Crambidae, Noctuidae, and Pyralidae) Pests on Corn Containing Pyramided Bt Traits and a Blended Refuge in the Southern United States

Journal of Economic Entomology Advance Access published June 21, 2016

Journal of Economic Entomology, 2016, 1–13 doi: 10.1093/jee/tow109 Insecticide Resistance and Resistance Management Research article

Impact of Lepidoptera (Crambidae, Noctuidae, and Pyralidae) Pests on Corn Containing Pyramided Bt Traits and a Blended Refuge in the Southern United States

F. P. F. Reay-Jones,1 R. T. Bessin,2 M. J. Brewer,3 D. G. Buntin,4 A. L. Catchot,5 D. R. Cook,6 K. L. Flanders,7 D. L. Kerns,8 R. P. Porter,9 D. D. Reisig,10 S. D. Stewart,11 and M. E. Rice12,13

1Department of Agricultural and Environmental Sciences, Pee Dee Research and Education Center, Clemson University, 2200 Pocket Rd., Florence, SC 29506-9727 ([email protected]), 2Department of Entomology, University of Kentucky, S-225 Ag North, Lexington, KY 40546 ([email protected]), 3Texas A&M AgriLife Research & Extension Center, 10345 State Hwy 44, 4 Corpus Christi, TX 78406 ([email protected]), Department of Entomology, UGA-Griffin Campus, 1109 Experiment Downloaded from Street, Griffin, GA 30223 ([email protected]), 5Department of Entomology & Plant Pathology, Mississippi State University, Mississippi State, MS 39762 ([email protected]), 6Delta Research & Extension Center, Mississippi State University, P.O. Box 197, Stoneville, MS 38776 ([email protected]), 7201 Extension Hall, Auburn University, AL 36849 (fl[email protected]), 8LSU AgCenter, Macon Ridge Station, 212A Macon Ridge Rd., Winnsboro, LA 71295 (dkerns@agcen ter.lsu.edu), 9Texas AgriLife Research & Extension Center, Lubbock, TX 79403 ([email protected]), 10Department of Entomology, North Carolina State University, Vernon G. James Research and Extension Center, Plymouth, NC 27962 (ddrei http://jee.oxfordjournals.org/ [email protected]), 11West Tennessee Research and Education Center, 605 Airways Blvd., Jackson, TN 38301 (sdstewart@ut k.edu), 12DuPont Pioneer, P. O. Box 1150, Johnston, IA 50131 ([email protected]), and 13Corresponding author, e-mail: [email protected]

Received 2 February 2016; Accepted 27 April 2016

Abstract by guest on June 21, 2016 Blended refuge for transgenic plants expressing Bacillus thuringiensis (Bt) toxins has been approved in the northern United States as a resistance management strategy alternative to a structured refuge. A three-year study (2012–2014) was conducted with 54 trials across nine states in the southern United States to evaluate plant injury from lepidopteran pests of corn and yield in a corn hybrid expressing Cry1F Cry1Ab Vip3Aa20 (Pioneer Brand Optimum Leptra) planted as a pure stand and in refuge blends of 5, 10, and 20% in both early and late plantings. Injury by corn earworm, Helicoverpa zea Boddie (Lepidoptera: Noctuidae), and fall armyworm, Spodoptera frugiperda (J. E. Smith) (Lepidoptera: Noctuidae), was generally proportional to the percentage of non-Bt corn within each refuge blend. Across locations, ear injury in plots with 100% Cry1F Cry1Ab Vip3Aa20 (Optimum Leptra) corn ranged from no injury to a maximum of 0.42 cm2 per ear in Mississippi in 2013. Leaf injury ratings in 100% non-Bt plots in early and late planted trials in 2014 were 86- and 70-fold greater than in 100% Cry1F Cry1Ab Vip3Aa20 (Optimum Leptra) plots. Plants in plots with blended refuges had significantly greater leaf injury in 2012 (5, 10, and 20% refuge blends), in the early- planted corn in 2013 (10 and 20% only), and in both early- and late-planted corn in 2014 (20% only) as compared with leaf injury in a pure stand of Cry1F Cry1Ab Vip3Aa20 (Optimum Leptra) seen during these years. Corn ears in plots with blended refuges also had significantly greater area of kernels injured in 2012 (5, 10, and 20%), in early- and late-planted corn in 2013 (5, 10, and 20%), and in early (10 and 20% only)- and late-planted corn (5, 10, and 20%) in 2014 as compared with ear injury in a pure stand of Cry1F Cry1Ab Vip3Aa20 (Optimum Leptra) seen during these years. Infestations of southwestern corn borer, Diatraea grandiosella Dyar (Lepidoptera: Crambidae), were also significantly reduced by Cry1F Cry1Ab Vip3Aa20 (Optimum Leptra). Despite these differences in injury, yield averaged across locations varied among refuge blends only in the late-planted trials in 2013, with greater yields in the 0% refuge blend than in the 20% blend; however, when examining yield separately by location, only two of nine locations had higher yields in the 100% Bt plots than in any of the blended refuge plots. As a complement to studying the contribution of blended refuge to delaying resistance, quantifying injury and yield in a range of

VC The Authors 2016. Published by Oxford University Press on behalf of Entomological Society of America. All rights reserved. For Permissions, please email: [email protected] 1 2 Journal of Economic Entomology, 2016, Vol. 0, No. 0

refuge blends is a necessary step to provide management information on the range of lepidopteran pests that occur in the southern United States.

Key words: Helicoverpa zea, Spodoptera frugiperda, Bacillus thuringiensis, maize, Optimum Leptra

Transgenic corn, Zea mays L., expressing insecticidal Cry proteins the rate of resistance of homozygous individuals is likely to be from Bacillus thuringiensis (Bt) Berliner represented 81% of the initially low and because maintaining pest populations below total corn planted in the United States in 2015 (USDA Economic thresholds is important, the high-dose/refuge strategy has been in- Research Service 2015). The European corn borer, Ostrinia nubilalis strumental in delaying resistance in the United States (Huang et al. (Hu¨bner) (Lepidoptera: Crambidae) was the initial target pests for 2011). Because of issues of grower noncompliance with refuge re- Bt corn in the United States (Koziel et al. 1993). Widespread plant- quirements (Jaffe 2009), a blended refuge, or refuge-in-the-bag, is ing of Bt corn was effective in controlling the European corn borer now used in the Corn Belt (Cullen et al. 2013). However, blended (Rice and Pilcher 2001) and has led to regional reductions in popula- refuges have not been approved in cotton-growing areas of the tions in the Corn Belt (Hutchinson et al. 2010), while being rela- southern United States without a structured block refuge. tively safe to nontarget natural enemies (Pilcher et al. 1997, Naranjo A recent study across 12 states in the southern United States 2009, Carpenter 2011). Although the European corn borer can be summarized the efficacy of a Cry1F Cry1Ab pyramid with a found in the southern United States, it is generally not considered a blended refuge of a non-Bt hybrid (Reisig et al. 2015). Results key pest of corn (Buntin et al. 2004). Bt traits in corn provide con- showed that injury in the non-Bt plants in the blended refuge did not Downloaded from trol or suppression of a number of other lepidopteran pests that are differ significantly from non-Bt plants in plots with a pure stand of more frequent pests of corn in the southern United States, including non-Bt, suggesting that, with the southern corn insect complex, a corn earworm, Helicoverpa zea Boddie (Noctuidae); fall armyworm, blended non-Bt refuge may not diminish resistance management. A Spodoptera frugiperda (J. E. Smith); southwestern corn borer, concern for the blended refuge concept is larval movement among Diatraea grandiosella Dyar (Crambidae); sugarcane borer, Diatraea non-Bt plants in the refuge and Bt plants, which may accelerate evo- http://jee.oxfordjournals.org/ saccharalis (F.) (Crambidae); and lesser cornstalk borer, lution of resistance (Bates et al. 2005, Goldstein et al. 2010, Elasmopalpus lignosellus (Zeller) (Pyralidae) (Buntin et al. 2004, Wangila et al. 2013). Another concern is that cross-pollination may Reay-Jones and Reisig 2014, Reisig et al. 2015). lead to Bt toxins expressed in blended refuge plants, which may im- Corn earworm eggs are laid on the silks, and young larvae feed pact refuge insects (Yang et al. 2014). In addition, only limited data on the silks before moving to the developing ear. Fall armyworm are available on field efficacy in the southern United States, with can cause defoliation at the whorl stage of corn and also feed on de- Reisig et al. (2015) reporting only one refuge blend (5% non-Bt) veloping kernels in the ear, particularly in late planted corn (Buntin et al. 2004). Early-instar southwestern corn borer and sugarcane without recording yield in trials generally planted within optimum borer can also infest whorl stage corn and cause leaf injury; how- planting dates. Because Lepidopteran insect pests of corn can be by guest on June 21, 2016 ever, later-instar injury can cause more serious injury by tunneling more abundant as the season progresses (Buntin et al. 2004), the ob- into corn stalks, which can stunt or even kill young plants (Hensley jectives of this research were: 1) to evaluate plant injury to a and Arbuthnot 1957, Davis et al. 1972). The use of insecticides is Cry1F Cry1Ab Vip3Aa20 (Pioneer Brand Optimum Leptra) hy- particularly challenging in corn because of the difficulty of timing brid with a range of refuge blends and early and late planting dates, applications to control the lepidopteran complex, the lack of com- and 2) to evaluate the impact of injury on yield, which has yet to be plete coverage by aerial application, and the relative concealment of reported in the southern United States with blended refuges. most insects on or within the plant. Corn expressing insecticidal toxins is therefore well suited for managing these Lepidoptera pests. Materials and Methods The first commercialized Bt hybrids produced single-gene toxins with events Bt11 (Cry1Ab, Syngenta Seeds, Golden Valley, MN), Locations and Design of Field Experiments MON810 (Cry1Ab, Monsanto Co., St Louis, MO), and TC1507 A three-year study was conducted across 12 southern states to evalu- (Cry1F, Dow AgroSciences, Indianapolis, IN) (Ostlie et al. 1997). ate four (2012–2013) or five (2014) treatments and two planting More recent pyramided Bt corn hybrids produce multiple toxins tar- dates (Table 1). Treatments were corn hybrids with the following geting Lepidoptera insect pests with Cry1Ab Cry1F, Cry1A.105 transgenic traits: Cry2Ab2, Vip3Aa20 Cry1Ab, Cry1Ab Cry1F Vip3Aa20, and Cry1A.105 Cry2Ab2 Cry1F (Reisig et al. 2015). Expression of 1. Cry1F Cry1Ab Vip3Aa20 pyramid of TC1507 multiple toxins with different modes of action may provide enhanced MON810 MIR162 NK603 (event TC1507, Cry1F protein, resistance management compared to single-toxin transgenic crops Dow Agro-Sciences, Indianapolis, IN; event MON810, Cry1Ab (Bates et al. 2005). Pyramiding toxins can also provide improved effi- protein, Monsanto Company, St. Louis, MO; event MIR162, cacy against corn earworm and fall armyworm (Burkness et al. 2010, Vip3Aa20 protein, Syngenta, Research Triangle Park, NC; event Reay-Jones and Reisig 2014, Reisig et al. 2015). NK603, mEPSPS protein conferring glyphosate herbicide toler- Insecticide resistance management (IRM) strategies for Bt crops ance, Monsanto Co.); in the United States and Canada aim to delay resistance by using 2. TC1507 MON810 MIR162 NK603 pyramid (same as “high-dose” expression of Bt toxins that kill >95% of heterozygotes Treatment 1) plus a 5% blended refuge non-Bt hybrid with only for single allele-type Bt resistance (Gould 1998). Corn growers in NK603; the United States are also required to plant a refuge of non-Bt corn 3. TC1507 MON810 MIR162 NK603 pyramid (same as to provide a refuge of susceptible genotypes that presumably would Treatment 1) plus a 10% blended refuge non-Bt hybrid with mate with any rare, resistant individuals (Bates et al. 2005). Because only NK603; Journal of Economic Entomology, 2016, Vol. 0, No. 0 3

