Sie sind auf Seite 1von 100

International Rice Research Notes

The International Rice Research Notes (IRRN) expedites communication among scientists concerned with the development of improved technology for rice and ricebased systems. The IRRN is a mechanism to help scientists keep each other informed of current rice research findings. The concise scientific notes are meant to encourage rice scientists to communicate with one another to obtain details on the research reported. The IRRN is published three times a year in April, August, and December by the International Rice Research Institute.

August-December 1996
Germplasm improvement
Genetic resources Distribution of rice land races in China 5 Some suggestions for in situ conservation of wild rices 6 Panicle culture and karyotype analysis from callus cells of a diploid wild rice, Oryza meyeriana 7 A study of Neolithic carbonized rice grains excavated from Hemudu, China 8 Classification of A genome species in the genus Oryza using nuclear DNA markers 8 Detection of D genome chromosomes bygenomic in situ hybridization 10 Fingerprinting rice germplasm using ALP and PCR-based RFLP 10 Distribution and evolutionary differentiation of mitochondrial plasmidlike DNAs in the genus Oryza 12 Genetic diversity of Chinese wild rice populations 13

The specter of food shortages is looming once again, with the annual rate of increase of rice production slowing to where it is lower than the rate of increase of rice consumers. Recent advances in cellular and molecular genetics of rice have come perhaps in the nick of time to provide us with new tools to develop rice varieties for the future. Only 10 years ago, the status of rice genetics was considered far behind that of other food crops, such as maize and wheat. The past decade, however, has seen an explosion of knowledge in this arena. Rice is now considered a model plant for such research on cereal crops. In October 1995, IRRI hosted the Third International Rice Genetics Symposium. More than 500 scientists from 31 countries attended. Along with a dramatic increase in the attendance over the years has come a major shift in the complexion of the program. During the first symposium in 1985, around 90% of the papers were on classical genetics; at this symposium, about 80% of the papers addressed topics on cellular and molecular genetics. The key papers presented have been published as an IRRI book. The posters displayed at the symposium appear as notes (in a modified format) throughout this double issue of IRRN, and will also be featured in the next issue. They are denoted by the symbol We hope you find these notes to be a valuable source of information.

Focus on rice genetics

IRRN production team

Genetics Molecular analysis of introgression in Oryza sativa/O. brachyantha and O. sativa/O. granulata derivatives 14 Acrotrisomics in rice 15 Molecular mapping of fertility-restoring gene Rf3 in rice 16 Characterization of common cis-regulatory elements responsible for endosperm-specific expression of rice glutelin gene 17 A fertility-restoring revertant, controlled by a single gene induced from a rice cytoplasmic male sterile line 18 Effects of Se1 gene on basic vegetative growth of rice 19 Genetic and cytochemical analysis of high 57-kD polypeptide mutants in rice 19 Molecular mapping of genes for F1 pollen sterility in rice 20 Chloroplast ribosomal protein genes encoded in rice nuclear genome 21 Stability of mitochondrial plasmidlike DNAs of rice 22 Sensitivity of plant height genes to gibberellic acid and their regulation by endogenous plant hormones in rice 22 Genetic diversity estimated among rice cultivars using restriction landmark genomic scanning method 24 Character expression of a rice dwarf mutant with lax panicle 25 Genetic differentiation between Japanese lowland and upland rices 26 Evolutionary variations in the Gramineae: rearrangements of DNA fragments transferred from chloroplast genomes to mitochondrial genomes 27 Identifying transposonlike element Tnr2 in rice 29 Breeding methods New cytoplasmic male sterile lines developed in Andhra Pradesh, India 30 Maintainers and restorers identified in some rice cultivars of Pakistan 31 Development of rice cytoplasmic male sterile line 47456 A in Kala Shah Kaku, Pakistan 31


Editors: Carolyn Dedolph and Theresa A. Castillo Assistant editor: Teresita Rola Layout and design: Erlie Putungan Artwork: Doris Rifareal

IRRN 21:2-3 (August-December 1996)

Comparison of promoters and selectable marker genes for indica rice transformation 32 Transfer of cytoplasmic male sterility in indica rice through protoplast fusion 33 Characters of plants regenerated from protoplasts of photoperiodsensitive genic male sterile rice 34 Attempted hybridization between Oryza sativa L. and Porteresia coarctata T. 35 DNA content in rice hybrids and heterosis 36 Induction of Agrobacterium tumefaciens vir genes by rice 36 Isolation of a tapetum-specific gene and promoter from rice 37 Production and characterization of Oryza sativa L./O. minuta Presl. hybrids and backcross progenies 38 Histological observation of callus morphology in rice 39 Genetic analysis of epicuticular structure involving a dripping-wet leaf mutant of rice 39 Embryogenic cell suspensions established from a high-protein purple black rice 40 Factors affecting pollen embryogenesis of rice anther culture 41 Genetic studies on purple pigmentation in rice and its use in breeding two-line rice hybrids 42 Grain quality Calorific values of 60 rice varieties 43 Cibodas, a high-yielding variety with good grain quality 43 Improving provitamin A (carotenoid) content of rice endosperm 43 Phytoene-forming activities in wild-type and transformed rice endosperm 44 Pest resistancediseases Screening of Basmati rice genotypes against blast 45 Resistance spectrum, race specificity, and expression of the Xa21 gene family 45 A bacterial blight-resistant, wide-compatible indica line 46 RFLP mapping of blast resistance gene Pi-k m in rice 47 Indica/japonica doubled haploid population as a model for mapping rice yellow mottle virus and blast resistance genes 47 Characterization of a pathogenesis-related gene family induced in rice infected with Magnapor the grisea 49 Genes controlling field resistance to blast in Japanese upland rice detected using RFLP markers 50 Genetic analysis for true resistance to blast in rice varieties 51 Pest resistance-insects Preliminary evidence of aphid resistance in transgenic plants 52 Role of host plant resistance in successful control of brown planthopper in Central Luzon, Philippines 53 Graphical genotypes of rice parental lines for resistance to green rice leafhopper 53 RFLP mapping of brown planthopper resistance gene Bph1 in rice 54 Stress tolerancesalinity Principal component analysis and variety classification in relation to rice seedling salinity tolerance 55 lntegrated germplasm improvementirrigated Variability, heritability, correlation, path analysis, and genetic divergence studies in M 2 generation of gamma-irradiated upland rice 56

RCPL-3-2 and RCPL-3-6: two promising rice lines for the mid-hills of Sikkim, India 59 Toag92: a short-duration rice cultivar for Turkey 59 KK15-36-C: a modern high-yielding rice variety for irrigated lowlands in Papua New Guinea 60 Birsa Dhan 201 and Birsa Dhan 202: early-maturing varieties for Bihar, India 60 lntegrated germplasm improvement-rainfed lowland Shakuntla: a rice variety for rainfed lowlands in Bihar, India 61 CMS lines for shallow rainfed lowland situation 62 lntegrated germplasm improvementupland Majhera 7, a direct seeded rainfed upland spring rice for Uttar Pradesh, India 62 Improved upland rice for the hillsides of Colombia 63 MTU9993, a promising rainfed upland rice for Andhra Pradesh, India 64

Crop and resource management

Physiology and plant nutrition Analysis of growth duration and heat units of different rice genotypes 65 The processes of pollination and fertilization in rice 65 Studies on intercellular space of embryo and seedling tissues in rice plants 66 Fertilizer management Effects of time of split application of K on soil N forms and lowland rice yield 67 Effect of Biozyme and NPK on rice yield 68 Effects of fertilizer and green manure on survival and productivity of detached deepwater rice plants 69 lntegrated N management with Gliricidia maculata and Sesbania aculeata for transplanted rice 70 Fertilizer managementinorganic sources Response of late-transplanted rice to NPK under irrigated conditions 70 Fertilizer managementorganic sources Evaluating stem-nodulating green manure crops for lowland rice 71 Managing crop residue in rice 71 Evaluation of green manures and grain legumes 72 Effect of soil amendments on rice yield 72 Crop management Use of DWR varietal mixtures to minimize hydrological risk in basin lands of Bihar, India 73 Effect of transplanting times on hybrid rice in Haryana, lndia 74 Effect of solar radiation and temperature on rice yields in different planting dates 75 Integrated pest managementdiseases Effects of neem derivatives on sheath rot in scented rice 76 Seedborne nature and seed transmission of rice sheath blight 76 Serological evidence for inducing resistance to rice tungro viruses using antiviral principles 77

IRRN 21:2-3 (August-December 1996)

Integrated pest managementinsects Hibernation sites of the rice stalk stink bug Tibraca limbativentris in the central region of Rio Grande do SUI, Brazil 78 Integrated pest managementweeds Predicting the effects of Echinochloa crus-galli on rice ( Oryza sativa L.) using the ecophysiological model INTERCOM 78 Weed control in direct seeded puddled rice 79 Integrated pest managementother pests Effect of soil solarization of rice nursery beds to suppress plant parasitic nematodes 80 Water management Supplemental irrigation for dry seeded upland rice 81 Rice yield under sprinkler irrigation 81 Farming systems Effect of different rice cultures on crops yield in rice-based systems 82 Farm machinery A low-cost, in-store dryer for small-scale farmers 83

Research methodology
A simple and rapid method for isolation of bacterial genomic DNA 84 A simulation model of rice sheath blight epidemics (I) Structure and model development 85 A simulation model of rice sheath blight epidemics (II) Model performance derived from sensitivity analysis 86 Use of CERES-Rice model to assess potential yield 87

Effect of contour hedgerow on runoff, soils loss, and upland rice production 87

Recommendations of the 3rd International Symposium on Hybrid Rice 88


Note to IRRN readers We sincerely apologize for the delay in producing this special combined issue. Regular production of the IRRN will resume in 1997. Thank you for your continued support.

IRRN 21:2-3 (August-December 1996)

Germplasm improvement
Genetic resources
Distribution of rice land races in China
Cao Yongsheng and Zhang Xianzhen, Institute of Crop Germplasm Resources, Chinese Academy of Agricultural Sciences, Beijing 100081, China

Cultivated rice (Oryza sativa L.) is widely distributed in China (see figure and table). These germplasm resources are highly diverse and extremely localized because of the country's vast territory, the complicated terrain, the continental and oceanic monsoon climates, and various other ecological factors. These factors make the

distribution of rice land races in China regional and noncontinuous. The summer monsoon brings warm temperatures and abundant precipitation to southeastern China, making for a long ricegrowing season and a concentration of ricefields. The landscape is very hilly, which causes significant differences in the climate. The rice germplasm resources in the region are rich. The majority (93%) of the rice land races in China are distributed in the area south of the Qing Mountains and Huaihe River, where the annual precipitation is more than 800 mm and the annual mean temperature is above 15C. The Changjiang Delta, the Zhujiang Delta, the central Anhui Plain, and the Chendu Plain are the main regions of

rice land race distribution; the plains in Yunnan and Guizhou provinces and the coastal plains in Zhejiang, Fujian, and Hainan provinces are also important areas. The rice-growing season is short in the area to the north of the Huaihe River in eastern China because of the high latitude and low mean temperature. Rice land races are less common (only 6% of the total accessions) than in southern China because most of the area's annual precipitation (400800 mm) occurs in the summer and is significantly less than that in southern China. Additionally, rivers and lakes in the area are few. In the southern part of the Huanghuai Plain, the abundant rainfall at the end of spring and the beginning of summer


IRRN 21:2-3 (August-December 1996)

Rice land races from several provinces in China. Province Anhui Beijing Fujian Gansu Guangdong Guangxi Guizhou Hainan Hebei Heilongjiang Henan Hubei Hunan Jiangsu Jiangxi Jilin Liaoning Neimenggu Ningxia Shaanxi Shandong Shanghai Shanxi Sichuan Taiwan TianJin Xinjiang Xizang Yunnan Zhejiang Total No. 725 9 1894 7 4863 8967 4279 668 325 113 365 1417 5001 2320 2869 68 78 10 18 642 128 304 169 3255 1190 27 16 38 5532 1585 46,882

Some suggestions for in situ conservation of wild rices

S. N. Singh and M. Gadgil, Centre for Ecological Sciences (CES) , Indian Institute of Science, Bangalore 560012, India

India has seven species of wild rices. Four of these, Oryza nivara, O. rufipogon, O. sativa f. spontaea, and Posteresia coarctata, occur in the coastal and hilly districts of Uttara Kannada (13 55' N to 15 32' N latitude and 74 5' E to 75 5' E longitude) and Shimoga in Karnataka. We examined 125 populations of wild rices to determine their habitat require-

ments and how ongoing ecological changes are affecting them. From 22 Oct 1992 to 6 Oct 1993, we collected seed and plant specimens from 43 locations and deposited them in a herbarium at the CES (see table). Karnataka had not previously been explored for wild rices, so the intraspecific variation that remains can still be captured. Moreover, large patches of wild rices still exist that can be easily conserved in situ. These wild rices are locally known as Nyarai, Nyarai batta, Kadu batta, Kadu hullu, Uddinakardi battu, Uddirige Urbu, Chungu Nyari, and Navane. A few plants of P. coarctata that were seen on 22 Oct 1992 at Kumta-Vannalli

Collection date, locality, and other information for some wild rice samples. Karnataka, India. Collection date 22-10-92 28-10-92 06-10-93 23-10-92 23-10-92 06-10-93 28-10-92 23-10-92 30-10-92 23-10-92 06-10-93 05-11-92 06-11-92 05-11-92 04-11-92 22-10-92 03-11-92 03-11-92 03-11-92 03-11-92 05-11-92 04-11-92 04-11-92 04-11-92 06-11-92 05-11-92 04-11-92 04-11-92 03-11-92 05-11-92 04-11-92 05-11-92 05-11-92 05-11-92 03-11-92 05-11-92 03-11-92 03-11-92 03-11-92 03-11-92 05-11-92 03-11-92 Locality Kumta-Vannalli Dhareshwar Dhareshwar Handigon Handigon Handigon Handigon Handigon Manki Manki Manki Gudwi Gudnapur Yelasi Kondli Sashitlu-Kumta Mattigar Menasi Menasi Mattigar Tyavgod Balekoppa Avragoppa Aigod Kantraji Konan mane Andawalli Andawalli Akkunji Kallambi Andawalli Sirlige Sirlige Hale Sorab Kawachur Hale Sorab Nagarbawi Hosur Hosur Hasvante Gundasettikoppa Talguppa Taluka Latitude Longitude Altitude (m) 3.5 3 3 4 4 4 4 4 5 5 5 550 570 580 590 3.25 580 580 580 580 585 590 585 570 555 560 570 570 570 570 570 575 575 580 580 580 590 595 595 595 595 615 Habitat typeb E S S F F F F F C F F P T T T F F F F F T T T T T T T T T T T T T T T T T T T T T T Relative Speciesd abundancec E F R A A E F A R A O A A A A F A A A A A F A A A A A A A A A A A A A A A R A A A A P s/n s/n s/n s/n s/n s/n s/n s/n s/n s/n n/r n/r n/r n/r S S S S S r r r r r r r r r r r r r r r r r r r r r r

negatively affects rice growth. The ricegrowing area and the number of land races there are much less than those in areas to the south of the Chanjiang River. In the Huanghuai Plain, the land races are concentrated in the lower valley of the Huaihe River and the coastal regions where water resources are abundant. The area near the Bohai Sea in Hebei Province, the lower valley of the Liaohe River in northeastern China, and the Songhuajiang River Valley are the main areas for rice land races in regions at high latitudes. The climate in northwestern China is continental and dry with a short wet season. Rice land races are mainlybut sparsely distributed over river valleys. They constitute only about 1% of all the land races in China.

K K K K K K K K K K K So So So Si K Si Si Si Si So Si Si Si So So So So Si So So Si Si So Si So Si Si Si Si So Sa

1425.25 1425.5' 1425.5' 1423.5' 423.5' 1423.5' 1423.5' 1423.5' 1426.75' 1426.75' 1426.75' 1427' 1433' 1422' 1421' 1425.5' 1418' 1417' 1417' 1418' 1425' 1421' 1422' 1424' 1422' 1422' 1422' 1422' 1418' 1426' 1422' 1421' 1421' 1423' 1416' 1423' 1419' 1420' 1420' 1414' 1422' 1413'

7424' 7424' 7424' 7424.5' 7424.5' 7424.5' 7424.5' 7424.5' 7426' 7426' 7426' 7501' 7459' 7503' 7454.5' 7424' 7452.5' 7449' 7449' 7452.5' 7503' 7454' 7452' 7456' 7459' 7501' 7457' 7457' 7453.5' 7502' 7457' 7456.25' 7456.25' 7505' 7454' 7505' 7451.8' 7453' 7453' 7454' 7504' 7454.25'

a Taluk is an administrative unit below the district level; K = Kumta, Si = Siddapur, So = Sorab, Sa = Sagar. b E = estuarine rice cultivation, F = farmer's field, P = pond, S = seasonal stream, T = tank. c A = abundant, F = frequent, O = occasional, R = rare, E = locally extinct. d n = Oryza nivara, r = O. rufipogon, s = O. sativa f. spontanea. p = Porteresia coarctata.

IRRN 21:2-3 (August-December 1996)

have already become locally extinct when a tidal wetland area was converted into prawn culture. Other wild rice populations are threatened by the encroachment of small irrigation tanks, freshwater aquaculture, and a change in the cultivation system from broadcasting to transplanting. In situ conservation of these important genetic resources must be initiated and complement ex situ conservation efforts. These measures could include Integration into conservation programs of tidal and freshwater swamps. For example, the Karnataka State Forest Department has a program to conserve mangrove

swamps at Kundapur in Dakshina Kannada and a freshwater wetland at Gudwi Pakshidham in Shimoga. Other wetlands, such as those in Bharatpur in the state of Rajasthan. are being protected under the Ramsar Convention. Special attention should be paid to in situ conservation. including deliberately introducing into the wetlands appropriate indigenous wild rice species. Integration into ecotourism-based conservation programs. One of the study sites, Gudwi. is a bird sanctuary close to the tourist attraction of the Gerusoppa Waterfalls. Conservation culture media alternately. The N6 medium contained 2 mg 2.4-D L -1 and 4.5% sucrose while the MS medium was supplemented with 2 mg kinetin L -1, 0.5 mg NAA L -1, and 3.0% sucrose. Green plant regeneration frequency of the selected calli was up to 76%, but that of the unselected calli was only 33% after the 10th culture. The free amino acids in the calli were determined. The methionine glutamic acid and the total amino acid contents in the selected calli were lower than those in the control, while the alanine and phenylalanine contents in the selected calli were higher than those in the control. The main components of the free amino acid pool in the selected calli were alinine. glutamic acid, and phenylalanine, which together constituted 65.8% of the total amino acids. It appears that alanine. glutamic acid, methionine, and phenylalanine occupy a key position in amino acid metabolism and influence the regeneration capacity of calli.

of wild rices, along with environmental awareness programs, could be effectively integrated with the tourist attraction of aquatic birds. Use of indigenous knowledge. Some local indigenous communities, such as the Mushahars of eastern Uttar Pradesh and Bihar, depend on wild rice grain and other wild foods for survival. These people have an intimate knowledge of the distribution and habitat requirements of wild rices. It would be highly useful to involve these communities in in situ conservation programs.

Panicle culture and karyotype analysis from callus cells of a diploid wild rice, Oryza meyeriana
Xiao-Ling Wang, Li-Hui Shu, Wen-Jing Yuan, and Lan-Jie Liao, School of Life Science, Wuhan University, Wuhan, Hubei 430072, China

The genus Oryza, to which the cultivated rice (O. sativa 2n=24) belongs, has 20 wild species with 2n=24 or 48 chromosomes. Oryza meyeriana (2n=24) is highly resistant to bacterial blight. The karyotype of this species has not been well documented and only a few tissue culture studies have been made. We report the results of in vitro culture of young panicles and karyotype analysis of this species. Embryogenic calli with high plant regeneration ability. We obtained embryogenic calli with high plant regeneration ability using N6 and MS

High IAA oxydase activity was helpful in regenerating the calli. The regenerated plants were tested for reaction to three bacterial blight strains: PXO61, T7174, and Jiang Ling 691. Fifty regenerated plants were inoculated with the three strains on different tillers of the same plant. The regenerated plants were resistant to the strains, but the degree of resistance differed among the plants. Karyotype analysis. O. meyeriana has strong seed shattering and low seed fertility, making karyotype analysis using seed difficult. So instead, we used the method of Kurata et a1 to conduct karyotype analysis using calli. The O. meyeriana had 2n=24 (see figure). Based on the arm ratio. the karyotype of O. meyeriana consisted of 4 pairs of metacentrics. 7 pairs of submetacentrics, and a pair of satellitesubmetacentric chromosomes (see table).

a) The somatic metaphase showing 24 chromosomes in O. meyeriana; b) the karyotype of O. meyeriana (x7000).

IRRN 21:2-3 (August-December 1996)

Relative length, arm ratio, and classification of the chromosomes of Oryza meyeriana. Chromosome number 1 2 3 4 5 6 7 8 9 10 11 12 Relative length (%) 11.80 11.48 9.87 8.80 8.69 8.31 8.11 7.26 6.97 6.79 6.17 5.74 b Arm ratio (short/long) 0.82 0.90 0.67 0.70 0.73 0.69 0.73 0.63 0.93 0.88 0.69 0.37 Classification a M M SM SM SM SM SM SM M M SM SAT

IRRN reminder Multiple submissions. Normally, only one report for a single experiment will be accepted. Two or more items about the same work submitted at the same time will be returned for merging. Submitting at different times multiple notes from the same experiment is highly inappropriate. Detection will result in the rejection of all submissions on that research.

a M = metacentric, SM = submetacentric, SAT = satellite. Arm ratio: M = (1.0-0.76), SM = (0.75-0.25). b The length

of the satellite is not inluded in the length of the chromosome.

A study of Neolithic carbonized rice grains excavated from Hemudu, China

Shengxiang Tang and Hanyong Yu, China National Rice Research Institute, Hangzhou 310006, China

Rice grains excavated from Hemudu, China (6950130 BC) are known to be some of the worlds oldest remains of rice cultivation. We studied the variation in

these grains to improve our understanding of the origin and domestication of cultivated rice in China. The length and width of 105 of the grains were measured using an enlarged photograph. About half of the grains were awnless. The grain length ranged from 5.2 to 8.6 mm, with an average of 7.1 mm, and the grain width ranged from 2.1 to 3.2 mm, with an average of 2.8 mm. The grain length-width ratio (L:W) was 1.7-3.2. with an average of 2.6. These findings suggest

that the Hemudu rice population had large variation in grain shape and awn characteristics. Our earlier study found a few wild rice grains (O. rufipogon) among the rice grains excavated at Hemudu. We therefore infer that about 7,000 yr ago, people in Hemudu did primitive rice cultivation, and that the middle and lower basin of Yangtze River is a homeland of cultivated rice in China.

Length, width, and shape distribution of 105 rice grains excavated from Hemudu, China. Grain trait Length (mm) (no.) (%) Width (mm) (no.) (%) L-W ratio (no.) (%) <5.4 1 0.9 2.1 2 1.9 1.7 1 0.9 5.6 2 1.9 2.2 3 2.9 1.8 0 0 5.9 3 2.9 2.3 3 2.9 1.9 1 0.9 6.2 6 5.7 2.4 1 0.9 2.0 2 1.9 6.5 7 6.7 2.5 5 4.8 2.1 5 4.8 6.8 21 20.0 2.6 9 8.6 2.2 7 6.6 7.9 29 27.6 2.7 20 19.0 2.3 5 4.8 Distribution 7.4 11 10.5 2.8 24 22.9 2.4 12 11.4 7.7 9 8.6 2.9 18 17.1 2.5 12 11.4 8.0 11 10.5 3.0 15 14.3 2.6 16 15.2 8.3 4 3.8 3.1 4 3.8 2.7 13 12.4 8.6 1 0.9 3.2 1 0.9 2.8 5 4.8 2.9 16 15.2 3.0 4 3.8 >3.1 6 5.7

Classification of A genome species in the genus Oryza using nuclear DNA markers
K. Doi, A. Yoshimura, M. Nakano, and N. Iwata, Plant Breeding Laboratory, Faculty of Agriculture, Kyushu University, Fukuoka 812-81, Japan; and D. A. Vaughan, National Institute of Agrobiological Resources, Tsukuba 305, Japan

genus Oryza, which contains O. rufipogon, O. nivara, O. glaberrima, O. barthii, O. longistaminata, O. glumaepatula, and O. meridionalis, and of the materials previously analyzed by Nakano et al (1992) (see figure). A total of 192 accessions of the A genome species were analyzed. RFLPs were detected for combinations of Dra I-digested total DNA and 21 single copy genomic clones. Genetic distances between accessions were quantified as D = -1n [2Mxy/(Mx+My)]

We did a new analysis of the restriction fragment lcngth polymorphisms (RFLPs) of 67 accessions of A genome species in the

where Mx and My were the total fragments in accessions X and Y, respectively, and Mxy were the common fragments observed between accessions X and Y. A dendrognm was then constructed by the UPGMA method using 64 of the 67 new accessions and 12 previously analyzed accessions as a reference (see figure). A genome species were classified into five major groups: Asian (O. sativa, O. rufipogon, and O. nivara), O. glumaepatula, O. glaberrimaO. barthii, O. longistaminata, and O. meridionalis. O. glumaepatula had

IRRN 21:2-3 (August-December 1996)

RFLP-derived dendrogram of 76 accessions. Sixty-four out of 67 accessions provided by IRRI, and 12 previously analyzed accessions of O. sativa, O. rufipogon, and O. glaberrima (Nakano et al 1992), were included for reference; three O. rufipogon accessions from Papua New Guinea were excluded.

IRRN 21:2-3 (August-December 1996)

closer affinity to the O. glaberrimaO. barthii group than to the Asian O. rufipogon group, although it had been considered a subtype of O. rufipogon. The Asian group comprised four or five loosely knit groups, each corresponding to the indica type (including some O. rufipogon ), japonica type (including O. rufipogon from China), O. rufipogon (consisting of several subgroups), and O. nivara (see figure). Some O. rufipogonO. nivara showed close affinity to cultivated rice, as suggested by previous studies (Wang et al 1992, Nakano et al 1992). Three accessions of perennial O. rufipogon from Papua New Guinea (excluded from the figure) carried both fragments of O. rufipogon and O. meridionalis. Although no O. meridionalis has been reported from Papua New Guinea so far, it is possible that some exist in the country and that the natural hybrids between O. rufipogon and O. meridionalis could have survived until now. The classification among species matched well with the findings of previous studies (Morishima 1969, Second 1985, Wang et al 1992). Further analysis is needed to reveal the relationships among Asian subgroups.

Detection of D genome chromosomes by genomic in situ hybridization

R. Shishido and T. Kinoshita, Hokkaido University, Sapporo 060, Japan; N. Ohmido and K. Fukui, Hokuriku National Agricultural Experiment Station, Joetsu 943-01, Japan

Cited references

Morishima H. 1969. Phenetic similarity and phylogenetic relationships among strains or Oryza perennis, estimated by methods of numerical taxonomy. Evolution 23:429-443. Nakano M, YoshimuraY, Iwata N. 1992. Phylogenetic study of cultivated rice and its wild relatives by RFLP. Rice Genet. Newsl. 9:132-134. Second G. 1985. Evolutionary relationships in the sativa group of Oryza based on isozyme data. Genet. Sel. Evol. 17:89-114. Wang ZY, Second G, Tanksley SD. 1992. Polymorphism and phylogenetic relationships among species in the genus Oryza as determined by nuclear RFLPs. Theor. Appl. Genet. 83:565-581.

Based on chromosome pairing at meiosis, it is known that the genus Oryza has five genomes (A, B, C, E, and F) in the diploid species. Tetraploid species with BC or CD genomes have also been found, although a diploid species with the D genome has not yet been discovered. The D genome is only found in tetraploids (O. alta, O. grandiglumis, and O. latifolia from Central and South America) in combination with the C genome. In this study, we visualized D genome chromosomes on a glass slide in a CD genome tetraploid species. We used a chromosome painting method in which we applied the genomic in situ hybridization (GISH) method, using the total genomic DNA isolated from a C genome diploid species, O. officinalis (2n=24), labeled with biotin as the probe. O. latifolia was used to prepare chromosomes. Chromosome samples were prepared by an enzymatic maceratiod/air drying method. The chromosome samples were treated with an enzymatic mixture (2% cellulase onozuka RS, 1.5% macerozyme R-200,0.3% Pectolyase Y-23), proteinase K (1 mg ml -1), 4.5% acetic acid, and RNase A (1 mg ml -1) prior to in situ hybridization to get clear genome-specific fluorescent signals on small chromosomes.

The post-treatment removed cytoplasmic debris, cellular protein, and RNAs that cause nonspecific signals. In addition, a modified thermal cycler with a flat aluminum plate was developed and used to precisely control temperature. The results of GISH were analyzed using CHIAS 2, an image-analyzing system that provides clear fluorescent signal reproduction. This system can capture the fluorescent signals of GISH within a second, enhance weak signals, and even superimpose the signal images onto the original chromosome We identified 24 C genome chromosomes, detected by the fluorescent probe of O. officinalis, from a mixture of 48 O. latifolia chromosomes belonging to the C and D genomes. All 24 D genome chromosomes were clearly identified using the imaging method, as were 24 C genome chromosomes from the 48 chromosomes with B and C genomes. We also found differences in the relative strength of fluorescent signals between C and D genomes, and between B and C genomes. The differences in the fluorescent signal strength between B and C genome chromosomes were more distinct than those between C and D genome chromosomes, despite applying the same probe. These differences reflect the sequence homology between the C and D or B genomes. The D genome is considered to be more similar to the C genome than to the B genome. We will continue to use the GISH method to provide new information about the relationships among the six genomes in Oryza.

Fingerprinting rice germplasm using ALP and PCR-based RFLP

B. Ghareyazie, Faculty of Agriculture, University of Guilan, Rasht, The Islamic Republic of Iran; N. Huang, G. Second, J. Bennett, and G. S. Khush, IRRI

Morphological and isozyme markers provide a simple method for classifying rice collections. However, the markers available and their level of polymorphism are inconveniently low. DNA markers, on the

other hand, are abundant, high in their level of polymorphism, and known to be useful for germplasm surveys. We evaluated the usefulness of converting mapped restriction fragment length polymorphism (RFLP) markers to polymerase chain reaction (PCR)-based markers, such as amplicon length polymorphism (ALP), and PCR-based RFLP (Ghareyazie et al 199.5). This is the first report on applying ALP and or PCR to classify rice and on classifying Iranian rice varieties at the DNA level.


IRRN 21:2-3 (August-December 1996)

We used a set of 35 Iranian varieties, along with three indica and two japonica varieties as references (Fig. 1). A set of 15 pairs of primers were used to amplify genomic DNA isolated from the 40 varieties. To examine ALPS, PCR products were separated in agarose gels containing ethidium bromide. The products were then digested with nine different restriction endonucleases (recognizing 4-5 bases) and separated on either agarose or polyacrylamide gels. Amplified PCR products and patterns generated upon their digestion were scored for the presence or absence of bands in each variety. These data were used to construct a resemblance matrix according to Dice distance coefficients. We used the NTSYS program to perform cluster analysis and to build the dendrogram according to the UPGMA clustering method. Agarose and or polyacrylamide gel analysis of the digested PCR products showed a variety of banding patterns.

Diversity at loci RG118, RG13, and RG235 was exhibited prior to and/or after digestion of PCR products (Fig. 2). Six of the 15 pairs ofprimers used showed ALPS (Fig.2c, e). The others generated monomorphic PCR products across 40 varieties (Fig. 2a). These monomorphic banding patterns were converted to polymorphisms by digesting PCR products with Mval (Fig, 2b). Using the formula
NP = S n1 n (Xi S X j ) i = 1 j=i+1

where n is the total number of alleles in a given locus, and Xi and Xj are the number of varieties carrying the ith and jth alleles. respectively, we calculated the number of polymorphic pairs (NP) as a measure of the abundance of ALP (13%) and PCR (28%) in rice. A dendrogram was built from the genetic distances matrix (Fig. 1). A clear distinction exists between the two major rice subspecies, indica and japonica. Three groups were distinguished among Iranian

varieties, (indicas and japonicas, and varieties that are genetically distinct from both indica and japonica types). The varieties are traditionally divided into the major groups of Sadri, Champa, and Gerdeh. One major group included all of the Sadri-type varieties from northern Iran (Guilan and Mazandaran provinces), where rice is the major crop. This group was clustered with two japonica standards. The second group included only six varaties (entries 5, 9, 14, 15, 21, and 16, Fig. 1). It was not clustered with any of the standard varieties, indicating the varieties probably evolved independently within the country. The third group. the smallest one, included only three Champa varieties from Iranian germplasm and was clustered with the three indica standard varieties. Sadri-type varieties, which are strongly scented and well known for their cooking quality, are genetically closer to japonicas

1. Dendrogram based on Dice's genetic distances demonstrating relationships among 40 rice varieties. Genetic distance was calculated according to Dice's method based on the pooled ALP and P-BR data of 13 loci. The NTSYS program was used to build the dendrogram according to the UPGMA clustering method.

IRRN 21:2-3 (August-December 1996)


2. Ethidium bromide staining of PCR products for RG118 (A), RG13 (C) and RG235 (E) loci amplified from total genomic DNA of Iranian varieties and restriction fragments generated from the products digested with Mval (B) and Hin Fl (D). The numbers for each lane are the entry numbers used in Figure 1. The approximate size of bands is shown in the base pair.

than to indicas. Despite their very typical indica morphology (long and cylindrical seed, tall stature, narrow leaves), these varieties, as well as some of the Champatype varieties, are clustered with standard japonicas. This suggests that the morphological characteristics are not representative of the genetic status of a given variety, and as revealed by Glaszmann (1987), these varieties may belong to isozyme group V

and are therefore neither pure indica nor pure japonica. We found PCR-based RFLP to be faster, easier, and relatively cheaper than Southern-based RFLP. PCR markers are now abundant, with known map locations. Their application is highly reproducible and more informative than randomly amplified polymorphic DNA.

Cited references
Ghareyazie B, Huang N, Second G, Bennett J, Khush GS. 1995. Classification of rice germplasm. I. Analysis using ALP and PCRbased RFLP. Theor. Appl. Genet. 91:218227. Glaszmann JC. 1987. Isozymes and Asian rice varieties. Theor. Appl. Genet. 74:21-30.

Distribution and evolutionary differentiation of mitochondrial plasmidlike DNAs in the genus Oryza
S. Miyata, A. Kanazawa, N. Tsutsumi, Faculty of Agriculture, The University of Tokyo, Tokyo 113, Japan; Y. Sano, Faculty of Agriculture, Hokkaido University, Sapporo 060, Japan; and A. Hirai, Faculty of Agriculture, The University of Tokyo

We investigated the distribution of plasmidlike DNAs homologous to those of O. sativa in the genus Oryza and found some evolutionary differentiations of plasmidlike DNAs. Four kinds of circular mitochondrial plasmid-like DNAs, designated B1, B2, B3, and B4, have been detected in many Oryza sativa strains and analyzed in DNA sequences.

We first analyzed the distribution of plasmidlike DNAs homologous to those of O. sativa in 40 strains of the genus Oryza with AA, BB, BBCC, CC, CCDD, and EE genomes (Miyata et al 1995). Using the B1, B2, B3, and B4 clones as probes, plasmidlike DNAs were observed only in varieties with AA, CC, and CCDD genomes. The distribution patterns of strains with the AA genome were highly polymorphic. This suggests that the disappearance of plasmidlike DNAs occurred during the diversification of strains with the AA genome. We then amplified the plasmidlike DNAs from strains with the AA genome by polymerase chain reaction (PCR) and examined restriction fragment length polymorphisms (RFLPs). RFLPs were detected among plasmidlike DNAs amplified from different strains (Miyata et al 1995), indicating some mutations, such as base substitutions.

In CC and CCDD genomes, some signals hybridized with the B1 probe showed rather different restriction patterns, indicating that these signals are new plasmidlike DNAs homologous to B1. We cloned them and named them M1, M2, and M3. Their entire nucleotide sequences were determined; M1 is 2,430 bp in size, M2 is 2,034 bp, and M3 is 2,096 bp. We compared each of these sequences and found that M1, M2, and M3 share high homology to B1. B1, M2, and M3 each lack about 300 bp of a region that is present in M1, and small repeats were found at the sites of deleted sequences (see figure). We therefore propose the hypothesis that the B1 family differentiated from a common ancient molecule that was similar to M1 through slipped mispairing during DNA replication at several stages in the evolution in the genus Oryza.


IRRN 21:2-3 (August-December 1996)

The schematic representation of the explanation for the generation of members of the B1 family diversification of the genus Oryza a . of mitochondrial plasmidlike DNAs during the
aShaded portions indicate M1 and the regions

various phenotypic characters, isozymes, and nuclear genome restriction fragment length polymorphism (RFLP). Four populations collected from Thailand, India, and Indonesia were used as the controls. A total of 292 plants were examined. O. rufipogon is differentiated into two types: perennial and annual (often referred to as O. nivara). We have found Chinese wild rices that are perennial, but they tend to differ from the typical perennial type from tropical countries in terms of root and shoot regeneration pattern from basal nodes, dry weight proportion of newly emerged tillers after heading, and panicle number per plant. The result of factor analysis based on 11 characters indicated that the populations examined, though they contained intrapopulational variability, showed interpopulational differentiation into annual and perennial types. Chinese populations were

intermediate between the annual and perennial types. Polymorphism was found in 29 isozyme loci and 15 probe/restriction enzyme combinations. Differences among populations seemed to be clearer in RFLP than in isozyme variation. Chinese populations share some common markers that are rarely found in other populations, indicating that they differ from one another (Table 1). The result of factor analysis of isozyme data revealed geographic variation, but not ecological variation (perennial vs annual differentiation). Interestingly, populations of Chinese rice O. rufipogon contain some japonicaspecific alleles, such as Est-13(t)-1 and Cat1-2, at higher frequencies than in other populations (Table 1). At other loci, however, they carry some indica-specific alleles, as in other populations.

Table 1. Allelic frequencies at six isozyme loci and four RFLP markers. Population China DX Pa YJ P GG P Thailand NE3 A NE88 P Indonesia W1981 P India W120 P

homologous to M1. Large arrows indicate the differentiation of the B1 family. The dashed lines of A, B, and C indicate the deleted regions of M1 and the corresponding sites in B1, M2, and M3, respectively. White and black arrowheads indicate the small repeats of 3 bp and 15 bp at the deleted sites of A and B, respectlvely.


Cited reference
Miyata S, Kanazawa A, Tsutsumi N, Sano Y, Hirai A. 1995. Polymorphic distribution and molecular diversification of mitochondrial plasmid-like DNAs in the genus Oryza, Jpn. J. Genet. (in press.)

Isozyme Sdh1-2 Amp1-3 Acp7 (t)-1 Est9-1 Est13 (t)-1 Cat1-2 RFLP G282-22 G187-12 G20-20 G249-12

0.90 1.00 1.00 0.92 0.98 0.99 0.65 1.00 1.00 0.65

0.89 1.00 1.00 0.47 1.00 1.00 1.00 1.00 0.65 1.00

0.92 0.96 0.81 0.29 0.59 0.04 0.10 0.96 0.63 0.78

0.00 0.00 0.19 0.00 0.00 0.00 0.00 0.65 0.00 0.00

0.00 0.04 0.33 0.01 0.00 0.00 0.07 0.01 0.00 0.01

0.00 0.50 0.56 0.00 0.00 0.00 0.00

0.68 0.43 0.21 0.07 0.00 0.25 0.64 0.00

Genetic diversity of Chinese wild rice populations

P = perennial, A = annual.

Cai Hong-wei, National Institute of Genetics, Mishima, 411 Japan; Wang Xiang-kun, Beijing Agricultural University, Beijing, 100094, China; and H. Morishima, National Institute of Genetics, Mishima, 411 Japan

Table 2. Parameter values showing intrapopulational genetic diversity estimated from data of 29 isozyme loci and 15 RFLP markers. Parameter Isozyme (no. of plants examined) Av gene diversity (H) Proportion of polymorphic loci Alleles/locus (av no.) Frequency of heterozygote RFLP (no. of plants examined) Av gene diversity (H) Proportion of polymorphic loci Alleles/locus (av no.) Frequency of heterozygote Population DX 54 0.07 0.55 1.55 0.13 46 0.13 0.31 1.38 0.54 YJ 14 0.04 0.10 1.10 0.14 14 0.19 0.46 1.46 0.00 GG 68 0.20 0.83 2.10 0.53 63 0.27 0.73 1.80 0.30 NE3 74 0.14 0.41 1.45 0.07 74 0.20 0.53 1.53 0.03 NE88 46 0.21 0.69 1.97 0.46 41 0.12 0.60 1.67 0.27 W1981 22 0.22 0.55 1.72 0.46 W120

The genetic diversity of Chinese rice Oryza rufipogon and its relationships with the strains distributed in tropical Asia have not been fully investigated. We examined three samples from populations collected in southern China from Dongxiang, Jiangxi (DX, the northernmost site of this species), Yuangjing, Yunnan (YJ), and Guigang, Guangxi (GG) for

14 0.25 0.59 1.75 0.71 -

IRRN 21:2-3 (August-December 1996)


Several parameters for intrapopulational genetic diversity were estimated for each population using isozyme and RFLP data (Table 2). These parameters showed no correlation between isozyme and RFLP level. The DX and YJ populations, which are completely isolated from cultivated ricefields, showed less polymorphism than the perennial populations from other countries at the isozyme level, but this trend was not found at the RFLP level. The high gene diversity at both the isozyme and DNA levels in the GG population adjacent to cultivated ricefields is probably due to the flow of genes from these neighboring cultivated rices.

Molecular analysis of introgression in Oryza sativa/ O. brachyantha and O. sativa/O. granulata derivatives
R. K. Aggarwal, D. S. Brar, N. Huang, and G. S. Khush, IRRI

The genus Oryza, to which cultivated rice belongs, has 20 wild species that provide an important reservoir of useful genes for rice improvement. O. brachyantha (2n=24, FF) and O. granulata (2n=24), two distantly related species, possess genes for resistance to yellow stem borer, bacterial blight, and brown planthopper. Several backcross progenies of these two species with elite breeding lines of O. sativa have been developed and are being analyzed for

possible introgression at the phenotypic and molecular levels. The objective of our investigation was to determine the nature and magnitude of introgression from O. brachyantha and O. granulata into rice and to characterize the nature of extra chromosome of wild species in monosomic alien addition lines (MAALs) using molecular markers. The material comprised parents (O. sativa cultivar IR56 and an elite breeding line IR31917-45-3-2, O. brachyantha accession 101232, and O. granulata accession 100879), F1, BC1, and advanced backcross progenies. We analyzed 30 derivatives (12 BC3F3, 11BC 3 F5 , 7 MAALs) of O. sativa/O. brachyantha and 40 derivatives (32 BC2F3, 5 BC3F1, 3 BC4F1) of O. sativa/O. granulata. RFLP analysis was carried out using mapped markers on chromosomes 6, 7, 9, 10, 11, and 12 (Causse et al 1994). Forty-seven markers comprising 27 cDNA and 20 genomic probes were used with seven restriction enzymes (EcoRI, EcoRV, Dra1, HindIII. Bam HI, Xba1, Sca 1). The hybridizations and washings were carried out at 60 C under high salt conditions. All of the marker-enzyme combinations, except a few, were polymorphic (see table). The hybridization signal intensities for the marker-specific target DNA sequences. done under conditions permissible to detect 70-75% DNA homology (washing stringency of 2X SSC at 60 C), were moderate to low for most of the markers in both the wild parents, especially O. granulata (see table). The divergence was pronounced for chromosomes 9 and 12

specific-marker sequences in both wild species. Five MAALs of O. brachyantha, which were identified based on phenotypic and cytological observations. were confirmed at the molecular level. In most of the MAALs, the alien chromosome was modified. Despite the high divergence of alien genomes from cultivated rice, introgressions were detected for a few of the markers in the derivatives from both of the crosses (Fig. 1a). The analysis of O. granulata derivatives revealed two new MAALs for chromosomes 9 and 11 (Fig. 1b). In general, the frequency of introgression was low. Several markers detected nonparental bands in wide-cross derivatives (Fig. 1b). About half of these involved modification(s) of Eco RI-specific alleles. Further analysis is under way to determine why such bands occurred. The results showed that the genomes of O. brachyantha and O. granuluta are highly diverged from that of cultivated rice, with O. granulata being more diverse. Despite high divergence, chromatin material can get introgressed into cultivated rice from distantly related genomes of Oryza. The occurrence of nonparental bands in wide crosses is probably due to genome repatterning resulting from the interaction between unrelated genomes.

Cited reference
Causse MA, Fulton TM, Cho YG, Ahn SN, Chunwongse J, Wu K, Xiao J, Yu Z, Ronald PC, Harrington SE, Second G. McCouch SR. Tanksley SD. 1994. Saturated molecular map of the rice genome based on an interspecific backcross population. Genetics 138:12511274.

Polymorphism and hybridization signal intensities for marker-specific target DNA sequences in the wild parents relative to those of O. sativa. Chromosome Markers tested (no.) 9 9 8 6 8 7 47 Polymorphic markers (%) 100bc 100 100 100 100 100 100 Monomorphic markerenzyme combinations O. granulata O. brachyantha 1/63 2/63 0/56 1/42 1/56 0/49 5/329 2/63 2/63 0/56 1/42 1/56 0/49 6/329 Markers with null phenotype O. granulata O. brachyantha 2 1d 1d 1 1 1d 8 2 1d 0 1 1 0 5 Hybridization signal a (%) O. granulata O. brachyantha 20-50 30-50 10-20 30-50 20-50 10-20 50-100 35-100 20- 40 35-100 40-100 20- 30

6 7 9 10 11 12 Total

a Under relaxed conditions of hybridizations and washings permissible to detect 70.75% DNA homology relative to O. sativa. b For at least five restriction enzymes. c Markers with null phenotype are considered polymorphic. d No discrete bands: signal highly dispersed: O. sativa alleles multicopy.


IRRN 21:2-3 (August-December 1996)

Dra 1 - RFLP patterns of introgression lines (O. sativa/O. granulata). Lanes. l - molecular weight marker; 1-10 and 14-23 - different derived lines; 11 - O. sativa lR31917-45-3-2; 12- O. granulata (acc. 100879); 13-F 1 hybrid. a) Probe RZ884, chromosome 6. See O. granulata - specific allele in lanes 2 and 3 (marked with arrow). b) Probe R2797, chromosome 11. Nonparental band in lane 1 and O. granulata alleles in derived-MAALs for chromosome 11 (lanes 6 and 7).

