Behavioural, physiological and morphological correlates of life-history in killifishes − a macroevolutionary approach

Behavioural, physiological and morphological correlates of life- history in killifishes –

a macroevolutionary approach

Simon Eckerström-Liedholm

Simon Eckerström-Liedholm Behavioural, physiological and morphological correlates of life-history in killifishes – a macroevolutionary approach

Doctoral Thesis in Ethology at Stockholm University, Sweden 2019

Department of Zoology

ISBN 978-91-7797-759-9

Simon Eckerström-Liedholm

Behavioural, physiological and morphological correlates of life-history in killifishes − a

macroevolutionary approach

Simon Eckerström-Liedholm

Academic dissertation for the Degree of Doctor of Philosophy in Ethology at Stockholm University to be publicly defended on Friday 6 September 2019 at 10.00 in Vivi Täckholmsalen (Q-salen), NPQ-huset, Svante Arrhenius väg 20.

Abstract

Life-histories commonly evolve along a continuum from short-lived and fecund, to long-lived and less fecund. Because life-history traits are mostly components of reproduction and survival, understanding the causes and consequences of life- history variation is at the core of evolutionary biology. This thesis aims to identify what other key traits (e.g. behavioural, physiological and morphological traits) covary with life-history, and why. Numerous hypotheses describe how life-history might be associated with other traits, with life-history trade-offs often considered to be a primary driver of any such relationships. For example, since resources are limited, increased investment in one trait must lead to decreased investment in one or several other traits, all else equal. Hypotheses on the relationship between life-history and other traits have been tested in many studies, but empirical studies in controlled experimental settings are rare. In this thesis I explore how behaviour, physiology and morphology relate to variation along the life-history continuum from fast to slow, in a system with substantial variation in life-history traits - the killifishes.

I began by exploring the patterns of egg to body size allometry in killifishes (Paper I), where species with faster life- histories showed indications of constraints on the independent evolution of egg size and body size. Furthermore, I found evidence of differences in variance and in the rates of evolution of egg size and body size across species, potentially caused by the colonisation of ephemeral habitats, which could have selected for adaptations that lead to differences in size.

I then performed a comparative common garden study (Paper II) of the pace-of-life syndrome hypothesis, which predicts that species with fast life-histories should take larger risks in order to maintain their increased reproductive rate. I obtained data on risk taking behaviours, including movement, tendency to enter an open environment, and aggressiveness, in addition to metabolic rate, for 20 species of killifish, with multiple replicates per species. The results indicated trait dependent associations with life-history, where aggression seemed to correlate positively with speed of life-history, in congruence with our prediction.

Next, my colleagues and I assessed the association between life-history and sexual selection (Paper III), in order to determine if investment in secondary sexual traits might be traded off against survival in killifish. Fin size was found to be negatively associated with escape performance in a simulated predator attack, suggesting survival costs for individuals with large fins. Importantly, fin size was also positively associated with the speed of life-history, supporting the hypothesis that costs to survival probability is lower in fast-living species.

Lastly, I tested the hypothesized negative covariation between relative brain size and speed of life-history, by collecting and analysing brain size measurements for 21 species of killifish (Paper IV). Surprisingly, a positive relationship between speed of life-history and relative brain size was found for adults, although juveniles did not differ in relative brain size. This implies at least one of two things: either there is no need to trade off brain size with life-history since resource acquisition is higher, or brain size and life-history are traded-off with other traits.

In conclusion, I show that previously found trade-offs between life-history and investment in other costly traits are only sometimes present, when tested in a system with substantial divergences in the speed of life-history. I also provide evidence for a trait dependent association between life-history and among species differences in risk-taking and metabolic rate.

Keywords: brain size evolution, comparative analysis, life-history continuum, risk-taking behaviour, sexual selection,

Stockholm 2019

http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-169666

ISBN 978-91-7797-759-9 ISBN 978-91-7797-760-5

Department of Zoology

Stockholm University, 106 91 Stockholm

BEHAVIOURAL, PHYSIOLOGICAL AND MORPHOLOGICAL CORRELATES OF LIFE-HISTORY IN KILLIFISHES –

A MACROEVOLUTIONARY APPROACH

Simon Eckerström-Liedholm

Behavioural, physiological and morphological correlates of life-history in killifishes –

a macroevolutionary approach

Simon Eckerström-Liedholm

©Simon Eckerström-Liedholm, Stockholm University 2019 ISBN print 978-91-7797-759-9

ISBN PDF 978-91-7797-760-5

Cover photo: a male Nematolebias whitei. Copyright of Simon Eckerström-Liedholm.

Printed in Sweden by Universitetsservice US-AB, Stockholm 2019

1

The thesis is based on the following articles, which are referred to in the text by their Roman numerals:

I Eckerström-Liedholm, S, W Sowersby, A Gonzalez-Voyer, and B Rogell. (2017).

Time-limited environments affect the evolution of egg–body size allometry.

Evolution., 71, 1900–1910.

II Eckerström-Liedholm, S, W Sowersby, S Morozov, W van der Bijl, P Rowinski, A Gonzalez-Voyer and B Rogell. Macroevolutionary evidence suggests trait-dependent coevolution between behaviour and life-history. - Manuscript.

III Sowersby W, S Eckerström-Liedholm, P Rowinski, J Balogh, S Eiler, J Upstone, A Gonzalez-Voyer and B Rogell. Costly sexual ornaments coevolve with fast life- histories in killifishes. - Manuscript.

IV Eckerström-Liedholm, S, W Sowersby, A Kotrschal, J Näslund, P Rowinski, A Gonzalez-Voyer and B Rogell. Fast life-histories are associated with larger brain size in killifishes. - Manuscript.

Candidate contributions to thesis articles*

* Contribution Explanation

Minor: contributed in some way, but contribution was limited.

Significant: provided a significant contribution to the work.

Substantial: took the lead role and performed the majority of the work.

