Evolution of Reproductive Traits in Sharks and Rays

Home , Argiope lobata, Australian ghostshark, Eretmodus, Figaro (genus), Macrostomum, Metaplastes

EVOLUTION OF REPRODUCTIVE TRAITS IN SHARKS AND RAYS

A thesis submitted to the University of Manchester for the degree of Doctor of Philosophy (PhD) in the School of Medical Sciences, Faculty of Biology, Medicine and Health

2018

AMY ROWLEY

FACULTY OF BIOLOGY, MEDICINE AND HEALTH

2 Contents

LIST OF FIGURES 6

LIST OF TABLES 9

LIST OF APPENDICES 12

GENERAL ABSTRACT 13

DECLARATION 14

ACKNOWLEDGEMENTS 16

1. GENERAL INTRODUCTION 19 1.1 SEXUAL SELECTION 19 1.2 SPERM COMPETITION 22 1.3 CRYPTIC FEMALE CHOICE AND SEXUAL CONFLICT 33 1.4 OUTSTANDING QUESTIONS IN HOW SPERM COMPETITION INFLUENCES THE EVOLUTION OF REPRODUCTIVE TRAITS 34 1.4.1 SPERM NUMBER 35 1.4.2 SPERM MORPHOLOGY 36 1.4.3 SPERM VARIANCE 37 1.4.4 GENITAL MORPHOLOGY 38 1.5 STUDYING EVOLUTIONARY RESPONSES OF REPRODUCTIVE TRAITS TO SPERM COMPETITION 39 1.6 SPERM COMPETITION AND EVOLUTIONARY RESPONSE IN SEXUAL TRAITS IN ELASMOBRANCHS 39 1.6.1 ELASMOBRANCHS 40 1.6.2 SHARKS VS RAYS 41 1.6.3 REPRODUCTIVE BEHAVIOURS IN ELASMOBRANCHS 41 1.6.4 GENETIC MATING SYSTEMS 43 1.6.5 VARIATION IN REPRODUCTIVE TRAITS 46 1.7 REPRODUCTIVE VARIATION IN MALES 47 1.7.1 TESTES 47 1.7.2 SPERM MORPHOLOGY 48 1.7.3 CLASPERS 49 1.8 REPRODUCTIVE VARIATION IN FEMALES 50 1.8.1 REPRODUCTIVE MODE 50 1.8.2 FECUNDITY 51 1.8.3 SPERM STORAGE 52 1.9 CHALLENGES IN STUDYING ELASMOBRANCH REPRODUCTION 54 1.10 AIMS OF THE THESIS 55 1.11 REFERENCES 56

2. TESTES SIZE INCREASES WITH SPERM COMPETITION RISK AND INTENSITY IN BONY FISH AND SHARKS 72 2.1 ABSTRACT 73 2.2 INTRODUCTION 74 2.3 METHODS 76

3 2.3.1 DATA COLLECTION 76 2.3.2 PHYLOGENY 78 2.3.4 PHYLOGENETIC ANALYSES 79 2.4 RESULTS 81 2.4.1 VARIATION IN SPERM COMPETITION RISK AND INTENSITY AMONG FISHES 81 2.4.2 SPERM COMPETITION RISK, INTENSITY AND TESTICULAR INVESTMENT 83 2.5 DISCUSSION 87 2.6 ACKNOWLEDGMENTS 89 2.7 REFERENCES 89 CHAPTER 2: SUPPORTING INFORMATION 96 SUPPORTING INFORMATION REFERENCES 105

3. SPERM COMPETITION AND THE EVOLUTION OF TESTES ORGANISATION IN SHARKS AND RAYS 113 3.1 ABSTRACT 114 3.2 INTRODUCTION 115 3.3 METHODS 117 3.3.1 SAMPLE COLLECTION 117 3.3.2 HISTOLOGICAL PROCESSING 119 3.3.3 IMAGE ANALYSIS 119 3.3.4 ESTIMATING SPERM COMPETITION RISK 120 3.3.5 PHYLOGENETIC ANALYSES 121 3.4 RESULTS 123 3.5 DISCUSSION 132 3.6 ACKNOWLEDGMENTS 136 3.7 REFERENCES 137 CHAPTER 3: SUPPORTING INFORMATION 142 SUPPORTING INFORMATION REFERENCES 143

4. SEXUAL SELECTION DRIVES DIVERGENT PATTERNS OF SELECTION ON SPERM FLAGELLUM LENGTH IN SHARKS AND RAYS 146 4.1 ABSTRACT 147 4.2 INTRODUCTION 148 4.3 METHODS 151 4.3.1 SAMPLE COLLECTION 151 4.3.2 SPERM ANALYSIS 152 4.3.3 PHYLOGENETIC LINEAR MODELS 154 4.3.4 COMPARING EVOLUTIONARY RATES 156 4.4 RESULTS 157 4.4.1 SPERM COMPETITION AND SPERM MORPHOLOGY 157 4.4.2 RATES OF EVOLUTION OF SPERM COMPONENTS 163 4.5 DISCUSSION 164 4.6 ACKNOWLEDGEMENTS 171 4.7 REFERENCES 172 CHAPTER 4: SUPPORTING INFORMATION 179 SUPPORTING INFORMATION REFERENCES 183

5. THE EVOLUTION OF WEAPONIZED GENITALS IN SHARKS 191 5.1 ABSTRACT 192 5.2 INTRODUCTION 193

4 5.3 METHODS 197 5.3.1 FIELD COLLECTED SAMPLES 197 5.3.2 LITERATURE COLLECTED DATA 197 5.3.3 COMPILING THE FINAL DATASET 199 5.3.4 PHYLOGENETIC ANALYSES 199 5.3.5 ANCESTRAL STATE RECONSTRUCTION 200 5.3.6 PHYLOGENETIC LINEAR MODELS 200 5.4 RESULTS 201 5.4.1 ANCESTRAL STATE RECONSTRUCTIONS AND THE GAINS AND LOSSES OF GENITAL APPENDAGES 201 5.4.2 PHYLOGENETIC LINEAR MODELS 204 5.5 DISCUSSION 208 5.6 ACKNOWLEDGMENTS 213 5.7 REFERENCES 214 CHAPTER 5: SUPPORTING INFORMATION 220 SUPPORTING INFORMATION REFERENCES 229

6. GENERAL CONCLUSION 238 6.1 THESIS SUMMARY 238 6.2 LIMITATIONS 245 6.3 FUTURE DIRECTIONS 248 6.4 REFERENCES 250

5 List of Figures

Chapter 1: General introduction

Figure 1.1: Summary of variation observed in male and female reproductive behaviours and how this variation influences the risk of sperm competition experienced by males ……………………………………………………………………………………….... 24 Figure 1.2: Comparison of a large guarding male and a smaller sneaking male in plainfin midshipman (Porichthys notatus), a species with two alternative male reproductive tactics. Testis size of a guarding and sneaking male demonstrating increased investment in testes in sneaking males ………………………………………………. 26 Figure 1.3: Diversity in genital morphology among cartilaginous fishes (sharks, skates, rays, sawfish and chimaeras) .……………………………………………………………………27 Figure 1.4: Variation in the percentage of litters sired by multiple males across shark and ray species ..………………………………………………………………………………………… 45 Figure 1.5: Variation in elasmobranch testes organisation in three shark and ray species, as illustrated by haematoxylin and eosin-stained histological testes sections ….………………………………………………………………………………………………………………………….48 Figure 1.6: Elasmobranch sperm ultrastructure. A single sperm cell from a nervous shark (Carachrinus cautus), with the components indicated and head coloured with the nuclear-binding stain DAPI ..……………………………………………………………………………49

Chapter 2: Testes size increases with sperm competition risk and intensity in bony fish and sharks

Figure 2.1: Variation in sperm competition risk and intensity across fishes. Phylogenetic relationships between fish species in our dataset, with points representing the percentage of litters sired by multiple males for each species (risk), and the average number of sires per brood within each species (intensity) ……………………………………………………………………………………………………………………………..82

6 Figure 2.2: Relationships between testes mass relative to body size and (a) the percentage of litters sired by more than one male (sperm competition risk) and (b) the mean number of sires per brood (sperm competition intensity) across species. ……………………………………………………………………………………………………………………………. 86

Chapter 3: Sperm competition and the evolution of testes organisation in sharks and rays

Figure 3.1: Variation in testes organisation traits across sharks and rays, as illustrated by haematoxylin and eosion-stained histological testes sections from (a) Ginglymostoma cirratum, (b) Mustelus asterias, (c) Chiloscyllium punctatum and (d) Scyliorhinus stellaris, shown at 20x magnification. …………………………………….……… 125 Figure 3.2: Traitgrams showing phenotypic divergence of (a) the proportion of sperm-producing tissue, (b) maximum follicle cross-sectional area and (c) body mass over time. ………………………………………………………………………………………………... 129 Figure 3.3: Relationships between testes mass corrected for body size (a proxy measure for sperm competition risk) and (a) the percentage of sperm-producing tissue within the testes and (b) maximum cross-sectional area of the seminiferous follicles across shark (black points) and ray (white points) species. …………………… 131

Chapter 4: Sexual selection drives divergent patterns of selection on sperm flagellum length in sharks and rays

Figure 4.1: Sperm collection and variation in sperm morphology across sharks and rays. ………………………………………………………………………………………………………………….. 159 Figure 4.2: Associations between sperm morphological traits (plotted on a log- scale) and body size corrected testes mass, a proxy measure for sperm competition risk, in sharks (black points) and rays (white points) …………………………………………. 160 Figure 4.3: Rates of phenotypic evolution of sperm components in (a) sharks and (b) rays and (c) all elasmobranch species in our dataset. ……………………………………….. 164

7 Chapter 5: The evolution of weaponized genitals in sharks

Figure 5.1: Variation in shark clasper morphology. …………………………………………… 196 Figure 5.2: Stochastic posterior probability density map of the presence or absence of clasper appendages in shark species, calculated from 1000 simulated phylogenies under an ‘all rates different’ (ARD) model of transitition probabilities. ……………. 203 Figure 5.3: Relationships between body length and clasper length (plotted on a log10-scale) across shark species, in species for which genital appendages are present (white points) and absent (black points) ……………………………………………... 208

Chapter 6: General conclusion No figures.

8 List of tables

Chapter 1: General introduction

No tables.

Chapter 2: Testes size increases with sperm competition risk and intensity in bony fish and sharks

Table 2.1: Phylogenetically controlled generalised least squares (PGLS) regressions between testes size and sperm competition risk/intensity …………………………………. 84 Table S2.1: Multiple paternity, body size and testes data collected for all n=34 fish species used in our analyses. ………………………………………………………………………………. 96

Chapter 3: Sperm competition and the evolution of testes organisation in sharks and rays

Table 3.1: Evaluation of phylogenetic signal in the proportion of sperm-producing tissue within the testes and maximum cross-sectional area of the seminiferous follicles. …………………………………………………………………………………………………………..… 127 Table 3.2: Summary of model fits for Brownian motion (BM), single stationary peak (SSP, or single-optimum Ornstein-Uhlenbeck (OU) model), and accelerating/decelerating, where the change in rate is either exponential (ACDC exponential) or linear (ACDC linear), models of evolution for the percentage of seminiferous tissue within the testes (% Seminiferous tissue), the maximum cross- sectional area of the seminiferous follicles (Maximum follicle area), and adult body mass. ………………………………………………………………………….…………………………………….. 128 Table 3.3: Phylogenetically controlled linear regressions between testes organisation traits and testes mass in (a) sharks and (b) rays. ………………………….. 130

9 Table S3.1: Testes mass (g), body mass (g), mean percentage of sperm-producing tissue within the testes and maximum seminiferous follicle cross-sectional area for 18 shark and ray species. ………………………………………………………………………………….. 142

Chapter 4: Sexual selection drives divergent patterns of selection on sperm flagellum length in sharks and rays

Table 4.1: Phylogenetically controlled generalized least squares (PGLS) regressions between sperm traits and testes mass. …………………………………………………………….. 161 Table S4.1: Mean, standard deviation (SD) and coefficient of variation (CV) of sperm component length, number of males from which sperm samples were collected (n), and (where applicable) body mass and testes mass data collected for (a) sharks, (b) rays, and (c) chimera species. …………………………………………………………………………… 179 Table S4.2: Phylogenetically controlled generalized least squares (PGLS) regressions between sperm variance and testes mass. ……………………………………………………….. 184 Table S4.3: Summary of model fits for Brownian Motion, Ornstein-Uhlenbeck, and Early-burst models of character evolution for total head and midpiece length and flagellum length (ln-transformed) for (a) sharks (n = 23) and (b) rays (n = 9) included in our evolutionary rates models ……………………………………………………………….…….. 186 Table S4.4. Phylogenetically controlled generalized least squares (PGLS) regressions between sperm traits and testes mass and taxonomic group (shark vs. ray)…….. 187

Chapter 5: The evolution of weaponized genitals in sharks

Table 5.1: Phylogentically controlled linear regressions between (a, c) clasper length and the presence or absence of genital appendages, and (b, d) testes mass and the presence or absence of genital appendages. ……………………………………..… 205 Table 5.2: Phylogentically controlled linear regressions between clasper length and body length and testes mass. ……………………………………………………………………..…….. 207

10 Table S5.1: Summary of model fits for Brownian Motion (BM), Ornstein-Uhlenbeck (OU) and Early-burst (EB) models of trait evolution for clasper length and body length. ………………………………………………………………………………………………………………. 220 Table S5.2: Mean clasper length, presence or absence of clasper appendages (hooks, spines, spurs or claws), mean body length and mean testes mass data collected for shark species. ………………………………………………………..…………………….. 222

Chapter 6: General conclusion No tables.

11 List of appendices

Appendix 1: ‘Sperm Competition’ by A. Rowley and J. L. Fitzpatrick, as published in the Encyclopaedia of Evolutionary Biology (2015)

12 General abstract

Females frequently mate with multiple males in a single reproductive episode, which creates the potential for sperm competition, a potent selective force that shapes male reproductive anatomy across the animal kingdom. However, relationships between sperm competition and reproductive traits at the interspecific level are generally inconsistent across taxa, and some key taxonomic groups are understudied or remain to be assessed, which limits the potential for evaluating the consistency of selection. In this thesis, I take a comparative approach to examine how sperm competition influences the evolution of reproductive traits in the elasmobranch fishes (sharks and rays). In spite of the extraordinary variation in reproductive systems and structures in this group, prior to this thesis we knew very little about the selective processes that have shaped them. First, I show that in shark species where females are more likely to mate with multiple males, or mate with a greater number of males, males invest in larger testes relative to body mass. This validates the use of body-size corrected testes mass as a proxy for sperm competition risk and intensity. Secondly, I demonstrate that reproductive traits exhibit variable responses to the level of sperm competition (as measured by body size-corrected testes mass). The length of the sperm flagellum appears to be a target of selection in both sharks and rays, though flagellum length is positively associated with sperm competition risk in sharks, whereas in rays the relationship is negative. In contrast, the structural organisation of the testes is not influenced by sperm competition in sharks, although there is tentative evidence to suggest that males from species at higher sperm competition risk have a higher proportion of sperm-producing tissue in the testes in rays. Finally, in sharks, the length of the male genitalia and presence or absence of genital appendages (spines, hooks and claws) are not associated with sperm competition risk, and only evolutionary losses of spines were detected across shark phylogeny. Overall, these findings provide new insights into post-mating selective dynamics in an ancestral vertebrate lineage and contribute to a wider understanding of how sexual selection acts on reproductive traits across the vertebrate tree of life.

13 Declaration

No portion of the work referred to in this thesis has been submitted in support of an application for another degree or qualification of this or any other institute of learning.

14 Copyright statement

The author of this thesis (including any appendices and/or schedules to this thesis) owns certain copyright or related rights in it (the “Copyright”) and s/he has given The University of Manchester certain rights to use such Copyright, including for administrative purposes.

Copies of this thesis, either in full or in extracts and whether in hard or electronic copy, may be made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part of any such copies made.

The ownership of certain Copyright, patents, designs, trademarks and other intellectual property (the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproductions.

Further information on the conditions under which disclosure, publication and commercialisation of this thesis, the Copyright and any Intellectual Property and/or Reproductions described in it may take place is available in the University IP Policy (see http://documents.manchester.ac.uk/DocuInfo.aspx?DocID=2442 0), in any relevant Thesis restriction declarations deposited in the University Library, The University Library’s regulations (see http://www.library.manchester.ac.uk/about/regulations/) and in The University’s policy on Presentation of Theses.

15 Acknowledgements

To borrow the words of Oscar Wilde: to lose one supervisor may be regarded as a misfortune; to lose two looks like carelessness. However, I consider myself extremely fortunate to have had not one, or two, but three excellent PhD supervisors. The first and largest thanks are owed to John Fitzpatrick, for trusting me with this (frankly, ridiculous) project that turned into a great adventure. I learned an incredible amount from working with you for the last four years, and had a lot of fun along the way. Thank you for constantly challenging me, for throwing me into the deep end almost literally with the first few field seasons, and for believing in my ability even when I didn’t. It’s been a weird and eventful four years and I couldn’t have wished for a better person to guide me through it.

To Richard Preziosi and Daniel Brison, thank you both for your helpful advice as co- supervisors, and for later taking up the responsibilities of primary supervisor. During all the changes to my supervisory team, I never felt abandoned and always knew I had someone I could reply on for guidance or support. This project benefitted from your unique perspectives and it was a pleasure to share it with you.

It’s been great to see the Fitzpatrick lab grow exponentially over the years. Ale, Ariel, Raissa, Charel, Madicken, Hannah and Erika, thank you for all of your advice, encouragement and sarcastic Whatsapps. There’s a solid chance that I wouldn’t have survived the last few months of writing without the steady stream of snacks arriving at my desk. You all make me want to be a better scientist, and a much better mini-golfer, and I’m so excited to see all the great work that you’ll continue to do in the future.

The work presented in this thesis, and in particular the data collected in the field, is the result of a collaborative effort of epic proportions. I am indebted to all of the many people who provided samples, assistance in the field and access to field sites. While there are too many to name, special thanks are due to Paco, Trudi, Vicente,

16 Curro, Mariana and the Rego family, Toby, Dean, Mark, Lisa, Carlotta and many more. Without you, this thesis could not exist.

I had the immense privilege of travelling during my PhD, both for field work and as part of the BBSRC DTP placements program. I learned so much from every place and especially from the people I met along the way, and am indebted to the BBSRC for making this possible. To the Bimini Sharklab PIT crew 2017 (special mention to the Putting Queens, Clemency, Chessie and Cheyenne, and honorary Room 2 member Hamdan), getting to live and work with you was an absolute highlight of my PhD, and I hope to see all of you again soon, somewhere in the world.

An unexpected bonus of my PhD was getting to work in the Zoology department at Stockholm University for much of the last two years. I’m so thankful for the warm welcome I received here; you made Stockholm feel like home straight away. Special thanks to the Friday pub regulars, and to Chris and Christina for keeping the beer flowing. The value of a friendly and supportive environment to work in can’t be overstated, and Zootis is one of the best.

Thanks also to my family, and in particular my parents, for supporting and encouraging everything I do without question, and to my friends, especially Charley, Doug, Kyle, Georgie, Adam and Alice. Every small disaster was a little bit easier because I knew it would be a funny story to tell you later.

Finally, thank you Wouter. Your constant support has meant so much over the last few years, and in particular the difficult last months. Thank you for always giving an honest opinion, for all your help and advice (both in person and via your old Stack Overflow posts), and for making me laugh through all the stress. I’ve had a crazy amount of good fortune over the last four years, but meeting you was the best.

Chapter 1: General Introduction

18 1. General Introduction

1.1 Sexual selection

In his theory of natural selection, Darwin described how traits that increase an organism’s chance of survival would inevitably increase in frequency in the population through subsequent generations (Darwin 1859). However, this explanation was not sufficient to describe some of the more extreme features he observed in the animal kingdom, which seemed to have no effect on, or even to negatively influence, chances of survival. This is because, as Darwin later realised, survival without reproduction is meaningless in evolutionary terms. In addition to merely surviving beyond sexual maturity, organisms belonging to sexually- reproducing species may have to attract a conspecific (usually) of the opposite sex, compete against rivals, and mate successfully in order to produce offspring and thereby transfer genes to the next generation. Variation among animals in the ability to perform these tasks results in differential reproductive success within the population, or sexual selection (Darwin 1871). Sexual selection is now recognized as a powerful evolutionary force that exerts a strong influence on morphology, physiology and behaviour, and is responsible for many of the most extreme traits seen in nature (Andersson 1994; Birkhead and Møller 1998).

Males invest in traits that increase reproductive success, even if such traits appear maladaptive from the perspective of natural selection; they produce elaborate ornaments to attract females, perform conspicuous courtship behaviours, and grow to huge sizes or develop weapons to fight rivals (Darwin 1871; Andersson 1994). These traits are shaped by sexual selection prior to copulation, which can occur in two main ways. Firstly, males may compete against each other for access to

19 females (i.e. intra-sexual selection). In many species, notably those in which males defend a harem of females, male-male competition is intense and can lead to extremely skewed reproductive success, with few males accounting for the majority of copulations and most males failing to reproduce at all (Le Boeuf 1974). Consequently, male-male competition is a potent selective force that drives the evolution of weapons such as horns or antlers (Bro-Jørgensen 2007), sexual size dimorphism (Fitzpatrick et al. 2012a) and traits that signal dominance to other males (Wagner 1989). Secondly, pre-copulatory sexual selection on males may arise through female mate choice (i.e. inter-sexual selection). Females can increase their fitness by selectively mating with more attractive males (Drickamer et al. 2000; Andersson and Simmons 2006), and may distinguish between potential mates using vocal signals (Ryan et al. 1990), ornaments (Zuk et al. 1990) or colour (Milinski and Bakker 1990; Waitt et al. 2003). As a consequence of directional preference for these traits, female mate choice is frequently implicated in the evolution of exaggerated male traits that may be detrimental to survival, such as long tails in birds (Andersson 1982).

For much of the century following Darwin’s (1871) introduction of the concept, the study of sexual selection was dominated by the pre-copulatory processes of male- male competition and female choice, largely because females were assumed to be predominantly monogamous (Birkhead 2000). However, we now know this assumption to be false. Behavioural observations and molecular techniques have shown that females frequently mate with multiple males during a single reproductive period, and as a consequence of this polyandry, sexual selection often does not end with mating (Eberhard 2009). Post-copulatory sexual selection can be broadly divided into two main mechanisms, sperm competition and cryptic female choice. When females mate multiply, the ejaculates of rival males will occupy the fertilisation environment contemporaneously. This creates the potential for competition between sperm from each male to fertilise the limited supply of ova (i.e. sperm competition; Parker 1970). Sperm competition is widespread throughout the animal kingdom and imposes intense selective pressures on the male reproductive anatomy, which I discuss in more detail below. However, in

20 recent years it has become more widely recognised that females can physiologically discriminate between the ejaculates of different males that simultaneously occupy the fertilisation environment. Thus, females may be able to exert control over the fertilisation of their ova to bias paternity in favour of preferred males, a process known as cryptic female choice (Thornhill 1983; Eberhard 1996).

Although species may experience sexual selection both before and after mating, the relative importance of pre- and post-copulatory sexual selection will vary across taxa depending on their reproductive behaviour and breeding biology (Parker et al. 2013). In the rare cases in which females are truly monogamous, post-copulatory sexual selection will, for obvious reasons, be virtually non-existent. In contrast, in many species the potential for pre-copulatory sexual selection is limited, and postcopulatory processes such as sperm competition and cryptic female choice are therefore the dominant mechanisms of sexual selection. For example, females may lack the opportunity to assess males prior to mating, or be unable to avoid mating with non-preferred males or do so to avoid injury or harassment (convenience polyandry). In the absence of pre-copulatory mate choice, females can limit the potential costs of mating with a genetically incompatible or substandard male by ‘bet-hedging’ through multiple mating and/or subsequent cryptic female choice (e.g. Welke and Schneider 2009). In addition, males may be unable to monopolise access to females, in which case male contest competition will be superseded by scramble competition, thereby making sperm competition more likely and causing males to invest more in post-copulatory reproductive traits (Lüpold et al. 2014). Due to the overwhelming prevalence of female multiple mating across the animal kingdom (Birkhead 2000), post-copulatory sexual selection is likely to be an extremely common and important selective force in many species. Therefore, in this thesis I focus on how post-copulatory sexual selection shapes the evolution of reproductive traits, with a particular emphasis on understanding how animal sexual traits respond to sperm competition.

21 1.2 Sperm Competition

(Note: This section was originally published as an entry of ‘Sperm Competition’ in the Encyclopedia of Evolutionary Biology by Rowley and Fitzpatrick (2015). See Appendix I for the published version).

Darwin (1871) recognized that males who are better able to attract mates or outcompete rival males would leave more offspring to future generations, and would therefore be favoured by evolution. Perhaps not surprisingly then, the pursuit of reproductive opportunities has led to the evolution of some of the most striking features in the animal kingdom. Males produce conspicuous ornaments (e.g. the peacocks tail) that are used to attract more mates than rivals or arm themselves with costly weapons (e.g. a stag’s antlers) that are used to outcompete rivals for access to mates. However for Darwin, and evolutionary biologists who followed for the next century, competition among males ended at mating. We now know that this is not the case. Molecular evidence has revealed that females frequently mate with multiple males during a single reproductive period and different males commonly sire eggs from a single clutch (Birkhead and Møller 1998). Such female promiscuity means that competition between males continues after mating in the form of sperm competition, the contest between sperm from rival males to fertilise an egg(s). The evolutionary significance of sperm competition was first recognised by Parker (1970), sparking a paradigm shift that resulted in an explosion of interest in the importance of sperm competition in shaping the evolution of reproductive behaviours and phenotypes.

The pervasive influence of sperm competition has played a vital role in the evolution and extraordinary diversification of reproductive traits. Adaptation to the intense selective pressures imposed by sperm competition has generated an astonishing variety of reproductive phenotypes, from male genitals that resemble medieval weapons (Crudgington and Siva-Jothy 2000), to testes so large that they account for more than 10% of adult male body mass (Montgomerie and Fitzpatrick 2009), to gigantic sperm measuring a remarkable 6 cm in length (20 times longer

22 than the male who produced them!) (Hosken 2003). Here we explore how sperm competition generates the tremendous diversity in reproductive traits by describing the conditions under which sperm competition occurs and examining how sperm competition influences the evolution of animal behaviours, anatomy and physiology.

Among species, female mating behaviour varies widely and this has important implications for the risk of sperm competition experienced by males (Figure 1.1). In species with monogamous females, who only mate with one male, the risk of sperm competition is low (or absent) and in these species sperm competition is unlikely to represent an important selective force shaping reproductive traits. However, genetic monogamy is exceedingly rare in animals. More often, socially monogamous females mate with, and produce offspring by, males outside of their social pairing. Furthermore, in many species female promiscuity is common, as females either seek out multiple mating opportunities or have little control over the number of males attempting to fertilize their eggs (as is often the case in externally fertilizing species). Therefore, the prevalence of female multiple mating means that males of the vast majority of species will experience sperm competition, albeit to varying extents, and researchers have capitalized on this variation in female promiscuity and animal mating systems to gain a better understanding of how sperm competition shapes reproductive traits.

Figure 1.1. Summary of variation observed in male and female reproductive behaviours and how this variation influences the risk of sperm competition experienced by males. Male and female behaviours can work synergistically or antagonistically to influence sperm competition risk. Solid lines between males and females represent matings, while dashed lines represent thwarted mating attempts.

As with females, males too have evolved a suite of behaviours that influence the risk of sperm competition (Figure 1.1). Males can actively guard their mates to prevent subsequent matings by rival males and thus reduce their risk of sperm competition. In an effort to prevent sperm competition males can even take the extreme action of detaching and lodging their genitalia inside the female to physically block access to the females reproductive tract (Fromhage 2006). In contrast, male reproductive behaviours can also increase sperm competition risk. For example, socially subordinate males who are unable to successfully attract females often adopt ‘sneaking’ behaviours (Figure 1.2), attempting to surreptitiously fertilize eggs without the knowledge of socially dominant males (Gross 1996). Therefore, despite the best efforts of males to reduce their risk of sperm competition, both male and female reproductive behaviours ensure that sperm competition persists.

For internally fertilizing species, male genitalia, which deliver sperm to the female’s reproductive tract, are at the front line of competition among rival males. Male genital morphology is extraordinarily diverse (Figure 1.3), even among closely related species, and this diversity is attributable in large part to the selective pressures imposed by sperm competition (Hosken and Stockley 2004). Male genitalia exhibit an astonishing variety of adaptations to maximise their chances of success in sperm competition. For example, male genitalia can displace sperm from previous matings, as is the case in the damselfly Calopteryx maculata, where male genitalia are covered in spines that remove almost all previously deposited sperm of rival males from the female reproductive tract, thereby virtually eliminating sperm competition (Waage 1979). When looking across species, genitalia have more elaborate sexual ‘weaponry’, including adaptations for sperm removal or displacement, in species where the risk of sperm competition is high compared with closely related monogamous species (Stockley 2002). Similarly, populations of seed beetles (Callosobruchus maculatus) evolved under experimentally enforced monogamy show reduced genital spine length relative to polygamous populations (Cayetano et al. 2011).

Figure 1.2. Sperm competition and sneaky matings. a. Comparison of a large guarding male and a smaller sneaking male in plainfin midshipman (Porichthys notatus), a species with two alternative male reproductive tactics. Both males are reproductively mature. Guarding males court females, defend territories and provide parental care for developing embryos, while sneaking males attempt to fertilize eggs surreptitiously and then leave the costly parental care to guarding males. Because sneaking males exclusively release sperm in the presence of a rival male they experience a higher sperm competition risk compared with guarding males. Testis size of a b. guarding and c. sneaking male demonstrating increased investment in testes in sneaking males, which represents a characteristic response to increased sperm competition risk. Photo credit: John Fitzpatrick.

Figure 1.3. Diversity in genital morphology among cartilaginous fishes (sharks, skates, rays, sawfish and chimaeras). Clasper (paired genitalia in cartilaginous fish) morphology differs dramatically in size and shape among a. Australian ghostshark, Callorhinchus milii (Photo credit: Eduardo Garza Gisholt), b. common skate, Dipturus batis (Photo credit: Amy Rowley), which shows paired claspers with the tail extending beyond the claspers and outside of the margins of the photo and c. New Zealand lanternshark, Etmopterus baxteri (Photo credit: Eduardo Garza Gisholt). Note the genital hooks and spurs observed at the terminal portion of claspers in a. and c.

Sperm competition also influences genital size and shape. For example, in house mice (Mus domesticus), males maintained under breeding regimes where sperm competition occurred for 27 generations had thicker penis bones (baculum) than males maintained under enforced monogamy (Firman and Simmons 2014). Across rodent and carnivores, the baculum is longer in species where sperm competition is prevalent (Ramm 2007a). There is good reason for genital morphology to respond to sperm competition risk, as several studies reveal that genital morphology predicts male reproductive success during sperm competition (Simmons et al. 2009; Stockley et al. 2013).

Following the release of sperm, either from male genitalia inside the female’s reproductive tract or into the external environment in the case of internal and external fertilizing species, respectively, the sperm themselves are the primary combatants in male-male competition. Consequently, sperm are under intense selection, as those sperm traits that provide an advantage during sperm competition will be favoured by selection. In particular, sperm competition influences the evolution of sperm number and sperm quality, both of which predict male fertilization success during sperm competition, albeit to varying degrees (Simmons and Fitzpatrick 2012).

The number of sperm competing to fertilize an egg(s) can dramatically influence a male’s competitive fertilization success. In many species, sperm competition is thought to follow a ‘raffle principle’, where success in the raffle (in this case fertilizing eggs) is related to the number of ‘tickets’ (in this case sperm) a male holds (Parker 1982). Under the raffle principle, all sperm have an equal chance of fertilizing eggs and therefore the probability of fertilisation during sperm competition increases with the number of sperm transferred. Therefore, males are expected to invest more in sperm number when the risk of sperm competition is elevated (Figure 1.2; Parker 1982).

28 The testes are the site of sperm production, and therefore an important target of selection for sperm number. Indeed, increases in testes size (in relation to body size) represent one of the most robust responses to sperm competition (Figure 2). The impact of sperm competition on testes size is especially well documented among primates (Harcourt et al. 1981). For example, gorillas (Gorilla gorilla) experience a very low risk of sperm competition, as a single dominant male controls a harem of females and mates with them (almost) exclusively, and have remarkably small testes for their body size. In contrast, the closely related chimpanzee (Pan troglodyte) is highly promiscuous, facilitating very high levels of sperm competition, and has testes four times larger than those of a gorilla, despite weighing a quarter of a gorilla’s body mass (Harcourt et al. 1981). A similar pattern of increasing investment in testes in response to increase sperm competition risk is observed across a much broader range of primates (Harcourt et al. 1981) and indeed across a wide range of other taxonomic groups, including birds, fish, amphibians and reptiles (Simmons and Fitzpatrick 2012).

Sperm competition not only selects for increases in sperm number but also influences the way males allocate their sperm during mating (Wedell et al. 2002). While the costs of producing an individual sperm may be negligible, the ejaculate as a whole can be costly to produce (Dewsbury 1982). Consequently, to maximize their reproductive success males are expected to strategically allocate their sperm during mating in response to cues of sperm competition (Wedell et al. 2002; Parker and Pizzari 2010). Males in many species indeed show strategic patterns of sperm allocation, but only under specific conditions. There is now clear evidence that males allocate more sperm when mating in the presence of a single rival male compared with matings where no rival males were present (Delbarco-Trillo 2011; Kelly and Jennions 2011).

However, sperm number is not the sole determinant of reproductive success during sperm competition, and under a broad range of conditions sperm quality (i.e. sperm morphology and performance) plays an important role in determining male fertility. Unique among cells, sperm must survive and travel outside the body

29 in order to fulfil their function of fertilising ova. Thus, as sperm move towards the egg, various aspects of sperm quality, including sperm motility and swimming speed, will experience intense selection (Simmons and Fitzpatrick 2012). For example, in domestic fowl (Gallus domesticus) and Atlantic salmon (Salmo salar), when females are artificially inseminated with an equal number of sperm from two males, males with greater relative sperm velocity sire more offspring (Birkhead et al. 1999; Gage et al. 2004). Thus, in most species (but for an exception see Dziminski et al. 2009; Fitzpatrick et al. 2012), sperm competition selects for increased sperm swimming speed (Simmons and Fitzpatrick 2012).

Selection on sperm quality is predicted to impact on both sperm size and speed due to an assumed link between sperm swimming speed and sperm morphology, with longer flagella expected to generate greater propulsive force and allow sperm to swim faster (Gomendio and Roldan 1991). While the underlying relationship between sperm morphology and swimming speed is far from clear (Humphries et al. 2008; Fitzpatrick et al. 2010; Simpson et al. 2014), numerous studies have evaluated how sperm size responds to varying levels of sperm competition risk. In almost every taxonomic group studied to date (including mammals, birds, fish, non-avian reptiles, amphibians, and insects) there is evidence that sperm size increases with sperm competition risk (Simmons and Fitzpatrick 2012). Moreover, a handful of recent studies have demonstrated that sperm competition also selects for faster swimming sperm across species, and this appears to be due to a positive relationship between sperm size and speed in these groups (Gomendio and Roldan 1991; Fitzpatrick et al. 2009; Lüpold et al. 2009a).

However, gaining a robust understanding of how sperm competition influences sperm size and speed has remained a contentious issue, as results contrary to the general pattern outlined above abound. Increasingly, efforts to understand how selection shapes sperm size and speed are focusing on better understanding the relationship and potential trade-off between sperm number and sperm size (Immler et al. 2011). Although this work remains limited in scope, what is becoming increasingly clear is that sperm number, size and speed are all important targets of

30 selection by sperm competition, and which of these sperm traits are favoured depends on the mechanism of sperm competition operating in a species (Immler et al. 2011). For example, sperm are produced in greater numbers at the expense of size in large species with dilute female reproductive tracts where sperm competition adheres to the raffle principle. Conversely, where sperm displacement is the primary mechanism of sperm competition, as in many insects, selection acts to increase sperm size at the expense of producing greater sperm numbers (Immler et al. 2011).

Increasingly researchers are recognizing the importance of the non-sperm component of the ejaculate - seminal fluid - in mediating male reproductive success during sperm competition (Chapman 2001). Sperm are released from males in the company of seminal fluid, a medium rich in proteins and other molecules produced by the accessory glands. These proteins exert a powerful influence on sperm competitive dynamics. First, seminal fluid proteins act on female physiology and behaviour, which has knock-on effects for the risk of sperm competition faced by a male’s ejaculate. For example, seminal proteins can reduce the risk of sperm competition experienced by an already inseminated ejaculate by reducing female receptivity to future matings (Chapman 2001), aiding the displacement of rival males’ sperm (Harshman and Prout 1994), and contributing to the production of mating plugs that block access to the female’s reproductive tract (Chapman 2001). Second, seminal proteins can act to diminish the performance of sperm from rival males. In some social insect species, sperm perform better in the presence of the male’s own seminal fluid than seminal fluid from a different, competing male (den Boer et al. 2010). However, the influence of seminal fluid on rival male sperm performance was not observed in monogamous social insect species, where sperm competition risk is low, indicating that seminal fluid function evolves in response to sperm competition risk (den Boer et al. 2010). Similarly, in the grass goby (Zosterisessor ophiocephalus), sperm of males using sneaky behaviours when attempting to fertilize eggs, and thus experiencing a higher risk of sperm competition, swim faster in the presence of the seminal fluid from a rival male, while sperm from males that use conventional courtship to woo females

31 showed reduced fertilisation rate when exposed to the seminal fluid of a sneaky male (Locatello et al. 2013). Thus seminal fluid has an important function during competitive matings and as such both the accessory gland and the proteins they produce evolve rapidly in response to sperm competition (Ramm et al. 2005; Linklater et al. 2007; Crudgington et al. 2009).

In the four decades following Parker’s (1970) recognition of the evolutionary significance of sperm competition, enormous advances have been made towards developing a comprehensive understanding of the evolutionary dynamics that govern sperm competition and its pervasive influence on male reproductive behaviour, anatomy and physiology. However, the study of sperm competition is still relatively young and many novel and exciting discoveries undoubtedly remain as advances in analytical tools and genomic approaches allow an ever-widening set of questions to be addressed. Moreover, efforts to move the focus of sperm competition studies from the researcher’s microscope to the female’s reproductive tract (Manier et al. 2010), the site of sperm competition in internal fertilizing species, promise to revolutionize our understanding of how sperm competition proceeds under ‘real world’ conditions. In this review, we focused our attention exclusively on evolutionary responses observed in males, paying particular attention to how sperm competition influences the evolution of male behaviours, genitalia, sperm and seminal fluid. However, any discussion of sperm competition inevitably runs into the related topics of cryptic female choice, where females can influence the outcome of sperm competition by biasing fertility in favour of preferred males, and sexual conflict, where the evolutionary interests between the sexes differ. While these related topics are outside the immediate scope of this entry, we acknowledge their impact on shaping reproductive traits and encourage interested readers to follow up on these topics.

(Note: This marks the end of the published entry by Rowley and Fitzpatrick (2015). See Appendix I for the published version).

32 1.3 Cryptic female choice and sexual conflict

While sperm competition is an undeniably important agent of selection, it does not happen in isolation. It is now widely recognised that females are not merely passive bystanders in post-copulatory sexual selection, but can actively influence its outcome through cryptic female choice (Eberhard 1996), post-copulatory biases towards certain reproductive phenotypes imposed by the female. There are many ways in which cryptic female choice can occur, including, but not limited to, selective sperm ejection, storage or use, which are likely to influence patterns on selection on sperm (Pizzari and Birkhead 2000; Bretman et al. 2009). For example, variation in female sperm storage organ length in Drosophila melanogaster mediates the relationship between sperm length and fertilisation success, with males that produce longer sperm siring a greater proportion offspring than males producing shorter sperm only when competing within a longer female storage organ (Miller and Pitnick 2002). Distinguishing between the processes of sperm competition and cryptic female choice is notoriously difficult (Pitnick and Hosken 2010), and it is likely that both processes interact to shape selection on male reproductive traits.

Moreover, when the reproductive interests of the sexes differ, there is potential for sexual conflict (Chapman et al. 2003b), whereby traits that are advantageous to one sex cause harm to the other, resulting in an evolutionary arms race (Parker 1979). Sexual conflict can arise as a consequence of male adaptations either to prevent or to increase performance in sperm competition, providing that they reduce female fitness (Stockley 1997; Edward et al. 2014). For example, sexual conflict over female remating rate may occur when males guard females (Jormalainen et al. 2001), insert copulatory plugs (Koprowski 1992), or inflict copulatory wounding (Reinhardt et al. 2015) to prevent females from mating with additional males, while females can obtain fitness benefits from doing so. Even seemingly benign male adaptations to sperm competition, such as higher sperm production, could present a cost to females by increasing the chances that a single ovum could be fertilised by multiple

33 sperm (polyspermy), which in most species results in a non-viable embryo (Edward et al. 2014). The hostility of the female reproductive tract to sperm may constitute a counter-adaptation to reduce the risk of polyspermy by limiting the number of sperm that ultimately reach the site of fertilisation (Birkhead et al. 1993). As a consequence of the division of reproductive function, the evolutionary interests of males and females rarely align exactly (Arnqvist and Rowe 2005), and sexual conflict is therefore likely to be a common phenomenon influencing the outcome of sperm competition. Clearly, due to myriad chemical, physical and behavioural interactions between males and females during and after mating, the study of sperm competition is closely linked to cryptic female choice and sexual conflict. Therefore, while sperm competition is the primary focus of this thesis, it must be considered within the wider selective environment to appropriately contextualise the findings.

1.4 Outstanding questions in how sperm competition influences the evolution of reproductive traits

Our understanding of sperm competition as a selective force and its influence on male anatomy, physiology and behaviour has grown exponentially in the decades since Parker’s (1970) initial description of the concept. However, there remain key gaps in our understanding of how sperm competition shapes the evolution of reproductive traits in animals. Some fundamental predictions from sperm competition theory remain surprisingly thinly supported, and the taxonomic scope of studies assessing evolutionary responses to sperm competition is generally limited, with some taxa underrepresented or entirely absent. Where comparative studies do span numerous taxonomic groups, relationships between sperm competition risk and reproductive traits at the interspecific level repeatedly show marked inconsistencies across taxa (Simmons and Fitzpatrick 2012). However, we still lack convincing explanations for these differences. Here I outline some major outstanding questions concerning the role of sperm competition in shaping the evolution of reproductive traits.

34 1.4.1 Sperm number

As detailed above (see ‘1.2 Sperm Competition’), a key prediction of sperm competition theory is that selection acts to increase male investment in testicular tissue. For this reason, body size-corrected testes mass is used as a proxy for sperm competition risk in the vast majority of comparative studies assessing evolutionary responses to sperm competition. Testes and body mass data are relatively easy to collect for a large number of species compared with, for example, behavioural observations of mating frequency, and are likely to provide a more reliable indicator of sperm competition risk than social mating system (e.g. Sefc et al. 2008). However, the key assumption that relative testes size is positively associated with the level of sperm competition males experience across has not been evaluated across all taxonomic groups. Most comparative studies of the relationship between sperm competition risk and testes mass use either social mating system or other behavioural measures as a proxy for sperm competition, and evidence of positive associations between body size-corrected testes mass and genetic measures of polyandry are restricted to a handful of studies of birds (Møller and Briskie 1995) and mammals (Ramm et al. 2005; Soulsbury 2010). Therefore, in many taxa which are frequently used as model systems for studies of post-copulatory sexual selection, the use of relative testes mass as a proxy for sperm competition is poorly validated.

Furthermore, selection may act not only on testes size but also increased sperm production via structural changes to the cellular organisation of the testes (Lüpold et al. 2009). Specifically, selection imposed by sperm competition is hypothesized to act on testes structure by increasing the proportion of sperm-producing tissue relative to interstitial space contained in the testes (Lüpold et al. 2009). However, to date only a handful of studies on mice (Montoto et al. 2012; Firman et al. 2015) and passerine birds (Lüpold et al. 2009) have evaluated the relationship between sperm competition risk and testes organisation. While all taxonomic groups examined thus far demonstrate a positive relationship between sperm competition

35 risk and the proportion of sperm-producing tissue in the testes, whether this response is universal remains unclear due to the limited taxonomic scope of these studies. For example, both mice and passerine birds are likely to be subject to functional or spatial constraints that limit the potential for increasing testes size. Therefore, increasing the proportion of sperm-producing tissue in the testes may represent an adaptation to increase sperm production limited to species in which the need to maintain a small body size or low body mass precludes the development of larger testes (e.g. Wright and Cuthill 1989). Investigating how sperm competition risk influences testes organisation in a wider variety of taxonomic groups, including those not subject to strong size and weight constraints, would shed further light on evolutionary responses of testes to sperm competition.

