Vous êtes sur la page 1sur 52

International Rice Research Notes

March 1995
The International Rice Research Notes (IRRN) expedites communication among scientists concerned with the development of improved technology for rice and ricebased systems. The IRRN is a mechanism to help scientists keep each other informed of current rice research findings. The concise scientific notes are meant to encourage rice scientists to communicate with one another to obtain details on the research reported. The IRRN is published quarterly in March, June, September, and December by the International Rice Research Institute; annual subject and variety indexes are also produced. The IRRN is divided into three sections: notes, news about research collaboration, and announcements.


Germplasm improvement Germplasm improvement

Genetics Analysis of yield and yield components in rice 4 General combining ability for kernel traits in rice 4 Evaluation of rice diallel crosses in semidry condition by combining ability analysis 5 Combining ability of some varieties for grain quality traits in diallel mating system 6 Breeding methods Heterosis for kernel characters in rice 7 High-yielding somaclonal variants of Basmati 370 7 Effect of regeneratlon media on shoot production from anther cultures of rice 8 A new source of thermosensitive genic male sterility for 2-line hybrid rice breeding 10 Effect of NAA and 2,4-D on plant regeneration from seedling leaf bases in indica rice 11 Grain quality Distribution of high density and low density grains in the panicle 12 A simple, low-cost, on-farm technique for accelerated aging of rice 12 Pest resistancediseases Development of near-isogenic lines as International differentials of the blast pathogen 13 Pest resistanceinsects Screening local varieties for resistance to whitebacked planthopper Sogatella furcifera in the Mekong Delta, Vietnam 14 Stress toleranceadverse temperature Influence of low night temperatures on rice spikelet differentiation in high-altitude swamps in Burundi 15 Stress toleranceadverse soils Aluminum uptake in relation to K, Ca, and P concentrations in the roots of two rice cultivars with different AI sensitivity 16

IRRN production team . . . . . . . . . . . .

Editor: Carolyn Dedolph Assistant editor: Teresita Rola Layout and design: Ariel Paelmo Production Supervisor: Millet Magsino Editorial assistant: Luisa Gelisan Typesetting: Erlie Putungan Cecilia Gregorio Artwork. Juan Lazaro IV Jesus Recuenco

Integrated germplasm improvementirrigated IRRI rice hybrids evaluated at Kala Shah Kaku, Pakistan 17 Performance of commercial rice varieties in lowlands of western Venezuela 18 lntegrated germplasm improvementrainfed lowland Three rice varieties of medium duration released for rainfed lowland areas in Cambodia 18 Integrated germplasm improvementupland Jaldi Dhan: an extra early rice suitable for upland production Seed technology Seed technology for cold tolerance in rice 20 19

ISSN 0115-0944

Crop and resource management

Fertilizer management Potassium fertilizer for higher yields in scented rice Fertilizer managementorganic sources Green manure: a substitute for inorganic fertilizer N in lowland rice 22

Considering rice quality factors in farm management decisions 39 21

News about research collaboration

Crop management Managing rice nurseries during winter season in the Hill Zone, Karnataka, India 23 Effect of planting date on grain yield and quality of semidwarf scented rice varieties 24 Integrated pest managementdiseases Comparison of Xanthomonas oryzae pv. oryzae strains from Africa, North America, and Asia by restriction fragment length polymorphism analysis 25 Relationship between DNA fingerprints and virulence of Pyricularia grisea from rice and other hosts in Korea 26 lntegrated pest managementinsects African rice gall midge pest in Sierra Leone 27

IRRI builds greenhouse for transgenic rice 40 Low-cost chopper turns farm residues into useful products 41 Cambodian farmers actively involved in technology testing 41 IRRI Library reaches out 41 Anaerobic seeding ready for farmer evaluation 41 Encouraging farmers to experiment with IPM 42 Climate changes and rice 42

lntegrated pest managementother pests Indian wild boar Sus scrofa L.: a rice pest in Madhya Pradesh, India 28 Effect of crop rotation and solarization on the population densities of rice root nematode Hirschmanniella spp. in Nepal 28 Farming systems Deepwater rice establishment in boro rice in the flood-prone ecosystem of West Bengal, lndia

1994-95 short course in rural development, project planning, and resource management 42 Biotechnologia Habana '95 42 New IRRI publications 43 Advanced courses from IHE 43 New publication 43 2nd International Symposium on Systems Approaches for Agricultural Development 43 Rice dateline 43 IRRI group training courses for 1995 43 Rice literature update reprint service 44 Call for news 44 IRRI address 44
Instructions contributors Instructions for contributors


Farm machinery Seeder developed for direct sowing of rice under the puddled soil surface 29 A low-cost chopper for farm byproduct use 30 Using rice hulls to fuel a portable gasifier-engine system 33

Effect of climatic changes on rice production in Punjab, India 32 Rice ecosystems in Tanzania 33 Effects of temperature on host feeding in Cardiochiles philippinensis (Hymenoptera: Braconidae), a larval parasitoid of rice leaffolder Cnaphalocrocis medinalis 34
Research methodology

page 30

page 31

Detection of rice tungro bacilliform virus in nonvector insect species following genomic amplification 35 Modified method of gel consistency test with small samples 36 A simple method to assess the relative amount of leaf wetness in rice canopies 36 Inoculum efficiency in sheath blight as affected by contact frequency, leaf wetness regime, and nitrogen supply 38

page 13

Germplasm improvement
Genetics Genetics
Analysis of yield and yield components in rice
K. Ganesan, W. Wilfred Manuel, and T. Sundaram, Rice Research Station (RRS), Ambasamudram 627401, Tamil Nadu, lndia
Genetic parameters for 12 traits in very early inbreds and hybrids of rice. RRS, Ambasamudram, Tamil Nadu, India. 1993-94. Trait Mean (X) Standard deviation (SD) 0.85 14.35 1.80 2.24 24.83 0.21 3.74 8.08 0.06 0.44 0.22 0.35 Phenotypic coefficient of variation (PCV) 10.4 13.8 19.3 9.2 23.3 9.4 28.1 24.9 16.1 7.2 9.4 13.7 Genotypic Heritability (broad coefficient of variation sense) (GCV) (h 2 ) 10.4 13.7 19.1 9.1 23.3 9.4 28.0 24.9 15.8 7.1 9.1 13.5 98.8 99.0 98.4 97.7 99.6 99.6 99.5 99.9 95.9 97.1 94.7 96.7 Genetic advance as % of mean (GA%) 21.2 28.2 39.1 18.6 47.9 19.4 57.5 51.3 31.9 14.4 18.3 27.4

Mean and genetic variability are the important factors for selection. Heritability estimates and genetic advance are helpful in predicting the effects of selecting the best genotypes. We undertook this study to select good genotypes in very early rice cultivars and hybrids. Four early-maturing (105-115 d) varieties ADT36, ASD16, CO 37, and IR50, and seven very early-maturing (less than 100 d) rice varieties and cultures, ASD8, Heera, Kalyani II, Sattari, AS18696, and AS89011, were crossed in line tester mating design. The 11 parents and their 28 F1 hybrids were studied in an experiment laid out in a randomized block design, replicated three times, during 1993-94 late Pishanam season (Nov-Mar) at RRS. Observations were recorded on five randomly selected plants per replication. Mean (X), standard deviation (SD), phenotypic coefficient of variation (PCV), genotypic coefficient of variation (GCV), heritability in the broad sense (h2 ), and genetic advance as a percentage of mean (GA%) were calculated for 12 traits. Significant differences were observed for all of the traits, indicating wide variability. PCV was, in general, higher than GCV, but the difference was very low, indicating less environmental influence on the expression of different traits. GCV ranged from 7.1 for kernel length to 28.0% for grain yield per plant (see table). Grains per panicle, grain yield per plant, and dry matter production exhibited high GCV and PCV, indicating a wide scope for improvement through selection. A moderate amount of variability (1020%) was observed for days to panicle emergence, plant height, panicles per

Days to panicle emergence Plant height Panicles per plant Panicle length Grains per panicle 100-grain weight Grain yield per plant Dry matter production Harvest index Kernel length Kernel breadth Kernel shape

66.7 104.6 9.4 24.6 106.7 2.22 13.4 32.46 0.42 6.17 2.39 2.58

plant, harvest index, and kernel shape, while low GCV was observed for panicle length, 100-grain weight, kernel length, and kernel breadth. All of the traits had very high heritability but only days to panicle emergence, plant height, grain yield per plant, dry matter production, harvest index, and kernel shape showed high h2 coupled with a high GA%,

indicating a predominance of additive gene effects in controlling these traits. Grains per panicle, grain yield per plant, dry matter production, and days to panicle emergence had high estimates of GCV, h 2 , and GA%. Direct phenotypic selection based on these traits would be effective for varietal improvement in very early rice cultivars.

General combining ability for kernel traits in rice

P. Vivekanandan and S. Giridharan, Tamil Nadu Rice Research lnstitute (TNRRI), Aduthurai, India

We studied the general combining ability of five lines of high-yielding cosmopolitan varieties and three testers of Basmati varieties for kernel traits in a line tester mating design. The experiment was laid out in a randomized block design with three replications in 1991-92 wet season at TNRRI. Each treatment (parents and their resulting 15 F1 hybrids) was raised in two 3-m rows with 30- 20-cm spacing. We randomly measured 10 plants per replication from each treatment. Among the lines, Improved White Ponni and IR50 were good general combiners for kernel breadth, kernel length-breadth ratio, kernel length after

cooking, kernel breadth after cooking, and elongation index (see table). High general combining ability was recorded in Improved White Ponni for grain yield and linear elongation ratio, and in IR50 for kernel length. ADT37 had desirable general combining ability for breadthwise expansion ratio and elongation index, and ADT39 for grain yield and kernel breadth. Considering the testers, Pusa Basmati 1 was a good general combiner for six kernel traits, and ADT 41 for kernel length, kernel length-breadth ratio, and grain yield. High general combining ability effects are related to additive genetic effects that represent the fixable genetic component of variation. Improved White Ponni was determined to be the best parent, followed by Pusa Basmati 1, and IR50, which possess desirable general combining ability for the majority of kernel characteristics studied.

IRRN 20:1 (March 1995)

General combining ability effects of parents for grain traits a at TNRRI, Aduthurai, India. 1991-92 wet season. Parent Grain yield (g) Kernel length (mm) Kernel breadth (mm) Kernel lengthbreadth ratio Kernel length after cooking (mm) 0.28* 0.80* 0.15* 0.96* 0.27* 0.06 Kernel breadth after cooking (mm) 0.18* 0.14* 0.06* 0.25* 0.14* 0.03 0.10* 0.15* 0.05* 0.02 Linear b elongation ratio Breadthwise c expansion ratio Elongationd index

Line TKM9 ADT37 ADT39 IR50 Improved White Ponni SE Tester ADT41 Pusa Basrnati 1 Kasturi

5.61* 2.23* 4.69* 2.62* 5.77* 0.83 4.39* 2.51* 1.88* 0.64

0.04* 0.52* 0.05* 0.64* 0.04* 0.01 0.16* 0.13* 0.04* 0.01

0.12* 0.22* 0.05* 0.16* 0.11* 0.01 0.003 0.04* 0.04* 0.01

0.20* 0.54* 0.02 0.58* 0.14* 0.01 0.08* 0.01 0.09* 0.01

0.09* 0.01 0.001 0.02 0.06* 0.01

0.01 0.07* 0.08* 0.03 0.02 0.01

0.06* 0.06* 0.05* 0.03* 0.03* 0.01


0.05 0.26* 0.31* 0.05

0.03* 0.06* 0.03* 0.01

0.05* 0.04* 0.01 0.01

0.07* 0.08* 0.01 0.01

L-B ratio of cooked rice to mean L-B ratio of milled rice.

aSignificant at 5% level. b Ratio of mean length of cooked rice to mean length of milled rice. c Ratio of mean breadth of cooked rice to mean breadth of milled rice. d Ratio of mean

Evaluation of rice diallel crosses in semidry condition by combining ability analysis

S. M. Ibrahim, Agronomy and Range Science Department, University of California, Davis, CA 95616, USA; and M. Subramanian, Rice Research Institute, Tamil Nadu Agricultural University, Tirur, Tamil Nadu, India

GCA and SCA effects across locations for grain yield and other traits. a Tamil Nadu, India, 1992 wet season. Parent/ hybrid GCA P1 P2 P3 P4 P5 P6 SCA P1/P2 P1/P3 P1/P4 P1/P5 P1/P6 P2/P3 P2/P4 P2/P5 P2/P6 P3/P4 P3/P5 P3/P6 P4/P5 P4/P6 P5/P6 GCA SE CD (5%) CD (1%) SCA SE CD (5%) CD (1%) Plant height Heading date Productive tillers Root weight Biomass Grain yield

Rice is grown on 2 million ha in Tamil Nadu, India. Only about 8% of this area is under limited irrigation, meaning that early drought is common. Drought warrants systematic research on developing rice varieties that withstand considerable moisture stress at early growth stages. We studied the general combining ability (GCA) of six parents: TKM9 (P1) and PMK1 (P3), improved droughttolerant varieties; AS26556 (P2), recently released early drought-tolerant variety; Poongar (P6), a traditional unimproved low-yielding local droughttolerant variety; and IR64 (P4) and ADT37 (P5), improved high-yielding varieties susceptible to drought. We also examined the specific combining ability (SCA) of 15 F1s of direct crosses in a 6 6 diallel design under semidry conditions during 1992 wet season. The experiment was laid out in a randomized block design with three replications in three locations: Agricultural College and Research Institute, Madurai; Agricultural Research Station,

6.28** 1.76** 9.08** 5.73** 0.40** 4.72**

0.40** 1.24** 5.28** 1.45** 7.35** 7.35**

0.30** 0.36** 0.01 0.74** 0.81** 0.81**

0.49** 0.45** 1.39** 1.13** 0.34** 1.01**

0.77** 2.01** 2.77** 1.33** 0.25 1.09**

0.54** 0.49** 0.41** 0.07 0.71** 0.71**

1.99** 1.00** 3.16** 2.33** 2.78** 3.36** 6.98** 0.37 0.91** 2.82** 2.84** 1.87** 0.20 1.29** 0.53**

0.03 0.68** 0.79** 0.71** 0.55** 2.60** 0.66** 3.02** 1.25** 1.26** 2.87** 2.35** 0.56** 1.94** 0.01

0.41* 0.93** 2.71** 0.99** 0.64** 1.95** 1.62** 1.92** 3.14** 1.32** 1.29** 1.81** 0.08 0.19 0.42

0.56* 0.87* 0.18 0.42* 1.14** 0.07 0.73** 1.01** 1.01** 1.07** 0.65** 0.06 0.07 1.82** 0.33

0.94 1.68** 0.37 1.25* 0.67 4.33** 2.00** 1.49** 1.49** 2.91** 0.18 0.01 3.17** 0.46 0.68

0.76* 1.04** 2.02** 0.39 0.14 4.74** 0.30 0.08 0.08 1.44** 0.50 0.60** 2.86** 1.33** 0.08

0.09 0.25 0.33

0.06 0.17 0.22

0.08 0.22 0.29

0.09 0.25 0.33

0.22 0.61 0.80

0.13 0.36 0.43

0.21 0.58 0.76

0.13 0.36 0.47

0.19 0.53 0.69

0.21 0.58 0.77

0.51 1.41 1.86

0.31 0.86 1.13

a *, ** = significant at 5 and 1% level, respectively.

IRRN 20:1 (March 1995)

Paramakudi, and Cotton Research Station, Srivilliputtur, Tamil Nadu, India. The crop was directly sown and maintained under rainfed condition up to 45 d after sowing, at which time the crop was irrigated. Plots were 3-m rows with 15 plants each at 30- 20-cm row and plant spacing. Among the many physiological indices available, the most reliable, such as root weight and biomass productivity, were used to measure drought. Five competitive plants were randomly selected from each plot for observation and analysis.

Among the six parents tested, PMK1 (P3) was the superior general combiner for special drought characters (root weight, biomass productivity, plant height, and earliness). TKM9 (P1) for grain yield and IR64 (P4) for productive tillers were good general combiners. PMK1 (P3), a good combiner for all traits except grain yield and productive tillers, and IR64, a poor combiner for many characters, were crossed and showed positive significant SCA effects for all traits studied (see table). This shows that superior hybrids may also be obtained Ponni) and their 15 F1 hybrids obtained through diallel mating (without reciprocals) for combining ability of 10 grain quality traits. The experiment was laid out in a randomized block design during 1993 dry season and replicated three times. Hills were spaced at 20 10 cm, with a single seedling per hill. Five plants per replication were evaluated following standard procedures for 10 grain quality traits:

from high/low GCA parental combinations. After PMK1/IR64, hybrids AS26556 (good combiner)/PMK1 (good combiner) and TKM9 (medium combiner)/PMK1 (good combiner) were next in superiority. The differential combining ability of PMK1 with other parental types reveals that using this genotype as a potential parent in breeding programs to develop semidry-tolerant types may produce useful transgressive segregants in the segregating generation.

Combining ability of some varieties for grain quality traits in diallel mating system
K. Paramasivam, S. Giridharan, A. P. M. K. Soundararaj, P. Vivekanandan, and P. Parthasarathy, Tamil Nadu Rice Research Institute (TNRRI), Aduthurai 612101, India

We evaluated six popular high-yielding rice genotypes (IR50, TKM9, ADT37, ADT39, ADT40, and Improved White
Table 1. Analysis of variance for combining ability. Source of variation df

kernel length, kernel breadth, kernel thickness, L-B ratio, kernel length after cooking, kernel breadth after cooking, linear elongation ratio, breadthwise expansion ratio, hulling, and milling. The variances due to general combining ability (GCA) and specific combining ability (SCA) were significant except for kernel length. This indicated that the genotypes differed widely for the traits studied. Higher GCA variance than SCA

Mean squares Kernel length (mm) Kernel breadth (mm) Kernel thickness (mm) L: B Kernel length after cooking (mm) 0.368** 0.374** 0.002 Kernel breadth after cooking (mm) 0.094** 0.042** 0.001 Linear elongation ratio 0.011** 0.011** 0.001 Breadthwise expansion ratio Hulling (%) Milling (%)

General combining ability (GCA) Specificcombining ability (SCA) Error

5 15 40

0.205**a 0.005 0.016

0.205** 0.005** 0.001

0.046** 0.016** 0.001

0.378** 0.160** 0.001

0.018** 0.015** 0.001

3.874** 2.601** 0.002

4.037** 2.648** 0.011

a** = significant at 1% level.

Table 2. Estimates of general combining ability effects of parents for different grain quality traits. Parent Trait Kernel length (mm) Kernel breadth (mm) Kernel thickness (mm) L:B Kernel length after cooking (mm) 0.40** 0.14** 0.05** 0.07** 0.02 0.21** 0.014 Kernel breadth after cooking (mm) 0.03** 0.21** 0.04** 0.01 0.04** 0.11** 0.010 Linear elongation ratio 0.01** 0.06** 0.01 0.01 0.05** 0.02** 0.004 Breadthwise expansion ratio Hulling (%) Milling (%)

IR50 ADT37 TKM9 ADT39 ADT40 Improved White Ponni SE (gi)

0.09*a 0.18** 0.01 0.02 0.06 0.01 0.041

0.22** 0.15** 0.16** 0.05** 0.10** 0.14** 0.005

0.01** 0.02** 0.05** 0.07** 0.10** 0.10** 0.002

0.29** 0.32** 0.04** 0.15** 0.15** 0.07** 0.002

0.06** 0.02** 0.07** 0.02** 0.02** 0.03** 0.004

0.43** 0.37** 0.03** 1.19** 0.52** 0.64** 0.035

0.33** 0.70** 0.34** 1.08** 0.49** 0.63** 0.034

a *,** = significant at 1 and 5% level, respectively.

IRRN 20:1 (March 1995)

variance for all the traits (except kernel length after cooking and linear elongation ratio) indicated the predominance of additive gene action (Table 1). IR50 possessed desirable GCA effects for kernel length, kernel breadth, kernel

thickness, L-B ratio, kernel length after cooking, kernel breadth after cooking, and linear elongation ratio. IR50 was found to be the best parent because of its desirable GCA effects for five traits (Table 2). ADT39 was the next best

parent. IR50 and ADT39 could be used in hybridization programs for improving grain quality traits through recombination breeding and for getting desirable segregants.

Breeding methods Breeding methods

Heterosis for kernel characters in rice
P. Vivekanandan and S. Giridharan, Tamil Nadu Rice Research Institute (TNRRI), Aduthurai, India

Relative heterosis (di) and heterobeltiosis (dii) in % for kernel characters in selected crosses. a TNRRI, Aduthurai, India. 1991-92 wet season. Kernel length after cooking di 2.7* 1.4 2.8* 12.5** 13.2** 3.6* dii 16.6** 14.2** 11.2** 4.4** 3.6** 9.9** Kernel elongation di 9.2** 4.0** 8.8** 12.2** 11.9** 8.0** dii 5.2** 3.3* 8.8** 11.5** 11.9** 7.3**

Kernel length Cross ADT37/ADT41 ADT39/ADT41 ADT39/Kasturi IR50/Kasturi Improved White Ponni/ADT41 Improved White Ponni/Kasturi di 3.1** 1.9** 2.6** 0.7* 1.9** 1.2** dii 22.1** 17.2** 16.0** 5.6** 14.7** 15.5**

Kernel L:B ratio di 11.4** 6.8** 11.5** 5.1** 4.1** 9.2** dii 36.9** 26.0** 29.3** 15.6** 21.5** 25.2**

Information is limited on the extent of heterosis for kernel characters in rice. We studied the heterotic manifestations of four kernel characters in 15 hybrids involving eight parents. The experiment was laid out in a randomized block design with three replications during 1991-92 wet season (Oct-Feb) at TNRRI. Each genotype (parents and hybrids) was planted in two 3-m rows at 30- 20-cm spacing. Observations were recorded for 10 randomly selected plants per replication. Seeds were hulled in a McGill sample sheller. Kernel length and L-B ratio were measured with a Mitutoyo micrometer. Grain was milled in a Kett rice polisher and cooked by the method of Juliano and Perez. Kernel elongation was computed as the ratio of mean length of cooked rice to that of milled rice. The performance of hybrids over mid-parent (di) and better parent (dii) was estimated following standard procedures. Heterosis for kernel length was low. Improved White Ponni/ADT41 and IR50/Kasturi expressed significant and positive relative heterosis. Heterosis for kernel L-B ratio was significant and negative, indicating that none of the hybrids were superior for grain fineness. Relative heterosis for kernel length after cooking was positive and significant in five crosses. Heterosis over mid-parent and better parent for kernel elongation was positive and highly significant in six hybrids (see table).

a*, ** = significant at 1 and 5% level, respectively.

