FARMER FRIENDLY SUMMARY. Cotton Insect Pest Management, Agreement TN Scott Stewart, The University of Tennessee

Transcription

1 FARMER FRIENDLY SUMMARY Cotton Insect Pest Management, Agreement TN Scott Stewart, The University of Tennessee Funding is used to partially support general extension activities such as scout training, moth trapping, resistance monitoring, insecticide testing and on-farm evaluations of various insect control technologies and treatment thresholds. Funding is also used to help support regional projects, several of which are also partially supported by core grants from Cotton Incorporated. Pheromone moth trapping for bollworm and tobacco budworm are improving the decision making of crop managers. In 2015, moth catches for each trap location were reported weekly on the UTcrops News Blog ( This information and other IPM news updates were distributed to agents, producers, consultants and other agricultural professionals. Statewide, average moth catches for bollworm (i.e., corn earworm) were well below average trap catches for the previous decade. As typical, statewide catches of tobacco budworm in 2015 were lower than bollworm but were higher than typically observed. Moth traps did well at predicting the general lack corn earworm (bollworm) pressure observed in flowering cotton. Assays using bollworm moths indicated substantially higher resistance to pyrethroid insecticides than in previous years, a phenomenon that also occurred in other states. An assessment of moth traps showed that Hartstack-style trap are better suited than bucket-style traps for estimating the size of bollworm moth flights. Many experiments were done in Tennessee to assess control options for thrips, tarnished plant bug, bollworms, and spider mites. The results of most experiments have been individually summarized and published on the UTcrops.com website ( Examples presented in this report include the influence of nozzle type on control of tarnished plant bugs with Acephate and a demonstration of the value of insect management in blooming cotton. Regional research efforts included 1) an evaluation of insecticide and insecticide/nematicide seed treatments on thrips control, 2) determining the value of supplemental insecticide applications for control of caterpillar pests in commercially available Bt cotton technologies, and 3) experiments to assess when insecticide applications for tarnished plant bug can be terminated. Comprehensive reports for these efforts have been submitted by the appropriate project leaders, but abbreviated reports for these projects are included. We observed generally poor performance of thiamethoxam (e.g., Cruiser ) in controlling thrips populations and injury. This is consistent with the recent years observations. Regional monitoring efforts indicate continued resistance of tobacco thrips to thiamethoxam (e.g., Cruiser) and developing resistance to imidacloprid (e.g., Gaucho ). Bt cottons are providing substantial protection form caterpillar pests, but our data show than foliar insecticide applications are sometimes warranted when bollworm pest pressure in high. Other data indicate that the first four weeks of bloom is a critical window to protect cotton for threshold infestations of tarnished plant bug. The data generated from these above experiments are used to validate and modify extension insect control recommendations in Tennessee. Over 1,500 copies of the UT insect control recommendations for cotton were distributed to stakeholders. This and additional information was also distributed over the internet.

2 ANNUAL REPORT Cotton Insect Pest Management, Agreement TN Scott Stewart, The University of Tennessee Funding for this project was used to partially support general extension activities such as scout training, moth trapping, evaluation of existing and new Bt cotton technologies, resistance monitoring, and insecticide testing and validation of treatment thresholds. In addition, funds were also used to help support several regional projects as described herein. 1) Moth Trapping. Despite the use of Bt-transgenic cotton on over 98% of the acreage in Tennessee, bollworm and tobacco budworm still compose an important pest complex. Bollworms may cause significant economic damage to Bt cotton fields, and the bollworm/budworm can be even more damaging to non-bt cotton. More importantly, the threat of tobacco budworm infestations results in high adoption of Bt cotton. Resistance to pyrethroid insecticides in tobacco budworm populations makes distinguishing between budworm and bollworm infestations is very critical in non-bt cotton. Area-wide monitoring remains a valuable tool in predicting the occurrence and size of pest populations. Pheromone trapping programs for bollworm, tobacco budworm, and beet armyworm provide insight into the timing and intensity of moth flights. For example, unusually high trap catches for a particular species can alert consultants and producers to the potential for impending outbreaks. When performed on a regional level and over a number of years, moth trapping can indicate historical and geographical patterns in the distribution of pest populations. Moth monitoring improves the decision making process, helping crop managers in the selection of insecticides and to indicate the need for intensified sampling efforts. This ultimately helps to minimize control costs and/or yield losses incurred by producers. Traps are used to collect moths used in assays for resistance to pyrethroid insecticides. Pheromone moth traps for corn earworm (CEW or bollworm) and tobacco budworm (TBW) were run on a weekly basis from early May through late August. Pheromone lures were obtained from Great Lakes IPM (Vestaburg, MI) and were changed at two week intervals. At least one, and usually two, sets of bollworm and tobacco budworm traps were run in each of the following 12 counties in West Tennessee: Carroll, Crockett, Dyer, Fayette, Gibson, Hardeman, Haywood, Shelby, Tipton, Lake, Lauderdale, and Madison. A comparison between Hartstack- and universal-style (bucket) moth traps was done for bollworm at a total of 24 locations in Arkansas, Louisiana and Tennessee. Traps at each of 24 locations were established in Arkansas (12), Louisiana (6) and Tennessee (6). Each location, included a Hartstack trap baited with one Hercon pheromone lure, a bucket trap baited with one lure, and a bucket trap baited with two lures. Vapona kill strips were used in bucket traps. The trapping period for at least some traps began in mid-may and extended through mid-september, but the primary trapping period was June through August. Traps were checked weekly, and the number of moths in each trap was recorded. Pheromone lures were replaced every second week.

