Dissertations / Theses on the topic 'Parasitoid'

This thesis examines the relationship between insect life history and behavioural decisions and underlying cellular biochemistry, with particular focus on bethylid parasitoid wasps in the genus Goniozus. This comprises the first major body of work attempting to draw links between the underlying metabolome of an organism and its behaviour. This thesis further optimised the first known example of a combined LC-MS and NMR metabolomic approach capable of analysing extremely low biomass samples (<1 mg), a vital requirement when studying the behaviour of individual organisms. Part 1 of this thesis details the optimisation and validation of this metabolomic approach, whilst also examining the effects of aging on the metabolome of adult Goniozus wasps. Part 2 applies this approach to examine the effects of diet, host species and host aging on Goniozus wasp behaviour and biochemistry. Comparisons of the metabolomes of starved and honey fed wasps indicate that G. legneri is capable of utilising a carbohydrate rich diet as an energy source. Aged honey fed wasps possessed higher levels of large storage lipids, such as tri- and diacylglycerides, than starved wasps of the same age. Metabolomic analysis also detected a legacy effect on the metabolome of G. legneri associated with differences in the species of host each wasp was reared on. A similar legacy effect was confirmed when examining the metabolomes of wasps reared on artificially aged hosts. Whilst Goniozus wasp oviposition behaviour was altered by the species of host presented, no links between changes in a wasp’s metabolome and its resulting contest behaviour were found. Part 3 of this thesis examines the morphological, behavioural and chemical mimicry of another wasp, the hyperparasitoid Gelis agilis. G. agilis demonstrated an enhanced predation avoidance rate compared with control species, similar to that of the black garden ant Lasius niger. Agitation of G. agilis also resulted in the chemical emission of a known ant alarm pheromone.

A capacidade de dispersão e parasitismo de D. longicaudata, proveniente de larvas de Ceratitis capitata (linhagem Ceratitis) e de Anastrepha fraterculus (linhagem Anastrepha) foi avaliada após liberação em pomar de goiaba. Onze liberações de cerca de 3.000 parasitoides foram realizadas entre março de 2014 e abril de 2015 no centro de um talhão de 15 ha de goiaba localizado no Projeto de Irrigação do Jaíba (15°14\'03.7\"S 43°56\'54.0\"W), em Jaíba, MG, Brasil. A dispersão foi monitorada, a partir de um ponto central de liberação, nas direções cardeais e colaterais, a uma distância de 6 m até 272 m, distantes 12 m entre si nas direções cardeais e 17 m nas colaterais. Em cada ponto uma \"unidade de parasitismo\", ou seja, um sache com cerca de 20 larvas de terceiro instar do hospedeiro mais dieta envoltos em tecido voile, foi pendurada. Larvas de C. capitata e de A. fraterculus foram oferecidas para recuperar os parasitoides em cinco liberações onde os parasitoides não tiveram escolha pelo hospedeiro e outras seis onde tiveram escolha. As \"unidades de parasitismo\" permaneceram 24 h no pomar quando foram substituídas por novas, também retiradas após 24h. Após sete e 15 dias novas \"unidades de parasitismo\" foram distribuídas e mantidas por 24 h. Foram verificadas a porcentagem de parasitismo, razão sexual, mortalidade das larvas hospedeiras e a direção de dispersão. Os dados foram analisados utilizando modelos lineares generalizados de efeito misto através do sistema estatístico R. A distância média de dispersão (DM) e a área de dispersão (S²) de D. longicaudata foram determinadas utilizando-se o modelo proposto por Dobzhansky e Wright. O total de 10.351 parasitoides foi recuperado, todos pertencentes à espécie D. longicaudata, 769 provenientes do hospedeiro A. fraterculus e 9.582 de C. capitata. A mortalidade das larvas foi maior para o hospedeiro A. fraterculus. A razão sexual foi preferencialmente (ou tendeu) para machos e diferiu apenas entre os hospedeiros, sendo maior em parasitoides recuperados de A. fraterculus (0,34), independente da linhagem liberada. Nenhuma variável climática analisada influenciou o parasitismo ou a mortalidade das larvas. A linhagem Ceratitis foi a que mais se dispersou e parasitou larvas até a distância máxima de 173 m e atingiu uma área de dispersão de até 34.067,17m². As duas linhagens do parasitoide se dispersaram em todas as direções, mas demonstraram uma tendência a se dispersar para a direção leste. As duas linhagens foram capazes de sobreviver e parasitar larvas de A. fraterculus e C. capitata por até 15 dias após sua liberação, mas, número maior de exemplares da linhagem Anastrepha foi recuperado após 15 dias parasitando larvas dos dois hospedeiros. A linhagem Anastrepha proporcionou os mais altos índices de parasitismo nas maiores distâncias do ponto de liberação e a linhagem Ceratitis em pontos mais próximos. Com base na maior capacidade de sobrevivência em condição de semiárido, maior capacidade de parasitismo em distâncias maiores do ponto de liberação e na área de dispersão média obtida (27.368,24 m²) recomenda-se a liberação de 3.000 parasitoides da linhagem Anastrepha a cada 2,7 ha de goiaba<br>The ability of dispersal and parasitism of Diachasmimorpha longicaudata reared with the larval host of Ceratitis capitata (Ceratitis strain) and Anastrepha fraterculus (Anastrepha strain) was evaluated in a guava orchard. Eleven releases of approximately 3,000 parasitoids were conducted between March 2014 and April 2015 at the center of a 15-ha guava orchard located in the Jaíba Irrigation Project area (15°14?03.7?S, 43°56?54.0?W) in Jaíba, State of Minas Gerais, Brazil. The dispersal of Diachasmimorpha longicaudata was monitored within a distance of 6 m to 272 m from a central point of release to the cardinal and collateral directions; each cardinal and collateral direction point was located at a distance of 12 m and 17 m from each other, respectively. At each point, a \"parasitism unit\", that comprised a sachet made of voile fabric containing 20 third instar larval host more diet was hung. Larvae of C. capitata and A. fraterculus were offered to recover the parasitoids into five releases where the parasitoids had no choice by the host and six where they had choice. The \"parasitism units\" remained for 24 hours in the orchard when they were replaced with new ones, also remained for 24 hours. After 7 and 15 days others \"parasitism units\" were distributed and maintained for 24 h. Observations on parasitism, sex ratio, mortality of host larvae, and the direction of dispersal were made for each time point. Data were analyzed using the general linear models and mixed effect models by the R statistical system. The average distance dispersion (DM) and the dispersion area (S²) of D. longicaudata were determined using the model proposed by Dobzhansky and Wright. A total of 10,351 parasitoids was recovered, all belonging to D. longicaudata, 769 from the host A. fraterculus and 9,582 from C. capitata. Larval mortality was higher in A. fraterculus. The sex ratio showed a tendency for males and differed between the hosts; the number of males was higher in the parasitoids recovered from A. fraterculus (0.34), regardless of the strain of D. longicaudata released. Parasitism or larval mortality was not influenced by any of the climatic variables analyzed. The Ceratitis strain was most dispersed and parasitized larvae up to a maximum distance of 173 m and scattering area of up to 34,067.17 m². Both parasitoid strains scattered in all directions, but showed a tendency to disperse toward the east. The two strains were able to survive and parasitize the larvae of A. fraterculus and C. capitata for up to 15 days after its release, but more number of the Anastrepha strain were recovered after 15 days of parasitizing the larvae of the two hosts. The Anastrepha strain showed the highest parasitism rate over larger distances from the point of release, while the Ceratitis strain parasitized the larval hosts close to the release points. Based on the highest survival in semiarid conditions, the higher parasitism capacity at greater distances from the point of release, and the average obtained dispersion area (27,368.24 m²), we recommend the release of 3,000 parasitoids of Anastrepha strain in every 2.7 ha of guava orchards

Islands constitute natural laboratories for the study of evolutionary and ecological processes due to their discrete and isolated nature. Island biotas tend to be species–poor and disharmonic compared to the mainland; typically, interspecific competition is low, and entire groups of predators, parasitoids or pathogens are absent from their biotas, so the ecological space is often not fully saturated. Consequently, species from island assemblages often use a wider range of resources than their counterparts from the source mainland. Here, I investigate whether island parasitoid communities have proportionally more generalist species than their source mainland, and which factors determine island community structure. These questions were approached using data on the distribution of Ichneumonoidea (Hymenoptera) species worldwide and with data from a survey conducted in the Macaronesian region. Prior to the global analyses, I assessed whether islands and archipelagos follow the same species–area relationship, and identified which islands have comparable inventories. Globally, islands have proportionally more idiobionts (i.e. generalists) than continental areas. However, there is a latitudinal gradient in the level of generalism of island parasitoid faunas that correlates with some environmental factors and island characteristics; the species pool is the most important determinant of island community structure, together with temperature (for braconids) or biogeographical region (for ichneumonids). Host and parasitoid larvae collected in different islands of the Macaronesian region and adjacent mainland were assigned to Molecular Operational Taxonomic Units using a protocol based on host dissection and DNA barcoding. At this scale, mainland faunas have proportionally more koinobiont species and island communities have a greater proportion of idiobionts. Although overall parasitism rates were similar between islands and mainland, islands had higher idiobiont parasitism rates than expected by chance. In summary, results from this thesis indicate that indeed island parasitoid faunas are biased towards generalist species.

