Spodoptera exempta
(black armyworm)

S. exempta is found in Africa, on the Arabian Peninsula, and in South-East Asia, Australasia and Oceania (including Hawaii, USA). Moths have been taken in Japan and New Zealand but there is no evidence that the species is established there. S. exempta is widespread in Africa south of the Sahara, being most prevalent in the east and eastern central regions of the continent. In South-East Asia, it is most frequently recorded from the Indonesian islands, including Kalimantan, Sulawesi and the Philippines. In Australasia, it is confined to Papua New Guinea, the Solomon Islands and New Caledonia, and to the northern and eastern seaboard of Australia.

Records of S. exempta in California, Kansas, Oregon, Washington and Wisconsin (EPPO, 2014) published in previous versions of the Compendium are invalid as the original source of the records (Arnett, 1993) does not mention the distribution of the pest in these states. There is no evidence of S. exempta in the continental United States (USDA APHIS, 2020, communication to CABI).

S. exempta is not listed as a quarantine pest by EPPO.

Hosts of S. exempta are primarily plants of the Poaceae and Cyperaceae. Major economically important hosts are: barley, Pennisetum typhoides, Eleusine coracana, maize, oat, rice, sorghum, sugarcane, Eragrostis tef, wheat and pasture grasses, especially Cynodon and Pennisetum species. The larvae exhibit strong host preferences within the Poaceae and there are major differences between varieties of cereal crops in their susceptibility to attack. S. exempta has once been found causing serious damage to coconut seedlings and once to young tea and, during high-density outbreaks, non-host plants including tobacco and cotton may be eaten, though not extensively. See Rose et al. (1996) for a full list of wild and cultivated host plants.

S. exempta larvae develop in markedly different forms or 'phases' depending on larval densities in the early instars. Although similar 'phase polyphenism' occurs in other Noctuidae, it is most strongly developed in S. exempta, in which the highly conspicuous, heavily pigmented and active 'gregarious' ('gregaria') form, typical of outbreak populations, is the familiar armyworm. The cryptic, sluggish 'solitarious' ('solitaria') form, typical of low-density populations, is difficult to find and often not recognized as an armyworm. A second key feature of the biology of S. exempta is the absence of any pre-imaginal diapause; a single report of an extended pupal period has never been confirmed.

In eastern Africa, there are typically six to eight outbreak generations per year, with four to five in southern Africa. There is a period of 3-5 months when no outbreaks are reported, during which low density populations persist in areas supporting continued growth of host plants, where a total of 13 generations per year is possible.

Oviposition usually begins at 20.00-21.00 h, eggs being laid in batches of 10-600 and covered with black hair scales. A single female lays about 1000 eggs over a period of up to 6 nights but little or no oviposition occurs at temperatures below 20°C. Oviposition often occurs on substrates other than host plants, including dry grasses, leaves of tall plants, twigs of bushes and trees, or on buildings. Eggs hatch in 2-5 days (typically 3 in normal outbreak conditions), with a temperature threshold for development of 10-12°C. Eggs generally hatch in the early morning and the translucent larvae, after feeding on the egg shells, drop from the oviposition site on silken threads to be dispersed on the wind, often over several metres. Mortality can be high, with larvae failing to reach suitable hosts, being drowned by rainfall, or taken by predators. On reaching a host plant, positive phototaxis and negative geotaxis takes them to the upper, often youngest, foliage where they commence feeding by rasping the lower lateral epidermis of the leaves, creating 'windows' and gradually becoming green. They remain green for the first three instars, diverging into the solitarious or gregarious forms at the III-IV instar moult. At this moult, the form of the mandibles also changes, enabling the larvae to feed on the leaf edge. There are normally six larval instars and food consumption increases rapidly in the latter stages, reaching about 0.2 g/day for a fully grown larva on maize. The increase in weight from first to last instar is about 1000-fold.

