Suitability of the Pest-Plant System absoluta (: )- for Trichogramma (Hymenoptera: Trichogrammatidae) Parasitoids and Insights for Biological Control Anais Chailleux, Antonio Biondi, Peng Han, Elisabeth Tabone, Nicolas Desneux

To cite this version:

Anais Chailleux, Antonio Biondi, Peng Han, Elisabeth Tabone, Nicolas Desneux. Suitability of the Pest-Plant System Tuta absoluta (Lepidoptera: Gelechiidae)-Tomato for Trichogramma (Hy- menoptera: Trichogrammatidae) Parasitoids and Insights for Biological Control. Journal of Economic Entomology, Entomological Society of America, 2013, 6 (106), pp.2310-2321. ￿10.1603/EC13092￿. ￿hal-02648694￿

HAL Id: hal-02648694 https://hal.inrae.fr/hal-02648694 Submitted on 29 May 2020

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. BIOLOGICAL AND MICROBIAL CONTROL Suitability of the Pest–Plant System Tuta absoluta (Lepidoptera: Gelechiidae)–Tomato for Trichogramma (Hymenoptera: Trichogrammatidae) Parasitoids and Insights for Biological Control

ANAI¨S CHAILLEUX,1,2,3 ANTONIO BIONDI,1,4 PENG HAN,1 ELISABETH TABONE,5 1 AND NICOLAS DESNEUX

J. Econ. Entomol. 106(6): 2310Ð2321 (2013); DOI: http://dx.doi.org/10.1603/EC13092 ABSTRACT The South American tomato leafminer, Tuta absoluta Meyrick (Lepidoptera: Gelechii- dae), is a major pest that has recently invaded Afro-Eurasia. Biological control, especially by Tricho- gramma parasitoids, is considered to be promising as a management tool for this pest. However, further development of Trichogramma-based biocontrol strategies would beneÞt from assessing the impact of released parasitoid offspring on the pest. Under laboratory conditions, we 1) compared the parasitism of Þve Trichogramma species-strains on the pestÐplant system T. absolutaÐtomato, and 2) assessed various biological traits of parasitoids, mass-reared on a factitious host (Ephestia kuehniella Zeller), when developing on T. absoluta. In addition, we evaluated the overall efÞciency of two speciÞc Trichogramma species when released under greenhouse conditions in combination with a common natural enemy in tomato crop, the predator Macrolophus pygmaeus Rambur. Parasitoids emerging from T. absoluta on tomato showed lower parasitism rates and poor biological traits, for example, wing deformations, reduced longevity, when compared with the control reared on the factitious host. Under greenhouse conditions, the parasitoids that developed on T. absoluta after initial releases contributed little to biological control of T. absoluta, and parasitism tended to be lower when the predator was present. However, a slightly higher T. absoluta control level was achieved by combining the predator and release of the parasitoid Trichogramma achaeae Nagaraja and Nagarkatti. This study shows that Trichogramma parasitoids may not build up populations on the T. absolutaÐtomato system, but that Trichogramma parasitoids can be used in combination with M. pygmaeus to enhance biological control of the pest in tomato crops.

KEY WORDS Þtness, generalist predator, invasive species, inundative release, mass-rearing

The South American tomato leafminer, Tuta absoluta logical control is considered for the development of Meyrick (Lepidoptera: Gelechiidae), is one of the environmentally and economically sound pest man- most devastating tomato pests in (Gon- agement tools for this pest in (Desneux et al. tijo et al. 2013). The pest has recently invaded and 2010, Zappala` et al. 2013). Agricultural pests are spread in the Afro-Eurasian continent, and in few largely controlled via topÐdown inßuences of natural years has become a major pest in tomato crops enemies in a wide range of agro-ecosystems (van Dri- (Desneux et al. 2010, 2011a). Appearance of T. ab- esche and Bellows 1996, Hawkins et al. 1997, Symond- soluta has led to extensive insecticide use by tomato son et al. 2002, Lu et al. 2012). However, their use as growers, potentially causing a multitude of undesired biocontrol agents, either through augmentation or side effects on nontarget organisms (Desneux et al. conservation biological control, is often slowed down 2007; Arno´ and Gabarra 2011; Biondi et al. 2012, because their effectiveness in the Þeld is difÞcult to 2013a). Among possible management methods cur- predict. rently studied (Amer et al. 2012, Cagnotti et al. 2012, Parasitoids belonging to the Trichogramma genus Caparros Megido et al. 2012, Cocco et al. 2013), bio- (Hymenoptera: Trichogrammatidae) are generalist egg parasitoids, mainly of Lepidoptera. They are fre- quently used in biological control programs, notably 1 French National Institute for Agricultural Research (INRA), ISA, through inundative releases (Smith 1996, Pintureau 400 Route des Chappes, 06903 SophiaÐAntipolis, France. 2 InVivo AgroSolutions, 83 avenue de la Grande Arme´e, 75782 Paris, 2009, Mills 2010). In addition, inoculative releases of France. Trichogrammatids have been tested on agricultural 3 Corresponding author, e-mail: [email protected]. pests in various crops (Thomson et al. 2003, Hoffmann 4 Department of Agri-Food and Environmental Systems Manage- et al. 2006, Mills 2010). In South America, T. absoluta ment, University of Catania, Via Santa SoÞa 100, 95123 Catania, . 5 French National Institute for Agricultural Research (INRA), is parasitized by Trichogramma pretiosum Riley under UEFM, 90 Chemin Gustave Raymond, 06160 Antibes, France. laboratory conditions (Pratissoli and Parra 2000), and

0022-0493/13/2310Ð2321$04.00/0 ᭧ 2013 Entomological Society of America December 2013 CHAILLEUX ET AL.: SUITABILITY OF T. absoluta FOR Trichogramma 2311