Reproductive traits of wasp parasitoids are thought to be strong indicators of their success as biological control agents. Our study looks at the number of eggs produced by the whitefly parasitoid Eretmocerus eremicus. A series of experiments conducted on female wasps reared in the absence of whitefly hosts demonstrated that adult wasps emerge with a large (approximately 54) number of eggs that is retained during the first 2 days of adult life. Eggs are then absorbed steadily until at least 8 days following emergence. The results of this study suggest that the mode of egg production exhibited by E. eremicus is the type where they emerge with all, or nearly all, of their eggs, i.e. they do not produce additional eggs as they age. This information is significant when considering how they find their whitefly hosts and how effective they might be in controlling whitefly numbers.

This thesis presents laboratory investigations on the competitive interactions which take place within and between bethylid parasitoids. Part one investigates the compatibility of three bethylids (Cephalonomia hyalinipennis, Cephalononlia stephanoderis and Prorops nasuta) for biocontrol releases against the principal pest of coffee, the coffee berry borer (CBB), Hypothenemus hampei. Cephalonomia hyalinipennis is able to hyperparasitise and consume pupae of C stephanoderis and P. nasuta. Cephalonomia stephanoderis also engages in intra-guild predation, consuming pupae of C hyalinipennis. In contests for CBB hosts, fatal fighting occurs in 69% of inter-specific replicates but never occurs in intra-specific replicates. This suggests that interspecific competition is stronger than intraspecific competition and that species coexistence may be compromised. Cephalonomia tephanoderis is the superior interspecific contestant while P. nasuta is the least successful and never kills an opponent. Where CBB infested coffee berries are provided to the three bethylids, coexistence between species is possible, but rare, within a single coffee berry. Prorops nasuta is the most successful species in interspecific replicates and replicates containing C. hyalinipennis generally have low production, regardless of the species combination added. Part two investigates contest interactions, the variables that influence contest outcome between Goniozus nephantidis females and chemical release. Prior ownership and difference in contestant weight have positive influences on contest outcome. Host weight positively influences the outcome of contests between two 'owners' and 'intruder' take-over success increases when intruders are older than owners. Seven bethylid species are found to release volatile chemicals when stressed. A pilot study identifies the volatile chemical in G. nephantidis and employs Atmospheric Pressure Chemical Ionisation-Mass Spectrometry for real-time analysis of chemical release during contest interactions. The appendix contains an advanced investigation using this technique. Bethylids are useful model organisms for the study of competitive interactions but appear to be generally ineffective as biological control agents.

This thesis focuses on a question central to the field of evolutionary ecology; how variation is maintained within populations. Focusing on a model system based around the pea aphid (Acyrthosiphon pisum) I reveal substantial levels of genetic variation in every fitness related trait examined. Theory suggests that such variation may commonly be maintained by life-history trade-offs operating between different traits, but experimental examples of these traits are exceedingly rare within the literature. Populations of the pea aphid show high levels of variation in the ability to successfully defend against natural enemy attack, and to date this reason for this prolonged variation has remained unexplained.

Made available in DSpace on 2014-06-11T19:25:18Z (GMT). No. of bitstreams: 0 Previous issue date: 2007-05-25Bitstream added on 2014-06-13T18:53:12Z : No. of bitstreams: 1 peres_fsc_me_jabo.pdf: 331825 bytes, checksum: 18b3c8bc6500ae52044825067eb8d2af (MD5)<br>Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)<br>Foi avaliada a atratividade de Tagetes patula (cravo-de-defunto) sobre tripes (Thysanoptera) e micro-hymenoptera em cultivo protegido de melão. Foram estudadas a abundância, dominância, freqüência e constância de espécies de insetos utilizando-se a análise faunística, análise de agrupamento (AA) e análise de componentes principais (ACP). Nas extremidades das estufas com cultivo de melão foram plantadas duas faixas transversais de cravo-de-defunto. As amostras foram tomadas nas plantas de melão, batendo-se três ponteiros e nas plantas de T. patula 1m2 sobre bandeja branca. Os pontos de amostragem consistiram em: T. patula, melão consorciado com T. patula e melão à distancia de 6m, 12m, 18m e 24m de T. patula. Onze espécies de tripes e 21 espécies de himenópteros parasitóides foram observados. As espécies dominantes de tripes e superdominantes e constantes de himenópteros foram analisadas para avaliar a distribuição na estufa. Os resultados permitiram verificar que houve três grupos diferentes em relação a abundância de espécies de tripes: (1) T. patula, (2) melão consorciado com T. patula e (3) melão a distancia de 6m, 12m, 18m e 24m do cravo-de-defunto. Também foi possível observar que as espécies de tripes foram mais abundantes em T. patula e que a bordadura com essa planta apresenta maior população de himenópteros parasitóides. Com isso, bordaduras de T. patula podem ser utilizadas para implementar o controle biológico bem como para servir de cultura atrativa.<br>The attractiveness of Tagetes patula (marigold) on thrips (Thysanoptera) and parasitic wasps (Hymenoptera) was ingestigated in protected melon crop. Insect abundance, dominance, frequency, and constancy were evaluated using faunistic analysis, cluster and principal component analyses. Transversal strips of T. patula were grow at both ends of the protected melon greenhouse. Samplings were taken by shaking three melon vine tips and all T. patula plants from 1 m2 on white trays. Samplings sites were T. patula, melon along with T. patula and melon plants at 6m, 12m, 18m, and 24m from T. patula. Eleven thrips species and 21 parasitic wasps were observed. The dominant species of thrips as well as superdominant and dominant and constant species of parasitic wasps were analysed to evaluated species distribution on the crop. The results showed that there were three different groups according to thrips species abundance: (1) T. patula, (2) melon along with T. patula, and (3) melon alone at 6m, 12m, 18m, and 24m from T. patula. It was also possible to note that thrips species were more attracted to T. patula, and that the border presented a higler population of parasitic wasps. Thus, T. patula border can be used to improve biological control as well as serve as trap crop.

In this paper, 48 ichneumonid species of 14 genera, two tribes of the subfamily Pimplinae were listed on the basis of material collected during the time from 2003 to 2011 in four provinces from the Central Highlands of Vietnam and a lesser degree of previous records. Among 48 aforemention pimpline no less than 16 species were described recently as new for science. The diversity of topography and vegetation types resulted in the high level of species richness and new discoveries of the region<br>Bài báo này giới thiệu danh sách 48 loài ong cự thuộc 14 giống, 2 tộc của phân họ Pimplinae dựa trên việc phân tích các mẫu vật thu tại Tây Nguyên trong thời gian từ năm 2003–2011 và một số ít loài dựa trên các ghi nhận trước đây. Trong số đó, 16 loài mới được mô tả trong thời gian 5 năm trở lại đây. Sự đa dạng về địa hình và các kiểu thảm thực vật khác nhau đã tạo nên tính đa dạng thành phần loài và các phát hiện mới của ong cự ở khu vực Tây Nguyên đầy tiềm năng này

Reproductive success is heavily influenced by life-history traits; a series of energy investment trade-offs that organisms must optimise according to their environmental conditions. These include considerations such as how many offspring and when to reproduce? The consequences of multiple trade-offs can be extremely complex, making research difficult. However, there are notable exceptions. Simple clutch size theory enabled great strides in assessing trade-offs in resource allocation, though it quickly becomes more complicated when considering investment in current versus future reproduction. Arguably, even greater success has come from consideration of investment in a particular sex. Sex allocation theory provides simple models that can be empirically tested, and has provided some of the strongest evidence for natural selection and evolution. Much of this work has focused on certain parasitoids due to their extraordinary sex ratios and the finite resources available to offspring in a host. Whilst clutch size and sex allocation theory have provided many answers, there are still questions regarding the impact of other life-history traits. In this thesis I have used the gregarious parasitoid wasp Nasonia vitripennis in laboratory experiments to assess some of these traits. I have focused on the impact of larval competition, inbreeding, host condition and host feeding on longevity, fecundity, sex allocation and mating success. By manipulating host quality through host-feeding, I was able to vary the level of resources available to offspring. Simultaneously, by manipulating the matedstatus and number of females ovipositing on a host, I was able to vary the number and sex ratio of offspring competing for resources. My research has provided an insight into how larval competition and host-feeding impact on optimal clutch size and sex allocation. Furthermore, I have attempted to assess the extent to which body size, which is commonly associated with reproductive success, can be used to predict fitness. The appendix includes work using molecular data to understand the mating behaviour and population structure of N. vitripennis in the wild, enabling models based on assumptions of laboratory-based behaviour to be applied to wild populations.

The ecology and evolution of a species is often considered only within the context of pairwise interactions even though a species' distribution and abundance may be determined by interactions with many species within and between trophic levels. Multiple herbivores often share the same host and may interact indirectly by altering the relationships between herbivores, their host plants and their parasitoids. However, the relationships between parasitoids and herbivore hosts have typically been studied in isolation of other herbivore species. I examined how the outcomes of species interactions change when multiple relationships are considered. Chapter 1 examined the potentially conflicting selection pressures Manduca sexta exerts on Nicotiana tabacum (tobacco), since M. sexta has pollinating adults but herbivorous larvae. I demonstrated that high nectar amino acids do not affect floral visitation, but increased oviposition of herbivores on leaves. Thus, the relative costs and benefits of nectar rewards may depend on the community of pollinators and their life histories. In the remaining chapters I examined how feeding on tobacco by the aphid Myzus persicae altered the interactions between a parasitoid (Cotesia congregata) and its hornworm host (M. sexta). Chapter 2 demonstrated that aphids reduced hornworm abundance and parasitism. Changes in hornworm abundance were not due to density-dependent changes in moth oviposition, but the proportion of caterpillars attacked by parasitoids was inversely density dependent with hornworm density. Chapter 3 examined whether changes in hornworm abundance and parasitism reflected aphid-induced changes in host plant quality or volatile emissions. Aphids increased hornworm mortality, did not affect parasitoid performance, and increased parasitoid search time. In combination with Chapter 2, results suggest that aphids can mediate parasitoid-caterpillar interactions through changes in host plants that reduce hornworm survival and alter parasitoid behavior. Chapter 4 addressed how the outcome of interactions that are altered by aphids changed with spatial scale, and found no effect spatial scale on hornworm abundance and parasitism. In this system, aphids alter parasitoid-caterpillar interactions through changes in plant quality that reduce hornworm performance and abundance, and in turn, influence parasitoid attack. This work demonstrates that the outcome of multispecies interactions may not be predictable from pairwise interactions.<br>Ph. D.