Gregarious larvae are predominantly black, very active, and feed on the upper parts of host plants, exposed to the sun and avoiding shade. They are adapted behaviourally and physiologically for accelerated development; in particular, their behaviour and pigmentation result in elevated body temperatures by the absorption of solar radiation. The development of gregarious larvae in outbreaks is often highly synchronized, with most larvae pupating over a 3-7 day period. There may also be fewer than six larval instars. Older gregarious larvae may 'march' in large numbers, especially when food plants are locally depleted or when development is complete and they are searching for pupation sites. Solitarious larvae are predominantly greenish and highly cryptic in their coloration and behaviour. They are sluggish, actively avoiding the sun, and sheltering and feeding at the bases of grasses. Although usually sparsely distributed and difficult to find, they can occur at relatively high densities (e.g. 10/m²) where thick vegetation prevents contact between early instars.

The full grown larva burrows into soft, damp soil to construct a silk-lined chamber 2-3 cm below the surface, where it pupates. High mortality may occur if the ground is too dry and hard.

In addition to temperature and larval density, development rates are influenced by food-plant type and quality, sex and the effects of parasitism. The larval period lasts from 11-24 days (21 days in typical outbreak conditions). Pupal periods range from 7-12 days (typically 10 days in outbreaks). The temperature threshold for larval and pupal development is 13-14°C. The larval period of females is longer, and the pupal stage shorter, than for males and, as a result, the peak of female emergence during outbreaks is usually one night earlier than that of males.

Moths emerge in the early part of the night, peak emergence being between 20.00 and 22.00 h. In a single outbreak, emergence may extend over about 12 days with a peak during the first 4 days. Moths are ready for flight in 1.5-2 h when some ascend in migratory flight to be carried downwind. However, the majority fly into nearby trees greater than 1.5 m in height where they may accumulate in numbers of several thousands. Numbers of these moths may embark on migratory flights at any time of the night, forming ascending 'plumes'. Those that remain in the trees at dawn undertake short flights to shelter during the day in grass clumps, under cowpats or in crevices in the bark of trees. The following dusk, they emerge from their day shelters to embark on downwind migratory flights, some moving briefly into trees before ascending. Where nectar is available, moths feed voraciously.

Current evidence suggests that S. exempta is an obligatory migrant, i.e. all individuals emigrate from the site of their emergence. Reports of successive generations at the same site are rare; they appear to occur only when emigration has been prevented by rainfall after moth emergence or by re-invasion by immigrant moths from distant sources. When they occur, they are likely to suffer high mortality from natural enemies, including pathogens and exhaustion of local food plants.

During the main outbreak season in East Africa, the prevailing easterly to south-easterly winds are strong in the first half of the night so migrants embarking at dusk or early in the night can be carried for long distances, of 100 km or more. Radar studies show migratory flight at up to 420 m above ground level in open savanna country and double that altitude in the Rift Valley near the Nairobi escarpment. The winds tend to lighten and become variable in the second half of the night when migrants may orientate across or upwind and displacements are consequently more limited. Radar also shows that the moths do not fly in cohesive swarms but are dispersed as they move downwind. In moderate to heavy rain, migrants descend to the ground.

In females, migration terminates when reproductive development reaches the stage at which the moths call (release sex pheromone) and is, therefore, exclusively pre-reproductive. Thereafter, only local flights occur during mating and oviposition. The oocytes are not developed when the females emerge but develop to approximately half full size within 24 h. Oocyte development may then be arrested for up to 13 days. Whether this arrest occurs, and for how long, is genetically determined. Once it is over, females require water to hydrate the maturing oocytes and complete development. Moths have been seen to feed avidly on nectar and to drink dew before, between, and after migratory flights but the availability of food or water does not terminate the arrest of oocyte development.

Variation between individuals in flight capacity (duration) on individual nights is also genetically determined although there is evidence that it is moderated by larval density (moths from gregarious larvae show enhanced flight performance). A high level of variation in both of the traits which determine migratory potential (capacity for displacement), i.e. flight capacity and pre-reproductive period (the number of nights over which migratory flights occur) has been demonstrated in field populations. This variation is thought to reflect the selection imposed by the pattern of distribution, within and between seasons, of suitable habitats. On average, male moths reach maturity earlier than females and the available evidence suggests that they may continue migration after attaining maturity, mating with females at different locations on different nights.

Marked moths have been recovered up to 147 km from an emergence site but studies of the spread of outbreaks suggest the frequent occurrence of much longer displacements of 200-700 km. In New Zealand, they have been recovered 3200 km from the nearest known source although these moths must have been constrained to fly by day and night across the sea.