I II III IV

Conceived the study Significant Significant Minor Substantial

Designed the study Substantial Substantial Minor Significant

Collected the data Significant Substantial Minor Substantial

Analysed the data Substantial Substantial Substantial Substantial

Manuscript preparation Substantial Substantial Significant Substantial

2

CONTENTS

INTRODUCTION ... 3

Life-history evolution ... 3

The Pace-of-Life Syndrome hypothesis ... 4

Links between life-history and morphological traits ... 4

Brain size ... 4

Offspring size ... 5

Sexual ornaments ... 5

OBJECTIVES ... 6

STUDY SYSTEM ... 6

METHODS ... 6

RESULTS AND CONCLUSIONS ... 8

Paper I ... 8

Paper II ... 9

Paper III ... 10

Paper IV ... 11

Concluding remarks ... 12

REFERENCES ... 13

SVENSK SAMMANFATTNING ... 17

ACKNOWLEDGEMENTS ... 18

3 INTRODUCTION

Life-history evolution

Life-history theory is a framework for studying and describing the magnitude, timing and duration of key life cycle events (Roff 1992; Stearns 1992; Flatt & Heyland 2011). The fitness components of any individual include a combination of life- history traits such as growth rate, fecundity and longevity (McGraw & Caswell 1996), therefore the causes and consequences of life-history variation are understandably an important and central topic in evolutionary biology. Even though the estimated heritability of life-history traits is relatively low compared to other types of traits, likely due to the substantial phenotypic variation in life-history traits caused by the often inherent randomness of extrinsic mortality, there is ample genetic variation for evolution to act on (Houle 1992). Selection for different life-history trait combinations have given rise to massive differences in life-history strategies among different species, and biologists have been working on uncovering the evolutionary mechanisms that have given rise to this variation in life-history strategies for decades (Stearns 1976). The results of these empirical and theoretical studies shed light on questions such as how ageing evolves (Medawar 1952; Williams 1957), the optimal number and size of eggs (Lack 1947; Smith & Fretwell 1974), the timing of reproductive events and investment (Gustafsson & Pärt 1990), optimal growth rate (Lee et al. 2013), and the existence of suboptimal life-histories due to sexually antagonistic coevolution (Rice & Holland 1997).

Trade-offs are essential to life-history evolution, since all organisms are constrained by the finite resources in their environments. In other words; no organism as of yet has been able to embody a so-called Darwinian Demon (Law 1979), i.e an organism that becomes sexually mature the moment it is born, has an infinite reproductive output and lives forever. Thus, the optimal allocation of finite resources will represent a trade-off between different traits. Empirical studies have uncovered several life-history trade-offs, such as the trade-off between reproduction and longevity (Zwaan et al. 1995; Xie & Klerks 2004; Kalra & Parkash 2014), between current and future reproductive investment (Stearns 1989) and between growth rate and longevity (Lee et al. 2013). A likely cause of the negative relationships between traits is the pleiotropic effects of certain genes that, for instance, direct resources to reproduction at the expense of somatic maintenance (Williams 1957). What are the factors that influence the optimal trade-off between different life-history traits? In the case of the trade-off between longevity, which is to a large extent determined by somatic maintenance in many species, and reproduction, the age specific mortality rate seems to play an important role (Charlesworth 1980; Reznick et al. 2002). Increased adult mortality for instance, all else equal, should lead to an increased investment in current reproduction, due to the reduction of Residual Reproductive Value caused by the lower survival probability (Hirshfield & Tinkle 1975; Promislow & Harvey 1990; Ricklefs 2000).

The life-history trade-offs mentioned above often result in life-history traits aligning with each other to form a general axis of variation from slow to fast, called the life-history continuum, where typically slow species live longer, grow slower, mature later and are less fecund, while fast species possess the opposite suite of traits. Empirical studies among species have provided evidence of such a continuum across a broad range of taxa, for example: generation time and age at onset of senescence correlates positively in mammals and birds (Jones et al. 2008), fecundity correlates positively with mortality rate and life span in mammals, birds and fish (Read & Harvey 1989; Promislow & Harvey 1990; Gunderson 1997; Ricklefs 2000) and fecundity and age at maturity correlate negatively in insects (Blackburn 1991). Within species however the trade-offs between key life- history traits are less clear, and it is common to find for instance that fecundity and life-span can correlate positively in some individuals (Smith 1981; Olijnyk & Nelson 2013). Such results could be taken as evidence against the presence of life-history trade-offs, but a more plausible explanation is that among individual differences in resource acquisition capacity are obscuring the trade-offs that occur more generally at a macro-evolutionary scale (van Noordwijk & de Jong 1986; Stearns 1989). This

4

could partly be due to the fact that individuals of the same species often have more similar life-history strategies than individuals of different species, which makes trade-offs between strategies harder to find, and partly could be the consequence of environmental heterogeneity, or health differences, that allows some individuals to acquire a greater sum total of resources (van Noordwijk & de Jong 1986).

The Pace-of-Life Syndrome hypothesis

Since differences in life-histories are generally associated with differences in resource allocation, it seems plausible that shifts in life-history could influence other traits associated with life-history, such as mate searching, foraging and even metabolic rate. According to a relatively recent hypothesis, the Pace-of-Life Syndrome (POLS) hypothesis (Réale et al. 2010; Dammhahn et al. 2018), the pace of an organism's life-history ought to covary with behavioural and physiological traits, both within species and across species. For instance; selection that favours current reproductive investment over future reproductive investment, might lead to selection for a higher metabolic rate (physiology), and risk-taking in terms of more intense foraging (behaviour), in order to sustain the increased energetic demands on offspring production and mate searching. Furthermore, with increased early reproductive investment rather than late, the individual has less to lose in terms of fitness by accepting a certain risk, since they will att any point in time as an adult have produced a larger fraction of their reproductive output (Clark 1994). Thus, the POLS hypothesis predicts that fast life-history strategies should co-occur with risk-taking behaviour and higher metabolic rates. However, to date, most empirical studies on the POLS hypothesis have been conducted on a within species scale (e.g. Ward et al. 2004; Mas-Muñoz et al. 2011; Le Galliard et al. 2013; White et al. 2016, but see Careau et al.

2009; Sol et al. 2018), which have yielded mixed support for the hypothesis. There is clearly a knowledge gap when it comes to among species tests of the POLS hypothesis, where one might possibly expect stronger effect sizes due to the fact that life- history traits often vary to a greater extent among species (van Noordwijk & de Jong 1986).

Links between life-history and morphological traits

In addition to the possible relationships with behaviour and physiology, there is theory and empirical evidence that suggests the existence of associations between the pace of life-histories and morphological traits such as offspring size, brain size and sexual ornaments, which will be described below.

Brain size

The brain is one of the most energetically costly organs across many taxa, measured as energy consumption per unit tissue (i.e.

specific metabolic rate) (Mink et al. 1981; Wang et al. 2012). The Expensive Brain hypothesis (Isler & van Schaik 2009a) predicts that increased allocation into brain size must be paid for somehow, for example, by allocating more energy towards developing and maintaining the brain at the expense of other traits (as part of the Energy trade-off Hypothesis; Isler & van Schaik 2006a). Furthermore, numerous other hypotheses predict that relative brain size ought to correlate negatively with pace of life-history for various reasons, such as the Cognitive Buffer hypothesis, the Delayed Benefits hypothesis, and the Brain Malnutrition Risk hypothesis, reviewed in Deaner et al. (2003). The predictions that follow from the Expensive Brain hypothesis has been supported by empirical tests in a range of taxa (van Schaik & Deaner 2003; Isler & van Schaik 2006b, 2009b, a; Barrickman et al. 2008; González-Lagos et al. 2010; Kotrschal et al. 2013; Gonzalez-Voyer et al. 2016; Liao et al.

