Ecosystem engineers can govern ecosystem dynamics, yet ecosystem consequences of trait variation within engineering species are often overlooked. Combining field and greenhouse experiments with mathematical modelling, this study aimed to assess the relative importance of heritable and non-heritable trait variation within the engineer species Spartina alterniflora in controlling salt marsh erosion. In the field experiment, plots along a devegetated shoreline were restored with wild and cultivated sources to test whether populations exerted different control on erosion. The greenhouse experiment investigated whether genotypic trait differences were conserved when genotypes were exposed to elevated nutrients. A modelling approach was used to extrapolate empirical findings to temporal and spatial scales involved in landform evolution, considering spatial patterns in trait variation. The field experiment revealed that erosion rates were higher in plots planted with a wild, non-local source population as compared to plots planted with cultivars or local genotypes. Differential erosion could not be explained by differences in biomass, suggesting that other traits and resource use are stronger determinants of erosion. In the greenhouse experiment, cultivars and wild genotypes exhibited trait-specific differences in phenotypic plasticity under changing nutrient availability. Nutrient regime and heritable trait differences explained 70% of observed variation in soil shear strength. Soil shear strength increased when plants received more nutrients, but plant genotype had an equal or larger influence on soil characteristics. Model simulations suggested that older marshes (with large clones) and genetically diverse marshes (with high spatial variance in soil shear strength) may experience higher mean erosion rates. However, simulations also showed that average erosion rates are easily underestimated if the observation period is short, as variability of annual erosion rates and the probability of mass failure events were also mediated by clone size and composition. These findings illustrate that heritable and non-heritable trait variation interact with environmental conditions and landform history, together driving geomorphological processes crucial to the persistence of coastal marshes. Consideration of these interacting factors is needed when deploying ecosystem engineers for habitat restoration. / 1 / Brittany Marie Bernik

Abiotic and biotic variation has been shown to be important in regulating nutrient cycling and belowground communities in natural systems. However, genetic variation in dominant plants as a driver of rates of nutrient cycling is still poorly understood and few studies have looked at genotype interactions across multiple environments. Using Populus angustifolia and a common garden approach, we hypothesized that all three factors: tree genetic variation, environmental conditions and genetic by environment (G x E) interactions would affect soil carbon (C) storage and nitrogen (N) cycling. Replicated copies of five different reciprocally planted Populus genotypes were studied in three separate 18-21 year old common gardens at different elevations (1300m, 1384m and 1587m) in northern Utah, to measure the genotype and environmental effects on pools of soil C and N as well as rates of soil net N nitrification and net mineralization. Our results indicate that genotypes influence pools of soil C, total N and C:N, but genotype did not influence net rates of nitrogen mineralization. Environmental variation significantly influenced pools of soil C, total N, soil C:N and rates of net nitrification and net N mineralization. As predicted, G x E interactions significantly influenced both pools and processes of soil C and N cycling. Overall, we found that genetic variation in plant traits (tree diameter and leaf/root chemistry) as well as soil texture across gardens were significant predictors of soil C and N pools and fluxes across seasons. These data help us understand the relative role of genotypic variation on above- and belowground interactions in different environments and the consequences of these interactions on ecosystem processes. The results from this study show that across an environmental gradient Populus angustifolia genotypes can influence nitrogen mineralization through feedbacks between environmental variation, tree phenotype and soils.

