Program Report For 1999

Research programs Irrigated rice ecosystem
BREEDING TO BREAK YIELD CEILINGS: A SYSTEMS APPROACH 4 Comparison of dry and wet season performance of new plant type lines (APPA, PBGB) 4 Dry season NPT yields 5 Wet season NPT yields 5 Hybrid rice (PBGB) 5 Release of IRRI hybrids 6 Elite hybrids 6 New cytoplasmic male sterile lines 7 Development of thermosensitive genic male sterile (TGMS) lines 7 IR lines named as varieties 7 Hybrid rice network 7 Genetic divergence of maintainer and restorer lines used to breed tropical rice hybrids (PBGB) 7 Association between SSR diversity, pedigree record, quantitative trait variation, and hybrid performance (PBGB) 9 Toward a synthetic apomixis for rice (PBGB) 10 Inducing the asexual embryo in the rice nucellus 11 Arresting the sexual embryo development 11 Characterization of root system in a new plant type (APPA) 13 Japonica germplasm utilization for value added (PBGB) 14 Inheritance of RTSV resistance in TW5, the nearisogenic line from Utri Merah (EPP, PBGB) 14 Breeding rice for resistance to tungro (PBGB) 15 SUSTAINING SOIL QUALITY IN INTENSIVE RICE SYSTEMS 16 Yield declines in long-term rice experiments in Asia (SS, SWS) 16
IMPROVING THE PRODUCTIVITY AND SUSTAINABILITY OF RICE-WHEAT SYSTEMS 18 Improving rice-wheat systems (SWS, APPA, EPP, SS) 18 Ecoregional approach to natural resource management (APPA, SS) 18 Soil- and seedborne pathogens of rice in a rice-wheat system (EPP) 19 Field sampling and disease assessment 19 Disease incidence in farmers fields 19 Results from a principal component analysis 20 Interpretation of trends 20 IMPROVING PEST MANAGEMENT 20 Determinants of farmers pest management decisions (EPP) 20 Field evaluation of prototype Tungro Screen B kits (EPP) 21 Alternative prey in irrigated rice: implications for biological control of hoppers (EPP) 21 PROGRESS OF UNREPORTED PROJECTS 22 Increasing water use efficiency in rice culture (SWS, APPA, SS) 22 Coping with global climate change: reducing methane emission from rice fields (SWS) 23 IRRIGATED RICE RESEARCH CONSORTIUM Coordination 23 Research 23 PROGRAM OUTLOOK 23 23
Irrigated rice ecosystem
Rice production will have to increase as much as 60% in the next 2030 years to meet growing demand for food, sustain global food security, and mitigate poverty. Most of that production will come from intensive irrigated rice systems. Competition for land and labor will require increased productivity to meet needs for rice. Rates of yield increase have recently declined in some countries, indicating that further increases in productivity may be difficult to achieve. A critical issue is maintenance of environmental quality while achieving increases in rice productivity. The irrigated rice ecosystem research program focuses on q breaking current yield barriers, q increasing efficiency of crop production inputs, q closing the yield gap and sustaining the irrigated lowlands, and q mitigation of the interaction between irrigated rice and global and atmospheric changes.
Breeding to break yield ceilings: a systems approach

Comparison of dry and wet season performance of new plant type lines S. Peng, G.S. Khush, R. Visperas, and A. Pamplona We reported in 1998 that the average yield of 43 new plant type (NPT) lines was 7.5 t ha1 as compared with 8.1 t ha1 of check cultivar IR72 in the dry season (DS). However, 51 NPT lines had an average yield of 5.2 t ha1, higher than 5 t ha1 of the check cultivar IR72 in the wet season (WS) (Table 1). These data suggest that the relative performance of NPT is better in WS than in DS. Because of limited irrigation, the area planted to rice during DS is decreasing in Southeast Asia while the WS area is increasing. That makes increasing WS yield by developing new rice cultivars as important for overall rice production as increasing yield potential in DS. Fifty-two lines were grown in 1999 DS and 58 NPT lines were grown in 1999 WS. The check cultivar was PSBRc 52 in DS and IR72 in WS. For DS, 22-d-old seedlings were transplanted on 22 Dec 1998. WS transplanting was done 1 Jul 1999 with 21-d-old seedlings. Hill spacing was 0.10 0.15 m with 1 seedling hill1. Fertilizer N was 120 kg ha1. Samples were taken at maturity from a 0.5-m2 area to determine panicle number, spikelets per panicle, grain-filling percentage, and 1000-grain weight. Grain yield was determined from a 5-m2 area and adjusted to moisture content of 0.14 g H2O g1 fresh weight.
IRRI program report for 1999
Table 1. Minimum, maximum, and average yield (t ha1) of new plant type (NPT) lines and a check cultivar grown in the dry (DS) and wet season (WS) of 1998-99. The check cultivar was PSBRc 52 in the dry season of 1999 and was IR72 in the other seasons. IRRI, 1999. 1998 DS NPT lines (no.) Minimum NPT yield Maximum NPT yield Av NPT yield Yield of check NPT 3 checka (no.)
a
1999 WS 51 3.7 7.7 5.2 5.0 30 DS 52 3.7 7.4 5.6 6.7 4 WS 58 2.9 6.5 4.9 4.3 49
43 6.6 8.8 7.5 8.1 7
Number of NPT lines with the same or greater yield than the check.
DRY SEASON NPT YIELDS
Four NPT lines produced the same or greater yield than PSBRc 52 in 1999 DS. NPT line IR70479-452-3 had highest yield at 7.4 t ha1, about 10% higher than PSBRc 52. Daily average radiation from January to April was 12% lower than the 10-year average and yields were low. Grain-filling percentage limited the yield of PSBRc 52, which was below 70%. There was a negative relationship between grain-filling percentage and spikelets per panicle. Six NPT lines (IR66160-121-4-4-2, IR66160-121-4-5-3, IR65600-42-5-2, IR66160-121-4-1-1, IR68552-843-1, and IR65600-54-6-3) had better than 75% grain filling but their panicle size was small, ranging from 110 to 135 spikelets panicle1. All NPT lines had bigger panicles than PSBRc 52. Nine NPT lines had more than 200 spikelets per panicle, but their grain filling was only 25-50%. Six NPT lines had 150-200 spikelets panicle1 and 6575% of grain filling. The next step in the NPT breeding program is to improve the grain filling of those lines.
WET SEASON NPT YIELDS
IR72. The results confirm that performance of NPT lines is better in WS than in DS. There are several reasons that may explain the better performance of NPT lines in WS. q NPT lines have better lodging resistance than the semidwarf indica type. More lodging occurs in WS than in DS. q Most NPT lines have larger panicles than the indica type. In WS when solar radiation is limiting, panicle-weight type is more favorable than panicle-number type because radiation has a large effect on tillering capacity and productive tiller percentage. q NPT lines usually have a better canopy structure for light interception than the semidwarf indica type, an important characteristic when light is a limiting factor. q Preliminary results suggest that NPT lines have better shading tolerance than the indica type in terms of single-leaf photosynthetic rate. Hybrid rice S.S. Virmani, R. Toledo, C. Casal, R. Ona, D. Sanchez, M. Nas, and M. Ilyas Ahmed Hybrid rice research at IRRI seeks to exploit the phenomenon of hybrid vigor to increase yield potential of rice cultivars beyond that of high-yielding semidwarf varieties. During 1999, 531 elite inbred lines were tested for their ability to maintain sterility or restore fertility of three cytoplasmic male sterility (CMS) systems (CMS-WA, CMS-ARC, CMS-mutagenized IR62829B) used in the breeding program. In all, 1,554 test crosses were evaluated and 330 new backcrosses initiated to develop new CMS lines in BC1 to BC6 generations. We evaluated 454 experimental hybrids in observational yield trials, 245 in preliminary yield trials, and 61 in advanced yield trials. Twenty rice hybrids were nominated for national trials by the Philippine Rice Research Institute (PhilRice) and 37 hybrids were nominated for inclusion in the International Rice Hybrid Observation Nursery (IRHON). Nucleus and breeder seeds of 81 CMS lines and 31 restorer lines were also produced for sharing with public and private sector institutions working on hybrid rice in national programs.
Forty-nine NPT lines produced the same or greater yield than IR72 in 1999 WS. The average yield of NPT lines was 4.9 t ha1 compared with 4.3 t ha1 for IR72. NPT line IR68552-55-3-2 produced the highest yield of 6.5 t ha1, about 50% higher than
RELEASE OF IRRI HYBRIDS
NEW CYTOPLASMIC MALE STERILE LINES
The IRRI hybrid IR69690H (IR58025A/BR 82735-3-1-1-1R), earlier released in Maharashtra State of India, was released for the Red River Delta in Vietnam, the first IRRI hybrid released in northern Vietnam. IR69690H has better grain quality and adaptability to the summer season than (and yield comparable with) Chinese hybrids already commercialized in the region. IR69690H has also been registered (along with IR68877H) for on-farm testing in Bangladesh.
ELITE HYBRIDS
Nine hybrids showed significant yield advantage in advanced yield trials at IRRI during 1999. Three of those hybrids yielded consistently high in both seasons (Table 2).
Sixteen new CMS lines (Table 3) were designated possessing WA (6), Gambiaca (5), and Dissi (5) cytoplasm, indicating a diverse cytoplasmic base of new IRRI-bred CMS lines. A CMS line IR70369A, designated in 1994, was found stable for complete pollen sterility, has good outcrossing, good phenotypic acceptability, acceptable grain quality (without aroma), and good combining ability. During 1999, several heterotic combinations derived from IR70369A were identified in observation and preliminary yield trials. Evaluation of maintainer lines for different biotic stresses resulted in identification of IR69618B and IR72079B lines possessing multiple resistance to diseases and insects and showing high phenotypic acceptability score. The corresponding CMS lines had fair to excellent outcrossing scores. These lines will be used extensively to develop new heterotic combinations.
Table 2. Growth duration and yield of three elite rice hybrids in comparison with check variety PSBRc 28 in advanced yield trials. IRRI, 1999. 1999 DS Hybrid or check variety 1999 WS
Growth Yield Growth Yield duration (t ha1) duration (t ha1) (d) (d) 112 110 111 115 0.69 0.91 7.2 7.0 6.9 6.2 115 111 113 116 .15 .20 4.6 4.6 4.9 4.3
IR68888A/IR62161-184-3-1-3-2R(IR75207H) IR58025A/IR59606-119-3R (IR69688H) IR68897A/IR59673-93-2-3-3R (IR75582H) PSBRc 28 (check) LSD (5%) LSD (1%)
Table 3. IRRI-bred CMS lines designated in 1999. CMS line IR76765A IR76766A IR76767A IR76768A IR76769A IR76770A IR77285A IR77286A IR77287A IR77288A IR77289A IR77290A IR77291A IR77292A IR77293A IR77294A Source BCN BCN BCN BCN BCN BCN BCN BCN BCN BCN BCN BCN BCN BCN BCN BCN 0022 0030 0042 0054 0064 0072 0862 0868 0874 0880 0884 0890 0894 0902 0944 0924 Female parent G46A/6*IR71564B G46A/6*IR69628B D297A/6*IR58025B D297A/6*IR70959B D297A/6*IR69628B IR68897A/6*IR68903-4-1-1-3 G46A/8*IR69623B GA46A/7*IR69618B G46A/*PMS1B D297A/7*IR69618B D297A/7*IR72079B IR68897A/7*IR68907-9-3-1-1 IR68897A/7*IR68908-3-1-2-2 IR68897A/7*IR68908-9-3-2-1 IR68281A/7*Basmati 385-1 IR68897A/7*IR68952-8-1-9-4 Male parent IR71564B IR69628B IR58025B IR70959B IR69628B IR68903-4-1-1-3 IR69623B IR69618B PMS1B IR69618B IR72079B IR68907-9-3-1-1 IR68908-3-1-2-2 IR68908-9-3-2-1 Basmati 385-1 IR68952-8-1-9-4
DEVELOPMENT OF THERMOSENSITIVE GENIC MALE STERILE (TGMS) LINES
We evaluated 2,149 F3-F6 progenies in the pedigree nursery. Among F5-F6 progenies, 215 good looking male sterile plants were selected and transferred to the phytotron (temperature 27 C/20 C) to induce fertility. Fifty-three plants reverted to partial fertility. Seeds produced from them were used for evaluation of male sterility in the field at high temperature (above 30 C). Based on their good phenotypic acceptability score (15), good fertility expression (score 15), and complete male sterility in the field (using the natural temperature variation), six TGMS lines were identified for sharing with national programs. Those are IR71007-9-51-2-122, IR72085-21-19-15-19, IR72093-10-13-26-4, IR72094-12-1-3-30, IR72096-14-10-4-12, and IR72097-10-1-11-25.
IR LINES NAMED AS VARIETIES
Nine IRRI lines from the irrigated breeding program were named as varieties in six countries (Table 4). This brought the number of IRRI breeding lines named as varieties by national programs to 317.
HYBRID RICE NETWORK
Several consultancies were completed by the IRRI/ Asian Development Bank (ADB) project on Development and Use of Hybrid Rice in Asia. A consultancy organized by IRRI in Bangladesh prepared a national plan for hybrid rice research and development for the next 5 years. IRRI also had
Table 4. IRRI breeding lines from the irrigated breeding program named as varieties in 1999. Country where named Myanmar Vietnam Bhutan Bhutan Vietnam Philippines Afghanistan China Philippines
consultancies in India, Indonesia, and Sri Lanka to review ongoing research and development activities for hybrid rice. Training on hybrid rice breeding was presented at IRRI during Mar-Apr 1999 and special on-thejob training was organized for research personnel from Bangladesh, Indonesia, and Vietnam. A coordinated international hybrid rice trial involved 10 elite hybrids, two international checks, and two national-local checks in six member countries. The trial helped identify specific hybrids adaptable to the countries. These data are being analyzed to develop a breeder-friendly selection index using the ranks of mean and cv under low- and high-yielding environments. A benchmark survey in member countries assessed their current capacity to develop and use hybrid rice technology. Good progress was made in developing new parental lines useful for developing rice hybrids for the member countries. Consultancies in collaboration with the Food and Agriculture Organization (FAO) assessed the seed industry support available for large-scale hybrid rice seed production and distribution and gave recommendations for establishment of large-scale seed production on a sustainable basis. A consultancy organized in collaboration with the Asia Pacific Seed Association identified policy and institutional constraints to development and use of hybrid rice in the member countries and recommended some policy interventions to develop and promote the technology. China was accepted as a full member of the network effective January 2000. Genetic divergence of maintainer and restorer lines used to breed tropical rice bybrids S.S. Virmani, Z.K. Li, Weijun Xu, J.E. Hernandez,1 L.S. Sebastian,1 and E.D. Redoa1 The success of three-line hybrid rice in China since the late 1970s is known to be partially due to high level of heterosis in crosses between the Chinese local maintainer lines and the restorers from tropical areas, which are considered to represent two heterotic groups within indica rice. In order to identify heterotic groups further, IRRI, the University of the Philippines Los Baos (UPLB), and PhilRice
Breeding line
Name given
IR35366-90-3-2-1-2 (IR72) IR59606-119-3 IR61328-1-136-2 IR61331-2-148 IR62032-189-3-2-2 IR62141-114-3-2-2-2 IR62871-166-2-2 IR64446-7-10-5 IR64683-87-2-2-3-3
Yezin 1 OMCS94 Bajo Kaap 2 Bajo Kaap 1 IR62032 PSBRc80 Baghlan 98 Dianchao 3 PSBRc82
Table 5. Microsatellite markers, allele distribution, and gene diversity in parental lines of tropical hybrid rice. Relative gene diversity Chromosomea Markers Alleles detected (no.) 8 4 3 4 2 4 4 5 3 7 3 4 3 5 6 2 4 5 4 2 5 3 5 9 3 5 4 4 6 6 2 3 7 5 2 3 3 4.24 1.71 Within B lines 0.25 0.25 0.07 0.11 0.18 0.10 0.11 0.19 0.22 0.22 0.23 0.20 0.36 0.25 0.18 0.22 0.16 0.25 0.21 0.19 0.21 0.24 0.20 0.25 0.24 0.24 0.18 0.25 0.24 0.24 0.27 0.26 0.10 0.24 0.12 0.12 0.14 0.20 Within lines Between B and R lines 0.53 0.50 0.58 0.57 0.49 0.52 0.52 0.50 0.53 0.52 0.50 0.53 0.51 0.53 0.55 0.50 0.52 0.53 0.52 0.49 0.50 0.50 0.52 0.51 0.52 0.50 0.55 0.50 0.54 0.51 0.49 0.50 0.55 0.55 0.54 0.53 0.51 0.52 Total gene diversity 0.77 0.69 0.35 0.45 0.07 0.13 0.23 0.20 0.56 0.71 0.40 0.62 0.19 0.69 0.63 0.34 0.23 0.56 0.63 0.13 0.22 0.54 0.60 0.69 0.45 0.37 0.53 0.62 0.78 0.69 0.09 0.61 0.49 0.70 0.32 0.31 0.18 0.45 0.22
9 10
11
12
Unmapped Av SE
a
RM1 RM5 RM212 RM6 RM221 RM250 RM16 RM22 RM168 RM226 RM241 RM261 RM255 RM164 RM13 RM122 RM3 RM30 RM217 RM10 RM11 RM18 RM25 RM80 RM42 RM215 RM201 RM216 RM228 RM258 RM167 RM254 RM209 RM19 RM17 RM12 RM15
0.22 0.25 0.35 0.32 0.33 0.38 0.38 0.31 0.25 0.26 0.27 0.27 0.12 0.22 0.27 0.28 0.32 0.22 0.27 0.32 0.28 0.26 0.29 0.24 0.24 0.26 0.26 0.25 0.22 0.25 0.24 0.24 0.34 0.21 0.34 0.35 0.34 0.28
The assignment of markers on each of the rice chromosomes was based on Chen et al (1997).
collaborated to investigate molecular divergence of some maintainer and restorer lines of the WA-CMS system used in tropical hybrid rice breeding programs. Thirty-seven simple sequence repeat (SSR) markers distributed in 12 rice chromosomes were selected for the molecular diversity assay of 37 maintainer lines (B) from IRRI and 43 restorer (R) lines from PhilRice. The SSR assays followed the standard procedure and Nie and Lis statistic was
used as a measure of genetic similarity. Gene diversity was partitioned into B population, within R population, and between B and R populations. The mean number of alleles per SSR locus was 4.24 1.71, ranging from 2 to 9 (Table 5). There were 14 loci where more than five alleles were detected. Among 157 detected alleles, 115 (73.2%) alleles were common to both B and R lines, 11 alleles at 9 loci were unique to B lines, and 3 alleles at 14 loci were unique in R lines. The frequency of
group-specific alleles ranged from 0.02 to 0.16. Relative gene diversity averaged 0.20 within B lines, 0.28 within R lines, and 0.52 between two groups of lines, suggesting a significant divergence between the parental groups. Significant allelic frequency differences between B and R groups were present for the two most common alleles at a majority of the SSR loci. The mean genetic distance was 0.39 between pairwise B lines, and 0.49 between R lines, indicating the B lines shared greater similarity within group than that of R lines. This is understandable because, among the tropical indica elite lines, frequency of B lines (less than 5%) is much lower than that of R lines (2030%) for the WA-CMS system. Cluster analyses were performed based on Neis genetic distances between lines within the B and R groups (Figs. 1 and 2). The B lines formed four clusters. Cluster A consisted of 33 cultivars (87%). Cluster B contained three closely related lines, B38 (IR72081B), B40 (IR64608B), and B16 (IR69616B). Cluster C contained only a single line B39 (IR72082B), which was known to have widecompatible genes from BPI 76 and Palawan. The line B2 (IR62829B) formed another single line cluster, D. The 43 R lines formed three clusters with cluster A containing 40 most recently developed advanced lines. Cluster B consisted of only two lines, R54 (C4160-B-31) and R7 (MRC 2219482156), which are respectively derived from crosses of Ostralia/Pinilian and BE-3/BPI 121. The line R62 (PR23765-23R) formed a single line cluster. Our results indicate a clear differentiation between the B and R lines in tropical hybrid rice breeding programs. Future research should be devoted to increasing diversity among B lines and between B and R lines to broaden the genetic base of rice hybrids. Association between SSR diversity, pedigree record, quantitative trait variation, and hybrid performance S.S. Virmani, Z.K. Li, Weijun Xu, J.E. Hernandez,1 L.S. Sebastian,1 and E.D. Redoa1 Knowledge of genetic diversity among prospective parental lines is important for the success of a hybrid rice breeding program. Genetic diversity is usually measured using pedigree information, pheno-
0.32
0.48
0.64
0.80
0.96
B1 B6 B33 B35 B24 B32 B17 B18 B4 B7 B10 B11 B20 B21 B3 B9 B14 B8 B15 B28 B27 B5 B34 B36 B13 B22 B31 B25 B37 B30 B29 B38 B40 B18 B39 B2
B C D
1. Dendrogram of 37 maintainer lines of hybrid rice based on Nei and Li distance using UPGMA clustering. IRRI, PhilRice, and UPLB, 1999.
0.32
0.48
0.64
0.80
0.96
R5 R83 R72 R66 R10 R63 R38 R65 R18 R81 R60 R33 R85 R71 R64 R76 R17 R13 R37 R75 R13 R15 R59 R56 R32 R57 R58 R52 R53 R61 R23 R82 R28 R51 R17 R36 R55 R20 R29 R15 R7 R51 R62
B C
2. Dendrogram of 43 restorer lines of hybrid rice based on Nei and Li distance using UPGMA clustering. IRRI, PhilRice, and UPLB, 1999.
typic data of plant characteristics, and molecular markers. We studied the relationship between those measures and hybrid performance in rice. Materials used included 37 maintainer and 43 restorer lines of the WA-CMS system, which represented germplasm of different origins (IRRI, Philippines, and China) used in the hybrid rice breeding program of PhilRice, IRRI, and UPLB. The pedigrees of those parental lines could be traced back to ultimate ancestors that had no known pedigree information. Ten maintainer and 18 restorer lines were randomly selected to produce 34 F1 hybrids in 1997. Field evaluation of the hybrids and their parents were accomplished at three sites in the Philippines during 1997 and 1988 DS. Coefficients of co-ancestry were calculated for the parental lines using the pedigree information. Euclidean distances, on the other hand, were derived with data from 12 quantitative charactersdays to flowering, flagleaf length and width (cm), plant height (cm), number of productive tillers, panicle length (cm), 100-grain weight (g), grain yield per plant (g), grain length and width (mm), and panicle weight (g). SSR assays with 37 primer pairs were made at the PhilRice Genetic Laboratory and Nei-Li coefficients were derived. Correlation analysis between various diversity measures and hybrid performance, or mid-parent heterosis, was performed on plant height, grain yield, total plant weight, 100-grain weight, grain length, and grain width. In analysis of variance across three sites, significant to highly significant differences were observed for all quantitative traits investigated, except flagleaf width. The mean number of alleles per SSR locus was 4.24 1.71, ranging from 2 to 9. There were 14 loci where more than five alleles were resolved. There was no, or poor, correlation among different measurements of diversity for both B and R lines. These results suggest that the methods for measuring genetic diversity are not consistently associated with each other. There was no correlation between the diversity measure based on SSR markers and F1 performance or the midparent heterosis, indicating that molecular diversity at a random set of SSR markers is not useful in predicting hybrid performance or mid-
parent heterosis. Prediction power may be improved if selected markers are linked to QTLs affecting heterosis. Genetic diversity measured by the pedigreebased coefficient of co-ancestry was significantly correlated with the F1 mean performance but not with midparent heterosis for 100-grain weight, grain length, and plant height, which are known to have high heritability. It appears that pedigree information can be useful to trace genes of additive action. Euclidean distance (D2) estimates based on the quantitative trait differences between the parents were significantly associated with F1 performance for plant height and midparent heterosis for the total plant weight, plant yield, and 100-grain weight. Toward a synthetic apomixis for rice X.Z. Bi, A. Kathiresan, J. Bennett, and G.S. Khush Cost of seed production limits adoption of hybrid rice technology. Whether the three-line or two-line system is used, hybridization must occur on the field scale, where yields are low (typically 1-3 t ha1). Scientists at IRRI and elsewhere have been searching for several years for apomictic relatives of rice with the idea that this form of asexual reproduction could simplify the hybrid rice industry. If an apomictic mechanism of reproduction could be activated in a manually produced hybrid, seed could be multiplied asexually in subsequent generations without loss of hybridity. Seed production would be cheaper, more convenient and versatile, and farmers would be able to reproduce their own apomictic hybrid seed. Apomixis occurs in a large number of plant species, including wild relatives of cereals such as maize, pearl millet, and wheat, but a careful search has failed to reveal any apomictic rice. We began exploring the possibility of synthesizing a form of apomixis resembling the adventitious embryony seen in citrus. Although several apomictic mechanisms exist, the generation of adventitious embryony would in principle require only two steps: 1) induction of an asexual embryo in the nucellar tissue around the embryo sac and 2) removal of the sexual embryo. Endosperm production would be unaltered.
10
INDUCING THE ASEXUAL EMBRYO IN THE RICE NUCELLUS
The nucellus is the large mass of cells surrounding the embryo sac within the ovary. About 10 d prior to fertilization, a single nucellar cell develops into the megaspore mother cell and undergoes meiosis and mitosis to form the embryo sac. This sac contains eight haploid nuclei, of which one is found in the egg cell and two are found in the central cell. After pollination, the egg cell is fertilized by one sperm cell and the central cell is fertilized by a second sperm cell. The embryo develops from the fertilized egg and the endosperm develops from the fertilized central cell. At about the time of fertilization, the nucellus begins to undergo programmed cell death (PCD) to supply the young zygotic embryo and the young endosperm with nutrients. This time would also be suitable for the start of adventitious embryo formation. We sought to isolate a rice promoter that could trigger embryogenesis in the nucellus. The nucellin gene of barley is expressed exclusively in the nucellus, starting about 1 d prior to fertilization. It appears to encode an aspartate protease involved in PCD. We used published data on that gene to isolate the rice homologue by a technique based on the polymerase chain reaction (PCR). Basing our PCR primer design on those segments of the barley nucellin gene that show its relationship to aspartate proteases, we used reverse transcriptase-PCR (RT-PCR) to amplify a partial complementary DNA (cDNA) from RNA isolated from the ovary 1 d prior to fertilization. The sequence of the cDNA fragment confirmed its relationship to barley nucellin. We used the rice cDNA fragment to determine gene copy number and found only a single copy of this gene. Mapping of the gene using the IR64/ Azucena mapping population showed that it is located on chromosome 11. We used the cDNA probe to isolate a series of clones from the IRRI library of bacterial artificial chromosomes (BAC) derived from IR64 DNA. The clones all proved to contain the same gene, which has now been sequenced from >3 kb upstream to >0.5 kb downstream of the nucellin coding sequence. The upstream sequence will presumably contain the desired promoter.
3. In situ hybridization of rice ovary tissue (5 d after fertilization) with nucellin cDNA. Paraffin-embedded ovary sections (8 m) were probed with DIG-labeled sense (A) and antisense (B) RNA transcribed from a 363-bp nucellin cDNA fragment. IRRI, 1999.
RT-PCR of RNA isolated from a range of rice tissues and from the panicle at various stages of development established that it is ovary-specific and expressed from about 1 d before fertilization to at least 5 d after fertilization. Preliminary in situ hybridization with a fluorescent anti-sense probe derived from the 3-untranslated region (3UTR) of the nucellin gene showed that the gene is indeed expressed in the nucellus (Fig. 3). The next step is to splice the nucellin promoter with the coding regions of rice homologues of genes implicated in inducing embryogenesis in Arabidopsis (e.g., lec1 and mi-2). The resulting constructs will then be introduced into rice to determine whether they induce adventitious embryony.
ARRESTING THE SEXUAL EMBRYO DEVELOPMENT
The simplest way of arresting the development of the sexual embryo is to generate an inhibitory protein under the control of an egg- or zygote-specific promoter. Because such a promoter is to date unknown for plants, we took a more involved (twogene) approach to achieve the same goal. In this approach, we required two promotersone expressed specifically during meiosis and another expressed during early embryo development. The meiosis-specific promoter is required to activate the embryo-specific promoter only in embryos derived from meiosis (sexual embryo). As the adventitious
11
embryo is derived from nonmeiotic cells, its development should not be arrested by this mechanism. Following a published report that the Arabidopsis recA homologue DMC1 is meiosisspecific, we decided to clone and characterize rice DMC1. Based on Genbank sequence data, we designed primers complementary to DMC1 and amplified a rice genomic fragment. Southern blotting with this partial fragment showed that the rice genome contains two copies of DMC1. We isolated and sequenced both copies (DMC1A and DMC1B) from IR64 BAC library. At the nucleotide level, the two DMC1 copies show 97.4% identity for coding regions and 70% identity for 3'-UTR. RT-PCR showed that although expression of both copies coincides with meiosis in panicles, they are also expressed in mitotically active cells such as calli. Figure 4 shows the data for DMC1B.
It is clear that DMC1 in plants is not strictly meiosis-specific, unlike homologues in yeast and animals. We are now exploring the potential utility of the rice homologue of SPO11, another meiosisspecific protein of yeast (type II topoisomerase subunit). For a promoter that is embryo-specific among post-meiotic cells, we turned to REE5, a 254-bp rice cDNA clone isolated from zygotic embryos. REE5specific primers and DNA gel blots showed that the rice genome contains at least two REE5-like genes. To generate gene-specific probes, we cloned the 3UTR of REE5 that is expressed in spikelets 2 d after fertilization through rapid amplification of 3'cDNA ends (3'-RACE). Using the 3'-UTR of the RACE clone, we screened IR36 lambda genomic library and subcloned a 5.0-kb HindIII fragment. Sequence analyses suggested that REE5 encodes a novel phosphoprotein. We are currently determining its expression pattern in rice ovules.
1 Kb plus 0-3 4-6 7-9 10-12 13-15 16-18 19-21 22-24 calli leaves roots dWater IR72 genome BAC clone (DMCIE) 100 bp ladder
1 Kb 600 bp
4. Expression profile for rice DMC1B. Embryogenic calli, leaves, and roots from 3-wk-old seedlings, and various lengths of panicles from primary tillers of IR64 were harvested. RT-PCR was done using 1 mg DNase-treated total RNA each as template and DMC1B sequence-specific primers. The primers flank two introns totaling a size of 511 bp. The expected size of RT-PCR product is 553 bp and that of the genomic product is 1064 bp. IRRI, 1999. 12 IRRI program report for 1999
Characterization of root system in a new plant type S. Kubota, H. Samejima,2 E. Laureles, and O. Ito A vigorous root system has been given as one of selection criteria in breeding of the NPT, but its root system has not been studied in detail. There is no evidence to show morphological and physiological superiority of NPTs root system over existing IRRI varieties. A 1997 WS field experiment at IRRI examined root systems. Seedlings (14 d old) of IR72 and NPT (IR65598-112-2) were transplanted on 27 Jun. Fertilizers were applied 1 d before transplanting as a basal application of 150 kg N ha1, 60 kg P ha1 and 60 kg K ha1 combined with four types of slow-release coated ureaLP40, LP70, LP100, and LP140. Plant density was 25 hills m2 for IR72 and 50 hills m2 for NPT. Root samples of 1 hill of IR72 and 2 hills of NPT were taken by a monolith (20 cm 20 cm 50 cm) at 3-wk intervals from transplanting to panicle initiation (PI), and at 2-wk intervals from PI onward. The soil and roots inside the monolith were divided into six horizons02, 25, 510, 1020, 2030, and 3040 cm from the soil surface. Collected roots were washed and measured by image analysis. Stem sap was sampled at 35, 46, 61, 76, 90, and 104 d after transplanting (DAT). Four hills of each plot were cut at 15 cm height at 1800 h on the sunny day and absorbent cotton was put on each stump until 0600 h the next morning. The cotton was covered with a plastic bag and a black plastic pot to avoid vaporization and night dew. Amount of stem sap was estimated by weighing the cottons. Eight hills were harvested for the measurement of yield and yield components and grain yield was adjusted to 14% moisture content. Total dry matter at harvest was larger in NPT than in IR72. IR72 had significantly larger panicle number, higher grain filling, and larger 1,000-grain weight than NPT. On the other hand, number of grains was larger for NPT than for IR72. There consequently was no significant difference between IR72 and NPT in rice yield. Total root length was longer in NPT than in IR72 throughout the growth stage (Fig. 5) and NPT recorded maximum root length (ML) two times longer than IR72 at 55 DAT. After 55 DAT, NPT showed a sharp decline in root length. The length
decreased by 50% of ML at flowering (FL), followed by further reduction. In contrast, IR72 flowered at 6 d after ML, resulting in slower reduction in length than NPT at FL. Changes in total root dry weight (DW) of both varieties were almost identical with the changes in root length (Fig. 6). The root DW was significantly larger in NPT than in IR72. Difference in the DW was particularly obvious at 0 2 cm and 25 cm of soil profiles, suggesting that NPT developed a vigorous root system in relatively shallow soil profiles.
Root length (km m-2) 40
PI NPT IR72
30
FL
20
MT
10
MT and PI FL
30
60 90 Days after transplanting
120
5. Changes in root length of NPT and IR72. IRRI, 1997 WS.
Root dry weight (g m-2) 60 120 Whole root 0-2 cm 50 100 40 80 NPT 30 60 20 40 10 20 IR72 0 0 40 40 2-5 cm 5-10 cm 30 30 20 20 10 10 0 0 30 30 10-20 cm Below 20 cm 20 20 10 10 0 0 0 30 60 90 120 0 30 60 Days after transplanting
90 120
6. Changes in root dry weight of whole root and each soil profile in NPT and IR72. IRRI, 1997 WS.
13
Amount of stem sap (g m-2) 600

NPT
Japonica germplasm utilization for value added M.H. Lee, J.R.T. Chavez, and G.S. Khush Since 1992, 475 IR designated crosses have been selected as advanced generations in research to develop japonica germplasm for use in the tropics. More than 3,000 pedigree lines were maintained each season to accommodate breeding selection pressures. Grain yield improved gradually from a fluctuating low potential yield of 3 t ha1 to an average of more than 5 t ha1. Twenty-five elite lines were tested at IRRI during 1999. Five of those lines had yield equal to, or greater than, yield of the check variety (Table 6). These new elite lines are adapted to tropical cultural management. Entries were sent to the International Network for Genetic Evaluation of Rice (INGER) and PhilRice to test their yield potential in different tropical regions. Inheritance of RTSV resistance in TW5, the near-isogenic line from Utri Merah O. Azzam, T. Imbe, R. Ikeda, P.D. Nath,3 M. Muhsin,4 and E. Coloquio Adaptation in green leafhopper populations and the presence of virus variants for the RNA (rice tungro spherical virus [RTSV]) and DNA (rice tungro bacilliform virus [RTBV]) viruses of tungro, suggest that tungro resistance, when deployed, will be easily overcome. Hence, an integrated approach for
500 400
FL
300
IR72
200 100 0
FL
20
40 60 80 Days after transplanting
100
120
7. Changes in the amount of stem sap in NPT and IR72. IRRI, 1997 WS.
IR72 recorded 300 to 380 g m2 of stem sap until 62 DAT, which was just before FL. Stem sap linearly decreased after FL (Fig. 7). NPT showed the maximum amount of 570 g m2 at 46 DAT, which was 40 d before FL. Decline in stem sap in NPT started during the vegetative stage and decreased by an estimated 200 g m2 at FL. This study indicates that the root system and stem sap as an index of root function in NPT is superior to IR72 at the early vegetative stage, but senescence of the root system begins earlier in NPT, resulting in poor root activities during the reproductive stage.
Table 6. Plant characteristics and grain yield of promising elite lines developed from japonica germplasm at IRRI during 1992-99. IRRI, 1999. 1999 DS Line HD (d)
a
1999 WS
c
PL (cm) 20.4 20.1 20.1 22.4 20.0 20.1
NP Yield (no. hill1) t ha1) 10.9 9.4 10.1 8.9 10.3 9.8 7.2 6.0 6.6 6.4 6.0 5.6
HD (d) 76 68 68 74 72 68
PL (cm) 18.3 18.5 18.7 19.7 19.4 19.4
NP Yield (no. hill1) (t ha1) 10.6 11.0 10.3 10.0 9.8 11.1 4.6 4.7 4.8 4.4 4.7 4.1
IR68333-R-R-B-22 IR68349-131-2-2-3 IR68352-14-1-1-1 IR68399-78-2-3-3-1 IR68373-R-R-B-22-2-2 Jinmibyeo check)

a
74 64 62 62 68 70
Hd = heading date. bPL = panicle length. cNP = no. of panicles.
14
combining several sources of resistance may be effective in tungro management. Understanding the mode of inheritance of currently available tungro resistance genes will allow integration of a diversified type of resistance. Utri Merah (IRGC 16680 and referred to as UM80) is highly resistant to RTSV and tolerant of RTBV in artificial inoculation and field evaluation. Analysis of a large number of F3 progenies derived from UM80/TN1 revealed that UM80 may have two independent recessive genes to RTSV (tsv-1 and tsv-2). In addition, the nearisogenic lines (TW5 and TW6), which were derived by backcrossing UM80 with IR22, a recurrent tungro-susceptible parent, also showed resistance to tungro infection by artificial inoculation. Based on their reaction with different RTSV variants, TW5 and TW6 were thought to have the recessive gene tsv-1. We studied the inheritance of that gene using F3 progenies of TW5/TN1 combination and two RTSV variants: RTBV + RTSV-V and RTSV-VI. The inoculation conditions were standardized and the RTSV inocula were monitored by enzyme-linked immunosorbent assay (ELISA) and RT-PCR to ensure the quality and type of inoculum. Thirty plants per line were insect-inoculated and their reaction was evaluated at 3 wk post inoculation by ELISA. The F3 analysis of 203 lines inoculated with RTBV + RTSV-V inoculum showed complete susceptibility to RTBV infection and a segregation of RTSV resistance into 13 resistant lines and 190 segregating and susceptible lines, resulting in a good fit to a 1:15 ratio rather than to a 1:3 as expected. These results triggered the retesting of the F3 lines with RTSV-VI, the avirulent RTSV inoculum source that is maintained on TN1 plants. Among 203 progenies, 118 were insect-inoculated and evaluated by ELISA for their resistance to the virus. Results showed that the F3 progenies segregated into 34 resistant lines and 84 segregating and susceptible lines, resulting in a good fit to a 1:3 ratio. These results suggest that the inheritance of resistance to RTSV in TW5 depends on the nature of RTSV inoculum used. Because of the differential reaction of TW5/TN1 F3 progenies to two RTSV inocula, the evolutionary relationships between RTSV variants were evaluated using sequence analysis of the amplified RTPCR coat protein region of the RTSV genome (nt
CP1-CP2- 1.1kb 96 80 48 74
Pe21II Pc88II Pc41II 99 Pc20I 100 Pc34I PgVt6-lll, 1994 P71III 80 89 98
52
Pc46III Pc17III Pg16VI, 1999 100 Pg09VI, purified virus stock-1996 99 Pc03V Pg08V, 1999 Pc12V 72 34 Pg05V, purified virus stock-1997 PgA, 1988
8. Parsimony tree for RTSV coat protein sequence alignments (1.1 kb) of IRRI greenhouse and field isolates in the Philippines. Pg = IRRI-greenhouse isolates; Pc = isolates from north Cotabato, Pe = isolates from Nueva Ecija. PgA = RTSVavilurent strain sequenced in 1988; PgVt6 = RTSV-virulent source on TKM6 sequenced in 1996; Pg16VI and 09VI = RTSVavirulent source on TN1; Pg08V and Pg05V = tungro source on TN1. IRRI, 1999.
2453-3607). The currently used RTSV-V and RTSV-VI inocula and their corresponding 1997 and 1996 purified virus stocks were included. In addition, the sequence of the original RTSV avirulent source (PgA) and that of the current RTSV-Vt6 (PgVt6-III) were also used. For general comparison, field RTSV isolates were included when phylogenetic grouping was attempted with the parsimony method. The tree generated for the coat protein nucleotide sequence alignment revealed a separate clustering for each of the four RTSV variants found at IRRI greenhouse (Fig. 8). Although Pg16VI and Pg08V variants were originally derived from PgA, the current data show that they are distinct variants. These results confirm that RTSV inoculum maintained with RTBV source on TN1 differs from the one maintained alone on the same host. Breeding rice for resistance to tungro G.S. Khush, E.R. Angeles, A.M. Pamplona, and P.S. Virk Resistance to tungro has always been emphasized in IRRIs breeding program. Most of the IR varieties developed to date have resistance to green leaf-
15
hopper, the vector of the tungro virus. However, the vector resistance erodes after several years. We have tried for the last 10 years to develop improved germplasm with resistance to the virus. Several donors for resistance were used, such as Habiganj DW8 from Bangladesh, Utri Merah and Utri Rajapan from Indonesia, and Oryza rufipogon. Donors were crossed with improved-plant-type lines susceptible to green leafhopper, and several backcrosses made. We were able, in the absence of vector resistance, to select for tungro resistance in the segregating generations. Several lines with virus resistance were selected from the crosses of each of the donors (Table 7). These lines have high yield potential, excellent grain quality, and short growth duration. They were tested for tungro resistance at Midsayap, Mindanao, Philippines, where tungro incidence is always high and showed tungro resistance. Evaluation with ELISA shows that these lines are resistant to RTSV.
Table 7. Elite tungro-resistant lines developed from crosses with donors of tungro resistance and improved plant types. IRRI, 1999. Donor for resistance Growth duration (d) 116 116 119 120 118 120 123 127 132 118 122
Selection
IR71606-1-1-4-2-3-1-2 IR71606-1-2-1-3-2-3-1 IR73885-1-4-3-2-1-4 IR73885-1-4-3-2-1-6 IR73885-1-4-3-2-1-10 IR73888-1-2-7 IR69726-41-2-3 IR69726-41-2-3 IR70458-87-2-2-3-1 IR69727-37-2-1-3-2 IR72 (check)
Habiganj DW8 Habiganj DW8 O. rufipogon O. rufipogon O. rufipogon O. rufipogon Utri Merah Utri Merah Utri Rajapan Utri Rajapan None
menon in long-term experiments (LTE) in Asia, we analyzed data from 30 LTEs at 24 different sites in China, India, Indonesia, Bangladesh, Vietnam, Philippines, and Malaysia. The data represent a variety of soil types and constitute all the experiments for which we were able to obtain data sets (data for one particular season) for at least 9 years. Twenty-two LTEs were from rice monoculture systems. Eight rice-upland crop LTEs included five rice-wheat sites and three experiments with triple-cropping rice-rice-wheat (Sichuan), rice-rice-barley (Zhejiang), and rice-wheat-jute (Barrackpore). The only data included in the analysis are those from the treatment with the NPK rate that consistently produced the highest yield. Rice varieties used in all LTEs were modern varieties with a harvest index of 0.450.50 and a growth duration of 110130 d. Transplanting was used for rice crop establishment. Yields were measured from a 4- to 5-m2 harvest area in each replicate plot per treatment. Soil and plant data were collected over time, but had widely differing frequencies and sampling and measurement procedures, and are not included in the formal statistical analysis. However, they are used in the interpretation of the results where appropriate. To test the hypothesis that yield trends over a period of at least 9 years are significantly different from zero, data were analyzed by ordinary least squares linear regression of yields (in logarithmic form) against a time trend variable: ln(Y) = a + bt where Y is the grain yield (kg ha1), a is a constant, t is the year, and b is the slope or magnitude of the yield trend (percentage change in yield per year). A statistically significant positive or negative yield trend was recorded only if the null hypothesis of a zero slope for the time trend variable could be rejected at a 5% level of significance (two-tail test). The WS and DS data were analyzed separately. Only two of the 21 data sets from the 15 sites outside of IRRI showed declining DS yield trends that were statistically different from zero at a 5% level of significance or less. Eight of these 21 data sets showed positive yield trends, with one statistically different from zero (Omon-LTFE, +4.63% y1). In contrast, DS experiments at IRRI had statis-
Sustaining soil quality in intensive rice systems

Yield declines in long-term rice experiments in Asia D. Dawe, A. Dobermann, P. Moya, S. Abdulrachman,5 Bijay-Singh,14 P. Lal,7 S.Y. Li,17 B. Lin,19 G. Panaullah,20 O. Sariam,21 Y. Singh,7 A. Swarup,18 P.S. Tan,22 and Q.-X. Zhen23 Yield declines are an important phenomenon at IRRI. To investigate how widespread this pheno-
16
tically significant yield declines ranging from 1.45% to 1.61% y1. High DS grain yields of 78 t ha1, with no substantial decline over time, were measured in the four LTEs at the PhilRice and Bicol, Philippines, sites. By the late 1980s, yields in those experiments exceeded those in comparable experiments at IRRI by 1 t ha1, although yields at the IRRI experiments had started out at higher levels. The DS Omon-LTFE experiment that showed a statistically significant positive yield trend started on a P-deficient, acidsulfate-influenced soil and regular P addition has been the main factor causing the yield increase. In the WS, two of the 14 data sets from 11 sites outside of IRRI showed a statistically significant declining yield trendthe PhilRice LTFE (1.35% y1) and the Omon-LTFE (2.22% y1). Four of these 14 data sets showed positive yield trends, although none were statistically significant. In contrast, all four WS data sets at IRRI showed yield declines, and three of them were statistically significant. Yield declines appear to be more common in the WS, although most of these declines are not statistically significant. Of the eight rice-upland crop LTEs, there was a statistically significant yield decline in only one of 10 rice data setsthe Pantnagar-LTFE (2.3% y1). The rice crop at the Ludhiana-LTE suffered a substantial yield decline of 2.7% y1, but it was not statistically significant and 2 of the years with low yields were due to pest outbreaks. In all other LTEs, rice yields in the best fertilizer treatments remained virtually unchanged over periods of 1025 years. Four of the 10 data sets showed positive yield trends, and the average yield trend was relatively small at just 0.45% y1. Of the seven wheat data sets, there were no statistically significant yield declines, but one statistically significant yield increase occurred in the Pantnagar LTE (+2.4% y1). Three of the seven data sets showed a positive yield trend, and the average yield trend was slightly positive at +0.04% y1. On the sodic soil of the Karnal-LTE, high rice yields of about 7 t ha1 and wheat yields of about 5 t ha1 were sustained for 12 years, and continuous rice-wheat cropping decreased the soil pH from 9.2 to 8.5. The Nangong-LTE is noted because of high rice yields (68 t ha1) sustained for 14 years. The Sichuan-LTE with three crops grown per year (ricerice-wheat), where annual grain production was
sustained at a level of 15 t ha1 for 10 years, had no distinct trend of a decline or increase. Our results question whether the yield declines observed in various LTEs at IRRI are currently representative of other irrigated rice areas in Asia. Nevertheless, yield declines do occur in some LTEs, and their causes must be understood. Based on a review of the evidence from these LTEs, several possible important causes stand out. q One possibility is a decline in indigenous N supply associated with increased phenol content of soil organic matter (SOM). That could abiotically immobilize N or reduce the rate of N mineralization per unit of organically bound N. With increased frequency of cropping (longer anaerobic vs aerobic periods), more partly degraded lignin residues accumulate, the phenolic content of SOM increases, and the degree of humification of SOM decreases. The IRRI-LTCCE, with short and wet fallow periods, might represent the most extreme case of such SOM changes. For example, a long, dry fallow period occurs at PhilRice before the DS, whereas soil drying in this period is much less intense at IRRI. This difference may explain why there was no significant decline in DS rice yields at PhilRice but a significant decline in all DS experiments conducted at IRRI. q Phosphorus deficiency in the form of low initial available soil P, or a negative P balance, appears to have contributed to yield declines in several LTEs, including Luisiana, Pantnagar, Shipai, Jinxian, and Omon. q In several LTEs, fertilizer K rates were not sufficient to sustain a neutral or positive K input-output balance so that soil K depletion occurred, often much below a commonly used critical level (1 N NH4OAc-extractable K) of 0.2 cmol c kg1. Such LTEs include Gazipur, PhilRice, Pantnagar, Jinxian, and Shipai. Available information suggests that the K rates used in many LTEs were sufficient in the initial years, but not after a period of decline in soil K reserves. q A decline in the quantity of SOM may also be an important factor in some rice-upland crop LTEs (e.g., Pantnagar), but this appears not to be a serious issue in rice monoculture LTEs.
17
Prolonged submergence, insufficient soil drying during fallow periods, and soil P depletion or soil K depletion, or both, were associated with many of the yield declines that did occur. If gradual changes in soil nutrient supply or root nutrient uptake are mainly driven by oxygen and carbon supply, they could be generic in nature for particular cropping systems. Yield declines caused by nutrient mining are less troublesome because they can be corrected more easily. The proper interpretation of future LTEs will require more detailed measurements than have been done in the past
Improving the productivity and sustainability of rice-wheat systems

Improving rice-wheat systems The rice-wheat production system in the Asian subtropics occupies nearly 24 million ha in South Asia and another 10.5 million ha in central China. The favorable production environment has induced farmers to use a highly intensified production system with increased use of chemical fertilizers and pesticides. Those contributed to an impressive increase in per capita cereal production in the region during 1965-85. The 1965-85 gains in per capita cereal production from the rice-wheat system are threatened by stagnant yields of both rice and wheat and a declining trend in total factor productivity. The decline in SOM, imbalances of soil nutrients, the lowering of groundwater tables, and buildup of insect and disease pressures are indicators of the threat to sustainability of the system. The Rice-Wheat Consortium for the IndoGangetic Plains was formed in 1994. Bangladesh, India, Nepal, and Pakistan provide the leadership in partnership with IRRI, the International Maize and Wheat Improvement Center (CIMMYT), the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), and the International Irrigation Management Institute (IIMI). Ecoregional approach to natural resource management R. Roetter, P.K. Aggarwal,6 N. Kalra,6 A.G. Laborte, and C.T. Hoanh New approaches and methods are being developed for research at the systems level to project future
food demand in relation to nutrient management and possible effects on environment. IRRIs Systems Research Network for Ecoregional Land Use Planning in Tropical Asia (SysNet) aims to develop and evaluate methodologies for exploring land use options at the subnational level. For that purpose, regional case studies were set up in Haryana State (India), Kedah-Perlis Region (Malaysia), Ilocos Norte Province (Philippines), and Can Tho Province (Vietnam). During a stakeholder-scientist workshop for Haryana in March 1999, various scenarios were formulated and analyzed. Stakeholders gave the following priority objectives: q Double food production for Haryana. q Maximize agricultural production while setting limits on labor migration. In the future, the supply of labor from outside Haryana may be more restricted. q Minimize nitrogen loss. q Minimize pesticide residues. q Improve water management-intervention measures to reduce groundwater depletion. q Maximize income from agriculture. Initial explorations for the scenario to maximize food production suggest that, with currently available land and water, annual cereal production could be increased to 16 million t (currently 10.7 million t) if appropriate technologies were adopted by farmers. When actual capital and labor were taken into account as constraints, the result did not change, which indicates that water is the most limiting factor. The scenarios assume that improved technologies that lead to increased fertilizer and water use efficiency can be applied. We used five levels of technology that led to different efficiencies of nutrient use. With technology level 5 (N use efficiency 75% for wheat), the fertilizer requirement (200250 kg ha1) and leaching losses of N (10% of applied N) are lower for achieving the maximum attainable (90% of potential) yield, but this requires extra capital for procuring the inputs. At the current technology level (N use efficiency 50% for wheat), however, the existing inputs will do, but fertilizer requirements (400500 kg ha1) as well as leaching losses of N (2530% of applied N) will be high for the same target yield. With current technology (average farmers practice) and current input use, the model suggested food production 10% higher than it is now being
18
achieved. This shows the potential of the model to develop, analyze, and optimize different scenarios along with their impacts on food production, required resources, production technologies, and environmental consequences to help stakeholders identify feasible solutions and select the best option. Soil- and seedborne pathogens of rice in a rice-wheat system L. Willocquet, S. Savary, A. Kumar,7 and U.S. Singh7 Pests (insects, weeds, pathogens) are an important component for sustainability of the rice-wheat system and soil- and seedborne pathogens may play a specific role. Cropping practices may affect the dynamics of those pathogens, both for their survival and their epidemic phases. Interaction among pathogens may also alter the patterns of diseases in a given field.
FIELD SAMPLING AND DISEASE ASSESSMENT
number of panicles, number of panicles infected by ShR, PB, and FS, and number of panicles showing glume discoloration were recorded for each hill.
DISEASE INCIDENCE IN FARMERS FIELDS
Quantification of disease incidence due to soilborne and seedborne pathogens was done in farmers fields. Pathogens included sheath blight (ShB, Rhizoctonia solani), stem rot (SR, Sclerotium oryzae), brown spot (BS, Cochliobolus miyabeanus), sheath rot (ShR, Sarocladium oryzae), panicle blast (PB, Pyricularia grisea), crown sheath rot (CShR, Gaeumannomyces graminis), and false smut (FS, Ustilaginoidea virens). Some (BS, ShR, PB, CShR) can be transmitted by seeds. Others (ShB, SR, BS, PB, CShR, FS) survive as propagules in the soil or in plant debris. Glume discoloration was also condidered because it is associated with the presence of different fungi, among which are P. grisea and C. miyabeanus. Assessments were done in eight fields selected to represent a range of crops preceding rice. We monitored two fields with wheat (B3 and B6), one field with rice (B4), one field with sorghum (B1), one field with maize (PP5), one field with lentil (B9), one field with a fallow (PP2), and one field with mint (PP3) as a crop preceding rice. Disease assessments were done at milk stage by a 20-hill sample. Total number of tillers, number of tillers infected by ShB, SR, BS, and CShR, total
SB, SR, ShB, BS, SHR and glume discoloration were observed in all fields (Fig. 9). The other diseases, when present, occurred at low incidences (08%) and were not considered for multivariate analyses (see below). Highest incidences were observed for SB (up to 68%) and BS (up to 85%), whereas ShR incidence remained below 20% in all the fields. Sheath blight incidence was relatively high in all the fields (around 20%) and reached 75% in field PP5. Brown spot incidence varied from 8 to 90% among fields. Stem rot was also extremely variable too, but within a smaller range of incidence (235%). Sheath rot was observed in all fields, but at low incidence. Glume discoloration incidence was around 25% in all fields except B4 and PP5, which had lower levels. Pearson coefficients of correlation were computed based on the disease incidence data collected
Fraction of infected tillers or panicles
0.8 0.6 0.4 0.2 0 0.4 0.2 0 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2
CShR GD ShR CShR GD
B1
B3
B4
B6
B9
PP2
PP3
PP5
0
SR ShR ShB BS
ShB
BS SR
PB
FS
PB
9. Fraction of infected tillers or panicles in eight farmers fields under rice-wheat in West Uttar Pradesh, India, 1999.
19
FS
Factor 2 1.0 BS 0.5 0.0 -0.5 -1.0 -1.0 ShB GD ShR SR
Factor 3 ShR
Factor 3 ShR
ShB GD SR BS
ShB SR
GD
BS
-0.5
0.0 0.5 Factor 1
1.0
-1.0
-0.5
0.0 0.5 Factor 1
1.0
-1.0
-0.5
0.0 0.5 Factor 2
1.0
10. Output of the principal component analysis on incidence of different diseases in eight farmers fields. IRRI, 1999.
at the hill scale. Sheath blight and SR were negatively correlated (0.388; P<0.001), and BS was positively correlated with glume discoloration (+0.19; P=0.013). Panicle blast correlated positively with SB, negatively with SR, and negatively with glume discoloration, but those correlations should be interpreted cautiously, given the low levels of PB encountered.
RESULTS FROM A PRINCIPAL COMPONENT ANALYSIS
Sheath rot seems negatively associated with BS but that must be further documented. Linkages between rotations and disease levels can be cautiously hypothesized because of small number of fields monitored. For instance, field PP5 shows the ShB level nearly twice that of other fields. PP5 was previously planted with maize, which can host R. solani AG 11A. The effect of a maize-rice system on ShB blight must be further documented.
A principal component analysis was done based on the disease incidence data collected at the hill level (Fig. 10). The two first axes explained 54% of the total variability. The third axis explained 19% of the total variability. The first axis mainly reflects a strong opposition between SB and SR incidence. Axis 2 reflects the association between glume discoloration and BS. Axis 3 reflects the opposition between ShR and BS.
INTERPRETATION OF TRENDS
Improving pest management

