Investigating New Management Approaches for Picture-Winged Flies in Sweet Corn

Objective 1.

In the Fall experiment, water sensitive cards exposed to the 103 liter/ha water rate received less overall spray coverage (13.5 ± 3.0 %) than the 207 liter/ha rate (30.1 ± 4.9 %; (F = 8.59, df = 1, 82, P = 0.004). Spray coverage was significantly affected by location of the spray card on the plant (F = 35.74, df = 2, 82, P <0.001). Cards on the ears received the least coverage (3.0 ± 1.2 %), stem-located cards received intermediate coverage (18.5 ± 4.0 %) and tassel-located received the greatest mean coverage (42.2 ± 5.9). Presence of the spreader adjuvant was not significant (no adjuvant = 27.7 ± 4.7%, adjuvant = 15.6 ± 3.4%; F = 0.77, df = 1, 82, P = 0.384). All interaction terms between adjuvant, location, and water volume were not significant (P > 0.05).

In the Spring trial, spray pressure and water rate did not have a significant effect on coverage (high pressure = 10.9 ± 0.8 %, low pressure = 8.6 ± 0.7 %; F = 3.27; df = 1, 7.46; P = 0.111). Water rate also did not have a significant effect on coverage (high water rate = 10.6 ± 0.8 %, low water rate = 9.0 ± 0.7 %; F = 5.04; df = 1, 7.46; P = 0.057). As in the Fall experiment, spray card location resulted in significant coverage differences (F = 181.70; df = 4, 169.1; P<0.001). Leaf coverage was greater than stem coverage for all locations (Table 1). The upper leaf received the greatest coverage, followed by the lower leaf. Among the stem sections, the greatest coverage was received on the upper stem, followed by the middle stem and the least coverage was on the lower stem. There was a significant rate x location interaction (F = 3.65; df = 4, 169.1; P = 0.007). Coverage was greatest with a high water rate (347 liters/ha) on the upper leaf and upper stem than with 187 liters/ha. Water rate did not affect card coverage at the middle stem, lower stem, and lower leaf. No other interaction terms were significant (P >0.05).

As with the spray cards, the spray residues significantly affected E. eluta survivorship depending on which section of the plant the flies had been exposed to (F = 44.26; df = 4,158; P<0.001). Fewer E. eluta survived exposure to leaf residues than stem residues (Table 2). Among stem locations, fewer E. eluta survived exposure to upper-stem residues, followed by mid-stem residues and then low-stem residues. With the spray cards, slightly greater coverage was achieved with the 374 liters/ha, but more flies survived when caged on these residues (28.0 ± 3.1 %) than flies exposed to residues deposited by a 187 liters/ha water rate (15.8 ± 2.5 %; F = 7.99; df = 1,58; P = 0.005). Although the location x water rate interaction was not significant (P > 0.05), this effect was pronounced with stem residues, probably because stems received far less coverage than leaves.

Table 1. Mean + SEM percentage spray droplet coverage on spray cards placed at different locations on sweet corn plants in Spring 2016. Sprays were applied with two different water rates and pressures.

Significantly different means are followed by different letters within a column (Tukey-Kramer HSD, P < 0.05). 

Table 2. Mean + SEM percentage of alive E. eluta flies after 24 hours caged on plant sections treated with a high label rate of Hero¹ insecticide in Spring 2016.

Significantly different means are followed by different letters within a column (Tukey-Kramer HSD, P < 0.05).

¹zeta-cypermethrin + bifenthrin; FMC Corporation, Philadelphia, PA.

Objective 2. During the fall experiment, the dominant species observed in the pyrethroid plots at the beginning of the trial was C. massyla, but by the end of the trial, E. stigmatias was the most prevalent species. Silk fly populations in the untreated check plots were as great as 3 flies per row-m. A sum total of 456 healthy silk flies were observed on the plants from 240 row-m 4 h following 8 pyrethroid applications (Fig. 1). Early in the trial, the steepest population reductions followed the first Baythroid and Mustang applications. More flies were observed on plants following the two later Warrior applications, possibly due to greater populations in the field following the earlier Baythroid application. During this time, more E. stigmatias were entering the field. However, it is unclear if the population fluctuations were due to differential physiological effects of insecticide application or from silk fly movement in and out of the treated block.

