New Approaches to Gall Midge
Resistance in Rice

Book is available

Edited by J. Bennett, J.S. Bentur, I.C. Pasalu, and K. Krishnaiah
2004

Suggested citation:
Bennett J, Bentur JS, Pasalu IC, Krishnaiah K, editors. 2004. New approaches to gall midge resistance in rice. Proceedings of the International Workshop, 22-24 November 1998, Hyderabad, India. Los Baños (Philippines): International Rice Research Institute and Indian Council of Agricultural Research. 195 p.

Rice gall midge, Orseolia oryzae—an overview
K. Krishnaiah

Rice gall midges Orseolia oryzae and O. oryzivora are important pests of rice in Asia and in Africa, respectively. Breeding resistant varieties has been a viable, ecologically acceptable approach for managing the pest. But diversity in pest population and rapid selection of virulent biotypes of the insect call for new approaches to studying gall midge resistance in rice. The International Workshop on New Approaches to Gall Midge Resistance in Rice took stock of the progress made so far and identified new thrusts for the future.

Biology and ecology of rice gall midge
S. Rajamani, I.C. Pasalu, K.C. Mathur, and Mangal Sain

Gall midge (Orseolia oryzae Wood-Mason) is an important pest of rice in Asia. It is essentially a monsoon pest and causes damage wherever higher humidity and moderate temperature prevail. Galls (silver shoots) occur generally during the tillering stage. The insect’s life cycle is completed in about 19-23 d, under normal conditions (at 22 to 28 °C with about 85% humidity). The sex ratio (male to female) is usually 1:3. Sex pheromone involvement has been observed.

During April and May, premonsoon rains in India usher in insect activity in rice stubbles, self-sown rice, and other hosts. The early planted rice crop is damaged extensively. Insect activity peaks between the last week of August and the first week of October. Graminaceous weeds (Leersia hexandra and Echinochloa crus-galli) and wild rice varieties (Oryza nivara, O. barthii, and O. rufipogon) serve as alternate hosts.

Rice gall midge has several natural enemies. Adoption of high-yielding resistant cultivars on a large scale has suppressed this pest, but the development of new biotypes may result in its resurgence. Consumers’ preference for susceptible rice cultivars continues to be an important factor in the prevalence of this insect.

Genetic conflict over sex determination in rice gall midge
S.C. Sahu, B. Kar, S.K. Behura, S. Nair, and M. Mohan

Insects belonging to the lower Diptera (suborder: Nematocera) have a bizarre type of sex determination and sex ratio of the progeny. The rice gall midge is known to produce unisexual progenies; the individual female produces all males or all females. An analysis of the mitotic and meiotic apparatus of the pest has shown that rice gall midge is karyologically dimorphic; males have six and females have eight somatic chromosomes. The haploid number of chromosomes is 4—evident from the somatic pairing of the polytene chromosomes in the salivary gland basal cells. As in other dipteran insects, the rice gall midge is characterized by having more chromosomes in germ line cells than in somatic cells, the presence of somatic pairing, lack of synapses in pachytene and no true bipolar anaphase I during spermatogenesis, highly irregular oogenesis, and elimination of chromosomes during early cleavage divisions of embryogenesis. Genetic conflict over sex determination is evident by eliminating the paternal set of chromosomes during spermatogenesis, highly irregular oogenesis, transmitting eliminated chromosomes to eggs along with the haploid set, elimination of sex chromosomes during early embryogenesis, and the presence of sex ratio-distorting endosymbiotic bacterium only in the female midges. Sex is thought to be female biased, contributed by nuclear and cytoplasmic factors, but compensatory male-determining genes have decisive roles to play.

Insect–host plant interactions in gall midge–rice
J.S. Bentur, S. Amudhan, I.C. Pasalu, N.P. Sarma, and U. Prasada Rao

The Asian rice gall midge Orseolia oryzae (Wood-Mason) (Diptera: Cecidomyiidae) is an important pest of rice that has been successfully managed through breeding and cultivation of resistant varieties. The development of leaf sheath gall, silver shoot, is the product of compatible interaction between insect and plant, wherein the insect establishes in the plant and develops to adult stage. The incompatible interaction between insect and resistant rice varieties manifests itself mainly as antibiosis in two distinct ways, i.e., with or without the expression of hypersensitive reaction (HR), both leading to insect mortality. At the species level, O. oryzae has a limited host range, while at the subspecies level biotypes are selected in specific response to antibiotic factors of resistant varieties. Interactions between biotypes and varieties appear to be on a gene-for-gene basis although this is not shown unambiguously. Resistance in many cultivars is governed by a single dominant gene, whereas virulence in biotype 4 against rice variety Phalguna with resistance gene Gm2 is inherited as a single recessive gene. Resistance in this variety against a virulent biotype can also be induced by prior infestation with an avirulent biotype.