2016; Sol et al. 2016; Yu et al. 2018, but see Isler & van Schaik 2006a). Hence the inverse relationship between pace of life-

5

history and brain size seems to be a general pattern across vertebrates. However, most studies on this relationship have been conducted on study systems where life-history covaries with body size. The rate of evolution of brain size is often lower than the rate of evolution for body size (Gonzalez-Voyer et al. 2009; Smaers et al. 2012), and such lags could lead to brain size being above or below the evolutionary allometry simply due to selection on body size. Thus, if selection on life-history is correlated with selection on body size, it will appear as if life-history is correlated with relative brain size, even though no causal relationship exists. To what extent these considerations affect the conclusions that can be drawn from comparative studies on the relationship between the pace of life-history and relative brain size is unclear. What is clear however, is that comparative and experimental studies on study systems where life-history variation is not confounded by body size are critical for our understanding of this important evolutionary pattern.

Offspring size

Species with fast life-histories favour investment into current rather than future reproductive events, which means that they require more resources for reproduction per unit time than species with slow life-histories. This larger investment is subject to a trade-off, between resources being spent on larger (often higher quality) offspring, or a larger number of offspring, or a combination of these traits (Smith & Fretwell 1974; Brockelman 1975; Lloyd 1987). Furthermore, since selection often favours high growth rates and early maturation in species with fast life-histories, and the starting size of the offspring will have a large influence on the time it takes to reach maturity, the size of the offspring might be constrained towards larger values compared to slow species (Morita et al. 2009). Many ecological factors could potentially influence the offspring size - offspring number trade-off, such as temperature and the productivity of the habitat (Steigenga & Fischer 2007; Robertson &

Collin 2015). Furthermore, if there is selection to produce a large number of offspring, for instance in order to be able to sufficiently spread risks in unpredictable environments, there might be a lower bound on the number of offspring produced.

Sexual ornaments

Sexual selection theory predicts that secondary sexual traits should incur costs to the bearer, such as costs imposed by increased risk of predation or increased aggression from competitors (Kotiaho 2001). There is some support for these theoretical predictions within species; for instance, the size and conspicuousness of secondary sexual ornaments, such as colouration or elaborate fins and feathers, have been associated with increased risk of predation (Endler 1980; Rudh et al.

2011; Hernandez-Jimenez & Rios-Cardenas 2012). However, differences in the pace of life-history, and consequently the relative investment in current over future reproduction, will affect the fitness costs of predation, which in turn could affect the optimal level of investment in a secondary sexual trait (Andersson 1982). This means that even though the benefit of the sexually selected trait stays the same, a reduction in the costliness of that trait might affect the cost-benefit ratio and lead to selection for higher investment in the trait. To see why this is so, consider the fact that the costs and benefits cancel out each other at the fitness optimum of signalling for the specific trait. If the costs are reduced while the benefits stays the same, the fitness optimum will shift upwards until it reaches a new equilibrium where costs balances out the benefits again. However, since many studies only measure the costs in terms of energy expenditure, there is a scarcity of knowledge about what the direct fitness costs of sexually selected traits are (Kotiaho 2001).

6 OBJECTIVES

The broad objective of this thesis was to better understand whether, and if so, how life-history traits coevolve with behavioural, physiological and morphological traits on a macroevolutionary scale. In Paper I my colleagues and I analysed data on body size and egg size for a large number of killifish species, and estimated egg-body size allometries, rates of evolution and fecundity. In Paper II we obtained behavioural and physiological measures for 20 species of killifish, in order to answer questions about the integration of life-history with physiology and different dimensions of behaviour. In Paper III we examined the relationship between life-history, fin size, and swimming performance, shedding light on the possible trade-off between investment into a sexually selected trait (ornamental fins) and predator escape performance. Finally, in Paper IV, we estimated differences in relative brain size of adult and juvenile killifish of different species, to test hypotheses about energetic trade-offs between neural investment and reproductive investment.

STUDY SYSTEM

In this thesis I have used different species of killifish (Figure 1 A and B, Figure 2), native to large parts of Africa and South America, to test various hypotheses related to the integration of life-history traits and behavioural, physiological and morphological traits. The main motivation for using killifishes is twofold. Firstly, there is substantial variation in life-history among killifish species of the suborder Aplocheiloidei, where some species have adapted to life in ephemeral habitats through the production of drought-resistant eggs, fast development, high investment in current reproductive events and early senescence (Genade et al. 2005; Blažek et al. 2013; Furness et al. 2015; Berois et al. 2016; Furness 2016), while others live in more stable habitats and have slower life-histories. For Paper I, III and IV, we used the ability to enter diapause stage II, which is generally the longest diapause stage in which individuals of some species can survive for years, as a proxy for fast (or so called ‘annual’) life-histories. These life-history strategies have most likely evolved at least five times independently (Furness et al. 2015), giving us an opportunity to test hypotheses on what is in a sense a replicated natural experiment. The fact that the effect sizes of these life-history differences seem to be large increases the chances of detecting the hypothesized relationships between life-history and other traits, given that hypotheses are correct. In Paper II however we constructed, using a principal component analysis, a continuous life-history axis from fast to slow, using data from our lab on growth rates, reproductive rates, and maturation time. The second reason for using killifish as a model system is that even the slow-living

‘non-annual’ killifish live short lives (compared to humans), which makes it possible to collect substantial amounts of data during a relatively short time period.

METHODS

In Paper I we obtained data for egg size and body size from a curated database, where researchers and dedicated aquarists have contributed valuable and impressive amounts of data, and modelled allometric slopes and rates of evolution for body size and egg size. For Paper II, III and IV, and for the analysis of fecundity in Paper I, we used killifish that were reared in a common garden setting in our lab. We obtained our study subjects from aquarists that had collected them in the wild, although for some species it was only possible to obtain individuals from strains that had been in the aquarium trade for several years. It should be noted though that the strains were not actively bred for certain characteristics. In Paper II we measured activity, boldness and aggression using either tracking software (Figure 1 C) or manual measurements, and metabolic rate using an intermittent-flow respirometer.