A total of 67 dead or twitching flies were removed from plants and plastic sheets after pyrethroid treatments (red bars in Fig 1). Of these flies, only a single C. massyla and a single E. stigmatias fully recovered after Mustang and Warrior, respectively.

Figure 1. Total number of live healthy flies (blue line) and insecticide-affected flies (red bars) 4 h after each insecticide application. Insecticide-affected flies were taken to the laboratory for further observation. Of the insecticide-affected flies, a single C. massyla recovered from a Mustang application and a single E. stigmatias recovered from a Warrior application.

In the Spring trial, silk fly populations were routinely monitored by visual inspection. Five days before treatments were applied, pre-treatment population density was 0.2 per 30.5 row-m. Fly populations in the untreated check plot rapidly increased, averaging 70.3 per 30.5 row-m over the course of the experiment, but pyrethroid applications stabilized silk fly populations at a much reduced level compared to the untreated check plots. Average silk fly population density in the pyrethroid plots was 3.3 per 30.5 row-m. Populations in the pyrethroid treatment peaked at 7.3 per 30.5 row-m at which time the population in the check plots increased from 11.8 to 67.3 per 30.5 row-m (Figure 2).

Figure 2. Silk fly populations in the untreated check plots and pyrethroid-treated plots the day after insecticide application in the Spring 2016 trial.

A total of 3,814 silk flies were observed from the untreated check plots; 66.0% were E. eluta, 29.6% were E. stigmatias, 2.9% were C. massyla, and 1.4% were unidentified Euxesta spp. (including E. annonae). Pyrethroid treatments affected the proportion of E. eluta and E. stigmatias observed in the plots (E. eluta proportion t = 8.91; df = 47.0; P <0.001; E. stigmatias: t = 6.20; df = 35.8; P < 0.001). Of the 172 silk flies observed from the pyrethroid plots, 22.1% were E. eluta, 65.7% were E. stigmatias, 8.1% were C. massyla, and 4.1% were unidentified Euxesta spp. (including E. annonae.

Time of day had a significant effect on the fly distribution in untreated check plots. The proportion of flies in the upper canopy was greatest in the evening and morning hours (F = 19.18; df = 3, 20; P = <0.001). Although the proportion of flies in the middle canopy was numerically great in the late morning hours, the difference with the other times of day was not significant (F = 3.66; df = 3, 5.32; P = 0.093). The proportion of flies in the lower canopy at the ear level or below was greatest at noon, and significantly less in the middle morning and evening hours (F = 7.30; df = 3, 5.44; P = 0.024; Figure 3).

Figure 3. Percentage of silk flies observed in the sweet corn canopy at four different times of day. Observations were conducted between 16 April and 5 May 2016. Fly locations: upper canopy = the stem or leaves between the tassel and the second leaf below the flag leaf on the tassel; middle canopy = the second leaf and the leaf above the ear; and at the ear and below were considered to be in the lower canopy. Observation periods: morning = between 8:00 AM and 9:30 AM (EDT); mid-morning between 10:00 and 11:30 AM (EDT); noon = between 11:30AM and 1:30 PM (EDT); evening = after 5:30 PM (EDT).

Objective 3.

In the Fall experiment, pyrethroid treatments resulted in the lowest ulidiid fly damage to ears and the lowest maggot load in the ears (Table 3). Radiant + Nu-Lure fly damage ratings and maggot load were intermediate, but closer to the pyrethroid treatment than the check plots. The Radiant + Nu-Lure treatment had the lowest proportion of FAW damaged ears. Although there were fewer FAW damaged ears in the pyrethroid treatment than the check plots, the pyrethroid treatment was more similar to the untreated check than the Radiant + Nu-Lure treatment.

Table 3. Mean + SEM insect damage to sweet corn ears following 9 different insecticide applications in the Fall 2015 experiment.

Significantly different means are followed by different letters within a column (Tukey-Kramer HSD, P < 0.05).

¹Fly damage rating: 0 = no maggots in the ear, 1 = silk damage only, 2 = tip damage, 3 = maggots feeding in the top ¼ of the ear, 4 = maggots feeding in the top ½ of the ear, and 5 = maggots feeding throughout the entire ear.

