Why high grain yield can be achieved in single seedling machinetransplanted hybrid rice under dense planting conditions?

Southern Regional Collaborative Innovation Center for Grain and Oil Crops (CICGO), Hunan Agricultural University, Changsha 410128, P.R.China

1. Introduction

World rice yield must increase by at least 1% annually to meet the growing demand for food due to population growth and economic development (Normile 2008). To achieve this goal, great efforts should be made to breed rice cultivars with higher yield potential (Penget al. 2008). The development of hybrid cultivars is a major approach since hybrid cultivars have a yield advantage of about 10−20%over improved inbred cultivars (Penget al. 1999; Chenget al. 2007). Moreover, it is also important to improve the crop management practices to get the greatest possible expression of yield potential in rice cultivars (Zouet al.2003).

China is the largest producer and consumer of rice in the world and a pioneer in applying hybrid rice technology (Wanget al.2005). Manual transplanting is the traditional but still dominant method for rice establishment in China (Chenet al. 2007). However, the operation of manual transplanting requires a large amount of manpower (about 400 man-hour ha–1) and the task is very laborious involving working in a stooping posture and moving in muddy field (Thomas 2002).Because of labor migration and increases in labor wages,the labor input for rice production has decreased significantly in China (Penget al. 2009). As a result, many rice farmers have greatly simplified crop management practices (Cai and Chen 2000). Typically, some rice farmers transplant rice at extremely wide spacing to reduce labor cost. This may cause a reduction in panicle number per unit land area and consequently a decline in rice yield (Penget al.2009;Huanget al. 2011a,2013b). In addition, this may also result in an increased environmental risk, because the farmers generally consider that the potential reduced panicle number caused by low planting density can be compensated for by applying more basal N fertilizer (Penget al. 2009; Huanget al. 2013b).

Machine transplanting is an alternative rice establishment method that can help achieve dense planting with less labor input. However, in machine-transplanted rice production,a high seed rate is generally used to minimize missing hill rate. To cut down the production cost, Chinese rice farmers prefer inbred cultivars with cheap seed price in machinetransplanted rice production. Partly because of this, the planting area to hybrid rice has started declining in recent years in China (Peng 2016). Therefore, it is important to reduce the seed rate in machine-transplanted hybrid rice production to reverse the declining trend of hybrid rice planting area in China.

Reducing seedling number per hill is the only way to reduce seed rate for machine-transplanted hybrid rice production under dense planting conditions. The main difficulty arising in this way is how to ensure a low hill missing rate. In recent years, we have addressed this problem by improving both seed germination rate and sowing accuracy, and have established a new seed sowing system for single seedling machine-transplanted hybrid rice production (Fig. 1). The preliminary production tests showed that single seedling machine-transplanted hybrid rice produced about 10% higher yield than did conventional machine-transplanted hybrid rice under dense planting conditions (data not shown). This is not surprise: A single seedling per hill is a component of a highyielding rice cultivation methodology called the system of rice intensification (Uphoffet al. 2002). However, it is not clear what factors contribute to the higher yield in single seedling machine-transplanted hybrid rice.

Fig. 1 A seed sowing system for single seedling machinetransplanted hybrid rice production.

Rice yield is determined by four components: panicle number per unit land area, spikelet number per panicle,spikelet filling percentage and grain weight (Yoshida 1981).However, in cereal crops including rice, the compensation mechanisms among yield components always arise, either from the physiological competition or from the developmental allometry (Grafiuset al.1976; Grafius 1978). For example,there is a tight negative relationship between panicle number per unit land area and spikelet number per panicle (Yinget al. 1998; Huanget al.2011a). A more spikelet number per panicle generally results in a lower spikelet filling percentage (Yanget al. 2002; Wanget al. 2006; Islamet al.2010). Therefore, establishing a harmonious relationship among the yield components is critical to achieve high rice yield. In this regard, it is recognized that increasing biomass production plays a key role in coordinating the relationships among the yield components in rice (Yinget al. 1998; Huanget al. 2013a).

In our present study, we compared grain yield and yield attributes between single seedling and conventional machine-transplanted hybrid rice at a high planting density in 2015 and 2016. Our objective was to identify the factors associated with high grain yield in single seedling machinetransplanted hybrid rice under dense planting conditions.

