DISCIPLINE: Arthropod Management

AUTHORS: Megha N. Parajulee (Corresponding Author)

Texas Agricultural Experiment Station

1102 East FM 1294

Lubbock, TX 79403-9803

Phone: (806) 746-6101

Fax: (806) 746-6528

Email: m-parajulee@tamu.edu

Apurba K. Barman

Texas Agricultural Experiment Station

1102 East FM 1294

Lubbock, TX 79403-9803

Phone: (806) 746-6101

Fax: (806) 746-6528

Acknowledgements: This research was supported by Cotton Incorporated/Texas State Support Committee, Plains Cotton Growers, Inc., and Texas Agricultural Experiment Station Hatch Project 8810. Stanley Carroll, Ram Shrestha, Chen Chen, and Bryan Theall provided the technical support.

Abstract

A field study was conducted at the Texas Agricultural Experiment Station farm located near Lubbock to quantify the compensation ability of cotton to Lygus bug induced fruit loss at two phenological stages: three weeks pre-flower and the first three weeks of flowering. Experiments were designed to achieve different levels of pre-flower square loss and the loss of fruiting structures (squares and bolls) during early flowering by augmenting natural populations of Lygus bugs with laboratory reared nymphs. Four treatments each were utilized for pre-flower and during-flower stages: 1) 3 bugs per plant augmented (3PP), 2) 1 bug per plant augmented (1PP), 3) naturally occurring background density (NC), and 4) 0 bug achieved through insecticide spray applications (SC). Artificial infestations of Lygus bugs released at two different crop growth stages caused variable fruit loss. In 2005, percent fruit loss due to the highest infestation level of bugs (3 per plant) in the pre-bloom and early-bloom studies was 48 and 36%, respectively. Pre-bloom fruit loss was not compensated through yield, where there was a major loss in first position bolls in the bug-release treatments. In the pre-flower Lygus release test, reduction in yield was primarily due to plant’s inability to compensate lost first fruiting positions because the first positions contributed up to 80% of the total yield. A compensatory response of cotton was observed when Lygus bug treatments were established during the early-bloom stage. Particularly, loss due to Lygus released at one per plant was fully recovered and there was no significant difference in yield compared with sprayed and natural control treatments. In a similar Lygus study conducted in 2006, 30 and 34% fruit loss were observed when 3 bugs per plant were released during pre-bloom and early-bloom stages of cotton, respectively. Lower lint yields were recorded in all treatments during 2006 due to harsh drought conditions during the mid- to late portions of the growing season. No significant yield difference was observed across treatments in 2006. Nevertheless, second and third fruiting positions over compensated the yield loss caused by Lygus augmentation during the early bloom stage. With an additional year of ongoing study in 2007, these data should produce some useful management recommendations for cotton producers and consultants in making a decision related to what extent the crop can tolerate a Lygus infestation without a significant economic loss.

Key Words: compensation, cotton, crop maturity, lint yield, Lygus

Introduction

A three-year study evaluating the compensation capacity of cotton grown in the southern High Plains was completed utilizing PM 2326RR (Baugh et al. 2003). Results using manual square removal treatments ranging from 0-100% indicated irrigated cotton could lose 100% of first position squares with no loss in yield or fiber quality. Dryland cotton could lose 50% with no significant impact on yield or fiber quality. Maturity was impacted under the most severe treatments. Even the removal of 100% of all squares prior to bloom did not reduce yield but did cause a substantial delay in maturity. The authors concluded that irrigated cotton could tolerate a 40% loss without any impact whatsoever. The current recommended pre-flower management approach places an emphasis on early square protection with a cumulative retention level of 75% after three weeks. Clearly this is higher than what the compensation study indicated. Additionally we use 25-30 fleahoppers per 100 plants or one Lygus per 3 feet of row (about 8 per 100 plants) as an indicator of the need to treat. If cotton plants can truly compensate for loss levels exceeding 25% during the first three weeks of squaring, then action thresholds will need to be elevated. It should be noted that as growing seasons become shorter, such as in the northern High Plains of Texas, the protection of 1^st position fruit become more important due to less time for the plants to compensate losses.