Miranda et al. (1998) reported that nearly 10% of T. To document the possible biocontrol impact of Þve absoluta mortality occurring at the egg stage is because Trichogramma species or strains, hereafter named of natural parasitism by this parasitoid. Trichogram- “strains,” when developing on T. absoluta eggs in the matids have been used in tomato crops against T. tomato crops after inundative releases, we evaluated absoluta through inundative releases both in the na- their parasitism and the subsequent development of tive (mainly with T. pretiosum; Parra and Zucchi 2004, their offspring. Thus, in a Þrst experiment, in the Pratissoli et al. 2005) and the invaded areas (with laboratory, we compared 1) a “released generation”: Trichogramma achaeae Nagaraja and Nagarkatti; Ca- developed in E. kuehniella eggs and offered T. absoluta bello et al. 2012, Calvo et al. 2012). Considering the eggs, with 2) a “Þeld generation”: developed in T. effectiveness of these programs and the natural par- absoluta eggs and offered T. absoluta eggs, and a con- asitism of this pest by Trichogrammatids in South trol treatment: parasitoids developed in E. kuehniella America and in Europe (Desneux et al. 2010, Zappala` eggs and offered E. kuehniella eggs. This aimed at et al. 2012, Biondi et al. 2013b), screenings for more assessing possible reduced parasitoid performance effective species in the invaded area are ongoing when parasitoids switched from the mass-rearing host (Chailleux et al. 2012, Khanh et al. 2012). to the targeted pestÐplant system. In a second exper- Various factors can inßuence the impact of inun- iment, in greenhouse, we tested the overall efÞciency dative releases of mass-reared parasitoids. For in- of caged release and the subsequent generation (that stance, host selection by parasitoids may be inßuenced corresponds to the Þeld generation) of two Tricho- by the characteristics of both the host and host plant gramma strains against T. absoluta. In addition, to as- (Chau and Mackauer 2001, Desneux and RamirezÐ sess the usefulness of Trichogramma parasitoids within Romero 2009). In addition, host preference in para- the framework of current biocontrol programs in to- sitoids tends to correlate with the Þtness gained from mato crops in Europe, we included the predator Ma- the host (van Alphen and Vet 1986, Driessen et al. crolophus pygmaeus Rambur (Hemiptera: Miridae) in 1991, Chau and Mackauer 2001), this being deÞned as the greenhouse experiment because it is widely used for biocontrol of whiteßies on tomato, and is also the “preferenceÐperformance” hypothesis (Jaenike 1978, Desneux et al. 2009b). However, very-low qual- released against T. absoluta because it preys on eggs ity hosts can also be parasitized (Janssen 1989; Heim- (Desneux et al. 2010, Bompard et al. 2013). pel et al. 2003; Desneux et al. 2009a, 2012). Studies on acceptance of Trichogramma reared on factitious host Materials and Methods species toward natural hosts yielded variable results. Biological Materials. The plants used in the exper- El-Wakeil (2007) showed differences in pest parasit- iments were tomato plants, Solanum lycopersicum L. ism, which depended on the rearing host. The size of ÔMarmande,Õ 5 wk old for the laboratory experiment the natal host affected natural host parasitism rates in and 7 wk old for the cage experiment. Plants were the Þeld, and female parasitoids usually accepted host grown in climatic chambers (25 Ϯ 1ЊC, 65% relative eggs of the same size or larger as their natal host (Salt humidity [RH], and a photoperiod of 16:8 [L:D] h); 1940, Nurindah et al. 1999). By contrast, KollikerÐOtt pesticide applications were strictly avoided, and a nu- et al. (2003) reported no preferenceÐperformance re- trient solution was applied daily. A colony of T. ab- lationship for Trichogramma brassicae Bezdenko mass- soluta was set up using greenhouse-collected individ- reared on the factitious host Ephestia kuehniella Zeller uals in July of 2009 (initial number of individuals ϭ when parasitizing the target host, the European corn 190) at the French National Institute for Agricultural borer Ostrinia nubilalis Hu¨ bner (Lepidoptera: Pyrali- Research (INRA), Alenya, France. T. absoluta was dae). reared in growth chambers (25 Ϯ 1ЊC, 70 Ϯ 10% RH, The effectiveness of Trichogramma releases may and a photoperiod of 16:8 [L:D] h) in mesh netting depend not only on the biological characteristics of cages (55 by 75 by 80 cm) containing potted tomato the parasitoid species or strains used, but also on their plants (see Biondi et al. 2013c). Adult were fed interactions with a speciÞc pestÐplant system (Tabone on honey placed on the mesh inside the cages and on et al. 2010, Andrade et al. 2011, Yuan et al. 2012). the plant. The parasitoids used for the experiments Depending on the afÞnity of the parasitoids with the originated from collections (Table 1). Dr. B. Pintureau pestÐplant system, the biocontrol services provided (INRA, Lyon, France) identiÞed the species before by the released parasitoids most likely vary. Tricho- the experiments, and voucher specimens were depos- gramma are mainly used as biocontrol agent through ited at the INRA (SophiaÐAntipolis, France). Tricho- inundative releases, but the impact of Trichogramma gramma euproctidis Girault, and both Trichogramma generations developing within a crop can play a cru- evanescens Westwood strains were chosen because cial role in the success of biological control programs they were shown to be promising strains for biocontrol (Mills 2010). For example, use of T. brassicae for bi- of T. absoluta during a previous study (29 strains ological control of the European corn borer in maize tested, Chailleux et al. 2012). T. pretiosum was con- takes into account the effect of parasitoids developing sidered because it is mass reared and released to con- in the Þeld for the long-term control of the pest (Pin- trol this pest in South America. T. achaeae was tested tureau 2009). Additional mass-releases may be re- because it has already been commercialized in Europe duced and overall crop protection costs can be low- and Africa against T. absoluta. Colonies of parasitoids ered. were reared on ultraviolet-irradiated eggs of a substi- 2312 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 106, no. 6

Table 1. Country of origin, initial host plant, and host and year of collection of the five Trichogramma strains tested

