VOLUME 23 NUMBER 2 1998

InternationalRice Research

Notes

INTERNATIONAL RICE RESEARCH INSTITUTE

2 IRRN 1998

International Rice Research Notes

The International Rice Research Notes (IRRN) expeditescommunication among scientists concerned with thedevelopment of improved technology for rice and rice-based systems.

The IRRN is a mechanism to help scientists keep eachother informed of current rice research findings. Theconcise scientific notes are meant to encourage ricescientists to communicate with one another to obtaindetails on the research reported.

The IRRN is published three times a year in April,August, and December by the International RiceResearch Institute.

IRRN production team

Editor: Bill HardyAssistant editor: Tess RolaLayout and design: Erlie PutunganArtwork: Erlie Putungan

○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○

Focus on hybrid riceAchieving self-sufficiency in rice production and maintainingprice stability are important in countries where rice providesfood security and generates employment and income forpeople. In the past three decades, most rice-growingcountries, particularly in Asia, have done remarkably well inmeeting their rice needs. But by 2030, the world must produce60% more rice than it produced in 1995 to meet the demandcreated by increasing population and rising income. Thisincrease must be achieved using less land, less labor, lesswater, and fewer pesticides, and it must be sustainable. Tomeet this challenge, increasing the yield potential of ricebeyond that of the semidwarf varieties is an importantstrategy.

Hybrid rice is a technology for meeting this challenge. Thistechnology has enabled China to increase its rice productionsignificantly during the past 20 years. IRRI, in collaborationwith several national agricultural research systems, hasdeveloped rice hybrids for the tropics and helped India,Vietnam, and the Philippines to begin commercializing them.By the 21st century, about 3 million hectares are expected to becovered with hybrid rice, which should produce about 3million tons of extra rice (worth $450 million) annually.

The Third International Symposium on Hybrid RiceResearch was held 14-16 November 1996, at the Directorate ofRice Research in Hyderabad, India. The key papers presentedare being published as an IRRI-Indian Council of AgriculturalResearch book. The posters displayed at the symposiumappear as notes (in a modified format) throughout this issueof the IRRN. They are denoted by the symbol.

We hope that you find these notes to be a valuable sourceof information.

This issue is dedicated to Dr. DharmawansaSenadhira.

It is with the deepest regret that IRRI announces the tragicand untimely death of Dr. Dharmawansa Senadhira, plantbreeder and leader of the Institute's Flood-prone RiceEcosystem Program. Dr. Senadhira was killed in a highwayaccident on 7 Jul in Bangladesh while returning to Dhakafrom a field trip associated with a deepwater fish-riceproduction project.

Dr. Senadhira, a citizen of Sri Lanka, had worked at IRRIfor 13 yr. With his colleagues, he had successfullydeveloped and refined screening techniques indetermining the inheritance of tolerance for soil-relatedstresses, low temperature, and submergence in rice.Working in close collaboration with national agriculturalresearch systems (NARS) in several countries, he had alsodeveloped improved germplasm for tidal wetlands anddeeply flooded ricelands, and for irrigated ricelandsaffected by low temperatures and salinity. Dr. Senadhiraalso had administrative responsibility for IRRI's office inBangkok, Thailand.

Before joining IRRI in 1985, Dr. Senadhira had beendeputy director of agriculture (research) and senior ricebreeder in the Sri Lankan Department of Agriculture. Whilein that position, he had won his country's President'sAward for Scientific Achievement and the CERES Medalfrom the United Nations Food and AgricultureOrganization. Just prior to his death, he had been informedthat he had been awarded the prestigious FukuiInternational Koshihikari Rice Prize for 1998 in recognitionof his outstanding achievements in rice culturedevelopment.

Dr. Senadhira will be sorely missed by his colleagues atIRRI and his collaborators in the NARS, not only for hisexcellent scientific achievements and leadership but alsofor his open, warm, ever-friendly personality, hisirrepressible sense of humor, and his willingness to take onnew tasks and responsibilities whenever asked. TheInstitute especially appreciated his dedication as a reviewerto improving the quality of contributions to the IRRN. Thenews of his death has already brought many messages ofsympathy and condolences from friends around the world.Dr. Senadhira, who was unmarried, leaves behind a familyin Ranala, Sri Lanka.

Vol. 23, No. 2 3

ContentsVol. 23, No. 2, 1998

Germplasm improvement

Genetic resourcesA new CMS line with a wide compatibility gene 5New cytoplasmic male sterile (CMS) lines with diversified CMS

sources and better outcrossing traits in rice 5Diversifying cytoplasmic sources in rice 5An indica source of male sterile cytoplasm and its fertility

restoration for hybrid rice breeding 6Male sterile lines for hybrid rice breeding 6Fertility-altering conditions of promising thermosensitive genic

male sterile lines in rice 6Identifying rice genotypes with the TGMS trait suitable for

north India 7

GeneticsZhenong 921: a new indica rice variety with high yield and

blast resistance 7Cytogenetics of alien chromosome addition lines in rice (Oryza

sativa L.) with single chromosomes of O. punctataKotschy 8

Genetic, histological, and histochemical evidence for rever-sion to partial fertility in WA CMS line IR54752A 9

Genetics of the thermosensitive genic male sterility trait inrice 9

Biochemical markers for characterizing rice genotypes 10Identifying molecular markers for the gene(s) governing

thermosensitive genic male sterility in rice 10

Breeding methodsHeterosis over environments in crosses involving indica and

tropical japonica rice cultivars 11Plant regeneration from protoplasts of wild rice, Oryza

meyeriana Baill. 11Chinoor: a promising spontaneous mutant and high-quality

rice variety of Madhya Pradesh, India 12Combination selection method for rice breeding efficiency 13Yield stability analysis of rice hybrids 14Yield system analysis in rice hybrids 14Studying heterosis for grain yield and its components in hybrid rice 15Standard heterosis of rice hybrids for yield and yield components 15Developing Pusa 5A, a stable indica CMS line with high

outcrossing potential 15Field evaluation of thermosensitive genic male sterile

lines 16Developing thermosensitive genic male sterile lines in

rice 16

Male sterility and fertility behavior of suspected thermosensitivegenic male sterile (TGMS) lines 16

Hybrid rice research in Pakistan 17Hybrid rice: status and future in Bangladesh 17

Grain qualityDonors for quality traits from the international aromatic

nursery 18Effect of bran removal on oil recovery 19Jaymati, a high-yielding summer rice variety with good grain quality

for Assam 20

Integrated germplasm improvementWide adaptability of new rice cultivars developed by the Japan-

China collaborative project in Yunnan, China 21

Integrated germplasm improvement — irrigatedKarnataka rice hybrids 22TNRH16: a salt-tolerant rice hybrid 22

Pest resistance — diseasesDifferentiation of rice tungro spherical virus variants by RT-PCR

and RFLP 22Mechanism of resistance to rice tungro spherical virus

(RTSV) 24

Seed technologyLeaf number: a reliable parameter for determining seeding

intervals between parental lines in hybrid rice seedproduction 25

Adapting hybrid rice seed production technology 26Analyzing leaf number in parental lines for hybrid rice seed

production 27Identifying some favorable environments for hybrid rice seed

production in Andhra Pradesh, India 27Seasonal influence of flowering behavior and plant growth charac-

ters on parental lines of hybrid rice 27Determining seeding intervals of parents in hybrid rice seed

production 28Adjusting flowering of a CMS line in hybrid rice seed

production 28

Stress tolerance — adverse soilsEffective amount of N fertilizer for direct seeding on wet surface of

reclaimed saline soil in Korea 29Identifying optimum seeding time for direct seeding on a wet field

surface in reclaimed saline soil in Korea 29

Stress tolerance — excess waterDistinguishing seedling traits in deepwater rice

(Oryza sativa L.) 30Tolerance for submergence in rainfed lowland rice under repetition

of flooding 31

4 IRRN 1998

Yield potentialPhysiological basis of heterosis in rice 32Physiological efficiency of rice hybrids 32Performance of hybrid rice in south Karnataka 33

Crop and resource management

Physiology and plant nutritionEffect of salinity on growth, chlorophyll content, and flag leaf

area of rice (Oryza sativa L.) genotypes 33

Crop managementA labor-saving technique in direct-sown and transplanted

rice 35Optimum seedlings per unit area for high-yielding rice

varieties in the hill zone of Karnataka 36Drill sowing of pregerminated rice seed: effect of rainfall on

plant stand 38Effect of planting geometry and N levels on grain yield of

hybrid cultures 38

Farm machineryAn improved suction apparatus for sampling invertebrate

communities in flooded rice 38A new Senegalese thresher/cleaner responds to small-farmer

postharvest needs 39Technology transfer from Asia to Africa sets the stage for

public- and private-sector collaboration in new technologyin Senegal 41

Polyurethane rollers: a substitute for rubber rollers in ricedehuskers 42

The International Rice Research Institute (IRRI) was established in 1960 by the Ford and RockefellerFoundations with the help and approval of the Government of the Philippines. Today IRRI is one of 16nonprofit international research centers supported by the Consultative Group on International AgriculturalResearch (CGIAR). The CGIAR is cosponsored by the Food and Agriculture Organization of the UnitedNations (FAO), the International Bank for Reconstruction and Development (World Bank), the UnitedNations Development Programme (UNDP), and the United Nations Environment Programme (UNEP). Itsmembership comprises donor countries, international and regional organizations, and private foundations.

As listed in its most recent Corporate Report, IRRI receives support, through the CGIAR, from a numberof donors including UNDP, World Bank, European Union, Asian Development Bank, and RockefellerFoundation, and the international aid agencies of the following governments: Australia, Belgium, Canada,People’s Republic of China, Denmark, France, Germany, India, Indonesia, Islamic Republic of Iran, Japan,Republic of Korea, The Netherlands, Norway, Philippines, Spain, Sweden, Switzerland, United Kingdom,and United States.

Fertilizer managementResponse of rice hybrids to N sources and time of application of N

and K 43Nitrogen response of Karnataka Rice Hybrid 2 43

Integrated pest management — diseasesThe effect of Trichoderma and antifungal agents on rice

germination 43

Integrated pest management — insectsEvolving insecticide use and practices at IRRI 44

Integrated pest management — weedsButachlor safener combinations for weed control in direct-seeded

puddled rice 46

Soil microbiologyEffect of immobilization of N2-fixing cyanobacteria on solid

matrices and their influence on N2-fixing activity and ammoniaexcretion 47

Socioeconomic impact

Economics of hybrid rice seed production in seed growers’ fields inKarnataka, India 48

Economics of commercial hybrid rice cultivation 48

Announcement

Scholarships available 48

Vol. 23, No. 2 5

Germplasm improvementGenetic resources

A new CMS line with awide compatibility gene

P. Jayamani, K.N. Ganesan, K. Thiyagarajan,M. Rangaswamy, and P. Rangasamy, School ofGenetics, Tamil Nadu Agricultural University,Coimbatore 641003, India

In the three-line system (A/B/R) ofheterosis breeding in rice, geneticdiversity of the A line is very important.The F1 hybrids of indica/japonicacrosses show semisterility. Some widecompatibility varieties do producefertile F1 hybrids when crossed withindica as well as with japonicavarieties. A cytoplasmic male sterile(CMS) line with a wide compatibilitygene will be useful in developingindica/japonica hybrids that canovercome sterility problems.

Variety Dular, with a widecompatibility gene, was crossed withdifferent CMS lines during the 1990 dryseason. The F1s of V20A/Dular showed100% pollen sterility and substitutionbackcrosses were then made. In eachbackcross generation, 36 plants wereraised along with the parental lines.The completely male sterile plants wereidentified based on pollen sterility andused for subsequent backcrossing up tothe BC6 generation. Finally, the genomeof Dular was transferred into a V20Acytoplasmic background.

In the BC7 generation, a completelymale sterile population was built up.This new male sterile line (COMS8A)has many advantages over the existingmale sterile lines, although it has anopen plant type. COMS8A has the widecompatibility allele, which can be usedto exploit indica/japonica heterosis.The line also possesses a highpercentage of panicle exsertion, whichminimizes the requirement of applyingGA3 in seed production. ■

New cytoplasmic malesterile (CMS) lines with

diversified CMS sources andbetter outcrossing traits in rice

M.S. Ramesha, M.I. Ahmed, B.C. Viraktamath,C.H.M. Vijayakumar, and S. Singh, Directorate ofRice Research, Rajendranagar, Hyderabad500030, India

About 95% of the commercial hybridsin China and elsewhere are based on awild abortive (WA) cytosterilitysystem. This excessive reliance on asingle source of sterility may givehybrids genetic vulnerability to asudden outbreak of diseases and insectpests. Another drawback of the WAsystem is poor panicle exsertion andundesirable flowering behavior, whichlead to low seed yield in seedproduction plots.

To overcome these disadvantages,three new and diversified CMS sourceshave been identified and many CMSlines possessing these sources havebeen developed. One CMS line fromOryza nivara possessed a sporophytictype of male sterility with a very highfrequency of typically abortive pollengrains. The other two sources, from O.rufipogon and O. nivara, possessed agametophytic type of male sterility asevidenced by a very high frequency ofround, sterile-type pollen grains.

The new CMS lines were comparedwith other CMS lines belonging to WA,O. perennis, and MS577A CMS sourcesduring 1994-96 for pollen sterilitycharacteristics and outcrossing traits.Besides stable sterility, the new CMSlines were found to have very highpanicle exsertion (92-96%), goodstigma exsertion (48-65%), and highoutcrossing ability (38-52%) comparedwith other CMS sources. The search forrestorers for the new CMS sources is inprogress. The use of good hybridcombinations involving new CMS linescan pave the way for maximizing

hybrid seed yield without the use ofGA3, a costly input in hybrid rice seedproduction. A reduction in the cost ofhybrid seed will help in large-scaleadoption. ■

Diversifying cytoplasmicsources in rice

M. Rangaswamy and P. Jayamani, School ofGenetics, Tamil Nadu Agricultural University,Coimbatore 641003, India

Most commercial hybrids of indica riceare based on a wild abortive (WA)source of cytoplasmic male sterility(CMS). This cytoplasmic uniformitycould lead to genetic vulnerability todisease and insect pests. There is thusan urgent need for cytoplasmicdiversification of the male sterilitysource for hybrid rice breeding.

Direct and reciprocal crosses weremade between AA genome speciesOryza nivara (Accessions 105879,101508, 102464, 105343, 101871, and106046), O. spontanea (106137), O.rufipogon (105616), O. barthii (100934),O. glaberrima (100139), and O. sativacultivars Co 43, IR50, Co 45, ASD16,White Ponni, and IR64. These cultivarsare restorers/weak restorers of WAcytoplasm.

The F1 was evaluated and based onpollen or spikelet fertility. Progenieswith >99% pollen or spikelet sterility—O. nivara (105343)/Co 45, O. barthii(100934)/ASD16, O. barthii (100934)/IR50, and O. nivara (101508)/IR64—were identified by their reciprocaldifference. These sterile hybrids werebackcrossed with the respectiverecurrent parents. Pollen sterility wasdetermined in the BC1, BC2, BC3, andBC4 generations and these lines werefound to be stable for complete malesterility. Subsequent backcrosses are inprogress to transfer the character of therecurrent parents. ■

6 IRRN 1998

An indica source of malesterile cytoplasm and its

fertility restoration for hybridrice breeding

U.S. Natarajan, F.U. Zaman, M.J. Abraham,A. Mahendru, and E.A. Siddiq, Division ofGenetics, Indian Agricultural Research Institute,New Delhi 110012, India

The search for alternate sources ofsterile cytoplasm in hybrid ricebreeding is a priority because morethan 90% of the rice hybrids releasedthroughout the world are based on asingle sterile cytoplasm—wild abortive(WA). Because of the very narrowgenetic base provided by the singlecytoplasmic source, hybrids carryingthis cytoplasm can become vulnerableto pests and diseases found associatedwith it.

We observed reciprocal differencesfor pollen sterility in the F2 generationin a number of crosses. In one suchcross (Pusa 743/Pusa 33), male sterilesegregants were observed during 1992,whereas in another reciprocal (Pusa33/Pusa 743) all the F2 plants werefertile. This indicated that Pusa 743 hassterility-inducing cytoplasm. Ofseveral breeding lines tested for fertilityrestoration in this cytoplasm, 30% werefound to be restorers. Initially, 21% ofthe lines were identified as maintainers.But only 10% showed a stable reactionfor sterility maintenance in the laterbackcross generations. This indicatedthat fertility restoration was governedby a single dominant gene, although amajority of the crosses showedsignificant deviations from Mendeliansegregation because of a deficit ofrecessives. Japonica rices are likely tomaintain sterility in this cytoplasm.Three cytoplasmic male sterile lineswith a stable sterility reaction weredeveloped. Pusa 1127A was found to beagronomically promising. ■

Male sterile lines for hybridrice breeding

C.P. Rao, Regional Sugarcane and Rice ResearchStation, Rudrur, Nizamabad 503188, India; andL.P. Yuan, China National Hybrid Rice ResearchCenter, Mapoling, Changsha 410125, HunanProvince, China

Experience with hybrid maize indicatesthe utility of diverse cytoplasms as asource of cytoplasmic male sterility.Crop heterogeneity is desirable for thesustainability of commercial hybrids.We studied the morphological andphysiological characters of male sterilelines belonging to diverse cytoplasm.We studied eight male sterile linesbelonging to wild abortive types (V20A, Zhengshan 97 A, BO A, Jin 23 A, andZhi A), IDR or Indonesian paddy ricetype (U1 A), dwarf abortive type(Xieqinzao A), and BT or japonica type(80-4 A) during the winter season of1993-94 in a randomized block designwith three replications. Each line wasplanted in a single row, with a spacingof 20 × 20 cm. The crop was maintainedas per recommended practices.

From a nursery 10- , 20-, and 30-d-old seedling samples were taken forstudy. Clean plant samples werecollected from the nursery using adigging plate and the roots were freedof soil by washing them with water.From five seedlings per replication,observations were recorded on length,volume by water displacement in aburette, and weight of root and shootportions. Total tillers, panicle-bearingtillers, and plant height at the vegeta-tive (60 d) and flowering stages werealso recorded for each test line. Pollenreaction was studied using an I-KI (1%)solution. The percentage of a certaintype of pollen was calculated by theratio of its occurrence to the totalnumber of pollen grains in a micro-scopic field. Seed set on male sterile

Review of notes. The IRRN editor will send an acknowledgment card or an e-mail message when a note is received. An IRRIscientist, selected by the editor, reviews each note. Reviewer names are not disclosed. Depending on the reviewer’s report, a notewill be accepted for publication, rejected, or returned to the author(s) for revision.

Comments. If you have comments or suggestions about the IRRN, please write to the editor.

lines was studied in isolation bybagging.

Some lines that expressed goodseedling growth and vigor in thenursery were Zhi A (root length, shootvolume, and shoot weight at 30 d),Xieqinzao A (root and shoot length at30 d, root and shoot volume at 30 d),and Jin 23 A (shoot length at threestages of sampling and shoot weight at20 and 30 d).

Jin 23 A, Zhi A, and 80-4A were tall atthe vegetative and flowering stages.The lines flowered in 85-95 d, with Jin23 A being early and Zhengshan 97 Alate to flower. The exsertion of stigmasoutside the floret was well expressed inBO A, Jin 23 A, U1 A, and Zhi A. LinesJin 23 A, V20 A, and Zhengshan 97 Ashowed 100% unstained and irregular-shaped pollen, whereas Zhi A, BO A,and U1 A showed 99-99.5% sterileirregular pollen. Xieqinzao A had 5%stained pollen and 80-4 A had all pollengrains circular, of which 50% werestained. Line BO A had a maximumpanicle-bearing tillers and was on a parwith Jin 23 A, Zhi A, V 20 A, andXieqinzao A. There was no seed set inthe panicle bag. ■

Fertility-altering conditionsof promising thermo-

sensitive genic male sterile linesin rice

O.U.K. Reddy, Directorate of Rice Research(DRR), Rajendranagar, Hyderabad, 500030;E.A. Siddiq, Indian Council of AgriculturalResearch, New Delhi 110001; J. Ali, CropImprovement Department, Agriculture Collegeand Research Institute; Navalur Kuttapattu,Trichy 620009; A.J. Hussain and M.I. Ahmed,DRR, Rajendranagar, Hyderabad 500030, India

Characterizing thermosensitive genicmale sterile (TGMS) lines with respect

Vol. 23, No. 2 7

Identifying rice genotypeswith the TGMS trait suitable

for north India

N. Saxena and J.P. Singh, Genetics and PlantBreeding Department, G.B. Pant University ofAgriculture and Technology, Pantnagar, India

The discovery of the environment-sensitive genic male sterility (EGMS)system laid the foundation for repla-cing the three-line system with thesimpler and more efficient two-linesystem for hybrid rice seed production.EGMS includes photoperiod-sensitivegenic male sterility (PGMS) andthermosensitive genic male sterility(TGMS) systems. The required day-length differences do not occur in ourrice seasons for PGMS seed production.Therefore, we attempted to developsuitable TGMS lines in India.

We studied fertility responses totemperature in 15 TGMS lines during1995-96 under natural environmentalconditions at Pantnagar, situated at29° N latitude, 79.3° E longitude, and244 m above sea level. The aim was toidentify suitable TGMS lines for thenorthwestern regions of Uttar Pradesh.The TGMS lines were sown on variousplanting dates and maintained in thenethouse or glasshouse.

Five TGMS lines showed seasonalfertility-sterility transformation underfield conditions. These lines werecharacterized by sterility under hightemperature (mean daily average>29 °C) and fertility transformation atlow temperature (mean daily average26-28 °C). IR68292-10-34-6, IR68292-10-34-25, and IR68949-5-31-34 werecompletely sterile when sown inFebruary to April but partially fertilewhen planted after the second fortnightof May. IR68945-33-4-14-4 and IR68945-33-4-14-7 showed complete sterilitywhen sown in February to May andwere partially fertile when sown inJune and July. IR68292-10-34-6,IR68292-10-34-25, and IR68949-5-31-34may be exploited for hybrid seedproduction with off-season sowing inFebruary to April. Seeds of these linescan be multiplied by sowing in July toAugust, whereas TGMS lines IR68945-33-4-14-4 and IR68945-33-4-14-7 can beused to produce hybrid seed in themain season, and parental lines can bemultiplied by sowing in July andAugust. To exploit these lines indeveloping commercial hybrids,further studies are in progress to testtheir stability at several locations in theregion and also under controlledtemperature conditions. ■

Genetics

to fertility alteration and evaluatingtheir utility in two-line heterosisbreeding were the objectives of thepresent investigation.

TGMS lines SA2, F61, JP8-8-1s, JP1,IC10, ID24, and JP24A identified atDRR, Hyderabad, and IR32364,IR68292, IR68294, IR68945, andIR68949, received from IRRI,Philippines, were screened. TheseTGMS lines were grown underconstant and varying temperature anddaylength combination regimes in agrowth chamber. They were also grownin fields under staggered sowing andratooning over seasons. Lines SA2, JP8-8-1s, F61, IC10, and ID24 were found tobe interactive with photoperiod;therefore, they are designated asPTGMS. JP1 and the TGMS lines fromIRRI were more or less independent ofphotoperiod influence. The criticalfertility point, the temperature at whichmaximum fertility occurred, rangedfrom 20 to 26 °C in various PTGMSlines and from 20 to 30 °C for otherTGMS lines. The critical sterility point,the temperature at which total sterilityoccurred, was at 25 °C and above forPTGMS lines and at 32 °C for TGMSlines. Reverse TGMS line JP24Ashowed total sterility below 26 °C andcomplete fertility at 28 °C and above.

The temperature at which physio-logical sterility occurred (biologicallower limit) and the temperature atwhich sensitivity for various lines wasidentified differed. PTGMS lines werestable for sterility under fieldconditions (March-June), whereasTGMS lines reverted back to fertilityunder cloudy weather and cooler nighttemperatures. Stable sterility in PTGMSlines may be because of the dual role oftemperature and photoperiod or theirinteraction. Spikelet fertility under fieldconditions ranged from 7 to 21% and11-70% seed set was recorded in agrowth chamber for PTGMS lines. Thiswarrants a search to identify criticalareas for their seed multiplication.■

Zhenong 921: a new indica ricevariety with high yield and blastresistance

Shi Chunchai, Department of Agronomy,Zhejiang Agricultural University (ZAU),Hangzhou 310029, China; Chen Wenguang, SeedCorporation of Shaoxin County, Zhejiang 312000,China; Chen Guolin, ZAU; Zhang Genxian,Agriculture Popularization Station of KaihuaCounty, Zhejiang 324300, China; Shen Zongtan,ZAU; and Wu Jianguo, ZAU

Zhenong 921, a new semidwarf indicarice variety, is derived from Zhongzhe1/K 125-34 and is suitable for thedouble-cropped area of southernChina. It was registered by ZhejiangProvince in 1997 as an early rice variety.

About 2,000 ha in southern China areplanted to Zhenong 921, which has agrowth duration of about 109 d.Zhenong 921 has a high, stable yieldpotential under normal fertilization. Ityielded 6.7-7.5 t ha-1 in yield trials andits highest yield of 8.1 t ha-1 was 9.3%more than that of the check in 1994. Itsleaf blast and neck blast resistancescores were 1.6 and 1.9, compared withcheck Zhe 852, which scored 4.6 and3.4, respectively. Zhenong 921 has coldresistance in the seedling stage andgood lodging resistance.

Tables 1 and 2 show morpho-agronomic characters. Zhenong 921 isawnless and medium-grained. Themilled rice length and the ratio oflength to width are 6.5 mm and 2.5,

8 IRRN 1998

Table 1. Plant traits of Zhenong 921.

Plants Panicles ProductiveYear Duration tillers Plant height

(d) (no. m-2) (%) (cm)

1994 104.9 195.0 406.5 77.4 68.91995 112.8 156.0 339.0 76.6 73.81996 109.3 162.0 411.0 73.5 80.7 Mean 109.0 171.0 385.5 75.8 74.5

Table 2. Agronomic traits of Zhenong 921.

Panicle Grains Fertilized Fertility 1,000-grainYear length (no. panicle-1) grains (%) weight

(cm) (no. panicle-1) (g)

1994 17.4 85.9 64.3 74.9 27.61995 17.6 98.3 70.6 71.8 27.61996 18.3 88.3 75.0 84.9 27.2 Mean 17.8 90.8 70.0 77.2 27.5

Cytogenetics of alien chromo-some addition lines in rice(Oryza sativa L.) with singlechromosomes of O. punctataKotschy

H. Yasui and N. Iwata, Plant Breeding Labora-tory, Faculty of Agriculture, Kyushu University,Fukuoka 812-81, Japan

Monosomic alien addition lines(MAALs) of O. sativa carrying singlechromosomes of O. punctata weredeveloped (Yasui and Iwata 1991). Inthe progenies of these MAALs, weidentified plants in which the extrachromosome was telocentric oracrocentric.

Three monotelosomic alien additionlines (MtAALs) and one monoacro-somic alien addition line (MaAAL),each carrying single chromosomes ofO. punctata, were isolated from theprogenies of the respective MAALs forchromosomes 2, 4, 7, and 9. Mitotic andmeiotic chromosomes of the MtAALsand an MaAAL were analyzed (seefigure). A telosome as an extrachromosome was observed in the

respective MtAALs and one of themwas for a short arm of chromosome 9.An acrosome that consisted of acomplete short arm and a heterochro-matic proximal region of a long arm forchromosome 4 was observed in theMaAAL. Three MtAALs were

respectively. It has 83.9% brown ricerecovery, 75.5% milled rice recovery,43.9% head milled rice recovery, 26.3%amylose content, 50 mm gel

consistency, an alkali spreading valueof 3, and acceptable cooking or eatingquality. ■

designated as MtAALs 2, 7, and 9S,respectively, and one MaAAL wasdesignated as MaAAL 4S4L.

Morphology, seed fertility, andtransmission of the extra chromosomeof the MtAALs and the MaAAL werecompared with the respective primarytrisomics and MAALs. The plantmorphology of MtAAL 2 and MaAAL4S4L was similar to that of the respec-tive MAALs, but that of MtAAL 9S wassimilar to disomics and that of MtAAL7 was similar to secondary trisomics forthe short arm of chromosome 7. Seedfertility of the MAAL, which rangedfrom 6.6% for MtAAL 2 to 94.5% forMtAAL 7, was higher than the respec-tive primary trisomics and MAALs.The transmission rates of the extrachromosome, which ranged from22.0% for MtAAL 2 to 28.2% forMaAAL 4S4L, were similar to those ofthe respective MAALs.

