Drumming was recorded for 11 of 13 Nearctic stonefly species, representing 4 families. Both male and female signals were obtained from 5 species, and were either 2-way or 3-way communications. Signals were species-specific; those of males and females varied from 3-39 and 1-14 beats/ signal, respectively. Duration of male signals varied from 105-8,016 ms; those of females, except Perlinella drymo (1 beat), varied from 402-1318 ms. Signals among related taxa showed greatest similarities. Duration of male signals of Perlinella drymo became progressively shorter at each of 4 temperatures from 7-29 0C. Females of Perlinella drymo would only repeatedly answer male signals recorded at near their own temperature, and would not repeatedly answer recorded male signals of 8 other species.

Acoustic communication in orthopterans and anurans provides a suitable model system for studying the evolutionary mechanisms of sexual selection mainly because males use acoustic signals to attract females over long distances for pair formation. Females use these signals not only to localize conspecific males but also to discriminate between potential mates. Investigations on the effect of sexual selection on acoustic signals requires an understanding of how female preferences for different features of the acoustic signal affect male mating success under ecological constraints in wild populations. The effect of female preferences on male mating success depends on the mate sampling strategy that females employ to search for potential mates. Despite its relevance, female mate sampling strategies based on male acoustic signals have rarely investigated in orthopterans and anurans, especially in the field. Considering the elaborate knowledge of the role of sensory physiology in female phonotaxis behaviour and characterization of the male acoustic signal, I used the field cricket species Plebeiogryllus guttiventris as a model system in this study. In this thesis, I first investigated the ecology of callers in wild populations. I then investigated female mate sampling strategies by incorporating relevant information on the ecology of signalers and the sensory physiology of receivers. Amount of calling activity is a strong determinant of male mating success in acoustically communicating species such as orthopterans and anurans. While many studies in crickets have investigated the determinants of calling effort, patterns of variability in male calling effort in natural choruses remain largely unexplored. I therefore investigated the spatio-temporal dynamics of acoustic chorusing behaviour in a wild population. I first studied the consistency of calling activity by quantifying variation in male calling effort across multiple nights of calling using repeatability analysis. Callers were inconsistent in their calling effort across nights and did not optimize nightly calling effort to increase their total number of nights spent calling. Next, I investigated calling site fidelity of males across multiple nights by quantifying movement of callers. Callers frequently changed their calling site across calling nights with substantial displacement but without any significant directionality. Finally, I investigated trade-offs between within-night calling effort and energetically expensive calling song features such as call intensity and chirp rate. Calling effort was not correlated with any of the calling song features, suggesting that energetically expensive song features do not constrain male calling effort. The two key features of signaling behaviour, calling effort and call intensity, which determine the duration and spatial coverage of the sexual signal, are uncorrelated and function independently Acoustic signal variation and female preference for different signal components constitute the prerequisite framework to study the mechanisms of sexual selection that shape acoustic communication. Despite several studies of acoustic communication in crickets, information on both male calling song variation in the field and female preference in the same system is lacking for most species. First, I quantified variation in the spectral, temporal and amplitudinal characteristics of the male calling song in a wild population, at two temporal scales, within and across nights, using repeatability analysis. Carrier frequency (CF) was the most repeatable call trait across nights, whereas chirp period (CP) had low repeatability. I further investigated female preferences based on song features with high and low repeatability (CF and CP respectively). Females showed no consistent preferences for CF but were more attracted towards calls with higher rates (shorter CP). I also examined the effect of signal intensity, which is known to play a critical role in female phonotaxis behaviour, on female preferences for faster calls. Females preferred louder calls over faster ones, implying a dominant role for signal intensity in female evaluation of potential mates based on acoustic signals. Call intensity was also the only signal feature that was positively correlated with male size. In the final chapter, I investigated female mate sampling strategies based on acoustic signals using both theoretical and empirical approaches. Analytical models of mate sampling have demonstrated significant differences in individual fitness returns for different sampling strategies. However these models have rarely incorporated relevant information on the ecology of signalers and the sensory physiology of receivers. I used simulation models to compare the costs and benefits of different mate sampling strategies by incorporating information on relative spacing of callers in natural choruses and the effect of signal intensity on female phonotaxis behaviour. The strategy of mating with males that were louder at the female position emerged as the optimal sampling rule in the simulations. When tested empirically in the field using callers in natural choruses, females seemed to follow the optimal strategy of mating with males that were perceived as louder at their position.

