Cookies on Plantwise Knowledge Bank

Like most websites we use cookies. This is to ensure that we give you the best experience possible.


Continuing to use means you agree to our use of cookies. If you would like to, you can learn more about the cookies we use.

Plantwise Knowledge Bank

Your search results

Species Page

rice stem gall midge

Orseolia oryzae


You can pan and zoom the map
Save map

Host plants / species affected

Main hosts

show all species affected
Oryza sativa (rice)

List of symptoms / signs

Growing point - dwarfing; stunting
Leaves - abnormal colours
Leaves - abnormal forms
Leaves - abnormal forms
Leaves - internal feeding


The most obvious symptom caused by O. oryzae is the development of 'onion shoot' galls in growing crops. These galls are easily seen, especially as they often have a light-coloured silvery appearance ('silver shoots') and are quite unlike the symptoms produced by other pests or diseases. There is a superficial resemblance to 'dead heart' symptoms produced by lepidopterous stem borers, but the midge galls are easily distinguished as they are hollow, elongate cylinders, not dead rolled leaves. When midge infestations are at a low level there will not be any obvious effect on the incidence of ear-head formation, but severe infestations will stunt plants, suppress heading and enhance tillering.

Prevention and control


The literature on control measures against O. oryzae is extensive and the emphasis placed on the various methods available has changed with time. Current thinking favours the development of host-plant resistance, combined with cultural and biological measures to restrict midge populations below economic injury levels.

Cultural Control

Many cultural measures have been proposed (flooding or draining rice fields; manipulating planting dates; eliminating alternative host plants; using fallow periods between crops; varying sowing rates; reducing applications of nitrogen or potassium in fertilizers; running light traps to reduce adult populations; planting early and using early-maturing varieties) and some of these methods have been incorporated in IPM programmes.

Biological Control

Classical biological control has not been attempted for O. oryzae but merits careful consideration as there is a distinct complex of natural enemies on the African rice gall midge, O. oryzivora, and it might be possible to establish one or more of these species successfully in Asia.

Conservation and management of indigenous natural enemies would also be worthwhile, but there seem to be no instances where this approach has been implemented. There would also seem to be scope for the management of interactions between natural enemies and rice varieties with partial resistance to the gall midge.

Host-Plant Resistance

Heinrichs and Pathak (1981) reviewed progress in the development of host-plant resistance in Asian countries, especially Bangladesh, India, Indonesia, Sri Lanka and Thailand. They concluded that significant advances had been made in developing rice varieties resistant to gall midge in Asia, but noted that it had proved difficult to provide farmers with varieties that had acceptable grain quality and multiple resistance to the gall midge and other insect pests.

Screening for resistance started in India in the late 1940s and gathered momentum there and in other Asian countries during the 1960s. By 1980, varietal screening trials had identified about 170 resistant varieties, almost all of which were of Indian origin. Mass-rearing techniques were developed in Thailand (Leuamsang et al., 1968) and in Sri Lanka (Perera and Fernando, 1969) and were widely adopted to accelerate evaluation programmes. In India, work on resistance was centred at the Central Rice Research Institute, Cuttack and at AICRIP, Hyderabad. With the advent of high-yielding, semi-dwarf varieties, such as IR8, hybridization programmes were established to incorporate gall midge resistance into the improved plant types from various sources, especially, Ptb 18, Ptb 21, Eswarakora, Siam 29, CR55 and CR56 series and Warangal cultures (W1251, W1253, W1257 and W1263). After multi-location tests, CR93-4-2 (CR55-12/IR8) was released to farmers in Orissa State as 'Shakti' and IR8/Siam 29 was released in Karnataka as 'Vikram'. These and other releases (up to 1979) of improved midge-resistant varieties for commercial cultivation in India, Sri Lanka, Thailand and the Philippines are listed in a bibliography of varietal resistance to the rice gall midge by Pathak and Heinrichs (1982). Heinrichs et al. (1985) detailed methods used in screening for resistance to the rice gall midge in a manual for the genetic evaluation of insect resistance in rice.

Mechanisms of resistance had been little studied by 1981 but were thought to be mainly antibiotic effects on first-instar larvae.

During the 1980s and early 1990s, screening for resistance and the development of IR8-type rices with multiple disease and insect resistance (including resistance to the rice gall midge) continued, resulting in the production of the modern varieties IR20, IR36, IR42, IR64 and IR72. The effects of biotypes also became apparent, and Kalode and Bentur (1989) proposed designation of three biotypes, all of which occur in India, and noted that reactions in Indonesia and Sri Lanka were typical of biotype 2 responses while those in Thailand resembled the biotype 3 pattern. Singh (1996) noted that five biotypes had been recorded in India and proposed recognition of a sixth virulent biotype based on a new pattern of reaction observed in Manipur since 1991.

