The evolutionary processes that shape individual variation in continuous behavioural traits remain poorly understood. While the emerging discipline of animal personality is providing increasing evidence that consistent individual differences in behaviour have significant fitness consequences, cognitive traits are yet to be explored in the same manner. My general objective in this thesis was twofold. First, I aimed to examine the ecological significance and fitness consequences of the cognitive trait innovative problem solving-performance, using a population of great tits (Parus major). Second, I aimed to explore the mechanisms underlying the functional significance of ‘exploration behaviour’ a captive measure of the reactive- proactive personality axis, focusing specifically on foraging and risk-taking behaviour. This two-trait approach was expected to shed light on whether personality and cognition simultaneously influence fundamental behaviours. By carrying out behavioural assays on birds temporarily held in captivity, I showed that success at solving a food-motivated problem was repeatable within individuals, consistent between two different tasks and independent of exploration behaviour. Problem-solving performance was positively related to clutch size and fledgling number, established when birds were released back into the wild. Furthermore, when rearing offspring, solvers had shorter working day lengths than non-solvers and foraged over smaller ranges without compromising either provisioning quantity or quality. However, solver females were also more likely to desert their broods than non-solvers and consequently there was little evidence to suggest that directional selection acted on problem-solving performance. In comparison to non-solvers, solver males were also found to be poorer at competing for limited food resources during the winter. Together these findings suggest that costs and benefits are associated with problem-solving performance, which together may act to maintain variation in this trait. My thesis also provides some of the first evidence that exploration behaviour is related to both foraging and risk-taking behaviour in the wild. In comparison to relatively slow explorers, fast exploring males were better competitors at feeders during the winter. Relatively fast and slow explorers also differed in a number of foraging behaviours during offspring provisioning, although not always in the direction predicted from captive work on other populations. Finally, while exploration behaviour was positively correlated with risk- taking behaviour amongst breeding females during incubation, it did not predict nest desertion in response to the risk associated with being trapped by field workers. Collectively these findings suggest that personality measured in captivity has specific but not general power to predict behaviour in the wild. An important facet of evolution is that natural selection is rarely likely to act on a single trait in isolation owing to correlations between traits. This thesis demonstrates how important behaviours in the wild can be simultaneously linked to multiple sources of consistent behavioural variation. It also represents the first large-scale investigation of how variation in a cognitive trait relates to natural behaviour, reproductive life-history variation and fitness in wild animals. Using this individual-based approach in a natural setting may prove to be a useful tool for understanding how selection acts on cognitive traits.

About 70% of the plant species in the temperate zone are characterised by clonal growth, clonal species are also in majority in the Arctic and Subarctic where they affect the structure and composition of the vegetation. It is therefore of great importance to increase our knowledge about clonal plants and their growth and life histories. I have investigated how ramets of the stoloniferous plant Linnaea borealis are affected by the naturally occurring variation in environmental factors, such as: light, nutrient and water availability. Moreover, I examined the seed set and how supplemental hand pollination affects seed set in L. borealis, and also investigated the significance of the apical meristem for shoot population fitness. All studies were performed under field conditions in northern Sweden in a Subarctic environment and most are experimental. The results show that nutrient resorption from senescing leaves is not significantly affecting the growth and nutrient pools of the ramet. This implies that the growth of L. borealis ramets is not governed by micro-site resource availability. However, removal of light competition resulted in increased branching and number of lateral meristems produced, reduced growth, and decreased root:shoot ratio on a per ramet basis. Thus, ramets of L. borealis can efficiently exploit favourable light patches through plastic growth. Apical dominance exerts a significant effect on shoot population fitness and can be lost through rodent grazing. However, loss of apical dominance is dependent on the timing of grazing, if the apical meristem is removed early in the autumn the ramet can repair the loss until the next summer. If grazing occur during spring the dry weight and leaf area production is affected negatively. Seed production in L. borealis in the Abisko area varies between years and sites, and was unaffected by supplemental hand pollination treatment, implying that there is no lack of pollinator activity.