Acrotrisomics in rice

K. Ikeda, H. Furuumi, A. Yoshimura, H. Yasui, and N. Iwata, Plant Breeding Laboratory, Faculty of Agriculture, Kyushu University, Fukuoka 812-81, Japan

Researchers have extensively used primary trisomics in genetic mapping of rice (Oryza sativa L.). However, no information

existed previously on the occurrence of acrotrisomics in rice. We isolated six acrotrisomic plants from selfed progenies of primary trisomics and from the F 1 obtained from pollination of primary trisomics with irradiated pollen. An extra chromosome was identified in these acrotrisomics. One of the acrotrisomics was used in chromosome mapping. Triplo 4, with liguleless gene (lg lg lg), and Triplo 7 were used for crossing as female plants. Among the F1 plants,

acrotrisomic-like plants were selected based on the phenomenon of pseudodominance (Triplo 4) and abnormal morphological characters (Triplo 7). In the progenies of primary trisomics (Triplo 4, 5, 7, and 11), acrotrisomic-like plants showed abnormal morphological characters spontaneously. Among the selected plants, six acrotrisomics were identified by mitotic chromosome analysis (Acro4S 4L -a, Acro4S 4L -b, Acro5, Acro 11) and/or restriction fragment

IRRN 21:2-3 (August-December 1996)


length polymorphism (RFLP) gene dosage analysis (Acro4S 4L -a, Acro7a, Acro7b). To distinguish two Acro4S 4L isolated from different origins, the acrotrisomics were designated as Acro4S 4L -a and Acro4S 4L -b, and Acro7. Acro4S 4L -a and Acro7a were selected in the F1 plants; the other acrotrisomics were obtained from the progenies of primary trisomics. Acro4S 4L -a and Acro4S 4L -b each had a complete short arm (4S) and heterochromatin region of the long arm (4L) on chromosome 4 in addition to a normal chromosome complement as identified by mitotic chromosome analysis. RFLP gene dosage analyses were studied based on RFLP map (Saito et a1 1991). RFLP gene dosage effects of Acro4S 4L -a were observed at the loci on XNpb203, 237, and 311 but not at the loci on XNpb49, 114, and 120 on chromosome 4. These plants had narrow grains, as did those of Acro4S 4L -b. Acro5 had a fragment of about one-third of chromosome 5, and the plants had small grains and short panicles. RFLP gene dosage effects of Acro7a and Acro7b were observed at the locus of XNpb338 but not at the loci XNpb33 to 22 on chromosome 7. These plants had compact panicles and round grains. Acrol1 had a fragment containing heterochromatin. Morphological features of the plants were late heading and short stature. The acrocentric chromosome of each acrotrisomic was easily transmitted to the

Breeding behavior of acrotrisomics upon selfing. Acrotrisomlc Acro4S 4L -a Acro4S 4L -b Acro5 Acro7a Acro7b Av (total) Plants (%) Disomics 55.2 61.9 59.4 71.3 66.4 63.5 Acrotrlsomics 35.8 37.1 38.1 27.3 30.6 33.3 Primary trisomics 9.1 0.9 2.5 1.5 3.0 3.2

Plants (no.)

618 854 286 891 265 (2914)

next generation. The average frequency of acrotrisomics in the selfed progenies of five acrotrisomics was 33.3%, ranging from 27.3% in Acro7a to 38.1% in Acro5 (see table). RFLP gene dosage analysis of Acro4S 4L a suggested that the break point of the acrocentric chromosome was in the region between XNpb311 and 49 (10.5 cM) on chromosome 4, and that the chromosome lacked the region from XNph49 to lg on the long arm. These findings and integrated linkage map (Ideta et a1 1992) revealed the orientation of the marker genes on chromosome 4 (see figure). The arrangement of classical and RFLP marker genes from the long arm are lg-XNph120 114 - d - 11 - 49- 311 - 237 - 203.

Construction of chromosome 4 map using

acrotrisomic-Acro4S 4L -a: classical linkage map (left), RFLP linkage map (center), and ideogram of chromosome 4 (right).
a Italicized numbers indicate RFLP markers.

Cited references
Ideta O, Yoshimura A, Matsumo T. Tsunematsu H, Iwata N. 1992. Integration of conventiona1 and RFLP linkage maps in rice. I. Chromosomes 1, 2, 3, and 4. Rice Genet. Newsl. 9:128-129.

Saito A, Yano M, Kishimoto N, Nakagahra M, Yoshimura A, Saito K, Kuhara S, Ukai Y, Kawase M, Nagamine T, Yoshimura S, Ideta O, Ohsawa R, Hayano Y, Iwata N, Sugiura N. 1991. Linkage map of restriction fragment length polymorphism loci in rice. Jpn. J. Breed. 41:665-670.

Molecular mapping of fertility-restoring gene Rf3 in rice

G. Zhang and Y. Lu, South China Agricultural University, Guangzhou 510642, China; and N. Huang, IRRI

Cytoplasmic male sterility in WA-Zhenshan 97 A (ZSA), a male sterile line widely used for developing hybrid rice in China, is caused by the interaction between wild abortive cytoplasm and two nuclear genes. A series of near-isogenic lines (NILS) carrying different genotypes for restoration was developed by nine backcrosses to ZSA using IR24, a restorer line with two fertilityrestoring genes, as the donor (Zhang et a1 1994).

To tag the restorer genes, the set of NILS and the donor IR24 were used for randomly amplified polymorphic DNA (RAPD) analysis. From a survey of 720 random primers, we identified six RAPD markers that were associated with one of the two restorer genes. Three of the six RAPD markers (OPKO5-800,OPU10-1100, and OPWOl-350) were used as probes. They were located on chromosome 1 in a population of doubled haploid lines derived from cross IR64/Azucena. An F2 population was developed from the cross between ZSA and ZSRR21. ZSR21 is a NIL carrying two restorer genes. Thirty-six sterile plants in the F2 segregating population were selected for linkage analysis. The three RAPD markers and three restriction fragment length

polymorphism markers (RG532, RG140, and RG458 located on chromosome 1) were found to be closely linked to the restorer gene. The distances between each of the six markers and the restorer gene were less than 4 cM in this population (see figure). For fertility restorer genes in rice, Rf1 was the notation given to the gene located on chromosome 10, which restores male fertility in line CMS-B (Shinjyo 1975). Rf2 was the notation given to the gene located on chromosome 2, which restores male fertility in line CMS-L (Shinjyo and Sato 1994). In this study, the restorer gene located on chromosome 2, named Rf2 in a previous paper, is named Rf3 (Zhang et a1 1994).


IRRN 21:2-3 (August-December 1996)

Location of the restorer gene Rf3 on chromosome 1. The map was developed using the F 2 population of 36 sterile plants derived from the cross Zhenshan 97 A/ZSR21. Molecular markers are shown on the right side of the figure. Map distances (in cM) are based on the Kosambi function.

Cited references
Shinjyo C. 1975. Genetical studies of cytoplasmic male sterility and fertility restoration in rice, Oryza sativa L. Sci. Bull. Coll. Agric. Univ. Ryukyus 22:1-57. Shinjyo C., Sato S. 1994. Chromosomal location of fertility-restoring gene Rf2. Rice Genet. Newsl. 11:93-95. Zhang G, Lu Y, Huang N. 1994. Molecular analysis of introgressed chromosomal segments in a set of near-isogenic lines of rice for fertility-restoring genes. Rice Genet. Newsl. 11:147-149.

Characterization of common cis -regulatory elements responsible for endospermspecific expression of rice glutelin gene
F. Takaiwa, Cell Biology Department, National Institute of Agrobiological Resources (NIAR), Tsukuba 305, Japan; T. Yoshihara, Biotechnology Department, Central Research Institute Electric Power Industry, Abiko 1646, Japan; H. Washida, F. Tanabe, and K. Ozawa, Cell Biological Department, NIAR, Japan; U. Yamanouchi, and K. Yamada, Biology Department, Toyama University, Toyama 930, Japan; and C-Y Wu and S. Toki, Cell Biology Department, NIAR, Japan

Glutelin is the major seed storage protein of rice, accounting for about 80% of total endosperm protein. Since glutelin genes are expressed only in maturing endosperm tissue and are regulated mainly at the transcriptional level, they provide excellent

model systems of regulation of tissuespecific gene expression. Glutelin is coded for by a small multigene family composed of about 10 copies per haploid genome (Okita et al 1989, Takaiwa and Oono 1991, Takaiwa et al 1991). These genes are clearly classified into two groupsdesignated as GluA and GluB subfamiliesthat share about 60-65% homology between members of each subfamily at the nucleotide sequence level (Takaiwa et al 1991). Comparison of the nucleotide sequence in the 5' flanking regions of six glutelin genes showed that the conserved sequences are restricted to the 13 bp of AACA motif around -70, the GCN4 motif around -160 and -90, and the GCAA motif around -350 from the transcriptional start site. Therefore, these motifs may be candidate for cis-regulatory elements controlling the endospermspecific expression found in the glutelin genes. To identify the common cis-regulatory elements responsible for the endospermspecific expression of glutelin genes, the 5' and 3' nested deletions and internal deletions of the 5' flanking regions of the GluB-1 and GluA-3 genes were transcriptionally fused to either glucuronidase (GUS) or firefly luciferase (LUX) reporter genes. These chimeric genes were transferred into the tobacco genome via Agrobacterium-mediated transformation system or into the rice genome using electroporation method. The GUS or LUX activities were assayed in maturing seeds. The GUS reporter gene was expressed at high levels in the sub-aleurone layer of the maturing rice seed. Essential cis -regulatory elements governing the spatially and temporally specific expression of the glutelin genes were located within the first 437 bp and 245 bp of the promoter regions of the GluA-3 and GluB-1 genes, respectively (Yoshihara and Takaiwa 1995). The AACA motif and the GCNA4 motif within the 100 bp of the 5' flanking region, common to all the glutelin penes, were the key elements determining endospermspecific expression of glutelin genes. Deletion or site-specific mutagenesis of this proximal AACA motif remarkably reduced

promoter activities. The GCN4 motif also acted as a major cis-regulatory element determining the quantitative level. Each of these elements is necessary, but not sufficient, for endosperm-specific expression because tissue specificity is retained irrespective of the absence of each element. However, deleting the entire region, which includes both the AACA and GCN4 motifs, abolished seed-specific expression, suggesting that the synergistic interaction between both elements determines the endosperm-specific expression of the glutelin genes. To confirm the significance of combining the AACA and GCN4 motifs to confer the endosperm-specific expression, the regions (-245 to -145, - 113 to - 40), including both motifs, were fused to the truncated (-90 to +9) or core (-46 to +1) CaMV35S promoter/GUS cassette, and then introduced into tobacco and rice. Although neither the AACA motif nor the GCN4 motifby itselfis capable of enhancing the truncated promoter of CaMV35S promoter, their combination conferred the endosperm-specific expression. However, the combination failed to activate promoter activity of the core CaMV promoter. The results suggest that the combination of two sets of AACA and GCN4 motifs or the combination of one set of these two motifs and the G-box motif is required for the endosperm-specific expression of the rice glutelin gene.

Cited references
Okita TW, Hwang YS, Hnilo J, Kim WT, Aryan AP, Larson R, Krishnan HB. 1989. Structure and expression of the rice glutelin mutigene family. J. Biol. Chem. 265:12573-12581. Taikawa F, Oono K. 1991. Genomic DNA sequences of two new genes for new storage proteinglutelin in rice. Jpn. J. Genet. 66:161171. Takaiwa et al. 1991, Sequences of three members and expression of a new major subfamily of glutelin gene from rice. Plant Mol. Bid. 17:875-885. Yoshihara T, Takaiwa F. 1995. cis-regulatory elements responsible for quantitative regulation of rice seed storage protein glutelin GluA-3 gene. (in press)

IRRN 21:2-3 (August-December 1996)


A fertility-restoring revertant, controlled by a single gene induced from a rice cytoplasmic male sterile line
Shen Yuwei, Institute of Genetics (IG), Fudan University, Shanghai 200433, China: Cai Qihua and Gao Mingwei, Institute of NuclearAgricultural Sciences, Zhejiang Agricultural University, Hangzhou 310029, China; and Wang Xunming, IG, Fudan University

Cytoplasmic male sterility is inherited steadily through the maternal parent. However, reversions of cytoplasmic male sterile (CMS) lines to male fertile lines, either by cytoplasmic or nuclear gene mutation, have been documented (Nawa et al 1987, Smith 1987, He et al 1989). We report a fertility-restoring revertant induced from an indica rice CMS line II-32A treated with -rays, which is perhaps the first case in rice. Dry seeds (13% moisture) of CMS rice line II-32A, which possesses the cytoplasm of an Indonesian rice variety, were irradiated with 60C-g rays at a dose of 290 Gy. Eight plants with an open air seed-setting rate of more than 30% were selected from about 1,000 M 1 II-32A individuals planted in an isolated area. These plants showed segregation in later generations. Among them, line T24 segregated for completely sterile and fully fertile plants. The sterile plants had 100% pollen grains that were shrunken and not stainable with I2-KI. The fertile plants had a seed-setting rate of about

80% in the open air and more than 90% dark-stained pollen grains. In the 1993 early season, revertant T24 line was testcrossed with Zhenshan 97 A and II-32 A for its fertility-restoring ability. The results were grouped into two sets. In set one, both the parental T24 and the F1 plants were fertile. In set two, the parental T24 plants segregated into a 3 fertile-1 sterile ratio. F1 plants showed a 1 fertile-1 sterile ratio, indicating that the revertant could restore the fertility of CMS lines and one pair of nuclear genes of T24 might control the restoring ability. To clarify the above facts, the seedsetting rate distribution in F2s of the crosses between II-32 A, Zhenshan 97 A, DShan A, Xieqinzao A, and T24 were investigated in the 1993 late season in Hangzhou and in the 1994 early season in Hainan Island. The F2 seed-setting rate distributions of all four crosses were clearly divided into sterility (0-40%) and fertility(70- 100%). The supplemental pollen vigor assay with I2-KI revealed the pollen grains of the sterile plants in the open air to be completely abortive, the seed setting of which resulted from cross pollination. The ratio of the fertile plants to sterile ones in F2s of the four crosses was 3:1 ( c 2 = 1.12,0.20<p>0.30 for II-32 A/T24, c 2 = 0.2735, 0.50<p>0.70 for Zhenshan 97 A/T24, c 2 =0.003,p>0.99 for DShan A, c 2 =0.1077, p=0.30 for Xieqinzao A/T24). The genetic mode of the fertility-restoring ability of T24 was different from that of Minghui 63 and 20964, the leading

restorers in southern China (see table). The bagged seed-setting rate and the seedsetting rate in open air of II-32 A/20964 and II-32A/Minghui 63 were widely distributed. The fertility-restoring ability possessed by 20964 and Minghui 63 was thought to be controlled by two pairs of nuclear genes, based on the proportion of sterile individuals to total segregating F 2 plants. The fertility-restoring revertant created in this study is of significance because first, the reversion process inferred that rice CMS is not as complicated as previously thought (conditioned by the incompatibility between cytoplasm and nucleus). The revertant, still possessing sterile cytoplasm, provides not only a restoring gene source, but also a restorer selection system. The revertant also provides excellent material for CMS molecular studies.

Cited References
He P, Li Z, Li T. 1989. Fertility restoring mutation in T type wheat cytoplasmic male sterile line irradiated with 60Co- grays. Acta Genet. Sin. 16:1-6. Nawa S, Sano Y, Yamada M, Fuji T. 1987. Cloning the plasmids in cytoplasmic male sterile rice and changes of organization of mitochondrial and nuclear DNA in cytoplasmic reversion. Jpn. J. Genet. 62:301314. Smith RL. 1987. Mitochondrial DNA rearrangements in Pennisetum associated with reversion from cytoplasmic male sterility to fertility. Plant Mol. Biol. 9:277286.

Distribution of F 2 seed-setting rate for crosses of T24 with indica CMS rice lines. Cross combinationb llA/T24 ZSA/T24 DSA/T24 XQZA/T24 llA/MH63 llA/20964 Seed-setting rate distribution a 0 10 16 13 19 18 6 6 20 12 11 3 4 6 10 30 5 2 3 9 7 40 1 50 60 70 8 2 6 3 3 7 80 16 3 7 11 1 3 90 25 21 21 18 1 100 c 2 (3:1)

3 3

9 7

10 18

3 5 3 9 17

45 1.1200 0.20-0.30 42 0.2735 0.50-0.70 30 0.0003 >0.99 33 0.1077 0.3 k=(logN-logM)/0/6021c

a Data for llA/MH63 and llA/20964 are bagged seed-setting rate while others are seed-setting rate in open air. b ZSA = Zhenshan 97 A, DSA = DShan A, ZQZA = Xieqinzao A, MH63 = Minghui 63. c The number of restoring genes in MH63 and 20964 were estimated with this formula.


IRRN 21:2-3 (August-December 1996)

Effects of Se1 gene on basic vegetative growth of rice

M. Niwa, K. Terakado, and N. Takeda, School of Agriculture, lbaraki University, Ami-machi, Inashiki-gun, Ibaraki, 300-03, Japan

24-h photoperiod as LD in the open air chamber with a mean temperature of 32C. Three plants were grown per pot; two pots were used per treatment. White fluorescent light (600 1x at the soil surface) was supplemented with sunlight to prolong photoperiod as needed.

The Se1 gene, located on the sixth chromosome of rice (Oryza sativa), plays an important role in determining the heading time in many rice varieties (Yokoo et al 1980). Through backcrossing, one of the alleles on the Se1 locus, Se1-u, was introgressed from Malaysian rice variety Morak Sepilaito to a genetic background of Fujisaka 5, which has Se1-e (Yokoo and Fujimaki 1971). This occurrence suggested that Se1-u brought late heading by increasing photoperiod sensitivity. Yokoo and Kikuchi (1982) proposed that Se1-u controlled basic vegetative growth. Our objectives in this study were to clarify the effects of Se1-e on vegetative growth and determine whether those effects varied with photoperiod. Two near-isogenic lines of rice with the genetic background for Fujisaka 5 and different alleles on the Se1 locus, ER (Se1e) and LR (Se1-u), were exposed to various photoperiod regimes in two series of experiments. In the first series, the plants were seeded and grown in a phytotron at 28C for 10 d under a 24-h, long-day (LD) photoperiod. Then some of the plants were shifted to a 12-h, short-day (SD) photoperiod at 4-d intervals (LD-SD shifting). In the second series, LD-SD shifting was conducted, using a 10- or 13-h photoperiod as SD and a

1. Days after sowing to heading of isogenic lines ER and LR, as affected by days after sowing when the plants were alternately shifted from 12 to 24 h photoperiod every 4th day.

In the first series, the horizontal parts of the line graphs were determined by averaging days to heading (DTH), which did not differ significantly from those of the control plants grown under 12 h SD from sowing to heading (Fig. 1). Increase in DTH with days from sowing to shifting was approximated by linear regression using significantly larger DTH than those of the control. Consequently, the turning points of the graphs project estimated days after sowing during which SD did not accelerate DTH. These days are the photoperiod-insensitive phase (PIP) for each isoline. The estimated length of PIP was 24.8 d for ER and 18.0 d for LR. In the second series. the estimated length of PIP for each isoline and for each photoperiod regime was 16.2 d for ER-10 SD, 18.8 d for ER-13 SD, 12.7 d for LR-10 SD, and 10.7 d for LR-13 SD (Fig.2). Here, ER also exhibited longer PIP than LR. PIP did not differ significantly between regimes within the same isoline. We concluded that Se1 affected not only photoperiod sensitivity but also duration of PIP, and that duration of PIP did not vary with photoperiod but rather with temperature. Cited references
Yokoo, M, Fujimaki H. 1971. Tight linkage of blast resistance with late maturity observed in different indica varieties of rice. Jpn. J. Breed. 21:35-39. Yokoo M, Kukuchi F, Nakane A, Fujumaki H. 1980. Genetical analysis of heading time by aid of close linkage with blast resistance in rice. Bull. Natl. Inst. Agric. Sci. D31:96 - 126. Yokoo M, Kukuchi F. 1982. Monogenic control of basic vegetative phase and photoperiodsensitive phase in rice. Jpn. J. Breed. 32: 1-8.

2. Days from sowing to heading of isogenic lines ER (a) and LR (b) as affected by days from sowing when the plants were shifted from a 10or 13-h to a 24-h photoperiod.

Genetic and cytochemical analysis of high 57-kD polypeptide mutants in rice

Y. Takemoto, M. Y. Son, T. Ito, and H. Satoh, Faculty of Agriculture, Kyushu University, Fukuoka 812, Japan; T. Kumamaru, Plant Genetics and Breeding Laboratory, Japan Tobacco Inc., lwata, Sizuoka 438, Japan; and M. Ogawa, Faculty of Home Economics, Yamaguchi Women's University, Yamaguchi 753, Japan

High 57-kD polypeptide mutants (57H mutants) were induced by mutagen N-

methyl-N-nitrosourea (MNU) treatment (Kumamaru et al 1988), while the 57H spontaneous mutants were found in northern Asian varieties. Four 57H mutant loci (esp2, Glup1, glup2, and glup3) have been located on chromosomes 11, 9,9, and 4, respectively (Satoh et al 1994). The 57-kD polypeptides of these mutants reacted with anti-glutelin subunit antibody, indicating that 57H mutants accumulate 57 kD glutelin precursors. Glutelins in wild-type Kinmaze were extracted using 1% lactic acid solution (acid solution). The presence of prolamins did not affect extraction of glutelins at all. In

esp2 mutant CM1787, the glutelin precursors could not be extracted without first removing the prolamins. In Glup1 mutant EM61 and glup2 mutant EM305, the glutelin precursors were extracted to a slight extent by acid solution, whereas removing the prolamins promoted extensive extraction of glutelin precursors. These results indicate that glutelin precursors in these mutants coexist with prolamins. In glup3 mutant HO1274, the glutelin precursors were extracted as easily as the glutelins of Kinmaze, suggesting that the deposition site of the glutelin precursors is different from that of the prolamins.
IRRN 21:2-3 (August-December 1996) 19

Molecular mapping of genes for F1 pollen sterility in rice

Zhuang Chuxiong, Zhang Guiquan, Mei Mantong, and Lu Yonggen, South China Agricultural University (SCAU), Guangzhou 510642, China

1. Electron micrographs of the developing endosperm of a) Kinmaze and b) CM1787 ( esp2 mutant). Arrow denotes new type of protein bodies. Bar = 0.5 m.

2. Electron micrographs of the developing endosperm of esp2 mutant showing the specificities of a) anti- b subunit antibody and b) anti-13b prolamin polypeptide antibody for a new type of protein body (denoted by arrow). Bar = 0.5 m.

Two types of protein bodies (PB) (PB-I and PB-II) were observed in the endosperm of Kinmaze, as reported by Tanaka et al (1980). In CM1787, although new types of PB and PB-II were observed, PB-I was absent. Many ribosomes were observed on the surface of the new type of PB (Fig. 1 b), suggesting that these PBs, such as PB- 1, were derived from endoplasmic reticulum. Immunogold labeling (Fig. 2) showed that the glutelin precursors of esp2 mutants were deposited in the new type of PB together with prolamin, suggesting that the presence of glutelin precursor in PB-I leads to the formation of the new PB type. In three other types of mutants, both PB-I and PB-II were observed; the new type of PB was not found. The localization of glutelin precursors in these mutants was examined. These observations indicate that the different mutations in the processing pathway of glutelin precursors can cause PB variations among mutants.

We conclude that these mutations are concerned with the genes controlling the processing of glutelin precursors. The study confirmed that esp2 mutation can result in the deposition of 57-kD glutelin precursors in PB-I, and that Esp2 is a gene controlling the proteins responsible for targeting glutelin precursors toward the protein vacuole.

Kumamaru T, Satoh H, Iwata N, Omura T, Ogawa M, Tanaka K. 1988. Mutant of rice storage proteins. 1. Screening of mutants for storage proteins in the starchy endosperm. Theor. Appl. Genet. 76:11-16. Satoh H, Kumamaru T, Yoshimura S, Ogawa M. 1994. New 57kDa glutelin genes on chromosome 9 in rice. Rice Genet. Newsl. 11:158-161. Tanaka K, Sugimoto T, Ogawa M, Kasai Z. 1980. Isolation and characterization of two types of protein bodies in the rice endosperm. Agric. Biol. Chem. 44:1633-1639.

Cited references

F1 hybrid sterility, probably caused mainly by pollen sterility constrains the use of heterosis in the subspecific hybrid between indica (Hsein) and japonica (Keng). Zhang and Lu (1993) proposed that at least six loci are involved in controlling F1 hybrid pollen sterility. To map S-c, one of the F 1 pollen sterility genes, Taichung 65 (a japonica variety, denoted as E1) and E5, its isogenic F1 pollen sterile line carrying this gene, were used as parents. E5 was developed using indica variety Pehku as the donor. It was then backcrossed with E1 for 11 generations. One hundred and four F 2 plants from E5/E1 were used for the segregation analysis. Pollen fertility was 98.5% for E1 and 98.6% for E5; however, pollen fertility for F1 plants from the cross E1/E5 was 53.2%. Segregation of fertile and sterile plants in the F2 population from this cross fit well to a 1-1 ratio (c2 =0). One hundred and seventy restriction fragment length polymorphism (RFLP) markers were used to survey polymorphism between E1 and E5, and only three markers located on the same chromosome detected different RFLP patterns. The three positive markers (RG227, RG166, RG369A) were tightly linked to each other and were mapped on chromosome 3 (Causse et al 1994). Polymorphism was low between E1 and E5, which could be explained by the short length of introgressed segments in the isogenic line, because it was developed through repeated backcrossing. The polymorphic RFLP markers were then used to survey the filters with DNA from 104 F 2 plants, and cosegregation of fertility and RFLP patterns were analyzed. The genetic distances between the S-c locus and markers RG227 and RG369 were 0.5 and 2.5 cM, respectively (see figure).


IRRN 21:2-3 (August-December 1996)

Location of the S-c gene on chromosome 3. Markers are on the right and map distances (in cM) are on the left.
F2 Segregation Patterns for RFLP markers in the cross between E 1 and E 5. Locus RG227 RG166 RG369 j/j 0 2 2 j/i 51 52 52 i/i c 53 50 50

(1:2:1) 50.1 44.3 44.3

Distorted F 2 ratios were observed for the three RFLP markers in the cross (see table). The results suggest the use of the one-locus sporo-gametophytic model in explaining F1 pollen sterility in cultivated rice (Zhang and Lu 1993).

Causse MA, Fulton TM, Cho YG. Ahn SN, Chunwongse J, Wu K, Xiao J, Yu Z, Ronald PC, Harrington SE, Second G. McCouch SR, Tanksley SD. 1994. Saturated molecular map of the rice genome based on an interspecific backcross population. Genetics 138: 12511274. Zhang G, Lu Y. 1993. Genetic studies on the hybrid sterility in cultivated rice (Oryza sativa). II. A genic model for F1 pollen sterility. Chin. J. Genet. 20:249-255.

Cited references

functionally very similar to eubacterial ribosomes, most likely reflecting their endosymbiotic origin. Several chloroplast ribosomal protein genes have been cloned and sequenced, mostly in dicotyledonous plants. Publication of complete nucleotide sequences of the rice chloroplast genome (Hiratsukaet al 1989) has allowed scientists to predict about two-thirds of the 55-60 chloroplast ribosomal proteins encoded in the nucleus. These nucleus-encoded chloroplast ribosomal proteins are synthesized on cytoplasmic 80S ribosomes as precursor polypeptides with a transit peptide sequence, which is cleaved off during transport into the organelle. We isolated and sequence CDNA clones for nuclear-encoded genes rpl13 and rpl28, which encode chloroplast ribosomal proteins L13 and L28, respectively, in rice. With poly(A)+-RNA isolated from green leaves of 10-d-old rice seedlings (Oryza sativa L. cv Nipponbare), cDNA libraries were constructed in ZAPII (Stratagene). Based on the highly conserved region among chloroplast ribosomal protein sequences reported in other plants, such as tobacco and spinach, the corresponding rice DNA fragments were amplified using reverse transcriptase-polymerase chain reaction and used as probes. Nucleotide sequences of positive clones were determined using the dideoxy nucleotide chain termination method on doublestranded DNA templates. The cDNA clone of rpl13 consists of 1,076 nucleotides, including poly(A) 29 tail, and contains an open reading frame of 702 bp. The cDNA clone of rpl28 consists of 781 nucleotides, including poly(A) 20, tail, and contains an open reading frame of 411 bp. Compared with sequences from spinach and tobacco, amino acid sequences deduced from cDNA sequences of rpl13 and rpl28 are more conserved in the mature peptide region, which is actually assembled into chloroplast ribosome. than in the transit peptide region. The mature L13 protein has 85 and 58% amino acid sequence identity with the counterparts of spinach (Phua et al 1989) and Escherichia coli (Isono et al 1985), respectively, in the homologous overlapping regions (see figure, region C). The amino acid identity of the mature

A schematic representation of local alignment of amino acid sequences of rp128. Predicted amino acid sequence of rice rp128 cDNA was aligned with the homologues of spinach and E. coli using MACAW algorithm (Lawrence et al 1993). Darker shade indicates sequence blocks with higher homology scores.

protein of L28 is 75 or 32% relative to tobacco (Yokoi and Sugiura 1992) and E. coli (Lee et al 1981). The mature L13 protein of rice contains 24 amino acid residues of N-terminal exlension (see figure, region B), which is 16 residues shorter than that of spinach.

Cited references
Isono S, Thamm S, Kitakawa M, Isono K. 1985. Cloning and nucleotide sequencing of the genes for ribosomal proteins S9 (rpsl) and L13 (rplM) of Escherichia coli. Mol. Gen. Genet. 198:779-282. Lawrence CE, Altschul SF, Boguski MS, Liu JS, Neuwald AF, Wootton JC. 1993. Detecting subtle sequence signals: a gibbs sampling strategy for multiple alignment. Science 262:208-214. Lee JS, Am G, Friesen JD. Isono K. 1981. Cloning and the nucleotide sequence of the genes for Escherichia coli ribosomal protein L28 (rpmB) and L33 (rpmG). Mol. Gen. Genet. 184:218-223. Phua SH, Srinivasa BR, Subramanian SR. 1989. Chloroplast ribosomal protein L13 is encoded in the nucleus and is considerably larger than its bacterial homologue. J. Biol. Chem. 264: 1968-1971. Hiratsuka J, Shimada H, Whiottier RF, Ishibashi T, Sakamoto M, Mori M, Kondo C, Honji R, Sun CR, Mong BY, Li YQ, Kanno A, Nishizawa Y, Hirai A, Shinozaki K, Sugiura M. 1989. The complete sequence of the rice (Oryza sativa) chloroplast genome: intermolecular recombination between distinct RNA genes accounts for a major plastid DNA inversion during the evolution of the cereals. Mol. Gen. Genet. 217:185194. Yokoi F, Sugiura M. 1992. Tobacco chloroplast ribosomes contain a homologue of E. coli ribosomal protein L28. FENS Lett. 308:258260.

Chloroplast ribosomal protein genes encoded in rice nuclear genome

N. Tsutsumi, S. Takusagawa, M. Kurashige, Y. Li, and A. Hirai, Laboratory of Radiation Genetics, Faculty of Agriculture, The University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113, Japan

Plant cells contain three distinct types of ribosomes that are found in the cytosol, mitochondria, and chloroplasts. Chloroplast ribosomes are structurally and

IRRN 21:2-3 (August-December 1996)


Stability of mitochondrial plasmidlike DNAs of rice

A. Kanazawa, N. Tsutsumi, and A. Hirai, Laboratory of Radiation Genetics, Faculty of Agriculture, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113, Japan

B1 and B4 are small, circular plasmidlike DNAs found in rice mitochondria. We previously investigated the distribution of four kinds of plasmidlike DNA molecules in 85 Oryza sativa accessions (Kanazawa et al 1992). These molecules were found mostly in indicas, with a few in javanicas, and none in japonicas. Copy number also differed among the four kinds of molecules. Copy number was not conserved among cultivated rice varieties with different cytoplasms. It seems possible that mitochondrial plasmidlike DNAs could be replicated and stably maintained for at least several generations under normal environmental conditions. However, transmission of these DNAs may become irregular as a result of suppressed replication under unstable environmental conditions. We examined whether temperature influences copy numbers of mitochondrial plasmidlike DNAs and main mitochondrial genomic DNAs, especially during cell proliferation. The observed quantitative fluctuations in levels of plasmidlike DNAs were larger and less closely related to cell proliferation than those of main mitochondrial genomic DNAs on which genes indispensable for mitochondrial biogenesis were located. The copy number of plasmidlike DNAs was reduced significantly in calli cultured at high temperatures (Fig. 1). We developed an assay system for monitoring DNA synthesis in isolated mitochondria to examine the contribution of mtDNA replication to the quantitative fluctuations. Synthesis of the main mitochondrial genomic DNAs occurred at all the temperatures examined, whereas that of plasmidlike DNAs occurred only over a limited range of temperatures (Fig. 2). These results indicate the instability of plasmidlike DNAs relative to main mitochondrial genomic DNAs.

1. Gel blot analysis of DNAs from calli cultured at various temperatures. Total DNA (10 g) isolated from calli cultured at 20 C (lane 1), 25 C (lane 2), 30 C (lane 3), and 35 C (lane 4) was digested with Stul to linearize plasmidlike DNA, B1. After electrophoresis, the gel blot was probed with indicated on the right.

We have detected sequences homologous to mitochondrial plasmidlike DNAs that are integrated within limited chromosomal regions on the rice nuclear genome (Kanazawa et al 1993). This suggests interorganellar sequence transfer between mitochondria and nucleus. The unstable transmission of plasmidlike DNAs shown in this study implies that plasmidlike DNA sequences have been transferred from the mitochondria to the nucleus during at least the early differentiation phase among rice varieties, before the sequences were lost from the mitochondria of some rice varieties.

Cited references
Kanazawa A, Sakamoto W, Nakagahra M, Kadowaki K, Tsutsumi N, Tano S. 1992. Distribution and quantitative variation of mitochondrial plasmid-like DNAs in cultivated rice (Oryza sativa L.) Jpn. J. Genet. 67:309-319. Kanazawa A, Kishimoto N. Sakamoto W, Ohsawa R, Ukai Y, Tsutsumi N, Hirai A, Saito A. 1993. Restriction fragments homologous to mitchondrial plasmid-like DNAs are located within limited chromosomal regional on the rice nuclear genome. Theor. Appl. Genet. 87:3577-586.

P-labeled B1 and

atpA. Signals corresponding to B1 and atpA are

Sensitivity of plant height genes to gibberellic acid and their regulation by endogenous plant hormones in rice
He Zuhua and Li Debao, Institute of Biotechnology, Zhejiang Agricultural University, Hangzhou 310029, China

2. Gel electrophoresis of products of DNA synthesis in isolated mitochondria. MtDNAs purified from reaction mixtures incubated at 5 C (lane 1), 13 C (lane 2), 20 C (lane 3), 25 C (lane 4), 30 C (lane 5), 35 C (lane 6), 42 C (lane 7), and 50 C (lane 8) were separated on an agarose gel. After electrophoresis, the gel was placed onto an imaging plate and radioactivity was visualized and quantified using a bio-imaging analyzer. One signal of low mobility and two signals of higher mobility indicate the incorporation of labeled nucleotide into the main mitochondrial genomic DNAs and the plasmidlike DNAs, respectively.

We studied the sensitivities to gibberellic acid (GA3) of genes governing plant height in five rice genotypes and how endogenous hormones regulate them. We also investigated proteins related to plant elongation. Sensitivity of plant height genes to GA3. Five rice genotypes. Er-jiu-feng (EJF. semidwarf, sd1 gene), tall EJF (nearisogenic line [NIL] of EJF, tall Sd1 gene), tall Zhen-shan 97 (TZS, NIL of ZS, tall, eui gene), Chao ai (CA, dwarf, sdss(t) gene), and Kyushu-3 (D53, dwarf, D53 gene) were treated with 20 and 100 mg GA3 L-1 at seedling, tillering, and heading stages. Plant elongation was measured 8 d after treatment. The sensitivities of rice genes to GA3


IRRN 21:2-3 (August-December 1996)

1. Endogenous plant hormone contents in different rice genotypes: (a) GA 3 content (g g -1 FW), (b) IAA content (nmol g -1FW), (c) ZR content (pmol g-1 FW), (d) ABA content (nmol g -1 FW). GA contents of dwarf genotypes were 16.9-55.6% lower than those of tall genotypes. IAA contents were lower by 12.0-81.7% in the dwarf genotypes than in the tall genotypes. ZR and ABA contents varied among genotypes during plant growth.

2. Protein patterns in GA3-treated and nontreated NILS. Lanes 1-6 consisted of nontreated seedlings of TEJF, EJF, TZS, ZS, 320T, and CA; lanes 7-12 consisted of TEJF, EJF, TZS, ZS, 320T, and CA seedlings treated with GA 3 for 3 d. Proteins a (about 50 kD), b (about 62 kD), c (about 50 kD), which existed in nontreated EJF, TZS, and CA, are shown with arrows; these proteins appeared in other lines of the NIL pairs after 3 d of GA3 treatment. Another new protein d (about 52 kD) also appeared in CA. Protein standard molecular weights are shown on the left.

were found to be ui>D53>sd1>Sd1>sds(t) at the seedling stage. However, sensitivities of the genes changed to D53>sd1>eui>sds(t)>Sd1 during tillering and sd1>eui>D53>Sd1>sds(t) during heading. The genes eui and Sd1 were

classified as sensitive at all growth stages, D53 as sensitive during seedling and tillering stages, and Sd1 and sds(t) as insensitive at all stages. Using endogenous hormones to regulate plant height genes. Plant hormones were

assayed using the enzyme-linked immunosorbent assay method. More GA3 was found in the uppermost internodes of heading plants of the tall genotypes (Fig. 1a) than in the other genotypes. Although the eui plants have the highest GA3, they were highly sensitive to GA 3. The sds(t) plants were less sensitive to GA3 despite their lower GA3 content. GA3 and indole acetic acid (IAA) were higher in the tall genotypes than in the dwarf genotypes (Fig. 1b), the highest IAA content was also found in the eui plants. The tall genotypes had higher zatin ribozide (ZR) content than the dwarf genotypes at seedling and tillering stages (Fig. 1c), while at heading, higher ZR contents appeared in the dwarf genotypes. Except for the tall genotypes Sd1 and eui, in which the abscisic acid (ABA) contents remained stable from the tillering stage (Fig. 1d). increases in ABA of the three dwarf genotypes began during the seedling stage. Dwarf rice plants had more ABA and less GA3 and IAA at tillering and heading stages. Dwarf seedlings had lower ABA. Seedlings and tillering plants with low ABA were more sensitive to GA3 than those with higher ABA. The results indicate that genes governing plant height play an important role in regulating plant hormones. Identifying GA3 ,-induced proteins. According to sodium dodecylsulfate polyacrylamide gel electrophoresis of seedling proteins, three proteins (Fig. 2) with molecular weights of about 50, 62, and 50 kD, existed only in the EJF (sdl) (a), TZS (eui) (b), and CA (sds(t)) (c) lines, respectively. After a 3-d GA3 treatment, these proteins appeared in other lines of the NIL pairs and a new protein (d) also appeared in the CA plants. The proteins were possibly related to plant height and could have been induced by GA 3. Hormone receptors promote the responses of plants to exogenous hormones. Genes governing plant height may take part in regulating the receptor system through some message molecules, or the receptors may induce plant height genes to affect plant elongation. In another study, we found that amounts of peroxidases and esterases differed among seedlings with different genes.

IRRN 21:2-3 (August-December 1996)


Genetic diversity estimated among rice cultivars using restriction landmark genomic scanning method
M. Kawase, Laboratory of Plant Biotechnology, Crop Improvement Department, Shikoku National Agricultural Experiment Station, 3-1, Sen'yu-cho 1, Zentsuji, Kagawa 765, Japan

The restriction landmark genomic scanning (RLGS) method is a new technique developed for analyzing genomic DNA of higher organisms (Hatada et al 1991). RLGS, which uses restriction enzyme sites as landmarks, requires direct labeling at the restriction sites and two-dimensional electrophoresis to resolve these landmarks. It gives a two-dimensional pattern with thousands of scattered spots after autoradiography. The RLGS method provides unbiased information on genetic polymorphism throughout the whole

genome, an accurate estimation of genetic similarity, and is applicable to rice cultivars (Kawase 1994). We extracted DNA from leaves of 14-dold seedlings of two Japanese improved cultivars (Nipponbare and Koshihikari) and a Chinese land race (Liu Zhou Bao Ya Zao) (Kawase 1994). RLGS was conducted based on the procedure described by Hatada et al (1991), which consisted of blocking reaction, landmark cleavage, labeling, first fractionation, and second fractionation. Genetic similarities between cultivars were calculated using the measure devised by Dice (1945), which was suggested for DNA polymorphism by Nei and Li (1979): GS(i,j) = 2N(i,j)/[N(i)+N(j)], where GS(i,j) = the measure of genetic similarity between cultivars i and j; N(i,j) = the total spots common to i and j; and N(i) and N(j) = the number of spots for cultivars i and j, respectively. More than 3,000 landmarks were detected as scattered spots on the auto-

radiogram (Fig. I). The relative position and the intensity of each spot were highly reproducible. An RLGS profile was unique to each cultivar used. The RLGS profiles of Nipponbare and Liu Zhou Bao Ya Zao showed quite different patterns, indicating a wide genetic difference between the two cultivars. Restriction fragment length polymorphism analysis also revealed that they were distantly related (Kawase et al 1991). A limited area (Fig. 1) could be compared, although the different patterns made it rather difficult to identify each spot. Nipponbare and Liu Zhou Bao Ya Zao have 387 and 403 spots in the area, respectively, of which 136 were scored as common or indistinguishable spots. The low GS value, calculated as 0.344, suggests large genetic diversity among rice cultivars. Nipponbare and Koshihikari showed similar patterns. An area showing the clearest resolution contained 1,156 spots common to both cultivars (Fig. 2).

1. More than 3,000 spots were scattered on the RLGS profile obtained by autoradiography (Kawase 1994). Nipponbare (a) and Liu'Zhou'Bao'Ya'Zao (b) showed quite different patterns. The enclosed area was used for estimating the GS value between the two cultivars.


IRRN 21:2-3 (August-December 1996)

2. RLGS profiles of Nipponbare and Koshihikari were highly similar. However, the cultivars were distinguished based on the presence and absence of particular spots in the enclosed area (Kawase 1994). Arrow heads indicate the spots unique to each of the two cultivars compared.

Additionally, 24 and 25 spots were found unique to Nipponbare and Koshihikari, respectively. The GS value was calculated as 0.980. The RLGS method can be a powerful fingerprinting technique for estimating genetic diversity in rice as well as for identifying cultivars.

Dice LR. 1945. Measures of the amount of ecologic association between species. Ecology 26:297-302. Hatada I, Hayashizaki Y, Hirotsune S, Komatsubara H, Mukai T. 1991. A genomic scanning method for higher organisms using restriction sites as landmarks. Proc. Natl. Acad. Sci. USA 88:9523-9527. Kawase M, Kishimoto N, Tanaka T, Yoshimura A, Yoshimura S, Saito K, Saito A, Yano M,

Cited references

Takeda N, Nagamine T, Nakagahra M. 1991. Intraspecific variation and genetic differentiation based on restriction fragment length polymorphism in Asian cultivated rice, Oryza sativa L. Rice genetics II. Manila (Philippines): International Rice Research Institute. p 467-473. Nei M, Li WH. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proc. Natl. Acad. Sci. USA 765269-5273.

Character expression of a rice dwarf mutant with lax panicle

Y. Futsuhara, T. Horio, and S. Tanaka, Faculty of Agriculture, Meijo University, Tenpaku, Nagoya 468, Japan; and J. Yonemaru, Tohoku National Agricultural Experiment Station, Morioka, 020-01, Japan

A dwarf mutant with lax panicle was induced from the parent cultivar Fujiminori by 200 Gy of X-ray irradiation. The mutant exhibits leaf rolling during the early vegetative phase until the appearance of the 6th or 7th leaf. Inheritance of this mutant trait

was studied from an F 2 segregation involving a cross between the dwarf mutant and the parent cultivar. The F 1 was normal and the F 2 segregated into 103 normal plants and 39 dwarf plants, giving a good fit to a 3:1 ratio ( =0.460, P=0.498). Pleiotropic action of the dwarf mutant gene appears to be responsible for the lax panicles and rolled leaves. Panicle density is a complex character, which consists of many component characters such as spikelet number, panicle length, and number of branches (Futsuhara et al 1979). Panicle laxness in this mutant could be attributed to the decrease in total spikelets and branches, especially secondary branches (see table).

Rolled leaf is a marker trait that can be used in selecting dwarf mutant types during juvenile stage. To determine the annual and seasonal variations in various agronomic traits, the dwarf mutant and its parent cultivar were tested in 1992 and 1994. The rice plants were raised under two seeding times (early and late seedings) only in 1994. The dwarf mutant showed different responses to various seeding conditions (see table). The difference in panicle density between parent and mutant grown under late seeding condition was not significant. Seed fertility of the mutant, which was significantly less than that of the parent

IRRN 21:23 (August-December 1996) 25

Related characters of panicle density in a dwarf mutant with lax panicles and its parent cultivar, Fujiminori. a Year Seeding time Cultivar and strain Culm length (cm) 88.0 74.8** 95.3 84.2** 62.2 52.6** Panicle length (cm) 24.1 21.6* 21.8 18.8* 14.0 12.8 Panicle densityb 1.58 1.11* 1.24 1.01** 1.10 1.09 Spikelets (no.) 158.4 58.6** 111.5 68.1** 47.7 39.9 Fertility (%) 96.0 66.2** 89.5 52.1** 87.6 34.5** Primary branch (no.) 10.8 8.3** 10.5 7.2** 5.8 6.2 Primary branch length (cm) 8.0 7.3 6.3 6.8 6.0 4.7* Secondary branches (no.) 32.8 16.5** 18.3 11.3** 6.6 5.1 Rachis length (cm) 18.7 17.2 16.6 12.7** 8.6 8.7

1992 1994 1994

Normal seeding Early seeding Late seeding

Fujiminori Dwarf mutant Fujiminori Dwarf mutant Fujiminori Dwarf mutant

a * and ** = significant at the 1 and 0.1% level, respectively. b Panicle density = no. of spikelets/total length of rachis and primary branches (in cm) (Futsuhara et al 1979).

cultivar under all three conditions, was further reduced under late seeding condition (during an abnormally hot summer). Premature heading occurred in all the mutant plants grown under late seeding condition. Most of the panicles with premature heading showed malformed floral organs such as degenerated ovary,

multi-ovary, and long empty glumes. Late sowing also promoted premature heading. The results indicate that heat stress tolerance was depressed and premature heading was enhanced in the dwarf mutant. Induced dwarf mutants with simultaneous changes in many traits are being used

in developmental genetic research to understand pleiotropic gene action.

Cited references
Futsuhara Y, Kondo S, Kitano H. 1979. Genetical studies on dense and lax panicles in rice. I. Character expression and mode of inheritance of lax paniclerice. Jpn. J. Breed. 29:151-158.

Genetic differentiation between Japanese lowland and upland rices

R. Ishikawa, Y. Harata, T. Harada, and M. Niizeki, Plant Breeding Laboratory, Faculty of Agriculture, Hirosaki University, Hirosaki 036, Japan

Upland and lowland varieties are cultivated under different conditions in Japan. Lowland rice with useful agronomic traits is a major crop. In contrast, upland rice. useful as a genetic source of stress resistance to blast, is regarded as a minor crop. Thus, diagnostic genetic markers to evaluate upland rice will be useful to help improve resistance in modern rice varieties. Isozyme genotypes were surveyed among Japanese lowland and upland traditional cultivars. Seventeen isozyme loci were monitored, and different genotypes were scored. A total of 27 genotypes were recognized, 16 for lowland rice and 13 for upland cultivars. Genotypic diversity for upland cultivars was 0.072, which is surprisingly higher than that of lowland cultivars (0.020). Such high genotypic diversity could not be seen from the data of Glaszmann (1987). Four hundred and fifty lowland and 200 upland cultivars were classified either as japonica or indica using discriminant score based on 11 loci (Ishikawa et al 1991, Sano and Morishima

1991). Except for five lowland and five upland cultivars classified as indicas, the rest of the cultivars were japonicas (Table 1 ). The 10 indicas were characterized with allele 3 for Pgd1, which is not common among other cultivars in Japan. Characteristics of the indica cultivars, such as short apiculus hair (APH) averaging 0.36 0.11 mm, slender hull type represented by hull
Table 1. Genotypic variation for Pgd among Japanese lowland and upland rices. Varietal type Lowland Indica Japonica Upland Indica Japonica Pgd1-1 Pgd1-2 Pgd1-3

0 445 1 26

0 0 0 169

5 0 4 0

length-width ratio (L/W; av of 2.86 0.14), and positive phenol reaction (Table 2), however, seemed to point to the cultivars common origin. Allele 2 for Pgd1, which is never observed in lowland rices, is also a remarkable diagnostic allele for specifying upland varietal groups. Upland cultivars can be classified into three groups based on Pgd and other trait combinations. Group 1 : characterized by the presence of allele 1 and represents longer APH (0.65 0.12 mm) and round hull type (L/W; 2.14 0.16) same as 445 lowland cultivars (APH; 0.72 0.19 mm, L/W; 2.09 0.34). However, only 19% of cultivars in this group showed positive phenol reaction. Group 2: characterized by the presence of allele 2 and represents slightly shorter

Table 2. Morphological and physiological characteristics of Japanese lowland and upland varietal groups having different alleles for Pgd. Group Allele Cultivars (no.) Apiculus hair length L/W Mean SD (mrn) Mean SD (mm) Phenol reactiona Cultivars (no.) +

Lowland Japonica Indica b Upland Japonica A Japonica B Indica b

Pgd1-1 Pgd1-3 Pgd1-1 Pgd1-2 Pgd1-3

445 5 26 169 5

0.72 0.37 0.65 0.44 0.34

0.19 0.12 0.12 0.14 0.13

2.09 2.79 2.14 2.38 2.98

0.34 0.16 0.16 0.21 0.19

32 4 5 131 4

413 1 21 38 1

a + = hull of cultivars blackened. - = hull did not blacken. b Total of 10 indica cultivars were found. Av APH and L-W ratios

were 0.96 0.11 and 2.8 0.14, respectively.


