International Rice Research Notes

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

Contents
Vol. 22 No. 1 1997
Germplasm improvement Genetic resources Spontaneous interspecific hybrids in Oryza in Lao PDR 4 Genetics Genetics of anthocyanin pigmentation in rice 5

Focus on rice genetics

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

Breeding methods Death of thermosensitive genic male sterile seedlings in Malaysian ricefields 6 Performance of rice mutants in different seasons 7 Using anther culture to generate fertile, doubled-haploid interspecific progeny 7 Response of rice anther culture to short-day treatment 8 Expression of an engineered cysteine proteinase inhibitor (Oryzacystatin-l 86) in transgenic rice plants 9 An improved biolistic method for transformation and production of fertile transgenic Pusa Basmati rice plants 10 Genetic analysis of a d1 chimeric rice plant 12 Increasing yield potential of irrigated rice through recurrent selection 13 Transgenic rice plants expressing rice yellow mottle virus coat protein gene 14 Plant regeneration toward transformation of several javanica rice cultivars 16 Variation among plants regenerated from microspore-derived cell suspension protoplasts of rice 17 Toward introgression of high response to anther culture into indica rices 18 Random mating of composite populations for improving restorers in rice 19 Yield potential Influence of brassinosteroid on rice seedling growth 20

Grain quality Physical and milling characters of popular Maruteru rice varieties 22 Variability of quality indices in aromatic rice germplasm 22 Biochemical composition of principal components of rice seeds Pest resistance Activity of the promoter of a rice lipid transfer protein gene in transgenic rice 24 Pest resistancediseases Sources of resistance to sheath blight 25 Effect of blast disease on rice yield 25

23

IRRN production team

Editors: Carolyn Dedolph and Domenic Fuccillo Assistant editor: Teresita Rola Layout and design: Erlie Putungan Artwork: Erlie Putungan

2 IRRN 1997

Pest resistanceinsects Resistance of varieties derived from Oryza sativa/Oryza officinalis to brown planthopper in the Mekong Delta, Vietnam 26 Erra Mallelu, Kavya, and Orugallu: fine-grained, gall midge (biotype 1)-resistant rice varieties 27 Stress toleranceother stresses Flowering behavior of rainfed lowland rice varieties during dry season in West Bengal, India 28 Integrated germplasm improvementirrigated Vijetha: a high-yielding, short-duration rice variety for Andhra Pradesh, India 29 Two rice hybrids released in Andhra Pradesh, India 30 Performance of hybrid rice in irrigated lowlands, Uttar Pradesh, India 30 FKR42 and FKR44: irrigated rice varieties released to farmers in Burkina Faso 31 Integrated germplasm improvementupland Evaluating local germplasm for the upland rice ecosystem in western Nepal 31 FKR33: a popular upland variety in Burkina Faso 32 Seed technology Revitalizing stored rice seeds 33

Effect of Azolla caroliniana and Sesbania rostrata on rice yield 40 Contribution of green manure in controlling the loss of applied fertilizer nitrogen from rainfed rice soil 41 Crop management Influence of planting dates on productivity of traditional scented rice varieties 42 Seedling vigor affects stand establishment and performance of flood-prone lowland rice 42 Integrated pest management A survey of rice constraints in the Mekong Delta 43 Integrated pest managementinsects Impact of methyl parathion on the natural enemies of rice insect pests in Cambodia 44 Effect of soil amendments on grain yield and incidence of rice leaffolder in iron-toxic soils of north Orissa, India 45 Integrated pest managementweeds Crop rotation in red rice control 46 Herbicide use and occurrence of Echinochloa spp. in ricefields in dry and intermediate zones of Sri Lanka 47 Integrated weed management through smother intercrops in rainfed lowland rice 47 Research methodology

Crop and resource management Physiology and plant nutrition Effect of hydroquinone and phenylhydrazine on yield and nitrogen use efficiency of rice 34 Fertilizer management Integrated nutrient management in a rice-based crop sequence 35 Transplanting geometry improves timing of uptake of deep point-placed P by rice hills 36 Fertilizer managementinorganic sources Effect of Zn on grain yield and Zn uptake by lowland rice in South Gujarat 37 Optimization of potassium application in acid soils of Tamil Nadu 37 Increased yield of lowland rice with late N application in the reproductive phase and at high N rates 38 Fertilizer managementorganic sources Management of urea briquettes containing diammonium phosphate increases rice yields of small farmers 39

Modified method for determination of amylose content using a single rice kernel 48 Comparison of two scoring systems for evaluating stem rot resistance in rice, and a new proposed rating scale 48 A procedure for continuous screenhouse rearing of the yellow stem borer 50

Vol. 22. No. 1

Germplasm improvement
Genetic resources
Spontaneous interspecific hybrids in Oryza in Lao PDR
S. A. Rao, Lao-IRRI Project, P. O. Box 4195, Vientiane, Lao PDR; V. Phetpaseut and C. Bounphanousay, National Agricultural Research Center, Ministry of Agriculture and Forestry, Vientiane, Lao PDR; and M. T. Jackson, IRRI

Three wild species of rice, Oryza rufipogon, O. nivara (diploid AA genome), and O. granulata (diploid, GG), are found in Lao PDR. Recent reports of O. officinalis still need to be confirmed. While collecting rice germplasm in four central and southern provinces, we found several O. rufipogon and O. nivara populations growing near cultivated rice (O. sativa) in roadside ditches, isolated ponds, canals, and in ricefields. In six populations we discovered what appeared to be intermediate forms between wild and cultivated rice (see table)probably spontaneous hybrids or/and different progenies at segregations. The inter-

mediate forms were mixed with the wild species and cultivated rice in five populations (see figure). We observed only intermediate weedy forms, however, in one population (Thapangthong). Flowering is synchronous in cultivated rice, making the duration short. The wild species, however, flower over a much longer period. For example, O. nivara (an annual) flowers in September and completes its life cycle by the end of October. O. rufipogon (a perennial) flowers in October and completes its life cycle in December or January of the next year. So the possibility of cultivated and wild rices hybridizing is real. Cultivated rice is mostly selfpollinated, while in the wild species (particularly O. rufipogon ), outcrossing has been reported to be as high as 50%. We believe the gene flow must have been from the cultivated to the wild forms. Intermediate weedy populations, such as those we identified, develop from such spontaneous hybrids.

Panicles of wild, weedy, and cultivated rice (left to right) found in a farmers field in Salakham, Lao PDR.

Location and habitat of six hybrid populations of rice found in Lao PDR.

Bolek population Province/ town District Village Coordinates Altitude (m) Habitat Predominant cultivated variety Wild species Hybrid features Remarks Vientiane Xaithani Bolek 18 05' N, 103 06' E 180 Pond, adjacent to ricefields Not identified

Nongpin population Vientiane Chanthabouli Nongpin 18 03' N, 102 46' E 180 Pond, 50 ha Idomcultivated for 20 yr, nonglutinous O. rufipogon Backcrossed to cultivated rice Wild and weedy forms

Salakham population Vientiane Hatsayfong Salakham 18 02' N, 102 45' E 180 Ricefield, alongside lake, 20 ha Kao chao malee slightly improved variety from Thailand O. rufipogon Backcrossed to cultivated rice Cultivated and weedy forms, wild forms nearby

Mahaxai population Khammouane Mahaxai Ten 17 28' N, 105 15' E 220 Roadside pond lloupe - cultivated for several years O. nivara More vigorous than wild rices Wild, with some intermediate types

Thapangthong population Savannakhet Thapangthong Nahuahay 15 50' N, 105 55' E 170 New pond, 200 m 2, in forest clearing Not identified

Samakhixai population Attapeu Samakhixai Sekhaman 14 50' N, 106 52' E 150 Roadside canal, 300 m 2 RD6 an improved glutinous variety Not identified Segregating for awn and glume traits Wild, with some intermediate types

O. rufipogon More vigorous than wild rices Wild, with some intermediate types

Not identified Uniform, mimicking cultivated rice Weedy form only

IRRN 1997

From the four populations we studied, the predominant rice varieties cultivated nearby are most likely to have hybridized with the wild species (see table). In two populations we could not identify either the wild species or the probable cultivated varieties. The intermediate forms had characters from both wild and cultivated rice. Until flowering, the intermediate forms resembled the cultivated forms for most morphological characters (culm size, leaf blade length

and width, and tillering) and in gross morphology. After flowering, however, their panicle and grain characters made them easy to distinguish. The intermediate forms also produced panicles and grains that resembled the wild formsslightly larger than the wild forms, but considerably smaller than the cultivated forms. We also observed some of the dominant characters of cultivated rice, such as leaf sheath pigmentation. All the wild forms had long green bristles

(absent in the cultivated forms), and the intermediate forms had red or purple bristles of intermediate length. We observed segregation for bristle length and color in the Samakhixai population, and for awn and grain characters in the Nongpin population. The weedy forms are useful for identifying desirable segregates combining traits from the wild and cultivated forms. It may be worthwhile to screen the weedy forms for stress resistance traits.

Genetics
Genetics of anthocyanin pigmentation in rice
A. R. Panda and P. M. Mohapatra, Plant Breeding and Genetics Division, Central Rice Research Institute (CRRI), Cuttack 753006, India
Behavior of parents, F 1 , and F 2 with respect to purple pigmentation (15 d after seeding).

Plants tested (no.) Parents Khao Y khao TK deep straw 34-774 Jalamagna NDGR 410 Kariawa

Observed frequency Purple Green (%) (%)

Expected ratio of purple to green

c 2

We studied the inheritance of anthocyanin pigmentation of leaf sheaths in rice for eight crosses during the 1995 wet season at CRRI. Three deepwater varieties, Jalamagna, NDGR 410, and Kariawa (with purple pigmentation), and two tall indica varieties (TK deepstraw 34-774 and Khao Y Khao) were crossed during the 1993 wet season. Seeds of the parents, their F1 s, and F2 s were sown in small trays (50 40 cm). Twenty seedlings of each parent, 15 seedlings of each F1, and about 200300 seedlings of each F2 population were uprooted and washed 15 d after sowing. The frequency of plants with purple leaf sheaths was determined. All F1 seedlings of crosses between purple and green parents had purple pigmentation like that of the purple parent, indicating the trait's dominance (see table). F2 populations of crosses between purple and green parents segregated into a 9 purple-7 green ratio, implying that two dominant complementary genes control purple pigmentation. F1 and F2 plants from two purple parents were all purple, and those from two green parents were all green.

20 20 20 20 20 15 15 15 15 15 15 15 15 255 276 266 226 287 328 305 250

0 0 20 20 20 15 15 15 15 15 15 15 0 140 158 151 120 173 201 305 0

20 20 0 0 0 0 0 0 0 0 0 0 15 115 118 115 106 114 127 0 250 143.4:111.6 155.3:120.8 149.6:116.4 127.1: 98.9 161.4:125.6 184.5:143.5 (9:7) (9:7) (9:7) (9:7) (9:7) (9:7) 0.19 0.11 0.03 0.91 1.89 3.37 0.50-0.75 0.50-0.75 0.75-0.90 0.25-0.50 0.10-0.25 0.05-0.10 -

F1

Khao Y khao/Jalamagna Khao Y khao/NDGR 410 Khao Y khao/Kariawa TK deep straw 34-774/Jalamagna TK deep straw 34-774/NDGR 410 TK deep straw 34-774/Kariawa Jalamagna/NDGR 410 Khao Y khao/TKdeep straw 34-774 Khao Y khao/Jalamagna Khao Y khao/NDGR 410 Khao Y khao/Kariawa TK deep straw 34-774/Jalamagna TK deep straw 34-774/NDGR 410 TK deep straw 34-774/Kariawa Jalamagna/NDGR 410 Khao Y khao/TK deep straw 34-774

F2

The data are consistent with the two dominant complementary genes for anthocyanin pigmentation of leaf sheath at the seedling stage. Purple pigmentation occurred when both genes were present; the absence of either or both resulted in green pigmentation like that in the green parents.

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

Vol. 22. No. 1

Breeding methods
Death of thermosensitive genic male sterile seedlings in Malaysian ricefields
H. Kato and M. Sobri, Japan International Research Center for Agricultural Sciences, Ohwashi 1-2, Tsukuba, Ibaraki 305, Japan; and Guok Hup Ping, Malaysian Agricultural Research and Development Institute (MARDI), 13200 Kepala Batas, Seberang Perai, Penang, Malaysia

One hundred and fifty-four thermosensitive genic male sterile (TGMS) lines were introduced in Malaysia from the National Agricultural Research Center in Japan. The lines were bred from indica and japonica crosses and comprised 31 family lines, ranging from B1F9 to F4. These TGMS lines were transplanted in ricefields at MARDI during the 1994-95 dry season (NovFeb) and the 1995 wet season (AprAug). Some of the plants in 18 family lines died during the dry season (see figure). Seedlings died in four of the five lines of the H91-4 family line, and no sterile plants were found (see table). In

Dead seedlings in Malaysian ricefield, 3 wk after transplanting.

the next generation of TG-3, which originated from the fertile plants of the TG-3 line, plants were segregated into fertile and sterile. This trait came from PL12, which contains a single and recessive TGMS gene. The segregation ratio of the TGMS plants, however, was less than the expected ratio (1/4). Three out of five lines of the H92-62 family line also had seedling deaths. TG-22, TG-23, TG-24, and TG-25 had some sterile plants. In the next generation of TG-23, plants were segregated into

Dead seedlings in four TGMS family lines. MARDI, Penang, Malaysia. 1994-95.

Line

Family line

Combination

Generation

Dead/total Sterile plants seedlings (no.) 1/20 0/20 3/20 1/20 2/20 0/20 6/20 3/10 12/20 4/20 4/20 8/20 11/20 10/20 4/20 7/20 5/20 2/20 5/20 2/20 0 0 0 0 0 0 2/14 2/7 2/8 6/16 0 0 1/9 1/10 0 3/13 3/15 0 0 1/18

Next generation

TG-1 TG-2 TG-3 TG-4 TG-5 TG-21 TG-22 TG-23 TG-24 TG-25 TG-41 TG-42 TG-43 TG-44 TG-45 TG-56 TG-57 TG-58 TG-59 TG-60

H91-4

H89-1//H89-1/ Mangetsumochi

B 1F 9

Fertlle; sterile segregation

H92-62

H89-1/X88

F10

Fertile; sterile segregation

H92-150

H89-1/Mansoek//H87-35

F9

All fertile plants All fertile plants

H93-106

PL12/90SL495

F7

Fertile, sterile segregation

fertile and sterile. The same phenomenon was observed in the H93-106 family line. Thirteen out of 18 family lines showed the same phenomenon. All five lines of the H92-150 family line showed more seedling deaths than the lines previously mentioned. TG-43 and TG-44 had a few sterile plants. All plants were fertile in the next generation of TG-43 and TG-44. Two out of 18 family lines showed the same phenomenon. This type of seedling death rate has not been observed in nonTGMS lines introduced from Japan, and the phenomenon has never occurred in Japan. The 30-yr average temperature during the seedling growth month for Tokyo is 18.4 C and that for Kuala Lumpur is 25.9 C. Seedling death was observed only during high temperatures. Germination of TGMS seeds was also poor. We suspect that the high temperature caused the TGMS seedlings to die. The phenomenon observed in H91-4 can be explained as follows: all the TGMS seedlings died from high temperature, so no TGMS plants lived to flower in the hot dry season. But the TGMS gene in heterozygous plants was expressed and both fertile and sterile plants were produced in the next generation. The phenomenon of H92-62 was different from that of H91-4. Perhaps only a few of the TGMS seedlings died from high temperature. For H92-150, we believe most of the TGMS seedlings and the fertile plants, including the TGMS heterozygous plants, died from the high temperature. The conclusions drawn from this study are speculative and need further confirmation through studies conducted with controlled temperature. Family lines with dead seedlings, however, cannot be used to produce hybrid seed. Further selection has to be done to eliminate this trait. Thirteen out of 31 family lines did not show this phenomenon and can therefore be used as parents for hybrids.

6 IRRN 1997

Performance of rice mutants in different seasons

N. Kulkarni, A. Gangaram, and M. Kashikar, Agricultural Research Institute, Rajendranagar, Hyderabad 500030, India

Samba Mahsuri, a quality rice variety grown in Andhra Pradesh, India, has a 150-d duration. Through mutation breeding, early mutants of 120-140 d duration were spotted. We evaluated 12 of them in the M5 (monsoon 1994), M6 (winter 1994-95), and M7 (monsoon 1995) generations along with their parents.

The experiment was laid out in a randomized block design in a 7.2-m 2 plot with three replications and 15- 15-cm spacing. The low temperature in the monsoon season ranged from 22.3 to 24.4 C during the 2 mo of seedling and vegetative growth. During the corresponding period in the winter, the low temperature was 9.2-15.5 C, resulting in a 19 d longer flowering duration (Table 1). Individual mutants behaved differently in the two seasons, with flowering delayed from 5.0 to 30.5 d in the winter compared with the monsoon season. Among these, M12 and M41

showed the least delay in flowering (5.0 and 11.5 d, respectively). Grain yield in the winter was more than during the monsoon season because of the dry weather and prolonged duration. The pooled analysis of variance (data not shown) indicated that seasonal effects were significant (Table 2). Differences among the mutants and their interactions with seasons were also significant. Mutants suitable for one or both seasons were identified: M34, M41, and M47 for the monsoon season; M9, M16, and M22 for the winter; and M16 and M22 for both seasons.
Table 2. Mean sum of squares from pooled ANOVA. Days to 50% flowering 433.0**a 2444.5** 29.2** 1.1 44.9 67.1 9.3 Grain yield (t ha -1) 694.0** 2335.0** 266.0** 94.5 52.0 64.9 43.9

Table 1. Mean values for days to flowering and grain yield of different rice mutants and their parent varieties. Days to 50% flowering Mutant Monsoon 1994 1995 M7 M8 M9 M11 M12 M14 M16 M22 M34 M41 M47 Samba-Mahsuri (check) Mean SE 109 100 91 97 110 107 99 105 100 105 114 120 105 0.4 112 102 94 105 106 108 101 108 102 102 110 12 2 106 0.3 Mean 110.5 101.0 92.5 101.0 108.0 107.5 100.0 106.5 101.0 103.5 112.0 121.0 105.5 Winter 1994-95 129 127 123 123 113 125 125 127 114 115 140 140 125 0.4 Difference 18.5 26.0 30.5 22.0 5.0 17.5 25.0 20.5 13.0 11.5 28.0 19.0 19.5 Monsoon 1994 1995 4.8 6.5 5.1 4.0 4.5 5.3 6.2 5.8 6.2 6.9 6.6 4.9 5.6 0.2 5.3 3.4 4.7 4.7 4.6 5.1 3.4 5.2 5.2 5.3 4.4 5.1 4.7 0.3 Grain yield (t Mean 5.1 5.0 4.9 4.4 4.6 5.2 4.8 5.5 5.7 6.1 5.5 5.0 5.2 ha-1) Winter 1994-95 5.6 6.8 9.1 8.4 2.8 78.4 9.7 11.1 5.8 6.6 2.6 7.5 6.9 0.4 Difference 0.5 1.8 4.2 4.0 1.8 2.2 4.9 5.6 0.1 0.5 2.9 2.5 1.8

Source

df

Mutant 11 Season 2 Mutant season 22 Error 42 s2 Mutant s2 Season s2 Mutant season

a ** = significant at the 0.01 level.

Using anther culture to generate fertile, doubled-haploid interspecific progeny

M. P. Jones, S. Mande, A. Daleba, and H. Sehi, West Africa Rice Development Association (WARDA), 01 BP 2551, Bouak, Cte d'Ivoire, West Africa

Anther culture enables the rapid genetic fixation of progeny from crosses. In the context of wide crosses, anther culture can also provide fertile doubled haploid progeny with fixed introgressions that might otherwise

disappear in the course of an extended series of backcrosses. We are using anther culture and embryo rescue techniques to accelerate and improve the efficiency of the wide O. sativa/ O. glaberrima hybridization program at WARDA. The program was expanded to classify representative japonica, indica, and O. glaberrima genotypes and their F1 and F2 progeny for suitability for anther culture. We carried out anther cultures for 10 parental lines and the progeny of 55 japonica/japonica, japonica/indica, and O. sativa/O. glaberrima crosses.

Main stem boots were collected when flag leaves were exserted 4-8 cm. Panicles were disinfected and cold pretreated at 8 C for 10-20 d depending on varietal type. Anthers were incubated in the dark at 251 C in modified N6 medium. Calli obtained were regenerated in Murashige and Skoog (MS) medium with a 16-h photoperiod at 25 C. Plantlet regeneration was induced with the base MS salts and 3 mg thiamine-HCl L-1, 1 mg IAA L -1, 4 mg GA3 L-1, and 2% (w/v) sucrose. Response to callus induction and green plantlet regeneration depended

Vol.

22.

No

on the genetic origin of the crosses, with japonica/O. glaberrima hybrids responding best, followed by pure japonica varieties and intra-japonica hybrids, japonica/indica hybrids, and pure indica and intra-indica hybrids. The least responsive were pure O. glaberrima and intra- O. glaberrima hybrids (see figure). Callus production generally peaked 4-5 wk after induction. Calli produced during this period had the greatest probability of yielding green plantlets. We found a 5:4:1 ratio for haploids, doubled haploids, and polyploids among the regenerated plants regardless of the crosses' genetic origins (see table). The spontaneous doubled haploid lines frequently displayed partial fertility, particularly when derived from O. sativa/O. glaberrima crosses and, to a lesser extent, from japonica/indica crosses. High sterility levels might have been due to aneuploidy of the regenerated plants or fixation of sterility genes.

a. Percentage of calli induced and green and albino plantlets produced. J = japonica, I = indica, G = O. glaberrima. b. Percentage of calli produced over time.

More than 500 progenies of spontaneous doubled haploid fertile lines of O. sativa/O. glaberrima and 600 japonica/indica lines have been evaluated in the field to study their fertility, genetic stability, and agronomic traits. Using the seed from these trials, we are conducting specific screening trials for resistance to the major biotic and abiotic stresses in the region.

Ploidy levels of plants regenerated.

Background of pedigrees Japonica Japonica and indica Japonica and O. glaberrima Mean

Haploid lines (%) 53 50 51

Dihaploid lines (%) 40 45 39

Polyploid lines (%) 7 5 10

51

41

7.3

Response of rice anther culture to short-day treatment

Lianqun Yang, Deying Gong, and Shoujiang Yuan, Shandong Rice Research Institute (SRRI), Shandong 272177, China

Japonica rice is widely cultivated in northern China, where the rice-grow ing season is only 5 mo (May to September) because of the cold climate. It takes 8-10 yr to select a pure line from a cross using conventional rice breeding and at least 4 yr using anther culture. To find ways to further reduce the time needed for rice anther culture, we conducted a study at SRRI (35.4 N latitude). Five parental varieties were sown on 18 Feb 1994 and transplanted on 21 Apr. When the seedlings had 7-8 leaves, they were exposed to 9 h of sunshine and 15 h of darkness for 25 d. Before the treatment, 150 ppm multi-effects triazole (MET) was applied to seedlings to

control excessive growth. Crosses were made on 4 Jun, and seeds from five F1 plants were harvested on 28 Jun. They were sown in pots outside on 2 Jul. The panicles of the five F1 plants were large enough for sampling on 20 Sep, at which time the micro-spores were in the early- to mid-uninucleate stage. The same crosses were made in 1993 as checks. They were sown in 1994 on 1 May, and their panicles were sampled on 15 Aug. Anther calli were induced on N6 medium with 2 mg 2,4-D L-1, 60 g sucrose L -1, 8 g agar L -1, and pH 5.85. The plant regeneration medium was Murashige and Skoog (MS) with 2 mg 6 BAL-1, 0.5 mg NAAL-1, 30 g sucrose L-1, 8 g agar L -1, and pH 5.85. Japonicas in northern China generally head from 15 to 25 Aug and have medium photoperiod sensitivity and medium temperature response. By combining early seeding and short-day treatment in the greenhouse, the five

parents headed from 1 to 5 Junnearly 80 d earlier than the same varieties grown in the ricefield under normal cultivation practices (Table 1). When the panicles of the five F1 crosses were ready for sampling on 20 Sep, the average day temperature was 19C, 6C below that of 15 Aug (the regular sampling date). The anther callus induction and green plantlet regeneration of the five F1 crosses and their controls were then compared to check the effect of low temperature on anther culture ability. No large differences were found among the five F1
Table 1. Heading date of five parents. Shandong, China. 1994.

Variety Akenohoshi 89-7013 Xiangqueno Kinuhikari Yakomagokari

Short-day treatment 4 Jun 3 Jun 1 Jun 31 May 1 Jun

Regular culture 23 Aug 25 Aug 21 Aug 20 Aug 16 Aug

IRRN 1997

crosses and their controls in inducing and differentiating calli (Table 2). Using early sowing and short-day treatments, the parent plants headed 80 d sooner than those grown under conventional cultivation. It only took 10 mo for the entire anther culture process, from cultivating the parents to growing enough anther lines for transplanting. The improved procedure took 1 yr less than the anther culture technique routinely used in rice breeding.
Comments. If you have comments or suggestions about the IRRN, please write to the editor.

Table 2. Comparison of anther culture ability of five F 1 crosses. Shandong, China. 1993-94.

Anthers Cross plated (no.) 391 385 407 416 395 1994 400 400 400 400 400 2000
b

Calli produced (%) a 36.57 30.58 38.56 56.49 42.87 41.01

Calli plated (no.)

Green plantlets regenerated (%) b 8.33 51.79 28.67 36.21 34.05 31.81 11.60 50.40 26.40 38.40 32.80 31.92

89-7013/Kinuhikari 89-7013/Yamauta 89-7013/Xiangqueno Akenohoshi/Xiangqueno Akenohoshi/Yakomagokari Total 89-7013/Kinuhikari 89-7013/Yamauta 89-7013/Xiangqueno Akenohoshi/Xiangqueno Akenohoshi/Yakomagokari Total
a

Treatment 1 c 240 256 247 243 255 1241

Treatment 2 d 34.50 250 33.50 250 40.25 250 52.50 250 46.25 250 41.40 1250 100.
c d

Anthers producing calli Total anthers plated

100.

Calli producing green plantlets Total calli plated

Crosses made in June 1994. Check; crosses made in 1993.

