HORTICULTURAL ENTOMOLOGY Integrating Chemical and Biological Control of European Corn Borer in Bell Pepper

ANNA V. CHAPMAN, THOMAS P. KUHAR,1 PETER B. SCHULTZ,2 TIMOTHY W. LESLIE,3 3 4 5 SHELBY J. FLEISCHER, GALEN P. DIVELY, AND JOANNE WHALEN

Department of Entomology, Virginia Polytechnic Institute & State University, Eastern Shore AREC, Painter, VA 23420

J. Econ. Entomol. 102(1): 287Ð295 (2009) ABSTRACT Using multiple locations and a series of Þeld trials over 2 yr, we evaluated an integrated pest management program for Ostrinia nubilalis (Hu¨ bner) (Lepidoptera: Crambidae) management in peppers involving biorational chemistries, inundative releases of Trichogramma ostriniae (Pang & Chen), and conservation of generalist predators. In small plot trials, three biorational insecticides (spinosad, indoxacarb, and methoxyfenozide) provided comparable control of O. nubilalis as two broad-spectrum conventional insecticides (acephate and lambda-cyhalothrin). However, lambda- cyhalothrin at most locations, and indoxacarb at one location, resulted in outbreaks of green peach aphids. We also observed signiÞcant effects on the generalist predator community: beneÞcial com- munities in methoxyfenozide-treated plots were most similar to untreated controls, and acephate- treated plots were the least similar. Management systems comparing untreated controls, inundative release of T. ostriniae with methoxyfenozide applied when lepidopterans exceeded thresholds, or weekly applications of acephate or lambda-cyhalothrin, showed no effects on marketable fruit or percentage of fruit damaged, but the conventional insecticide approach caused aphid ßares. Inun- dative releases of T. ostriniae and biorational chemistries provide a more environmentally sound approach to managing O. nubilalis in peppers, due, in part, to conservation of generalist predators.

KEY WORDS Ostrinia nubilalis, Trichogramma ostriniae, pepper, biocontrol, IPM

Sweet bell pepper (Capsicum annuum L.) is a valuable enemies in the agroecosystem, environmental and hu- vegetable crop, grossing nearly $500 million annually man health risks, and reduced proÞts. in the United States (NASS 2006). Commercial grow- For pepper growers in the northeastern and central ers have a large Þnancial investment once plants begin United States, European corn borer, Ostrinia nubilalis to set fruit, and they often apply insecticides preven- (Hu¨ bner) (Lepidoptera: Crambidae), is the primary tatively to protect the pepper fruit from pests. target of insecticide sprays (Welty 1995, Hazzard et al. Pepper growers in the Mid-Atlantic United States gen- 2001). The insect has two to three generations per year erally apply the organophosphate acephate for two and is a season-long risk to bell pepper. Egg masses are sprays (maximum allowed per crop per season) and deposited on the undersides of leaves (Barlow and then apply a pyrethroid insecticide every 5Ð10 d until Kuhar 2004). After a brief period of leaf feeding, larvae Þnal harvest. This conventional (preventative spray- bore into fruit if available or stems if no fruit or only ing) approach may result in Þve to 10 insecticide very small immature fruit are present (Hitchner and applications per crop. Although it is generally effec- Ghidiu 2006). Direct damage is caused by the tunnel- tive (Welty 1995, Kuhar and Speese 2002, Kuhar et al. ing larvae that feed on the pericarp, placenta, and 2003), it has numerous drawbacks associated with un- seeds of developing fruit. Moreover, tunnel holes fre- warranted applications, including potential buildup of quently are entry points for fruit-rotting pathogens pesticide residues, destruction of important natural such as Erwinia caratovora (Hazzard et al. 2001). Fruit damage can range from 40 to 60% (Welty 1995, Kuhar and Speese 2002, Kuhar et al. 2003, Welty and Vitanza 2005). Even greater fruit damage can occur to ma- 1 Corresponding author, e-mail: [email protected]. ture (colored) bell peppers because of the in- 2 Hampton Roads AREC, Virginia Tech, Virginia Beach, VA 23455- 3315. creased time that fruit are in the Þeld. O. nubilalis 3 Department of Entomology, Pennsylvania State University, Uni- is particularly difÞcult to control because larvae are versity Park, PA 16802. only exposed to insecticide sprays from egg hatch 4 Department of Entomology, University of Maryland, College until tunneling. Park, MD 20742. 5 Department of Entomology and Applied Biology, University of Other that may infest pepper fruit include Delaware, Newark, DE 19717-1303. the noctuid pests Helicoverpa zea (Boddie); Spodopt-

