Hybrid rice for food security, poverty alleviation, and environmental protection

Proceedings of the 4th International Symposium on Hybrid Rice held in Hanoi, Vietnam, 14-17 May 2002, which was attended by 187 participants from 19 countries and three international agencies (IRRI, the Food and Agriculture Organization, and the Asia-Pacific Seed Association).

Book is available

Recent progress in breeding super hybrid rice in China
LP Yuan

No abstract

Advances in hybrid rice research and development in the tropics
SS Virmani

During the past five years, hybrid rice technology in the tropics has entered the commercialization phase in India, Vietnam, the Philippines, Bangladesh, and Indonesia. Almost all the commercial rice hybrids are derived from the cytoplasmic male sterile (CMS) system. An IRRI CMS line, IR58025A, has been widely used in the tropics. Even in China, this line has been used in Jiangxi and Hunan provinces to improve the grain quality of local hybrids. IRRI's hybrid rice breeding program has focused on developing improved CMS lines possessing cytoplasmic and genetic diversity, improved grain quality, a higher outcrossing rate, and resistance to diseases and insects. Most widely used wild abortive cytoplasm has been analyzed for its effect on grain quality and resistance to/tolerance of biotic and abiotic stresses. There was no negative effect on the traits except for grain chalkiness, which needs to be studied further. As reported earlier, there was no lack of restorer lines among elite indica lines bred in the tropics. Marker-aided selection (MAS) using the sequence-tagged site marker RG140 with PvuII digestion linked with the Rf3 gene on chromosome 1 was useful for increasing screening efficiency for restorers. Progress was also made in developing stable TGMS lines possessing a low critical sterility point and good outcrossing. Several public and private hybrids were released in national agricultural research and extension systems and new elite hybrids were identified. Average seed yields were further improved in the national programs by further fine-tuning of the seed production technology, training, the choice of appropriate locations/seasons, and, above all, the experience of seed growers. A significant increase occurred in private-sector participation in the hybrid rice seed industry during the past five years. About 40 private companies are actively involved in hybrid rice research and/or seed production and marketing in tropical Asia. During 2001, private companies were producing 90% of the hybrid rice seeds in India. Hybrid rice in the tropics covered about 770,000 ha in 2002. Major constraints identified during the commercialization phase of the technology in the tropics were inconsistent performance, inferior grain quality, an inadequate level of disease/insect resistance of hybrids, the inadequate supply and high cost of hybrid seeds, and inadequate policy support. Future opportunities include the development of hybrids possessing stronger heterosis and more stable yield performance, improved grain quality, an adequate level of disease/insect resistance, and higher seed yields. Both conventional and molecular methods will be used in hybrid breeding and in assessing seed purity. Agronomic management of hybrid rice in the tropics will also be improved to improve hybrid yield and stability in performance. Hybrid rice, which made a significant impact in China in the 20th century, should contribute significantly to food security and environmental protection in the tropical rice-growing countries in the 21st century.

Breeding strategies to enhance heterosis in rice
P.S. Virk, G.S. Khush, and S.S. Virmani

Indica germplasm has served as a rich reservoir for the identification of heterotic combinations in rice. However, the yield potential of indica parental lines is about 9-9.5 t ha-1 under tropical conditions. Further improvement in the yield potential of parental lines and broadening of their genetic base should lead to the identification of hybrid combinations with even higher heterosis. To develop elite lines with higher yield potential, breeding work on the development of the new plant type (NPT), using tropical japonica (TJ) germplasm, began in 1989 at IRRI. Since then, major progress has been made in developing high-yielding NPT lines. Two high-yielding NPT lines have been released as varieties in China from this program. To further improve the new plant type for wider acceptability and its ability to withstand harsh tropical environments, elite indica lines were crossed with NPT lines. NPT lines derived from IJ (indica/japonica) ? indica elite lines possess higher yield, the desired grain quality, and resistance to several diseases. They also have strong stems and hence don't lodge even after heavy rains and strong winds. We are continuing to improve the NPT lines for grain quality and for resistance to various diseases and insects. Improved NPT lines with high yield potential, developed from the crosses between indicas and tropical japonicas, offer new opportunities in heterosis breeding in rice.

We are now investigating whether the present level of heterosis (15% to 20%) observed in indica ? indica crosses can be further increased by exploiting the newly developed germplasm. In fact, enhanced heterosis (>40%) observed at IRRI in indica/tropical japonica crosses encouraged us to develop CMS lines in the genetic background of elite TJ lines possessing wide compatibility (WC) genes. A very low frequency of restorer lines among elite TJ lines did not allow us to use them as the male parent. In addition, TJ CMS lines showed two major weaknesses: degeneration of panicles and low outcrossing rate.

Consequently, we are emphasizing developing TGMS lines in the genetic background of elite TJ and indica/TJ derivative lines. The latter appear to be better adapted, and higher yielding in the tropics. They also possess acceptable indica grain type. Heterosis between indica (TGMS)/(indica/TJ) derivative lines is being studied to identify heterotic combinations. We have also identified six putative heterotic groups in rice that are likely to help us in using heterotic combinations.

Two-line hybrid rice breeding in and outside China
Tong-Min Mou, Lu Xing-Gui, N.T. Hoan, and S.S. Virmani