1.4.2 Sperm morphology

Various morphological traits have been shown to be associated with sperm competition risk (Simmons and Fitzpatrick 2012), including the size and/or shape of each of the sperm components: the head (Byrne et al. 2003; Montoto et al. 2011), midpiece (Anderson et al. 2005; Immler and Birkhead 2007) and flagellum (Gomendio and Roldan 1991; Immler and Birkhead 2007; Tourmente et al. 2011). However, comparative studies yield inconsistent relationships between sperm traits and sperm competition risk across taxa (Simmons and Fitzpatrick 2012), and the reason for this is unclear. The varying evolutionary responses of sperm morphology to sperm competition risk could reflect differences in the intraspecific relationships between sperm form and function depending on the mechanics of fertilisation, which are similarly inconsistent across species. For example, in externally fertilising species sperm with shorter heads and longer flagella tend to swim faster, whereas the opposite is true in internal fertilisers (Simpson et al. 2014). In internally fertilising species, various aspects of the female’s reproductive biology can also influence sperm evolution (Miller and Pitnick 2002; Karr et al. 2009). In addition, associations between sperm morphology and sperm competition risk can be influenced by trade-offs, for example between sperm size and sperm number

36 (Immler et al. 2011), which can lead to a negative relationship between sperm competition risk and sperm size as a correlated response to selection for increased sperm number. Resolving these inconsistent relationships is a crucial step in understanding how sperm competition shapes sperm morphology.

1.4.3 Sperm variance

Sperm competition is expected to influence not only sperm morphology, but also how sperm are produced. Where sexual selection targets sperm morphological traits, it is predicted that selection should erode variation in sperm morphology across the ejaculate (Birkhead et al. 2005), resulting in consistent production of an optimal sperm phenotype. However, the relationship between sperm variation and sperm competition risk has only been assessed by a handful of studies of passerine birds and social insects, with all evidence to date showing a consistent pattern of decreasing variation in sperm morphology in response to sperm competition (Calhim et al. 2007; Immler et al. 2008; Kleven et al. 2008; Fitzpatrick and Baer 2011). However, the generality of this pattern across broader taxonomy is not known. For example, both passerine birds and social insects exhibit sperm storage, which could reduce variation across the ejaculate if certain sperm phenotypes are more likely to enter or be retained in the female sperm storage organs. Is the relationship between sperm competition and morphological variation therefore mediated by the presence or length of sperm storage? Intriguingly, Kleven et al. (2009) document a negative relationship between variation in total sperm length and clutch size – a proxy measure for the duration of sperm storage – in passerines. A wider taxonomic focus is needed to more fully explore the evolutionary factors that influence sperm variation.

37 1.4.4 Genital morphology

As highlighted above (see ‘Sperm Competition’ section), sperm competition is also expected to influence the organs that deliver sperm to the fertilisation environment – the genitalia. While longer genitalia confer an advantage in some species (Evans et al. 2011; Dougherty et al. 2015), this relationship is not universal (Gasparini et al. 2011; Stockley et al. 2013; Booksmythe et al. 2016). Across species, positive associations between genital length and sperm competition risk have been found in some taxa (Coker et al. 2002; Ramm 2007b; Fitzpatrick et al. 2012a), but not others (Ramm et al. 2007), and the reason for this inconsistency remains unclear. Male genitalia often display appendages such as hooks, barbs, spurs spines or claws (Von Helversen and Von Helversen 1991; Crudgington and Siva-Jothy 2000; Blanckenhorn et al. 2002; Orr and Brennan 2016). While not directly involved in sperm transfer, these genital appendages have been implicated in numerous roles, including the removal of sperm from rival males (Waage 1979), acting as an ‘anchor’ during mating (Edvardsson and Tregenza 2005), stimulating the female (Dixson 1987; Stockley 2002) or discouraging remating by inflicting harm on the female (Hotzy and Arnqvist 2009). However, despite the widespread occurrence of these genital appendages, comparative studies of their evolution are rare, and the factors that drive their emergence and loss remain obscure. The size and number of genital spines are associated with a number of reproductive behaviours that are likely to increase fertilisation success (Ronn et al. 2007; Polak and Rashed 2010; Hotzy et al. 2012; Kwan et al. 2013; Friesen et al. 2014), but studies investigating how sperm competition influences their presence or absence across species are extremely limited in number and taxonomic breadth. The presence of genital spines is associated with sperm competition risk (Orr and Brennan 2016) and shorter durations of female receptivity (Stockely et al. 2002) in primates, suggesting that sexual selection promotes increasing ‘weaponisation’ of genitalia. However, the scarcity of studies directly assessing the potential covariance between genital appendages and the strength of sperm competition make it impossible to draw general conclusions at present.

38 1.5 Studying evolutionary responses of reproductive traits to sperm competition

Phylogenetic comparative methods are used to test evolutionary hypotheses by examining variation in traits across a phylogeny. In this way, it is possible to infer the macroevolutionary processes that influence broad-scale patterns (Felsenstein 1985; Harvey and Pagel 1991). As molecular genetic techniques have become more sophisticated, comprehensive, well-resolved phylogenies are increasingly available. This has greatly expanded the potential for using comparative methods, which are dependent on estimates of phylogenetic relationships between species. Modern phylogenetic methods offer researchers a powerful toolbox for investigating a range of questions about trait evolution. For instance, phylogenetically controlled regressions can be used to examine the relationship between two (or more) variables while controlling for statistical non-independence arising from shared ancestry. In recent years, the number of methods and approaches available to researchers has grown exponentially. It is now possible to control for confounding variables, which may obscure a relationship between variables of interest (Price 1997), reconstruct the evolutionary history of trait in a form of ‘statistical palaeontology’ (Cunningham et al. 1998), assess different models of trait evolution (Harmon et al. 2008), model covariance of traits under alternative evolutionary models (Tung Ho and Ané 2014) and measure and compare rates of phenotypic diversification (Adams 2013). These methods allow us to examine evolutionary responses to sperm competition risk across longer timescales and broader taxonomic scopes, providing unprecedented insight into the selective processes that have shaped the extraordinary variation in reproductive traits across the tree of life.

1.6 Sperm competition and evolutionary response in sexual traits in elasmobranchs

In this thesis, I explore how post-copulatory sexual selection arising from sperm competition influences the evolution of reproductive traits in the elasmobranch fishes (sharks and rays). Elasmobranchs represent a unique and valuable

39 opportunity for studying post-copulatory sexual selection. They occupy a critical phylogenetic position in early vertebrate evolution, display an extraordinary variety of reproductive systems, and the frequency of genetic polyandry varies widely across species. However, despite this extensive variation, the potential for postcopulatory sexual selection in elasmobranchs has until recently been largely overlooked (Fitzpatrick et al. 2012). Consequently, how sperm competition influences the evolution of elasmobranch reproductive traits and contributes to their diversification has never before been evaluated. Nevertheless, many features of their biology strongly suggest that sperm competition is a widespread and potent evolutionary force shaping reproductive traits in this group, as I summarise below.

1.6.1 Elasmobranchs

Elasmobranchs are a subclass of cartilaginous fishes (Elasmobranchii) within the class Chondrichthyes (sharks, rays and chimaeras), representing an ancient, highly conserved lineage as the oldest extant gnathstomes. The oldest recorded elasmobranchs existed over 455 million years ago, in the Paleozoic era (Klimley 2013b). Modern elasmobranchs have a circumglobal distribution and have evolved to occupy a wide range of ocean (and even freshwater) habitats and ecological niches, from shallow coastal waters and coral reefs to the abyssal plains and below the Arctic sea ice. Due to their pivotal phylogenetic position as a basal sister group to all other extant jawed vertebrates, elasmobranchs offer a unique and valuable opportunity to examine selective pressures at the base of vertebrate evolution. They represent one of the earliest vertebrate groups to develop internal fertilisation and specialised intromittent organs (Long et al. 2015) and their central nervous system represents an early yet highly conserved example of the vertebrate brain archetype (Yopak 2012). Thus, elasmobranchs have the potential to provide a window into selective patterns in early vertebrate evolution to evaluate the consistency of selection across the vertebrate tree of life.

40 1.6.2 Sharks vs rays

The elasmobranchs are divided into two superorders, the sharks (Selachii) and rays (Batoidea). The earliest rays diverged from ancient sharks approximately 300 million years ago during the adaptive radiation of the Carboniferous era. This deep ancestral split is reflected in key differences in biology. Perhaps the most significant divergence between the two lineages is in body plan. Sharks are fusiform in shape, with elongated, cylindrical bodies, whereas rays are dorsoventrally flattened and disc-shaped. These differences in body shape are likely to influence investment in reproductive traits. For example, the more restricted coelomic space of rays could impose spatial constraints that limit the size of the reproductive organs (Musick and Ellis 2005), and consequently we might expect to see differences in how sharks and rays respond to selective pressures such as sperm competition. In addition, rays are generally less mobile and have smaller distributions than sharks, possibly as a consequence of more specialised diets (Last et al. 2016; Navia et al. 2017). Thus, it is important to consider biological differences between sharks and rays when evaluating how selection acts across elasmobranchs as a whole, and it may be necessary to examine them separately due to these confounding effects.

1.6.3 Reproductive behaviours in elasmobranchs

All sharks and rays reproduce via internal fertilisation. Males initiate copulation by following females, then biting the female’s pectoral fin. Females appear to exert little pre-copulatory mate choice, though behavioural observations of wild matings indicate that they may attempt to avoid mating at this stage by twisting their body away from the male and pressing the cloaca toward the ground (Pratt and Carrier 2001; Whitney et al. 2004). Although the exact mechanics of copulation differ slightly between species depending on body shape, generally the female rolls over so that the ventral side faces upward and the male inserts one (or, rarely, both) of the paired intromittent organs known as ‘claspers’ (scroll-shaped modifications of

41 the pelvic fins - see below) into the female’s cloaca to transfer sperm. The handful of matings that have been observed in the wild indicate that females commonly mate with multiple males in quick succession during the breeding period (Pratt and Carrier 2001; Chapman et al. 2003a; Whitney et al. 2004). Female multiple mating is also commonly observed in captivity. For example, a female spotted eagle ray (Aetobatus narinari) was observed mating with four different males within an hour at the Okinawa Expo Aquarium (Klimley 2013). This evidence of polyandry suggests that sperm competition is likely to be prevalent in elasmobranchs.

Due to the challenges of observing elasmobranch matings in the wild, reproductive behaviours of sharks and rays remain relatively enigmatic. Mating seasons tend to be loosely defined in many species as a consequence of difficulty in finding mating locations or observing neonates. This is particularly true of endangered or rarely seen species, for which it is difficult or ethically dubious to catch females for dissection and/or hormone analysis in the necessary numbers. However, the available evidence suggests that both reproductive seasonality and breeding cycles are variable across species. While some species appear to breed year-round, others have clearly defined breeding seasons (Stevens and McLoughlin 1991). Similarly, females of some species are capable of producing two broods a year (Fahy et al. 2007), while others mate once a year, every two years, or even more infrequently (Klimley 2013). In contrast, males may continue to breed every year. This variation in breeding cycles likely reflects differences in the length of gestation across species. Gestation ranges from approximately 4 – 5 months in the bonnethead shark (Sphyrna tiburo) to at least 3.5 years in the frilled shark (Chlamydoselachus anguineus) (Tanaka et al. 1990). Females belonging to species with longer gestation will necessarily exhibit longer breeding cycles. Where male and female reproductive cycles differ, there is increased potential for sexual conflict, as breeding populations are likely to be strongly male-biased. For example, in a species in which females are capable of reproducing in alternate years, breeding males will likely outnumber breeding females by two to one. Under this scenario, males will experience intense competition to mate with a small number of available females, and females will suffer the costs of mating with more males, leading to conflict over mating rate.

42 Therefore, we might expect to see patterns of sexually antagonistic coevolution, with selection acting on males to overcome female reluctance to mate, and on females to avoid mating or mitigate damage caused.

1.6.4 Genetic mating systems

Advancements in molecular techniques in recent decades have confirmed that polyandrous mating is indeed a common feature of elasmobranch mating systems, and revealed that shark and ray litters are frequently sired by multiple males. When a gravid female is captured, the litter can be genotyped at highly polymorphic microsatellite loci. The maternal genotype is then effectively ‘subtracted’ and the number of remaining alleles present in the litter correspond to the likely number of putative sires (e.g. 4 extra-maternal alleles indicate a minimum of two contributing sires). The first evidence of multiple paternity in elasmobranchs was documented nearly 20 years ago in nurse sharks (Ginglymostoma cirratum) (Ohta et al. 2000). Multiple paternity was first assessed in an oviparous species (Raja clavata) in 2007 (Chevolot et al. 2007). To date, paternity has been assessed in 24 elasmobranch species, of which 22 are sharks and 2 are rays. Of the species assessed, 22 are viviparous, compared with only 2 oviparous species. As such investigations are often opportunistic due to the infrequency and unpredictability of catching pregnant sharks and rays, many studies are conducted on a very small number of broods, which limits the probability of detecting multiple paternity in a population. Viviparous species are overrepresented in paternity studies, as a whole litter can be collected with relative ease and the maternal genotype can be determined, which is often not possible in oviparous species as the female does not guard or otherwise remain associated with the eggcases.

As yet, there is no evidence of monogamy in any shark species for which more than a single brood has been examined (Fitzpatrick et al. 2012). However, both the frequency at which multiply sired litters occur within a population and the number of males that contribute offspring to a single litter varies widely across species

43 (summarised in Fitzpatrick et al. 2012), indicating that shark and ray species experience very different levels of sperm competition risk and intensity (Figure 1.4). The prevalence of multiple mating in sharks and rays, coupled with the apparent lack of pre-copulatory female choice, suggests that post-copulatory processes are likely to be more important. Females incur high costs of mating (Pratt and Carrier 2001), and assessments of the potential benefits of female multiple mating have thus far found no effect of polyandry on fecundity (Portnoy et al. 2007; DiBattista et al. 2008), inbreeding avoidance (Feldheim et al. 2004; Verissimo et al. 2011), genetic diversity (DiBattista et al. 2008; Daly-Engel et al. 2010) or offspring survival (DiBattista et al. 2008; Daly-Engel et al. 2010). This suggests that female multiple mating in elamsobranchs is likely an example of convenience polyandry, whereby females are unable to avoid mating, or do so to avoid further harassment or injury (Daly-Engel et al. 2010).

Figure 1.4. Variation in the percentage of litters sired by multiple males across shark (blue points) and ray (green points) species. Only species for which 4 or more broods were analysed are presented.

The use of multiple paternity estimates to assess sperm competition risk is not without challenges. The frequency of multiple paternity is likely to represent a conservative estimate of sperm competition risk, as by definition only males who sire offspring will be represented, while other males may be eliminated at an earlier stage through sperm competition, cryptic female choice or intrauterine cannibalism. Thus, the number of males a female mates with (i.e. the true number of sperm ‘competitors’) may be higher than the number of sires who contribute offspring to her litter. However, due to the extreme difficulty of measuring behavioural remating frequency in the vast majority of elasmobranch species over the course of an entire reproductive cycle (or even longer, given the existence of long-term sperm storage in this group (see below)), estimates of the frequency of multiple paternity represent by far the best estimates of sperm competition risk in this group.

1.6.5 Variation in reproductive traits

As a consequence of the aquatic habitat, wide distributions, and migratory lifestyles of many species, the study of elasmobranch reproduction has historically presented many challenges and thus lagged behind other taxonomic groups. In 1988, Pratt described our knowledge of elasmobranch reproductive systems and structures as ‘ranging from a modest understanding to pure conjecture’ (Pratt 1988). Fortunately, decades of study and advancements in research techniques have since catalogued the diverse reproductive physiology and breeding biology of the elasmobranch fishes (Hamlett 2005; Klimley 2013b). However, we still have very little understanding of the selective forces that shape this variation, and investigating this represents a crucial next step.

46 1.7 Reproductive Variation in Males

1.7.1 Testes

Elasmobranch testes are paired, cylindrical structures suspended from the dorsal body wall by the mesorchium, connective tissue that extends from the dorsal surface of the testis to the epididymis, and which contains the ductus efferens (small sperm transport tubules). While there is a strong correlation between testes mass and body length in sharks, many species have smaller or larger testes than would be predicted from body size alone (Fitzpatrick et al. 2012). Such variation in testes investment could be indicative of varying levels of sperm competition, which has been demonstrated in other taxa (Møller and Briskie 1995; Stockley et al. 1997; Byrne et al. 2002; Ramm et al. 2005; Soulsbury 2010; Simmons and Fitzpatrick 2012). Furthermore, a preliminary examination of the relationship between the percentage of multiple paternity and testes mass relative to body size was consistent with the hypothesis that species invest in larger testes when sperm competition risk is higher, though as data was only available for five species, a statistical analysis was not performed (Fitzpatrick et al. 2012).

In addition to variation in size, elasmobranch testes also vary in structure. Unlike many other commonly-studied taxa (e.g. birds and mammals) in which sperm is produced in permanent seminiferous tubules, which are fixed in position, elasmobranch testes are cystic in structure (in common with newts) (Schlatt and Ehmcke 2014). The basic unit of sperm production is the spermatocyst, which consists of a number of clonal germ cells (or developing spermatozoa) each associated with a single Sertoli cell, enclosed within a basement membrane (Callard 1991). Spermatocysts are formed in the germinal zone (Figure 1.5) and migrate across the testis as they are displaced by newly formed spermatocysts. This creates a gradient of sperm developmental stages extending from the germinal zone, with spermatocysts containing new germ cells located closest to the germinal zone and spermatocysts containing mature sperm most distant. On reaching the distal

47 margins of the testes, the membrane of the spermatocyst ruptures to release mature sperm to the ductus efferens, from where they are transported to and stored in the epididymis and eventually reach the ductus deferens, where they are packaged into spermatophores. There are three ‘types’ of testes, classified based on the pattern of development of the spermatocysts across the testis (Figure 1.5): diametric, radial and compound. Therefore, the locations of sperm in the testes at each stage of development will differ based on the testes type.

(a) (b) (c)

GZ GZ GZ

Figure 1.5. Variation in elasmobranch testes organisation in three shark and ray species, as illustrated by haematoxylin and eosin-stained histological testes sections. Elasmobranch testes can be classified as (a) diametric (b) radial or (c) compound, depending on the direction (indicated with white arrows) of spermatocyst progression from the germinal zone (GZ).

1.7.2 Sperm morphology

Elasmobranch sperm are comprised of three components: an elongated, helical head, a shorter midpiece, and a single flagellum (Figure 1.6). Sperm morphology is species-specific (Tanaka et al. 1995), and is conspicuously variable across species, both in terms of total length and the length of the components, each of which show approximately three-fold variation across species (Tanaka et al. 1995; Jamieson 2005). No study has yet considered the evolutionary drivers that contribute to the diversification of sperm morphology across elasmobranchs. While this high level of

48 morphological variation may result from post-copulatory selection promoting the evolution of more competitive sperm phenotypes under conditions of high sperm competition risk, this remains to be investigated. Various features of elasmobranch reproduction are likely to influence selection on sperm morphology. For example, female sharks are able to store sperm, which is likely to impose selection on sperm to enable them to enter the storage organ and to maintain viability for prolonged periods (Orr and Brennan 2015). In other taxa with sperm storage, sperm with longer flagella are better able to displace rival sperm and maintain an advantageous position prior to and during storage (Lüpold et al. 2012).

Midpiece

Head Flagellum

Figure 1.6. Elasmobranch sperm ultrastructure. A single sperm cell from a nervous shark (Carachrinus cautus), with the components indicated and head coloured with the nuclear-binding stain DAPI. Image by Rodolfo Jaffe.

1.7.3 Claspers

Male sharks and rays transfer sperm to the female reproductive tract using paired intromittent organs (claspers, Figure 1.3). Clasper morphology is so variable, even among closely related species (Muñoz-Chápuli and Ramos 1989), that it is frequently used as a diagnostic feature in elasmobranch taxonomy and species identification (e.g. Marouani et al. 2012). Elasmobranch claspers vary in both length and the presence of appendages such as across species (Leigh-Sharpe 1921, 1922, 1924, 1926). In many species, the terminal components, which are exposed when the distal tip of the clasper spreads open inside the female during copulation, consist of curved or pointed structures alternatively referred to as spurs, spines,

49 claw and/or hooks in the literature. However, these ‘weapons’ are absent in other species. Such variation is surprising, given that the mechanics of elasmobranch copulation (see above) appear broadly similar across species (Pratt and Carrier 2001; Whitney et al. 2004), and a functional explanation for differences in the presence/absence of genital weapons therefore seems unlikely. Instead, it seems likely that they serve a sexually selected role; for example, acting as a holdfast to prolong sperm transfer (Pratt and Carrier 2001), or inflicting copulatory wounding (Matthews 1950; Pratt and Carrier 2001). Although clasper length and morphological data is widely available due to its importance as a diagnostic feature in elasmobranch taxonomy and to determine sexual maturity (Clark and von Schmidt 1965), to date no study has considered how clasper morphology is shaped by sexual selection, despite clear evidence of an association between sperm competition and both genital length (Coker et al. 2002; Ramm 2007a; Fitzpatrick et al. 2012a) and the presence of spines (Orr and Brennan 2016) in other taxonomic groups.

1.8 Reproductive Variation in Females

1.8.1 Reproductive mode

Elasmobranchs exhibit an exceptionally wide range of reproductive modes. Approximately 60% of sharks are live-bearing, while the remaining 40% lay eggs. A broad spectrum of fetal-maternal trophic relationships are present in this group, from lecithotrophic oviparity to placental viviparity (though lecithotrophic, or yolk sac, viviparity is the most common mode of reproduction, and most matrotrophic species initially have a yolk sac) (Hamlett 2005). Unusually among vertebrates (with the exception of reptiles), reproductive mode is evolutionarily labile in this group, with a high number of transitions between modes of reproduction throughout Chondrichthyan evolution (Dulvy and Reynolds 1997; Musick and Ellis 2005). In modern elasmobranchs, reproductive mode is phylogenetically clustered. The majority of rays are viviparous, with the exception of the order Rajiformes, which

50 are oviparous. In sharks, the Hexanchiformes, Pristiphoriformes, Squatiniformes, Echinorhiniformes and Squaliformes are viviparous (though the exact form of embryonic development varies), and most of the Carcharhiniformes and approximately half of the Orectolobiformes are viviparous (Klimley 2013). The Heterodontiformes are oviparous, and the Lamniformes exhibit oophagy, a system in which the developing embryos obtain nourishment from unfertilised ova (Klimley 2013). Seemingly uniquely among elasmobranchs, this mode of reproduction reaches its apotheosis in the sandtiger shark (Carcharias taurus), in which the first embryo to ‘hatch’ inside the uterus eats all of the subsequent hatchlings, and can then monopolise the remaining ova (Gilmore et al. 1983). As the level of maternal investment in offspring is largely determined by foetal-maternal trophic relationship, the wide array of systems in elasmobranchs represents an excellent opportunity to study how differences in maternal provisioning influence other physiological and life history traits from a comparative perspective (Mull et al. 2011). For example, reproductive mode has been shown to influence the relationship between body size and fecundity in elasmobranchs (Goodwin et al. 2002). In addition, males are expected to invest less in testicular tissue in oviparous species than viviparous species (Coleman et al. 2009; Pollux et al. 2014), and we might therefore expect reproductive mode to influence the relationship between sperm competition and relative testes mass in elasmobranchs.

1.8.2 Fecundity

Elasmobranchs are generally slow to reach sexual maturity. While some species may be capable of reproducing in as little as 3-4 years, elasmobranch species commonly take 15-20 years to reach sexual maturity, and in some exceptional cases maturation can be as long as 156 years, as in the Greenland shark (Somniosus microcephalus) (Nielsen et al. 2016). In comparison with teleosts, sharks and rays tend to have low fecundity and large, well-developed offspring. However, fecundity varies greatly both within and between species (Musick and Ellis 2005). In 1995, a gravid whale shark (Rhincodon typus) was found to contain 304 embryos

51 encompassing a spectrum of developmental stages, which remains the largest shark litter recorded (two subsequent captures of pregnant whale sharks revealed litter sizes of 16 embryos and 200 eggs) (Schmidt et al. 2010). While brood sizes appear similar between oviparous and viviparous species overall (Wourms and Lombardi 1992), this is likely confounded by the positive association between fecundity and body size (Cortés 2000); oviparity has evolved only in taxonomic groups that consist mostly of small-bodied species (Callard et al. 1995). Indeed, oviparity is likely to represent an adaptation to allow small-bodied species to achieve higher fecundity (Holden 1973). When comparing species of similar body sizes, the fecundity of oviparous species is on average at least an order of magnitude higher than viviparous species (Goodwin et al. 2002; Musick and Ellis 2005). For example, scyliorhinid species lay an average of 60 eggs per year, whereas similarly-sized squaliforms give birth to only 4.6 pups on average over the same period (Musick and Ellis 2005). The lifetime fecundity of viviparous species is likely to be even lower, as many do not breed annually (Dodd 1983). Spatial constraints may also play a role in determining fecundity. The coelomic space in rays is restricted by their flattened, disc-shaped body plan, compared with sharks, which have a large, elongated coelom. Thus, there is likely to be more limited space available for reproductive organs in rays than in sharks, and ray uteri may therefore be smaller and have a lower capacity.

1.8.3 Sperm storage

The possibility of female sperm storage in elasmobranchs was first introduced in 1909, when it was noticed that that female skates separated from males in aquaria nevertheless continued to lay fertilised eggs (Lo Bianco 1909). It was later discovered that sperm are stored in the terminal zone of the oviducal gland (Metten 1939; Prassad 1944). Fertilisation takes place in, or immediately anterior to the oviducal gland, which also plays a role in the formation of egg cases. The size and structure of the oviducal gland varies among species; for example, it is notably larger in oviparous species (likely due to its function in shell secretion) and more

52 complex in the Heterodontiformes (Musick and Ellis 2005). Determining the exact length of sperm storage in elasmobranchs is extremely challenging, as it requires either isolating females from males in captivity after mating, which is often impractical, or dissecting the oviducal gland to look for stored sperm, which, for obvious reasons, can only be done once. Therefore, many females would have to be caught and sacrificed over the course of the entire reproductive cycle – which can be many years – to determine for how long sperm is stored with reasonable certainty. In addition, it is often not known exactly when females mated and therefore it cannot be determined how long sperm have been in storage for if they are present. However, the available evidence indicates that the likely length of storage varies across species (Klimley 2013a). Pratt (1993) hypothesised that long- term sperm storage should be common in wide-ranging, nomadic species such as the blue shark (Prionace glauca) and dusky shark (Carcharhinus obscurus), where encounter rates between males and females are likely to be lower, and short-term storage in species with prolonged ovulation, such as the Atlantic sharpnose shark (Rhizoprionodon terranovae). Sperm storage enables these females to successively fertilise ova released over a period of weeks, or even months, following mating (Walker 2005), which allows them to circumvent spatial constraints limiting the number of eggcases that can be contained in the uteri at any one time. However, this does not account for the presence of prolonged storage in species which do not meet these criteria (Klimley 2013a), or the extreme storage durations observed in captive females, which can produce offspring after as long as five years of isolation from males (Bernal et al. 2015). By uncoupling mating from fertilisation, sperm storage increases the potential for sperm competition risk as males may be confronted with rival ejaculates from past mating seasons, and raises the intriguing possibility that females may exert cryptic female choice (Eberhard 1996) by selectively storing or utilising sperm to bias paternity in favour of preferred males.

53 1.9 Challenges in studying elasmobranch reproduction

Clearly, there is an abundance of diversity in reproductive systems and structures across sharks and rays. However, capitalising on this variation presents significant challenges. Elasmobranchs are widely (and in the case of many species, sparsely) distributed, and occupy hard-to-reach habitats, such as the open ocean or abyssal plains. Comparative studies of elasmobranch biology therefore require a global perspective and intense sampling effort. In addition, studying the reproductive anatomy requires access to dead specimens for dissection. As elasmobranchs are experiencing a conservation crisis, with one third of oceanic sharks threatened with extinction according to the IUCN red list (Stein et al 2018), sacrificing sharks and rays for this purpose is both challenging and raises important ethical considerations.

For these reasons, I took a three-pronged approach to collect reproductive data from as many shark and ray species as possible. First, I visited commercial and artisanal fisheries around the world to opportunistically sample shark and ray specimens available for commercial sale. Elasmobranchs are routinely caught in both targeted fisheries and as bycatch. As the reproductive organs are usually discarded before the fish are sold, it is possible to sample a wide variety of species within a fraction of the time that would be required to catch them specifically for research, at a much lower cost and without inflicting additional or unnecessary mortality. Using this approach, I was able collect morphological data from shark and ray species at field sites in 9 countries across 5 continents. Second, to supplement this field sampling, I also compiled data collected from the wealth of published sources spanning more than a century of study of elasmobranch biology. Third, I collected samples from captive individuals that died of natural causes or were euthanized in aquaria.

54 1.10 Aims of the thesis

The characteristics of elasmobranch reproduction suggest that sharks and rays experience a high level of sperm competition, though the exact level appears to vary across species. This has important implications for our understanding of how reproductive traits evolve in elasmobranch fishes, and across vertebrates more generally. However, despite the sustained public fascination with sharks and rays, and a long history of academic study, we know surprisingly little about how sexual selection acts in this group (Fitzpatrick et al. 2012). However, this has never before been investigated, which is surprising as sharks and rays represent a highly promising model for studying the reproductive evolution, and are likely to yield fresh insights into post-copulatory sexual selection. In this thesis, I address this crucial gap and present the first investigation of how sperm competition risk influences male reproductive traits in sharks and rays. The specific questions I aimed to address are as follows:

(1) Does the percentage of multiple paternity (sperm competition risk) and the number of sires per litter (sperm competition intensity) influence investment in testes across sharks and bony fishes? (Chapter 2) (2) Does sperm competition risk affect the proportion of sperm-producing tissue in sharks and rays? (Chapter 3) (3) Does sperm competition risk influence the length of sperm components and variation in sperm morphology across the ejaculate in sharks and rays? (Chapter 4) (4) Do elasmobranch sperm components evolve at different rates? (Chapter 4) (5) Where and how often do clasper appendages evolve across the shark phylogeny? (Chapter 5) (6) Does sperm competition risk influence clasper length and the presence or absence of clasper appendages across sharks? (Chapter 5)

55 1.11 References

Adams, D. C. 2013. Comparing evolutionary rates for different phenotypic traits on a phylogeny using likelihood. Syst. Biol. 62:181–192. Anderson, M. J., J. Nyholt, and A. F. Dixson. 2005. Sperm competition and the evolution of sperm midpiece volume in mammals. J. Zool. 267:135–142. Andersson, M. 1982. Female choice selects for extreme tail length in a widowbird. Nature 299:818–820. Andersson, M. 1994. Sexual Selection. Princeton University Press, Princeton, New Jersey, USA. Andersson, M., and L. W. Simmons. 2006. Sexual selection and mate choice. Trends Ecol. Evol. 21(6):296-302 Arnqvist, G., and L. Rowe. 2005. Sexual Conflict. Princeton University Press, Princeton, New Jersey, USA. Bernal, M. A., N. L. Sinai, C. Rocha, M. R. Gaither, F. Dunker, and L. A. Rocha. 2015. Long-term sperm storage in the brownbanded bamboo shark Chiloscyllium punctatum. J. Fish Biol. 86:1171–1176. Birkhead, T. R. 2000. Promiscuity: An Evolutionary History of Sperm Competition. Harvard University Press, Cambridge, Massachusetts, USA. Birkhead, T. R., J. G. Martínez, T. Burke, and D. P. Froman. 1999. Sperm mobility determines the outcome of sperm competition in the domestic fowl. Proc. R. Soc. London B 266:1759–64. Birkhead, T. R., and A. P. Møller. 1998. Sperm Competition and Sexual Selection. 1st ed. Academic Press, San Diego, California. Birkhead, T. R., A. P. Møller, and W. J. Sutherland. 1993. Why do females make it so difficult for males to fertilize their eggs? J. Theor. Biol. 161:51–60. Birkhead, T. R., E. J. Pellat, P. Brekke, R. Yeates, and H. Castillo-Juarez. 2005. Genetic effects on sperm design in the zebra finch. Nature 434:383–387. Blanckenhorn, W. U., D. J. Hosken, O. Y. Martin, C. Reim, Y. Teuschl, and P. I. Ward. 2002. The costs of copulating in the dung fly Sepsis cynipsea. Behav. Ecol. 13:353–358. Booksmythe, I., M. L. Head, J. S. Keogh, and M. D. Jennions. 2016. Fitness

56 consequences of artificial selection on relative male genital size. Nat. Commun. 7:11597. Bretman, A., D. Newcombe, and T. Tregenza. 2009. Promiscuous females avoid inbreeding by controlling sperm storage. Mol. Ecol. 18:3340–3345. Bro-Jørgensen, J. 2007. The intensity of sexual selection predicts weapon size in male bovids. Evolution. 61:1316–1326. Byrne, P. G., J. D. Roberts, and L. W. Simmons. 2002. Sperm competition selects for increased testes mass in Australian frogs. J. Evol. Biol. 15:347–355. Byrne, P. G., L. W. Simmons, and J. D. Roberts. 2003. Sperm competition and the evolution of gamete morphology in frogs. Proc. R. Soc. B Biol. Sci. 270:2079– 2086. Calhim, S., S. Immler, and T. R. Birkhead. 2007. Postcopulatory sexual selection is associated with reduced variation in sperm morphology. PLoS One 2:e413. Callard, I. P., O. Putz, M. Paolucci, and T. J. Koob. 1995. Elasmobranch reproductive life histories: Endocrine correlates and evolution. Pp. 204–208 in F. Goetz and P. Thomas, eds. Proceedings of the Fifth International Symposium on the Reproductive Physiology of Fish. Austin, Texas. Callard, G. V. 1991. Spermatogenesis. Pp. 303–341 in P. K. T. Pang and M. P. Schreibman, eds. Vertebrate Endocrinology: Fundamentals and Biomedical Implications, Volume 4. Academic Press. Cayetano, L., A. a Maklakov, R. C. Brooks, and R. Bonduriansky. 2011. Evolution of male and female genitalia following release from sexual selection. Evolution 65:2171–83. Chapman, D. D., M. J. Corcoran, G. M. Harvey, S. Malan, and M. S. Shivji. 2003a. Mating behavior of southern stingrays, Dasyatis americana (Dasyatidae). Environ. Biol. Fishes 68:241–245. Chapman, T. 2001. Seminal fluid-mediated fitness traits in Drosophila. Heredity. 87:511–521. Chapman, T., G. Arnqvist, J. Bangham, and L. Rowe. 2003b. Sexual conflict. Trends Ecol. Evol. 18:41–47. Chevolot, M., J. R. Ellis, A. D. Rijnsdorp, W. T. Stam, and J. L. Olsen. 2007. Multiple paternity analysis in the thornback ray Raja clavata L. J. Hered. 98:712–715.

57 Clark, E., and K. von Schmidt. 1965. Sharks of the Central Gulf Coast of Florida. Bull. Mar. Sci. 15:13–83. Coker, C. R., F. McKinney, H. Hays, and K. M. Cheng. 2002. Intromittent Organ Morphology and Testis Size in Relation to Mating System in Waterfowl. Auk 119:403–413. Coleman, S. W., A. Harlin-Cognato, and A. G. Jones. 2009. Reproductive isolation, reproductive mode, and sexual selection: empirical tests of the viviparity- driven conflict hypothesis. Am. Nat. 173:291–303. Cortés, E. 2000. Life History Patterns and Correlations in Sharks. Rev. Fish. Sci. 8:299–344. Crudgington, H. S., S. Fellows, N. S. Badcock, and R. R. Snook. 2009. Experimental manipulation of sexual selection promotes greater male mating capacity but does not alter sperm investment. Evolution. 63:926–38. Crudgington, H. S., and M. T. Siva-Jothy. 2000. Genital damage , kicking and early death. Nature 407:855–856. Cunningham, C. W., K. E. Omland, and T. H. Oakley. 1998. Reconstructing ancestral character states: A critical reappraisal. Trends Ecol. Evol. 13(9):361-366 Daly-Engel, T. S., R. D. Grubbs, K. A. Feldheim, B. W. Bowen, and R. J. Toonen. 2010. Is multiple mating beneficial or unavoidable? Low multiple paternity and genetic diversity in the shortspine spurdog Squalus mitsukurii. Mar. Ecol. Prog. Ser. 403:255–267. Darwin, C. 1859. On the Origin of Species by Means of Natural Selection. John Murray, London. Darwin, C. 1871. The Descent of Man and Selection in Relation to Sex. John Murray, London. Delbarco-Trillo, J. 2011. Adjustment of sperm allocation under high risk of sperm competition across taxa: a meta-analysis. J. Evol. Biol. 24:1706–14. den Boer, S. P. a, B. Baer, and J. J. Boomsma. 2010. Seminal fluid mediates ejaculate competition in social insects. Science. 327:1506–9. DiBattista, J. D., K. a Feldheim, S. H. Gruber, and A. P. Hendry. 2008. Are indirect genetic benefits associated with polyandry? Testing predictions in a natural population of lemon sharks. Mol. Ecol. 17:783–95.

58 Dixson, A. F. 1987. Observations on the evolution of the genitalia and copulatory behaviour in male primates. J. Zool. 213:423–443. Dodd, J. M. 1983. Reproduction in Cartilaginous Fishes (Chondrichthyes). Pp. 31–95 in W. S. Hoar, D. J. Randall, and E. M. Donaldson, eds. Fish Physiology. Academic Press, New York. Dougherty, L. R., I. A. Rahman, E. R. Burdfield-Steel, E. V. Greenway, and D. M. Shuker. 2015. Experimental reduction of intromittent organ length reduces male reproductive success in a bug. Proc. R. Soc. B Biol. Sci. 282:20150724 Drickamer, L. C., P. A. Gowaty, and C. M. Holmes. 2000. Free female mate choice in house mice affects reproductive success and offspring viability and performance. Anim. Behav. 59:371–378. Dulvy, N. K., and J. D. Reynolds. 1997. Evolutionary transitions among egg-laying, live-bearing and maternal inputs in sharks and rays. Proc. R. Soc. B Biol. Sci. 264:1309–1315. Dziminski, M. a, J. D. Roberts, M. Beveridge, and L. W. Simmons. 2009. Sperm competitiveness in frogs: slow and steady wins the race. Proc. R. Soc. London B 276:3955–61. Eberhard, W. G. 1996. Female Control: Sexual Selection by Cryptic Female Choice. Princeton University Press, Princeton. Eberhard, W. G. 2009. Postcopulatory sexual selection: Darwin’s omission and its consequences. Proc. Natl. Acad. Sci. 106:10025–10032. Edvardsson, M., and T. Tregenza. 2005. Why do male Callosobruchus maculatus harm their mates? Behav. Ecol. 16:788–793. Edward, D. A., P. Stockley, and D. J. Hosken. 2014. Sexual Conflict and Sperm Competition. Cold Spring Harb. Perspect. Biol. 7:1–11. Evans, J. P., C. Gasparini, G. I. Holwell, I. W. Ramnarine, T. E. Pitcher, and A. Pilastro. 2011. Intraspecific evidence from guppies for correlated patterns of male and female genital trait diversification. Proc. R. Soc. B Biol. Sci. 278:2611–2620. Fahy, D. P., R. E. Spieler, and W. C. Hamlett. 2007. Preliminary observations on the reproductive cycle and uterine fecundity of the yellow stingray, Urobatis jamaicensis (Elasmobranchii: Myliobatiformes: Urolophidae) in Southeast Florida, U.S.A. Raffles Bull. Zool. Supplement:131–139.

59 Feldheim, K. A., S. H. Gruber, and M. V. Ashley. 2004. Reconstruction of parental microsatellite genotypes reveals female polyandry and philopatry in the lemon shark, Negaprion brevirostris. Evolution. 58:2332–2342. Felsenstein, J. 1985. Phylogenies and the Comparative Method. Am. Nat. 125:1–15. Firman, R. C., F. Garcia-Gonzalez, E. Thyer, S. Wheeler, Z. Yamin, M. Yuan, and L. W. Simmons. 2015. Evolutionary change in testes tissue composition among experimental populations of house mice. Evolution. 69:848–55. Firman, R. C., and L. W. Simmons. 2014. Experimental evolution of sperm quality via postcopulatory sexual selection in house mice. Evolution. 64:1245–1256. Fitzpatrick, J. L., M. Almbro, A. Gonzalez-Voyer, N. Kolm, and L. W. Simmons. 2012a. Male contest competition and the coevolution of weaponry and testes in pinnipeds. Evolution. 66:3595–3604. Fitzpatrick, J. L., and B. Baer. 2011. Polyandry reduces sperm length variation in social insects. Evolution. 65:3006–3012. Fitzpatrick, J. L., F. Garcia-gonzalez, and J. P. Evans. 2010. Linking sperm length and velocity : the importance of intramale variation. Biol. Lett. 6:797–799. Fitzpatrick, J. L., R. M. Kempster, T. S. Daly-Engel, S. P. Collin, and J. P. Evans. 2012. Assessing the potential for post-copulatory sexual selection in elasmobranchs. J. Fish Biol. 80:1141–58. Fitzpatrick, J. L., R. Montgomerie, J. K. Desjardins, K. a Stiver, N. Kolm, and S. Balshine. 2009. Female promiscuity promotes the evolution of faster sperm in cichlid fishes. Proc. Natl. Acad. Sci. U. S. A. 106:1128–32. Fitzpatrick, J. L., L. W. Simmons, and J. P. Evans. 2012b. Complex patterns of multivariate selection on the ejaculate of a broadcast spawning marine invertebrate. Evolution. 66:2451–60. Friesen, C. R., A. R. Kerns, and R. T. Mason. 2014. Factors influencing paternity in multiply mated female red-sided garter snakes and the persistent use of sperm stored over winter. Behav. Ecol. Sociobiol. 68:1419–1430. Fromhage, L. 2006. Emasculation to plug up females: the significance of pedipalp damage in Nephila fenestrata. Behav. Ecol. 17:353–357. Gage, M. J. G., C. P. Macfarlane, S. Yeates, R. G. Ward, J. B. Searle, and G. A. Parker. 2004. Spermatozoal Traits and Sperm Competition in Atlantic Salmon. Curr.

60 Biol. 14:44–47. Gasparini, C., A. Pilastro, and J. P. Evans. 2011. Male genital morphology and its influence on female mating preferences and paternity success in guppies. PLoS One 6:e22329. Gilmore, R. G., J. W. Dodrill, and P. A. Linley. 1983. Reproduction and Embryonic Development of the Sand Tiger Shark, Odontaspis taurus. Fish. Bull. 81:201– 226. Gomendio, M., and E. R. Roldan. 1991. Sperm competition influences sperm size in mammals. Proc. R. Soc. London B 243:181–5. Goodwin, N. B., N. K. Dulvy, and J. D. Reynolds. 2002. Life-history correlates of the evolution of live bearing in fishes. Philos. Trans. R. Soc. B Biol. Sci. 357:259– 267. Gross, M. R. 1996. Alternative reproductive tactics : diversity within sexes. Trends Ecol. Evol. 11:92–98. Hamlett, W. C. 2005. Reproductive Biology and Phylogeny of Chondrichthyes: Sharks, Batoids, and Chimaeras. CRC Press. Harcourt, A. H., P. H. Harvey, S. G. Larson, and R. V. Short. 1981. Testis weight, body weight and breeding system in primates. Nature 293:55–56. Harmon, L. J., J. T. Weir, C. D. Brock, R. E. Glor, and W. Challenger. 2008. GEIGER: Investigating evolutionary radiations. Bioinformatics 24:129–131. Harshman, L. G., and T. Prout. 1994. Sperm Displacement Without Sperm Transfer in Drosophila melanogaster. Evolution. 48:758–766. Harvey, P. H., and M. D. Pagel. 1991. The Comparative Method in Evolutionary Biology. Oxford University Press. Holden, M. J. 1973. Are long-term sustainable fisheries for elasmobranchs possible? Rapp. Process. des Reun. du Cons. Int. pour l"Expoloration la Mer 164:360– 367. Hosken, D. J. 2003. Sperm Biology: Size Indeed Matters. Curr. Biol. 13:355–356. Hosken, D. J., and P. Stockley. 2004. Sexual selection and genital evolution. Trends Ecol. Evol. 19:87–93. Hotzy, C., and G. Arnqvist. 2009. Sperm Competition Favors Harmful Males in Seed Beetles. Curr. Biol. 19:404–407.

