BioOne Navigation Bar. This is an image map. Text links are at the bottom of the page. Get Help Using BioOneBioOne HomepageSearch BioOneView ArchivesCurrent Issues
[PDF] [Create Reference]


[Materials and...] [Results] [Discussion] [References] [Tables]

Journal of Economic Entomology: Vol. 93, No. 3, pp. 800°V804.

Comparison of Greenhouse Screening Materials for Excluding Whitefly (Homoptera: Aleyrodidae) and Thrips (Thysanoptera: Thripidae)

Michelle L. Bell,a, b and James R. Bakerb

aCurrent address: SePRO Corporation, 11550 N. Meridian Street, Ste. 600, Carmel, IN 46032

bDepartment of Entomology, Box 7613, North Carolina State University, Raleigh, NC 27695

Manuscript Received by the Society 23 August 1999
Manuscript Accepted 17 February 2000


Twenty-eight greenhouse screening materials, with predetermined airflow resistance values, were evaluated for exclusion of the silverleaf whitefly Bemisia argentifolii Perring & Bellows and thrips from a mixed-species population. Screens differed in exclusion efficacy, expressed as a percentage of the fiberglass window screen control and at an approach velocity of 92 m/min, from −35 to 94% for silverleaf whitefly and from −13 to 95% for thrips. Seventeen screens excluded more silverleaf whitefly than did the window screen control, whereas only seven excluded more thrips. One material differentially excluded whitefly over thrips; many more differentially excluded thrips over whitefly. Airflow resistance, indicative of mesh hole size, did not necessarily correspond with degree of exclusion. Not all materials characterized as highly resistant to airflow provided significant exclusion. Exclusion of both types of pests was attained with several moderate- and one low-resistance screen. Another low-resistance screen excluded silverleaf whitefly only.

Keywords: Bemisia argentifolii, greenhouse screening, exclusion, physical control.

Whiteflies and thrips are among the most important pests of greenhouse crops because of both direct feeding damage and disease transmission (Hussey 1985 ). Because of pest-acquired resistance, management practices that rely on insecticides are growing increasingly less effective, and less environmentally and economically appropriate. Exclusion of pests should be one of the first tactics considered in an integrated pest management system. With losses and rising costs of registered pesticides, the use of screens to exclude greenhouse pests is now more cost effective than in the past (Neal 1992 ). Reductions in pest populations (Berlinger et al. 1983 ; Berlinger et al. 1991 ; Berlinger et al. 1992 ; Robb and Parrella 1988 ; Baker and Jones 1989 ), lower incidence of insect-transmitted diseases (Berlinger et al. 1983 ; Berlinger et al. 1991 ; Berlinger et al. 1992 ; Baker and Jones 1989 ; Baker and Jones 1990 ), and fewer needed pesticide applications (Berlinger et al. 1983 , Robb and Parrella 1988 ) have been documented when screening is used. Exclusion screens for the greenhouse may become a necessary alternative to pesticide use. At the very least, screens will provide a valuable addition to other management practices, especially in areas where greenhouse pests overwinter outside.

With the rising popularity of exclusion screens among greenhouse manufacturers and operators (Hudson et al. 1996 ), many new screening products have become available. Selection of the screen most beneficial to a particular greenhouse situation requires much information. Of greatest need is characterization, by an independent laboratory, of effects on airflow restriction and ability to exclude pest insects for the variety of screening products available. Previous work investigating the mechanism of screen efficacy focused on hole size and geometry in relation to thoracic width of pest insects (Bethke and Paine 1991 ), whereby laboratory studies were performed in the absence of forced air. In a mechanically ventilated, screened greenhouse situation, however, we observed instances where insects wider than the barrier holes were able to pass through the screens. We hypothesized that the velocity with which air moved through a greenhouse screen was a basic determinant of its exclusion performance. The objective of this work was to compare efficacy of screens for excluding whiteflies and thrips under summer ventilation conditions resembling those of a typical commercial greenhouse.

Materials and Methods Return to TOC

Equalizing Air Approach Velocities.

Twelve wood-framed and polyethylene plastic-covered cages (0.5 by 0.5 by one m) were constructed with open fronts to allow covering with test materials. A 2,085 cm3/min (265 feet3/min) squirrel cage blower was installed at the rear of each cage to pull air through the screening materials and into the cage. In separate wind tunnel experiments (unpublished data), we generated resistance curves for screening materials by plotting fabric pressure drops versus a range of air velocities. Resistance curves were used in the current study to equalize the velocity of air entering the cages through the test material. Air approach velocities were equalized by measuring the pressure drop across the screen with a Dwyer Mark II, model 25 manometer (Dwyer Instruments, Michigan City, IN). A damper to restrict blower output was adjusted until the manometer reading was equal to the pressure drop needed to achieve an approach velocity of 92 m/min (300 feet/min) for each fabric; an airflow velocity recommended for well-designed production greenhouses. Additionally, in 1996 a vane-type, digital anemometer (Omega Engineering, Stamford, CT) was used at the front of each cage to confirm that air was entering at the desired approach velocity.