The two-line system for hybrid breeding is more efficient than the three-line system for exploiting heterosis in rice to increase the yield per unit area. Three PGMS genes and five TGMS genes have been identified by scientists from China, Japan, IRRI, Vietnam, and India. The PGMS genes pms1, pms2, and pms3 were located on chromosomes 7, 3, and 12, respectively. The TGMS genes tms1, tms2, tms3, tms4(t), and tms5 were located on chromosomes 8, 7, 6, 9, and 2, respectively. The molecular markers linked with these genes have also been found. In China, more than ten EGMS lines were used to breed commercial two-line hybrid rice. Thirty two-line rice hybrids were released in various rice-growing regions up to 2001. The area planted to two-line hybrid rice was 1.54 million ha in 2000. EGMS lines are multiplied in three ways: in the autumn season, in the winter season, and under cool-water irrigation. Yields of seed multiplication were 2-5 t ha-1. A technological package for seed production of two-line hybrid rice has been developed. The area of seed production was around 13,000 ha across China in 2001. The average yield was 2.5 t ha-1, which was similar to that of three-line hybrid rice seed production. The two-line hybrid rice breeding program at IRRI focuses on developing the technology for the tropics by deploying the TGMS system. TGMS genes tms2 (from Norin PL12) and tms3 (from IR32364S mutant) have been primarily deployed for breeding new TGMS lines. ID24, a TGMS line introduced from India, has also been used. TGMS lines IR73827-23S and IR73834S, derived from ID24 mutant, have stable sterility. Identification of such lines can be done in the field by evaluating them during the wet season when the mean temperature is 1-2 oC lower than in the dry season. These lines can be multiplied by growing them under high-altitude conditions. To facilitate incorporation of the tms2 gene, a simple sequence repeat (SSR) marker, RM11 located on chromosome 7, was identified and found to be useful in identifying heterozygous fertile plants in F2 populations and F3-F4 progenies for generation advance in selected crosses. The tms genes are transferred in indica and indica/TJ derivative lines from IRRI and elite lines from collaborating countries (Bangladesh, the Philippines, and Sri Lanka). TGMS lines are also being developed at IRRI through a shuttle breeding procedure for some temperate countries such as China, Iran, and Egypt. Using the genetic male sterility-facilitated recurrent selection procedure, a composite population is also being developed at IRRI for extracting TGMS lines. This population will be shared with national agricultural research and extension systems to enable them to extract locally adapted TGMS lines. In Vietnam and India, several TGMS lines with a relatively low critical sterility point were developed. Some of them are being evaluated in the field and used for developing two-line hybrid rice. Some two-line rice hybrids were tested in multilocation trials. Other countries, such as the Philippines, Bangladesh, Sri Lanka, Iran, Egypt, and Indonesia, were evaluating EGMS lines introduced from IRRI. In the tropics, TGMS lines can be multiplied by growing them under high-altitude conditions. Experimental seed production of two-line hybrid rice was carried out at IRRI and in the Philippines, Vietnam, and India. Two-line hybrid rice would be commercialized in 3-5 years in tropical countries. Future strategies and constraints to developing two-line hybrid rice worldwide are discussed in this paper.

Opportunities for and challenges to developing and using hybrid rice technology for temperate countries
T. Tsuchiya, A. Bastawisi, Z.Y. Yang, H.P. Moon, J.A. Mann, and H. Ikehashi

Despite the adverse climate for cereal breeding under the sufficient supply of cereals, breeding of hybrid rice was reestablished through the 1990s, and a series of new hybrids has been released and have shown a remarkable performance in Japan, the United States, and the temperate region in China, thus demonstrating an enormous potential for hybrid rice in the area. Superior experimental hybrids have been developed in Egypt and Korea. In that development, the private seed industry is playing a key role, as has been shown in India and China, while public institutes are still in a leading position in Egypt and Korea and for japonica hybrids in China. Even in Japan, where incentives for cereal production have been very low, a few private companies are trying to establish hybrid rice breeding in addition to Mitsui Chemicals Inc., which released hybrids to the market. Initiatives by the private sector seem to be decisively important in the development of hybrids. Strong demand for high rice grain quality is also common in the countries and the region being considered. The new series of hybrids is now much closer to the high quality standards than preceding ones. Hybrid rice technologies are not simple because of the coexistence of the two subspecies and the lack of heterogeneity in japonica types in temperate countries. Breeding methods are diversified because both indica and japonica types or intermediate types are planted in the countries. There is a strong preference for japonica-type rice except for the United States. The japonica characteristic is also essential for cultivation in cooler climates, though adaptability to such climates is not strongly required in Egypt and the southern states of the U.S. While the broad diversity in the U.S. hybrid programs makes it easy to derive intersubspecific hybrids, which may account for the early success in high-yielding hybrids, the development of a heterotic genetic background is a central issue in the seed production system for japonica-type hybrids.

Improving grain quality in hybrid rice
F.U. Zaman, B.C. Viraktamath, and S.S. Virmani

Grain quality preferences in rice vary from region to region and country to country. In most countries, long-grain indica rice, which is soft and nonsticky on cooking, is preferred; in others, low-amylose japonica rice, which is soft and sticky on cooking, is liked. Premium basmati rice is yet another group possessing some specific characteristic features, such as the presence of aroma, tenderness, and linear elongation on cooking. In the case of hybrids, quality considerations assume greater significance as their produce is formed by F2 seed generation.

Most of the hybrids commercialized in the tropics are based on IR58025A. Although these hybrids yielded 15-20% higher than high-yielding popular varieties, their grain quality did not find acceptability by consumers in some parts of India and Bangladesh. In some regions of India, the Philippines, and Vietnam, hybrids derived from IR58025A were widely accepted. This led to the development of hybrids to meet specific regional quality requirements. With this in view, as a first step, efforts have begun at IRRI and other centers to critically analyze the quality characteristics of all available CMS lines, maintainers, restorers, and their hybrids. This has helped to identify some new CMS and TGMS lines combining desirable quality characteristics for the development of hybrids. Simultaneously, the directed development of parental lines combining good grain quality, including basmati characteristics, has begun in India. Efforts in this direction have helped in the development of the first basmati high-quality hybrid, Pusa RH-10. Among other hybrids released in India so far, ADT RH-1 possesses good grain quality. In China also, greater emphasis is being given to the improvement of grain quality of hybrids.

Indica/japonica hybrids have shown high heterosis, but their grain quality seems to be a serious impediment. This has prompted the use of indica/japonica-derived lines to overcome quality problems. Studies on the effects of nucleo-cytoplasmic interaction on quality traits reveal that they vary from cross to cross. Cytoplasmic effects on chalkiness, head rice recovery, and amylose content need to be given due consideration when choosing parental lines for the development of hybrids. However, further studies to overcome the problem are required.

Opportunities for and challenges to improving hybrid rice seed yield and seed purity
C.X. Mao and S.S. Virmani

After success in China, more than 20 countries worldwide have started hybrid rice programs and about 40 rice hybrids have been released for commercial production. Hybrid rice seed production area and yield have increased rapidly as hybrid rice growing area expanded in many countries. Seed yield in China has reached 2.5-2.7 t ha-1 and about 1-1.5 t ha-1 outside China in commercial seed production plots. The world hybrid rice growing area should increase to 19-20 million ha in 2005, of which 16.5 million ha are in China and the rest outside China. The total area for F1 seed production and parental line multiplication should reach 123,400 ha in China and 35,000 ha outside China. Considering the Chinese experience, there are tremendous opportunities to improve hybrid rice seed yield and seed quality.