IRRN 21:2-3 (August-December 1996)

APH (0.44 0.14 mm) and more slender hull type (L/Wp; 2.38 0.21) than the first group. Group 3: possesses mainly allele 3 belonging to the indica type mentioned earlier. Pgd1 is located on chromosome 11 along with la, v4, and Adh1 (Ishikawa et al 1991). Several key resistance genes have already been located on this chromosomal area (Goto 1970), which seemed to be a specific region for upland characters. Landmarks on chromosome 11 were used to obtain restriction fragment length polymorphism (RFLP) markers for constructing a detailed map around Pgd1. Sensho and Taichung 65 (Acc. 504) were used to ensure polymorphic regions on chromosome 11. Eleven probes and seven restriction enzymes were used; four of these probes showed polymorphism with some enzymes. Among the probes, adh1, G181, and G189 showed allelic band

trends specific to the upland group. The trend of RFLP patterns, however, was less specific than that of Pgd. These RFLP markers and Pgd are used to arrange several agronomic and stress resistance traits on the chromosome. Recent plant breeding programs have failed to introduce upland-specific resistance genes into lowland modem varieties. This implies the presence of linkage blocks for useful stress resistance traits of upland cultivars, and also of existing undesirable agronomic traits in lowland varieties. Confirming the chromosomal region where upland-specific markers are located would increase understanding of the origin of Japanese upland rices. Also. detecting RFLP between upland and lowland cultivars will help rice breeders to introduce only useful genetic resources from upland rice into modem varieties.

Cited references

Glaszmann JC. 1987. Isozyme and classification of Asian rice varieties. Theor. Appl. Genet. 74:21-30. Goto I. 1970. Genetic studies on the resistance of rice plant to the blast fungus. I. Inheritance of resistance in crosses Sensho x H79 and Imochi-Shiraru X H79. Ann. Phytopathol. Soc. Jpn. 36:304-312. Ishikawa R, Maeda K, Harada T, Niizeki M, Saito K. 1991. Classification of Japanese rice varieties into Indica and Japonica types by using isozyme genotypes. Jpn. J. Breed. 41:605-622. Ishikawa R, Morishima H, Kinoshita T, Harada T, Niizeki M, Saito K. 1991. Linkage analysis of nine isozyme genes on the conventional linkage map in rice. Jpn. J. Breed. 41:265-272. Sano R, Morishima H. 1992. Indica-Japonica differentiation of rice cultivars viewed from the variations in key characters and isozymes, with special reference to landraces from the Himalayan hilly areas. Theor. Appl. Genet. 84:266-274.

Evolutionary variations in the Gramineae: rearrangements of DNA fragments transferred from chloroplast genomes to mitochondrial genomes
A. Kanno, Institute of Genetic Ecology, Tohoku University, Sendai 980-77, Japan; M. Nakazono, N. Watanabe, N. Tsutsumi, Faculty of Agriculture, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113, Japan; T. Kameya, Institute of Genetic Ecology, Tohoku University; and A. Hirai, Faculty of Agriculture, The University of Tokyo, Japan

The transfer of DNA fragments from chloroplast to mitochondrial genomes is considered a general phenomenon in higher plants (Stem and Lonsdale 1982, Stern and Palmer 1984). For rice, Nakazono and Hirai (1993) have determined the entire set of transferred chloroplast sequences present in the mitochondrial genome. Total length of the transferred chloroplast sequences, whose lengths range from 32 bp to 6.8 kb. accounted for about 6% of the rice mitochondrial genome. Two of the 16 identified ct-derived fragments in rice mitochondrial genome, one comprising

rps 19 - trnH - rp12 - rp123 and the other comprising y rp123-rbcL-atpB-atpE-trnMtrnV, were found to be separated in the chloroplast genome but were joined in the mitochondrial genome, which might be the cause for homologous recombination between rp123 and rp123. Maize mitochondrial genome also contains these gene clusters (Lonsdale et al 1983). Watanabe et al (1994) compared this region, which is homologous to chloroplast rps l9 in the mitochondrial genomes, among five gramineous plants (rice, maize, sorghum, Italian rye grass, and wheat) by Southern blot hybridization, polymerase chain reaction (PCR), and DNA sequencing techniques. In all the mitochondrial DNAs from the five gramineous plants except for that from wheat the ct-derived fragments ofchloroplast DNA were found to be maintained and the same junctions of mitochondrion-specific and chloroplastlike sequences were found at one terminus (rps 19 side). Subsequent analysis revealed that the fragments had been variously rearranged among species with respect to the other terminus. For this region, however, rice rp123 includes a 135-bp deletion in both chloroplast and mitochondrial genomes. In contrast, neither gene in maize has such a deletion. This

suggests that this region must have been transferred separately in rice and maize after their divergence from one another, with subsequent homologous recombination between rp123 and Yrp123 in the mitochondrial genomes of rice and maize occurring independently (Fig. 1). These findings indicate that the transfer of the choloroplast sequence occurred in the distant past during the evolution of gramineous plants. Maintenance of a common junction on one side (the rps19 side), despite the extensive rearrangements of mitochondrial genome, points to the possibility that these sequences might function in the mitochondria. For the result of reverse transcriptase(RT)-PCR and Northern blot hybridization in this study, the chloroplastderived trnH gene was expressed in rice mitochondria (Fig. 2). Comparisons of such ct-derived fragments among or within species may provide some interesting information about the evolution of mitochondrial genomes.

Cited references
Stem DB, Lonsdale DM. 1992. Mitochondrial and chloroplast genomes of maize have a 12kilobase DNA sequence in common. Nature 299:698-702.

IRRN 21:2-3 (August-December 1996)


1. Schematic representation of possible transfer and homologous recombination of ct-derived DNA fragments. Event I (transfer of DNA fragment from chloroplast to mitochondrial genome) occurred before the divergence of rice and maize. Events II (secondary DNA transfer from chloroplast to mitochondrial genome) and III (homologous recombination) occurred after the divergence of rice and maize.

Stern DB, Palmer JD. 1984. Extensive and widespread homologies between mitochondrial DNA and chloroplast DNA in plants. Proc. Natl. Acad. Sci. USA 81:19461950. Nakazono M, Hirai A. 1993. Identification of the entire set of transferred chloroplast DNA sequences in the mitochondrial genome on rice. Mol. Gen. Genet. 236:341-346. Watanabe N, Nakazono M, Kanno A, Tsutsumi N, Hirai A. 1994. Evolutionary variations in DNA sequences transferred from chloroplast genomes to mitochondrial genomes in the Gramineae. Curr. Genet. 26:512-5 18. Lonsdale DM, Hodge TP, Howe CJ, Stern DB. 1983. Maize mitochondrial of DNA contains a sequence homologous to the ribulose-1, 5bisphosphate carboxylase large subunit gene ofchloroplast DNA. Cell 34:1007-1014.

2. Gene expression of the ct-derived DNA fragment in mitochondrial genome. (a) Positions of primers for PCR experiment are indicated by P1 and P2. (b) RT-PCR experiment. Total RNA was prepared from rice green leaves and cDNA were synthesized (or not synthesized) with reverse transcriptase (RT+/RT-) using P2 primer. M = molecular weight marker of FX174 DNA digested with Hae lll. (c) Northern hybridization was carried out using nonradioactively labeled probes, indicated under each Northern hybridization pattern. Oligo-trnfM and B9 Kpnl 1.8 kb are specific probes for mitochondrial and chloroplast tRNA, respectively. Lane 1 = rice mtRNA; lane 2 = total RNA extracted from green leaves; lane 3 q total RNA extracted from etiolated leaves.


IRRN 21:2-3 (August-December 1996)

Identifying transposonlike element Tnr2 in rice

Y. lida, H. Ohtsubo, and E. Ohtsubo, Institute of Molecular and Cellular Biosciences, The University of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo 113, Japan

Tnr2 was found as an insertion sequence within p-SINEl-r4, a member of the retroposon p-SINE1, in rice. This sequence is 157 bp in length and present in multiple copies in the rice genome. Tnr2 has the 56bp terminal inverted repeat sequence (TIR) and is flanked by an 8-bp direct repeat sequence. Tnr2 shows no homology with known transposable elements, suggesting that Tnr2 is a new transposable element (Fig. 1) (Mochizuki et al 1993, Ohtsubo et al 1993).

To analyze the inner region of Tnr2, we synthesized primers that hybridize the sequences of inverted repeats and carried out polymerase chain reaction (PCR) using genomic DNA of Oryza sativa L. cultivars IR36 (indica) and Nipponbare (japonica) as the template. The PCR-amplified fragments were cloned, and their nucleotide sequences determined. The Tnr2 members contained substitutions and deletions in their sequences. A consensus sequence derived from the nucleotide sequences of these members showed a 20-bp tandem repeat in the region between the TIR sequences (Fig. 2). A few members, however, had only one of the repeating sequences. We identified three more Tnr2 members with flanking sequences by screening from the genomic library of IR36. Nucleotide sequencing of these members revealed that

1. The stem-loop structure of Tnr2 element, 157 bp in length. Arrows indicate TSD of 8 bp.

2. Nucleotide sequences of Tnr2 members. The TIR sequences (56 bp) are underlined. - - = bases identical to those of consensus sequence of Tnr2, / = bases deleted, * = primer sequences used for PCR. Tandem repeat sequences of 20 bp are double-underlined. Clones pYK2A and pYK4 lack one of the repeating units.

IRRN 21:2-3 (August-December 1996)


each member had target site duplication (TSD) of 8 bp, confirming that Tnr2 is a transposable element (Fig. 2). PCR using oligonucleotides derived from the Tnr2 flanking sequences as primers revealed that several accessions of AA genome species O. longistaminata and O. glaberrima also had Tnr2 in the corresponding loci. This indicates that Tnr2 was inserted into each locus before the rice varieties with AA genome had differentiated during evolution. The PCR-amplified fragments,

however, showed length polymorphism caused by deletion within the Tnr2 sequence or in the region flanking Tnr2. All theTnr2 members identified in the study are small and are assumed to be a nonautonomous element derived by deletion from an autonomous one that can transpose by itself. Several hundred copies of Tnr2 exist in the rice genome, one of which may correspond to the autonomous Tnr2.

Cited references
Ohtsubo E, Mochizuki K, Tenzen T, Ohtsubo H. 1993. Gamma Field Symp. 32:71-83. Mochizuki K, Ohtsubo H, Hirani H, Sano Y, Ohtsubo E. 1993. Classification and relationship of rice strains with AA genome by identification of transposable elements at nine loci. Jpn. J. Genet. 68:205-217.

Breeding methods
New cytoplasmic male sterile lines developed in Andhra Pradesh, India
R. V. Kumar, P. V. Satyanarayana, and M. S. Rao, Agricultural Research Station (ARS), Maruteru 534122, Andhra Pradesh, India

Successful use of hybrid vigor in rice largely depends on availability of local cytoplasmic male sterile (CMS) and restorer lines. In India, many IRRI-bred CMS lines from the wild abortive (WA) source are being used to develop rice hybrids, which could lead to genetic vulnerability. Use of local CMS lines will help to alleviate this problem and to develop adaptable, heterotic hybrids. We therefore screened several hundred elite genotypes with diverse genetic backgrounds for their maintaining ability at ARS, Maruteru, India. Two effective maintainers for the WA cytoplasmic source and one effective maintainer for the ARC cytoplasmic source were identified and successfully converted into local CMS lines through backcrossing. Lines with complete pollen sterility identified from BC6 generation were

designated as APMS1 A (IR54755 A/WGL 3010-407 - ARC source), APMS2 A (V20 A/WGL 3010-407 - WA source) and APMS5A(IR58025 A/MTU4870 - WA source) during 1994-95. These lines, along with IRRI-bred CMS line IR58025 A, were evaluated for their agronomic and floral traits and natural outcrossing potential. The experiment was laid out in a randomized block design with five replications during the 1995 wet season. Plants were raised at 20- 20-cm spacing in a 4-m2 plot. Five plants from the central row in each replication were observed. To study their natural outcrossing potential, the CMS lines were planted adjacent to the corresponding maintainer lines. Good flowering synchrony was achieved. Seed set was attained without resorting to any supplementary pollination techniques. Data on six characters were statistically analyzed. The four CMS lines significantly differed in agronomic and floral traits, except for productive tillers plant -1 (see table). Early-duration (110-115 d) APMS1 A (ARC) and APMS2 A (WA) possess field resistance to gall midge and bacterial leaf

blight (BLB). Their anthers are white and shivered with moderately well-exserted stigmas. These lines may be extremely valuable in developing early to super-early rice hybrids adaptable to different cropping systems. Long-duration (147 d) APMS5 A has sturdy culms and possesses resistance to brown planthopper, BLB, and rice tungro virus. Spikelet opening angle, stigma exsertion, and natural outcrossing potential are comparable with those of popular CMS line IR58025A. APMS5 A is the first longduration CMS line with desirable floral and agronomic traits developed in India. This line will be useful in developing longduration rice hybrids suitable for wetseason cultivation in coastal Andhra Pradesh. We used the three CMS lines for testcrossing with several elite lines, and effective restorers and maintainer were identified. The three CMS lines are potential female parents for developing heterotic rice hybrids adapted to different ecological conditions, including coastal regions of Andhra Pradesh, India.

Agronomic and floral traits of new CMS lines in rice. CMS line APMS1 A APMS2 A APMS5 A lR58025 A SEM CD (0.05) CV (%) Days to 50% flowering 78.0 80.0 116.7 90.5 0.57 1.75 1.40 Plant height (cm) 86.3 87.5 115.5 90.8 0.63 1.94 1.30 Productive tillers/plant (no.) 10.3 10.1 9.8 10.1 0.23 ns 5.00 Duration of spikelet opening (min) 192.5 183.5 194.0 222.5 4.36 13.39 4.90 Angle of spikelet opening () 26.8 27.0 31.0 30.0 0.21 0.63 1.60 Seed set on natural outcrossing (%) 10.0 10.8 12.1 12.6 0.3 0.93 6.00 Pollen sterility (%) 100 100 100 100 Moderately well exserted Moderately well exserted Well exserted Well exserted White and shriveled White and shriveled White and shriveled White and shriveled Stigma exsertion Anther characteristic


IRRN 21:2-3 (August-December 1996)

Maintainers and restorers identified in some rice cultivars of Pakistan

S. S. Ali and M. G. Khan, Rice Research Institute (RRI), Kala Shah Kaku, Lahore, Pakistan

Maintainer and restorer lines identified for IRRI CMS lines at Kala Shah Kaku, Pakistan. 1994 95. Spikelet fertility of hybrids witha lR58025 A R B PR B R PR B PR PR PR B PR B B B B PR PR B B PR PR B B B B R B B B B B B B B B B PR lR62829 A R B PR B PR PR B PR PR PR B PR B B B B PR PR B B PR PR B B B B R B B B B B B B B B B PR

Development of rice cytoplasmic male sterile line 47456 A in Kala Shah Kaku, Pakistan
S. S. Ali and M. G. Khan, Rice Research Institute, Kala Shah Kaku, Pakistan

Pollen parent

We crossed 38 aromatic and nonaromatic local rice varieties and lines with IR58025 A and IR62829 A wild abortive (WA) cytoplasmic male sterile (CMS) lines from IRRI during 1994 kharif season. Seventysix F1 hybrids, their male parents, and two isogenic maintainers were transplanted in rows of 30 plants with 23-cm spacing on each side. The crop received 120 kg N ha-1 and 60 kg P ha-1. Standard agronomic and plant protection measures were used. Ten plants from each hybrid were labeled. Three panicles from each of these plants were marked and bagged before flowering. Spikelet fertility was calculated as a percentage of filled grains. Five spikelets from the upper part of each panicle were collected before anthesis and fixed in 70% alcohol. Two to three anthers from each spikelet were placed onto a glass slide, squashed in 1% IKI solution, and screened for sterile and fertile pollens. Fertile pollens, which became deeply stained, were round and fully developed. Male parents of the hybrids with 100% pollen and spikelet sterility were classified as potential maintainers, those with 80-100% pollen and spikelet fertility as restorers, and the rest as partial restorers. Frequency of maintainers (63%) was much higher than that of restorers among 76 hybrids tested (see table). 1021-8 and KS282 were the only potential restorers for both CMS lines, whereas Basmati 385 restored fertility only in IR58025 A (>80%), indicating that marked cytoplasmic nuclear interaction exists and expression of restorer genes varies with the genetic background of

KS282 IR6 49744 4048-3 Basmati 385 4029-A 4029-B 4439 50189-8-6 40555 GP-6 GP-10 GP-15 GP-16 GP-43 4029-2 4029-3 Basmati 5854 47456 PK3732-28 C622 Jhona 349 PK3717-12 PK3727-2 DR82 Basmati 370 1021-8 4289 4334 4365 PK3355-5-1-4 PK3303-7-2 49818 33608 50020 50021 Basmati 198 Basmati 6129

R = restorer (80-100% pollen and spikelet fertility). PR = partial restorer (5-79% pollen and spikelet fertility). B = maintainer (100% pollen and spikelet sterility).

female parents. Apparently, restorers among Pakistani rice cultivars are scarce. Maintainer lines identified are being used in a backcrossing program for inducing CMS in local germplasm. Restorer lines 1021-8, KS282, and Basmati 385 will be used to develop new hybrid combinations.

Hybrid rice research was initiated at Kala Shah Kaku in 1991. We crossed IRRI cytoplasmic male sterile (CMS) lines IR58025 A and IR62829 A, with wild abortive (WA) cytoplasm, with aromatic rice breeding line 47456 in 1992. 47456 [4048-3/PK4112 (Basmati 370/4439)] has extra-long grains and short stature, and flowers early. Plants in the F 1 were raised in 1993. Crosses of IR58025 A and IR62829 A with 47456 showed complete pollen sterility as well as spikelet sterility under bagged conditions (see table). Backcrosses were made with recurrent parent 47456. The derived progenies (F1, BC1, and BC2) and the parents were transplanted in the field in rows of 30 plants at 23-cm spacing from each side in Jul 1995. The crop received 120 kg N ha -1 and 60 kg P ha-1. Standard agronomic and plant protection measures were followed. Ten plants from each generation were selected, with three panicles from each plant bagged before flowering. Spikelet fertility was calculated as the percentage of filled grains. Five spikelets from the upper part of each panicle were collected before anthesis and fixed in 70% alcohol. Two to three anthers from each spikelet were squashed in 1% IKI solution to determine pollen sterility. Complete pollen sterility was observed in the F1 and backcross generations, indicating successful transfer of CMS from IR58025 A and IR62829 A into Basmati line 47456. Restorers for the Basmati CMS line 47456 A are being identified.

IRRN 21:2-3 (August-December 1996)


Pollen and spikelet sterility of parents, F 1, BC 1, and BC 2 generations at Kala Shah Kaku, Pakistan. 1995. Plant height (cm) Parent lR58025 A lR62829 A 47456 F1 lR58025 A/47456 lR62829 A/47456 BC1 lR58025 A/(47456) a IR62829 A/(47456) BC2 lR58025 A/(47456) b lR62829 A/(47456)

Days to 50% flowering 84 83 85

Pollen sterility (%) 100 100 10

Spikelet sterility on bagging (%) 100 100 13.5 100 100

97.1 85.5 90.1

99.1 92.2

92 90

100 100

99.3 94.4 99.5 94.6

95 90 90 89

100 100 100 100

100 100 100 100

= generation 1.

= generation 2.

Comparison of promoters and selectable marker genes for indica rice transformation
Z. Li, N. M. Upadhyaya, and P. M. Waterhouse, Commonwealth Scientific and Industrial Research Organization (CSIRO) Division of Plant Industry, GPO Box 1600, Canberra, ACT 2601, Australia; and A. J. Gibbs, Research School of Biological Sciences, Australia National University, Canberra, ACT, Australia

Several selectable marker genes (nptII, hptII, dhfr; bar; and csr1) have been used for cereal transformation. However, the bar gene and, to a lesser degree, the hpt gene, have been the most commonly used in recent years. A comparison of the efficacy of these selectable marker genes in indica rice has not been previously reported. Transient expression of GUS in electroporated protoplasts and bombarded suspension cells. Plasmids with the gus gene under the control of promoters pCaMV35S-gus, pEmu-gus, pRolC-gus, and pActl-gus were electroporated into protoplasts. Similarly, pCaMV35S-gus, pEmu-gus, pActl-gus, and pUbi1-gus were used to bombard suspension cells. Except for the RolC promoter, which directs vascular-specific expression, GUS expression was detectable under the control of all promoters tested. Constitutive promoters Act1 and Ubi1 gave the highest expression levels. The order of promoter

strength (from strong to weak) in nonorganized cells of indica rice Chinsurah Boro II was Ubi1 > Act1 > Emu > CaMV35S>RolC. The Ubil, Actl, and Emu gave 12-, 8-, or 2-fold higher expression levels, respectively, compared with the CamV35S promoter. Comparison of hpt and bar genes driven by various promoters for stable transformation. Embryogenic calli were cobombarded with pUbi1-gus and a range of plasmids containing either the hph or bar gene controlled by various promoters. Calli staining for GUS activity 1 d after shooting produced more than 300 blue expression units (BEUs) per petri dish of calli bombarded with each plasmid. Although most of the BEUs diminished or disappeared, a few from the pUbi1-gus-bombarded calli and any of the three hph -containing plasmids gradually increased in size over 2 wk of selection. Four weeks after selection. they formed nodular structures from the surface of the original calli. Cobombardment of pUbi1-gus and pUbil-hph gave the most large BEUs after 4 wk, pEmu-hph produced a similar number of smaller BEUs, and pCaMV35Shph gave the fewest and smallest BEUs. Some cultures cobombarded with gus and bar constructs had modest increase in size and number of BEUs. None of the BEUs developed nodular structures. Selection of stably transformed cell lines and regeneration of transgenic plants. Embryogenic calli were bombarded with

the hph gene constructs (pEmu-hph and pUbi1-hph) either alone or combined with the gus constructs (p40CSD35SAcl-gus and pUbi1-gus). With the cobombarded calli, about 300-1,500 BEUs were obtained 1 d after shooting. Without staining for GUS activity, resulting transformed cells or microcalli cannot be identified under the dissecting microscope for at least 1 mo after bombardment. However, staining a few plates of target calli for GUS activity after 2 wk of selection allowed us to observe transformed cell clusters. After three to four subcultures on selection media, fast-growing callus lines could be distinguished among the bombarded calli. These callus lines were individually subcultured onto a new selection medium for further growth. From 10 experiments and after 2 mo of selection on media, 50 transformed callus lines were obtained from 72 bombarded plates (see table). Of these, 16 regenerated into 79 putative transgenic plants (about 30% regeneration frequency) when placed onto the regeneration medium. Of the six putative transgenic plant lines from cotransformation of gus and hph genes, two were GUS-positive when the plants were about 5 cm tall. However, GUS activity of one line ceased as the plants matured. All 79 putative transgenic plants (24 sterile and 55 fertile) were transferred to soil in the glasshouse. Most of the sterile plants were derived from experiment 1, which used foundation callus line 1. About 87% of the transgenic plants derived from foundation callus lines 2 and 3 were fertile. Fertile plants were derived from 12 different transformed callus lines (6 from pUbil-hph and 6 from pEmu-hph). Southern blot analysis showed that at least one plant from each line contained between one and eight copies of the transgene. Ten of the 12 transgenic plant lines contained both selectable and nonselectable genes, which were introduced on separate plasmids. giving an 83% cointegration frequency. Calli bombarded with bar constructs grew poorly on bialaphos-containing media irrespective of the promoter used to control the bar gene. The few BEUs that expanded over time failed to form nodular cell clusters. This suggests the limited use of the bar gene in indica rice transformation. Rapid growth of hygromycin-resistant calli


IRRN 21:2-3 (August-December 1996)

Summary of stable transformation of indica rice cv Chinsurah Boro II. Experiment Callus age (mo)/ callus line 8.5/1 9/1 7.5/2 8/2 7.5/2 8/2 8.3/2 8.3/2 5.3/3 5.3/3 Plates bombarded (no.) 8 4 3 12 7 7 8 9 7 7 72 hphresistant callus lines 13 5 3 5 7 5 1 2 3 6 50 GUS+ callus lines (no.) 10 4 1 2 17 Embryogenic lines (no.) 3 0 2 1 3 1 1 1 3 1 16 Plants (no.) GUS+ plant lines (no.) 1 (1) a 0 0 1 Fertile plants (no.) 0 7 1 22 4 1 1 13 6 55 Selectable marker gene pEmu- hph pEmu- hph pUbi1- hph pUbi1- hph pEmu- hph pEmu- hph pEmu- hph pEmu- hph pUbi1- hph pUbi1- hph Nonselectable gene

1 2 3 4 5 6 7 8 9 10 Total

16 8 2 23 4 1 2 17 6 79

p40CSD35S- gus p40CSD35S- gus pUbi1- gus pUbi1- gus RRSV b RRSV RRSV RRSV RRSV RRSV

a Number in parentheses shows transgenic plant line expressing GUS when about 5 cm tall, but not at 20 cm tall. b Coding region of rice ragged stunt virus segment 5 drlven by various promoters.

and BEU expansion were observed when the hph gene was driven by the Ubi 1 promoter, followed by Emu and then CaMV35S. In all three cases, expanding BEUs formed nodular structures indicative of regeneration. The Ubi 1>Emu>CaMV35S ranking

follows the relative levels of GUS expression driven by these promoters in our transient and stable expression studies. From this we infer that stronger promoters enable the most effective selection. The data show that about 2,000-4,000 BEUs 1 d

after shooting are required to achieve each hygromycin-resistant callus line and that such BEUs regenerate into fertile transgenic rice plants with 1-8 copies of the transgene.

Transfer of cytoplasmic male sterility in indica rice through protoplast fusion

H. S. Gupta, B. Bhattacharjee, and A. Pattanayak, Plant Breeding Division, Indian Council for Agricultural Research (ICAR) Research Complex for NEH Region, Umroi Road, Barapani 793103, Meghalaya, India

Cytoplasmic male sterility (CMS) is widely used in hybrid seed production and requires routine transfer of CMS to new genetic backgrounds. Transfer of CMS through conventional breeding requires five to six backcrossings, taking 2-3 yr. However, a one-step transfer of CMS through asymmetric protoplast fusion can be used to shorten this period (Kyozuka et al 1989). Plant regeneraion, from protoplastsof CMS, maintainer, and restorer lines. Embryogenic calli initiated from the scutella obtained from mature embryos of CMS line V20A were repeatedly subcultured to make them friable. Five hundred 1,000-mg calli were used to initiate cell suspension, which was established after 3-4 mo of regular subculturing at a 4-5 d

interval. Protoplasts, which were isolated, purified, and cultured as described by Bhattacharjee and Gupta (1995), underwent sustained division into to microcolonies that became macroscopic after 30-40 d. Protocalli differentiation was obtained 20-30 d after transfer to regeneration medium. Plants were transferred to pots where they flowered and produced sterile pollen grains. Likewise, protocols for protoplasts to plant systems were developed for RCPL1-2C, V20 B, and IR36. Inactivation of donor and recipient protoplasts and asymmetric fusion. Because protoplasts of all four lines were dividing on culture, we standardized the dose for their inactivation to undertake asymmetric fusion. Protoplasts of V20A were irradiated with 30 krad gamma ray (Fig. 1a); those of RCPL1-2C, V20 B, and IR36 were inactivated with 10 mM iodoacetamide for 15 min at 20C (Fig. 1b). Fusion of gamma ray- and iodoacetamide-inactivated protoplasts of V20A with RCPL1-2C, V20 B, and IR36 was attained with 25 V/1 KHz AC amplitude for 25 s, followed by 900 V DC pulse (2.52 KV/cm field strength) amplitude for 30 s. Seven 10% fusions were obtained.

Culture of fusion products and plant regeneration, from putative cybrid calli. Fused protoplasts of V20 A with RCPL12C, V20 B, and IR36, along with inactivated protoplasts of V20 A, V20 B, RCPL1-2C, and lR36, were cultured separately. The inactivated protoplasts served as the control. Only fused protoplasts underwent sustained division and formed microcolonies (Fig. 1c), which became macroscopic after 25-30 d of culture. Macrocolonies of putative cybrids between V20A and IR36 differentiated into plantlets when transferred to regeneration medium (Fig. 1 d). The plantlets transferred to pots (Fig. 1 e) grew to maturity and, as expected. were fertile. Putative cybrid calli between V20 A/RCPL1-2C and V20A/V20 B also differentiated and gave rise to plantlets, which were grown to maturity to check the expression of male sterility. Molecular analyses of mitochondrial DNA of both the sets and parents are in progress.

Cited reference
Bhattacharjee B, Gupta HS. 1995. Fertile plants regenerated from protoplasts of cold-tolerant rice line RCPL1-2C. J. Biochem. Biotechnol. 4:61-65.

IRRN 21:2-3 (August-December 1996)


variation in ploidy, and agronomic traits. Promising PGMS line Zhenongdal 1s (X126- 11) has already been developed through protoplast culture. A cell suspension with small globular calli was established from the scutellum of mature embryos of PGMS cultivar N5047s. A high density (1.5 106 ml-1) of protoplasts was obtained, and green regenerates (plantlets) were produced following Xue and Earles (1995) method. To distinguish autotetraploids from diploids, the DNA content of regenerates was measured using flow cytometry according to the procedure of Arumugamuthan and Earle (1991). Sowing at different times under natural conditions at Hangzhou (30 15' N) promoted changes in fertility and agronomic traits of the four diploid PGMS protoplast clones and their control cultivar N5047s. Variation in ploidy. Flow cytometric analysis indicated that nuclear DNA content in protoplasts of diploid plants ranged between 0.715 and 0.873 pg 2c-1, and that from autotetraploids ranged between 1.537 . and 1.786 pg 2c -1 This accounted for 60%. diploid plants and 40% autotetraploid plants. Diploid regenerates were taller (78.0 cm) compared with the autotetra-

ploids (44.5 cm). The diploid plants had 96.5 spikelets per panicle whereas autotetraploids had few spikelets (19.3). Change in fertility. Typical sterile pollen was >85% at the sterile stage and ranged from 10 to 30% at the fertile stage. However, the spherical sterile pollen ranged from 5 to 20% at the two stages. Percentage of staining pollen exhibited one or three peaks at the fertile stage. Four PGMS protoplast clones had obvious differences both in duration of sterility and in stability before mid-August. The change from sterility to fertility was clear. Seed set was more than 30% under shorter daylength in September. Agronomic traits. Agronomic characters significantly differed among the four diploid protoplast clones and control cultivar N5047s. Spikelets panicle -1 , panicle number plant-1, spikelet density, and 100-grain weight of X126-11 were significantly higher than those of other protoplast clone and NS047s. Photoperiod and temperature responses. We used computer simulation analysis of PGMS protoplast clones to measure photoperiod-sensitive and temperature-sensitive coefficients for defining their model of seed set rate. The

Protoplast fusion between V20 A and IR36 and regeneration of putative cybrid. (a) gammairradiated protoplasts of V20 A; (b) iodoacetamide-treated protoplasts of IR36; (c) microcolonies resulting from sustained division of the fusion product; (d) plantlets regenerated from putative cybrid calli; and (e) putative cybrid plants.

Characters of plants regene rated from protoplasts of photoperiod-sensitive genic male sterile rice
Qingzhong Xue and K. Etoh, Agronomy Department, Zhejiang Agricultural University, Hangzhou 310029, People's Republic of China; and S. McCouch and E. D. Earle, Plant Breeding and Biometry Department, Cornell University, Ithaca, NY 14853-1902, USA

Identifying photoperiod-sensitive genic male sterility (PGMS) in rice (Shi 1981) makes it possible to produce hybrid rice seed through the two-line system. We studied the response of PGMS protoplast clones to photoperiod and temperature,

Comparison of the anatomical structure of anther connective tissue and thrum at the fertile and sterile stages X126-11. (a) Fertile anther connective tissue, (b) sterile anther connective tissue, (c) fertile thrum, and (d) sterile thrum. NVB = normal vascular bundle. AVB = abnormal vascular bundle. V = vascular bundle. S = sieve tubes.


IRRN 21:2-3 (August-December 1996)

analysis showed that photoperiod and temperature can affect fertility of PGMS protoplast clones. Anatomical differences between and fertile periods in PGMS rice. The pollen sac and thrum (or filament) at the sterile stage were smaller than those at the fertile stage. Poorly developed vessel elements and sieve tubes of the anther vascular bundle and thrum vascular bundle (see figure) are abnormalities that can affect the transport of nutrients to the pollen sac, leading to microspore sterility.

Arumugamuthan K, Earle ED. 1991. Estimate of nuclear DNA content of plants by flow cytometry. Plant Mol. Biol. Rep. 9:229-231. Shi MS. 1981. Preliminary research report on breeding and utilization of natural two-uses line in late Japonica rice. Sci. Agric. Hubei 7:1-3. Xue QZ, Earle ED. 1995. Plant regeneration from protoplasts of cytoplasmic male sterile lines of rice ( Oryza sativa L.). Plant Cell Rep. 15:76-81.

Cited references

Attempted hybridization between Oryza sativa L. and Porteresia coarctata T.

Z. I. Seraj, M. O. Faruque, Biochemistry Department, University of Dhaka (UD), Dhaka 1000, Bangladesh; K. G. Hossain, Institute of Post Graduate Studies in Agriculture, Salna, Gazipur, Bangladesh; R. H. Sarker, Botany Department, UD; T. Devi, Z. Islam, Biochemistry Department, UD; and A. S. Islam, Botany Department, UD

(4x)/O. sativa (2n) and tetraploid O. sativa (4x-c)/ P. coarctata. Hand pollinations were made between the two genera. Gibberellic acid (75 ppm) with 0.1 % Tween was sprayed once 20 min after crossing. Embryos were rescued 1012 d after pollination and cultured on 1/4 Murashige and Skoog (MS) medium with 0.45% agarose. Hybrids showing slow growth were kept on 1/4 MS for 8-10 wk. Plants were then transferred to the soil. Rapidly dividing root tips were stained with Feulgen. Seed set percentage and survival among crosses were recorded (see table) The tetraploid hybrid from the cross BR-7 (4xc)/ P. coarctata grew vigorously and was taller than the induced tetraploid mother variety BR-7 (4x-c). Leaf texture was rough or intermediate between that of the parents. Leaves of the hybrid taper gradually compared with the abrupt taper of BR-7 (4x-c) leaves but not as gradually as those of P. coarctata. The hybrid did not set any seed on selfing. It has spikelets showing short awn and the curled morphology and asynchronized flowering pattern of P. coarctata. The hybrid from cross IR36/ P. coarctata, which showed parental isozyme bands both of O. sativa and P. coarctata, did not survive. The hybrid from BR-7(4x)/P. coarctata showed a faint band corresponding to that of P. coarctata. Both P. coarctata and the tetraploid hybrid had 48 chromosomes. Chromo-

somes in the hybrid, counted from 50 cells, appear to be a mixture of the parents. We examined 70 cells in triploid hybrids P. coarctata Rajashail/Binnatoa and sterile found them to have 36 chromosomes. The morphology of the triploid hybrid is close to that of P. coarctata. Three F1s involving P. coarctata with Binnatoa 1, Rajashail 1, and Rajashail 4 showed 24.3, 15.3, and 30.8% seed set, respectively. Seed set in selfed P. coarctata was 94%. Embryos from three seeds obtained after backcrossing P. coarctata/Rajashail 4 with Rajashail pollen were rescued 11-13 d after anthesis and are currently being grown in 1/4 MS medium with 0.45% agarose. Chromosomes of P. coarctata and O. sativa are small and not morphologically distinct from each other, making studies difficult on the tetraploid hybrid nature through standard karyotype analysis. We have multiplied the hybrid through vegetative propagation. We will examine it for hybridity based on meiotic chromosome and restriction fragment length polymorphism analysis.

Jena KK. 1994. Production of intergeneric hybrid between Oryza sativa L. and Porteresia coarctata T. Curr. Sci. 67:744-746. Sarkar RH, Samad MA, Seraj ZI, Hoque MI, Islam AS. 1993. Pollen tube growth in crosses between Porteresia coarctata and Oryza sativa. Euphytica 67:744-746.

Cited references

Percent seed set and survival in crosses of O. sativa and P. coarctata.

Wild halophytic Porteresia coarctata (2n=4x=48), endemic to the coastal areas in Bangladesh, is a potential source of salt tolerance for Oryza sativa. Crosses between these two genera were unsuccessful until Jena (1994) reported a hybrid between IR36 and P. coarctata. This triploid hybrid was much smaller than either parent and was completely male sterile. Sarker et al (1993) found that P. coarctata accepts 0. sativa pollen readily, but not vice versa. Our crossing strategy therefore involved P. coarctata

Cross combination IR36 (4x)/ P. coarctata IR64 (4x)/ P. coarctata Latisail (4x)/ P. coarctata BR-7 (4x)/ P. coarctata P. coarctata/Rajashail (2x) P. coarctata /Binnatoa (2x) P. coarctata/BR-7 (2x)

Seed set (%)a 1.6 1.6 0.16 0.44 2.34 4.16 2 (180) (120) (1250) (900) (940) (240) (1140)

Germinated plants (no.) 3 2 2 4 22 10 6

Surviving plants (no.) 0 0 0 1 4 4 0

aValues in parentheses indicate total number of florets used in crossing.

IRRN 21:2-3 (August-December 1996)


DNA content in rice hybrids and heterosis

X. C. Liu, J. L. Wu, S. Q. Tang, and S. K. Min, China National Rice Research Institute, Hangzhou 310006, Zhejiang, People's Republic of China

Certain metabolic functions in the nucleus are enhanced in heterotic plants compared with those of their parents. However, what happens to the nucleus under heterozygosity is still unknown. Nebiolo et al (1983), in a study on RNA metabolism in protoplasts isolated from seedlings of a heterotic maize hybrid and its parents, suggested that a hybrid nucleus may provide advantages in the rate of DNA and RNA synthesis. We studied the nuclear behavior and DNA content in hybrids to improve understanding of the molecular basis of heterosis. Using a MIPS-1 imaging analytical instrument with analytical software, intact

nuclei were isolated from root tips on slides, stained with Feulgen, and screened for different mitotic stages (prophase, metaphase, anaphase, and telophase). Screened nuclei were synchronically determined based on optic density (OD), integrated optic density (IOD), and size. We selected five female indica parental lines, six male indica parental lines, and nine hybrids derived from them. We examined 4,401 nuclei from root tips based on OD, IOD, and nucleus size (or an av of 55 nuclei for each mitotic stage for each material). Average OD value of nuclei from the nine hybrids was less than that from the 11 parental lines. However, average IOD value was greater in hybrids than in the parents (see table). Average nucleus size, which significantly expanded during prophase and telophase stages (at the 5% level), was greater in the hybrids. OD, IOD, and nucleus size were compared for hybrids and their parents (see figure). We found that nucleus size is directly proportional to IOD based on the regres-

sion index of about 0.977. Increases in DNA content and nucleus size and decrease in DNA density occurring in the hybrids indicate that gene expression is probably enhanced. These factors could be important for understanding karyological and molecular bases of rice heterosis. However, we have not yet determined which part of the DNA increases and what its functions are.

Variation in DNA content and nuclear size in hybrids compared with those in their parental lines (represented at zero).

Comparison of nuclear size and DNA content between nine hybrids and their parental lines. Determinant a Prophase b F 333 0.158 5.69 36.77 M 374 0.157 5.87 38.38 H 592 0.146 6.22 43.41* F 143 0.300 3.89 13.25 Metaphase M 218 0.286 3.98 14.09 H 298 0.283 4.04 14.56 F 254 0.240 2.19 9.38 Anaphase M 313 0.239 2.18 9.29 H 406 0.231 2.27 10.16 F 316 0.176 2.5 14.67 Telophase M 416 0.176 2.56 14.91 H 738 0.170 2.73 17*

Nuclei (no.) OD (arbitrary) IOD (arbitrary) Size (m) 2

a OD = optic density, IOD = integrated optic density. b = female line, M = male line, H = hybrld. * = signlficant at the 5% level. F

Induction of Agrobacterium tumefaciens vir genes by rice

K. Vijayachandra, K. Palanichelvam, and K. Veluthambi, Plant Biotechnology Department, School of Biotechnology, Madurai Kamaraj University, Madurai 625021, India

Transformation of monocotyledonous plants, such as rice, by Agrobacterium has low efficiency and has not yet become a routine laboratory technique. We studied whether rice tissues have limited ability to induce Agrobacterium vir genes and to promote synthesis of T-strands.

We used Agrobacterium tumefaciens strain A348 harboring a cosmid with vir ElacZ fusion to monitor ability of plant tissues to induce vir genes. Tobacco leaf segments preincubated for 48 h in Murashige and Skoog (MS) medium and incubated with Agrobacterium for 24 h induced vir E expression to 4,500 Miller units, which equaled vir induction level obtained using the chemical signal, acetosyringone. Induction of vir E expression by leaf segments of Oryza sativa L. cv Co 43 was low (only about 5% of that achieved by tobacco leaf segments). We analyzed various parts of rice seedlings of different ages for their ability to induce vir genes. Rice roots, leaf bases,

coleoptiles, and leaf segments induced vir genes to very low levels (see table). Only scutellum from 4-d-old seedlings induced vir E to a high level (about 40% of induction by tobacco). We found that a 72-h preincubation is essential for vir gene induction. Most of the vir gene induction was associated with the preincubated scutellum per se. The conditioned medium exhibited very little induction. Agrobacterium efficiently transformed scutellum-derived calli of japonica rice cultured for 25 d on a 2,4-D-containing medium (Hiei et al 1994). However, scutella excised from 5-d-old seedlings and grown on 2,4-D-containing medium were transformed at very low efficiency.

36 IRRN 21:2-3 (August-December 1996)

Induction of A. tumefaciens vir E gene by various parts of rice plants of different ages. Age of rice plant (d) 4 8 Part of rice plant preincubated for 72 h Coleoptile Scutellum Leaf blade Leaf base Root Leaf blade Leaf base Leaf blade Leaf base MS medium MS medium + 60 M AS MS medium + tobacco leaf segments Mean Miller units SE

14 20

203 2919 501 378 362 353 531 528 455

4 284 47 42 39 17 22 75 15 8 90

Control values:

175 5129

6840 235

induced vir E only to about 646 Miller units. The scutellum-derived calli that induced vir genes more efficiently exhibited a higher level of susceptibility to Agrobacterium mediated transformation. However, scutella excised from 4-d-old seedlings and grown on hormone-free medium induced vir gene expression to a level (2,547 Miller units) higher than that achieved with scutellum-derived calli. The results suggest that these scutella may be more susceptible to Agrobacterium than the scutellum-derived calli. We also found that rice scutellum promotes generation of the T-DNA transfer intermediates, T-strands, in Agrobacterium.

each other. A 0.52-kb EcoRI/XhoI fragment of a cDNA clone (RTS1) was used as a probe for screening the IR54 genomic library. A positive genomic clone (RTS2) was isolated and sequenced. Northern blot analysis (Fig. 1) and in situ hybridization (Fig. 2) showed that the gene was predominantly expressed in the anther's tapetum during vigorous meiosis and disappeared before anthesis. Sequence comparison between the cDNA clone

Scutellum-derived calli induced vir E expression to 1,745 Miller units, whereas scutella excised from 5-d-old seedlings and grown on 2,4-D-containing medium

Hiei Y, Ohta S, Komari T, Kumashiro T. 1994. Efficient transformation office (Oryza sativa L.) mediated by Agrobacterium and sequence analysis of the boundaries of the T-DNA. Plant J. 6:271-282.
1. Northern blot of total RNA isolated from leaves, seeds, and panicles at various

Cited references

Isolation of a tapetumspecific gene and promoter from rice

J.-Y. Lee, R. R. Aldemita, and T. K. Hodges, Botany and Plant Pathology Department, Purdue University, West Lafayette, IN 47901, USA

DNA from rice cultivar IR54. A panicle cDNA library was constructed and differentially screened for panicle and anther specificity. We selected two cDNA clones that could be cross-hybridized with

development stages and probed with putative anther-specific cDNA. Lanes with RNA are (A) leaf, (B) panicle, 4 d before C, (C) panicle when auricles of first and second leaves are at same level, (D) panicle, 4 d after C, (E) panicle, 8 d after C, (F) panicle, 12 d after C, and (G) seed.

Potential vulnerability of cytoplasmic male sterility (CMS)-derived hybrids calls for developing an alternative method of controlling male fertility/sterility for rice breeding and hybrid development. A tapetal-specific promoter-driving barnase induced sterility in tobacco and oil seed rape plants (Mariani et al 1992). Fertility was restored with the same promoterdriving barstar. Although tapetum-specific cDNA clones have recently been isolated from rice (Tsuchiya et al 1992) no regulating sequences have been reported. Developing a system by which male fertlity in rice can be controlled requires a promoter that can regulate gene expression. specifically in the male organ, during anther development. We cloned a rice anther (tapetum)specific cDNA and corresponding genomic

2. In situ hybridization of transverse sections through rice spikelet, including anthers with denatured cDNA (RTS1) probe. Spikelets were sampled when auricle of first leaf was 2.5 cm above the auricle of the second leaf. (A) and (B) anthers and glumes: (C) lower magnification of A; (D) and (E) RNAsetreated anthers before hybridization; and (F) leaf sheath, a = anther; t = tapetum; lo = locule; g = glume; ep = epidermis; and en = endothecium.

IRRN 21:2-3 (August-December 1996)


(RTS1) and the corresponding genomic clone (RTS2) and a primer extension assay showed that this gene has no introns and a continuous coding region of 285 base pairs yielding a putative polypeptide of 94 amino acids. The coding region of the gene has a 75.8% GC content (216/285). The nucleotide and deduced amino acid sequences. when compared with those in data banks, did not show any significant homology to known sequences. However, GAATTTGTTA, a sequence in the promoter region, differs only by one or two

nucleotides from one of the conserved sequence motifs in the promoter region of two pollen-specific genes of tomato (Twell et al 1991). The promoter function and regulation of RTS2 for tapetum specificity may be limited to rice. Experiments to obtain rice plants transformed with gus A genecontaining constructs under the control of this tapetum-specific promoter are in progress. These studies will allow us to determine the DNA sequences responsible for controlling tapetum-specific expression.

