locations for most measurements, SLA at tillering and heading stages, F v / F m and LAI at heading stage, stem biomass at heading and maturity stages, and leaf nitrogen concentration at tillering and heading stages remained little changed, indicating their possible applications as selectable characters in breeding programs. It was also found that stem nitrogen accumulation at tillering stage is one of the most important and stable traits for high yield formation. Environment Photosynthesis Nutrition Rice Yield INTRODUCTION 1 Rice ( Oryza sativa L.) is one of the most important food crops in the world. Yield potential of modern rice cultivars in irrigated tropical areas has been greatly increased since the first semi-dwarf tropical indica cultivar, IR8, was released three decades ago (Khush et al., 2001 2001 ), which, to a great extent, alleviated the pressure of food shortage caused by increasing population. However, food security is still a serious issue facing the world, and there is a need for more food production to keep pace with increasing world population. The demand for increasing rice production is particularly urgent, because the population of traditional rice-producing countries will require 70% more rice by year 2025 (IRRI, 1995 1995 ; Swaminathan, 2007 2007 ). Hence, the world rice production must increase by approximately 1% annually to meet the growing demand (Rosegrant et al., 1995 1995 ). Recently, so-called super-rice cultivars or hybrids have been released in China, which have shown a markedly high yield potential when planted in the special areas (Amano et al., 1996 1996 ; Cheng and Min, 2001 2001 ). However, these super-rice cultivars or hybrids are commonly characterized by unstable yields across environments (locations and years) (Horie et al., 1997 1997 ; Ying et al., 1998 1998 ). Therefore, it is imperative to clarify the responses of their growth and yield formation to environments. Understanding of the physiological processes associated with these responses may provide effective information for improving breeding and agronomy practices. There have been some reports on determination of the physiological traits closely linked with yield formation. Horie et al.( 1997 1997 ) suggested the importance of increasing sink size and radiation use efficiencies during grain filling. Ying et al.( 1998 1998 ) reported that further improvement in rice yield potential depended more on the ability of increasing biomass production than on that of increasing harvest index (HI). As a matter of fact, in the breeding programs focused on improving plant type at IRRI (International Rice Research Institute, Philippines), large panicle and high biomass production are considered as the main selection criteria (Khush et al., 2001 2001 ). It is well known that final yield (phenotype) of a given cultivar is dependent on both genetic factors (genotype) and growth conditions (environment). Much effort has been done in breeding to improve genetic potential of rice cultivars or hybrids by optimizing plant type and light utilization. However, little research has been done in understanding the effect of environmental factors on the performance of physiological traits in super-rice cultivars or hybrids. The objectives of the current experiments were to determine the relationship between yield formation and environments, and to clarify the physiological traits closely linked with high yield under field conditions. MATERIALS AND METHODS 1 Experimental sites and soil properties 2 Field experiments were conducted at experimental stations of Qingpu and Jinshan, Shanghai (30°53′ N, 120°08′ E, and altitude 6 m) in 2005 and 2006. The two locations, around 35 km apart, have the same climate. Basic soil fertility in these two locations is shown in Table 1 1 . On the whole, year 2005 had more total precipitation and lower mean temperature than year 2006. A japonica hybrid Jinfeng and a japonica cultivar 9746 were used. Location Organic C (g/kg) Available P (mg/kg) Available K (mg/kg) Available N (g/kg) pH Jinshan 1.14 12.39 68.27 3.55 7.35 Qingpu 0.65 21.23 54.42 1.45 5.97 Rice seeds were immersed in water for 2 d before sowing, and were sown on May 25 and transplanted