The challenges to improving hybrid rice seed yield and seed quality are (1) establishing a suitable system for parental line multiplication and purification, F1 seed production, and quality control; (2) finding the best sites and season for large-scale seed production; (3) training seed growers to increase their skills and experience; (4) refining seed production techniques locally; (5) improving field management approaches, including agronomic and pest control; (6) developing seed parent lines with a high outcrossing rate and resistance to major pests; (7) purifying existing parental lines; and (8) setting up national purity or quality standards for hybrid rice seed production.

International cooperation and the extensive participation of public, private, and nongovernment organization sectors in hybrid rice seed production and marketing would contribute toward both food security and additional rural employment in rice-growing countries.

Hybrid rice for mechanized agriculture
M. Walton

Hybrid rice breeding in the United States has been done primarily by RiceTec Inc., which has established a collaboration agreement with the Chinese National Hybrid Rice Research Center at Changsha, Hunan. The latter provides RiceTec with exclusive access to CMS and EGMS line germplasm. Restorer lines are based upon South China indica sources that are traced in large part to materials developed by the International Rice Research Institute. U.S. cultivars that are derived primarily from tropical japonica germplasm are used extensively as pollen parents for RiceTec's two-line rice hybrids. In 2000, RiceTec released XL6 as its first rice hybrid ever commercialized in the U.S. This hybrid yielded about 10 t ha-1 (about 20% higher than the best U.S. varieties). However, it suffered from high lodging and poor milling factors, preventing it from achieving a significant market share. In 2002, two new hybrids, XL7 and XL8, were released; these have less lodging and better milling quality. XL7 is also several days earlier than any variety currently grown in the market and offers producers the option of producing a ratoon crop or planting after wheat. XL8 competes with two leading long-grain varieties, Cocodrie and Wells, which account for 80% of the southern U.S. long-grain production. Challenges to the mechanization of hybrid rice in the U.S. are discussed in this paper.

Improving hybrid rice through anther culture and transgenic approaches
S. Balachandran, G. Chandel, M.F. Alam, J. Tu, S.S. Virmani, K. Datta, and S.K. Datta

Among the available genetic means to break the yield barrier in rice, agreement is widespread that rice hybrids have the potential to increase yield by 1-1.5 t ha-1 in farmers' fields under irrigated conditions. However, rice hybrids are affected by as many pests and diseases as common rice varieties. For the large-scale adoption of this technology, hybrids need to be resistant to the major pests and diseases prevailing in the target areas. Chemical control of yellow stem borer has been either ineffective or expensive, besides leading to environmental pollution. As an effective plant protection strategy, genetic engineering approaches now enable the production of insecticidal protein within the rice plant itself, which offers environment-friendly protection against insect pests. The incorporation of resistance genes in CMS or maintainer or restorer lines is expected to make hybrids resistant to the target disease or pest. While it is difficult to produce transgenic CMS plants every time, the transformed maintainer line with resistance genes will result in a resistant CMS line by backcrossing. The transgenic restorer line, in contrast, can be directly used for hybrid production.

Recently, we have successfully transformed two transgenic maintainer lines (IR68899B and IR68897B) and two restorer lines (MH-63 and BR-827-35R) with the truncated chimeric Bt gene, cryIAb (driven by 35S and PEPC promoters), and/or the hybrid Bt gene cryIAb/cryIAc (driven by the actin I promoter). These lines showed a wide range of expression (low to high) of Bt proteins, which were stably inherited. For BR-827-35R (T0 plants), protein expression was 0.61 mg g-1 of fresh leaf vis-à-vis 20 ng mg-1 soluble protein found in the Bt MH-63 line. A selected homozygous MH-63 Bt was hybridized with CMS line Zhenshan 97A to produce the first-ever hybrid Bt rice (Shanyou 63). The hybird rice was field-evaluated for the first time in Wuhan, China, in 1999-2000. The transgenic CMS restorer plant and its hybrid exhibited excellent protection against extremely high, repeated infestations of yellow stem borer and natural outbreaks of leaffolder. The transgenic hybrid not only showed high protection against both insects but also recorded 28.9% more yield than the non-Bt hybrid. Similarly, Xa21 has been incorporated in maintainer (B) and restorer (R) lines. The combinations of Xa21 and Bt will further enhance the yield of hybrid rice and ensure a pesticide-free environment and improved rice breeding. Further work on developing transgenic IR58025B with plant protection is in progress. Our recent research on nutrition (iron, protein, and provitamin-A) improvement and tolerance for abiotic stresses will probably diversify and improve the hybrid rice-breeding program.

Advances in understanding the genetic basis of heterosis in rice
Qifa Zhang and Zhikang Li

Findings revealed by the two most recent sets of studies designed to characterize the genetic basis of heterosis in rice are summarized in this paper. The first set of studies used a population of recombinant inbred lines (RILs) from a cross between Lemont and Teqing, the backcross populations of RILs with the two parents and two testcross populations. The second set of studies used an "immortalized F2" population produced by intercrossing RILs from a cross between Zhenshan 97 and Minghui 63. Heterosis measurements were used as data input for identifying a single-locus QTL and epistasis governing heterosis. Results from the two studies were in agreement that epistasis played an important role as the genetic basis of heterosis, but the types of epistatic gene actions identified in the two studies were different. The first set of studies suggested the importance of dominant types of interaction in the genetic basis of heterosis. While the second set of studies indicated the prevalence of additive-by-additive interactions, it was also shown that heterotic effects at the single-locus level in combination with the small advantageous effects of double heterozygotes at the two-locus level could adequately explain the genetic basis of heterosis in Shanyou 63. The implications of these findings were discussed.

Molecular approaches for fixing the heterozygosity of hybrid rice
Xuezhi Bi, J. de Palma, R. Oane, G.S. Khush, and J. Bennett

The heterotic yield advantage of hybrid rice is not available to poor farmers because costly F1 seed must be purchased every season. If the heterozygosity of the hybrid could be fixed, through some novel form of sexual or asexual seed formation, production costs would fall and farmers would be able to reproduce hybrid seed in their own fields. Here we consider two approaches: (1) a sexual pathway similar to that leading to permanent hybridity in a few plant genera such as Oenothera and Isotoma and (2) an asexual pathway based on the much more common phenomenon of apomixis. We have focused our experiments on the second pathway.