61 Hotzy, C., M. Polak, J. L. Rönn, and G. Arnqvist. 2012. Phenotypic engineering unveils the function of genital morphology. Curr. Biol. 22:2258–2261. Humphries, S., J. P. Evans, and L. W. Simmons. 2008. Sperm competition: linking form to function. BMC Evol. Biol. 8:319. Immler, S., and T. R. Birkhead. 2007. Sperm competition and sperm midpiece size: no consistent pattern in passerine birds. Proceedings. Biol. Sci. 274:561–8. Immler, S., S. Calhim, and T. R. Birkhead. 2008. Increased postcopulatory sexual selection reduces the intramale variation in sperm design. Evolution. 62:1538– 1543. Immler, S., S. Pitnick, G. A. Parker, K. L. Durrant, S. Lüpold, S. Calhim, and T. R. Birkhead. 2011. Resolving variation in the reproductive tradeoff between sperm size and number. Proc. Natl. Acad. Sci. 108:5325–5330. Jamieson, B. G. M. 2005. Chondrichthyan Spermatozoa and Phylogney. Pp. 201–236 in W. C. Hamlett, ed. Reproductive Biology and Phylogeny of Chondrichthyes: Sharks, Batoids and Chimaeras. Science Publishers. Jormalainen, V., S. Merilaita, and J. Riihimäki. 2001. Costs of intersexual conflict in the isopod Idotea baltica. J. Evol. Biol. 14:763–772. Karr, T. L., W. J. Swanson, and R. R. Snook. 2009. The evolutionary significance of variation in sperm-egg interactions. Pp. 305–365 in Sperm Biology. Elsevier. Kelly, C. D., and M. D. Jennions. 2011. Sexual selection and sperm quantity: meta- analyses of strategic ejaculation. Biol. Rev. 86:863–884. Kleven, O., F. Fossøy, T. Laskemoen, R. J. Robertson, G. Rudolfsen, and J. T. Lifjeld. 2009. Comparative evidence for the evolution of sperm swimming speed by sperm competition and female sperm storage duration in passerine birds. Evolution. 63:2466–2473. Kleven, O., T. Laskemoen, F. Fossøy, R. J. Robertson, and J. T. Lifjeld. 2008. Intraspecific variation in sperm length is negatively related to sperm competition in passerine birds. Evolution. 62:494–499. Klimley, A. P. 2013a. Courtship and Reproduction. in The Biology of Sharks and Rays. University of Chicago Press. Klimley, A. P. 2013b. The Biology of Sharks and Rays. University of Chicago Press, Chicago.

62 Koprowski, J. L. 1992. Removal of Copulatory Plugs by Female Tree Squirrels. J. Mammal. 73:572–576. Kwan, L., Y. Y. Cheng, F. H. Rodd, and L. Rowe. 2013. Sexual conflict and the function of genitalic claws in guppies (Poecilia reticulata). Biol. Lett. 9:20130267. Last, P. R., W. T. White, M. R. De Carvalho, B. Séret, M. F. W. Stehmann, and G. J. P. Naylor. 2016. Rays of the World. Cornell University Press, Ithaca, NY. Le Boeuf, B. J. 1974. Male-male competition and reproductive success in elephant seals. Integr. Comp. Biol. 14:163–176. Leigh-Sharpe, W. H. 1921. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir II. J. Morphol. 35:359–381. Leigh-Sharpe, W. H. 1926. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir IX. J. Morphol. 42:321–334. Leigh-Sharpe, W. H. 1924. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir VI. J. Morphol. 39:553–566. Leigh-Sharpe, W. H. 1922. The comparative morphology of the secondary sexual characters of holocephali and elasmobranch fishes the claspers, clasper siphons, and clasper glands. Memoir V. J. Morphol. 36:221–243. Linklater, J. R., B. Wertheim, S. Wigby, and T. Chapman. 2007. Ejaculate depletion patterns evolve in response to experimental manipulation of sex ratio in Drosophila melanogaster. Evolution. 61:2027–34. Lo Bianco, S. 1909. Notizie biologiche reguardanti specialmente il periodo di maturita sessuale degli animali del golfo di Napoli (Biological notes on the animals of the Gulf of Naples with special emphasis on periods of sexual maturation). Mitth. Zool. Stat. Napol. 19:513–761. Locatello, L., F. Poli, and M. Rasotto. 2013. Tactic-specific differences in seminal fluid influence sperm performance. Proc. Biol. Sci. 280: 20122891. Long, J. A., E. Mark-Kurik, Z. Johanson, M. S. Y. Lee, G. C. Young, Z. Min, P. E. Ahlberg, M. Newman, R. Jones, J. den Blaauwen, B. Choo, and K. Trinajstic.

63 2015. Copulation in antiarch placoderms and the origin of gnathostome internal fertilization. Nature 517:196–199. Lüpold, S., S. Calhim, S. Immler, and T. R. Birkhead. 2009a. Sperm morphology and sperm velocity in passerine birds. Proc. Biol. Sci. 276:1175–81. Lüpold, S., G. M. Linz, J. W. Rivers, D. F. Westneat, and T. R. Birkhead. 2009. Sperm competition selects beyond relative testes size in birds. Evolution (N. Y). 63:391–402. Lüpold, S., M. K. Manier, K. S. Berben, K. J. Smith, B. D. Daley, S. H. Buckley, J. M. Belote, and S. Pitnick. 2012. How multivariate ejaculate traits determine competitive fertilization success in Drosophila melanogaster. Curr. Biol. 22:1667–1672. Lüpold, S., J. L. Tomkins, L. W. Simmons, and J. L. Fitzpatrick. 2014. Female monopolization mediates the relationship between pre- and postcopulatory sexual traits. Nat. Commun. 5:3184. Manier, M. K., J. M. Belote, K. S. Berben, D. Novikov, W. T. Stuart, and S. Pitnick. 2010. Resolving mechanisms of competitive fertilization success in Drosophila melanogaster. Science. 328:354–7. Marouani, S., R. Chaâba, H. Kadri, B. Saidi, A. Bouain, F. Maltagliati, P. Last, B. Séret, and M. N. Bradai. 2012. Taxonomic research on Squalus megalops (Macleay, 1881) and Squalus blainvillei (Risso, 1827) (Chondrichthyes: Squalidae) in Tunisian waters (central Mediterranean Sea). Sci. Mar. 76:97–109. Matthews, L. H. 1950. Reproduction in the basking shark, Cetorhinus maximus (Gunner). Philos. Trans. R. Soc. Lond. B. Biol. Sci. 234:247–316. Metten, H. 1939. Studies on the Reproduction of the Dogfish. Philos. Trans. R. Soc. B Biol. Sci. 230:217–238. Milinski, M., and T. C. M. Bakker. 1990. Female sticklebacks use male coloration in mate choice and hence avoid parasitized males. Nature 344:330–333. Miller, G. T., and S. Pitnick. 2002. Sperm-female coevolution in Drosophila. Science. 298:1230–1233. Møller, A. P., and J. V. Briskie. 1995. Extra-pair paternity, sperm competition and the evolution of testis size in birds. Behav. Ecol. Sociobiol. 36:357–365. Montgomerie, R., and J. L. Fitzpatrick. 2009. Reproductive Biology and Phylogeny of

64 Fishes. 1st ed. Science Publishers, Enfield, New Hampshire. Montoto, L. G., L. Arregui, N. M. Sanchez, M. Gomendio, and E. R. S. Roldan. 2012. Postnatal testicular development in mouse species with different levels of sperm competition. Reproduction 143:333–346. Montoto, L. G., M. V. Sánchez, M. Tourmente, J. Martín-Coello, J. J. Luque-Larena, M. Gomendio, and E. R. S. Roldan. 2011. Sperm competition differentially affects swimming velocity and size of spermatozoa from closely related muroid rodents: Head first. Reproduction 142:819–830. Mull, C. G., K. E. Yopak, and N. K. Dulvy. 2011. Does more maternal investment mean a larger brain? Evolutionary relationships between reproductive mode and brain size in chondrichthyans. Marine and Freshwater Research. 62:567– 575 Muñoz-Chápuli, R., and F. Ramos. 1989. Morphological comparison of Squalus blainvillei and S. megalops in the Eastern Atlantic, with notes on the genus. Japanese J. Ichthyol. 36:6–21. Musick, J. A., and J. K. Ellis. 2005. Reproductive Evolution of Chondrichthyans. Pp. 45–79 in W. C. Hamlett, ed. Reproductive Biology and Phylogeny of Chondrichthyes: Sharks, Batoids and Chimaeras. Science Publishers, Inc., Plymouth, UK. Navia, A. F., P. A. Mejía-Falla, J. López-García, A. Giraldo, and V. H. Cruz-Escalona. 2017. How many trophic roles can elasmobranchs play in a marine tropical network? Mar. Freshw. Res. 68:1342–1353. Nielsen, J., R. B. Hedeholm, J. Heinemeier, P. G. Bushnell, J. S. Christiansen, J. Olsen, C. B. Ramsey, R. W. Brill, M. Simon, K. F. Steffensen, and J. F. Steffensen. 2016. Eye lens radiocarbon reveals centuries of longevity in the Greenland shark (Somniosus microcephalus). Science. 353:702–704. Ohta, Y., K. Okamura, E. C. McKinney, S. Bartl, K. Hashimoto, and M. F. Flajnik. 2000. Primitive synteny of vertebrate major histocompatibility complex class I and class II genes. Proc. Natl. Acad. Sci. U. S. A. 97:4712–4717. Orr, T. J., and P. L. R. Brennan. 2016. All Features Great and Small - The Potential Roles of the Baculum and Penile Spines in Mammals. Integrative and Comparative Biology. 56:635–643.

65 Orr, T. J., and P. L. R. Brennan. 2015. Sperm storage: Distinguishing selective processes and evaluating criteria. Trends Ecol. Evol. 30:261-272. Parker, G. A. 1982. Experimental evidence for the evolution of the mammalian baculum by sexual selection. J. Theor. Biol. 96:281–294. Parker, G. A. 1979. Sexual selection and sexual conflict. Pp. 123–166 in M. Blum, ed. Sexual Selection and Reproductive Competition in Insects. Academic Press. Parker, G. A. 1970. Sperm competition and it’s evolutionary consequences in the insects. Biol. Rev. 45:525–567. Parker, G. A., C. M. Lessells, and L. W. Simmons. 2013. Sperm competition games: A general model for precopulatory male-male competition. Evolution. 67:95– 109. Parker, G. a, and T. Pizzari. 2010. Sperm competition and ejaculate economics. Biol. Rev. Camb. Philos. Soc. 85:897–934. Pitnick, S., and D. J. Hosken. 2010. Postcopulatory sexual selection. Pp. 379–399 in D. F. Westneat and C. W. Fox, eds. Evolutionary Behavioural Ecology. Oxford University Press. Pizzari, T., and T. R. Birkhead. 2000. Female feral fowl eject sperm of subdominant males. Nature 405:787–789. Polak, M., and A. Rashed. 2010. Microscale laser surgery reveals adaptive function of male intromittent genitalia. Proc. R. Soc. B Biol. Sci. 277:1371–1376. Pollux, B. J. a., R. W. Meredith, M. S. Springer, and D. N. Reznick. 2014. The evolution of the placenta drives a shift in sexual selection in livebearing fish. Nature 513:233–236. Portnoy, D. S., A. N. Piercy, J. a Musick, G. H. Burgess, and J. E. Graves. 2007. Genetic polyandry and sexual conflict in the sandbar shark, Carcharhinus plumbeus, in the western North Atlantic and Gulf of Mexico. Mol. Ecol. 16:187–97. Prassad, R. R. 1944. The structure, phylogenetic significance, and function of the nidamental glands of some elasmobranchs of the Madras coast. Proc. Nat. Inst. Sci. India, Part B, Biol. Sci. 11:282–302. Pratt, H. L. 1988. Elasmobranch gonad structure: a description and survey. Copeia 1988:719–729.

66 Pratt, H. L. 1993. The storage of spermatozoa in the oviducal glands of western North Atlantic sharks. Environ. Biol. Fishes 14:139–149. Pratt, H. L., and J. C. Carrier. 2001. A review of elasmobranch reproductive behavior with a case study on the nurse shark , Ginglymostoma cirratum. Environ. Biol. Fishes 60:157–188. Price, T. 1997. Correlated evolution and independent contrasts. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 352:519-529. Ramm, S. 2007. Sexual selection and genital evolution in mammals: a phylogenetic analysis of baculum length. Am. Nat. 169:360–369. Ramm, S. A., G. A. Parker, and P. Stockley. 2005. Sperm competition and the evolution of male reproductive anatomy in rodents. Proc. R. Soc. London B 272:949–55. Reinhardt, K., N. Anthes, and R. Lange. 2015. Copulatory wounding and traumatic insemination. Cold Spring Harb. Perspect. Biol. 7:1–26. Ronn, J., M. Katvala, and G. Arnqvist. 2007. Coevolution between harmful male genitalia and female resistance in seed beetles. Proc. Natl. Acad. Sci. 104:10921–10925. Ryan, M. J., J. H. Fox, W. Wilczynski, and A. S. Rand. 1990. Sexual selection for sensory exploitation in the frog Physalaemus pustulosus. Nature 343:66–67. Schlatt, S., and J. Ehmcke. 2014. Regulation of spermatogenesis: An evolutionary biologist’s perspective. Semin. Cell. Dev. Biol. 29:2-16. Schmidt, J. V., C. C. Chen, S. I. Sheikh, M. G. Meekan, B. M. Norman, and S. J. Joung. 2010. Paternity analysis in a litter of whale shark embryos. Endanger. Species Res. 12:117–124. Sefc, K. M., K. Mattersdorfer, C. Sturmbauer, and S. Koblmüller. 2008. High frequency of multiple paternity in broods of a socially monogamous cichlid fish with biparental nest defence. Mol. Ecol. 17:2531–2543. Simmons, L. W., and J. L. Fitzpatrick. 2012. Sperm wars and the evolution of male fertility. Reproduction 144:519–34. Simmons, L. W., C. M. House, J. Hunt, and F. García-González. 2009. Evolutionary response to sexual selection in male genital morphology. Curr. Biol. 19:1442–6. Simpson, J. L., S. Humphries, J. P. Evans, L. W. Simmons, and J. L. Fitzpatrick. 2014.

67 Relationships between sperm length and speed differ among three internally and three externally fertilizing species. Evolution. 68:92–104. Soulsbury, C. D. 2010. Genetic Patterns of Paternity and Testes Size in Mammals. PLoS One 5:e9581. Stein, R. W., C. G. Mull, T. S. Kuhn, N. C. Aschliman, L. N. K. Davidson, J. B. Joy, G. J. Smith, N. K. Dulvy, and A. O. Mooers. 2018. Global priorities for conserving the evolutionary history of sharks, rays and chimaeras. Nat. Ecol. Evol. 2:288–298.

Stevens, J., and K. McLoughlin. 1991. Distribution, size and sex composition, reproductive biology and diet of sharks from Northern Australia. Stockley, P. 1997. Sexual conflict resulting from adaptations to sperm competition. Trends Ecol. Evol. 12:154-9 Stockley, P. 2002. Sperm competition risk and male genital anatomy: comparative evidence for reduced duration of female sexual receptivity in primates with penile spines. Evol. Ecol. 123–137. Stockley, P., M. J. Gage, G. A. Parker, and A. P. Møller. 1997. Sperm competition in fishes: the evolution of testis size and ejaculate characteristics. Am. Nat. 149:933–954. Stockley, P., S. a Ramm, A. L. Sherborne, M. D. F. Thom, S. Paterson, and J. L. Hurst. 2013. Baculum morphology predicts reproductive success of male house mice under sexual selection. BMC Biol. 11:66. Tanaka, S., H. Kurokawa, and M. Hara. 1995. Comparative morphology of the sperm in chondrichthyan fishes. Pp. 313–320 in B. G. M. Jamieson, J. Ausio, and J.-L. Justine, eds. Advances in Spermatozoal Phylogeny and Taxonomy. Tanaka, S., Y. Shiobara, S. Hioki, H. Abe, G. Nishi, K. Yano, and K. Suzuki. 1990. The Reproductive-Biology of the Frilled Shark, Chlamydoselachus anguineus, from Suruga Bay, Japan. Japanese J. Ichthyol. 37:273–291. Thornhill, R. 1983. Cryptic Female Choice and Its Implications in the Scorpionfly Harpobittacus nigriceps. Am. Nat. 122:765–788. Tourmente, M., M. Gomendio, and E. R. Roldan. 2011. Sperm competition and the evolution of sperm design in mammals. BMC Evol. Biol. 11:12. Tung Ho, L. si, and C. Ané. 2014. A linear-time algorithm for gaussian and non-

68 gaussian trait evolution models. Syst. Biol. 63:397–408. Verissimo, A., D. Grubbs, J. McDowell, J. Musick, and D. Portnoy. 2011. Frequency of Multiple Paternity in the Spiny Dogfish Squalus acanthias in the Western North Atlantic. J. Hered. 102:88–93. Von Helversen, D., and O. Von Helversen. 1991. Pre-mating sperm removal in the bushcricket Metaplastes ornatus Ramme 1931 (Orthoptera, Tettigonoidea, Phaneropteridae). Behav. Ecol. Sociobiol. 28:391–396. Waage, J. K. 1979. Dual Function of the Damselfly Penis : Sperm Removal and Transfer. Science. 203:916–918. Wagner, W. E. 1989. Fighting, assessment, and frequency alteration in Blanchard’s cricket frog. Behav. Ecol. Sociobiol. 25:429–436. Waitt, C., A. C. Little, S. Wolfensohn, P. Honess, A. P. Brown, H. M. Buchanan-Smith, and D. I. Perrett. 2003. Evidence from rhesus macaques suggests that male coloration plays a role in female primate mate choice. Proc. R. Soc. B Biol. Sci. 270:S144–S146. Walker, T. I. 2005. Reproduction in Fisheries Science. Pp. 81–127 in W. C. Hamlett, ed. Reproductive Biology and Phylogeny of Chondrichthyes: Sharks, Batoids and Chimaeras. Science Publishers. Wedell, N., M. J. G. Gage, and G. A. Parker. 2002. Sperm competition, male prudence and sperm-limited females. Trends Ecol. Evol. 17:313–320. Welke, K., and J. M. Schneider. 2009. Inbreeding avoidance through cryptic female choice in the cannibalistic orb-web spider Argiope lobata. Behav. Ecol. 20:1056–1062. Whitney, N. M., H. L. Pratt, and J. C. Carrier. 2004. Group courtship, mating behaviour and siphon sac function in the whitetip reef shark, Triaenodon obesus. Anim. Behav. 68:1435–1442. Wourms, J. P., and J. Lombardi. 1992. Reflections on the evolution of piscine viviparity. Integr. Comp. Biol. 32:276–293. Wright, J., and I. Cuthill. 1989. Manipulation of sex differences in parental care. Behav. Ecol. Sociobiol. 25:171–181. Yopak, K. E. 2012. Neuroecology of cartilaginous fishes: The functional implications of brain scaling. J. Fish. Biol. 80:1968-2023

69 Zuk, M., R. Thornhill, J. D. Ligon, K. Johnson, S. Austad, S. H. Ligon, N. W. Thornhill, and C. Costin. 1990. The Role of Male Ornaments and Courtship Behavior in Female Mate Choice of Red Jungle Fowl. Am. Nat. 136:459–473.

70 Chapter 2: Testes size increases with sperm competition risk and intensity in bony fish and sharks

Status: Manuscript, in revision at Behavioral Ecology

Authors: Amy Rowley, Toby S. Daly –Engel, and John L. Fitzpatrick

Author contributions:

Conceptualization: AR, JLF

Data collection: All authors

Data analysis: AR, JLF

Writing – original draft: AR, JLF

Writing – review and editing: All authors

Funding acquisition: JLF

71 2. Testes size increases with sperm competition risk and intensity in bony fish and sharks

Amy Rowley1,2, Toby S. Daly –Engel3, and John L. Fitzpatrick2

Affiliation 1 Faculty of Biology, Medicine and Health, University of Manchester, UK 2 Department of Zoology, Stockholm University, Stockholm, Sweden 3 Department of Biological Sciences, Florida Institute of Technology, Melbourne, FL, USA

72 2.1 Abstract

Female multiple mating provides the opportunity for sexual selection to continue after gamete release, generating strong selection on male reproductive traits. In particular, in species where female multiple mating is common, males are expected to invest more in testicular tissue to afford them a numerical advantage during sperm competition. However, although testes size (correcting for body size) is a commonly used proxy of the strength of sperm competition, there is surprisingly scant direct evidence linking male investment in testes with genetic estimates of multiple paternity across species. Here we test the hypothesis that testes size is influenced by genetic estimates of sperm competition risk (multiple paternity percentage) and intensity (number of sires per brood) in fishes, the most diverse and specious vertebrate group. We provide conclusive evidence that testes size is larger in species experiencing a higher risk and intensity of sperm competition, a finding that remains consistent among sharks and bony fishes (including in separate analyses focused only on cichlids). These findings shed new light on evolutionary processes governing sperm competition in a basal vertebrate lineage and validate the now-widespread use of body-size corrected testes mass as a proxy for sperm competition risk and intensity.

73 2.2 Introduction

Female genetic monogamy appears to be the exception rather than the rule in animals (Birkhead & Møller, 1998; Birkhead et al., 2009; Taylor et al., 2014). Females frequently mate with multiple males during a single reproductive period (Birkhead & Møller, 1998), leading to sperm competition, an extended form of male-male competition between sperm from rival males to fertilise ova (Parker, 1970). Sperm competition generates strong selection on males to produce ejaculates that are better at fertilising ova than those of their rivals (Simmons & Fitzpatrick, 2012; Rowley & Fitzpatrick, 2016). Critically, sperm number is a crucial factor influencing the outcome of sperm competition, as males that ejaculate a greater number of sperm enjoy a competitive advantage in fertilisations (Martin et al., 1974; Parker, 1982; Stolz & Neff, 2006; Boschetto et al., 2011). At a functional level, males with larger testes are capable of producing more sperm (Marconato & Shapiro, 1996; Schärer et al., 2004; Lüpold et al., 2009; Ramm & Stockley, 2010; Rowe & Pruett-Jones, 2011; Ramm & Schärer, 2014). But due to the costs associated with sperm production (Wedell et al., 2002), male investment in testicular tissue should be tailored to the level of sperm competition their ejaculates are likely to experience. Therefore, a long-standing prediction from sperm competition theory is that selection will act to increase male investment in testes size as the probability that females will mate multiply (i.e. sperm competition risk) and the number of partners a female has (i.e. sperm competition intensity) increases (Parker & Ball, 2005; Parker & Pizzari, 2010).

Indeed, increases in testes size (relative to body size) represents a well- characterized response to sperm competition risk both within and across species (Simmons & Fitzpatrick, 2012). Within species, comparisons between males with alternative reproductive tactics (Montgomerie & Fitzpatrick, 2009), among males from populations where the level of sperm competition differs (Parker et al., 1997; Firman & Simmons, 2008), and between populations experimentally exposed to different levels of sperm competition (Hosken & Ward, 2001; Hosken et al., 2001; Simmons & García-González, 2008) all demonstrate a positive association between

74 sperm competition and male investment in testicular tissue. Similarly, comparative studies routinely demonstrate increased testicular investment in response to sperm competition risk (reviewed by Simmons & Fitzpatrick, 2012). Increases in testes size in response to sperm competition are thought to be so common that the use of body-size corrected testes mass is used extensively as a proxy measure for sperm competition risk across species (Balshine et al., 2001; Gage & Freckleton, 2003; Calhim et al., 2007; Immler & Birkhead, 2007; Gómez Montoto et al., 2011; Simmons & Fitzpatrick, 2012). However, direct evidence that testicular investment reflects inter-specific variation in sperm competition risk remains limited. Most comparative studies examining how sperm competition influences male investment in testes rely on comparisons between species with different social mating systems, or use behavioural metrics as an indicator of sperm competition risk, two approaches that may not accurately reflect competitive dynamics at the genetic level (Birkhead & Møller, 1998; DeWoody & Avise, 2001; Griffith et al., 2002). Surprisingly, direct evidence of positive correlations between body-size corrected testes size and the frequency of genetic polyandry has been demonstrated in less than a handful of studies, all of which focused on either mammals (Ramm et al., 2005; Soulsbury, 2010) or birds (Møller & Briskie, 1995).

Here, we evaluate the relationship between testicular investment and multiple paternity in fishes, the largest and most diverse vertebrate group, which correspondingly display the most complex and varied range of mating strategies and behaviours of any vertebrate group (Desjardins & Fernald, 2009). Fish reproduction encompasses an astonishing variety of behaviours from strict monogamy to extreme promiscuity (DeWoody & Avise, 2001), and multiple paternity frequencies vary considerably across species (Coleman & Jones, 2011; Fitzpatrick et al., 2012). Therefore, fishes represent an ideal model for investigating the relationship between multiple paternity rates and male investment in testes size. Moreover, as is the case with other taxa (e.g. Griffith et al., 2002), there are discrepancies between behavioural and genetic mating systems in fishes (e.g. Sefc et al., 2008; Chapman et al., 2013). Yet, the few available comparative studies of fishes to date have examined how investment in testicular tissue is influenced by

75 behavioural estimates of mating systems (Stockley et al., 1997; Pyron, 2000; Balshine et al., 2001; Molloy et al., 2007; Erisman et al., 2009; Fitzpatrick et al., 2009). Here we take advantage of the wealth of information now available on the genetic mating systems of fish species to consider the effect of female promiscuity on patterns of investment in testes across both major fish lineages, elasmobranchs (sharks and rays) and bony fishes.

2.3 Methods

2.3.1 Data collection

Estimates of multiple paternity were obtained from the literature from 152 studies, representing 107 fish species, including bony fishes (n=77) and elasmobranchs (sharks and rays n=30). We define the multiple paternity percentage as the proportion of the total number of broods/nests/litters (henceforth collectively referred to as ‘broods’) tested for parentage where two or more males sired progeny, which served as a proxy of sperm competition risk. Where available, we also recorded the mean number of sires per brood for each species, which was used as a proxy of sperm competition intensity. We only considered data derived from wild populations, where multiple paternity was assessed using polymorphic microsatellite loci, and where at least one parent was genotyped at the candidate loci. When multiple paternity estimates were available from multiple populations or from a number of studies for a given species, we combined the data from the populations/studies by calculating the mean percentage of multiple paternity weighted by brood number.

The methods used to estimate multiple paternity typically differ between egg laying (oviparous) and live bearing (ovoviviparous or viviparous) species based on variation in their reproductive biology. In many egg laying (oviparous) species, at least one parent typically remained associated with fertilized eggs. In these oviparous species, one parental genotype can be determined from the associated

76 individual and the presence of offspring showing an alternative genotype indicates the genetic contribution of non-associated individuals to the brood (as is the case in nest guarding species (e.g. Micropterus salmoides), mouth brooding cichlids (e.g. Tropheus moorii), and seahorses and pipefishes. In some bony fish species, males display fixed alternative reproductive tactics, with typically larger ‘guarding’ males monopolizing a female or guarding a nest, while smaller ‘satellite’ or ‘sneaker’ males try to fertilise eggs by stealth (Montgomerie & Fitzpatrick, 2009). Alternative reproductive tactics are common in fish (Montgomerie and Fitzpatrick 2009), and likewise are common in our dataset (e.g. Porichthys notatus, Telmatochromis temporalis, Pomatoschistus minutus). In species with alternative reproductive tactics, guarding males may occasionally take over the nest of another male, which already contains eggs that have been fertilized by the displaced male. In such nest takeover events, the presence of offspring unrelated to the guarding male would not indicate sperm competition. Therefore, in species with alternative reproductive tactics, we classified any litters where the dominant or guarding male was found to be unrelated to the entire brood as probable nest takeover events and excluded such cases when calculating the proportion of multiply-sired litters. For live bearing (ovoviviparous or viviparous) species, where eggs or embryos are retained in the female’s body cavity, the maternal genotype was subtracted from the offspring genotypes and the number of contributing sires inferred from the number of remaining alleles represented in the brood.

We then searched the literature for data on testes and body mass of sexually mature males for each of these 107 species, using their common and binomial names in combination with search terms including ‘testes’/’testis mass’, ‘body mass’ and ‘gonadosomatic index’. We also used testes and body mass data from our own field collections or obtained through personal communications with researchers in the field (Supplementary Table S2.1). As testes mass varies temporally in many species (Montgomerie & Fitzpatrick, 2009), we used maximum measurements taken during the breeding period of each species whenever possible. Where raw data were not reported, testes and body mass values were obtained from personal communication with the authors of the original studies or

77 extracted when possible using the program GraphClick v3.0.3 (Boyle et al., 2012). In some studies, gonadosomatic index (GSI, calculated as testes mass/body mass x 100) was reported in lieu of testes mass. Despite criticisms of its use as a metric of investment in testicular tissue (Tomkins & Simmons, 2002), GSI continues to be widely used by researchers in fish biology. When only GSI data were available, we calculated testes mass from GSI values using male body mass data. In the case of species that display morphologically distinct alternative reproductive tactics, we only included testes mass of the dominant or guarding males, as satellite or sneaker males experience a higher risk of sperm competition than guarding males and thus their inclusion is likely to yield an artificially high measure of investment in testes. Testes and body mass data was available for 40 of the 107 species with multiple paternity estimates. However, the number of broods sampled differs dramatically among the species where multiple paternity data were available (across all 107 species mean brood size = 43.05; range: 4 – 408). Because small broods can provide unreliable estimates of multiple paternities, we excluded species from which <10 broods were sampled. This exclusion criteria removed four species from our dataset (in these four species 5 broods were examined). Thus, the final sample size for analyses was 36 species (n = 24 bony fish, n = 12 sharks). When we included the four species where <10 broods were examined in our analyses the results did not qualitatively change (data not shown).

2.3.2 Phylogeny

Despite recent advances in phylogenies for both bony fishes (e.g. Rabosky et al., 2013), and elasmobranchs (e.g. Naylor et al., 2010), no detailed phylogeny that encompasses both lineages is currently available. We therefore built a phylogeny incorporating the 36 species for which both multiple paternity and testes mass data was available. We obtained sequences for three mitochondrial genes (NADH dehydrogenase subunit 2, 16S ribosomal RNA and cytochrome B) and one nuclear gene (recombination activating gene 1) from GenBank using Geneious v9.1.5

78 (Kearse et al., 2012). Sequences were aligned using the MUSCLE plugin for Mesquite v3.04 (Maddison & Maddison, 2015) and used to estimate phylogenetic relationships between species within a Bayesian framework. At this stage two species (Symphmodus ocellatus and Petrochromis fasciolatus) were removed as their sequences failed to align, reducing the total sample size to 34. The best fitting substitution model for each aligned gene sequence was determined using jModelTest v2.1.1 (Darriba et al., 2012) by comparing Akaike Information Criterion (AIC) values. The most appropriate substitution model was found to be GTR + I + G for all 4 genes. However, these models were later simplified to HKY to aid convergence. Phylogenies were estimated from the aligned sequences using the programs BEAUTi and BEAST (Bayesian evolutionary analysis by sampling trees) version 1.8.3 (Drummond et al., 2012) under an uncorrelated lognormal relaxed molecular clock, a Yule speciation process and unlinked speciation model for all genes. We constrained a number of well-resolved monophyletic clades (bony fishes, elasmobranch fishes, carcharhiniformes, carcharhindae, cichlids, poecilids) and set the ucld.mean (the mean of the prior distribution for mutation rate) to 0.66 with upper and lower values of 1 and 0, respectively, with a uniform distribution based on sensitivity testing that resulted in chain convergence. We ran a Markov Chain Monte Carlo (MCMC) simulation using a chain length of 50 million iterations with parameters logged every 1000 generations. Effective sample size (ESS) values were assessed in Tracer to determine convergence of the Bayesian chain, with a minimum ESS score of 200 indicating adequate mixing of the MCMC. We then used TreeAnnotator v2.1.2 (Rambaut & Drummond, 2014) to construct a maximum clade credibility tree using mean node heights and a burn-in of 10%. A consensus phylogeny was viewed in FigTree version 1.4.2 (Rambaut & Drummond, 2015) and was generally consistent with existing phylogenies of bony and cartilaginous fishes (Naylor et al., 2010; Rabosky et al., 2013, Figure 1).

2.3.4 Phylogenetic Analyses

We tested the relationship between the percentage of multiple paternity and testes

79 mass using phylogenetically controlled generalized (PGLS) linear models (Freckleton et al., 2002) to account for statistical non-independence as a result of shared ancestry. To assess phylogenetic dependence of the data, likelihood ratio tests were used to calculate the phylogenetic scaling parameter λ, a measure of phylogenetic signal ranging from 0 (low phylogenetic signal) to 1 (high phylogenetic signal) (Pagel, 1999; Freckleton et al., 2002). Importantly, estimates of λ become less robust when the sample size in PGLS models drops below 20 species (Freckleton et al., 2002), which occurs in some of the models presented below. To determine if our results were sensitive to uncertainty in λ estimates driven by low sample sizes, we examined the confidence intervals around λ for each model. We then repeated all analyses with λ estimates constrained to the upper and lower extremes (i.e. λ = 0 or λ = 1, respectively). Our results did not qualitatively change when the constrained λ estimates were within the confidence intervals calculated in likelihood ratio tests (data not shown). However, analyses with low sample sizes should be interpreted cautiously, particularly in taxon-specific analyses. To control for the allometric relationship between testes mass and body size, body mass was included as a covariate in all models that included testes mass. All data were log10 transformed prior to analyses (this included multiple paternity percentage data, where we used log10 transformation rather than an arcsine square root transformation, as the latter has been criticized as generating uninterpretable model outputs; Warton & Hui, 2011). Although our proxy for sperm competition intensity (i.e. the number of sires per brood) constitutes count data, we used Gaussian distribution in our models as sire number was never included as a response variable in any model. All analyses were performed in R version 2.15.1 (R Core Team, 2012).

We began by comparing broad differences in multiple paternity percentage and the number of sires per brood between bony fish and elasmobranchs (hereafter referred to as “sharks”, as no rays were present in our final dataset). To investigate evolutionary responses in testes to sperm competition risk/intensity, we first examined the relationship between multiple paternity percentage/number of sires per brood and body size-corrected testes mass across all fish species in the dataset.

80 We then subdivided the dataset into two major taxonomic groups, bony fish and sharks, to examine taxa-specific responses in testes size to sperm competition. Among bony fish, there was sufficient data among one family of bony fish, the Cichlidae (n=8), to evaluate the relationship between body size-corrected testes mass and sperm competition risk and intensity at a lower taxonomic level. We refrained from directly comparing the relationship between testes size and sperm competition risk/intensity between bony fish and sharks, as such an analysis confounds inherent differences in fertilization and reproductive mode between these taxonomic groups (i.e. most bony fish are external fertilizers while all sharks are internal fertilizers, and most bony fish are oviparous while all sharks in this study are ovoviviparous or viviparous). Mating systems and multiple paternity frequency can vary among populations in fishes (e.g. Daly-Engel et al., 2007; Portnoy et al., 2007). To account for possible population-specific effects, we performed a secondary set of analyses restricted to species for which testes measurements came from individuals from the same geographical location as those used to determine multiple paternity frequency. Finally, to examine the possibility that our results are confounded by inclusion of the seahorses and pipefishes (n=5) in our sample, which exhibit monogamous or sequentially polygynous mating systems due to the uncommon strategy of male pregnancy (Wilson et al., 2003; Stölting & Wilson, 2007), we also performed all analyses on a dataset that excluded seahorses and pipefishes.

2.4 Results

2.4.1 Variation in sperm competition risk and intensity among fishes

Sperm competition risk and intensity varied widely across fishes (Figure 2.1). Multiple paternity percentage among bony fishes ranged from 0% in the cichlids Eretmodus cyanostictus and Neolamprologus caudopunctatus and all seahorses and pipefishes, to 100% in the plainfin midshipman (Porichthys notatus) and Variabilichromis moorii. Among sharks, multiple paternity ranged from 13% in the

81 rig (Mustleus lenticulatus) to 92% in the small-spotted catshark (Scyliorhinus canicula) (Figure 2.1). The mean number of sires per brood among bony fishes ranged from 1 in Eretmodus cyanostictus, Neolamprologus caudopunctatus and all seahorses and pipefishes to 4.7 in Variabilichromis moorii. Among sharks, mean number of sires ranged from 1.1 in the rig to 2.3 in the small-spotted catshark (Scyliorhinus canicula) and sandbar shark (Carcharhinus plumbeus) (Figure 2.1).

Figure 2.1. Variation in sperm competition risk and intensity across fishes. Phylogenetic relationships between fish species in our dataset, with points representing the percentage of litters sired by multiple males for each species (risk), and the average number of sires per brood within each species (intensity).

Mean multiple paternity percentage did not differ between bony fishes (33.8 ± 7.9%, n = 22) and sharks (44.5 ± 7.2% (SE), n = 12; PGLS, λ = 0.68, slope = 0.75 ± 0.69 (SE), t=1.08, p=0.29, r = 0.33, lower CI = -0.05, upper CI = 3.98). When seahorses and pipefish were removed from the analysis, the mean multiple paternity percentage in bony fishes increased to 43.7 ± 8.9% (n=17), which again did not differ between bony fishes and sharks (PGLS, λ < 0.001, slope = 0.23 ± 0.20, t=1.16,

82 df=27, p=0.26, r = 0.22, lower CI = -0.83, upper CI = 3.13). The mean number of sires contributing offspring to a brood did not differ significantly between bony fish (1.59 ± 0.22, range: 1−3.3, n = 17) and sharks (1.64 ± 0.13, range: 1.2−2.3, n=12; λ < 0.001, slope = 0.06 ± 0.07, t=0.81, df=25, p=0.43, r = 0.17, lower CI = -1.13, upper CI = 2.82), and this result was unchanged when seahorses and pipefishes were removed from the analysis (λ < 0.001, slope = -0.01 ± 0.08, t=-0.11, df=20, p=0.91, r = -0.02, lower CI = -2.07, upper CI = 1.85). Multiple paternity percentage was tightly correlated with mean number of sires per brood (λ = 0.60, slope = 2.57 ± 0.46, df=25, t=5.61, p<0.001, r = 0.75, lower CI = 3.08, upper CI = 8.07).

2.4.2 Sperm competition risk, intensity and testicular investment

We found a significant positive relationship between body size corrected testes mass and both multiple paternity percentage, our proxy measure of sperm competition risk, and the mean number of sires per brood, our proxy measure of sperm competition intensity, across all fishes in our dataset (Table 2.1, Figure 2.2). The positive association between body size corrected testes size and sperm competition risk or intensity remained significant when seahorses and pipefish were excluded from the analyses (sperm competition risk: λ = 0.28, df = 26, multiple paternity percentage: slope = 0.48 ± 0.1, t =4.48, p<0.001, r = 0.66, lower CI = 2.14, upper CI = 6.75; body mass: slope = 0.90 ± 0.04, t=20.29, p<0.001, r = 0.97, lower CI = 14.44, upper CI = 26.08; sperm competition intensity: λ = 0.82, df = 19, no. of sires: slope = 1.53 ± 0.33, t=4.57, p<0.001, r = 0.72, lower CI = 2.10, upper CI = 6.96; body mass: slope = 0.85 ± 0.07, t=12.65, p<0.001, r = 0.95, lower CI = 8.17, upper CI = 17.06). Likewise, the results remained qualitatively unchanged when only geographically matched data were included in the analyses (sperm competition risk: λ = 0.38, df = 23, multiple paternity percentage: slope = 0.48 ± 0.11, t=4.47, p<0.001, r = 0.68, lower CI = 2.09, upper CI = 6.78; body mass: slope = 0.88 ± 0.05, t=17.19, p<0.001, r = 0.96, lower CI = 11.85; sperm competition intensity: λ = 0.81, df = 18, number of sires: slope = 1.60 ± 0.34, t=4.69, p<0.001, r =0.74 , lower CI = 2.17, upper CI = 7.13; body mass: slope = 0.83 ± 0.07 t=11.12, p<0.001, r = 0.93, lower CI = 6.98, upper CI = 15.18).

Table 2.1. Phylogenetically controlled generalised least squares (PGLS) regressions between testes size and sperm competition risk/intensity. Sperm competition risk (a) assesses the relationship between testes mass and the percentage of litters sired by multiple males, while sperm competition intensity (b) assesses the relationship between testes mass and the mean number of sires per brood. Body mass was included as a covariate in all models to control for the allometric relationship between body and testes size. The phylogenetic scaling parameter λ indicates the level of phylogenetic dependence of the data, ranging from 0 (low phylogenetic signal) to 1 (high phylogenetic signal). The t-statistic (t), degrees of freedom (df) and slope of the regression are presented for each model, with significant results (p<0.05) highlighted in bold.

Predictors λ df Slope SE t p r Lower CI Upper CI

(a) Sperm competition risk Multiple paternity percentage 0.24 31 0.49 0.08 6.10 <0.001 0.74 3.59 8.54 Body mass 0.90 0.04 23.53 <0.001 0.97 17.36 29.65

(b) Sperm competition intensity Number of sires per brood 0.81 24 1.61 0.31 5.16 <0.001 0.73 2.69 7.56 Body mass 0.87 0.06 14.58 <0.001 0.95 10.01 19.09

84 The positive association between testicular investment and sperm competition risk and intensity remained consistent across a range of taxonomic levels of analysis. When we examined bony fish and sharks separately, body size corrected testes size was significantly related to multiple paternity percentage in both taxonomic groups (bony fish: λ < 0.001, df = 19, multiple paternity percentage: slope = 0.45 ± 0.08, t=5.79, p<0.001 r = 0.80, lower CI = 3.07, upper CI = 8.43; body mass: slope = 0.90 ± 0.06, t=15.64, p<0.001, r = 0.96, lower CI = 10.30, upper CI = 20.92; sharks: λ < 0.001, df = 9, multiple paternity percentage: slope = 0.58 ± 0.21, t=2.72, p=0.02, r = 0.67, lower CI = 0.35, upper CI = 4.99; body mass: slope = 0.65 ± 0.06 t=10.68, p<0.001, r = 0.96, lower CI = 5.42, upper CI = 15.87). Similarly, both bony fish and sharks exhibited significantly positive associations between body size corrected testes mass and the mean number of sires per brood (bony fish: λ < 0.001, df = 13, number of sires: slope = 1.83 ± 0.39, t=4.76, p<0.001, r = 0.80, lower CI = 2.04, upper CI = 7.38; body mass: slope = 0.88 ± 0.07, t=12.13, p<0.001, r = 0.96, lower CI = 7.09, upper CI = 17.10; sharks: λ < 0.001, df = 8, number of sires: slope = 1.44 ± 0.42, t=3.40, p<0.01, r = 0.77, lower CI = 0.79, upper CI = 5.89; body mass: slope = 0.61 ± 0.06, t=10.62, p<0.001, r = 0.97, upper CI = 5.13, lower CI = 16.05). Moreover, consistent with the results from our wider analyses, male cichlids also invested more in testes mass (correcting for body size) as sperm competition risk and intensity increased, although it should be noted that only eight species were included in this analysis (sperm competition risk: λ < 0.001, df = 5, multiple paternity percentage: slope = 0.42 ± 0.15, t=2.77, p=0.04, r = 0.78, lower CI = 0.12, upper CI = 5.27; body mass: slope = 1.29 ± 0.46, t=2.83, p=0.04, r = 0.78, lower CI = 0.16, upper CI = 5.35; sperm competition intensity: λ < 0.001, df = 4, number of sires: slope = 1.32 ± 0.48, t=2.76, p=0.05, r = 0.81, lower CI = -0.01, upper CI = 5.37; body mass: slope = 1.63 ± 0.44, t=3.70, p=0.02, r = 0.88, lower CI = 0.48, upper CI = 6.78).