Expression of an engineered cysteine proteinase inhibitor (Oryzacystatin-186) in transgenic rice plants

P. Vain and B. Worland, John Innes Centre (JIC) Colney Lane, Norwich NR4 7UH, UK; M. C. Clarke and M. Beavis, Centre for Plant Biochemistry and Biotechnology (CPBB), University of Leeds (UL), Leeds LS2 9JT, UK; A. Kohli, M. Leech, J. Snape, and P. Christou, JIC; and H. Atkinson, CPBB, UL

Meloidogyne spp. (root knot nematodes) damage upland rice in both Asia and Africa. Their wide host range ensures other crops in rotation with rice are at risk in both upland and intermittently flooded, lowland rice. Hirschmanniella spp. (rice root knot nematodes) also damage lowland rice but only when cropping is intensive. Pratylenchus spp. (root lesion nematodes) occur in virtually all Asian and many African upland rice crops and can cause up to 50% yield loss. Our goal is to provide a basis for concomitant control of all nematode pests of rice. Efficient tissue culture, regeneration, DNA delivery, and selection methodologies have been established for elite African (ITA212, IDSA6, LAC23,

WAB56-104) and Asian (IR64) genotypes. The plasmids pWRG4517(D 86 (CaMV 35S-promoter~oc-I D 86~NOSpolyA) + CaMV35S~ promoter- aphIV ~ S-poly A) and pWRG2920 (ubi-5' region ~ uida ~ NOS-polyA) were constructed and used for stable transformation of rice. The oc-ID 86 gene was designed by protein engineering to improve the efficacy of native rice oc-I against nematodes (Urwin et al 1995). Oc-I D 86 differs from Oc-I by only one amino acid (Asp86 deleted, Urwin et al 1995). Particle gun bombardment, selection, and regeneration of transformed plants were carried out as previously described (Christou et al 1991). Two days after particle bombardment, the explants were transferred to a medium containing hygromycin to minimize the establishment of chimeric calli (a mix of transformed and nontransformed tissues, Christou and Ford 1995). Twelve days after bombardment, the immature embryos exhibited necrotic areas and large proliferations of embryogenic calli from the scutellum. Embryogenic tissue was repeatedly selected at each subculture. After 3-6 wk, we observed clear differential growth between transformed (hyg+) and nontransformed (hyg-) calli during

proliferation, regeneration, and germination. Transformation of the hyg+ clones was further confirmed by GUS, PCR, Southern blot, and Western blot analyses of regenerated plants. Transformation frequencies were 1-6% from the elite African and Asian genotypes tested (av: ITA212, 2.6%; LAC23, 1.8%; WAB56-104, 1.1%; IDSA6, 1.8%; and IR64, 1.1%). More than 80 independent transformed lines have been obtained to date. When both pWRG4517D 86 and pWRG2920 were used for transformation, hyg+ clones and regenerated plants became intensely blue after GUS histochemical staining. In some cases, transformed GUS-negative plants were regenerated from hyg+ clones. When only pWRG4517D 86 was used, the regenerated plants were first screened by PCR using two primers (AAGAAGACGTTCCAACCACG and GATCTCCAATCTGCGGGATC), amplifying a 1.4-kb fragment containing the aphIV gene. Southern blot analysis confirmed the integration of the pWRG4517D 86 plasmid in the rice genomic DNA (Fig. 1). Levels of oryzacystatin (Oc-ID 86) in plant roots were detected in four of nine transformed rice lines by Western blot analysis (Fig. 2).
Vol.22 No.1 9

Cited references
Christou P, Ford TL, Kofron M. 1991. Production of transgenic rice (Oryza sativa L.) plants from agronomically important indica and japonica varieties via electric discharge particle acceleration of exogenous DNA into immature zygotic embryos. Bio/ Technology 9:957-962.

Christou P, Ford TL. 1995. The impact of selection parameters on the phenotype and genotype of transgenic rice callus and pants. Trans. Res. 4:44-51. Urwin PE, Atkinson HJ, Waller DA, McPherson MJ. 1995. Engineered oryzacystatin-I expressed in transgenic hairy roots confers resistance to Globodea pallida. Plant J. 8:121-131.

An improved biolistic method for transformation and production of fertile transgenic Pusa Basmati rice plants
R. K. Jain, Genetics Department, Haryana Agricultural University (HAU), Hisar 125004, India; S. Jain, B. Y. Wang, and R. Wu, Biochemistry Section, Molecular and Cell Biology, Biotechnology Building, Cornell University, Ithaca NY 14853-2703, USA

1. Southern analysis of one ITA212 hygromycin-resistant callus line (C) and regenerated plant (P). DNA was either undigested (UD) or digested by Not 1, Sac 1, or Kpn 1. The Not 1 sites are flanking the 2-kb CaMV35S-promoter~aphIV~NOS-poly A sequence in pWRG4517D 86 plasmid and only one Sac 1 site is present in this plasmid. Nontransformed (NT) and previously obtained transformed (T) rice plants (Christou et al 1991) were used as negative and positive controls, respectively. The membrane was hybridized with the aphlV gene.

2. Western blot of an SDS PAGE gel probed with a polyclonal antibody raised to oryzacystatin Oc-l (Urwin et al 1995). Lanes 1-3 were loaded with 2 g total protein from ITA212-transformed rice plants regenerated from three independent lines. Lanes 4-8 were loaded with 2 g total protein from an ITA212-untransformed rice plant to which 0, 0.25, 0.5, 0.75, or 1% Oc-l protein was added.

An improved biolistic procedure for the transformation of embryogenic suspension cells of Pusa Basmati 1 (aromatic, fine grained, semidwarf plant type, Group 1 indica), has been developed. The procedure involves using of suitable reporter or other useful genes, osmotic preconditioning of cells on a medium supplemented with 0.25 M mannitol prior to bombardment, gold particles for DNA delivery, and plant regeneration medium with high agarose concentration (1.0%). This procedure allowed us to produce more than 600 transient transformants and 12-50 fertile transgenic plants per bombarded filter carrying 0.5 mL settled cell volume (scv) of rice cells. We have been working on biolistic transformation of Basmati rice (Pusa Basmati 1) using gene constructs that can potentially improve the defense against insect attack and water stress tolerance. Pusa Basmati 1 is an improved, fine grain, aromatic, and semidwarf indica rice variety that essentially belongs to Varietal Group 1 (Jain et al 1995). Plasmids carrying b -

glucuronidase (GUS) gene or other agronomically important genes driven by the actin 1 gene promoter ( Act1), were constructed for rice transformation. Act1F-GUS plasmid (McElroy et al 1990) was used for transient gene expression studies; other plasmids carrying potato protease inhibitor 2 (PIN2) gene, or a late embryogenesisabundant protein (Lea3) gene from barley (see figure), were used for optimization of the biolistic process and production of useful transgenic plants. The structure of these plasmids is shown in the figure. In the beginning, we used the biolistic japonica rice transformation procedure developed in this laboratory (Cao et al 1992) for DNA delivery into Pusa Basmati 1 cells. The procedure involved bombardment of M10 tungsten particles coated with plasmid DNA using the DuPont Helium PDS-1000 device keeping the bombardment pressure of 1500 psi, gap distance (distance between the rupture disc and the launch point of macroprojectile) of 1.0 cm, and target cell distance (distance between the target cells and the launch point of the microprojectiles) of 12 cm. Each filter, carrying 0.5 mL scv of suspension cells, was bombarded twice with tungsten particles coated with 2.5 g plasmid DNA. This procedure produced per filter paper an average of 24 blue spots (transient transformants, see table, line 1) or 4.06.8 ammonium glufosinate-resistant (AG R) calli (see table, lines 6 and 12). For practical plant genetic transformation work, at least several hundred potentially generated cells

10

IRRN 1997

Structure of pAct1F-GUS, pBY520 and pTWa plasmids. Act1 5', rice actin 1 gene promoter; Act1 In, intron from Act1 5' promoter region; bar, phosphinothricin acetyl transferase gene; HVA1, late embryogenesis abundant (LEA3) protein gene; Nos3', nopaline synthase gene 3' region; Pin2, potato protease inhibitor 2 gene; Pin2 3', Pin2 3' region; Pin2 5', Pin2 promoter region; 35S 5', cauliflower mosaic virus 35S promoter. Only important restriction endonuclease sites are indicated.

Effect of metallic particles, osmotic preconditioning of cells, and target cell distance on GUS expression and stable transformation in Pusa Basmati 1 cells. a

Plasmid

Osmotic preconditioning b No No Yes Yes No No Yes Yes No No No No Yes Yes No No No No Yes Yes

Particle type

Target cell distance (cm) 12, 9, 12, 9, 12, 9, 12, 9, 12, 9, 9, 9, 9, 9, 12 9 12 9 12 9 12 9 12 9 12 12 12 12

Number of blue spots c or resistant callid per filter 24.4 37.3 207.8 215.0 132.5 158.7 510.3 684.8 6.8 14.4 13.3 27.0 43.3 102.7 4.0 10.8 10.7 24.5 24.1 52.7 5.2 c 8.4 c 27.8 c 41.5 c 25.8 c 16.5 c 142.1 c 91.1 c 1.4d 4.4d 1.9d 4.6d 0.6d 25.0d 4.4 d 1.2 d 1.6 d 3.4 d 4.7 d 8.5 d

pAct1F-Gus

Tungsten Tungsten Tungsten Tungsten Gold Gold Gold Gold Tungsten Tungsten Tungsten Gold Tungsten Gold Tungsten Tungsten Tungsten Gold Tungsten Gold

pTWa

pLEA3

12, 12 9, 9 9, 12 9, 12 9, 12 9, 12

aData represent the mean standard error of three independent experiments. For each experiment, three filters were used and the values averaged. b Filters carrying target cells were kept on the MS2.5 medium supplemented with 0.25 M mannitol for 24 h before bombardment. c Number of blue spots per filter. d Number of ammonium glufosinate-resistant calli per filter.

should receive and transiently express the introduced DNA per bombardment (Yang and Christou 1994). To improve transformation frequencies, we investigated the effects of osmotic preconditioning of cells, metallic nature of microcarriers, and target cell distance on transformation of Pusa Basmati 1 cells (data presented in the table). The optimal conditions of each parameter were then combined to form a new improved protocol. The new improved protocol. Cell suspension cultures of Pusa Basmati 1 were established using the mature seed scutella-derived calli, as described by Jain et al (1995). Plasmid DNAs were adsorbed to the gold or tungsten particles (mean diameter of 1 m, SylvaniaGTE Products Corp., Towanda, PA) by CaCl2 and spermidine precipitation and bombarded to the target cells (Cao et al 1992). To prepare filters with competent cells, 0.5 mL scv of sieved (1000 m nylon mesh) suspension cells obtained after 4 d of subculture were spread on each of the 5.5-cm-diameter Whatman #1 filter disc. For osmotic preconditioning, the filter discs were placed on the surface of 25 mL of 0.5% (w/v) agarose-solidified Murashige and Skoog (1962) medium with 2.5 mg 2,4-D L -1 (MS2.5) and 0.25 M mannitol in 9-cm petri dishes. These dishes were used for the biolistic experiment the following day. Each filter disc was bombarded twice with the metallic particles coated with 2.5 g of plasmid DNA, sequentially keeping the target cell distances of 9 and 12 cm in the first and second bombardment. Microscopic observations showed that the bombardments made using the two different target cell distances allowed a more even distribution of coated-particles to cells, with minimum damage to the cells. After 24 h, filters were transferred to the new dishes containing mannitol-free MS2.5 medium. Histochemical GUS expression assay was carried out 2 d after bombardment (Battraw and Hall 1990). Blue loci, indicative of transient GUS expression, were counted 2 d after adding the X-Gluc substrate solution.
Vol. 22. No. 1 11

Three days after the bombardment, filter discs with overlaying cells were transferred to the petri dishes containing 25 ml MS2.5 medium supplemented with 8 mg L-1 ammonium glufosinate for selecting transformed calli. These filters were transferred every 10 d onto the fresh selection medium. Petri dishes were always sealed with the parafilm and incubated at 26 C in the dark. After 30-40 d, the new calli showing active cell proliferation on the selection medium and growing up about 1.0 cm diameters (referred to as resistant calli), were transferred to the MS-based regeneration medium containing maltose (30 g L-1), kinetin (2.0 mg L-1), a -naphthalene acetic acid [NAA] (0.5 mg L -1), ammonium glufosinate (5.0 mg L-1); the medium was solidified with 1.0% agarose. For plant regeneration, cultures were incubated at 26 C in the dark for 2 wk and then transferred to light (55 mol m-2 s-1, daylight fluorescent tubes, 16-h photoperiod). After 4 wk, these calli were transferred onto regeneration medium with lower concentrations of agarose (0.5%, w / v) and ammonium glufosinate (3 mg L-1). Regenerated shoots were transferred to the rooting medium (0.25%, w/v phytagel-

solidified MS medium with 1.5 mg NAA L -1) and 4 wk later, they were transferred to pots in the greenhouse. This new procedure produced an average of 510-684 transient transformants and 24-103 transformed AGR calli per filter; the number of transformants was higher using gold particles and with pTWa plasmid (see table). Water stress created by using higher agarose concentration (1.0% instead of 0.5%) promoted somatic embryogenesis, and shoot regeneration frequencies of 69-91% were obtained from AGR calli. A total of 235 plants obtained after transformation with the two plasmids were analyzed by PCR using a suitable set of primers derived from the promoter, bar or PIN2 gene sequences. As an internal control, a set of primers that amplify a 304-bp product during PCR from Act1 promoter region was used. Of these, 188 plants showed the amplification of expected size of DNA fragment, indicating that 80% of selected AGR plants are true transgenic plants. All the transformed plants transferred to the greenhouse were fertile and showed good seed set within 110-145 d after planting. Further molecular and progeny analyses of these transgenic plants are in progress.

Battraw MJ, Hall TC. 1990. Histochemical analysis of CaMV 35S promoter-glucuronidase gene expression in transgenic rice plants. Plant Mol. Biol. 15:526-538. Cao J, Duan X, McElroy D, Wu R. 1992. Regeneration of herbicide-resistant transgenic rice plants following transmicroprojectile-mediated formation of suspension culture cells. Plant Cell Res. 11:586-591. Jain RK, Khehra GS, Lee S-H, Blackhall NW, Merchand R, Davey MR, Power JB, Cocking EC, Gosal SS. 1995. An improved procedure for plant regeneration from indica and japonica rice protoplasts. Plant Cell Rep. 14:515-519. McElroy D, Zhang W, Cao J, Wu R. 1990. Isolation of an efficient actin promoter for use in rice transformation. Plant Cell 2:163-171. Murasighe T, Skoog F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiol. Plant. 15:473-497. Yang N-S, Christou P. 1994. Particle bombardment technology for gene transfer. New York, Oxford University Press.

Cited references

Genetic analysis of a d1 chimeric rice plant

M. Ashikari, O. Ideta, A. Yoshimura, and N. Iwata, Plant Breeding Laboratory Faculty of Agriculture, Kyushu University, Hakozaki 6-1-10, Higashi-ku, Fukuoka 812-81, Japan

A chimeric rice plant has been maintained through seed for more than 50 yr in our laboratory. It has both normal and dwarf tillers. The dwarf tillers have wide dark green leaves, short panicles, and small round grains. Differences in chimeric manifestations were observed in chimeric plants (see figure). Both the dwarf-type progenies and dwarf tillers of the chimeric plant

are short, with wide dark green leaves, short panicles, and small round grains. These characters are almost the same as the Daikoku dwarf carrying the d1 gene. We therefore call the chimeric plants d1 chimeric type. Progenies of both the dwarf and normal tillers of the chimeric plant were dwarf, chimeric, or normal (see figure). To test the inheritance of the chimeric expression, dwarf and normal plants obtained from chimeric plants were crossed. The 26 F1 hybrids produced by crossing a dwarf plant with a normal plant were all normal. Furthermore, 18 and 34 F1 hybrids from crosses of dwarf and normal tillers of chimeric plants with IR24 and Taichung 65 were all normal. Segregation in F2 from the

reciprocal crosses of the dwarf plants with IR24, Taichung 65, and the normal plant segregated into 3 normal:1 dwarf (see table). The results indicate a single recessive gene controls chimeric expression of the d1 chimeric plant. To test for allelism between dl and the d1 chimeric plants, dwarf plants obtained from the chimeric plant were crossed to a genetic marker carrying the d1 gene. Seventeen F1 hybrids from the reciprocal crosses of d1/dwarf plant were all d1-like plants, indicating that chimeric locus is allelic to the dl locus. High frequency mutation at the d1 locus could cause the occurrence of a dl chimeric plant. When the three types of progenies from the plant were selfed, the average frequency of chimeric type

12 IRRN 1997

Differences in chimeric manifestations: a) almost all tillers are dwarf type with few normal tillers, b) half the tillers are dwarf, the other half normal, and c) almost all the tillers are normal with few dwarf tillers.
F2 segregation for plant types in the crosses involving d1 chimeric plant.a

Phenotypes of the progenies derived from chimeric plants: d) dwarf type, e) chimeric type having both dwarf and normal tillers, and f) normal type.

Segregation in F 2 Cross Normal 47 224 260 318 383 259 Dwarf 15 71 88 94 134 102 Chimera 0 3 0 5 3 0 Total 62 298 348 417 520 361

c2 3:1 (N:D) b 0.022 0.137 0.015 1.049 0.233 2.040 P

Dwarf plant/lR24 IR24/dwarf plant Dwarf plant/TC65 TC65/dwarf plant Dwarf plant/normal plant Normal plant/dwarf plant

0.70-0.90 0.70-0.90 0.90-0.99 0.30-0.50 0.50-0.70 0.10-0.30

aDwarf plant = dwarf plant progenies derived from d1 chimeric rice. Normal plant = normal plant progenies derived from d1 chimeric rice. b N = normal, D = dwarf.

from dwarf, chimeric, and normal types were 26.7, 44.4, and 9.8%, respectively. The frequency of mutation in dwarf to normal is higher than in normal to dwarf. Based on the results obtained from the inheritance and allelic tests, dwarf plant progenies of chimeric plants have

the d-1 alleles while the normal progenies have the wild D-1 alleles. We hypothesize that originally the chimeric plant had the D-1 allele, but then the gene mutated to the recessive d-1 allele. As a result, a chimeric plant has both dwarf (d-1 allele) and normal tillers ( D-1 allele). If the progenies of a chi-

meric plant inherited the d-1 allele, they become dwarf plants. If progenies inherited the D-1 allele, they become normal plants. Throughout the growth stages of the rice plant, the d-1 allele may revert to D-1 allele or vice versa causing a dwarf plant to produce normal tillers and a normal plant to have dwarf tillers. We also hypothesize that the d1 chimeric rice plant, a very unstable mutant, occurs due to the on and off expression of the D-1 allele, which appears to be controlled by factors such as a mutable gene, transposons, and methylation. A fine restriction fragment length polymorphism linkage map for d1 is being developed, and we are cloning the d1 gene to understand the mechanism of high mutation in the d1 chimeric rice plant.

Increasing yield potential of irrigated rice through recurrent selection

P.H.N. Rangel, F.J.P. Zimmermann, and P.C.F. Neves, Empresa Brasileira de Pesquisa Agropecuaria/Centro Nacional de Pesquisa Arroz e Feijo (EMBRAPA/CNPAF), Caixa Postal 179, Goinia, Gois, Brazil 74001-970

Recurrent selection allows scientists to increase the frequency of favorable genotypes from a large population

through selecting and intermating generations quickly. To assess the possibility of using this breeding method in irrigated rice in Brazil, we divided 162 S0:2 families (S 0 plant-derived families self-pollinated twice) into two groups (early-maturing and medium-maturing) and evaluated them in two environments. Each group was divided into two subgroups and then tested in the field with controls BR-IRGA 409 (early-maturing) and CICA 8 (medium-maturing).

The experiment was laid out in two triple-square lattices (10 10 m and 8 8 m). The S0:2 families came from the early-maturing population CNA-IRAT 4PR/1/1 and medium-maturing population CNA-IRAT 4ME/1/1. Individual and joint analyses of variance were done; we also estimated variance components within each maturation cycle group. In all cases, the F test revealed highly significant differences (P<0.01) among the mean grain yields for the early- and
Vol. 22. No.1 13

Average of original (Mo) and selected (Ms) populations with respective mean standard errors, and estimates of gains through direct selection on grain yield and indirect selection over the other characteristics, based on the classic selection index with a 30% selection intensity.

Characteristic

Early-maturinga Mo Ms

Medium-maturing a Mo Ms

Direct selection on grain yield (%) EarlyMediummaturing maturing 9.8 0.1 11.9 0.1

Selection based on classical index (%) EarlyMediummaturing maturing 9.6 0.1 10.1 1.3

Grain yield (t ha -1 ) Days to 50% flowering Panicle blast b score Brown spot b score

4.6 (.03) 98 (.11) 4.5 (.03) 3.8 (.03)

5.5 (.03) 98 (.11) 4.6 (.03) 3.9 (.03)

4.5 (.03) 104 (.12) 3.8 (.02) 4.1 (.03)

5.3 (.03) 107 (.12) 3.7 (.02) 3.7 (.03)

1.2

2.6

0.1

1.1

0.2

1.7

0.4

4.6

a Mean standard error based on 983 and 981 observations for the early- and medium-maturing families, respectively. b Scored using the scales in the Standard evaluation system for rice.

medium-maturing families. The average grain yield was 4.6 t ha -1 for the early-maturing families and 4.5 t ha -1 for the medium-maturing families (see table). However, in the early group, six families yielded more than 6 t ha -1 , and in the medium group, two families yielded more than 7 t ha -1 . The average

of the selected families equaled 5.5 t ha -1 in the early-maturing group and 5.3 t ha -1 in the medium-maturing group. These findings show the possibility of positively altering the population means through recurrent selection (see table). If families are selected (at an intensity of 30%) based only on grain yield

(direct selection), the genetic gains estimated for the cycle would be 9.8% for the early-maturing families and 11.9% for the medium-maturing families (see table). However, the indirect gain for panicle blast would be 1.2 and 2.6%, respectively for each group, demonstrating that selecting directly for grain yield increases the populations' susceptibility to the disease. The classical index of Smith (1936) and Hazel (1943), which simultaneously considers all characters, can be used to select the best families to be recombined for initiating a new cycle of recurrent selection. Using this index, genetic gains for grain yield are similar to those obtained with direct selection with the additional advantage of simultaneously increasing (over the original populations) resistance to panicle blast and brown spot in all improved populations (see table). In the populations studied, one recurrent selection cycle was efficient for increasing grain yield.

Transgenic rice plants expressing rice yellow mottle virus coat protein gene
N. Kouassi, Plant Biology Division, International Laboratory for Tropical Agricultural Biotechnology The Scripps Research Institute (ILTAB/ TSRI), MTC7, 10666 North Torrey Pine Road, La Jolla, CA 92037, USA; C. Brugidou, ILTABInstitut franais de recherche scientifique pour le dveloppement en coopration (ORSTOM); L. Chen, M. Ngon A. Yassi, and R. N. Beachy, ILTAB/TSRI; C. M. Fauquet, ILTAB-ORSTOM

Coat protein-mediated resistance (CPMR) is used to induce resistance caused by the expression of a virus coat protein (CP) gene in transgenic plants (Powell et al 1986). We investigated the use of this strategy to produce transgenic plants resistant to rice yellow mottle virus (RYMV), an important viral disease in Africa.

This virus was first reported in Kenya (Bakker 1970). The genome organization was achieved at ILTAB (Ngon A Yassi et al 1994). Transformation experiments using the biolistic method, which involves microprojectile bombardment of embryogenic calli or immature embryos (Christou et al 1991; Li et al 1993; Sivamani et al 1996; Zhang et al 1996), followed. Four different constructs containing the cDNA sequence of the RYMV coat protein gene were inserted into the ubiquitin promoter-Nos terminator cassette and engineered in different ways to produce the CP (CP+), a truncated coat protein (CP D NTS), a RNA sense (mRNA), and a RNA antisense (CP-). The biolistic method was used to introduce these into embryogenic calli of japonica rice variety Taipei 309. We obtained a relatively high transformation efficiency (see table).

Cotransformation with various ratios (gene of selection/gene of interest) was achieved using the different CP chimeric plasmids and the plasmid pMON410 carrying the hygromycin resistance gene. After several rounds of selection on hygromycin, the plantlets were transferred to soil. Molecular analysis was performed using DNA extracted from leaves of putative R0 transgenic plants following the method. We amplified by polymerase chain reaction (PCR) the hph gene and the entire cassette containing the gene of interest (ubiquitin-CPNos=3000 bp). Southern blot experiments using a CP probe 32 P radiolabeled were also performed and confirmed the integration of the gene in R0 plants (see figure). We tested 117 transgenic lines and all were found to carry the hph gene, while an average percentage (62%) of independent R0 lines were positive by both

14

IRRN 1997

PCR and Southern experiments for the ubiquitin-CP-Nos cassette (see table). Northern blot experiments showed different levels of accumulation of CP mRNA in the transgenic lines. Western blot analyses using the RYMV polyclonal antibodies revealed a weak accumulation of the coat protein in some appropriate transgenic plants (from CP+ lines). The level of accumulation of the coat protein and mRNA differed among lines.
Preliminary experiments were done

to screen the transgenic plants against the virus. Based on observation of RYMV symptoms, different levels of tolerance for the RYMV virions have been recorded. Further experiments are in progress to quantify the virus replication in transgenic plants by enzyme-linked immunosorbent assay and to select the most resistant lines. In the meantime, variety BG 90-2 (widely used in West Africa and highly susceptible to RYMV) is being transformed for the same purpose with appropriate constructs. Other indica rice varieties cultivated in West Africa, such as Bouak 189 and Jaya, will be transformed with RYMV coat protein.

Amplification and detection of the DNA fragment comprising the ubiquitin promoter, the RYMV-CP gene, and the Nos terminator cassette (Ubi-CP-Nos = 3 kbp). a) Agarose gel electrophoresis of DNA fragment amplified by PCR using total DNA extracted from hygromycinresistant T309 rice leaves. Line 1 = 1 kb ladder used as MW marker; lines 2, 3, 4, 5, 6, 7, 8, 9, 11 = transgenic CP positive plants; line 10 = transgenic CP negative plant; line 12 = nontransgenic TP309 plant. b) Southern analyses of transformants (R 0 generation): detection by radioactive method using a CP fragment as a probe. Total DNA from transgenic plants (T309) were Hin dlll/Afl lldigested, which releases the 3 kb Ubi-CP-Nos cassette. T = transgenic plants; NT = nontransgenic plant; P = CP chimeric plasmid used for the transformation.

Cited references

Bakker W. 1970. Rice yellow mottle virus, a mechanically transmitted virus disease of rice in Kenya. Neth. J. Plant Pathol. 76:53-63. Christou P, Ford TL, Kofron M. 1991. Production of transgenic rice (Oryza sativa L.) plants from agronomically important indica and japonica varieties via electric discharge particle acceleration of exogenous DNA into immature zygotic embryos. Bio/Technology 9:957-962. Ngon AYassi M, Ritzenthaler C, Brugidou C, Fauquet CM, Beachy RN. 1994. Nucleotide sequence and genome characterization of rice yellow mottle virus RNA. J. Gen. Virol. 75:249-257. Li L, Qu R, Kochko de A, Fauquet C, Beachy RN. 1993. An improved rice transformation system using the biolistic method. Plant Cell Rep. 12:250-255. Powell PA, Nelson RS, De B, Hoffman N, Rodgers SG, Fraley RT, Beachy RN. 1986. Delay of the disease development in transgenic plants that express the tobacco mosaic virus coat protein gene. Science 232:738-743.

Efficiency of biolistic transformation and integration of the RYMV-CP gene in regenerated plants (T309).

Plasmid construct CP (+) CP NTS D

Rice tissue (no. of explants) Embryogenic calli (480) Embryogenic calli (100) Cell suspension (360) Cell suspension (360)

hph-resistant calli (no.)

hph-resistant plants (no.)