Mariani C, Gossele V, De Beukeleert M, De Block M, Goldberg RB, De Greef M, Leeman J. 1992. Achimaeric ribonucleaseinhibitor gene restores fertility to male sterile plants. Nature 357:348-387. Twell D, Yamaguchi J, Wing RA, Ushiba J, McCormic S. 1991. Promoter analysis of genes that are coordinately expressed during pollen development reveals pollen-specific enhanced sequences and shared regulatory elements, Gene Dev. 5:496-507. Tsuchiya T, Toriyama K, Nasrallah ME, Ejiri S-I. 1992. Isolation of genes abundantly expressed in rice anther at the microscope stage. Plant Mol. Biol. 20:1189-1993.

Cited references

Production and characterization of Oryza sativa L./ O. minuta Presl. hybrids and backcross progenies
A. L. Mariam, Biology Department, Faculty of Science and Natural Resources, Universiti Kebangsaan Malaysia, Sabah Campus, Locked Bag No. 62, 88996 Kota Kinabalu, Sabah, Malaysia; A. H. Zakri, M. C. Mahani, Genetics Department, Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor DE, Malaysia: and M. N. Normah, Botany Department, Faculty of Life Sciences, Universiti Kebangsaan Malaysia

We attempted to produce hybrid and backcross progenies between rice and wild species to enable the transfer of some important traits from O. minuta into locally cultivated rice. Rice varieties Mahsuri, Setanjung, MR84, and MR103 were crossed with one accession of O. minuta (IRRI Acc. 101141), which also served as the male parent. The original F 1 hybrids and two colchicine-treated F 1s were then backcrossed to the respective O. sativa varieties as the recurrent parent to produce backcross 1 (BC1). Successive backcrosses to the recurrent parents were made to produce BC2 progenies. Seeds of the F 1, BC1, and BC2 with poorly developed endosperm began to degenerate 2 wk after pollination. To ensure survival of the hybrids and BC progenies, the 14-d-old embryos were rescued on 1/4 MS medium following the procedure of Jena and Khush (1984).

Morphological characteristics, pollen fertility, somatic chromosome number, and meiotic chromosome pairing of the F 1 hybrids and BC progenies were observed. The hybrids and BC progenies were also subjected to starch gel electrophoresis and stained for shikimate dehydrogenase (SDH), phosphogluconate dehydrogenase (PGD), and glutamate oxaloacetate transaminase (GOT) activities using the procedure described by Glaszmann et al (1988). Seed set in O. sativa/O. minuta crosses ranged from 9.5 to 25.1%, depending on the rice variety used (see table). By rescuing 10- to 14-d-old embryos. 414 F 1 hybrids were successfully raised to maturity. The hybrids closely resembled their wild parent, although some of their morphological characteristics were intermediate. The male sterile F 1 plants were vigorous and tillered profusely. All had the expected chromosome number (2n=3x=36) and showed irregular meiotic chromosome pairing, with mostly univalents. Mean chromosome configuration was 29.31 (16-

36) Is + 3.32 (0-10)IIs + 0.02 (0-1)IIIs + 0.002(0-1)IVs. Seed set in the original F 1 backcrossed with O. sativa parents was extremely low (0.03%); however, seed set was slightly higher (0.88%) when backcrosses were done on the colchicine-treated plants (see table). The 17 BC1 plants obtained after embryo rescue appeared more like the F1 and were male sterile. These plants, for which chromosome numbers varied from 44 to 48, exhibited irregular meiosis with mostly univalents. Seed set in BC 2 was about 2.4% (see table). Of the 47 embryos cultured, 18 BC2 plants grew to maturity and had different morphological characteristics, pollen fertility, and somatic chromosome number. Of these, one plant closely resembled the O. sativa parent and had 24 chromosomes. The partially fertile plant had about 58.8% pollen stainability and 12.47% spikelet fertility and showed 12 bivalents in 60% of the cells and 9-11 bivalents in the other 40% of the cells. The results showed the recovery of O. sativa phenotype in the

Production of Oryza sativa/O. minuta F 1 hybrids and backcross progenies. Cross combination a Spikelets pollinated (no.) 1,742 1,729 2,465 2,346 37,080 3,733 4,452 Seed set (%) 15.1 25.1 12.5 9.5 0.03 0.88 2.47 Embryos cultured (no.) 132 138 110 112 10 33 110 Plants obtained (no.) 108 123 89 94 1 16 32 (18) a Chromosome number (2n) 36 36 36 36 45 44-48 24-37

Mahsuri/O. minuta (F 1) Setanjung/O. minuta (F 1) MR84/O. minuta (F 1) MR103/O. minuta (F 1) (O. sativa/ O. minuta) O. sativa (BC 1) (CT F1)/O. sativa (BC1) (BC1 from CT F 1) O. sativa (BC 2)

a CT F 1 = colchiclne-treated F 1 (O. sativa/O. minuta). Figure in parentheses shows number of plants with normal growth.


IRRN 21:2-3 (August-December 1996)

second backcross, which probably resulted from the restricted recombination between parental genomes indicated by the low frequency of rod bivalents in the meiosis of the F1 hybrids. Furthermore, none of the O. minuta allozyme (Sdh1, Pgd2, Got1, and Got3) was detected in the 24-chromosome plant, indicating lack of introgression of chromosome segment from the donor wild species. Cited references
Glaszmann JC, delos Reyes BG, Khush GS. 1988. Electrophoretic variation of isozymes in plumules of rice (Oryza sativa L.) - a key to the identification of 76 alleles at 24 loci. IRRI Res. Pap. Ser. 134.14 p. Jena KK, Khush GS. 1984. Embryo rescue of interspecific hybrids and its scope in rice improvement. Rice Genet. Newsl. 1:133134.
1. Dehusked grains of a) indica (IR54), japonica (Nipponbare), and javanica (Rinatte) and b) their corresponding callus types.

Histological observation of callus morphology in rice

J. O. Narciso and K. Hattori, Laboratory of Plant Genetics and Breeding, School of Agricultural Sciences, Nagoya University, Chikusa, Nagoya 464-01, Japan

species were yellowish, compact, and smooth-surfaced. Slight changes in callus morphology were noted 1 mo after culture. Green spots occupied more than 50% of the whole callus mass in IR54. Calli of Nipponbare and Tsutsu became globular and easily disintegrated into small pieces. Callus of Rinatte became friable. Callus types of the three subspecies represented by IR54, Nipponbare, and Rinatte are shown in Figure lb. Histological observation through a scanning electron microscope showed that

each callus type had a distinct cell cumposition (Fig. 2a-c). Callus of IR54 showed a large, highly compact cell mass with domelike structures. The surface was rough with corrugations (Fig. 2a). Callus of Nipponbare was made up of compact, globular cell masses of about the same size (Fig. 2b). The cell masses were composed of round cells arranged in layers, like a flower. Callus of Rinatte was composed of small. round, compactly arranged cells (Fig. 2c). Light microscopic observation of the resin sections generally showed darkly stained, meristematic cell clusters. The results of a study using mungbean suggest that cell types in a callus mass relate to embryogenic potential (Narciso and Hattori 1995). This relationship has not been fully studied in rice. Results of this research could provide the basic information needed to establish this relationship. Cited reference
Narciso JO, Hattori K. 1995. Different cell compositions in mungbean ( Vigna radiata (L.) Wilczek) cotyledon calli. Breed. Sci. 45(2):173-177.

We evaluated callus morphology, emphasizing cell compositions of rice varieties. through histological observation. Callus morphology of five indica (IR54. Rc2, Oboshi, 2757, 2764), two japonica (Nipponbare, Tsutsu), and two javanica (Lemonte, Rinatte) varieties was evaluated in two callus induction media. Varieties showing good callus growth 2 and 4 wk after culture were evaluated through histological observations using scanning electron and light microscopes. Of the varieties tested, IR54, Toboshi, Nipponbare, Tsutsu, and Rinatte exhibited good callus growth. Callus morphology of each subspecies differed from one another. At 2 wk after culture, calli of IR54 and Toboshi were characterized by two callus masses: one was yellow, compact, and smooth-surfaced; the other yellowish with hairy protuberance and green spots. Japonica varieties Nipponbare and Tsutsu and javanica variety Rinatte had almost similar morphology. Calli of both sub-

2. Scanning electron microscopic observation of the cell compositions in the calli of a) IR54, b) Nipponbare, and c) Rinatte. Scale bars = 300 .

Genetic analysis of epicutitular structure involving a dripping-wet leaf mutant of rice

K. Nakamura and K. Hattori, Laboratory of Plant Genetics and Breeding, Nagoya University, Chikusa, Nagoya 464-01, Japan

We obtained a dripping-wet leaf mutant among doubled haploid progenies regenerated from rice ( Oryza sativa L. cv Nipponbare) anther culture. This mutant

was derived from callus treated with 20 Gy of gamma rays 2 d after transfer to regeneration medium. Rice leaves, in general, can shed drops of water; however, leaves of the dripping-wet leaf mutant do notbecause of their highly hydrophilic leaf surface. Many of these mutant types, which are controlled by a single recessive gene, have been obtained in rice. Some of these gene loci have already been identified Satoh et al 1983). Although factors causing high leaf wettability in these mutants are not clear, leaf surface structure

IRRN 21:2-3 (August-December 1996)


Embryogenic cell suspensions established from a high-protein purple black rice

Huihua Fu, Tingliang Tian, Biology Department, Central Chinal Normal University, Wuhan 430070, China; and Yingguo Zhu, Biology Department, Wuhan University, Wuhan 430072, China

Scanning electron microscopy observations of the flag leaf blade surface at the abaxial side of a dripping-wet mutant. Nipponbare (wild type) (a and b) dripping-wet leaf mutant (c and d). LP = large papillae, SP = small papillae, S = stomata, P = pili.

is one of the putative causes of wettability. To study the histological basis of leaf wettability, we observed the leaf surface of the newly obtained dripping-wet leaf mutant under a scanning electron microscope. We observed that the dripping-wet leaf mutant was also an epicuticular waxless mutant. Large and small papillae, stomata, and pili on the mutant leaves were not so different from those on wild types (see figures 1a, 1c). However, the mutant leaf had a plain surface with epicuticular wax less developed than that found in wild types (see figure 1b, 1d). We also investigated the inheritance of epicuticular wax and dripping-wet leaf traits in crosses of mutant type/wild type and wild type/mutant type. The F1 plants in both crosses were wild types. In the F 2 of both crosses, drippingwet leaf plants always had the epicuticular waxless trait. The segregation ratio of the wild type to mutant type (dripping-wet leaf and epicuticular waxless) fitted to a 3-1 ratio

in both crosses, indicating that the drippingwet leaf trait was caused b) reduced leaf epicuticular wax. Many mutants with less epicuticular wax accumulation were already obtained in Arabidopsis, barley, and wheat (Lemieux et al 1994). Reduced wax on the leaf surface altered the reflection of light, which allowed isolation of the mutants through visual inspection. These mutants were designated as glossy or nonglaucous mutants. The dripping-wet leaf mutant of rice might be one of these mutants with reduced epicuticular wax.

Lemieux B, Koornneef M. Feldmann KA. 1994. Epicuticular wax and eceriferum mutants. In: Meyerowitz and Somerville, editors. Arabidopsis. New York: Cold Spring Harbor Laboratory Press. p 1031 - 1047. Satoh H, Iwata N, Omura T. 1983. Gene analysis of some dripping-wet leaf mutants in rice. Jpn. J. Breed. 33 (Suppl. 2): 243-243.

Cited references

Establishing embryogenic cell suspensions is important in regenerating plants from rice protoplasts. We studied the effects of explants, media, hormones, and culture procedures on callus induction and cell suspension of indica variety HuaHei 01, a new high-protein purple black rice derived from progenies of Hua 03/Hong Jing 501. Effects of explants. We cultured explants from young panicles (1-10 mm), anther at mid-uninucleate stage, and mature seeds on Murashige and Skoog (MS) medium supplemented with 2 mg 2,4-D L-1 + 0.2 mg Kin L -1 . The lowest callus induction (0.4%) was obtained from anthers. Young panicles had high callus induction (24.6%), as did mature seeds (21.3%). Calli from anthers and mature seed were compact. globular, and yellow, and could be used to establish embryogenic cell suspensions. Calli from young panicles were scattered and pale yellow. Effects of media and hormone ratios. We chose MS, N6, and GM supplemented with different ratios of hormones for incubating explants. AA, N6, and MS with 2 mg 2,4-D L-1 + 0.2 mg Kin L-1 was used for the liquid suspension culture. Although the calli induction rate on N6, MS, and GM media did not significantly differ, the quality of calli induced on the various media did. Most of the calli induced on N6 were yellow, compact, and globular, and grew rapidly; those induced on MS and GM grew slowly. Calli in the AA liquid medium remained yellow, while those in the N6 and MS media turned brown in the first month of suspension. Hormones significantly affected induction frequency and callus quality. Time required for callus induction was shortened, and some calli induced were pale yellow with root hairs when 4 mg NAA L-1, 1 mg 2,4-D L-1, and 0.2 mg Kin L-1 were supplemented. However, induction


IRRN 21:2-3 (August-December 1996)

Effects of hormones on callus induction from mature seeds of rice. a Hormone (mg L -1) NAA (4) + 2,4-D (1) + kinetin (0.2) 2,4-D (2) + kinetin (0.2)
a Basic medium used was Murashige and Skoog (MS).

Callus Induction frequency (%) 20.8 24.2

Callus characteristics Pale yellow; some calli with root hairs Globular, compact, and yellow

frequency was slightly higher and calli were fresh yellow, compact, and globular when supplemented with 2 mg 2,4-D L-1 and 0.2 mg Kin L-1 (see table). Effect of culturing procedures. When calli reached 1-2 mm in diameter, half were transferred into AA, MS, and N6 liquid media supplemented with 2 mg 2,4-D L-1 and 0.2 mg Kin L-1 and shaken at 120 rpm.

The other half was subcultured onto the same induction medium with 500 mg proline L-1 for 10 d to 6 mo (subcultured once a month) and then transferred into liquid suspension culture. Calli turned brown and died in 3-4 d in liquid culture with immediate transfer or subculture for less than 1 mo. On the other hand, calli could propagate quickly when transferred

into liquid media after subculture on solid induction media for 1 mo or longer. At the start of suspension culture, the calli were subcultured every 3 d by replacing all original medium with the same fresh medium. After 4 wk, the aggregates grew well, and cultures were subcultured every 6 d by replacing two-thirds of the original medium with the same fresh medium. Subsequently, the well-grown cultures released many miniclusters into the liquid medium. Thus, through about 3 mo of suspension culture, we obtained embryogenic cell suspensions composed mainly of cell aggregates consisting of 10-60 cells with fresh yellow color and dense cytoplasm.

Factors affecting pollen embryogenesis of rice anther culture

J. K. Sohn, Agronomy Department, Kyungpook National University, Taegu, Republic of Korea (ROK); G. H. Yi, B. G. Oh, National Youngnam Agricultural Experiment Station, 1085, Milyang, ROK; S. J. Yang, IRRI: and T. S. Kwak, Agronomy Department, Sangi University, Weonju, ROK

Table 1. Effects of abscisic acid (ABA) on embryo and callus formation in rice anther cutture. ABA (mg L -1) 0 0.5 1.0 5.0 10.0 20.0

Responding anthers (%) 19.7 28.8 19.7 18.1 15.1 7.8 2.3 a 2.2 2.1 1.8 1.6 1.4

Calli formed (%) 8.4 9.1 7.2 6.6 7.3 2.9 1.1 1.1 1.2 1.0 0.9 0.7

Embryos formed (%) 1.6 10.8 7.2 8.8 5.5 3.2 0.4 1.2 0.8 0.9 0.9 0.5

Mean SE.

Table 2. Effects of growth regulators on germination of pollen embryos in rice anther culture. Growth regulators (mg L-1) Zeatin 1 5 1 5 0 0 IAA a 0 0 0.2 0.2 0.5 0 Kinetin 0 0 0 0 2 0 Green 6.1 13.9 8.6 9.8 26.7 8.0 2.0 3.9 1.8 0.2 1.9 1.0

Anther culture techniques as a means of breeding are considered effective and timesaving methods for obtaining pure linea in rice and for other major crop improvement. Pollen embryogenesis is believed to provide new prospects for anther culture because of its stability and enhanced recovery of regenerants. We occasionally observed pollen embryogenesis during rice anther culture in many cultivars and F1 plants in our laboratory. Growth regulators. water stress, and media composition affect the induction rate of pollen embryogenesis (Torrizo and Zapata 1986). Abscisic acid (ABA) in the callus formation medium was most effective for pollen embryogenesis of rice; callus formation from the anthers plated did not increase significantly. However, embryo formations were markedly increased.

Plants (%) Albino 5.0 5.0 2.0 0 0 8.0 5.0 5.0 2.0


aIAA = indole acetic acid. b Mean SE.

Maximum frequency (10.8%) of embryo formation was obtained at 0.5 ppm ABA (Table 1). Also, ABA treatment during the cold pretreatment period (15 C, 15 d) showed the same tendency as that under in vitro condition. We tested several combinations of plant growth regulators for germinating pollen embryos. The combination of 0.5 ppm

indole acetic acid and 2.0 ppm kinetin was suitable for germinating embryos (Table 2). The results indicate that ABA is important in plant regeneration and embryo formation in rice anther culture. Cited reference
Torrizo LB, Zapata FJ. 1986. Anther culture of rice. IV. The effect of abscisic acid on plant regeneration. Plant Cell Rep. 5:136-139.

IRRN 21:2-3 (August-December 1996)


Genetic studies on purple pigmentation in rice and its use in breeding two-line rice hybrids
M. Tongmin, L. Chunhai, Y. Guocai, and L. Xinggui, Food Crop Research Institute, Hubei Academy of Agricultural Sciences, Wuhan 430064, China

Many scientists, particularly those in Japan, have systematically studied how purple color is inherited in rice leaves. Three pairs of basic genes (C, A, P) control the inheritance of anthocyanin pigments (Nagao 1951 and Takashi 1957). The C-AP gene system in japonica rice is suitable for indica rice (Kinoshita 1984). All C, A, P genes have their multiple alleles as well as several main genes with multiple effects. The gene P1, for example, is responsible for giving leaves, nodes, internodes, sheath, and leaf rings their purple color. At least one or two pairs of inhibitor genes influence the expression of purple leaf color.

We studied the inheritance of two newly discovered purple rice lines and the possibility of their use as gene markers in hybrid rice breeding. OPL and PL 184 with purple leaves were discovered in fields planted to normal green varieties Ketan Nangka and W6184S in 1988 and 1990, respectively. (They were crossed with each other and then crossed with IRRI variety IR1552 in 1991 and 1992.) The resulting F 1 s and F2 s were purple, and no segregation in F2 populations was found, inferring that OPL, PL 184, and IRI552 may have allelic genes controlling purple pigmentation of seedlings. However, when OPL and PL 184 were crossed with 23 normal green rice cultivars or lines (12 indica thermosensitive genic male sterile (TGMS) lines, 2 japonica photoperiod-sensitive genic male sterile (PGMS) lines, 2 indica wild abortive (WA) lines, 1 indica BT lines, 4 conventional indica cultivars. and 2 japonica cultivars), all F 1 s had green leaves, indicating that the purple leaf character of OPL and PL 184

Table 1. Segregation and genetic ratio of seedling leaf color in F 2 populations from crosses between purple and green rice varieties. Cross W9056 S/PL 184 W8013 S/PL 184 Shuang-guang S/PL 184 Zhen-shan 97 A/PL 184 Mi-ai 64 S/PL 184 Pei-Ai 64 S/PL 184 Yue-tai A/PL 184 W91273 S/PL 184 W6068 S/PL 184 An-nong S/OPL 3285 S/PL 184 W91238 S/PL 184 E47 S/PL 184 HN5 S/PL 184 Total plants (no.) 1,081 1,361 5,360 12,034 17,898 6,025 2,034 2,699 8,127 4,809 4,720 3,969 1,671 4,143 Green plants (no.) 827 1,038 4,352 9,833 14,520 4,852 1,655 2,173 6,947 4,158 4,030 3,439 1,448 3,698 Purple plants (no.) 254 323 1,008 2,201 3,378 1,175 379 526 1,180 651 690 530 223 445 Genetic ratio 3:1 3:1 13:3 13:3 13:3 13:3 13:3 13:3 55:9 55:9 55:9 55:9 55:9 229:27 c 2 P

1.2239 1.0994 0.0077 1.6425 0.1711 1.9957 0.0116 0.9153 1.3669 1.0557 1.1624 1.5882 0.6549 0.1454

0.50-0.25 0.50-0.25 0.90-0.75 0.25-0.10 0.75-0.50 0.25-0.10 0.90-0.75 0.50-0.25 0.25-0.10 0.50-0.25 0.50-0.25 0.25-0.10 0.50-0.25 0.75-0.50

Table 2. Segregation and genetic ratio of seedling leaf color in testcross F 1 and three way cross F 1 . Cross Generation Total plants (no.) 148 110 8434 Green plants (no.) 76 62 7400 Purple plants (no.) 72 48 1034 Genetic ratio c 2 P

was recessive. This trait can therefore be used as a gene marker in hybrid rice breeding. We investigated leaf color at the seedling stage (about five leaves) in the F 2 generation of 14 crosses between PL 184 or OPL and green lines (Table 1). Crosses 1 and 2 had a 3:1 genetic ratio for green:purple, indicating that PL 184 and W9056S or W8013S differed in one pair of genes. Crosses 3-8 had a 13:3 genetic ratio for green:purple, indicating differences between parents were based on two pairs of genes, one of which was an inhibitory gene. Crosses 9-13 had a 55:9 genetic ratio, indicating that differences between green and purple parents were based on three pairs of genes, one of which was an inhibitory gene. The F 2 s from HN5S/PL 184 had a 229:27 genetic ratio, indicating that differences between parents were based on four gene pairs, one of which was an inhibitory gene. Although the green:purple ratios in the F2 s of Zhen-shan 97 A/PL 184 and HN5S/ PL 184 differed, both backcrosses had a genetic ratio of 1:1 (Table 2), which confirmed an inhibitory gene in the gene system of the purple character in rice. Furthermore, the ratio of green:purple in triple cross Shuang-guang S//W9056S/PL 184 was 113:15, indicating that inhibitory genes in W9056S and Shuang-guang S were nonallelic. Purple-leafed sterile plants were found in all F 2 s of crosses with TGMS or PGMS lines, indicating that purple genes and TGMS or PGMS can be combined into one line. Several purple-leafed TGMS lines have been developed. However, crossing these lines with normal green rice varieties produced F1 s with green leaves, and PL 184-5 and IR1552 had better maintainer ability than Zhen-shan 97 A. The results suggest possibility of using purple leaf character as a gene marker in hybrid rice breeding.

Cited references

Zhen-shan 97 A/PL 184// PL 184 HN5 S/PL 184//PL 184 Shuang-guang S//W9056 S/ PL 184


1:1 1:1 113:15

0.0608 1.5364 2.3311

0.90-0.75 0.25-0.10 0.25-0.10

Kinoshita T. 1984. Gene analysis and linkage map. In: S. Tsund, N Takashi. editors. Biology or rice. Tokyo: JSSP/Elsevier. p 187274. Nagao S. 1951. Gene analysis and linkage relationship of characters in rice. Adv. Genet. 4:181-212.


IRRN 21:2-3 (August-December 1996)

Grain quality
Calorific values of 60 rice varieties
P. V. Nandini, L. Prema and C. E. Ajithkumar, College of Agriculture, Thiruvananthapuram, Kerala, India

varieties after parboiling, with Aryankali again having the highest calorific value
Calorific values of 60 rice varieties. Variety Raw rice (kcal g -1) 330 317 342 345 307 345 342 321 335 341 312 Parboiled rice (kcal g -1) 357 337 380 375 358 353 345 329 372 358 333

(383 kcal) and Bhadra, the lowest (301 kcal).


We determined the calorific values of 60 rice varieties (30 hybrid derivatives, 28 traditional, and 2 improved varieties) in raw and parboiled forms (see table). Calories significantly differed among the varieties. In raw rice, traditional variety Aryankali had the highest value (371 kcal) and the hybrid derivative Bhadra. the lowest (279 kcal). We observed a significant variation in calories in all rice

Raw rice (kcal g -1) 312 332 336 346 357 316 348 344 316 341 332

Parboiled rice (kcal g -1) 333 348 356 359 365 364 354 356 373 358 362

Kavungin-poothala Veluthavattan Kattamodan Arvan Vadakken Chitteni Thekken Chenkayama Chuvannamodan Elappapoochemban Navara Thrissur local 1

Thrissur local 2 Ponnaryan Chuvannari Thavalakannan Veluthari Thavalakannan Thkkencheera Cheriya Aryan Aruvakkari Kutticheradi Sinduram lmproved CO 25 Mashuri

Cibodas, a high-yielding variety with good grain quality

A. Partoatmodjo, Allidawati, and Z. Harahap, Research Institute for Food Crop Biotechnology, JI. Tentara Pelajar 3A, Bogor 16002, Indonesia

B9775b-Mr-8-1-1 (derived from the cross B7004d-Mr-10-1/B6992f-Mr-262) is a

high-yielding line with good grain quality. Released as Cibodas for the irrigated rice ecosystem in Sep 1995, the variety is resistant to brown planthopper biotype 1 and bacterial blight strain III. It yields about 30% more than IR64 and Cisadane at 200- to 500-m above sea level (Table 1). Cibodas seems more suitable for cropping on intermediate elevation (about 400 m).

We compared grain quality characteristics of Cibodas with those of IR64 and Cisadane (Table 2). Large grain size (1,000-grain weight = 349) probably contributes to the varietys high yield potential. Cibodas has intermediate amylose content and tender cooking quality that rice consumers in Java prefer.

Table 1. Average yield of Cibodas in seven provinces in Indonesia. 1993-95. Province Locations (no.) Elevation (m) 50 200-450 400-600 2-18 86 750 320 Yleld (t ha-1) 5.7 8.2 7.0 5.8 6.2 6.3 6.9 Increase over check (%)a 2 30 29 12 5 14 12 a b a a a a a

Improving provitamin A (carotenoid) content of rice endosperm

E. T. Wurtzel, Z. H. Li, R. B. Luo, D. Matias, D. Mozoub, P. D. Matthews, V. N. Upasani, G. Valdez, A. Yoganathan, and J. Yu, Biological Sciences Department, Lehman College of The City University of New York, Bronx, New York 10468, USA

Jakarta West Java West Java Central Java East Java Lampung Bengkulu
a a = IR64, b = Cisadane.

1 3 3 3 1 1 1

Table 2. Grain quality characteristics of Cibodas, IR64, and Cisadane. Characteristic Yield (t ha -1) Plant height (cm) Days to maturity 1,000-grain weight (g) Head rice recovery (%) Grain length (mm) Grain length-breadth ratio Amylose content Cooking quality Cibodas 5.7 - 8.2 115 123 34 80 6.5 2.4 24 Tender IR64 3.8 - 6.2 99 115 25 80 7.0 3.2 22 Tender Cisadane 4.8 - 7.8 110 135 28 80 6.7 2.6 20 Tender

Genetic engineering is a feasible approach to alter or improve carotenoid content of endosperm, an agronomically valuable tissue. However, to plan a strategy for engineering rice endosperm, we must identify the biosynthetic block preventing carotenoid accumulation. Three alternative possibilities are being considered: 1) genes encoding one or more of the biosynthetic enzymes may not be expressed in the endosperm, 2) these enzymes are expressed but are not functional, and/or 3) a limitation

IRRN 21:2-3 (August-December 1996)


2. Western blot analysis: PSY protein expressed throughout rice endosperm development.

1. RT-PCR analysis: PSY and PDS transcribed throughout rice endosperm development. a M = mature endosperm. b DAF = days after flowering. c Sh = Sucrose synthetase.

exists in precursors available to the pathway. Rice endosperm lacks both end products of the carotenoid biosynthetic pathway and any intermediates, suggesting that biosynthetic enzymes may not be expressed. Only geranylgeranyl diphosphate (GGPP), a substrate of the first enzyme specific to the carotenoid biosynthetic pathway, is present. GGPP is a precursor common to other pathways, such as the gibberellic acid biosynthetic pathway. We developed gene and protein probes to test the expression of the PSY and PDS, the first two enzymes specific to the pathway. Genes encoding both enzymes were each mapped to a single genetic locus in rice and in maize and behaved as single

copy genes by Southern blot analysis. Reverse transcriptase-polymerase chain reaction showed transcripts in the leaves and endosperm. In developing rice endosperm. the PSY transcript was constant, whereas the PDS transcript increased over time (Fig. I). Using a polyclonal antiserum raised against maize PSY, Western blot analysis showed PSY to be constantly expressed in developing rice endosperm (Fig. 2). We used a starch gene encoding sucrose synthetase to compare the transcript and protein analyses. Finally, high-pressure liquid chromatography analysis of a rice carotenoid mutant showed phytoene (a carotenoid intermediate) in the leaves and embryos but not in the endosperm.

Several factors likely cause the carotenoid deficiency in rice. Genes encoding PSY and PDS, the first two biosynthetic enzymes, are expressed in developing rice endosperm. However, the absence of intermediates, even in rice mutants blocked in the pathway, suggests that the endosperm PSY does not function or that precursors feeding the pathway are limiting. Also, the level of PDS transcripts was not constant, suggesting the possibility of discordant expression of enzymes at inappropriate levels or times during endosperm development, which would contribute to the general problem of carotenoid deficiency.

Phytoene-forming activities in wild-type and transformed rice endosperm

P. Beyer, University of Freiburg, Institute of Biology II, D79104 Freiburg, Germany; P. K. Burkhardt, Swiss Federal Institute of Technology (ETH), Institute for Plant Sciences, CH-8092 Zurich, Switzerland; M. Schledz, S. Al-Babili, M. Bonk, and J. von Lintig, University of Freiburg; G. A. Armstrong and I. Potrykus, ETH

Rice endosperm does not contain carotene, including phytoene, the first carotene in the biosynthetic pathway. To develop a strategy for rice transformation with the aim

of obtaining phytoene synthesis in endosperm (as a first step toward b -carotene accumulation), we investigated in initial experiments the enzymatic prenyl lipidforming activity of this tissue. We incubated immature endosperm with [1-14Clisopentenyl diphosphate and [14 C] geranylgeranyldiphosphate (GGPP). After hydrolysis with alkaline phosphatase to convert prenyl phosphates into their corresponding alcohols, followed by extraction, reverse phase-high performance liquid chromatography (RP-HPLC) analysis revealed farnesol, geranylgeraniol, a third unidentified alcohol, and squalene as the main products in incubations with radio-labeled isopentenyl diphosphate.

Geranyl-geraniol was most important because its diphosphate, besides being a substrate in the formation of various prenyl lipids, is also the first carotenoid-specific precursorthe substrate of the enzyme phytoene synthase. Both phytoene synthase and GGPP are localized in the plastids of plants. We cloned a cDNA coding for phytoene synthase from the flower Narcissus pseudonarcissus and proved its identity by functional expression (enzymatic activity) in insect cells, using the baculovirus system. To confirm the presence of functional transit sequence, we studied the import of a radio-labeled in vitro translation product into isolated pea chloroplasts.


IRRN 21:2-3 (August-December 1996)

These experiments showed a high extent of protein import as measured by inaccessibility to thermolysin digestion. We conclude that the cDNA from N. pseudonarcissus codes for a functional phytoene synthase and thus is suitable for rice transformation. The cDNA was subcloned under the control of the CaMV35S promoter and glutelin promoter (Burkhardt et al 1996, Potrykus et al 1996). We examined the resulting transformants for phytoene synthase expression. First, reverse-transcriptase polymerase chain reaction was established to detect mRNA of the introduced transgenes in rice endosperm. The results revealed the presence of specific

mRNA, which was absent in endosperm of control plants. Second, using homologous anti-phytoene synthase antibodies, phytoene synthase was detected in transformed endosperm, whereas no signal was obtained in the control plants. Third, using RPHPLC, the phytoene content was determined in mature rice seeds deprived of the embryo. Phytoene content was usually below detection limit in the controls: in some cases, however, insignificantly low amounts probably derived from seed coat were present. In contrast, high amounts of phytoene found in seeds of some transformants allowed clear UV/VIS spectra-proving identity to be obtained. Constructs under glutelin promoter control were successful in

phytoene formation. The best line obtained so far showed a phytoene content of 1 mol per gram (after quantification by calibrated integration on HPLC and use of a molar extinction coefficient of 68,125 mol -1 cm-1). This corresponds to 545 g phytoene per gram, which is safely within the nutritionally useful range (Burkhardt et al 1996). Cited references
Burkhand PK, Beyer P, Terada R, Kloti A, Wnn J, Lintig JV, Armstrong GA, Potrykus I. 1996. Genetic engineering of provitamin A biosynthesis in rice endosperm. In: Khush GS, ed. Rice genetics III. Manila (Philippines): International Rice Research Institute, p 818-821.

Pest resistancediseases
Screening of Basmati rice genotypes against blast
R. Singh and D. S. Dodan, CCS, Haryana Agricultural University, Rice Research Station, Kaul 132021, Haryana, India
Basmati rice genotypes showing resistance to blast. Haryana, India, 1993 and 1994 kharifs. Leaf blast HKR86-416 HKR90-403 HKR90-407 HKR90-410 HKR91-401 HKR91-402 HKR91-404 HKR91-405 HKR91-408 HKR91-413 HKR91-424 HKR91-427 HKR91-456 HKR91-459 HKR91-460 HKR91-467 HKR92-405 HKR92-409 HKR239 Neck blast HKR90-403 HKR90-404 HKR91-405 HKR91-431 RP3121-14-10-2 RP3138-60-91-0-6 RP3238-38-15-7-1 RPST328 NDR637-90 UPR-BS-92-5 Both leaf and neck blast HKR90-403 HKR91-405

Resistance spectrum, race specificity, and expression of the Xa21 gene family
Guo-Liang Wang, Wen-Yuan Song, Li-Li Chen, R. Ruan, S. Sideris, and P. C. Ronald, Plant Pathology Department, University of California, Davis, CA 95616, USA

Blast caused by Pyricularia grisea Sacc.= Magnaporthe grisea (Hebert) Barr., is a devastating disease of scented, tall varieties in Haryana. We screened 175 rice genotypes in station and coordinated trials at the Rice Research Station in Kaul. For leaf blast screening in the Uniform Blast Screening Nursery, we alternated each test genotype with susceptible variety Taraori Basmati on all sides during the 1993 kharif (dry season). For neck blast screening, 1-mo-old seedlings of test entries were transplanted in two 5-m-long rows at 20- 15-cm spacing. One line of Taraori Basmati was planted between each test entry to ensure maximum disease development. We recorded leaf blast incidence during the last week of September when the disease was at its peak. Tillers of 25 randomly selected hills of each test entry were examined for neck blast 1 wk before harvest. We scored the disease severity using the Standard evaluation system for rice (1988). Genotypes showing

resistance to leaf or neck blast were retested during 1994 kharif to confirm their reaction to the disease. We found 19 genotypes resistant to leaf blast and 10 genotypes to neck blast during both years (see table). Only genotypes HKR90-403 and HKR91-405 showed resistance to both leaf and neck blast. Besides blast, three genotypes (HKR864 16, HKR90-404, and HKR92-405) also exhibited resistance to stem rot.

The rice gene Xa21 confers resistance to the bacterial pathogen Xanthomonas oryzae (Xoo) (Ikeda et al 1990, Ronald et al 1992). The Xa21 locus consists of a small multigene family, and transgenic lines (T0) expressing a single member of the Xa21 gene family confer resistance to Xoo race 6 (Song et al 1996). Unlike all other cloned plant disease resistance genes, the deduced amino acid sequence of Xa21 encodes a glycosylated leucine-rich repeat (LRR) extracellular domain, a single-pass transmembrane domain, and a serine threonine kinase intracellular domain. The sequence of the predicted protein of Xa21 strongly suggests its role in cell-surface recognition of a pathogen ligand and subsequent activation of an intracellular defense response. To determine the resistance spectrum of cloned resistance gene Xa21. we tested 300 T1 progeny from the transgenic line 106-17 containing Xa21 for Xoo resistance.

IRRN 21:2-3 (August-December 1996)


Inoculation results indicate that adding a single gene (Xa21 ) is sufficient to confer resistance to all seven Philippine races of Xoo and 20 isolates from China, Colombia, India, Indonesia, Malaysia, Nepal. and Thailand. Resistance to Xoo strains of Philippine races 1 through 6 cosegregated with polymerase chain reaction (PCR) profile of the Xa21 transgene in a 3:1 ratio. Southern blot analysis confirmed the PCR results, indicating that a single member of the Xa21 multigene family can confer resistance to diverse strains isolated from seven countries. The Xa21 gene family contains eight members, all located at the Xa21 locus. Each member contains an LRR motif.

Differences in the LRR domain of the family may play a role in the specificity of pathogen recognition. To test this idea. we generated transgenic plants carrying subclones from other members of the Xa21 gene family. The resistance of these lines to the six Xoo races has been determined. Preliminary screening showed that at least one copy (copy D) confers partial resistance to races 1 and 6 in transgenic plants. To analyze the expression of the Xa21 gene family, we made a cDNA library from the resistant line infected with Xoo. We identified 17 cDNAs hybridizing with LRR or/and kinase domains. Sequencing results indicate that at least four copies of the gene family are expressed.

Ikeda R, Khush GS, Tabien RE. 1990. A new resistance gene to bacterial blight derived from O. longistaminata. Jpn. J. Breed. 40 [suppl.l]:280-281. Ronald PC, Albani B, Tabien R, Abenes L, Wu KS, McCouch S, Tanksley SD. 1992. Genetic and physical analysis of the rice bacterial blight disease resistance locus, Xa21. Mol. Gen. Genet. 236: 113-120. Song WY, Wang GL, Chen L, Kim KS, Holsten T, Wang B, Zhai Z, Zhu LH, Fauquet C, Ronald PC. 1996. The rice disease resistance gene, Xa-21, encoded a receptor kinase-like protein. Sci. (in press)

Cited references

A bacterial blight-resistant, wide-compatible indica line

Zhu Lihong and Zhang Hongsheng, Nanjing Agricultural University, Nanjing 210095, China

In China, scientists started to develop innovative elite lines with genes for resistance to bacterial blight (BB) Xanthomonas oryzae pv. oryzae (Xoo ) in the early 1980s. Scientists are now using wide compatibility as an approach to develop indica/japonica hybrid rice through pyramiding genes for wide compatibility, disease resistance, and agronomically important characters. We determined resistance to BB by evaluating lesion length 20 d after inoculation (9 108 cells ml-1 concentration) using the clipping method at booting stage. We used 10 Xoo isolates (provided by the Plant Protection Department, Nanjing Agricultural University) of major pathotypes I through VII from different provinces in China: OS14 (I) (Liaoning),

KS-6-6 (II) (Jiangsu), HB84-17 (II) (Heibei), OS105 (II) (Guangdong), FJ856 (III) (Fujian), ZHE173 (IV) (Zhejiang). AH023 (IV) (Anfei), GD1358 (V) (Guangdong), GX325 (V) (Guangxi), and JS49-6 (VII) (Hunan). We evaluated wide compatibility using spikelet fertility from testcrosses to both typical indica and japonica cultivars. Mean fertility of unbagged F 1 plants was assessed at maturity. Spikelet fertility of up to 70% was normal and compatible. 85-21416 is a promising indica line developed from a 1978 cross between Taichung Native 1, a susceptible and noncompatible indica cultivar from Taiwan. China, and DZ78, a highly resistant, tall indica from Bangladesh. 85-21416 has strong resistance to BB and is extremely compatible in intersubspecific crosses. We grew F 3 lines of this cross combination in early spring of 1980. We then selected individuals each year through artificial inoculation evaluation. By 1985,85-21416 along with other advanced lines (F 7-8) were selected for stable resistance and other

characters. 85-21416 was used in 1988 as a resistance donor in intersubspecific genetic studies. The shorter lesion lengths of 85-21416 confirmed its stable BB resistance (Table 1). A recessive gene allelic to xa5 (data not shown) controls the resistance. 85-21416 may have inherited the resistance
Table 2. F 1 spikelet fertilities of 85-21416
crossed with various tester varieties. a Mean 87.3 91.7 87.9 79.4 62.5 3.58 3.20 5.04 5.28 1.41

Cross combination b Balilla/85-21416

Plants (no.) 15 14 6 12 18 18 14 12 10 10 15 10

Aikihikari/85-21416 85-21416/IR36
02428/85-21416 Ketan Nangka/ 85-21416 85-21416/CPSL017 85-21416 (I) Balilla (J) Aikihikari (J) 02428 (J) IR36 (I) CPSL017 (I)

68.4 1.73 80.6 5.31 67.9 4.86 89.3 3.51 71.4 7.06 92.9 1.05 82.6 3.00

21416/IR36 and IR36, which were collected in Sep 1995. b l = indica, J =japonica.

a Data were collected in 1992, except those for 85-

Table 1. Reactions of 85-21416 to bacterial blight isolates compared with Nanjing 11, DV85, and IRBB7. Isolates Line/variety Gene OS14 1.7 1.2 19.8 HB84-17 1.4 1.1 3.0 26.1 KS-6-6 1.8 1.2 25.8 OS105 0.8 1.5 20.6 FJ856 1.1 1.1 13.1 ZHE173 1.1 1.3 3.0 19.8 AH023 1.5 1.3 23.7 GD1358 3.9 0.7 5.0 9.7 GX325 1.6 1.6 25.6 JS94-6 0.6 0.9 3.0 10.2 Evaluation criterion Year

85-21416 xa5(t) DV85 xa5, Xa7 IRBB117 Xa7 Nanjing 11 None

Lesion length (cm) Lesion length (cm) Leaf area necrotized (%) Lesion length (cm)

1994 1994 1995 1994


IRRN 21:2-3 (August-December 1996)

gene from DZ78, which carries xa5 and xa7. F1 hybrids derived from crosses between 85-21416 and 16 japonica cultivarsin 1989 were fertile, with 84.2 2.8% average spikelet fertility. 85-21416 carries a

compatibility gene that appears allelic to S-5 n . Fairly fertile hybrids were obtained from crosses between 85-21416 and cultivars known to have the widecompatibility gene S-5 n (Ketan Nangka, CPSL017, and 02428) (Table 2). The

compatibility gene carried by 85-21416 could have come from indica DZ78, which produced a compatible intersubspecific hybrid when crossed with japonica tester cultivar Akihikari (69.1 1.1% spikelet fertility in 1992).

RFLP mapping of blast resistance gene Pi-k m in rice

R. Kaji and T. Ogawa, Kyushu National Agricultural Experiment Station, Chikugo, Fukuoka 833, Japan

Blast resistance gene Pi-k is located on chromosome 11 (Goto et al 1981) and has multiple alleles Pi-k h, Pi-k m, and Pi-k p (Kiyosawa 1978). Although restriction fragment length polymorphism (RFLP) linkage analysis indicated that Pi-k is located on chromosome 11, no RFLP marker closely linked to Pi-k was found at that time. For fine-mapping of Pi-k loci, we conducted RFLP analysis of Pi-k m and
Linkage analysis among Pi-k m and RFLP markers. Gene pair A Pi-k m B L190 R1506 G181 G1465 R1506 G181 G1465 G181 G1465 G1465 AABB 36 29 28 31 26 26 20 23 18 19 AABb 61 47 62 80 5 3 13 3 10 8 AAbb 1 3 5 12 1 1 3 0 1 1

RFLP markers using F 3 lines from the cross between Tohoku IL4 (japonica nearisogenic line with Pi-k m introgressed from Tsuyuake) and CO 39 (susceptible indica variety). Each F 3 line was inoculated with Japanese blast race 003 at 3-4 leaf stage using the spray method. The F 3 lines were classified either as resistant (or segregating) lines or susceptible lines 30 d after inoculation. DNA was extracted from the leaves of each F 3 line using the cetyltrimethyl ammonium bromide method. Total DNA was digested with restriction enzymes (Eco RI, HindIII, and Dra I) and blotted onto a positively charged nylon membrane filter using capillary transfer after electrophoresis. RFLP landmarkers on chromosome 11 and

cDNA clone R1506 (provided by NIAR/ STAFF) were used. Linkage analyses of Pik m and the RFLP markers are shown in the table. Recombination values calculated using the maximum likelihood method were converted into genetic map distances (cM) using the Kosambi function. All RFLP markers used in this study were linked to Pi-k m, notably L190, which was closely linked to Pi-k m at about 1.5 cM. Pi-k m is located on chromosome 11 in the RFLP linkage map between L190 and R1546 (see figure). Because L190 is located on the terminal region of chromosome 11 (Nagamura et al 1993), results of our study correspond to those of a classical linkage map. For further finemapping of the Pi-k m loci, we will subject more F 3 lines to RFLP analysis.

Segregation mode in F3 AaBB AaBb Aabb aaBB 0 0 0 1 0 0 1 0 0 0 aaBb 1 2 7 12 1 6 10 6 9 8 aabb 37 32 34 32 27 29 27 27 26 31

Genetic map distance (cM)

L190 R1506 G181

1 0 6 3 8 6

31 42 48 37 33 54

2 5 8 3 8 9

1.5 4.3 8.4 15.8 6.0 7.4 19.1 7.7 18.5 13.3

1.2 1.0 1.9 2.4 2.6 1.8 3.0 1.8 1.9 2.8

Linkage map of Pi-k m and RFLP markers on chromosome 11.

Indica/japonica doubled haploid population as a model for mapping rice yellow mottle virus and blast resistance genes
A. Ghesquiere, M. Lorieux, lnstitut Franais de recherche scientifique dveloppement en coopration (ORSTOM), BP 5045, 34032 Montpellier Cedex 1, France; E. Roumen, Centre de coopration internationale en recherche agronomique por le dveloppement (CIRAD), BP 5035, 34032 Montpellier Cedex 1, France; L. Albar, ORSTOM and CIRAD; D. Fargette, ORSTOM; N. Huang, IRRI; and J. L. Notteghem, CIRAD

Rice yellow mottle virus (RYMV) is the most damaging rice pest in irrigated fields of West Africa. Blast is a widespread disease causing severe yield losses. To map several resistance genes, we used two doubled haploid (DH) populations derived from the F 1 hybrids IR64/Azucena and IRAT 177/Apuraj and segregating for many agronomic traits. This analysis relied on core restriction fragment length polymorphism (RFLP) maps developed at IRRI for the IR64/Azucena population (Huang et al 1994) and at Cornell University for the IRAT 177/Apura population (Yu et al 1991).

RYMV scoring. The resistance analysis involved 74 IR64/Azucena DH lines and 48 IRAT 177/Apura DH lines. Nineteen-dayold rice plants were mechanically inoculated with a RYMV isolate collected from Mali. Two weeks after inoculation, newly emerged leaves were harvested for assessment of virus concentration by serological methods involving ACP- or DAS-enzyme-linked immunosorbent assays. Blast scoring. An initial survey involving the parents and 20 IR64/Azucena DH lines allowed us to select six diverse strains of Magnaporthe grisea apparently inter-

IRRN 21:2-3 (August-December 1996)


acting with different genes or quantitative trait loci (QTLs) in this population (Roumen et al, 1996, unpubl. data). The 135 mapped DH lines were evaluated for resistance to the six blast strains. Twenty to twenty-five plants from each DH line were grown in the glasshouse and inoculated with spore suspensions by spraying or injecting into sheaths of rice plants at the 4-5 leaf stage. Resistance scores ranged from 0 (complete resistance) to 6 (high susceptibility). Genetic mapping. For chromosome 12, additional RFLP markers were mapped on the available 183 DH lines of IR64/ Azucena progeny to refine QTL/major gene locations. Clones mapped on an interspecific backcross were used as probes (Causse et al 1994), and RFLP was identified using a chemiluminescent kit following CIMMYT protocols (Hoisington et al 1994). In addition, we conducted bulked segregant analysis (BSA) as described by Michelmore et al (1991) to compare the RYMV susceptible/resistant pools with 300 randomly amplified polymorphic DNA (RAPD) primers. Interval mapping identified the loci involved in resistance to RYMV or blast (Lander and Botstein 1989) and confirmed by the Kruskall & Wallis nonparametric test using MapQTL 2.4. RYMV resistance. For the IR64/ Azucena population, interval mapping revealed only one locus associated with significant effects on RYMV resistance on chromosome 12 (LOD=2.57), close to the RG341 marker. This locus probably corresponds to a QTL with a relatively high percentage of explained trait variance (R2=0.30). This localization was found highly consistent with that obtained by BSA. A 800-bp band amplified with Op O10 primer was also detected by BSA and mapped at 2.7 cM from RG341. A QTL was detected at the same location for IRAT 177/Apuraj population (LOD=3.27; R2=0.33). Because the two recombinant populations shared the five RFLP markers on chromosome 12, an integrative map of chromosome 12 was established on 2.58 individuals, and the QTL was reinforced (LOD=5.57; R2 =0.37). This result clearly indicates that the resistance genes from IRAT 177 and Azucena were allelic or closely linked.