2. Materials and methods

Field experiments were conducted at Yongan Town(28°09´N, 113°37´E, 43 m a.s.l.), Hunan Province, China in late rice-growing season in 2015 and 2016. The experimental site has a moist subtropical monsoon climate.The soil at the experimental site was clayey with pH=5.85,organic matter=38.4 g kg–1, available N=75.4 mg kg–1,available P=12.8 mg kg–1, and available K=115 mg kg–1. The soil test was based on samples collected from the upper 20 cm of the soil.

Treatments were a factorial combination of two machine transplanting methods and two hybrid rice cultivars. The experiment was arranged in a randomized block design with three replications and a plot size of 80 m2. Two machine transplanting methods were single seedling machine transplanting (SMT) and conventional machine transplanting(CMT). Two hybrid rice cultivars were Taiyou 390 and Wuyou 308 in 2015 and Taiyou 390 and Longjingyou 1212 in 2016. For SMT, seeds were sown according to the procedures described in Fig. 1. For CMT, seeds were manually sown in seedling trays. Seed rates were 13.5–14.4 g per tray for SMT and 80 g per tray for CMT.

The 17- and 20-day-old seedlings were transplanted with a high-speed rice transplanter (PZ80-25, Dongfeng Iseki Agricultural Machinery Co., Ltd., Xiangyang, China)in 2015 and 2016, respectively. Transplanting was done at a spacing of 25 cm×11 cm, which is the highest planting density achieved by the transplanter. SMT had a hill missing rate of about 10%, while CMT had hardly any missing hills. The missing hills were replanted by hand at 7 days after transplanting to ensure a uniform plant population.Nitrogen was applied in three splits (75 kg N ha–1at basal,30 kg N ha–1at mid-tillering, and 45 kg N ha–1at panicle initiation). Phosphorus (75 kg P2O5ha–1) was applied at basal. Potassium (150 kg K2O ha–1) was split equally at basal and panicle initiation. The experimental field was kept flooded from transplanting until 7 days before maturity.Insects, diseases, and weeds were intensively controlled by chemicals to avoid yield loss.

Three seedling samples (30 seedlings per sample) were randomly selected for each machine transplanting method at the transplanting day to determine seedling traits including plant height, basal stem width, shoot dry weight, and root dry weight. Twenty hills were marked in each plot to count tillers at a 5-day interval from 10 to 40 days after transplanting.Ten hills were sampled in each plot at heading and maturity stages. Plants were separated into leaves, stems and panicles at heading stage. Leaf area was determined with a leaf area meter (LI-3000C, Li-Cor, Lincoln, NE, USA). Each plant organ was oven-dried at 70°C to constant weight to determine dry weight. At maturity, plants were separated into straw and panicles. Panicle number was counted in each hill to calculate panicle number per m2. Primary branch number per panicle, the secondary branch number per panicle, and panicle length were determined. Panicles were hand-threshed and the filled spikelets were separated from unfilled spikelets by submerging them in tap water.Three subsamples of 30 g of filled spikelets and all unfilled spikelets were taken to count the number of spikelets. Dry weight of straw, rachis, and filled and unfilled spikelets were determined after oven-drying at 70°C to constant weight. Total biomass was the summation of straw, rachis,and filled and unfilled spikelets dry matter. Leaf area per stem at heading stage, dry weight per stem at heading and maturity stages, harvest index, panicle-bearing tiller rate,spikelet number per cm of panicle length, spikelet number per panicle, spikelet filling percentage, and grain weight were calculated. Grain yield was determined from a 5-m2area in each plot and adjusted to the standard moisture content of 0.14 g H2O g–1.

Data were analyzed by analysis of variance (Statistix 8.0,Analytical software, Tallahassee, FL, USA). The statistical model included replication, machine transplanting method,cultivar, and the interaction between machine transplanting method and cultivar. Means of treatments were compared based on the least significant difference test at the 0.05 probability level for each year.

3. Results

Plant height in seedlings for SMT was higher than that in seedlings for CMT by 22 and 38% in Taiyou 390 and Wuyou 308 in 2015 and by 5 and 7% in Taiyou 390 and Longjingyou 1212 in 2016, respectively (Table 1). Seedlings for SMT had larger basal stem width than did seedlings for CMT by 43 and 59% in Taiyou 390 and Wuyou 308 in 2015 and by 19% in both Taiyou 390 and Longjingyou 1212 in 2016, respectively. Shoot dry weight was greater in seedling for SMT than for CMT by 1.58 and 2.06 times in Taiyou 390 and Wuyou 308 in 2015 and by 60 and 48% in Taiyou 390 and Longjingyou 1212 in 2016, respectively. Root dry weight in seedlings for SMT was greater than that in seedlings for CMT by 1.30 and 2.04 times in Taiyou 390 and Wuyou 308 in 2015 and by 91 and 68% in Taiyou 390 and Longjingyou 1212 in 2016, respectively.

