Nicolas Desneux.annu. Rev. Entomol. 2007

ANRV297-EN52-05 ARI 21 November 2006 10:17

The Sublethal Effects ofPesticides on BenecialArthropodsNicolas Desneux,1 Axel Decourtye,2

and Jean-Marie Delpuech31Department of Entomology, University of Minnesota, St. Paul, Minnesota 55108;email: [email protected], Institut Claude BourgelatENVL, 69280 Marcy Letoile, France;email: [email protected] de Biometrie et Biologie Evolutive (UMR 5558); CNRS; UniversiteLyon 1, 69622, Villeurbanne Cedex, France; email: [email protected]

Annu. Rev. Entomol. 2007. 52:81106

First published online as a Review inAdvance on July 14, 2006

The Annual Review of Entomology is online atento.annualreviews.org

This articles doi:10.1146/annurev.ento.52.110405.091440

Copyright c 2007 by Annual Reviews.All rights reserved

0066-4170/07/0107-0081$20.00

Key Words

ecotoxicology, insecticide, behavior, honey bee, natural enemy

AbstractTraditionally, measurement of the acute toxicity of pesticides to ben-ecial arthropods has relied largely on the determination of an acutemedian lethal dose or concentration. However, the estimated lethaldose during acute toxicity tests may only be a partial measure ofthe deleterious effects. In addition to direct mortality induced bypesticides, their sublethal effects on arthropod physiology and be-havior must be considered for a complete analysis of their impact.An increasing number of studies and methods related to the identi-cation and characterization of these effects have been published inthe past 15 years. Review of sublethal effects reported in publishedliterature, taking into account recent data, has revealed new insightsinto the sublethal effects of pesticides including effects on learn-ing performance, behavior, and neurophysiology. We characterizethe different types of sublethal effects on benecial arthropods, fo-cusing mainly on honey bees and natural enemies, and we describethe methods used in these studies. Finally, we discuss the potentialfor developing experimental approaches that take into account thesesublethal effects in integrated pest management and the possibilityof integrating their evaluation in pesticide registration procedures.

81

Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


Pollinator: anorganism that carriespollen from oneower to another

IPM: integratedpest management

LD50: lethal dose50%

Sublethal effect: aneffect (physiologicalor behavioral) onindividuals thatsurvive an exposureto a pesticide (thepesticidedose/concentrationcan be sublethal orlethal)

Sublethaldose/concentration:a dose or aconcentrationdened as inducingno statisticallysignicant mortalityin the experimentalpopulation

INTRODUCTION ANDDEFINITIONS

For the past 20 years, the effects of pesticideson benecial arthropods have been the sub-ject of an increasing number of studies, andthe potential effects have been reviewed sev-eral times (24, 62, 128). Two groups of organ-isms, natural enemies and pollinators, havereceived the most attention in this regard be-cause of their value in integrated pest manage-ment (IPM) (131) and pollination processes(105), respectively.

Methods to test the side effects of pesti-cides have been developed as a function of thebenecial arthropods and pesticides studied.In each country, regulatory insect risk assess-ment related to agrochemical use and regis-tration follows specic guidelines (EuropeanCouncil Directive 91/414 in Europe, and theFederal Insecticide Fungicide and Rodenti-cide Act in the United States). For a longtime, the classical laboratory method for esti-mating the side effects of chemicals on bene-cial arthropods was to determine a medianlethal dose (LD50) or lethal concentration(LC50) estimate. In a second step, the ef-fects of pesticides on benecial arthropodswere examined further by running selectivitytests (pest/benecial arthropods) to identifyproducts with the lowest nontarget activity(24). However, estimation of selectivity wasbased on LD50 values, and side effects of pes-ticides on benecial arthropods still occurredbecause of the lack of attention to sublethaleffects. Because of the increasing economicimportance of benecial arthropods in agri-culture and the recognition of limitations as-sociated with traditional methods for study-ing sublethal effects of pesticides (80), agrowing body of literature is aimed at ad-dressing this issue. Now, it is importantto step back and review what these studieshave documented to determine the directionsof future studies and applications. Sublethaleffects are dened as effects (either physio-logical or behavioral) on individuals that sur-vive exposure to a pesticide (the pesticide

dose/concentration can be sublethal or lethal).A sublethal dose/concentration is dened asinducing no apparent mortality in the experi-mental population.

We review the sublethal effects of pesti-cides on benecial arthropods reported in thepublished literature and divide these effectsinto two major groups: physiology and be-havior. We focus on the side effects and noton the indirect effects of pesticides, such ashabitat destruction and damage to nesting,oviposition, resting, and mating sites. This re-view aims to (a) provide a better understand-ing of the different types of sublethal effectsassociated with pesticide exposure, (b) clarifythe range of methods used to address sub-lethal effects and permit new insights intothe development of better experimental ap-proaches, (c) determine if evaluation of theseeffects could be included in the pesticides reg-istration process, and (d ) elucidate the pos-sible consequences of the sublethal effectsof pesticides on the efciency of benecialarthropods (pest limitation or pollination) andcommunity dynamics.

PHYSIOLOGICAL EFFECTS

General Biochemistry andNeurophysiology

Studies on effects of pesticides on insect bio-chemistry have been conducted with bothpollinator and natural enemy models. Morein-depth studies have been performed us-ing honey bees primarily because more isknown about their biochemical systems. Ex-periments on bee physiology have been donemainly by measuring the activity of enzymesafter or during exposure to pesticides. Af-ter injection of emerging honey bees inthe laboratory, fenitrothion (organophospho-rus) and cypermethrin (pyrethroid) led todecreases in Na+/K+ ATPase and acetyl-cholinesterase (AChE) activities (12). Re-lated glycemic disorders were also linkedto enzyme inhibition. Na+/K+ ATPase is a

82 Desneux Decourtye Delpuech

Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


transmembrane enzyme that releases energynecessary for cell metabolism and establishesthe ionic concentration balance that main-tains the cell potential. Thus, the inhibitionof Na+/K+ exchange provoked by pyrethroidsmight affect a wide range of cellular functions.For example, the pyrethroid deltamethrincauses marked dysfunctions in myocardialcells. Indeed, Papaefthimiou & Theophilidis(97) have demonstrated the cardiotoxicity ofdeltamethrin using intracellular recordingsfrom the myocardial cells of the semi-isolatedhearts of honey bee. The frequency and theforce of spontaneously generated cardiac con-tractions were modied by deltamethrin. Theimidazole fungicide prochloraz had a simi-lar impact, but its effects were more intense.When prochloraz and deltamethrin are com-bined there is a synergistic interaction. Thejoint effects of both compounds were also in-vestigated on honey bee thermoregulation byinfrared thermography. When associated withprochloraz, deltamethrin elicited a joint hy-pothermia at doses that did not induce a sig-nicant effect on thermoregulation when usedalone (133). One hypothesis was that imida-zoles delayed the metabolism, detoxication,and excretion of pyrethroids by inhibition ofmicrosomal oxidation and thus enhanced thetoxicity of the pyrethroid to the honey bees(98). However, the results of sublethal toxic-ity suggest other mechanisms for synergistictoxic effects, such as combined action on acommon target (97).

In contrast to studies conducted on honeybees, few studies have investigated the effectsof pesticides on the general biochemistryand enzymatic processes in natural enemies.In a study aiming to use enzyme activityas a biomarker of sublethal exposure toinsecticides, Rumpf et al. (109) demonstratedthat acute toxicity tests (LD50 determination)could miss sublethal perturbations involvingeffects on enzymes. This study (on lacewings)showed that the correlation between the de-gree of AChE and glutathione-S-transferaseinhibition and corresponding mortalitycaused by a given insecticide (ve classes

CO: cytochromeoxidase

tested) was toxin specic as well as speciesspecic. The inhibition of AChE could leadto general perturbation in all systems becauseit is a major component in all synaptictransmission (74), especially when inhibitioncontinues for a long time after exposure. Forexample, eight days were required after a 24-or 48-h exposure to the organophosphorusdiazinon and chlorpyrifos for AChE inhibi-tion in wolf spiders (Lycosidae) to disappear(132). Thus, pesticide effects on importantenzyme systems cannot be extrapolated ordeduced from LD50 values.

Effects on neurophysiology have also beendescribed. The metabolic activity in the honeybee brain was investigated using cytochromeoxidase (CO) histochemistry. Because CO isthe terminal enzyme in the electron transportchain of the mitochondrial respiratory pro-cesses, histochemistry is used as an endoge-nous metabolic marker for neuronal activity(145). Using CO histochemistry to carry outmetabolic mapping of discrete brain regions,Armengaud et al. (7) showed that CO histo-chemistry could be used to identify the tar-get structures of cholinergic ligands in thehoney bee brain, particularly in the case ofthe neonicotinoid imidacloprid. In a behav-ioral and histochemical analysis of the effectof imidacloprid on olfactory learning in thehoney bee, oxidative metabolism in the ca-lyces of the mushroom body was increased af-ter treatment (27). In parallel, the impairmentof olfactory memory by imidacloprid was ob-served. The structure-specic increase of COactivity in the brain observed after treatmentsuggests that imidacloprid impairs olfactorymemory by a physiological effect at the level ofthe mushroom body, which is reported to havean essential role in olfactory memory (51a).