Thus, moths emerging from a high-density outbreak become widely dispersed downwind, in both space and time, as a result of: emergence and emigration over a period of about 10-12 days; differences between individuals in migratory potential; and dispersal in flight on winds which vary in speed and direction across both time and space. The usual outcome is widely dispersed oviposition resulting in low-density, solitarious populations.

However, during the rains, flying moths have been shown, on radar, to be concentrated by wind convergence associated with rainstorms and topography. Particularly important is the strongly convergent airflow in rainstorm outflows, at the gust front, where further dispersal is often curtailed by the descent of flying moths as they encounter the rain. The complex patterns of eddies established in the lee of hills and other topographical features may also accumulate moths flying into them. It is the localized oviposition by these concentrations of moths that result in outbreak populations of gregarious caterpillars.

At the end of their migratory flight, mature female S. exempta settle in trees, releasing sex pheromone to attract males. The pheromone has two major (Z-9-tetradecen-1-yl acetate and (Z,E)- 9, 12-tetradecen-1-yl acetate) and four minor components. Mated pairs are found from 21.00-05.00 h with a peak between 24.00 and 02.00 h. Multiple mating is common.

S. exempta has an extremely high reproductive capacity and thus potential for population increase. Laboratory studies indicate Ro (net reproductive rate) from 15 on poor quality grasses to 125 on Cynodon sp. and 142 on maize. Assuming Ro = 100 and 80% mortality, increases of 10,000-fold are possible in two generations. These levels are easily achieved for early season infestations feeding on flushes of new grass growth.

The annual cycle of S. exempta in eastern and southern Africa can be summarized as follows.

In eastern Africa, the rains are associated with the passage of the Inter-Tropical Convergence Zone (ITCZ) which moves northwards and southwards seasonally, moving over northern Sudan and the southern Arabian Peninsula in July to August, southward through Ethiopia and Somalia during October, across Kenya during November, and reaching Tanzania by December and as far south as northern Mozambique and Zimbabwe in January. During February and March, it moves north through Tanzania, crossing Kenya in April and May and southern Sudan, Ethiopia and Somalia in May to June, to reach western Arabia by July. The southward movement of the ITCZ is associated with the 'short rains', and the northward movement with the 'long rains', in southern Ethiopia, Somalia, Kenya and northern Tanzania. Regions at the extremes of its traverse usually have only one rainy season.

During the dry season, grasses dry out over most of eastern Africa, becoming unsuitable for larval development. However, populations persist where host plants continue to grow, for example, where there is intermittent rainfall through this period. In the absence of concentrating weather systems, these populations remain at low densities in the solitarious form. In eastern Africa, dry season populations have been detected in Malawi, western Uganda, on the shores of Lake Victoria and in highland areas, and probably most importantly, in coastal regions of Kenya and Tanzania. Similar areas presumably support dry-season populations elsewhere in Africa. These populations are the source of moths initiating the first outbreaks of the subsequent rainy season.

First outbreaks typically occur in Kenya and Tanzania where there are two rainy seasons. They are associated with the first rainstorms of the short rains which tend to occur on the eastern side of high ground where moths, flying from coastal source areas on the dominant easterly winds at this time, meet storm-induced wind convergence. These 'primary' outbreaks, derived from low-density populations, are distinguished from subsequent 'secondary' outbreaks which are initiated by moths originating from earlier outbreaks. Outbreaks (usually primary) leading to significant sequences of secondary outbreaks are termed 'critical'. The sequence of outbreaks that occurs seasonally follows the pattern of seasonal winds. In eastern Africa, outbreaks move progressively westwards from eastern Tanzania and Kenya, in some years as far as Burundi, before the start of the major northward movement with the passage of the ITCZ through Kenya and Uganda, Ethiopia, Somalia and southern Sudan, in some years reaching the Yemen. In southern Africa, the spread of outbreaks seems to originate in southern Tanzania, northern Mozambique and Malawi, moving into Zimbabwe, Swaziland and South Africa, with less frequent invasions of Zambia and Botswana.