Table 1. Location, planting date, and data collected for trials evaluating efﬁcacy of Bt corn deployed against lepidopteran pests in locations across nine states in the southern United States

Location (county, state) Year Planting dates Data collected

Tallassee, AL 2012 25 April Leaf injury, kernel area consumed Williamson, GA 2012 3 April Kernel area consumed Lexington, KY 2012 23 April Leaf injury, kernel area consumed Winnsboro, LA 2012 20 April Kernel area consumed Starkville, MS (MS1) 2012 5 June Kernel area consumed Stoneville, MS (MS2) 2012 27 April Kernel area consumed Plymouth, NC 2012 19 April Kernel area consumed Florence, SC 2012 24 April Kernel area consumed Jackson, TN 2012 18 May Kernel area consumed, percent infested with southwestern corn borer Corpus Christi, TX 2012 28 March Leaf injury, kernel area consumed Lubbock, TX 2012 24 May Leaf injury, kernel area consumed Tallassee, AL 2013 3 April Kernel area consumed Tallassee, AL 2013 24 April Kernel area consumed Plains, GA 2013 10 April Kernel area consumed Plains, GA 2013 30 May Leaf injury, kernel area consumed Lexington, KY 2013 3 May Leaf injury, kernel area consumed Lexington, KY 2013 28 May Leaf injury, kernel area consumed Downloaded from Winnsboro, LA 2013 17 April Leaf injury,a kernel area consumed, tunnel length (southwestern corn borer and sugarcane borer)b Starkville, MS (MS1) 2013 18 April Kernel area consumed Starkville, MS (MS1) 2013 29 May Kernel area consumed Stoneville, MS (MS2) 2013 30 April Kernel area consumed Stoneville, MS (MS2) 2013 29 May Kernel area consumed http://jee.oxfordjournals.org/ Plymouth, NC 2013 24 April Kernel area consumed Plymouth, NC 2013 14 May Kernel area consumed Florence, SC 2013 8 April Kernel area consumed Florence, SC 2013 8 May Kernel area consumed Jackson, TN 2013 23 April Kernel area consumed, tunnel length, and infested plants (southwestern corn borer) Jackson, TN 2013 4 May Kernel area consumed, tunnel length, and infested plants (southwestern corn borer) Corpus Christi, TX (TX1) 2013 15 March Leaf injury, kernel area consumed Corpus Christi, TX (TX1) 2013 7 April Leaf injury, kernel area consumed Lubbock, TX (TX2) 2013 17 April Leaf injury, kernel area consumed Lubbock, TX (TX2) 2013 21 May Leaf injury, kernel area consumed by guest on June 21, 2016 Tallassee, AL 2014 11 April Kernel area consumed Tallassee, AL 2014 8 May Kernel area consumed Plains, GA 2014 25 April Leaf injury, kernel area consumed Plains, GA 2014 20 May Leaf injury, kernel area consumed Lexington, KY 2014 2 May Leaf injury, kernel area consumed Lexington, KY 2014 25 May Leaf injury, kernel area consumed Winnsboro, LA 2014 20 March Leaf injury, kernel area consumed Winnsboro, LA 2014 17 April Leaf injury, kernel area consumed Starkville, MS (MS1) 2014 17 April Kernel area consumed Starkville, MS (MS1) 2014 12 May Kernel area consumed Stoneville, MS (MS2) 2014 2 April Leaf injury, kernel area consumed Stoneville, MS (MS2) 2014 1 May Leaf injury, kernel area consumed Plymouth, NC 2014 6 May Kernel area consumed Plymouth, NC 2014 30 May Leaf injury, kernel area consumed Florence, SC 2014 9 April Kernel area consumed Florence, SC 2014 1 May Leaf injury, kernel area consumed Jackson, TN 2014 11 April Kernel area consumed, tunnel length, and infested plants (southwestern corn borer) Jackson, TN 2014 6 May Kernel area consumed, tunnel length, and infested plants (southwestern corn borer) Corpus Christi, TX (TX1) 2014 19 March Leaf injury, kernel area consumed Corpus Christi, TX (TX1) 2014 11 April Leaf injury, kernel area consumed Lubbock, TX (TX2) 2014 17 April Leaf injury, kernel area consumed Lubbock, TX (TX2) 2014 2 May Leaf injury, kernel area consumed

a Leaf injury resulted from artiﬁcial infestation of fall armyworm on 11 June 2013. All other data regarding insect injury resulted from natural infestations. b Stalk tunnels resulted from artiﬁcial infestation of egg masses of southwestern corn borer per plant on 19 June. Sugarcane borers were also present (50:50 borer species). 4 Journal of Economic Entomology, 2016, Vol. 0, No. 0

4. TC1507 MON810 MIR162 NK603 pyramid (same as planting date as the main plot and refuge blend treatment as the Treatment 1) plus a 20% blended refuge non-Bt hybrid with sub-plot treatment. only NK603; Leaf injury (log x þ 1), proportions (arcsine of the square root), 5. Non-Bt hybrid with only NK603 (2014 trials only). and kernel injury (log x þ 1) were transformed prior to ANOVA to normalize their distribution (Zar 1999). The Kenward–Roger Pioneer Brand Optimum Leptra hybrids from the P1615 hybrid method (Kenward and Roger 1997) was used to compute denomin- platform with a relative maturity of 116 d were used in 2012–2013 ator degrees of freedom for the test of fixed effects for all variables. and hybrids from the P1319 hybrid platform with a relative matur- Significant interactions were analyzed using the SLICE option of the ity of 113 d were used in 2014. Refuge seed were blended and incor- LSMEANS statement. Means were separated using Tukey’s HSD porated with the Bt hybrid prior to planting. All seeds were (Tukey 1953). Data from all trials were included in ear injury ana- provided by DuPont Pioneer with 0.25 mg a.i./kernel of thiame- lyses, as injury occurred in all trials. For leaf injury, only data from thoxam (Cruiser 250, Syngenta Crop Protection, Greensboro, NC) trials with more than one plot with injury were included in analyses. for secondary pest management. The early planting dates were generally within the range of dates recommended by State Cooperative Extension guidelines. The late planting dates were generally three to Results four weeks later, and were outside of the recommended planting Leaf Injury range (Table 1). Leaf injury caused either by corn earworm or fall armyworm Randomized block designs were used in each trial with four rep- occurred in four out of 11 trials in 2012, four out of 11 early- lications. Plots were 12.2 m in length and four rows wide with 76 to planted trials in 2013, three out of 10 late-planted trials in 2013,