Results demonstrating negative effects ofpesticide at the biochemical and neurophysi-ological levels are difcult to interpret becausetheir consequences at individual or populationlevels are often unknown. However, this is notthe case with studies concerning communica-tion between insects or the development ofbenecial arthropod larvae.

www.annualreviews.org Sublethal Effects of Pesticides 83

Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


IGR: insect growthregulator

Parasitoid: an insectthat completes itslarval developmentwithin the body ofanother insecteventually killing itand is free-living asan adult

JHA: juvenilehormone analog

Development

Sublethal effects on larval development mayresult from perturbations in development ofneural tissues by neurotoxic substances. Giventhe importance of the cholinergic systemin insect development (117), many kinds ofsublethal effects are possible. Insect growthregulators (IGRs) are also likely to perturbthe development of benecial arthropods. In-deed, IGRs are commercial hormone mim-ics that disrupt molting (juvenile hormoneor ecdysone mimics) and cuticle formation(chitin inhibitors) and more generally act onendocrine systems (49). Studies reporting pes-ticide impacts on the development of naturalenemies typically differ with the biology ofthe experimental subject (i.e., predators ver-sus parasitoids). Studies using parasitoids of-ten report effects on adult emergence from thepupal stage (75, 110, 114). Adult emergencehas also been studied for the lacewing preda-tor Mallada signatus exposed to the botanicalinsecticide azadirachtin A (AzaA) in the pu-pal stage (101). In most of these studies, how-ever, it has remained unclear whether reducedadult emergence is related to the direct lethaleffects of pesticides or if other perturbationssuch as organ malformation are primarily re-sponsible. Other studies have further clariedthis subject. Schneider et al. (115) reporteda decrease in emergence from parasitizedhost after exposure to spinosad (spinosyns) inthe endoparasitoid Hyposoter didymator; how-ever, they related their ndings to the appar-ent inability of the larvae to produce silk, anecessary material for cocooning. A similarnding has been reported for the predatorChrysoperla carnea following fenoxycarb (juve-nile hormone analog, JHA) exposure (13).

Another parameter often reported in asso-ciation with the effects of pesticides on insectdevelopment is the developmental rate. De-velopmental rate can have a large impact on anatural enemys intrinsic rate of increase (rm)and phenological synchrony with the host orprey. An increase in developmental rate couldpresent a signicant disadvantage for a para-sitoid if it disrupts synchrony with a critical

window of susceptibility in the host. Fenoxy-carb is reported to prolong the developmenttime of the predator Chrysoperla rulabris inall stages but the pupae (81). Consoli et al.(21) reported that Trichogramma pretiosum pu-pae displayed a higher sensitivity to pesticidesin terms of development time than did eithereggs, larvae, or prepupae. Increases in devel-opment time have also been reported in otherpredators exposed to neurotoxic insecticides(5355) and on parasitoids exposed to botani-cal insecticides (18). The impact of pesticideson development time may also be a function ofgender. In the pentatomid predator Supputiuscincticeps, exposure to permethrin decreaseddevelopment time for females, whereas thistime increased for males (146). Malforma-tions also occur in natural enemies after ex-posure to pesticides and may lead to reduc-tion in predator or parasitoid efciency andtness. In a study describing sublethal effectsin the predators Coccinella septempunctata andChrysoperla carnae exposed to AzaA-treatedaphids, Ahmad et al. (3) reported morpho-logical deformities and thus calculated a rateof deformity to express all the visible defor-mations in adult individuals exposed to pesti-cides during the larval stage. Although thisstatistic did not provide a qualitative mea-sure of deformation, it could be useful forestimating the potential developmental ef-fects of pesticides. In another study, the hindtibia length was shorter in males of the para-sitoid Cotesia plutellae emerging from Plutellaxylostella larvae that fed on cabbage treatedwith botanical insecticide (18). Such a malfor-mation in males may lead to a strong reduc-tion in their tness because their capacity tomate is correlated with overall body size (69).In the reduviid predator Rhynocoris kumarii,adults developed severe abnormalities in thealimentary canal, testis, and ovary whentreated with the organophosphorus insec-ticides monocrotophos, dimethoate, methylparathion, or quinalphos at sublethal doses(57). The authors determined that abnormal-ities in the alimentary canal were due to lysisof intercellular cementing material, pycnotic


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


nuclei, vacuolated cells, obliteration of theperitrophic membrane, and exfoliation ofcells. In the same study, pesticides causedpredator size reduction, sperm cell distortion,vacuolated spermatocytes in the testis, andcrumpled follicular epithelium and vacuoliza-tion of the germarium in the ovaries. Malfor-mation of ovaries can also occur in parasitoidsexposed to IGRs (115).

For hymenopterous social pollinators, per-turbations in larval development must be seenas a major threat for colonies. Reductions inbrood and numbers of emerging Apis melliferamay be more damaging to colony health thanthe loss of foragers, because exibility in thedivision of labor can replace foragers if thereare sufcient brood and nurse bees (128).IGRs can interfere with development, partic-ularly when exposure occurs during the lar-val stage. Most observations on the effects ofIGRs rely on measurements of brood quantitytaken under eld or semield conditions. Onecommon approach uses the number of hivecells containing different bee brood stages todetect possible adverse effects, because distur-bances to larval development could be accom-panied by failure to emerge. Oral exposureof worker bees with diubenzuron (IGR) re-duced brood surface area (17). The impact offenoxycarb on brood was manifested by thepresence of malformed larvae or pupae, whichwere ultimately found dead in front of the hive(42). However, opposite effects were reportedon Bombus terrestris exposed to imidacloprid(126), and these authors assumed that the re-duced larval ejection rate in treated groupswas due to reduction of brood size (due tomortality). Because it is impossible to disso-ciate effects on brood size from direct effectson larvae using these measures, an accurateassessment of the effects of pesticide exposureon larvae cannot be obtained.

Few quantitative studies have assessed theimpact of IGRs on larval development, possi-bly because of difculties associated with rear-ing larvae under consistent conditions. How-ever, a new in vitro brood test described inseveral reports (10, 85) may ensure a more

precise assessment of the effects of pesticideson honey bee larvae. After collection of rstinstars, and grafting in articial rearing cells inlaboratory, the larvae are fed with a diet con-taining a pesticide until adult emergence. Thismethod appears to be promising to screen outsublethal effects of pesticides on the physiol-ogy of larval development.

IGRs may also have physiological effectson honey bee adults, particularly by inhibit-ing the formation of imaginal organs, whichmay have indirect effects on larval develop-ment. Newly emerged adults of A. melliferaand Apis cerana treated with diubenzuronshowed reduced weight gain and suppresseddevelopment of hypopharyngeal glands (59).This study demonstrated the morphogeniccapacity of a chitin inhibitor, an action for-merly showed with juvenile hormone mimics(65). Because hypopharyngeal glands of nursebees produce the royal jelly used to feed therst instars of worker larvae, and all instars ofqueen larvae, their malformations might re-sult in undernourished larvae and so mightpotentially lead to a decline in colony popu-lation or to no renewal of the queen. Simi-larly, nurse bees use vitellogenin to produceroyal jelly (6), and pyriproxyfen ( JHA) im-pairs vitellogenin synthesis in the hemolymph(99), suggesting repercussions on brood care.Such physiological effects might also causedisturbances in longevity, immunity, or repro-duction, because vitellogenin, a lipoprotein, isof fundamental importance in each of theseprocesses (6).

Adult Longevity

Effects on longevity after exposure to lethalor sublethal doses of pesticides have been de-scribed mostly for parasitoid species (5, 43,47, 75, 108, 110, 115, 118) and to a lesser ex-tent for predators (60, 82). Depending on thestudy, reduced longevity may be considered asublethal effect or latent mortality. Extrapo-lation of these effects to the population levelis difcult because, depending on the biologyof the particular natural enemy [proovigenic


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


or synovigenic (66), parasitoid or predator],they may be more or less likely to reproduceand/or to kill pests before their prematuredeath. From a practical perspective, it is theresulting amount of feeding and reproductionthat occurs between exposure and death thatis important. The consequences of reducedlongevity on population dynamics were re-cently emphasized by studies assessing pes-ticide impacts on arthropods using life tableanalysis (reviewed in Reference 120). Whenthe rm is determined for risk assessment ofpesticides, a reduction of survival (lx) couldlead to a strong reduction of the rm and con-sequently a negative effect at the populationlevel (120).

