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 of maximizing access to critical research.

Within-Plant Distribution of Frankliniella species (Thysanoptera:Thripidae) and Orius insidiosus (Heteroptera: Anthocoridae) inField PepperAuthor(s): Eric A. Hansen, Joe E. Funderburk, Stuart R. Reitz, SureshRamachandran, Joe E. Eger, and Heather McAuslaneSource: Environmental Entomology, 32(5):1035-1044. 2003.Published By: Entomological Society of AmericaDOI: http://dx.doi.org/10.1603/0046-225X-32.5.1035URL: http://www.bioone.org/doi/full/10.1603/0046-225X-32.5.1035

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in thebiological, ecological, and environmental sciences. BioOne provides a sustainable onlineplatform for over 170 journals and books published by nonprofit societies, associations,museums, institutions, and presses.

Your use of this PDF, the BioOne Web site, and all posted and associated content indicatesyour acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.

Usage of BioOne content is strictly limited to personal, educational, and non-commercialuse. Commercial inquiries or rights and permissions requests should be directed to theindividual publisher as copyright holder.

POPULATION ECOLOGY

Within-Plant Distribution of Frankliniella species (Thysanoptera:Thripidae) and Orius insidiosus (Heteroptera: Anthocoridae)

in Field Pepper

ERIC A. HANSEN,1 JOE E. FUNDERBURK,2 STUART R. REITZ,3 SURESH RAMACHANDRAN,4

JOE E. EGER,5 AND HEATHER MCAUSLANE6

Department of Entomology and Nematology, University of Florida, Gainesville, FL 32611

Environ. Entomol. 32(5): 1035Ð1044 (2003)

ABSTRACT We evaluated the within-plant distribution of Frankliniella spp. and the predatorOriusinsidiosus (Say) in pepper (Capsicum anuum L.), over a range of Þeld conditions, and we conductedbehavioral experiments to examine the time budgets of F. occidentalis (Pergande) and F. tritici (Fitch)females on pepper plant parts. In the Þeld experiments Frankliniella species composition varied byseasonand location. Still, all populationsof thrips andO. insidiosus inuntreatedand insecticide-treatedpepper were highly concentrated in the ßowers, with 82Ð99% of individuals of each taxa found inßowers. This preference for ßowers was corroborated by laboratory-choice experiments. Adultfemales of F. occidentalis and F. tritici showed a strong preference for pepper ßowers over leaves andbuds. In laboratory observations, females of F. occidentalis spent 3.6� as much time on ßowers as onall other plant parts, and females of F. tritici spent over 6.3� as much time on ßowers as on all otherplant parts. Therefore, the concentration of these thrips in ßowers appears to be behaviorally basedand not an artifact of insecticide applications or sampling. Using estimates of populations fromßowersof Þeld pepper is sufÞcient for understanding the local dynamics of Frankliniella spp. and the predatorO. insidiosus, and for estimating the beneÞts of biological control in scouting programs based onpredator to prey ratios.

KEY WORDS pepper, Anthocoridae, Thripidae, within plant distribution, biological control

THRIPS IN THE GENUS Frankliniella (Thysanoptera:Thripidae) are ubiquitous, polyphagous pests of veg-etable, fruit and ornamental crops. Feeding by thesethrips can result in distortion, discoloration, stunting,and silvering of foliage, ßowers and fruits of crops(Childers 1997). An even greater concern with Frank-liniella thrips is the ability of some species to transmitpathogens in the genus Tospovirus (Bunyaviridae) to awidevarietyofplant species(SetherandDeangelis1992,Ullmanet al. 1997).Theseplantpathogens areespeciallydifÞcult to control in crops grown in mild climates andgreenhouses, which favor nearly continuous croppingand year-round presence of thrips vector populations.Direct feeding damage and pathogen transmission bythrips cost growers billions of dollars worldwide in con-trol measures and lost productivity (Ullman et al. 1997).

Not only do species of Frankliniella feed on manydifferent species of host plants, they are able to feedin different microhabitats within a particular hostplant (Kirk 1997, Mound 1997). The adults and larvaeof Frankliniella occidentalis (Pergande) are mostabundant in ßowers of a variety of plants (Gonzalezand Wilson 1982, Pickett et al. 1988), but these pat-terns can be variable. For example, Higgins (1992)found the majority of F. occidentalis adults in ßowersof greenhouse-grown peppers (Capsicum anuum L.)and cucumbers (Cucumis sativus [L.]) but the major-ity of larvae on the leaves. In contrast, Tavella et al.(1996) showed that 96% of adult and larval Frank-liniella spp. occurred in ßowers of greenhouse-grownpepper. The factors leading to such variation amongstudies are unknown.Although a similar ßower-biaseddistribution to that found by Tavella et al. (1996) isbelieved to occur in Þeld-grown pepper (Funderburket al. 2000, Ramachandran et al. 2001), extensive stud-ies of within-plant patterns of distribution and thebehavioral mechanisms producing those distributionshave not been conducted.Understanding the within plant distribution of

thrips and the underlying behavioral mechanisms is ofparticular relevance for understandinghowcropdam-age occurs, planning efÞcient sampling strategies, and

1 CurrentAddress:DepartmentofZoology,UniversityofOtago,POBox 56, Dunedin, New Zealand.

2 North Florida Research & Education Center, 155 Research Road,Quincy, FL 32351.

3 USDAÐARS Center for Biological Control, Florida A&M Univer-sity, 310 Perry-Paige Bldg., South, Tallahassee, FL 32307 (e-mail:sreitz@nettally.com).