Salmonid fish such as brown trout (Salmo trutta) are known to adapt to local environments in large water systems such as in Lake Vänern. Since brown trout returns to spawn at the location of birth, they become reproductively isolated from populations spawning elsewhere. Thus, genetically-based physiological and ecological differences may develop between them. There is evidence that two of the stocks of Lake Vänern brown trout, Gullspångälven brown trout (G-trout) and Klarälven brown trout (K-trout) differ, with G-trout growing faster and becoming bigger than K-trout, but there is no study identifying why this is the case. The purpose of this study was therefore to investigate whether there were differences in three variables: metabolic rate (gill-beat frequency, referred to as ventilation rate), foraging rate (number of captured prey per unit time) and aggressive behavior (number of displays) between the two stocks of brown trout. By conducting laboratory experiments on G-trout and K-trout in artificial streams with a constant supply of food through drift, behavioral observations were made to quantify these variables. The results showed that G-trout had significantly higher gill-beat frequency and foraging rate than did K-trout. In terms of aggression there was no significant difference observed between the two stocks, though there was a trend for G-trout to show more display behavior. The differences in metabolic rate and foraging rate are consistent with the different growth rates reported for the two stocks. The differences between the stocks suggest that they have developed local adaptations and have maintained these after a number of years of being raised in hatcheries. The observed differences should be taken into account when it comes to management and restoration of populations adapted to local environments. / Laxfisk som till exempel öringen (Salmo trutta) är kända att anpassa sig till lokala miljöer i stora vattensystem, såsom i sjön Vänern. Eftersom öringen återvänder och leker där de är födda blir de reproduktivt isolerade från populationer som leker på annan plats. Därmed kan ärftliga fysiologiska och ekologiska skillnader utvecklas dem emellan. Där finns bevis på att Gullspångsöringen tillväxer snabbare och blir större än Klarälvsöringen men det finns ingen riktig vetskap om varför. Syftet med denna studie var därmed att undersöka om där fanns skillnader i tre variabler: metabolisk hastighet (gälslagsfrekvens), furagering (antal tagna byten) och aggressivt beteende (antal display uppvisanden) mellan öringstammarna. Med separata försök av G-öring och K-öring i ett försöksakvarium med en konstgjord vattenström och konstant tillförsel av mat genom drift, gjordes beteendeobservervationer av dessa variabler. Resultaten visade på att G-öring hade en signifikant högre gälslagsfrekvens och furageringshastighet än K-öring. Vad gällde displaybeteende fanns ingen signifikantskillnad mellan stammarna, dock observerades en trend att G-öringen uppvisade mer display. Dessa resultat tyder på att metabolisk hastighet och furagering kan ha en roll i tillväxttakten. Skillnaderna mellan stammarna kan tyda på att de har utvecklat lokala anpassningar och har kvar dessa efter ett antal års fiskodling. De observerade skillnaderna är viktiga att ta med i beräkning när det kommer till förvaltning och restauration av populationer anpassade till lokala miljöer.

The aim of this study is to provide a further insight into influence of environment on clonal plants. The study focuses particularly on effects of fertilization level and light availability on production and growth of clonal organs. Three experiments were carried out within the study, targeted to elicit influence of these environmental conditions or clonal interactions on six species of clonal plants. Interspecies dependencies on these conditions was compared, regarding habitat occurrence of these species. A comparison was made also between species producing rhizomes and stolons. The experiments revealed that five of six studied species show significant relationship among at least one environmental condition and parameters of clonal reproduction. Most of the species showed higher elongation and production of clonal organs in relation to fertilization level. On the contrary, only three species reacted significantly to the light availability level by alternation of at least one parameter of clonal reproduction and the light availability level affected each species differently. A strong influence on production and elongation of clonal organs had also a size of a plant. There was not found significant difference in influence of environmental conditions on clonal reproduction among plant families. It...