102 cm row width. Planting densities and thinning of seedlings were Downloaded from seven out of 11 early-planted trials in 2014, and eight out of 11 late- used to achieve a target population of 80,000 seed per ha. When planted trials in 2014. Averaged across locations, leaf injury gener- two planting dates were used, two separate trials were planted adja- ally increased with the percentage of non-Bt plants (Table 2; Fig. 1). cent to one another in the same field. Differences were not significant between the 5 and 10% non-Bt ref- Natural infestations of one or more target pests occurred in each uge blends. The 20% refuge blend had significantly more injury trial, except for fall armyworm and southwestern corn borer in the than the 0% (no blend) blend in all trials, except in the late planting http://jee.oxfordjournals.org/ Louisiana trial in 2013; artificial infestations of fall armyworm were date in 2013. made on 6 June to every other plant on rows two and three of each Location interacted significantly with refuge blend for leaf injury plot at a rate of 10 neonates per whorl in corn cob grits (Wiseman in all trials except the late planted trials during 2013 (Table 2). Out et al. 1980). Also, artificial infestations of southwestern corn borer of 26 trials with leaf injury, significant differences among ref- were made at this location to every other plant in rows two and uge blends occurred in only 12 trials based on the overall F-test three of each plot on 19 June by stapling egg masses onto leaves. (Table 3), which dropped to only 10 trials when using the conservative Tukey mean separation test. Among these trials with significant Data Collection differences, three trials (Lexington, KY 2012, Lubbock, TX 2012, by guest on June 21, 2016 Leaf and ear injury ratings were recorded in all plants of rows two and Corpus Christi, TX in the early-planted 2013 trial) had signifi- and three of each plot. Leaf injury was recorded in whorl-stage cantly more leaf injury in the 20% refuge blend than the 0% or 5% plants when 99% of larvae had finished feeding in the non-Bt refuge blends. In 2014, the 100% non-Bt plots had significantly plots. Leaf feeding injury was evaluated on a reverse 9–1 scale ac- more leaf injury than all refuge blends in all seven trials with signifi- cording to the Pioneer research proposal and then converted to the cant differences; in only one of these seven trials (Lubbock, TX) did Davis et al. (1992) 0–9 scale (0 ¼ lowest injury and 9 ¼ highest in- the 20% refuge blend have significantly more injury than the 0% jury). Ear injury was recorded when 99% of larvae had finished refuge. feeding based on sampling of border rows (rows one and four). When analyzing data with planting date and refuge blend as Variables recorded were proportion of ears injured, the area of ker- fixed effects, averaged across locations in 2013, neither refuge blend 2 nels injured (cm ), and the number (and species) of larvae per ear. (F ¼ 3.33; df ¼ 3, 12; P ¼ 0.0564) nor planting date (F ¼ 1.15; Injury from stalk-boring species was assessed by either visual exam- df ¼ 1, 1.59; P ¼ 0.4187) affected leaf injury, and the two factors ination (exit holes) or by splitting stalks and measuring tunnel length did not interact significantly (F ¼ 0.85; df ¼ 3, 12; P ¼ 0.4947). of all plants in rows two and three of each plot (Table 1). Averaged across locations in 2014, leaf injury varied with refuge blend (F ¼ 15.72; df ¼ 4, 58.2; P < 0.0001), but not with planting Data Analyses date (F ¼ 3.36; df ¼ 1, 63.97; P ¼ 0.0714) and the two factors did Leaf injury, proportion of ears injured, kernel area consumed by not interact significantly (F ¼ 1.56; df ¼ 4, 58.2; P ¼ 0.1966). corn earworm, length of stalk tunnels by southwestern corn borer or Though not significant, there was a trend for increased leaf injury in sugarcane borer, and yield data were analyzed using a two-way the late planting date (0.244 6 0.083) than in the early planting date ANOVA with refuge blend and location as fixed effects and repli- (0.113 6 0.045). cate (location) as a random effect for each year and planting date (Littell et al. 2006; PROC MIXED, SAS Institute 2011). Because Ear Injury planting dates were not randomized at each location, planting date All trials had ear feeding by either corn earworm or fall armyworm was not included in ANOVA models as a fixed effect with refuge or both species. Based on larval sampling in ears of non-Bt plots, blend and location. A second ANOVA with the same variables was corn earworm was the only pest found in ears in 29 of the 33 trials. therefore conducted with planting date and refuge blend as fixed ef- A combination of corn earworm and fall armyworm was found in fects, and location and location planting date as random effects Lubbock, TX, in 2012 (proportion of each species unknown), 2013 for each year; all variables were averaged across the four replicates (68% fall armyworm, 32% corn earworm), 2014 (53% fall army- prior to ANOVA to simplify the design to a split-plot design with worm, 47% corn earworm), and in Plains, GA, in 2014 (10% fall Journal of Economic Entomology, 2016, Vol. 0, No. 0 5

Table 2. ANOVA statistics for leaf injury and H. zea injury and corn yield collected for trials evaluating efﬁcacy of Bt corn deployed against lepidopteran pests in locations across nine states in the southern United States

Year Planting Effect Leaf injury Proportion of injured ears Area of kernels injured (cm2) date df FP df FPdf FP

2012 – Refuge 3, 36 26.14 <0.0001 3, 97.9 95.81 <0.0001 3, 97.1 53.36 <0.0001 Location 3, 12 62.41 <0.0001 10, 33.2 34.46 <0.0001 10, 32.4 10.05 <0.0001 Refuge location 9, 36 8.22 <0.0001 30, 97.8 5.36 <0.0001 30, 97 3.39 <0.0001 2013 Early Refuge 3, 36 17.19 <0.0001 3, 99 53.79 <0.0001 3, 132 22.29 <0.0001 Location 3, 12 138.01 <0.0001 10, 33 28.68 <0.0001 10, 132 25.42 <0.0001 Refuge location 9, 36 5.93 <0.0001 30, 99 4.03 <0.0001 30, 132 2.83 <0.0001 Late Refuge 3, 36 1.11 0.3588 3, 89.7 146.94 <0.0001 3, 119 61.03 <0.0001 Location 2, 36 34.53 <0.0001 9, 30.2 36.33 <0.0001 9, 119 40.21 <0.0001 Refuge location 6, 36 0.67 0.6759 27, 89.6 4.27 <0.0001 27, 119 3.27 <0.0001 2014 Early Refuge 4, 83.4 120.71 <0.0001 4, 131 515.70 <0.0001 4, 131 352.07 <0.0001 Location 7, 21.2 23.12 <0.0001 10, 33.2 17.48 <0.0001 10, 33.2 9.39 <0.0001 Refuge location 24, 83.3 20.47 <0.0001 40, 131 7.17 <0.0001 40, 131 6.58 <0.0001 Late Refuge 4, 96 124.94 <0.0001 4, 132 1,014, 66 <0.0001 4, 132 720.42 <0.0001 Location 7, 14 20.80 <0.0001 10, 33 65.18 <0.0001 10, 33 41.31 <0.0001 Refuge location 28, 96 12.63 <0.0001 40, 132 17.35 <0.0001 40, 132 20.10 <0.0001 Downloaded from

2012 2013 2014

1.2 1.2 1.2 a Early planting date http://jee.oxfordjournals.org/ Late planting date Early planting date 1.0 1.0 1.0 Late planting date a 0.8 0.8 0.8

0.6 0.6 0.6 a b bc a Leaf injury Leaf Leaf injury 0.4 injury Leaf 0.4 0.4 b c a a b a b 0.2 c 0.2 a 0.2 c b by guest on June 21, 2016 c c c c bc 0.0 0.0 0.0 051020 0 5 10 20 051020100 Refuge scenario Refuge scenario Refuge scenario (% non-Bt) (% non-Bt) (% non-Bt)