Species Geographic origin Host plant Most species Year of collection T. achaeae Nagaraja and Nagarkatti, 1969 Canaries islands Tomato Chrysodeixis chalcites 2010 T. euproctidis Girault, 1911 Switzerland Ñ Ñ Ñ T. evanescens Westwood, 1833 (1) Northern France Vine Lobesia botrana 1990 T. evanescens Westwood, 1833 (2) Northern France Caulißower Argyrotaenia sphaleropa 2002 T. pretiosum Riley, 1876 Uruguay Vine Argyrotaenia sphaleropa 1995 tute host, E. kuehniella (18 Ϯ 1ЊC, 70 Ϯ 10% RH, and Control (Treatment 3) and Released Generations a photoperiod of 12:12 [L:D] h), until used in the (Treatment 1). Females emerging from E. kuehniella experiments. Rearing was carried out in glass tubes eggs (obtained as described in the “Biological Mate- (4.5 cm in length and 0.7 cm in diameter), and the rials” section) were used both for the control treat- parasitoids were fed on honey droplets. E. kuehniella ment, when releasing the females on E. kuehniella eggs were glued on a piece of cardboard (3 by 10 mm) eggs, and for the released generation treatment, when with 10% arabic gum (Pizzol et al. 2010). Parasitoids releasing the female on T. absoluta eggs. Therefore, were maintained for at least three generations at the young mated females were released individually in temperature of 25ЊConE. kuehniella eggs before start- glass tube containing 30Ð40 eggs of E. kuehniella (the ing experiments. same technique as previously described for the rear- In the laboratory trials, the T. absoluta eggs used ing) and fed with honey droplets during 24 h to mea- were 0Ð12 h old, and the parasitoids were 0Ð24 h old sure the control (“rearing generation”) parameters. (as both host and parasitoid ages can play a role in However, for the released generation, young mated development of the Trichogramma offspring, see Piz- females were presented individually with 30Ð40 T. zol et al. 2012). To obtain parasitoid females that had absoluta eggs on a tomato leaßet in ventilated Plexiglas developed in E. kuehniella eggs, females from the col- tubes (4 cm in diameter and 14 cm in length) with onies were released on E. kuehniella eggs in the same honey during 24 h. The stems of leaßet, sticking out of conditions as described for the rearing, but female the tube, were planted into ßoral foam for watering. emergence date was checked precisely. Parasitoid fe- Field Generations (Treatment 2). To evaluate the males developed on T. absoluta eggs were obtained by Þeld generation parameters, young Trichogramma fe- releasing parasitoids from the colonies on T. absoluta males emerging from T. absoluta-parasitized eggs (ob- eggs on tomato leaßets for 12 h inside ventilated Plexi- tained as described in the “Biological Materials” sec- glas tubes (4 cm in diameter and 14 cm in length). The tion) were presented individually to 30Ð40 T. absoluta parasitoids used in all the experiments were collected eggs on a tomato leaßet in the ventilated tube (with daily from these tubes, namely those containing par- honey droplets as food source) during 24 h. Thus, we asitized eggs of E. kuehniella or T. absoluta. M. pyg- used the same oviposition conditions as the released maeus specimens for the greenhouse experiment were generation, except that females had previously devel- reared on E. kuehniella eggs on artiÞcial substrate by oped in T. absoluta eggs vs. E. kuehniella eggs. the private company Biotop (Valbonne, France) and For the three treatments, tubes containing parasit- after being purchased, they were reared for at least ized eggs were kept in a climatic chamber (25ЊC, 70 Ϯ two generations in the laboratory. Rearing was carried 10% RH, and a photoperiod of 16:8 [L:D] h) and out in cages (55 by 75 by 80 cm) covered with a mesh maintained until offspring emergence for 15 d. For in climatic chambers (23 Ϯ 1ЊC, 70 Ϯ 5% RH, and a each treatment, the number of parasitized eggs (black photoperiod of 16:8 [L:D] h) on tobacco plant, and E. eggs) and of aborted eggs (white and unhatched eggs) kuehniella eggs were provided ad libitum. New eggs was counted after5dofparasitization. The sex ratio were added twice a week, and tobacco plants were (females per total) and the wings deformation occur- changed monthly. Adult predators used in the exper- rence (absent, stump, or crumple wings) of offspring iments were 1Ð3 d old. individuals (F1) were recorded 2 d after emergence, Impact of T. absoluta–Tomato Pest–Plant System namely 15 d after the parasitization. Ten to 15 repli- on Parasitism and Offspring Development. The ex- cates were carried out per parasitoid strain and per periment was conducted in growth chambers (25ЊC, treatment. 70 Ϯ 10% RH, and a photoperiod of 16:8 [L:D] h). Two Impact of the Host on the Longevity. We assessed different combinations were tested mimicking the de- the longevity of female parasitoids that had developed velopment of two generations of Trichogramma and in T. absoluta eggs (ϭthe Þeld generation) and in E. the rearing situation (control treatment): 1) the re- kuehniella eggs as a control. Females were obtained as leased generation (developed in E. kuehniella eggs on described in the “Biological Materials” section. These cardboard and offered T. absoluta eggs on tomato were placed in glass tubes (4.5 cm in length and 0.7 cm leaßets), 2) the Þeld generation (developed in T. ab- in diameter), closed with a cotton ball and provided soluta eggs on tomato leaßets and offered T. absoluta with honey or left without in climatic chambers (25ЊC, eggs on tomato leaßets), and 3) parasitoids in the 70 Ϯ 10% RH, and a photoperiod of 16:8 [L:D] h). rearing situation were used as a control. Control in- Twenty to 25 females per strain and per rearing host dividuals developed on E. kuehniella eggs and were were tested. The survival was checked daily with a offered E. kuehniella eggs on cardboard. binocular microscope, and parasitoids were consid- December 2013 CHAILLEUX ET AL.: SUITABILITY OF T. absoluta FOR Trichogramma 2313 ered dead when they did not react after being touched of water. Five days later, leaßets were collected and by a Þne paint brush. parasitized eggs (black eggs) and predated eggs Cage Experiment. A cage experiment was carried (empty crumpled eggs) were counted under a bin- out with the aim of testing the efÞciency of the Þeld ocular microscope and percent parasitism and percent generation under realistic conditions. The experiment rates were calculated. Then, the leaßets was carried out in cages (100 cm in height, 70 cm in were put back inside their respective cages. width, and 100 cm in length, made with -proof To evaluate parameters in the same way for the Þeld mesh) that were placed in a greenhouse at the INRA generation, produced by the released parasitoid, we Sophia AgroBiotech Institute (SophiaÐAntipolis, provided new sentinel leaßet at the Þeld generation France). The greenhouse was similar to the ones used emergence time. The Trichogramma juveniles (egg to by many tomato growers in France and more broadly adult) developmental time is8dat30ЊC (Foerster and in Southern Europe (semicontrolled temperatures, Foerster 2009); therefore, the new leaßets were min Ͻ mean temperature Ͻ max: 15.5 Ͻ 24.2 Ͻ 37.4ЊC; placed in each cage 8 d after the Trichogramma re- min Ͻ mean RH Ͻ max, 18.0 Ͻ 63.0 Ͻ 90.0%; natural leases, following the same procedure previously de- ambient light: mid-July mid-August 2011). The two scribed. The leaßets were collected 7 d later to enable Trichogramma species tested, T. achaeae and T. eu- an effective assessment of overall parasitism under- proctidis, which showed the highest parasitism at the taken by the second-generation parasitoid females, as Þeld generation in the laboratory (see “Results” sec- they emerged for up to 3 d and adults longevity in tion), were chosen for this experiment. Both species greenhouse conditions is Յ3 d (A. C. and N. D., un- were tested alone and in combination with the pred- published data). Parasitized eggs and predated eggs ator M. pygmaeus. The predator was also tested alone were counted under a binocular microscope as for the and a control with the pest alone was carried out. The released generation. Plants were destructively sam- following six treatments or combinations were stud- pled at the end of the experiment (after 16 d). The ied: 1) T. absoluta, 2) T. absoluta ϩ M. pygmaeus, 3) T. ϩ ϩ number of T. absoluta larvae and M. pygmaeus juve- absoluta T. achaeae, 4) T. absoluta T. euproctidis, niles were counted in each cage on four plants chosen 5) T. absoluta ϩ T. achaeae ϩ M. pygmaeus, and 6) T. ϩ ϩ randomly. absoluta T. euproctidis M. pygmaeus. The exper- Statistical Analyses. All statistical analyses were per- iment lasted 16 d to allow Þrst Þeld generation obser- formed using the R software with the multcomp and vation, and six replicates (cages) were carried out per agricolae packages. For the laboratory experiments, treatment. the number of parasitized and aborted eggs were an- Eight tomato plants (development stage from seven alyzed using a generalized linear model (GLM) based to eight fully expanded leaves) were placed into the on Poisson distributed data with a log link function. cages. First, 22-d-old T. absoluta adults (10 males and Sex ratio (females per total) and wing malformations 10 females) were released per cage. After 24 h, pred- were analyzed using a GLM designed for modeling ators, parasitoids, or both, were introduced into the cages. Depending on the treatment, 400 Tricho- binomial data with a logit link function. For these four gramma adults and eight pairs of M. pygmaeus (1:1, parameters the effect of the factors Rearing host (host male/female) were released. These release rates were egg used to rear parasitoid), Oviposition host (host chosen according to the recommendations of produc- egg parasitized by the females in the experiments), ers of natural enemies. To release Trichogramma para- and Trichogramma strain were tested. Longevity was sitoids, we mimicked a Þeld release similar to what is also analyzed using a GLM, but it was designed for recommended by the producers. One open glass tube modeling gamma data with an inverse link function. containing a small cardboard strip bearing 400 para- The effect of the factors Rearing host, Honey supply, sitized E. kuehniella eggs was placed in the center of and Trichogramma strain were tested. each cage. Honey droplets were provided as food For the cage experiments, the percentages of par- source to the parasitoids. A quality control carried out asitism, aborted eggs, and predated eggs were ana- on a sample of 1,000 parasitized eggs per parasitoid lyzed using a GLM designed for modeling binomial strain (randomly selected from the stock of parasitized data. For the parasitism and aborted eggs, the effect of eggs used in this experiment) showed that parasitoid the factors, M. pygmaeus, Trichogramma generation emergence was always Ͼ90%. Individuals showed a (released or Þeld generations) and Trichogramma sex ratio of Ϸ0.70 (females/total) for both species. strain factors were tested. For the predated eggs the Therefore, we assumed that releasing 400 parasitized effect of the factors Trichogramma strain and Tricho- eggs corresponded to a release of Ϸ252 parasitoid gramma generation were tested. A GLM designed for females. Poisson distributed data were used to analyze the To facilitate the parasitization monitoring, we number of T. absoluta larvae and M. pygmaeus juve- added four sentinel tomato leaßets bearing 30 T. ab- niles. For T. absoluta larvae, the factor Trichogramma soluta eggs (15 on adaxial and 15 on abaxial surfaces of strain and M. pygmaeus were tested, and for the M. each leaßet) before the parasitoids release. These leaf- pygmaeus juvenile, the Trichogramma strain factor was lets were placed on the top of a stake so that they tested. Finally, multiple comparisons were performed always touched a tomato leaf in the upper plant part. using a Fisher least signiÞcant difference (LSD) post The stems of these leaßets were put inside a tube full hoc test for both laboratory and cage experiments. 2314 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 106, no. 6

Fig. 1. Mean (ϮSEM) number of parasitized eggs per parasitoid female for the Þve Trichogramma strains tested. Released generation: developed in E. kuehniella eggs and offered T. absoluta eggs; Þeld generation: developed in T. absoluta eggs and offered T. absoluta eggs; control: developed in E. kuehniella eggs and offered E. kuehniella eggs. Post hoc tests (Fisher LSD test) were carried out separately for each strain tested. For each generation tested (released and Þeld generations) and for the control, the histograms bearing the same letter are not signiÞcantly different at P Ͻ 0.05.

