BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors,nonprofit publishers, academic institutions, research libraries, and research funders in the common goal ofmaximizing access to critical research.

Comparison of Cotton Square and Boll Damage andResulting Lint and Seed Loss Caused by Verde PlantBug, Creontiades signatusAuthor(s): Michael J. Brewer , Darwin J. Anderson and J. ScottArmstrongSource: Southwestern Entomologist, 37(4):437-447. 2012.Published By: Society of Southwestern EntomologistsDOI: http://dx.doi.org/10.3958/059.037.0401URL: http://www.bioone.org/doi/full/10.3958/059.037.0401

BioOne (www.bioone.org) is a nonprofit, online aggregation of coreresearch in the biological, ecological, and environmental sciences. BioOneprovides a sustainable online platform for over 170 journals and bookspublished by nonprofit societies, associations, museums, institutions, andpresses.

Your use of this PDF, the BioOne Web site, and all posted and associatedcontent indicates your acceptance of BioOne’s Terms of Use, available atwww.bioone.org/page/terms_of_use.

Usage of BioOne content is strictly limited to personal, educational, andnon-commercial use. Commercial inquiries or rights and permissionsrequests should be directed to the individual publisher as copyright holder.

VOL. 37, NO. 4 SOUTHWESTERN ENTOMOLOGIST DEC. 2012 Comparison of Cotton Square and Boll Damage and Resulting Lint and Seed Loss Caused by Verde Plant Bug, Creontiades signatus1 Michael J. Brewer2*, Darwin J. Anderson2, and J. Scott Armstrong3

Abstract. Retention of cotton, Gossypium hirsutum L., bolls and squares (referred to as fruit retention), boll damage, and resulting lint and seed weight loss were assessed when two (2010) and three (2011) age classes of sympodial fruiting branches with different ages of squares and bolls where exposed to verde plant bug, Creontiades signatus Distant (Hemiptera: Miridae), in cages in the field. An increasing trend was observed in fruit retention from the youngest to the oldest branch age treatment for both years, and the effect was primarily determined by the age of the fruiting body. Low fruit retention (<20%) was seen in young bolls <7 days old in the first two fruiting positions from the main stem compared with >80% retention in checks not exposed to verde plant bug. In infested cages, fruit retention of young bolls was significantly less (P = 0.016) than older bolls and squares 2-3 days before bloom or younger. Similar to fruit retention results, damage of harvested bolls and lint and seed weight (using zero for abscised fruit) differed among the age of the fruiting body primarily for fruiting positions one (2010 and 2011) and two (2011) (P < 0.0001). The first two fruiting positions were the main contributors to plant productivity differences in lint and seed weight. Overall, verde plant bugs given a feeding choice reduced fruit retention in young bolls <7 days old, damaged retained bolls <11 days old, and larger bolls and young squares at least 2-3 days from bloom incurred significantly less abscission and damage. These results supported the interpretation that less fruit retention and more damage of young bolls justified a focus on protecting young bolls in a pest management program, especially during early to peak bloom when young bolls are abundant.

Introduction

Use of insecticide in cotton, Gossypium hirsutum L. (Malvaceae), has been reduced following the success of transgenic Bt (Bacillus thuringiensis Berliner) cotton and boll weevil eradication (Edge et al. 2001, Allen 2008). These advances probably have released stink bugs and plant bugs (Hemiptera: Pentatomidae and Miridae) from indirect insecticide control (e.g., Lu et al. 2010). Associated cotton boll damage caused by feeding stink bugs increased during the last two decades (Greene et al. 2001, Hopkins et al. 2009). Selected plant bugs are also known to damage bolls, and the verde plant bug, Creontiades signatus Distant (Hemiptera: Miridae), has emerged as a threat to cotton along the Gulf Coast of south Texas ________________________ 1Hemiptera: Miridae 2Texas AgriLife Research and Department of Entomology, Texas AgriLife Research and Extension Center at Corpus Christi, 10345 State Hwy 44, Corpus Christi, TX 78406 3USDA ARS, 1301 N. Western Rd., Stillwater, OK 74075 *Corresponding author: mjbrewer@ag.tamu.edu

437

438

(Armstrong et al. 2010, Brewer et al. 2012). Other mirid pests of cotton, western tarnished plant bug, Lygus hesperus Knight, and cotton fleahopper, Pseudatomoscelis seriatus (Reuter), primarily damage squares (Leigh et al. 1988, Ring et al. 1993).