CP transgenic lines (PCR, Southern) (%)

78 28 118 111

49 19 70

64 64 59 62

CP (-) rnRNA CP (+) mRNA

79

Sivamani E, Shen P, Opalka N, Beachy RN, Fauquet CM. 1996. Selection of large quantities of embryogenic subcultured calli from Indica rice seeds for production of fertile transgenic plants using the biolistic method. Plant Cell Rep. 15:322327.

Zhang S, Chen L, Qu R, Marmey P, Beachy RN, and Fauquet CM. 1996. Regeneration of fertile transgenic Indica (Group 1) rice plants following micropojectiletransformation of embryogenic suspension culture cells. Plant Cell Rep. 15:465-469.
Vol. 22, No. 1 15

Plant regeneration toward transformation of several javanica rice cultivars

I.H. Slamet, A. Estiati, W. Rahayu, s. Hutajulu, S. Nugroho, Research and Development Centre for Biotechnology-Indonesian Institute of Science, Jl. Raya Bogor Km 46, Cibinong 16911, Indonesia

The successful culturing of rice in vitro strongly depends on the genotype and culture used (Cooley et al 1995). Most successful transformation work has involved japonica rice cultivars; only a few indica cultivars have so far been transformed. The competency of plant material to regenerate in vitro is a prerequisite for transgenic research. Here we describe plant regeneration studies from a range of indica and javanica rice cultivars. We cultured scutella of mature rice seeds on three media compositions (Li et al 1993, Poonsapaya et al 1989, Abdullah et al 1986) for embryogenic callus induction. Regenerated calli were identified and proliferated for transformation experiments using Agrobacterium strain EHA 101 pIG121 Hm (Hiei et al 1994) or DNA coated microprojectile bombardment techniques. Based on the results of preliminary embryogenesis studies of 45 rice cultivars on Linsmaier and Skoog medium (unpublished results), five indica cultivars (Cisadane, Dodokan, Gajah Mungkur, Maninjau, and Kelimutu) and 10 javanica cultivars (Gebang, Rajalele, Rajalele KA, Asemandi, Aselapan, Kencana Bali, Caping Gajah, Jalawara, Pandan Wangi, and Raja Imut) were selected. Calli were induced from 50 mature seeds in three different solid media designated as IK1 (Li et al 1993), IK2 (Poonsapaya et al 1989), and IK3 (Abdullah et al 1986) and incubated at 28 C in the dark for 30 d. At 4 wk, each callus of 3-5 mm diameter was numbered and divided into two parts. The first part was subcultured onto the original induction callus medium while
16 IRRN 1997

the second part was transferred onto regeneration described by Poonsapaya et al (Linsmaier and Skoog basal medium, 0.5 mg IAA L-1, 0.3 mg BAP L-1, 3% sucrose) followed by subculturing plantlets onto Murashige and Skoog basal medium. Regenerated calli were identified and corresponding calli were proliferated on the induction media for transformation experiments. Regenerable calli from rice cultivars Cisadane, Maninjau, Gajah Mungkur, and Gebang were bombarded with tungsten particles (0.5-1 mm diam) coated with plasmid pAHC 25 (Christensen et al 1992) using a particle inflow-helium gun (He pressure of 800 kPa, vacuum pressure -50 kPa) and selected on medium containing bialaphos. Regenerable calli from Caping Gajah, Rojolele KA, Jalawara, and Koshihikari (control) were infected with Agrobacterium strain EHA101 harboring pIG121Hm according to the method described by Hiei et al (1994).

Plasmid pAHC25 was provided by the United States Department of Agriculture and the Agrobacterium strain pIG121Hm was obtained from Nagoya University, Japan. Bombarded and infected calli were histochemically assayed for their GUS activity with assay buffer described by Rueb et a1 (1989). The most regenerated plants from the 15 cultivars tested were obtained from indica cultivar Maninjau and javanica cultivars Rojolele KA, Caping Gajah, and Jencana Bali (Table 1). In general, the Linsmaier and Skoog basal medium with the addition of 6% coconut water and 2,5 mg 2,4-D L-1 (Poonsapaya et al 1989) combined with the plant regeneration medium gave more regenerated plants. However, different cultivars responded differently when they were cultured following the same plant regeneration medium scheme. The experiment was repeated for the three most responsive varieties

Table 1. Number of regenerated plants a of several indica and javanica cultivars on three callus induction media.a

Callus induction medium Cultivar Kelimutu Gajah Mungkur Cisadane Dodokan Maninjau Rojolele Rojolele KA Asemandi Aselapan Jalawara Kencana Bali Raja lmut Gebang Caping Gajah Pandan Wangi
a

IK1 (no. of plants) 0 3 5 0 0 0 8 0 1 5 5 2 9 9 0

IK2 (no. of plants) 10 9 0 70 10 59 3 0 0 24 7 0 8 0

IK3 (no. of plants) 4 0 2 0 0 0 0 0 0 0 24 11 0 41 0

Number of plants obtained from 50 seed scutella after 8-10 wk.

Table 2. Number of regenerated plants of selected indica and javanica cultivars on three callus induction media.

Cultivar

IK1 medium

IK2 medium (no. of plants) 24 5 9 (no. of plants seed-1) 0.48+0.14 0.10+0.06 0.26+0.11

IK3 medium (no. of plants) 39 95 7 (no. of plants seed -1) 0.18+0.24 0.19+0.28 0.18+0.09

(no. of plants)
Caping Gajah Rojolele Maninjau
a

(no. of plants seed -1) 1.70+0.15 0.33+0.13 0

35 13 0

Number of plants obtained from 50 seed scutella after 4 wk.

(Table 2); data were recorded for plant regeneration after 4 wk. We conclude that an efficient plant regeneration system has been obtained from embryogenic calli of javanica cultivars Rojolele KA and Caping Gajah despite earlier claims that javanicas are resistant to tissue culture. Bialaphos-resistant calli expressing gus-A gene resulting from bombardment experiments were obtained from indica rice cultivars Gajah Mungkur and Maninjau (data not shown). Scoring of gus transient expression of rice calli infected with Agrobacterium tumefaciens strain EHA 101 (pIGl21Hm) gave a high percentage of gus positive calli. We are now growing these calli on medium containing hygromycin to obtain stable transformed material.
Abdullah R, Cocking EC, Thompson JA. 1986. Efficient plant regeneration from rice protoplasts through somatic embryogenesis. Bio/Technology 4:10871090. Cooley J, Ford T, Christou P. 1995. Molecular and genetic characterization of elite transgenic rice plants produced by electric-discharge particle acceleration. Theor. Appl. Genet. 90:97-104. Christensen AH, Sharrock RA, Quail PH. 1992. Maize polyubiquitin genes: structure, thermal perturbation of expression and transcript splicing, and promoter activity following transfer to protoplast by electroporation. Plant Mol. Biol. 18:675-689. Li LC, Qu R, de Kochko A, Gauquet C, Beachy RN. 1993. An improved rice transformation system using the biolistic method. Plant Cell Rep. 12:250-255. Murashige T, Skoog R. 1962. A revised medium for rapid growth and bioassay with tobacco tissue cultures. Physiol. Plant. 15:473-497. Poonsapaya P, Nabors MW, Wright K, Vajrabhaya M. 1989. A comparison methods for callus culture and plant regeneration of RD25 rice ( Oryza sativa L.) in two laboratories. Plant Cell Tissue Organ cult. 16:175-186. Rueb S, Hensgens LAM. 1989. Improved histochemical staining for beta-Dglucuronidase activity in monocotyledonous plants. Rice Genet. Newsl. 6:168-169.

Variation among plants, regenerated from microspore-derived cell suspension protoplasts of rice
T. Cauchy, Centre de coopration internationale en recherche agronomique pour le dveloppement (CIRAD-CA), Centre fraais du riz, Mas du Sonnailler, Route de Gimeaux, F-13200 Arles, France; S. Pichot, CIRAD-CA, Biotrop-Gerdat Laboratory, BP 3035, F-31032 Montpellier Cedex, France; G. Clement, CIRAD-CA, Centre franais du riz; H. Char and E. Guiderdoni, CIRAD-CA, Biotrop-Gerdat Laboratory

Cited references

Rice embryo and microspore-derived calli are suitable for establishing embryogenic cell suspensions from which protoplast preparations can be readily isolated, engineered, and regenerated to plants. In that aim, somaclonal variation should be minimized. However, ploidy, chromosomal, morphological, and molecular changes have been extensively reported in plants regenerated from embryo callus-cell suspension protoplasts and their progenies. We recently reported variation in ploidy level among plants regenerated from microspore-derived, haploid cell suspension protoplasts of japonica rice cultivar Miara (Guiderdoni and Chair 1992), as well as a high frequency of agronomical changes in progenies of the diploid protoclones generated in that study (Mzencev et al 1995). To determine whether the importance of this variation is related to the genotype rather than to our haploid protoplast-

to-diploid plant system, we conducted ploidy analyses and field evaluation among plants regenerated from haploid protoplasts of the temperate japonica cultivar Ariete, the top rice variety in Italy and France. Flow cytometric analyses of nuclei isolated from protoplast preparations of two and three 4-mo-old cell suspensions established from microsporederived calli of cultivars Miara and Ariete, respectively permitted detection of haploid nuclei in four suspensions and a mixture of haploid and diploid nuclei in two suspensions (Table 1). We later found ploidy to remain stable over culture periods of up to 18 mo in fully haploid cell suspensions, whereas the haploid to diploid cell ratio varied among suspension cells exhibiting two ploidy levels. Protoplasts isolated from haploid suspensions yielded a majority of diploid plants whereas the majority of plants derived from suspensions consisting of both haploid and diploid cells exhibited ploidy levels superior or equal to 4n. Distribution of ploidy levels was similar among plants derived from haploid protoplasts of cultivars Miara and Ariete. This was consistent with our previous findings that attributed ploidy changes to early polyploidization events during protoplast culture (Guiderdoni and Char 1992). R1 progenies of 72 diploid plants derived from haploid protoplasts of variety Ariete (A4 cell suspension)

Table 1. Ploidy of plants regenerated from microspore-derived cell suspension protoplasts of cultivars Miara and Ariete.

Genotype

Cell suspension M1b M2 M3 A4 a A4 a A6 A7 TB1c

Ploidy level at time of isolation n n n+2n n n n n+2n 2n

Samples analyzed (no.) a 422 65 75 59 113 57 60 74

Frequency of plants exhibiting a ploidy level of n 2.1 1.5 0 1.7 2.5 1.8 10.0 0 2n 59.9 60.2 10.7 50.9 59.5 43.8 15.0 74.3 3n 12.6 21.5 4n 24.4 13.8 24.0 30.5 14.4 38.6 28.3 18.9 5n 0.9 0 12.0 3.4 1.0 5.3 1.7 1.4 >6n 0 3.0 52.0 0 5.5 7 43.3 2.7

Miara

1.3

Ariete

Thaibonnet
a

13.5 17.1 3.5 1.7 2.7

Results of two experiments, 4 and 10 mo after initiation of the cell suspension. b Mean data from 4 experiments (from Guiderdoni and Char 1992). c Seed embyro-callus-derived cell suspension (diploid control).

Vol. 22

No. 1

17

Table 2. Mean and standard deviation for six traits in the population of protoclonal lines compared with control lines.

Trait Days to flowering Fertile tillers (8 plants) Plant height (cm) Panicle length (cm) Grain length (mm) Grain width (cm)

Control lines Mean 91.1 64.6 98.1 17.3 9.1 2.9 SD 1.4 16.2 2.8 0.3 0.12 0.06

Protoclonal lines Mean 88.7 67.9 89.9 18.9 9.4 2.9 SD 1.8 13.7 6.5 1.2 0.22 0.07

Significance of the differences at 5% level (Fisher t test) NSa NS S NS NS NS

Miara protoclonal lines/Miara control lines b Later** Fewer** Taller** Longer** No difference Wider**

aNS due to difference in earliness between the two replicated experimental plots. b From Mzencev et al (1995); ** = significant at the 1% level.

were scored for six traits along with control lines in a replicated field experiment (Table 2). Overall, the range of variation observed among the Ariete protoclonal lines was much narrower than that in the protoclonal population generated from Miara following the same experimental procedures, though it remained for most traits wider than that of the control line population. A unidirectional drift toward earlier, shorter, and longer grained lines was observed. The mean of the protoclonal lines differed significantly from that of the control lines for plant height, and 22 and 44% of the lines exhibited significantly shorter culms and longer grains, respectively. On the other hand, we found no significant difference in fertile tillering, grain width, and panicle length between protoclonal and control lines.

This limited occurrence of variation among the protoclonal lines contrasts with our previous results from evaluating Miara protoclones in the field (Table 2). These findings, along with the analyses of other reports, confirm that the range, direction, and frequency of variation is more related to the genotype than to a given tissue culture procedure.

Guiderdoni E, Char H. 1992. Plant regeneration from haploid, microsporederived cell suspension protoplasts of Mediterranean rice (Oryza sativa L. cv. Miara). Plant Cell Rep. 11:618-622. Mzencev N, Clement C, Guiderdoni E. 1995. Variation among progenies of diploid plants regenerated from haploid, microspore-derived haploid cell suspension protoplasts of rice (Oryza sativa L.). Plant Breed. 114:149-154.

Cited references

AC, with long, slender grains) and japonica (highly responsive to AC, with short, thick grains) genotypes. We examined F1 (and their corresponding reciprocal), F2 , and BC populations with one BC to each parent. The cosegregation of AC response and grain type was evaluated. The diallel study indicated that the low AC response shows incomplete dominance with respect to the high response. Thus, the AC response in the F1 generation is below that of the most responsive parent (Table 1). Possible maternal effectsalthough not statistically differentwere only noted in the crosses when CT8707 was used as the nonresponsive parent (Table 1). The generation mean analyses suggest that simple (olygogenic) genetic systems of different sets of genes control callus induction and green plant regeneration. For callus induction, additive and dominant effects were highly significant, whereas for green plant regeneration, only additive effects were statistically significant (Table 2). The cosegregation analyses indicate that callus induction ( c2 = 0.853) and green plant regeneration (c2 = 0.978)
Table 1. Anther culture response of crosses between japonicas CT6241-17-1-5-1 and Todoroki Wase, and indicas IR43 and CT8707. a

Parent or cross IR43 CT8707 CT6241 Todoroki Wase IR43/CT8707 CT8707/IR43 CT6241/Todoroki Todoroki/CT6241 CT8707/Todoroki Todoroki/CT8707 CT8707/CT6241 CT6241/CT8707 IR43/Todoroki Todoroki/lR43 Todoroki/lR43 F 2 Todoroki/lR43//IR43 Todoroki/lR43//Todoroki IR43/CT6241 CT6241/IR43 CT6241/IR43 F 2 CT6241/IR43//IR43 CT6241/IR43//CT6241

Callus Green plants anther-1 (%) callus -1 (%) 0.0 0.0 19.8 42.2 0.0 0.0 58.3 65.2 2.2 15.9 6.6 11.8 15.4 17.9 12.7 5.2 24.9 13.9 13.4 4.5 5.4 12.3 d d d d bcd bc bcd bcd b b bc c b bc bc cd c bc 0.0 d 0.0 d 43.9 a 28.8 b 0.0 d 0.0 d 28.3 ab 23.2 ab 0.0 d 3.3 cd 3.3 cd 0.0 d 12.3 bc 12.8 b 11.7 b 7.2 c 17.6 b 11.9 bc 22.1 b 7.5 b 7.9 b 15.9 b

Toward introgression of high response to anther culture into indica rices

Z. Lentini, C. P. Martinez, E. Nossa, Rice Genetics, International Center of Tropical Agriculture (CIAT), A. A. 6713, Cali, Colombia

Anther culture (AC) is routinely used in our program to reduce the generation time for broadening the genetic diversity of breeding gene pools and to facilitate molecular tagging of genes. Most of these applications have been mainly restrained to crosses containing at least one japonica parent due to the recalcitrance of indica genotypes.

Replacing sucrose with maltose and adding silver nitrate to the callus induction medium substantially increases the AC response of indicas (Lentini et al 1995); however, the yield of doubled haploids per anther from indicas is still about 20-fold less than that from japonicas. Evidence exists that a few genes control AC response in rice (Miah et al 1985, Quimio and Zapata 1990). Therefore, the introgression of the high AC response from japonicas into indicas could facilitate applying AC as a routine tool for breeding indica rice. We conducted diallel and backcross (BC) inheritance analyses using crosses between true indica (nonresponsive to

a a a

aMeans with the same letter are not significantly different

according to the Waller-Duncan K-ratio test, p < 50.01.

18

IRRN 1997

Table 2. Generation mean analysis based on data from the populations P 1, P 2 , F 1, F 2, BC indica, and BC japonica of the crosses Todoroki/lR43 and CT6241/IR43. a Parameter b Callus anther -1 Todoroki/lR43 [m] [a] [d] [axa] [axd] [dxd] c 2 (2df) P 7.02* c 20.75** c 10.29* 13.79** 1.20 0.55 CT6241/IR43 5.11 9.12** 20.33** 14.76** 3.70 0.16 Green plant callus -1 Todoroki/lR43 16.29** 15.72** 1.58 17.85** 3.90 0.14 CT6241/IR43 21.02** 21.46** 3.86 28.19** 4.10 0.13

genies from each selected line for its in vitro response and agronomic characteristics. Selected plants will be used as parents in crosses with plants from a recurrent selection population to develop a genetically diverse indica gene pool with increased response to AC.

Cited references

a According to the simplest model that explains the observed values. b [m] = mean value between parents, [a] = additive effects, [d] = dominant effects, [axa], [axd], and [dxd] interactions between additive and/or dominant effects. c *, ** = different from zero

at p = 0.05 and 0.01, respectively.

segregate independently from the grain type. Plants combining high response to AC from the japonica parents (up to 70% callus induction and 90% green plant regeneration) and grain type from the indica parents (up to 8.7 mm long

and 2.9 mm wide) were recovered as early as in the F2 generation and, subsequently, in the BC indica generations. Twenty lines with high response to AC and long, slender grains were selected. We are evaluating individual pro-

Lentini Z, Reyes P, Martinez CP, Roca WM. 1995. Androgenesis of highly recalcitrant rice genotypes with maltose and silver nitrate. Plant Sci. 110:127-138. Miah MAA, Earle ED, Khush GS. 1985. Inheritance of callus formation ability in anther cultures of rice, Oryza sativa L. Theor. Appl. Genet. 70:113-116. Quimio CA, Zapata FJ. 1990. Diallel analysis of callus induction and greenplant regeneration in rice anther culture. Crop Sci. 30:188-192.

Random mating of composite populations for improving restorers in rice

T. S. Bharaj and S. S. Virmani, IRRI

Most commercial indica hybrid rices have been developed through the wild abortive (WA) cytoplasmic male sterility (CMS) system, which involves using CMS (A), maintainer (B), and restorer (R) lines. Desirable restorer lines should possess good fertility restoration ability, high performance, multiple disease and insect resistance, good general combining ability, high pollen production and/or dispersal, and good grain quality. Hybrid rice breeding programs continuously need these lines to breed superior hybrids. Most hybrid rice breeders usually select R lines among elite lines bred by inbred breeding programs where crosses are generally made without considering the parents restoration abilities. Alternatively, R lines are also developed from R R or A R crosses specifically made for this purpose. Thus, frequency of R lines among a group of breeding lines depends on the

Procedure used at IRRI to develop random mating composite populations for improving restorer lines.

Vol. 22, No. 1

19

latters origins and pedigrees. To regularly extract a high frequency of R lines, two random mating composite populations of restorers were developed by exercising male sterility-facilitated recurrent selection. The procedure (see figure) involves 1. Developing a base population derived from the crosses of a genic male sterile line with a series of selected restorer lines (steps 1-4). 2. Random mating of male sterile plants with the pollen from fertile plants occurring in the base and subsequently derived populations until both male sterile and fertile plants are in equilibrium (steps 5-8). 3. Evaluating the random mating population and selecting desirable fertile plants to extract improved restorer lines, and then selecting desirable male sterile plants to form a new random mating population (step 9). Following the procedure, two random mating composite populations of restorers were constituted at IRRI. The first population (IR69701 CP138) was derived from nine early-maturing (less than 120 d) elite restorer lines and IR36 (ms) and the second population (IR69702 CP139) was derived from 14 medium-maturing (120-140 d) elite restorer lines and IR29723-143-3-2-1 (ms) (see table). The two ms lines used were in the genetic background of restorers. During the 1995 dry season, we reached step 8 of the procedure, in which the populations showed equilibrium among fertile and sterile plants. These populations are being used as a source for extracting genetically diverse improved restorer lines for WA CMS in rice. We are sharing seeds of these composite populations with scientists in national agricultural research systems to help them extract genetically diverse R lines for their local hybrid rice breeding programs. During the 1995 wet season and 1996 dry season, we selected about 300

Component lines of early-maturing random-mating composite restorer population (lR69701 CP 138) and mediummaturing random-mating composite restorer population (lR69702 CP 139) and their salient features.

Component line IR69 701 CP 138 lR36 ms IR64R IR66R IR72R IR9761-19-1R IR19058-107-1R lR44675-101-3-3-2-2R lR47310-94-4-3-1R Milyang 46R LA2877-2-7 IR69 702 CP 139 lR29723-143-3-2-1 ms lR32809-26-3-3R lR34686-179-12-1R lR40750-82-2-2-3R lR54055-142-2-1-2-3R lR54742-22-19-3R lR54959-41-2-2R lR55628-79-1-3-3-3R IR24R IR46R IR64R IR70R ITA121 LA1451-5R LA2827-2-7R

Salient features a

Genic male sterile, resistant to multiple diseases and insects, widely adapted High yielding, good grain quality, and resistant to major diseases and insects High yielding, resistant to tungro virus and BPH biotype 4 High yielding, resistant to major diseases and insects, widely adapted High pollen producer, good general combiner High yielding, resistant to BPH and BB Resistant to BPH, GLH, and BB High yielding, resistant to major diseases and insects Elite Indica/japonica derivative line originating in Republic of Korea, good general combiner Long anthers derived from O. sativa/ O. longistaminata Genic male sterile, high yielding, good general combiner Resistant to BPH, cold tolerant, good general combiner Resistant to major diseases and insects, tolerant of cold and salinity Good general combiner and resistant to several diseases and insects Resistant to BPH, GLH, and BB Elite line derived from O. sativa/O. officinalis Resistant to BPH, GLH, and BB Resistant to GLH High yielding, good general combiner, possesses broad spectrum of restoration ability High yielding, good general combiner, also adaptable to favorable rainfed lowland ecosystem High yielding, good grain quality, and resistant to major diseases and insects Resistant to major diseases and insects Elite line introduced from IITA, adapted to irrigated ecosystems in West Africa Long anther derived from O. sativa/O. longistaminata Long anther derived from O. sativa/O. longistaminata

a BPH = brown planthopper; GLH = green leafhopper; BB = bacterial blight.

single plants from these populations at IRRI. We are now evaluating these in a pedigree nursery. The promising R lines identified can again be made into

composites by repeating this procedure, thus developing new random mating composite populations for further improving R lines.

Yield potential
Influence of brassinosteroid on rice seedling growth
Wang Sangen, Agronomy Department, Southwest Agricultural University, Chongqing 630716, China

Since brassinosteroid (BR) was discovered as a new family of plant hormones, scientists have recognized its importance in regulating plant growth and reproduction. BR and its analogues have been synthesized or isolated and are available to biologists for basic physiological and biochemical studies

and field experiments. BR has been isolated in rice plants. Seeds of hybrid rice variety D You 63 and conventional variety Fujiang 2 were surface-sterilized, rinsed in sterile water, and then dried with blotting paper. BR, which was isolated from beeswax, was dissolved with a little alcohol and then diluted with distilled water to various concentrations. Seeds were presoaked in different concentrations of BR or distilled water (as a control). Imbibed seeds were placed in 30- 20-cm plastic culture pans. Nutrient solution was added when the first true leaf emerged.

20

IRRN 1997

The effects of brassinosteroid (BR) on rice seedling growth.

BR concentration (ppm) Variety Fujiang 2 Character Shoot height (cm) Dry weight (mg plant-1) Root-shoot ratio Chlorophyll content (mg g -1 fresh weight) Root dehydrogenase activity (mg g -1 h -1 ) a - amylase b activity b - amylase activity D You 63 Shoot height (cm) Dry weight (mg plant-1) Root-shoot ratio Chlorophyll content (mg g -1 fresh weight) Root dehydrogenase activity (mg g -1 h -1) a - amylase activity b - amylase activity 0 8.18 0.15a c (100) 9.31 0.05 b (100) 0.536 0.010 b (100) 0.786 0.031 a (100) 2.11 0.82 a (100) 61.20 5.24 c (100) 34.43 3.31 bc (100) 9.15 0.92 ab (100) 9.97 0.27 a (100) 0.524 0.021 b (100) 0.854 0.045 a (100) 2.59 0.34 ab (100) 78.17 2.65 d (100) 41.03 3.54 c (100) 10-5 8.53 0.07 b (104.3) 9.73 0.36 ab (104.5) 0.590 0.014 ab (110.1) 0.801 0.015 a (101.9) 2.26 0.24 a (107.1) 84.40 6.14 b (137.9) 42.63 2.14 b (123.8) 9.47 0.91 ab (103.5) 10.37 0.51 a (104.0) 0.608 0.025 a (116.0) 0.851 0.22 a (99.7) 3.15 0.42 a (121.6) 100.83 7.74 bc (129.0) 47.43 4.17 bc (115.6) 10-3 9.24 0.58 a (113.0) 10.21 0.28 a (109.7) 0.631 0.029 a (117.7) 0.824 0.091 a (104.8) 2.59 0.36 a (122.8) 114.03 3.71 a (186.3) 47.50 1.37 ab (138.0) 9.98 0.74 a (109.1) 10.69 0.43 a (107.2) 0.600 0.045 a (114.5) 0.895 0.036 a (104.8) 3.67 0.91 a (141.7) 130.90 9.13 a (167.5) 61.70 5.29 a (150.4) 10 -1 8.11 0.27 cd (99.1) 9.890.76 ab (106.2) 0.629 0.025 a (117.4) 0.811 0.083 a (103.2) 2.21 0.51 a (104.7) 116.70 9.21 a (190.7) 55.03 3.74 a (159.8) 9.21 0.26 ab (100.7) 10.23 0.76 a (102.6) 0.603 0.036 a (115.1) 0.865 0.034 a (101.3) 2.95 0.73 ab (113.9) 114.70 4.27 b (146.7) 55.17 6.02 ab (134.5) 10 7.37 0.59 d (90.1) 8.41 0.31 c (90.3) 0.605 0.023 a (112.9) 0.795 0.027 a (101.2) 2.15 0.26 a (101.9) 70.17 4.05 bc (114.7) 31.32 4.16 c (91.0) 8.76 0.31 b (95.7) 9.84 0.46 a (98.7) 0.528 0.009 b (100.8) 0.843 0.027 a (98.7) 2.40 0.21 b (92.7) 97.70 6.35 c (125.0) 46.70 3.16 bc (113.8)

aValues given are followed by the standard deviation. In a row, means followed by the same letter are not significantly different at the 5% level by LSD statistical test. Numbers in parentheses are % of control. b The amylase activities were expressed as g maltose kernel -1s -1.