on June 20 as single season rice (mono-cropping), and hill space was 25 cm×20 cm with one plant per hill. A completely random block design with 3 replications was used and plot area was 50 m 2 . N fertilizer was supplied in the form of urea at a rate of 270~300 kg N/ha, with 30% of total N being applied before seeding, 30%, 20% and 20% top-dressed at tillering, booting and heading stages, respectively. K fertilizer was applied in the form of potash chloride at a rate of 370~450 kg/ha, with 70% being applied before transplanting and 30% top-dressed at booting stage. During growth, weeds, insects and diseases were chemically controlled as required. Measurements of photosynthesis, fluorescence and SPAD value 2 The plant sampling and measurements were carried out at mid-tillering (35 d after transplant), booting (56 d after transplant) and heading (82 d after transplant) stages. Net photosynthetic rate ( P n ), stomatal conductance and transpiration rate were measured with a LI-6400 portable photosynthesis system (LI-COR, Lincoln, NE, USA) on the upper-most fully expanded leaves. Meanwhile, chlorophyll fluorescence was measured using a portable pulse-modulated fluorometer (mini-PAM, H. Walz, Germany) and chlorophyll content, expressed as SPAD value, was determined using an SPAD meter (SPAD-502, 1989 Minolta Co. Ltd., Japan). P n and chlorophyll fluorescence were determined on 6 and 10 leaves respectively for each replication, while SPAD measurement was done on 15 leaves. Biomass and nutrient analysis 2 Fifteen plants were harvested at each sampling stage and used for biomass and nutrient analyses. The plants were separated into different parts, dried in an oven at 70 °C and weighted. N concentration in plant tissue was determined with Kjeldahl method. K concentration was determined by atomic absorption spectrophotometer and P concentration by ammonium-vanadomolybdate method. Nutrient accumulation was calculated by product of nutrient concentration and tissue biomass (dry weight). At maturity stage, all plants of each plot were harvested for yield determination (expressed as grain yield with water content of 14%). Statistical analysis 2 All data were subjected to ANOVA using statistical software SAS 8.0 for Windows, and comparisons between the treatments with P < 0.05 were considered significantly different. RESULTS 1 Yield and its components 2 Table 2 2 shows grain yield and its components for all the treatments. Grain yield varied between cultivars and environments (locations and years), ranging from 8.61~10.68 t/ha for 9746 and 7.61~10.58 t/ha for Jinfeng. There was a significant difference in grain yield between years or locations, with the yield in 2006 being higher than that in 2005, and the one of Qingpu being higher than that of Jinshan. In addition, the higher yield of 9746 was associated with more panicles per square meter and greater grain weight. In contrast, higher yield of Jinshan is attributed to more grains per panicle. Obviously, the yield components showed the great changes between years. Thus, panicles per square meter, the percentage of filled grains and grain weight were greater in 2006 than in 2005, but spikelet per panicle was just opposite. The difference in yield components between the two locations was relatively small. Means followed by different letters in the same column are significantly different at the P < 0.05 level according to Turkey test. Var.: Variety; Loc.: Location; NS: Nonsignificant; S: Significant Panicles (m −2 ) Grains per panicle Percentage of filled grains (%) Grain weight (mg) Yield (t/ha) Variety 9746 243.29a 184.32b 91a 29.02a 9.97a Jinfeng 223.48b 205.45a 92a 24.30b 9.01b Location Jinshan 234.11a 190.82b 91a 26.71a 9.36b Qingpu 232.66a 199.50a 93b 26.61a 9.62a Year 2005 221.49b 199.60a 91b 26.07b 9.10b 2006 245.28a 190.18b 93a 27.24a 9.88a Interaction Var.×loc. NS NS NS NS NS Var.