4 DE-NOCIL Crop Protection Ltd., Mumbai 400 079, India.5 Dow Agrosciences, 2608 S. Dundee Blvd., Tampa, FL 33629.6 Department of Entomology and Nematology, University of Flor-

ida, Gainesville, FL 32611.

implementing pest management tactics. The ability ofthrips to deplete yields and render some crops un-economical has demanded that effective pest man-agement practices be developed. Attempts to controlpopulations by regular insecticide applications haveled to the development of resistance tomost classes ofinsecticides (Immaraju et al. 1992). However, sustain-able control may be achieved through the manage-ment of natural enemies. Anthocorid predators (Het-eroptera: Anthocoridae) can be effective biologicalcontrol agents of thrips in greenhouses (Brødsgaardand Enkegaard 1995, Tavella et al. 1996). Althoughtheir effectiveness in open Þeld crops has only re-cently been studied, this recent research has revealedthat Orius insidiosus (Say) is an effective predator ofF. occidentalis, F. tritici (Fitch), and F. bispinosa (Mor-gan) in the ßowers of Þeld-grown pepper (Funder-burk et al. 2000, Ramachandran et al. 2001).Our study had two primary objectives. The Þrst

objective was to determine the within plant distribu-tion of Frankliniella spp. of thrips and the key thripspredator O. insidiosus in Þeld-grown pepper. In addi-tion to behavioral alterations because of the presenceofnatural enemies (Coll and Izraylevich1997), factorssuchaspopulationdensity andmigration areknown toinßuence within-plant distributions among species ofthrips (Shipp and Zariffa 1991, Theunissen and Legu-towska 1991, Higgins 1992). Therefore, we evaluatedpatterns of distributionover a rangeof Þeld conditionsand insecticide use patterns. Insecticideswere used tomanipulate densities of thrips and O. insidiosus todetermine if distributions of thrips and/or O. insidio-sus change in the presence or absence of the other.The second objective was to evaluate the behavioraltime budget of F. occidentalis and F. tritici on pepper.The behaviors of thrips are difÞcult to observe underÞeld conditions; therefore thrips preferences weredetermined under optimal conditions (absence ofpredators, chemicals, and environmental stresses), inthe laboratory.

Materials and Methods

Field Experiments. Sweet pepper (ÔCamelotÕ) wasplanted at the North Florida Research and EducationCenter, Quincy, FL, during the fall 1998 and spring1999, and in the spring 1999 in Riverview, FL. Quincy,GadsdenCounty, is located in the panhandle of north-ern Florida, and Riverview, Hillsborough County, islocated in central Florida.Pepper plants were grown according to standard

horticulturalpractices, inraisedbedscoveredwithwhiteplastic mulch in the fall experiment and black plasticmulch in the spring experiments.Differentmulch colorsare used as a typical production practice to regulate soiltemperature. Each bed consisted of two linear rows ofplants, with a 30-cm spacing between and within rows.Plasticmulch bedswere spaced 0.9m apart. Plants wereirrigatedbasedon theplantneeds, througha trickle tubeplaced at the center of each bed. The design of eachexperimentwas a randomized complete blockwith fourreplicates of each treatment.

Plants in the experiment conducted in the fall of1998 in Quincy were transplanted on 12 August. Plotsizewas two adjacent beds, each 4-m long. Treatmentsincluded an untreated control, esfenvalerate (Asana0.66 SC, DuPont, Wilmington, DE, at 0.057 kg [AI]/ha), and spinosad (Spintor 2 SC, Dow Agrosciences,Indianapolis, IN, at 0.076 kg [AI]/ha). Insecticidetreatments were applied in a water solution at 180liters/ha, with a gas pressurized backpack sprayer Þt-ted with three hollow-cone nozzles (D7Ð45). Sidenozzles were directed straight into the plants of the2-row bed and a nozzle was placed over the top of thebed. Esfenvalerate is a pyrethroid toxic to a broadspectrum of arthropods. Spinosad is derived from thefermentation of Saccharopolyspora spinosa Mertz &Yao and has minimal negative effects on certain spe-cies of natural enemies, such as O. insidiosus (Eger etal. 1998, Elzen et al. 1998, Pietrantonio and Benedict1999, Funderburk et al. 2000). Plots were treated on14, 22, 28 September and 5, 12, and 19 October.Five ßowers, Þve ßower buds, Þve leaves from the

upper half of the plants, and Þve leaves from the lowerhalf of the plants were randomly collected from eachplot 1, 3, and 7 d after spraying. Samples were placedin vials containing 70% ethyl alcohol, and thrips andnatural enemies were extracted under a stereomicro-scope at 40�. Based on taxonomic characters, num-bers of adult males and females of each thrips speciesin each sampleweredetermined. Because larval thripsare not identiÞable to species, they were counted as agroup. The numbers of O. insidiosus and other pred-atory insects were also determined. Representativevoucher specimens are deposited at the University ofFlorida, North Florida Research and Education Cen-ter, Quincy, FL.Plants in the experiments conducted in the springof

1999were transplanted on 22March inQuincy and on29 March in Riverview. Plot size at Quincy was onebed7.5m long; plot size atRiverviewwas twoadjacentbeds each 6-m long. Treatments in both experimentsconsisted of the same chemicals and rates as the fall1998 season, but also included acephate (Orthene75SP, Valent, Walnut Creek, CA, at 0.85 kg [AI]/ha).Acephate is an organophosphate toxic to a broad spec-trum of arthropods through direct contact and sys-temic activity. Dates of insecticide applications were3, 10, 17, 24 May and 1, 8, 15 June in Quincy and 18,25, April and 2, 9, 16 May in Riverview. Samplingprocedures were the same as described previously forthe 1998 season.A split-plot analysis of variance (ANOVA), at � �

0.05, was used to test for signiÞcant effects of insec-ticide treatment and plant part on the seasonal totalnumbers of thrips and O. insidiosus. Before analysis,datawere examined to determine that assumptions forANOVA were satisÞed. Insecticide treatments werethe whole plots and plant parts were the subplots.Means were separated by least squares means (SASInstitute 1990) when the ANOVA was signiÞcant.