The question arises, whether the manual removal or mechanical injury to the fruiting bodies of cotton is similar to injury by plant bugs. Study on this aspect has been investigated by different workers (Pitman et al. 2000, Herbert and Abaye 1999, Mann et al. 1997, Phelps et al. 1997, Montez and Goodell 1994) and they inferred that there was a difference in plant response to square loss when injured by insect versus manually removed. Plant responses differ to feeding injury because of time, duration and feeding nature of insects, where insects secrete enzymes or other chemicals besides feeding on plant parts (Sadras 1995).

The present research, conducted during 2005-2006 at the Texas Agricultural Experiment Station located near Lubbock, was adopted from the concept of previous work on the plant’s ability to compensate manual removal of pre-flower squares. The objective of this research was to quantify the cotton plant response in terms of yield compensation resulting from the bug-induced injury. Specifically, this study was designed to quantify the compensation ability of cotton after Lygus induced pre-flower and early flower square loss in a limited irrigation production regime.

Materials and Methods

Plant bugs. A Lygus bug colony was maintained in the Cotton Entomology Laboratory at the Texas Agricultural Experiment Station (Lubbock) from a nucleus culture received from the USDA-ARS laboratory at Phoenix, AZ. In order to minimize insect movement away from release sites, only second or third instars were used for field releases. In some instances, local field-collected Lygus nymphs were added to the released lots.

Cultivar. The cotton cultivar used in the study was PM 2326RR in 2005 and ST 4554B2R in 2006, mid-season stripper varieties well adapted for the Texas High Plains with consistent yield and good fiber package.

Experimental procedure. A field study was conducted at the Texas Agricultural Experiment Station farm located near Lubbock to quantify the compensation ability of cotton to Lygus bug induced fruit loss at two phenological stages: three weeks pre-flower and the first three weeks of flowering. Experiments were designed to achieve different levels of pre-flower square loss and the loss of fruiting structures (squares and bolls) during early flowering by augmenting natural populations of Lygus bugs with laboratory reared nymphs. Two test sites were set up for pre-flower and early flower studies. Each test site was prepared for 16 experimental plots (8 rows x 75 ft long) following standard cultivation practices with furrow irrigation and supplemental fertilizer (80 lbs N/acre). Each test consisted of four treatments including 3 bugs per plant (3PP); 1 bug per plant (1PP); 0 bugs (naturally occurring background density; NC) and no bugs achieved through insecticide applications (SC). Three consecutive weekly releases were made for each study (Table 1). Each treatment was replicated four times, so there were 16 plots each for pre-bloom and early-bloom studies. An insecticide (Intruder® @ 0.6 oz/acre) was used to achieve the SC treatment which represents the sprayed control. The bug release area was restricted to 10 row ft on each of the middle two rows of any plot or about 30 to 35 plants per row whichever was appropriate. Thus, each 1PP and 3PP plot received 60-70 and 180-210 Lygus nymphs per plot per release, respectively, with about 960 to 1,120 Lygus nymphs required for a single-day release for each study.

Observations for fruit (square/boll) loss were recorded before and after each release. Plant height and number of nodes were recorded from 10 plants in each plot in order to determine differences due to treatment effects. Standard COTMAN sampling was conducted which included plant density, plant height-to-node ratios, square retention, and Nodes Above White Flower (NAWF). One week after the third release, final counts on fruit retention were recorded and all plots were sprayed with an insecticide to kill any remaining plant bugs. Frequent monitoring occurred during the remainder of the season and necessary control measures were taken to avoid damaging insect infestations in the experimental plots. Plots were harvested on 12 November (2005) and 7 November (2006), hand stripping the entire area that received Lygus release treatments or equivalent areas in control plots. Cotton samples were ginned at the Texas A&M University Agricultural Research and Extension Center, Lubbock.