×year NS S NS NS NS Biomass accumulation 2 Stem biomass reached the maximum at heading stage and then declined during grain filling (Table 3 3 ). The two rice cultivars showed obvious difference in the change of stem biomass: 9746 had little increase after booting stage, whereas Jinfeng had a marked increase from booting stage to heading stage. The change of leaf biomass was basically similar for the two cultivars, with the maximum value occurring at heading stage and slight reduction during grain filling. Although grain biomass was mainly accumulated during grain filling for the two cultivars, there was a distinct difference between the cultivars in the relative proportion of biomass accumulated during grain filling, with the one of 9746 being much less than that of Jinfeng. Means followed by different letters in the same column are significantly different at the P < 0.05 level according to Turkey test. MT: Mid-tillering stage; BS: Booting stage; HS: Heading stage; MS: Maturity stage; Var.: Variety; Loc.: Location; NS: Nonsignificant; S: Significant 2 2 2 MT 2 BS 3 HS 3 MS Stem Leaf Stem Leaf Stem Leaf Panicle Stem Leaf Panicle Variety 9746 109.99a 107.13a 350.32a 248.41a 397.66b 263.77a 450.61a 356.68b 227.46b 742.31a Jinfeng 103.29a 103.38a 324.07b 248.23a 498.95a 254.24a 344.63b 439.42a 253.93a 839.41a Location Jinshan 102.50a 98.67a 356.07a 257.58a 456.60a 260.74a 428.24a 392.26a 243.43a 847.91a Qingpu 109.79a 110.70a 319.08b 240.28a 447.87a 256.74a 364.01b 408.58a 240.09a 739.22b Year 2005 72.11b 82.96b 251.07b 211.30b 428.44a 229.47b 391.57a 389.60b 223.64b 758.37b 2006 140.62a 127.23a 421.13a 285.32a 476.61a 287.75a 394.83a 413.40a 259.96a 831.44a Interaction Var.×loc. NS NS S S NS NS S NS NS NS Var.×year S S NS S NS S NS NS NS NS The environmental effect on the tissue biomass accumulation varied with years and locations. For stem biomass, little difference was found after heading stage regardless of years or locations, but a significant difference was found at mid-tillering stage between years, with the stem biomass in 2006 being greater than that in 2005, and at booting stage between years and locations, with the one in 2006 being greater than that in 2005, and the one of Qingpu being greater than that of Jinshan. For leaves, the biomass in 2006 was constantly higher than that in 2005 and no significant difference was found between locations through the whole growth stages, except for the leaf biomass of Qingpu being greater than that of Jinshan at mid-tillering stage. For panicles, the biomass of Jinshan was higher than that of Qingpu at both heading and maturity stages. Photosynthetic rate, fluorescence parameter and SPAD value 2 Photosynthetic rate ( P n ) of leaf at mid-tillering stage was significantly higher than that at heading stage, while for fluorescence parameter (maximum quantum yield of PSII, F v / F m ), the opposite was observed (Table 4 4 ). However, there was no significant difference in SPAD value between the two measurements. Photosynthetic rate ( P n ) differed significantly between the two cultivars at heading stage, with P n of 9746 being higher than that of Jinfeng, and there was a significant difference between years, with P n in 2005 being higher than that in 2006. For F v / F m , significant differences were not detected between cultivars, locations and years at heading stage, but were observed at mid-tillering stage. The F v / F m ratios of 9746, Jinshan and year 2005 were significantly higher than those of Jinfeng, Qingpu and year 2006, respectively. For SPAD value, no significant difference was found between cultivars or locations, whereas the difference was significant between years, with SPAD in 2005 being higher than that in 2006, irrespective of growth stage. Means followed by different letters in the same column are significantly different at the P < 0.05 level according to Turkey test. MT: Mid-tillering stage; HS: Heading stage; Var.: Variety; Loc.: Location; NS: Nonsignificant; S: Significant 2 2 2 P n (µmol CO 2 /(m 2 ·s)) 2 F v / F m 2 SPAD 2 SLA (dm 2 /g) 2 LAI MT HS MT HS MT HS MT HS MT HS Variety 9746 24.63a 18.65a 0.77a 0.80a 43.93a 43.73a 2.34a 1.99a 2.85a 6.51a Jinfeng 22.87a 17.11b 0.69b 0.79a 44.33a 44.29a 2.25a 2.17a 2.56a 5.45b Location Jinshan 24.54a 19.12a 0.76a 0.79a 43.20b 44.17a 2.19a 2.09a 2.43b 6.04a Qingpu 22.96a 16.64b 0.70b 0.80a 45.05a 43.85a 2.39a 2.07a 2.98a 5.91a Year 2005 25.57a 19.21a 0.75a 0.79a 44.85a 47.52a 2.29a 2.18a 2.15b 5.74a 2006 21.93b 16.56b 0.71b 0.79a 43.40b 40.50b 2.28a 1.98a 3.25a 6.22a Interaction Var.