Dans un monde changeant et incertain, pouvoir moduler sa réponse phénotypique selon les conditions locales représente un avantage adaptatif majeur. Cette aptitude passe par la récolte d’informations, et plus précisément, dans le cadre de la plasticité du comportement, par la combinaison d’informations passées et présentes. Ceci est rendu possible par l’apprentissage et la mémoire, qui permettent aux animaux de stocker l’information en vue d’une utilisation ultérieure. Cette thèse a pour vocation de renforcer les liens entre écologie comportementale et cognition. Tout d’abord, nous nous sommes penchés sur deux questions propres à l’utilisation d’information passée. (i) Quelle importance est accordée aux informations anciennes face aux récentes ? (ii) En cas de redondance, toutes les informations récoltées sont-elles conservées ? Nous avons montré que (i) l’hyménoptère parasitoïde Venturia canescens accorde un poids égal à l’information issue de l’exploitation de parcelles successives d’hôtes. (ii) Parmi trois sources possibles d’information passée, une seule est utilisée. Dans un second temps, nous avons testé l’hypothèse selon laquelle apprentissage et mémoire ont évolué en réponse au milieu de vie. Nous avons mené une étude comparative sur des V. canescens arrhénotoques et thélytoques vivant dans des habitats contrastés. Les résultats obtenus lors d’un conditionnement olfactif concordent en partie avec nos prédictions, renforçant l’idée d’un lien entre distribution de la ressource et apprentissage. Du point de vue théorique, la question de la valeur adaptative de la durée des différentes phases de mémoire n’a jusqu’ici pas été traitée. Un modèle de simulation est proposé pour commencer à combler cette lacune / In an uncertain and changing world, being able to shape its phenotype according to the current environmental conditions should provide individuals with a major fitness benefit. This depends on information gathering. In the context of behavioural plasticity, it more specifically implies to combine past and current information pieces. Past information use happens through learning and memory, that enables animals to store information in memory for a subsequent decision making. This thesis aims at strengthen the links between behavioural ecology and cognition. First, we tackled two questions specific to past information use. (i) Which weight is given to an older information piece in front of a more recent one? (ii) If several redundant information pieces are gathered, are they all stored? We showed that (i) the parasitoid wasp Venturia canescens seems, at the time scaled we worked with, to weight equally information pieces coming from older and more recent host patches successively foraged. (ii) Among three possible past information sources, only one is stored for a later use. Second, we tested the hypothesis that learning and memory evolved in response to ecological constraints. We conducted a comparative study with arrhenotokous and thelytokous V. canescens thriving in different habitats. Results obtained through olfactory conditioning partially matched our initial predictions; this support the idea of a link between resource distribution and learning features. From the theoretical point of view, the question of the adaptive significance of the length of memory phases has not been tackled yet. Here is shown a model, as a first attempt to fill this gap

As ectothermic marine megafauna, sea turtle physiology and ecology are tightly intertwined with temperature, seasonality, and oceanography. Identifying how turtles respond when exposed to cold water, how they adapt to cold environments when they need to explore cold environments in order to forage, and what foraging resources are exploited by sea turtles are all components central to their conservation. Cold-stunning is a well-documented phenomenon that occurs when water induced decreases in sea turtle body temperature cause turtles to become immobilized and wash ashore. While most cold-stunned turtles are rescued and rehabilitated, we do not know whether cold-stunning is an acute transient occurrence, or a symptom of a bigger environmental problem. Further, while in some environments avoiding cold water is preferential, in other habitats, sea turtles need to inhabit cold environments in order to forage. Along the Eastern Pacific Rim, discrete upwelling locations are characterized by high primary productivity and unusually cold water. In these environments, avoidance is not possible and sea turtles require physiological adaptions to mitigate body temperature decreases in cold water. Little is known about how turtles handle upwelling environments, despite the fact that sea turtles remain in these habitats regardless of water temperature fluctuations. Because upwelling habitats provide increased nutrient presence, and sea turtles are opportunistic foragers, quantification of diet composition will further our understanding of why sea turtles remain in cold water environments year-round. Diet composition in multiple populations of cohabitating sea turtles revealed partitioning that results in reduced inter-specific competition. Further, flexibility in diets provides a wide range of ecosystem services central to habitat resiliency. Therefore, conservation of endangered sea turtles requires complete ecosystem conservation, and complete understanding of the interconnectivity of sea turtles and their environments is crucial.<br>

