Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Thermal adaptation along a latitudinal gradient in damselflies Viktor Nilsson Akademisk avhandling som med vederbörligt tillstånd av Rektor vid Umeå universitet för avläggande av filosofie doktorsexamen framläggs till offentligt försvar i sal N450, Naturvetarhuset. Fredagen den 18 januari, kl. 13:00. Avhandlingen kommer att försvaras på engelska. Fakultetsopponent: Docent Karl Gotthard, Zoologiska institutionen: Ekologi, Stockholms Universitet, Sverige. Ecology and Environmental Science Umeå University Umeå 2012 Organization Document type Umeå University Doctoral thesis Ecology and Environmental Science Date of publication 14 December 2012 Author Viktor Nilsson Title Thermal adaptation along a latitudinal gradient in damselflies Abstract Understanding how temperature affects biological systems is a central question in ecology and evolutionary biology. Anthropogenic climate change adds urgency to this topic, as the demise or success of species under climate change is expected to depend on how temperature affects important aspects of organismal performance, such as growth, development, survival and reproduction. Rates of biological processes generally increase with increasing temperature up to some maximal temperature. Variation in the slope of the initial, rising phase has attracted considerable interest and forms the focus of this thesis. I explore variation in growth rate-temperature relationships over several levels of biological organization, both between and within species, over individuals’ lifetime, depending on the ecological context and in relation to important life history characteristics such as generation length and winter dormancy. Specifically, I examine how a clade of temperate damselflies have adapted to their thermal environment along a 3,600 km long latitudinal transect spanning from Southern Spain to Northern Sweden. For each of six species, I sampled populations from close to the northern and southern range margin, as well from the center of the latitudinal range. I reared larvae in the laboratory at several temperatures in order to measure individual growth rates. Very few studies of thermal adaptation have employed such an extensive sampling approach, and my finding reveal variation in temperature responses at several levels of organization. Keywords Growth rate, metabolic theory of ecology, universal temperature dependence, environmental gradients, thermal performance, thermal sensitivity, environmental variability, optimality theory, life history, acclimation, size structure, competition, cannibalism, intraguild predation Language English ISBN 978-91-7459-529-1 Number of pages 35 + 5 papers Thermal adaptation along a latitudinal gradient in damselflies Viktor Nilsson Department of Ecology and Environmental Science Umeå 2012 Copyright © Viktor Nilsson ISBN: 978-91-7459-529-1 Cover image: Coenagrion puella from August Johann Rösel von Rosenhof’s Der Insecten Belustigung, Zweiter Theil (1749). Distributed under a CC-BY 2.0 licence by Pecay @ Bibliodyssey. Elektronisk version tillgänglig på http://umu.diva-portal.org/ Printed by: VMC, Umeå University Umeå, Sweden 2012 “You behold there a considerable number of a remarkable species of beautiful insects, the elegance of whose appearance, and their attire, have procured for them the name of damsels” Sonnini de Manoncourt, Travels to Upper and Lower Egypt (1799) LIST OF PAPERS This thesis is a summary and discussion of the following papers, referred to in the text by the roman numerals. I. Nilsson-Örtman, V., R. Stoks, M. De Block and F. Johansson. 2012. Generalists and specialists along a latitudinal transect: patterns of thermal adaptation in six species of damselflies. Ecology 93: 1340-1352. II. Nilsson-Örtman, V., R. Stoks, M. De Block, H. Johansson and F. Johansson. In press. Latitudinally structured variation in the temperature dependence of damselfly growth rates. Ecology Letters III. Johansson, H., R. Stoks; V. Nilsson-Örtman, P. Ingvarsson and F. Johansson. In press. Large-scale patterns in genetic variation, gene flow and differentiation in five species of European Coenagrionid damselfly provide mixed support for the central-marginal hypothesis. Ecography. IV. Nilsson-Örtman, V., R. Stoks, M. De Block and F. Johansson. Latitudinal patterns of phenology and age-specific thermal performance across six Coenagrion damselfly species (Odonata). Submitted manuscript V. Nilsson-Örtman, V., R. Stoks and F. Johansson. Competition modifies the temperature dependence of damselfly growth rates. Manuscript Papers I, II and III are reproduced with kind permission from the respective publishers. TABLE OF CONTENTS INTRODUCTION 1 Background 1 Adapting to thermally heterogeneous environments 1 Studying thermal adaptation 2 The Universal Temperature Dependence hypothesis 4 Context-dependent thermal performance 4 Thermal performance and life history theory 5 The evolutionary ecology of thermal adaptation 6 Aims 7 Research questions STUDY SYSTEM METHODS Experiment 1: Thermal sensitivity of growth rates across latitudes Experiment 2: Effects of competition and temperature on growth rates PAPER I: THERMAL BREADTH DECREASE WITH LATITUDE 7 8 11 11 11 12 Model 1: lake temperature across latitudes 13 Model 2: larval life cycles and temperature variation 13 Results and discussion PAPER II & III: GROWTH RATES DO NOT SCALE WITH METABOLISM Results and discussion PAPER IV: THERMAL PERFORMANCE SHIFTS DURING ONTOGENY Results and discussion 14 15 16 17 18 PAPER V: COMPETITION AND TEMPERATURE MODIFIES GROWTH RATES 19 Results and discussion 20 CONCLUDING REMARKS 20 ACKNOWLEDGEMENTS 22 AUTHOR CONTRIBUTIONS 23 Paper I 23 Paper II 23 Paper III 23 Paper IV 23 Paper V 23 REFERENCES 24 THANKS 31 TACK 32 ABSTRACT Understanding how temperature affects biological systems is a central question in ecology and evolutionary biology. Anthropogenic climate change adds urgency to this topic, as the demise or success of species under climate change is expected to depend on how temperature affects important aspects of organismal performance, such as growth, development, survival and reproduction. Rates of biological processes generally increase with increasing temperature up to some maximal temperature. Variation in the slope of the initial, rising phase has attracted considerable interest and forms the focus of this thesis. I explore variation in growth rate-temperature relationships over several levels of biological organization, both between and within species, over individuals’ lifetime, depending on the ecological context and in relation to important life history characteristics such as generation length and winter dormancy. Specifically, I examine how a clade of temperate damselflies have adapted to their thermal environment along a 3,600 km long latitudinal transect spanning from Southern Spain to Northern Sweden. For each of six species, I sampled populations from close to the northern and southern range margin, as well from the center of the latitudinal range. I reared larvae in the laboratory at several temperatures in order to measure indiviudal growth rates. Very few studies of thermal adaptation have employed such an extensive sampling approach, and my finding reveal variation in temperature responses at several levels of organization. My main finding was that temperature responses became steeper with increasing latitude, both between species but also between latitudinal populations of the same species. Additional genetic studies revealed that this trend was maintained despite strong gene flow. I highlight the need to use more refined characterizations of latitudinal temperature clines in order to explain these findings. I also show that species differ in their ability to acclimate to novel conditions during ontogeny, and propose that this may reflect a cost-benefit trade-off driven by whether seasonal transitions occur rapidly or gradually during ontogeny. I also carried out a microcosm experiment, where two of the six species were reared either separately or together, to determine the interacting effects of temperature and competition on larval growth rates and population size structure. The results revealed that the effects of competition can be strong enough to completely overcome the rate-depressing effects of low temperatures. I also found that competition had stronger effects on the amount of variation in growth rates than on the average value. In summary, my thesis offers several novel insights into how temperature affects biological systems, from individuals to populations and across species’ ranges. I also show how it is possible to refine our hypotheses about thermal adaptation by considering the interacting effects of ecology, life history and environmental variation. INTRODUCTION Background Temperature not only dictates the limits for life, it sets the pace. For most organisms, the biochemical processes that are necessary for their existence can only occur within a narrow range of temperatures, roughly between 0°C and 45°C (Pörtner 2002). Within this temperature range, the rates of virtually all biological processes, from enzyme kinetics to species interactions and populations growth rates, depend strongly on temperature (Angilletta 2009; Daniel & Danson 2010). With the exception of a small number of model organisms kept in laboratories, very few organisms live in a world of constant temperatures. Rather, temperature varies considerably between geographic regions, over the seasons, throughout the course of the day and between sun and shade (Bradshaw et al. 2004). Unlike processes like predation and competition, that only exist when individuals interact with other individuals, temperature is an inescapable physiological reality that acts directly on each molecule, enzyme and cell within an organism (Hochachka & Somero 2002; Angilletta 2009). But temperature does not affect all organisms in the same way. Temperatures that would cause the sudden death of an Antarctic brachiopod (Peck 2002) can represent the optimum temperature for activity in a Sahara ant (Gehring & Wehner 1995). Thus, temperature not only affects organisms – organisms adapt to temperature. Adapting to thermally heterogeneous environments Organisms have evolved a multitude of strategies that allows them to cope with temperature heterogeneity. A striking example is thermoregulation, the ability of an organism to maintain a constant body temperature over a range of environmental temperatures. Famous examples include the basking behavior of lizards and snakes and the tightly controlled internal body temperature of mammals and birds. Thermoregulatory strategies, however, are available to a small number of organisms. These mainly include large, terrestrial vertebrates (Huey & Slatkin 1976; Chappell 1983). For the vast majority of species on earth, body temperature is highly or completely dependent on ambient temperatures. This is especially true for small-bodied (Unwin & Corbet 2008) and aquatic organisms, due to the high