Blast resistance. Significant resistance loci observed in the IR64/Azucena DH population are summarized in the table. Some loci were strain-specific while others were common to different strains. For instance, a QTL on chromosome 2 was detected with strains Br26, Ch66, and Ch72. Other strains, such as Br26, could reveal up to four different QTLs. Differential QTLs for strain Ph68 were detected following inoculation by spray or injection. QTLs for blast resistance tend to cluster on rice chromosomes (McCouch et al 1994). Two QTLs or major genes are lying on the opposite sides of RG869 (chromosome 12) in the same pattern as observed by Yu et al (1991) and could correspond to Pi4(t) and Pi6(t) resistance genes. For CD69 strain,

the locus detected on chromosome 12 probably corresponds to a major gene because it explained about 90% of the trait variance. In addition, the results suggest that one or two other different QTLs could exist in the vicinity of RG457. Similarly, loci detected on other chromosomes could correspond to previously located genes (see table). A cluster of resistance genes or QTLs involving a major QTL for resistance to RYVM and several blast resistance genes or alleles was observed on a relatively small portion of chromosome 12 (see figure). This chromosome segment will need special attention and further evaluation for determining its implications in breeding for resistance in rice.

Loci identified for resistance to 6 blast isolates by a QTL detection method (interval mapping) in IR64/Azucena cross. a BI isolates b Chromosome 1 2 2 5 Marker/ interval RG173 RG532 RG520 RZ123 RG654 RG313 RZ70 RZ225 6 6 8 RG213 RZ144 RZ667 G104 RZ617 RG667 RG451 10 11 12 RZ500 CD093 RG1094 RG118 RG457 010 010 RG341 RG869 RZ816 Map position (cM) 147.4 177.8 0 16.6 27.4 37.7 132.8 152.5 20.7 33.4 35.2 34.6 41.6 92.8 111.1 0.0 25.0 100.2 111.8 57.9 73.5 73.5 76.4 76.4 103.2 6.1 23.2 (i) 100.0i 34.0 (i) 10.8 37.0 (i) 25.0 90.0 (i) Pi4 (t) =Pi62? =Pi-ta? Pi6 (t) 2.2i 8.0 (a) 4.8 21.6 (i) 2.2 10.2 (a) 4.2 20.2 (i) Pi-a? 1.9 7.1 (i) 1.5 8.2 (i) 6.1 22.2 (i) 2.3 8.9 (i) Pi11 (t) 5.8 (20.5 (i) d 5.7 18.8 (i) 2.8ie 14.1 (a) 3.1 15.2 (a) 3.0 13.8 (i) Pi2 (t) or Pi9 (t) 3.0 11.6 (i) 2.3 9.5 (i) Br26 spr Ph68 spr or inj CD69 spr Ch66 inj Ch72 inj C16 spr Possible correspondence with known loci

2.9 24.1 (a) c



with injection method only.

aLoci with low significance (1.5 < LOD score < 2.3) are reported if detected with at least two Isolates. Letters in regular font = LOD score; in italics = % of explained variance by the locus. b spr = inoculation by spray: inj = inoculation by injection. c (a) = resistance locus brought by Azucena. d(i) = resistance locus brought by IR64. e i = locus detected


IRRN 21:2-3 (August-December 1996)

Characterization of a pathogenesis-related gene family induced in rice infected with Magnaporthe grisea
J. D. McGee, Botany and Plant Pathology Department, Purdue University (PU), West Lafayette, IN 47907, USA; N. J. Talbot, T. Bhargava, and J. E. Hamer, Biology Department, PU; and T. K. Hodges, Botany
Alignment of QTLs/major genes for RYMV and blast resistance on rice chromosome 12. Loci were detected using DH lines derived from two indica/japonica F 1 hybrids (IR64/Azucena and IRAT 177/Apura). Vertical dashed line indicates position of RG869 marker, which separates Pi4 and Pi6 in other studies. The peaks observed on each side of the line could correspond to these two genes.

and PIant Pathology Department, PU

Fungal pathogens, such as Magnaporthe grisea and Rhizoctonia solani can cause significant yield losses. Increased disease resistance can be obtained in plants transformed with genes encoding known antifungal proteins, such as the hydrolytic enzymes chitinase and b -1.3-glucanase

(Jach et al 1995, Zhu et al 1994). We isolated a M. grisea-induced promoter that was later used to drive antifungal genes in rice to suppress the effects of fungal infection. Rice cDNAs, w hose genes exhibited elevated expression during rice blast infection. have been isolated through differential screening (Talbot et al 1993). One of these cDNAs, MAG-7, was chosen as a source of the M. grisea-induced promoter. MAG-7 was found to share homology with a known class of PR proteins (PR-10 proteins) found in various plants (van Loon et al 1994). Southern blot hybridization of rice genomic DNA (varieties Co 39, IR36, and IR54) probed with the MAG-7 cDNA indicated that these varieties contain multiple (2-5) copies of the MAG-7 gene

Cited references
Causse M, Fulton TM, Cho YG, Ahn SN, Chungwongse J, Wu K, Xiao J. Yu Z, Ronald PC, Harrington SE, Second G, MacCouch SR, Tanksley SD. 1994. Saturated molecular map of the rice genome based on an interspecific backcross population. Genetics 138:1251-1274. Hoisington D, Khairallah M, Gonzalez-de-Leon D. 1994. Laboratory protocols: CIMMYT Applied Molecular Genetics Laboratory. Second edition. Mexico D. F.: International Maize and Wheat Improvement Center. Huang N, McCouch S, Mew T, Parco A, Guiderdoni E. 1994. Development of an RFLP map from a doubled haploid population rice. Rice Genet. Newsl. 11:134137. Lander ES, Botstein D. 1989. Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121:185-199. McCouch SR. Nelson KJ, Tohme J. Leigler KS. 1993. Mapping of blast resistance genes in rice. In: Rice blast disease. RS Zeigler, SA Leong, PS Teng, editors. London: CABI and Manila: International Rice Research Institute. p 167- 186. Yu ZH. I991. Molecular mapping of rice (Oryza sativa L.) genes via linkage to restriction fragment polymorphism, (RFLP) markers. Ph D thesis, Cornell University, Ithaca, N.Y. Yu ZH, Mackill DJ, Bonmann JM, Tanksley SD. 1991, Tagging genes for blast resistance via linkage to RFLP markers. Theor. Appl. Genet. 81:471-476.

1. Northern blot analyses comparing the presence of different rice mRNA transcripts at various times following Magnaporthe grisea infection. 10-g samples of total RNA isolated either from uninfected 3wk-old Co 39 seedlings (0 h) or from 3-wk-old seedlings infected for 12, 18, 24, 36, 48, 72, or 144 h were hybridized to different rice PR-10 probes. (a) Total RNA hybridized with a 814-bp nonspecific probe from a PR-10a cDNA, which shares homology with all three rice PR-10 genes. (b) Total RNA hybridized to a 198-bp probe derived from the 5' untranslated region of the PR-10a gene. (c) Total RNA hybridized to a 230-bp probe derived from the 3' untranslated region of the PR-10b gene. (d) Total RNA hybridized to a 231-bp probe derived from the 5' untranslated region of the PR-10c gene. The intensity of chloroplast 23S rRNA in b, c, and d shows similar amounts of RNA loaded onto gels.

IRRN 21:2-3 (August-December 1996)


per haploid genome. A 16-kb genomic clone isolated from a Co 39 genomic library was found to contain three copies of the MAG-7 gene (named rice PR-10a, rice PR10b, and rice PR-10c). Rice PR-10a, sequenced in its entirety, including about 1.4 kb of the 5' promoter region, shared 100% homology with the MAG-7 cDNA. The rice PR-10b and PR-10c genes have been partially sequenced. We constructed gene-specific probes to observe transcriptional differences among the three rice PR-10 genes in Northern blots of total RNA isolated from Co 39 seedling leaves at 0, 12, 18, 24, 36, 48, 72, and 144 h after M. grisea infection (Fig. 1). Transcripts of rice PR-10a were induced from a low basal level within 12 h following M. grisea infection, with increasing accumulation overtime. M. grisea also enhanced transcripts of rice PR-10b, for which induction became strongly visible 48 h after inoculation. Unlike rice PR-10a and rice PR-10b, rice PR-10c exhibited no signs of induction by M. grisea in RNA samples isolated from infected rice leaves throughout the 144 h. Tissue prints of M. grisea -infected Co 39 leaves were made using the PR-10a gene-specific probe (Fig. 2). Hybridization of the rice PR-10a probe occurred only at the infection site, illustrating that M. grisea induces the rice PR-10a transcript in a localized fashion.

Barley genes encoding class II chitinase, class II 1,3-gluconase, and Type-I ribosome-inactivating protein (Jach et al 1995) will be introduced into rice via Agrobacterium tumefaciens transformation to study their ability to control fungal diseases. Both the rice PR-10a promoter and the constitutively expressed ubiquitin promoter will be used to drive each of these genes; the effectiveness of both promoters will be compared.

Cited references
Jach G, Gornhardt B, Mundy J, Logemann J, Pinsdorf E, Leah R, Schell J, Maas C. 1995. Enhanced quantitative resistance against fungal disease by combinatorial expression of different barley antifungal proteins in transgenic tobacco. Plant J. 8:97-109. Talbot NJ, Ebbole DJ, Hamer JE. 1993. Identification and characterization of MPG1, a gene involved in pathogenicity from the rice blast fungus Magnaporthe grisea. Plant Cell 5:1575-1590. van Loon LC, Pierpoint WS, Boller T, Conjero V. 1994. Recommendations for naming plant pathogenesis-related proteins. Plant Mol. Biol. Rep. 12:245-264. Zhu Q, Maher EA, Masoud S, Dixon RA, Lamb CJ. 1994. Enhanced protection against fungal attack by constitutive co-expression of chitinase and glucanase genes in transgenic tobacco. Bio/Technology 12:807812.

2. Tissue print illustrating the localized transcription of the rice PR-10a gene during Magnaporthe grisea infection. A 2-wk-old (Co 39 rice seedling was inoculated with two adjacent 20-l drops of M. grisea conidial suspension (0.4% gelatin; 1,000 conidia drop -1). Following 72 h of incubation, the seedling was pressed onto a nylon membrane and hybridized to an antisense RNA probe specific for the rice PR10a transcript. Inset photo: a 4X magnification of the infection site.

Genes controlling field resistance to blast in Japanese upland rice detected using RFLP markers
S. Fukuoka, K. Okuno, National Institute of Agrobiological Resources, Tsukuba, lbaraki 305, Japan; M. Kawase, Shikoku National Agricultural Experiment Station, Zentsuji, Kagawa 765, Japan; K. Miura, Hokkaido National Agricultural Experiment Station, Sapporo, Hokkaido 062, Japan; and S. Kiyosawa, Tsukuba International Agricultural Training Center, Tsukuba, lbaraki 305, Japan

After observing the rapid breakdown of vertical resistance genes to blast, such as Pi-k, rice breeders are now developing

breeding lines with field resistance to blast. Field resistance is defined as the resistance that allows effective control of a parasite under natural field conditions. Japanese upland rice varieties are potential gene donors for field resistance. Studies using linkage analysis with conventional genetic markers have pointed to several different loci controlling field resistance to blast (Goto 1970, Shinoda et al 1971, Higashi and Saito 1985). We studied the chromosomal location of genes controlling field resistance to blast in Japanese upland rice using restriction fragment length polymorphism (RFLP) markers. RFLP is frequently detected among Japanese varieties. About 40% of RFLP

markers located on the rice genetic linkage map revealed RFLP between lowland and upland rice varieties using DNA probes polymorphic between lowland and upland varieties. Field resistance to leaf blast was genetically analyzed. Three hundred twelve F 3 lines from a cross between Nipponbare (moderately susceptible, lowland) and Owarihatamochi (resistant, upland) were field-tested for resistance to blast. Of these, 27 F 3 lines expressed resistance equivalent to that of Owarihatamochi while 27 lines were susceptible. Total DNA was extracted from leaves of resistant and susceptible F3 lines and their parents. Using 138 RFLP markers, we determined the relationship between


IRRN 21:2-3 (August-December 1996)

Genetic analysis for true resistance to blast in rice varieties

M. Yamaguchi, Tohoku National Agricultural Experiment Station, Omagari, Akita 014-01, Japan; H. Saitoh, National Agriculture Research Center (NARC), Tsukuba, lbaraki 305, Japan; H. Yaegashi, National Institute of Agro-Environmental Science, Tsukuba, Ibaraki 305, Japan; R. Ikeda, NARC; and T. Higashi, Tohoku National Agricultural Experiment Station

Location of the regions (loci) with significant correlation between genotype of RFLP markers and blast field resistance in Japanese upland rice.

markers and resistance/susceptibility in the F3 lines. We found a close relationship between blast resistance and genotypic frequency in eight regions of six chromosomes. One region was detected each on chromosomes 2, 6, 7, and 12; two regions were detected on chromosomes 4 and 11 (see figure). The loci linked to RFLP markers on chromosomes 4 and 11 played a major role in expressing field resistance, confirming the results of an earlier study (Wang et al 1994).

Cited references

Goto I. 1970. Genetic studies on the resistance of rice plant to the blast fungus 1. Inheritance of resistance in cross Sensho x H-79 and Imochi-shirazu x H-79. Ann. Phytopathol. Soc. Jpn. 36:304-3 12. Higashi T, Saito S. 1985. Linkage group of field resistance genes of upland rice variety Sensho to leaf blast caused by Pyricularia oryzae Cav. Jpn. J. Breed. 35:438-448. Shinoda H, Toriyama K, Yunoki T, Ezuka A, Sakurai T. 197 1. Studies on varietal resistance of rice to blast. 6. Linkage relationship of blast resistance genes. Bull. Chugoku Natl. Exp. Stn. A20: 1-25. Wang L, MacKill D, Bonman J, McCouch S, Champoux M, Nelson R. 1994. RFLP mapping of genes confering complete and partial resistance to blast in a durably resistant cultivar. Genetics 136:1421-1434.

Analysis of unknown blast resistance gene(s) of introduced or improved indica varieties showing high blast resistance is useful. We estimated the true resistance genes of indica varieties IR50, Suweon 288 (IR24//IR24/IR747), and Takanari (Milyang 42/Milyang 25) using the method of Toriyama et al (1983), in which backcrossed progenies are tested for reactions to blast isolates of different races. IR50 was crossed with Ouu308, a japonica line having only the Pi-a gene. Resulting F1 plants were backcrossed with parent Ouu308 to obtain B 1 F2 lines. We inoculated 141 lines of the B 1 F2 with blast isolate TH68-141 (race 003, compatible to Pi-a only) to estimate the number of true resistance genes involved. The ratio of the B1F2 lines fixed for susceptibility to the total B1F2 lines tested was 17.0%, indicating that IR50 has two or three true resistance genes, exclusive of Pi-a (Table 1). All 24 B1 F2
Table 1. Expected and actual ratios of B1F2 lines fixed for susceptibility to the total B1F2 lines tested for different numbers of true resistance genes involved (except for Pi-a). Genes involved (no.) Expected ratio Actual ratio IR50 Suweon 288 Takanari 2 25.0 17.0 13.0 3 12.5 4 6.3 5 3.1 6 1.6


lines fixed for susceptibility to TH68-141 showed resistance to blast isolate Naga6616 (race 001, incompatible to Pi-a gene), indicating that IR50 has the Pi-a gene. We selected 29 of the B1 F2 lines showing the segregation ratio of 3 resistant: 1 susceptible plants against TH68-141 and inoculated these with seven isolates of different races (Table 2). Of these, eight lines showed susceptibility to isolate TH87-20 (race 007b+, compatible to Pi-a, Pi-i, and Pi-b ) despite showing segregation against isolate Naga69-50 (race 007, compatible to Pi-a and Pi-i ). This indicates that IR50 also has the Pi-b gene. Based on the response of the lines to other isolates, it was difficult to estimate other true resistance genes in IR50 (Table 2). The results suggest that IR50 has the Pi-a, Pi-b, and one or two true resistance genes. Suweon 288 and Takanari were backcrossed with variety Nekken 2, which has the Pi-a and S-5 n (wide-compatibility gene to reduce hybrid sterility). We inoculated 100 lines of these B 1F2 with blast isolate TH68-141. The ratio of B 1F2 lines fixed for susceptibility to the total lines tested were 3.0 and 13.0%, respectively, indicating that Suweon 288 has four to six genes and Takanari has three genes, exclusive of Pi-a (Table 1). Both varieties were verified to have the Pi-a gene after inoculation with isolate Naga66-16, with 3 and 13 of the B 1F2 lines fixed for susceptibility to TH68-141, respectively. When seven isolates (Table 2) were used to check 28 and 50 of the B 1F2 lines showing the segregation against TH68- 141 respectively for the two varieties, only Takanari was proven to have the Pi-b gene based on the lines response to isolates Naga69-150 and TH87-20. The results suggest that Suweon 288 has Pi-a and four to six other genes. Takanari, on the other hand, has Pi-a, Pi-b, and two other genes. Some of the genes we failed to identify in this study are probably new resistance genes, including recessive ones.

Table 2. Blast isolates of different races used and the corresponding true resistance genes. Isolate (race) Compatible gene(s) for true resistance TH 68-141 (003) Pi-a Naga 69-150 (007) Pi-a Pi-i NAO -02 (033) Pi-a Pi-k Pi-k m Ina N86-95A (047) Pi-a Pi-i Pi-z TH 67-106 (103) Pi-a Pi-ta Ina 85-101 (303) Pi-a Pi-ta Pi-ta 2 5A (403) Pi-a Pi-z t TH 87-20 (007-b + ) Pi-a Pi-i Pi-b

IRRN 21:2-3 (August-December 1996)


Cited reference
Toriyama K, Ezuka A, Asaga K, Yokoo M. 1983. A method of estimating true resistance genes to blast in rice varieties by testing their backcrossed progenies for race-specific reactions. Jpn. J. Breed. 33:448-456.

To save on costs, the IRRN is now being published three times a year: April, August, and December. We have also discontinued the IRRN sections on news about research collaboration and announcements. The material previously covered in these sections is now included in the Rice Reporter.

Note to readers

Pest resistanceinsects
Preliminary evidence of aphid resistance in transgenic plants
Zhen Zhu, Zhaolan Zhou, Zhaoyang Deng, Yuefeng Gao, Yu Zhu, and Xianghui Li, Institute of Genetics, Academia Sinica, Beijing 100101, China

Sap-sucking insect pests (mainly Homoptera) can severely damage rice crops and also transmit virus diseases. Promising results have already been achieved with transgenic plants expressing the Bacillus thuringiensis (Bt) delta-endotoxin gene and proteinase inhibitor genes, which protect plants from Lepidoptera and Coleoptera insect pests, but are not effective against Homoptera insects. Information from this study. which describes pea lectin (P-lec) transgenic tobacco plants showing apparent aphid resistance, may be valuable for controlling rice Homoptera insect pests. Plant expression plasmid constructed. We cloned and identified the P-lec gene as described by Liu et al(1995). A HindIII/ SacI fragment from pBS-lecI, including 0.9kb P-lec gene, was inserted into mini Tiplasmid pK7-1, regulated by CaMV35S promoter in the 5'-end and rbcS Poly(A) signal in the 3'-end. The recombinant was

then transferred into Agrobacterium tumefaciens LBA4404 by electroporation. Plant transformation. Leaf discs of tobacco line NC89 were transformed via A. tumefaciens. Transformants were selected using kanamycin, and the transgenic plants were confirmed using Npt-II assay. Bioassay on transgenic plants.Young transgenic tobacco plants and control tobacco plants (4-5 leaves) were individually infested on the top leaves with 80 peach aphids (Myzus persicae). Aphid population density was calculated 4 and 6 wk after infestation and used to assign the resistance score (see table). We evaluated aphid resistance (divided into four resistance levels) based on density of aphid population on the top leaves. The ratio of plants with high resistance to aphids (scores 3 and 4) and the resistance mean in transgenic plants were higher than in the control (see table), indicating that the transgenic plants had better protection from aphids than the control (see figure). A dramatic difference exists in aphid resistance between the transgenic tobacco and the control, possibly caused by transgenic plants expressing P-lec. However, evidence indicating the effect of pea lectin against some sap-sucking insect pests fed

Leaves from transgenic tobacco plant (right) and from control tobacco plant. The photo was taken 6 wk after aphid infestation.

Bioassay of transgenic and control tobacco plants. Weeks after infestation b Score a Plants (no.) 1 2 3 4 Total 21 5 2 6 34 Control (%) 61.8 14.9 5.9 17.6 4 Plants (no.) 8 4 7 10 29 P-lec (%) 27.6 13.8 24.1 54.5 Plants (no.) 27 5 2 0 34 Control (%) 79.4 14.7 5.9 0.0 6 Plants (no.) 8 8 4 9 29 P-lec (%) 27.6 27.6 13.8 31.0

on an artificial diet is not available (Powell et al 1993). Possible explanations for this area) the effects of pea lectin on rice brown planthopper and rice green leafhopper used in Powells study differed from the effects on aphids, and b) the pea lectin could have lost its antimetabolic activity on Homoptera insects during the artificial diet preparation. The P-lec gene expression plasmid suitable for rice transformation has already been constructed, and the transformation of rice is under way.

Cited references
Liu CM, Yu ZY, Zhu Z, Zhou ZL, Xiao GF, and Li XH. 1995. Cloning and sequencing of the pea lectin gene. Acta Genet. Sin. 22(4):302306. Powell KS, Gatehouse AMR, Hilder VA, and Gatehouse JA. 1993. Antimetabolic effects of plant lectins and plant and fungal enzymes on the nymphal stages of two important rice pests, Nilaparvata lugens and Nephotettix cincticeps. Entomol. Exp. Appl. 66:119126.

a4 = plants where peach aphid population density on top leaves is less than 2 aphids cm 2 leaf: 3 = density is between 2 and 5 aphids cm -1 leaf; 2 = denslty is in the range of 1-50; and 1 = denslty of more than 10 aphids cm -2 b Data represent percentage of particular plants to total plants Infested.



IRRN 21:2-3 (August-December 1996)

Role of host plant resistance in successful control of brown planthopper in Central Luzon, Philippines
E. B. Medina, C. C. Bernal, and M. B. Cohen, IRRI

Planting a single rice variety over a large area year after year may promote the breakdown of rice resistance to brown planthopper (BPH) Nilaparvata lugens (Stl). This cropping scenario, when combined with other factors such as overuse of insecticides, may lead to BPH outbreaks. Most farmers in Central Luzon, Philippines, have been planting for almost 10 yr the popular variety IR64. Despite this situation, BPH populations in Central Luzon are extremely low. We conducted life table studies and seedbox resistance tests on four BPH populations from Central Luzon to determine if IR64 is still resistant to BPH in Central Luzon, in the process determining the extent to which host plant resistance is responsible for low and stable BPH populations in that region. We collected BPH in Oct 1994 from four sites, 5-50 km apart, and took the insects to IRRI to evaluate on four varieties: IR22 (no genes for BPH resistance), IR36 (bph2 gene), IR64 (Bph1 gene plus additional

minor gene(s)), and IR72 (Bph3 gene). Initially, we studied the data by analysis of variance using PROC GLM (general linear model) of the Statistical Analysis Software package, with site as a random effect. Two variables (developmental time and nymphal survival) showed a significant site-byvariety interaction; hence, means were not averaged across sites. Fecundity among female BPH reared on IR22 and IR64 did not differ significantly across populations (see table). In three populations, survival to the adult stage among BPH reared on IR22 and IR64 either did not differ significantly or survival was higher on IR64. In two populations, the developmental period of female BPH reared on IR64 was significantly longer than those reared on IR22. At three sites, IR64 had a lower damage rating than did IR22 in seedbox tests. Results indicate that IR64 maintains a slight-to-moderate resistance to BPH populations in Central Luzon. It is possible that the minor gene(s) in IR64 contribute to a greater durability of resistance in this variety. The results also suggest that high levels of host plant resistance are not necessary for successfull BPH management in areas, such as Central Luzon, where insecticide use is low and biological control is not disrupted.

Graphical genotypes of rice parental lines for resistance to green rice leafhopper
K. Tamura, Y. Fukuta, M. Hirae, K. Fukui, and S. Oya, Hokuriku National Agricultural Experiment Station, Joetsu, Niigata 943-01, Japan

Performance of BPH from Central Luzon, Philippines, on four rice varieties. a 1994 wet season. Population Matingkis Variety IR22 IR36 IR64 IR72 IR22 IR36 IR64 IR72 IR22 IR36 IR64 IR72 IR22 IR36 IR64 IR72 Fecundity b 623.90 574.20 477.88 403.00 467.60 482.00 435.20 271.80 596.40 459.90 490.10 321.10 357.10 398.80 474.90 317.60 52.24 a 40.30 a 83.81 a 52.84 b 36.19 a 56.95 a 56.66 a 59.76 b 55.10 a 60.03 ab 59.14 a 75.65 b 50.07 ab 47.04 ab 37.11 a 66.29 b Development time c 13.34 13.38 13.16 13.63 13.06 13.93 14.32 14.12 13.84 13.09 13.52 13.67 13.12 14.59 14.35 14.50 0.17 0.18 0.18 0.24 a a a a Nymphal survival d 13.00 13.38 14.77 10.00 11.80 11.20 13.90 12.20 16.20 14.70 13.30 11.30 13.80 12.20 12.20 9.80 0.89 abc 0.18 ab 0.83 a 0.93 c 1.03 0.83 0.82 1.04 a a a a Damage rating e 9.0 7.6 7.0 5.6 9.0 6.3 5.6 5.6 8.33 5.00 5.00 3.66 9.0 5.6 4.3 3.0 0 a 0.66 ab 0.66 bc 0.66 c 0 1.76 1.33 1.33 a a a a

Green rice leafhopper (GRLH), Nephotettix cincticeps, infects rice by transmitting viral diseases such as dwarf disease. Screening for resistance to GRLH has been conducted in Japan since the mid-1960s. The degree of resistance to GRLH was found to differ considerably among several indica varieties, but all Japanese rice varieties were susceptible (Ishii et al 1969). GRLH resistance was introduced from indica donors into japonica backgrounds. Norin PL5 was reported to express GRLH resistance by means of two complementary genes (Imbe and Iwasaki 1987) or by two dominant multiple genes (Takita and Nishiyama 1989). However, scientists have not yet determined chromosomal location of the resistance genes in those lines. Restriction fragment length polymorphism (RFLP) linkage map and RFLP markers have been developed in rice (Kurata et al 1994). Graphical genotype, as designated by Young and Tanksley (1989), is a powerful method for evaluating progeny strains generated from wide crosses. We studied the resistance to GRLH of genotype method and examined which regions of the rice chromosomes remain as indica domains of those parental lines. We used five lines (Norin PL2, Kanto PL6, Norin PL5, Norin PL6, and Aichi42) with resistance genes derived from indica varieties Pe-bi-hun, Tadukan, C203-1, Lepedumai, and Rantaj-emas 2, respectively (see table). Total DNAwas extracted from parental lines, original indica varieties, and three japonica varieties (Nipponbare, Reiho, Toyonishiki) using the cetyltrimethyl ammonium bromide method. The extracted DNA was digested with four restriction enzymes (Hind III, BamHI, Eco Rv, and BgIII) and separated by agarose gel (0.8%) electrophoresis. A nonradioactive method

five parental linesusing thegraphical


0.32 b 0.35 ab 0.45 a 0.16 a 0.16 a 0.23 b 0.23 ab 0.24 a 0.10 b 0.41 a 0.28 a 0.14 a

San Miguel

0.80 a 0.79 ab 0.73 bc 0.95 c 1.04 a 1.04 ab 0.90 ab 1.22 b

0.66 a 1.54 b 1.54 b 0.66 b 0 a 0.66 b 0.66 bc 0 c


a MeanSE. Within a column and site, means followed by the same letter are not significantly different (P>0.05) by LSD statistical test. b Number of eggs laid per female, n = 10. c Days for completion of female development, n=10. d Number of nymphs reaching adult stage (out of 20 nymphs female -1). n = 10 females. e 0-9 scale of the Standard

evaluation system for rice.

IRRN 21:2-3 (August-December 1996)


(ECL system: Amersham) was used for labeling and detection. Graphical genotypes of the five lines are shown in the figure. Several chromosome domains originating from each indica donor parent were detected in all lines. We found four domains in Norin PL 2 (chromosomes 5, 11, 12); six in Kanto PL6 (chromosomes 1, 2, 3, 4, 8, 12); seven in Norin PL 5 (chromosomes 1, 3, 4, 8, 9,11); two in Norin PL 6 (chromosomes 3, 11); and one in Aichi 42 (chromosome 6). We also detected two domains (on chromosomes 3 and 11) common to both Norin PL 5 and Norin PL 6. Some undetermined domains
Parental lines resistant to green rice leafhopper. Line Norin PL 2 (Kanto PL 3) Kanto PL 6 Cross combination a Pe-bi-hun /Kanto 98// 2*Kanto 100 Kanto 107/3/ Tadukan /Kanto 98 / / Kanto 100 / 4 / Nipponbare Reiho/ C203-1 //2*Reiho/3/ Saikai 137 Toyonishiki/ Lepedumai // Ou284/3/2*Toyonishiki Nipponbare/4/Alchi24/3/Kanto 105//Nipponbare*5/ Rantaj-emas 2

did not show polymorphism between indica donors and the three japonica cultivars. To determine these domains, other japonica varieties in those crosses should be tested further. Chromosomal regions that remained indica domains of the five parental lines were detected in the study. But because GRLH resistance genes may be located on those domains, we will conduct linkage analysis between resistance to GRLH and RFLP markers in a hybrid population to identify the gene loci.

Takita T, Nishiyama H. 1989. Selection of biotypes of green rice leafhopper and genetic analysis for the resistance in rice. Bull. Kyushu Natl. Agric. Exp. Stn. 25(3):51-259. Young ND, Tanksley SD. 1989. Restriction fragment length polymorphism maps and the concept of graphical genotypes. Theor. Appl. Genet. 77:95-101.

RFLP mapping of brown planthopper resistance gene Bph1 in rice

H. Hirabayashi and T. Ogawa, Kyushu National Agricultural Experiment Station, Chikugo, Fukuoka 833, Japan

Cited references
Imbe T, Iwasaki M. 1987. Inheritance of resistance to the green rice leafhopper Nephottettix cincticeps Uhler and dwarf disease in rice Norin PLS (in Japanese, English summary]. Jpn. J. Breed. 37:177-184. Ishii M, Yasuo S, Yamaguchi T. 1969. Testing methods and analysis of the varietal resistance to rice dwarf disease (in Japanese, English summary]. J. Cent. Agric. Exp. Stn. Jpn. 13:23-44. Kurata N, NagamuraY, Yamamoto K. Harushima Y, Sue N, Wu J, Antonio BA, Shomura A, Shimizu T, Lin Y-S, Inoue T, Fukuda A, Sinamano T, Kuboki Y, Toyama T, Miyamoto Y, Kirihara T, Hayasaka K, Miyao A, Manna L, Zhong HS, Tamura Y, Wang ZX, Momma T, Yan M, Sasaki T, Minobe Y. 1994. A 300-kilobase interval genetic map of rice including 883 expressed sequences. Nat. Genet. 8:365.372.

Norin PL 5 (Saikai PL 2) Norin PL 6 (Ou PL 1) Aichi42

a Indica varietles from which resistance genes were

derived are in boldface. Kanto 98, Kanto 100: Kochikaze/ Nipponbare.

Graphical genotypes of five rice parental lines. In each of the 12 chromosomes, the 5 strips stand for, from left to right, Norin PL 5, Norin PL 2, Kanto PL 6, Aichi 42, and Norin PL 6. Each bar beside the strips shows the position of 123 RFLP markers used.

The nine brown planthopper (BPH) resistance genes identified so far have not yet been located on the linkage map in detail. In this study, we determined the Bph1 locus using several restriction fragment length polymorphism (RFLP) markers. BPH-resistant indica variety IR28 (Bph1) was crossed with Koshihikari, a susceptible japonica variety. We tested 92 F3 lines of the cross for resistance in a mass screening using biotype 1 BPH and IR28 and Koshihikari. DNA was extracted from young leaves of the parents and F 3 lines using the cetyltrimethyl ammonium bromide method. The RFLP probes, except XNpb, were provided by NIAR/STAFF. The F3 lines from the cross Koshihikari/ IR28 segregated into 68:24 for resistant homozygous or heterozygous (R+H) and homozygous susceptible (S). Although susceptible lines died after infestation, we could not distinguish between resistant homozygous and heterozygous plants. This segregation showed a good fit to the expected 3(R+H):1 (S) ratio (c 2 = 0.058). We initially used several markers located on chromosome 4, based on the study of Ikeda and Kaneda (1983). that reported the location of Bph1 on chromosome 4 by trisomic analysis. However, our RFLP analysis showed that Bph1 was not linked to 11 markers on chromosome 4. We then used several RFLP markers located on chromosome 1. The bph2 gene, allelic or closely linked to Bph1, was linked to the d2 gene on chromosome 4 at a 39.4% recombination value (Ikeda 1985). Ideta et al (1984) reported that d2 was located on chromosome 1 using RFLP analysis.


IRRN 21:2-3 (August-December 1996)

Linkage analysis between Bph1 and RFLP markers on chromosome 12. Gene pair AB Bph1 Bph1 Bph1 Bph1 Bph1 Bph1 Bph1 Bph1 Bph1

Segregation mode in F3 A_BB A_Bb A_bb aaBB 20 61 20 19 18 21 20 21 18 0 0 42 42 42 40 38 33 36 41 5 6 6 6 6 9 14 7 1 7 0 0 0 0 1 2 2 aaBb aabb 0 0 4 5 5 5 6 9 14 24 17 20 19 19 19 17 13 8 10.9 39.1 54.5 48.6 48.6 49.1 33.0 12.5 6.6

c2 a

Recombination value (%) 18.4 11.5 10.7 11.9 11.9 12.1 18.9 30.3 32.7 10.1 3.6 3.4 3.6 3.6 3.6 4.5 5.6 5.9

Genetic map distance (cM) 19.3 11.7 10.9 12.1 12.2 12.3 19.9 35.2 38.1 10.3 3.6 3.4 3.6 3.6 3.6 4.5 5.6 6.0

G148 C185 XNpb248 XNpb304-1 XNpb319 XNpb304-2 XNpb336 XNpb316 G124-1

(<0.01) (<0.001) (<0.001) (<0.001) (<0.001) (<0.001) (<0.001) (<0.01) (<0.01)

Calculated based on the ratio of 3:6:3:1:2:1 (df 2).

However, Bph1 was not linked to the 21 RFLP markers on chromosome 1. We therefore used several RFLP markers located on the other 10 chromosomes to obtain the marker linked to Bph1. The results showed that Bph1 was linked to G148, C185, XNpb248, and XNpb304-1 on chromosome 12 at recombination values of 18.4, 11.5, 10.7, and 11.9%, respectively (see table). Thus, we conclude that Bph1 is located on chromosome 12. These loci are arranged as G148 - C 185 - Bph1 - XNpb248 - XNpb304- 1 - XNpb319 - XNpb304-2 (see figure). However, results of earlier studies (Ikeda and Kaneda 1983, Ikeda 1985) were inconsistent with our findings. Mistakes during trisomic analysis for insect and

disease resistance are possible. Weak and scanty trisomic plants that have the resistance gene may appear susceptible under resistance selection pressure. The recombination value of 39.4% between bph2 and d2 is also too distant to confirm a linkage relationship. Previous studies did not use marker genes on chromosome 12. Use of several RFLP markers on all chromosomes enabled us to determine the location of Bph1 on chromosome 12. Cited references
Idetd O, Yoshimura A, Iwata N. 1994. lntegration of convention and RFLP linkage map in rice. V. The locus ebisu dwarf (d2 ) [in Japanese]. Breed. Sci. 44 (suppl.2): 185.

Arrangement of Bph1 and RFLP markers on chromosome 12. The genetic distances are given in centiMorgan units.

Ikeda R. 1985. Studies on the inheritance of resistance to the rice brown planthopper (Nilaparvata lugens Stl) and the breeding of resistant rice cultivars [in Japanese]. Bull. Natl. Agric. Res. Cent. 3:1-54. Ikeda R, Kaneda C. 1983. Trisomic analysis of the gene Bphl for resistance to the brown planthopper, Nilaparvata lugens Stl., in rice. Jpn. J. Breed. 33:40-44.
Results of principal component analysis. Characteristica STI STI STI STI based based based based on on on on Ht RL FM DM Principal components 1 0.746 0.290 0.538 0.732 2.306 68.80 variability 68.80 2 0.009 0.634 0.281 0.010 0.934 14.30 83.60

Strees tolerancesalinity
Principal component analysis and variety classification in relation to rice seedling salinity tolerance
L. M. Gonzles, Soil Science and Agricultural Chemistry Department, Agricultural Research Institute Jorge Dimitrov, Bayamo 2360, Cuba

Salinity is one of the major contraints to rice production in Cuba; improved salt-tolerant varieties are essential for achieving economically important gains in yield. We evaluated the salinity tolerance of some varieties important to rice breeding programs in eastern Cuba. Distilled water (EC 0.02 dS m -1 ) was used as the laboratory medium and as a control. Salinity stress (EC 12 dS m -1 ) was

adjusted by adding NaCl. Twenty seeds were kept in each solution in petri dishes, with five replications. Water volume was maintained by adding solution. Root length, seedling height, and fresh and dry seedling matter were measured after 7 d. Salinity tolerance indexes (STI) were calculated using the equation STI (%) = 100 (IS/IC), where IS is the mean of saline solution indicator and CL is the mean of the control solution indicator. Principal component analysis showed that more than 83% of the total variability between varieties was accumulated in the first two components (see table). The STI (based on seedling height and seedling dry matter) in the first component and the STI (based on root length) in the second component appear to have the highest value of correlation with the principal axis, indicating the usefulness of these indexes in classifying varietal tolerance.

Eigenvectors Contribution to variability (%) Total

aSTI = salinity tolerance indexes. Ht = seedling height,

RL = root length, FM = seedling fresh matter, and DM = seedling dry matter.

Based on the first two components, cultivars were placed in four groups (see figure), where the variability in cultivar response to salinity was observed. Group I (Caribe 1, IA Cuba-26,2 196, Ecia 22-8, J-

IRRN 21:2-3 (August-December 1996)


112, 2006, IR42, and IR5931), which gave the highest STI values (seedling height = 79%, root length = 81%, and seedling dry matter = 71%), was the most salt-tolerant group.

Use of this method will improve efficiency and save on expenses in breeding programs for seedling tolerance because the most salt-tolerant cultivars can de determined and used.

Integrated germplasm improvementirrigated

Variability, heritability, correlation, path analysis, and genetic divergence studies in M 2 generation of gamma-irradiated upland rice
S. S. Mehetre, C. R. Mahajan, P. A. Patil, and P. M. Dhumal, Botany Section, College of Agriculture, Kolhapur 416004, Maharashtra, India

We studied the genotypic and phenotypic coefficients of variation, heritability, genetic advance (GA), coefficients of correlation, path analysis, and genetic divergence in 75 M2 families raised from

eight upland varieties treated with 10-, 20-, 30-, 40-, and 50-kR gamma rays. Analysis of variance showed significant differences among genotypes for all characters (Table 1). Considerable range of variation was expressed for percent sterility (43.36), tillers plant-1 (14.91), grain yield plant-1 (14.67), and spikelets panicle -1 (13.51), indicating better scope for selection for genetic improvement. Percent sterility had maximum genotypic coefficient of variation (32.23), followed by grain yield plant-1 (29.75), plant height (26.07), spikelets panicle -1 (19.94), tillers plant-1 (13.85), and days to flowering (12.26) and maturity (9.33).

Estimates of heritability ranged from 91.20% for plant height to 35.60% for sterility. Although plant height (91.20), and days to flowering (88.00) and maturity (87.00) had high heritability, these characters had low genotypic coefficients of variation (GCV) values. This might be due to the variation of environmental components involved in these traits. Expected GA ranged from 6.92% (panicle length) to 54.91% (grain yield plant-1). Grain yield plant-1, plant height, percent sterility, and spikelets panicle-1 showed high GA with high GCV values. These characters should therefore be considered in efforts to obtain high genetic gains. Grain yield plant-1 was positively correlated with tillers plant-1 but was negatively , and significantly correlated with days to 50% flowering (0.409) and maturity (0.421), plant height (0.717), panicle length (0.302), and percent sterility (0.771). Negative correlation of characters (tillers plant -1, panicle length, and spikelets panicle-1) might be due to the highest correlation (r 2 = 0.771) with sterility. Percent sterility was positively and significantly correlated with plant height (0.579) and panicle length (0.575). Panicle length was found significantly and positively correlated with days to 50% flowering (0.382) and maturity (0.370) and plant height (0.68). Both plant height and tillers plant-1 were significantly and positively correlated with days to flowering (0.416 and 0.376) and maturity (0.425 and 0.361). Days to 50% flowering had a highly significant and positive (0.999) correlation with days to maturity. Correlation and path analysis studies , revealed that filled grains panicle -1 plant height, and panicle length are important yield-contributing characters that must be considered when adopting selection criteria in an upland rice breeding program. Genetic divergence studies using D2 analysis showed that the 75 M 2 families had formed 14 genetically diverse groups (Table 2). Selection pressure applied on the M2 families is likely to provide targeted yield and desired combinations of yieldcontributing traits in future generations.


IRRN 21:2-3 (August-December 1996)

Table 1. Analysis of variance and genetic parameters of variation in genotypic and phenotypic coefficients of correlation and path analysis for 8 quantitative characters in M 2 generation of upland rice. df Source Analysis of variance Replication Treatment Error CD (5%) Genetic parameters of variationc Mean CV (%) Range (min) (max) Variance (P) (G) Coefficient (GCV) Variance (PCV) Heritability (BS) (%) Genetic advance Expected genetic advance Coefficients of correlation and path analysisd Days to 50% flowering rp rg Days to maturity pcg pcp Days to maturity rp rg pcg pcp Plant height (cm) rp rg pcg pcp Tillers plant-1 (no.) rp rg pcg pcp Panicle length (cm) rp rg pcg pcp rp rg pcg pcp rp rg pcg pcp rp rg pcg pcp

Days to 50% flowering

Days to maturity

Plant height (cm) 2.94 1382.29** 63.41 15.60 98.48 8.08 69.40 161.10 122.85 659.44 27.30 26.07 91.20 50.51 51.28 0.411 0.466** 0.187 (0.575) 0.420** 0.425** 0.191 (0.588)

Tillers plant-1 (no.)

Panicle length (cm)

Spikelets panicle -1 (no.) 188.25 1501.70** 280.49 32.82 123.93 13.51 78.60 259.60 891.10 610.60 24.08 19.94 68.50 42.12 33.99 0.297* 0.288** 0.017 (0.121) 0.295* 0.281 0.017 (0.118) 0.261** 0.265 0.015 0.112 0.136 0.142 0.008 (0.060) 0.212 0.136 0.012 (0.057)

Sterility (%)

Grain yield plant -1 (g)

1 44 44

26.06 313.42**a 20.03 8.77 98.76 4.53 71.00 121.00 166.73 146.69 13.08 12.26 88.00 23.40 23.69

36.00 303.17** 21.00 8.98 127.31 3.59 100.00 156.00 162.09 141.09 10.00 9.33 87.00 22.81 17.92 0.998 0.999**

0.17 10.87** 3.99 3.91 13.39 14.91 9.00 19.40 7.44 3.44 20.37 13.85 46.30 2.60 19.42 0.419 0.376** 0.001 (0.931) 0.140 0.361** 0.001 (0.894) 0.288 0.284 0.002 (0.704)

0.78 3.55** 1.39 2.31 20.51 5.74 16.80 26.10 2.47 1.08 7.66 5.07 43.90 1.42 6.92 0.196 0.382** 0.012 (0.807) 0.186 0.370** 0.011 (0.782) 0.480** 0.658** 0.029 (1.392) 0.013 0.128 0.001 (0.271)

2.74 10.72*b 5.09 4.42 5.20 43.36 1.49 14.58 7.91 2.81 54.08 32.23 35.60 2.06 39.62 0.126 0.078 0.035 (0.161) 0.134 0.088 0.037 (0.178) 0.365** 0.579** 0.102 (1.173) 0,010 0.222 0.003 (0.457) 0.084 0.575** 0.241 (1.165) 0.025 0.150 0.007 (0.303)

9.08 45.05** 4.88 4.33 15.06 14.67 5.15 30.00 24.96 20.08 33.17 29.75 80.40 8.27 54.91 0.379** 0.409** 0.393** 0.421** 0.265** 0.717** 0.118 0.147 0.207 0.302** 0.201 0.130 0.473** 0.771**

1.584 (52.458)

1.878 (50.593)

1.741 (50.634)

1.581 (52.416)

0.455 (1.383)

0.651 (21.815)

0.730 (21.523)

0.009 (2.479)

0.237 (19.705)

(0.243) (18.260)

(0.104) (0.393)

Spikelets panicle -1 (no.)

0.061 (2.114)

0.310 (20.028)

0.324 (18.731)

0.219 (0.910)

0.000 (0.317)

Sterility (%)

0.058 (0.420)

0.471 (15.105)

0.513 14.208

0.119 (0.367)

0.001 10.351

0.013 (0.288)

Grain yield/plant (g)

0.280 (2.027)

0.200 (4.156)

0.233 4.444

0.166 (0.800)

0.000 (0.559)

0.005 (0.215)

0.001 (0.063)

= significant at the 1% level. b* = significant at the 5% level. c GCV = genotypic coefficient of variation; PCV = phenotypic coefficient of variation: BS = broad sense. d Figures in parentheses indicate genotypic and underlined figures indicate direct effects; residual effect = 1,354 rg, rp, pcg and pcp = genotypic correlation, phenotypic correlation, genotypic path and phenotypic path coefficient, respectively.