Biology and host-plant relationships of the Hessian fly: past and current research
M.O. Harris, C.E. Williams, R.H. Ratcliffe, S.P. Foster, W. Griffin, H. Kanno, B.D. Morris, H.W. Ohm, R.A. Pickering , U. Rani, R.H. Shukle, J.J. Stuart, and J. Ziesmann

The Hessian fly is unique among insect herbivores in that many aspects of its biology and relationships with grass host plants have been studied in detail. As with many other holometabolous insects, there is a strict division of tasks between the adult and immature stages. The adult stage is short-lived and focus on reproductive tasks, finding virgin females in the case of the adult male, and attracting males and finding oviposition sites in the case of the adult female. Mate-finding and host-finding occur, respectively, when males fly upwind to a sex pheromone released by virgin females and when mated females orient to the visual stimuli originating from grasses. Host identification occurs after females land on a plant and examine the chemical and physical properties of the leaf surface. Although wheat, barley, and rye are the best-known host plants of the Hessian fly, larvae have been found feeding on 16 different grass genera. The reaction of grasses to Hessian fly larval feeding depends on the plant’s growth stage and includes seedling death, plant stunting, stem breakage, and reductions in grain quantity and quality. In wheat crops, planting of resistant cultivars is the most convenient and effective means of control. Cultivar resistance is conferred by major resistance genes that prevent the neonate larva from establishing at feeding sites at the base of the plant. A problem related to the deployment of resistant wheat cultivars is the evolution of adapted or virulent Hessian fly biotypes that can survive on cultivars that were formerly resistant. Because of adapted biotypes, breeding programs for Hessian fly include the monitoring of biotypes and screening for novel resistance genes. Several current Hessian fly research programs are briefly mentioned and contact addresses are provided for researchers.

Insect-plant relationships of the sorghum midge, Stenodiplosis sorghicola
H.C. Sharma

Sorghum midge, Stenodiplosis sorghicola (Coquillett), is one of the most important pests of grain sorghum worldwide, while plant resistance is one effective means of keeping its populations below threshold levels. Physical stimuli from viable pollen and receptive stigma influence oviposition by sorghum midge. Screening for resistance can be carried out using infester row and headcage techniques. Genotypes AF 28, TAM 2566, IS 2579C, IS 3461, IS 8891, DJ 6514, and IS 10712 are diverse sources of resistance. Several high-yielding cultivars such as ICSV 197, ICSV 745, ICSV 804, and ICSV 88032 with resistance to sorghum midge have been developed. Resistance to sorghum midge has also been transferred into male-sterile lines. Antixenosis to oviposition, antibiosis, and tolerance are important mechanisms of resistance. Short, tight, and hard glumes, initial faster rate of grain development, and tannin content of grain are associated with resistance to sorghum midge. Resistance to sorghum midge is largely governed by additive gene action, with some cytoplasmic effects. Resistance is needed in both parents to produce midge-resistant hybrids. Resistance to sorghum midge breaks down in Kenya. Temperature, relative humidity, and solar radiation influence the expression of resistance to sorghum midge.

Rice gall midge: pest status, distribution, and yield losses
K.C. Mathur and K. Krishnaiah

The rice gall midge Orseolia oryzae has been one of the major insect pests in many Asian countries. Its status as a major pest is influenced by rice ecology and prevailing climatic and agronomic conditions. Natural selection of resistance-breaking biotypes, cultivation of other insect-resistant varieties or popular local susceptible varieties, and the growing popularity of susceptible quality rice are some of the factors responsible for the recent upsurge in economic importance of this pest. Conservative estimates of yield loss from the pest were US$500 million in Asia and US$80 million in eastern and southern India alone. Experimental determination of pest damage influence on yield parameters suggested large variations due to varieties under study. Because of the internal feeding habit of gall midge maggots and the late appearance of damage symptoms, determination of the economic threshold level as a tool to initiate pest control measures has proven to be a formidable task.

Gall midge in Cambodian lowland rice
G.C. Jahn and K. Bunnarith

The rice gall midge, Orseolia oryzae, is a pest of lowland rice in portions of southern Cambodia. In June and July, gall midges emerge from wild rice and grasses to infest wet-season seedbeds. To a lesser extent, gall midges also infest dry-season seedlings in December to February. Other than rice and wild rice, several grass species found in Cambodia possibly serve as alternate hosts to the gall midge, including Echinochloa oryzoides, E. crus-galli, E. colona, Leersia hexandra, Ischaemum rugosum, Panicum repens, Brachiaria mutica, Paspalum dilatatum, and P. distichum. Among the parasites of gall midge in Cambodia are the Hymenoptera Platygaster sp. and Propicrocystus mirificus. More than 90% of gall midge pupae and prepupae collected from galls are parasitized in any given wet-season field. Predators of the gall midge in Cambodia include the spiders Tetragnatha mandibulata, Argiope catenulata, and Araneus inustus.

Status of gall midge in Lao PDR
S. Inthavong, J.M. Schiller, V. Sengsoulivong, and P. Inthapanya

The rainfed lowland rice ecosystem accounts for approximately 70% of the national rice area and 78% of production in Lao PDR. More than 90% of the rainfed lowland area is sown to glutinous genotypes. In provinces of the central and southern agricultural regions, improved varieties are grown on more than 60% of the area. In the northern region, improved varieties occupy less than 20% of the cultivated area. Gall midge damage is the most consistently reported cause of yield loss due to insects in this environment. Attempts to demonstrate economic yield loss from gall midge in areas with a history of the problem have generally not been successful, except on a localized basis. The significance of the problem varies from year to year and from locality to locality. Factors responsible for the variability are not yet understood. DNA typing (by IRRI) of gall midge samples collected from Savannakhet Province in the central agricultural region has shown similarities with those of Guangdong Province in southern China, and Manipur State of northeastern India. There are no known rice varieties in Lao PDR with resistance to the local gall midge biotypes. Differences in relative susceptibility of varieties currently grown, however, appear to exist. Surveys have established high populations of efficient natural enemies (egg, larval, and larval-pupal parasitoids) that should be capable of regulating the gall midge population in the natural environment.