Apomixis is known among relatives of maize, wheat, and millet but is unproven for relatives of rice. In collaboration with CSIRO Plant Industry, IRRI is pursuing the goal of a synthetic form of apomixis that would be compatible with hybrid rice production ("one-line hybrid rice"). CSIRO has already accomplished fertilization-independent (FI) endosperm formation in Arabidopsis thaliana through mutagenesis and is now attempting to reproduce this achievement in rice through RNA interference. If this objective can be attained in conjunction with another major objective-FI embryo formation in the nucellus-we would be well on the way to achieving a form of FI apomixis for hybrid rice. However, when fertilization fails for genetic or environmental reasons, the ovary normally fails to expand. If this were to happen in FI apomicts, it would greatly restrict the size of the endosperm.

We have now succeeded in bypassing this important checkpoint in more than 50 independent transformants by using OsASP1::gus constructs. OsAsp1 is a putative aspartyl proteinase that is the rice orthologue of the barley nucellin gene. The OsASP1::gus transformants show two phenotypes: (1) total male sterility through poor pollen development and failure of dehiscence and (2) the formation of full-size pericarp and testa devoid of embryo and endosperm. They are, however, completely female fertile and have been propagated as hemizygotes. This discovery opens the way to FI formation of full-size endosperm. We report also on progress toward achieving FI embryogenesis in rice through ectopic expression of transcription factor genes such as OsLEC1.

Physiological bases of heterosis and crop management strategies for hybrid rice in the tropics
S. Peng, J. Yang, R.C. Laza, A.L. Sanico, R.M. Visperas, and T.T. Son

Commercial production of three-line hybrid rice has expanded in some tropical Asian countries. Heterotic level, crop management, hybrid seed production, and grain quality are the major limiting factors for the further expansion of hybrid rice in this region. Traits responsible for heterosis for grain yield are different under different growing conditions. Under favorable growing conditions, the higher yield of hybrid rice is attributed to greater biomass production. Under suboptimum growing conditions such as low radiation, a higher harvest index contributes to the higher yield of hybrid rice. Greater leaf area production rather than tiller production is responsible for the higher biomass accumulation and grain yield of hybrid rice compared with the best conventional check variety in the tropics. A high tillering rate causes the early vigor of hybrid rice grown in temperate China. Hybrid rice technology is more profitable in rice-growing areas with medium and high yields (³5 t ha-1). Hybrid rice is suitable for both the dry and wet seasons. Both heterosis and yield differences are greater at a high N rate than at a low N rate. Hybrid rice needs a different N management strategy from conventional rice to maximize yield and minimize disease damage. Leaf area and tiller and biomass production of A lines do not limit the seed yield of hybrid rice. In fact, crop growth could be too high to cause a reduction in the outcrossing rate. Widening transplanting spacing up to 20 ´ 30 cm and transplanting split tillers are feasible ways to reduce the seed requirement to as low as 2.6 kg ha-1 without reducing grain yield.

Public, private, and NGO-sector partnership for developing and promoting hybrid rice technology
M. Ilyas-Ahmed, Ish Kumar, B.C. Viraktamath, J.S. Sindhu, and Y. Yogeswara Rao

There is no greater incentive for public, private, and NGO-sector partnership in agricultural research and development (R&D) than to together meet the enormous challenge posed by global food security, which will require that our limited global resources be used effectively to develop sustainable systems that also conserve natural resources. Comparative advantages of each sector are to be identified and harnessed for achieving the common goal of global food security. In agricultural systems around the world, the roles of the public and private sector and the relationship between them is changing. Recently, there has been a growing awareness, in both sectors, of significant benefits that can be derived from partnership. Investment by the public sector in agricultural R&D has been declining, whereas investment by the private sector has been increasing in both developed and developing countries. Concern is growing that the current level of investment, by both the public and private sector pooled together, may also be inadequate to ensure food security in the coming decade, thus highlighting the urgent need for partnership.

In this article, several ways to establish effective partnership are outlined. The current status of hybrid rice in tropical Asian countries has been considered and the strength and level of involvement of the public, private, and NGO sectors have been described. The comparative advantages of these sectors in the development and promotion of hybrid rice are enumerated. As a case study, the current status of partnership among these sectors in India is given. Finally, a model for establishing effective partnership for various activities among different sectors for developing and promoting hybrid rice in tropical Asian countries is proposed.

The public sector, which has a comparative advantage in technology generation, can complement the expertise of the private sector in large-scale seed production and distribution and both sectors, along with NGOs, can jointly undertake technology transfer to develop and promote hybrid rice. IRRI can play an effective role in developing and implementing a detailed action plan for partnerships for the member countries under the IRRI/ADB project on hybrid rice.

Policy and institutional support to expedite the development and use of hybrid rice technology
M.A. Sombilla, B. Ba Bong, B. Mishra, L.S. Sebastian, and S.B. Siddique

Alarms are once more being sounded for the beginning of a long-term trend in which staple grain harvests, particularly of rice, have slackened and have failed to outpace demand in several countries in Asia. Despite the remarkable production performance of the region in the past, several countries were threatened by declining rice supply and demand balances while others continued to rely on their neighbors to overcome their increasing domestic deficits. Although recent projection studies indicated a rosy baseline scenario for rice in which Asia would have a net surplus of about 16.4 million tons in 2025, the estimates were contingent on the continuation and maintenance of the favorable interrelationship of several factors that affect trends in technology development, flows in investment, environmental protection, and human behavior. Failures in any one of these factors could have a devastating impact on the future of rice. Hence, the big challenge is to promote an environment to prevent the key development factors from turning against the agricultural sector. The adoption of appropriate and well-balanced policy reforms could help ensure sustained rice production growth and equity in Asia.
The great potential of hybrid rice technology to increase rice production has been exhibited in China and other countries. However, some constraints and challenges hamper further development and the wide-scale adoption of the technology. These constraints have both technical and policy dimensions. The technical constraints include, primarily, further improvement of breeding lines not only to further increase and stabilize their yields but also to improve their quality. The policy constraints involve the challenge of promoting and intensifying dialogue with policymakers to increase and sustain investments in hybrid rice research, strengthen collaboration among various stakeholders, and create a positive environment for the expanded production and marketing of hybrid rice seed.