Grain yield was higher under SMT than under CMT by 10 and 12% in Taiyou 390 and Wuyou 308 in 2015 and by 10 and 20% in Taiyou 390 and Longjingyou 1212 in 2016,respectively (Table 2). In 2015, panicle number per m2was 30 and 32% less under SMT than that under CMT in Taiyou 390 and Wuyou 308, respectively. In 2016, SMT had significantly but slightly less panicle number per m2than did CMT. Spikelet number per panicle under SMT was more than that under CMT by 61 and 46% in Taiyou 390 and Wuyou 308 in 2015 and by 10 and 15% in Taiyou 390 and Longjingyou 1212 in 2016, respectively. SMT had higherspikelet filling percentage than CMT, and the difference was significant for Wuyou 308 in 2015 and Longjingyou 1212 in 2016. The difference in grain weight between SMT and CMT was relatively small and inconsistent across years. SMT produced higher total biomass than CMT, and the difference was significant for Taiyou 390 and Longjingyou 1212 in 2016.

Table 1 Seedling traits in hybrid rice cultivars for single seedling and conventional machine transplanting in 2015 and 2016

Table 2 Grain yield and yield attributes in hybrid rice cultivars grown under single seedling and conventional machine transplanting in 2015 and 2016

Harvest index was higher under SMT than that under CMT, and the difference was significant for Taiyou 390 and Wuyou 308 in 2015.

Tiller number per m2was significantly less under SMT than that under CMT during 10 to 40 days after transplanting(Fig. 2-A–D). The maximum tiller number per m2under SMT was less than that under CMT by 33% in both Taiyou 390 and Wuyou 308 in 2015 and by 22 and 31% in Taiyou 390 and Longjingyou 1212 in 2016, respectively. The difference in panicle-bearing tiller rate between SMT and CMT was not significant in 2015 (Fig. 3-A and B). In 2016, panicle panicle-bearing tiller rate was 14 and 22% higher under SMT than that under CMT in Taiyou 390 and Longjingyou 1212, respectively (Fig. 3-C and D).

Primary branch number per panicle under SMT was more than that under CMT by 21 and 15% in Taiyou 390 and Wuyou 308 in 2015 and by 7 and 13% in Taiyou 390 and Longjingyou 1212 in 2016, respectively (Table 3). SMT had more secondary branch number per panicle than did CMT by 84 and 47% in Taiyou 390 and Wuyou 308 in 2015 and by 38 and 26% in Taiyou 390 and Longjingyou 1212 in 2016, respectively. In 2015, panicle length was 13 and 10%longer under SMT than that under CMT in Taiyou 390 and Wuyou 308, respectively. In 2016, there was no significant difference between SMT and CMT in panicle length. Spikelet number per cm of panicle length was more under SMT than that under CMT by 42 and 33% in Taiyou 390 and Wuyou 308 in 2015 and by 9 and 12% in Taiyou 390 and Longjingyou 1212 in 2016, respectively.

Dry weight per stem at heading stage was greater under SMT than that under CMT by 30 and 56% in Taiyou 390 and Wuyou 308 in 2015 and by 15 and 16% in Taiyou 390 and Longjingyou 1212 in 2016, respectively (Fig. 4-A–D).SMT produced greater dry weight per stem at maturity than did CMT by 48 and 50% in Taiyou 390 and Wuyou 308 in 2015 and by 11 and 15% in Taiyou 390 and Longjingyou 1212 in 2016, respectively. Leaf area per stem at heading under SMT was larger than that under CMT by 18 and 42%in Taiyou 390 and Wuyou 308 in 2015 and by 7 and 13%in Taiyou 390 and Longjingyou 1212 in 2016, respectively(Fig. 5-A–D).

Fig. 2 Tillering dynamic in hybrid rice cultivars grown under single seedling and conventional machine transplanting in 2015(A and B) and 2016 (C and D). Cultivars in A–D are Taiyou 390,Wuyou 308, Taiyou 390, and Longjingyou 1212, respectively.SMT, single seedling machine transplanting; CMT, conventional machine transplanting. Vertical bars represent SE.