²Maggot load: where 0 = no maggots, 1 = <5 maggots, 2 = 5-20 maggots, and 3 = >20 maggots per ear.

In the Spring trial, silk fly damage was less severe than in the fall, and all treatments reduced silk fly infestation and damage to the ears (Table 4). There was no separation between spinetoram with and without Nu-lure for any insect damage category. Even though worm pressure was less severe than during the fall experiment, ear protection from fall armyworm was best in the spinetoram treatments; pyrethroid protection was intermediate between spinetoram and the untreated check.

Table 4. Mean + SEM insect damage to sweet corn ears following 9 different insecticide applications in Spring 2016.

Significantly different means are followed by different letters within a column (Tukey-Kramer HSD, P < 0.05).

¹Fly damage rating: 0 = no maggots in the ear, 1 = silk damage only, 2 = tip damage, 3 = maggots feeding in the top ¼ of the ear, 4 = maggots feeding in the top ½ of the ear, and 5 = maggots feeding throughout the entire ear.

²Maggot load: where 0 = no maggots, 1 = <5 maggots, 2 = 5-20 maggots, and 3 = >20 maggots per ear.

Objective 4.

In the Fall trial, a total of 2,582 silk flies were removed from the 12 cages (Fig. 4). The vast majority of these (95.5%) were E. stigmatias, followed by C. massyla (2.9%). Euxesta eluta was extremely uncommon during the fall trial, and comprised only 1.0% of the emerging flies. There were no significant differences between the residue destruction treatments and the standing corn treatment on fly emergence from the cages (F = 1.58, df = 3, 120, P = 0.199). Although there was a significant date x treatment interaction (F = 2.34, df = 42, 120, P <0.001), no clear, consistent treatment pattern on fly emergence was observed. Fly emergence from the disked and mowed plots peaked on 7 December, and peak emergence from the standing corn plots occurred on 18 December. Extrapolating the number of silk flies captured from the emergence cages by the total area that the cages covered to a full acre would mean that one acre of infested sweet corn could potentially produce 1.5 million silk flies.

Figure 4. Fall trial mean adult silk fly emergence from soil within cages erected over the control plots where plants were left standing, mowed, disked once or disked twice.

In the spring trial, a total of 5,604 silk flies were removed from the cages. The species complex composition during the first two weeks (3,175 flies captured) was 60.2% E. stigmatias, 31.1% E. eluta, 4.1% C. massyla, 0.6% were E. annonae, and 4.1% unidentifiable. During the second half of the cage study (2,429 flies captured), the proportion of E. stigmatias decreased to 43.3%, while E. eluta increased to 45.8%, C. massyla increased to 7.2%, 1.2% were E. annonae, and 2.5% were unidentifiable. The species complex captured by the emergence cages differed from that observed in the field during the corn reproductive stages when out of 3,267 flies observed during the silking period, 64.2% were E. eluta, 29.8% were E. stigmatias and 1.7% were C. massyla.

Flies had already begun emerging out of the soil by the first sampling date which occurred 5 d after harvest, indicating that many larvae had already finished development and left the ear before harvest and their development in the soil was nearly complete by the time cages were installed. Fly emergence per m² per day in the cages over standing corn initially peaked on 31 May, decreased, and then increased to a greater second peak on 17 June (Fig 5). A similar ‘double emergence peak’ was observed from the emergence cage experiment in December 2015.

Treatment, date, and the interaction between treatment and sampling date significantly affected silk fly counts in the emergence cages (F = 98.78; df = 2,270; P <0.001; F = 11.57; df = 11, 270; P < 0.001; F = 2.19;df = 22, 270; P = 0.002). For all sampling dates except 17 May, fewer flies emerged from disked plots than standing corn plots. Fly emergence from ploughed plots did not differ significantly from standing corn plots on 17 May, 27 May, and 7 June. Data from 7 June were included in the analysis even though 9 cages were blown down by a heavy windstorm the evening before; treatment capture did not significantly separate (P = 0.062). Fly emergence between disked and ploughed plots only differed significantly on 26 May (Fig 5). Extrapolating the number of silk flies captured from the standing corn plots by the total area that the cages covered to a full acre would mean that one acre of infested sweet corn in the spring would have produced more than 570,000 silk flies.