As the title suggests, this paper aims to identify policy and institutional options and strategies to help expedite and sustain the development and wide-scale use of hybrid rice technology, primarily in the original six member countries of the International Hybrid Rice Network: Bangladesh, India, Indonesia, the Philippines, Sri Lanka, and Vietnam. The options and strategies vary across the countries because of the differential stage of development of hybrid rice technology. In all of them, however, there are three keys to successful commercialization of the technology: higher and more stable profits, availability of an appropriate and dependable package of technology, and a strong public-private-sector linkage.

International Task Force on Hybrid Rice
S.S. Virmani, Dat Van Tran, and J.S. Sindhu

The establishment of an International Task Force on Hybrid Rice (INTAFOHR) was first proposed during the 2nd International Symposium on Hybrid Rice held at IRRI in 1992 to promote the technology outside China. Subsequently, this was endorsed by several countries in various international fora organized by FAO and IRRI. Subsequently, goals, objectives, participation, strategies, and implementation arrangements for INTAFOHR were proposed. A hybrid rice network project was launched in 1998 in the form of an IRRI-ADB project titled "Development and Use of Hybrid Rice in Asia." This project aimed to expedite the development and use of hybrid rice technology in tropical Asia. Three international agencies (IRRI, FAO, and APSA) and six member countries (Bangladesh, India, Indonesia, the Philippines, Sri Lanka, and Vietnam), possessing high proportions of irrigated area and a high labor/land ratio, collaborated in the project. In 2000, China also joined as a collaborating partner in this project. This paper describes the roles of the coordinating partners, major achievements, and future directions and strategies of the project.

Hybrid rice: how to go forward?
Dat Van Tran

All stakeholders, including decision-makers, rice scientists, extension workers, F1 seed producers, and rice growers, have recognized the advantages of hybrid rice technology. Rice hybrids not only outyield commercial inbred varieties by 15% to 20% but also generate rural employment opportunities and potential land saving for other uses. These promising benefits have encouraged many governments to focus on developing and disseminating hybrid rice, especially in countries where rice imports have become a burden on the national economy, population continues to grow, and expansion of rice area is not possible. However, progress in large-scale adoption under tropical conditions seems slow, although, in countries such as Vietnam and India, the government has taken aggressive measures to solve technical and nontechnical problems.
How should we go forward to expedite farmers' adoption of this technology? This paper reviews the factors thought to be responsible for the slow progress and suggests that technological constraints could be the major limitation to technology transfer. Obviously, technological development has not been able to keep pace with the public interest in hybrid rice, in terms of research activities and hybrid seed production. Regardless of substantial improvement in human capacity, national hybrid rice programs would stagnate should progress in research be slow for a long period, if the level of F1 seed yield remained unchanged, and if the government decreased its interest in the matter.

Hybrid rice research and development in Bangladesh
A.W. Julfiquar, M. Jamil Hasan, A.K. Azad, M. Anwar Hossain, and S.S. Virmani

Hybrid rice technology has been recognized as a viable technology for meeting the increased demand for rice in Bangladesh. Research and development (R&D) of hybrid rice technology began in 1993 with the introduction of rice hybrids and parental materials from IRRI. The technology was initially constrained by (1) a lack of conviction by researchers, policymakers, and administrators about the economic viability of the technology, (2) a lack of trained human resources to develop and use the technology locally, (3) a lack of knowledge about its seed production technology, and (4) the poor performance of introduced hybrids over conventionally bred released varieties. Over the past few years, some of these constraints have been overcome and the government is convinced about the potential of this technology and is committed to giving priority to accelerating hybrid rice R&D.

Several IRRI-developed hybrids were tested in preliminary to advanced yield trials during the last four years at different locations in Bangladesh. The objective was to test the yield potential of the IRRI-developed hybrids under ideal boro conditions and then select the best hybrids for Bangladesh. Two hybrids, IR69690H and IR68877H, were nominated in the National Hybrid Rice Trial in the boro season (1999-2000) coordinated by the Seed Certification Agency and the Department of Agriculture Extension. After extensive evaluation, the National Seed Board recommended IR69690H as BRRI hybrid dhan-1, primarily in the Jessore and Barisal region, with a view to subsequent cultivation in farmers' fields in other regions of the country.
Seed production is an important component of the development and use of hybrid rice technology. Results of the past two years show the possibility of raising the yield of F1 hybrid seed on average to 2.0 t ha-1.

The Bangladesh Rice Research Institute (BRRI) maintains close collaboration with IRRI and receives technical support, including seed materials and training, to establish an effective hybrid rice program in Bangladesh. Nucleus and breeders' seeds are being produced under the supervision of the plant breeder at BRRI and those are being supplied to public and private seed-producing agencies for commercial seed production. BRRI is sharing its parental materials of promising hybrids with NGOs to multiply seed of parental lines and produce F1 hybrid seeds.
This paper reports on the progress of hybrid rice R&D in Bangladesh, including recent accomplishments in technology generation and seed production.

Hybrid rice achievements and development in China
Ma Guohui and Yuan Longping

With the hybrid rice research begun by Professor Yuan Longping in 1964, only in the past 12 years did China become the first country to commercialize hybrid rice in the world. Now, the annual area under hybrid rice is about 15.5 million ha, 50% of the total rice area, and the production of hybrid rice occupies nearly 60% of the total rice production in China. Hybrid rice has a yield advantage of more than 20% over conventional rice.

The technology of hybrid rice seed production was optimized in 1976, resulting in national average yield rapidly increasing from 274.5 kg ha-1 in 1975 to 2.25 t ha-1 in 1990. Yield on a large scale can be 3.0-4.5 t ha-1 and the area ratio of A line multiplication to F1 hybrid seed production to F1 hybrid commercial production is about 1:50:6,000, which greatly lowers the cost of hybrid seeds.
China has made great achievements in research on two-line hybrid rice. Many viable PTGMS lines developed that possess the characteristics of low critical sterility-inducing temperature (CSIT) and safe hybrid seed production have been technically identified and approved. Up to now, more than 20 two-line hybrids have been released to farmers. The area of two-line hybrid rice increased to 2.67 million ha in 2001, about 17.2% of the total hybrid rice area and showing a yield increase of 5-10% compared with current three-line hybrid rice as well as better grain quality and good pest resistance.