×loc. NS NS S NS NS S NS NS NS NS Var.×year S NS S S S NS NS S S S Specific leaf area (SLA) and leaf area index (LAI) 2 SLA showed obvious changes with the growth stages, and the higher changes were observed in mid-tillering stage compared with the heading stage. However, no significant difference was found between cultivars, locations or years, indicating that it is a relatively stable character. For LAI, the two cultivars showed no significant difference at mid-tillering stage, but at heading stage, 9746 displayed significantly higher LAI than Jinfeng. The difference in LAI between locations or years was found at early stage, and became smaller at heading stage. Nutrient concentration and accumulation 2 The nutrient concentrations in different plant parts of the two rice genotypes under the two locations are shown in Table 5 5 . N concentration in leaves and stems of the two genotypes showed dramatic decreases from mid-tillering stage to booting stage, remained less changed from booting stage to heading stage, and then declined rapidly during grain filling. N concentration in panicles remained relatively constant or had a slight decrease during grain filling. The difference in tissue N concentration between genotypes, locations and years varied with growth stages. At mid-tillering and heading stages, there was no significant difference in leaf N concentration between genotypes, locations and years. However, the difference was significant between genotypes and years at both booting and mid-tillering stages. The 9746 had constantly higher N concentration than Jinfeng, but for panicle N concentration, the opposite was observed. Means of the same nutrient followed by different letters in the same column are significantly different at the P < 0.05 level according to Turkey test. MT: Mid-tillering stage; BS: Booting stage; HS: Heading stage; MS: Maturity stage; Var.: Variety; Loc.: Location; NS: Nonsignificant; S: Significant 2 3 10 N, P, K concentrations (%) 2 MT 2 BS 3 HS 3 MS Stem Leaf Stem Leaf Stem Leaf Panicle Stem Leaf Panicle Variety 9746 2.02a, 0.45a, 1.96b 3.89a, 0.36a, 1.32a 1.25a, 0.52a, 1.60b 2.72a, 0.37a, 1.37a 1.11a, 0.48a, 2.44a 2.30a, 0.38a, 1.28a 1.42a, 0.32a, 0.49b 0.72a, 0.19a, 1.84a 1.12a, 0.13a, 0.71b 0.89b, 0.30a, 0.24a Jinfeng 1.79b, 0.41a, 2.29a 3.69a, 0.36a, 1.46a 1.18a, 0.53a, 2.02a 2.53b, 0.34a, 1.43a 0.89b, 0.43a, 2.09b 2.44a, 0.35a, 1.44a 1.34b, 0.27b, 0.55a 0.60b, 0.14b, 1.62b 0.87b, 0.14a, 0.88a 0.98a, 0.29a, 0.24a Location Jinshan 1.91a, 0.43a, 2.20a 3.86a, 0.36a, 1.41a 1.22a, 0.47b, 1.82a 2.60a, 0.32b, 1.49a 0.96a, 0.44a, 2.33a 2.29a, 0.34a, 1.42a 1.42a, 0.29b, 0.50b 0.63a, 0.18a, 1.67a 0.91b, 0.13b, 0.75a 0.80b, 0.30a, 0.23b Qingpu 1.90a, 0.43a, 2.05a 3.71a, 0.35a, 1.37a 1.21a, 0.59a, 1.79a 2.66a, 0.39a, 1.31b 1.04a, 0.48a, 2.19a 2.46a, 0.38a, 1.30a 1.33b, 0.31a, 0.55a 0.69a, 0.16b, 1.80a 1.07a, 0.15a, 0.84a 1.07a, 0.29a, 0.25a Year 2005 2.05a, 0.47a, 2.21a 3.89a, 0.43a, 1.34a 1.35a, 0.49b, 2.06a 3.08a, 0.39a, 1.62a 1.01a, 0.38b, 2.41a 2.46a, 0.25b, 1.65a 1.29b, 0.24b, 0.62a 0.65a, 0.17a, 1.78a 1.00a, 0.13a, 0.87a 0.95a, 0.32a, 0.32a 2006 1.76b, 0.39b, 2.04a 3.63a, 0.29b, 1.45a 1.07b, 0.58a, 1.56b 2.17b, 0.32b, 1.19b 1.00a, 0.53a, 2.11b 2.28a, 0.46a, 1.07b 1.49a, 0.36a, 0.42b 0.67a, 0.16a, 1.68a 0.98a, 0.14a, 0.72b 0.92a, 0.29b, 0.16b Interaction Var.×loc. S, S, S S, NS, NS NS, S, NS S, S, NS NS, S, NS S, NS, S NS, NS, NS S, NS, S S, NS, S S, NS, NS Var.×year NS, S, S NS, NS, S NS, S, S NS, NS, S NS, S, S S, NS, S NS, NS, NS S, NS, NS S, S, NS NS, NS, S For P concentration, there was no significant difference between the two genotypes in leaves through the whole growth duration and in stems before maturity stage. At maturity stage, 9746 had higher P concentration than Jinfeng. The difference between locations was dependent on growth stages. At mid-tillering and heading stages, no significant difference was detected in both leaves and stems, while at booting and maturity stages, the difference was significant. Relatively, the difference between years was more obvious. Thus at heading stage, the significant difference was found in both leaves and stem. For K, there is no significant difference in leaf