Abiotic and biotic variation has been shown to be important in regulating nutrient cycling and belowground communities in natural systems. However, genetic variation in dominant plants as a driver of rates of nutrient cycling is still poorly understood and few studies have looked at genotype interactions across multiple environments. Using Populus angustifolia and a common garden approach, we hypothesized that all three factors: tree genetic variation, environmental conditions and genetic by environment (G x E) interactions would affect soil carbon (C) storage and nitrogen (N) cycling. Replicated copies of five different reciprocally planted Populus genotypes were studied in three separate 18-21 year old common gardens at different elevations (1300m, 1384m and 1587m) in northern Utah, to measure the genotype and environmental effects on pools of soil C and N as well as rates of soil net N nitrification and net mineralization. Our results indicate that genotypes influence pools of soil C, total N and C:N, but genotype did not influence net rates of nitrogen mineralization. Environmental variation significantly influenced pools of soil C, total N, soil C:N and rates of net nitrification and net N mineralization. As predicted, G x E interactions significantly influenced both pools and processes of soil C and N cycling. Overall, we found that genetic variation in plant traits (tree diameter and leaf/root chemistry) as well as soil texture across gardens were significant predictors of soil C and N pools and fluxes across seasons. These data help us understand the relative role of genotypic variation on above- and belowground interactions in different environments and the consequences of these interactions on ecosystem processes. The results from this study show that across an environmental gradient Populus angustifolia genotypes can influence nitrogen mineralization through feedbacks between environmental variation, tree phenotype and soils.

L’expression du phénotype des organismes dépend en partie d’organismes symbiotiques avec qui ils sont en interaction étroite. Selon le mode de transmission du symbiote, ce dernier va être en conflit d’intérêt plus ou moins intense avec l’hôte pour l’expression du phénotype, conduisant parfois le symbiote à évoluer vers la manipulation du phénotype de l’hôte. Nous avons tenté d’identifier différents facteurs génétiques et environnementaux influençant l’expression et l’évolution de la manipulation chez l’insecte parasitoïde de larves de drosophiles, Leptopilina boulardi, et son virus manipulateur du comportement, LbFV. Ce virus bénéficie d’une transmission mixte, verticale et horizontale, cette dernière étant favorisée par l’induction de superparasitisme induite par le virus. L’étude de la contribution du génotype du parasitoïde dans l’expression de la manipulation a révélé la présence de gènes de résistance partielle à la manipulation. Le potentiel évolutif de cette résistance a ensuite été évalué par des expériences d’évolution expérimentale. Nous avons également montré que LbFV augmente la virulence du parasitoïde envers les larves de drosophiles, révélant ainsi une évolution vers une forme de mutualisme sur ce trait. Par ailleurs, le travail montre qu’un même parasitoïde peut être non seulement infecté par plusieurs souches du virus LbFV mais également infecté par un virus à ARN, décrit pour la première fois dans cette thèse. La transmission verticale, la prévalence élevée et les forts effets phénotypiques de ce virus soulignent de nouveau l’importance des virus dans l’expression du phénotype en population naturelle. / The expression of the phenotype of an individual depends partly on the presence of symbiotic organisms with which it engages in intimate interactions. According to the symbiont’s mode of transmission, the conflict of interest between the host and its symbiont for the expression of the phenotype can be more or less stronger, sometimes leading to the evolution of manipulation of the host phenotype by the symbiont. We attempted to identify different genetic and environmental factors influencing the expression and the evolution of manipulation in the Drosophila parasitoid wasp, Leptopilina boulardi, and its behavior manipulating virus, LbFV. The virus undergoes both vertical and horizontal transmission, the latter being favoured by the induction of superparasitism behaviour by the virus. The study of the contribution of the parasitoid genotype to the expression of the manipulation revealed a polymorphism in the resistance to the manipulation. The evolutionary potential of this resistance was then investigated using a protocol of experimental evolution. We also demonstrated that LbFV increases the virulence of the parasitoid towards its Drosophila host, revealing a form of mutualistic interaction on this trait. Additionally, we showed that an individual parasitoid can be coinfected by several LbFV strains but also by an RNA virus, described for the first time in this thesis. The vertical transmission, the high prevalence and the strong phenotypic effects of this new virus further highlights the importance of viruses in the expression of the phenotype in natural populations.