85 (a)

(b)

Figure 2.2. Relationships between testes mass relative to body size and (a) the percentage of litters sired by more than one male (sperm competition risk) (y = 1.24 + 1.15*log10(x)) and (b) the mean number of sires per brood (sperm competition intensity) across species (y = 0.18 + 0.25*log10(x)). Data presented is this figure are not controlled for phylogeny.

2.5 Discussion

Testes mass (corrected for body size) is a commonly used metric of the strength of post-copulatory sexual selection (Møller, 1991; Briskie & Montgomerie, 1992; Dunn et al., 2001; Calhim & Birkhead, 2007). Yet surprisingly few comparative studies have linked testicular investment with genetic estimates of female promiscuity (Møller & Briskie, 1995; Ramm et al., 2005; Soulsbury, 2010). Moreover, despite the widespread use of fish as a model for studying sexual selection (Coleman et al., 2009; Montgomerie & Fitzpatrick, 2009; Evans et al., 2010), how female promiscuity shapes male investment in testicular tissue remained previously unclear in fishes. In this study, we found that males’ investment in testes increased commensurately with sperm competition risk (i.e. multiple paternity percentage) and intensity (i.e. the number of sires per brood) in fishes. This primary finding of our study is consistent with the theoretical prediction that an increase in either risk or intensity of sperm competition should select for greater investment in sperm production (Parker & Ball, 2005). Our results therefore demonstrate a robust response in testes size to sperm competition in fishes and validate the use of relative testes mass as a proxy for sperm competition risk and intensity.

Sperm competition risk is highly variable across fishes, from strict genetic monogamy in some bony fish species (though no sharks), to universal polyandry in others (see Figure 2.1). While this variation is evident across the various taxonomic scales we tested, the positive relationship we observed between testicular investment and sperm competition remained remarkably consistent between all fishes in our dataset and when we examined fishes at increasingly reduced taxonomic scales (i.e. sharks, bony fish and cichlids). In addition, we also detected a positive relationship between testicular investment and sperm competition when we included or excluded seahorses and pipefish from our analyses and when we constrained our analyses to only consider species where testicular and genetic data were collected from the same geographical location. Thus, overall, our results suggest that female promiscuity imposes selection on sperm number, driving males

87 to increase sperm-production capacity by investing in relatively larger testes in fishes.

The similar evolutionary responses in testicular investment in response to sperm competition risk and intensity between sharks and bony fish was surprising at first glance, particularly given the dramatic differences in reproductive biology between these taxonomic groups. For example, all sharks in our dataset are internal fertilizers, while most of the bony fish examined were external fertilizers. Recent theoretical arguments suggest that evolutionary shifts from external to internal fertilization are hypothesized to relax selection on testes size, as sperm dilution effects are less extreme in internal fertilizers (Parker, 2014). However, the shark species examined in our dataset were significantly larger than the bony fish we examined (PGLS: λ = 0.00, df = 32, taxonomic class: slope = 3.41 ± 0.38, t=9.08, p<0.001), and large female body size can generate strong dilution effects, even in the absence of external fertilization (Immler et al., 2011; Lüpold & Fitzpatrick, 2015). However, our dataset did not allow us to adequately disentangle the role of fertilization in shaping the strength of postcopulatory sexual selection, as the difference between sharks and bony fish was largely synonymous with differences in fertilization mode. Moreover, our sample of 34 fish species represents a small fraction (~0.1%) of this large and diverse taxonomic group. Therefore, a crucial next step is to clarify the role of fertilization mode in shaping testicular investment across a wider range of fish species.

Our findings demonstrate that the extent of female polyandry strongly determines the strength of sexual selection on males, shaping the evolution of male reproductive physiology in fishes. These results are consistent with patterns found in other taxonomic groups (Møller & Briskie, 1995; Ramm et al., 2005; Soulsbury, 2010) and validate the use of testes mass corrected for body size as a metric of sperm competition in fishes. Our proxy measures of sperm competition risk and intensity are conservative estimates of the degree of female polyandry, as these proxies are unable to incorporate the potential effects of post-mating processes such as cryptic female choice or intrauterine cannibalism (Sefc et al., 2008a;

88 Chapman et al., 2013) that may eliminate certain males either before fertilisation occurs or before the brood is analysed for parentage. Moreover, future studies should also consider the potential for interactions between male investment in sperm producing tissue, male investment in other sexually selected traits (e.g. sperm and gential morphology) and the size and number of eggs available to fertilize during a reproductive event (sensu Stockley et al., 1996). Nevertheless, the broad patterns we observed between testicular investment and sperm competition risk and intensity at various levels of taxonomic scales points to a strong evolutionary response among fishes.

2.6 Acknowledgments

We thank Geremy Cliff, Sheldon Dudley, Edward Farrell, Wally Bubley and the National Oceanographic and Atmospheric Association (NOAA) for contributing unpublished data and Wouter van der Bijl for help producing figures. This work was supported by the Biotechnology and Biological Sciences Research Council (BBSRC); Knut and Alice Wallenberg Academy Fellowship to John L. Fitzpatrick; Stockholm University and the University of Manchester.

2.7 References

Balshine, S., Leach, B.J., Neat, F., Werner, N.Y. & Montgomerie, R. 2001. Sperm size of African cichlids in relation to sperm competition. Behav. Ecol. 12: 726–731. Birkhead, T., Hosken, D.J. & Pitnick, S. 2009. Sperm biology : an evolutionary perspective. Elsevier/Academic Press. Birkhead, T.R. & Møller, A.P. 1998. Sperm Competition and Sexual Selection, 1st ed. Academic Press, San Diego, California. Boschetto, C., Gasparini, C. & Pilastro, A. 2011. Sperm number and velocity affect sperm competition success in the guppy (Poecilia reticulata). Behav. Ecol. Sociobiol. 65: 813–821. Boyle, M.A., Samaha, A.L., Rodewald, A.M. & Hoffmann, A.N. 2012. Evaluation of

89 the reliability and validity of GraphClick as a data extraction program. Comput. Human. Behav. 29: 1023-1027. Briskie, J. V. & Montgomerie, R. 1992. Sperm size and sperm competition in birds. Proc. R. Soc. London B 247: 89–95. Calhim, S. & Birkhead, T.R. 2007. Testes size in birds: quality versus quantity— assumptions, errors, and estimates. Behav. Ecol. 18: 271–275. Calhim, S., Immler, S. & Birkhead, T.R. 2007. Postcopulatory Sexual Selection Is Associated with Reduced Variation in Sperm Morphology. PLoS One 2: e413. Chapman, D.D., Wintner, S.P., Abercrombie, D.L., Ashe, J., Bernard, A.M., Shivji, M.S., et al. 2013. The behavioural and genetic mating system of the sand tiger shark, Carcharias taurus, an intrauterine cannibal. Biol. Lett. 9: 20130003. Coleman, S.W., Harlin-Cognato, A. & Jones, A.G. 2009. Reproductive isolation, reproductive mode, and sexual selection: empirical tests of the viviparity- driven conflict hypothesis. Am. Nat. 173: 291–303. Coleman, S.W. & Jones, A.G. 2011. Patterns of multiple paternity and maternity in fishes. Biol. J. Linn. Soc. 103: 735–760. Daly-Engel, T.S., Grubbs, R.D., Bowen, B.W. & Toonen, R.J. 2007. Frequency of multiple paternity in an unexploited tropical population of sandbar sharks (Carcharhinus plumbeus). Can. J. Fish. Aquat. Sci. 64: 198–204. Darriba, D., Taboada, G.L., Doallo, R. & Posada, D. 2012. jModelTest 2: more models, new heuristics and parallel computing. Nat. Methods 9: 772–772. Desjardins, J.K. & Fernald, R.D. 2009. Fish sex: why so diverse? Curr. Opin. Neurobiol. 19: 648–53. DeWoody, J.A. & Avise, J.C. 2001. Genetic Perspectives on the Natural History of Fish Mating Systems. J. Hered. 92: 167–172. Drummond, A.J., Suchard, M.A., Xie, D. & Rambaut, A. 2012. Bayesian Phylogenetics with BEAUti and the BEAST 1.7. Mol. Biol. Evol. 29: 1969–1973. Dunn, P.O., Whittingham, L.A. & Pitcher, T.E. 2001. Mating systems, sperm competition, and the evolution of sexual dimorphism in birds. Evolution 55: 161–175. Erisman, B.E., Craig, M.T. & Hastings, P.A. 2009. A phylogenetic test of the size- advantage model: evolutionary changes in mating behavior influence the loss

90 of sex change in a fish lineage. Am. Nat. 174: E83-99. Evans, J.P., Pilastro, A. & Schlupp, I. 2010. Ecology and Evolution of Poeciliid Fishes. The University of Chicago Press, Chicago. Firman, R.C. & Simmons, L.W. 2008. The frequency of multiple paternity predicts variation in testes size among island populations of house mice. J. Evol. Biol. 21: 1524–1533. Fitzpatrick, J.L., Kempster, R.M., Daly-Engel, T.S., Collin, S.P. & Evans, J.P. 2012. Assessing the potential for post-copulatory sexual selection in elasmobranchs. J. Fish Biol. 80: 1141–1158. Fitzpatrick, J.L., Montgomerie, R., Desjardins, J.K., Stiver, K. a, Kolm, N. & Balshine, S. 2009. Female promiscuity promotes the evolution of faster sperm in cichlid fishes. Proc. Natl. Acad. Sci. U. S. A. 106: 1128–32. Freckleton, R.. P., Harvey, P.H. & Pagel, M. 2002. Phylogenetic analysis and comparative data: A test and review of evidence. Am. Nat. 160: 712–726. Gage, M.J.G. & Freckleton, R.P. 2003. Relative testis size and sperm morphometry across mammals: no evidence for an association between sperm competition and sperm length. Proceedings. Biol. Sci. 270: 625–32. The Royal Society. Gómez Montoto, L., Varea Sánchez, M., Tourmente, M., Martín-Coello, J., Luque- Larena, J.J., Gomendio, M., et al. 2011. Sperm competition differentially affects swimming velocity and size of spermatozoa from closely related muroid rodents: head first. Reproduction 142: 819–30. Griffith, S.C., Owens, I.P.F. & Thuman, K.A. 2002. Extra pair paternity in birds: a review of interspecific variation and adaptive function. Mol. Ecol. 11: 2195– 2212. Hosken, D.J., Garner, T.W.J. & Ward, P.I. 2001. Sexual conflict selects for male and female reproductive characters. Curr. Biol. 11: 489–493 Hosken, D.J. & Ward, P.I. 2001. Experimental evidence for testis size evolution via sperm competition. Ecol. Lett. 4: 10–13. Immler, S. & Birkhead, T.R. 2007. Sperm competition and sperm midpiece size: no consistent pattern in passerine birds. Proc. R. Soc. London B 274: 561–568. Immler, S., Pitnick, S., Parker, G.A., Durrant, K.L., Lüpold, S., Calhim, S., et al. 2011. Resolving variation in the reproductive tradeoff between sperm size and

91 number. Proc. Natl. Acad. Sci. 108: 5325–5330. Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., et al. 2012. Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28: 1647–1649. Lüpold, S. & Fitzpatrick, J.L. 2015. Sperm number trumps sperm size in mammalian ejaculate evolution. Proc. R. Soc. B 282: 20152122. Lüpold, S., Linz, G.M., Rivers, J.W., Westneat, D.F. & Birkhead, T.R. 2009. Sperm competition selects beyond relative testes size in birds. Evolution. 63: 391– 402. Maddison, W.P. & Maddison, D.R. 2015. Mesquite: a modular system for evolutionary analysis. Marconato, A. & Shapiro, D.Y. 1996. Sperm allocation, sperm production and fertilization rates in the bucktooth parrotfish. Anim. Behav. 52: 971–980. Martin, P.A., Reimers, T.J., Lodge, J.R. & Dziuk, P.J. 1974. The effect of ratios and numbers of spermatozoa mixed from two males on proportions of offspring. J. Reprod. Fertil. 39: 251–8. Møller, A.P. 1991. Sperm Competition, Sperm Depletion, Paternal Care, and Relative Testis Size in Birds. Am. Nat. 137: 882–906. Møller, A.P. & Briskie, J. V. 1995. Extra-pair paternity, sperm competition and the evolution of testis size in birds. Behav. Ecol. Sociobiol. 36: 357–365. Molloy, P.P., Goodwin, N.B., Côté, I.M., Reynolds, J.D. & Gage, M.J.G. 2007. Sperm competition and sex change: a comparative analysis across fishes. Evolution. 61: 640–652. Montgomerie, R. & Fitzpatrick, J.L. 2009. Reproductive Biology and Phylogeny of Fishes, 1st ed. (B. G. M. Jamieson, ed). Science Publishers, Enfield, New Hampshire. Naylor, G.J.P., Caira, J.N., Jensen, K., Rosana, K.A.M., Straube, N. & Lakner, C. 2010. Elasmobranch Phylogeny: A Mitochondrial Estimate Based on 595 Species. In: Biology of Sharks and Their Relatives, pp. 31–56. Taylor & Francis Group, LLC. Pagel, M. 1999. Inferring the historical patterns of biological evolution. Nature 401: 877–884.

92 Parker, G.A. 1970. Sperm competition and it’s evolutionary consequences in the insects. Biol. Rev. 45: 525–567. Parker, G.A. 2014. The sexual cascade and the rise of pre-ejaculatory (Darwinian) sexual selection, sex roles, and sexual conflict. Cold Spring Harb. Perspect. Biol. 6: a017509. Parker, G.A. 1982. Why are there so many tiny sperm? Sperm competition and the maintenance of two sexes. J. Theor. Biol. 96: 281–294. Parker, G.A. & Ball, M.A. 2005. Sperm competition, mating rate and the evolution of testis and ejaculate sizes: a population model. Biol. Lett. 1: 235–8. Parker, G.A., Ball, M.A., Stockley, P. & Gage, M.J. 1997. Sperm competition games: a prospective analysis of risk assessment. Proceedings. Biol. Sci. 264: 1793–802. Parker, G. a & Pizzari, T. 2010. Sperm competition and ejaculate economics. Biol. Rev. Camb. Philos. Soc. 85: 897–934. Portnoy, D.S., Piercy, A.N., Musick, J. a, Burgess, G.H. & Graves, J.E. 2007. Genetic polyandry and sexual conflict in the sandbar shark, Carcharhinus plumbeus, in the western North Atlantic and Gulf of Mexico. Mol. Ecol. 16: 187–97. Pyron, M. 2000. Testes mass and reproductive mode of minnows. Behav. Ecol. Sociobiol. 48: 132–136. Rabosky, D.L., Santini, F., Eastman, J., Smith, S.A., Sidlauskas, B., Chang, J., et al. 2013. Rates of speciation and morphological evolution are correlated across the largest vertebrate radiation. Nat. Commun. 4: 1958. R Core Team. 2012. R: A language and environment for statistical computing. R Foundation for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Rambaut, A. & Drummond, A.J. 2015. FigTree, ver. 1.4. 2. Rambaut, A. & Drummond, A.J. 2014. TreeAnnotator v2. 1.2. University of Edinburgh, Institute of Evolutionary Biology. Ramm, S.A., Parker, G.A. & Stockley, P. 2005. Sperm competition and the evolution of male reproductive anatomy in rodents. Proc. R. Soc. London B 272: 949–55. Ramm, S.A. & Schärer, L. 2014. The evolutionary ecology of testicular function: size isn’t everything. Biol. Rev. 89: 874–888. Ramm, S.A. & Stockley, P. 2010. Sperm competition and sperm length influence the

93 rate of mammalian spermatogenesis. Biol. Lett. 6: 219–21. Rowe, M. & Pruett-Jones, S. 2011. Sperm Competition Selects for Sperm Quantity and Quality in the Australian Maluridae. PLoS One 6: e15720. Rowley, A.G. & Fitzpatrick, J.L. 2016. Sperm Competition. In: Encyclopaedia of Evolutionary Biology (R. Kilman, ed), pp. 245–248. Academic Press. Schärer, L., Ladurner, P. & Rieger, R. 2004. Bigger testes do work more: experimental evidence that testis size reflects testicular cell proliferation activity in the marine invertebrate, the free-living flatworm Macrostomum sp. Behav. Ecol. Sociobiol. 56: 420–425. Sefc, K.M., Mattersdorfer, K., Sturmbauer, C. & Koblmüller, S. 2008. High frequency of multiple paternity in broods of a socially monogamous cichlid fish with biparental nest defence. Mol. Ecol. 17: 2531–2543. Simmons, L.W. & Fitzpatrick, J.L. 2012. Sperm wars and the evolution of male fertility. Reproduction 144: 519–34. Simmons, L.W. & García-González, F. 2008. Evolutionary reduction in testes size and competitive fertilization success in response to the experimental removal of sexual selection in dung beetles. Evolution. 62: 2580–2591. Soulsbury, C.D. 2010. Genetic Patterns of Paternity and Testes Size in Mammals. PLoS One 5: e9581. Stockley, P., Gage, M.J., Parker, G.A. & Møller, A.P. 1997. Sperm competition in fishes: the evolution of testis size and ejaculate characteristics. Am. Nat. 149: 933–54. Stockley, P., Gage, M.J.G., Parker, G.A. & Moller, A.P. 1996. Female Reproductive Biology and the Coevolution of Ejaculate Characteristics in Fish. Proc. R. Soc. B Biol. Sci. 263: 451–458. Stölting, K.N. & Wilson, A.B. 2007. Male pregnancy in seahorses and pipefish: beyond the mammalian model. BioEssays 29: 884–896. Stolz, J.A. & Neff, B.D. 2006. Sperm competition in a fish with external fertilization: the contribution of sperm number, speed and length. J. Evol. Biol. 19: 1873– 1881. Taylor, M.L., Price, T.A.R. & Wedell, N. 2014. Polyandry in nature: A global analysis. Trends Ecol. Evol. 29: 376-383.

94 Tomkins, J.L. & Simmons, L.W. 2002. Measuring relative investment: a case study of testes investment in species with alternative male reproductive tactics. Anim. Behav. 63: 1009–1016. Warton, D.I. & Hui, F.K.C. 2011. The arcsine is asinine: the analysis of proportions in ecology. Ecology 92: 3–10. Wedell, N., Gage, M.J.G. & Parker, G.A. 2002. Sperm competition, male prudence and sperm-limited females. Trends Ecol. Evol. 17: 313–320. Wilson, A.B., Ahnesjö, I., Vincent, A.C.J. & Meyer, A. 2003. The Dynamics of Male Brooding, Mating Patterns, and Sex Roles in Pipefishes and Seahorses (Family Syngnathidae). Evolution. 57: 1374.

95 Chapter 2: Supporting Information

Supporting Table S2.1

Multiple paternity, body size and testes data collected for all n=34 fish species used in our analyses. For each species we present the number of broods examined for paternity and the number of loci used to determine parentage (in parentheses), the program used in the parentage analysis, the percentage of litters sired by multiple males and the mean number of sires per brood. Where multiple values are presented for multiple paternity percentage, we show the weighted mean percentage with values from individual studies in parentheses. We also present total body length, body mass and testes mass.

96 Species Multiple Paternity Data Somatic and Testes Data c No. broods % Multiple Paternity b Number of sires Body Body Testes Reference (loci) a CERVUS GERUD Other Method CERVUS GERUD Other Length Mass Mass Method (cm) (g) (g) d (a) bony fish Eretmodus 14 (3) 0 - - 1 - - 6.47 9.09 0.008 Taylor et al., cyanostictus e 2003; Fitzpatrick et al., 2009 Gambusia holbrooki 50 (3) - - 90 f - - 2.2 2.27 0.201 0.004 g Zane et al., 1999; Orlando et al., 2007 Gasterosteus 58h - - 20.24I (21, 20) - - - - 1.79 0.0056 j Andersson et aculeatus (3-5) al., 1988; Largiadèr et al., 2001; Blais et al., 2004 Gobiusculus 21 (4) - 5 - - - - 3.93 0.46 0.005k,l Mobley et al., flavescens e 2009; Utne- Palm et al., 2015

Heterandria 36h - - 46.08 m (15, 54, - - 1.5 - 0.0094 0.0001 Soucy & Travis, formosa e (3) 66) 8n 4n 2003; Schrader et al., 2012 Hippocampus 29h - - 0 - - 1 - 9.65 p 0.0183 Jones et al., subelongatuse,o (4-3) q 1998; Kvarnemo et al., 2000; Kvarnemo & Simmons, 2004 Julidochromis 63 (4) - - 23 r - - - 5.59 4.02 0.03 Awata et al., ornatus e 2005; Fitzpatrick et al., 2009 Lepomis 98h - - 94.87 i (92, 96.7) - - - - 140 s 1.82 s Neff, 2001; Neff macrochiruse (3-11) et al., 2003; Neff & Clare, 2008 Micropterus 26 (5) - - 4 - - 1.04 32.69 584.36 1.88t DeWoody et al., salmoides 2000; Lorenzoni et al., 2002 Neolamprologus 32 0 - - 1 - - 5.09 3.08 0.004 Fitzpatrick et al., caudopunctatus e (11) 2009; Schaedelin et al., 2015

98 Neolamprologus 12 (5) 66.7 66.7 - 1.4 - - 8.19 12.7 s 0.11 s Hellmann et al., modestus e 2015b(Fitzpatric k et al., 2009; Schaedelin et al., 2015) Neolamprologus 71h 50.80 i (33.3, - 1.93 - - 5.92 5.3 u 0.02 u Fitzpatrick et al., pulcher e (3-12) 61.3, 41.7, 60) 2006; Dierkes et al., 2008; Stiver et al., 2009; Bruintjes et al., 2011; Hellmann et al., 2015a Nerophis ophidion 11 (3) - - 0 - - 1 - 0.36 v 0.0014 Mccoy et al., 2 v 2001; Kvarnemo & Simmons, 2004 Poecilia reticulata 408h - - 63.79i (94, 24.6, - - 3.31 1.74 0.082 0.0035 Kelly et al., (2-9) 62.4, 95, 97.5, k 1999; Hain & 93.4) Neff, 2007; Skinner & Watt, 2007; Neff et al., 2008; Elgee et al., 2012

99 Pomatoschistus 41h - - 43.5i (35, 52) - - - - 0.24 w 0.0011 Jones et al., minutus e (3) 8 w 2001a; b; Kvarnemo et al., 2010 Porichthys notatus 47 (6) - - 100 - - - 20.44 121.91 1.50 s Cogliati et al., s 2013; Fitzpatrick et al., 2016 Syngnathus floridae 141h 0 - 0 - - 1 - 0.7 u 0.0010 Jones & Avise, (4) 2 u 1997b; Kvarnemo & Simmons, 2004; Mobley & Jones, 2007, 2009 Syngnathus scovelli 41h - - 0 - - 1 - 0.61 u 0.0007 Jones & Avise, (4) 6 u 1997a; Jones et al., 2001; Kvarnemo & Simmons, 2004 Syngnathus typhle 58 h 0 - 0 - - 1 - 2.68 u 0.0033 Jones et al., (3-4) 4 u 1999; Kvarnemo & Simmons,

100 2004; Rispoli & Wilson, 2008 Telmatochromis 21 (2) - - 29 - - 1.29 5.94 4.7 0.72 Katoh et al., temporalis e 2005; Fitzpatrick et al., 2009 Tropheus moorii e 19 (4) - 5 x - - 1.05 - 7.46 15.43 0.04 Egger et al., 2006; Fitzpatrick et al., 2009 Variabilichromis 10 (6) 100 100 100 4.7 - - 72.4 y 11.6 y 0.16 Sefc et al., 2008; moorii e Ota et al., 2012 (b) elasmobranchs Carcharias taurus e 15 - 40 - - - - - 92000 z 416 z Chapman et al., (10) 2013; Geremy Cliff & Sheldon Dudley, personal communication

Carcharhinus 14 (5) - 35 - 1.4 3.2 - - 200000 250 Rossouw et al., obscurus e 2016; Geremy Cliff & Sheldon Dudley,

101 personal communication

Carcharhinus 20 (5) - - 85 - - 2.3 162. 49935. 322.1aa Cliff et al., 1988; plumbeus 7 83 k Portnoy et al., 2007; Baremore & Hale, 2012 Mustelus 29 (8) - 24 31 (COLONY) - 1.32 - 86 1900.0 14.8 Fitzpatrick et al., antarcticus e 2012; Boomer et al., 2013 Mustelus asterias e 12 (5) - 58 83 (COLONY) - 1.58 - 87.7 2418.7 25.488 Farrell et al., 4jk 75 5 2014; Edward Farrell, personal communication

Mustelus 19 (8) - 13bb 42.0 (COLONY) - 1.13 - 98.4 3644.1 31.0 cc Massey & lenticulatus e ,dd Francis, 1989; Boomer et al., 2013 Mustelus mustelus e 19 (9) - 47 - - 1.6 - 116. 5275.4 43.41j Saïdi et al., 23k 5cc 2008; Marino et al., 2015

102 Mustelus 13 (9) - 54.00 - - 2 - 88.8 2122.9 23.80 ee Saïdi et al., punctulatus e 3j 2cc 2009; Marino et al., 2015 Scyliorhinus 13 - 92 GERUD/COLONY - 2.31 - 42.3 253.5 11.98 Griffiths et al., canicula (12) 2012 Sphyrna lewini e 13 (5) - 46 GERUD/COLONY 2 - 97020 ff 340 gg de Bruyn et al., 2005; Rossouw et al., 2016

Sphyrna tiburo e 22 (4) - 18.8 - - 1.2 - 82.2 2270.0 8.1 Chapman et al., 2004; Fitzpatrick et al., 2012 Squalus acanthias e 39 (7) - 21 - - 1.17 - 81.5 2085 32.4 Lage et al., (17,30) 2008; Verissimo et al., 2011; Wally Bubley, personal communication

103

Notes: a. Loci indicates number or range of loci evaluated. b. Multiple paternity frequency was calculated as the percentage of broods/nests/litters where offspring were sired by multiple males. When a guarding male was found to be unrelated to the entire brood, the brood was removed from percentage calculations as a probable nest takeover event. Weighted mean values are provided in cases where multiple studies or populations were available for a single species. Values in brackets represent frequencies of multiple paternity found in each individual study or population from which mean weighted by brood number was calculated. c. In cases where species have alternative male reproductive tactics we only report somatic and testes data from parental/guarding males. d. Since testes exhibit seasonal variation in size in many species, when we were able to account for seasonal variation we only report peak testes values. e. Multiple paternity and testes samples were collected from the same geographical location. f. Rises from 86% to 90% when criteria of Mendelian segregation of paternal alleles is added g. Males from uncontaminated reference site h. Sum of brood numbers analysed in different studies i. Weighted mean of multiple paternity frequencies from multiple studies j. Calculated from GSI k. Values calculated from figure using Graphclick l. Calculated from samples collected in the early breeding season m. weighted mean of multiple paternity frequencies of different populations within the same study n. Mean values derived from 15 populations o. Formerly Hippocampus angustus p. Testes and body mass data found in appendix, Table 2, species listed as Hippocampus angustus q. Body length, mass and testes mass of 2 year old males

r. 23% unrelated or half-related siblings, 13% unrelated s. Body and testes mass data from parental/territorial males only t. Calculated from peak GSI value (May) u. Body and testes mass data from breeder males only v. Testes and body mass data found in appendix, Table 2 w. Testes and body mass from males caught on-nest with breeding colour (Table 1) x. The authors (Egger et al. 2006) argue that one brood of the 19 assayed could result either from a multiple mating event or a genotype mutation. We chose to conservatively treat this as a multiply sired litter to avoid bias. y. Body mass, length and testes mass taken from males collected at shallow depths z. Testes and body mass data from mature males during months characterised by peak GSI (June-August) aa. Peak testes mass (April) calculated from Figure 7 using Graphclick bb. Only 16 broods analysed using GERUD cc. Calculated from length value and length-weight equation given in paper dd. Calculated from peak GSI (December), body size based on graphclick values of age when 100% of males reach sexual maturity ee. Calculated from peak GSI value ff. Calculated from mean pre-caudal length (PCL) obtained from Figure 7 using Graphclick, using PCL-mass equation given in Figure 4a gg. Calcuated from peak GSI value for which more than one data point was available (June)

Supporting Information References

Andersson, E., Mayer, I. & Borg, B. 1988. Inhibitory effect of 11- ketoandrostenedione and androstenedione on spermatogenesis in the three-

105 spined stickleback, Gasterosteus aculeatus L. J. Fish Biol. 33: 835–840. Awata, S., Munehara, H. & Kohda, M. 2005. Social system and reproduction of helpers in a cooperatively breeding cichlid fish (Julidochromis ornatus) in Lake Tanganyika: field observations and parentage analyses. Behav. Ecol. Sociobiol. 58: 506–516. Baremore, I.E. & Hale, L.F. 2012. Reproduction of the Sandbar Shark in the Western North Atlantic Ocean and Gulf of Mexico. Mar. Coast. Fish. 4: 560–572. Blais, J., Rico, C. & Bernatchez, L. 2004. Nonlinear effects of female mate choice in wild threespine sticklebacks. Evolution. 58: 2498–510. Boomer, J.J., Harcourt, R.G., Francis, M.P., Walker, T.I., Braccini, J.M. & Stow, A.J. 2013. Frequency of multiple paternity in gummy shark, Mustelus antarcticus, and rig, Mustelus lenticulatus, and the implications of mate encounter rate, postcopulatory influences, and reproductive mode. J. Hered. 60: 371–9. Bruintjes, R., Bonfils, D., Heg, D. & Taborsky, M. 2011. Paternity of Subordinates Raises Cooperative Effort in Cichlids. PLoS One 6: e25673. Chapman, D.D., Prodöhl, P.A., Gelsleichter, J., Manire, C.A. & Shivji, M.S. 2004. Predominance of genetic monogamy by females in a hammerhead shark, Sphyrna tiburo: Implications for shark conservation. Mol. Ecol. 13: 1965–1974. Chapman, D.D., Wintner, S.P., Abercrombie, D.L., Ashe, J., Bernard, A.M., Shivji, M.S., et al. 2013. The behavioural and genetic mating system of the sand tiger shark, Carcharias taurus, an intrauterine cannibal. Biol. Lett. 9: 20130003. Cliff, G., Dudley, S.F.J. & Davis, B. 1988. Sharks caught in the protective gill nets off Natal, South Africa. 1. The sandbar shark Carcharhinus plumbeus (Nardo). South African J. Mar. Sci. 7: 255–265. Cogliati, K.M., Neff, B.D. & Balshine, S. 2013. High degree of paternity loss in a species with alternative reproductive tactics. Behav. Ecol. Sociobiol. 67: 399– 408. de Bruyn, P., Dudley, S., Cliff, G. & Smale, M. 2005. Sharks caught in the protective gill nets off KwaZulu-Natal, South Africa. 11. The scalloped hammerhead shark Sphyrna lewini (Griffith and Smith). African J. Mar. Sci. 27: 517–528. DeWoody, J.A., Fletcher, D.E., Wilkins, S.D., Nelson, W.S. & Avise, J.C. 2000. Genetic monogamy and biparental care in an externally fertilizing fish, the largemouth

106 bass (Micropterus salmoides). Proceedings. Biol. Sci. 267: 2431–7. The Royal Society. Dierkes, P., Taborsky, M. & Achmann, R. 2008. Multiple paternity in the cooperatively breeding fish Neolamprologus pulcher. Behav. Ecol. Sociobiol. 62: 1581–1589. Egger, B., Obermüller, B., Phiri, H., Sturmbauer, C. & Sefc, K.M. 2006. Monogamy in the maternally mouthbrooding Lake Tanganyika cichlid fish Tropheus moorii. Proceedings. Biol. Sci. 273: 1797–802. Elgee, K.E., Ramnarine, I.W. & Pitcher, T.E. 2012. Multiple paternity, reproductive skew and correlates of male reproductive success in a wild population of the Trinidadian guppy. Ecol. Freshw. Fish 21: 109–118. Farrell, E.D., O’Sullivan, N., Sacchi, C. & Mariani, S. 2014. Multiple paternity in the starry smooth-hound shark Mustelus asterias (Carcharhiniformes: Triakidae). Biol. J. Linn. Soc. 111: 119–125. Fitzpatrick, J.L., Desjardins, J.K., Stiver, K.A., Montgomerie, R. & Balshine, S. 2006. Male reproductive suppression in the cooperatively breeding fish Neolamprologus pulcher. Behav. Ecol. 17: 25–33. Fitzpatrick, J.L., Earn, D.J.D., Bucking, C., Craig, P.M., Nadella, S., Wood, C.M., et al. 2016. Postcopulatory consequences of female mate choice in a fish with alternative reproductive tactics. Behav. Ecol. 27: 312–320. Fitzpatrick, J.L., Kempster, R.M., Daly-Engel, T.S., Collin, S.P. & Evans, J.P. 2012. Assessing the potential for post-copulatory sexual selection in elasmobranchs. J. Fish Biol. 80: 1141–1158. Fitzpatrick, J.L., Montgomerie, R., Desjardins, J.K., Stiver, K. a, Kolm, N. & Balshine, S. 2009. Female promiscuity promotes the evolution of faster sperm in cichlid fishes. Proc. Natl. Acad. Sci. U. S. A. 106: 1128–32. Griffiths, A.M., Jacoby, D.M.P., Casane, D., McHugh, M., Croft, D.P., Genner, M.J., et al. 2012. First analysis of multiple paternity in an oviparous shark, the small- spotted catshark (Scyliorhinus canicula). J. Hered. 103: 166–173. Hain, T.J.A. & Neff, B.D. 2007. Multiple paternity and kin recognition mechanisms in a guppy population. Mol. Ecol. 16: 3938–3946. Hellmann, J.K., Ligocki, I.Y., O’Connor, C.M., Reddon, A.R., Garvy, K.A., Marsh-Rollo,

107 S.E., et al. 2015a. Reproductive sharing in relation to group and colony-level attributes in a cooperative breeding fish. Proceedings. Biol. Sci. 282: 20150954. Hellmann, J.K., O’Connor, C.M., Ligocki, I.Y., Farmer, T.M., Arnold, T.J., Reddon, A.R., et al. 2015b. Evidence for alternative male morphs in a Tanganyikan cichlid fish. J. Zool. 296: 116–123. Jones, A.G. & Avise, J.C. 1997a. Microsatellite analysis of maternity and the mating system in the Gulf pipefish Syngnathus scovelli, a species with male pregnancy and sex-role reversal. Mol. Ecol. 6: 203–213. Jones, A.G. & Avise, J.C. 1997b. Polygynandry in the dusky pipefish Syngnathus floridae revealed by microsatellite DNA markers. Evolution. 51: 1611–1622. Jones, A.G., Kvarnemo, C., Moore, G.I., Simmons, L.W. & Avise, J.C. 1998. Microsatellite evidence for monogamy and sex-biased recombination in the Western Australian seahorse Hippocampus angustus. Mol. Ecol. 7: 1497–1505. Jones, A.G., Rosenqvist, G., Berglund, A. & Avise, J.C. 1999. The genetic mating system of a sex-role-reversed pipefish ( Syngnathus typhle ): a molecular inquiry. Behav. Ecol. Sociobiol. 46: 357–365. Jones, A.G., Walker, D. & Avise, J.C. 2001. Genetic evidence for extreme polyandry and extraordinary sex-role reversal in a pipefish. Proceedings. Biol. Sci. 268: 2531–5. Jones, A.G., Walker, D., Kvarnemo, C., Lindström, K. & Avise, J.C. 2001a. How cuckoldry can decrease the opportunity for sexual selection: data and theory from a genetic parentage analysis of the sand goby, Pomatoschistus minutus. Proc. Natl. Acad. Sci. U. S. A. 98: 9151–6. Jones, A.G., Walker, D., Lindström, K., Kvarnemo, C. & Avise, J.C. 2001b. Surprising similarity of sneaking rates and genetic mating patterns in two populations of sand goby experiencing disparate sexual selection regimes. Mol. Ecol. 10: 461– 469. Katoh, R., Munehara, H. & Kohda, M. 2005. Alternative Male Mating Tactics of the Substrate Brooding Cichlid Telmatochromis temporalis in Lake Tanganyika. Zoolog. Sci. 22: 555–561. Zoological Society of Japan . Kelly, C.D., Godin, J.-G.J. & Wright, J.M. 1999. Geographic variation in multiple paternity within natural populations of the guppy (Poecilia reticulata). Proc. R.

108 Soc. B Biol. Sci. 266: 2403–2408. Kvarnemo, C., Moore, G.I., Jones, A.G., Nelson, W.S. & Avise, J.C. 2000. Monogamous pair bonds and mate switching in the Western Australian seahorse Hippocampus subelongatus. J. Evol. Biol. 13: 882–888. Kvarnemo, C. & Simmons, L.W. 2004. Testes investment and spawning mode in pipefishes and seahorses (Syngnathidae). Biol. J. Linn. Soc. 83: 369–376. Kvarnemo, C., Svensson, O. & Manson, W. 2010. Investment in testes, sperm-duct glands and lipid reserves differs between male morphs but not between early and late breeding season in Pomatoschistus minutus. J. Fish Biol. 76: 1609– 1625. Lage, C.R., Petersen, C.W., Forest, D., Barnes, D., Kornfield, I. & Wray, C. 2008. Evidence of multiple paternity in spiny dogfish ( Squalus acanthias ) broods based on microsatellite analysis. J. Fish Biol. 73: 2068–2074. Largiadèr, C.R., Fries, V. & Bakker, T.C.M. 2001. Genetic analysis of sneaking and egg-thievery in a natural population of the three-spined stickleback (Gasterosteus aculeatus). Heredity. 86: 459-468. Lorenzoni, M., Martin Dörr, A.J., Erra, R., Giovinazzo, G., Mearelli, M. & Selvi, S. 2002. Growth and reproduction of largemouth bass (Micropterus salmoides Lacépède, 1802) in Lake Trasimeno (Umbria, Italy). Fish. Res. 56: 89–95. Marino, I.A.M., Riginella, E., Gristina, M., Rasotto, M.B., Zane, L. & Mazzoldi, C. 2015. Multiple paternity and hybridization in two smooth-hound sharks. Sci. Rep. 5: 12919. Massey, B.R. & Francis, M.P. 1989. Commercial catch composition and reproductive biology of rig (Mustelus lenticulatus) from Pegasus bay, Canterbury, New Zealand. New Zeal. J. Mar. Freshw. Res. 23: 113–120. Mccoy, E.E., Jones, A.G. & Avise, J.C. 2001. The genetic mating system and tests for cuckoldry in a pipefish species in which males fertilize eggs and brood offspring externally. Mol. Ecol. 10: 1793–1800. Mobley, K.B., Amundsen, T., Forsgren, E., Svensson, P.A. & Jones, A.G. 2009. Multiple mating and a low incidence of cuckoldry for nest-holding males in the two-spotted goby, Gobiusculus flavescens. BMC Evol. Biol. 9: 6. Mobley, K.B. & Jones, A.G. 2009. Environmental, demographic, and genetic mating

109 system variation among five geographically distinct dusky pipefish (Syngnathus floridae) populations. Mol. Ecol. 18: 1476–1490. Mobley, K.B. & Jones, A.G. 2007. Geographical variation in the mating system of the dusky pipefish (Syngnathus floridae). Mol. Ecol. 16: 2596–2606. Neff, B.D. 2001. Genetic Paternity Analysis and Breeding Success in Bluegill Sunfish (Lepomis macrochirus). J. Hered. 92: 111–119. Neff, B.D. & Clare, E.L. 2008. Temporal variation in cuckoldry and paternity in two sunfish species (Lepomis spp.) with alternative reproductive tactics. Can. J. Zool. 86: 92–98. Neff, B.D., Fu, P. & Gross, M.R. 2003. Sperm investment and alternative mating tactics in bluegill sunfish (Lepomis macrochirus). Behav. Ecol. 14: 634–641. Neff, B.D., Pitcher, T.E. & Ramnarine, I.W. 2008. Inter-population variation in multiple paternity and reproductive skew in the guppy. Mol. Ecol. 17: 2975– 2984. Orlando, E.F., Bass, D.E., Caltabiano, L.M., Davis, W.P., Gray, L.E. & Guillette, L.J. 2007. Altered development and reproduction in mosquitofish exposed to pulp and paper mill effluent in the Fenholloway River, Florida USA. Aquat. Toxicol. 84: 399–405. Ota, K., Hori, M. & Kohda, M. 2012. Testes Investment along a Vertical Depth Gradient in an Herbivorous Fish. Ethology 118: 683–693. Portnoy, D.S., Piercy, A.N., Musick, J. a, Burgess, G.H. & Graves, J.E. 2007. Genetic polyandry and sexual conflict in the sandbar shark, Carcharhinus plumbeus, in the western North Atlantic and Gulf of Mexico. Mol. Ecol. 16: 187–97. Rispoli, V.F. & Wilson, A.B. 2008. Sexual size dimorphism predicts the frequency of multiple mating in the sex-role reversed pipefish Syngnathus typhle. J. Evol. Biol. 21: 30–38. Rossouw, C., Wintner, S.P. & Bester-Van Der Merwe, A.E. 2016. Assessing multiple paternity in three commercially exploited shark species: Mustelus mustelus , Carcharhinus obscurus and Sphyrna lewini. J. Fish Biol. 89: 1125–1141. Saïdi, B., Bradaï, M.N. & Bouaïn, A. 2009. Reproductive biology and diet of Mustelus punctulatus (Risso, 1826) (Chondrichthyes: Triakidae) from the Gulf of Gabès, central Mediterranean Sea. Sci. Mar. 73: 249–258.

110 Saïdi, B., Bradaï, M.N. & Bouaïn, A. 2008. Reproductive biology of the smoothhound shark Mustelus mustelus (L.) in the Gulf of Gabès (south-central Mediterranean Sea). J. Fish Biol. 72: 1343–1354. Schaedelin, F.C., van Dongen, W.F.D. & Wagner, R.H. 2015. Mate choice and genetic monogamy in a biparental, colonial fish. Behav. Ecol. 26: 782–788. Schrader, M., Apodaca, J.J., Macrae, P.S.D. & Travis, J. 2012. Population density does not influence male gonadal investment in the least killifish, Heterandria formosa. Ecol. Evol. 2: 2935–2942. Sefc, K.M., Mattersdorfer, K., Sturmbauer, C. & Koblmüller, S. 2008. High frequency of multiple paternity in broods of a socially monogamous cichlid fish with biparental nest defence. Mol. Ecol. 17: 2531–2543. Skinner, A.M.J. & Watt, P.J. 2007. Phenotypic correlates of spermatozoon quality in the guppy, Poecilia reticulata. Behav. Ecol. 18: 47–52. Soucy, S. & Travis, J. 2003. Multiple paternity and population genetic structure in natural populations of the poeciliid fish, Heterandria formosa. J. Evol. Biol. 16: 1328–1336. Stiver, K.A., Fitzpatrick, J.L., Desjardins, J.K. & Balshine, S. 2009. Mixed parentage in Neolamprologus pulcher groups. J. Fish Biol. 74: 1129–1135. Taylor, M.I., Morley, J.I., Rico, C. & Balshine, S. 2003. Evidence for genetic monogamy and female-biased dispersal in the biparental mouthbrooding cichlid Eretmodus cyanostictus from Lake Tanganyika. Mol. Ecol. 12: 3173– 3177. Utne-Palm, A.C., Eduard, K., Jensen, K.H., Mayer, I. & Jakobsen, P.J. 2015. Size dependent male reproductive tactic in the two-spotted goby (Gobiusculus flavescens). PLoS One 10: e0143487. Verissimo, A., Grubbs, D., McDowell, J., Musick, J. & Portnoy, D. 2011. Frequency of Multiple Paternity in the Spiny Dogfish Squalus acanthias in the Western North Atlantic. J. Hered. 102: 88–93. Zane, L., Nelson, W.S., Jones, A.G. & Avise, J.C. 1999. Microsatellite assessment of multiple paternity in natural populations of a live-bearing fish, Gambusia holbrooki. J. Evol. Biol. 12: 61–69.

