Effect of Infestation of Alabama argillacea ( Lepidoptera : Noctuidae ) at Different Ages of Cotton Plants in Yield

The cotton leafworm is an important defoliating pest of cotton in Brazil, and occurs in all regions where the crop is cultivated. The purpose of this study was to evaluate the effect of larval densities of A. argillacea after different infestation periods on the yield components of four cotton cultivars. The experiment was carried out in Pindorama, SP, Brazil, in the 2008/2009 growing season. The cultivars IAC-25, DeltaOPAL, Fibermax 966 and Fibermax 993 were artificially infested with A. argillacea larvae at three times (30, 60 and 90 DAE) and four densities (0, 2, 4, and 6 larvae per plant). The average boll weight (g), fiber percentage (%), 100-seed weight (g) and yield (kg∙ha−1) were evaluated. With increasing infestation density of A. argillacea, the cotton yield of the cultivars decreased. The presence of larvae significantly reduced the weight of 100 seeds of cultivar Fibermax 966. Initial infestations reduced the boll weight of IAC-25 and DeltaOPAL, while cultivars Fibermax 966 and Fibermax 993 were most affected by late infestations. Early infestation compromised fiber percentage of cultivar DeltaOPAL and late infestations were most harmful to cultivar Fibermax 966. Early A. argillacea infestation reduced the yield of DeltaOPAL, while infestations 60 DAE caused the most damage to IAC-25, and the other cultivars were not affected by the moment of infestation. Corresponding author. M. D. Michelotto et al.


Introduction
Cotton industry is a socially and economic relevant activity in the agricultural scenario of Brazil.The centralwest region is the country's leading producer, with about 62.7% of the cotton production in Brazil in the 2012/ 2013 growing season [1].
The cotton leafworm, Alabama argillacea (Hübner, 1818) (Lepidoptera: Noctuidae), is the important defoliating pest of the crop, causing great yield losses if not controlled properly [2] [3].This pest causes great damage by the high intensity of plant defoliation to meet the larval food demand.The pest can occur from the initial crop development stages to maturation, in all cotton-cultivating regions of Brazil [2] [4] and often reaches the threshold level that makes control measures necessary [5].According to the researchers [6] [7], the plants are not resistant to high losses of leaf area in the first 45 days of development.
The cotton leafworm consumes a cotton leaf area of 88 cm 2 of all instars [8], while other author [9] found that at a constant temperature of 27.5˚C, the average larval consumption was 117.95 cm 2 , and that the last instar larvae consumed approximately 73% of the total.
Development of efficient strategies to control A. argillacea requires the understanding of its biological relationship with the host plant.Therefore, an important component is to understand the host's susceptibility to the pest [10].
In view of the high destructive capacity of cotton leafworm and lack of information about the impact of infestation at different times and levels on novel cotton cultivars, the purpose of this study was to evaluate the effect of different larval densities after different infestation periods on the yield components of four cotton cultivars.

Material and Methods
The experiment was conducted on an experimental field and in the laboratory of Entomology of the Agência Paulista de Tecnologia dos Agronegócios (APTA), Polo Regional do Centro Norte, in Pindorama, São Paulo, Brazil, (21˚11'9''S and 48˚4'25''W).The experiment was arranged in a randomized factorial block design with 48 treatments (4 cultivars × 4 larval densities per plant × 3 infestation periods), with 4 replications.

Rearing A. argillacea
Larvae of A. argillacea were raised artificially by the methodology proposed by workers [11].Pupae of A. argillacea were collected from cotton plantations of the APTA in Pindorama, SP, sexed and transferred to Petri dishes (diameter 9 cm, height 1.5 cm), and placed in PVC cages (height 21.5 cm, diameter 20 cm) lined with printing paper as substrate for oviposition until adult emergence.The top of the cage was covered with voile and the base set on a plastic dish lined with paper towel.Sponges (thickness 0.5 cm, diameter 5 cm) were placed in the cage, soaked with a 50% honey solution, to feed the adults.During oviposition, the paper and voile containing eggs were transferred daily to cages similar to those used for the adults.After two days, this egg-containing material was distributed on cotton leaves, with the stem inserted in water, in glasses sealed with a cotton wool ball.The newly-hatched larvae were fed with leaves of cotton cultivar Coodetec 407, collected in a greenhouse and washed in tap water and in a 2% sodium hypochlorite solution for 2 min, followed by four washes in tap water.