In response, research on plant sensitivity to feeding by verde plant bug established insecticide use guidelines to protect bolls, similar to studies with stink bugs (e.g., Siebert et al. 2005). Armstrong et al. (2013) found that bolls <2.7 cm in diameter (equates to about <166 degree-days [15.6ºC base] or <12 days old from first day of bloom) were readily injured by feeding verde plant bugs while older bolls incurred little injury. We considered here whether this boll protection guideline based on no-choice tests with different boll ages was applicable when verde plant bug has free choice to feed on squares and bolls. The specific objective was to assess retention of bolls and squares (referred hereafter as fruit retention), boll damage, and resulting lint and seed loss when cotton branches with different ages of squares and bolls where exposed to adult verde plant bug.

Materials and Methods

Protocol for Infesting Branches and Measurements. Two (2010) and three (2011) age classes of sympodial fruiting branches were caged in the field and exposed to verde plant bug for 72 hours, along with noninfested checks. Fruit retention, boll damage, and resulting lint and seed weight were evaluated on bolls and squares along the branch from the first fruiting position nearest the main stem outward to the fourth position. FiberMax 840 RRB2 cotton (Bayer CropScience, Research Triangle Park, NC) was planted in early April on 91-m rows and 96-cm row centers in a field approximately 0.4 ha (Corpus Christi, TX). The cotton was grown using standard agronomic practices under dryland conditions, but no insecticide was used. Beginning early bloom, 1 month after planting, colored tags were looped onto cotton branches to designate selected branch age treatments in a randomized complete block of 11 to 15 (2010) and 28 to 30 (2011) replications. Replication numbers differed because of branch loss from handling and wind. The following guidelines were used to randomly choose fruiting branches: a) the branch had four fruiting positions with no abscised positions and was in good health, b) fruiting position one nearest the main stem was a bloom or would bloom the following morning (0-1 day old boll), and c) no more than one branch was chosen per plant. The cages were made from white organza fabric (26 cm in length and 12 cm in diameter) with draw strings on each end to enclose the branches.

In 2010, two branch age treatments were established by caging branches at two time periods and infesting at the same time. When branches were caged, a set was left noninfested to serve as a check. Plants were infested 7-8 days and 11-12 days post bloom of fruiting position one on 22 June. In 2011, the protocol was modified to cage all branches at the same time and infest at different times, which allowed an additional branch age treatment meeting the criteria for using branches in the test. Three branch-age treatments were established by infesting at 3-4 days (10 June), 9-10 days (16 June), and 14-15 (21 June) days post bloom of fruiting position one. Because branches for all treatments were caged at the same time, one set of cages was left noninfested to serve as a common check.

Adult verde plant bugs used for infesting were obtained from a laboratory colony reared on green beans, Phaseolus vulgaris L., and shucked maize, Zea mays L., and periodically replenished with bugs collected from several wild and

439

cultivated host plants in the Lower Rio Grande Valley (Armstrong 2010). Two days before each infestation date, the branches were sprayed with short-residual pyrethrin (0.02% by volume, United Industries, St. Louis, MO). Also, periodic inspection of the surrounding cotton showed negligible boll damage. On the designated infestation date, cages excluding the noninfested check cages were infested with 10 adult verde plant bugs. The assumption that a new bloom was expected along the branch approximately every 6 days (Ritchie et al. 2004) was used to estimate the square and bolls ages on the first four fruiting positions relative to expected bloom. Therefore, the two branch age treatments in 2010 were designated (listed from the first oldest to the fourth youngest fruiting body position) a) bolls 8 days old to squares 11 days prebloom and b) bolls 12 days old to squares 7 days prebloom. In 2011, the treatments were designated a) bolls 4 days old to squares 15 days prebloom, b) bolls 10 days old to squares 9 days prebloom, and c) bolls 15 days old to squares 4 days prebloom (Table 1). The infestations were terminated after 72 hours by application of pyrethrin as used at the beginning of the experiment.

