1996 MANAGEMENT GUIDELINES FOR THE BEET ARMYWORM ON COTTON

In the last 10 years the beet armyworm, Spodoptera exigua (Hübner) (BAW), has become a frequent and, at times, explosive pest of cotton in the southeastern United States (Ruberson et al. 1994a, b, Stewart et al. 1996). The BAW caused extensive damage to cotton in the lower Rio Grande Valley and San Angelo areas of Texas in 1995. A number of factors appear to have contributed to these outbreaks, including weather, agronomic practices, soil type, stand density, early season insecticide applications and plant vigor (Ruberson et al. 1994a). The BAW is inherently tolerant to most available conventional insecticides and can be expensive to control. When used properly, the insect growth regulators (IGR), Dimilin (Uniroyal) and Confirm (Rohm and Haas), and a viral insecticide, Spod-X (biosys) have shown good control. Two insecticides, Pirate (American Cyanamid) and Confirm, are available for 1996 under Section 18 labeling. The following is an updated review of available information concerning the biology and recommendations for management of the BAW in Texas.

Distribution. The BAW is an introduced pest, thought to be a native of southern Asia. It has become widespread throughout the tropical and temperate regions of the world. In the United States it is common in the southern and western states and occurs northward into Montana. The BAW is becoming an important pest of soybeans from the Mississippi Delta eastward to Florida and northward to North Carolina and southern Virginia.

Life History. The adult BAW is a gray to brown moth with a lighter spot near the center of each wing and a wing span of 1.25 inches (Fig. 1G). Heavy infestations of BAW may occur suddenly due to migrations of large numbers of adults. Moths are capable of migrating over one hundred miles in a single night when weather conditions are favorable (French 1969, Mitchell 1979). Migrations typically move northward as the seasons progress from spring into summer. Most northern states are infested by these migrating BAW adults. A total of 500-600 eggs per female are laid in masses (Fig.1A-C), averaging about 80 eggs each, and covered with hairs and scales from the female's body. The egg laying period lasts from 3 to 7 days. Eggs may be laid on either the upper or lower surface of leaves, but more commonly the lower. Females also prefer to lay eggs on drought stressed and more exposed plants. Egg masses are similar to those of the fall armyworm (FAW), which is a closely related species sometimes found in cotton. However FAW egg masses are usually larger with over 100 eggs per mass. FAW eggs may tend to be light grey while BAW eggs are white to pink. Both BAW and FAW masses are covered by moth scales but viewed from above a FAW egg mass looks light brown, while a BAW egg mass looks white to dingy white in color. Both FAW and BAW masses can have stacks or layers of eggs, but BAW masses commonly consist of a single layer of eggs. FAW eggs tend to be laid lower on the plant than BAW eggs. BAW eggs hatch in 2-5 days, and larvae feed for 10-24 days then pupate. The pupal period lasts from 5 to 10 days. Larvae pupate within 3 inches of the soil surface (Fig 1F). Moths die 4 to 10 days after emerging
from the pupa.

There are typically 5 BAW larval stages or instars before pupation. The first 3 instars feed for several days in groups on the underside of a leaf close to the site of the egg mass. As the larvae develop, they tend to disperse from the plant where the egg mass was deposited. FAW larvae may not skeletonize leaves as much as BAW larvae, and survival of FAW larvae on cotton is usually not as high as BAW, so few FAW larvae may result from numerous egg masses. Newly hatched BAW larvae are white with a black head and are not distinguishable from newly hatched FAW larvae until about 3 days after hatching (second instar). BAW larvae are distinctively green in color with a conspicuous black dot on each side of the second body segment behind the head and just above the second pair of legs. FAW larvae have an inverted AY@ marking (white to orange in color) on the front of the dark brown head (Fig. 1E) and are brown in color. BAW larvae do not have this light colored inverted AY@ on the head. BAW larvae have smooth skin with few or no hairs on the body, a character that separates them from bollworms and tobacco budworms. BAW larvae reach a maximum length of about 1.25 inches.

The length of the life cycle varies with temperature and may range from 21 to 46 days (Fig. 2). At lower temperatures, development is slower resulting in a longer generation time. In South Texas in 1995, fairly distinct generation peaks occurred about 21 days apart and larvae fed for about 10-14 days.