Determinants of farmers pest management decisions K.L. Heong Farmers make decisions on pest management practices every season. Those decisions often seem to lack economic rationality. We introduced two concepts from social psychology to gain better understanding of the determinants of farmers stem borer management decisions: 1) the pest belief model and 2) Fishbein and Ajzens theory of reasoned action. Farmers spent an average of US$39 ha1 on insecticides, believing that they would lose an average of 1,004 kg ha1 or US$402 if no action was taken. Farmers estimates of the worst attack loss from stem borers averaged 1,038 kg ha 1 or US$415, which was similar to farmers loss estimates for last season. This implied that farmers decisions were based on preventing the worst case
Our preliminary results point at trends that must be interpreted with caution, given the small number of sampled fields: q SB and SR appear to be negatively correlated. This may be due to differences in environments that favor these two diseases, or to competition between the two types of propagules, or both. q Glume discoloration is correlated with BS. Indeed, C. miyabeanus is reported as a fungus associated with glume discoloration.
20
occurring. However, farmers estimates of the highest number of whiteheads averaged 19 whiteheads m2 and the loss computed from known agronomic characteristics of IR64, the most commonly planted variety, was only 351 kg ha1 valued at US$140 ha 1. We found that more than 50% of the farmers spent more than US$18 ha1 on stem borer control and expected to prevent a loss equivalent to $237 ha1, a cost-benefit ratio of 1:13. The cost-benefit from the worst infestation was only 1:4, implying that farmers perceived benefits were about three times higher than the worst case. Perceived benefits from insecticides were directly related with farmers insecticide use and perceived severity. Perceived susceptibility was also high, with 59% of farmers believing that a loss of 450 kg ha1 would be likely. Farmers believed insecticides could destroy natural enemies but placed only moderate importance on conserving them. Human health was believed to be important but farmers had mixed beliefs that spraying could cause poor health. This study also provided evidence suggesting high peer pressure on farmers spray decisions, which directly influenced perceived benefits from sprays, insecticide expenditures, and spray frequency. Field evaluation of prototype Tungro Screen B kits O. Azzam, P. Nath,4 P. Cabauatan, R. Cabunagan, T. Chancellor, and L. Kenyon8 A diagnostic screen kit was successfully developed for RTBV (Fig. 11) and tested by collaborators in Bangladesh, India, Indonesia, and Philippines. Encouraging results were obtained. Collaborators were asked to complete a questionnaire on the utility of the kit after completing the indexing of rice samples for the presence of RTBV. Based on their responses, the prototype kit was then modified to improve its efficacy. The kit was field-tested in July 1999 at six sites in north Cotabato, Philippines. Rice stems from plants at different growth stages were collected from six known and three unknown varieties and tested locally. Plant material was retained for later testing by DAS ELISA at IRRI. There was a high level of similarity in the scoring of the membranes
11. An example of a tissue printing blot for field testing of rice samples collected from North Cotabato, Philippines, in July 1999. The prints were scored based on the positive (+) and negative () reference checks found at the bottom of the page. IRRI, 1999.
by different assessors, with between 83% and 89% of the positive assessments from the kit also positive for RTBV in DAS ELISA tests. The kit had a good degree of accuracy, which can be improved with further refinement of the method. The method is sufficiently robust to be used outside the laboratory and simple enough to be used by persons after limited training. Alternative prey in irrigated rice: implications for biological control of hoppers L. Sigsgaard, S. Toft,9 and S. Villareal The sheet web spider (Atypena formosana) is the most important linyphiid in the rice ecosystem. Adults and immature spiders prefer to live in the
21
rice stem or at the base of rice hills, where they hunt for nymphs of planthoppers and leafhoppers, collembola, and small dipterans. The sheet web spider and the wolf spider (Pardosa pseudoannulata) are the dominant predators in rice until about 35 DAT in the Philippines. Indirect evidence of A. formosanas association with brown planthopper (BPH) and green leafhopper (GLH) is that populations fluctuate with the densities of BPH, GLH, whitebacked planthopper, and all other hoppers. Previous research at IRRI suggests that spiders can make a significant contribution to a reduction in the number of BPH and GLH, if there is a high density of predators in the field early in the cropping season. We made a laboratory assessment of the survival and development time of A. formosana juveniles on different diets that represented the most common prey in the field. We tested the hypothesis that differences in diets would have different effects on their survival. Survival and development differed significantly among diets. No spider survived until adulthood on the diet of GLH, only a single male individual survived on a diet of BPH, 70% survived until adulthood on a collembola diet, and all spiders survived on the mixed diet. Spiders developed fastest on the mixed diet and the collembola diet. Because no adult females developed on the two hopper diets, it was possible to compare fecundities only on collembola and mixed diets. Of these, the females on mixed diet had the highest fecundity. We concluded that alternative prey not only increase the population of predators but are a necessity because BPH and GLH alone did not provide food enough to increase the number of predators.
Progress of unreported projects

Increasing water use efficiency in rice culture T.P. Tuong, A.M. Mortimer, B.A.M. Bouman, and D.C. Dawe
q
Established a database on results of 31 historic water-saving irrigation (WSI) experiments. The analysis shows that WSI generally saves water and increases water productivity
(grains produced per unit water used) at the field scale, but that land productivity (yield) decreases. The crop growth and water balance simulation model ORYZA was developed from former ORYZA1 and ORYZA-W models to support the empirical data analysis of WSI experiments. Started multilocation field monitoring of water flows within 1) District I of the Upper Pampanga River Integrated Irrigation System; 2) Sta Cruz irrigation systems in Laguna, Philippines; and 3) Cu Chi irrigation system in Vietnam to study water (re-use and water productivity at different spatial scale levels. Started a field water balance study at three spatial scale levels (field, irrigation subsystem, and whole irrigation system) in the Zang He irrigation system, China, together with field experiments into cropping under the ACIAR-SWIM project. Conducted a survey of farmers in China to assess the economic profitability of watersaving irrigation (in collaboration with scientists from WUHEE). Analyzed the impact of large-scale adoption of direct seeding on water productivity in the Muda irrigation scheme and completed a survey of 200 farm households to analyze the economics of alternative crop establishment methods. Analyzed secondary time-series data on production, area, and yield for the Philippines. Found that El Nio events (which cause drought in the Philippines) are responsible for most of the important fluctuations in production at the national level. In relation to the El Nio drought of 1998, found that most farmers were unable to plant alternative crops, and that the most common coping mechanism was to rely on monetary transfers from family and friends who had access to sources of nonfarm income. Completed a survey of rice crop establishment practices of farmers in Nueva Ecija, Philippines. The results indicate that farmers are likely to establish rice early by seeding only when irrigation authorities are able to assure water supply when rainfall is insufficient.
22
Characterized variation in growth traits of weedy rice populations in the Philippines. Found that variation in trait combination was highly population (farm)-specific, suggesting strong inter-population selection. Developed and began validation of a simulation model DSRICE1 for examining weedcrop competition for direct-seeded rice. Illustrated the importance of vertical leaf canopy distribution in governing yield determinants. Conducted experiments into role of nitrogen in governing root-shoot bioma allocation in weed species and its subsequent effect on rice weed competition and yield. Developed methodology for measuring micro field levels in dry and saturated soils and of methods to quantify variability of emergence in wet direct-seeded fields. Fabricated a ride-on boom sprayer suitable for small Asian rice fields for improved pesticide application and safety.
Supported participants from the Consortium (networks) to participate in IRRI/Networkconducted training course (SRINM, SWBE, Ethnoscience). Conducted the second joint meetings of the IRRC, INMNet, and IPMNet steering committees and the IRRC external review (Hanoi, Vietnam).
Coping with global climate change: reducing methane emission from rice fields R. Wassmann and R. Lantin
q
Research q Conducted data management workshop for pest impact assessment (PIA) research collaborators. q Conducted pest survey in RTDP farms (PIA) for three seasons at seven IRRC sites (crop 1 data had been collated and initially analyzed; crop 2 data had been collated; and most of crop 3 data are still with research collaborators). q Conducted crop residue management (CRM) for three seasons (out of 5) of research at four sites (CRM research will be a continuing activity until the 3rd quarter of 2000).
Program outlook
The Irrigated Rice Ecosystem program will continue research to raise the yield barrier through collaboration in a systems approach to development of new plant types and hybrid rice. High yield-determinant growth patterns, new genes for lodging resistance, tungro viruses, and stem borer and blast resistance will be the focus, along with improved grain-filling capacity in the NPT. Research on highyielding hybrid rice from CMS and TGMS lines will continue with advanced breeding materials introduced to the national agricultural research systems (NARS). The program will continue research on soil quality and nutrient management. Research on optimal N applications, internal nutrient efficiencies, and modeling the nutritional balance will be directed toward sustainable nutrient management in intensively cropped irrigated lowlands. Identification and quantification of key biotic and soil organic determinants of sustainability and ecological resilience will also be addressed. Studies on the constraints to nutrient supply, and the development of practical approaches through site-specific nutrient management technologies in partnership with
Comprehensively evaluated, documented, and published data obtained over 6 years at eight stations of the UNDP project. Overall, assessed mitigation technologies regarding their potential for emission reduction and specifics of application. Clarified the ultra-structural pathway of methane gas through the different plant compartments. Upscaling of emission rates through a process model and GIS techniques yielded conclusive estimates on national source strengths.
Irrigated Rice Research Consortium

Coordination q Provided general coordination of the IRRC: membership, agenda, resource allocation, etc. q Continued implementation of the nutrientpest interaction research. q Supported training workshop (Uttar Pradesh, India) on database management and a site visit (Tamil Nadu, India) for the nutrient-pest interaction research collaborators.
23
NARS and farmers are nearing the stage for largescale technology transfer. Efforts to improve water use on-farm and in irrigation districts will focus on land preparation, cultivation period, and weed and crop establishment in direct-seeded rice under intermittent irrigation. Research on water quality degradation and the development of feasible mitigation strategies will continue. Characterization of pest problems and generation of practical pest management strategies will be sustained. Adaptation of integrated pest management (IPM) will focus on motivating farmers through printed materials and radio. Surrogate taxa analysis of rice invertebrates and assessment of predators directional movement in irrigated rice are under way.
The IRRC will continue to link with NARS and other institutions to address important interdisciplinary regional problems. The networks in IRRC will continue to address NARS-driven IPM, integrated nutrient management, and hybrid rice research priorities. Two SDC-supported projects, the IPM Project (within IPMNet) and RTDP Project (within INMNet), and the ADB-supported Hybrid Rice Network (HRNet) have been formally linked in the consortium. The initial results of the nutrient-pest interactions research in intensive irrigated rice systems are under analysis and will direct the next phase of this work. The IRRC will broaden to include other aspects of irrigated rice research on water management, farm mechanization, and global climate change.
24
Research programs Rainfed lowland rice ecosystem
CHARACTERIZING AND ANALYZING RAINFED RICE ENVIRONMENTS 26 Water-balance modeling to study regional drought risk and crop management strategies (SS, SWS) 26 Land use dynamics and changes in rice production in the Mekong River Delta in the 1990s (SS) 29 Rice yield and yield stability patterns in eastern India (APPA) 32 Categorizing rice systems according to yield and yield stability 32 Production systems within ecosystems 32 ADDRESSING GENDER CONCERNS IN RICE RESEARCH AND TECHNOLOGY DEVELOPMENT 33 Eliciting male and female farmers perceptions of rice varieties (SS, APPA) 33 Description of the farming systems 34 Gender division of labor 35 Male and female farmers criteria for traits of rice varieties 36 Farmer participation in rice varietal selection 37 RAINFED LOWLAND RICE RESEARCH CONSORTIUM 38 Climate, agrohydrology, and management of rainfed rice production in Central Java: a modeling approach (SWS) 38 Effect of climate and agrohydrology on rice production 38 Management options to increase yield 39 Dynamics, balance, and recycling of residual soil N in a lowland rice-sweet pepper system (SWS) 39 Rice yield, N uptake, and N use efficiency 40 Nitrogen balance 40 Carbon management for sustainability of an intensive rice-based cropping system (SWS) 41 PROGRESS OF UNREPORTED PROJECTS 43 Managing crop, soil, and water resources for enhanced productivity and sustainability of lowland areas 43 Germplasm improvement for rainfed lowland rice 44 PROGRAM OUTLOOK 45
Rainfed lowland rice ecosystem
Rainfed lowland rice is grown on more than 48 million ha, with one-third of that in South and Southeast Asia. In most rainfed lowland areas, rice is often the only crop grown. The objectives of the rainfed lowland program are: q improve the understanding of the biotic and socioeconomic constraints to increase productivity and stability of rice yield; q develop technology for better management of soil, water, and biotic resources to increase rice yield and sustain the natural resource base; and q improve germplasm to overcome the constraints imposed by poor soils, drought, and submergence.
Characterizing and analyzing rainfed rice environments

Research through 1997 developed the methodology and database for characterizing the biophysical aspects of the systems at different scales and quantified biotic stresses in rainfed ricefields. Since 1997, research on socioeconomic components has increased. Focus has been on integrating various components of the biophysical components to better describe the complex environment. Water-balance modeling to study regional drought risk and crop management strategies S.P. Kam, T.P. Tuong, B. Bouman, S. Fajardo, and J.P. Reyes Drought is a main cause of low and unstable rice yields in the Korat Plateau of Thailand. An understanding of its temporal and spatial occurrence is important for identifying strategies for geographical targeting of improved varieties. A water-balance model that takes into account climatic, soil, and crop factors was developed to quantify the frequency and severity of drought at different stages of lowland rice. The weekly field-water balance components were calculated as follows: Wt = Wt-1 + Pt ETt Gt Rt where W = field-water storage (mm) (includes water storage within the root zone and standing water on the soil surface); P = precipitation (mm wk1); ET = evapotranspiration (mm wk1). When the soil is bare, this is replaced by evaporation from bare soil
26
[ES]; G = percolation (mm wk1), which occurs when there is standing water in the field; R = runoff (mm wk1), which occurs when the depth of standing water (H) in the field is higher than the effective bund height (Hmax); subscript t signifies the week under calculation; and t-1 signifies the previous week. Starting with a given field-water storage of the previous week (Wt-1), the model consecutively calculates ETt (or ESt if the soil is bare), Gt, Rt, and Wt. The water-balance model was run using data of three typical soil types ranging from the coarsest (sandy loam) to the finest (clayey loam) soils encountered in the Korat Plateau. Percolation rates and detailed measurement of water retention functions reported by previous investigators were used. The water-balance model also took into account the effect of late planting on delayed flowering and harvest. The week of disappearance of standing water at the end of the rainy season was determined relative to the flowering week. That and the relationship between yield and number of weeks of water disappearance before flowering were used to determine
the relative yield reduction due to late-season drought stress. The model was used for 1) a time-series analysis and 2) a spatial surface analysis by linking with geographic information systems (GIS). The time-series analysis used yearly records of weekly climatic data to model annual variations in weekly field-water storage. That analysis was done at eight sites. Frequency analysis covered the effect of drought on the timeliness of transplanting, harvest week, and yield reduction due to late-season drought stress. In the spatial-surface analysis, long-term average rainfall data of 126 stations and average Penman evapotranspiration data of eight stations were interpolated to generate gridded surfaces as input into the water-balance model. The outputs of this analysis included 5- 5-km gridded surfaces of weekly fieldwater storage, timeliness of transplanting, residual soil moisture at harvest, and yield reduction due to late-season drought stress. Figure 1 illustrates the spatial distribution of the field-water storage at week 31 for clay loam soil. Similar figures were produced for each week of the
297 mm
350 300 250 200 150 100 50 0 1 5 9 13 17 21 25 29 33 37 41 45 49
400 300 200

350 300 250
350
100 0 1 6 11 16 21 26 31 36 41 46 51
300
Loei
200 150 100 50 0 1 5 9 13 17 21 25 29 33 37 41 45 4 9
250
Sakon Nakhon
35 0 25 0
Nakhon Phanom
200
150
100
50
Field Capacity
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52
0
15 0 50
Stand ard meteorol ogical week

1 6 11 16 21 26 31 36 41 46 51
400 300 200 100 0 1 6 11 16 21 26 31 36 41 46 51
-5 0
Khon Kaen
Roi Et
237 mm
3 50 300 2 50 200 1 50 100 50
350 300 250
0 1 5 9 13 17 21 25 29 33 37 4 1 45 49
3 50 2 50
200 150 100 50 0 1 6 11 16 21 26 31 36 41 46 51
1 50 50 - 50
Ubon Ratchathani
11 16
21 26
31
36
41
46
51
Surin
Chok Chai
103 mm
1. Field water storage at week 31, based on average weekly rainfall on clay loam soil. Korat Plateau, Thailand, 1999.
27
year. Field-water storage decreases markedly across the Korat Plateau from the northeast to the southwest. Table 1 shows the frequency at which transplanting was possible (when there were at least 3 consecutive wk of standing water) before week 35 (start of September) for three soil types at eight selected sites. Sites in the northeastern and eastern parts of the plateau had higher frequency of transplanting than sites in the western part. Figure 2a shows yield reduction due to late-season drought stress for transplanted rice on clay loam. Figure 2b is the yield reduction for direct-seeded rice. Comparison of the two maps suggests three major zones, indicated in Figure 3, for which different crop management strategies may be relevant. Zone A (east-northeastern belt): Transplanting induces less yield reduction when compared with
Table 1. Effect of soil type on possibility of transplanting rice (TPR) before week 35 for eight selected sites in the Korat Plateau, Thailand, 1999. Years of possible TPR in 12 years (no.) Site Sandy loam Nakhon Phanom Sakon Nakhon Ubon Ratchathani Surin Roi Et Khon Kaen Loei Chok Chaia
a
Loamy sand 12 8 10 4 3 1 3 0
Clay loam 12 12 11 10 10 8 10 2
12 6 8 3 0 1 0 0
direct seeding. However, because rainfall ceases rather abruptly at the end of the season, late-maturing photoperiod-sensitive varieties (such as KDML 105) may encounter drought around anthesis, causing yield reduction. The main strategy here is to avoid late drought by the use of nonphotoperiodsensitive varieties. Zone B (lower central part of the plateau): Transplanting is possible but often occurs later in the monsoon season (between wk 26 and 34) and is normally later than direct seeding. Under optimum crop management, direct seeding would be more advantageous. Zone C (western part): Yield loss in transplanted rice is distinctly higher than dry direct-seeded rice. Drought risk in this zone occurs in the early part of the season and midseason. Therefore the more appropriate management strategy here is the use of direct-seeded, drought-tolerant varieties. Figure 2c depicts the yield reduction for late-season drought stress on direct-seeded rice for the loamy sand soil. Percolation rates used in Figures 2b and 2c were similar. The clay loam soil in Figure 2b, however, had higher water retention than the loamy sand soil in Figure 2c. The high degree of similarity between Figures 2b and 2c shows that percolation, rather than the inherent water retention properties of the soil, is the overriding factor that influences fieldwater storage. This suggests potential for improving water storage through management such as soil compaction to reduce percolation losses.
Only 8 years of weather data available.
a. Transplanting case, clay loam scenario
b. Direct seeding case, clay loam scenario
c. Direct seeding case, loamy sand scenario, with halving of percolation rate % Yield Reduction
4 8 15 21 32 48 73 Not applicable
2. Simulated yield reduction due to late-season drought stress of a photoperiod-sensitive rainfed lowland rice variety. Korat Plateau, Thailand, 1999.
28
Zone A
Zone C
Zone B
3. Zonation (A,B,C) for rice crop management strategies to cope with drought stress. Korat Plateau, Thailand, 1999.
The accuracy of the output of this study is constrained by a paucity of data for key soil variables, particularly percolation rates. Nevertheless, the study demonstrates how a simple water-balance model linked with GIS can be used for regional drought analysis. It has implications for broad-scale geographical targeting of crop management strategies. The results provide the basis for selecting target areas for in-depth studies of agrohydrological influence on field-water balance. Land use dynamics and changes in rice production in the Mekong River Delta in the 1990s C. Edmonds, S.P. Kam, L. Villano, and H.C. Viet10 Biophysical and socioeconomic constraints influence land use decisions and affect the production and income of rice farmers. Characterizing both types of constraints and understanding the relationship between them are essential to development of technologies and policies to increase rice production and income. Research in collaboration with the Institute of Agricultural Sciences, Vietnam, characterized changes in agricultural environment and farm responses to changes that enabled production increases of about 6.3% y1 in the Mekong River Delta during the 1990s.
Farm-level changes in rice output were captured in farm survey data (1994-97) for 149 farms (eight villages) in the Mekong River Delta. The villages represent a range of agroecological and production situations. GIS data on soil types, flooding, water salinity levels, and transport routes were used to characterize the biophysical environment. Integration of econometric techniques with data organized in GIS offers promise for modeling the effect of the main constraints to rice production. We applied microeconomic modeling to explore relationships between biophysical and socioeconomic characteristics, to derive hypotheses, and to form an estimation equation that can be tested using available data. Estimates assess the importance of new technology adoption, changes in input application, increased cropping intensity, and local infrastructure development in explaining observed production increases. Figure 4 superimposes land use as reported by farms in each of the eight villages on a 1996 landuse map for the Mekong Delta. The figure highlights the benefit of integrating GIS with farm survey data. The map gives a complete characterization of land use, while information from the survey adds a time dimension and provides detailed data on farm resources and activities. For our model, accessibility to markets and rice plots plays a key role in determining the rice cropping intensity likely to be adopted by farms. The GIS generates accessibility indicators. These are used to predict land use and production decisions. Estimates predicting the rice cropping intensity of farms were significant in each of the 4 years. Greater distances between farms and markets were associated with a reduced probability of intensive rice cultivation. The distance between the farm plot and homesteads had a negative and statistically significant effect on rice cropping intensity in 1995 and 1997 estimates. The study estimated other land use, as well as rice production and supply functions implied by the analytical model or basic microeconomic theory, which included general cropping intensity, farm cropping pattern, and rice production and farm supply of rice to the market. Together, the estimates provide a clear indication of the factors driving farmland use, production, and marketing decisions. The availability of saline-free irrigation water to farms had positive and statistically significant effects on the intensity of land use. The magnitude of the
29
4. Land use in the Mekong River Delta in the 1990s. Source of land use map is the Soil Science Department, Can Tho University, Vietnam, and the IAS-IRRI project. IRRI, 1999.
30
effect of high-quality irrigation was much greater than the effects of other variables included in the model. The level of rainfall had a mixed effect on cropping intensity. In years with normal to above-average rainfall, higher levels of rainfall were associated with increased cropping intensity. In 1996, increased rainfall was associated with significantly reduced cropping intensity. The size of families landholdings relative to their available labor had mixed signs across models and years, but in general supported the hypothesis that relative scarcity of land in relation to labor led to higher cropping intensity. Other variables such as education, age, or farming experience had inconsistent or insignificant effects on land use. The estimates of price elasticity of supply ranged between 0.145 and 0.319 in yearly estimates. The implications of regression model estimates are made clearer by using them to formulate a simulation model to assess the effect of policy changes, or investments in infrastructure, on land use and rice production. The results of a model derived from our empirical estimates are summarized in Tables 2 and 3. Table 2 shows the distribution of mono-, double, and triple-cropped rice among surveyed farms for 1994-97 with three alternative infrastructure investment scenarios. The implied changes in the share of farms with double- or triple-cropped rice can be used to derive an implied increase in aggregate rice output using results from production function estimates. The estimated changes in total rice production are seen on Table 3. The simulation illustrates how the results of land use and production estimates based on the observed behavior of farms can be used to assess the likely effects of different investments on rice production levels. In this case, it shows the large effect of investments in irrigation, and a more moderate effect of improvements in the transportation system, on rice production. The integration of behavioral parameters from econometric estimates is broadly applicable and can provide an empirical foundation to larger simulation models. Because this study relied on existing data, we encountered severe data constraints. However, the research provided insight into the farm-level changes in land use and production systems that enabled the rice production increases in the Mekong River Delta in the 1990s.
Table 2. Simulation of effects of infrastructure investments on distribution of farm rice cropping intensity in the Mekong River Delta, Vietnam. IRRI, 1999.
Predicted distributiona: 10% better water management
1997 1996 1995 1994 1995 1996 1997 1996 1997 1994 1995 1996 1997 1994 1995 1994
Predicted distributiona: home to plot reduced by 1 km
Predicted distributiona: less than 10 min travel to market
Actual distributiona
Monocrop 37 Double cropped 7 Triple cropped 16
Rice cropping system
33 55 26
a
31 51 32
Farms (no.)
12 45 20
36 7 17
21 62 31
26 51 37
12 45 20
40 6 14
29 57 28
33 51 30
12 44 21
15 8 37
13 67 34
0 52 62
12 38 27
31
Table 3. Simulation of effects of infrastructure investments on rice production (in t) among surveyed farms in the Mekong River Delta, Vietnam. IRRI, 1999.
Predicted productiona: 10% better water management
1996
0 24 51 75 31.5
Rice yield and yield stability patterns in eastern India V.P. Singh and A.S.R.A.S. Sastri24 Quantification of yield, yield stability, and their relationship to environmental parameters is beneficial to identifying strategies for sustainable rice production. The districtwise yearly average rice databases from 1970-71 to 1994-95 were used to calculate the yield coefficient of variation (CV) and do a timetrend analysis of the yield for all of Bihar, Orissa, West Bengal, and eastern parts of Uttar Pradesh and Madhya Pradesh.
CATEGORIZING RICE SYSTEMS ACCORDING TO YIELD AND YIELD STABILITY
1997 1995 1994 1995 1996 1997
Predicted productiona: home to plot reduced by 1 km
20 43 23 86 1.0
12 24 25 60 2.1
7 28 17 51 0.4
10 6 31 47 7.6
9 50 28 87 4.9
7 24 21 52 2.6
1996 1997 1994 1995 1996 1997 1994
Columns may not add to total rice production due to rounding.
7 28 16 51
Yields and CV values in each state vary considerably. Therefore, criteria were developed (Table 4) for categorizing the rice production systems based on yield and its variability. From those, rice production can be categorized in four groups (Table 5). The spatial distribution of the four groups in Bihar is presented in Figure 5. The overall analysis of eastern India shows that among 93 districts, 54 have a low and unstable system of rice production, 16 have a low and stable system, 22 have a high and unstable system, and only one has a high and stable system.
PRODUCTION SYSTEMS WITHIN ECOSYSTEMS
Predicted productiona: less than 10 min travel to market
Actual productiona
11 24 26 61
24 5 14 44 0.4
14 47 26 86 3.0
10 24 30 64 5.2
7 28 16 51 0
27 5 12 43 0.9
District rice area under different categories of production systems was classified under different rice ecosystems (Table 6). Almost 16 million ha of rice are in the low and unstable category, of which 26.1% is in the upland ecosystem, 45.3% is in the rainfed lowland ecosystem, 12% is in the deepwater ecosystem, and the remaining 16.6% is in the irrigated system. About 2.7 million ha of rice area are in the low and stable category of production system. This entire area has a good potential for increasing rice yield. Likewise, about 5.7 million ha are in the high and unstable category, which would need intervention to stabilize yields. Long-term strategies would be required for improving rice production in the low and unstable production system. The logical approach would be in
1995 1994
Monocrop 25 Double cropped 5 Triple cropped 13 Total product 44 Change (%) in total product
Rice cropping system
22 41 22 85
32
Table 4. Rice yield and variability classification for eastern India. 1999. Yield (t ha ) <0.8 0.81 to 1.0 1.01 to 1.20 1.21 to 1.40 1.41 to 1.60 >1.60
-1
Variability CV (%) <10.0 10.1- 15.0 15. -20.0 20.1-25.0 25.1-30.0 >30.0 Code 6 5 4 3 2 1 Classification Very stable Stable Moderately stable Moderately unstable Unstable Very unstable
Code 1 2 3 4 5 6
Classification Extremely low Very low Low Moderately high Fairly high High
Table 5. Rice production system classification in eastern India. 1999. Combination of the codes of Yield 1, 2, & 3 1, 2, & 3 4, 5, & 6 4, 5, & 6 Variability 1, 2, & 3 4, 5, & 6 1, 2, & 3 4, 5, & 6 Category of rice production system Low and unstable Low and stable High and unstable High and stable
Eliciting male and female farmers perceptions of rice varieties T R. Paris, A. Singh,11 M. Hossain, J. Luis, H.N. Singh,11 O.N. Singh,11 S. Singh,11 and R.K. Singh Encouraging women farmers to participate in the process of germplasm enhancement and conservation is important for achieving positive impact on rice farming families in rainfed environments. Farmer participatory plant breeding (PPB) for rainfed rice was developed at IRRI in 1997 in collaboration with national agricultural research systems (NARS) in eastern India. The aims were to test the hypothesis that farmer participation for rainfed rice breeding can help develop suitable varieties more efficiently. The research has two components: q a plant breeding component, which aims to develop and evaluate a methodology for screening improved germplasm for heterogeneous environments through farmer participation; and q a social science component, which aims to establish a farmers typology to guide establishment of breeding goals that suit the needs of various groups. In 1998, we began to incorporate gender concerns in the ongoing PPB project. The strategies were q Develop methodologies for assessing male and female farmers criteria of useful traits of rice varieties. q Develop participatory approaches that include male and female farmers in selecting new rice lines.
stepsincrease some yield and stabilize it at that level before taking the next step of increasing yield, and so on. The stepwise approach would not require large expenditure on inputs at once and the yield stabilization period would provide options for building the resource base to some extent. That can then be used for increasing the yield to the next level.
Addressing gender concerns in rice research and technology development

Research in this project aims to incorporate womens knowledge and concerns in identifying and adapting mechanical postharvest technologies that add value to rice and byproducts. It also examines methodologies for incorporating male and female farmers knowledge of traditional cultivars and their criteria for varietal selection into plant breeding strategies. IRRI participates in the CGIAR systemwide initiative on Participatory Research and Gender Analysis for Technology Development and Institutional Innovation.
33
W. Champaran
E. Champaran Sitamarhi Gopalganj Madhubani Siwan Saran Muzaffarpur Darbhanga Vaishali Samastipur Bhojpur Patna Nalanda Rohtas Jehanabad Munger Gaya Aurangabad Nawada Bhagalpur Godda Sahibganj Madhepura Saharsa Purnia
Begusarai
Katihar
Hazaribag Daltenganj
Giridih
Deoghar Dumka
Dhanbad Low and unstable Ranchi Low and stable High and unstable High and stable Data not available Gumla West Singhbhum East Singhbhum
5. Rice production systems in Bihar District, India. 1999.

q
Enhance womens knowledge and skills in germplasm conservation. Enhance NARS capacities for conducting male and female farmer participatory approaches in rice germplasm enhancement and conservation in rainfed rice environments.
DESCRIPTION OF THE FARMING SYSTEMS
We initiated the gender study in Mungeshpur in Faizabad District and Basalatpur in Siddathnagar District, eastern Uttar Pradesh. PPB research had included the same villages. Basalatpur represents favorable lowland rainfed areas and Mungeshpur
34
Table 6. Area under various rice production systems and ecosystems in eastern India. 1999. Rice area ('000 ha) Rice production system and ecosystem Eastern Uttar Pradesh Eastern Madhya Pradesh
Bihar
Orissa
West Bengal
Total
Low and unstable Upland Rainfed lowland Irrigated Deepwater Total Low and stable Rainfed lowland Irrigated Deepwater Total High and unstable Rainfed lowland Irrigated Deepwater Total High and stable Rainfed lowland Irrigated Deepwater Total Total Upland Rainfed lowland Irrigated Deepwater Total
531 1,142 552 807 3,032
691 1,667 844 495 3697
714 1,723 648 606 3,691
1,349 2,695 608 4,652
883 883
4,168 7,227 2,652 1,908 15955
762 100 187 1,049
437 41 55 533
563 260 268 1,091
1,762 401 510 2,673
248 660 30 938
125 177 302
451 334 183 968
1,592 1,064 792 3,448
2,416 2,235 1,005 5,656
11 200 30 241
11 200 30 241
531 2,163 1,512 1,054 5,260
691 2,229 1,062 550 4,532
714 2,174 982 789 4,659
1,349 2,695 608 4,651
883 2,155 1,324 1 060 5,422
4168 11416 5,488 3,453 24525
represents shallow and submergence-prone areas that are favorably rainfed during years of low rainfall. Although modern varieties are grown in Mungeshpur, they are not suitable for cultivation as rainfed lowland rice and often suffer from submergence and drought and stress at the reproductive and ripening phases when the crop is planted late. Farmers in Mungeshpur have more access to supplementary irrigation than farmers in Basalatpur, which enables farmers to grow vegetables and to increase crop diversification during the winter season (Nov-Dec to Mar-Apr). Most farm households in Mungeshpur belong to the lower caste (backward and scheduled), while Muslims dominate the population in Basalatpur. The Yadavs, a subcaste of the caste in Mungeshpur, take care of milch animals. Degree of market orientation is high in Basalatpur, which is near a city. Rice grown in
Mungeshpur is mainly used for home consumption. Rice/wheat mixed with mustard is the dominant cropping pattern in both villages, occupying more than half the cultivated area.
GENDER DIVISION OF LABOR
The extent of female participation in rice production is high in both villages. Land preparation and application of chemicals are mostly mens responsibilities. In Mungeshpur, women from the lower social status dominate in pulling of seedlings (100%), transplanting (70%), weeding (80%), application of farmyard manure (60%), harvesting (82%), and threshing (82%). In Basalatpur, men and women equally share in the pulling of seedlings and harvesting. Women do the transplanting of seedlings
35
(100%), and most of the weeding (75%) with men doing most of the spraying (90%). Basalatpur farms are mechanized, using tractors for land preparation and threshing, but in Mungeshpur bullocks are used for land preparation and threshing is done by hand. Livestock is more important in Mungeshpur than in Basalatpur. Postharvest activities such as seed selection, storage, manual dehulling, hand threshing, parboiling rice, and making rice puffs are major responsibilities of the women of small farming households. Aside from their significant contributions in crop production, they take care of dairy cattle, collect green animal fodder, and feed and tend livestock.
MALE AND FEMALE FARMERS CRITERIA FOR TRAITS OF RICE VARIETIES
We interviewed 15 males and 15 females from separate households and different social groups in Mungeshpur and Basalatpur. Each farmer was asked what traits he, or she, considered in selecting rice varieties for each major land type in their fields. We asked: If you had 100 paisa, how much would you
pay for each trait? to obtain the weight or importance given by farmers to a particular trait. We summed the weights per trait of all respondents in each land type and took the proportion of each trait to all traits mentioned. Table 7 shows the selection criteria of male and female farmers for different land types and villages. Male and female farmers in Basalatpur agreed that grain yield, grain price, and taste are the most important traits they consider in choosing rice varieties. Grain price is an important consideration for commercial farmers. Many produce the traditional variety, Kalanamak, as it commands a higher price because of good taste and aroma even though it yields 1.52 t ha 1. In contrast, grain price is not an important consideration in Mungeshpur because rice is mainly used for home consumption. Male and female farmers in Mungeshpur cited grain yield and duration (125135 d) as the most important traits they consider in growing rice. Women, but not men, mentioned competitiveness with weeds and postharvest quality as important traits in selecting rice varieties. Weeds are the ma-
Table 7. Traits farmers mentioned as useful when selecting rice varieties in Basalatpur and Mungeshpur, eastern Uttar Pradesh, India, 1999. Basalatpur Upland Traita Male (%) Grain yield 36.8 Duration 25.9 Grain price 0.0 Resistance to abiotic stress 8.4 Biomass quality 3.3 Taste 1.7 Bold and pure grain 7.7 Adaptation to specific soil type 3.0 Postharvest quality 0.8 Resistance to biotic stress 4.2 Cooking characteristics 0.8 Response to fertilizer 2.5 Competitiveness with weeds 0.0 Resistance to lodging 1.7 Early vegetative vigor 0.9 Culm strength/diameter 0.0 Adaptation to several preparations 1.8 Female (%) 39.5 34.5 0.0 6.5 2.5 0.5 1.5 3.0 3.0 2.5 1.0 1.0 0.0 0.0 0.5 0.5 3.5 Male (%) 48.7 0.7 15.6 0.8 5.3 10.3 1.7 2.3 6.7 1.0 1.7 2.7 0.0 2.3 0.3 0.0 0.0 Female (%) 49.7 1.0 16.0 0.5 5.7 12.3 0.0 0.7 7.8 1.3 2.0 1.3 1.0 0.7 0.0 0.0 0.0 Male (%) 41.6 20.5 1.7 5.6 5.0 2.8 4.4 5.0 0.0 3.9 3.9 5.0 0.0 0.0 0.6 0.0 0.0 Female (%) 36.0 25.9 2.8 6.2 2.3 2.8 4.4 4.4 5.1 1.7 3.9 2.3 2.2 0.0 0.0 0.0 0.0 Male (%) 42.1 20.6 3.0 5.1 5.4 2.1 3.4 5.4 0.0 4.3 3.4 4.3 0.0 0.9 0.0 0.0 0.0 Female (%) 40.4 15.0 1.8 5.0 8.6 3.2 5.0 6.4 2.3 3.2 5 1.8 2.3 0.0 0.0 0.0 0.0 Lowland Upland Mungeshpur Lowland
a Grain yield includes tillering, panicle length, and number of grain; resistance to biotic stress includes resistance to pests, bugs, and blast; resistance to abiotic stress includes resistance to Zn deficiency and drought; biomass quality includes height, quality, and quantity of straw; postharvest quality includes easiness to dehull and milling recovery; and cooking characteristics include cooking time, elongation ability, aspect after cooking, and impression in the stomach.
36
jor problem in the uplands, particularly when rice is direct seeded. In the lowlands, weeds are more prevalent during drought. Gender-specific tasks influenced choice of variety. Tradeoffs among the different traits in variety selection between men and women representing different socioeconomic groups will be studied in 2000.
FARMER PARTICIPATION IN RICE VARIETAL SELECTION
Two farmers from each of the villages of Mungeshpur and Sariyawan (rainfed neighboring village) and Basalatpur were selected to check the performance of 13 rice cultivars on their fields during the 1998 monsoon season. Included were 10 advanced lines and three released varieties for lowlands. Two were scented varieties: Kamini, which flowers in 136 d, and Sugandha, which flowers in 124 d. Seed was distributed through the PPB project and breeders selected the most promising lines with farmers input. Female farmers were included to give them an equal chance to participate in decisions in selecting rice genotypes. Ten farmers (five females and five males) visited individual plots on farmers fields and ranked the genotypes at maturity. Farmers were asked to rank the lines from 1 (excellent)
to 13 (worst) on the basis of visual assessment. The farmer rankings generated an n k matrix, where n are the lines being evaluated and k are the farmers evaluating the crop performance. Kendalls coefficient of concordance (W) was used to measure the agreement in rankings among male farmers and female farmers, and the correlation between male and female farmers ranking. High and significant correlation values indicate close agreement on the ranking of the 13 rice genotypes by men and women in the sample. Table 8 shows that male and female evaluators in the two villages were in close agreement in the ranking of the 13 lines. The Ws were highly significant, revealing that farmers ranking is acceptable. In Mungeshpur, NDR973004 was ranked highest by both males and females because it matures early and yields better than the local check (Mashuri). Both NDR973003 and NDR9730012 are resistant to drought. In Basalatpur, both male and female farmers rated Kamini as the best variety because of its higher yields (33.5 t ha1) and early maturity compared with Kalanamak, which matures in 160165 d. The initial findings indicate that although women have less access to education and extension services, they are as knowledgeable as men in identifying the
Table 8. Average scores and farmers rankings of 13 rice genotypes in on-farm plots at Basalatpur and Mungeshpur, Uttar Pradesh, India, 1999. Basalatpur Genotypea Av score NDR-40032 Kamini-KMJ-1-17-2 NDR-973004 NDR-95003 Satyam-Sugandha NDR-973000 NDR-9730015 NDR-9730020 Malasia-Mashuri RAU-136- RAU-1306 NDR-9730012 RAU-1326 NDR-9730025 Kendalls Wc
a
Mungeshpur Females Males Av score 5.2 8.0 1.2 3.4 11.2 8.6 8.4 9.6 5.6 4.6 1.8 11.8 11.6 0.87* Rank 5 7 1 3 11 9 8 10 6 4 2 13 12 Females Av score 4.8 3.8 1.4 4.2 12.6 12.0 10.8 9.4 7.0 6.0 1.6 7.4 10.0 0.88* Rank 5 3 1 4 13 12 11 9 7 6 2 8 10
Males Rankb 2 1 5 7 6 8 3 9 11 12 4 13 10 0.75*
Av score 3.8 1.4 4.0 7.4 6.8 6.6 3.3 9.0 11.2 11.6 4.2 13.0 8.8
Rank 3 1 4 8 7 6 2 10 11 12 5 13 9 0.85*
3.6 2.0 5.4 7.0 5.8 7.0 3.6 8.2 11.0 11.6 4.2 13.0 8.6
Where genotypes differ, the first given is that used in Basalatpur, the second is that used in Mungeshpur. bRankings from 1 (excellent) to 13 (worst) on basis of visual assessment. cKendalls coefficient of concordance (W); * = significant at P = 0.01
37
useful traits in rice. If given proper guidance, technical knowledge, and access to new seeds suited to their environment and needs, women can contribute strongly to faster adoption of improved rice varieties. We will continue to validate these findings at other sites and further develop methodologies for involving women in variety selection and sensory evaluation.
EFFECT OF CLIMATE AND AGROHYDROLOGY ON RICE PRODUCTION
Rainfed Lowland Rice Research Consortium

The Rainfed Lowland Rice Research Consortium (RLRRC) implements research in rainfed lowland environments in Bangladesh, India, Indonesia, Philippines, and Thailand. Each site represents a subecosystem for research focus. Climate, agrohydrology, and management of rainfed rice production in Central Java: a modeling approach A. Boling, T.P. Tuong, B.A.M. Bouman, M.V.R. Murty, and S.Y. Jatmiko4 The typical rainfed cropping system in Central Java includes dry-seeded rice (DSR) grown from November to February (gogorancah), followed by a transplanted rice (TPR) from March to June (walik jerami). Earlier studies showed that the yield of the walik jerami crop was lower and less stable than that of the gogorancah crop. We collaborated with the Central Research Institute for Food Crops, Indonesia, to assess the climatic and agrohydrologic constraints to rice production and explore management strategies to increase the yield and yield stability of the double rice cropping system in Central Java. The crop growth simulation model ORYZA was used. The ORYZA model was validated with data from 1995-96 field experiments at Jakenan. Long-term simulation of the potential and rainfed rice yield was done for IR64 grown on a 15-d planting interval for 1977-98. Three water table depth scenarios (medium, shallow, and deep), corresponding to the mean, mean + standard error, and mean standard error of measurements in six rice cropping seasons were used in the simulation. A series of simulation runs explored the effect on yield of supplementary irrigation and of varieties with duration shorter than IR64.
Potential yield. The simulated, long-term average potential rice yield ranged from 6 to 8 t ha1 (Fig. 6). The standard errors of the average values are low, indicating little variation in potential yields across years. Rice sown in the typical walik jerami period had a higher yield potential than rice planted in the typical gogorancah period. Radiation and temperature regime are thus not the determinant factors for the reportedly low and unstable yield of the walik jerami rice. Rainfed yield. Figure 6 shows that the yield of rice sown from mid-November to the end of March differed significantly among the three water table depths. Rice yields reached the potential yield level with the shallow water table, but yields were much reduced in the other two water table depths. For crops sown at the end of the wet season (WS) and in the dry season (DS) (roughly April-November), the yield was low and yield differences among the water table scenarios diminished. The low yields in this season were attributed to a deep water table (>1.2 m) and to inadequate rainfall. The results highlight the important effect of the groundwater depths on rice yield. More research is needed to better quantify groundwater depth.
Simulated yield (t ha-1) 12 10 8 6 4 2 0 O N D J F M A M J J A S O Dry seeded O N D J F Transplanted M A M J J Dry seeded A S O

Rainfed, shallow water table Rainfed, medium water table Rainfed, deep water table Potential
Day of seeding
6. Simulated potential and rainfed rice yield (meanSE) in Jakenan, Central Java, Indonesia. IRRI, 1999.
38
MANAGEMENT OPTIONS TO INCREASE YIELD
Establishment date. Figure 6 shows that the yield of rainfed rice started to decline at different seeding dates in the three water table scenarios. The yield decline was attributed to drought stress during the reproductive stage of the crops. The results indicate that timely planting is critical to sustain yield of the walik jerami crop. Cut-off dates could be determined for transplanting of walik jerami. Beyond those dates, the farmer could choose to forgo the walik jerami crop or plant it and risk low yield. Supplementary irrigation. Yield increase in irrigation scenarios is seen in Figure 7. Irrigation whenever the topsoil fell below field capacity (scenario I1) increased the yields of crops sown in a typical walik jerami period (FebMar) by 2.13.2 t ha1 when compared with yields as nonirrigated rainfed rice. A daily irrigation of 7.5 mm water from panicle initiation to maturity of the crop (scenario I2) increased the yield by 1.53.1 t ha1, and a daily irrigation of 3.3 mm water from panicle initiation to maturity (scenario I3) increased the yield by 0.5 1.2 t ha1. The maximum yield increases for every m3 of irrigation water applied in the three scenarios occurred for crops sown in early March. Water stored in on-farm reservoirs (now widespread in Central Java) can thus be used effectively to increase yield and yield stability of walik jerami crop. The simulation study may supply necessary information for
Yield increase (kg ha-1 m-3 irrigation)
the required capacity of the reservoirs as well as for the cost-benefit of using the reservoirs for supplementary irrigation. Short-duration rice variety. The combined yield of the gogorancah and walik jerami crops for IR64 (V1) and for a hypothetical variety (V3) with 10-d reduction in growth duration as function of sowing date of the gogorancah crop is shown in Figure 8. In these simulations, a 10-d turnaround time between the harvest of the gogorancah and the transplanting of the walik jerami crop was used. The combined yield of variety V3 sown on 15 Nov was lower than that of IR64, but the combined yield when the gogorancah was sown in December was higher than that of IR64 at all probability levels. Early planting (and hence early harvest) of gogorancah did not expose the succeeding walik jerami to late-season droughts and switching to shorter duration varieties reduced the combined yield because the potential yield was reduced with shorter duration. In the years with late arrival of rainfall, growing a short-duration variety in the gogorancah season advanced the transplanting of the walik jerami crop and helped it escape the late-season droughts. In such years, planting varieties with shorter duration may help increase yield and yield stability. Dynamics, balance, and recycling of residual soil N in a lowland rice-sweet pepper system J.K. Ladha, R.K. Shrestha, and S.R. Pascua25 Lowlands in Ilocos Norte, Philippines, have a system of WS rice followed by one or two nonrice, highvalue DS crops (sweet pepper, tomato, garlic,
Yield (t ha-1) 16 12
Nov 15 V1 Dec 01 V1 Dec 15 V1 Nov 15 V3 Dec 01 V3 Dec 15 V3
1.0 0.8 0.6 0.4 0.2 0.0 J F

-3 I1: 0.34 cm3 cm3 0.34 cm3 1 mm -1 I2: PI to M, 7.5 mm dd1 PI to 2 1 I3: PI to M, 3.3 mm d-1 PI to 3
4 0 0.00
Day of seeding 7. Yield increase in three irrigation scenarios for rice planted from February to April in Jakenan. Irrigation scenarios include irrigation when moisture of the topsoil falls below field capacity (0.34 cm3 cm3, I1); and when irrigated daily with 7.5 mm (I2) and 3.3 mm (I3) from panicle initiation to maturity. IRRI, 1999.
0.20
0.40
0.60
0.80
1.00
Probability of exceedance
8. Combined yield of gogorancah and walik jerami rice crops using IR64 (V1) and a variety with a 10-d reduction in duration (V3) with three seeding dates. IRRI, 1999.
39
mungbean, maize, and tobacco). Farmers usually apply high rates of N fertilizer to the DS crops, resulting in N losses ranging from 34 to 549 kg ha1. The largest N loss is in rice-sweet pepper, a predominant cropping sequence. The reasons for the excessive use of N fertilizer are not known, but high economic returns discourage farmers from being concerned about sustainability. The N not used by the crop is prone to loss through leaching or denitrification or both, which results in a net reduction of the mineral N pool at the beginning of the rice crop. A fallow period during the dry-to-wet season (DTW) transition provides an opportunity to grow a crop to capture NO3-N and recycle it through residue incorporation. Experiments with rice-sweet pepper in farmers fields at four sites aimed to 1) quantify the amounts of N released from the residues of different transition crops and their effects on rice yields and N use efficiency, and 2) estimate the soil mineral N balance. Nitrate catch crops were grown during the DTW transition period. At site 1, the DTW transition treatments differed in the species of transition crops grownindigo, indigo+mungbean, and maize in addition to fallow (farmer practice). Transition crop residues equivalent to 87 kg N ha1, after determining the N content of all transition crops on the day of harvest, were returned to the soil during WS. Nitrogen fertilizer (87 kg N ha1) for rice, if applied, was prilled urea (PU) or tablet urea (TU). At other sites, the DTW transition treatments included fallow and indigo, and WS treatments included no fertilizer N, fertilizer N, and residue N equivalent to 87 kg N ha1.
RICE YIELD, N UPTAKE, AND N USE EFFICIENCY
residue treatments because of high residual soil mineral N. At site 3, yield and N uptake with residue were lower than with PU for an unknown reason. The increase in grain yield from catch-crop residue N ranged from 39% to 90% of fallow, which was equivalent to, or higher than, the equivalent amount of N (87 kg PU N ha1). Although NH4+-N release was highest with residue or residue with PU, N uptake was highest with fertilizer N, especially with TU. However, grain yields with residue alone or in combination with either PU or TU were similar to those with fertilizer N but significantly higher than those with the control (zero N). The results indicate a better synchrony of N release and crop demand when a slow-release fertilizer such as TU or residue plus urea is used.
NITROGEN BALANCE
Rice grain yield ranged from 3.1 to 6.6 t ha1 across sites and treatments, with a corresponding N uptake of 48 to 146 kg ha1 (Table 9). Grain yield and N uptake were significantly different with residue vs without residue, and with PU vs TU. But both parameters did not differ among residues. At site 1, yield with TU was about 1.2 t ha1 higher than with the equivalent amount of PU. Yields were similar with the integrated use of indigo or mungbean, or both, with mineral N. At site 2, rice yield and N uptake were similar in the control and
An apparent mineral N balance was calculated to evaluate the effect of DTW transition and WS management on reducing N loss (Table 10). At site 1, the N loss from the 100-cm soil profile ranged from 208 to 516 kg ha1. It was highest in fallow and lowest in the plot with maize grown during the DTW transition and its residue incorporated. The efficiency of DTW transition crops in decreasing N loss was in the order of maize > indigo+mungbean > indigo. Maize was highly efficient due to its ability to capture a larger amount of soil N (177 kg ha1) than indigo (151 kg ha1) and indigo plus mungbean (118 kg ha1). At sites 2 and 3, losses were similar to those for site 1. At site 4, the loss in the indigo plot was negligible. The reduction in N loss due to integration of a catch crop during DTW and its incorporation for rice ranged from 33% to 72% of the N lost in the fallow. The inability of the transition crop to reduce N loss at sites 1, 2, and 3, as compared with site 4, was due to the high buildup of soil mineral N (range from 494 to 908 kg ha1 with 97% as NO3-N), which was beyond the capacity of the transition crop to capture. At site 4, indigo was able to reduce N loss by 72% due to lower soil mineral N (260-313 kg ha1). Residual mineral N was 23 times higher than the uptake by the transition crop at sites 1, 2, and 3. That resulted in as much as 83% loss of residual N. This indicates that the strategy of use of a N transition crop alone is not effective enough to reduce sig-
40
Table 9. Effect of dry-to-wet season (DTW) transition crops on grain yield, N uptake, and agronomic N use efficiency of rice in Ilocos Norte, Philippines. IRRI, 1999. Treatment DTW transition Wet seasona Grain yield (t ha1) N uptakeb (kg ha1) Agronomic N use efficiency (kg grain kg N1 applied)
Site 1 Fallow Fallow Fallow Indigo Indigo Indigo Indigo + mungo Indigo + mungo Indigo + mungo Maize Maize Maize F test LSD (5%) Site 2 Fallow Indigo F-test LSD (5%) Site 3 Fallow Fallow Indigo F-test LSD (5%) Site 4 Fallow 2 Fallow 3 Indigo F-test LSD (5%)
a
0N PU TU IR IR+PU IR+TU IMR IMR+PU IMR+TU MR MR+PU MR+TU
3.1 4.9 6.1 5.4 5.7 5.4 4.7 5.2 5.8 4.3 4.9 5.1 ** 1.0 5.9 6.6 Ns 18 3.3 5.3 4.3 * .69 3.9 5.4 4.9 ** .612
48 87 101 79 74 78 72 80 83 68 71 73 ** 13 106 146 **
22 35 27 31 27 19 25 32 14 22 24 ** 12 8
0N IR
0N PU IR
48 94 64 ** 10 53 89 62 ** 15
23 11 ns
0N PU IR
17 12 ns
0 N= no applied N, PU = prilled urea, TU = tablet urea, IR = indigo residue, IMR = indigo + mungo residue, MR = maize residue. b Aboveground N uptake including BNF-N.
nificant N loss. There is an urgent need to explore other avenues such as optimizing N applications to DS crops. Carbon management for sustainability of an intensive rice-based cropping system J.K. Ladha and R.K. Shrestha The sustainability of agroecosystems depends on maintaining the reserve of soil organic matter (SOM). The release of nutrients from SOM is largely through microbial activity, so a supply of readily usable C is essential to maintain nutrient release. It is important to differentiate SOM pools into labile (active) and nonlabile (stable) pools because they
play different roles in regulating C and nutrient flows of soil and in maintaining all aspects of soil fertility. A procedure based on the degree of oxidation of organic C by KMnO4 was used to fractionate labile C. That technique allows monitoring the impact of different cropping systems on the change in lability of SOM and hence a measure of sustainability of the system. Experiments in four rice-sweet pepper fields in Batac, Magnuang, Ilocos Norte, Philippines, aimed to q examine the effects of the integration of different NO3 catch crops grown during the DTW transition,
41
Table 10. Apparent mineral N balance (kg ha1) from the 0100 cm soil depth in sweet pepper transition crops for a rice cropping system in Ilocos Norte, Philippines. IRRI, 1999. KCl-extractable soil mineral N (kg ha1) Soil N uptaked (kg ha1) Rice (D) Apparent loss of soil mineral N (kg ha1) (A-B-C-D)
Treatment
DTW transition
Wet seasona
Maximumb Minimumc Transition (A) (B) crope (C)
Site 1 Fallow Fallow Fallow Indigo Indigo Indigo Indigo + mungo Indigo + mungo Indigo + mungo Maize Maize Maize Site 2 Fallow Indigo Site 3 Fallow Fallow Indigo Site 4 Fallow Fallow Indigo
a
0N PU TU IR IR+PU IR+TU IMR IMR+PU IMR+TU MR MR+PU MR+TU 0N IR 0N PU IR 0N PU IR
602 602 602 604 604 604 539 539 539 494 494 494 908 744 506 506 514 313 313 260
43 44 55 51 51 53 40 55 50 49 43 43 216 156 43 76 54 58 78 46
151 151 151 118 118 118 177 177 177 205 126 131
43 78 91 71 66 70 65 72 74 61 64 66 95 131 38 80 57 47 80 58
516 480 456 338 343 337 318 303 299 208 211 209 597 252 425 350 277 208 155 25
0 N = zero N; PU = prilled urea; TU = tablet urea; IR = indigo residue; IMR = indigo + mungo residue; MR = maize residue. bMaximum KCl extractable mineral N in most of the transition crop was at 29 May, whereas in fallow it was at 23 Jun. About 97% of mineral N was NO3-N. c Minimum KCl extractable mineral N in most of the treatment was observed at 30 Sep sampling but some on 30 Oct. About 90% of mineral N was NO3-N. dSoil N uptake = aboveground N uptake-BNF-N. eTransition crops were not planted in fallow plot during DTW transition period.
examine the effect of incorporation of crop residues on total C (CT) and labile soil C (CL) pools, and q determine whether the C L pool and the C management index (CMI) are suitable parameters to explain the C status of a soil. All transition treatments (either fallow or used to grow indigo, an indigo/mungbean intercrop, or maize) followed a DS sweet pepper crop and WS rice crop. Residues were returned to the soil. Nitrogen fertilizer for rice was either not applied (0), applied as PU, or applied as TU. CT measured by combustion and the amount of oxidizing agent (KMnO4) consumed were used to
q
calculate two fractions of organic Clabile C = C oxidized by 333 mM KMnO4 and nonlabile C (CNL) = the remainder. After determining the concentration of CL in the soil, a CMI was calculated. Across all sites and residue treatments, the CL decreased (9%) with cropping and increased (47%) with residue incorporation, but CT remained unchanged (Fig. 9). An apparent balance of the CL pool (difference between initial and final) in the sweet pepper-rice system showed the highest positive balance (2.53 t ha1) when an indigo catch crop was grown and incorporated, either alone or in relay with mungbean, during DTW transition at site 1 (Table 11). Fallow during the DTW transition resulted in a
42
Total carbon (mg g-1) 7.5
Labile carbon (mg g-1) 3.0
7.0
Total C Labile C
2.5
6.5
2.0
Catch crop management in an intensive system can play a role in maintaining soil health, as indicated by improvement in CL. Labile C is sensitive to soil management practices and provides a better measurement of C dynamics in the short- to mediumterm than CT alone, and CMI provides an expression of C pools or changes in C pools.
6.0 9 Nov 25 Jul Soil sampling date 30 Oct
1.5