Rice gall midge in Thailand: current status and biotype characterization
C. Tayathum, T. Attathom, D. Thongphak, and K. Sripongpankul

One constraint to rice production is gall midge. Most Thai-recommended and local commercial varieties are susceptible to gall midge. Consequently, severe rice gall midge infestation has been reported from many locations throughout the country, especially in the north, northeast, and central plain. Resistant varieties such as Muey Naung 62M, RD4, and RD9 have now become susceptible to some rice gall midge populations collected from various locations. Therefore, to assess the differential reaction of rice gall midge, phenotyping and DNA analysis were conducted. The results of five years (1991-95) of studies based on the reaction of differential varieties and phenotyping indicate the possibility of many rice gall midge biotypes in Thailand. The reaction of RD9 to gall midge populations from Ubon Ratchathani (northeast region) and Chachoengsao (central plain) indicated that these populations contain different biotypes. The reaction of Muey Naung 62M, however, was similar for gall midges from these locations and also from Nan (north region). Biotype characterization of Thai rice gall midge was made using the random amplified polymorphic DNA (RAPD) method. Fifteen populations of the rice gall midge were collected from 14 provinces in different geographic areas of Thailand and were compared using presence-absence data 50 polymorphic bands amplified by eight primers. UPGMA analysis for indicated genetic clustering of presence-absence data populations into three distinct biotypes, which corresponded with geographical distance. Similarities between populations from the central plain and those from the north and northeast were observed. This may suggest an evolutionary history of the rice gall midge originating in the central plain area of Thailand followed by dispersion to the north and east. Data from this study could form the basis for a national and international breeding program on gall midge-resistant varieties that will benefit all rice-growing countries in the region.

Genetics and breeding for gall midge resistance in India
U. Prasada Rao, J.K. Roy, I.C. Pasalu, and K. Krishnaiah

Sources of gall midge resistance in the primary pool of rice germplasm have been known for a long time. The systematic evaluation during the past three decades of more than 20,000 accessions of rice germplasm both in the field at pest-endemic locations and in greenhouses under artificial infestation has identified more than 250 sources of resistance. The majority of these are landraces from the northeastern states of India. Some accessions of wild species of the rice genus Oryza are also known to be resistant to the pest. After the detection of gall midge biotypes within the country, resistant germplasm accessions were screened against characterized biotypes to note the range in resistance. Multilocation testing of primary donors against all known biotypes revealed that none of the donors displayed resistance against all six biotypes. Only one of the donors tested, Orumundakan (mutant), showed resistance to five biotypes, whereas 10 donors had resistance to four biotypes. Studies on inheritance of resistance have, in general, indicated that resistance is governed by major genes. So far, five dominant genes and one recessive gene have been well characterized. Of these, one gene, Gm6(t), is effective and confers resistance against all four biotypes in China but is not effective against Indian biotypes. An additional five genes are suspected to confer biotype-specific resistance in recent studies. The systematic breeding for gall midge resistance in India was boosted with the introduction of high-yielding varieties in 1965. Based on the performance of cultivars in multilocation testing, five cultures with built-in resistance to gall midge were released as varieties for commercial cultivation in the country during the 1970s, 28 during the 1980s, and more than 15 since then. The majority of these varieties carried a single gene from three known resistance sources. These varieties made a big impact on yield advantage and lowering of pest incidence during the first decade of cultivation. Subsequently, the development of virulent biotypes accompanied by a breakdown in resistance is being reported in many pest-endemic regions in the country. The initial response of breeders to this situation has been to look for alternate sources of resistance against new biotypes and incorporate the resistance into local popular varieties. Deploying a single gene in any variety and its wide cultivation, however, would only lead to a new cycle of boom and bust as evidenced in the past. Some alternative and promising strategies are being considered for the future.

Breeding for Asian rice gall midge resistance in China
Bing-Chao Huang, Yang Zhang, Yu-Jan Tan, Yan-Kang Xu, Zhen-Wen Xie, Li-Xia Zhao, Jian-Wei Chen, Hong Li, Jian-Yuan Yang, Li-Hua Lu, S.K. Katiyar, N. Huang, S.V. Constantino, G.S. Khush, and J. Bennett