Figure 5. Spring trial mean adult silk fly emergence from soil within cages erected over the control plots where plants were left standing, ploughed, and disked.

Discussion      

The spray cards indicated that coverage significantly decreases between the upper canopy, middle canopy and lower canopy. This decrease in coverage could influence spray efficacy for silk flies, and different ground application methods were tested for improving coverage. Spray coverage was slightly improved with increased water rate but not increased pressure. Nozzle selection has been assessed using similar methods in cotton; several were recommended for canopy penetration sprays targeting stink bugs (Sumner et al. 2007). In soybean, the denser the plant canopy, the fewer droplets penetrate the canopy. This results in decreased insecticide efficacy targeting Lepidoptera pests (Hutchins and Pitre 1984). Further areas of research include nozzle selection, incorporating drop nozzles, and comparing aerial application methods with each other for improved canopy penetration and insecticide efficacy.

Euxesta stigmatias is not as susceptible to some of the commonly used pyrethroids as E. eluta (Owens et al. 2016), and if flies are not located in the upper canopy, this decreased susceptibility might allow some to survive an insecticide application. Unexpectedly, E. eluta survivorship was greater when exposed to residues from the high water rate, possibly due to a decrease in active ingredient concentration. The observation data further indicate that E. stigmatias is not as impacted by pyrethroids as E. eluta. While the fly populations were significantly reduced in the pyrethroid plots, the species composition shifted from majority E. eluta to majority E. stigmatias.

A greater proportion of the silk flies in the field were observed in the upper canopy in the early morning and early evening, approximately 2 hours after sunrise and before sundown. These observations confirm earlier reports of greater silk fly activity in the early morning and evening hours (Seal et al. 1995).  Silk flies congregate on the tassel in the evening, and females on the tassel late in the day had fewer eggs in their ovaries (Seal et al. 1996), which may stimulate flies to seek food sources and therefore come into greater contact with an insecticide. Because flies remain in the upper canopy during the night, it is possible that an evening spray would be more effective than a morning spray.

The silk fly population and subsequent damage to ears was greatest in the Fall experiment than in the Spring experiment. There was also severe armyworm pressure during the vegetative and reproductive corn stages in the Fall trial compared with the Spring trial. Scouts have previously observed greater silk fly abundance in armyworm infested fields. Furthermore, C. massyla is reported to be attracted to noctuid damaged grasses (Allen and Foote 1992), and the armyworm infestation in the Fall experiment may have contributed to the greater silk fly population.

A spinosad bait used for tephritid management, GF-120^® NF Naturalyte^® (Dow Agro Sciences, Indianapolis, IN), was used in shade houses to control C. massyla, where it was hampering efforts to grow the giant reed Arundo donax as a host for its gall wasp biological control agent (Goolsby and Mangan 2010). If applied in an efficacious manner, spinetoram’ s unique mode of action could help reduce the possibility of silk flies developing resistance to pyrethroids by reducing pyrethroid selection pressure. The equivalent protection spinetoram provided compared with pyrethroids in the spring could be due to silk flies ingesting the active ingredient while foraging on other material or exploring the environment with their mouthparts. During the Fall trial, when silk fly populations were greater, spinetoram provided intermediate efficacy, but there were no plots in which Nu-Lure was not included in the spinetoram treatment. However, in both Fall and Spring, spinetoram treated plots had the least amount of armyworm damage. It is possible that if spinetoram were used to treat armyworms in reproductive stage corn, some selection pressure would be removed from the pyrethroids. More research is needed in order to understand how feeding stimulants might improve spinetoram efficacy and what rates would be more effective in a rotation with pyrethroids.