Managing crop, soil, and water resources for enhanced productivity and sustainability of lowland areas
q
9. Total and labile C as affected by crop and residue management. IRRI, 1999. Table 11. Apparent labile soil C (CL) balance (kg ha1) from the 0- to 25-cm depth during catch crop and rice. IRRI, 1999. Treatment Dry-to-wet transition (DTW) Fallow Ricea Residual CL gain or CL before CL left in loss in catch crop soil after soil (A) rice harvest (B-A) (B) 6107 6107 6107 6107 6107 6107 6107 6107 6107 6107 6107 6107 5971 5462 6140 8708 8809 9142 8659 9037 9184 7850 8815 7359 136 645 33 2601 2702 3035 2552 2930 3077 1743 2708 1252
0N PU TU IR IR+PU IR+TU
Indigo
Indigo + IMR mungbean IMR+PU IMR+TU Maize MR MR+PU MR+TU
q
a 0 N = no N, PU = prilled urea, TU = tablet urea, IR = indigo residue, IMR = indigo/mungbean residue, MR = maize residue.
negative to low positive balance (1.4 to 0.8 t ha1) of CL at all sites. The maize residue mixed with PU produced a CL balance that was equivalent to that of the indigo systems. Maize alone or in combination with TU did not have a highly positive balance. The CMI increased with residue incorporation compared with the conventional practice adopted by farmers of keeping the DTW transition fallow. Indigo alone or relayed with mungbean produced a better CMI than maize. The major effect of residue incorporation on CMI was observed in the soil surface layer at all four sites.
Completed special issue of Field Crops Research, featuring two manuscripts derived from the nutrient + water (N+W) research: one characterizing site conditions and the other examining nutrient requirements of rice in rainfed lowlands. Over 78 locations, yield obtained without applied fertilizer were not closely related to soil test values. The greatest nutrient response was to N, with NPK increasing yields from 2.25 to 4.00 t ha-1 on average. The effect of adding micronutrients was small and PK was of little benefit unless N was added. But the magnitude of the N response varied substantially with the water regime. Substantial yield grains were possible in rainfed systems by application of appropriate nutrients, especially if used in conjunction with cultivars suitably adapted to the target environments. Continued studies into selective effects of water depth in determining germination and survivorship of selected weed species. Continued long-term monitoring of the impact of rotational cropping systems on weed species shifts in direct-seeded rice. Quantitative differences were found in relative abundance of weed species in the WS, which were related to DS farming practices. Weed community composition after weed management was characterized in gogorancah rice (Indonesia) in relation to toposequence and nutrient status. Relative abundance of major weeds was correlated with position on toposequence and soil K status.
43
Analyzed research data from past 4 years from mechanical seeder development project in Jakenan, Indonesia. Along with NARS collaborators, designed and initiated a new onfarm participatory study to assess acceptability of the developed seeder technology. Repeated the experiment on yield constraint analysis in Jakenan for two seasons. Analyzed the panel data for 6 years on risk management practices of farmers from eastern India (Faizabad). The results indicate the importance of income diversification strategies in managing risk at the farm level. Where farmers have highly diversified income strategies, benefits from stabilization of rice yield were quite small. However, in rainfed areas where income diversification is constrained due to environmental conditions or policy factors, stabilization of rice yield can result in major gains, especially to poor farmers who have very limited means for coping with risk. Completed a study of the economic cost of drought in eastern India and farmers coping mechanisms in the event of severe drought. The average production loss in eastern India was estimated to be 8% of the total value of rice and nonrice output. Drought caused a reduction in the output not only of rice but also of nonrice post-rainy season crops such as pulses and oilseeds. Using farm-level data from Orissa, additional costs such as the loss in future production potential due to asset depletion, loss of land, deterioration of human health, and other long-term environmental and social costs were also documented. Completed two papers analyzing the changes in variability of rice area, yield, and production in eastern India. District-level time series data for 71 districts from eastern India covering the period 1969-94 were analyzed. The major findings of the study are: a) productivity growth in eastern India as a whole has led to an increase in production variance, but the relative variability as measured by the coefficient of variation has declined; b) increased yield instability (both absolute and relative) in parts of Bihar and Orissa are the major contributors to increased production instability in eastern India; c) even though there has been a
substantial growth in productivity in eastern Uttar Pradesh and West Bengal due to expansion of irrigation, the yield variability still closely tracts the variability of rainfall; and d) the correlation between yield and area has increased over time in most eastern districts. Conducted stochastic dominance analyses of farm- and district-level yields of modern and traditional varieties. The results indicate that modern varieties are often less risky than the traditional varieties. In several cases, modern varieties had not only a higher average yield but their probability distributions did not intersect with those of traditional varieties. Assessed the economic value of rainfall forecast to rainfed rice farmers in the Philippines. Farm-level panel data showed that the economic value of simple forecasts such as rainfall being above or below average was substantial. Initiated a joint ICAR/IRRI project to study the patterns of changes in rice production systems. This comprehensive project uses a common methodology and cover all states of eastern India.
Germplasm improvement for rainfed lowland rice q In Thailand, IR62558-SRN-17-2-1-B was recommended for release as Chao Surin 72 and in Lao PDR, IR43070-UBN-501-2-1-1-1 was released as Tha Dawk Kam 4. CN1035-61 (IR57540) was recommended for release in semi-deep water (4070 cm) in West Bengal. Six lines were retained in the All India Coordinated Testing Program in 1999. q In G E studies, combined analysis of 47 varieties across 37 environments confirmed repeatable G E interaction across widely dispersed sites in the rainfed lowlands, which was well characterized by pattern analysis and provided interpretable groupings of genotypes and environments. Hydrology is identified as the major discriminating factor and strategies of drought and submergence tolerance as the main feature of adaptation groups. q In hybrid rice studies, RL lines were checked for the CMS maintainer and restorer traits, and these appear to be less common than in irriga-
44
ted lines. Some existing CMS lines and hybrids from the irrigated program were tested for adaptability to RL conditions. Developed and deployed methodologies for eliciting male and female farmers perceptions of useful traits in rice varietal selection in Faizabad District and Raipur, Madhya Pradesh, and Orissa as well as in the Philippines and Vietnam. Conducted phenotypic evaluation of a DH population for root-pulling resistance at Ubon. QTLs for this trait were located on chromosome 4. In addition, identified QTLs associated with root penetration and root thickness on the same chromosome and located QTLs for osmotic adjustment on chromosome 8. Conducted fine mapping of genes controlling submergence tolerance at Kasetsart University using the DH population of CT6241/IR49830 and the RIL population of FR13A/CT6241. The genes controlling submergence tolerance are located on chromosome 9. Started a MAS program for selection of genotypes tolerant of this stress using F2 and BC populations. Found significant QTL environment interaction for constitutive root traits with few QTLs common over populations or runs within populations. Found a large transgressive segregation for deep roots in CT9993/ IR62266 and for thick roots in IR58821/ IR52561. In drought studies, showed that genotypes differed in their ability to perform under drought and recover on rewatering. Those that extended roots to deeper layers, extracted water, and maintained LAI in drought also recovered on rewatering. Early vigor and capacity to develop roots under drought conditions were advantageous. Data from studies of enzyme activity in submergence-tolerant and -intolerant varieties suggest that an active oxygen-scavenging system is involved in alleviating damage caused by submergence and desubmergence. With several lines identified for cross-tolerance to submergence and drought by our breeders, obtained a close correlation between visual score of damage cause by submergence and drought separately imposed in the greenhouse.
With a computer model, compared and analyzed (in terms of root-induced solubilization of P) growth and P uptake by RL cultivars in soil that was either continuously flooded, continuously moist, or flooded then moist. In all moisture regimes, the plants relied on solubilization for the bulk of their P, but solubilization was less effective in re-oxidized, flooded soil than continuously moist soil. Tested a screening technique for external Zn efficiency based on oxygen release from roots by comparing an efficient genotype (Madhukar) with an inefficient one (IR26). Confirmed varietal differences for bacterial rhizosphere associations governing ammonium oxidation. Based on molecular genetic analyses covering a broad spectrum of nitrifying bacteria including N2-fixing bacteria, variety IR63087-1-17 strongly suppressed this target group in its rhizosphere in contrast to Mahsuri and KDML 105. Through field experiments at URRC (Thailand), confirmed that root-associated microbial communities, unlike those of the bulk soil, maintain their richness of functions in biogeochemical C and N cycling during an entire cropping season.
Program outlook
Starting in 2000, the Rainfed Lowland Rice Ecosystem program will be restructured to include research in the flood-prone ecosystem. Ecosystem characterization and resource management activities will be combined into one project, with emphasis on understanding biophysical and socioeconomic constraints, characterizing weed and pest problems, developing management practices for efficient use of nutrients and water, and understanding farmers risk management strategies. The crop physiology and breeding activities for rainfed lowland and flood-prone environments will be in a project for germplasm improvement. This will also include activities on farmer participatory breeding and gender-related studies. Collaborative interaction with NARS partners will continue through the RLRRC and a new collaborative project, the Flood-Prone Research Consortium (FPRC). FPRCs main objective will be to establish collaboration and partnerships among sci-
45
entists and institutes in developing new technology and to create a mechanism for rapid dissemination of technology to flood-prone rice farming communities.
46
Research programs Upland rice ecosystem
GENETIC IMPROVEMENT OF UPLAND RICE 48 Marker-assisted transfer of quantitative trait loci for root depth to IR64 (PBGB) 48 Understanding genetic control of yield loss due to water stress at flowering (APPA, PBGB) 51 Ricea model plant for allelopathy research (APPA, PBGB) 52 Confirmation of allelopathic potential in rice cultivars 52 Progress in identification of rice allelochemicals 52 Identifying quantitative trait loci correlated with allelopathy 53 Candidate gene profiling to accumulate qualitative and quantitative blast resistance (PBGB, EPP) 53 Progress toward a perennial upland rice (PBGB, APPA) 54 Evaluation of perenniality in O. sativa/O. rufipogon F1 progeny 54 Growth and partitioning in perennial rice in flooded and aerobic soils 55 Drought tolerance in annual and perennial Oryza species 56 Participatory varietal selection of upland rice in eastern India (PBGB, SS) 58 Upland rice research consortium (URRC) 58 PROGRESS OF UNREPORTED PROJECTS 60 Improved productivity and sustainability of farming systems in upland rice areas (APPA, SS, SWS) 60 PROGRAM OUTLOOK 60
Upland rice ecosystem
Upland rice is grown on about 17 million ha, mostly in Asia (10.5 million ha). About 3.7 million ha of upland rice are grown in Latin America and 2.8 million ha are grown in Africa. High levels of rural poverty generally characterize upland rice areas. Technological interventions are needed to increase the productivity, profitability, and stability of all upland rice-based systems. The goal of upland rice ecosystem research is to develop technology that will improve the productivity and sustainability of the crop, with the ultimate objective of reducing poverty. Much of the research in the upland rice ecosystem program generates knowledge with application to other rice ecosystems and other upland crops. New tools in the field of functional genomics are being applied to problems of drought and blast disease. A novel approach to improving the sustainability of upland rice systems in sloping areas is the development of a perennial upland rice through wide hybridization with wild rice species. Crop management research seeks to develop soil-specific recommendations for economical fertilizer rates. Because rice competes poorly with weeds in directseeded (DSR) systems, weed control options that are specific to rice are required. Research is under way to understand the underlying biology of weed interference, both in terms of competition for resources and in terms of allelopathic effects. Socioeconomic factors are particularly significant in upland cropping systems. Improved technologies have traditionally been adopted slowly by upland rice farmers and improved understanding of farmers circumstances and logic is required to develop technologies that will be adopted.
The diversity of upland systems requires that technology be developed and adapted through experimentation in the target regions. This is achieved through collaboration with national agricultural research system (NARS) partners in the Upland Rice Research Consortium.
Genetic improvement of upland rice

IRRI has focused germplasm improvement on the genetic control of plant response to drought and disease. Different breeding approaches are evaluated and high priority placed on sharing information and germplasm with NARS partners. Novel sources of genetic variation are used to develop varieties that can reduce weed pressure and reduce erosion. Grain quality and plant acceptability to the farmer are particularly important in many smallholder systems where upland rice is grown. Participatory approaches to varietal evaluation are used to address that issue. Marker-assisted transfer of quantitative trait loci for root depth to IR64 B. Courtois, L. Shen, K. McNally, S. Robin, and Z. Li Drought is a main abiotic constraint in upland rice production. A deep root system contributes to the maintenance of crop water status during stress. In 1997, we performed quantitative trait loci (QTL) analysis for root parameters on a double haploid (DH) population of rice derived from the cross IR64/Azucena and started a marker-assisted backcross program to transfer four of the QTLs detected on chromosomes 1, 2, 7, and 9 from selected DH lines to IR64. The selection scheme is outlined in
48
Table 1. We selected plants with the proper allelic combination at the target regions based strictly on the genotype up to BC2F2. Microsatellite and restriction fragment length polymorphism (RFLP) markers flanking the targeted regions used for the genotyping included q RZ19, RG690, RZ730, and RZ801 for the target region of chromosome 1; q RM29, RG171, RG157, and RZ318 for the target region of chromosome 2; q RM234, CDO418, RZ978, CDO38, and RM248 for the target region of chromosome 7; and q RZ228 and RZ12, replaced by RM201 and RM242 after BC2 generation, for the target region of chromosome 9. We evaluated the proportion of remaining alleles from Azucena in the nontarget areas of the BC2F2 plants. The theoretical frequency of Azucena alleles in BC3 is 3.1% or less. The whole genome survey with microsatellite markers showed that the frequency of the Azucena alleles in the nontarget regions ranged from 0.0 to 6.7% with an average of 2.4%. The selected BC3F2 plants are regarded as near-isogenic lines (NILs) for the target QTLs. We evaluated the root system of 30 BC3F3 NILs selected to represent various recombinant genotypes in the target regions and compared it with that of IR64 in a replicated greenhouse experiment. Three of the lines carried Azucena alleles at the target region of chromosome 1 (target 1), five carried target 2, seven carried target 7, eight carried both target 1 and target 7, and six carried target 9. The design was an alpha-lattice with six replications, five blocks per replication, and the check IR64 replicated in all blocks. The plants were grown in tubes 1 m long and 0.2 m diameter filled uniformly with sandy loam soil. The plants were watered three times a week. The root measurements were made 43 d after sowing. Maximum root length (point reached by the longest nodal root), total root weight in the column, and deep root weight (root weight below 30 cm) were determined. The main characteristics of the NIL root system are presented in Table 2. Of the three NILs carrying target 1, one had significantly improved deep root weight over IR64. Three of the seven NILs carrying target 7 as well as three of the eight NILs carrying both targets 1 and 7 showed significantly improved root mass at soil depths below 30 cm. This confirms
Table 1. Sequence of operations in the marker-aided selection scheme for introgression of QTL for root depth using four DH lines derived from the cross IR64/Azucena as donors and IR64 as recurrent parent. IRRI, 1999.
Plants selected for further backcrossing or selfing (no.)
18
15 50 120 BC3F1 seeds
Plants with desired genotypes (no.)
33
Plants genotyped (no.)
120
BC1F1 seeds
BC2F1 seeds
BC3F2 seeds Yes Yes Homozygous
75.0% IR64 25.0% Azucena 87.5% IR64 12.5% Azucena 93.7% IR64 6.3% Azucena 96.9% IR64 3.1% Azucena 96.9% IR64 3.1% Azucena Heterozygous Yes Heterozygous Yes Yes Yes Whole-genome survey 49
a
Theoretical status of the product
Selection Selection for for primary secondary target target
Type of plant selected
No
No
Yes
DS = dry season, WS = wet season.
No
1996 WS
No
No
1997 WS
Hybridization DH lines/IR64 BC1
Backcross generation
BC2
Selfing
BC3
1998 WS
Seasona
1996 DS
1997 DS
1998 DS
BC3F3 seeds
Product
F1 seeds
74 312
32 66
32 58
Table 2. Root characteristicsa and other important agronomic traits of the NILs of IR64 evaluated at IRRI, 1999.a Line Target MRL (cm) 69.1 75.9 73.5 70.7 72.1 65.0 70.4 73.2 72.6 64.3 70.8 78.5* 67.9 65.5 68.2 65.6 66.2 71.6 74.9 75.6 69.7 66.8 67.9 68.5 88.7** 71.5 72.8 79.7* 77.9* 88.1** TRW (kg) 1.392 1.699~* 1.645 1.669~* 1.230 1.173 1.333 1.424 1.656 1.510 1.593 1.572 0.988* 1.288 1.897** 1.471 1.411 1.365 1.807** 1.616 1.685~* 1.353 1.467 1.581 1.585 1.271 1.212 1.160 1.430 1.306 DRW (kg) 0.175 0.146 0.279* 0.233 0.154 0.117 0.127 0.228 0.194 0.116 0.278* 0.260* 0.100 0.103 0.265* 0.082* 0.222 0.185 0.263* 0.308** 0.302** 0.150 0.135 0.106 0.225 0.187 0.143 0.148 0.153 0.187 HGT (cm) 107.4 107.7 150.2** 112.0* 110.8 105.3 101.6** . 100.8* 101.9* 122.7** 119.5** 99.0** 94.9** 137.1** 101.9* 120.9** 124.3** 148.4** 147.4** 123.8** 142.1** 165.8** 155.6** 109.0 107.0 109.8 105.8 107.4 105.2 HMG TIL
IR64 IR74392-108-6 IR74392-118-4 IR74392-135-1 IR74392-201-14 IR74399-204-10 IR74401-215-5 IR74401-215-18 IR74401-216-7 IR74405-711-1 IR74405-720-7 IR74405-720-12 IR74409-730-8 IR74409-730-9 IR74409-730-10 IR74409-734-4 IR74409-735-2 IR74409-735-12 IR74409-736-11 IR74409-737-5 IR74409-737-12 IR74409-738-11 IR74409-739-4 IR74409-739-7 IR74418-910-2 IR74418-910-3 IR74418-910-12 IR74418-913-7 IR74419-921-1 IR74419-921-8
a
1 1 1 2 2 2 2 2 7 7 7 7 7 7 7 1+7 1+7 1+7 1+7 1+7 1+7 1+7 1+7 9 9 9 9 9 9
H H H OT H H H H H S S H H H H S S H H S H H H H H H H H H
26.3 21.9** 23.4* 27.4 19.4** 20.0** 23.3* . 23.6* 24.6 25.1 21.0** 20.9** 20.3** 24.5 24.7 21.4** 20.9** 23.9 20.2** 23.0* 22.2** 23.8 24.9 23.3* 22.0** 23.3* 20.6** 21.6** 21.7**
MRL = maximum root length, TRW = total root weight, DRW = root weight below 30 cm, HGT = plant height, HMG = homogeneity for plant height (H = homogeneous, S = segregating, OT = off-types), TIL = no. of tillers. ~*, *, ** = significantly different from IR64 at 10%, 5%, and 1% thresholds, respectively.
the existence of both QTLs and narrows their locations. Five NILs carrying target 2 were phenotyped, but none had a root phenotype significantly different from IR64. A re-analysis of the initial data with composite interval mapping technique showed that two QTLs with opposite effects existed in this area, even though the initial analysis, which was conducted using regression on flanking markers, detected only one. The tested lines did not include a root-increasing recombinant for this segment. Four of the six NILs carrying target 9 had significantly improved maximum root length. Their other root characteristics were similar to those of IR64. We looked at the influence of these segments on other traits in a field experiment using an alpha-lattice design with four replications. Most introgres-
sions seem to decrease tiller number. Improved deep-root mass (over IR64) seems to be linked with the presence of the tall allele at the sd-1 locus or heterozygosity at this locus. sd-1 is a recessive semidwarfism gene located on chromosome 1 between RZ730 and RZ801. However, some tall plants did not display improved root mass. This could indicate tight linkage rather than pleiotropy. The rate of QTL transfer was not high in this study. There are several possible reasons for that. The first relates to the quality of the initial QTL analysisrisk of type I error on the QTL existence, uncertainty on QTL position, or existence of nonallelic interactions. The second is that the target QTL can be lost during the successive backcrosses through double crossovers between markers flank-
50
ing the QTLs. The size of some of the intervals we were manipulating does not rule out this possibility. Moreover, the fact that the root measurement method is destructive and prevents phenotyping during the NIL production process does not allow elimination of such recombinants. Our results demonstrated that it is possible to transfer QTLs for root system characteristics in rice and record significant improvement in the root system. The lines with improved root systems are being tested in the field. Understanding genetic control of yield loss due to water stress at flowering R. Lafitte and B. Courtois The reproductive stage is the period when water shortage most affects grain yield. Our objective was to document genetic variation in tolerance for water deficit during flowering and grain filling among DH lines from a japonica/indica cross, to identify genetic markers associated with performance, and to identify secondary traits that cosegregate with yield or yield components. Evaluating drought tolerance at flowering is complicated by the fact that genotypes flower at different times. Three approaches have addressed that problem: sprinkler irrigation with staggered planting dates (1995), furrow irrigation with repeated mild stress periods that spanned flowering for all entries (1998), and individual plot drip irrigation (1999). The upland japonica cultivar Azucena and the lowland indica cultivar IR64 were sown in 1999 DS along with DH lines from the Azucena/IR64 mapping population. The stress treatment with staggered planting consisted of no irrigation for 10 d, with the stress starting at anthesis of most of the lines followed by 1 h of irrigation every 3 d until the end of the season. The control was irrigated 1 h d1 by sprinkler (application of about 10 mm daily). Irrigation was applied twice a week in the future irrigation experiment except for two periods near flowering (5264 and 70 84 d after sowing [DAS]) in the stress treatment. In the individual drip irrigation experiment, water was withheld from the stress plots 10 d before 50% flowering of the control plots, and was reapplied after 14 d. The staggered sowing in 1995 was not completely effective in inducing synchronous flower-
ing. Some of the lines started to flower well into the stress period. All lines that flowered more than 15 d after the start of the stress were fully sterile, as were some that flowered 10 d after start of stress. Lines that flowered late were eliminated, and the population used in the analysis comprised 85 DH lines. Genetic differences were observed for most traits. As intended, the stress primarily affected sterility, yield, biomass after stress, and harvest index. The objective of having two stress periods in 1998 was to ensure that all lines would experience stress during the sensitive period. Water deficit had little effect on tiller number or spikelets per panicle and primarily affected the sterility of panicles and spikelets, and final grain weight. Genetic differences were detected for these yield components and for most other traits. Data from the first two experiments were subjected to QTL analysis, using the genetic map available for the population. Means for each entry were used to map QTLs for each trait using QTL Mapper 1.0. The QTLs identified were compared with those found when the population was examined in a lowland ecosystem, under vegetative stress in other studies, and with other published data on rice QTLs for leaf traits and root penetration. They were also compared with QTLs for osmotic adjustment identified in other rice populations. QTLs were identified for almost all traits. Some were common between stress and control treatments, particularly in 1995. When common QTLs were observed, their effects were in the same direction in the two environments, indicating minimal crossover interaction for those QTLs. A number of the QTLs identified for yield components were also identified when the population was grown in fully irrigated lowland in other studies. On the other hand, some QTLs were not common across the stress and control plots. These loci represent variation that is water treatment-specific. Common QTLs were identified across the two experiments for plant height, panicle length, tillering, and sterility in the region of chromosome 1 that contains the sd1 locus and confers the semidwarf habit. The same interval has been shown to be associated with root parameters and also with field scores of leaf rolling and drying under water stress. There were several areas where QTLs were associated in repulsion phase (e.g,. for thousand-grain weight under stress on chromosome 3 and for yield
51
under stress on chromosome 6). QTLs for some secondary traits were found to cosegregate with intervals associated with yield or yield components. For many traits, both parents contributed alleles with positive effects. Epistasis seemed to explain a nonnegligible part of the variability for most traits. Our results show the potential of molecular markers to identify chromosome regions that control the response of yield to water deficit at flowering. Further studies under other drought conditions and with other populations are needed to evaluate the repeatability of these QTLs. It should be possible to identify QTLs for performance under stress that are not linked with biomass, and introduce them through marker-aided selection into varieties with good yield potential. Ricea model plant for allelopathy research M. Olofsdotter, L. Bach-Jensen, and B. Courtois Attempts to increase the competitive ability of rice have had limited success, in part because of limited understanding of the components and mechanisms of competition. The importance of chemical interference, including allelopathy, in crop competition has often been debated. The goals for research on allelopathy have included q laboratory, greenhouse, and field studies to illustrate the effect of released allelochemicals; q isolation, identification, and characterization of allelochemicals; q establishment of correlation between growth inhibition and allelochemicals in controlled experiments; q genetic mapping of QTLs correlated with allelopathy; and q development of cultivars with confirmed allelopathic potential. IRRIs research on allelopathy in rice has focused on these goals and has been conducted in collaboration with several institutions. Rice has become a model plant for progress toward utilization of allelopathy.
CONFIRMATION OF ALLELOPATHIC POTENTIAL IN RICE CULTIVARS
Field evaluation of weed suppressive activity of rice cultivars was conducted against Echinochloa crusgalli in irrigated lowland fields and against Trianthema portulacastrum and E. colona in upland fields. Weeds and rice were direct seeded the same day in all experiments. During establishment and early growth of rice and weeds, a range of variables was measurede.g., tillering, height, and density of rice and weed dry weight 8 wk after seeding. Two N levels were included in the experiments with E. colona to test if N availability interacted with weed suppressive ability. The results indicated cultivar differences in the ability to suppress weeds in upland and lowland irrigated fields. Rice height was the single most important factor influencing weed suppression, accounting for 15 50% of weed reduction. However, looking only at the most weed suppressive cultivars, rice height was not an important factor, meaning that allelopathy in some cases plays a more significant role than any of the other features measured. The number of weeds in the field was not affected by rice cultivars but weed height and biomass were strongly dependent on cultivar. The experiments also indicated that maximum weed suppression is obtained in cultivars that have both allelopathy and competitive traits.
PROGRESS IN IDENTIFICATION OF RICE ALLELOCHEMICALS
Phenolic acids have frequently been described as allelochemicals, but it is unlikely that they are responsible for allelopathic potential in rice. In collaboration with the Natural Compound Utilization Laboratory, USA, isolation and identification of rice allelochemicals by use of bioassay-guided isolation have started. Several putative allelochemicals have been isolated from roots of Taichung Native 1. Ten of at least 23 compounds have been identified, but none of those alone explains the allelopathic effect seen in the laboratory and the field. If the allelochemicals involved are identified, and if the allelochemicals correspond to genes that are already sequenced, the genes will be used as candidate probes to detect whether they cosegregate with QTLs detected in mapping populations.
52
IDENTIFYING QUANTITATIVE TRAIT LOCI CORRELATED WITH ALLELOPATHY
A population from the cross IAC165/Co 39 was used for mapping the genes involved in allelopathy. The population of 144 recombinant inbred lines was derived through single-seed descent. A molecular map of this population contains 184 markers, both microsatellite and RFLPs. The length of the genetic map is close to 2000 cM. This population was evaluated both for allelopathic potential and also for characteristics related to drought tolerance. The upland cultivar IAC165 showed strong and consistent allelopathic activity against E. crus-galli in a study using a relay seeding technique, while the irrigated cultivar Co 39 was weakly allelopathic. IAC165 is a highly allelopathic cultivar from Brazil. Laboratory screening showed that a large proportion of the Brazilian upland rice germplasm is allelopathic. However, weed selectivity, autotoxicity, and residual effects from Brazilian rice germplasm differ from those for Asian germplasm. Candidate gene profiling to accumulate qualitative and quantitative blast resistance J.L. Wu,12 P.K. Sinha,13 K.L. Zhang,12 B. Courtois, and H. Leung Genetic resistance to blast in upland rice is often short-lived due to the disease-conducive environment in the upland ecosystem and the high variability exhibited by the pathogen. Previous work has demonstrated that both qualitative and quantitative
resistances are important components of durable resistance, but the masking effect of major resistance genes has made it difficult to combine complementary resistance mechanisms. We have developed an advanced backcross population (BC3F3) using Moroberekan as the donor of qualitative and quantitative blast resistance and Vandana as the recurrent parent with good quality traits and adaptation to eastern India. Eighty BC3F3 lines were evaluated at seedling stage in the greenhouse and in the blast nursery at IRRI. When inoculated with a single pathogen isolate in the greenhouse, the BC3F3 lines segregated in a bimodal pattern, suggesting the presence of major resistance gene or genes. In the blast nursery, the BC3F3 and BC3F4 populations showed different levels of resistance among the lines against a diverse pathogen population, indicating successful introgression of resistance genes from Moroberekan (Fig. 1). Molecular markers associated with both qualitative and quantitative resistance were identified by use of 26 candidate resistance genes from rice, maize, and barley as probes to analyze DNA of the 80 BC3F3 lines by Southern hybridization. The candidate genes included 1) major resistance genes with nucleotide-binding site (NBS) and leucine-rich repeats (LRR), which are conserved domains involved in pathogen recognition, and 2) disease-response genes involved in general defense mechanisms. Five NBS-LRR sequences and three diseaseresponse genes showed significant contribution to disease resistance (Table 3).
Table 3. Candidate resistance genes associated with resistance observed in BC3F3 lines derived from Vandana/Moroberekan. IRRI, 1999. Gene categorya (specific name) NBS-LRR (Rxo-6.31) NBS-LRR (Rp1) NBS-LRR PR protein (PR5 thaumatin) PR protein PR protein NBS-LRR NBS-LRR Chromosomeb 9 1,2 11 Fc 16.01 22.54 10.16 11.49 11.25 5.83 5.46 4.7
Marker CG54a CG36b CG52 CG7b CG16a CG1 R2d R6e

a
Source Maize Maize Maize Barley Barley Barley Rice Rice
P 0.0001 0.0001 0.0001 0.0011 0.0013 0.0181 0.022 0.0333
11 11
PR = pathogenesis-related. bChromosomal locations determined using a recombinant inbred mapping population derived from Gumei 2/Zhong 156. cF values from analysis of variance using the general linear model. A significant F value supports the hypothesis that the gene marker has an effect on reducing disease severity as measured by percent diseased leaf area.
53
12 8 4 0
Moroberekan
16
BC3F3 BC3F4
Vandana
Lines (no.) 20
Progress toward a perennial upland rice E. Sacks, K. McNally, S. Kubota, L. Liu, R. Lafitte, and O. Ito Upland rice is the main subsistence crop in the hilly areas of Southeast Asia, providing food security via low (~1 t1 ha) but stable yields using little or no purchased inputs. It is grown in a traditional shifting cultivation system that has long rotations and is relatively sustainable, but there is a trend toward shorter rotations and more intensive use of marginal lands because of increasing population pressure. The resulting soil degradation directly limits upland rice yields and indirectly limits water availability and farm productivity in lowlands because of siltation of reservoirs. A perennial upland rice is one of several possible interventions to address the problem of erosion. Efforts to develop a perennial upland rice focus on combining genes for perenniality and drought tolerance from O. rufipogon and O. longistaminata, with genes for agronomic acceptability from upland cultivars of O. sativa. Work in 1999 included more than 500 cross combinations, initial field testing of more than 1,500 O. sativa/O. rufipogon F1 progeny (32 families), and field testing of more than 7,000 O. sativa/O. longistaminata progeny (98 S1 families). A backcross intermating scheme for the O. sativa/O. longistaminata population was established. Mapping populations were developed from O. sativa/O. longistaminata crosses, and were evaluated for rhizome formation. Production of the genetic map is under way to allow identification of genetic markers for rhizome formation.
EVALUATION OF PERENNIALITY IN O. SATIVA/O. RUFIPOGON F1 PROGENY
10
15
20 25 30 35 40 45 Diseased leaf area (%)
50
55
1. Percent disease leaf area (DLA) of BC3F3 and BC3F4 lines derived from Vandana/Moroberekan. The plants were exposed to natural pathogen popul1ation in the blast nursery at IRRI. Note the normal distribution of the phenotypes. IRRI, 1999.
To determine whether phenotypic expression of resistance is associated with specific profiles of candidate genes, the backcross lines were clustered based on the similarity of their DNA banding patterns produced by candidate gene probing. About 20% of the top performing lines were found in three distinct clusters in the dendrogram. Genotypes in one cluster were associated with a high level of resistance, whereas genotypes in the other two clusters showed partial resistance. In contrast, the same high-performing lines were scattered in a dendrogram generated by microsatellite markers, which are presumably not associated with disease resistance except through linkage. The results suggest that candidate gene profiling is indicative of functional resistance and may provide an effective means of combining both qualitative and quantitative resistance in advanced breeding lines. Because the backcross lines have a substantial amount (theoretically 87%) of the Vandana background, suitable commercial varieties could be extracted directly from the backcross lines, or produced by an additional cycle of selection and intercrossing. We are working to extend the phenotypic analysis to correlate seedling resistance with neck blast resistance in upland sites with high disease pressure.
Variation for perenniality and drought tolerance was observed among and within 10 accessions of O. rufipogon in a previous study (IRRI Program Report for 1997). Selections from that experiment were crossed with upland rice cultivars to produce F1 seed. We evaluated perenniality and drought tolerance of the F1s during 1998-99. A total of 51 O. sativa/O. rufipogon F1 combinations were obtained from six accessions of O. rufipogon crossed with one or more of 14 annual upland rice cultivars. The upland parents included
54
indica and japonica varieties from Asia, Latin America, and Africa. We evaluated 501 F1 individuals and their annual parents in an IRRI upland field. Plots were established in Jul 1998 as two rows of plants in each of 10 concrete-lined beds. To evaluate drought tolerance, no supplemental irrigation was provided from 15 Mar 1999 (during DS) until the end of the experiment. Due to above-average rainfall for March, April, and May, entries were exposed to modest drought stress. Entries were scored for survival 30 Jun 1999 and survivors were rated for vigor. Survival of the F1 individual O. sativa parent and O. rufipogon parent, and their interaction, were assessed using SAS procedures CATMOD and FREQ. None of the annual controls had survived by the end of the experiment, indicating that the O. sativa parents were not perennial. Ninety-five F1 plants (~19%) survived, indicating that genes for perenniality from O. rufipogon were expressed in some F1s. Moreover, the perenniality observed in the F1s and derived from the O. rufipogon parents must have been the result of nonrecessive gene action. Interactions among the O. sativa and O. rufipogon parents and the main effect of O. sativa parent were not significant. Although the O. sativa parent and parental combination did not have a detectable effect on survival, we interpret these data cautiously in light of the incomplete crossing design. The effect of O. rufipogon parent was highly significant (P <0.001 for chi-square of 115 with 11 df). Survival of F1s ranged from 0% for parental accession 106133 (plant 9) to 71% for accession 105832 (plant 1) (Table 4). The second, third, and fourth ranked O. rufipogon parents for survival were from accession 106138 and the fifth ranked parent was from accession 106114. Those two accessions were the best in a previous report of perenniality in wild species (IRRI Program Report for 1997). Our results are encouraging. Recombination and selection should allow identification of progeny that are perennial and agronomically desirable. F1s that had the highest vigor scores at the end of the study are currently being intercrossed using nested and factorial designs.
Table 4. Variation in survival and vigor associated with O. rufipogon parents of F1 hybrids of upland rice cultivars after 1 year of cultivation and 2.5 mo of moderate drought stress. IRRI, 1999. Vigorb Accession no.-plant no.a 105832-1 105832-11 105832-18 106114-11 106133-9 106138-4 106138-7 106138-18 106144-2 106144-18 106340-1 106340-2 Hybrids Survival Rank (no.) (%) 73 26 7 8 54 29 92 10 42 26 93 41 10 8 71 25 0 66 35 70 7 12 11 12 9 10 1 5 12 3 4 2 11 7 8 6 Mean Std dev 5.6 7.0 5.8 4.5 3.8 6.0 3.4 4.7 6.3 4.2 5.4 2.6 0.0 1.3 3.5 2.1 2.3 2.5 2.3 2.1 1.5 4.5
a Identification of O. rufipogon parents of F1 hybrids. bData for survivors only. Based on a 1-9 scale, where 1 is best.
GROWTH AND PARTITIONING IN PERENNIAL RICE IN FLOODED AND AEROBIC SOILS
An experiment determined how different soil conditions (flooded vs moist-aerobic) and daylength affected growth and development in O. longistaminata, O. sativa, and their interspecific progeny. Two O. longistaminata accessions (WL02-2 and SL313-13), two O. sativa cultivars (BS125 and UPLRi-5), two BCLF2, and two BCLF1 intercrosses were studied. Culm emergence was recorded each day, and flowering date was recorded for each culm. Five plants of each genotype were collected at flowering, separated by plant part (rhizome, aboveground shoots, and roots), freeze-dried, and weighed. The upland cultivars and perennial genotypes grew better in flooded soil than in moist aerobic soil, producing more culms, panicles, total plant dry weight, aboveground shoot dry weight, and root dry weight in flooded soil (Table 5). The perennial genotypes had greater rhizome dry weight in flooded soil than in aerobic soil. The data (Table 5) demonstrate that variation in tolerance for aerobic soils exists in O. sativa and O. longistaminata. For the perennial genotypes, the proportion of rhizome dry weight to total dry weight was larger for aerobic soil than for flooded soil, primarily because plants maintained the production of rhizomes better than the production of aboveground parts. The maintenance of rhizome mass under moisture
55
Table 5. Dry weight of each plant part, fraction of dry matter in rhizome, culm and panicle production at flowering for annual and perennial rice grown in flooded and moist aerobic soil. IRRI, 1999. Distribution Total ratio in culms Panicles rhizome (%) (no. plant1) (no. plant1) 0.0 0.0 12.0 17.0 3.4 4.0 4.4 2.6 0.0 0.0 26.6 30.2 5.7 7.8 5.9 7.9 ** ** 8.1 5.9 26.7 37.9 83.4 43.3 49.5 37.0 2.7 2.7 8.9 10.2 38.3 18.7 19.8 8.4 ** ** 8.1 5.9 8.6 9.5 26.0 15.7 26.0 22.8 2.7 2.7 2.1 1.0 3.9 4.8 7.0 1.2 ** **
Soil
Entry
Total (g plant1) 86.3 42.8 135.3 76.2 127.0 56.8 129.4 66.0 21.5 22.2 43.1 37.4 27.9 13.0 32.2 18.0 ** **
Top (g plant1) 78.4 39.5 107.3 58.3 105.6 51.1 109.3 61.4 18.0 18.4 25.1 22.7 21.3 10.2 23.4 14.8 ** **
Rhizome (g plant1) 0.0 0.0 16.1 12.7 4.3 2.4 5.6 1.7 0.0 0.0 11.7 10.3 1.6 1.0 1.9 1.3 ** **
Root (g plant1) 4.9 3.2 5.5 3.1 7.8 2.2 6.4 2.0 2.0 2.1 3.3 2.9 3.8 0.8 3.6 1.1 ** **
Flooded
BS125 UPLRi-5 WL02-2 SL313-13 IR73352.30-4 IR73354.3-2 B52.19/54.36-1 B54.17/6-2 BS125 UPLRi-5 WL02-2 SL313-13 IR73352.30-4 IR73354.3-2 B52.19/54.36-1 B54.17/6-2 Water Progenies
Moist, aerobic
Significance
stress should allow long-term survival in difficult environments. The perennial genotypes flowered over a longer time period (4560 d) than the cultivars (~20 d). The ability to flower over a long period of time may be a useful trait for developing cultivars that have stable yield because some panicles will avoid water stress at critical stages. The effect of flowering duration on yield stability and yield potential should be investigated further because subsistence farmers typically prefer yield stability over yield potential. The number of flowering culms produced by the O. longistaminata genotypes was similar to that of the cultivars even though the wild species produced many more culms. The plants grown in aerobic soil showed patterns of shoot development and flowering similar to the plants in flooded soil. Because of poor seed set in the perennial plants, it was not possible to determine the implications of competition between vegetative and reproductive sinks during grain filling.
DROUGHT TOLERANCE IN ANNUAL AND PERENNIAL ORYZA SPECIES
Perennial upland rice will have to survive a dry season (DS) and also withstand dry periods during the rainy season. Fifteen genotypes, including upland and lowland varieties of O. sativa and accessions of O. rufipogon and O. longistaminata, were evaluated in long tubes in the greenhouse during DS. Treatments included well-watered, medium water stress, and severe water stress. Stomatal resistance and root bleeding differences were detected early in the water-stress period. The correlation between stomatal resistance and biomass production, root depth, and root dry weight was significant. A close correlation was also found between stomatal resistance and visual drought screening score (r=0.84**) and stomatal resistance and leaf relative water content (RWC) (r=0.88**), traits that are often used to assess drought tolerance under severe water deficit.
56
Table 6. Stomatal resistance, leaf relative water content, and drought screening score based on green leaf retention, root dry weight, and root dry weight distribution in control (Trt. I) and after 25 d water deficit (Trt. III). IRRI, 1999.
a O.r. = Oryza rufipogon, O.l. = Oryza longistaminata. bStomatal resistance (s cm1). cLeaf relative water content. dDrought screening score where 0 = no stress symptoms and 5 = half of leaves fully dried. eRoot dry weight.
Significant differences in drought tolerance (measured as biomass production) were found among the varieties and among accessions of the wild species. All O. longistaminata accessions and most O. rufipogon accessions were more tolerant than O. sativa, but only one O. rufipogon accession was as tolerant as the O. longistaminata entries. The wild species produced relatively more biomass under stress, had deeper root systems and higher leaf RWC and stomatal conductance (Table 6). Some wild accessions had much greater membrane stability, root sap production, and osmotic adjustment than the cultivated varieties. Contrasting strategies for productivity and survival were apparent among the genotypes tested. The superiority of the tolerant O. rufipogon accession was confirmed in field evaluations of O. sativa/O. rufipogon progeny. Among all the traits tested, root dry weight distribution with depth appeared to be particularly important (Table 6). This trait was closely correlated with other traits, such as leaf RWC, stomatal resistance, shoot biomass production, and root sap production. Under water deficit, varieties with dramatically reduced shoot biomass production also suffered reductions in root growth. No decline was observed in root growth in varieties that maintained better shoot biomass production. Furthermore, the root systems of tolerant varieties showed a redistribution of dry matter to deeper soil levels under stress. Differences in root growth and distribution between drought-tolerant and -susceptible rice varieties may be due to abscisic acid analog (ABA) or other chemical signals. Although root characteristics may be important for drought tolerance, it is difficult to analyze root dry weight and distribution in large-scale screening. Our results indicate that stomatal resistance under moderate water deficit may reflect root distribution. Shoot traits such as leaf membrane stability and osmotic adjustment are important for survival under more severe stress and for recovery ability. Plants that combine superior root and shoot traits will be sought among the interspecific progeny being produced.
60100 cm 3060 cm 030 cm DSSd Trt. III RWCc Trt. III Rsb Trt. III Varietya Total RDWTe (g) Trt. I
Root distribution (%) in Trt. III
RDWT (%) Root distribution (%) in Trt. I in Trt. III 30100 cm 030 cm 3060 cm 60100 cm RDWT (%) of Trt. I Trt. III
Wab 56-50 Salumpikit IR65907-216-1-b Azucena IRAT212 IR60080-46a IR64 IRAT104 106138-18 O.r. 106407-18 O.r. 106144-18 O.r. 105832-11 O.r. Wol 2-2 O.l. Sl 313-13 O.l.
8.5 20.0 9.7 12.7 10.4 8.0 21.3 6.8 9.0 18.5 15.3 20.8 13.3 5.5
80.5 42.9 78.1 85.6 80.3 84.6 51.0 84.0 87.5 70.2 86.0 40.6 74.0 84.7
2.0 5.0 2.6 4.0 2.8 2.8 4.8 2.4 1.6 4.0 3.0 5.6 2.0 1.6
2.0+/0.35 7.3+/1.64 4.3+/0.72 6.6+/0.96 1.6+/0.45 3.0+/0.89 2.3+/0.47 4.2+/0.15 3.3+/0.61 13.5+/2.9 3.4+/1.3 4.5+/0.66 19.6+/9.4 15.6+/3.4
150 41 114 50 150 100 96 102 94 59 26 56 93 96
21 11 29 15 12 22 1 37 5 13 1 5 10 18
89 91 82 82 95 83 97 78 99 85 98 98 86 87
11 8 18 15 4 17 3 20 1 11 2 2 14 12
0.5 0.9 0.4 3.3 0.3 0.2 0.4 2.6 0 3.7 0 0.3 0.4 1.0
79 89 71 85 88 78 99 63 95 87 99 95 90 82
16 11 22 11 10 14 1 21 5 11 1 4 9 18
4.9 0.6 6.7 4.1 2.1 7.5 0 16.4 0.4 1.6 0 1.1 0.7 0.4
57
Participatory varietal selection of upland rice in eastern India B. Courtois, D. Chaudhary, G.N. Mishra,13 N.P. Mandal,13 K. McAllister, S. Pandey, T. Paris, K. Prasad,13 A.T. Roy,26 R.K. Sahu,24 V.N. Sahu,24 S. Sarkarung, S.K. Sharma,24 A. Singh,11 H.N. Singh,11 O.N. Singh,11 N.K. Singh,27 R.K. Singh,13 S. Singh,11 P.K. Sinha,13 B.V.S. Sisioodia,11 and R. Takhur27 A participatory plant breeding project for rainfed rice was developed by plant breeding institutes in eastern India in collaboration with IRRI to test the hypothesis that farmer participation in one or more steps of the breeding process can substantially increase the suitability of modern rice varieties for rainfed environments. The project includes the upland and rainfed lowland ecosystems and encompasses plant breeding and social science. Plant breeders and social scientists work with farmers in a participatory mode. Field trials represented various hydrological situations in three to five villages in areas where upland and rainfed lowland rice is grown. Sets of 1525 varieties were tested both in farmers fields and onstation in 1997 and 1998 and ranked by farmers and breeders. The quality of their agreement was measured by the Kendall coefficient of concordance for the agreement within group (W=1.0 in case of perfect agreement) and by rank correlation coefficients for the agreement between groups. Coefficients of concordance among farmers were generally significant, showing that farmers rankings were not randomly allocated, even though the agreement within the group was not always strong. This indicates that socioeconomic diversity among farmers, which we assumed would have generated differences in selection criteria, was lower than expected. Concordance among breeders was high, except at one site, but a limited number of scorers was involved and it was seldom significant. In about two-thirds of the cases, there was a good agreement among farmers and breeders rankings. The consensus was particularly strong when severe constraints induced contrasting behavior in the genotypes. Genotype-by-environment (G E) interactions for grain yield were evaluated through various methods, showing more effect of G E interactions at some sites than at others. Crossover interaction,
which changes varietal ranks from site to site, and is the most damaging G E interaction component for breeders, represented a limited part of the yearly G E interactions at all sites. Farmers in several villages identified varieties of interest among the ones they evaluated and decided to test them in their own fields to confirm their first impression. This indicates that access to new varieties can also be a constraint to adoption. Postharvest traits such as cooking quality and taste are additional factors that influence the probability of a varietys adoption. A sensory evaluation of varieties was conducted with farmers in Korahar, Bihar. Twenty-four farmers (12 women and 12 men) evaluated 15 upland varieties as raw rice and parboiled rice for milled rice appearance, cooked rice appearance, color, odor, texture, stickiness, taste, and overall acceptability. The rice samples were milled and cooked by the women farmers following their ordinary practices. One variety recorded good results with both modes of preparation (Table 7). The preferences of women and men farmers did not differ significantly. Most of the traits ranked by farmers were highly correlated with overall acceptability, indicating that it is possible to reduce the number of traits scored by farmers. The rankings based on preferences were poorly correlated with the rankings of the varieties for various physicochemical characteristics measured in the laboratory. The correlation between ranking based on preferences and ranking based on field performance of the varieties was positive for raw rice and negative for parboiled rice. The farmers trade-off between field performance and grain quality is therefore important to assess, at least for parboiled rice. The results of the first sensory evaluation will be used to simplify the methodology, extend it to other villages involved in the project, and improve varietal evaluation in the formal breeding process. Upland Rice Research Consortium (URRC) R. Lafitte Workplan meetings for 1999 were held at each URRC site. The annual Steering and Technical Committee Meeting was held at Ranchi, India, 1114 Oct. Sixteen scientists from India, 23 scientists from other NARS, and 12 IRRI scientists at-
58
Table 7. Sum of scoresa given by 24 farmers for cooking quality characteristics of upland rice varieties, Korahar, Bihar, India, 1998. Cooked rice appearance Odor Raw Parboiled 11 18 10 23 17 16 23 16 17 17 16 20 23 21 12 11 0.12 0.20 0.06 2 17 20 17 6 14 12 13 21 16 4 18 19 15 16 19 7 21 9 19 8 15 23 21 17 20 19 20 23 22 10 5 2 13 18 13 12 5 6 12 14 15 5 17 16 10 7 16 10 17 8 15 10 12 17 18 14 16 14 17 20 22 13 8 8 11 16 10 9 8 8 9 12 15 11 16 7 11 13 11 12 20 10 22 13 15 19 19 16 15 13 21 22 19 15 8 0.19 Raw Parboiled Raw Parboiled Raw Parboiled Color Stickiness Raw Parboiled 13 19 9 19 18 14 21 18 17 20 16 21 24 24 14 9 0.10 2 13 18 12 11 9 8 10 17 13 6 20 14 15 12 15 Raw Texture (soft-hard)
Milled rice appearance Parboiled 11 10 1 16 11 6 13 18 6 22 13 10 23 16 7 2 5 18 20 16 11 13 9 13 20 19 7 20 15 14 11 15
Taste-flavor Parboiled 10 15 17 18 11 13 9 13 14 17 8 21 18 14 13 12 0.26 13 20 9 18 16 15 20 16 13 14 14 20 22 22 11 9
Acceptability Raw Parboiled 4 12 18 16 9 13 8 9 12 15 6 17 17 14 11 12 0.23 9 16 6 19 18 11 20 18 12 16 13 22 24 21 13 6
Variety
Raw
Brown Gora RR139-1 RR151-3 RR151-4 RR166-645 RR203-16 RR2-6 RR265-1 RR347-166 RR348-5 RR348-7 RR352-1 RR354-1 RR50-5 RR51-1 Vandana
1 4 17 17 4 8 8 19 21 1 1 22 12 21 9 12
Rank correlation Raw-parboiled
-0.12
Varieties with high scores are the preferred ones.
59
tended. Participants examined farmer-participatory varietal selection experiments as well as on-station trials in Hazaribagh. More than 40 collaborative URRC experiments were conducted among IRRI and NARS scientists. A complete flora of upland weeds was published and distributed to research partners. More than 20 research publications and communications were published based wholly or in part on collaborative studies undertaken under the URRC. Through collaboration with Centre de cooperation internationale en recherche agronomique pour le developpement (CIRAD, France), computer software for climate analysis and prediction of water deficit in upland rice cropping areas was translated into English and made available to URRC partners.