4. Discussion

China’s rice production is in the unprecedented period of transition to mechanization (Peng 2014). Reducing seed rate is critical for hybrid rice to adapt to the period of transition (Peng 2016). In the present study, seed rate for machine-transplanted hybrid rice was largely reduced by using a single seed sowing system established by our group.However, there is a general worry that low seed rate may cause weak root twining power and consequently fractured seedling blocks in machine-transplanted rice production(Tenget al. 2015). Interestingly, we did not observe such a phenomenon in the present study. This might be attributed to that (1) hybrid rice plants generally had larger root systems (Huanget al.2012), and (2) root growth was largely improved under low seed rate conditions (Table 1). The single seedling sowing system also resulted in a tremendous improvement in seedling quality. This is easy to understand that reducing seed rate allows more space for seedling development. It is well known that high seedling quality is of benefit to yield formation in rice (Huanget al.2012). This might also be partly responsible for the higher grain yield under SMT than under CMT in the present study. Averaged across cultivars and years, SMT produced 12% higher grain yield than did CMT, which is similar to that observed in the preliminary production tests (data not shown). Moreover,there is a noteworthy difference between this study and the preliminary production tests. The missing hills were replanted in this study but not in the preliminary production tests. The missing hill rates in the preliminary production tests were 8–14%. This indicates that a hill missing rate of less than 15% may has small effect on grain yield for SMT under dense planting conditions, but further investigations are required to confirm this speculation.

Fig. 3 Panicle-bearing tiller rate in hybrid rice cultivars grown under single seedling and conventional machine transplanting in 2015 (A and B) and 2016 (C and D). Cultivars in A–D are Taiyou 390, Wuyou 308, Taiyou 390, and Longjingyou 1212,respectively. SMT, single seedling machine transplanting; CMT,conventional machine transplanting. Vertical bars represent SE.

Prior to this study, limited information was available on the yield formation processes in single seedling machinetransplanted hybrid rice under dense planting conditions.In our present study, we compared yield attributes between single seedling and conventional machine-transplanted hybrid rice at a high planting density (25 cm×11 cm). Our results showed that SMT generally had less panicle number per m2than did CMT. Panicle number per m2is a function of the maximum tiller number per m2and panicle-bearingtiller rate (Huanget al.2011a). In this study, the less panicle number per m2under SMT was mainly attributed to less maximum tiller number per m2than under CMT, because the difference in panicle-bearing rate between SMT and CMT was inconsistent across years. This is in agreement with that reported by Huanget al.(2011a), who observed that there was a tight positive relationship between panicle number per m2and the maximum tiller number per m2. In contrast to panicle number per m2, spikelet number per panicle was more under SMT than under CMT. In this regard, it is suggested that a strong compensation mechanism exists between panicle number per unit land area and spikelet number per panicle (Yinget al. 1998; Huanget al. 2011a).Spikelet number per panicle can be divided into four subcomponents: the primary branch number per panicle, the secondary branch number per panicle, spikelet number per primary branch, and spikelet number per secondary branch(Cuiet al. 2002). Kato (1997) reported that spikelet number per panicle was strongly positively associated with primary branch number per panicle, whereas Cuiet al. (2002) and Meiet al. (2006) stated that spikelet number per panicle was more closely correlated with the secondary branch number per panicle. In the present study, the more spikelet number per panicle under SMT was driven more by increased secondary branch number per panicle than by increased primary branch number per panicle. In another approach, spikelet number per panicle is determined by panicle length and spikelet number per unit panicle length(Wanget al. 2007). In this study, because the difference in panicle between SMT and CMT was inconsistent across years, the more spikelet number per panicle under SMT was mainly attributed to more spikelet number per cm of panicle length than that under CMT. This observation is consistent with the result of Cuiet al.(2002) that spikelet number per panicle was more closely related with spikelet number per unit panicle length. However, rice plants with more spikelet number per cm of panicle length(i.e., compact panicle) are always lower in spikelet filling percentage (Yanget al. 2002; Wanget al.2006; Islamet al.2010). Interestingly, in this study, SMT had more spikelet number per cm of panicle length and higher or equal spikelet filling percentage than did CMT. These results indicate that the potential compensation between spikelet number per panicle and spikelet filling percentage was detached under SMT.