Experimental Design and Sampling.

Whitefly and thrips exclusion experiments were conducted in 1995 and 1996. The silverleaf whitefly Bemisia argentifolii Perring & Bellows exclusion studies were performed in a greenhouse where the cages were placed side-by-side on one bench across from whitefly-infested poinsettias [Euphorbia pulcherrima (Willd. ex Klotzch)] on a facing bench. In thrips exclusion studies, the cages were placed side-by-side on a 1-m high bench outdoors in late April through July. Thrips species trapped included Frankliniella fusca (Hinds), F. occidentalis (Pergande), F. tritici (Fitch), Sericothrips variabilis (Beach), Thrips tabaci Lindeman, and undetermined Chirothripidae and Plaeothripidae. Voucher specimens were deposited at the North Carolina State University Insect Collection, Raleigh, NC.

Testing of a given screening material on a given cage constituted one experimental unit. One run of the 12 cages comprised a block. Fiberglass window screen (standard mesh: 7 by 6 threads per centimeter [18 by 16 threads per inch]) served as the control treatment in all studies. In each block, the fiberglass window screen control was randomly assigned to two cages. Twenty-six (in 1995) and 28 (in 1996) test materials were assigned to the remaining 10 cages with a cyclic incomplete block design generated by Gendex software (Commonwealth Scientific and Industrial Research Organization-IAPP Biometrics Unit, Clayton, Victoria, Australia). Three, two, and one discontinued screens in the 1995 whitefly exclusion test, 1995 thrips exclusion test and 1996 tests, respectively, were included in the design, but data were not included in analysis or reported. For any block where a cage was left unassigned, a third fiberglass window screen control was added and assigned to that cage. In each study, four replications of three blocks each were performed for a total of 12 blocks.

Before each installation of the test material, one 7.5 by 13-cm yellow sticky trap was vertically mounted in each cage perpendicular to the airflow. Cards were collected and numbers of trapped whitefly or thrips were counted after the completion of a block.

Data Analysis.

Data were tested by analysis of variance (ANOVA). Trap count data were transformed into a percentage of the coinciding control counts. These data, in turn, were transformed with a square-root transformation for thrips exclusion data and logarithmic transformation for whitefly exclusion data to correct for non-normality as indicated by residual plots from the ANOVA. Treatment means were pooled over years because there was no significant screen °— year interaction. Pairwise comparisons of means were made by the least squares means test (PROC GLM, SAS Institute 1990 ) at P < 0.05. As a result of repeated pairwise comparisons, screens were categorized as group 1, 2, or 3 screens, whereby group 1 screens excluded more pest insects than the control and similar to the screen with the highest percentage efficacy. Group 2 screens excluded more insects than the control and less than the screen with the highest percentage efficacy. Group three screens excluded similar to the window screen control. As a derivation of Abbott°¶s formula (Abbott 1925 ), exclusion efficacy was computed by subtracting counts as percentages of the control from 100% as exclusion efficacy (%) = [1 − (test screen count °“ control count)] °— 100.

Results Return to TOC

Seventeen screens gave greater whitefly exclusion than the fiberglass window screen control, and of these, the seven group 1 screens were the most efficacious screens for whitefly exclusion (Table 1) . Eleven group 3 screens did not exclude whitefly better than the window screen control. Of these, four screens (Kontrol 45204, Pak WP87, Lumite 32 °— 32, and Kontrol 45304) exhibited a lower mean efficacy than the control. However, in pairwise comparisons with the control, these screens were not significantly different and may be considered equal to other group 3 screens.

Seven screens gave greater thrips exclusion than the control, and of these, the two group 1 screens were the most efficacious screens for thrips exclusion (Table 1) . Twenty-one group 3 screens excluded thrips no better than the window screen control. Of these, four screens (Insecta 500, Kontrol 45304, Kontrol 45204, and Econet L) exhibited a lower mean efficacy than the control. However, in pairwise comparisons with the control, these screens were not significantly different and may be considered equal to other group 3 screens.