Behavioral Assays.Feral adults of F. occidentalis andF. triticiwerecollected fromwildhost plants, andplacedon green bean pods for at least 24 h. Individual adult

1036 ENVIRONMENTAL ENTOMOLOGY Vol. 32, no. 5

females were placed into a petri dish between one leaf,one bud, and one ßower of ÔCamelotÕ sweet pepper.Thrips were observed under a videomicroscope at 40�for 1 h. Pepper plants weremaintained in a greenhouse,andplantparts fromthesameplantwereusedtoobserveboth F. occidentalis and F. tritici on the sameday (n � 12per species). The time spent in different behaviors andlocation of events were recorded, using Observer v. 2.0software(Noldus InformationTechnology Inc., Sterling,VA). Behavior was categorized as feeding, resting, orwalking. A feeding session occurred when an individualstood still with its forelegs apart, antennae still, usuallynodding its head up and down, and probing with itsmandible. Resting includedwhen a thripswas quiescentor grooming. Because thrips rarely engaged in locomo-tion other than walking, all types of locomotion wereclassiÞed as walking. Rarely observed behaviors werecombined into “other” behaviors. Occasionally, thripsmoved deep into the ßower. This time was categorizedas unaccounted, and later combined with “other” timefor analyses.Locationswereonßower, budor leaf, or offplant parts.Time spent in each behavior and location was con-

verted to a proportion of the total observation time. Amultivariate ANOVA (MANOVA) was used to com-pare the time budgets of F. tritici and F. occidentalis onpepper (Aebischer et al. 1993, Cisneros and Rosen-heim 1998). These proportional data were logarithmi-cally transformed to satisfy the assumption of normality.Appropriate univariate ANOVAs were performed tocompareproportionof time spent indifferentbehaviors,such as feeding and locomotion, given that a signiÞcantMANOVAindicated therewasa signiÞcantdifference intime budgets between the species.

Results

Field Experiments. The demography of thrips spe-cies varied by site and time of year. Virtually all of theadults collected at Riverview were F. bispinosa (Fig.1). Other species collected were F. occidentalis and F.

schultzei (Trybom) (collectively �1% of the adults insamples of untreated pepper). Species collected inQuincy were F. occidentalis, F. tritici, and F. bispinosa,but the species composition in untreated pepper dif-fered greatly between the spring and fall seasons (Fig.1). Although each species was found commonly in thespring, F. tritici comprised 84% of the adults collectedin untreated pepper in the fall. Over 99% of all pred-ators collected in untreated pepper of each experi-ment were adults and nymphs of O. insidiosus.Adults and larvae of each Frankliniella species as

well as adults and nymphs ofO. insidiosuswere highlyconcentrated in the pepper ßowers (Table 1; Fig. 2).Numbers on the leaves and buds were always ex-tremely low; therefore, the pooled means for theseplant parts are shown. �98% of the total thrips in thespring at Quincy (n � 6,566) and Riverview (n �10,117) were found in the ßowers (Fig. 2). The per-centage collected in the ßowers in the fall at Quincywas 92%(n� 974).Over 95%of the adults andnymphsof O. insidiosus were collected in the ßowers in thespring at Quincy and Riverview (Fig. 2). There was�71% of the total O. insidiosus collected in the fall atQuincy in ßowers.SigniÞcant interactions between insecticide treat-

ments and plant parts on the abundance of thrips andO. insidiosus occurred in nearly all cases (Table 1). Inthese cases, the interactions reßect the distribution ofthe insects. The abundances of thrips andO. insidiosuson plant parts other than the ßower were so low thatinsecticide treatments would result in minimal differ-ences occurring. However, insecticide treatments didproduce signiÞcant differences in insect abundancesin ßowers (Table 1; Figs. 3-5). In cases where signif-icant plant part by insecticide interactions occurred,simple effects of insecticide treatments within ßowersonly were examined.The insecticides tended to have species-speciÞc ef-

fects. The abundance of F. occidentalis in experimentsatQuincywas signiÞcantly higher inplots treatedwitheither esfenvalerate or acephate than in untreatedplots or plots treated with spinosad (Table 1; Figs. 3and 4). Numbers of F. occidentalis at Quincy weresigniÞcantly lowerboth seasons inpepper treatedwithspinosad compared with untreated pepper (Table 1;Figs. 3 and 4). Possibly because of their variable, lowoverall numbers, no differences among insecticidetreatments were found for F. bispinosa and F. triticiadults, and Frankliniella spp. larvae during the fallseason in Quincy (Table 1; Fig. 3). In contrast duringthe spring season in Quincy, acephate and esfenval-erate reduced numbers of F. tritici and F. bispinosa inthe ßowers compared with untreated plots (Table 1;Fig. 4). A similar pattern for F. bispinosa occurredduring the spring season inRiverview(Table 1;Fig. 5).During the spring season in Quincy, numbers of larvalthrips in plots treated with acephate or esfenvaleratewere signiÞcantly lower than numbers in untreatedplots or plots treated with spinosad (Table 1; Fig. 4).Larvae were least abundant in the spinosad treatedplots. However, larvae were signiÞcantly more abun-

Fig. 1. Composition of Frankliniella spp. adults in sam-ples of untreated pepper in experiments conducted inQuincy (fall: n � 247; spring: n � 920) and Riverview (n �1749), FL.