0022-0493/09/0287Ð0295$04.00/0 ᭧ 2009 Entomological Society of America 288 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 102, no. 1 era exigua Hu¨ bner; Spodoptera frugiperda (J.E. Smith); Materials and Methods and pepper maggot, Zonosemata electa (Say) (Diptera: Biorational Insecticide Trials. We evaluated the Tephritidae) (Boucher and Ashley 2001). However, effects of biorational chemistries on O. nubilalis dam- Kuhar et al. (2004) found that this complex of pests age and natural enemy communities in 2005, by using only accounted for Ϸ5% of insects found attacking bell small-plot Þeld experiments at four locations: Virginia pepper fruit in Virginia, with most (95%) being O. Tech Eastern Shore Agricultural Research and Exten- nubilalis. In addition, green peach aphid, Myzus per- sion Center (AREC), Painter, VA; University of Del- sicae (Sulzer), can be an important indirect pest of aware REC, Georgetown, DE; University of Maryland peppers by feeding on plant phloem sap, reducing Wye REC, Queenstown, MD; and the Russell E. Lar- plant vigor, vectoring viruses, and depositing honey- son Research Farm, Rock Springs, PA. Each Þeld con- dew, where black sooty mold fungus may grow. Green sisted of treatments in a randomized complete block peach aphid has a tremendous reproductive capabil- design with four replications of six insecticide treat- ity, and it can quickly build to damaging levels if ments. Individual plots were four rows wide by 6 m populations are not contained (van Emden et al. long with Ͼ3-m alleys between plots. Crops were 1969). established and maintained using standard agricul- Biological control of O. nubilalis eggs using aug- tural procedures including fertilizer, herbicide, and mentative releases of Trichogramma parasitoids may fungicide applications according to a commercial veg- provide an alternative to reliance on insecticides etable production manual for the Mid-Atlantic United (Smith 1996), and thus help conserve beneÞcials that States (Kuhar et al. 2006b, see Table 1 for details at could regulate aphid densities. Trichogramma ostriniae each location). Insecticide treatments included 1) in- Pang & Chen (: Trichogrammatidae) doxacarb at 0.072 kg (AI)/ha (Avaunt 30 WG; E. I. du was introduced into the United States from China in Pont de Nemours and Co.); 2) methoxyfenozide at the early 1990s (Hoffmann et al. 1995), and it has 0.112 kg (AI)/ha (Intrepid 2 F; Dow AgroSciences shown promise for controlling O. nubilalis in corn LLC); 3) spinosad at 0.026 kg (AI)/ha (SpinTor 2 SC, (Wang et al. 1999, Hoffmann et al. 2002, Kuhar et al. Dow AgroSciences LLC); 4) acephate at 1.09 kg 2002, Wright et al. 2002) and solanaceous crops (Ku- (AI)/ha (Orthene 97SG; Valent USA Corporation, har et al. 2004). After four to Þve inundative releases Walnut Creek, CA); 5) lambda-cyhalothrin at 0.03 kg of T. ostriniae in bell pepper, O. nubilalis egg parasitism (AI)/ha (Syngenta Crop Protection Inc., Greensboro, Ϸ averaged 50%, and cumulative fruit damage was re- NC); and 6) an untreated control. Insecticides were duced almost 70% compared with nonrelease control applied with a CO2 backpack sprayer with a one-row plots (Kuhar et al. 2004). However, under heavy pest boom having three hollow-cone nozzles per row (one pressure, pepper fruit damage still exceeded 10% in over the top and one drop nozzle on each side) de- the T. ostriniae release plots. livering 39 gpa at 40 psi. Treatments were applied Reduced-risk (biorational) insecticides offer an al- weekly from Þrst fruit until Þnal harvest. Dates of ternative to organophosphates and pyrethroids for applications for each location are presented in Table controlling O. nubilalis in pepper. In recent years, a 1. On three postspray sample dates, aphids were few narrow-spectrum (more lepidopteran speciÞc) counted on 50 randomly picked leaves per plot. Rel- insecticides have been registered for use on bell pep- ative density of natural enemies (generalist per, including spinosad (Spintor, Dow AgroSciences predators and parasitoid species) were estimated by LLC, Indianapolis, IN), methoxyfenozide (Intrepid, slowly walking the entire length of one middle row per Dow AgroSciences LLC), and indoxacarb (Avaunt, plot while vacuuming plants using a Craftsman 200 E. I. DuPont de Nemours and Co., Wilmington, DE). mph leaf blower-vac. Samples were collected using a These insecticides provide effective control of most Þne mesh bag Þtted on the intake tube of diameter 12 lepidopteran pests, whereas generally having greatly cm. After each sample was collected, bags were tied reduced toxicity to natural enemies (Brunner et al. shut and frozen. Any dirt or plant debris was removed 2001; Elzen 2001; Baur et al. 2003; Hewa-Kapuge et al. and all insects were stored in 70% ethanol for later 2003; Hill and Foster 2003; Schneider et al. 2003; Stude- identiÞcation and sorting into seven natural enemy baker and Kring 2003a, 2003b; Carton et al. 2003; Has- groups: ladybird beetles; predatory bugs; parasitic eeb et al. 2004, 2005; Villanueva and Walgenbach, hymenoptera; lacewings; predatory Diptera; spiders, 2005). and other. Here, we evaluated strategies to reduce broad-spec- On three separate dates during a one month span in trum insecticide use on bell pepper in the Mid-Atlan- late summer (Table 1), all market-sized fruit were tic United States. In one experiment, the efÞcacy and harvested from the middle two rows of each plot and relative impact on beneÞcial of spinosad, assessed for insect damage. O. nubilalis damage to bell methoxyfenozide, and indoxacarb were compared pepper can manifest itself as either reduced number of with the conventional broad-spectrum insecticides harvested fruit from rotting, or percentage of har- acephate and lambda-cyhalothrin; and in a second vested fruit exhibiting damage. The cumulative num- experiment, the efÞcacy of a T. ostriniae-based bio- ber of marketable fruit and the percentage damaged logical control program was compared with a conven- by insects were recorded. Damaged fruit were dis- tional insecticide-based program. sected and inspected for any insects present. February 2009 CHAPMAN ET AL.: EUROPEAN CORN BORER IPM IN PEPPERS 289

Data were analyzed using analysis of variance (ANOVA) (Analytical Software 2003) to test for sig- niÞcant treatment effects on marketable fruit yield, proportion damaged fruit, and peak density of aphids on leaves. To stabilize variances all proportion damage data were arcsine square-root transformed, and aphid

density data were log10 x-transformed. If the treatment evaluation dates source of variation was signiÞcant, differences among

Harvest and fruit damage treatment means were tested using FisherÕs protected least signiÞcant difference (LSD) at the P Յ 0.05 level of signiÞcance.