Submitted by (edna.saturno@ufrpe.br) on 2016-11-28T12:51:34Z No. of bitstreams: 1 Marciene Dantas Moreira.pdf: 367488 bytes, checksum: 1c62f7268fdd87f4772b0530a6eaab2b (MD5)<br>Made available in DSpace on 2016-11-28T12:51:34Z (GMT). No. of bitstreams: 1 Marciene Dantas Moreira.pdf: 367488 bytes, checksum: 1c62f7268fdd87f4772b0530a6eaab2b (MD5) Previous issue date: 2007-07-02<br>Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES<br>During the parasitism process many factors can lead to host rejection by the parasitoid. Thus, it was evaluated bio-ecological aspects of Trichogramma pretiosum Riley (Hym.: Trichogrammatidae) parasiting Sitotroga cerealella Oliver (Lep.: Gelechiidae) eggs. The study was conducted in free-choice test, alternating parasitized and unparasitized eggs, and no-choice test, using only parasitized eggs. For the parasitism study, mated females of T. pretiosum 12 to 24h-old were used. And, parasitized eggs of S. cerealella 24, 72 and 120h-old were used as host. To evaluate the biological characteristics of T. pretiosum superparasitism, S. cerealella eggs were exposed to the parasitism by one or more parasitoid females aiming oviposition of one or two eggs of the parasitoid per host. The parasitism rejection on 72 e 120h previously parasitized S. cerealella eggs was approximately 100% in relation to the 24h-parasitized eggs (ca. 71.3%). On the other hand, the contact time of the female on 24h-parasitized eggs was around twice longer than on 72 and 120h-parasitized eggs. The offspring produced from single egg laid per host werelarger, exhibited no deformation and greater capacity of parasitism, and in contrast of those produced from two eggs laid per host. We concluded that T. pretiosum exhibited ability inrecognizing eggs previously parasitized and that the superparasitism resulted in reduction of reproductive success of the parasitoid.<br>Durante o processo de parasitismo diversos fatores podem levar à rejeição do hospedeiro pelo parasitóide. Dessa forma, avaliou-se aspectos comportamentais e biológicos de Trichogramma pretiosum Riley (Hym.: Trichogrammatidae) parasitando ovos de Sitotroga cerealella Oliver (Lep.: Gelechiidae). O estudo foi conduzido com chance de escolha, alternando-se ovos parasitados e não parasitados, e sem chance de escolha, utilizando-se apenas ovos parasitados. Para o parasitismo, foram utilizadas fêmeas acasaladas de T. pretiosum de 12 a 24h de idade. Foram utilizados ovos de S. cerealella com 24, 72 e 120h de parasitados. Para a avaliação das características biológicas em situação de parasitismo e superparasitismo, ovos de S. cerealella foram submetidos a uma ou mais fêmeas de T. pretiosum, possibilitando a oviposição de 1 a 2 ovos do parasitóide por hospedeiro. A porcentagem de rejeição de ovos de S. cerealella com 72 e 120h de parasitados foi de aproximadamente 100% em relação aos ovos com apenas 24h de parasitados (71,3%). Em contrapartida, o tempo de contato da fêmea sobre ovos de 24h deparasitados foi, aproximadamente, duas vezes maior que em ovos com 72 e 120h de parasitados. Os descendentes originados de um único ovo do parasitóide/hospedeiro apresentaram-se todos sem deformações, maiores e com maior capacidade de parasitismo, ao contrário dos descendentes emergidos de dois ovos do parasitóide/hospedeiro. Dessa forma, conclui-se que T. pretiosum apresentou habilidade em reconhecer ovos previamente parasitados e que o superparasitismo resultou em redução do sucesso reprodutivo do parasitóide.

Dans un contexte de mondialisation et de changement climatique, de plus en plus d'insectes deviennent invasifs et affectent notamment l'agriculture à l'échelle mondiale. Au cœur de ce réseau d'invasions, les Etats-Unis subissent l'arrivée de nombreuses espèces provenant principalement d'Asie. Parmi elles, les punaises pentatomidés menacent la plupart des agrosystèmes américains et sont aujourd'hui principalement régulées par les pesticides de synthèse. Bagrada hilaris (bagrada), originaire d'Asie et d'Afrique, est la plus récente à avoir colonisé les Etats-Unis où elle sévit depuis 2008 dans les cultures de chou et de brocoli (Brassicaceae). Pour répondre aux besoins des agriculteurs, l'USDA-ARS a financé un programme de lutte biologique classique visant à sélectionner et introduire en Californie un ou plusieurs ennemis naturels de bagrada depuis son aire d'origine. Les travaux inclus dans cette thèse sont une partie de ce programme. Ils ont cherché à évaluer en serre de quarantaine si le parasitoïde oophage Gryon gonikopalense (Scelionidae), originaire du Pakistan, pourrait s'établir en Californie et contrôler efficacement bagrada. Quatre principaux objectifs ont ainsi été définis : 1) décrire la biologie générale de G. gonikopalense ; 2) décrire son comportement de recherche d'hôte ; 3) examiner la possibilité de le produire en masse et 4) évaluer sa spécificité vis-à-vis de bagrada avec des punaises euro-méditerranéennes. L'objectif 1 a permis de montrer que la physiologie de G. gonikopalense était similaire à celle de nombreux autres scelionidés : en fonction de la température, il complète son développement en 7 à 25 jours, peut vivre entre 30 et 150 jours et parasiter une centaine d'œufs au cours de sa vie. Il partage avec bagrada un optimum thermique entre 25 et 35 °C qui pourrait lui permettre de s'établir en Californie. En plus d'être un parasite efficace de bagrada, l'atout de G. gonikopalense réside particulièrement dans son comportement de recherche des œufs de son hôte qui sont enterrés dans le sol dans 90% des cas. Nous avons découvert que G. gonikopalense était capable de creuser le sol pour parasiter les œufs. En outre, nous avons montré dans un système tri-trophique que le parasitoïde attaquait principalement les pontes enterrées de bagrada, tout en étant capable d'atteindre les œufs déposés sur la plante hôte (env. 10%). Pour le 3ème objectif, nous avons montré que le stockage des œufs de bagrada à 5°C sur trois semaines permettait d'optimiser l'élevage du parasitoïde et faciliter une production de masse. Concernant l'objectif 4, toujours en cours, il en résulte que les sur 11 pentatomidés testées, au moins 4 ont été parasitées, dont deux essentiellement localisées autour du bassin méditerranéen. Ces résultats complémentent les tests menés en Californie et permettent de discuter proactivement de l'introduction de G. gonikopalense dans les populations méditerranéennes invasives de bagrada. Finalement, nous avons décrit au cours de cette thèse plusieurs aspects, jusqu'alors inconnus, de la biologie de G. gonikopalense. Sa spécialisation comportementale envers bagrada en fait notamment un candidat de premier choix pour le programme de lutte biologique Une pétition rédigée sur la base de ces connaissances remise aux autorités sanitaires des Etats-Unis restent un objectif à court terme. Ceux-ci évalueront alors les risques et les bénéfices liés à l'utilisation de G. gonikopalense pour contrôler bagrada en Californie<br>In a context of globalization and climate change, more and more insects are becoming invasive, particularly affecting agriculture on a global scale. At the heart of this network of invasions, the United States of America is experiencing the arrival of many species, mainly from Asia. Among them, stink bugs (Pentatomidae) threaten most American agrosystems and are now mainly regulated by synthetic pesticides. Bagrada hilaris (bagrada), native to Asia and Africa, is the most recent species to invade the USA, where it has been reported since 2008 in Brassicaceae crops. To answer the needs of farmers, USDA-ARS funded a classical biological control program to select and introduce from its native range one or more natural enemies of bagrada into California. The research included in this PhD thesis is part of this program. It sought to assess in a quarantine greenhouse whether the oophagous parasitoid Gryon gonikopalense (Scelionidae), native to Pakistan, could establish in California and efficiently control bagrada. Thus, four main objectives were defined: 1) to describe the general biology of G. gonikopalense; 2) describe its host foraging behavior; 3) examine the possibility of its mass production and 4) assess its specificity towards bagrada using Euro-Mediterranean non-target stink bugs. Objective 1 allowed to show that the physiology of G. gonikopalense was similar to that of many other scelionids: depending on the temperature, it completed its development in 7 to 25 days, its lifespan ranges from 30 to 150 days and it is able to parasitize hundred eggs during its lifetime. It shares with bagrada a thermal optimum between 25 and 35 °C which could allow it to settle in California. In addition to being an effective parasitoid of bagrada, the advantage of G. gonikopalense particularly involves its foraging behavior for host's eggs, which are buried in the ground in 90% of cases. We found that G. gonikopalense was able to dig into the soil to parasitize eggs. In addition, we have shown in a tri-trophic system that the parasitoid mainly attacks the buried eggs of bagrada, while being able to reach the eggs deposited on the host plant (about 10%). For the 3rd objective, we have shown that the storage of bagrada eggs at 5 °C over three weeks made it possible to optimize the breeding of the parasitoid and facilitate mass production. Concerning the objective 4, still in progress, we showed that among the 11 pentatomids tested, at least four were suitable for the parasitoid, and two of them were mainly located around the Mediterranean basin. These results complement the tests carried out in California and open a proactive discussion on the introduction of G. gonikopalense in invasive Mediterranean populations of bagrada. Finally, during this thesis, we have described several aspects, hitherto unknown, of the biology of G. gonikopalense. Its behavioral specialization towards bagrada makes this parasitoid particularly promising for the biological control program. A petition-to-release drafted on the basis of this knowledge submitted to the health authorities of the USA remains a short-term objective. They will then assess the risks and benefits of using G. gonikopalense to control bagrada in California