Host Plants. The host plant range of the BAW is very broad and includes many weeds, trees, grasses, legumes, truck crops and field crops. Crops attacked are sugar beets, asparagus, alfalfa, corn, cotton, lettuce, tomatoes, potatoes, onions, peas, citrus and sunflower. Important wild hosts are sunflower, redroot pigweed, mallow (lambsquarters), saltbush, and grasses.

Historically, the BAW has been considered to be an occasional, late season pest in cotton that seldom occurred in economically important numbers. Several entomologists familiar with the BAW have stated that it may be a changing pest (Ruberson et al. 1994b). For example, it may be infesting cotton earlier in the season and possibly choosing cotton as one of its favored hosts. Also, it may be able to survive the winter much further north than was previously thought.

The most obvious indications of infestation are the characteristic egg masses coupled with leaf skeletonization by early instar larval feeding. Larger larvae can feed on foliage but also feed on squares, flowers and immature bolls; and may move back and forth from leaves to fruit. Damage to squares may appear similar to bollworms, but the BAW frequently feeds on the square bracts, while the bollworm usually does not. Overall, the BAW cannot be thought of as behaving like the bollworm, nor does a single BAW apparently cause as much fruit damage as a single bollworm.

In seedling cotton (pre-squaring), the BAW can skeletonize leaves and damage terminals, even though natural mortality of larvae is high on this stage of growth. The BAW is most damaging during the squaring and blooming stages of growth. In post cut-out cotton, the BAW is generally not a heavy feeder on mature bolls and, therefore, is less damaging.

The BAW can be a devastating pest as evidenced by previous outbreaks in the U.S.; including 1977, 1980, 1981, 1988, 1990, and 1993. It caused $26 million worth of damage in Georgia in 1990 alone. Infestations occurred in far West Texas in 1994, and many areas of Texas in 1995. Over 3 million acres in Texas were infested to varying degrees in 1995 and about 1.2 million acres were treated for BAW at a cost of over 31 million dollars. However, the amount of damage varied widely, ranging from none to severe. Different crop stages were attacked and the efficacy of insecticides varied. The areas receiving the most damage were the lower Rio Grande Valley (LRGV), Southern Rolling Plains, and parts of the Coastal Bend. In the LRGV, the BAW infestation peaked at an estimated 1.1 million larvae per acre. For a more detailed description of the BAW in Texas in 1995, refer to the 1996 Beltwide Cotton Conference papers on that subject (Huffman 1996, Stewart et al. 1996, Summy et al. 1996).

Several factors can contribute to BAW outbreaks. The likelihood of a heavy outbreak increases as more of these key factors occur in a given location (Ruberson et al. 1994a, 1994b, Stewart et al. 1996). These factors are: mild winters (e.g., absence of freezing temperatures); late planting; delayed crop maturity; heavy early season organophosphate or pyrethroid insecticide use, especially for aphids and/or plant bugs that kill natural enemies of BAW; prolonged hot, dry weather conditions; presence of BAW early in the production season; and weather conditions that support long-distance atmospheric transport (i.e., migration of BAW into a cotton production area.). Recently, entomologists at Mississippi State concluded that dry weather and early season insecticide applications were the two factors which best correlated with BAW outbreaks over a 10 year period in 4 states (Stewart et al. 1996). During 1995 the LRGV averaged 4 insecticide applications for early season insects. Additional characteristics of "high risk" fields that consistently seem to fit a pattern for developing BAW problems are: sandy and doughty soils; skip-row planting; fields with skippy, open canopies; drouth stressed plants; and fields infested with pigweed. However, when BAW populations are high all fields are susceptible.

The most important BAW management practices to minimize crop loss are early planting and early maturity, preservation of beneficial insects, thorough scouting, early detection and the use of treatment thresholds. There are two general types of thresholds discussed in this fact sheet: "early detection" and "remedial". The term, "early detection", refers to the scouting and discovery of unhatched and newly hatching egg masses only; while "remedial" refers to fields where scouting reveals relatively larger larvae that have dispersed to some degree from the egg masses and are generally distributed in the field. The time from early detection to a more generally developed infestation may only be a few days. Control efforts will be more effective when directed at newly hatched to 3 inch long larvae, before they begin feeding on fruit. Also, BAW is easier to control on pre-squaring cotton than on squaring or blooming cotton, probably due to better spray coverage, better exposure of the larvae and relatively poor survival on pre-squaring cotton.