To determine seed and seedling vigor, we measured shoot height, plant fresh and dry weight, root-shoot ratio, chlorophyll content, and root dehydrogenase. The dry weight was measured by holding samples at 80 C to constant weight. Chlorophyll was extracted with acetone and measured using the method of Arnon. 2, 3, 5-triphenyl tetrazolium chloride was used to quantitatively analyze root dehydrogenase activity, hence estimating root viability. The enzyme solution extracted from the kernel was incubated with a citric acid buffer (pH 5.6) and starch solution for 5 min. a and b amylase activities were measured using the 3, 5-dinitrosalicylic acid method to understand the efficiency of matter transformation. Four replicates of 50 plants comprised each sample. After soaking in BR solution, both D You 63 and Fujiang 2 had increased shoot height, seedling dry weight,

Root and shoot dry weights and root-shoot ratio of rice seedlings treated with different brassinosteroid (BR) concentration.

chlorophyll content, and root dehydrogenase activity in available BR solutions (see table). Root weight and rootshoot ratio were especially enhanced (see figure). a and b amylase activities rose significantly during germination and were promoted by BR solutions. BR has been found to have an extensive physiological role in plant development. Amylases are also key enzymes for starch catabolism during rice seed germination. Higher amylase activity in kernels would enhance seedling vigor. BR-enhanced seed and seedling vigor, including shoot height, dry weight, root-shoot ratio, chlorophyll content, and root dehydrogenase activity, was associated with the increase of a and b amylase activities in seeds. BR could therefore be applied in solution to improve germination and seedling vigor.

Vol. 22, No. 1

21

Grain quality
Physical and milling characters of popular Maruteru rice varieties
P. V. K. Jagannadha Rao and P. S. S. Murty, Agricultural Research Station (ARS), Maruteru 534122, Andhra Pradesh, India

We evaluated the physical and milling characters of eight rainfed lowland rice varieties commonly grown in Andhra Pradesh and parts of Maharashtra,

Karnataka, and Orissa, India. Sampling was done 1 mo after harvest; grain had 14% moisture content. The 1,000-grain weight of Vijetha was heaviest (see table). All have medium-shaped grain and mediumlength grain, extra for Vijetha, which has long grain. The brown rice recovery was higher for MTU4870 and Krishnaveni compared with other varieties. The milled rice recovery was highest for MTU4870.

The bran and germ percentage for Nandi was highest (10.1) followed by that for Krishnaveni (9.4) and Chaitanya (9.1). Among these varieties, head rice recovery was highest in MTU4870. Except for Swarna and Chaitanya, all other varieties yielded more than the acceptable head rice recovery of 60% (see table). Brokens were lowest in Vijetha (3.4%) and highest in Swarna (7.6%).

Physical and milling characters of popular rice varieties. Agricultural Research Station, Maruteru, India. 1995.

Variety Duration (d) Grain length (mm) 5.77 6.35 6.13 6.10 5.81 5.69 6.63 5.88 6.05 0.32 5.34 Grain breadth (mm) 2.24 2.16 2.31 2.25 2.35 2.24 2.37 2.28 2.28 0.07 2.98

Physical characters Grain thickness (mm) 1.72 1.70 1.71 1.75 1.62 1.72 1.85 1.79 1.73 0.07 3.89 L-B ratio Grain length Grain shape 1000grain weight (g) 19.90 20.60 21.10 21.10 20.90 20.20 27.50 22.00 21.60 2.44 4.50 Brown rice

Milling characters (%) Bran and germ Milled rice recovery 66.60 68.20 69.20 66.60 61.60 67.70 65.80 70.90 67.08 2.75 4.10 Broken rice Head rice recovery 58.99 63.90 63.98 61.10 55.10 62.30 62.40 66.30 61.76 3.45 5.58

Swarna MTU7029 Pratibha MTU5293 Vajram MTU5249 Nandi MTU5182 Chaitanya MTU2067 Krishnaveni MTU2077 Vijetha MTU 1001 MTU4870 X SD CV

150 165 150 150 150 150 140 150

2.57 2.94 2.64 2.72 2.46 2.54 2.80 2.57 2.67 0.15 5.74

Medium Medium Medium Medium Medium Medium Long Medium

Medium Medium Medium Medium Medium Medium Medium Medium

75.50 8.90 74.50 6.30 76.60 7.40 76.70 10.10 70.70 9.10 77.10 9.40 74.50 8.70 77.20 6.30 75.38 8.28 2.17 1.44 2.88 17.34

7.61 4.30 5.22 5.50 6.50 5.40 3.40 4.60 5.32 1.31 24.56

Variability of quality indices in aromatic rice germplasm

V. P. Singh, Indian Agricultural Research Institute, New Delhi, India; G. S. Khush and N. dela Cruz, IRRI

Aromatic rice varieties are generally preferred over nonaromatic types and therefore earn premium prices. Among the aromatics, those with long slender grains (length >6.5 mm and breadth <2.0 mm) and high lengthwise volume expansion after cooking are popularly called basmati rice. These varieties have helped countries such as India, Pakistan, and Thailand earn considerable foreign exchange. We evaluated 356 traditional aromatic rice varieties for eight grain and
22 IRRN 1997

cooking characteristics: amylose content (AC), gelatinization temperature (GT) as indicated by the alkali spreading score, gel consistency (GC), milled rice kernel length (KL), kernel breadth (KB), kernel length-breadth ratio (L/B), kernel elongation after cooking (KE) (length of cooked rice divided by

original kernel length using the average of 10 unbroken milled rice kernels), and 1,000-grain weight (WT). Wide variation was observed for all the characteristics, with the maximum being for GC followed by WT and AC. Most of the aromatic varieties from India, Myanmar, Nepal, and Pakistan

Table 1. Varieties (no.) in different groups based on quality indices.

Quality index Amylose content (%) Gelatinization temperature (Alkali spreading score) Gel consistency (mm) Kernel elongation Length-breadth ratio

Very low 2.5-9.0 27 6&7 54 1.31-1.60 66 1.3-2.0 44

Low 9.0-20.0 169 0-40 77 1.61-1.80 100 2.01-3.0 140

Intermediate 20.1-25.0 155 4&5 273 41-60 148 1.81-2.0 101 3.01-4.0 159

High >25.0 5 1&3 29 >60 131 >2.0 69 >4.01 13

possess intermediate AC, GT, and GC. WT, one of the key selection indices for high grain yield, ranged from 7.5 to 42.2 g. Milled rice KL ranged from 3.20 to 8.06 mm, and KB ranged from very thin (1.38 mm) to very thick (3.08) mm. Based on the Food and Agriculture Organization of the United Nations' milled rice scale, the varieties were grouped as long-slender (110), medium-slender (9), long-bold (82), medium-bold (19), short-bold (33), short-round (42), and medium (63). High volume expansion after cooking (maximum linear KE with minimum breadthwise expansion) is one of the characters of basmati rice. Varieties that double their length with cooking are preferred. KE range was wide among the varieties (Table 1). We determined the Pearsons correlation coefficients for eight quality

indices. AC showed highly significant negative correlation with all the traits except LB (0.10) and KE (0.19). WT had significant negative correlation with AC and KE, suggesting that lower grain weight types must be selected to identify basmati types. The association between KE and KL as well as LB was negative and significant, indicating that many of the slender grains do not

elongate after cooking. KE had a significant positive association with AC, while that with GT was significant and negative. Most of the aromatic varieties with greater linear KE also have intermediate AC and GT. Among the aromatic germplasm evaluated, 39 accessions are basmati based on grain shape and elongation; details are provided for nine (Table 2).

Table 2. Some typical Basmati varieties and their grain and cooking quality characteristics.a

Variety Barah Mussa Tarom Mulai Sadri Pakistani Fine Sathi Basmati Basmati 370 Karnal Local Hansraj
a

Origin Afghanistan Iran Iran Iran Pakistan India India India India

AC 17.6 19.2 19.5 18.2 20.4 18.4 21.0 21.2 18.1

GT 4.58 4.20 3.58 4.54 5.40 3.83 3.83 3.66 4.50

GC 70 40 55 35 50 58 57 60 60

KL 7.55 7.62 7.76 7.66 6.96 7.24 6.84 7.06 6.88

KB 1.88 1.88 1.96 1.94 1.96 1.92 1.84 1.90 1.94

LB 4.02 4.05 3.96 3.45 3.55 3.78 3.72 3.72 3.55

KE 2.18 1.83 1.93 2.10 1.93 1.95 2.10 1.85 2.17

AC = amylose content, GT = gelatinization temperature, GC = gel consistency, KL = kernel length, KB = kernel breadth, LB = lengthbreadth ratio, KE = kernel elongation.

Biochemical composition of principal components of rice seeds

N. M. Anigbogu, Federal University of Agriculture, PMB 7267 Umuahia, Abi State, Nigeria

Rice is an important source of carbohydrates in the diets of several Nigerian populations, especially in the forest belt. We determined the nutritional

values of rough rice, unpolished rice, polished rice, hulls, bran, and rice mill feed for rice grown in Nigeria. Rice components were collected from the rice mill of the College of Agriculture, Enugu State University, Abakaliki Campus, Nigeria, ground, and then mixed to obtain homogeneous samples. The methods of the Association of Official Analytical Chemists were used to determine moisture,

crude protein, ash, crude fiber, crude fat, dry matter, and nitrogen free extract. Calorific values were calculated by multiplying the mean values of crude protein, fat, fiber, and nitrogen free extracts by Atwater factors. The mineral composition was assayed for each sample in 5 N HCl and then analyzed with a Perkin-Elmer atomic absorption spectrophotometer and flame photometry method.

Chemical composition of the principal nutritional components of rice grown in Nigeria.

Composition (%)a Constituent Rough rice Unpolished rice Polished rice Rice hulls Rice bran Rice mill feed Rice starch refuse

Proximate composition (n=3) Moisture Dry matter Ash Crude protein Crude fat Crude fiber Nitrogen free extract Calorific value (kcal 100 g -1) Mineral content (n=3) Calcium Phosphorus

12.3 87.7 6.4 9.6 1.7 11.1 58.9 334.1

0.4 0.1 0.2 0.0 0.5 0.8 1.8

12.3 87.8 2.1 9.1 1.1 0.5 75.0 348.5

0.2 0.4 0.1 0.6 0.1 0.5 1.0

10.8 89.2 1.6 8.4 1.0 0.2 78.1 355.2

0.1 0.7 0.1 0.4 0.1 0.0 0.4

11.4 88.6 17.2 3.4 0.6 34.7 32.7 288.5

0.2 0.4 1.0 0.2 0.2 1.6 1.1

11.2 88.9 10.8 12.0 15.1 8.7 42.2 387.6

0.2 0.5 0.3 0.4 0.4 1.2 1.5

6.1 93.9 15.9 6.9 4.8 27.9 38.5 336.2

0.3 3.3 2.8 1.2 0.7 6.8 4.5

9.1 90.9 2.1 9.9 0.9 1.9 78.2 388.1

1.0 0.2 0.1 0.9 0.0 0.0 0.9

0.1 0.0 0.3 0.0

0.0 0.0 0.1 0.0

0.0 0.0 0.3 0.0

0.3 0.0 0.1 0.0

0.2 0.0 1.7 0.1

0.1 0.0 0.4 0.0

0.1 0.0 0.4 0.0

aValues are means standard deviations of three replications.

Vol. 22. No. 1

23

Moisture content was moderate among the different samples (see table). The percentage of ash was larger in rice hulls (17.2%), rice mill feed (15.9%), and rice bran (10.8%) than in other samples; it was lowest in polished rice (1.6%). All principal components were good sources of crude protein except rice hulls (3.4%), and all were low in crude

fat except for rice bran (15.1%). Crude fiber was greatest for rice hulls (34.7%) and rice mill feed (27.9%) and moderately low for rough rice (11.1%) and rice bran (8.7%). The different components had relatively high calorific values and were identical in inorganic element, calcium, and phosphorus contents.

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.

Pest resistance
Activity of the promoter of a rice lipid transfer protein gene in transgenic rice
E. Guiderdoni, N. M. Ferrire, Centre de cooperation international de recherche agronomique pour le dveloppement (CIRADCA), Biotrop-Gerdat Laboratory, BP5035,34032, Montpellier cedex, France; F. Vignols, Centre nationale de la recherche scientifique (CNRS), University of Perpignan, Laboratoire de physiologie et biologie molculaire des plantes, URA565, 52 Avenue de Villeneuve, 66860 Perpignan Cdex, France; T. Legavre, H. Char, and M.J. Cordero, CIRAD-CA; and M. Delseny, CNRS, University of Perpignan

Scientists have assumed that plant lipid transfer proteins (Ltp) participate in membrane biogenesis, which facilitates the movement of lipids between membranes. Expression analysis of these genes revealed complex patterns characterized by high cell and tissue specificity. Homologous or heterologous transfer of translational Ltp promoter-gus fusions in transgenic plants should facilitate localization of that expression. Recently, Kalla et al (1994) described aleurone cell-specific activity of the promoter of the barley Ltp2 gene in transgenic rice. We report here activity of the promoter of a rice Ltp genea member of a small multigenic familyin transgenic rice. A construct bearing a translational fusion between the promoter region of a rice lipid transfer protein (Ltp) gene (Vignols et al 1994), the uidA coding sequence, and the nos 3' terminator

(p D Gus1177) was transferred along with a plasmid consisting of the bar gene driven by the 35S promoter (p35SAc, Hoechst, Germany) to haploid, microspore callus cell suspension protoplasts in four separate experiments of PEG-mediated transformation, as described in Char et al (1995). Transient activity of the Ltp promoter quantified using the fluorimetric GUS assay 48 h after transformation was not significantly different from that of control protoplasts. However, protoplasts treated with the pEMUGN plasmid (Last et al 1991) exhibited very high GUS activity, suggesting that the Ltp promoter is not active in unorganized, undifferentiated cells. The four experiments yielded 274 ammonium glufosinate-resistant calli that regenerated 136 plants. Histochemical assay of protoplast-derived colonies revealed faint GUS activity in some calli. Histochemical assay of 103 out of 136 regenerated plants enabled us to detect GUS activity in 47 plants, suggesting that the cotransformation efficiency reached at least 45%. The Ltp promoter appeared more active in the innermost developing leaves and variation in intensity of staining was noted among the plants. No GUS activity was found in the roots of plants exhibiting GUS-positive leaves. Southern blot analysis of 13 plants confirmed integration of both uidA and bar genes in the genome of regenerated plants (see figure). Histological localization of GUS activity in transverse sections of shoots

of these plants permitted us to detect GUS activity in all cell types of the first innermost leaf, whereas that activity appeared localized in the outer epidermis and subepidermis of the second innermost leaf. In leaves of higher rank, activity became restricted to the outer epidermal cell layer. In fully mature leaves, GUS activity was mainly detected in stomate guard cells and in vascular bundles.

5.9Kb

Southern blot analyses of 8 T 0 plants regenerated from protoplasts treated with the p D Gus1177 plasmid and exhibiting GUS activity in leaves. Genomic DNA was digested by Bam HI, which cuts the plasmid once, resolved in a 0.8% agarose gel, transferred to a nylon membrane, and then probed with an internal 1.8-kb fragment corresponding to the gus coding sequence. UT= DNA from an untransformed plant. The expected position of 5.9-kb fragments corresponding to tandem insertions of the plasmid is marked with an arrow.

24

IRRN

1997

Ongoing histochemical assay of reproductive organs of plants transformed with the p D Gus1177 plasmid permitted detection of GUS activity in the sterile lemmas and in the lemma and palea of the spikelets and in the maternal tissues of young developing anthers containing uninucleated microspores, but not in ovaries or in the walls of mature anthers containing binucleated pollen. We also found activity in pollen and in the pericarp of the mature seed.

Tissue specificity and developmental regulation of the rice ltp promoter make it a good candidate for driving genes conferring resistance to leaffolder and blast that mainly develops on young leaves.

Cited references

Char H, Legavre T, Guiderdoni E. 1997. Transformation of haploid microsporederived cell suspension protoplasts of rice (Oryza sativa L.) Plant Cell Rep. (in press) Kalla R, Shimamoto K, Potter R, Stein Nielsen P, Linnestad C, Olsen OA. 1994.

The promoter of the barley aleuronespecific gene encoding a putative 7kDa lipid transfer protein confers aleurone cell-specific expression in transgenic rice. Plant J. 6:849-860. Last DJ, Brettel RIS, Chamberlain DA, Chaudury AM, Larkin PJ, Marsh EL, Peacock WJ, Dennis ES. 1991. pEMU, an improved promoter for gene expression in cereal cells. Theor. Appl. Genet. 81:581-588. Vignols F, Lund G, Pammi S, Trmoussaygue D, Grellet F, Kader JC, Puigdomench P, Delseny M. 1994. Characterization of a rice gene coding for a lipid transfer protein. Gene 142:265-270.

Pest resistancediseases
Sources of resistance to sheath blight
M. M. Reddy, T. Madhusudan, N. Kulkarni, and M. Kashikar, Rice Section, Agricultural Research Institute (ARI), Rajendranagar, Hyderabad 500030, India
Table 2. Reaction to sheath blight disease of promising lines and other characters. Rajendranagar, Hyderabad, India. 1993-95.

Line

Parentage

Disease score a 1993 1.8 0.6 6.2 1994 0.0 1.6 8.6 1995 0.6 1.2 7.8

Mean

Plant height (cm) 89 89 91

Tillers hill -1 (no.) 9.0 8.4 10.2

Grain yield (t ha -1 ) 5.7 5.5 4.2

RNR15336 RNR82096 TN1 (check)

a

Sheath blight of rice, caused by Rhizoctonia solani Kuhn, is becoming a major rice disease because of the increased use of nitrogen fertilizer and shorter, high-tillering cultivars, which together produce a microclimate that enhances disease incidence. Because chemical control is costly and not effective, we screened breeding lines developed at ARI to identify resistant sources. This experiment was conducted with 166, 147, and 144 entries during the 1993, 1994, and 1995 wet
Table 1. Frequency distribution of sheath blight disease among entries. Rajendranagar, Hyderabad, India. 1993-95.

Saleem/ W12708 Tellahamsa/ Akaswari

0.8 1.1 7.5

Mean of 10 hills.

Disease scale 0 1 2 3 4 5 6 7 8 9

Entries (no.) 1993 0 3 4 9 5 11 6 71 39 19 1994 1 0 1 0 2 5 1 12 39 86 1995 0 1 1 0 0 1 6 16 40 80

seasons, respectively, under irrigated transplanted conditions. TN1 was the susceptible check. Each entry was grown in two 2-m rows at 15- 15-cm spacing. Artificial inoculation was done at maximum tillering by placing 2-3 typha bits containing the inocu1um in the middle of the hill just above the water level, tied with a rubber band. Ten hills were inoculated per entry. To compute relative lesion height, the average vertical height of the upper most lesion and average plant height were measured at grain filling. Results revealed that most of the entries were highly susceptible (Table 1) but two entries, RNR15336 and RNR82096, produced low relative lesion height (Table 2) and appear to be resistant to sheath blight. RNR15336 (120 d duration) and RNR82096 (140 d duration) have long slender grains and are being evaluated in advanced yield trials.

Effect of blast disease on rice yield

H. Surek and N. Beser, Thrace Agricultural Research Institute, P. O. Box 16, Edirne, Turkey

We studied the effect of blast disease on 12 rice varieties and lines at Edirne and Uzunkpr, Turkey, in 1995. Eighty kilometers separate the two sites. The experiment was laid out at both sites in a randomized complete block design with four replications under continuous irrigation. Plot size was 20 m 2 . Four hundred and fifty pregerminated seeds m-2 were broadcast in standing water on 20 May. The same agronomic techniques were practiced at both sites. Fertilizer was applied at 150 kg N ha-1 in three equal splits (basal dressing and topdressing at tiller initiation and panicle initiation) and 80 kg P ha-1 as a single basal dressing.

Vol. 22. No.1 25

Mean values for yield, yield components, and blast disease infection in Edirne (E) and Uzunkpr (U), Turkey. 1995.

Variety E

Grain yield (t ha -1) U Mean E

1,000-grain weight (g) U Mean

Total rice recovery (%) E U Mean E

Head rice yield (%) U Mean

Disease infection a Leaf blast U Node and neck blast U MS S MS HS HS MS MS MS S MS S S

Ribe Ergene Serhat-92 Ana/Mar Lap/PG TR-427 TR-475 TR-489 TR-648 TR-765 lpsala Surek-95 Mean F values Variety Location Location x variety LSD (0.05) CV (%)

5.8 d 6.7 bcd 6.8 bc 6.1 cd 4.7 e 7.4 ab 6.3 cd 6.3 cd 6.9 bc 6.9 bc 6.4 cd 7.8 a 6.5

4.6 abc 4.4 bc 3.8 bcd 3.6 cd 2.2 e 5.7 a 4.6 abc 4.8 ab 3.1 de 5.1 ab 3.2 de 3.9 bcd 4.0

5.2 bcde 5.6 bcd 5.3 bcde 4.9 de 3.5 f 6.5 a 5.4 bcde 5.6 bc 5.0 cde 6.0 ab 4.8 e 5.9 b 5.3

31.7 fg 36.0 cd 32.5 ef 33.5 e 30.5 g 32.8 ef 35.0 d 36.1 cd 39.0 b 36.5 c 40.8 a 35.5 cd 35.0

31.0 b 28.3 cd 25.8 e 26.9 de 25.1 e 28.3 cd 30.4 bc 32.6 ab 32.3 ab 31.0 b 34.2 a 26.7 de 29.4

31.4 eg 32.2 def 29.1 hi 30.2 gh 27.8 i 30.6 fgh 32.7 cde 34.3 bc 35.6 b 33.7 cd 37.5 a 31.1 efg 32.2

70.7 de 72.4 bc 73.5 ab 69.2 ef 68.3 f 72.1 bcd 73.1 ab 73.3 ab 73.1 ab 74.2 a 71.1 cd 72.2 bcd 72.1

70.5 ab 70.5 ab 69.7 abc 69.2 bc 69.1 bc 71.6 a 70.0 abc 71.7 a 70.1 abc 69.0 bc 71.6 a 68.0 c 70.1

70.6 bc 71.5 abc 71.6 ab 69.2 de 68.7 e 71.9 ab 71.5 ab 72.5 a 71.6 ab 71.6 ab 71.3 abc 70.1 cd 71.0

63.8 bc 63.2 c 68.4 ab 64.9 abc 55.4 d 69.3 a 65.5 abc 63.3 c 64.4 abc 68.1 abc 64.4 abc 64.4 abc 64.6

56.3 a 53.2 ab 42.4 c 47.4 bcd 45.6 cd 54.7 a 55.3 a 56.8 a 53.3 ab 55.1 a 55.0 a 50.3 abc 52.1

60.1 bc 58.2 de 55.4 g 56.1 fg 50.5 h 62.0 a 60.4 abc 60.0 bc 58.8 cde 61.6 ab 59.7 cd 57.3 ef 58.3

MS MS MR S HS MR MR MR MS MR MS MR

6.637** b 6.01** 0.90 9.46 1.10 19.42

8.64** 267.52** 3.10** 0.70 13.81

40.80** 1.35 2.68

12.14** 2.43 5.75

21.42** 262.02** 3.11** 1.69 5.27

9.47** 1.67 1.61

2.22* 2.25 2.23

5.21** 41.81** 4.02** 1.40 1.97

3.91** 5.17 5.56

3.54** 7.11 9.49

4.11** 189.22** 2.83** 1.68 7.62

a Based on field Evaluation of 70 and 100-d-old plants in Uzunkpr only. No lnfection was reported in Edirne. MR = moderately resistant, MS = moderately susceptible, S = susceptible, HS = highly susceptible. b * and ** = significant at 0.05 and 0.01 level, respectively.

initiation) and 80 kg P ha -1 as a single basal dressing. We examined the effects of blast disease infection on rice yield, total rice recovery, head rice, and 1,000-grain weight. The blast disease infection in 1995 was the most severe ever recorded in the Uzunk pr region. It caused a 20% yield loss over 25,000 ha of riceland, with some farmers not even harvesting their crops. There was no disease infection, however, in Edirne. Significant differences in all characters studied were recorded for the two locations, with all being less for rice grown in Uzunk pr (see table). The varieties with moderate susceptibility to node and neck blast (Ribe, TR-427, TR-475, TR-489, and TR-765) differed less for yield and yield components
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.

between the two sites than the susceptible and highly susceptible varieties (Ergene, Serhat-92, Ana / Mar, Lap/ PG, TR-648, Ipsala, and S rek-95). The environmental factors did not, in general, affect 1,000-grain weight very much, although huge differences did exist for some varieties between the

locations. Blast infection, plus other environmental factors, was therefore the main reason for smaller yields at Uzunk pr. Node and neck blast caused more damage to the varieties than did leaf blast because none of the varieties were even moderately resistant to it.

Pest resistanceinsects
Resistance of varieties derived from Oryza sativa/Oryza officinalis to brown planthopper in the Mekong Delta, Vietnam
Luong Thi Phuong, Luong Minh Chau, Plant Protection Department, Cuu Long Delta Rice Research Institute (CLRRI), Vietnam; and M. B. Cohen, IRRI

One hundred lines containing a brown planthopper (BPH) resistance gene from the wild rice species Oryza officinalis were sent from IRRI to CLRRI in 1990. Several of these lines were re-

leased to farmers and have been widely grown in the Mekong Delta, although susceptibility to blast has limited their popularity. It is not known whether the BPH resistance gene from O. officinalis is a novel gene or one of about 10 BPH genes already identified from other sources. In tests at IRRI, the gene appears to be dominant. We report here on the resistance of varieties with the O. officinalis gene to BPH in the Mekong Delta, and compare them with a series of test varieties containing known resistance genes.

26

IRRN 1997

BPH resistance was evaluated using the standard seedbox screening test and scored with the Standard evaluation system for rice. Each year (1993-96),

plants were infested with a fresh BPH population collected at CLRRI and reared in a screenhouse for one to three generations on TN1.

Brown planthopper resistance of selected varieties. Cuu Long Delta Rice Research Institute, Vietnam. 1993-96.