Improved productivity and sustainability of farming systems in upland rice areas T. George, S. Pandey, M. Kondo, J.C. Castella, M. Mortimer, V. Murty, and G. Kirk
q
A survey of 980 upland farmers in northern Vietnam indicated that land and labor productivity and income were highest in marketaccessible areas. Upland rice provided food security in areas with poor market access. A methodology to study village accessibility to market, education, health services, and technical information has been developed. Work started on an inventory of technical innovations for upland rice-based cropping systems. Interventions that might increase production in sloping areas include the introduction of short-duration upland varieties, setting up production groups for high-quality seeds of Bao Thai rice in partnership with GRET (a French NGO), and introduction of cover crops and mulching techniques. These are being evaluated. A mechanistic mathematical model describing the processes involved in amelioration of subsoil acidity was developed. We concluded that management options could be developed based on the ideas of leaching down the effects of surface-applied lime and using acid-tolerant rice to help capture the leached solutes in the subsoil.
Field- and laboratory-determined P buffer coefficients from the Long-Term Phosphorus Experiment sites are being compared to arrive at better estimates of this coefficient. Laboratory-determined coefficients seem to be stable across crop seasons. In contrast, field coefficients decline with time after an initial spike that follows P fertilization. Collaborative experiments with the Philippine Rice Research Institute indicated high production potentials for Philippine upland Ultisols when adequate lime, N, and P were applied. Ultisols cover more than 8 million ha in the Philippines. Yields of 2 t ha1 for peanut and 6 t ha1 for maize were achieved on an upland site previously considered unsuitable for cropping. Data are being used to test model parameters for a Decision Support System and a Nutrient Management Support System, and to evaluate the yield predictions from these models. Upland rice in field experiments in India, Philippines, and Thailand was found to be more sensitive to water stress than N uptake and sensitivity was greater during the reproductive stage. Water stress seemed to cause a rapid decrease in water potential in the stem base before leaf water potential decreased, possibly indicating poor water extraction by upland rice roots compared with water transport within the plant.
Program outlook
The Upland Program will continue to focus on areas where problems of poverty, population pressure, and environmental degradation demand improved technologies. The largest area of upland rice in Asia is in eastern India, where upland rice is grown in permanent rice-based cropping systems. We will increase our emphasis on understanding the socioeconomic circumstances of farmers in the area and their capacity to adopt improved technologies. We will continue to use emerging technologies in functional genomics to discover the genes and traits that confer greater drought tolerance and disease resistance. We will develop additional populations from superior donor lines and evaluate those in target environments through collaboration
60
with NARS. This upstream research is being linked to impact by involving endusers of technology in the evaluation process, through farmer-participatory varietal selection. The long-term sustainability and profitability of higher yielding upland rice systems is being examined in multiyear fertility trials. Much of this work will have direct application to upland rice systems in Bangladesh and elsewhere. Another important upland rice system is found in Indonesia and the Philippines. Key issues in these highly market-integrated systems include yield losses due to disease and the need for technologies to reduce production costs. We will build on our initial candidate gene analysis to understand the specific resistance genes in different varieties and seek durable resistance to blast disease. Upland rice grown in zero-tillage systems is highly profitable in Brazil. We will work to bring zero-tillage technology to Asian upland rice farmers. Upland rice area in the traditional sloping regions of Southeast Asia has declined in the past decade due to government policies, improved market integration, and declining yields due to reduced fallow periods in traditional swidden systems. Nonetheless, recent reports indicate that adequate upland rice production remains essential for the food security of rural households in remote and marginal areas. Even when rice surpluses exist in the lowlands in a country, these do not translate into adequate rice for upland farm families.
Our project on perennial upland rice has made significant progress, and we anticipate that cultivars will soon be available for testing by farmers in erosion-prone upland areas. Spillover benefits from this project include the identification of valuable traits for nematode resistance and tolerance for water deficit in wild Oryza species. We will work to transfer these traits to annual varieties. Weeds are a major constraint to upland rice production. We will continue to evaluate genetic variation in weed competitiveness, with allelopathy as a component trait. We will focus the allelopathy research on identifying the chemicals responsible for observed variation and on its underlying genetic control. We will also examine management and germplasm options to improve the early growth of upland rice, to make the crop more competitive with weeds. Upland rice area is increasing dramatically in the tropical and subtropical areas in China due to scarcity of irrigation water to grow lowland rice. To improve the ability of rice to function as a fully aerobic crop in these systems, we will undertake work to clarify the underlying biology of aquatic adaptation. At the same time, we will devise management systems that will control weeds and supply nutrients effectively to aerobic rice. Our research focus will be guided by the priority problems identified through our URRC partners. We will reorganize our consortium activities to build on the capacity for upland rice research that has developed in NARS.
61
Research programs Flood-prone rice ecosystem
CROP AND RESOURCE MANAGEMENT TO IMPROVE PRODUCTIVITY AND SUSTAINABILITY OF FLOOD-PRONE RICE 64 Land and water management effect on acid sulfate soils, Mekong Delta, Vietnam (SWS) 64 GERMPLASM IMPROVEMENT 65 Genetic analysis of rice grain mineral density (PBGB) 66 Rice grain mineral density 66 Combining ability analysis of rice 66 QTL underlying phosphorus deficiency tolerance in rice (PBGB, SWS) 69 Phenotypic performance to P deficiency 69 AFLP-RFLP-SSLP map 70 QTLs for P-deficiency tolerance 70 IRRI lines named as varieties for saline-prone areas (PBGB) 76 PROGRAM OUTLOOK 76
Flood-prone rice ecosystem
There are about 12 million ha of flood-prone rice worldwide, 95% of which is in South and Southeast Asia. Priorities of the flood-prone ecosystem program include socioeconomic studies on changes occurring in the ecosystem, research on nutrient availability and soil toxicity, and resource management. The research is in three projects: 1) crop and resource management, 2) germplasm improvement, and 3) the Floodprone Rice Research Consortium. The consortium was established in partnership with concerned national agricultural research systems (NARS). The main goal is to formulate and test technology that will promote sustainable, stable production systems.
Crop and resource management to improve productivity and sustainability of flood-prone rice
The project objectives are: q evaluate socioeconomic and biotic constraints to productivity of flood-prone rice, q analyze the role of mineral elements in flooding tolerance and productivity of rice cultivars, and q determine the potential effects of global warming and changing weather on floodprone rice land. Land and water management effect on acid sulfate soils, Mekong Delta, Vietnam T.P. Tuong Understanding of changes in soil quality as affected by land use and water management will help in development of land and water management strategies for fragile acid sulfate soil environments. We collaborated with the Southern Institute of Water Resources Planning, Vietnam, to investigate 1) changes in chemical properties of a newly reclaimed acid sulfate soil with different land uses (rice cultivation vs fallow) and 2) variations in soil properties as a function of distance from adjacent drainage canals. The research was done at the Tan Thanh experimental farm in the Mekong River Delta. Soil at the site was classified as very fine Typic sulphaquept. The whole area is subjected to annual flooding to a depth of 11.5 m from September to December. The 5-ha farm was divided into four plots. Drainage canals at 100-m intervals were along the length of the
64
Al3+ (meq L-1) 30 a 25 20 15 10 5 0 3 9

LSD5% depth 10-20 cm 10/20 40-50 LSD5% depth 40-50 cm
Depth (cm) 10-20 40-50
Al3+ (meq L-1) 20 a 15 10 5 0 25
3 9 25 50 25 Distance from drainage canal (m)
50
b 20 15 10 5 0 May 1990 May 1991 May 1992 May 1993 Time
1. Soil solution aluminum (Al 3+) concentration at 10 20 cm and 4050 cm depths in (a) plot double-cropped with rice and (b) fallow plot, as affected by distance to adjacent drainage canals. Tan Thanh, 1999.
plots. Two plots were left fallow in their natural condition. The other plots were cleared. They had been in rice cultivation since 1990 with two rice crops annually from January to August. Aluminum concentration at 1020 cm and 4050 cm depth in all plots was measured monthly at 3, 9, 25, and 50 m from the adjacent canals. Figure 1 shows Al3+ concentration in soil solution for both types of plot as a function of distance from the adjacent canal. Aluminum concentration tended to increase in both plot types as distance from the drainage canals increased. The surface soil (020 cm depth) in the cultivated plots had been tilled, during which farmers drained surface water to remove acidity. Those activities made the soil more homogeneous across the field, resulting in no significant differences in Al3+ concentration across the field (Fig. 1a). Soils in the fallow plots and at 4050 cm depth in the rice plots were not affected by land preparation and their Al3+ concentration depended on vertical leaching. As a result, Al3+ concentrations were highest at the center of the field. Leaching was effective only at distances less then 25 m from the canals. Thus drainage canals should be no more than 50 m apart. Figure 2 shows the 1990-93 changes in soil solution Al3+ concentration at 40-50 cm depth. The Al3+ concentration in the rice plots decreased significantly (p <0.001), while there was no significant
2. Temporal changes in soil solution aluminum (Al 3+) concentration at 4050 cm depths in (a) plot double-cropped with rice and (b) fallow plot. Points designate the means of seven sampling site; vertical bars are standard deviations of the means. Tan Thanh, 1999.
change in the Al3+ concentration in the fallow plots. Leaching by the ponded water layer was responsible for the decrease of Al3+ in the rice plot. Land preparation, surface water drainage, and leaching by ponded surface water improved the topsoil of acid sulfate rice fields. Flushing and leaching activities, however, transfer acidity from the fields to the surrounding area and may contribute to environmental degradation.
Germplasm improvement
Germplasm improvement focuses on developing cultivars adapted to flood-prone environments. DNA-based molecular marker-aided selection (MAS) techniques were used to increase selection efficiency and permit simultaneous selection for a number of traits. Research focused on abiotic stresses and high micronutrient in rice grain.
65
Genetic analysis of rice grain mineral density Tin Htut, G.B. Gregorio, D. Senadhira, R.D. Graham,28 G.S. Khush, R.O. Mendoza, and A.N.R. Monroy Enhancing essential micronutrients in food crops through breeding requires understanding of the nature of combining ability among donor and recipient rice germplasm and development of appropriate breeding strategies. Only limited information is available for such a study in rice. Crossing nine parental rice varieties in all possible combinations developed a nine-parent diallel population. The parental rice varieties involved were Azucena, Basmati 370, BG300, IR64, IR68144-2B-2-2-3, IR72, IR74, Madukhar, and Xua Bue Nuo. Azucena, Basmati 370, Madukhar, and Xua Bue Nuo were selected as donor rices for their high grain Fe and Zn density as a result of preliminary germplasm screening at IRRI. Azucena is an upland variety classified as a tropical japonica. Basmati 370 is an aromatic variety classified as indica. Madhukar is a traditional Indian variety adapted to Zn-deficient soils. Xua Bue Nuo is a Chinese traditional red variety whose name means iron. The recipient parental varieties (BG300, IR64, IR72, and IR74) have high yield potential. IR68144-2B-2-2-3 is an improved rice line with enhanced micronutrient density. Seedlings of all 72 crosses and 9 parental genotypes were transplanted in experimental plots. Extra care was taken to avoid mineral contamination of seeds throughout sampling and sample preparation. Manually dehulled rice samples were analyzed for grain mineral density by inductively coupled plasma atomic emission spectrometry at the Plant Science Laboratory, Waite Agriculture Research Institute, University of Adelaide, Australia. The mean of two subsamples for grain Fe, Zn, Ca, Mg, Cu, Mn, P, S, and K density was subjected to diallel analysis. Only F1s and reciprocal F1s were included in the data analysis for unbiased estimates of general combining ability (GCA) of the parent rice varieties.
RICE GRAIN MINERAL DENSITY
The mean grain mineral density (GMD) values of the nine parents, when compared with their respective crosses, exhibited large differences in either positive or negative direction, suggesting heterosis of GMD. The range of grain Fe density was from 9.01 to 19.18 ppm, with maximum value in the cross Azucena/Basmati 370 followed by IR681442B-2-2-3/Madukhar, Azucena/Madukhar, Basmati 370/Xua Bue Nuo, and Azucena/Xua Bue Nuo, and then by Madukhar/Xua Bue Nuo and Azucena/ IR68144-2B-2-2-3. All high-Fe density crosses were also found to be high for grain Zn and other GMD. The maximum value (49.25 ppm) of grain Zn density was observed for Azucena/Basmati 370 followed by IR68144-2B-2-2-3/Madukhar and Azucena/Madukhar. The variability of the crosses for GMD was determined by their coefficient of variation (CV) across the population. The largest CV was observed for grain Zn (30.38%), followed by Fe (19.54%), Mn (17.31%), and P (14.56%). The diallel crosses varied largely for grain Zn density followed by grain Fe. Thus the selection for grain Zn density would be the most effective in this set of crosses.
COMBINING ABILITY ANALYSIS OF RICE
Highly significant variations among 72 crosses for the density of Fe, Zn, Ca, Mg, Cu, Mn, P, S, and K in grain were observed (Table 1). Highly significant GCA, specific combining ability (SCA), and reciprocal mean square for all GMD were also observed when variations among the crosses were partitioned into GCA, SCA, and reciprocal sums of square. Except for grain Cu density, R2 (coefficient of determination) for all GMD was high, indicating fitness with the additive dominance model. The GCA effects of the parent rice varieties were both positive and negative for all GMD. Varieties with positive GCA effect for most of the grain minerals were Azucena, Basmati 370, IR68144-2B-2-23, Madukhar, and Xua Bue Nuo (Table 2). This suggests GMD is heritable. All high-yield-potential varieties showed negative GCA for GMD, suggesting that they did not contribute to increase in mineral density traits of rice. Xua Bue Nuo showed highest GCA effect on grain Fe density. Madhukar
66
0.08ns 10.186ns 9594.91ns 3272.69ns 328692.13* 1.60** 94.799** 1030020.87** 70142.95** 281390.19* 7.80** 627.375** 5461345.90** 294795.25** 1511549.27**
238042.33**
40532.41**
23128.75 0.86
was the best combiner for Zn density. IR68144-2B2-2-3 was the best combiner for Ca density. Because magnitude of the GCA effect of a parent varies depending on the mineral elements, the association of GCA effects of parents on nine mineral nutrients was examined. High correlation coefficients among GCA effects on some mineral density were generally observed (Table 3). However, GCA effects of the parents on Zn and Ca mineral densities showed low correlation. The GCA effects of the parent rice varieties on grain Mn and other grain minerals, except S, also had low correlation coefficients. This suggests that basing parental selection on GCA to breed for enhancing the GMD cannot be applied for enhancing the overall mineral concentration of the rice grain. This may be due to the fact that parental varieties were originally selected only for high Fe and Zn. Choosing parent rice varieties based on one or two mineral nutrients may upset the biologically important ratio of grain minerals in future rice varieties. Therefore, breeding for enhancing grain mineral elements should aim for optimization of mineral elements important for human nutrition and for selection strategies for multiple mineral traits. Highly significant SCA across mineral elements was observed (Table 3). However, there were significant negative correlation coefficients between SCA effect on grain Ca and other minerals such as Mg, Mn, P, S, and Zn. The SCA effect on grain Mn also showed negative correlated effect on grain Cu, Fe, K, Mg, Mn, P, S, and Zn. Therefore, selection based on SCA of specific crosses for GMD would have negative effect on other grain minerals. Our results suggest that selection based on SCA should be done on the most nutritionally limited grain minerals and with the condition that negative consequences on other minerals not alter either the human or plant nutritional requirement. An alternative is selection based on the index form of multiple GMD. The correlation among GCA and SCA effects on the nine mineral densities could be considered as genotypic correlation. This relationship can still be modified by environment, cancellation effect, and opposition between GCA and SCA effects as shown in phenotypic correlation among the mean density of each mineral nutrient based on F1s and reciprocal F1s. There is no negative relationship, suggesting some cancellation effect of GCA
Table 1. Analysis of variance for rice grain mineral density of a 9 9 diallel population according to model 1, method 3 of Griffing. IRRI, 1999.
85218.98**
963051.15**
8913.19** 95509.26** 18.505** 0.77* 8935.30** 74.90** 19.27** 1.68* 36
38.725**
Mean square
Mn
0.87**
8.23ns 351.97** 15468.52* 188.26** 324.14** 90820.97** 941.71** 2101.06** 429487.57**
99656.20**
129.96**
Ca
190.32**
Zn
2 1.53ns 72 18.13** 8 97.53**
27 16.55**
Replications Crosses General combining ability (GCA) Specific combining ability (SCA) Reciprocal sum of squares Error R2
Source
142
df
1.03 0.90
Fe
3.59 0.96
25.76 0.87
4393.05 0.91
Mg
0.46 0.63
Cu
5.439 0.90
37024.48 0.93
4333.72 0.90
67
Table 2. General combining ability effects of nine rice genotypes for grain mineral density in 9 9 diallel population. IRRI, 1999. Mineral elements Parent Fe Azucena Basmati 370 BG300 IR64 IR68144 IR72 IR74 Madhukar Xua Bue Nuo SE 1.13 0.61 1.84 1.55 0.87 1.63 1.18 1.75 1.82 0.15 Zn 6.26 3.10 6.24 4.72 0.03 4.15 2.56 6.50 1.83 0.28 C 6.59 2.19 2.37 6.25 9.79 3.33 9.95 7.54 6.68 0.74 Mg 124.51 22.13 138.11 94.66 44.63 102.16 62.28 91.30 114.63 9.64 Cu 0.14 0.20 0.57 0.52 0.02 0.04 0.20 0.07 0.88 0.10 Mn 4.38 6.74 4.17 3.36 3.41 1.03 2.44 2.57 0.96 0.34 P 395.11 171.30 451.32 362.04 108.20 367.99 266.80 351.06 422.49 27.99 S 34.22 12.57 61.73 51.49 16.01 79.35 92.21 160.77 86.36 9.57 K 298.2 126.85 177.91 188.62 37.57 199.34 181.48 119.71 164.95 22.12
Table 3. Correlation coefficients among general combining ability (GCA) effect and specific combining ability (SCA) (in parenthesis) of parent rice varieties on grain mineral density. IRRI, 1999. Ca Cu Fe K Mg Mn P S Zn 0.35 (.15) 0.53 (.29) 0.27 (.27) 0.39 (38*) 0.07 (0.69**) 0.41 (38*) 0.61 (34*) 0.26 (44*) Cu Fe K Mg Mn P S
0.66* (0.79**) 0.52 (0.66**) 0.69* (0.74**) 0.15 (6**) 0.67* (0.75**) 0.49 (0.81**) 0.49 (0.73**)
0.91** (0.74**) 0.97** (0.92**) 0.49 (3**) 0.98** (0.90**) 0.89** (0.91**) 0.90** (0.89**)
0.95** (0.88**) 0.66 0.53 (35*) (0**) 0.96** 0.99** (0.91**) (0.99**) 0.75* 0.83* (0.83**) (0.96**) 0.92* 0.92** (0.83**) (0.96**)
0.54 (9**) 0.14 (2**) 0.56 (0**)
0.85** (0.95**) 0.94** (0.96**)
0.81* 0.94**
and SCA or environment. Phenotypic correlation among GMD also showed that Fe and Zn were largely and significantly correlated with P, S, and K (Table 4). Crosses with high SCA effects were IR64/Xua Bue Nuo, IR72/Madukhar, IR68144-2B-2-2-3/ IR72, IR64/IR74, and IR64/IR72. However, selections based on SCA alone cannot guarantee higher GMD because SCA was composed of dominance and some type of epistasis components. When the relative importance of GCA and SCA for GMD was considered, the suggested ratio of GCA and SCA showed 0.87 and higher for all grain minerals (Table 5). Our results indicate that GMD of a single cross can be highly predictable based on the GCA of the
parent rice varieties. Primary selection should be done based on GCA, followed by selection based on SCA for better cross combination coupled with their mean performance. When the correlation between SCA effects of each cross and their GMD were considered, only moderately-significant correlation was observed in each mineral density. Therefore, selection based on GCA is more warranted than selection based on SCA effect. There were significant reciprocal effects on GMD traits, but those were not considered important because the size of reciprocal mean square was relatively smaller than for GCA and SCA effects. Parents and crosses selected by GCA and SCA effects were mostly traditional rice varieties and their combinations require further breeding to com-
68
Table 4. Phenotypic correlations among grain mineral density of 71 crosses (F1 + RF1) of 9 9 diallel population. IRRI, 1999. Ca Cu Fe K Mg Mn P S Zn 0.25* 0.31** 0.13ns 0.16ns 0.23ns 0.20ns 0.30* 0.07ns Cu Fe K Mg Mn P S
0.66** 0.51** 0.64** 0.10ns 0.63** 0.58** 0.55**
0.84** 0.94** 0.21ns 0.94** 0.89** 0.89**
0.90** 0.38** 0.93** 0.77** 0.87**
0.20ns 0.99** 0.88** 0.93**
0.25* .07ns 0.22ns
0.88** 0.94**
0.86**
Table 5. Suggested ratio of general combining ability (GCA) to specific combining ability (SCA) determined by 2 MSGCA/(2 MSGCA+MSSCA) for eight minerals. IRRI, 1999. Grain mineral Fe Zn Ca Cu Mn P S K GCA/SCA 0.92 0.91 0.97 0.95 0.97 0.92 0.87 0.93
bine high GMD and agronomically important characteristics. Evidence of existing heterosis should be exploited and needs to characterize for GMD. The heterosis and significant SCA in this diallel population indicates need for in-depth genetic analysis to investigate the possibility of releasing the existing latent genetic variability by either epistasis or linkage for GMD. It is essential to understand the nature of GMD in rice grain so as to not upset the biologically important ratio of all grain minerals. QTL underlying phosphorus deficiency tolerance in rice J. Ni, G.B. Gregorio, P. Wu,29 G.V. Vergara, Z. Li, C. Quijano-Guerta, and G. Kirk Phosphorus deficiency is a major limiting factor for rice production in highly weathered soils in many parts of the world. In addition to the natural low level of P of those soils, there is high P fixation in rice soils such as acid, acid sulfate, and alkaline soils. Genotypic variation in P-deficiency tolerance
by rice has been extensively documented, and it is possible to develop varieties with increased P deficiency tolerance. A previous study used a recombinant inbred (RI) population for genetic mapping. The marker loci linked to a major quantitative trait locus (QTL) underlying P deficiency tolerance were identified based on segregation of phenotypic parameters in solution culture. Because of the heterogeneity and complexity of soils, it is crucial to confirm the result in soil culture before initiating MAS or mapbased cloning programs. Based on the results from the previous study, 42 each of the tolerant and sensitive RI populations were used in mapping P-deficiency tolerance QTLs. The RI lines were from the cross IR20/IR55178-3B9-3, where IR20 is P-deficiency tolerant and IR55178-3B-9-3 is sensitive.
PHENOTYPIC PERFORMANCE TO P DEFICIENCY
The selected RI lines (RIL) were phenotyped for tolerance for P deficiency in a greenhouse pot experiment using soil collected from a P-deficient field in Pangil, Laguna, Philippines (soil pH =4.7; Olsen P = 3 mg kg1). The field had not received P fertilizer for at least 30 years. The soil was air-dried, ground to pass a 2-mm sieve, and fertilized at either low P level (0.01 g P2O5 kg soil1) or medium P level (0.06 g P2O5 kg soil1) with sufficient N (0.08 g N kg soil1) and K (0.06 g K2O kg soil1) for all the treatments. Four pregerminated seeds of each RIL were sown in pots containing 6.5 kg soil with either low P or medium P level. Sixteen plants of each parent were also sown in pots with two different P levels.
69
At 2, 3, 4, 5, and 6 wk after sowing, the tillers of each plant were counted. At 42 d after sowing, the shoots of each plant were removed, oven-dried at 65 C for 3 d and weighed to determine shoot dry weight (SDW). The dry plant tissue was ground and analyzed for P content. P uptake (PUP as mg plant1) and P use efficiency (PUE as plant dry weight mg1 of absorbed P) were calculated. Relative values of the growth parameters (relative parameter) were calculated. The parameters determined were relative tillering ability (RTA), relative shoot dry weight (RSDW), relative P uptake (RPUP), and relative P use efficiency (RPUE) (Table 6). IR20 had higher RTA and RSDW than IR55178, a result consistent with earlier reports, indicating that IR20 was more tolerant of low-P stress than IR55178. Segregation among the 84 selective RILs for RTA and RSDW was observed and the distributions of the parameters were normal after log-transformation. The population means of the parameters were near the parental means. Transgressive variations in all four parameters were observed. For RSDW, about 18.9% of the 84 selective RILs had a RPUP higher than IR20, and 38.9% had lower RPUP than IR55178. Correlation analysis among RSDW and RTAs for the 84 RILs showed highly significant correlation among these relative parameters.
AFLP-RFLP-SSLP MAP
veyed for simple sequence length polymorphism (SSLP) and polymorphic markers were used for selected RILs. Using SSLP markers together with previous AFLP and RFLP anchors, all other AFLP markers were assigned to the 12 linkage groups at LOD >3.0 based on their linking to the anchor markers. The map distances were estimated by the Kosambi function. The parental survey revealed 27 polymorphic SSLP markers. All these SSLP markers were mapped to the expected location verifying the previous AFLP-RFLP map. The 27 SSLP markers, 26 RFLP markers, and 30 AFLP markers were used as anchor markers to generate the linkage map (Fig. 3). The linkage map had a total map length of 1830.1 cM. The average interval size was 8.39 cM with the smallest on chromosome 4 (2.96 cM) and largest on chromosome 7 (17.23 cM).
QTLs FOR P-DEFICIENCY TOLERANCE
Amplified fragment length polymorphism (AFLP)restriction fragment length polymorphism (RFLP) genotyping and map construction of the RI population were done earlier. The parents were also sur-
Data for RTA, RSDW, RPUP, and RPUE were used for QTL analysis after being normalized by log transformation. Single-marker locus analysis at a probability of less than 0.005 for error I and interval mapping with LOD value >2.4 were used for detecting marker loci associated with the variations in the parameters measured and most likely position of the gene loci. The mean comparisons among marker genotypes for each trait were conducted using t-test analysis. The proportion of the phenotypic variance explained by the marker loci linked to QTLs detected was provided by regression analysis. The RSDW and RTAs were used for both singlemarker analysis and interval mapping analysis (Tables 7, 8). For RTA2w, only one QTL accounting
Table 6. Means and standard deviation of relative tillering ability (RTA) and relative shoot dry weight (RSDW) for the parents and selected genotypes grown in soil culture. IRRI, 1999. Parent Parametera RSDW (%) RTA2w RTA3w RTA4w RTA5w RTA6w
a
Selective genotype IR55178 25.22.9 32.51.5 28.22.6 23.63.3 25.24.1 33.14.7 Tolerant 38.9 7.8 66.425.9 59.325.9 44.918.4 47.230.2 49.030.4 Sensitive 14.5 5.1 53.619.5 38.614.2 27.710.8 22.712.2 21.910.5 Range 2.9146.5 30.8116.7 18.2155.6 10.8130.0 8.0155.6 8.3160.0
IR20 34.13.7 45.84.8 60.35.7 47.55.5 49.48.4 51.67.3
RSDW = relative shoot dry weight, RTA2w = relative tillering ability 2 wk after sowing, RTA3w = RTA 3 wk after sowing, RTA4w = RTA 4 wk after sowing, RTA5w = RTA 5 wk after sowing, RTA6w = RTA 6 wk after sowing.
70
CHRO.1a CHRO.1b CHRO.2 Marker cM 0.0 20.0 21.5 33.3 36.9 60.5 79.2 P1M10_3C P1M8_4CH (E3M6-6) E3M3_6 P2M4_5CH (P4M8-1, E1M7-11) RM263 P1M4_6
(P1M8-2CH, P2M6-3, P2M9-5, P4M4-2, P4M9-3)
cM cM Marker E4M7_1
(E1M7-10)
Marker 0.0 E4M7_2
0.0
RG140
(OSR2, RM234, P1M3-1,
RG147 P1M7-10, P3M4-6,
P4M7-8, E1M7-6)
0.0 19.7
RG173
41.8
P1M1_6CH
50.6 58.0
E3M5_4 P2M1_2
77.4
E3M3_7
(P1M6-6, E7M5-4)
(P2M5-1, E3M8-10, E4M6-8)
95.9
E3M5_8
108.0 CHRO.4 cM 0.0 4.8 8.0 9.9 14.8
101.8 107.6 110.5 112.6 114.7 117.2 124.3
P1M5_8 P3M1_1 P3M2_4 (P4M7-6) P3M9_8 (P2M4-3, P3M3-6) E3M7_9 P2M8_4CH E1M3_12
RM207
7.0 11.7 20.8 22.8 22.8 24.1 26.0 28.6 28.6 31.9 34.5 38.5 40.4 43.4 46.4 59.2 72.8 76.9 80.0 82.2 92.5 102.2 111.3 E4M6_4 E7M5_6 (E5M4-8) E5M1_3 E4M1_8 E4M3_8 E1M5_8 P4M7_4 P4M3_1 P1M6_2CH P3M9_10 P1M2_2 P3M1_3 RG236 E3M3_5 E4M1_5 E4M3_11 (E5M4-5) (P4M8-3) P4M1_1 P4M7_3 P2M9_3 E4M2_11 (P3M3-1) (P1M4-5,E6M5-4, P3M1_7 E7M5-7) E5M4_6 (P3M2-6) E1M3_4 Marker E4M1_1
E1M7_9 E4M4-10, P1M9_3CH (E4M1-6,E7M6-5, E7M5-2, E3M6_11 E1M7-1) P2M5_2
3. AFLP-RFLP-SSLP linkage map of the IR20/IR55178 recombinant inbred population showing the location of quantitative trait loci associated with RSDW and RTA. IRRI, 1999.
71
72
Figure 3. continued.
CHRO.3 CHRO.5 CHRO.6
cM cM cM 0.0 5.4 20.6 P3M10_6 (P3M3-8)

(RM148)
Marker Marker E7M5_5 P1M10_4 0.0 RZ394 RZ576
Marker E3M8_3 E1M7_7

(E5M4-3)
0.0
17.3
24.8
(P2M8-2, E7M6-3)
53.5 65.5 71.5 78.6 93.4
P1M7_2 P3M2_5
27.1 31.5 33.6 36.6 41.0 64.9

(P1M5-13, P2M2-2, P4M2-1,P4M7-2)
RG179 (E4M2-10) E1M5_1 (E3M5-6) E4M6_9 E6M1_9 E7M6_1 (P1M8-5, E4M7-4) P1M5_14CH
(P4M4-4)
E3M3_2 P2M4_6 (P1M5-9)

(E4M7-3)
E4M6_7
110.4 RM227 P1M7_8

(P3M4-3, P3M9-5) (E5M4-13)
P2M6_1CH
118.5 124.0 E3M6_1 P1M9_1CH RG418 P3M6_1 (E6M1-1)
139.5
28.3 32.0 33.4 37.2 42.7 53.0 63.4 74.1 88.5 88.5 102.4 105.5 110.0 112.0 113.3 136.7
E1M3_10 (P3M9-9) E1M3_13 (P3M7-5) E3M6_3 (RM249, E3M3_9 P1M8-1) P1M2_1 (P1M5-11) RM164 RG573 P1M1_8CH P1M5_12C P1M5_10C E6M5_5 RM31 RG493 RG474 RG119 P3M10_3 (OSR34, P3M2-7)
101.4 109.6 113.9 117.7 130.2 140.5
P2M9_8 (E3M8-1) RM3 P1M8_3CH (P3M5-2) (E1M3-11) RG123 P4M4_6 RG64
156.0
161.5
183.1
144.4 146.9 150.9 157.3 183.1
E4M2_2 P1M6_8CH E6M1_7 E4M1_7 P4M7_9

(RM204)
cM cM Dist (cM)
(E4M3-14) (E1M3-7)
Marker Marker Marker E1M3_5 0.0 16.8 32.5 42.0 P3M1_5

(RM219)
(P1M1-2, P2M6-5)
(P1M1-7, P1M6-3, P4M7-5)
0.0
P1M2_4
0.0
(P1M1-3, P1M6-5)
P1M3_2CH
(P4M4-5)
22.8
RM51
(P2M9-10)
P3M10_1 E3M7_2
46.0 57.2 RM38 OSR30 67.0

(P1M1-1, E1M3-15)
RG128
5.5 6.8 9.4 23.7 27.1 32.7
E3M7_6 P4M3_3 (E4M2-6) E3M6_10 (P2M9-9) E3M8_6 P2M10_2 (E7M5-10) E1M5_6
(P1M5-4, P3M5-4, P3M6-4, P4M3-2, E5M1-2, E5M4-4, E5M4-14) (P1M6-1CH)
100.8
E1M7_8
126.8 138.3 RG136
RM248
123.3
P3M5_3
144.1
P4M9_4
(P2M4-4, P3M10-2,
(P2M6-6, P4M7-7)
166.4
P2M4_7
(E1M5-2)
45.2 47.5 49.4 52.6 56.5 64.4 82.1 83.1 84.5 86.7 89.9 95.3 101.5 109.3 121.1 140.3 169.6
(OSR7)
P1M5_7 (P3M7-1) (E6M5-3) E4M3_15 P2M7_2CH E1M3_6 E4M6_5 E4M2_8 (E1M3-8) P1M7_13 E3M5_5 E1M7_3 P1M7_11 P1M3_3CH (E1M3-3) E3M6_2 E7M6_6 E5M4_2 E1M7_5 (RM257) RM242 (E5M1-4) RM205 181.5 182.8 183.4 186.9 190.1 194.0 204.4 RG451 RZ404 RZ792 (P1M5-6CH) P4M10_2 (P4M1-4) E7M5_1 E1M3_14 E3M3_11
171.6 175.5 177.6 180.8 206.8
E3M6_8 (E4M1-3) P1M2_5CH E6M5_2 (E3M7-5) P1M7_14CH P2M6_2
140.9 140.9 152.0 154.9 157.3 162.3
E7M6_7 E3M5-9, E1M5-5) P1M3_4CH E1M7_10 P2M5_3CH E6M1_6 RM256
73
74
Figure 3. continued.
cM Dist(cM) Dist(cM)
(E4M6-6)
Marker Marker RZ797

(P2M1-4)
Marker
(RG241)
0.0 0.0 13.9 21.1

(P4M4-3) (P3M5-1, P3M9-2 P4M1-2, P4M1-3)
E7M5_8 E4M2_7 P1M6_4CH P2M4_8 8.6 15.4
0.0
OSR32
(P2M4-ICH, E5M4-7) E5 M4-7)
(E3M6-7, E7M6-4)
RG9 (P1M7-9,E3M8-8) P1M7_7 (P1M1-5CH,E4M2-1) P1M9_2 E5M6_3 P2M9_2 E1M5_7 P1M4_2
25.7 35.1 20.0 22.7 29.3
P2M2_3CH
37.7
P1M1_4
35.1 47.7 54.5 54.5 65.8 85.1 91.8 106.0 P4M8_2

(E4M2-4, E4M6-1)
42.5 45.9 47.1 49.0 52.8 P1M5_2CH (P1M7-1) P3M2_1 (E4M3-16, P3M2_3 E4M3-17) P2M2_4CH (P3M2-2 P3M9-3) E5M4_12 RM209 RM21
(P1M7-6)
P1M4_4 E6M1_5 E6M1_4 P2M8_3 P1M5_3
(P1M10-2, E5 M6-6) E5M6-6)
39.4 49.0 57.2 64.2 73.2 75.2 77.9 84.0 88.0 92.8 96.1 118.9 128.4
P1M7_12 E1M5_9 RZ76 RZ261 RG413 CDO344
(P1M7-3, P2 M9-6) P2M9-6)
57.6 61.1 66.2 74.7 90.8 111.0 116.2 118.3 121.7 129.4 140.4 143.3 144.9 146.6 146.6 149.4 150.8 160.9
E4M3_12 E7M6_2 E3M7_10 (P3M1-4, E1M3-9) (E5M4-1) P4M10_3 P1M4_3 E5M6_5 OSR33 P3M1_6 P3M9_7 (P3M6-5, E5M4-11) E5M4_9 P3M2_8 RG561 P3M3_3 P3M3_4 P3M1_2 P3M2_9 E3M5_1 (P3M10-4) P3M6_5 (P3M6-2)
E3M6_5 E1M5_12 P2M1_1 E7M5_9 RM235
Table 7. Putative QTLs underlying tolerance for P stress detected using interval mapping program based on relative parameters measured. IRRI, 1999. Traita Marker Chromosome Additive effect 0.200 0.228 0.180 0.109 0.121 0.115 0.128 0.086 0.149 0.177 0.151 0.193 0.116 % VARb LODc
RSDW RTA2w RTA3w RTA4w
RTA5w RTA6w
E3M8-3~ E1/M7-7 OSR32~RG9 E3M8-3~ E1/M7-7 E3M8-3~ E1/M7-7 OSR32~ RG9 E3M8-3~ E1/M7-7 OSR32~ RG9 RG413~CDO344 E3M8-3~ E1/M7-7 OSR32~ RG9 E3M8-3~ E1/M7-7 OSR32~ RG9 RG413~CDO344
6 12 6 6 12 6 12 12 6 12 6 12 12
28.6 36.4 27.8 33.2 40.0 28.0 34.6 15.9 28.5 38.4 31.7 50.1 19.2
4.96 6.97 4.39 6.17 7.45 5.12 6.25 2.83 5.09 7.13 5.65 10.11 3.48
a RSDW = relative shoot dry weight, RTA2w = relative tillering ability 2 wk after sowing, RTA3w = RTA 3 wk after sowing, RTA4w = RTA 4 wk after sowing, RTA5w = RTA 5 wk after sowing, RTA6w = RTA 6 wk after sowing. bVariation contributed by single QTL across the population. cLOD scores (log10-likelihood ratio).
Table 8. Mean comparisons for the relative parameters measured between genotypic classes of marker loci linked to QTLs detected. IRRI, 1999. Genotype Parametera Marker Chromosome IR20 IR55178 Peb Fc R+ Av. (%)d 72.6 79.6 73.3 102.0 34.1 49.1 49.0 50.7 48.2 -50.9 -55.8 50.9 36.4 66.9 73.5 73.7 67.6 65.8 72.4 74.5 46.9 49.9
RSDW
RTA2w RTA3w
RTA4w OSR32
RTA5w
RTA6w OSR32
E3M8-3 E1/M7-7 OSR32 RG9 E1/M7-7 E3M8-3 E1/M7-7 OSR32 RG9 E3M8-3 E1/M7-7 12 RG9 RG413 E3M8-3 E1/M7-7 OSR32 RG9 E3M8-3 E1/M7-7 12 RG9 RG413 CDO344
6 6 12 12 6 6 6 12 12 6 6 45.49 12 12 6 6 12 12 6 6 48.36 12 12 12
14.21 15.53 36.27 39.17 50.67 35.52 37.80 60.79 60.47 25.70 27.11 26.42 44.98 40.71 21.46 23.12 47.31 45.89 21.99 23.53 21.24 47.61 39.95 43.74
33.62 36.79 16.69 11.91 70.99 59.55 61.77 35.99 36.90 44.13 47.31 19.07 26.56 27.52 44.82 48.77 21.57 22.27 45.29 49.19 27.12 21.23 23.35 26.06
19.41 21.26 19.58 27.26 20.32 24.03 23.97 24.80 23.57 18.43 20.2 20.65 18.42 13.19 23.36 25.65 25.74 23.62 23.3 25.66 24.52 26.38 16.60 17.68
10.99 13.61 9.40 26.28 16.14 23.74 26.14 27.47 24.22 18.81 24.70 0.212 21.78 8.98 18.19 22.70 22.35 20.33 18.32 22.81 0.246 27.64 8.83 9.42
0.133 0.155 0.107 0.268 0.172 0.253 0.262 0.269 0.246 0.208 0.248 52.7 0.224 0.104 0.204 0.234 0.229 0.214 0.205 0.235 76.7 0.273 0.103 0.112
a RSDW = relative shoot dry weight, RTA2w = relative tillering ability 2 wk after sowing, RTA3w = RTA 3 wk after sowing, RTA4w = RTA 4 wk after sowing, RTA5w = RTA 5 wk after sowing, RTA6w = RTA 6 wk after sowing. bPhenotypic effects. c The F statistic as determined by the PROC GLM procedure of SAS. dCoefficient of determination, the percent of phenotypic variation explained by individual markers as determined from the PROC GLM procedure of SAS.
75
Table 9. IRRI breeding lines from the flood-prone breeding program named as varieties in the Philippines, 1999. Breeding line Parentage PSB name PSBRc84 PSBRc86 PSBRc88 Local name Sipocot Matnog Naga
IR65185-3B-8-3-2 IR65195-3B-13-2-3 IR52713-2B-8-2B-1-2
CSR10/TCCP266-B-B-B-10-3-1 IR10198-66-2/TCCP266-B-B-B-B-10-3-1 IR64//IR4630-22-2-5-1-3/IR9764-45-2-2
for about 27.8% of the total variations was closely linked to P3M8-3 and and E1/M7-7 on chromosome 6. For RSDW and RTA3w~6w, two common QTLs were detected by using both single-marker analysis and interval mapping. First common QTL flanked by marker loci RG9 and OSR32 on chromosome 12, respectively, explained about 36.4% and 33.2 50.1% of the total variations across the 84 selective lines for RSDW and RTA3w~6w. Second common QTL accounting for about 28.6% and 28.033.2% of the total variations in RSDW and RTA3w~6w were closely linked to E3/M8-3 and E1/M7-7 on chromosome 6. For RTA4w and RTA6w, another common QTL flanked by marker loci RG413 and CDO344 on chromosome 12 was identified, and it explained about 15.919.2% of total variation. To investigate allelic information of the QTLs for P-deficiency tolerance, mean comparisons were performed for the relative parameters between genotypic classes of IR20 and IR55178 at the marker loci linked to the detected QTLs. For most of significant marker loci, the genotypic class of the tolerant parent, IR20, was superior to other marker classes. However, the classes identified by the sensitive genotype, at E3/M8-3 and E1/M7-7 on chromosome 6, were superior to tolerant classes (Table 8).
IRRI lines named as varieties for saline-prone areas G.B. Gregorio, D. Senadhira, R.O. Mendoza, J.P. Roxas, G.S. Khush, P.S. Bonilla,1 H.C. dela Cruz,1 and T.F. Padolina1 Three IRRI breeding linesPSBRc 84, PSBRc 86, and PSBRc 88from the flood-prone breeding program were named as varieties for saline-prone areas in the Philippines (Table 9). These varieties had improved tolerance for salinity, were early maturing, resistant to diseases, and had good grain qualities. This brought the number of IRRI breeding lines named as varieties in saline-prone areas in the Philippines to five. Two others were released in 1995. In addition, the Rice Varietal Improvement Group of the Philippines identified another salt-tolerant line, IR58443-6B-10-3, for pre-release, multiplication, and distribution to farmers in coastal areas.
Program outlook
The program will merge with the Rainfed Lowland Ecosystem Program in 2000. Flood-prone ecosystem research will be combined under two projects crop and resource management and breeding. Use of MAS techniques will continue in screening and in understanding abiotic stresses in flood-prone rice areas.
76
Research programs Cross-ecosystems research
BIOTECHNOLOGY TO ACCELERATE RICE BREEDING AND BROADEN THE RICE GENEPOOL 78 Construction of a genomewide physical map of rice using the IR64 BAC library (PBGB) 78 Molecular cloning of xa5 (PBGB) 79 Physical map position of xa5 80 Identification of candidate cDNA clones 80 BLAST analysis of the terminal sequences of cloned markers from the genetic map of rice 81 Development and field evaluation of transgenic IR72, M.H. 63, and hybrid rice Shan You 63 with a Bt gene for stem borer resistance (PBGB, EPP) 81 Evaluation of transgenic IR72 with Xa21 for bacterial blight resistance (PBGB, EPP) 83 Evaluation of transgenic rice with Xa gene for bacterial blight resistance (EPP, PBGB) 84 NARS development of elite lines through breeding and marker-aided selection (EPP, PBGB) 85 Philippines 85 India 85 Indonesia 86 EXPLOITING BIODIVERSITY FOR SUSTAINABLE PEST MANAGEMENT 86 Rice blast management through varietal diversification (EPP) 86 Field variants of tungro (EPP) 87 BIOLOGICAL NITROGEN FIXATION 88 Diazotrophic endophytic bacteria (SWS) 89 Growth-promoting activities (SWS) 89 Working group meeting 89
RICEA WAY OF LIFE FOR THE NEXT GENERATION OF RICE FARMERS 89 Using multisector partners to deliver IRRI rice technology to rainfed rice farmers in northeastern Thailand (SSD, AE, TC, CIAP) 90 1998-99 activities 90 Lessons learned (1998-99) 91 Post-production systems research (AE) 91 Philippine Post-production Research Consortium 91 Rice quality management (AE) 92 SOCIOECONOMIC STUDIES FOR TECHNOLOGY IMPACT, GENDER, AND POLICY ANALYSIS 92 Production of improved rice varieties in South and Southeast Asia (SS, PBGB, Biometrics) 93 Production of varieties 93 NARS uses of IRRI materials 93 Demand for specialty rices: implications for technology development (SS) 98 Changing structure in food demand 98 Policy implications 98 IMPLEMENTING ECOREGIONAL APPROACHES TO IMPROVE NATURAL RESOURCE MANAGEMENT IN ASIA 98 Market integration, agricultural diversification and erosion risk in northern Thailand (APPA, SS) 99 Changes in cropping systems to reduce erosion risk 99 Erosion risk decrease 99 PROGRAM OUTLOOK 99
Cross-ecosystems research
The Cross-Ecosystems Research Program is designed to develop knowledge and tools to solve problems that are common across rice-growing ecosystems. The program is anticipatory and seeks to create new opportunities, in addition to research on current problems that can enhance ecosystem-based research. The program includes six research projects that span the disciplines of biological science, natural resource management, and anthropology and socioeconomics. One focus of the program is to apply advances in biological sciences to develop practical tools to improve germplasm and pest management, and to explore opportunities in biological N2 fixation in rice. The biotechnology project involves the application of molecular and genetic tools to assist plant breeding. The Asian Rice Biotechnology Network (ARBN) establishes close collaboration with national agricultural research systems (NARS) to develop elite breeding lines that address local problems. Systems analysis is used to delineate the biophysical and socioeconomic factors affecting land use and resource allocation and their implications to rice production. Through the Systems Analysis and Simulations in Rice Production Systems (SysNet), we form partnerships with NARS to develop methodologies and tools for improving land use planning at the national level and below. The program examines factors affecting technology adoption and the mechanisms needed to accelerate technology transfer to improve the livelihood of farmers. The socioeconomic component is concerned with how rice technologies can benefit farmers and with its impact on socioeconomic equality and poverty.
Biotechnology to accelerate rice breeding and broaden the rice genepool

Biotechnology research at IRRI covers a broad spectrum of activities. Notable successes were achieved throughout the spectrum during 1999. At the basic end, the development of a physical map for chromosomes 11 and 12 of rice passed the half-way point and terminal sequencing was conducted on 350 cloned markers from the genetic map of rice. At the applied end of the spectrum, several transgenic plants, with enhanced resistance to bacterial blight or stem borers, were tested in China. Molecular breeding through marker-aided selection was successfully applied in several NARS laboratories. Construction of a genomewide physical map of rice using the IR64 BAC library A.C. Sanchez, B. Fu, J. Talag, R. Maghirang, C. Aquino, S. Yu, J. Domingo-Rey, J. Mendoza, L. Nguyen, D. Brar, G.S. Khush, and Z. Li A physical map consists of a series of ordered, contiguous recombinant DNA clones spanning the entire genome. It is measured physically in terms of the number of base pairs in the DNA and is essential for the analysis of genome structure. A physical map is also the first requirement for the isolation and characterization of agronomically important genes based on their map positions. Our objective is to use the IR64 bacterial artificial chromosome (BAC) library to construct a physical map of rice that has overlapping BAC clones spanning all the rice chromosomes and to facilitate more efficient cloning of important rice genes-quantitative trait loci (QTLs).
78
The physical map was constructed in three steps. First, the BAC library was screened with anchor restriction fragment length polymorphism (RFLP) and sequence-tagged site (STS) markers to establish anchor BAC islands across the genome. Second, the expected gaps between anchor BAC contigs were filled by random chromosome landing. In doing so, three-dimensional DNA pools prepared from the whole library were amplified with large numbers of random primers to generate random BAC islands for gap filling. Putative-positive clones identified by random primers were isolated and confirmed individually with the corresponding DNA markers. Assignment of these random BAC islands to individual chromosomes was simultaneously conducted by genetic mapping (randomly amplified polymorphic DNA [RAPD] markers associated with BAC clones were mapped to their corresponding chromosomes using a recombinant inbred mapping population) and by hitting anchor BAC islands. In the third step, BAC contigs were assembled from the BAC islands by HindIII fingerprinting. A total of 453 RFLP and STS markers have been used to screen the BAC library, establishing the anchor BAC islands across the genome. On average, 3.3 BAC clones were identified by each marker, agreeing with the size of the BAC library (3.28 genome equivalents) although chromosomes 710 have the least number of assigned BAC clones. We also screened the BAC library with 498 RAPD primers. Of those, 262 RAPD primers identified 614 BAC islands consisting of 1,600 clones. Eighty-nine of the 614 BAC islands were localized to respective chromosomes by hitting clones of the
anchor BAC islands. Hence, a total of 546 islands consisting of more than 1,800 clones have been assigned to the different chromosomes (Table 1). Based on these results, the framework physical maps for chromosomes 11 and 12 were largely established. The physical map of chromosome 11 consisted of 93 anchor BAC islands and 34 random BAC islands with a total of 438 clones. This physical map is estimated to cover 50% of chromosome 11. The primary physical map of chromosome 12 was constructed with a total of 414 clones, which were assembled into 54 large contigs, giving an approximate coverage of 52% of the whole chromosome. Physical maps for several gene-rich regions on chromosomes 3, 4, 5, and 9 have also been constructed. This physical map will be useful for mapbased cloning of important rice genes-QTLs and help in understanding the genomic structures of many chromosomal regions. Molecular cloning of xa5 A.C. Sanchez, B. Fu, D. Yang, G.S. Khush, and Z. Li The project goal is to clone a recessive gene, xa5, which confers resistance to bacterial blight (BB) by positional cloning. Rice genes conferring resistance to BB have been the targets of intensive investigation of the host and bacterial pathogen relationships in plants. Two of these resistance genes, Xa21 and Xa1, had been cloned. Xa21 was reportedly a transmembrane protein containing extracellular leucinerich-repeat (LRR) and a cytoplasmic kinase domain while Xa1 contains a nucleotide-binding site and an LRR. Unlike the dominant resistance genes such as
Table 1. Chromosomal distribution of anchor BAC islands. IRRI, 1999. Chromo- RFLP STS RAPD SSR Total some markers markers markers markers markers number (no.) 1 2 3 4 5 6 7 8 9 10 11 12 20 12 46 51 16 15 4 13 12 6 68 74 Total 337 16 12 7 11 14 12 9 8 8 5 13 5 120 7 8 7 2 6 9 2 5 4 34 5 89 43 32 60 64 36 36 15 21 25 15 127 96 570 Total clones (no.) 122 78 162 133 140 101 56 40 76 43 438 414 1803 Clones marker1 (av no.) 2.8 2.8 2.7 2.1 3.9 2.8 3.7 1.9 3.0 2.9 3.8 4.8 3.3
12 12
79
Xa21 and Xa1, our target gene for cloning, xa5, is recessive. This gene has been physically mapped and a single BAC clone carrying the gene identified. Information on the gene product of xa5 will help to further elucidate the molecular mechanisms behind specific host-pathogen recognition in plants.
PHYSICAL MAP POSITION OF XA5
The xa5 gene has been mapped to the distal end of the short arm of chromosome 5, tightly linked to three RFLP markersRG556, RG207, and RZ390. We used these three tightly linked markers as a probe to screen the IR64 BAC library by colony hybridization. Through fingerprinting of the BAC clones and chromosome walking, a physical map of the xa5 region of chromosome 5, consisting of 14 BAC clones, was established. The integrated physical and genetic map indicates that xa5 is probably in the region between RG556 and RG207.
IDENTIFICATION OF CANDIDATE CDNA CLONES
To check if xa5 is expressed in the homozygous resistance plants and to identify the candidate cDNA clone, an expression library from an incompatible BB strain (race 2, PXO 85)-induced mRNA of IRBB5 plants were constructed into the pSPORT1 (GIBCO BRL) vector. We dot-hybridized three
overlapping BAC clones (44B4, 9E8, and 28N22) with total RNA isolated from induced IR24 and IRBB5 and obtained a differential signal with 9E8 and 28N22. The cDNA library was then screened with the BAC clones 9E8 and 28N22. The clone 28N22 identified 14 cDNA clones while 9E8 hybridized to two clones. The common cDNA clone identified by both 28N22 and 9E8 was 5P2. Hybridization of 5P2 to the BAC clones revealed two strong bands (8 and 4 kb) and a weak band (2 kb) on 28N22 and only the 2-kb band on 9E8 (Fig. 1). Northern hybridization of 5P2 indicated a 5-kb RNA band, suggesting that 5P2 appeared to be constitutively expressed. The clone 5P2 was sequenced, revealing a complete insert length of 1.7 of 5P2. Database search of the 5P2 sequence using BLAST 2.0 shows that a great length (~1.5 kb) of the sequence is homologous to a rice retrotransposon while the remaining 200 bp has homology to some known kinases. Analysis using the GENESCAN program predicts a single peptide of 72 amino acids potentially encoded in the initial 218 bp of the 5P2 sequence. The PROSITE program identified five probable kinase domains within the 72-amino-acid-long peptide. The 8-kb subclone from the BAC clone 9E8 is being transformed into IRBB5 plants and sequencing of the whole 9E8 clone is in progress. We expect to clone xa5 in the near future.
1. Southern hybridization of BAC clones on the xa5 contig using 5P2 cDNA as probe. IRRI, 1999.
80
BLAST analysis of the terminal sequences of cloned markers from the genetic map of rice S. Constantino, A. Resurreccion, B. Albano, J.A. Champoux, C. Villareal, G.S. Khush, and J. Bennett Functional genomics seeks to assign biological function to the sequences of genes and intergenic regions. This assignment may be accomplished by a variety of approaches, including sequence comparison with known genes. We attempted to sequence 350 of the markers of the rice genetic map that was established principally at Cornell University. These data supplement the large database assembled from sequenced markers of the genetic map developed by the Rice Genome Research Project, Tsukuba, Japan. Terminal sequencing was conducted principally on both RZ clones (cDNA clones derived from RNA of etiolated leaves of IR36) and RG clones (genomic clones derived from IR36 DNA). Some barley (BCD) and oat (CDO) cDNA clones were also sequenced. Both ends of each clone were sequenced to allow polymerase chain reaction (PCR)-based amplification of longer DNA segments than is possible with data derived from the single-pass sequencing. The nucleotide sequences and the deduced amino acid sequences of each terminus were then compared with the Genbank databases using programs of the Basic Local Alignment Search Tool (BLAST). Table 2 shows hits with the BLAST-N program accessed through the Entrez Web site at www.ncbi.nlm.nih.gov. A total of 76 mapped clones, representing all 12 chromosomes of rice, recorded hits on known sequences. The termini of each clone were designated as F or R, depending on whether they were sequenced using the universal forward (F) or reverse (R) sequencing primer. As all inserts had been spliced into the vectors in random orientation, there was no relationship between the F and R ends and the direction of transcription of the gene. Most of the hits (71%) were among the RZ clones, consistent with these clones being from a cDNA library. Relatively few hits (24%) were found among the RG clones, which were random PstI genomic clones. The remaining hits (5%) were BCD and CDO cDNA clones. In 20% of the 76 clones registering hits, both termini hit on known genes, and in all of these cases, the same class of protein was re-
vealed, even if the name was different, as in the case of RZ244. For the remaining 80% of clones, only the F or R terminus hit on a known gene. The usual reason for the lack of homology at the other terminus was that the terminus corresponded to a poorly conserved region of the gene such as the 3untranslated region. This research was supported in part by grants from the Rockefeller Foundation and the German Federal Ministry for Economic Cooperation and Development (BMZ). Development and field evaluation of transgenic IR72, M.H. 63, and hybrid rice Shan You 63 with a Bt gene for stem borer resistance S.K. Datta, K. Datta, J. Tu, G.S. Khush, Q. Zhang,30 G.Y. Ye,29 M. Cohen, and Y.L. Fan19 The plasmid pFHBT1, which contains a hybrid Bt toxin gene made from synthetic cryIA(b) and cryIA(c) under control of rice ActinI promoter with its first intron, was used for rice transformation in this study. The recipient genomes included elite indica rice IR72 and Chinese cytoplasmic male sterile (CMS) restorer line Minghui 63 (M.H. 63). Both cultivars are restorer lines that have been used in rice production for more than a decade and are still considered among the best for their yield potential and combining ability. Rice transformation was done by the biolistic method. The transgenic plants obtained were evaluated on the different criteria such as DNA integration, protein production, and phenotypic acceptability. Homozygous transgenic lines were then evaluated in field plots in collaboration with Huazhong Agricultural University and Zhejiang University, China. The target insects included yellow stem borer, striped stem borer, and leaffolder. Molecular analysis and bioassay demonstrated that the hybrid Bt toxin gene made from the synthetic cryIA(b) and cryIA(c) was successfully integrated and expressed in the genome of both IR72 and M.H. 63. The typical expression levels of Btendotoxin detected in both rices was estimated to be 0.010.2% of the total soluble protein in leaf tissue or stem tissue. The bioassay results showed that larval mortality after feeding on cutting stems for 4 d reached 100% on all tested Bt-expressing plants.
81
82 Score Organism Chr Locus Putative identification Score Organism 231 Rice RZ244 R Lipoamide dehydrogenase 431 Soybean 5 5 5 6 6 6 6 6 6 7 7 7 7 8 8 8 8 RZ997 R 8 8 8 RZ572 F RG20 F RZ143 F RZ617 F RZ649 F RZ997 F RG156 F 565 194 468 748 372 236 155 160 323 233 665 792 1160 160 519 223 848 539 712 173 661 228 483 Mouse Arabidopsis Rice Rice Soybean Tomato Rice Rice Rice Rice Castor bean Maize Carrot Barley Maize DM virus Arabidopsis Azotobacter Wheat Rice Rice Maize Maize RZ390 F RZ455 R CDO580 R RZ508 F RZ508 R RG64 F RG64 R RZ140 F RZ2X R RZ405 R RZ682 R RZ488 F RZ488 R RZ509 R RG128 F Rice Tomato Human Rice Rice Rice Rice Tomato Human Wheat E. coli Rice Rice Barley Arabidopsis Rat virus Rice Wheat Rice Rice Rice Maize Arabidopsis
Table 2. DNA markers from the genetic map of rice with significant sequence homology to known genes. IRRI, 1999.
Chr
Locus
Putative identification
RZ566 F
Metallothionine
1 1 1 1 1 1
1 2
2 2
RZ566 R RZ836 F RZ836 R RZ588 R RZ730 R RZ744R RG345 F RZ382 R RG233 F RG233 R BCD828 R RZ204 F RZ204 R RZ643 F RZ643 R
Metallothionine +tubulin (RIP3) +tubulin (RIP3) ketoacyl-ACP synthase Ubiquitin conjugate enzyme B2 protein Peptidase of D1 protein Coat protein of dwarf mottle virus Enoyl-CoA hydratase FeS cluster assembly gene Mitochondrial ATP-2F1-ATPase Mitochondrial ATP/ADP translocator Mitochondrial ATP/ADP translocator Ferredoxin III Ferredoxin III
RZ962 F
Ribosomal protein L17a
2 2 2 2
RZ962 R RZ273 R RZ876 F RZ87 R
Ribosomal protein L17a Mitochondrial ATP/ADP translocator EF-tuf a ATP dependent protease ClpC
RG256 F
Plastidic aspartate amino transferase
Panicum
2 2
RG120 F RZ476 R
DNA repair protein EF-gamma
2 3
RG322 R RZ16 F RZ16 R RZ589 F
DNA binding protein Ascorbate peroxidase Ascorbate peroxidase +tubulin (RIP3)
161 231 887 348
Pea Rice Rice Rice
9 9
RZ698 F RZ698 R RZ792 F CDO590 R
Cytochrome b5 1035 GTP binding protein 184 Isovaleryl coA dehydrogenase 446 Catalase 1095 Catalase 516 Blast resistance clone 1051 Blast resistance clone 891 Homeobox protein 524 Human PRP8 protein 330 High affinity K transporter 475 Nodulation protein nodK 296 mRNA for thioredoxin h 383 mRNA for thioredoxin h 307 bpw1 water transport protein 135 mRNA for acyl-(acyl protein) 165 thioesterase Rat cytomegalovirus DNA 164 biosynthetic enzyme Cytosolic GADPH 655 Poly-A binding protein 680 Fructose diphosphate aldolase 336 Plastidic Cu/Zn superoxide 487 dismutase Plastidic Cu/Zn superoxide 974 dismutase Cytosolic APX 294 Meristem L1 layer homeobox 265 protein Plastidic Pi/PEP transporter 313 Plastid Pi/PEP transporter 641 Nucleic acid binding protein 269 Protein phosphatase 523
Maize Maize Maize Tobacco
RZ589 R RG754 F RG754 R RZ142 F 870
730 261 393 50
Rice 9 Maize 9 Ricinus 9 Caenorhabiditis 10 Maize 10
RZ206 F RZ228 F RZ 337 R RZ500 R RZ561 F
VDAC protein/porin HSP82 Suppressor-like protein Sugar transporter Cell division cycle protein cDc48 mRNA
325 650 431 379 393
Wheat Rice Arabidopsis Arabidopsis
RZ448 R
+tubulin (RIP3) Phosphoglycerate mutase Phosphoglycerate mutase Dolichol monophosphate mannose synthase Phosphoglucomutase
Arabidopsis
Barley Apple Rice S. chinensis Tobacco
Organism
Lotus Rice C. roseus
Panicum
Wheat Potato 216 484 Gluthathione-S-transferase Plastidic transketolase
Score
657 716 746 dehydratase Peptide transporter 176 Exon of ACC oxidase 158 Lipid transfer protein 1071 -ketoacyl CoA synthase 201 Dehydroquinate dehydratase 299
787
Alanine amino transferase
GTP binding protein Rubisco activase S-adenosyl homocysteine
125
Rice
2. Pest reaction of transgenic line T51-1 (right) and nontransgenic Minghui 63 control plants (left) to a heavy infestation of yellow stem borer. IRRI, 1999. EF-1 a
Organism
Arabidopsis
Rice Maize Arabidopsis
Soybean Arabidopsis Rice Rice Maize
Barley Arabidopsis
Arabidopsis Soybean
The data from the field trial confirmed that season-long protection from heavy infestation of Lepidoptera insects was obtained in both IR72 and M.H. 63 transgenic lines as well as the M.H. 63-produced rice hybrid (Fig. 2). This protection was achieved without affecting the combining ability of original restorer line M.H. 63 and its resultant hybrid Shan You 63. Twenty-eight percent yield advantage was recorded using Bt hybrid rice when compared with non-Bt hybrid rice. Evaluation of transgenic IR72 with Xa21 for bacterial blight resistance S. K. Datta, J. Tu, K. Datta, G.S. Khush, Q. Zhang,30 and T.W. Mew Gene Xa21 is known to confer resistance to most known bacterial blight (BB) races in India and the Philippines. We introduced Xa21 into the genome of IR72 via the biolistic method. The transformant obtained was designated as TT103. Molecular analysis of T0 and T1 plants of TT103 demonstrated that the intact coding sequence of Xa21 is present in the recipient genome, without rearrangement, and the inheritance of the transgene in the T1 generation fits the one-locus integration pattern. The T1 plants positive for the transgene proved to be highly resistant to prevalent races 4 and 6 of Xoo. Based on the characterization of resistance phenotype and molecular analysis, several homozygous lines carrying Xa21 were obtained from the transformed IR72. The homozygous line, TT10310, with the best phenotype and seed-setting was
RG247 F RG118 R RZ737 F RG218 F RZ261 F
10 10 11
11 11 11 12 12
10
12 12 RZ86 F RZ250 R 4 4 716 355
RZ397 F RG396 R
Score
Chr
194 181 216
484 254 135 811 227
542
UDPG dehydrogenase KAT C/KAT A kinesin EF-1 a S-adenosyl methionine decarboxylase GADPH (pseudogene)
Table 2. continued.
RG179 F RG418X F RZ993X F RZ740 F CDO1328 F
RZ585 F RZ614 F RZ574 R
Chr
3 3 3
3 3 3 4 4
4 5
RG449 F RZ244 F
RZ576 F
Locus
Cytosolic GADPH Molybdenum co-factor biosynthetic enzyme Serine/threonine protein kinase Ferric leghemoglobin reductase
Plastidic ribosomal protein
Histone H3 Ferredoxin III Cell wall protein
190 48
12
RZ993X F
RZ400 F RZ536 R RZ900 F
RZ892 R
Locus
83
Table 3. Resistance performance of transgenic homozygous line, TT103-10, against multiple races of Xoo in field trials at Huazhong Agricultural University, Wuhan, China, 1999. Plants tested (no.) 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 60 60 60 60 Lesion length (cm) 1.04 0.31 16.43 0.97 9.00 0.61 14.11 0.82 9.37 2.43 15.69 8.00 0.72 0.39 7.60 1.02 1.41 0.79 11.16 1.58 0.71 0.61 20.01 1.94 0.12 0.05 1.32 0.24 0.86 0.24 1.46 0.40 1.21 0.53 1.24 1.20 0.13 0.09 1.11 0.62 0.53 0.31 1.89 0.73 0.45 0.31 1.89 0.91 Performancea R HR HS HR S HR HS HR S R HS S HR HR S R R HR S R HR HR HS R
Xoo race
Philippine race 1 (PXO 61)
Variety IR72 TT103 IR24 IRBB21 IR72 TT103 IR24 IRBB21 IR72 TT103 IR24 IRBB21 IR72 TT103 IR24 IRBB21 IR72 TT103 IR24 IRBB21 IR72 TT103 IR24 IRBB21
Philippine race 6 (PX O99)
Chinese race 4 (Zhe 173)
Japanese race 2 (T2)
a R = resistant, S = susceptible, HS = highly susceptible, MS = moderately susceptible, HR = highly resistant.
repeatedly tested in the field in Wuhan, China, in 1998 and 1999 to evaluate its levels of resistance to BB. The races of Xoo used in the experiments were isolated from the Philippines, Japan, and China. The results demonstrated that the transgenic homozygous line expressed the same resistance spectrum, but with a shorter lesion length, to each inoculated race as Xa21 donor line IRBB21 (Table 3). Nontransformed control IR72 carrying Xa4 was resistant to Philippine races 1 and 5, Chinese race 4, and Japanese race 2 but susceptible to Philippine races 6 and 3. Negative control variety IR24 was susceptible to all races. The Xa21 transgene led to excellent field performance of the induced-BB resistance trait on the recipient plants. The yield performance of the transgenic homozygous line TT103-10 is comparable with that of IR72 in field plots.
Evaluation of transgenic rice with Xa gene for bacterial blight resistance T.W. Mew, S.K. Datta, A. Ona, and C.M. Vera Cruz A microplot experiment was done on the basis of a protocol approved by the National Biosafety Committee of the Philippines. The Xa21 gene in the transgenic line TT103 significantly reduced BB infection. Transgenic rice had a slow rate (r) of disease increment (r = 0.11, 0.07, and 0.17 caused respectively by races, 1, 3, and 6) when compared with nontransformed IR72 (r = 0.41, 0.41, and 0.51). Five rows of purple rice (recessive gene for purple color and highly susceptible to BB) were grown around each plot to determine the gene flow from transgenic to nontargeted rice. A pathogenicity test of purple rice grown from seeds of the first row (closest to transgenic plants) showed that all plants remained purple and were highly susceptible to BB, suggesting no evidence of gene flow in screenhouse tests.
84
NARS development of elite lines through breeding and marker-aided selection H. Leung, D. Brar, C.M. Vera Cruz, L. Sebastian,1 R. Tabien,1 D. Tabanao,1 S. Singh,14 M. Sodhi,14 G.S. Mangat,14 Y. Vikal,14 M. Bustamam,15 Suwarno,15 E. Lubis,15 A. Warsun,15 S. Christiyanthy,15 D. Agisimanto,15 and T. Kadir15 Current research in the Philippines, Indonesia, and India has focused on pyramiding genes for BB resistance to improve local commercial cultivars before release to farmers (Table 4).
PHILIPPINES
highest yield of 4.9 t ha1 was obtained from AR322-82-6, a progeny of IR64//IRBB5-21/PSBRc 14. In a wet-season (WS) replicated yield trial, 10 elite lines yielding 4-5 t ha1 performed well despite high disease and insect pressure (Table 4). Crosses aimed to combine the three genes (xa5, Xa21, and gene from O. minuta) into PSBRc 14, IR64, and BPI-Ri10 were made and will be advanced in 2000 dry season (DS).
INDIA
Molecular markers associated with xa5, Xa21, and the gene from Oryza minuta were used by the Philippine Rice Research Institute (PhilRice) in pyramiding genes for BB resistance in PSBRc 14, BPI Ri-10 and IR64. The markers were used in the initial plant selection during three backcrosses made in the three recurrent parents. Lines from six crosses were evaluated at three different stages of breeding, and the homogeneous lines were included in two yield trials. In a nonreplicated yield trial, 10 of 30 lines were as good as, or better than, the check PSBRc 28. The
A cross was made at Punjab Agricultural University (PAU) between local commercial cultivar PR106 and pyramid line NH56 carrying xa5, xa13, and Xa21. BC3F3 lines were developed from this cross by subsequent backcrossing with PR106. Three sequence-tagged site (STS) markers corresponding to xa5, xa13, and Xa21 were used to pyramid the genes into PR106. A combination of two or more genes was more effective in providing resistance to Punjab and Philippine BB strains than the individual genes. A PR106 pyramid line carrying xa5+xa13+Xa21 genes showed a high level of resistance to BB strains from both countries. BC3F3 lines of PR106 carrying individual genes and their combinations were evaluated at 31 farmers fields in Punjab for resistance to the predo-
Table 4. Elite lines carrying one, or a combination of, bacterial blight resistance genes developed by NARS institutes through marker-aided selection into popular commercial cultivars, 199899. Yield(t ha1)a NARS Cross combination Generation Genes 1999 DS Philippines BPI Ri-10 3*/IRBB5-21 BPI Ri-10 3*/IR59183 PSBRc 14 3*/IRBB5-21 IR64//IR59183/PSBRc 14 IR72 (control) PSBRc 28 (control) PR106 3*/NH56 PR106 3*/NH56 PR106 3*/NH56 PR106 3*/NH56 PR106 3*/NH56 PR106 3*/NH56 PR106 3*/NH56 Bio9-Mr-V-V/4-8-Pn-1 Bio9-Mr-V/4-5-Kn-5-1 Bio10-Mr-V1/1-1-Pn-1-1 Bio8-Mr-V/3-4/Pn-3-1 BC3F6 (4 lines) BC3F6 (4 lines) BC3F6 (1 line) BC3F6 (1 line) BC3F3 BC3F3 BC3F3 BC3F3 BC3F3 BC3F3 BC3F3 BC5F5 BC5F5 BC6F4 BC5F5 1999 WS 4.35.7 3.95.1 4.0 5.0 4.6 4.4
India
Indonesia
xa5, Xa21 O. minuta gene xa5, Xa21 O. minuta gene Xa4 xa5 xa13 Xa21 xa5 + xa13 xa5 + Xa21 xa13 + Xa21 xa5 +xa13 + Xa21 Xa7 Xa7 xa5 xa5
6.27.2 7.17.8 7.0 7.9 6.2 5.9
= no data available.
85
minant field population of BB. Replicated yield trials of the BC3F3 pyramided lines are in progress at PAU. BC4F1 of PR106/PR106 (xa5+xa13+ Xa21) were planted in the field to advance gene pyramids with xa5, xa13, and Xa21.
INDONESIA
Rice blast management through varietal diversification Zhu Youyong,16 Chen Hairu,16 Fan Jinghua,16 Wang Yunyue,16 Li Yan,16 Fan Jinxiang,16 Yang Shisheng,16 Wang Zonghua, C. Mundt, T. Mew, and H. Leung Increasing the genetic diversity of crop populations through deployment of multilines or mixture of varieties has been successfully used to manage cereal diseases, such as stripe rust of wheat and powdery mildew of barley. An Asian Development Bankfunded project, Exploiting Biodiversity for Sustainable Rice Pest Management, supported collaborative research by Yunnan Agricultural University and IRRI to diversify varieties in farmers field as a means to control blast. On-farm field trials started in 1997 found that interplanting glutinous varieties (Huangkenuo or Zinuo) with indica hybrid varieties (Xianyou 63 or Xianyou 22) significantly reduced the incidence and severity of rice blast. The field design was a repeating pattern of one row of glutinous rice interplanted with four or six rows of indica rice. Interplanting japonica varieties Hexi 41, Chujing 12, and Luxuan 1 did not result in disease reduction. Farmers in Yunnan adopted interplanting on 812 ha in 1998 and 3,342 ha in 1999. Yield and disease data obtained in 1999 from 15 demonstration fields at five sites in Jiangsui county are in Table 5. Disease incidence and severity were lower for varieties used when grown in mixture than when grown in pure stand. Reductions were remarkable in glutinous varieties Zinuo and Huangkenuo where control efficiency ranged from 83.8 to 96.5%. Total yield in mixtures was 0.80.9 t ha1 higher than with yield of pure indica rice. The effectiveness of interplanting to control blast seemed to be associated with the functional diversity of the component varieties. DNA fingerprints generated by polymerase chain reaction (PCR) primers corresponding to the conserved motifs of disease resistance genes, reveal a high degree of genetic polymorphism between the glutinous and the indica varieties but not among the japonica varieties used. The genetic relationships among the varieties interplanted are depicted in the dendrogram shown in Figure 3. Data from studies at IRRI also indicate that the polymorphism revealed by re-
The genes xa5, Xa7, and Xa21 were respectively introgressed singly in IR64 from the IR24 nearisogenic lines, IRBB5, IRBB7, and IRBB21. These genes were effective against the predominant population of BB pathogen in Indonesia. STS primers for xa5 and Xa21, and G1091 probe for Xa7 were used in selecting and confirming lines introgressed with the genes (Table 4). Field and greenhouse evaluation of the lines for resistance to diagnostic strains at Sukamandi Rice Research Institute and IRRI showed a high level of resistance when compared with the donor cultivars. Bio-xa5, the IR64 line carrying xa5, showed higher resistance to the diagnostic strain in the Philippines than to the donor cultivar due to quantitative complementation of xa5 from IRBB5 and Xa4 in the recurrent parent IR64. Both Bio-xa5 and Bio-Xa7 (IR64 line carrying Xa7 gene) are currently at BC5F5 generation and are comparable with IR64 in agronomic characters such as height, maturity, tillering ability, and amylose content, but are more resistant to BB. Lines carrying either xa5 or Xa7 have been selected by STS marker and DNA probe for seed purity for multilocation tests at different collaborating institutions and in farmers fields. Selected breeding lines of Bio-xa5+Xa7 and Bio-Xa21 are being evaluated for their resistance to the diagnostic BB strains in Indonesia. Incorporation of tungro resistance into Bio-xa5+Xa7 is currently under way.
Exploiting biodiversity for sustainable rice pest management