The extent and severity of outbreaks appears to depend largely on the successful development of primary outbreaks at the start of the rains. This has been shown to be associated with preceding drought, frequent early storms to concentrate flying moths and initiate flushes of grass growth, and dry and sunny periods during the larval stages to promote survival and rapid development. Therefore, major upsurges occur in seasons of sporadic rainstorms and long sunny periods throughout the outbreak period. Prolonged cool, cloudy and rainy weather in the short rains is associated with high larval mortality, particularly from viral disease, and suppresses the development of subsequent outbreaks.

With its rapid development, high reproductive capacity, and mobility by migration, S. exempta is well adapted to exploit the highly seasonal and ephemeral habitats provided by the rain-induced growth of host plants in tropical grasslands.

This account is based on extensive, long-term studies in East and southern Africa. Very little is known of the biology and ecology of S. exempta elsewhere in its range.

See Rose et al. (1996) and references therein.

Natural enemies sometimes have a significant impact on S. exempta populations though their influence is difficult to quantify. Lists for eastern and southern Africa of species attacking all stages are given by Rose et al. (1996). There is no published data on natural enemies elsewhere in the species' range. Of the 21 genera of Diptera and 23 of Hymenoptera represented among armyworm parasitoids, only four Diptera and four Hymenoptera are regularly found in significant numbers. Because of its short life cycle and obligatory migration, there is little opportunity for parasitoid populations to build up to levels capable of constraining population development significantly. Only on rare occasions, when successive generations occur at one location, have high levels of parasitism (exceeding 70%) been recorded. Avian predators (storks [Ciconiidae] and crows [Corvus]) may decimate small outbreaks but have little influence on larger ones. The major cause of mortality in S. exempta larvae is probably the species-specific nuclear polyhedrosis virus (SpexNPV). First outbreaks of the season may be virus-free because of the lack of transmission in low density populations but later outbreaks may be almost eliminated by this virus, mortalities of >98% being frequent. Widespread virus attack is often associated with overcast, cool, wet weather. A Cytoplasmic Virus (CPV) is also an important pathogen, killing pre-pupae and pupae. Fungi and bacteria are thought to be of minor importance. See Rose et al. (1996) and references therein.

Infestation of the crop by S. exempta is evident from the `windowing' or skeletonizing of younger leaves caused by rasping of the epidermis by young larvae, or gross feeding by older larvae. The horizontal middle leaves of the plant, where photosynthetic activity is greatest, are particularly affected, but the growing points, other leaves, and the stems of younger plants are also affected. Large numbers of black larvae feeding on poaceous plants can be assumed to be S. exempta, whereas larvae feeding in significant numbers on leaves of dicotyledonous plants are unlikely to be (although damage to non-poaceous plants has occasionally been reported). S. exempta egg batches can be distinguished from those of all other Spodoptera spp. by the presence of a covering of black hair-scales. The main distinguishing features of gregaria-phase larvae are:

* velvety black-dorsal surface
* three pale parallel lines on the dorsal surface of the prothoracic segment
* central longitudinal stripe along the top of the body always paler than the black areas on either side
* pale lateral lines
* green or yellow ventral surface
* lack of hairs on the body
* shiny black head

They can be distinguished from the following species, with which they are sometimes confused, by the following features:

* Spodoptera exigua - often very similar but paler dorsally than laterally
* Helicoverpa spp. - the larvae are green or brownish, covered with short hairs, and do not have a black head-capsule
* sawflies (Symphyta) - larvae are typically black and hairless with more than five (usually seven) pairs of prolegs on the abdominal segments. Spodoptera larvae lack prolegs on the first two abdominal segments.

Solitaria-phase larvae may be difficult to distinguish from other grass-feeding caterpillars. Their green, pink or brown coloration is cryptic and the larvae hide in the grass mat during the day, emerging to feed at night. When found, they appear fat, are usually curled up, and behave very sluggishly. The distinguishing characters of older larvae of both phases are the three parallel pale lines on the dorsum of the prothoracic segment, the white spot at the rear of each abdominal segment, and the shape of the mandibles. Pupae of all Spodoptera species are very similar and can only be distinguished with difficulty. However, if very large numbers of pupae are found over large areas by digging, they are likely to be from an outbreak of S. exempta. Adults may be distinguished from other Spodoptera spp. by their forewing pattern and genitalia using keys published by Brown and Dewhurst (1975) and modified by Rose et al. (1996). The black hair-scales at the tip of the abdomen of the female moth and the grey, racket-shaped scales on the outer part of the genitalia of males are characteristic of S. exempta, and useful for identifying moths which have lost their wing scales, e.g. in light traps.