Early Detection Threshold (Hatching Egg Masses). The early detection threshold from initiation of squaring to cutout is 2 "active hits" (i.e., recently hatched egg masses with actively feeding larvae) per 100 row feet (see below for Pirate thresholds). Scout several areas of the field, because BAW infestations can be spotty. Treatment should be considered if the early detection threshold is reached or exceeded, and conditions are optimal for a BAW outbreak (i.e., many of the key outbreak factors exist). Ovicides alone apparently are not very effective against BAW because the eggs are well protected by the cotton canopy and moth scales and hairs covering the eggs.

The insect growth regulator (IGR), Dimilin 2L (diflubenzuron), is best used in an early detection approach. Dimilin inhibits formation of the larva's exoskeleton. To be effective, it should be applied before a BAW infestation becomes well established. Dimilin binds to the leaf and has long residual activity (4 weeks or more), but new plant growth is not protected so multiple applications of Dimilin may be necessary. Dimilin does not work well against bollworm or tobacco budworm.

There are some Dimilin label changes for 1996. It can be applied in sequential applications of 2-4 ounces per acre (before bloom) and 4-8 ounces of formulated material per acre (mid-season), on a 5-7 day interval. A total of at least 8 ounces per acre should be applied and additional applications made (maximum 24 ounces or 6 applications per season) as necessary, based on plant growth and subsequent BAW egg and larval pressure. It may also be used ULV by ground or air. For conventional aerial application, apply in 3-5 gallons total volume per acre. For ULV application, use 20-48 fluid ounces total volume per acre. Refer to the label for specific mixing directions and additional changes or restrictions, including the use of oil under certain conditions. Note that food or feed crops cannot be planted in Dimilin treated soils within 6 months following the last application (unless Dimilin is labeled for use on such crops).

Another IGR, Confirm (tebufenozide), has a specific exemption (Section 18) for the 1996 season, and is effective against all sizes of BAW larvae (but is not effective against bollworm or tobacco budworm). Confirm causes the larvae to molt prematurely, but larvae cannot complete molting and subsequently die 15 - 72 or more hours after exposure. Ingestion of Confirm results in feeding inhibition 12 - 36 hours after exposure and feeding ceases within 15 - 48 hours post-exposure. The speed of action of Confirm is dependent on the larval stage; it works faster on smaller larvae. The Section 18 label requires that there are 2 active hits per hundred row feet before Confirm can be used. It is used at the rate of 8 ounces of formulated material per acre, for a maximum of 2 applications. If treating with Confirm by air, use a minimum of 5 gallons total volume per acre. Confirm has a residual of 2 weeks after binding to the leaf surface, and must be used with the spreader-binder, Latron CS-7. Again, new plant growth following application will not be protected. Refer to the Section 18 label for additional information or restrictions.

Insect growth regulators will be relatively slow acting and may require 4-7 days or more for maximum activity. In the LRGV and the Coastal Bend in 1995, activity of Confirm was evident within 3 days. However, in areas with cooler night temperatures, such as the Southern Rolling Plains, Confirm took 7 or more days for activity to become evident. Confirm works on the insect faster than Dimilin.

When using any insecticide, coverage is critical. This is especially critical for the IGR insecticides, Dimilin and Confirm, which are acquired by the larvae primarily through feeding. If treating with conventional ground equipment, high nozzle pressure (60 PSI) and high spray volume (>10 GPA) will improve control.

A new insecticide, Pirate (chlofenapyr), also has a Section 18 label for 1996 and is very efffective against the BAW. Pirate represents a new class of insecticide chemistry known as the pyrroles. The mode of action is to stop ATP formation in the mitochondria resulting in cell, tissue and insect death. Before Pirate can be used, thresholds of either 5 active hits per 100 row feet, or 20,000 larvae per acre must be reached according to the Section 18 label. Threshold levels of BAW must be verified by an Extension Entomologist as being present in at least one field in an Extension District before Pirate can be sold in that District. A maximum of 2 applications will be allowed. Refer to the Section 18 label for specific information.