Crop Management
The soil was tilled and limed as recommended for the crop.The cotton cultivars DeltaOPAL, IAC-20, Fibermax 966, and Fibermax 993 were sown mechanically on 15 December 2008, for a final germination of 12 plants per meter.Each plot consisted of three 4-m long rows spaced 0.9 meters apart.Five plants of each cultivar per replication were infested 30, 60 and 90 DAE with 0, 2, 4, and 6 third-instar larvae (length 15 mm, mass 60 ± 10 mg).After infestation, the plants were protected by rectangular cages consisting of metal frames (0.8 × 1.2 × 1.0 m) (W × H × D), completely covered by voile, corresponding to the size of each plot.Insecticide was not needed for pest control.To prevent excessive growth of cotton plants and to facilitate the management practices, a plant growth regulator (mepiquat chloride) was applied twice (50 and 70 DAE, respectively, at 300 and 500 mlp.c.ha −1 ) to obtain a final plant height of 1.10 -1.20 m, as currently recommended for mechanical harvesting [12].

Parameters Evaluated
At the end of the cycle, when all bolls had opened, the plants in the cages were harvested and the following parameters assessed: average boll weight (g), lint percentage (%), 100-seed weight (g), and average yield per cultivar (kg•ha −1 ).To determine the 100-seed weight and fiber content, samples of all plots were sent to the laboratory for Fiber Technology of the Grain and Fiber Center of the Agronomic Institute of Campinas (IAC).

Data Analysis
Data were subjected to analysis of variance and treatment means were compared by Tukey's test at 5% probability.The data were not transformed for statistical analyses.

Results
Significant differences were observed in the parameters average boll weight, lint percentage, 100-seed weight and cotton yield of the cultivars evaluated (Table 1).
The average boll weight (5.82 and 5.50 g respectively) and average yield (2344.0 and 2350.0 kg•ha −1 , respectively) of the cultivars DeltaOPAL and Fibermax 993 were highest while the cultivars Fibermax 966 and Fibermax 993 had the highest lint percentage (38.94%and 39.63%, respectively) and IAC-25 the highest 100-seed weight (13.90 g) (Table 1).The higher infestation density of A. argillacea larvae, the lower average boll weight (5.72 g for 0 larva per plant to 5.25 g for 6 larvae per plant) and the lower yield of the cultivars (2753.0kg•ha −1 for 0 larva per plant to 1566.6 kg•ha −1 for 6 larvae per plant), whereas lint percentage and 100-seed weight were not influenced by the larval density (Table 1).The time of larval infestations was significantly related with yield, and inoculations 30 DAE and 60 DAE reduced the mean cultivaryield more (2075.8and 2038.3 kg•ha −1 , respectively) than late inoculations 90 DAE (2273.0kg•ha −1 ) (Table 1).
The partition analysis of the cotton cultivars and infestation period for the parameter average boll weight indicated that all cultivars were influenced by the time of infestation, so that the average boll weight was lowest when infestation occurred 60 DAE on IAC-25 (4.75 g), 30 DAE (5.18 g) on cultivar DeltaOPAL, and 90 DAE (4.85 and 4.99 g, respectively) on cultivars Fibermax 966 and Fibermax 993 (Figure 1).The boll development of cultivar DeltaOPAL was most affected by initial infestation (30 DAE), while for IAC-25 infestations in the middle of the cycle (60 DAE) were most damaging, and late infestations (90DAE) were most harmful to boll development of IAC-25, Fibermax 966, and Fibermax 993 (Figure 1).
The larva infestation period did not influence the lint percentage of IAC-25 and Fibermax 993; infestations 30 DAE were more harmful to cultivar DeltaOPAL than later ones, whereas late infestations (90 DAE) affected cultivar Fibermax 966 most (Figure 1).
The interaction between cultivars and larval density was significant for the parameters 100-seed weight and yield among cultivars and infestation period on the parameters average boll weight, lint percentage, 100-seed weight, and cultivar yield.No significant interaction was detected between infestation period and density (Table 1).
The partition analysis of cotton cultivars and larval density indicated that only 100-seed weight of cultivar Fibermax 966 was reduced in the presence of larvae, where a reduction of 100-seed weight was observed in the presence of larvae (Figure 2).With regard to the performance of cultivars at each larval density, the average 100-seed weight of Fibermax 993 without larva infestation was lower than of the other cultivars, while cultivar Fibermax 966 had lowest average 100-seed weight with two and four A. argillacea larvae per plant and Delta-OPAL, Fibermax 966 and Fibermax 993 with six larvae per plant (Figure 2).
The partition analysis of cotton cultivars and A. argillacea infestation period in relation to 100-seed weight indicated that IAC-25 was more damaged by initial infestations (30 DAE), the cultivar DeltaOPAL was more compromised when infested 60 DAE and cultivars Fibermax 966 and Fibermax 993 were most affected by late infestation (90 DAE), which reduced the 100-seed weight significantly (Figure 1).
The weight of 100 seeds did not vary significantly among cultivars after infestation 30 DAE; 60 DAE, the 100-seed weight of cultivar IAC-25 was the highest, and 90 DAE of IAC-25 and DeltaOPAL (Figure 1).The partition analysis of cotton cultivars and larval density indicated that an increased number of A. argillacea larvae per plant significantly reduced the yield of all cultivars, so that 6 larvae per plant resulted in lower yield of IAC-25, DeltaOPAL, Fibermax 966 and Fibermax 993 (1255.3, 1578.3, 1474.3, and 1958.4 kg•ha −1 respectively) (Figure 2).
The yield of plants without A. argillacea larvae was the same, while DeltaOPAL had the highest yield with two larvae, and Fibermax 993 had the highest yield when infested with four and six larvae per plant (Figure 2).