Asian rice gall midge Orseolia oryzae Wood-Mason (GM) is one of the major insect pests in South China with several outbreaks in recent years. In this paper, screening and evaluation of germplasm against GM, biotypes of GM, breeding for GM resistance in China, tagging and mapping of the new resistance gene Gm6, and genetic analysis of gene Gm6 are reviewed. Among the resistant resources from China, cultivar Yangshanzhan was resistant to biotype 1 but susceptible to biotype 4. Cultivar Daqiuqi was confirmed as an excellent resistance donor for breeding, with resistance to four biotypes of GM identified in Guangdong and in other provinces. Resistance in Daqiuqi was transferred into five new resistant varieties, including cultivars Kangwen 2, Kangwen 5, Duokang 1, Duokang 2, and Kangwenqingzhan. These five varieties were released and tested in 50 locations in five provinces of South China. They showed high resistance against GM. The resistance mechanism is antibiosis. From 1993 to 1997, the Guangdong Academy of Agricultural Sciences (GAAS) and the Asian Rice Biotechnology Network (ARBN) of IRRI conducted a project to further analyze the resistance gene in Duokang 1 against the four biotypes of Asian rice gall midge in South China. A single dominant gene tentatively named Gm6 in 1994 governs the resistance to biotypes 1 and 4 in this variety. Gene mapping was carried out on 160 F3 progenies of hybrid combination Duokang 1/Fengyingzhan 1 (susceptible). Based on the principle of bulked segregation analysis (BSA), the resistant gene Gm6 was tagged successfully for the first time by applying the RAPD marker OPM6 (1.4 kb). After this RAPD product was labeled with 32P and used as a probe to detect the referential mapping population in IR64/Azucena, the resistance gene was mapped between RFLP markers RG214 and RG163 on the fourth rice chromosome. Furthermore, research results indicated that the gall midge resistance gene of Duokang 1 was not allelic to all of the gall midge resistance genes of W1263 (Gm1), Leuang 152 (Gm2), BG404-1 (Gm3), and OB677 [gm4 (t)]. It was confirmed that Gm6 is a gene with resistance to rice gall midge. Gm6 was transferred into the restorer line RL301 by the cross combination Minghui 63//Changzhan/Duokang 1. Hybrid Shanyou 301 crossed to Shan A/RL301 was bred successfully in GAAS. Future research directions on GM are also suggested.

Breeding for resistance to African rice gall midge, Orseolia oryzivora Harris and Gagné
B.N. Singh, C.T. Williams, M.N. Ukwungwu, and A.T. Maji

African rice gall midge (AfRGM), Orseolia oryzivora, is indigenous to Africa. It was first reported from southern Sudan in 1947, and since then it has been reported from 20 countries south of the Sahara. Severe outbreaks and yield losses have been reported. Infestation by AfRGM is more common in humid and Guinea savanna zones, but it also occurs in the Sudan savanna zone. Oryza longistaminata and other wild rice species are the main alternative host for AfRGM in the dry season. Field screening for host resistance began in 1982 at Edozhigi in Nigeria, and at Karfiguela in Burkina Faso. Screening has shown that most of the cultivars resistant to Asian gall midge are susceptible to African rice gall midge. Since there is a lack of strong antibiosis and antixenosis resistance for AfRGM in O. sativa, tolerant and moderately resistant lines were used for hybridization in the breeding program. In early 1998, Cisadane (an Indonesian cultivar) was released as FARO 51 in Nigeria. It is tolerant of AfRGM. Another line, BW 348-1 from Sri Lanka, has been identified as tolerant of AfRGM and iron toxicity, and is undergoing on-farm testing in Nigeria. O. glaberrima lines were observed as immune to AfRGM under field screening in Nigeria. From screening of African O. sativa germplasm collections at the International Institute of Tropical Agriculture (IITA) Genetic Resources Unit, one indica line (TOs 14519 from Gambia) showed high resistance to AfRGM. It is a tall rainfed lowland plant type. Based on host-plant reactions in the field, there seem to be at least two biotypes of AfRGM in West Africa. A genetic study of host-plant resistance in O. glaberrima lines (TOg 7206 and TOg 7442) is ongoing. Crosses are being made to incorporate resistance from TOs 14519 into released cultivars and elite breeding lines. Evidence for biotypes of AfRGM suggests that breeding and selection activities should be carried out at different sites for different biotypes.

Current status of rice gall midge biotypes in India and China
I.C. Pasalu, Bing-Chao Huang, Yang Zang, and Yu-Juan Tan

Biotypes of Asian rice gall midge Orseolia oryzae have been reported and distinguished based on the reaction of differential rice varieties. In India, six distinct biotypes have been characterized, whereas in South China four biotypes are recorded. A distinct link between the resistance gene in the host plant and biotypes exists. Based on reaction against resistance genes Gm1 and Gm2, Indian biotypes designated as 1, 2, 3, and 4 show a similar pattern of reaction to that displayed by Chinese biotypes. Extensive cultivation of resistant varieties and mass migration appear to be the causes of biotype evolution.

Pyramiding of gall midge resistance genes in rice: different approaches and their implications
S.K. Katiyar, S.B. Verulkar, G. Chandel, Y. Zhang, B. Huang, and J. Bennett

The Asian rice gall midge (Orseolia oryzae) is a major pest across South and Southeast Asia. Several genes that condition resistance to gall midge have been identified in cultivated rice (Oryza sativa) and are being used in rice improvement programs of different countries. Some of these genes, however, are ineffective because of the evolution of new gall midge biotypes. It is therefore important to manage known sources to maximize the effectiveness and durability of genes, as there is no guarantee that new resistance sources will be identified at appropriate times in the future. Gene pyramiding in rice has already been done to incorporate wide-ranging and durable pest resistance. Increased understanding of genetics, allelic relationships, and linkage is necessary for pyramiding of resistance genes. Gall midge biotypes differing in their reactions to host resistance genes offer an excellent tool for pyramiding of gall midge resistance genes of dissimilar effect with or without the use of DNA marker technology. For pyramiding genes with similar effect, however, DNA markers are necessary. We have already demonstrated the pyramiding of two gall midge resistance genes (Gm-2 and Gm-6t) using a unique host-pest interaction approach. Pyramid lines with desirable alleles of both genes are ready for field evaluation against different biotypes in India and China. To speed up the gene-pyramiding process and to facilitate durable resistance for gall midge, possibilities of using various strategies separately or in combination are discussed.