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Forecasting and Control

Effective forecasting and management of pests like S. exempta, whose long-range migrations result in the rapid spread of outbreaks from country to country, requires international cooperation. This is most easily achieved if regional organizations have responsibility for coordinating information and, if possible, the infrastructure required for both forecasting and control. Several national crop protection services, especially in eastern, central and southern Africa and the Yemen, have departments with special responsibility for control of migrant pests, including armyworm. These may assist farmers in countries where armyworm is a notifiable pest, by providing and/or applying insecticide. The Desert Locust Control Organization for Eastern Africa (DLCO-EA) and the International Red Locust Control Organization for Central and Southern Africa (IRLCO-CSA) have regional responsibilities for armyworm. Where infestations are widespread, governments have the option of requesting (or hiring) spray aircraft from them. International cooperation also allows the development of a strategic approach to limiting the development and spread of outbreaks. By focusing attack on early season infestations providing sources of migrants, it should be possible to reduce or eliminate the occurrence of subsequent outbreaks, or sequences of outbreaks, downwind. Such an approach has been developed and implemented by DLCO-EA and its member countries. For further discussion of strategic control of S. exempta, see Cheke and Tucker (1995), Rose et al. (1996), and references therein. Forecasting An effective forecasting system is essential for successful management of Spodoptera exempta, to allow preparations to be made in time to control infestations and reduce crop losses. There may be little time to respond as infestations frequently go unnoticed until larvae become conspicuous at the fourth instar and the amount of damage then increases exponentially with every day that passes. A forecasting system for S. exempta has been in operation in East Africa since 1969. All forecasts and warnings are currently based on: the distribution of armyworm populations currently reported through the monitoring system; the distribution of previously reported populations, with allowances made for their development with time (timing of oviposition following large trap catches of moths and subsequent hatching and development of larvae, timing of emergence of, and emigration by, moths from earlier outbreaks); predominant winds during the periods of moth migration, and synoptic and meso-scale zones of wind convergence in which moths could become concentrated; the distribution of rainstorms with associated wind convergence capable of concentrating moths, particularly early in armyworm seasons; and historical precedents for the development anticipated. The necessary information is derived from monitoring outbreaks and sampling larvae. Accurate monitoring and prompt reporting of armyworm outbreaks is essential for forecasting and control. The procedures developed in member countries of DLCO-EA involve: searches for newly hatched larvae one week after first high catches in pheromone traps, especially catches associated with the first heavy rains after a drought period; larval sampling to determine the age of outbreaks; reporting outbreaks to local district agricultural offices and national plant protection services immediately they are found, preferably with samples of larvae; assessment of larval age by national plant protection service laboratories. Details of these procedures are given by Rose et al. (1996). Trap data from moth trap networks Networks of moth traps in each country are the most effective way of monitoring armyworm populations. Pheromone traps are recommended for widespread use in national networks. However, they catch only sexually receptive males so some light traps, which catch immature, migrating moths (sometimes in large numbers), should be included at selected sites where electricity and expertise for sorting and identification of catches is available. Detailed information on trap design, pheromone lures (chemistry, formulation and availability), and siting and operating traps is provided by Rose et al. (1996). Meteorological data Forecasting services require rainfall, windfield and rainstorm distribution data. These are generally available through national meteorological services with which close cooperation is essential. In recent years, satellite remote-sensing data have become available at an affordable cost. In East Africa, Cold Cloud Duration data from Meteosat are used to identify locations of rainstorms with the potential to concentrate moths, as well as to determine whether there has been little or no, intermittent and scattered, or prolonged and widespread, rainfall on a regional scale. This information is used to guide monitoring to locate outbreaks and to assess the level of infestation that can be expected, given the known influence of weather, especially rainfall, on the population dynamics of S. exempta (see section on Biology and Ecology). Historical data archives Archives of S. exempta trap and outbreak data have been accumulated nationally and regionally in eastern Africa (these archives are stored also at DLCO-EA, Nairobi and, up to 1988, at the Natural Resources Institute, Chatham, UK). The archive data are now routinely accessed by forecasters and provide analogues of outbreak distribution and trap characteristics for the evaluation of current armyworm situations. Computerised databases, using the specific data management system 'WormBase', are established in eastern Africa (Crop Protection Branch in Kenya, Pest Control Services, Tanzania, DLCO-EA), and are being introduced in IRLCO-CSA countries of central and southern Africa. Historical meteorological data are available from national meteorological services. Forecasts and warnings have different levels of urgency: forecasts are prepared weekly, or every 2 weeks, based on information received at the national or regional offices from the monitoring systems and describe expected future armyworm developments. They may be regional or national in scale. An expert system to produce computer-assisted forecasts has been developed at DLCO-EA. A warning is issued as an alert to the immediate potential occurrence or redistribution of infestations, ideally while they are still in the moth or early larval stages. Warnings should be sent by the quickest possible means (telephone, radio, or broadcast media and newspapers) to agricultural offices in the affected areas, to be acted on immediately. Verification of the reliability of forecasts and warnings is essential if their accuracy and value are to be improved. This is now undertaken routinely by the Regional Forecasting Office at DLCO-EA, Nairobi, by plotting locations of reported outbreaks in relation to predictions, using a computer. Reasons for errors are analysed to avoid repetition. Further information on forecasting is given by Rose et al. (1996).