IRRN 21:2-3 (August-December 1996)


Table 2. Distribution of 45 M2 families of 8 upland rice varieties treated with different irradiation doses and cluster means for different characters studied. Cluster no. Treatments (no.) Variety Treatment Gamma ray doses (kR) 10 50 30, 20, 10, 50, 30 Days to 50% flowering (I) 98.3 (12.20) Maturity (II) 123.35 Plant height (cm) Tillers plant -1 (no.) Panicle length (cm) (V) 20.16 Spikelets panicle -1 (no.) (VI) 106.65 Sterility (%) Yield plant -1 (g) (VIII) 17.05 (IX) (X) (XI) (XII) (XIII) (XIV)

(III) 76.34

(IV) 14.12

(VII) 5.77


Jaya JS180 R24 Kundalika ACK5 Ghansal R24 HS17 Kundalika Jaya Kundalika Ghansal JS180 HS17 Basmati Basmati Ghansal Basmati Ghansal JS180 R24 R24 R24 Basmati Basmati


50 30, 40, 50 20, 30, 40, control


5 5 3

20 30, 40, control Control 20, 30, 40, control 10 10, control 10, 20, 30, 40 10, 20 10 30, 40 40, 50 50 20 Control 10 20 Control 20 Control

89.73 (13.63)

118.67 (19.26)







103.46 (27.52) 109.74 (101.97) 86.42 (117.75) 92.00 (66.03) 79.33 (80.20) 108.00 (101.71) 123.50 (132.60) 102.67 (18.75) 99.50 (56.86) 95.00 (29.18) 95.75 (154.12) 121.00 (88.45)

131.75 (47.28) 147.81 (130.12) 115.00 (53.62) 121.50 (62.28) 108.50 (46.67) 136.38 (101.10) 152.50 (207.45) 130.34 (46.73) 129.00 (41.47) 123.50 (48.26) 124.00 (134.13) 150.50 (134.12)

84.56 (14.07) 140.86 (121.15) 128.83 (131.35) 115.50 (99.58) 89.47 (123.03) 139.70 (108.53) 153.90 (193.08) 84.50 (40.45) 89.00 (44.45) 78.00 (48.53) 132.60 (206.10) 115.60 (112.03)

15.03 13.25 (15.65) 11.53 (155.88) 13.15 (54.42) 11.65 (148.80) 16.00 (27.45) 11.75 (40.96) 11.47 (136.58) 13.00 (82.27) 20.30 (109.14) 30.90 (57.07) 10.60 (48.35)



1.75 7.34 6.30


21.66 22.27 (19.29) 21.60 (68.06) 19.33 (46.88) 20.98 (93.86) 22.50 (222.77) 21.34 (120.47) 22.10 (107.19) 18.40 (118.15) 22.69 (142.60) 17.30 (183.24)

155.48 118.47

7.65 15.68


2 3

106.50 (23.88) 107.20 (53.76) 173.95 (47.62) 38.08 (74.69) 210.96 (124.50) 116.10 (44.69) 71.30 (74.11) 101.85 (13.18) 98.80 (79.90)

6.91 7.87 (15.46) 4.24 (110.96) 28.65 (207.52) 3.68 (112.18) 2.76 (64.66) 17.73 (74.41) 6.27 (128.98) 3.85 (140.27)

11.36 10.60


1 1 1 1 1 1 1

9.17 (0.00) 8.76 (93.05) 14.41 (103.89) 12.46 (38.84) 15.39 (88.25) 15.06 (64.44) 7.51 (75.72)

(0.00) (257.53) (14.88) (180.88) (44.24) (86.45)

(0.00) (69.50) (65.05) (238.28) (164.27)

(0.00) (50.33) (118.45) (61.92) (0.00) (152.75) (117.37)

(0.00) (111.74)


RCPL-32 and RCPL-3-6: two promising rice lines for the mid-hills of Sikkim, India
M. Singh, G. Singh, L. S. Srivastava, and R. P. Awasthi, Indian Council for Agricultural Research Complex for NEH Region, Tadong, Gangtok, Sikkim; H. S. Gupta, ICAR Research Complex for NEH Region, Meghalaya; and A. C. Sharma, CAR Research Complex for NEH Region, Imphal, Manipur, India

Table 1. Grain yield (t ha -1) of promising lines and checks over years and locations in Sikkim, India. Location Tadong Tadong Tadong Tadong Tadong Pakyong Pakyong Dharmdin Nazitam Nazitam Year 1990 1991 1992 1993 1994 1992 1993 1992 1992 1993 RCPL-3-2 5.0 5.0 4.1 4.3 4.2 4.0 3.9 6.0 4.9 4.9 RCPL-3-6 4.3 4.5 4.6 4.4 4.4 4.6 3.5 5.5 4.6 4.6 DR-92 3.8 2.9 2.5 3.0 3.6 3.3 2.6 3.8 3.5 3.2 Attey 2.0 1.2 1.3 1.8 1.4 1.9 2.2 2.2 2.1 2.0

Rice is the second most important crop in Sikkim, India covering about 19,000 ha. Yields average 1.3 t ha-1. In Sikkim, rice is cultivated on terraces using gravity flow water. Commonly grown local cultivars (Attey, Chirkey, Krishnabhog, Tapre, Kalonunia, Tulasi, Champasare, and Bhuinphool) are late-maturing, tall, and low-tillering. Their yields are poor but they have good quality attributes. Suitable substitutes for these cultivars must be semitall, medium-maturing, nonlodging, have low photoperiod sensitivity, high yield potential, and good quality. After 4 yr of evaluation, lines RCPL-3-2 and RCPL-3-6 outyielded all other lines and checks (Table 1). RCPL-3-2, a selection from cross Moirangphou/Jaya, is semidwarf (98-105 cm), while RCPL-3-6, a selection from cross CH-988/IR24, is also semidwarf but slightly taller. RCPL-3-2 is nonglutinous, whereas RCPL-3-6 is a semiglutinous rice. Both are medium in maturity and resistant

Table 2. Important characteristics of RCPL-3-2 and RCPL-3-6. Characteristic Plant height (cm) Days to flowering Days to maturity Ear-bearing tillers (no.) Panicle length (cm) Flag leaf Awn type Husk grain color Kernel color 1,000-grain weight (g) Grain length (cm) Grain width (cm) L-B ratio Endosperm type Grain yield (t ha -1) Reaction to leaf blast (0-9 scale) Reaction to neck blast (%) RCPL-3-2 98-105 60-70 115-120 9-15 20-23 Erect Awnless Straw Dull 25.4 0.83 0.30 2.77 Nonglutinous 4.5-5.0 0-2 0-1 RCPL-3-6 110-115 67-72 120-125 8-14 25-27 Erect Awnless Straw Dull 23.5 0.84 0.28 2.94 Semiglutinous 4.2-4.6 0-2 0-1

to lodging, leaf and neck blast. Distinguishing traits for both include panicle length, 1,000-grain weight, L-B ratio, panicle-bearing tillers. and endosperm type (Table 2).

RCPL-3-2 and RCPL-3-6 are suitable for early sowing, and will enable farmers to adopt more profitable rice-based cropping systems such as rice-mustard/pea/barley, that promise to increase cropping intensity under rainfed conditions.

Tooag92: a shortduration rice cultivar for Turkey

N. Aikgz and M. N. Gevrek, Field Crops Department, Faculty of Agriculture, Ege University, Bornova, Izmir, Turkey

Table 1. Comparison of some agronomic characters of Toag92 and commercial varieties. Bornova, Turkey (3-yr av) . Variety Yield (t ha -1) 5.3 4.1 4.5 4.4 Height (cm) 64 92 99 101 1,000grain weight (g) 33 38 33 31 Yield (kg d -1) 5.95 3.66 4.24 4.40 Duration (d) 89 112 106 100 Sterility (%) 16 24 30 24 Leaf blast a

Toag92, a rice variety derived from the cross Nucleoryza/Labella, was released in 1992. The main aim of the breeding program was to create a variety that does well in the second cropping season in the Aegean and Mediterranean regions of Turkey. Toag92, however, also has high yields in northern Turkey, where it did particularly well when sown late and in areas where existing varieties cannot be grown because of their long vegetative

Toag 92 Baldo Ribe Krasnodorsky 424

0 8 4

aScored using the Standard evaluation system (SES) 0-9 scale where 0 = highly resistant and 9 = highly susceptible.

durations. Toag92 matures 2 wk earlier than Ribe. Its low sterility under Mediterranean conditions makes it promising for the southeastern Anatolia region, where August temperatures are quite high.

Toag92 is of medium height (88 cm), 30 cm shorter than Ribe, and is resistant to lodging. The yield d-1 of Toag92 is almost 50% more than that of other improved cultivars. Toag92 is also resistant to leaf blast (Table 1).

IRRN 21:2-3 (August-December 1996) 59

Toag92 has medium grain and is free of white belly. Quality is good and similar to that of Baldo, Ribe, and Krasnodorsky-424 (Table 2).

Table 2. Quality traits of Toag92. Bornova, Turkey. Variety Toag92 Baldo Ribe Krasnodorsky 424 Amylose content (%) 18.8 16.8 15.6 Protein content (%) 13.7 9.8 7.8 7.4 KOH reaction White belly a

Low Low Low Low

5 0 5

aScored using the SES scale where 0 = no belly. 9 = more than 20% of kernel area.

KK1536-C: a modern high-yielding rice variety for irrigated lowlands in Papua New Guinea
M. S. Sajjad, Agriculture and Livestock Department, Food Management Branch, Erap Research and Development Centre, P. O. Box 1984, Lae, Papua New Guinea

Out of 122 genotypes we evaluated in 1991, KK15-36-C (IR 5657-33-2/BKNDR10317, Nepalese in origin) performed the best

and was selected as one of six genotypes for further evaluation. We confirmed the high yield potential (10.5 t ha-1) of KK15-36-C in local and regional yield trials conducted during 199295. At the OISCA Farm in Kokopo, Rabaul, KK15-36-C and Ayung yielded the most of the varieties tested. At Bubia during 1991 and 1995, the variety was either highest yielding (6-7 t ha-1 during 1994, 9.5 t ha-1 during 1995) or at par with the highest yielding genotypes.

The variety is semidwarf (<110 cm) with medium-tillering ability. It has long (>25 cm), fully filled panicles with more than 100 grains panicle-1. The 1,000-grain weight is 30 g. Its foliage remains green up to late maturity. The varietys grains are long and medium bold, awned and translucent. The milling recovery is very good. KK 15-36-C is extremely well suited for general cultivation in the irrigated lowland fields of Papua New Guinea.

Birsa Dhan 201 and Birsa Dhan 202: early-maturing varieties for Bihar, India
M. P. Singh, D. N. Singh, and M. F. Haque, Plant Breeding and Genetics Department, Birsa Agricultural University, Ranchi 834006, India

Performance of Birsa Dhan 202 in the All-India coordinated trials. 1989 and 1990. Mean of different locations Variety Yield (t ha -1) Birsa Dhan 202 Vikas Ratna Rasi Local (control) 6.3 6.1 6.1 6.2 1989 Days to 50% flowering 101 99 95 99 Yield (t ha -1) 3.5 2.6 3.1 1990 Days to 50% flowering 94 85 92

Birsa Dhan 201 was developed by crossing TN1/Brown Gora 23-19, and Birsa Dhan 202 by crossing Jaya/BR34. Both varieties were released during 1985 and re-released in 1995 for use in the Chotanagpur and Santhal Parganas region of Bihar, India. Birsa Dhan 201 is an early-duration variety that matures in 105-110 d under transplanting. It is nonlodging, nonshattering, and highly responsive to chemical fertilizers, yielding 3.0-3.5 t ha-1. It has compact, long panicles with bold grains and white kernels. It is moderately resistant to

blast, bacterial leaf blight, stem borer, and gall midge. Birsa Dhan 202 is intermediate in plant type and provides adequate rice straw for cattle feed, as required by farmers in the region. The variety has very good seedling vigor and produces heavy, drooping panicles. It is stiff-strawed, making it resistant to lodging. The grains are long and bold. producing white kernels. It is moderately

resistant to blast, stem borer, and bacterial leaf blight, but moderately susceptible to gall midge. The variety yields reasonably well, producing 5-6 t ha-1 when 40-60 kg N ha-1 is applied (see table). A lot of this region has been planted to Birsa Dhan 202. Farmers have been sharing its seed with each other, increasing the area planted to it. Both varieties have acceptable taste. and grain and cooking quality.


IRRN 21:2-3 (August-December 1996)

Integrated germplasm improvementrainfed lowland

Shakuntla; a rice variety for rainfed lowlands in Bihar, India
Year Site Pusa Sabour Chinsurah Ghaghraghat Ranital Canning Patna Pusa Patna Pusa Sabour Patna Pusa Sabour Bikramganj Chinsurah Masodha Patna Pusa Raipur Titabar Chinsurah Masodha Patna Pusa Raipur Titabar Pooled mean
a ns = not significant.

Table 1. Grain yield of Shakuntla at different sites in India. 1990-94. Yield (t ha-1) Shakuntla 2.2 3.1 2.7 2.2 3.4 2.8 2.5 2.6 4.1 2.8 1.7 4.8 2.7 1.9 4.6 3.0 6.5 5.2 5.5 2.8 5.5 5.1 2.5 3.6 2.2 4.6 2.6 3.5 Mahsuri or Utkal prabha 1.4 2.4 2.6 1.2 2.8 2.1 1.4 2.2 3.4 2.6 1.1 3.9 1.7 0.5 3.3 0.9 1.0 0.4 0.5 0.4 0.5 0.6 0.3 0.1 nsa 0.8 0.1 0.6 0.8 0.5 27.1 15.0 10.9 7.1 8.1 10.8 15.3 7.9 24.0 23.3 16.3 16.2 20.8 5.7 10.7 9.0 17.4 6.9 27.9 13.5 9.3 27.6 6.4 32.1 LSD (5%) CV (%)

V. N. Sahai, B. K. Singh, and S. Saran, Agricultural Research Institute, Mithapur, Patna 800001, Bihar, India; R. C. Chaudhary, IRRI

Shakuntla (RAU73-16-1-40) was released in 1995 for cultivation in rainfed lowland areas of Bihar, India. Rainfed lowland rice is grown on about 40% (2.2 million ha) of the states area. The current yield averages less than 1.0 t ha-1 . Developed from the cross Pankaj/Br8, Shakuntla is weakly photoperiod-sensitive, of long duration (140- 145 d), and is medium in height. Its grains are long and slender. In the All India coordinated yield trials, Shakuntla ranked fifth in 1990 and first in 1991. In state uniform variety trials during 1993 and 1994, Shakuntla yielded 12.350.5% more than check Mahsuri. In the 1994 eastern India rainfed lowland shuttle breeding trial, the variety was tested using normal planting with 30-d-old seedlings and delayed planting with 60-d-old seedlings. The yield advantage was greater under normal than delayed planting (Table 1). From 1988 to 1990, Shakuntla outperformed check Mahsuri by an average of 20% under direct seeding and transplanting (Table 2).





Normal planting with 30-d-old seedlings 3.4 0.4 0.9 5.0 0.2 6.7 0.5 3.0 0.7 2.3 0.9 3.4 Delayed planting with 60-d-old seedlings 0.6 3.3 2.1 1.0 0.3 2.2 0.2 1.3 nsa 2.2 nsa 2.2 2.6

Table 2. Grain yield (t ha -1) of Shakuntla in farmers' fields under direct seeding and transplanting. ICAR-IRRI collaborative research, 1988-90 wet seasons. Variety Direct seeding 1988-90 3.5 2.9 Transplanting 1989-90 3.7 3.0 Av Increase over check (%) 20

Shakuntla Mahsuri (check)

3.6 3.0

Manuscript preparation. Arrange the note as a brief statement of research objectives, a short description of project design, and a succinct discussion of results. Relate results to the objectives. Do not include abstracts. Do not cite references or include a bibliography. Restrain acknowledgments Manuscripts must be in English. Limit each note to no more than two pages of double-spaced typewritten text. Submit the original manuscript and a duplicate, each with a clear copy of all tables and figures. Authors should retain a copy of the note and of all tables and figures.

IRRN 21:2-3 (August-December 1996)


CMS lines for shallow rainfed lowland situation

S. B. Pradhan and P. J. Jachuck, Central Rice Research Institute (CRRI), Cuttack 753006, Orissa, India

Agronomic characters of CMS lines and their maintainers developed at CRRI, Cuttack, Orissa, India. 1994 wet season. CMSa or maintainer lines Moti A Moti B Kalashree A Kalashree B Padmini A Padmini B IET11350 A IET11350 B IET10428 A IET10428 B IET11668 A IET11668 B BKS64 A BKS64 B Manipur A Manipur B Mirai A Mirai B Blazin A Blazin B IET10983-I A IET10983 B IET10983-ll IET10983 B Tharrangsong A Tharrangsong B Pedigree of A lines Pollen sterility (%) 100 100 100 100 100 100 100 100 100 100 100 100 100 Spikelet fertility (%) 0 87.9 0 82.9 0 89.5 0 82.5 0 90.0 0 74.4 0 86.7 0 82.5 0 80.6 0 74.8 0 80.4 0 81.5 0 78.7 Plant height (cm) 90 120 95 113 119 145 95 100 104 105 93 102 85 99 82 92 79 88 70 126 75 82 65 85 71 75 Days to 50% flowering 127 124 134 130 119 118 97 95 98 90 96 95 94 95 100 99 100 99 95 102 100 97 98 97 88 85 Panicle exsertion (%) 73.1 100 68.7 100 80.3 100 80.4 100 79.3 100 81.3 100 77.8 100 67.3 100 70.1 100 66.1 100 66.6 100 62.3 100 69.3 100

V20A/Moti (WA) V20A/Kalashree (WA) V20A/Padmini (WA) V20A/IET11350 (WA) V20A/IET10428 (WA) V20A/IET11668 (WA) V20A/BKS 64 (WA) Annada A/Manipur (WA) Krishna A/Mirai (Kalinga-I) Krishna A/Blazin (Kalinga-I) IR62829A/IET10983 (WA) lR66707 A/IET10983 (O. perennis) lR66707 A/Tharrangsong (O. perennis) -

We are working to develop parental lines for hybrid rice for the rainfed lowland ecosystem in eastern India. Although hundreds of cytoplasmic male sterile (CMS) lines have been developed for breeding hybrids for the irrigated ecosystem in India, hardly any have been reported so far for the rainfed lowlands. We therefore converted 13 popular lowland cultivars into CMS lines through repeated backcrossing with three male sterility sources: wild abortive, Kalinga-I, and O. perennis. These sources, in BC6 generation, were grown in the field along with their maintainers during the 1994 wet season. We recorded pollen sterility, spikelet fertility in bagged panicles, plant height, days to 50% flowering, and panicle exsertion for each of the CMS and maintainer lines. All 13 CMS lines were completely pollen sterile with white anthers and did not set any seed in bagged panicles (see table). Moti A, Kalashree A, and Padmini A were late-duration varieties while the rest were medium-duration. The CMS lines are semidwarf to semitall in stature, with stiff straw. All were shorter (1-56 cm) than their respective isonuclear

aCMS lines are in the BC generatlon. 6

maintainers. Their plant heights and durations are considered ideal for use in devel-

oping hybrids for the shallow rainfed lowland situation in India.

Integrated germplasm improvementupland

Majhera 7, a direct seeded rainfed upland spring rice for Uttar Pradesh, India
B. V. Singh, A. Prasad, V. P. Singh, and C. S. Prasad, G. B. Pant University of Agriculture and Technology, Research Station, Majhera, Nainital, U. P. 263135, India

Spring rice is a major rainfed crop in the uplands of Uttar Pradesh, India. It is direct dry seeded from mid-March to mid-April and harvested in September-October. Only two spring rice varieties had been released to date: Majhera 3 in the early 1960s and VL 206 in 1983. Varieties grown in this

region must have long duration and drought tolerance, especially during May and June. To address this need, we developed Majhera 7 from germplasm collected from the Kumaun uplands. After numerous trials (Table 1), Majhera 7 was recommended for release as a commercial variety in the Uttar Pradesh hill zone under direct dry seeded rainfed upland conditions. In agronomical evaluation trials involving different sowing dates and fertilizer doses, Majhera 7 yielded more than checks Majhera 3 and VL 206. Maximum yield for the variety was obtained using a 7 Apr sowing date. Majhera 7 yielded 23.9% more than Majhera 3 and 6.3% more than

VL 206 when 75 kg N and 11 kg P ha-1 were applied. Majehra 7 matures in 174-187 d and fits well in common crop rotations: spring rice wheat - ragi - toria - fallow and spring rice wheat - barnyard millet - barley. It is 95105 cm tall, erect, and with broad, darkgreen leaves. It has medium-bold grain, a length-breadth ratio of 2.24. and a 21 g thousand-grain weight. Milling recovery is 80%. Majhera 7 is widely adaptable and shows field resistance to stem borer, leaffolder, grasshopper, sucking insects, and blast. (See Table 2 for morpho-agronomic characters of Majhera 7, Majhera 3, and VL 206.)


IRRN 21:2-3 (August-December 1996)

Table 1. Performance of Majhera 7 and checks in various trials. Uttar Pradesh, India. 1991-94. Mean yield (t ha -1 ) in Uttar Pradesh uplands Majhera 7 VL 206 (check) 3.5 2.0 1.0 2.1 1.5 1.5 1.6 1.5 3.0 3.0 1.2 1.8 2.1 1.7 1.7 2.0 1.7 1.1 2.0 2.0 1.7 66.6 22.2 9.5 22.4 17.6 15.0 35.3 81.8 10.0 15.0 29.4 Majhera 3 (check) 2.9 1.8 0.8 1.8 1.5 1.8 1.6 1.6 Yield increase (% ) over checks VL 206 22.8 10.0 50.0 23.8 40.0 33.3 12.5 26.6 Majhera 3 48.3 22.2 87.5 44.4 40.0 11.1 12.5 18.7 -

Improved upland rice for the hillsides of Colombia

A. M. Moreno-B, Centro Nacional de Investigaciones de Caf (CENICAFE). A. A. 2427, Manizales, Colombia; E. P. Guimaraes, M. Chatel, and J. Borrero, Centro International de Agricultura Tropical (CIAT), A. A. 6713, Cali, Colombia


Standard varietal trials 1991 4.3 1992 2.2 1993 1.5 Mean Station trials 1992 1993 1994 Mean Adaptive trials 1992 1993 Mean 2.7 2.1 2.0 1.8 1.9 2.5 3.0 -

Demonstration at research station 1992 2.0 1993 2.2 1994 2.3 Mean 2.3

Demonstration in farmers' fields 1994 2.0 Total mean 2.3

Table 2. Morphoagronomic character of Majhera 7 and check varieties Majhera 3 and VL 206. Genotype Days to Days to flowering maturity 1,000Plant Panicle Panicles m -2 Grains height length (no.) panicle-1 grain (cm) (cm) (no.) weight (g) 104 104 103 21.8 22.1 22.3 198.3 227.7 215.8 115.2 116.9 11.4 19.3 20.9 20.5 Milling Grain recovery color (%) 78.0 80.0 81.0 Brown White White

Majhera 3 Majhera 7 VL 206

139 139 139

174 176 175

Manuscript preparatlon. Arrange the note as a brief statement of research objectives. a short description of project design, and a succinct discussion of results. Relate results to the objectives. Do not include abstracts. Do not cite references or include a bibliography. Restrain acknowledgments. Manuscripts must be in English. Limit each note to no more than two pages of double-spaced typewritten text. Submit the original manuscript and a duplicate, each with a clear copy of all tables and figures. Authors should retain a copy of the note and of all tables and figures.

The Andean mountain range runs across Colombia from south to north, rising up to almost 6,000 m above sea level (masl). Coffee, grown by small-scale farmers, is the most important agricultural crop in the middle altitudes. Once the coffee is planted, it takes at least 3 yr to start commercial production. In the meantime, farmers use considerable resources to control weeds and prevent erosion. CENICAFE has been working to develop different cropping alternatives to help farmers have income before coffee starts to produce. CENICAFE, together with the CIAT Rice Program, worked to identify upland rice germplasm suitable for these hillsides. In 1993, we evaluated 21 upland lines at three locations, each replicated three times. in the heart of the coffee-growing area at 1,300 masl. Average temperature ranges from 20.6 to 23.1 C. The monthly average maximum (28.5 C) occurs in February and the minimum (16.9 C) in September. So germplasm must be cold tolerant. Breeding lines were selected for this trial based on results in Africa, where some of their parents performed well. Only results of the performance of the top six lines grown in trials in La Catalina, Risaralda state, are reported in the tables. The percentage of empty grains ranged from 12 to almost 100%, indicating that the germplasm presents variability for the trait (data not shown). Selection concentrated on the lines with at least 60% fertility. The average grain yield in those lines was higher than originally expected, ranging from 3.8 to 5.6 t ha -1 (Table 1). Even though the yield was relatively high, little is known about agronomic management under hilly conditions in Colombia. No checks were used because rice was never before planted in this area. The growth duration of the lines was extended to around 150 d after sowing, compared with 120 d under acid soil

IRRN 21:2-3 (August-December 1996)


Table 1. Average data from the best lines in the 1993 observational trial conducted in the coffeegrowing area of Colombia, Naranjal Experimental Station, Risaralda. Line Panicles Tillers (no.) (no.) Grains Filled (no.) CT10037-9-4-M-4-8P-1-M CT6196-33-11-1-3-M CT9997-5-3-M-4-M CT10069-27-3-1-4 CT10037-9-7-M-1-M CT10037-30-3-M-1-2P-2-M

Sterile (%) GT

Quality traits a WB GL Disp

Empty (no.) 123 99 136 66 176 78

Yield (t ha-1)

113 120 120 132 120 116

141 143 135 160 155 124

376 348 345 526 512 433

33 28 39 12 34 18


0.2 0.6 0.4 2 0.2 0.2


4 5 3.6 3.4 5 5

4.4 4.0 3.8 5.2 5.6 3.8

GT = gelatinization temperature, WB = white belly, GL = grain length, and Disp = dispersion.

Table 2. Yield ability, tiller number, and sterility percentage of the six breeding lines evaluated. a Naranjal Experimental Station, Risaralda, Colombia. 1994. Line CT10037-9-4-M-4-8P-1-M CT6196-33-11-1-3" CT9997-5-3-M-4-M CT10069-27-3-1-4 CT10037-9-7-M-1-M CT10037-30-3-M-1-2P-2-M Yield (t ha -1 ) 3.6 bc 3.3 c 3.2 c 4.3 ab 4.7 a 3.2 c Tillers (no.) 82.53 85.00 84.01 88.13 90.85 76.05 bc ab abc ab a c Filled grain (%) 81.9 81.0 58.0 87.5 83.0 82.5 a a a a a

a Means in a column followed by the same letter are not significantly different (P = 0.05) by DMRT.

conditions at 700 masl. Rainfall during the cropping season ranged from 128 mm in

March, the end of the growing season. to 394 mm in January.

CT10069-27-3-1-4 was selected in 1993 as one of the most promising lines, so agronomic evaluation was done. A simple unreplicated trial was conducted. Row planting produced a higher yield than spaced or hill planting, independent of the N level (0, 60 kg N ha -1) and seed density (60, 80, 100 kg ha-1). The rice responded to N at all densities, and the ideal seed density was 80 kg ha+ (data not shown). The best six lines were evaluated again in 1994 in a yield trial in a randomized block design with three replications (Table 2). The yield potential of these lines was confirmed, with the highest yielding line (CT10037-9-7-M-1-M) averaging 4.7 t ha-1, similar to that of CTl0069-27-3-1-4 and statistically different from the four others. These preliminary results indicate that CIAT upland breeding lines are an alternative for the coffee-growing region of Colombia. Further research will aim to identify the best line for release and suitable agronomic practices.

MTU9993, a promising rainfed upland rice for Andhra Pradesh, India

P. S. S. Murthy, S. S. R. Prasad, K. R. K. Murthy, and N. S. R. Reddi, Agricultural Research Station, Maruteru 534122, West Godavari District, Andhra Pradesh, India

Performance of MTU9993 under rainfed upland conditions at multiple locations during dry seasons 1988-92. Grain yield (t ha-1 ) 1988 1989 1990 1991 1992

Location and variety

Chinthapally MTU9993 MTU17 (check) % increase Lam, Guntur MTU9993 MS. Vari (check) % increase Ragolu MTU9993 MTU17 (check) % increase Jagtial MTU9993 Rasi (check) % increase V.R. Gudem MTU9993 MTU17 (check) % increase Maruteru MTU9993 MTU17 (check) % increase

3.2 2.9 10.3 1.8 1.6 12.5

3.5 2.5 40.0 3.9 3.1 25.8

2.5 1.8 38.8

2.3 1.5 53.5

1.7 1.2 41.7

MTU9993 is a high-yielding variety identified from the progeny of Rasi/Fine Gora (white) for the rainfed uplands of coastal Andhra Pradesh. It was released in I993 to replace tall traditional varieties that are low yielding and prone to lodging. The short-duration (105-110 d) variety has straw-colored glumes and white kernels. MTU9993 is superior to the existing rainfed upland rice cultures, such as MTU17 and Mettasannavari, in grain quality and yield, shortness, tolerance for lodging, strong seed dormancy, tolerance for iron deficiency, and resistance to leaf blast. The variety was tested from 1988 to 1992 and yielded 10-57% more than local checks (see table).

4.4 3.3 33.3

3.4 2.6 30.8

2.2 2.0 10.0 1.8 1.2 50.0

2.7 2.2 22.7

2.8 2.2 27.3

0.9 0.8 12.5

2.6 2.2 18.2

1.4 1.2 16.7

2.3 1.8 27.5

2.2 1.4 57.1

4.3 2.9 48.3

2.5 2.0 25.0

2.2 1.6 37.5

2.8 2.0 40.0


IRRN 21:2-3 (August-December 1996)

Crop and resource management

Physiology and plant nutrition
Analysis of growth duration and heat units of different rice genotypes
J. L. Chaudhary, A.S.R.A.S. Sastri, and R. K. Sahu, lndira Gandhi Agricultural University, Raipur, Madhya Pradesh 492012, India

In general, higher temperatures accelerate the rate of development at all phenological stages of a crop and thereby reduce the length of a given phenological stage (Oldemann et al 1986). We analyzed variation in duration of some prominent rice genotypes from sowing to 50% flowering and the corresponding variation in accumulated degree days (DD) (with base temperature of 10 C) in Raipur, India. Long-term phenological observations and meteorological data from a nearby agrometeorological observatory were used. Duration from sowing to 50% flowering of 12 selected genotypes varied from 66 d (Poorva) to 111 d (Surekha) [see table). The

DD requirement for the same period was lowest for Poorva (1127 DD) and highest for Surekha (1935 DD). The coefficient of variation for duration varied from 2.6%; for Samridhi to 6.8% for Cauvery. indicating least variability for Samridhi and greatest variability for Cauvery. Variability in DD requirement varied from 4.8% for Samridhi to 6.9% for Phalguna. For Samridhi, this indicates that variability in duration was lower than variation in DD requirement. This is contrary to the general belief that variability in duration of different phenological stages can well be explained by DD. This trend also suggests that variability in duration of different phenological stages is influenced more by thermal regime within the vicinity of the crop environment. The meteorological observatory was situated in a nearby field and the microclimate there is quite different from the one existing in the ricefields. Cited reference
Oldemann LR, Seshu DV, Cady FB. 1986. Response of rice to weather variables. Report on the IRRI/WMO Project. Manila (Philippines): International Rice Research Institute.

The processes of pollination and fertilization in rice

K. Hattori and T. Wada, Laboratory of Plant Genetics and Breeding, School of Agricultural Sciences, Nagoya University, Chikusa, Nagoya 46401, Japan

Analysis of variation in growth duration (from germination to 50% flowering) and degree days of some prominent rice genotypes. Duration (d from germination to 50% flowering) Mean 92 89 66 76 75 91 89 105 88 104 111 102 SD 4.95 2.33 4.35 4.6 5.1 4.6 4.9 5.3 4.7 5.9 5.6 5.7 Cv (%) 5.4 2.6 6.6 6.1 6.8 5.0 5.5 5.0 6.4 5.7 5.0 5.6 Mean 1584 1527 1127 1311 1295 1575 1539 1794 1523 1771 1935 1762 Accumulated degree days SD 100 73 65 81 81 89 86 115 94 123 113 91 CV (%) 6.3 4.8 5.8 6.1 6.2 5.6 5.6 6.3 6.1 6.9 5.8 5.1


Observations (no.) 31 6 12 16 25 13 41 10 38 24 9 8

IR36 Samridhi Poorva Tripti Cauvery Deepti Ratna Madhuri Anupama Phalguna Surekha Usha

The processes of pollination and fertilization in higher plants have, to date, been only well studied in Liliaceae species. In japonica rice variety Nipponbare, we observed conditions of pollen tube elongation on the stigmatic hair and the portion of the style through which the pollen tube passed. Pollen tube behavior in the ovary was also studied. At the flowering stage, every flower was sampled and fixed in formalin-acetic-alcohol after pollination. These samples were examined under both fluorescent and light microscopes after being sectioned using a resin-embedding method. Light microscopy of a cross-section taken from the top of the stigma to the bottom of the ovule showed that each stigmatic hair had a hole in the center (the intercellular space formed with stigmatic hair cells) (Fig. 1a). These holes are connected to an area (transmitting tissue) made up of cells that are shaped differently than other parenchymatous cells located inside the vascular bundle of the stigma (Fig. 1b). This area continued through the stylar tissue up to the ovary. In the ovary, a narrow space was observed between the outer wall of the ovule and the ovary wall, except at the dorsal portion of the locule (Fig. 1c). This space shows where the pollen tube passed through the micropyle. The longitudinal sections of the ovary reveal that the ovary wall and the outer wall of the ovule fused from the top to near the bottom of the ovule (Fig. 1d).

IRRN 21:2-3 (August-December 1996)


1. Observations of some parts of the rice pistil using a light microscope. a) cross-section of a stigmatic hair, b) cross-section of a stigma, c) cross-section toward the bottom of an ovary, and d) longitudinal section of the dorsi-ventral direction of an ovary. Bars show 100 m. 2. Observations of some parts of the rice pistil after pollination using an epifluorescence microscope. a) cross-section of stigmatic hair, b) longitudinal section of stigmatic hair, c) cross-section of the uppermost part of style, and d) cross-section of ovary. Bars show 100 m.

Fluorescent microscopy showed that pollen germination occurred 1 h after the pollen reached the stigmatic hairs. The pollen tube then elongated, passing through the center hole of the stigmatic hair (Fig. 2a, b). In the stigmatic and stylar tissues. the pollen tube passed through the transmitting tissue located in the space between the center area and the vascular bundle (Fig. 2c). (Among Liliaceae species, the pollen tube is believed to pass through the stylar cavity.) The cells comprising the transmitting tissue differed from other parenchymatous cells. They appear as a darkened area under the fluorescent microscope. implying that the cells were thin walls. The absence of a narrow space toward the dorsal side (downward) of the ovary explained the observance of pollen tube elongation at both sides of the locule but not in the dorsiventral direction of the ovary (Fig. 2d). The pollen tube reached the micropyle 3 h after pollination. However, time of fertilization could not be determined.

Studies on intercellular space of embryo and seedling tissues in rice plants

T. Wada, R. Goto, S-Y. Kang, School of Agricultural Sciences, Nagoya University, Chikusa, Nagoya, 464-01, Japan; and K. Hattori, Chungnam Rural Administration, Chungnam 320-860, Korea

The development of intercellular spaces into aerenchyma in rice plants is thought to be important in improving oxygen gas diffusion through the plant body. But details of the developmental process during seedling and early vegetative stages are unclear. To better understand the process, a morphological study of intercellular spaces was conducted using high-resolution microscopy as well as cryo-scanning electron microscopy. Intercellular spaces form in mature embryos of dry seeds (see figures a, b, c).

The spaces are located mainly in the parenchymatous cell regions including the scutellum, mesocotyl, and coleoptile and radicle, and are evident in the inner regions of these tissues. Intercellular spaces near the apical region of the radicle procortex were found in mature embryos, confirming previous findings (Armstrong 1979) (see figures f, g). However, all these spaces. which appear at the corner of neighboring cells, are very small in dry seeds. They increase in volume during imbibition and germination (see figures a-e). During the early seedling stages after germination, no obvious changes in intercellular space were observed, but a tubular structure along the longitudinal cell alignment was formed (this occurred after the elongation and development of cortical cells along the shoot-root axis) (see figures e, f, h). In the shoot region, air space was formed in chlorenchymatous cell regions in the coleoptile and young leaves, especially under nonsubmerged culture condition.

The intercellular space near the axis during the early seedling stage seemed to be mostly in the water-rich state as revealed by cryo-scanning electron microscopy. Similar results have been obtained in other plants (Canny and Huang 1993). Seven days after germination, root air space was formed in the basal elongated seminal root region of the nonsubmerged plants. In the submerged condition, inner air space was formed among the elongated coleoptile and young leaves, while seminal root growth was reduced and the development of the cortical air space was delayed. Cited references
Armstrong W. 1979. Aeration in higher plants. Adv. Bot. Res. 7:225-332. Canny MJ, Huang CX. 1993. What is in the intercellular spaces of roots? Evidence from the cryo-analytical-scanning electron microscope. Physiol. Plant. 87:561-568.


IRRN 21:2-3 (August-December 1996)

Fertilizer management
Effects of time of split application of K on soil N forms and lowland rice yield
D. Muthumanickam, M. Ravichandran, and M. V. Sriramachandrasekaran, Faculty of Agriculture, Annamalai University, 608002, India

Light micrographs of rice embryo tissues, showing intercellular spaces among the cells. A low magnification view of the radicle and mesocotyl regions of a mature embryo (a); higher profiles of the radicle procortex region (b); inner coleoptile cells (c) in dry seeds; inner scutellar parenchyma cells 48 h after imbibition (HAI) (d); meristematic cortical cell region of seminal root (e); and a lower view of the root (f), 72 HAI; apical region of the root tip, 96 HAI (g); and meristematic cells of the cortex, 120 HAI (h). d-h: nonsubmerged condition. Arrows indicate intercellular spaces. Bar = 50 m.

An experiment was conducted to study the effect of timing of split application of K on N forms in the soil and on rice yield. The experiment was carried out using clay soil (Typic Chromusters. pH 8.04, electrical conductivity 0.32 dS m -1 , organic carbon 0.41 %. and 294, 10, 290 kg available NPK ha -1 ). The crop received 100-22-41.5 kg NPK ha -1. Half of the N and K and the entire dose of P were applied basally. The remaining amount of N and K was equally split and topdressed as per treatment schedule (see table). The experiment was carried out in two different seasons (February to June and July to November) using IR20 and a factorial completely randomized block design. Treatments were replicated thrice. Twelve replications were used (three replications at each stage of soil analysis). Soil samples were taken at a depth of 15 cm during tillering, panicle initiation, flowering, and harvest stages. Wet soil samples were analyzed for NH 4-N, NO3 -N, available N, and fixed NH4 -N. Plants collected at harvest were threshed carefully to separate the grains and straw and weighed separately after drying at 60 C. In both seasons, grain and straw yields were highest when K was always applied 7 d earlier than N (see table). Yields were lowest in the control and when K was applied 7 d after N (see table). The higher grain and straw yield with early application of K was caused by the greater availability of NH 4 -N and NO 3 -N at tillering and panicle initiation (see figure),

IRRN 21:2-3 (August-December 1996)


Effect of time of split application of K on grain and straw yield (t ha -1) of rice. Treatmenta Grain yield Feb-Jun Control (T1) Entire basal K (T2) K 7d earlier than N at T and PI (T3) K + N at T and PI (T4) K 7d after N at T and PI (T5) K 7d earlier than N at T and K + N at PI (T6) K 7d earlier than N at T and K 7d after N at PI (T7) K + N at T and K 7d earlier than N at PI (T8) K + N at T and K 7d after N at PI (T9) K 7d after N at T and K 7d earlier than N at PI (T10) K 7d after N at T and K + N at PI (T11) CD (p = 0.05)

Straw yield JuI-NOV 3.3 4.4 6.4 4.9 3.6 5.5 4.9 5.0 4.1 4.0 4.0 0.405 Feb-Jun 4.8 6.6 8.2 7.0 5.3 7.2 7.6 7.3 6.6 6.5 6.7 0.834 Jul-Nov 4.4 6.4 8.1 6.7 5.0 7.0 7.4 7.1 6.3 6.1 6.4 0.625

3.4 4.4 6.5 5.1 3.8 5.7 5.1 5.2 4.3 4.2 4.2 0.544

T = tillering. PI = panicle initiation, K + N = joint application of K and N.

Grain yield was positively correlated with NH 4 -N at maximum tillering in the February-June (Y= 1.51 -1.85, r = 0.89**) and July-November seasons (Y = 1.49 -1.91, r = 0.91**). Application of K before N presumably suppressed fixation of NH4+ from applied fertilizer

because selective exchange sites in the 2:1 layer clay minerals were occupied by K+. On clay soils with high K+ and NH 4 + fixation potential, rice yields and N recovery efficiency can be increased by applying K fertilizer 1 wk before N topdressing.

Changes in ammoniacal (a), nitrate (b), and available nitrogen (c) due to time of split application of potassium.

Effect of Biozyme and NPK on rice yield

A. Sen and K. Santosh, Agronomy Department, Institute of Agricultural Sciences, Banaras Hindu University (BHU), Varanasi 221005, Uttar Pradesh, India

Biozyme is a commercial product produced from Ascophyllum nodosum, a seaweed alga known to be rich in cytokinin and auxin precursors, enzymes, and hydrolyzed protein. Farmers apply it in the field to enhance growth and yield of crop by effecting better utilization of nutrients. We conducted a field trial during the 1993-94 wet season to investigate the effect of Biozyme granules in combination with NPK on yield performance of rice at the BHU research farm. The trial consisted of nine treatments: control (i.e., no Biozyme or NPK) (T1); 15 kg Biozyme ha -1 + 12060-60 kg NPK ha-1 (100% fertilizer) (T2); 15 kg Biozyme ha-1 + 50% fertilizer (T3); 20 kg Biozyme ha-1 + 100% fertilizer (T4);

20 kg Biozyme ha-1 + 50% fertilizer (T5); 40 kg Biozyme ha-l + 100% fertilizer (T6); 40 kg Biozyme ha-1 + 50% fertilizer (T7); 50 kg Biozyme ha-1 + 100% fertilizer (T8); and 50 kg Biozyme ha-1 + 50% fertilizer (T9). The experiment was laid out in a randomized block design with three replications. N, P, and K were applied as urea, diammonium phosphate, and muriate of potash, respectively. All of the P, K, and Biozyme were applied basally, while N was applied in three splits (in 1:2:1 ratio) during transplanting, maximum tillering, and panicle initiation. In treatments 6, 7, 8, and 9, Biozyme was applied in two equal splits, at transplanting and 30 d after transplanting (DAT). The variety was Early Monsoori, a medium-duration cultivar. Soil was slightly alkaline (pH 7.4), has low available N (210 kg ha-1) and P (11 kg ha-1) but medium K (179 kg ha-1) content, and 0.32% organic C. The field was kept flooded throughout the growing period.

All treatments recorded significant increases in number of spikelets panicle -1. panicles m-2, and grain and straw yield over the control in both years. An increase in yield with an increasing dose of Biozyme + fertilizer was observed up to the 40 kg Biozyme + 100% fertilizer ha-1 treatment, after which a decreasing trend was noted. T6 produced a significantly superior yield compared with T2 and T3; the rest remained statistically on par with each other. It was further observed that the yield gap between 100 and 50% fertilizer doses along with different doses of Biozyme was considerably reduced. A similar trend was also found in all other yield-attributing characters. The stimulatory effect of Biozyme was most probably due to the presence of cytokinin and auxin precursors, which resulted in elevated cellular energy and more effective tillering. The addition of Biozyme at 20 kg ha-1 with 50% of the recommended NPK dose could save as much as Rs 690.35 ha-1 over 100% fertilization.


IRRN 21:2-3 (August-December 1996)

Yield-attributing characters of rice as affected by Biozyme and NPK. Uttar Pradesh, India, 1993-94. Panicle length (cm) Treatment Control (no Biozyme and no NPK) 15 kg Biozyme ha -1 + 100% fertilizer a 15 kg Biozyme ha -1 + 50% fertilizer 20 kg Biozyme ha -1 + 100% fertilizer 20 kg Biozyme ha -1 + 50% fertilizer 40 kg Biozyme ha -1 + 100% fertilizer 40 kg Biozyme ha -1 + 50% fertilizer 50 kg Biozyme ha -1 + 100% fertilizer 50 kg Biozyme ha -1 + 50% fertilizer S dLSD (0.05) CV (%) Treatment Control (no Biozyme and no NPK) 15 kg Biozyme ha -1 + 100% fertilizer a 15 kg Biozyme ha -1 + 50% fertilizer 20 kg Biozyme ha -1 + 100% fertilizer 20 kg Biozyme ha -1 + 50% fertilizer 40 kg Biozyme ha -1 + 100% fertilizer 40 kg Biozyme ha -1 + 50% fertilizer 50 kg Biozyme ha -1 50 kg Biozyme ha -1 + 50% fertilizer S dLSD (0.05) CV (%)
a 120-60-60 kg NPK ha -1.

Panicle weight (g) 1993 3.21 3.69 3.51 4.10 4.00 4.29 4.17 3.98 3.74 0.80 ns 11.97 Filled 1993 81.12 82.67 81.57 85.03 83.19 85.96 85.49 84.96 84.08 3.35 ns 4.87 1994 3.83 3.90 3.84 4.22 4.18 4.47 4.37 4.11 4.00 0.33 ns 10.29 spikelets (%)

Spikelets panicle-1 (no.) 1993 173.17 187.30 183.40 197.23 195.13 202.63 199.33 189.23 181.60 7.23 15.35 13.33 1994 190.10 210.32 199.62 220.30 212.53 225.87 215.40 207.05 203.21 8.17 17.32 13.46

Spikelets m-2 ('000) 1993 29.13 40.44 38.75 45.04 42.11 51.37 48.17 38.72 45.07 0.74 1.58 3.35 1994 36.00 50.63 47.34 55.41 51.25 65.79 60.83 48.08 46.81 0.83 1.75 3.71

1.000-grain weight (g) 1993 16.87 16.97 16.93 17.07 17.02 17.60 17.33 17.20 17.00 0.43 ns 3.09 1994 16.58 17.01 16.99 17.10 17.08 17.85 17.58 17.33 17.23 0.39 ns 4.58

1993 22.60 23.34 23.62 22.92 22.86 22.85 23.74 23.74 23.11 0.62 ns 3.27 Panicles 1993 175.31 227.33 218.44 243.62 231.23 265.19 258.92 218.65 264.91 18.85 39.96 12.13 m -2

1994 21.34 22.04 22.22 22.66 22.26 22.00 21.65 22.46 22.38 0.61 ns 3.38 (no.) 1994 200.11 255.73 249.42 261.15 255.80 309.43 297.37 247.13 245.56 21.22 44.99 13.34

Grain yield (t ha -1) 1993 2.1 2.9 2.8 3.3 3.2 3.5 3.4 3.0 2.9 0.3 0.6 11.2 1994 2.2 3.2 3.2 3.5 3.4 3.7 3.6 3.3 3.3 0.3 0.5 12.0

Straw yield (t ha -1 ) 1993 3.8 4.8 4.6 5.4 5.3 5.2 5.6 5.0 5.0 0.7 1.5 13.9 1994 4.2 6.0 5.8 6.4 6.2 7.0 6.6 5.1 5.9 0.5 1.2 11.7

1994 84.34 86.83 86.18 87.17 86.99 89.96 88.89 86.93 86.77 3.21 ns 5.07

Effects of fertilizer and green manure on survival and productivity of detached deepwater rice plants
M. M. Panda, Regional Research Station (RRS), Judia Farm, Keonjhar 758002, Orissa, India

of stem detachment and improve the survival and productivity of deepwater rice. Detached deepwater rice plants (128 d old) having only nodal roots were floated in concrete tanks containing 2400 liters of water. Treatments consisted of applying various fertilizers and burying of Sesbania aculeata plants in water (see concentration of 5 or 10 ppm N in water. Temperature and dissolved O2 content of the water were measured at 6:20 a.m. and at 4:00 p.m. Average temperature was 29 C in the morning and 31 C in the afternoon. Water pH was measured at 9:00 a.m. and at 3:00 p.m. It increased by 0.5-1.0 unit in the

table). The aim was to maintain a nutrient

Deepwater rice plants sometimes get detached from the soil during the vegetative stage due to damage by crabs and other organisms. They then float in water. The survival of such plants depends on nodal roots that draw nutrients from the floodwater. Nodal roots can appear within 24 h

afternoon. After application of NP/NPK fertilizers, pH increased to within the 9.39.7 range due to algal growth. Decomposition of green manure lowered water pH, maintaining it between 7.2 and 8.2. Dissolved O2 concentration was generally higher in the afternoon than in the morning. In the presence of decomposing green manure, dissolved O2 was 0.0-0.2 ppm in the morning and 1.0-6.6 ppm in the afternoon. At crop maturity, four samples were harvested from each tank. There was a positive response to applied nutrients (see table). Application of NP gave NPK gave the best yields and were superior to application

IRRN 21:2-3 (August-December 1996)


Effect of fertilizers and green manure on yield of detached deepwater rice plants grown in water. Treatment Yield (g m-2 tank area) Grain No fertilizer (control) 5 ppm N as urea 10 ppm N as urea 5 ppm N through fresh Sesbania aculeata plants (3 kg) 5 ppm N as urea and 5 ppm P2O5 as single superphosphate 5 ppm N as urea, 5 ppm P2O5 as single superphosphate, and 5 ppm K 2O as muriate of potash LSD (0.05) 14.4 25.1 27.4 22.6 37.5 38.7 Straw 261 321 329 326 394 378

Effect of green manure and fertilizer N on rice yield and comparative economic data. Maharashtra, India. Rice yield (t ha -1) 1989 100 kg N ha-1 as urea in 3 splits: 1/2 at planting + 1/4 at tillering + 1/4 at panicle initiation (recommended) 5 t G. maculata ha-1 (27.4 kg N ha -1 + 50 kg N ha -1 as urea at planting 7.5 t G. maculata ha -1 (41.1 kg N ha -1) + 25 kg N ha -1 as urea at planting 5 t S. aculeata ha -1 (29 kg N ha -1 + 50 kg N ha -1 as urea at planting 7.5 t S. aculeata ha-1 (43.5 kg N ha -1 ) + 25 kg N ha -1 as urea at planting Control (no N) SE LSD (0.05)


1990 2.6

Pooled mean 3.1

Cost of cultivation (US$ ha-1) 344.99

Gross returns (US$ ha-1) 404.71

Net B:C a returns (US$ ha -1) 59.72 1.17


3.1 3.4

2.2 2.5

2.6 2.9

330.10 334.80

342.95 385.00

12.85 50.20

1.04 1.15



3.2 3.4

2.3 2.1

2.8 2.7

335.70 333.35

363.98 357.41

28.28 24.06

1.08 1.07

of N alone, either through urea or green manure. The detached plants could complete their life cycle to grain filling and ripening without anchorage and nutrition from the soil. However, grain yield was low, with a maximum harvest index of 0.1.