Dans l’océan, les bancs de proies représentent une source d’alimentation ponctuelle. Pour pallier à cela, les oiseaux marins profitent de l’information sociale provenant de conspécifiques, hétérospecifiques, ou bateaux de pêche. Dans cette thèse, nous cherchons à évaluer l’influence de différentes sources d’information sur les stratégies de recherche alimentaire des oiseaux marins en milieu pélagique tropical.Les oiseaux détectent des prédateurs ou des bateaux à des distances dépassant dix kilomètres. Nous étudions donc leur distribution à sub-méso-échelle, ceci grâce à des images d’un radar à bord d’un thonier senneur.Les images radar offrent une visualisation, dans les 30 km autour du navire, de la distribution instantanée de la communauté d’oiseaux en vol et de leurs déplacements. Nous mettons au jour des motifs d’agrégation temporaires, d’une dizaine de kilomètres de diamètre, au sein de la distribution des oiseaux. Les distances entre groupes d’oiseaux sont compatibles avec l’échange d’information au sein de ces agrégations.Une étude plus fine révèle des vols coordonnés de groupes d’oiseaux distants de plusieurs centaines de mètres («râteaux»), suggérant un haut degré de coopération lors de la recherche de proies. Nous questionnons ensuite le degré de perturbation du senneur dans la recherche alimentaire des oiseaux. Les résultats indiquent que les oiseaux marins discriminent les différentes activités du navire.Ce travail constitue, à notre connaissance, la première utilisation des radars embarqués pour l’étude de l’écologie des oiseaux en haute mer, et ouvre nombre de perspectives pour comprendre la dynamique des interactions entre prédateurs marins supérieurs. / In the ocean, prey is patchily distributed. To overcome this challenge, pelagic seabirds benefit from social information from conspecifics, heterospecifics, or fishing boats.As part of this dissertation, we aim at evaluating the influence of different information sources in seabird foraging strategies in the tropical pelagic environment.Seabirds can detect visible predators or boats at distances of over ten kilometres. We thus study the distribution of seabirds at sub-meso-scale, analysing images recorded from a radar on board a tuna purse-seiner. Radar images provide a visualisation of the instantaneous distribution of the in-flight seabird community as well as seabird movements within thirty kilometres of the seiner. We detect over-aggregatedand temporary patterns, spanning about ten kilometers, within flying seabirds’ distribution. Distancesbetween seabird groups are compatible with information exchanges within these aggregations. A finer scale study reveals coordinated flights of seabird groups distant of hundreds of meters from each others (« rakes »), and suggesting a high level of coordination during foraging. We then investigate the potential disturbance induced by the seiner in the seabirds’ foraging network.First results indicate that seabirds can discriminate the different seiner’s activities.To our knowledge, this work is the first contribution of on board radars use for the study of seabird ecology in the high seas, and opens interesting perspectives, such as the understanding of interactions’ dynamics within the marine top-predators guild.

Grandes mamíferos herbívoros exercem papel fundamental para a função e estrutura dos ecossistemas terrestres, através principalmente da herbivoria e da deposição de fezes e urina, que influenciam a germinação e crescimento de plantas, a composição e diversidade de comunidades vegetais, a ciclagem e translocação de nutrientes, o estoque de carbono e a freqüência de distúrbios como o fogo. Entretanto, em florestas tropicais, o papel de grandes mamíferos herbívoros foi pouco estudado até o momento. De acordo com a Teoria de Forrageio Ótimo, a seleção de recursos alimentares é resultado do balanço entre o ganho energético e o gasto com a procura, captura e ingestão do alimento. Na escala da paisagem, um dos processos que pode gerar manchas com diferentes qualidades nutricionais para herbívoros é a sucessão da vegetação. Estádios iniciais de sucessão, por serem dominados por espécies de plantas de baixa estatura e crescimento rápido, com folhas tenras, alto conteúdo de nitrogênio, poucos compostos secundários e mais palatáveis, deveriam ser preferidos por grandes mamíferos herbívoros. No entanto, para herbívoros podadores, que incluem frutos na dieta, a disponibilidade destes itens, que são mais nutritivos que a folhagem, deve também influenciar o forrageio. Neste trabalho investigamos como varia a intensidade de forrageio e a deposição de fezes por Tapirus terrestris entre estádios sucessionais, visando contribuir para o entendimento do papel do maior herbívoro terrestre sul-americano para a regeneração da floresta e a translocação de nutrientes. Investigamos se T. Terrestris forrageia mais intensamente em manchas de vegetação em estádios mais iniciais de sucessão, se esta preferência é mais forte quando não há a oferta de um recurso alimentar muito nutritivo, os frutos de cambuci (Campomanesia phaea), e se deposita maior quantidade de fezes onde forrageia mais intensamente e assim não transloca nutrientes. Para tanto, em uma paisagem de 20.000 ha de Floresta Atlântica contínua, foram alocados 12 sítios de amostragem em manchas com diferentes proporções de vegetação em estádios mais iniciais de sucessão. Em cada sítio, a amostragem foi realizada em quatro transecções perpendiculares de 160 m cada durante seis sessões entre março e agosto de 2011. A anta foi registrada