In the honey bee, the possible long-termexposure to a toxic agent by contaminationof stored food has been established by study-ing the transfer of pesticides sprayed on cropsinto the hive (137). Thus, the lethal dose esti-mated during acute toxicity tests appears to bea partial measure of the lethal effect becauseof the short duration of these tests (1 to 3 daysin most cases). Studies concerning long-termsurvival of honey bees raise the problem ofstatistical analysis of survival data. In chronictoxicity tests, most often only the end result oflong-term poisoning (i.e., an increase of cu-mulative mortality) is analyzed (113). Someapproaches consider how the mortality ratevaried during the time of pesticide exposureby a graphic interpretation (124, 126), but notwith statistical analysis. Conversely, when sta-tistical methods are employed in survival anal-ysis a parametric model is often used (63, 138).However, these analyses depend strongly onthe validity of the assumption that the survivaltime has a particular probability distribution.Moreover, these statistical methods are gener-ally based on the hypothesis of independencebetween bees belonging to the same group,which is not realistic. Indeed, food exchanges,contacts, and pheromonal communication oc-curring among workers make survival of a beedependent on the survival of its nestmates.Dechaume-Moncharmont et al. (26) demon-strated this density dependence in pesticide

effects with the use of a Cox proportionalhazard model.

Immunology

Insecticides can interact with the immunecapacity of insects. Depending on the typeof insecticide, they can decrease or increasethis capacity. Monocrotophos and methylparathion applied at one tenth of the LC50decreased the number of plasmatocytes in thehemolymph of the predator R. kumarii by16% and 13%, respectively, whereas endo-sulfan (organochlorine) increased this num-ber by 15% (56). Plasmatocytes have a di-rect role in the immune response of insectsby enabling the encapsulation of foreign bod-ies (111a). George & Ambrose (56) reportedthat decreases in the number of plasmato-cytes were associated with an increase in thenumber of granular hemocytes, which playa role in detoxication through phagocyto-sis. They hypothesized that plasmatocytes aretransformed into granular hemocytes duringthe detoxication process, indicating that thetested pesticides acted on the predators im-munological response indirectly by mobiliz-ing immunity cells for detoxication tasks. Inhost-parasitoid relations, pesticides may indi-rectly affect the parasitoids by lowering theimmune reaction of the host. Dieldrin (cyclo-diene) and endosulfan, applied at LD30, de-creased by 25% and 23%, respectively, theimmune reaction of Drosophila melanogasteragainst larvae of its parasitoid Leptopilinaboulardi (35). However, insecticides may alsoincrease the encapsulation of parasitoid lar-vae. When L. boulardi was exposed to an LD50of chlorpyrifos, the encapsulation of its eggswas increased by 4.5% (41). Therefore, insec-ticides may have an impact on both the im-mune capacity of a host and the capacity ofparasitoids to evade the host immune reaction.

Fecundity

Reductions in fecundity associated with pes-ticides may be due to both physiological


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


and behavioral effects (the effects on be-haviors are described later in the review).Many authors have reported general effectson fecundity of natural enemies regardlessof the nature of perturbations (14, 22, 54,75), but mechanistic insights into the ef-fects of pesticides on natural enemy fe-cundity have been obtained. Consoli et al.(21) described a reduction of fecundity ofthe parasitoid T. pretiosum when exposedto lambda-cyhalothrin (pyrethroid), teuben-zuron (IGR), or tebufenozide (ecdysone ag-onist) before oogenesis, but not after. Theyhypothesized that tebufenozide may inter-fere with ecdysteroid receptors, leading toa general perturbation of insect reproduc-tive process involving ecdysteroids (vitello-genesis, ovulation of mature eggs, promo-tion of spermatocyte growth). A reduction inthe number of hosts parasitized by C. plutel-lae (during a 10-h period) after ingestion ofthe IGRs chloruazuron, ufenoxuron, andteubenzuron has been reported (61). The ef-fect was linked to a reduction in viable eggsbecause of the known effect of ufenoxuronand teubenzuron on female fertility (67).Considering both neurotoxic and IGR pes-ticides, the IGRs may induce more long-termeffects on fecundity than neurotoxics. Indeed,the life-table parameters (which include fe-cundity) of the lacewing predator Micromustasmaniae after exposure to several IGR andneurotoxic pesticides were more seriously af-fected by the IGRs than by the neurotoxicinsecticides (108). Moreover, Rumpf et al.(108) emphasized that long-term sublethaleffects described in their study may inter-fere with the phenological synchrony betweenpest species and natural enemies, leading to aglobal decrease in their ability to regulate pestpopulations.

Sex Ratio

Physiological effects of pesticides include al-teration of the sex ratio of benecial insectsvia differential survival as a function of sex(5, 24), but additional effects are expected be-

cause pesticides can induce deformations ofovaries (57, 88, 115) and testes (57). How-ever, very few studies have documented po-tential mechanisms of sex ratio alteration bypesticides for benecial arthropods. Overall,two major causes are thought to alter the sexratio of the offspring when adults are exposedto pesticides: (a) an effect on the fertilizationof ova, especially in haplodiploid species inwhich the fertilization of ova is a voluntaryact by females when they are laying eggs, and(b) differential survival of sexes when exposureis before the adult stage (64).

Chlorpyrifos modies the sex ratio of hy-menopteran parasitoids by decreasing thenumber of females in the offspring whenonly parental females are exposed. This phe-nomenon has been observed for Aphytismelinus. The offspring of females that sur-vived the insecticide (LD50) were 58% fe-male and offspring of the control group were73% female (107). In Trichogramma brassicae,the offspring of females surviving exposureto chlorpyrifos (LD20) were 61% female andprogeny of the control group were 73% fe-male (40). Similar results were obtained withtwo pyrethroids (deltamethrin and lambda-cyhalothrin) that decreased the number of fe-male offspring of Aphidius uzbekistanicus whenadults were exposed to insecticides (75). Thisdecrease in the number of female offspringmay be related to the fact that hymenopteranfemales result from fertilized eggs, whereasmales result from unfertilized eggs. Egg fer-tilization is a voluntary act by females. There-fore, this behavior of fertilizing eggs may bealtered through the impacts of insecticides onnerve transmission in exposed females.

BEHAVIORAL EFFECTS

Mobility

The mobility of benecial arthropods af-ter exposure to pesticides is often not di-rectly studied. Moreover, studies are usuallynot accompanied by precise measures withquantitative data or statistical analysis. Effects


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


Kairomone:chemical substanceproduced by anindividual that servesas a stimulus to otherindividuals ofanother species forbehavioral responses

Foraging: thebehavior ofsearching for food,host, or prey

on the mobility of benecial arthropods havebeen observed, but they are mostly dueto (a) direct intoxication by the pesticides,resulting in knock-down effect (23, 124),uncoordinated movement (5, 14, 116), trem-bling, tumbling, abdomen tucking, and/or ro-tating and cleaning of the abdomen whilerubbing the hind legs together (124); (b) sec-ondary consequences of behavioral modica-tions (111) such as disruption in the detec-tion of kairomones that result in an increaseof angular speed due to higher arrestment bykairomone patches and hydrous stress (34);and (c) a repellent (72, 84, 106) or irritant ef-fect of pesticides (144). Several authors (111,144) reported increases in mobility of naturalenemies with the assumption that these in-creases would result in greater activity againstpests. The predator C. septempunctata walkedand groomed more frequently when releasedin a plot sprayed with deltamethrin (144)mainly because of irritation caused by thepesticide. The grooming behavior associatedwith increased mobility is thought to be a re-ex action initiated by irritation of chemore-ceptors located on the surface of the insectbody (103). This irritant effect may inducemovement of the insects away from the treatedareas. Consequently, increased mobility can-not be associated with increased natural en-emy efciency. In contrast, perturbations ofmobility can increase natural enemy vulnera-bility to predation in the eld (77).

To study chemical effects on the motoractivity of benecial arthropods, more sub-tle endpoints that provide quantitative datamight be more useful. The amount of inac-tive time and the position of topically treatedworker bees (in an open-eld-like arena al-lowing observation of bee vertical displace-ment) were compared with those of controlbees (78). Adverse effects of imidacloprid onmotor activity were dependent on insecticidedose. The lowest dose (1.25 ng per bee) re-sulted in increased motor activity, whereas thehigher doses (2.5 to 20 ng per bee) decreaseddisplacements in the arena. The inuence ofimidacloprid on mobility could also change

with time (124). Therefore, we can assumethat with the same dose of imidacloprid it ispossible to observe inverse effects accordingto the time of observation. Using the sameparadigm described by Lambin et al. (78), astudy reported that pronil (pyrazole) had noeffect on motor activity whatever the routeof exposure (oral or topical) (51). Here, otherendpoints were used: distance covered andtime spent in each of the six levels of the arena.This test is based on negative geotaxis or pos-itive phototaxis because honey bees tend tomigrate upward against the force of gravity tothe light source. This test provides an accu-rate assessment of motor function of walkingbees, but it does not measure ying activity,which is essential in the process of foraging.