K concentration between the two genotypes until maturity, while difference in stem K concentration between the two genotypes was significant in all growth stages, with Jinfeng having higher values at mid-tillering and booting stages and smaller values at heading and maturity stages. The significant difference between the locations was only found for leaf at booting stage and for panicle at heading stage. On the whole, K concentration was found dramatically higher in 2005 than that in 2006. Table 6 6 shows the nutrient accumulation in different plant parts of the two genotypes under different conditions. For N, the difference in stem between the genotypes was only found at mid-tillering stage, whereas in the leaves the difference was significant at booting and maturity stages. It may be suggest that more N in leaf and stem was remobilized and transferred into panicle for Jinfeng than for 9746 from booting stage to maturity stage. Means of the same nutrient followed by different letters in the same column are significantly different at the P < 0.05 level according to Turkey test. MT: Mid-tillering stage; BS: Booting stage; HS: Heading stage; MS: Maturity stage; Var.: Variety; Loc.: Location; NS: Nonsignificant; S: Significant 2 3 10 N, P, K accumulations (g/m 2 ) 2 MT 2 BS 3 HS 3 MS Stem Leaf Stem Leaf Stem Leaf Panicle Stem Leaf Panicle Variety 9746 2.19a, 0.48a, 2.15a 4.14a, 0.38a, 1.43a 4.27a, 1.86a, 5.58a 6.66a, 0.90a, 3.41a 4.45a, 1.95b, 9.75b 6.08a, 0.98a, 3.42a 6.49a, 1.48a, 2.21a 2.58a, 0.70a, 6.56b 2.54a, 0.31b, 1.63b 6.58b, 2.25a, 1.78b Jinfeng 1.80b, 0.43b, 2.26a 3.81a, 0.35a, 1.50a 3.75a, 1.80a, 6.02a 6.11b, 0.84b, 3.40a 4.54a, 2.21a, 10.47a 6.07a, 0.95a, 3.43a 4.65b, 0.98b, 1.90b 2.69a, 0.64b, 7.18a 2.25b, 0.35a, 2.21a 8.02a, 2.44a, 1.98a Location Jinshan 1.92b, 0.43b, 2.21a 3.79a, 0.34a, 1.39a 4.29a, 1.81a, 6.05a 6.53a, 0.81b, 3.78a 4.38a, 2.05a, 10.50a 6.02a, 0.93a, 3.63a 6.17a, 1.30a, 2.12a 2.47b, 0.73a, 6.51b 2.24b, 0.31b, 1.86a 6.97b, 2.57a, 1.97a Qingpu 2.03a, 0.48a, 2.21a 4.11a, 0.38a, 1.53a 3.74a, 1.85a, 5.61a 6.22a, 0.91a, 3.03b 4.59a, 2.13a, 9.81a 6.12a, 1.00a, 3.25a 4.93b, 1.14b, 1.98b 2.79a, 0.61b, 7.23a 2.51a, 0.35a, 2.01a 7.68a, 2.16b, 1.81b Year 2005 1.49b, 0.35b, 1.54b 3.23b, 0.36a, 1.09b 3.42b, 1.21b, 5.11b 6.53a, 0.83b, 3.43a 4.27a, 1.64b, 10.43a 5.57a, 0.59b, 3.73a 5.13b, 0.98b, 2.41a 2.53a, 0.67a, 6.91a 2.21b, 0.30b, 1.96a 7.13b, 2.47a, 2.46a 2006 2.46a, 0.56a, 2.88a 4.69a, 0.36a, 1.84a 4.57a, 2.45a, 6.52a 6.19b, 0.91a, 3.39a 4.72a, 2.55a, 9.84a 6.57a, 1.34a, 3.11b 5.86a, 1.44a, 1.68b 2.75a, 0.67a, 6.89a 2.55a, 0.37a, 1.92a 7.59a, 2.23a, 1.31b Interaction Var.×loc. S, S, S NS, S, NS S, S, S S, S, S NS, S, NS S, NS, S S, S, S S, NS, S S, S, S S, NS, NS Var.×year S, S, S S, S, S S, S, S NS, S, NS S, NS, S S, S, S NS, NS, NS S, NS, S NS, NS, NS NS, NS, S For P accumulation, there was a significant difference between the two genotypes, with 9746 having greater P accumulation at early stage, but Jinfeng having greater P accumulation in panicles at maturity stage. The difference between the locations varied with growth stages. At mid-tillering and booting stages, Qingpu had more leaf and stem P accumulations, whereas at heading and maturity stages, Jinshan had more. The significant difference was found between years, with more leaf and stem P accumulation in 2006 than in 2005. For K accumulation, the difference in leaves between the two genotypes was only found at maturity stage, while in the stems the difference was also significant at heading stage. Although 9746 had more K accumulation in the panicles than Jinfeng at heading stage, the opposite was true at maturity stage. There was no difference between locations in leaf and stem K accumulations through the whole stage, except for the stems in maturity stage and the leaves in booting stage. The difference between years was quite dominant and varied with growth stages. At early stage, the plants in 2006 had more K accumulation than those in 2005, whereas at later stage the case was just opposite. DISCUSSION 1 There have been a great number of studies on the relationships between yield components and yield in rice. Li et al.