The cages were left on the branches until the bolls were fully opened and the surrounding field was ready for harvest. The caged branches were clipped from the plants the fourth week of July (about 100 days post emergence). Fruiting positions on each branch were observed for fruit retention. Bolls at harvest were scored using a five class locule scale for damage by insects. The scale ranged from 0 representing no damage detected; 1, 2, and 3 representing a progression of damage from localized in one locule to damage in most locules; and 4 representing severe damage in all locules (Lei et al. 2003). The scale was implemented for visual inspection by equating the number of damaged locules to the same scalar. Plant productivity loss because of square and boll injury was assessed, setting yield measures to zero when a fruiting position abscised. Seed cotton was hand-harvested. In 2010, lint and seed were separated by hand and weighed. In 2011, lint and seed were separated by using a roller gin (Porter Morrison and Son, Athens, TX) and weighed.

Data Analysis. Mean percentage of fruit retention, damage score of harvested bolls, and lint and seed weight were calculated by fruiting position for the Table 1. Estimated Age of Fruiting Bodies (Squares and Bolls) Relative to when First Bloom Occurreda on Two (2010) and Three (2011) Age Classes of Sympodial Fruiting Cotton Branches Exposed to Verde Plant Bug for 72 Hours in a Field, Corpus Christi, TX

Year: Branch age Position 1 Position 2 Position 3 Position 4 2010: Bolls 8 days oldsquares 11 d prebloom

Bolls 7-8 days old Bolls 1-2 days old Squares within 4-

5 days of bloom Squares within 10-11 days of bloom

2010: Bolls 12 days oldsquares 7 days prebloom

Bolls 11-12 days old Bolls 5-6 days old Squares within 0-

1 days of bloom Squares within 6-7

days of bloom 2011: Bolls 4 days oldsquares 15 days prebloom

Bolls 3-4 days old

Squares within 2-3 days of bloom

Squares within 8-9 days of bloom

Squares within 14-15 days of bloom

2011: Bolls 10 days oldsquares 9 days prebloom

Bolls 9-10 days old Bolls 3-4 days old Squares within 2-

3 days of bloom Squares within 8-9

days of bloom 2011: Bolls 15 days oldsquares 4 days prebloom

Bolls 14-15 days old Bolls 8-9 days old Bolls 2-3 days

old Squares within 3-4

days of bloom aA new bloom is expected approximately every 6 days (Ritchie et al. 2004).

branch age treatments and noninfested check. The branch means for percentage of fruit retention and boll damage score, and the total lint and seed weight for the branch, also were calculated. No differences in the two sets of noninfested checks caged in 2010 were detected; therefore, the check data were aggregated.

An analysis of variance conforming to a randomized block design was used to detect differences in overall means and totals among the branch ages and checks for each year. Analyses also were done separately by fruiting position to assess relative measurement differences among different ages of squares and bolls along the branch. Fruit retention data were transformed by the arcsine square root of the proportion before analysis. Damage was scored only for bolls retained until harvest, which limited analysis to the first position in 2010 (n = 37 across the three treatments) because of little fruit retention of the other positions. A branch mean was not calculated. Sufficient open bolls were available to analyze damage scores in 2011 (n = 101, 73, and 19 across the four treatments for positions one, two, and three, respectively). When the main factor was significant (P < 0.05), contrast statements were used to test measurements for the whole branch and by fruiting position for differences of the two infested branch ages of 2010, and for linear and quadratic trends across the three infested branch ages of 2011 (Littell et al. 1991). Average values from the infested branch ages also were compared to the noninfested check using contrasts for both years.

Results and Discussion

Fruit Retention. Fruit retention averaged across the fruiting branch positions varied among treatments in 2010 (F = 4.64; df = 2, 44; P = 0.015) and 2011 (F = 8.02; df = 3, 84; P < 0.0001). In 2010, there were no detectable differences in mean fruit retention between the two infested branch ages (P = 0.49), but average fruit retention of infested branch ages was less than that of the noninfested check (check contrast: F = 8.92; df = 1, 44; P = 0.005) (Fig. 1A). The greatest fruiting body loss occurred from bolls 1 to 6 days old on position two of the two infested branch ages compared to the check even though retention was <50% on position two (check contrast: F = 10.86; df = 1, 44; P = 0.002) (Fig. 1A, Table 1). Most bolls 7 days old on position one were retained (>90%), and fruit retention did not differ among branch ages (P = 0.53) (Fig. 1A).