IR50/Kasturi and Improved White Ponni/ADT41 show promise for improving kernel length, kernel length after cooking, and kernel elongation through heterosis breeding.

High-yielding somaclonal variants of Basmati 370

F. M. Abbasi, S. T. Abbas, and M. A. Sagar, Rice Program, National Agricultural Research Centre (NARC), P. O. NIH, Park Road, Islamabad 45500, Pakistan

Improving basmati varieties through cross breeding results in recombination of various genes for yield and quality. But

these new combinations often do not yield the desired basmati quality. We have been researching ways to accelerate the improvement of both quality and yield of basmati rices. We obtained a series of somaclonal variants with desired characteristics. Seeds of Basmati 370 were inoculated on MS medium with 2 mg 2,4-D/liter. Calli were subcultured 12 times on the same

Table 1. Agronomic performance of somaclonal variants. a NARC, Islamabad, Pakistan. Variant Flowering (d) 114 de 127 b 129 ab 115 de 116 cd 113 e 116 cd 118 c 114 de 130 a 2.142 1.05 Plant height (cm) f 120 c 140 147 b 116 g h 113 e 123 d 127 e 123 122 e 150 a 2.187 0.99 Tillers/ hill (no.) 12 e 14 cd 14 cd 15 bc 13 de 13 de 20 a 16 b 15 bc 9 f 1.265 5.23 1,000-grain weight (g) 17.33 cd 16.00 de 14.33 e 13.67 e 19.00 abc 17.67 bcd 17.33 cd 16.00 de 21.00 a 20.00 ab 2.38 8.08 Fertility (%) 89.0 a 65.0 de 65.0 de 63.3 de 70.0 c 65.0 de 66.3 cd 61.7 e 80.0 b 70.0 c 3.514 2.95 Yield (t/ha) 5.0 a 3.0 bc 2.5 cd 2.1 d 3.5 b 3.5 b 3.2 b 3.1 bc 5.2 a 3.5 b 0.5868 9.90

TF4 SN85 SN1-80 TF11 SN12 TF8 TF9 TF10 Basmati 385 Basmati 370 LSD (5%) CV (%)

a Means in a column followed by the same letter are not significantly different at 5% level by DMRT.

IRRN 20:1 (March 1995)

Table 2. Grain quality characteristics of TF4 and Basmati 370. NARC, Islamabad, Pakistan. Genotype Head rice (%) 55.4 55.6 Kernel Length (mm) 6.9 6.7 Breadth (mm) 1.73 1.70 Thickness (mm) 1.55 1.60 Quality index L:BT 2.6 2.5 Kernel Size Long Long Shape Slender Slender Type Fine Fine

TF4 Basmati 370

MS medium, with a 15-d interval between subcultures. Three hundred 3-mm calli were transferred to LS medium for regeneration. Plants regenerated were termed R0 plants. Seeds collected from these plants

were the source of R1 plants, and R2 plants were raised from seed collected from R1 plants. About 60% of the lines differed from the parental variety. Eight R2 lines (TF4, TF8, TF9, TF10, TF11, SN12, SN85, and SN1-80) were early

flowering, stiff stemmed, short, and had acceptable grain appearance. They were evaluated in the field for agronomic performance. Results indicated that plant height and days to flowering were reduced significantly compared with those of Basmati 370. TF4 was the most promising variant (Table 1). This line had comparatively greater grain length than Basmati 370 (Table 2). TF8, TF9, and SN12 yielded at par with Basmati 370. Somaclonal variants produced through tissue culture may be used as additional sources of interesting variability for rice breeding.

Effect of regeneration media on shoot production from anther cultures of rice

N. M. Ayres, Biochemistry and Biophysics Department, Texas A&M University, College Station, TX 77843-2128, USA; and A. M. McClung, Texas A&M/USDA Agricultural Research and Extension Center, Beaumont, TX 77713, USA; G. A. Walker and W. D. Park, Biochemistry and Biophysics Department, Texas A&M University

Plants produced from 11 crosses on 4 regeneration media. Crossa Total (no.) (%)b no. N6NB % MSNBC no. % MSMU no. % MSCIB no. %

To identify a medium that could be used on crosses between a wide variety of rice (Oryza sativa) cultivars, we investigated the ability of four regeneration media to induce shoots from antherderived calli. Panicles were selected from F1 plants of 11 crosses of cultivars and breeding lines at the Texas A&M/ USDA Agricultural Research and Extension Center, Beaumont, Texas, USA. Parents included the US longgrained javanica rice cultivars Cypress, Jodon, L202, Lemont, and Toro-2 and the Chinese indica cultivar Teqing. Three unreleased US breeding lines were also used, including javanica line 90102065, indica line MB231, and MB302, derived from an indica japonica cross. Panicles containing microspores at the early- to mid-uninucleate stage were collected from field-grown plants and were kept at 6-8 C in the dark for 3-4 d. They were sterilized with 40% bleach for 16 min, followed by three sterile water washes. Anthers were then removed from the florets, placed on callus induction medium, and incubated

CPRS/MB231 Anthers Calli Shoots CPRS/Toro-2 Anthers Calli Shoots Jodon/Toro-2 Anthers Calli Shoots L202/MB231 Anthers Calli Shoots L202/TQNG Anthers Calli Shoots L202/Toro-2 Anthers Calli Shoots LMNT/Toro-2 Anthers Calli Shoots TQNG/MB231 Anthers Calli Shoots TQNG/Toro-2 Anthers Calli Shoots 2065/MB231 Anthers Calli Shoots 2065/MB302 Anthers Calli Shoots All crosses Anthers Calli Shoots

5730 543 15 5600 727 20 6685 1506 9 3360 275 14 5535 668 4 8230 2345 20 7430 2021 15 10,815 600 58 6950 1937 60 2750 197 10 5125 205 16 68,210 10,960 221

9.5 2.8 13.0 2.8 22.5 0.6 8.2 5.1 12.1 0.6 28.9 0.9 27.2 0.7 5.6 9.7 27.9 3.1 7.2 5.1 4.0 7.8

138 0 203 5 547 0 40 0 212 0 821 4 785 0 213 9 486 3 -

200 4 274 9 503 0 102 10 254 1 819 12 732 8 184 23 777 32 115 4 94 2


148 11 166 6 310 7 75 3 137 3 377 2 315 5 138 20 504 20 82 6 111 14


57 0 84 0 146 2 60 1 65 0 328 2 189 2 65 6 170 5






























16.1 2.0

3445 21


4023 92


2328 90


1164 189


a CPRS = Cypress, TQNG = Teqlng, LMNT = Lemont. bCalculated as percentage of anthers forming calli and percentage

of calli forming shoots.

IRRN 20:1 (March 1995)

at 25 C in the dark for 1-2 mo. The callus induction medium consisted of Chus (N6) salts and vitamins, 60 g sucrose, 0.5 g casein hydrolysate, and 2 mg 2,4-D per liter. When calli had developed to 1-2 mm in diameter, they were randomly placed onto one of four regeneration media. Medium N6NB was Chus basal salts supplemented with 0.1 g inositol, 30 g sucrose, 2 mg benzyl adenine (BA), and 0.005 mg naphthalene acetic acid (NAA) per liter of medium. The other three media were comprised of MS salts, 0.1 g inositol, and 30 g sucrose per liter. In addition, MSMU and MSNBC contained MS vitamins (0.5 mg nicotinic acid, 0.5 mg pyridoxine-HCl, 1 mg thiamine-HCl, and 2 mg glycine per liter). MSCIB was supplemented with 0.8 mg BA, 0.5 mg indole acetic acid (IAA), and 0.5 g casein hydrolysate. MSMU was supplemented with 1 mg BA, 0.5 mg IAA, and 0.5 g casein hydrolysate. MSNBC was supplemented with 2 mg BA, 0.5 mg NAA, and 1 g casein hydrolysate. All media were adjusted to a pH of 5.7 and were solidified with 4 g of agarose per liter. Calli on regeneration media were cultured at 25 C under 16 h light, 8 h dark and were transferred to fresh media every 2-3 wk. Shoots were transferred to basal MS medium with 3% sucrose for rooting when they were several centimeters long, typically 4 wk after calli were placed on regeneration medium. Calli with green spots were maintained on regeneration medium for up to 2.5 mo to allow for shoot production; most shoots, however, were obtained within 4-6 wk. Once a vigorous root system developed, plants were transferred to soil. All green shoots but one grew into mature plants when transferred to the field. The 68,210 anthers plated produced 10,960 calli (1 6.1 %) which were randomly placed onto four types of regeneration media (see table). Overall, 221 of these calli (2.0%) produced green shoots and 586 (5.3%) produced albino shoots. However, as observed in other studies, the rate of green shoot formation differed among media as well as among genotypes. With MSMU, 3.9% of the calli formed green shoots, compared with

0.6% for N6NB. 1.5% for MSCIB, and 2.3% for MSNBC. The superior performance of MSMU was because of better production of shoots from most of the crosses rather than from being skewed by a few crosses producing many shoots. Data in the table are listed by number of independent calli producing green shoots rather than by total number of shoots produced to eliminate bias from crosses producing many shoots per callus. Eight of the 11 crosses tested performed best on MSMU, and one produced equal numbers of shoots on MSMU and MSNBC (see table). Only two crosses produced more shoots on MSNBC; L202/ MB231 and L202/Toro-2. This might be attributed to L202 performing better on this medium. L202 was also a parent in another cross that performed better on MSMU than on MSCIB. Many albino shoots were also produced on all four media (data not shown). Many of them resulted from crosses involving Teqing, possibly reflecting the large genetic distance between the indica Teqing and the other cultivars. While most calli produced no shoots, those that did had an average of 4.3 green shoots per callus, with no apparent difference between the four media. Plants derived from seed of multiple plants taken from a single callus had very similar phenotypes in the field while the phenotype differed among plants derived from different calli (see figure). We did not appear to obtain chimeric calli, perhaps because of calli being

Field view of some of the anther culturederived plants. Every 4 rows are from seed of a single independent plant.

placed on regeneration medium when they were only 1-2 mm in diameter. Overall, MSMU gave a 2-7-fold greater rate of regeneration from calli than did the other media tested for a wide range of cultivars and breeding lines. For the crosses tested, this corresponded to 2-11 calli that produced one or more green plants per 1,000 anthers plated (av of 6). Thus MSMU should be a generally useful regeneration medium for rice anther culture.

Routine research. Reports of screening trials of varieties, fertilizer, cropping methods, and other routine observations using standard methodologies to establish local recommendations are not ordinarily accepted. Examples are single-season, single-trial field experiments. Field trials should be repeated across more than one season, in multiple seasons, or in more than one location as appropriate. All experiments should include replications and an internationally known check or control treatment.

IRRN 20:1 (March 1995)

A new source of thermosensitive genic male sterility for 2-line hybrid rice breeding
P. V. Satyanarayana, I. Kumar, and M. S. S. Reddy, Genetics and Plant Breeding Department, College of Agriculture, Andhra Pradesh Agricultural University, Rajendranagar, Hyderabad 500030, India

Table 1. Pollen fertility-sterility behavior of male sterile mutant rice line under field and glasshouse conditions. Hyderabad, India, 1990. Flowering date 7 Nov 8 Nov 9 Nov 10 Nov 11 Nov 12 Nov 13 Nov 14 Nov 15 Nov 16 Nov 17 Nov 18 Nov 19 Nov 20 Nov 21 Nov-14 Dec Swarna (MTU7029) (control) 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 Pollen sterility (%) Field (20.1/30.8) a (20.2/30.9) (19.8/30.7) (18.9/30.9) (18.6/30.6) (18.4/29.3) (18.1/29.3) (18.2/28.2) (18.1/27.5) (18.2/27.3) (18.5/27.5) (18.3/26.5) (18.0/27.8) (17.8/27.9) (<18.0/<27.5) 15b 15 16 12 16 11 12 10 15 10 8 10 95 100 100 100 Glasshouse (23.2/33.9) (23.0/33.8) (23.2/33.8) (23.0/33.7) (23.1/33.9) (23.0/33.6) (23.2/33.9) (23.8/34.2) (23.0/34.1) (23.1/34.0) (23.0/33.8) (22.2/33.6) (21.1/33.5) (21.0/33.4) (<21.0/<33.5) 15b

Scientists in Japan and at IRRI have reported a thermosensitive genic male sterility (TGMS) line that shows complete pollen sterility at day/night temperatures of 31/24 C and partial to complete fertility at 28/21 C. Scientists in China, Japan, and at IRRI are attempting to use this male sterility system to develop two-line hybrids that are relatively easier to breed than three-line hybrids produced using the cytoplasmic genetic male sterility (CMS) system. To identify such genes from Indian germplasm for use in the two-line breeding system, we conducted a study at the Directorate of Rice Research. Sixteen spontaneously occurring sterile plants from indica rice cultivar IET10726 were identified during 1990 kharif (wet) season. They had 100% pollen and spikelet sterility under a 30/18 C regime and partial to complete fertility when the minimum temperature was higher (22 C). One plant, MSM652, was selected for studying its fertility sterility behavior under different temperature regimes in the field, glasshouse, and photoperiodic chambers. The clones of this mutant showed differential behavior under different temperature regimes in the field and glasshouse, indicating the primary effect of temperature on sterility-fertility behavior (Table 1). The same clones, however, did not show differential sterility-fertility behavior in photoperiodic chambers, thus eliminating the possibility of daylength having any major effect in inducing sterility (Table 2). The major role of temperatureinduced male sterility was further elucidated during the 2nd wk after panicle initiation (PI) with MSM65-2 exhibiting 100% male sterility from 2nd wk of October to 1 st wk of April when average minimum temperature was < 22 C at Hyderabad. It showed partial

a Figures in parentheses indicate corresponding average minimum/maximum temperature in C during the critical week b

(2nd week after panicle initiation). Control Swarna (MTU7029) showed pollen and spikelet sterility <15% in the field and glasshouse under different temperature regimes.

Table 2. Pollen fertility-sterility behavior of mutant rice line in the field and in photoperiodic chambers. Hyderabad, India. Flowering date Pollen sterility (%) Field Photoperiodic chamber (12-h photoperiod) 100 100 100 100 100 100 100 100 8 10 9 10 10 8 9 (20.8/35.6) (20.8/34.9) (20.7/34.4) (21.3/33.8) (21.2/33.6) (21.4/33.6) (21.4/33.8) (21.6/33.9) (22.1/33.8) (22.4/33.9) (22.9/34.2) (23.4/34.2) (23.5/34.1) (23.7/34.0) (>23.6/>34.1) 15b Photoperiodic chamber (14-h photoperiod) 100 100 100 100 100 100 100 100 12 10 10 7 5 6 8 (20.7/35.7) (20.7/34.9) (20.6/34.5) (21.2/33.9) (21.8/33.8) (21.5/33.7) (21.6/33.4) (21.6/34.1) (22.2/34.1) (22.5/34.0) (22.8/34.3) (23.4/34.4) (23.5/34.2) (23.5/34.2) (>23.4/>34.1) 15b

1 Feb 2 Feb 3 Feb 4 Feb 5 Feb 6 Feb 7 Feb 8 Feb 9 Feb 10 Feb 11 Feb 12 Feb 13 Feb 14 Feb 15 - 20 Feb Swarna (MTU7029) (control)

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

(12.1/27.9) a (12.1/28.3) (11.9/28.6) (11.8/28.2) (12.1/29.2) (12.1/29.4) (12.4/29.6) (12.8/30.0) (13.0/30.1) (12.8/30.0) (12.9/29.9) (13.0/29.7) (12.1/29.8) (12.2/30.1) (<12.8/<30.2) 15b

a Figures in parentheses indicate corresponding average minimum/maximum temperature in C during the critical (2nd week after panicle initiation). b Control Swarna (MTU7029) showed pollen and sterility <15% in the field and

photoperiodic chambers under different temperature regimes.

to complete fertility during the rest of the year. The mutants angle and duration of glume opening were comparable with those of CMS line V20 A. We assessed this mutants potential by crossing it with genotypes exhibiting fertility even at 20/ 16 C. The resulting two-line hybrids were evaluated during 1992 rabi (dry) season and found to be heterotic. The TGMS line, when grown in a fertility-inducing temperature regime, can be multiplied by selfing without using a

B line, which is essential for multiplication of a CMS line. The two-line system has an advantage over the CMS system because rice varieties without the fertility restoration gene can be used effectively in a hybrid rice development program.


IRRN 20:1 (March 1995)

Effect of NAA and 2,4-D on plant regeneration from seedling leaf bases in indica rice
G. S. Oinam and S. L. Kothari, Botany Department, University of Rajasthan, Jaipur 302004, India

Plant regeneration from embryogenic calli derived in the second subculture (MS + 0.1 mg NAA/liter and 0.1 mg 2,4-D/liter). Cultivar Calli plated (no.) 46 64 64 44 60 58 Regenerating callia (no.) 30 40 32 28 0 0 Green plantlets produced (no.) 124 4534 342 158 -

Plant regeneration from indica rice is considered to be much more difficult than from japonica rice. By using a combination of 2,4-D and NAA in the medium, we obtained a high frequency of plant regeneration from indica rice. Seedling leaf bases from six indica rice varieties were used. Dehulled seeds were surface-sterilized for 5 min in 0.1 % HgCl2 solution and then washed four times in sterile water. Aseptic seedlings were raised on MS agar (0.8%) medium (pH 5.8) with 3% sucrose. Leaf bases (1 cm long) were dissected from 7-d-old seedlings after removing the outer two leaves. Seedling leaf base explants were cultured on MS medium supplemented with 2,4-D and NAA. Cultures were kept in a growth room at 26 C with a 16-h photoperiod of 1400 lux intensity provided by fluorescent tubes and incandescent bulbs. Preliminary experiments were conducted by culturing the seedling leaf bases of genotype CH1039 only. Explants cultured on medium supplemented with 2,4-D or NAA alone produced poor and nonembryogenic calli. In another experimental set, both 2,4-D and NAA were added in the medium in 16 different combinations. The hormonal combination containing NAA (0.1 mg/liter) and 2,4-D (0.1 mg/ liter) gave the best embryogenic callusing response in CH1039. Seedling leaf bases from five other cultivars were cultured on this medium, out of which three formed embryogenic callus. Embryogenic calli initiated from the basal portion of the leaf bases (Fig. 1a). These calli were compact, creamy, globular, and shiny (Fig. 1b). Embryogenic calli from the primary cultures were subcultured on the induction medium. After 28 d, the calli were subcultured on the same medium. Plant

Chambal CH 1039 HPU2202 HPU5101 Himdhan Kaladhan

a = Approximately 1 g calli (fresh weight) at the time of second subculture.

regeneration was observed in CH1039, Chambal, and HPU2202 in the first subculture (Fig. 1c). Regenerated plantlets were counted after 35 d of the second subculture. A very high frequency of regeneration was recorded in CH1039 whereas Chambal, HPU2202, and HPU5101 had low frequencies. Himdhan and Kaladhan did not show any plant regeneration (see table). During the early stage of plantlet development, coleoptiles were seen emerging, indicating germination of embryoids (Fig.1d).

a) Development of calli from basal portion of leaf segments; b) Embryogenic calli; c) Plant regeneration in the first subculture (cl = coleoptile); and d) High frequency plant regeneration in the second subculture.

The results indicated that genotypic variation for plant regeneration exists among different cultivars. The combination of NAA and 2,4-D in the medium influenced plant regeneration.

IRRN 20:1 (March 1995)


Grain quality
Distribution of high density and low density grains in the panicle
A. R. Gomosta, H. A. Quayyum, A. H. Molla, and M. Z. Haque, Plant Physiology Division, Bangladesh Rice Research Institute, Gazipur 1701, Bangladesh

Distribution of filled grains in the panicles of varieties with different grain sizes. a
a Grain size: length breadth in mm) BR11 = 16.06, BR25 = 11.57, BR5 = 9.22, and Tulsimala = 8.14.

Increasing the number of high density grains (HDG) can increase yield potential of a variety. Varieties differ in the number of HDG per panicle. In a medium grain variety, most spikelets on the primary rachis branches are HDG and those on the secondary or tertiary rachis branches are low density grains (LDG). This study was conducted to find the distribution of HDG and LDG in relation to different grain sizes. We collected 50 panicles at maturity from each of four varieties: BR11, BR25, BR5, and Tulsimala. From each panicle, grains on the primary rachis branches, secondary rachis branches, and tertiary rachis branches were hand-threshed and separated into HDG (specific gravity of

1.2 or more) and LDG (specific gravity of less than 1.2). Most of the grain weight of panicles with coarse grain, (BR11 ), medium grain (BR25), and fine grain (BR5) was attributed to HDG. However, LDG mainly made up the panicle weight of a finegrained variety (Tulsimala) (see figure). The spikelet numbers per panicle were 248 10 for BR11, 270 for BR25, 264 10 for BR5, and 363 10 for Tulsimala. Most of the LDG in Tulsimala appeared on secondary and tertiary rachis branches. The position of HDG and LDG probably depends on grain size and variety. Secondary and tertiary rachis branches contributed significantly to the panicle weight compared with primary rachis branches. preferred because of its excellent cooking and organoleptic characteristics. However, the natural aging process takes at least 3-4 mo and requires proper storage. We undertook this study to explore techniques for aging rice that could be practiced in households without much investment or technical competence. IR50 and ADT36 rice samples harvested during late January 1993 were shade-dried and milled in a modern rice mill. From 100 kg of rice, IR50 yielded 60.2 kg head rice, 18.5 kg broken rice, and 21.3 kg bran and hull mixture, and ADT36 yielded 60.2 kg head rice, 18.2 kg broken rice, and 21.6 kg bran and hull mixture. Sixteen-kg capacity metal tins were coated with blackboard paint on the outer surface and lids to enhance their heat absorption capacity (see figure). The tins were then filled with rice. Two small closed cups with perforated lids were filled with water and left in the rice at
Amylosea content (g%) Amylasec activity (units) Starch a (g%)

A simple, low-cost, on-farm technique for accelerated aging of rice

S. Jesudas and S. Neelakantan, Food Science and Nutrition Department, Agricultural College and Research Institute, Madurai 625104, India

Freshly harvested grain of all rice varieties exhibits poor cooking and organoleptic properties and cooks to a pasty, sticky mass. Aged rice is highly

Changes in the physicochemical, cooking, and organoleptic characteristics of rice varieties due to accelerated aging. Variety Aging period (d) Cooked volume (ml) Total soluble solids a of gruel (g%) Cooking time (min) Volume expansion ratio Length elongation ratio Organoleptic scores b Texture of cooked rice 2.5 3.0 4.0 2.0 3.0 4.0 Overall acceptability 2.5 3.0 3.0 3.0 3.0 4.0


0 6 10 0 6 10

26.0 29.0 33.0 25.0 28.5 33.0

6.6 5.5 3.7 6.5 5.3 3.7

23.0 26.0 26.6 23.0 26.0 27.0

4.0 4.8 5.4 4.1 4.7 5.4

1.5 1.6 1.7 1.5 1.5 1.7

10.8 12.6 17.0 10.1 13.1 17.8

09.1 21.8 36.4 10.9 21.8 36.4

75.8 73.9 70.4 70.0 68.1 64.5


ag/100 g rice. b Maximum score is 4. c mg maltose produced during 15 min with 1% starch.