Navigation/Orientation

In natural enemies, navigation and orientationcould involve multiple sensory cues, eitherchemical (135) or visual (140). Natural ene-mies spend a signicant proportion of theirlife searching for hosts or prey. Navigationdepends entirely on nervous transmissions,which are targeted by neurotoxic insecticidesthrough different modes of action. Therefore,effects on navigation are frequently reported.Longley & Jepson (83, 84) and Umoru et al.(130) reported perturbations of the foragingpattern in parasitoids, but specic effects ofthe pesticides were not isolated and repul-sive and direct behavioral effects remained un-known. In general, insects have been connedto pesticide-treated plants and the positionof the natural enemies was recorded at var-ious times. The authors described a reductionin time spent on treated plants and an inver-sion in leaf side preference, but direct effectson orientation behaviors remained unknown.However, other studies have more preciselydescribed potential effects on navigation be-havior by combining a controlled exposuretime and dose followed by the use of a spe-cic behavioral apparatus. Exposure methodscan mimic natural exposure conditions, for ex-ample, tarsal exposure on pesticide deposits


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


(34, 36, 48), exposure via feeding on contam-inated sources (118), and direct exposure bytopical application (50). Behavioral tests canassess most important steps involved in thenavigation process. Results show that pesti-cides induce different and sometimes oppo-site effects on host searching by parasitoidsdepending on the species and insecticide used.Indeed, positive sublethal effects of pesticideson natural enemy orientation behaviors havebeen reported (34, 73).

However, most of the studies reportednegative effects on orientation behavior.When the parasitoid Microplitis croceipes con-sumed extraoral nectar of cotton con-taminated with imidacloprid or aldicarb(carbamate), its response to odors of the host-plant complex in a wind tunnel decreased by71% and 62%, respectively (118). A lowerresidence time on the contaminated hostpatch was observed with females of the para-sitoid Trissolcus basalis exposed to deltamethrinat LD25, compared with unexposed females(111). In a four-armed olfactometer, the ca-pacity of aphid parasitoids to orient towardhost-induced plant odors (synomones) couldbe decreased by exposure to a sublethal doseof lambda-cyhalothrin (45) and to increasingdoses of triazamate (carbamate) (46). Desneuxet al. (45) also emphasized that these ef-fects could be temporary and that insectscould recover after a period without expo-sure. With predators, studies designed toassess the effects of pesticides on navigationtypically focus on relatively short-range preydetection and hunting. Cypermethrin, at rec-ommended eld rates, reduced the attack rateof Acanthaspis pedestris 2.4- to 6.4-fold, withthe effect increasing with prey density (19).

Disruption of sexual communication andmate-nding has also been reported. Pes-ticides modify chemical communication be-tween sexual partners by altering the capac-ity for stimulus creation by the emitter orstimulus perception by the receiver. Stimulusdetection and integration by the CNS are po-tential targets for perturbations by pesticides(62). For example, T. brassicae males exposed to

a LD20 of chlorpyrifos are less arrested by fe-male sexual pheromones, and exposed femalesemit less of these pheromones (36). Sublethaldoses may also disrupt sexual communica-tion. T. brassicae males exposed to a LD0.1 ofchlorpyrifos were less arrested by female sex-ual pheromones; however, pheromones emit-ted by exposed females (LD0.1) were more ar-resting for untreated males (37). In contrast,when T. brassicae males were treated with thepyrethroid deltamethrin at LD0.1 there wasan increase in arrestment, whereas when fe-males were treated, their pheromones wereless arresting for males (39). These effects off-set each other when both sexes are exposed,with a mean response to sexual pheromonessimilar to that of the control. However, thekinetics of the response are modied (38).

For pollinators, visual learning of land-marks is important in spatial orientation.Honey bees use visual landmarks to navigateto a food source as well as to communicate ac-curately to their nest mates the distance anddirection to y to reach it (139). A bee exposedto pesticide during a foraging trip may in-correctly acquire or integrate visual patterns,causing disorientation and loss. Aside fromimpairing the orientation behavior of exposedforagers, insecticides could affect the accuracyof information relayed through the dances ofthe returning foragers. Recently, the effects ofdeltamethrin on the homing ability of foragerswere investigated. Honey bees were trainedto forage on an articial feeder lled with su-crose solution and were individually markedwith colored number tags. In an insect-prooftunnel with the feeder located 8 m from thehive, deltamethrin altered the homing ightin foragers treated topically at sublethal doses(134). The percentage of short-term ightsback to the hive decreased in treated foragers,which ew in the direction of the sun.

Still, a relatively small number of studieshave investigated the impact of pesticides onhoming ight, perhaps because of the dif-culty of measuring parameters such as direc-tion of ight or the route time between thefood source and the hive. Most techniques


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


ppb: part per billion

are limited by the number of individuals whomight be monitored simultaneously and bythe time span during which observations canbe made. However, techniques of automatictracking and identication of individuals havethe potential to revolutionize the study of be-havioral ecotoxicology. In this regard, severaldifferent types of transponders such as har-monic radar (104) and radio frequency iden-tication devices (RFID) (123) may be use-ful for studies using the honey bee. Presently,RFID tags offer the most advantages (unlim-ited number of individual insects, large num-bers of events recorded, rapid reading) (79)and they cause less disturbance to the in-sects than harmonic radar, which requires theattachment of an antenna. Given the largerange of biological parameters potentially af-fected by pesticides, another approach mea-suring the orientation performance of bees ina complex maze relies on associative learn-ing between a visual mark and a reward ofsugar solution (147). Using this experimentalsetup, researchers examined whether foragersreceiving 1 ppb (parts per billion) pronil (ad-ministered orally) can learn to y through amaze according to the presence or absence ofa visual cue (A. Decourtye, unpublished data).The bees learned the maze by making cor-rect and incorrect decisions. The maze sim-ulates learning of complex routes under eldconditions. Results for experimental controlsshowed that 89% of bees ew through the en-tire path and arrived at the goal (reward ofsugar solution). However, when the bees wereexposed to pesticide, the rate fell to 60%. Inparallel, the percentage of bees that did notnd the goal within 5 min of entering the mazeincreased dramatically when exposed (34%and 4% in exposed and control groups, re-spectively). Thus, the orientation capacity offoragers in a complex maze was highly affectedby pronil.

Feeding Behavior

Pesticides may interfere with the feeding be-havior of exposed insects in three general

ways. First, some pesticides are well docu-mented to have repellent effects on bene-cial insects, and this effect may conict withfeeding behavior. Second, some pesticides areused specically for their antifeedant proper-ties (100) with the possibility that benecialinsects may also be discouraged from feed-ing when exposed. Third, disruption in theability to locate food may occur after expo-sure to pesticides because of reduced olfac-tory capacity (31). However, the consequencesof effects may depend on the organisms con-sidered. For proovigenic natural enemies,reduced feeding may inuence the overallparasitism/predation rate because of reducedlongevity. However, this effect may be lim-ited because these insects do not require en-ergy for egg production (66). In contrast, re-duced feeding by the adults of synovigenicspecies may reduce egg production, leadingto reduced tness. Moreover, perturbation ofhost feeding behavior exhibited by many par-asitoids (66) and predation by predators maydrastically reduce the efciency of natural en-emies. In the case of honey bees, impairedfeeding behavior can induce a drastic declinein hive population. In large-scale farming ar-eas, when food resources are reduced to culti-vated plants, the repellent effect of pesticidesmay reduce pollen and nectar uptake, poten-tially leading to a demographic decrease of thecolony.

Honey bees change their behavior in re-sponse to pesticides through reduced feed-ing stimulation (62). For example, topical ap-plication of deltamethrin at concentrationsranging from 0.08 to 0.16 ppm increasedsyrup uptake in bees, although similar concen-trations delivered orally decreased uptake oftreated syrup (127). Despite this work, manyof the reported effects on the foraging activ-ity of bees relate to avoidance behavior and,in a few cases, feeding stimulation. In thesame manner, ofcial guidelines for assessingthe impact of pesticides on foraging activityconsist mainly of measuring repellency (93).These bioassays were developed in higher-tierstudies, under semield and eld conditions.


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


In this regard, pyrethroids are probably thebest-known repellent insecticides (33, 106).For a long time, repellency associated withpyrethroids has been considered a behavioraladaptation for reducing the risk of exposure.However, it is now known that pyrethroidapplications during peak foraging activity (inbroad daylight) result in high exposure levels(32). Therefore, a repellent effect must notbe misconstrued as providing any protectionagainst exposure to pesticides.

In the case of parasitoids, Aphidiusrhopalosiphi responded strongly to patchesof aphid honeydew on lter paper, butthe addition of increasing concentrations ofdeltamethrin caused increasingly early de-parture from the honeydew/deltamethrin-treated areas because of a repellent and/or irri-tant effect (84). The inuence of deltamethrinon the general feeding behavior of the preda-tory ladybeetle C. septempunctata was exam-ined via visual observations (144). The move-ment of ladybeetles on deltamethrin-treatedplants increased, although the ladybeetlestended to stay on plant parts known to re-ceive less pesticides during treatment (72),thus demonstrating a clear repellent effect. Asimilar effect was observed when ladybeetlesforaged on dimethoate-treated plants (116).The authors indicated that adult and lar-val stages ate fewer aphids when exposed todimethoate because they avoided the treatedareas. They also reported less consumptionof aphids that were previously treated (inchoice experiment), indicating that a com-bined repellent/antifeedant effect could oc-cur. An antifeedant effect was demonstratedfor the predatory carabid Nebria brevicollisfeeding on deltamethrin-treated aphids (143).The authors reported that 53% to 80% of bee-tles ingesting treated aphids exhibited a regur-gitation response after consumption.