2.4 0.2 0.6

1.5 0.2 0.5

1.9 0.1 0.2





Benefit-cost ratio.

Integrated N management with Gliricidia maculata and Sesbania aculeata for transplanted rice
L. S. Chavan, B. P. PatiI, and T. S. Ingole, Agricultural Research Station, Palghar 401404, Thane District, Maharashtra, India

To save on fertilizer costs, green manure (GM) is used in conjunction with fertilizer N. We evaluated the effect of integrating GM Gliricidia maculata and Sesbania aculeata with different levels of fertilizer N (urea) on grain yield of transplanted rice in the 1989-90 rainy season. Gliricidia seedlings were planted on ricefield bunds at a distance of 0.5-1 m. The plants produced sufficient green loppings after 3 yr with 2.74% N, 0.5% P, and 1.1 % K in the leaves (dry weight basis). Sesbania was grown in situ at a seeding rate of 50 kg ha -1 and incorporated in the soil after 45 d. Sesbania contained 2.9% N, 0.2% P, and 0.98% K. Soil was clay with pH 8.1, 0.6% organic C, and 200-42-346 kg available NPK ha -1 . The experiment was laid out in a randomized block design with four replications. GM crops were incorporated in the soil at transplanting. On the same day, 30-

to 35-d-old seedlings of 125-d-duration rice variety PLG1 were transplanted. P and K at 50 kg ha -1 each were basally applied. N was applied as per treatment schedule (see table). Fertilizer application at the recommended dose and integration of fertilizer N with GM proved superior to the control. Basal application of 7.5 t G. maculata with 25 kg urea N ha -1 was on par with 100 kg urea N ha-1 in terms of yield. Costs of cultivation, gross returns, net returns, and benefit-cost ratio of this integrated N management treatment were comparable with those of the recommended dose. With this treatment, a savings of 75 kg urea N ha -1 may be achieved.

Fertilizer management inorganic sources

Response of late-transplanted rice to NPK under irrigated conditions
B. Hasan, A. S. Bali, and K. N. Singh, Agronomy Division, S. K. University of Agricultural Sciences and Technology (SKUAST), P. 0. Box 706, G. P. 0. Srinagar, Kashmir 180001, India

Rice followed by rapeseed - mustard is the most feasible double cropping system in Kashmir Valley (1650 m asl). But aberrant weather conditions (low temperature and cloudiness) delay transplanting. The NPK requirement of a crop under such a situation may differ from that of a timely planted rice crop, which usually requires 80-45-20 kg NPK ha -1. We conducted a 2-yr field trial at SKUASTs Shalimar Campus, on silty clay loam soil (pH 6.8, low available N and P, medium available K) during the 1990-91 wet seasons. The 18 treatments, tested in a randomized block design experiment with three replications, comprised different combinations of NPK levels (see table). Half the N and a full P and K dose were applied basally while the remaining N was applied in two splits: at tillering (18 d after transplanting [DAT]) and at panicle initiation (37 DAT). Rice variety K39 was transplanted at 15 1.5-cm spacing at 3 seedlings per hill by the 1 st week of Jul. During the experiment. temperatures ranged from 23.0 to 31.1 C (maximum) and from 11.3 to 19.6 C (minimum). Relative humidity was 4576%. Total rainfall averaged 234 mm, spread over 35 d.


IRRN 21:2-3 (August-December 1996)

Effect of NPK levels on yield and yield-attributing characters of rice. Kashmir, India. 1990-91. Nutrient level (kg ha -1 ) N 40 80 120 LSD (P = 0.05) P 8.8 17.6 26.4 K 16.6 33.2 Yield (t ha -1) Panicles m -2 (no.) Grains panicle -1 (no.) 57.8 65.8 69.0 1.7 1,000-grain weight (g)

4.0 4.5 4.7 0.37

318 331 339 5

21.25 21.70 23.05 0.50

LSD (P = 0.05)

3.6 4.6 5.0 0.37

315 331 344 5

58.7 64.7 69.0 1.7

20.90 21.50 22.15 0.50

Results showed that up to 80 kg of N and up to 26.4 kg of P ha-1 significantly increased rice grain yield with concomitant improvement in all yield attributes (see table). The yield levels obtained were, however, lower than those of a timely sown crop (5.5 - 6.0 t ha-1). This implies that a combination of 80-26.4-33.2 kg NPK ha-1 will give satisfactory yields under late transplanting conditions in the valley.

LSD (P = 0.05)

4.9 5.5 0.30

322 335 4

63.0 65.4 1.3

21.40 21.65 0.40

Fertilizer managementorganic sources

Evaluating stem-nodulating green manure crops for lowland rice
D. N. Medhi, Agronomy Department, Assam Agricultural University (AAU), Jorhat 785012, Assam, India; and B. Deka Medhi, Agricultural Chemistry Research Unit, AAU

We studied the effect of stem-nodulating green manure (GM) crops Sesbania rostrata, S. speciosa, Aeschynomene afraspera, A. indica, A. schemperi, and A. nilotica on N2 fixation and biomass production in lowland rice. The experiment was conducted during Aug 1993 and Jul 1994 wet season on a sandy loam acidic soil (pH 5.1) with 205 kg available N ha-1. S. aculeata was the check.

Treatments were replicated four times in a randomized block design. The GM seeds were broadcast at 50 kg ha-1 in all the 3- 3m plots. The GM plants were allowed to grow for 45 d after emergence; they were then cut at ground level from an area of 0.32 m2 in each plot. N concentration of the whole plants was determined by microKjeldahl method. Significant differences in dry biomass production, N concentration. and N content of the GM crop were recorded (see table). S. rostrata and A. afraspera were found to be superior to all other species. indicating that these two are the most suitable GM crops for lowland rice soils.

Managing crop residue in rice

R. D. Misra, V. K. Gupta, and D. S. Pandey, Department of Agronomy, G. E. Pant University of Agriculture and Technology (GBPUAT), Pantnagar, Nainital 263145, India

Dry biomass production, N concentration, and N content of different green manure crops. Assam, India, 1993-94. Species Dry biomass (t ha-1 ) 1993 3.1 5.3 3.1 4.3 2.5 2.3 2.4 2.5 1994 3.2 6.1 3.3 5.4 3.1 2.2 3.5 2.3 N concentration (%) 1993 1.4 3.0 2.7 3.6 2.7 2.8 2.6 1.8 1994 1.5 3.2 2.9 3.8 2.9 2.8 2.7 1.9 N content (kg ha -1) 1993 41 142 94 139 65 61 61 30 1994 44 188 88 185 89 58 84 32

Sesbania aculeata Sesbania rostrata Sesbania speciosa Aeschynomene afraspera Aeschynomene indica Aeschynomene schemperi Aeschvnomene nilotica LSD (0.05)

Burning of crop leads to loss of organic matter and nutrients. Incorporating straw helps in nutrient recycling and maintains soil health. We carried out an experiment on crop residue management at GBPUAT, Pantnagar, during the 1993-94 kharif season. Soil was loam with pH 7.4,0.084% total N, 1.10% organic C, 17.0 kg available P ha-1, and 194.0 kg available K ha-1. The experiment, laid out in randomized block design, had 12 treatments with four replications, (see table). Green manure (GM) crop Sesbania aculeata was grown and incorporated in the Same plot before transplanting rice, This provided 60 kg excess N ha-1 in 1993 and 57 kg excess N ha-1 in 1994. Berseem (Trifolium alexandruim L.) was sown in a standing crop of rice at the milk stage and was incorporated in the field 1 mo before transplanting, giving an additional 80 and 70 kg N ha-1 in the respective years. Incorporation of wheat straw. done without a starter dose, provided an additional 74 and 65 kg N ha-1. With 20 kg of starter dose of N (as urea), straw incorporation provided

IRRN 21:2-3 (August-December 1996)


Effect of different crop residue management practices on rice yield. Treatment Grain yield (t ha-1) 1993 Rice and what straw removed (control) Rice straw burned and wheat straw removed Only rice straw incorporated Only rice straw incorporated with 20 kg N ha -1 Only rice straw incorporated and excess 20 kg N ha-1 to wheat Rice and wheat straw incorporated Rice and wheat straw incorporated with 20 kg N ha -1 Only rice straw incorporated with berseem 20 t FYM a before rice 20 t FYM a before wheat Sesbania aculeata before rice Berseem before wheat LSD (P = 0.05)
a FYM = farmyard manure.

Evaluation of green manures and grain legumes

S. Porpavai, S. P. Palaniappan, and S. Purushothaman, Centre for Soil and Crop Management Studies, Tamil Nadu Agricultural University (TNAU), Coimbatore 641003, India

1994 4.6 4.7 4.7 4.9 4.8 5.0 5.1 5.0 5.4 5.0 5.7 4.8 .23

5.6 5.6 5.7 5.7 5.7 5.9 6.1 5.8 6.7 5.9 6.9 5.8 .15

We compared several green manures and grain legumes in terms of dry weight, N percentage, and N accumulation at 60 d after sowing in two seasons, Jul-Sep 1991 and Feb-May 1992. The experiment was laid out in a randomized block design with, three replications.

All of the green manures and grain legumes recorded higher dry matter production in the February-May season. Sesbania rostrata accumulated more N (165 kg ha-1 ) than did the other green manures. Cowpea (Vigna unguiculata ) outperformed the other grain legumes, producing 10 t green manure on dry-weight basis. After the rice harvest in January, farmers may grow S. aculeata or S. rostrata in situ and plow it down during transplanting of the next rice crop that is grown during kuruvai season (June-September).

Dry matter production and N accumulation in green manures and grain legumes at 60 DAS. a Coimbatore, India. Jul-Sep Dry weight (t ha -1) S. aculeata S. rostrata S. speciosa S. grandiflora Tephrosia purpurea Glycine max Crotalaria juncea Phaseolus trilobus Vigna aureus Vigna unguiculata CD

Green manure

1991 N accumulation (kg ha -1) 91.8 110.4 27.2 28.8 16.2 22.0 75.4 55.1 31.1 77.6 3.8 Dry weight (t ha -1) 8.6 7.5 6.5 7.2 5.0 8.1 9.3 9.5 8.4 10.0 3.5

Feb-May 1992 N (%) N accumulation (kg ha-1) 137.6 165.0 136.5 100.8 45.0 81.0 107.0 114.0 84.0 123.0 5.2

N (%)

90 and 83 kg N ha -1 in 1993 and 1994, respectively. Rice straw was incorporated or burned before the wheat crop. Twenty-two-day-old seedlings of Pant Dhan 10 was transplanted on 24 Jul 1993 and 7 Jul 1994. The recommended fertilizer dose of 120 kg N ha -1 was applied in all plots and the usual cultural operations were done. Burning of rice straw was at par with the treatment where straw was removed. Incorporating both rice and wheat straw gave significantly higher yield over rice alone because wheat straw supplied some extra nutrient to the rice crop. Incorporating straw with a starter dose of 20 kg N was slightly better than the treatment without a starter dose because the initial dose helped straw to decompose faster, thereby providing more nutrients to rice. Green manuring with S. aculeata gave the highest production and was at par with 20 t farmyard manure ha-1 before rice. Use of GM berseem gave a slightly better yield than the control. The higher grain yield obtained from incorporating straw with a starter dose of N, FYM, and GM with S. aculeata was attributed to extra N (more than 120 kg ha-1) applied to these treatments. Besides the direct contribution of N, the better soil physical condition must have positively contributed to yield.

5.4 4.6 1.6 2.4 1.4 2.0 5.8 3.8 2.1 5.8 0.61

1.7 2.4 1.7 1.2 1.2 1.1 1.3 1.5 1.5 1.4 0.62

1.6 2.2 2.1 1.4 0.9 1.0 1.2 1.2 1.0 1.2 1.0

DAS = days after sowing.

Effect of soil amendments on rice yield

R. D. Sharma and S. Ali, G. B. Pant University of Agriculture and Technology (GBPUAT), Western Campus, Modipuram Meerut, Uttar Pradesh 250110, India

We conducted a field experiment on a Typic Natrustalf at GBPUAT in the 1987-92 wet seasons (WS) to evaluate the performance of gypsum, pyrite, mud press, and farmyard manure (FYM) as soil amendments. Surface soil (0-15 cm) is silty clay loam (53.5% sand, 25.2% silt, and 21.3% clay), with pH 9.4-9.8, EC 6.9 dS m -1 , 22.6% exchangeable Na, 0.12% organic C, 8.2 kg Olsen P ha -1 , 0.35 c mol exchangeable K kg -1 , and CEC 12.4 c mol kg -1 .

Amendments were applied in Apr 1987 and 35-d-old seedlings (3 hill -1) of rice cultivar Saket 4 were transplanted at 15- 15-cm spacing. A uniform dose of 150 kg N had in three splits, 21.8 kg P ha -1 , 4 1.5 kg K ha -1 , and 25 kg ZnSO 4 ha-1 (basal), was applied. A wheat crop was raised in the rabi (winter) season, with 150 kg N applied in 3 splits, 26.2 kg P, and 41.5 kg K (basal) ha -1. The same rice - wheat cropping sequence and NPK fertilization were followed during subsequent years on the same layout without additional soil amendments. The control treatment had an average rice yield of 2.7 t ha -1 (Table 1). The higher yields recorded were with gypsum application at 12 t ha -1 followed by pyrite at 4 t ha-1. The lowest yield increase over the control was observed with pyrite at 2 t ha -1 , mud press at 6 t ha-1 , and FYM at 20 t ha -1.


IRRN 21:2-3 (August-December 1996)

Table 1. Effects of soil amendments on rice yield. Uttar Pradesh, India, 1987-92. Grain yield (t ha -1) Rice Soil amendment 1987 1988 1989 1990 1991 1992 Yield (t ha -1 ) 2.7 3.4 3.8 4.2 3.0 3.1 3.9 3.0 3.4 3.0 3.6 3.7 Increase over control (%) 25.9 40.7 55.6 11.1 14.8 44.4 11.1 25.9 11.1 33.3 37.0 Yield (t ha -1) Six-year average Wheat Increase over control (%) 66.7 100.0 141.7 41.7 58.3 108.3 41.7 66.7 25.0 66.7 83.3

Control Gypsum (6 t ha -1) Gypsum (9 t ha -1) Gypsum (12 t ha -1) Pyrite (2 t ha -1) Pyrite (3 t ha -1) Pyrite (4 t ha -1) Mud press (6 t ha -1) Mud press (12 t ha -1 ) FYMa (20 t ha -1) FYM (10 t) + pyrite (2 t ha -1) FYM (10 t )+ mud press (6 t ha -1 ) SEM () LSD (5%)
a FYM = farmyard manure.

0.9 1.8 2.2 2.6 1.3 1.6 2.5 1.5 1.7 1.4 2.2 2.4 0.12 0.4

2.2 3.1 3.4 3.8 2.5 2.9 3.5 2.7 3.1 2.7 3.2 3.3 0.05 0.2

2.8 3.5 3.8 4.2 3.0 3.1 3.9 3.1 3.4 2.9 3.6 3.7 0.07 0.2

3.1 3.8 4.1 4.6 3.3 3.4 4.3 3.4 3.8 3.3 3.9 4.1 0.06 0.2

3.5 4.1 4.5 4.9 3.7 3.8 4.5 3.7 4.1 3.6 4.2 4.3 0.09 0.3

3.8 4.3 4.8 5.1 3.9 4.0 4.6 3.8 4.3 3.9 4.4 4.5 0.016 0.5

1.2 2.0 2.4 2.9 1.7 1.9 2.5 1.7 2.0 1.5 2.0 2.2

Table 2. Physicochemical properties of surface soil (0-15 cm) a after 6 yr of experimentation. Uttar Pradesh, India, 1987-92. pH (1:2 soil-water ratio) 8.2 8.1 7.8 7.3 8.2 8.1 7.8 8.1 7.8 7.7 8.0 7.7 EC (dSm-1) (1:2 soil-water ratio) 0.40 0.36 0.33 0.33 0.36 0.34 0.35 0.35 0.29 0.30 0.33 0.32 0.02 CEC (c mol (p+ ) kg-1) 12.0 12.1 12.0 12.0 12.1 12.1 12.2 12.1 11.8 12.3 12.0 12.1 Exchangeable cations (c mol (p + ) kg -1) Na + 0.58 0.42 0.41 0.40 0.49 0.43 0.40 0.40 0.39 0.42 0.39 0.38 0.03 0.09 K 0.19 0.12 0.11 0.11 0.16 0.14 0.12 0.13 0.12 0.19 0.19 0.16 0.03 ns Ca ++ 5.27 7.00 7.13 7.47 6.80 7.00 7.10 6.73 7.20 6.13 6.67 7.47 0.29 0.84 Mg ++ 1.27 1.20 1.00 1.13 1.27 1.10 1.00 1.00 1.27 0.93 1.00 0.86 0.13 ns Exchangeable sodium percentage 4.83 3.50 3.45 3.13 3.95 3.58 3.28 3.28 3.31 3.44 3.22 3.22 -


Control Gypsum (6 t ha -1) Gypsum (9 t ha -1) Gypsum (12 t ha -1) Pyrite (2 t ha -1) Pyrite (3 t ha -1) Pyrite (4 t ha -1) Mud press (6 t ha -1) Mud press (12 t ha-1) FYM (20 t ha -1 ) FYM (10 t) + pyrite (2 t ha -1) FYM (10 t) + mud press (6 t ha -1) SEM () LSD (5%)


a The Initial values were pH 9.4-9.8, EC 6.9 dS m -1, CEC 12.4 cmol(p+) kg -1, and ESP 22.6. b ns = not significant.

Applying FYM at 10 t ha-1 in combination with pyrite at 2 t ha-1 or gypsum at 6 t ha-1 resulted in comparable yields. The amendments were effective in this order: gypsum > pyrite > mud press > FYM. Wheat yield averaged 1.2 t ha-1 in the control treatment (Table 1). Maximum yield increase was observed with gypsum application at 12 t ha-1 and minimum with FYM at 20 t ha-1. The pattern of effectiveness of soil amendments in the wheat crop was similar to that in the rice crop.

After 6 yr, the pH and exchangeable sodium percentage in the control and treated plots were appreciably reduced, and the level of amendment applied was reflected in this reduction (Table 2). Besides the amendments, the rice - wheat cropping sequence has a profound influence on soil improvement, as reflected in the yield levels from year 1 to year 6. Gypsum alone or gypsum with FYM outperformed pyrite.

Crop management
Use of deepwater varietal mixtures to minimize hydrological risk in basin lands of Bihar, India
M. M. Sinha, A. Kumar, N. Chaudhary, Rajendra Agricultural University, Bihar Agricultural Research Institute, Mithapur, Patna 800001, India

Chaurs (basin lands) are low-lying fields of variable shape and size having poor or no drainage. Crop production depends mainly on the unpredictable monsoon rainfall. Drought or flooding of varying durations

IRRN 21:2-3 (August-December 1996)


and intensities are the main hydrological constraints to rice productivity. Cultivating mixtures of deepwater rice (DWR) is a prevalent practice in chaurs. The varieties used have almost identical morphological characters and are highly adapted to sowing and transplanting conditions in the area. We studied the submergence tolerance of DWR varieties Darmi and Jagar under different hydrological conditions. Five samples of 100 panicles each were collected randomly from deepwater ricefields (200 cm maximum water depth) cropped with mixtures of Darmi and Jagar during 1990 kharif in chaur. The area was stratified into 50-100, 100-150, and 150200 cm water depths. Based on kernel color and panicle characteristics, samples from each water regime were classified into threeDarmi, Jagar, and others. Pregerminated seeds of Darmi and Jagar were sown 5 cm apart in five trays filled with soil on 12 Jun 1991. The trays with 14-d-old seedlings were completely submerged in a pond for 7 d with 30 cm of water above the tip of the seedling. On the 10th day of the recovery period, survival count and phenotypic scoring for submergence tolerance were done. The IRRI standard scoring system was used; a low score indicates good tolerance.

Table 1. Number of panicles of Darmi, Jagar, and others per sample of 100 panicles, collected from three water depths. Vaishali, Bihar, India. Water regimes Sample Jagar 1 2 3 4 5 Total Av 52 51 40 49 56 248 49.6 50-100 cm Darmi 45 36 50 35 40 206 41.2 Others 3 13 10 16 4 46 9.2 Jagar 34 41 28 40 30 173 34.6 100-150 Darmi 56 51 63 54 59 283 56.6 crn Others 10 8 9 6 11 44 8.8 Jagar 21 33 48 26 43 181 36.2 150-200 Darmi 77 61 50 66 54 307 61.4 crn Others 2 6 2 8 3 21 4.2

Table 2. Submergence tolerance scores of Darmi and Jagar. Bihar, India. Variety Darmi Jagar Tray 1 7 8 Tray 2 6 9 Tray 3 8 9 Tray 4 6 9 Tray 5 7 8 Av 6.8 8.6

Among samples collected from the 50-100 cm water regime, Jagar had more panicles (49.6%) than Darmi (41.2% ), suggesting that Jagar was the better adapted cultivar with a greater number of effective tillers. Among samples obtained from the lower niches of the chaur. Darmi had higher panicle number (56.6 and 61.4%) in the 100-1.50 and 150-200 cm water regimes compared with Jagar, which had 34.5 and 36.2%, respectively (Table 1). This indi-

cates that Darmi was better adapted to deep water than was Jagar. Darmi had better submergence tolerance (av 6.8) than Jagar (av 8.6) (Table 2). This suggests that Darmi possesses genes that make it suitable for deepwater areas. Developing DWR varietal mixtures with different levels of submergence tolerance but similar agronomic characters could be one way of minimizing hydrological constraints to growing rice in rainfed low-lying areas.

Effect of transplanting times on hybrid rice in Haryana, India

H. Om, Rice Research Station (RRS), Kaul 132021; S. K. Katyal, CCS Haryana Agricultural University, Hisar 125004; S. D. Dhiman and A. Singh, RRS, Kaul, India

subplots. Plant spacing was 20 15 cm. The crop was fertilized with 150 kg N.26.4 kg P, and 5.75 kg Zn ha-1, with N applied in three equal splits: at transplanting, active tillering, and panicle initiation. Thirty-day-

old seedlings were transplanted in all dates except in the 25 Jul 1993 planting, where 40-d-old seedlings were used. Highest grain yields were obtained when the genotypes were transplanted on 25 Jun

Effect of transplanting time and genotype on grain yield and harvest index. Kaul, India. 1993-94. Treatment 1993 Time of transplanting 15 Jun 25 Jun 5 Jul 25 Jul LSD (0.05) Genotypes ORI 161 PMS2 A/IR31802 PMS10 A/PR106 HKR106 LSD (0.05) Grain yield (t ha -1 ) 1994 Mean Harvest index 1993 1994 Productivity (kg d-1) 1993 1994 Spikelet sterility (%) 1993 1994

We evaluated the performance of newly developed hybrids in India with respect to staggered planting. The field study involved four transplanting dates (15 Jun, 25 Jun, 5 Jul, and 25 Jul) and four genotypes (hybrids ORI 161, PMS2A/IR31802, and PMS01A/PR106 and check variety HKR126) and was conducted during the 1993-94 wet seasons at RRS, Kaul. Soil was clay loam, with pH 8.2, low available N. but high available P and K. The experiment was laid out in a splitplot design with four replications; planting dates were main plots and varieties were

7.2 7.9 7.6 5.5 0.4 7.8 7.3 6.3 6.9 0.2

6.8 7.5 7.2 4.2 0.3 7.4 6.6 5.8 6.1 0.2

7.0 7.7 7.4 4.9 -

0.51 0.54 0.53 0.46 0.01 0.54 0.52 0.48 0.50 0.01

0.50 0.53 0.52 0.41 0.02

46.5 51.2 50.9 41.9 -

51.0 57.1 54.6 32.1 -

26.2 20.4 22.1 28.8 2.1

27.6 23.4 25.4 37.9 1.8 24.4 29.7 33.9 27.6 2.0

7.6 7.0 6.1 6.5 -

0.52 0.49 0.45 0.49 0.01

52.9 49.1 42.4 47.3 -

56.5 49.8 43.7 46.4 -

19.2 25.3 30.1 23.5 2.0

74 IRRN 21:2-3 (August-December 1996)

(7.7 t ha-1) and on 5 Jul (7.4 t ha -1) in both years (see table). A drastic yield reduction was observed in the 25 Jul planting, but the rate of decrease was slower in 1993 than in 1994. This may be due to the use of aged (40 d) seedlings in 1993. There was 10.1, 4.1, and 57.1% increase in grain yield in the 25 Jun planting over that of 15 Jun, 5 Jul, and 25 July, respectively. Harvest index and productivity (kg d-1) were highest in the 25 Jun planting, followed by the 5 Jul and 15 Jun plantings.

Lowest values of harvest index and productivity were registered with the 15 Jun planting. Spikelet sterility was lower in 1993. It was similarly lower for 25 Jun and 5 Jul than for 15 Jun and 25 Jul. Percentage sterility was relatively higher for the late planting dates in 1994. Among genotypes, ORI 161 (PHB71) produced the highest grain yield (7.6 t ha -1). Hybrids ORI 161 and PMS2 A/IR31802 were found significantly superior to

HKR126 (6.5 t ha-1). Productivity and harvest index were highest in ORI 161, followed by PMS2 A/IR3 1802 and HKR126. Spikelet sterility was minimum in ORI 161, followed by HKR126. Planting hybrid OR1 161 resulted in a 17% increase in grain yield over the best available pureline genotypes. In northwestern India, the best time to transplant hybrid rice is between 25 Jun and 5 Jul.

Effect of solar radiation and temperature on rice yields in different planting dates
F. J. Osuna-Canizales, Campo Experimental Zacatepec, Apartado Postal 12, Zacatepec, Morelos 62780, Mexico

Solar radiation and temperature are important climatic factors affecting rice yields under irrigated conditions. In an experiment to define optimum planting dates for the newly released high-yielding variety Morelos A92 and breeding line CAEZ401, the effect of these climatic factors on grain yield was assessed through regression techniques. Soil is a Vertisol with pH 7.6,3% organic matter, 0.31% total N, and 22 ppm Olsen P. Five planting dates with

3-wk intervals were evaluated, the first in 29 Apr and the last in 22 Jul 1993. The experiment was laid out in a splitplot design at the experimental station of Zacatepec, Morelos with four replications, with planting dates in the main plot and genotypes in the subplots. Forty-five-dayold seedlings were planted at 20- 20-cm spacing in 4- 4-cm plots. Water management consisted of periodical irrigation. Basal fertilization with 90 kg N ha-1 40 kg P ha -1 and 40 kg K ha-1 was , , applied 25 d after transplanting (DAT); additionally, 45 kg N ha-1 each were applied at panicle initiation and at booting stage. Grain yield in a 9-m2 area was measured and adjusted to 14% moisture on a wet basis. Regression analysis with grain yield as the dependent variable and global radiation

(GR, estimated after sunshine hours) and minimum mean temperature (MMT) as independent variables were performed in three periods: from 20 d before flowering up to flowering (f-20). from flowering to 20 d after flowering (f+20), and from 20 d before to 20 d after flowering (f-20 to f+20). Growth duration was shortened in the last planting dates because of photoperiod sensitivity of the two genotypes (see table). In the first planting date, Morelos A92 and CAEZ401 were harvested 137 and 125 DAT, respectively, whereas in the last planting date. Morelos A92 and CAEZ401 were harvested 120 DAT. Grain yield declined in Morelos A92 as planting was delayed (see figure), but in CAEZ401, decline was significant only for the late planting date.

Regression equations and parameters to assess the fitness of different models in the three periods. Zacatepec, Morelos, Mexico, 1993. Model Equation f-20 c 1 2 3 4 Yield d Yield Yield Yield = = = = 10328 - 4.6 (GR)e 437393 - 49234 (MMT) e + 1410 (MMT2) 10599 - 93 (GR/MMT) 916660 + 980 (GR) - 78858 (MMT) + 1496 (MMT2 ) - 16751 (GR/MMT) f + 20 1 2 3 4 Yield Yield Yield Yield = = = = 20 (GR) - 2411 144952 - 17855 (MMT) + 571 (MMT 2) 12114 - 137 (GR/MMT) 163592 + 37 (GR) - 18847 (MMT) + 568 (MMT 2 ) - 652 (GR/MMT) f - 20 to + 20 1 2 3 4 Yield Yield Yield Yield = = = = 28 (GR) - 6705 47782 (MMT) - 1277 (MMT 2 ) - 437848 305 (GR/MMT) - 796 454149 + 367 (GR) - 44589 (MMT) - 1051 (MMT 2) - 6035 (GR/MMT) 0.25 0.34 0.09 0.49 14.4 23.2 40.2 40.9 2595018 2594505 3138594 2799678 0.20 0.81 0.02 0.81 19.9 0.3 67.4 4.9 2768937 758453 3365802 1060182 0.16 0.07 0.19 0.85 25.9 78.1 20.2 2.6 2909314 3670335 2776786 813472 R2 (%) a MSEb

aObserved level of significance. b Mean square error. c f - 20 = from 20 d before flowering up to flowering, f + 20 = from flowering to 20 d after flowering. f - 20 to + 20 = from 20 d before to 20 d after flowering. d kg ha-1. e GR = global radiation. MMT = mean minimum temperature.

IRRN 21:2-3 (August-December 1996)


Filled grains panicle -1 and grain weight were the most affected by late transplanting. For the two genotypes, filled grains panicle-1 went down from 63 in the first date to 47 in the last date while 1,000 grain weight was reduced from 43.6 to 40.7 g. The regression equations obtained and the parameters used to evaluate fitness of the models in each period are shown in the table. Within each period, the best fit was obtained when GR and MMT were
Effect of planting date on grain yield of cultivar Morelos A92 and breeding line CAEZ401. Zacatepec, Morelos, Mexico. 1993.

evaluated together (model 4), except in the period f+20, when MMT and MMT2 (model 2) were comparable. The fitness of model 4 was higher for f-20 and f+20 than their combination. These results show that variation in incident global radiation and mean minimum temperature during the period 20 d before or 20 d after flowering may explain much of the variation in grain yield in the different planting dates for irrigated rice.

Integrated pest managementdiseases

Effects of neem derivatives on sheath rot in scented rice
K. Sivaprakasam and R. Jagannathan, Plant Pathology Department, Agricultural College and Research Institute, Madurai, Tamil Nadu, India
Comparison of biocides to control sheath rot incidence and grain yield of scented rice JJ92. Treatment Neem oil (3%) Neem seed kernel extract (5%) Monocrotophos (1250 ml ha -1) Carbendazim (500 g ha -1) Mancozeb (1 kg ha-1) Monocrotophos (1250 ml) + carbendazim (500 g ha -1) Monocrotophos (1250 ml) + mancozeb (1 kg ha -1) Nochi (10%) (Vitex negundo ) Untreated check LSD (P=0.05) Percent ShR incidencea 30.22 26.30 25.86 24.27 30.47 27.57 22.87 26.80 37.52 1.10 (33.25) (29.70) (30.48) (29.80) (33.38) (31.61) (28.49) (31.10) (37.67) Graln yield (t ha-1) 5.2 5.0 4.4 4.1 3.4 3.8 4.5 4.2 3.2 .5

We assessed the efficacy of botanical biocides and pesticides (fungicides carbendazim and mancozeb) for controlling sheath rot (ShR) of scented rice. Leaf extract and fungicides were applied twice as foliar sprays at a 2-wk interval beginning at booting. The field trial was laid out in a randomized block design with three replications using scented hybrid rice JJ92. Plot size was 5 2 m. ShR incidence was measured as the percentage of infected tillers in 25

a Means of three replications. Data in parentheses are arcsine-transformed values.

randomly selected hills/plot 20 d after the last spray. Higher yields were recorded with neem oil (3%), neem seed kernel extract (5%), and monocrotophos (1,250 ml ha-1) + mancozeb (1 kg ha -1). Standard fungicides carbendazim (500 g ha -1) and monocrotophos (1,250 ml ha -1), and

mancozeb (1 kg ha-1) effectively reduced ShR incidence (see table). Using a spray of neem seed kernel extract generated higher grain yield and effectively reduced disease incidence, although to a lesser degree, than using regular fungicides.

Seed borne nature and seed transmission of rice sheath blight

S. Acharya and P. K. S. Gupta, Plant Pathology Department, Bidhan Chandra Krishi Viswavidyalaya (BCKV), Kalyani 741235, Nadia, West Bengal, India

Effect of R. solani infection on rice germination and seed health. West Bengala , India. Treatment Seeds yielding R. solani (%) 37.83 56.35 12.17 0.682 1.520 50.92 19.98 0.545 1.214 0.9315 2.07 Germination (%) Seedling infection (%) 14.88 23.58 0 0.728 1.624 8.70 6.93 0.594 1.326 1.03 2.29 Seedling collapse (%) 4.35 9.21 0 0.127 0.284 9.09 0 0.018 0.024 0.032 0.072 Shoot fresh weight (mg) 279.46 361.20 467.26 0.57 1.286 334.66 403.94 0.47 1.050 0.81 1.819 Shoot dry weight (mg) 46.83 66.05 94.3 0.376 0.838 60.52 77.60 0.307 0.684 0.532 1.85 Root fresh weight (mg) 485.7 673.75 922.4 0.575 1.281 608.66 779.23 0.469 1.04 0.813 1.81 Root dry weight (mg) 64.1 120.13 174.3 0.324 0.721 109 130.02 0.265 0.590 0.459 1.022

Naturally Infested seeds Artificially inoculated seeds Apparently healthy seeds SE LSD (0.05) Unsterilized seeds Surface-sterilized seeds SE LSD (0.05) Interaction SE LSD (0.05)

59.09 70.66 73.19 0.655 1.459 65.63 69.67 0.535 1.19 0.9266 2.06

Although rice sheath blight (ShB) caused by Rhizoctonia solani is documented as a seedborne disease, evidence of this nature is lacking. We investigated spotted and apparently healthy seeds of rice cultivar Swarna Mashuri to find out the mode of transmission involved. The seeds were collected from a heavily ShB-infected rice crop from the BCKV Regional Research Farm.

a All percentage data are transformed angular values.


IRRN 21:2-3 (August-December 1996)

Healthy seeds (200 g) were placed in a sterilized erlenmeyer flask and 10 cm of airdried inoculum of R. solani (grown in sandmaize-meal medium for 15 d at 20 C) was added to it. The flask was shaken for 5 min. R. solani was isolated from spotted, artificially inoculated, and apparently healthy seeds. The seeds were placed in 2% sterilized water agar in petri dishes and incubated at 28 C for 72 h. One lot was seeded without surface sterilization; the other lot was seeded after surface sterilization with NAOCl solution for 1 min.

Among naturally infected seeds (spotted), 40% of unsterilized seeds and 36% of surface-sterilized seeds yielded R. solani (see table). This indicates that in naturally infested seeds, the fungus was both externally and internally seedborne. The data for artificially inoculated and apparently healthy seeds were 100 and 15, and 17 and 0, respectively. Isolation of the fungus from some unsterilized apparently healthy seeds is probably due to surface contamination. None of the apparently healthy seeds showed seedling infection. Seedling infection was higher in artificially

inoculated unsterilized seeds. Surface sterilization resulted in reduced seedling infection, indicating that the fungus in artificially inoculated seeds was externally borne. Infection of seedlings grown from R. solani -infected seeds had considerable effect on seedling health. This was indicated by the much higher fresh and dry weight of the seedlings grown from healthy seeds as compared with those from naturally infected seeds (both unsterilized and surface-sterilized) and artificially inoculated unsterilized ones.

Serological evidence for inducing resistance to rice tungro viruses using antiviral principles
P. Muthulakshmi, and P. Narayanasamy, Plant Pathology Department, Centre for Plant Protection Studies, Tamil Nadu Agricultural University, Coimbatore, 641003, India; and H. Koganezawa, IRRI

Detection by indirect ELISA of rice tungro-associated viruses in rice plants following AVP application and RTBV and RTSV inoculation. Coimbatore, India. a Plants infected (no.)/plants inoculated (no.) 24/25 4/25 5/25 Absorbance at 405 nm RTSV Infected 1.111-1.525 0.860-0.939 0.550-0.925 Uninfected 0.060 0.069-0.125 0.070-0.090 0.070-0.152 Infected 1.200-1.600 0.831-0.945 0.745-0.940 RTBV Uninfected 0.025 0.068-0.120 0.070-0.085 0.075-0.150


Use of antiviral principles (AVPs) from plant sources is found to reduce rice tungro virus (RTV) infection. We conducted a study to obtain serological evidence for the effectiveness of AVP application in managing rice tungro disease. AVPs from seed sprouts of pigeonpea (PP-AVP) and mungbean (MB-AVP) were sprayed at 10% concentration on rice variety Co 43. The following day, each plant was inoculated with two viruliferous Nephotettix virescens (Distant). Twentyfive plants were inoculated in each treatment; inoculated plants without AVP application served as the control. Leaf samples were collected 25 d after inoculation and tested using enzyme-linked immunosorbent assay (ELISA) to determine the presence of rice tungro bacilliform (RTBV) and rice tungro spherical virus (RTSV). The RTV antigen was optimally diluted in carbonate-bicarbonate buffer (pH 9.6), coated at 100 1 well-1 of ELISA plate and incubated overnight at 4 C. The plate was washed three times with PBS Tween 20 (PBS-T). Bovine serum albumin (2%) was added to block nonspecific reaction and the

RTV-inoculated (control) AVP-treated and RTV-inoculated 10% Pigeonpea AVP 10% Mungbean AVP Healthy (ELISA check)

a RTV = rice tungro virus, AVP = antiviral principles, ELISA = enzyme-linked immunosorbent assay, RTSV = rice tungro spherical virus, RTBV = rice tungro bacilliform virus.

plate was washed with PBS-T. The rabbit Ig G antiserum specific to RTBV and RTSV was diluted 1:100 with PBS-T buffer and incubated for 90 min at 37 C, then washed three times with PBS-T. The alkaline phosphatase conjugated goat antirabbit was diluted 1:7,000, placed in the well, and incubated for 1 h at 37 C. The plate was then washed five times with PBS-T. Freshly prepared substrate solution containing P-nitrophenyl phosphate (PNP) (1 mg ml -1 in 10% diethanolamine buffer) was added and incubated at room temperature for 1 h. The reaction was stopped with 3 M NaOH. Absorbance at 405 nm was recorded using the ELISA reader. Absorbance values greater than 0.152 (maximum value for healthy plants) indicate a positive reaction. The ELISA test was repeated. Application of PP-AVP and MB-AVP reduced virus infection to 16 and 20%,

respectively (untreated control has 96% infection). RTBV and RTSV could not be detected in plants that did not show any visible symptoms, indicating their absence in AVP-treated plants. But the presence of RTSV and RTBV was detected by ELISA in all plants showing infection symptoms. Applying PP-AVP protected the plants from both RTSV and RTBV infection, shown by the reduced percentage of infection in AVP-treated plants. The concentration of both viruses was reduced in AVP-treated plants, even though they are infected by the viruses, indicating that AVPs may interfere with the replication of rice tungro-associated viruses (see table). The AVPs do not appear to have any effect on leafhopper feeding. Studies on the characterization of AVPs and molecular biology of resistance induced by AVPs in rice are in progress.

IRRN 21:2-3 (August-December 1996)


Integrated pest managementinsects

Hibernation sites of the rice stalk stink bug Tibraca limbatiwentris in the central region of Rio Grande do SUI, Brazil
D. Link, J. G. Naibo, and J. P. Pelentir, Departamento de Defesa Fitossanitaria, Centro de Ciencias Rurais-UFSM, 97119900, Santa Maria-RS, Brazil
Average number of bugs on some plant species. Santa Maria, Rio Grande do SUI, Brazil, 199294. Species Bushes (no.) Bugs bush-1 (no.) 33.99 14.00 9.17 5.73 2.20 2.14 1.10 Standard Range deviation

We surveyed the preferential sites where the rice stalk stink bug (RSSB) Tibraca limbativentris (Hemiptera: Pentatomidae), an important irrigated rice pest in southern Brazil, hibernates between growing seasons. Visual observation of densely massed plants was made. The insects were counted in each plant located 2.5-100 m away from field borders. In 1992, we surveyed fallow fields in three counties: five in Sao Sepe, two in

1992-93 Eryngium eburneum a Erianthus sp. b Andropogon lateralis b Paspalum urvillei b Baccharis trimerac Panicum prionitesb Senecio brasiliensisc 1993-94 Andropogon lateralis b Schizachyrium microstachyum b Eryngium eburneum a Paspalum urvillei
a Apiaceae

330 330 330 330 330 330 330

0.86 1.57 0.69 0.93 1.20 0.70 1.15

5-100 1-101 1-24 1-26 1-10 1-3 1-5

600 600 600 600

11.58 6.89 5.02 2.86

1.22 0.71 1.01 0.69

1-100 1-30 1-29 3-23

Agudo, and three in Santa Maria. In 1993, 20 fallow fields in Santa Maria were visited. Fields that were flooded after harvest or were not infested during the growing season were not included. Higher RSSB populations were observed on the bigger and more compact clusters, which provide an ideal refuge for both adult and young forms. The bugs were near the soil surface. where air moisture is higher. More than 90% of the hibernating forms were located up to 10 m from the field border. The distribution of the hibernating forms of RSSB was observed at 2.5, 5.0, 7.5, 10, 15, 25, 50, and 100 m from the field border. For each site, 10 samples were collected from a 2-m2 area. More than 40 plant species were found to shelter RSSB, but the last four species listed were preferred by the bugs (see table).

c Asteraceae (=Compositae).

(=Umbelliferae), b Poaceae (=Gramlneae),

Integrated pest managementweeds

Predicting the effects of Echinochloa crus-galli on rice (Oryza sativa L.) using the ecophysiological model INTERCOM
F. J. Abamu, IRRI

Weed research is now focused on finding alternative methods to control weeds as well as on understanding new concepts that underlie weed management. To support this new approach, tools that can be used to understand the complexities of crop-weed relations and to predict weed effects are important. Between 1994 and 1995, we conducted field experiments and simulation studies at IRRI to evaluate the ability of an ecophysiological crop-weed competition model (INTERCOM) to simulate the effects on rice (Oryza sativa L.) of competition from different densities of a weed (Echinochloa crus-galli L. Beauv), emerging at different times, and allowed to grow with the crop for the rest of the season. A split-plot expe-

rimental design was used. Weed density and time of weed emergence/flush were the treatments. There were three weed density levels 0, 20, and 40 plants m -2 (designated as weedfree check, low, and high densities, respectively). All weed populations were introduced artificially. To imitate natural conditions in which weeds emerge in different flushes, the weed density treatments were controlled in such a way that weeds emerge at two separate times in the ricefield. These were 2-5 d after rice transplanting (DAT) (early emergence) and 12-15 DAT (late emergence). Nitrogen was applied at a rate of 150 kg ha-1. Simulation runs were made with weather data of 6 yr, including the actual year of experimentation. Results showed that time of weed emergence was a more critical factor than weed density. Competition from weeds that emerged late (both high and low densities) did not have significant (P>0.05) effects on rice. This is because the cultivar used (IR72)

closed its canopy at about 25-30 DAT, thereby increasing the shade over the lateemerging weeds and limiting the light capture ability of this set of weeds for the benefit of the rice crop. It has been reported that the growth, development, and competitiveness of the weeds (Echinochloa) are adversely affected by shade. When the seeds emerged early, rice plant height did not significantly (P>0.05) respond to competition from both high and low weed densities, whereas there were significant (P<0.05) reductions in rice leaf area index (LAI) from maximum tillering. The model comparison for plant height prediction showed that INTERCOM adequately simulated rice height from transplanting to maturity. Simulated and observed heights at maturity did not differ significantly (P>0.05). The pattern of simulated LAI development in time was also similar to that observed, but deviation of the observed from the simulated was significant (P>0.05) around maximum LAI. LAI was underpredicted at this growth stage in most


IRRN 21:2-3 (August-December 1996)

2. Comparison between simulated and observed dry matter of rice from transplanting to maturity. In the legend, ds and ws = crop seasons (dry and wet), H and L = weed densities (high [40], 1. Response of observed and simulated grain yield of rice to competition from different densities of weeds ( Echinochloa crus-galli ) emerging at different times. IRRI, 1994-95. and low [20] plants m -2) 1 and 2 = time of weed emergence (1 = early [2-5 DAT], 2 = late [12-15 DAT]). IRRI, 1994-95.

treatments, suggesting a need to further evaluate the model's ability to predict LAI. Dry matter production and grain yield (Fig. 1) were significantly (P<0.05) reduced by competition from early-emerging weeds. Reduction was greater at high than at low weed density. Simulated and ob-served grain yields across all weed densities and time of weed emergence treatment combinations were highly correlated (R=0.97**) and did not differ significantly (P>0.05). Simulated and observed dry matter content from transplanting to maturity were also correlated (R=0.92**) (Fig. 2). INTERCOM can provide insights into the effects of Echinochloa competition on rice when the crop receives adequate N, irrespective of time of weed flush or weed density.

Weed control in direct seeded puddled rice

P. S. Bisht, P. C. Pandey, and P. Lal, Agronomy Department, G. B. Pant University of Agriculture and Technology, Pantnagar, Nainital 263145, India

Weed control in direct seeded rice is a serious problem in northern India. Manual weeding controls weeds effectively but availability and cost of labor limit its adoption. There are safe and nonphytotoxic herbicides, but their effectiveness has not yet been evaluated.