( 2005 2005 ) reported that there were positive correlations between yield and harvest index (HI), grain weight per plant, and grains per panicle. Rice yield is mainly dependent on producing ability of dry matter before heading (Katsura et al., 2007 2007 ). It is commonly observed in rice as well as other cereals that fertile floret percentage and floret number per unit area are negatively correlated (Fischer et al., 1977 1977 ; Matsushima, 1957 1957 ). Matsushima ( 1957 1957 ) proposed that sink size was mainly responsible for yield difference. Moreover, grain number, which is the major part of sink size, is positively correlated with yield (Matsushima, 1957 1957 ; Yoshida et al., 2006 2006 ). In the current study, it was found that the higher yield of 9746 was associated with more panicles per square meter and greater grain weight, but the higher yield of Jinfeng was mainly attributed to more grains per panicle (Table 2 2 ). It was reported that hybrid rice accumulated more biomass before heading than conventional cultivars. Sarker et al.( 2001 2001 ) showed that the degree of heterosis in dry matter accumulation was relatively lower at the later growth stage than at the early growth stage. Khan et al.( 1998 1998 ) found that dry matter accumulation during ripening stage would be slightly lower in hybrid rice than in their mid-parental varieties. In this study, two rice genotypes showed a similar change in leaf biomass, but an obvious difference in stem biomass; 9746 had little increase after booting stage, while Jinfeng showed a marked increase from booting stage to heading stage. The results suggest that biomass accumulated before heading has a great contribution to final yield performance. Obviously, the biomass accumulated in the stems (and leaf sheath) will be transported into developing grains during filling stage. Murayama et al.( 1987 1987 ) considered that hybrid rice had a higher photosynthetic rate due to heterosis. However, photosynthetic rate of the same genotypes showed the marked difference during growth stage (Sarker et al., 2001 2001 ; Tang et al., 2002 2002 ). It is well known that there was an optimum LAI for CGR (crop growth rate) (Bunce, 1989 1989 ). In this study, the photosynthetic rate and LAI of 9746 at heading stage were found significantly higher than those of Jinfeng (Table 4 4 ). However, little difference in biomass between the two genotypes was found, indicating that photosynthetic rate, at least at heading stage, might not be a main factor responsible for high grain yield in “super rice”. It is well documented that tillering capacity is one of the most important characters determining yield potential, as it is closely related with the number of panicle per unit area (Zou et al., 1991 1991 ). On the other hand, the stem acts as an important storage organ and affects yield formation by accumulating biomass (Katsura et al., 2007 2007 ). It has been shown that N concentration in rice plants is an important factor affecting tiller initiation and higher N concentration enhances tillering and stem development (Weerakoon et al., 1999 1999 ; Zhong et al., 2003 2003 ). Thus more N accumulation at mid-tillering stage means greater tillering capacity and better stem development. It can be assumed that N accumulation at mid-tillering stage is one of the most important and stable traits for determination of high yield formation. Cooper and Somrith ( 1997 1997 ) reported that the interaction between genotype and environment (G×E) played a relatively large role in the yield of rice grown in the rain fed lowlands. It was also reported that the physiological traits, such as photosynthesis rate (Hu et al., 2007 2007 ), nitrogen uptake (Shan et al., 2005 2005 ), floret number per unit area (Yoshida et al., 2006 2006 ), which are closely related to the yield, showed wide variation over genotypes and environments. Furthermore, the G×E interaction may be partitioned into the interactions of genotype-by-location (G×L), genotype-by-year (G×Y) and genotype-by-location-by-year (G×L×Y). In this study, almost all the traits examined showed the significant differences between years or locations. On the other hand, some traits, including SLA at mid-tillering and heading stages, F v / F m and LAI at heading stage, stem biomass at heading and maturity stages, and leaf N concentration at mid-tillering and heading stages, kept little change over years and locations, indicating that these characters may be used in evaluation of yield potential or breeding program. References T Amano , C Shi , D Qin , M Tsuda , Y Matsumoto . High-yielding performance of paddy rice achieved in Yunnan Province, China: I. High yielding ability of Japonica F1 hybrid rice, Yu-Za 29 . Japanese Journal of Crop Science , 1996 . 