Reddy et al. (1997) investigated the inheritance of biotype-specific resistance in India and reported a simple mode of inheritance in which genes conferring resistance are dominant.

Chemical Control

Due to the variable regulations around (de-)registration of pesticides, we are for the moment not including any specific chemical control recommendations. For further information, we recommend you visit the following resources:

This information is part of a full datasheet available in the Crop Protection Compendium (CPC); For information on how to access the CPC, click here.


Most assessments of the economic significance of O. oryzae have been subjective, based on general observations rather than rigorous objective assessments of effects on yields. It is generally accepted, however, that the rice gall midge is a major pest in Asia and may be very damaging in some locations and seasons. Average crop losses of 50-100% over wide areas have been reported in some seasons (Reddy, 1967). These are exceptional cases but they do indicate the potential that this pest has to cause widespread and severe damage. In 1983, Hill reported that in some areas of the tropics, O. oryzae is a very serious pest causing losses of 30-50% with some regularity and occasionally losses of 100%.

More recently, Waterhouse (1993), in a survey of the major pests of agriculture in South-East Asia, ranked this species as very widespread and very important in Cambodia and Indonesia, widespread and important in Myanmar and Vietnam and locally important in Thailand and Laos.

Circumstances vary with changes in climate, agronomic practices and the degree of resistance or tolerance in cultivated varieties of rice. There also seems to be a tendency for midge infestations to become more damaging as rice production is intensified. When rice is damaged, the number of useless tillers increases while the number of useful spikes decreases (Huang et al., 1996). The larvae typically move down between the leaf sheaths until they reach the apical bud or a lateral bud where they lacerate the bud tissues and feed until pupation (Hill, 1983).

Grover and Prasad (1980) recorded that, following the 12th FAO Regional Conference in 1972, the Governments of Bangladesh, India, Indonesia, Sri Lanka, Thailand and Vietnam strongly supported the regional co-ordinated approach to the rice gall midge problem. Grover and Prasad (1980) also noted that epidemics of this pest used to occur in cycles of 3-5 years and that 100% of rice plants could be damaged, but that the pest now appears annually and has become a limiting factor in rice production in many areas.

Pathak and Dhaliwal (1981) reported that damage caused by O. oryzae had recently increased in Bangladesh, India, Indonesia and Thailand. In Thailand, O. oryzae had started to attack irrigated dry season rice in the Central Plains in the mid-1970s and was continuing to spread southwards. Similarly, in Java it had been present in mountainous areas but began to occur in coastal rice fields in the 1960s. Pathak and Dhaliwal (1981) also reported that O. oryzae was spreading in the Indian subcontinent and was first recorded in Uttar Pradesh in 1971. They noted that, although it was first known as a pest only in the wet season, large infestations had been recorded on dry season crops in Indonesia.

In Madhya Pradesh, India, the effect of transplanting date during the kharif season (July-October) and nitrogen level on the abundance of O. oryzae and rice yield was determined. Seedlings transplanted in late July had significantly lower silver shoot percentage due to O. oryzae and higher grain yield than other dates tested. Grain yield also increased with increasing levels of nitrogen; the percentage silver shoots was significantly lower up to 90 kg nitrogen/ha than with higher doses of nitrogen (Harinkhere et al., 1995). In variety trials in 1978, some varieties showed higher yields at low levels of infestation (1-10%) than at nil infestation. It was suggested that the infestation stimulated the production of an extra healthy tiller that developed into a panicle and subsequently increased grain yield. At higher levels of infestation, yield was progressively adversely affected (Deshmukh et al., 1989).

Insecticide trails in Tamil Nadu, India, revealed that successful treatments (control by chemical while the parasitoid Platygastor oryzae was present) reduced pest incidence to 5.44% and resulted in the highest yield being obtained (Logiswaran et al., 1993). In trials assessing resistance to O. oryzae, the rice cultivar with the lowest pest incidence also had the highest grain yield (5.3 t/ha in 120 days, 32% higher than that in a susceptible cultivar) (Sevugaperumal et al., 1988). Foliar sprays of neem extracts to control rice pests (including O. oryzae) have also shown that where the control methods were successful the highest yields were also obtained (Saroja, 1986).

In 1987, rice lines resistant and susceptible to O. oryzae were sown on 13 July and 11 August in Andhra Pradesh, India. Gall midge incidence was higher in late-sown crops and was negatively correlated with yield (Kulkarni et al., 1989). In 1989, a serious outbreak occurred in the West Godavari district. Crops planted early (up to December 1988) suffered no damage or less than 10% silver shoots and had higher than expected yields (Rao and Rao, 1989).