IRRN 20:1 (March 1995)

equidistance. Tins were kept in direct sunlight from 0800 to 1700 h, thus exposing rice to moist heat as the water evaporated. During the experiment, the temperature of rice in the tin was 40-44 C and the ambient temperature was 30.839.7 C. A sample of rice was drawn from each tin at a 2-d interval and analyzed for changes in physical, chemical, cooking, and organoleptic properties (see table). During 10-d of accelerated aging, cooked volume, cooked weight, volume expansion ratio, length elongation ratio, amylose content, and amylase activity increased and showed a positive correlation with the number of days of aging, while the total soluble solids of the gruel and breadth elongation ratio decreased. Amylase activity caused starch content to decrease. Accelerated aging improved all qualities of the rice much more than did naturally aging rice for 90 d. No insect attack was noticed in either the control or the accelerated aged rice samples after being stored for 90 d. The accelerated aged rice samples did not give off a smell or go rancid during storage. This simple, low-cost technique may be practiced in rural and urban households.

Setup used for the accelerated aging of rice.

Pest resistancediseases
Development of nearisogenic lines as international differentials of the blast pathogen
Zhongzhuan Ling, lnstitute of Crop Breeding and Cultivation, Chinese Academy of Agricultural Sciences (CAAS), Beijing, China and IRRI; T. V. Mew, IRRI; Jiulin Wang and Cailin Lei, CAAS; and N. Huang, IRRI

genic race analysis in different areas may not be comparable. To overcome these problems, we need an international set of differentials, each with a single major gene in a common highly susceptible background. We initiated our attempt to develop such a differential set by identifying a highly susceptible rice variety, which was then used as a recurrent parent to develop a set of NILS for establishing international differentials.

Lijiangxintuanheigu (LTH) is a local japonica variety from Yunnan Province in China. It is one of the Chinese differentials established in the late 1970s. This variety was test-inoculated by the Collaborative Research Group of All China with 1,739 blast isolates from different rice-growing regions in China from 1976 to 1979. It showed a highly susceptible reaction to all the isolates. Since 1980, the susceptibility of LTH has been tested in various provinces and autonomous regions of China. No

The existing differential sets for classifying pathogenic races of Pyricularia grisea cannot be used commonly in different rice-growing regions of the world. Some differential sets consist of rice varieties with various genetic backgrounds. For differential sets consisting of near-isogenic lines (NILS), the recurrent parents used carry resistance genes. Therefore, results of patho-

Table 1. Isolines, parents, genotypes, and homozygous lines tested at IRRI. Isoline Donor parent Genotype Recurrent parent Number of resistant lines (BC7F3) 22 67 55 14 53 14 Homozygous lines tested 1 1 1 1 1 1

F-80-1 F-98-7 F-124-1 F-128-1 F-129-1 F-145-2

Kusabue Tsuyuake K1 Pi No. 4 K60 BL 1

Pi-K Pi-K m Pi-ta Pi-ta 2 Pi-K p Pi-b


IRRN 20:7 (March 1995)


Table 2. Reaction patterns of NILS and their parents to Philippine blast isolates.a Test isolate 1 LTH F-98-7 Tsuyuake F-145-2 BL1 F-1281 Pi No.4 F-124-1 K1 F-129-1 K60 F-80-1 Kusabue + Pi-K m Pi-K m Pi-b Pi-b Pi-ta 2 Pi-ta 2 Pi-ta Pi-ta Pi-K p Pi-K p Pi-K Pi-K 2 3 4 5 6 7 8 9 10 11 12 13 14 15 S S R S R S R S R S S S R S R R R R R R S R S R R R S R R R R R R R R S R S R 16 17 18 19 20 21 22 23 S S R R R R R R R S R R R S S R R R S S S R S R R R S S R S S R R S S S R S R S S R R R R R R R R R S R S S R S R R R S R S S S R S S R S R R R S R S R S R S S R S R R R S R S R S R S R R R R R R R R R R S R 24 S R R R R R R R R R R S R 25 S S 26 S R R R R R R R R R R S R 27 S R R R 28 S S R R R S R S R S R S R 29 S S R S R S R S R S R S R 30 S S







a+ = no resistance gene. R = resistant, S = susceptible, NT = not tested, 1 = 840328, 2 = 850220, 3 = 840505, 4 = V85076, 5 = V85068, 6 = 8401, 7 = 55, 8 = 56, 9 = 57,

10 = P06-6, 11 = V86010, 12 = 86013, 13 = G907, 14 = 840775, 15 = 86046, 16 = 9021, 17 = 9022, 18 = NBG-A8401, 19 = 9081-25-3, 20 = V85016, 21 = BN111, 22 = CA34, 23 = CA79, 24 = CA36, 25 = C9228-33, 26 = B90169, 27 = CA77, 28 = CA57, 29 = C93, 30 = CA16.

incompatible isolates were found. This variety was recently test-inoculated with 30 Philippine blast isolates, which were selected based on differential reactions to Philippine varieties and sets of differential varieties at IRRI. LTH showed susceptible reactions to all isolates used. Thus, LTH appears to lack any known major genes for blast resistance. It was therefore used as the recurrent parent for developing a series of NILS. Selected varieties with known resistance genes from Kiyosawas differential varieties (KDV) (Kusabue, Tsuyuake, K1, Pi No. 4, K60, and BL 1) were crossed to LTH. The F1s were testinoculated with Japanese strain Hoku 1 and the resistant plants were backcrossed to LTH. Seven backcrosses were done, and in each backcross the resistant F1s were screened by test inoculations and were then backcrossed to LTH. The derived BC 7F1 plants were selfed in Beijing, and the BC 7F2 seeds were sown in Hainan Province, China, in 1992. BC7F3 populations from each BC7F2 plant were inoculated with strain Hoku 1 to identify homozygous resistant lines in Beijing in 1993. These homozygous lines are F-80 from Kusabue; F-98, F-102, and F-105 from Tsuyuake; F-114, F-123, and F-124 from K1; F-128 from Pi No.4; F-129, F-131, and F-136 from K60; and F- 145 from BL 1. Finally, a set of NILs that includes F-98-7 ( Pi-K m ), F-145-2 ( Pi-b), F-128-1 ( Pi-ta 2), F-1241 ( Pi-ta), F-129-1 ( Pi-K p ), and F-80-1

(Pi-K) (Table 1) was established. This was confirmed by inoculating with Hoku 1 strain. To evaluate the reactions of the derived NILs with the Philippine isolates, the NILs and the donor KDV were testinoculated with 30 Philippine blast differential isolates from IRRI. Inoculation results showed that the NILS and the corresponding KDV donor parents exhibited different reactions while the donor Tsuyuake showed highly resistant reactions to isolates 840328, V85076, 8401, 55, 57, PO6-6, and 86-13 (Table 2). The same trend was observed in the

other set of NILs and their corresponding donor parents. This indicated that the donor parents have additional unknown genes besides the known genes that were transferred to the isolines. Hence, the isolines exhibited clear-cut resistant or susceptible reactions, whereas the donors, having different genetic backgrounds and additional unknown genes, showed resistant reactions to all the isolates used. Because the isolines are capable of differentiating pathogenic races of blast fungus in both indica and japonica ricegrowing areas, they may be considered as a new set of international differentials.

Pest resistanceinsects
Screening local varieties for resistance to whitebacked planthopper Sogatella furcifera in the Mekong Delta, Vietnam
Le Thi Sen, University of Cantho, Omon, Cantho, Vietnam

We tested 953 local rice varieties for resistance to whitebacked planthopper (WBPH). The bulk seedling test showed

Table 1. Population of WBPH settling on seedlings of susceptible and resistant rice varieties 2, 4, and 6 DAI. Variety lR2035-117-3 BR46 Xoai cat Base Keo Cha A Ba Khieu La Rong Lem TN1 CV (%) Nymphs on 10 seedlings (no.) 2d 15 20 22 22 24 30 29 38 57 16 a ab ab ab ab abc abc bc c 4d 12 18 20 18 33 39 34 43 64 a a a a 6d 16 a 23 ab 21 ab 21 ab 31 bc 30 bc 39 d 35 d wilted 10 Score 0 1 1 1 3 1 3 1 9

b b b b



IRRN 20:1 (March 1995)

Table 2. Settling response of WBPH females on rice varieties at 24, 48, 72, and 96 h after release. Individuals settled on a hill (no.) Variety lR2035-117-3 BR46 Ba Khieu Rong Lem Xoai cat Base La Keo Cha A TN1 CV (%) 24 h 2 a 5 ab 7 bc 8 bc 8 bc 8 bc 10 c 15 c 19 c 29 48 h 1 a 2 b 8 c 8 c 8 c 7 c 8 c 13 d 30 e 10 72 h 1 a 2 a 5 6 5 5 4 14 37 27.6 96 h 2a 1 a 6 b 4 b 4 b 4 b 6 b 13 c 43 d 21.6 Eggs/hill (no.) 152 a 103 a 521 d 572 d 325 b 410 c 267 b 804 d 2,974 e 6

bc bc bc bc b d e

that about 30% were moderately resistant to WBPH. Among these, seven varieties were tested for the nonpreference mechanism.

Two days after infestation (DAI), a significantly higher population of nymphs was found on susceptible Taichung Native 1 (TN1) than on the

other varieties. TN1 wilted 5 DAI. Insect settling response was significantly lower on Base, BR46, Xoai cat, and the resistant varieties than on the other varieties (Table 1). Settling response of WBPH females and number of eggs laid/hill were lowest on BR46 (Table 2). Egg density varied significantly across the varieties and was positively correlated with the density of females settled on a hill. The nonpreference mechanism due to pubescence and hardness of plant tissue makes local rice varieties unsuitable hosts for WBPH settling, feeding, and egg laying.

Stress toleranceadverse temperature Stress toleranceadverse temperature

Influence of low night temperatures on rice spikelet differentiation in high-altitude swamps in Burundi
A. Nizigiyimana, University of Bujumbura, Burundi, and J. Bouharmont, Catholic University of Louvain, Belgium

Rice was only recently introduced in highaltitude swamps in Burundi. Several factors account for reduced yield, one of which is spikelet sterility. We conducted an experiment to determine the origin of sterility and the stage at which the plant is most sensitive to low night temperatures. Plants grown from seedling stage to harvest in cylindrical plastic posts were used. Soil was taken from a fallow bean field. We compared varieties Yunnan 3, Tokombana, Kirundo 9, Facagro 71 , and Facagro 57 at two sites at different altitudes: Bujumbura, located on Tanganyika plain at 800 m altitude with a warm, humid climate and temperature not below 15 C; and Kizunga swamp, at 1900 m altitude on the Zire-Nile ridge, with a cold climate and night temperature often below 10 C, which results in poor rice-growing conditions. Five pots containing three plants each were prepared per variety for each of two treatments. In the first, plants were grown in Bujumbura from seedling to harvest. In the second, plants were grown in

1. Distribution of spikelet types in five varieties of rice grown in Bujumbura, Burundi. 2. Effect of 20 d at Kizunga during panicle initiation on the distribution of spikelet types.

Bujumbura, transferred to Kizunga at panicle initiation for 20 d, and then returned to Bujumbura. The second treatment caused very strong spikelet sterility in five varieties. Some sterile spikelets were poorly differentiated, characterized by small size and discolored glumes. Well-differentiated sterile spikelets stayed green until harvest. Different spikelet types were counted on 10 healthy panicles randomly selected (two per pot) per variety for each treatment (Figs. 1 and 2). More than 70% of the spikelets were fertile in all varieties and the rate of poorly differentiated sterile spikelets was very low with the first treatment. In contrast, 20 d at Kizunga between panicle initiation and the end of microsporogenesis stage considerably increased the frequency of poorly differentiated sterile spikelets, especially in Facagro 71, Facagro 57, and Tokombana. Glumes and floral organs of poorly differentiated sterile spikelets were observed under a binocular microscope. Three types of morphological abnormalities due to low night temperatures were detected: presence of a supernumerary envelope around the floral organs between two glumes, presence of two or more fused ovaries, and reduced number or size of stamens. Sterility rises when rice is grown at high altitude with low night temperatures. When there is cold stress during panicle initiation and spikelet differentiation, morphological malformations of spikelets and flowers contribute to sterility.
IRRN 20:1 (March 1995) 15

Stress toleranceadverse soils

Aluminum uptake in relation to K, Ca, and P concentrations in the roots of two rice cultivars with different AI sensitivity
F. Jan and S. Pettersson, Plant Physiology Department, Swedish University of Agricultural Sciences, Box 7047, S-75007 Uppsala, Sweden

The presence of biologically active Al in acid upland soils and its effect on upland rice are inevitable. In general, rice appears to be a species with a wide range of Al tolerance. A better understanding of Al tolerance mechanisms could lead to the development of more tolerant plants and more effective liming and fertilization practices for existing varieties. Upland

rice cultivars BG35 and IR45, which differ in response to Al (as determined in a previous study), were selected to investigate the Al uptake by roots in relation to root concentrations of K, Ca, and P. The plants were grown for 9 d in aerated nutrient solutions at pH 4.1 under controlled conditions in a growth cabinet (70% RH, 29/21 C day/night temperature) that simulated tropical upland conditions. Nutrients were added daily in small amounts from a complete stock nutrient solution for rice developed by IRRI using nutrient addition technique. This technique ensures less probable interactions between Al and other ions present in the solution, especially P. On day 10, the plants were treated with 560 M Al 3+ as Al2(SO4)3 for 30, 60, and 120 min at pH 4.1

Al concentrations in the roots of both cultivars increased during the experiment but were consistently higher in Alsensitive IR45 than in Al-tolerant BG35 (Fig. 1 a). Al absorption at 0 min may reflect Al contents of the seeds before the start of the experiments. Al uptake in both cultivars seems to be biphasic. A rapid phase of uptake occurred within the first 30-60 min, followed by a slower phase of uptake up to 120 min. The rapid phase was interpreted as representing uptake into the cell walls (free space) and the slow phase across plasma membranes. In IR45, the Al uptake was accompanied with a decrease in internal root concentration of K and Ca during the experiment (Figs. 1b and 1c). We suggest that the decreased K and Ca concentrations in the roots over time are due to Al exchange with K and Ca at Al-sensitive sites within the cell walls or from the inner parts of the cells. In BG35, K and Ca concentrations in roots were practically unchanged with time, indicating that Al exchange with K and Ca did not occur in its roots. The increased P concentration of the roots in both cultivars (Fig. 1d) can be attributed to the parallel increase of Al concentration (Fig. la). It is known that Al/P complexes may precipitate on the surface and/or in the free space of roots and may render P unavailable for transport within the plant. The results suggest that there is a cultivar difference in the binding of Al in the roots. It appears that one of the possible mechanisms of Al tolerance is associated with low Al concentration in the cell walls combined with less disturbed nutrient (K and Ca) binding in the roots.

Root concentrations of aluminum (a), potassium (b), calcium (c), and phosphorus (d) of two rice cultivars during the time interval of 30-120 min. Each value is the mean of 4 replicates, SE. a
aStatistical analysis of the data

was based on a confidence level of 95% (P<0.05) as computed by Students t-test.


IRRN 20:1 (March 1995)

Integrated germplasm improvementirrigated

IRRI rice hybrids evaluated at Kala Shah Kaku, Pakistan
S. S. Ali and M. G. Khan, Rice Research Institute, Kala Shah Kaku, Lahore, Punjab, Pakistan

We evaluated 26 IRRI rice hybrids using CMS lines IR62829 A and IR58025 A, exotic checks IR72 E and IR68 L, and local checks IR6 and KS282 during 1993 kharif (dry) season. Our main objective was to find hybrid combinations that could outperform check varieties in yield and related traits. The experiment was laid out in a randomized complete block design with three replications. Thirty-d-old seedlings were transplanted in the field on 30 Jun with a plant to plant distance of 23 cm. Fertilizer was applied at 120 kg N/ha and 60 kg P/ha. Normal agronomic and plant protection measures were followed.

Fifteen plant/genotype per replication were randomly selected to record plant height, panicle length, and spikelets/ panicle. One hundred seed samples from the 15 randomly selected plants were used to record 100-grain weight in each replication. Panicles/m 2 per replication were counted. For maturity duration, days were counted as days from transplanting to physiological maturity, when about 90% of the grains/panicle were mature. After harvesting, yield was adjusted to 14% moisture level and recorded in t/ha using the net plot size of 1.75 m 7.5 m. Panicles/m2 ranged from 234 to 394, plant height from 86 to 116 cm, and panicle length from 22 to 30 cm. Spikelets/panicle ranged from 125 to 205 and 100-grain weight from 1.88 to 2.76 g. Maximum spikelet sterility (48.0%) was observed in IR62829 A/IR44675-

101-3-3-2 R. IR58025 A/IR72 R had the earliest maturity at 103 DAT. Checks, except IR68, showed minimum spikelet sterility among the hybrid combinations. IR58025 A/IR72 R matured in 103 d and IR58025 A/ IR40750-82-2-2-3 R in 123 d. IR58025 A/IR10198-66-2 R showed 48.44% standard heterosis over KS282. Of the hybrids, only IR58025 A/ IR10198-66-2 R was significantly better than high-yielding check KS282. This hybrid showed standard heterosis over KS282 of 7.5% for panicles/m 2, 16% for panicle length, and 27.2% for spikelets/ panicle. The hybrid IR58025 A/IR29723143-3-2-1 R yielded 24.4% more than the check, but the difference was not statistically significant. These two hybrids should be evaluated on a multilocational basis in Pakistan.

Evaluation of IRRI rice hybrids for yield and related traits at Kala Shah Kaku, Pakistan. 1993 kharif season. Hybrid lR62829 A/IR13603-30-1E-P18 R lR62829 A/lR29723-143-2-1 R lR62829 A/lR32809-26-3-3 R lR62829 A/lR40750-82-2-2-3 R lR62829 A/lR44675-101-3-3-2 R lR62829 A/IR46 R IR62829 A/lR47310 -94-4-3-1 R lR62829 A/lR49461-128-3-3-3 R lR62829 A/IR53901-11-1-2-3 R lR62829 A/lR53915-43-3-3-3 R lR62829 A/lR54742-22-19-3 R lR58025 A/IR10198-66-2 R lR58025 A/IR19058-107-1 R lR58025 A/lR29723-143-3-2-1 R lR58025 A/lR32358-90-3-3 R lR58025 A/lR32809-26-3-3 R lR58025 A/lR32809-314-2-3-1 R lR58025 A/lR34686-179-1-2-1 R lR58025 A/lR35366-28-3-1-3-2-2-2 R lR58025 A/lR35454-18-1-2-2-2 R lR58025 A/lR40750-82-2-2-3 R lR58025 A/IR46 R lR58025 A/lR52256-5-2-2-1 R lR58025 A/lR54742-22-19-3 R lR58025 A/IR70 R lR58025 A/IR72 R IR72 early (exotic check) IR68 late (exotic check) IR6 (local check) KS282 (local check) CV% = 12.97 LSD at 5% for yield = 1.36 t/ha Days to maturity 117 110 105 113 112 115 114 105 105 106 122 111 109 114 107 111 122 115 107 105 123 115 111 120 109 103 117 114 114 110 Panicles/m2 (no.) 301 268 346 323 346 387 394 320 336 382 374 315 318 308 357 295 234 253 242 281 320 358 323 224 284 327 282 255 300 293 Plant height (cm) 96 91 91 102 104 103 107 97 90 86 101 106 102 103 108 97 116 105 99 101 110 103 100 106 102 98 88 107 97 100 Panicle length (cm) 28 25 23 26 25 24 26 23 22 23 28 29 26 27 25 26 30 28 26 26 28 25 27 30 27 26 23 30 25 25 Spikelets/ panicle (no.) 163 149 115 161 154 172 187 146 137 125 172 178 181 180 188 178 205 146 157 145 155 146 144 205 175 147 105 148 128 140 Spikelet sterility (%) 31 34 45 23 48 29 22 28 30 28 32 19 37 34 26 40 24 20 39 36 30 45 33 46 25 37 13 27 16 16 1,000-grain weight (g) 2.1 2.0 2.0 1.9 1.9 1.9 2.0 2.1 2.6 2.4 1.9 2.8 1.9 2.0 2.2 2.3 2.4 2.1 2.3 2.1 2.4 2.7 2.8 2.0 2.0 1.9 1.9 1.9 2.6 2.7 Yield (t/ha) 4.4 6.3 2.7 5.7 5.4 6.3 5.4 5.9 4.9 5.6 4.7 8.1 4.7 6.8 5.2 4.6 5.0 5.5 4.4 5.3 4.6 5.5 5.6 5.0 5.3 4.4 4.4 4.9 5.0 5.5 % heterosis for yield over KS282 19.63 15.78 50.64 4.22 1.65 16.33 0.55 8.07 9.91 2.39 14.68 48.44**a 13.39 24.40 3.85 15.78 7.89 1.28 20.00 3.49 14.85 0.18 2.94 8.44 2.20 19.45

a ** = significantly better than high-yielding check KS282 at 5% level.