The parasitoid Anagyrus kamali Moursi [Hymenoptera: Encyrtidae] has been recently introduced into the Caribbean as a biological control agent against the Hibiscus Mealybug (HMB), Maconellicoccus hirsutus Green [Homoptera: Pseudococcidae]. Information on this parasitoid is scarce and investigations of its biology and behaviour were conducted to optimize its mass-production, and improve our understanding of the A. kamali/M. hirsutus system. Host selection experiments showed that, of nine species tested, HMB was the only suitable mealybug species for A. kamali development. Third nymphal instar was the optimal stage for wasp mass-production since A. kamali progeny from this stage had a female biased sex ratio and it had a weaker immune response as evidenced by lower encapsulation rate than adult female HMB. Parasitoid functional response was of type III in variable-time conditions, ensuring the preservation of HMB population at low density, and of type II--III in fixed-time conditions, leading to a better use of HMB population in mass-rearing. Temperature and photoperiod had a profound impact on parasitoid fecundity, 26°C and LD 0:24 being the optimal condition to maximize A. kamali oviposition. Parasitoid body size was positively correlated to its fitness. Large parasitoids lived significantly longer, had a higher fecundity, and progeny emergence with a lower sex ratio than small ones. Studies on mating showed that A. kamali is arrhenotokous and that multiple mating is required to optimize progeny sex ratio. Progeny production by mated females was also higher than that of virgin females. Female parasitoid fecundity was negatively correlated to their density due to mutual interference. Under mass-rearing conditions, 75 female A. kamali per cage was the optimum. Finally, under stored condition, food supply and low temperature (20°C) appeared essential to sustain parasitoid longevity. However, storage periods up to 14 days did not affect parasitoid lifetime fec