Fig. 1. Impact of refuge blend (0, 5, 10, 20, and 100% non-Bt) and planting date on leaf injury from corn earworm or fall armyworm feeding in trials evaluating efﬁ- cacy of Bt corn deployed against lepidopteran pests at locations across nine states in the southern United States. Bars for each planting date with the same letter are not signiﬁcantly different (Tukey’s HSD [1953]). armyworm, 90% corn earworm). Averaged across locations, the trend (e.g., Tallassee, AL, in 2012, Stoneville, MS, in late-planted percentage of ears injured increased with refuge blend in each year trial in 2013). With the addition of a 100% non-Bt treatment in (Table 2; Fig. 2). Mean separation tests showed significant differ- 2014, the variability in the levels of injury was further emphasized ences between each refuge blend except between 5 and 10% in the among locations, ranging from 6.37 6 0.14% in Lexington, KY, in early-planted corn in 2014 and between 0 and 5% in the late- the early-planting date in 2014 to 100% in Starkville, MS, in the planted corn in 2014. late-planting date in the non-Bt treatment. Among trials with no sig- The percentage of injured ears changed with refuge blend, but nificant variation in the percentage of ears injured among refuge this change was not consistent across locations (Table 2). Out of 54 blends, injury was often low (<1% injured ears averaged across ref- trials, the percentage of ears injured varied among refuge blends in uge blends in nine of the 15 trials with no significant differences) or 47 trials (Table 3), though only 39 trials had significant differences injury was relatively high across the 0–20% refuge blends as was the among means using the conservative Tukey’s test (Fig. 3). Among case in 2012 (Starkville, MS: average of 12.89 6 1.89%; Stoneville, such trials, the 0% blend had significantly fewer ears injured than MS: 12.96 6 1.53%) and 2013 (Starkville, MS: 6.16 6 0.86% in the 5% blend only in Winnsboro, LA, Corpus Christi, TX, and early-planting date and 18.44 6 1.04 in late-planting date; Lubbock, TX, in 2012, and in the late planting in Plymouth, NC, Stoneville, MS: 8.71 6 1.61% in early-planting date; Jackson, TN: and Corpus Christi, TX, in 2013. The interaction between location 2.29 6 0.59). When planting date was included as a fixed effect and and refuge blend resulted from some trials having an increase in in- averaging across locations in 2013, the percentage of ears injured jury that was proportional to the percentage of non-Bt plants (e.g., varied with refuge blend (F ¼ 39.74; df ¼ 3, 54; P < 0.0001), but not Corpus Christi, TX, in 2012, Williamson, GA, and Florence, SC, in with planting date (F ¼ 0.46; df ¼ 1, 9; P ¼ 0.5166), and the two fac- late-planted trial in 2013) and other trials having a less apparent tors did not interact significantly (F ¼ 2.71; df ¼ 3, 54; P ¼ 0.0541). 6

Table 3. F-statistics from the SLICE option of the LSMEANS statement in PROC MIXED indicating signiﬁcance of the refuge blend treatment at each location

Location 2012 2013 2014

Early planting Late planting Early planting Late planting

Proportion of Area of kernels Proportion of Area of kernels Proportion of Area of kernels Proportion of Area of kernels Proportion of Area of kernels injured ears injured (cm2) injured ears injured (cm2) injured ears injured (cm2) injured ears injured (cm2) injured ears injured (cm2)

Tallassee, AL 9.52b 3.70e 0.48NS 1.04NS 3.39e 0.45NS 49.32b 50.88b 55.75b 50.56b Plains, GAa 1.01NS 0.02NS 0.47NS 0.06NS 39.65b 22.65b 33.34b 23.99b 182.73b 205.67b Lexington, KY 0.27NS 0.03NS 1.53NS 0.10NS 6.55c 0.83NS 4.10c 0.60NS 13.94b 7.31b Winnsboro, LA 27.65b 9.08b 11.73b 1.40NS – – 23.61b 18.24b 24.92b 7.42b Starkville, MS 2.17NS 3.21e 3.42e 10.03b 5.32d 14.34b 61.43b 36.80b 209.45b 95.46b Stoneville, MS 3.36e 2.64NS 3.62e 2.74e 9.91b 5.36d 77.34b 68.12b 147.75b 144.62b Plymouth, NC 18.11b 22.97b 1.48NS 0.16NS 14.99b 6.11d 24.96b 21.96b 79.12b 70.98b Florence, SC 30.14b 21.84b 14.62b 7.54c 28.40b 9.69b 58.38b 21.87b 161.88b 90.75b Jackson, TN 13.54b 8.68b 11.30b 2.91e 6.29c 0.45NS 33.55b 16.85b 18.40b 3.26e Corpus Christi, TX 21.52b 12.33b 19.57b 21.61b 44.52b 23.95b 97.86b 67.74b 106.52b 45.22b Lubbock, TX 20.27b 2.15NS 25.84b 3.00e 29.03b 6.34c 117.17b 84.87b 187.67b 199.85b df 3, 97.5 3, 96.6 3, 99 3, 132 3, 89.4 3, 119 4, 131 4, 131 4, 132 4, 132

Leaf injury Yield Leaf injury Yield Leaf injury Yield Leaf injury Yield Leaf injury Yield Tallassee, AL 0.32NS 0.10NS – 0.11NS – 10.03b – 0.58NS – 0.47NS Plains, GA – 0.00NS – 0.08NS 0.01NS 0.48NS 4.81d – 35.93b – Lexington, KY 24.28b 1.04NS 0.80NS 0.32NS 0.11NS 0.48NS 16.59b 0.21NS 0.66NS 1.75NS

Winnsboro, LA – 1.66NS 1.18NS 0.48NS – – 3.39e 1.55NS 0.23NS 1.12NS Entomology Economic of Journal Starkville, MS – 0.95NS – 5.79d – 0.36NS – 0.06NS – 0.78NS Stoneville, MS – 0.68NS – 1.02NS – 1.45NS 0.06NS 4.18d 0.80NS 5.21c Plymouth, NC – 4.77d – 1.89NS – 0.62NS – 0.33NS 42.77b 0.33NS Florence, SC – 1.09NS – 12.12b – 10.38b – 1.61NS 1.84NS 4.53c Jackson, TN – 0.05NS – 0.34NS – 0.06NS 0.04NS 1.26NS – 1.39NS Corpus Christi, TX 0.06NS 0.50NS 32.56b 0.64NS 2.33NS 0.11NS 134.22b 0.31NS 48.38b – Lubbock, TX 26.15b 0.98NS 0.43NS 0.34NS – 0.18NS 93.21b 4.40c 82.75b 8.77b df 3, 36 3, 92.9 3, 36 3, 99 3, 36 3, 88.3 4, 83.2 4, 119 4, 96 4, 108

NS, not signiﬁcant. a For Georgia, all trials were conducted in Plains except in 2012 (Williamson). b P < 0.0001. 06 o.0 o 0 No. 0, Vol. 2016, , c P < 0.001. d P < 0.01.

e P < 0.05.

Downloaded from from Downloaded http://jee.oxfordjournals.org/ by guest on June 21, 2016 21, June on guest by Journal of Economic Entomology, 2016, Vol. 0, No. 0 7

2012 2012

16 ) 0.5 a 2 14 0.4 a 12 b 10 0.3 b 8 c 6 0.2 c d

Ears injured (%) Ears 4 0.1 d 2

0 Area of injured kernels (cm 0.0 0 5 10 20 0 5 10 20 Refuge scenario Refuge scenario (% non-Bt) (% non-Bt)

2013 2013

16 ) 0.6 2

Early planting date Early planting date Downloaded from 14 a Late planting date 0.5 Late planting date 12 a 0.4 10 a a 8 0.3 b

b b http://jee.oxfordjournals.org/ 6 b c 0.2 b c b Ears injured (%) Ears 4 d 0.1 c c 2 d

0 Area of injured kernels (cm 0.0 0 5 10 20 0 5 10 20 Refuge scenario Refuge scenario (% non-Bt) (% non-Bt) by guest on June 21, 2016 2014 2014

100 ) 5 2 Early planting date Early planting date Late planting date 80 Late planting date a 4 a a 60 3 a 40 2

Ears injured (%) 20 bb 1 cc bb d d cd d cdde c c

0 Area of injured kernels (cm 0 0 5 10 20 100 0 5 10 20 100 Refuge scenario Refuge scenario (% non-Bt) (% non-Bt)

Fig. 2. Impact of refuge blend (0, 5, 10, 20, and 100% non-Bt) and planting date on corn ear injury in trials evaluating efﬁcacy of Bt corn deployed against lepidopteran pests at locations across nine states in the southern United States. Bars for each planting date with the same letter are not signiﬁcantly different (Tukey’s HSD [1953]).

Averaged across locations in 2014, the percentage of ears injured tests showed significant differences between each refuge blend ex- varied with refuge blend (F ¼ 126.76; df ¼ 4, 80; P < 0.0001), but cept between 5 and 10% in the early- and late-planted corn in 2013 not with planting date (F ¼ 0.42; df ¼ 1, 10; P ¼ 0.5333), and the and early-planted corn in 2014. The area of injured kernels varied two factors did not interact significantly (F ¼ 0.54; df ¼ 4, 80; with refuge blend, but this change was not consistent across loca- P ¼ 0.7052). tions (Table 2). The area of injured kernels varied significantly Averaged across locations, the area of injured kernels increased among refuge blends in 41 of 54 trials (Table 3), though only 35 tri- with refuge blend in each year (Table 2; Fig. 2). Mean separation als had significant differences among means using the conservative 8 Journal of Economic Entomology, 2016, Vol. 0, No. 0 Downloaded from http://jee.oxfordjournals.org/ by guest on June 21, 2016

Fig. 3. Impact of refuge blend (0, 5, 10, 20, and 100% non-Bt) and planting date on percent corn ear injury in trials evaluating efﬁcacy of Bt corn deployed against lepidopteran pests at locations across nine states in the southern United States. Bars for each planting date with the same letter are not signiﬁcantly different (Tukey’s HSD [1953]) using the SLICE option of the LSMEANS statement in PROC MIXED to interpret the refuge blend location interaction.