7 A

B C

Figure 1. A; a young adult male of the species Nothobranchius guentheri, native to Tanzania, Africa. B; a fully developed egg of a killifish, ready to hatch. Many killifish have the ability to enter a third and final stage of diapause (diapause III) just before hatching, when the embryo is fully developed. C; an adult killifish in one of the open field test arenas, where activity was tracked (ellipse) using a camera mounted above the arena. A tracking software was used to obtain the exact position of the fish during the trial.

All measurements were, to the extent possible, taken from all individuals, except for aggression that was only measured in males. We then tested for the predicted associations between these traits and life-history. Paper III included geometric morphometric measures of fin shapes and body shapes, and swimming performance measures such as escape speed, that were obtained both from fish in our lab, and from published high-quality illustrations of killifish. For Paper IV we collected whole brains (fixed in formalin) from juvenile and adult fishes, and compared the relative brain sizes of fish with fast and slow life- histories. The analyses for the association between brain size and life-history was performed both at the level of the whole brain, and in terms of the relative sizes of different sub-regions. Virtually all statistical models were performed using Bayesian models. The main reason for using this approach was the flexibility of the method, which allowed us to correct for phylogenetic effects while simultaneously including multiple individuals per species.

8

Figure 2. Phylogenetic relationships of the species included in Paper II. Fast-living species are coloured red and slow-living species are coloured blue (note however that this division of fast- and slow-living species was not used in Paper II). This tree represents a pruned version of the tree produced by Furness et al. (2015), created with auto-correlated rates and soft bound constraints, with the species Nothobranchius kadleci and Ophtalmolebias constanciae added to their corresponding sister taxa prior to pruning.

RESULTS AND CONCLUSIONS

Paper I

Previous studies have indicated that egg size might be related to the pace of life-history (e.g. Morita et al. 2009); thus in Paper I we analysed the relationship between body size and egg size for 189 species of either fast-living (annual) and slow-living (non-annual) killifish, and found a difference in the allometric slope of egg size to body size between these two groups.

Specifically, we observed that fast-living species had a steeper allometric slope than the slow living species. This difference, we hypothesize, was due to stronger constraints on the independent evolution of body size and egg size in fast living species, since small eggs are likely not favoured when a large body size has to be obtained quickly, and very large eggs will negatively affect fecundity (i.e. reproductive rate), which might be suboptimal due to the fact that fast-living species are selected for high reproductive rates. Thus, we found that adults of large species seem to be forced to produce large eggs, possibly in order to complete development in time, while small species produce relatively small eggs, possibly due to constraints on fecundity. The fast-living and slow-living species did not differ significantly in terms of average body size (Figure 3) or egg size.

Furthermore, we found significant differences in the variance and the rate of phenotypic evolution for both egg size and body size, where the variance and the rate of phenotypic evolution was larger in fast-living species. This difference in the rate of evolution could be related to differences in habitat composition between fast-living and slow-living killifish, where it is conceivable that the ephemeral habitats of fast-living killifish offer a larger breadth of different niches to fill, which could explain the larger phenotypic variance as well as the higher rates of phenotypic evolution in fast-living species (Cooney et al.

2017).

Notholebias minimus Nematolebias whitei

Austrolebias nigripinnis Austrolebias wolterstorffi Ophtalmolebias constanciae Rivulus Cynodonichthys fuscolineatus

Pterolebias longipinnis Gnatholebias zonatus Rivulus Anablepsoides iridescens Rivulus Anablepsoides amphoreus Pachypanchax playfairii Callopanchax toddi Aphyosemion splendopleure Aphyosemion striatum

Nothobranchius guentheri Nothobranchius kadleci Fundulopanchax filamentosus Fundulopanchax cinnamomeus Fundulopanchax scheeli Scriptaphyosemion cauveti

Fast Slow

9

Figure 3. Log10-transformed body size (total length) of fast-living (annual) and slow-living (non-annual) species of killifish. The box-plots represent median (black bar), 1st and 3rd quartiles (box), highest and lowest value within 1.5 * the interquartile range (whiskers), and data points outside 1.5 * the interquartile range (dots). Gray violin plots show smoothed kernel density of the data points. The data was obtained from the database KilliData for Paper I.

Paper II

Based largely on the Pace-of-Life Syndrome hypothesis (Réale et al. 2010; Dammhahn et al. 2018) applied at an among species scale, we predicted that the pace of life-history among killifish species ought to be positively associated with proactive and risk-taking behavioural traits, such as being bold, active and aggressive, and that fast-living species ought to have higher metabolic rates. We tested these predictions in Paper II, but interestingly could not find any clear support for the hypothesis.

In other words, life-history, behaviour and metabolic rate did not seem to covary in our study system, with the exception of a significant positive association between aggression and pace of life-history (Figure 4). Neither did we find any substantial among species covariation among the different behavioural assays and metabolic rate. We observed large intraclass correlations for most measured traits, indicating that individuals from different species differed consistently from each other in behaviour and metabolic rate. Such differences among species is a prerequisite for finding effects on the among species scale, and indicates that if there were biologically meaningful effects, we should be able to detect them.

0.6 0.8 1.0 1.2

Non−annual Annual

Life−history strategy Log10 body size (cm)

10

Figure 4. Relationships between life-history and different variables measured in Paper II. Points represent species means and error bars represent standard error of the mean. PC1 of life-history was obtained by extracting the first principal component of a PCA on log10- transformed time until sexual maturity in males (days), log10-transformed growth rate (cm/day), and reproductive rate (eggs/female/month). The life-history variables were collected from several separate studies in our lab, while activity, boldness, metabolic rate and aggression were measured in Paper II. Activity is represented by log10-transformed mean speed in an open field test, boldness is represented by log10-transformed, mean centered and inverted latency to emerge +1 (s) in an emergence test, metabolic rate is represented by log10-transformed standard metabolic rate (mg O2/kg/h), and aggression is represented by log10-transformed number of attacks +1 in a mirror test.