Rice pest management can be made more effective and sustainable by increasing biodiversity at the genetic, species, and community levels.
86
Table 5. Performance of varieties in mixture and pure stands during the 1999 cropping season in Jianshui County, Yunnan Province, China. Blast severity index Disease Yield of Combined Yield control components yield increase (%) (t ha1) (t ha1) (t ha1) 69.1 95.4 58.1 85.3 50.7 96.5 49.8 83.8 9.6 0.9 9.6 0.8 9.1 0.9 9.1 0.9 9.7 9.1 5.1 4.9 10.5 10.5 10.0 10.0 0.86 0.8 0.9 0.9
Treatment
Variety
Mixture Mixture Mixture Mixture Pure Pure Pure Pure
Xianyou 63/ 1.16 Huangkenuo 1.65 Xianyou 63/ 1.57 Zinuo 1.83 Xianyou 22/ 2.24 Huangkenuo 1.25 Xianyou 22/ 2.28 Zinuo 2.01 Xianyou 63 3.75 Xianyou 22 4.55 Huangkenuo 36.19 Zinuo 12.45
Xianyou 63 Xianyou 22 Zhinuo Huangkenuo Hexi 41 Chujing 12 8126 0.45 0.56 0.67 0.79 Coefficient 0.90
The interplanting approach presented an opportunity for Yunnan farmers to successfully grow glutinous varieties, which are highly valued but susceptible to blast, without high inputs of fungicide. The gain in yield and reduction in fungicide use translated to about US$189 ha1 improvement in farmers income during 1999. Interplantig is expected to extend to more than 100,000 ha by 2002. Field variants of tungro O. Azzam, K. Umadhay, F. Sta Cruz, and K. McNally The heterogeneity of field populations of both rice tungro bacilliform virus (RTBV) and rice tungro spherical virus (RTSV) was previously established using genetic markers. However, to examine the extent of sequence variability and the possible evolutionary relatedness among the different tungro isolates, an RTBV genome region (nt 7747-289), which includes the intergenic region and part of the open reading frame I, was amplified from nine different isolates and sequenced directly. In addition, the RTSV coat protein region 2 (nt 3070-3607) was amplified from 24 different isolates and sequenced directly. Three RTBV isolates from the Philippines, three from Vietnam, and two from Indonesia were examined. The published sequence of Serdang, a Malaysian isolate, was also included as a check. For RTSV, 13 isolates from Indonesia and eight from the Philippines were examined. The published sequence of two more isolates from the Philippines and one from Malaysia was included.
3. Dendrogram constructed using RGA profiles generated with the PCR primers XLRR, Pto kinase, and NLRR for varieties used in interplanting trials. Variety 8126 is a sister line of Luxuan 1, the japonica variety used in interplanting experiments and is genetically similar to Hexi 41 and Chujing 12. Jianshui County, Yunnan Province, China, 1999.
sistance gene analogs (RGA markers) correlates with the diversity in resistance to blast and is potentially useful as predictor of functional diversity for blast resistance. The mechanisms by which interplanting diverse varieties reduces blast disease may involve physical obstruction to the spread of inoculum on susceptible plant, inoculum density reduction, microclimate change, and induced resistance as a result of incompatible interaction. PCR fingerprints of blast isolates obtained from mixture and monoculture fields showed that mixtures tend to support diverse blast pathogen populations with no single dominant strain. However, pathogen populations in monoculture fields tended to be dominated by one or few dominant strains.
87
Sequence alignment and phylogenetic grouping were attempted on the sequences with the neighborjoining, parsimony, and maximum likelihood method of quartets. Sequence alignment was performed using CLUSTALW. Phylogenetic reconstruction by neighbor-joining and parsimony methods was conducted with programs from PHYLIP 3.5c, whereas the maximum likelihood analysis was accomplished by the quartet puzzling method of PUZZLE 4.0. Graphics of the trees were drawn using TREEVIEW 1.5. Phylogenetic analysis of RTBV sequences show three well-defined groups. The first group clustered the isolates from the Philippines, the second cluster had the Vietnamese isolates, and the third group the Indonesian isolates. The Malaysian isolate of Serdang, which was included as a check, did not cluster with any of the three groups. Bootstrap values for the parsimony analysis indicated that these three major groups were respectively observed in more than 70%, 54%, and 92% of the 1,000 trees determined. Phylogenetic analysis of RTSV coat protein region 2 sequences also showed three major lineages (Fig. 4). Most of the Indonesian isolates clustered in the first lineage with high significances found for the nodes in the bootstrap analysis of the puzzle tree (74%). The other two lineages also had high significances for nodes clustering mostly the Philippines isolates (47% and 88%). These results confirm the geographical distribution of both RTBV and RTSV populations but show a less discrete distribution because two RTSV Indonesian and Philippine isolates cluster in one lineage of RTSV puzzle 2 tree. Migration or gene flow seems to occur between some RTSV populations, while location-specific isolates were detected as well. The geographic isolation of tungro populations has implications for the deployment of tungro resistance. Disease outbreaks may be localized and the appearance of resistance-breaking strains in one geographic location may not mean that they will easily spread to other locations. Therefore, targeted deployment of relevant resistance genes to a particular environment may be effective.
Biological nitrogen fixation

J.K. Ladha, P.M. Reddy, P. Gyaneshwar,32 N. Mathan,33 S. Peng, B. Reinhold-Hurek,34 W.L. Barraquio,35 A.K. Tripathi,36 A. Kumar,36 F.B. Dazzo,37 B.G. Rolfe,38 and Y.G. Yanni39 The strategies for developing endophytic/symbiotic nitrogen fixation in rice include the establishment of effective endophytic associations and the development of legume-like nodulation.
97 60 Is72M 74 Ib121 Ib22VIII Is75VIII 23 Is73M 19 6.8 Is85M 52 Ib36V 93 Ib52V 57 Ib01V 75 76 Is81V PgA Pg08V 47 32 Is76IV Ib60M
100 47 5.2
41
Pc03Vl Pg09Vl PgVt6lll 94 Pc17lll 49 P71lll 90 M70lll Pc34l 64 lb42ll 88 43 Pc41ll Pe21ll
0.01
4. Puzzle tree for nucleotide sequence alignments for coat protein 2 of 24 RTSV isolates. The numbers above an internal branch are percentages of times a grouping occurred in 500 bootstrap replicates of the data. Isolate name includes the country (I for Indonesia and P for Philippines), the province (Ib for Bali or Is for Subang or Pc for North Cotabato, Pe for Nueva Ecija, Pg for IRRI-Philippines greenhouse isolates), the isolate field number, if any, and the coat protein genotype of that isolate. The bar indicates the branch length corresponding to a maximum likelihood distance of 0.01. IRRI, 1999.
88
Diazotrophic endophytic bacteria Predominant diazotrophic endophytic bacteria were isolated from surface-sterilized roots and stems as well as seeds of several rice varieties grown in nonsterile soil. Most abundant stem- and rootborne diazotrophs were identified by 16S rDNA sequencing and SDS-PAGE analysis. A dominant endophyte identified as Serratia marcescens represents a novel diazotroph and seems to possess alternate nitrogenase. The seedborne endophytes were grouped by fingerprinting using PCR with BOX primers. Some of these promising endophytes were further characterized by whole cell protein analysis and 16S rDNA sequencing, and identified as species belonging to Alcaligenes, Rhizobium, and Klebsiella. Several endophytes, including seedborne bacteria, showed the ability to systemically spread to various rice plant parts during plant growth. Seed transmission as an important mechanism of endophytic establishment has an advantage in the development of seed-based bacterial inocula. A novel DNA-based detection method to tag bacteria by using signature sequence for monitoring in field was developed. The detection method will allow the verification of the abundance of the inoculated bacterial strains in rice tissues and soils from various agroecosystems. Growth-promoting activities Six diazotrophs from a wide host range were investigated to determine their growth-promoting activities in the lowland rice cultivar Pankaj. Inoculation with Rhizobium leguminosarum bv. trifolii E11, Rhizobium sp. IRBG74, and Bradyrhizobium sp. IRBG271 respectively increased grain and straw yields by 822% and 419%. Nitrogen, P, and K uptake were increased by 1028% due to rhizobial inoculation. Iron uptake increased by 1564% at different N rates. Inoculated diazotrophs produced about 2 mg L1 indole-3-acetic acid in rice root exudate. It is likely that the increases in rice growth and nutrient uptake were due to growth-stimulating substances produced by rhizobia. Inoculation with Rhizobium sp. IRBG74 significantly and consistently promoted the single-leaf net photosynthetic rates in rice with different N treatments.
Working group meeting The third biological N2 fixation working group meeting was held at IRRI. Twenty-four research papers covering a wide range of topics on rice-endophyte and rice-rhizobial interactions and nif gene transfer were presented. The working group formulated future directions for research and recommended the inclusion of studies on rice-mycorrhizal associations as part of the program to understand the genetic predisposition of rice to form symbiotic associations. A project review by a panel of external experts was conducted during the working group meeting. The panel recommended q assessment of N2 fixation and plant growth promotion by endophytic diazotrophs in the laboratory, greenhouse, and field; q initiation of studies on rice-mycorrhizal associations with an aim to identify symbiotic genes common between legumes and rice, and functional genomics to characterize symbiosis-related genes in rice; q investigations to incorporate N2 fixation (nif) genes into rice; and q establishment of international steering committees to function as advisory bodies to help identify appropriate research on biological N2 fixation in rice.
Rice, a way of life for the next generation of rice farmers

This research acknowledges the growing knowledge gap between researchers and farming communities and seeks to bridge that gap. This is done through identifying and fostering partnerships with community development and extension organizations including nongovernment organizations (NGO), the private sector, farming organizations, and kinship networks. The main objectives are: q Improve rural incomes and reduce drudgery in rice farming through the introduction of appropriate engineering and knowledge-intensive crop management systems. q Understand the social, economic, and technical factors that influence knowledge acquisition and absorption, and develop communication mechanisms for faster adoption of technologies.
89
Using multisector partners to deliver IRRI rice technology to rainfed rice farmers in northeastern Thailand C. Edmonds, M.A. Bell, R. Raab, S. Morin, and J. Rickman Thailands Population and Community Development Association (PDA), Department of Agriculture (DOA), and IRRI are principal partners in a project in a rainfed area of Buriram Province. The objective is to improve understanding of institutional arrangements to enhance rural livelihoods. The project includes participation in and research about nontraditional partnerships in technology development and local adaptation. The participating organizations form a unit in which the DOA assists IRRI with the diagnostic and technical phase, and the PDA delivers the technology to farmer groups through existing community networks. Alhough PDAs services were focused on urban areas, it identified a need to set up programs in rural areas to improve the earnings of rural households and thereby preempt the migration of families to urban areas. PDA established community-based integrated rural development (CBIRD) centers in several towns throughout Thailand. A key component of the centers involves hosting of small private sector manufacturing facilities. The centers provide relatively high-paying manufacturing jobs to rural workersoften the children of farming families without requiring the families to leave the countryside. PDAs interest in increasing the agricultural income of rural families where rice is an important crop led to the development of the multisector partners project.
1998-99 ACTIVITIES
Project development involved IRRI scientists accompanying PDA and DOA staff members to meet with farmers. A series of problems were collectively identified (water stress, nutrient and weed management, blast, and timeliness of input availability) along with technological options that interested the farmers and appeared to offer potential to increase farm welfare. Work in 1998 involved families in two localities around Nan Rong and Lam Plaimat. Demonstration
plots were established on each participating farm and a combination of interventions (including land leveling, integrated weed management, improved nutrient management, and a new blast management technology) were applied. Early in the implementation of the project, a need for greater input in field technical support became evident and the active participation of DOA increased. PDA field staff and a few participating farmers were trained in the fundamentals of rice cultivation. Project field staff were provided with additional training and technical backstopping by IRRI and Thai DOA scientists. Training materials included a CD ROM-based rice pest compendium and instructional materials on proper rainfed rice crop management. Farmer field days just prior to harvest provided scientists and project staff members an opportunity to discuss the technologies and issues of concern with local farmers. The field days highlighted substantial farmer interest and identified additional problems of seed health and grain quality. The demonstration plots during 1998 had average yield increases of more than 840 kg ha1or 45% higher than average yields reported by farms for 1995-97. The average yield increase on the demonstration plots across all farms for 1998 was 83 kg ha1, a 25% increase. The yield increases varied across technologies applied and participating farms. Additional collaborators from the private sector joined the project in 1999. Word-of-mouth regarding results from the initial year, farmer field days, and advertising at the PDA-CBIRD centers substantially increased the number of farmers participating in the project. The interest of local private-sector firms in collaborating with the project increased as well. A number of political conflicts unfortunately emerged and required the attention of project leaders from the various organizations. The conflicts slowed progress for the project. In addition, early rains hampered land leveling, the most promising of the technological interventions tried. Despite setbacks, more than 100 farmers from the same two localities participated in 1998. Krung Thai Bank joined the project as a partner in 1999 and initiated a rice seed cultivation component. DOA training on nutrient and weed management and seed health continued.
90
LESSONS LEARNED (1998-99)
A number of issues emerged and lessons were learned through pilot projects in Bangladesh and Thailand. Those are summarized in Table 6. Post-production systems research
PHILIPPINE POST-PRODUCTION RESEARCH CONSORTIUM
Table 6. Issues and lessons learned during 1998-99 in projects in Bangladesh and Thailand aimed at on-farm identification and evaluation of technology and knowledge delivery in partnership arrangements. Issue Stability
q
Lessons learned It is vital that all partners have organizational stability (including financial, personnel, and institutional stability). Projects in Bangladesh and Thailand were slowed because of changes in key NGO staff. Good programs and powerful partnerships will fail without the financial resources for program activities. Programs require adequate resources to carry out activities and pay staff. An additional potential benefit of multisector partnerships may be an increase in the public profile of programs to assist the rural poor. This can help to build a coalition of support for these issues, therefore assisting fund-raising efforts. In implementing partnerships, it is important that project leaders talk to the leaders from each participating organization and ensure buy-in at the top, middle, and bottom. Fostering trust and commitment to the stated goals of the partnership is essential. It should be recognized at the outset that it takes time and shared experiences in working together to build trust and establish the confidence that each partner has the good of the project and each other at heart. Individual self-interest or egos can override any institutional commitment to partnerships. An individual or individuals (preferably on-site) from one of the organizations must assume a leadership role in the project. This person, or persons, should coordinate project activities, control quality, and ensure that deadlines and project commitments are met on time. There are considerable transaction costs involved in establishing and maintaining working relationships. Time is required for project staff to establish a certain level of trust through meetings and project implementation. Expect differences in the organizational cultures of partners. This can lead to misunderstandings and miscommunications in the early stages of the project.
Money
R. Bakker, D.B. de Padua, and M.A. Bell IRRI signed a memorandum of agreement in June 1999 with the four main Philippine agencies involved in rice post-production researchthe National Food Authority (NFA), the Bureau of Postharvest Research and Extension, the College of Engineering and Agrotechnology of the University of the Philippines Los Baos, and PhilRiceto form a research consortium. The objective is a comprehensive approach in rice post-production research, better communication of results, and an increase in the participation of technology end-users. Consortium staff members visited rice farmers, cooperatives, rice processing centers, and government grain handling facilities to identify problems and prioritize research activities. Case studies of successful and failed technology introductions defined an appropriate dissemination strategy for postproduction technologies. The consortium initiated a partnership with the Manufacturing Industries Association of the Philippines (MIAP) with the objective to localize production of needed post-production technologies in the Philippines. Consortium members will provide the know-how for technology design and development. MIAP members will produce prototypes for testing and eventually standardize parts for mass production under a subcontracting scheme. A 3-d mechanical dryer design workshop was conducted during which consortium engineers and members of MIAP reviewed suitability of different drying concepts. Four different mechanical drying systems were conceptualized, and a multi-agency dryer design group was formed. The first dryer design was transferred to MIAP in November 1999. The prototype, a 6-t recirculating batch dryer, will be field-tested early in 2000.
Institutional commitment
Project driver
Transaction costs
Private and public sector perspectives
91
Table 6 continued. Issue

q
Lessons learned These differences need to be considered in the design of the project and the strategy for carrying it out. For example, working with the private sector typically introduces a sense of urgency, which can be an advantage or disadvantage. When proposing and carrying out change, a common byproduct is an upset in the status quo and vested interests. Expect opposition to the project and attempt to engage potential critics. Project responses to criticism should consider the underlying motivation for the criticism. Also consider how to encourage many sectors or organizations on-board and how to avoid unfounded or unnecessary criticism, while responding to genuine concerns in developing partnerships. Recognize that opposition to change derives from several motivesopposition to change itself, opposition to the objectives of the project, disagreement with the approach applied, different agendas (e.g., political or individual attacks), or misunderstandings of project activities. Recognize that because multisector partnerships involve organizations with their own unique histories, prejudices may already exist on the part of some critics. Engage all the collaborating organizations in the development of project goals and its guiding principles. Project development should be conducted openly with proactive efforts to explain the project to potential critics. Project partners should define project milestones and set up a system for monitoring project implementation and performance.
Change and dealing with fallout
Develop project goals collectively, explicitly, and openly
rice that are based on consumer preferences are needed. We made an exploratory assessment of rice quality from selected retail markets in Laguna Province and Metro Manila, Philippines. The objectives were to evaluate the quality of the milled rice with respect to the existing grades and standards, and to make recommendations based on our quality assessment. Among 55 samples obtained from rice retailers, only 18 had grades indicated on the label or package. Milled rice quality in terms of head rice was comparatively low based on NFA standards, with the majority falling in Grades 2 and 3 (Fig. 5). Evidence of mislabeling was observed. The persistence of names of traditional varieties that are no longer grown suggested that consumers associate particular quality characteristics with these varietal names, standards, and grades. Results of our study were presented at two major national conferences, and recommendations were communicated to the appropriate government agencies. The research methodology we developed will be used in a nationwide follow-up study in which IRRI and NFA will collaborate. Drying is a key operation to maintain rice quality. As part of a study that explores drying characteristics of modern rice cultivars, a series of experiments determined the effects of drying air temperature on grain quality, including head rice, whiteness, and seed viability. A two-stage drying approach was used to optimize drying time while maintaining grain quality. Results indicated that
Price (P kg-1) 40 35 30
Grade 3 50-65%
Grade 2 65-80%
Grade 1 80-95%
Premium 95-100%
Rice quality management R. Bakker, D.B. de Padua, A.R. Elepao, R.D. Billate, I.R. Barredo, and M.A. Bell Quality of milled rice in many countries is low due to inappropriate management techniques in all aspects of grain handling. Changes in post-production management practices will occur only if there is sufficient economic incentive. In this respect, implementation of proper standards and grades for milled
25 20 15 10 40
Not graded Graded
50
60 80 70 Head rice (%)
90
100
5. Head rice and retail market price of 55 milled rice samples collected in Laguna province and Metro Manila, Philippines. IRRI, 1999.
92
Table 7. Drying rate, drying time, and head rice as affected by drying air temperature in thin layer drying of rough rice (PSBRc 54, first stage = 2518% moisture content, second stage = 1814% moisture content). IRRI, 1999. Drying air temperature (C) First Second stage stage 40 50 60 70 80 50 60 70 80
a
Drying rate (% h1) First stage 2.5 10.9 22.4 34.6 60.2 6.79 21.6 38.4 62.0 5.3 5.4 5.8 2.3 2.5 2.1 2.3 Second stage
Total drying timea
Head rice (%)
40 40 40 40 40 50 50 50 50
1.00 0.52 0.44 0.49 0.35 0.36 0.24 0.20 0.17
91.6a 90.1a 87.6a 88.5a 84.9b 79.3c 84.7c 84.2c 83.4c
We surveyed 28 rice breeding stations in South and Southeast Asia to collect information on q number of varieties released, q crosses involved in the released varieties, q breeding objectives, q pool of genetic materials used in breeding, and q most popular varieties adopted by farmers in the area served by the breeding stations. Rice breeders were asked to enumerate the genetic traits they sought from each parent in each of the crosses. Pedigree analysis identified genetic materials used in the development of multiple-trait varieties and determined the contribution of IRRI and NARS in the development of varieties widely grown by farmers.
PRODUCTION OF VARIETIES
Relative to continuous drying at 40 C.
higher drying rates can be used without compromising head rice recovery and whiteness, and result in reductions of drying time compared with single-stage drying (Table 7). However, results also showed that grain subjected to drying air of 60 C or higher lose their seed viability. This study provided useful guidelines for the design of mechanical dryers for rough rice.
Socioeconomic studies for technology impact, gender, and policy analysis

This research generates information to support planning and prioritization of rice research and to improve the understanding of the interface between the diffusion of new technology, socioeconomic equity (including the gender dimension), and alleviation of poverty. Production of improved rice varieties in South and Southeast Asia M. Hossain, V. Cabanilla, D. Gollin,31 G.S. Khush, and G. McLaren The Impact Assessment and Evaluation Group of the Consultative Group on International Agricultural Research (CGIAR) initiated a study on the economic evaluation of germplasm improvement research conducted by the CGIAR centers. The IRRI Social Sciences Division did the study on rice for South and Southeast Asia.
Since 1960, the NARS of South and Southeast Asia have released about 1,100 varieties of rice (Table 8). They have developed about 35 new varieties per year since the mid-1970s. Vietnams research system has increased its releases of new varieties dramatically since the 1980s. Laos released a number of varieties in the 1990s after a long period with few releases. Cambodias research system also increased the pace of variety releases in the 1990s. In all of these countries, the increased rate of releases corresponded to recent investments in the human capital for the rice research system, as well as to increased cooperation with the international scientific community. IRRI contributed directly to the germplasm improvement research in these countries through bilateral country programs.
NARS USES OF IRRI MATERIALS
Direct release of IRRI lines or varieties. Of the 1,132 released varieties in the data for which ancestry could be traced, 161 (14%) were IRRI lines released directly in other countries. Differences across countries are noted from the data. Vietnam, Cambodia, and Laos released many IRRI lines in recent years. In earlier years, the Philippines was a major user of IRRI-developed varieties. By contrast, Thailand did not release any IRRI-developed varieties. India and Sri Lanka had fewer than 10% of releases as direct IRRI-developed varieties.
93
Table 8. Number of improved rice varieties released, by country and by time period. IRRI, 1999. Country Varieties (no.) Pre-1970 1971-80 Bangladesh Cambodia India Indonesia Laos Malaysia Myanmar Pakistan Philippines Sri Lanka Thailand Vietnam Total 11 0 35 16 8 10 5 18 9 23 61 30 226 24 0 151 21 0 14 16 4 32 10 12 9 292 1981-90 19 6 202 24 0 12 27 4 16 13 15 35 373 1991-98 19 24 66 14 8 5 5 1 37 11 8 43 241 Total 73 30 454 75 16 40 53 27 94 57 96 117 1132 Riceland (million ha) 10.2 1.9 42.5 11.2 0.57 0.65 5.6 2.3 3.5 0.66 9.2 7.1 95.4 Varieties (no. million ha1) 7.2 15.8 10.7 6.7 28.1 61.5 9.5 11.8 26.9 86.4 10.4 16.5 11.9
0.45 0.40
0.25
0.21
0.42 0.39 0.32 0.24 0.31 0.39
0.35 0.30
0.15 0.14
0.20 0.15 0.10 0.05 0.00

0.10
0.18
0.25 0.20 0.15

0.04 0.05
0.10 0.05
0.05
Pre-1970 1971-75 1976-80 1981-85 1986-90 1991-95 1996-98
0.00
Pre-1970
1971-75 1976-80 1981-85 1986-90 1991-95 1996-98
Time period 6. IRRI crosses released as varieties, as percent of total releases. IRRI, 1999.
Time period
7. Releases with IRRI parents, as percent of total releases. IRRI, 1999.
IRRI-developed varieties appear to have reached their highest level in the late 1970s, when more than 20% of all releases in South and Southeast Asia were IRRI-developed lines (Fig. 6). This corresponds with the period in which the first modern varieties were developed with resistance to a number of important diseases and pests. Across countries, the proportion of IRRI-developed varieties is notably high in Cambodia, Myanmar, the Philippines, and Vietnam, accounting for about one-fourth to one-third of total released varieties. Uses of IRRI materials as parents. We estimate that about 44% of released varieties originated from one or more parents developed at IRRI. The number of IRRI-parents releases appears to have reached a peak in the 1976-80 period (Fig. 7), but there is evi-
dence for a later increase in the 1990s period due to IRRI participation in the germplasm improvement research in Cambodia, Laos, and Vietnam. Bangladesh had the highest proportion of releases in this category, with 48% of all locally bred varieties using at least one IRRI parent. By contrast, only 8% of Thai varieties had an IRRI parent. In a number of major rice-producing countries, the use of IRRI parents has decreased since the early years of the Green Revolution. In India, for example, 45% of varieties released during 1971-75 period had at least one IRRI parent (in addition to the 10% of releases that were wholly IRRI-developed). By the 1990s, however, only about 20% of Indian releases had an IRRI parent. This suggests a change in the respective roles of IRRI and the Indian national program.
94
IRRI provision of other ancestors. As NARS capacity grew, IRRI increasingly provided them with elite lines for use in breeding. Those were used as parents of released varieties but sometimes they appear as grandparents or more remote ancestors. This has begun to create a pool of genetic resources that have IRRI ancestry but not at the parental level. Excluding IRRI crosses and varieties with IRRI parents, this pool accounted for 78 released varieties (7% of the total). The varieties that fall in this category are of relatively recent origin (Fig. 8). The overall contribution of IRRI to the improved germplasm released by NARS can be seen from
0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00
0.00
Pre-1970 1971-75 1976-80 1981-85 1986-90 1991-95 1996-98
0.12 0.11 0.10 0.10
0.03 0.01
Time period
8. Percent of released varieties with IRRI ancestry other than direct release or immediate parents. IRRI, 1999.
Table 9. The contribution was highest in the Philippines, Bangladesh, and Indonesia and lowest in Thailand, Cambodia, and Laos, where farmers are still growing mainly traditional varieties. Types of IRRI materials in use. Are national programs today using IRRI materials that were already available a decade or two ago? To what extent are newer IRRI materials being used? To answer those questions, we identified the date of crossing for every IRRI material appearing in the genealogy of a released variety. We then asked how many of the released varieties make use of materials crossed at IRRI after 1990, after 1980, etc. The aggregated data for all 12 countries are summarized in Table 10. IRRIs first semidwarf varieties (IR5 and IR8) were crossed during 1962-64, although they did not reach a usable stage for breeding or multiplication until 1966. A second generation of IRRI materials was developed from crosses made in the period from 1965 to 1971. This included the widely used varieties IR20, IR22, and IR24, along with less widely used IR26, IR28, and IR29. Some of those varieties displayed useful insect and disease resistance, with several drawing on the hardy Indian variety TKM6. In our data set, 137 released varieties can be traced to the second-generation IRRI materials, but have no subsequent IRRI lines in their genealogies.
Table 9. Contribution of IRRI to released varieties in 12 countries of South and Southeast Asia. Numbers are proportion (%) of total releases. IRRI, 1999. IRRI crosses released as varieties Varieties with Varieties linked IRRI material with IRRI in the previous materials ancestors 0.10 0.03 0.07 0.11 0.00 0.07 0.00 0.00 0.04 0.26 0.03 0.05 0.07 0.70 0.33 0.44 0.67 0.38 0.50 0.64 0.48 0.71 0.53 0.11 0.67 0.50
Country
Varieties with an IRRI parent
Bangladesh Cambodia India Indonesia Laos Malaysia Myanmar Pakistan Philippines Sri Lanka Thailand Vietnam Total
0.12 0.27 0.06 0.20 0.00 0.13 0.30 0.26 0.35 0.04 0.00 0.32 0.14
0.48 0.03 0.31 0.36 0.38 0.30 0.30 0.22 0.32 0.23 0.08 0.30 0.29
95
Table 10. Distribution of released varieties by the generation of IRRI materials in their genealogies. IRRI, 1999. Period of release Without IRRI With IRRI material (%) crossed during material (%) Pre-1965 1965-71 1972-76 1977 and later 90.3 41.0 32.9 38.3 47.0 52.5 48.2 8.7 34.4 28.2 25.5 16.2 5.8 18.1 0.0 12.3 19.4 16.5 9.2 5.8 12.1 0.0 12.3 19.4 18.1 21.6 20.6 15.8 0.0 0.0 0.0 1.6 5.9 15.3 5.8
Pre-1970 1971-75 1976-80 1981-85 1986-90 1991-98 Total
In the succeeding years, disease and insect resistance continued as a major goal of IRRI breeding. A major innovation was the incorporation of Oryza nivara, conferring resistance to the grassy stunt virus into the IRRI breeding pool. Varieties based on that germplasm developed from 1972 to 1976 included IR32, IR38, and IR40. Another set of varieties with multiple disease resistance was represented by IR36 and IR42. This third generation of material provided the foundation for 20% of the varieties released in the 1980s and the early 1990s. The following generation of IRRI breeding material consisted of crosses made at IRRI from 1977 through 1980. Those incorporated the characteristics of shorter growth duration and improved grain quality. An additional 49 varieties in the data set include crosses made at IRRI in this period in their pedigrees. All but three of those were released after 1986, accounting for about 11% of all releases in the period. The total number of varieties based on this germplasm is not large, but is significant. One of the crosses made resulted in IR64, which is the most widely cultivated rice variety in the world. Only 17 varieties (2%) include crosses made at IRRI in 1981-98 in their pedigrees. Varieties in the data set from such relatively recent IRRI crosses were not effected. It is normal to see time lags of 10 years or more between the time a cross is made at IRRI and the time it results in a variety released by a national program. IRRI as a source of traits. IRRI materials were initially attractive to national breeding programs as a source of a single traitthe semidwarfing gene. However, IRRI varieties soon incorporated bundles of other useful traits and characteristics.
The number of landraces in the genealogy of a released variety offers a useful measure of breeding intensity or complexity. To assess the IRRI contribution to the overall bundling process, we measured the number of distinct landraces in the genealogies of released varieties and the extent to which that number is dependent on IRRI breeding. The average number of landraces per released variety is just above 6.4 for the whole data set. From a low of about two landraces released variety1 in the pre-1970 period, the average number increased to more than 11 for 1996-98. This implies a substantial increase in the genealogical complexity of succeeding generations of varieties and substantial genetic diversity within a single variety. In turn, this implies greater bundling of desirable traits. Of the 6.4 landraces in the genealogy of an average released variety, IRRI has contributed about 4.6, meaning that national programs are typically combining an IRRI variety with a single landrace (Table 11). IRRI has put together packages of traits that the national programs find useful. The IRRI rices are occasionally useful in their existing form and are released directly by NARS. But more often, the IRRI rices lack one or more locally important traits. In those cases, the NARS use IRRI materials as building blocks to get desirable bundles of traits. Farmer adoption of improved rices. By the early 1990s, nearly 75% of the rice area in Asia was growing so-called high-yielding or modern varieties. The rate of adoption varied across countries, often depending on the development of irrigation infrastructure. The adoption of modern varieties in East Asia, where most of the rice land is irrigated,
96
Table 11. Average number of landraces in released varieties. IRRI, 1999. Landraces per variety (no.) Landraces introduced through IRRI material (no.) 4.9 5.8 3.1 6.9 5.3 4.6 6.1 2.6 10.8 2.8 0.8 7.3 4.6 Landraces introduced independent of IRRI (no.) 1.7 1.1 2.0 2.5 0.9 2.6 1.3 1.3 1.2 3.8 1.3 1.1 1.8
Country
Bangladesh Cambodia India Indonesia Laos Malaysia Myanmar Pakistan Philippines Sri Lanka Thailand Vietnam Total
6.6 6.9 5.1 9.4 6.2 7.2 7.4 3.9 12.0 6.6 2.1 8.4 6.4
was complete in the 1970s. In Myanmar, Cambodia, and Laos, where rice is grown mostly as rainfed lowland, farmers still grew traditional varieties in the early 1990s. The Philippine rate is more than 90% and Indonesia is about 75%. In Bangladesh, farmers grow more than 700 varieties, with 40 modern varieties accounting for 60% of land use. IRRIs contribution to popularly grown varieties. National-level data on the actual use of specific varieties by farmers are typically lacking except for
Indonesia and Bangladesh. We created a list of popular varieties grown in several regions from village studies and sample surveys IRRI conducted in collaboration with NARS. For regions for which the information was not available, we requested breeders to report the five most widely grown varieties in the area served by their station. The popularly grown varieties are listed in Table 12. We did a pedigree analysis of those varieties to assess the relative contribution of IRRI vis--vis the NARS. IR64 is the most popular variety in the region and is grown in extensive areas in Indonesia, Philippines, and Vietnam and in the Indian states of Andhra Pradesh and Orissa. We estimate that IR64 is grown to nearly 13 million ha of rice land in Asia. IR8 and Jaya, which were released in 1960s, still remain popular among farmers in several Indian states and in Bangladesh. Other IRRI varieties among the top five popular varieties are IR36, IR42, and IR66. Mashuri, a variety introduced in the early 1960s, and Swarna, a selection from Mashuri in the early 1980s, are the most popular varieties grown in several Indian states and in Nepal. An earlier IRRI study similar to ours found that in 1975, Jaya (a semidwarf short-duration variety developed in India) was the most popular variety in Asia, followed by IR8. By 1982, IR8 was replaced by IR36 and was estimated grown on about 11 mil-
Table 12. Top five most popular rice varieties grown by farmers in 12 countries of South and Southeast Asia. IRRI, 1999. Country Bangladesh Cambodia India Andhra Pradesh Bihar Madhya Pradesh Punjab Orissa Karnataka Tamil Nadu Uttar Pradesh West Bengal Indonesia Laos Malaysia Myanmar Philippines Pakistan Thailand Sri Lanka Vietnam Popular varieties BR11, BR14, BR3, IR8, BR10 IR66, Kesar, Neang Minh, Phka Khney BPT5204,IR64, MTU5182, Swarna, BPT3291, Tella Hamsa Sita, Rajashree, IR36, Mashuri, Kanihar Safri, IR36, Swarna, Mahamaya, Kranti PR111, Pusa44, PR106, PR113, IR8 Swarna, CR1030, Mashuri, Sabiti, IR64 MTU1001, Jaya, IR64, IR20, IR8 Ponni (Mashuri), ADT36, ADT39, IR50, CO 43 Sarjoo 52, Pant Dhan 4, Mashuri, NDR118, NDR97 Swarna, IET5656, IET4786, IR36, Joya IR64, Cisadane, Membermo, IR42, IR36 TDK1, RD6, KDML 105, Dokmay, Mueng Nga MR84, MR167, MR77, IR42, Semerak IR13240, IR92224, Mashuri, IR5, Inmayebaw IR64, PSBRc 14, PSBRc 28, PSBRc 18, PSBRc 34 Super Basmati, Basmati 1385, KS 282, IR6, Basmati 198 RD6, KDML 105, SPR60, RD23, RD10 BG300, BG352, BG94, BG350, BG450 IR64, OM997, IR50404, IR56279, DT10
97
lion ha. By the early 1990s, the most popular position was taken over by IR64. In 1975, about 60% of the widely grown varieties were locally developed, 30% were introduced from IRRI, and 10% from another country. The situation has changed little since then. Our data show that 67% of the popular varieties were developed in the country, 29% came from IRRI, and only 4% originated from another country. We analyzed the distribution of the origin of all distinct ancestors that appeared in the genealogy of the widely grown varieties. Of the total ancestors, 45% originated from IRRI materials, 23% from India, 7% from Sri Lanka, and 4% each from Indonesia and Thailand. IR8 was the most frequently used IRRI material occurring in the ancestry of these popular varieties, followed by IR8608 (51%). Of the 24 IRRI materials that were found in the ancestry of 15% or more of these popular varieties, 11 were developed in 1960s and 13 in 1970s. Materials developed since then have had insignificant contribution to the development of the popularly grown varieties. Demand for specialty rices: implications for technology development M.A. Sombilla and M. Hossain Rice regarded as a top-quality variety in one country may be considered low-quality rice in another country. Quality is primarily classified based on grain length, head rice content, and method of milling. A secondary classification is based on amylose content, which accounts for degree of stickiness and aroma. Of the different types of rice, long- and medium-grain rice with intermediate amylose content (nonglutinous) dominates production, consumption, and trade.
CHANGING STRUCTURE IN FOOD DEMAND
for rice grows less than proportionately and such demand shifts to better quality rice. Similarly, as rice prices decline, consumers do not proportionately increase quantity consumed but shift to purchasing more expensive, better quality rice.
POLICY IMPLICATIONS
The evidence suggests that grain quality will become increasingly important in demand for rice. This has strong implications on future development of production and marketing technologies. Rice scientists need to continue focusing on the development of varieties that are able to yield high and at the same time have the potential to capture the price margin currently existing between the higher priced (high-quality) traditional varieties and the lower priced (low-quality) modern varieties. Iron- and Znenriched traditional varieties as well as those with high vitamin A content are being crossed with modern rice varieties to increase yields and improve resistance to insects and diseases. The availability of those varieties will prove valuable to millions of poor people in Asia and Africa. The absence of formal grades and standards of rice in the international markets adds to market inefficiency in the commodity. The proxy measurements used, based on physical quality (head rice recovery, length, and cleanliness), do not tell the entire story of quality preferences. A reliable method for variety classification that takes into consideration all characteristics, including cooking quality, is needed for rice trading to gain more credence.
Implementing ecoregional approaches to improve natural resource management in Asia

The focus of the ecoregional approach is conservation and management of natural resources to develop sustainable food production systems, taking into account socioeconomic factors in biophysically defined ecoregions. One focus of the ecoregional approach in natural resource management is to identify agricultural production systems that are sustainable and have least impact on the state of the natural resource base.
A notable pattern in rice consumption is that with growing income, people express preferences for higher quality rice once caloric needs have been met. With respect to incomes and prices, demand for better quality rice is more elastic than demand for quantity. This means that as income grows, demand
98
Market integration, agricultural diversification, and erosion risk in northern Thailand G. Trebuil, S.P. Kam, T. Boch, Charal ThongNgam, F. Turkelboom, and A. Begue In the fragile agroecosystems of mountainous areas in Southeast Asia, accelerated erosion by concentrated runoff is a major problem that limits the productivity and sustainability of cropping systems. Cropping systems are rapidly diversifying in many of the upland areas, often with less emphasis on upland rice and more on cash crops. The effect of changes in cropping systems on erosion risk is poorly documented. A study in the uplands of northern Thailand documented land use changes between 1990 and 1994 and assessed the impact of the changes on erosion risk at the field- and small-watershed level. The study site is the Mae Salaep watershed in the extreme northern region of Thailand. This watershed, which is farmed by Akhas highlanders, is at an advanced stage of diversification of agricultural production systems. Most of the cultivated fields are on slopes of 1040%, and sometimes as steep as 60%. Expansion of farmland into surrounding forest areas is limited due to a national forest protection law. As a result, the land is subjected to more intensive cultivation. A multiscale approach was adopted. We linked results of a 2-year field-level agronomic survey (1998 Program Report) with a geographic information system (GIS) analysis of the spatial distribution of land use changes in the study sites. From the detailed, field-level agronomic survey, we identified key indicators and critical thresholds of soil erosion.
CHANGES IN CROPPING SYSTEMS TO REDUCE EROSION RISK
A decrease in the production of upland rice, which is a subsistence and high erosion-risk crop. Limited expansion of irrigated rice on bench terraces (Fig. 9). Recent appearance of a diversity of horticultural and perennial cash crops as indicated by map legends in Figure 9.
EROSION RISK DECREASE
Table 13 summarizes the categories of erosion risk, based on critical thresholds of slope length and slope steepness, for the main cropping systems in the Mae Salaep watershed. Figure 12 shows the spatial pattern of erosion risk in the Mae Salaep watershed for the 1990 WS. Figure 13 shows the changes in erosion risk between 1990 and 1994, displayed in three-dimensional perspective in relation with the topography of the Mae Salaep watershed. The agricultural diversification process for the Mae Salaep watershed observed during the early 1990s tends to decrease the risk of land degradation caused by accelerated erosion. The cropping situations that are of high erosion risk in the watershed can be identified spatially for on-farm soil and water conservation interventions to decrease erosion risk in Mae Salaep fields.
Program outlook
A new functional genomics project will be in full operation in 2000. It provides the backbone for gene discovery that will drive germplasm improvement. Through the use of genomewide approaches, we aim at understanding complex biological pathways controlling desirable agronomic traits. We have built considerable genetic resources essential for assigning biological functions to DNA sequences produced by genome sequencing projects of rice and allied species. The utility of these genetic materials will be amplified with the application of gene array analysis. We will continue to build infrastructure and facilities to apply high-throughput, genomewide analysis of genetic stocks. We will strengthen the bioinformatics component with recruitment of new staff members. Emphasis will be on building phenotype databases that would link with international genome databases.
Cross-ecosystems research 99
Figure 9 shows the spatial distribution of farm lots and crops in the Mae Salaep watershed in the 1990 and 1994 WS. A number of features of the changes in the cropping systems during this 4-year period had the effect of reducing erosion risk: q A decrease in the size of fields, hence reducing their slope lengths (Fig. 10). q An increase in the share of fallow land, especially on steep slopes (Fig. 11).
100
1990 1994
North

North Uncultivated Paddy Swidden rice Maize Paddy/soy Sw. rice/maize Sw. rice/misc Sw. rice/hedgerow Sw. rice/sesame Maize/Sw. rice Maize/bamboo Bamboo Fallow Uncultivated Paddy Upland rice Maize Paddy/soy Soybean Groundnut Ginger Sesame Orchard Jackfruit Tea Bamboo Litchee Mango Tamarind Meters 537.81 Vetiver Fallow Fish pond
Meters
537.81
9. Spatial distribution of crops in the Mae Salaep, Thailand watershed in the 1990 and 1994 wet seasons. IRRI, 1999.
Percent of farmed area 60 40 20 0