Esta dissertação de mestrado aborda a controvérsia em torno de qual é o nível biológico no qual a seleção natural atua, com ênfase na proposta de Richard Dawkins do gene egoísta e nas questões que surgem em torno da mesma. Examina-se um panorama de questões de filosofia da biologia abordadas a partir do problema dos níveis nos quais a seleção natural atua. Esperamos que ao avaliar o impacto da teoria do gene egoísta na problemática evolutiva, consigamos compreender sua importância. O objetivo deste trabalho é filosófico, delineando as questões e clarificando alguns termos do debate, sem se propor a tomar partido por uma ou outra posição. O primeiro capítulo apresenta as origens históricas do debate, partindo do ponto de vista original de Charles Darwin no qual o indivíduo era a entidade efetivamente selecionada. Em seguida, buscamos entender como novas questões empíricas, em especial a busca de explicações biológicas para o altruísmo, conduziram a propostas de seleção de grupo. No segundo capítulo delineamos como o desenvolvimento da genética possibilitou que um novo nível de seleção fosse proposto: o gene, e acompanhamos a exposição de Dawkins sobre o ponto de vista do gene egoísta, em especial a partir de seus dois livros mais relevantes sobre o tema: O gene egoísta e O fenótipo estendido. O terceiro capítulo examina diversas aproximações filosóficas no contexto de resposta à pergunta: o que é uma unidade de seleção?. Nosso estudo é consistente com a tese de que as forças seletivas atuam simultaneamente em diversos níveis. / This Masters thesis studies the controversy over what is the biological level in which natural selection takes place. Emphasis is given to Richard Dawkins proposal of the selfish gene and to the issues that arise therefrom, which include many questions in the philosophy of biology. We hope that by assessing the impact that the theory of the selfish gene has had on the problems of evolution, one may understand its importance. The aim of this study is philosophical, raising questions and clarifying the terms of the debate, without taking side on one or another position. The first chapter presents the historical origins of the debate, starting with the original view of Charles Darwin that the individual is the entity that is effectively selected. We then set out to understand how new empirical problems, specifically the search for biological explanations for altruism, led to proposals of group selection. In the second chapter, we depict how the development of genetics allowed that a new level of selection be proposed: the gene. We analyze Dawkins exposition of the point of view of the selfish gene, especially in the two most important books on the subject: The selfish gene and The extended phenotype. The third chapter examines several philosophical approaches to the question what is a unit of selection?. Our study is consistent with the thesis that selective forces act simultaneously in different levels.

Esta dissertação de mestrado aborda a controvérsia em torno de qual é o nível biológico no qual a seleção natural atua, com ênfase na proposta de Richard Dawkins do gene egoísta e nas questões que surgem em torno da mesma. Examina-se um panorama de questões de filosofia da biologia abordadas a partir do problema dos níveis nos quais a seleção natural atua. Esperamos que ao avaliar o impacto da teoria do gene egoísta na problemática evolutiva, consigamos compreender sua importância. O objetivo deste trabalho é filosófico, delineando as questões e clarificando alguns termos do debate, sem se propor a tomar partido por uma ou outra posição. O primeiro capítulo apresenta as origens históricas do debate, partindo do ponto de vista original de Charles Darwin no qual o indivíduo era a entidade efetivamente selecionada. Em seguida, buscamos entender como novas questões empíricas, em especial a busca de explicações biológicas para o altruísmo, conduziram a propostas de seleção de grupo. No segundo capítulo delineamos como o desenvolvimento da genética possibilitou que um novo nível de seleção fosse proposto: o gene, e acompanhamos a exposição de Dawkins sobre o ponto de vista do gene egoísta, em especial a partir de seus dois livros mais relevantes sobre o tema: O gene egoísta e O fenótipo estendido. O terceiro capítulo examina diversas aproximações filosóficas no contexto de resposta à pergunta: o que é uma unidade de seleção?. Nosso estudo é consistente com a tese de que as forças seletivas atuam simultaneamente em diversos níveis. / This Masters thesis studies the controversy over what is the biological level in which natural selection takes place. Emphasis is given to Richard Dawkins proposal of the selfish gene and to the issues that arise therefrom, which include many questions in the philosophy of biology. We hope that by assessing the impact that the theory of the selfish gene has had on the problems of evolution, one may understand its importance. The aim of this study is philosophical, raising questions and clarifying the terms of the debate, without taking side on one or another position. The first chapter presents the historical origins of the debate, starting with the original view of Charles Darwin that the individual is the entity that is effectively selected. We then set out to understand how new empirical problems, specifically the search for biological explanations for altruism, led to proposals of group selection. In the second chapter, we depict how the development of genetics allowed that a new level of selection be proposed: the gene. We analyze Dawkins exposition of the point of view of the selfish gene, especially in the two most important books on the subject: The selfish gene and The extended phenotype. The third chapter examines several philosophical approaches to the question what is a unit of selection?. Our study is consistent with the thesis that selective forces act simultaneously in different levels.