através de armadilhas fotográficas e de rastros e fezes localizados e removidos no início e no final de cada sessão. Foram também quantificados a proporção das transecções ocupada por estádios iniciais de sucessão (estádios pioneiro, inicial e médio), o número de clareiras, e o número de pontos das transecções onde a densidade do sub-bosque, do estrato herbáceo e de árvores com DAP até 5 cm foi considerada alta ou muito alta. A frutificação do cambuci foi acompanhada ao longo do período de estudo, tendo sido registrada em metade das sessões de amostragem. Através da abordagem de seleção de modelos, comparamos um conjunto de modelos candidatos para três variáveis dependentes: (a) parâmetro abundância (&lambda;, interpretado como a intensidade de forrageio ao longo das sessões de amostragem) em modelos de abundância (que assumem que não há variações temporais na abundância), considerando apenas modelos simples com cada uma das variáveis explanatórias associadas à proporção de vegetação em estádios mais iniciais de sucessão; (b) número de setores com rastros (interpretado como a intensidade de forrageio dentro das sessões de amostragem) em modelos mistos (GLMMs), considerando estes mesmos modelos simples e mais um grupo de modelos compostos que incluem uma variável temporal relacionada à frutificação do cambuci; e (c) número total de fezes em modelos lineares (GLMs), considerando todos os modelos descritos anteriormente e um modelo com o número total de setores com rastros (interpretado como a intensidade de forrageio total). Nossos resultados indicam que, como esperado pela Teoria de Forrageio Ótimo, Tapirus terrestris seleciona manchas de vegetação em estádios mais iniciais de sucessão, particularmente áreas com maior densidade de árvores pequenas e com maior número de clareiras, e especialmente quando não há oferta de frutos de cambuci. Embora o número total de fezes seja melhor explicado pelo número total de setores com rastros, a relação do número total de fezes com a proporção de estádios iniciais tende a ser negativa, sugerindo que possa haver alguma translocação de nutrientes dos estádios jovens para os tardios. Embora inferências sobre o efeito de T. terrestris sobre a comunidade de plantas dependam de estudos complementares em escalas menores, nossos resultados sugerem que: a herbivoria afeta estádios iniciais de sucessão, onde pode resultar no aumento da diversidade de plantas; antas podem ser agentes de translocação de nutrientes de estádios ricos (iniciais) para aqueles onde nutrientes tendem a ser limitantes (tardios), e a espécie é um potencial agente dispersor do cambuci, espécie ameaçada de extinção / Large herbivorous mammals play a crucial role to the function and structure of terrestrial ecosystems, mainly through herbivory and deposition of dung and urine, which influence plant germination and growth, composition and diversity of plant communities, nutrient cycling and translocation, carbon storage and the frequency of disturbances such as fire. So far, however, the role of large herbivorous mammals in tropical forests has been poorly studied. According to the Optimal Foraging Theory, selection of food resources results from the balance between energy intake and costs to search, capture and ingest the food. At the landscape scale, a process that can generate patches with different nutritional quality for herbivores is vegetation succession. Early successional stages should be preferred by large herbivorous mammals because they are dominated by plant with low height and fast growth, with leaves that are tender, present high nitrogen content and few secondary compounds, and are more palatable. However, for herbivore browsers, which include fruits in their diet, the availability of these items that are more nutritious than foliage should also affect foraging. Here, we investigate how foraging intensity and dung deposition by Tapirus terrestris vary among successional stages, aiming at contributing to the understanding of the role of the largest South American terrestrial herbivore to forest regeneration and nutrient translocation. We investigated if T. terrestris forages more intensively in earlier successional stages, if this preference is stronger when there is no availability of a highly nutritious food resource, the fruits of cambuci (Campomanesia phaea), and if it deposits larger amounts of dung where it forages more intensively, not translocating nutrients. In a 20,000 ha landscape of continuous Atlantic Forest, we allocated 12 sampling sites in patches with different proportion of vegetation in early successional stages. At each site, samplings were conducted in four 160 m long perpendicular transects during six sessions between March and August 2011. Tapir were registered by camera traps and tracks and dung located and cleared at the beginning and end of each session. We also quantified the proportion of transects occupied by early successional stages (pioneer, initial, and mid stage), the number of gaps, and the number of points in transects where the density of the understory , of herbaceous vegetation and of trees with DBH up to 5cm was considered high or very high. The fruiting of cambuci was registered throughout the study period, and was recorded in half of the sampling sessions. Using a model selection approach, we compared a set of candidate models for three dependent variables: (a) the parameter abundance (&lambda;, interpreted as foraging intensity along sampling sessions) in abundance models (which assume no temporal variations in abundance), considering only simple models containing each of the explanatory variables associated with the proportion of earlier successional stages, (b) the number of transect sectors with tracks (interpreted as foraging intensity within sampling sessions) in mixed-effects models (GLMMs), considering these same simple models and a group of composite models that included a temporal variable related to cambuci frutification, and (c) the total number of dung piles in linear models (GLMs), considering all models described above and a model with the total number of transect sectors with tracks (interpreted as total foraging intensity). Our results indicate that, as expected by the Optimal Foraging Theory, Tapirus terrestris selects patches of vegetation in earlier successional stages, in particular areas with higher density of small trees and higher number of gaps, and especially when there is no availability of cambuci fruits. Although the total number of dung piles is better explained by the total number of transect sectors with tracks, the relationship between the number of dung piles and the proportion of earlier successional stages tends to be negative, suggesting that there may be translocation of nutrients from initial to later successional stages. Although conclusions on the effect of T. terrestris on plant communities depend on complementary studies at smaller scales, our results suggest that: herbivory affects early successional stages, where it can result in increased plant diversity; tapirs may act as agents of nutrient translocation from nutrient-rich stages (earlier) to those where nutrients tend to be limiting (later); and tapirs are a potential disperser of cambuci, an endangered plant species

Les écosystèmes forestiers naturels du Sud-Ouest de l’océan Indien (SOOI) sont fortement impactés par la déforestation et font face à de nombreuses invasions biologiques pouvant altérer leur processus, en particulier les interactions de type plantes-pollinisateur. C’est dans ce contexte que s’inscrit cette thèse, décrivant les relations entre un pollinisateur généraliste indigène : Apis mellifera unicolor et la flore de deux écosystèmes de Ranomafana : RA et de Mare Longue : ML. La description palynologique de 135 espèces issues de 52 familles de plantes mellifères de la formation de RA, a fait ressortir les caractères spécifiques des pollens liés à une pollinisation entomophile. Dans un second temps, des suivis phénologiques mensuels de 131 espèces (90% d’indigènes) à RA et 120 espèces (53% d’exotiques) à ML ont permis d’identifier et d’estimer les ressources florales disponibles. Pendant un an, l’analyse mensuelle des miels et pollens collectés a permis de dresser l’inventaire des ressources florales réellement exploitées. Les espèces indigènes ont été significativement plus butinées que les espèces exotiques, malgré une diversité de ressources exotiques supérieures à celles des indigènes (ML). Le comportement de butinage de l’abeille sur le genre Weinmannia a été analysé sur la base de 104 h de vidéo (W. bojeriana et W. rutenbergii à Madagascar, et W. tinctoria à La Réunion). Les fleurs ont été visitées par un cortège de pollinisateurs potentiels : coléoptères, diptères, lépidoptères et autres hyménoptères. A. m. unicolor était le visiteur le plus fréquent à Madagascar. L’ensemble de ces résultats a permis de confirmer le comportement généraliste d’A. m. unicolor, avec toutefois une préférence forte et significative pour les espèces florales indigènes des strates arborées et arbustives et permet d’émettre des hypothèses sur la complexité des interactions entre l’abeille et les écosystèmes du hotspot de biodiversité SOOI. / The natural forest ecosystems of the Southwest Indian Ocean (SOOI) have been strongly impacted by deforestation and face many biological invasions that are alter their balance, particularly plant-pollinator interactions. It is in this context that this thesis is set, describing the relations between an indigenous general pollinator: Apis mellifera unicolor and the flora of two ecosystems of Madagascar (Ranomafana: RA) and Reunion (Mare Longue: ML). The palynological description of 135 species from 52 families of melliferous plants from the RA formation, highlighted the specific criteria of pollens associated with entomophilous pollination. In a second time, monthly phenological monitoring of 131 species (90% native) to RA and 120 species (53% exotic) to ML allowed to identify and estimate the available floral resources. During one year, the monthly analysis of honeys and pollen collected enabled an inventory of the floral resources actually exploited. Native species were significantly more visited than exotic species, despite a diversity of exotic resources superior to that of the natives (ML). The foraging behaviour of the honeybee on the genus Weinmannia was analysed on the basis of 104 hours of video (W. bojeriana and W. rutenbergii in Madagascar, and W. tinctoria in Reunion Island). The flowers were visited by many potential pollinators: Coleoptera, Diptera, Lepidoptera and other Hymenoptera. A. m. unicolor was the most frequent visitor to the two species from Madagascar. The results confirmed the generalist behaviour of A. m. unicolor, with, however, a strong and significant preference for native floral species of tree and shrub strata and makes it possible to speculate on the important place of this bee in the ecosystems of the SOOI biodiversity hotspot.