Variety TN1 Mudgo .. ASD7 Rathu Heenati Babawee ARC10550 Swarnalata T12 Chin Saba Pokkali IR64 Ptb33 MTL 103 (lR54751-2-34-10-6-2) MTL 110 (lR54742-23-19-16-10-3) MTL 114 (lR54751-2-44-15-2-2)
a

BPH resistance gene None Bph1 bph2 Bph3 bph4 Bph5 bph6 Bph7 bph8 Bph9 Bph1 plus Minor gene(s) bph2, Bph3 O. officinalis O. officinalis O. officinalis

SSST damage scores a 1993 9.0 a def 3.7 5.0 dc 1.0 hi 5.0 dc 7.7 ab 1.7 gh 4.3 de 6.3 bc 4.3 de 0.0 2.3 3.0 3.7 fgh efg def i 1994 9.0 5.0 7.7 2.3 6.3 9.0 3.0 7.7 7.7 5.0 0.0 3.0 5.0 5.7 a c ab bc d 1996 9.0 5.7 7.7 3.7 4.3 9.0 5.7 7.0 7.7 7.7 5.7 0.0 2.3 5.7 3.7 cd ef a ab cd ef de

ab ab

c e

cd bc ab ab cd g f

d c c

The resistance score of varieties containing the O. officinalis gene varied from 2.3 (resistant) to 5.6 (moderately resistant) (see table). We did not find a trend of decreasing resistance over time. However, hopperburn was observed in some farmers' fields planted to these varieties, probably as a result of insecticide overuse. In all 3 yr of testing, varieties with the O. officinalis gene were significantly more resistant to BPH than the test varieties with two other dominant genes, Bph5 and Bph9. This suggests that the O. officinalis gene is distinct from these genes, although minor genes in the O. officinalis -derived varieties could be enhancing their resistance. Interestingly, varieties containing the genes bph4, Bph5, Bph7, bph8, and Bph9 scored susceptible or only moderately resistant to BPH in 2 or 3 of the test years, even though varieties containing these genes are not known to have been grown in the Mekong Delta.

Scores are the means of three replicates. Means within a column followed by the same letter are not significantly different (P>0.05, LSD test).

Erra Mallelu, Kavya, and Orugallu: fine-grained, gall midge (biotype 1)-resistant rice varieties
P. P. Reddy, N. Kulkarni, N. S. Reddy, A. G. Ram, C. P. Rao, T. N. Rao, R. V. Kumar, B. Narendra, A. Sudarshanam, and A. S. Rao, Agricultural Research Station, Andhra Pradesh Agricultural University, Warangal 506007, Andhra Pradesh, India

Characters of fine-grained gall midge (biotype I)-resistant rice varieties.

Cultivars Character Parentage Erra Mallelu Sabarmati/W12708 Kavya WGL 27120/// WGL 17672/ Mahsuri//Surekha 135 Wet, winter (under irrigation) 90-95 10-12 lnsensitive Responsive Absent Compact 24.5 220 Straw 20.5 73 Medium slender 3.86 2.78 Absent 6.5-7.0 Resistant to gall midge biotype 1 Orugallu OBS677/IR2070-423-2-5

Duration (d) Suitable season Height (cm) Panicle-bearing tillers hill -1 (no.) Photoperiod sensitivity Response to fertilizer Anthocyanin pigmentation Plant type Panicle length (cm) Grains panicle-1 (no.) Glume color 1,000-grain weight (g) Head rice recovery (%) Grain type L-B ratio of grain (mm) L-B ratio of kernel (mm) Abdominal white Yield potential (t ha -1) Resistance to pests

120 Wet, winter, summer 80-85 10-12 Insensitive Responsive Absent Semicompact 22.2 125 Light brown 21.0 70 Long slender 4.37 3.61 Absent 6.0-6.5 Resistant to gall midge biotype 1

140 Wet (up to 30 Jun) 85 15-16 Insensitive Responsive Absent Compact and erect 21.7 180 Straw 24.5 68 Long slender 3.74 3.00 Absent 7.0 Resistant to gall midge biotype 1 and tolerant of bacterial leaf blight

Rice gall midge (biotype 1) is a serious pest when rice is planted late because of delayed rains and late filling of tanks, or planted in the tailend areas of canals in the Telangana zone of Andhra Pradesh, India. Rice varieties Erra Mallelu (1991), Kavya (1991), and Orugallu (1993) were released to control gall midge in this area. Erra Mallelu (UGL 20471) is a shortduration rice variety (see table) that outperformed popular Tellahamsa,

Vol. 22. No. 1 27

which was susceptible to gall midge and other pests, in station trials (198182) and minikit trials (1988-91) at different sites in Telangana zone. Erra Mallelu had no or negligible incidence of galls compared with checks Jaya and TN1 in trials at numerous locations. The variety is now grown on 50,000 ha in the zone.

Kavya (WGL48684) is a mediumduration variety (see table) that is similar in grain type but 10 d earlier than the locally popular Samba Mahsuri, which is susceptible to gall midge and other pests. Kavya showed no or negligible incidence of galls in many trials at several locations from 1987 to 1989. Its yield performance in multi-

location and national trials from 1987 to 1989 was promising. Kavya is grown on 40,000 ha. Orugallu (WGL47970) is a longduration, high-yielding (4.7 t ha-1) rice variety (see table). It performs well in slightly saline soils and was resistant to gall midge in the 1987-89 trials. It is grown on 20,000 ha.

Stress toleranceother stresses

Flowering behavior of rainfed lowland rice varieties during dry season in West Bengal, India
S. K. Sinha and S. D. Chatterjee, Rice Research Station, Chinsurah, Hooghly, West Bengal 712102, India
Table 1. Mean daylength and minimum and maximum temperature of each month. Rice Research Station, Chinsurah, India.

Month H January February March April May June July August September October November December 10 11 11 12 13 13 13 12 12 11 10 10

Day length Min 43 14 51 22 06 24 11 50 11 31 55 36

Minimum temperature (C) 10.2 13.2 22.9 25.0 25.5 25.7 25.7 25.5 22.9 16.0 10.0

Maximum temperature (C) 25.6 28.7 33.7 36.8 36.4 33.9 31.7 31.3 31.5 31.0 28.8 25.6

18.5

We studied the flowering behavior of five popular rainfed lowland rice cultivars (CN540, CNM539, Mahsuri, Pankaj, and IR42) compared with that of two photoperiod-insensitive boro checks (Jaya and IR26). Seeds were sown on three dates during 1992-93 and 1993-94 dry seasons. Rice is traditionally sown in JuneJuly and harvested in NovemberDecember in West Bengal. Daylength peaks in July (13 h 11 min) (Table 1). Low temperature from November to January limits seedling growth, which extends crop duration. Earlyto medium-duration photoperiod-

insensitive modern boro varieties are popular in this season. We studied the performance of longer duration rainfed lowland rice varieties under these conditions, which had not been done to date. Seeds were sown on 23 Nov, 12 Dec, and 28 Jan and transplanted on 10 Jan, 8 Feb, and 3 Mar, respectively, for two

consecutive dry seasons (1992-93 and 1993-94), with three replications for each sowing. The data were recorded for initial flowering (first panicle in the plot), 50% flowering (half of the panicle in the plot), 100% flowering (every panicle in the plot), and grain yield (t ha-1) pooled over 2 yr (Table 2).

Table 2. Flowering behavior and grain yield of different lowland rice varieties under different sowing dates in West Bengal, India. 1992-94 a.

Sowing date Variety Initial Pankaj IR42 CN540 CNM539 Mahsuri Jaya IR36 CD (variety sowing date) at 0.05 level
aData pooled over 1992-93 and 1993-94.

23 Nov Days to flowering 50% 144 140 160 143 154 140 130 4 100% 147 144 167 145 158 145 135 4 Initial 130 120 130 135 123 116 8

12 Dec Days to flowering 50% 136 125 155 139 127 122 4 100% 140 129 169 143 131 127 4 Initial 145 105 150 137 102 94 8

28 Jan Days to flowering 50% 152 108 162 145 105 98 4 100% 162 110 175 152 108 103 4

Grain yield (t ha-1) 23 Nov 5.3 5.4 4.7 4.8 4.5 5.6 5.3 0.6 12 Dec 4.4 4.8 3.0 3.7 5.1 5.0 0.6 28 Jan 3.8 4.2 2.5 3.3 4.4 4.2 0.6

140 135 147 139 147 135 127 8

28

IRRN

1997

For the November planting, the varieties were generally consistent in duration between initial and 100% flowering, regardless of the magnitude of photoperiod sensitivity (Table 2). The exception was CN540, which took as long as 20 d to finish flowering. Lowland varieties sown after November, however, had different flowering patterns and the variety sowing date interaction was significant for duration to initial flowering, 50%) flowering, and 100% flowering. CNM539 did not flower at all.

While time to flowering gradually decreased for IR42, Jaya, and IR36 in successive sowings, duration of Pankaj, CN540, and Mahsuri decreased from November to December but increased again in January. Though all other varieties flowered, none, except IR72, retained their flowering synchrony (a short interval between initial and 100% flowering) that simplifies harvesting. IR42 appeared to be the best variety with synchronous flowering behavior under all the sowing dates.

Flowering during mid- to late June for varieties such as Pankaj, Mahsuri, and CN540 is not conducive for higher grain yield. Long vegetative growth exposes the crop to the beginning of the southwest monsoon, often leading to lodging followed by seed sprouting. A good harvest with these varieties cannot be assured. On the other hand, the earliest variety, IR42, yielded comparably with Jaya and IR36. The advantage of IR42 is its long slender grain, which receives a premium in the market.

Integrated germplasm improvementirrigated

Vijetha: a high-yielding, shortduration rice variety for Andhra Pradesh, India
Y. Suryanarayana, P. Sankara Rao, N. Sree Rama Reddi, K. R. K. Murthy, and P. S. S. Murty, Andhra Pradesh Agricultural University, Agricultural Research Station (ARS), Maruteru 534122, West Godavari District, Andhra Pradesh, India

Vijetha was developed from a cross between MTU5249 and MTU7014, using the pedigree breeding method at ARS. Vijetha was highly promising in research station trials conducted during the 1991-94 dry seasons (Table 1). It yielded a mean 7.5 t ha-1 and consistently produced at least 1 t ha-1 more than check IR64an average yield advantage of 22.8%. In the All India Coordinated Trials during 1993 dry season, Vijetha re-

corded a mean yield of 4.9 t ha-1 (Table 2) across 15 locations, indicating its wide adaptability. At Maruteru, we used the seedbox technique to determine that the variety is resistant to brown planthopper (BPH), scoring a 3 on the Standard evaluation system for rice scale. Vijetha is also moderately resistant to gall midge (biotype 1), BPH, whitebacked planthopper, stem borer, leaffolder, and blast. Vijetha is a semidwarf (105 cm) with dark green foliage and a broad, erect flag leaf. It possesses 8-10 productive tillers plant -1. Flowering is completed in 3-4 d, contributing to uniform maturity. The panicle length is 26 cm and grains panicle-1 total about 150; 1,000-grain weight is 24 g. The grain is mediumshaped and translucent, with a lengthbreadth ratio of 2.6 (length is 6.2 cm, breadth 2.4 cm). Dormancy ranges from 6 to 8 wk during kharif (dry

season). Milling percentage is 65.8%, with a 62.4% head rice recovery, and 5.0% broken rice percentage, classifying it as a fine-grained rice. Results from 530 minikit trials conducted during 1994 and 1995 confirmed the superiority of Vijetha over check IR64 across Andhra Pradesh. Vijetha can be grown in either the wet season (145 d) or dry season (125 d). It is a nonlodging, fertilizer responsive variety with resistance to several pests. Even before 3yr of minikit testing was completed, Vijetha was planted on more than 150,000 ha in Andhra Pradesh. Because of its outstanding potential, the State Seed Subcommittee released Vijetha in 1995.

Table 2. Performance of Vijetha at different locations in India. 1993 kharif.

Yield (t ha -1) a State Locations (no.) Vijetha Ratna (check) 2.9 4.8 5.4 4.6 0.8 3.8 3.4 4.0

Table 1. Yield performance of Vijetha at Agricultural Research Station, Maruteru, India. 1991-94 dry seasons.

Grain yield (t ha-1) Year Vijetha 7.5 7.9 7.7 6.8 7.5 IR64 5.6 6.6 6.2 6.0 6.1

% increase over IR64

CV (%)

LSD (5%)

1991 1992 1993 1994 Mean

33.9 19.7 24.2 13.3 22.8

11.3 7.2 13.0 7.5

0.9 0.7 1.0 0.6

Kerala Tamil Nadu Karnataka Andhra Pradesh Gujarat Haryana Uttar Pradesh Meanb
a

3 3 2 3 1 1 2

5.6 6.5 4.8 4.6 3.5 4.9 2.2 4.9

Mean of the locations in the state. b Mean of 15 locations.

Vol. 22, No.1 29

Two rice hybrids released in Andhra Pradesh, India

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

Mean yield performance of MTU HR 2003 (APHR-1) and MTU HR 2008 (APHR-2) in minikit trials. Andhra Pradesh, India. 1992 and 1993 wet seasons.

District

Locations (no.)

Mean yield of hybrid (t ha -1 ) 7.0 3.6 7.8 7.5 4.1 5.3 5.9

Mean yield of best check (t ha-1 ) 6.2 2.8 6.4 6.0 3.7 4.0 4.8

Yield advantage over the best check (t ha-1 )

Since 1988, we have been working to develop rice hybrids by using effective local restorer lines. The five top performers (MTU HR 2000, MTU HR 2001, MTU HR 2002, MTU HR 2003, and MTU HR 2008) were tested in farmers fields throughout Andhra Pradesh, India, during the 1992 and 1993 wet seasons. Each hybrid was grown in a 5001000-m2 plot along with a popular local cultivar as a check. Plants were transplanted at 1-2 seedlings hill -1, at a population of 40-45 hills m -2. Local cultural practices were followed. Of the five hybrids, MTU HR 2003 and MTU HR 2008 performed exceedingly well across locations (see table). In 1992, MTU HR 2003 produced a 1.1 t ha-1 mean yield advantage over the best check (4.8 t ha-1) across 11 locations. In 1993, it produced a 1.3 t ha-1 mean yield advantage over the best check (5.8 t ha-1) in six out of seven districts. MTU HR 2008 performed well at three locations during 1992, with a mean yield advantage of more than 3 t ha-1 over the check. During 1993, the same hybrid outperformed the best check (5.7 t ha-1) with a mean yield advantage of 1.3 t ha-1. The hybrid produced maximum yields of 1011 t ha-1. The Andhra Pradesh Agricultural University released MTU HR 2003 as APHR 1 and MTU HR 2008 as APHR 2 in December 1993. APHR 1 (IR58025 A/Vajram R) is a medium-duration (130-135 d) hybrid, while APHR 2 (IR62829 A / MTU 9992 R) is short in duration (120 d). Both are tolerant of lodging, intermediate in height, and have long slender grains with good cooking quality. APHR 2 performs well under low inputs and possesses tolerance for bacterial leaf blight.

MTU HR 2003 (1992 wet season) Warangal Ranga Reddy Kurnool Chittoor East Godavari West Godavari Mean MTU HR 2003 (1993 wet season) Nalgonda Ranga Reddy Waragal Kurnool Krishna Guntur West Godavari Mean MTU HR 2008 (1992 wet season) Ranga Reddy Krishna Anantapur Mean MTU JHR 2008 (1993 wet season) Nalgonda Ranga Reddy Warangal Kurnool Krishna Guntur West Godavari Mean

3 2 2 1 2 1 11

0.8 0.9 1.3 1.5 0.4 1.4 1.1

2 3 3 5 5 1 3 22 1 1 1 3

9.3 6.3 7.5 8.1 7.3 5.5 4.7 7.1 10.0 9.4 10.4 9.9

6.6 5.2 6.1 7.0 5.4 3.9 4.8 5.8 6.5 6.6 7.5 6.9

2.7 1.0 1.4 1.0 1.9 1.5 0.2 1.3

3.5 2.8 2.9 3.1

2 3 3 6 6 3 4 27

11.0 6.4 6.8 8.1 6.8 6.3 5.3 7.1

6.6 5.7 5.4 6.9 5.2 5.9 4.8 6.7

4.4 1.2 1.4 1.2 1.5 0.4 0.5 1.3

Performance of hybrid rice in irrigated lowlands, Uttar Pradesh, India

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

Yields of hybrid PRH-1 (PMS 2A/IR31802) and highyielding inbred rice varieties at two N rates. Pantnagar, India, 1994-95 wet seasons.

Treatment N rate (kg ha -1) 120 160 SE CD(5%) Varieties PRH-1 Jaya Pant Dhan 4 Pant Dhan 10 Pant Dhan 12 S E CD(5%) lnteraction

Grain yield (t ha -1) 1994 6.4 7.0 0.1 ns 7.7 6.7 6.2 6.3 0.1 0.3 ns 1995 6.3 6.6 0.1 ns 7.1 6.4 6.0 6.3 0.1 0.3 ns 4 Mean 6.4 6.8

In the Tarai region of northwestern India, irrigated rice yields are high, with many farmers harvesting 7 t ha-1 or more. Hybrid rice varieties, with their higher yield potential, should have a place in these areas. We evaluated the performance of newly developed hybrid PRH-1 (PMS2A/IR31802) compared with that of high-yielding inbred varieties (Jaya,

7.4 6.5 6.2 6.0 6.3

CV (%) based on error b 4

30 IRRN 1997

Pant Dhan 4, Pant Dhan 10, and Pant Dhan 12) at two N fertility levels. The experiment was conducted in the 1994 and 1995 wet seasons at Pantnagar, where the soil is a silt loam (Aquic Hapludoll) with pH 7.9, 1.1% organic C, 0.1 % total N, and CEC 20 meq 100 g -1 soil. The experiment was laid out in a split-plot design with three replications. N level (120 and 160 kg N ha-1) was in the main plot and varieties (PRH-1, Jaya, Pant Dhan 4, Pant Dhan 10, and Pant Dhan 12) in the subplots. A single basal dose of 17.5 kg P ha-1, 33.2 kg K ha-1, and 10 kg Zn ha-1 was uniformly applied in all plots. N was applied as per treatment in three splits: 1/2 as basal, 1/4 topdressed at active tillering, and 1/4 topdressed at panicle initiation. Twenty-five-day-old seedlings were transplanted on 7 Jul in both years at 20- 20-cm spacing with 2-3 seedlings hill-1. Varieties did not respond differently to N. An additional dose of 40 kg N (bringing total N to 160 kg ha-1) helped plants yield 7% more grain than at 120 kg N ha-1 the reccomended dose for high-yielding inbreds. The difference, however, was not significant (see table). Hybrid rice PRH-1 yielded significantly more (17% in 1994 and 13% in 1995) than the best-performing inbreds: Jaya (12% less), Pant Dhan 4 (17% less), Pant Dhan 10 (19% less), and Pant Dhan 12 (15% less). Based on these findings, suitable hybrids can produce 15-20% more than todays best inbred varieties in intensive input systems.
Routine research. Reports of screening trials of varieties, fertilizer, cropping methods, and other routine observations using standard methodologies to estab lish 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.

FKR42 and FKR44: irrigated rice varieties released to farmers in Burkina Faso
M. Sie, Y. Sere, and A. Sanou, Institut d' etudes et de recherche agricoles (INERA), station de Farako-B, 01 BP 910, Bobo-Dioulasso, 01 Burkina Faso, West Africa

Rice varieties IR64 and IR13240-108-22-3 were introduced into INERA's breeding program through collaboration with the International Network for

Genetic Evaluation of Rice and the West Africa Rice Development Association. Both varieties proved promising in advanced yield trials in 1992-95 (see table). They also performed well in adaptive on-farm trials under farmers' cultural practices. The varieties are early-maturing with long, slender grains and are suited for irrigated rice - rice double cropping. Both were released in 1995: IR64 as FKR42 (Farako-B Riz) and IR13240108-2-2-3 as FKR44.

Grain yield (t ha -1) in irrigated rice variety trialsa at the INERA research stations in the Kou and Sourou valleys. Burkina Faso. 1992-95 wet (WS) and dry seasons (DS).

Kou Valley Variety 1992 WS 4.3 4.1 4.1 1993 DS 4.3 2.1 3.1 1994 WS 4.0 4.7 4.3 4.4 1995 DS 2.4 2.8 3.0 2.9

Sourou Valley 1994 WS 6.1 3.2 5.4 1995 DS 6.9 7.7 7.7 6.6

IR64 (FKR42) lR13240-108-2-2-3 (FKR44) ITA123 (FKR28) 4456 (FKR16)

aFertilizer rate = 88-69-42 kg NPK ha -1.

Integrated germplasm improvement upland

Evaluating local germplasm for the upland rice ecosystem in western Nepal
K. D. Joshi, Local Initiatives for Biodiversity, Research and Development, Pokhara, Nepal; B. R. Sthapit, Nepal Agriculture Research Council, Khumaltar, Nepal

Upland rice, locally known as ghaiya dhan in Nepal, is grown as the major crop in the tars (unirrigated ancient river fans), on terraced lands, and on hills where forests have recently been cleared. Most of the 126,000 ha (9% of the country's total rice area) under ghaiya is hilly. Resource-poor upland farmers from ethnic groups such as the Kumal, Derai, and Bote are the main growers of this rice. Of the 42 rice varieties released in Nepal, only Ghaiya 2 (MV10) is suitable for subtropical upland areas. Hill farmers, however, do not like it because it is

short and does not perform well under fertility. Indigenous ghaiya varieties have not been properly studied or used in variety development programs that could benefit large farming communities. Researchers at the Lumle Agricultural Research Center have done some preliminary work on evaluating local germplasm. They screened 53 ghaiya land races from Gorkha and Tanahun districts and evaluated them in replicated randomized complete block designs in a target environment (Chyanglitar, 400 m) for two seasons. The crop was broadcast and grown using farmers practices, except for maintaining a row spacing of 15 cm between the entries. Farmers were involved in ranking the varieties at maturity. Ghaiya varieties that performed poorly during the first year were dropped. Only 24 varieties were tested during the second season.

Vol. 22, No. 1

31

Table 1. Variability in 53 Ghaiya varieties for different traits.

Table 2. Performance of Ghaiya varieties at Chyanglitar, Nepal (400 m). 1993-94.

1993 Variety Plant height (cm) 148 139 141 132 135 137 132 136 142 138 129 139 137 140 154 139 134 140 143 127 140 125 138 77 133 10.63 Days to maturity 116 116 116 113 120 112 114 117 112 113 111 115 116 114 120 115 116 113 114 114 110 121 113 125 115 4.31 Yield (t ha -1) 3.1 3.1 3.5 3.6 3.7 3.5 3.4 3.4 3.4 3.5 3.2 3.4 3.6 3.8 4.1 3.5 3.3 4.2 2.2 2.4 2.8 1.5 4.2 2.6 2.89 1.16 Plant height (cm) 157 114 143 142 155 143 133 144 149 146 144 152 148 139 161 149 140 150 142 139 139 144 146 85 142.9 9.64

Character Leaf length (cm) Leaf width (cm) Leaves tallest tiller -1 (no.) Internode length (cm) Culms (no.) Plant height (cm) Panicle length (cm) Grains tallest tillers-1 (no.)

Range 16-85 1-2.6 4-8 29-47

Mean 50.2 1.61 5.58 38.57

SD 9.77 0.25 0.85 3.0 2.39 11.28 1.49 50.97

SE 0.42 0.01 2.22 0.29 0.10 1.10 0.15 0.04

1994 Days to maturity 124 123 122 120 123 122 118 122 121 123 121 124 118 119 125 120 121 119 121 120 119 118 122 117 120.9 2.94 Grain yield (t ha -1) 4.1 4.5 5.4 3.6 3.6 4.4 3.9 4.5 4.0 3.4 4.3 4.5 4.2 4.4 3.3 3.8 3.1 4.2 4.6 3.9 4.4 3.4 4.5 4.9 4.1 1.09 Phenotypic acceptance 1-9 3 4 3 4 4 3 4 3 3 4 3 3 4 3 2 4 3 3 2 4 1 2 2 3 3.08 -

4.99 1-14 80-156 133.0 24.59 20.3-28 3-332 110.2

Indigenous ghaiya varieties have a large leaf area, which may have important implications for drought tolerance. Because of its tall, weak architecture, the plant commonly lodges under high fertility. The varieties showed variability in traits contributing to yield potential (such as panicle length, amount of grain, and leaf number on the tallest tiller) and other traits (such as leaf characteristics and internode length) that are important for varietal improvement programs (Table 1). The study revealed that there was a lot of variation for grain yield. The majority of indigenous ghaiya varieties were at par with semidwarf rice variety Ghaiya 2 for grain yield (Table 2). At maturity, local farmers helped assess the phenotypic acceptance of the rice entries. They rated the majority of entries as good to excellent, with most of the entries having plant height and maturity periods suitable for local conditions. The study identified variability between and within different genotypes that can be used both for developing varieties through pureline selection and in crossing programs. Although farmers are only growing these land races in a few places, more could be benefiting from them through participatory variety selection and promotion of them in similar ghaiya-growing areas. To obtain varieties that are useful for farmers, indigenous knowledge and genetic resources must be tapped to develop high-yielding, suitable ghaiya varieties.

Kalo Basaune Kalo Dhan Sindure Ghaiya Pahenle Ghaiya Seto Ghaiya Bichare Chobo Rato Ghaiya Phalame Chobo Sano Thantar Nani Maiya Pakhe Masino Phalame Chobo Nani Dhan Langre Jabaka Begani Dudhe Tauli Dare Ghaiya Chiuri Jhyalebicharo Jire Ghaiya Linde Basaha Ghaiya 2 (check) Mean LSD (0.05)

FKR33: a popular upland variety in Burkina Faso

M. Sie, Y. Sere, and A. Sanou, Institut d' etudes et de recherche agricoles (INERA), station de Farako-B, 01 BP 910, Bobo-Dioulasso, 01 Burkina Faso, West Africa

INERA scientists are developing new promising upland rice varieties from F 4 generations of pedigree lines screened at the Institut des Savanes research station in Bouak, Cte d'Ivoire. The release of IRAT144 as FKR5 (Farako-B Riz) in Burkina Faso was a result of this collaboration, but the variety has poor grain quality. In 1982, we developed the line 1119-

5-2 by pedigree selection from the cross IRAT 112/IRAT 13. This variety was released as FKR33 in 1992. FKR33 matures slightly earlier and is shorter than FKR5. Their grains are of the same length (9.8 mm), but FKR33 grains are more slender (2.8 mm) than that of FKR5 (3.0 mm). Superior grain type and good head rice recovery are FKR33s best qualities. FKR33 produced 35% more grain than FKR5 in farmers fields with less than 500 mm of rainfall. In trials conducted in 25 farmers fields in Como Province in southwest Burkina Faso, FKR33's average yield was 3 t ha-1. In Farako-B its yield potential was 6 t ha-1 during the wet season (see table).
Panicles m-2 (no.) 144 228 268 128
b

Agronomic characteristics of upland rice varietiesa , Farako-B Research Station, Burkina Faso. 1992.

Variety FKR5 (IRAT144) FKR33 (1195-5-2) IREM194 IDSA13

a

Plant height (cm) 152 124 162 134

Maturity (d) 107 102 105 105

Leaf blast b 1 1 3 1

Neck blast b 1 1 1 1

Grain yield (t ha -1) 5.9 6.2 5.9 5.5

Sowing date: 20 Jun 1992: fertilizer rate: 76-46-28 kg NPK ha -1. IRRI 1988.

Scored using the Standard evaluation system for rice.

32 IRRN 1997

Seed technology
Revitalizing stored rice seeds
K. Sivasubramaniam, Seed Technology Department, Agricultural College and Research Institute, Madurai 625104, India

Seeds of rice varieties ADT36, JJ92, ASD16, ASD38, ADT39, Co43, IR20, MDU2, MDU4, Ponni, and White Ponni stored under ambient conditions (75% relative humidity, 25-35C, 12% moisture content) for 28 mo were subjected to hydration-dehydration treatment by soaking in water (1:2v/v) for 6 h. The seeds were then dried to their previous moisture content and stored for another 6 mo.