The specic sensory mechanism throughwhich repellency operates is not well under-stood but may depend on the mode of expo-sure. Fipronil (87) and AzaA (92) reduced vis-itation of honey bees to treated sucrose butdid not have an effect when applied to ow-

ering canola. Several observations on the re-pellency of pesticides after spraying indicatethat it could not be attributed to the activeingredient itself, but rather to additives incommercial formulations (13a) or the phys-ical characteristics of the spray (wetting veg-etal). These observations may explain the dif-ferences in behavior observed between treatedarticial food and eld applications. Althoughthe means by which foragers detect the treat-ment on the crop remains uncertain, the gus-tatory perception of active ingredients can beassumed when there is a notable decline inthe level of foraging on an articial feeder atshort time intervals (28, 87, 112). The delayin the inhibition of foraging with imidaclopridvaries according to concentration tested (112).The author attributed this delay to a processoccurring inside the hive rather than to ef-fects on foragers. A decrease in feeding activ-ity may be the result of changes in commu-nication processes that alter foraging activity.This hypothesis was reinforced by studies re-porting the impact of imidacloprid on dances.Dances produced by returning foragers con-tain coded information about the distance anddirection of food sources and are aimed at therecruitment of foragers (139). Decreases inthe frequency of waggling dances were ob-served when foragers had previously visiteda sugar solution contaminated with approxi-mately 20 ppb of imidacloprid (25). A lowermotivation to perform waggling dances canresult in a reduction of recruitment activity.

The process of food detection, whetherthe food is prey (predators and host-feedingparasitoids), nectar, or honeydew (proovi-genic parasitoids or honey bees), involvessophisticated nervous activity that can be dis-rupted by neurotoxic pesticides (62). The re-duviid predator Acanthaspis pedestris exhibiteda decrease in excitation by prey and a de-creased ability to paralyze prey after expo-sure to the cypermethrin (19). These per-turbations were accompanied by a reductionin food intake, haphazard movement, andsigns of restlessness. Reductions in the rate ofaphid consumption by the predators Coccinella


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


PER: proboscisextension response

septempunctata and Chrysoperla carnae at thelarval stage were also reported when aphidswere treated with AzaA (3). The impact onfood detection may also have a negative im-pact on parasitism capacity of parasitoids. Thedecrease in parasitism capacity may resultfrom reduced energy intake, as well as from in-direct effects on host detection. Indeed, it hasbeen suggested that aphid parasitoids exploithoneydew deposits on the surfaces of leaves ascues for evaluating the numbers of aphids onthe plant and therefore the amount of effort tobe invested in further search (83). Thus, if theability to detect honeydew is compromised, aneffect on foraging pattern and patch time al-location may occur, resulting in a reduction inparasitism rate. In honey bees, the proboscisextension reex (PER) can be elicited throughantennal stimulation with a sucrose solutionand can be used to assess the gustatory thresh-old to sugary foods (8). The gustatory thresh-old is dened as the lowest concentration ofa sucrose solution, applied to the antennae,that is capable of eliciting a PER. A reductionof sensitivity for low-sucrose concentrations(55% to 60% in control versus 15% to 20%in treated bees) was observed following tho-racic application of pronil at a dose of 1 ngper bee (51), demonstrating that pesticide ex-posure can drastically reduce the capacity ofhoney bees to detect food sources.

Oviposition Behavior

Most studies concerning the effects of pesti-cides on oviposition behavior have been doneon parasitoids because of the direct linkagebetween oviposition and parasitism rate andconsequently pest regulation. However, fewstudies in this regard have been conducted onpredators (11), and to our knowledge nonehave been conducted on pollinators. Pesti-cides can disrupt the very precise coordinationbetween the insect nervous and hormonal sys-tems, resulting in a breakdown in the complexseries of behavioral and physiological eventsrelated to oviposition. Indirect perturbationsin oviposition behavior may be induced by the

repellent effect of pesticides, which can re-duce the chances that a natural enemy willnd a suitable host or oviposition site (83,130), and also by occurrence of uncoordinatedmovements after pesticide exposure (5, 46). Inthese two last studies, after exposure to lethaland sublethal doses of pesticides, Aphidius erviand Trybliographa rapae females exhibited anirreversible uncoordinated ovipositor extru-sion and consequently failed to lay eggs.

Kuhner et al. (76) described the negativeeffects of herbicides on the parasitic behav-ior of Diaeretiella rapae, which included a re-duction in the number of attempted stings.For another aphid parasitoid, A. ervi, femalesshowed signicantly less oviposition activitycompared with the controls after exposureto a LD20 of lambda-cyhalothrin (45). Thefrequency of sting attempts and related be-haviors were signicantly reduced. The par-asitoid Neochrysocharis formosa exhibited a re-duction in the number of ovipositor insertionsinto a host, host mine drumming frequency,and the number of eggs laid when foraging onimidacloprid-treated leaves (129). Similar re-duction in the number of hosts stung has beenreported in the parasitoid Colpoclypeus orus af-ter exposure to two commercial formulationsof spinosad (14). These authors also reportedthat for one formulation no offspring wereproduced. Egg deposition may have been dis-rupted in these experiments, as uncontrolledegg laying associated with egg losses could oc-cur after pesticide exposure (5). Effects wereformulation dependent, which implied thatadjuvants may be worthy of consideration.An effect on egg deposition may also be dueto perturbation of chemoreceptors or infor-mation integration during host acceptance[occurring during ovipositor insertion intohost (136)], but this effect has not been welldescribed.

Short-range detection of hosts may also bealtered by pesticides, resulting in disruptionof oviposition. When parasitoids are closeto a host, they often rely on short-rangechemical cues (135) or color (140). Desneuxet al. (45) determined that after exposure to


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


lambda-cyhalothrin the parasitoid A. erviexhibited less host antennal contact andantennal examination behavior than didunexposed individuals and that these obser-vations were not associated with perturbedmobility. Similarly, the parasitoid H. didy-mator had difculties nding hosts afterexposure to AzaA (115). The parasitoid didnot actively search for hosts and often walkedaway from hosts. This effect was independentof any repellent effect, because parasitoidswere tested after their exposure to pesticides.However, sublethal effects of pesticides onhost detection and oviposition may not alwaysbe unfavorable to parasitoids. Chlorpyrifosat LD20 caused an increase (5.1-fold) inhost searching by Leptopilina heterotoma, aparasitoid of Drosophila larvae that probessubstrate with its ovipositor (102). Further-more, the authors reported that treatedfemales found and oviposited into host larvae46% faster than did control females.

Learning Performance

Effects of pesticides on learning processesof benecial arthropods have been studiedmostly in pollinator models and, more specif-ically, in honey bees because of the better un-derstanding of their learning processes andthe importance of learning in the foragingprocess (91). In contrast, very few studies haveinvestigated the effects of pesticides on thelearning capacity of natural enemies, and im-pairment of specic learning traits has notbeen reported. Odor conditioning in the par-asitoid L. heterotoma (probing into substrate)was not modied by tarsal exposure to dryresidues of chlorpyrifos (LD20) (102). In theaphid parasitoid A. ervi, learning capacity forsynomones and consequent olfactory orien-tation in an olfactometer were not modiedafter tarsal exposure to lambda-cyhalothrin(LD0.1 and LD20) (45).

When landing on a ower, each honey beeforager is subjected to a conditioning processin which oral cues (smell, color, and shape)are memorized after being associated withfood (91). Once memorized, the odors play

a prominent role in ower recognition dur-ing subsequent trips (90). Under laboratoryconditions, olfactory learning can be studiedusing a bioassay based on conditioning of thePER in restrained individuals (125). The PERassay simulates natural honey beeplant inter-actions that take place when landing on theower; the forager extends its proboscis as areex when the gustatory receptor set on thetarsi, antennae, or mouthparts are stimulatedwith nectar. This reex leads to the uptakeof nectar and promotes memorization of con-comitant oral odors. The PER assay has beenused with restrained workers to investigate thebehavioral effects of about 20 different pesti-cides (1, 29, 30, 86, 122, 142). However, inorder to conrm that the effect of a pesticideon conditioned PER levels is due strictly tofailure of learning or memory ability, it is nec-essary to consider impacts on motor functionsand gustatory and olfactory senses that under-lie the endpoint (8, 31, 51).

Toxicant exposure can be carried out be-fore (122), during (1), or after (86) PER con-ditioning. In an ecological context, long-termexposure to low concentrations correspondsto the case of inexperienced bees involved inforaging duties based on their learning abil-ity after being fed a contaminated food withinthe hive. With this approach, reduced learn-ing performance was observed in bees surviv-ing 11 days of oral exposure to imidacloprid,5-OH-imidacloprid, pronil, deltamethrin,endosulfan, and prochloraz (29, 30). Thisbioassay can also help to assess how chem-ical treatments can interfere with the mem-ory process and provides an indication of theability of foragers to return to a crop wherethey have been exposed to a toxin while theywere collecting food and memorizing the o-ral cues. In this regard, imidacloprid adminis-tered after trial conditioning of PER impairedmedium-term olfactory memory (27). By con-trast, short-term and long-term memory wasunaffected. It was assumed that the consoli-dation process that ensures the transfer fromshort-term memory to medium-term mem-ory was affected by imidacloprid. Because


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


ower choice in successive foraging bouts isinterrupted by return to the hive (89), imi-dacloprid may affect this process. However,the precise consequences for foraging behav-ior are still unclear.