d States in 2005 and 2006 To determine whether insecticide treatments inßu- enced the beneÞcial community and to investigate patterns attributable to insecticide, we performed a

release dates redundancy analysis (RDA) by using location as a covariable with CANOCO 4.5 (ter Braak and Sˇmilauer in IPM plots 2002). RDA is an ordination technique that can be useful for identifying associations between response T. ostriniae variables and explanatory variables when working with complex multivariate data. RDA identiÞes the prominent gradients (i.e., orthogonal axes) among the response variables, as constrained by the treatment variables, and determines the signiÞcance of the axes using Monte Carlo permutational procedures (ter Braak and Sˇmilauer 2002). To visualize trends, the response and treatment variables were plotted on the Þrst two axes (a biplot), in which proximity indicates degree of associa- tion. Forward selection was used to identify signiÞcant treatment factors. All natural enemy data were square- in conventional plots root transformed in CANOCO 4.5. Insecticide application dates Management Systems Experiments. In 2005 and 2006, we established three spatially isolated plots of Ϸ Ϸ 8, 15, 22, 29 July, and 5 Aug.24, 31 July, and 7, 14, 21, 28 Aug. 19, 27 21, July, 28 and July 3, and 14 4 Aug. Aug. 26 July and 8, 21 Aug. 7, 17, 24, 31 Aug. and 7 Sept. 29 July, and 3, 11, 17, 24, 31 Aug. 3, 10, 17 Aug. 5, 17, 29 Aug. 25 July, and 1, 9, 15, 22, 29 Aug. 25 July, and 3, 12, 20 Aug. 20 Aug., and 1, 16 Sept. 16, 22, 28 June, 6, 13, 20, 27 July, and 5 Aug. 29 June, and 12, 19 July 12, 26 July, and 15 Aug. peppers (each 0.03 ha and 500 plants) at each of Þve locations: Virginia Tech Eastern Shore AREC, Painter, VA; Virginia Tech Hampton Roads AREC, VA Beach, VA; University of Delaware AREC, George- town, DE; University of Maryland Wye AREC, Queen- stown, MD; and the Russell E. Larson Research Farm, Rock Springs, PA (Table 1). Each location repre- sented a replicate (n ϭ 5) in a randomized complete block experiment that was analyzed separately by year. Crops were established and maintained using standard agricultural procedures including fertilizer, Crop production method herbicide, and fungicide applications following a com- drip irrigation drip-line irrigation drip-line irrigation plastic with drip irrigation drip-line irrigation mercial vegetable production manual designed for the Mid-Atlantic United States (Kuhar et al. 2006b; details at each location are presented in Table 1). At each location, the three pepper plots were ran- date release, all information for the four locations in 2005domly marked by an asterisk was the same for the biorational insecticide experiments. designated as 1) conventional, which involved Transplant two applications of acephate at 1.09 kg (AI)/ha initi- ated at Þrst fruiting followed by four or Þve weekly applications of lambda-cyhalothrin at 0.03 kg (AI)/ha from Þrst fruit until Þnal harvest; 2) IPM, which in- Trichogramma cluded three or four weekly inundative releases of T. PaladinKing Arthur 13 June 16 June Double-staggered rows Single on rows black with overhead irrigation 24 July, 1, 8, 15, 21, 29 Aug., and 8 Sept. 19, 26 Aug., and 16 Sept. PaladinAristotle 14 June 27 May Ground with overhead irrigation Single rows on black plastic with 21, 28 July, and 3, 9, 16, 23, 31 Aug. 21 July, and 2, 12 Aug.ostriniae 11, 22 Aug., and 7 Sept. starting with Þrst fruiting and an application *

* of methoxyfenozide at 0.112 kg (AI)/ha only if lepi- * dopteran pests exceeded action thresholds; and 3) an

* untreated control. In the IPM plot, we released 100,000 T. ostriniae per acre on three or four dates

Location Variety (Table 1) by using perforated cardboard release car-

Table 1. Crop and pest management information for each location of bell pepper management systems experiments in the Mid-Atlantic Region of the Unite tons and methods described in Kuhar et al. (2004). The With the exception of the Rock Springs, PA Paladin 8 June Single rows on black plastic with Queenstown, MDGeorgetown, DE Paladin Aristotle 6 June 5 June Single rows with overhead Single irrigation rows on black plastic with 12, 20, 31 July, 17, 24 Aug. and 4 Sept. 4, 12 Aug. 27 July, 11 and 23 Aug. Rock Springs, PA Painter, VAVirginia Beach, VA Paladin Paladin 7 June 7 June Single rows on black plastic with Ground with overhead irrigation 28 July, and 4, 11, 18, 25 Aug., and 5 Sept. 3, 10, 17 Aug. 14, 24, and 5 Sept. Queenstown, MD Virginia Beach, VAGeorgetown, DE Paladin 6 June Ground with drip-line irrigation 22, 29 July, and 4, 12, 19, 30 Aug. 22 July, and 3, 12 Aug. 26 July, 4, 16 Aug., and 1 Sept. Painter, VA 2006 2005 parasitoids were obtained from M. P. Hoffmann (Cor- 290 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 102, no. 1

Table 2. Influence of insecticide applications on green peach aphid on bell peppers at four locations in the Mid-Atlantic United States

Mean Ϯ SE no. green peach aphids per 10 leaves Rate kg Georgetown, DE Rock Springs, PA Painter, VA Treatmenta (AI)/ha After 4 After 6 After 7 After 2 After 5 After 6 After 3 After 6 After 7 sprays sprays sprays sprays sprays sprays sprays sprays sprays Lambda-Cyhalothrin 0.03 44.2 Ϯ 33.2a 81.2 Ϯ 23.8a 139.4 Ϯ 38.0a 0.1 Ϯ 0.1a 21.9 Ϯ 10.4a 35.9 Ϯ 7.9a 0.6 Ϯ 0.4a 76.3 Ϯ 25.6a 158.0 Ϯ 46.7a Acephate 1.09 0.0 Ϯ 0.0b 0.2 Ϯ 0.1b 0.2 Ϯ 0.2b 0.0 Ϯ 0.0a 0.0 Ϯ 0.0b 0.0 Ϯ 0.0b 0.0 Ϯ 0.0a 0.0 Ϯ 0.0b 0.0 Ϯ 0.0b Spinosad 0.026 0.8 Ϯ 0.4b 0.8 Ϯ 0.4b 3.4 Ϯ 0.7b 0.5 Ϯ 0.2a 4.6 Ϯ 1.2b 4.6 Ϯ 4.2b 0.2 Ϯ 0.1a 0.0 Ϯ 0.0b 0.5 Ϯ 0.5b Indoxacarb 0.072 7.1 Ϯ 5.2ab 55.3 Ϯ 39.7a 139.8 Ϯ 84.5a 0.5 Ϯ 0.3a 1.9 Ϯ 1.0b 0.3 Ϯ 0.1b 0.2 Ϯ 0.1a 0.8 Ϯ 0.3b 3.2 Ϯ 1.7b Methoxyfenozide 0.112 0.7 Ϯ 0.4b 0.5 Ϯ 0.4b 2.9 Ϯ 1.1b 0.4 Ϯ 0.3a 2.4 Ϯ 0.5b 13.0 Ϯ 12.2b 0.2 Ϯ 0.2a 0.3 Ϯ 0.3b 0.7 Ϯ 0.7b Untreated control 0.9 Ϯ 0.5b 1.0 Ϯ 0.3b 2.0 Ϯ 0.4b 0.4 Ϯ 0.3a 0.4 Ϯ 0.2b 0.0 Ϯ 0.0b 0.7 Ϯ 0.6a 0.0 Ϯ 0.0b 0.4 Ϯ 0.4b