In 2011, mean fruit retention linearly increased from the bolls 4 days old to squares 15 days pre-bloom, to the bolls 15 days old to squares 4 days prebloom (linear contrast: F = 5.30; df = 1, 84; P = 0.024), and the check had more fruit retention than the average fruit retention of infested treatments (check contrast: F = 18.79; df = 1, 84; P < 0.0001) (Fig. 1B). The least fruit retention relative to the check was on bolls 3-4 days old on position one (bolls 4 days old to squares 15 days prebloom) and two (bolls 10 days old to squares 9 days prebloom) (check contrast: F > 6.06; df = 1, 84; P < 0.016) (Fig. 1B). In contrast, fruit retention for bolls 8 days old and squares 2-3 days before bloom was about as great as in the check (at least 75%) (Fig. 1B, Table 1). The fourth position for both years was not considered in this and other measures because of low retention (<20%) in all treatments and the check. Overall, an increasing trend was observed in fruit retention from the youngest to the oldest branch age for both years, and differences in fruiting retention were determined primarily by the age of the fruiting body on positions one and two. Scarce fruit retention was seen mostly in young bolls <7 days old and not older bolls and young squares at least 2-3 days before bloom.

440

Fig. 1. Percentage of fruit retention when two (A: 2010) and three (B: 2011) age classes of cotton branches were caged and exposed to 10 verde plant bugs for 72 hours, along with noninfested checks. Data were taken on the fruiting positions along the branch. Data for positions three and four in 2010 and four in 2011 were not used to calculate the branch mean because of poor retention. Lines above bars are SEMs.

Boll Damage. Damage of harvested bolls differed among the age classes of infested branches and noninfested checks for fruiting position one in 2010 (F = 37.14; df = 2, 31; P < 0.0001) (Fig. 2A). Bolls 11-12 days old were damaged less than bolls 7-8 days old in position one (F = 5.08; df = 1, 31; P = 0.032), and average boll damage of infested treatments was greater than that of the noninfested check (check contrast: F = 72.62; df = 1, 31; P < 0.0001).

441

Fig. 2. Boll damage score (0 to 4 scale) when two (A: 2010) and three (B: 2011) age classes of cotton branches were caged and exposed to 10 verde plant bugs for 72 hours, along with noninfested checks. Data were taken on the fruiting positions along the branch. Data for positions two to four in 2010 and four in 2011 were not used to calculate the branch mean because of poor boll retention. deleted from analysis. Lines above bars are SEMs.

In 2011, differences were found in damage scores between fruiting positions one (F = 37.68; df = 3, 69; P < 0.0001) and two (F = 16.61; df = 3, 41; P < 0.0001).

442

The first two fruiting positions were the main contributors to branch mean differences across treatments (F = 12.69; df = 3, 33; P < 0.0001) (Fig. 2B). Non-significant differences for position three (P = 0.13) probably reflected the small number of bolls available for analysis. Bolls 10 days old were damaged more than bolls 14-15 days old in position one (linear and quadratic contrasts: F > 17.35; df = 1, 69; P < 0.0001), and bolls 3-4 days old and 8-9 days old incurred more damage than squares within 2-3 days of bloom in position two (linear contrast: F = 16.09; df = 1, 41; P = 0.0002) (Fig. 2B, Table 1). The noninfested check had less boll damage than the average damage from infested treatments for fruiting position one (check contrast: F = 47.72; df = 1, 69; P < 0.0001) and two (check contrast: F = 26.70; df = 1, 41; P < 0.0001) (Fig. 2B). These data were consistent with previous findings that most bolls >12 days old (equates to bolls >2.7 cm diameter) were mostly not damaged in a no-choice test (Armstrong et al. 2013). In the feeding choice situation of the test reported here, bolls >10 days old and young squares at least 2-3 days before bloom were less damaged than were younger bolls.