specific heat and thermal conductivity of water (Feder & Hofmann 1999). It is therefore not surprising that ectotherms display a range of adaptations that allows them to move, feed, grow and develop over a wide 1 range of body temperatures. Adaptation to thermal heterogeneity frequently involves changes in the degree to which an organism’s performance depends on changes in body temperature. This is referred to as the thermal sensitivity or dependence of performance. Important performance traits include growth, development and locomotion. Thus, some organisms are able to maintain relatively constant levels of performance over a wide range of temperatures. These are referred to as thermal generalists. Others can only tolerate a very narrow range of temperatures, with even small changes in temperature causing major changes in performance. These are referred to as thermal specialists (Angilletta 2009). Studying thermal adaptation Thermal performance curves (TPCs) forms an unifying, conceptual framework for studying differences in the thermal sensitivity of performance among organisms (Fig 1). TPCs describe the relationship between a temperature and a continuous phenotypic character (e.g. growth, development, or activity). These curves represent a special type of continuous reaction norms, and the slope of TPCs represents a measure of plasticity. TPCs of many traits have a distinct shape: a more or less steeply increasing “rise” phase, a single optimum, and a sharply declining “fall” phase (Figure 1; a,b,c). Six characteristics of performance curves has attracted considerable interest: 1) the optimal temperature (Topt) – the body temperature at which performance is maximized; 2) the maximum performance (Pmax) – the height of the curve at the optimal temperature; 3) the thermal breadth –the temperature range over which an individual can maintain an arbitrary level of performance, for example 80% of Pmax; 4) the amount of thermal plasticity of performance – the slope of the rising and falling parts of TPCs (a and c in Figure 1); 5) critical thermal thresholds – the lower (CTmin) and upper (CTmax) body temperatures at which performance is zero; and 6) the tolerance range – the difference between CTmin and CTmax. The complex shape of TPCs reflects the contribution of several underlying mechanisms. There is, however, little consensus about what mechanisms give rise to these different aspects of thermal performance curves. In fact, different processes may be important for TPCs at different hierarchical levels, in different organisms and for different aspects of organism performance (Pörtner 2002). For example, lower critical thresholds may reflect temperature-induced changes in the structure of cellular membranes (Hochachka & Somero 2002) or the ability of mitochondria to generate ATP (Pörtner et al. 2006). Upper critical limits have been suggested to reflect the denaturation of important proteins (Hochachka & Somero 2002), or limits 2 Figure 1. Thermal Performance Curve (TPC) showing important characteristics, including the lower (CTmin) and upper (CTmax) critical thermal thresholds, a single thermal optimum (Topt), maximal performance (Pmax) and performance breadth (defned as the range of temperatures over which a given fraction. e.g. 80% of Pmax, is maintained. In the text I also refer to the rising (a), plateau (b) and falling (c) phases. Note that the amount of plasticity (i.e. the slope of the curve) is great during the rising and falling phases, but zero during the plateau phase. to the ability to supply mitochondria with oxygen (Pörtner 2010). The slope of the rising phase is thought to reflect the effect of temperature on ratelimiting enzymes and/or metabolism (Brown et al. 2004). While the falling phase has attracted less interest than the rising phase, it has been suggested to reflect the rate of degradation of important enzymes (Ratkowsky et al. 2005; Dell et al. 2011). Finally, the location of Topt may represent an independently evolving trait related to the thermal optima of performancelimiting enzymes (Somero 2002), a balance between the rate-increasing and rate-depressing processes of the rise and fall phases (Logan et al. 1976), or a temperature window where an organism’s supply and demand of oxygen is in balance (Pörtner 2002). 3 The Universal Temperature Dependence hypothesis The slope of the rising phase has recently attracted considerable interest, largely due to the fact that Brown et al. (2004) used this relationship as one of the cornerstones in their influential Metabolic Theory of Ecology (MTE). According to the MTE, the effect of temperature on biological rate traits can be modelled using the Boltzmann-Arrhenius equation: R = b0 e-E/kT (1) where R is the rate, E is the activation energy (in units eV), k is Boltzmann’s constant (8.62 x 10-5 eV K-1) an T is the temperature in Kelvin. The parameter E describes the slope of the relationship between the natural logarithm of a rate trait and temperature (at the scale of -1/kT). MTE postulates that the activation energy term, E, can only take on values between 0.6 eV and 0.7 eV, with a mean value across traits of 0.65 eV. This is known as the “Universal Temperature Dependence” (UTD) hypothesis (Brown et al. 2004). Recently, Dell et al. (2011) showed that E varies considerably between traits, species and habitats. Paper II is the first empirical test of this hypothesis below the species level and for full-sib families. It is also the first to use a strong, geographic sampling design. Context-dependent thermal performance TPCs estimated under controlled laboratory conditions are sometimes interpreted as if they represent a fundamental physiological property of an organism. We may, for example, be inclined to call a species a thermal specialist if we find that it only grows under a very narrow range in the laboratory. This, however, represents an oversimplification. Instead, TPCs represent the performance of an individual under one specific set of conditions and at a given point in time. There is a large literature on how experimental manipulations of environmental variables (light, temperature, water, etc.) in the laboratory can modify various aspects of physiological traits (e.g. location of thermal optima or tolerance range) (Huey & Berrigan 1996; Chown et al. 2009; Cooper et al. 2010). When this occurs in the laboratory it is called acclimation. Whereas facultative modification of physiological traits by environmental factors in the field is called acclimatization (Wilson & Franklin 2002). Furthermore, the expression of physiological traits can also change over time, both in relation to an individuals’ physiological age (Spence et al. 1980; Berger et al. 2011) and in 4 relation to the time-scale of exposures to environmental factors (Schulte et al. 2011). Traditionally, acclimation responses have been studied mainly from a mechanistic, physiological perspective and the observed patterns have been assumed to be of adaptive significance. This is known as the beneficial acclimation hypothesis (Leroi et al. 1994; Huey & Berrigan 1996). More recently, there has been a surge in interest in critically evaluating whether acclimation responses really are adaptive, and to quantify their costs and benefits (Wilson & Franklin 2002; Woods & Harrison 2002; Kristensen et al. 2008) Thermal performance and life history theory The performance trait that I focus on in this thesis is individual growth rates – the rate at which an individual increase in size during. Growth rate is an important performance trait as size is positively correlated with mating success, fecundity, survival and development time (Kingsolver & Huey 2008) and for the outcome of competitive and aggressive encounters (Rudolf 2007). But growing fast comes at a cost, and growth rates are involved in many trade-offs, making it necessary to consider the temperaturedependence of growth rates within the framework of life history theory (Roff 2002). The trade-off between the age and size at reproduction is perhaps one of the most important in organisms with complex life cycles (i.e. those that make an abrupt habitat shift when transitioning between life stages). Early reproduction allows the completion of more generations in a given amount of time. This affects the intrinsic rate of increase, r (Kingsolver & Huey 2008). Later reproduction, on the other hand, leaves more time available for growth, which allows individuals to grow bigger. This mainly affects the net reproductive rate, R0 (Kingsolver & Huey 2008). The optimal investment in development or growth will shift depending on the costs and benefits of attaining a large size or maturing early (Abrams et al. 1996). High growth rates are also involved in trade-offs with predation risk. Fast growth requires more resources, making it necessary for fast-growing individuals to forage more. This likely causes them to be more exposed to predators (Rowe & Ludwig 1991). These and other trade-offs are necessary to consider in order to derive realistic predictions about patterns of thermal adaptation, and for interpreting the results from comparative physiological studies. 5 The evolutionary ecology of thermal adaptation Adaptation to temperature represents a balancing act between many complex processes, both abiotic and biotic. Latitudinal, continental and local patterns of temperature variation form the abiotic backdrop of thermal adaptation. From the equator to the poles, average temperatures decrease, and seasonality of temperature increase (Piper & Stewart 1996; Angilletta 2009). Seasonality of temperature also varies along longitudinal clines: temperature seasonality increase from the coasts to the interior of large land masses. Light, similarly, varies with latitude, with nearly equally long days and nights year round at the equator, but with increasingly long summer days and winter nights towards higher latitudes (Bradshaw & Holzapfel 2007). This affects patterns of diurnal temperature fluctuation. Average, seasonal and diurnal temperature patterns also depend on local factors such as altitude and topography (Kang et al. 2000). Aquatic and especially marine habitats display greatly reduced fluctuations due to the high capacity of water to store heat (Angilletta 2009). Different terrestrial vegetation types also differ strongly in thermal characteristics. Together, this ensures that each population along a latitudinal gradient will experience a unique set of thermal conditions. Because of gene flow from populations that may be exposed to a radically different thermal regime, thermal strategies at a local scale will in part depend on the local selection regime, but also on selection acting in other parts of a species’ range. To what extent patterns of thermal adaptation reflect optimization to local environmental conditions or evolutionary constraints represents a major unresolved question in thermal biology (Gilchrist & Kingsolver 2001). But thermal strategies also strongly depend on the biotic environment. Changes in