Means within a column with a letter in common are not signiÞcantly different according to ANOVA and FisherÕs protected LSD at the P ϭ 0.05 level of signiÞcance. a All insecticides were applied Ϸweekly beginning at Þrst appearance of fruit. nell University, Ithaca, NY), details on the T. ostriniae cations of these insecticides resulted in even greater colony maintenance can be found in Hoffmann et al. aphid densities, whereas aphid densities remained low (1995). Insect monitoring followed guidelines out- in control plots (Table 2). lined in Boucher and Ashley (2001), and involved Lepidopteran pest pressure was relatively low blacklight or pheromone trap monitoring of O. nubi- across all locations, and there was no signiÞcant treat- lalis and S. frugiperda and S. exigua moths (Lepidop- ment effect on cumulative number of marketable (un- tera: Nocutidae). Weekly inspections of 40 plants per damaged) fruit in Maryland (F ϭ 0.67; df ϭ 5, 23; P ϭ plot involved looking for aphids and eggs on the un- 0.6534), Pennsylvania (F ϭ 1.80; df ϭ 5, 23; P ϭ 0.1730), derside of leaves and examining any present fruit for and Virginia (F ϭ 0.27; df ϭ 5, 23; P ϭ 0.9243); how- tunnels or frass, indicating the presence of larvae of ever, there was a consistent numeric trend toward the lepidopteran pest species. As described in Boucher untreated control plots yielding the least number of and Ashley (2001), insecticide applications for O. nu- marketable fruit (Fig. 1A). There was a signiÞcant bilalis were applied on IPM plots 1 wk after phero- treatment effect on the percentage of fruit damaged mone trap counts reached seven moths per week and by O. nubilalis in Pennsylvania (F ϭ 6.35; df ϭ 5, 23; the pepper crop had reached bloom stage. Insecticide P Ͻ 0.0024) and Virginia (F ϭ 7.56; df ϭ 5, 23; P Ͻ spray dates at each location are shown in Table 1. 0.0010), where the untreated control plots at both When they reached marketable size, pepper fruit locations had more damage than any of the insecticide were hand-harvested and evaluated for damage from treatments, and no differences were found among the July to October (speciÞc dates are listed in Table 1). Data insecticide treatments (Fig. 1B). No signiÞcant treat- were pooled across the Þve locations and analyzed by ment effect on fruit damage was found at the Maryland year using ANOVA (Analytical Software 2003) to test for location (F ϭ 1.56; df ϭ 5, 23; P ϭ 0.2308). Fruit signiÞcant treatment effects on marketable fruit yield, damage and concomitant number of marketable fruit proportion damaged fruit, and peak density of aphids on was not assessed at the Delaware location. leaves. Proportion data were square-root transformed The predominant natural enemies were parasitic before analysis to stabilize variance. Differences among Hymenoptera (mostly Chalcidoidea, Braconidae, and treatment means were tested using Fisher Protected Ichneumonidae), predatory bugs (mostly Anthocori- LSD at the P Յ 0.05 level of signiÞcance. dae, Orius spp.), ladybird beetles (Coccinellidae), and spiders (Aranea). RDA identiÞed a signiÞcant primary (F ϭ 8.78, P ϭ 0.002) and secondary (F ϭ 2.41, P ϭ Results 0.002) gradient among the beneÞcial insect commu- Biorational Insecticide Trial. Green peach aphid nity as constrained by our treatment variable (i.e., infested pepper plots at all four locations, but it did not insecticide) and together they explained 11.6% of the reach damaging levels in the untreated control plots variance in the community data. The biplot (Fig. 2) on any sample date at any location. However, after indicates that the communities associated with lamb- four or more insecticide applications, there was a da-cyhalothrin and acephate differed from the other highly signiÞcant treatment effect on numbers of treatments along the Þrst axis, and forward selection aphids at three of the four locations: Virginia (F ϭ identiÞed lambda-cyhalothrin as the predominant 23.43; df ϭ 5, 23; P Ͻ 0.0001), Pennsylvania (F ϭ 6.35; variable inßuencing this gradient (F ϭ 4.57, P ϭ 0.002). df ϭ 5, 23; P Ͻ 0.0024), and Delaware (F ϭ 7.99; df ϭ Lacewings, spiders, and predatory bugs responded 5, 23; P Ͻ 0.0008). In Virginia and Pennsylvania, aphid positively to indoxacarb, methoxyfenozide, and the densities were signiÞcantly higher in the lambda-cy- untreated control, and they were negatively corre- halothrin plots compared with any other insecticide lated to lambda-cyhalothrin and acephate (Fig. 2). treatment or the control, and in Delaware, aphid den- The second axis represented community differentia- sities were signiÞcantly higher in the lambda-cyhalo- tion between lambda-cyhalothrin and acephate (F ϭ thrin and indoxacarb plots compared with any other 5.26, P ϭ 0.002). Ladybird beetles, braconids, and insecticide or the control (Table 2). Additional appli- other hymenopterans (excluding ichneumonids) February 2009 CHAPMAN ET AL.: EUROPEAN CORN BORER IPM IN PEPPERS 291