Results evanescens 2 (Fig. 1). Parasitism of all the strains was signiÞcantly lower when females were offered T. ab- Impact of T. absoluta–Tomato Pest–Plant System on Parasitism and Offspring Development. There was soluta (released and Þeld generations) than E. kueh- a signiÞcant effect of the Rearing host (host in which niella eggs (control). Parasitism rates of T. euproctidis ϭ Ͻ and T. evanescens 2 were also lower when females had ovipositing parasitoid developed; F(1,181) 80.980; P 0.001) and of the Oviposition host (host parasitized in developed on T. absoluta (Þeld generation) than when ϭ Ͻ females had developed on E. kuehniella eggs (control the experiment; F(1,182) 91.780; P 0.001) on the number of parasitized eggs. Parasitism also varied sig- and released generation; Fig. 1). ϭ niÞcantly among the Trichogramma strains (F(4,177) The host egg species in which the females devel- 8.902; P Ͻ 0.001). The number of parasitized eggs oped (Rearing host) signiÞcantly affected the sex ratio ϭ Ͻ ranged from 35.5 Ϯ 4.4 for the rearing generation of T. of their offspring (Fig. 2; F(1,127) 5.149; P 0.025). euproctidis to 3.0 Ϯ 1.0 for the Þeld generation of T. Indeed, the T. achaeae Þeld generation was more male-

Fig. 2. Mean sex ratio (ϮSE; females per total) of offspring for the Þve Trichogramma strains tested. Released generation: females developed in E. kuehniella eggs and offered T. absoluta eggs; Þeld generation: females developed in T. absoluta eggs and offered T. absoluta eggs; control: females developed in E. kuehniella eggs and offered E. kuehniella eggs. Post hoc tests (Fisher LSD test) were carried out separately for each strain tested. For each generation tested (released and Þeld generations) and for the control, the histograms bearing the same letter are not signiÞcantly different at P Ͻ 0.05. December 2013 CHAILLEUX ET AL.: SUITABILITY OF T. absoluta FOR Trichogramma 2315

Fig. 3. Proportion (ϮSE) of offspring with wings malformation: absent, stump, or crumple wings, for the Þve Tricho- gramma strains tested. Released generation: females developed in E. kuehniella eggs and offered T. absoluta eggs; Þeld generation: females developed in T. absoluta eggs and offered T. absoluta eggs; control: females developed in E. kuehniella eggs and offered E. kuehniella eggs. Post hoc tests (Fisher LSD test) were carried out separately for each strain tested. For each generation tested (released and Þeld generations) and for the control, the histograms bearing the same letter are not signiÞcantly different at P Ͻ 0.05. biased than its generations developing on E. kuehniella honey, longevity did not exceed 4 d (for T. pretiosum eggs. By contrast, the Oviposition host factor did not emerged from E. kuehniella eggs; Fig. 4B). ϭ Ͻ signiÞcantly affect the sex ratio (F(1,126) 0.0736; P Cage Experiment. Trichogramma strain and M. pyg- 0.787). The sex ratio of the offspring also varied sig- maeus had signiÞcant effects on the percentage of niÞcantly among strains tested (signiÞcant Tricho- parasitism observed on the sentinel leaßets (Fig. 5; ϭ Ͻ ϭ Ͻ ϭ ϭ gramma strain factor; F(4,122) 4.937; P 0.001). The F(1,29) 22.648; P 0.001 and F(1,30) 8.981; P sex ratio signiÞcantly differed among generations only 0.007, respectively). The parasitism of the Þeld gen- in the case of T. achaeae when strains were tested eration was signiÞcantly lower than of the released ϭ Ͻ individually (Fig. 2; Fisher LSD test). The sex ratio of generation (F(1,28) 33.542; P 0.001). The highest offspring from the Þeld generation was signiÞcantly level of percent parasitism was obtained with T. lower than in the two other treatments (control and achaeae alone. On the contrary, no parasitism was released generation). We observed similar trends for observed for the Þeld generation of T. euproctidis in T. evanescens 1 (Fig. 2), but differences among treat- the presence of M. pygmaeus (Fig. 5). ment were not signiÞcant either for T. evanescens 1or In the three treatments with M. pygmaeus, the num- the three other strains (most likely owing to a low ber of eggs eaten did not vary signiÞcantly depending number of offspring individuals in the Þeld genera- on the Trichogramma strain (T. achaeae or T. euproc- ϭ ϭ tion). tidis or absence) (F(2,21) 0.080; P 0.923). No Trichogramma strain, Rearing host, and Oviposition signiÞcant effect of the Trichogramma generation host factors all had a signiÞcant impact on wing mal- factor (released, Þeld generations) was observed ϭ ϭ ϭ formations occurring in offspring (F(4,122) 39.107; (F(1,20) 0.876; P 0.363). T. absoluta eggs were Ͻ ϭ Ͻ ϭ Ϯ P 0.001; F(1,127) 24.765; P 0.001; F(1,126) 7.479; preyed on an average 71.7 3.6%, ranged between 25 P ϭ 0.007, respectively). The strain presenting the and 96.6%. The factors Trichogramma strain (T. highest proportion of offspring with malformed wings achaeae, T. euproctidis or absence), M. pygmaeus, and was T. euproctidis (Fig. 3), the highest proportion was Trichogramma generation had a signiÞcant effect on ϭ ϭ observed for the T. euproctidis Þeld generation, with the number of aborted eggs (F(2,44) 3.918; P 0.029; Ϯ ϭ ϭ ϭ ϭ 38 4% of offspring with wing malformations (Fig. 3). F(1, 46) 12.429; P 0.001; and F(1,43) 12.429; P Impact of the Host System on the Longevity. The 0.001, respectively). However, the overall percentages effects of the Rearing host (␹2 ϭ 173.319; df ϭ 1; P Ͻ of aborted eggs were low throughout the course of the 0.001) and the Honey supply factors (␹2 ϭ 173.319; experiments in cages (Յ5%). The maximum values df ϭ 1; P Ͻ 0.001) were signiÞcant on female parasitoid were observed in groups with Trichogramma parasi- longevity (Fig. 4). The Trichogramma strain factor also toids without M. pygmaeus (released generation). had a signiÞcant effect on Trichogramma longevity At the end of the experiment, an overall average of (␹2 ϭ 28.530; df ϭ 4; P Ͻ 0.001), with a maximum of 18 d 2.12 Ϯ 0.50 M. pygmaeus juveniles per plant was re- on average for T. evanescens 1 developed in E. kueh- corded, ranging from one to three per plant. No sig- niella egg and supplied with honey (Fig. 4A). Without niÞcant differences related to the Trichogramma pres- 2316 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 106, no. 6

Fig. 4. Mean (ϮSE) longevity of the Þve Trichogramma strains tested for females developed in E. kuehniella eggs (equivalent to the control and the released generation); and for females developed in T. absoluta eggs (equivalent to the Þeld generation); (A) in presence of honey as food source and (B) without any food sources. Post hoc tests (Fisher LSD test) were carried out separately for each strain tested. For each emergence host tested the histograms bearing the same letter are not signiÞcantly different at P Ͻ 0.05.