Beaucoup de parasites à cycle complexe ont développé la capacité d’altérer le phénotype de leurs hôtes. Il est reconnu que ces modifications d’apparence ou de comportement sont responsables d’une augmentation de la probabilité de prédation de l’hôte intermédiaire par l’hôte définitif. Ce phénomène de manipulation parasitaire peut avoir de nombreuses conséquences à l’échelle de l’écosystème, modifiant les interactions entre les populations d’hôtes et bouleversant leur rôle écologique. Cependant, les parasites manipulateurs sont peu étudiés sur le plan écologique, et l’effet de l’environnement sur l’interaction entre ces parasites et leurs hôtes, notamment en termes de manipulation, est encore largement inconnu. Au cours de cette thèse, j’ai étudié l’effet de l’environnement sur l’interaction entre les gammares, des crustacés amphipodes de grande importance écologique dans les rivières, et leurs parasites acanthocéphales. Mes travaux ont montré que les conditions de ressources alimentaires et de température subies par les hôtes durant le développement des parasites influençaient plusieurs paramètres de l’infection mais n’affectaient pas la manipulation en termes d’utilisation de refuges. Toutefois, tandis que la géotaxie ne dépendait pas non plus de la température, l’impact des parasites sur la phototaxie des gammares était plus fort à haute température. De plus, mes travaux ont montré que l’impact conjoint de la température et des parasites pouvait modifier le rôle détritivore des gammares. Dans un contexte de changements globaux, ces travaux permettent de mieux comprendre l’importance de l’impact des parasites sur leurs hôtes et en prévoir les conséquences écologiques. / Many parasites with complex life cycle have developed the ability to alter the phenotype of their hosts. It is recognized that such changes in appearance and behavior are responsible for an increase in the probability of predation of their intermediate hosts by their definitive hosts. This phenomenon of parasite manipulation can have numerous consequences at the scale of the ecosystem, modifying the interactions between host populations and altering their ecological role. However, manipulative parasites received little attention from an ecological point of view. Thus, the effect of the environment on the interaction between these parasites and their hosts, in particular in terms of manipulation, is largely unknown. In this thesis, I studied the effect of the environment on the interaction between gammarids – ecologically important crustacean amphipods in rivers – and their acanthocephalan parasites. My experiments showed that the conditions of food resources and temperature experienced by gammarids during the development of their parasites influenced several infection parameters, but did not affect behavioral manipulation in terms of use of refuges. Nevertheless, while the geotaxis was not either effected by temperature, the impact of parasites on gammarids phototaxis was stronger at high temperature. Moreover, my studies showed that the cumulative effect of temperature and parasitism could alter the shredder role of gammarids. In a context of global changes, this work provides a better understanding of the importance of the impact of parasites on their hosts, allowing to make previsions on their subsequent ecological consequences.