111 Chapter 3: Sperm competition and the evolution of testes organisation in sharks and rays

Status: Manuscript

Authors: Amy Rowley, Mark F. Stidworthy, Ariel Kahrl, Kyle Martin, Mariana Rego, Fabio Hazin, Toby S. Daly-Engel, Francisco Garcia-Gonzalez and John L. Fitzpatrick

Author contributions:

Conceptualization: AR, FGG, JLF

Data collection: AR, MFS, MR, FH, TSDE, FGG, JLF

Data analysis: AR, JLF, AK, KM

Writing – original draft: AR, JLF

Writing – review and editing: All authors

Funding acquisition: JLF

112 3. Sperm competition and the evolution of testes organisation in sharks and rays

Authors Amy Rowley1,2, Mark F. Stidworthy3 Ariel Kahrl2, Kyle Martin4, Mariana Rego5, Fabio Hazin6, Toby S. Daly-Engel7, Francisco Garcia-Gonzalez8,9 and John L. Fitzpatrick2

Affiliations

1 Faculty of Biology, Medicine and Health, University of Manchester, UK 2 Department of Zoology, Stockholm University, Stockholm, Sweden 3 International Zoo Veterinary Group Pathology, Keighley, UK 4 Department of Physiology and Pharmacology, Molecular and Cellular Exercise Physiology, Karolinska Institutet, Stockholm, Sweden 5 Laboratório de Histologia Animal, Departamento de Morfologia e Fisiologia Animal, Universidade Federal Rural de Pernambuco, Recife, Brazil 6 Laboratório de Oceanografia Pesqueira, Departamento de Pesca e Aquicultura, Universidade Federal Rural de Pernambuco, Recife, Brazil 7 Department of Ocean Engineering and Marine Sciences, Florida Institute of Technology, Melbourne, FL, USA

8 Estación Biológica de Doñana-CSIC, Sevilla, Spain

9Centre for Evolutionary Biology, School of Biological Sciences, University of Western Australia, Crawley, Western Australia 6009, Australia

113 3.1 Abstract

When sperm from multiple males compete for fertilisations after mating (i.e. sperm competition), selection is expected to favour traits that allow males to outperform rivals and increase their probability of siring offspring. Sperm competition is expected to generate selection for increased sperm production, as one of the strongest predictors of success in sperm competition is the number of sperm that are released into the fertilisation environment. Increased sperm production can be achieved by increasing the size of the testes and/or through changes in the structural organisation of the testes. However, while there is ample evidence that sperm competition promotes increases in testes size across taxa, we know far less about how testes organization responds to variation in the strength of sperm competition. Here we use a comparative approach to evaluate testes evolution and examine how sperm competition risk influences testes organisation in sharks and rays (elasmobranchs), two basal vertebrate groups displaying wide variation in female promiscuity across species. We show that the proportion of sperm- producing tissue, spermatocyst size, and body mass all follow a late-burst model of evolution across elasmobranchs. In sharks, neither the proportion of sperm- producing tissue within the testes nor the size of the spermatocysts are associated with sperm competition risk, a result that stands in contrast to the pattern observed in other taxonomic groups. However, in rays, we found limited evidence that the proportion of sperm-producing tissue is positively associated with sperm competition risk. These results indicate that testes organisation may exhibit different responses to sperm competition between sharks and rays. We discuss the potential role of spatial constraints in shaping evolutionary responses of testes to sperm competition.

114 3.2 Introduction

Competition to sire offspring imposes strong selective pressure on males to outperform rivals. When females mate with multiple males, competition among rivals can continue after mating in the form of sperm competition (Parker 1970), a post-mating process that exerts a profound influence on the evolution of male reproductive traits (Simmons and Fitzpatrick 2012). The outcome of sperm competition is hypothesized to be largely determined by the relative number of sperm released into the fertilisation environment by each competitor (Parker and Pizzari 2010), and this prediction is well supported by empirical evidence across a broad range of taxa (Martin et al. 1974; Gage and Morrow 2003; Boschetto et al. 2011; Firman and Simmons 2011). Due to the positive relationship between testes size and the number of sperm produced (Marconato and Shapiro 1996; Schärer et al. 2004; Lüpold et al. 2009; Ramm and Stockley 2010; Rowe and Pruett-Jones 2011; Ramm and Schärer 2014), sperm competition is expected to select for an increase in testes size relative to body size (Parker and Ball 2005; Parker and Pizzari 2010). In accordance with this key theoretical prediction, a positive relationship between body size-corrected testes mass and various metrics of sperm competition (e.g. rates of multiple paternity, mating system, female remating rates) has been demonstrated in numerous comparative studies (Chapter 2; Møller and Briskie 1995; Stockley et al. 1997; Byrne et al. 2002; Ramm et al. 2005; Soulsbury 2010; Simmons and Fitzpatrick 2012), as well as in intraspecific studies of populations experiencing different levels of sperm competition (Firman and Simmons 2008). In addition, selection lines that experimentally manipulate sperm competition risk reveal that males invest more in testes size in selection lines where females mate multiply (i.e. increased sperm competition risk) compared with selection lines with enforced female monogamy (Hosken and Ward 2001; Pitnick et al. 2001; Simmons and García-González 2008, but see Firman and Simmons 2010). However, while the link between sperm competition and testes size is well-documented (Simmons and Fitzpatrick 2012), the response of the structural organisation of the testes to sperm competition and its potential role in increasing sperm production has received relatively little attention (Ramm and Schärer 2014).

115

The number of sperm a male produces is determined not only by the size of the testes, but also by the amount of sperm-producing tissue they contain and the speed at which sperm are produced (Lüpold et al. 2009; Ramm and Schärer 2014; Ramm et al. 2014). However, the duration of the spermatogenic cycle is not associated with sperm competition risk (at least in primates, Wistuba et al. 2003; Luetjens et al. 2005), suggesting that selection may target the amount of sperm producing tissue rather than the rate of spermatogenesis. In addition to, or instead of, developing larger testes, selection may therefore act to increase sperm production through structural changes to the testes (Ramm and Schärer 2014), such as increasing the proportion of sperm-producing tissue (Lüpold et al. 2009). While the proportion of sperm-producing tissue varies widely across species (Russell et al. 1990), the effect of sperm competition in influencing testes organisation has been evaluated by only a handful of studies. Comparative studies of passerine birds and mice (Lüpold et al. 2009; Rowe and Pruett-Jones 2011; Montoto et al. 2012) show that males that experience a greater risk of sperm competition have testes comprised of a higher proportion of sperm-producing tissue. Similarly, in populations of house mice experimentally evolving under monogamous and polygamous conditions, the testes of polygamous males have a higher proportion of sperm-producing tissue than monogamous males (Firman et al. 2015). These results demonstrated evolutionary responses of testes organisation to increases in the levels of sperm competition. In the capybara, more socially dominant males, who are expected to experience lower sperm competition risk due to their ability to control access to females, display a lower proportion of sperm producing tissue than subordinate males (López et al. 2008). However, as all studies have thus far focused on either birds or mammals, it remains unclear how widespread this pattern of increasing seminiferous tissue density in response to sperm competition is across a broader taxonomic spread.

Here we examine the relationship between sperm competition risk and the structural organisation of the testes in sharks and rays, a basal vertebrate lineage characterised by widely varying reproductive behaviours (Byrne and Avise 2012;

116 Fitzpatrick et al. 2012). Female sharks and rays have been observed to mate with multiple males in a single reproductive cycle (Yano et al. 1999; Chapman et al. 2003) and show substantial variation across species in the frequency of litters sired by multiple males (ranging from 11-92%, Chapter 2). This variation indicates that elasmobranch species experience radically different levels of sperm competition, which may impose selection on testes architecture to increase sperm production. We take advantage of the variation in sperm competition risk in this group to test how body size corrected testes mass, a widely-used proxy measure for sperm competition risk recently validated against genetic mating system in sharks (Chapter 2), affects the proportion of the testes comprised of sperm-producing tissue and the size of the seminiferous follicles. We hypothesized that males belong to species at higher sperm competition risk would have a greater proportion of sperm-producing tissue within the testes. We also evaluate the pattern of diversification of the proportion of sperm-producing tissue in the testes and maximum follicle size over time to provide insight into the evolution of testes organisation in sharks and rays.

3.3 Methods

3.3.1 Sample collection

Testes samples were collected from 95 individuals across 26 species (18 sharks and 8 rays). Of these, testes were sampled from 43 individuals across 10 species (8 shark and 2 ray) from field collections in 6 countries across 5 continents between 2014-2016 (Supporting Material Table S3.1). Among wild caught samples, only sharks and rays that were caught as part of commercial or artisanal fisheries for subsequent sale or as bycatch were sampled, and in no case were specimens sacrificed solely for the purpose of this study. Sharks and rays were either dissected immediately post-mortem, or frozen at -25C and thawed before dissection. Testes samples were also obtained from 51 individuals representing 16 species (10 sharks

117 and 6 rays) by the International Zoo Veterinary Group Pathology division in the United Kingdom (henceforth referred to as IZVG). Specimens were obtained from 24 public aquaria, 23 in the UK or Europe and one in South-East Asia. Thirty-two individuals were wild-caught, and 6 were bred in captivity. The origin of the remaining 13 individuals could not be determined, though it is likely that most were wild-caught as relatively few shark species breed reliably in captivity. Deaths were mostly spontaneous or occasionally by humane methods of euthanasia (terminal anaesthesia either by injectable agent or anaesthetic bath) under the direction of a veterinary surgeon in accordance with animal care policies within the respective institutions. All aquaria were appropriately licensed under United Kingdom or their respective national zoo and aquaria licensing legislation. In almost all cases, fresh carcasses were dissected by aquarists on-site and formalin-fixed tissues were submitted to the pathology laboratory for further processing and examination. Carcasses were not frozen prior to examination. In all cases, we assessed the maturity of sharks by examining the claspers, with fully calcified claspers indicating a sexually mature individual (Hamlett 2005). The testes of mature males were excised and small cross-sections of the testis (approximately 1cm thick) from each male were preserved in 10% formalin neutral buffered solution.

Assessing sharks and rays collected from the field and from aquaria raises a potential confound, as animals likely differ in age and social environments, which could potentially influence testes organization. Although the impact of these potentially confounding variables on testes organization has not been considered in elasmobranchs, available evidence from the mammalian literature suggests that these variables are unlikely to affect testes organization. For example, the proportion of sperm producing tissue present in the testes remains remarkably consistent across age classes in stallions (Johnson and Neaves 1981) and a recent study on house mice found plasticity in sperm production in males reared under different social conditions (competitive vs. non-competitive environments), but no evidence that this plasticity could be explained by changes in the spatial organisation of the testes (Firman et al. 2018).

118

3.3.2 Histological processing

Tissues were processed routinely for histological examination with haematoxylin and eosin staining. A cross-section of each testis was trimmed to approximately 10mm2 in size and processed overnight in a Leica TP1050 ASP300 enclosed tissue processor, in which tissues were dehydrated by successive immersion in increasing concentrations of ethanol from 70 – 100%, then immersed in xylene and finally in paraffin wax. After processing, tissues were embedded in molten paraffin wax, left to dry completely, and sectioned on a rotary microtome to 4 - 5 µm thickness. Five sections were taken from each testis, at 200 µm intervals, mounted on microscope slides and allowed to dry overnight in an incubator at 37°C. The sections were then stained with haematoxylin and eosin using a Leica XL autostainer.

3.3.3 Image analysis

All slides were scanned using a 20x/0.80 Plan Apo objective on a 3D Histech Panoramic 250 Flash II slide scanner. Image analysis was performed using 3D Histech Panoramic Viewer software (v1.15.4) and ImageJ (Rasband 1997). Elasmobranch testes show substantial differences in structure from other commonly-studied taxa (Pratt 1988; Schlatt and Ehmcke 2014); unlike birds and mammals, where sperm is produced in fixed tubular structures (seminiferous tubules, Schlatt and Ehmcke 2014), shark and ray seminiferous tissue is organised in a cystic structure (called seminiferous follicles or spermatocysts), which are produced in the germinal zone and migrate across the testis as sperm mature (Pratt 1988). Therefore, our analyses focused on the size and proportion of spermatocysts within the testes.

We haphazardly placed 10 circular boundaries of radius 600 µm in testicular regions showing mature spermatocysts (sperm stages 5-6), avoiding any areas showing

119 obvious signs of tissue damage or artefacts of tissue processing (e.g. tears, staining irregularities). Within each boundary, we measured the area of each spermatocyst using the freehand trace tool in 3D Histech Panoramic Viewer. In species with cystic testes, the size of the spermatocysts is determined by the number of pre-meiotic divisions of the germ cells (Ramm et al. 2014; Schlatt and Ehmcke 2014). Therefore, the size of the spermatocyst is indicative of the final number of mature sperm produced by the cyst, and a likely target of selection to increase sperm production capabilities (Ramm et al. 2014; Schlatt and Ehmcke 2014). However, estimating the size of the spermatocyst is challenging due to the cystic structure of elasmobranch testes, as it is not possible to section through the equatorial diameter of every spermatocyst in a single section. We therefore used the maximum spermatocyst area for each individual sampled as the most reliable estimate of spermatocyst size at the equatorial diameter, as mean spermatocyst area would necessarily underestimate true spermatocyst size. We calculated the proportion of the total testes area measured that was comprised of seminiferous tissue (sum of follicle areas/total area*100). We then calculated the mean proportion of seminiferous tissue and the mean maximum spermatocyst area for each species.

Due to the constraints of sampling sharks in the field, there were unavoidable differences in how samples were treated before fixing. For example, individuals that could not be dissected immediately post-mortem were frozen before dissection, which resulted in freeze artefacts in the tissue; specifically, the shrinking of the spermatocysts. To avoid this confounding factor, we restricted our analysis of the proportion of sperm-producing tissue and maximum follicle size to only those samples that did not display freeze artefacts. Our final dataset incorporated testes samples from 52 individuals across 18 species (11 sharks and 7 rays, mean = 2.89 males per species).

3.3.4 Estimating sperm competition risk

120 Body size-corrected testes mass was used as a proxy for sperm competition risk due to the close association between testes mass relative to body size and other measures of sperm competition risk across a wide range of taxa (Simmons and Fitzpatrick 2012). Body size-corrected testes mass has also been shown to correlate positively with genetic mating system and the number of sires per brood in sharks (Rowley et al. submitted), which supports its validity as a reliable estimate of sperm competition risk within the species that are the subject of our analyses. Where possible (i.e. when sharks were sampled in our own field collections), testes and body mass (g) was measured for the same individuals that testes sections were collected from. However, in cases where testes and body mass could not be measured directly (e.g. when animals were too large to be weighed or because total body weight and gonad weight are not routinely measured during necropsies conducted by aquaria), we searched the literature using the species name in combination with the words ‘testes mass’, ‘body mass’ or ‘gonadosomatic index’ (GSI, typically quantified as the percentage of body mass made up of testicular tissue). If the raw data were not reported in the relevant studies found in our search, we contacted the authors of those studies directly to request the data or calculated body and testes mass from GSI, or used the program GraphClick v3.0.3 (Boyle et al. 2012) to extract data from figures (summarised in Supporting Information Table S3.1). As testes mass can vary throughout the reproductive cycle, where GSI data were available throughout the year, we calculated mean testes mass at peak GSI.

3.3.5 Phylogenetic analyses

We applied a range of phylogenetic models to assess the evolution of testes structural organization in sharks and rays. To do this, we used a recent phylogeny constructed by Stein et al. (2018) using genetic data from more than 600 species of sharks and rays. We used a set of 500 phylogenetic trees to generate a consensus tree using the function consensus.tree with the function consensus.edges to set

121 branch lengths in the package phytools (Revell 2012). Sixteen of the 18 species in our dataset were also present in the phylogeny (10 sharks and 6 rays).

To evaluate trait evolution across elasmobranchs, we assessed phylogenetic signal, a measure of the dependence of species’ trait values on their phylogenetic relationships, using the phylosig function in the package phytools (Revell 2012) in R version 2.15.1 (R Core Team 2017). Phylogenetic signal was assessed using two measures: Blomberg’s K (Blomberg et al. 2003), which compares observed pattern of trait variance to a null model under Brownian motion, and Pagel’s λ (Pagel 1999), which evaluates the phylogenetic dependence of the trait independently. A K value of 1 indicates that traits evolve as expected under Brownian motion, while a value lower than 1 indicates less phylogenetic signal and a value higher than 1 indicates more phylogenetic signal than would be expected under Brownian motion. A λ value of 0 indicates no phylogenetic signal, while a value of 1 indicates complete phylogenetic dependence. We then used likelihood ratio tests to compare the maximum-likelihood λ value to λ estimates constrained to 0 and 1. Due to recent criticism of the use of likelihood ratio tests to model modes of trait evolution across phylogenies with low sample sizes, which is likely to generate spurious results (such as a tendency for Ornstein-Uhlenbeck, (OU) models of evolution to be incorrectly favoured over other models) (Cooper et al. 2016), we evaluated patterns of trait evolution using the function fitContinuousMCMC in the R package geiger (Slater et al. 2012). This method utilises a Markov Chain Monte Carlo (MCMC) approach to identify the best-fitting evolutionary model from Brownian motion (BM), single stationary peak (SSP, or single-optimum OUI model), accelerating/decelerating with an exponential rate of evolutionary change (ACDC.exp) and accelerating/decelerating with a linear rate of evolutionary change (ACDC.lin) for each trait measured (proportion of sperm-producing tissue, maximum follicle area and body mass). Models fits were compared using the Akaike Information Criterion for MCMC samples (AICm) (Rafferty et al. 2007). We set uninformative priors and iteratively tweaked proposal widths based on chain convergence, which was assessed in Tracer (version 1.5, Drummond et al. 2003). Each analysis was run for 10,000,000 iterations, with sampling from the chain every 1000 iterations and the

122 first 10% of samples discarded as burn-in. The models assume equal rates of evolution across all branches of the tree. For ACDC.exp and ACDC.lin models, evolutionary rates change can occur across time, but the same rates apply to all branches at any given time.

We used phylogenetically controlled linear regressions models to compare testes organization between sharks and rays. Separate regressions were used to examine the relationship between the proportion of spermatogenetic tissue within the testes and sperm competition risk, and between maximum follicle area and sperm competition risk in sharks and rays. We treated sharks and rays separately in our analyses as previous work suggests that there are dramatically different evolutionary responses to sperm competition in these two groups (Chapter 4). As our earlier analyses indicated that an ACDC.lin model of trait evolution provided the best fit to both testes organisation traits and body mass (see Results), we ran models specifying an early-burst (EB) model of evolution (which is equivalent to the ACDC transformation) using the phylolm function in the R package ‘phylolm’ (Tung Ho and Ané 2014). Data were log10 transformed prior to analysis in R version 3.4.1 (R Core Team 2017). Of the 16 species from which we collected histological sections that were also present in the phylogeny, testes and body mass data were available for 11 species (7 sharks and 4 rays). To control for the allometric relationship between testes mass and body size, body mass was included as a covariate in the model.

3.4 Results

Testes structural organization varied among sharks and rays (Figure 3.1). In sharks, the proportion of spermatogenic tissue within the testes varied from 0.69 in the nurse shark (Ginglymostoma cirratum) to 0.89 in the nursehound (Scyliorhinus stellaris), while in rays, the proportion of spermatogenic tissue varied from 0.71 in the smalleyed ray (Raja microocellata) to 0.95 in the Southern stingray (Dasyatis americana) (Figure 3.1). The maximum spermatocyst area showed considerable

123 variation across species, ranging in sharks from 37041 m2 to 110689 m2 in the sandbar shark (Carcharhinus plumbeus) and nursehound (S. stellaris), respectively, and in rays from 31303 m2 in the spotted eagle ray (Aetobatus narinari) to 93860 m2 in the thornback ray (Raja clavata). This variation in maximum spermatocyst area represents a nearly three-fold difference in both sharks and rays.

Despite this variance in testes organization, sharks and rays did not differ in either the proportion of spermatogentic tissue (AIC = -56.00, lnL = 32.00, df = 14, slope = - 0.01  0.05, t = -0.36, P = 0.73, r = -0.10, lower CI = -2.32, upper CI = 1.61) or the maximum spermatocyst area (AIC = 3.32, lnL = 2.34, df = 14, slope = 0.13  0.31, t = -0.41, P = 0.69, r = -0.11, lower CI = -2.37, upper = 1.56). However, in sharks there was a significant positive relationship between the proportion of spermatogenetic tissue in the testes and the maximum spermatocyst area (AIC = -39.17, lnL = 23.59, df = 8, slope = 0.13  0.04, t = 3.14, P = 0.01, r = 0.74, lower CI = 0.61, upper CI = 5.56), suggesting covariance in testes organizational structure in sharks. In contrast, there was no association between the proportion of spermatogenetic tissue and the maximum spermatocyst area in rays, although the low sample size suggests there may be limited power in this analysis (AIC = -20.98, lnL = 14.49, df = 4, slope = 0.08  0.04, t = 1.80, P = 0.15, r = 0.67, lower CI = -0.58, upper CI = 4.02). In both sharks and rays, neither the proportion of spermatogentic tissue (sharks: AIC = - 31.15, lnL = 19.58, df = 8, slope = -0.00  0.01, t = -0.05, P = 0.96, r = -0.02, lower CI = -2.01, upper CI =1.91 ; rays: AIC = -21.89, lnL = 14.95, df = 4, slope = 0.03  0.01, t = 1.75, P = 0.16, r = 0.66, lower CI = -0.61, upper CI = 3.96) nor the maximum spermatocyst area (sharks: AIC = 3.90, lnL = 2.05, df = 8, slope = -0.01  0.07, t = - 0.19, P = 0.85, r = -0.07, lower CI = -2.15, upper CI =1.78; rays: AIC = 6.04, lnL = 0.98, df = 4, slope = 0.05  0.15, t = 0.35, P = 0.74, r = 0.17, lower CI = -1.64, upper CI = 2.30) were associated with body mass.

124

Mustelus_asteriasMustelus asterias

Mustelus_mustelusMustelus mustelus

Carcharhinus_melanopterusCarcharhinus melanopterus

Carcharhinus_plumbeusCarcharhinus plumbeus

Galeus_melastomusGaleus melastomus

Scyliorhinus_caniculaScyliorhinus canicula

Scyliorhinus_stellarisScyliorhinus stellaris

Hemiscyllium_ocellatumHemiscyllium ocellatum

Chiloscyllium_punctatumChiloscyllium punctatum

Ginglymostoma_cirratumGinglymostoma cirratum

Raja_asteriasRaja asterias

Raja_clavataRaja clavata

Raja_undulataRaja undulata

Urobatis_jamaicensisUrobatis jamaicensis

Dasyatis_americanaDasyatis americana

Aetobatus_narinariAetobatus narinari 0 25 50 75 100 0 30000 60000 90000 120000 30 60 90 % Spermatogenic Tissue Maximum Follicle Area (..m) Body mass (kg)

% Sperm-producing tissue Maximum spermatocyst area (µm2) Body mass (kg)

Figure 3.1. Variation in testes organisation traits across sharks and rays, as illustrated by haematoxylin and eosion-stained histological testes sections from (a) Ginglymostoma cirratum, (b) Mustelus asterias, (c) Chiloscyllium punctatum and (d) Scyliorhinus stellaris, shown at 20x magnification. Phylogenetic relationships between the 18 species in our dataset are presented with bars representing the mean proportion of sperm-producing tissue within the testes, the maximum cross- sectional area of the seminiferous follicles and mean body mass (plotted on a log10-scale) for each species.

Across sharks and rays, the evaluation of phylogenetic signal with Blomberg’s K and Pagel’s λ showed qualitatively similar results, with the proportion of seminiferous tissue, maximum follicle size, and body mass all displaying low phylogenetic signal (Table 3.1), indicating that none of the trait we considered appears to evolve under a Brownian motion model of evolution. When we compared alternative models of trait evolution we found complementary results. Comparisons of AICm weights among alternative evolutionary models showed that the best model of trait evolution for the proportion of sperm-producing tissue, maximum follicle size and body size is ACDC linear; a model of accelerating evolution with a linear rate of change through time (Table 3.2). This suggests that both testes structural organization and body size exhibit late-burst patterns of evolution, with the rate of diversification of all three traits becoming more rapid toward the tips of the phylogeny (Figure 3.2).

Table 3.1. Evaluation of phylogenetic signal in the proportion of sperm-producing tissue within the testes and maximum cross-sectional area of the seminiferous follicles. Phylogenetic signal was assessed using Blomberg’s K and Pagel’s λ. K and λ values closer to 1 indicate stronger phylogenetic signal. The maximum-likelihood estimate of Pagel’s λ was compared to models where λ=0 (no phylogenetic signal) and λ=1 (complete phylogenetic dependence) using likelihood ratio tests, with significant p-values indicating that the maximum-likelihood value of λ deviated

significantly from these models.

Blomberg’s K Pagel’s λ

Trait K P λ lnL λ lnL λ = 0 (P‐value) lnL λ = 1 (P‐value) % Seminiferous tissue 0.46 0.19 <0.001 -49.77 -49.77 (1.00) -51.96 (0.04) Maximum follicle area 0.35 0.41 0.03 -184.82 -184.82 (0.97) -189.32 (0.003) Body Mass 0.55 0.12 <0.001 -182.57 -182.57 (1.00) -183.35 (0.21)

Table 3.2. Summary of model fits for Brownian motion (BM), single stationary peak (SSP, or single-optimum Ornstein-Uhlenbeck (OU) model), and accelerating/decelerating, where the change in rate is either exponential (ACDC exponential) or linear (ACDC linear), models of evolution for the percentage of seminiferous tissue within the testes (% Seminiferous tissue), the maximum cross- sectional area of the seminiferous follicles (Maximum follicle area), and adult body mass. Comparisons of evolutionary models was performed using Aikake Information Criterion for Markov Chain Monte Carlo samples (AICm), with delta AIC

(ΔAICm) and AICm weights (AICm ω) for each model presented for each trait.

% Seminiferous tissue Maximum follicle area Body Mass Model AICm ΔAICm AICm ω AICm ΔAICm AICm ω AICm ΔAICm AICm ω BM -61.53 5.33 0.07 50.93 7.20 0.03 139.92 7.83 0.02 SSP (OU) -38.38 28.47 0.00 68.76 25.03 0.00 157.33 25.24 0.00 ACDC exponential 267.19 334.04 0.00 93.73 50.00 0.00 187.99 55.89 0.00 ACDC linear -66.85 0.00 0.93 43.73 0.00 0.97 132.09 0.00 0.98

In sharks, there was no association between either the proportion of spermatogenic tissue or the maximum follicle area and body size corrected testes mass, a proxy measure for the sperm competition risk (Figure 3.3, Table 3.3). In rays, we detected a significant positive relationship between the proportion of spermatogenic tissue and body size corrected testes mass (Figure 3.3, Table 3.3). However, despite the surprisingly tight correlation between these variables (adjusted r2 = 0.99, see Figure 3.3) the low sample size (n = 4) demands caution in interpreting this result. In rays there was no relationship between maximum follicle area and body size-corrected testes mass (Figure 3.3, Table 3.3).

128 Dasyatis americana

(a) 4.55

Scyliorhinus stellaris 4.50 Carcharhinus melanopterus Mustelus asterias Chiloscyllium punctatum Aetobatus narinari 4.45 Mustelus mustelus Raja undulata Scyliorhinus canicula Galeus melastomus

4.40 Urobatis jamaicensis Hemiscyllium ocellatum producing Tissue Raja clavata − Carcharhinus plumbeus

4.35 Raja asterias 4.30 % Sperm

4.25 Ginglymostoma cirratum

0.00 69.21 138.42 207.64 276.85

(b) Scyliorhinus stellaris 11.6 Carcharhinus melanopterus Raja clavata Raja undulata Galeus melastomus 11.4 Chiloscyllium punctatum Scyliorhinus canicula 11.2

Mustelus mustelus

11.0 Mustelus asterias Hemiscyllium ocellatum Raja asterias Dasyatis americana

10.8 Ginglymostoma cirratum

10.6 Carcharhinus plumbeus Urobatis jamaicensis

10.4 Aetobatus narinari Maximum SpermatocystMaximum Area 10.2

0.00 69.21 138.42 207.64 276.85

Ginglymostoma cirratum

11 Carcharhinus plumbeus (c) Aetobatus narinari Dasyatis americana 10 Chiloscyllium punctatum

9 Carcharhinus melanopterus Mustelus mustelus Raja undulata Raja clavata 8 Scyliorhinus stellaris Mustelus asterias Body Mass Raja asterias 7 Hemiscyllium ocellatum Galeus melastomus

6 Urobatis jamaicensis Scyliorhinus canicula

0.00 69.21 138.42 207.64 276.85 Relative Time Figure 3.2. Traitgrams showing phenotypic divergence of (a) the proportion of sperm-producing tissue, (b) maximum follicle cross-sectional area and (c) body mass over time. All data are ln-transformed. All traits follow a late-burst pattern of evolution, with the rate of evolutionary diversification increasing over relative time.

129 Table 3.3. Phylogenetically controlled linear regressions between testes organisation traits and testes mass in (a) sharks and (b) rays. Body mass is included as a covariate in all models to control for the allometric relationship between testes and body size. The phylogenetic scaling parameter λ measures the level of phylogenetic dependence of the data, ranging from 0 (low phylogenetic signal) to 1 (high phylogenetic signal). We present the degrees of freedom (df), Aikake Information Criterion (AIC), log-likelihood (lnL), slope of the regression with standard error, t-statistic and p-value for each model. Significant results (p<0.05) are highlighted in bold.

Testes Trait Predictors df AIC lnL Slope SE t p r Lower Upper CI CI (a) Sharks

Proportion of Testes mass 4 -21.93 15.96 -0.02 0.11 -0.21 0.84 -0.10 -2.16 1.77 seminiferous tissue Body mass 0.03 0.09 0.32 0.77 0.16 -1.67 2.27

Maximum follicle Testes mass 4 6.93 1.54 -0.20 0.84 -0.24 0.82 -0.12 -2.19 1.74 area Body mass 0.12 0.67 0.18 0.86 0.09 -1.79 2.13

(b) Rays Proportion of Testes mass 1 -32.70 21.35 0.35 0.02 14.08 <0.05 1.00 0.08 31.64 seminiferous tissue Body mass -0.12 0.01 -10.05 0.06 -1.00 -22.64 0.18

Maximum follicle Testes mass 1 2.03 3.99 1.83 1.89 0.97 0.51 0.70 -1.43 3.12 area Body mass -0.10 0.90 -0.11 0.93 -0.11 -2.05 1.88

130

(a) (b)

Figure 3.3. Relationships between testes mass corrected for body size (a proxy measure for sperm competition risk) and (a) the percentage of sperm-producing tissue within the testes and (b) maximum cross-sectional area of the seminiferous follicles across shark (black points) and ray (white points) species (rays: log10(y) = 1.91 + 0.34x). Body size-corrected testes mass is calculated as residual values from a linear regression of testis mass on body mass. Data in this figure are not phylogenetically controlled.

131 3.5 Discussion

Our results provide little support for the hypothesis that sperm competition influences the organisation of testes in sharks and rays. Both the proportion of sperm-producing tissue within the testes and the maximum spermatocyst area showed considerable variation across the shark and ray species we assessed (Figure 3.1). Yet, in sharks, neither trait was associated with body size-corrected testes mass, a proxy measure of sperm competition risk. In rays, we found suggestive evidence that sperm competition influences testes architecture by increasing the proportion of sperm-producing tissue in the testes. This result is in keeping with previous empirical demonstrations that testes have increased sperm production capacity in species experiencing increased sperm competition risk in birds and mammals (Lüpold et al. 2009; Rowe and Pruett-Jones 2011; Montoto et al. 2012). However, we interpret the relationship found in rays with caution, as we were only able to assess a limited number of species in our analyses. Nevertheless, our results raise the possibility that testes organization in sharks and rays exhibit divergent responses to sperm competition risk.

Unlike other reproductive traits in animals (e.g. Simmons and Fitzpatrick 2016), including sperm length in sharks (Chapter 4), the proportion of sperm producing tissue and spermatocyst area showed low phylogenetic signal. When assessing alternative models of trait evolution of testes organisational structure and body mass, we found that both testes architecture and body mass conform to a late- burst linearly-accelerating model of trait evolution. This indicates that the rates of evolutionary diversification of these traits has increased over time, with the greatest phenotypic divergence occurring relatively late. The similarity in evolutionary models of trait evolution that we detected could arise from underlying correlations between body size and testes architecture. However, body size was not associated with either measure of testes architecture we examined, suggesting that selection on body size is not driving the evolutionary pattern detected in testes organizational structure. A late-burst pattern of evolution suggests that testes architecture and body mass evolve rapidly and could be associated with a strong

132 ecological driver, such as a rapid niche change (e.g. Supriya et al. 2016). However, without specific testing of this hypothesis, the selective forces that drive this accelerating evolutionary diversification in sharks and rays remain enigmatic.

In contrast to the inter- and intraspecific patterns observed in birds and mice (Lüpold et al. 2009; Rowe and Pruett-Jones 2011; Montoto et al. 2012; Firman et al. 2015), we found limited support for a pattern of increasing sperm-producing tissue in response to increasing sperm competition risk in elasmobranchs. This discrepancy raises the question of why sharks and rays may differ from other vertebrates, and possibly from one another. In sharks and rays, the proportion of seminiferous tissue ranged from 68% to 92% across the species in our dataset. This range exceeds the level of variation found in blackbirds (88 – 96%, Lüpold et al. 2009) and Australian fairy-wrens, emu-wrens and grasswrens (~1% difference among species, Rowe and Pruett-Jones 2011), two passerines groups where the proportion of sperm producing tissue covaries with sperm competition risk. In contrast, mammals show exaggerated variation in sperm producing tissue compare with sharks and rays (32.7% to 92.7%, Russell et al. 1990). Thus, a lack of sufficient variation in testes architecture in sharks and rays is an unlikely explanation for the limited support we report for a link between proportion of sperm producing tissue and sperm competition risk in elasmobranchs. Instead, the inclusion of captive animals housed from aquaria into our examination of testes could have obscured any association between testes organisation and sperm competition risk. However, previous mammalian research demonstrated relationships between sperm competition risk and reproductive traits when assessing zoo animals exclusively (e.g. Anderson et al. 2005). Moreover, very few individuals included in our study were bred in captivity, and they are therefore unlikely to show evolutionary responses to the captive environment. It is therefore also unlikely that the absence of a relationship between testes organisation and sperm competition risk in sharks is a result of sampling from aquaria.

Instead, we suggest that the inconsistent responses in testes organizational structure to sperm competition risk across taxonomic groups might reflect

133 differences in the functional constraints limiting investment in testicular tissue. Both birds and small mammals are likely to experience energetic trade-offs or functional constraints that could limit the potential for increasing testes size. In flying species, such as birds, increased body weight negatively affects flight ability (Wright and Cuthill 1989). As testes mass can represent a substantial proportion of body mass (e.g. Vahed et al. 2010), increases in testes size in response to sperm competition risk may be functionally limited in species that fly. Indeed, gonad size trades-off against flight ability in insects (Saglam et al. 2008; Almbro and Kullberg 2012). Such constraints led Lüpold et al. (2009) to suggest that spatial constraints within the body cavity may drive changes in testes architecture, even in the absence of changes in overall testes mass. Similar spatial constraints may also be present in small mammals, where selection for small body size may impose limits on increases in testes mass in response to sperm competition risk. In keeping with this hypothesis, testes architecture is shaped by sperm competition risk in mice species (Montoto et al. 2012; Firman et al. 2015). In contrast, sharks and rays live in an aquatic medium that alleviates constraints on body weight. Therefore, selective pressure to maximize spatial efficiency in the testes by increasing sperm producing tissue may be relatively relaxed in aquatic organisms. However, even within an aquatic medium, functional constrains may differ between sharks and rays due to divergent body plans. Most sharks are fusiform in shape with large, elongated body cavities, whereas rays are dorsoventrally flattened, which limits the size of the coelomic space. This restriction may impose constraints on the size of the reproductive organs in rays (Musick and Ellis 2005). Interestingly, in our limited sample of ray species, we detected a positive correlation between the proportion of sperm-producing tissue and sperm competition risk, a pattern not seen in sharks. Although this idea remains speculative due to the limits in our sample size, our results provide provisional support for the hypothesis that spatial constraints are a key driver influencing the evolution of testes organizational structure in vertebrates.

We found no relationship between spermatocyst size and sperm competition risk across either shark or ray species. The substantial variation we observed in

134 spermatocyst size across species (Figure 3.1) therefore remains unexplained. Other selective forces besides sperm competition may influence spermatocyst size, such as selection imposed by differences in sperm morphology due to the functional link between the spermatocysts and the sperm they produce. For example, sperm length is positively correlated with the size of the seminiferous tubules in birds, suggesting that selection for longer sperm influences testes architecture (Lüpold et al. 2009). However, we were unable to collect sufficient sperm length data for the species in our dataset to test this hypothesis here. Future work integrating the role of variation in sperm morphology in determining the organisation of the testes would provide valuable insight into the evolution of testes architecture in this group.

Our results show that sperm competition is not a major driver of testes architecture in sharks and rays. This finding contrasts with recent work in other taxonomic groups (Lüpold et al. 2009; Rowe and Pruett-Jones 2011; Montoto et al. 2012) and highlight the potential role of spatial constraints in influencing responses in testes organizational structure to sperm competition risk. Moreover, the limited evidence that the proportion of sperm producing tissue present in the testes is influenced by sperm competition risk in sharks and rays suggests that responses in testes organisation to selection arising from sperm competition may not be universal in vertebrate groups. Future work that capitalises on this extraordinary variation among sharks and rays, while simultaneously assessing differences in fecundity, sperm length, the duration of sperm storage among species, would aid in evaluating patterns of testes evolution in response to sperm competition across vertebrates.

135 3.6 Acknowledgments

We thank Trudi, Vicente and Curro at the Lonja Pesquera de Algeciras for providing access to sharks and rays caught as bycatch, and Samantha Hook, Lou Ruddell and Phil Root for assistance with field sampling. IZVG Pathology thanks the aquarium staff at multiple aquaria involved in the collection of original histological samples. Research was supported by a Biotechnology and Biological Sciences Research Council (BBSRC) Doctoral Training Program studentship to AR, Manchester University and Stockholm University (AR and JLF) and a Knut and Alice Wallenberg Academy Fellowship to JLF. FGG was supported by grants (CGL2012-34685 and CGL2016-76173-P, co-funded by the European Regional Development Fund) from the Spanish Ministry of Economy. Funding for the project was also provided by the Western Australia Government, the University of Western Australia, the Australian Research Council, and a Doñana Biological Station project financed by the Spanish Ministry of Economy, through the Severo Ochoa Program for Centres of Excellence (SEV-2012-0262).

136 3.7 References

Almbro, M., and C. Kullberg. 2012. Weight Loading and Reproductive Status Affect the Flight Performance of Pieris napi Butterflies. J. Insect Behav. 25:441–452. Anderson, M. J., J. Nyholt, and A. F. Dixson. 2005. Sperm competition and the evolution of sperm midpiece volume in mammals. J. Zool. 267:135–142. Blomberg, S. P., T. Garland, and A. R. Ives. 2003. Testing for phylogenetic signal in comparative data: Behavioral traits are more labile. Evolution. 57:717–745. Boschetto, C., C. Gasparini, and A. Pilastro. 2011. Sperm number and velocity affect sperm competition success in the guppy (Poecilia reticulata). Behav. Ecol. Sociobiol. 65:813–821. Boyle, M. A., A. L. Samaha, A. M. Rodewald, and A. N. Hoffmann. 2012. Evaluation of the reliability and validity of GraphClick as a data extraction program. Comput. Human. Behav. 29:1023-1027. Byrne, P. G., J. D. Roberts, and L. W. Simmons. 2002. Sperm competition selects for increased testes mass in Australian frogs. J. Evol. Biol. 15:347–355. Byrne, R. J., and J. C. Avise. 2012. Genetic mating system of the brown smoothhound shark (Mustelus henlei), including a literature review of multiple paternity in other elasmobranch species. Mar. Biol. 159:749-756. Chapman, D. D., M. J. Corcoran, G. M. Harvey, S. Malan, and M. S. Shivji. 2003. Mating behavior of southern stingrays, Dasyatis americana (Dasyatidae). Environ. Biol. Fishes 68:241–245. Cooper, N., G. H. Thomas, C. Venditti, A. Meade, and R. P. Freckleton. 2016. A cautionary note on the use of Ornstein Uhlenbeck models in macroevolutionary studies. Biol. J. Linn. Soc. 118:64–77. Firman, R. C., F. Garcia-Gonzalez, E. Thyer, S. Wheeler, Z. Yamin, M. Yuan, and L. W. Simmons. 2015. Evolutionary change in testes tissue composition among experimental populations of house mice. Evolution. 69:848–55. Firman, R. C., and L. W. Simmons. 2011. Experimental evolution of sperm competitiveness in a mammal. BMC Evol. Biol. 11:19. Firman, R. C., and L. W. Simmons. 2010. Experimental evolution of sperm quality via postcopulatory sexual selection in house mice. Evolution. 64:1245–1256.

137 Firman, R. C., and L. W. Simmons. 2008. The frequency of multiple paternity predicts variation in testes size among island populations of house mice. J. Evol. Biol. 21:1524–1533. Fitzpatrick, J. L., R. M. Kempster, T. S. Daly-Engel, S. P. Collin, and J. P. Evans. 2012. Assessing the potential for post-copulatory sexual selection in elasmobranchs. J. Fish Biol. 80:1141–1158. Gage, M. J. ., and E. H. Morrow. 2003. Experimental Evidence for the Evolution of Numerous, Tiny Sperm via Sperm Competition. Curr. Biol. 13:754–757. Hosken, D. J., and P. I. Ward. 2001. Experimental evidence for testis size evolution via sperm competition. Ecol. Lett. 4:10–13. Johnson, L., and W. B. Neaves. 1981. Age-related changes in the Leydig cell population, seminiferous tubules, and sperm production in stallions. Biol. Reprod. 24:703–712. López, M., M. G. Muñoz, and E. A. Herrera. 2008. Reproductive morphology of capybaras, Hydrochoerus hydrochaeris (Rodentia: Hystricognathi): No evidence for sperm competition? Mamm. Biol. 73:241-244 Luetjens, C. M., G. F. Weinbauer, and J. Wistuba. 2005. Primate spermatogenesis: New insights into comparative testicular organisation, spermatogenic efficiency and endocrine control. Biol. Rev. Cam. Philos. Soc. 80:475-88 Lüpold, S., G. M. Linz, J. W. Rivers, D. F. Westneat, and T. R. Birkhead. 2009. Sperm competition selects beyond relative testes size in birds. Evolution. 63:391–402. Marconato, A., and D. Y. Shapiro. 1996. Sperm allocation, sperm production and fertilization rates in the bucktooth parrotfish. Anim. Behav. 52:971–980. Martin, P. A., T. J. Reimers, J. R. Lodge, and P. J. Dziuk. 1974. The effect of ratios and numbers of spermatozoa mixed from two males on proportions of offspring. J. Reprod. Fertil. 39:251–8. Møller, A. P., and J. V. Briskie. 1995. Extra-pair paternity, sperm competition and the evolution of testis size in birds. Behav. Ecol. Sociobiol. 36:357–365. Montoto, L. G., L. Arregui, N. M. Sanchez, M. Gomendio, and E. R. S. Roldan. 2012. Postnatal testicular development in mouse species with different levels of sperm competition. Reproduction 143:333–346. Musick, J. A., and J. K. Ellis. 2005. Reproductive Evolution of Chondrichthyans. Pp.

138 45–79 in W. C. Hamlett, ed. Reproductive Biology and Phylogeny of Chondrichthyes: Sharks, Batoids and Chimaeras. Science Publishers, Inc., Plymouth, UK. Pagel, M. 1999. Inferring the historical patterns of biological evolution. Nature 401:877–884. Parker, G. A. 1970. Sperm competition and it’s evolutionary consequences in the insects. Biol. Rev. 45:525–567. Parker, G. A., and M. A. Ball. 2005. Sperm competition, mating rate and the evolution of testis and ejaculate sizes: a population model. Biol. Lett. 1:235–8. Parker, G. a, and T. Pizzari. 2010. Sperm competition and ejaculate economics. Biol. Rev. Camb. Philos. Soc. 85:897–934. Pitnick, S., G. T. Miller, J. Reagan, and B. Holland. 2001. Males’ evolutionary responses to experimental removal of sexual selection. Proc. R. Soc. B Biol. Sci. 268:1071–1080. Pratt, H. L. 1988. Elasmobranch gonad structure: a description and survey. Copeia 1988:719–729. R Core Team. 2017. R: A language and environment for statistical computing. R Foundation for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Ramm, S. A., G. A. Parker, and P. Stockley. 2005. Sperm competition and the evolution of male reproductive anatomy in rodents. Proc. R. Soc. London B 272:949–55. Ramm, S. A., and L. Schärer. 2014. The evolutionary ecology of testicular function: size isn’t everything. Biol. Rev. 89:874–888. Ramm, S. A., L. Schärer, J. Ehmcke, and J. Wistuba. 2014. Sperm competition and the evolution of spermatogenesis. Mol. Hum. Reprod. 20:1169 -1179. Ramm, S. A., and P. Stockley. 2010. Sperm competition and sperm length influence the rate of mammalian spermatogenesis. Biol. Lett. 6:219–21. Rasband, W. . 1997. ImageJ,. U. S. National Institutes of Health, Bethesda, Maryland, USA. Revell, L. J. 2012. phytools: An R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol. 3:217–223.