IRRN 20:1 (March 1995)


Performance of commercial rice varieties in lowlands of western Venezuela

L. E. Alvarez Larrauri, Universidad Ezequiel Zamora, UNELLEZ, Guanare, Estado Portuguesa, Venezuela

Grain and milling yield and yield components of 6 commercial rice varieties. a Guanare, Venezuela. 1993 wet season. Variety Yield (t/ha) 5.3 a 5.2 ab 4.8 bc 4.6 cd 3.9 e 3.6 e 14.1 Head rice (%) 57.5 ab 59.7 a 58.8 ab 56.8 b 51.1 c 56.9 ab 5.6 Broken grain (%) 6.7 cd Panicles/m 2 (no.) 224.2 bc Grains/ panicle (no.) 126.0 a 80.4 80.0 105.0 62.0 70.0 16.0 c c cd cd 1,000-grain weight (g) 19.8 c

We evaluated the response of rice varieties Cimarron, Araure 4, Oryzica 3, Oryzica 1, Oryzica Llanos 5, and La Toma to increasing N levels. The experiment was laid out in a split-plot design with varieties in the main plots and N levels in subplots during the 1993 wet season. Soil was an Inceptisol with 4.7% organic matter, pH 5.8, 10 ppm P, 147 ppm K, and 53.4% clay content in the upper 17 cm. Rice was dry seeded at 160 kg/ha in a farmers field. Half of the N dose, 20.2 kg P/ha as P 2O5, and 49.8 kg K/ha as K2O were broadcast and incorporated into the soil at sowing. The other half of the N dose was applied at panicle

Oryzica 3 Oryzica Llanos 5 Araure 4 La Toma Cimarron Oryzica 1 CV (%)

4.5 e 5.2 de 7.6 c 16.1 a 9.9 b 7.9

233.6 bc 240.0 abc 245.3 ab 263.6 a 215.6 c 12.3

28.2 a 25.4 b 18.1 c 24.5 b 23.7 b 9.7

aNumbers in a column followed by the same letter are not different according to LSD test at P = 0.05.

initiation. Rice was sown on 31 Mar and the first irrigation was on 1 Apr. The field was flooded on 8 May and the water level was kept at 20 cm until harvest. Cimarron and Oryzica 1 were harvested on 6 Aug, Oryzica 3 and La Toma on 16 Aug, and Araure 4 and Oryzica Llanos 5 on 20 Aug. Oryzica 3 and Oryzica Llanos 5 produced the highest grain yields (see

table), rice grain yield response increased up to 120 kg N/ha. Oryzica Llanos 5 produced the highest head rice yield, lowest % broken grains, and highest grain weight. Oryzica 3 produced the most grains/panicle. Araure 4 lodged severely. Pyricularia grisea, Helminthosporium oryzae, Cercospora oryzae, and Ustilaginoidea virens affected Cimarron and Oryzica 1.

Integrated germplasm improvementirrigated

Three rice varieties of medium duration released for rainfed lowland areas in Cambodia
R. C. Chaudhary, Cambodia-IRRI-Australia Rice Project, P. O. Box 1, Phnom Penh, Cambodia; M. Sarom, O. Makara, S. Sovith, and P. K. Hel, Ministry of Agriculture, Phnom Penh, Cambodia
Table 1. Yield (t/ha) of Santepheap 1, Santepheap 2, and Santepheap 3 compared with that of checks in varietal trials in Cambodia. 1989-92. Variety 1989 (3) Santepheap 1 Santepheap 2 Santepheap 3 Check (IR42, locals) CV (%) 5.0 a a 4.0 c 4.8 b 3.1 d 9.0 Year (no. of locations) 1990 (15) 3.5 b 4.1 a 3.5 b 3.1 c 12.0 1991 (17) 4.0 b 3.9 b 4.2 a 3.5 c 11.3 1992 (87) 2.7 3.1 3.2 2.6 0.22 0.27 0.21 0.22

Cambodia grows rainfed lowland rice of medium duration on about 30% of its area. Popular varieties of medium duration are usually weakly photoperiod sensitive with less than 150 d duration. No recommended modern varieties of medium duration were previously available in the country, so a massive varietal introduction and testing program was initiated in 1989. More than 11,000 introduced and indigenous varieties were tested during 1989-92 using a multilocational approach and a four-tier testing system: observational yield trials, preliminary yield trials, advanced yield trials, and on-farm adaptive trials (OFAT). Three varieties performed very well and were released in 1992 as Santepheap

a Values with a common letter in the same column are not significantly different by DMRT.

Table 2. Main characteristics of the recommended rice varieties. Characteristic Duration (d) Height (cm) Vegetative vigor score Panicle length (cm) Grains per panicle (no.) 100-grain weight (g) Grain size (LB in mm) Grain typeb Leaffolder damage score Gall midge damage score BPH reactionc GLH reactionc Tungro reactionc Blast reactionc Phenotypic acceptability Santepheap 1 131 WS 108 4 24 110 2.7 6.8 2.2 LS 4 6 R R MR MR 4

Santepheap 2 135 WS 108 3 24 102 2.7 6.8 2.2 LS 2 6 R R MR MR 4

Santepheap 3 137 WS 106 4 24 105 2.5 6.1 2.6 LB 3 5 R R R MR 3

aWS = weakly photoperiod sensitive. b LS = long slender, LB = long bold. c R = resistant, MR = moderately resistant.


IRRN 20:1 (March 1995)

(Khmer word for peace) 1, Santepheap 2, and Santepheap 3. Santepheap 1 (IR43342-10-1-1-3-3) is a cross of Meedom Hmwe/IR21313-39-3-2; Santepheap 2 (IR45411-40-2-1 ), a cross of SPR7215-1-25-1-5/IR20925-238-2; and Santepheap 3 (OR142-99), a cross of Pankaj/Sigadis. All three entries were introduced through International Network for Genetic Evaluation of Rice nurseries in Cambodia. During 1989 tests at three locations and 1990 tests at 15 locations, the varieties outyielded the checks (Table 1).

More than 100 farmers tested the varieties in OFAT during 1992 using their best medium varieties as checks. Eighty-seven farmers gave their yields and reactions; 54 of them grew the trial with no fertilizer application. Average yield of Santepheap 1 was 2.7 t/ha; Santepheap 2, 3.1 t/ha; and Santepheap 3, 3.2 t/ha compared with 2.6 t/ha for the best check. The yield advantage was much higher for the 33 farmers who grew the varieties using fertilizer. When asked to choose the best variety, 57 farmers picked Santepheap 1, 62

farmers picked Santepheap 2, and 69 farmers picked Santepheap 3 over their best medium variety. Morphoagronomic and resistance characteristics of these varieties, scored using the Standard evaluation system for rice, are in Table 2. The varieties have reasonable levels of resistance to brown planthopper, green leafhopper, leaffolder, gall midge, blast, and tungro. Farmers are rapidly adapting these varieties on large areas in Cambodia.

Integrated germplasm improvementupland Integrated germplasm improvementupland

Jaldi Dhan: an extra early rice suitable for upland production
S. N. Chakrabarti, Genetics Division, lndian Agricultural Research Institute (IARI), New Delhi 110012, India

A series of Jaldi Dhan (JD) varieties has been developed through radiationinduced mutant genes, mostly derived from traditional cultivars (Table 1). These types showed superior harvest index and productivity efficiency over tall traditional varieties (Table 2). They have appreciable tolerance for moisture stress, are very early (60-80 d to maturity), have aggressive early vigor to suppress weed growth, and possess
Table 1. Features of Jaldi Dhan (JD) varieties. Variety Cross

adaptability to low fertilizer and water inputs. Their capacity to produce biomass under adverse conditions is considerably higher than that of modern varieties. Based on their desirable performance in the national coordinated program during 1988-92, JD3 and JD8 were recommended for on-farm programs in various upland areas. Jaldi Dhan varieties have changed rice production scenarios in drought-prone uplands, flood-prone areas, high altitude areas, coastal lands, and tracts with problem soils. These varieties are useful for increasing upland productivity, as a short-duration dry season (DS) or wet season (WS) crop, and in postjute and postpotato rotations. They may also be

Days to maturity 60-75 65-75 65-75 70-75 65-70 70-75 75-80

Height group a MT MT MT SD SD MT SD

Potential yield (t/ha) b

JD1 (PNR570-17) JD3 (PNR555-5) J D4 (PNR580-6) JD5 (PNR556-26) JD6 (PNR551-4) JD8 (PNR550-1-2) JD10 (PNR555-28)

Gora M c / MW-10M//N-22M Dular M/N-22// PNR351/Mtu-17 M Kalinga Ill// PNR-351/MW-10M Dular M/N-22MM// MW-10 M Dular M/N-22 M Dular M/N-22 M Dular M/N-22// PNR351/Mtu-17 M

5.3 3.9

2.7 3.5 4.2 5.8

used as a contingency crop in areas affected by natural calamities, as a quick crop between two major crops to increase per unit production, and as a component crop in two successive rice crops in the same WS, particularly in areas where only single rice crops are usually grown. In the famine-prone Kalahandi District of Orissa, farmers could harvest yields of 3.5 t/ha with JD5 and JD6, which is double the yield of the traditional local variety, Satika. In Manipur, JD1 produced stable yields of about 4 t/ha. JD3 has become popular in coastal areas of Parganas District in West Bengal as an early DS crop because it can be harvested before salinity becomes a problem. At Central Rice Research Institute, Cuttack, JD3 was the highest yielding extra early type, producing up to 4.6 t/ha under direct seeded, puddled conditions in DS. With its tolerance for moisturestress conditions, JD3 is capable of producing 2 t/ha with less than 500 mm of rain. JD8 had a yield potential of more than 3 t/ha in rainfed uplands in Bihar. An important feature of the Jaldi Dhan plant type is its superior response at low input applications. The response of JD3 was 48% in WS and 91.6% in DS more than that of semidwarf Ratna when 30 kg N/ha was applied.

a MT = medium tall (>110 cm), SD = semidwarf(<110 cm). b Potential yield represents the maximum grain yield obtained

in coordinated trials or in on-farm trials.

M = mutant induced by gamma radiation.

IRRN 20:1 (March 1995)


Table 2. Features of Jaldi Dhan (JD) showing tolerance for moisture stress, New Delhi, India. 1989 WS. Stressed Variety Drought scorea JD1 JD3 JD6 JD8 Dular (traditional check) Cauvery (semidwarf check) 2 3 2 2 2 8 Initial Recovery score b 2 3 2 3 3 5 Spikelet fertility (%) 52.0 45.5 54.0 50.8 50.0 0.0 Terminal Relative water content (%) 62.5 58.0 62.0 61.5 57.5 30.5 Grain yield (% of nonstressed control) 58 50 52 48 30 0 Nonstressed Harvest index Grain yield (t/ha) control Gram yield/d (kg/ha)

.38 .40 .44 .38 .30 .45

3.2 3.5 3.4 3.0 2.0 4.0

52.0 54.0 54.0 45.0 23.0 42.0

a 1 = least affected, 9 = most affected. b 1 = maxlmum recovery, 9 = minimum recovery.

Seed techonology

Seed technology for cold tolerance in rice

S. Nandi, S. Ganguli, and S. Sen-Mandi, Seed Biology Laboratory, Botany Department, Bose Institute, Calcutta 700009, India

Most indica rices exhibit susceptibility to cold stress at seedling stage, which causes slow or staggered seedling growth and delayed maturation. As a result, fields remain occupied longer, causing late sowing of the subsequent crop. We developed a simple technique to overcome the constraints of cold susceptibility of rice seedlings. Rice seeds of cultivar CBI were incubated (for pretreatment) on moist germination paper at 4 C for 0 (control), 1, 2, 3, and 5 d. Untreated (0 d) and pretreated sets (20 seed set) were then germinated at 18 2 C on a germination plate. One hundred seeds were sown in earthen pots under field conditions (1523 C) to assess the ability of the seeds to adapt to cold conditions at the seedling stage. The experiment was repeated three times, maintaining a constant water volume. Root and shoot lengths were measured after 5 d of incubation on germination plates and above soil shoot length (in pots) was measured after 17 d (Fig. 1). To determine the rate of cellular activity during germination under cold stress, we estimated the incorporation of 35S-methionine in untreated and coldpretreated seeds. Seeds were imbibed in water for 0-22 h at 18 2 C. Germinating embryos were then excised and

1. Root and shoot lengths of seedlings being assessed for cold tolerance.

35 S-methionine

2. Incorporation of (sp.act. 200 ci/ mmol; 50 l) into TCA insoluble material in rice embryos. a
a 35S-methionine counting

incubated in germinating medium (sterile 3% sucrose) + 35 S-methionine for 22-24 h at 18 2 C. The embryos were then homogenized in Tris-HCl buffer (pH 7.5). To this homogenate, an equal volume of 10% trichloroacetic acid (TCA) was added and the precipitate was collected in a GF/A filter. The filtrate was washed and dried and incorporation counts in the precipitate were estimated in a liquid scintillation counter using PPO and POPOP as scintillants and toluene as a solvent.

efficiency of 90-95%; counts were taken at SE of 2%. Bar indicates the standard deviation of the mean (SDM) of repeated experiments.


IRRN 20:1 (March 1995)

Cold-pretreated seeds behave differently with different durations of exposure to cold temperatures. Under cold stress in both the laboratory and the field, the 1-d

pretreated seeds exhibited significantly higher root/shoot growth and emergence above soil compared with the control and other pretreated sets. This improved

seedling growth was associated with significantly higher protein synthesis ability during early hours (22-24 h) of germination under cold stress.

Crop and resource management

Fertilizer management
Potassium fertilizer for higher yields in scented rice
D. S. Mehla, Regional Research Station (RRS), Kaul, India; A. P. Gupta, Soil Science Department, CCS, Haryana Agricultural University, Hisar 125004, India; and D. V. S. Panwar, RRS

The area under scented rice varieties is growing because of the premium price they command compared with that of high-yielding nonscented varieties. The fertilizer recommendation for nonscented varieties is high (150, 26.4, 49.8 kg NPK/ ha) compared with scented varieties (60 kg N/ha and 13.2 kg P/ha for tall, and 90 kg N/ha and 13.2 kg P/ha for dwarf). High N fertilizer doses lead to lodging of scented rices, whereas K enhances the plants resistance against lodging. Applying K is not recommended for these varieties. We conducted a field study to compare the responses of three scented rices (HB1, HBC19, Basmati 370) at four fertility levels: 60-13.2-0, 60-13.2-24.9, 60-13.2-49.8, and 90-19.8-37.3 kg NPK/ ha. Thirty-five-d-old seedlings (2/hill) were transplanted at 20- 15-cm spacing. All of the P and K was applied before transplanting and the N in two equal splits at 21 and 42 d after transplanting. The grain yield and yield components were recorded at harvesting. Plants were harvested and then dried for 3 d before threshing. Grain yield was recorded at about 16% moisture.
Effect of fertility levels a on yield and yield components of some scented rices. RRS, India.
a1 = 60-13.2-0 kg NPK/ha, 2 = 60-13.2-

24.9 kg NPK/ha, 3 = 60-13.2-49.8 kg NPK/ha, 4 = 90-19.8-37.3 kg NPK/ha.

IRRN 20:1 (March 1995)


Panicles/m2 were significantly different among varieties across treatments, with HBC19 having the most. Panicles/m2 increased when more fertilizer K was applied in all varieties but their interactive effect was not significant (see figure). Tillers/hill increased with increasing fertility. Spikelet sterility decreased with K application but increased for Basmati 370

as N increased, irrespective of increase in K. Spikelet sterility was drastically reduced in HB1 (see figure). Grains/panicle increased with increasing NPK levels and was highest in HB1. The 1,000-grain weight of all the varieties increased with increasing K application and the grain yield increased with increasing NPK fertility. By increasing the K application, the yield

increase was highest in HB1, then Basmati 370 and HBC19 (see figure). Yield increase in the scented rices with K application was because 1,000grain weight increased, sterility decreased, and the crop did not lodge. Thus, applying K in scented rices is important for producing a good yield.

Fertilizer managementorganic sources Fertilizer managementorganic sources

Green manure: a substitute for inorganic fertilizer N in lowland rice
P. C. Pandey, P. S. Bisht, and P. Lal, Agronomy Department, G. B. Pant Universlty of Agriculture and Technology, Pantnagar, Nainital 263145, Uttar Pradesh, India

Legume green manuring and recycling of crop residues and farmyard manure (FYM) are becoming increasingly important in making todays agriculture sustainable. We studied growing green manure (GM) (Sesbania rostrata and S. aculeata) during summer fallow and

Table 1. N supplied through GM, FYM, and urea fertilizer to meet the basal requirement of 80 kg N/ ha. Pantnagar, India. 1988-91. Proportion of N (kg/ha) applied basally through Treatment S. rostrata b + prilled urea S. aculeata c + prilled urea FYM d + Drilled urea S. aculeata S. aculeata GM + PU a 1988 60 + 20 (17) e 40 + 40 (10) 50 + 30 (20) 60 + 00 (15) 40 + 00 (11) GM + PU 1989 50 + 30 (14) 60 + 20 (15) 50 + 30 (20) 60 + 00 (15) 53 + 00 (15) GM + PU 1990 50 + 30 (14) 76 + 94 (19) 50 + 30 (20) 60 + 00 (15) 54 + 00 (15) GM + PU 1991 60 + 20 (17) 63 + 17 (16) 50 + 30 (20) 47 + 00 (12) 56 + 00 (16)

a 46% N. b 1.75% N, 80% moisture. c 2% N, 80% moisture. d 0.5% N, 50% moisture. e Numbers in parentheses equal the fresh weight of GM and FYM (t/ha).

recycling crop wastes and FYM to supplement inorganic N at the recommended level of 120 kg N/ha for transplanted rice in the irrigated lowlands. The experiment was laid out in a randomized block design with 4 replications during 1988-91 at Pantnagar, India (29 N, 7929' E, 244 m altitude). The soil is silt loam (Aquic Hapludoll) with pH 7.9, 1.1% organic C, 0.1% total N, and CEC 20 meq/100 g soil. S. rostrata and S. aculeata GM treatments were grown in situ and incorporated at 50-60 d into the top 15 cm of soil 1 wk before transplanting rice. The green biomass during 1988-91 vaned from about 10 to 19 t/ha (40-76 kg N/ha) (Table 1). This N addition through GM, however, was underestimated because root weight could not be recorded. FYM at 10 t/ha (dry weight basis) was incorporated 25-30 d before transplanting. Pant Dhan 4 was sown in nursery on 7 Jun and transplanted 2nd wk of July. All plots received 17.5-33.2-10 kg P-K-Zn/ha

Table 2. Effects of different GM, FYM, and urea treatments on grain yield of rice. Pantnagar, India. 1988-91. N source a Control Urea supergranules Prilled urea Prilled urea S. rostrata + prilled urea S. aculeata + prilled urea FYM + prilled urea S. rostrata (varying, see Table 1) S. aculeata (varying, see Table 1) Control (extra) Control (extra) LSD (5%) CV (%)

N rate (kg/ha) 0 120 120 90 120 120 120

Application method 1988 Placed at 8-10 cm 2/3 N basal and 1/3 N 6-7 d before panicle initiation 2/3 N basal and 1/3 N 6-7 d before panicle initiation S. rostrata incorporated in situ before transplanting + prilled urea applied in splits S. aculeata incorporated in situ before transplantlng + prilled urea applied in splits FYM incorporated 1 mo before transplanting + prilled urea applied in splits S. rostrata incorporated in situ S. aculeata incorporated in situ 3.8 5.8 5.7 5.4 6.1 6.1 6.1 5.2 5.2 b 3.8 0.4 6

Grain yield (t/ha) 1989 3.9 7.7 6.8 -b 7.3 7.6 6.9 6.5 6.6 4.4 3.8 0.5 6 1990 3.9 6.5 6.2 6.9 6.6 6.3 6.3 6.2 6.3 3.8 -b 0.5 6 1991 2.4 7.1 5.8 5.6 6.8 6.3 6.9 6.7 6.0 2.5 2.6 0.7 9

Mean 3.5 6.8 6.1 5.6 6.7 6.6 6.6 6.2 6.0 3.6 3.4

N content in Sesbania and FYM was analyzed. After deducting N supplied through these sources, basal N requirement was met using urea. The remaining portion (1/3 of 120 kg N) was applled 6-7 d before panicle initiation. b = not tested.


IRRN 20:1 (March 1995)

before transplanting. N was applied as per treatment (Table 1). Sesbania alone in 1989 and 1990 produced more than 6 t grain/ha (about 2.5 t more than the control), which was statistically equal to 120 kg N/ha as prilled urea (PU) (Table 2). In 1991, S. rostrata alone (16 t green biomass = 56 kg N) was significantly superior to PU. Initially in 1988, Sesbania alone could contribute to producing rice yield equal to only 90 kg N/ha. But response to GM increased with time, probably due to the buildup of organic matter. S. rostrata with PU and FYM + PU produced yields significantly higher than the urea split in 1991 (Table 2). Initially in 1988 and 1989, GM (both S. rostrata and S. aculeata) had some superiority over PU, but in 1990, the differences were nonsignificant. These findings indicated that Sesbania GM alone was capable of producing and sustaining more than a 6 t/ha rice yield, equal to more than 100 kg inorganic N. Green manuring with about 60 kg fertilizer N boosted rice yield to 6.8 t/ha. Economic analysis revealed that using FYM increased the cultivation cost by US$15 and GM by US$27.20, but this was compensated by their yield increases. Integrating organic sources with organic N economized fertilizer N by 50% with GM and 42% with FYM. GM + PU and FYM + PU gave an extra benefit of US$36 and US$24 over the recommended dose of 120 kg N. GM alone gave an extra benefit of US$260/ha over no N and US$30/ha over 90 kg of urea N. These findings are useful for poor and marginal farmers who possess little purchasing power.

Crop management Crop management

Managing rice nurseries during winter season in the Hill Zone, Karnataka, India
M. S. Anwarulla, Y. G. Shadakshari, H. S. Vasudev, and N. M. Poonacha, Regional Research Station (RRS), Mudigere 577132, University of Agricultural Sciences, Bangalore, India

The Hill Zone of Karnataka, India, is situated between 1156' and 1439' N, 7438' and 764' E. Because low temperatures prevail during the winter season (21-25 C maximum and 7-13 C minimum temperatures between 1st wk of December and 3rd wk of January), it

takes 60-65 d for rice seedlings to attain an acceptable minimum height of 15 cm for transplanting. This slow growth and development delays transplanting, and harvesting may not be completed before the onset of monsoon. Multilocation field experiments were conducted during 1992-93 winter months at RRS, Mudigere, Agricultural Research Station, Mudigere, and Agricultural Research Station, Ponnampet, with the objective to determine suitable methods for raising healthy, vigorous rice seedlings within 25-30 d. Each experiment was laid out in a randomized block design with four

Table 1. Air and soil temperature (C) as influenced by different treatments in winter rice nurseries. a Hill Zone, Karnataka, India, 1992-93. Air temperature near plant zone (C) 1992 19.8 16.1 32.1 20.1 20.1 17.6 33.1 21.8 22.5 1.3 1993 20.5 22.0 31.2 22.5 21.6 21.0 32.5 21.6 23.3 1.4 Meanb 20.1 19.0 31.6 21.3 20.8 19.3 32.8 21.7 1992 18.1 12.5 26.1 26.6 20.4 13.6 25.8 22.1 20.2 1.4 Soil temperature near root zone (C) 1993 20.0 15.7 24.3 23.7 22.1 17.2 25.2 22.7 21.3 1.5 Meanb 19.0 14.1 25.2 23.1 21.2 15.4 25.5 22.4

Treatment Raised bed Control Rice straw Polythene Double P Flat bed Control Rice straw Polythene Double P Mean LSD 0.05

aValues presented are the average for 15 d after sowing at 5 cm above and 5 cm below the soil surface. b Mean of 3 sites.