ϭ ϭ ence were observed (F(2,9) 0.022; P 0.978). The a control treatment, which is reared in the laboratory number of T. absoluta larvae was inßuenced by Tricho- on the factitious host E. kuehniella. The highest par- gramma presence, though only numerically different asitism was recorded for the generations under the ϭ ϭ (Fig. 6; F(2, 21) 2.870; P 0.086), and it was signif- standard rearing conditions on the factitious host. Par- ϭ ϭ icantly affected by M. pygmaeus (F(1,20) 9.178; P asitism decreased when oviposition occurred on the 0.008). The highest number of larvae was obtained in novel pestÐplant system. Among the three treatments, the control (T. absoluta alone) and the lowest infes- the lowest parasitism was recorded for the Þeld gen- tation was observed in the treatment combining T. eration although it was not consistent for T. euproctidis achaeae and M. pygmaeus, although it was not signif- and T. pretiosum, which showed same parasitism levels icantly different from the treatments with T. achaeae in both released and Þeld generations. alone and M. pygmaeus alone and with T. euproctidis Overall, parasitism efÞciency decreased on the T. (Fig. 6). absolutaÐtomato system, so that this pestÐplant system appeared unsuitable for the Trichogramma strains tested. Our results showed a decrease of female prog- Discussion eny sex ratio, increased malformations of individuals The parasitism of Þve Trichogramma strains, belong- (deformed wings), and reduced longevity (for T. ing to four species, on T. absoluta was evaluated in the achaeae, T. evanescens 1, and T. pretiosum, with up to laboratory by testing two different parasitoid gener- a 6-d reduction) in the pestÐplant system studied com- ations, particularly the ones potentially occurring in pared with rearing conditions. tomato crop after inundative releases of parasitoids. Plausibly, differences among parasitoid efÞciency The parasitism of T. absoluta eggs on tomato by para- recorded for the control and the released and Þeld sitoids from the released and Þeld generations was generations are linked to host physiological suitability related to the parasitism observed in individuals from for the Trichogramma strains tested. Although the host December 2013 CHAILLEUX ET AL.: SUITABILITY OF T. absoluta FOR Trichogramma 2317

Fig. 5. Mean (ϮSE) percent parasitism of T. absoluta eggs obtained on sentinel leaßets placed in greenhouse cages for the four treatments with Trichogramma releases, for the released generation (females developed in E. kuehniella eggs) and the Þeld generation (females developed in T. absoluta eggs). Two types of post hoc tests (Fisher LSD test) have been done separately 1) each strain separately, generations bearing the same lowercase letters are not signiÞcantly different for P Ͻ 0.05; and 2) each generation separately, strains bearing the same capital letters are not signiÞcantly different at P Ͻ 0.05. sizeÐparasitoid Þtness relationship is not always con- threefold larger than those of T. absoluta. Therefore, sistent among parasitoids species (Harvey et al. 2013), the decrease in parasitoids quality (parasitism efÞ- parasitoids largely depend on the quality and quantity ciency, sex ratio, longevity, and wing malformations) of resources provided by the host for their larval de- that developed in T. absoluta eggs may result from the velopment (Mackauer 1996, Pennacchio and Strand small size of the host eggs, for example, smaller para- 2006). A close relationship has been generally ob- sitoids emerged from T. absoluta eggs (A. C., unpub- served between the size of the host parasitized and the lished data). Wing malformations in Trichogramma size and Þtness of parasitoid offspring, for example, have been documented in other studies. Wings and egg load of emerging females (Harvey et al. 2013). In abdominal malformations were observed in Tricho- addition, offspring sex ratio and longevity can be gramma parasitoids reared on artiÞcial diet (Consoli linked to the size of parasitoid individual (Fidgen et al. and Parra 1996), and sexual dimorphism (apterous 2000) and Trichogramma parasitoids commonly show males and fully winged females) of T. semblidis was preference for relatively large host eggs (Roriz et al. described by Salt (1937) when the parasitoids devel- 2006). T. achaeae is reared on E. kuehniella eggs by oped in small eggs (Sialis lutaria L. [Megaloptera: producers of natural enemies, and eggs of this host are Sialidae]) vs. E. kuehniella eggs. Finally, we cannot

Fig. 6. Mean (ϮSE) number of T. absoluta larvae per plant 16 d after insect releases in greenhouse cages for the six treatments tested. Values marked with the same letter do not differ signiÞcantly at P Ͻ 0.05 (Fisher LSD test). 2318 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 106, no. 6 exclude that parasitoid rearing, at least concerning gramma efÞciency (Kauffman and Kennedy 1989, Far- Trichogramma for many generations on a factitious rar et al. 1994). host, could have affected parasitoids capacity to attack Our results suggest that the pestÐplant system T. other hosts (Hopper and Roush 1993, Hoffmann et al. absolutaÐtomato shows unfavorable characteristics for 2001, Chailleux et al. 2012). the effective establishment of Trichogramma parasi- Understanding the suitability of the pestÐplant sys- toids in tomato crops. This contrasts with other pestÐ tem that underpins the parasitism pattern in the lab- plant systems in which Trichogramma Þeld parasitoid oratory has important practical implications, as high- generations provide biocontrol services to some ex- lighted by our greenhouse cage experiments, where tent after the initial release in the crop, for example, the parasitism activity of the Þeld generation of both on the European corn borer in maize crop (Thomson Trichogramma species was quite low in cages under et al. 2003, Hoffmann et al. 2006, Pintureau 2009). greenhouse conditions. However, overall T. absoluta Thus, the expected biocontrol service provided by the control achieved by T. achaeae was similar to those Trichogramma Þeld generations in tomato crop against achieved with M. pygmaeus when these natural ene- T. absoluta should be considered low. Trichogramma- mies were used alone (Ϸ46% decrease in T. absoluta based biocontrol programs targeting T. absoluta on larvae per plant, see Fig. 6), though the experiment tomato crop therefore may need to rely on periodical was carried out over a short period as infestation levels inundative releases of the parasitoid, at least for were measured 16 d after the pest was introduced into Trichogramma strains so far identiÞed as potential nat- the cages. As opposed to the results under laboratory ural enemies of T. absoluta in Europe. conditions, T. euproctidis both released and Þeld gen- We also evaluated the effect of the generalist pred- erations (with or without predator) showed the low- ator M. pygmaeus presence on the effectiveness of T. est parasitism rates in the greenhouse cage experi- achaeae and T. euproctidis releases against T. absoluta. ments. This species was not able to control T. absoluta M. pygmaeus decreased the abundance of parasitoids populations; there was no signiÞcant difference in T. hence their effectiveness on T. absoluta, both through absoluta density compared with the control (see Fig. direct competition for the resource ( eggs), and 6). These results indicate that populations of parasi- direct predation on Trichogramma-parasitized eggs toids released in the tomato crop may disappear soon (Desneux et al. 2011b, Chailleux et al. 2013). Parasit- ism exerted by the released and Þeld generations on after the releases. Therefore, the impact of Tricho- both parasitoid strains was lower when the predator gramma Þeld generation(s) on T. absoluta population was present. The highest T. absoluta control level growth could be considered negligible for biological (Ͻ68% larvae per plant than in the control) was control, at least with strains tested in our study. achieved when combining the parasitoid T. achaeae This study further documents the possible im- with the predator, although it was not signiÞcantly pacts of Trichogramma generations when parasi- different from T. achaeae or M. pygmaeus when tested toids emerged in the greenhouse from T. absoluta- alone. In addition, the presence of M. pygmaeus parasitized eggs. Poor effectiveness of the Þeld nymphs suggested effective predator establishment generation parasitoid in T. absoluta parasitism may be after introduction into cages. Previous studies re- because of various factors. First, continuous tomato ported that supplementary releases of T. achaeae in plant growth during the course of the cage experiment tomato crops, in which another mirid predator (Ne- has led most likely to an increasing plant architecture sidiocoris tenuis Reuter) was released during the crop- complexity between the beginning of the experiment, ping cycle, improved the biological control of T. ab- in which the parasitoids were Þrst released (released soluta (Desneux et al. 2010; Cabello et al. 2012). Our generation), and the end of the experiment, when results back up these conclusions. However, addi- parasitoids developing in T. absoluta-parasitized eggs tional releases of T. achaeae did not improve control of emerged. Increased habitat complexity is known to T. absoluta when N. tenuis was inoculated by preplant negatively affect efÞciency (host searching, dispersal releases in nurseries (Calvo et al. 2012), namely when ability, or both) of natural enemies in Þnding hosts or the predator populations were already well estab- prey (Rutledge and OÕNeil 2005, Tabone et al. 2012). lished in the tomato crop before arrival of T. absoluta. Second, parasitism effectiveness may have decreased Nevertheless, such a strategy based on very early es- because of 1) the small size of the parasitoids emerging tablishment of N. tenuis in nursery plants often leads from T. absoluta eggs (such small host eggs make to use of insecticides when the omnivorous predator parasitoid females mature fewer eggs, Kazmer and reaches high densities. Indeed, this predator is known Luck 1995), and 2) increased sex ratio in offspring to inßict damage to tomato plants, causing necrotic produced (more males in the population). Third, para- rings on stems and ßowers and punctures in fruits sitoid mobility may have been reduced on tomato (Calvo et al. 2009, Castan˜ e´ et al. 2011). plants because of the high pubescence of tomato. Our study suggests that the integration of Tricho- Foraging ability was consequently reduced (Carrillo gramma parasitoids for biological control strategies et al. 2008). Trichogramma parasitoids are known to be against T. absoluta in tomato crops should rely on the poor ßyers and to forage primarily by walking and use of several inundative releases and should also take jumping on substrates (Keller 1987, Olson and Andow into account the population levels of generalist pred- 2006). In addition, previous studies reported that to- ators either artiÞcially released, naturally occurring in mato trichomes have negative effects on Tricho- the tomato crop, or both. December 2013 CHAILLEUX ET AL.: SUITABILITY OF T. absoluta FOR Trichogramma 2319