139 Rowe, M., and S. Pruett-Jones. 2011. Sperm Competition Selects for Sperm Quantity and Quality in the Australian Maluridae. PLoS One 6:e15720. Russell, L. D., H. P. Ren, I. S. Hikim, W. Schulze, and A. P. S. Hikim. 1990. A comparative study in twelve mammalian species of volume densities, volumes, and numerical densities of selected testis components, emphasizing those related to the sertoli cell. Am. J. Anat. 188:21–30. Saglam, I. K., D. A. Roff, and D. J. Fairbairn. 2008. Male sand crickets trade-off flight capability for reproductive potential. J. Evol. Biol. 21:997–1004. Schärer, L., P. Ladurner, and R. Rieger. 2004. Bigger testes do work more: experimental evidence that testis size reflects testicular cell proliferation activity in the marine invertebrate, the free-living flatworm Macrostomum sp. Behav. Ecol. Sociobiol. 56:420–425. Schlatt, S., and J. Ehmcke. 2014. Regulation of spermatogenesis: An evolutionary biologist’s perspective. Semin. Cell. Dev. Biol. 29:2-16. Simmons, L. W., and J. L. Fitzpatrick. 2016. Sperm competition and the coevolution of pre- and postcopulatory traits: Weapons evolve faster than testes among onthophagine dung beetles. Evolution. 70:998–1008. Simmons, L. W., and J. L. Fitzpatrick. 2012. Sperm wars and the evolution of male fertility. Reproduction 144:519–34. Simmons, L. W., and F. García-González. 2008. Evolutionary reduction in testes size and competitive fertilization success in response to the experimental removal of sexual selection in dung beetles. Evolution. 62:2580–2591. Slater, G. J., L. J. Harmon, and M. E. Alfaro. 2012. Integrating fossils with molecular phylogenies improves inference of trait evolution. Evolution. 66:3931–3944. Soulsbury, C. D. 2010. Genetic Patterns of Paternity and Testes Size in Mammals. PLoS One 5:e9581. Stein, R. W., C. G. Mull, T. S. Kuhn, N. C. Aschliman, L. N. K. Davidson, J. B. Joy, G. J. Smith, N. K. Dulvy, and A. O. Mooers. 2018. Global priorities for conserving the evolutionary history of sharks, rays and chimaeras. Nat. Ecol. Evol. 2:288–298. Stockley, P., M. J. Gage, G. A. Parker, and A. P. Møller. 1997. Sperm competition in fishes: the evolution of testis size and ejaculate characteristics. Am. Nat. 149:933–954.

140 Supriya, K., M. Rowe, T. Laskemoen, D. Mohan, T. D. Price, and J. T. Lifjeld. 2016. Early diversification of sperm size in the evolutionary history of the old world leaf warblers (Phylloscopidae). J. Evol. Biol. 29:777–789. Tung Ho, L. si, and C. Ané. 2014. A linear-time algorithm for gaussian and non- gaussian trait evolution models. Syst. Biol. 63:397–408. Vahed, K., D. J. Parker, and J. D. J. Gilbert. 2010. Larger testes are associated with a higher level of polyandry, but a smaller ejaculate volume, across bushcricket species (Tettigoniidae). Biol. Lett., doi: 10.1098/rspb.2008.0122. Wistuba, J., A. Schrod, B. Greve, J. K. Hodges, H. Aslam, G. F. Weinbauer, and C. M. Luetjens. 2003. Organization of Seminiferous Epithelium in Primates: Relationship to Spermatogenic Efficiency, Phylogeny, and Mating System1. Biol. Reprod. 69:582–591. Wright, J., and I. Cuthill. 1989. Manipulation of sex differences in parental care. Behav. Ecol. Sociobiol. 25:171–181. Yano, K., F. Sato, and T. Takahashi. 1999. Observations of mating behavior of the manta ray, Manta birostris, at the Ogasawara Islands, Japan. Ichthyol. Res. 46:289–296.

141 Chapter 3: Supporting Information

Supporting Information Table S3.1

Testes mass (g), body mass (g), mean percentage of sperm-producing tissue within the testes and maximum seminiferous follicle cross-sectional area for 18 shark and ray species.

Species Testes Body N % Seminiferous Maximum mass mass tissue follicle area Aetobatus_narinari - 40000a 1 85.98 31303.3 Carcharhinus_melanopterus 56.59b 7055.77b 3 86.50 102669.5 Carcharhinus_plumbeus 322.1 49935.83c 1 78.98 37041 Chiloscyllium_punctatum - 10900d 3 86.24 88493.9 Dasyatis_americana - 27500e 1 94.91 51464.1 Galeus_melastomus 6.87 495 1 83.12 88655.5 Ginglymostoma_cirratum - 77150f 1 68.76 47866.8 Hemiscyllium_ocellatum - 600g 1 82.50 56904.6 Heterodontus_japonicus - - 2 75.33 69884.1 Mustelus_asterias 25.49g 2418.78h 12 86.37 57791.8 Mustelus_mustelus 43.41 5275.45 3 84.62 59120 Raja_asterias 27.67 2008 1 77.97 51629.5 Raja_clavata 34.80i 2950.62i 4 80.89 93860.8 Raja_microcellata - - 2 71.43 89335.3 Raja_undulata 41.26 3500 5 83.76 93295 Scyliorhinus_canicula 12.65 286.34 7 83.46 87199.1 Scyliorhinus_stellaris 75.70 2870 3 89.21 110689.3 Urobatis_jamaicensis 19.09j 421.06j 1 82.65 36249.5

a Body mass data from Swider et al. 2017 bTestes and body mass data from Lyle et al. 1987 c Body mass data obtained from Cliff et al 1988 using Graphclick. Testes mass calculated from peak GSI value (July) in Baremore & Hale 2012 d Body mass data from Harahush et al. 2007 e Body mass data from Myagkov 1991 f Body mass from Castro 2000, using body length obtained from Graphclick and length-weight equation supplied g Body mass data from Yopak et al. 2007 h Testes and body mass data from Edward Farrell, personal communication

142 i Testes from Saglam et al 2012, body mass calculated from length-weight relationship given in Filiz and Bilge 2004 jTestes and body mass from Daniel Fahy, personal communication

Supporting Information References

Baremore, I. E., and L. F. Hale. 2012. Reproduction of the Sandbar Shark in the Western North Atlantic Ocean and Gulf of Mexico. Mar. Coast. Fish. 4:560– 572. Castro, J. I. 2000. The biology of the nurse shark, Ginglymostoma cirratum, off the Florida east coast and the Bahama Islands. Environ. Biol. Fishes 58:1–22. Cliff, G., S. F. J. Dudley, and B. Davis. 1988. Sharks caught in the protective gill nets off Natal, South Africa. 1. The sandbar shark Carcharhinus plumbeus (Nardo). South African J. Mar. Sci. 7:255–265. Filiz, H., and G. Bilge. 2004. Length-weight relationships of 24 fish species from the North Aegean Sea, Turkey. J. Appl. Ichthyol. 20:431–432. Harahush, B. K., A. B. P. Fischer, and S. P. Collin. 2007. Captive breeding and embryonic development of Chiloscyllium punctatum Muller & Henle, 1838 (Elasmobranchii: Hemiscyllidae). J. Fish Biol. 71:1007–1022. Lyle, J. M. 1987. Observations on the biology of Carcharhinus cautus (Whitley), C. melanopterus (Quoy & Gaimard) and C. fitzroyensis (Whitley) from Northern Australia. Mar. Freshw. Res. 38:701–710. Myagkov, N. A. 1991. The brain sizes of living elasmobranchii as their organization level indicator. I. General analysis. J. Hirnforsch. 32:553–561. Saglam, H., and O. Ak. 2012. Reproductive biology of Raja clavata (Elasmobranchii: Rajidae) from Southern Black Sea coast around Turkey. Helgol. Mar. Res. 66:117–126. Swider, D. A., A. L. Corwin, T. Y. Kamerman, S. L. Zimmerman, G. C. Violetta, J. Davis, and M. Janse. 2017. Reproduction of spotted eagle rays, Aetobatus narinari, in

143 aquaria. Pp. 422–442 in The Elasmobranch Husbandry Manual II: Recent Advances in the Care of Sharks, Rays and their Relatives. Yopak, K. E., T. J. Lisney, S. P. Collin, and J. C. Montgomery. 2007. Variation in brain organization and cerebellar foliaton in chondrichthyans: sharks and holocephalans. Brain Behav. Evol. 69:280–300.

144 Chapter 4: Sexual selection drives divergent patterns of selection on sperm flagellum length in sharks and rays

Status: Manuscript, in review at Evolution

Authors: Amy Rowley*, Lisa Locatello*, Ariel Kahrl, Mariana Rego, Annika Boussard, Eduardo Garza-Gisholt, Ryan M. Kempster, Shaun P. Collin, Eva Giacomello, Maria C. Follesa, Cristina Porcu, Jonathan P. Evans, Fabio Hazin, Francisco Garcia-Gonzalez, Toby S. Daly-Engel, Carlotta Mazzoldi and John L. Fitzpatrick

* Shared first authorship

Author contributions:

Conceptualization: AR, LL, JPE, FGG, CM, JLF

Data collection: AR, LL, MR, AB, EGG, RMK, SPC, EG, MCF, CP, FH, FGG, TDE, JLF

Data analysis: AR, LL, AK, JLF

Writing – original draft: AR, JLF

Writing – review and editing: All authors

Funding acquisition: LL, FGG, CM, JLF

145 4. Sexual selection drives divergent patterns of selection on sperm flagellum length in sharks and rays

Amy Rowley1,2*, Lisa Locatello3*, Ariel Kahrl2, Mariana Rego4, Annika Boussard2, Eduardo Garza-Gisholt5,6, Ryan M. Kempster5,6, Shaun P. Collin5,6, Eva Giacomello7, Maria C. Follesa8, Cristina Porcu8, Jonathan P. Evans9, Fabio Hazin4, Francisco Garcia-Gonzalez9,10, Toby S. Daly-Engel11, Carlotta Mazzoldi3 and John L. Fitzpatrick2

Affiliation 1 Faculty of Biology, Medicine and Health, University of Manchester, UK 2 Department of Zoology, Stockholm University, Stockholm, Sweden 3 Department of Biology, University of Padova, Via U. Bassi 58/B, 35131 Padua, Italy 4 Laboratório de Histologia Animal, Departamento de Morfologia e Fisiologia Animal, Universidade Federal Rural de Pernambuco, Recife, Brazil 5 Oceans Graduate School, The University of Western Australia, Crawley, WA 6009, Australia 6 The UWA Oceans Institute, The University of Western Australia, Crawley, WA 6009, Australia 7 MARE – Marine and Environmental Sciences Centre, IMAR- Institute of the Sea, OKEANOS Centre- University of the Azores, Horta, Portugal

8 Department of Life and Environmental Sciences, University of Cagliari, Cagliari, Italy

9 Centre for Evolutionary Biology, School of Biological Sciences, University of Western Australia, Crawley, WA 6009, Australia

10 Estacion Biologica de Doñana-CSIC, Sevilla, Spain

146 11 Department of Ocean Engineering and Marine Sciences, Florida Institute of Technology, Melbourne, FL, USA

* Shared first authorship

4.1 Abstract

Postcopulatory sexual selection is a powerful selective force commonly credited with driving the extraordinary diversification of sperm morphology observed among species. In particular, sperm from multiple males often compete for fertilisations (i.e. sperm competition), which is hypothesized to generate selection on sperm morphology, sperm production efficiency, and increased rates of evolutionary diversification in sperm form. However, contradictory empirical results and limited taxonomic scope has led to debate about how sperm competition influences the evolution of sperm morphology. Here, we use phylogenetically controlled analyses to explore the evolutionary diversification and variance in sperm morphology traits in two basal vertebrate groups, sharks and rays (elasmobranchs). Our analyses reveal divergent patterns of selection on sperm flagellum length in these groups. In sharks, species experiencing greater levels of sperm competition produce sperm with longer and less variable flagella, while in rays, sperm flagellum length decreased and head and midpiece variance increased with the level of sperm competition. Across elasmobranchs, sperm flagellum length exhibited elevated rates of evolutionary diversification compared to sperm head and midpiece length. Our findings suggest that the flagellum is an important target of sexual selection in elasmobranchs and provide insight into patterns of selection on the ejaculate in basal vertebrate lineages.

Keywords: Sperm competition, sperm morphology, evolutionary rate, elasmobranch

147 4.2 Introduction

Sperm exhibit extraordinary morphological diversity and are among the most variable of all known cell types (Pitnick et al. 2009a). However, the evolutionary processes that promote sperm diversification remain hotly debated (Lüpold and Pitnick 2018). Although fertilization environments and phylogenetic effects undoubtedly influence sperm evolution (Pitnick et al. 2009a; Simpson et al. 2014; Lüpold and Pitnick 2018), sexual selection is now recognized as a particularly powerful selective force driving the evolution of sperm morphological diversity (Pizzari and Parker 2009; Simmons and Fitzpatrick 2012; Fitzpatrick and Lüpold 2014). When females mate with multiple males, the temporal and spatial overlap of sperm from rival males within the fertilisation environment can result in sperm competition, where sperm from different males compete to fertilise the available ova (Parker 1970), and cryptic female choice, where females bias the outcome of sperm competition in favour of preferred males (Eberhard 1996). These episodes of postcopulatory sexual selection therefore impose strong selective pressures on males to produce more competitive ejaculates. When male fertility is influenced by the number of sperm present at the site of fertilization, sperm competition is expected to favour increases in the number of sperm that males produce (Parker 1998; Pizzari and Parker 2009). Indeed, evolutionary increases in sperm number in response to sperm competition are commonly observed across species (Chapter 2; Simmons and Fitzpatrick 2012). However, if trade-offs exist between sperm number and size (Parker 1982), postcopulatory sexual selection for increased sperm production may result in evolutionary reductions in sperm size (Immler et al. 2011). Consequently, the evolutionary processes that shape selection on sperm morphology and drive its evolutionary diversification appear numerous and complex.

Each component of the sperm cell (i.e. the head, midpiece, and flagellum) is responsible for different aspects of sperm function, and therefore sperm competition may act on each, or all, of these components provided they influence male fertilisation success (Gage et al. 2004; Simmons and Fitzpatrick 2012). For

148 example, the size of the sperm midpiece is expected to increase in response to sperm competition to provide the cell with greater energy (Anderson et al. 2005), and can influence the beat frequency of the flagellum (Cardullo and Baltz 1991), which may in turn be targeted by selection to increase thrust (Gomendio and Roldan 1991; Fitzpatrick et al. 2009). Sperm competition is also hypothesized to influence sperm head size, which is expected to create drag that opposes the thrusting force of the flagellum (Humphries et al. 2008) and thus can be an important determinant of swimming speed (e.g Malo et al. 2006). Despite substantial research attention, comparative studies evaluating how sperm morphological traits respond to variation in the strength of sperm competition show mixed, and often taxon-specific, results (Immler and Birkhead 2007; Simmons and Fitzpatrick 2012). Moreover, recent research suggests that sperm competition acts not only on individual sperm components but also on the efficiency of sperm production as a whole (Birkhead et al. 2005). When sperm competition risk (i.e. probability that an individual’s sperm will compete with those from another male; Parker et al. 1997) is high, selection should favour consistent production of an optimal sperm phenotype, and correspondingly act to erode variation in sperm morphology within the ejaculate (Parker 1993; Birkhead et al. 2005). Indeed, the evidence to date from phylogenetically controlled studies has revealed consistent negative relationships between intraspecific variation in sperm morphology and the level of sperm competition, although such relationships have been evaluated in only a handful of studies of passerine birds and social insects (Calhim et al. 2007; Immler et al. 2008; Kleven et al. 2008; Fitzpatrick and Baer 2011). However, the way in which selection acts on sperm morphology and sperm production across a broader taxonomic scale remains unclear.

Sexually-selected traits are generally expected to exhibit rapid rates of phenotypic evolution (Arnqvist 1998; Gonzalez-Voyer and Kolm 2011; Fitzpatrick et al. 2012a). Consequently, we should expect strong signatures of sexual selection on ejaculate traits across species experiencing varying levels of sperm competition. Yet, few studies have considered how selection drives phenotypic divergence in sperm morphology at the macro-evolutionary scale (Simmons and Fitzpatrick 2016), and it

149 remains unclear whether sperm components exhibit distinct rates of evolutionary divergence or evolve as a functionally integrated unit. If genetic correlations between head, midpiece and flagellum size are generally weak, then sperm components are free to evolve independently (Simmons and Moore 2009). Indeed, among passerine birds, sperm components show different rates of diversification, with the sperm head evolving at a different rate than the midpiece and flagellum (Immler et al. 2012; Rowe et al. 2015). In contrast, strong genetic correlations among sperm components can result in sperm components responding as a functionally integrated trait (Moore et al. 2013). Thus, comparing rates of evolution of different sperm components is a crucial step in providing insight into overall patterns of how selection acts on sperm.

Here we take an integrative approach to consider how sperm competition shapes sperm morphology in sharks and rays (elasmobranchs), two ancestral vertebrate groups with divergent body plans, behaviours, life histories, and reproductive traits that can influence how selection acts on ejaculates (Wourms 1977; Pratt and Carrier 2001). Observations of matings in the wild suggest that female sharks and rays commonly mate with more than one male within a reproductive cycle (Yano et al. 1999; Chapman et al. 2003), but multiply-sired litters occur at vastly different frequencies across a wide variety of species, ranging from infrequent (e.g. 11% multiple paternity in the shortspine spurdog, Squalus cf. mitsukurii) to frequent (e.g. 92% multiple paternity in the small-spotted catshark, Scyliorhinus canicula) (Chapter 2.; Byrne and Avise 2012; Fitzpatrick et al. 2012b). Moreover, sharks are one of few taxonomic groups in which body size corrected testes mass has been validated against genetic estimates of sperm competition risk (the percentage of litters sired by more than one male) and intensity (the mean number of sires per litter) (Chapter 2), thereby allowing broad comparative studies to be conducted using validated proxy measures for the level of sperm competition. In addition, female sharks and rays retain sperm in specialised storage organs (i.e. oviducal glands), allowing offspring to be produced using stored sperm for up to 4 years after mating in some species (Bernal et al. 2015). The potential for long-term sperm storage uncouples mating from fertilization, increasing competition between rival

150 ejaculates, and imposes selection on sperm morphology to enter and remain viable in sperm storage organs (Fitzpatrick et al. 2012; Orr and Zuk 2014; Orr and Brennan 2015). We take advantage of the considerable variation in sperm competition risk and intensity observed among sharks and rays to consider how male competitive dynamics shape sperm morphology, variance and diversification in these two basal vertebrate groups. Specifically, we contrast patterns of selection in sharks and rays by taking the three-step approach of i) examining the relationship between the size of each sperm component (head, midpiece and flagellum) and the level of sperm competition, ii) considering how sperm competition acts on within-male variation in sperm component size and iii) comparing rates of evolutionary divergence between sperm components. We hypothesized that sperm components would be longer and less variable in species experiencing a higher risk of sperm competition.

4.3 Methods

4.3.1 Sample collection

Sperm was collected opportunistically from 141 individuals representing 34 shark and ray species (25 sharks, 9 rays, mean number of males per species = 4.15, range = 1 – 19, Supporting Information Table S4.1). Only individuals caught previously as part of commercial and artisanal fisheries, during scientific surveys, for research purposes, or as bycatch were sampled. Samples were collected at 13 field sites spanning 10 countries across five continents over a 12-year period (Supporting Information Table S4.1). We assessed males for maturity by examining the claspers, with fully calcified claspers indicating a sexually mature individual (Hamlett 2005). Immature males were not sampled. Semen (sperm and seminal fluid) was extracted from mature males in breeding condition (defined as those males currently producing sperm) by manually applying pressure to the sperm sac or claspers (Figure 4.1a). Whenever possible we measured the total length (mm) and body mass (g) of each individual prior to dissection and the testes of mature males were excised and weighed to the nearest 0.01g. However, we were unable to collect

151 body and testes mass data directly from all individuals sampled in the field due to logistical constraints (e.g. when sharks or rays were too large to be weighed or could not be dissected because they were being sold at market). In cases where we lacked body or testes mass data, we searched the literature using the species name in combination with the words ‘testes mass’, ‘body mass’ or ‘gonadosomatic index’ (GSI). If raw data were not reported in the studies examined, we contacted the authors directly to request the data, or calculated body and testes mass from GSI, using the program GraphClick v3.0.3 (Boyle et al. 2012) to extract data from figures (summarised in Table S4.1). In this way, we supplemented the data we collected in the field with data on male body mass and testes mass from an additional two and four species, respectively (Table S4.1). In addition, we searched the literature for cases where data on sperm morphology, body mass and testes mass were collected from the same population by a single research group to supplement our field- collected data. This search uncovered one species that could be added to the dataset, bringing the total sample size to 35 shark and ray species (Table S4.1).

4.3.2 Sperm analysis

Semen samples were either examined fresh or preserved in 1 ml of 10% neutral buffered formalin for subsequent examination. Fresh ejaculates were processed at field sites within a few hours of extraction, while preserved samples were taken back to the laboratory where they were examined. Sperm component length did not differ between fresh and preserved samples (paired t-test performed on six species where sperm was preserved using both methods: head+midpiece: t = - 0.431, P = 0.684, d.f.=5; flagellum: t = -0.739, P = 0.493, d.f.=5). Microscope slides were loaded with 10 l of seawater-diluted (for fresh samples) or formalin-diluted (for preserved samples) semen samples and covered with a coverslip. Slides were left for up to two hours after loading to allow sperm to settle onto a single plane of focus prior to viewing under the microscope. For each male sampled, we haphazardly selected and photographed between 20-30 individual morphologically

152 normal sperm cells. All sperm images were captured at 400X magnification (Figure 4.1b).

The number of field sites, number of years of data collection, and the logistical constraints of sampling elasmobranch fishes introduced some differences in how images were captured. Images of sperm taken for later use in standardized downstream analyses (see below) were collected using different microscope and camera systems depending on field sites and sampling conditions. Specifically, when collecting semen samples at field stations, sperm images were captured using three different microscope and camera systems, which differed based on sampling locations (Adriatic Sea: a Leica DMLB30 light microscope fitted with a Leica DFC 420 camera (n = 3 species); Sardinian Sea: a Zeiss Axioskop light microscope fitted with a Canon EOS 1100D camera (n = 8 species); Azores: a Leica DM 6000B light microscope fitted with a Leica DFC340 camera (n = 6 species)). When microscopes and cameras were not available during sampling (e.g. in remote field locations or on boats), we preserved sperm in the field, then performed subsequent analyses in the laboratory using a Leica DM750 light microscope fitted with a Canon 600D camera (n = 24 species; note that six species were sampled in multiple locations and thus our total sample of field-collected samples remains at 35 species). Importantly, while differences in microscope and camera systems may add noise to the overall dataset, there is no a priori reason to assume that the variation in sampling will systematically bias the results in favour of the hypotheses being tested, but rather will attenuate regression coefficients toward zero (Hansen and Bartoszek 2012; Hansen 2016). Moreover, because all downstream analyses on field-collected sperm images were performed using standardized methods (see below) our final dataset likely has less error than would be introduced from analyses that collect data from the literature, which is a common practice in comparative analyses in general, and in comparative analyses of sperm evolution in particular e.g.Gage and Freckleton 2003; Gomendio et al. 2011; Lüpold and Fitzpatrick 2015).

Sperm components were measured from digital images using the segmented line tool in ImageJ (Rasband 1997). Mean sperm component lengths (m) were

153 calculated for each species. The division between the sperm head and midpiece was difficult to distinguish in some species. Therefore, we examined the combined head and midpiece length and the flagellum length for all species in our dataset. For 20 species where the head and midpiece could be clearly distinguished, sperm components were assessed separately.

4.3.3 Phylogenetic Linear Models

We used phylogenetically controlled general least squares (PGLS) multiple regressions to examine associations between sperm morphology and sperm competition risk. Phylogenetic relationships were derived from a recent elasmobranch phylogeny constructed using genetic data from 610 species (Stein et al. 2018). Using the original set of 500 phylogenetic trees, we generated a consensus tree using the function consensus.tree with the function consensus.edges to set branch lengths in the package phytools (Revell 2012). To assess phylogenetic dependence of the data, likelihood ratio tests were used to calculate the phylogenetic scaling parameter λ (Pagel 1999; Freckleton et al. 2002), where a value of 0 indicates no phylogenetic signal, and 1 indicates total phylogenetic dependence. All data were log10 transformed prior to analysis in R version 3.4.1 (R Core Team 2017). Thirty-two of the 35 elasmobranch species in our dataset were present in the phylogeny (Figure 4.1c), and of these 32 species testes and male body mass data were available for 25 species (18 sharks and 7 rays).

To investigate evolutionary responses in sperm morphology to sperm competition in sharks and rays, we examined the relationship between each sperm component (the combined length of the head and midpiece, and the length of the flagellum) and body size-corrected testes mass. Body size corrected testes mass was used as a proxy for sperm competition risk, due to the close association between body size- corrected testes mass and the level of sperm competition across a wide range of taxa (Simmons and Fitzpatrick 2012). Specifically, body size corrected testes mass is correlated positively with multiple paternity rates and the number of males siring

154 offspring in a brood among shark species (Chapter 2), supporting the assertion that body size corrected testes mass represents a valid estimate of the sperm competition risk in our analyses. Initial analyses across all elasmobranchs revealed stark differences in the associations between sperm traits and sperm competition risk between sharks and rays (Table S4.4). Such taxon-specific responses may arise due to differences in reproductive biology, life history, and ecology between taxa (Hamlett 2005), which may influence how selection acts on the ejaculate (sensu Immler and Birkhead 2007) . Therefore, we conducted separate analyses for sharks and rays. For the subset of shark species (n = 16) for which the head and midpiece could be distinguished, we also considered the effect of relative testes size on each component separately. For rays, we lacked a sufficient number of species (n = 4) where the head and midpiece could be clearly differentiated and therefore did not examine these sperm components separately. To control for the allometric relationship between testes mass and body size, body mass was included as a covariate in all models.

We calculated the mean within-male coefficient of variation (CV) for both head and midpiece and flagellum length using the formula CV = (standard deviation/mean)*100. We did not evaluate between-male CV, as only six of the 35 species we examined had 10 individuals sampled per species. We used PGLS regressions to test the relationship between the within-male CV of each sperm component and body size corrected testes mass. Although CV is commonly used to assess standardized variation in sperm morphology (e.g. Malo et al. 2006; Calhim et al. 2007; Immler et al. 2008; Kleven et al. 2008), the use of CV as an estimate of variation has been criticized for potentially yielding biased results in the absence of an isometric relationship between the mean and variance (Tomkins and Simmons 2002; Fitzpatrick and Baer 2011). Therefore, we performed additional analyses to examine the association between the standard deviation of sperm component length and body size corrected testes mass while accounting for mean-variance relationships by adding the mean sperm component length as a covariate in the model (Fitzpatrick and Baer 2011). All results remained consistent when we assessed variance by inspecting the response of the standard deviation of sperm

155 component length to body size-corrected testes mass while controlling for mean sperm component length (Table S4.2).

4.3.4 Comparing Evolutionary Rates

To examine and compare rates of phenotypic divergence of sperm component length, we used a recently developed likelihood approach to directly contrast the evolutionary Brownian rate parameter (2) among multiple traits evolving on the same phylogeny (Adams 2013). This analysis focused on the 23 shark and 9 ray species for which sperm component lengths were measured that were present in the phylogeny. A key assumption of Adams’s (Adams 2013) likelihood approach is that traits evolve under a Brownian motion process. To validate this assumption, we first compared the fit of three different models of character evolution (Brownian motion, Ornstein-Uhlenbeck, Early-burst) for the sperm head and midpiece length and sperm flagellum length for sharks and rays separately using the fitContinuous function in the package geiger (Harmon et al. 2008; Simmons and Fitzpatrick 2016). Based on AICc model comparisons, Brownian motion was the best model of character evolution for all traits (Supporting Information Table S3), which fits the assumptions of the models used in the subsequent evolutionary rate analysis. We then applied Adams’s (Adams 2013) likelihood method to compare rates of phenotypic divergence among traits when assuming trait covariance among sperm components. This method estimates the observed rate of evolution for each

2 trait independently ( obs) and then compares the likelihood value from the observed model to a common model in which all traits are constrained to have a

2 common rate of evolution ( common) using likelihood-ratio tests. Because dimensionality can impact trait variance (Adams 2013), we ln-transformed all data prior to analysis. We then estimated the rates of evolution of species means for the combined sperm head and midpiece, and the sperm flagellum length. Rates models assumed evolutionary covariance in the observed evolutionary rate matrix among sperm traits (Adams 2013; Simmons and Fitzpatrick 2016). Comparisons of evolutionary rates can also incorporate within-species measurement error (i.e.

156 sampling error around the mean) in the models. However, incorporating among- individual measurement error requires sampling of several different individuals for each species, which was not the case for the majority of species in our dataset (less than one fifth of the species we examined had 10 individuals sampled per species). Therefore, we did not incorporate within-species measurement error into our analyses. All models converged under the Nelder-Mead optimization function assuming covariance between the traits (Adams 2013).

4.4 Results

4.4.1 Sperm competition and sperm morphology

Sperm morphology was variable across sharks and rays (Figure 4.1). However, the association between sperm morphology and the level of sperm competition exhibited striking differences between sharks and rays. In sharks, flagellum length was significantly positively associated with body size-corrected testes mass (Figure 4.2, Table 4.1), while in rays, the association between flagellum length and body size-corrected testes mass was negative (Figure 4.2, Table 4.1). Similarly, flagellum length was negatively associated with body mass in sharks, suggesting that larger- bodied shark species produce sperm with smaller flagella, but there was a trend suggesting a positive relationship in rays (Table 4.1). Sperm head and midpiece length was not associated with body size-corrected testes mass in either sharks or rays (Figure 4.2, Table 4.1). When we examined the head and midpiece separately, we found no relationship between either sperm head length or midpiece length and body size-corrected testes mass (sharks: sperm head length: λ = 1.00, df = 13, testes mass: slope = 0.04  0.07, t = 0.53, P = 0.60 , r = 0.16, lower CI = -1.45, upper CI = 2.49; body mass: slope = 0.002  0.06, t = 0.03, P = 0.98, r = 0.01, lower CI = - 1.93, upper CI = 1.99; sperm midpiece length: λ = 1.00, df = 13, testes mass: slope = 0.12  0.11, t = 1.10, P = 0.29, r = 0.29, lower CI = -0.92, upper CI = 3.08; body mass: slope = -0.08  0.09, t = -0.86, P = 0.40 , r = -0.23, lower CI = -2.83, upper CI = 1.14;

157 note that there were insufficient ray species for which sperm head and midpiece could be measured separately to facilitate this analysis).

In sharks, there was a significant negative association between within-male CV of sperm flagellum length and body size-corrected testes mass, but no such association in rays (Figure 4.2, Table 4.1). Rays exhibited a positive association between within-male CV of sperm head and midpiece length and body size- corrected testes mass, while no association was detected in sharks (Figure 4.2, Table 4.1).

158

(a) (b) Midpiece

Head Flagellum

Etmopterus spinax (c) Etmopterus princeps Etmopterus pusillus Centroscyllium fabricii Centroscymnus coelolepis Squalus megalops Squalus mitsukurii Squalus acanthias Cirrhigaleus asper Centrophorus granulosus Centrophorus squamosus Deania calcea Hexanchus griseus Carcharhinus cautus Carcharhinus tilstoni Carcharhinus acronotus Rhizoprionodon acutus Mustelus antarcticus Galeus melastomus Scyliorhinus canicula Scyliorhinus stellaris Isurus oxyrinchus Raja miraletus Raja asterias Raja undulata Dipturus oxyrinchus Leucoraja melitensis Taeniura lymma Neotrygon kuhlii Dasyatis pastinaca Himantura signifer Rhinochimaera pacifica

0 20 40 60 80 0 20 40 60 80 100 120 140 160 Head and midpiece length (µm) Flagellum length (µm)

Figure 4.1. Sperm collection and variation in sperm morphology across sharks and rays. (a) Semen being stripped from the clasper of a male marbled torpedo ray, Torpedo marmorata. (b) Sperm cell from a nervous shark, Carcharhinus cautus, with the head stained with DAPI, with sperm components labelled (image by Rodolfo Jaffe). (c) Phylogenetic relationships between species in our dataset, with bars representing sperm head and midpiece length, and sperm flagellum length (µm). Sperm was collected from one chimaera species, Rhinochimaera pacifica, which was included in the figure but not in the analyses presented in the main text.

159

Figure 4.2. Associations between sperm morphological traits (plotted on a log-scale) and body size corrected testes mass, a proxy measure for sperm competition risk, in sharks (black points) and rays (white points). Body size-corrected testes mass represents residual values from a linear regression of testis mass on body mass for the shark and ray species present in the analysis. Data are from the association between (a) sperm head and midpiece length (b) the within-male coefficient of variation (CV) of head and midpiece length (rays: log10(y) = 0.61 + 0.42x), (c) flagellum length (sharks: log10(y) = 1.99 + 0.19x; rays: log10(y) = 1.99 – 0.38x) and (d) within-male CV of flagellum length and body size-corrected testes mass (sharks: log10(y) = 0.66 – 0.45x).

Table 4.1. Phylogenetically controlled generalized least squares (PGLS) regressions between sperm traits and testes mass. Models assess the relationship between sperm length and variance and testes mass in (a) sharks and (b) rays. Body mass was included as a covariate in all models to control for the allometric relationship between body and testes size. The phylogenetic scaling parameter λ indicates the level of phylogenetic dependence of the data, ranging from 0 (low phylogenetic signal) to 1 (high phylogenetic signal). The degrees of freedom (df), slope of the regression with standard error, t-statistic (t), and p-value are presented for each model, with significant results (p<0.05) highlighted in bold

161 Table 4.1.

Sperm trait Predictors λ df Slope SE t p r Lower CI Upper CI (a) Sharks (Superorder: Selachimorpha, n = 18) Head and midpiece length Testes mass 1.00 15 0.07 0.07 0.94 0.36 0.24 -1.06 2.91 Body mass -0.05 0.06 -0.94 0.36 -0.24 -2.91 1.06 Flagellum length Testes mass 0.00 15 0.18 0.07 2.49 0.02 0.54 0.31 4.61 Body mass -0.17 0.06 -2.86 0.01 -0.59 -5.03 -0.62 Within-male head and midpiece length CV Testes mass 1.00 15 -0.19 0.18 -1.09 0.29 -0.27 -3.07 0.93 Body mass 0.13 0.14 0.90 0.38 0.23 -1.10 2.87 Within-male flagellum length CV Testes mass 0.00 15 -0.50 0.17 -2.92 0.01 -0.60 -5.10 -0.67 Body mass 0.29 0.14 2.06 0.06 0.47 -0.06 4.12

(b) Rays (Superorder: Batoidea, n = 7) Head and midpiece length Testes mass 0.00 4 0.11 0.11 0.99 0.38 0.44 -1.13 3.01 Body mass -0.13 0.13 -0.95 0.40 -0.43 -2.96 1.16 Flagellum length Testes mass 0.00 4 -0.47 0.16 -2.92 0.04 -0.83 -5.61 -0.08 Body mass 0.45 0.19 2.36 0.08 0.76 -0.24 4.80 Within-male head and midpiece length CV Testes mass 1.00 3 0.87 0.20 4.21 0.02 0.92 0.43 7.90 Body mass -1.07 0.22 -4.78 0.02 -0.94 -8.86 -0.64 Within-male flagellum length CV Testes mass 0.00 3 0.02 0.37 0.06 0.96 0.03 -1.91 2.02 Body mass 0.09 0.42 0.21 0.85 0.12 -1.77 2.16

4.4.2 Rates of evolution of sperm components

Sperm flagellum length evolved 1.58 and 3.23 times as fast as sperm head and midpiece length in sharks and rays, respectively. However, despite these elevated rates of diversification, likelihood ratio tests revealed that rates of phenotypic divergence of sperm head and midpiece length and sperm flagellum length did not differ in either sharks or rays when assessed separately. In sharks, sperm components evolved at a single common evolutionary rate (n = 23, LRT = 1.20, P =

2 -4 0.27, AICobserved = -37.98, AICcommon = -38.78, head and midpiece:  obs = 1.47 x 10 ,

2 -4 flagellum:  obs = 2.33 x 10 ). In rays, there was a non-significant trend suggesting that sperm flagellum evolved faster than sperm head and midpiece length (n = 9,

2 LRT = 3.24, P = 0.07, AICobserved = 0.07, AICcommon = 1.30, head and midpiece:  obs =

-4 2 -4 1.83 x 10 , flagellum:  obs = 5.92 x 10 ). Since the PGLS models indicated that sperm flagellum length, but not sperm head and midpiece length, responded in opposing directions to increases in the level of sperm competition in sharks and rays, we performed an additional analysis considering sperm evolution across all elasmobranchs in our sample. When sharks and rays were assessed together in a model, flagellum length evolved 2.11 times faster than sperm head and midpiece length, a difference that was statistically significant (n = 32, LRT = 4.48, P = 0.03,

2 -4 AICobserved = -40.09, AICcommon = -37.61, head and midpiece:  obs = 1.59 x 10 ,

2 -4 flagellum:  obs = 3.36 x 10 , Figure 3).

(a) (b) (c)

) 2

 *

Brownianrate parameter (

Figure 4.3. Rates of phenotypic evolution of sperm components in (a) sharks and (b) rays and (c) all elasmobranch species in our dataset. Estimates of the observed rate of phenotypic divergence (2) of combined head and midpiece length and flagellum length are shown with bars representing 95% confidence intervals. Note, rate comparisons can only be made within each model. Flagellum length showed a significantly higher rate of divergence than head and midpiece length across all species (c), as indicated with an asterisk (*), but not in either sharks (a) or rays (b) individually.

4.5 Discussion

Our results show striking differences in how postcopulatory sexual selection acts to shape the evolution of sperm morphology between sharks and rays. In sharks, we found that species experiencing higher levels of sperm competition produce ejaculates with longer and less variable sperm flagella, while in rays, flagellum length decreased at higher levels of sperm competition. Sperm head and midpiece length was not associated with sperm competition risk in either sharks or rays. These results suggest that flagellum length is targeted by selection in sharks and rays, albeit in opposite directions, to increase fertilization success during sperm

164 competition. The opposing patterns of selection on flagellum length in sharks and rays were manifest in the elevated rate of phenotypic divergence in flagellum length compared to head and midpiece length across elasmobranchs. Taken in combination, these results provide compelling evidence for distinct patterns of selection on different sperm components between sharks and rays and suggest that the flagellum is an important target of sexual selection in elasmobranchs. More broadly, our findings contribute to the growing appreciation that ejaculate traits can exhibit divergent evolutionary responses to sperm competition in different taxonomic groups (Immler and Birkhead 2007b). As we discuss below, sharks and rays display both intra- and inter-taxonomic variation in key reproductive traits that could explain the different responses of sperm traits to selection imposed by sperm competition risk.

This positive association between flagellum length and the level of sperm competition observed in sharks supports the theoretical prediction that longer sperm should be favoured in sperm competition (Gomendio and Roldan 1991). Longer flagella can be advantageous during competitive fertilizations under three possible mechanistic scenarios, none of which is mutually exclusive. First, longer flagella may provide greater thrusting force to propel sperm more quickly as they swim towards the egg (Fitzpatrick et al. 2009; Lüpold et al. 2009), particularly as sperm swimming speed is an important predictor of competitive fertilization success in a wide range of taxa (Simmons and Fitzpatrick 2012). However, finding a link between sperm swimming speed and sperm morphology is notoriously challenging, and among internally-fertilizing species the evidence supporting a mechanistic link between sperm form and function is weak at the intraspecific level (Lüpold et al. 2009; Mossman et al. 2009; Simpson et al. 2014). The relationship between sperm morphology and speed has not yet been examined in sharks, although such an assessment would help shed light on sperm evolution in these fishes. Second, sperm with longer flagella may be better able to displace rival sperm from advantageous positions within the female’s reproductive tract, for example by being better positioned to fertilise the egg or enter sperm storage organs (Lüpold et al. 2012). In sharks, sperm are commonly retained in specialised tubules in the

165 oviducal gland after mating, although both the structural organisation of the oviducal gland and the length of storage vary among species (Pratt 1993; Hamlett 2005; Marongiu et al. 2015). It is therefore plausible that sperm storage may impose specific selective pressures on sperm morphology, for example if sperm with longer flagella are able to reach – and fill – the storage site more quickly or are better able to displace rival sperm from the storage tubules. Interestingly, Pratt (Pratt 1993) suggested that Lamniform sharks have short-term sperm storage (lasting only days), and the only Lamniform in our dataset (Isurus oxyrinchus) had one of the shortest sperm flagellum lengths recorded among sharks. While this observation hints at a functional relationship between sperm storage duration and selection on sperm length, the duration of sperm storage is not well characterized for most shark species (or indeed in most species; Orr and Brennan 2015). Clearly, further work assessing the interaction between sperm storage duration and sperm competition risk will help to illuminate how selection acts on flagellum length in sharks. Finally, longer sperm may be selected for if cryptic female choice favours the use and storage of sperm with longer flagella (Miller and Pitnick 2002; Baer et al. 2003) and such choice co-varies positively with sperm competition risk. Distinguishing between these alternative mechanistic possibilities explaining evolutionary responses in flagellum length in sharks remains challenging at present.

In rays, sperm flagella length was negatively associated with the estimated level of sperm competition. A classic prediction from sperm competition theory is that selection will favour smaller sperm in the face of sperm competition, provided that sperm number predicts competitive fertilization success and that sperm size and number trade-off against one another (Parker 1993). Under this scenario, sperm number may be favoured during competitive mating in rays, resulting in indirect selection against longer sperm flagella. But why would sperm number be selected for over sperm size in rays? Systematic differences in the level of sperm competition or reproductive mode (oviparity, ovoviviparity vs. viviparity) between sharks and rays could influence how selection acts on ejaculates (Zeh and Zeh 2000). However, this appears to be an unlikely explanation for the differences between these two groups because sharks and rays exhibit similar variation in the

166 level of sperm competition among species (based on body size corrected testes mass, see Figure 4.2) and both groups show a broad range of reproductive modes (Musick and Ellis 2005). Instead, differences in reproductive physiology may be driving differences in how selection acts on ejaculate traits. Rays have claspers (i.e. male genitalia) that are approximately twice as long, relative to body size, as sharks (Rowley A. unpublished data), suggesting that rays can deposit sperm farther into the female’s reproductive tract. However, whether reduced distance between ejaculated sperm and the unfertilised egg would favour the evolution of smaller, more numerous sperm remains unclear. There is also some evidence to suggest that sperm storage duration may be relatively short (i.e. a few days) in some stingray species of the genus Dasyatis (Maruska et al. 1996). Notably, the Dasyatis species in our dataset (D. pastinaca) was an outlier in two of the reproductive traits we considered: D. pastinaca has the smallest flagellum and one of the largest body size-corrected testes mass values of any species in our dataset and drove the negative relationship between flagellum length and sperm competition risk we observed in rays (see Figure 4.2). When this species was removed, there was no association between flagellum length and body size-corrected testes mass in rays (λ = 1.00, df = 3, testes mass: slope = -0.18  0.14, t = -1.34, P = 0.27, r = -0.61, lower CI = -3.47, upper CI = 0.95; body mass: slope = 0.18  0.15, t = 1.23, P = 0.31, r = 0.58, lower CI = -1.02, upper CI = 3.33) though it is unclear whether this represents an outlier effect or reflects the reduced sample size. Shorter sperm storage durations may reduce the potential for sperm-female coevolutionary dynamics to drive the evolution of longer sperm, particularly if the shorter period of time between insemination and fertilization favour the production of smaller, more numerous sperm. However, the opposite pattern has been observed in bats, where males invest more in sperm number when fertilization is delayed (Orr and Zuk 2013). Disentangling these potential hypotheses requires more information on the female reproductive anatomy, genital fit during copulation (e.g. Orbach et al. 2017) , and sperm-female interactions during storage in both sharks and rays.