3 Outcomes: Moth catches for each trap were reported weekly on the UTcrops News Blog ( which was distributed to agents, cotton producers, consultants and other agricultural professionals. As usual, catches of corn earworm (i.e., bollworm) moths in pheromone traps were greater than tobacco budworm, although the difference was not as great in 2015 compared with previous years (Fig. 1). The highest moth catches were observed in Lake County (Fig. 2). Fewer corn earworm moths were caught 2015 than normal (Fig. 3). The bollworm is Tennessee s most significant caterpillar pest in cotton because this species is able to cause economic injury to Bt cotton which composes the vast majority of the acreage. However, as predicted by trap catches, late-season infestations observed in July and August were generally low. Few if any cotton fields were treated for tobacco budworm. This is not surprising considering the very low acreage of non-bt cotton. However, tobacco budworm moth catches were relatively high compared with previous years (Fig. 4). The highest single-trap capture was recorded in Crockett County (Maury area) where 141 moths were caught the week preceding August 13. This trap caught many tobacco budworms moths compared with other locations. Figure 1. Average number of tobacco budworm and corn earworm (bollworm) moths caught per trap in pheromone traps across West Tennessee, Trapping did not necessarily reflect all local variations in pest densities observed in cotton fields, in part because trap density was not high and because other factors influence oviposition and survival of these pests in cotton. However, the pheromone trapping program did a good job of predicting the generally low infestations of bollworms observed in cotton. We also participated in a national fall armyworm trapping effort, PestWatch ( Results are available online and will not be reported here.

4 Figure 2. Seasonal average number of corn earworm (bollworm) moths caught per trap per week in each county, Figure 3. Average statewide catches of corn earworm (bollworm) moths, comparing 2015 to the average catches for

5 Figure 4. Average statewide catches of tobacco budworm moths, comparing 2015 to the average catches for The comparison of Hartstack and bucket traps indicated that bucket traps may be suitable or perhaps better when monitoring invasive moths such as the old world bollworm that are expected to occur in low numbers, and two lures would increase the numbers of moths caught. Because average catches in bucket traps plateaued at about 20 moths/trap/week (Fig. 5), it appears Hartstack-style traps are more suitable for estimating the size of moth flights for the purposes of making IPM decisions or for collecting moths for other purposes. The complete results of this experiment were submitted for publication in the 2016 Proceedings of the Beltwide Cotton Conferences. Figure 5. Regression fit for the average number of H. zea moths caught in bucket traps baited with one (left) or two (right) pheromone lures versus the number of moths caught in Hartstack traps.