65 ( 1 ): 16 - 21 . . High-yielding performance of paddy rice achieved in Yunnan Province, China: I. High yielding ability of Japonica F1 hybrid rice, Yu-Za 29 JA Bunce . Growth rate, photosynthesis and respiration in relation to leaf area index . Annals of Botany , 1989 . 63 ( 4 ): 459 - 463 . . Growth rate, photosynthesis and respiration in relation to leaf area index SH Cheng , SK Min . Super-rice breeding in China . Chinese Rice Research Newsletter , 2001 . 9 ( 2 ): 13 - 15 . . Super-rice breeding in China M Cooper , B Somrith . Implications of Genotype-by-Environment Interactions for Yield Adaptation of Rain Fed Lowland Rice: Influence of Flowering Date on Yield Variation . Proc. Workshop, Plant Breeding Strategies for Rain fed Lowland Rice in Drought Prone Environments , 1997 . : Canberra, Australia ACIAR , 104 - 114 . . Implications of Genotype-by-Environment Interactions for Yield Adaptation of Rain Fed Lowland Rice: Influence of Flowering Date on Yield Variation RA Fischer , I Aguilar , DR Laing . Post-anthesis sink size in a high-yielding dwarf wheat: yield response to grain number . Australian Journal of Agricultural Research , 1977 . 28 ( 2 ): 165 - 175 . DOI: 10.1071/AR9770165 http://doi.org/10.1071/AR9770165 . Post-anthesis sink size in a high-yielding dwarf wheat: yield response to grain number T Horie , M Ohnishi , JF Angus , LG Lewin , T Tsukaguchi , T Matano . Physiological characteristics of high-yielding rice inferred from cross-location experiments . Field Crops Research , 1997 . 52 ( 1-2 ): 55 - 67 . DOI: 10.1016/S0378-4290(96)03458-2 http://doi.org/10.1016/S0378-4290(96)03458-2 . Physiological characteristics of high-yielding rice inferred from cross-location experiments ML Hu , YX Zhang , LN Kong , QH Yang , CM Wang , HQ Zhai , JM Wan . Quantitative trait locus for photosynthesis and its related physiological traits in rice ( Oryza sativa L.) . Acta Agronomica Sinica , 2007 . 33 ( 2 ): 183 - 188 . (in Chinese) . Quantitative trait locus for photosynthesis and its related physiological traits in rice (Oryza sativa L.) IRRI (International Rice Research Institute) Rice Facts , 1995 . : Los Banos, Philippines IRRI , . Rice Facts K Katsura , S Maeda , T Horie , T Shiraiwa . Analysis of yield attributes and crop physiological traits of Liangyoupeijiu, a hybrid rice recently bred in China . Field Crops Research , 2007 . 103 ( 3 ): 170 - 177 . DOI: 10.1016/j.fcr.2007.06.001 http://doi.org/10.1016/j.fcr.2007.06.001 . Analysis of yield attributes and crop physiological traits of Liangyoupeijiu, a hybrid rice recently bred in China MNA Khan , S Murayama , Y Ishimine . Physio-morphological studies of F1 hybrids in rice ( Oryza sativa L.). Photosynthetic ability and yield . Plant Production Science , 1998 . 1 ( 4 ): 233 - 239 . . Physio-morphological studies of F1 hybrids in rice (Oryza sativa L.). Photosynthetic ability and yield GS Khush , WR Coffman , HM Beachell . WG Rockwood . The History of Rice Breeding: IRRI’s Contribution . Rice Research and Production in the 21st Century: Symposium Honoring Robert F Chandler, Jr. , 2001 . : Metro Manila, Philippines 117 - 135 . . The History of Rice Breeding: IRRI’s Contribution CS Li , SH Ye , YZ Chen , GH Ruan , FD Huang , XM Zhang . The study on yield component factors of high yielding Japonica rice varieties . Acta Agriculture Zhejiangensis , 2005 . 17 ( 4 ): 177 - 181 . (in Chinese) . The study on yield component factors of high yielding Japonica rice varieties S Matsushima . Analysis of developmental factors determining yield and yield prediction in lowland rice . Bulletin of the National Institute of Agricultural Science Japan , 1957 . 