Variation in sperm component length also showed divergent responses to sperm competition risk between sharks and rays. In sharks, within-male variance in sperm

167 flagellum length – but not sperm head and midpiece length – is reduced in species that experience higher levels of sperm competition. Such variance reduction in response to increases in sperm competition suggests that polyandrous mating selects for increased ‘quality control’ in male sperm production (Hunter and Birkhead 2002; Birkhead et al. 2005) and supports the pattern of decreasing variation in sperm morphology in response to sexual selection previously documented in social insects (Fitzpatrick and Baer 2011) and passerine birds (Calhim et al. 2007; Immler et al. 2008; Kleven et al. 2008). Thus, postcopulatory sexual selection appears to exert both directional (resulting in longer flagellum lengths) and stabilizing selection (resulting in reduced variance in flagellum length) on the sperm flagellum in sharks. Cryptic female choice, specifically at the site of sperm storage, may exert stabilizing selection for optimal sperm length, particularly given the long-term sperm storage observed in some sharks species (Hamlett 2005; Bernal et al. 2015). If female sperm storage organs preferentially retain specific sperm morphologies, competition among sperm for access to the oviducal gland will likely drive the evolution of less-variable sperm (Fitzpatrick and Baer 2011). It is well known that female storage organs impose selective pressures on sperm morphology (Briskie et al. 1997; Pitnick et al. 1999; Pattarini et al. 2006; García- González and Simmons 2007), and variation in total sperm length is negatively related to the duration of sperm storage in passerine birds (Kleven et al. 2009). In contrast with sharks, the level of sperm competition in rays did not influence variation in sperm flagellum length, and surprisingly sperm head and midpiece length exhibited greater variation as the level of sperm competition increased in this group. As argued above, the shorter duration of sperm storage in rays compared to sharks may explain the inconsistent patterns observed between these groups. Nevertheless, a commonality among the taxa where sperm variance is negatively related with sperm competition risk (i.e. sharks, social insects and passerine birds) is that females retain sperm for prolonged periods (i.e. several weeks or years) in specialised storage organs after copulation (Birkhead et al. 1990; Baer 2005; Storrie et al. 2008; den Boer et al. 2009). Thus, our findings suggest that sperm-female interactions, specifically mediated by sperm storage organs, may represent a convergent mechanism underpinning reductions in sperm variation

168 across phylogenetically distinct taxa. Future studies comparing the relationships between the level of sperm competition and sperm variance across species both with and without female sperm storage organs would represent an important test of this hypothesis.

Consistent with the idea that sperm components exhibit distinct rates of phenotypic divergence (Immler et al. 2012; Rowe et al. 2015), we found that evolutionary diversification of sperm flagellum length was approximately twice as fast as the head and midpiece length across elasmobranchs. However, we did not detect distinct rates of evolutionary diversification of sperm components when we compared sharks and rays separately. This inconsistency may reflect the different evolutionary responses in sperm flagellum length to sperm competition risk we observed between sharks and rays. Put simply, as sperm competition increases in elasmobranchs, sperm flagellum length either increases (in sharks) or decreases (in rays), which leads to increased measures of phenotypic diversification in sperm flagellum length compared to sperm head and midpiece length when assessed across elasmobranchs. Critically, the increased evolutionary rate of diversification observed in sperm flagellum length reflects both evolutionary increases (in sharks) and decreases (in rays) in flagellum length in response to sperm competition risk. Rapid phenotypic diversification is characteristic of reproductive traits, and sexually-selected characters have been shown to evolve relatively quickly (Price and Whalen 2009; Fitzpatrick et al. 2012a; Seddon et al. 2013). Therefore, the varying rates of diversification among different sperm components across elasmobranchs suggest that the flagellum evolves under stronger directional selection than the head and midpiece in the sharks and rays and may therefore play a more important role in determining male reproductive success. Alternatively, the head and midpiece may be subject to stronger functional constraints on length than the flagellum, thus impeding diversification. For example, the head is required to interact closely with the female reproductive tract during storage (Pitnick et al. 2009b) and with the ovum during fertilization (Karr et al. 2009), critical functions that may constrain morphology by imposing strong stabilising selection on head size. Rapid phenotypic divergence of traits under sexual selection has often been

169 invoked to explain the enormous diversity of sperm morphology across species, and our results suggest that sexual selection likely contributes to the diversification in flagellum length in sharks and rays.

In conclusion, our findings in an ancestral vertebrate lineage suggest that sperm competition arising from female polyandry targets sperm length and variance in strikingly divergent patterns in sharks and rays and drives the rapid diversification of sperm flagellum length in this group. Sharks and rays represent a useful and novel model for studying the evolution of reproductive traits, due to their wide range of reproductive systems and behaviours, and unique position as the first vertebrates to develop internal fertilization. Our efforts to evaluate the implications of this reproductive variation on the selective pressures influencing sperm evolution in this group represent a first step towards understanding sperm evolution in these groups. Further research will contribute to the evaluation of the generality of the patterns of reproductive trait evolution observed in other species. While we found intriguing evidence that selective patterns differ among shark and ray species, further study incorporating a wider range of ray species is needed to further substantiate lineage-specific effects. As sharks and rays are internal fertilisers, it is very likely that selection on sperm is imposed not only by male-male competition to fertilise ova, but also by interactions between sperm and the female reproductive physiology. Further phylogenetically controlled analyses to test for evolutionary relationships between the female sperm storage organ and patterns of sperm morphology in sharks and rays would be a logical first step towards testing this possibility. Future work considering female effects on ejaculate evolution will aid in moving towards a more comprehensive understanding of how postcopulatory sexual selection operates in sharks and rays.

170 4.6 Acknowledgements

We thank Clinton Duffy and Malcolm Francis for contributing unpublished data, the Lonja Pesquera de Algeciras (and in particular Trudy, Vicente, Rosa and Curro) for facilitating access to fish caught as bycatch, Samantha Hook, Lou Ruddell, Phil Root, Emilio Riginella, Vittoria Correale and Licia Finotto for helping in data collection, and Niclas Kolm for helpful comments on the manuscript. Thanks to staff at the NOAA Southeast Fisheries Science Center’s Pelagic Longline Observer Program for samples from the northern Gulf of Mexico, and the University of West Florida’s Jim Hammond for logistical support. Research was supported by a Biotechnology and Biological Sciences Research Council (BBSRC) Doctoral Training Program studentship to AR, Manchester University and Stockholm University (AR and JLF) and a Knut and Alice Wallenberg Academy Fellowship to JLF. CM was financially supported by Clodia project funded by the Veneto Region (Italy) Law 15/2007 (DGR no 4069). LL was supported by Erasmus+ mobility program 2017, University of Padova (Italy). EG acknowledges Fundação para a Ciência e Tecnologia (FCT), through the strategic project UID/MAR/04292/2013 granted to MARE. FGG was supported by grants (CGL2012-34685 and CGL2016-76173-P, co-funded by the European Regional Development Fund) from the Spanish Ministry of Economy. Funding for the project was also provided by the Western Australia Government, the University of Western Australia, the Australian Research Council, and a Doñana Biological Station project financed by the Spanish Ministry of Economy, through the Severo Ochoa Program for Centres of Excellence (SEV-2012-0262).

171 4.7 References

Adams, D. C. 2013. Comparing evolutionary rates for different phenotypic traits on a phylogeny using likelihood. Syst. Biol. 62:181–192. Anderson, M. J., J. Nyholt, and A. F. Dixson. 2005. Sperm competition and the evolution of sperm midpiece volume in mammals. J. Zool. 267:135–142. Arnqvist, G. 1998. Comparative evidence for the evolution of genitalia by sexual selection. Nature 393:784–786. Baer, B. 2005. Sexual selection in Apis bees. Apidologie 36:187–200. Baer, B., P. Schmid-Hempel, J. T. Høeg, and J. J. Boomsma. 2003. Sperm length, sperm storage and mating system characteristics in bumblebees. Insectes Soc. 50:101–108. Bernal, M. A., N. L. Sinai, C. Rocha, M. R. Gaither, F. Dunker, and L. A. Rocha. 2015. Long-term sperm storage in the brownbanded bamboo shark Chiloscyllium punctatum. J. Fish Biol. 86:1171–1176. Birkhead, T. R., E. J. Pellat, P. Brekke, R. Yeates, and H. Castillo-Juarez. 2005. Genetic effects on sperm design in the zebra finch. Nature 434:383–387. Birkhead, T. R., J. E. Pellatt, and F. M. Hunter. 1990. Numbers and Distribution of Sperm in the Uterovaginal Sperm Storage Tubules of the Zebra Finch. Condor 92:508. Boyle, M. A., A. L. Samaha, A. M. Rodewald, and A. N. Hoffmann. 2012. Evaluation of the reliability and validity of GraphClick as a data extraction program. Briskie, J. V., R. Montgomerie, and T. R. Birkhead. 1997. The Evolution of Sperm Size in Birds. Evolution. 51:937. Byrne, R. J., and J. C. Avise. 2012. Genetic mating system of the brown smoothhound shark (Mustelus henlei), including a literature review of multiple paternity in other elasmobranch species. Mar. Biol. 159:749-756. Calhim, S., S. Immler, and T. R. Birkhead. 2007. Postcopulatory sexual selection is associated with reduced variation in sperm morphology. PLoS One 2:e413. Cardullo, R. A., and J. M. Baltz. 1991. Metabolic regulation in mammalian sperm: Mitochondrial volume determines sperm length and flagellar beat frequency. Cell Motil. Cytoskeleton 19:180–188.

172 Chapman, D. D., M. J. Corcoran, G. M. Harvey, S. Malan, and M. S. Shivji. 2003. Mating behavior of southern stingrays, Dasyatis americana (Dasyatidae). Environ. Biol. Fishes 68:241–245. den Boer, S. P. A., J. J. Boomsma, and B. Baer. 2009. Honey bee males and queens use glandular secretions to enhance sperm viability before and after storage. J. Insect Physiol. 55:538–543. Eberhard, W. G. 1996. Female Control: Sexual Selection by Cryptic Female Choice. Princeton University Press, Princeton. Fitzpatrick, J. L., M. Almbro, A. Gonzalez-Voyer, N. Kolm, and L. W. Simmons. 2012a. Male Contest Competition And The Coevolution Of Weaponry And Testes In Pinnipeds. Evolution. 66:3595–3604. Fitzpatrick, J. L., and B. Baer. 2011. Polyandry reduces sperm length variation in social insects. Evolution. 65:3006–3012. Fitzpatrick, J. L., R. M. Kempster, T. S. Daly-Engel, S. P. Collin, and J. P. Evans. 2012. Assessing the potential for post-copulatory sexual selection in elasmobranchs. J. Fish Biol. 80:1141–1158. Fitzpatrick, J. L., and S. Lüpold. 2014. Sexual selection and the evolution of sperm quality. Mol. Hum. Reprod. 20:1180-9 Fitzpatrick, J. L., R. Montgomerie, J. K. Desjardins, K. a Stiver, N. Kolm, and S. Balshine. 2009. Female promiscuity promotes the evolution of faster sperm in cichlid fishes. Proc. Natl. Acad. Sci. U. S. A. 106:1128–32. Freckleton, R. . P., P. H. Harvey, and M. Pagel. 2002. Phylogenetic analysis and comparative data: A test and review of evidence. Am. Nat. 160:712–726. Gage, M. J. G., and R. P. Freckleton. 2003. Relative testis size and sperm morphometry across mammals: No evidence for an association between sperm competition and sperm length. Proc. R. Soc. B Biol. Sci. 270:625–632. Gage, M. J. G., C. P. Macfarlane, S. Yeates, R. G. Ward, J. B. Searle, and G. A. Parker. 2004. Spermatozoal Traits and Sperm Competition in Atlantic Salmon. Curr. Biol. 14:44–47. García-González, F., and L. W. Simmons. 2007. Shorter sperm confer higher competitive fertilization success. Evolution. 61:816–824. Gomendio, M., and E. R. Roldan. 1991. Sperm competition influences sperm size in

173 mammals. Proc. R. Soc. London B 243:181–5. Gomendio, M., M. Tourmente, and E. R. S. Roldan. 2011. Why mammalian lineages respond differently to sexual selection: Metabolic rate constrains the evolution of sperm size. Proc. R. Soc. B Biol. Sci. 278:3135–3141. Gonzalez-Voyer, A., and N. Kolm. 2011. Rates of phenotypic evolution of ecological characters and sexual traits during the Tanganyikan cichlid adaptive radiation. J. Evol. Biol. 24:2378–2388. Hamlett, W. C. 2005. Reproductive biology and phylogeny of Chondrichthyes : sharks, batoids and chimaeras. Science Publishers. Hansen, T. F. 2016. On bias and precision in meta-analysis: the error in the error. J. Evol. Biol. 29: 1919-1921. Hansen, T. F., and K. Bartoszek. 2012. Interpreting the evolutionary regression: The interplay between observational and biological errors in phylogenetic comparative studies. Syst. Biol. 61:413–425. Harmon, L. J., J. T. Weir, C. D. Brock, R. E. Glor, and W. Challenger. 2008. GEIGER: Investigating evolutionary radiations. Bioinformatics 24:129–131. Humphries, S., J. P. Evans, and L. W. Simmons. 2008. Sperm competition: linking form to function. BMC Evol. Biol. 8:319. Hunter, F. M., and T. R. Birkhead. 2002. Sperm viability and sperm competition in insects. Curr. Biol. 12:121–123. Immler, S., and T. R. Birkhead. 2007. Sperm competition and sperm midpiece size: no consistent pattern in passerine birds. Proc. R. Soc. London B 274:561–568. Immler, S., S. Calhim, and T. R. Birkhead. 2008. Increased postcopulatory sexual selection reduces the intramale variation in sperm design. Evolution. 62:1538– 1543. Immler, S., A. Gonzalez-Voyer, and T. R. Birkhead. 2012. Distinct evolutionary patterns of morphometric sperm traits in passerine birds. Proc. R. Soc. B Biol. Sci. 279:4174–4182. Immler, S., S. Pitnick, G. A. Parker, K. L. Durrant, S. Lüpold, S. Calhim, and T. R. Birkhead. 2011. Resolving variation in the reproductive tradeoff between sperm size and number. Proc. Natl. Acad. Sci. 108:5325–5330. Karr, T. L., W. J. Swanson, and R. R. Snook. 2009. The evolutionary significance of

174 variation in sperm-egg interactions. Pp. 305–365 in Sperm Biology. Elsevier. Kleven, O., F. Fossøy, T. Laskemoen, R. J. Robertson, G. Rudolfsen, and J. T. Lifjeld. 2009. Comparative evidence for the evolution of sperm swimming speed by sperm competition and female sperm storage duration in passerine birds. Evolution. 63:2466–2473. Kleven, O., T. Laskemoen, F. Fossøy, R. J. Robertson, and J. T. Lifjeld. 2008. Intraspecific variation in sperm length is negatively related to sperm competition in passerine birds. Evolution. 62:494–499. Lüpold, S., S. Calhim, S. Immler, and T. R. Birkhead. 2009. Sperm morphology and sperm velocity in passerine birds. Proc. Biol. Sci. 276:1175–81. Lüpold, S., and J. L. Fitzpatrick. 2015. Sperm number trumps sperm size in mammalian ejaculate evolution. Proc. R. Soc. B 282:20152122. Lüpold, S., M. K. Manier, K. S. Berben, K. J. Smith, B. D. Daley, S. H. Buckley, J. M. Belote, and S. Pitnick. 2012. How multivariate ejaculate traits determine competitive fertilization success in Drosophila melanogaster. Curr. Biol. 22:1667–1672. Lüpold, S., and S. Pitnick. 2018. Sperm form and function: What do we know about the role of sexual selection? Reproduction. 155:R229-R243 Malo, A. F., M. Gomendio, J. Garde, B. Lang-Lenton, A. J. Soler, and E. R. S. Roldan. 2006. Sperm design and sperm function. Biol. Lett. 2:246–249. Marongiu, M. F., C. Porcu, A. Bellodi, D. Cuccu, A. Mulas, and M. C. Follesa. 2015. Oviducal gland microstructure of Raja miraletus and Dipturus oxyrinchus (Elasmobranchii, Rajidae). J. Morphol. 276:1392–1403. Maruska, K. P., E. G. Cowie, and T. C. Tricas. 1996. Periodic gonadal activity and protracted mating in elasmobranch fishes. J. Exp. Zool. 276:219–232. Miller, G. T., and S. Pitnick. 2002. Sperm-female coevolution in Drosophila. Science. 298:1230–1233. Moore, A. J., L. D. Bacigalupe, and R. R. Snook. 2013. Integrated and independent evolution of heteromorphic sperm types. Proc. R. Soc. B Biol. Sci. 280:20131647. Mossman, J., J. Slate, S. Humphries, and T. Birkhead. 2009. Sperm morphology and velocity are genetically codetermined in the zebra finch. Evolution. 63:2730–

175 2737. Musick, J. A., and J. K. Ellis. 2005. Reproductive Evolution of Chondrichthyans. Pp. 45–79 in W. C. Hamlett, ed. Reproductive Biology and Phylogeny of Chondrichthyes: Sharks, Batoids and Chimaeras. Science Publishers, Inc., Plymouth, UK. Orbach, D. N., D. A. Kelly, M. Solano, and P. L. R. Brennan. 2017. Genital interactions during simulated copulation among marine mammals. Proc. R. Soc. B Biol. Sci. 284:20171265. Orr, T. J., and P. L. R. Brennan. 2015. Sperm storage: Distinguishing selective processes and evaluating criteria. Trends Ecol. Evol. 30:261-272. Orr, T. J., and M. Zuk. 2013. Does delayed fertilization facilitate sperm competition in bats? Behav. Ecol. Sociobiol. 67:1903–1913. Orr, T. J., and M. Zuk. 2014. Reproductive delays in mammals: An unexplored avenue for post-copulatory sexual selection. Biol. Rev. 89:889–912. Pagel, M. 1999. Inferring the historical patterns of biological evolution. Nature 401:877–884. Parker, G. A. 1970. Sperm competition and it’s evolutionary consequences in the insects. Biol. Rev. 45:525–567. Parker, G. A. 1998. Sperm Competition and the Evolution of Ejaculates: Towards a Theory Base. Pp. 3–54 in Sperm competition and Sexual selection. Academic Press. Parker, G. A. 1993. Sperm competition games: Sperm size and sperm number under adult control. Proc. R. Soc. B Biol. Sci. 253:245–254. Parker, G. A. 1982. Why are there so many tiny sperm ? Sperm competition and the maintenance of two sexes. J. Theor. Biol. 96:281–294. Parker, G. A., M. A. Ball, P. Stockley, and M. J. Gage. 1997. Sperm competition games: a prospective analysis of risk assessment. Proceedings. Biol. Sci. 264:1793–802. Pattarini, J. M., W. T. Starmer, A. Bjork, and S. Pitnick. 2006. Mechanisms underlying the sperm quality advantage in Drosophila melanogaster. Evolution. 60:2064– 2080. Pitnick, S., D. J. Hosken, and T. R. Birkhead. 2009a. Sperm morphological diversity.

176 Pp. 69–149 in Sperm Biology. Elsevier. Pitnick, S., T. Markow, and G. S. Spicer. 1999. Evolution of Multiple Kinds of Female Sperm-Storage Organs in Drosophila. Evolution. 53:1804. Pitnick, S., M. F. Wolfner, and S. S. Suarez. 2009b. Ejaculate-female and sperm- female interactions. Pp. 247–304 in Sperm Biology. Elsevier. Pizzari, T., and G. A. Parker. 2009. Sperm competition and sperm phenotype. Pp. 207–245 in Sperm Biology. Elsevier. Pratt, H. L. 1993. The storage of spermatozoa in the oviducal glands of western North Atlantic sharks. Environ. Biol. Fishes 14:139–149. Pratt, H. L., and J. C. Carrier. 2001. A review of elasmobranch reproductive behavior with a case study on the nurse shark , Ginglymostoma cirratum. Environ. Biol. Fishes 60:157–188. Price, J. J., and L. M. Whalen. 2009. Plumage evolution in the oropendolas and caciques: Different divergence rates in polygynous and monogamous taxa. Evolution. 63:2985–2998. R Core Team. 2017. R: A language and environment for statistical computing. R Foundation for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Rasband, W. . 1997. ImageJ,. U. S. National Institutes of Health, Bethesda, Maryland, USA. Revell, L. J. 2012. phytools: An R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol. 3:217–223. Rowe, M., T. Albrecht, E. R. A. Cramer, A. Johnsen, T. Laskemoen, J. T. Weir, and J. T. Lifjeld. 2015. Postcopulatory sexual selection is associated with accelerated evolution of sperm morphology. Evolution. 69:1044–1052. Seddon, N., C. A. Botero, J. A. Tobias, P. O. Dunn, H. E. A. MacGregor, D. R. Rubenstein, J. A. C. Uy, J. T. Weir, L. A. Whittingham, and R. J. Safran. 2013. Sexual selection accelerates signal evolution during speciation in birds. Proc. R. Soc. B Biol. Sci. 280:20131065. Simmons, L. W., and J. L. Fitzpatrick. 2016. Sperm competition and the coevolution of pre- and postcopulatory traits: Weapons evolve faster than testes among onthophagine dung beetles. Evolution. 70:998–1008.

177 Simmons, L. W., and J. L. Fitzpatrick. 2012. Sperm wars and the evolution of male fertility. Reproduction 144:519–34. Simmons, L. W., and A. J. Moore. 2009. Evolutionary quantitative genetics of sperm. Pp. 405–434 in Sperm Biology. Elsevier. Simpson, J. L., S. Humphries, J. P. Evans, L. W. Simmons, and J. L. Fitzpatrick. 2014. Relationships between sperm length and speed differ among three internally and three externally fertilizing species. Evolution. 68:92–104. Stein, R. W., C. G. Mull, T. S. Kuhn, N. C. Aschliman, L. N. K. Davidson, J. B. Joy, G. J. Smith, N. K. Dulvy, and A. O. Mooers. 2018. Global priorities for conserving the evolutionary history of sharks, rays and chimaeras. Nat. Ecol. Evol. 2:288–298. Storrie, M. T., T. I. Walker, L. J. Laurenson, and W. C. Hamlett. 2008. Microscopic organization of the sperm storage tubules in the oviducal gland of the female gummy shark (Mustelus antarcticus), with observations on sperm distribution and storage. J. Morphol. 269:1308–1324. Tomkins, J. L., and L. W. Simmons. 2002. Measuring relative investment: a case study of testes investment in species with alternative male reproductive tactics. Anim. Behav. 63:1009–1016. Wourms, J. P. 1977. Reproduction and development in chondrichthyan fishes. Integr. Comp. Biol. 17:379–410. Yano, K., F. Sato, and T. Takahashi. 1999. Observations of mating behavior of the manta ray, Manta birostris, at the Ogasawara Islands, Japan. Ichthyol. Res. 46:289–296. Zeh, D. W., and J. A. Zeh. 2000. Reproductive mode and speciation: the viviparity- driven conflict hypothesis. BioEssays 22:938–946.

178 Chapter 4: Supporting Information

Supporting Informaton Table S4.1. Mean, standard deviation (SD) and coefficient of variation (CV) of sperm component length, number of males from which sperm samples were collected (n), and (where applicable) body mass and testes mass data collected for (a) sharks, (b) rays, and (c) chimera species. Note that the chimera species was sampled but not included in analyses. Fish from Brazil, Morocco, Australia, Canada, Spain and the UK were previously caught for commercial sale or as bycatch. Fish from the northern Adriatic Sea (offshore Chioggia and Ancona, Italy) were collected from local fishermen for commercial sale or as bycatch. Specimens from Sardinia (Italy) were collected during scientific surveys of the projects MEDITS (International Bottom Trawl Survey in the Mediterranean) (http://www.sibm.it/SITO%20MEDITS/principalemedits.htm). Specimens from the Azores (Portugal) were collected during longline scientific surveys conducted between 150-1400 m depth around Condor seamount, southwest of Faial island (details on survey methodology and Condor seamount available at Giacomello et al. 2013; Menezes & Giacomello 2013).

Species Sampling Body mass Testes Head + Head + Flagellum Flagellum Head + Flagellum Sperm Location (g) mass (g) midpiece midpiece length (m) SD (m) midpiece CV n length (m) SD (m) CV

(a) Sharks Carcharhinus acronotus Brazili 6979.65 41.87 38.95 2.77 97.44 7.31 7.26 7.41 2

Carcharhinus cautus Australiai 5100 53.2 40.06 2.84 100.96 3.90 7.11 3.87 2

Carcharhinus tilstoni Australiai 3950 35.5 38.31 2.82 102.99 6.19 7.37 6.01 1

Centrophorus granulosus Moroccoi 2287.9iii 14.35iii 60.02 3.00 87.68 5.54 5.00 6.26 2

Centrophorus squamosus Azoresii/ 7653.85 30.62iv 63.33 2.69 86.11 4.31 4.24 5.04 10 Moroccoi Centrophorus zeehani Australiai NA 14.55 56.56 1.91 96.91 1.41 3.38 1.46 1

Centroscyllium fabricii Azoresii 1675 NA 64.99 2.09 110.08 8.14 3.22 7.38 2

Centroscymnus coelolepis Azoresii/ 6383.68 43.612v 77.00 2.48 75.53 3.65 3.25 4.93 12 Moroccoi Cirrhigaleus asper Brazili 6028.13 21.2 55.13 3.36 105.23 5.43 6.10 5.16 1

Deania calcea Azoresii 2525 NA 63.89 2.13 94.85 2.06 3.33 2.21 2

Etmopterus baxteri New Zealandi NA NA 77.34 1.86 98.90 7.07 2.40 7.18 2

Etmopterus princeps Azoresii 991 NA 79.52 2.94 116.83 6.56 3.70 5.68 8

Etmopterus pusillus Azoresii 410 NA 81.75 4.00 104.89 4.89 4.89 4.66 1

Etmopterus spinax Sardinian Seaii 129.77 2.37 77.85 3.61 110.34 5.82 4.65 5.29 2

Galeus melastomus Moroccoi/ 495 6.87 48.42 3.82 146.13 10.33 12.95 6.81 3 Sardinian Seaii

Hexanchus griseus Australiai/ 135000 1360 59.55 2.74 121.21 3.14 4.60 2.58 2 Sardinian Seaii Isurus oxyrinchus Moroccoi 97333vi 250vi 60.40 2.12 74.16 1.92 3.51 2.59 1

Mustelus antarcticus Australiai 1900 14.8 50.74 2.29 67.88 4.74 4.49 7.04 2

Rhizoprionodon acutus Australiai 2085 8.8 32.21 1.34 68.59 1.71 4.18 2.49 1

Scyliorhinus canicula UKi/Adriatic 286.34 12.65 59.49 1.43 112.04 1.86 1.78 1.90 13 Seaii Scyliorhinus stellaris Adriatic Seaii 2870 75.7 57.87 0.98 114.05 2.10 1.70 1.85 3

Squalus acanthias Adriatic 1351.20 24.39 50.26 1.69 115.63 4.72 3.43 4.09 15 Seaii/Northwest Pacific i Squalus blainville Sardinian Seaii 864.50 13.31 61.67 3.38 125.95 4.22 5.48 3.35 1

Squalus cf. mitsukurii Brazili 1314.67 7.33 53.79 2.62 107.92 10.54 4.95 10.07 14

Squalus megalopsviI Brazili 527.33 4.63 52.70 2.78 106.99 10.61 5.17 9.89 19

(b) Rays Dasyatis pastinaca Spaini 3773.33 84.47 58.90 4.76 61.50 5.12 7.89 8.87 3

Dipturus oxyrhinchus Sardinian Seaii 3859.5 23.93 44.88 5.63 104.74 7.23 2.21 6.91 2

Himantura signifer viiI Thailand 750.0 5.32 60 - 147 - - - -

Leucoraja melitensis Sardinian Seaii 251.04 2.65 50.59 3.30 93.91 4.71 6.56 4.93 2

Neotrygon kuhlii Australiai NA NA 43.81 5.97 63.85 1.58 13.62 2.48 1

Raja asterias Spaini 2008 27.67 51.31 2.23 95.24 4.25 4.34 4.46 4

Raja miraletus Sardinian Seaii 299.4 2.43 54.27 2.00 100.72 7.18 3.69 7.13 1

181 Raja polystigma Sardinian Seaii 871.40 8.74 52.51 1.90 96.25 2.45 3.62 2.55 1

Raja undulata Spaini 3500 41.26 58.94 1.81 91.40 8.71 3.08 9.63 4

Taeniura lymma Australiai NA 11.52 41.11 5.40 61.21 2.47 13.14 4.03 1

i Semen samples preserved in 10% formalin before measuring, ii Semen samples measured when fresh, iii Testes and body mass values collected from Megalofonou and Chatzispyrou 2006, iv Testes mass calculated from GSI value reported in Clarke 2001, v Testes mass reported in Clarke 2001, vi Testes and body mass from Clinton Duffy and Malcom Francis, unpublished data. vii Species referred to as S. megalops may represent a complex of similar species rather than a single wide-ranging species. As this taxonomic uncertainty remains unresolved, we continue to refer to the species from north-eastern Brazil as S. megalops, as elsewhere in the literature (e.g. Hazin et al. 2006). viiI Testes, body mass and sperm morphology data from Chatchavalvanich et al. 2005a,b. Dashes (-) indicate that data were not available.

182 Supporting Information References

Chatchavalvanich, K., Thongpan, A. and Nakai, M. 2005a. Structure of the testis and genital duct of freshwater stingray, Himantura signifer (Elasmobranchii: Myliobatiformes: Dasyatidae). Icthyol. Res. 52: 123 - 131 Chatchavalvanich, K., Thongpan, A. and Nakai, M. 2005a. Ultrastructure of spermiogenesis in a freshwater stingray, Himantura signifer. Ichthyol. Res. 52: 379 - 385 Clarke, M. W., P. L. Connolly, and J. J. Bracken. 2001. Aspects of reproduction of the deep water sharks Centroscymnus coelolepis and Centrophorus squamosus from west of Ireland and Scotland. J. Mar. Biol. Assoc. United Kingdom 81:1019–1029. Cambridge University Press. Giacomello, E., Menezes, G.M., Bergstad, O.A., 2013. An Integrated Approach for Studying Seamounts: CONDOR Observatory. Deep Sea Research Part II, vol.98 Part A, 1-6. Hazin, F. H. V., A. F. Fischer, M. K. Broadhurst, D. Veras, P. G. Oliveira, and G. H. Burgess. 2006. Notes on the reproduction of Squalus megalops off northeastern Brazil. Fish. Res. 79:251–257. Megalofonou, P., and A. Chatzispyrou. 2006. Sexual maturity and feeding of the gulper shark, Centrophorus granulosus, from the eastern Mediterranean Sea. Cybium 30:67–74. Menezes, G.M., Giacomello, E., 2013. Spatial and temporal variability of demersal fishes at Condor seamount (Northeast Atlantic). Deep Sea Research Part II, vol.98 Part A, 101-113.

Supporting Table S4.2. Phylogenetically controlled generalized least squares (PGLS) regressions between sperm variance and testes mass. Models assess the relationship between the standard deviation (SD) of sperm component length and testes mass while controlling for the mean sperm component length in (a) sharks and (b) rays. Body mass was included as a covariate in all models to control for the allometric relationship between body and testes size. The phylogenetic scaling parameter λ indicates the level of phylogenetic dependence of the data, ranging from 0 (low phylogenetic signal) to 1 (high phylogenetic signal). The degrees of freedom (df), slope of the regression with standard error, t-statistic (t), and p-value are presented for each model, with significant results (p<0.05) highlighted in bold.

184 Supporting Table S4.2.

Sperm trait Predictors λ df Slope SE t p r Lower Upper CI CI (a) Sharks (Superorder: Selachimorpha, n = 18) Within-male head and midpiece length Mean head and midpiece 0.00 14 0.35 0.37 0.93 0.37 0.24 -1.08 2.90 SD length Testes mass -0.26 0.14 -1.83 0.09 -0.44 -3.87 0.27 Body mass 0.20 0.12 1.68 0.12 0.41 -0.40 3.71 Within-male flagellum length SD Mean flagellum length 0.00 14 2.08 0.59 3.53 0.003 0.69 1.14 5.84 Testes mass -0.71 0.19 -3.66 0.003 -0.70 -5.99 -1.24 Body mass 0.50 0.17 2.94 0.011 0.62 0.66 5.14

(b) Rays (Superorder: Batoidea, n = 7) Within-male head and midpiece length Mean head and midpiece 1.00 2 -2.98 1.93 -1.54 0.26 -0.74 -3.85 0.97 SD length Testes mass -0.14 0.55 -0.26 0.82 -0.18 -2.21 1.74 Body mass 0.22 0.61 0.36 0.76 0.25 -1.67 2.31 Within-male flagellum length SD Mean flagellum length 0.00 2 -0.72 1.77 -0.41 0.72 -0.28 -2.36 1.63 Testes mass -0.66 0.75 -0.88 0.47 -0.53 -2.91 1.32 Body mass 0.73 0.75 0.97 0.43 0.57 -1.27 3.03

Supporting Table S4.3. Summary of model fits for Brownian Motion, Ornstein-Uhlenbeck, and Early-burst models of character evolution for total head and midpiece length and flagellum length (ln-transformed) for (a) sharks (n = 23) and (b) rays (n = 9) included in our evolutionary rates models. Values of 2 (Brownian rate parameter),  (rate of evolutionary change parameter), and a (selection strength parameter) are presented for Brownian Motion, Ornstein-Uhlenbeck, and Early-burst models, respectively. Models were compared using the Aikake information criterion (AICc). We report delta AICc (Di) and Akaike weights (wi) below, which indicate that Brownian Motion is the preferred model of evolution for each trait.

Brownian Motion Ornstein-Uhlenbeck Early-burst

2  lnL AICc i i  lnL AICc i i a lnL AICc i i

(a) Sharks Sperm head+midpiece length 1.47 x 10-4 14.12 -23.64 0 0.76 ~0 14.12 -20.98 2.66 0.20 0.007 12.44 -17.62 6.02 0.04

Sperm flagellum length 2.33 x 10-4 8.85 -13.10 0 0.67 ~0 8.85 -10.44 2.66 0.18 0.007 8.73 -10.19 2.91 0.16

(b) Rays Sperm head+midpiece length 1.83 x 10-4 4.36 -2.73 0 0.64 0.02 5.49 -0.17 2.56 0.18 0.04 5.49 -0.17 2.56 0.18 Sperm flagellum length 5.92 x 10-4 -0.92 7.84 0 0.82 0.007 -0.74 12.27 4.44 0.09 0.01 -0.74 12.27 4.44 0.09

186 Supporting Table S4.4. Phylogenetically controlled generalized least squares (PGLS) regressions between sperm traits and testes mass and taxonomic group (shark vs. ray). Models assess the relationship between sperm (a) head and midpiece length, (b) flagellum length, (c) within-male head and midpiece CV, and (d) within-male flagellum CV and testes mass while assessing potential differences between sharks and rays. Body mass was included as a covariate in all models to control for the allometric relationship between body and testes size. The phylogenetic scaling parameter λ indicates the level of phylogenetic dependence of the data, ranging from 0 (low phylogenetic signal) to 1 (high phylogenetic signal). The degrees of freedom (df), slope of the regression with standard error, t-statistic (t), and p-value are presented for each model, with significant results (p<0.05) highlighted in bold.

Supporting Table S4.4.

Sperm trait Predictors λ df Slope SE t p r Lower Upper CI CI (a) Head and midpiece length Testes mass 1.00 19 0.05 0.05 1.05 0.30 0.23 -0.95 3.02 Body mass -0.05 0.05 -1.07 0.30 -0.24 -3.05 0.93 Group 0.02 0.08 0.25 0.80 0.06 -1.71 2.21

(b) Flagellum length Testes mass 0.00 19 -0.47 0.17 -2.71 0.01 -0.53 -4.82 -0.54 Body mass 0.45 0.20 2.19 0.04 0.45 0.09 4.24 Group 1.21 0.47 2.59 0.02 0.51 0.44 4.68 Group:Testes 0.65 0.19 3.46 <0.01 0.62 1.18 5.67 mass Group:Body mass -0.62 0.21 -2.92 <0.01 -0.56 -5.05 -0.72

(c) Within-male head and midpiece Testes mass 1.00 18 0.87 0.36 2.45 0.02 0.50 0.31 4.54 length CV Body mass -1.07 0.39 -2.72 0.01 -0.54 -4.84 -0.54 Group -2.50 0.87 -2.87 0.01 -0.56 -5.01 -0.67 Group:Testes -1.06 0.39 -2.71 0.01 -0.54 -4.83 -0.53 mass Group:Body mass 1.20 0.42 2.88 0.01 0.56 0.68 5.02

(d) Within-male flagellum length CV Testes mass 0.00 20 -0.33 0.16 -2.07 0.05 -0.42 -4.11 0.01 Body mass 0.21 0.14 1.48 0.16 0.31 -3.47 0.55 Group -0.16 0.09 -1.73 0.10 -0.36 -3.74 0.32

189 Chapter 5: The evolution of weaponized genitals in sharks

Status: Manuscript

Authors: Amy Rowley, Lisa Locatello, Toby Daly-Engel, Francisco Garcia-Gonzalez, Ariel Kahrl, Mariana Rego, Fabio Hazin, R. Dean Grubbs, Carlotta Mazzoldi, John Fitzpatrick

Author contributions:

Conceptualization: AR, LL, FGG, CM, JLF

Data collection: AR, LL, MR, FH, FGG, AK, RDG, TDE, JLF

Data analysis: AR, JLF

Writing – original draft: AR, JLF

Writing – review and editing: AR, JLF

Funding acquisition: LL, FGG, CM, JLF

5. The evolution of weaponized genitals in sharks

Amy Rowley1,2, Lisa Locatello3, Toby Daly-Engel4, Francisco Garcia-Gonzalez5,6, Ariel Kahrl2, Mariana Rego7, Fabio Hazin7, Dean Grubbs8, Carlotta Mazzoldi3, John Fitzpatrick2

Affiliation

1 Faculty of Biology, Medicine and Health, University of Manchester, UK 2 Department of Zoology, Stockholm University, Stockholm, Sweden 3 Department of Biology, University of Padova, Via U. Bassi 58/B, 35131 Padua, Italy 4 Department of Ocean Engineering and Marine Sciences, Florida Institute of Technology, Melbourne, FL, USA 5 Centre for Evolutionary Biology, School of Biological Sciences, University of Western Australia, Crawley, WA 6009, Australia

6 Estacion Biologica de Doñana-CSIC, Sevilla, Spain

7 Laboratório de Histologia Animal, Departamento de Morfologia e Fisiologia Animal, Universidade Federal Rural de Pernambuco, Recife, Brazil

8 Florida State University Coastal and Marine Laboratory, Tallahassee, FL, USA

191 5.1 Abstract

Male genitalia exhibit high levels of morphological diversity across species, and are targeted by post-copulatory sexual selection to increase male performance in sperm competition, cryptic female choice and/or sexual conflict. In many species, male genitalia are adorned with elaborate appendages that do not function directly in sperm transfer (e.g. hooks, spines and spurs). However, relatively few studies have considered the evolution of elaborate genital appendages, and the selective forces generating the extreme variation observed in genital morphology remain a source of ongoing debate. Here we take a comparative approach to examine the evolution of genital morphology across sharks, one of the earliest vertebrate lineages to develop internal fertilisation and, consequently, specialised intromittent organs. Combining field-collected data on clasper morphology from five continents with data collected from published sources spanning over a century, we show that evolutionary transitions between the presence and absence of clasper appendages are unidirectional throughout the shark phylogeny. Specifically, only evolutionary losses of genital appendages were detected. The level of sperm competition (as measured by body-size corrected testes mass) was not associated with either clasper length or the presence of genital appendages. However, we detected a complex relationship between clasper length and the presence or absence of genital appendages. These findings suggest that clasper morphology is unlikely to have been shaped by sperm competition risk in sharks, and we discuss potential constraints and alternative hypotheses to explain the extreme variation in genital morphology in this group.

192 5.2 Introduction

Male genital morphology exhibits an extraordinary level of variation across species (Eberhard 1985; Hosken and Stockley 2004; Simmons 2014), and is widely considered to be one of the fastest-evolving phenotypic traits (Eberhard 2010). Such diversity is puzzling, given that male genitalia share the common function of delivering sperm to the fertilisation environment. While a number of hypotheses have historically invoked functional (e.g. lock and key hypothesis, Dufour 1844) or genetic (e.g. pleiotropy, Mayr 1963) mechanisms to explain the evolutionary diversification of genital morphology, sexual selection is now recognised as an important force in driving the evolution of male genitalia (Eberhard 1996; Arnqvist 1998; Hosken and Stockley 2004; Simmons 2014). Sexual selection on male genitalia can arise before mating (i.e. precopulatory) when genitalia are used in sexual displays (Langerhans et al. 2005) or to overcome female resistance to mating (Grieshop and Polak 2012). However, more commonly, genitalia are influenced by selection occurring after mating (i.e postcopulatory) through the processes of sperm competition (Parker 1970), cryptic female choice (Eberhard 1996) and/or sexual conflict (Arnqvist and Rowe 2005). These postcopulatory processes represent powerful selective agents influencing the evolutionary diversification of a wide range of genital traits, including genital complexity, size, and shape (Simmons 2014).

Male genitalia are often adorned with elaborate appendages, including hooks, barbs, spurs, and spines, that are attached to the male intromittent organ and may be subject to postcopulatory sexual selection (Von Helversen and Von Helversen 1991; Crudgington and Siva-Jothy 2000; Blanckenhorn et al. 2002; Orr and Brennan 2016). The functional roles of these structures are varied and include the removal of sperm from rival ejaculates (Waage 1979), acting as a holdfast during copulation (Edvardsson and Tregenza 2005), stimulating the female during courtship or mating (Dixson 1987; Stockley 2002), or inflicting copulatory wounding, possibly as a mechanism to discourage female remating (Hotzy and Arnqvist 2009). As all these functions should theoretically increase fertilisation success, sexual selection is

193 expected to promote the evolution of elaborate genital appendages (Orr and Brennan 2016). Indeed, studies of fish, snakes and various insect species demonstrate that males with more or longer genital spines are more successful in achieving copulations (Polak and Rashed 2010), copulate for longer durations (Friesen et al. 2014, 2016), transfer more sperm (Kwan et al. 2013), outcompete rival males during sperm competition (Nessler et al. 2007; Hotzy and Arnqvist 2009; Hotzy et al. 2012), and inflict more damage on the female reproductive tract (Ronn et al. 2007; Hotzy and Arnqvist 2009). Insect studies also reveal that populations experimentally evolving under monogamous conditions show reduced genital spine length in comparison with polygamous populations (Cayetano et al. 2011). These intraspecfic findings suggest that sexual selection routinely targets genital appendages. However, the few comparative studies that have examined the evolution of genital appendages have produced mixed results. In primates, phylogenetically controlled analyses reveal that penile spines are associated with body size corrected testes mass, a common proxy for the strength of postcopulatory sperm competition, and shorter durations of female receptivity (Stockley 2002; Orr and Brennan 2016). In contrast, genital length, which is itself subject to sexual selection in some taxa (see below), is not associated with the presence of genital spines in poecilids fishes (Langerhans 2011). Thus, the selective forces driving the evolution of genital appendages at the macroevolutionary level remain unresolved.