6 2) Insecticide Efficacy and Threshold Verification Trials. Multiple replicated experiments are done annually to evaluate insecticide efficacy, new or alternative treatments, and validate IPM recommendations. Small-plot experiments are done as randomized and replicated trials on UT experiment stations or in grower fields. Individual plot size varies from 4-8 rows and feet in length. The data generated from these activities are used to validate and modify extension insect control recommendations for Tennessee. These evaluations included insecticide efficacy trials for thrips, plant bugs, stink bugs, and bollworm. Two examples are presented below. However, the results of most experiments have been individually summarized will be made available on UTcrops.com at the following link: 2.1) Threshold Verification. Studies were conducted in 2010, 2014, and 2015 to demonstrate the value of foliar insecticides as part of an integrated pest management approach to reduce the threat of crop loss from several important cotton pests that have thrived after the eradication of the boll weevil, despite the use of Bt cotton. These tests were done at the West Tennessee Research and Education Center in Jackson. Phytogen 375 WRF (WideStrike) cotton was planted no-till May 7, 2010 and May 12, Phytogen 333 WRF (WideStrike) cotton was planted no-till May 11, Individual plots were 8 rows (38 inch centers) x feet. Treatments were replicated in a randomized complete block design. All agronomic practices such as fertilization, seeding rates and control of insect pests followed University of Tennessee recommendations. Plots were scouted using proven sampling methods throughout the growing season. Pre-bloom foliar insecticides were made as needed to all plots. Foliar insecticide applications were not made prior to flowering in 2010, but applications for plant bugs were needed in 2014 and Foliar applications for the control of insect pests during the flowering period of cotton were made only when established economic thresholds were met or exceeded. Five applications were made during bloom in 2010, four in 2014, and three in Insecticides used were local standards such as Acephate, Acephate + Brigade (bifenthrin), Transform WG (sulfoxaflor), Bidrin XP II (bifenthrin + dicrotophos) depending on the pest spectrum present at the time of application. Yield data were collected on September 20, 2010, October 22 and November 2, 2014 and October 1, 2015 by harvesting the center two rows of each plot. Lint yield was calculated assuming a 40% gin turn out and a lint value of $0.70 per pound. Outcomes: Plant bug, stink bugs and bollworm were the primary targets of insecticide applications made in these tests. The tarnished plant bug was the most common pest, and populations in nontreated plot greatly exceeded the recommended treatment threshold of six or more insects per 10 row feet (Fig. 6). Foliar insecticides successfully controlled damaging insect pests during the flowering period (not shown). The use of foliar insecticides as part of an integrated pest management program resulted in significant increase in cotton yield for all three years. Compared to plots not treated with insecticide, lint yield was increased 1002, 632, and 515 pounds per acre in 2010, 2014, and 2015, respectively (Fig. 7). This provided a substantial increase in crop value of $701, $387, and $317 per acre in 2010, 2014, and 2015, respectively. This included the cost of insecticide and application. These data demonstrate the value of managing insect pests in cotton.

7 Figure 6. Total plant bugs per 10 row feet in non-treated plots during flowering. Figure 7. Lint yield (Lbs/acre in parentheses) and gross crop value ($/acre; $0.70/lb) 2.2) Evaluation of Nozzle-type Effects on Control of Tarnished Plant Bug. Tests were done in 2015 to evaluate how nozzle type might influence the control of insect pests with foliar insecticides. The control of tarnished plant bug with Orthene (0.67 Lb AI/A) was used as an example. This issue may become more important as new GMO herbicide technologies require the use of low drift nozzles to apply dicamba or 2,4-D. These same nozzles may often be used to apply insecticides. A single application of Orthene 97S was made July 31 using several nozzle

8 types. Efforts were made to maintain an application volume of 15 GPA and 40 PSI. Numbers of plant bugs were counted by taking two drop cloth samples in each plot at five days after application. Outcomes: Nozzle type did not affect the efficacy of Orthene in controlling tarnished plant bugs (Table 1). Indeed, the numerically highest number of bugs was found in the flat fan nozzle (treatment 1) often recommended for insect control. Reduced drift nozzles (treatments 2-4) performed similarly to each other. Although encouraging in that nozzle type did not appear critical, further tests are needed to determine how nozzle selection and other application parameters influence the efficacy of foliar applied insecticides. Table 1. Influence of nozzle type on control of tarnished plant bug with Orthene. Tarnished Plant Bugs/10 Row Feet Treatment Immature Adults + Immatures 1 Orthene 97S (0.67 LB) 7.5 b 7.8 b TeeJet XR Orthene 97S (0.67 LB) 5.0 b 5.3 b Greenleaf TTI002 3 Orthene 97S (0.67 LB) 5.8 b 6.5 b Greenleaf TADF02 4 Orthene 97S (0.67 LB) 4.0 b 4.5 b Wilger MR Not treated 33.3 a 34.8 a Means followed by same letter do not significantly differ (P< 0.05, LSD). 3) Support of General Extension Activities. Funding for this project is used to support general IPM Extension activities in Tennessee. This includes the delivery of the annual Cotton Scout School held at the West Tennessee Research and Education Center. Scouts are delivered classroom-style and in-field training related to cotton plant development, insect management (identification, sampling, etc.) and disease and weed identification. A scouting notebook was prepared for each attendee. This project also supports the preparation and publication of Insect Control Recommendations for Field Crops (UT Publication, PB1768) which contains IPM information for cotton and other crops. This publication is also available at Outcomes: Approximately 60 scouts participated in the Cotton Scout School during A scouting notebook was prepared for each attendee. 1,500 copies of the Insect Control Recommendations for Field Crops (PB1768) were distributed to clientele groups. 4) Regional Projects. In 2015, a limited amount of core Cotton Incorporated funding was received to support regional research efforts (Project Helicoverpa zea Pyrethroid Resistance Monitoring; Project Insect Management in Mid-South Cotton). Additionally, unsupported regional efforts are also performed. Shortened summaries are included here because state-support funding is substantially used to complete these efforts.