We evaluated them every month after that for germination rate and dry matter production of seedlings (mg seedling-1) to determine the rejuvenation ability of these varieties. All showed significant improvement in germination rate with the treatment compared with untreated controls. ADT38 recorded the highest recovery, with a difference of 34% over the untreated seed. Untreated seeds of all the varieties were nonviable at 31 mo, while treated seeds showed some degree of viability even after 33 mo (Table 1). Hydration of stored seeds has been hypothesized to quench the free radi-

cals, thereby arresting membrane damage and promoting repair mechanisms. The effect of hydration-dehydration nearly doubled the germination rate and slowed the rate of decline in viability. The effectiveness of the treatment, however, varied among varieties. This method can be used to reenergize seeds showing declined viability after being stored for some time. The increase in seedling dry matter of hydrated-dehydrated seeds (Table 2) also shows the re-energization of cells, leading to improved overall synthesis of metabolites for growth and development.

Table 1. Response of rice varieties (germination percentage) to hydration-dehydration treatment. Months Variety 28 ADT36 JJ92 ASD16 ADT38 ADT39 Co 43 IR20 MDU2 MDU4 Ponni White Ponni LSD (0.05) 35 31 60 49 57 61 47 56 51 40 44 9.07 29 36 23 49 35 51 56 29 36 40 35 24 9.67 30 23 13 36 23 39 47 28 28 35 21 19 9.04 Hydrated-dehydrated (treated) 31 15 8 26 21 31 24 19 21 25 13 13 6.17 32 11 0 12 12 16 14 12 11 16 8 8 4.05 33 3 0 0 3 4 4 3 1 0 0 0 1.58 Mean 20.5 12.5 30.5 23.8 33.0 34.3 23.0 25.5 27.8 19.5 18.0 28 11 19 39 15 13 33 24 33 19 19 28 6.12 29 8 11 21 16 14 20 16 24 11 5 11 4.21 30 5 4 9 7 8 9 0 11 4 4 4 2.18 Months Control (untreated) 31 0 0 4 0 0 0 0 0 0 0 0 0 32 0 0 0 0 0 0 0 0 0 0 0 0 33 0 0 0 0 0 0 0 0 0 0 0 0 Mean 2.17 5.67 12.20 6.33 5.83 10.30 6.67 11.30 4.67 7.17

Table 2. Response of rice varieties (mg dry matter seedling -1) to hydration-dehydration treatment. Months Variety Hydrated-dehydrated (treated) 28 ADT36 JJ92 ASD16 ADT38 ADT39 Co 43 lR20 MDU2 MDU4 Ponni White Ponni LSD (0.05) 17.17 19.42 19.18 23.54 18.04 15.51 15.15 26.33 20.15 21.67 20.17 3.01 29 14.60 18.61 17.16 21.88 17.63 14.36 13.97 24.32 19.89 19.72 19.59 2.93 30 14.58 16.13 15.78 21.64 14.63 13.79 13.20 21.20 16.45 18.33 16.70 2.68 31 13.92 11.39 13.47 17.42 12.92 12.95 10.10 20.75 15.04 17.22 12.08 2.80 32 10.91 12.63 0.0 11.63 16.88 9.68 10.12 10.10 9.56 0.0 0.0 5.56 33 1.58 10.16 0.0 0.0 14.11 8.62 5.18 5.36 8.18 0.0 0.0 4.58 Mean 13.84 10.93 12.87 19.24 13.58 11.98 11.99 18.39 11.92 12.83 11.42 28 2.62 6.83 4.43 10.0 3.36 4.00 2.33 6.00 9.32 12.75 12.62 3.49 29 1.12 3.62 3.18 8.62 3.00 3.68 2.00 5.72 8.18 5.28 8.58 2.41 30 1.06 2.11 1.18 5.11 2.52 3.00 0.0 5.13 6.63 4.15 3.16 1.82 Months Control (untreated) 31 0.0 0.0 0.0 2.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 32 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 33 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Mean 1.60 2.09 1.47 4.32 1.48 1.78 0.72 2.82 4.02 3.69 4.06

Vol. 22. No. 1 33

Crop and resource management

Physiology and plant nutrition
Effect of hydroquinone and phenylhydrazine on yield and nitrogen use efficiency of rice
H. Pathak, J. Singh, and K. Sarkar, Environmental Sciences Division, Indian Agricultural Research Institute (IARI), New Delhi 110012, India

This field experiment evaluated the N use efficiency (NUE) of using two urease inhibitors, hydroquinone (HQ) and 2,4-dinitrophenylhydrazine (PH). Rice (cv Pusa 169) was grown in the wet season (June to November) of 1994. The soil was an Ustochrept clay loam with 0.45% organic C, 667 mg total N kg-1, 232 kg available N ha-1, 37 kg available P ha-1, 291 kg available K ha-1 , pH 8.2, EC 0.48 dS m-l. The table shows the treatments in a completely randomized block design with three replications in 5- 4-m plots. Urea, at the rate of 120 kg N ha-1 was applied in three , splits: one-half at the time of transplanting, one-fourth 30 d after transplanting (DT), and one-fourth 50 DT.

The plots were basally dressed with P and Kat the rate of 26.4 and 49.8 kg , ha-1 respectively, and irrigated to maintain 5 cm of standing water. Residual effect was studied by growing wheat (cvHD2285) after rice harvest in the dry (December to April) season of 199495. Abooster dose of 50 kg N ha-1 was applied 40 d after sowing in all the plots. The experiment was repeated in the same site during 1995-96. Adding either HQ or PH with urea increased yield of rice more than using urea alone in both the years (see table). In 1994-95, the highest yield (6.89 t ha -1) was recorded with urea plus 10%) PH, and the next highest yield (6.55 t ha-1) with urea plus 10% HQ. Harvest index showed no significant difference. Application of PH at 5 and 10% and HQ at 10% with urea resulted in a higher uptake of N than with application of urea alone. The NUE increased when HQ and PH were added. With urea alone, the efficiency was 14.75 kg of grain kg-1 N. It increased to 18.08 and

25.58 kg of grain kg-1 N with the application of 5 and 10% of HQ and 25.25 and 28.42 kg of grain kg-1 with 5 and 10% PH. Application of 10% of HQ and PH increased the N recovery by almost 60 and 100%, respectively. Total N content of the soil after harvest showed no significant difference among treatments. Residual effect of various treatments was also not significant. During 1995-96, NUE and N recovery were less than in 1994-95. Higher NUE and N recovery found in both the years with the use of HQ were attributed to reduced volatilization loss of ammonia. With the use of PH, higher NUE was attributed to increased mineralization of immobilized fertilizer N. Although these inhibitors increased yield, NUE, and N recovery, the benefit-cost ratio (B/C) was poor. In both years, B/C was less than 1.0, indicating net economic loss. At the present price of HQ and PH, use of these chemicals in rice cannot be recommended.

Yield, N uptake, N use efficiency, and N recovery of rice, total N content of soil after rice, grain yield of wheat, and benefit-cost ratio (B/C) of using hydroquinone (HQ) and dinitrophenylhydrazine (PH). New Delhi. 199495.

Yield (t ha -1) Rice grain 1994 3.5 5.3 5.7 6.6 6.5 6.9 1.2 8.5 1995 3.8 5.1 5.8 6.7 6.4 6.8 1.1 9.2 Rice straw 1994 6.1 7.6 8.0 7.7 7.5 7.5 1.5 9.7 1995 6.2 7.6 7.9 7.7 7.5 7.9 1.2 8.8 Wheat grain 1994 3.6 4.5 4.7 4.9 4.5 4.6 6.6 1995 3.8 4.7 4.8 4.9 4.8 4.9 ns 7.8 Harvest index 1994 0.36 0.41 0.41 0.46 0.47 0.48 0.09 7.4 1995 0.38 0.40 0.43 0.47 0.46 0.46 0.07 6.2 N uptake (kg ha -1 ) 1994 58.95 100.15 111.13 126.96 123.27 140.36 12.84 8.7 1995 66.32 98.86 108.46 122.58 128.12 136.64 13.46 9.4 Use efficiency 1994 14.75 18.08 25.58 25.25 28.42 1995 11.08 17.08 24.50 22.08 24.75

N (kg kg -1) Recovery 1994 34.33 43.48 56.67 53.60 67.84 1995 27.12 35.12 46.88 46.88 58.60 Content in soil 1994 539.1 684.0 633.1 664.5 601.7 644.4 ns 9.5 1995 552.6 646.2 654.3 686.4 628.5 638.4 ns 8.6 B/C 1994 0.54 0.80 0.46 0.32 1995 0.79 0.90 0.50 0.34

Treatment

Control Urea U + 5% HQ U + 10% HQ U + 5% PH U + 10% PH CD (0.05) CV (%)

U = urea, B/C = benefit-cost ratio (rice plus wheat), cost of PH = US$25 kg -1, cost of HQ = US$9.4 kg -1, price of rice = US$117 t-1, price of wheat = US$125 t -1.

34

IRRN

1997

Fertilizer management
Integrated nutrient management in a rice-based crop sequence
C. R. S. Devi, G. K. B. Nair, K. K. Sulochana, and V. R. Nair, Cropping Systems Research Centre, Karamana, Trivandrum 695002, Kerala, India

yield. Use of organic sources along with inorganic sources enhanced yield. Using an equal proportion of each

source of nutrients would be optimum for producing higher and sustained yield over time.

Effect of integrated nutrient supply on grain yield of rice (t ha -1). 1987-93.

Treatment Kharif No NPK fertilizers No organic manures (Control) NPK in the ratio 45:22.5:22.5 kg ha -1 through fertilizers NPK in the ratio 45:22.5:22.5 kg ha -1 through fertilizers NPK in the ratio 67.5:33.75:33.75 kg ha-1through fertilizers NPK in the ratio 90:45:45 kg ha-1 through fertilizers NPK in the ratio 45:22.5:22.5 kg ha -1 through fertilizers + 45 kg N ha-1 throughFYM NPK in the ratio 67.5:33.75:33.75 kg ha -1 through fertilizers + 22.5 kg ha -1 throughFYM NPK in the ratio 45:22.5:22.5 kg ha -1 through fertilizers + 45 kg N ha-1through crop residues NPK in the ratio 67.5:33.75:33.75 kg ha-1through fertilizers + 22.5 kg N ha -1through crop residues NPK in the ratio 45:22.5:22.5 kg ha -1 through fertilizers + 45 kg N ha-1 through green manuring Rabi Same as in Kharif 1987 6.6 1988 4.9 1989 5.3 1990 4.9 1991 6.5 1992 5.4 1993 6.8

This field experiment was designed to develop a system for integrated nutrient supply for a rice - rice cropping sequence from 1987 to 1993. It was done in an irrigated double-cropped lateritic alluvium (Inceptisol) acidic sandy loam whose mean values were 0.72% organic C, 156 kg available N ha-1, 31 kg available P ha-1 , and 198 kg available K ha-1. A randomized block design was used with 4 replications, 12 treatments, and a medium-duration variety, Jaya. Different combinations of organic and inorganic sources were used during the first and second crop seasons as well as controls without fertilizer or organic manure (see table). Application of 50% NPK through farmyard manure (FYM) during the first season followed by 100% NPK through fertilizers during the second (treatment 6) significantly increased grain yield in all the years of study except 1993. During 1993, application of 50% NPK through green manure instead of FYM resulted in the highest
Manuscript preparation. Arrange the note as a brief statement of research objectives, a short description of project design, and a succint discussion of results. Relate results to the objectives. Do not include abstracts. Do not cite references or include a bibliography. Restrain acknowledgments. Manuscripts must be in English. Limit each note to no more than two pages of double-spaced typewritten text. Submit the original manuscript and a duplicate, each with a clear copy of all tables and figures. Authors should retain a copy of the note and of all tables and figures.

Same as in kharif

8.2

5.4

6.2

5.9

7.0

5.8

7.0

NPK in the ratio 90:45:45 kg ha-1 through fertilizers Same as in kharif

8.9

5.5

6.5

5.9

6.7

6.1

7.0

8.8

5.5

6.4

6.2

6.7

5.7

7.1

Same as in kharif

8.7

5.8

6.3

6.0

7.4

6.1

7.2

NPK in the ratio 90:45:45 kg ha-1 through fertilizers

9.4

6.1

7.4

6.5

8.6

7.5

7.4

NPK in the ratio 67.5:33.75:33.75 kg ha-1 through fertilizers

8.8

5.7

6.2

6.5

7.9

6.6

7.3

NPK in the ratio 95:45:45 kg ha-1 through fertilizers

8.6

5.8

6.1

5.8

6.8

6.3

7.6

NPK in the ratio 67.5:33.75:33.75 kg ha -1 through fertilizers

8.8

5.5

6.0

5.9

6.9

5.8

7.2

NPK in the ratio 90:45:45 kg ha-1 through fertilizers

8.9

6.0

6.6

6.1

7.5

6.4

7.6

continued on next page

Vol. 22, No. 1

35

Table continued. Treatment Kharif NPK in the ratio 67.5:33.75:33.75 kg ha -1 through fertilizers + 22.5 kg N ha -1 through green manuring Farmers' practice (NPK in the ratio 90:22.5:45 kg ha -1 through fertilizers; no organic manure) CD (0.05) Rabi NPK in the ratio 67.5:33.75:33.75 kg ha -1 through fertilizers 1987 8.3 1988 5.9 1989 6.2 1990 5.9 1991 7.5 1992 6.2 1993 7.3

Same as in kharif

7.4

4.6

5.3

6.0

6.9

6.0

6.3

0.7

0.6

0.6

0.7

1.0

0.8

0.8

Transplanting geometry improves timing of uptake of deep point-placed P by rice hills

N. K. Savant, R. G. Menon, and D. K. Friesen, International Fertilizer Development Center, Muscle Shoals, Alabama 35662, USA

This greenhouse experiment tested the hypothesis that establishing a shorter distance between deep point-placed water-soluble phosphorus (P) and transplanted rice hills may help rice plant roots to intercept and recover fertilizer Pearlier. Urea briquettes (UB) containing diammonium phosphate tagged with 32P (UB-DAP) were deepplaced after transplanting rice (Fig. 1). The soil used was Typic Ustochrept (pH 8.2; organic matter, 11 g kg -1; CEC, 37 cmol kg -1; and Olsen P, 11 mg kg-1). The soil was flooded with water for 2 wk and then puddled prior to transplanting. Nitrogen and P were applied at 58 kg N ha-1 and 12 kg P ha-1. Estimated distances between the deep point-placed 32P and the rice hills were about 14.1 and 10.6 cm for regular and modified spacing. As a standard for comparison, 32P-tagged triple superphosphate (TSP) with a specific activity 74 MBq 32P g -1 P was broadcast and incorporated before transplanting, and N was applied as prilled urea in two splits

(two-thirds basally broadcast and incorporated before transplanting, and one-third topdressed at the panicle initiation stage). Four hills with two 3-wkold-rice (IR36) seedlings hill -1 were transplanted in each box and treatments were replicated three times under continuously flooded soil conditions.

Fertilizer P was calculated from 32P activity in 7-ml aliquots of floodwater whose volume was estimated from its depth (measured before sampling) and surface area of soil in a box. Interception and absorption of 32P from basally applied TSP and deep-placed UB-DAP were determined in situ by monitoring 32 P activity on a well-shielded 1-cm 2 area of the youngest fully developed leaf. The results suggest that basally broadcast and incorporated soluble P fertilizers are prone to surface runoff losses under rainfed, lowland field conditions. Other experiments show practically no 32P in the floodwater when UB-DAP is deep-placed. For the mcorporated TSP, P uptake reached near-plateau 15 d after transplanting (DT) (Fig. 2). For the deep point-placed Pin regular spacing, radioactivity in the youngest leaves was not detected until about 30 and 20 DT. Thus P diffusion alone is not sufficient for onset of its uptake by rice roots. On the other hand, the onset of P uptake in the youngest leaves was observed about 10 d earlier for the modified spacing (Fig. 2). These and other results suggest that modified spacing can reduce the lag period of spatial nonavailability of deep-placed UB-DAP and help to explain improved agronomic performance reported in rainfed transplanted rice on farmers fields.

1. Schematic sketches of regular and modified 20- 20-cm spacings (with 25 hills m-2) used in the boxes.

2. In situ 32P activity in youngest leaves of transplanted rice (IR36) as influenced by basal incorporation of triple superphosphate (TSP) and deep placement of urea briquettes containing diammonium phosphate (UB-DAP) in regular (R) and modified (M) 20- 20-cm spacings.

36 IRRN 1997

Fertilizer management inorganic sources

Effect of Zn on grain yield and Zn uptake by lowland rice in South Gujarat
B. S. Trivedi, R. C. Gami, and G. G. Patel, Research Project on Soil Science, GAU-Navsari Campus, Navsari 396450, India

One-fourth of South Gujarat rice soils have Zn deficiency (less than 0.60 mg kg-1 DTPA-extractable Zn). Another 39%, classified as medium-deficient (0.6-1.0 mg kg-1 DTPA- extractable Zn), will probably worsen from continuous waterlogging of ricefields. Presently, a blanket application of 2 kg Zn ha-1 is recommended but considered insufficient for low Zn soils. We examined the effect of 0, 5.6, 11.2, and 16.8 kg Zn ha-1 on Jaya (coarse grain variety) and GR.11 (fine grain variety) in field experiments during 1992 and 1993 monsoon seasons. The experiments were conducted in a randomized block design with four replications in plots of 6.0- 3.0- m on Vertic Ustochrept soils. Chemical analysis showed that the soil was low in DTPAextractable Zn (0.60 mg kg -1) and alkaline KMnO4 -extractable N (230 kg ha-1), medium in Olsen P (12.4 kg ha-1) and high in neutral N NH 4 O Ac-extractable K (283 kg ha -1) with a pH of 7.9 without problems of salinity or sodicity. Nitrogen was applied at 100 kg ha-1 and P at 22 kg ha-1. The entire amount of Zn (as ZnSO4. 7H2 O), P, and 40 kg N were applied before puddling. The remaining N was applied in two equal splits each after 20 and 45 d of transplanting. The sources of N and P were diammonium phosphate (DAP) and urea. The crop was irrigated during drought periods. At harvest, diacid extracts of grain and straw samples were analyzed for Zn using an AA spectrophotometer. Pooled analysis on grain yield and total Zn uptake data indicated nonsignificant first- and second-order inter-

where X( e ) = dose of Zn for economic yield, q = cost of Zn at Rs 54.76 kg-1, p = price of rice at Rs 4500 t-1), indicated an economic dose of 11.9 kg Zn ha-1. Thus the present recommendation of 2 kg Zn ha-1 is inadequate for rice production in Zn-deficient lowland soils of South Gujarat. Our results using these two rice varieties indicate the need to increase Zn application to about 12 kg ha-1 for these soils.

1. Relationship between Zn level and grain yield.

Optimization of potassium application in acid soils of Tamil Nadu

A. Ithayarajan, R. Mariappan, and P. Vivekanandan, Rice Research Station, Ambasamudram 627401, Tamil Nadu, India

2. Relationship between Zn level and total Zn uptake.

actions between Zn levels, years, and varieties. In main effects, only the Zn levels were significant for yield and uptake. A quadratic relationship was found between Zn levels and pooled yield and pooled total Zn uptake (Figs. 1 and 2). Solving the equation

To assess the response of K in Typic Ustropepts soils (pH 5.6; EC 0.24 dS m-1) and to optimize its application for low-land irrigated rice (37.3 kg available K ha-1), we laid out a trial of variety ASD18 on 3.4- 4.5-m plots with 15- 10-cm hill spacing during the 1995 dry season (June-September) with six levels of K2O and two levels of lime (as CaO) using a factorial randomized

Table 1. Influence of K on grain yield of rice in acid soils.

K2O (kg ha -1 ) 0 25 50 75 100 125 CD (p=0.05) Potash Lime Potash lime

Grain yield (t ha -1) L0a 4.6 4.7 4.9 5.1 5.3 5.5 0.09 0.11 0.28 L1 5.1 5.2 5.5 6.5 6.1 6.0 Mean 4.8 5.0 5.2 5.8 5.7 5.8

where X( opt ) = optimum Zn dose for uptake, indicated a continuation of Zn uptake up to 14.3 kg Zn ha-1 and a yield increase up to 13.6 kg Zn ha-1. Solving the equation,

a L0 = zero Iime, L1 = lime at 2.8 t ha-1.

Vol. 22, No. 1

37

block design with three replications (Table 1). Potash was applied per treatment in two equal splits as basal and 30 d after transplanting (DT) (panicle initiation stage); K2O as basal at 50 kg ha-1; and N at 125 kg ha-1 with 50 kg ha-1 as basal and 25 kg ha-1 on 15, 30, and 45 DT. Grain yield was assessed by plot, and yield component traits were recorded on five plants in a single replication. Significant differences were found between the levels of K, lime, and their interaction. At all K levels, lime application increased grain yield. Application at 75 kg K 2O ha-1 with lime resulted in the highest grain yield of 6.5 t ha-1.

Without lime application, K response was low and at 125 kg K2O ha-1 the grain yield was only 5.5 t ha-1 (Table 1). Increases were also observed in plant height, productive tillers, panicle length, filled grains panicle-1, and 1000grain weight (Table 2). Potassium is an excellent replacer of Al in exchangeable clay complexes. Soluble Al (hindrance to P uptake) is precipitated into aluminum hydroxide by the application of lime, which mitigates the toxic effect of Al and increases the uptake of K in acid soils. In these acid soils, 75 kg K2O ha-1 applied with 2.8 t lime ha-1 were optimum for rice cultivation.

Table 1. Effect of N applications on grain yield of rice.

Grain yield (t ha -1 ) Treatment N rate (kg ha -1) 120 180 LSD (0.05) 1988 1989 1991 1992 Mean

6.7 7.2 0.4

7.0 7.7 0.7

5.2 6.7 0.7

7.4 7.7 0.3

6.6 7.3

Time of N application a T1 T2 T3 T4 T5 T6 T7 LSD (0.05) Interaction CV (%) 6.5 6.7 17.4 7.2 7.2 6.7 0.5 ns 6 7.2 7.3 7.4 7.4 7.4 7.4 7.5 ns ns 7 6.0 6.2 5.8 5.8 0.4 ns 5 7.3 7.8 8.0 7.4 0.5 ns 5 6.8 7.0 7.2 7.1 7.3 7.1 6.9

Table 2. Effect of K and lime application on plant characters of rice grown on acid soil.a

a T1 = 1/2 basal, 1/4 at tillering, 1/4 at 6-7 d before panicle

Plant height K 2O (kg ha -1) 0 25 50 75 100 125 (cm) L0

b

tillers hiIl -1(no.) L0 6.1 6.2* 5.9 6.2* 6.2* 6.2* L1 6.3 6.5 6.8 6.8 6.9* 6.9*

Productive

Panicle length (cm) L0 18 20 20 21* 21* 21* L1 20 21 21 22* 22* 22*

panicle -1 (no.) L0 57 80 84 89 93* 104* L1 64 81 87 97* 106* 106*

Filled grains

1000-grain weight (g) L0 21.0 21.0 21.6 21.8 22.0* 22.0 L1 21.4 21.5 22.0 22.1 22.1 22.6*

L1 b 82 83 83 88* 89* 92*

80 80 81 82* 81 83*

initiation (DBPI); T2 = 2/3 basal, 1/3 at 6-7 DBPI; T3 = 1/5 basal, 1/5 a tillering, 1/5 at 6-7 DBPI, 1/5 at booting, 1/5 at heading; T4 = 1/5 basal, 1/5 at tillering; 1/5 at 6-7 DBPI, 1/ 5 at booting, 1/5 (foliar spray) at heading; T5 = 1/4 basal, 1/ 4 at tillering, 1/6 at 6-7 DBPI, 1/6 at booting, 1/6 at heading; T6 = 1/4 basal, 1/4 at tiliering, 1/6 at 6-7 DBPI, 1/6 at booting, 1/6 (foliar spray) at heading; T7 = 1/2 basal, 1/8 at tillerlng, 1/8 at 6-7 DBPI, 1/8 at booting, 1/8 at heading.

a * = significant at the 5% level. b L0 = zero lime, L1= lime at 2.8 t ha -1 .

Table 2. Economics of late N application at booting and heading stages over recommended N split application at two N rates. a

Economic factor

N rates (kg ha -1 ) 120 1 180 1

Increased yield of lowland rice with late N application in the reproductive phase and at high N rates
P. C. Pandey, P. S. Bisht, and P. Lal, G. B. Pant University of Agriculture and Technology, Pantnagar 263145, Nainital, Uttar Pradesh, India

Nitrogen deficiency results from current recommendations of local bestsplit N applications. A field experiment was designed to test the hypothesis that supplying a large proportion of N during the reproductive phase (1 wk before panicle initiation, at booting, and at heading) would eliminate this deficiency and improve yields. Applications at recommended N (120 kg ha-1) and at an increased level (180 kg N ha-1)

were carried out in a split-plot design (N rates in main plots and N splits in subplots). The experiments used four replications during the wet seasons of 1988, 1989, 1991, and 1992 at Pantnagar (29 N, 79' E, and 244 m) on silt loam with pH 7.9, 1.1% organic C, CEC 20 meq 100 g -1 soil, 0.1% total N, 13.9 kg available P (Olsen) ha-1, and 202 kg available K ha-1 (NH4OAc extraction). Rice variety Pant Dhan 4 (134 d duration) was sown 1 Jun in a nursery and transplanted between 5 and 15 Jul in the 4 yr with basal applications of 18 kg , P ha-1, 33 kg K ha-1 and 10 kg Zn ha-1. Time of N application had a significant effect in all years except 1989 (Table 1). In general, the results suggest that controlling the supply of N during initial growth stages and following

1. Added labor for topdressing of urea per ha required at booting and heading 2. Wage of labor (US$ d -1 ) 3. Added labor cost (US$) 4. Added yield of rice obtained from late application (t ha -1 ) 5. Price of rice (US$ t -1 ) 6. Added return (US$) 7. Added net return from late application (item 6-item 3)

1.2 1.2 0.5

1.2 1.2 0.4

100 50 48.8

100 40 38.8

aCost of cultivation was the same for both treatments except for the extra cost incurred in topdressing of urea at booting and heading. Av of 1988, 1989, 1991, and 1992 data.

with applications of larger amounts of N topdressing in the reproductive stages bring higher yields than the recommended N split application.

38

IRRN

1997

Based on added return-added cost ratio, topdressing N at late stages (booting and heading) would be more profitable over the recommended N split (Table 2). Late application at booting and heading at 120 and 180 kg N ha -1 gave an additional return of

US$48.80 and US$38.80, respectively, over the recommended N split (T1). This study also found a significant response of rice after applications of 180 kg N ha-1 in all years (Table 1). The grain yield over 120 kg N ha-1 was 0.5 t in 1988, 0.7 t in 1989, 1.5 t in 1991, and 0.3 t in

1992, or an average increase of 0.75 t. With the present cost of fertilizer and price of rice, the net return is US$62.40 ha-1. Thus economic doses of N for transplanted rice now reach as high as 180 kg N ha -1 in intensive cropping in the Tarai region of Uttar Pradesh, India.