Table 3 Panicle architecture in hybrid rice cultivars grown under single seedling transplanting (SMT) and conventional machine transplanting (CMT) in 2015 and 2016

Fig. 4 Dry weight per stem at heading and maturity stages in hybrid rice cultivars grown under single seedling and conventional machine transplanting in 2015 (A and B) and 2016 (C and D). Cultivars in A–D are Taiyou 390, Wuyou 308,Taiyou 390, and Longjingyou 1212, respectively. SMT, single seedling machine transplanting; CMT, conventional machine transplanting. Vertical bars represent SE.

Fig. 5 Leaf area per stem at heading stage in hybrid rice cultivars grown under single seedling and conventional machine transplanting in 2015 (A and B) and 2016 (C and D).Cultivars in A–D are Taiyou 390, Wuyou 308, Taiyou 390, and Longjingyou 1212, respectively. SMT, single seedling machine transplanting; CMT, conventional machine transplanting.Vertical bars represent SE.

It has been recognized that the way to decouple the potential negative relationships between yield components in rice crops is to increase biomass production (Yinget al.1998; Huanget al. 2013a). Consistently, in this study, SMT had heavier dry weight per stem than did CMT. The heavier dry weight per stem under SMT was partly attributed to lager leaf area per stem than that under CMT. Moreover,we observed that radiation use efficiency (RUE) was 8–12% higher under SMT than that under CMT in 2015(data not shown). RUE depends on gross photosynthesis,maintenance respiration, and growth respiration (Charles-Edwards 1982). Because large reductions in respiration are unlikely (Byrdet al.1992), increasing photosynthesis is probably the only option for raising the value of RUE(Mitchell and Sheehy 2006). Therefore, we deduced that improved leaf photosynthetic rate might also be partly responsible for the heavier dry weight per stem under SMT. This deduction, to some extent, can be supported by a study on direct-seeded rice by San-ohet al. (2006),who reported that rice crops planted with a single seed per hill maintained higher levels of Rubisco and N in leaves and consequently higher leaf photosynthetic rate during ripening. These highlight that the need for greater fundamental understanding on the physiological processes governing dry weight per stem in single seedling machinetransplanted hybrid rice.

5. Conclusion

High seedling quality, large panicle size, and heavy dry weight per stem are critical factors to the high grain yield in single seedling machine-transplanted hybrid rice under dense planting conditions.

Acknowledgements

This work was supported by the National Key R&D Program of China (2017YFD0301503) and the earmarked fund for China Agriculture Research System (CARS-01).

Byrd G T, Sage R F, Brown R H. 1992. A comparison of dark respiration between C3and C4plants.Plant Physiology,100, 191–198.

Cai H, Chen Q. 2000. Rice research in China in the early 21st century.Chinese Rice Research Newsletter,8, 14–16.

Charles-Edwards D A. 1982.Physiological Determinants of Crop Growth. Academic Press, Sydney, Australia.

Chen S, Xia G M, Zhao W M, Wu F B, Zhang G P. 2007.Characterization of leaf photosynthetic properties for notillage rice.Rice Science,14, 283–288.

Cheng S H, Zhuang J Y, Fan Y Y, Du J H, Cao L Y. 2007.Progress in research and development on hybrid rice: A super-domesticate in China.Annual Botany,100, 959–966.

Cui K H, Peng S B, Xing Y Z, Yu S B, Xu C G. 2002. Genetic analysis of the panicle traits related to yield sink size of rice.Acta Genetica Sinica,29, 144–152.

Grafius J E. 1978. Multiple characters and correlated response.Crop Science,18, 931–934.

Grafius J E, Thomas R T, Barnard J. 1976. Effect of parental component complementation on yield and components of yield in barley.Crop Science,16, 673–677.

Huang M, Jiang L, Xia B, Zou Y, Jiang P, Ao H. 2013a. Yield gap analysis of super hybrid rice between two subtropical environments.Australian Journal of Crop Science,7,600–608.

Huang M, Xia B, Zou Y, Jiang P, Shi W, Hongthong P, Xie X.2012. Improvement in super hybrid rice: A comparative study between super hybrid and inbred varieties.Research on Crops,13, 1–10.

Huang M, Yang C, Ji Q, Jiang L, Tan J, Li Y. 2013b. Tillering responses of rice to plant density and nitrogen rate in a subtropical environment of southern China.Field Crops Research,149, 187–192.