Resums pendents<br>Les comunitats ecològiques són entitats dinàmiques l’estructura i composició de les quals està conformada per processos ecològics i per variabilitat ambiental. Degut a que les espècies interaccionen entre elles, els canvis en l’estructura i la composició de les comunitats poden alterar les xarxes d’interacció i el funcionament dels ecosistemes. En aquesta tesi estudio com un procés ecològic (la colonització d’un hàbitat per una nova espècie) i un component important de la variabilitat espacial (la grandària de l’hàbitat) afecta a les comunitats de pol·linitzadors, d’abelles i vespes, així com a les seves interaccions mutualistes i antagonistes. També investigo si aquests canvis afecten l’estructura de la xarxa d’interaccions i les conseqüències potencials sobre la funció ecosistèmica (pol·linització, reproducció de les plantes, i parasitisme).Els canvis en la composició de les comunitats de plantes són cada vegada més freqüents en el context actual de les invasions biològiques. Tot i així, les invasions de plantes normalment representen escenaris extrems de dominància ecològica i els seus efectes han sigut estudiats relativament bé. Contràriament, els efectes de la colonització dels hàbitats per espècies autòctones (per exemple, com a resultat dels canvis en la distribució geogràfica mediats pel canvi climàtic) han rebut molta menys atenció.Al Capítol 1 analitzem els efectes de l’arribada d’una nova espècie de planta autòctona a un matollar, incloent els canvis en la composició de la comunitat de pol·linitzadors, els patrons d’interacció i l’estructura de la xarxa planta-pol·linitzador, y les conseqüències per a la pol·linització i per a l’èxit reproductiu de les espècies de plantes residents. Per a això vam manipular experimentalment taques de matollar introduint una planta autòctona (planta colonitzadora d’ara endavant), ajustant el número de flors introduïdes a la quantitat de recursos florals localment disponibles. La planta colonitzadora va atraure borinots, facilitant les interaccions a les plantes residents. Les comunitats colonitzades van exhibir una generalització de la xarxa més gran, un aniuament més gran, i modificacions en la composició dels mòduls de la xarxa. Tots aquests canvis van resultar en una deposició de pol·len heteroespecífic (però no conespecífic) més gran, i va tenir efectes contrastats en l’èxit reproductiu de les plantes residents (neutral, positiu, o negatiu depenent de l’espècie). El nostre estudi mostra que les interaccions planta-pol·linitzador es reorganitzen ràpidament com a resposta a noves situacions (inclús quan la nova planta no és excessivament dominant), amb conseqüències funcionals importants en la pol·linització i en la reproducció de les plantes. La relació espècies-àrea (increment en la riquesa d’espècies amb l’àrea) és una de les lleis de l’Ecologia més antigues i fonamentals i, recentment, ha sigut estesa a la riquesa d’interaccions (relació enllaços-àrea) i a l’estructura de la xarxa (relació xarxa-àrea). Tot i així, tot i que les relacions espècies-àrea i enllaços-àrea han rebut un ampli suport d’estudis empírics, la relació entre l’estructura de la xarxa i l’àrea roman en gran part inexplorada. Entendre aquesta relació és important perquè la fragmentació antropogènica dels hàbitats està tenint impactes profunds en la configuració dels hàbitats i està reduint la grandària dels mateixos.Al Capítol 2 testem vàries hipòtesis relacionades amb les relacions espècies-àrea, enllaços-àrea, i xarxa-àrea utilitzant comunitats d’abelles i vespes nidificants en cavitats preestablertes així com la fauna associada als seus nius (hostes i parasitoides d’ara endavant, respectivament) com a sistema d’estudi. Per a obtenir nius d’abelles i vespes vam instal·lar estacions de niadors en taques d’hàbitat obert (favorable) –envoltats de bosc dens (hàbitat desfavorable)- de tres categories de grandària (petites, ~0.1 ha; mitjanes, ~5ha; i grans, hàbitat obert continu). Els nius obtinguts en cada taca van ser disseccionats i els seus contingut van ser analitzats anotant la identitat i l’abundància de cada espècie d’hoste i parasitoid. La riquesa d’hostes, parasitoides, i interaccions va incrementar de les taques petites a les mitjanes, però no de les taques mitjanes a les grans. Els enllaços per espècie van incrementar amb la grandària de la taca, però la taxa de parasitisme no va seguir cap tendència relacionada amb la grandària de la taca. L’especialització de la xarxa va disminuir de les taques petites a les mitjanes, però les taques grans van mostrar una especialització similar a la de les taques petites i mitjanes. La modularitat va disminuir de les taques petites a les mitjanes, però no va haver-hi diferències entre les taques mitjanes i les grans. El número de mòduls va incrementar amb la grandària de la taca. Els nostres resultats estan, majoritàriament, parcialment d’acord amb les relacions espècies-àrea i enllaços-àrea. L’absència de diferències entre les taques mitjanes i grans indica l’existència d’un llindar per damunt del qual la riquesa d’espècies i d’interaccions ja no està limitada per la grandària de l’hàbitat, amb implicacions importants per a la conservació. D’altra banda, els nostres resultats estan majoritàriament en desacord amb les prediccions de la relació xarxa-àrea. Suggerim que l’empobriment de les comunitats a les taques petites resulta en una limitació en l’elecció dels hostes per part dels parasitoides generalistes, donant lloc a nivells alts d’especialització ecològica i de modularitat.S’espera que els canvis en la composició de plantes afectin a les comunitats planta-pol·linitzador de múltiples maneres depenent del context floral, de l’abundància relativa de la planta colonitzadora, del seu grau de generalització, i del seu síndrome floral. La facilitació o la competència pels pol·linitzadors poden resultar en canvis predictibles en l’estructura de la xarxa i en la pol·linització, però s’espera que els efectes sobre la reproducció de les plantes siguin altament contingents a cada espècie. S’espera que la grandària de l’hàbitat afecti a les espècies i a les interaccions seguint les relacions espècies-àrea i enllaços-área. Unint les evidències empíriques amb les prediccions teòriques, desenvolupem un marc de referència per a la relació xarxa-àrea. Les respostes de les xarxes mutualistes i antagonistes als canvis són predictibles. Tot i així, les conseqüències sobre la funció ecosistèmica segueixen sent altament contingents.<br>Las comunidades ecológicas son entidades dinámicas cuya estructura y composición está conformada por procesos ecológicos y por variabilidad ambiental. Debido a que las especies interaccionan entre ellas, los cambios en la estructura y composición de las comunidades pueden alterar las redes de interacción y el funcionamiento de los ecosistemas. En esta tesis estudio cómo un proceso ecológico (la colonización de un hábitat por una nueva especie) y un componente importante de la variabilidad espacial (el tamaño del hábitat) afecta a las comunidades de polinizadores, de abejas y avispas, así como a sus interacciones mutualistas y antagonistas. También investigo si estos cambios afectan a la estructura de la red de interacciones y las consecuencias potenciales sobre la función ecosistémica (polinización, reproducción de las plantas, y parasitismo).Los cambios en la composición de las comunidades de plantas son cada vez más frecuentes en el contexto actual de las invasiones biológicas. Sin embargo, las invasiones de plantas normalmente representan escenarios extremos de dominancia ecológica y sus efectos han sido estudiados relativamente bien. Por el contrario, los efectos de la colonización de los hábitats por especies autóctonas (por ejemplo, como resultado de los cambios en la distribución geográfica mediados por el cambio climático) han recibido mucha menos atención.En el Capítulo 1 analizamos los efectos de la llegada de una nueva especie de planta autóctona en un matorral, incluyendo los cambios en la composición de la comunidad de polinizadores, los patrones de interacción y la estructura de la red planta-polinizador, y las consecuencias para la polinización y para el éxito reproductivo de las especies de plantas residentes. Para ello manipulamos experimentalmente parches de matorral introduciendo una planta autóctona (planta colonizadora de ahora en adelante), ajustando el número de flores introducidas a la cantidad de recursos florales localmente disponibles. La planta colonizadora atrajo a abejorros, facilitando las interacciones a las plantas residentes. Las comunidades colonizadas exhibieron una mayor generalización de la red, un mayor anidamiento, y modificaciones en la composición de los módulos de la red. Todos estos cambios resultaron en una mayor deposición de polen heteroespecífico (pero no conespecífico) y tuvo efectos contrastados en el éxito reproductivo de las plantas residentes (neutral, positivo, o negativo dependiendo de la especie). Nuestro estudio muestra que las interacciones planta-polinizador se reorganizan rápidamente en respuesta a nuevas situaciones (incluso cuando la nueva planta no es excesivamente dominante), con consecuencias funcionales importantes en la polinización y en la reproducción de las plantas.La relación especies-área (incremento de la riqueza de especies con el área) es una de las leyes de la Ecología más antiguas y fundamentales y, recientemente, ha sido extendida a la riqueza de interacciones (relación enlaces-área) y a la estructura de la red (relación red-área). Sin embargo, a pesar que las relaciones especies- y enlaces- área han recibido un soporte amplio de estudios empíricos, la relación entre la estructura de la red y el área permanece en gran parte inexplorada. Entender esta relación es importante porque la fragmentación antropogénica de los hábitats está teniendo impactos profundos en la configuración de los hábitats y está reduciendo el tamaño de los mismos.En el Capítulo 2 testamos varias hipótesis relacionadas con las relaciones especies-área, enlaces-área y red-área utilizando comunidades de abejas y avispas nidificantes en cavidades preestablecidas así como la fauna asociada a sus nidos (huéspedes y parasitoides de ahora en adelante, respectivamente) como sistema de estudio. Para obtener nidos de abejas y avispas instalamos estaciones de nidales en parches de hábitat abierto (favorable) –rodeados por bosque denso (hábitat desfavorable)- de tres categorías de tamaño (pequeños, ~0.1 ha; medianos, ~5 ha; y grandes, hábitat abierto continuo). Los nidos obtenidos en cada parche fueron diseccionados y sus contenidos fueron analizados anotando la identidad y la abundancia de cada especie de huésped y parasitoide. La riqueza de huéspedes, parasitoides, e interacciones incrementó de los parches pequeños a los medianos, pero no de los parches medianos a los grandes. Los enlaces por especie incrementaron con el tamaño del parche, pero la tasa de parasitismo no siguió ninguna tendencia relacionada con el tamaño del hábitat. La especialización de la red disminuyó de los parches pequeños a los medianos, pero los parches grandes mostraron una especialización similar a la de los parches pequeños y medianos. La modularidad disminuyó de los parches pequeños a los medianos, pero no hubo diferencias entre los parches medianos y los grandes. El número de módulos incrementó con el tamaño del parche. Nuestros resultados están, en su mayor parte, parcialmente de acuerdo con las relaciones especies-área y enlaces-área. La ausencia de diferencias entre los parches medianos y grandes indica la existencia de un umbral por encima del cual la riqueza de especies y de interacciones ya no está limitada por el tamaño del hábitat, con implicaciones importantes para la conservación. Por otro lado, nuestros resultados están mayoritariamente en desacuerdo con las predicciones de la relación red-área. Sugerimos que el empobrecimiento de las comunidades en los parches pequeños resulta en una limitación en la elección de los huéspedes por parte de los parasitoides generalistas, dando lugar a altos niveles de especialización ecológica y de modularidad.Se espera que los cambios en la composición de plantas afecten a las comunidades planta-polinizador de múltiples maneras dependiendo del contexto floral, de la abundancia relativa de la planta colonizadora, de su grado de generalización, y de su síndrome floral. La facilitación o la competencia por los polinizadores pueden resultar en cambios predecibles en la estructura de la red y en la polinización, pero se espera que los efectos sobre la reproducción de las plantas sean altamente contingentes en cada especie. Se espera que el tamaño del hábitat afecte a las especies y a las interacciones siguiendo las relaciones especies-área y enlaces-área. Uniendo las evidencias empíricas con las predicciones teóricas, desarrollamos un marco de referencia para la relación red-área. Las respuestas de las redes mutualistas y antagonistas a los cambios son predecibles. Sin embargo, las consecuencias sobre la función ecosistémica siguen siendo altamente contingentes.<br>Ecological communities are dynamic entities whose structure and composition are shaped by ecological processes and environmental variability. Because species interact with one another, changes in community structure and composition are likely to alter interaction networks and ecosystem functioning. In this thesis I study how an ecological process (the colonization of a habitat by a new species) and an important component of spatial variability (habitat area) affect pollinator and bee-wasp communities and their mutualistic and antagonistic interactions. I also investigate whether these changes affect interaction network structure and the potential consequences on ecosystem function (pollination, plant reproduction, and parasitism). Changes in plant community composition are becoming increasingly frequent in the current context of pervasive biological invasions. However, plant invasions usually represent extreme scenarios of ecological dominance and their effects have been relatively well studied. By contrast, the effects of habitat colonization by native species (for example, as a result of changes in geographical distribution mediated by climate change) have received much less attention.In Chapter 1 we analyze the effects of the arrival of a new native plant into a scrubland community, including changes in pollinator community composition, plant–pollinator interaction patterns and network structure, and the consequences for pollination and reproductive success of resident plant species. To do so, we experimentally manipulated patches of scrubland by introducing a native plant (henceforth colonizing plant), adjusting the number of flowers introduced to the amount of floral resources locally available. The colonizing plant attracted bumblebees, facilitating interactions to resident plants. Colonized communities exhibited higher network generalization, nestedness, and modifications in the composition of network modules. All these changes resulted in higher heterospecific (but not conspecific) pollen deposition and had contrasting effects on the reproductive success of resident plants (neutral, positive, or negative depending on the species). Our study shows that plant-pollinator interactions are rapidly rearranged in response to novel situations (even when the new plant is not overly dominant), with important functional consequences on pollination and plant reproduction.The species-area relationship (increasing species richness with area) is one of the oldest and most fundamental laws in Ecology and, in recent years, has extended to interaction richness (link-area relationship) and network structure (network-area relationship). However, although the species- and link-area relationships have received ample support from empirical studies, the relationship between network structure and area remains largely unexplored. Understanding this relationship is important because anthropogenic habitat fragmentation is having a profound impact on habitat configuration and effectively reducing the size of natural habitats.In Chapter 2 we test several hypothesis related to the species-area, link-area and network-area relationships using cavity-nesting bee-wasp communities and their nest associates (henceforth hosts and parasitoids, respectively), as a study system. To obtain bee-wasp nests we set up nest-trapping stations in patches of (favorable) open habitat -surrounded by dense forest (unfavorable habitat)- of three size categories (small, ~0.1 ha; medium, ~5 ha; and large, continuous open habitat). The nests obtained in each patch were dissected and their contents analyzed by noting the identity and abundance of each host and parasitoid species. Host, parasitoid, and interaction richness increase from small to medium patches, but not from medium to large patches. Links per species increased with patch size, but parasitism rate did not follow any patch-size-mediated trend. Network specialization decreased from small to medium patches, but large patches showed similar specialization to that of small and medium patches. Modularity decreased from small to medium patches but there were no differences between medium and large patches. The number of modules increased with patch size. Our results are mostly in partial agreement with the species-area and the link-area relationships. The lack of differences between medium and large patches indicates the existence of a threshold above which local species/link richness are no longer limited by habitat size, with important implications for conservation. On the other hand, our results are mostly in disagreement with the predictions of the network-area relationship. We suggest that community impoverishment in small patches results in limited host choice by generalist parasitoids, leading to high levels of ecological specialization and modularity. Changes in plant composition are expected to affect plant-pollinator communities in multiple ways depending on the flower context, the relative abundance of the colonizing plant, its degree of generalization, and its pollination syndrome. Facilitation or competition for pollinators may result in predictable responses in network structure and pollination, but the effects on plant reproduction are expected to be highly contingent on each species. Habitat size is expected to affect species and interactions following the species-area and the link-area relationships. Merging empirical evidence with theoretical predictions, we developed a framework for the network-area relationship. Mutualistic and antagonistic network responses to changes are predictable. However, consequences on ecosystem function remain highly contingent.