Durable deployment of gall midge-resistant varieties
M.B. Cohen, J.S. Bentur, and F. Gould

Through the use of a simulation model, we exa.mined the durability of various deployment strategies for gall midge-resistant rice varieties in farmers’ fields. The strategies compared included sequential release of varieties with a single resistance gene, release of two genes “pyramided” in a single variety, and seed mixtures composed of a susceptible variety and a single-gene or pyramided variety. We also examined the effect on each strategy of varying levels of dominance of gall midge virulence alleles, the initial frequency of virulence alleles in gall midge populations, and epistasis between rice resistance genes. Two important results of the simulations are that (1) release of a variety with two pyramided resistance genes will generally provide more total years of resistance to gall midge than will sequential release of two genes in separate varieties, and (2) in most cases the incorporation of a susceptible variety into a seed mixture will greatly prolong the durability of resistant varieties. The release of a new resistant variety as a seed mixture containing some seeds of a susceptible variety might not be readily accepted by plant breeders or farmers, but this difficulty should be balanced against the potentially substantial increases in the effective lifetime of genes for gall midge resistance.

Mapping and cloning avirulence genes in insects
D.G. Haeckel and L.J. Gahan

Many of the interactions between plants and their pathogens appear to be controlled by gene-for-gene systems, in which a positive reaction occurs only in the presence of specific genes in the plant and in the pathogen. Here we describe genetical approaches to analyzing this interaction for rice gall midge on rice that can eventually lead to cloning the genes controlling virulence in the insect. As an example of the methodology that could be applied, we described our work on the mapping and positional cloning of genes conferring resistance to Bacillus thuringiensis endotoxins in the tobacco budworm Heliothis virescens. We further described strategies of comparative insect genomics that could be designed to maximize the benefit of similar studies in more closely related species, such as Hessian fly on wheat.

Posters

Poster 1
Emergence of rice gall midge biotypes in Andhra Pradesh—an overview
A. Ganeswara Rao, P.R.M. Rao, P. Rama Mohan Rao, P. Seshagiri Rao, and C. Srinivas Acharya N.G. Ranga Agricultural University (ANGRAU), Hyderabad 500030, Andhra Pradesh, India

Andhra Pradesh State ranks as the sixth largest in rice-growing area and is among the top three rice producers in India. Although improved and high-yielding rice varieties and technologies have helped increase yields in the state for a certain period, rice production has again stagnated in recent years. Among the different factors responsible for the plateauing of rice yields are pests and diseases. Gall midge (GM) is one of the destructive pests of rice widely distributed in Andhra Pradesh. In the southern Telangana region and the north coastal districts, it was a major production constraint until the 1970s. The state witnessed severe pest epidemics seven times from 1947 to 1954. Subsequently, with GM-resistant varieties covering 0.6 to 0.7 million ha of rice, GM was almost eliminated from this region. Most of these varieties were totally resistant to GM with one or two major genes governing resistance. This vertical resistance seemed to have exerted considerable selection pressure over time on the pest population, resulting in the emergence of virulent biotypes.

The widespread and severe outbreak of GM six times during 1986-93 in epidemics in the north coastal region showed that the most popular resistant varieties such as Phalguna and Surekha were totally susceptible to the new population, which was identified as biotype 4. In the early 1990s, the Siam 29-derived GM-resistant varieties that were resistant to biotype 1 became susceptible to the GM populations of northern Telangana. This population has been recognized as biotype 3. In addition, the GM population of southern Telangana, recognized as biotype 1, has now spread to new areas along the southern coast and is increasing in the Godavari-Krishna river deltas in both the wet (kharif) and dry (rabi) seasons, with the latter being more severe. The existing GM populations show consistent field reactions against the 10 differentials in four groups and exhibit the patterns R-R-R-S for biotype 1, S-S-R-S for biotype 4, and R-S-R-S for biotype 3.

Pioneering work was done at the Warangal station of ANGRAU in identifying GM-resistant donors and evaluating germplasm. Many GM-resistant varieties were bred and released for cultivation in the state. The history of insect pest-host-plant relationships, however, underscores the need to review and reorient the breeding program for GM resistance. Varieties with moderate resistance (60-70%) and with multigene resistance (horizontal resistance) to GM and to other pests are needed to preserve resistance over a long time and prevent emergence of more virulent biotypes.