temperature will affect the performance and competitive ability of individuals differentially provided they differ in any aspect of thermal performance. Thus, thermal performance will have an important influence on ecological processes, and ecological interactions will have an important influence on the evolution of thermal performance. In many ways, temperature can be considered as one of the axes of an organism’s ecological niche. The benefits that it gains from performing at a given temperature depend not only on the frequency by which it experience that temperature, but also on the behavior of the individuals with which it interacts with at that temperature (Magnuson et al. 1979; Mitchell & Angilletta Jr 2009). Thus, thermal adaptations must be considered as the joint outcome of both environmental (e.g. temperature differences across latitudes, seasonality of temperature, the magnitude of diurnal temperature fluctuations, etc.), ecological (e.g. seasonal variation in predator abundance, prey availability, community composition, and the life histories and thermal strategies of predators, mutualists and competitors) and evolutionary factors 6 (evolutionary constraints on the shape of TPCs, gene flow from other populations, frequency dependent selection from biotic and abiotic factors, genetic drift, etc.). Recently, several theoretical models have produced testable hypotheses about how environmental, ecological and evolutionary factors may drive patterns of thermal adaptation (Hughes & Grand 2000; Mitchell & Angilletta Jr 2009). Empirical studies with the aim to test predictions about the interacting effects of these factors are necessary to advance the field of thermal biology. Aims This introduction was meant to convey one important point: that an understanding of how organisms adapt to thermally heterogeneous environments will require a careful consideration of many different aspects of thermal performance – both separately and jointly, over both time and space, and from the level of enzymes to organisms and ecosystems (Clarke 2003; Pörtner et al. 2006). The focus of this thesis lies on two of aspects of thermal adaptation: performance breadth and variation in the plasticity (i.e. the slope) of the rising phase of TPCs. My aim was to test theoretical predictions about how thermal breadth and plasticity of the rising phase of TPCs vary across levels of biological organization, over time, and in relation to the ecology and life history of different species. Thus, I have studied variation in temperature-responses among species in relation to environmental variation and life history (Paper I), over different hierarchical levels from families, populations, latitudes and species (Paper II), over the lifetime of individual organisms (Paper IV) and in relation to interactions between competing species (Paper V). I also evaluated the genetic diversity and differentiation of the studied populations using microsatellites (Paper III) to back up observed patterns of thermal adaptation with estimates of gene flow and differentiation at neutral loci. Research questions 1) Can latitudinal variation in thermal performance breadth be predicted based on how thermal variation is partitioned among and within generations as predicted by optimality models? How can we quantify within- and amonggeneration temperature variation across latitudes in a group of aquatic organisms with a complex life cycle? (Paper I) 7 2) Is the slope of the rising phase of the thermal performance curves of growth rates constrained by metabolism? If not, over what levels of biological organization do we find variation in the temperature-dependence of growth rates? What is the role of gene flow in preventing or enabling thermal adaptation across species’ range? (Paper II and III) 3) What is the role of thermal acclimation and phenological timing for species living in seasonal environments? What aspects of a species’ thermal environment or life history strategy favour strong acclimation responses? (Paper IV) 4) To what extent can information from controlled laboratory experiments be used to predict the growth performance of individuals in nature? How does inference competition affect the growth performance of damselflies at different temperatures? (Paper V) STUDY SYSTEM My study system consist of six species of damselflies (Insecta: Odonata: Zygoptera) in the genus Coenagrion. The six species were chosen on the basis of their latitudinal pattern of distribution in Europe (Figure 2). Thus, these species represent two northern species (C. armatum and C. johanssoni) that are exclusively found in the boreal parts of northern Europe, two central species (C. puella and C. pulchellum) whose ranges spans much of Central Europe, and two southern species (C. mercuriale and C. scitulum) that are restricted to Southern Europe. My sampling scheme involved collecting eggs and DNA samples of each species from populations near the northern and southern range margin, as well as within the core part of each species’ range (represented by red squares in Figure 2) Odonates (dragonflies and damselflies) are an ancient group of insects, with an evolutionary history deeply rooted in tropical areas (Grimaldi & Engel 2005). Some of the thermal characteristics of ancestral, tropical odonates appears to have remained largely unchanged as they have expanded into temperate areas. Thermal requirements for egg development are consistently among the highest of any insect order (Pritchard et al. 1996). Optimal temperatures for growth and activity fall above 20°C in all stages (Leggott & Pritchard 1985, 1986). Odonates are nevertheless a highly successful group in temperate environments, often being dominant predators in many aquatic ecosystems, in particular those that lack fish (Benke 1976). By studying how odonates have adapted to temperate 8 Figure 2. Map showing the European distribution of the six studied species. The location of the sampled latitudinal populations are shown as red squares. Each square represent three nearby sub-populations. Note differences in distribution pattern, with two North European (a,b), two Central European (c,d) and two South Eropean (d,e) species. Maps are reproduced from Askew (2004) with permission from Apollo books. environments it is possible to gain an understanding of the types and sequence of adaptations that enables or hinders low-latitude taxa to colonize temperate environments. The genus Coenagrion (Odonata: Zygoptera: Coenagrionidae) is one of the most species-rich temperate zygopteran lineages with about 20 species in the Palaearctic region and some additional representatives in the Nearctic region (Dijkstra & Lewington 2006). Adults are typically active in spring and summer. Mating takes place near the aquatic larval habitat. Depending on the species, the habitat consists of a pond, a shallow lake or a slowly running small stream (Askew 2004). A single female usually mates with multiple males. This has resulted in an intense sperm competition between males. Upon contact, a male first remove the sperm package deposited by the previous male before transferring his own sperm (Miller & Miller 1981; Waage 1986). Eggs are deposited in aquatic plants. 9 The aquatic larvae hatch after a brief period of embryonic development (Waringer & Humpesch 1984). Damselfly larvae are generalist predators of smaller invertebrates. Both interference and exploration competition as well as intra-guild predation and cannibalism is common in damselfly larvae (McPeek & Crowley 1987; Johansson 1993, 1996; Anholt 1994). Winters are spent in the larval stage. In fall, Coenagrion larvae enter a facultative diapause in which cold tolerance is increased and development is retarded or completely stopped (preventing premature emergence) (Norling 1984; Corbet 1999). Diapause is initiated by both low temperatures and short photoperiods (Pritchard 1989). Odonates are well known for their highly plastic life histories. The number of generations per year (referred to as voltinism) decrease with increasing latitude (Norling 1984; Corbet et al. 2006). Thus, the southern species in this study are univoltine (one generation/year) over most of their range, although there are indications that some species may complete two generations per year in North Africa (Jacquemin & Boudot 1999). Central European species are likely univoltine over much of their range whereas north European species require two or even more years to complete their development (Norling 1984; Corbet et al. 2006). Variation in voltinism is important with regards to the subject of this thesis for several reasons. The first is related to the thermal niche: variation in voltinism have a strong influence on how seasonally varying temperatures are partitioned within and between larval generations, which I explore in detail in Paper I. The second is related to the fitness consequences of growing fast: depending on the size individuals reach at the end of the growth season, large individuals will develop directly into adult at the beginning of the next growth season, whereas smaller individuals will prolong their life cycle and require one more year to complete their development (Norling 1984; Buskirk 1992). This phenomenon is known as cohort splitting and is explored further in Paper IV. Individuals reaching maturity in one instead of two years will effectively double the intrinsic rate of population increase, r (Roff 2002; Kingsolver & Huey 2008). This means that growing fast can be of immense importance for individual damselflies, further motivating the choice of growth rates as the trait studied in this thesis. Besides affecting development time, variation in growth rates is important for damselfly larvae as growth rates affect the relative size of individuals. Size differences is an important aspect of the life history of damselfly larvae: the strength and direction of intraguild predation, cannibalism and competition strongly depend on relative size differences between larvae (Polis et al. 1989; Buskirk 1992; Rudolf 2007). In addition, the size of final-instar larvae is correlated with adult size, which is an important determinant of longevity, fecundity and mating success in damselflies (Banks & Thompson 1987; Thompson et al. 2011). 