Acknowledgments Effects of X-rays on Tuta absoluta for use in inherited sterility programmes. J. Pest Sci. 85: 413Ð421. We thank Ce´cile Thomas, Philippe Bearez, and EttyÐAm- Caparros Megido, R., E. Haubruge, and F. J. Verheggen. bre Colombel (INRA SophiaÐAntipolis) for technical assis- 2012. First evidence of deuterotokous parthenogenesis tance; Me´line Beal and Jacques Frandon (Biotop, InVivo in the tomato leafminer, Tuta absoluta (Meyrick) (Lep- AgroSolutions) for providing biological materials; ANRT idoptera: Gelechiidae). J. Pest Sci. 85: 409Ð412. (National Association of Research and Technology) and Carrillo, D., J. E. Pena, and J. L. Capinera. 2008. Effect of InVivo AgroSolutions for funding to A. C. (Ph.D. fellowship); host plants on successful parasitism by Haeckeliania spe- the University of Catania for funding to A. B. (Ph.D. fellow- rata (Hymenoptera: Trichogrammatidae) on Diaprepes ship); the French ministry of agriculture for funding to N. D. abbreviatus (Coleoptera: Curculionidae) Eggs. Environ. (CASDAR [Account for Agricultural and Rural Develop- Entomol. 37: 1565Ð1572. ment] project 10063); and the anonymous referees as well as Castan˜ e´, C., J. Arno, R. Gabarra, and O. Alomar. 2011. Plant the handling editor for their helpful comments. damage to vegetable crops by zoophytophagous mirid predators. Biol. Control 59: 22Ð29. Chailleux A., N. Desneux, J. Seguret, H. Do Thi Khanh, P. References Cited Maignet, and E. Tabone. 2012. Assessing European Egg parasitoids as a mean of controlling the invasive South Amer- Amer, R.A.M., A. E. Hatem, and M. A. El-Sanady. 2012. ican tomato pinworm Tuta absoluta. Plos ONE 7: e48068. Microbial control of the tomato leaf-miner, Tuta absoluta (doi: http://dx.doi.org/10.1371/journal.pone.0048068). (Povolny) (Lepidoptera: Gelechiidae) by biopesticides Chailleux, A., J. Pizzol, E. Amiens–Desneux, R. Ramirez– in relation to some biological parameters. Egypt. J. Biol. Romero, and N. Desneux. 2013. Potential for combining Pest Control 22: 161Ð167. indigenous natural enemies for biological control of the Andrade, G. S., D. Pratissoli, L. P. Dalvi, N. Desneux, and key invasive tomato pest, Tuta absoluta, in Afro-Eurasia. H.J.G. dos Santos. 2011. Performance of four Tricho- J. Pest Sci. 86: 533Ð541. gramma species (Hymenoptera: Trichogrammatidae) as Chau, A., and M. Mackauer. 2001. Host-instar selection in biocontrol agents of Heliothis virescens (Lepidoptera: the aphid parasitoid Monoctonus paulensis (Hymenop- Noctuidae) under various temperature regimes. J. Pest tera: Braconidae, Aphidiinae): assessing costs and bene- Sci. 84: 313Ð320. Þts. Can. Entomol. 133: 549Ð564. Arno´, J., and R. Gabarra. 2011. Side effects of selected in- secticides on the Tuta absoluta (Lepidoptera: Gelechii- Cocco, A., S. Deliperi, and G. Delrio. 2013. Control of Tuta dae) predators Macrolophus pygmaeus and Nesidiocoris absoluta (Meyrick) (Lepidoptera: Gelechiidae) in green- tenuis (Hemiptera: Miridae). J. Pest Sci. 84: 513Ð520. house tomato crops using the mating disruption tech- nique. J. Appl. Entomol. 137: 16Ð28. Biondi, A., N. Desneux, G. Siscaro, and L. Zappala`. 2012. Using organic-certiÞed rather than synthetic pesticides Consoli, F. L., and J.R.P. Parra. 1996. Biology of Tricho- may not be safer for biological control agents: selectivity gramma galloi and T. pretiosum (Hymenoptera: Tricho- and side effects of 14 pesticides on the predator Orius grammatidae) reared in vitro and in vivo. Ann. Entomol. laevigatus. Chemosphere 87: 803Ð812. Soc. Am. 89: 828Ð834. Biondi A., L. Zappala`, J. D. Stark, and N. Desneux. 2013a. Do Desneux, N., and R. Ramirez–Romero. 2009. Plant charac- biopesticides affect the demographic traits of a parasitoid teristics mediated by growing conditions can impact para- wasp and its biocontrol services through sublethal ef- sitoidÕs ability to attack host aphids in winter canola. J. fects? Plos ONE 8(9): e76548. doi:10.1371/journal. Pest Sci. 82: 335Ð342. pone.0076548. Desneux, N., A. Decourtye, and J. M. Delpuech. 2007. The Biondi, A., A. Chailleux, J. Lambion, P. Han, L. Zappala`, and sublethal effects of pesticides on beneÞcial . N. Desneux. 2013b. Indigenous natural enemies attack- Annu. Rev. Entomol. 52: 81Ð106. ing Tuta absoluta (Lepidoptera: Gelechiidae) in Southern Desneux, N., P. Stary, C. J. Delebecque, T. D. Gariepy, R. J. France. Egypt. J. Biol. Pest Control 23: 117Ð121. Barta, K. A. Hoelmer, and G. E. Heımpel. 2009a. Cryptic Biondi, A., N. Desneux, E. Amiens–Desneux, G. Siscaro, and species of parasitoids attacking the Soybean Aphid L. Zappala`. 2013c. Biology and developmental strategies (Hemiptera: Aphididae) in Asia: Binodoxys communis and of the Palaearctic parasitoid Bracon nigricans (Hymenop- Binodoxys koreanus (Hymenoptera: Braconidae: Aphidii- tera: Braconidae) on the Neotropical moth Tuta absoluta nae). Ann. Entomol. Soc. Am. 102: 925Ð936. (Lepidoptera: Gelechiidae). J. Econ. Entomol. 106: 1638Ð Desneux, N., R. J. Barta, K. A. Hoelmer, K. R. Hopper, and 1647. G. E. Heimpel. 2009b. Multifaceted determinants of Bompard, A., C. Jaworski, P. Bearez, and N. Desneux. 2013. host speciÞcity in an aphid parasitoid. Oecologia 160: Sharing a predator: can an invasive alien pest affect the 387Ð398. predation on a local pest? Soc. Popul. Ecol. 55: 433Ð440. Desneux, N., E. Wajnberg, K.A.G. Wyckhuys, G. Burgio, S. Cabello, T., J. R. Gallego, F. J. Fernandez, M. Gamez, E. Vila, Arpaia, C. A. Narvaez–Vasquez, J. Gonzalez–Cabrera, M. Del Pino, and E. Hernandez. 2012. Biological control D. C. Ruescas, E. Tabone, J. Frandon, et al. 2010. Bio- strategies for the South American tomato moth (Lepi- logical invasion of European tomato crops by Tuta ab- doptera: Gelechiidae) in greenhouse tomatoes. J. Econ. soluta: ecology, geographic expansion and prospects for Entomol. 105: 2085Ð2096. biological control. J. Pest Sci. 83: 197Ð215. Calvo, J., K. Blockmans, P. A. Stansly, and A. Urbaneja. 2009. Desneux, N., M. G. Luna, T. Guillemaud, and A. Urbaneja. Predation by Nesidiocoris tenuis on Bemisia tabaci and 2011a. The invasive South American tomato pinworm, injury to tomato. Biocontrol 54: 237Ð246. Tuta absoluta, continues to spread in Afro-Eurasia and Calvo, F. J., M. J. Lorente, P. A. Stansly, and J. E. Belda. 2012. beyond: the new threat to tomato world production. J. Preplant release of Nesidiocoris tenuis and supplementary Pest Sci. 84: 403Ð408. tactics for control of Tuta absoluta and Bemisa tabaci in Desneux, N., J. Pizzol, C. Thomas, E. Pautrat, P. Bearez, C. greenhouse tomato. Entomol. Exp. Appl. 143: 111Ð119. Poncet, E. Tabone, F. Kabiri, and J. Frandon. 2011b. Cagnotti, C. L., M. M. Viscarret, M. B. Riquelme, E. N. Botto, Potential for direct interference between natural ene- L. Z. Carabajal, D. F. Segura, and S. N. Lo´pez. 2012. mies of Tuta absoluta on tomato. Acta Hortic. 917: 31Ð37. 2320 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 106, no. 6