The mode of meiotic chromosomebehavior of the MtAALs and theMaAAL was compared with that of therespective primary trisomics andMAALs (see table). The ratios of thepollen mother cells (PMCs) with atrivalent were clearly different betweenprimary trisomics (32-67%) and

Comparison of meiotic chromosome behavior of primary trisomics and MAALs each with a single chromosomeof O. punctata.

Number of cellsTrisomics/ Cells withMAAL 12II+1I trivalent

11II+3Ia 11II+1III Total (%)

Triplo 2 45 21 66 31.8MAAL 2 158 5 163 3.1MtAAL 2 26 0 26 0

Triplo 4 31 62 93 66.7MAAL 4 171 0 171 0MaAAL 4S4Lb 50 0 50 0

Triplo 7 47 83 130 63.9MAAL 7 54 0 54 0MtAAL 7 155 1 156 0.6

Triplo 9 25 51 76 67.1MAAL 9 60 0 60 0MtAAL 9Sc 41 0 41 0

aTwo PMCs of MtAAL 2 and three PMCs of MtAAL 7 showed 11II+3I. bMonoacrosomic alien addition line carrying the completeshort arm and the proximal long arm of chromosome 4 derived from O. punctata. cMonotelosomic alien addition line carryingthe short arm of chromosome 9 derived from O. punctata.

Vol. 23, No. 2 9

Genetic, histological, andhistochemical evidence for

reversion to partial fertility inWA CMS line IR54752A

S.L. Kumari, Regional Agricultural ResearchStation, Pattambi; and M. Mahadevappa,University of Agricultural Sciences, Dharwad,India

The widely used wild abortive (WA)cytoplasmic male sterile (CMS) lineIR54752A for the hybrid rice programat IRRI and in India was reported to beunstable for male sterility charactersand therefore restricted to use in seedproduction. We investigated the causesfor the breakdown in sterility. All evi-dence gathered suggested the presenceof multiple nuclear genes (minor) forfertility in the CMS line. Evidence in-cluded the presence of partially fertileplants in a cross with the isonuclearmaintainer line, genetic studies usingselfed and test-crossed progenies of thepartially fertile plants, and the compa-rative ease of restoration of fertility ofthis CMS line in crosses with restorers

Genetics of the thermo-sensitive genic male

sterility trait in rice

Li Rongbai and M.P. Pandey, Genetics and PlantBreeding Department, G.B. Pant University ofAgriculture and Technology, Pantnagar 263145,India

UPRI 95-140 (P1), the thermosensitivegenic male sterile (TGMS) line of low-temperature fertility transformationtype, was analyzed for its inheritanceand transformation behavior. Parents,F1, F2, and six test crosses betweenTGMS (female parent) and UPRI 95-141(P2), UPRI 95-165 (P3), UPRI 95-124(P4), UPRI 95-117 (P5), and IR36 (P6)

Chromosome constitution of three monotelosomic and one monoacrosomic alien addition linesof rice each carrying a telocentric or acrocentric single chromosome of O. punctata. Mitoticchromosomes of MtAAL 2 (a), MaAAL 4S4L (b), and MtAAL 7(c), and meiotic chromosomes ofMtAAL 9S (d and e). Each arrow indicates the telocentric and/or acrocentric chromosomes de-rived from O. punctata.

MAALs or Mt(a)AALs (approximately0%). The telocentric and acrocentricchromosomes would originate from amisdivision of an alien chromosomeand the following chromosome break-age in the PMCs of the respectiveMAALs at the first or second anaphase.The addition lines with small chromo-somal fragments such as alien telo-centric and acrocentric chromosomeswill serve as a source to transfer usefulgenes from wild relatives to cultivatedrice.

ReferenceYasui H, Iwata N. 1991. Production of

monosomic alien addition lines of Oryzasativa having a single O. punctatachromosome. In: Rice genetics II.Proceedings of the Second InternationalRice Genetics Symposium, 14-18 May1990. Manila (Philippines): InternationalRice Research Institute. p 147-155. ■

as well as maintainer lines of otherCMS lines belonging to the same cyto-plasmic source. Histological studiesrevealed normal behavior and func-tioning of anther wall tissues, includingtapetum and endothecium, as well asnormal development of microspores inCMS line IR54752A up to the liberationof microspores. The CMS pollen grainsdiffered from other fertile pollen inhistochemical components such asstarch, proteins, and RNA at maturity,indicating that mobilization ofmetabolites to the pollen grains washindered to some extent at maturity.This showed that factors within themicrospore were responsible for thebreakdown in sterility in IR54752A. ■

Routine research. Reports of screening trials of varieties, fertilizer, croppingmethods, and other routine observations using standard methodologies toestablish local recommendations are not ordinarily accepted. Examples aresingle-season, single-trial field experiments. Field trials should be repeatedacross more than one season, in multiple seasons, or in more than one locationas appropriate. All experiments should include replications and an internationallyknown check or control treatment.

10 IRRN 1998

were screened in the field during themain season.

In F1 progenies from all crosses,dominance of fertility over completesterility was observed. F2 populationsfrom P1/P5 and P1/P6 showed segre-gation for fertility and complete sterili-ty in a ratio of 15:1. This indicated thatthermosensitive male sterility wascontrolled by two pairs of recessivegenes with duplicate inheritance. Theremaining three crosses displayedmonogenic control of this trait. The testcross progenies from P1//P1/P5showed 3:1 segregation (χ2=0.0126).The results suggested two pairs ofindependent recessive genes

Biochemical markers forcharacterizing rice

genotypes

V. Santhy, V. Niral, and M. Dadlani, Division ofSeed Science and Technology, Indian AgriculturalResearch Institute, New Delhi 110012, India

Eight frequently used genotypes in thehybrid rice program were charac-terized on the basis of (1) phenol colorreaction; (2) electrophoresis profiles oftotal soluble seed proteins, albumin,and globulin; and (3) polymorphismwith respect to esterase (EST), malatedehydrogenase (MDH), peroxidase(POX), alcohol dehydrogenase (ADH)and glutamate dehydrogenase (GDH)isozymes. These genotypes wereIR58025A, IR58025B, IR62829A,IRR62929B, IR54742-22-19-3R, IR40750-

controlling the TGMS trait in UPRI 95-140. Monogenic inheritance observedin the F2 populations of crosses P1/P2,P1/P3, and P1/ P4 was due to theallelic nature of one of the two pairs ofrecessive genes present in both theparents involved.

Fertility transformation of TGMSplants in segregating populations ofcrosses depended on the genetic back-ground of the male parents used. In theF2 generation, average maximum ferti-lity of TGMS plants during spring 1996was 37.7, 45.3, 60.5, and 65.4% incrosses P1/P5, P1/P6, P1/P2, and P1/P4, respectively, as compared with41.5% in UPRI 95-140, the female and

TGMS donor parent. Differentialbehavior of TGMS plants in the F2 fortransformation from fertility to com-plete sterility was observed during theheading period: 25 Apr to 20 May forcross P1/P5, 1-10 May for P1/P6, 1-25May for P1/P2, 1-10 May for P1/P4,and 1 May for the TGMS female parent.Eight lines identified with early trans-formation behavior were 97-7S fromP1/P2; 34-2S and 69-2S from P1/P5;206-8S, 206-9S, and 206-10S from P1/P4; and 79-2S and 79-4S from P1/P6.These lines are being further evaluatedfor their agronomic performance andfor possible exploitation in hybridbreeding.■

82-2-2-2-3R, IR10198-66-2R, andIR20933-68-21-1-1-2-1-R.

Seven genotypes were grouped intotwo classes on the basis of phenol colorreaction; IR20933-68-21-1-1-2-1R wasdistinct from the others because it didnot develop color. A comparison ofsodium dodecyl sulfate polyacry-lamide gel electrophoresis profiles ofsoluble proteins, albumin, and globulinrevealed maximum polymorphismamong the genotypes for the albuminfraction. All the genotypes weredistinct from each other except for the Aand B lines, in which only the presenceof a weak band of approximately 14,000kD molecular weight differentiatedIR58025A from its maintainer line.

Maximum polymorphism wasdetected in the isozyme patterns of ESTand POX, which differentiated all A

Identifying molecularmarkers for the gene(s)

governing thermosensitive genicmale sterility in rice

O.U.K. Reddy, Directorate of Rice Research(DRR), Rajendranagar, Hyderabad; E.A. Siddiq,Indian Council of Agricultural Research, NewDelhi 110001; J. Ali, Crop Improvement Depart-ment, Agriculture College and Research Institute,Navalur Kuttapattu, Trichy 620 009; A.J. Hussain,P. Arti, M.I. Ahmed, and N.P. Sarma, DRR, India

and R lines. The greater intensity of aPOX band having an Rm of 0.93differentiated IR58025A from its B line.A faint MDH band having an Rm of0.67 was detected in IR62829A, but wasabsent in its corresponding B line.Thus, a combination of the albuminprofile with a POX, EST, or MDHisozyme pattern could identify all eightlines studied. The phenol color reactionwas useful in verifying the identity ofIR20933-685-21-1-1-2-1R. Such charac-terization is useful in establishing orverifying the identity of a genotype andin differentiating one genotype fromanother. The uniformity of thesepatterns within the population of agenotype needs to be tested. Thepossibility of using isozyme markers totest genetic purity, particularly of Alines, is now being investigated.■

Thermosensitive genic male sterility(TGMS) in rice is advantageous inhybrid seed production compared withthree-line breeding. Genetic studiesrevealed four putative genes impartingthermosensitive male sterility in rice. Itwas observed that the segregationpattern of this trait in the F2, F3, andbackcrosses indicated that sterilitycaused by the TGMS trait wascontrolled by a single recessive gene. Itwas noted, however, that individualTGMS segregants drawn from the same

F2 population exhibited a differentialpattern for fertility alteration,suggesting the influence of modifiergenes. This makes the transfer of thischaracter difficult to achieve, asselection has to be made for theappropriate fertility alteration level inaddition to selection for the TGMS trait.Identifying a suitable molecularmarker linked to genes of the TGMStrait may therefore aid in selection. Thepresent study aimed at tagging genespertaining to TGMS through a bulked

Vol. 23, No. 2 11

Breeding methods

Heterosis over environ-ments in crosses involving

indica and tropical japonica ricecultivars

D.K. Dwivedi, M.P. Pandey, S.K. Pandey, and LiRongbai, Genetics and Plant Breeding Depart-ment, G.B. Pant University of Agriculture andTechnology, Pantnagar 263145, India

The discovery of wide compatibilitygenes in rice recently shifted theemphasis on enhancing heterosis fromintervarietal hybrids to intersubspecifichybrids. We therefore studied thenature and magnitude of heterosis ininter- and intrasubspecific crossesinvolving improved tropical japonica(J) parents (BSI 10 [P1], BSI 16 [P2],B4116 [P3], and B4122 [P4]) and indica(I) parents (Govind [P5], Manhar [P6],Pant Dhan 4 [P7], Sarjoo 52 [P8], PantDhan 12 [P9], and Narendra 359 [P10]).We evaluated all 45 hybrids andparents in a randomized completeblock design under environments with

optimum sowing and high fertility(E1), optimum sowing and optimumfertility (E2), and late sowing and highfertility (E3). We studied 10 agronomictraits, including harvest index andgrain yield, for heterosis, hetero-beltiosis, and standard heterosis usingPant Dhan 4 as the standard variety.

Analysis of variance indicated alarge variation among hybrids andparents. Moderate to high heterosis foryield and agronomic traits acrossenvironments was recorded. Trends inthe magnitude of heterosis were I/J >I/I > J/J for grain yield and plant heightand I/J > J/J > I/I for days to flowering.Standard heterosis for grain yieldpercentage in the respective E1, E2, andE3 environments ranged from -64.5 to146.1, -70.4 to 82.2, and -67.2 to 63.8. E1(optimum sowing and high fertility)gave a better response for heterosisexpression.

Higher heterosis in grain yieldaccompanied heterosis in paniclenumber, total dry matter and/or

Plant regeneration fromprotoplasts of wild rice, Oryzameyeriana Baill.

Lahong Sheng, Guangcun He, Lihui Shu, andLanjie Liao, Department of Biochemistry andBiophysics, College of Life Sciences, WuhanUniversity, Wuhan 430072, China

Oryza meyeriana is a wild rice speciesdistributed in southern China andSoutheast Asia. An accession found inChina’s Yunan Province has a strongresistance to bacterial leaf blight. It is

segregant analysis. TGMS line SA2,selected from mutagenified popu-lations, was nonallelic to other sourcesof TGMS genes and was designated astms4. TGMS and fertile bulks weremade from the F2 population of thecross SA2 (TGMS)/N22 (fertile).Seventy-five operon primers were usedto examine polymorphism in SA2, N22,

and bulks. Seventeen polymorphicproducts were specific to the fertileparent N22 and 19 were specific to theTGMS parent SA2. Of these, the 0.7-kbamplicon of OPA12 and 1.9-kbamplicon of OPS1 were specific to theTGMS trait.

TGMS line ID24 was allelic to NorinPL 12 (tms2) and was nonallelic to SA2.

Earlier studies on tagging the ChineseTGMS source reported that the 1.2-kbamplicon was closely linked. A bulkedsegregant analysis of the F2 of the crossID24/N22 revealed that a 1.2-kbamplicon of OPB19 was specific toTGMS. Further studies related tomapping are in progress. ■

spikelet number, and grain numberpanicle-1. Almost 95% of the hybridsrecorded negative heterosis forflowering and some were earlier by11-27 d across environments. HybridsP3/P8 in E1, P4/P6 in E2, and P1/P5 inE3 recorded 19.1, 20.1 and 20.4%standard heterosis for earliness and146.1, 66.8, and 60.2% for grain yield,respectively. Standard heterosis forheight was 2.0-13.7% across environ-ments. Both parents having Sd1 genescaused F1 height to increase onlyslightly. Promising hybrids for directexploitation were P3/P8, P1/P9, andP1/P10 in E1; P4/P7, P4/P6, and P4/P10 in E2; P4/P7, P7/P9, and P7/P8 inE3; and P3/P8, P4/P7, and P4/P10over environments. These hybridsshowed good prospects for increasingpotential yields and per-day produc-tivity in the subtropics. Studies are inprogress to improve parents for thethermosensitive genic male sterilitytrait and resistance against bioticstresses. ■

difficult, however, to transfer thedisease resistance character of O.meyeriana to cultivated rice (O. sativa)by sexual hybridization because ofsexual incompatibility between thetwo species.

Plant regeneration from protoplastsis important for rice breeding and is aprerequisite for genetic manipulationvia somatic hybridization. Progress hasbeen made in protoplast culture andregeneration of cultivated rice andother Oryza species. Plants have beenrecovered from cryopreserved calli in

O. meyeriana. Here, we report onsuccessful plant regeneration fromprotoplasts of this wild rice.

The immature panicles in the stamenand pistil differentiation stage weretaken from cryopreserved callus-regenerated plants and used asexplants. Calli were induced on N6medium containing 2 mg L-1 2,4-dichlo-rophenoxyacetic acid, 45 g L-1 sucrose,and 5 g L-1 agar at pH 5.8. Pale yellowprimary calli with a dry appearancewere transferred to liquid AA2medium. Fine dispersed calli were

12 IRRN 1998

selectively subcultured every 3 d andmaintained in the dark at 25 °C on agyratory shaker at 120 rpm. The sus-pension cultures with a high regenera-tion frequency were established within1 mo and were suitable for protoplastisolation. The suspension cells wereprecultured for 3 d on MS liquidmedium before they were isolated byenzyme solution.

The isolated protoplasts had densecytoplasm and many starch grainswere seen in them. The highest yield ofprotoplasts was 2 × 107 g-1 fresh weightcells. Protoplasts were cultured in KPRmedium with the membrane nurseculture method. An aliquot of about0.4 mL liquid KPR medium containingprotoplasts (4 × 104) was plated onto afilter (Whatman, 0.2 µm) placed over alayer of nurse cells. Nurse cells wereembedded in KPR medium with 1%agarose. After 20-30 d of culture, the cellclusters formed could be seen by thenaked eye. The rate of colony formationwas about 0.0075%.

When protocalli grew to 1-2 mm indiameter, they were transferred to N6medium for proliferation and then toMS regeneration medium for plantregeneration. Green plantlets wereregenerated from protocalli after2-3 wk. The average plant regenerationfrequency was 3 plants per 1 × 107

protoplasts. The total time from callusinitiation to plant regeneration ofprotoplasts was 4-6 mo.

Our results showed that thepresence of nurse cells was required toinduce division of protoplasts. In con-trast to suspension cultures, it was diffi-cult to isolate pure populations of intactprotoplasts directly from the primarycalli. Suspension cells with a goodgrowth status, dense cytoplasm, andembryogenic characters played a cri-tical role in successful culture.Protoplast division and colonyformation also depended on the ageand status of the suspension cells usedfor preparation of the layer of nursecells. Establishment of a protoplastregeneration system makes it possibleto transfer resistance genes fromO. meyeriana into cultivated rice. ■

Chinoor: a promising spontane-ous mutant and high-quality ricevariety of Madhya Pradesh,India

D. Sharma, Department of Plant Breeding andGenetics, I. G. Agricultural University, Raipur;and M.P. Janoria, J.N. Agricultural University,Jabalpur, Madhya Pradesh, India

Chinoor is a low-yielding, late, tall, andlodging-prone indigenous rice varietyof Madhya Pradesh, India. It continuesto be grown in spite of its low yieldbecause of its high-quality aromaticslender grains that sell for more thantwice the price of Mahsuri. Chinoor isalso exported to Middle East countriesto a limited extent.

We began efforts to reduce durationand plant height of Chinoor in 1995with a survey in farmers’ fields tocollect possible spontaneous mutantsof interest. Progenies of the 25 single-plant selections made were trans-planted along with Chinoor in non-replicated plots for observation. Eachprogeny plot consisted of 50 plants

grown in 2 rows 20 cm apart. Plants in arow were spaced 15 cm apart. Fertilizerwas applied at the rate of 60-40-20 kgNPK ha-1. Twenty plants from each plotwere selected randomly and taggedbefore flowering for observation.

One of the progenies flowered 31 dearlier, matured 38 d earlier, and had 28cm less height than Chinoor. A compar-ison of the mutant with Chinoor forgrain yield, its components, and somephysical quality characters was made(see table).

Although kernel aroma and trans-lucence apparently stayed the same,the mutant recorded 80% higher grainyield plant-1 than Chinoor. This highyield advantage of the mutant resultedlargely from its 30% more paniclesplant-1 and 20% higher grain weightpanicle-1.

The Chinoor mutant, with its higheryield potential and reduced durationand plant height, seems to have poten-tial to not only replace Chinoor but alsoto be extended to new rice areas whereit may enhance productivity and thusprofitability of rice cultivation. ■

Comparison of Chinoor mutant with Chinoor, Waraseoni, India, 1996 wet season.

Character Chinoor mutant Chinoora

(mean ± SD) (mean ± SD)

Days to 50% flowering 89.6 ± 2.4 120.7 ± 2.5Days to maturity 114.7 ± 1.6 153.0 ± 1.3Plant height (cm) 113.2 ± 1.9 141.0 ± 2.5Panicles plant-1 (no.) 12.8 ± 1.7 8.8 ± 1.5Grains panicle-1 (no.) 211.3 ± 4.1 184.1 ± 7.61,000-grain weight (g) 15.3 ± 0.2 16.8 ± 0.2Panicle weight (g) 2.4 ± 0.3 2.0 ± 0.2Grain length (mm) 8.1 ± 0.1 7.5 ± 0.1Grain width (mm) 2.1 ± 0.1 1.9 ± 0.1Kernel length (mm) 6.0 ± 0.1 5.8 ± 0.1Kernel width (mm) 1.9 ± 0.1 1.7 ± 0.1L/B (grain) 3.9 ± 0.2 4.0 ± 0.1L/B (kernel) 3.2 ± 0.2 3.4 ± 0.2Plant yield (g plant–1) 31.6 ± 2.0 17.5 ± 2.7

aAv of 20 individuals.

Multiple submissions. Normally, only one report for a single experiment will beaccepted. Two or more items about the same work submitted at the same timewill be returned for merging. Submitting at different times multiple notes fromthe same experiment is highly inappropriate. Detection will result in the rejectionof all submissions on that research.

Vol. 23, No. 2 13

Combination selection methodfor rice breeding efficiency

Pak Chang Hong, Rice Research Institute,Academy of Agricultural Sciences, PyongyangDemocratic People's Republic of Korea

A new method of breeding wasdeveloped in which cross combinationsare selected based on productivity inearly generations of hybrids. The goalis to develop high-yielding varieties.In this method, crosses are madebetween superior lines adaptable to theclimatic and soil conditions of a specificarea as females, and improved lines,intersubspecies derivations, disease-resistant lines, and new high-yieldingvarieties as males.

The new method consists of thefollowing steps (see figure):

In the F2 generation, plants are trans-planted at 3 hill-1 in a 10-m2 area, and acheck variety is grown for every 4combinations. Only those combina-tions with good character traits andthat yield at least 90% of what the checkyields are selected.1. In the F3 generation, the same

method as that for the F2 generationis followed, but combinations withgood character traits and that yieldat least 80% of what the check yieldsare selected.

2. In the F4 generation, 1 plant istransplanted hill-1 in an area of 66 m2

~100 m2 for each combination, andindividual plants are selected.

3. In the F5 generation, the plantsselected in the F4 generation are line-cultivated (3.3 m2 line-1), and lineswith good productivity and traitsare selected.

4. In the F6 generation, line tests, uni-formity tests, and other preliminarytests are carried out with thepromising lines.

5. In the F7 and subsequent genera-tions, line tests, uniformity tests,preliminary yield tests, and compa-rison tests are made, and adaptabili-ty to specific agroclimatic conditionsis evaluated and lines are selected.

The advantages of the method are asfollows:1. Selection of combinations can be

made scientifically based on produc-tivity and promising combinationscan be adapted.

2. The numbers of lines can be reducedand the field experiments carriedout in smaller areas with less labor.

3. The effect of selection is very highbecause a choice is made with onlythe relevant fixed plants.

Design for combination selection in breeding.

Female × male

F1

20 plants,1 plant hill-1,selfing test,quality test,trait test

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

F2

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

vv

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

F5

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

F6

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

:

vvvvv

vvvvv

vvvvv

vvvvv

F3

F4

Line comparison Uniformitytest

Trait test Preliminaryand comparison

test

F7 and further generations: line comparison, uniformity test, preliminary and comparison tests,test of adaptation, state variety comparison.

3 plants hill-1, in 10 m2

(1 control per 4 combinations),1,000 plants,quantity comparison,combination selection,quality test

3 plants hill-1, in 10 m2

1,000 plants,quantity comparison,combination selection,quality test

1 plant hill-1, in 66 m2,2,400 plants,individual selection

3.3 m2 line-1,100 plants,30-50 lines combination-1,quantity comparison,quality test

14 IRRN 1998

Germination rate (%) of three varieties under low-temperature conditions at Pyongyang in 1995.

Treatment 1a Treatment 2b

Variety6 d 12 d 18 d 3 d 6 d 9 d

Pyongyang 24 80 98 100 90 92 91Yomju 1 (tolerant check 1) 12 96 100 100 100 100Olbyeo 1 (tolerant check 2) 10 78 82 91 98 100

a15/10 °C day/night temperature. b30/25 °C day/night temperature.

4. Accurate observations of plants andlines are possible because selectionwas made only from combinationsof highly productive populations.This technique has been applied to

develop high-yielding varieties since1983 and two new high-yielding ricevarieties, Toenonbyeo 2 and Pyong-yang 24, were developed successfully.

The mean yield of Toenonbyeo 2, incomparative tests from 1992 to 1994,was 9,475 kg ha-1, outyielding the checkby 1,852 kg ha-1. The yield of this vari-ety, under conditions of no-fertilizerapplication in 1995, was 8,101 kg ha-1 inRakrang District, where soil fertility ishigh. Its record yield under experi-

mental conditions was 11,706 kg ha-1

with 120 kg ha-1 of applied N.The mean yield of Pyongyang 24 for

3 tests from 1993 to 1995 was 8,732 kgha-1, which was 1,873 kg ha-1 higherthan that of the check. This varietyshowed high yield under conditions oflow-fertilizer application. It yielded

Yield stability analysis ofrice hybrids

S. Hegde and B. Vidyachandra, RegionalResearch Station, VC Farm, Mandya 571405,Karnataka, India

The stability of some promising ricehybrids released for yield and yieldcomponents (panicles plant-1, spikeletspanicle-1, fertility, and grain weight)was studied at Mandya, Bangalore,Shimoga, Kathalagere, and Brahmavarin Karnataka. These locations repre-sented different agroclimatic zones.Data from the 1994 wet season on fivehybrids and four checks were analyzedas per the Eberhart and Russell model.

Highly significant genotype × envi-ronment interactions were observed foryield and yield components. No hybridor check variety showed stability overthe environments studied for yield. Butall the hybrids showed an averageperformance with a regression coeffi-cient (bi) equal to unity. This was main-ly due to instability in the number ofspikelets panicle-1 and the degree offertility.

Even though the regression coeffi-cient deviated from unity, medium-duration hybrid IR58025A/KMR3(KRH2) was the best performer at allthe locations, except Brahmavar incoastal Karnataka. Another short-duration hybrid, IR58025A/IR9761-19-

1R (KRH1), also performed similarly,with a yield much superior to that ofchecks Rasi and Mangala at all loca-tions except Brahmavar. The resultsfrom the correlation analysis betweenstability parameters (bi and s2di) forgrain yield and yield componentsindicate that stability parameters in ricehybrids may be governed by differentgenes and gene interactions. ■

7,650 kg ha-1 in 1995 in the demonstra-tion plot in Sariwon City, with theapplication of 200 kg ha-1 of ammoniumsulfate. Its highest recorded experi-mental yield was 11,282 kg ha-1.Pyongyang 24 is also highly tolerant ofcold (see table). ■

Yield system analysis inrice hybrids

M.S. Ramesha, B.C. Viraktamath, M. Ilyas-Ahmed, C.H.M. Vijayakumar, and S. Singh,Directorate of Rice Research, Rajendranagar,Hyderabad 500030, India

Grain yield is a complex trait that isinfluenced by genetically controlledphysiological components. The rate ofaccumulation of biomass and actualyield, as well as the ratio between thesetwo components, can determine agenotype’s yielding ability. At a similarlevel of harvest index, increased grainyield can come from increased totalbiomass or increased harvest index, orincreased biomass and harvest index.We therefore investigated thephysiological and genetic causes ofyield superiority in rice hybrids in the1994 wet season.

The test materials involved eight ricehybrids belonging to different maturitygroups, their A, B, and R parents, andthree check varieties—Jaya, IR72, andTellahamsa. They were grown in threereplications using a randomized com-plete block design. Observations onshoot and root dry weights at differentgrowth stages were recorded on fiverandom and competitive plants in eachreplication. Grain yield, panicle dryweight, and spikelet fertility were alsorecorded. Based on the observations,total biomass at different growth stagesand harvest index were calculated.

Heterosis over the best check andbetter parent was calculated andsubjected to a test of significance. OnlyIR58025A/Swarna, PMS3A/PR103,and IR58025A/IR32809 showed signi-ficant commercial heterosis for grainyield, total dry matter, and panicle dryweight at harvest despite a negativeheterosis for spikelet fertility. Thesehybrids also showed heterosis for drymatter accumulation at many growthstages. IR58025A/IR52256 andIR62829A/IR53901 showed negativecommercial heterosis for grain yield,harvest index, and spikelet fertilitywith no or very low heterosis for othergrowth parameters. The negativeheterosis for spikelet fertility in all thehybrids clearly indicated a need toimprove spikelet fertility in hybrids by

Vol. 23, No. 2 15

resorting to special breeding to increaserestoration ability. The superior yieldingability of the hybrids over the checkswas found to come from increased totalbiomass and increased panicle weight,with almost the same level of harvestindex. Harvest index and panicleweight, however, can be further in-creased by increasing spikelet fertility inhybrids, which can ensure an additionalincrease in total grain yield. ■

Studying heterosis forgrain yield and its

components in hybrid rice

D.V.S. Panwar, R. Kumar, A. Singh, andB.S. Mehla, CCS Haryana Agricultural Univer-sity, Rice Research Station, Kaul 136021, India

Heterosis in hybrid rice was studied inthe 1995 wet season under irrigatedconditions. Some 22 F1 hybridsdeveloped by using cytoplasmic malesterile lines IR58025A, PMS1A, PMS2A,PMS3A, and PMS10A and 18 restorerswere grown along with the standardvariety HKR126 in a randomized blockdesign with three replications. Eachline was planted in a single row 3 mlong. Spacing between rows and plantswas kept at 30 × 15 cm. Observationswere recorded for grain yield plant-1,panicles plant-1, grains panicle-1, and1,000-grain weight.