10 METHODS Experiment 1: Thermal sensitivity of growth rates across latitudes This experiment was used for Paper I, II and IV and was conducted over a period of three years, between 2008 and 2010. I reared larval damselflies individually at three or four temperatures to estimate the effects of temperature on larval growth rates. For each species, I acquired eggs from females collected near the northern and southern limits of its range, and from the center of its range (Figure 2). At each of these latitudes, I sampled between two and four populations (i.e. separate water bodies). From each population, I collected eggs from about 5 females, making the total sample size about 15 egg clutches at each latitude. Egg clutches were brought to the laboratory in Umeå, Sweden or Leuven, Belgium and reared at the following fixed temperatures: 16.3°C, 19.5°C, 21.5°C and 24.0°C (24.0°C added during 2009 and 2010). The experiment was a full-sib design: 15 or 20 offspring from each female were randomly divided between the three or four temperature treatments (five per temperature). Thus, full-sib larvae were nested in families nested in populations nested in latitudes nested in species. In total, I reared more than 5000 individual larvae from over 200 familygroups. Larvae were fed six days a week on brine shrimps (Artemia salina) and the head width of each individual was measured at the age of 0, 42, 84 and 126 days. Using these data, I fitted individual growth trajectories as a thirddegree polynomial function between head width and time. Individual, sizecorrected relative growth rates (RGRs) were calculated as the slope of the growth trajectory when an individual had reached 1/5th the adult head width. This method accounts for differences between species in both initial and final size (Rose et al. 2009). See below under Paper I, II and IV for details and results from this experiment. Experiment 2: Effects of competition and temperature on growth rates The second experiment, used for Paper V, was a microcosm experiment designed to test how the thermal sensitivity of growth rates depend on competitive interactions between and within species. I selected two species from the first experiment: the northern C. armatum and the central C. pulchellum. The reasons for choosing these species were threefold. First, the latter species is expanding its range northwards into the range of the former (Hickling et al. 2005). Thus, I was interested in the population dynamics of 11 these species when they come into contact. Second, Experiment 1 revealed that the two species differ in the thermal sensitivity of growth rates. And finally, the two species were possible to distinguish from each other in the larval stage based on morphological characters. I set up 48 microcosms (3L water volume), each containing 30 newly hatched damselfly larvae, aquatic snails, macrophytes and microorganism from a natural pond. Larvae were continuously fed Artemia salina and Daphna magna. The experimental design was as follows: two damselfly species, two temperatures (16°C and 21°C) and two competition treatments (intra- and interspecific competition). Thus, one third of the microcosms contained only C. armatum (intraspecific), one third contained only C. pulchellum (intraspecific) and one third contained both species (interspecific competition). The microcosms were run for 126 days and the head width of individuals was checked at day 0, 42, 84 and 126 as in the first experiment. To calculate relative growth rates from these data, I used the same growth model as in Experiment 1. However, since I could not keep track of individual larvae, I used a stratified resampling approach. I ranked individual observations based on their relative size at each measurement event, and iteratively fitted the growth model using those data points that had the same rank order within each measurement event. I was interested both in the growth rates and in changes in the size distribution of larvae over time. More details and results from this experiment can be found in Paper IV. PAPER I: THERMAL BREADTH DECREASE WITH LATITUDE In Paper I, I show that the thermal breadth of growth rates becomes narrower towards higher latitudes. I propose that this reflect a combination of rapid seasonal transitions in high-latitude environments together with that damselfly larvae enter diapause at low temperatures. The aim of this paper was to test the predictions from two theoretical models that have explored how thermal breadth depends on environmental variation. The first is by Lynch & Gabriel (1987) and the second by Gilchrist (1995). The main difference between these two models lies in how performance is assumed to affect fitness. In Lynch & Gabriel’s model, performance affects fitness by increasing survival, whereas in Gilchrist’s model, performance affects fitness through increasing fecundity. Either of these fitness mechanisms may apply to growth rates (Arendt 1997; Roff 2002). If growth rates mainly affect survival, thermal breadth should 12 increase with increasing thermal heterogeneity within generations (Lynch & Gabriel 1987). If growth rates affect fitness by increasing fecundity, thermal breadth should depend on how environmental variation is partitioned among relative to within generations (Gilchrist 1995). Few studies have explicitly tested these predictions (Angilletta 2009). In part, this is related to the challenges involved in quantifying temperature variation among and within generations. This is especially challenging in aquatic organisms because water temperature data is not generally available in the same way as air temperature data. To deal with these problems, I combine simulations of water temperature and damselfly life cycles with empirical data on the thermal breadth of the six species from Experiment 1. This paper is a comparative study of difference between species. Thus, I only used data from central populations. Model 1: lake temperature across latitudes My first step was to produce realistic temperature profiles for representative water bodies along the latitudinal gradient. To do this, I took advantage of the recently developed lake model FLake (Mironov 2008). FLake was intended to model feedback loops between the atmosphere and lakes for weather forecasting, but it can also be used to model the temperature evolution of lakes of a user-specified depth and turbidity anywhere in the world. The model requires input data such as air temperature, solar radiation, cloud cover and wind speed. This information is available in publicly available global databases. I ran a total of 18 FLake simulations using weather data from each of the areas where the six damselfly species were sampled (see Figure 2). This allowed me to characterize patterns of temperature variation along the entire 3,600 km latitudinal transect. The weather data came from the ECMWF Era-Interim dataset for the years 19992009 at 12h intervals (Simmons et al. 2007). Model 2: larval life cycles and temperature variation In this model, I considered how environmental variation was partitioned among and within damselfly generations. I incorporated two important aspects of the life history of damselflies in this model: 1) that damselfly larvae enter diapause during winter; and 2) that the generation length increase with latitude (Corbet 1999; Corbet et al. 2006). With regards to diapause, I assumed that temperatures below 8°C were not meaningful for the evolution of thermal breadth of growth rates, as damselflies do not grow at temperatures below 8-10°C (Corbet 1999). Next, I defined the length of a 13 Figure 3. Representative temperature profiles (surface water temperature) for shallow lakes at different latitudes. Temperature data comes from FLake simulations using meteorological data from Northern Sweden (a), Belgium (b) and Southern Spain (c). Data from three separate years (2000-2002) are shown in light gray, medium gray and black, respectively. generation based on the number of degree-days needed to complete larval development (based on my own laboratory data). Running these life cycle simulations using water temperature data from Model 1, I was able to quantify how water temperature variation was partitioned within and between damselfly generations across latitudes. Results and discussion The life cycle simulations revealed the following patterns. First, withingeneration thermal variability decreased towards higher latitudes. Second, the relative amount of among-to-within generation temperature variation peaked at mid-latitudes (~50°N). The latter was due to a voltinism shift from a univoltine to a semivoltine life cycle. The finding that within-generation variation decrease with latitude may at first appear surprising, since highlatitude environments are characterized by dramatic seasonal fluctuations in temperature. This result, however, arise due to the fact that seasonal 14 transitions are more rapid in high-latitude environments together with that larvae enter diapause at low temperatures. Thus, in low-latitude environments, temperatures change slowly over the year and only rarely reach below the diapause threshold (Figure 3C). As a consequence, lowlatitude damselflies spend a considerable amount of time growing at lowand intermediate temperatures (e.g. 10-15°C), (Figure 3C), which greatly increase within-generation variability. At higher latitudes, a considerable part of the year is spent in diapause, during which time no selection on growth performance is expected to occur (Figure 3A). When winter is over, temperatures rise rather rapidly up to about ~15-20°C. As a consequence, individuals spend much less time at low and intermediate temperatures and within-generation variability is greatly reduced during the relatively brief period each year when growth occurs (Figure 3A). Comparing the TPCs of growth rates of the six species, I found that thermal breadth decreased with latitude. I was unable to measure thermal breadth directly as the entire experimental thermal range (16°C - 24°C) was found to lie within the initial, rising phase of all six species. However, the differences between species were sufficiently strong that it was clear that high-latitude species focus their growth to a narrow range of relatively high temperatures (i.e. thermal specialization), whereas low-latitude species grow relatively fast over a much broader range of temperatures (i.e. thermal generalization), with central species being intermediate. Thus, the latitudinal pattern of thermal breadth of these species supported the Lynch & Gabriel (1987) prediction that performance breadth decrease with decreasing within-generation variation across latitudes. But it is important to note that the direction of both the predicted and observed pattern of thermal breadth evolution was not that which may intuitively be expected. Instead, thermal breadths were narrower in the most thermally heterogeneous environments. But due to the presence of a larval diapause, much of this variation may not be of very great importance to the growth of individual larvae. PAPER II & III: GROWTH RATES DO NOT SCALE WITH METABOLISM In Papers II and III, I show that the slope of the rising phase of TPCs (E) of growth rate increase with latitude both within and between species, and that the within-species latitudinal trend in E is maintained despite strong gene flow. I therefore propose that E is a trait involved in local adaptation. This 15 rejects a key prediction from the Metabolic Theory of Ecology (MTE) (Brown et al. 2004), that the slope of the rising phase of TPCs should be tightly constrained by metabolism. In Paper II, I estimated E (the slope of the relationship between growth rate and temperature at the Arrhenius scale of -1/kT; see Eq. [1]) for each full-sib family in Experiment 