220 200 Lambda-cyhalothrin A 180 Acephate

160 Spinosad

140 Indoxacarb

120 Methoxyfenozide

100 Untreated control fruit / 100 plants 80

60 Cumulative no. of marketable 40

20

0 Salisbury, MD Rock Springs, PA Painter, VA 50 45 B 40

35

30

25 a 20

15 b a Percentage fruit damage 10 b b b b b 5 b b b b 0 Salisbury, MD Rock Springs, PA Painter, VA Fig. 1. Inßuence of insecticide applications on marketable fruit yield (A) and percentage of fruit damaged by lepidopteran pests (B) on bell peppers at three locations in the Mid-Atlantic United States in 2005. All insecticides were applied approximately weekly beginning at Þrst appearance of fruit. Means within a location with a letter in common are not signiÞcantly different according to ANOVA and FisherÕs protected LSD at the P ϭ 0.05 level of signiÞcance.

0.3 were more highly associated with lambda-cyhalothrin

Ladybird and not acephate (Fig. 2). Ichneumonids and chalci- 0.2 λ-cyhalothrin dids were negatively correlated with acephate and Lacewing spinosad and uncorrelated with the remaining treat- Ichneum Hymenop 0.1 Spider ments (i.e., found equally among them). BigEyeBg Chalcid Braconid Indox UTC Management Systems Experiments. In both years, 0.0 Methox lepidopteran pest pressure to bell peppers was mod- Orius Spinosad erately low to low at all Þve Mid-Atlantic locations. O. -0.1 Acephate nubilalis was the primary insect pest encountered, and weekly moth catch at each location and year is shown -0.2 in Fig. 3. In 2005, conventional plots averaged seven -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 insecticide applications (two acephate sprays ϩ Þve Fig. 2. Redundancy analysis bipolot, depicting associa- pyrethroid sprays) compared with less than one in- tions between beneÞcial insect taxa (vectors) and six insec- secticide spray of methoxyfenozide in the IPM plots. ticide programs (black diamonds). Abbreviations used in the In 2006, conventional plots averaged 5.8 insecticide Þgure were as follows: Indox, indoxacarb; Methox, methoxy- applications (two acephate sprays ϩ additional pyre- fenozide; UTC, untreated control; Ladybird, Coccinellidae adults and larvae; Lacewing, larvae of Chrysopidae and throid sprays) compared with no insecticide sprays in Hemerobiidae; Ichneum, adult Ichneumonidae; BigEyeBg, the IPM plots. Georcoris spp. adults and nymphs; Chalcid, adult Chal- There was no signiÞcant treatment effect on cumu- cidoidea; Braconid, adult Braconidae; Hymenop, all adult lative number of marketable fruit in 2005 (F ϭ 1.37; Hymenoptera; and Orius, Orius spp. adults and nymphs. df ϭ 2, 6; P ϭ 0.3035) or 2006 (F ϭ 0.7; df ϭ 2, 6; P ϭ 292 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 102, no. 1

1600 Conventional 1400 NS A IPM 1200 Control 1000 800 NS 600 400 fruit / 100 plants fruit / 100 200 0 Cumulative no. of marketable Cumulative 18 16 14 B NS 12 NS 10 8 6 4 2

Percentage fruit damage Percentage 0 1600 a 1400 C 1200

1000

800 600 a 400 200 b b b

Peak no. aphids/ 100 leaves Peak no. aphids/ b 0 2005 2006 Fig. 4. Cumulative yield of marketable fruit (A), per- centage of fruit damaged by lepidopteran pests (B), and peak green peach aphid densities (C) from bell pepper in the Fig. 3. Weekly moth catch of O. nubilalis in blacklight or Mid-Atlantic States (n ϭ 5 locations per year) under three pheromone traps located at Rock Springs, PA (A), Queen- pest management systems: 1) a conventional system of stown, MD (B), Georgetown, DE (C), Painter, VA (D), and weekly acephate or lambda-cyhalothrin applications from Virginia Beach, VA (E), in 2005 and 2006. Þrst fruiting until Þnal harvest; 2) an IPM system of three or four inundative releases of T. ostriniae to control O. nubilalis 0.5239). However, in 2005, across the Þve locations, plus zero, one, or two applications of the insect growth regulator (IGR) insecticide methoxyfenozide; and 3) an un- conventional plots tended to have the most market- treated control. Means within the same year with a letter in able fruit followed by the IPM plots and then the common are not signiÞcantly different according to ANOVA untreated control plots (Fig. 4A). There was also no and FisherÕs protected LSD at the P ϭ 0.05 level of signiÞ- treatment effect on percentage of fruit damaged by cance. lepidopteran larvae in either 2005 (F ϭ 0.15; df ϭ 2, 6; P ϭ 0.8649) or 2006 (F ϭ 2.60; df ϭ 2, 6; P ϭ 0.1351). Percentage of damaged fruit averaged between 4 and conventional acephate or lambda-cyhalothrin treat- 12% among the treatments with the untreated control ments. However, lepidopteran pest pressure was rel- plots typically having the most damage numerically atively low in our study. Nonetheless, similar efÞcacy (Fig. 4B). results with some or all of the aforementioned insec- There was a signiÞcant treatment effect on peak ticides on bell peppers have been shown by Nault and density of green peach aphids in 2005 (F ϭ 5.71; df ϭ Speese (2000), Kuhar and Speese (2002), Kuhar et al. 2, 6; P Ͻ 0.041) and nearly a signiÞcant effect in 2006 (2003, 2006a), and Sorensen and Cooke (2004). In (F ϭ 3.70; df ϭ 2, 6; P Ͻ 0.0896). In both years, the contrast, Welty and Vitanza (2005) found acephate to conventional insecticide plots had the most aphids, be the superior insecticide under extreme O. nubilalis and no differences were found between the IPM and pest pressure on mature red bell pepper in Ohio. untreated control plots (Fig. 4C). Timing of the insecticide applications is critical for O. nubilalis control. Most do not kill the egg stage, and after hatch, larvae quickly tunnel into plants where Discussion they are protected from chemical sprays. A systemic In the 2 yr of our study, the biorational insecticides insecticide such as acephate could exhibit greater ef- spinosad, indoxacarb, and methoxyfenozide provided Þcacy against O. nubilalis because of a longer residual similar control of O. nubilalis in bell pepper as the in the Þeld compared with the newer biorationals. February 2009 CHAPMAN ET AL.: EUROPEAN CORN BORER IPM IN PEPPERS 293