Overall, work using the PER assay hasemployed olfactory Pavlovian conditioning.However, pesticide exposure can affect notonly associative learning task but also nonas-sociative learning procedures such as habit-uation. Habituation of the PER is a simpleform of learning in which the repetition ofgustatory stimulation leads to a decrease ofresponse probability. This test sheds light onthe ability of an organism to constrain a re-ex response. The habituation procedure hasbeen used to demonstrate the effect of a sub-lethal dose of imidacloprid on PER suppres-sion (58, 78). These results clearly indicatetask-dependent behavioral effects associatedwith sublethal doses of imidacloprid and canbe generalized to other insecticides.

IMPACT OF SUBLETHALEFFECTS ON COMMUNITYECOLOGY

Because sublethal effects of pesticides inter-act with numerous life-history traits involvedin the reproduction of benecial arthropods(i.e., foraging, fecundity, sexual communica-tion, and sex ratio), they likely have an impacton insect communities. However, althoughthe effects of pesticides on insect communi-ties have been described, sublethal effects havenot specically been analyzed. Pesticides tendto lower the abundance of both parasitoidsand their hosts, and lead to the disappear-ance of scarce species (species that are notnaturally abundant in a given agroecosystem)(J.M. Delpuech & R. Allemand, unpublisheddata). A potentially useful tool for evaluatingthese different impacts would be to integratethem in a modeling approach. However, syn-thetic pesticides generally had a lower impacton natural enemy populations than predictedby database analyses, and recolonization ofchemically treated plots can be rapid (9). In-

deed, effects of pesticides on natural enemiescan be short-lived (4).

Serious losses of pollinators have been at-tributed to pesticides (96), and honey beescan be used effectively as bioindicators todetect environmental pollution (71). How-ever, effects of pesticides on honey bees arepoorly representative of effects on other polli-nators, including other bees (68). Indeed, bees(Apoidea) constitute a highly diverse group,and bees from different taxonomic groups dif-fer widely in their vulnerability to pesticideexposure. In honey bees, pesticides may af-fect social organization (reduction of food up-take or reduction of worker/brood popula-tion), but these effects may be compensatedfor because the queen does not take part inforaging and is probably less likely to be ex-posed than workers. In contrast, in other so-cial pollinators such as bumble bees, the queenmust nd food during spring in order to foundthe colony. In this case, the potential negativeeffects of pesticides may substantially affectcolony establishment. In summary, social pol-linators having no perennial colony and no so-cial pollinators are more likely to suffer frominsecticide exposure.

IMPLICATIONS ININTEGRATED PESTMANAGEMENT ANDPOLLINATION

The economic gains due to beekeeping andagricultural pollination might be reduced byintoxication of colonies with pesticides, eventhough there are few data to support this as-sertion. The best example is a long-term studyconducted in eastern Canada. In that region,blueberry production, which depends largelyon pollination by as many as 70 species ofnative insects, failed in 1970, and subsequentyears, because of aerial spraying of fenitroth-ion (70). Although the impacts of mass mortal-ities of honey bees on pollination of crops aredocumented, less understood and often over-looked is the problem of sublethal effects thatreduce agricultural production.


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


Studies of the sublethal effects of pesti-cides on natural enemies often aim to assessthe suitability of pesticides for IPM. How-ever, sublethal effects on natural enemies arerarely taken in account when IPM programsare established. To reduce nontarget effectsof pesticides on natural enemies, selectivitytests are performed with the aim of choos-ing pesticides with a high degree of lethaltoxicity against the target pests and mini-mal nontarget lethal toxicity (24). However,given the potential importance of sublethaleffects on natural enemies (reported in thisreview), pesticide choice should also con-sider those with minimal sublethal effects onkey components of benecial efciency. Forexample, a comparison of sublethal effectsof two pyrethroids (lambda-cyhalothrin anddeltamethrin) on key behaviors of aphid para-sitoids demonstrated that lambda-cyhalothrindisrupted olfactory orientation toward host-infested plants and oviposition behavior (45)and that deltamethrin did not (43, 47, 48).These results were partially conrmed by astudy in semield conditions in which theauthors showed that aphid parasitoids re-leased on deltamethrin-treated plants signi-cantly limited aphid population growth evenwhen introduced one day after treatment (44).Moreover, the pesticide and parasitoid effectswere additive. We expect that more thor-ough consideration of potential sublethal ef-fects on natural enemies in the future will helpto optimize IPM programs involving use ofboth natural enemies and pesticides againstpests.

METHODS PROMISINGFOR INTEGRATION INREGISTRATION PROCEDURES

Pollinators

Environmental risk assessment of pesticideson honey bees takes into account mainly thesurvival of adult bees exposed to pesticidesover a relatively short time frame. Further-more, sublethal effects are generally not con-

sidered. Whereas lethal effects are rather easyto observe and can lead to loss of product reg-istration, more subtle effects on bee physiol-ogy or behavior may also affect honey beepopulations. The honey bee risk assessmentscheme tentatively takes into account thesedifferent aspects of exposure. U.S. EPA guide-lines indicate that abnormal behavior dur-ing acute toxicity tests should be preciselyrecorded (i.e., kind, time of onset, duration,severity, numbers of bees affected). In Europe,when the standard procedures cannot provideclear conclusions on the harmlessness of apesticide, the ofcial decision-making schemerecommends the use of additional studiesin order to provide adequate information(93). However, no specic protocols are out-lined even though issues related to pesticidesand bees are intensively discussed. In recentyears, many beekeepers in European coun-tries have complained about unusual honeybee losses and hive depletions. These lossesmay be due to the use of seeds dressed withnewly registered pesticides (20). This issue hasrevealed limitations in standardized regula-tory methods: underestimation of bee expo-sure after seed-dressing application, failure toaccount adequately for larval and sublethaltoxicities, and absence of measures of chronictoxicity. In this context, more standardizedmethods to evaluate sublethal effects of pesti-cides may be needed.

According to guidelines, a brood feedingtest is required to evaluate whether bee larvaemay be at risk when exposed to a compoundshowing IGR activity (94). Because of envi-ronmental variation, the recommended meth-ods (95) may not be easily reproducible. Thus,an in vitro method for rearing bee larvae hasbeen improved and may be recommended forregulatory trials assessing pesticide toxicity tolarvae (10). Sublethal effects may be investi-gated by measuring weight and larval devel-opmental variation and morphology changesin adults. Further studies are necessary to de-termine if the method can reliably detect be-havioral effects in individuals exposed duringthe larval stage.


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


The PER assay can be considered a quan-tiable and reliable method to assess sublethaltoxicity (31). The PER procedure enables re-searchers to compare responses to differentchemicals and different concentrations of thesame chemical, determine no-effect concen-trations, and investigate the nervous circuitryunderlying the olfactory learning processeswhen neurotoxic molecules that affect the pe-ripheral nervous system or CNS are tested.Although the associative learning of workersinvestigated with PER assays may be ecologi-cally signicant, it is unknown how well theytranslate to the colony level under naturalconditions (128). However, preliminary stud-ies using PER assays indicate that the decreasein learning performance induced by imidaclo-prid at the individual level translate well to thecolony level in olfactory discrimination tasks(28).

Natural Enemies

To date, no standardized methods for assess-ing sublethal effects on parasitoids and preda-tors are described in the regulatory texts. TheEuropean standard characteristics of nontar-get arthropod regulatory testing (ESCORT 2)workgroup has developed a method to im-prove the risk assessment of pesticides on nat-ural enemies and propose the adoption of ahazard quotient (HQ) approach. HQ is cal-culated by dividing crop-specic applicationrates by the LR50 (lethal rate 50) derived fromworst-case-scenario laboratory studies gener-ated using two sensitive indicator species (16).This method is a welcome development, butthere are still important questions to con-sider due to the fact that HQ is calculatedfrom LR50 values, and thus potential sub-lethal effects are not included. This integra-tion can be solved in part by using a de-mographic approach to estimate toxicity asreviewed by Stark & Banks (120). Life-tableexperiments are conducted by exposing indi-viduals or groups to increasing doses of a tox-icant over their life span and daily mortalityand reproduction are recorded, providing datato calculate the rm. Sublethal effects on fer-

tility, fecundity, developmental rate, survival,and sex ratio can be detected when estimat-ing the rm. Life-table experiments provide amore accurate measure of toxic effect than dolethal concentration estimates (52) and havebeen used successfully to evaluate side ef-fects of pesticides on several natural enemies(2, 121).