Poster 2
Rice gall midge Orseolia oryzae—an emerging problem in the Visveswaraya Canal tract of Karnataka
Gubbaiah and H.P. Revanna Regional Research Station (RRS), VC Farm, Mandya 571 405, Karnataka, India

Rice gall midge, Orseolia oryzae, is one of the production constraints in the coastal belt of Karnataka, especially for the wet-season crop. In recent years, late release of irrigation water from reservoirs necessitated late planting and caused a moderate level of gall midge infestation in the Visveswaraya Canal (VC) tract of Karnataka during the 1994-97 wet seasons. Gall midge incidence in late-planted experimental plots at the Regional Research Station, VC Farm, Mandya, and in farmers’ fields was 92% and 100% hills, with 25% and 35% tillers, respectively, during 1996-97. Thus, gall midge has assumed the status of a major pest in the VC tract. The shift in pest scenario is attributed to late planting of susceptible rice varieties such as Jaya, IR20, Madhu, and IR64 (beyond the second fortnight of August); increased temperature and increased relative humidity; and prevalence of cloudy weather coupled with drizzle during the first 30 d after planting. This level of infestation is likely to cause a 10% yield loss in the VC tract area if proper care is not taken.

In the gall midge-endemic areas of coastal Karnataka, the early planted crop normally escapes pest attack, but the late-planted crop is severely infested. Many gall midge-resistant rice cultivars such as Shakti, Phalguna, Mahaveer, Nethravathi, MO4, and Latha have been identified and released for the endemic coastal belt. These resistant varieties have been accepted by coastal farmers and now occupy 60% of the area. Of late, however, the popular gall midge-resistant cultivar Phalguna has shown a moderate level of gall midge infestation (19% silver shoots), indicating the possible emergence in Karnataka of a new biotype in addition to the prevalent biotype 2.

Research to evaluate rice varieties with a moderate level of gall midge resistance under late planting for the VC tract is ongoing. Timely planting, growing resistant varieties, and protecting susceptible varieties using granular insecticide(s) in both nurseries and the transplanted crop are suggested as integrated gall midge management practices for Karnataka.

Poster 3
Rice gall midge Orseolia oryzae as an imporant pest in Kerala
D. Ambikadevi, N. Remabai, R. Devika, and K.P.V. Nair Rice Research
Station (RRS), Moncompu, Thekkekara, Alapuzha 688503, Kerala, India

With 1.1% of the country’s area, Kerala has to support about 3.4% of the population. Rice area and production, which were steadily increasing until the mid-seventies, succumbed to economic pressure from other remunerative crops during the last two decades, resulting in a decline of more than 0.3 million ha of land under paddy cultivation. The total rice area in Kerala during 1996-97 was 0.472 million ha (0.187 million ha during the wet season and 0.285 million ha during the dry season), while production was 0.953 million t. Average productivity peaked at 2.0 t ha–1 in 1992-93 and decreased to 1.9 t ha–1 in 1995-96. Major biological constraints to achieving higher and more stable productivity include the incidence of pests such as gall midge, stem borer, brown planthopper, and leaf blight.

With the introduction of high-yielding varieties, gall midge has attained the status of a major pest in Kerala. The Kuttanad region of Kerala—comprising Alleppy district and parts of Kottayam and Pathanamthitta districts—is an area endemic to this pest. Gall midge is also a major pest in Palghat, Trichur, and Thiruvananthapuram districts. During the 1990 wet season (April-May to August-September), the rice crop in Kuttanad suffered extensive damage. Earlier, gall midge was a problem only in the wet-season crop, but, over the last few years, the pest has attacked the dry-season crop with equal severity. It has also shown adaptability to climatic conditions prevailing during the period. A severe outbreak of gall midge in the 1996-97 dry season (October-November to December-January) resulted in a 90% crop loss (worth 80 million rupees) in Kuttanad.

Variety Bhadra released in 1978 from RRS, Moncompu, showed tolerance for gall midge during the 1996-97 dry-season outbreak. In 1998, three medium-duration, red kernel varieties—Pavithra, Panchami, and Uma—all with a yield potential of 5-6 t ha–1 were released from RRS, Moncompu. The results of the coordinated biotype studies at RRS, Moncompu, and RARS, Pattambi, under the All-India Coordinated Rice Improvement Project proved the emergence of the virulent gall midge biotype 5 in Kerala. Farmers’ preference for short-duration, red kernel varieties with long slender grains, low market value, lack of sufficient gall midge-resistant material for farmers, and the absence of a full-fledged surveillance program covering the entire endemic area are some problems constraining the implementation of integrated pest management for gall midge in Kerala.

Poster 4
The rice gall midge problem in Orissa
A.N. Dash All-India Coordinated Rice Improvement Project (AICRIP), Regional Research Station (RRS), Chiplima, Sambalpur 780002, Orissa, India

Rice in Orissa is grown on more than 4 million ha. It is cultivated on 68% and 7% of the gross cultivated area in the wet (kharif) and dry (rabi) seasons, respectively. During 1995-96 rice production was 6.23 million t and productivity was 1.4 t ha–1. It was projected that production must reach 7.15 million t and productivity increase to 1.6 t ha–1 by 2000 to meet the growing food demand.
Of the various insect pests attacking rice, gall midge Orseolia oryzae (Wood-Mason) is considered to be a regular and major pest throughout Orissa. The Sambalpur district in the west-central tableland zone is considered to be the endemic pocket for gall midge in the state. The hot and humid climate, rice-rice croppings, and application of higher doses of fertilizer have favored gall midge incidence in this zone. Moderate to high pest incidence is also observed in coastal districts such as Puri, Cuttack, Balasore, and Ganjam, and in Bolangir, Keonjhar, and Dhenkanal districts. Low pest incidence is observed in the remaining districts of the state. Results of experiments under AICRIP at RRS, Chiplima, for the last 10 years revealed that 25% of the average yield loss was due to insect pests, of which gall midge contributed a major share.
Many resistant and moderately resistant varieties are being used by the farming community in the state to reduce gall midge damage considerably. These include Heera, Kalinga-II, Neela, Tara, Khandagiri, Udaya, Daya, Gouri, Pratap, Shakti, Phalguna, Meher, Birupa, Bhanja, and Samanta for medium lands and Samalei, Manika, and Urbashi for lowlands. Monitoring the prevalence of biotypes in the state revealed the existence of biotype 1 at Sambalpur and biotype 2 in coastal districts Puri and Cuttack.