1. With six species, three latitudes from each species, three populations from each latitude and about five families from each population, I was able to quantify variation in E both within and between species along the full 3,600 km latitudinal gradient. In total, I was able to estimate E for 209 family-groups, represented by 2924 individuals. In Paper III, I analyzed microsatellite data (between 7 and 12 loci) from about 30 adults collected from each of the populations that were used to estimate variation in E (with the exception of C. scitulum). With these data, we estimated genetic diversity in each latitudinal population, differentiation between sub-populations within latitudes, and migration rates between latitudes. Results and discussion I found that the average activation energy of damselfly growth rates was significantly higher (i.e. the slope of the rising phase of TPCs were steeper) than the 0.6-0.7 eV range predicted by the MTE. In addition, E varied considerably both within and between species. The strongest trend was that temperature-responses became steeper with increasing latitude. In four out of six species, I found a significant, positive latitudinal trend within species. Subsequent analyses showed that the best environmental predictors of the slope of temperature-responses were a positive correlation between E and temperature seasonality, and a negative correlation with average temperature during the summer months. The population genetic analyses revealed that gene flow is strong over much of the ranges of these species (although the Pyrenees formed a strong barrier to dispersal in C. puella and C. mercuriale). Clearly, latitudinal differences in E are not a product of genetic drift. Thus, the slope of the rising phase of TPCs is very likely involved in how latitudinal populations adapt to local environmental conditions. The finding that E varies with latitude has important implications for several aspects of our understanding of temperature-responses. First, all damselfly populations that had slopes close to the “universal” value of 0.65 eV where located within the southern range of the latitudinal gradient (between 35° and 50°N). Coincidentally, this perfectly matches the latitudinal band where the vast majority of previous research on insect temperature responses has been carried out (reviewed by Irlich et al. 16 [2009]). Based on this, I suggest that the universal value of 0.65 eV may simply represent the average value within a restricted geographic area where the vast majority of researchers work. It is clear that carefully designed, spatially extensive studies such as mine will be important for gaining a better understanding of how temperature-responses have evolved in response to environmental and ecological factors. The population genetic analyses revealed that migration is asymmetrical in these species: total migration towards south (i.e. from northern to central, or from central to southern populations) was higher than total migration towards north. This could arise if there is a higher propensity to disperse towards south, although there is no clear reason why this would be the case here. Instead, it seems likely that migrants going south are more successful than migrants going north. This may reflect differences in the growth strategies of “northern“ and “southern” genotypes, elaborated in Paper I and Paper II. Northern genotypes focus growth to periods of beneficial conditions, and stop growing relatively early in the season. Southern genotypes, on the other hand, continue to grow later in the season and appears to invest more in growing fast over a broader range of temperatures. A southern strategy in a northern environment may thus be strongly selected against, as an individual will still attempt to grow under adverse conditions towards the end of the growth season. A northern strategy in a southern environment, on the other hand, may be under less strong selection. Although the northern genotypes will likely grow sub-optimally compared to locally adapted genotypes because they fail to exploit the full length of the growth season, this may not translate into increased mortality to the same extent as when southern genotypes confront high-latitude environments. PAPER IV: THERMAL PERFORMANCE SHIFTS DURING ONTOGENY In Paper IV, I look at how the temperature-dependence of individual growth rates change over the ontogeny of individual larvae. I contrast this with information about how the thermal environment differs between larvae of different age-classes in the field. The aim was to test the hypothesis that acclimation of temperature-specific growth rates have evolved to allow individuals to track changes in environmental temperatures during ontogeny. Few studies have quantified differences between species in the strength of acclimation responses. Even fewer have considered the 17 environmental factors that may select for strong or weak acclimation responses. I modelled the life cycle of damselflies in relation to temperature variation in the field in order to derive predictions about how the strength of acclimation depends on changes in the thermal niche among age-classes. In this model, I bring together two types of phenological information on the timing of damselfly reproduction. First, I reviewed the zoological literature for information about the flight periods of Coenagrion sp. in different parts of Europe. Using this information, I was able to predict the start and end of the flight period for each species in the area where it was sampled. Second, I used data from Lowe et al. (2009) on the seasonal timing of egg laying within the flight season. This revealed that most eggs are laid relatively early during the flight season. I incorporated this information into a model that I used to simulate the hatching date of 1000 individuals of each species in the area where they were sampled. The thermal environment that these indiviudals would experience during different parts of their life cycle were estimated by coupling individuals’ hatching dates to temperature data from the FLake simulations. I quantified the growth rates of individuals during early (day 0-42), intermediate (day 42-84) and late (day 84-126) ontogeny using data from Experiment 1. In organisms with determinate growth, absolute growth rates decrease with size. To account for this, I used the growth trajectories of the fastest-growing individuals as a “reference trajectory” in order to calculate size-corrected growth rates during early, intermediate and late ontogeny. I quantified the strength of acclimation by assessing how much the slope of temperature responses changed during ontogeny – if acclimation responses are strong, individuals will grow almost equally well at all temperatures after having spent some time at the rearing temperatures (and the slope of temperature responses will decrease over the ontogeny). Results and discussion My main finding were that acclimation responses differed between species. Acclimation responses were strongest in species inhabiting the least seasonal environments (i.e. southern European species), with thermally specialized, north European species displaying the weakest acclimation responses. The environmental factor that best predicted the strength of acclimation was whether a species experienced slowly or rapidly changing temperatures during ontogeny. My interpretation of this is that acclimation may indeed be important for allowing individuals to track changes in temperature during ontogeny, but that acclimation responses are costly and that the strength of acclimation will depend on the relative costs and benefits of acclimating. If 18 acclimation was not costly, we would expect high-latitude species to display very strong acclimation responses, the opposite to what I found. Thus, as long as temperatures change slowly (in low-temperate environments), individuals are able to track changes in temperature by acclimating. These individuals will be favoured by selection as they can grow better than nonacclimating individuals both immediately following hatching when temperatures are warm and stable, as well as towards the end of the growth season when temperatures are lower and more variable. At higher latitudes, temperatures drop so rapidly in autumn, that acclimating at a sufficient rate will be exceedingly costly or even physiologically impossible. In addition, there will be less to gain from acclimating, as less time will be spent at any given temperature. This is to my knowledge a novel hypothesis about the strength of acclimation responses that should be further tested in systems that display greater differences between species in the seasonal timing of reproduction and growth. PAPER V: COMPETITION AND TEMPERATURE MODIFIES GROWTH RATES In paper V, I show that intra- and interspecific interactions alters the slope of temperature-responses. This paper is based on data from Experiment 2. This study was motivated by recent interest in incorporating information about the slope of the rising phase of TPCs into predictive models of ecosystem responses to climate change (Montoya & Raffaelli 2010; Woodward et al. 2010). Individual growth rates are relevant in this context as they determine the strength and direction of ecological interactions (Rudolf 2006; Binzer et al. 2012). They also affect population size structure, with important implications for community dynamics (De Roos et al. 2003; Persson & De Roos 2006). Growth rate is a complex trait that depends on food density, food quality, behavior, size-dependent mortality, metabolism and respiration, among other factors (Baker 1982; Anholt 1990; McPeek 2004). It is therefore far from certain that differences in growth rates observed under controlled laboratory experiments are representative of individuals’ performance under more realistic conditions. To test this, I reared C. armatum and C. pulchellum in microcosms (30 individuals in 3L microcosms) at two temperatures, either with members of its own species, or in equal proportions of both species. The growth patterns of individuals 19 reared under these conditions were contrasted with the results from Experiment 1. Results and discussion I found that temperature responses were steeper when individuals were reared in the presence of competitors than when reared individually. Although I was not able to determine the mechanistic basis for this finding, it clearly demonstrates that it will be very difficult to predict ecosystem responses to climate change by extrapolating from laboratory experiments of single species. Growth rates and other complex traits do not simply reflect a physiological property of an organism at a given temperature. Instead, they are shaped by trade-offs between many temperature-dependent processes such as predation rates, attack rates, inference competition, food acquisition, digestion, respiration and metabolism (Clarke 2003). I also found that the size structure of larval cohorts were dependent on both temperature and competion. However, the