Desneux, N., R. Blahnik, C. J. Delebecque, and G. E. Heim- Keller, M. A. 1987. Inßuence of leaf surfaces on movements pel. 2012. Host phylogeny and specialisation in parasi- by the hymenopterous parasitoid Trichogramma-ex- toids. Ecol. Lett. 15: 453Ð460. iguum. Entomol. Exp. Appl. 43: 55Ð59. Driessen, G., L. Hemerik, and B. Boonstra. 1991. Host se- Kolliker–Ott, U. M., F. Bigler, and A. A. Hoffmann. 2003. lection behavior of the parasitoid Leptopilina clavipes, in Does mass rearing of Þeld collected Trichogramma bras- relation to survival in hosts. Neth. J. Zool. 41: 99Ð111. sicae wasps inßuence acceptance of European corn borer El-Wakeil, N. E. 2007. Evaluation of efÞciency of Tricho- eggs? Entomol. Exp. Appl. 109: 197Ð203. gramma evanescens reared on different factitious hosts to Lu, Y. H., K. M. Wu, Y. Y. Jiang, Y. Y. Guo, and N. Desneux. control Helicoverpa armigera. J. Pest Sci. 80: 29Ð34. 2012. Widespread adoption of Bt cotton and insecticide Farrar, R. R., J. D. Barbour, and G. G. Kennedy. 1994. Field- decrease promotes biocontrol services. Nature 487: 362Ð evaluation of insect resistance in a wild tomato and its 365. effects on insect parasitoids. Entomol. Exp. Appl. 71: 211Ð Mackauer, M. 1996. Sexual size dimorphism in solitary para- 226. sitoid wasps: inßuence of host quality. Oikos 76: 265Ð272. Fidgen, J. G., E. S. Eveleigh, and D. T. Quiring. 2000. In- Mills, N. 2010. Egg parasitoids in biological control and in- ßuence of host size on oviposition behaviour and Þtness tegrated pest management, pp. 389Ð411. In F. L. Consoli, of Elachertus cacoeciae attacking a low-density population J.R.P. Parra, and R. A. Zucchi (eds.), Egg Parasitoids in of spruce budworm Choristoneura fumiferana larvae. Agroecosystems With Emphasis on Trichogramma. Ecol. Entomol. 25: 156Ð164. Springer, Dordrecht, The Netherlands. Foerster, M. R., and L. A. Foerster. 2009. Effects of tem- Miranda, M.M.M., M. Picanco, J. C. Zanuncio, and R.N.C. perature on the immature development and emergence Guedes. 1998. Ecological life table of Tuta absoluta of Þve species of Trichogramma. Biocontrol 54: 445Ð450. (Meyrick) (Lepidoptera : Gelechiidae). Biocontrol Sci. Gontijo, P. C., M. C. Picanc¸o, E.J.E. Pereira, J. C. Martins, M. Technol. 8: 597Ð606. Chediak, and R.M.C. Guedes. 2013. Spatial and tempo- Nurindah, B. W. Cribb, and G. Gordh. 1999. Inßuence of ral variation in the control failure likelihood of the tomato rearing hosts on host size acceptance by Trichogramma , Tuta absoluta. Ann. Appl. Biol. 162: 50Ð59. australicum. Biocontrol 44:129Ð141. Harvey, J. A., E. H. Poelman, and T. Tanaka. 2013. Intrinsic Olson D. M., and D. A. Andow. 2006. Walking pattern of inter- and intraspeciÞc competition in parasitoid wasps. Trichogramma nubilale Ertle & Davis (Hymenoptera; Annu. Rev. Entomol. 58: 333Ð51. Trichogrammatidae) on various surfaces. Biol. Control Hawkins, B. A., H. V. Cornell, and M. E. Hochberg. 1997. 39. Predators, parasitoids, and pathogens as mortality agents Parra, J.R.P., and R. A. Zucchi. 2004. Trichogramma in Bra- in phytophagous insect populations. Ecology 78: 2145Ð zil: feasibility of use after twenty years of research. 2152. Neotrop. Entomol. 33: 271Ð281. Hoffmann, M. P., P. R. Ode, D. L. Walker, J. Gardner, S. Van Pennacchio, F., and M. R. Strand. 2006. Evolution of devel- Nouhuys, and A. M. Shelton. 2001. Performance of opmental strategies in parasitic hymenoptera. Annu. Rev. Trichogramma ostriniae (Hymenoptera: Trichogramma- Entomol. 51: 233Ð258. tidae) reared on factitious hosts, including the target host, Pintureau, B. 2009. La lutte biologique et les Tricho- Ostrinia nubilalis (Lepidoptera: Crambidae). Biol. Con- grammes: application au controle de la pyrale du maõs, p. trol 21: 1Ð10. ˆ ¨ Hoffmann, M. P., S. A. Pitcher, S. A. Cheever, J. Gardner, J. E. 262. Le Manuscrit, Paris, France. Losey, T. P. Kuhar, C. A. Laub, and R. R. Youngman. Pizzol, J., B. Pintureau, O. Khoualdia, and N. Desneux. 2010. 2006. EfÞcacy of inoculative releases of Trichogramma Temperature-dependent differences in biological traits ostriniae (Hymenoptera: Trichogrammatidae) against between two strains of Trichogramma cacoeciae (Hyme- European corn borer Ostrinia nubilalis (Lepidoptera: noptera: Trichogrammatidae). J. Pest Sci. 83: 447Ð452. Crambidae) in Þeld corn. Biol. Control 36: 345Ð349. Pizzol, J., N. Desneux, E. Wajnberg, and D. Thiery. 2012. Hopper, K. R., and R. T. Roush. 1993. Management of ge- Parasitoid and host egg ages have independent impact on netics of biological-control introductions. Genetics 38: various biological traits in a Trichogramma species. J. Pest 27Ð51. Sci. 85: 489Ð496. Heimpel, G. E., C. Neuhauser, and M. Hoogendoorn. 2003. Pratissoli, D., and J.R.P. Parra. 2000. Fertility life table of Effects of parasitoid fecundity and host resistance on Trichogramma pretiosum (Hym., Trichogrammatidae) in indirect interactions among hosts sharing a parasitoid. eggs of Tuta absoluta and Phthorimaea operculella (Lep., Ecol. Lett. 6: 556Ð566. Gelechiidae) at different temperatures. J. Appl. Entomol. Jaenike, J. 1978. Optimal oviposition behavior in phytopha- 124: 339Ð342. gous . Theor. Popul. Biol. 14: 350Ð356. Pratissoli, D., R. T. Thuler, G. S. Andrade, L.C.M. Zanotti, Janssen, A. 1989. Optimal host selection by Drosophila para- and A. F. da Silva. 2005. Estimate of Trichogramma pre- sitoids in the Þeld. Funct. Ecol. 3: 469Ð479. tiosum to control Tuta absoluta in stalked tomato. Pes- Kauffman, W. C., and G. G. Kennedy. 1989. Relationship quisa Agropecua´ria Brasileira 40: 715Ð718. between trichome density in tomato and parasitism of Roriz, V., L. Oliveira, and P. Garcia. 2006. Host suitability Heliothis spp. (Lepidoptera, Noctuidae) eggs by Tricho- and preference studies of Trichogramma cordubensis gramma spp. (Hymenoptera, Trichogrammatidae). En- (Hymenoptera: Trichogrammatidae). Biol. Control 36: viron. Entomol. 18: 698Ð704. 331Ð336. Kazmer, D. J., and R. F. Luck. 1995. Field-tests of the size- Rutledge C. E., and R. J. O’Neil. 2005. Orius insidiosus (Say) Þtness hypothesis in the egg parasitoid Trichogramma- as a predator of the soybean aphid, Aphis glycines Mat- pretiosum. Ecology 76: 412Ð425. sumura. Biol. Control 33:56Ð64. Khanh, H.D.T., M. Tiradon, E. Colombel, A. Bonnet, A. Salt, G. 1937. The egg parasite of Sialis lutaria: a study of the Chailleux, N. Desneux, and E. Tabone. 2012. Using new inßuence of the host upon a dimorphic parasite. Parasi- egg parasitoids (Trichogramma spp.) to improve inte- tology 29: 539Ð53. grated management against Tuta absoluta. OEPP/EPPO Salt, G. 1940. Experimental studies in insect parasitism. VII. Bull. 42: 249Ð254. The effect of different host on parasite Trichogramma December 2013 CHAILLEUX ET AL.: SUITABILITY OF T. absoluta FOR Trichogramma 2321