Uno de los principales objetivos de la ecología es comprender cómo la biodiversidad está estructurada espacial y temporalmente, y cuáles son los mecanismos subyacentes a los gradientes de biodiversidad en diferentes escalas espaciales y temporales. En esta tesis, analizo la variabilidad espacio-temporal de comunidades de abejas/avispas (huéspedes) y de sus parasitoides, y de las redes de interacción huésped-parasitoide que se establecen entre ellas. Las especies de abejas y avispas muestran notables diferencias temporales en su fenología, y, por otro lado, las especies de abejas muestran diferentes capacidades termorreguladoras en relación con su tamaño corporal (cuanto más grandes es una, mayor es su capacidad termoreguladora). Por tanto, se podría hipotetizar una relación entre el tamaño corporal (~’grado de endotermia’) y la temperatura ambiente durante el período de vuelo del adulto. Las comunidades de abejas y avispas también muestran una considerable heterogeneidad espacial en respuesta a sus recursos alimentarios y de nidificación. Estos cambios espacio-temporales en las comunidades de abejas/avispas podrían conllevar cambios en sus ‘rasgos funcionales’, y podrían tener un impacto en sus comunidades de parasitoides y, en consecuencia, esto podría reflejarse en cambios en la estructura de sus redes de interacción y en las funciones ecosistémicas asociadas. En el capítulo 1 se analizó la relación entre el tamaño corporal y la temperatura a lo largo de un gradiente de temperatura ambiental intra-anual, utilizando una fauna regional de abejas mediterráneas. Esperábamos encontrar especies más grandes (más endotérmicas) en las estaciones más frías, y especies progresivamente más pequeñas hacia estaciones más cálidas. Esto se puede considerar un test a la ‘norma de Bergmann’ a lo largo de un gradiente de temperatura temporal (en lugar de su formulación clásica a lo largo de gradientes geográficos). Encontramos una relación diferente entre el tamaño corporal y la temperatura ambiente de las especies para las abejas grandes ('endotérmicas') y para las pequeñas (ectotérmicas): las especies mayores que 27,81 mg (peso seco) siguieron la norma de Bergmann, mientras que las especies por debajo de este umbral no mostraban ningún patrón. Nuestros resultados extienden la norma de Bergmann a un gradiente temporal y son coherentes con el mecanismo fisiológico propuesto originalmente por el propio Bergmann ("hipótesis termorreguladora"). Para estudiar las redes de interacción huésped-parasitode se utilizaron comunidades de abejas y avispas nidificantes en cavidades preestablecidas (AANCP), que actúan como 'huéspedes', y sus comunidades de parasitoides, en una zona templada (Capítulos 2 y 3). En el capítulo 2 se estudiaron los efectos de la estacionalidad (primavera vs verano) sobre la estructura y composición taxonómica y funcional de las comunidades de AANCP y de sus parasitoides, y sobre sus redes de interacción. Se encontraron notables cambios estacionales en la estructura taxonómica y funcional, y en la composición tanto de la comunidad de AANCP como de parasitoides. Sin embargo, no encontramos cambios estacionales en el porcentaje de parasitismo, y los pocos cambios estacionales en la estructura de la red de interacción parecían principalmente motivados por cambios en el tamaño de la red. Por último, en el capítulo 3 se estudiaron los efectos de los factores espaciales locales (ambiente de nidificación: granjas vs agrupaciones de árboles) y paisajísticos (gradiente de cobertura agrícola) sobre la estructura taxonómica y la composición de las comunidades de AANCP y de sus parasitoides, y sobre sus redes de interacción. La estructura y composición de la comunidad AANCP, así como la estructura de la red, fueron mucho más dependientes de los factores locales que de los factores del paisaje. Los hábitats abiertos asociados con explotaciones extensivas favorecen la diversidad local de AANCP (especialmente abejas) lo que origina redes de interacción huésped-parasitoide más complejas en comparación con áreas boscosas.<br>One of the main goals in ecology is to understand how biodiversity is spatial and temporally structured, and which are the mechanisms underlying biodiversity gradients at different spatial and temporal scales. In this thesis, I analyze spatial and temporal variability in bee/wasp (hosts) and their parasitoid communities, and in the antagonistic interaction networks between them. Bees, wasps and their parasitoids are related to key ecosystem functions (e.g., pollination or herbivore populations control). Bee and wasp species show notably seasonal differences in their phenology. Bee species also show different thermoregulatory capabilities in relation with their body size (the bigger the bee species, the more ‘endothermic’ the species are). So, it could be hypothesized a relationship between body size (~endothermic capabilities) and ambient temperature in the period of adult flying activity. Bee and wasp communities also have been shown to be spatially heterogeneous in response to food and nesting resources. Temporal and spatial changes in bee/wasp communities are expected to impact in their parasitoid communities, as they depend on their host communities. Moreover, if host and parasitoid community structure and composition change over space and time, their functional traits, interaction patterns, network structure and ecosystem functionality are also expected to change spatio-temporally. In Chapter 1 we tested the body size-temperature relationship along an intra-annual, seasonal environmental temperature gradient using a Mediterranean regional bee fauna. We expected to find larger bee species (i.e. more endothermic species) in colder seasons, and progressively smaller bee species towards warmer seasons. This approaches to the Bergmann’s rule along a temporal temperature gradient (instead of their classical formulation along geographical gradients). We found a different relationship between body size and ambient temperature for large (‘endothermic’) and small (ectothermic) bee species: species larger than 27.81 mg (dry weight) followed Bergmann’s rule, whereas species below this threshold did not (no relationship at all). Our results extend Bergmann’s rule to a temporal gradient and are coherent with the physiological mechanism proposed originally by Bergmann himself (“thermoregulatory hypothesis”). In order to analyze spatial and temporal variability in antagonistic interaction networks, we used cavity-nesting bees and wasp communities (‘CNBW’, acting as ‘hosts’), and their interacting ‘parasitoid’ communities in a temperate zone (Chapters 2 and 3). In Chapter 2, we studied the effects of seasonality (spring vs. summer) on taxonomic and functional structure and composition of CNBW and their parasitoid communities, and on their interaction networks. We found strong seasonal changes in taxonomic and functional structure and composition of both the CNBW host and their parasitoid communities. However, we did not find seasonal shifts in percent parasitism, and the few seasonal changes in the structure of the host-parasitoid interaction network appeared to be mostly driven by changes in network size. Our results underscore the need to consider functional traits and to incorporate a temporal component into network analysis if we are to understand the global relationship between network structure and ecosystem function. Finally, in Chapter 3 we studied the effects of local (nesting environment: farms vs tree stands) and landscape (forest-cropland gradient) spatial factors on taxonomic structure and composition of CNBW hos and their parasitoid communities, and on their interaction networks. CNBW host community structure and composition, as well as network structure, were much more dependent on local than on landscape factors. Open habitats associated with extensively farmed exploitations favor local CNBW diversity (especially bees) and result in more complex host–parasitoid interaction networks in comparison to forested areas. This study highlights the conservation value of this kind of open habitat in view of the progressive abandonment of extensively cultivated farmland in favor of agricultural intensification and reforestation taking place in Europe.

Bibliography: leaves 108-128. Maternal protein secretions from the parthenogenetic endoparasitoid wasp Venturia canescens Gravenhorst (Hymenoptera: Ichneumonidae) were analysed regarding their ability to protect the parasitoid egg against the host immune system. These secretions include nucleic acid-free glycoproteins, called virus-like particles (VLPs) that are produced in the ovarian calyx gland. Results found that VLP-free calyx fluid has the ability to actively suppress host defence reactions, rather than VLPs on the egg surface being solely responsible for the protection of the egg. It was shown the VLP-free calyx fluid contains serine protease-inhibitor activity capable of transiently inhibiting melanogenisis and hemocyte spreading in the host. This indicates that immediately after oviposition the egg surface is not fully protected against host defence reactions. However, at later stages the egg seems to have acquired an immunologically inert surface, and is therefore protected after the suppressive activity of the calyx fluid has disappeared.