The standard heterosis of F1 hybridsfor these traits was calculated overHKR126. The standard heterosis wasfound to be significant for all four traits.It was both negative and positive forgrain yield plant-1 (-0.57 to 54.75%),panicles plant-1 (-14.84 to 89.14%), andgrains panicle-1 (-16.04 to 43.28%), andnegative for 1,000-grain weight (-34.55to -5.82%). Positive and significantstandard heterosis was observed in 6hybrids for grain yield plant-1, in 12hybrids for panicles plant-1, and in 7hybrids for grains panicle-1. Thehybrids PMS1A/Pusa 44-33 (54.75%),PMS3A/RP2151-40-1 (40.86%),PMS2A/IR 31802-56-4-3-3R (37.37%),IR58025A/IRON89-54 (36.48%),

PMS1A/RP2151-173-1-3 (33.21%), andPMS2A/Pusa 44-33 (32.86%) showedpositive and signi-ficant standardheterosis for grain yield plant-1,panicles plant-1, and grains panicle-1. ■

Standard heterosis of ricehybrids for yield and yield

components

K.V. Sitaramaiah, Ch.V.D. Rani, and N.S. Reddi,Rice Research Unit, Agricultural College Farm,Bapatla 522101, India

The exploitation of hybrid vigor iswidely recognized as the only readilyavailable means to raise the geneticyield ceiling in areas where yields havealready approached their potential. Inthis approach, developing highlyheterotic rice hybrids with superioryield performance and evaluatingthem across environments are im-portant. We assessed the performanceof 10 promising rice hybrids to estimatethe standard heterosis over popularmedium-duration rice variety Prabhatfor yield and yield components.

Significant differences existedamong the hybrids for all charactersstudied, thus indicating considerablevariability. Hybrids were of early tomedium duration and had a semidwarfplant stature. Six hybrids had signifi-cantly higher grain yield over Prabhat.MTUHR2015 had the highest grainyield (5.65 t ha-1), followed byMTUHR2002 (5.13 t ha-1). The releasedhybrids APHR1 and APHR2 had the4th and 5th positions, respectively.Though hybrids had long panicles withmore spikelets panicle-1, the expectedgrain yields were not realized becauseof spikelet sterility and lower grainweight.

The standard heterosis of hybrids forplant height was insignificant except inMTUHR2006, which showed a nega-tive response (-17.8%). Similarly,MTUHR2020 and MTUHR2015 exhi-bited significant and positive heterosisfor panicle length. All hybrids exceptMTUHR2006 exhibited a very highpositive standard heterosis for number

of spikelets panicle-1. This was reflectedin the positive high heterosis for grainyield. All hybrids showed negative andsignificant standard heterosis for 1,000-grain weight because Prabhat had abold grain with a high 1,000-grainweight of 32 g. Most of the hybridspossessed slender grains with lesserweight.

Heterosis is a highly cross-specificphenomenon. To use heterosis success-fully to improve grain yield, parentalgenotypes need to have a high potentialyield. We need to reduce spikelet steri-lity and increase grain weight to realizethe effect of high heterosis for numberof spikelets panicle-1 in rice hybrids. ■

Developing Pusa 5A, astable indica CMS line with

high outcrossing potential

M.J. Abraham, F.U. Zaman, U.S. Natarajan,A. Mahendru, and F. Mohammad, Division ofGenetics, Indian Agricultural Research Institute(IARI), New Delhi 110012, India

From available germplasm andbreeding lines at IARI, we haveidentified Pusa 150-2, a maintainerwith excellent agronomic traits. Pusa150-2 is a high-yielding semidwarfindica variety with excellent grainquality.

When test-crossed with IR58025A, itproduced F1 progeny with 100% pollensterility. The F1 progeny was back-crossed to Pusa 150-2 and a populationof approximately 750 BC1 plants wasraised. All plants showed 100% pollensterility under the microscope butsegregated for various agronomiccharacters. Thirty plants showingcharacters comparable with those ofPusa 150-2 were backcrossed. All theprogenies were 100% sterile.

Fifteen of these progenies wereselected for further backcrossing. Allplants in the BC3 population showed100% sterility. Five of these populationsshowing close similarity to Pusa 150-2were further backcrossed and the BC4population was raised. All five popula-tions showed perfect pollen and

16 IRRN 1998

spikelet sterility. In the BC5 also, pollenand spikelet sterility were maintainedin the progenies. The CMS line derivedfrom progeny 1 was named Pusa 5A.

The outcrossing potential of this linewas evaluated against IR58025 A,PMS2A, PMS 3A, and PMS10A usingthe respective B lines. Pusa 5A recordeda seed yield of 2.8 t ha-1 vs 2.0, 0.6, 0.7,and 0.8 t ha-1 of IR58025A, PMS2A,PMS3A, and PMS10 A, respectively.Pusa 5A had more tillers, longer pani-cles, more spikelets panicle-1, and betterstigma exsertion than IR58025A. Pusa5A may therefore prove to be a goodalternative to IR58025A, the mostpopular cytoplasmic male sterile lineused in India. ■

Field evaluation of thermo-sensitive genic male sterile

lines

C.H.M. Vijayakumar, M.Ilyas-Ahmed,B.C. Viraktamath, M.S. Ramesha, and S. Singh,Directorate of Rice Research, Rajendranagar,Hyderabad 500030, India

The use of thermosensitive genic malesterility (TGMS) is well recognized intwo-line heterosis breeding. The use ofthe TGMS system has several advan-tages over the cytoplasmic genic malesterility system, such as (1) its simpleand efficient seed production system,and (2) its greater scope for enhancingheterosis because any genotype can beused as a male parent. To develophybrids using the TGMS system, it isnecessary to identify or develop linesthat show a desirable transformationfrom fertility to sterility and vice versa.

We evaluated several TGMS linesthat carried TGMS genes introducedfrom IRRI. Besides intensive studyunder field conditions for five seasons,starting in 1994, we made simultaneousindividual plant selections in theselines. The TGMS lines were (1) IR68945-4-33-14, (2) IR68948-12-3-7, (3) IR68949-11-5-31, (4) IR68945-4-33-4-14, and (5)IR32364-120-1-3-2. Lines 1, 2, and 3showed very good transformationbetween seasons. At Hyderabad,

during the wet season, the weather isunstable and daily mean temperaturesbelow 25 °C occur frequently. The dailymaximum temperature is 29 °C, whichinduces fertility in most of the IRRI-bred TGMS lines. During the dry sea-son, however, the daily mean tempe-rature is always >25 °C, with a dailymaximum of >32 °C beginning in thesecond week of March, which inducessterility in most lines, including theselected lines 1, 2, 3, and 4.

During the 1995-96 dry season, 325plant progenies of TGMS lines 1, 2, 3,and 4 were evaluated in two setsplanted at 10-d intervals. The resultsshowed that IR68945-4-33-14 stillsegregated for plant type, grain type,and sterility, whereas IR68948-12-3-7and IR68949-11-5-31 were almoststable. Another line, IR68945-4-33-4-14,was uniform, but partially fertile, andappeared to be a very high-tempe-rature sterile type. Plants showingstable sterility were selected and theirstubble was planted in the 1996 wetseason, in which all those lines trans-formed to fertility further confirmedtheir behavior. IR32364-120-1-3-2 didnot set seeds in either season after 1994.Although the anthers were light yellow,they did not shed any pollen, whichwas >99% abortive type. Perhaps it is avery low-temperature fertile type.Lines IR68948-12-3-7 and IR68949-11-5-31, which also have desirable floraltraits that influence outcrossing, will beused in hybrid rice breeding in thecoming years. ■

Developing thermo-sensitive genic male sterile

lines in rice

K. Thiyagarajan, P. Jayamani, R. Latha,P. Suthamathi, and M. Rangaswamy, School ofGenetics, Tamil Nadu Agricultural University,Coimbatore 641003, India

The three-line method (A/B/R) ofheterosis breeding is cumbersome andwarrants development of alternateapproaches to exploit hybrid vigorcommercially in rice. Two-line breeding

is one such possibility that emergedfollowing the chance discovery ofphotoperiod-sensitive genic male sterile(PGMS) lines and thermosensitive genicmale sterile (TGMS) lines in China.

A study was conducted to identify astable TGMS source from suspected andintroduced TGMS lines under naturalconditions. Thirty-eight lines wereevaluated at Coimbatore (high tempera-ture 38/23 °C) and at Gudalur (lowtemperature 30/16 °C) simultaneouslyduring summer 1995. Pollen fertility inthese lines was recorded during theflowering stage using I-KI (1%) solution.Panicles were bagged and spikeletfertility recorded.

Lines TS 15 and TS 16 showed 100%pollen sterility at Coimbatore and 55-80% pollen fertility at Gudalur. Thesterile lines are maintained as stubble atCoimbatore for further evaluation in thewet season. These lines are also evalua-ted under controlled conditions todetermine the critical sterility and fertili-ty points of temperature. Crosses arealso made simultaneously with agrono-mically superior lines to transfer theTGMS genes. ■

Male sterility and fertilitybehavior of suspected

thermosensitive genic malesterile (TGMS) lines

C.R. Elsy and M. Rangaswamy, School ofGenetics, Tamil Nadu Agricultural University,Coimbatore 641003, India

The male sterility-fertility response inthermosensitive genic male sterile(TGMS) lines was studied to identifytheir adaptability to different tempera-ture regimes. Twenty suspected TGMSlines available at the Paddy BreedingStation, Coimbatore, were used for thestudy. The plants of these lines, whichwere completely male sterile duringsummer 1995, were tagged and ob-served for their male sterility-fertilitybehavior in the ratooned tillers duringwinter 1995 and summer 1996 bystaining pollen grains with I-KI. Panicledevelopment stages, from the

Vol. 23, No. 2 17

formation of pollen mother cells to thelate uninucleate stage of pollen grains,were considered as sensitive totemperature. These stages occurred9-19 d prior to heading (or 9-17 d afterpanicle initiation). Therefore, thecritical temperature at which plantsshowed variation in pollen sterility andfertility was calculated by consideringthe average daily temperature andaverage maximum temperatureprevalent during 9-19 d prior toheading.

Observations on the mean daily andmaximum temperature during thesensitive stages of the identified TGMSlines indicated that these lines showedmore pollen fertility when the meandaily temperature was ≤26 °C, with amean maximum temperature of<32 °C. Similarly, a majority of the linesbecame male sterile when the meantemperature was >27 °C, with a meanmaximum temperature of >33 °C.TGMS 6 (RP2161-106-1 S), TGMS 9(mutant of IR50), TGMS 10 (C12japonica purple S), TGMS 16 (IR68945-33-4-14-40 S), and TGMS 18 (IR68949 S)showed sterility-fertility changes undersuch temperature conditions. Theselines showed high pollen fertility inDecember and January and pollensterility in April and May. TGMS 16 andTGMS 18 remained pollen-free withempty anther sacs in the peak summermonths, indicating the more stablenature of these lines. TGMS 18remained pollen-free up to the firstfortnight of June.

In November, TGMS 6 showed 25%pollen fertility at a mean daily tempera-ture of 27 °C and mean maximum of31 °C. When the mean daily tempera-ture remained the same during March,but the mean maximum rose to 35 °C, itshowed 100% pollen sterility. Thisindicated the pronounced influence ofmaximum temperature on the sterility-fertility behavior of this line. For TGMS18 also, the influence of maximumtemperature became clearer, when theline showed 80% pollen fertility duringSeptember at a mean daily temperatureof 27 °C and maximum temperature of32 °C. The line turned out to be 100%

sterile when plants were exposed to thesame mean daily temperature (27 °C),but a higher maximum temperature(34 °C).

These TGMS lines can be success-fully used in two-line breeding byundertaking hybrid seed productionprograms in areas where constantlyhigh temperatures (mean daily tempe-rature >27 °C, mean maximum tem-perature >33 °C) prevail in the summermonths. Similarly, these lines can bemaintained at high-altitude areaswhere the temperature is low or even inthe plains during December-January. ■

Hybrid rice research in Pakistan

Syed Sultan Ali and M.G. Khan, Rice ResearchInstitute, Kala Shah Kaku, Pakistan

In Pakistan, hybrid rice research beganin 1991-92. In evaluations made during1992-95, IRRI hybrids outyielded localvarieties KS 282 and IR6 by more than50%. For practical exploitation ofhybrid rice technology, local Basmatiand non-Basmati germplasm is beingconverted into commercially usablecytoplasmic male sterile (CMS) lines.17156A is the newly developed BasmatiCMS line (male sterility source,IR58025A). Restorers are beingidentified for this line. Less restorerand higher maintainer frequency wereobserved in local germplasm.Therefore, A × R and R × R crosses arebeing made to increase restorerfrequency. Maintainer frequency isbeing increased using B × B crosses.IR58025A, IR62829A, IR68897A,IR68896A, IR69617A, and 913A showedan excellent stigma exsertion ratio,outcrossing rate, and other floralcharacteristics that influenceoutcrossing. One recessive gene foraroma and two dominant genes forfertility restoration of wild abortivecytoplasm were observed. Antherlength and pollen grain size wereobserved to be monogenicallycontrolled traits. Thermosensitive genicmale sterility and the wide compati-bility trait are being studied. Recently

developed hybrid IR58025A/KS 282 isbeing field-tested. Target areas for seedproduction are being located. Grainquality analysis shows that bothparents should have accep-table grainquality to develop rice hybridspossessing that same quality. ■

Hybrid rice: status and future inBangladesh

A.W. Julfiquar, Bangladesh Rice ResearchInstitute, Gazipur-1701, Bangladesh

The Bangladesh Rice Research Institute(BRRI) started hybrid rice research in1993 through an informal collaborationwith the International Rice ResearchInstitute (IRRI). Initial work involvedtesting F1 hybrids, and evaluatingcytoplasmic male sterile (CMS) andrestorer lines received from IRRI. InCMS lines IR58025A and IR62829A, anoutcrossing rate of 3% was recorded.The maintainers showed 80-90%spikelet fertility and a number ofrestorers were also identified. Somewell-adapted varieties/lines wereidentified as maintainers or goodrestorers for the wild abortive cyto-sterility system. Some elite restorerlines are BR29, BR5876-6-2-1, BR4839-17-2-2-HR42, BR5690-62-23, BR5882-12-2-1 and BR5892-32-5-3. SomeChinese CMS lines and their main-tainers were also evaluated for theiradaptability and performance. TheseChinese lines were not adapted toBangladesh conditions and were highlysusceptible to disease and insects.

A number of IRRI-developedhybrids were tested in multilocationalyield trials during the 1995-96 boroseason at Gazipur, Comilla, Bhanga,and Habiganj. A number of rice hybridsoutyielded the standard check varietyof the same duration by more than 1 tha-1. Grain quality characteristics of thetested hybrids were comparable withthose of our recommended checkvarieties. These results have encour-aged rice scientists and policymakers todevelop and use hybrid rice technologyin Bangladesh.

18 IRRN 1998

Grain quality

A working group for hybrid rice hasbeen formed with a plant breeder, seedproduction specialist, and plant patho-logist and entomologist. This programinvolves collaborative research withthe Bangladesh Agricultural Develop-

ment Corporation, and other publicand private agencies engaged in seedproduction in the country. A specific,time-bound, and goal-oriented workplan has been made for developinghybrid rice technology for the boro

season. Routine evaluation of elite linesand introduced rice hybrids is conti-nuing. Four hybrid combinations werefound promising and these hybrids areundergoing on-farm trials. ■

Number of entries promising for key quality traits, 1996 wet season.

Table 1. Number of entries identified as promising for key quality traits, 1996 wet season.

Quality traits

Category Number Head rice Intemerdiate Intermediaterecovery gelatinization amylose(>60%) temperature content (%)

Long slender 23 9 3 9Long bold 5 3 1 4Medium slender 2 2 1 1Short bold 3 2 1 3 Total 33 16 6 17

Donors for quality traits from theinternational aromatic nursery

N. Shobha Rani, B. Krishna Veni, P. BhaskarReddy, G.S.V. Prasad, and U. Prasada Rao,Directorate of Rice Research, Rajendranagar,Hyderabad 500030, Andhra Pradesh, India

To identify some varieties with basmatibackground to be used as donors, thefirst International Fine Grain AromaticRice Observational Nursery (IRFAON),consisting of 33 entries including theinternational check Basmati 370 alongwith local check Pusa Basmati 1, wasgrown under normal cultural practicesduring the 1996 wet season. All thevarieties were analyzed for 14 physi-cochemical characteristics at theDirectorate of Rice Research, Hyde-rabad, using standard procedures.These characteristics involve grain sizeand shape, endosperm appearance,and milling and head rice recovery(HRR). Among the cooking and eatingquality traits, water uptake (WU),volume expansion ratio (VER), kernellength after cooking (KLAC), elonga-tion ratio (ER), gelatinization tempera-ture (GT), and amylose (AC) contentwere studied.

Of the entries tested, 22 belonged tolong slender, 5 to long bold, 2 tomedium slender, and 3 to short boldgroups (Table 1) (see figure). In the longslender group, DR28 (7.38 mm), DR29(7.37 mm), and Shah Pasand (7.35 mm)recorded high kernel length. Sixteenvarieties showed >60% head rice,ranging from 60% (PK1656-48-2-2-1) to69.7% (ARC11554). Among theseentries, Basmati 385 and Shah Pasandexhibited high KLAC and ER, compa-rable with those of the checks. Basmati385 recorded 14.5 mm KLAC and an

elongation ratio of 2.15 and ShahPasand exhibited 14.2 mm KLAC andan elongation ratio of 1.93. Otherentries that possessed moderate to high

KLAC combined with moderate ERwere Binam, Domsiah, DR28, DR29,PK1379-9-1-1, PK1399-12-1-1, andSuper Basmati.

Total

35

30

25

20

15

10

5

0

Shortbold Medium

slender Longbold Long

slender

Amylose content

Gelatinizationtemperature

Head ricerecovery

Number

17

16

6

3

11 1 9 3

2 2 3

32

5

23

9

14

33

Number

Vol. 23, No. 2 19

Table 2. Summary of grain quality characteristics of promising varieties in the International Fine Grain Aromatic Rice Observational Nursery, 1996 wet season.

Variety Origin HRRa (%) KL (mm) L/B Alk val. WU (mL) VER KLAC (mm) ER AC (%) Scent Grn. chalk

Long slender groupDR28 Pakistan 61.0 7.38 3.53 7.0, 7.0 310 4.0 13.2 1.82 26.1 3 VOCDR31 Pakistan 61.8 6.88 3.31 7.0, 7.0 230 3.8 11.3 1.64 26.4 2 VOCDomsiah Iran 52.0 6.80 3.54 4.8, 5.0 290 4.7 13.2 1.94 18.2 3 VOCPK1379-9-1-1 Pakistan 56.3 6.62 3.15 6.0, 6.0 250 4.3 13.2 1.99 19.0 3 VOCPK165648-2-2-1 Pakistan 60.0 6.50 3.16 7.0, 7.0 270 4.3 12.2 1.88 22.6 2 ADM 38 Pakistan 64.2 6.53 3.27 3.1, 3.1 130 4.0 12.1 1.85 21.4 3 OC

Long bold groupAzucena Philippines 64.8 6.39 2.88 4.2, 3.2 130 4.0 11.3 1.77 20.9 2 OC

Short bold groupARC 11554 India 69.7 4.17 2.12 5.0, 4.3 110 4.3 7.6 1.82 21.8 2 ATulsimanjari 14-2 India 67.5 4.20 2.19 3.0, 3.0 100 4.3 7.9 1.88 21.0 2 A

aHRR = head rice recovery, KL = kernel length, L/B = length/breadth, Alk val. = alkali value, WU = water uptake, VER = volume expansion ratio, KLAC = kernel length after cooking, ER = elongation ratio,AC = amylose content, scent: 3 = strong, 2 = moderate, Grn. chalk = grain chalkiness: A = absent, VOC = very occasionally present, OC = occasionally present.

Seventeen varieties had the mostdesirable AC range of 20-25%, and sixvarieties showed intermediate GT.Three varieties—ARC11554, Azucena,and Basmati 385—possessed bothintermediate AC and GT. Except forPK1385-6-3-1-2, all other varieties hadan aroma, which is the typicalcharacteristic of basmati rice.

With excellent phenotypic acceptabi-lity, DR28 and DR31 (Pakistan) had>60% HRR, moderate to high KL andKLAC, and slightly higher AC(Table 2). Domsiah (Iran) was anothervariety that possessed all the qualitytraits in a desirable range, with a slight-ly low AC. Other varieties fromPakistan having moderate to high

HRR, KL, and KLAC coupled with goodphenotypic acceptability are PK1379-9-1-1, PK1756-49-2-2-1, and DM38. Withthe emphasis now shifting toward thedevelopment of dwarf short-grainedaromatic rice for domestic and exportpurposes as well, Azucena, ARC 11554,and Tulsimanjari 14-2 may also be usedas donors depending on the objectives. ■

Effect of bran removal on degree of polish and oil recovery of rice bran.

Time of Basmatia UPR79milling(s) Dp (%) Or (%) Wg (g) Dp (%) Or (%) wg (g)

0 0 0 0.02299 0 0 0.02145 10 3.33 15.70 0.02146 2.75 14.57 0.02054 20 3.94 18.57 0.02055 3.49 14.90 0.02032 30 5.08 19.25 0.02010 4.46 15.56 0.01962 40 6.32 19.52 0.01956 5.06 16.55 0.01732 50 7.02 20.91 0.01909 5.34 17.46 0.01730 60 7.83 20.91 0.01801 5.81 17.90 0.01676 70 8.13 19.95 0.01755 6.25 17.02 0.01615 80 8.51 19.51 0.01760 6.56 16.34 0.01610 90 8.84 18.76 0.01700 6.97 15.72 0.01600100 8.94 17.60 0.01620 7.18 15.58 0.10536110 9.05 17.07 0.01618 7.57 15.07 0.01480

Statistical parameters of Or = aDp - bDp2

Variety a b r SEE

Basmati 6.367 0.4859 0.999 0.721UPR79 6.348 0.5822 0.991 0.699

aDp = degree of polish on brown rice basis, %; Or = oil recovery of rice bran, %; Wg = weight of one grain, g; r = correlationcoefficient; SEE = standard error of estimate.

Effect of bran removal on oilrecovery

J.P. Pandey, Department of Post-HarvestProcessing and Food Engineering, G.B. PantUniversity of Agriculture and Technology,Pantnagar 263145, Nainital, India

For the past two decades, India hasbeen meeting a shortage of oils byimporting this commodity at extremelyhigh costs in terms of foreign exchange.The demand for vegetable oils in theyear 2000 will be 8.75 million t. Toaugment oil and fat resources in India,rice bran oil needs to be made availableto the maximum extent.

Experiments were conducted withtwo popular and common varieties ofparboiled rice: coarse (UPR79) and fine(Basmati) rice grown in the Tarai regionof Uttar Pradesh, India. The objective ofthis study was to evaluate the effect ofbran removal on oil recovery. Clean ricesamples were collected from a commer-cial rice mill.

The moisture content of collectedsamples of UPR79 and Basmati wasdetermined by the oven method. Bothvarieties had an average moisture

content of about 14%. The collected ricesamples were shelled and milled/polished via a Satake rice sheller andpolisher. Milling time varied from 0 to

20 IRRN 1998

110 s, with an interval of 10 s to controlbran removal (degree of polish). Thedegree of polish ranged from 0 to 9.05%and 0 to 7.57% for Basmati and UPR79rice, respectively. The collected bran atthe end of each interval of milling wasused to extract the oil using Soxtec HT.

The table shows the effect of time ofmilling on the degree of polish and oilrecovery of bran. Oil recovery rangedfrom 15.70 to 20.91% for Basmati and14.57 to 17.90% for UPR79. The max-imum oil content was observed at 10 smilling time, for which the degree ofpolish was 3.33% for Basmati and2.75% for UPR79. For both varieties ofrice bran, the oil recovery starts increas-ing from 0 to 60 s time of milling sam-ples of bran and then decreases from 70to 110 s. The oil content of both varie-ties was maximum at 7.83 and 5.81%degree of polish for Basmati andUPR79, respectively. Weight of a ricegrain at the end of each time of millingis shown in the table.

Various mathematical models weretested for their suitability (least errorprediction criterion) to describe the oilrecovery (Or) of bran. The followingmodel was found to correlate the expe-rimental observations satisfactorily:

Or = aDp - bDp2

The coefficients a and b andstatistical parameters of the equationare given in the table. In view of thehigh r value and low associated error,the above model has been accepted andtested for validity in commercial ricemills. ■

Jaymati, a high-yielding sum-mer rice variety with good grainquality for Assam

T. Ahmed, K.C. Sarma, A.K. Pathak, H. Borah,K. Sharma, S. Hussain, and S. Ali, RegionalAgricultural Research Station (RARS) , AssamAgricultural University, Titabar 785630, Assam,India

Flooding causes a heavy decrease inwinter rice area, with a resulting

decline in production and productivityin the state of Assam. There is thereforescope for bringing chronically flood-affected and waterlogged winter riceareas under summer (boro) rice cultiva-tion. With the increase in irrigation faci-lities, summer rice area is increasingfast. But a suitable summer variety forflood-affected areas must be of160-170 d duration so that it can beharvested before the onset of floodingin mid-May.

We therefore developed Jaymati(TTB 103-2), a medium-tall (130 cm)rice variety from the cross Jaya/Mahsuri by the pedigree method ofbreeding. Jaymati is photoperiod-insensitive and suitable for transplant-ing in summer and autumn, and haswell-exserted panicles and a brownishhusk. Grains are awnless and mediumin size. The variety has a 1,000-grainweight of 20.2 g. Kernels are white,with a nonglutinous endosperm. The

length and width of the kernels are 6.34mm and 2.16 mm, respectively. Millingrecovery is 66.5%. Jaymati is moderate-ly resistant to bacterial blight, gallmidge, and stem borer.

The duration of Jaymati from seed toseed is 170 d for summer and 130 d forautumn and winter rice. The variety’sgood grain quality gives it an advantagefor growing autumn rice. Most autumnvarieties grown by Assam farmers havecoarse grains. Because farmers need aMahsuri type of grain for autumncultivation, some farmers grow Mahsuriin autumn. But because its longergrowth duration is not suitable in thatseason, Jaymati would be a better choice.

Jaymati was tested at several loca-tions throughout the state and it per-formed well as a summer crop (seetable). It was recommended for summercropping in the Central and Brahma-putra Valley zone of Assam by the statefor 1996. ■

Yield performance of Jaymati.

Yield (t ha-1)Location Year Season

Jaymati Jaya/local LSD (0.05)

RARS, Titabar1989 Autumn 3.4 3.0 0.701990 Autumn 4.8 3.5 1.001991 Autumn 3.3 3.9 1.251991 Winter 4.4 4.0 1.701991 Summer 3.2 2.2 0.30

National trialAICRIP (IET13316), 1991, winter

Chiplima – – 7.7 7.0 1.26Sindri – – 6.0 5.3 1.35Masodha – – 4.3 4.2 0.50Kanpur – – 3.7 2.9 0.26Raipur – – 2.8 2.8 0.72Titabar – – 4.4 4.0 1.72

Yield (t ha-1)Location Year Season

Jaymati Mahsuri Jagliboro LSD (0.05)

RARS, Shillongani, Nagaon1991-92 Summer 7.3 6.3 4.0 0.651992-93 Summer 6.9 6.2 4.0 0.521993-94 Summer 6.4 6.2 4.0 0.81

On-farm testing1994-95 Summer

Tokowbari – – 6.2 3.2 – –Dablongati – – 4.0 3.2 – –Bengenati – – 3.3 3.2 – –Bahuabheti – – 6.0 4.8 – –

Pooled average over season and yearAutumn 3.83 3.40Winter 4.76 4.31Summer 5.41 3.58Overall 4.88 3.84

Vol. 23, No. 2 21

Integrated germplasm improvement

Wide adaptability of new ricecultivars developed by theJapan-China collaborativeproject in Yunnan, China

K. Ise, Japan International Research Center forAgricultural Sciences (JIRCAS), Ohwashi 1-2,Tsukuba, Ibaraki, Japan; Sun Youquan, Liu Jishin,Zhou Tiande, and Jiang Zhinong, YunnanAcademy of Agricultural Sciences (YAAS),Kunming, Yunnan, China; S. Kudo, MountainousRegion Agricultural Research Institute, Inabu,Aichi, Japan; and Y. Sunohara, Fujisaka Branch ofthe Aomori Agricultural Experiment Station,Towada, Aomori, Japan

China’s Yunnan Province is consideredto be one of the centers of geneticdiversity in Asian cultivated rice. Acollaborative research project betweenJIRCAS and YAAS for lowlandjaponica rice breeding using a widediversity of genetic resources has beenconducted in Yunnan since 1982. Newrice cultivars developed through thejoint project have been grown widely inand around the province since 1989. In1997, the total area of these cultivarscovered more than 200,000 ha,accounting for more than 40% of thejaponica rice-growing area and about20% of the total rice-growing area inYunnan.