joint effect of these processes would not be possible to predict if they were studied in isolation. A notable finding was that some individuals were able to grow as fast as, or even faster, at 16°C than at 21°C. This occurred even though the vast majority of individuals grew considerably slower at the former temperature. A combination of high food intake rates and high growth efficiency is thus able to completely overcome the rate-depressing effects of low temperatures. CONCLUDING REMARKS To conclude, I found variation in the thermal sensitivity of damselfly growth rates over multiple levels of biological organization: within and among species, over individuals’ ontogeny and depending on the ecological context. My main finding is that the slope of temperature responses increases with latitude, both within and between species. This has important implications for our understanding of how temperature affects biological systems, as it rejects the hypothesis that the slope of the rising phase of TPCs is fundamentally constrained by metabolism. On the contrary, my findings revealed that differences in the slope of the rising phase of TPCs can evolve despite strong gene flow between populations. Based on this, and the results from several recent reviews (Irlich et al. 2009; Dell et al. 2011; Englund et al. 2011), it is clear that temperature affects biological rate traits in much more complex ways than what has been widely appreciated. I have highlighted several aspects of the life history and ecology of damselflies that may 20 generate variation in temperature responses. In order to understand latitudinal patterns of thermal adaptation, we must clearly investigate the interacting effects of life history, ecology and environmental variation. I also found that damselflies become increasingly thermally specialized (steeper reaction norms) towards higher latitudes. This may reflect that damselflies spend an increasingly large part of the year in diapause towards higher latitudes, which results in that growth only takes place during periods of relatively warm and stable temperatures. Consequently, my findings support the idea that thermal specialists are favoured when the amount of within-generation variation is small (Lynch & Gabriel 1987). However, this shows that we risk to mischaracterize latitudinal temperature clines if we do not carefully consider the life history of the studied species (Angilletta 2009; Clusella-Trullas et al. 2011).The fact that food availability, predation and other important ecological factors also display seasonal fluctuations may further cause damselflies to focus their growth efforts to periods of relative warmth. Although I cannot conclusively show that the evolution of a larval diapause have preceded thermal specialization in this case, these species appears to conform with the model proposed by Bradshaw and colleagues (Bradshaw et al. 2000, 2004). This stipulates that organisms first evolve diapause thresholds that allow them to survive in a novel temperate environment, followed by changes in thermal sensitivity. I also highlight the importance of considering the seasonal timing of major life history events when deriving predictions about thermal adaptation. In particular, I show how the seasonal timing of reproduction can have strong effects on the conditions that individual larvae experience during ontogeny. I find that the rate of change of the thermal niche during ontogeny may be an important factor for understanding the costs and benefits of thermal acclimation. Finally, I show that the inferences we make about how temperature affects growth rates of damselflies depends on the methodological context. Temperature-responses were steeper when larvae were able to compete with other individuals, and species differed in how strong this effect was. Information on temperature-responses estimated in the laboratory is currently used to predict ecosystem responses to climate change. My findings raise questions about how realistic these models are. I have only scratched the surface of the complex ways in which ecology, life history and environmental variation shapes the evolution of temperature responses. My research suggests several fields of inquiry that merits further attention. In particular, is the timing of reproduction in damselflies shaped by the thermal requirements of early-instar larvae? Or has the thermal requirements of larvae evolved to match the environment they experience due to the flight periods of adults? Similar studies that investigate thermal sensitivity in different instars in species that differ in the timing of 21 reproduction would be able to answer these questions. It would also be important to identify the mechanisms that give rise to differences in temperature responses with or without competition. Investigations of the effects of temperature on feeding, digestion, predation and several other traits, both in isolation and in the presence of competitors, would help us understand how these processes scales with temperature and competition. By focusing on differences between species in feeding and aggression behavior it would be possible to find out why one of our species was strongly affected by competition, whereas the other was not. My work shows that it is impossible to understand the costs and benefits of growth rates under realistic conditions without considering the trade-offs between rapid growth and size structure, survival, development rate and mating success. This will require a much better understanding of the fitness consequences of performance. Many optimality models assume that performance affect fitness either purely additively through fecundity or purely multiplicatively through survival. This is most likely unrealistic. We should instead focus on quantifying the contribution of performance to different components of fitness, such as development time, fecundity, survival and mating success. This needs to be done across a range of environmental and ecological conditions. Organisms depend on so many different aspects of their biotic and abiotic environment, that each display seasonal and diurnal oscillations. I believe that the best way to gain an understanding of how performance and species interactions depend on temperature is to simultaneously study as many aspects of performance and fitness as possible under natural conditions. Transplantation experiments under near-natural conditions can provide the much needed realism that allows organisms to display their full suite of behavioral adaptations, while still allowing researchers to modify important environmental variables. With this thesis, I hope to have shed some light on the processes that shape latitudinal patterns of thermal adaptation in organisms with complex life cycles and where body temperature is highly or completely dependent on ambient temperatures. ACKNOWLEDGEMENTS I would like to thank Frank Johansson and Malin Nilsson for providing valuable comments on earlier versions of this thesis. The research was funded by grants from the Swedish Research Council Formas to Viktor Nilsson-Örtman and Frank Johansson; from the Fund for Scientific Research-Flanders (FWO) to Marjan De Block; from FWO and the KULeuven research fund to Robby Stoks; and from Umeå University and the Academy of Finland to Helena Johansson. 22 AUTHOR CONTRIBUTIONS Paper I Viktor Nilsson-Örtman, Robby Stoks and Frank Johansson conceived the study. VNÖ, RS, FJ and Marjan De Block performed the laboratory experiment. VNÖ developed the computer model, analyzed data, and wrote the paper; and RS, MDB, and FJ contributed to writing. Paper II Viktor Nilsson-Örtman, Robby Stoks, Marjan De Block, Helena Johansson and Frank Johansson conceived the study and performed the laboratory experiment; VNÖ analyzed data, developed the computer models, and wrote the manuscript; RS, MDB, HJ and FJ contributed to writing. Paper III Viktor Nilsson-Örtman, Robby Stoks, Helena Johansson and Frank Johansson conceived the study and carried out the fieldwork. HJ genotyped the samples, analyzed the data and wrote the manuscript. FJ, RS and Pär K. Ingvarsson contributed to writing. Paper IV Viktor Nilsson-Örtman, Robby Stoks and Frank Johansson conceived the study. VNÖ, RS, FJ and Marjan De Block performed the laboratory experiment. VNÖ analyzed data, developed the computer models, and wrote the manuscript. FJ contributed to writing. Paper V Viktor Nilsson-Örtman , Robby Stoks and Frank Johansson concieved the study. VNÖ carried out the field work, performed the laboratory experiment, analyzed data and wrote the paper. FJ contributed to writing. 23 REFERENCES Abrams, P.A., Leimar, O., Nylin, S. & Wiklund, C. (1996). The effect of flexible growth rates on optimal sizes and development times in a seasonal environment. The American Naturalist, 381–395. Angilletta, M.J. (2009). Thermal adaptation: a theoretical and empirical synthesis. Oxford University Press, USA. Anholt, B.R. (1990). An experimental separation of interference and exploitative competition in larval damselfly. Ecology, 1483–1493. Anholt, B.R. (1994). Cannibalism and early instar survival in a larval damselfly. Oecologia, 99, 60–65. Arendt, J.D. (1997). Adaptive intrinsic growth rates: an integration across taxa. The Quarterly Review of Biology, 72, 149. Askew, R.R. (2004). The dragonflies of Europe (revised edition). Harley Books, Colchester, UK. Baker, R.L. (1982). Effects of food abundance on growth, survival, and use of space by nymphs of Coenagrion resolutum (Zygoptera). Oikos, 47–51. Banks, M.J. & Thompson, D.J. (1987). Lifetime reproductive success of females of the damselfly Coenagrion puella. The Journal of Animal Ecology, 815–832. Benke, A.C. (1976). Dragonfly production and prey turnover. Ecology, 57, 915–927. Berger, D., Friberg, M. & Gotthard, K. (2011). Divergence and ontogenetic coupling of larval behaviour and thermal reaction norms in three closely related butterflies. Proceedings of the Royal Society B: Biological Sciences, 278, 313–320. Binzer, A., Guill, C., Brose, U. & Rall, B.C. (2012). The dynamics of food chains under climate change and nutrient enrichment. Philosophical Transactions of the Royal Society B: Biological Sciences, 367, 2935–2944. Bradshaw, W. & Holzapfel, C. (2007). Evolution of animal photoperiodism. Annual Review of Ecology Evolution and Systematics, 38, 1–25. Bradshaw, W.E., Fujiyama, S. & Holzapfel, C.M. (2000). Adaptation to the thermal climate of North America by the pitcher-plant mosquito, Wyeomyia smithii. Ecology, 81, 1262–1272. Bradshaw, W.E., Zani, P. a. & Holzapfel, C.M. (2004). Adaptation to temperate climates. Evolution, 58, 1748–1762. 