Lint and Seed Weight. Mean lint and seed weights followed similar trends across the fruiting positions, and varied among the age classes of infested branches and noninfested checks in 2010 (F > 7.37; df = 2, 44; P < 0.002) and 2011 (F > 15.05; df = 3, 84; P < 0.0001). In 2010, lint and seed weights did not differ significantly (P > 0.33) between the two infested branch ages. But average total lint and seed weights of the infested branches were less than weights of the noninfested check (check contrast: F > 14.01; df = 1, 44; P < 0.006). Lint and seed weights in the infested branch ages were less than in the noninfested check for fruiting positions one and two (check contrast: F > 4.61; df = 1, 44; P < 0.04) (Fig. 3). Lint and seed weights for all treatments were less in position two because of much fruit abscission (Fig. 1A).

In 2011, lint and seed weights increased from the bolls 4 days old to squares 15 days prebloom, to the bolls 15 days old to squares 4 days prebloom (linear contrast: F > 4.29; df = 1, 84; P < 0.04), and the noninfested check had the greatest total lint and seed weights (check contrast: F > 41.33; df = 1, 84; P < 0.0001) (Fig. 4). Lint and seed weights increased as age of the fruiting bodies increased from 3-4 day old bolls to 14-15 day old bolls on position one (linear contrast: F > 29.40; df = 1, 84; P < 0.0001). On position two, lint and seed weight loss was greatest for bolls 3-4 days old, while squares 2-3 days before bloom and bolls 8-9 days old were less affected by feeding verde plant bugs (quadratic contrast: F > 15.09; df = 1, 84; P < 0.0002) (Fig. 4, Table 1). Lint and seed weights in the infested branch ages were less on average than those of the noninfested check for positions one and two (F > 22.31; df = 1, 84; P < 0.0001) and to a lesser degree for position three (F > 5.58; df = 1, 84; P < 0.025) (Fig. 4).

Overall, verde plant bugs when given a feeding choice reduced fruit retention in young bolls <7 days old, damaged retained bolls <11 days old, and larger bolls and young squares at least 2-3 days before bloom incurred significantly less abscission and damage. Fruiting body abscission was the main contributor to plant productivity differences in lint and seed weights. These results were consistent with and added to the previous finding that mature bolls 12 days or older incurred negligible damage from verde plant bug feeding in a no-choice test (Armstrong et al. 2013). The tendency for more damaged younger bolls, including significant abscission, may have been magnified by a feeding preference for younger bolls in the feeding choice experiment. Three- to 6-day-old bolls had the least retention in the first two fruiting positions (10-60%) (Fig. 1). In contrast, other mirids, western

443

tarnished plant bug and cotton fleahopper, are very damaging to squares and meristematic growth (Leigh et al. 1988, Ring et al. 1993). Fruiting body age was the primary determinant of injury potential, as found with other boll feeders (Allen et al. 2009).

Fig. 3. Lint (A) and seed (B) weights for the 2010 test when two age classes of cotton branches were caged and exposed to 10 verde plant bugs for 72 hours, along with noninfested checks. Data were taken on the fruiting positions along the branch, and weights were set to zero for abscised bolls. Data for positions three and four were not used to calculate the branch total because of poor retention. Lines above bars are SEMs.

444

Fig. 4. Lint (A) and seed (B) weights for the 2011 test when three age classes of cotton branches were caged and exposed to 10 verde plant bugs for 72 hours, along with noninfested checks. Data were taken on the fruiting positions along the branch, and weights were set to zero for abscised bolls. Data for position four was not used to calculate the branch total because of poor retention. Lines above bars are SEMs.

Currently, verde plant bug infests cotton around peak bloom in the coastal cotton-growing region of south Texas, and management programs have focused on boll protection because maturing bolls are abundant and will be major contributors

445

to yield (Ritchie et al. 2004). Based on results from this study, losses from feeding verde plant bugs are expected to be greatest when verde plant bug infests cotton from early to peak bloom (when young bolls are abundant). Infestations beginning later in bloom (when bolls are older and harvestable) and any infestations detected before first bloom (when young squares are dominant) are expected to be of less concern. These results supported the interpretation that less fruit retention and much damage of young bolls, especially during early to peak bloom when young bolls are abundant, justified a focus on protecting young bolls as recommended by Armstrong et al. (2013).