In nest-building species, offspring survival and reproductive success of parental individuals are strongly influenced by nest location and quality. Thus, quantifying the influence of abiotic conditions on nest integrity is important to predict the effects that weather variability may have on offspring survival and parental reproductive success. Here we investigated how rainfall affects nest integrity and how nest integrity influences male attractiveness and nest tenure. Our study species was the harvestman Quindina limbata, in which males build cup-like mud nests on fallen logs and protect the eggs against predators and fungi infestation. Our dataset is based on 12 months of regular inspections of over 150 nests in a tropical rainforest from Costa Rica. We found that 43% of the nests were destroyed by rainfall. The drag force promoted by rainfall running on the log surface negatively affected nest integrity by decreasing the wall perimeter and the floor area. The intensity of fungi cover was not explained by nest position or by precipitation. No matter the body size of the owner males, nests with high integrity received more eggs than nests with low integrity. Curiously, nests with high values of fungi cover received more eggs. Finally, nest integrity and fungi cover did not affect nest tenure, but the probability of males abandoning their nests increased with time they did not receive eggs. Considering that intense rainfall occurs all year long in tropical forests, the best males can do to decrease the chances of nest destruction is to select protected places to build their nests. Protected sites may keep nest structure better preserved, improve offspring survival, attract more females, and ultimately increase male reproductive success / Em espécies que constroem ninhos, a sobrevivência da prole e o sucesso reprodutivo dos indivíduos parentais são fortemente influenciados pela localização e qualidade dos ninhos. Portanto, quantificar a influência das condições abióticas sobre a integridade dos ninhos é importante para prever os efeitos que a variabilidade climática pode ter na sobrevivência da prole e no sucesso reprodutivo dos pais. Neste estudo, investigamos como a chuva influencia a integridade dos ninhos e como a integridade dos ninhos influencia a atratividade masculina e o tempo de posse do ninho. Nossa espécie de estudo foi o opilião Quindina limbata, em que os machos constroem ninhos de barro em troncos caídos e protegem os ovos contra predadores e infestação por fungos. Nosso conjunto de dados é baseado em 12 meses de inspeções regulares de mais de 150 ninhos em uma floresta tropical na Costa Rica. Descobrimos que 43% dos ninhos foram destruídos pela chuva. A força de arrasto promovida pela chuva na superfície dos troncos caídos influenciou negativamente a integridade dos ninhos, diminuindo o perímetro da parede e a área do piso. A intensidade de cobertura de fungos não foi explicada pela posição do ninho ou pela precipitação. O tamanho corporal dos machos não influenciou o ganho de ovos, mas ninhos com alta integridade receberam mais ovos do que ninhos com baixa integridade. Curiosamente, ninhos com altos valores de cobertura de fungos receberam mais ovos. Por fim, a integridade dos ninhos e a cobertura de fungos não influenciaram o tempo de posse do ninho, mas a probabilidade de os machos abandonarem seus ninhos aumentou com o tempo em que não receberam ovos em seus ninhos. Considerando que temporais ocorrem durante todo o ano em florestas tropicais, os melhores machos devem selecionar locais protegidos para construção dos seus ninhos a fim de diminuir as chances de destruição promovida pela chuva. Locais protegidos podem manter a estrutura do ninho melhor preservada, melhorar a sobrevivência da prole, atrair mais fêmeas e, finalmente, aumentar o sucesso reprodutivo dos machos

In nest-building species, offspring survival and reproductive success of parental individuals are strongly influenced by nest location and quality. Thus, quantifying the influence of abiotic conditions on nest integrity is important to predict the effects that weather variability may have on offspring survival and parental reproductive success. Here we investigated how rainfall affects nest integrity and how nest integrity influences male attractiveness and nest tenure. Our study species was the harvestman Quindina limbata, in which males build cup-like mud nests on fallen logs and protect the eggs against predators and fungi infestation. Our dataset is based on 12 months of regular inspections of over 150 nests in a tropical rainforest from Costa Rica. We found that 43% of the nests were destroyed by rainfall. The drag force promoted by rainfall running on the log surface negatively affected nest integrity by decreasing the wall perimeter and the floor area. The intensity of fungi cover was not explained by nest position or by precipitation. No matter the body size of the owner males, nests with high integrity