In 1996, two rice cultivars, Hexi 34and Hexi 35, were developed and re-leased by the project; they were official-ly registered in 1997 by the Yunnanprovincial government. Hexi 34 wasdeveloped from the cross Yunxi 2/Dianyu 1, and Hexi 35 from the crossHexi 15/Hexi 4. The pedigree of thesenew cultivars can be traced to someJapanese rice genetic resources: high-yielding cultivar Todorokiwase, blast-resistant germplasm accessions BL 1and BL 6, and others. Todorokiwase,especially, was the most usefulJapanese cultivar, with resistance tocool weather and blast resistance forimproving the grain quality and highyielding ability of Yunnan rice in thebreeding program.

The principal reason for releasingthese two cultivars is that they have ahigher yielding ability than currentjaponica rice cultivars grown inYunnan, which is the most importantfactor in crop production in China.Hexi 34 and Hexi 35 performed verywell in the uniform trials conducted at12 sites representative of japonica rice-growing areas in the central andnorthern parts of Yunnan (see table). In24 tests across 12 locations during a 2-yrperiod, the average yields of Hexi 34and Hexi 35 were 8.4 and 8.6 t ha-1,respectively, compared with 6.9 t ha-1

for the standard cultivar Yunkeng 9.We conducted a statistical analysis of

adaptation based on the data of theuniform trials using the linearregression method. The regressioncoefficients on the mean yield of eachenvironment for Hexi 34 and Hexi 35were smaller than those for other culti-vars tested (see figure). The findingsindicate that these cultivars with highyield potential are well adapted todifferent environments in Yunnan.

The cooking and processing quali-ties of Hexi 34 and Hexi 35 are superiorto those of standard cultivar Yunkeng 9in Yunnan. Milled kernels of Hexi 34and Hexi 35 are nonglutinous and non-aromatic; they are translucent in con-trast to those of Yunkeng 9, which showa pronounced white belly. Tastepanelists rated Hexi 34 and Hexi 35 as

satisfactory in sensory tests of steamedrice. The rapidly rising living standardin urban areas of China has resulted in aremarkable increase in demand for ricewith a good taste. These two new culti-vars will meet the demand for good-quality rice and could be widely grownin and around Yunnan Provincebecause of their high yielding abilityand high grain quality.

Finally, we should pay careful atten-tion to the shift in the frequency of blastfungus races. Hexi 34 and Hexi 35 exhi-bit a race-specific resistance to rice blastdisease, the most devastating disease inYunnan’s japonica rice-growing areas.

Agronomic performance of Hexi 34 and Hexi 35 in regional trials in Yunnan, China.

Character Hexi 34 Hexi 35 Yunkeng 9(standard)

Growth duration (d) 187 178 185Culm length (cm) 92 93 111Panicle length (cm) 19.0 17.8 16.4Panicles (no. m-2) 417 388 387Grains panicle-1 (no.) 109 118 131Panicle fertility (%) 79.3 81.4 71.5Single-grain weight (mg) 26.3 25.8 22.7Yielding ability (%) 123 126 100Lodging resistance HRa HR SCool-weather resistance MR MR MRBlast resistance HR M RGrain appearance Good Good PoorEating quality Good Good Poor

aHR = high resistance, MR = moderate resistance, M = moderate, S = susceptible.

Genotype mean (t ha-1)

Regression coefficient

Hexi line

Standard

Others

9.0

8.5

8.0

7.5

7.0

6.5

6.00.6 0.8 1.0 1.2 1.4

Cultivar performance by regressioncoefficient and mean yield in 24 environ-ments in Yunnan, China.

22 IRRN 1998

But the breakdown of blast resistance iscommon in many rice-growing areas,often shortly after the release of culti-

Integrated germplasm improvement — irrigated

vars with race-specific resistance. Weshould therefore develop breedingstrategies for durable resistance to

reduce the impact of rice blast diseaseby using the abundant rice geneticresources of Yunnan and Japan. ■

Karnataka rice hybrids

R.M. Radhakrishna, B.V. Chandra, S. Lingaraju,and S. Gangadhariah, Regional Research Station,VC Farm, Mandya 571405, Karnataka, India

For more than a decade, farmers in theirrigated area of Karnataka harvestedyields of up to 7 t ha-1. No new techno-logies could help them increase yieldfurther. Therefore, the major objectiveof the hybrid rice research begun atMandya was to identify a suitablehybrid that could yield at least 1 t morethan the check varieties. This objectivewas fulfilled by the release in 1994 ofKarnataka Rice Hybrid 1 (KRH1) andin 1996 of Karnataka Rice Hybrid 2(KRH2). The hybrid KRH1 (IR58025A/IR9761-19-1R) has a yield potential of7.5-8.5 t ha-1 and matures in 125 d. In150 on-farm irrigated trials conductedin farmers’ fields in different districts,KRH1 had a yield advantage of 1.5 tha-1 over the check varieties. Thishybrid possesses field tolerance formajor pests and diseases.

Medium-duration hybrid KRH2(IR58025A and KMR3) matures in 135 dand yields at least 1.0 t ha-1 more thanthe best check variety, Jaya. In addition,it has a better straw yield and exhibitstolerance for blast. Therefore, farmersare impressed by its performance. Thishybrid is semitall with long, slendergrains. In 15 other trials conducted atMandya from 1992 to 1995, KRH2produced an average of 9.3 t ha-1 with ayield advantage of 1.5 t ha-1 over Jaya.Similarly, multilocation, multiseasonalexperiments conducted at differentresearch stations in Karnataka in irriga-ted areas also confirmed the superiorityof KRH2, with at least 1.2 t ha-1 moreyield than Jaya. This hybrid was foundto be more stable over locations. KRH2recorded the highest yields in nationaltrials during 1995, with a yield

advantage of more than 0.9 t ha-1 overJaya.

Agronomic experiments with thesehybrids indicated that the recom-mended dose of 100-50-50 kg NPK ha-1

for semidwarf varieties is also suitablefor these hybrids. Planting one seedlinghill-1 and 20- × 10-cm spacing arerecommended for this hybrid. A seedrate of 20 kg ha-1 is sufficient. ■

TNRH16: a salt-tolerant rice hybrid

A.J. Ali, M. Rangaswamy, R. Rajagopalan, S.E.N.Mohamed, and T.S. Manickam, Crop Improve-ment Department, Agricultural College andResearch Institute, Navalur Kuttappattu,Tiruchirappalli 620009, India

Among abiotic stresses, salinity-alkalinity is the major constraint toyield in rice. Experience in other cropsindicates that hybrids perform betterthan varieties under adverse growingconditions. A study was therefore madeto evaluate hybrids involving salt-tolerant parental lines. Ten male sterilelines with their respective B lines and24 restorer lines were screened undernatural salt-stress conditions (pH 9.0

and EC 0.21 dS m-1). Of these,IR62829A, IR66707A, and IR58025Awere more tolerant. A similar responsewas recorded in their respective B lines.Of the 24 restorer lines, only six hadgood phenotypic acceptability at allstages of growth. Assuming that theadditive complementary effect oftolerant A and R lines in the hybridwould enhance the level of salttolerance, hybrids involving suchparental lines were evaluated for salttolerance. All F1 hybrids derived usingIR62829A and IR58025A as male sterilelines, though poor yielders, provedquite tolerant, with high phenotypicacceptability at all stages of growthfrom seedling to flowering. A highpollen or spikelet number during thelast phase possibly resulted in poorgrain yields. TNRH16, derived frommoderately salt-tolerant parents(IR58025A/C20R), was the onlyexception. This hybrid recorded a grainyield of 5,002 kg ha-1, whereas the salt-tolerant check, Co 43, recorded 4,160 kgha-1. The standard heterosis was 23%.The higher grain yield of TNRH16 mayalso be attributed to the fact thatparental characters for sodicitytolerance get complemented in the F1hybrid. ■

Pest resistance — diseases

Differentiation of rice tungrospherical virus variants by RT-PCR and RFLP

M.L.M. Yambao, P.Q. Cabauatan, and O. Azzam,IRRI

Rice tungro spherical virus (RTSV) isone of the two causal agents of ricetungro disease, the most importantviral disease of rice in South andSoutheast Asia. RTSV assists in thesemipersistent transmission of rice

tungro bacilliform virus (RTBV), theother causal agent, which causes thesymptoms.

RTSV is a single-stranded RNA virusof 12,180 nucleotides (Hull 1996).Genome organization shows that itencodes a large polyprotein of 393 kDa,which contains the three coat proteins(CP1, CP2, and CP3), motifs for nucleo-tide triphosphate (NTP) bindingdomain, protease (PRO), and poly-merase (POL). The polyprotein isthought to be cleaved by the virus-

Vol. 23, No. 2 23

and/or cell-encoded proteases. Thegenome also contains two short openreading frames at the 3' end. Recentstudies have demonstrated that ricecultivars react differently to RTSVvariants. TKM6, resistant to typevariant A, is susceptible to the RTSV-Vt6 variant (Cabauatan et al 1995).Discrimination between these twovariants cannot be achieved byserological means (Fig. 1).

To look for polymorphic molecularmarkers that will differentiate the twoRTSV variants, specific oligonucleo-tides were used to amplify the coatprotein gene fragments of RTSV byreverse transcriptase-polymerase chainreaction (RT-PCR). The CP1-2 regionwas then selected, amplified usingRSCP1V2453 and RSCP2C3607primers, and digested with differentrestriction enzymes. RSCP1Vcorresponds to nucleotides (nt) 2453 to2472 and RSCP2C corresponds to nt3548 to 3607 in the RTSV-A RNAnucleotide sequence.

Briefly, the first-strand cDNAsynthesis was performed using 0.1 ngof purified virus or 1 µg of total RNAextracts and RSCP2C primer. Themixture was denatured at 70 °C for 10min, chilled on ice for at least 1 min,and after the adjustment to 200 mMTris-HCl (pH 8.4), 500 mM KCl, 25 mMMgCl2, 10 mM dNTP, and 0.1 M DTT in20 µl total volume, the reaction wasincubated at 42 °C for 5 min. Fifty U of

SuperScript II reverse transcriptase(Gibco BRL) were added and thereaction proceeded at 42 °C for 50 min.The reaction was terminated at 70 °Cfor 15 min followed by a chill on ice.The first-strand cDNA was amplifiedusing PCR mixtures containing 200mM Tris-HCl (pH 8.4), 500 mM KCl, 25mM MgCl2, 10 mM dNTP, a mix ofRSCP1V and RSCP2C primers, 5 U ofTaq polymerase, and the cDNA tomake a total volume of 50 µL, and wasoverlaid with mineral oil. Reactionmixtures were heated at 95 °C for 1 min,95 °C for 1 min, 50 °C for 1 min, and68 °C for 5 min for 30 cycles, and at72 °C, for 7 min as a final extension.

PCR aliquots were analyzed in 1.5%agarose gels, using 45 mM Tris-borate,pH 8.0, 1 mM EDTA (TBE) as electro-phoresis buffer. Five µL of the productwere restricted with HindIII or BstYI(XhoII) at 37 °C for 16 h in a finalvolume of 50 µL in the buffer suppliedby New England Biolabs Co. Restric-tion fragments were also observed on1.5% agarose gels.

Using the RT-PCR technique, thethree coat protein regions of both RTSVvariants were amplified (Fig. 2). Theidentity of the PCR products wasconfirmed by Southern blot hybrid-ization using the RTSV-specific CP2-3probe (a gift from Dr. R. Hull, JIC,England). Digestion of the RT-PCRproducts from the CP1-2 region usingHindIII and BstY1 showed a distinct

banding pattern between the two RTSVvariants (Fig. 3). For RTSV-A,restriction digestion with HindIIIproduced two fragments of about 600bp each and with BstYI two fragmentsof about 800 bp and 300 bp.

These patterns were identical tothose expected from the publishedsequence of the RTSV CP1-2 region:two fragments of 579 bp for HindIIIdigestion and two fragments of 807 bpand 284 bp for BstYI digestion (Shen etal 1993).

For RTSV-Vt6, restriction digestionwith HindIII did not produce any sizechange in the PCR product, but with

1. Western blot of coat proteins (CP) 1, 2,and 3 of RTSV strains A and Vt6. CPs wereseparated by SDS-PAGE, transferred to anylon membrane, and incubated with poly-clonal antiserum and antiserum specific toeach CP. The size of the molecular weightmarkers (in kiloDaltons) is indicated on theleft. H = healthy extract.

2. Amplification of coat protein regions of RTSV variants by reverse transcriptase-polymerasechain reaction (RT-PCR). (A) 1.5% agarose gel electrophoresis analysis of RT-PCR amplified cDNAusing CP1 primers (lane 2), CP2 primers (lane 3), CP3 primers (lane 4), CP1-2 (lane 5), CP2-3(lane 6), and CP1-3 (lane 7). (B) Autoradiograph of the same gel when blotted onto a nylonmembrane and hybridized with an RTSV CP2-3 specific probe. Lane 1, high marker. U = undi-gested PCR product, H = HindIII digested PCR product, B = BstY1 digested product.

3. RFLP differences between the RT-PCRproducts of RTSV variants when digestedwith HindIII or BstY1. Lane 1, lowbiomarker; 2, undigested RTSV A (U); 3,HindIII digested RTSV A (H); 4, BstY1digested RTSV A (B); 5, undigested RTSV Vt6(U); 6, HindIII digested RTSV Vt6 (H); 7,BstY1 digested RTSV Vt6 (B).

2 3 4

RTSV-A RTSV-V16

5 6 71U H B U H B

1000

700500400300200

A B

300015001000700500400300200

1 2 3 4 5 6 7 2 3 4 5 6 7 2 3 4 5 6 71 2 3 4 5 6 7

RTSV-A RTSV-V16 RTSV-A RTSV-V16

24 IRRN 1998

BstYI, two fragments of about 700 bpand 300 bp were observed. Sequenceanalysis of the CP1-2 region fromRTSV-Vt6 (unpublished results)confirmed the absence of the HindIIIsite and the presence of three fragmentsafter BstYI digestion: 707 bp, 284 bp,and 108 bp. Variation between the twoviral strains was confirmed at ntpositions 2556 and 3032 based on thepublished sequence. The RT-PCRmethod can therefore be used to studyRTSV coat protein variation in naturalfield populations.

ReferencesCabauatan PQ, Cabunagan RC, Kogane-

zawa K. 1995. Biological variants of ricetungro viruses in the Philippines.Phytopathology 85:77-81.

Hull R. 1996. Molecular biology of ricetungro viruses. Annu. Rev. Phytopathol.34:275-297.

Shen P, Kaniewska M, Smith C, Beachy RN.1993. Nucleotide sequence and genomicorganization of rice tungro sphericalvirus. Virology 193:621-630. ■

Mechanism of resistance to ricetungro spherical virus (RTSV)

P.Q. Cabauatan and O. Azzam, IRRI

RTSV is one of two viruses that causetungro disease. RTSV is independentlytransmitted, whereas the other virus,rice tungro bacilliform virus (RTBV), isdependent on RTSV for its transmis-sion by the green leafhopper (GLH),Nephotettix virescens. The occurrenceand spread of tungro disease thereforedepend on the presence of RTSV in thefield. Resistance to RTSV infectionwould slow down the spread of thedisease.

One of the most important compo-nents of a tungro management strategyhas been the use of resistant varieties.After a few years of intensive cultiva-tion, however, these varieties succumbto tungro infection. Changes in fieldresponse to tungro infection have been

attributed to a shift in GLH virulence tothe varieties over time as shown byexperiments conducted under naturaland artificial conditions (Dahal 1988,Dahal et al 1990, Cabunagan and Ling1982). Although these studies indicatethat the mecha-nism of resistance totungro might be vector-dependent,other results showed a differentphenomenon. For example, GLHselected on Pankhari 203, TAPL 796,and IR20 for 9-19 generations did notcause a significant increase in RTSVtransmission to these varieties(Heinrich and Rapusas 1984, Dahal1988, Hibino et al 1988). Hence, themechanism of resistance to tungro maynot be vector-dependent only and GLHadaptation may not be the only factorinvolved in resistance breakdown.

In 1995, a new strain of RTSV, desig-nated as Vt6, was isolated from ricevariety TKM6 in Midsayap, NorthCotabato, Philippines (Cabauatan et al1995). TKM6 is highly resistant to theIRRI strain of RTSV (strain A) and is thecommon parent of IR20, IR26, IR30, andIR40. These varieties have moderateresistance to GLH and have shownconsistent resistance to RTSV since theywere released in the 1970s (Hibino et al1988). We used IR26 to study the role ofGLH adaptation and strain variation inthe mechanism of resistance to RTSVand found that resistance to RTSV isboth vector- and virus-dependent.GLHs were collected from a field inPolangui, Albay, Philippines, andreared on rice variety TN1 for twogenerations. The progenies were then

tested for their transmission of RTSVstrains A and Vt6 to TN1 (RTSV-susceptible) and IR26 (RTSV-resistant).Afterwards, the GLH colony wasdivided into two; one-half was rearedon TN1 (GLH-susceptible) and theother half on IR26 (GLH-resistant) for11 generations. The GLH colonies werethen tested again for RTSV straintransmission on both TN1 and IR26.Transmission of RTSV strains A and Vt6by each GLH colony was compared onboth varieties before and after selection.RTSV transmission efficiency ofselected colonies was also tested on ricevarieties TKM6 and Adday Sel.,varieties with known resistance toRTSV. All transmission tests wereconducted in test tubes at 1 insectseedling-1 for an overnight inoculationaccess. RTSV was detected ininoculated plants by the enzyme-linked immunosorbent assay (ELISA).

After 11 generations, the GLHcolony on IR26 reproduced as well asthe one on TN1. It was reported that abiotype of GLH could be selected bybeing reared on a resistant rice variety(Kobayashi et al 1983); therefore, theGLH colony on IR26 can be consideredto be adapted to its host. Transmissiontests showed that the reaction of IR26 toRTSV-Vt6 changed from resistant (23%infection rate before selection) tosusceptible (62.8% infection rate after11 generations on IR26), while thereaction of IR26 to RTSV-A remainedresistant (0% infection rate) (Table 1).

These results indicate that theresistance of IR26 to RTSV is both

Table 1. Transmission (%) of RTSV strains A and Vt6 to IR26 and TN1 by field-collected GLH before selectionon IR26, TN1, TKM6, and Adday Sel. and after selection for 11 generations.

Variety Before selectionb After selection on

TN1 IR26

AC Vt6c A Vt6 A Vt6

TN1 97 78 82 82 59 82IR26 0 23 0 29 0 63TKM6 – –d 0 59 0 31Adday Sel. – – 0 0 0 0

aOne insect seedling–1, overnight inoculation access; 40 seedlings treatment–1. bField-collected (Bicol) GLH was reared on TN1 fortwo generations and adult progenies were tested for transmission of RTSV A and Vt6 before selection on IR26. cTested by ELISA 3wk after inoculation. dNot tested.

Vol. 23, No. 2 25

vector- and virus-dependent. In thepresence of a resistance-breakingstrain, such as Vt6, a shift in GLHvirulence (adaptation) results in abreakdown of resistance. On the otherhand, the resistance of IR26 did notbreak down in the presence of RTSV-A.Selection of GLH on this variety for 11generations did not result in increasedRTSV-A infection. TN1 is susceptible toboth virus strains regardless of theGLH colony used for inoculation.TKM6 is a differential host for bothstrains. Although the selected coloniesdid not transmit RTSV-A, they trans-mitted RTSV-Vt6 at varying efficiency,depending on the GLH colony used.Variety Adday Sel. (IRGC Acc. No. 180)was highly resistant to both strainsregardless of the GLH colony used.Table 2 summarizes the mechanism ofresistance of IR26 to RTSV.

This study demonstrated thatbreakdown of resistance to tungrodisease can be both virus- and vector-dependent. This study also showedthat some rice cultivars have “true”

Table 2. Mechanism of resistance of IR26 to RTSV.

Virus strain GLH colony Reaction Mechanism of resistance

RTSV-A Nonadapted Resistant R to both virus and vectorRTSV-A Adapted Resistant R to virusRTSV-Vt6 Nonadapted Resistant R to vectorRTSV-Vt6 Adapted Susceptible

resistance, that is, race-specificresistance, to the virus strain(s) asexhibited by IR26 and TKM6 againstRTSV-A and Adday Sel. against bothRTSV strains. This information shouldbe considered when formulatingeffective control strategies againsttungro.

ReferencesCabauatan PQ, Cabunagan RC, Koganezawa K.

1995. Biological variants of rice tungroviruses in the Philippines. Phytopathology85:77-81.

Cabunagan RC, Ling KC. 1982. Resistance totungro: a case of IR34 variety. Philipp.Phytopathol. 18:18. (abstr.)

Dahal G. 1988. Transmission of tungro-associated viruses by field and selectedcolonies of Nephotettix virescens (Distant) and

their mode of feeding on selected cultivars.Ph D thesis, University of the Philippines LosBaños, Laguna, Philippines. 139 p.

Dahal G, Hibino H, Cabunagan RC, Tiongco ER,Flores ZM, Aguiero VM. 1990. Changes incultivar reaction due to changes in virulenceof the leafhopper vector. Phytopathology80:659-665.

Heinrich EA, Rapusas HR. 1984. Feeding,development, and tungro transmission bythe green leafhopper, Nephotettix virescens(Distant) (Homoptera: Cicadelidae) afterselection of resistant rice cultivars. Environ.Entomol. 13:1074-1078.

Hibino H, Daquioag RD, Cabauatan PQ, DahalG. 1988. Resistance to rice tungro sphericalvirus in rice. Plant Dis. 72:893-847.

Kobayashi A, Supaad A, Othman O. 1983.Inheritance of resistance of rice to tungro andbiotype selection of green leafhopper inMalaysia. JARQ 16:307-311. ■

Leaf number: a reliable param-eter for determining seedingintervals between parental linesin hybrid rice seed production

B.C. Viraktamath, C.H.M. Vijayakumar, M.I.Ahmed, S. Singh, and M.S. Ramesha, Hybrid RiceLaboratory, Directorate of Rice Research (DRR),Rajendranagar, Hyderabad 500030, AndhraPradesh, India

Because parental lines of rice hybridsusually differ in their growth duration,obtaining well-synchronized floweringis a major problem in hybrid rice seedproduction. The present method ofstaggered sowing of parental linesbased on their growth durationdifference, though simple, is not alwayseffective, especially in areas wheretemperature changes are frequent.Reports from China indicate thatbecause the leaf number of a rice

cultivar is fairly stable, leaf numberdifference can be used to decide theseeding dates of parental lines inhybrid rice seed production. Beforeusing this method, however, the leafnumber of the prospective parentallines to be used in hybrid rice seedproduction should be ascertained.We conducted an experiment todetermine the leaf number of twocytoplasmic male sterile (CMS) linesand five promising restorers during the1994 wet season (WS), 1995 WS, and1995 dry season (DS) at the DRRresearch farm. We also compared theextent of variation for leaf number anddays to flowering over the seasons.

Seedlings were raised in wet beds,10 seedlings were marked, and leafcounting started from the tenth day.After 25 d, the same seedlings weretransplanted to the main field andleaves were counted at intervals of 5 d

until the opening of the flag leaf.Incomplete leaves were numbered bycomparing their length with theprevious leaf, using a 10-point scale(see figure).

The table presents data for leafnumber and days to 50% flowering.Leaf number ranged from 15.3(IR58025A) to 18.5 (Vajram). Thecoefficient of variation for leaf numberover seasons varied from 0.65 to 1.89and a wide range of variation wasobserved for growth duration (CV 6.23-12.53). Variation for leaf numberbetween WS and DS was negligible,whereas growth duration variedwidely between the two seasons. Inview of the relative stability of leafnumber over seasons, this parametercan be used to determine the seedingintervals between parental lines toobtain better synchronization inflowering.

Seed technology

26 IRRN 1998

Adapting hybrid rice seedproduction technology

A.S. Ponnuswamy, M. Rangswamy, P. Ranga-swamy, and K. Thiyagarajan, Hybrid Rice Scheme,School of Genetics, Tamil Nadu AgriculturalUniversity, Coimbatore 641003, India

The success of hybrid rice cultivation inIndia depends on a successful seedproduction program. Optimizing rowratio, determining the economic dose of

Leaf number difference between twoparental lines can be used to determinethe seeding interval when the leafgrowth rates (leaf growth rate =number of days taken to produce thefirst 7 leaves divided by 7) of theparental lines are the same. Forexample, the leaf growth rate of CMSline IR58025A and restorer IR40750-82-2-2-3R is the same (6.4) and the leafnumber difference between them is 1.3during the DS. Therefore, for seedproduction of hybrid IR58025A/IR40750-82-2-2-3R during the DS, themale parent is sown first and the femaleparent is sown later, when the earliersown male parent has produced 1.3leaves.

In cases where parents differ in theirleaf growth rate, the seeding interval isdetermined by using the followingformula: seeding interval (d) = leafnumber difference × leaf growth rate ofthe early sown parent. ■

GA3 and its stage of application, andidentifying a suitable substitute for GA3are required for hybrid rice seed produc-tion.

An experimental study revealed thatthe row ratio of female to male of 6:1recorded the highest hybrid seed yield of1,921 kg ha-1. GA3 application increasedplant height, panicle exsertion, flag leafangle, seed setting percentage, and seedyield. Comparatively, the proportionateincrease in seed set and yield was high

when applying GA3 at 125 g ha-1. ForGA3 application, the 15-20% panicleexsertion stage was found to be idealfor obtaining maximum seed set andseed yield.

Besides GA3, applying a 2% foliarspray of juvenile leaf extract of Albiziaamara and 2% urea spray enhancedseed yield by increasing panicleexsertion and seed set in thecytoplasmic male sterile line. ■

3rd leaf

4th leaf

2nd leaf

1st leaf 1st leafColeoptile Coleoptile 1st leaf

Coleoptile

2nd leaf 2nd leaf

3rd leaf3rd leaf

4th leaf4th leaf

5th leaf

Leaf no. 3.3 Leaf no. 3.7 Leaf no. 4.4

Rice seedlings showing different leaf numbers.

Leaf number and growth duration of parental lines of promising hybrids at Hyderabad, Indiaa.

Leaf number Days to 50% floweringParental line

1994 WS 1995 WS 1995 DS Mean CV 1994 WS 1995 WS 1995 DS Mean CV

IR58025A 15.7 15.3 15.0 15.3 1.83 103 98 113 104.6 6.23IR62829A 16.2 16.0 15.9 16.0 0.81 96 89 106 100.0 7.59Vajram 18.3 18.6 18.6 18.5 0.75 112 119 144 125.0 10.98IR10198-66-2R 15.6 15.4 15.0 15.3 1.64 90 87 101 92.6 6.49IR40750-82-2-2-3R 16.1 16.1 16.3 16.2 0.61 111 102 125 112.6 8.40IR54742-22-19-3R 18.5 18.0 18.8 18.4 1.89 111 109 142 120.6 12.53IR29723-143-3-2-1R 17.1 17.8 17.3 17.4 1.69 119 116 143 126.0 9.58

aWS = wet season, DS = dry season, CV = coefficient of variation.

Vol. 23, No. 2 27

Analyzing leaf number inparental lines for hybrid

rice seed production

C.H.M. Vijayakumar, M.S. Ramesha, B.C. Virak-tamath, S. Singh, and M. Ilyas-Ahmed, Directo-rate of Rice Research, Rajendranagar, Hyderabad500030, India

Determining the seeding intervalbetween two parents in any hybridcombination at a given location is ofprime importance for achieving perfectsynchronization in a hybrid rice seedproduction plot. Fixed growth dura-tion, commonly practiced in India, is asimple and easy method to adopt. Butfluctuations in temperature and otherweather conditions influence maturity.Leaf number, unlike growth durationof a rice variety, is reported to be rela-tively stable over seasons and years.