Spod-X is a viral product, which is specific for BAW, and has worked well in some trials. Multiple applications (3-4) on a 2-3 day interval may be required for this product.

Remedial Thresholds.

Pre-squaring. The BAW can damage terminals, however no thresholds have been established for pre-squaring cotton. A rule of thumb might be 20-30 larvae per 100 plants, though high larval mortality in seedling cotton should be expected. Certain insecticides may provide better control in pre-squaring cotton than in later cotton (including B.t., Larvin, Lorsban and others). Keep in mind early season application of conventional synthetic insecticides may predispose the crop to a BAW outbreak.

Mid-season. The remedial threshold for advanced infestations in mid-season (where 20-60 days of squaring and blooming remain) is not well defined, but we suggest continuing to use a threshold of 20,000 BAW larvae per acre with a general field distribution of 10% or more of the plants infested. This is about 7-12 larvae per 6 feet of row, depending on row spacing and plant population (or, for example, 40 larvae per 100 plants if 50,000 plants per acre are present). See the early detection threshold section above for a discussion on the use of Confirm or Pirate, which can also be used if the remedial threshold is reached. Conventional insecticides listed for BAW in the TAEX Cotton Guide are Lorsban, Lannate, Curacron, Bolstar, and Larvin. None of these conventional insecticides are very effective when used alone against BAW in mid to late season, but Dimilin in combination with high rates of Lorsban or Larvin has provided good control in various tests in South Texas (Sparks et al. 1996). There is some indication, however, that differences in BAW susceptibility to various insecticides varies with populations. Pyrethroids are not effective for BAW control.

Post-cutout. For this growth stage, a threshold of 10 or more BAW per row foot appears reasonable, because feeding on squares in late season does not result in serious yield loss. The BAW does not usually cause extensive damage to mature bolls on post-cutout plants.

Transgenic cottons (B.t.). Thresholds for genetically engineered B.t. cottons have not been established. There are some data, however, which indicate that mortality of BAW on B.t. cottons may be 25% (or possibly more). We suggest using the remedial thresholds discussed above, because this would allow both natural mortality from beneficial insects and mortality from the insecticidal protein in B.t. cotton to be considered in making the decision to treat with a foliar insecticide.

Parasites and Predators. Natural enemies play a key role in suppressing BAW populations. At least 10 parasite species and 13 predator species have been identified (in Georgia) as attacking BAW eggs and larvae (Ruberson et al. 1993, 1994b). Most of these parasite/predator species, or close relatives, are also found in Texas (Benedict et al. 1996, Harding 1976, Puterka et al. 1985, Shepard and Sterling 1972). This parasite/predator complex normally helps keep BAW populations at low numbers. For example, mortality of BAW eggs, larvae and pupae in untreated cotton may exceed 95% due to natural mortality factors (Ruberson et al. 1994b). Many of these BAW parasite/predator species also attack tobacco budworm and bollworm in cotton.

One of the most important predators of BAW eggs, larvae and pupae is the red imported fire ant (Ruberson et al. 1994b), while one of the most effective parasites is a tiny braconid wasp, Cotesia marginiventris, which attacks the first through fourth larval stages of the BAW. It is especially effective at finding and attacking small larvae while they are still feeding in a group. C. marginiventris is a widely distributed parasite that attacks a wide range of caterpillar species, including BAW, in a variety of habitats. Its population dynamics are not exclusively dependent on insects in cotton. In fact, C. marginiventris is a common parasite attacking bollworm in cotton, corn and sorghum in the lower Gulf Coast (Benedict et al. 1996). The pupal stage of this parasite is small, white, fuzzy and oblong; and is often found attached to the underside of a cotton leaf. Important egg parasites of bollworm and tobacco budworm include several species of Trichogramma wasps. These are also general parasites of many insect eggs, but do not commonly parasitize egg masses of the BAW. However, Chelonus insularis, a braconid wasp, (Ruberson et al. 1994b) has been found parasitizing BAW eggs and larvae in cotton as well as bollworm in corn and sorghum in Texas (Benedict et al. 1996).