Cultural Control

Weed-free maize crops greater than 50 cm high are unlikely to become infested by newly hatched larvae of S. exempta because the leaves are too tough to allow them to establish. However, if larvae are able to develop on grass weeds, subsequent infestation of the crop may occur. Farmers are advised to keep crops free of grass weeds but, if fields do become infested, to leave the weeds until the larvae have pupated or been controlled. Some maize varieties are more susceptible to attack than others, e.g. Katumani, a dryland variety grown widely in Kenya. These varieties are most at risk where probabilities of armyworm infestation are high.

Biological Control

Predators and parasitoids of S. exempta are never numerous enough to achieve natural control of outbreaks. The nuclear polyhedrosis virus (SpexNPV) has been used to control infestations by spraying with a water suspension of diseased larvae. The feasibility of laboratory production and formulation of the virus has been demonstrated. Recent field trials in Tanzania indicate that SpexNPV could have a potential role as a substitute for chemical insecticides in strategic armyworm management programmes (Grzywacz et al., 2008). A formulation of Bacillus thuringiensis has also been identified as promising. Full laboratory and field evaluation of these products may result in their adoption in the future.

Chemical Control

Only the larval stage is accessible to control by insecticides; eggs are difficult to find, pupae are underground and moths fly at night at extremely low aerial densities. S. exempta larvae are susceptible to a wide range of insecticides and there is no record of resistance; insecticides are never applied consistently enough, in space or time, to impose adequate selection for it to evolve. The major problem is that larvae are not usually detected until they turn black at the III-IV instar moult, when they have already been developing for 7-10 days and serious damage is imminent. Any delay in applying control measures, during the further 8-12 days until pupation, will result is escalating levels of damage. Rapid response is, therefore, essential and can only be achieved by being well prepared and equipped for control operations and acting on forecasts and warnings. In the absence of warnings or trap data, farmers are advised to check crops for newly hatched larvae one week after a rainstorm, particularly the first storms of the rainy season. Recommended insecticides include organophosphorous compounds, carbamates and synthetic pyrethroids, several of which (especially fenitrothion and cypermethrin) are effective when applied at low dose rates by Ultra Low Volume (ULV) spraying. For environmental and safety reasons, organochlorine insecticides are not recommended for armyworm control. Insect Growth Regulators (IGRs) have recently been shown to have considerable promise for use against S. exempta (Fisk et al., 1993) but require further field evaluation. ULV spraying is the most efficient method of armyworm control, allowing rapid treatment of large areas with low volumes of insecticide. Application may be achieved by hand-held, vehicle-mounted, or aircraft-mounted rotary atomisers, according to the scale and accessibility of outbreaks and the resources available. Lever-operated knapsack sprayers, fitted with low-volume hydraulic nozzles, are frequently used for armyworm control. Further details of chemical control (calibration of equipment, conditions for spraying, application techniques, assessment of kill, environmental considerations, safety, logistics and responsibilities) are provided by Rose et al. (1996).