We studied leaf number and growthduration in promising parental lines

and their utility in hybrid rice seedproduction. The experimental materialincluded male sterile lines (IR58025Aand IR62829A), restorer lines (IR10198R,IR40750R, IR54742R, and IR29723R), andinbred variety Vajram. They were grownin the field in the 1994-95 wet season(WS) and dry season (DS). A three-division method of counting leaves wasadopted; the values for developingleaves are given as 0.2, 0.5, and 0.8 forjust-emerged, half-emerged, and almost-opened leaf, respectively, keeping thefully opened previous leaf as a reference.Besides growth duration, observationswere recorded on leaf number by using12 competitive plants in each entry andcounting leaves at an interval of 5 d fromthe 3rd leaf stage until the flag leaf stage.

Results revealed that growth dura-tion fluctuated and depended on seasonand year. This fluctuation was less(3-14 d) in cytoplasmic male sterile(CMS) and restorer lines of medium

duration—IR58025A, IR62829A,IR10198R, and IR40750R—comparedwith late-maturing restorers (3-29 d).The leaf number of a variety or A and Blines did not vary over seasons oryears. The rate of leaf emergencedepended on season and stage of plantgrowth. Leaf emergence is faster in theWS (5 d to emerge) than in the DS (7-8d). During the growth period, the rateof leaf emergence declined after 65 and75 d after sowing in the WS and DS,respectively. In general, early-durationvarieties showed a lower number ofleaves than late-maturing varieties.

We have not found any appreciabledifference among CMS lines and earlyand late restorers for their rate of leafemergence. Based on this study, wesuggest adopting 3-4 staggeredsowing of parental lines at an intervalof 5 d so that leaf number index can beused as an aid in hybrid rice seedproduction. ■

Identifying somefavorable environments

for hybrid rice seed production inAndhra Pradesh, India

R.V. Kumar, M.S. Rao, and P.V. Satyanarayana,Andhra Pradesh Agricultural University,Agricultural Research Station, Maruteru 534122,India

Outcrossing in rice mainly depends onclimatic factors. Favorable environ-ments increase hybrid seed production.Likewise, synchrony in flowering ofmale and female parents also increasesseed set. We therefore conducted twotrials to investigate these two importantaspects. A trial was conducted insemiarid and humid zones of AndhraPradesh during the 1991-92 wet season(WS) and dry season (DS), usingIR62829A and IR62829B in a row ratioof 4A:2B. Sowings of A and B lines werestaggered. Leaf clipping and ropepulling were adopted. The outcrossingpercentage was higher at Palem andWarangal, located in the semiarid zone(>40%), than at Maruteru, located in the

humid zone (<20%) in both seasons.Seed yield was also higher in the semi-arid regions in the 1991 WS (Palem, 959kg ha-1; Warangal, 1,191 kg ha-1) and 1992DS (Palem, 1,527 kg ha-1) than in thehumid zone (Maruteru, WS 408 kg ha-1

and DS 851 kg ha-1).To achieve flowering synchronization

in the seed production plots, preciseinformation on the phenologicalbehavior of the parental lines is essentialin the target environments. With thisobjective, a second trial was conductedduring the 1995 WS at seven locationsrepresenting four different agroclimaticzones in Andhra Pradesh. The materialfor this study included parental lines (Aand R lines) of released and prereleasehybrids. In different agroclimatic zones,from location to location, in all threeCMS lines and five restorer lines,flowering duration varied: 81-105 d inIR58025A, 77-105 d in IR62829A, and 90-155 d in PMS 3A. A similar trend wasalso observed in the restorer lines. Thedifferences in flowering duration in eachR line varied from 26 to 34 d in differentagroclimatic zones. ■

Seasonal influence offlowering behavior and

plant growth characters onparental lines of hybrid rice

S.R. Prabagaran, Center for Biotechnology,Pondicherry University, Pondicherry 605104; andA.S. Ponnuswamy, Seed Technology Department,Tamil Nadu Agricultural University, Coimbatore,India

Parental lines of hybrid rice—A lines(IR62829A and IR58025A), B lines(IR62829B and IR58025B), and R lines(IR10198-66-2R, AS 89044, and Pusa150R)—were raised at monthlyintervals from August 1994 to July 1995.The recommended cultivation prac-tices were adopted and observationswere made in 10 randomly selectedplants from each population. Duringthe cropping period, data on maximumand minimum temperature and hoursof sunshine and daylength wereobtained from agrometeorologicaldocumentation records. From this, heatunit concepts—growing degree days(GDD), relative temperature disparity

28 IRRN 1998

(RTD), photothermal units (PTU), andheliothermal units (HTU)—werederived for correlation and regressionstudies. Variations were observed indifferent parameters, not only becauseof genotype characters and seasonalinfluence but also because of theinteraction between them. Most floralcharacters were better expressed in thesummer season (February to May) thanin the samba season (October toJanuary).

Among the parental lines, the A linesrecorded higher values for days to firstflowering, 50% flowering, duration offlowering in a panicle, period ofanthesis, stigma exsertion, andproductive tillers than their respectivemaintainers. Correlation studies withdays to first flowering, duration offlowering in a panicle, and panicleexsertion proved significant in all theparents except AS 89044. All the heatunit concepts showed a significantcorrelation for IR10198-66-2R. Amongthe heat unit concepts, HTU played alimited role in altering floral characters.Regression equations were fitted topredict the flowering behavior of theseparental lines. These studies will beuseful in identifying parental lines withstability against seasonal influences.Staggered sowing of B/R lines basedon the season of seed production can bedone based on such results. ■

Determining seedingintervals of parents in

hybrid rice seed production

A.H. Krishnamurthy, B.V. Chandra,R.M. Radhakrishna, and S. Lingaraju, RegionalResearch Station, VC Farm, Mandya 571405,Karnataka, India

For hybrid seed production, twoparents—a cytoplasmic male sterile(CMS) line (female) and restorer (male)lines—are planted in alternate rows.Similarly, to multiply CMS lines, theCMS and maintainer lines are plantedin alternate rows. Simultaneousflowering in both male and female linesis highly essential for seed set.

Flowering duration in the male andfemale lines differs, and, therefore,there will not be any synchrony inflowering of the male and female linesif they are sown on the same day.Parental lines differing in their growthduration can be sown on different daysin nursery beds so that they can reachflowering at the same time in the mainfield. To determine the seeding intervalof the parents of KRH2 and CMS lineIR58025A, studies were conducted inthe 1995 wet season. IR58025Aflowered in 90 d whereas its maintainer,IR58025B, flowered in 85 d. KMR-3R,the restorer parent of KRH2, floweredin 98 d. Therefore, the growth durationdifference between IR58025A andKMR-3R was 8 d. But maintainer lineIR58025B flowered 5 d earlier than theA line. The leaves on the main culm ineach of the parents were counted. Theparents, IR58025A, IR58025B, andKMR-3, produced 15, 14, and 16 leavesat flowering, respectively.

The results indicated that IR58025Bshould be sown 5 d after sowingIR58025A or when IR58025A producesone leaf in the nursery for CMSmultiplication. For KRH2 hybrid seedproduction, IR58025A should be sown8 d after sowing restorer line KMR-3Ror when the restorer line produces 1.4leaves in the nursery. ■

Adjusting flowering of aCMS line in hybrid rice

seed production

S. Lingaraju, B.V. Chandra, V. Bhaskar, andS. Gangadhariah, Regional Research Station, VCFarm, Mandya 571405, Karnataka, India

Seed production of hybrids, unlike thatof inbred varieties, involves difficultprocedures. In hybrid rice seed produc-tion, two parents—a cytoplasmic malesterile (CMS) line (female) and restorerline (male)—are grown in alternatefixed row ratios, one after the other. Thefemale line, which is male sterile,receives pollen from the male, which isfertile and set the seed. Therefore, it is

essential that both male and femaleparents flower at the same time. Tech-niques such as the leaf count methodand growth duration differencemethod are used for the differentialseeding of male and female parents.Flowering in male and female parentsoften fails to synchronize because ofenvironmental conditions. In China,where hybrid seed is produced on alarge scale, several techniques havebeen adopted to synchronize floweringat the early stages of panicle develop-ment.

Because seed growers in Karnatakafound it difficult to identify the earlystages of panicle development, weattempted to develop simple tech-niques that could be adopted at thelater stages of panicle development (atthe half and full boot leaf stage).IR58025A was the common CMS lineused for different hybrids in Karnataka.Adjustment of flowering in this linewas studied during the 1995 wetseason. Techniques such as sprayinggibberellic acid (GA3) and urea andapplying potash, phosphorus, and ureato the base of plants were tried.

The results indicated that the abovetreatments at the half boot leaf stagehad no effect on flowering. At the fullboot leaf stage, however, a GA3 spray at60 ppm advanced 50% flowering by 3 dand full flowering by 5 d. The totalduration of flowering was reduced by5 d. Applying urea at 20 kg ha-1 to thebase of plants at the full boot leaf stagedelayed both 50 and 100% flowering by2 d. Similarly, the total duration offlowering also increased by 2 d.Applying phosphorus at the rate of 20kg ha-1 at the full boot leaf stage delayed50 and 100% flowering by 2 d, butflowering duration decreased by 3 d.Therefore, by spraying GA3 at 60 ppmat the full boot leaf stage, we canadvance flowering in CMS lineIR58025A. With an application of ureaor phosphorus at 20 kg ha-1, floweringcan be delayed, when necessary, by 2 din CMS line IR58025A. This method canbe used to synchronize floweringadequately. ■

Vol. 23, No. 2 29

Stress tolerance — adverse soils

Effective amount of N fertilizerfor direct seeding on wet surfaceof reclaimed saline soil in Korea

K.S. Lee, J.H. Jun, and H.T. Shin, HonamNational Agricultural Experiment Station, Iksan570080, Korea

The demand for changing culturalmethods from machine transplantingto direct seeding is a current trend inKorea. We identified the appropriateamount of N fertilizer for direct seedingon a wet surface of reclaimed saline soilin Korea.

The experiment was conducted atthe Gyehwa substation of the HonamNational Agricultural ExperimentStation during 1996-97 using a ran-domized complete block design withthree replications on saline soil thatcontained 0.2-0.4% NaCl in soilsolution. Four N levels were testedwith a control of 210 kg N ha-1 asstandard fertilization for machinetransplanting. The results showed thatseedling establishment increasedslightly up to 180 kg N ha-1 butdecreased at 210 kg N ha-1 in directseeding. Culm length is higher in directseeding with 210 kg N ha-1. Fieldlodging in the maturing stage is moreserious in direct seeding than in trans-planting, as noted by the score of 5 outof 9 under visual assessment (Table 1).

Panicle number and spikeletnumber per unit area increased signifi-cantly at above 180 kg N ha-1 comparedwith machine transplanting at 210 kg Nha-1 as a control. The percentage ofripened grains and 1,000-grain weightwere higher for machine transplantingthan direct seeding at the same amountof N. Milled rice yield showed no signi-ficant differences between direct seed-ing and machine transplanting at thesame amount of N (Table 2). As a result,we found that the appropriate N levelfor direct seeding in reclaimed salinesoil was 180 kg ha-1. This amount willbest prevent field lodging and save onfertilizer and labor costs. ■

Table 1. General characteristics of rice as influenced by N fertilizer under wet seeding in reclaimed saline soil,Gyehwa, Korea, 1996-97.

Cultural method N Seedling Heading date Culm length (cm) Field lodginga

amount establishment(kg ha-1) (no. m–2)

Direct seeding 100 100 12 Aug 79 0150 110 12 Aug 81 1180 115 13 Aug 83 3210 99 14 Aug 84 5

Machine transplanting 210 _ 16 Aug 80 3(standard)

aOn a scale of 0-9, where 0 = none, 1 = low, and 9 = high.

Table 2. Yield components and yield according to amount of N fertilizer under wet seeding in reclaimed salinesoil, Gyehwa, Korea, 1996-97a.

Cultural N Panicles Spikelets Ripened 1,000- Milled Yieldmethod amount (no. m-2) grain grain rice yield index

(kg ha-1) (no. m-2) × 103 (%) weight (g) (t ha–1)

Direct seeding 120 375 b 26.6 c 85.1 b 24.6 a 4.6 b 91150 381 b 28.8 b 85.3 b 24.5 a 4.7 b 93180 397 a 30.9 a 84.1 bc 24.9 a 4.8 a 96210 401 a 31.9 a 83.2 c 23.8 b 4.7 ab 94

Machine 210 346 c 27.2 bc 89.7 a 25.3 a 5.1 a 100transplanting(standard)

aMeans followed by a common letter are not significantly different at the 5% level by Duncan’s multiple range test.

Identifying optimum seedingtime for direct seeding on a wetfield surface in reclaimed salinesoil in Korea

K.S. Lee and J.H. Jun, Honam National Agricul-tural Experiment Station, Iksan 570080, Korea

Rice cultivation in Korea is changingfrom machine transplanting to directseeding to save on labor costs and toincrease profits. We therefore identifiedthe appropriate seeding time for directseeding on a wet field surface in

reclaimed saline soil in the southernpart of Korea.

The experiment was conducted atthe Gyehwa substation of the HonamNational Agricultural ExperimentStation during 1996-97 using a split-plot design with three replications in asaline soil that contained 0.2-0.4% NaClin soil solution. Three varieties wereused to represent each maturing type:Shinunbongbyeo (early maturing),Gancheokbyeo (mid-maturing), andGyehwabyeo (late maturing). Theresults showed that for Shinunbong-

Table 1. Seedling establishment (no. m-2) according to seeding times on a wet field surface in reclaimed salinesoila, Gyehwa, Korea, 1996-97.

Seeding time Shinunbongbyeo Gancheokbyeo Gyehwabyeo

10 May 103 a 104 a 108 a20 May 97 a 109 a 107 a30 May 92 a 98 a 93 a10 Jun 89 a 79 b 76 b

aMeans followed by a common letter are not significantly different at the 5% level by Duncan’s multiple range test.

30 IRRN 1998

byeo, seeding time had no effect onseeding establishment and for Gan-cheokbyeo and Gyehwabyeo, seedingestablishment adversely affected themif they were seeded after June (Table 1).

Panicle number per unit area forShinunbongbyeo was not significantlydifferent for seeding time; for Gan-cheokbyeo and Gyehwabyeo, paniclenumber per unit area decreased if theywere seeded after 10 Jun. For 1,000-grain weight, seeding time had noeffect on all three varieties.

Spikelet number per unit area andpercentage of ripened grains werereduced significantly when the seedingtime was 10 Jun for all maturing types.Milled rice yield for Shinunbongbyeowas not affected by seeding time. ForGancheokbyeo and Gyehwabyeo,milling yield was reduced significantly

Table 2. Yield components and yield according to seeding times on wet field surface in reclaimed saline soila,Gyehwa, Korea, 1996-97.

Cultivar Seeding Panicles Spikelets Ripened 1,000- Milledtime (no. m-2) (no. m-2) × 104 grain grain rice yield

(%) weight (g) (t ha-1)

Shinunbongbyeo 10 May 417 a 30.7 a 89 a 21.7 a 4.5 a20 May 414 a 30.6 a 89 a 21.4 a 4.4 a30 May 406 a 28.3 ab 87 ab 21.5 a 4.3 a10 Jun 399 a 27.2 b 86 b 21.3 a 4.1 a

Gancheokbyeo 10 May 411 a 31.6 a 90 a 22.7 a 4.5 a20 May 405 a 32.9 a 92 a 23.0 a 4.6 a30 May 397 a 31.5 a 92 a 23.1 a 4.6 a10 Jun 361 b 25.3 b 86 b 22.2 a 3.8 b

Gyehwabyeo 10 May 421 a 29.3 a 90 a 24.0 a 4.5 a20 May 412 a 31.0 a 90 a 24.3 a 4.8 a30 May 398 a 28.4 ab 87 b 23.9 a 4.4 b10 Jun 355 b 24.4 b 84 c 23.1 a 3.4 c

aMeans followed by a common letter are not significantly different at the 5% level by Duncan's multiple range test.

when the seeding time was late Mayand later (Table 2). The results indicatedthat the best seeding time for early-maturing types was from May to early

June. The most appropriate seedingtime for mid- and late-maturingvarieties was the middle of May. ■

Stress tolerance — excess waterDistinguishing seedling traits indeepwater rice (Oryza sativa L.)

P.M. Mohapatra and A.R. Panda, Division ofGenetics and Plant Breeding, Central RiceResearch Institute, Cuttack 753006, Orissa, India

Deepwater rice genotypes are charac-terized by their elongation abilityunder excess water. This character isexpressed approximately 4-6 wk afterseeding. However, young germinatingseedlings within 7-15 d after seedingcan be identified for elongation abilityfrom morphological characters such asmesocotyl elongation and length of thesecond leaf blade, which can be used todistinguish deepwater types from theirnondeepwater counterparts.

A comparative assessment wasmade for these two parameters taking10 genotypes from both deepwater andnondeepwater rice during the dry andwet seasons, 1996, at the Central RiceResearch Institute, Cuttack. Seedingwas done at three depths—4, 6, and 8cm—under normal pot culture as wellas under dark with seeds germinated inpetri dishes. Observations wererecorded for mesocotyl elongation andlength of the second leaf blade on 1-wk-

Length of mesocotyl and second leaf blade of deepwater and nondeepwater rice varieties (data pooled over twoseasons).

Mesocotyl length (mm) Length ofsecond

Variety Seeding depth Germination leaf bladein dark (mm)

4 cm 6 cm 8 cm

Deepwater riceJalamagna 26.15 37.10 53.40 3.75 37.40TCA4 28.75 42.40 63.35 5.20 38.25Jalanidhi 25.95 31.65 38.05 2.15 49.10NDGR410 27.25 44.40 63.75 3.08 49.35NDGR413 28.00 34.50 43.75 1.75 47.80LPR85 27.05 36.75 50.80 2.25 47.10TCA12 28.45 33.90 43.55 1.35 49.90Chakia 59 26.55 38.00 43.25 1.60 48.10Madhukar 26.55 34.50 46.05 1.87 48.15Jalapriya 32.40 39.55 47.20 2.25 45.45 Mean 27.71 37.28 49.32 2.33 46.06

Nondeepwater riceKhao Y Khao 17.25 26.25 29.55 0.38 59.15IET7590 21.85 28.15 36.15 0.40 59.90Kwan-Fu-1 21.10 20.75 27.20 0.23 63.50Mtu-6 22.10 28.75 38.35 0.58 56.45TK Deep straw 34-774 17.55 26.20 27.40 0.10 50.60Deep straw 24-500 12.40 19.35 21.20 0.05 58.95Samson Polo (YS386) 21.45 25.05 37.85 0.55 54.25Daeng Laem 17.85 18.85 22.90 0.45 62.65Basmati 802 19.25 21.35 24.05 0.43 57.10Savitri 13.25 15.80 19.90 0.28 51.00 Mean 18.41 23.05 28.46 0.34 57.36

old seedlings by taking a sample of 10seedlings from each type.

The two seasons were found to behomogeneous, with nonsignificant dif-ferences for the two characters as tested

by Bartlett’s test. The pooled data overtwo seasons (see table) indicated highermesocotyl elongation for deepwaterrice than for nondeepwater rice vari-eties for both pot culture and dark petri

Vol. 23, No. 2 31

Tolerance for submergence inrainfed lowland rice underrepetition of flooding

R.K. Sarkar, R.K. Sahu, and R.N. De, Central RiceResearch Institute, Cuttack 753006, Orissa, India

In South and Southeast Asia, rainfedlowland rice is generally affected bysubmergence during flash floods whenthe plant remains completely underwater. The frequency and duration offloods differ from place to place andyear to year, causing different degreesof plant mortality. Information ontolerance for single submergence isavailable, but it is lacking on cultivarsurvival because of recycling offlooding. These experiments wereconducted to identify rice cultivars thatcould withstand different cycles offlooding.

The experiments were carried outduring the wet season with 242cultivars of the Assam Rice Collection(ARC), which is considered to be a richsource of germplasm because of itshigh degree of phenotypic divergence.The genotypes were direct-seeded witha spacing of 20 × 10 cm. Two seedlingshill-1 were maintained by thinning after10 d of germination. Twenty-day-oldseedlings were submerged for 12 dunder 60 cm of water followed bynormal conditions with 10 cm of water.After 10 d from the removal of waterpressure, that is, 42 d after sowing, asevere submergence treatment was

measured around 12 noon was abovethe air saturation (>7.36 ppm) level(Table 1), but susceptible checks wereruined because of the severe pressure ofsubmergence (Table 2).

After second submergence, a fewgenotypes survived. Survival per-centage was maximum in Matia (87%),followed by FR13A (85%), CR380-10(60%), and ARC18112 (56%) and 18101(51%). Sabita, a submergence-tolerantcultivar, identified earlier on the basisof single-submergence treatment,showed only 35% survival. After firstsubmergence, when water receded, theelongated top leaves descended to theground and ultimately decayed.Therefore, plant height just beforesecond submergence was less than theheight just after the completion of firstsubmergence.

Yield and yield-contributingcharacters showed great variation

Table 1. Floodwater characteristics at around 12 noon. Data given as mean +SE.

Water Days after first submergence Days after second submergencedepth(cm) 2 5 9 2 5 5

Light transmission (%) 5 72.6 ± 0.4 70.4 ± 0.3 73.3 ± 1.2 68.4 ± 0.3 67.1 ± 0.6 62.2 ± 0.330 62.2 ± 0.6 64.5 ± 0.9 62.3 ± 1.0 56.0 ± 0.7 53.6 ± 1.1 45.9 ± 0.460 53.7 ± 0.5 53.6 ± 0.4 53.5 ± 1.0 50.1 ± 0.4 42.2 ± 1.1 37.6 ± 0.4 Mean 62.8 62.8 63.0 58.2 54.3 48.6

Oxygen concentration (ppm) 5 7.37 ± 0.02 7.92 ± 0.14 7.40 ± 0.07 7.65 ± 0.04 7.82 ± 0.03 7.95 ± 0.0530 7.47 ± 0.06 7.75 ± 0.13 7.57 ± 0.02 7.57 ± 0.01 7.77 ± 0.04 7.90 ± 0.0560 7.92 ± 0.02 7.50 ± 0.12 7.55 ± 0.04 7.95 ± 0.03 7.90 ± 0.02 7.72 ± 0.05 Mean 7.59 7.72 7.51 7.72 7.83 7.86

given to the regenerated seedlings foranother 10 d under 80 cm of water. Asurvival count was taken visuallyimmediately after drainage of the water,and was confirmed after 15 d.

Dissolved oxygen concentration wasmeasured using a submersible polaro-graphic oxygen electrode (SylandModel 610, Heppenheim, Germany).Light transmission was measured usinga quantum underwater light sensor(Licor, Model 185B).

To determine yield and yield-contributing characters, an experimentwas conducted under rainfed lowlandconditions where water depth variedbetween 0 and 50 cm.

Light penetration inside the flood-water was comparatively less in secondsubmergence than in first submergence,which suggested that the floodwaterwas more turbid during secondsubmergence. Oxygen concentration

dish culture. It was also evident thatelongation was greater at higherseeding depths in all the test genotypes.On the contrary, the length of thesecond leaf blade as measured inseedlings raised under normal potculture showed a reverse trend, i.e., thelength was less for deepwater ricegenotypes than for nondeepwatertypes.

These two seedling characters wereused as morphological markers to eval-uate germplasm and segregating linesfor tolerance for deepwater conditionsin our subsequent experiments. ■

Table 2. Survival percentage and changes in plant height because of repetition of flooding.

Seedling height (cm)Germplasm/ Survival (%)cultivar First submergence Second submergence after second

submergenceBefore After Before After

FR13A 23 76 48 72 85Sabita 29 97 44 74 35CR380-10 34 100 53 98 60Matia 35 107 61 100 87ARC18112 27 91 34 86 56ARC18101 28 90 41 85 51ARC18104 28 87 36 83 46ARC18172 32 91 41 81 42ARC18198 29 83 42 80 40Tulasi 25 70 34 52 03Mahsuri 23 76 35 63 02 Mean 26 88 43 79 46

32 IRRN 1998

Yield potential

Physiological basis ofheterosis in rice

A.K. Singh and F.U. Zaman, Division of Genetics,Indian Agricultural Research Institute, NewDelhi 110012, India

New plant types and hybrids arephysiologically more efficient. Theyhave a higher biomass than inbredvarieties and a comparable harvestindex. Increasing biomass yield andmaintaining a high harvest index arethe main breeding strategies for achiev-ing another breakthrough in rice yield.Biomass is a complex trait governed bya number of morphophysiologicalcomponents. We attempted to analyzethe physiological components ofbiomass and to study the characterassociation and path relationshipamong its components. A strategy toselect parents for developing hybridsand pedigree breeding based onphysiological efficiency are outlined.

The material for this study consistedof 15 F1s and 6 parents from a 6 × 6 halfdiallel involving diverse parents forphysiological efficiency. The experi-ment was conducted in a randomizedblock design with three replications.Data were recorded on characters suchas biological yield, grain yield, harvestindex, days to 50% flowering, grain-filling duration, days to maturity,cumulative growth rate (CGR), relativegrowth rate (RGR), net assimilation

rate (NAR), leaf area ratio (LAR), leafarea index (LAI), leaf area duration(LAD), and leaf area plant-1 at differentgrowth stages. Correlation, path effects,and heterosis were estimated.

Grain yield, CGR, LAI, LAD, andleaf area at seedling, tillering, flower-ing, and maturity showed a highlysignificant positive association withbiological yield. The cross, Jaya/Swarnaprabha, which exhibited a highheterobeltiosis for biological yield(54.9%), also showed a significantpositive heterosis for grain yield,harvest index, CGR, LAI, and leaf areaplant-1. Characters that showed anegative association with biologicalyield did not exhibit any positiveheterosis. It is therefore evident that, todevelop heterotic hybrids, we shouldidentify parents with high physio-logical efficiency in terms of CGR, LAI,LAD, and leaf area plant-1. Becausecharacters such as NAR, LAR, LAD,and leaf area showed a high positivedirect path effect, direct selection forthese traits should be effective.

With this understanding of the keyphysiological determinants of biolo-gical yield, the interrelationship amongthese components, and their direct andindirect effects on biological yield, wecan make a better selection of parents.This approach should help developsuperior hybrids and generate valuablebreeding material for pedigreeselection. ■

Physiological efficiency ofrice hybrids

K.V. Sitaramaiah, J. Madhuri, and N.S. Reddy,Rice Research Unit, Agricultural College Farm,Bapatla 522101, India

Heterosis observed in hybrids is theresult of various physiological pro-cesses that occur at different stages ofplant growth. A precise knowledge ofthe factors that contribute to heteroticvigor would help to manipulateheterosis expression in hybrids. Wemade a study to assess the physio-logical efficiency of rice hybrids duringthe crop growth period and to analyzethe factors that contribute to heterosisexpression. Six promising experimentalrice hybrids were evaluated along withthe popular check varieties SambaMahsuri and Chaitanya during the1995 wet season. Crop growth rate(CGR), leaf area index (LAI), andbiomass production (BMP) wereestimated at 15-d intervals from 15 to 60d after planting (DAP). Harvest index(HI) and grain yield were recorded atharvest.

CGR increased up to 45 DAP anddecreased later in all hybrids andchecks. LAI, however, consistentlyincreased from 15 to 60 DAP andhybrids recorded a higher LAI from 30DAP onward compared with thechecks. The high initial CGR was notmaintained to the later growth stages

Comments. If you have commentsor suggestions about the IRRN,please write to the editor.

Table 3. Grain yield and its contributing characters under waterlogged (0-50 cm depth) conditions.

Germplasm/ Plant Days to Single PBTa Harvest Graincultivar height 90% panicle (no. m–2) index yield

(cm) flowering weight (g) (%) (g m-2)

FR13A 158 145 3.63 135 23.1 386Sabita 184 141 4.46 95 31.0 421CR380-10 169 149 4.20 138 33.5 472Matia 178 149 2.58 110 21.6 215ARC18112 154 98 2.59 118 24.9 260ARC18101 158 121 3.03 100 31.2 261ARC18104 152 131 3.75 104 29.4 347ARC18172 185 110 2.50 109 27.9 212ARC18198 122 92 2.34 133 38.4 267Tulasi 145 141 4.23 153 29.4 516 Mean 160 128 3.33 119 29.0 336 CV 11.4 15.8 23.2 15.2 16.3 30.8

aPBT = panicle-bearing tillers.