Acknowledgment

We thank Jonathan Martinez and Charlene Farias for providing field and other technical support. This work was partially supported by a Texas State Support Committee, Cotton Incorporated, award (11-845TX) to MJB and JSA.

References Cited

Allen, C. T. 2008. Boll weevil eradication: an areawide pest management effort, pp. 467-559. In O. Koul, G. Cuperus, and N. Elliott [eds.], Areawide Pest Management. CAB International, Wallingford, UK.

Allen, K. C., R. G. Luttrell, and C. D. Parker, Jr. 2009. Influence of within-season densities of Heliothines and tarnished plant bugs on variability in end-of-season cotton yield mapping. J. Cotton Sci. 13: 11-22.

Armstrong, J. S. 2010. Rearing Creontiades signatus (Heteroptera: Miridae) on green bean pods and pea plants: a comparative mass rearing method. J. Entomol. Sci. 45: 193-196.

Armstrong, J. S., M. J. Brewer, R. D. Parker, and J. J. Adamczyk, Jr. 2013. Verde plant bug (Hemiptera: Miridae) feeding injury to cotton bolls characterized by boll age, size, and damage ratings. J. Econ. Entomol. (in press).

Armstrong, J. S., R. J. Coleman, and B. L. Duggan. 2010. Actual and simulated injury of Creontiades signatus Distant (Hemiptera: Miridae) feeding on cotton bolls. J. Entomol. Sci. 45: 170-177.

Brewer, M. J., J. A. Armstrong, E. G. Medrano, and J. F. Esquivel. 2012. Association of verde plant bug, Creontiades signatus (Hemiptera: Miridae), with cotton boll rot. J. Cotton Sci. 16: 1-8.

Edge, J. M, J. H. Benedict, J. P. Carroll, and H. K. Reding. 2001. Bollgard cotton: an assessment of global economic, environmental, and social benefits. J. Cotton Sci. 5: 121-136.

Greene, J. K., S. G. Turnipseed, M. J. Sullivan, and O. L. May. 2001. Treatment thresholds for stink bugs (Hemiptera: Pentatomidae) in cotton. J. Econ. Entomol. 94: 403-409.

Hopkins, B. W., A. E. Knutson, J. S. Bernal, M. F. Traecy, and C. W. Smith. 2009. Species composition, damage potential, and insecticide susceptibility of stink bugs in cotton in the Lower Gulf Coast region of Texas. Southwest. Entomol. 35: 19-32.

Lei, T., M. Khan, and L. Wilson. 2003. Boll damage by sucking pests: an emerging threat, but what do we know about it?, pp. 1338-1344. In Proceedings, World Cotton Research Conf. 3, Cape Town, Rep. South Africa.

446

Leigh, T. F., T. A. Kerby, and P. F. Wynholds. 1988. Cotton square damage by the plant bug, Lygus hesperus (Hemiptera: Heteroptera: Miridae), and abscission rates. J. Econ. Entomol. 81: 1328-1337.

Littell, R. C., R. J. Freund, and P. C. Spector. 1991. SAS System for Linear Models, 3rd ed. SAS Institute Inc, Cary, NC.

Lu, Y., K. Wu, Y. Jiang, B. Xia, P. Li, H. Feng, K.A.G. Wyckhuys, and Y. Guo. 2010. Mirid bug outbreaks in multiple crops correlated with wide-scale adoption of Bt cotton in China. Science 328: 1151-1154.

Ring, D. R., J. H. Benedict, M. L. Walmsley, and M. F. Treacy. 1993. Cotton yield response to cotton fleahopper (Hemiptera: Miridae) infestations on the Lower Gulf Coast of Texas. J. Econ. Entomol. 86: 1811-1819.

Ritchie, G. L., C. W. Bednarz, P. H. Jost, and S. M. Brown. 2004. Cotton Growth and Development. Bulletin 1252, U. Georgia Coop. Ext. Serv.

Siebert, M. W., B. R. Leonard, R. H. Gable, and L. R. LaMotte. 2005. Cotton boll age influences feeding preference by brown stink bug (Heteroptera: Pentatomidae). J Econ. Entomol. 98: 82-87.

447

448