October 2003 HANSEN ET AL.: POPULATION DYNAMICS AND DISTRIBUTION OF THRIPS IN PEPPER 1037

Table 1. Split plot ANOVA showing effects of plant parts and insecticide treatments on Frankliniella species and O. insidiosusabundance in field pepper

Season Location Species Source of Variation df F P

Fall 1998 Quincy, FL F. occidentalis Insecticide 2 5.13 0.050Block � Insecticidea 6Plant Part 3 43.10 �0.0001Plant Part � Insecticide 6 7.80 0.0001Block � Plant Part (Insecticide)b 27

Fall 1998 Quincy, FL F. tritici Insecticide 2 1.45 0.31Block � Insecticidea 6Plant Part 3 37.04 �0.0001Plant Part � Insecticide 6 1.19 0.34Block � Plant Part (Insecticide)b 27

Fall 1998 Quincy, FL F. bispinosa Insecticide 2 1.92 0.226Block � Insecticidea 6Plant Part 3 17.22 �0.0001Plant Part � Insecticide 6 1.29 0.29Block � Plant Part (Insecticide)b 27

Fall 1998 Quincy, FL Frankliniella Larvae Insecticide 2 1.88 0.23Block � Insecticidea 6Plant Part 3 42.07 �0.0001Plant Part � Insecticide 6 2.68 0.36Block � Plant Part (Insecticide)b 27

Fall 1998 Quincy, FL O. insidiosus Insecticide 2 3.16 0.11Block � Insecticidea 6Plant Part 3 16.41 �0.0001Plant Part � Insecticide 6 4.58 0.002Block � Plant Part (Insecticide)b 27

Spring 1999 Quincy, FL F. occidentalis Insecticide 3 18.36 0.0004Block � Insecticidea 9Plant Part 3 963.47 �0.0001Plant Part � Insecticide 9 27.77 �0.0001Block � Plant Part (Insecticide)b 36

Spring 1999 Quincy, FL F. tritici Insecticide 3 1.50 0.28Block � Insecticidea 9Plant Part 3 632.91 �0.0001Plant Part � Insecticide 9 2.78 0.014Block � Plant Part (Insecticide)b 36

Spring 1999 Quincy, FL F. bispinosa Insecticide 3 6.51 0.012Block � Insecticidea 9Plant Part 3 280.34 �0.0001Plant Part � Insecticide 9 7.94 �0.0001Block � Plant Part (Insecticide)b 36

Spring 1999 Quincy, FL Frankliniella Larvae Insecticide 3 29.03 �0.0001Block � Insecticidea 9Plant Part 3 424.36 �0.0001Plant Part � Insecticide 9 13.45 �0.0001Block � Plant Part (Insecticide)b 36

Spring 1999 Quincy, FL O. insidiosus Insecticide 3 18.59 0.0003Block � Insecticidea 9Plant Part 3 382.25 �0.0001Plant Part � Insecticide 9 15.55 �0.0001Block � Plant Part (Insecticide)b 36

Spring 1999 Riverview, FL F. occidentalis Insecticide 3 4.01 0.046Block � Insecticidea 9Plant Part 3 12.96 �0.0001Plant Part � Insecticide 9 4.69 0.0004

Spring 1999 Riverview, FL F. schultzei Insecticide 3 3.26 0.737Block � Insecticidea 9Plant Part 3 6.40 0.0014Plant Part � Insecticide 9 3.86 0.0017Block � Plant Part (Insecticide)b 36

Spring 1999 Riverview, FL F. bispinosa Insecticide 3 13.79 0.001Block � Insecticidea 9Plant Part 3 223.45 �0.0001Plant Part � Insecticide 9 12.11 �0.0001Block � Plant Part (Insecticide)b 36

Spring 1999 Riverview, FL Frankliniella Larvae Insecticide 3 27.53 �0.0001Block � Insecticidea 9Plant Part 3 302.15 �0.0001Plant Part � Insecticide 9 14.00 �0.0001Block � Plant Part (Insecticide)b 36

Spring 1999 Riverview, FL O. insidiosus Insecticide 3 6.78 0.01Block � Insecticidea 9Plant Part 3 50.58 �0.0001Plant Part � Insecticide 9 5.39 0.0001Block � Plant Part (Insecticide)b 36

a Error term for the whole plot insecticide treatment effect.b Error term for the subplot plant part treatment effect and plant part � insecticide effect.

1038 ENVIRONMENTAL ENTOMOLOGY Vol. 32, no. 5

dant in the untreated plots than in any of the insec-ticide treated plots in Riverview (Table 1; Fig. 5).The abundance of O. insidiosus was affected by

insecticide treatments (Table 1; Figs. 3Ð5). Esfenval-erate and acephate suppressed populations of O. in-sidiosus in the ßowers in each experiment comparedwith untreated plots. Spinosad had a less detrimentalimpact on O. insidiosus compared with these otherinsecticides.