Tukey’s test (Fig. 4). Among such trials, the 0% blend had signifi- and refuge blend resulted from some trials having an increase in in- cantly less kernel injury than the 5% blend only in Starkville, MS, in jury proportional to the percentage of non-Bt plants (e.g., Plymouth, the early- and late-planted trials during 2013. The greatest levels of NC, in 2012, Corpus Christi, TX, in early- and late-planted trials in injured kernels in both the 0% blend (0.42 cm2) and in the 20% 2013) and other trials having a less apparent trend (e.g., Starkville, blend (1.27 cm2) occurred in the same trial (late-planted trial in MS, in early- and late-planted trials in 2013). With the addition of a Mississippi in 2013). The significant interaction between location 100% non-Bt treatment in 2014, kernel injury in non-Bt treatments Journal of Economic Entomology, 2016, Vol. 0, No. 0 9 Downloaded from http://jee.oxfordjournals.org/ by guest on June 21, 2016

Fig. 4. Impact of refuge blend (0, 5, 10, 20, and 100% non-Bt) and planting date on area of kernels injured in trials evaluating efﬁcacy of Bt corn deployed against lepidopteran pests at locations across nine states in the southern United States. Bars for each planting date with the same letter are not signiﬁcantly different (Tukey’s HSD [1953]) using the SLICE option of the LSMEANS statement in PROC MIXED to interpret the refuge blend location interaction. ranged from 0.14 6 0.03 cm2 in Lexington, KY, in the early-planted 0.26 6 0.07 cm2; Stoneville, MS: 0.30 6 0.04 cm2) and 2013 trial to 8.01 6 0.96 cm2 in Plains, GA, in the late-planting date. (Starkville, MS: 0.50 6 0.09 cm2 in early-planting date and Among trials with no variation in the percentage of ears injured 0.81 6 0.14 cm2 in late-planting date). among refuge blends, kernel injury was often low (<0.1 cm2 injured When planting date was included as a fixed effect and averaging averaged across refuge blends in 16 of the 19 trials with no signifi- across locations in 2013, kernel injury varied with refuge blend cant differences) or injury was relatively high across the 0–20% ref- (F ¼ 24.10; df ¼ 3, 54; P < 0.0001), but not with planting date uge blends, as was the case in 2012 (Starkville, MS: average of (F 2.53; df ¼ 1, 9; P ¼ 0.1463), and the two factors did not interact 10 Journal of Economic Entomology, 2016, Vol. 0, No. 0 significantly (F ¼ 2.08; df ¼ 3, 54; P ¼ 0.1131). Averaged across lo- P ¼ 0.0032). Averaged across locations, the only significant differ- cations in 2014, kernel injury varied with refuge blend (F ¼ 87.71; ence was between in the 0% refuge blend and the 20% refuge blend df ¼ 4, 80; P < 0.0001), but not with planting date (F ¼ 0.54; df ¼ 1, (Fig. 5). Among the two trials with a significant refuge effect in the 10; P ¼ 0.4790), and the two factors did not interact significantly second planting date in 2013 (Table 3), yield in Tallassee, AL, was (F ¼ 0.52; df ¼ 4, 80; P ¼ 0.7228). significantly greater in the 0% refuge blend (9,004.3 6 869.5 kg/ha) than in the 10% refuge blend (6,610.4 6 1,368.2 kg/ha); yields in Southwestern Corn Borer Florence, SC, were significantly greater in the 0% refuge blend Natural infestations of southwestern corn borer occurred only in (10,210.1 6 327.3 kg/ha) than in the 20% refuge blend Tennessee. Infested stalks varied among refuge blends in 2012 (6,768.7 6 462.5 kg/ha). (F ¼ 26.62; df ¼ 3, 9; P < 0.0001), with averages (%) of 0 6 0a, In the early-planted trials in 2014, yield varied among locations 0.62 6 0.36 b, 2.50 6 0.51 c, and 4.69 6 0.60 c in the 0, 5, 10, and (F ¼ 179.48; df ¼ 9, 30.1; P < 0.0001) and the interaction (F ¼ 1.58; 20% refuge blends, respectively. Percentages of damaged stalks df ¼ 36, 119; P ¼ 0.0351), but not among refuge blends (F ¼ 0.24; were low in 2013 and 2014 and did not vary among refuge blends in df ¼ 4, 119; P ¼ 0.9148). Out of 10 early-planted trials in 2014 the early (average of 0.31 6 0.31%; F ¼ 1.00; df ¼ 3, 9; P ¼ 0.4363)- (Table 3), yield varied among refuge blends only in Stoneville, MS and late (0.62 6 0.43%; F ¼ 0.67; df ¼ 3, 9; P ¼ 0.5934)-planted tri- (F ¼ 4.18; df ¼ 4, 120; P ¼ 0.0033), and Lubbock, TX (F ¼ 4.40; 6 als in 2013, and in the early (1.00 1.00%; F ¼ 1.01; df ¼ 4, 12; df ¼ 4, 119; P ¼ 0.0023); however, differences among yields were P 0.4419)- and late (2.00 6 1.56%; F 2.60; df 4, 12; ¼ ¼ ¼ not significant using the more conservative Tukey’s test. In the late- P ¼ 0.0897)-planted trials in 2014. Percentages of damaged stalks planted trials in 2014, yield varied among locations (F ¼ 137.64; also did not vary among refuge blends in the early (0.31 6 0.31%; < df ¼ 8, 27; P 0.0001) and the interaction (F ¼ 2.90; df ¼ 32, 108; Downloaded from F ¼ 1.00; df ¼ 3, 9; P ¼ 0.4363)- and late (2.50 6 1.77%; F ¼ 0.67; P < 0.0001), but not among refuge blends (F ¼ 1.18; df ¼ 4, 108; df ¼ 3, 9; P ¼ 0.5867)-planted trials in 2013, and in the early P ¼ 0.3246). Out of 10 late-planted trials in 2014 (Table 3), yield (1.50 6 1.50%; F ¼ 1.01; df ¼ 4, 12; P ¼ 0.4419)- and late varied among refuge blends only in Stoneville, MS (F ¼ 5.21; df ¼ 4, (2.00 6 2.04%; F ¼ 2.40; df ¼ 4, 12; P ¼ 0.1079)-planted trials in 108; P ¼ 0.0007), and Lubbock, TX (F ¼ 8.77; df ¼ 4, 108; 2014. In the Louisiana trial in 2013 with artificial infestations of P < 0.0001); in Stoneville, MS, yield in the 0% refuge blend

southwestern corn borer, the 20% refuge blend (2.15 6 0.65 a) had http://jee.oxfordjournals.org/ significantly greater damaged stalks (%) than the 0 (0 6 0 b), 5 (8,803.0 6 587.9) was significantly less than yields in the other ref- (0.34 6 0.20 b), and 10% (0.18 6 0.18 b) refuge blends (F ¼ 12.93; uge blends (range of 10,425.8 to 11,585.0 kg/ha). In Lubbock, TX, df ¼ 3, 9; P ¼ 0.0013). yields in the non-Bt plots (11,316.5 6 420.8 kg/ha) were significantly less than in plots in the 0% (14,043.6 6 576.5 kg/ha), 5% Yield (14,612.4 6 351.8 kg/ha), and 20% refuge blend (13,841.8 6 In 2012, yield varied among locations (F ¼ 39.72; df ¼ 10, 31.7; 187.7 kg/ha). P < 0.0001), but not among refuge blends (F ¼ 1.25; df ¼ 3, 93.3; Averaged across locations in 2013, yield varied with the inter- P ¼ 0.2968), and the two factors did not interact significantly action between refuge blend and planting date (F ¼ 2.89; df ¼ 3, 54; by guest on June 21, 2016 (F ¼ 1.06; df ¼ 30, 93.2; P ¼ 0.4067; Fig. 5). In the early-planted tri- P ¼ 0.0435), but not with planting date (F ¼ 0.00; df ¼ 1, 9; als in 2013, yield varied among locations (F ¼ 44.14; df ¼ 10, 33; P ¼ 0.9791) or refuge blend (F ¼ 0.51; df ¼ 3, 54; P ¼ 0.6762). P < 0.0001), but not among refuge blends (F ¼ 0.78; df ¼ 3, 99; When using the more conservative Tukey’s test, yields did not vary P ¼ 0.5061), and the two factors did not interact significantly among planting dates and refuge blends in 2013. Averaged across (F ¼ 1.09; df ¼ 30, 99; P ¼ 0.3702; Fig. 5). In the late-planted trials locations in 2014, yield did not vary with refuge blend (F ¼ 0.50; in 2013, yield varied among refuge blends (F ¼ 4.19; df ¼ 3, 88.7; df ¼ 4, 64; P ¼ 0.7379), planting date (F ¼ 3.12; df ¼ 1, 8; P ¼ 0.0080), locations (F ¼ 85.93; df ¼ 9, 30.3; P < 0.0001), and the P ¼ 0.1155), and the two factors did not interact significantly two factors interacted significantly (F ¼ 2.19; df ¼ 27, 88.7; (F ¼ 0.16; df ¼ 4, 64; P ¼ 0.9562).

