Studies assessing the evolution of plant traits frequently focus on pollinators as the primary drivers of floral trait evolution. However, herbivores can also play a role, and, under some circumstances, may even impose stronger selection on floral display than pollinators. This is especially true when the traits under selection are linked to anti-herbivore defense. Here I describe a study in which I quantified the relative role of herbivores and pollinators in selection for floral traits in the North American invasive plant, purple loosestrife (Lythrum salicaria). Because L. salicaria responds to leaf-chewing herbivores by producing compensatory tissue growth, and this in turn alters the architecture of the floral display, I further tested whether herbivores can indirectly modify pollinator-mediated selection through this pathway. Using a split-plot design, I measured pollen limitation and reproductive output in experimentally manipulated plants in the presence and absence of simulated herbivore damage in order to quantify the effects of damage and pollination on natural selection for floral display. My results showed that damage significantly increased direct selection (beta-i) for earlier flowering time and decreased selection on the number of inflorescences, even more than pollinators did. Because damage did not modify pollinator-mediated selection for floral display traits, the selection imposed by herbivores is likely only having direct effects. My findings demonstrate the importance of considering multiple agents of selection and their potential interactions when quantifying natural selection in a study system. In particular, it is important to consider that the agent most frequently studied may not be imposing the brunt of selection.

Within a lifetime of an individual, there are changes in the organism leading to an an impairment of phenotype, survival and reproduction. The expression of phenotypic traits often increases with age. Higher expression of these traits in the young age can lead to decreased probability of survival till the next season, influencing reproductive success of an individual. That is why organisms are challenged with trade-offs, when they have to allocate their limited resources either into the reproduction in form of expression of extravagant sex ornaments or into the life span. If the ornament evolved as a handicap, its exaggerated size could reduce viability of low quality individuals. Therefore, the ornament becomes a predictor of quality of its bearer and its size indicates the ability to survive till the next season. Both sexual ornamentation and inter-individual variation in life span may result in differences in reproductive success (fitness) and targeted by sexual or natural selection. Variability in reproductive success, mainly in monogamous species, can be increased also by sexual promiscuity and extra-pair mating and extra-pair copulations. This strategy can increase the opportunity for selection of traits that play role in choosing of extra-pair mate. In this work, I focused on two potential...

La théorie de la sélection sexuelle a été largement élaborée à partir du constat de dimorphisme sexuel chez les espèces à sexes séparés. Une de ses caractéristiques générales est une sélection plus forte pour l'augmentation du nombre de partenaires sexuels chez les mâles que chez les femelles qui résulterait d'un investissement différentiel dans les descendants entre les deux sexes (anisogamie). Si hermaphrodisme et sélection sexuelle sont considérés comme compatibles depuis les travaux de Charnov (1979), les études sur le sujet restent rares que ce soit chez des animaux ou des plantes. Une raison importante est que la méthodologie disponible pour quantifier la sélection sexuelle ne prend pas en compte les particularités des hermaphrodites (par ex., corrélations ou effets croisés entre les deux fonctions sexuelles d'un même individu, autofécondation). Le premier objectif de cette thèse est de combler cette lacune méthodologique en proposant un cadre travail adapté aux hermaphrodites, que nous appliquons dans une étude empirique chez Physa acuta, un gastéropode hermaphrodite d'eau douce. Nous observons que la sélection sexuelle est plus intense sur la fonction mâle, comme généralement chez les espèces gonochoriques. Par ailleurs, nous ne détectons aucun effet des particularités des hermaphrodites dans cette expérience. Dans un deuxième temps, nous nous intéressons de manière plus détaillée aux composantes du succès reproducteur mâle (RSm). Nous montrons que chez P. acuta il existe une priorité spermatique au premier partenaire mâle lorsque plusieurs individus sont en compétition. Enfin, nous proposons une décomposition de la variance de RSm en ses composantes pré- et post-copulatoires, qui représentent respectivement 60 et 40% de la variance. Dans la troisième partie, nous intégrons la sélection sexuelle à l'étude de l'évolution de l'allocation sexuelle d'un hermaphrodite, via une approche d'évolution expérimentale chez P. acuta. Menée sur plus de 10 générations, elle vise à faire évoluer l'allocation sexuelle de manière disruptive (lignées mâle ou femelle) en sélectionnant les composantes mâle et femelle du succès reproducteur. Les résultats préliminaires suggèrent qu'il est possible de manipuler l'allocation sexuelle chez un hermaphrodite simultané en sélectionnant sur son régime d'appariement. Nous concluons que l'anisogamie suffit à justifier l'existence de la sélection sexuelle sans avoir à supposer un dimorphisme sexuel. Son étude chez les hermaphrodites simultanés ouvre des perspectives pour la compréhension du rôle de l'allocation sexuelle dans l'évolution des systèmes de reproduction. / A cornerstone of the theory of sexual selection in gonochoric species is sexual dimorphism. A very general result is stronger selection on males than on females for increasing mating success, and this fundamentally relies on differential investment in offspring between the sexes (anisogamy). Although sexual selection does operate in hermaphroditic species as well, few empirical studies have been performed whether in animals or in plants. The main reason is that the current framework for studying sexual selection does not incorporate the particularities of hermaphrodites, including correlations or cross-sex effects between sex functions and self-fertilization. The first goal of this thesis is to fill this gap by proposing an appropriate framework for hermaphrodites (generalizing that available for gonochoric species). It was applied to approach sexual selection in the hermaphroditic freshwater gastropod Physa acuta. Sexual selection turns out to be stronger on the male than on the female function, as classically observed in gonochorists. Moreover, we do not detect any effect in relation to hermaphrodites' particularities. We then focus on the components of male reproductive success (RSm) in more details. We detect a pattern of sperm precedence in conditions of sperm competition. We develop a new method for decomposing the variance in RSm into pre- and post-copulatory components (representing 60 and 40% of the variance respectively). The third section aims at integrating sexual selection in studies of sex allocation and its evolution. It relies on a protocol of experimental evolution in P. acuta. Conducted over more than 10 generations. Its aim is to observe the evolution of sex allocation by disruptively selecting male and female components of reproductive success. Preliminary results indeed suggest that it is possible to manipulate sex allocation in a simultaneous hermaphrodite by manipulating its mating system. We conclude that anisogamy alone is a sufficient condition for sexual selection to proceed, and that sexual dimorphism is not required. Study sexual selection in simultaneous hermaphrodites gives insights for understanding the role of sex allocation in the evolution of mating systems.