Behavioral Assays. In the laboratory choice exper-iments, females of both F. occidentalis and F. triticidemonstrated a preference for ßowers comparedwithbuds and leaves on pepper (Fig. 6). There was asigniÞcant difference in the overall time budgets forthe two species (Wilks lambda � 0.428, P � 0.05). Onaverage, females of F. occidentalis spent 70.1� 5.3% ofthe time on the ßower compared with 59.6 � 8.7% ofthe time for females of F. tritici. Females of F. occi-

dentalis did spend more time on the buds and leaves(17.0 � 2.5% of the time) than did females of F. tritici(8.1 � 2.2%). In turn, the females of F. tritici weremoreactive than thoseofF. occidentalis(Fig. 6;F�5.1,df � 1,22; P � 0.034). F. tritici were moving amongplant parts 22.9 � 8.0% of the time compared with7.2 � 1.3% of the time for F. occidentalis.Although no difference was observed between the

feeding behaviors of F. occidentalis and F. tritici, thetwo species differed in their utilization of host plantresources. F. occidentalis females fed almost threetimes longer (4.6� 0.4% of the time) than did F. triticifemales (1.7 � 0.2%; F � 6.70, df � 1, 22; P � 0.017).The females of F. occidentalis fed on all of the pepperplant parts (i.e., ßowers, buds and leaves) (Fig. 6). Incontrast, the females of F. triticiwere observed to feedalmost exclusively (�93%of total feeding time)on theßowers. Feeding sessions lasted from 2 s to �1.5 min.

Fig. 2. Proportion of Frankliniella spp. and O. insidiosus on the ßowers, buds, upper leaves, and lower leaves of pepperfrom samples collected in experiments conductedduring the fall 1998 inQuincy, FL(A), andduring the spring 1999 inQuincy(B) and Riverview, FL (C). Numbers in parentheses are totals for each taxon.

October 2003 HANSEN ET AL.: POPULATION DYNAMICS AND DISTRIBUTION OF THRIPS IN PEPPER 1039

Discussion

The different compositions of Frankliniella speciesfound in our study (Fig. 1) reßect geographical dif-ferences between northern and central Florida in thedominance of the species, and seasonal differences.Previous studies have reported the abundance ofthrips species oncropandwildplanthosts in these twogeographical areas. In central Florida, F. bispinosa isthe predominant species; F. occidentalis is present butnot abundant, and F. tritici is extremely rare (Childersand Beshear 1992, Childers et al. 1998). In northernFlorida, F. occidentalis, F. tritici, and F. bispinosa are allabundant, while F. schultzei is absent (Salguero-Navaset al. 1991, Chellemi et al. 1994, Toapanta et al. 1996).Also in northern Florida, the proportion of F. occiden-talis in the species complex becomes extremely low inthe fall, and F. tritici is the predominant species at thattime of year (Ramachandran et al. 2001, Reitz 2002).Pepper is a reproductive host for F. occidentalis, F.

tritici,andF.bispinosa(Funderburketal.2000),although

its suitability may differ for each species (Frantz et al.1995, Nuessly and Nagata 1995). Nevertheless, the dif-ference in species composition between northern andcentralFloridacouldnotbeaccountedforbydifferentialreproductive success only. These differences may beattributed either to competition between the Frank-liniella species or to other factors that regulate popula-tion abundance. Seasonal differences in the abundanceof these thrips may be related to activity of natural en-emies. Ramachandran et al. (2001) showed that localpopulations of thrips in pepper between late spring andearly fall are lowbecauseofcontinuedsuppressionbyO.insidiosus,andtheyspeculate thatpopulationsofF.bispi-nosa, F. occidentalis, and F. tritici are regulated by thispredator during summer and fall in northern Florida.Decreasing daylength, as occurs in the late fall and win-ter, triggers a reproductivediapause inO. insidiosus(Ru-berson et al. 1991). This diapause may allow for a resur-genceof thripspopulations in thespringwhenadultscanbe found on many crop and wild plant hosts (Chellemi

Fig. 3. Meannumber(�SEM)ofFrankliniella spp. adults and larvae, andO. insidiosus inuntreatedand insecticide-treatedpepper in the experiment conducted during the fall of 1998 in Quincy, FL. For the abundance of each taxon within ßowers,insecticide treatment means marked with the same letter are not signiÞcantly different (P � 0.05, least squares means [SAS1990]). Data for buds and upper and lower leaves have been pooled for presentation. Data are pooled over all sample dates.

1040 ENVIRONMENTAL ENTOMOLOGY Vol. 32, no. 5

etal. 1994,Toapantaetal. 1996). Incontrast,O. insidiosusdo not appear to enter diapause duringwinter in centralFlorida, which may prevent buildup of F. occidentalispopulations (Bottenberg et al. 1999). However, F. bispi-nosa may be better able to avoid predation than F. occi-dentalis because of its greater mobility (Ramachandranetal.2001,Reitzetal.2002),andthereforeitspopulationsare able to persist.Responses to insecticides alsowere species-speciÞc.