received more eggs than nests with low integrity. Curiously, nests with high values of fungi cover received more eggs. Finally, nest integrity and fungi cover did not affect nest tenure, but the probability of males abandoning their nests increased with time they did not receive eggs. Considering that intense rainfall occurs all year long in tropical forests, the best males can do to decrease the chances of nest destruction is to select protected places to build their nests. Protected sites may keep nest structure better preserved, improve offspring survival, attract more females, and ultimately increase male reproductive success / Em espécies que constroem ninhos, a sobrevivência da prole e o sucesso reprodutivo dos indivíduos parentais são fortemente influenciados pela localização e qualidade dos ninhos. Portanto, quantificar a influência das condições abióticas sobre a integridade dos ninhos é importante para prever os efeitos que a variabilidade climática pode ter na sobrevivência da prole e no sucesso reprodutivo dos pais. Neste estudo, investigamos como a chuva influencia a integridade dos ninhos e como a integridade dos ninhos influencia a atratividade masculina e o tempo de posse do ninho. Nossa espécie de estudo foi o opilião Quindina limbata, em que os machos constroem ninhos de barro em troncos caídos e protegem os ovos contra predadores e infestação por fungos. Nosso conjunto de dados é baseado em 12 meses de inspeções regulares de mais de 150 ninhos em uma floresta tropical na Costa Rica. Descobrimos que 43% dos ninhos foram destruídos pela chuva. A força de arrasto promovida pela chuva na superfície dos troncos caídos influenciou negativamente a integridade dos ninhos, diminuindo o perímetro da parede e a área do piso. A intensidade de cobertura de fungos não foi explicada pela posição do ninho ou pela precipitação. O tamanho corporal dos machos não influenciou o ganho de ovos, mas ninhos com alta integridade receberam mais ovos do que ninhos com baixa integridade. Curiosamente, ninhos com altos valores de cobertura de fungos receberam mais ovos. Por fim, a integridade dos ninhos e a cobertura de fungos não influenciaram o tempo de posse do ninho, mas a probabilidade de os machos abandonarem seus ninhos aumentou com o tempo em que não receberam ovos em seus ninhos. Considerando que temporais ocorrem durante todo o ano em florestas tropicais, os melhores machos devem selecionar locais protegidos para construção dos seus ninhos a fim de diminuir as chances de destruição promovida pela chuva. Locais protegidos podem manter a estrutura do ninho melhor preservada, melhorar a sobrevivência da prole, atrair mais fêmeas e, finalmente, aumentar o sucesso reprodutivo dos machos

Les insectes endophytophages, tels que les insectes foreurs de tiges, les galligènes et les mineurs de feuilles, vivent et se nourrissent à l’intérieur des végétaux. L'hypothèse de l'alimentation sélective stipule que ces organismes endophytes possèdent un avantage adaptatif par rapport aux ectophages en accédant aux tissus les plus nutritifs tout en évitant les principaux composés défensifs de la plante. Ce comportement d’alimentation sélective peut être également renforcé par une manipulation de la physiologie de la plante comme cela a été démontré chez les insectes galligènes mais également suggéré chez certains insectes mineurs. Ces derniers sont en effet capables d’induire un phénotype « îles vertes » qui se manifestent par la persistance de la photosynthèse au niveau de la zone minée à l'automne alors que le reste de la feuille entre en sénescence et jaunit. L’objectif de notre étude a été d’étudier, en conditions de terrain, les capacités de manipulation du végétal dans le système Malus domestica / Phyllonorycter blancardella. Cet insecte hautement spécialisé complète l’ensemble de son développement dans une zone restreinte d’une seule feuille. / Endophytophagous insects, such as stem-boring, gall-forming and leaf-mining insects, live within plant tissues and feed internally. The selective feeding hypothesis states that this life-style presumably provides adaptive advantages for the insect over other external-feeding modes by allowing access to most nutritional tissues while avoiding main plant defensive compounds. This selective feeding behavior can be reinforced by manipulating the plant physiology which has been clearly demonstrated in gallers but also suggested in leaf-miner insects due to the autumnal formation of “green islands” around mining caterpillars in yellow leaves. This study aimed to investigate, under field conditions, the ability of insects to manipulate their host-plant in the Malus domestica / Phyllonorycter blancardella biological system. This insect is highly specialized and entirely develops within a restricted area of a single leaf. We first characterized the plant-insect interface by describing larval mouthparts and leaf anatomy alterations resulting from the insect feeding activity.