Insecticide Toxicity to Beneficials. An important question when treating BAW and other insect pests of cotton is: Which insecticides are least harmful to the parasite/predator complex yet still provide good control of the pest insect? The general order of toxicity of insecticides to beneficial insects is as follows (highest to lowest): organophosphates > pyrethroids > carbamates, pyrroles, endosulfan > fipronil, amitraz > insect growth regulators, B.t.=s, viruses. However, generalities about all insecticides are difficult to make. There can be considerable variation in response between individual insecticides in a class and between beneficial insect species.

Within the organophospate (OP) class, the gram-for-gram active ingredient toxicity to parasites and predators, including C. marginiventris, varies among individual insecticides. Certain OP insecticides (phosphorothionates), such as methyl parathion and Lorsban, are very lethal to parasites. Another group of OP's (phosphorodithionates), such as Guthion, malathion and Bolstar, are relatively intermediate in parasite toxicity. A third group of OP's (phosphorothiolates), such as Curacron, may be less detrimental (gram-for-gram) than the above. The carbamates at low rates, such as Larvin, appear slightly Asafer@ than the OP=s.

The gram-for-gram toxicity of pyrethroids to beneficial parasites also varies. Single isomer pyrethroids, which includes most of the pyrethroids, are quite toxic; mixed isomer pyrethroids, such as Ammo and Cymbush, are less toxic. Pyrethroids are also very toxic to fire ants.

Based on several field studies with C. marginiventris and Microplitis croceipes (a braconid wasp), the Asafest@ synthetic insecticides among those tested in the studies appear to be (using lowest labeled rates): cypermethrin (Ammo), esfenvalerate (Asana), oxamyl (Vydate) and thiodicarb (Larvin) (Powell and Scott 1991, Ruberson et al. 1993). However, under field application conditions, toxicity to beneficials depends on differences in their exposure to the insecticide, on the rate of insecticide used, and on the beneficial species of concern. It should be kept in mind that most of these conventional synthetic insecticides have toxic residues that may last 5 or more days, and over time, most appear to be fairly toxic, killing 50 to 100% of adult parasites. Therefore, even those carbamates and pyrethroids considered to be relatively "safe" are quite toxic to many beneficials, especially when used at relatively higher rates and in multiple applications.

The topic of insecticide selectivity (i.e., identifying insecticides that are safe to beneficial insects but highly toxic to pest insects) is quite complex, with many variables involved, and more research is needed (Croft 1990). At this time we believe that the Asafest@ insecticides are the biological insecticides (Bacillus thuringiensis and various viruses), Bt cotton, Dimilin and Confirm (not listed in order of safety). The use of scouting and economic thresholds for cotton pests is recommended to avoid unnecessary insecticide applications and preserve naturally occurring beneficial arthropods as much as possible.

Crop Earliness. The BAW has demonstrated a strong preference for late-planted or late-maturing cotton. Efforts should be made to maintain a short season cotton production system, where possible. In South Texas, this includes planting as early as feasible, making variety selection to achieve relatively rapid fruiting and maturity, and protecting early fruit from insect damage. Uniform delayed planting programs for boll weevil management in parts of West Texas should not be modified, but standard management practices to promote earliness should be used.

Pheromone Traps. The use of BAW pheromone traps (which are compact and relatively easy to use) can be a valuable tool as an early warning device to detect the presence of male moths early in the season. If distinct generation cycles are present, as occurred in South Texas in 1995, pheromone trap catches of adults might also be used to alert producers as to when to intensify efforts to scout for egg masses and newly hatching larvae. Trap catches should, however, be interpreted with caution, because they do not always correlate well with BAW activity in the field.

Information in this publication was obtained from various sources. We would like to especially acknowledge the contributions of the following: Dr. Ron Smith, Extension Entomologist, Auburn University; Dr. Blake Layton, Extension Entomologist, Mississippi State University; Dr. Gary Herzog, Associate Professor, Entomology, University of Georgia; Dr. Bill Lambert, Extension Entomologist, University of Georgia; Dr. John Ruberson, Research Entomologist, University of Georgia; Dr. Bill Plapp, Department of Entomology, Texas A&M University (retired); Ken Buchert, Rohm and Haas; Dr. Al Dalrymple, Uniroyal; Dr. Tim Weiland, Uniroyal; Darwin Anderson, Planter's Coop, Odem; Dr. Sid Hopkins, Hopkins Ag Services; Dr. Mike Treacy, American Cyanamid.