Discussion Return to TOC

Only BugBed 123 appeared in group 1 in both sets of studies, providing exclusion >93% compared with window screen. We consider BugBed 123 the best screen for excluding both whitefly and thrips. No-Thrips was rated as a group 1 screen for thrips exclusion and as a group 2 screen for whitefly exclusion. Although only a group 2 screen, No-Thrips gave whitefly exclusion at 86.8% ranking it in the top third of all screens. BugBed 110UV, BugBed 85, and Econet S were rated as group 1 screens for whitefly exclusion and as group 2 screens for thrips exclusion. If whiteflies are the major pest problem in a particular greenhouse, these three screens will probably give excellent whitefly exclusion as well as provide the added benefit of good thrips exclusion.

Several screens that did not exclude thrips excluded whiteflies. According to Bethke and Paine (1991) greenhouse pests are likely to be excluded by screens with hole sizes smaller than the insects thoracic width. They also noted that projecting body parts such as the wings of whiteflies further limit their ability to penetrate many screens. In general, species of thrips attacking greenhouse crops are narrower than species of whitefly pests, including the silverleaf whitefly. These data suggest that the holes of several screens allow differential passage of thrips pests over the silverleaf whitefly and undoubtedly other whiteflies.

FlyBarr was the only screen that excluded thrips and not whiteflies. FlyBarr is a uniquely three-dimensional, polyspun material. Perhaps whiteflies are able to wiggle or are pulled through between the relatively slender, moveable fibers, whereas the thrips are caught in these irregular fibers by their fringed wings.

Screen Properties and Exclusion.

According to our unpublished data, five screens used in this study, No-Thrips, Econet S, Typar, PakWP87, and Econet T, were characterized in wind tunnel studies as highly or very highly resistant to airflow. In general, the higher the airflow resistance of a screen, the smaller the hole size through which insects must pass. However, of the five high-resistance screens tested, only No-Thrips and Econet S excluded both whiteflies and thrips better than did window screen. In addition to hole size, hole geometry may play a role in insect penetration through screens (Bethke and Paine 1991 ). The holes of No-Thrips and Econet S are small and square, whereas Typar, marketed primarily as a crop blanket, is a polyspun material with holes very inconsistent in size and shape. Holes of Pak WP87, a woven material, are inconsistent in size and shape because of a relatively thick coating of a UV light degradation-inhibiting acrylic. Econet T very effectively excludes whiteflies as a group 1 screen but does not exclude thrips. Econet T has rectangular holes that allow thrips to pass readily.

Four of the seven screens that exclude thrips in this study were characterized (unpublished data) as moderately resistant to airflow; several that exclude whiteflies are considered moderately resistant. Two screens characterized as low in airflow resistance, BugBed 110UV and Protex 1 (unpublished data), excluded whiteflies; additionally, BugBed 110UV excluded thrips. Again, hole geometry may explain differential exclusion of whiteflies over thrips as holes of Protex 1 are rectangular.

Although it may seem best in a commercial situation to use the most efficacious screen available, other considerations must be weighed. Many efficacious screens have a small hole size and are more resistant to airflow than are more open-meshed screens. As screen hole size decreases, effort by greenhouse fans to move air through it increases, and a larger pressure drop is created in a greenhouse. A large drop in static pressure may result in inadequate air exchange, higher energy consumption by fans, excessive wear on the fans, and high greenhouse temperatures (NGMA 1993 ). In selecting the proper greenhouse screening material, the grower should first determine the most serious pest(s) of the crop that need to be excluded. If, for example, thrips and whiteflies are the major pests of concern and a low threshold of thrips could be tolerated, a screen with slightly larger holes than one providing total thrips exclusion may suffice. However, if the reason for excluding thrips is to prevent transmission and spread of disease, such as viral pathogens, then nearly complete exclusion is necessary, and screens with smaller holes are essential. Where exclusion of thrips or thrips-transmitted diseases is of paramount importance but where whiteflies also may be pests, BugBed 123 and No-Thrips should be strongly considered. An advantage of BugBed 123 over No-Thrips is that BugBed 123 is only moderately resistant to airflow, whereas No-Thrips is very highly resistant to airflow. The two screens also may differ in longevity. On a demonstration and research greenhouse at North Carolina State University on which BugBed 123 was installed, the screen barely lasted one season before tearing occurred from weathering. No-Thrips lasts longer under similar conditions. BugBed 110UV, the screen exhibiting the highest percent exclusion for whiteflies at 94%, has enhanced protection from UV light degradation and thus may be longer-lived than BugBed 123 and other screens that lack UV protection. Another distinct advantage of BugBed 110UV is that its resistance to airflow is lower than BugBed 123; we characterize BugBed 110UV as a low-resistance screen (unpublished data). The degradation inhibitor that is incorporated in fibers of BugBed 110UV is dark, giving the screen a charcoal gray appearance, and may, in part, result in the high efficacy of this low-resistance screen. Because many greenhouse pest insects, such as whiteflies, thrips, and aphids, are attracted to light colors, BugBed 110UV may be less attractive to pest insects that may encounter the screen while in flight.