24 Brown, J.H., Gillooly, J.F., Allen, A.P., Savage, V.M. & West, G.B. (2004). Toward a Metabolic Theory of Ecology. Ecology, 85, 1771– 1789. Buskirk, J.V. (1992). Competition, cannibalism, and size class dominance in a dragonfly. Oikos, 65, 455–464. Chappell, M.A. (1983). Metabolism and thermoregulation in desert and montane grasshoppers. Oecologia, 56, 126–131. Chown, S.L., Jumbam, K.R., Sørensen, J.G. & Terblanche, J.S. (2009). Phenotypic variance, plasticity and heritability estimates of critical thermal limits depend on methodological context. Functional Ecology, 23, 133–140. Clarke, A. (2003). Costs and consequences of evolutionary temperature adaptation. Trends in Ecology & Evolution, 18, 573– 581. Clusella-Trullas, S., Blackburn, T.M. & Chown, S.L. (2011). Climatic predictors of temperature performance curve parameters in ectotherms imply complex responses to climate change. The American Naturalist, 177, 738–751. Cooper, B.S., Czarnoleski, M. & Angilletta, M.J. (2010). Acclimation of thermal physiology in natural populations of Drosophila melanogaster : a test of an optimality model. Journal of Evolutionary Biology, 23, 2346–2355. Corbet, P., Suhling, F. & Soendgerath, D. (2006). Voltinism of Odonata: a review. International Journal of Odonatology, 9, 1–44. Corbet, P.S. (1999). Dragonflies: behaviour and ecology of Odonata. Cornell University Press, Ithaca, New York, USA. Daniel, R.M. & Danson, M.J. (2010). A new understanding of how temperature affects the catalytic activity of enzymes. Trends in Biochemical Sciences, 35, 584–591. Dell, A.I., Pawar, S. & Savage, V.M. (2011). Systematic variation in the temperature dependence of physiological and ecological traits. Proceedings of the National Academy of Sciences, 108, 10591– 10596. Dijkstra, K.B. & Lewington, R. (2006). Field Guide to the Dragonflies of Britain and Europe. British Wildlife Publishing, Gillingham, Dorset, UK. Englund, G., Öhlund, G., Hein, C.L. & Diehl, S. (2011). Temperature dependence of the functional response. Ecology Letters, 14, 914– 921. 25 Feder, M.E. & Hofmann, G.E. (1999). Heat-shock proteins, molecular chaperones, and the stress response: evolutionary and ecological physiology. Annual review of physiology, 61, 243–282. Gehring, W.J. & Wehner, R. (1995). Heat shock protein synthesis and thermotolerance in Cataglyphis, an ant from the Sahara desert. Proceedings of the National Academy of Sciences, 92, 2994–2998. Gilchrist, G.W. (1995). Specialists and generalists in changing environments. i. fitness landscapes of thermal sensitivity. The American Naturalist, 146, 252. Gilchrist, G.W. & Kingsolver, J.G. (2001). Is Optimality Over the Hill? In: Adaptationism and optimality (ed Orzack, S.& Sober E.). Cambridge Univ. Press, Cambridge, U.K. pp. 219-241. Grimaldi, D. & Engel, M.S. (2005). Evolution of the Insects. Cambridge University Press, Cambridge, U.K. Hickling, R., Roy, D., Hill, J. & Thomas, C. (2005). A northward shift of range margins in British Odonata. Global Change Biology, 11, 502–506. Hochachka, P.W. & Somero, G.N. (2002). Biochemical adaptation: mechanism and process in physiological evolution. Oxford University Press, USA. Huey, R.B. & Berrigan, D. (1996). Testing evolutionary hypotheses of acclimation. In: Animals and temperature: Phenotypic and evolutionary adaptation (ed Johnston, I.A. & Bennett, A.F.). Cambridge University Press, Cambridge, pp. 205–237. Huey, R.B. & Slatkin, M. (1976). Cost and Benefits of Lizard Thermoregulation. The Quarterly Review of Biology, 51, 363–384. Hughes, N.F. & Grand, T.C. (2000). Physiological ecology meets the ideal-free distribution: predicting the distribution of size-structured fish populations across temperature gradients. Environmental Biology of Fishes, 59, 285–298. Irlich, U.M., Terblanche, J.S., Blackburn, T.M. & Chown, S.L. (2009). Insect rate-temperature relationships: environmental variation and the metabolic theory of ecology. The American Naturalist, 819–835. Jacquemin, G. & Boudot, J.P. (1999). Les libellules (odonates) du Maroc. Société franc̨ aise d’odonatologie, Bois-d'Arcy, France. Johansson, F. (1993). Intraguild predation and cannibalism in odonate larvae: effects of foraging behaviour and zooplankton availability. Oikos, 66, 80–87. 26 Johansson, F. (1996). The influence of cannibalism and prey density on growth in the damselfly Coenagrion hastulatum. Archiv für Hydrobiologie, 137, 523–535. Kang, S., Kim, S., Oh, S. & Lee, D. (2000). Predicting spatial and temporal patterns of soil temperature based on topography, surface cover and air temperature. Forest Ecology and Management, 136, 173–184. Kingsolver, J. & Huey, R. (2008). Size, temperature, and fitness: three rules. Evolutionary Ecology Research, 10, 251–268. Kristensen, T.N., Hoffmann, A.A., Overgaard, J., Sørensen, J.G., Hallas, R. & Loeschcke, V. (2008). Costs and benefits of cold acclimation in field-released Drosophila. Proceedings of the National Academy of Sciences, 105, 216–221. Leggott, M. & Pritchard, G. (1985). The effect of temperature on rate of egg and larval development in populations of Argia vivida Hagen (Odonata: Coenagrionidae) from habitats with different thermal regimes. Canadian Journal of Zoology, 63, 2578–2582. Leggott, M. & Pritchard, G. (1986). Thermal preference and activity thresholds in populations of Argia vivida (Odonata: Coenagrionidae) from habitats with different thermal regimes. Hydrobiologia, 140, 85–92. Leroi, A.M., Bennett, A.F. & Lenski, R.E. (1994). Temperature acclimation and competitive fitness: an experimental test of the beneficial acclimation assumption. Proceedings of the National Academy of Sciences, 91, 1917–1921. Logan, J.A., Wollkind, D.J., Hoyt, S.C. & Tanigoshi, L.K. (1976). An analytic model for description of temperature dependent rate phenomena in arthropods. Environmental Entomology, 5, 1133– 1140. Lowe, C.D., Harvey, I.F., Watts, P.C. & Thompson, D.J. (2009). Reproductive timing and patterns of development for the damselfly Coenagrion puella in the field. Ecology, 90, 2202–2212. Lynch, M. & Gabriel, W. (1987). Environmental tolerance. The American Naturalist, 129, 283–303. Magnuson, J.J., Crowder, L.B. & Medvick, P.A. (1979). Temperature as an ecological resource. The American Zoologist, 19, 331–343. McPeek, M.A. (2004). The growth/predation risk trade-off: so what is the mechanism? The American Naturalist, 163, E88–E111. McPeek, M.A. & Crowley, P.H. (1987). The effects of density and relative size on the aggressive behaviour, movement and feeding of 27 damselfly larvae (Odonata: Coenagrionidae). Animal Behaviour, 35, 1051–1061. Miller, P.L. & Miller, C.A. (1981). Field observations on copulatory behaviour in Zygoptera, with an examination of the structure and activity of male genitalia. Odonatologica, 10, 201–218. Mironov, D.V. (2008). Parameterization of lakes in numerical weather prediction. Description of a lake model. COSMO Technical Report. Mitchell, W.A. & Angilletta Jr, M.J. (2009). Thermal games: frequency‐dependent models of thermal adaptation. Functional Ecology, 23, 510–520. Montoya, J.M. & Raffaelli, D. (2010). Climate change, biotic interactions and ecosystem services. Philosophical Transactions of the Royal Society B: Biological Sciences, 365, 2013–2018. Norling, U. (1984). Life history patterns in the northern expansion of dragonflies. Advances in odonatology, 2, 127–156. Peck, L.S. (2002). Ecophysiology of Antarctic marine ectotherms: limits to life. Polar Biology, 25, 31–40. Persson, L. & De Roos, A.M. (2006). Size-structured interactions and the dynamics of aquatic systems. Polish Journal of Ecology, 54, 621–632. Piper, S.C. & Stewart, E.F. (1996). A gridded global data set of daily temperature and precipitation for terrestrial biospheric modeling. Global biogeochemical cycles, 10, 757–782. Polis, G.A., Myers, C.A. & Holt, R.D. (1989). The ecology and evolution of intraguild predation: potential competitors that eat each other. Annual review of ecology and systematics, 297–330. Pritchard, G. (1989). The roles of temperature and diapause in the life history of a temperate-zone dragonfly: Argia vivida (Odonata: Coenagrionidae). Ecological Entomology, 14, 99–108. Pritchard, G., Harder, L.D. & Mutch, R.A. (1996). Development of aquatic insect eggs in relation to temperature and strategies for dealing with different thermal environments. Biological Journal of the Linnean Society, 58, 221–244. Pörtner, H.O. (2002). Climate variations and the physiological basis of temperature dependent biogeography: systemic to molecular hierarchy of thermal tolerance in animals. Comparative Biochemistry and Physiology-Part A: Molecular & Integrative Physiology, 132, 739–761. 28 Pörtner, H.O. (2010). Oxygen-and capacity-limitation of thermal tolerance: a matrix for integrating climate-related stressor effects in marine ecosystems. Journal of Experimental Biology, 213, 881. Pörtner, H.O., Bennett, A.F., Bozinovic, F., Clarke, A., Lardies, M.A., Lucassen, M., et al. (2006). Trade-offs in thermal adaptation: the need for a molecular to ecological integration. Physiological and Biochemical Zoology, 79, 295–313. Ratkowsky, D.A., Olley, J. & Ross, T. (2005). Unifying temperature effects on the growth rate of bacteria and the stability of globular proteins. Journal of theoretical biology, 233, 351–362. Roff, D.A. (2002). Life history evolution. Sinauer Associates Sunderland, Massachusetts, USA. De Roos, A.M., Persson, L. & McCauley, E. (2003). The influence of size-dependent life-history traits on the structure and dynamics of populations and communities. Ecology Letters, 6, 473–487. Rose, K.E., Atkinson, R.L., Turnbull, L.A. & Rees, M. (2009). The costs and benefits of fast living. Ecology Letters, 12, 1379–1384. Rowe, L. & Ludwig, D. (1991). Size and timing of metamorphosis in complex life cycles: time constraints and variation. Ecology, 72, 413–427. Rudolf, V.H.. (2006). The influence of size-specific indirect interactions in predator-prey systems. Ecology, 87, 362–371. Rudolf, V.H.. (2007). Consequences of stage-structured predators: cannibalism, behavioral effects, and trophic cascades. Ecology, 88, 2991–3003. Schulte, P.M., Healy, T.M. & Fangue, N.A. (2011). Thermal performance curves, phenotypic plasticity, and the time scales of temperature exposure. Integrative and Comparative Biology, 51, 691–702. Simmons, A., Uppala, S., Dee, D. & Kobayashi, S. (2007). ERAInterim: New ECMWF reanalysis products from 1989 onwards. ECMWF Newsletter, 110, 25–35. Somero, G.N. (2002). Thermal physiology and vertical zonation of intertidal animals: optima, limits, and costs of living. Integrative and Comparative Biology, 42, 780–789. Spence, J.R., Spence, D.H. & Scudder, G.G.E. (1980). The effects of temperature on growth and development of water strider species (Heteroptera: Gerridae) of central British Columbia and implications for species packing. Canadian Journal of Zoology, 58, 1813–1820. 