Benedict, J. H., M. H. Walmsley, K. M. Schmidt, and B. R. Maher. 1996. Parasitism of bollworm and tobacco budworm larvae in corn, cotton, and sorghum in the lower Gulf Coast of Texas. Southwest. Entomol. (In Preparation).

Croft, B. A. 1990. Arthropod biological control agents and pesticides. John Wiley, New York.

French, R. A. 1969. Migration of Laphygma exigua (Hhbner (Lepidoptera: Noctuidae) to the British Isles in relation to large-scale weather systems. J. Anim. Ecol. 38: 199-210.

Harding, J. A. 1976. Heliothis spp.: Parasitism and parasites plus host plants and parasites of the beet armyworm, diamondback moth and two tortricids in the Lower Rio Grande Valley of Texas. Environ. Entomol. 5: 669-671.

Huffman, R. 1996. Beet armyworm, Texas/Oklahoma. Cotton Insect Management Workshop. In D. A. Richter and J. Amour (eds.) Proc. Beltwide Cotton Production Conference, National Cotton Council of Am., Memphis TN. (In Press).

Insects and related pests of field crops. 1982. Department of Ag. Comm., North Carolina State University, Raleigh, NC. Pub # Ag-271.

Mitchell, E. R. 1979. Migration of Spodoptera exigua and S. frugiperda C North American style, pp. 386-393. In Movement of highly mobile insects. Concepts and methodology in research. North Carolina State University, Raleigh, NC.

Powell, J. E. and W. P. Scott. 1991. Survival of Microplitis croceipes (Hymenoptera: Braconidae) in contact with residues of insecticides on cotton. Environ. Entomol. 20: 346-348.

Puterka, G. J., J. E. Slosser, and J. R. Price. 1985. Parasites of Heliothis spp. (Lepidoptera: Noctuidae): Parasitism and seasonal occurrence for host crops in the Texas Rolling Plains. Environ. Entomol. 14: 441-446.

Ruberson, J. R., G. A. Herzog, and W. J. Lewis. 1993. Parasitism of beet armyworm, Spodoptera exigua, in south Georgia cotton, pp. 993-997. In D. J. Herber and D. A. Richter (eds.) Proc. Beltwide Cotton Production Conference, National Cotton Council of Am., Memphis TN.

Ruberson, J. R., G. A. Herzog, W. R. Lambert, and W. J. Lewis. 1994a. Management of beet armyworm: Integration of control approaches, pp. 857-859. In D. J. Herber and D. A. Richter (eds.) Proc. Beltwide Cotton Production Conference, National Cotton Council of Am., Memphis TN.

Ruberson, J. R., G. A. Herzog, W. R. Lambert, and W. J. Lewis. 1994b. Management of beet armyworm (Lepidoptera: Noctuidae) in cotton: Role of natural enemies. Florida Entomol. 77: 440-453.

Shepard, M. and W. Sterling. 1972. Incidence of parasitism of Heliothis spp. (Lepidoptera: Noctuidae) in some cotton fields of Texas. Entomol. Soc. Am. 65: 759-760.

Sparks, A. N., J. W. Norman, and D. Wolfenbarger. 1996. Efficacy of selected insecticides against beet armyworm, Spodoptera exiguaCField and laboratory evaluations. In D. A. Richter and J. Amour (eds.) Proc. Beltwide Cotton Production Conference, National Cotton Council of Am., Memphis TN. (In Press).

Stewart, S. D., M. B. Layton, and M. R. Williams. 1996. Occurrence and control of beet armyworm outbreaks in the cotton belt. In D. A. Richter and J. Amour (eds.) Proc. Beltwide Cotton Production Conference, National Cotton Council of Am., Memphis TN. (In Press).

Summy, K. R., J. R. Raulston, and J. Vargas. 1996. An analysis of the 1995 beet armyworm outbreak on cotton in the lower Rio Grande Valley. In D. A. Richter and J. Amour (eds.) Proc. Beltwide Cotton Production Conference, National Cotton Council of Am., Memphis TN. (In Press).

The information given herein is for educational purposes only. Reference to commercial products or trade names is made with the understanding that no discrimination is intended and no endorsement by the Cooperative Extension Service is implied.

Return to top of page

Return to Departmental Home Page