Paper III

In Paper III we collected data on fin morphology and swimming performance for different species of killifish, in order to test hypotheses related to trade-offs between survival and reproductive success. We found that males of fast-living species had larger anal and dorsal fins. We also found that species with longer and more elaborate fins performed worse in one test of swimming performance, escape speed, that was devised to simulate a predator attack, although the ability to turn quickly did not seem to depend on fin morphology (Figure 5). These findings thus suggest that there is a trade-off between survival and investment into current reproduction, and that a shift towards a faster life-history in killifish could have shifted the optimal level of investment into secondary sexual traits towards larger and more elaborate fins. An alternative explanation is that female choice differ between fast-living and slow-living species of killifish. This is certainly possible, given the fact that female choice is likely an important component in what is driving the investment into elaborate fins among males (Andersson

& Simmons 2006), although it would then have to be selected for in fast-living species for some reason. Differences in operational sex ratio could for instance lead to greater choosiness among females (Berglund 1994), and it would therefore be useful for future studies to test whether the operational sex ratio differs consistently between fast-living and slow-living

PC1 of life−history

Activity

−2 0 1 2 3

−0.5 0.0 0.5 1.0

A

PC1 of life−history

Boldness

−2 0 1 2 3

−1.5

−1.0

−0.5 0.0 0.5 1.0 1.5

B

PC1 of life−history

Metabolic rate

−2 0 1 2 3

1.3 1.4 1.5 1.6 1.7

C

PC1 of life−history

Aggression

−2 0 1 2 3

−0.2 0.0 0.2 0.4 0.6 0.8 1.0

D

11

killifish species. In summary however, our results suggest that the trade-off between current and future reproduction can drive the evolution of costly secondary sexual traits.

Figure 5. Relationship between escape swimming speed and fin shape variation explained on RW2 (relative warp 2, the component that explained most variation in fin morphology), with lower values of RW2 representing more elongated fins. Fast-living species are represented as red data points, slow-living species are represented as blue data points, males are represented by the solid trend-line and the triangles, and females are represented by the dashed trend-line and the circles.

Paper IV

According to The Expensive Brain hypothesis (Isler & van Schaik 2009a), and several other hypotheses (Deaner et al. 2003), the pace of life-history ought to be causally linked to brain size corrected for body size, i.e. relative brain size, for a certain species. Specifically, according to the hypothesis, we should expect a reduction in relative brain size among species with fast life-histories. To test this, we collected brain volume and body weight measures for juveniles and adults across multiple species of killifish, and compared their relative brain sizes. We found that adult brain sizes differed significantly, but in the opposite direction of what we predicted, with fast-living species having larger relative brain sizes (Figure 6), while fast- and slow-living juvenile species did not differ in terms of relative brain size. There are a number of plausible explanations for this somewhat surprising result. Firstly, even though numerous other studies have found the predicted relationship between life- history and relative brain size, there has almost always been a correlation between body size and life-history in those study systems, which could lead to a spurious relationship between life-history and relative brain size (Smaers et al. 2012), if the rate of evolution differs between brain size and body size (Gonzalez-Voyer et al. 2009; Smaers et al. 2012). Luckily, body size and life-history does not to covary in the killifish system (see Figure 1 and Paper IV). Another potential reason for the difference between the results in our study and the results in other studies, is that fast-living species might have a substantially higher energy acquisition rate, which gives them the opportunity to invest in both increased reproduction and increased relative brain size.

12

Figure 6. Allometric slopes of total brain volume to body weight for adult killifish, where brain volume is a composite measure of the volumes of the major subregions added together. Each data point represents an individual, and each line represents an allometric slope for a species calculated using separate regression models. The two life-history strategies, that were determined by the presence or absence of the diapause II stage, are represented by blue (slow) and red (fast) colours, and the two sexes are represented by circles (females) and triangles (males).

Concluding remarks

In this thesis I have tested multiple different, but related, hypotheses about how life-history trade-offs influence behaviour, physiology and morphology. Generally, our results seem to show both support, lack of support, and in some cases completely contradict, the predictions that follow from these hypotheses. For instance, we find the opposite of a trade-off between investment in brain size and investment in current reproduction, and we do not find an association between life-history and risk-taking behaviour, despite clear predictions. The lack of a body size - life-history correlation in the killifish system makes studies on killifish valuable but also hard to compare with studies in other systems. Future studies testing similar questions in study systems where body size does not covary with life-history, if possible, would thus likely be informative in determining what role body size has on the associations between life-history and other traits.

The studies in this thesis are all correlational, meaning that they are exploring preexisting differences in traits shaped by evolutionary forces over millions of years. This means causality is difficult to determine, although it can generally be said that unless there is a very obvious alternative explanation, the presence of correlations among traits makes it ever so slightly more likely that they are causally related in some way. In other words; observing a correlation ought to shift your belief towards there being a causal relationship a bit, although by how much depends on how plausible the alternative hypotheses are. Still, the main type of criticism against comparative studies is a valid one, which points to the possibility of unknown third variables causing the appearance of a causal relationship between the measured variables. In the case of our model system, the killifish, there is a possibility of a correlation between life-history and predation, where such a hypothetical difference in predation could have affected the expression of the other traits that we measured. Due to logistic constraints related to accurately estimating predation in remote habitats spread out across multiple countries on two continents, we have not been able to estimate predation in the natural habitats. Previous work studying the effects of predation on the spatial distribution of prey has

0.25 0.50 0.75 1.00

-0.8 -0.4 0.0

Log

body weight (g) L og

tot al bra in vol um e (m m

)

Life-history strategy

Fast Slow

Sex Female Male

13

shown that predation can cause displacement of prey fish into tributaries (Gilliam & Fraser 2001). It is unclear however to what extent this observation translates into differences in terms of total predation pressure in the different habitats of killifish, accounting for avian predation for instance. Furthermore, we have analysed the presence of other killifish species in the habitats of killifish with both fast and slow life-histories (‘annual’ and ‘non-annual’) as a proxy for potential predation pressure by piscivorous fish, and we did not find any relationship (Paper II). Ultimately, the best way to answer the question is to measure predation pressure and other plausible third variables in order to determine whether it covaries with life-history.

Comparative data is often the best evidence available, but until data on third variables such as predation is available, we should interpret the results in this thesis with the aforementioned considerations in mind.

The results presented in this thesis expand the current knowledge of correlates of life-history variation, without the potentially confounding effects of body size; by for instance revealing how fast life-histories might constrain the evolution of relative offspring size. The results also point towards the difficulty of demonstrating trade-offs related to life-history that seem plausible to expect, such as between risk-taking or brain size and reproductive effort, while they at the same time provide evidence in favour of certain trade-offs, such as between escape performance and investment in ornamental fins.

REFERENCES

Andersson, M. (1982). Sexual selection, natural selection and quality advertisement. Biol. J. Linn. Soc., 17, 375–393.

Andersson, M. & Simmons, L.W. (2006). Sexual selection and mate choice. Trends Ecol. Evol., 21, 296–302.

Barrickman, N.L., Bastian, M.L., Isler, K. & van Schaik, C.P. (2008). Life history costs and benefits of encephalization: a comparative test using data from long-term studies of primates in the wild. J. Hum. Evol., 54, 568–590.

Berglund, A. (1994). The operational sex ratio influences choosiness in a pipefish. Behav. Ecol., 5, 254–258.