However, this method also has some limi-tations (120): It is expensive, time consuming,and performed under laboratory conditionsthat do not reect wild conditions (such asdensity dependence). Several of these pointshave been addressed (119, 141). However, ifwe consider the caging conditions used dur-ing experiments, lack of detection of behav-ioral perturbations induced by pesticides islikely. Indeed, the ability of parasitoids to de-tect host-induced plant odors (synomones) iscrucial because these odors are used to detecthost patches at long range (135). However, inlife-table experiments, parasitoids may be soclose to the hosts that they detect their hostswithout using long-range cues. Thus, an im-pairment of important aspects of foraging be-havior can be missed during evaluation. It maytherefore be important to add a standard be-havioral test to any toxicological tests, includ-ing the rm evaluation.

CONCLUSION AND FUTUREOUTLOOK

This review reports a wide variety of sub-lethal effects of pesticides on the physiologicaland behavioral processes in benecial arthro-pods. Effects are documented according to thetechnical issues associated with studying var-ious processes and also according to ecologi-cal knowledge of basic mechanisms involvedin the traits that are potentially perturbed bypesticides. In most cases only one dose, punc-tually administered and not necessarily sub-lethal, was studied. Thus, misinterpretationmay result from lethal concentrations mistak-enly used to study sublethal effects. Acute ex-posure to high concentrations of a chemicalcan result in selection of insects that are less


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


sensitive to the pesticide tested. Such resis-tant insects may provide responses that are notrepresentative of the population. This pointshould be carefully considered in future stud-ies of the sublethal effects of pesticides be-cause errors in choice of doses/concentrationstested may provide misleading results. Assess-ment of sublethal effects should be conductedby testing effects of both sublethal and lethaldoses/concentrations.

Methods using the honey bee model arewell dened, particularly in the eld of be-havior, because of its importance for studyingbehavioral and learning processes in insects(91). Moreover, increasing requirements re-lated to the nontarget effects of pesticides onpollinators for new pesticide registration havemotivated expansion of tests on this model(93). Thus, tests on honey bees in registrationprocedures are better developed than tests onnatural enemies, and the development and in-clusion of several promising methods into reg-ulatory procedures is in progress (32, 128).Methods to assess nontarget effects on nat-ural enemies are also progressing. Choice ofindicator species is also being made (15, 16)and was a rst step to help the developmentand inclusion of methods on natural enemies.However, sublethal effects are not a majorconcern yet, and further development of stan-dardized methods assessing sublethal effectson key components of natural enemy ef-ciency must be achieved before incorporatingthese effects into regulatory procedures.

The link between sublethal effects of pesti-cides and consequences at the population andcommunity levels are still not well understoodin either pollinators or natural enemies, andthe same can be said when considering howsublethal effects are taken into account for thedevelopment of IPM programs. Even thoughmany studies have documented sublethal ef-fects of pesticides on natural enemies, onlymortality tests are considered when a choicebetween several pesticides must be made. Tofully assess risk, it is crucial to establish a linkbetween the toxicity of a given product in lab-oratory assays and the risk associated with ex-posure under eld conditions. Although thispoint is often overlooked, it emphasizes theneed for studies on the dynamics of exposureto pesticides. It will require the quanticationof residues in different locations visited by in-sects and also an estimation of the degradationof pesticides under eld conditions. The useof multistep bioassays to evaluate the potentialeffects of pesticide on benecial insects wouldalso help to assess risk in a more completeway by including evaluation of pesticide ef-fects on key behavioral and physiological pro-cesses instead of considering mortality only asan endpoint. These assays, although slightlymore laborious than lethal concentration es-timates, will help researchers to evaluatethe nontarget impacts of pesticides and pro-mote discovery of crucial ecological side ef-fects before pesticide registration rather thanafter.

SUMMARY POINTS

1. Physiological sublethal effects on the development of benecial arthropods occur atmultiple levels. The parameter generally recorded is the developmental rate. However,new parameters such as malformation rates in natural enemies (when emerging frompupae) and in pollinators (in the cells inside the hive) are now used.

2. Studies have generally reported perturbations of the foraging pattern in parasitoidsand honey bee. Other studies have described more precisely the potential effects onnavigation behavior by combining a controlled exposure time and dose followed bythe use of a specic behavioral apparatus.


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


3. Pesticides may interfere with the feeding behavior by repellent, antifeedant, or re-duced olfactory capacity effects. A more drastic effect should be observed for synovi-genic species that need feeding for egg production all life long.

4. Learning processes depend on a high functionality of sensory and integrative nervoussystems, which in particular have high importance in the honey bee (oral and nestrecognition, spatial orientation). Therefore, the impact of neurotoxic pesticides onthese processes has been largely studied and identied in this insect.

5. Even though many studies have documented sublethal effects of pesticides on naturalenemies, only mortality tests are considered when a choice between several pesticidesmust be made in an IPM context. To fully assess risk, it is crucial to establish a linkbetween the toxicity of a given product in laboratory assays and the risk associatedwith exposure under eld conditions (including lethal and sublethal effects).

6. Methods to test sublethal effects on benecial arthropods are currently being devel-oped, and inclusion of several promising methods into regulatory procedures is inprogress (more advanced work on pollinators). However, sublethal effects are not amajor concern yet, and further development of standardized methods assessing sub-lethal effects on key components of natural enemy efciency will need to be achievedbefore incorporating these effects into regulatory procedures.

ACKNOWLEDGMENTS

We wish to thank Dr. D.S. Richmond for helpful comments on the review. N. Desneux alsothanks Dr. L. Kaiser for encouraging his interest in sublethal effects of pesticides. This workwas supported in part by a grant from French Ministry of Agriculture (European fund forFrench beekeeping).

LITERATURE CITED

1. Abramson CI, Squire J, Sheridan A, Mulder PG. 2004. The effect of insecticides consid-ered harmless to honey bees (Apis mellifera): proboscis conditioning studies by using theinsect growth regulators tebufenozide and diubenzuron. Environ. Entomol. 33:37888

2. Acheampong S, Stark JD. 2004. Effects of the agricultural adjuvant Sylgard 309 and theinsecticide pymetrozine on demographic parameters of the aphid parasitoid, Diaeretiellarapae. Biol. Control 31:13337

3. Ahmad M, Ossiewatsch HR, Basedow T. 2003. Effects of neem-treated aphids asfood/hosts on their predators and parasitoids. J. Appl. Entomol. 127:45864

4. Al-Deeb MA, Wilde GE, Zhu KY. 2001. Effects of insecticides used in corn, sorghumand alfalfa on the predator Orius insidiosus (Hemiptera: Anthocoridae). J. Econ. Entomol.94:135360

5. Alix A, Cortesero AM, Nenon JP, Anger JP. 2001. Selectivity assessment of chlorfen-vinphos reevaluated by including physiological and behavioral effects on an importantbenecial insect. Environ. Toxicol. Chem. 20:253036

6. Amdam GV, Norberg K, Hagen A, Omholt SW. 2003. Social exploitation of vitel-logenin. Proc. Natl. Acad. Sci. USA 100:1799802


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


7. Armengaud C, Causse N, Ait-Oubah J, Ginolhac A, Gauthier M. 2000. Functionalcytochrome oxidase histochemistry in the honeybee brain. Brain Res. 859:39093

8. Armengaud C, Lambin M, Gauthier M. 2002. Effects of imidacloprid on the neuralprocesses of memory in honey bees. See Ref. 48a, pp. 85100

9. Armenta R, Martinez AM, Chapman JW, Magallanes R, Goulson D, et al. 2003. Im-pact of a nucleopolyhedrovirus bioinsecticide and selected synthetic insecticides on theabundance of insect natural enemies on maize in southern Mexico. J. Econ. Entomol.96:64961

10. Aupinel P, Fortini D, Dufour H, Tasei JN, Michaud B, et al. 2005. Improvement ofarticial feeding in a standard in vitro method for rearing Apis mellifera larvae. Bull.Insectol. 58:10711

11. Banken JAO, Stark JD. 1998. Multiple routes of pesticide exposure and the risk of pesti-cides to biological controls: a study of neem and the sevenspotted lady beetle (Coleoptera:Coccinellidae). J. Econ. Entomol. 91:16

12. Bendahou N, Bounias M, Fleche C. 1999. Toxicity of cypermethrin and fenitrothionon the hemolymph carbohydrates, head acetylcholinesterase, and thoracic muscle Na+,K+-ATPase of emerging honeybees (Apis mellifera mellifera. L). Ecotoxicol. Environ. Safety44:13946

13. Bortolotti L, Sbrenna AM, Sbrenna G. 2005. Action of fenoxycarb on metamorphosisand cocoon spinning in Chrysoperla carnea (Neuroptera: Chrysopidae): identication ofthe JHA-sensitive period. Eur. J. Entomol. 102:2732

13a. Bos C, Masson C. 1983. Repellent effect of deltamethrin on honey bees. Agronomie3:54553

14. Brunner JF, Dunley JE, Doerr MD, Beers EH. 2001. Effects of pesticides on Colpoclypeusorus (Hymenoptera: Eulophidae) and Trichogramma platneri (Hymenoptera: Tri-chogrammatidae), parasitoids of leafrollers in Washington. J. Econ. Entomol. 94:107584

15. Candol MP, Bakker F, Canez V, Miles M, Neumann C, et al. 1999. Sensitivity of non-target arthropods to plant protection products: Could Typhlodromus pyri and Aphidiusspp. be used as indicator species? Chemosphere 39:135770