0-0.5 0.5-1 1-1.5 1.5-2 2-2.5 2.5-3 3-3.5 3.5-4 More than 4 1990 1994
Field size (ha)
10. Distribution of field size in the Mae Salaep, Thailand watershed in the 1990 and 1994 wet seasons. IRRI, 1999.
Perennial crops Fallow 6% 10%
Perennial crops 7% Annual cash crops 23%
Annual cash crops 27% Subsistence crops 17% Subsistence crops 57%
Fallow 53% 1994
1990
11. Land use dynamics in the Mae Salaep, Thailand watershed between 1990 and 1994. IRRI, 1999.
Table 13. Matrix of erosion risk for main cropping systems in the Mae Salaep, Thailand watershed, 1990 WS. IRRI, 1999. Slope characteristics Angle (5) Length (m) Fallow perennial paddy paddy/soy Low Low Low Low Land use classes Cash crop with low value added Low Medium Medium Medium Cash crop with high value added Low Medium High High
Upland rice
<47 <47
<25 >25 <25 >25
Low Medium High High
101
Ban Mae Salaep Bon
250
500 meters
12. Spatial distribution of erosion risk in the Mae Salaep, Thailand watershed in 1990. IRRI, 1999.
Ban Mae Salaep Bon
250
500 meters
N
Changes in erosion risk Decreasing Stable Increasing Uncultivated
13. Changes in erosion risk in the Mae Salaep, Thailand watershed between 1990 and 1994, IRRI, 1999.
102
Our biotechnology project will be strengthened by investment in functional genomics research. The project will emphasize development of tools and products for immediate applications. Priority will be given to delivery of products that help remove key constraints in different rice ecosystems. We will also emphasize the development of productive varieties in heterogeneous and marginal lands where rice is often the only source of nutrients in peoples diet. The ARBN will continue to be our pipeline to deliver these products to NARS. Because of the inherent variability of pest-induced problems, there is a continual need to develop management tactics using sound ecological principles. We will foster strong disciplinary research in pest science to address issues central to sustainable pest management. In particular, we will continue efforts to understand the role of biodiversity in the prevention of pest outbreak in environments with different levels of production intensity. We will apply ecological principles to develop practices in farmers fields where sustainable pest management and economic impact can be achieved. Knowledge-intensive technologies will be increasingly important in rice farming. However, without effective means to transfer knowledge and technologies, there will be a delay in bringing improved technology to benefit farmers. We will fo-
cus on identifying and evaluating technology and knowledge delivery systems. We will continue to use case studies to determine the social, economic, and technical factors that determine successful transfer of technology and knowledge acquisition. Our project on socioeconomic studies will be closely linked to work on transfer of technology and knowledge acquisitions in that we are interested in understanding technology diffusion and impact. The primary goal is to generate information to provide the rationale for effective planning and prioritization of rice research in the near and long term. The unique feature of this project is to understand socioeconomic dimensions of technological impact on poverty alleviation. The use of improved germplasm will continue to be central to improving productivity and food security in marginal lands. We will analyze the implications of quality preference of rice and marketing on research priority and income generation for farmers. In research initiated as a systemwide initiative to address issues on natural resource management at a regional scale, focus will be on integrating methodologies and available data to develop practical recommendations on resource management policies relevant to rice production. We will concentrate on pilot areas to generate data for natural resource management in larger regions.
103
Research programs Rice genetic resources: conservation, safe delivery, and use
CONSERVATION OF RICE AND BIOFERTILIZER GENETIC RESOURCES 106 Germplasm and information exchange 106 Indonesia 106 Lao PDR 106 Madagascar 106 Malaysia 107 Myanmar 107 Nepal 107 Philippines 107 Swaziland 107 Vietnam 107 Genebank management 107 Germplasm characterization 108 Data management 108 Training 108 Conservation of biofertilizer germplasm 108 Biosystematic studies of wild rices 108 Dynamic systems of genetic conservation 108 DELIVERY OF GENETIC RESOURCES: THE INTERNATIONAL NETWORK FOR GENETIC EVALUATION OF RICE (INGER) 112 1999 INGER nurseries 112 Distribution of nurseries 112 Preparation of the 2000 INGER nurseries 112 Utilization of 1998 INGER entries 112 Data management 113 Processing and distribution of test materials for the Upland Rice Research Consortium and Breeding Network 113 Maintaining genetic diversity of blast inoculum in the International Rice Blast Nursery (IRBN) 113 THE INTERNATIONAL RICE INFORMATION SYSTEM 113 ICIS development 114 IRIS development 114 SEED HEALTH TESTING SERVICES 115 PROGRAM OUTLOOK 116
Rice genetic resources: conservation, safe delivery, and use
Applications of information technology for data management were important program activities in 1999. Improvements were made to the genebank data management system, the recently developed integrated data management system for the International Network for Genetic Evaluation of Rice (INGER) was installed on the Institutes intranet, and a database for the Seed Health Unit was implemented. Additional modules were developed for International Rice Information System (IRIS), and more pedigree data were added. A genetic resources home page was opened on the IRRI web site. This was the last year of the rice biodiversity project funded by the Swiss Agency for Development and Cooperation (SDC). Partners in national agricultural research systems (NARS) completed most of the germplasm collecting planned at the initiation of the project in 1995.
Conservation of rice and biofertilizer genetic resources

Germplasm and information exchange M.T. Jackson, B.R. Lu, G.C. Loresto, and S. Appa Rao IRRI received more than 3,300 samples of Oryza sativa and 41 samples of wild species for long-term conservation.
INDONESIA
A collecting trip to Irian Jaya resulted in 29 samples of wild species from MeraukeO. officinalis (9), O. longiglumis (7), O. rufipogon (6), and O. meridionalis (7). Eight traditional rice varieties were also collected. The collection of O. meridionalis, an annual diploid species considered endemic to Australia, was its first documentation in Irian Jaya. Seventeen wild rice samples and eight cultivated rice were brought to IRRI for long-term conservation.
LAO PDR
IRRI and Lao PDR collectors brought 2,402 samples of cultivated rice and 16 samples of wild rice to IRRI for long-term preservation.
MADAGASCAR
National personnel explored the provinces of the northeast coastal area (Toamasina, Vavatenina, Fenoarivo-Est, and Soanierana-Ivongo) in midJanuary. The varieties Marotia and Botojingo were collected.
106
Three upland varieties were collected in the field. Nine photoperiod-sensitive varieties were collected from on-farm storage. Populations of O. longistaminata were found along a river near the village of Anisobe, 20 km from Toamasina. Their panicles were sterile and only vegetative samples were taken. IRRI received 253 samples of cultivated rice collected during 1997-99. Seed samples collected in 1997 were multiplied in Madagascar.
MALAYSIA
PHILIPPINES
The Philippine Rice Research Institute sent 51 samples of O. sativa to IRRI for long-term conservation. They were collected from Surigao del Sur, Catanduanes, Quirino, and Palawan provinces.
SWAZILAND
IRRI received 17 samples of wild species and one cultivated rice from Swaziland.
VIETNAM
Researchers and extension workers from the Malaysian Agriculture and Research Development Institute, the Agricultural Research Center of Sarawak, and extension workers from Sabah collected 435 samples of cultivated rice from Sarawak, Sabah, and Peninsular Malaysia. Nineteen samples of wild species and six samples of weedy rice were collected from Terengganu and Kelantan areas of Peninsular Malaysia.
MYANMAR
Work in three Vietnam provinces in the north, one in the central plateau, and three in the south resulted in 157 varieties collected. Seeds collected in 1997 were regenerated in 1998 and processed for duplication at IRRI. Germplasm collected in 1998 was regenerated during 1999. Genebank management M.T. Jackson, F. de Guzman, R. Reao, and S. Almazan Germplasm multiplication activities during the 1999 dry season (DS) included 4,600 newly acquired samples of O. sativa and 508 samples of wild species. An additional 3,100 O. sativa accessions were rejuvenated with 71 entries transferred to the nursery screenhouse because they were either photoperiod-sensitive or had poor germination. Highquality seeds were harvested from all but four of those accessions. About 250 O. glaberrima accessions were also successfully rejuvenated. About 300 wild species accessions were rejuvenated in the nursery screenhouse. We added 5,278 accessions from the 1998 wet season (WS), 1998 DS, and 1999 DS harvests for long-term conservation in the International Rice Genebank Collection. The viability of about 10,000 seed samples from the Active Collection was monitored. The viability of more than 5,200 samples was determined before assigning new accession numbers and placing them in long-term conservation. A marked decline in viability was observed for the glutinous varieties, and some wild species accessions that had been stored for more than 6 years. A special study of glutinous varieties will be initiated to understand their poor storage potential.
Extension workers from Kachin, Kayah, Kayin, Rakhine, and Shan states and from Magwe, Sagaing, and Tanintharyi divisions collected 647 samples of cultivated rice during 1998-99. More than 45 wild rice samples were collected from Yangon and Bago divisions. IRRI received 649 samples of cultivated rice and eight samples of wild rice for long-term conservation.
NEPAL
Scientists from the Agricultural Botany Division, Nepal Agricultural Research Council, and an IRRI collector, collected wild rice species in central and eastern Nepal. Fifty samples of O. rufipogon, O. nivara, and some weedy types were collected. Three rare local rice varieties were also collected. Oryza rufipogon is less common in the central and eastern Nepal than in mid-western and western Nepal. It is found as large populations in swamps and fishponds with adequate water supply. All populations of O. rufipogon encountered appeared to be photoperiod-sensitive. Duplicates of the 53 samples collected were brought to IRRI for long-term conservation.
107
We distributed more than 6,000 seed samples in response to 169 from 26 countries. Furthermore we restored 1,189 samples to Bangladesh (700), India (487), and Surinam (2). As part of the existing agreement for security backup storage, 9,321 accessions were sent to the USDAs National Seed Storage Laboratory (NSSL) at Fort Collins, Colorado. More than 95% of IRRI-registered accessions are now safely duplicated at the NSSL. Germplasm characterization M.T. Jackson, R. Reao, and S. Almazan We completely characterized 2,600 accessions for vegetative and reproductive traits during 1999 WS. Panicles were taken for postharvest characterization. Almost 200 O. glaberrima and about 90% of the more than 2,000 O. sativa accessions from the 1998 WS planting were completely described for postharvest traits. Newly acquired wild species samples (370) were characterized for both vegetative and reproductive traits during their initial multiplication in the nursery screenhouse. Data management M.T. Jackson. A.P. Alcantara, and E.B. Guevarra We made several modifications to the International Rice Genebank Collection Information System (IRGCIS) to improve its operation, to generate reports on the status of the newly acquired samples, and to facilitate selection of samples for initial seed increase and seed management. Because the IRGCIS database was developed in Oracle v.7.2 and its application was in Developer 2000 v.1.2, we reviewed the system for Y2K compliance and made necessary modifications to upgrade the system to Oracle v.8 and Developer 2000 v.1.5. During 1999, we updated IRGCIS with information on newly received samples (22 batches with a total of 8,497 samples). Master files were upgraded with morphoagronomic characterization data from the 1995 WS (2,297 O. sativa accessions), and the 1996 WS (1,707 O. sativa and 197 O. glaberrima accessions). We responded to 36 requests from IRRI staff members and 29 requests from outside IRRI for information on conserved germplasm. The status of germplasm designated to the Food and Agriculture Organization (FAO) was reviewed and a revised list submitted to FAO.
Our data management staff improved several features of the Lao genebank data management system (LaoRIS). That work included automatic generation of different reports on germplasm conserved such as the lists of unique variety names and their meaning, and conserved germplasm by province and district. LaoRIS was updated with more than 2,400 newly collected accessions. Training G.C. Loresto, S. Appa Rao, B.R. Lu, and M.T. Jackson A 2-d workshop in Irian Jaya covered identification, field collection, and conservation of wild rice species. Fifteen workers from the agricultural extension offices of Wamena, Monokwari, and Merauke, the agricultural research station of Merauke, and from different government units of Jayapura participated. A Lao PDR scientist worked with Lao germplasm in long-term conservation at IRRI during a 3-wk on-the-job training in genebank and data management. Conservation of biofertilizer germplasm J.K. Ladha and T.S. Ventura During 1999, we supplied 10 Azolla, 45 N2-fixing bacteria, 28 Rhizobium, 11 blue-green algae, and 30 aquatic legume samples in response to 25 requests from 10 countries. We sent duplicate samples of the bacteria and Rhizobium collections at IRRI to the Philippine Network of Microbial Culture Collections. Biosystematic studies of wild rices B.R. Lu, M.T. Jackson, A. Juliano, and M.E. Naredo Intraspecific crosses of African O. longistaminata and interspecific hybridization between O. longistaminata and AA genome species from Asia, South America, and Australia resulted in more than 300 hybrids from more than 1,000 crosses and 30,000 pollinated spikelets. The crossability of O. longistaminata with other AA species was relatively high. However, germination of the hybrid seeds was low compared with those from intraspecific crosses of O. longistaminata and from interspecific combinations between other AA ge-
108
nome rice species. A post-fertilization barrier between O. longistaminata and other rice species is evident. The reproductive isolation mechanism of O. longistaminata is therefore different from other AA genome rice species studied. The biosystematic relationships of the O. ridleyi complex were studied through morphological analysis, inter- and intraspecific hybridization, meiotic analysis, and fertility determinations of the pa-
1. Meiosis of the interspecific hybrid between Oryza longiglumis (IRGC 106525) and O. ridleyi (IRGC 100821), showing full chromosome pairing (24 ring bivalents). IRRI, 1999.
rental accessions and inter- and intraspecific hybrids. Seed set of the crosses was largely variable among the combinations. Meiosis was normal in intraspecific hybrids of O. ridleyi and O. longiglumis, as well as in hybrids between O. ridleyi and O. longiglumis (Fig. 1), indicating a close genomic relationship between the two species. The fertility data showed a more complicated pattern (Tables 1 and 2), suggesting a geographic differentiation pattern of the populations of the two species. The taxonomy of the O. ridleyi complex must be revised to take into account the geographical distribution and associated variation patterns of populations within the complex. Morphological and randomly amplified polymorphic DNA (RAPD) data (Fig. 2) strongly support the geographic differentiation pattern of the O. ridleyi complex. The genetic diversity of O. meridionalis (35 accessions) was studied using RAPD and isozyme markers. O. meridionalis samples from Queensland have RAPD variation patterns distinct from those in samples from the Northern Territory and Western Australia, demonstrating the genetic diversity of the species across its geographic distribution in northern Australia. The herbarium of wild species was upgraded, the identity of the samples was confirmed, and a database in Microsoft Access 7 was developed to sup-
Table 1. Origin and pollen and panicle fertilities of O. ridleyi and O. longiglumis accessions. IRRI, 1999. Species and IRGC accession number Pollen stainability (%)a Panicle fertility (%)a
Origin
O. ridleyi 100820 100821 100877 101453 105366 105973 106028 106259 106471 O. longiglumis 100974 105147 105148 105562 106525
a
Thailand Thailand Thailand Malaysia Thailand Indonesia Thailand Papua New Guinea Malaysia
63.0 (56.169.9) 81.2 (60.597.2) 63.0 (10.891.6) 84.1 (83.697.0) 85.3 (78.691.1) 71.7 96.6 63.0 57.4
40.2 (29.347.0) 46.8 (30.961.5) 43.9 ( 5.166.6) 34.0 (16.758.5) 36.3 (19.257.4) 9.6 61.6 (46.779.9) 53.5 0.9
Indonesia Indonesia Indonesia Indonesia Papua New Guinea
69.8 46.1 84.0 61.2 76.3
(23.487.2) ( 3.293.4) (79.089.7) (50.966.7) (63.993.3)
39.5 21.4 59.8 67.9 70.4
( 1.260.8) ( 1.451.4) (55.264.4) (52.963.8) (52.179.2)
Range (%) in parentheses.
109
Table 2. Pollen and panicle fertilities of intra- and interspecific hybrids of O. ridleyi and O. longiglumis. IRRI, 1999. Combination Pollen stainabilitya (%) Panicle fertilitya (%)
O. ridleyi O. ridleyi 101453 105366 105366 106259 106259 105366
88.1 (72.597.4) 1.0 ( 0 2.0) 2.7
50.9 (40.563.8) 1.6 (0.67 2.5) 0.2
O. longiglumis O. longiglumis 105147 106525 81.4 (67.692.8) 106525 105147 80.6 106525 105148 71.1 (40.290.3) O. ridleyi O. longiglumis 105366 105147 105366 106525 106259 106525 O. longiglumis O. ridleyi 105147 106028 106525 100821 106525 100877 106525 101453 106525 105366 106525 106259
a
59.0 (51.070.6) 66.4 51.4 (40.964.2)
10.1 (1.021.0) 4.8 1.4 (02.8)
3.9 (1.09.0) 0.6 0
11.9 13.9 7.0 5.6 9.5
(0.118.9) (3.427.5) (2.410.5) (0 13.0) (3.120.2) 10.7
4.6 ( 0.8 8.8) 20.4 (17.722.2) 26.3 (16.231.3) 15.0 ( 9.518.4) 4.4 ( 0.219.2)
Range (%) in parentheses.
port the biosystematic studies and conservation of the wild rice collection. Dynamic systems of genetic conservation J.L. Pham, S.R. Morin, M. Calibo, M. Belen, and S. Quilloy The continuation of genetic and socioeconomic studies in India, Vietnam, and the Philippines led to a more detailed description and analysis of the rice diversity on-farm and its management by farmers. We analyzed the genetic diversity of 179 accessions (78 variety names) from 16 central Vietnam villages. Surveys had shown the sets of varieties grown by farmers for DS and WS are different. In DS, traditional and modern varieties respectively contribute three and nine specific alleles (i.e., alleles not found in the other category), but in WS, the reverse is true with traditional and modern varieties respectively contributing nine and three specific alleles. However, when only the five most frequently grown traditional or modern varieties are considered, we observed that WS varieties had only three specific alleles contributed by modern varieties vs 11 by the traditional ones. The contributions
of traditional and modern DS varieties were identical (five specific alleles). These results demonstrate that traditional and modern varieties contribute specifically to the overall genetic diversity. This is particularly true for the varieties grown during WS, the main rice-cropping season. It also shows that the diversity contributed by the modern varieties in DS is threatened because it is essentially contributed by rare varieties. Onfarm conservation strategies should take into account both traditional and modern varieties. In India, the analysis of the distribution of varieties across eight villages on the Bastar Plateau showed that most modern varieties were locally distributed. They were present in a small number of villages. Moreover, they were grown by a small number of farmers per village (less than three out of the 14 farmers per village surveyed). On the other hand, traditional varieties may be widespread or only locally distributed. When locally distributed, they may be grown by few or by many farmers. This reflects the fact that specific varieties are limited in their distribution because they fulfill critical functions within given situations. Others have a more general distribution because
110
100820_R/T 100820_R/T 100820_R/T 100820_R/T 100821_R/T 100821_R/T 100821_R/T 100821_R/T 101453_R/M 101453_R/M 101453_R/M 101453_R/M 106028_R/T 106028_R/T 106028_R/T 106028_R/T 100877_R/T 105366_R/T 105366_R/T 105366_R/T 105366_R/T 100877_R/T 100877_R/T 100877_R/T 100974_L/I 105146_L/I 105146_L/I 105146_L/I 105562_L/I 105562_L/I 100974_L/I 100974_L/I 100974_L/I 105148_L/I 105148_L/I 105148_L/I 105148_L/I 105146_L/I 105562L/I 105147_L/I 105147_L/I 105147_L/I 105147_L/I 106525_L/P 106525_L/P 106525_L/P 106525_L/P 106259_R/P 106259_R/P 106259_R/P 106259_R/P 105973_R/I 105973_R/I 106471_R/M 106471_R/M 106471_R/M 106471_R/M 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Similarity coefficient
2. A UPGMA dendrogram generated from cluster analysis of 15 accessions of Oryza ridleyi (R) and O. longiglumis (L), from Indonesia (I), Malaysia (M), Papua New Guinea (P), and Thailand (T), based on 135 RAPD markers (including the monomorphic and polymorphic bands). IRRI, 1999.
111
they perform better across environments. Farmers choose specific or general varieties based on several criteria, such as soils, hydrology, and slope. Most farmers grow between two and five varieties. We observed no significant farm-level correlation between landholding size and number of varieties grown. The number of varieties grown is also highly variable within a class of landholding. However, landholding is a limiting factor to the number of varieties grown by a farmer. We also observed that large landholders tend to grow fewer varieties. Preliminary results from the microsatellite analysis of 26 varieties using 15 primers showed their high diversity, and 49 alleles were detected. This confirms the analysis of agromorphological traits of more than 300 samples that revealed an impressive continuum of diversity for vegetative and reproductive traits. Overall, these results demonstrate the need to define on-farm conservation strategies based on a thorough assessment of the agronomic, genetic, and socioeconomic criteria. We conceptualize two different approaches to on-farm conservation: 1) make diversity a viable option for farmers and 2) strengthen farmers access to diversity. Research examined the two approaches. In the Philippines, we designed a new cropping pattern for the traditional Wagwag varieties in order to shorten their duration. We implemented trials in the Cagayan Valley to check whether a delayed planting date would permit rainfed-lowland farmers to grow Wagwag varieties without losing the opportunity to grow modern varieties during the following cropping season. We initiated an experiment in Hu, central Vietnam, to test a new strategy for on-farm conservation. The first generation of three composite populations was multiplied, with two objectives: 1) build genetic pools to represent and help preserve the overall genetic diversity grown in this region, and 2) involve farmers in the conservation of genetic resources. The three populations will be split into subpopulations to be grown by farmers. This approach to on-farm conservation is expected to be dynamic in terms of evolution, and additionally increase farmers exposure and access to diversity.
Delivery of genetic resources: The International Network for Genetic Evaluation of Rice (INGER)
1999 INGER nurseries E.L. Javier, S.W. Ahn, C. Toledo, V. Lopez, and R. Reao Four hundred and thirty-one breeding lines from 41 NARS and 189 lines from IRRI, the International Institute of Tropical Agriculture (IITA), the International Center for Tropical Agriculture (CIAT), and the West Africa Rice Development Association (WARDA) were organized into five ecosystem-oriented and four stress-oriented nurseries. The stressoriented nurseries were composed to screen for cold tolerance and resistance to bacterial blight, brown and whitebacked planthopper, and gall midge.
DISTRIBUTION OF NURSERIES
Two hundred and thirty-nine nursery sets were distributed to 30 countries and evaluated at 129 test sites. Seventy-six percent of INGER trials were conducted by 22 NARS in Asia. Cooperators in Africa (Nigeria, Senegal, and WARDA), Latin America (Bolivia, Brazil, Surinam, and CIAT), and Europe (Italy) tested the remaining nursery sets. Rice scientists from 32 countries requested 358 breeding lines and varieties. NARS and international agricultural research centers nominated 786 breeding lines for evaluation in 15 nurseries. They were multiplied during the 1998-99 DS to be included as test materials in 2001 INGER nurseries.
PREPARATION OF THE 2000 INGER NURSERIES
Observational nurseries for four ecosystems (irrigated, rainfed lowland, upland, deepwater) and screening sets for tungro, blast, and problem soils were processed for evaluation in 2000.
UTILIZATION OF 1998 INGER ENTRIES
Numbers of INGER entries used by participating NARS for yield testing and hybridization in 1998 are in Table 3. Three hundred and ninety-two entries from the ecosystem-based nurseries and 283 from the stress-oriented nurseries were used as par-
112
Table 3. Utilization of 1998 INGER nursery entries by participating NARS. Entries utilized (no.) Region and country Yield testing Hybridization
Processing and distribution of test materials for the Upland Rice Research Consortium and Breeding Network B. Courtois, S.W. Ahn, and M. Laza We processed 124 upland rice breeding lines from Brazil, Colombia, France, India, Cte dIvoire, Thailand, and IRRI for evaluation at key test sites in China, India, Indonesia, and Vietnam. Maintaining genetic diversity of blast inoculum in the International Rice Blast Nursery (IRBN) S.W. Ahn Effective evaluation of blast resistance requires maintenance of a diverse blast race with different genetic backgrounds at a test site. The use of a mixture of broadly susceptible rice cultivars is recommended for spreader rows on which the blast pathogen can multiply and spread as the inoculum source. We monitored the population dynamics of the blast pathogen over 11 cycles of sequential plantings of spreader rows of different compositions (Table 4). The sequential planting technique permits monitoring of the long-term interaction of rice cultivar and blast population. We found that a mixture of several broadly susceptible cultivars could sustain a high diversity of blast pathotypes. A mixture of several host cultivars for a distinct blast lineage also maintained a high diversity of lineagepathotype combinations. However, use of only one susceptible cultivar as the spreader row did not support a wide range of pathogenic diversity. Therefore, we recommend use of a mixture of susceptible cultivars at sites where there is no genetic information on blast population structure.
East Asia China Taiwan (China) Southeast Asia Cambodia Indonesia Malaysia Myanmar Philippines Thailand Vietnam South Asia Bangladesh India Nepal Pakistan West Asia and North Africa Egypt Iran Turkey Latin America Colombia
48 15
54 70 10 82 16 61 10
4 29 1 35 32 30 13
38 81 9 13
22 83 8
16 16 5
42 3
ents in the varietal improvement programs of 13 countries. Sixteen NARS evaluated 491 nursery entries in advanced yield trials.
DATA MANAGEMENT
INGER data were managed in several databases before 1999. During 1999, we developed the INGER Information System (INGERIS), with applications running in Oracle Version 8, on a dedicated server. Three modules that manage data on processing incoming seeds, the seed inventory, and nursery data processing were completed and are used routinely by staff members. Modules for seed multiplication, nursery composition, and making queries were also completed. Pedigree and trial data are managed within the Data and Genealogy Management Systems of the IRIS.
The International Rice Information System

C.G. McLaren We continued to refine the dynamic link library (DLL) for the Genealogy Management System (GMS) to ensure compatibility with Microsoft Access and Oracle database systems, and have also developed a functional DLL for the Data Management System (DMS). DLLs contain the programs that allow access to the International Crop Information System (ICIS) data in any database system. The ICIS installation program was improved and modi-
113
Table 4. Genetic diversity of blast pathogen population of Pyricularia grisea after 11 cycles of sequential plantings of mixed rice cultivars in different combinations. IRRI, 1999. Treatmenta Level Hg Lineage Pathotype Lineage-pathotype Total isolatesc
a b
1
b
2 Hs 0.8 2.2 2.3 Hg 0.5 4.9 3.2 Hs 0.6 2.7 1.4 Hg 1.0 3.8 4.5
3 Hs 1.0 2.4 2.6 Hg 1.1 4.2 4.8
4 Hs 1.1 2.5 2.6
1.1 3.1 3.5
17/28
9/27
18/30
15/28
1= IR50, 2 = mixture of broadly susceptible varieties, 3 = six host cultivars, 4 = nine cultivars. Genetic diversity indices: Hg = Gleason index; Hs = Shannon index. cIsolates used for lineage analysis/isolates used for pathotype analysis.
fied to make it more reliable and easier to use. It was also modified to install the TexasInternational Wheat and Maize Improvement Center (CIMMYT) interface for the ICIS CD release. ICIS development The plant breeding interface for ICIS comprising SETGEN and FLDBK was redesigned and improved at IRRI in response to feedback from breeders. The new versions are easy to use, are quick, and link to familiar Microsoft Excel outputs as well as customized field books. A prototype ICIS Workbook was developed at IRRI as a DMS input and query tool, consisting of a series of Excel macros that access the DMS DLL. A GMS input program was developed at IRRI to download large sets of breeding cross histories and accession lists. This tool facilitates the capture of historical data in the development of new ICIS implementations. It was used to link IRIS to the Systemwide Information Network for Genetic Resources (SINGER) and to form a base implementation for maize (International Maize Information SystemIMIS) by downloading CIMMYT maize accession genealogy from SINGER to ICIS. The GMS Browse program was improved to allow calculation of coefficients of parentage, tracing the use and distribution of genetic resources, and analysis of contributions of different sources to particular lines. IRRI hosted an ICIS Workshop for 43 participants from CIAT, CIMMYT, the International
Center for Agricultural Research in the Dry Areas (ICARDA), the International Centre for Research in Agroforestry (ICRAF), IITA, IRRI, and the International Service for National Agricultural Research as well as NARS representatives from China, India, Korea, Philippines, Australia, France, and the USA. We verified that the DMS can store data from molecular studies, and a start was made on the design of the Gene Management System to handle unique identification, nomenclature, source, location, and function of genetic elements, including molecular polymorphisms, sequences, and traditional genes. IRIS development Further capture of historical pedigrees has enlarged IRIS to more than 700,000 lines. Breeding records from Thailand (20,000), Malaysia (7,573), and Pakistan (4,939) were added. CIAT breeding records (5,000) were updated as well as IRGC and GRIN rice accessions. Considerable effort went into verifying and correcting pedigrees for released varieties as part of the Economic Evaluation of Germplasm Improvement Impact Study. There are now records on 3,000 released varieties from 72 countries. Historical characterization and evaluation data from the Plant Breeding, Genetics, and Biochemistry Division and the Genetic Resources Center were loaded into the IRIS DMS155 studies including the hybridization block data, observational yield trial data, and replicated yield trial data from 1977 to 1996. Characterization data for 65,315 genebank accessions have also been loaded, and response data from
114
two INGER nurseries (International Rice Soil Stress Nursery 1980-91 and International Irrigated Rice Yield Nursery-Early 1984-96) were added to DMS. IRIS is fully deployed in the irrigated rice breeding program at IRRI and is used by the Upland Perennial Rice Breeding project. It is also being deployed in the Rainfed Lowland Shuttle Breeding Program. The features of IRIS that allow full participation in data sharing are ideal for management of data from a shuttle-breeding program. However, the number of nodes in the network and their distribution make training and logistics difficult. NARS partners in Cambodia, China, India, Indonesia, Laos, Philippines, and Thailand are interested in using IRIS. Researchers from all those countries have been trained in various aspects of its use. China, Korea, Philippines, and Thailand have returned historical genealogy data and some countries are now setting up local databases for use in managing breeding programs.
Seed health testing services

T.W. Mew S.D. Merca, P.G. Gonzales, C.C. Huelma, and J.O. Guevarra The Seed Health Unit processed 59 incoming seed shipments from 22 countries for post-entry clearance (Table 5). Their consignment is shown in Table 6. Post-entry examination showed that 32 shipments were contaminated with quarantinable objects such as soil, insects, and weed seeds. Contamination with weed seeds was only 0.6%, mainly Echinochloa sp. and Ischaemum rugosum. Seed lots affected by insects were 11.3%, mainly by Sitophilus granarius and S. oryzae. Seed health tests
showed that Bipolaris oryzae affected 66% of the seed lots, followed by Microdochium oryzae (40%), Fusarium moniliforme (22%), Sarocladium oryzae (19%), Tilletia barclayana (14%), Aphelenchoides besseyi (2%), and Pyricularia oryzae (1%). The prescribed ASEAN standard seed treatments were applied to all incoming seeds. Incoming seeds were multiplied only in the designated post-entry area on the Experiment Station and were monitored for crop health. Almost 13,000 entries were inspected for preexport certification, of which 12% were affected with leaf scald, 6% with tungro virus, and 6% with bacterial leaf streak. In WS crop selection, seedling blight affected 11% of the entries, leaf scald 73%, and bacterial leaf streak 25%. Crop health inspections were made on 8,035 entries in the DS and 3,339 in the WS. The most common DS disease was sheath rot, affecting 34% of the entries, while bacterial leaf streak affected 23%. No diseases were seen on 46% of the entries. Fewer diseases were observed in WS than in DS.
Table 6. IRRI consignees of imported rice seed shipments in 1999. Division APPAa EPPDb GRCc INGERd IRGe PBGBf Total Shipments (no.) 5 3 10 13 28 59 Seed lots (no.) 1,249 329 78 3,491 7,062 12,209 Weight (kg) 15.9 10.2 22.1 120.8 117.1 286.1
a Agronomy, Plant Physiology, and Agroecology. bEntomology and Plant Pathology. cGenetic Resources Center. dInternational Network for Genetic Evaluation of Rice. eInternational Rice Genebank. fPlant Breeding, Genetics, and Biochemistry.
Table 5. Origin of incoming seed shipments to IRRI, 1999. Region Countries (no.) Shipments (no.) 19 8 18 1 4 5 4 59 Seed lots (no.) 3,552 92 7,969 15 110 344 127 12,209 Weight (kg) 34.5 40.9 166.4 0.9 2.7 36.1 4.6 286.1
East Asia 3 South Asia 4 Southeast Asia 7 Latin America 1 North America 1 Sub-Saharan Africa 3 West Asia and North Africa 3 Total 22
115
Table 7. Distribution of rice seed shipments with phytosanitary certification by the Seed Health Unit, IRRI, 1999. Region Countries (no.) 5 6 9 11 8 2 1 8 4 54 Shipments (no.) 101 97 95 44 16 23 10 20 20 426 Seed lots (no.) 14,517 13,496 14,774 1,933 1,743 10,614 2,563 2,628 4,908 67,176 Weight (kg) 425.8 483.1 487.4 133.0 57.1 156.2 15.8 62.1 155.0 1,975.5
East Asia South Asia Southeast Asia Europe Latin America North America Oceania Sub-Saharan Africa West Asia and North Africa Total
Phytosanitary certificates were issued for 426 rice seed shipments covering 67,176 seed lots (1,975.5 kg) sent to 54 countries worldwide (Table 7). Routine seed health testing of 9,440 nontreated, outgoing seed lots revealed that S. oryzae (sheath rot) affected 58% of the seed lots, followed by F. moniliforme (41%), B. oryzae (36%), M. oryzae (10%), T. barclayana (1%), P. oryzae (0.7%), and A. besseyi (0.3%). All exported rice seeds were cleaned for quarantinable objects, tested for health, and treated with prescribed ASEAN standard seed treatment for ricehot-water 52-57 C/15 min. This was followed by fungicide slurry treatment with benomyl and mancozeb both at 0.1% by seed weight, except for countries that do not allow seed treatment. Fumigation with phosphine was also administered to all outgoing seeds. A database for the management of the Seed Health Unit operations was developed in MS-SQL 7.0. Researchers from Bangladesh (5), Cambodia (1), China (1), Indonesia (1) Lao PDR (1), Thailand
(1), and Vietnam (1) were trained in rice seed health for crop management.
Program outlook
Satisfying demand for information on germplasm and its characteristics and performance will be a major challenge for the next year. We are working to place the genebank database, IRGCIS, and IRIS on the World Wide Web to permit interactive access to germplasm data. We are evaluating different ways of improving the distribution and testing of germplasm through INGER. The SDC-funded project will terminate in mid-year, and much of the germplasm collecting in many countries will be completed by that date. Our biosystematic and genetic diversity studies continue to provide a framework for management and use of the wild species. Seed health testing protocols will continue to be modified in response to demands from plant quarantine authorities around the world.
116
Accelerating the impact of rice research
STRENGTHENING PARTNERSHIP WITH NARS 120 Institutionalization of rice research in Cambodia 120 Rice research in the Lao PDR 121 DELIVERY OF KNOWLEDGE-INTENSIVE TECHNOLOGIES: CROP AND RESOURCE MANAGEMENT NETWORK (CREMNET) 122 Paddy drying practices and evaluation of a simple dryer 122 Bangladesh 123 India 123 Myanmar 123 Farming systems technology in Cambodia 123 COLLECTING, EXCHANGING, AND DISTRIBUTING KNOWLEDGE AND INFORMATION ABOUT RICE 124 Scientific publishing 125 Public awareness 125 IRRI visitors, exhibitions, and conferences 126 Library and documentation service 126 HUMAN CAPITAL DEVELOPMENT OF NARS RICE PROFESSIONALS 126 Degree and postdegree training 126 Language proficiency development for scholars and trainees 126 Development and implementation of short-term group courses 129 Development of new training methodology 129 Training materials development 130 Collaborative in-country training 132 Regional courses 132 National courses 132 PROGRAM OUTLOOK 132 International Programs Management Office (IPMO) 132 Training Center 132 CREMNET 132 Access to IRRI's media assets 133 Computer Services 133
The impact of rice research on low-income rice producers and consumers is our foremost concern. The ability to make positive and lasting contributions to poverty alleviation, food security, and sustainable management of natural resources depends on two key factors: q quality and relevance of research, and q effective evaluation, adaptation, and delivery of research products to clients/users. Accelerating the Impact of Rice Research (IM) takes results from research programs, works those into usable technologies, evaluates the technologies, and adapts them to specific rice-growing conditions in collaboration with national agricultural research systems (NARS). Technical support and infrastructure strengthening are provided to NARS. IM has four projects: q Strengthening partnership with NARS q Delivery of knowledge-intensive technologies: Crop and Resource Management Network (CREMNET) q Collecting, exchanging, and distributing knowledge and information about rice q Human capital development of NARS rice professionals
Strengthening partnership with NARS

The need for strengthening of NARS in Asia is declining and IRRI focuses on technical support and training in strategic research areas for the advanced NARS. For NARS at an earlier stage of development, IRRI provides technical support for strengthening national research infrastructure, with funding derived mainly from bilateral sources. Institutionalization of rice research in Cambodia H.J. Nesbitt Agricultural research in Cambodia received renewed attention over the past decade. Thirty-four new rice varieties were released, an agronomic based soil identification system was developed, and recommendations extended to improve soil productivity. Pest prediction tables and control recommendations using integrated pest management (IPM) techniques adapted to local conditions were released. Farming systems techniques, including agricultural engineering, to improve overall farm productivity were developed and extended. These research programs were implemented within a framework of improving resources, especially in human capital. The Cambodia-IRRI-Australia Project (CIAP) played a significant role, training 1,128 individuals between 1987 and 1998, with each person attending an average of three courses. Thirteen project participants either completed or are in the process of fulfilling the requirements for PhD or MS degrees in Australia. Concurrent improvement in the physical infrastructure of research stations over the past 10 years
120
facilitated the installation of research trials. That infrastructure, and some research, was assisted by a number of nongovernment organizations (NGO) and the United Nations. All agencies work in close collaboration with the Cambodian Ministry of Agriculture, Forestry and Fisheries (MAFF). About 200 of 11,500 MAFF staff members work in agronomic research under the Department of Agronomy (DOA). Their research is supported by a central administrative office, a soils laboratory, and a plant protection center. The DOA has 14 research centers that cover a range of crops. A new agricultural research and extension policy was adopted by the MAFF in August 1999. Research and extension became separate functions of MAFF and will soon have separate budget allocations. An Agricultural Research Council (ARC) will oversee the direction and monitoring of research institutes. Research institutes will be semi-autonomous with a charter of responsibilities under the direction of the ARC. Technical departments will be responsible for production, statistics, regulation, surveillance, compliance, and technical backstopping for the provinces. The newly formed Cambodian Agricultural Research and Development Institute (CARDI) was designed to fit into the new system to take primary responsibility for rice-based farming systems research in addition to overseeing upland crops, forage crops, and horticulture research. The research priority-setting process had considerable activity during 1997-99. Setting a charter under which CARDI will operate was initiated and the charter will be developed further in 2000 to specify the nature and scope of the institute. CARDI will receive support from q government (staff and laborer salaries, operating costs); q foreign donors (infrastructure, salary support, technical assistance); q international institutes (technical assistance, collaborative research funds, knowledge and information databases); and q private sector (contract research, levies and research equipment). CARDI and other semi-autonomous institutes will receive overall direction from the ARC and through MAFF and a Board of Directors.
CARDI now has the physical and personnel resources plus the institutional support to carry itself into the 21st century. The institute is expanding its funding base to cover operational costs for another decade. Rice research in the Lao PDR J. Schiller Rice is the single most important crop in the Lao PDR, occupying more than 80% of the cropped area and accounting for 80% of the caloric intake of people in rural areas. Most of the planted area is rainfed, which makes rice production highly susceptible to climatic variations. Significant levels of rice deficit exist in some areas, particularly in the upland environment. The Swiss Agency for Development and Cooperation (SDC)-funded Lao-IRRI Project was initiated in 1990 to address rice research, development, and training. Almost 95% of the rainfed lowland and 100% of the uplands were being planted to traditional varieties and there was no defined national rice research program. The level of fertilizer inputs was also low. Yields in the lowlands often did not exceed 2 t ha1. By the end of 1999, with the cooperation of the Ministry of Agriculture and Forestry and provincial agriculture and forestry offices, a rice research network had been established throughout the country. Seven research centers serve the central, southern, and northern agricultural regions. Research activities are coordinated by the National Agricultural Research Center (NARC) located in Vientiane Municipality. At the end of 1999, more than 130 coordinators, scientists, and technicians were collaborating in the national rice research network and a National Agricultural Training Center had been established at NARC. The national rice research network includes varietal improvement, germplasm conservation, soil fertility management, plant protection (IPM and weed research), agronomy, seed technology, and integrated production systems research. The allocation of resources and related research emphasis is divided among the three rice environments at about 60% for rainfed lowland, 25% for irrigated, and 15% for upland rice.
121
Improved rice varieties (with mostly IRRI parentage) contributed to more than 80% of production in the rainfed lowland in 1999. All Lao improved varieties are glutinous. Soil fertility research has developed fertilizer recommendations for the rainfed lowland and irrigated environments. Farming systems research has demonstrated the benefits of adopting an integrated package of technical recommendations in rainfed lowland. Training has been a major part of the Lao National Rice Research Program. By the end of 1999, more than 1,070 places had been provided in various types of in-country training, including English language studies. Lao language versions of IRRI nondegree training on Basic Rice Research and Farming Systems Research were developed and adopted to local needs. A total of 197 NARC staff members were supported in various kinds of nondegree training and special work experience at IRRI and in third countries during 1990-99. Five research staff members either completed or are in the process of completing their MS studies. Constraints still being faced in relation to the sustainability of the Lao National Rice Research Program include the need for q greater Lao leadership of the program; q improved systems for monitoring, evaluation, and review; and q development of a greater critical mass of trained staff such that the effect of the movement of individual staff on the research programs is minimized. The upland component of the research program has yet to develop technical recommendations that can achieve the objectives of food security, reduction of income disparities, and ecological stabilization of upland areas. Further development of the national extension services is also essential to maximize the potential benefits of research to small-scale producers.
Delivery of knowledge-intensive technologies: Crop and Resource Management Network (CREMNET)

Rice production technologies are becoming more location-specific, complex, and knowledge-intensive as crop intensification achieves high yields (810 t ha-1) in farmers fields. CREMNET is designed to facilitate the identification, free exchange, participatory evaluation, and promotion of promising rice farming technologies. Paddy drying practices and evaluation of a simple dryer V. Balasubramanian, A.C. Morales, M.A. Baqui,20 Myint Wai,40 and S. Hooper41 Drying of harvested rough rice (paddy), especially during rainy season, is a major problem for rice producers in many Asian countries. More than 80% of farmers sun-dry paddy on open floors with high losses in quantity and quality of grain. Many farmers lack drying floors and use public roads to dry paddy. Farmers need a simple, low-cost dryer to dry paddy economically and efficiently. The University of Agriculture and Forestry (UAF), Vietnam, participated during 1994-95 in a technical cooperation project with IRRI and the donor agency, German Agency for Technical Cooperation, Germany, to develop a low-cost, low-temperature dryer called the SRR-1. It can dry 1 t of paddy in 23 d, with the same unit used as storage. Targeted users are resource-poor farmers with less than 0.5-ha rice land and who need to store 12 t paddy each season. The dryer consists of a 1/2-hp, two-stage axial fan, a 1000-watt electric heater, and a drying bin of bamboo mat or other material. The drying bin has two concentric bamboo-mat cylinders of 0.4 m and 1.5 m diameter, with a height of 1.1 m. The inner cylinder is supported by a frame made of 6-mm steel wire. A 0.5-hp, single-phase, 2800-rpm electric motor drives the fan. Two 350-mm-diameter, seven-blade plastic rotors are mounted on both ends of the motor shaft and enclosed in a steel-wire casing. The fan is positioned on top of the inner bamboo-mat cylinder. At 40 Pa static pressure, the airflow is 0.3 m3 s1.
122
The heater is a 1000-w resistor from an electric stove, mounted beneath the motor. Supplemental heat from the resistor is used at selective times e.g., cool nights or during continuous rains. The SRR-1 dryer costs US$55 in Vietnam, and the calculated drying cost is US$6 t1. We surveyed the paddy drying practices in selected areas in Bangladesh, India, and Myanmar in a collaborative project with UAF and NARS. Local engineers and technicians, with the help of UAF engineers, built and commissioned a prototype dryer in each area and collected users feedback for followup action.
BANGLADESH
tion on hard mud-plastered surfaces near the homestead or sometimes on public roads. The SRR-1 dryer was installed on a farm at Sivaranthakam village and the drying of wet paddy was demonstrated to about 100 farmers assembled from nearby villages. One SRR-1 unit was given to the LAM Research Institute, Andhra Pradesh, for testing and demonstration to farmers in the coastal districts of Andhra Pradesh where drying is a serious problem for rice harvested in wet season.
MYANMAR
Selected Bangladesh villages of Sylhet and Chittagong districts were surveyed. About 54% of the sample farmers were educated above secondary school level and 83% had the national medium level of income. Only 27% of total income came from agricultural sources. Almost all farmers had a landholding of less than 2 ha and mean rice yield was 3 t ha1. All farmers used the traditional methods of rice harvesting with sickles, manual threshing, and sun drying. About 65% of the farmers reported a drying loss of 13%, and 95% expressed the need for a simple mechanical dryer. Respondents stored paddy for 17 mo. Most farmers (82%) sold their paddy in different lots 14 mo after harvest. Pilot testing of the SRR-1 dryer was conducted at the Bangladesh Rice Research Institute station, Gazipur. The cost of the locally manufactured dryer was US$117 and the drying cost was US$6.57 t1 paddy. The head rice recovery was higher for paddy dried in the dryer than for sun-dried paddy. Farmers, extension workers, farm machine manufacturers, and rice millers witnessed the dryer demonstration. Ten dryers were produced in 1998 for tests and demonstrations in 1999 and 2000.
INDIA
A survey covered six villages from five townships in Yangon and Bogo divisions. Among farmers in the study area, landholding ranged from 2.4 to 5.8 ha, with a mean of 4.1 ha. Farm income constituted 87100% of the total income. Harvested paddy is kept on the threshing floor for 510 d before threshing, a practice that can adversely affect grain quality and head rice recovery. About 70% of the farmers use hired mechanical threshers. The machine threshing requires about 4 h ha1 at a cost of US$10 ha1. Traditional sun-drying of rice is predominant. About 1315 workers ha1 are employed for sundrying at a cost of US$6.857.17 t1. The estimated drying loss is 3.04.3%. The dried paddy is stored in bamboo bins for 3-5 mo. The dryer was demonstrated to farmers, engineers, and technicians of the Agricultural Machine Manufacturers Cooperative. The farmers preferred a high-capacity, village-level dryer. Moreover, electricity to operate the SRR-1 dryer is not available in villages. Farming systems technology in Cambodia C. Phaloeun, T. Vuthy, S. Sakkhunthea, and H.J. Nesbitt Constraints to rice production in Cambodia are related to infertile soils, poor on-farm water control, pest incidence, and low-yielding indigenous varieties. The CIAP has developed soil fertility improvement recommendations for each soil group, appropriate land leveling and on-farm water conservation techniques, and environment-friendly pest control techniques that decrease crop damage and reduce production costs.
Three villages in Pondicherry State, India, were surveyed. Farmers faced difficulties in drying rice harvested in wet weather in September. Many farmers sell excess paddy immediately after harvest. Most of the farmers sun-dried paddy for home consump-
123
Project plant breeders have released a range of varieties suitable for most of the rice-growing environments in Cambodia. Farming systems agronomists have combined these technologies into packages to evaluate in farmers fields. A study during 1996-99 compared the technology packages with farmers practices on recession (partially irrigated) and rainfed lowland rice grown in a series of soil types and water depths. CIAP technology increased yields in the rainfed lowland ecosystem by as much as 140% based on comparisons with neighboring farms. Farmers received an average 69% additional net income by using CIAP technology if the cost of the land leveling was not included and if loans were not necessary to purchase inputs. It is, however, necessary for the farmer to increase the level of purchased inputs to adopt the technology, thus increasing economic risk. Farmers were surveyed twice during the growing season to monitor farm activities, labor use, and cash flows. Cash values were placed on home-produced and -consumed products by surveying their potential value at local markets. Off-farm income was calculated on a family basis and, in some cases, included the sale of gold or other valuables. Expenditures were composed of farm, off-farm, and household and hired labor. Family farm labor was not charged to the budget. Major farm activities of all farms were cultivation of rice and attending to cattle. Time was also spent on caring for other animals, nonrice crops, and catching fish in rice fields and nearby streams. Four of the farms surveyed raised fish on their properties. Working persons constituted just over half of the family members and each worked on the farm for an average 100 d y1. Farm sizes ranged from 0.9 to 3.8 ha with an average of 1.7 ha. Household sizes were 1.5 persons above the national average of 5.5, making it difficult for the smaller farms to produce surpluses for sale. The 12 farms averaged net cash incomes of $254 in 1 year, with income ranging from a loss of $31 to an income of $914. Two-thirds of the farms received off-farm incomes. These were usually from off-farm labor or small businesses. Farm income was regularly derived from the sale of rice but sale of pigs or other animals, including fish, was the major cash income source.
Except in those cases where the farmers had small businesses, household costs consumed a major proportion of the cash income. Only a small amount of money was spent on purchasing farm inputs each year. Surplus funds were generally used for medicine or school fees. Where off-farm activities were minimal, net income was extremely small or negative. Nine of the 12 farm families were highly dependent on farm-grown products, foraging, and fishing. The average value of food produced or gathered or caught near the farm exceeded the net cash income by about 50%. The study emphasizes the high degree of risk the farmers in southeastern Cambodia face if their cash income or subsistence product levels are destabilized. It is important that one farm component is not improved at the risk of decreasing overall farm productivity. The development of technology suitable for these farmers must take into consideration the low levels of household cash income and the fact that they are dependent on home-produced goods for survival. Acceptance of new technology for many farmers will be a gradual process with the low-cost technologies being adopted first and different approaches being taken to reduce input costs. Other farmers may opt to work off-farm to purchase the technology they believe will provide sustainable farm productivity increases.
Collecting, exchanging, and distributing knowledge and information about rice

G. Hettel, B. Hardy, D. Macintosh, S. Inciong, M. Ramos, M. Movillon, and staff IRRIs mandate requires that the Institute publish and disseminate research findings and promote current rice research knowledge to scientists worldwide. Activities include q publishing and disseminating high-quality scientific materials about rice; q promoting IRRI and rice research through public awareness activities; q maintaining the IRRI Library and Documentation Service as the worlds foremost collection of materials on rice; q maintaining and further developing the Riceworld Museum and Learning Center; and
124
taking advantage of new information delivery technologies, particularly in the area of electronic communication media, to speed the exchange of information among rice scientists worldwide.
Scientific publishing IRRIs four web sitesthe IRRI Home site (www.cgiar.org/irri), Riceweb, Riceworld, and the IRRI Library sitecontinue to grow in popularity. There were nearly 120,000 visitors to our web sites during 1999. They made more than 500,000 hits or movements within the sites. The web sites were improved by addition of electronic versions of q 1999 issues of the International Rice Research Notes (www.cgiar.org/irri/irrn.htm), q the 1998 Program Report (www.cgiar.org/irri/ 98programreport.htm), q INGER-derived rice varieties (www.cgiar.org/ irri/INGERforeword.html), q popular discussion papers, q the Medium-Term Plan 2000-2002, q IRRIs monthly newsletter Sandiwa (www.cgiar.org/irri/Sandiwa.htm), and q abstracts of recent IRRI conference and workshop proceedings (www.cgiar.org/irri/ absidx.htm). More than 200 direct links to useful web sites were added to the Library web site. Improvements continued to the in-house IRRI Intranet. Several Intranet sites, created by IRRIs divisions, centers, units, programs or research networks, were elevated to the external Internetgenetic resources (www.cgiar.org/irri/geneticresources.htm), Asian Rice Biotechnology Network (www.cgiar.org/ irri/ARBN/arbnindex.htm), and the Systems Research Network for Ecoregional Land Use Planning in Tropical Asia (www.cgiar.org/irri/sysnet/index.htm). Credit card sales of IRRI books continued successfully on the Internet via the German vendor TRIOPS. Twenty-five titles were produced and distributed, including six IRRI books, eight installments of the IRRI discussion paper series, six installments of the new technical bulletin series, and one installment of the new limited proceedings series. One of the books, A rice village saga: three decades of green revolution in the Philippines, was a triple im-
print with Macmillan Press Limited and Barnes & Noble. A bibliography is provided in the section on publications and seminars near the end of this program report. The 25-year-old International Rice Research Notes (IRRN) was given a fresh new format with expanded features and sections. More than 1,100 h of historic videotapes were indexed, work that will continue into the IRRI still photography collection in 2000. With an appropriate search engine, IRRI staff and clients will then be able to search for available photos and videos on CD. Staff members in Communication and Publications Services received one Outstanding Professional Skill Award, three Gold Awards, and one Bronze Award at the 1999 meeting of the U.S.based Agricultural Communicators in Education as recognition of outstanding skills in writing, photography, and graphic design in publications and web sites. Public awareness A new head of public awareness (PA) arrived in May. New initiatives focused on donor organizations, the international media, and other key audiences, including IRRIs local community. In August, the PA Unit launched The IRRI Hour, a three-times-a-week chat and music program on the FM radio station on the campus of the University of the Philippines Los Baos. Many IRRI scientists have been local radio guests. PA arranged 25 radio or television interviews in which IRRI scientists appeared before international audiences. About 60 news releases and more than 100 photographic releases were issued to the world news media during the year, covering research developments, new appointments, important visitors, and major conferences. Four issues of the Hotline newsletter were printed and distributed. PA helped organize 19 special events, including NGO, Host Country, and Ambassador days. PA also arranged the first corporate sponsorship of an IRRI event during which the airline Lufthansa sponsored a week-long IRRI-Germany event in November. The PA unit will continue seeking corporate sponsorship to carry word of IRRI achievements to the international public.
125
IRRI visitors, exhibitions, and conferences 1999 was an active year for international exhibitions. IRRI staff members staged eight international displays, and one in the Philippines, including first-ever appearances at the World Bank annual conference in Washington and the American Society of Agronomy meeting, as well as the week-long IRRI-Germany event. A fire in Chandler Hall in September extensively damaged the popular Riceworld Museum and Learning Center, depriving IRRI of its principal showpiece for visitors. Renovations began immediately to the fire-, smoke-, and water-damaged museum and a new, updated Riceworld was ready for IRRIs 40th anniversary celebration in April 2000. Four months without Riceworld and the Chandler Hall auditorium reduced the number of visitors in 1999. Despite the disruption, more than 38,250 people visited IRRI, including a royal head of state, 8 state ministers, 39 ambassadors and members of the diplomatic corps, and 11 representatives of various international and donor organizations. IRRI sponsored or cosponsored and hosted 24 regional and international conferences, workshops, symposia, reviews, and meetings, attended by more than 1,150 participants from 48 countries (Table 1). Library and documentation service During 1999, 7,445 references were added to the rice bibliography at the IRRI Library, bringing the total to nearly 180,000. One hundred four dissertations, mainly from China and the Philippines, were acquired. The librarys computer equipment was upgraded to scan and transmit documents via email. An audiovisual learning center was opened at the library for use by the IRRI staff. Plans were announced for expansion of the librarys rice database and its online public access catalog, and for placing 50 year of citations from its Rice Literature Update on CD-ROM.
Human capital development of NARS rice professionals