Some constraints exist in three-line hybrid rice, such as yield and area stagnation, the lack of japonica hybrid rice with strong heterosis, and the need for grain quality as well as sources of male sterility-inducing cytoplasm to develop better CMS lines.
One of our breeding strategies is to use heterosis in hybrid rice breeding by means of systems such as the three-line, two-line, and one-line methodology and to improve heterosis by means of intervarietal, intersubspecific, and distant hybrids.

Hybrid rice research and development in Egypt
A.O. Bastawisi, H.F. El-Mowafi, M.I. Abo Yousef, A.E. Draz, I.R. Aidy, M.A. Maximos, and A.T. Badawi

Egypt reached its ceiling in rice productivity in 1987. Although the national average yield in 2001 was 9.3 t ha-1, research trials showed an increase in the productivity of hybrids ranging from 15% to 20% in normal soils and from 20% to 30% in saline soils over inbreds.

Five Egyptian hybrid combinations were recognized under multilocation yield trials; some of these are highly adapted to normal soils, some to saline soils, and others to both. The two-line system (PGMS and TGMS) is in the testcrossing stage. Nucleus seed for initial seed production and purification in three hybrid combinations was started.

In the training of human resources, 32 researchers were trained on the scientific basis and applied technology for hybrid rice breeding and seed production. By the end of 2002, extension and seed production specialists, seed growers, and farmers will have been trained.

The economic evaluation and advantage of hybrids vis-à-vis inbreds are discussed.

Hybrid rice development and use in India
B. Mishra, B.C. Viraktamath, M. Ilyas Ahmed, M.S. Ramesha, and C.H.M. Vijayakumar

Recognizing the potential of hybrid rice to enhance production and productivity, India launched a national program in 1989 for the development and large-scale adoption of hybrid rice to sustain self-sufficiency in rice production. As a result of intensive efforts to develop and evaluate hybrids during the last decade, 16 hybrids have been released for commercial cultivation. A systematic multilocational evaluation of released hybrids during three seasons indicated that six were high-yielding and widely adaptable. These, in the order of their performance, are KRH-2, PHB-71, Sahyadri, PA6201, NSD-2, and DRRH-1. A significant achievement has been the development of the first super-fine-grain aromatic hybrid, Pusa RH-10. This hybrid has a 40% yield advantage over the highest-yielding basmati check variety, Pusa Basmati-1, with comparable grain and cooking quality characteristics. Many promising hybrids with enhanced magnitude of heterosis and better grain quality are in the pipeline. Farmer adoption of hybrid rice has been much slower than expected because of several constraints. India is now estimated to have around 200,000 hectares under hybrid rice annually.

With the added emphasis on parental line improvement through recombination breeding and population improvement approaches by developing and using intra- and intersubspecific germplasm, more than 3,000 genetically diverse promising derivatives have been developed and are being used to develop experimental hybrids. In recent years, many promising hybrids identified are based on indigenously developed CMS lines such as PMS 10A, PMS 12A, Pusa 6A, CRMS 31A, DRR 2A, and APMS 6A. New TGMS lines developed through recombination breeding are at the final stage of evaluation (F6 generation). Initial evaluation of intersubspecific hybrids involving indica/tropical japonica derivatives indicated enhanced heterosis of 5-10% over indica/indica hybrids.

Refinement of the seed production package has resulted in average seed yields improved from 1.5 to 2.0 t ha-1 by experienced seed growers in large-scale seed production. Effective public- and private-sector partnership in seed production has been a key to the successful adoption of this technology. Of the estimated 3,000 t of hybrid seed produced in the country annually, more than 90% is produced by the private seed sector.

Efforts to transfer the technology already developed, through extensive compact block front-line demonstrations and training programs, has created awareness and motivated farmers to begin adopting hybrid rice in target states. Policymakers and senior research managers have been sensitized and the government has planned to popularize hybrids on a priority basis during the tenth five-year plan period (2002-07).

Improving grain and cooking quality characteristics of hybrids, incorporating resistance to some major pests and diseases, enhancing the magnitude of heterosis, and further increasing average seed yields to 2.5 t ha-1 on a large scale to reduce seed costs are the research priorities during the next five years. Policy interventions by the government for increased support, aggressive popularization of hybrids, and assured procurement of hybrid rice at a minimum support price are needed. If these problems can be solved, hybrid rice could be cultivated on 3-4 million hectares in India during the next decade to partially sustain food security.

Hybrid rice research and development in Indonesia
Suwarno, N.W. Nuswantoro, Y.P. Munarso, and M. Direja

The plateauing trend of rice production and the success of the development and commercial use of hybrid rice technology outside China, including India, Vietnam, and the Philippines, encouraged the government of Indonesia to intensify research on and development of hybrid rice in 1998. Some hybrid rice varieties, including two public and five private ones, were released recently. The public hybrids, Rokan and Maro, yielded 1.0-1.5 t ha-1 higher than the inbred variety IR64 in suitable conditions. The hybrids are susceptible to the major pests brown planthopper, bacterial leaf blight, and rice tungro virus, and may not always express heterosis across locations. Cultivation of the hybrids should be followed by application of integrated pest management, but cultivation is not recommended in the endemic area of the pests. The breeding program of developing better hybrids is conducted by public as well as private institutions and some promising hybrids have been developed. The development of CMS lines resistant to the major pests is also in progress. A program for disseminating hybrid rice technology has begun in 15 districts identified as having potential for hybrid rice cultivation. The total area of hybrid rice cultivation during this season is about 60 ha and will be increased gradually to at least 500,000 ha in the wet season of 2004-05. A seed system suitable for hybrid rice, including seed production, inspection, and certification, is being established.

Hybrid rice research and development in the Democratic People's Republic of Korea
Ri Tae Sik and Rim Yun Uk

In recent years, the development of hybrid rice has focused on intersubspecific hybrids for high-yielding rice varieties. Yield per hectare in intersubspecific hybrids may increase by 30-50% over that of the best check variety because of its high heterosis. Many constraints and challenges exist, however, for the development of intersubspecific hybrids. The main problem in breeding a hybrid between two subspecies by the three-line method is to solve the problem of unstable sterility, F1 semisterility, and delay of heading date and to improve hybrid seed production.