2012 2013 2014

8000 14000 16000 a a a a Early planting date Early planting date 12000 Late planting date 14000 Late planting date a a a a a a a a a a a 6000 a a ab a ab a b 12000 10000 10000 8000 4000 8000 6000 6000

Yield (kg/ha)Yield a 4000 2000 4000

2000 2000 c

0 0 0 051020 0 5 10 20 0 5 10 20 100 Refuge scenario Refuge scenario Refuge scenario (% non-Bt) (% non-Bt) (% non-Bt)

Fig. 5. Impact of refuge blend (0, 5, 10, 20, and 100% non-Bt) and planting date on yield at locations across nine states in the southern United States. Bars for each planting date with the same letter are not signiﬁcantly different (Tukey’s HSD [1953]). Journal of Economic Entomology, 2016, Vol. 0, No. 0 11

Discussion has not been formerly determined, the authors suggested that Cry1Ab Vip3Aa20 corn is likely the first to express a high dose Injury by corn earworm and fall armyworm was generally propor- for this species. This is particularly important, as resistance to tional to the percentage of non-Bt corn within each refuge blend. Cry1F in fall armyworm has been documented in Puerto Rico Previous work has shown efficacy of Cry1Ab Vip3Aa20 corn in (Storer et al. 2010), North Carolina, and Florida (Huang et al. reducing corn earworm injury, with an average of 1.03 cm2 injured 2014). In a greenhouse trial, larvae from a fall armyworm strain re- kernels per ear across two years and 17 locations in the southern sistant to Cry1F were unable to develop on pyramided corn express- United States compared to 10.8 and 11.3 cm2 in two non-Bt hybrids ing Cry1A.105 Cry2Ab2, Cry1A.105 Cry2Ab2 Cry1F, and (Reisig et al. 2015). In our study, ear injury in 100% Cry1Ab Vip3Aa20 (Huang et al. 2014), underlining the value of Cry1F Cry1Ab Vip3Aa20 corn averaged across locations ranged pyramided corn hybrids in managing fall armyworm. from <0.01 cm2 per ear in the second planting date in 2014 to 0.05 Plots with blended refuges had significantly greater leaf injury in cm2 per ear in 2012. While Reisig et al. (2015) did not report variabil- 2012 (5, 10, and 20%), in the early-planted corn in 2013 (10 and ity among locations, injury in our study in plots with 100% 20% only), and in both early- and late-planted corn in 2014 (20% Cry1F Cry1Ab Vip3Aa20 (Optimum Leptra) corn ranged across only) compared with a pure stand of Cry1F Cry1Ab Vip3Aa20 locations from no injury up to 0.42 cm2 per ear in Mississippi in (Optimum Leptra). Plots with blended refuges also had greater area 2013. When examining the refuge blend effect at each location, the of kernels injured in 2012 (5, 10, and 20%), in early- and late- only significant differences in plots with 100% Cry1F planted corn in 2013 (5, 10, and 20%), and in early (10 and 20% Cry1Ab Vip3Aa20 (Optimum Leptra) corn were between only)- and late-planted corn (5, 10, and 20%) in 2014 compared to Starkville, MS, and all other locations in the 2013 early-planting date. a pure stand of Cry1F Cry1Ab Vip3Aa20. In the 100% non-Bt This suggests that either 1) corn earworm populations were less sus- plots used in the 2014 trials, 62 and 71% of the ears had corn ear- Downloaded from ceptible at this location to the Bt toxins, 2) expression of the toxins worm feeding in the early- and late-planting dates, respectively. varied among locations, as has been shown for Cry1Ab (event Despite these differences in injury, yield averaged across locations MON810) in corn (Nguyen and Jehle 2007) and for Cry1Ac in cotton varied among refuge blends only in the late-planted trials in 2013, (Greenplate 1999), and/or 3) the higher corn earworm pressure in this with greater yields in the 0% refuge blend than in the 20% blend; trial led to higher levels of injury in 100% Cry1F Cry1Ab however, when examining yield separately by location, only two of http://jee.oxfordjournals.org/ Vip3Aa20. nine locations had higher yields in the 100% Bt plots than in any of Previous work has shown that hybrids expressing Cry1Ab, the blended refuge plots. Refuge blend also interacted significantly Cry1F, or Cry1Ab Cry1F provide poor to fair control of corn ear- with location in early- and late-planted trials in 2014, with reduced worm infestation in the corn ear (Reay-Jones and Reisig 2014, yields associated with increased refuge only in Lubbock, TX. These Reisig et al. 2015). Based on the number of injured kernels per ear, a data indicate that the addition of non-Bt refuge seed generally did study in the Carolinas showed that corn plants expressing Cry1F, not have a negative impact on yield. Ear injury in the 100% non-Bt Cry1Ab, or Cry1Ab Cry1F reduced injury by 30, 51, and 55%, re- hybrid in the 2014 trials averaged 2.71 cm2, with a range of 0.14 spectively (Reay-Jones and Reisig 2014). Kernel consumption by cm2 in the early-planted trial in Kentucky to 8.02 cm2 in the late- corn earworm was not significantly different between a non-Bt hy- planted trial in Georgia. Lack of yield response to ear injury from by guest on June 21, 2016 brid and a Cry1F hybrid in trials in Mississippi, Tennessee, and corn earworm feeding had previously been reported in trials in Georgia (Siebert et al. 2012). Corn expressing Vip3Aa20 generally North and South Carolina (with injury up to 3.75 cm2 in a non-Bt provides excellent control of corn earworm based on trials in the hybrid; Reay-Jones and Reisig 2014), which can be partially due to southern United States and in Minnesota (Burkness et al. 2010, intraear compensation for kernel injury to ear tips (Steckel and Reay-Jones and Reisig 2014, Reisig et al. 2015, Yang et al. 2015). In Stewart 2015). When planting within recommended dates, yield ad- our study, the excellent levels of control of kernel injury in the vantages of Bt hybrids relative to non-Bt near isolines have rarely 100% Cry1F Cry1Ab Vip3Aa20 (Optimum Leptra) plots were been documented when corn earworm is the major target pest (Allen therefore likely due to the expression of Vip3Aa20 rather than and Pitre 2006; Bohnenblust et al. 2013; Buntin et al. 2001, 2004; Cry1F and Cry1Ab. Reay-Jones and Reisig 2014). Whereas the vast majority of kernel injury was caused by corn Late-planted corn can sometimes lead to increased pressure from earworm feeding, the insect species (corn earworm or fall army- corn earworm and fall armyworm (Buntin et al. 2001, 2004). In our worm) responsible for leaf injury was generally not recorded, as was study, planting date did not affect the percentage of injured ears, the the case in Reisig et al. (2015). Leaf injury was therefore the best in- area of injured kernels, or leaf injury, although a strong trend dicator of fall armyworm injury. In trials where leaf injury occurred, (P ¼ 0.07) was noted for increased leaf injury in the late planting ratings in plots of 100% Cry1F Cry1Ab Vip3Aa20 (Optimum date during 2014. Fall armyworm are not known to diapause and Leptra) ranged from <0.01 in the early-planted trial in 2014 to 0.23 do not survive freezing temperatures (Sparks 1979). Long-range dis- in the early-planted trial in 2013. Ratings in 100% non-Bt plots in persion occurs every year from overwintering sites in Texas and early- and late-planted trials in 2014 were 86- and 70-fold greater Florida to central and eastern United States and Canada (Luginbill than in 100% Cry1F Cry1Ab Vip3Aa20 (Optimum Leptra) 1928). Within the range of our study, the northward movement plots (Fig. 1). Limited field data are available on efficacy of from these overwintering sites generally leads to moths arriving Vip3Aa20 Bt corn on fall armyworm. Leaf injury in Reisig et al. from April–May in Louisiana, Texas, Mississippi, Alabama, and (2015) was not significantly different between Vip3Aa20 Cry1Ab, Georgia, to late June–July in North Carolina and Tennessee (Snow Cry1F Cry1Ab, and Cry1F, though all three hybrids had signifi- and Copeland 1969). Leaf injury occurred in 27 out of 54 trials, cantly less injury than two non-Bt hybrids. A laboratory study based with injury occurring in every state at least once during the three on field collected fall armyworm larvae from Louisiana and Florida years of this study. However, injury was generally limited and did showed that none of 150 F2 families screened on leaf tissue from a not reach levels that can sometimes lead to substantial yield loss in Cry1Ab Vip3Aa20 corn hybrid survived after seven days (Yang corn (e.g., 71% yield loss in a trial in Georgia [Roberts and All et al. 2013); although the high-dose qualification for fall armyworm 1993]). 12 Journal of Economic Entomology, 2016, Vol. 0, No. 0