The use of moderately resistant varieties and effective granular insecticides both in the nursery and in the main field has considerably reduced pest damage in recent years. For effective pest management, emphasis should be given to surveillance and determination of an accurate ETL (economic threshold level). Continued breeding of suitable resistant varieties for different rice ecosystems, developing appropriate cultural management tactics, and protecting natural enemies by applying suitable selective pesticides on a need basis should be the focus of future research.

Poster 5
The rice gall midge problem in Madhya Pradesh
B.C. Shukla, R. Gupta, and U.K. Kaushik Indira Gandhi Agricultural University (IGAU), Raipur (M.P.), India

Rice is cultivated on about 5.4 million ha in Madhya Pradesh, India. Of this, the district of eastern Chhattisgarh accounts for about 70%. But the productivity of this area is only 1.1-1.2 t ha–1. Many constraints have been identified to contribute to such a low productivity, including weeds, insect pests, and diseases. Among insect pests, gall midge, stem borer, leaf- and planthoppers, and leaffolder are major concerns. Since the first report of economic damage by gall midge in Raigarh district in 1969, the pest has spread to Dhamtari, Raipur, Bilaspur, and Durg districts and resulted in epidemics in the late 1970s. Losses of about Rs. 2,000 million (~US$47.6 million) were estimated in 1977-78. No insecticide was found to be effective in controlling this pest. Resistant varieties or donors for resistance such as Surekha, RPW 6-17 (Phalguna), and IR36 were identified and were grown in some areas. Later, Asha, Usha, Ruchi, Abhaya, and Mahamaya were bred by IGAU. These varieties, with 5-6 t ha–1 yield potential, occupy about 15–20% of the rice area. Continuous drought or low rains in the last 5–6 y resulted in low gall midge populations, and farmers have shifted back to cultivating susceptible high-yielding varieties such as Kranti, Annada, and Swarna. For the last 7–8 y, differentials such as Phalguna (group II) and CR-MR 1523 (group III) have been showing low levels of susceptibility, which may indicate that there might be some biotypic shift under way in the Raipur gall midge population. Gall midge is now a major pest to combat in the area.

Poster 6
Alternate hosts of rice gall midge Orseolia oryzae (Wood-Mason)
M. Sain and M.B. Kalode Directorate of Rice Research, Rajendranagar, Hyderabad 500030, Andhra Pradesh, India

Several weed species have been reported as alternate hosts of rice gall midge Orseolia oryzae in India and other parts of the world. Some workers reported successful cross-infestations of gall midges from weeds to rice and vice versa, while others have recorded failure of such cross-infestations. Studies were thus undertaken under greenhouse and field conditions at the Directorate of Rice Research, Hyderabad.

Regular surveys in and around Hyderabad revealed only three weeds, Paspaladium geminatum, P. flavidium, and Cynodon dactylon with pronounced galls. But midges emerging from these weeds were identified by the Commonwealth Institute of Entomology as Orseolia sp.1 and Orseolia sp. 2. Gall midges emerging from C. dactylon and P. geminatum failed to establish and produce galls in rice plants. Of the 24 weed species tested in the greenhouse as plants grown from seeds, only in Echinochloa crus-galli, Leersia hexandra, Ischaemum rugosum, and wild rice Oryza nivara did the rice gall midge survive and produce galls. The pests, however, laid eggs on all weed plants, and maggots were observed at the growing point 5 d after oviposition. Further, adults from E. crus-galli and O. nivara infested rice plants again. When grownup plants of these species from the field were tested, however, none of the weed species could support O. oryzae development. The weeds in general received significantly fewer eggs compared with rice plants in the choice study. E. crus-galli could serve as a potential alternate host for both Orseolia oryzae from rice and Orseolia fluvialis from the weed P. geminatum.

Poster 7
Identification and molecular mapping of new genes for resistance against rice gall midge
A. Kumar, B.C. Shukla, M.N. Shrivastava, Indira Gandhi Agricultural University (IGAU), Regional Agricultural Research Station (RARS), Bilaspur 495001; Raipur 492012, Madhya Pradesh, India; S. Nair, and Madan Mohan, International Centre for Genetic Engineering and Biotechnology (ICGEB), Aruna Asaf Ali Marg, New Delhi 110067, India

Rice gall midge is one of the most destructive pests in South and Southeast Asian countries including India. The insect is endoparasitic; thus, the use of resistant varieties is the most economical and feasible approach to its control. Identifying diverse sources of resistance is necessary to develop resistant varieties with different genetic backgrounds to effectively counter biotypic variation in gall midge populations. The systematic work on the genetics of gall midge resistance at IGAU has led to the identification of five diverse genes [Gm1, Gm2, gm3(t), Gm4(t), Gm5(t)] for gall midge resistance. Gm2 has been mapped for its location on a rice chromosome by ICGEB, New Delhi, in collaboration with the Directorate of Rice Research, Hyderabad. Similarly, the marker linked to the Gm4 resistance gene has also been mapped at ICGEB, in collaboration with IGAU. A further genetic study on gene identification revealed the presence of a nonallelic resistance source in rice line RP 2333-156-8 (Ratna/ARC 10659). The gall midge resistance gene in this line was different from the other known genes. The new gene has been designated as Gm6(t) and molecular mapping of this new gene is in progress at ICGEB under a collaborative program with IGAU
.