In Orissa, India, yield losses due to O. oryzae ranged from -0.06 to 1.1% for every unit percent increase in silver shoots (Rao, 1987). Insecticide trials revealed that when the average incidence of O. oryzae was only 5.1% silver shoots, yields of 3935 kg/ha and 3840 kg/ha were recorded in 1980 and 1981, respectively. This compared favourably with the control yields that were 875 and 2167 kg/ha in the two years, respectively (Samalo et al., 1983). Another trial reported that infestation was reduced to 19% compared with 39.6% in untreated crops and increased the yield from 1.46 tons/ha to 2.04 tons/ha (Satpathy, 1970). Kulshreshtha and Rajamani (1973) reported that paddy yields were reduced from 3.19 t/ha in gall midge free crops to 1.97, 1.6 and 1.35 t/ha in those having 5, 10 and 15% silver shoots, respectively. This indicated that the presence of even 5% silver shoots decreased yields by 37%.

In Goa, the damage caused by O. oryzae and its impact on grain yield were assessed in 1981-82. Early planted crops showed no signs of damage up to 45 days after transplanting, and later infestation had a minimal impact on grain yield. Late-planted crops were attacked 30 days after transplanting and yield was significantly reduced (Sundararaju, 1986). In Bihar, it was shown that tillering was positively correlated with silver shoots and also with panicles. Partial regression analysis indicated that the net effect of a unit increase in the number of silver shoots is a 0.4 reduction in panicle number (Chand and Acharya, 1983). The number of galls caused by O. oryzae in Karnataka averaged 14.1/100 and 19.4/100 plants for seedlings planted on 8 and 15 July 1977, respectively. The corresponding yields were 1.6 and 1.4 kg/plot. It was concluded that later planting resulted in higher pest incidence and lower yield (Rai and Vidyachandra, 1979).

Other work in India generally has also been undertaken. In 1982-83, the incidence of silver shoots was positively correlated with the number of tillers/hill and negatively correlated with grain yield, number of panicles/hill and average grain weight/panicle. For every 1% increase in silver shoots, there was a 1.15% increase in number of tillers/hill and a 1.71, 1.13 and 1.42 % decrease in yield/hill, number of panicles/hill and grain weight/panicle, respectively. (Shrivastava et al., 1988). Mandal et al. (1990) studied the relationship between gall midge infestation and yield-attributing characters of paddy in the field in India. Highly significant positive correlations were found between damaged leaves and tillers 35 days after transplanting. Onion shoots showed negative correlations with panicle length, number of grains/panicle, grain yield and 1000-grain weight, but showed no correlation with number of panicles/hill and straw yield. Correlation between tillers and these characters followed similar trends. Onion shoots produced 25 days after transplanting showed significant negative correlations with grain yield.

In Sri Lanka, yield losses in nine rice cultivars due to O. oryzae were studied during the 1987-88 wet season. Grain yield was negatively correlated with the incidence of O. oryzae in all cultivars tested (Kudagamage et al., 1988).

Tests in Thailand determined that the cost of applying insecticides was more than compensated for where more than 20% of the tillers were infected (Katanyukul et al., 1982). In a field experiment, yield losses attributed to gall midge ranged from 8 to 33% for different cultivars. When gall midge was controlled, average grain yields ranged from 2.03 to 2.91 t/ha for the different cultivars (Katanyukul et al., 1980a). In northern Thailand, studies showed that insecticide treatment reduced the number of damaged tillers from 19.2-26.5 for no treatment to 0.4-2.2%, 55 days after transplanting. The yield loss in untreated plots compared with treated plots ranged from 7.9 to 32.9% (Katanyukul et al., 1980b).

The effect of plant density on infestation and yield was also studied in northern Thailand. At a spacing of 15 x 15 cm and 3 seedlings/hill, 34% of tillers were infested and the yield was 3.4 t/ha. At 25 x 25 cm and 3, 5 or 7 seedlings/hill, the percentage of damaged tillers was 23, 25 and 32, and the yield was 5, 4.6 and 4.9 t/ha, respectively. It is suggested that the high-tillering rice grown in northern Thailand can rapidly compensate for tillers damaged by the gall midge (Katanyukul et al., 1979).

Soenarjo and Hummelen (1976) reported results of field surveys in Java, Indonesia, during the 1975 and 1976 growing seasons and provided detailed maps showing the incidences of infestations in both years. O. oryzae was present at almost all of the 190 locations visited. Counts of silver shoots ranged from 0 to 70% but were mostly below 5%. The northern coastal plain of West Java was most severely attacked, with 190,000 ha affected in 1975 and 250,000 ha in 1976, and a smaller area was heavily infested in Central Java.

A survey in China showed that the distribution and damage caused by O. oryzae was increasing (Huang et al., 1996). Damage to late rice was more serious than damage to early rice in double-cropping areas. In single-crop rice, damage to late-cultivated crops was more serious than that to early-cultivated crops.

In Bangladesh, an economic threshold of 5% onion shoots (caused by O. oryzae) was proposed based on field trials carried out between 1970 and 1979 (Alam et al., 1981).