Zahlreiche Forschungsarbeiten haben gezeigt, dass nicht nur Menschen, sondern auch Tiere konstante individuelle Unterschiede im Verhalten aufweisen. Zu verstehen warum sich diese Verhaltensunterschiede im Laufe der Evolution entwickelt haben, ist ein Ziel dieses Forschungsbereiches. In dieser Arbeit wurde untersucht wie verschiedene Modulatoren das Nahrungssuchverhalten von Blütenfledermäusen (Glossophaginae) beeinflussen um individuelle Verhaltensunterschiede zu quantifizieren und theoretische Vorhersagen zu testen. Alle Experimente wurden in naturnaher Umgebung mit programmierbaren, künstlichen Blüten durchgeführt. Es wird angenommen, dass die Plastizität von Verhalten ein generelles Merkmal ist in dem sich Tiere unterscheiden, da manche Individuen allgemein stärker auf Reize aus der Umwelt reagieren könnten als andere. Um diese Vorhersage zu testen, wurde die Nahrungsverfügbarkeit experimentell manipuliert und zwei Arten von Verhaltensplastizität in denselben Individuen gemessen. Die Ergebnisse unterstützen diese Annahme jedoch nicht, da die beiden Arten von Verhaltensplastizität nicht korrelieren. Neben Umwelteinflüssen können auch innere Merkmale wie die Stoffwechselrate das individuelle Nahrungssuchverhalten beeinflussen. Die Ergebnisse zeigen, dass sich die Individuen in ihrem täglichen Energieverbrauch unterscheiden und dass diese Unterschiede mit dem Explorationsverhalten während der Nahrungssuche korrelieren. Zusätzlich kann das individuelle Nahrungssuchverhalten auch von sozialen Faktoren beeinflusst werden. Bei limitierter Nahrungsverfügbarkeit verteidigen einzelne Männchen Blüten gegen andere Männchen, jedoch nicht gegenüber Weibchen. Individuelle Unterschiede in der Aggression und Aktivität werden dagegen nicht von der sozialen Gruppenzusammensetzung beeinflusst. In dieser Arbeit wurden nicht nur individuelle Unterschiede im Nahrungssuchverhalten von Blütenfledermäusen bestimmt, sondern auch Vorhersagen aus dem Bereich der Persönlichkeitsforschung bei Tieren überprüft. / Animal personality research has shown that animals express individual differences in their behavior that are consistent over time and/or across situations. Furthermore, animal personality research aims to understand how these individual differences in behavior evolve and how they are maintained within populations. This thesis focuses on how different modulators influence the foraging behavior of nectar-feeding bats (Glossophaginae) in order to investigate consistent individual differences in their behavior and to test predictions proposed in the field of animal personality. All experiments were conducted in a semi-natural environment by using a setup of computer-controlled artificial flowers. One prediction of animal personality research is that behavioral plasticity is a single trait in which individuals differ because some individuals might be generally more responsive to changes in the environment than others. In order to test this prediction, resource availability was manipulated and two types of behavioral plasticity were quantified within the same individual. However, the two types of plasticity did not correlate and therefore this hypothesis was not supported. Individual foraging behavior can not only be modulated by changes in the environment but also by individual differences in internal traits like metabolic rates. Experiments confirmed that individual differences in daily energy expenditure correlated with consistent individual differences in exploration. Additionally, foraging behavior can also be modulated by social factors. When resources are limited some males started to defend flowers against other males but not against females. Furthermore, it could be shown that individual differences in aggression and activity are internally driven and not influenced by changes in the social group composition. This thesis not only assessed individual differences in the foraging behavior of nectar-feeding bats but also investigated predictions proposed in the field of animal personality research.