Poster 8
Reversal of dominance of resistance against gall midge in rice
J. Pani, L.K. Bose, and S.C. Sahu Rice Genome Laboratory, Central Rice Research Institute, Cuttack 753006, India

Reversal of dominance is a unique phenomenon of gene interactions. Gall midge resistance in rice is known to be controlled by a few dominant or recessive genes. Cutting plants at 40 d after transplanting (DAT) has been recommended in screening against gall midge. Studies on inheritance of resistance to biotype 2 in five rice cultivars (Velluthacheera, T1477, ARC5984, ARC5158, and Phalguna) versus two susceptible cultivars (Jaya and IR20) showed reversal of the dominant effect of the resistance gene when F2 plants were cut at 40 DAT. Crosses with Velluthacheera and T1477 showed an F2 segregation ratio of 3:1 and 11:5 (resistant:susceptible), respectively, before cutting but a ratio of 1:3 and 7:9, respectively, after cutting. The F2 segregation ratio in crosses with the other three resistant cultivars remained unchanged. There was no change in the reaction of parents before and after cutting. The hypersensitivity test on parent plants showed that Phalguna, ARC5984, and ARC5158 were strongly positive, while Velluthacheera and T1477 were weakly positive with larvae surviving on secondary tillers and susceptible parents not expressing a hypersensitive reaction. Cutting the plants may have resulted in more tertiary tillers and hence enhanced gall development in heterozygous F2 plants of crosses with Velluthacheera and T1477. Two gene pools may be controlling gall midge resistance in rice, one for gall development and the other for larvae response to hypersensitivity.

Poster 9
Biological observations on rice gall midge biotypes
J.S. Bentur, S. Amudhan, and M. Rajasri Directorate of Rice Research (DRR), Rajendranagar, Hyderabad 500030, India

The Asian rice gall midge Orseolia oryzae (Wood-Mason) is known to rapidly evolve into biotypes that are capable of overcoming host-plant resistance in popularly grown GM-resistant varieties. At DRR we successfully reared as many as five different biotypes and studied their biology and response to rice cultivars under greenhouse conditions.

Developmental duration of different biotypes for the adult to adult stage was mainly influenced by ambient temperature, but during the same time of the year biotypes 4, 5, and 6 took longer to develop than biotype 1. Interbiotypic matings were not generally successful. Biotype 6 adults (originally from Manipur) did not mate successfully to produce fertile eggs when caged with biotype 1 or 4 adults. Likewise, biotype 5 insects from Kerala did not mate with adults of biotype 1 or 4. Successful mating, however, was observed between adults of biotype 4 or 3 and those of biotype 1, which are closely distributed within Andhra Pradesh.
Reactions of the four standard sets of differential rice varieties against different biotypes under greenhouse conditions at Hyderabad were similar to those noted under field conditions at sites where the population originated. Thus, the R-R-R-S pattern of reaction was observed in the biotype 1 population native of Hyderabad, R-SR-S against the biotype 3 population of Jagtiyal, S-S-R-S against biotype 4 of Srikakulam and Sakoli, R-R-S-S against biotype 5 of Moncompu, and R-S-S-S against biotype 3M (now 6) of Imphal. Biotype 2 insects prevalent in parts of Orissa were not covered in these studies.

Poster 10
Genetics of gall midge resistance in rice
U. Prasad Rao, V.R. Manoj Kumar, and S. Amudhan Directorate of Rice Research, Rajendranagar, Hyderabad 500030, India

The antibiosis mechanism of resistance against rice gall midge Orseolia oryzae manifests in two distinct modes in resistant rice varieties. In one, there is pronounced tissue necrosis, which is a typical hypersensitive reaction (HR) accompanied by maggot mortality. This is referred to as HR positive [HR(+)]. The other case does not involve HR and is referred to as HR negative [HR(-)] reaction. Phalguna with the Gm2 gene is a typical HR(+) rice variety, whereas W1263 with the Gm1 gene is a typical HR(-) rice variety. Thus, it appears that the two types of gall midge resistance are conferred by two nonallelic sets of genes. We studied the inheritance pattern of resistance and hypersensitive reaction in crosses involving both types of donors and susceptible parents. Our results showed that the gene for HR(+) resistance is epistatic to HR(-) resistance in some crosses studied.

Resistance against gall midge biotype 1 was further studied in F1, F2, and F3 generations of crosses involving genetically characterized donors and new sources of resistance. While the gene in NHTA8 was allelic to Eswarakora (Gm1), that in S2204 was allelic to Phalguna (Gm2). The single dominant gene observed in Orumundakn, T1432, T1477, Aganni, and Banglei was allelic to that in Bhumansan, but segregated independently of Gm1 or Gm2.