Berois, N., García, G. & O de Sá, R. (2016). Annual fishes: life history strategy, diversity, and evolution. Book. CRC Press, Boca Raton, Florida.

Blackburn, T.M. (1991). Evidence for a `fast-slow’ continuum of life-history traits among parasitoid Hymenoptera. Funct.

Ecol., 5, 65–74.

Blažek, R., Polačik, M. & Reichard, M. (2013). Rapid growth, early maturation and short generation time in African annual fishes. Evodevo, 4, 24.

Brockelman, W.Y. (1975). Competition, the fitness of offspring, and optimal clutch size. Am. Nat., 109, 677–699.

Careau, V., Bininda-Emonds, O.R.P., Thomas, D.W., Réale, D. & Humphries, M.M. (2009). Exploration strategies map along fast--slow metabolic and life-history continua in muroid rodents. Funct. Ecol., 23, 150–156.

Charlesworth, B. (1980). Evolution in age-structured populations. Cambridge University Press, Cambridge, New York.

Clark, C. (1994). Antipredator behavior and the asset-protection principle. Behav. Ecol., 5, 159–170.

Cooney, C.R., Bright, J.A., Capp, E.J.R., Chira, A.M., Hughes, E.C., Moody, C.J.A., et al. (2017). Mega-evolutionary dynamics of the adaptive radiation of birds. Nature, 542, 344–347.

Dammhahn, M., Dingemanse, N., Niemelä, P. & Réale, D. (2018). Pace-of-life syndromes: a framework for the adaptive integration of behaviour, physiology and life history. Behav. Ecol. Sociobiol., 72, 62.

Deaner, R.O., Barton, R.A. & van Schaik, C.P. (2003). Primate brains and life histories: Renewing the connection. In: Primate Life Histories and Socioecology. pp. 233–264.

Endler, J.A. (1980). Natural selection on color patterns in Poecilia reticulata. Evolution (N. Y)., 34, 76–91.

Flatt, T. & Heyland, A. (2011). Mechanisms of life history evolution: the genetics and physiology of life history traits and trade-offs. Oxford University Press, Oxford (United Kingdom).

14

Furness, A. (2016). The evolution of an annual life cycle in killifish: adaptation to ephemeral aquatic environments through embryonic diapause. Biol. Rev. Camb. Philos. Soc., 91, 796–812.

Furness, A., Reznick, D., Springer, M. & Meredith, R. (2015). Convergent evolution of alternative developmental trajectories associated with diapause in African and South American killifish. Proc. R. Soc. B, 282, 20142189.

Le Galliard, J., Paquet, M., Cisel, M. & Montes-Poloni, L. (2013). Personality and the pace-of-life syndrome: variation and selection on exploration, metabolism and locomotor performances. Funct. Ecol., 27, 136–144.

Genade, T., Benedetti, M., Terzibasi, E., Roncaglia, P., Valenzano, D., Cattaneo, A., et al. (2005). Annual fishes of the genus Nothobranchius as a model system for aging research. Aging Cell, 4, 223–233.

Gilliam, J. & Fraser, D. (2001). Movement in corridors: enhancement by predation threat, disturbance, and habitat structure.

Ecology, 82, 258–273.

González-Lagos, C., Sol, D. & Reader, S.M. (2010). Large-brained mammals live longer. J. Evol. Biol., 23, 1064–1074.

Gonzalez-Voyer, A., González-Suárez, M., Vilà, C. & Revilla, E. (2016). Larger brain size indirectly increases vulnerability to extinction in mammals. Evolution, 70, 1364–1375.

Gonzalez-Voyer, A., Winberg, S. & Kolm, N. (2009). Distinct evolutionary patterns of brain and body size during adaptive radiation. Evolution (N. Y)., 63, 2266–2274.

Gunderson, D.R. (1997). Trade-off between reproductive effort and adult survival in oviparous and viviparous fishes. Can. J.

Fish. Aquat. Sci., 54, 990–998.

Gustafsson, L. & Pärt, T. (1990). Acceleration of senescence in the collared flycatcher Ficedula albicollis by reproductive costs. Nature, 347, 279.

Hernandez-Jimenez, A. & Rios-Cardenas, O. (2012). Natural versus sexual selection: predation risk in relation to body size and sexual ornaments in the green swordtail. Anim. Behav., 84, 1051–1059.

Hirshfield, M. & Tinkle, D. (1975). Natural selection and the evolution of reproductive effort. Proc. Natl. Acad. Sci. U. S. A., 72, 2227–2231.

Houle, D. (1992). Comparing evolvability and variability of quantitative traits. Genetics, 130, 195–204.

Isler, K. & van Schaik, C. (2006a). Costs of encephalization: the energy trade-off hypothesis tested on birds. J. Hum. Evol., 51, 228–243.

Isler, K. & van Schaik, C.P. (2006b). Metabolic costs of brain size evolution. Biol. Lett., 2, 557–560.

Isler, K. & van Schaik, C.P. (2009a). The expensive brain: a framework for explaining evolutionary changes in brain size. J.

Hum. Evol., 57, 392–400.

Isler, K. & van Schaik, C.P. (2009b). Why are there so few smart mammals (but so many smart birds)? Biol. Lett., 5, 125–129.

Jones, O., Gaillard, J.-M., Tuljapurkar, S., Alho, J., Armitage, K., Becker, P., et al. (2008). Senescence rates are determined by ranking on the fast-slow life-history continuum. Ecol. Lett., 11, 664–673.

Kalra, B. & Parkash, R. (2014). Trade-off of ovarian lipids and total body lipids for fecundity and starvation resistance in tropical populations of Drosophila melanogaster. J. Evol. Biol., 27, 2371–2385.

Kotiaho, J.S. (2001). Costs of sexual traits: a mismatch between theoretical considerations and empirical evidence. Biol. Rev., 76, 365–376.

Kotrschal, A., Rogell, B., Bundsen, A., Svensson, B., Zajitschek, S., Brännström, I., et al. (2013). Artificial selection on relative brain size in the guppy reveals costs and benefits of evolving a larger brain. Curr. Biol., 23, 168–171.

Lack, D. (1947). The significance of clutch-size. Ibis (Lond. 1859)., 89, 302–352.

Law, R. (1979). Optimal life histories under age-specific predation. Am. Nat., 114, 399–417.

Lee, W., Monaghan, P. & Metcalfe, N.B. (2013). Experimental demonstration of the growth rate - lifespan trade-off. Proc. R.

Soc. B Biol. Sci., 280, 20122370.