Results and Discussion

Pre-Bloom Lygus Release Test

Percent fruit loss. In this experiment, cotton was exposed to four different regimes of Lygus bug pressure including two augmented populations receiving 1 bug per plant (1PP) and 3 per plant (3PP). For both years, fruit retention in “bug free” treatment (SC) was 95% first week into the bloom (Fig. 1). Overall, bug-augmented plots had higher fruit shed rate compared with SC or NC plots, with significantly higher fruit shed rate in 3PP followed by 1PP and the lowest shed rate in control plots one week after the third release in both 2005 and 2006 (Fig. 1, Table 2). In 2005, final fruit shed monitoring and spray application were deployed two weeks after the third release that resulted in 48 and 32% fruit loss in 3PP and 1PP treatments, respectively, compared with 17 and 10% fruit loss in NC and SC treatments, respectively (Fig. 1). In 2006, 3PP, 1PP, NC, and SC treatments had 30, 19, 6 and 5% fruit loss, respectively, one week after the third release when a spray application was applied to remove plant bugs from all test plots.

Harvestable bolls. If the early fruit loss is not fully compensated, one would assume that the percent fruit loss in the bug-augmented plots will reflect a corresponding reduction in harvestable bolls and final yield. In 2005, average number of harvestable bolls per plant was highest (9.2 bolls/plant) in SC plots and the lowest in 3PP (6.42 bolls/plant) (Fig. 2). There was a significant reduction in the number of harvestable bolls (2.8 bolls/plant) between the SC and 3PP treatments. However, number of harvestable bolls was similar between NC and 1PP treatments, indicating some level of compensation. In 2006, the number of harvestable bolls was similar across four treatments, suggesting a full compensation of the fruit loss.

Lint yield. In 2005, plants were unable to compensate for fruit loss caused by Lygus augmented infestations due to plant’s inability to compensate for lost first fruiting positions (Fig. 3). Total lint yield was in SC plots (992 lbs/acre) and lowest in 1PP (766 lbs/acre) and 3PP (792 lbs/acre) plots, with no significant difference between the two bug-augmented treatments. The yield difference between the SC and 3PP treatments was about 200 lbs/acre. Also, the in-season fruit loss of 48% in 3PP due to Lygus infestation (Fig. 1) resulted in 2.8 fewer harvestable bolls per plant in 3PP (Fig. 2). This reduced number of bolls (2.8 per plant) translated into a final lint yield reduction of about 200 lbs/acre (Fig. 3). Further, the average yield from first position bolls was reduced by 202 lbs/acre in 3PP as compared to the SC treatment while the second position contribution was similar across four treatments. These data suggest that plants were unable to compensate for the bug infestation damage under the prevailing condition. In 2006, total lint yield was similar across the four treatments, suggesting the plant’s ability to compensate early square loss up to 25%. Nevertheless, the environmental stress (high heat and drought) prevented cotton from realizing its full yield potential in SC plots in 2006 and it appeared that SC and 3PP were similar in yield with 25% additional fruit loss in 3PP. We suspect that the results could have been somewhat different had the 2006 summer been a more typical year. For example, percent fruit loss was about 5% in the first week into bloom in the control plots, with 20 and 30%, respectively, in 1PP and 3PP plots (Fig. 4), but the pre-harvest fruit loss increased to 17 and 23% for SC and NC plots, respectively, while the pre-harvest fruit loss in 1PP and 3PP were similar to their respective in-season losses. These data clearly indicate that the perceived “full compensation” in bug-augmented treatments in 2006 may be spurious. In fact, it is safe to say that SC and NC plots underwent physiological shedding of more fruit and “equalized” their fruit load with 1PP and 3PP treatments in response to the environmental stress. We hope to gain more insight in this area of research in 2007.