5 203 - 206 . (in Japanese) . Analysis of developmental factors determining yield and yield prediction in lowland rice S Murayama , K Miyazatv , A Nost . Studies on matter production of F1 hybrid in rice. I. Heterosis in the single leaf photosynthetic rate . Japanese Journal of Crop Science , 1987 . 56 ( 2 ): 198 - 203 . . Studies on matter production of F1 hybrid in rice. I. Heterosis in the single leaf photosynthetic rate MW Rosegrant , M Agcaoili-Sombilla , ND Perez . Global Food Projections to 2020: Implications for Investment. Food, Agriculture, and Environment Discussion , 1995 . : Washington, DC IFPRI , Paper No. 5 . Global Food Projections to 2020: Implications for Investment. Food, Agriculture, and Environment Discussion MAZ Sarker , S Murayama , Y Ishimine , E Tsuzuki . Heterosis in photosynthetic characters and dry matter production in F1 hybrids of rice . Nippon Sakumotsu Gakkai Koenkai Yoshi, Shiryoshu , 2001 . 211 48 - 49 . . Heterosis in photosynthetic characters and dry matter production in F1 hybrids of rice YH Shan , YL Wang , XB Pan . Mapping of QTLs for nitrogen use efficiency and related traits in rice ( Oryza sativa L.) . Agricultural Sciences in China , 2005 . 4 ( 10 ): 721 - 727 . . Mapping of QTLs for nitrogen use efficiency and related traits in rice (Oryza sativa L.) MS Swaminathan . Can science and technology feed the world in 2025? . Field Crops Research , 2007 . 104 ( 1-3 ): 3 - 9 . DOI: 10.1016/j.fcr.2007.02.004 http://doi.org/10.1016/j.fcr.2007.02.004 . Can science and technology feed the world in 2025? JJ Tang , X Chen , S Katsuyoshi . Varietal differences in photosynthetic characters and chlorophyll fluorescence induction kinetics parameters among intergeneric progeny derived from Oryza × Sorghum , its parents, and hybrid rice . Journal of Zhejiang University SCIENCE B , 2002 . 3 ( 1 ): 113 - 117 . . Varietal differences in photosynthetic characters and chlorophyll fluorescence induction kinetics parameters among intergeneric progeny derived from Oryza×Sorghum, its parents, and hybrid rice MW Weerakoon , DM Olszkyb , DN Mossc . Effects of nitrogen nutrition on responses of rice seedlings to carbon dioxide . Agriculture, Ecosystems and Environment , 1999 . 72 ( 1 ): 1 - 8 . DOI: 10.1016/S0167-8809(98)00166-2 http://doi.org/10.1016/S0167-8809(98)00166-2 . Effects of nitrogen nutrition on responses of rice seedlings to carbon dioxide JF Ying , SB Peng , QR He , H Yang , CD Yang , RM Visperas , KG Cassman . Comparison of high yield rice in tropical and subtropical environments I. Determinants of grain and dry matter yields . Field Crops Research , 1998 . 57 ( 1 ): 71 - 84 . DOI: 10.1016/S0378-4290(98)00077-X http://doi.org/10.1016/S0378-4290(98)00077-X . Comparison of high yield rice in tropical and subtropical environments I. Determinants of grain and dry matter yields H Yoshida , T Horie , T Shiraiwa . A model explaining genotypic and environmental variation of rice spikelet number per unit area measured by cross-locational experiments in Asia . Field Crops Research , 2006 . 97 ( 2-3 ): 337 - 343 . DOI: 10.1016/j.fcr.2005.11.004 http://doi.org/10.1016/j.fcr.2005.11.004 . A model explaining genotypic and environmental variation of rice spikelet number per unit area measured by cross-locational experiments in Asia XH Zhong , SB Peng , AL Sanico , HX Liu . Quantifying the interactive effect of leaf nitrogen and leaf area on tillering of rice . Journal of Plant Nutrition , 2003 . 26 ( 6 ): 1203 - 1222 . DOI: 10.1081/PLN-120020365 http://doi.org/10.1081/PLN-120020365 . Quantifying the interactive effect of leaf nitrogen and leaf area on tillering of rice Y Zou , Q Tang , C Hu , S Liu , D Xiao . Dynamic simulation for rice growth and yield. II. The comparison and application of rice tillering statistical models . Crop Research , 1991 . 5 ( 4 ): 18 - 22 . . Dynamic simulation for rice growth and yield. II. The comparison and application of rice tillering statistical models 10.1631/jzus.B0710603 Project (No. 2005C12024) supported by the Department of Science and Technology of Zhejiang Province, China 2007-11-29 2008-03-07 2022-01-05 Copyright © 2008, Journal of Zhejiang University Science 2008 Journal of Zhejiang University-SCIENCE B (Biomedicine and Biotechnology) 05 09