<p>Tolerance to herbivory is the ability of plants to maintain fitness in spite of damage. The goal of this thesis is to investigate the genetic variation and expression of tolerance within species, determine whether and in what conditions tolerance has negative side-effects, and how tolerance is affected by different ecological factors. Tolerance is investigated with special focus on the effects of different damage types, competitive regimes, history of herbivory, and polyploidization in plants. Studies are conducted as a literature review and three experiments on two cruciferous species Raphanus raphanistrum and Cardamine pratensis.</p><p>In the tolerance experiments, plants are subjected to artificial damage solely, or in a combination with natural damage. A literature review was conducted in order to investigate the effects of damage method. We found that traits related to tolerance, such as growth and fitness were not as sensitive in regard to damage method as measures of induced chemical traits, or measures of secondary herbivory.</p><p>Genetic variation of tolerance was demonstrated within populations of R. raphanistrum and between subspecies of C. pratensis. In R. raphanistrum, traits involved in floral display and male fitness were positively associated with plant tolerance to herbivore damage. A potential cost of tolerance was demonstrated as a negative correlation between levels of tolerance in high and low competitive regimes. I found no evidence of other proposed costs of tolerance in terms of highly tolerant plants suffering of reduced fitness in the absence of herbivores or trade-offs in terms of a negative association between tolerance to apical and leaf damage, or between tolerance and competitive ability. In C. pratensis, higher ploidy level in plants involved higher levels of tolerance measured as clonal reproduction. Furthermore, populations exposed to higher levels of herbivory had better tolerance than populations exposed to lower levels of herbivory. In this thesis, I demonstrate evidence of different components for the evolution of tolerance in plants: genotypic variation, selective factors in terms of costs and ploidization, and selective agents in terms of changing environment or herbivore pressure.</p>

Tolerance to herbivory is the ability of plants to maintain fitness in spite of damage. The goal of this thesis is to investigate the genetic variation and expression of tolerance within species, determine whether and in what conditions tolerance has negative side-effects, and how tolerance is affected by different ecological factors. Tolerance is investigated with special focus on the effects of different damage types, competitive regimes, history of herbivory, and polyploidization in plants. Studies are conducted as a literature review and three experiments on two cruciferous species Raphanus raphanistrum and Cardamine pratensis. In the tolerance experiments, plants are subjected to artificial damage solely, or in a combination with natural damage. A literature review was conducted in order to investigate the effects of damage method. We found that traits related to tolerance, such as growth and fitness were not as sensitive in regard to damage method as measures of induced chemical traits, or measures of secondary herbivory. Genetic variation of tolerance was demonstrated within populations of R. raphanistrum and between subspecies of C. pratensis. In R. raphanistrum, traits involved in floral display and male fitness were positively associated with plant tolerance to herbivore damage. A potential cost of tolerance was demonstrated as a negative correlation between levels of tolerance in high and low competitive regimes. I found no evidence of other proposed costs of tolerance in terms of highly tolerant plants suffering of reduced fitness in the absence of herbivores or trade-offs in terms of a negative association between tolerance to apical and leaf damage, or between tolerance and competitive ability. In C. pratensis, higher ploidy level in plants involved higher levels of tolerance measured as clonal reproduction. Furthermore, populations exposed to higher levels of herbivory had better tolerance than populations exposed to lower levels of herbivory. In this thesis, I demonstrate evidence of different components for the evolution of tolerance in plants: genotypic variation, selective factors in terms of costs and ploidization, and selective agents in terms of changing environment or herbivore pressure.