R. Raab and staff IRRIs training efforts anticipate, and respond to, the needs of rice research and development programs. It simultaneously addresses the human resource development needs of NARS and the objectives of IRRI in fulfilling its global mandate. Between 1961 and 1999, IRRIs training program provided 12,439 opportunities for degree and postdegree scholarships, and headquarters and incountry group training. That training benefited NARS rice scientists and professionals from 98 countries. Degree and postdegree training The objective of degree and postdegree training is to increase the number of highly trained professionals in NARS institutions and develop leaders in rice science. In 1999, a total of 130 rice scientists from 19 countries participated in the degree and postdegree training program: 74 PhD scholars, 23 MS scholars, 22 on-the-job (postdegree) fellows, and 11 interns (Table 2). Fifty-five scholars and fellows completed their program during the year: 22 PhD scholars, 7 MS scholars, 21 on-the-job fellows, and 5 interns (Table 3).
LANGUAGE PROFICIENCY DEVELOPMENT FOR SCHOLARS AND TRAINEES
Scholars and trainees require a good level of proficiency in the English language for success in learning. During 1999, we offered four language proficiency development courses: Basic English, Selfediting, Conversational English, and English for Agriculture. Sixty-three individuals from 13 countries, including scholars, trainees, IRRI staff members, and some spouses, participated in the courses. The English as a Second Language (ESL) lab provided a parallel service for those who had individual programmed lessons. Six graduating scholars had assistance in the editing of their thesis manuscripts for grammar and style.
126
Table 1. International and regional conferences, workshops, symposia, and meetings hosted or cosponsored by IRRI, 1999. Participants (no.) 10 Countries (no.) 1
Date 1122 Jan
Title Workshop on Data Analysis of the Participatory Plant Breeding Project SysNet Workshop on Methodologies for Land Use Planning for Haryana State, India International Crop Information System (ICIS) Development Workshop SysNet Workshop on Analysis of Land Use Scenarios for Kedah-Perlis Region, Malaysia Mid-Term Meetings of the Agricultural Technology and Utilization and Transfer (ATUT) SysNet Workshop on Methodologies for Land Use Planning for Can Tho Province, Vietnam Workshop on Carbon and Nitrogen Dynamics in Flooded Soils The Red River Basin Ecoregional Project(RRBEP) Regional Working Group Meeting and Technical meetings Rice Orientation Workshop for News Media Pest Impact Assessment Research: Workshop on Data Management and Analysis; Discussion on Red Stripe Disease, and visit to IRRC site
Venue IRRI
716 Mar
India
31
812 Mar
IRRI
43
13
712 Apr
Malaysia
51
1516 Apr
IRRI
29
1619 Apr
Vietnam
45
1922 Apr
IRRI
25
2126 Apr
Vietnam
28
2124 Apr 312 May
IRRI India
24 25
9 6
712 May
Workshop on Analysis of Land Use Scenarios for Ilocos Norte and Stakeholders Meeting Bioinformatics Workshop Workshop on Geo-Information Techniques for Understanding and Analyzing Rainfed Rice Environments in Eastern India Third Working Group Meeting and Review of the Project on Assessing Opportunities for Nitrogen Fixation in Rice Second CREMNET-India Workshop Cum Group Meeting on Innovative Nitrogen and Other Crop Management Technologies for Intensive Rice Systems of South Asia Workshop on Genebanks and Comparative Genetics
Philippines
57
1214 May 2426 May
IRRI India
14 50
5 2
912 Aug
IRRI
49
2327 Aug
India
52
2427 Aug
Netherlands
52
21
610 Sep
Workshop on Decision Processes for Determining Diagnostic and Predictive Criteria for Soil Nutrient Management PhilRice-UPLB-IRRI Tripartite Meeting Human Nutrition Workshop
Philippines
54
20
2829 Sep 57 Oct
IRRI IRRI
75 93
1 20
127
Table 1 continued. Participants (no.) 75 Countries (no.) 8
Date 1113 Oct
Title SysNet 99: Systems Research for Optimizing Future Land Use Upland Rice Research Consortium Steering and Technical Committee Meeting Planning Workshop on Ecoregional Approaches to Natural Resource Management in the Korat Basin, Northeast Thailand Workshop on Seed Health for Disease Management Council for Partnership on Rice Research
Venue IRRI
1114 Oct
India
50
11
2629 Oct
Thailand
50
1719 Nov 2728 Nov
Thailand Thailand IRRI Indonesia
41 19 25 70
10 11 6 10
29 Nov.3 Dec Redesigning Rice Photosynthesis Workshop 59 Dec International Workshop on Understanding and Characterizing Rainfed Environments 10th Rainfed Lowland Rice Research Consortium Steering Committee Meeting
10 Dec
Indonesia
15
Table 2. Participants in degree and postdegree training at IRRI, 1999.
PhD scholars Alam, Muhammed Murshedul Alam, Murshed Cui, Kehui Fu, Binying Jianli, Wu Li, Luping Liu, Bin Lu, Wenjing Xu, Jianlong Yuequi, He Zhong, Daibin Zhong, Xuhua Ziqi, Wang Bagali, Prashanth G. Baisakh, Niranjan Deka, Nivedita Dey, Moul Kaur, Jatinder Lourdusamy, Gabriel Stephen Mathan, Natarajan Mitra, Sudip Narayanan, Narayanan N. Nath, Palash Deb Rangan, Latha Sreevidya, V.S. Abdullah, Buang Muhsin, Muhammad Esfahany, Masoud Fallah, Allahyar Hosseini Salekdeh, Seyed G. Moradi, Foad Moumeni, Ali Kobayashi, Sohei Naoyoshi, Kawano Okada, Kanako
Bangladesh Bangladesh China China China China China China China China China China China India India India India India India India India India India India India Indonesia Indonesia Iran Iran Iran Iran Iran Japan Japan Japan
Samejima, Hiroaki Andrianilana, Fidelis J. Ramanantsoanirina, Alain Mari Razafinjara, Aime Lala Htet, Kyu Htut, U Tin Thet, Khin Maung Winn, Tun Bhandari, Hum Nath Ojha, Gana Pati Regmi, Anant Prasad Trolove, Stephen Neil Arif, Muhammad Faiz, Faiz Ahmad Hussain, Fayyaz Ijaz, Muhammad Briones, Lucia M. Cabuslay, Gloria delos Reyes, Jeannelyn B. Enriquez, Emilie C. Lumbo, Susanita G. Rubia, Leila G. Mnzava, Moses N.W. Lersupavithnapa, Boontium Linwattana, Grisana Nieuwenhuis, Pailin Graw, Stephen M. Cuong, Ngo Luc Huu Ho, Nguyen Le, Cam Loan Le, Thi Phuong Nghia, La Tuan Ngo, Ngoc Hung Nguyen, My Hoa Nguyen, Van Hong Thanh, Duong Ngoc
Japan Madagascar Madagascar Madagascar Myanmar Myanmar Myanmar Myanmar Nepal Nepal Nepal New Zealand Pakistan Pakistan Pakistan Pakistan Philippines Philippines Philippines Philippines Philippines Philippines Tanzania Thailand Thailand Thailand United States Vietnam Vietnam Vietnam Vietnam Vietnam Vietnam Vietnam Vietnam Vietnam
128
Table 2 continued. Thi Ut, Tran+A26 Tran, Chi Thien Truong, Van Tuyen Vietnam Vietnam Vietnam Denmark Ethiopia India India Indonesia Indonesia Indonesia Korea Lao-PDR Madagascar Nepal Nepal Nepal Netherlands Philippines Philippines Philippines Philippines Philippines Vietnam Vietnam Vietnam Vietnam Bangladesh Bangladesh Cambodia Mot, Sana Pao, Sinath Nath, Palash Deb Sikdar, Samir Ranjan Prasetiyono, Joko Adachi, Shimpei Ki-Do, Park Boualaphan, Chanthakhone Khangsuthor, Khamdok S. Shrestha, Sundar M. Pedrera, Maria Jehan G. Parakam, Jagkrit Sengpaseuth, Rasabandith Urairong, Hathairat Cuong, Nguyen Minh Luo, Wen Yong Luu, Van Quynh Phan, Van Tuan Tran, Duc Thach Cambodia Cambodia India India Indonesia Japan Korea Lao-PDR Lao-PDR Nepal Philippines Thailand Thailand Thailand Vietnam Vietnam Vietnam Vietnam Vietnam
MS scholars Jensen, Morten Shoatatek, Negussie Singh, Bhupinder Pal Varghese, Pulickal A. Sattarai, Majid Suganda, Husein Widowati, Ladiyani Retno Chung, Dae-Hwan Inthavong, Soulaphone Rakotomalala, R. Mbolarinosy Bhattarai, Kiran Joshi, Madan Raj Upadhyay, Bhawana De Vries, Sander C. Alcantara, Jovencio Avendano, Bita S. Madamba, Reina Suzette B. Pantua, Sheila Marie E. Ulat, Victor Jun Ngo, Dang Phong Nguyen, Thi Phong Lan Thach, Thi Ngoc Anh Thao, Hoang Thi Bich Postdegree on-the-job trainees Hossain, Shahadat Rahman, Shamsur Bounphanousay, Chay
Interns Manio, Denise Chung, Carrie Lee Golinowsky, Shawn Paul Gray-Donald, James Scholosser, David Schaffer, Sebastian Widharto De Ponti, Tomek Oort, Pepjin Ponsioen, Thomas Christian Riksen, Barbara
Australia Canada Canada Canada Canada Germany Indonesia Netherlands Netherlands
Development and implementation of shortterm group courses Specialized short-term group training courses assist NARS professionals in increasing their competencies in rice science and in disseminating appropriate technology to end-users in NARS. Ten group training courses were conducted in 1999, with participation by 95 scientists from 18 countries (Table 4). Two of the courses were offered for the first time in 1999: q Modern Rice Farming for Asia q Transgenic Rice Production and Deployment with Special Reference to Sheath Blight and Stem Borer Resistance Development of new training methodology We continued to improve our distance education capability. A 3-d meeting on the use of information communication technology (ICT) in support of the training program was organized. Thirteen distance
education and ICT experts from the US, UK, Canada, Japan, India, Thailand, the Philippines, and organizations such as the World Bank, Commonwealth of Learning, and the International Development Research Centre (IDRC), attended and proposed strategies to upgrade the IRRI ICT capability. Development of the IDRC-funded project Application of Distance Learning Technologies to Human Capital Development continued. Simon Fraser University finalized the design for the online course on Experimental Design and Data Analysis and is developing its components. IRRI and the Directorate for Rice Research, Hyderabad, India, identified two priority online courses for their scientists: Digital Literacy and Presentation Skills for Scientists. The online course on digital literacy was developed. The course on presentation skills was designed and content is being developed. We used net-based video conferencing for remote delivery of topics to participants in a Rice Production Research course in Thailand. It was also
129
Table 3. Number of scholars and postdegree trainees who completed training at IRRI in 1999. Type Ia Type IIa Type IIIa Country PhD MS PhD MS ND interns Total
Africa Madagascar Subtotal Asia Australia Bangladesh Cambodia China India Indonesia Iran Japan Korea Lao-PDR Myanmar Nepal Pakistan Philippines Thailand Vietnam Subtotal Europe Germany Netherlands New Zealand Subtotal North America Canada Subtotal Total
0 0
0 0
2 2
0 0
0 0
2 2
0 0 0 2 4 0 1 1 0 0 0 1 0 2 0 3 14
0 0 0 0 0 0 0 0 1 0 0 0 0 3 0 0 4
0 1 0 0 0 0 0 0 0 0 1 0 1 0 1 1 5
0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2 3
1 2 3 0 1 1 0 1 1 2 0 1 0 1 3 5 22
1 3 3 2 5 1 1 2 2 3 1 2 1 6 4 11 48
0 0 1 1
0 0 0 0
0 0 0 0
0 0 0 0
1 1 0 23
1 1 1
0 0 15
0 0 4
0 0 7
0 0 3
2 2 26
2 2 55
a Type I = MS and PhD scholars, thesis research only at IRRI. Type II = MS and PhD scholars, coursework and thesis at IRRI. Type III = on-the-job nondegree trainees and interns.
used for an interactive lecture by a resource person at Nanyang University, Singapore, to participants in a bioinformatics course at IRRI. Training materials development Training materials development supports group course implementation, both at IRRI and in-country. Twelve sets of training materials were produced in 1999: q five course and reference manuals for three new headquarters courses: Application of Molecular Tools to Study Rice Viruses, Transgenic Rice Production and Deployment, and Soil and Water Biochemistry and Ecotoxicology;
two instructional video materials on the leaf color chart (LCC) and the portable chlorophyll meter; three Web publications based on print or sound-slide presentation formats of the training module on Growth Stages of the Rice Plant, the course manual for the Trainers Training Course, and the skills development booklet How to Take Notes; and two manuals on tungro disease identification and management to support a two-level course for field technicians and for farmers.
130
Table 4. Participants in short-term group training at IRRI, 1999.
Hybrid Rice Breeding 1 Mar23 Apr Haque, Md. Azizul El-Mowafi, Hamdi Fetouh Kanwar, Omvir S. Mutchukota, Subba Rao Maibam, Rohinikumar S. Mukherjee, Sumita Lekkala, Madan Mohan Reddy Kitani, Shigekazu Kaskon, Mohd. Asrore Bin Aye, Mar Mar Cheema, Md. Bashir dela Rosa, Analen M. Gaspar, Manuel G. Silva, Jayasena T. Dao, Thi Phuong
Bangladesh Egypt India India India India India Japan Malaysia Myanmar Pakistan Philippines Philippines Sri Lanka Vietnam
Introduction to New Developments in G E Analysis and Interpretation of Results 711 Jun Sarom, Men Cambodia Singh, Ashok Kumar India Roy, Ashutosh India Sarwoto Indonesia Sengpaseuth, Rasabandith Lao-PDR Bhandari, Hem Singh Nepal Bordeesorn, Juntra Thailand Suriya-Arunroj, Duangjai Thailand Chatwachirawong, Prasert Thailand Youngsuk, Piboonwat Thailand Quynh, Luu Van Vietnam Tran, Duc Thach Vietnam Phan, Van Tuan Vietnam Soil and Water Biochemistry and Ecotoxicology 23 Aug10 Sep Min, Liao China Kalaichelvan, Gurumurthy India Collado, Wilfredo B. Philippines Malabayabas, Myrna D. Philippines Ruensuk, Nittaya Thailand Kunnoot, Laddawan Thailand Trang, Le Thi Dieu Vietnam Rice Seed Health for Crop Management 30 Aug8 Oct Harun, M. Eusuf Bangladesh Jahan, Quazi Shireen Akhter Bangladesh Islam, Md. Shahidul Bangladesh Hossain, Mohammod Bangladesh Hossain, Md. Mosaddeque Bangladesh Khiev, Bunnarith Cambodia Wei, Zhang China Suprihanto Indonesia Sengsoulivong, Viengkham Lao-PDR Thavong, Porntip Thailand Thuan, Tran Thi Vietnam Transgenic Rice: Production and Deployment with Special Reference to Sheath Blight and Rice Stem Borer Resistance 1129 Oct Alam, Mohammad Firoz Bangladesh Islam, Rafiqul Bangladesh Gongyin, Ye China Sobhakumari, Valiya P. India Tummala, Rama Kumar India Ayyagari, Phani Padma K. India Sellapan, Krishnan India Daradjat, Aan Andang Indonesia Machmud, Muhammad Indonesia Kim, Hyun-Soon Korea Abad, Joseph D. Philippines Angeles, Amelita T. Philippines Soontrajarn, Kasem Thailand Applications of Molecular Tools to Study Rice Viruses 2 Nov3 Dec Rai, Ramesh C. India Nath, Palash Deb India Shrestha, Chandra Laxmi Nepal Quan, Hoang Anh Vietnam
Introduction to IRRISTAT Statistical Software 1 Mar23 Apr Roongrawee, Puttana Thailand Tacho, Chonticha Thailand Modern Rice Farming for Asia 328 May Rahman, Habibur Uddin, Md. Mahatab Thol, Nou Sarun, Sam Lang, Mondul Sarith, Hin Poch, Kep Ravi, Venkatachalam Vijayalakshmi, B. Ibrahim, M. Nur Mindarti, Susi Corpuz, Artemio A. Chaikot, Kaensri Lacornket, Chainarong Satienras, Worwit Huan, Tran Thi Ngoc Sau, Tran Van Advanced Experimental Design 31 May4 Jun Mollah, Md. Islam Uddin Hun, Yadana Jatmiko, Sigit Yuli Tirtoutomo, Susanto Phan, Van Tuan Quynh, Luu Van Tran, Duc Thach Instructional Video Production 7 Jun2 Jul Quraishi, Nazrul Norbu, Tandin Qin, Zelin Prayogo, Kuscahyo Budi Varquez, Frenciso L. Le, Don Dai
Bangladesh Bangladesh Cambodia Cambodia Cambodia Cambodia Cambodia India India Indonesia Indonesia Philippines Thailand Thailand Thailand Vietnam Vietnam
Bangladesh Cambodia Indonesia Indonesia Vietnam Vietnam Vietnam
Bangladesh Bhutan China Indonesia Philippines Vietnam
131
Collaborative in-country training Collaborative in-country training is a 10-y-old initiative to help develop NARS institutions training capability and to complement IRRI training activities at headquarters through partnerships for group courses away from IRRI. Fourteen courses were conducted collaboratively with NARS partner institutions. Those had 361 participants211 rice scientists and 150 farmersin 12 countries.
REGIONAL COURSES
The IM program will increase integration of its activities with research. The number of projects will be reduced to bring similar activities closer together. It is anticipated that many of the Training Center, CREMNET, and the International Programs Management Office activities will merge. Similarly, CPS, Library, and Computer Services activities expect greater integration. The aim will then be to imbed IM activities in the research activities. The Irrigated Rice Research Consortium will be an initial focal point for the merging. International Programs Management Office (IPMO) The year 2000 will be a period of review and consolidation of IPMOs existing mandate and responsibilities. The goal is to have better integration of country activities with other research and training activities and to better tap the country representatives as sources of research needs, priorities, and opportunities. A review of existing protocols will continue the standardization of operating procedures. The review will include assessment of our presence in different countries and the roles needed with collaborating NARS. We will develop investment criteria and databases that refine IRRI strategies for involvement in the different countries. A multicountry coverage for smaller countries in implementing collaboration will be considered. Training Center The activities of the Training Center will be highlighted by a pro-active posture in the design and delivery of training events and materials. We are in the process of seeking opportunities to assist in the development and delivery of field training based on the research product and in association with the IRRI research program. Examples of this pro-active approach are: q Indonesian AARD/World Bank Trainee Program. Ten trainees from Indonesia who have been through the Rice Production and Bio-Statistics courses are currently enrolled in the Farming Systems course, and will complete their 3+-mo on-the-job training with an IPM course. Under past procedure, they would develop a survey instrument and implement the practice study at IRRI. For 2000, they will return to Indonesia to
Regional in-country courses are conducted collaboratively with a NARS institution for an international group of trainees. Two courses were conducted in 1999. One was the eighth offering of the course on Rice Production Research in collaboration with the Pathumthani Rice Research Center of Thailand. It attracted 18 rice professionals from six countries. The other course was the second offering of the course on Adaptive Research in Farming Systems Development done in collaboration with the Farming Systems Research Institute of the University of the Philippines Los Banos (UPLB) for eight participants from four countries.
NATIONAL COURSES
National in-country courses are IRRI courses revised to address the specific needs of requesting NARS. Twelve in-country courses were conducted by IRRI for 185 rice professionals in nine countries. Among the courses was the sixth offering of the course on Problem-based Technology Generation for Rice Environments conducted in Cambodia. Twenty rice scientists and agriculture educators participated in the course.
Program outlook
During 2000, IM will develop its role as a delivery facilitation program and its role in linking research and development. Although IRRI has a long history of linkage with development organizations, that linkage is not widely recognized. Various groups within IM will review their roles, functions, products, clients, and targets to improve the visibility of their activities.
132
implement the farming systems study, including IPM. This will provide the field test of the document and the method and, at the same time, serve as a baseline study for the Indonesian scientists. An example of our proactive approach with our research program is with the Hybrid Rice Working Group. We are in the process of designing a model workshop with the Philippine Rice Research Institute that will use IRRI research output in seed breeding and seed production. A multiorganizational team will create the materials and workshop design. The workshop will then be replicated in the participating countries, with the materials adapted to local conditions, facilitated by NARS partners, and backstopped by IRRI specialists.
velop and evaluate TRC, a location-specific package of improved technologies for sustaining high rice productivity in the irrigated ecosystem, in one country. Access to IRRIs media assets We will follow up the 1999 creation of a searchable database for video by adding IRRIs still photography collection (an estimated 40,000 images as print and transparency). Key rice-related images will be selected, indexed, and digitized. The video and still image databases will then be merged to allow IRRI staff members and clients to access them for their communication needs in rice research and training. Computer Services Computer Services will move across campus to the CPS building for better integration of activities and increased efficiency of operation. Other improvements in efficiency include the planned implementation of several changes to the IRRI network. Staff housing will be brought onto the network, beginning with the guesthouse and then with the homes of scientists. IRRIs IVDN connection will also be upgraded to provide a redundant connection to Manila. Operating system software will be upgraded to Windows 2000 on servers and on selected PCs. Computer Services will collaborate actively with CPS in helping the organization shift away from paper-based information flows. The improvement of IRRIs Intranet and a move toward electronic rather than printed forms will be key steps in this. The benefits of increasing digitalization and improved communications will flow also to IRRIs country activities.
CREMNET Plans for 2000 include the tighter integration of CREMNETs applied research activities with other IRRI and NARS research partners. Activities will be defined in light of their fit with other nutrient management research activities. For example, CREMNET will continue to refine the chlorophyll meter technique with variable rate N application based on a range of SPAD threshold values observed at critical crop growth stages. Village-level evaluation and promotion of the LCC, impact evaluation of the LCC method in Vietnam, and the evaluation of controlled-release fertilizers for hybrid rice, rice-rice and rice-wheat systems will continue in three countries. Development and adaptation of technologies for integrated crop management for direct-seeded rice and the refinement and promotion of drum seeders will continue in four countries. CREMNET will de-
133
Affiliations of collaborating researchers
1Philippine 2Hokkaido
Rice Research Institute, Philippines. University, Japan. 3Assam Agricultural University, India. 4Research Institute for Rice, Indonesia. 5Central Research Institute for Food Crops, Indonesia. 6Indian Agricultural Research Institute, India. 7G.B. Pant University of Agriculture and Technology, India. 8Natural Resources Institute, UK. 9University of Aarhus, Denmark. 10Institute of Agricultural Sciences, Vietnam. 11Narendra Deva University of Agriculture and Technology, India. 12China National Rice Research Institute, China. 13Central Rice Research Institute, India. 14Punjab Agricultural University, India. 15Research Institute for Food Crop Biotechnology, Indonesia. 16Yunnan Agricultural University, China. 17Zhejiang Academy of Agricultural Sciences, China. 18Indian Institute of Soil Science, India. 19Chinese Academy of Agricultural Sciences, China. 20Bangladesh Rice Research Institute, Bangladesh. 21Malaysian Agricultural Research and Development Institute, Malaysia. 22Cuu Long Delta Rice Research Institute, Vietnam. 23Sichuan Academy of Agricultural Sciences, China. 24Indira Gandhi Agricultural University, India. 25Mariano Marcos State University, Philippines. 26Orissa University of Agricultural Technology, India. 27Rajendra Agricultural University, India. 28University of Adelaide, Australia. 29Zhejiang Agricultural University, China. 30Huazhong Agricultural University, China. 31Williams College, USA. 32Indian Institute of Science, Center for Genetic Engineering, India. 33Murugappa Chettiar Research Center, India. 34Universitat Bremen, Germany. 35University of the Philippines Diliman, Philippines. 36Banaras Hindu University, India. 37Michigan State University, USA. 38Australian National University, Australia. 39Sakha Agricultural Research Station, Egypt. 40Myanmar Agricultural Service, Myanmar. 41M.S. Swaminathan Research Foundation, India.
135
Research support services
ANALYTICAL SERVICE LABORATORIES
The Analytical Service Laboratories (ASL) provides analytical and analysis-related services to IRRI research projects. Guidelines for a User Laboratory were developed and distributed during 1999. Aside from providing facilities for projects involving the use of radionuclides, the User Laboratory offers use of equipment and facilities for organic extraction and chromatography. Analytical services ASL completed 61,191 routine analyses for soil, plant, fertilizer, and water samples; elemental analysis for total C/N, stable isotope ratio analysis for 13C and 15N, and N and carbohydrates by nearinfrared reflectance (NIR) (Table 1). Improvement of protocols A library of NIR spectral information for rice plant samples was developed from an assortment of samples (different plant parts collected at various growth stages) sent for different analyses for the
past 3 years. This necessitated the development of specific calibrations for the diverse types of samples and resulted in a large collection of NIR spectra of samples with reference chemical analyses. The spectral library was organized by analyses, plant part, sample presentation (cup insert used), and by growth stage. Spectra of new samples can now be readily matched to the samples in the library and appropriate samples selected for developing the calibration. NIR analysis was initially routine after ASL developed global calibrations applicable to plant samples commonly submitted for N analysis. However, certain analyses required specific calibrations, e.g., analysis of sugar and starch in rice seedlings. Also, the small amounts (<100 mg) of samples available for NIR scanning demanded careful packing and cleaning of the sample cups. A protocol was developed to allow researchers to perform their own NIR analysis. With ASL training and supervision, a scientist researcher can do NIR scanning, select suitable samples from the library, and perform the calibration and prediction.
Table 1. Analyses completed by ASL for IRRI research programs, 1999. Research programsa served (no. of analyses) Analysis I Plant analysis Soil analysis Water analysis Fertilizer Mass spectrometry Radioisotope counting Total
a
R 10,733 2,185 619 12 30 13,579
U 2,084 651 676 4,741 1,057 9,209
F 1,270 68 132 1,470
C 286 278 3,154 3,718
IM 3,592 91 350 4,033
Others 2,895 2,236 2,301 55 747 55 8,289
12,427 4,081 377 2,952 1,056 20,893
I = irrigated, R = rainfed, U = upland, F = flood-prone, C = cross-ecosystems, IM = accelerating impact.
138
WINISI v1.02a software was acquired to upgrade NIR data management software from the DOS to Windows version. A new and useful feature is the local calibration, which automatically matches each individual sample to the library, selects similar samples, and performs calibration for each sample. This was found to be as much as 25% more accurate than global calibration wherein equations are developed from all samples in the product library. Data management ASL was successful in implementing electronic transactions (submission of sample identification forms, receipt notification, analysis query or request) over the IRRI Intranet. ASL clients have access to the ASL database anytime, anywhere in the world. Moreover, requests for various services can be submitted anytime. The only form without an online equivalent is the Charge Slip Form. A software program was developed to implement a secure, reliable, and fast system to validate budget code and user ID number in electronic forms. The program still needs permission to access human resources development and accounting central databases and has to go through beta testing before it can be implemented. User Laboratory Guidelines for the use of special analytical instruments and facilities by non-ASL staff were developed. The User Laboratory includes facilities for chromatography (high-performance liquid chromatography, gas chromatography (GC), GC-mass spectrometry), radioisotope work (LSC, radioisotope management), and organic extraction. These facilities serve a multidisciplinary demand and require particular expertise for operation and maintenance. ASL staff members are available to help develop and optimize instrument parameters for specific applications in collaborative arrangements. Soil organic matter characterization by thermochemolysis with tetramethylammonium hydroxide (TMAH) ASL continued to provide run-ready facilities for soil organic matter characterization by thermochemolysis as part of the User Laboratory.
Thermochemolysis involves the breakdown of organic macromolecules such as humic acids into smaller fragments by reaction with 25% TMAH in methanol at elevated temperature (250 C). In a trial run, four humic acid samples from a rice-upland crop rotation experiment were analyzed to identify chemical changes associated with timing of crop residue incorporation or crop rotation. Preliminary results suggest that aerobic decomposition of crop residues might provide small amounts of several phenolic compounds, implying a more decomposed and possibly less cross-linked nature. This finding agreed with independent conclusions drawn from crop nutrient uptake and soil N mineralization data of the same experiment, indicating that early residue incorporation leads to formation of more easily mineralizable organic matter. The types of phenolic compounds detected did not vary by treatment. More reliable conclusions will be possible in 2000 after a complete set of samples from this experiment has been analyzed. Radioisotope Laboratory The ASL provided liaison with the Philippine Nuclear Research Institute on matters related to the use of radioactive materials. The following are new projects that benefited from those services: q Use of amplified fragment length polymorphism in the study of variability in rice root-knot nematode populations q Perenniality and nematode resistance, genomic mapping, and quantitative trait locus (QTL) analysis q Drought tolerance, root morphology, and allelopathy genetic mapping and QTL analysis q Transferring osmotic adjustment to japonica upland rice genotypes through marker-aided selection q Allelopathy in upland rice breeding programs q Fine-mapping the major gene underlying Pdeficiency tolerance in rice
BIOMETRICS
The Biometrics Units functions involve research and service activities. Service activities include providing a consultation facility, quality assurance advice, biometrics training, statistical software, and computer services. Research activities include in-
139
volvement of staff members in the statistical and data management aspects of research projects and in the development of statistical techniques and software required by IRRI research programs. Service functions Biometrics provided consultation services to researchers, scholars, and fellows ranging from planning and layout of experiments, data collection and analysis, to interpretation and presentation of results. Ad hoc consultations involved problems in SAS, IRRISTAT, and other software, database design, and questions on statistical concepts. Some consultations during 1999 were for q dynamics of rice tungro bacilliform virus (RTBV) accumulation on a susceptible and tolerant host infected with RTBV alone or coinfected with rice tungro spherical virus (RTSV), q dispersal of yellow stem borers in the field, q estimation of heritability of root-pulling resistance in rainfed lowland rice, q assessment of milled rice quality in the Philippines, q survey of RTBV/RTSV occurrences in different sites in the Philippines, and q assessment of resistance to brown planthoppers and whitebacked planthoppers of several A, B, and R lines. Training Biometrics participates in Training Center courses that contain a statistical component and provides training workshops and courses for IRRI researchTable 2. Training by the Biometrics Unit at IRRI, 1999. Training Duration
ers, scholars, and trainees on special topics (Table 2). Software development International Crop Information System (ICIS) development continued with refinement of a breeders interface and development of the Data Management System (DMS). An Excel program called Workbook was developed to access the DMS. It was used to load INGER data and breeders data to a local DMS. The local DMS for PBGB data for the irrigated rice ecosystem now contains the evaluation data from 1997 DS to 1999 WS for the replicated yield trial, observational yield trial, and hybridization block trial. IRIS is used for all data management. The DMS structure is being revised to accommodate multiple labeling of factors. Simultaneous with the Workbook revision is the modification of the central DMS, the local DMS for the irrigated ecosystem, the DMS for the Hybrid Rice project, and the DMS for INGER.
COMMUNICATION SUPPORT
The Communication and Publications Services unit of the Information Center provides communication support for the entire institute. The services include editing, graphic design, art and illustration, audiovisual, photography, video, and printing. The unit printed 1,935,244 pages of text, not including IRRI books, which were contracted out. About 19,965 original slides were produced; 3,728 slides duplicated; and 3,469 black-and-white photo-
Students (no.) 15 14 9 9 9
Comments
Experimental design and data analysis (EDDA1) Introduction to IRRISTAT for Windows Introduction to SAS for Windows Introduction to SAS for Windows Advanced experimental design Hybrid rice breeding training course Introduction to new developments in GE analysis and interpretation Rice seed health
112 Feb 15 Mar 812 Mar 2630 Apr 31 May4 Jun 1 Mar23 Apr 711 Jun Aug 30 8 Oct
Includes demonstration of distance education
21
Participants from different IRRI outreach offices. Resource persons for experimental design and GE Participants from different RLRRC countries plus Cambodia, Laos, and Vietnam Resource person for experimental design, in-class sampling experiment, and IRRISTAT
140
graphs printed. IRRI graphic artists produced 204 illustrations, laid out 1,335 pages for publication, and prepared 116 posters. Other activities of the Information Center (public awareness, scientific publication, library and documentation service, and conference and workshops) are reported in the Accelerating the Impact of Rice Research Program. In addition to the work reported there, IRRI editors worked on 129 journal articles and miscellaneous papers (conference papers, proposals, and others) totaling about 2,745 pages of text, tables, and figures.
COMPUTER SERVICES
IRRI electronic mail A change from an old VAX-based email system to Microsoft Exchange and an upgrade to a server capable of supporting more than 1,000 mailboxes were completed. One mail server is reserved for use as a standby system in future. Dial-in facilities for the campus network were improved with the installation of additional phone lines and the replacement of old modems. Integrated voice and data network (IVDN) The IVDN was upgraded from a 64-k leased line to a 128/256-k frame relay circuit with TMI via Hong Kong. The old circuit was grossly overloaded with email queues routinely backing up half a day and longer. A contractual standoff with Globe Telecommunications continued unresolved. A proposal to replace Globe is expected in the first quarter of 2000. IRRI and the University of the Philippines Los Banos (UPLB) agreed to interconnect their campus networks. Shared aims were to improve communications between them (currently routed via the United States), improve joint access to communication facilities off campus, and interconnect satellite locationsIRRI staff housing and the Open University. Connection is expected in 2000 after UPLB upgrades its campus network. IRRI staff members have used historically the IVDN as a convenience. In 1999, staff members were given detailed information on the costs of using the IVDN vs international direct dialing and many began to use the cheaper alternative. At 1999 prices, IRRI can save a further $25,000$30,000 per annum on telephone calls. Monitoring of the use of the IRRI Internet link was introduced and tested and found useful in discouraging nonbusiness use. Asset management The Y2K inventory and assessment exercise proved to be a wake-up call for asset management. Problems were identified in existing records and management. Computer Services worked on possible solutions and initiated discussions on terms of reference with potential outside providers.
Computer Services began work to ensure year 2000 (Y2K) compliance in 1998. Computer hardware, software, and other equipment were examined for compliance. IRRI intranet facilities were used to disseminate information and software. Remedial actions during 1999 included q shutting down old information systems and moving applications and data to new systems and media; q amending and recompiling IRRI software applications; q replacing or updating commercial software applications (IRRI accounting system); q patching operating systems software (Windows 95, NT); q adding software corrections for noncompliant hardware or disposal or replacement of noncompliant hardware; and q updating affected noncomputer systems (telephone). Communications Expenditures in all areas of communications came under scrutiny in 1999 and a number of areas were identified as offering significant potential for savings, most notably in fax costs. Close attention to billing and cost control detected fraudulent use of a SITA line by email. The password required for access was changed, and negotiations for a rebate of the charges are ongoing. New email and communications software was dispatched to outposted staff members, along with up-to-date antivirus software.
141
EXPERIMENT STATION
The Experiment Station (ES) served 222 requests for land and facilities. Land use was: Division Dry season (ha) 42.88 14.41 8.69 4.54 4.25 3.91 1.99 1.36 92.03 Wet season (ha) 55.93 4.29 9.43 5.92 4.25 5.11 1.99 1.36 88.28
PBGB GRC APPA ES1 CIAT EPPD SWS AED Total
A total of 7.3 ha (4 ha dry bed and 3.3 ha wet bed) of nurseries were provided and maintained by ES. The dry-bed nurseries were constructed with a modified tractor-powered rototiller with a fluted roller used to mark rows as furrows. The wet-bed nurseries were used mainly by PBGB for rice hybrid trials. Sixty-eight tons of different kinds of fertilizers were used in 1999. ES planted and harvested breeder seed increase and seed production materials on 20.46 ha of lowland. Kabesilya transplanters were used for seed increase plots and a mechanical transplanter, seed blower-spreader, and direct seeding were used for seed production plots. A Thai combine harvester was used for all seed harvesting. Integrated pest management (IPM) eliminated the need for insecticide applications in all seed increase and production plots. A total of 143 t of different rice materials were processed from harvests of seed increase, seed production, and border rows. Underground irrigation systems using polyvinyl chloride (PVC) pipes were installed in Blocks UE14, UU1, and UU4. Five hundred irrigation and drainage manhole covers were fabricated and installed. Drainage outlets were developed and constructed in plots C27-37, D10-D14, D15-D23, E6E27, F1-F10, G2-G21, J2-J13, K2-K10, L2-L8, Lower MN1-7, and Upper MN1-8. Drainage rehabilitation continued at Blocks 518-531, 602-617, UE, UK, UU, and UV.
Land development work consisted of reorientation of Block UE (4 ha) and conversion of upland to lowland use in Blocks UU1 and UU4 (2 ha) as an additional area for post-entry quarantine. The upper half of Blocks UF4-5 was developed as alternate wet-bed area. Major road rehabilitation and repair covered 4.5 km at Blocks E, F-KL, 706-733, series 900, UC-UF, UE, UJ-UK, US, UT, UY2, and the road between the upland farm and the Institute of Plant Breeding, UPLB. Pesticide use increased by 31% in 1999, mainly due to increased molluscicide use on experimental fields and herbicide use in perimeter fences, fallow areas, and levees. Application of sublethal doses of nonselective herbicides reduced the cost of mowing in maintaining fallow areas, levees, and perimeter fences. ES provided rat control services consisting of 1,866 baiting stations with 1,800 kg of rat baits for field and outreach areas. Likewise, 0.92 ha of bird net and 115 ha of rat fences with 2,259 live traps were installed. Expenditures on contract work are 21% higher in 1999 than for 1998. Birdboy hours were 10.5% higher than last year but expenditures grew by 82% mainly from an increase in pay rate. One hundred sixty-five requests for repair and maintenance of light and heavy equipment and reservoir pumps were serviced. Pumps at Blocks JK, MN, UT, and UW reservoirs were converted from turbine to submersible units. A multipurpose boom sprayer mounted on a ride-on mechanical transplanter with dual nozzles was tested for field use.
CONTROLLED GROWTH FACILITIES AND GROUNDS
Controlled Growth Facilities and Grounds (CGFG) supported 48 experiments at the phytotron, 143 experiments in greenhouses, and more than 16 experiments in the confinement level 4 (CL4) transgenic greenhouses. A total of 1,174 maintenance service requests were served. Those included provision of 728 t of ground soil to experiments in the greenhouses and the fields.
142
Controlled growth facilities Work in controlled growth facilities during 1999 covered: q New, reliable, user-friendly software and hardware were installed for the upgrade of the phytotron automatic climate control system. q Renovation of the cooling system and the canopies of the 12 outdoor growth cabinets (OGC) was completed. q Rewiring and reconditioning of the indoor growth cabinets pumping system and the sump pit tank was completed. q Twelve stainless humidifier tanks were fabricated specifically for the OGCs to replace old units, facilitate shutdown operations, and reduce OGC humidifier servicing time. q Repair and replacement of 16 solar panels was done as part of a gradual upgrade of the 64 units of solar panels of the phytotron heating system. q The quarantine screenhouse and associated work areas plus the cooling tower that services the indoor growth cabinets were refurbished. q Brackets of the condenser fan motors of the CL4 greenhouse cooling system were modified to protect the motors from the elements and simplify machine maintenance and operation procedures. The greenhouse unit maintained 44 glasshouse and screenhouse structures and provided a wide range of services to facilitate conduct of experiments. Increased demand for soil and expanded number of experiments in the greenhouses made soil grinding the main workload. Acquisition of an additional soil grinder with associated covered space, plus an appropriate soil hauling mini-dump truck, was recommended to improve productivity and efficiency of the unit. Industrial grade exhaust fans were installed in 16 glasshouses in cluster B and cluster N to improve air circulation and help reduce inside summer temperatures. The lighting and environment control systems of the cluster N darkrooms were renovated. Extended roofing and concrete flooring were in-
stalled in one processing area in cluster B. Concrete walls were installed in one cluster A screenhouse to replace rusted metal walls. Refurbishment of the structures and replacement of the screening of the screenhouse in the upland area was completed. An annual, staggered, 1-mo shutdown period was fully implemented as a standard procedure for all greenhouses, CL4, and the phytotron plantgrowing areas. This allowed general cleanup and washdown of all structures and provided for prophylactic treatment against pests, as well as preventive maintenance of cooling systems and structures. The shutdown disrupted insect and disease cycles and reduced the frequency of chemical spraying against pests by as much as 40%. Grounds The grounds unit provided landscape maintenance and development services including garbage collection for the research center and 104 residential units. Garbage collection routes and schedules were revised to improve efficiency of the service. A waste segregation program for residences and the research center resulted in high earnings from sale of materials. Sales in the 1999 last-quarter scrap auction were about US$1.2 million pesos. Two indoor gardens in the F.F. Hill Hall were landscaped to improve aesthetic appeal and reduce maintenance costs. Improvements in the safety, security, and mobility in the plant propagation work areas were done through installation of fences, plant benches, and ground surface cover, and through improvements in the blocking and arrangement of pot plants. Removal of some more hedges in applicable areas of the research center further streamlined grounds maintenance operations. Other improvements to grounds and landscaping were construction of a pylon marker at the entrance of the research center and installation of an underground sprinkler system in lawn areas fronting Chandler Hall and F.F. Hill Hall.
143
Publications and seminars
Institute publications
Books Genetic improvement of rice for waterlimited environments. 1999. 353 p. Increasing rice production in Bangladesh: challenges and strategies. 1999. 167 p. INGER-derived rice varieties directly released in various countries. 1999. 37 p. IRRI 1998-1999: Ricehunger or hope. 1999. 57 p. Program report for 1998. 1999. 187 p. Resource management in rice systems: nutrients. 1999. 355 p. Upland rice weeds of South and Southeast Asia. 1999. 156 p. Periodicals/serials International rice research notes, vol. 24, nos. 1-3. IRRI discussion paper series, No. 37. IRRI limited proceedings series, No. 1. IRRI technical bulletin series, Nos. 1-6. Agricultural Engineering Bell MA. 1999. A brighter future. Resource:13-14. Bell MA, Cedillo P. 1999. Mechanization in Asia: statistics and principles for success. Agric. Mech. Asia, Africa, Latin Am. 30(4):70-75. Agronomy, Plant Physiology, and Agroecology Azmi M, Mortimer AM. 1999. Effect of tillage practices, seeding rates and herbicides on barnyardgrass infestation (Echinochloa crusgalli) in direct seeded rice. In: Proceedings of the 17th Asian-Pacific Weed Science Society Conference, Bangkok, Thailand. p 199-204.
Bach Jensen L, Courtois B, Olofsdotter MO, Shen L, Mauleon RP, Li Z. 1999. See Plant Breeding, Genetics, and Biochemistry. Banziger M, Edmeades GO, Lafitte HR. 1999. Selection for drought tolerance increases maize yields across a range of nitrogen levels. Crop Sci. 39:1035-1040. Bird JA, Horwath WR, van Kessel C, Hill JE. 1999. Effect of flooding and straw residue management on soil bulk density in two flooded Vertisols: implications for nutrient cycling research. ASA-CSSA-SSSA Annual Meeting Abstracts, Salt Lake City, Utah, 31 Oct-4 Nov. p 237. Breen JL, Hill JE, Kusanagi T. 1999. Tiller density determines competitive outcome between water-seeded rice (Oryza sativa L.) and Monochoria vaginalis var. vaginalis. J. Weed Sci. Technol. 44(3):180-188. Castella J-C, Gayte O, Do Minh Phuong. 1999. Between micro- and macrolevel studies, the quest for a meso-level for effective community-based natural resource management in Vietnam uplands. In: Kam SP, Hoanh CT, editors. Scaling methodologies in ecoregional approaches for natural resource management. Limited proceedings of an international workshop, 22-24 Jun 1998, Ho Chi Minh City, Vietnam. Makati City, Philippines: Intenational Rice Research Institute. p 93-107. Castella J-C, Husson O, Le Quoc Doanh, Ha Dinh Tuan. 1999. Implementing the ecoregional approach in the Red River Basin uplands, Vietnam. In: Kinh NN et al, editors. Towards an ecoregional approach for natural resource management in the Red River Basin of Vietnam. Hanoi, Vietnam: The Agricultural Publishing House. p 75-94.
145
Castella J-C, Husson O, Le Quoc Doanh, Ha Dinh Tuan. 1999. Mise en oeuvre de lapproche corgionale dans les montagnes du bassin du Fleuve Rouge au Vietnam [in French]. Cah. Rech.-Dv. 45:114-134. Castella J-C, Jourdain D, Trbuil G, Napompeth B. 1999. A systems approach to understanding obstacles to effective implementation of IPM in Thailand: key issues for the cotton industry. Agric. Ecosyst. Environ. 72:17-34. Castella J-C, Kam SP, Hoanh CT. 1999. The Red River Basin (Vietnam) pilot region for operational ecoregional approach: new mechanisms for reconciling macro and micro studies, biophysical and socioeconomic perspectives in natural resource management. Contribution to SAAD-3. The Third International Symposium on Systems Approaches for Agricultural Development: Methodologies for Interdisciplinary, Multi-scale Perspectives, 8-10 Nov 1999, Lima, Peru. 15 p. Caton BP, Foin TC, Hill JE. 1999. A plant growth model for integrated weed management in direct-seeded rice. I. Development and sensitivity analyses of monoculture growth. Field Crops Res. 62:129-143. Caton BP, Foin TC, Hill, JE. 1999. Plant growth model for integrated weed management in direct-seeded rice. II. Validation testing of waterdepth effects on monoculture growth. Field Crops Res. 62:145-155. Caton BP, Foin TC, Hill JE. 1999. A plant growth model for integrated weed management in direct-seeded rice. III. Interspecific competition for light. Field Crops Res. 63(1):47-61. Caton BP, Mortimer AM, Foin TC, Hill JE, Gibson KD, Fischer AJ. 1999. Weed morphology effects on competitiveness for light in directseeded rice. In: Proceedings of the 17th AsianPacific Weed Science Society Conference, Bangkok, Thailand. p 116-120. Chauhan RPS, Singh BB, Singh VP. 1999. Strategic nutrient management for sustained rice production in sodic soils. Fert. Newsl. 44(8):1326. Chopart LJ, Siband P. 1999. Development and validation of a model to describe root length density of maize from root counts on soil profiles. Plant Soil 214:61-74.
Cohen MB, Jackson MT, Lu BR, Morin SR, Mortimer AM, Wade LJ. 1999. Predicting the environmental impact of transgene outcrossing to wild and weedy rices in Asia. In: Lutman PJW, editor. Gene flow and agriculture: relevance for transgenic crops. British Crop Protection Council Symposium. p 151-157. Cooper M, Fukai S, Wade LJ. 1999. How can breeding contribute to more productive and sustainable rainfed lowland rice systems? Field Crops Res. 64:199-209. Dizon MA, Migo TR, Mortimer AM. 1999. Understanding weed competition in direct seeded rice: an examination of Echinochloa crus-galli at the plant level. Philipp. J. Crop Sci. 2:81. Eagle AJ, van Kessel C, Horwath WR, Bird JA, Hill JE. 1999. Nitrogen cycling dynamics under alternative rice straw management practices. ASA-CSSA-SSSA Annual Meeting Abstracts, Salt Lake City, Utah, 31 Oct-4 Nov. p 220. Edmeades GO, Bolanos J, Chapman SC, Lafitte HR, Banziger M. 1999. Selection improves drought tolerance in tropical maize populations. I. Gains in biomass, grain yield, and harvest index. Crop Sci. 39:1306-1315. Fukai S, Cooper M, Wade LJ. 1999. Adaptation of rainfed lowland rice. Field Crops Res. 64:1-2. (Preface) Fukai S, Cooper M, Wade LJ. 1999. Adaptation of rainfed lowland rice. Field Crops Res. 64:1210. (Special issue) Garcia CO, Piggin CM, Janiya JD, Bell MA, Castro E, Razote E, Hill JE. 1999. Establishment of irrigated lowland rice under zero and conventional tillage systems in the Philippines. In: Proceedings of the 17th Asian-Pacific Weed Science Society Conference, Bangkok, Thailand. p 226-231. George T, Quiton J, Manoharan V, Corton T, Yost R. 1999. Adaptation of the nutrient management support system (NuMaSS) in Asia. In: Proceedings of the International Symposium on Systems Research for Optimizing Future Land Use. International Rice Research Institute, 1113 Oct 1999, Los Baos, Laguna, Philippines. Huang S, Yu L, Watson AK. 1999. Study of conidia production characterization of barnyardgrass fungi Alternaria alternata and Curvularia lunata. In: Proceedings of the 6th Weed Science
146
Conference of China, 19-21 Mar 1999, Nanning, China. p 165-169. Inderjit, Olofsdotter M. 1999. Allelopathy in agroecosystem. In: Proceedings of the 11 th EWRS (European Weed Research Society) Symposium, Basel. p 63. Janiya JD, Dizon MA, Mortimer AM, Piggin C, Hill JE. 1999. The impact of cropping practices on rice-weed communities. In: Proceedings of the 17th Asian-Pacific Weed Science Society Conference, Bangkok, Thailand. p 146-151. Kam SP, Castella J-C, Hoanh CT. 1999. Methodology Integration under the Ecor(I)Asia. Contribution to the Symposium on Systems Research for Optimizing Future Land Use (SysNet99), 11-13 Oct 1999, International Rice Research Institute, Los Baos, Philippines. 5 p. Kamoshita A, Wade LJ, Siopongco J, Sarkarung S, Nguyen HT. 1999. Phenotyping and tagging genes for constitutive root traits. Philipp. J. Crop Sci. 24:8. Kamoshita A, Wade LJ, Yamauchi A. 1999. Prospect for crop-soil research for rainfed lowland rice improvement. In: Horie T, Geng S, Amano T, Inamura T, Shiraiwa T, editors. World food security and crop production technologies for tomorrow. Kyoto: Crop Science Society of Japan. p 217-220. Kim JK, Peng S, Park ST. 1999. Direct seeding technology for irrigated rice in Asia. J. Korean Soc. Int. Agric. 11(3):225-233. Kinh NN, Teng PS, Hoanh CT, Castella JC. 1999. Towards an ecoregional approach for natural resource management in the Red River Basin of Vietnam. Hanoi, Vietnam: The Agricultural Publishing House. 254 p. Kumar D, Chauhan RPS, Singh BB, Singh VP. 1999. Response of rice to zinc sulfate incubated and blended with organic materials. Indian J. Agric. Sci. 66(6):402-405. Kyu H, Amarante ST, Gomez A, Samonte HP, Robles RP, Wade LJ. 1999. Variation in seed and seedling vigor in wet-seeded rice. Philipp. J. Crop Sci. 24:42. Latore J, Gould P, Mortimer AM. 1999. Effects of habitat heterogeneity and dispersal strategies on population persistence in annual plants. Ecol. Model. 58:382-389.
Lhoste P, Tonneau JP, Trebuil G. 1999. Recherche corgionale et dveloppement rgional: enjeux, dmarche et outils [in French]. Cah. Rech.-Dv. 45:7-36. Mabbayad MO, Watson AK. 1999. Rejection of Fusarium pallidoroseum as a biological control agent of Mimosa invisa in upland rice. In: Program Abstracts, Tenth International Symposium on Biological Weed Control, 4-14 Jul 1999, Bozeman, Montana. p 121. Manichon H, Trebuil G. 1999. An ecoregional approach for development-oriented research on agricultural systems. In: Kinh NN, Teng PS, Hoanh CT, Castella JC, editors. Towards an ecoregional approach for natural resource management in the Red River Basin of Vietnam. Hanoi, Vietnam: The Agricultural Publishing House. p 13-28. Masangkay RF, Mabbayad MO, Paulitz TC, Watson AK. 1999. Host range of Alternaria alternata f. sp. sphenocleae causing leaf blight of Sphenoclea zeylanica. Can. J. Bot. 77:103112. Masangkay RF, Paulitz TC, Hallett SG, Watson AK. 1999. Factors influencing biological control of Sphenoclea zeylanica with Alternaria alternata f. sp. sphenocleae. Plant Dis. 83:1019-1024. Mortimer AM, Caton BP, Hill JE. 1999. On ecological issues in the development of sustainable weed management. In: Proceedings of the 17th Asian-Pacific Weed Science Society Conference, Bangkok, Thailand. p 45-50. Mortimer AM, Hill JE. 1999. Weed species shifts in response to broad-spectrum herbicides in sub-tropical and tropical crops. BCPC 2:425437. Murchie EH, Chen Y, Hubbart S, Peng S, Horton P. 1999. Interactions between senescence and leaf orientation determine in situ patterns of photosynthesis and photoinhibition in field-grown rice. Plant Physiol. 119:553-563. Olofsdotter M. 1999. Allelopathya future component in weed management. In: Proceedings of the 16th Danish Crop Protection Conference on Weeds. p 101-112. Olofsdotter M. 1999. Lessons to be learned from rice allelopathy. In: Proceedings of the Second World Congress on AllelopathyCritical
147
Analysis and Future Prospects, Thunder Bay, Canada, 8-13 Aug 1999. (Abstract). p 141. Olofsdotter M, Inderjit. 1999. Allelopathy. In: Jensen JE, Andreasen, Sieden P, editors. Weed science. KVL. p 337-355. Olofsdotter M, Navarez D, Rebulanan M, Streibig JC. 1999. Weed suppressing rice cultivars does allelopathy play a role? Weed Res. 39(6):441-454. Peng S, Cassman KG, Virmani SS, Sheehy JE, Khush GS. 1999. Yield potential trends of tropical rice since the release of IR8 and the challenge of increasing rice yield potential. Crop Sci. 39:1552-1559. Peng S, Sanico AL, Garcia FV, Laza RC, Visperas RM, Descalsota JP, Cassman KG. 1999. Effect of leaf phosphorus and potassium concentration on chlorophyll meter reading in rice plant. Plant Prod. Sci. 2(4):227-231. Pheng S, Adkins S, Olofsdotter M, Jahn G. 1999. Allelopathic effects of rice (Oryza sativa) on the growth of awnless barnyardgrass (Echinochloa colona (L.) Link.): A new form for weed management. Cambodian J. Agric. 2(1):42-49. Pheng S, Adkins S, Olofsdotter M, Jahn G. 1999. The search for allelopathic rice in Cambodia: allelopathic effects of modern rice varieties, Cambodian traditional rice varieties (Oryza sativa L.) and wild rice (Oryza nivara Sharma et Shastry and Oryza rufipogon Griff.) on awnless barnyardgrass [Echinochloa colona (L.) Link]. Proceedings of the 17th Asian-Pacific weed Science Society Conference, Weeds and Environmental Impact, Bangkok, 22-27 Nov 1999. Ram PC, Singh AK, Singh BB, Singh VK, Singh HP, Setter TL, Singh VP, Singh RK. 1999. Characterization of floodwater in eastern India: relevance to submergence tolerance of rice. Exp. Agric. 33:141-152. Rodriguez R, Mazid MA, Harnpichitvitaya D, Wade LJ, Raab R, Balasubramanian V, Bell MA, Teng PS. 1999. Transfer of rice-related technologies in partnership with researchers, farmers, and nongovernment organizations. Philipp. J. Crop Sci. 24:31. Roetter R, Hoanh CT. 1999. Exploring land use options under multiple goals in support of natural
resource management at subnational level. In: Kinh NN, Teng PS, Hoanh CT, Castella JC, editors. Towards an ecoregional approach for natural resource management in the Red River Basin of Vietnam. Hanoi, Vietnam: The Agricultural Publishing House. p 29-57. Roetter R, Van de Geijn SC. 1999. Climate change effects on plant growth, crop yield and livestock. Clim. Change 43:651-681. Sheehy JE, Mitchell PL, Tsukaguchi T, Dionora MJA, Ferrer A, Torres R. 1999. Breaking the yield barrier in rice: problems and solutions. Does radiation use efficiency limit rice yield in the tropics. In: Horie T, Geng S, Amano T, Inamura T, Shiraiwa T, editors. World food security and crop production technologies for tomorrow. Kyoto University, Kitashirakawa, Sakyou-ku, Kyoto, Japan. p 147-151. Sinclair TR, Sheehy JE. 1999. Erect leaves and photosynthesis in rice. Science 283:1456-1457. Singh G, Singh OP, Singh RK, Singh VP, Nayak R. 1999. Production potential of rice-based cropping systems in rainfed lowland situations. Oryza 36(1):57-62. Singh U, Patil SK, Das RO, Padilla JL, Singh VP, Pal AR. 1999. Nitrogen dynamics and crop growth on an alfisol and vertisol under rainfed lowland rice-based systems. Field Crops Res. 61:237-252. Singh VP, Minh VQ, Singh AN, Kam SP. 1999. Ecosysem analysis-based methodology for technology extrapolation. In: Balasubramanian V et al, editors. Resource management in rice systems: nutrients. The Netherlands: Kluwer Academic Publishers. p 213-229. Singh VP, Singh RK, Sastri ASRAS, Baghel SS, Chaudhry JL. 1999. Agroclimatic atlas of Eastern India. Indira Gandhi Agricultural University and Makati (Philippines): International Rice Research Institute. 76 p. Smitaman P, Panyafoo S, Watson AK. 1999. Screening of selective biocontrol agents for Leptochloa chinensis (L.) Nees. in Thailand. In: Proceedings of the 17th Asian-Pacific Weed Science Society Conference, 22-27 Nov, Bangkok, Thailand. p 567-572. Srivastava PC, Gangwar MS, Singh VP. 1999. Adsorption-desorption of zinc in mollisols and their relationship with the uptake of fertilizer
148
applied zinc by rice. Commun. Soil Plant Anal. 30(3-4):471-481. Srivastava PC, Ghosh D, Singh VP. 1999. Evaluation of different zinc sources for lowland rice production. Biol. Fertil. Soil 30(1 & 2):168172. Tsukaguchi T, Sheehy JE, Ito O, Horie T. 1999. Analysis of grain filling of individual tillers of rice based on their emergence time and NSV content in genotypes of IR72 and new plant type. In: Horie T, Geng S, Amano T, Inamura T, Shiraiwa T, editors. World food security and crop production technologies for tomorrow. Kyoto University, Kitashirakawa, Sakyou-ku, Kyoto, Japan. p 298-299. Turkelboom F, Trebuil G. 1999. Approche intgre de lrosion des sols en milieu paysan : application aux collines du nord de la Thalande. In: Gestion agrobiologique des sols et des systmes de culture. Rasolo F, Raunet M, editors. Actes de latelier international, Antsirabe, Madagascar, 23-28 mars 1998, Anae, Cirad, Fafiala, Fifamor, Fofifa, Tafa. Montpellier, France, CIRAD, Collection Colloques. p 367383. Van Keer K, Trebuil G, Thiraton A. 1999. Le sel: un herbicide populaire sur riz pluvial au nordThalande. Agric. Dv. 21:99-109. Wade LJ. 1999. Research balance, impact delivery and future challenges in crop-soil management systems. In: Horie T, Geng S, Amano T, Inamura T, Shiraiwa T, editors. World food security and crop production technologies for tomorrow. Crop Science Society of Japan, Kyoto. p 247-249. Wade LJ, Amarante ST, Olea A, Harnpichitvitaya D, Naklang K, Wihardjaka A, Sengar SS, Mazid MA, Singh G, McLaren CG. 1999. Nutrient requirements in rainfed lowland rice. Field Crops Res. 64:91-107. Wade LJ, Fukai S, Samson BK, Ali A, Mazid MA. 1999. Rainfed lowland rice: physical environment and cultivar requirements. Field Crops Res. 64:3-12. Wade LJ, McLaren CG. 1999. Exploiting varietal and genetic suitability in variable environmentsexploiting the E in GXE. In: Geo-information techniques for understanding and analysing rainfed rice environments in eastern India. Lucknow, India, 24-26 May 1999.
Wade LJ, McLaren CG, Quintana L, Harnpichitvitaya D, Rajatasereekul S, Sarawgi AK, Kumar A, Ahmed HU, Sarwoto, Singh AK, Rodriguez R, Siopongco J, Sarkarung S. 1999. Genotype by environment interactions across diverse rainfed lowland rice environments. Field Crops Res. 64:35-50. Watson AK. 1999. Can viable weed control be attainable with microorganisms? In: Hong LW, Sastroutomo SS, Caunter IG, Ali J, Yeang LK, Vijaysegaran S, Sen YH, editors. Biological control in the tropics. 18-19 Mar 1999, Serdang, Malaysia. CAB International. p 5963. Zhang W, Watson AK. 1999. Isolation and partial characterization of phytotoxins produced by Exserohilum monoceras, a potential bioherbicide for control of Echinochloa species. In: Program abstracts. Tenth International Symposium on Biological Weed Control, 4-14 Jul 1999, Bozeman, Montana. p 71. Zhong X, Peng S, Sheehy JE, Liu H, Visperas RM. 1999. Parameterization, validation and comparison of three tillering models for irrigated rice in the tropics. Plant Prod. Sci. 2(4):258266. Yu L, Watson AK, Huang S, Xu Z, Wang Y. 1999. Control of Monochoria vaginalis in rice with the indigenous pathogenic fungi. China Rice Res. Newsl. 7(2):6-7. CREMNET Balasubramanian V. 1999. Farmer adoption of improved nitrogen management technologies in rice farming: technical constraints and opportunities for improvement. Nutr. Sci. Agroecosyst. 53:91-101. Balasubramanian V, Fischer KS. 1999. Rice: global achievements and advances in innovative technology development and dissemination. In: Proceedings of the 19th Session of the International Rice Commission, Cairo, Egypt, 7-9 Sep 1998. International Rice Commission on assessment and orientation towards the 21st century. Food and Agriculture Organization, Rome, Italy. p 181-193. Balasubramanian V, Morales AC, Cruz RT, Abdulrachman S. 1999. On-farm adaptation of knowledge-intensive nitrogen management
149
technologies for rice systems. Nutr. Cycl. Agroecosyst. 53:59-69. Balasubramanian V, Morales AC, Nagarajan R, Ramanathan SP. 1999. Efficiency of resin + wax- and polyolefin-coated urea in tropical flooded rice. 1999 Annual Meeting Abstracts. ASA-CSSA-SSSA, Salt Lake City, Utah, 31 Oct4 Nov 1999. p 40. Chuahan RPS, Singh BB, Singh RK, Singh VP. 1999. Strategic nutrient management for sustained rice production in sodic soil. Fert. News 44(8):13-16, 19-23, 25-26. Cox P, Jahn GC, Mak S, Chhorn N. 1999. Farmer participatory research for the design of improved rat management systems. Workshop on Developing a Research Framework for Farmer Participation in Rat Management, 19-23 July 1999, Ho Chi Minh, Vietnam. Cox P, Mak S. 1999. Participatory farmer research for the management of rats in Cambodia. Cambodian J. Agric. 2(1):23-28. Jahn GC, Mak S, Cox PG, Chhorn N. 1999. Farmer participatory research on rat management in Cambodia. In: G. Singleton, Hinds L, Leirs H, Zhang Z, editors. Ecologically based rodent management. p 358-371. Canberra (Australia): Australian Centre for International Agricultural Research. Khiev B, Jahn GC, Pol C, Chhorn N. 1999. Simulating rice pest damage to determine effects on yield. Cambodian J. Agric. 2(1):29-31. Kongchum M, Wuti Niponkit, Buresh RJ, Puckridge DW. 1999. Soil nitrate as a source of nitrogen for rainfed and deep-flooded rice. Thai Agric. Res. J. 17(2):174-184. Lantican M, Lampayan RM, Bhuiyan SI, Yadav MK. 1999. Determinants of improving productivity of dry-seeded rice in rainfed lowlands. Exp. Agric. (35):127-140. Mohanty SK, Singh U, Balasubramanian V, Jha KP. 1999. Nitrogen deep-placement technologies for productivity, profitability, and environmental quality of rainfed lowland rice systems. Nutr. Cycl. Agroecosyst. 53:43-57. Phaloeun C, Vuthy T, Sakkhunthea S, Nesbitt HJ. 1999. New technology and farming systems in Cambodia. Cambodian J. Agric. 2(1):38-41. Pheng S, Adkins S, Olofsdotter M, Jahn G. 1999. Allelopathic effects of rice on awnless
barnyardgrass. Cambodian J. Agric. 2(1):4248. Pheng S, Adkins S, Olofsdotter M, Jahn GC. 1999. The search for allelopathic rice in Cambodia. In: Proceedings of the 17th Asian-Pacific Weed Science Society Conference, 22-27 Nov 1999, Bangkok, Thailand. Ram PC, Singh AK, Singh BB, Singh VK, Singh HP, Setter TL, Singh VP, Singh RK.1999. Environmental characterization of flood water in eastern India: relevance to submergence tolerance of lowland rice. Exp. Agric. 35(2):141152. Rickman JF, Som B, Pao S, Meas P. 1999. Rice milling in Cambodia. Cambodian Agric. J. 2(1):49-51. Singh G, Singh OP, Singh RK, Singh VP, Nayak R. 1999. Production potential of rice-based cropping systems in rainfed lowland situation of eastern Uttar Pradesh, India. Oryza 36(1):5762. Singh RK, Prasad K. 1999. Farmers' participatory plant breeding. In: Proceedings of Summer School on Advances in Seed Sciences and Technology, New Delhi, India, 28 Apr-27 May 1999. p 263-270. Singh ON, Singh S, Singh RK, Sarkarung S. 1999. Screening for submergence tolerance using double-haploid rice lines for rainfed lowlands. Oryza 36(1):38-41. Singh RK. 1999. Genetic resources and the role of international collaboration in rice breeding. Genome 42:635-641. Singh RK. 1999. Planning for improved productivity and sustained rice production in eastern India. India Grains 1(3):15-16. Singh VP, Singh RK, Sastri ASRAS, et al. 1999. Rice-growing environments of eastern India: an agro-climatic atlas. IRRI-GAU. 76 p. Tang SX, Ding L, Wang ZQ. 1999. Exploiting biodiversity for sustainable pest management in rice field. World Agric. (237):28. Tang SX, Jiang YZ, Li SS. 1999. Observation on the amyloplasts in endosperm of early Indica rice with scanning electron microscpe. Acta Agron. Sin. 25(2):269-272. Tang SX, Jiang YZ, Zhang BD, Lu YL. 1999. Biodiversity of rice growing regions in China. Chin. Biodiversity 7(1):73-78.
150
Tang SX, Zhang WX, Liu J. 1999. The study on the Bi-Peak-tubercle on lemma of Hemudu and Luojiajiao ancient excavated rice grains with electric scanning microscope. Acta Agron. Sin. 25(3):269-272. White PF, Seng V, Ros C, Nesbitt HJ, Chan P. 1999. Nutrient and organic matter management in cash-poor, rice-based cropping systems in Cambodia: research and extension needs in nutrient and organic matter management in cashpoor rice-based farming systems. In: SE Asia: research and adoption needs in Laos, Thailand, Cambodia and Vietnam. Proceedings of an international workshop held in Vientiane, Laos, 21-22 Apr 1999. Canberra (Australia): Australian Centre for International Agricultural Research. Wade LJ, Amarante ST, Olea A, Harnpichitvitaya D, Naklang K, Wihardjaka A, Sengar SS, Mazid MA, Singh G, McLaren CG. 1999. Nutrient requirements in rainfed lowland rice. Field Crops Res. 64:91-107. Wade LJ, CG McLaren, L Quintana, Harnpichitvitaya D, Rajatasereekul S, Sarawgi AK, Kumar A, Ahmed HU, Sarwoto, Singh AK, Rodriguez R, Siopongco J, Sarkarung S. 1999. Genotype by environment interactions across diverse rainfed lowland rice environments. Field Crops Res. 64:35-50. Entomology and Plant Pathology Adhikari TB, Basnyat RC, Mew TW. 1999. Virulence of Xanthomonas oryzae pv. oryzae on rice lines containing single resistance genes and gene combinations. Plant Dis. 83:46-50. Arboleda M, Azzam O. 1999. Inter- and intrasite genetic diversity of natural field populations of rice tungro bacilliform virus in the Philippines. Arch. Virol. 144:1-15. Arboleda M, Sta Cruz F, Azzam O. 1999. Preliminary analysis of genetic variation of rice tungro bacilliform virus in two provinces of the Philippines. In: Chancellor TCB, Azzam O, Heong KL, editors. Proceedings of the Workshop on Rice Tungro Disease Management, 9-11 Nov 1998, Los Baos, Philippines. Azzam O, Yambao Ma. LM, Muhsin M, McNally K, Umadhay K. 1999. Genetic diversity of rice tungro spherical virus in tungro-endemic prov-
inces of the Philippines and Indonesia. Arch. Virol. 145:1183-1197. Cabauatan PQ, Melcher U, Ishikawa K, Omura T, Hibino H, Koganezawa H, Azzam O. 1999. Sequence changes in six variants of rice tungro bacilliform virus and their phylogenetic relationships. J. Gen. Virol. 80 (8):2229-2237. Cohen MB, Jackson MT, Lu BR, Morin SR, Mortimer AM, Wade LJ. 1999. Predicting the environmental impact of transgene outcrossing to wild and weedy rices in Asia. In: Gene flow and agriculture: relevance for transgenic crops. BCPC Symposium Proceedings no. 72. British Crop Protection Council: Brighton. p 151-156. Datta K, Velazhahan R, Oliva N, Ona I, Mew T, Khush GS, Muthukrishnan S, Datta SK. 1999. See Plant Breeding, Genetics, and Biochemistry. Escalada MM, Heong KL, Huan NH, Mai V. 1999. Communications and behavior change in rice farmers pest management: the case of using mass media in Vietnam. J. Appl. Commun. 83 (1):7-26. Heong KL. 1999. From research to farmer practicea case study in rice pest management. In: Balasubramanium V, Ladha JK, Denning GL, editors. Resource management in rice systems: nutrients. The Netherlands: Kluwer Academic Publishers. p 199-212. Heong KL. 1999. New paradigms and research opportunities in rice pest management. In: Zhang R, Gu D, Zhang W, Zhou C, Pang Y, editors. Proceedings of the International Symposium on IPM in Rice-Based Ecosystems, Guangzhou, Peoples Republic of China. 20-24 Oct 1997. Guangzhou, (China): Zhongshan University. p 3-14. Heong KL, Escalada MM. 1998. Changing rice farmers pest management practices through participation in a small-scale experiment. Int. J. Pest Manage. 44:191-197. Heong KL, Escalada MM. 1999. Quantifying rice farmers pest management decisions beliefs and subjective norms in stem borer control. Crop Prot. (18):315-322. Huan NH, Mai V, Escalada MM, Heong KL. 1999. Changes in rice farmers pest management between 1992 and 1997 in the Mekong Delta, Vietnam. Crop Prot. (18):557-563.
151
Islam Z, Heong KL. 1998. Abundance and predatory behavior of wasps of non-rice habitats in rice ecosystems in the Philippines. Bangladesh J. Entomol. 8:31-44. Islam Z, Heong KL. 1999. Effects of tillage on arthropod predators of rice insect pests in irrigated rice. In: Zhang R, Gu D, Zhang W, Zhou C, Pang Y, editors. Proceedings of the International Symposium on IPM in Rice-Based Ecosystems, Guangzhou, Peoples Republic of China, 20-24 Oct 1997. Guangzhou (China): Zhongshan University. p 198-208. Mew TW, Merca SD, Tuazon ED. 1999. In: Kahn RP, Mathur SB, editors. Containment facilities and safeguards for exotic plant pathogens and pests. St. Paul, Minnesota: APS Press. p. 38-43. Mew TW, Castilla CP, Huelma CC. 1999. Current status and future needs for inoculum threshold of seedborne pathogensthe basis for plant quarantine regulations. Abstract. p 9. Proceedings of the 3rd International Seed Testing Association-PDC, 16-19 Aug 1999, Ames, Iowa, USA. Mun JH, Song YH, Heong K. Roderick GK. 1999. Genetic variation among Asian populations of rice planthoppers, Nilaparvata lugens and Sogatella furcifera (Hemiptera:Delphacidae): mitochondrial DNA sequences. Bull. Entomol. Res. 89:245-253. Soriano IR, Schmit V, Brar DS, Prot J-C, Reversat G. 1999. Resistance to rice root-knot nematode Meloidogyne graminicola identified in Oryza longistaminata and O. glaberrima. Nematology 1(4):395-398. Sta Cruz FC, Boulton MI, Hull R, Azzam O. 1999. Agroinoculation allows the screening of rice for resistance to rice tungro bacilliform virus. J. Phytopathol. 147(12):653-659. Wang GL, Leung H. 1999. Molecular biology of host-pathogen interactions in rice diseases. In: Shimamoto K, editor. Molecular biology of rice. Berlin: Springer-Verlag. p 201-232. Zhang R, He X, Heong KL. 1999. Assessing impacts of climate change on paddy borer Scirpophaga incertulas (Walker) in Guangdong province, China. In: Zhang R, Gu D, Zhang W, Zhou C, Pang Y, editors. Proceedings of the International Symposium on IPM in Rice-Based Ecosystems, Guangzhou, Peoples
Republic of China. 20-24 Oct 1997. Guangzhou (China): Zhongshan University. p 245-253. Genetic Resources Center Cohen MB, Jackson MT, Lu BR, Morin SR, Mortimer AM, Pham JL, Wade LJ. 1999. Predicting the environmental impact of transgene outcrossing to wild and weedy rices in Asia. In: 1999 PCPC Symposium Proceedings No. 72. Gene flow and agriculture: relevance for transgenic crops. Proceedings of a symposium held at the University of Keele, Staffordshire, UK, 12-14 Apr 1999. p 151-157. Ge S, Sang T, Lu BR, De-Yuan Hong. 1999. Phylogeny of rice genomes with emphasis on origins of allotetraploid species. Proc. Natl. Acad. Sci. 96:14400-14405. Jackson MT. 1999. Managing genetic resources and biotechnology at IRRIs rice genebank. In: Cohen JI, editor. Managing agricultural biotechnologyaddressing research program and policy implications. The Hague, The Netherlands: International Service for National Agricultural Research (ISNAR), and UK: CAB International. p 102-109. Naredo MEB, Juliano AB, Lu BR, de Guzman F, Jackson MT. 1998. Responses to seed dormancy-breaking treatments in rice species (Oryza L.). Seed Sci. Technol. 26:675-689. Parsons B, Newbury HJ, Jackson MT, Ford-Lloyd BV. 1999. The genetic structure and conservation of aus, aman and boro rices from Bangladesh. Genet. Res. Crop Evol. 46:587-598. Thurston HD,Salik J, Smith M, Trutmann P, Pham JL, McDowell R. 1999. Traditional management of agrobiodiversity. In: Wood D, Lenne JM, editors. Agrobiodiversity: characterization, utilization and management. CABI. p 211-244. Zhu JH, Stephenson P, Laurie DA, Li W, Tang D, Jackson MT, Gale MD. 1999. Towards rice genome scanning by map-based AFLP fingerprinting. Mol. Gen. Genet. 261:184-295.
152
Plant Breeding, Genetics, and Biochemistry Aggarwal RK, Brar DS, Nandi S, Huang N, Khush GS. 1999. Phylogenetic relationships among Oryza species as revealed by AFLP markers. Theor. Appl. Genet. 98:1320-1328. Alam MF, Datta K, Vasquez A, Tu J, Virmani SS, Datta SK. 1999. Transgenic insect resistant maintainer line (IR68899B) for improvement of hybrid rice. Plant Cell Rep. 18:572-575. Alam MF, Datta K, Abrigo E, Oliva N, Tu J, Virmani SS, Datta SK. 1999. Transgenic insect-resistant maintainer line (IR68899B) for improvement of hybrid rice. Plant Cell Rep. 18:571-575. Arakawa A, Fukuta Y, Yagi T, Tamura K, Kudo K. 1999. QTL analysis of awn length in rice Milyang23/Akihikari recombinat Inbred lines using DNA marker. Hokuriku Crop Sci. 34:9496. Ashikawa I, Fukuta Y, Tamura K, Yagi T. 1999. Application of AFLP technique that uses nonradioactive fluorescent primers to the detection of genetic diversity in Japanese rice cultivars and cloning of DNA sequences derived from an Indica genome. Breed. Sci. 49:225-231. Chareonpornwattana S, Krishnarajapuran VT, Wang L, Datta SK, Panbangred W, Muthukrishnan S. 1999. Inheritance, expression, and silencing of a chitinase transgene in rice. Theor. Appl. Genet. 98:371-378. Cho Y-C, Shui Y-S, Ahn S-N, Gregorio GB, Kang K-H, Brar DS, Moon HP. 1999. DNA fingerprinting of rice cultivars using AFLP and RAPD markers. Korean J. Crop Sci. 44:26-31. Datta K, Datta SK. 1999. Transformation of rice via PEG-mediated DNA uptake into protoplasts. In: Hall RD, editor. Methods in molecular biology, plant cell culture protocols. Vol. 3. New Jersey: Humana Press, Inc. p 335-347. Datta K, Muthukrishnan S, Datta SK. 1999. Expression and function of PR-protein genes in transgenic plants. In: Datta SK, Muthukrishnan S, editors. Pathogenesis-related proteins in plants. USA: CRC Press. p 261-277. Datta K,Velazhahan R, Oliva N, Mew T, Khush GS, Muthukrishnan S, Datta SK. 1999. Over expression of cloned rice thaumatin-like protein (PR-5) gene in transgenic rice plants enhances environmental friendly resistance to
Rhizoctonia solani causing sheath blight disease. Theor. Appl. Genet. 98:1138-1145. Datta SK, Muthukrishnan S, editors. 1999. Pathogenesis-related proteins in plants. USA: CRC Press. Datta SK. 1999. Transgenic rice: development, products, acceptance and economic impact. In: Plant biotechnology and in vitro biology in the 21st century. The Netherlands: Kluwer Academic Publisher. p 737-740. Datta SK. 1999. Transgenic cereals: Oryza sativa (rice) In: Vasil IK, editor. Molecular improvement of cereal crops. Vol. 5. The Netherlands: Kluwer Academic Publisher. p 149-187. Fukuta Y, Fujita Y, Tamura K, Yagi T. 1999. QTL analysis for resistance to rice leaf blast using Milyang 23/Akihikari recombinant inbred lines. Hokuriku Crop Sci. 34:90-93. Fukuyama T, Sasahara H, Fukuta Y. 1999. Variation of vascular bundle system corresponds to indica, tropical- and temperate-japonica differentiation of Asian rice (Oryza sativa L.). Breed. Sci. 49:15-19. Graham RD, Senadhira D, Beebe S, Iglesias L, Monasterio I. 1999. Breeding for micronurient density in edible portions of staple food crops: conventional approach. Field Crop Res. 60:5780. Jelodar NB, Blackhall NW, Hartman TPV, Brar DS, Khush GS, Darvey MR, Cocking EC, Power JB. 1999. Intergeneric somatic hybrids of rice [Oryza sativa L. (+) Porteresia coarctata (Roxb.) Tateoka]. Theor. Appl. Genet. 99:570577. Jensen LB, Courtois B, Olofsdotter M, Shen LS, Mauleon RP, Li ZK. 1999. Locating genes controlling allelopathic effects against Echinochloa crus-galli in upland rice. In: Proceedings of the Second World Congress on AllelopathyCritical Analysis and Future Prospects, Thunder Bay, Canada, 8-13 Aug 1999. (Abstract). p 53. Khush GS. 1999. New plant type of rice for increasing the genetic yield potential. In: Nanda JN, editor. Rice breeding and geneticsResearch priorities and challenges. Enfield, New Hampshire (USA): Science Publishers, Inc. p 99-108. Khush GS, Bennett J, Datta SK, Brar DS, Li Z. 1999. Advances in rice genetics and biotechnology. In: Proceedings of the 19th Session of
153
the International Rice Commission, Cairo, Egypt, 7-9 Sep 1998. International Rice Commission. Rome: FAO. p. 64-76. Khush GS, Sombilla MA, Hossain M. 1999. Livestock, ethics and quality of life in Asia: the food-feed dimension of grain demand. In: Hodges J, Han IK, editors. Livestock, ethics and quality of life. Wallingford, UK: CABI Publishing. p 175-198. Li ZK, Luo L, Tabien R, Mei H, Paterson AH, Zhao XH, Zhong DB, Wang DL, Wang YP, Ying CS, Stansel JW. 1999 A defeated resistance gene acts as a QTL against a virulent strain of Xanthonomas oryzae pv. oryzae. Mol. Gen. Genet. 261:58-63. Li ZK, Paterson AH, Pinson SRM, Stansel JW. 1999. RFLP facilitated analysis of tiller and leaf angles in rice (Oryza sativa L.). Euphytica 109(2):79-84. Lijun L, Li ZK, Mei H, Wang D, Zhou X, Wang Y, Zhong D, Yu X, Ying C, Paterson A. 1999. Heterosis performance and parental genetic diversity. Chin. J. Rice Sci. 13(1):6-10. Mei HW, Li ZK, Wang YP, Yu XQ, Zhong DB, Luo LJ, Ying CS, Paterson AH. 1999 Classification of the recombinant inbred lines from Lemont/Teqing cross based on the six indicajaponica differentiation characteristics. Chin. J. Rice Sci. 11(4):193-197. Mei HW, Luo LJ, Wang YP, Yuan XP, Zhao XH, Zhong DB, Yu XQ, Wang DL, Ying CS, Paterson AH, Li ZK. 1999. QTL mapping of horizontal resistance to bacterial blight Xanthomonas oryzae pv. oryzae in rice. Acta Genet. Sin. 26(4):345-349. Peng S, Cassman KG, Virmani SS, Sheehy J, Khush GS. 1999. Yield potential trends of tropical rice since the release of IR8 and the challenge of increasing rice yield potential. Crop Sci. 39:1552-1559. Reddy PM, Aggarwal RK, Ramos MC, Ladha JK, Brar DS, Kouchi H. 1999. See Soil and Water Sciences. Sanchez AC, Khush GS. 1998. Inheritance and linkage relationships of twenty-one genes in rice, Oryza sativa L. SABRAO J. 30:51-60. Sanchez AC, Ilag LL, Yang DC, Brar DS, Ausubel F, Khush GS, Yano M, Sasaki T, Li ZK, Huang N. 1999. Genetic and physical mapping of
xa13, a recessive bacterial blight resistance gene in rice. Theor. Appl. Genet. 98:10221028. Sasahara H. Fukuta Y, Fukuyama T. 1999. Mapping of QTLs for vascular bundle system and spike morphology in rice, Oryza sativa L. Breed. Sci. 49:75-81. Singh KN, Nandi R, Shanmugasundram P, Sadasivam S, Huang N, Brar DS, Khush GS. 1999. High-resolution DNA fingerprinting of Indian rice (Oryza sativa L.) varieties by amplified fragment length polymorphism. Euphytica 46:427-433. Soriano IR, Schmit V, Brar DS, Prot JC, Reversat G. 1999. Resistance to rice root-knot nematode Meloidogyne graminicola identified in Oryza longistaminata and O. glaberrima. Nematology 14:395-398. Tamura K, Fukuta Y, Hirae M, Oya S, Ashikawa I, Yagi T. 1999. Mapping of the Grh1 locus for green rice leafhopper resistance in rice using RFLP markers. Breed. Sci. 49:11-14. Virk PS, Pooni HS, Syed NH, Kearsey MJ. 1999. Fast and reliable validation of lines and crosses using microsatellite markers in Arabidopsis thaliana. Theor. Appl. Genet. 98:462-464. Social Sciences Ballabh V, Pandey S. 1999. Transitions in rice production systems in Eastern India: evidences from two villages in Uttar Pradesh. Econ. Polit. Week. Estudillo JP, Fujimura M, Hossain M. 1999. New rice technology and comparative advantage in rice production in the Philippines. J. Dev. Stud. 35(5):162-184. Godilano EC, DeGloria SD. 1998. Spatial analysis framework for community-based watershed management in the Philippines. Philipp. Remote Sensing Newsl. 9(1). Hayami Y, Marciano E, Kikuchi M. 1998. Did Philippine land reform reduce inequality? A perspective from a Laguna village. J. Agric. Econ. Dev. 26(1 and 2):17-38. Hossain M. 1999. Long-term outlook of the demand-supply balance for staple grains in Asian deltas: implications for food security strategy. In: Proceedings of the First International Sym-
154
posium on Sustainable Ecosystem Management, Planetary Garden 99, Chambery, 14-18 Mar 1999. p 348-357. Hossain M. 1999. Longterm perspective of the demand-supply balance for rice in Asia: implications for technological challenges. In: Horie T et al., editors. Proceedings of the International Symposium on World Food Security and Crop Production Technologies for Tomorrow. Kyoto University, 8-9 Oct 1999. p 31-39. Hossain M. 1999. Grassroots organizations for promoting sustainable agriculture and food security: the experience of Grameen Bank in Bangladesh. In: Balasubramanian V, Ladha JK, Denning GL, editors. Resource management in rice systems: nutrients. The Netherlands: Kluwer Academic Publishers. p 297-311. Hossain M, Sombilla M. 1999. Emerging trends in demand for cereal crops. In: Bindraban PS et al., editors. Food security at different scales: demographic, biophysical and socioeconomic considerations. Wageningen: DLO Research Institute for Agrobiology and Soil Fertility and The C.T. de Wit Graduate School for Production Ecology. Hossain M, Sombilla M. 1999. World grains market: implications for a food security strategy. In: Cabanilla L, Paunlagui M, editors. Food security in the Philippines. Institute of Strategic Planning and Policy Studies in cooperation with UP Center for Integrative and Development Studies. University of the Philippines. Kinh NN, Kam SP, Hoanh CT, Castella JC. 1999. Ecoregional approaches for natural resource management in the Red River Basin, Vietnam. Summary Proceedings of a technical workshop held in MARD Hanoi on 9-10 Nov 1998. MARD-IRRI, Los Banos. 42 p. Kinh NN, Teng PS, Hoanh CT, Castella JC, editors. 1999. Towards an ecoregional approach for natural resource management in the Red River Basin of Vietnam. Ministry of Agriculture and Rural Development, Hanoi, Vietnam and International Rice Research Institute, Los Baos, Philippines. 254 p. Kinh NN, Kam SP, Hoanh CT, Castella JC. 1999. Summary proceedings of Technical Workshop on Ecoregional Approaches for Natural Resources Management in the Red River Basin (RRB), Vietnam. 48 p.
Lapar L, Pandey S. 1999. Adoption of soil conservation: the case of the Philippine uplands. Agric. Econ. 21(3):241-246. Liew SCP, Chen CP, Kam SP, Tuong TP, Minh VQ, Lim H. 1999. Monitoring changes in rice cropping systems using space-borne SAR imagery. Proceedings of the 1999 International Geoscience and Remote Sensing Symposium. 2:741-743. Lucas MP, Pandey S, Villano RA, Culanay DR, Obien SR. 1999. Characterization and economic analysis of intensive cropping systems in rainfed lowlands of Ilocos Norte, Philippines. Exp. Agric. 35:211-224. Pandey S. 1999. Adoption of nutrient management technologies for rice production: economic and institutional constraints and opportunities. Nutr. Cycl. Agroecosyst. 53:103-112. Pandey S, Rajatsereekul S. 1999. Economics of plant breeding: the value of shorter breeding cycles for rice in northeast Thailand. Field Crops Res. 64(1/2):187-197. Pandey S, Masicat P, Velasco L, Villano R. 1999. Risk analysis of a rainfed rice production systems in Tarlac, Central Luzon, Philippines. Exp. Agric. 35:225-237. Paris T, Singh A, Luis J, Hossain M, Singh HN, Singh S, Singh ON. 1999. Incorporating gender concerns in participatory plant breeding and varietal selection: preliminary results from eastern India. In: Ashby J, Sperling L, editors. Proceedings of the International Workshop on Participatory Plant Breeding and Gender Analysis. Roetter R, Hoanh CT. 1999. Exploring land use options under multiple goals in support of natural resource management at sub-national level. In: Kinh NN, Teng PS, Hoanh CT, Castella JC, editors. Towards an ecoregional approach for natural resource management in the Red River Basin of Vietnam. Ministry of Agriculture and Rural Development, Hanoi, Vietnam and International Rice Research Institute, Los Baos, Philippines. p 29-57. Truong Thi Ngoc Chi, Price LL, Hossain M. 1998. Effect of IPM-farmer field school on the male and female rice farmers insect management knowledge and pest control practices in Can Tho, Vietnam. Philipp. J. Crop Sci. 23(1): 5358.
155
Wang DL, Zhu J, Li ZK, Paterson AH. 1999. Mapping QTLs with epistatic effects and genotype x environment interactions by mixed linear model approaches. Theor. Appl. Genet. 99:1255-1264. Soil and Water Sciences Adhikari C, Bronson KF, Panuallah GM, Regmi AP, Saha PK, Dobermann A, Olk DC, Hobbs PR, Pasuquin E. 1999. On-farm soil N supply and N nutrition in the rice-wheat system of Nepal and Bangladesh. Field Crops Res. 64:273-286. Agustin EO, Ortal CI, Pascua SR, Sta. Cruz PC, Padre AT, Ventura WB, Obien SR, Ladha JK. 1999. Role of indigo in improving the productivity of rainfed lowland rice-based cropping systems. Exp. Agric. 35:201-210. Alberto MCR, Arah JRM, Neue HU, Wassmann R, Lantin RS, Aduna JB, Bronson KF. 1999. A sampling technique for the determination of dissolved methane in soil solution. Chemosphere: Global Change Sci. 2(2000):5763. Boling A, To Phuc Tuong, Anil Kumar Singh, Wopereis MCS. 1998. Comparative root growth and soil water extraction of dry-seeded, wet-seeded, and transplanted rice in a greenhouse experiment. Philipp. J. Crop Sci. 23(1):45-52. Briones Jr. AM, Reichardt W. 1999. Estimating microbial population counts by most probable number using Microsoft Excel. J. Microbiol. Meth. 35:157-161. Chalk PM, Ladha JK. 1999. Estimation of legume symbiotic dependence: an evaluation of techniques based on 15N dilution. Soil Biol. Biochem. 31:1901-1917. Gumtang RJ, Pampolino MF, Tuong TP. 1999. Ground water dynamics and quality under intensive cropping systems. J. Exp. Agric. 35:153-166. Kirk GJD. 1999. A model of phosphate solubilization by organic anion excretion from plant roots. Eur. J. Soil Sci. 50:369-378. Kirk GJD, Santos EE, Findenegg GR. 1999. Phosphate solubilization by organic anion excretion from rice (Oryza sativa L.) growing in aerobic soil. Plant Soil 211:11-18.
Kirk GJD, Santos EE, Santos MB. 1999. Phosphate solubilization by organic anion excretion from rice growing in aerobic soil: rates of excretion and decomposition, effects on rhizosphere pH, and effects on phosphate solubility and uptake. New Phytol. 142:185-200. Kouchi H, Takane K, So R, Ladha JK, Reddy PM. 1999. Rice Enod40: isolation and expression analysis in rice and transgenic soybean root nodules. Plant J. 18:121-129. Kouchi H, Takane K, So RB, Ladha JK, Reddy PM. 1999. Rice ENOD40: Isolation and expression analysis in rice and transgenic soybean root nodules. Plant J. 18:121-129. Kronzucker HJ, Siddiqi MY, Glass ADM, Kirk GJD. 1999. Nitrate-ammonium synergism in rice: a subcellular flux analysis. Plant Physiol. 119:1041-1045. Kundu DK, Ladha JK. 1999. Sustaining productivity of lowland rice soils: issues and options related to N availability. Nutr. Cycl. Agroecosyst. 53:19-33. Malarvizhi P, Ladha JK. 1999. Influence of available nitrogen and rice genotype on associative dinitrogen fixation. Soil Sci. Soc. Am. J. 63:9399. Pascua SR,Ventura W, Agustin EO, Padre AT, Valencia DA, Marcos TF, Sta. Cruz PC, Obien SR, Ladha JK. 1999. Yield trends and apparent nutrient balances in intensified and diversified rice-based cropping systems. Exp. Agric. 35:181-199. Reddy PM, Aggarwal PK, Ramos MC, Ladha JK, Brar DS, Kouchi H. 1999. Widespread occurrence of the homologues of the early nodulin (ENOD) genes in Oryza species and related grasses. Biochem. Biophys. Res. Commun. 258:148-154. Reddy PM, Aggarwal RK, Ramos MC, Ladha JK, Brar DS, Kouchi H. 1999. Widespread occurrence of the homologues of the early nodulin (ENOD) genes in Oryza species and related grasses. Biochem. Biophys. Res. Commun. 258:148-154. Revsbech NP, Pedersen O, Reichardt W, Briones A. 1999. Microsensor analysis of oxygen and pH in the rice rhizobia under field and laboratory conditions. Biol. Fertil. Soils 29:379-385.
156
Tuong TP, Bhuiyan SI. 1999. Increasing water use eficiency in rice poduction: farm level perspectives. Agric. Water Manage. 40:117-122. Wihardjaka A, Kirk GJD, Abdulrachman S, Mamaril CP. 1999. Potassium balances in rainfed lowland rice on a light-textured soil. Field Crops Res. 64:237-247.
Rice research seminars