evanescens Wests. (Hym.: Chalcidoidea). Proc. R. Ento- van Driesche R. G., and T. S. Bellows. 1996. Biological con- mol. Soc. Lond. A, General Entomol. 15: 81Ð95. trol. Chapman & Hall, New York, NY. Smith, S. M. 1996. Biological control with Trichogramma: Yuan, X. H., L. W. Song, J. J. Zhang, L. S. Zang, L. Zhu, C. C. advances, successes, and potential of their use. Annu. Rev. Ruan, and G. Z. Sun. 2012. Performance of four Chi- Entomol. 41: 375Ð406. nese Trichogramma species as biocontrol agents of the Symondson, W.O.C., K. D. Sunderland, and M. H. Green- rice striped stem borer, Chilo suppressalis, under var- stone. 2002. Can generalist predators be effective bio- ious temperature and humidity regimes. J. Pest Sci. 85: control agents? Annu. Rev. Entomol. 47: 561Ð594. 497Ð504. Tabone, E., C. Bardon, N. Desneux, and E. Wajnberg. 2010. Zappala` L., U. Bernardo, A. Biondi, A. Cocco, S. Deliperi, G. Parasitism of different Trichogramma species and strains Delrio, M. Giorgini, P. Pedata, C. Rapisarda, G. Tropea on Plutella xylostella l. on greenhouse caulißower. J. Pest Garzia, et al. 2012. Recruitment of native parasitoids by Sci. 83: 251Ð256. the exotic pest Tuta absoluta (Meyrick) in Southern Italy. Tabone, E., C. Bardon, and N. Desneux. 2012. Study of dis- persal as a selection criterion for Trichogrammatidae for Bull. Insectol. 65: 51Ð61. biological control in caulißower greenhouses. Acta Hor- Zappala`, L., A. Biondi, A. Alma, I. J. Al-Jboory, J. Arno`,A. tic. 927: 227Ð236. Bayram, A. Chailleux, A. El-Arnaouty, D. Gerling, Y. Thomson, L. J., D. M. Bennett, D. C Glenn, and A. A. Hoff- Guenaoui, L. Shaltiel-Harpaz, G. Siscaro, M. Stavrinides, mann. 2003. Developing Trichogramma as a pest man- L. Tavella, R. Vercher Aznar, A. Urbaneja, N. Desneux. agement tool, pp. 65Ð85. In O. Koul and G. S. Dhaliwal 2013. Natural enemies of the South American moth, Tuta (eds.), Predators and Parasitoids. Taylor and Francis, absoluta, in Europe, North Africa and Middle-East, and London, United Kingdom. their potential use in pest control strategies. J. Pest Sci. (in van Alphen J.J.M., and L.E.M. Vet. 1986. An evolutionary press) (doi:10.1007/s10340-013-0531-9). approach to host Þnding and selection, pp. 23Ð61. In J. K. Waage and D. J Greathead (eds.), Insect Parasitoids. Academic, London, United Kingdom. Received 21 February 2013; accepted 10 September 2013.