Although signiÞcant insecticide by plant part interac-tions existed, changes in insect abundance were con-sistently greater for ßowers than for other plant parts,most likely as a result of the insects being so concen-trated in the ßowers. F. occidentalis populations wereactually higher in plots treated with esfenvalerate oracephate than in untreated plots. This may stem froma combination of insecticide resistance on the part ofF. occidentalis (Immaraju et al. 1992, Broadbent andPree 1997) and exclusion of O. insidiosus by thoseinsecticides (Funderburk et al. 2000, Ramachandranet al. 2001). Although little is known of the suscepti-bility of F. bispinosa and F. tritici to insecticides, our

results suggest that these species aremore susceptibleto broad-spectrum insecticides than is F. occidentalis.Despite these geographic and temporal differences

in the Frankliniella species composition and abun-dance in pepper, and variable responses to insecti-cides, the within plant distribution of thrips remainedconstant. Extensive populations of F. occidentalis, F.tritici and F. bispinosa did not develop on the leavesandbuds of Þeld pepper in the current study, but largepopulations were in the ßowers. Similar ßower-biaseddistributions for populations of F. occidentalis occur ingreenhouse pepper (Shipp andZariffa 1991, Tavella etal. 1996). As our behavioral study shows, adults of F.occidentalis and F. tritici are able to move throughoutaplant, but theyhavea strongattraction to theßowers.Oviposition may occur, and larvae can develop on theleaves of some plant hosts, such as cucumber (Kiers etal. 2000), but even on such hosts the vast majority ofindividuals reside in the ßowers.Anthophilous thrips, such as Frankliniella spp. use

both visual and olfactory cues to locate hosts (Teulonet al. 1993, Terry 1997, Teulon et al. 1999). Low UV

Fig. 4. Meannumber(�SEM)ofFrankliniella spp. adults and larvae, andO. insidiosus inuntreatedand insecticide-treatedpepper in the experiment conducted during the spring of 1999 inQuincy, FL. For the abundance of each taxonwithin ßowers,insecticide treatment means marked with the same letter are not signiÞcantly different (P � 0.05, least squares means [SAS1990]). Data for buds and upper and lower leaves have been pooled for presentation. Data are pooled over all sample dates.Note different scale for O. insidiosus.

October 2003 HANSEN ET AL.: POPULATION DYNAMICS AND DISTRIBUTION OF THRIPS IN PEPPER 1041

white, blue, and yellow tend to be the most attractivecolors (Kirk 1984). In addition to color, size, shape,and background contrast inßuence host location. Cer-tain plant volatiles, such as benzenoids and monoter-penes, are attractive to F. occidentalis (Teulon et al.1993, 1999; Koschier et al. 2000). These types of visualand olfactory cues are associatedwith ßowers ofmanyplant species, including peppers.Adult females of F. tritici spentmore time than adult

females of F. occidentalis moving between differentplant parts of pepper. These results support the con-clusions of Ramachandran et al. (2001) that localmovement by F. occidentalis between ßowers of Þeldpepper was limited, whereas F. tritici moved rapidlybetween the ßowers. This rapid movement may assistF. tritici in avoiding predation, but adults of O. insid-iosus also are capable ofmoving rapidly among pepperßowers.O. insidiosusnymphs alsomust be attracted toßowers. Eggs of O. insidiosus are laid mostly on otherplant parts of greenhouse pepper (van denMeirackerand Sabelis 1993), yet we found populations of O.

insidiosus nymphs and adults were aggregated withtheir prey in the ßowers of Þeld pepper. Shipp et al.(1992) reported the same pattern of aggregation ofO.insidiosus in the ßowers of greenhouse pepper.The ecological cues that attract O. insidiosus to

prey-infested patches are not well understood. A por-tion of prey searching byOrius spp. relies on olfactoryperception of chemical cues emitted by herbivore-infested plants (Reid and Lampman 1989, VanLaer-hoven et al. 2000). Constitutive odors of thrips also areimportant as prey-searching cues. An alarm phero-mone produced by larvae of F. occidentalis also isattractive toO. tristicolorWhite (Teerling et al. 1993).There was no evidence in our experiments that the

presence of other thrips species or O. insidiosus orinsecticides affected the within plant behavior of F.tritici, F. bispinosa, and F. occidentalis on Þeld pepper.While various methods of sampling for thrips are avail-able (e.g., Shipp and Zariffa 1991, Shipp et al. 1998),estimatesofpopulationsbasedonnumbers intheßowersofpepperseemtobest reßect the localdynamicsof these

Fig. 5. Meannumber(�SEM)ofFrankliniella spp. adults and larvae, andO. insidiosus inuntreatedand insecticide-treatedpepper in the experiment conducted during the spring of 1999 in Riverview, FL. For the abundance of each taxon withinßowers, insecticide treatmentmeansmarkedwith the same letter are not signiÞcantly different (P � 0.05, least squaresmeans[SAS 1990]). Data for buds and upper and lower leaves have been pooled for presentation. Data are pooled over all sampledates. Note different scale for O. insidiosus.

1042 ENVIRONMENTAL ENTOMOLOGY Vol. 32, no. 5

Frankliniella thrips and O. insidiosus in open Þeld situa-tions. This method would be best for assessing the ben-eÞts of biological control in scouting programs based onpredator to prey ratios (Sabelis and Van Rijn 1997,Funderburk et al. 2000, Ramachandran et al. 2001).

Acknowledgments

We thank Brent Brodbeck (University of Florida) andNorm Leppla (University of Florida) for comments on themanuscript. We are indebted to John Petti, Andrew Brown,and Julianne Stavisky for their invaluable technical assis-tance. The contents of this report comprise a portion of athesis submitted by EAH as partial fulÞllment of the require-ments for a M.Sc. degree at the University of Florida. Con-tribution from the Florida Agricultural Experiment StationJournal Series number R-07764.

References Cited

Aebischer, N. J., P. A. Robertson, and R. E. Kenward. 1993.Compositional analysis of habitat use from animal radio-tracking data. Ecology 74: 1313Ð1325.

Bottenberg, H., G. Frantz, and H. Mellinger. 1999. Refugeand cover crop plantings for beneÞcial insect habitats.Proc. Fla. State Hort. Soc. 112: 339Ð341.