<p>In this thesis, I combined comparative and experimental approaches to examine selection on reproductive traits, and population differentiation in the insect-pollinated, outcrossing, perennial herb <i>Arabidopsis lyrata</i>. More specifically, I (1) determined whether selection on flowering phenology and floral display can be attributed to interactions with pollinators and herbivores, (2) examined whether population differentiation in flowering phenology and floral display is correlated with current selection on these traits, and (3) tested for local adaptation from contrasting environments in Europe.</p><p>A field experiment conducted in a Swedish population demonstrated, that interactions with pollinators may markedly affect selection on both floral display and phenology of flowering. In an alpine population in Norway, grazing damage to inflorescences strongly influenced selection on floral display. The results suggest that variation in the abundance of pollinators and herbivores should contribute to spatio-temporal variation in selection on flowering phenology and floral display in <i>A. lyrata</i>. </p><p>A common-garden experiment showed that flowering phenology and floral display vary among Scandinavian populations of <i>A. lyrata</i>. For some traits patterns of population differentiation were consistent with differences in the direction and strength of phenotypic selection determined in comparisons (a) between an alpine population in Norway and a coastal population in Sweden, and (b) among coastal populations in Sweden. This suggests that current selection contributes to the maintenance of genetic differentiation in these traits.</p><p>Adaptive differentiation among populations was examined in a reciprocal transplant experiment that included populations from three contrasting environments, alpine Norway, coastal Sweden and lowland, continental Germany. The experiment provided evidence for local adaptation, and indicated that populations have diverged in traits affecting plant establishment and early growth.</p>

In this thesis, I combined comparative and experimental approaches to examine selection on reproductive traits, and population differentiation in the insect-pollinated, outcrossing, perennial herb Arabidopsis lyrata. More specifically, I (1) determined whether selection on flowering phenology and floral display can be attributed to interactions with pollinators and herbivores, (2) examined whether population differentiation in flowering phenology and floral display is correlated with current selection on these traits, and (3) tested for local adaptation from contrasting environments in Europe. A field experiment conducted in a Swedish population demonstrated, that interactions with pollinators may markedly affect selection on both floral display and phenology of flowering. In an alpine population in Norway, grazing damage to inflorescences strongly influenced selection on floral display. The results suggest that variation in the abundance of pollinators and herbivores should contribute to spatio-temporal variation in selection on flowering phenology and floral display in A. lyrata. A common-garden experiment showed that flowering phenology and floral display vary among Scandinavian populations of A. lyrata. For some traits patterns of population differentiation were consistent with differences in the direction and strength of phenotypic selection determined in comparisons (a) between an alpine population in Norway and a coastal population in Sweden, and (b) among coastal populations in Sweden. This suggests that current selection contributes to the maintenance of genetic differentiation in these traits. Adaptive differentiation among populations was examined in a reciprocal transplant experiment that included populations from three contrasting environments, alpine Norway, coastal Sweden and lowland, continental Germany. The experiment provided evidence for local adaptation, and indicated that populations have diverged in traits affecting plant establishment and early growth.