Parasitoides interferem no sistema imunológico de seus hospedeiros, influenciando a expressão de genes relacionados à resposta celular e humoral, podendo interferir na relação hospedeiro - microbiota intestinal. Além disso, parasitoides induzem alterações fisiológicas no hospedeiro que alteram o consumo e a utilização de alimento, e que podem influenciar a microbiata intestinal do mesmo. Alterações nessa microbiota poderiam afetar as relações e contribuições ao hospedeiro e, consequentemente, influenciar o desenvolvimento do próprio parasitoide. O objetivo deste trabalho foi o de verificar o efeito do parasitismo por Cotesia flavipes (Cameron) (Hymenoptera: Braconidae) na estrutura e no potencial funcional de contribuição da microbiota intestinal de Diatraea saccharalis (F.) (Lepidoptera: Crambidae), hospedeiro permissivo, e de Spodoptera frugiperda (J.E. Smith) (Lepidoptera: Noctuidae), hospedeiro não-permissivo. Além disso, buscou-se verificar se as secreções utilizadas pelo parasitoide (veneno, fluidos do cálice e virus simbionte) na regulação hospedeira estariam associadas à manipulação da microbiota intestinal do hospedeiro. O efeito do parasitismo na microbiota intestinal associada às porções antero-mediana e posterior do intestino dos hospedeiros estudados foi avaliado na fase inicial (1 DAP - dia após o parasitismo), intermediária (5 DAP) e final (9 DAP) do desenvolvimento larval do parasitoide. A avaliação foi feita por meio da comparação da diversidade e abundância de bactérias associadas ao trato intestinal de D. saccharalis e S. frugiperda parasitadas ou não por C. flavipes. A caracterização das bactérias foi feita via análise metagenômica em plataforma Illumina MiSeq utilizando a região V4 do gene ribossomal 16S. O pacote de softwares QIIME foi utilizado para a atribuição taxonômica das mesmas e o potencial funcional foi inferido por meio do software PICRUSt. O parasitismo afetou a abundância e diversidade de unidades taxonômicas operacionais (UTOs) da microbiota intestinal da porção antero-mediana e posterior de ambos hospedeiros. As alterações observadas para as duas regiões intestinais investigadas não seguiram o mesmo padrão ao longo do desenvolvimento do parasitoide. As análises realizadas também demonstraram que as alterações da microbiota induzidas pelo parasitismo refletiram em alterações significativas no potencial funcional de contribuição da microbiota associada ao trato digestivo de D. saccharalis e S. frugiperda. As análises da microbiota de lagartas pseudo-parasitadas demonstraram que as secreções maternas injetadas pela fêmea do parasitoide no momento do parasitismo estão envolvidas, pelo menos parcialmente, com os processos que levam às modificações na diversidade e abundância da microbita intestinal hospedeira, assim como de seu potencial de contribuição funcional. Esses resultados indicam que outros fatores/alterações produzidos em condições normais de parasitismo, seja pela influência de secreções de teratócitos e das próprias larvas do parasitoide em desenvolvimento também estão envolvidos na manipulação da microbiota hospedeira. Várias das alterações observadas no potencial de contribuição da microbiota intestinal do hospedeiro podem refletir sua qualidade nutricional e, consequentemente, favorecer sua exploração pelo parasitoide. Assim, o processo de regulação hospedeira por parasitoides se estende ao conjunto de organismos associados que compõem o holobionte representado pela lagarta hospedeira.<br>Parasitoids interfere with the immune system of their hosts by influencing the expression of genes related to cellular and humoral responses, which may interfere with the host - gut microbiota relationship. Furthermore, parasitoids induce physiological changes in the host, modifying food consumption and utilization, influencing then the host gut microbiota. These changes can affect the relationship and contributions of the gut microbiota to the host and therefore influence parasitoid development. This study aimed to evaluate the effect of parasitism by Cotesia flavipes (Cameron) (Hymenoptera: Braconidae) in the structure and potential functional contribution of the gut microbiota of the permissive host Diatraea saccharalis (F.) (Lepidoptera: Crambidae) and the non-permissive host Spodoptera frugiperda (JE Smith) (Lepidoptera: Noctuidae) . In addition, the participation of the secretions female parasitoids (venon, calyx fluid and symbiotic virus) use in host regulation in the manipulation of the host gut microbiota was also investigated. The effects of host parasitization on the microbiota associated with the anterior (foregut-midgut) and posterior (hindgut) portions of host gut were evaluated at the early (1 DAP - day after parasitism), intermediate (5 DAP) and final (9 DAP) stages of parasitoid larval development. The diversity and abundance of the gut microbiota of D. saccharalis and S. frugiperda was compared in between parasitized and non-parasitized larvae by C. flavipes. The gut microbiota was characterized by sequencing the V4 region of the 16S ribosomal gene using the Illumina MiSeq platform. The software package QIIME was used for taxonomic attribution and the PICRUSt software was used to infer the potential funcional contribution of the gut microbiota. Host parasitization affected the abundance and diversity of operational taxonomic units (OTUs) of the two gut regions investigated (foregut-midgut and hindgut) in both hosts. The changes observed for both gut regions did not follow the same pattern throughout parasitoid development. Changes in the gut microbiota induced by parasitization reflected in significant changes in the potential of the functional contribution of the gut microbiota associated with D. saccharalis and S. frugiperda. Analyses of pseudo-parasitized larvae demonstrated that the maternal secretions female parasitoids inject when ovipositing are involved, at least partially, with the processes that lead to changes in the abundance, diversity and potential functional contribution of the host gut microbita. These results indicate that other factors / changes produced during normal parasitization, such as secretions from teratocytes and/or the developing parasitoid larvae can also be involved in the manipulation of host gut microbiota. Several of the changes observed in the potential contribution of the host gut microbiota may reflect its nutritional quality and therefore favor host exploitation by parasitoids. Thus, the process of host regulation by parasitoids also involves the regulation of the gut-associated bacteria, which altogether comprise the holobionte represented by the host larvae.

Submitted by LUCIANE KATARINE BECCHI null (luciane.becchi@hotmail.com) on 2017-10-04T11:36:12Z No. of bitstreams: 1 Dissertação final LUCIANE.pdf: 1506577 bytes, checksum: 7466cc527263feec458778405126fce6 (MD5)<br>Approved for entry into archive by Monique Sasaki (sayumi_sasaki@hotmail.com) on 2017-10-04T17:35:11Z (GMT) No. of bitstreams: 1 becchi_lk_me_bot.pdf: 1506577 bytes, checksum: 7466cc527263feec458778405126fce6 (MD5)<br>Made available in DSpace on 2017-10-04T17:35:11Z (GMT). No. of bitstreams: 1 becchi_lk_me_bot.pdf: 1506577 bytes, checksum: 7466cc527263feec458778405126fce6 (MD5) Previous issue date: 2017-07-14<br>Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)<br>O percevejo-bronzeado, Thaumastocoris peregrinus Carpintero & Dellapé, 2006 (Hemiptera: Thaumastocoridae), nativo da Austrália, está presente no Brasil desde 2008. De forma agregada, a distribuição e disseminação do percevejo-bronzeado foi rápida nos estados produtores de eucalipto, sendo relatado causando danos em plantios em 14 estados brasileiros. Em 2012, o parasitoide de ovos Cleruchoides noackae Lin & Huber, 2007 (Hymenoptera: Mymaridae) foi introduzido no país, para criação massal em laboratório e liberação à campo, para o controle biológico da população do percevejo-bronzeado. No entanto, ainda há poucos estudos de biologia e comportamento de C. noackae e da dinâmica entre o parasitoide e o hospedeiro. O objetivo foi determinar as exigências térmicas e duração do desenvolvimento de C. noackae, em ovos de T. peregrinus, em diferentes temperaturas; avaliar o parasitismo de ovos de T. peregrinus, por C. noackae em diferentes temperaturas, e estudar, em condições de laboratório, o comportamento de corte, acasalamento e oviposição e o parasitismo de fêmeas virgens e acasaladas de C. noackae em ovos de T. peregrinus. A temperatura, influenciou o tempo de desenvolvimento de C. noackae, notando-se redução do período ovo-adulto com o aumento da temperatura. Fêmeas e machos do parasitoide necessitam de temperaturas acima de 7,34°C e 7,59°C e acúmulo de 298,50 e 289,85 graus-dia, respectivamente, para completar o desenvolvimento. A viabilidade de emergência foi também, afetada pela temperatura, com 20% de parasitoides retidos (não emergidos) a 30ºC. A temperatura influenciou, o parasitismo de C. noackae em ovos de T. peregrinus. Temperaturas entre 21 e 27ºC foram as melhores para o parasitismo de C. noackae em ovos de T. peregrinus, com maior taxa de parasitismo nas primeiras 24 horas. No estudo comportamental de C. noackae, não foi observado comportamento de côrte e ocorre apenas uma cópula entre o casal. Fêmeas virgens e copuladas encontraram o primeiro hospedeiro em 15,21 e 17,14 minutos e os demais em 3,85 e 0,86 minutos, respectivamente. O tempo de forrageamento e inserção do ovipositor de fêmeas virgens e copuladas C. noackae em ovos de T. peregrinus, foi de 24 e 21 segundos e de 5,13 e 3,69 minutos, respectivamente. Fêmeas virgens e copuladas inseriram o ovipositor com maior frequência nas laterais e opérculo do ovo de T. peregrinus. Em uma hora, mais de 50% dos ovos de T. peregrinus oferecidos às fêmeas de C. noackae foram parasitados, indicando alto e rápido parasitismo. A razão sexual de 0,00 (fêmeas virgens) e 0,68 (fêmeas copuladas) confirmam a partenogênse arrenótoca de C. noackae. Fêmeas virgens e copuladas de C. noackae tem capacidade semelhante de parasitismo em laboratório, 89%, porém, a viabilidade de emergência do parasitoide é menor para fêmeas copuladas. Os resultados apresentados neste trabalho podem ser utilizados para o ajuste de técnicas de criação massal em laboratório deste parasitoide e manejo do percevejo-bronzeado.<br>The bronze bug Thaumastocoris peregrinus Carpintero & Dellapé, 2006 (Hemiptera: Thaumastocoridae) is an eucalypt sap-sucking, native of Australia, present in Brazil since 2008. Its distribution and dissemination, in aggregate form, was fast in the producing states of eucalyptus, being reported causing damages in plantations in 14 states Brazilians. In 2012, the parasitoid of eggs Cleruchoides noackae Lin & Huber, 2007 (Hymenoptera: Mymaridae) was introduced in the country for laboratory mass-rearing and release into the field for the biological control of the population of the bronze bug. However, there are few studies on a C. noackae biology and behavior and dynamics between the parasitoid and its host. The objective was to determine the thermal requirements and the duration of C. noackae development in eggs of T. peregrinus at different temperatures; to evaluate the parasitism of C. noackae in T. peregrinus eggs at different temperatures and to study, under laboratory conditions, the parasitoid courtship, mating and oviposition behaviour and the parasitism of virgin and mated females of C. noackae on T. peregrinus eggs. The temperature, influenced the development time of C. noackae, observing reduction of the biological cycle (egg-adult) with the increase of the temperature. Females and males of the parasitoid require temperatures above 7.34°C and 7.59°C and accumulation of 298.50 and 289.85 degrees day, respectively, for their development. The viability of emergence was affected by temperature with 20% of parasitoids retained (not emerged) at 30ºC. The temperature also influenced the parasitism of C. noackae in T. peregrinus eggs. Temperatures ranging from 21 to 27ºC were better for the parasitism of C. noackae in eggs of T. peregrinus with the highest rate of parasitism occurs within the first 24h. In the behavioral study of C. noackae, no courtship behavior was observed and only one copula between the couple occurred. Virgins and copulated females found the first host in 15.21 and 17.14 minutes and the next host in 3.85 and 0.86 minutes, respectively. The foraging time and duration of ovipositor insertion into T. peregrinus eggs was 24 and 21 seconds and 5.13 and 3.69 minutes, respectively. Virgins and copulated females of C. noackae inserted the ovipositor more frequently on the sides of the egg and operculum of the T. peregrinus egg. At one hour, more than 50% of T. peregrinus eggs offered to C. noackae females were parasitized, indicating high and rapid parasitism. The sex ratio of 0.00 (virgin females) and 0.68 (females copulated) confirming the arrhenotokous parthenogenesis of C. noackae. Female virgins and copulated C. noackae have similar parasitism capacity in the laboratory, 89%, however, the viability of emergence is lower for copulated females. The results presented in this work can be used for adjustment of techniques for parasitoid mass-rearing and bronze bug management.