Green peach aphid did not reach damaging levels in controls or methoxyfenozide, along with indoxacarb, the untreated control plots, or in plots treated with were most different than those treated with either of acephate, spinosad, or methoxyfenozide. In contrast, the broad-spectrum insecticidesÑthe organophos- at most locations, multiple sprays of the pyrethroid phate (acephate) or the pyrethroid (lambda-cyhalo- lambda-cyhalothrin caused severe aphid outbreaks. thrin). However, there was a measurable difference in Also at one location, multiple sprays of indoxacarb communities between these two broad-spectrum caused similar aphid ßares. Destruction of arthropod plots (shown by the signiÞcant second axis; Fig. 2). natural enemies by insecticides was likely the cause Plots treated with the systemic organophosphate for the aphid outbreaks (van Emden et al. 1969, Croft acephate were not associated with any beneÞcial taxa and Brown 1975). lambda-Cyhalothrin and acephate in contrast with plots treated with lambda-cyhalo- are highly toxic to most arthropod natural enemies thrin, which associated with some hymenopterans and (Baur et al. 2003). However, acephate provides ex- ladybird beetles. Indeed, although lambda-cyhalo- cellent control of green peach aphid, but pyrethroid thrin is known to be detrimental to beneÞcial insects, insecticides, such as lambda-cyhalothrin, do not the highly mobile parasitic hymenopterans and lady- (Speese 1994 and 1995). Therefore, multiple sprays of bird beetles were more closely associated with the lambda-cyhalothrin or any pyrethroid will typically pyrethroid than the systemic acephate, which may exacerbate green peach aphid problems. Our data have been due to recolonization (Baur et al. 2003) in showed that at least in one Þeld, multiple sprays of response to the aphid ßare documented in these plots. indoxacarb could result in similar aphid outbreaks. The microbial metabolite, spinosad, had no clear pat- Indoxacarb does not control aphids, and is moderately tern of association. In general, this suggests that newer toxic to some natural enemies including predatory biorational insecticides, coupled with multivariate an- hemipterans (Baur et al. 2003) and ladybeetles (Kuhar, alytical tools, enable analysis of beneÞcial communi- unpublished data). Several factors can inßuence the ties in vegetable entomological research, and that me- nontarget impact of an insecticide in the Þeld. For thoxyfenozide had the least affect on beneÞcial example, spinosad was extremely toxic to the hyme- communities among the treatments we tested. nopteran parasitoid Diadegma insulare (Cresson) in Our demonstration plots in the Mid-Atlantic Region leaf dip assays (Hill and Foster 2000); however, in Þeld of IPM versus conventional insecticide-based pro- experiments D. insulare parasitism of diamondback grams in bell pepper did not reveal signiÞcant treat- moth larvae was not affected by spinosad applications ment effects on cumulative number of marketable (Hill and Foster 2003). In addition, the aforemen- fruit or percentage fruit damage. This was clearly tioned insecticides may have sublethal effects on nat- impacted by low lepidopteran pest pressure across the ural enemies, such as parasitoid oviposition rate and two years of our study. The economic implications of emergence (Brunner et al. 2001, Schneider et al. this study suggest a conventional insect management 2004). However, although sublethal effects of both approach could reduce costs by adopting some IPM indoxacarb and spinosad have been documented un- practices during low insect pressure. Based on typical der lab conditions, it is difÞcult to know to what extent grower practices, approximately seven insecticide these effects may occur in the Þeld. Moreover, rapid sprays were made throughout the growing season on degradation of insecticide surface residues in the our conventional plots. Two sprays of acephate ($32/ Þeld would improve their compatibility with natural ha) followed by Þve sprays of lambda-cyhalothrin enemies. This would likely be the case with spinosad ($21/ha) resulted in an approximate cost of $169/ha degradation by photolysis (Viktorov et al. 2002). In for insect control (Knodel 2007). The IPM plots were addition, the translocation and translaminar proper- sprayed once with methoxyfenozide ($39/ha) in 2005 ties of some insecticides make them available in the and none in 2006. Combined with three scheduled host plant tissues for control of leaf feeders, but sur- releases of Trichogramma wasps ($50/ha) insect con- face residues disappear quickly, thus making them safe trol totaled approximately $189/ha. Despite the for parasitoids and most natural enemies (Hoy and weekly spray regimen, the conventional-insecticide Cave 1985, Brunner et al. 2001). treated plots did not have signiÞcantly more fruit or We worked under constraints commonly associated less insect damage than either the IPM or the control with efÞcacy trials in high-value crops, including small plots. Presumably, if a conventional insect manage- plots and the presence of toxins in proximity to all ment program adopted monitoring for European corn plots. The proximity of plots could have masked dif- borer and limited insecticide sprays to when an action ferences in mobile beneÞcial insects associated with threshold was met, less insecticide sprays would be treatments. Combined with small plot size, there was needed thus lowering costs. Under greater insect pres- likely a homogenizing effect on the treatment re- sure, the cost beneÞt of the treatments would likely be sponse. Nonetheless, we were able to demonstrate more apparent. As it stands, future research is needed signiÞcant community-scale effects (Fig. 2). Commu- for continued improvement for the IPM systems of nity gradients among insecticide management tactics interest in this study. Inundative releases of the para- stood out as the Þrst axis, and showed strong similarity sitoid, Trichogramma ostriniae, have been shown to in beneÞcial communities in untreated controls and signiÞcantly reduce O. nubilalis damage in bell pepper plots treated with methoxyfenozide. In both, heterop- in previous studies (Kuhar et al. 2004). Unfortunately, teran predator abundance and density were strongest. in our study, we were unable to determine whether In contrast, beneÞcial communities from untreated this pest management approach can provide compa- 294 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 102, no. 1 rable control of O. nubilalis as a conventional insec- ticides on the beneÞcial egg parasitoid Trichogramma nr. ticide approach. It should be noted, however, that the brassicae (Hymenoptera: Trichogrammatidae) under conventional insecticide treatment resulted in signif- laboratory and Þeld conditions. J. Econ. Entomol. 