Hybrid rice research and current status in Korea
S.J. Yang, Y.C. Song, and H.P. Moon

Hybrid rice research in Korea started in the early 1970s, but the national hybrid breeding program started in the early 1980s. The Korea-IRRI collaborative research project has played a significant role in the progress of the hybrid rice program. Hybrid breeding was active in 1985, when some wild abortive (WA)-type cytoplasmic male sterile (CMS) systems were introduced. Since then, tongil-type (derived from indica/japonica) cultivars Samgangbyeo, Taebagbyeo, Cheongcheongbyeo, and Milyang 23 and some breeding lines have been identified as maintainers. Most tongil-type cultivars showed good fertility restoration for WA-type lines. Several experimental hybrids yielded 4-34% more than elite cultivars until 1997. However, Korean consumers do not prefer their high chalkiness and high-amylose grain, so the tongil-type hybrid rice breeding program was reduced. Nevertheless, WA-type lines, maintainers, and restorers are being continuously developed. Japonica CMS lines were developed by transferring the boro-type CMS source into elite Korean cultivars and breeding lines in 1994 and backcrossing is being used to develop japonica restorer lines with some success. Forty CMS lines have been developed during the past six years by transferring WA types and COA types into leading Korean cultivars and breeding lines. Twenty lines are tongil-type, derived from indica/japonica hybridization, and 17 lines have the japonica genetic background. Commercial rice hybrids are not yet available to farmers because poor grain quality and labor-intensive seed production technology are major constraints. We hope to have 10 t ha-1 of milled rice with the use of hybrid rice in the 21st century; therefore, hybrid rice research in Korea restarted in the early 2000s.

Hybrid rice breeding in Russia
I.K. Gontcharova and S.V. Gontcharov

Research on hybrid rice problems has been carried out for a long time in Russia, but it is mainly theoretical. Hybrid rice research faces several problems in Russia: (1) the lack of suitable restorers in japonica rice, which is mainly grown in Russian conditions; and (2) the low level of allogamy does not allow using hybrid seed for direct sowing; and (3) combining ability evaluations and screening for hybrid combinations with high productivity or heterosis.

During the first period of research, we studied available CMS sources for possible use. Wild abortive (WA) cytoplasmic male sterility was transferred into our local lines. All tested elite lines appeared to be maintainers. So, several CMS lines were successfully developed and a restorer line breeding program started. Rf genes were transferred from IR54. Studying the genetics of fertility restoration showed that at least three different Rf genes are found in the japonica background: two genes described in the literature and the last one with regulation activity-the dominant allele of this gene allows other Rf genes to act. We think that indica WA-CMS lines have this dominant allele too.

The low level of cross-pollination in rice is a serious problem in Russia. Only direct sowing is used in commercial rice production. Therefore, we studied floral behavior and allogamy-connected traits. Studying the inheritance of allogamy-connected traits showed that the inheritance of floral and flag-leaf traits was determined by polygenes in many cases. Additive variance prevailed in a population for all traits except for two characteristics (stigma exsertion and stigma diameter). Gene interaction (complementary epistasis) was also noticed for all traits studied. For stigma diameter, dominance was higher; for all other traits, dominance was lower. As proved by ANOVA, the number of recessive genes in varieties is correlated with the character value for all traits except for flag-leaf length. There was overdominance in the locus of floral traits. Length and breadth of the flag leaf were partially dominant.

We analyzed the correlation between all studied allogamy-connected traits and only stigma traits were closely correlated with each other. Therefore, when breeding lines with increased stigma characters, it is enough to measure only one parameter because of the high correlation between them.
Different patterns of flowering behavior during the day have a large effect on outcrossing because usually the peak of flowering of CMS lines and fertile lines does not match. We found three different patterns of flowering behavior among CMS lines and one of them was the best but the least frequent among the lines.

During two years, we studied our germplasm collection for allogamy-connected traits to screen and develop lines with allogamy good enough for industrial hybrid seed production. Many samples selected as donors of allogamy traits also have good productivity traits and a suitable vegetative period. This allows us to recommend the use of the given samples for hybridization and in the search for hybrid combinations with high heterosis.

Combining ability evaluation and screening of hybrid combinations for the highest productivity are obligatory in hybrid breeding. We studied 51 hybird combinations. Among these, 15 showed 15% heterosis for the following productivity traits: panicle length, spikelet number per panicle, panicle weight, plant height, and others.

We can conclude that the development of hybrid rice for moderate climatic conditions (south of Russia) is quite possible, although this will require strong breeding efforts.

Hybrid rice research and development in Myanmar
Khin Than Nwe, Myint Yee, Hmwe Hmwe, Myint Aung, and Aye Aye Myint

Rice production is the backbone of the national program on economic development as well as for attaining rice self-sufficiency in Myanmar. To increase rice production, hybrid rice technology is considered as a new approach. Under the Myanma Agriculture Service (MAS), a hybrid rice research and development program started at the Central Agricultural Research Institute (CARI) in 1991 with collaboration from the International Rice Research Institute (IRRI) and strengthened its activities with cooperation and collaboration from seed companies from China and Japan and technical support of the Food and Agriculture Organization (FAO).

The hybrid rice breeding program focuses on developing parental lines better adapted to local environmental conditions. Two cytoplasmic male sterile (CMS) lines, IR58025A and IR62829A, were initially introduced from IRRI and IR58025A was found to be more adaptable to local conditions. Two more lines, IR68897A and IR68887A, were selected as good CMS lines from further evaluation. Some prospective maintainers and restorers were identified and some were used in the backcross nursery. Yield tests were carried out at CARI and different research stations to evaluate the yield and agronomic characters of experimental hybrids and introduced hybrids. Research to determine the optimum male:female row ratio, the gibberellic acid dosage test, and a study on the best time for the use of GA3 for CMS lines are being done at CARI. Some hybrid combinations from the International Rice Hybrid Observational Nursery seem to be promising.

Commercial F1 seed production was handled by MAS and Chinese seed companies. At present, the area planted to hybrid rice has decreased because Chinese hybrids were observed to have some constraints: poor grain quality and susceptibility to disease.

In Myanmar, hybrids must show about 20% yield heterosis over existing inbred cultivars with acceptable grain quality. Myanmar also became a member country of the IRRI-ADB hybrid rice project phase II.

Hybrid rice research and development in Japan
T. Takita

No abstract

Hybrid rice research and development in Sri Lanka
S.W. Abeysekera, S.N. Jayawardena, K.D.S. Kiriwaththuduwage, and D.S. de Z. Abeysiriwardena

Hybrid rice (HR) technology is considered as the best option to overcome yield stagnation in rice in Sri Lanka. In Sri Lanka, the research and development (R&D) program on hybrid rice began in late 1994. Initially, this program was academic rather than goal-oriented. However, in 1996, the hybrid rice R&D program was revised to be goal-oriented in collaboration with the International Rice Research Institute (IRRI). The R&D program on hybrid rice has three main components: developing and evaluating F1 hybrids, developing locally adapted management practices for seed production, and developing locally adapted management practices for HR cultivation.