Broadbent, A. B., and D. J. Pree. 1997. Resistance to insec-ticides in populations of Frankliniella occidentalis (Per-gande)(Thysanoptera: Thripidae) fromgreenhouses in theNiagara region of Ontario. Can. Entomol. 129: 907Ð913.

Brødsgaard, H. F., and A. Enkegaard. 1995. Interactionsamong polyphagous anthocorid bugs used for thrips con-trol and other beneÞcials in multi-species biological pestmanagement systems. Med. Fac. Landbouww. Univ.Gent. 60: 893Ð900.

Chellemi, D. O., J. E. Funderburk, and D. W. Hall. 1994.Seasonal abundance of ßower-inhabiting Frankliniellaspecies (Thysanoptera: Thripidae) on wild plant species.Environ. Entomol. 23: 337Ð342.

Childers, C. C. 1997. Feeding and oviposition injuries toplants, pp. 505Ð537. InT.Lewis [ed.],Thrips as croppests.CAB International, New York.

Childers, C. C., and R. J. Beshear. 1992. Thrips Thysan-optera species associated with developing citrus ßowersin Florida and a key to adult Terebrantian females. J.Entomol. Sci. 27: 392Ð412.

Childers, C. C., S. Nakahara, and R. J. Beshear. 1998. Thys-anoptera collected during bloom on white and coloreddisposable sticky cards in Florida citrus groves. J. Ento-mol. Sci. 33: 49Ð71.

Cisneros, J. J., and J. A. Rosenheim. 1998. Changes in theforaging behavior, within-plant vertical distribution, andmicrohabitat selection of a generalist insect predator: anage analysis. Environ. Entomol. 27: 949Ð957.

Coll, M., and S. Izraylevich. 1997. When predators also feedon plants: effects of competition and plant quality onomnivore-prey population dynamics. Ann. Entomol. Soc.Am. 90: 155Ð161.

Eger, J. E., J. Stavisky, and J. E. Funderburk. 1998. Com-parative toxicity of spinosad to Frankliniella spp. (Thys-anoptera: Thripidae), with notes on a biossay technique.Fla. Entomol. 81: 547Ð551.

Elzen, G. W., P. J. Elzen, and E. G. King. 1998. Laboratorytoxicity of insecticide residues to Orius insidiosus, Geo-coris punctipes, Hippodamia convergens, and Chrysoperlacarnea. Southwest. Entomol. 23: 335Ð342.

Frantz, G., F. Parks, and C. Mellinger. 1995. Thrips popu-lation trends inpeppers in southwestFlorida, pp. 111Ð114.In B. L. Parker, M. Skinner and T. Lewis [eds.], Thripsbiology and management. Plenum, New York.

Funderburk, J., J. Stavisky, and S. Olson. 2000. Predation ofFrankliniella occidentalis (Thysanoptera: Thripidae) inÞeld peppers by Orius insidiosus (Hemiptera: Anthoco-ridae). Environ. Entomol. 29: 376Ð382.

Gonzalez, D., and L. T. Wilson. 1982. A foodweb approachto economic thresholds: a sequence of pests/predaceousarthropods in California cotton. Entomophaga 27: 31Ð43.

Higgins, C. J. 1992. Western ßower thrips (Thysanoptera:Thripidae) in greenhouses: population dynamics, distri-bution on plants and associations with predators. J. Econ.Entomol. 85: 1891Ð1903.

Immaraju, J. A., T.D. Paine, J. A. Bethke,K. L.Robb, and J. P.Newman. 1992. Western ßower thrips (Thysanoptera:Thripidae) resistance to insecticides in coastal Californiagreenhouses. J. Econ. Entomol. 85: 9Ð14.

Kiers, E., W. J. de Kogel, A. Balkema-Boomstra, and C. Mol-lema. 2000. Flower visitation and oviposition behaviourof Frankliniella occidentalis (Thysan., Thripidae) on cu-cumber plants. J. Appl. Entomol. 124: 27Ð32.

Kirk, W.D.J. 1984. Ecologically selective coloured traps.Ecol. Entomol. 9: 35Ð41.

Kirk, W.D.J. 1997. Feeding, pp. 119Ð174. In T. Lewis [ed.],Thrips as crop pests. CAB International, New York.

Koschier, E. H., W. J. de Kogel, and J. H. Visser. 2000. As-sessing the attractiveness of volatile plant compounds towestern ßower thrips Frankliniella occidentalis. J. Chem.Ecol. 26: 2643Ð2655.

Mound, L. A. 1997. Biological diversity, pp. 197Ð216. In T.Lewis[ed.],Thrips as croppests.CABInternational,NewYork.

Nuessly, G. S., and R. T. Nagata. 1995. Pepper varietal re-sponse to thrips feeding, pp. 115Ð118. In B. L. Parker, M.Skinner and T. Lewis [eds.], Thrips biology andmanage-ment. Plenum, New York.

Pickett, C. H., L. T. Wilson, and D. Gonzalez. 1988. Popu-lationdynamics andwithin-plant distributionof thewest-ern ßower thrips (Thysanoptera: Thripidae), an early-season predator of spidermites infesting cotton. Environ.Entomol. 17: 551Ð559.

Fig. 6. Proportion of time (mean� SEM) spent engagedin different behaviors by adult female F. occidentalis and F.tritici in laboratory choice experiments.