Percent lint yield from 2^nd and 3^rd position bolls. Cotton produces fruit at several positions in a single nodal fruiting branch. In our study, we found the maximum to be a 4^th position on a fruiting branch arising from a mainstem node, but most of the plants produced 2^nd and 3^rd position fruits. This is a strategy of plants to compensate their yield loss due to limiting factors. The results showed that percent lint yield from 2^nd and 3^rd position bolls contributed about 15% of the total yield, but a slightly higher value was observed in treatment 3PP (20%) which is not significantly different from the other treatments. This result leads us to conclude that a Lygus infestation did not have any effect in evoking more secondary fruiting positions and so the yield compensation. About 85% of the total yield came from first fruiting positions and yield losses due to Lygus were mainly a result of the loss of first position harvestable bolls per plant.

Early-Bloom Lygus Release Test

Percent fruit loss. In 2005, final fruit shed monitoring and spray application were deployed one week after the third release that resulted in 10, 24, 26, 36% fruit loss in SC, NC, 1PP and 3PP treatments, respectively (Fig. 5). In 2006, SC, NC, 1PP, and 3PP treatments had 15, 28, 32, 35% fruit loss one week after the third release when a spray application was applied to remove plant bugs from all test plots. As stated earlier, mid-summer in 2006 was hot and dry, resulting in 15% fruit shed in SC plots as opposed to 10% fruit loss in SC plots in 2005.

Harvestable bolls. Number of harvestable bolls per plant was similar across treatments in both 2005 and 2006 (Fig. 6). However, numerically there were 2.2 fewer bolls in 3PP compared with that in SC plots in 2005 although the differences were not statistically different. These data indicate that plants were able to either partially or fully compensate the fruit loss caused by bug augmentation treatments in the early bloom stage.

Lint yield. Lint yields from the Lygus early-bloom experiment showed identical trends between the two years, with bug-augmented treatments (3PP in 2005 and 1PP & 3PP in 2006) resulting in lower yields compared with control treatments (Fig. 7). An 85 lb/acre difference in control versus bug-augmented treatments in 2006 was not statistically significant. However, the 2.2 boll deficit in 3PP compared with that in SC in 2005 (Fig. 6) is reflected in a 127 lb/acre deficit in total yield in the 3PP treatment which was significant (Fig. 7). These data show that the plant was unable to fully compensate the 127 lbs/acre lint loss due to insect-induced fruit loss when 3 Lygus bugs per plant were introduced during the first three weeks of flowering. Further examining the lint yield by fruiting positions, the yield contribution of first position bolls in SC treatment was about 73% whereas the first position contribution in 3PP and 1PP treatments were 59 and 69%, respectively. Therefore, it is evident from this observation that there existed much higher contributions of 2^nd and 3^rd position bolls in total lint yield in insect-augmented treatments compared with that in the SC treatment. This result also leads to another question regarding contribution of first position bolls with respect to time of Lygus infestation. At what stage can a cotton plant produce more secondary position bolls under bug pressure? As we have seen in the pre-bloom Lygus release experiment, all treatments had more than an 80% contribution from first position bolls towards total lint yield (Fig. 3). In the early-bloom release, the contribution of first position bolls to total lint yield ranged from 79 (SC) to 59% (3PP) (Fig. 7).

Lint yield from 2^nd and 3^rd position bolls. The contribution of 2^nd and 3^rd position bolls towards total lint yield ranged from 40% in 3PP to 27% in SC in 2005, whereas it ranged from 10% in 3PP to 7% in SC in 2006. These data clearly indicate that the 2006 season was environmentally very harsh for the growth and development of lateral branches and second/third fruiting nodes. Nevertheless, it is apparent that the compensatory fruit production in 3PP was through horizontal contribution by as much as 13% in 2005 and 3% in 2006. In 2005, 13% over-compensation was not sufficient to fully compensate the total insect-induced fruit loss in this treatment. In 2006, plants were perceived to have the loss fully compensated while in fact the environmental stress did not allow plants to realize the full yield potential. As speculated in the Pre-bloom study above, we suspect that the compensatory response could have been different in the 2006 Early-bloom study had the 2006 summer been not so abnormally hot and dry. For example, percent fruit loss was about 15% the fourth week into bloom in SC plots with 28, 32, and 35%, respectively, in NC, 1PP and 3PP plots (Fig. 8), but the pre-harvest fruit loss in SC plots increased to 29% while the pre-harvest fruit loss in NC, 1PP, and 3PP were similar to their respective in-season losses (Fig. 8).