Butachlor, pendimethalin, and thiobencarb are the recommended herbicides, but at least one additional hand weeding is required with these. The use of safener, an antiphytotoxicity compound that eliminates the toxic effect of herbicides on rice seedlings, offers opportunities to control

Effect of herbicides and cultural methods on grain yield, panicles m -2, panicle weight, and weed dry weight in direct-seeded puddled rice. Pantnagar, India, 1993-94. Time of application (DAS)a Dose (kg ai ha -1) Concentration 1993 Grain yield (t ha -1) 5.3 5.1 2.5 EC b EC EC EC EC EC EC EC EC EC EC EC EC EC EC EC 3.7 2.6 4.7 4.6 3.4 3.0 3.6 4.0 3.4 3.0 3.7 1.2 1.1 21.0 Panicles No. m -2 475 357 432 392 378 400 385 322 358 373 355 389 429 391 252 97 18 Weight (g) 1.42 1.66 0.98 1.23 0.93 1.06 1.39 1.22 1.08 1.07 1.16 1.30 1.13 1.15 0.99 ns 26.00 Weed dry weight (g m-2) 10 91 247 305 430 225 267 279 394 422 404 305 373 360 473 200 46 Grain yield (t ha -1) 5.9 5.4 5.4 3.7 4.4 3.4 4.1 3.8 3.4 1.5 0.9 3.5 2.2 1.9 0.8 1.5 23.0 1994 Panicles No. m-2 468 486 562 512 463 484 498 440 425 244 141 355 250 202 150 150 28 Weight (g) 1.05 1.18 0.93 0.99 1.06 0.87 0.89 0.90 0.85 0.69 1.31 1.05 1.15 1.06 0.78 0.34 24.00 Weed dry weight (g m-2) 35 175 87 563 304 678 357 376 444 601 960 356 607 678 935 213 31


Weed-free (check) Hand weeding twice High plant population (125 kg seed ha-1) + 1 hand weeding Pretilachlor + safener Pretilachlor + safener Pretilachlor + safener Pretilachlor + safener Butachlor + safener Butachlor + safener Butachlor + safener Butachlor + safener Butachlor Butachlor Butachlor Anilophos Anilophos Anilophos Thiobencarb Thiobencarb Weedy (nonweeded control) LSD (5%) CV (%)

20 and 40 20 (1993) 30 (1994) 3 6 3 6 3 6 3 6 3 3 6 3 3 8 3 3

0.4 0.4 0.6 0.6 1.0 1.0 1.5 1.5 1.0 1.5 1.5 0.3 0.4 0.4 1.0 1.5 -

30 30 30 30 50 50 50 50 50 50 50 30 30 30 50 50 -

aDAS = days after seeding. b EC = emulsifiable concentrate.

IRRN 21:2-3 (August-December 1996)


weeds in direct seeded rice more effectively. A field experiment was conducted during the 1993-94 wet season at the Crop Research Centre, Pantnagar (29 N, 79 E, 244 m altitude). Soil is silt loam (Aquic Hapludoll) with pH 7.9, 1.1% organic C, 0.1 % total N, and CEC 20 meq 100 g -1 soil. The treatments (see table) were tested in a randomized block design with four replications. Short-duration variety Govind (90 d) was used in 1993 and Saket 4 (110 d) was planted in 1994. Pregeminated seeds were uniformly broadcast in puddled soil at 100 kg dry seeds ha-1 on 10 Jul l993 and 2 Jul 1994. A single dose of 18 kg P ha-1 and 33.2 kg K ha-1 was applied along with three splits of 120 kg N ha-1 (1/2 as basal dressing, 1/4 at tillering, and l/4 at panicle initiation). Soil was kept near saturation during seedling establishment and was later flooded with 0-5 cm water. Weed samples were taken from a 0.25m2 area and were dried at 70 5 C. Dry weights were recorded. The common weeds found were Echinochloa colona, Panicum sp., Digitaria sp., Cyperus iria, Cyperus difformis, Ischaemum rugosum, Commelina henghalensis, Casulia axillaris, and Eclipta prostrata. Grain yield was recorded from a 6-m 2 net plot area. Plant samples from 0.5 m 2 were taken at harvest to evaluate yield components. Phytotoxicity problems were not observed in treatments with the new herbicide formulation (herbicides with safener). Weeds were controlled effectively in both years. Pretilachlor + safener at 0.6 kg ai ha-1 applied 3 d after sowing was found most effective in controlling weeds. Yield was statistically at par with that of weed-free check and that of the crop that received two hand weedings. Compared with butachlorand thiobencarb-treated plots, high plant density with one manual weeding at 30 d was found to suppress weeds significantly (see table).

Integrated pest managementother pests

Effect of soil solarization of rice nursery beds to suppress plant parasitic nematodes
A. K. Ganguly, Pankaj, and A. Sirohi, Nematology Division, Indian Agricultural Research Institute (IARI), New Delhi 110012, India

An advantage of solarizing nursery beds, aside from a favorable cost-benefit ratio, is its potential to suppress other soilborne pests besides the target pest. We explored the possibility of using this nonchemical method to suppress plant parasitic nematodes that infect rice seedlings.
Table 1. Effect of solarization and organic amendments on growth characters of rice seedlings. New Delhi, India. Treatment Length (cm) Shoot Control Solarization Organic amendment SE LSD (5%) 26.20 39.30 38.40 1.31 2.75 Root 6.70 13.80 14.80 0.78 1.65 Weight [g) Shoot 0.42 1.22 1.39 0.11 0.23 Root 0.165 0.631 0.644 ns ns

The experiment used a randomized block design and was conducted in the IARI farm during the 1995 summer season. The beds were infested with moderate to heavy population of plant parasitic nematodes. mostly root-knot nematodes Meloidogye spp. Tillage was achieved by digging to a depth of about 10 cm, followed by light irrigation. Mulching was done with 60 mthick, transparent, white polyethylene sheet buried on all sides. Daily temperature was recorded for each treatment at 0800 and 1500 h for the entire 3-wk duration of solarization and organic amendment treatments. The organic amendment treatment consisted of incorporating mustard cake (0.5% w/w) into the soil 15 d prior to sowing of rice seeds. At the end of the solarization treatment, polyethylene mulch was removed and soil samples at 0-15 cm depth were drawn. The modified sieving and decanting method by Cobb was used to extract nematodes from 150 cm3 of soil sample. The nursery beds were loosened and irrigated; seeds of rice cultivar Basmati 370 were then sown. Seedlings were harvested 30 d after sowing and data on plant growth characters and nematode population in the soil were recorded.

Table 2. Effect of solarization and organic amendment on population of nematodes infecting rice seedlings. New Delhi, India. Nematode species Meloidogyne spp. Control Solarization Organic amendment Helicotylenchus spp. Control Solarization Organic amendment Tylenchorhynchus spp. Control Solarization Organic amendment Pratylenchus spp. Control Solarization Organic amendment Aphelenchus avenae Control Solarization Organic amendment
a+ = increase; - = decrease.

Nematode population 250 cm3 of soil Pre-treatment 172 115 116 13 71 41 23 61 33 26 37 14 105 56 78 Post-treatment 250 15 33 33 30 23 40 25 22 45 18 7 139 13 57

Percentage Increase/ decrease a 45.3 (+) 87.0 (-) 71.5 (-) 153.8 (-) 57.7 (-) 43.9 (-) 74.0 (+) 59.0 (-) 33.3 (-) 73.0 (+) 51.3 (-) 50.0 (-) 32.4 (+) 76.8 (-) 27.0 (-)


IRRN 21:2-3 (August-December 1996)

Polyethylene mulching increased mean maximum temperature at all observation times (data not shown). An increase of 713 C in the mulched bed over the nonmulched bed was observed during the 3-wk period of solarization. Mulching significantly increased seedling growth (Table 1).

With regard to plant growth parameters, no significant differences were observed between treatments. The improved growth of seedlings in the solarized bed could have been due to the good control of plant parasitic nematodes, other soilborne pathogens, and weeds.

Good control of plant parasitic nematodes (50-87% reduction) was achieved in the solarized beds (Table 3). The decrease in root-knot nematode populations was most notable. Solarization controlled nematodes better than did organic amendment.

Water management
Supplemental irrigation for dry seeded upland rice
V. V. Angadi, Agricultural Research Station, Sankeshwar 591314, Belgaum; and P. N. Imapathy, Main Research Station, University of Agricultural Sciences, Dharwad 580005, Karnataka, India

Upland rice in Karnataka is often exposed to moisture stress due to dry spells that coincide with the critical growth stages of the crop. Supplemental irrigation using water from small tanks found all over the rainfed rice tract may increase rice yield. We studied the effect of supplemental irrigation at various growth stapes on grain yield of rice during the 1992 and 1994 wet seasons. The experiment consisted of rainfed rice and rice grown with supplemental irrigation (5 cm water depth) at booting, flowering, and grain-filling stages and their combination (see table for treatments). Daily irrigation from booting to 10 d before harvest was also one of the treatments. Due to high percolation in the unpuddled soil, the fields were flooded only for several hours after irrigation or heavy rainfall.

The 5.0- 3.4-m plots were laid out in a randomized complete block design with three replications. Soil was silty clay loam with 0.4% slope. Plots were separated by well-compacted bunds and buffer channels. Land was plowed and harrowed and direct seeded with rice cultivar Avinash (140 d) at the onset of the rains. A common fertilizer dose (100 kg N, 22 kg P, and 41 kg K ha-1) was applied to all plots. Total rainfall was 1150 in 1992 and 1135 mm in I994 with 75 and 79 rainy days during the respective cropping season. Except for a dry spell of 11 d before booting in 1992, good rainfall was received during other stages in both years. Irrigation at booting, alone or in combination with other stages, relieved the stress caused by the dry spell in 1992 and significantly increased yields. Irrigation only at flowering and grain-filling stages in 1992 and all irrigation treatments in 1994 did not increase yield. This was attributed to suffi cient rainfall in 1994. During flowering and grain filling in 1992, plants were not stressed, even without irrigation. Best results are obtained when timely supplemental irrigation is given only when dry spells occur during critical stages (booting, flowering, and grain filling).

Rice yield under sprinkler irrigation

H. Srek, H. Aydin, Thrace Agricultural Research Institute (TARI), Edirne, Turkey; R. Cakir, H. Karaata, Atatrk Soil and Irri, gation Research Institute (ASIRI), Kirklareli, Turkey; M. Negis and H. Kusku, TARI

Grain yield of rice as influenced by supplemental irrigation management. a Karnataka, India. 1992 and 1994. Treatment Grain yield (t ha -1) 1992 3.8 5.3 4.4 3.9 5.4 4.2 5.4 5.5 5.2 1.3 1994 5.4 5.4 4.9 5.6 5.7 5.9 5.0 5.2 5.3 nsb Mean 4.6 5.4 4.7 4.8 5.5 5.0 5.2 5.3 5.3 % increase over rainfed treatment 17 1 3 20 9 16 16 14

Rainfed Irrigation at booting Irrigation at flowering Irrigation at grain filling Irrigation at booting and flowering Irrigation at flowering and grain filling Irrigation at booting and grain filling Irrigation at booting, flowering, and grain filling Daily irrigation from booting up to 10 d before harvest LSD (0.05)
a Water depth applied at each irrigation: 5 cm. b ns = not significant.

We investigated rice yield under sprinkler irrigation during three growing seasons 1991-93 in Turkey. The experiment was laid out in a split plot design with three replications. Main plots were two irrigation intervals: 4d (A1) and 8d (A2). The subplots were irrigation depths (or amount): same amount of pan evaporation (B1), 50% more than pan evaporation (B2), and 100% more than pan evaporation (B3). Plots were 12 m2. Seeds of rice variety Ergene (120 d) were drill-seeded in early May at I80 kg ha-1 using 25-cm row spacing. N (150 kg ha -1) was applied in two splitsl/2 basal and 1/2 60 d after planting. P (35.2 kg ha -1 ) was applied during soil preparation. Weeds were controlled with two herbicide applications, before drilling and 40 d after drilling. Soil was sandy clay silt with 1.3% organic matter and pH 6.5. All treatments received similar amounts of irrigation water until seedling emergence to ensure adequate stand establishment. Water use efficiency (WUE) was calculated by dividing grain yield (kg ha-1) by total amount of water applied (mm) for the whole crop. Irrigation interval has no significant effect on yield (see table). No interaction between irrigation interval and irrigation depth was observed. However, irrigation water depth influenced yield significantly. As the amount of sprinkler irrigation water decreased, the yield decreased. B3 had the highest yield while B2 gave the highest WUE.

IRRN 21:2-3 (August-December 1996)


Average yield of Ergene under flooded irrigation conditions (2000 mm water application) was about 6 t ha-1. With the B3 treatment (1172 mm water application), yield was 4.7 t ha-1.

Sprinkler irrigation can be thus used to increase WUE in conditions where water is scarce. Extra investment is, however. needed to install and operate the system.

Yield of rice under sprinkler irrigation. Edirne, Turkey. 1991-93.

Treatmenta 1991 A1B1 A1B2 A1B3 A2B1 A2B2 A2B3 2.7 d 5.5 ab 5.9 a 3.7 cd 4.6 bc 4.8 ab Grain yield (t ha -1) 1992 1.8 b 3.2 abc 3.6 ab 1.6 c 3.5 ab 4.4 a 1993 2.2 c 3.2 b 4.8 a 2.4 bc 4.4 a 4.8 a Mean 2.3 c 3.9 b 4.7 a 2.5 c 4.2 ab 4.7 a 1991 641 874 1107 641 874 1107 Water applied (mm) 1992 681 944 1206 681 944 1206 1993 804 1004 1203 804 1004 1203 Mean 709 940 1172 709 940 1172 Water use efficiency (kg ha -1 mm -1) 1991 4.4 6.3 5.3 5.7 5.3 4.4 1992 2.7 3.3 3.0 2.3 3.7 3.6 1993 2.7 3.1 4.0 2.9 4.4 3.9 Mean 3.2 4.2 4.0 3.6 4.4 4.0

4 d

8 d

F values A B AB CD (0.05) CV (%)

at 1% level.

0.87 24.36**b 4.18 1.07

0.35 9.28** 0.50 1.76 12.49

14.62 49.67** 3.77 0.82 18.06

0.43 52.59** 0.32 0.66 12.03


aA: Irrigation interval: A1 = 4 d; A2 = 8 d. B: lrrigation depth (amount): B1 = same as pan evaporation; B2 = 50% of pan evaporation, B3 = 100% of pan evaporation. b ** = significant

Farming systems
Effect of different rice cultures on crop yield in rice-based systems
T. S. Verma and P. K. Sharma, Soil Science Department, Himachal Pradesh Agricultural University (HPAU), Palampur 176062, Himachal Pradesh, India

We investigated the performance of three rice cultures(i) irrigated lowland (IL, rice seedlings transplanted in puddled soil under irrigated conditions); (ii) rainfed lowland (RL, crop established by dry seeding, followed by wet tillage at the 3-4 leaf stage under rainfed conditions [locally called halod]); and (iii) rainfed upland rice (DSR, rice dry seeded and raised as an upland crop under rainfed conditions)and their subsequent effects on the yield of the

following upland crops of wheat (Triticum aestivum), linseed (Linum usitatissinum), and raya (Brassica juncea). We conducted this experiment at the HPAU experimental farm in Palampur (326'N, 763'E, 1300 m altitude) during the 1993 and 1994 wet seasons. Soil was silty clay loam (Typic Hapludalf). Plots (5 3 m) were arranged in randomized complete block design with nine replications for rice and three replications for each upland crop. All crops received fertilizers at the recommended rates. In both years, rice grain yield followed this trend: IL > RL > DSR. The differences were statistically significant. Grain yield of wheat and raya were highest in DSR and lowest in IL plots; linseed yield was not affected by the land preparation technique

Grain yield of crops (t ha-1) in rice-based cropping systems involving three rice cultures. Rice culture Irrigated lowland Rainfed lowland Direct seeded rice LSD (0.05) Rice 1991 4.2 3.3 2.9 0.3 1992 4.5 3.5 3.1 0.3 1991 3.0 3.6 3.9 0.3 Wheat 1992 3.3 3.9 4.1 0.2 Linseed 1991 1.2 1.4 1.5 ns 1992 1.3 1.4 1.6 ns 1991 0.9 1.3 1.6 0.1 Raya 1992 1.0 1.4 1.7 0.2

adopted in the preceding rice crop. The energy required to prepare the land after rice harvest was significantly affected by the land preparation technique used in rice. Puddling is the most intensive tillage practice in rice, followed by halod and dry tillage. Manual land preparation with the use of spades required about 967 h ha -1 in IL, 633 h ha-1 in RL, and 483 h ha-1 in DSR. Considering that the energy of an adult worker equals 1.96 MJ human h -1, land preparation after IL, RL, and DSR rice required 1895, 1241, and 947 MJ ha -1, respectively. Thus, the energy required to prepare the seedbed after RL rice was about 2/3, and the energy needed for the crop after DSR was about 1/2 that of IL rice. Rice in silty clay loam soil performed best under lowland irrigated conditions. But this rice culture did not favor the following crops of wheat and raya. They performed best after DSR. Linseed was unaffected by rice culture. Because wheat is next to rice in importance, appropriate tillage technology has to be developed to maximize crop yields in rice - wheat cropping systems.


IRRN 21:2-3 (August-December 1996)

Farm machinery
A low-cost, in-store dryer for smallscale farmers
Pham Hieu Hien, Le Van Ban, and Bui Ngoc Hung, University of Agriculture and Forestry (UAF), Ho-Chi-Minh City, Vietnam

We developed a low-cost in-store dryer, called SRR-1, at UAF to meet the needs of small-scale farmers in Vietnam. They require a dryer with adequate capacity, high-quality dried grain. and reasonable investment and drying cost. The design of the SRR-1 dryer is based on the principle of low-temperature in-store drying. The dryer consists of three components: a two-stage axial fan, an electric heater, and a bamboo mat drying bin (see figure). The drying bin consists of two concentric bamboo mat cylinders, 0.4 and 1.5 m in diameter and 1.1 m high. The bin can contain 1 t of rough rice. A 0.5-hp electric motor drives the fan. Two 350-mm-diameter rotors are mounted on both ends of the motor shaft inside a steel casing. (Plastic rotors are locally made and readily available in the market.) The fan is positioned on top of the inner bamboo mat cylinder; it can deliver 0.3 m 3 s-1 at 400 Pa static pressure. The heater is a 1000-W resistor from an electric stove; it is mounted beneath the lower rotor. Supplemental heat from the resistor is sometimes used at night or during times of continuous rain. During the day and the first two nights, the fan is turned on to aerate the rice. Supplementary heat is provided on the third night. In succeeding nights, the blower is turned off because the grain moisture content is low enough for temporary storage. At an initial moisture content of 26-28% (wet basis), drying time is 80-85 h batch-1, for which total power consumption is 80

Schematic of SRR-1 dryer.

kWh batch-1. The dryer does not require much, and its manufacturing cost is less than US$l00. (SRR stands for very cheap dryer in Vietnamese). The aired grain quality from SRR is outstanding. Moisture gradients are less than 1 %. and head rice recovery is increased by at least 2% when compared with sun dryingand the weather risk is eliminated. After drying, the

rice is stored in the drying bin and the fan is moved to the next bin. Drying costs at UAF were $5 batch -1, including dryer depreciation. The dryer is designed for farm-level use, mainly for drying and storing rough rice for household consumption. More than 200 units were sold wihtin 1 yr of introduction in Vietnam. Many farmers in southern Vietnam are enthusiastically accepting the SRR-1 dryer.

IRRN 21:2-3 (August-December 1996)


Research methodology
A simple and rapid method for isolation of bacterial genomic DNA
M. L. C. George, V. Quinto, M. Villamayor, and R. J. Nelson, IRRI

The polymerase chain reaction (PCR) is now an important tool for ecological studies because of its simplicity and capacity to rapidly screen a large number of samples with a minimal amount of DNA. It is especially useful for studying the population biology of plant pathogens, such as Xanthomonas oryzae pv. oryzae (Xoo) (which causes bacterial blight of rice), where massive sampling is required for analysis. While it is possible to obtain PCR fingerprints directly using whole bacterial cells, more consistent results are obtained using good-quality DNA as a template. Genomic DNA is isolated in a multistep process, typically involving sodium dodecyl sulfate-mediated cell lysis. DNA extraction, alcohol precipitation, and ethanol wash. Because it is a lengthy process, the isolation of genomic DNA is a bottleneck in large-scale application of PCR as a fingerprinting tool. Here we report a simple alcohol lysis procedure to isolate PCR-quality genomic DNA of bacterial cells. In this method, ethanol functions both to lyse the cells and to precipitate the DNA. Cells from Xoo, X. o. pv. oryzicola, Pseudomonas blumei, Burkholderia glumae, Asorhizobium sp., and Escherichia coli plate cultures were collected using the wide end of a sterile toothpick (Fig. 1) and washed in 400 l of water to remove the extracellular polysaccharides. Cells were pelleted in a microcentrifuge by spinning for 2 min (12,400 rpm) and the still turbid supernatant was discarded. The cells were resuspended in the residual liquid, and 200 l of 95% ethanol was added to a final concentration of approximately 70% alcohol. After 10-30 min, the suspension was spun for 30 s to pellet the cell debris, and the supernatant was transferred to a new tube. DNA was pelleted by spinning for 5 min, dried, and

then dissolved in 10 l of sterile water. To check the viability of the ethanol-treated cells, the pelleted cells were resuspended in the residual alcohol, and the whole suspension was plated. One l of the DNA preparation was used as template in a 25-l reaction volume. Primers corresponding to conserved sequences in bacterial repetitive elements (REP1R-I, 5'IIIICGICGICATCIGGC 3' and REP2-I, 5'ICGICTTATCIGGCCTAC 3') were used for all the strains under conditions prescribed for rep-PCR. Additionally for the Xanthomonas strains, 0.5 M of the primers JELl (5'CTCAGGTCAGGTCGCC 3') and JEL2 (5'GCTCTACAATCGTCCGC 3') were used with 185 M each of four dNTPs. approximately 2.5 units of Taq polymerase in an incubation buffer amended with 10% dimethylsulfoxide (v/v) and 7.5 l of 1 M Tris, pH 9.5. The reaction mixture was initially denatured for 1 min at 94 C and then subjected to 30 cycles of PCR (10s denaturation at 94 C, 1 min annealing at 62 C, and 10 min extension at 65 C) and a final extension for 15 min at 65 C using a Perkin Elmer Cetus DNA thermal cycler. The PCR products were resolved in a gel containing 0.5% agarose and 0.75% Synergel , visualized by staining with ethidium bromide and photographed using Polaroid 667 film.

With DNA prepared by this method as template for PCR, fingerprint patterns generally comparable with those obtained using DNA prepared by conventional means were obtained from 3-d-old, 1 -wkold, and 5-wk-old plate cultures, and from 1-d-old broth culture (Fig. 2). Similar patterns were also obtained from cells stored in ethanol for 6 wk prior to DNA isolation. The alcohol lysis method is simple and fast, requiring about 15 min, and produces enough DNA for several amplification reactions. It also renders cells nonviable and thus suitable for storage and transport and eventual PCR fingerprinting of pathogen sample ex situ.

2. PCR fingerprint patterns of Xanthomonas oryzae pv. oryzae strains PX0215 (A), PX0143

(B), PX01 (C) and PX0204 using template DNA prepared by the potassium acetate method (lanes 1, 6, 11, and 16), and by the alcohol lysis

method using 3-d old (lanes 2, 7, 12, and 17), 1wk-old (lanes 3, 8, 13, and 18), and 5-wk-old 1. Cells collected using a sterile toothpick from a 3-d-old plate culture of Xanthomonas oryzae pv. oryzae. DNA isolated from the cells by the alcohol lysis method was resuspended in 10 I of water, and 1 I was used as a template for PCR amplification. (lanes 4, 9, 14, and 19) plate cultures, and from 1-d-old broth cultures (lanes 5, 10, 15, and 19). The primers used in the PCR method were JEL1/JEL2. The DNA molecular size markers are on the lanes labeled M on the left (1-kb ladder) and right (BioMarker EXT).


IRRN 21:2-3 (August-December 1996)

A simulation model of rice sheath blight epidemics (I) Structure and model development
S. Savary, IRRI-Institute Franqais de recherthe scientifique pour le dveloppement en coopration (ORSTOM) Joint Project on Rice Pest Characterization, IRRI: and L. Willocquet, ORSTOM

A systems model was developed to simulate rice sheath blight (ShB) epidemics over time. The model considers the two phases of ShB epidemics: mobilization of soilborne primary inoculum and secondary spread. The mechanism involved in secondary spread is a key feature of ShB epidemiology. Once infected, a tiller may in turn infect its neighbors via direct contact between leaves, which enable mycelial strands of the fungus to spread in the canopy. While the primary phase of epidemics is dependent on soilborne inoculum, the secondary phase involves leafborne inoculum and depends on crowding of canopy and climatic factors such as leaf wetness, light, and relative humidity for its multiplication. The system under consideration in the model consists of a 1-m2 rice crop. represented by a growing population of tillers. Tillers may belong to two categories: healthy or diseased. Two independent infection processes may lead to the accumulation of diseased tillersprimary infections generated from primary inoculum and secondary infections resulting from the progress of the pathogen in the crop canopy. The rate of change of infected tillers is thus represented by the sum of a rate of primary infection and of a rate of secondary infection: dNi/dt=(dNi/dt) p + (dNi/dt)s . The rates of primary and secondary infections are described by a monomolecular model and a modified logistic model: dNi/dt=rp P(1-(N/(N+Ni))+rs Ni(l-N/ (N+Ni)) a where N is the number of healthy tillers (per m2 ); Ni is the number of infected tillers (per m2 ); rp and rs are respectively intrinsic rates of primary and secondary infection; Pis the current amount of primary inoculum (per m2 ); and a is a parameter for disease aggregation. The structure of the model is described in the flow chart of Figure 1.

1. A systems model for rice ShB epidemics. State variables are indicated by rectangles, flows of individuals by double arrows, parameters (estimated or computed) by circles, and numerical relationships by simple arrows. The intrinsic rate of primary inoculum decay under flooded conditions was derived from Roy (1986).

The term (1-N/(N+Ni)) is commonly called correlation factor. It represents the fractionof healthy tissues (tillers)i.e., tissues available for infection. One major underlying assumption of the model is materialized by a difference between correction factors for rates of primary and secondary infection. and the introduction of a new term, a, that reflects the limited accessibility of available (healthy) tissues for infection in the secondary phase. The distribution of primary, soilborne inoculum is considered random due to field plowing, harrowing, and flooding, which results in a randomly distributed probability of each healthy tiller being infected by soilborne inoculum. This assumption is translated in the model by the use of a correction factor (COFR) in this simplest form for the rate of primary infection: (1 - (N/N+Ni)). When disease becomes established in the crop and progresses within the canopy. however, the typical aggregated structure of

the disease develops as a result of the short range of dispersal of the pathogen. In the course of an epidemic, an additional constraint to disease multiplication therefore develops. In addition to the progressively declining proportion of tillers available to infection. there is a limited number of tillers that actually are accessible to infection from the diseased tillers. The parameter a of the correction factor for secondary infection, (1 - (N/ N+Ni)) a , represents disease aggregation, or alternately, the limited accessibility of tillers available for infection. The amount ofprimary inoculum, P, also varies over time, as result of the flooding of the soil on which the crop is grown. Variation of P is assumed to follow an exponential decay: dP/dt=k P. Some additional features were added to the model and are summarized in Figure 1: 1) the size of the tiller population is made variable over time, reflecting crop growth; a logistic increase (with relative growth rate

IRRN 21:2-3 (August-December 1996)


[RCRI) was assumed to adequately represent total tiller growth; 2) tillers may recover from infection, and a rate of disease recovery (Recov) was included; 3) severe infection on individual tillers may lead to their death, and a rate of tiller death (Rmort) was included; 4) tiller senescence (Tsen) was also included, and represented as an exponential decay function. This series of equations was linked together and integrated using a 1-d time-step. Empirical values were derived for the rates of senescence, mortality, and recovery from field experiments at IRRI. An intrinsic rate of primary inoculum decay was derived from published data (k = -0.11; Roy [1986]). Data from field experiments (Gou et al 1983) were then used as a basis for model verification, and to estimate the three parameters of the equation for the rate of infection: rp , rs , and a. This was achieved

, ,

2. Comparison of observed (dots) data and simulated (line) ShB incidence. Observed data were derived from Gou et al (1983).

The results of the simulation indicate that this system model has the potential of adequately describing ShB epidemics, and may serve as a basis for further improvement (Figure 2). Experiments are currently under way at IRRI to further assess the performance of the model and to better document rs in terms of climatic factors and contact frequency among tillers in a rice crop.

Cited references
Gou FS, Li XQ, Xu CL. 1983. Study on the spatial distribution patterns of the rice sheath blight plant in rice field and its practical amplications. Acta Phytopathol. Sin. 13:7734. Roy AK. 1986. Survival of sclerotia of Rhizoctonia solani F. sp. sasakii in relation to moisture regime of soil. Indian Phytopathol. 39:259-263.

using disease incidence (i.e.. Ni/(N+Ni) as a synthetic variable representing the processes underlying a ShB epidemic, and the DUD interactive technique of the NLIN procedure of the statistical package SAS. Using this technique, parameter estimates were r p = 0.04, r s = 0.09, and a = 1.77.

A simulation model of rice sheath blight epidemics (II) Model per formance derived from sensitivity analysis
S. Savary, IRRI-ORSTOM Joint Project on Rice Pest Characterization, IRRI; and L. Willocquet, ORSTOM

A preliminary simulation model of rice sheath blight (ShB) epidemics was developed, which incorporates a few important characteristics of the disease: a rate of primary infection (which is proportional to the amount of primary, soilborne inoculum, the amount of healthy tissues [tillers] available for new infection, and an intrinsic rate of primary infection, rp ) and a rate of secondary infection (which is proportional to the amount of infected tissues (tillers) and an intrinsic rate of secondary infection, rs). The rate of secondary infection also depends on the amount of tissues still available for infection and on an aggregation parameter, a. This model addresses a few research issues, such as 1) the relative effect of variation of parameter values on the behavior of the model; 2) the contribution

A series of simulation runs using a sheath blight simulation model, with varying values of three parameters: r p (a), rs (b), and a (c).

of primary inoculum, relative to that of secondary inoculum, resulting in spread of the disease within the rice crop canopy; and 3) the magnitude of the effect of disease aggregation on disease spread. The figure shows a series of scenarios where the three parameters were independently varied, using optimized parameter values (rp =0.04, rs = 0.09, and a = 1.77) as references. A series of ShB incidence progress curves were generated by increasing or decreasing r p and rs by 25% or 50% of their initial values. Because a cannot, by definition, be smaller than 1, the following values were tested: 1 (random distribution of disease throughout the epidemic), 1.39, 1.77 (optimized value). 2.16, and 2.54. Variation in rp (Fig. a) has a small effect on the shape of disease epidemics, a decrease of rp resulting primarily in a delay of the epidemics with the same speed. Variation in rs (Fig. b) has strong effects on both the slopes of the curves and the terminal incidences. As expected, variation in a (Fig. c) affects the shape of curves in a later stage of disease epidemicthe larger the a value, the stronger the disease aggregation. the lower the accessibility of healthy tillers


IRRN 21:2-3 (August-December 1996)

to diseases, and the slower the epidemic when it enters the polycyclic phase. This behavior of the model indicates that it is sensitive to variation in rs. Future development of the model must consider the effect of time-dependent factors on

variation in rs so as to reflect variation of environment under which ShB epidemics develop. This suggests that the polycyclic nature of ShB must be considered a key characteristic of the disease for its management, and that measures taken in the course

of a cropping season should prove effective. It also indicates a comparatively high sensitivity of the model to variation in the aggregation parameter, and experiments are necessary for its estimation.

Use of CERES-Rice model to assess potential yield

G. S. L. H. V. Prasada Rao and N. Subash, Regional Agricultural Research Station (RARS), Pilicode 671353, Kasaragod District, Kerala, India

We conducted a field experiment at RARS, Pilicode (1 2 12' N; 75 10' E; 2755 mm rainfall, 4.4 h of bright sunshine d-1) to test

a. Actual and simulated grain yield (t ha -1) of rice varieties on different dates of transplanting during 1993 kharif.

b. Actual and simulated grain yield (t ha -1) of rice varieties on different dates of transplanting during 1994 kharif.

the CERES-Rice model developed under the IBSNAT Project. Soils were welldrained sandy loam of lateritic origin with pH 5.5. The model was tested for five cropping seasons using different dates of transplanting during 1993 and 1994 kharif and 1992-93, 1993-94, and 1994-95 rabi. Popular varieties Jaya (120- 125 d), Jyothi (110-125d), and Triveni (95-l05d) were used. The simulated grain yield through the CERES-Rice model was in good agreement with the observed yield during 1993 kharif while simulated grain yield was higher than that of the observed during 1994 kharif on all transplanting dates (Figs. a and b). Heavy rains during I994 kharif brought floods. Incidence of gall midge and rice bug was also high at the time, leading to a high percentage of grain chaffing (see table). The CERES-Rice model during rabi runs for a floodwater management scheme and takes into account bund height as well as floodwater maintained through irrigation at three different early stages of crop growth. The actual rabi crop at the field site was in a totally different condition from that mentioned previously. However, it should be noted that the CERES-Rice model simulated higher grain when run during rabi in all three test varieties on all transplanting dates. This study reveals that the CERES-Rice model needs validation for rabi while it could very well be used during kharif to assess potential grain yields based on simple daily weather variables.

Percentage of grain chaffing of three test varieties during 1994 kharif on different dates of transplanting. Kerala, India, 1992-95. Date of transplanting 8 15 22 29 6 Jun Jun Jun Jun Jul Av Jaya Good grains (%) 68.7 82.7 75.2 67.9 58.3 70.6 Chaffing (%) 31.3 17.3 24.8 32.1 41.7 29.4 Jyothi Good grains (%) 81.1 83.6 71.6 63.0 65.0 72.9 Chaffing (%) 18.9 16.4 28.4 37.0 35.0 27.1 Triveni Good grains (%) 90.3 88.7 82.8 84.0 64.2 82.0 Chaffing (%) 9.7 11.3 17.2 16.0 35.8 18.0

Effect of contour hedgerow on runoff, soil loss, and upland rice production
C. R. Subudhi, P. C. Pradhan, and P. C. Senapati, Dry Land Agricultural Research Project (DLAP), Orissa University of Agriculture and Technology, Phulbani 7602001, Orissa, India

Most of the hilly areas in Orissa lose a lot of soil during the wet season. Soil loss and runoff, as affected by different contour hedgerows, were quantified and their effects on upland rice production determined. The study was conducted at DLAP sites in Bhawanipatna in 1993 and in Phulbani in 1994. Soil at both sites is sandy loam with about 2% slope.

Rice (variety 9991 in 1993 and Heera in 1994) was dry-seeded in rows with 20-cm spacing along contours. Hedgerows separated rice terraces. The treatments (9 in 1993 and 7 in 1994) arranged in randomized complete block with three replications were the vegetation used to form the hedgerows (see table). The hedgerows were planted with 1-yrold grass seedlings (except for stylo where seeds were sown) 1 mo before the experiment. The distance between two hedgerows was 8 m. Plots were 25 m 3 m. Lengthwise, each plot comprised three terraces separated by two hedgerows. Daily runoff was monitored using a multislot divisor. Runoff collected by one slot is fed into in a drum installed at the end of the divisor. A sample of 250 ml was taken from each drum after thorough stirring and

IRRN 21:2-3 (August-December 1996)


Runoff, soil loss, and crop yield during 1993-94 cropping seasons. 1993 Runoff Soil loss (mm) (t ha-1) T1 No hedgerows T2 Cynodon dactylon hedgerow T3 Pennisetum purpureum hedgerow T4 Vetiveria zizanoidea hedgerow T5 Eulaliopsis binata hedgerow T6 Cymbopogon flexuosus hedgerowa T7 Stylosanthes hamata hedgerow T8 Hybrid napier hedgerow b T9 Fallow land T10 Broadcasting of dry seed, no hedgerow c Mean LSD (0.01) SE (m) 184.1 154.9 149.9 142.4 150.7 151.8 156.3 155.2 201.9 3.5 3.0 2.7 2.2 2.4 2.6 2.8 3.1 7.9 Crop yield (t ha -1 ) 1.2 1.0 1.6 2.1 1.4 1.9 1.3 1.6 Runoff (mm) 204.0 172.2 151.7 162.4 172.6 248.5 235.7 1994 Soil loss (t ha-1 ) 10.4 5.5 4.2 4.8 5.8 20.7 11.1 Crop yield (t ha -1 ) 1.3 1.7 2.0 1.9 1.8 1.0 Mean yield (t ha -1)


1.3 1.4 1.6 2.1 1.7 1.9 1.6 1.6 1.01

was evaporated to determine soil loss. The vetiver treatment (T4) produced the highest yield, having less runoff and less soil loss. Vetiver, with root penetration up to 1 m and consequently greater chances of survival than other grasses, was able to check the velocity of flow. Highest soil loss was observed in the fallow. Compared with leaving the land fallow (T9), 20% (1993) and 80% (1994) soil loss could be reduced by contour planting using vetiver.



1.5 0.2 0.1



1.5 0.08 0.03


a Total rainfall was 859 mm in 1993 and 1023 mm in 1994. bDone only in 1993. c Done only in 1994.

Recommendations of the 3rd lnternational Symposium on Hybrid Rice
The global requirement for rice by 2020 is expected to be around 800 million t compared with the current production of 520 million t. With shrinking resourcesparticularly arable land area, irrigation water, and energythe only option left is to increase production. Increasing rice production by 300 million t during the next 25 yr will be a very challenging task. Of the possible genetic approaches to meet this challenge, hybrid rice technology is an immediate option since it has been a proven technology over the past 2 decades in China and is now showing commercial prospects in India. The theme of the recent 3rd International Symposium on Hybrid Rice held in Hyderabad, India, 14-16 Nov 1996, was enhancement and sustenance of hybrid rice technology. The symposium was cosponsored by the Indian Council of Agricultural Research (ICAR), the United Nations Development Programme (UNDP), and IRRI. About 150 Indian delegates and 50 international delegates from 20 countries addressed various aspects and issues for improving the technology and making it available outside of China. These issues were deliberated in six sessions (current scenario, increasing breeding efficiency and enhancing yield heterosis, sustainability of hybrid rice technology, tissue culture and molecular approaches in heterosis breeding, toward true-breeding hybrids, and status of development and adoption of hybrid rice technology in various countries) and a meeting of the International Task Force on Hybrid Rice. During the 3 d of sessions, the following major recommendations emerged. Technology development Formation of a research network and effective collaboration with international agencies have been the main reasons for the remarkable success achieved in India. Use these approaches as models for similar achievements in other countries of South and Southeast Asia. Intensify efforts for development of hybrid rice technology in countries, such as India, Philippines, Vietnam, and Indonesia, where there is a demand for hybrid rice seeds and the capability to produce them. To meet consumer preference, emphasize improving milling, head rice recovery, and other quality characteristics. Especially emphasize incorporating resistance to major pests and diseases in the promising parental lines.

To enhance the level of heterosis, the

Chinese are successfully using two-line hybridsphotoperiod-sensitive gene male sterility (PGMS) and thermosensitive gene male sterility (TGMS) and are developing hybrids through intersubspecific crossing of indicas and japonicas. Vigorously pursue these approaches in the tropics using indica/ tropical japonica lines. Countries that grow japonica rices should be exploring the prospects of temperate japonica/ tropical japonica hybrids. For full exploitation of rice hybrids, develop special management practices that promote efficient use of nitrogen, phosphorus, and other nutrients and water. Improve parental lines through the use of random mating populations-the private sector, particularly, should pursue this to minimize dependence on the public sector for the supply of improved parental lines. Develop rice hybrids adapted to different ecosystemsespecially for the shallow lowlands, which are similar to the irrigated ecosystem. Intensify research to develop rice hybrids for the boro season (India, Bangladesh) and the spring season (Vietnam, Myanmar).


IRRN 21:2-3 (August-December 1996)

Technology transfer

To speed up large-scale adoption, create

possible tools including genetic engineering.

International Task Force on Hybrid Rice

To strengthen the hybrid seed industry

awareness and the demand for hybrid rice by conducting extensive on-farm trials, front-line demonstrations, and training programs. Expeditiously develop mechanisms for registration of parental lines so that they can be shared freely among collaborating countries.
Seed production

and the linkage between it and the hybrid rice research centers.

The higher cost of hybrid seed is a constraint to adoption of the technology. The following measures are recommended to reduce prices to affordable levels. Intensify studies on proper flowering synchronization of parental lines to get higher seed yields in the target areas; low seed yield is a major problem faced by seed growers. Emphasize producing and supplying parental lines with only the highest purity; identify suitable agencies to perform this task. Strengthen the breeding of cytoplasmic male sterile (CMS) lines (possessing higher outcrossing potential) and restorers (providing high pollen load). Identify in each country locations/ seasons that are most favorable for seed production. Economize the use of gibberellic acid (GA3) and simultaneously intensify the search for cheaper alternatives. Develop special management practices to obtain higher seed yields. Wherever feasible, involve interested nongovernment organizations in producing hybrid rice seed. Encourage governments to develop policies that aggressively advocate private sector participation in hybrid rice seed production and research. Conduct mass-scale in-country training on hybrid seed production.
Basic studies

The establishment of INTAFOHR was first proposed during the 2nd International Symposium on Hybrid Rice at IRRI in 1992, to promote the technology outside of China. Subsequently, this was endorsed by several countries in various international forums organized by FAO and IRRI. In October 1995, IRRI and FAO convened a joint meeting with country representatives and prospective donors (United Nations Development Programme, Asian Development Bank, and MAHYCO Research Foundation of India). The participants concluded that establishing INTAFOHR involving IRRI-FAO and the national agricultural research systems (NARSs) was an excellent idea and the prospective donors asked IRRI to submit a project profile for their consideration. After this was done, the donors asked IRRI to submit a detailed project proposal for possible funding. During the 3rd symposium. a panel discussion was held on 15 Nov 1996. During this discussion attended by the representatives of 16 countries, consensus was reached on the following points:



Participation in the Task Force is open to all interested countries.


Improved food security and sustainable development through increased rice production using hybrid rice.


molecular markers associated with quantitative trait loci for yield heterosis and subsequently incorporate these heterotic blocks into the parental lines. Speed up the work on apomixis as a long-term strategy by utilizing all

Initiate intensive work to identify

germplasm, information, and data from ongoing research and development programs on hybrid rice among interested partners. To strengthen national systems capabilities for applied and strategic research so that they can develop hybrid rice technology expeditiously. To intensify collaborative strategic research on hybrid rice by establishing effective regional/interregional collaboration on hybrid rice research and development. To assist member countries in formulation of appropriate policy incentives for hybrid rice development and use.

To promote free exchange of registered

General framework. Development of hybrid rice technology in member countries, particularly the charter members, will be expedited by 1) establishing goaloriented hybrid rice research and development programs aiming to strengthen human resources for research and development of hybrid varieties and hybrid seed production capacity, 2) establishing collaborative research linkages, and 3) freely exchanging breeding materials and information. Technology transfer will be expedited by strengthening on-farm testing and promoting the technology through appropriate channels and identifying policy interventions and by respective governments encouraging investment in hybrid rice research and seed production. Phasewise development. In recognition of the conditions governing rice production and the situations concerning research, development, and use of hybrid rice in different countries as well as the necessity to obtain time-bound outputs to sustain the activities of the Task Force, the Task Force will initially have three categories of membership: 1) charter (or core) members, 2) observers, and 3) affiliate members. All members will gradually become core members as conditions permit. The charter member countries should have the following features: Rice is an important crop and planted on a large area so that an economically viable local hybrid seed industry is possible. Hybrid rice research, development, and use have been adopted as national priorities. Staff and facilities are adequate to make effective and positive contributions to the Task Force activities, especially in terms of exchange of germplasm, information, and data. Bangladesh, India, Indonesia, Philippines, and Vietnam agreed to participate as

IRRN 21:2-3 (August-December 1996)


charter member countries. China and Sri Lanka agreed in principle to serve as charter member countries but first need to get approval from their respective governments. Brazil, Colombia, Egypt, and Thailand agreed to participate as observers. Australia, Japan, and the USA agreed to participate as affiliates. Implementation. IRRI, possessing a strong multidisciplinary hybrid rice research program linked with several NARSs, will be the executing agency that coordinates Task Force activities especially for technical assistance in the matters of developing hybrid rice technology and establishing necessary

collaboration with selected institutions that have advance programs concerning strategic research for hybrid rice development. FAO, having expertise in agricultural development, will provide guidance and technical backstopping for seed industry development and transfer of hybrid rice technology in member countries. FAO will also facilitate institutional linkage among agencies dealing with hybrid rice research and seed production within and among member countries. Member countries will designate respective national coordinators to work collaboratively with IRRI, FAO, and other

member countries to implement the work plans as prepared and adopted in annual meetings of the Task Force. Management. The Task Force should be a component of the Council for Collaborative Research in Asia (CORRA) and should be managed on the pattern of the rainfed and upland rice consortia already in operation at IRRI in collaboration with several NARSs.


Replace leaf CHO with stem CHO in Figure 1 in Leaf carbohydrate analysisa simple method for integrating daily canopy photosynthesis, IRRN 21:1, p. 52.


IRRN 21:2-3 (August-December 1996)




Stem borers of rice (RP4-18)

1988 edition. 80 colored slides, one cassette tape, and two booklets (self-learning and self-testing)10.50x21.30 cm. Paperback. HDC US$254.00, LDC US$67.00 plus airmail (US$17.00) postage.

Morphology of the rice plant (RP4-01)

1987 edition. 79 colored slides, one cassette tape, and two booklets (self-learning and self-testing) 10.50x21.30cm. Paperback. HDC US$254.00, LDC US$67.00 plus airmail (US$17.00) postage.

Growth stages of the rice plant (RP4-02)

1988 edition. 80 colored slides, one cassette tape, and two booklets (self-learning and self-testing)10.50x21.30cm. Paperback. HDC US$254.00, LDC US$67.00 plus airmail (US$17.00) postage.

Stem borers of rice presents an overview of the life cycle of rice stem borers, the damage they cause, and control measures. The lesson lasts for about 35 min. It will enable the user to describe and identify damage to rice caused by stem borers; identify and describe the five destructive stem borers of rice, and relate stem borer species with geographic distribution; identify and describe the four stages in the life cycle of the major stem borers including their appearance, habitat, behavior, mode of feeding, and duration of each life cycle; and describe the most effective methods of controlling the five major stem borers of rice. Stem borers can damage rice plants from seeding to maturity. A knowledge of their life cycle and damage symptoms is important to implement effective control measures.

Morphology of the rice plant introduces the terms associated with the different parts of the rice plant. The lesson lasts for about 20 min. It will enable the user to identify and describe the morphological characters or parts of the following: the germinating rice seed and seedling; a rice tiller roots, culm, and leaves; the rice inflorescence or panicle. Understanding of standard terms describing rice morphology is important to effectively communicate with other scientists about the rice plant.

Growth stages of the rice plant relates and describes the various stages of growth of the rice plant from seed to maturity. The lesson lasts for about 26 min. It will enable the user to recognize the three basic growth phases of the rice plant and the stages of development in each phase; identify the growth stages of a rice plant according to a 0-9 numerical scale (each number in the scale corresponds to a specific growth stage); and explain the specific physical changes in a growing rice plant. It is essential to know the stages of the growth cycle of rice so that appropriate management practices can be employed at the right time. The effectiveness of management practices, such as fertilization and irrigation, and weed, pathogen, and insect control, largely depends on correct timing.