La conservation au niveau population est extrêmement nécessaire pour limiter la perte de biodiversité au sein d'une espèce ou d'un complexe d'espèces. Ainsi, l'évaluation de la variabilité inter-populationnelle dans le complexe est reconnue comme première étape importante pour bien définir les plans de conservation des espèces menacées. Arenaria grandiflora form un complexe d'espèces herbacées pérennes à courte durée de vie (4 ans en moyenne) menacé dans certains sites de ses zones de distribution en Europe. A ce jour, sa taxonomie n'est pas bien résolue, ce qui entraîne des problèmes potentiels pour mettre en oeuvre une conservation efficace de ce taxon. Une variation inter-populationnelle du complexe d'espèces A. grandiflora est présentée dans cette étude aux niveaux génétiques, cytogénétiques et morphométriques. Quatre méthodes ont été utilisées : des marqueurs microsatellites nucléaires, une approche cytogénétique, la cytométrie en flux, et enfin la morphométrie sur les feuilles. De plus, les études phénotypiques de variation de taux de germination entre stocks de graines ont été développées. Une différenciation significative entre les profils de variations moléculaires, cytogénétiques et phénotypiques a été détectée dans le complexe d'espèces. Deux cytotypes (diploïdes 2n=2x=22 et tétraploïdes 2n = 4x = 44) ont été mis en évidence en utilisant à la fois des méthodes classiques et des méthodes plus récentes (marqueurs microsatellites, nombres chromosomiques et cytométrie de flux). Le complexe d'espèces d'A; grandiflora présente une forte variation de la valeur de l'ADN 2C, la taille du génome varie de 2.11 ± 0.74 pg à 2.70 ± 0.11 pg pour les populations diploïdes et de 4.30 ± 1.51 pg à 5.27 ± 0.14 pg pour les populations de tétraploïdes. En outre, les grains de tétraploïdes germent significativement mieux que les graines des diploïdes. Les feuilles diffèrent considérablement entre les diploïdes (aciculaires et linéaires) et les tétraploïdes (lancéolées). Cette étude peut être considérée comme préliminaire pour une révision taxonomique de ce complexe d'espèces. D'autre part, grâce à l'ensemble des résultats obtenus, il est également possible de revisiter le concept d'unités évolutives significatives (ESUs) dans le complexe d'espèces A. grandiflora et donc de définir les groupes de populations devant faire l'objet de mesures distinctes. Ainsi, il est possible d'évaluer la pertinence de plans déjà entrepris et de proposer de nouveaux plans de restauration efficaces pour l'avenir de ce complexe d'espèces. / Population-level conservation is being extremely required to restrain the biodiversity loss within a species. So, the assessment of the variability within the species complex is being renowned as an important first step to well implement the future conservation settings for threatened species. The species complex of Arenaria grandiflora is a short-lived perennial herbaceous and a threatened taxon in certain of sites of its distribution areas in Europe, with unresolved gentics and taxonomy, which lead to potential problems in the conservation and utilization of the resource. A differenciation among populations of the species complex of A. grandiflora is presented in this study based on the genetic, cytogenetic and phenotypic patterns. Intraspecific ploidy level varaition is an important aspect of numerous species, so, the present study explores this phenomenon within the A. grandiflora species complex in some type of populations (27 natural populations). To infer the intraspecific genetic and cytogenetic patterns of variability among the studied natural populations of the investigated species complex (A. grandiflora), three methods were used : nuclear microsatellite markers, cytogenetic and flow cytometry approaches. Moreover, the phenotypic patterns of variation among both the stock of seeds and the herbarium materials of A. grandiflora were defined. These patterns were detected using three methods of seed germination (in vitro culture, filter papers and potting soil) and morphometric approaches. A significant differentiation among populations' patterns of molecular, cytogenetic and phenotypic variation was detected within the A. grandiflora species complex. Presence of two closely related cytotypes (diploids 2n=2x=22 and tetraploids 2n=4x=44) was detected using both classical and more recent methods (chromosome number count and flow cytometry respectively). The species complex of A. grandiflora exhibits high variation in 2C-DNA value, the genome size ranges from 2.11 ± 0.74 pg to 2.70 ± 0.11 pg for the diploid populations and from 4.30 ± 1.51 pg to 5.27 ± 0.14 pg for the tetraploid populations. Moreover, the seeds of tetraploids germinate well and in high proportion than the seeds of the diploid ones. In addition, both acicular and linear leaves from the diploid populations differ significantly within the diploids and with the lanceolate leaves of the tetraploid ones. New protocol of seed germination for the tetraploids by in vitro culture after scarifying was described for th first time. The affected factors on seed germination percentages were determinated by an explanatory model of six predictors (altitude, longitude, latitude, ploidy levls, both period and condition of seed storage). Consequently, all these findings are fundamental for the determination of the evolutionarily significant units (ESUs) within A. grandiflora species complex and thus the definition of efficient restoration plans in the future. This study would consider as the preliminary signal for necessary revision for the intraspecific taxonomic keys problematic for this species complex.