References

Herbert, D.A., and A.O. Abaye. 1999. Compensation from systemic square removal by Virginia cotton, pp. 968-971. In Proceedings, Beltwide Cotton Conf., National Cotton Council, Memphis, TN.

Baugh, B., J. Leser, R. Boman, and T. Doederlein. 2003. Plant response to different levels of pre-bloom square removal and its relevance to plant bug management, pp. 1131-1137. In Proceedings, Beltwide Cotton Conf., National Cotton Council, Memphis, TN.

Mann, J.E., S.G. Turnipseed, M.J. Sullivan, P.H. Adler, J.A. Durant, and O.L. May. 1997. Effect of early-season loss of flower buds on yield, quality, and maturity of cotton in South Carolina. J. Econ. Entomol. 90:1324-1331.

Montez, G.H., and P.B. Goodell. 1994. Yield compensation in cotton with early season square loss, pp. 916-919. In Proceedings, Beltwide Cotton Conf., National Cotton Council, Memphis, TN.

Phelps, J.B., J.T. Ruscoe, and W.H. McCarty. 1997. Cotton development following early square removal, pp. 1412-1413. In Proceedings, Beltwide Cotton Conf., National Cotton Council, Memphis, TN.

Pitman, V.O., A.O. Abaye, D.A. Herbert, and D. Oosterhuis. 2000. Comparison of cotton to square and boll removal with different varieties and planting dates, p. 955. In Proceedings, Beltwide Cotton Conf., National Cotton Council, Memphis, TN.

Sadras, V.O. 1995. Compensatory growth in cotton after loss of reproductive organs. Field Crops Res. 40:1-18.

Table 1. Calendar of field research activities during compensation studies, 2005 and 2006, Lubbock, TX.

Year	Activities	Pre-bloom	Early-bloom
2005	Plant mapping	6, 13, 20, and 27 Jul, 6 Aug	16 and 23 Jul, 2 and 11 Aug
	Insect release (1PP and 3PP)	7, 14, and 21 Jul	16 and 23 July, 3 Aug
	Insecticide spraying	28 Jun, 5, 22, and 29 Jul	6, 13, 20, and 28 Jul, 4 Aug
2006	Plant mapping	28 Jun, 5, 13, and 22 Jul	20 and 27 Jul, 7 and 14 Aug
	Insect release (1PP and 3PP)	29 Jun, 6 and 13 Jul	24 Jul, 1 and 7 Aug
	Insecticide spraying	23 Jun, 3, 7, and 23 Jul	27 and 31 Jun, 7 and 16 Jul

Table 2. Fruit retention and loss in each nodal position after three releases of Lygus bugs, 2006, Lubbock, TX.