Discussion
The results showed that the larval density infestation influenced the boll weight and cultivar yield, while time of infestation influenced yield.The partition analysis of cotton cultivars and larvae infestation period for yield indicated that IAC-25 was more impaired by infestations 60 DAE and DeltaOPAL 30 DAE, while infestations at different phenological stages of the plants did not affect the yield of Fibermax 966 and Fibermax 993 (Figure 1).Silva et al. [13] to simulate the defoliation at different phenological stages of cotton, verified that defoliation done in the early fruiting (opening of the first flower) resulting in major reductions in productivity.
For the parameter weight of 100 seeds, the partition analysis of the cotton cultivars and infestation period showed that the 100-seeds weight of cultivar IAC-25 was highest at 60 and 90 DAE, suggest that plantations of cultivars DeltaOPAL, Fibermax 966 and Fibermax 993 are more prejudiced than of IAC-25.This scenario is more severe in seed production fields, where a reduction in seed weight is likely to affect the seed quality.
The partition analysis of cotton cultivars and larval density indicates that at a low infestation (up to 2 larvae per plant), DeltaOPAL can overcome the pest attack better than the other cultivars, while at high infestations (4 -6 larvae per plant), cultivar Fibermax 993 supports A. argillacea infestation better than the other cultivars.
The reduction in cotton yield by A. argillacea was evaluated by other researchers [14], who found that the reduction in cotton yield was a consequence of the reduced boll weight or number.According to [15], this can be explained by the ease with which assimilates reach these plant parts.Moreover, the leaves of the main stem are the most important because they nourish branch growth, being one of the main factors contributing to cotton yield [16].
The consumption of the primary leaves of cotton plants by A. argillacea (early infestation) was more damaging because these leaves are grown first by the plant and account for over 80% of the cotton yield [17].In addition, vegetative leaves live longer and have a greater leaf area than fruit leaves [18], and are responsible for vegetative production and growth of the cotton plant [3].The development of cotton plants is compromised by the removal of leaves from the main stem [19], which drastically reduces the yield, number of fruit branches and plant height [20].

Conclusion
The result on this paper shows that higher infestation densities of A. argillacea increase yield reduction and the presence of larvae reduces the 100-seed weight of cultivar Fibermax 966.Initial infestations affect the boll weight of IAC-25 and DeltaOPAL, whereas the cultivars Fibermax 966 and Fibermax 993 are most affected by late infestations.Initial infestations of A. argillacea affect the fiber percentage of cultivar DeltaOPAL and late infestations are more harmful to Fibermax 966.Initial infestations of A. argillacea reduce the yield of cultivar DeltaOPAL, while infestations 60 DAE cause the most damage to IAC-25 and the time of infestation does not influence the other cultivars.

Figure 1 .
Figure 1.Mean values of the unfolding analysis of the significant interactions for (a) boll weight (g), (b) lint percentage (%), (c) 100-seed weight (g) and (d) yield (kg•ha −1 )in cotton cultivars infested withAlabama argillacealarvae at different times after plant emergence.Bars followed by the same lowercase (comparisons for cultivars across plant age) or capital (comparisons for cultivars within the same plant age) letters do not differ significantly by Tukey's (p > 0.05).

Figure 2 .
Figure 2. Mean values of the unfolding analysis of the significant interactions for (a) 100seed weight (g), and (b) yield (kg•ha −1 ) in cotton cultivars infested with different densities of Alabama argillacea larvae.Bars followed by the same lowercase (comparisons for cultivars across larvae density) or capital (comparisons for cultivars within the same larvae density) letters do not differ significantly by Tukey's test (p > 0.05).