(Table 3). For the rainfed lowlands ofeastern India, plants with differentmaturity periods are desirable fordifferent locations. Therefore, plantbreeders could employ the genotypesmentioned here. ■

Vol. 23, No. 2 33

or translated into higher yields. Earlyvigor was probably sacrificed to ensurea better canopy structure at the lategrowth stages and better partitioning ofphotosynthates to grain. This was wellexhibited by the hybrids, especiallyMTUHR2033, MTUHR2020, andMTUHR2037, which had significantlyhigher grain yields. These hybrids alsorecorded greater BMP and HI. Thoughthe check varieties and hybrids did notdiffer for BMP, the hybrids showed asuperior performance because of moregrains per panicle, which was indicatedby a higher HI. Two hybrids,MTUHR2024 and MTUHR2029,however, showed an inferior perform-ance compared with the best checkChaitanya in BMP, HI, and grain yield.

Heterosis is a highly cross-specificphenomenon. To use heterosis success-fully to increase grain yield, parentalgenotypes need to be selected frommodern cultivars with a high potential.The physiological efficiency of F1hybrids at the vegetative stage is auseful indication of heterosis for grainyield. MTUHR2033 and MTUHR2020,which showed such physiologicalefficiency, hold good promise. ■

Performance of hybrid ricein south Karnataka

M. Rudraradhya, S. Panchaksharaiah, and B. R.Patil, University of Agricultural Sciences,Shimoga Unit, Bangalore, India

At the agricultural research stations atHonnaville and Kathalagere, we evalu-ated elite hybrids. We also investigatedthe fixing of isolation distance, numberof seedlings to be planted hill-1, and Nmanagement. Among the varioushybrids tested in 1991-94, IR58025A/IR9761-19-1R was found to be superiorto inbred variety Mangala. This hybridrecorded an average yield of 5.6 and 4.9t ha-1, respectively, at Honnaville andKathalagere. Mangala yielded 3.8 and3.6 t ha-1, respectively, at those twocenters. This hybrid also performedwell in on-farm trials in the districts ofChikamagalore, Hassan, Mysore, andShimoga, with a mean grain yield of 5.9t ha-1 vs 4.7 t ha-1 for Mangala. Thishybrid was named Karnataka RiceHybrid 1 (KRH1) and released forgeneral cultivation in the southerntransition zone.

Crop and resource managementPhysiology and plant nutrition

Effect of salinity on growth,chlorophyll content, and flag leafarea of rice (Oryza sativa L.)genotypes

M. Yasin Ashraf and Yousaf Ali, Nuclear Institutefor Agriculture and Biology (NIAB), Jhang Road,P.O. Box 128, Faisalabad, Pakistan

Rice is the most important food crop,feeding more than half of the world’spopulation. To feed this ever-growingpopulation, an increase in the cultiva-tion of food crops is essential, which ispossible by increasing cultivable lands.

Lands in Pakistan that are availablefor cultivation suffer from problems of

salinity/sodicity. So we urgently needto select or develop salt-tolerant ricevarieties or methods to increase theyield per unit area.

Although yield is the result ofinteractions in the genetic makeup ofgenotypes, it has been suggested thatby increasing photosynthetic efficiency,crop production could be increased(Ashraf et al 1995). Photosynthetic effi-ciency depends on leaf area, chloro-phyll content, stomatal response, gasexchange, etc. We therefore used aphysiological-genetic approach to testwhether genotypes sensitive to ortolerant of salt differ in their photosyn-thetic efficiency.

For this purpose, a gravel cultureexperiment was conducted in a growthcabinet maintained at 30/25 + 2 °Cday/night temperature and a photo-period of 10 h, to study the effect ofsalinity created by mixed salt(Na2SO4, CaCl2, MgCl2, and NaCl in theratio of 6.33:3.58:1:2.28) and NaCl onlyon growth and photosynthetic activityof rice genotypes. Thirty-day-old seed-lings of five rice genotypes—BH-5-89,BH-3-89, RST-1-86, RST-2-84, andBas370×NR1—were acquired from therice section of the Mutation BreedingDivision of the NIAB, Faisalabad,Pakistan.

Similarly, another F1 hybrid,IR58025A/ KMR 3, had an averageyield of 6.6 t ha-1 vs 5.1 t ha-1 for Jaya.This hybrid was found promisingbecause it matured in 125-130 d, similarto the popular inbred variety. In 12 on-farm trials, this hybrid registered amean grain yield of 7.3 t ha-1 vs 6.0 t ha-1

for the standard check Jaya. This hybridwas named Karnataka Rice Hybrid 2and released for general cultivation in1996. The present recommended doseof 100-50-50 kg NPK ha-1 for thevarieties was found to be equallyeffective for the two hybrids. There wasno difference in yield among plotsplanted with one, two, three, four, orfive seedlings hill-1 in both the dry andwet seasons. Tillering ability in thehybrids was found to be very good.Therefore, planting 1 seedling hill-1

ensured an economical seed input cost.In the studies conducted on isolationdistance, seed set decreased with everyincrease in the isolation distance. Thiscalled for a revision of isolationdistance for each hybrid throughinvestigations at specific locations. ■

34 IRRN 1998

Table 1. Effect of different salinity levels on biomass (dry matter plant–1) and flag leaf area (cm2) of different ricegenotypes.

Treatment Genotypes(dS m-1)

BH-5-89 BH-3-89 RST-1-86 RST-2-84 Bas370XNR1

0 (control)DMa 2.29 2.31 4.53 3.45 3.88

(100)b (100) (100) (100) (100)LA 16.60 20.62 36.67 22.31 21.87

(100) (100) (100) (100) (100)5 (mixed salt)

DM 2.11 3.58 4.41 3.57 3.09(92) (154) (97) (103) (74)

LA 13.80 18.60 21.19 19.35 13.72(83) (90) (58) (86) (63)

10 (mixed salt)DM 1.69 2.04 2.23 1.37 2.44

(74) (83) (43) (34) (63)LA 10.30 15.41 17.90 8.37 12.75

(62) (75) (44) (38) (54)5 (NaCl)

DM 2.01 2.09 2.35 2.59 2.96(88) (90) (52) (75) (63)

LA 10.26 13.05 14.27 11.47 13.10(61) (63) (34) (51) (54)

10 (NaCl)DM 1.54 1.68 1.42 1.36 2.10

(67) (54) (31) (33) (54)LA 6.30 4.26 4.77 4.11 10.52

(38) (21) (13) (18) (43)

aDM = dry matter (g plant-1), LA = flag leaf area (cm2). bValues in parentheses show % of control.

Table 2. Effect of different levels of salinity on chlorophyll (a, b, and total) (mg g–1 fresh weight) of differentrice genotypes.

Treatment Genotypes(dS m-1)

BH-5-89 BH-3-89 RST-1-86 RST-2-84 Bas370XNR1

0 (control)Chl aa 0.424 0.427 0.299 0.383 0.397

(100)b (100) (100) (100) (100)Chl b 0.486 0.494 0.517 0.490 0.424

(100) (100) (100) (100) (100)Chl t 0.910 0.921 0.816 0.873 0.821

(100) (100) (100) (100) (100)5 (mixed salt)

Chl a 0.407 0.412 0.299 0.302 0.399(96) (95) (100) (74) (101)

Chl b 0.438 0.496 0.481 0.481 0.443(90) (100) (64) (98) (104)

Chl t 0.845 0.908 0.781 0.783 0.842(93) (99) (89) (90) (103)

10 (mixed salt)Chl a 0.360 0.367 0.233 0.297 0.389

(85) (86) (78) (77) (98)Chl b 0.382 0.476 0.261 0.440 0.418

(79) (76) (35) (90) (99)Chl t 0.742 0.743 0.494 0.737 0.807

(82) (81) (57) (84) (98)5 (NaCl)

Chl a 0.403 0.384 0.289 0.262 0.392(95) (84) (96) (63) (99)

Chl b 0.411 0.424 0.406 0.364 0.384(85) (86) (54) (74) (91)

Chl t 0.814 0.808 0.695 0.626 0.776(89) (88) (80) (72) (94)

Continued on next page

This experiment consisted of fivetreatments (0, 5, and 10 dS m-1 ECprepared with mixed salt and EC 5, 10dS m-1 with NaCl only) with three repli-cations in a completely randomizedblock design. The plants were sown inplastic pots (30 cm diam and 25 cmlength) filled with washed gravelcontaining full-strength Hoaglandsolution. The treatments were preparedby adding salt solutions from 2.5 dS m-1

and maintained at desired concentra-tions. This was done to prevent suddenshock to the plant from salinity. Thevolume of the solution was maintaineddaily by adding Hoagland solution.The plants were harvested after 1 moand their dry weight plant-1 (afterdrying the plants in an oven at 70 °C for72 h), flag leaf area (Ashraf et al 1995),and chlorophyll content (Arnon 1949)were estimated.

The results indicated that biomassdecreased with an increase in saltconcentration in all the genotypesexcept at 5 dS m-1 of mixed salt, whereenhancement was recorded in ricegenotype BH-3-89 (Table 1). Themaximum reduction in biomass wasunder 10 dS m-1 (NaCl). In BH-5-89,Bas␣ 370×NR1, and BH-3-89, thereduction in biomass at 10 dS m-1

(NaCl) was less than 50%. On the otherhand, RST-1-86 and RST-2-84 had abiomass of only 31 and 33% that of thecontrol under the same treatment.Similar genotypic differences havebeen reported for rice, sorghum, andwheat.

A reduction in flag leaf area wasobserved in all the genotypes withincreases in salt concentration (Table 1).The maximum decrease was againunder 10 dS m-1 (NaCl). The genotypeswith a higher biomass had a higher flagleaf area (Table 1). The minimum flagleaf area was recorded for RST-1-86 andRST-2-84, and with mixed salt. Hussainand Ismail (1994) also reported areduction in leaf area caused by salinityin sunflower.

The chlorophyll content (a, b, andtotal) of rice leaves was generallyreduced by salinity (Table 2). RST-1-86

Vol. 23, No. 2 35

Crop management

A labor-saving technique indirect-sown and transplantedrice

P. Santhi, K. Ponnuswamy, and N. Kempuchetty,Department of Agronomy, Tamil Nadu RiceResearch Institute (TNRRI), Aduthurai 612101,India

The Cauvery Delta zone accounts forabout 22.3% of the rice area and 25.3%of rice production in Tamil Nadu State.Rice is cultivated in three distinctseasons—kuruvai (Jun-Oct), followedby thaladi (Oct-Feb) in double-crop

wetlands and samba (Sep-Jan) in single-crop wetlands.

Transplanting continues to be themajor method of crop establishment forrice in this area. It requires more labor,time, and efforts. The scarcity and highcost of farm labor invariably delaytransplanting and often lead to the useof aged seedlings. Uncertain release ofwater from the Mettur dam aggravatesthe situation for seedling age and timeof transplanting.

Because of these problems, manyrice farmers wish to switch to directseeding under puddled conditions.

Table 2 continued.

Treatment Genotypes(dS m-1)

BH-5-89 BH-3-89 RST-1-86 RST-2-84 Bas370XNR-1

10 (NaCl)Chl a 0.300 0.310 0.163 0.222 0.357

(71) (73) (54) (58) (90)Chl b 0.370 0.405 0.172 0.297 0.354

(76) (62) (23) (61) (83)Chl t 0.670 0.615 0.335 0.519 0.711

(74) (67) (39) (59) (87)

aChl a = chlorophyll a, Chl b = chlorophyll b, Chl t = chlorophyll (total). bValues in parentheses show % of control.

and RST-2-84 had the lowest chloro-phyll content of the five genotypes.

Biomass production is a function ofphotosynthetic efficiency (Terry andWaldron 1984), which is decreasedbecause of the decrease in leaf area andchlorophyll. A decrease in leaf areacauses a reduction in area for light

interception and absorption of thespecific wavelength necessary forphotosynthesis. The results of thisexperiment showed that NaCl is moretoxic and injurious than mixed salts andthat BH-5-89, BH-3-89, and Bas370 aremore tolerant of salt stress and RST-1-86and RST-2-84 are sensitive to salt stress.

ReferencesArnon DT. 1949. Copper enzymes in

isolated chloroplasts. Polyphenol-oxidase in Beta vulgaris. Plant Physiol.24:1-15.

Ashraf MY, Khan AH, Naqvi SSM. 1995.Relationship of chlorophyll content andleaf area with grain yield in wheatgenotypes. Indian J. Plant Physiol.38(2):162-163.

Hussain S, Ismail S. 1994. Effect of salt andwater stress on growth and biomassproduction in Helianthus annuus L. Pak. J.Bot. 26(1):127-138.

Terry N, Waldron LJ. 1984. Salinity photo-synthesis and leaf growth. Calif. Agric.38:38-39. ■

Besides reducing labor requirements, itcan help to shorten the growth cyclefrom seeding to maturity. Experimentswere conducted during 1995-96 and1996-97 at the TNRRI to evaluate wetseeding for establishing rice crops.

The methods of crop establish-ment—transplanting (S1), sowingsprouted seeds in lines manually (S2),and using a seed drum (S3)—werecompared. The drum seeder used inthis trial was developed by the Depart-ment of Agricultural Engineering,Tamil Nadu Agricultural University,Coimbatore. The 8-row drum seeder

Table 1. Effect of crop establishment methods on yield and economics of rice, 1995-96.

Days to 50% Grain yield Straw yield Net return Return Re–1

Treatment flowering (t ha-1) (t ha-1) (Rs ha-1) invested

Kuruvaia Thaladi Kuruvai Thaladi Kuruvai Thaladi Kuruvai Thaladi Kuruvai Thaladi

Transplanting (S1) 85.8 105.6 5.5 4.8 6.6 5.1 16,165 11,982 2.32 1.98Sowing sprouted seeds in lines manually (S2) 78.6 98.6 5.5 4.9 6.7 5.3 16,561 13,545 2.47 2.20Drum seeding of sprouted seeds (S3) 78.9 98.7 5.6 5.0 6.8 5.3 17,836 13,960 2.65 2.29 CD (P = 0.05) 0.62 0.55 nsb ns ns ns ns ns 0.204 ns

aKuruvai = Jun-Oct, thaladi = Oct-Feb. bns = not significant.

36 IRRN 1998

requires only 9 kg of pulling force tooperate. The machine weight is 11 kgwithout seed and 19 kg with seed (2 kgseed hopper-1). It requires 14 person-hto cover 1 ha.

Direct seeding reduced the time to50% flowering by 7 d in both thekuruvai and thaladi seasons in bothyears (Tables 1 and 2). The delay inflower emergence by 1 wk in trans-planted rice might be due to the trans-planting shock experienced by theseedlings.

Table 2. Effect of crop establishment methods on yield and economics of rice, 1996-97.

Days to 50% Grain yield Straw yield Net return Return Re–1

Treatment flowering (t ha-1) (t ha-1) (Rs ha-1) invested

Kuruvaia Thaladi Kuruvai Thaladi Kuruvai Thaladi Kuruvai Thaladi Kuruvai Thaladi

Transplanting (S1) 85.9 105.4 5.9 5.0 7.2 5.5 17,542 12,813 2.40 2.02Sowing sprouted seeds in lines manually (S2) 78.8 98.5 5.8 5.2 7.4 5.7 18,303 14,657 2.58 2.27Drum seeding of sprouted seeds (S3) 78.9 98.7 6.0 5.3 7.4 5.7 19,453 15,046 2.75 2.30 CD (P = 0.05) 0.62 0.55 ns ns ns ns ns ns 0.180 0.172

aKuruvai = Jun-Oct, thaladi = Oct-Feb. bns = not significant.

The methods of crop establishmentshowed no significant influence ongrain yield of rice in both the kuruvaiand thaladi seasons in both years ofstudy. In both seasons, however, drumseeding gave a slightly higher grainyield. A similar trend was noticed in thecase of straw yield. Yield parameterssuch as number of panicles, paniclelength, number of filled and immaturegrains, and 1,000-grain weight were notaffected by the method of cropestablishment.

Net savings averaged Rs 1,911(US$47) and Rs 2,233 (US$56) ha-1 in thedrum-seeding method versus thetransplanting method in the kuruvaiand thaladi seasons, respectively. Thedrum-seeding method required only 8person-h whereas line sowing ofsprouted seeds and transplantingmethods required 200 and 400 person-hha-1, respectively. There is thus asignificant savings of labor in thedrum-seeding method. ■

Optimum seedlings per unit areafor high-yielding rice varieties inthe hill zone of Karnataka

B. Jagannath, K.S. Kamath, H.M. Chandrappa,Y.G. Shadakshari, and N.E. Thyagaraj, RegionalResearch Station, Mudigere 577132, Karnataka,India

Rice is the major field crop grownduring monsoon under rainfedconditions in the hill zone of Karnataka(located between 11° 56' and 15° 46'north latitude and 74° 31' and 76° 4' eastlongitude). For improved low- andmidland rice cultivation, 20 × 10-cmspacing (50 hills m-2) is beingrecommended for improved ricevarieties. Adoption of this recommen-dation by farmers was low on thegrounds that the number of seedlingsrecommended per m2 was too high.

To verify whether the number ofseedlings per unit area could bereduced, an experiment was conductedwith the popular high-tillering variety

Intan during 1993 and 1994. The experi-ment was continued during 1995 andlow-tillering, high-yielding, blast-tolerant variety DWR 4107 (Hemavati)was included. DWR 4107 has beenreleased to replace Intan in high-blastpressure areas of the zone.

A randomized block design repli-cated four times was employed in 1993and 1994. A split-plot design was em-ployed in 1995 with varieties as mainplots and plant spacing as subplots. Inaddition to the recommended spacing,an additional four spacings were usedto produce 45, 40, 35, and 30 hills m-2.Depending on climatic factors, nurserysowing and transplanting varied overthe years. Sowing and transplantingwere done in Jul 1993. Sowing wasdone in Jun 1994 and 1995 and trans-planting was done in Jul 1994 and Aug1995. Net plot size was 12.5 m2 during1993 and 1994 and 9.6 m2 during 1995.Recommended P (33 kg P ha-1) and K(72.6 kg K ha-1) along with 50% N (37.5kg N ha-1) were applied at the time of

transplanting and the balance of N wastopdressed in two splits at 30 and 60 dafter transplanting. There was no needfor plant protection measures in any ofthe 3 yr.

Compared with 1993 and 1995,grain yield of Intan was significantlyhigher in 1994 (see table) because sow-ing and transplanting were done ontime during that year. The seedlingswere not sufficiently aged in 1993 andthey were overmatured in 1995 (seetable). Grain yield from the recom-mended 50 seedlings m-2 was signifi-cantly different from that produced by30, 35, 40, and 45 seedlings m-2 eachyear, indicating that a similar grainyield could be obtained by adoptingwider spacing, accommodating as fewas 30 seedlings m-2. There were signifi-cant differences between treatments forother characters, but these differenceswere not reflected in grain yield.

A pooled analysis over the 3 yrindicated that there was a significantincrease in tillers plant-1 and panicles

Vol. 23, No. 2 37

Effect of number of seedlings m-2 on rice yield and its components in the hill zone of Karnataka.a

1,000-Variety Year/ Plant Tillers Panicles Panicle Grains panicle-1 Panicle grain Yield (t ha-1)

seedlings height plant-1 plant-1 length sterility weighthill-1 (cm) (no.) (no.) (cm) Total Chaffy (%) (g) Grain Straw

Intan 1993 102 b 9.3 b 9.2 b 21.0 b 120.5 b 22.6 b 19.0 25.3 a 4.98 b 7.50 a1994 104 a 10.1 a 9.9 a 21.6 a 127.2 b 21.7 b 17.1 24.5 b 6.24 a 7.92 a1995 100 c 7.6 c 7.6 c 21.9 a 149.0 a 27.8 a 18.5 23.5 c 4.43 c 5.72 b

CD (5%) 0.92 0.59 0.57 0.35 7.55 3.69 ó 0.48 0.26 0.5530 102 10.0 a 9.8 a 21.3 135.8 25.0 18.9 24.1 b 5.12 6.9635 102 9.2 b 9.1 b 21.6 134.0 23.9 17.5 24.3 ab 5.09 7.0840 102 9.0 b 8.9 b 21.7 131.0 24.7 19.0 24.7 a 5.26 6.8845 102 8.7 bc 8.6 bc 21.3 126.9 24.4 19.1 24.2 ab 5.21 6.9950 102 8.1 c 8.1 c 21.7 133.5 22.0 16.7 24.8 a 5.39 7.31

Mean 1993-95 102 9.0 8.9 21.5 132.2 24.0 18.2 24.4 5.21 7.05CD (5%) – 0.76 0.73 – – – – 0.70 – –CV (%) 1.40 10.17 9.93 2.54 8.90 23.94 22.01 3.09 7.81 12.09

DWR4107 199530 106 7.2 a 7.2 a 22.9 ab 159.3 11.9 ab 7.3 ab 21.1 2.96 4.4635 105 6.8 b 7.0 ab 22.1 ab 159.0 13.4 a 8.4 a 20.5 3.11 4.6940 106 6.6 b 6.6 bc 23.1 a 166.5 13.1 a 7.7 ab 21.6 3.28 4.4045 106 6.1 c 6.1 c 22.0 ab 144.2 9.2 b 6.7 ab 20.7 3.18 4.5350 105 6.0 c 6.0 c 21.9 b 146.3 9.3 b 6.1 b 21.1 3.23 4.67

Mean 106 6.5 6.6 22.4 135.0 11.4 7.3 21.0 3.15 4.55CD (5%) – 0.44 0.61 1.20 – 3.56 2.12 – – –CV (%) 1.91 4.34 6.00 3.47 10.91 20.26 18.90 3.76 6.98 12.83

Intan 1995 100 7.6 a 7.6 21.9 149.0 b 27.8 a 18.5 a 23.5 a 4.43 a 5.72DWR4107 1995 106 6.5 b 6.6 22.4 135.0 a 11.4 b 7.3 b 21.0 b 3.15 b 4.55

Mean 103 7.1 7.1 22.4 152.0 19.6 12.9 22.3 3.79 5.13CD (5%) – 0.81 – – 3.72 15.34 11.17 2.28 0.58 –CV (%) 1.78 7.46 7.9 3.40 10.38 31.63 27.98 3.34 11.32 14.47

aMean values followed by dissimilar letters are significantly different from one another.

Influence of plant population on grain yield of rice.

plant-1 and a significant decrease ingrain weight in the treatment withwider spacing (30 hills m-2) comparedwith the treatment with narrowspacing (50 hills m-2) and these traitswere similar in treatments with 35, 40,and 45 hills m-2. But grain and strawyield were statistically similar in all the

xx

Grain yield (t ha-1)

6

5

4

3

2

1

0

10

9

8

7

6

5

4

3

2

1

0

Tillers hill-1 (no.)

Grain yield, IntanGrain yield, HemavatiTillers hill-1, IntanTillers hill-1, Hemavati

Seedlings m-2 (no.)30 35 40 45 50

xxxx xx

xx xxx

treatments. Thus, wider spacingbetween hills as a result of highertillering compensates for reduced hillsand grain yield will be on a par withnarrow spacing, which requires 20seedlings more per m2. The dataobtained with DWR 4107 alsoconfirmed the same trend (see figure).

The results obtained with twovarieties significantly different fromeach other for tillering ability revealedthat grain and straw yield withrecommended spacing of 50 hills m-2

was on a par with spacings for 30-45hills m-2. We therefore suggest revisingthe number of seedlings recommendedper m2 and reducing it by 20.

Intan was significantly superior toDWR4107 in tillering, grain yield, andgrain weight, though it showed signifi-cantly higher chaffy grains panicle-1

and panicle sterility. The significantlylower grain yield of DWR4107 isattributed to aged seedlings. ■

Multiple submissions. Normally, onlyone report for a single experiment willbe accepted. Two or more items aboutthe same work submitted at the sametime will be returned for merging.Submitting at different times multiplenotes from the same experiment ishighly inappropriate. Detection willresult in the rejection of all submis-sions on that research.

38 IRRN 1998

Plant stand (plants m–2)

150

100

50

00 20 40 60 80 100 120

Cumulative rainfall (mm)

Plant stand (plants m–2) vs cumulative rainfall(from day prior to sowing to day 3 after sow-ing).

Drill sowing of pregerminatedrice seed: effect of rainfall onplant stand

M.A. Bell, G.F. Pateña, and D.C. Mendoza, IRRI

In recent years, wet seeding of rice(Oryza sativa) has gained in popularitythroughout Asia, primarily because ofthe scarcity of available labor. With thissystem, rice is usually broadcast, butdrill-seeded rice in rows above orbelow (anaerobic) the soil is an optionthat has the advantage of facilitatingmanual weed control. During the 1995dry season (DS), the anaerobic seedingmethod was used on 32 fields covering8 ha at the IRRI experiment station.Results were promising—standestablishment was visually rated asgood and the average yield of 5.4 t ha-1

was reasonable.Given the success in the 1995 DS, the

anaerobic seeding method was againused on 19 fields during the 1995 wetseason (WS). Fields were prepared bywet disk harrowing and puddlingusing tractors before a final landleveling using hand tractors. Seed wasgermination-tested, presoaked (24 h),and then incubated for 24 h. Seed wassown at a rate of 80 kg ha-1. Afteremergence, crop establishment in allexcept two fields was consideredextremely poor, with less than 100plants m-2. (Fifty percent emergencewould equate to approximately 160plants m-2.) Germination was not afactor because all seed had a germi-nation percentage above 95%. Thisresult was particularly worrisomegiven the success of the method in theprevious season.

When rainfall results for the 1995WS were compared with planting datesand emergence, it was apparent thatrainfall at planting had a serious nega-tive effect on stand establishment (seefigure), with the relation described bythe equation:

log (crop establishment [plants m-2])= 1.97 - 0.0058 (cumulative rainfall [mm]);r2 = 0.44

We include rainfall on the day priorto sowing because this will affect soilconsistency at the time of sowing. Smallamounts of rain at planting clearly hadstrongly negative effects on crop estab-lishment, with the effect appearing tobe significant at rainfall rates as low as10 mm (see figure). This study supportsthe work of others on the importance ofgood water management at planting.

Effect of planting geometryand N levels on grain yield

of hybrid cultures

S.P. Singh, K.V. Rao, S.V. Subbiah, and K.G. Pillai,Directorate of Rice Research, Rajendranagar,Hyderabad 500030, India

The influence of planting geometry (20× 15, 15 × 15, and 20 × 10 cm) and Nlevels (90 and 150 kg ha-1) on theproductivity of two rice hybrids (TNH1and TNH2) was evaluated in compari-son with Rasi and Jaya as standardchecks during the 1992 wet season.Planting densities of 0.33-0.50 millionhills ha-1 did not significantly influencethe productivity of the test hybrids andvarieties. The yield response to Napplication was significant up to 150 kgN ha-1. The interaction effects amongtreatments were nonsignificant. Jayaproduced the highest grain yield (5.1 tha-1). Rasi and TN1 were on a par ingrain yield, but TNH2 recorded thelowest grain yield of 3.1 t ha-1. Ourresults indicate that maintaining aplant population of 0.33-0.50 millionhills ha-1 and a higher N application(150 kg ha-1) are required to achievehigher grain yields in hybrid rice. ■

Field observations suggested that atleast three factors were involved:(1) rainfall buries seed under a layer ofmud and water, (2) rainfall trapped in afield hinders germination, and (3) seedwashed to the top of the soil was moreprone to be eaten by birds. One possibleresponse to the first two factors is toimprove field drainage, including theuse of small canals or canalettes.Because anaerobic seeding seems tofunction in some regions even underrainy conditions, we speculate that soiltexture and drainage that will affect thedevelopment of anaerobic conditionsaround the seed are important to thesuccess of the technology.

Although broadcast wet seedingwas not included in this study, thenegative effect of rain on germinationin that system has been observed inother fields. ■

Farm machinery

An improved suction apparatusfor sampling invertebrate com-munities in flooded rice

I. Domingo and K.G. Schoenly, IRRI

Arida and Heong (IRRN 17(6):30-31,1992) invented a petrol-driven suctionapparatus (hereafter called the originalBlower-Vac sampler) to replace thebattery-powered FARMCOP for sam-pling invertebrates in flooded rice. Inthis note, we report on several designand performance improvements in theoriginal Blower-Vac sampler.

The improved suction apparatus(see figure) shortens (by 0.6 m) the paththat air, water, and invertebrates travelthrough, beginning with a 1.5-m-longrubber hose (2 cm diam), an invertedplastic soft drink bottle with its bottomremoved, and a pair of open-endedglass vials glued end-to-end with a finemesh strainer in between that permitsonly water to pass. The water trapattaches to a 12-cm rubber stopper andfastens underneath the machineblower. Elastic cords, hooked to the

Vol. 23, No. 2 39

water trap, fix the unit firmly to theblower. A PVC reducer (measuring 10cm high and 5.8 cm in diam) attached tothe blower base facilitates a strongervacuum than the original model.