16. Candol MP, Barrett KL, Campbell P, Forster R, Grandy N, et al. 2001. Guidancedocument on regulatory testing and risk assessment procedures for plant protectionproducts with non-target arthropods. In SETAC/ESCORT 2 Workshop Rep. 2123,March 2000, Wageningen

17. Chandel RS, Gupta PR. 1992. Toxicity of diubenzuron and penuron to immaturestages of Apis cerana indica F and Apis mellifera L. Apidologie 23:46573

18. Charleston DS, Kr R, Dicke M, Vet LEM. 2005. Impact of botanical pesticides derivedfrom Melia azedarach and Azadirachta indica on the biology of two parasitoid species ofthe diamondback moth. Biol. Control 33:13142

19. Claver MA, Ravichandran B, Khan MM, Ambrose DP. 2003. Impact of cypermethrinon the functional response, predatory and mating behaviour of a non-target potentialbiological control agent Acanthaspis pedestris (Stal) (Het., Reduviidae). J. Appl. Entomol.127:1822

20. Colin ME, Bonmatin JM, Moineau I, Gaimon C, Brun S, Vermandere JP. 2004. Amethod to quantify and analyze the foraging activity of honey bees: relevance to thesublethal effects induced by systemic insecticides. Arch. Environ. Contam. Toxicol. 47:38795

21. Consoli FL, Parra JRP, Hassan SA. 1998. Side effects of insecticides used in tomatoelds on the egg parasitoid Trichogramma pretiosum Riley (Hym., Trichogrammatidae), anatural enemy of Tuta absoluta (Meyrick) (Lep., Gelechiidae). J. Appl. Entomol. 122:4347


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


22. Corrales N, Campos M. 2004. Population, longevity, mortality and fecundity of Chrysop-erla carnea (Neuroptera, Chrysopidae) from olive orchards with different agriculturalmanagement systems. Chemosphere 57:161319

23. Cox RL, Wilson WT. 1987. The behavior of insecticide-exposed honey bees. Am. BeeJ. 127:11819

24. Croft BA. 1990. Arthropod Biological Control Agents and Pesticides. New York: Wiley.723 pp.

25. Dechaume-Moncharmont FX. 2003. Butinage collectif chez labeille Apis mellifera L.: etudetheorique et experimentale. PhD thesis. Univ. Pierre-et-Marie-Curie, Paris. 309 pp.

26. Dechaume-Moncharmont FX, Decourtye A, Hennequet-Hantier C, Pons O, Pham-Dele`gue MH. 2003. Statistical analysis of honeybee survival after chronic exposure toinsecticides. Environ. Toxicol. Chem. 22:308894

27. Demonstratesthat oxidativemetabolism(cytochromeoxidase) in thecalyces of themushroom bodycan be increasedafter a pesticideexposure, while theolfactory memoryis affected.

27. Decourtye A, Armengaud C, Renou M, Devillers J, Cluzeau S, et al. 2004. Imi-dacloprid impairs memory and brain metabolism in the honeybee (Apis melliferaL.). Pestic. Biochem. Phys. 78:8392

28. Decourtye A, Devillers J, Cluzeau S, Charreton M, Pham-Dele`gue MH. 2004. Effectsof imidacloprid and deltamethrin on associative learning in honeybees under semi-eldand laboratory conditions. Ecotoxicol. Environ. Safety 57:41019

29. Decourtye A, Devillers J, Genecque E, Le Menach K, Budzinski H, et al. 2005. Com-parative sublethal toxicity of nine pesticides on olfactory learning performances of thehoneybee Apis mellifera. Arch. Environ. Contam. Toxicol. 48:24250

30. Decourtye A, Lacassie E, Pham-Dele`gue MH. 2003. Learning performances of honey-bees (Apis mellifera L.) are differentially affected by imidacloprid according to the season.Pest Manag. Sci. 59:26978

31. Decourtye A, Pham-Dele`gue MH. 2002. The proboscis extension response: assessingthe sublethal effects of pesticides on the honey bee. See Ref. 48a, pp. 6784

32. Decourtye A, Tisseur M, Tasei JN, Pham-Dele`gue MH. 2005. Toxicite et risques liesa` lemploi de pesticides chez les pollinisateurs: cas de labeille domestique. In EnjeuxPhytosanitaires pour Lagriculture et Lenvironnement au XXI Sie`cle, ed. C Regnault-Roger,pp. 28399. Paris: Lavoisier. 1013 pp.

33. Delabie J, Bos C, Fonta C, Masson C. 1985. Toxic and repellent effects of cypermethrinon the honeybee: laboratory, glasshouse and eld experiments. Pestic. Sci. 16:40915

34. Delpuech JM, Bardon C, Bouletreau M. 2005. Increase of the behavioral response tokairomones by the parasitoid wasp Leptopilina heterotoma surviving insecticides. Arch.Environ. Contam. Toxicol. 49:18691

35. Delpuech JM, Frey F, Carton Y. 1996. Action of insecticides on the cellular immunereaction of Drosophila melanogaster against the parasitoid Leptopilina boulardi. Environ.Toxicol. Chem. 15:226771

36. Delpuech JM, Froment B, Fouillet P, Pompanon F, Janillon S, Bouletreau M. 1998.Inhibition of sex pheromone communications of Trichogramma brassicae (Hymenoptera)by the insecticide chlorpyrifos. Environ. Toxicol. Chem. 17:110713

37. Highlights thateven insecticidedoses inducing noapparent mortality(LD0.1) induceperturbations inbehaviors involvedin the reproductionof natural enemies.

37. Delpuech JM, Gareau E, Terrier O, Fouillet P. 1998. Sublethal effects of theinsecticide chlorpyrifos on the sex pheromonal communication of Trichogrammabrassicae. Chemosphere 36:177585

38. Delpuech JM, Legallet B, Fouillet P. 2001. Partial compensation of the sublethal ef-fect of delthamethrin on the sex pheromonal communication of Trichogramma brassicae.Chemosphere 42:98591


Ann

u. R

ev. E

ntom

ol. 2

007.

52:8

1-10

6. D

ownl

oade

d fro

m w

ww

.annu

alre

view

s.org

by U

nive

rsid

ade

Fede

ral d

e V

icos

a on

07/

12/1

3. F

or p

erso

nal u

se o

nly.


39. Delpuech JM, Legallet B, Terrier O, Fouillet P. 1999. Modications of thesex pheromonal communication of Trichogramma brassicae by a sublethal dose ofdeltamethrin. Chemosphere 38:72939

40. Delpuech JM, Meyet J. 2003. Reduction in the sex ratio of the progeny of a parasitoidwasp (Trichogramma brassicae) surviving the insecticide chlorpyrifos. Arch. Environ. Con-tam. Toxicol. 45:2038

41. Delpuech JM, Tekinel-Ozalp P. 1991. Epigenetic inuences of insecticide on host-parasitoid relations. Redia 74:41724

42. DeRuijter A, VanderSteen J. 1987. A eld study on the effect on honeybee brood ofinsegar (fenoxycarb) applied on blooming apple orchards. Apidologie 18:35657

43. Desneux N, Denoyelle R, Kaiser L. 2006. A multi-step bioassay to assess the effect ofthe deltamethrin on the parasitic wasp Aphidius ervi. Chemosphere 65:1697706

44. Desneux N, Fauvergue X, Dechaume-Moncharmont FX, Kerhoas L, Ballanger Y,Kaiser L. 2005. Diaeretiella rapae limits Myzus persicae populations following applica-tions of deltamethrin in oilseed rape. J. Econ. Entomol. 98:917

45. Shows thatorientation andovipositionbehaviors ofparasitoids may beimpaired by lowdoses oflambda-cyhalothrindepending on thedose, parasitoidexperience, andtype of behavior.

45. Desneux N, Pham-Dele`gue MH, Kaiser L. 2004. Effects of sublethal and lethaldoses of lambda-cyhalothrin on oviposition experience and host searching be-haviour of a parasitic wasp, Aphidius ervi. Pest Manag. Sci. 60:38189

46. Desneux N, Rafalimanana H, Kaiser L. 2004. Dose-response relationship in lethal andbehavioural effects of different insecticides on the parasitic wasp Aphidius ervi. Chemo-sphere 54:61927

47. Desneux N, Ramirez-Romero R, Kaiser L. 2006. Multi-step bioassay to predict recol-onization potential of emerging parasitoids after a pesticide treatment. Environ. Toxicol.Chem. 25:267582

48. Desneux N, Wajnberg E, Fauvergue X, Privet S, Kaiser L. 2004. Sublethal effects ofa neurotoxic insecticide on the oviposition behaviour and the patch-time allocation intwo aphid parasitoids, Diaeretiella rapae and Aphidius matricariae. Entomol. Exp. Appl.112:22735

48a. Devillers J, Pham-Dele`gue MH, eds. 2002. Honey Bees: Estimating the EnvironmentalImpact of

Nicolas Desneux.annu. Rev. Entomol. 2007

Documents

Transcript of Nicolas Desneux.annu. Rev. Entomol. 2007