9 4.1) Bollworm - Pyrethroid Resistance Monitoring (F. Musser, et al.). Monitoring pyrethroid resistance in bollworms has been conducted for numerous years. Throughout this time the methodology has remained consistent. Moths are individually placed in 20 ml scintillation vials that had been previously coated with 0 or 5 µg cypermethrin per vial. A concentration of 5 µg cypermethrin/vial is used with baseline survival generally less than 10%. Vial preparation was done at Mississippi State University by Dr. Fred Musser. Mortality was assessed after 24 h, and moths were considered dead if they could no longer fly. During 2015, moth traps at the West Tennessee Research and Education Center were used to collect bollworm moths for vial assays. Healthy moths were subsequently tested for pyrethroid resistance. Because the moth flight was relatively late and small, only 349 moths were tested. All of these moths were trapped in August or Early September. Outcomes: We observed < 40% mortality of 349 moths in vials treated with cypermethrin. Mortality in untreated vials was low. The survivorship in treated vials was much higher than observed in previous years, and this same phenomenon was observed in North Carolina and Mississippi (Table 2, Fig. 8). This spread of resistance to new regions is an indication of insect movement and continued selection for pyrethroid resistance. Table 2. Bollworm survival to 5 µg cypermethrin per vial in 24-h vial tests during State May June July Aug Sep Overall Total bollworms tested AR GA LA MS MO NC SC TN TX VA Average Prepared by F. Musser, Mississippi State University.

10 Figure 8. Beltwide bollworm average survival per year at 5 µg cypermethrin per vial from ) Tobacco thrips Neonicotinoid Resistance Monitoring. It is now well documented that tobacco thrips have developed resistance to thiamethoxam. The worsening of resistance, including resistance to imidacloprid, poses a substantial threat as these insecticides are used almost exclusively in Tennessee for the control of thrips in cotton. In early June, a population of tobacco thrips was collected from cotton from Madison Co., TN and submitted to Dr. Fred Musser (Mississippi State University) for evaluation of resistance levels to imidacloprid and thiamethoxam. Also, multiple populations from the Research and Education Centers in Milan and Jackson were submitted to Syngenta for the same purpose. Outcomes: The tobacco thrips population sent to MSU had similar levels of resistance to thiamethoxam as observed in 2014, and compared with a susceptible colony, it was only slightly more tolerant to imidacloprid. More concerning are that indications of resistance to imidacloprid are apparent, including populations from Tennessee submitted to Syngenta and assayed by a third party (Fig. 9). Figure 9. Survivorship of tobacco thrips in diagnostic bioassays. Dot size represents percent survival. Graph courtesy of Huseth et al. (Pest Management Science, DOI: /ps.4232).