Of the 60 IRRI, Indian, and locally bred CMS lines evaluated, 48 were adaptable and stable under local conditions. Some well-adapted varieties were identified as maintainers or good restorers for the wild abortive cytosterility system. Some elite maintainers identified were BR168-2B-23R, BR827-35-2-1-1-1R, IR54742-22-19-3R, IR34686-179-1-2-1R, Bg99-350, and H4.
Two CMS lines, BgCMS 1A and BgCMS 2A, have been developed at the Rice Research and Development Institute (RRDI). These lines have small round (samba) grain type with 100% sterility and 100% panicle exsertion rate (PER%). BgCMS 1A and BgCMS 2A are producing 14 and 18 panicles per plant and 203 and 243 spikelets per panicle, respectively. Three hybrid rice varieties, BgHR1, BgHR6, and BgHR12, developed at RRDI had significantly higher yields than the standard inbred varieties Bg 357 and Bg 403. The standard heterosis of these hybrids ranged from 28.9% to 58.6%. Grain quality characteristics of these hybrids were superior to those of inbred checks. Similarly, out of the 21 experimental hybrids evaluated in yield trials during the 2001 minor season, eight lines outyielded the standard checks Bg 359 and Bg 403 by more than 1 t ha-1. Grain quality characteristics of these hybrids were also superior to those of the check varieties. Of the 11 IRRI-bred hybrids tested during the 2001 minor season, two, IR69686H and IR75217H, outyielded the local and international check varieties of the same growth duration by more than 1 t ha-1. Grain quality characteristics of the tested hybrids were comparable with those of the recommended check varieties.

Comprehensive studies on cultural management, fertilizer management, and reaction to pests and diseases of the promising hybrids are being conducted at RRDI to develop an appropriate package of management practices for hybrid rice cultivation. Average seed production in large-scale CMS plots, in small-scale CMS plots, in large-scale F1 plots, and in small-scale F1 plots was 0.54-1.2 t ha-1, 0.59-1.09 t ha-1, 0.32- 1.42 t ha-1, and 0.21-0.74 t ha-1, respectively.

Hybrid rice development and use in Vietnam
Nguyen Tri Hoan and Nguyen Huu Nghia

No abstract

Hybrid rice technology and achievements in Iran
G.A. Nematzadeh, M. Sattari, A. Valizadeh, A. Alinejad, and M.Z. Nori

Rice is a second main staple crop after wheat in Iran. More than 600,000 hectares of cultivated land are allocated to rice cultivation. Rice cultivation has had three distinguished periods in Iran: (1) before 1979, when all local rice varieties (tall, low-yielding, and excellent quality) were used; (2) 1978-2000, when high-yielding semidwarf varieties; either local improved ones such as Dasht, Nemat, Neda, and Sepeadroad or introduced varieties such as Amo13, Khazar, and others, were used, besides local varieties; the contribution of improved semidwarf high-yielding varieties to total rice production varies highly annually and it has depended on the quantity of imported rice; and (3) a period that actually began in 1991 with the introduction of some CMS lines from IRRI, when hybrid rice technology began in Iran.

V20A, W32A, IR58025A, and IR28298A were the first CMS lines introduced from IRRI. Our primary attempt with hybrid rice technology started in 1991 and so far we have improved at least six new CMS lines from well-adapted high-yielding improved varieties such as Kazar A, Nemat A, Neda A, Dasht A, and Champa A. The development of restorer lines such as IR24R, IR60969R, IR56R, and Amol 1R for Neda A, including a study of allogamy-associated traits, for all new improved CMS lines was carried out. This is our starting point for hybrid seed production. Two books on hybrid rice-on hybrid seed production and a hybrid rice breeding manual-were also translated into Persian to explain hybrid rice technology in Iran.

Hybrid rice development and use in the Philippines, 1998-2001
E.D. Redoña, F.M. Malabanan, M.G. Gaspar, J.C. de Leon, and L.S. Sebastian

The development and use of hybrid rice technology as a major approach for further increasing rice productivity gained national prominence in the Philippines with the launching of a national hybrid rice program under the Department of Agriculture (DA) in January 1998. With the Philippine Rice Research Institute (PhilRice) at the helm, activities on research and development (R&D), seed production, training, technology demonstrations, and information dissemination on hybrid rice were undertaken nationwide. The hybrid rice research program at PhilRice, with 18 multidisciplinary projects conducted by more than 80 scientists, was strengthened with the establishment of a hybrid rice center and collaboration with IRRI and other national programs, the private sector, and Chinese hybrid rice R&D institutions. Season-long training activities for prospective hybrid seed growers in strategic areas led to the organization of cooperatives that began producing hybrid seed on a commercial scale starting in 2000. The more than 1,000 trainers, agricultural technicians, seed inspectors, and local government officials that were trained on hybrid rice seed production and cultivation techniques extended this new knowledge to hybrid rice seed growers and cultivators. On large-scale technology demonstration farms, the hybrid rice Mestizo showed more than a 1.2 t ha-1 yield advantage over the best inbred varieties during 1997-99 and 1.3 t ha-1 in commercial cultivation areas during the 2001 wet season. Private-sector participation also increased in the areas of breeding, seed production, and technology promotion. In 2002, the hybrid rice program was embraced as the banner program for agricultural modernization by the Philippine government.

Recent progress in breeding super hybrid rice in China

Advances in hybrid rice research and development in the tropics

Breeding strategies to enhance heterosis in rice

Two-line hybrid rice breeding in and outside China

Opportunities for and challenges to developing and using hybrid rice technology for temperate countries

Improving grain quality in hybrid rice

Opportunities for an challenges to improving hybrid rice seed yield and seed purity

Hybrid rice for mechanized agriculture

Improving hybrid rice through anther culture and transgenic approaches

Advances in understanding the genetic basis of heterosis in rice

Molecular approaches for fixing the heterozygosity of hybrid rice

Physiological bases of heterosis and crop management strategies for hybrid rice in the tropics

Public, private, and NGO-sector partnership for developing and promoting hybrid rice technology