Seed mixtures of pyramided Bt corn and non-Bt corn have been 2005, Goldstein et al. 2010, Wangila et al. 2013) are major concerns used commercially since 2010 (U.S. Environmental Protection that may diminish resistance management when using seed blends, Agency [USEPA] 2011) to ensure that growers are in compliance the burden of refuge compliance shifts from the grower to the seed with refuge requirements and that refuge plants are distributed producer when using refuge blends, which modeling suggests are throughout corn fields. Refuge blends may also favor random mat- more effective in delaying resistance with low levels of structured ing of resistant and susceptible individuals (Davis and Onstad refuge compliance (Carroll et al. 2012). As a complement to study- 2000). However, the use of blended refuges may make IRM riskier ing the contribution of blended refuge to delaying resistance, quan- because of larval movement, increased adoption of Bt corn (Onstad tifying injury and yield in a range of refuge blends as in our study is et al. 2001), and cross-pollination for ear feeding pests (Burkness a necessary step to provide management information to growers. et al. 2011). Interplant larval movement may hasten evolution of resistance through exposure to sublethal doses of Bt toxins, though the value of adopting blended refuges in IRM will vary with insect Acknowledgments species (Onstad et al. 2001). While corn earworm larval movement This study was funded in part by DuPont Pioneer. We thank key assistants in between plants is fairly low (range of 2–16%; Burkness et al. 2015), the field project: William Griggs and Terry Teal (Clemson University), Jessica a recent larval movement study showed that 50% of fall armyworm Bessin (University of Kentucky), and Clifton Moore and Steven Roberson were recovered within the infested row, and 91% within a radius of (North Carolina State University). Appreciation is also expressed to Dr. 1.1 m (Pannuti et al. 2016). While this was less than the western William Bridges (Department of Mathematical Sciences, Clemson University) bean cutworm, Striacosta albicosta (Smith), in the same study, sig- for help with statistical analysis. nificant dispersion of fall armyworm occurred across rows (Pannuti et al. 2016). Sugarcane borer larvae move up to four plants away Downloaded from References Cited and to adjacent rows, but the majority stay on the infested row (Wangila et al. 2013). Lesser cornstalk borer feeds in silken tubes Allen, K. C., and H. N. Pitre. 2006. Influence of transgenic corn expressing in- just below the soil surface (Tippins 1982); larvae can move readily secticidal proteins of Bacillus thuringiensis Berliner on natural populations and a blended refuge may not be effective for this species. While of corn earworm (Lepidoptera: Noctuidae) and southwestern corn borer more information is needed on interplant movement of larvae sus- (Lepidoptera: Crambidae). J. Entomol. Sci. 41: 221–231. http://jee.oxfordjournals.org/ ceptible to Bt toxins, structured refuges may be more appropriate Bates, S. L., J.-Z. Zhao, R. T. Roush, and A. M. Shelton. 2005. Insect resistance management in GM crops: past, present, and future. Nat. Biotechnol. than blended refuges for insects with significant interplant move- 23: 57–62. ment of larvae. Another issue is cross-pollination, which may lead to Bohnenblust, E., J. Breining, S. Fleischer, G. Roth, and J. Tooker. 2013. Corn Bt toxins expressed in blended refuge plants, leading to sublethal ex- earworm (Lepidoptera: Noctuidae) in northeastern field corn: infestation posure of target insects to Bt toxins (Burkness et al. 2011). Yang levels and the value of transgenic hybrids. J. Econ. Entomol. 106: et al. (2014) showed that 90% of the kernels in non-Bt blended ref- 1250–1259. uge ears expressed at least one Bt protein because of cross- Buntin, G. D., R. D. Lee, D. L. Wilson, and R. M. McPherson. 2001. pollination, thus potentially exposing ear-feeding Lepidoptera to Evaluation of YieldGard transgenic resistance for control of fall armyworm single rather than pyramided traits. Other field studies on seed mix- and corn earworm (Lepidoptera: Noctuidae) on corn. Fla. Entomol. 84: by guest on June 21, 2016 tures have identified refuge plants in the blend with the goal of 37–42. examining the contribution of these plants to the production of sus- Buntin, G. D., J. N. All, R. D. Lee, and D. M. Wilson. 2004. Plant-incorpo- ceptible individuals. While Reisig et al. (2015) showed that injury rated Bacillus thuringiensis resistance for control of fall armyworm and from corn earworm and fall armyworm in non-Bt plants within a corn earworm (Lepidoptera: Noctuidae) in corn. J. Econ. Entomol. 97: blended refuge did not differ significantly from non-Bt plants in 1603–1611. plots mimicking a structured refuge, Yang et al. (2014) showed in Burkness, E., G. Dively, T. Patton, A. Morey, and W. Hutchison. 2010. Novel Vip3A Bacillus thuringiensis (Bt) maize approached high-dose efficacy lab assays that survival from neonate to adult of corn earworm was against Helicoverpa zea (Lepidoptera: Noctuidae) under field conditions. reduced from 38.3% in non-Bt plants within a structured refuge to GM Crops 1: 337–343. 4.6% in non-Bt plants in a blended refuge. Crespo et al. (2016) re- Burkness, E. C., P. K. O’Rourke, and W. D. Hutchinson. 2011. Cross-pollin- ported that pure stand plants produced around 70% more larvae ation of nontransgenic corn ears with transgenic corn: efficacy against lepi- per ear than blended refuge plants mixed with plants containing the dopteran pests and implications for resistance management. J. Econ. MIR162 protein. While both larval movement and cross-pollination Entomol. 104: 1476–1479. represent significant challenges for blended refuges in the southern Burkness, E. C., T. M. Cira, S. E. Moser, and W. D. Hutchinson. 2015. Bt United States, a recent model showed that blended refuges provided maize seed mixtures for Helicoverpa zea (Lepidoptera: Noctuidae): Larval similar durability compared with structured refuges when compli- movement, development, and survival on non-transgenic maize. J. Econ. ance is 50% (Pan et al. 2016). Entomol. 108: 2761–2769. In conclusion, this study showed that yields varied among loca- Carpenter, J. E. 2011. Impact of GM crops on biodiversity. GM Crops 2: tions across nine states, but were generally not affected by refuge 7–23. blend. Corn expressing Cry1F Cry1Ab Vip3Aa20 (Optimum Carroll, M. W., G. P. Head, and M. Caprio. 2012. When and where a seed mix refuge makes sense for managing insect resistance to Bt plants. Crop Leptra) was effective in reducing injury to ears and leaf injury from Prot. 38: 74–79. corn earworm and fall armyworm, and in reducing stalk injury from Crespo, A.L.B., A. P. Alves, Y. Wang, B. Hong, J. L. Flexner, A. Catchot, D. southwestern corn borer in one trial with artificial infestations. The Buntin, and D. Cook. 2016. Survival of corn earworm (Lepidoptera: general lack of a yield response to the levels of injury recorded Noctuidae) on Bt maize and cross-pollinated refuge ears from seed blends. across nine states may facilitate adoption of a blended refuge by J. Econ. Entomol. 109: 288–298. growers, particularly as structured refuge areas can be perceived by Cullen, E. M., M. E. Gray, A. J. Gassmann, and B. E. Hibbard. 2013. growers as lost income because of lower yields, higher pest pressure, Resistance to Bt corn by western corn rootworm (Coleoptera: and need for separate management (Storer et al. 2012). While cross- Chrysomelidae) in the U.S. Corn Belt. J. Integr. Pest Mngmt. 4: 1–3. pollination (Yang et al. 2014) and larval movement (Bates et al. doi:10.1603/IPM13012. Journal of Economic Entomology, 2016, Vol. 0, No. 0 13