11 4.3) Evaluation of insecticide, fungicide and nematicide seed treatments (partly from prepared by Don Cook, Mississippi State University). Thrips are one of the first insect pests to infest cotton following seedling emergence, with tobacco thrips, Frankliniella fusca Hinds, being the predominate species in the Mid-South. Following the introduction of the neonicotinoid seed treatments, thrips management has been almost exclusively accomplished with seed treatments and supplemental foliar treatments. Currently there are two seed treatment insecticides commonly used for thrips management in cotton and include Gaucho (imidacloprid) and Cruiser (thiamethoxam). Generally, the seed treatments have provided satisfactory control. However, recently less than satisfactory control of tobacco thrips has been observed with Cruiser (thiamethoxam). During 2015 studies were conducted in Arkansas, Louisiana, Mississippi, and Tennessee to evaluate the performance of Cruiser and Gaucho seed treatments alone and in combination with their respective companion nematicide products (Aeris and Avicta), as well as, acephate seed treatment at several rates. Studies were conducted during 2015 in Arkansas, Louisiana, Mississippi, and Tennessee to evaluate the performance of selected insecticide seed treatments against thrips in cotton. Treatments were arranged in a randomized complete block design with four replications. Phytogen 333 WRF cotton seed was used in all trials. Cotton seed were treated by Dr. Gus Lorenz and received a common fungicide seed treatment (Trilex Advanced 300FS, 1.6 oz/cwt). Due to wet conditions planting dates varied among locations from 30 Apr to 5 Jun. Thrips densities were determined by sampling 5-10 plants per plot at the 1, 2, 3, and 4 leaf stage using a modified whole plant washing procedure. Also, plant damage was also estimated at these timing using a 1 5 scale, with a rating of 1 = no damage and 5 = severe damage. Lint yield data was also collected. Data were subjected to ANOVA procedures, with means separated according to Fisher s Protected LSD. Outcomes: In Tennessee, thrips numbers and plant protection was best in treatments containing imidacloprid (i.e., Gaucho and Aeris), and Acephate seed treatments provided better protection than either Cruiser of Cruiser Avicta (data not shown). In general, thrips populations were low across the region compared with studies done during On average across locations, Acephate provided thrips control and plant protection that was better than Cruiser and similar to that of Gaucho and Aeris, depending on rate and growth stage (e.g., Table 3). Across all tests, no significant differences in average yields were observed among treatments. However, yields from the imidacloprid treated plots were among the highest, while yields from thiamethoxam treated plots were among the lowest in the studies (Table 4).

12 Table 3. Impact of selected seed treatments on densities of thrips adults, immatures, and total thrips and thrips damage at the 3 leaf growth stage. Thrips per 5 Plants Thrips Treatment Rate Adults Immatures Total Damage Rating Fungicide a 28.0a 2.8a only Acephate b 14.0b 1.7c Acephate b 11.5b 1.7c Acephate b 11.8b 1.7c Cruiser a 23.8a 2.3b Avicta Duo a 25.4a 2.2b Gaucho b 13.8b 1.6c Aeris b 7.7b 1.5c P>F 0.39 <0.01 <0.01 <0.01 Means within a column followed by a common letter are not significantly different (FPLSD 0.05). 1 oz product / cwt. 2 mg AI / seed. 3 mg AI / seed. Avicta Duo and Aeris applied at the listed rates contain mg AI, thiamethoxam (Cruiser) and imidacloprid (Gaucho), respectively. Table 4. Impact of selected seed treatments on cotton yield. Treatment Rate Lb Lint / Acre Fungicide only - 1,101.8 Acephate ,152.3 Acephate ,107.4 Acephate ,146.0 Cruiser ,110.4 Avicta Duo ,182.2 Gaucho ,182.0 Aeris ,201.6 P>F 0.09 Means within a column followed by a common letter are not significantly different (LSD 0.05). 1 oz product / cwt. 2 mg AI / seed. 3 mg AI / seed. Avicta Duo and Aeris applied at the listed rates contain mg AI, thiamethoxam (Cruiser) and imidacloprid (Gaucho), respectively. 4.4) Impact of foliar insecticide application on non-bt and Bt cotton (partly from report prepared by Dr. David Kerns, LSU AgCenter). Since the introduction of Bt cotton in the United States in 1996, management of the Heliothines including the bollworm has become much less problematic. However, there are still incidents where unacceptable fruit injury is experienced and insecticidal oversprays are utilized to prevent yield loss. There has been much speculation surrounding the reasons for control failures among Bt cotton technologies including Bt resistance