29 Thompson, D.J., Hassall, C., Lowe, C.D. & Watts, P.C. (2011). Field estimates of reproductive success in a model insect: behavioural surrogates are poor predictors of fitness. Ecology Letters, 14, 905– 913. Unwin, D.M. & Corbet, S.A. (2008). Wingbeat frequency, temperature and body size in bees and flies. Physiological entomology, 9, 115–121. Waage, J.K. (1986). Evidence for widespread sperm displacement ability among Zygoptera (Odonata) and the means for predicting its presence. Biological Journal of the Linnean Society, 28, 285–300. Waringer, J.A. & Humpesch, U.H. (1984). Embryonic development, larval growth and life cycle of Coenagrion puella (Odonata: Zygoptera) from an Austrian pond. Freshwater Biology, 14, 385– 399. Wilson, R.S. & Franklin, C.E. (2002). Testing the beneficial acclimation hypothesis. Trends in Ecology & Evolution, 17, 66–70. Woods, H.A. & Harrison, J.F. (2002). Interpreting rejections of the beneficial acclimation hypothesis: when is physiological plasticity adaptive? Evolution, 56, 1863–1866. Woodward, G., Perkins, D.M. & Brown, L.E. (2010). Climate change and freshwater ecosystems: impacts across multiple levels of organization. Philosophical Transactions of the Royal Society B: Biological Sciences, 365, 2093 –2106. 30 THANKS First of all, I wish to thank all the people that made the field work part of this thesis possible. Either by guiding me to the right sites or helping me in the field. Before I give any names, I would like to thank everybody who has ever submitted a sighting to a records database, published a faunistic note, carried out a survey of their local patch, written a regional atlas, deposited a specimen in a museum, or in any other way contributed to our knowledge about the earth’s biodiversity. Thank you, science would be a dark and lonely place without you. In Andalucia, a special thanks goes to Florent Prunier – because of you, I am dying to get back to the Sierra de Cazorla. And thank you Raúl Santos Quirós, Daniel Grand and José Luis Pérez Bote for pointing me in the direction of the right people and regions. Thank you Adolfo Cordero-Rivera, for inspiring me and for helping me through the necessary paperwork. Y tambien un gran agradecimiento a La Dirección General de Gestión del Medio Natural, y a la Consejería de Medio Ambiente, Junta de Andalucia. Estoy muy agradecido. In Catalonia, I am forever indebted to Oxygastra, the odonate study group of Catalonia. You are doing an amazing work in documenting the dragonflies and damselflies of this spectacular region. I cannot fully express my gratitude towards the García family. Josep, Assumpta, David and Marta – thank you for your hospitality, invaluable help in the field and for taking me to the Fogueres de Sant Joan. Your kindness and generousity was overwhelming. A big thanks also goes to Mike Lockwood for getting me in contact with these wonderful people and for helping me while in Catalunya. And thank you Xavier Maynou. The day we spent in Mura and the evening at your house formed one of the best days of my travels. It was nice to meet someone that shares my passion for things written in old books. In France, I would had been completely lost without the help of Cyrille Deliry, Philippe Lambret, Cedric Vanappelghem and Xavier Houard. Thank you so much Cedric and Xavier for not only guiding me to some great places for damselflies, but also to some amazing dining. I Sverige vill jag särskilt tacka Åke Hedman. Dina kunskaper om Medelpads trollsländor har varit oersättiga. Tack också till Tommy Karlsson, utan din hjälp hade jag kanske aldrig lyckats få tag på den gäckande griptångsflicksländan eller fått se det vackra Appeldalen. 31 In my search for phenological data, I am grateful for the help of Vincent J. Kalkman, Maarten De Groot, Klaus-Jürgen Conze, Holger Hunger, Roberto Sindaco, Xavier Maynou and the Oxygastra team, David Molina and Ximo Baixeras in Valencia, Adolfo Cordero-Rivera and Frank Suhling. In Leuven, I had the best of times when I visited my dear collaborators Robby and Marjan. I would never had been able to complete this work without you. Also, I will never forget the night when I saw my first fire salamander, in the pitch dark and pouring rain in the Meerdaalbos. And thank you, Lieven and Janne. It was great to hang out at the department together with you Thank you Jochen Rodenkirchen for your beautiful photographs. And a big thanks to Richard Askew and Peder Skou for allowing me to use the maps from Richard’s book. TACK Mest av allt vill jag tacka Frank. Du har varit den bästa handledare jag kunnat tänka mig. Det första ord jag kommer att tänka på är förtroende. Oavsett om jag velat lägga ner ett par månader på att gräva efter slovensk litteratur om trollsländor eller stånga min panna blodig mot någon obegriplig metereologisk databas så har du alltid litat på mig. Du har alltid vetat att när tiden väl var mogen, så skulle jag skärpa mig och få saker gjorda. Och så fort mina irrfärder varit på väg att krascha ner i en mörk grop så har du funnits där och styrt mig på rätt spår igen genom att prata eller skriva några kärnfulla meningar som förklarar hur allt hänger ihop. Det, för mig, är handledning när den är som bäst. Att forska tillsammans med dig har varit bland det roligaste jag gjort. Tack Pelle, för att du alltid hjälpt mig när jag kommit instormande på ditt kontor efter att motvilligt tvingats inse att jag inte klarar allting själv. Jag önskar bara att jag hade haft vett att utnyttjat dina fantastiska kunskaper ännu mer. Och Anders Nilsson! Var ska jag ens börja? Du är en av de viktigaste orsakerna till att jag över huvud taget vill forska. Ända sedan första kursen på biologprogrammet har jag sett upp till din ständiga nyfikenhet. Varenda 32 gång jag klivit in på ditt kontor har jag vetat att du hittat någon ny tråd att nysta i. Jag har heller aldrig mött någon med din känsla för text. Du behandlar ett korrektur med samma känsla som Guy Clark behandlar sin gitarr. Tack Martin för den tid som vi överlappade varandra i forskargruppen. Jag var verkligen lyckligt lottad som fick dela kontor med dig under denna känsliga inkuberingstid. Tack för att du gjorde mig peppad på R och introducerade mig för Zotero – du är en matchmaker av rang. Helena, det har varit fruktansvärt roligt att göra fältjobb, prata text och forskning med dig. Jag undrar om jag någonsin kommer att bli så strukturerad som du. And also to you, Owen – big thanks for helping me with the writing, and for keeping me good company at the UFC games På EMG vill jag annars särskilt tacka Folmer och Göran, för goda samtal om forskning. Tack Katarina och Stefan för att ni gjort det roligt att vara fikarummet – och för att ni sett till att jag inte blivit för stor på mig. Tack Birgitta, Mattias och Ingrid för att ni stått ut med mitt obefintliga minne för hur man fyller i en blankett. Och tack så mycket till Szymon, Annelie and Andreas för all hjälp på plan 2! Szymon, you deserve a whole page of thanks all by yourself. You have been such a good friend and collaborator throughout these years. I hope you will be able to do precisely what you want for the rest of your life. And thank you Melanie for being such a generally awesome person. Tack till Norrlands Entomologiska Förening för att ni hjälpt mig att behålla perspektiven på tillvaron. Många skulle kanske säga att mina flicksländor var obskyra så de räcker och blir över. Men ni i NEF förstår att dessa bara är ett populistiskt kuttersmycke i kampen om forskningsbidragen, och att det som verkligen betyder något här i världen är alla slemsvampbaggar, bastflugor, smalbin, barkgnagare, broksteklar och andra skörvar som förtjänar att tilltalas vid sina rätta namn. Så tack Andreas, Folke, Sven, Ylva, Roger, Per, Nils, Benedicte, Sara och Magnus. Och särskilt tack till Johannes och Anders, som en gång i världen tände gnistan i mig genom att hålla i världens bästa kurs. Don't let the bastards get you down! 33 En liknande och minst lika viktig roll har Svenska Malaisefälleprojektet spelat både före och under min doktorandtid. Utan er hade jag inte stått här idag. Särskilt tack till Kajsa och Dave för att ni trodde på mig! Och tack Julia, Sibylle, Pelle, Mattias och Svante för en episk sommar. På Naturhistoriska Riksmuseet vill jag särskilt tacka Kevin och Julio, som inspirerat mig mer än ni kan ana. Tack också till Niklas, Rasmus, Hege, Bert och Yngve som fått mig att känna mig välkommen på denna plats som fascinerat mig sedan jag var barn. Men jag ska heller inte glömma fåglarna, och öarna. På Stora Fjäderägg vill jag tacka Lars och Niklas för att ni gav mig en plats att ständigt längta tillbaka till. Och tack till båtkonsortiet och Johan Forssell för att ni såg till att Fjäderägg fick konkurrens i form av Holmögadd, Lövöbådan, Svartlass, Måkläppen och Tärnören. Tack Jocke för alla kortare eller längre turer, i med- eller motgång, höst eller vår, storm eller stiltje, till stora eller små öar, och med mycket eller lite fokus på fåglarna. Jag kommer sakna detta mycket. I Umeå – ni är för många och ni har betytt för mycket för att jag ska kunna göra det rättvisa här. Jag tänker ta det i någon form av kronologisk ordning, som ett monument över mina sju år i Umeå. Tack Lovisa för att du fick mig att vilja flytta till Umeå, och för att jag fick bli din och dina vänners vän. Tack Anna Östman för att du är ett geni. Kjell Wågberg – jag kommer alltid att tänka på dig och ditt avundsvärda liv när jag har viktiga beslut att fatta. Tack Erik och Jenny för att ni alltid haft plats för mig i ert hem och i era liv. Mats och Ingeborg – tack för att Fågelstigen fick bli som ett andra hem för mig. Och Stina, Nina, Viktor Wennberg, Johanna Helander, Johanna Eriksson, Anna Philipsson, Anna Fransson, Kolbjörn, Laura Zagar, Jonas Gidlund och Ola Norlander – ni är de finaste vännerna som finns. Tack Micke från Lund, Pernilla Berglund, Maxime, Micke Data, Oskar Franklin, Martina Nygren, Carl Åkerlund och alla ni som gjort Umeå till en plats där jag bara behövt lämna lägenheten för att träffa någon som gör mig glad. Tack Björn för att det var så great att bo med dig, maxa bandbredden, trasha soffor och lyssna på Slayer, eller bara äta frukost i tystnad. Och tack Erik för det fina år vi fick bo tillsammans. Tack Smilla för all den värme och ro du gav varenda dag som jag fick vara med dig. Tack Pelle, Oscar, Mattias, Anna och Lovisa för kryssfezterna. Tack Krutröken för sommarkvällarna. Tack Ola och Lotta för värmen både på och utanför plan. 34 Och Pontus, Ida-Maria, David och Andreas – alla sitcoms på tv är bara bleka kopior av våra kaffepauser i KBC-fiket. I Uppsala, Portland och Ottawa – tack Björn för djur-, växt- och statistikpeppen. I Stockholm – tack Jörgen och Klara för stughäng, natur- & kultursamtal, fladdermuslyssning och vinterpromenader. Tack Anthon och Caroline, jag älskar att komma hem till er och se vad som hänt sedan jag var där senast. I Göteborg – tack Johan, Hanna, Erik, Helena, Alexander, Jakob och PerAnders för att ni sett till att jag inte behövt ångra mitt beslut att lämna Umeå. Tack också Linnéa och Per för de första av många (hoppas jag!) road trips och katthäng. Tack Lovisa för att du alltid står upp för neanderthalsmänniskornas rättigheter. Och tack Siska för att du räddade mig från garanterad katastrof under slutfasen av avhandlingsarbetet! I Åsaka – tack Eva och Åke för att ni är världens snällaste. Jag menar det. Slutligen tack till min familj. Mamma, Pappa – ni är bäst. Lisa, Fia, Johan, Jesper, Göran – älskar er. Tova, Sigge, Vega, Lilith – tack för att ni finns. Och Malin. 35