Table 2. Seedling dry weight (g/seedling) and height (cm) at 25 d after sowing as influenced by different treatments in management of rice nurseries during the winter. Hill Zone, Karnataka, India, 1992-93. Treatment Raised bed Control Rice straw Polythene Double P Flat bed Control Rice straw Polythene Double P LSD (0.05)
aMean of 3 sites.

Seedling dry weight 1992 0.276 0.080 0.974 0.768 0.307 0.042 0.935 0.720 0.245 1993 0.279 0.055 1.101 0.730 0.270 0.034 0.971 0.703 0.256 Meana 0.277 0.067 1.037 0.749 0.288 0.038 0.953 0.711 1992 10.10 0.52 17.83 12.86 9.82 0.78 20.52 17.21 0.97

Seedling height 1993 9.12 0.59 19.13 13.48 8.90 0.42 20.71 17.91 1.63 Meanb 9.61 0.55 18.45 13.17 9.36 0.60 20.16 17.56

IRRN 20:1 (March 1995) 23

replications. At each site, the recommended doses of farmyard manure (5 kg/ m 2) and inorganic fertilizer (90-45-45 g NPK/m 2) were applied before sowing. Pregerminated seeds of three varieties (IET6893, CTH1, and Mangala) were sown during 1st wk of December. The treatments comprised two types of seedbeds (raised and flat) and four seedbed conditions: control; nursery beds covered with rice straw at 5 t/ha; nursery beds covered with transparent polythene sheeting (300 gauge) at 45 cm above ground level, keeping the bed airtight for 15 d after sowing (DAS) and later exposing the seedlings to the natural environment; and a double dose of the

recommended P (88 kg/ha) applied before sowing. Air temperature at 5 cm above the soil surface outside and under the covering and soil temperature at 5 cm below the surface were recorded every day for 15 DAS (Table 1). Air and soil temperatures were approximately 20 C in the control treatment. Covering the nursery beds with polythene sheeting increased seedling dry weight from 0.28 to 1.0 g, presumably as a result of the increased temperature. In contrast, the lower soil temperature under the rice straw mulch resulted in a decline in seedling dry weight to 0.05 g.

A double dose of P increased seedling dry weight to 0.73 g without any significant change in soil temperature (Table 2). This suggests the most critical factor determining the growth rate of seedlings under low soil temperature conditions may be availability of an immobile nutrient such as P. Seedbed condition, raised or flat beds, nonpuddled or puddled condition, and not impounding or impounding water may also have influenced seedling height to a lesser extent in the double P treatment. Using either polythene sheeting or applying P could provide 25-d-old seedlings suitable for transplanting in the Hill Zone.

Effect of planting date on grain yield and quality of semidwarf scented rice varieties
S. P. Singh, K. G. Pillai, D. Pati, and N. Shobha Rani, Directorate of Rice Research (DRR), Rajendranagar, Hyderabad 500030, Andhra Pradesh (AP), India

With the release of semidwarf scented rice varieties, the optimum planting time had to be determined to exploit their grain yield potential while still maintaining grain quality. Field experiments were conducted during 1990-91 wet seasons at the DRR experimental farm in Rajendranagar. The

clay (Vertisols) soil has low to medium available N and P with pH 8.2. Semidwarf scented varieties Haryana Basmati 1, Pusa Basmati 1, and Kasturi were tested with check variety Basmati 370 under three planting dates: 15 Jul, 25 Jul, and 4 Aug. The experiment was laid out in a splitplot design with three replications, with varieties in main plots and planting dates in subplots. A single dose of 22 kg P/ha and 33.2 kg K/ha was applied along with three splits of the recommended N dose of 90 kg/ha: 45 kg as a basal dressing. 22.5 kg at tillering, and 22.5 kg at panicle initiation. Twenty-five-d-old seedlings (2-3/hill) were transplanted at 20- 15-cm spacing.

All three improved varieties were significantly superior over Basmati 370 during 1991, although differences among them were nonsignificant. Mean yield increase over that of Basmati 370 was 45.0% for Pusa Basmati 1, 42.2 % for Haryana Basmati 1, and 36.3% for Kasturi. A significantly linear reduction in grain yield was recorded with successive delay in planting from 3.8 t/ha from 15 Jul to 2.8 t/ha from 4 Aug. Average grain yield was reduced by 21.1 % for 25 Jul planting and 27.8% for 4 Aug compared with the grain yield for 15 Jul planting. Daylight hours were maximum for the crop planted on 25 Jul. The maximum mean day temperature was

Grain yield and quality parameters of semidwarf scented rice varieties under different planting dates. DDR, Rajendranagar, AP, India. 1990-91 WS. Grain quality parameters a Treatment Grain yield (t/ha) 1990 1991 Mean Hulling (%) Milling (%) Head rice recovery (%) Kernel length (mrn) Before cooking After cooking Elongation ratio Alkali value Amylose value (%) 1991 Aroma

Variety Haryana Basmati 1 Pusa Basmati 1 Kasturi Basmati 370 LSD (0.05) Planting date 15 Jul 25 Jul 4 Aug LSD (0.05)

3.6 3.5 3.3 0.2

3.5 3.8 3.5 2.5 0.3

3.6 3.6 3.4 2.5

76.8 75.5 75.9 77.2

71.1 70.6 70.1 72.5

27.3 33.8 42.7 24.7

6.8 6.6 6.7 6.5

11.5 12.4 12.0 11.1

1.7 1.9 1.8 1.7

6.56-6.57 5.65-5.67 5.48-5.51

20.3 20.3 18.0


3.6 3.5 3.0 0.2

4.0 3.2 2.6 0.3

3.8 3.3 2.8

76.4 75.3 76.5

71.0 70.3 70.8

28.8 34.1 40.8

6.6 6.7 6.8

11.3 12.4 12.3

1.7 1.9 1.8

5.72-5.80 6.05-6.03 5.88-5.92

20.5 19.2 19.4


aMean of 2 yr. b SS = strongly scented.


IRRN 20:1 (March 1995)

recorded for the 15 Jul planting, which accounts for its higher mean grain yield. Hulling (77.2%) and milling (72.50%) percentages were highest for Basmati 370, though its head rice recovery was the lowest (24.66%). Haryana Basmati 1 showed maximum kernel length (6.8 mm) and intermediate amylose content (20.3%) while Pusa Basmati 1 exhibited the highest linear elongation of cooked rice (12.4 mm) and elongation ratio (1.9) coupled with an acceptable amylose content (20.3%). Kasturi recorded the highest head rice recovery (42.7%) (see table). Among the different planting dates, maximum hulling (76.5%) and head rice recovery (40.8%) and highest kernel length (6.8 mm) were recorded for 4 Aug planting, although grain yield declined with delayed planting. Maximum kernel length after cooking (12.4 mm) and elongation ratio (1.9%) were recorded for 25 Jul planting, and high milling percentage (71.0) and intermediate amylose content (20.5%) were recorded for 15 Jul planting. Aroma was strong under all planting dates. In general, late planting promoted significant improvements in grain quality characteristics but at the cost of slightly reducing grain yield.

Integrated pest managementdiseases

Comparison of Xanthomonas oryzae pv. oryzae strains from Africa, North America, and Asia by restriction fragment length polymorphism analysis
M. Ryba-White, Plant Pathology Department, Kansas State Universlty, Manhattan, Kansas, USA; J. L. Notteghem, lnstitut de Recherches Agronomiques Tropicales et des Cultures Vivrieres, 34032 Montpellier Cedex, France; and J. E. Leach, Kansas State University, USA

Bacterial blight, caused by Xanthomonas oryzae pv. oryzae ( Xoo), is a major

disease of rice in Asia. The disease also has been reported to occur in North and South America and Africa. Although the symptoms of the disease caused by strains from the various countries are similar, the aggressiveness of strains to susceptible rice cultivars varies. For example, the strains from North America are less aggressive than those from Asia, Africa, and South America. We used restriction fragment length polymorphism (RFLP) analysis to compare the genomes of Xoo strains from Africa (strains CAMXO1 and CAMXO2 from Cameroon and MALXO1, MALXO2, and MALXO3 from Mali) and North America (20 strains from
1. Restriction fragment length polymorphism (RFLP) patterns of Xanthomonas oryzae pv. oryzae from the Philippines (PXO61 and PXO86), Mali (MalXO1), Cameroon (CamXO1), USA (TX57-5 and TX 135C), Japan (JXOT7133 and JXOH75373), India (IX0-10 and IX0-4), and Colombia (CIAT1186). Also included are X. o. pv. oryzicola strains BLS175 and BLS256. Southern blots of EcoRIdigested genomic DNA were probed with 32P-labeled pBSavrXa10, which contains a 3.1 kb BamHl fragment internal to avirulence gene avrXa10. 2. Same as Figure 1, except the blot was probed with pBS101, which contains the mobile insertion element IS1112 on 2.4 kb EcoR I- Hind lll fragment.


Reprint service. All items included in the Rice literature update are available at the IRRI Library and Documentation Service. Photocopies of original documents (not to exceed 40 pages) are supplied free to rice scientists of developing countries. Rice scientists elsewhere are charged US$0.20 for each page or part of a page copied, plus postage. Payment should be in check or money order, payable to Library and Documentation Service, IRRI. Address requests to Library and Documentation Service, IRRI, P.O. Box 933, Manila 1099, Philippines. Fax: (63-2) 817-2087, electronic mail: C.AUSTRIA@IRRI.CGNET.COM

IRRN 20:1 (March 1995)


Texas and three strains from Louisiana, USA) with representative strains from Asia (Philippines, Japan, India). Also included for comparison were two strains of X. oryzae pv. oryzicola (Xoz ) from the Philippines. Three repetitive DNA elements cloned into pasmid pBluescript were used as probes. pBSavrXa10 contains avrXa10, a member of an avirulence gene family found in many xanthomonads. pBSTnX1 and pBS101 contain mobile repetitive elements cloned from Xoo. Total genomic DNA from each strain was digested with Eco RI, separated by electrophoresis and blotted to membranes. The DNA was hybridized to 32Plabeled probes using high stringency conditions. Few polymorphisms were observed among DNAs from the five African strains after hybridization with probe pBS101; no polymorphisms were detected with probe pBSavrXa10. With the exception of one strain that hybridized to probe pBSTnX1, only probe pBS101 hybridized with DNA from the 23 strains from the USA. Strains representing two different pBS101 patterns for Africa (MALXO1 and CAMXO1) and for the USA (TX57-5 and TX13-5C) were used for comparison with strains from other rice-growing regions. The DNA hybridization patterns of strains from Africa and USA were different from other Xoo and Xoz strains with all three probes. When probed with pBS101, DNA from the African strains hybridized with no EcoRI bands smaller than 3 kb, whereas all other strains of Xoo and Xoz strains each hybridized with multiple bands smaller than 3 kb (Fig. 1). Probe pBS101 hybridized with fewer bands of DNA (<11 bands) from strains from the USA than from the other strains. Hybridization of Eco RI-digested DNA from Asian and South American strains of Xoo with pBASavrXa10 revealed 6-12 bands, whereas hybridization of the African strains revealed four bands (Fig. 2). The strains from the USA did not hybridize to pBASavrXa10 under high stringency conditions. Xoz strains hybridized with more bands than did the Xoo strains. Neither African strain hybridized to the probe (data not shown).

Only one fragment from one strain from the USA hybridized with pBSTnX1. The DNA banding patterns revealed by three different repetitive probes are substantially different between the genomes of African and Asian Xoo strains, suggesting these strains represent distinct, geographically isolated populations. The pattern revealed by hybridization of the South American strain was more similar to that observed for the Asian strains than to that for strains from North America or

Africa. Patterns revealed for strains from the USA were the least similar to any other group of strains. Although RFLP analysis is useful for comparing genomes within a pathovar, it cannot, on its own, be used for reclassifying strains. However, based on this RFLP analysis and the fact that the strains from the USA are weakly aggressive to rice compared with other strains, we suggest that the grouping of strains from the USA as Xoo be reevaluated.

Relationship between DNA fingerprints and virulence of Pyricularia grisea from rice and other hosts in Korea
Seong Sook Han and Dong Soo Ra, Plant Pathology Department, Agricultural Sciences Institute, Rural Development Administration, Suweon 441 707, Korea; and R. J. Nelson, IRRI

Ricefield weeds are often infected by the blast pathogen Pyricularia grisea. and possible cross-infection between riceand weed-infecting populations of the pathogen could have implications for blast management. In previous studies, populations of the blast pathogen infecting rice and weed hosts were shown to be distinct based on DNA fingerprinting

using the repetitive probe MGR586. To determine if rice- and weed-infecting populations of P. grisea were distinct in Korea, genetic relationships among isolates from rice and other hosts were analyzed by RFLP analysis using MGR586. Pathogenicity of isolates to rice was examined. The isolates collected from rice showed multiple bands hybridizing to the probe MGR586 (see figure). For the isolate from ginger (Zingiber officinale), no hybridizing bands were detected (see figure). Only a few hybridizing bands were detected in isolates from most of the nonrice hosts, such as Italian millet (Setaria italica), millet (Panicum miliaceum), wester world grass ( Lolium multiflorum), and green foxtail

Infection of rice by strains of Pyricularia grisea collected from different host species in Korea. Isolate no. G93-2 G90-5 G91-10 G88-7 G91-5 G88-5 G92-2 G88-4 G90-20 G91-3 G88-1 G91-11 G90-7 G91-15 G90-6 G90-4 G91-2 G90-1 G88-2 G91-4 Source plant Relative no. of bands a (probe MGR586) + + + + + + + + + + + ++ ++ ++ ++++ ++++ ++++ ++++ ++++ Infects rice b (identified race c ) nd nd nd nd nd (KJ-104) (KJ-301) (KJ-101) (KJ-107) (KJ-107)

Ginger (Zingiber officinale) Green foxtail (Setaria viridis) Green foxtail (Setaria viridis) Barnyard grass (Echinochloa crus-galli) Gyopul (Leersia sayanuka) Crab grass (Digitaria sanguinalis) Crab grass (Digitaria sanguinalis) Millet (Panicum miliaceum) Millet (Panicum miliaceum) Millet (Panicum miliaceum) Italian millet (Setaria italica) Italian millet (Setaria italica) Italian millet (Setaria italica) Wester world grass (Lolium rnultiflorum) Timothy (Phleum pratense) Wester world grass (Lolium multiflorum) Goose grass (Eleusine indica) Crab grass (Digitaria sanguinalis) Hybrid ryegrass (Lolium boucheanum) Tall fescue (Festuca elatior)

+ + + + +

a - = no hybridizing bands; + = 1-5 hybridizing bands; ++ = 6-30; ++++ = more than 50 bands. b - = does not infect rice; + = produces typical blast lesions on rice; nd = not determined. c Pathogenic race based on reaction of the Korean differential varieties.


IRRN 20:1 (March 1995)

1. DNA fingerprints produced by hybridization of probe MGR586 with Eco R1digested genomic DNA of Pyricularia grisea collected from rice and different hosts in Korea. Isolates used were a-e, from Oryza sativa; f, from Lolium boucheanum; g, from Lolium multiflorum; h, from Setaria viridis; i, from Zingiber officinale; j, k, from Panicum miliaceum; I, m, from Setaria italica; n, from Digitaria sanguinalis; M1, Hind lll-digested lambda DNA marker; M2, 1 kb ladder.

found to be susceptible to isolates from rice (see table). Several grass hosts in Korea were attacked by isolates of the rice blast fungus. It appears that the Pyricularia populations from these nonrice hosts could be a source of inoculum for the rice crop in the field.

Integrated pest management insects

African rice gall midge pest in Sierra Leone
D. R. Taylor, S. N. Fomba, S. J. Fannah, and H. M. Bernard, Rice Research Station (RRS), Rokupr, PMB 736, Freetown, Sierra Leone

(S. viridis ). Some isolates collected from nonrice hosts, however, showed DNA fingerprints with several to many hybridizing bands, similar to those of the rice blast fungus (see figure, table). Both rice type and nonrice type isolates were collected from crab grass. Cross-infectivity of isolates on rice was examined in greenhouse inoculation experiments (data for rice infection shown in table). Fifteen isolates from various hosts were inoculated to the Korean differential varieties (KDVs). The isolates from tall fescue, goose grass, and crab grass could infect rice and were classified as race KJ-107, KJ-101, and KJ-301, respectively, based on the reactions of the KDVs. The isolates from

hybrid ryegrass and wester world grass were classified as KJ-107 and KJ-104, respectively. Isolates from millet, Italian millet, ginger, and other grasses did not infect rice (see table). Isolates from nonrice hosts were not tested on their hosts of origin. Mixed rice blast isolates representing races KJ-301, KJ-101, KJ-313, and KJ315a were inoculated to the other host species. Susceptible lesions were produced by inoculation with P. grisea from rice on crab grass, goose grass, hybrid ryegrass, tall fescue, and wester world grass, but not on the other eight host species listed in the table. It was interesting to note that isolates attacking rice were collected from the same five species

Rice gall midge has been reported to cause considerable damage to rice in many Asian and African countries. In West Africa, rice gall midge is now ranked as the most serious insect pest of rice in Nigeria and Burkina Faso. We observed incidences of the typical onion leaf-like symptoms of gall midge infestation in upland nurseries, hydromorphic areas, and transplanted fields at RRS, Rokupr, Sierra Leone, in July 1994. Estimates of up to 27% hill infestation were found in farmers fields in the region. Infested tillers and larvae were cultured in the laboratory and the adults tentatively identified as Orseolia oryzivora. Specimen samples were sent to the Entomology Unit, West Africa Rice Development Association, for confirmation. This is the first report of the rice gall midge in Sierra Leone. Similar infestation symptoms were observed on wild rices Oryza longistaminata and O. barthii in infested fields. These wild rices could therefore be important overseasoning hosts for African rice gall midge in Sierra Leone.

IRRN 20:1 (March 1995)


Integrated pest managementother pest Integrated pest managementother pest

Indian wild boar Sus scrofa L.: a rice pest in Madhya Pradesh, India
M. Thomas and A. K. Naidu, College of Agriculture, Jawaharlal Nehru Krishi Vishwa Vidyalaya, Jabalpur 482004 Madhya Pradesh (MP), India

Yield losses of rice due to weeds, pathogens, insects, birds and rodents are well known, but losses due to Indian wild boar Sus scrofa L. have not previously been reported. We observed large-scale damage to rice by S. scrofa in areas adjacent to bamboo forests in Tindhini village, about 17 km away from the city of Jabalpur, MP, India. Jabalpur is located between 22 49'-248' N latitude and 7821'-8051' E longitude.

S. scrofa is an even-toed ungulate belonging to the order Artiodactyla and family Suidae. It is intelligent and vigilant, with highly developed olfactory, visual, and auditory senses. The Indian wild boar is nocturnal, moving under the cover of bushes and trees, rarely exposing itself. S. scrofa is generally herbivorous, often feeding on bamboo (Bambifera spp.) rhizomes and roots of plants. Its seasonal food includes Zizyphus mauritiana fruits, Muduca latifolia flowers, and stones of Mangifera indica. Rapid deforestation and depletion of food in the forest have caused S. scrofa to become a destructive rice pest. The animals now regularly invade cultivated fields in search of food. During June and

July, when food is scarce in the forest, they enter nearby rice nurseries to eat sprouting seeds. Herds also disturb entire fields with their digging and trampling. During the ripening stage of rice, the damage may be more severe, particularly for local, tall cultivars. Taller cultivars may provide greater shelter, and the animals may also be attracted by the plants aroma. Dwarf rice cultivars are less subject to trampling damage. Cultivars with many long auricles on the panicle are less preferred, presumably because of reduced palatability. Farmers in the village are now growing dwarf rice cultivars with long auricles, such as Kranti and Madhuri. in an attempt to lessen damage by S. scrofa.

Effect of crop rotation and solarization on the population densities of rice root nematode Hirschmanniella spp. in Nepal
R. R. Pokharel, Plant Pathology Department, Institute of Agriculture and Animal Science, Rampur, Chitwan, Nepal

High incidence of a mixed population of Hirschmanniella oryzae and H. mucronata was observed at Bhairahawa Agricultural Station, where clay soil predominates. We conducted two separate field experiments at the station in 1991 and 1992 in a rice - wheat crop rotation to understand the effect of different crops and solarization on population densities of rice root nematode in naturally infested soils.

Rice root nematode population density in different crop rotations.

aSoil/200 cm3. b Roots/3 g.

Effect of solarization on Hirschmanniella spp. population densities in rice roots.a Bhairahawa Agricultural Station, Nepal, 1991-92. Treatment May Soil Solarized Nonsolarized 6.6 a 49.4 b Root 5.6 a 38.4 b 1991b November SoiI 5.6 ns 5.8 ns Root 25.1 ns 17.4 ns Soil 6.2 a 50.8 b 1992 c May Root 7.2 a 62.2 b

a ln a column, means followed by the same letter are not signiflcantly different (P=0.01) by LSD test. Soi1/200 cm 3 and root/3 g. ns = nonsignificant. bAv of 10 replications. c Av of 5 replications.

Lentil, mustard, and wheat were grown in the winter season in a randomized complete block design with four replications followed by lowland irrigated rice variety Masuli in the normal growing season. Population densities of rice root nematode were monitored at maximum tillering stage of rice. Nematodes were extracted from soil by sieving with modified Baermann trays, and from roots using a blender followed by sieving with modified Baermann trays. There was no significant difference or trend in nematode population densities in soil and roots across treatments (see figure). This indicated that the crops have either no effect or are equally effective in reducing nematode population. Another experiment was conducted to study the effect of solarization on rice root nematode in naturally infested soil during the normal 1991-92 rice-growing seasons in completely randomized design. Solarizations were done in November and May in the first year and in May only in the second year. Fields were plowed, the soil loosened, and moisture ensured. The fields were then covered with high-gauge plastic sheeting that was sealed from all sides, and left for 21 d. The temperature reached 50-65 C during May solarization. After 21 d, the soil was opened and aerated for 7 d. Rice

28 IRRN 20:1 (March 1995)

was grown during the normal ricegrowing season. The nematode population was monitored during the maximum tillering stage of rice by the methods described previously. The results indicated that May solarization significantly reduced nema-

tode population compared with nonsolarized plots (see table), whereas November solarization did not. This may be because in May, the temperature was high enough to kill the nematodes but it was not in November. May solarization of ricefields may be an effective strategy

for reducing the initial inoculum of the rice root nematode in areas where it is a problem.