Huang M, Zou Y, Feng Y, Cheng Z, Mo Y, Ibrahim M, Xia B,Jiang P. 2011a. No-tillage and direct seeding for super hybrid rice production in rice-oilseed rape cropping system.European Journal of Agronomy,34, 278–286.

Huang M, Zou Y, Jiang P, Xia B, Ibrahim M, Ao H. 2011b.Relationship between grain yield and yield components in super hybrid rice.Agricultural Sciences in China,10,1537–1544.

Islam M S, Peng S, Visperas R M, Bhuiya M S U, Hossain S M A, Julfiquar A W. 2010. Comparative study on yield and yield attributes of hybrid, inbred, and NPT rice genotypes in a tropical irrigated ecosystem.Bangladesh Journal of Agricultural Research,35, 343–353.

Kato T. 1997. Selection responses for the characters related to yield sink capacity of rice.Crop Science,37, 1472–1475.

Mei H W, Xu J L, Li Z K, Yu X Q, Guo L B, Wang Y P, Ying C S,Luo L J. 2006. QTLs influencing panicle size detected in two reciprocal introgressive line (IL) populations in rice (Oryza sativaL.).Theoretical and Applied Genetics,112, 648–656.

Mitchell P L, Sheehy J E. 2006. Supercharging rice photosynthesis to increase yield.New Phytologist,171,688–693.

Normile D. 2008. Reinventing rice to feed the world.Science,321, 330–333.

Peng S. 2014. Reflection on China’s rice production strategies during the transition period.Scientia Sinica Vitae,44,845–850. (in Chinese)

Peng S. 2016. Dilemma and way-out of hybrid rice during the transition period in China.Acta Agronomica Sinica,42,313–319. (in Chinese)

Peng S, Cassman K G, Virmani S S, Sheehy J, Khush G S.1999. Yield potential trends of tropical rice since the release of IR8 and the challenge of increasing rice yield potential.Crop Science,39, 1552–1559.

Peng S, Khush G S, Virk P, Tang Q, Zou Y. 2008. Progress in ideotype breeding to increase rice yield potential.Field Crops Research,108, 32–38.

Peng S, Tang Q, Zou Y. 2009. Current status and challenges of rice production in China.Plant Production Science,12, 3–8.

San-oh Y, Sugiyama T, Yoshida D, Ookawa T, Hirasawa T. 2006. The effect of planting pattern in the rate of photosynthesis and related processes during ripening in rice plants.Field Crops Research,96, 113–124.

Teng F, Zhu D, Chen H, Cai X, Xu Y. 2015. Effects of seed rate on root twining power and seedling quality of machine-transplanted super rice.Agricultural Science and Technology,16, 2644–2648.

Thomas E V. 2002. Development of a mechanism for transplanting rice seedlings.Mechanism and Machine Theory,37, 395–410.

Uphoff N, Rafaralahy S, Rabenandrasana J. 2002. What is the system of rice intensification? In: Uphoff N, Fernandes E C M, Yuan L, Peng J, Rafaralahy S, Rabenandrasana J, eds.,Assessments of the System of Rice Intensification(SRI).Sanya, China. pp. 5–7.

Wang F, Cheng F M, Zhang G. 2006. The relationship between grain filling and hormone content as affected by genotype and source-sink relation.Plant Growth Regulation,49, 1–8.

Wang F, Cheng F M, Zhang G P. 2007. Difference in grain yield and quality among tillers in rice genotypes differing in tillering capacity.Rice Science,14, 135–140.

Wang Y, Xue Y, Li J. 2005. Towards molecular breeding and improvement of rice in China.Trends in Plant Science,10,610–614.

Yang J, Peng S, Zhang Z, Wang Z, Visperas R M, Zhu Q. 2002.Grain and dry matter yields and partitioning of assimilates injaponica/indicahybrid rice.Crop Science,42, 766–772.

Ying J, Peng S, He Q, Yang H, Yang C, Visperas R M, Cassman K G. 1998. Comparison of high-yield rice in tropical and subtropical environments. I. Determinants of grain and dry matter yields.Field Crop Research,57, 71–84.

Yoshida S. 1981.Fundamentals of Rice Crop Science.International Rice Research Institute, Los Baños, The Philippines.

Zou Y, Zhou S, Tang Q. 2003. Status and prospect of high yielding cultivation researches on China super hybrid rice.Journal of Hunan Agricultural University,29, 78–84.