Au sein des écosystèmes, les Composés Organiques Volatils (COVs) émis par les plantes jouent un rôle majeur dans les interactions trophiques. Ces signaux olfactifs vont renseigner les insectes phytophages sur la présence de leurs hôtes mais permettent aussi de recruter les ennemis naturels. Il a été montré que certains de ces composés pouvaient être utilisés pour manipuler le comportement des insectes phytophages s’attaquant aux cultures. La stratégie push-pull, vise à combiner des stimuli positifs et négatifs pour un insecte ravageur afin de le repousser d’une culture tout en l’attirant sur une plante piège implantée en périphérie du champ où il pourra être contrôlé. L’objectif de cette thèse est d’améliorer notre compréhension et nos connaissances sur la manipulation de la communication chimique entre les plantes et les insectes phytophages à l’aide de COVs de synthèse et de préférence d’hôte afin de mettre en place sur le terrain une stratégie de type push-pull pour protéger une culture de la mouche du chou (Delia radicum). Dans un premier chapitre, nous montrons que dans un système push-pull composé (i) de diffuseur de diméthyldisulfure DMDS (composante push) et (ii) d’une bordure de choux chinois supplémenté avec de l’acétate d’hexenyl (Z-3-HAC) (composante pull), il est possible de modifier fortement le comportement d’oviposition de D. radicum sans impacter de façon négative ses principaux ennemis naturels. Les limites agronomiques et d’utilisation du DMDS atteintes, nous avons recherché d’autres COVs ayant un effet répulsif. Lors d’une seconde expérimentation en push-pull, l’eucalyptol (1-8 cinéol) à permis de réduire le nombre de pupes de D. radicum retrouvées aux pieds des plants de brocoli de 60%. Suite à ce résultat encourageant, nous avons qualifié au laboratoire l’effet inhibiteur de ce monoterpène sur l’oviposition de la mouche du chou et en avons conclu qu’il permet de masquer le bouquet de COVs attractif d’un hôte. L’efficacité de la composante push étant fortement liée à la diffusion des COVs, nous avons ensuite testé au laboratoire et sur le terrain un nouveau type de diffuseur à base de cires végétales, permettant d’émettre l’eucalyptol de manière passive mais régulière pendant plusieurs semaines. L’étude réalisée montre le fort potentiel du diffuseur qui est à la fois bon marché, facile d’utilisation et fiable en terme de diffusion, permettant ainsi d’envisager son utilisation dans des stratégies de lutte intégrée à grande échelle. Parallèlement à l’étude de la composante push, nous avons cherché à améliorer l’efficacité de la composante pull. Le chou chinois (Brassica rapa) permet grâce à sa forte attractivité de capter et détourner la pression de phytophagie de la culture d’intérêt. Nous montrons qu’au sein de sa grande diversité, certaines variétés comme Richi (appartenant à la sous espèce pekinensis) sont préférentiellement attaqué par la mouche du chou, ce qui en fait de bons candidats au développement d’une composante pull efficace. A la lumière de la bibliographie et de nos résultats, nous proposons une réflexion générale sur le système étudié. Ainsi, nous concluons que, au moins dans le cas de la mouche du chou, les COVs ne peuvent probablement pas être les seuls outils considérés dans le développement d’une stratégie push-pull. / Within ecosystems, Volatile Organic Compounds (VOCs) emitted by plants play a major role in trophic interactions. These olfactive signals will inform phytophagous insects about the presence of their hosts, but they also allow to recruit their natural enemies. It has been shown that some of these compounds could be used to manipulate the behaviour of insect crop pests. The push-pull strategy aims at combining positive and negative stimuli to push a pest out of a crop and lure it into a trap crop located in the periphery of the field, where it can be controlled. The objective of this thesis is to improve our understanding and our knowledge on the manipulation of chemical communication between plants and phytophagous insects, using synthetic VOCs and host preference as tools, to protect a crop against the cabbage root fly (Delia radicum). In a first chapter, we show that in a push-pull made of (i) DMDS dispensers as a push component and (ii) chinese cabbage strips supplemented with hexényl acetate (Z-3-HAC) as a pull component, it is possible to strongly reduce oviposition behaviour in D. radicum without impacting its natural enemies. Agronomic limits to the use of DMDS led us to search for other repulsive VOCs. In a second push-pull field experiment, using eucalyptol (1-8 cinéol), reduced by 60% the number of D. radicum pupae produced in the crop. This encouraging result led us to characterize the inhibition effect of this monoterpene in the lab and concluded that it allows to mask the attractive blend of VOCs released by the host. The efficiency of the push component being strongly linked to that of the VOC diffusion, we have then tested both in the laboratory and in the field a new kind of dispenser made of vegetal waxes, allowing to diffuse eucalyptol passively but regularly during several weeks. This study shows the good potential of this device, which is affordable, easy to use and reliable in terms of diffusion, in integrated pest management strategies at a large scale. In parallel of our studies on the push component, we have also sought to increase the efficiency of the pull component. Thanks to its strong attractivity toward D. radicum, Chinese cabbage (Brassica rapa), allows to divert pest pressure from the cash crop. We show that within its important genetic diversity, some cultivars such as Richi (of the pekinensis subspecies) are preferentially attacked by D. radicum, which makes them good candidates to develop an efficient pull component. Using both the literature and our own results, we finally propose a general discussion on the system studied. We conclude that, at least in the case of the cabbage root fly, VOCs are probably not the sole tools to consider when trying to develop a push-pull strategy.