Farming systems Farming systems

Deepwater rice establishment in boro rice in the floodprone ecosystem of West Bengal, India
S. K. Bardhan Roy and A. K. Chakraborty, Section of Economic Botanists-Ill, West Bengal, Mission Compound, Midnapore, West Bengal, 721101, lndia

In flood-prone deepwater rice (DWR)growing areas of West Bengal, India, cultivating boro (dry season) rice has become common during the past 10-12 yr. Boro rice is usually transplanted in January and harvested between the end of April and mid-May. The standing boro crop, however, interferes with seeding wet season DWR in April/May. As a result, DWR sowing is either delayed or abandoned. The grain yield of boro rice is an assured 4-5 t/ha compared with an uncertain yield of 1-1.5 t/ha for DWR. So farmers grow boro rice and sacrifice growing DWR. However, innovative farmers have started to grow both boro and DWR by adjusting the establishment technique. This method was developed in the Moyna Basin, a perpetually flood-prone deepwater area of Midnapore District, West Bengal. It comprises 4,000 ha of land connected with the Kangsabati River through a channel. The whole area lies below the river bed, consequently rain and floodwater collect and remain stagnant except for a short time in April/ May. Maximum water depth in the basin center is 100-130 cm during August/ September. Boro rice is planted after stagnant flood water is pumped through the channel in January. We surveyed and interviewed 14 farmers in three villages during 1993. Their practices can be summarized as follows:

Boro rice varieties IET826, IET1444, and IR36 are sown in seedbeds in late November/early December. Land is prepared for transplanting boro seedlings, which includes applying diammonium phosphate @ 200 kg/ha and muriate of potash @ 70 kg/ha. Freshly harvested (Nov-Dec) seeds of photoperiod-sensitive traditional DWR varieties, such as Hatipajra, Amol, Raktapanikalash, and Kamouth, are broadcast at 70-90 kg/ha and incorporated into the soil during puddling for the boro season crop. They remain dormant for a considerable time. Boro seedlings (45-50 d) are transplanted in mid-January and urea at 140 kg/ha is applied at 30 and 60 d after transplanting.

Mature boro rice is harvested by the end of April, after which the soil starts to crack. With the onset of premonsoon rain, germinated DWR seedlings start to appear 30-35 d after the boro harvest. DWR germination rate is usually 60-70%. Seedlings are 20-25 cm tall with 1-2 tillers/plant after 1 mowhen the flood waters start to accumulate. Urea at 70 kg/ha is applied 45 d after the seedlings appear to improve their growth. DWR yields range from 1.0 to 1.8 t/ha depending on the severity of the flood. This establishment method for DWR in boro rice allows farmers to harvest 5-6.8 t/ha of rice per year, with DWR contributing 25-36% of the total yield.

Farm machinery machinery

Seeder developed for direct sowing of rice under the puddled soil surface
P. C. Borlagdan, M. Yamauchi, D. V. Aragones, and G. R. Quick, IRRI

Farmers in South and Southeast Asia are rapidly adapting wet direct sowing as a labor-saving technology for rice crop establishment. A drum seeder was developed at IRRI for direct row sowing of pregerminated seeds on the surface of puddled soil. When this practice is used, sown seeds are likely to be eaten by birds and rats or desiccated in the dry parts of the field. The seeder cannot be operated in the rain as seeds sown in a line would be dispersed. These unfavorable conditions constrain the adoption of the drum seeder by some farmers. It was found recently that pregerminated seeds can be sown under the

puddled soil surface (anaerobic seeding), where sown seeds are protected from bird and rat damage, rain splashing, and desiccation. To overcome the constraints to adoption of the drum seeder, we developed an anaerobic seeder. The drum seeder was modified with spring-loaded furrow openers and closers, rotary flotation drive wheels that prevent the unit from sinking in fields with deep plow layers, and a detachable rainguard, which prevents the seeds from getting wet, thus clogging the drum (see figure). The unit weighs 10 kg. Production cost in the Philippines is US$150. Row spacing is adjustable. The seeding rate was determined by the number of holes around the drum, seed size (a cultivar trait), and the germination conditions. The seeder was tested with five rice cultivars (see table). The seeding rate varied from 38 kg/ha (IR41996-50-2-1-3, seed weight 27.7 mg/seed) to 80 kg/ha (BG34-8, 22.6 mg/seed). Seedling

IRRN 20:1 (March 1995)


Anaerobic seeder designed to put rice seeds under the puddled soil surface. Positions of drum and furrow openers and closers are adjustable. The rainguard is attached to the lifting bar. Abbreviations: cc = center to center distance, = diameter.

Performance of 5 rice cultivars sown by anaerobic seeder. a IRRI, Philippines. 1994 dry season. Cultivar Seeding rate b (no./m2) 354 136 172 238 206 (kg/ha) 80 38 41 59 46 Crop establishment c Seedling (no./m2) 285 a 135 b 129 b 152 b 141 b Establishment (%) 81 ab 99 a 75 b 64 b 68 b Sowing depthd (mm) 7.1 5.3 5.0 3.8 4.9 a a a a a Grain yielde (t/ha) 5.4 a 5.2 a 4.8 a 3.9 b 3.0 c

BG34-8 lR41996-50-2-1-3 IR72 E4129 AUS351

at the 5% level by DMRT. Fertilizer was applied at 50, 30, and 50 kg N/ha at 21 and 42 d after sowing and heading, respectively. bEstimated by measuring the weights of seeds loaded into the seeder and left in the seeder after sowing. Calculation of number of seeds sown was based on the weight of single seed. Mean of four plots of replications. cMeasured 3 wk after sowing. d Estimated by measuring the length between seed and the portion of leaf sheath where the color changes from white to green. The 20 plants sampled for the measurement of crop establishment were used. eHarvest area = 6 m 2 .

aRandomized complete block design with four replications. Means having a common letter are not significantly different

establishment ranged from 64 to 99%. Sowing depth was 4-7 mm. Grain yield varied from 3.0 t/ha (due to lodging) to 5.4 t/ha. The ease with which seeding rate and row spacing may be changed will be useful for the agronomic studies of direct seeding. No seeder currently available is as flexible and as easy to operate as this one. The seeder might be useful for farmers shifting from transplanting to direct sowing. The blueprint is available free of charge from IRRI.

A low-cost chopper for farm byproduct use

M. C. Pasikatan and G. R. Quick, IRRI

When farm byproducts such as rice straw, cane tops, and corn stover are chopped, they can be used for livestock feed, mulch, compost, green manure, and handmade paper. By converting these farm wastes into useful products, wasteful burning may be minimized, and farmers may realize additional income. We have developed and tested a portable, low-cost ($250), inclined-axis chopper that is easy and safe to use (Fig. 1). It makes use of locally available materials, fabrication skills, and equipment.
30 IRRN 20:1 (March 1995)

The blades were made from leaf spring. The four rotating blades were angled 10 upward with respect to the stationary counterblade for effective and low-power cutting (Fig. 2). The angled blades also function as a centrifugal fan to throw the presented materials effectively when chopped. Both the rotating blades and counterblades are reversible to avoid frequent resharpening. To reduce costs, we excluded any feeding device, which ensures convenient feeding and controlled length of chop. Two compensating features were instead included: a blade configuration for suction-assisted feeding, and a 45 inclined blade housing with a hopper perpendicular to it for easy gravity

feeding. With these features, the materials need not be pushed, but simply thrown or dropped into the hopper. This ensures the safety of the operators hands and prevents him or her from stooping. The floor of the housing functions as a simple regulator by restricting length of introduced materials to 25 mm. The prototype was tested with napier grass (70.9 and 80.9% MC, wet basis [wb]), corn stalks (62.7% MC, wb), rice straw (47.9 and 65.9% MC, wb), and sesbania (newly cut, 82.0% MC, wb). The power requirement, capacity, and specific energy were determined for each of these materials at four speeds (900, 1050, 1200, and 1500 rpm) and with a set clearance for each. Power requirement

ping corn stalks required almost 4 times as much power as napier grass at the same clearance because of larger stalk cross-sections and more stiffness. Chopping rice straw with 65.9% moisture yielded 27% higher capacity than that with 47.8% moisture, which indicates straw must be chopped as fresh as possible. At 65.9% moisture, the highest capacity was obtained at 1500 rpm at which the power was 1.026 kW;

capacity, 0.744 t/h; specific energy, 1.475 kW-h/t, L25, 26%; and L50, 62%. The power was 0.960 kW; capacity, 0.664 t/h; specific energy, 1.446 kW-h/t; L25, 96%; and L50, 100% for chopping sesbania at 1500 rpm. Field tests showed that the blades became dull after continuous chopping of 8-10 t corn stalks. The 3.5-hp engine consumed 0.5 liter gasoline/t corn stalks chopped.

Using rice hulls to fuel a portable gasifier-engine system

Zhang Bao Zhao and Xu Yi Cheng, Agricultural Engineering Department, China National Rice Research lnstitute (CNRRI), 310006, Hangzhou, China

1. Front view of the portable, inclined-axis straw and forage chopper. 2. The chopper blades: four rotating blades angled 10 upward and a fixed counterblade.

was measured through an AC power meter and calibrated motor. Capacity was measured by weighing the chopped material output over time. Specific energy, a measure of energy efficiency, is the quotient of power and capacity. Fineness of chop was expressed as percent chopped materials with length of 25 mm or less (L25) and 50 mm or less (L50). Power and capacity generally increased with speed and inversely with clearance. Speed had no significant effect on length of chop. For napier grass, the highest capacity and the least specific energy were obtained at 1500 rpm. At 2 mm clearance, 70.9% moisture content, and 1500 rpm, the power was 0.613 kW; capacity, 1.114 t/h; specific energy, 0.552 kW-Nt; L25, 30%; and L50, 94%. For corn stalks at 2 mm clearance, the highest capacity and least specific energy were obtained at 1050 rpm, which was not significantly different from 1200 rpm. At 1050 rpm, the power was 2.302 kW; capacity, 1.148 t/h; specific energy, 2.01 6 kW-h/t; L25, 23%; and L50, 93%. Chop-

We investigated using biowaste, such as rice hulls, as a farm energy source to power a portable gasifier and engine system. The gasifier (25 cm diam), based on a design developed at the University of California at Davis, USA, was matched with a Chinese S-195 diesel engine rated at 8.8 kW/2000 rpm and with a compression ratio of 20:1. Fuel injection timing (1-1.5 crankshaft angle before top dead center) and range of air to gas ratio of 1.1:1.4 were slightly retarded to permit satisfactory dual-fueled engine operation. The raw gas from the gasifier was ventilated through a wet sieve plate scrubber, a packed bed, and a paper filter to remove tar, water, and particulates. A gas-air mixture device to produce a gas to air ratio of 1.0:1.5 was mounted upstream of the engine manifold for dual-fuel running. The gasifier generated a nearly uniform flow of raw gas, corrected to normal conditions (N) of 760 mm Hg atmospheric pressure and 20 C temperature, of 17-18.4 Nm3/h with a lower heating value (LHV)

of 4.02 MJ/Nm3. The specific gas output was 2.39 Nm3/kg rice hulls; the gasifier's cold gas efficiency was 52.6%. The optimal specific gasification rate was between 185 and 195 kg/m2 per h. One batch of rice hulls lasted for 1.5 h. Characteristics of the dual-fueled engine, without major modification, follow the appropriate torque level of 0.65-0.70 of the rated torque level when on straight diesel (see table). The drop of the brake parameter was observed because of the much lower LHV of the gas and flame velocity. Pilot diesel fuel injection was 10.2% of the injection amount with diesel fuel alone. Thermal efficiency was 30.6%, which was slightly lower than when diesel fuel was used. We observed a change of <3% variance in engine speed due to the change of torque level between 0 and 25 N-m, a slight difference in the sound level between operational modes, and an increase in pilot injection of 6.9% at onset of detonation and knocking at output of 6.3 kW at 2000 rpm. The system satisfactorily pumped water for a continuous 60 h. Diesel fuel consumption for this task could be reduced by 89.8% if rice hulls were used to produce gas to power a dual-fueled system. Total fuel cost for the dual-fueled mode was 5.9 times less than that of the diesel-fueled mode.

Engine performance and operational cost when run under diesel fuel and dual fuels (diesel and rice hull). CNRRI, China. Operational mode Power (kW) Brake thermal efficiency (%) 32.2 30.6 Specific fuel consumption Diesel fuel (g/kW-h) 254.36 25.97 Rice hull (kg/kW-h) 1.125 Total fuel cost a (US$/kW-h) 0.053 0.009

Diesel fuel Dual fuel

8.8 5.6

aDiesel fuel: US$0.21/kg; rice hull: US$.0035/kg.

IRRN 20:1 (March 1995) 31

Effect of climatic changes on rice production in Punjab, India
S. S. Hundal and P. Kaur, Agricultural Meteorology Department, Punjab Agricultural University (PAU), Ludhiana, 141004, lndla

1. Deviation from normal temperature and rainfall, Ludhiana, India. 1972-92.

Global climatic changes are projected to influence crop production. We conducted a study to assess rice yields in Punjab, India, as a consequence of changing climate scenarios. Historical weather data for 1970-93 for Ludhiana were analyzed to determine trends in temperature and rainfall. Temperature increased consistently by about 0.3-0.5 C but no trend was apparent for rainfall during this period (Fig. 1). Global warming scenarios have predicted a temperature rise of 0.52.5 C by the year 2030 and global increases in atmospheric CO2 concentration. We used rice growth simulation model CERES-rice. PR106, a commonly grown rice cultivar in the region, was used to validate it for local conditions. Using computer simulations, the yield under increasing temperature and CO2 levels was evaluated and compared with normal temperature and CO2 scenarios. We created future weather by increasing the daily temperature of the growing season by 0.5, 1.0, 1.5, 2.0, 2.5, and 3.0 C above normal, and using CO2 concentrations of 330 (normal), 430, 530, and 630 ppm. The interactive effects of increase in temperature and CO2 concentrations were also simulated to assess projected rice grain yield. Rice growth and yield were simulated under irrigated and no nutritional and pest stress conditions. As per normal practice, 40-d-old seedlings were transplanted on 15 Jun at 33 hills/m2 using two seedlings per hill. Under the normal temperature scenario, grain yield increased above the usual 7.0 t/ha by 2.6% with enhanced CO2 concentrations of 430, 7.3% with 530 ppm, and 8.8% with 630 ppm (Fig. 2). Under normal CO2 concentration, rice yield decreased by 4.4, 7.3, 9.4. 8.7, 14.2, and 24.5%, for temperature increments of
32 IRRN 20:1 (March 1995)

2. Effect of increasing CO2 and temperature on grain yield of rice.

0.5, 1.0, 1.5, 2.0, 2.5, and 3.0 C above normal, respectively. Adverse effects of temperature increases of nearly 0.3. 0.9, and 1.1 C on rice yield were nullified by enhanced CO2 concentrations of 430, 530, and 630 ppm, respectively. A World Meteorological Organization study predicted a global warming of 1.5-4.5 C with doubling of CO2 concentrations. Our results indicate that a near doubling of CO2 will neutralize the negative effects on rice yields if the corresponding temperature increase is up to about 1.1 C above normal. Global warming of 3.0 C predicted for well into the 21 st century could lead to drastic reductions in rice grain yields of 20.1 % even with a CO2 concentration of 630 ppm. Thus increase in CO2 level would be beneficial for rice yields only if corresponding temperature rise is limited to about 1 C above normal.


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

Rice ecosystems in Tanzania

Z. L. Kanyeka and S. W. Msomba, Kilombero Agricultural Training and Research Institute (KATRIN), Ifakara, Tanzania; A. N. Kihupi, Sokoine University of Agriculture, Box 3601, Morogoro, Tanzania; M. S. Penza, Dakawa Research Center, Box 11872, Morogoro, Tanzania; and K. Alluri, lnternational Institute of Tropical Agriculture (IITA), Ibadan, Nigeria

Classification of rice ecosystems in Tanzania. Rice ecosystem a and growing conditions Unfavorable upland (dryland) (12%) Grown under free draining conditions on undulating slopes of hilly regions and in flat, sandy, porous coastal areas. Slash-and-burn and shifting cultivation system is practiced. Mostly traditional varieties (Kikarati, Mwangulu, Salama, Africa, and Crindima) adapted to low moisture availability are sown directly, commonly by random dibbling or by broadcasting. Favorable upland (hydromorphic) (8%) Grown under impeded drainage or hydromorphic conditions in small depressions. Rice roots are saturated periodically, depending on the water table, which fluctuates with water from rain and subsurface interflow. Rainfed lowland (bunded) (32%) Grown in areas with annual rainfall ranging from 500 to 700 mm. Farmers plow the fields and divide them into small (10 5 m) bunded plots. Puddling within the plots is done with hand hoes after fields are flooded naturally. Transplanting of 4- to 6-wk-old seedlings raised on dry seedbeds is practiced. Rainfed lowland (unbunded) (42%) Ricefields are not bunded, but weeds pulled out and piled along borders between farms may resemble bunds. Depending on the amount and distribution of rainfall, land topography, and flooding from rivers, water depths vary and these lands may encounter drought and/or flooding damage. Cultural practices and soil moisture conditions prior to water accumulation, which starts around mid-tillering or early reproductive stage, resemble those of the upland ecosystem. This ecosystem is often mistaken characterized as upland. Irrigated lowland (mechanized, large-scale) (5%) Grown mainly on commercial, mostly mechanized farms with good water control. Farm sizes vary from 700 ha in Ruvu to 7,100 ha in Mbarali. Land is prepared when dry using tractors, and the crop is dry seeded with seed drills. After initial intermittent Irrigation until the crop is established, fields are continuously flooded. These farms are highly productive (4% of land area gives nearly 25% of the rice production). Recently, infestation of wild and red rice species has started to affect rice yields. Irrigated lowland (nonmechanired, smallholder) (1%) Grown mainly by smallholders in governmentassisted-irrigation schemes. Plots are small, about 200 m 2 and bunded. Improved varieties such as Afaa Mwanza 159 are grown, usually by transplanting. Grain yields of about 4 t/ha are obtained, which are much higher than those in the rainfed bunded ecosystem. Major regions Coast, Lindi and Tanga regions; and Mbeya-Kyela upper regions; Uluguru mountains near Morogoro in Ulanga district

We characterized and classified the rice ecosystems in Tanzania, which had not been previouly done. Out of nearly 40 million ha of arable land, about 17 million ha are potentially suitable for cultivating rice. Rice area in Tanzania is estimated to have varied from 300,000 to 370,000 ha since 1991. A broad classification, based only on the source of water, separates upland and lowland ecosystems. In the upland ecosystem, rice is grown under rainfed conditions without water accumulation on the soil surface. This includes the freedraining, drought-prone, unfavorable upland (dryland) rice and the favorable upland (hydromorphic) rice. In the lowland ecosystem, water accumulation occurs in the ricefields during most of the crop-growing periods. This includes various types of rainfed lowland and irrigated ecosystems (see figure). Characterized upland ecosystems in Tanzania occupy 20% of the total rice area and lowland ecosystems, 80%. (See the table for more details on surface hydrology, topography, growing conditions, and cultural practices.)

Coast, Lindi, Morogoro, Tanga regions; near Dares Salaam, Kibaha and Mikese areas

Bahi-Manyoni, Dodoma, Mbeya (Usanga plains), Mwanza, Shinyanga, Tabora, and Wembere swamps

River valleys of Kilombero, Pangani, Rufiji, Ruvu, Wami; shores of lakes Nyasa and Victoria; and along Kalimawe plains

National Food and Agriculture Corporation (NAFCO)-managed farms in Dakawa, Kapunga, Madibira, Mbarali, and Ruvu

Coast region, Dodoma, Kilimanjaro, Mbeya, Morogoro, Mtwara, Musoma, and Tanga

a The figure in parentheses for each ecosystem is Its approximate percentage of total rice area in Tanzania.

Rice ecosystem in Tanzania. Source: IRRI rice almanac.

IRRN 20:1 (March 1995)


Effects of temperature on host feeding in Cardiochiles philippinensis (Hymenoptera: Braconidae), a larval parasitoid of rice leaffolder Cnaphalocrocis medinalis
Zhang Runjie, Research Institute of Entomology, Zhongshan University, Guangzhou, China; and K. L. Heong and I. T. Domingo, IRRI

Parameters of functional response of C. philippinensis feeding on RLF larvae. Searching rate/h Handling time (h) 0.04 0.03 0.05 0.09 0.12 R2 0.85 0.93 0.86 0.76 0.74

Temperature (C) 25 28 30 33 35 40

F value 81.14 183.06 85.04 44.67 40.70

1.77 1.30 0.29 1.51 0.66 0.25 0.92 0.46 0.31 0.88 0.63 0.45 0.53 0.33 0.56 Parasitoid unable to survive

Many climatologists agree that ecologically significant warming will occur during the next century. Recent estimates suggested that warming as high as 3 1.5 C could be possible. This will affect not only the distribution and survival of species but also the interspecific interaction, such as competition and predation. The parasitoid, C. philippinensis, is a solitary endoparasitoid of the larvae of the rice leaffolder (RLF) C. medinalis. It is commonly found in ricefields of the Philippines, with field parasitism as high as 60%. We evaluated the effects of temperature on C. philipinensis larvae. The laboratory experiments were conducted at six temperatures with six host density levels of 1, 2, 4, 8, 16, and 32 third-instar larvae. Each density was replicated 5 times. Forty-day-old IR36 potted plants were used to support the RLF larvae. These were enclosed in cylindrical mylar cages (23 cm diam and 60 cm high). One pair of 2-d-old mated parasitoids were introduced into each arena and placed in the designated chambers maintained at 25, 28, 30, 33, 35, and 40 0.5 C, 70 4% relative humidity, and 12:12 illumination conditions. The larvae were collected after 24 h and dissected using a binocular microscope to determine parasitism. The data were fitted to the random parasitoid equation, Na = Nt{1-exp[-aTPt/(1+aThNt)]} where N a is the number of hosts attacked, Nt is the number of hosts available, a is the instantaneous search rate, T is the total time of the experiment, Pt is the number of parasitoids, and Th is the parasitoid handling time.