Fertilizer managementorganic sources

Management of urea briquettes containing diammonium phosphate increases rice yields of small farmers
S. Y. Daftardar, Janaseva Foundation, Dadra, 296230 Union Territory of DNH, India; S. M. Wagle, Bharatiya AgroIndustries Foundation, Jwhar, 401603 Thane District, Maharashtra, India; and N. K. Savant, StaSav International, 2551 Hough Road, Florence, AL, 35630, USA

We compared the use of conventional prilled urea (PU), diammonium phosphate (DAP), and urea briquettes (UB) containing DAP (UB-DAP)(4:1, N:P) as NP sources for rainfed transplanted rice. The experiments were done in the 1993, 1994, and 1995 wet seasons in a warm subhumid tropical region on the west coast of Maharashtra State, India.

Three regimes were tested in 117 farmer-managed field trials: (1) Improyed management. The UB-DAP was deep-placed by hand on the day of transplanting (DP) with modified 20- 20-cm spacing (see figure). One 2.7-g UB-DAP per 4 hills supplied 56 kg N ha-1 and 14 kg P ha-1 . (2) Conventional management. Nitrogen as PU was applied in two splits (one-half at DP and one-half topdressed at panicle initiation stage) and all P as single superphosphate (SSP) at DP was applied at the same N and P rates as that in improved and farmers management. (3) Farmers management. Traditional random transplanting and fertilizer at 0-40 kg N ha-1 and 0-10 kg P ha-1. Average plot size was about 100 m 2. Duration of rice cultivars ranged from

110 to 135 d. Four- to five-week-old seedlings, prepared by farmers, were transplanted at 5-6 seedlings hill -1 for the improved management trial. Valuecost ratio (VCR) of fertilizer management was determined by dividing the value of additional grain + straw yields by all additional variable costs of fertilizers and their application. Firstdegree stochastic dominance analysis was used to determine risk efficiency. Despite marked differences in seasonal rainfall, improved management significantly increased grain and straw yields over conventional management (PU + SSP) and was distinctly superior to farmers management (see table). Agronomic performance of PU + SSP was variable and unsatisfactory. Yields for improved management with improved varieties needed 40% less

a) Bamboo transplanting guide method; b) Two-row transplanting method used for achieving the modified 2020 cm spacing (plant population, 25 hills m -2 ).

Vol. 22, No. 1

39

fertilizer than locally recommended NP rates and a 30-35% lower plant population. Modified 20- 20-cm spacing could reduce up to 50% of the labor normally required for the UB placement by hand. Estimated VCRs for improved management of UB-DAP vis-a-vis the farmers management ranged from 5.0 to 8.5. The VCRs for the conventional management of PU + SSP were less than 2.9 and economically unattractive. Stochastic dominance analysis indicated that the improved management of UB-DAP was a riskfree practice and therefore would be accepted by small farmers and preferred by policymakers.

Average rice yields under farmer-managed field trials as influenced by improved management of UB-DAP vis-avis the conventional and farmers' management of PU+SSP. Maharashtra State, India. 1993-95 wet seasons. a

Regime 1993 wet season (n=26) Improved Conventional Farmers' 1994 wet season (n=51) Improved Conventional Farmers 1995 wet season (n=40) Improved Conventional Farmers
a

Fertilizer

Yield (t ha-1) Grain 4.6 1.1 a 3.0 1.0 b 2.4 1.1 c 3.9 1.1 a 3.1 1.0 b 2.7 1.0 c 4.1 1.2 a 3.0 1.2 b 2.7 1.1 b Straw 6.7 2.3 a 4.5 1.9 b 4.1 1.9 b 5.3 1.6 a 4.4 1.3 b 4.2 1.3 b 4.6 1.2 a 3.2 1.3 b 3.2 0.9 b

UBDAP PU+SSP PU+SSP UBDAP PU+SSP PU+SSP UBDAP PU+SSP PU+SSP

n = number of field trials. Fertilizer rates used for the improved management of UB-DAP and the conventional management of PU and SSP: 56 kg N ha-1 t + 14 kg P ha-1. Farmers' fertilizer rates varied and were lower than these rates. Av yields ( SD) for a given season followed by the same letter are not significantly different from each other by LSD (P=0.05).

Effect of Azolla caroliniana and Sesbania rostrata on rice yield

S. Sahoo and B. K. Datta, Vivekananda Institute of Biotechnology, P. O. Nimpith Ashram, South 24-Parganas, West Bengal 743338, India

We studied how rice yield was affected with the use of Azolla as a dual crop and as green manure along with a green leaf manure plant, S. rostrata. Yield was compared using N fertilizer in pot culture. Experimental plots contained poorly drained fine soils occurring on the nearly level upper delta of the Ganges (Haplaquepts), pH 7.6; organic C, 0.58%; EC, 0.05 mmho cm-1; N,

0.06%; 29.9 kg available P ha-1; and 265.6 kg K ha-1. S. rostrata was sown into the soil at 10 t ha-1 just 1 d before transplanting. Nitrogen was 1.5% of fresh biomass. Azolla caroliniana was grown in a multiplication tank. As a dual crop, fresh Azolla was inoculated after 6 d of transplanting at 2 t ha-1 , left to grow along with rice, and incorporated into the soil after 20 d of growth. For basal application, fresh Azolla biomass at 10 t ha-1 was incorporated into the soil 1 d before transplanting the rice. In the presence of 15 kg N ha-1 as urea, Azolla as basal/dual yielded much less than that of the control with

60 kg N ha-1 as urea (see table). In the presence of 30 kg N as urea, Azollabasal proves better compared with Azolla-dual. The use of Azolla as basal proved better compared with Azolladual. The use of Azolla as basal + dual was also better in the presence of 30 kg N as urea. The addition of S. rostrata as green manure (providing 60 kg N ha-1) in the presence of Azolla (dual) almost negated the requirement of the addition of chemical N, whereas the addition of 30 kg N (as urea) to the above combination resulted in a significant 100% increase in the grain and straw yield over the control.

Effect of Azolla caroliniana and Sesbania rostrata on IR36 yield on the upper delta of the Ganges, India. 1994 wet season.

Treatment

N from urea (kg ha -1)

N from Azolla (basal) (kg ha -1) 24 24 24 24

N from Azolla (dual) (kg ha-1) 48 48 48 48 48 48 48

N from S. rostrata (kg ha -1) 60 60 60

Tillers at panicle initiation (no.) 8 8 7 8 8 8 9 15 17 17

Increase over control (%) 12.5 12.5 87.5 112.5 112.5

Grain yield (g) 13.7 12.7 9.5 12.5 14.0 13.0 14.2 22.5 26.0 27.0

Increase over control (%) 7.29 30.65 8.75 2.18 5.10 3.65 64.23 89.78 97.08 6.74

Straw yield (g) 11.5 10.5 9.0 10.0 15.0 8.7 12.2 24.5 28.9 29.5

Increase over control (%) 8.69 21.73 0.13 30.43 24.34 6.08 113.04 143.47 186.53

60 kg N (control) 60 15 kg N + Azolla (B) a 15 b 15 kg N + Azolla (D) 15 15 kg N + Azolla (B+D) 15 30 kg N + Azolla (B) 30 30 kg N + Azolla (D) 30 30 kg N + Azolla (B+D) 30 No N + S. rostrata + Azolla (D) 15 kg N + S. rostrata + 5 Azolla (D) 30 kg N + S. rostrata + 30 Azolla (D) CD (P=0.05)
aB = basal. b D = dual.

40

IRRN 1997

Contribution of green manure in controlling the loss of applied fertilizer nitrogen from rainfed ricesoil
K. Bhattacharyya and S. R. Mandal, Agricultural Chemistry and Soil Science Department, Bidhan Chandra Krishi Viswavidyalaya, Mohanpur Nadia 741252, West Bengal, India

This field study evaluated the efficiency of green manure to restrict loss of applied fertilizer N through successive nitrification and denitrification. A factorially randomized block design with three replications was done using IR36 in 1991 on a sandy loam soil (pH 7.5; organic C, 0.40%, total N, 0.04%; CEC, 9.4 meq 100 g-1 soil). It was repeated in 1992 with IET5656 on a silty loam soil (pH 7.7; organic C, 0.53%; total N, 0.05%; and CEC, 12.6 meq 100 g-1 soil) at Kalyani (22 53' N, 88 31' E). Six-week-old Sesbania rostrata, a stemnodulating green manure, was incorporated at 3 kg m -2 into half of the 1- 1-m plots and seedlings transplanted 5 d later. The plots received 100 kg N (as 15N enriched urea; 5% AE) in three splits, i.e. 70 kg basal, 20 kg at 40 d after

transplanting (DT,) and 10 kg at 80 DT; 26.4 kg P (basally as single superphosphate); and 33.2 kg K (basally as muriate of potash). Four different rainfall conditions were maintained by imposing moisture stress using polyethylene sheets to intercept rain and constant monitoring with field tensiometers. Drought increased both leaching and gaseous losses of fertilizer 15N over continuous submergence (Table 1). Green manure significantly reduced both leaching and gaseous losses of fertilizer 15N and remained most efficient under drought prevailing at the vegetative stage. Maximum reduction in gaseous losses of fertilizer 15N through green manure was 66.31% in sandy loam soil under drought at the reproductive stage of IET5656. Savings of fertilizer N were reflected in rice grain yield, registering a maximum of 4.20 t ha-1 (IR36) and 8.09 t ha-1 (IET5656) under continuous submergence with green manure. Maximum leaching loss under vegetative-stage drought limited the grain yields to 1.9 t ha-1 (IR36) and 4.8 t ha-1 (IET5656). Drought imposed at maturation did

Table 2. Economic aspects of using green manure in rainfed rice. West Bengal, India. 1991-92.a Additional expenditure for green manure use (US$ ha -1) Continuous submergence Drought at vegetative stage Drought at reproductive stage Drought at maturation stage
a

Moisture status

Additional income from green manuring (US$ ha -1) Rice grain yield 164.40 Fertilizer N saved 9.92 Net profit

7.20

167.12

7.20

18.00

8.48

19.28

7.20

164.40

10.68

167.88

7.20

182.40

11.56

186.76

Av of 2 yr (1991-92).

Table 1. Yield and fertilizer N losses of rainfed rice as affected by green manure. West Bengal, India. 1991-92. Leaching losses of fertilizer 15N beyond 30 cm depth of soil (kg ha-1) Sandy loam (1991) Continuous submergence Drought at vegetative stage Drought at reproductive stage Drought at maturation stage GM C GM C GM C GM C Silty loam (1992) Gaseous losses of fertilizer 15N) (kg ha-1) Sandy loam (1991) 8.01(55.20) 17.88 22.40 (8.38) 24.45 6.16(66.31) 18.34 13.56(34.78) 20.79 Silty loam (1992) 6.32 (78.70) 29.67 27.97 (20.65) 35.25 17.77 (40.45) 29.84 4.48 (82.70) 25.89

Moisture status

Manure

Grain yield (t ha-1) IR36 (1991) IET5656 (1992)

25.59 (18.08) 37.82 35.04 31.24 17.71 (52.13) 25.93 (48.28) 37.00 50.14 34.99 (12.50) 29.14 (31.04) 39.99 42.26 28.20 (12.99) 32.33 32.41 33.00 (2.03)

4.2(31.7) 8.1(27.2) 3.2 6.4 3.0(57.6) 6.9(42.9) 1.9 4.8 3.0(54.9) 7.0(31.9) 1.9 5.3 4.1(38.0) 8.1(31.2) 3.0 6.2

not change the grain yields from continuous submergence, and thus conservation of irrigation water can be encouraged at this stage without hampering yield. Green manure significantly increased the grain yields of rice varieties, registering maximum increments of 57.59% (IR36) and 42.87% (IET5656) under drought during the vegetative stage. Drought at any growth stage was harmful to N economy. The use of green manure recovers fertilizer 15N losses from rainfed rice soil to a great extent and stimulates yield to benefit the economy of rainfed rice farming by a margin of US$19.28 to US$186.76 ha-1 yr-1 (Table 2) under various rainfall conditions.
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.

C.D. (at 5%) Manure Moisture Interaction

a

2.185 4.089 2.677

b

3.840 ns 10.159

5.548 10.330 ns

4.649 ns 12.300

0.745 ns ns

0.225 0.420 ns

GM = green manure, C = control.

Figures in parentheses indicate percent change over control. Av of three replications.

Vol. 22. No. 2

41

Crop management
Influence of planting dates on productivity of traditional scented rice varieties
K. S. Gangwar and S. K. Sharma, Project Directorate for Cropping Systems Research, Modipuram, Meerut 250110, Uttar Pradesh, India

To exploit the full yield potential of traditional scented rice varieties, it is necessary to determine their optimum planting time for specific locations. A field experiment was conducted during the 1993 and 1994 wet seasons using a split-plot design with four dates of planting as main plot treatments and three varieties of traditional Basmati as subplot treatments, replicated four times. Soil at the experimental site was sandy loam, low in organic C (0.41%), low in available P (7.7 kg ha-1 ), and medium in K (178.5 kg ha-1) with pH 7.8. The crop was fertilized with 60, 13.2, and 24.9 kg of N, P, and K ha-1 , respectively of which 50% N, full P, and K along with 5.8 kg Zn ha-1 were applied uniformly as basal dressing and incorporated. The other 50% N was applied in two equal splits at tillering and panicle initiation. Seedlings (21 d old) were transplanted at 20- 15-cm spacing with 2-3 seedlings hill-1. Dates of planting influenced the grain yield of Basmati rice significantly during both years of study (see table). The highest mean grain yield was recorded from the 1 Jul planting date and did not differ significantly from the 16 Jul planting, but yield differed from that of 31 Jul and 16 Aug. A linear reduction in grain yield with every 15-d delay in planting was recorded from 1 Jul (4.1 t ha-1) to 16 Aug (2.5 t ha-1). The percent reduction in grain yield averaged over 2 yr was 11.8, 20.8, and 37.4%, respectively, for 16 Jul, 31 Jul, and 16 Aug as compared with the grain yield of the 1 Jul planting. In yieldcontributing characters, panicle weight differed significantly during both years while panicles m -2 differed only during
42 IRRN 1997

1994. The maximum panicles m-2 were recorded for the 1 Jul planting, whereas panicle weight remained the same for the 1 Jul and 16 Jul plantings. The interaction between planting dates varieties was not significant. The test varieties did not differ significantly among themselves in grain yield. Basmati 370 had a slightly higher grain yield (3.40 t ha-1), followed by Taraori Basmati (3.38 t ha-1). Basmati 370 also had the highest average

panicles m -2 (387 m-2 ) while Taraori Basmati had the highest average panicle weight (1.67 g panicle-1). Results indicated that Basmati 370 could be used for early planting whereas Ranveer Basmati could be used for delayed planting. The higher grain yield of such traditional scented rice varieties could be achieved by transplanting them during the first fortnight of July in the western part of Uttar Pradesh.

Grain yield and yield-contributing characters of traditional scented rice varieties under different planting dates. Uttar Pradesh, India, 1993-94 wet seasons.

Treatment

Grain yield (t ha -1) 1993 1994 Mean

Panicles m -2 (no.) 1993 1994 Mean 1993

Panicle weight (g) 1994 Mean

Planting date 1 Jul 16 Jul 31 Jul 16 Aug CD (5%) Variety Basmati 370 Ranveer Basmati Taraori Basmati CD (5%)

4.2 3.6 3.3 2.8 0.9

3.9 3.5 3.1 2.3 0.7

4.1 3.6 3.2 2.5 -

406 369 385 378 ns 409 383 361 ns

358 345 330 309 15 365 355 360 9

382 357 358 344 -

1.74 1.71 1.67 1.62 0.03 1.64 1.67 1.73 0.02

1.68 1.70 1.59 1.55 0.05 1.60 1.52 1.61 0.06

1.71 1.71 1.63 1.59 -

3.5 3.5 3.5 ns

3.3 3.1 3.2 ns

3.4 3.3 3.4 -

387 369 361 -

1.62 1.60 1.67 0

Seedling vigor affects stand establishment and performance of flood-prone lowland rice
A. Ghosh and A. R. Sharma, Central Rice Research Institute (CRRI), Orissa 753006, India

We conducted a field experiment at Cuttack, India, during 1993 using three long-duration, photoperiod-sensitive rice varieties of different plant heights. Vegetative tillers were transplanted from conventional nursery seedbeds unfertilized or with N fertilization (100 kg N ha-1 at sowing). Their performance was compared with that of sowings at 400 kg N ha-1 in a nursery seedbed in a relatively upland field and 80 kg N ha-1 in a lowland field, from where about 100 tillers m -2 were up-

rooted and transplanted. Seedlings and vegetative tillers (45 d old) were transplanted into 8 cm of water at 20- 15-cm spacing of 2-3 hill-1 in a randomized block design with three replications and fertilized with a basal dose of 40 kg N ha-1 , 8.7 kg P ha-1, and 16.7 kg K ha-1. Vigor depended on variety and type of seedling (see table). Fertilizer N application in the seedbed improved height and dry weight compared with unfertilized seedlings. Two floods occurred within the month of transplanting (see figure). The taller and heavier plants were well established when subjected to 54 cm of water 4 d after transplanting. Plants raised from nursery seedlings suffered the greatest damage. Unfertilized seed-

lings (27-34 cm height) remained completely under water for nearly a week, whereas the top 2-3 leaves of fertilized seedlings were above water. The second flood killed most of the unfertilized seedlings. The mean grain yields of all rice varieties were similar, but crops raised from vegetative tillers

yielded best. Panicle number in the vegetative propagated crop was 112132 m-2 and their weight was 3.39-3.63 g, leading to a grain yield of more than 4 t ha-1. Nursery seedlings had very low yields and poor crop stand even with good panicle weight (2.70-3.32 g) because the flood reduced panicle num-

Effect of seedling vigor on growth and yield attributes of rice varieties under flood-prone conditions. Cuttack, India. 1993.

Seedling vigor a Treatment Height (crn) 27 33 71 32 56 71 34 57 98 Dry weight (g seedling-1) 0.08 0.15 0.86 0.08 0.18 0.79 0.17 0.22 1.25

Plant height at maturity (crn)

Panicles m -2 (no.)

Panicle weight (g)

Grain yield (t ha -1)

Straw yield (t ha -1)

ber (61-88%) and tillers. Straw yield was highest with Matangini, and its nursery seedlings were also significantly better than those of Utkalprabha and Gayatri. Seedling height and dry weight at transplanting determine survivability, crop stand, and productivity under flood-prone conditions. Suitable varieties can be obtained either by uprooting vegetative tillers from a field-grown crop or from a conventional nursery seedbed raised using relatively low seed density and basal N fertilization.

Gayatri (semidwarf) Unfertilized seedlings Fertilized seedlings Vegetative tillers Utkalprabha (semi-tall) Unfertilized seedlings Fertilized seedlings Vegetative tillers Matangini (tall) Unfertilized seedlings Fertilized seedlings Vegetative tillers

74 93 106 105 107 136 107 114 164

15 28 124 22 34 133 31 43 112

2.89 3.30 3.63 3.18 3.32 3.59 2.70 2.98 3.39

0.2 0.9 4.1 0.4 0.7 4.3 0.7 1.3 4.1

0.3 1.5 5.5 0.8 1.5 6.2 1.4 2.3 7.6

CD (0.05) Varieties Kind of seedlings Varieties kind of seedlings CV (%)

a

6 6 6 3.2

ns 8 13 12.9

0.18 0.18 ns 5.5

ns 0.33 ns 17.7

0.35 0.35 0.61 11.6

Based on mean of 25 seedlings.

Flooding patterns during the growth period of rice. Cuttack, India. 1993.

Integrated pest management

A survey of rice constraints in the Mekong Delta
P. V. Du, Cuu Long Delta Rice Research Institute, Can Tho, Vietnam; S. Savary, IRRI-ORSTOM Joint Project on Rice Pest Characterization; and F. A. Elazegui, IRRI

A survey was conducted in three ricecropping seasons of 1996 in the Can Tho Province of Vietnam. The sample consisted of 73 fields distributed over these successive seasons (20, 32, and 21 fields) and four villages. The Survey Portfolio of the Rice Research Prioritization Project supported by the Rockefeller Foundation was used. Analysis of variance was carried out on individual injuries attributable to

pests for the three seasons (see table). Leaf folder (LF), deadhearts (DH), and leaf blast (LB) injuries were higher in winter-spring; weed infestation (above the rice crop canopy WA) and sheath rot (ShR) injuries were higher in summer; and brown spot (BS) and bacterial blight (BLB) injuries were higher in autumn. Brown planthopper (BPH) populations were higher in winterspring, although remaining at low levels. Injury due to red stripe syndrome (RS) was also highest in winter-spring, and next highest in summer. Analysis of variance was likewise conducted on individual injuries for various levels of cropping practices. We found numerical, but not significant effects, of fertilizer inputs on levels of

injuries, except for sheath blight (ShB) (P< 0.01, increase with fertilizer application) and weed infestation below the rice crop (WB, P = 0.02, decrease with fertilizer application). No numerical trends, nor significant differences, were found to suggest a link between fertilizer input and BS or RS. Some injuries differed (P < 0.05) in intensity depending on location (village): WB, DH, BS, and NB, reflecting the influence of a number of possible factors (e.g., soil characteristics, varieties, crop husbandry). No significant insecticide effects on any injury attributable to insects were found except one, BPH, where injury was significantly (P = 0.039) increased when insecticide was used. No significant effect of fungicide on

Vol. 22. No. 1

43

any injury attributable to pathogens was found either, except for ShB, whose injury was also increased by fungicide use. Relationships between different cropping practices (e.g., water management, variety, etc.) did not show a distinctive pattern. A multivariate approach was then used to address multiple, complex associations. Results of chi square tests and multiple correspondence analysis can be summarized as follows: Winter-spring was characterized by higher levels of LB, RS, and BPH; summer by higher RS and WA; and autumn by higher BLB and BS, but lower levels of RS. ShB injury, high in all seasons, peaked in winter-spring. In winter-spring, water management was generally good; in summer, water stress was found in some fields, and there were low fertilizer inputs and higher use of herbicides; in autumn, water excess (floods) occurred in some fields. Winter-spring was characterized by higher yields (5 t ha-1 and more), summer by medium yields (2.5-5 t ha-1, and autumn by low yields (2.5 t ha-1 and less). Multiple correspondence analysis using patterns of cropping practices and injury variables described much of the yield variation: at least 88% of each yield class is accounted for on the first two

Variation in levels of injuries with rice cropping seasons. Can Tho Province, Mekong Delta. 1996.

Injurya
Leaffolderb Whiteheadsc Deadheartsc Weed below the rice canopyd Weed above the rice canopyd Neck blast c Stem rot c Sheath rot c Sheath blightc Red stripe b Brown spotb Leaf blast b Bacterial blightb Rice bugs e Brown planthopper e
a

Winter-spring (Dec-Mar) 517 0.19 2.50 170 385 2.83 0.75 0.77 3.28 352 649 253 21 0.35 75

Summer (Apr-Jul) 118 0.38 1.00 400 600 1.17 0.21 7.41 4.07 228 583 5 16 0.33 28

Autumn (Jul-Oct) 9 0.51 0.90 414 232 0.36 0.12 1.84 9.22 135 1885 3 195 3.08 10

32.2 0.80 4.69 2.74 4.56 1.98 2.72 4.96 3.32 4.67 17.3 18.0 4.38 2.58 9.70

<0.001 0.45 0.01 0.07 0.01 0.14 0.07 0.01 0.04 0.01 <0.001 <0.001 0.01 0.08 <0.001

Injury levels are indicated in different units. The magnitude of entries does not reflect the magnitude of the injurious effect on the crop and the yield-reducing effects. b Area under progress curve of mean proportion of injured leaves. c Maximum proportion of tillers injured over four successive visits. d Area under progress curve of proportion of soil covered by weed canopy. e Area under progress curve of mean insect catches over four successive visits.

combined axes. (These axes represent independent combinations of variables, except yield, using a chi-square distance.) This analysis indicates no apparent effect of pesticide usefungicides, insecticides, and herbicideson the pattern of distribution of injuries. In other words, pesticides do not appear to have any significant effect on the overall occurrence of pests, except for the two cases mentioned. This survey provides a first overview of injuries due to pests in Mekong Delta rice cultivation. It points to sea-

sonal patterns, a few environmental factors that may influence injuries, and some links among injuries. The survey results permit us to make hypotheses, especially to explain why pesticide use, which is very high, has so little effect on injury levels, and hypotheses about red stripe syndrome. Surveys in the Mekong Delta are ongoing. We expect to derive from an additional year of surveyand a larger survey data set production situations (combination of climatic and crop management practices) that will permit us to develop testable hypotheses.

Integrated pest managementinsects

Impact of methyl parathion on the natural enemies of rice insect pests in Cambodia
G. S. Arida, K. L. Heong, IRRI; P. Visarto and H. J. Nesbitt, Cambodia-IRRI-Australia Project

Methyl parathion is one of the most commonly used insecticides in Cambodia. An experiment at Kap Srau Station, Phnom Penh, during 1994 main crop season evaluated methyl parathion effects on major arthropod groups in rainfed lowland rice.

We planted IR72 in a randomized complete block design on 35- 20-m plots, with treatments of sprayed and unsprayed plots each replicated three times. Methyl parathion was sprayed 14 and 38 d after transplanting (DT) at 0.75 kg ai ha-1. Arthropods were sampled 1 d before (DBS) and at 3 d after spraying (DAS). Samples were taken from 10 randomly selected sites per plot using a blower-vac suction machine. Each sample site was enclosed with a bottomless plastic bucket fixed with a muslin cloth around its

opening to prevent escape of arthropods during suction. The enclosed area was 0.102 m 2 and covered 3-4 hills. Arthropods were placed in labeled vials with 80% alcohol, counted, and identified. Damage from defoliators was also recorded at 45 DT from 10 randomly selected hills plot-1. Yields were taken from 2- 5-m area in each plot and adjusted to 14% moisture content. The arthropod population collected at the first set of samplings was too low for useful comparisons. In the second sampling, green leafhopper density

44 IRRN 1997

Effect of soil amendments on grain yield and incidence of rice leaffolder in iron-toxic soils of north Orissa, India
H. P. Patnaik and U. C. Panigrahi, Regional Research Station, Orissa University of Agriculture and Technology, Keonjhar 758002, Orissa, India

1. Population of N. virescens and its parasitoids, Gonatocerus sp. 1 d before (DBS) and 3 d after spray (DAS). Kap Srau, Phnom Penh, Cambodia. 1994 wet season.