The modified version uses the samesampling enclosure as the earlier one(45-cm-diam plastic bucket with itsbottom removed and top-fitted with a1-m-long muslin/fiberglass net). Anadditional muslin sleeve can also beattached to the top netting to provide aside access. Arthropods inside theenclosure are sampled in a downwardspiral beginning with the muslinnetting, followed by the air column, theplant surfaces, and the water columnand soil. Because arthropod popula-tions in rice increase in both speciesrichness and abundance with crop age(at least up to maximum tillering),sampling time should also increase.

Vacuum tests revealed a morepowerful vacuum in the modified thanin the original model (see table); addi-tional vacuum in the modified versionallows easier capture of large inver-tebrates (e.g., rice bugs) inside theenclosure. A comparison of arthropodcatches in the modified Blower-Vac,FARMCOP, D-vac, and sweep netmethods is in pro-gress and will bepublished at a later date. Interestedreaders, however, can consult the tablein Arida and Heong (1992) that

compared arthropod catches from twoD-vac models (backpack and hand-carried), FARMCOP, and the originalBlower-Vac. Because the modifiedBlower-Vac catches more larger sizedinvertebrates than the original model,the modified method is expected torepresentatively sample a wider rangeof species than the original method.

The addition of a vial caddy and theattachment of a tripod to the blowerunit (see figure) allows 2-3 persons tocomfortably operate the unit. Thematerials and the machine blower unitcost approximately US$150, compara-ble in price to the original modelreported in 1992 ($158). For safeoperation, we recommend using anaspirator mask, earplugs, and goggles.

AcknowledgmentThe ingenuity of Mr. Ruben Abuyo inour Insect Ecology Unit helped to makethis improved version possible. ■

Modified Blower-Vac apparatus for sampling invertebrate communities. Arrows indicate theflow of air, water, and invertebrates through the apparatus. (n) = new, (m) = modified partfrom the original Blower-Vac.

Vacuum tests of the original and modified Blower-Vacmachines.

Time triala

Model AvT1 T2 T3

Original 27 23 23 24.3Modified 18 18 19 18.3

aTime (in s) required to fill the water trap.

A new Senegalese thresher/cleaner responds to small-farmer postharvest needs

M.C.S. Wopereis, K.M. Miézan, C. Donovan,A.M. Ndiaye, West Africa Rice DevelopmentAssociation (WARDA), 01 B.P. 2551, Bouake 01,Côte d’Ivoire; and B. Douthwaite, IRRI

Development agents cite harvest andpostharvest constraints as major factorsin the profitability of irrigated rice pro-duction in the Senegal River Valley.Existing systems are expensive, tootime-consuming, or too labor-intensiveduring periods of peak labor demand.An IRRI technology exchange withWARDA has provided a potentialsolution.

WARDA and its partners, with thehelp of an IRRI agricultural engineerand Senegalese expertise, developed aSenegalese prototype of the IRRI-designed TC800 axial flow thresher/cleaner. Here we report on the adapta-tions made, and the technical andfinancial performance. In an accom-panying article (Donovan et al, page41), we report on the collaborative

Rubber stopper

Machineblower

Machine vacuum

Braided tygonhose (n)

Vial

1.5 L plastic bottle (m)

To arthropods(inside enclosure)

Vial rack (n)Cap bag (n)

Woodenstick (n)

Rubber hose

PVC pipe (m)

Screen strainer (m)Elastic cords (m)

Tripod (n)

Rubber stopper

PVC reducer

Screen strainer (m)

H2O only(when open)

40 IRRN 1998

effort that was critical in producing theSenegalese thresher/cleaner, whichwas named ASI.

In 1995, research in the region indi-cated many problems with currentharvest and postharvest technology.Combine harvesters were agingrapidly and were not being replaced.The 1994 devaluation of the regionalcurrency, the CFA franc, increased thecost of imported equipment and spareparts. Plot-level surveys in 1994-95show harvesting dates as much as 2 mopast maturity because farmers waitedfor combine harvesters. Therefore,many farmers shifted to manualharvest and threshing. But, manualmethods are difficult and time-consuming, and extend the period ofexposure of rice crop to hot, drySahelian winds.

The Votex thresher was introducedinto the region during the early 1990s.Machine components for the Votex areimported for assembly, rather thanbuilt in the region. The Votex requires 5or 6 people to operate the machineefficiently for threshing, with an 85%separation rate. Another 6-9 laborersare then needed to sift the strawmanually to recover the remaining15%, as well as to winnow and clean thethreshed rice before bagging. In a 1994survey, farmers complained that theperformance of the Votex was insuffi-cient and the labor requirements toohigh.

In 1995, IRRI donated two stripper/gatherer (SG) systems, with the SG800stripper/harvester and the TC800thresher/cleaner, to WARDA for evalu-ation in West African irrigated rice con-ditions. Testing of the TC800 in Senegalshowed that the machine neededadjustments. For example, the feed trayand blower were too small to handlemanually harvested rice, and themachine was too fragile overall. Therewere no wheels to maneuver themachine in the fields or allow animal-drawn transport. Table 1 lists the majormodifications undertaken during the 21/2 yr of product development.

The performance characteristics andfinancial analysis of the ASI are given in

Table 1. Key adjustments to the original TC800 thresher-cleaner from IRRI for the Senegalese ASI prototype,under guidance from the ADRAO-SAED-ISRA engineering team.

Wheels fittedAbility to pull the machine by animal power addedAll housing assemblies fortifiedSecurity added to protect hands while feeding straw into the machineBlower enlarged (almost twice as large)Air entry to blower made adjustableFeed tray enlargedNumber of discharge louvers increased from 5 to 6Two oscillating screens installed instead of oneAll shaft diameters enlargedRings stabilizing oscillating screens modified to include ball bearings to reduce wear and tearLarger threshing cylinder assembly strengthened and enlarged; diameter increased from 270 to 445 mm andlength from 600 to 885 mmLength of peg teeth in threshing cylinder enlarged from 100 to 150 mmU-beam assembly added to eccentric arm to strengthen and stabilize itInstallment of alternating rows with 8 or 9 peg teeth instead of 5Straw exit in front of machine made adjustableExit for immature and damaged grains brought to opposite side of the machineAuger discharge spout enlarged from 80 to 100 mmGrain auger assembly: distance between auger flights enlarged from 56 to 100 mm12-hp diesel engine added

Table 2. Financial comparison of ASI and Votex machines.

ASI Votex

Technical and financial parametersType of machine Axial flow Tangential flowGrain separation rate (% of rough rice) 97-99 85Daily processing capacity (kg d-1) 6,000 4,300Days worked yr-1 55 55Hours worked d-1 6.00 6.00Rough rice price (US$) 0.17 0.17Charge for processing services (% of rough rice) 10 8Purchase price (US$) 4,138 3,276Fuel consumption (L h-1) 2.0 0.8Workers needed: operators 2 1

laborers (provided by farmers) 4 4Exchange rate (CFA francs US$–1) 580 580

Financial results (in US$)Fixed costs (annual)

Depreciation, interest 1,522 1,228Maintenance and repairs 414 328Taxes, insurance, etc. 20.69 16.38

Total fixed costs yr–1 1,986 1,572Total fixed costs t-1 6.02 6.65Variable costs (annual)

Fuel and oil 679 442Labor 379 190

Total variable costs yr–1 1,058 632Total variable costs t–1 3.21 2.67Total costs yr–1 3,044 2,204Total costs ha–1 41.51 41.94Total costs t–1 9.23 9.32Gross revenues t–1 17.24 13.79Gross revenues yr–1 5,690 3,262Net revenues t–1 8.02 4.47Net revenues yr–1 2,645 1,058Breakeven tonnage (annual) (t yr–1) 142 141Breakeven workdays (d yr–1) 23.7 32.8Debt repayment (yr) 1.0 1.8Net present value 10,780 4,047Internal financial rate of return (%) 65.6 34.1Benefit/cost ratio 1.7 1.4

Source: Donovan C, Douthwaite B. 1997. Financial analysis of the ASI (thresher/cleaner). WARDA, St. Louis, Senegal. (mimeo.)

Vol. 23, No. 2 41

Table 2, including a comparison withresults of the Votex thresher operatingunder similar conditions. The para-meters were chosen to be realisticassessments in the Senegalese context,based on earlier trials and currentwork. With a 12-hp diesel motor and 6workers, the ASI yields 6,000 kg d-1 ofclean rice from manually harvestedproduction in fields with yields ofapproximately 4.5 t ha-1. Under similaroperating conditions, the Votexcapacity is 4,300 kg d-1. The separationrecovery rate was 99% in on-farm testsfor the ASI, much higher than theVotex’s 85%. Thus, with the ASI, totallabor needed is 6 workers. No addi-tional labor is needed for sifting thestraw for unthreshed rice or for win-nowing, because of the efficient separa-tion of rice from straw. Daily perform-

ance for both machines depends on therelative grain to straw proportion, thegrain yield, the moisture content of thestraw, and the speed of input ofharvested crop.

The financial analysis evaluates theprofitability of the machine for a serviceoperator in order to assess potentialdistribution. With a purchase price ofUS$4,138 and the parameters in Table 2,the cost per ton for the ASI is US$9.23,when operator labor is the only laborincluded (the current system). For theVotex, per-ton costs are only slightlyhigher ($9.32). The ASI’s net revenuesare more than double the revenues forthe Votex, primarily because of thehigher processing rate. ASI’s benefit-cost ratio of 1.7 is in the acceptablerange for investments, even with only55 workdays yr-1.

In addition to revenues for the ASIowner, farmers also benefit. The ASIhas a high quality of output andreduces the time for postharvest work.This in turn lowers postharvest grainlosses that arise from delays and loss ofhumidity. Labor for winnowing,cleaning, and sifting, which is usuallysupplied by the farm household, is nolonger needed. The modified ASI helpsreduce calendar constraints to doublecropping, avoid grain losses fromdelays, reduce total harvest and post-harvest costs, and provide localemployment in artisan workshops andagroindustry. In terms of demand,SISMAR, the local manufacturer of theASI, is unable to meet current demandfor the machine. ■

Technology transfer from Asia toAfrica sets the stage for public-and private-sector collaborationin new technology in Senegal

C. Donovan, K.M. Miézan, M.C.S. Wopereis, B.S.Diack, West Africa Rice Development Association(WARDA), 01 B.P. 2551, Bouake 01, Côte d’Ivoire;and B. Douthwaite, IRRI

In the early 1990s, the productivity andprofitability of irrigated systems in theSenegal River Valley began to be ques-tioned. The Senegal River Valley Devel-opment Authority (known as SAED)found that harvest and postharvestconstraints were very important, bothin costs and in grain quality losses.SAED agents therefore turned toresearch, specifically to the SenegaleseAgricultural Research Institute (knownby the French acronym ISRA) and toWARDA to help find alternatives.

In response to WARDA’s request,IRRI donated prototypes of the TC800axial flow thresher/cleaner and theSG800 stripper gatherer. WARDA andISRA modified the TC800, with helpfrom SAED, a local artisan, farmers,and local agro-industry. By November1997, a new thresher/cleaner, known asthe ASI, was released in Senegal. A

companion article by Wopereis et al,page 39, presents the technical andfinancial results. Here, we present thedevelopment process of the ASI, anexample of how research, develop-ment, private industry, and farmers canwork together to meet technologyneeds.

IRRI researchers developed theSG800 and the TC800 to meet the needsof small-scale Asian rice farmers. Earlyin 1995, with the IRRI-donated proto-types and construction plans, IRRI andWARDA collaborated with SAED andISRA to assess the potential of themachines in Sahelian irrigated riceproduction systems. In November1995, during the wet-season harvest, anIRRI agricultural engineer (B. Douth-waite) assisted technicians andmachinists with field tests of themachines. These people developed alist of changes recommended tostrengthen the machine and adapt it,particularly for use with manuallyharvested fields and under wet fieldconditions. A workshop was organizedwith the major stakeholders involved,including farmers, rice millers, localmanufacturers, representatives fromthe development agencies (SAED, etc.),researchers from WARDA and ISRA,

and extension agents, to discuss testresults and future work.

WARDA, SAED, and ISRA decidedto work with a local artisan skilled withagricultural machinery in order tobuild a Senegalese prototype, using theexisting TC800 model and the IRRIplans. One of the challenges was tobuild the machines with locally avail-able materials. The artisan chosen wasskilled with agricultural machinery(including the existing Votex thresher),and his machine shop was sufficiently(and typically) equipped. By the dryseason of 1996-97, testing of the newlocal prototype took place in farmers’fields as well as on WARDA’s researchfarms, with the IRRI engineer againassisting. Farmers participated in theequipment tests, indicating areas ofconcern with the design and operation.The advantage of using both localmaterials and local expertise wasevident during the tests. Alternativescould be tested and minor changesmade rapidly in the field or in thenearby artisan workshop, withoutmajor time delays. Thus, local artisansare in a good position to build, main-tain, and repair the machines in thefuture.

42 IRRN 1998

In 1996, a second workshop washeld to present test results from thelatest thresher prototype and to discussthe commercial production of thethresher/cleaner. To reflect the inte-grated approach to product develop-ment, the name “ASI” was chosen,derived from ADRAO/SAED/ISRA,for the three agencies involved inSenegal (WARDA’s French acronym isADRAO). Then, in addition to theartisans, an agricultural machineryfirm, SISMAR, was given the technicalplans for the thresher/cleaner. Invest-ing its own capital, SISMAR built a new

Polyurethane rollers: a substi-tute for rubber rollers in ricedehuskers

K.P. Sudheer, Indian Agricultural ResearchInstitute, New Delhi; and R. Viswanathan, TamilNadu Rice Research Institute, Aduthurai 612101,Thanjavur District, Tamil Nadu, India

Thermoplastic polyurethane is formedby linking polyol and isocyanate. Aftera reaction, polyol becomes flexible andhighly hydrophobic, and forms hydro-lysis-stable polyurethanes. These twocompounds, when mixed and under-going reactions in different proportionsand conditions, result in a polyure-thane product of various hardness,tensile strength, elongation, andabrasion resistance. This polyurethanehas applications in various industriessuch as automobiles, cold storage, foot-wear, etc. The major quality of poly-urethane products is their abrasiveresistance, which is approximatelythree times that of products made ofrubber. In our study, polyurethanepieces were produced and tested toassess their suitability for replacingrubber rollers in rice dehuskers. Therubber rollers require frequent replace-ment because of the abrasive nature ofthe rice surface.

Polyol and isocyanate were mixed indifferent proportions (100:45, 100:50,100:55, and 100:60) without any airbubbles forming in the mixture. Thismix, poured in a mold for making test

Physical and mechanical properties of polyurethane specimens of various proportions of polyol and isocyanateand curing durations.

Curing Composition Tensile Elongation Hardness Abrasionduration polyol:isocyanate strength (%) (°A) loss

(h) (MPa) (cc 40 m-1)

100:45 6.19 158.50 82 0.35100:50 9.17 170.70 87 0.28

2.5 100:55 10.00 182.90 88 0.24100:60 9.52 161.00 86 0.36

100:45 9.42 161.00 83 0.30100:50 11.20 178:05 89 0.24

3.0 100:55 16.00 197.56 90 0.18100:60 12.43 170.70 87 0.32

100:45 8.00 143.90 83 0.30100:50 9.23 153.66 88 0.20

3.5 100:55 13.38 173.70 89 0.13100:60 10.22 158.50 86 0.22

Commercialrubber roller 12.59 139.00 89 0.59

prototype, and assisted the WARDA/SAED/ISRA team in the tests anddevelopment of recommendations forthe remaining modifications.

The ASI was officially released forcommercial distribution on 5 Novem-ber 1997. SISMAR has already manu-factured machines for use in theSenegal River Valley and is unable tomeet demand. Among the first pur-chasers of the ASI were two Senegalesefarmers—local seed producers—whohad participated in the tests and wereimpressed by the ASI performance rate,ease and safety of use, and quality of

output. Using the collaborative ap-proach, this research and developmenteffort was able to produce an adaptedtechnology that addresses postharvestconstraints with a domestically pro-duced and easily repaired thresher/cleaner. Collaborative research by IRRI,ISRA, SAED, and WARDA was criticalto ensuring a well-built, durable ma-chine. The private sector, both artisansand the agroindustrial firm SISMAR, aswell as farmers and their organizations,will determine the final diffusion of thismachinery in Senegal. ■

pieces, was cured at 90 ± 2 °C in an ovenfor 2.5, 3, and 3.5 h. These pieces weretested for hardness, tensile strength,elongation, and abrasion resistanceusing standard procedures (see table).

The table shows that all the proper-ties required for use in the paddydehusker-rubber roll type increasedwith a curing duration of 2.5 and 3 hand proportions of 100:45, 100:50, and100:55, and decreased with 3.5 h ofcuring duration and proportion of

100:60. The tensile strength, elongation,and hardness of the test pieces were ona par with those of commercially avail-able rubber rollers. The abrasion losswas in the range of 0.13-0.36 cc 40 m–1 vs0.59 cc 40 m–1 for the rubber rollers.Tensile strength, elongation, andhardness were maximum at the pro-portion of 100:55 cured at 3 h with 0.18cc 40 m–1 of abrasion loss. Thiscombination was thus identified as theoptimum. ■

Manuscript preparation. Arrange the note as a brief statement of researchobjectives, a short description of project design, and a succint discussion ofresults. Relate results to the objectives. Do not include abstracts. Do not citereferences or include a bibliography. Restrain acknowledgments.

Vol. 23, No. 2 43

Fertilizer managementResponse of rice hybrids toN sources and time of

application of N and K

K. Surekha, M.N. Reddy, and C.H.M.Vijayakumar, Directorate of Rice Research,Rajendranagar, Hyderabad 500030, India

Although NH4-N has been accepted asthe preferred N source in rice, NO3-Nwas reported to be the preferred sourcein China at late crop growth stageswhen hybrid exhibited a peak demandfor nutrients. A higher percentage ofunfilled grains is common in hybridrices unless the soil nutrient supply cancope with crop demands. Thus,delayed application of N and K coinci-ding with flowering can help realize thepotential of hybrid rices.

We studied the differential responseof recently released hybrids to variousN sources and split application of N

and K. In a Vertisol (typical Pellustert),during the 1994 and 1996 wet season,four hybrids (MGR1, KRH1, APRH1,and APRH2) received the following Ntreatments: 120 kg ha-1 at 33:33:33% atthe basal, tillering, and floweringstages, respec-tively; and 120 kg ha-1

likewise at the basal, tillering, plantinitiation, and flowering stages. Allphosphorus was applied basally (P at26.4 kg ha-1). Potash was either appliedas basal or in two splits (75% basal and25% at flowering). Data on grain andstraw yields and their N uptake wererecorded.

Of the two N sources tested, NH4-Nwas found to be superior for grain yield(5.3 t ha-1) and N uptake (128 kg ha-1).This indicated that the more stableNH4-N held by the soil cation exchangecomplex (CEC) affected yield. Al-though the usefulness of NO3-N forhybrids at the reproductive stage has

been reported, it leached and wasunstable. Three equal N split applica-tions were on a par with four N splits.Irrespective of N sources, a topdressingof N beyond the panicle initiation stagefor short- and mid-duration groups didnot bestow any additional yield advan-tage. Similarly, a split application of K(twice) did not affect either grain yieldor N uptake even though an adequateK supply at flowering was known toimprove grain filling via the efficienttranslocation of photosynthates to thesink. High levels of available K presentat the experimental site might havetaken care of the crop needs.

The present study indicated thatNH4-N is a better source of N forhybrids. In Vertisols with a high CECcontributing to the retention of appliedNH4-N and high available K status,there is no need to resort to delayed Nand K application. ■

Nitrogen response ofKarnataka Rice Hybrid 2

K.M. Devaraju, H. Gowda, and B.M. Raju,Regional Research Station, VC Farm, Mandya571405, Karnataka, India

The development of hybrids, becauseof their heterotic vigor, could increaserice productivity by breaking the yieldbarrier. With this objective, two ricehybrids—Karnataka Rice Hybrid 1(KRH1) and Karnataka Rice Hybrid 2(KRH2)— were released in Karnataka.Hybrids in most crops usually requirean increased use of various nutrients,especially N.

To determine the N requirement inKRH2, a medium-duration rice hybrid,a study was undertaken using IR20 as acheck variety. Experiments were laidout with different levels of N, rangingfrom 0 to 200 kg ha-1, on a sandy loamsoil with a pH of 6.5 and organic C of0.5% in a split-plot design with threereplications during the 1994-96 wetseasons.

KRH2 outyielded IR20 at all levels ofN application, ranging from 0 to 200 kgha-1. In IR20, the tiller number washigher than that of KRH2. The in-creased yield of KRH2 was mainlyattributed to the higher number of pro-ductive tillers, panicle weight, andnumber of filled grains. The nutrients

Integrated pest management — diseases

used for the production of unproduc-tive tillers in IR20 might have beenbetter used in KRH2 to increase seedset, panicle weight, and thus yield. IR20responded to applied N only up to100 kg ha-1. In KRH2, the response toapplied N beyond 100 kg ha-1 wasmarginal and insignificant. ■

The effect of Trichoderma andantifungal agents on rice germi-nation

K. Duncan and Youxan Su, University ofCanterbury, Bag 4800 Christchurch, New Zealand

Rice seeds were found to exhibit veryweak or no germination when sown inan antifungal seed-raising mix made bycultivating the forest green mold,Trichoderma viride, in nutrient-supplemented sawdust from Pinusradiata wood. (P. radiata is a commontemperate plantation softwood that isgrown extensively in southern

hemisphere countries for lumber.)Trichoderma is a genus of pathogenicfungi used widely in horticultureagainst such fungal diseases as silver-leaf in fruit trees. The aim of the anti-fungal mix is to provide a naturalcontrol agent against damping down orseedling wilt diseases using a cheapand abundant plant waste.

Common vegetables (carrot, rad-dish, lettuce, tomato) germinated andgrew extremely well in the antifungalmix with no disease mortality. Butwhen four rice varieties (IR60 HD 105,PSBRe4 HD 437, PSBRe2 HD 429, andPSBRc10 HD 439) were tried in the new

44 IRRN 1998

Germination and growth of lots of 100 seeds of rice variety PSBRc4 HD 437 on filter paper subjected toantifungal treatments.

Treatment

Factor None 1% hydrogen 1% copper 1% benomyl 0.01%peroxide oxychloride benlate Spirulina platensis

Germination (%) 98 98 98 100 100Total wet weight (g) at 20 d postplanting 5.627 6.305 6.496 5.652 7.730Maximum stem height (mm) at 19 d postplanting 50 60 57 53 59Max. root length (mm) at 6 d 20 25 35 22 30

mix, the germination percentage rangedfrom 0 to 2%. Growth of those seeds thatdid germinate was extremely poor.

This unexpected result could beexplained by T. viride being toxic to ricebut not to the other plants tested. Themechanisms for this toxicity could beeither exosomatic, through the ricebeing especially sensitive to T. viridetoxins released in the rhizosphere, orendosomatic, through the habit of riceof phagocytizing soil organic materialsin heterophagous nutrition—toxins ofT. viride origin might be brought into therice root to make intimate connectionwith the rice cytoplasm. Thus, T. viridewould especially affect rice but notother plants that lack the phagocytichabit.

An alternative hypothesis is that ricemight have a specific requirement for amycorrhizal symbiont that is killed bythe T. viride, hence the poor or zerogermination.

In an attempt to test these hypo-theses, trials were conducted in which100 seed lots of rice were germinated ondamp filter paper after being treatedwith dilute solutions of antifungicides.Copper oxychloride, benlate, andhydrogen peroxide were used in weak

concentrations that did not affect thegermination of other plant species(barley, oat, and wheat). One hundredseed lots were also germinated in thepresence of a small amount of Spirulinaplatensis, a commonly available cyano-bacterium. The experiments wererepeated with dehulled rice seeds. Threereplicates of all trials were conducted.

Compared with the controls, all theantifungicides stimulated growth.Germination and growth were en-hanced over the controls when rice wasgerminated in the presence of S. platensis(see table).

These results suggest that micro-organisms can affect the early growth ofrice. In practical management, fungi-

cides can improve germination. Seed-raising mixes contaminated withTrichoderma spp. or other fungi that areinhibitors to growth should be avoided.

Another interesting result from thiswork is that germination and growthmight be enhanced if a small quantityof S. platensis is added to the seeds atplanting time. S. platensis is a N2-fixingphotoautotroph that grows actively indamp conditions, thus possibly in-creasing the N available to the seed-lings as well as providing an additionalcarbon source. It also contains powerfulantifungal and antiviral compoundsthat could protect the seed andemerging seedling from attack bymicroorganisms. ■

Integrated pest management — insects

Evolving insecticide use andpractices at IRRI

M.A. Bell, K.L. Heong, G. Pateña, andT. Clemeno, IRRI

With growing awareness of theconcerns associated with pesticide(especially insecticide) use, IRRI hasacted over the past decade to play alead role in both reducing insecticideuse and promoting safe pesticide usefor people and the environment. Inaddition to the direct steps taken at theIRRI experiment station, a book hasalso been produced by IRRI (in 1995)addressing areas of concern related topesticides, farmer health, and the riceenvironment. The improvements madeat the IRRI experiment station include

1. Implementation of a trainedpesticide applicators (TPA)program (1987).

2. A ban on World HealthOrganization (WHO)Category One pesticides(1989).

3. Implementation of a regularscheme to monitor insect,disease, and weed populationsin fields (1991).

4. Endorsement of integratedpest management (IPM) as thestandard for pest managementat the IRRI station (1995).

5. Further changes in applicationmethods and monitoring offields (1997).

We detail each of these steps below.1. Trained Pesticide Applicators

program (1987)To improve the safe use of pesticides,IRRI introduced a system of “author-ized pesticide applications” in 1987.(The name was subsequently changedto trained pesticide applicators, TPA.)The system includes guidelines for(1) certification process (training), (2)routine medical examinations andhealth reporting, (3) use of personalprotective equipment, (4) job rotation,(5) decontamination facilities, and (6)safe product disposal.

This system ensures that pesticideapplications are made by trained staff,resulting in improved application withbenefits to both users and the environ-ment.

Vol. 23, No. 2 45

2. Ban on the use of Category Onepesticides (1989)

In 1989, IRRI management banned theuse at the IRRI experiment station ofproducts defined by WHO as CategoryOne such as parathion, that is, productsthat are extremely and highly hazard-ous to human health). Coming after theimplementation of the TPA system, thismove had further benefits for bothusers and the environment.

3. Implementation of regular pestmonitoring (1991)

To improve the basis for decidingwhether a pesticide application wasnecessary, IRRI introduced a regularmonitoring system in 1991. With thissystem, all fields and glasshouses weremonitored weekly by trained staff todetermine insect, disease, rat, and weedpressures; the golden apple snail(Pomacea sp.) was monitored atplanting. Critical pest levels forrecommended control wereestablished. If a doubt arose about theneed for an application, the monitoringteam would revisit a field and, wherenecessary, call upon expertise fromIRRI’s Entomology and PlantPathology Division.

4. Integrated pest management as astandard (1995)

The IRRI Experiment StationCommittee (chaired by the deputydirector general for research and withdivision heads as members) made IPMthe standard for pest control on theIRRI experiment station in 1995. Thegrowing knowledge base for IPM ofinsects indicates that insecticideapplications during the initial stages(e.g., first 30-40 d) of the rice crop aregenerally unnecessary. It is worthmentioning that although IPM is thestandard, more intensive insect controlis permitted depending on theexperimental objectives and/orsusceptibility of the materials in thetrial.

5. Further changes in application andresearcher monitoring (1997)

With individual researchers becomingincreasingly aware of the benefits ofIPM, the responsibility for fieldmonitoring was given to the individualscientists in 1997. The experimentstation retains expertise for validatingmonitoring results as required.In 1997, experiment station staffmembers introduced boom sprayingwhere possible and the use of acontrolled pressure valve (for backpacksprayers). Both practices improve theuniformity of spray applications andimprove safety for users. Recentimprovements in land leveling will alsohelp improve pest control (especiallyfor weeds and the golden apple snail).

In the IRRI glasshouses, pesticideuse has been reduced by implementingan annual month-long greenhouseshutdown (thus breaking insect lifecycles) and using colored sticky trapsand milder hand-spray insecticidalsoaps. Biological control agentsrepresent a further option to bepursued.

Changes in pesticide useInsecticide use (active ingredients ha-1

season-1) has dropped dramatically atIRRI in recent years, especially whenIPM was introduced in 1995 (seefigure). Both fungicide and mollusci-cide use have also fallen slightly.