Davis, P. M., and D. W. Onstad. 2000. Mixtures as a resistance management Reay-Jones, F.P.F., and D. D. Reisig. 2014. Impact of corn earworm on yield strategy for European corn borers (Lepidoptera: Crambidae) infesting trans- of transgenic corn producing Bt toxins. J. Econ. Entomol. 107: 1101–1109. genic corn expressing Cry1Ab protein. J. Econ. Entomol. 93: 937–948. Reisig, D. D., D. S. Akin, J. N. All, R. T. Bessin, M. J. Brewer, D. G. Buntin, Davis, F. M., C. A. Henderson, and G. E. Scott. 1972. Movements and feeding A. L. Catchot, D. Cook, K. L. Flanders, F-N. Huang, et al. 2015. of larvae of the southwestern corn borer on two stages of corn growth. J. Lepidopteran injury to corn containing single and pyramided Bt traits, and a Econ. Entomol. 65: 519–521. refuge blend with Cry1Ab x Cry1F, in the southern United States. J. Econ. Davis, F. M., S. S. Ng, and W. P. Williams. 1992. Visual rating scales for Entomol. 108: 157–165. screening whorl-stage corn for resistance to fall armyworm. Miss. Agric. Rice, M. E., and C. D. Pilcher. 2001. Potential benefits and limitations of For. Exp. Stn. Tech. Bull. 186: 1–9. transgenic Bt corn for management of the European corn borer Goldstein, J. A., C. E. Mason, and J. Pesek. 2010. Dispersal and movement be- (Lepidoptera: Crambidae). Am. Entomol. 44: 75–78. havior of neonate European corn borer (Lepidoptera: Crambidae) on non- Roberts, P. M., and J. N. All. 1993. Hazard for fall armyworm (Lepidoptera: Bt and transgenic Bt corn. J. Econ. Entomol. 103: 331–339. Noctuidae) infestation of maize in double-cropping systems using sustain- Gould, F. 1998. Sustainability of transgenic insecticidal cultivars: Integrating able agricultural practices. Fla. Entomol. 76: 276–283. pest genetics and ecology. Annu. Rev. Entomol. 43: 701–726. SAS Institute. 2011. User’s manual, version 9.3. SAS Institute, Cary, NC. Greenplate, J. T. 1999. Quantification of Bacillus thuringiensis insect control Siebert, M. W., S. P. Nolting, W. Hendrix, S. Dhavala, C. Craig, B. R. protein Cry1Ac over time in Bollgard cotton fruit and terminals. J. Econ. Leonard, S. D. Stewart, J. All, F. R. Musser, G. D. Buntin, et al. 2012. Entomol. 92: 1377–1383. Evaluation of corn hybrids expressing Cry1F, Cry1A.105, Cry2Ab2, Hensley, S. D., and K. D. Arbuthnot. 1957. Migration of larvae of southwest- Cry34Ab1/Cry35Ab1, and Cry3Bb1 against southern United States insect ern corn borer on corn. J. Econ. Entomol. 50: 103. pests. J. Econ. Entomol. 105: 1825–1834. Huang, F., D. A. Andow, and L. L. Buschman. 2011. Success of the high-dose/ Sparks, A. N. 1979. A review of the biology of the fall armyworm. Fla. refuge resistance management strategy after 15 years of Bt crop use in North Entomol. 62: 82–86. America. Entomol. Exp. Appl. 140: 1–16. Snow, J. W., and W. W. Copeland. 1969. Fall armyworm: Use of virgin female Downloaded from Huang, F., J. A. Qureshi, R. L. Meagher, Jr, D. D. Reisig, G. P. Head, D. A. traps to detect males and to determine seasonal distribution. USDA Prod. Andow, A. Ni, D. Kerns, G. D. Buntin, Y. Niu, et al. 2014. Cry1F resistance Res. No. 110: 9. in fall armyworm Spodoptera frugiperda: single gene versus pyramided Bt Steckel, S., and S. D. Stewart. 2015. Intraear compensation of field corn, Zea maize. PLoS ONE 9: e112958. (doi:10.1371/journal.pone.0112958) mays, from simulated and naturally occurring injury by ear-feeding larvae. Hutchinson, W. D., and N. P. Storer. 2010. Expanded use of pyramided trans- J. Econ. Entomol. 108: 1106–1114. genic maize hybrids expressing novel Bacillus thuringiensis toxins in the http://jee.oxfordjournals.org/ Storer, N. P., J. M. Babcock, M. Schlenz, T. Meade, G. D. Thompson, J. W. southern U.S.: Potential for areawide suppression of Helicoverpa zea Bing, and R. M. Huckaba. 2010. Discovery and characterization of field re- (Lepidoptera: Noctuidae) in the Mississippi Delta. Southwest. Entomol. 35: sistance to Bt maize: Spodoptera frugiperda (Lepidoptera: Noctuidae) in 403–408. Puerto Rico. J. Econ. Entomol. 103: 1031–1038. Jaffe, G. 2009. Complacency on the farm. Center for Insect Science in the Storer, N. P., G. D. Thompson, and G. P. Head. 2012. Application of pyra- Public Interest, Washington, DC. (http://cspinet.org/new/pdf/complacen mided traits against Lepidoptera in insect resistance management for Bt cyonthefarm.pdf) crops, GM Crops Food 3: 154–162. Kenward, M. G., and J. H. Roger. 1997. Small sample inference for fixed ef- Tippins, H. H. 1982. A review of information on the lesser cornstalk borer fects from restricted maximum likelihood. Biometrics 53: 983–997. Elasmopalpus lignosellus (Zeller). Ga. Agric. Exp. Stn. Spec. Publ. 17.

Koziel, M. G., G. L. Beland, C. Bowman, N. B. Carozzi, R. Crenshaw, L. by guest on June 21, 2016 Tukey, J. W. 1953. The problem of multiple comparisons. Department of Crossland, J. Dawson, N. Desai, M. Hil, S. Kadwell, et al. 1993. Field per- Statistics, Princeton University, NJ. formance of elite transgenic maize plants expressing an insecticidal protein (USDA ERS) U.S. Department of Agriculture Economic Research Service. derived from Bacillus thuringiensis. Nat. Biotech 11: 194–200. 2015. Adoption of genetically engineered in the U. S. (http://www.ers.usda. Littell, R. C., G. A. Milliken, W. W. Stroup, R. D. Wolfinger, and O. gov/data-products/adoption-of-genetically-engineered-crops-in-the-us. Schabenberger. 2006. SAS for mixed models, 2nd ed. SAS Institute, Cary, aspx) NC. (USEPA) U.S. Environmental Protection Agency. 2011. Biopesticides Luginbill, P. 1928. The fall armyworm. U.S. Dep. Agric. Tech. Bull. 34: 1–91. Registration Action Document. In MON89034 x TC1507 xMON88017 Naranjo, S. E. 2009. Impacts of Bt crops on non-target invertebrates and in- VR secticide use patterns. CAB Rev. Perspect. Agric. Vet. Sci. Nutr. Nat. xDAS-59122-7 (SmartStax ) B. t. Corn Seed Blend. US Environmental Resources 4: No. 011. (doi: 10.1079/PAVSNNR20094011) Protection Agency, Washington, DC. (https://www3.epa.gov/pesticides/ Nguyen, H. T., and J. A. Jehle. 2007. Quantitative analysis of the seasonal and chem_search/reg_actions/pip/smartstax-seedblend.pdf) tissue-specific expression of Cry1Ab in transgenic maize Mon810. J. Plant Wangila, D. S., B. R. Leonard, M. N. Ghimire, Y. Bai, L. Zhang, Y. Yang, K. Dis. Prot. 114: 82–87. D. Emfinger, G. P. Head, F. Yang, Y. Niu et al. 2013. Occurrence and larval Onstad, D. W., P. D. Mitchell, T. M. Hurley, J. G. Lundgren, R. P. Porter, C. movement of Diatraea saccharalis (Lepidoptera: Crambidae) in seed mixes H. Krupke, J. L. Spencer, C. D. DiFonzo, T. S. Baute, R. L. Hellmich, et al. of non-Bt and Bt pyramid corn. Pest Manag. Sci. 69: 1163–1172. 2011. Seeds of change: corn seed mixtures for resistance management and Wiseman, B. R., F. M. Davis, and J. E. Campbell. 1980. Mechanical infest- IPM. J. Econ. Entomol. 104: 343–352. ation device used in fall armyworm plant resistance programs. Fla. Ostlie, K. R., W. D. Hutchison, and R. L. Hellmich. 1997. Bt corn and Entomol. 63: 425–432. European corn borer. NCR Publication 602. University of Minnesota, St. Yang, F., H. Huang, J. A. Qureshi, B. R. Leonard, Y. Niu, L. Zhang, and D. S. Paul, MN. Wangila. 2013. Susceptibility of Louisiana and Florida populations of Pan, Z., D. Onstad, P. Crain, A. Crespo, W. Hutchison, D. Buntin, P. Porter, Spodoptera frugiperda (Lepidoptera: Noctuidae) to transgenic AgrisureVR A. Catchot, D. Cook, C. Pilcher, et al. 2016. Evolution of resistance by VipteraTM 3111 corn. Crop Prot. 50: 37–39. Helicoverpa zea (Lepidoptera: Noctuidae) infesting insecticidal crops in the Yang, F., D. L. Kerns, G. P. Head, B. R. Leonard, R. Levy, Y. Niu, and F. southern United States. J. Econ. Entomol. 109: 821–831. Huang. 2014. A Challenge for the seed mixture refuge strategy in Bt maize: Pannuti, L.E.R., S. V. Paula-Moraes, T. E. Hunt, E.L.L. Baldin, L. Dana, and impact of cross-pollination on an ear-feeding pest, corn earworm. PLoS J. V. Malaquias. 2016. Plant-to-plant movement of Striacosta albicosta ONE 9: e112962. (doi:10.1371/journal.pone.0112962) (Lepidoptera: Noctuidae) and Spodoptera frugiperda (Lepidoptera: Yang, F., D. L. Kerns, B. R. Leornard, I. Oyediran, T. Burd, Y. Niu, and F. TM Noctuidae) in maize (Zea mays). J. Econ. Entomol. in press. Huang. 2015. Performance of AgrisureVR Viptera 3111 corn against Pilcher, C. D., J. J. Obrycki, M. E. Rice, and L. C. Lewis. 1997. Preimaginal Helicoverpa zea (Lepidoptera: Noctuidae) in seed mixed plantings. Crop development, survival, and field abundance of insect predators on transgenic Prot. 69: 77–82. Bacillus thuringiensis corn. Environ. Entomol. 26: 446–454. Zar, J. H. 1999. Biostatistical analysis. Prentice Hall, Upper Saddle River, NJ.