In conclusion, about two-thirds of greenhouse screening products exclude whiteflies appreciably; one-third excludes thrips. High airflow resistance, often costly in terms of requiring greater screening area to maintain adequate airflow for effective cooling in a greenhouse, neither ensures nor is necessary for significant exclusion of whiteflies and thrips. Of the 28 screens tested, only Typar and Reemay are marketed primarily for other uses. Although several screens performed poorly in these exclusion tests, most are currently being marketed as insect exclusion screens for commercial greenhouses. This situation points to the pressing need for continued, independent evaluation of screens for their pest exclusion properties.


We thank Edwin Shearin (Department of Entomology, North Carolina State University) for assistance in generating resistance curves and for building the cages, and Cavell Brownie (Department of Statistics, North Carolina State University) for help with experimental design and statistical analyses. David Stephen (Plant Disease and Insect Clinic and Department of Entomology, North Carolina State University) was very helpful in confirmation and identification of thrips species. Earlier reviews of the manuscript by D. A. Bailey (Department of Horticultural Science, North Carolina State University); R. J. Kuhr (Department of Entomology, North Carolina State University), and D. M. Benson (Department of Plant Pathology, North Carolina State University) were appreciated. We are grateful for financial support from the Fred C. Gloeckner Foundation, and the White and Helmich Research Grants of the Horticultural Research Institute.

References Return to TOC

1. Abbott, W. S. A method of computing the effectiveness of an insecticide: J. Econ. Entomol.265°V276. 18 1925. Find this article on other systems

2. Baker, J. R., and R. K. Jones. Screening as part of insect and disease management in the greenhouse: N.C. Flower Growers°¶ Bull. Dec.1°V9. 34 1989.

3. Baker, J. R., and R. K. Jones. An update on screening as part of insect and disease management in the greenhouse: N.C. Flower Growers°¶ Bull. Dec.1°V3. 35 1990.

4. Berlinger, M. J., Alla M. Gol°¶berg, R. Dahan, and S. Cohen. The use of plastic covering to prevent the spread of tomato yellow leaf curl virus in greenhouses: Hassadeh.1862°V1865. 63 1983.

5.  Berlinger, M. J., S. Leblush-Mordechl, D. Fridja, and N. Mor. 1992. The effect of types of greenhouse screens on the presence of western flower thrips: a preliminary study, pp. 13°V16. In Contribution from the Agricultural Research Organization. No. 3716-E, 1992 series. The Volcani Center, Bet Dagan, Israel.

6. Berlinger, M. J., S. Mordechl, and A. Leeper. Application of screens to prevent whitefly penetration into greenhouses in the Mediterranean Basin: Proceedings, Working Group Integrated Control in Protecting Crops under Mediterranean Climate, 29 September°V2 October. Bulletin of IOBC/WPRS 1991/14/5 1991. 105-110. IOBC/WPRS Alassio, Italy.

7. Bethke, J. A., and T. D. Paine. Screen hole size and barriers for exclusion of insect pests of glasshouse crops: J. Entomol. Sci.169°V177. 26 1991. Find this article on other systems

8. Hudson, W. G., M. P. Garber, R. D. Oetting, R. F. Mizell, A. R. Chase, and K. Bondari. Pest management in the United States greenhouse and nursery industryV: Insect and mite control. HortTechnology.216°V221. 6 1996.

9. Hussey, N. W. Whitefly control by parasites: N. W. Hussey N. Scopes Biological controlthe glasshouse experience 1985. Cornell University Press Ithaca, NY.

10. NGMA] National Greenhouse Manufacturers Association. Recommendations for using insect screens in greenhouse structures. Addendum to NGMA ventilation and cooling standards. 1993. NGMA Buffalo, NY.

11. Neal, K. Screen pests out, reduce chemical use: Greenhouse Manager. April.54 11 1992.

12. Robb, K. L., and M. P. Parrella. Chemical and non-chemical control of western flower thrips: Proceedings, Fourth Conference on Insect and Disease Management on Ornamentals, 1°V3 March 1988 1988. 94-103. Society of American Florists Kansas City, MO.

13. SAS Institute. SAS/STAT user°¶s guide version 6. 1990. SAS Institute Cary, NC.

Tables Return to TOC

Table 1. Efficacy of silverleaf whitefly and thrips exclusion by commercial greenhouse screens, as percentages of the fiberglass window screen control, group ranking, and air resistance rating

© 2000, Entomological Society of America

Home | Current Issues | Archives | Search | Send Feedback | Help