October 2003 HANSEN ET AL.: POPULATION DYNAMICS AND DISTRIBUTION OF THRIPS IN PEPPER 1043

Pietrantonio, P. V., and J. H. Benedict. 1999. Effect of newcotton insecticide chemistries, tebufenozide, spinosadand chlorfenapyr, on Orius insidious and two Cotesiaspecies. Southwest. Entomol. 24: 21Ð29.

Ramachandran, S., J. Funderburk, J. Stavisky, and S. Olson.2001. Population abundance and movement of Frank-liniella species andOrius insidiosus in Þeld pepper. Agric.Forest Entomol. 3: 1Ð10.

Reid, C. D., and R. L. Lampman. 1989. Olfactory responsesof Orius insidiosus (Hemiptera: Anthocoridae) to vola-tiles of corn silks. J. Chem. Ecol. 15: 1109Ð1116.

Reitz, S. R. 2002. Seasonal and within plant distribution ofFrankliniella species in north Florida tomatoes. Fla. En-tomol. 85: 431Ð439.

Reitz, S. R., J. E. Funderburk, E. A. Hansen, I. Baez, S.Waring, and S. Ramachandran. 2002. InterspeciÞc vari-ation in behavior and its role in thrips ecology, pp. 133Ð140. In R. Marullo and L. A. Mound [eds.], Thrips andTospoviruses: Proceedings of the 7th International Sym-posium on Thysanoptera. Australian National Insect Col-lection, Canberra, Australia.

Ruberson, J. R., L. Bush, and T. J. Kring. 1991. Photoperi-odic effect ondiapause induction anddevelopment in thepredator Orius insidiosus (Heteroptera: Anthocoridae).Environ. Entomol. 20: 786Ð789.

Sabelis, M. W., and P. C. J. Van Rijn. 1997. Predation bymites and insects, pp. 259Ð354. In T. Lewis [ed.], Thripsas crop pests. CAB International, Wallingford, UnitedKingdom.

Salguero-Navas, V. E., J. E. Funderburk, R. J. Beshear, S. M.Olson, and T. P. Mack. 1991. Seasonal patterns of Frank-liniella spp. (Thysanoptera: Thripidae) in tomatoßowers.J. Econ. Entomol. 84: 1818Ð1822.

SAS Institute. 1990. SAS/STAT UserÕs guide, vol. 2. SASInstitute, Cary, NC.

Sether, D. M., and J. D. Deangelis. 1992. Tomato spottedwilt virus host list, bibliography, Special Report 888. Or-egon State Experiment Station, Corvallis, OR.

Shipp, J. L., and N. Zariffa. 1991. Spatial patterns and sam-pling methods for western ßower thrips (Thysanoptera:Thripidae) on greenhouse sweet pepper. Can. Entomol.123: 989Ð1000.

Shipp, J. L., N. Zariffa, and G. Ferguson. 1992. Spatial pat-terns and sampling methods for Orius spp. (Hemiptera:

Anthocoridae) on greenhouse sweet pepper. Can. Ento-mol. 124: 887Ð894.

Shipp, J. L., M. R. Binns, X. Hao, and K. Wang. 1998. Eco-nomic injury levels for western ßower thrips (Thysan-optera: Thripidae) greenhouse sweet pepper. J. Econ.Entomol. 91: 671Ð677.

Tavella, L., A. Alma, A. Conti, and A. Arizone. 1996. Eval-uation of the effectiveness of Orius spp. in controllingFrankliniella occidentalis. Acta Hortic. 431: 499Ð506.

Teerling, C. R., D. R. Gillespie, and J. H. Borden. 1993.Utilization of western ßower thrips alarm pheromone asa prey-Þnding kairomone by predators. Can. Entomol.125: 431Ð437.

Terry, L. I. 1997. Host selection, communication and repro-ductive behavior, pp. 65Ð118. In T. Lewis [ed.], Thrips ascrop pests. CAB International, New York.

Teulon, D.A.J., D. R. Penman, and P.M.J. Ramakers. 1993.Volatile chemicals for thrips (Thysanoptera: Thripidae)host-Þnding andapplications for thrips pestmanagement.J. Econ. Entomol. 86: 1405Ð1415.

Teulon, D.A.J., B. Hollister, R. C. Butler, and E. A. Cameron.1999. Colour and odour responses of ßying westernßower thrips: wind tunnel and greenhouse experiments.Entomol. Exp. Appl. 93: 9Ð19.

Theunissen, J., and H. Legutowska. 1991. Thrips tabaci Lin-deman (Thysanoptera: Thripidae) in leek: within-plantdistribution. J. Appl. Entomol. 112: 309Ð316.

Toapanta, M., J. Funderburk, S. Webb, D. Chellemi, and J.Tsai. 1996. Abundance of Frankliniella spp. (Thysan-optera: Thripidae) on winter and spring host plants. En-viron. Entomol. 25: 793Ð800.

Ullman, D. E., J. L. Sherwood, and T. L. German. 1997.Thrips as vectors of plant pathogens, pp. 539Ð565. In T.Lewis[ed.],Thrips as croppests.CABInternational,NewYork.

van den Meiracker, R.A.F., and M. W. Sabelis. 1993. Ovi-position sites of Orius insidiosus in sweet pepper. Bull.IOBC/WPR S. 16: 109Ð112.

VanLaerhoven, S., D. R. Gillespie, and R. R. McGregor.2000. Leaf damage andprey typedetermine searcheffortin Orius tristicolor. Entomol. Exp. Appl. 97: 167Ð174.

Received for publication 14 January 2003; accepted 11 June2003.

1044 ENVIRONMENTAL ENTOMOLOGY Vol. 32, no. 5