Liao, W.B., Lou, S.L., Zeng, Y. & Kotrschal, A. (2016). Large brains, small guts: the expensive tissue hypothesis supported

15 within anurans. Am. Nat., 188 6, 693–700.

Lloyd, D.G. (1987). Selection of offspring size at independence and other size-versus-number strategies. Am. Nat., 129, 800–

817.

Mas-Muñoz, J., Komen, H., Schneider, O., Visch, S. & Schrama, J. (2011). Feeding behaviour, swimming activity and boldness explain variation in feed intake and growth of sole (Solea solea) reared in captivity. PLoS One, 6, e21393.

McGraw, J.B. & Caswell, H. (1996). Estimation of individual fitness from life-history data. Am. Nat., 147, 47–64.

Medawar, P.B. (1952). An unsolved problem of biology. H.K. Lewis & Co., London, UK.

Mink, J. W., R. J. Blumenschine, and D. B. Adams. 1981. Ratio of central nervous system to body metabolism in vertebrates:

its constancy and functional basis. Am. J. Physiol. Integr. Comp. Physiol. 241:R203-212.

Morita, K., Tamate, T., Sugimoto, Y., Tago, Y., Watanabe, T., Konaka, H., et al. (2009). Latitudinal variation in egg size and number in anadromous masu salmon Oncorhynchus masou. J. Fish Biol., 74, 699–705.

van Noordwijk, A. & de Jong, G. (1986). Acquisition and allocation of resources: their influence on variation in life history tactics. Am. Nat., 128, 137–142.

Olijnyk, A.M. & Nelson, W.A. (2013). Positive phenotypic correlations among life-history traits remain in the absence of differential resource ingestion. Funct. Ecol., 27, 165–172.

Promislow, D. & Harvey, P. (1990). Living fast and dying young: a comparative analysis of life-history variation among mammals. J. Zool., 220, 417–437.

Read, A. & Harvey, P. (1989). Life history differences among the eutherian radiations. J. Zool., 219, 329–353.

Réale, D., Garant, D., Humphries, M., Bergeron, P., Careau, V. & Montiglio, P. (2010). Personality and the emergence of the pace-of-life syndrome concept at the population level. Philos. Trans. R. Soc. B Biol. Sci., 365, 4051–4063.

Reznick, D., Bryant, M.J. & Bashey, F. (2002). r- and K-selection revisited: the role of population regulation in life-history evolution. Ecology, 83, 1509–1520.

Rice, W.R. & Holland, B. (1997). The enemies within: intergenomic conflict, interlocus contest evolution (ICE), and the intraspecific Red Queen. Behav. Ecol. Sociobiol., 41, 1–10.

Ricklefs, R. (2000). Density dependence, evolutionary optimization, and the diversification of avian life histories. Condor, 102, 9–22.

Robertson, D.R. & Collin, R. (2015). Inter-and Intra-specific variation in egg size among reef fishes across the Isthmus of Panama. Front. Ecol. Evol., 2, 84.

Roff, D. (1992). Evolution of life histories: theory and analysis. Chapman and Hall, New York.

Rudh, A., Rogell, B., Håstad, O. & Qvarnström, A. (2011). Rapid population divergence linked with co-variation between coloration and sexual display in strawberry poison frogs. Evolution (N. Y)., 65, 1271–1282.

van Schaik, C. & Deaner, R. (2003). Life history and cognitive evolution in primates. In: Animal social complexity:

intelligence, culture, and individualized societies (eds. de Waal, F. & Tyack, P.). Harvard University Press, Cambridge, Massachusetts, pp. 5–25.

Smaers, J.B., Dechmann, D.K.N., Goswami, A., Soligo, C. & Safi, K. (2012). Comparative analyses of evolutionary rates reveal different pathways to encephalization in bats, carnivorans, and primates. Proc. Natl. Acad. Sci., 109, 18006–

18011.

Smith, C.C. & Fretwell, S.D. (1974). The optimal balance between size and number of offspring. Am. Nat., 108, 499–506.

Smith, J.N.M. (1981). Does high fecundity reduce survival in song sparrows? Evolution (N. Y)., 35, 1142–1148.

Sol, D., J. Maspons, A. Gonzalez-Voyer, I. Morales-Castilla, L. Z. Garamszegi, and A. P. Møller. (2018). Risk-taking behavior, urbanization and the pace of life in birds. Behav. Ecol. Sociobiol. 72:59. Springer.

Sol, D., Sayol, F., Ducatez, S. & Lefebvre, L. (2016). The life-history basis of behavioural innovations. Philos. Trans. R. Soc.

B, 371, 20150187.

16

Stearns, S. (1989). Trade-offs in life-history evolution. Funct. Ecol., 3, 259–268.

Stearns, S. (1992). The evolution of life histories. Oxford University Press, Oxford, UK.

Stearns, S.C. (1976). Life-history tactics: a review of the ideas. Q. Rev. Biol., 51, 3–47.

Steigenga, M.J. & Fischer, K. (2007). Ovarian dynamics, egg size, and egg number in relation to temperature and mating status in a butterfly. Entomol. Exp. Appl., 125, 195–203.

Wang, Z., J. Zhang, Z. Ying, and S. B. Heymsfield. (2012). Organ-tissue level model of resting energy expenditure across mammals: new insights into Kleiber’s law. ISRN Zool. 2012:1–9.

Ward, A., Thomas, P., Hart, P. & Krause, J. (2004). Correlates of boldness in three-spined sticklebacks (Gasterosteus aculeatus). Behav. Ecol. Sociobiol., 55, 561–568.

White, S., Kells, T. & Wilson, A. (2016). Metabolism, personality and pace of life in the Trinidadian guppy, Poecilia reticulata. Behavior, 153, 1517–1543.

Williams, G.C. (1957). Pleiotropy, natural selection, and the evolution of senescence. Sci. Aging Knowl. Environ., 11, 398–

411.

Xie, L. & Klerks, P.L. (2004). Fitness cost of resistance to cadmium in the least killifish (Heterandria formosa). Environ.

Toxicol. Chem. An Int. J., 23, 1499–1503.

Yu, X., Zhong, M.J., Li, D.Y., Jin, L., Liao, W.B. & Kotrschal, A. (2018). Large-brained frogs mature later and live longer.

Evolution (N. Y)., 72, 1174–1183.

Zwaan, B., Bijlsma, R. & Hoekstra, R.F. (1995). Direct selection on life span in Drosophila melanogaster. Evolution (N. Y)., 49, 649–659.