A nonlinear least squares technique using the PROCNLIN procedure available in SAS was applied to fit the data to the equation and to estimate the parameters a and Th (see table). The functional responses are presented in the figure. At 40 C, the parasitoids were unable to survive. Searching efficiencies decreased gradually with increasing temperatures.

Functional response curves of C. philippinensis.

This relationship was best described by a linear model: a = 4.80 - 0.12t where t is temperature in C. The handling time declined slightly at 20 C, but increased subsequently with increase in temperature. A quadratic relationship described this response well: T h = 3.88 - 0.27t + 0.005t2.

34 IRRN 20:1 (March 1995)

Elevated temperature regimes tend to reduce parasitism of C. philippinensis on RLF larvae. This implies that global warming may decrease mortality of RLF larvae caused by this parasitoid. This

study, however, is restricted to the Philippine population of C. philippinensis. These more tolerant populations may exist in other areas, particularly in environments frequently

subjected to higher temperature stresses. More tolerant strains may be introduced to augment biological control of RLF in the Philippines.

Research methodology
Detection of rice tungro bacilliform virus in nonvector insect species following genomic amplification
S. R. Venkitesh and H. Koganezawa, IRRI
1. PCR products resolved in 2% agarose gel, stained with ethidium bromide and visualized by UV illumination. The DNA template used was M = size marker (1001000 bp); lanes 1 and 2 = total genomic DNA from RTBV-infected plant (positive control); lanes 3-12 = total genomic DNA of N. virescens (top) and 10 N. nigropictus (bottom); lanes 13 and 14 = total genomic DNA from nonviruliferous insects (negative control). 2. PCR products resolved in 2% agarose gel, stained with ethidium bromide and visualized by UV illumination. The DNA template used was M = size marker (3001000 bp); top lane 1 = total genomic DNA from RTBV-infected plant (positive control); lane 2 = total genomic DNA from nonviruliferous insect (negative control): lanes 3-14 = total genomic DNA of 12 Nilaparvata lugens. Bottom: lanes 1-2 = total genomic DNA from RTBV-infected plant (positive control); lanes 3-12 = total genomic DNA of 10 Recillia dorsalis; lanes 13 and 14 = total genomic DNA from nonviruliferous insect (negative control).

Rice tungro disease (RTD) caused by rice tungro bacilliform virus (RTBV) and rice tungro spherical virus (RTSV) has a semipersistent relationship with its principal vector Nephotettix virescens (Distant). This implies there is no replication of virus particles in the vector, and the active site for association of the virus is in the anterior part of the gut, namely the pharynx, cibarium, or possibly the stylets. The very low quantities of virus(es) acquired during feeding are only retained for a relatively short period, up to 3-5 d. To detect RTBV in single viruliferous insects, we used Ampliwax polymerase chain reaction (PCR) amplification procedure in which a small fragment (569 bp) of RTBV DNA was amplified from the total nucleic acid extract of a single viruliferous insect and detected on agarose gel. Total nucleic acid extract of insect tissue was prepared by grinding the individual insect in a microcentrifuge tube containing the extraction buffer, (10 mM Tris-HC1, pH 8.3, 10 mM EDTA, 10 mM NaCl, 0.5% SDS, and 1 mg proteinase K/ml). The contents were transferred to eppindorf tubes and incubated at 65 C for 2 h. The samples were extracted with TE phenol/chloroform (1:1 v/v) and then with chloroform and isoamyl alcohol (24:1 v/v). Total genomic DNA was precipitated by adding 3 M ammonium acetate (1/10V), glycogen (1 l), and 2/3 volume of cold absolute alcohol to the aqueous phase and incubated at -20 C overnight. Pellets formed by the centrifugation at 13,000

rpm for 20 min were washed carefully with 70% alcohol. Excess alcohol was removed by vacuum drying. The pellets were suspended in 30 l of TE buffer (10 mM Tris-HC1, 1 mM EDTA, pH 8). The nucleic acid concentration was estimated by gel electrophoresis of samples on 1% agarose gel. Twenty-mer primers were designed according to the published sequence of RTBV. The plus strand primer (5'-CCTGTGAAACCCAATACTGC-3') and minus strand primer (5'-GCTGAGGTGCTACATAGGTT-3')

IRRN 20:1 (March 1995)


were complementary to the region of the P 194 open reading frame of the RTBV genome. In PCR amplifications using wax beads (Ampliwax PCR Gem l00), the PCR reagents were mixed separately as lower and upper components. The lower reagent mix contained 1X PCR buffer (50 mM KC1, 10 mM Tris-HC1, pH 8.3), 0.2 mM of each dNTP, 2.5 mM of MgCl 2 , and 0.5 M of each of two primers, made up to a total volume of 25 l per reaction tube using sterile deionized water. A wax bead was added in each reaction tube. The wax was melted by incubating the reaction tubes at 80 C for 5 min. Melted wax was layered over the lower reagent mix. After wax layering, the reaction tubes were either used directly for PCR reactions or stored at 4 C until needed. The upper reagent mix contained 1X PCR buffer, 2.5 units (0.5 l) of AmpliTaq polymerase and 100 ng (15-25 L) of total nucleic acid extract of insect tissue, made up to a total volume of 65 L per reaction tube. Upper reagent mix was aliquot to each reaction tube above the solid wax layer, PCR amplifications were carried out for 45 cycles with a denaturation temperature of 94 C for 1 min, annealing temperature of 53 C for 2 min, extension temperature of 72 C for 3 min, and a final extension temperature of 72 C for 5 min. PCR products were examined with ethidium bromide-stained 2% agarose gel in 1X TBE buffer using size markers ranging from 100 to 1000 bp. To obtain viruliferous insects, female adults of the four insect species reared on virus-free TNl were exposed to RTDinfected TN1 rice plants for 3 d. Among 30 insects from each species tested, PCR detected RTBV in several individual insects of vector species such as Nephotettix virescens, N. nigropictus, and Recillia dorsalis, and in one nonvector species, Nilaparvata lugens. (Results of PCR on 10 individual insect samples from each species are shown in Figures 1 and 2.) Amplification of RTBV DNA in the nonvector occurred presumably from RTBV particles that had entered the insect gut through ingestion of plant sap containing the virus. Absence of desired PCR products in some viruliferous insects may

be because of the presence of undigested enzyme inhibitors, variability in the amount of viral DNA in individual insects, or rapid viral nucleic acid degradation. PCR is an expensive technique, particularly if the reagents are not assembled in the lab and expensive kits must be purchased. As our results

show, the technique can be very sensitive, but that sensitivity does not necessarily indicate the identification of a vector. Further considering all these difficulties, we found that PCR is not a reliable virus detection method for the routine population analyses of viruliferous vectors of rice tungro.
Gel type of 12 varieties tested. Genotype Sitabhog T412 CR1009 CR1018 Ratna Karnal Local EC205265 CR1014 Tulasimanjari Basmati 370 Pakistan Basmati M. Sungsong Gel length (mm) Gel type 10.3 10.0 10.0 9.67 9.67 9.67 9.67 9.67 9.33 11.66 10.67 31.0 0.3 0.5 0.5 0.3 0.3 0.3 0.3 0.3 0.5 0.3 0.3 1.5 Hard Hard Hard Hard Hard Hard Hard Hard Hard Medium Medium Soft

Modified method of gel consistency test with small samples

B. Swain and M. Nagaraju, Central Rice Research Institute, Cuttack 753006, India

Gel consistency test is a reliable index for cooked rice texture. We modified Cagampangs method to develop a reliable method for small samples (10 mg). We analyzed the gel consistency of 10 varieties and two varieties with known gel consistenciesM. Sungsong (soft) and Basmati 370 (medium)as controls. The protocol for the modified gel consistency test is as follows: 1. Grind the samples into a fine powder (100 mesh). 2. Put a 10-mg sample of powder into a 10- 75-mm test tube. Add two drops (0.02 ml) of 95% ethanol containing 0.025% thymol blue as an indicator. 3. Mix the contents well in a cyclo-mixer for 1 min. 4. Immediately add 0.2 ml of 0.2 N KOH and mix well again for 1 min in the mixer. 5. Cover the test tube with a glass marble and place it over a boiling water bath for 5 min. Contents should not exceed 2/3 the height of the tube. 6. Cool the test tube at room temperature for 6 min, and then transfer it to an ice water bath, maintaining a constant temperature of 8-9 C for 5 min. 7. Lay the chilled test tube horizontally on a table for 30 min. 8. Measure length of the gel in mm from the bottom of the tube to the top of the gel. Gel is classified as hard (6-10 mm), medium (11-20 mm), and soft (more than 20 mm). Previously, time was not specified in steps 3 and 4, nor was specific temperature for the ice water bath. The classification of rice based on gel length was modified according to the results of the controls. Nine of the 12 varieties had hard gel consistency; only Pakistan Basmati and Basmati 370 had

medium type (see table). Based on this analysis, the modified method is reliable and time-saving and will be useful when handling many varieties.

A simple method to assess the relative amount of leaf wetness in rice canopies
N. P. Castilla, F. A. Elazegui, IRRI; R. M. Leao, Nueva Viscaya State lnstitute of Technology, Nueva Viscaya, Philippines; and S. Savary, IRRI-lnstitut Franais de Recherche pour le Dveloppement la Coopration (ORSTOM) Collaborative Project, IRRI

Leaf wetness is a critical factor in the development of many diseases. Leaf wetness in the plant canopy may come from dew, guttation, rain, or irrigation. Leaf wetness in rice crops is influenced by canopy structure (leaf area, density, and angle), which is, in turn, affected by plant development stage and cropping practices, including planting density. Leaf wetness measurement is possible using sophisticated and often expensive mechanical or electronic leaf wetness sensors or recorders. In many cases, several sensors must be used to address the vertical structure of the canopy and its successive layers. The


IRRN 20:1 (March 1995)

1. An assessment scale for rating leaf wetness in a plant canopy, where 0 = dry, 1 = with a few scattered big droplets, 2 = with a thin film of small droplets, and 3 = with a dense film of small droplets. Approximate weight (mg/cm 2) of water corresponding to each rating scale is 1 = 1.50, 2 = 2.50, and 3 = 4.50.

2. Leaf wetness rating and duration over time at different layers of the rice canopy at LAI = 3 (a), 7 (b), and 8 (c). Each point is the mean of 10 observations.

water corresponding to each rating scale is indicated in Figure 1. Assessments can be made at the upper layer, middle layer, and base of the rice canopy. The effect of leaf area index (LAI) on leaf wetness in three layers of a rice canopy is shown in Figure 2. Observations were made from 0600 to 1200 h. The average duration of leaf wetness increased as LAI increased. The upper layer of the canopy initially had the highest mean leaf wetness rating, which declined faster than that of the lower layers after 0800 h. The average leaf wetness duration of the middle layer was the longest. This simple method was used in field experiments during the 1992-93 wet season and dry season using IR72 with the following N input levels as treatments: no fertilizer (N0), 80 kg N/ha (N1), and 120 kg N/ha (N2). During the cropping season, 10 leaf wetness assessments were made and taken on 10 hills in each N treatment at 0600 h. During the wet season, mean leaf wetness ratings were N0 = 1.38 b, N1 = 1.70 a, and N2 = 1.78 a (values followed by the same letter are not significantly different at LSD 0.05 ). During the dry season, means were N0 = 1.61 b, N1 = 2.11 a, and N2 = 2.38 a. These results showed that variation in canopy structure, which is affected by the amount of N fertilizer, caused differences in leaf wetness. This method allows direct observations of leaf wetness in different layers of a canopy at desired frequency and interval. Although flexible, it is more time-consuming and difficult to use in determining the exact time of water deposition and disappearance than are leaf wetness sensors. This method is currently being used in understanding the relationship between leaf wetness and spread of sheath blight.

number of sensors is again increased when several treatments that affect the canopy structure are handled in a single experiment. Moreover, most sensors measure leaf wetness indirectly: wetness of artificial surfaces is actually recorded. These artificial surfaces may react differently from plant surfaces, leading to erroneous observations. We developed a

simple method to determine the amount of leaf wetness at different layers of the rice canopy. The method is based on direct observation, with a four-point rating scale where 0 = dry; 1 = few, scattered big droplets; 2 = thin film of small droplets; and 3 = dense film of fine droplets (Fig. 1). The approximate amount of

IRRN 20:1 (March 1995)


Inoculum efficiency in sheath blight as affected by contact frequency, leaf wetness regime, and nitrogen supply
N. P. Castilla and F. A. Elazegui, IRRI; and S. Savary, IRRI- lnstitut Franais de Recherche pour le Dveloppement en Coopration (ORSTOM) Collaborative Project, IRRI

An experimental design was developed at IRRI to study components of rice sheath blight (ShB) epidemics, caused by Rhizoctonia solani Khn (IRRN 18(3): 42-43, 1993). This design allows measurement of responses of the pathosystem to various stimuli, representing crop management or environmental factors. One response is the ratio of the density of daughter lesions on trap plants to the density of mother lesions on inoculated plants, which we term inoculum efficiency (IE). We report here effects of crop density, leaf wetness, and N supply on IE. The methodology involves two basic components: a quadrat of 3 3 hills where the disease is artificially established by inoculating the hills with a rice grain and hull mixture colonized by the fungus, and a healthy trap plant that probes the conduciveness of the quadrat to ShB at its center. Inoculation of the quadrat hills was done at maximum tillering stage by placing on each of them 5 g of inoculum at the leaf level and another 5 g at the base of the tillers. IR72, a semidwarf, high-yielding variety, was used in all experiments. Unless otherwise specified, all experiments had a 20- 20cm spacing between hills and a fertilizer rate of 120 kg N/ha. All were conducted in the screenhouse using a randomized complete block design with three successive batches of trap plants and the quadrats as the basic experimental unit. The experiment addressing contact frequency was conducted in Sep-Oct 1992. Seedlings were transplanted 14 d after sowing at 18- 18-cm spacing. They were later transplanted during maximum tillering stage according to preset spacing treatments: 15 15 cm, 15 20 cm, 20 20 cm, and 25 25 cm. The effect of leaf wetness duration was analyzed in the dry season (Feb-Apr

1993) with five leaf wetness regimes as treatments: permanently dry (T0), wet for one night (Tl), wet for two consecutive nights (T2), wet for three consecutive nights (T3), and permanently wet (PW). The experiment had five replications. Leaf wetness was obtained during the night with water sprays prior to caging the quadrats with plastic stapled onto wooden frames at 1700 h. Permanent wetness was achieved by continuous water sprays during the day and caging at night. The effect of N supply on the crop was analyzed during Jun-Jul 1992. The experiment had eight replications. Trap plants were grown in plots with three N input levels: no N (N0), 60 kg N/ha (Nl), and 90 kg N/ha (N2). At maximum tillering stage, the trap plants were transferred to the center of the quadrats that had been fertilized at the rate of 120 kg N/ha. At the start of a 3-d probing period in all experiments, trap plants were placed into the quadrats. One hill per quadrat was randomly chosen and five of its tillers were sampled to measure their leaf and stem area (cm 2). The lesion density on the quadrat (LDq, dimension: [lesion number/ area]) was determined by taking the average number of infection points on all the leaves and stems of each tiller. After each probing, the trap plants were removed from the quadrat and potted. The leaf and stem areas were measured from a sample of five tillers. and the lesion density on the trap plant (LD, dimension: [lesion number/area]) was taken by counting the number of infection points on the leaves and stems of all the tillers after a 3-d incubation period. Inoculum efficiency: IE = LD/LDq (dimension: [lesion number/lesion number]) was calculated for each probing period. It represents the progeny-parent ratio in terms of lesions present on the quadrat and on the trap plants. Because it involves lesion densities and not lesion numbers, it takes into account the differences in area of the source and trap plants across the different treatments and batches. IE was analyzed using repeated measures ANOVA with batch as the repeated measures factor (see table). IE generally declines across batches,

1. Relationship between plant spacing (a), leaf wetness regime (b), and N supply (c) and the inoculum efficiency in sheath blight. IRRI, 1992-93.
a T0 = permanently dry, T1 =

wet for one night, T2 = wet for two consecutive nights, T3 = wet for three consecutive nights, and PW = permanently wet.

suggesting that the number of infectious lesions decreased over time. Higher contact frequency facilitated disease spread as shown by the significantly higher IE (see table, P < 0.01) in


IRRN 20:1 (March 1995)

Repeated measures ANOVA for inoculum efficiency (IE) in ShB as affected by contact frequency, leaf wetness regime, and N supply to the crop. a IRRI, 1992-93. Effect of contact frequency Source of variation df 4 3 12 2 (1) (1) 6 (3) (3) 32 (16) (16) 59 MS 0.0023 0.0047 0.0006 0.0079 0.0148 0.0010 0.0002 0.0003 0.0001 0.0008 0.0013 0.0004 F 3.60* 7.47** df 4 4 16 2 (1) (1) 8 (4) (4) 40 (20) (20) 74 Effect of leaf wetness MS 0.0089 0.0509 0.0058 0.2187 0.3329 0.1045 0.0178 0.0269 0.0087 0.0094 0.0113 0.0074 F 1.55 8.84** df 7 2 14 2 (1) (1) 4 (2) (2) 42 (21) (21) 71 Effect of N supply MS 0.0113 0.0855 0.0148 0.1194 0.2327 0.0061 0.0136 0.0166 0.0105 0.0126 0.0146 0.0106 F 0.76 5.80*

Replication (R) Treatment (T) Error (a) Batch (B) Linear (BL) Quadratic (B q) B T BL T BQ T Error (b) EbL EbQ Total

9.88** 18.50** 2.63 0.25 0.23 0.26

23.26** 29.46** 14.12** 1.89 2.38 1.18

9.48** 15.91** 0.57 1.89 1.13 0.99

aFisher ratio values followed by * or ** are significant at P <0.05 or 0.01, respectively.

close plant spacings than in wide spacings (Fig.la). At closer spacing, inoculum from infected tissues is more easily mobilized to healthy tissues of the

same or adjacent plants via mycelial growth. IE was significantly higher (see table, P < 0.01) in the interrupted leaf wetness treatments (T3) than in the and quality production functions were specified ex ante by rice research scientists and specialists of the Texas Agricultural Experiment Station, Texas Agricultural Extension Service, and U. S. Department of Agriculture at Beaumont, Texas. Multiple regression analysis was used to determine the impact of rice production inputs on rice yields and quality. Scientists hypotheses about the relationship between N and ratoon quality are primarily drawn from work done on the main crop. High rates applied to the main crop have been shown to be positively associated with higher harvest grain moisture contents, thus N is not hypothesized to directly affect quality when rice is harvested at the proper moisture. For the ratoon crop, moisture content may vary more widely than that of the main crop because retillering of the ratoon crop is often uneven. Thus, it is hypothesized that high N rates contribute to higher yields but may contribute to lower quality. Statistically significant estimated relationships between N and ratoon field yield and N and ratoon head yield are presented in the figure for producers

continuously dry or in continuously wet treatments (Fig. 1b): alternate dry and wet periods enhance disease progress. While leaf wetness is necessary for the survival and growth of mycelium, dry periods may enhance contacts between hyphae and plant surfaces, thus facilitating penetration. IE declined with increased N supply. IE at 60 and 90 kg N/ha (resulting in 1.8 and 2.2% N content in the leaves, respectively) was significantly lower (see table, P < 0.05) than unfertilized plants (1.7% N content), with a twofold difference between the fertilized and nonfertilized treatments (Fig. 1c). Another independent experiment (not shown) was conducted, which led to the same conclusion. While N content decreases IE, it may enhance other components of the ShB monocycle, such as the rate of lesion expansion, latent period, and infectious period of ShB. These effects are currently being investigated.

Considering rice quality factors in farm management decisions

T. N. Thompson, M. E. Rister, W. R. Grant, Agricultural Economics Department, Texas A&M University, College Station, Texas 77843-2124, USA; G. N. McCauley, F. T. Turner, J. W. Stansel, and A. D. Klosterboer, Agricultural Research Center, Beaumont, Texas 77713, USA

Ignoring the relationship between inputs and quality when making rice farm management decisions may have negative implications. We compared, at the farm level, maximum returns above variable cost (RAVCR ) levels and associated ratoon crop nitrogen (N) levels using traditional profit maximization techniques that ignore rice quality factors with results for modified maximization techniques that incorporate rice quality determining factors, such as head yield, milling yield, and grade. The data used in the analysis are from a three-year survey (1987-89) of Texas rice producers main and ratoon crop production practices. Data for 464 ratoon-cropped ricefields covering more than 22,000 ha were collected. Quantity

using semidwarf varieties irrigated with high pH water. Ratoon crop field and head yield increase with higher levels of N until 90 kg/ha. Above that application level, ratoon head yields decrease while field yields continue to increase, resulting in a conflicting relationship. This relationship may be due to mechanically harvesting rice at improper grain moisture content; however, uneven maturity, ruts in the field, weather, and N rates above those used in previous experimental studies may hinder the replication of previous experimental research. The results, however, do support current Texas recommendations of no more than 112 kg N/ha for semidwarf varieties during the ratoon production phase. Rice producers in the USA are paid not only for quantity but also quality. To determine the impact of recognizing the relationship between rice quality and input usage, yield and quality production functions (head yield, milling yield, and grade) were used to calculate RAVCR . RAVCR-maximizing levels of N were determined using a static $0.15/kg nonquality-dependent rice price and a quality-dependent rice price. With regard to sensitivity of results to N input cost

IRRN 20:1 (March 1995)


Comparison of per hectare RAVC R and RAVC R-maximizing level of ratoon crop N for Texas rice producers using semidwarf rice varieties, considering and ignoring rice quality consequences, 198789 Texas ratoon crop survey, USA. RAVCR-maximizing level of ratoon crop N ignoring quality (kg/ha) RAVCR RAVC R ignoring maximizing RAVCR quality level of considering ($/ha) ratoon crop N quality considering ($/ha) quality (kg/ha) Normal pH water 71.45 112 61.41 89 53.16 65 High pH water 66.57 128 55.21 104 45.60 82 47.48 38.62 30.00 42.47 32.27 24.07 RAVC R N level difference difference ($/ha)a (kg/ha)a

N cost ($/kg)

0.40 0.48 0.57 0.40 0.48 0.05


124 105 84 139 119 98

12 16 19 11 15 16

23.97 22.79 23.16 24.10 22.94 21.53

Estimated relationship between N levels and ratoon crop field and head yield responses to N for semidwarf rice irrigated from the Colorado River, 1987-89 Texas ratoon crop survey, USA.

lgnoring quality minus considering quality.