Os inseticidas neonicotinóides são atualmente o principal grupo químico utilizado para o controle de insetos sugadores, constituindo-se uma boa opção para o controle de Diaphorina citri Kuwayama. Outra opção de controle do psilídeo tem sido a exploração do parasitóide Tamarixia radiata (Waterson). A compatibilidade dessas duas estratégias de controle poderia auxiliar na implementação de programas de manejo integrado de pragas na cultura do citros. No entanto, faltam estudos sobre a caracterização da suscetibilidade de D. citri para os inseticidas neonicotinóides e o impacto desses inseticidas sobre T. radiata. Sendo assim, os objetivos do trabalho foram o de caracterizar a suscetibilidade de D. citri a inseticidas neonicotinóides, realizar o monitoramento da suscetibilidade a esses inseticidas em populações de D. citri coletadas em pomares de diferentes regiões do Estado de São Paulo e avaliar os efeitos letais e subletais desses inseticidas sobre o parasitóide T. radiata. Os inseticidas avaliados foram: thiamethoxam, thiacloprid e imidacloprid. O método de bioensaio adotado foi o de contato residual para a caracterização da suscetibilidade de D.citri a esses inseticidas. O monitoramento da suscetibilidade a esses inseticidas em diferentes populações de D. citri foi realizado com concentrações diagnósticas baseadas na concentração letal 95 (CL95) de cada inseticida. Para avaliar os efeitos letais e subletais desses inseticidas sobre T. radiata foram realizados bioensaios de contato direto em adultos e pupas, toxicidade residual em adultos, e persistência da atividade biológica desses inseticidas sobre a superfície de folhas de citros. Entre os neonicotinóides testados, a maior toxicidade a D. citri foi observado com thiametoxam, seguidos por imidacloprid e thiaclopid. Os resultados do monitoramento apresentaram diferenças significativas na suscetibilidade das populações de D. citri aos inseticidas neonicotinóides. Para thiamethoxam, a sobrevivência estimada para as populações de D. citri testadas na concentração diagnóstica variou entre 5,5 e 16%, para thiacloprid variou entre 4,5 e 22,5 % e para imidacloprid entre 4 e 14%. Uma alta toxicidade desses inseticidas foi observada para adultos e pupas de T. radiata. No entanto, os efeitos subletais desses inseticidas a 10% da concentração recomendada sobre o estágio de pupa causou redução significativa no parasitismo somente com thiamethoxam. A emergência, longevidade e razão sexual de T. radiata não foram afetadas pelos inseticidas avaliados. A toxicidade residual de thiametoxam, thiacloprid e imidaclopid em folhas de mudas de citros foi relativamente elevada para adultos de T. radiata e com persistência de pelo menos 14 dias.<br>The neonicotinoids insecticides are currently the main chemical group used for controlling sucking pests and represent a good option for the control of Diaphorina citri Kuwayama. Another control alternative of this pest is the exploitation of parasitoid Tamarixia radiata (Waterson). The compatibility of these control strategies could be very helpful for the implementation of integrated pest management program in citrus. However, there are few studies on the characterization of the susceptibility of D. citri to neonicotinoid insecticides and the evaluation of the impact of these insecticides on T. radiata. Therefore, the objectives of this research were to evaluate the susceptibility of D. citri to neonicotinoid insecticides, to monitor the susceptibility to these insecticides in D. citri populations collected from different citrus groves in the State of São Paulo, and to evaluate the lethal and sublethal effects of these insecticides on T. radiata. The insecticides evaluated in this study were: thiamethoxam, thiacloprid e imidacloprid. A residual contact bioassay was used to characterize the susceptibility of D. citri to these insecticides. A diagnostic concentration bioassays based on lethal concentration 95 (LC95) of each insecticide were used for monitoring the susceptibility of D. citri populations. The lethal and sublethal effects of these insecticides on T. radiata were conducted by using direct contact bioassays on adult and pupal stages of T. radiata, residual contact bioassays and persistence of biological activity of these insecticides on citrus leaf surface. Among the neonicotinoid insecticides tested, the highest toxicity was observed with thiametoxam, followed by imidacloprid and thiaclopid. A significant difference in the susceptitibity to neonicotinoid insecticides was detected in D. citri populations. For thiamethoxam, the survivorship at diagnostic concentration varied from 5.5 to 16%, for thiacloprid varied from 4.5 to 22.5 %, and for imidacloprid from 4 to 14%. The toxicity of these insecticides was high to adult and pupal stages of T. radiata. However, the sublethal effects of these insecticides at 10% of the recommended rate on pupae stage caused the reduction of the parasitism capacity only with thiamethoxam. The emergence, longevity and sexual ratio of T. radiata were not affected by any insecticide tested. The residual toxicity of thiametoxam, thiacloprid and imidaclopid sprayed on citrus seedling leaves was relatively high to D. citri adults and lasted at least 14 days.

In this paper, 48 ichneumonid species of 14 genera, two tribes of the subfamily Pimplinae were listed on the basis of material collected during the time from 2003 to 2011 in four provinces from the Central Highlands of Vietnam and a lesser degree of previous records. Among 48 aforemention pimpline no less than 16 species were described recently as new for science. The diversity of topography and vegetation types resulted in the high level of species richness and new discoveries of the region.<br>Bài báo này giới thiệu danh sách 48 loài ong cự thuộc 14 giống, 2 tộc của phân họ Pimplinae dựa trên việc phân tích các mẫu vật thu tại Tây Nguyên trong thời gian từ năm 2003–2011 và một số ít loài dựa trên các ghi nhận trước đây. Trong số đó, 16 loài mới được mô tả trong thời gian 5 năm trở lại đây. Sự đa dạng về địa hình và các kiểu thảm thực vật khác nhau đã tạo nên tính đa dạng thành phần loài và các phát hiện mới của ong cự ở khu vực Tây Nguyên đầy tiềm năng này.

The aim of this thesis project was to elucidate patterns and processes associated with the biotic interactions in a natural plant–herbivore–parasitoid food web characterized by spatial and temporal heterogeneity with regard to species composition. The system examined is based on island populations of the perennial herb Meadowsweet (Filipendula ulmaria, Rosaceae), located in the Skeppsvik Archipelago. The area is subject to isostatic rebound, amounting to 0.85 cm per year; this makes it possible to calculate the age of the rising islands. Meadowsweet colonizes new islands when they are about 100 years old. Meadowsweet is consumed by two major herbivores in the study area: Galerucella tenella and Altica engstroemi (Coleoptera: Chrysomelidae). Both herbivores overwinter in the topsoil and successful colonization occurs when the islands reach a height that prevents the beetles from being removed or killed as a result of wave wash during the winter. I found that both herbivores significantly reduced individual plant fitness and population growth rate. A “cafeteria experiment” with Galerucella showed that this beetle discriminated between plants from different islands, avoiding plants from old islands which contained high concentrations of putative defence compounds, while readily accepting plants from younger islands which contained lower concentrations of these chemicals. Further, the plant species exhibited a trade-off between growth and production of the putative defence compounds. Taken together, these results were interpreted as providing evidence of herbivore-driven evolution of resistance in Meadowsweet. Further, laboratory studies suggested that Galerucella gradually includes a less preferred host plant (Rubus arcticus, Rosaceae) in its diet as Meadowsweet resistance increases. This implies that Galerucella drives its own host-breadth enlargement by selectively inducing a ‘rent rise’ in the original host, Meadowsweet. In a number of field studies I showed that the oligophagous parasitoid Asecodes mento (Hymenoptera: Eulophidae) has a strong positive effect on Meadowsweet seed set by removing large numbers of G. tenella larvae. This top-down effect is, however, altered by the presence of a close relative of G. tenella, namely G. calmariensis, which is monophagous on Purple loosestrife (Lythrum salicaria, Lythraceae). G. tenella experiences associational susceptibility when coexisting with G. calmariensis since the latter supports a higher and more fit pool of shared parasitoids and because Meadowsweet attracts a higher proportion of the shared parasitoid females than Purple loosestrife. This implies that G. tenella densities are very low in coexisting populations and that Meadowsweet experiences associational resistance and produces more seeds when co-occurring with Purple loosestrife. Thus, selection for increased resistance in Meadowsweet is likely to be relaxed in populations mixed with Purple loosestrife. I conclude that the evolution of plant resistance is likely to depend on the length of time and intensity of selection. When Meadowsweet colonizes new islands it experiences a period of enemy-free space; followed by a midlife and ageing with selection by herbivores. The intensity of this selection does, however, depend on the presence of additional plant and herbivore species.