Pre-bloom study: 2006		Treatment	All positions	First position	Second position	Third position
	Fruit intact	SC	12.30a	7.62a (61.9)	4.20a (34.1)	0.45a (3.6)
		NC	12.75a	6.62a (51.92)	4.65a (36.47)	0.47a (3.6)
		1PP	9.97b	6.40b (64.19)	2.85b (28.6)	0.72a (7.2)
		3PP	8.02b	5.32c (66.33)	2.35b (29.3)	0.35a (4.3)
	Fruit loss	SC	0.7c (5.3)	0.57c (4.4)	0.1b (0.8)	0.02a (0.1)
		NC	0.8c (6.0)	0.67c (5.0)	0.12b (1.0)	0a (0.0)
		1PP	2.42b (19.5)	1.37b (11.0)	1.02a (8.0)	0.02a (0.5)
		3PP	3.52a (30.6)	2.35a (20.4)	1.07a (9.3)	0.1a (0.9)
Early-bloom study: 2006	Fruit intact	SC	13.02a	8.55 a (65.6)	2.97a (22.81)	1.20a (9.0)
		NC	11.10a	7.32 ab (65.94)	3.07a (27.65)	0.57a (5.1)
		1PP	9.52a	7.25 ab (76.15)	1.90a (19.9)	0.37a (3.8)
		3PP	10.52a	6.82b (64.82)	2.85a (27.1)	0.80a (7.6)
	Fruit loss	SC	2.17b (14.92)	0.80b (5.2)	1.1a (7.2)	0.22a (1.4)
		NC	4.3a (28.26)	2.17a (14.1)	1.82a (11.8)	0.30a (1.9)
		1PP	4.4a (31.58)	2.45a (17.3)	1.85a (13.1)	0.12a (0.8)
		3PP	5.07a (34.27)	2.87a (18.4)	1.67a (10.7)	0.45a (2.8)

Values in parentheses are percentage contribution by each nodal position (fruit intact) and percentage square/boll loss (fruit loss). Values within columns within each plant parameter (pre- or early bloom) study followed by different letters are significantly different (P<0.10).

F

igure 1. Percent fruit loss in cotton plots receiving various levels of Lygus bug releases during the pre-bloom fruiting period, Lubbock, TX, 2005 (upper panel) and 2006 (lower panel). Bars within a year and sampling date with different letters are statistically different (P <0.10); NS = no statistical differences.

2006

2005
Figure 2. Average number of harvestable bolls, missing fruits, and non-harvestable, green bolls/plant in cotton plots receiving various levels of Lygus bug releases during the pre-bloom fruiting period. Lubbock, TX, 2005 (upper panel) and 2006 (lower panel). Bars within a fruit type (open, missing, or non-open) and study year with different letters above bars are statistically different (P<0.10).

a

b

b

b

b

a

a

a

a

a

a

a

a

a

a

a Figure 3. Total lint yield and yield from 1^st-position bolls in cotton plots receiving various levels of Lygus bug releases during the pre-bloom fruiting period in 2005 (upper panel) and 2006 (lower panel), Lubbock, TX. Bars within a lint source and study year with different letters are statistically different (P<0.10).

Figure 4. Percent fruit loss during the first week into flowering and at harvest in cotton plots receiving various levels of Lygus bug releases during the pre-bloom fruiting period, Lubbock, TX, 2006. Bars within a plant mapping period with different letters are statistically different (P<0.10).

Figure 5. Percent fruit loss in cotton plots treated with various levels of Lygus bug releases during the early-bloom period, Lubbock, TX, 2005 (upper panel) and 2006 (lower panel). Bars within a year and sampling date with different letters above the bars are statistically different (P <0.10).

Figure 6. Average number of harvestable bolls, missing fruits, and non-harvestable, green bolls/plant in cotton plots receiving various levels of Lygus bug releases during the early bloom fruiting period. Lubbock, TX, 2005 (upper panel) and 2006 (lower panel). Bars within a fruit type (open, missing, or non-open) and study year with different letters above bars are statistically different (P<0.10).

Figure 7. Total lint yield and yield from 1^st-position bolls in cotton plots receiving various levels of Lygus bug releases during the early-bloom fruiting period, Lubbock, TX, 2005-2006. Bars within a lint source and study year with different letters are statistically different (P<0.10).

Figure 8. Percent fruit loss four weeks into bloom (after three consecutive releases of bugs in bug treated plots) and the pre-harvest fruit loss profile in cotton plots receiving various levels of Lygus bug releases during the early-bloom fruiting period, Lubbock, TX, 2006. Bars within a plant mapping period with different letters are statistically different (P<0.10).