96: icantly higher densities of green peach aphids on pep- 1083Ð1090. pers compared with either the untreated control plots Hill, T. A., and R. E. Foster. 2000. Effect on insecticides on or the Trichogramma-based IPM plots. Thus, in con- the diamondback moth (Lepidoptera: Plutellidae) and its parasitoid Diadegma insulare (Hymenoptera: Ichneu- ditions of the degree of O. nubilalis pressure we ob- monidae). J. Econ. Entomol. 93: 763Ð768. served over 2 yr in the mid-Atlantic states, integration Hill, T. A., and R. E. Foster. 2003. Inßuence of selected of T. ostriniae releases, combined with conservation of insecticides on the population dynamics of diamondback generalist predators achieved through use of biora- moth (Lepidoptera: Plutellidae) and its parasitoid, Dia- tional insecticides, provides a more environmentally degma insulare (Hymenoptera: Ichneumonidae), in cab- sound pest management option in peppers. bage. J. Entomol. Sci. 38: 59Ð71. Hitchner, E. M., and G. M. Ghidiu. 2006. Fruit size and infestation by European corn borer, Ostrinia nubilalis Acknowledgments Hu¨ bner, in bell pepper. J. Veg. Sci. 12: 101Ð107. Hoffmann, M. P., D. L. Walker, and A. M. Shelton. 1995. This research was supported by a Region III Strategic Biology of Trichogramma ostriniae (Hym.: Trichogram- Agricultural Initiative Grant addressing Food Quality Pro- matidae) reared on Ostrinia nubilalis (Lep.: Pyralidae) tection Act issues from the U.S. Environmental Protection and a survey for additional hosts. Entomophaga 40: 387Ð Agency. 402. Hoffmann, M. P., M. G. Wright, S. A. Chenus, and J. Gardner. 2002. Inoculative releases of Trichogramma ostriniae for References Cited suppression of Ostrinia nubilalis (European corn borer) in sweet corn: Þeld biology and population dynamics. Analytical Software. 2003. Statistix 8.0 Analytical Software Biol. Control 25: 249Ð258. for Windows. Analytical Software, Tallahassee, FL. Hoy, M. A., and F. E. Cave. 1985. Laboratory evaluation of Barlow, V. M., and T. P. Kuhar. 2004. Distribution of egg avermectin as a selective acaricide for use with Metaseiu- masses of the European corn borer, Ostrinia nubilalis, in the bell pepper plant canopy. J. Entomol. Sci. 39: 670Ð672. lus occidentalis (Nesbitt) (Acarina: Phytoseiidae). Exp. Baur, M. E., J. Ellis, K. Hutchinson, and D. J. Boethel. 2003. Appl. Acarol. 1: 139Ð152. Contact toxicity of selected insecticides for non-target Knodel, J. 2007. North Dakota Þeld crop insect management predaceous hemipterans in soybeans. J. Entomol. Sci. 38: guide. North Dakota State University Extension Service, 269Ð277. Fargo, ND. Boucher, T. J., and R. A. Ashley [eds.]. 2001. Northeast Kuhar, T. P., and J. Speese. 2002. Evaluation of foliar insec- pepper integrated pest management (IPM) manual. Uni- ticides for control of European corn borer in peppers, versity of Connecticut Cooperative Extension, Storrs, CT. 2001. Arthropod Manag. Tests 27: E56. Brunner, J. F., J. E. Dunley, M. D. Doerr, and E. H. Beers. Kuhar, T. P., M. G. Wright, M. P. Hoffmann, and S. A. 2001. Effects of pesticides on Colpoclypeus florus (Hy- Chenus. 2002. Life table studies of European corn borer menoptera: ) and Trichogramma platneri (Hy- (Lepidoptera: Crambidae) with and without inoculative menoptera: Trichogrammatidae), parasitoids of leafrol- releases of Trichogramma ostriniae (Hymenoptera: Tri- lers in Washington. J. Econ. Entomol. 94: 1075Ð1084. chogrammatidae). Environ. Entomol. 31: 482Ð489. Carton, B., G. Smagghe, and L. Tirry. 2003. Toxicity of two Kuhar, T. P., J. Speese, R. J. Cordero, R. Y. Venkata, and V. M. ecdysone agonists, halofenozide and methoxyfenozide, Barlow, and R. Cordero. 2003. Evaluation of foliar in- against the multicoloured Asian lady beetle Harmonia secticides for control of lepidopterous pests in peppers, axyridis (Col., Coccinellidae). J. Appl. Entomol. 127: 240Ð 2002. Arthropod Manag. Tests 28: E37. 242. Kuhar, T. P., V. M. Barlow, M. P. Hoffmann, S. J. Fleischer, Croft, B. A., and A.W.A. Brown. 1975. Response of arthro- E. A. Groden, J. Gardner, R. Hazzard, M. G. Wright, pod natural enemies to insecticides. Annu. Rev. Entomol. S. A. Pitcher, and P. Westgate. 2004. Potential of Tri- 20: 285Ð335. chogramma ostriniae (Hymenoptera: Trichogrammati- Elzen, G. W. 2001. Lethal and sublethal effects of insecti- dae) as a biological control agent of European corn borer cide residues on Orius insidiosus (Hemiptera: Anthoco- (Lepidoptera: Crambidae) in solanaceous crops. J. Econ. ridae) and Geocoris punctipes (Hemiptera: Lygaeidae). J. Entomol. 97: 1209Ð1216. Econ. Entomol. 94: 55Ð59. Kuhar, T. P., A. V. Chapman, and E. M. Hitchner. 2006a. Haseeb, M., T. X. Liu, and W. A. Jones. 2004. Effects of Evaluation of insecticides for lepidopteran pest control selected insecticides on Cotesia plutellae, endoparasitoid on bell peppers, 2005. Arthropod Manag. Tests 31: E43. of Plutella xylostella. Biocontrol 49: 33Ð46. Kuhar, T. P., S. B. Phillips, R. A. Straw, C. M. Waldenmaier, Haseeb, M., H. Amano, and T. X. Liu. 2005. Effects of se- and H. P. Wilson. 2006b. Commercial vegetable produc- lected insecticides on Diadegma semiclausum (Hyme- tion recommendations. Virginia Cooperative Extension noptera: Ichneumonidae) and Oomyzus sokolowskii (Hy- Publication 456-420. menoptera: Eulophidae), parasitoids of Plutella xylostella [NASS] National Agricultural Statistics Service. 2006. Cen- (Lepidoptera: Plutellidae). Insect Sci. 12: 163Ð170. sus of agriculture. Historical data. (http://www.usda.gov/ Hazzard, R. V., G. M. Ghidiu, and T. J. Boucher. 2001. Eu- nass/pubs/histdata.htm). ropean corn borer management. P. 67Ð76 In T. J. Boucher Nault, B. A., and J. Speese. 2000. Evaluation of insecticides and R. A. Ashley [eds.], Northeast pepper integrated pest to control European corn borer and corn earworm on management (IPM) manual. University of Connecticut fall-grown peppers, 1999. Arthropod Manag. Tests 25: Cooperative Extension, Storrs, CT. E52. Hewa-Kapuge, S., S. McDougall, and A. A. Hoffmann. 2003. Schneider, M. A., G. Smagghe, A. Gobbi, and E. Vin˜ uela. Effects of methoxyfenozide, indoxacarb, and other insec- 2003. Toxicity and pharmokinetics of insect growth reg- February 2009 CHAPMAN ET AL.: EUROPEAN CORN BORER IPM IN PEPPERS 295