In valley areas of north Orissa, India, the incidence of rice leaffolder (LF), Cnaphalocrocis medinalis, and iron toxicity have been of concern in wetland rice production. This field experiment examined the effects of soil amendments on the incidence of LF and the grain yield of two varieties, Lalat and Jajati. A factorial randomized block design with three replications was laid out in farmers' fields adjacent to the iron ore mines of Joda, Keonjhar (Orissa), India. The soil (Oxic Paleustalf) of the experiment site had a pH of 4.8. The HCl extract of the soil evidenced Fe 2+ content of 3500 ppm (by ortho phenanthroline method). The treatments consisted of application of various fertilizer rates with organic manure as amendments on 12- 12-m plots. One-monthold seedlings were transplanted at a spacing of 15 10 cm on 24 Jul l993. The percentage of leaf damage caused by the LF in each plot was ascertained at the flag leaf stage by counting the total and infested leaves on 10 randomTable 1. Leaffolder (LF) incidence and grain yield in response to amendments to iron-toxic soils in rice, irrespective of variety.

2. Population of common predators of insect pests of rice and collembolans 1DBS and 3 d DAS. Kap Srau, Phnom Penh, Cambodia, 1994 wet season.

Treatment 60:13.2:24.9 kg NPK ha -1 T1 + cowdung @ 5 t ha -1 T1 + MnO (4%) seedling root dip 60:13.2:49.8 kg NPK ha -1 60:13.2:74.7 kg NPK ha -1 60:13.2:99.6 kg NPK ha -1 30:13.2:24.9 kg NPK ha -1 + FYM @ 3.3 t ha -1 + green leaves of Ipomoea @ 1.7 t ha -1 LSD (P=0.05)
b ns = not significant.

Leaf damage (%) by LF a

Grain yield (t ha -1 ) 2.7 3.4 2.9 2.9 3.0 2.9 2.8

increased in both treated and untreated plots by 3 DAS (42 DT). By contrast, the number of Gonatocerus spp., an egg parasitoid of Nephotettix virescens, declined (Fig. 1). Populations of spiders and collembolans were heavily affected by insecticide applications. The spraying had no effect on the populations of Cyrtorhinus lividipennis Reuter and Microvelia douglasi atrolineata Bergorth (Fig. 2). Leaf damage was similar (15 and 16%) and yield comparable (2970

209 and 2290 221 kg ha -1 ) in the treated and untreated plots. The insecticide did not appear to cause significant reduction in pest species and predators such as Cyrtorhinus and Microvelia. Instead, it had more significant effects on parasitoids, spiders, and collembolans. It is likely that pest infestations in Cambodia are inherently low, especially in an insect-resistant cultivar like IR72. Methyl parathion applications are thus not justified.

22.6 (27.2) 34.6 (35.8) 32.6 (34.7) 22.6 25.3 22.6 25.3 (28.3) (30.1) (28.1) (29.9)

3.8

ns b

a Figures in parentheses are in corresponding angular values.

Vol. 22, No. 1

45

ly selected hills. Leaf damage ranged from 22.6 to 34.6% and, among the amendments tested, application of cow dung (at 5 t ha-1) or seeding root dip in MnO (4%) appeared to accelerate leaf damage (32.6-34.6%) by LF appreciably (Table 1). Irrespective of soil amendments, however, Lalat suffered less

damage (23.05,) than Jajati (30.1%). Moreover, Lalat yielded significantly more (3.4 t ha-1) than Jajati (2.6 t ha -1) in iron-toxic soils (Table 2). Thus, adoption of a variety such as Lalat might be a possible option for stable rice production by resource-poor farmers working the iron-toxic soils of the region.

Table 2. Effect of variety on LF incidence and grain yield, irrespective of soil amendments in iron-toxic soils of north Orissa, India.

Variety Lalat Jajati LSD (P=0.05)

Leaf damage (%) by LF a 23.0 30.1 2.0 (28.2) (33.0)

Grain yield (t ha -1 ) 3.4 2.6 0.4

a Figures in parentheses are in corresponding angular values.

Integrated pest managementweeds

Crop rotation in red rice control
E. Marchezan, Universidade Federal de Santa Maria 97119-900, Santa Maria, RS, Brasil

There are currently no chemical weed control options for red rice, the most important weed of irrigated ricefields in Rio Grande do Sul, Brasil. This research evaluated red rice control through crop rotation in highly infested areas. Rotations included no-till or

tilled rice, soybean, and maize during summer and ryegrass or fallow in winter (Table 1). A trifactorial split-plot design was used with three replications. Treatment effects were evaluated by counting seeds within a 10-cm-deep soil layer and by counting the panicles before harvest. Soybean - maize rotations led to a 82% decrease of the seed reservoir in the soil from 1992 to 1996. Soybean rotation exerted greater control (88%)

than maize (76%). Conventional tillage controlled 93% of the red rice reservoir compared with no tillage at 76%, an amount considered low because of deficient control during the 1993-94 growing season. Continuous rice (S7) had about twice the number of red rice seeds in the soil but reduced by 45%) the number of weed panicles at harvest. Based on the number of red rice panicles (Table 2), crop rotation presented 95% average control. Weed control

Table 1. Number of red rice seeds m -2 in the soil (0-10 cm deep) in areas with ryegrass (c/a) and without ryegrass (s/a) before seeding, Santa Maria, Rio Grande do Sul, Brasil. 1992-96.

Crop sequencea S1 S2 S3 S4 S5 S6 S7 Av
a S1 =

1992-93 c/a 2261 1715 2594 1484 3026 2792 1844 2312
-

1993-94 Av 3552 2975 2947 1268 3047 2692 2002 2747

1994-95

1994-95 Av 1620 852 1404 972 1176 744 1224 1128 c/a 2814 7979 6450 2121 2759 3098 3024 4203 s/a 2418 1619 5177 2206 4541 1018 2716 2829 Av 2616 4795 5813 2163 3650 2058 2870 3516 c/a 308 127 647 435 148 1963 3820 605

1995-96 s/a 361 117 605 287 244 153 3966 294 Av 334 122 626 361 196 1061 3897 450

s/a 4843 4236 3300 1052 3069 2593 2161 3182

1993-94 rice (c) rice (c) rice (d) rice (d)

c/a 2041 744 1777 1368 1633 840 1016 1400

s/a 1200 960 1032 576 720 648 1432 856

% of control 90 96 79 72 94 60 82

S2 = S3 = S4 =

1991-92 Rice (c) Rice (c) Rice (d) Rice (d)

1992-93 soybean (c) maize (c) soybean (d) maize (d)

soybean (c) maize (c)

soybean (c) maize (d)

1992-93 1991-92 S5 = Rice (c) - soybean (d) S6 = Rice (c) - maize (d) S7 = Rice - rice c = conventional tillage, d =

1993-94 1994-95 - rice (c) - soybean (d) - rice (c) - maize (d) - rice - rice no tillage.

Table 2. Number of red rice panicles m -2 in areas with rye grass (c/a) and without rye grass (s/a) before harvest. Santa Maria, Brazil. 1992-96.

Crop sequence S1 S2 S3 S4 S5 S7 Av

1992-93 c/a 180 118 265 23 171 102 361 143 s/a 201 160 270 61 184 182 389 176 Av 190 139 267 42 177 142 375 160 c/a 136 104 135 63 127 107 159 112

1993-94 s/a 140 76 83 52 113 113 164 96 Av 138 90 109 57 120 110 161 104 c/a 9 7 4.6 3.6 3 4.3 213 5.2

1994-95 s/a 10 10.3 5.3 5.3 6.3 2.3 198 6.6 Av 9.5 8.7 5 4.5 4.7 3.3 206 6.0

% of control

assisted by soybean or maize rotations as well as by tillage systems were similar and always above 90% during the three seasons. Neither the ryegrass nor fallow during winter months affected the seed reservoir or the quantity of red rice panicles. Crop rotation is an efficient method to control red rice in lowland areas where efficient herbicides are also used. In the highly infested areas, it is recommended that the weed seed reservoir should be reduced before adopting crop rotations or no tillage. Ryegrass fallow does not contribute to the control of red rice.

S6

95 94 98 89 97 98 45 95

46

IRRN 1997

Herbicide use and occurrence of Echinochloa spp. in ricefields in dry and intermediate zones of Sri Lanka
P. Van Mele, V. Van Damme, X. Scheldeman, B. Meylemans, and P. Van Damme, Faculty of Agriculture and Applied Sciences, University of Gent, Coupure links 653, B-9000 Gent, Belgium

Table 1. Frequency of Echinochloa spp. during the 1994-95 survey in the dry and intermediate zones of Sri Lanka. Frequency (%) Weed Maha season Echinochloa Echinochloa P. Beauv. Echinochloa Echinochloa P. Beauv. colona (L.) Link crus-galli (L.) frumentacea Link stagnina (Retz.) 52.3 36.9 13.1 6.8 Yala season 81.3 57.3 21.3 13.3

This ricefield weed survey included 251 fields in the dry and intermediate rice growing zones of Sri Lanka in 1994 and 1995. All species within 5- 5-m plots were identified and farmers interviewed about practices. During the 1994 maha season (northeastern monsoon, October-December), we identified 132 different weeds belonging to 32 families in 176 fields. During the 1995 yala season (southwestern monsoon, April-August), 96 different weeds belonging to 21 different families were identified in 75 fields. In both seasons, Echinochloa colona (L.) Link and E. crusgalli (L.) P. Beauv. (Poaceae) ranked among the 10 most important weed

species, while E. frumentacea Link and E. stagnina (Retz.) P. Beauv. occurred less frequently. Overall, Echinochloa spp. frequencies were higher during the dry yala season (Table 1). The reason for this might be that during the yala season all cultivated rice is irrigated whereas during the rainy season (maha) upland rice is also grown. Canonical correspondence analysis (CCA), often used in ecological vegetation science, conducted on species occurring in more than three fields, revealed that during maha, E. crus-galli

is a typical species in frequently irrigated ricefields, whereas E. colona occurs as well under drier regimes. Water level in ricefields during yala, however, does not seem to have a significant influence on their distribution. From interviews with the farmers, it became clear that the range of herbicides used is very limited. Propanil, active against several grassy and broadleaf weeds, and MCPA, against broadleaves, are the most frequently applied herbicides. Only about a third of the farmers reported using the recommended amounts of herbicides (Table 2).
Table 2. Herbicide use (% of farmers interviewed) in ricefields during the 1994-95 survey in the dry and intermediate zones of Sri Lanka. Classa I II III Propanil 41.7 20.8 37.5 MCPA 37.5 30.6 31.9

a Class I: no application: class II: amount of herbicide applied

is less than recommended: class III: recommended amounts according to the Department of Agriculture.

Integrated weed management through smother intercrops in rainfed lowland rice

V. V. Angadi, Agricultural Research Station (ARS), Sankeshwar 591314, Belgaum, India; P. N. Umapathy, Main Research Station, University of Agricultural Sciences, Dharwad 580005, Karnataka, India

DE. Weed-free check, weedy check, and farmer's practices were included for comparison (see table). The rice and green manure seeds were mixed in a

100:25 kg ha-1 proportion and sown in 20-cm rows by seed drill in semidry soil at the onset of the monsoon season. Green manures were buried in stand-

Grain yield and weed dry weight in rainfed lowland rice as influenced by integrated weed management involving smother crops. Karnataka, India. 1992 and 1994. Grain yield (t ha -1 ) 1992 C. juncea + HW at 30 DE C. juncea + IC at 15 DE + HW at 40 DE V. sinensis + HW at 30 DE V. sinensis + IC at 15 DE + HW at 40 DE G. max + HW at 30 DE G. max + IC at 15 DE + HW at 40 DE S. rostrata + HW at 30 DE S. rostrata + IC at 15 DE + HW at 40 DE S. aculeata + HW at 30 DE S. aculeata + IC at 15 DE + HW at 40 DE Weed-free check Weedy check Farmer's practice Butachlor + HW at 30 DE LSD (0.05) 5.2 6.2 6.3 6.7 5.3 6.9 5.9 6.1 5.3 6.7 6.6 3.2 6.8 6.5 1 1994 5.9 6.1 5.8 6.8 5.9 7.2 5.8 5.9 4.6 5.2 6.6 2.7 6.7 6.7 1.2 Weed dry weight (g m -2 ) 1992 175 50 103 25 58 45 106 53 181 36 111 361 81 95 1994 78 49 89 49 137 69 90 53 124 100 8 461 40 49 42

Treatmenta

Among the management options in the first 40-50 d of rainfed lowland rice, smother crops can ensure effective weed control. We evaluated the weedsmothering effect of green manure, including Crotalaria juncea, Vigna sinensis, Glycine max, Sesbania rostrata, and S. aculeata during 1992 and 1994 at the ARS, Mugad, Karnataka, India. The crop treatments were combined factorially with either hand weeding (HW) alone at 30 d after rice emergence (DE) or with intercultivation (IC) at 15 DE followed by hand weeding at 40

aHW = hand weeding; DE = days after rice emergence; IC = intercultivation.

Vol. 22, No. 1

47

ing water at 40 DE by wet IC and followed by passing a wooden plank between crop rows. A randomized complete block design with three replications was used. Rainfall amounted to 1150 and 1135 mm in 75 and 79 rainy days during the cropping seasons of 1992 and 1994, respectively. The rice cultivar was Abhilash (155 d). Common weeds were Echinochloa colona, Panicum spp., Ischaemum rugosum, Cyanotis spp., and Eclipta prostata.

The results (see table) indicated that intercropped C. juncea, V. sinensis, G. max, and S. rostrata, when combined with 1 IC at 15 DE and I HW at 40 DE, could suppress weeds effectively. The rice yields obtained in these plots were similar to those recorded after butachlor applications with 1 HW at 30 DE (see table), suggesting that these crops could replace butachlor applica-tion when combined with 1 IC. Sesbania aculeata, which grows very slowly, could not suppress weeds effectively.

Vigna sinensis had better weedsmothering ability and rice yields were comparable with those using butachlor with 1 HW at 30 DE, even when combined with 1 HW at 30 DE. Apart from their weed-smothering ability these legumes can fix atmospheric N and add a considerable amount of organic matter to the soil. Therefore, intercropping of C. juncea, V. sinensis, G. max, and S. rostrata may be encouraged for sustainable weed management in rainfed lowland rice.

Research methodology
Modified method for determination of amylose content using a single rice kernel
B. R. Swain and M. Nagaraju, Central Rice Research Institute, Cuttack, India
Amylose content of 11 varieties tested by two methods. Cuttack, India. Genotype Modified Amylose content 3.5 19.9 20.8 20.8 21.3 21.4 21.4 23.7 24.7 25.2 25.2 Type Waxy Intermediate Intermediate Intermediate Intermediate Intermediate Intermediate High High High High Juliano Amylose content 3.7 19.0 20.1 20.1 20.8 20.7 20.9 23.9 24.9 26.3 25.5 Type Waxy lntermediate Intermediate Intermediate Intermediate Intermediate Intermediate High High High High

Amylose content determines the stickiness and eating quality of cooked rice. We modified Shen's (1990) method to determine amylose content in a single rice kernel. We tested the method by analyzing 11 varieties having waxy, intermediate, and high amylose content. The final protocol involved the following steps. A single rice kernel was milled by a hand-set miller and the embryo carefully removed with a blade. The material was weighed on an Afcoset electronic balance (0.1 mg-180 g) and placed in a 50-mL graduated tube. A total of 0.1 mL of 95% ethanol and 1.8 mL of 1 N NaoH solution was added. The mixture was kept in a hotwater bath (35 C) for 20 h. It was shaken well and brought to 20 ml with distilled water. A 5 mL aliquot was pipetted into a 100 mL volumetric flask. One milliliter of acetic acid and 2 mL of iodine solution were added. The solution was brought to full volume with distilled water and shaken well. Color was read at 620 nm after 20 min in a UV

Malagkit Sungsong Basmati 370 Karnal local Basmati 113 T412 Pakistan Basmati Sitabhog Tulsimanjari Savitri Gayatri Ratna

1201 spectrophotometer. Concentration of amylose was determined from a standard curve using amylose (Sigma). According to the amylose content of control varieties Malagkit Sungsong (waxy), Basmati 370 (intermediate), and Ratna (high), we changed the classification of amylose to waxy (0-5), low (5-17), intermediate (17-22), and high (>22). The amylose content of the 11 varieties is shown in the table. Most of them were intermediate, four were high, and one was waxy. This method is suitable for determination of amylose content in a large number of samples for genetic studies. The results agreed closely with those obtained using Juliano's standard method.

Comparison of two scoring systems for evaluating stem rot resistance in rice, and a new proposed rating scale
R. Singh and D. S. Dodan, CCS, Haryana Agricultural University (HAU), Rice Research Station, Kaul 136021, Haryana, India

Genotypes judged susceptible by the Standard evaluation system for rice (SES, 1988) often fall into the resistant or intermediate category when scored using the rating scale by Jackson et a1 (1977). The SES scale is based on disease incidence (% infected tillers) and excludes few resistant genotypes.

48

IRRN 1997

Jackson's scale uses disease severity and groups some genotypes as resistant. A new six-point rating scale proposed here accounts for both incidence and severity by using a coefficient of infection (CI):
Score 0 1 2 3 4 5 Reaction Highly resistant Resistant Moderately resistant Moderately susceptible Susceptible Highly susceptible Description (CI%) No symptoms Less than 5% Above 5-10% Above 10-20% Above 20-40% Above 40%

Categorization of rice genotypes scored by different rating scales.

Reaction Resistant

SES (1988) Gobind, lR35546-17-3-1-3, KJT72-1-173-44, RYT3/13

Jackson et al (1977)

New proposed scale

Basmati 370, Gobind, HKR86-196, Gond, HKR86-403, HKR86-403, HKR90-403, HKR90-408, HKR90-408, HKR90-410, HKR90-410, HKR90-413, HKR90-421, HKR91-438, lR35546-17HKR90-424, HKR90-426, HKR91-403, 3-1-3, KTJ72-1-173-44, HKR91-410, HKR91-417, HKR91-419, RP2633-15-2-5, RYT3/13, HKR91-420, HKR91-422, HKR91-423, Taraori Basrnati HKR91-438, HKR91-452, HKR91-453, HKR91-455, HKR91-459, HKR91-465, HKR91-471, HKR238, Haryana Basmati No. 1, IET10918, lR9761-19-1, lR31785-58-1-2-3-3, lR35366-62-1-2-2-3, lR35546-17-3-1-3, IRON84-142, KJT72-1-173-44, NDR84, NDR85, NDR86, Pusa 33, Pusa Basmati No. 1, Ratna, RP1674-690-390-14, RP1681-170-4-204, RP2632-249-5-5, RP2633-15-2-5, RP2633-30-4-10, RYT3/13, Taraori Basmati, UPR756-2-1-3 HKR86-1,HKR86-217,HKR90-407, HKR91-405, HKR91-458, HKR91-462, HKR91-464, HKR91-468, HKR91-470, IRON86-171, KAU8754, KAU8759, KAU8770, KAU8772, NDR118, RP2235-113-35-20, RP2434-17-9-5, RP2632-334-2-5, RP2633-30-4-7, RP2633-110-11-2, RP2633-225-10-8 Basrnati 370, HKR90-403, HKR90-421, HKR91-403, HKR91-417, HKR91-422, HKR238, IET10918, IR9761-19-1, IRON84-142, NDR84, NDR85, Pusa 33, Ratna, RP1681-170-4-204, RP2632-249-5-5, UPR756-2-1-3 AS26556, HKR86-1. HKR86-196, HKR86-217, HKR90-407, HKR90-413, HKR90-424, HKR90-426, HKR91-405, HKR91-410, HKR91-419, HKR91-420, HKR91-423, HKR91-424, HKR91-452, HKR91-453, HKR91-455, HKR91-458, HKR91-459, HKR91-462, HKR91-464, HKR91-465, HKR91-468, HKR91-470, HKR91-471, Haryana Basrnati No. 1, lR1674-690-390-14, lR31785-58-1-2-3-3, lR35366-62-1-2-2-3, IRON86-171, KAU8754, KAU8759, KAU8770, KAU8772, MTU7991, NDR86, NDR118, NLR33055, Pusa 834-13-101, Pusa 835-5-2-101, Pusa Basmati No. 1, RP2235-113-35-20, RP2434-17-9-5, RP2632334-2-5, RP2633-30-4-7, RP2633-30-4-10, RP2633-67-7-9, RP2633110-11-2, RP2633-22510-8, TNAU851979

Intermediate

To use this proposed rating scale, tillers are classified into Jackson's categories: 1 = no symptoms, 2 = lesions confined to outer leaf sheath, 3 = lesions extending through the leaf sheath to outer culm, 4 = lesions penetrating the culm resulting in partial rotting of the culm, 5 = mycelium and /or sclerotia formed within the culm, stem completely rotten. These categories are then assigned numerical values of 0, 0.25, 0.5, 0.75, and 1.0, respectively, to give relative weight to the tillers falling in different categories so that computed CI values do not exceed 100%. A zero value is assigned to healthy tillers instead of the 1 value used by Jackson. The formula for determining CI is similar to that given by Jackson in determining the weighted disease index except that (a) numerical values of 0 to 1 assigned to the scales of 0 to 5 are used in computations and (b) the weighted disease index so obtained is multiplied by 100 to obtain % CI. The table shows the screening of a collection of 78 rice genotypes against stem rot by use of the three methods in 1992 and 1993. We found 4 resistant, 9 intermediate, and 65 susceptible using SES; 48 resistant, 21 intermediate, and 9 susceptible using Jackson's scale; and 10 resistant, 17 intermediate, and 51 susceptible using the new scale. Four resistant, 4 intermediate, and 45 sus-

Basmati 370, HKR91-417. HKR91-423, HKR91-438, IET10918, IRON84-142, RP2633-15-2-5, Taraori Basmati, TNAU85-19-79

Susceptible

AS26556, BK779-50, AS26556, BK779-50, HKR91-424, Haryana Basmati No. 1, MTU7991, NLR33055, Pusa 834-13-101, Pusa 835-5-2-101, HKR86-1, HKR86-196, RP2633-67-7-9, TNAU851979 HKR86-217, HKR9-403, HKR90-407, HKR90-408, HKR90-410, HKR90-413, HKR90-421, HKR9-424, HKR90-426, HKR91-403, HKR91-405, HKR91-410, HKR91-419, HKR91-420, HKR91-423, HKR91-424, HKR91-452, HKR91-453, HKR91-455, HKR91-458, HKR91-459, HKR91-462, HKR91-464, HKR91-465, HKR91-468, HKR91-470, HKR91-471, HKR238, lR9761-19-1, lR31785-581-2-2-3, lR35366-62-1-2-2-3, IRON86-171, KAU8754, KAU8759, KAU8770, KAU8772, MTU7991, NDR84, NDR85, NDR86, NDR118, NLR33055, Pusa Basmati No. 1, Pusa 33, Pusa 83413-101, Pusa 835-5-2-101, Ratna, RP1674-690-390-14, RP1681-170-4-204, RP2235-113-35-20, RP2434-17-7-5, RP2632249-5-5, RP2632-334-2-5, RP2633-30-4-7, RP2633-30-4-10, RP263367-7-9, RP2633-110-11-2, RP2633-225-10-8, TNAU851979. UPR756-2-1-3

Vol. 22, No.1 49

ceptible genotypes were found in common between the new scale and SES. Ten resistant, no intermediate, and 8 susceptible genotypes were in common between the SES scale and Jacksons scale. Six genotypes (HKR86-403, HKR 90-408, HKR90-410, HKR91-438,

RP2633-15-2-5, and Taraori Basmati) found resistant with the new scale had 7.4, 10.0, 9.1,7.4,5.3, and 7.3% of infected tillers. These entries were excluded from the resistant category by SES because they just crossed the resistance limit (<5% infected tillers).

A procedure for continuous screenhouse rearing of the yellow stem borer

L. M. Sunio, J. S. Bentur, and M. B. Cohen, IRRI

Studies of yellow stem borer (YSB), Scirpophaga incertulas, are often disrupted because of the great difficulty of rearing this species. We have developed a procedure to maintain colonies of YSB in a screenhouse. This procedure has provided us with a daily supply of adults for our research. Our screenhouse has a floor space of 56 20 m. The planting area is divided into 16 plots of 15 m 2. Seedlings of rice variety IR62 are grown and transplanted in each of the plots at 7- to 10-d intervals. Each plot has 11 rows of 32 hills planted at 20- 20-cm spacing. To initiate colonies, adult moths of YSB are collected from the field and caged on plants for oviposition. Neonate larvae hatching from these egg masses are used to infest the plots at 10-15 larvae hill-1 when the plants are 65-100 d old. At this stage of plant growth, each hill

has 12-15 tillers and 4224 tillers are infested in each plot. Four weeks after infestation, the plants are trimmed 40 cm above ground level to facilitate adult moth emergence. Plots of trimmed plants are covered with fine fiberglass mesh to collect adults. Moths are collected every day and caged on 40-d-old potted plants for oviposition. Pieces of leaves with egg masses are
Performance of YSB colonies at IRRI over time. a Dependent variable Colony Sex ratio Total adults produced

Colony A (n = 6 generations) Generation Plant age Colony B1 (n = 10) Generation Plant age Colony B2 (n = 7) Generation Plant age Colony C (n = 8) Generation Plant age

0.510 0.004 0.413 0.896b 0.903 0.606 0.212 0.260

0.307 0.119 0.364 0.427 0.711 0.552 0.329 0.644

aValues are correlation coefficients. b Significant (P<0.05).

removed 7 d after oviposition and kept in a container for hatching. The larvae can either be used for experiments or for reinfesting a fresh plot to maintain the colony. Four separate, overlapping colonies can be maintained by infesting every 4th plot with larvae from a single earlier plot. Alternatively, because moth emergence from each plot occurs over a period of 12-18 d, colonies can be continuously mixed by infesting each new plot with larvae from 2-3 previous plots. Between August 1994 and January 1996, we reared 4 separate, overlapping YSB colonies for 6-10 generations each, assuring an uninterrupted supply of 200-600 female moths each week. The overall mean number of moths produced and sex ratio (proportion of females among total adults) were 865.1 11.5 and 0.57 0.001, respectively. Mean developmental time was 39.2 0.04 d and was highly dependent upon seasonal temperatures. Colony performance was measured by total number of moths produced and the sex ratio. The table shows that performance was not correlated with the number of generations a colony had been in culture, indicating that colony performance did not deteriorate over time. In 3 of the 4 colonies, performance was not correlated with age of plants at the time of infestation (range 65-100 d after sowing), indicating that reproductive-stage plants of a wide range of ages were suitable for colony maintenance.

50 IRRN 1997

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. 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 succint 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: ecosystems as irrigated, rainfed lowland, upland, and flood-prone (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. of Units for measurements. For example, express yield data in metric tons per hectare (t ha -1) for field studies. Do not use local units of measure. 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. 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.

Use generic names, not trade names, for all chemicals. Use the International System

Specify the rice production

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 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.

Review of notes. The IRRN editor will send an acknowledgment card or an e-mail message when a note is received. An IRRI scient' 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. Comments. If you have comments or suggestions about the IRRN, please write to the editor. Mailing address. Send notes and correspondence to the IRRN Editor, IRRI, P.O. BOX 933, Manila 1099, Philippines. Fax: (63-2) 845-0606 E-mail: c.dedolph@cgnet.com Home page: http://www.cgiar.org/irri Riceweb: http://www.riceweb.org Riceworld: http://www.riceworld.org