17 SVENSK SAMMANFATTNING

Livshistorieegenskaper, det vill säga egenskaper som bland annat påverkar hur många individer som finns i en population, såsom reproduktiv framgång och överlevnad, är centrala för vår förståelse av evolution. Flera andra egenskaper påverkar populationsdynamik mer indirekt, som till exempel beteende, fysiologi och morfologiska egenskaper, vilka, enligt flera hypoteser, bör samvariera med livshistorieegenskaper. Denna avhandling syftar till att identifiera vilka egenskaper (så som beteendemässiga, fysiologiska och morfologiska egenskaper) som samvarierar med livshistorieegenskaper, och varför. Ett flertal hypoteser beskriver varför livshistoria bör vara relaterat till andra egenskaper, där avvägningar mellan investering i olika livshistorieegenskaper anses ha en nyckelroll. Till exempel; livshistorier evolverar vanligtvis kontinuerligt längs med en axel från kortlivad och med hög fertilitet, till långlivad och med lägre fertilitet. En möjlig förklaring är att resurser är begränsade, vilket bör leda till att en ökad investering i en egenskap leder till en minskad investering i en eller flera andra egenskaper.

Hypoteser som rör sambandet mellan livshistoria och andra egenskaper har testats i många studier, men få studier på mellan- artnivå har genomförts under kontrollerade experiment. I denna avhandling undersöker jag hur beteende, fysiologi och morfologi relaterar till variation längs med livshistorieaxeln, som går från snabb till långsam, i ett system med en betydande mängd variation i livshistorieegenskaper - killifiskar.

Jag började med att utforska det allometriska sambandet mellan äggstorlek och kroppsstorlek hos killifiskar (Artikel I), där arter med snabbare livshistorier uppvisade hade starkare samevolution mellan äggstorlek och kroppsstorlek, jämfört med arter med långsammare livshistorier. Vidare fann jag skillnader i varians och evolutionshastighet av både äggstorlek och kroppsstorlek. Dessa skillnader beror troligtvis på att de snabbt levande arterna lever i temporära (uttorkande) habitat habitat,vilket kan ha resulterat i en diversifiering av både ägg och kroppsstorlek.

Därefter genomförde jag, i en ‘common-garden’ miljö, en jämförande studie (Artikel II) av the pace-of-life syndrome hypotesen, som förutsäger att arter med snabba livshistorier borde ta större risker för att bibehålla deras högre reproduktiva hastighet. Jag samlade in data på risktagande beteende, inklusive aktivitet, benägenhet att utforska en öppen arena och aggressivitet, utöver mätningar av metabolism, hos 20 killifiskarter, med flera replikat per art. Resultaten tydde på att förekomsten av samvariation med livshistoria berodde på vilken egenskap som studerades. Aggression visade sig korrelera positivt med livshistoriehastighet, i linje med förutsägelser baserade på pace-of-life syndrome hypotesen.

Därefter så undersökte jag och mina kollegor associationen mellan livshistoria och sexuell selektion (Artikel III), för att ta reda på om investeringar i sekundära könskaraktärer kan samvariera negativt med förmågan att överleva predation hos killifiskar. Fenstorlek visade sig vara negativt relaterat till flykthastighet vid en simulerad predatorattack, vilket tyder på en överlevnadskostnad för individer med större fenor. Fenstorlek var även positivt relaterat till livshistoriehastighet, vilket stödjer hypotesen att kostnader i termer av lägre överlevnad troligtvis är lägre i de snabblevande arterna.

Slutligen så testade jag även hypotesen att relativ hjärnstorlek bör korrelera negativt med livshistoriehastighet, genom att samla in och analysera hjärnstorleksmått från 21 arter av killifisk (Artikel IV). I motsats till förutsägelsen så återfanns ett positivt samband mellan livshistoriehastighet och relativ hjärnstorlek. Detta innebär en av två saker: anting så finns det inget behov av en avvägning mellan hjärnstorlek och livshistoria eftersom detta kompenserats med en högre tillgång på resurser, eller så betyder det att ökad hjärnstorlek och ökad livshistoriehastighet betalas med hjälp av minskade investeringar i andra egenskaper.

18

Sammanfattningsvis har jag visat att tidigare funna avvägningar mellan livshistoria och andra kostsamma egenskaper bara återfinns ibland, i ett system med stor variation i livshistoriehastighet.

19 ACKNOWLEDGEMENTS

There are many people I owe a debt of gratitude to, for supporting me throughout this PhD project. The largest share of this gratitude goes to my two supervisors, Alejandro Gonzalez-Voyer and Björn Rogell. Alejandro, it never really felt like you were on another continent, thanks to the time and energy you invested in me (and perhaps also thanks to fiber optic cables).

You have given me valuable support and feedback, and taught me how to think about comparative analyses among many other things. Björn, thank you for all the fun and insightful conversations about everything from science to the experiences of having new family members come into existence and growing up. Thank you for teaching me how to think about and apply a broad range of statistical methods, it has been incredibly useful and rewarding.

I also want to thank my colleagues in the killifish lab; Will, Piotr and Joacim. A special thanks to Will; you have been a mentor and a friend throughout my PhD. You helped me out with setting up and performing experiments, and you always had answers to my numerous questions. Piotr, you are always very friendly and helpful, and you have given me very useful feedback on ideas for new projects. Joacim, thanks for all the long and interesting conversations about biology. Deep diving in different topics and questions have helped clarify what it is I understand, and also what it is I do not understand (yet), about particular subjects.

To all the colleagues at zootis; I have truly enjoyed working at this department, and it is largely thanks to you. I am very grateful to the members of my follow-up committee, Gabriella, Niklas and Niclas, for helping me bring this PhD project to completion. Siw, Minna and Anette; thank you for helping me with all sorts of questions, it has been a smooth ride thanks to you. I have enjoyed numerous fun conversations with my office roommates Mariana, Anna and Madicken. I have invariably had entertaining and interesting lunch conversations with Olle, Loke, Wouter, Will, Sara, David, Alexandre, Hannah, Alberto, Joacim, Kalle, Meet, Mariana, Rasmus, Madicken, Erika, Stephanie, Ram, Masahito, Ciaran, Mats, Sandra, Bertil, Charel, Alessandro, Ariel, Philipp, Mirjam, Christina, Chris and Christen, among others.

I am fortunate to have grown up in two very loving families. Thank you Katarina, Peter, Per and Angelique, for supporting me and encouraging me to pursue my goals, and thanks to Johannes, Hanna, Adina and David, for being the best siblings I could have hoped for. I would further like to thank all of my amazing friends at Effektiv Altruism Sverige, who have given me new perspectives on rationality and epistemology - concepts that are undoubtedly central to science. Last, but definitely not least, a big thank you to my beloved Emeli. You have always been supportive and understanding, even when I had to work late in the evenings.