13 or tolerance, inadequate expression of Bt toxins expression due to plant phenology or environmental stressors. The objective of this project was to evaluate the efficacy of second and third generation Bt cotton at multiple locations within the Mid-South for efficacy towards Heliothines, predominately bollworms, and to determine if treating Bt cotton results in a reduction in damage and increased yields relative to non-bt cotton. Tests were conducted across the Mid-South at five locations in 2014 and seven location in Four Bt technologies and a non-bt variety were evaluated which included: St 5289 GLT [TwinLink (TL; Cry1Ab, Cry2Ae)], ST 5288 B2RF [Bollgard II (BG2; Cry1Ac, Cry2Ab)], PHY 299 WRF [WideStrike (WS; Cry1Ac, Cry1F)], and PHY 495 W3RF [WideStrike 3 (WS3; Cry1Ac, Cry1F, Vip3A)] and PHY 315 RF [non-bt, (NBT)]. All tests were 5 x 2 factorials with factor A being the cotton technology and factor B being entries sprayed for bollworms with Prevathon at 20 fl-oz/ac, or non-sprayed. Plots were 4 rows wide x ft in length. Each factorial combination was replicated 4 times. Foliar applications were made in accordance with the occurrence of larvae in the non-bt cotton plots at each individual location. When sprayed, Prevathon was used at a rate of 20 Oz/A. Insect densities, square and boll injury were determined prior to foliar treatment and weekly thereafter using one of two techniques depending on test site. Yields were estimated using 2-row mechanized cotton pickers with integrated scales for weighing plot samples. The middle two rows were harvested from each plot and yields were converted to pounds of seed cotton per acre. Outcomes: In Tennessee, the application of Prevathon significantly reduced fruit injury and increased yield for non-bt cotton during All Bt technologies significantly reduced injury and yield loss. Across 12 tests done in the Mid-South during 2014 and 2015, application of Prevathon increased the yield of non-bt cotton by 21.3%. Cotton yields were not affected by insecticide application in one-third of the tests, but a statistical difference in yield among treatments was observed in the other trials (Fig. 10). Treating WideStrike cotton with Prevathon significantly increased yield by an average of 8.1% across all 12 tests. On average, a significant increase in yield was not observed by spraying TwinLink, Bollgard 2, or WideStrike 3 cottons when treated with Prevathon. However, for all technologies, there were one or more locations where a significant increase in yield was observed by treating with Prevathon (Fig. 10). These data suggest there will be instances where insecticide applications for the control of bollworm may be needed in all commercially available Bt cottons. These data are also being used to develop action thresholds for the application of insecticides to control bollworm infestation in Bt cotton.

14 % trials benefited from spray % 66.7% 33.3% 12 trials across Mid-South Yield 16.7% 8.3% 8.3% 0 None Non-Bt Widestrike Twinlink Bollgard 2 Widestrike 3 Figure 10. Percentage of test sites where significant differences in yield (sprayed v. non-sprayed) within a Bt technology were detected 4.5) Termination of Insecticide Applications for Tarnished Plant Bug in Cotton (figures courtesy of Dr. Jeff Gore, MSU). Experiments were done a multiple locations in the Mid-South to evaluate when insecticides applications in flowering cotton could be terminated as judged by yield response. These replicated small-plot trials were done on Bt cotton. Insecticides were chosen at the researchers discretion but was typically Acephate ( 0.75 Lb AI/A) or Transform WG ( 1.5 Oz/A). Weekly insecticide applications were initiated at first bloom and terminated beginning the second week of bloom until the 6 th, 7 th or 8 th week of bloom, depending upon the location. Counts of tarnished plant bugs were made weekly in all plots using a black drop cloth. Outcomes: In Tennessee, tarnished plant bug numbers in untreated plots exceeded 45 insects per 10 row feet. Indeed, infestations in untreated plots were well over the recommended threshold of 6 plant bugs per 10 row feet during the first six weeks of flowering (Fig. 11). A significant increase in yield was not observed when insecticide applications were made after the fourth week of flowering (Fig. 12). However, when averaged across all eight locations, a yield increase was observed for application made up to the sixth week of flowering (Fig. 13). These data suggest that management of plant bug infestations may be needed during the first six weeks of bloom, but applications made during the first four weeks of bloom are most critical for protecting yield potential.

15 Figure 11. Tarnished plant bug numbers at Tennessee location during the first six weeks of flowering when weekly insecticides applications were terminated the 2 nd, 3 rd, 4 th, 5 th, 6 th, 7 th and 8 th week of bloom. UTC = untreated check, SLC = season long control. Figure 12. Seed-cotton yields at the Tennessee location when weekly insecticides applications were terminated the 2nd, 3rd, 4th, 5th, 6th, 7th and 8th week of bloom. UTC = untreated check, SLC = season long control.

16 Figure 13. Seed-cotton yields averaged across all locations when weekly insecticides applications were terminated the 2nd, 3rd, 4th, 5th, 6th, 7th and 8th week of bloom. UTC = untreated check, SLC = season long control.