Disease resistance, rice mutants, and functional genomics. Dr. H. Leung. The ecology of plant virus disease. Dr. M. Thresh, chair, Virus Epidemiology Committee, International Society of Plant Pathologists, UK. Plant breeding perspectives: balanced integration of approaches. Dr. A. Ashri, Faculty of Agriculture, The Hebrew University of Jerusalem, Israel. Rice research development through Thai-IRRI collaboration. Dr. B. Somrith, candidate, IRRI liaison scientist (consultant) for Thailand. Operationalizing the ecoregional approach in the uplands of the Red River Basin. Dr. J.-C. Castella. Juggling science and management: seven years in the IRRI circus. Dr. R.S. Zeigler. A balancing act: nutrient equilibrium for sustainable crop production. Dr. J.K. Ladha. Virginia techs involvement in global food security, food safety, and natural resources management issues. Dr. S.K. De Datta, director, Office of International Research and Development, and associate dean, College of Agriculture and Life Sciences, Virginia Tech, Blacksburg, Virginia, USA. Temporal and spatial variability of optimal nitrogen applications. Dr. D. Dawe. The road to impact. Dr. A. Dobermann. Evaluation, evaluators, and evaluation culture. Dr. G.T. Castillo. Fertile feedback: an evolutionary approach to achieving impact. Mr. B. Douthwaite. Proprietary property: what is it and why do I care? Mr. D. Kryder, intellectual property specialist, International Service for the Acquisition of Agri-Biotech Applications New York, USA. Spatial analysis framework for community-based natural resource management in the Philip-
pines, implications to new CGIAR mission. Dr. S.C. Godilano. The universe, the evolution of the perverse, and a rice problem. Dr. J.E. Sheehy. Transgenic rice: lab to field, constraints and impact. Dr. S.K. Datta. Farmer participatory research in rice technology development and dissemination. Dr. V. Balasubramanian. Conceptual and methodological issues in rice field biodiversity research: lessons from invertebrates. Dr. K.G. Schoenly. Changing rice production situations in lowland rice and implications to plant protection. Dr. S. Savary. The role of an R&D network in a CGIAR center: the case of CIP-UPWARD. Dr. D.M. Campilan, social scientist and UPWARD network coordinator, International Potato Center. Rice and the Philippine media: getting through. Mr. J.C. Burgos, Jr., publisher-editor, We Forum, Quezon City, Philippines. To direct-seed or not to direct-seed: some economic considerations. Dr. S. Pandey. Introducingthe new International Rice Research Notes (IRRN): content, creed, and continuing conundrums. Dr. M.B. Cohen. INM research in Cambodia: the end of one phase and the beginning of another. Dr. P. White. Everything you always wanted to know about the IRRI Librarybut were afraid to ask. Mr. I. Wallace. Food security in Asian countries in the 21st century. Dr. K. Ohaga, Professor, Department of Global Agricultural Sciences, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Japan. SysNet methodology development for agricultural land-use planning in South and Southeast Asia. Dr. R. Roetter. Molecular markers for common beans at CIAT. Dr. S. Beebe, common bean breeder, Centro Internacional de Agricultura Tropical, Cali, Colombia. Small RNA viruses of insects: Helicoverpa armigera stunt virus and new transgenic strategies for pest control. Dr. K. Gordon, Biotechnology Program, Commonwealth Scientific and Industrial Research Organization Entomology, Canberra, Australia.
157
Bacillus thuringiensis: an overview. Dr. R. Frutos, Centre de coopration intenationale en recherche agronomique pour le dveloppement, Montpellier, France. Pesticidesrumors, gossip, and a few facts. Mr. C. Meek, expert consultant on agrochemical applications, Experiment Station.
Division seminars
Agronomy, Plant Physiology, and Agroecology System of rice intensification. Dr. N. Uphoff, director, Cornell International Institute for Food, Agriculture, and Development, Cornell University, Ithaca, New York, USA. Plant-available silicon in slag fertilizer and soil. Dr. N. Kato, visitor collaborator, Laboratory of Lowland Soils, Tohoku National Agricultural Experiment Station, Omagari, Akita, Japan. Changes in root water uptake power of waterstressed upland rice. Ms. K. Okada. Integrated on-farm diagnosis on upland rice crop in northern Thailand. Dr. G. Trebuil and Dr. K. Van Keer. Progress toward natural weed suppression in rice and an overview of rice research objectives at the USDA-ARS Dale Bumpers National Rice Research Center, Stuttgart, Arkansas, USA. Dr. D. Gealy, plant physiologist, USDA-ARS, Stuttgart, Arkansas, USA. Upland rice-based research in northern Laos: stabilizing swidden systems. Dr. K. Fahrney. CREMNET Role of farmer participatory research in rice technology development and dissemination. Dr. V. Balasubramanian. Entomology and Plant Pathology Polycistronic expression units for increased and selectable expression of transgenes in rice for engineered resistance to RTD. C. Henrich, Institute of Plant Science, ETH, Zurich, Switzerland.
Dissecting disease resistance pathway by forward and reverse genetics: progress and application. Dr. Wu Changjian. Bacterial populations associated with rice seed. Mr. B. Cottyn. Genetic diversity of the rice blast pathogen and the evolution of races. Dr. J. Correll, visiting scientist from the University of Arkansas. Molecular characterization of tungro biological variants. Dr. O. Azzam and Dr. P. Cabauatan. Attempts to elucidate the role of rice dwarf virus genes in multiplication in cells. Dr. I. Uyeda, collaborator, IRRI-Japan Shuttle Research. Some important issues in stem borer management and the potential of rice plants to compensate for insect pest injury. Dr. E. R.-Sanchez. Using conserved motifs of plant resistance genes to characterize rice functional genetic diversity. Dr. Zonghua Wang. Simulation modeling of rice yield losses due to pests, diseases, and weeds. Dr. L. Willocquet and Ms. L.C. Fernandez. Rice sheath blight research: the disease, its cycle and importanceDr. S. Savary; Epidemiogy N. Castilla; Biological controlDr. T.W. Mew; Crop establishment methodsDr. L. Willocquet; Transgenic rice for resistanceK. Datta. Exploring synteny between rice and wild rice. Prof. R. L. Philips, regent professor, Department of Agronomy and Plant Genetics, University of Minnesota, USA. Research needs on seed pathology in developing countries: a future prospective. Dr. T.W. Mew. Bt rice for control of yellow stem borer: initial impressions. Dr. J.S. Bentur. Evaluation of a cryIAb -transformed Iranian rice variety against lepidopterous pests and crylAb resistance development in striped stem borer. Dr. F. Alinia. Prey-mediated and direct effects of Bt proteins on immature Chrysoperla carnea. Dr. A. Hilbeck, Swiss Federal Research Station for Agroecology and Agriculture. Engineering resistance against bacterial blight of rice. Dr. Wen Yuan Song, Plant Pathology Department, University of California Davis, USA.
158
Plant Breeding, Genetics, and Biochemistry Biotechnology for enhancement of gall midge resistance in rice: present status. Dr. S. Katiyar. Genetically modified organisms: issues and concerns. B.S. Ahloowalia, Agriculture and Food Development Authority, Dublin, Ireland. Cytogenetic and molecular analysis of Oryza sativa O. australiensis and O. sativa O. brachyantha derivatives. F.M. Abbasi. Biological significance of programmed cell death in rice. Prof. H. Uchimiya, Institute of Molecular and Cellular Biosciences, University of Tokyo, Japan. Positional cloning and phylogeny of rice blast resistance gene, Pi-b. Dr. S. Kawasaki, National Institute of Agrobiological Resources, Kannondai, Tsukuba, Ibaraki, Japan. Genetic engineering of wheat for improved phosphate bioavailability. Dr. P.B. Holm, Danish Institute of Agricultural Sciences, Denmark. Stability of performance in plant breeding. Dr. M.S. Kang, Louisiana State University, USA. Plastid transformation: an ecofriendly approach for crop improvement. Dr. S.R. Sikdar, Bose Institute, Calcutta, India. Nutritional and eating quality of rice. Dr. I. Tetens, Research Department of Human Nutrition, The Royal Veterinary and Agricultural University, Denmark. Studies of mineral bioavailability. Dr. M. Hansen, Research Department of Human Nutrition, The Royal Veterinary and Agricultural University, Denmark. Enhancing nutritional protein content in crops: results from sweet potato. Dr. C.S. Prakash, Tuskegee University, USA. Partial resistance to rice blast and current status of rice breeding for blast control in Japan. Dr. S. Koizumi, Ms. K. Zenbayashi, and Mr. N. Yokogami, Tohoku National Agricultural Experiment Station, Japan. Management, display, and analysis of expressed sequence tag (EST) data for the Triticeae in the graingenes database. Dr. G.R. Lazo, USDA/ ARS Western Regional Research Center, California, USA. Utilizing temperature japonica germplasm for tropical cultivation. Mr. J. Chavez.
Genetic diversity within and between maintainer and restorer lines used to develop tropical rice hybrids in the Philippines. Dr. Weijun Xu. Social Sciences Multi-temporal analysis of flooded areas in the Mekong River Delta using radar remote sensing. Ms. A. Alvaran. The impact of El Nino in rice production: a case of Bulacan. Ms. S. Valencia. Assessing the impact of agricultural research on poverty alleviation. Ms. C. Diaz and Dr. C. Edmonds. Diversification into fruit production on lowland rice farms in Thailand: a multiperiod linear programming analysis. Dr. S. Phuphak. Gender and economics: its relevance to rice research. Dr. S. Floro. The effects of overseas remittance income on lowland rice producers: a research plan. Dr. S. Graw. A socioeconomic study of land use and crop production systems in the uplands of northern Vietnam (household survey in Ha Giang Province, 1998). Dr. D.E. Pemsl. The conditions of farmer participation in irrigation management: a cross section analysis for the Phlippines. Dr. Y. Hayami and Dr. M. Fujita. Effect of improved technology on rice production and impact on income distribution and poverty alleviation: a case of Vietnam. Mr. Tran Thi Ut. Should developing countries invest more in less favored areas? An empirical analysis from rural India. Dr. P. Hazell, director, International Protection and Technology Division, International Food Policy and Research Institute, Washington, D.C., USA. Soil and Water Sciences Design and management of the RTDP databases (a sharing of lessons learned). Mr. G. Simbahan. Genetic diversity of arbuscular-vesicular fungi and their application in Japan. Dr. M. Saito, National Grassland Research Institute, Tochigi, Japan. Water and straw management long term study: insights, constraints, and potentials. Ms. Ma. J. Aduna.
159
yword index Keyword index
A agricultural diversification 99 Agricultural Research Center of Sarawak 107 agrohydrology 38 agromorphological traits 112 alleles 8, 9, 10, 49, 52, 110, 112 allelopathy 52, 53, 61 Aphelenchoides besseyi 115 apomixis 10 aquatic legume 108 ASEAN standard seed treatment 115, 116 asexual embryo 10 Asian Rice Biotechnology Network 78, 125 Atypena formosana 21, 22 Azolla 108 B BAC library 12, 78, 79, 80 bacterial blight resistance 83, 84, 85t bacterial leaf streak 115 Bastar Plateau 110 belief model 20 biofertilizer germplasm 108 bioinformatics 99, 130 biological control 21 biosystematic relationships 109 biotechnology 78, 103 biotic constraints 64 Bipolaris oryzae 115 blast 19, 20, 23, 48, 87t, 90, 112, 113, 114t blast management 86, 87 blast resistance 53, 54, 61 blue-green algae 108 Bolivia 112 Brazil 53, 61, 112, 113, 168 brown spot 19 Bt-endotoxin 81 C Cagayan Valley 112 Cambodia-IRRI-Australia Project 120, 123, 124, 162 Cambodian Agricultural Research and Development Institute 121
candidate genes 53, 54, 61. CARDI (see Cambodian Agricultural Research and Development Institute) Catanduanes 107 CIAP (see Cambodia-IRRI-Australia Project) CIAT (see International Center for Tropical Agriculture) CIMMYT (see International Maize and Wheat Improvement Center) Colombia 113 competitive ability 52 consortium 61, 64, 91 contig 80f crop establishment methods 22 crop intensification 122 crossability 108 crown sheath rot 19 cytoplasmic male sterility 5 D data management 23, 106, 108, 139, 140 data management system 106,108,113, 114, 115, 140 deepwater 32, 112 diversity in resistance 87 diversity of upland systems 48, 99 DLL (see dynamic link library) DMC1 12 DMS (see data management system) drought 26, 28, 44, 48 dynamic link library 113,114 E Echinochloa 52, 115 ecoregional approach 98 ecosystem-based nurseries 112 ELISA (see enzyme-linked immunosorbent assay) embryo 10, 11, 12 endosperm 11 enzyme-linked immunosorbent assay 15,16,21 erosion risk 99, 101t, 102f
172 IRRI program report for 1999
F false smut 19 FAO (see Food and Agriculture Organization) farmer-participatory varietal selection 60, 61 farming systems 34, 60, 121, 133 Fe density (see Iron density) field water balance 22 flood-prone rice 64, 76 flooding tolerance 64 Food and Agriculture Organization 108 food security 4, 54, 60, 61, 103, 120, 122 France 60 functional genomics 48, 60, 81, 89, 99, 103, 157 Fusarium moniliforme 115 G gender concerns 33, 93 gene flow 84, 88 genealogy 95, 96, 98, 114, 115 Genealogy Management System 113, 114 genepool 78 gene sequence 81, 82t, 83t genetic diversity 8, 9, 10, 86, 96, 109, 110, 112, 113, 114t, 116, 151, 153, 154, 158 geographic information system 23, 27, 29, 99 germplasm characterization 108, 116 germplasm collection 106 germplasm conservation 34,108 germplasm improvement 16,26,33, 44, 45, 48,64,78, 93, 94, 95, 99 germplasm multiplication 107 germplasm utilization 10, 14, 53, 61, 66, 96, 103 GIS (see geographic information system) glutinous varieties 86, 87, 107, 122 GMD (see grain mineral density) GMS (see Genealogy Management System) grain mineral density 66, 67,68, 69 grain quality 6, 16, 58, 90, 92, 96, 98, 123 grain yield 10, 13, 14t, 16, 36t, 40, 51, 58 green leafhopper 14, 15, 16, 22 GRIN 114 H head rice recovery 93, 98, 123
International Institute of Tropical Agriculture 112, 114 International Network for Genetic Evaluation of Rice 14, 196, 112, 113, 115, 116, 125, 140 International Rice Blast Nursery 105, 113 international Rice Genebank Collection 107 International Rice Genebank Collection Information System 108, 116 International Rice Information System 106, 113, 114, 115, 116, 140 International Maize and Wheat Improvement Center 18, 114 interplanting 86, 87 IPM (see integrated pest management) IR64 bacterial artificial chromosome 78 iron density 66 IRBN (see International Rice Blast Nursery) IRGC accessions 15, 109, 114 IRGCIS (see International Rice Genebank Collection Information System) Irian Jaya 106, 108 IRIS (see International Rice Information System) irrigated rice 74 53, 99, 115, 124 irrigated rice ecosystem 4, 21, 23, 24, 32, 45, 52, 112, 121, 122, 133 irrigation 39, 44, 51, 96 Ischaemum rugosum. 115 Italy 112 K Kachin 107 Kayah 107 Kayin 107 kinase domain 79 Korea 114, 115, 129t, 130t, 131t L land use planning 78 LaoRIS 108 Laos 93, 94, 95, 96, 97, 115 leaf scald 115 long-term experiments 16 low-cost technologies 124 M marker-aided selection 48, 52, 65, 78, 85, 139 market integration 61 meiosis 11, 12 micronutrients 43, 66 microsatellite analysis 112 mixed diet 22 multiscale approach 99 N N uptake 40 N2-fixing bacteria 45, 108 national rice research network 121 natural resource management 18, 78, 98, 103 National Seed Storage Laboratory 108
I ICIS (see International Crop Information System) IITA (see International Institute of Tropical Agriculture) improved rice varieties 38, 93, 94t INGER (see International Network for Genetic Evaluation of Rice) INGER Information System 113 INGER nurseries 105, 112, 115 INGERIS 113 integrated pest management 24, 120, 121, 132, 133, International Center for Tropical Agriculture 112, 114 International Crop Information System 113, 114, 127, 140
Keyword index 173
near-isogenic lines 14, 15, 49, 86 Nepal 18, 97, 107, 113t, 128t, 129t, 130t, 131t Nepal Agricultural Research Council 107 new plant type 4, 5, 13 nif gene 89 Nigeria 112 NSSL (see National Seed Storage Laboratory) nucellin gene 11 nucellus 11 O on-the-job training 7, 108 O. globerrima 107, 108 O. longiglumis 106, 109, 1104 111f O. longistaminata 54, 55, 56, 57, 107, 108, 109 O. meridionalis 106, 109 O. minuta 85 O. nivara 107 O. officinalis 106 O. ridleyi 109, 110t 111f O. rufipogon 17, 54, 55, 56, 57, 106, 107 O. sativa 54, 55, 56, 57, 107, 108 ORYZA model 38 P P deficiency 69 P uptake 45, 70 Pakistan 18, 94t, 95t, 97t, 113t, 114, 128t, 130t, 131t Palawan 9, 107 panicle blast 19 participatory varietal selection 60, 61 perenniality 54 perenniality genes 55 pest management decisions 20 physical map 77, 78, 79, 80 potential yield 14, 38, 39 poverty 4, 48, 60, 78, 93, 103, 120 problem soils 112 promoter 11, 12, 81 pyramiding genes 85 Q quantitative trait loci 48, 52, 71-74f, 78 Quirino 107 R rainfed lowland ecosystem 28, 35,. 58, 97, 124 rainfed lowland rice 26, 32, 35t, 38, 44, 45, 112, 121, 122, 124 Rainfed Lowland Shuttle Breeding Program 115 Rakhine 107 randomly amplified polymorphic DNA markers 79, 109, 111f RAPD markers (see randomly amplified polymorphic DNA markers) recombinant inbred lines/population 53,69, 71-74f, 79 REE5 12 relative P uptake 70 relative P use efficiency 70
relative shoot dry weight 69, 70, 75t, 76 relative tillering ability 69, 70, 75t, 76 resistance genes 15, 61 resistance gene analogs 87 Phizobium 89, 108 rice blast 86 root system 13, 14, 48, 49, 51 RTSV resistance 14, 15 S Sabah 107 Sarocladium oryzae 19, 115, 116 seed health 90, 116 Seed Health Unit 106, 115, 116 seedling blight 115 semidwarfism gene 50, 96 Senegal 112 Shan 107 Shan You 63 81, 83 sheath blight 19 sheath rot 19, 115, 116 simple sequence length polymorphism 70 SINGER (see Systemwide Information Network for Genetic Resources) Sitophilus granarius 115 socioeconomic diversity 58 soil fertility 41, 121, 123 soil organic matter 17, 41, 139 spatial distribution 27, 32, 34f, 99, 100f spatial surface analysis 27 spider 21, 22 stem borers 20, 78, 140 Stern rot 19 stem sap 13, 14 stress-oriented nurseries 112 sun drying 123 Surigao del Sur 107 Surinam 108, 112 survival 19, 22, 554 56, 57, 124 sustainability 18, 19, 23, 25, 40, 41, 43, 47, 48, 60, 61, 64, 99, 122 SysNet (see Systems Research Network for Ecoregional Land Use Planning in Tropical Asia) Systems Research Network for Ecoregional Land Use Planning in Tropical Asia 18, 78, 127t, 128t Systemwide Information Network for Genetic Resources 114 T technology adoption 29, 78 terminal sequencing 78 Thailand 90, 914 93, 94t, 95, 97t, 98, 99 Tilletia barclayana 115 trainees 128t, 129t, 130t, 131t training 120, 121, 122, 126, 138, 139 training courses 23, 129, 132, 140t training materials 90, 130, 133 training methodology 129 transgenic plants 78, 81, 84
174 IRRI program report for 1999
tungro 14, 15, 16, 17, 23, 87, 88, 112, 115, 140 tungro resistance 14, 15, 16, 17, 86, 88 tungro virus 16, 115 U upland ecosystem 32, 48, 53, 54, 58, 60, 61, 66, 99, 112, 121, 122, 138 upland rice 48, 53, 54, 55, 56, 58, 594 60, 61, 99, 107, 113, 121 upland weeds 60, 61 USA 52, 114 V vector resistance 1.6 vertical leaching 65 W Wagwag 112 (see West Africa Rice Development Association) watei-balance model 26, 27, 29 weed competitiveness 36t, 61 weed control 48 weed suppressive cultivars 52 weeds 19, 37, 43, 60 West Africa Rice Development Association 112 wheat 10, 16,17, 18, 19, 35, 86, 133 wild rices 108, 109 X xa5 gene 79, 80, 85, 86 Xa7 gene 86 xa13 gene 85,86 Xa2l gene 83, 84, 85, 86 Y yield decline 17, 39 yield stability 32, 38, 39, 56 yield trends 16, 17
WARDA
Keyword index 175

Program Report For 1999

Uploaded by

Copyright:

Available Formats

Program Report For 1999

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Program Report For 1999

Uploaded by

Copyright:

Available Formats

Research programs Irrigated rice ecosystem

Irrigated rice ecosystem

Breeding to break yield ceilings: a systems approach

IRRI program report for 1999

43 6.6 8.8 7.5 8.1 7

DRY SEASON NPT YIELDS

Irrigated rice ecosystem

RELEASE OF IRRI HYBRIDS

NEW CYTOPLASMIC MALE STERILE LINES

IRRI program report for 1999

DEVELOPMENT OF THERMOSENSITIVE GENIC MALE STERILE (TGMS) LINES

Irrigated rice ecosystem

IRRI program report for 1999

Irrigated rice ecosystem

IRRI program report for 1999

INDUCING THE ASEXUAL EMBRYO IN THE RICE NUCELLUS

Irrigated rice ecosystem

60 90 Days after transplanting

5. Changes in root length of NPT and IR72. IRRI, 1997 WS.

Irrigated rice ecosystem

Amount of stem sap (g m-2) 600

40 60 80 Days after transplanting

PL (cm) 20.4 20.1 20.1 22.4 20.0 20.1

PL (cm) 18.3 18.5 18.7 19.7 19.4 19.4

IR68333-R-R-B-22 IR68349-131-2-2-3 IR68352-14-1-1-1 IR68399-78-2-3-3-1 IR68373-R-R-B-22-2-2 Jinmibyeo check)

Hd = heading date. bPL = panicle length. cNP = no. of panicles.

IRRI program report for 1999

Pe21II Pc88II Pc41II 99 Pc20I 100 Pc34I PgVt6-lll, 1994 P71III 80 89 98

Irrigated rice ecosystem

Sustaining soil quality in intensive rice systems

IRRI program report for 1999

Irrigated rice ecosystem

Improving the productivity and sustainability of rice-wheat systems

IRRI program report for 1999

Fraction of infected tillers or panicles

Irrigated rice ecosystem

Factor 2 1.0 BS 0.5 0.0 -0.5 -1.0 -1.0 ShB GD ShR SR

0.0 0.5 Factor 1

0.0 0.5 Factor 1

0.0 0.5 Factor 2

Improving pest management

IRRI program report for 1999

Irrigated rice ecosystem

Progress of unreported projects

IRRI program report for 1999

Irrigated Rice Research Consortium

Irrigated rice ecosystem

IRRI program report for 1999

Research programs Rainfed lowland rice ecosystem

Rainfed lowland rice ecosystem

Characterizing and analyzing rainfed rice environments

IRRI program report for 1999

350 300 250 200 150 100 50 0 1 5 9 13 17 21 25 29 33 37 41 45 49

400 300 200

200 150 100 50 0 1 5 9 13 17 21 25 29 33 37 41 45 4 9

Stand ard meteorol ogical week

350 300 250

200 150 100 50 0 1 6 11 16 21 26 31 36 41 46 51

48 87 101 79 74 78 72 80 83 68 71 73 13 106 146