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

paddy armyworm

Mythimna separata
This information is part of a full datasheet available in the Crop Protection Compendium (CPC). Find out more information on how to access the CPC.
©CAB International. Published under a CC-BY-NC-SA 4.0 licence.


You can pan and zoom the map
Save map
Select a dataset
Map Legends
  • CABI Summary Records
Map Filters
Third party data sources:

Host plants / species affected

Main hosts

show all species affected
Avena sativa (oats)
Brassica rapa subsp. chinensis (Chinese cabbage)
Brassica rapa subsp. oleifera (turnip rape)
Brassica rapa subsp. rapa (turnip)
Cyperus rotundus (purple nutsedge)
Echinochloa colona (junglerice)
Echinochloa crus-galli (barnyard grass)
Eleusine coracana (finger millet)
Eleusine indica (goose grass)
Hordeum vulgare (barley)
Linum usitatissimum (flax)
Nicotiana tabacum (tobacco)
Oryza sativa (rice)
Panicum antidotale (elbow grass)
Panicum miliaceum (millet)
Pennisetum glaucum (pearl millet)
Pennisetum purpureum (elephant grass)
Phaseolus (beans)
Saccharum officinarum (sugarcane)
Secale cereale (rye)
Setaria italica (foxtail millet)
Sorghum bicolor (sorghum)
Sorghum halepense (Johnson grass)
Triticum (wheat)
Triticum aestivum (wheat)
Urochloa mutica (para grass)
Zea mays (maize)

List of symptoms / signs

Inflorescence - external feeding
Inflorescence - fall or shedding
Leaves - external feeding


During the vegetative stage of rice, M. separata damage is evident as massive leaf removal, often including leaf veins. They may also eat the lemma and palea of the developing grains as well as the anthers of flowers. Large angular notches can be cut away from young seedlings in a seedbed, giving an irregular appearance. Damage is often localized to one part of a field. During outbreaks many fields can be affected at the same time.

A characteristic peculiar to M. separata occurs during the grain filling stage when the mature larva cuts off panicles at their base causing some to bend while others are completely severed and fall to the ground.

Prevention and control

Cultural Control

A number of cultural control measures are discussed by Litsinger (1994). When armyworms attack a seedbed the water level can be raised to drown the larvae. In endemic areas, farmers should avoid rotating host crops such as wheat or maize after rice. The rice-wheat cropping pattern is popular in temperate regions of China, Pakistan, India, Nepal, and Bangladesh but encourages build up of M. separata (Litsinger and Barrion, 1988).

Farmers should keep weeds (particularly Gramineae) in check as they are also alternative hosts. Nitrogen fertilizer should be used with care as the improved nutrition causes greater armyworm fecundity and more larval feeding and survival. The water level can be raised when the population is in the pupal stage to drown them. Flooding also limits plant to plant dispersal of M. separata larvae.

Mechanical and Physical Control

Barriers can be constructed to divert migrating M. separata larvae (Litsinger, 1994). Water moats can protect a seedbed or kerosene can be poured in M. separata's path. Egg masses or larvae can be hand-picked from seedbeds or a young crop.


Ducks are often raised in rice areas and can be herded into damaged fields to feed on M. separata larvae (Chiu, 1984). Ducks can locate M. separata larvae hiding in the soil or at the base of plants. Perching in the fields can increase predation by insect-predatory birds.

Classical biocontrol has worked well in New Zealand, where a strain of Cotesia ruficrus originating from Pakistan was imported and released against M. separata. About 24 500 wasps have been released since 1979 (Burgess, 1987). Parasitism of M. separata greatly increased (from 50-55% to over 80%) and the numbers of larvae in the crops have significantly declined. Of even greater importance is that parasitized larvae have weaker appetites. Hill (1988) concluded that the ability of C. ruficrus to reduce the rate of feeding of M. separata larvae was probably the most important factor in its success as a biological control agent. What is even more instructive regarding this effort is that a local strain of C. ruficrus was already present but was replaced by the more aggressive and better-adapted Pakistan strain. A high rate of control has been evident despite high rates of hyperparasitism by the pteromalid Eupteromalus parnarae [Trichomalopsis apanteloctena] which normally ranges from 41-97%. The hyperparasite became abundant after the armyworm was under control.

Host-Plant Resistance

There are no currently known resistant varieties for M. separata. Wilde and Apostol (1983) screened a number of rice cultivars and wild rice accessions. Among the rice cultivars they noted differences in feeding preference and damage but no strong resistant sources were found. Some wild rices (Oryza spp.) had greater levels of resistance than cultivated rice and could be used in breeding programmes. Modern high-tillering rices offer the best level of tolerance against heavy defoliation caused by armyworms. No studies have evaluated rices for resistance to panicle severing. To date, there have been no reports on the development of M. separata-resistant transgenic plants.

Chemical Control

Economic thresholds have been established by Alam et al. (1980b), Khamparia et al. (1982), and Saini (1987) in the control of M. separata by chemical methods. The thresholds are higher during the pre-panicle stages (9 larvae/m² or 1-5 larvae/hill) than the panicle stage (0.4-1 larvae/hill = 10 panicles). Greater care should be taken in opting for selective insecticides during the early crop-growth stages when natural enemy populations are increasing. Less care would be needed during the panicle stage as the crop is about to be harvested. Selective materials include poisoned bait made from rice- or wheat-bran that can be placed along field borders (or in the field if it is dry) for control of larvae (Giraddi and Kulkarni, 1987). Hiremath et al. (1990) developed sugar-based insecticide baits for moths. Microbial insecticides with nuclear polyhedrosis virus are highly selective but should be applied sooner than conventional chemicals as they are slower acting (Dhaduti and Mathad, 1979; Ho, 1986; Okada, 1987). Farmers themselves can re-use the virus by collecting infected larvae from the field, macerating them, and straining the body mass directly into sprayers. Sharma and Davies (1983) reviewed the use of conventional insecticides.

Pheromonal Control

There are no reports of pheromones being used to control M. separata.

Integrated Pest Management

The management strategy for M. separata is based on early detection. High-risk areas can be identified based on the history of past outbreaks and vigilance can be intensified during seasons of expected occurrence. Early-warning systems have been developed in Japan and China to detect immigrating moths (Hirai, 1988; Chen et al., 1989).

Preventative measures can be taken such as planting a high-tillering cultivar and carrying out good crop management. The rice crop is then monitored for larval build-up on a weekly basis. Scouting should be done throughout the entire field area as populations are often highly aggregated. Control decisions should be made before the population attains the last larval stage which not only causes more damage, but is more difficult to control due to its large body-size. Larvae can be collected from the field and reared to determine the incidence of parasitoids. If parasitoid activity is low, corrective actions such as the use of ducks and perches in the field can be practiced before larval population reaches the economic threshold, while or a microbial pesticide can be taken when the larval population reaches the economic threshold.


M. separata's habit of severing panicles represents the greatest damage to the rice crop, and is analogous in effect to whitehead damage caused by stem borers in that the grains are not eaten by the pest but are removed from the plant.

The effect of larval damage on the growth and yield of rice at the booting and panicle stages was determined in field-plot tests by Alam et al. (1980a). The yield loss caused by a population of 15 larvae/hill at the two stages was 38 and 93%, respectively. With each increase of 1 larva/hill, the number of grains per panicle decreased by 4 and 9, respectively, for each growth stage. In cases of severe infestation, damage may be up to 60% or more (Dale, 1994). However, the incidence of M. separata has declined to some extent during the past 30-40 years because of the expansion of irrigated rice cultivation.