News about research collaboration

IRRI builds greenhouse for transgenic rice
A new greenhouse for transgenic rice was opened at IRRI in November 1994. The new facility houses transgenic rice, which are rice plants that contain one or two additional genes from other sources. Through gene transfer, scientists hope to improve rices resistance to pests and diseases and tolerance for salinity and flooding, says Dr. John Bennett, IRRI molecular biologist. Eventually, rice scientists also hope to improve rice yield and quality through greater tolerance for drought and reduced dependence on added fertilizer. Once a gene is successfully transferred, a transgenic rice plant needs to be grown and tested in a special confinement facility in order to establish its stability and effectiveness. The self-contained, 400-square meter greenhouse is the latest addition to IRRIs biotechnology facilities. The greenhouse is designed to withstand earthquakes, typhoons, and fires, and meets all of the safety features specified for growing transgenic rice by the National Committee on Biosafety of the Philippines. The greenhouse will be used by scientists from IRRI, the Philippine Rice Research Institute, and the University of the Philippines Los Baos (UPLB). The greenhouse has eight separate bays, each with its own airlock and washroom. The rice plants in bays 1 through 5 will be raised free of pests and pathogens, often through several generations. In bays 6 through 8, samples of these plants will be deliberately subjected to the same insect pests and diseases found in Philippine ricefields. The concept behind the greenhouse is to keep what is inside the greenhouse, in, and what is outside, out. IRRI is permitted by the National Committee on Biosafety of the Philippines to import approved transgenic rice seeds from collaborating laboratories and to produce transgenic rice plants. The Department of Healths Quarantine Service and IRRIs Biosafety Committee (IBC) will regularly monitor the operations of the transgenic greenhouse. The IBC consists of IRRI staff members and representatives of the local community, including UPLB scientists.

levels, three different N prices were considered: $0.40/kg, $0.48/kg, and $0.57/kg. RAVCR-maximizing levels for semidwarf varieties are presented in the table. Ignoring quality, optimal ratoon crop N levels range from 84 to 139 kg/ha and corresponding RAVC Rs range from $45.60 to $71.45/ha. Considering quality consequences, RAVC R-maximizing levels of ratoon crop N range from 65 to 128 kg/ha. RAVC,s considering quality range from $24.07 to $47.48/ha. Analyses indicate that results with and without consideration of quality are not equal. The difference in N levels ranges from 11 to 19 kg/ha and the difference in RAVC R ranges from $21.53 to $24.10/ha. Results indicate maximization of profit when considering rice quality requires lower N application rates compared with maximization of profit without consideration of rice quality for most situations analyzed. Considering the effect of N on quality results in lower profit than when quality effects are ignored.


IRRN 20:1 (March 1995)

Low-cost chopper turns farm residues into useful products

A common scene in rice-growing countries during the dry season is emaciated cattle and water buffalo scouring the ground for the last blade of grass. One reason for this is that farmers often burn rice straw and other farm residues instead of utilizing them for livestock feed. IRRIs Agricultural Engineering Division has developed a low-cost machine that chops farm by products, such as rice straw, into livestock feed,

compost, mulch, and green manure. Called the Flipper (flywheel-type, inclined-axis chopper), the 40-kg, gasoline-powered machine chops 0.5-0.75 t of rice straw per hour. It is easy to run, says Melchor Pasikatan, the IRRI engineer who designed the machine. Operators simply drop rice straw into the hopper. They need not stoop and push the materials as the rice straw slides by gravity into the inclined housing. It is also safe because the blades are designed to suck materials into the chopper housing. Between 1990 and mid- 1994, 12 varieties have been released to Cambodian farmers through OFAT mechanisms. These varieties could potentially be

The Flippers simple design means it can be fabricated in small workshops. It is sold in the Philippines at P8,000 (about US$310) without the engine. Operation and maintenance are simple; the blades, for instance, are reversible and need not be resharpened until they have chopped 8-10 t of materials. It takes only half a liter of gasoline to chop one ton of farm residues such as rice straw and corn stalks. The Flipper has been-field tested at IRRI and in Pangasinan Province, Philippines. grown on 1.1 million ha. The OFAT system is now being recommended by IRRI for use in other countries, says Dr. Chaudhary. The IRRI Library is the worlds largest single-crop library with a 100,000-entry general library catalog, about 2,000 serials, maps, and atlases, rare documents, and many translations of rice articles. In addition to producing a comprehensive rice bibliography, it provides clients around the world with document delivery, database searches, reference service, a newspaper index, and a directory of forthcoming scientific meetings.

Cambodian farmers actively involved in technology testing

Cambodian farmers are involved in a unique systemcalled on-farm adaptive trial (OFAT)for identifying and popularizing new rice varieties. Farmers use their own fields, labor, inputs, and crop management skills to test new varieties. The testing fields are near roads so that more people can observe the new varieties being tested. For comparison purposes, farmers plant their own best variety beside a maximum of three new test varieties. OFAT is a very cost-effective tool for the transfer of technology, says Dr. R. C. Chaudhary, global coordinator of the International Network for Genetic Evaluation of Rice (INGER). Since farmers themselves are involved in testing the technology, using their best bets as the control, any superior variety gets 100% acceptability by the farming community. Each year, INGER provides up to 2,000 varieties to Cambodian researchers who identify varieties for farmers to test through OFAT. The trials provide scientists with feedback about each varietys desirable and undesirable characteristics and serve as a mechanism to multiply seeds and distribute the new varieties faster. Started in the 1990 dry season, OFAT is a collaborative activity of the Cambodian Ministry of Agriculture, nongovernment organizations, and the Cambodia-IRRI-Australia Rice Project.

IRRI Library reaches out

The IRRI Library has a mandate to provide scientific information on rice to researchers around the world. To meet clients and librarians on their home grounds, IRRI librarian Ian Wallace recently visited 12 libraries in Vietnam and Cambodia. Generally, IRRI library services in these countries are little known and underused. Things are changing as a result of those visits. The IRRI Library, for example, now has improved access to local rice literature, including theses from Vietnama significant change since most theses in the library collection have usually come from the United States and the Philippines. Electronic mail communication has been established between the IRRI Library and Cantho University, and its chief librarian was trained for two weeks at IRRI. Future activities include training librarians and an overhaul of the catalog at the library of the Cambodia-IRRIAustralia office in Phonm Penh, which has Cambodias biggest collection of rice literature. We want to establish contacts with local users of IRRI library materials as well as find out what they need so that we can serve them better, says Mr. Wallace. They can also help us trace local rice literature and get it into the IRRI Library.

Anaerobic seeding ready for farmer evaluation

Anaerobic seeding can reduce most of the constraints of wet seeded rice. It involves seeds tolerant of oxygen-deficient, or anaerobic, conditions such as in flooded fields. These anaerobic seeds are sown about 5-10 mm under the soil. Anaerobic seeds covered by the soil are less vulnerable to pest damage. The seeds are protected from flooding, although there is enough soil moisture to nurture them. Roots growing under the soil are better anchored. The resulting rice plants also compete well with weeds and may lead to reduced herbicide use. Anaerobic seeding technology is now ready for evaluation on farmers fields, says Dr. Minoru Yamauchi, IRRI plant physiologist. Dr. Yamauchi has identified aus (early summer, rainfed rice) and

IRRN 20:1 (March 1995)


deepwater rice from northern India and Bangladesh that are better suited for anaerobic seeding. While grain yield was the same compared with that of the high-yielding semidwarf varieties, anaerobic seeding produces more seedlings that have

increased shoot weight, height, and leaf development, says Dr. Yamauchi. His research on anaerobic seeding is a collaborative project of the Japan International Research Center for Agricultural Sciences (JIRCAS), Central Agriculture Research Institute (Myanmar), Vietnam

Agricultural Science Institute, Cuu Long Delta Rice Research Institute (Vietnam), Philippine Rice Research Institute, and IRRI. The Government of Japan has renewed the project for another five years, beginning in 1995, this time with a focus on water stress problems.

Encouraging farmers to experiment with IPM

The Vietnamese Ministry of Agriculture and Food Industry (MAFI) and the IRRIcoordinated Rice Integrated Pest Management (IPM) Network are collaborating to evaluate a campaign strategy that encourages rice farmers to experiment with avoiding early-season insecticide sprays. Farmers will be encouraged to experiment with not applying insecticides to small parts of their fields for 30-40 days after planting and to compare the yield there with the rest of the field at the end of the season, explains Dr. K. L. Heong, IRRI entomologist and IPM network coordinator. They can then decide for themselves whether early insecticide use is necessary. Rice plants can compensate for earlyseason damage by leaffeeding insects, and no yield loss occurs as a result. Thus, by not spraying early, farmers can reduce their costs and also lessen the harmful effects of insecticides on their own health and on the environment. The campaign is a simple, inexpensive way to get farmers to experiment with IPM, says Mrs. Vo Mai, vice director of the Plant Production and Protection Department, MAFI. It will reinforce the IPM field school training that is now going on. Radio spots, posters, and comic strips were pretested before the campaign was launched in late 1994 in Long An Province. If successful, similar campaigns will be conducted in other provinces.

Climate changes and rice

Using computer models, IRRI researchers have evaluated the likely effects of climate change on rice yield and production in East and Southeast Asia. Climate change scenarios predicted by the General Fluid Dynamics Laboratory, the Goddard Institute of Space Studies, both in the United States, and the United Kingdom Meteorological Officeplus crop growth modelswere used in the computer models. A rise in temperature generally reduces yields. A rise in carbon dioxide level increases yields, says Dr. Robin Matthews, a crop modeler at IRRIs

Agronomy, Plant Physiology, and Agroecology Division. But it is hard to determine the exact effect of climate change on existing crop yields at a specific site as it depends on the combination of these two factors. All the computer models used predicted production declines in Thailand, Bangladesh, and southeastern China. Increases were predicted for Malaysia, Indonesia, northern China, and western India. Such changes are likely to have a significant effect on future rice production and trading relationships within East and Southeast Asia.

1994-95 short course in rural development, project planning, and resource management
Many middle management professionals associated with rural development and natural resource management are trained in biological or technical fields. They often have only a limited understanding of the economic and social impacts of their work. The Development Studies Program of the Department of Agricultural and Resource Economics, University of New England, Australia, offers a short course program that allows these people to upgrade their skills through training in the latest economic techniques and approaches to environmental or natural resource management and rural development problems. It also provides an opportunity to share experiences and develop future information networks with peers from similar countries and backgrounds. For information, contact the Executive Officer, Development Studies Program, P. O. Box U298, University of New England, Armidale NSW 2351, Australia.

Biotechnologia Habana 95
The Center for Genetic Engineering and Biotechnology, in collaboration with the United Nations Educational, Scientific, and Cultural Organization and the Food and Agriculture Organization of the United Nations, is sponsoring an international meeting on biotechnology. The event, with the theme Biotechnologia Habana 95: new opportunities in plant, animal, and industrial technology, will be held in Havana, Cuba, on 12-17 Nov 1995. For more information, contact the Organizing Committee, Biotechnology Habana 95, P.O. Box 6162, Havana, 10600, Cuba.


IRRN 20:1 (March 1995)

New IRRI publications

Breaking the yield barrier. 1994. 141 pages. US$12.00 in highly developed countries (HDC), US$3.00 in less developed countries (LDC), plus US$4.50 airmail or US$1.50 surface postage. In the next 30 years, global rice production must be raised by another 300 million tons. With little scope for expanding the irrigated areas, this challenge must be met by increased yields from the existing riceland. IRRI invited scientists to attend a workshop to help delineate the boundaries of our present understanding of yield potential, to explore the frontiers of science in plant biology relevant to yield formation, and to identify the most promising approaches to develop rice varieties with higher yield potential. This book is the result of that meeting. It contains papers by IRRI scientists and extended abstracts of the invited papers. There is also a set of recommendations for future research as prepared by the participants. The output of this meeting will help to establish the scientific foundation for our efforts to break through the existing yield potential barrier, a barrier that must be overcome in an environmentally conscious manner to ensure long-term food security in Asia.

Rice blast disease. 1994. 626 pages. Customers in HDC should send orders to CAB International, Wallingford, Oxon OX10 8DE, UK; US$25.00 in LDC, plus airmail US$18.00 or surface mail US$2.50 postage. Rice blast disease has long been recognized as the most explosive and potentially damaging disease of the rice crop. The complexity of the disease and the formidable challenge of the diversity of the pathogen have dictated that no institution or group can expect to solve the problem in isolation. The conference on which this publication is based was the fourth such international gathering of scientists from universities and international and national research centers working on this problem. More than 120 scientists from 20

countries shared their latest findings, which are published in this book. From the contents, it is clear that tremendous progress has been made in our understanding of, and our ability to manipulate, the rice-pathogen system in order to reduce loses. The book was copublished with CAB International.

New publications
Triops 1994: tropical scientific books Catalog listing scientific books from both academic publishers and from international institutes and universities on natural and applied sciences in the tropics and the subtropics. Order from Triops, Toeche-Mittler Distribution, Hindenburg Street 33, D-64295 Darmstadt, Germany. Fax: 06151/31 4048.

2nd International Symposium on Systems Approaches for Agricultural Development

The second International Symposium on Systems Approaches for Agricultural Development will be held at IRRI on 6-8 Dec 1995. The objectives of the symposium are to review the status of applications of systems research and modeling in agricultural research, with specific focus on countries where agricultural develop-

ment is facing major challenges, and to promote international collaborative activities and to increase awareness of the opportunities for using systems approaches as a tool in research and planning. It is organized by IRRI, the International Consortium for Agricultural Systems Applications (ICASA), and the Systems Analysis and Simulation in Rice Production Network (SARP). For details, contact Dr. P. K. Aggarwal, IRRI.

Advanced courses from IHE

The International Institute for Infrastructural Hydraulic and Environmental Engineering (IHE) in Delft, The Netherlands, is offering two senior advanced courses on polder development and appropriate modernization and management of irrigation systems. The first course (30 May-23 Jun) is designed for senior professionals working in land and water development, particularly in irrigation design, management, operation, and maintenance. The second course (19 Sep-6 Oct) is aimed at senior professionals involved in the development of lowlands and deltaic areas. For information, contact the Registrar of IHE, P.O. Box 3015,2601 DA, Delft, The Netherlands.

Rice dateline
2-4 May 16 May 16-17 May 1st Biological Nitrogen Fixation Working Group Meeting, IRRI ...................................................... J. K. Ladha, IRRI

INGER Project Support Team Meeting, IRRI ........ R. C. Chaudhary, IRRI PHILRICE-UPLB-IRRI Work Plan Meeting, Nueva Ecija, Philippines ............................................. G. L. Denning, IRRI

IRRI group training courses for 1995

IRRI provides a limited number of scholarships for participation in its shortterm group training courses for 1995. To be considered for an IRRI-funded scholarship, a scientist must be affiliated with a national institution that has an official collaborative agreement with

IRRI in a rice-related research and training project. A scientist interested in an IRRIfunded scholarship should apply directly to his or her institution and not to IRRI. IRRI also accepts scientists from other institutions and agencies for the courses if they are working in rice or rice-related areas. Their applications to participate in courses must be endorsed to IRRI by their employer and specify funding sources to

IRRN 20:1 (March 1995)


cover costs. IRRI's group course training fee is approximately US$1,200/month; this does not include participants roundtrip international airfare, enroute expenses, or shipping allowance upon return home.
Date 5 Jun-30 Jun 17 Jul-8 Sep 24 Jul-15 Sep 14-25 Aug 14 Aug-3 NOV 2-27 Oct 9 Oct-1 Dec 6 Nov- 15 Dec 13-24 NOV

The courses are conducted at IRRI headquarters unless otherwise indicated. For additional information, contact the Head, Training Center, IRRI.

Cornell University grants

Cornell University, USA, is sponsoring fellowships in environmental research in international agriculture. They are intended to encourage creative and imaginative research aimed at integrating environmental and natural resource protection with mainline international agriculture research and development. Qualified disciplines include ecology, botany, zoology, resource management, wildlife protection, climatology, entomology, atmospheric sciences, microbiology, and other environment-oriented fields. The program provides a stipend and research support for visiting scientists at international agricultural research centers. Funding is shared between the Rockefeller Foundation and the host institution. There is no formal deadline for applications, but it is strongly urged that they be submitted at least six months before the requested starting date. Applications should include a written project developed and submitted jointly by the candidate and the host center; curriculum vitae and copies of recent and relevant publications; a letter from the host center's administration recommending the candidate and the host center; and two letters of recommendation from established scientists familiar with the applicant's qualifications. Centers in Colombia, India, Mexico, Nigeria, Peru, Philippines, and Taiwan, China are participating in the program, but others will be considered. For more information, contact Ms. Lyn Hogan, Communications Officer, The Rockefeller Foundation, 1133 Avenue of the Americas, New York NY 10036, USA. Tel: (212) 869 850

Course Training on Video Production Integrated Pest Management a (University of the Philippines Los Baos) Adaptive Research with a Farming Systems Perspective Scientific Programming Genetic Evaluation and Utilization Geographic Information Systems Rice Production Researchb (Pathum Thani Rice Research Center, Thailand) Engineering for Rice Agriculturec (India Institute of Technology, Kharagpur, India) Gender Perspective and Analysis in Rural Developmenr d (International Institute for Rural Reconstruction, Silang, Cavite, Philippines) Frontiers of Social Science Research Methods for Agricultural Systems Analysis

20 Nov-1 Dec

aUniversity of the Philippines Los Baos, Philippine Rice Research Institute, Southeast Asian Regional Center for Graduate bThailand Rice Research Institute and IRRI. cIndia Institute of Technology and IRRI.

Study and Research in Agriculture, and IRRI

dInternational Institute for Rural Reconstruction and IRRI.

Rice literature update reprint service

Photocopies of items listed in the Rice literature update are available from the IRRI Library and Documentation Service. Reprints of original documents (not to exceed 40 pages) are supplied free to scientists of developing countries. Rice scientists elsewhere are charged US$0.20 for each page or part of a page copied, plus postage. Make checks or money orders payable to Library and Documentation Service, IRRI. Address requests to Library and Documentation Service, IRRI. E-mail: C.AUSTRIA@CGNET.COM

IRRI address
International Rice Research Institute P.O. Box 933 1099 Manila, Philippines Tel: (63-2) 818-1926 Fax: (63-2) 818-2087 Telex: (ITT) 40890 RICE PM E-mail: Postmaster@IRRI.CGNET.COM

Path analysis focus expansion in rice sheath blight, by N.P. Castilla et al, 19(4) (Dec 1994), 27-28. The equation on page 28 should have been written as

Call for news

Individuals, institutions, and organizations are invited to tell readers about upcoming events in rice research or related fields in the rice dateline. Send announcements to the Editor, International Rice Research Notes, IRRI.
44 IRRN 20:1 (March 1995)

Instructions for contributors

NOTES General criteria. Scientific notes submitted to the IRRN for possible publication should be original work, have international or pannational relevance, be conducted during the immediate past three years or be work in progress, have rice environment relevance, advance rice knowledge, use appropriate research design and data collection methodology, report pertinent, adequate data, apply appropriate statistical analysis, and reach supportable conclusions. Routine research. Reports of screening trials of varieties, fertilizer, cropping methods, and other routine observations using standard methodologles to establish local recommendations are not ordinarily accepted. Examples are singleseason, single-trial field experiments. Field trials should be repeated across more than one season, in multiple seasons, or in more than one location as appropriate. All experiments should include replications and an internationally known check or control treatment. Multiple submissions. Normally, only one report for a single experiment will be accepted. Two or more items about the same work submitted at the same time will be returned for merging. Submitting at different times multiple notes from the same experiment is highly inappropriate. Detection will result in the rejection of all submissions on that research. IRRN categories. Specify the category in which the note being submitted should appear. Write the category in the upper right-hand corner of the first page of the note. GERMPLASM IMPROVEMENT genetic resources genetics breeding methods yield potential grain quality pest resistance diseases Insects other pests stress tolerance drought excess water adverse temperature adverse soils other stresses integrated germplasm improvement Irrigated rainfed lowland upland flood-prone (deepwater and tidal wetlands) seed technology CROP AND RESOURCE MANAGEMENT soils soil microbiology physiology and plant nutrition fertilizer management inorganic sources organic sources crop management integrated pest management diseases insects weeds other pests water management farming systems farm machinery postharvest technology economic analysis ENVIRONMENT SOCIOECONOMIC IMPACT EDUCATION AND COMMUNICATION RESEARCH METHODOLOGY Manuscript preparation. Arrange the note as a brief statement of research objectives, a short description of project design, and a succinct discussion of results. Relate results to the objectives. Do not Include abstracts. Do not cite references or include a bibliography. Restrain acknowledgments. Manuscripts must be in English. Limit each note to no more than two pages of doublespaced typewritten text. Submit the original manuscript and a duplicate, each with a clear copy of all tables and figures. Authors should retain a copy of the note and of all tables and figures. Apply these rules, as appropriate, in the note: Specify the rice production ecosystems as irrigated, rainfed lowland, upland, deepwater, and tidal wetlands. Indicate the type of rice culture (transplanted, wet seeded, dry seeded). If local terms for seasons are used, define them by characteristic weather (wet season, dry season, monsoon) and by months. Use standard, internationally recognized terms to describe rice plant parts, growth stages, and management practices. Do not use local names. Provide genetic background for new varieties or breeding lines. For soil nutrient studies, Include a standard soil profile description, classification, and relevant soil properties. Provide scientific names for diseases, insects, weeds, and crop plants. Do not use common names or local names alone. Quantify survey data, such as Infection percentage, degree of severity, and sampling base. When evaluating susceptibility, resistance, and tolerance, report the actual quantification of damage due to stress, which was used to assess level or incidence. Specify the measurements used. Use generic names, not trade names, for all chemicals. Use international measurements. Do not use local units of measure. Express yield data in metric tons per hectare (t/ha) for field studies and in grams per pot (g/pot) for small-scale studies. Express all economic data in terms of the US$. Do not use local monetary units. Economic information should be presented at the exchange rate US$:local currency at the time data were collected. When using acronyms or abbreviations, write the name in full on first mention, followed by the acronym or abbreviation in parentheses. Use the abbreviation thereafter. Define any nonstandard abbreviations or symbols used in tables or figures in a footnote, caption, or legend. Tables and figures. Each note can have no more than two tables and/or figures (graphs, illustrations, or photos). All tables and figures must be referred to in the text; they should be grouped at the end of the note, each on a separate page. Tables and figures must have clear titles that adequately explain the contents. Review of notes. The IRRN editor will send an acknowledgment card when a note is received. An IRRI scientist, selected by the editor, reviews each note. Reviewer names are not disclosed. Depending on the reviewer's report, a note will be accepted for publication, rejected, or returned to the author(s) for revision. (continued on back cover)