Based on informal observations,reduced insecticide use at IRRI has

resulted in increases in the number ofbirds and beneficial insects seen infields. In addition, field observationshave shown that control of the ricewhorl maggot (Hydrellia philippina) anddefoliators such as leaffolders (e.g.,Cnaphalocrocis medinalis) is rarelyrequired. Planthoppers (e.g., Nilapar-vata lugens) have only been a problem infields that were sprayed early, and greenleafhoppers (GLH-Nephotettix virescens)have occasionally been a problem, butprimarily only when rice tungro virusdisease has been present (GLH beingthe primary vector of tungro). Rice bug(Leptocorisa oratorius) is occasionally aproblem on crops planted out ofsynchrony with other crops.

ChallengesThe practices introduced at IRRI haveled to a reduced and safer use of pesti-cides (especially insecticides), withbenefits to the environment and IRRIstaff. But several challenges remain,including better systems for stem borermonitoring and control, better under-standing of the effects of stem borerdamage at high yield potential levels,improved control of the golden applesnail with reduced use of molluscicide,and mechanization of pesticide appli-cation (boom spraying and the use oftractor-mounted sprayers with specialnarrow-design wheels), and thusfurther reductions in pesticide use.

Stem borers also merit attention.They are more problematic to control

Use of pesticides (kg ai ha-1) at IRRI experiment station (DS = dry season, WS = wet season).

Active ingredients (kg ha-1)

2.5

2

1.5

1

0.5

0

Fungicide

Insecticide

Molluscicide

Season

1993DS

1993WS

1997WS

1997DS

1996WS

1996DS

1995WS

1995DS

1994WS

1994DS

46 IRRN 1998

than other pests because once thedamage is apparent, it is essentially toolate to effectively control the pest; thus,farmers and researchers may betempted to use prophylactic applica-tions for “safety.” Infestations of stemborer during the 1996 dry season were

Integrated pest management — weeds

the highest seen for many years(infestation varied from a 0 to 8%proportion of total panicles infested).Although the damage lookedspectacular, it has been repeatedlyshown that damage at low to moderate

yield levels has little to no effect on cropyield. Data presently being collectedalso suggest that plants retain sufficientyield component plasticity even athigher yield levels (>8 t ha-1) tocompensate for damaged panicles. ■

Butachlor safener combinationsfor weed control in direct-seededpuddled rice

N.N. Angiras and S.S. Rana, Department ofAgronomy, Himachal Pradesh KrishiVishvavidyalaya, (HPKV), Palampur 176062,India

Weed competition is a major constraintto the productivity of broadcast-seededrice. Although manual weeding con-trols weeds effectively, because of theabsence of rows, it is difficult, time-consuming, and costly. In addition,labor is scarce. Butachlor, the recom-mended herbicide, controls weedseffectively, but is toxic to germinatingrice seedlings. The use of a safener, anantidote, protects rice seedlings frombutachlor injury effectively withoutany adverse effect on weed control.

A field experiment was conductedduring the 1994 and 1995 wet seasons atHPKV, Palampur (32° 6' N, 76° 3' E,

1,290 m altitude). The soil of the test sitewas silty clay loam, with pH 6.9, 1.2%organic C, 364 kg available N ha-1, 21 kgavailable P ha-1, and 326 kg available Kha-1. Sprouted seeds of varietyHPU2216 were sown at 100 kg ha-1 bythe broadcast method onto puddledsoil on 19 Jun 1994 and 26 Jun 1995. Thetreatments (see table) were evaluated ina randomized block design with threereplications. The data analyzed bystandard analysis of variance andtreatment means were compared withthe LSD test at the 5% level.

Weed samples were taken from a25 cm × 25 cm area and oven-dried at70 °C, and dry weight was recorded.Grain yield was recorded from a 7.5-m2

net plot area. Plant samples from a 0.5-m2 area were taken at harvest to recordyield components. The emergencecount was recorded from a 0.5-m2 areaat 18 d after sowing (DAS). The domi-nant weeds were Echinochloa crus-galli,

E. colona, Ammania baccifera, Scirpus sp.,Cyperus iria, C. difformis, C. esculentus,Aeschynomene indica, Monochoriavaginalis, and Commelina forskalli.

Butachlor without safener applied at2 or 7 DAS caused toxicity to rice seed-lings, but this was significantly lesswhen applied at 7 DAS than at 2 DAS.Other butachlor safener combinationsalso caused acute toxicity symptoms,but rice plants were able to recover.Butachlor at 2 kg ha-1 + safener at 0.188kg ha-1 (7 DAS) was the most effectivecombination for yield and weedcontrol. Rice yield of the butachlor at1.5 kg ha-1 + safener at 0.188 kg ha-1 (2and 7 DAS) treatments was equal tothat of the crop that received two handweedings. At each time of application(i.e., 2 or 7 DAS), butachlor with safenerat 0.094 or 0.188 kg ha-1 controlledweeds effectively and increased ricegrain yield significantly over its appli-cation without safener (see table). ■

Effect of butachlor safener combinations on emergence count, panicles m-2, panicle weight, grain yield, and weed dry weight in direct-seeded puddled rice.

EmergenceTime of count Panicles Panicle Grain Weed dry weight

Treatment appli- (kg ai ha-1) (g m-2) (no. m–2) weight (g) yield (t ha–1) (g m–2)cation(DAS)a 1994 1995 1994 1995 1994 1995 1994 1995 1994b 1995

Butachlorc 2 1.5 199 148 324 168 1.53 1.58 3.4 1.4 5.54 (30.3) 123Butachlor + safenerd 2 1.5 + 0.094 292 239 420 234 1.68 1.75 4.5 2.7 3.24 (10.1) 77Butachlor + safener 2 1.5 + 0.188 316 265 467 248 1.75 1.81 5.0 3.0 2.41 (6.8) 43Butachlor 7 1.5 257 195 373 195 1.55 1.61 4.0 2.0 4.76 (25.5) 81Butachlor + safener 7 1.5 + 0.094 323 261 427 242 1.71 1.77 4.5 3.0 3.93 (15.1) 47Butachlor + safener 7 1.5 + 0.188 342 281 463 261 1.77 1.88 4.9 3.2 3.37 (11.0) 31Safener 2 0.188 368 324 327 188 1.35 1.40 3.4 1.2 8.97 (80.2) 399Butachlor + safener 7 2.0 + 0.188 338 272 492 295 1.79 1.92 5.2 3.7 1.10 (0.8) 12Hand weeding twice 25 & 45 – 365 324 472 264 1.83 1.95 5.0 3.3 3.63 (12.7) 27Unweeded check – 361 329 296 184 1.32 1.39 3.0 1.2 9.58 (91.7) 406 LSD (0.05) 25 40 59 40 0.17 0.20 0.6 0.6 1.15 21

aDays after sowing. bData transformed to x + 0.5 transformation. Values in parentheses are the means of original values. cFormulation: Trapp 50 EC. dTrade name S901A, chemical name 4,6dichloro-2-phenyl pyrimidine.

√

Vol. 23, No. 2 47

Soil microbiology

Effect of immobilization in different solid ma-trices on ammonia excretion by A. variabilis-SAo and N. muscorum-DOH. PUF = poly-urethane foam, SCW = sugarcane waste,PW = paper waste.

Effect of immobilization of N2-fixing cyanobacteria on solidmatrices and their influence onN2-fixing activity and ammoniaexcretion

S. Suresh Babu and S. Kannaiyan, Department ofAgricultural Microbiology, Tamil Nadu Agricul-tural University, Coimbatore 641003, Tamil Nadu,India

Immobilization is the process of attach-ing cells or their constituent biocata-lysts to a solid matrix so that they donot move independently when placedin a fluid environment. Immobilizationcould be carried out by physical meanssuch as adsorption or entrapment in agel or foam matrix or by chemicalmethods such as covalent bonding.

In the present study, solid matricessuch as polyurethane foam (PUF),sugarcane waste, and paper waste wereused for immobilizing N2-fixingcyanobacteria Anabaena variabilis andNostoc muscorum. PUF was cut into 0.5-cm pieces before use. Sugarcane wastewas treated with 0.5% sodiumhydroxide to remove the phenols andthen thoroughly washed before use.

PUF at 1.5 g, sugarcane waste at 2.0g, and paper waste at 2.0 g were eachplaced in six 250-mL volumetric flaskscontaining 150 mL of N-free BG-11medium. These were sterilized in anautoclave at 15 lb pressure at 120 °C for30 min and then cooled to roomtemperature prior to inoculation of oneset with 5 mL of actively growinghomogenized cyanobacterial culturesof A. variabilis-SAo and another set withN. muscorum-DOH. The cultures wereincubated under nethouse conditionsfor 4 wk at 28 ± 1 °C with 3,000 lux lightintensity.

Both N2-fixing cyanobacterialcultures were also grown under free-living conditions with the same growthconditions as immobilized cyano-bacterial cultures. After 4 wk, theculture filtrate from each treatment wascollected and the ammonia excreted

Effect of immobilization in different solid matrices onnitrogenase activity of A. variabilis-SAo and N.muscorum-DOH.

Nitrogenase activitya

Solidmatrix A. variabilis-SAo N. muscorum-DOH

Free-living 134 174Polyurethane foam 247 344Sugarcane waste 300 428Paper waste 374 235 SE 7 9 SEd 10 12 CE 20 24

an moles of ethylene produced h-1 g-1 dry weight biomass.

was estimated at weekly intervals. Thesolid matrices were used to determinenitrogenase activity via the acetylenereduction assay method.

Anabaena variabilis-SAo and N.muscorum-DOH immobilized in solidmatrices such as PUF, sugarcane waste,and paper waste have registered signi-ficantly higher nitrogenase activity thanunder free-living conditions. N. mus-corum-DOH recorded relatively highernitrogenase activity than A. variabilis-SAo in free-living and immo-bilizedconditions except in the paper wasteimmobilized treatment (see table).

Ammonia excretion was more insugarcane waste than in PUF and paperwaste immobilized and free-livingcyanobacterial cultures. Ammoniaexcretion by both A. variabilis-SAo andN. muscorum-DOH was maximum onthe 14th day and lowest on the 28th dayof inoculation (see figure). The highernitrogenase activity under the immobi-lized state could be due to the coloniza-tion and accumulation of a highernumber of algal cells.

The solid matrices also providefavorable conditions for colonization ofcyanobacteria on the surface of paperwaste and also inside the solid matricesin PUF and sugarcane waste. Inte-restingly, the colonization of cyano-bacteria on the surface could play a vitalrole in increasing N2-fixing activity andammonia production.

Financial assistance by EEC,Belgium, is gratefully acknowledged. ■

Routine research. Reports ofscreening trials of varieties,fertilizer, cropping methods, andother routine observations usingstandard methodologies to estab-lish local recommendations are notordinarily accepted. Examples aresingle-season, single-trial fieldexperiments. Field trials should berepeated across more than oneseason, in multiple seasons, or inmore than one location as appropri-ate. All experiments should includereplications and an internationallyknown check or control treatment.

Ammonia excretion(n moles mL-1) A. variabilis-SAo

N. muscorum-DOH

CD (P = 0.05) = 8.17

7 14 21 28

140

120

100

80

60

40

20

0

Free-living PUF SCW PW

CD (P = 0.05) = 6.75Days

160

140

120

100

80

60

40

20

07 14 21 28

48 IRRN 1998

Socioeconomic impactEconomics of hybrid riceseed production in seed

growers’ fields in Karnataka,India

Honnaiah, B.V. Chandra, R.M. Radhakrishna, andS. Lingaraju, Regional Research Station, VC Farm,Mandya 571405, Karnataka, India

The importance of the contribution ofIndia’s agricultural sector to nationalincome, employment, and generaleconomic development has been wellrecognized. The introduction andadoption of technological innovations isone way to increase the productivity.Hybrid rice varieties have been releasedin Karnataka. Hybrid seed productionis an important component of thespread of hybrid rice. We studied theeconomics and employment generationof seed production in seed growers’fields.

A farmer in Mandya District grewhybrid rice, KRH1, for the KarnatakaState Seed Corporation on a contractual

basis, on 0.5 ha under irrigated condi-tions during the 1996 dry season. Theparental seeds were supplied to thefarmer from the Regional ResearchStation through the corporation.Standard seed production techniqueswere followed. Trained executives of thecorporation provided technicalguidance.

The hybrid seed yield was 1.5 t ha-1,with a total gross income of Rs 60,000ha-1 (US$1=Rs 35), and the corporationprocured seed at the rate of Rs 40 kg-1.The cost of cultivation was Rs 24,000ha-1, including transportation and pro-cessing. Thus, the seed production costwas Rs 16 kg-1.

For the total cost of cultivation perhectare, the major costs were labor andbullock pairs (Rs 2,375), followed byfertilizer and plant protection chemi-cals (Rs 5,985), GA3 (Rs 2,800), seeds(Rs 1,875), and transportation and pro-cessing (Rs 975). The special techniquesof seed production, such as leaf clip-ping, GA3 application, and rope pulling,

Economics of commercialhybrid rice cultivation

Honnaiah, S. Gangadharaiah, S. Lingaraju, andR.M. Radhakrishna, Regional Research Station,VC Farm, Mandya 571405, Karnataka, India

We studied the economics of the com-mercial cultivation of KRH1 in MandyaDistrict, Karnataka. This study involved2 farmers during the 1994 dry season, 5farmers during the 1994 wet season, and10 farmers during the 1996 dry season.Means were used for the economicanalysis. Rasi was used as a check varie-ty. Plot size ranged from 0.4 to 1.0 ha.

KRH1 yielded 6.5 t ha-1 on average,whereas Rasi yielded 5.0 t ha-1. KRH1recorded a yield advantage of 1.5 t ha-1.The cost of cultivation of KRH1 was Rs13,885 (US$1=Rs 35) vs Rs 12,170 forRasi. The increased cultivation cost forKRH1 was due to the increased cost ofseed and labor for nursery management

cost Rs 3,825 ha-1. The cost of additionaloperations such as roguing, separateharvest of A and R lines, among others,was Rs 450 more than for general seedproduction. For seed production, 150more men and 55 more women laborerswere required for hybrid rice than forinbred varieties. The major operationsfor these laborers included roguing,rope pulling, leaf clipping, planting,and harvesting. Because the groweralso received Rs 2,430 for straw and Rs8,750 through the sale of restorer grains,gross income was Rs 71,180 ha-1. Thus,the grower received a net income ofRs 47,180 ha-1. This represents a benefit-cost ratio of 1.97.

The study thus indicated that theseed grower can obtain a net profit ofRs 47,180 ha-1 in hybrid rice seedproduction. Seed production cost Rs16kg-1. This activity also generated addi-tional rural employment for 150 menand 55 women. Hybrid seed produc-tion is a highly profitable enterprise forwilling and innovative farmers. ■

and planting. Gross income from KHR1was Rs 37,500 vs Rs 28,500 from Rasi.KRH1 had an additional gross return ofRs 9,000 ha-1, which resulted in an addi-tional net return of Rs 7,285 ha-1.

It was also observed that theadditional labor required for KRH1lessened with each season as farmers

became more acquainted with new prac-tices such as planting single seedlingshill-1, among others. The benefit-costratio for KRH1 was 1.70; it was 1.34. forRasi. Because the cultivation of KRH1resulted in an additional net profit of Rs7,285 ha-1, farmers were impressed by itsprofit and accepted the new hybrid. ■

Scholarships availableIRRI is pleased to announce the avail-ability of an anticipated 14 scholar-ships to be awarded during 1999 tosupport highly qualified scientists fromrice-growing developing countriesinterested in pursuing a graduatedegree in areas related to rice science.These scholarships include thoseprovided by IRRI (PhD scholarshipsonly) as well as scholarship funds thatIRRI administers for other agencies,

Announcementprimarily the Asian Development Bank(ADB)-Japan scholarship fund (MS andPhD scholarships).

IRRI’s overall research effort isdivided into seven programs: fourfocus on ecosystems (irrigated, rainfed,upland, flood-prone), one addressesissues that cut across rice ecosystems(cross ecosystems), one focuses ongenetic conservation of rice (geneticresources), and one deals with capacitydevelopment of IRRI’s partners

Vol. 23, No. 2 49

(accelerating impact of rice research).Scholarship slots are allotted to eachprogram and scholars are selected towork in areas of critical importance toeach program. For 1999, scholarsinterested in the following broadthematic areas by IRRI researchprogram will be given firstconsideration:

Irrigated rice1. Modeling and measuring respi-

ration losses in rice.2. Modeling and measuring tempe-

rature effects on high-yielding rice.3. Modeling and measuring the

functional balance of shoots androots in rice.

4. Root morphology and rootphysiology for yield improvement.

5. Long-term trends in Asian ricefieldsand productivity at experimentstations and in farmers’ fields; whatis the evidence of degradation of thesoil resource base.

6. Physiological changes of rice plantsdamaged by insects.

7. Sociology of pest managementdecisions among rice farmers.

8. Weed ecology.9. Integrated weed management in

intensive rice production.

Rainfed lowland rice1. Economic risk analysis.2. Cross tolerance mechanisms for

drought and submergence.3. Physiology of drought tolerance

mechanisms related to root growthand water extraction in anaerobic/aerobic conditions of rainfedlowlands.

4. Genetics of physiologicalmechanisms for drought tolerance.

5. Establishment and vigor of dry-seeded rice.

6. Zero and reduced tillage and cropestablishment.

7. The role of microflora and effects ofdrought on microscale soil nutrientsupply.

8. Weed dynamics and biology inrainfed systems.

9. Dynamics and management ofcarbon and nitrogen in rainfedlowlands.

Upland rice1. Socioeconomic aspects of land use

dynamics in the uplands.2. P dynamics and management in the

acid upland soils.3. Understanding P × N interactions

and effects on C, N, and P cycling inthe uplands.

4. Economic sustainability of uplandrice systems.

5. Refining P coefficients fordevelopment/improvement of Pdecision aids for the uplands.

6. Soil- and slope-specific analyses ofupland rice productivity.

Flood-prone rice1. Effect of rice-shrimp farming

systems on the sustainability offlood-prone coastal ricelands.

Cross ecosystems1. Ex ante evaluation of the impact of

biotechnology applications.2. Decision support systems for crop

and pest management underdifferential water stresses.

3. Techniques for using pestbiodiversity.

4. Yield gap analysis.5. The role of scales in ecoregional

research.6. Applications of biotechnology for

crop improvement.

Genetic resources1. Biology and genetics of wild rice.2. Biosystematics of the genus Oryza.3. Dynamic conservation of rice genetic

resources.4. G × E analysis to study adaptation of

rice to stress environments.

Accelerating impact1. Impact evaluation of using the leaf

color chart for N management in ricein the Mekong Delta area of Vietnam.

2. Assessing the potential of controlled-release fertilizers for rice systems inIndia and the Philippines.

3. Evaluation of the decentralizedproduction and distribution of ureabriquettes and their use by farmersin Bangladesh.

4. Assessing farmer adoption patternsand constraints of wet-seedingmethods for rice in two selectedAsian countries.

5. Ex ante evaluation of adoptinghybrid rice in India and Bangladesh.

6. Evaluation of the public, private,cooperative, and NGO sectororganizations for the disseminationof new knowledge and technologiesto small-scale rice farmers in Asia.

7. Ex ante evaluation of the expectedimpact of two-way ham radiocommunication for linking farmerswith extension agencies andresearchers in an intensivelycultivated area.

8. Evaluating the potential impact ofdistance education materialsincluding video films for thedissemination of knowledge-intensive technologies.Although IRRI is not a degree-

granting institution, it does supportcandidates enrolled in one of the manyuniversities throughout the world withwhich IRRI has a formalized memoran-dum of agreement. IRRI scholars mustfulfill all requirements of that universi-ty and IRRI. What IRRI provides is theopportunity to conduct thesis researchunder the guidance and supervision ofan experienced IRRI scientist on a topicof global importance to rice science.Quite often, this involves being givenaccess to facilities and resources notreadily available to students from deve-loping countries. Upon successfuldefense of the research work, the scho-lar is conferred a degree through thecollaborating university.

Two kinds of scholarships aregranted—full and thesis only. Fullscholarships are generally granted onlyto qualified students from countrieswith relatively less developededucational systems. IRRI has definedthese to include developing countriesother than Bangladesh, China, India,Indonesia, Korea, Malaysia, the

50 IRRN 1998

Philippines, and Thailand. A fullscholarship provides support forcompletion of courseworkrequirements as well as thesis research.Scholars receiving a full scholarshipnormally attend the nearby Universityof the Philippines Los Baños for theircoursework requirements beforeundertaking thesis research throughIRRI, although this is not mandatory.Full PhD scholars must complete theirprogram within 3 1/2 yr. MS scholarsare allotted 2 1/2 yr.

A thesis-only scholarship is madeavailable to scholars who have finishedtheir formal studies and are ready toconduct thesis research. Scholars fromBangladesh, China, India, Indonesia,Korea, Malaysia, the Philippines, andThailand are eligible for thesis researchscholarships only. Thesis-only PhDscholars must complete their programwithin 2 1/2 yr.

Selection for all grants is highly com-petitive and applications must beendorsed by the applicant’s institution.IRRI encourages the application ofwomen candidates and IRRI selectionprocedures stipulate that “equally qua-lified women candidates will be givenpreference in IRRI’s nondegree anddegree training programs and work-shops.” Scholarships may be awardedto individuals working in governmentorganizations, universities, and NGOs.

Application procedureCandidates should be between 25 and45 yr old; have at least 2 yr of relevantprofessional experience; be proficientin English; and be physically fit, assupported by a medical report.Interested and qualified individualsshould contact the Head of IRRI’sTraining Center expressing interest inapplying for a scholarship andrequesting an application form. Forcandidates with access to the Internetand a WWW browser, applicationforms can also be printed from IRRI’sWeb page <http://www.cgiar.org/irri/>. The Head of the Training Centercan be contacted at the followingaddress and/or numbers:

HeadTraining CenterInternational Rice Research InstituteP.O. Box 933, 1099 Manila, PhilippinesTelephone: (63-2) 845-0563Fax: (63-2) 891-1292email: R.Raab@cgnet.com

The completed application formshould be sent to the Head of theTraining Center before 1 Nov 1998. The

Errata• In the third column of page 26, "...

with 450 m3 water . . ." should read"4,500 m3 water."

form should include a clear indicationof the thematic research area in whichthe candidate is interested in working,the kind of scholarship requested (full,thesis-only, Ph D, MS), and anendorsement from the applicant’semployer.

IRRI’s Scholarship Committee willscreen all applications and make finalscholarship awards before the end of1998. ■

• In Vol. 23, No. 1 (1998), on page 18,the first author's name should readK. L. Sahrawat, not L. Sahrawat.

• On page 17, the figure did not clearlyshow the bars representing meantungro disease incidence. Therevised figure is shown below.

70

60

50

40

30

20

10

0

Infection (%)/GLH (no.)

BB infSS infVis infGLH pop

IR62 IR64 IR68305-18-1 IR71026-3-2-4-3-5-2

IR69705-1-1-3-2-1

IR71030-2-3-2-1

Test lines/varieties

Mean tungro disease incidence ( ), infection with rice tungro bacilliform ( ) and spherical ( )viruses, and number of green leafhoppers ( ) per 10 sweeps of a 30-cm-diameter insect net inadvanced breeding lines and varieties at six locations in India, Indonesia, and the Philippines inreplicated field trials in 1995 and 1996.

Vol. 23, No. 2 51

Instructions for contributorsNOTES

General criteria. Scientificnotes submitted to the IRRN forpossible publication should● be original work,● have international or pan-national relevance,● be conducted during theimmediate past three years orbe work in progress,● have rice environmentrelevance,● advance rice knowledge,● use appropriate researchdesign and data collectionmethodology,● report pertinent, adequatedata,● apply appropriate statisticalanalysis, and● reach supportable conclu-sions.

Routine research. Reports ofscreening trials of varieties,fertilizer, cropping methods,and other routine observationsusing standard methodologiesto establish local recommenda-tions are not ordinarily ac-cepted. Examples are single-season, single-trial fieldexperiments. Field trials shouldbe repeated across more thanone season, in multipleseasons, or in more than onelocation as appropriate. Allexperiments should includereplications and an internation-ally known check or controltreatment.

Multiple submissions. Nor-mally, only one report for asingle experiment will beaccepted. Two or more itemsabout the same work submittedat the same time will bereturned for merging. Submit-ting at different times multiplenotes from the same experi-ment is highly inappropriate.Detection will result in therejection of all submissions onthat research.

IRRN categories. Specify thecategory in which the notebeing submitted should appear.Write the category in the upperright-hand corner of the firstpage of the note.

GERMPLASM IMPROVEMENTgenetic resourcesgeneticsbreeding methodsyield potentialgrain qualitypest resistance

diseasesinsectsother pests

stress tolerancedroughtexcess wateradverse temperatureadverse soilsother stresses

integrated germplasm improve-ment

irrigatedrainfed lowlanduplandflood-prone (deepwater and tidal wetlands)

seed technology

CROP AND RESOURCEMANAGEMENT

soilssoil microbiologyphysiology and plant nutritionfertilizer management

inorganic sourcesorganic sources

crop managementintegrated pest management

diseasesinsectsweedsother pests

water managementfarming systemsfarm machinerypostharvest technologyeconomic analysis

ENVIRONMENTSOCIOECONOMIC IMPACTEDUCATION AND COMMUNI-

CATIONRESEARCH METHODOLOGY

Manuscript preparation.Arrange the note as a briefstatement of research objec-tives, a short description ofproject design, and a succintdiscussion of results. Relateresults to the objectives. Do notinclude abstracts. Do not citereferences or include abibliography. Restrain acknowl-edgments.

Manuscripts must be inEnglish. Limit each note to nomore than two pages of double-spaced typewritten text. Submitthe original manuscript and aduplicate, each with a clear copyof all tables and figures. Authorsshould retain a copy of the noteand of all tables and figures.

Apply these rules, asappropriate, in the note:

● Specify the rice productionecosystems as irrigated, rainfedlowland, upland, and flood-prone(deepwater and tidal wetlands).● Indicate the type of riceculture (transplanted, wetseeded, dry seeded).● If local terms for seasons areused, define them by character-istic weather (wet season, dryseason, monsoon) and bymonths.● Use standard, internationallyrecognized terms to describerice plant parts, growth stages,and management practices. Donot use local names.● Provide genetic background fornew varieties or breeding lines.● For soil nutrient studies,include a standard soil profiledescription, classification, andrelevant soil properties.● Provide scientific names fordiseases, insects, weeds, andcrop plants. Do not use commonnames or local names alone.● Quantify survey data, such asinfection percentage, degree ofseverity, and sampling base.● When evaluating susceptibility,resistance, and tolerance,report the actual quantificationof damage due to stress, whichwas used to assess level orincidence. Specify the measure-ments used.

● Use generic names, not tradenames, for all chemicals.● Use the International Systemof Units for measurements. Forexample, express yield data inmetric tons per hectare (t ha–1)for field studies. Do not uselocal units of measure.● Express all economic data interms of the US$. Do not uselocal monetary units. Economicinformation should be pre-sented at the exchange rateUS$:local currency at the timedata were collected.● When using acronyms orabbreviations, write the namein full on first mention, followedby the acronym or abbreviationin parentheses. Use theabbreviation thereafter.● Define any nonstandardabbreviations or symbols usedin tables or figures in a foot-note, caption, or legend.

Each note can have nomore than two tables and/orfigures (graphs, illustrations, orphotos). All tables and figuresmust be referred to in the text;they should be grouped at theend of the note, each on aseparate page. Tables andfigures must have clear titlesthat adequately explain thecontents.

Review of notes. The IRRNeditor will send an acknowledg-ment card or an e-mailmessage when a note isreceived. An IRRI scientist,selected by the editor, reviewseach note. Reviewer names arenot disclosed. Depending onthe reviewer’s report, a notewill be accepted for publication,rejected, or returned to theauthor(s) for revision.

Comments. If you have com-ments or suggestions about theIRRN, please write to the editor.

Mailing address. Send notesand correspondence to theIRRN Editor, IRRI, P.O. Box 933,Manila 1099, Philippines.Fax: (63-2) 845-0606E-mail: b.hardy@cgnet.comHome page: http://www.cgiar.org/irriRiceweb: http://www.riceweb.orgRiceworld: http://www.riceworld.org

INTERNATIONAL RICE RESEARCH INSTITUTE

P. O. Box 933, 1099 Manila, Philippines

Printed Matter

ISSN 0115-0944