The effect of crop destruction on reducing the silk fly emergence from the soil was inconsistent. Seasonal weather variation may influence survivorship in the soil. For example, Caribbean fruit fly survivorship from composting infested fruit was lowest in compost piles that were warmest (Kendra et al. 2007). Heavy rains and waterlogged soil may have also contributed to reductions in fly emergence in the disked plots in the Spring. Between 17-21 May, the research station received 14.1 cm of rain. During the month of June, the research station received 18.1 cm of rain. In the Fall 2015 cage study when disking did not decrease fly emergence, only 6.4 cm of rain were recorded throughout the duration of the emergence cage study. Disked plots often contained standing water after heavy rain events that the standing corn plots did not. Increasing moisture content of the soil following harvest and disking is another possible management strategy. However, until more rigorously controlled experiments are performed, soil incorporation should not be considered as an effective means of preventing silk fly larvae from completing their life cycle. Therefore, the crop should be removed as quickly as possible to prevent adult females from continuing to utilize the field for oviposition. The species composition emerging from the soil differed from that observed in the field during the corn reproductive stages. The ecological interaction of the three species utilizing the same reproductive host needs to be further examined to determine if the three fly species differ in terms of damage potential and whether some level of larval competition within ears alters the final proportion of emerging fly species.

References

Allen, E. J. and B. A. Foote. 1992. Biology and immature stages of Chaetopsis massyla (Diptera: Otitidae), a secondary invader of herbaceous stems of wetland monocots. Proc. Entomol. Soc. Wash. 94: 320-328.

Anonymous. 2009. Sweet corn pest management strategic plan (PMSP). Belle Glade, Fl. (http://www.ipmcenters.org/pmsp/pdf/FLsweetcornPMSP.pdf).

Cowles, R. S., C. Rodriguez-Saona, R. Holdcraft, G. M. Loeb, J. E. Elsensohn, and S. P. Hesler. 2015. Sucrose improves insecticide activity against Drosophila suzukii (Diptera: Drosophilidae). J. Econ. Entomol. 108: 640-653.

Goolsby, J. A. and R. Mangan. 2010. Use of spinetoram bait (BF-120) for management of Chaetopsis massyla in shadehouse-grown Arundo donax. Southwest. Entomol. 35: 573 – 574.

Goyal, G. 2010. Morphology, biology and distribution of corn-infesting Ulidiidae. Ph. D. dissertation, University of Florida, Gainesville.

Hutchins, S. H. and H. N. Pitre. 1984. Effects of soybean row spacing on spray penetration and efficacy of insecticides applied with aerial and ground equipment. Environ. Entomol. 13: 948-953.

Kendra, P. E., M. K. Hennessey, W. S. Montgomery, E. M. Jones, and N. D. Epsky. 2007. Residential composting of infested fruit: a potential pathway for spreas of Anastrepha fruit flies (Diptera: Tephritidae). Fla. Entomol. 90: 314-320.

Mangan, R. L., D. S. Moreno, and G. D. Thompson. 2006. Bait dilution, spinosad concentration, and efficacy of GF-120 based fruit fly sprays. Crop Prot. 25: 125-133.

Nuessly, G. S. and M. G. Hentz. 2004. Contact and leaf residue activity of insecticides against the sweet corn pest Euxesta stigmatias (Diptera: Otitidae). J. Econ. Entomol. 97: 496-502.

Owens, D., G. S. Nuessly, D. R. Seal, T. A. Colquhoun. 2016. Variable pyrethroid susceptibility among the sweet corn infesting Ulidiidae (Diptera) in Florida and new baseline susceptibilities. J. Econ. Entomol. 109: 1283-1288.

SAS Institute. 2008. SAS/STAT 9.2 user’s guide. SAS Institute. Cary, NC.

SAS Institute. 2013. Using JMP 11. SAS Institute Inc. Cary, NC.

Seal, D. R., R. K. Jansson, and K. Bondari. 1995. Bionomics of Euxesta stigmatis (Diptera: Otitidae) on sweet corn. Environ. Entomol. 24: 917-922.

Seal, D. R., R. K. Jansson, and K. Bondari. 1996. Abundance and reproduction of Euxesta stigmatis (Diptera: Otitidae) on sweet corn in different environmental conditions. Fla. Entomol. 79: 413-422.

Sumner, P. E., P. M. Roberts, R. P. Edwards. 2007. Comparison of low-drift nozzles for canopy penetration in cotton. ASABE Annual International Meeting, Minneapolis MN.

United States Department of Agriculture [USDA]: National Agricultural Statistics Service (NASS). 2016. Vegetables 2015 Summary (February 2016). (http://usda.mannlib.cornell.edu/usda/current/VegeSumm/VegeSumm-02-04-2016.pdf).