ulators and other novel insecticides on pupae of Hypo- software for canonical community ordination, 4.5 edition. soter didymator (Hymeoptera: Ichneumonidae), a para- Microcomputer Power, Ithaca, NY. sitoid of early larval instars of lepidopteran pests. J. Econ. van Emden, H. F., V. F. Eastop, R. D. Hughes, and M. J. Way. Entomol. 96: 1054Ð1065. 1969. The ecology of Myzus persicae. Annu. Rev. Ento- Schneider, M. I., G. Smagghe, S. Pineda, and E. Vinuela. mol. 14: 197Ð270. 2004. Action of insect growth regulator insecticides and Viktorov, A. V., E. N. Preshokov, and V. A. Driniaev. 2002. spinosad on life history parameters and absorption in Comparative analysis of degradation of native avermec- third-instar larvae of the endoparasitoid Hyposoter didy- tins and spinosyns by UV-HPLC. Antibiot. Khimioter. 47: mator. Biol. Control 31: 189Ð198. 6Ð10. Smith, S. M. 1996. Biological control with Trichogramma: Villanueva, R. T., and J. F. Walgenbach. 2005. Develop- advances, successes, and potential of their use. Annu. Rev. ment, oviposition, and mortality of Neoseiulus fallacis Entomol. 41: 375Ð406. (Acari: Phytoseiidae) in response to reduced-risk insec- Sorensen, K. A., and D. G. Cooke. 2004. European corn ticides. J. Econ. Entomol. 98: 2114Ð2120. Wang, B., D. N. Ferro, and D. W. Hosmer. 1999. Effective- borer, corn earworm, and beet armyworm control with ness of Trichogramma ostriniae and T. nubilale for con- insecticides, 2002. Arthropod Manag. Tests. 29: E58. trolling the European corn borer Ostrinia nubilalis in Speese, J. 1994. Soil and foliar treatments to control insects sweet corn. Entomol. Exp. Appl. 91: 297Ð303. in peppers, 1993. Arthropod. Manag. Tests 19: 106Ð107. Welty, C. 1995. Monitoring and control of European corn Speese, J. 1995. Foliar sprays to control insects on spring borer, Ostrinia nubilalis (Lepidoptera: Pyralidae), on bell peppers, 1995. Arthropod. Manag. Tests 21: 137. peppers in Ohio. J. Agric. Entomol. 2: 145Ð161. Studebaker, G. E., and T. J. Kring. 2003a. Effects of insec- Welty, C., and S. Vitanza. 2005. Insecticide efÞcacy evalu- ticides on Orius insidiosus (Hemiptera: Anthocoridae) ation of European corn borer control in bell pepper, 2004. measured by Þeld, greenhouse and Petri dish bioassays. Arthropod Manag. Tests 30: E52. Fla. Entomol. 86: 178Ð185. Wright, M. G., T. P. Kuhar, M. P. Hoffmann, and S. A. Studebaker, G. E., and T. J. Kring. 2003b. Effects of various Chenus. 2002. Effect of inoculative releases of Tri- insecticide residues in cotton on gender and develop- chogramma ostriniae on populations of Ostrinia nubilalis mental stage of the insidious ßower bug (Hemiptera: and damage to sweet corn and Þeld corn. Biol. Control 23: Anthocoridae). J. Entomol. Sci. 38: 409Ð419. 149Ð155. ter Braak, C.J.F., and P. Sˇmilauer. 2002. CANOCO refer- ence manual and CanoDraw for Windows userÕs guide: Received 5 February 2008; accepted 24 August 2008.