Male genitalia also exhibit a high level of diversity in length, both at the intra- and inter-specific level (Hosken and Stockley 2004). Pre-copulatory selection on male genital length may be imposed by female mate choice (Kahn et al. 2010; Mautz et al. 2013). After mating, longer male genitalia (relative to body size) may increase male success in competitive matings by allowing sperm to be deposited in an optimal location (Parker 1984) and/or stimulating the female more effectively (Dixson 1987) and are therefore hypothesized to be favoured by postcopulatory sexual selection. Indeed, while genital length is positively associated with mating (Evans et al. 2011) and fertilisation success in some species (Dougherty et al. 2015), this relationship is far from universal (Gasparini et al. 2011; Stockley et al. 2013;

194 Booksmythe et al. 2016). Comparative studies of the relationship between genital length and the strength of postcopulatory sexual selection have also yielded mixed results. While positive associations between genital length and proxy measures of sperm competition risk are reported in carnivores, rodents, pinnipeds, cetaceans and waterfowl (Coker et al. 2002; Ramm 2007; Fitzpatrick et al. 2012a; Dines et al. 2014), no such associations were found in bats and primates (Ramm 2007). These contradictory results indicate that patterns of selection on genital length may be taxon-specific and highlights the need to assess how sexual selection acts on genitalia across a wider taxonomic scale.

Here we examine the evolution of genital morphology in sharks, a basal vertebrate fish lineage that (together with rays and chimaerids) represent one of the earliest vertebrate groups to evolve intromittent organs and develop internal fertilization (Long et al. 2015). Sharks copulate using either one or both of their paired intromittent organs, known as claspers. Shark claspers are modified, scroll-shaped extensions of the pelvic fins that are inserted into the female’s cloaca to deposit sperm in the reproductive tract. Once inserted into the female’s cloaca, the distal tip of the clasper changes in shape by flaring open, likely acting as a holdfast to anchor the clasper in place (Pratt and Carrier 2001). In some shark species, the claspers, and particularly the terminal cartilage elements, are adorned with curved or pointed appendages (e.g. spurs, claws, spines, and hooks, Figure 5.1). When present, clasper appendages interact with the female reproductive tract during copulation, often puncturing the vaginal wall and inflicting damage to the female’s reproductive tract (Pratt and Carrier 2001). Although it is possible that these structures serve a sexually-selected purpose (Fitzpatrick et al. 2012b), there has yet to be a systematic evaluation of the evolution of clasper appendages and the selective forces that shape their evolutionary emergence and persistence across the shark phylogeny. Clasper length also varies widely across sharks, even among closely related species (Fitzpatrick et al. 2012b), although we currently lack an evolutionary explanation for this variation. To address these outstanding questions in clasper evolution, we took a three-step approach. First, we reconstructed the evolution of clasper appendages to determine the number of gains and losses of

195 appendages during the evolution of sharks. Second, we investigated whether species with clasper appendages experienced greater levels of sperm competition, using testes mass (correcting for body size) as a proxy measure of sperm competition risk that has been validated in this group (Chapter 2), and invested more in clasper length. Finally, we examined the association between clasper length and sperm competition risk. We hypothesized that the presence of elaborate genital structures and clasper length would both be associated with greater sperm competition risk in sharks.

(a)

spurs

(b)

Figure 5.1. Variation in shark clasper morphology. Clasper appendages (spines/spurs) are present in (a) Etmopterus princeps, and absent in (b) Mustelus mustelus. Photo by Lisa Locatello.

196 5.3 Methods

5.3.1 Field collected samples

We collected clasper data from 24 shark species sampled opportunistically at nine field sites in seven countries. All individuals sampled were previously caught in commercial or artisanal fisheries, during existing scientific surveys, or as bycatch. Sexual maturity of male sharks was assessed through examination of the claspers, with fully calcified (inflexible) claspers indicating a mature individual. For all sexually mature males, we measured total body length (mm), and when possible body mass (g). Body mass was not collected when logistical constraints associated with field sampling made it impossible to collect this data (e.g. in the case of large species that could not be weighed or when animals were immediately sold at market). Males were dissected and the testes excised and weighed to the nearest 0.1g. The inner clasper length was measured (mm), from the tip of the clasper to the point of insertion on the pelvic fin, and examined for the presence and number of elaborate genital structures (i.e. spurs, hooks, claws and spines, Pratt and Carrier 2001). Sample sizes vary among traits collected from field-sampled sharks as we were unable to take all measures from every individual sampled.

5.3.2 Literature collected data

We augmented our field-collected data with data from the literature to increase the phylogenetic coverage of shark species in our dataset. We searched the literature for information on the presence or number of genital appendages and data on body length and mass, and clasper length. Searches were performed using Google Scholar using species names in combination with the search terms ‘clasper morphology’, ‘clasper spur’, ‘clasper hook’, ‘clasper spine’, ‘clasper claw’, ‘rhipidion’, ‘accessory terminal cartilage’, ‘testes mass’, ‘GSI’, ‘gonad’, and ‘reproductive biology’.

197

Data on genital appendages was collected from species descriptions and from drawings, photographs and microCT scans of clasper morphology, along with accompanying descriptions. We recorded the total (minimum) number of genital appendages reported for each species. Claspers were only classified as not possessing genital appendages if studies explicitly reported the absence of these structures in their description of clasper morphology. The literature sources we used spanned more than a century, from the late 1800s to present, creating challenges with reconciling variation in terminology used to describe clasper attributes. Genital appendages have alternatively been referred to as hooks, claws, spurs, or spines depending on their location and shape. As the developmental origin and homology of these various clasper structures is unclear, we avoided analyses that subdivided genital appendages into different classifications (e.g. spurs, hooks, claws). Using literature sources, we classified the presence or absence and number of clasper appendages for 62 additional shark species.

Data on clasper length was derived exclusively from studies that examined the relationship between body length and clasper length from sexually mature males (i.e. with calcified claspers), as our aim was to match data on body and clasper length from the same individuals. When mean values were not presented in the text, we extracted raw data from plots depicting the relationship between body length and clasper length using the program Graphclick (Boyle et al. 2012) and calculated the mean body and clasper length from sexually mature males. Whenever possible we collected data on testes and body mass data from species where clasper length data was available. When mean testes or body mass data were not presented in the text, we extracted this data from figures using the program Graphclick or calculated values using length-weight equations and gonadosomatic (GSI) values. As testes mass can vary across the reproductive cycle, we preferentially used peak testes mass values whenever available. Using this approach, we collected data on body length and clasper length for an additional 36 species. Of these additional species, mean male body mass and testes mass were available for 10 and 12 species, respectively.

198

5.3.3 Compiling the final dataset

Combining our field collected and literature collected data, our final dataset comprised data on the presence/absence of genital appendages for 80 species, and of these species data on body length, clasper length, and testes size was available for 28, 29, and 17 species, respectively. Data on body length and clasper length was collected for 57 species, of which 28 species had associated testes mass data. The final dataset included data from 111 species, representing all 8 shark orders and 29 of 34 families. In compiling the dataset on genital appendages, we focused on structures composed of cartilage components that extended from the clasper (e.g. spurs, hooks, claws and spines). This approach meant that we did not classify claspers with dermal denticles, small tooth-like projections made of dentine that cover the outer surface of the clasper in some species, as possessing genital appendages, as the presence or absence of dermal denticles is relatively infrequently reported in the literature and it is unclear if elaborate genital appendages and dermal denticles share a common function.

5.3.4 Phylogenetic Analyses

All phylogenetic analyses were performed using a recent elasmobranch phylogeny constructed using genetic data from 610 shark species (Stein et al. 2018). A consensus tree was generated from the original set of 500 phylogenetic trees using the function consensus.tree in the package phytools (Revell 2012), with the function consensus.edges used to set branch lengths. In our dataset, 88 of the 111 species were present in the phylogeny. All analyses were performed in the statistical program R version 3.4.1 (R Core Team 2017).

199 5.3.5 Ancestral state reconstruction

We reconstructed the presence/absence of elaborate genital structures across the shark phylogeny. We first compared equal rate (ER) and all rates different (ARD) models of transition probabilities between discrete character states using the ace function in the R package ape (Paradis et al. 2004). A likelihood ratio test was used to identify the best fitting model, which was used as the model of character evolution in subsequent analyses. The evolution of elaborate genital structures was reconstructed using stochastic character mapping sampled from Markov chain Monte Carlo (MCMC) posterior distributions within a Bayesian framework (Bollback 2006; Revell 2013) as implemented using the make.simmap function in the R package phytools (Revell 2012). The model was fit using 1000 simulations to estimate the average number of state changes occurring in the phylogeny.

5.3.6 Phylogenetic linear models

We used phylogenetic linear regression models to assess association between reproductive traits. We previously demonstrated that body size and testes characteristics evolve under a non-Brownian motion model of trait evolution (Chapter 3). Therefore, before implementing phylogenetic linear models we assessed trait evolution for all of the continuous traits in our dataset (body length, clasper length, and testes mass) using the fitContinuous function in the R package ‘geiger’ (Harmon et al. 2008). We compared the fit of Brownian motion (BM), Ornstein-Uhlenbeck (OU) and Early-burst (EB) models of trait evolution for each trait and compared fits using AICc values. For all traits, the OU and EB models best explained the observed trait evolution (Supporting Information Table S5.1). Therefore, when performing phylogenetic linear models, we used the phylolm function in the R package phylolm to fit models that specify alternative models of trait evolution. As both OU and EB model had equal AICc weights, we ran separate models specifying ‘OUrandomroot’ (Ornstein-Uhlenbeck model with a stationary

200 distribution of the ancestral state at the root, henceforth referred to as OU models) and ‘EB’ (Early-burst) models of covariance. We first tested if clasper length or testes mass is influenced by the presence of elaborate genital structures using models that incorporated body length. To consider how sperm competition influences genital size in sharks, we tested the relationship between clasper length and body length-corrected testes mass, a commonly used proxy for sperm competition risk (Chapter 2). Body length data was more widely available than body mass in our dataset (mean body length was available for 57 species, compared with only 29 for mean body mass), and body length and weight were tightly correlated (PGLS:  = 0.00, d.f. = 24, slope = 3.18  0.10, t = 32.88, p < 0.001, r = 0.99). Therefore, we included body length (rather than mass) as a covariate in all phylogenetic linear models to control for the allometric relationship between body size and both clasper length and testes size.

5.4 Results

5.4.1 Ancestral state reconstructions and the gains and losses of genital appendages

Genital appendages were present in five of the eight orders assessed. While the Carcharhinformes, Hexanchiformes and Squatiniformes entirely lacked genital appendages, they were present in all Heterodontiformes and Pristiophoriformes in our dataset. All Lamniformes possessed clasper appendages, with the exception of one basal species, Misukurina owstoni. In the Squaliformes, only one species (Deania calcea) lacked clasper appendages. Of the six Orectolobiform species in our dataset for which the presence of genital appendages could be determined, appendages were absent in two species and present in four. Ancestral state reconstructions revealed phylogenetic patterns of distribution of genital appendages across the shark phylogeny that were largely consistent at the order level (Figure 5.2). An ARD model of character evolution provided a significantly better fit to the data than an ER model (ARD lnL = -17.88; ER lnL = -21.10; df = 1, p =

201 0.01). Therefore, we used an ARD model in subsequent analyses. Stochastic character mapping analyses using an ARD model revealed an average of 7.32 changes in character states across the phylogeny. However, we detected only evolutionary losses in genital appendages (Figure 5.2).

202

Figure 5.2. Stochastic posterior probability density map of the presence or absence of clasper appendages in shark species, calculated from 1000 simulated phylogenies under an ‘all rates different’ (ARD) model of transitition probabilities. Branch colours represent the proportional probability of clasper appendages present (red) or absent (blue). Pie charts indicate the probable ancestral clasper appendage state at the nodes.

203 5.4.2 Phylogenetic linear models

We detected a complex relationship between clasper length (controlling for body length) and the presence or absence of genital appendages. A significant interaction term in the model revealed that species with genital appendages have longer clasper lengths at larger body sizes, while species without genital appendages have longer clasper lengths at smaller body sizes (Figure 5.2, Table 5.1). In contrast, male investment in testes mass, corrected for body length, did not differ based on the presence or absence of genital appendages (Table 5.1).

Clasper length, correcting for body length, varied markedly across species, ranging from 5% of male length in Hexanchus griseus to 16% of male length in Galeus melastomus. Yet despite this variation in clasper length, there was no relationship between clasper length and body length-corrected testes mass, irrespective of which model of trait evolution (OU or EB) was specified (Table 5.2). When we added the presence or absence of genital appendages as a factor in our analyses, clasper length remained unassociated with body length corrected testes mass (OU: AIC = -10.69, lnL = 11.34, d.f.= 11, testes mass: slope = 0.06  0.11, t = 0.50, p = 0.63, r = 0.15, lower CI = -1.48, upper CI = 2.46; body length: slope = 0.78  0.32, t = 2.47, p = 0.03, r = 0.60, lower CI = 0.22, upper CI = 4.64; genital appendages: slope = 0.08  0.11, t = 0.70, p = 0.50, r = 0.21, lower CI = -1.30, upper CI = 2.67; EB: AIC = - 17.82, lnL = 14.91, d.f.= 11, testes mass: slope = 0.09  0.07, t = 1.25, p = 0.24, r = 0.35, lower CI = -0.80, upper CI = 3.25; body length: slope = 0.71  0.22, t = 3.17, p = 0.009, r = 0.69, lower CI = 0.76, upper CI = 5.48; genital appendages: slope = 0.16  0.14, t = 1.10, p = 0.29, r = 0.31, lower CI = -0.94, upper CI = 3.09).

204 Table 5.1: Phylogentically controlled linear regressions between (a, c) clasper length and the presence or absence of genital appendages, and (b, d) testes mass and the presence or absence of genital appendages. We present models using both the Ornstein-Uhlenbeck (OU) and Early-burst (EB) models of trait evolution, as comparisons of model fits indicated that both models fit the data equally well (SI Table S1). Body length was included as a covariate in all models to control for the allometric relationship between clasper length and body size, and testes mass and body size. We present the degrees of freedom (df), Aikake Information Criterion (AIC), log-likelihood (lnL), slope of the regression with standard error, t-statistic and p-value for each model. Significant results (p<0.05) are highlighted in bold.

205 Trait Predictors df AIC lnL Slope SE t p r Lower CI Upper CI

Ornstein-Uhlenbeck

(a) Clasper length Body length 22 -37.28 24.64 0.72 0.12 5.91 <0.001 0.78 3.25 8.49 Genital appendages -1.93 0.53 -3.62 0.002 -0.61 -5.82 -31.36

Genital appendages: 0.67 0.18 3.78 0.001 0.63 1.49 6.00 Body length

(b) Testes mass Body length 16 7.73 1.14 2.50 0.25 10.06 <0.001 0.93 6.05 13.99 Genital appendages -0.22 0.13 -1.67 0.12 -0.39 -3.69 0.40

Early-burst

(c) Clasper length Body length 22 -37.73 24.87 0.79 0.12 6.53 <0.001 0.81 3.75 9.23 Genital appendages -1.82 0.50 -3.64 0.001 -0.61 -5.84 -1.38 Genital appendages: 0.65 0.17 3.89 <0.001 0.64 1.59 6.13 Body length

(d) Testes mass Body length 16 16.38 -3.19 2.63 0.35 7.43 <0.001 0.88 4.17 10.60 Genital appendages -0.22 0.32 -0.69 0.50 -0.17 -2.65 1.29

Table 5.2. Phylogentically controlled linear regressions between clasper length and body length and testes mass. We present models using both the Ornstein- Uhlenbeck (OU) and Early-burst (EB) models of trait evolution, as comparisons of models indicated that both were an equally good fit (SI Table S1). Body length was included as a covariate to control for the allometric relationship between testes mass and body size. We present the degrees of freedom (df), Aikake Information Criterion (AIC), log-likelihood (lnL), slope of the regression with standard error, t- statistic and p-value for each model. Significant results (p<0.05) are highlighted in bold.

Trait Predictors df AIC lnL Slope SE t p r Lower CI Upper CI

Ornstein-Uhlenbeck Clasper length Testes mass 22 -34.38 22.19 0.02 0.06 0.28 0.78 0.06 -1.68 2.24 Body length 0.89 0.18 4.98 <0.001 0.73 2.50 7.39

Early-burst Clasper length Testes mass 22 -36.32 23.16 0.02 0.06 0.40 0.69 0.08 -1.57 2.36

Body length 0.87 0.16 5.26 <0.001 0.75 2.73 7.72

Figure 5.3. Relationships between body length and clasper length (plotted on a log10-scale) across shark species, in species for which genital appendages are present (white points) and absent (black points) (appendages: log10(y) = -2.04 + 1.36x ; no appendages: log10(y) = -1.14 + 1.05x).

5.5 Discussion

Despite longstanding interest in shark reproductive biology (Wourms 1977), our understanding of how selection acts on reproductive traits in this group remains surprisingly limited (Fitzpatrick et al. 2012c). In this study, we addressed this key gap by combining field and literature collected data on clasper morphology from shark species spanning five continents and more than a century of scientific interest, respectively. Our results reveal multiple independent losses in genital appendages across the shark phylogeny. These evolutionary losses were phylogenetically clustered within orders, with the Carcharhinformes, Hexanchiformes and Squatiniformes showing losses. We also detected some losses

208 of clasper appendages within an order, in the Lamniformes, Squaliformes and Orectolobiformes. We did not detect any gains in genital appendages, and thus the evolution of these appendages in sharks appears unidirectional. Yet, despite the variation in the presence of genital appendages among shark orders, the functional significance of these structures remains unclear. While we detected a complex relationship between genital appendages and clasper length (relative to body size), both the presence of elaborate genital structures and clasper length were not associated with body size corrected testes mass, a proxy measure of sperm competition risk. Thus, in contrast to other taxonomic groups (Ramm 2007; Orr and Brennan 2016), our findings cast doubt on the role of sperm competition in driving the evolution of elaborate clasper appendages and length in sharks.

Despite the clear sexually selected role that genital appendages such as hooks and spines play in other taxonomic groups (Stockley 2002; Hotzy and Arnqvist 2009; Friesen et al. 2014), the function of these structures in sharks, and the factors that influence their evolution, remain enigmatic. The dramatic asymmetry in losses of elaborate genital structures compared to gains could reflect that clasper appendages may be difficult to evolve de novo, or are associated with a unidirectional ecological change or another morphological or life history trait that evolves asymmetrically. Moreover, the lack of an association between genital appendages and sperm competition risk we observed in sharks directly contradicts recent findings in primates (Orr and Brennan 2016), where the presence of penile spines is associated with greater investment in testes mass. At present, our results suggest that the evolution of elaborate genital structures in sharks may be driven by a selective agent other than sperm competition. For example, elaborate genital appendages may play a role in cryptic female choice. However, if the purpose of elaborate genital structures is to stimulate females, it isn’t clear why they would be lost in some clades but maintained in others. In addition, elaborate genital structures appear to be lost when claspers are relatively short, suggesting that clasper appendages do not evolve via a compensatory mechanism whereby either elaborate genital structures or increased clasper length can fulfil the role of stimulating females. Genital appendages may also evolve as a result of sexual

209 conflict between males and females. If elaborate genital appendages injure females during mating, as appears to be the case in sharks (Pratt and Carrier 2001), the loss of elaborate genital structures in certain groups may be associated with variation in ecological or life history factors that reduce the strength of sexual conflict; for example, when population density is low or sex ratios are less male-biased. In many shark species females exhibit biennial breeding cycles due to extended gestation periods, whereas males are capable of breeding annually (Klimley 2013). In biennial breeding species, breeding populations will be strongly male-biased, leading to increased sexual conflict over female mating rate. If genital spines function as a means of overcoming female resistance to mating in sharks, as in other taxa (Hotzy and Arnqvist 2009), we would therefore expect genital spines to be more common in species with extended gestation periods and, consequently, long breeding cycles. As female sharks can store sperm for prolonged periods (Pratt 1993), effectively uncoupling fertilisation from mating, there may be no associated decrease in sperm competition risk in species with shorter breeding cycles, despite the potentially lower female remating rate (Boomer et al. 2013). Determining how female breeding cycles influences the evolution of genital appendages and moderates the potential strength of sexual conflict would help to clarify the function of genital appendages in sharks.

We found no evidence that clasper length is influenced by sperm competition risk in sharks. This finding stands in contrast with the positive relationships demonstrated between genital length and proxy measures for sperm competition risk in some bird and mammal taxa (Coker et al. 2002; Ramm 2007; Fitzpatrick et al. 2012a; Dines et al. 2014). As argued above, it is possible that other forms of sexual selection may be shaping clasper size. For example, genital length can be influenced by pre-copulatory sexual selection, with females of some fish species showing a preference for mating with males that have longer intromittent organs (Langerhans et al. 2005; Kahn et al. 2010). It has been suggested that clasper ‘flaring’ behaviour in male sharks, where males extend their clasper while spreading its distal tips, may function as a pre-copulatory sexual display (Ritter and Compagno 2013), possibly analogous to gonopodial swinging in other internally fertilising fish (Langerhans et

210 al. 2005). However, pre-copulatory female mate choice appears relatively limited in sharks and convenience polyandry seems to dominate (Boomer et al. 2013, but see Pratt and Carrier 2001; Whitney et al. 2004). Thus, it seems unlikely that pre-mating mate choice explains differences in relative genital size in sharks. Instead, as claspers interact intimately with the female reproductive tract during copulation, clasper morphology may be influenced by male-female coevolutionary dynamics arising from cryptic female choice or sexual conflict (reviewed in Brennan and Prum 2015). While there is currently limited evidence of male-female genital coevolution in sharks (Matthews 1950), further investigation of covariance in male and female reproductive anatomy would be valuable in understanding selective dynamics in this group.

In addition to sexual selection, genital morphology may also be influenced by natural selection. For example, in Gambusia affinis and Gambusia holbrookii males with longer external intromittent organs have reduced burst-swimming speeds, which leaves males more vulnerable to predators (Langerhans et al. 2005). Indeed, male exhibited longer genital lengths in populations lacking predators in these Poecilid fish (Langerhans et al. 2005), highlighting the inherent antagonism between sexual selection for longer genitalia and natural selection for shorter genitalia in the face of increased predation risk. Like Poecilid fishes, shark genitalia cannot be retracted during locomotion, which likely generates drag. As many shark species are active swimmers and travel long distances, clasper length may be limited by hydrodynamic constraints (sensu Langerhans et al. 2005), even in the absence of increased predation risk (as is the case for shark species that are apex predators). Thus, the absence of a relationship between clasper length and sperm competition risk we observed in sharks could stem from natural selection for swimming performance acting as a ‘brake’ on increasing genital length in response to increasing levels of sperm competition risk. In contrast, genital length increases with sperm competition risk in cetaceans (whales, dolphins, and porpoises) (Dines et al. 2014), a group of large aquatic organisms that may also experience hydrodynamic constraints associated with locomotion. However, unlike in sharks and Poecilid fishes, longer genitalia is unlikely to negatively affect swimming

211 performance in cetaceans as the penis is retracted into the body when not in use (Miller 2007). Due to the hydrodynamic costs associated with non-retractable external genitalia in aquatic species (Langerhans et al. 2005), it appears likely that clasper morphology is shaped by a combination of natural and sexual selection in sharks, though further research is needed to disentangle these selective processes.

This study represents the first examination of the the evolution of genital morphology across sharks. Our findings showcase multiple independent evolutionary losses (and no gains) in genital appendages in sharks, but could not identify the selective processes that act on clasper length and appendages in this ancient vertebrate lineage. These findings contribute to a growing appreciation that patterns of selection on male genital morphology are complex and unlikely to be universal across taxa (Ramm 2007). However, while this represents an important first step, key questions remain unresolved. Our results do not support the hypothesis that clasper size and weaponry are shaped by sperm competition in sharks, leaving the evolutionary reason(s) for the wide variation in clasper morphology across shark species an open question. To clarify how selection acts on clasper morphology, future work should consider sexual conflict between males and females and coevolutionary dynamics between the claspers and the female reproductive tract arising from cryptic female choice and/or sexual conflict (Orr and Brennan 2016). There is also a clear need to move beyond simple linear measures of genital length that may not adequately describe biologically meaningful variation that is available for selection to act on. Indeed, genital shape predicts fertilisation success, even when genital length does not, in many species (Arnqvist and Danielsson 1999; House and Simmons 2003; Stockley et al. 2013). Claspers are complex three-dimensional structures comprised of various cartilaginous components that vary enormously across species in both size and shape (Hamlett 1999). Future work in sharks assessing how variation in clasper shape responds to sperm competition, cryptic female choice and sexual conflict would be a worthwhile endeavour. In particular, focus on the inter-specifically variable curvature of the terminal accessory cartilage components, including the ‘claw’ and clasper ‘spurs’ (Muñoz-Chápuli and Ramos 1989), would be valuable, as these

212 terminal components are likely key targets of selection due to their close interaction with the female reproductive tract during copulation. Finally, investigating female effects and utilising techniques that incorporate three- dimensional variation in clasper shape has the potential to greatly enhance our understanding of the selective processes that shape genital evolution in sharks.

5.6 Acknowledgments

We thank Clinton Duffy and Malcolm Francis for contributing unpublished data, the Lonja Pesquera de Algeciras (in particular Trudy, Vicente, Rosa and Curro) for facilitating access to fish caught as bycatch, and Samantha Hook, Lou Ruddell, Phil Root, Emilio Riginella, Vittoria Correale and Licia Finotto for helping in data collection. Thanks to staff at the NOAA Southeast Fisheries Science Center’s Pelagic Longline Observer Program for samples from the northern Gulf of Mexico, and the University of West Florida’s Jim Hammond for logistical support. Research was supported by a Biotechnology and Biological Sciences Research Council (BBSRC) Doctoral Training Program studentship to AR, Manchester University and Stockholm University (AR and JLF) and a Knut and Alice Wallenberg Academy Fellowship to JLF. CM was financially supported by Clodia project funded by the Veneto Region (Italy) Law 15/2007 (DGR no 4069). LL was supported by Erasmus+ mobility program 2017, University of Padova (Italy). EG acknowledges Fundação para a Ciência e Tecnologia (FCT), through the strategic project UID/MAR/04292/2013 granted to MARE. FGG was supported by grants (CGL2012- 34685 and CGL2016-76173-P, co-funded by the European Regional Development Fund) from the Spanish Ministry of Economy. Funding for the project was also provided by the Western Australia Government, the University of Western Australia, the Australian Research Council, and a Doñana Biological Station project financed by the Spanish Ministry of Economy, through the Severo Ochoa Program for Centres of Excellence (SEV-2012-0262).

213 5.7 References

Arnqvist, G. 1998. Comparative evidence for the evolution of genitalia by sexual selection. Nature 393:784–786. Arnqvist, G., and I. Danielsson. 1999. Copulatory behaviour, genital morphology, and male fertilization success in water striders. Evolution. 53:147-156. Arnqvist, G., and L. Rowe. 2005. Sexual Conflict. Princeton University Press, Princeton, New Jersey, USA. Blanckenhorn, W. U., D. J. Hosken, O. Y. Martin, C. Reim, Y. Teuschl, and P. I. Ward. 2002. The costs of copulating in the dung fly Sepsis cynipsea. Behav. Ecol. 13:353–358. Bollback, J. P. 2006. SIMMAP: Stochastic character mapping of discrete traits on phylogenies. BMC Bioinformatics 7:88. Booksmythe, I., M. L. Head, J. S. Keogh, and M. D. Jennions. 2016. Fitness consequences of artificial selection on relative male genital size. Nat. Commun. 7:11597. Boomer, J. J., R. G. Harcourt, M. P. Francis, T. I. Walker, J. M. Braccini, and A. J. Stow. 2013. Frequency of Multiple Paternity in Gummy Shark, Mustelus antarcticus, and Rig, Mustelus lenticulatus, and the Implications of Mate Encounter Rate, Postcopulatory Influences, and Reproductive Mode. J. Hered. 104:371–379. Boyle, M. A., A. L. Samaha, A. M. Rodewald, and A. N. Hoffmann. 2012. Evaluation of the reliability and validity of GraphClick as a data extraction program. Brennan, P. L. R., and R. O. Prum. 2015. Mechanisms and Evidence of Genital Coevolution: The Roles of Natural Selection, Mate Choice, and Sexual Conflict. Cold Spring Harb. Perspect. Biol. 7:a017749. Cayetano, L., A. a Maklakov, R. C. Brooks, and R. Bonduriansky. 2011. Evolution of male and female genitalia following release from sexual selection. Evolution 65:2171–83. Coker, C. R., F. McKinney, H. Hays, and K. M. Cheng. 2002. Intromittent Organ Morphology and Testis Size in Relation to Mating System in Waterfowl. Auk 119:403–413.

214 Crudgington, H. S., and M. T. Siva-Jothy. 2000. Genital damage, kicking and early death. Nature 407:855–856. Dines, J. P., E. Otárola-Castillo, P. Ralph, J. Alas, T. Daley, A. D. Smith, and M. D. Dean. 2014. Sexual selection targets cetacean pelvic bones. Evolution. 68:3296–3306. Dixson, A. F. 1987. Observations on the evolution of the genitalia and copulatory behaviour in male primates. J. Zool. 213:423–443. Dougherty, L. R., I. A. Rahman, E. R. Burdfield-Steel, E. V. Greenway, and D. M. Shuker. 2015. Experimental reduction of intromittent organ length reduces male reproductive success in a bug. Proc. R. Soc. B Biol. Sci. 282:20150724. Dufour, L. 1844. Anatomie générale des diptères. Ann. des Sci. Nat. 1:244–264. Eberhard, W. G. 1996. Female Control: Sexual Selection by Cryptic Female Choice. Princeton University Press, Princeton. Eberhard, W. G. 2010. Rapid Divergent Evolution of Genitalia. in The Evolution of Primary Sexual Characters in Animals. Oxford University Press. Eberhard, W. G. 1985. Sexual Selection and Animal Genitalia. Harvard University Press. Edvardsson, M., and T. Tregenza. 2005. Why do male Callosobruchus maculatus harm their mates? Behav. Ecol. 16:788–793. Evans, J. P., C. Gasparini, G. I. Holwell, I. W. Ramnarine, T. E. Pitcher, and A. Pilastro. 2011. Intraspecific evidence from guppies for correlated patterns of male and female genital trait diversification. Proc. R. Soc. B Biol. Sci. 278:2611–2620. Fitzpatrick, J. L., M. Almbro, A. Gonzalez-Voyer, N. Kolm, and L. W. Simmons. 2012a. Male contest competition and the coevolution of weaponry and testes in pinnipeds. Evolution. 66:3595–3604. Fitzpatrick, J. L., R. M. Kempster, T. S. Daly-Engel, S. P. Collin, and J. P. Evans. 2012b. Assessing the potential for post-copulatory sexual selection in elasmobranchs. J. Fish Biol. 80:1141–1158. Friesen, C. R., A. R. Kerns, and R. T. Mason. 2014. Factors influencing paternity in multiply mated female red-sided garter snakes and the persistent use of sperm stored over winter. Behav. Ecol. Sociobiol. 68:1419–1430. Friesen, C. R., E. J. Uhrig, R. T. Mason, and P. L. R. Brennan. 2016. Female behaviour

215 and the interaction of male and female genital traits mediate sperm transfer during mating. J. Evol. Biol. 29:952–964. Gasparini, C., A. Pilastro, and J. P. Evans. 2011. Male genital morphology and its influence on female mating preferences and paternity success in guppies. PLoS One 6:e22329. Grieshop, K., and M. Polak. 2012. The precopulatory function of male genital spines in Drosophila ananassae [Doleschall] (Diptera: Drosophilidae) revealed by laser surgery. Evolution. 66:2637–2645. Hamlett, W. C. 1999. Sharks, Skates, and Rays: The Biology of Elasmobranch Fishes. JHU Press. Harmon, L. J., J. T. Weir, C. D. Brock, R. E. Glor, and W. Challenger. 2008. GEIGER: Investigating evolutionary radiations. Bioinformatics 24:129–131. Hosken, D. J., and P. Stockley. 2004. Sexual selection and genital evolution. Trends Ecol. Evol. 19:87–93. Hotzy, C., and G. Arnqvist. 2009. Sperm Competition Favors Harmful Males in Seed Beetles. Curr. Biol. 19:404–407. Hotzy, C., M. Polak, J. L. Rönn, and G. Arnqvist. 2012. Phenotypic engineering unveils the function of genital morphology. Curr. Biol. 22:2258–2261. House, C. M., and L. W. Simmons. 2003. Genital morphology and fertilization success in the dung beetle Onthophagus taurus: An example of sexually selected male genitalia. Pp. 447–455 in Proceedings of the Royal Society B: Biological Sciences. Kahn, A. T., B. Mautz, and M. D. Jennions. 2010. Females prefer to associate with males with longer intromittent organs in mosquitofish. Biol. Lett. 6:55–58. Klimley, A. P. 2013. Courtship and Reproduction. P. in The Biology of Sharks and Rays. University of Chicago Press. Kwan, L., Y. Y. Cheng, F. H. Rodd, and L. Rowe. 2013. Sexual conflict and the function of genitalic claws in guppies (Poecilia reticulata). Biol. Lett. 9:20130267. Langerhans, R. B. 2011. Genital Evolution. P. in Ecology and Evolution of Poeciliid Fishes. Langerhans, R. B., C. A. Layman, and T. J. DeWitt. 2005. Male genital size reflects a

216 tradeoff between attracting mates and avoiding predators in two live-bearing fish species. Proc. Natl. Acad. Sci. 102:7618–7623. Long, J. A., E. Mark-Kurik, Z. Johanson, M. S. Y. Lee, G. C. Young, Z. Min, P. E. Ahlberg, M. Newman, R. Jones, J. den Blaauwen, B. Choo, and K. Trinajstic. 2015. Copulation in antiarch placoderms and the origin of gnathostome internal fertilization. Nature 517:196–199. Matthews, L. H. 1950. Reproduction in the basking shark, Cetorhinus maximus (Gunner). Philos. Trans. R. Soc. Lond. B. Biol. Sci. 234:247–316. Mautz, B. S., B. B. M. Wong, R. A. Peters, and M. D. Jennions. 2013. Penis size interacts with body shape and height to influence male attractiveness Proc. Natl. Acad. Sci. 110:6925-6930. Mayr, E. 1963. Animal Species and Evolution. Harvard University Press, Cambridge, Massachusetts, USA. Miller, D. L. 2007. Reproductive Biology and Phylogeny of Cetacea: Whales, Porpoises and Dolphins. CRC Press. Muñoz-Chápuli, R., and F. Ramos. 1989. Morphological comparison of Squalus blainvillei and S. megalops in the Eastern Atlantic, with notes on the genus. Japanese J. Ichthyol. 36:6–21. Nessler, S. H., G. Uhl, and J. M. Schneider. 2007. A non-sperm transferring genital trait under sexual selection: An experimental approach. Proc. R. Soc. B Biol. Sci. 274:2337–2341. Orr, T. J., and P. L. R. Brennan. 2016. All Features Great and Small - The Potential Roles of the Baculum and Penile Spines in Mammals. Pp. 635–643 in Integrative and Comparative Biology. Paradis, E., J. Claude, and K. Strimmer. 2004. APE: Analyses of phylogenetics and evolution in R language. Bioinformatics 20:289–290. Parker, G. A. 1970. Sperm competition and it’s evolutionary consequences in the insects. Biol. Rev. 45:525–567. Parker, G. A. 1984. Sperm competition and the evolution of animal mating strategies. P. in Sperm competition and the evolution of animal mating systems. Academic Press, Orlando, Florida. Polak, M., and A. Rashed. 2010. Microscale laser surgery reveals adaptive function

217 of male intromittent genitalia. Proc. R. Soc. B Biol. Sci. 277:1371–1376. Pratt, H. L. 1993. The storage of spermatozoa in the oviducal glands of western North Atlantic sharks. Environ. Biol. Fishes 14:139–149. Pratt, H. L., and J. C. Carrier. 2001. A review of elasmobranch reproductive behavior with a case study on the nurse shark , Ginglymostoma cirratum. Environ. Biol. Fishes 60:157–188. R Core Team. 2017. R: A language and environment for statistical computing. R Foundation for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Ramm, S. A. 2007. Sexual Selection and Genital Evolution in Mammals: A Phylogenetic Analysis of Baculum Length. Am. Nat. 169:360–369. Revell, L. J. 2012. phytools: An R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol. 3:217–223. Revell, L. J. 2013. Two new graphical methods for mapping trait evolution on phylogenies. Methods Ecol. Evol. 4:754–759. Ritter, E. K., and L. V. J. Compagno. 2013. Clasper Flaring : Maintenance Behavior , or a Normally Hidden Feature of Male Whitetip Reef Sharks, Triaenodon Obesus? Open Fish Sci. J. 10–12. Ronn, J., M. Katvala, and G. Arnqvist. 2007. Coevolution between harmful male genitalia and female resistance in seed beetles. Proc. Natl. Acad. Sci. 104:10921–10925. Simmons, L. W. 2014. Sexual selection and genital evolution. Austral Entomol. 53:1– 17. Stein, R. W., C. G. Mull, T. S. Kuhn, N. C. Aschliman, L. N. K. Davidson, J. B. Joy, G. J. Smith, N. K. Dulvy, and A. O. Mooers. 2018. Global priorities for conserving the evolutionary history of sharks, rays and chimaeras. Nat. Ecol. Evol. 2:288–298. Stockley, P. 2002. Sperm competition risk and male genital anatomy: Comparative evidence for reduced duration of female sexual receptivity in primates with penile spines. Evol. Ecol. 16:123–137. Stockley, P., S. A. Ramm, A. L. Sherborne, M. D. F. Thom, S. Paterson, and J. L. Hurst. 2013. Baculum morphology predicts reproductive success of male house mice under sexual selection. BMC Biol. 11:66.

218 Von Helversen, D., and O. Von Helversen. 1991. Pre-mating sperm removal in the bushcricket Metaplastes ornatus Ramme 1931 (Orthoptera, Tettigonoidea, Phaneropteridae). Behav. Ecol. Sociobiol. 28:391–396. Waage, J. K. 1979. Dual function of the damselfly penis: Sperm removal and transfer. Science. 203:916–918. American Association for the Advancement of Science. Whitney, N. M., H. L. Pratt, and J. C. Carrier. 2004. Group courtship, mating behaviour and siphon sac function in the whitetip reef shark, Triaenodon obesus. Anim. Behav. 68:1435–1442.

219 Chapter 5: Supporting Information

Supporting Table S5.1. Summary of model fits for Brownian Motion (BM), Ornstein-Uhlenbeck (OU) and Early-burst (EB) models of trait evolution for clasper length and body length. All trait values were ln-transformed prior to analysis. The

Brownian rate parameter (2), rate of evolutionary change parameter () and selection strength parameter (a) are presented for the BM, OU and EB models respectively. We used the Aikake Information Criterion corrected for small sample sizes (AICc) to compare model fits. We report delta AICc (Di) and Akaike weights

(wi), which indicate that Ornstein-Uhlenbeck and Early-burst are equally preferred models of evolution for both traits.

220 Supporting Table S5.1.

Brownian Motion Ornstein-Uhlenbeck Early-burst

2  lnL AICc i i  lnL AICc i i a lnL AICc i i

Clasper length 0.003 -45.92 96.08 7.74 0.01 0.008 -40.94 88.34 0.00 0.49 0.02 -40.94 88.34 0.00 0.49 Body length 0.003 -39.00 82.24 6.31 0.02 0.01 -34.72 75.93 0.00 0.49 0.016 -34.72 75.93 0.00 0.49

Supporting Table S5.2. Mean clasper length, presence or absence of clasper appendages (hooks, spines, spurs or claws), mean body length and mean testes mass data collected for shark species. Field-collected data was collected from sharks caught in Brazil, Morocco, Australia, Canada, Spain, Italy, Portugal, USA and the UK. All sharks sampled were previously caught for commercial sale or as bycatch in fisheries, during scientific surverys of the MEDITS (International Bottom Trawl Survey in the Mediterranean) project (http://www.sibm.it/SITO%20MEDITS/principalemedits.htm), or during scientific longline surveys conducted between 150-1400 m depth around the Condor seamount, southwest of Faial island (for details on methodology see Giacomello et al. 2013; Menezes & Giacomello 2013).

Species Clasper Clasper Appendages Body Length Testes References Length (mm) Mass (g) (mm) Alopias superciliosus 350.34 - 2849.44 - Chen et al. 1997

Apristurus brunneus 51.50 - 609.00 32.30 Flammang et al. 2008

Apristurus exsanguis - absent - - Sato et al. 1999

Apristurus microps 50.09 - 550.90 - Ebert et al. 2006

Apristurus saldanha 66.25 - 829.40 - Ebert et al. 2006

Asymbolus analis 59.78 absent 487.67 12.19 Leigh-Sharpe 1926a; Kyne et al. 2011 Atelomycterus erdmanni - absent - - Fahmi and White 2015

Atelomycterus marmoratus 64.36 absent 613.47 - Leigh-Sharpe 1926a; White 2007

Brachaelurus waddi - absent - - Leigh-Sharpe 1926b

Carcharhinus amblyrhynchos 154.80 - 1480.50 187.29 Stevens and McLoughlin 1991; Wetherbee et al. 1997 Carcharhinus brachyurus 211.52 - 2241.69 - Lucifora et al. 2005b

Carcharhinus brevipinna 301.00 - 1922.00 210.40 Stevens and McLoughlin 1991; Capapé et al. 2003 Carcharhinus cautus 127.50 - - 53.20 Field-collected

Carcharhinus falciformis 266.68 - 2000.83 90.49 Stevens and McLoughlin 1991; Galván-Tirado et al. 2015 Carcharhinus galapagensis 221.87 - 2330.04 - Wetherbee et al. 1996

Carcharhinus isodon 171.41 - 1429.99 - Castro 1993

Carcharhinus limbatus 170.40 absent 1271.67 255.57 Castro 1996

Carcharhinus melanopterus 151.11 absent 1052.70 - Leigh-Sharpe 1924; Papastamatiou et al. 2009 Carcharhinus plumbeus 253.21 - 1658.15 224.45 Stevens and McLoughlin 1991; Saïdi et al. 2005 Carcharhinus signatus 211.16 absent 1979.48 - Hazin et al. 2000

Carcharhinus tilstoni 97.25 - 739.50 35.49 Field-collected

Carcharodon carcharias - present - - Ainley and Klimley 1996

Centrophorus granulosus 91.50 present 908.50 17.19 White et al. 2013

Centrophorus squamosus 97.73 present 1100.00 30.62 Field-collected; Clarke et al. 2001 Centroscyllium excelsum - present - - Shirai and Nakaya 1990

Centroscyllium fabricii 64.25 present 680.00 - Field-collected

Centroscymnus coelolepis 70.13 present 934.21 43.61 Field-collected, Girard and Du Buit 1999 Centroscymnus owstoni - present - - Yano and Tanaka 1988

Cephaloscyllium ventriosum - absent - - Leigh-Sharpe 1926a

Cetorhinus maximus - present - - Jungersen 1899; Matthews 1950

Chaenogaleus macrostoma - absent - - Leigh-Sharpe 1924

Chiloscyllium plagiosum 95.07 - 752.94 26.26 Chen and Liu 2006

Chiloscyllium punctatum - absent - - Leigh-Sharpe 1922a

Chlamydoselachus anguineus - absent - - Leigh-Sharpe 1926

Cirrhigaleus asper 68.90 - 1007.77 21.20 Fischer et al. 2006

Deania calcea 71.00 absent 885.00 - Field-collected

224 Echinorhinus brucus - present - - Leigh-Sharpe 1926

Etmopterus baxteri - present - - Fitzpatrick et al. 2012

Etmopterus princeps 58.15 present 564.00 - Field-collected

Etmopterus pusillus 27.00 present 420.00 - Field-collected

Etmopterus spinax 27.00 present 344.00 2.37 Leigh-Sharpe 1922b

Etmopterus unicolor - present - - Yano and Tanaka 1989

Euprotomicrus bispinatus - present - - Hubbs T. Iwai, and K. Matsubara. 1967 Figaro boardmani 43.99 - 431.30 - Kyne et al. 2011

Galeorhinus galeus 103.43 absent 1365.50 - Leigh-Sharpe 1921; Lucifora et al. 2004 Galeus atlanticus 42.90 - 389.93 - Rey et al. 2010

Galeus longirostris - absent - - Tachikawa and Taniuchi 1987

Galeus melastomus 66.45 absent 412.00 5.68 Field-collected

Ginglymostoma cirratum - present - - Carrier et al. 1994

Gollum attenuatus 73.00 - 870.93 - Yano 1993

Halaelurus natalensis - absent - - Leigh-Sharpe 1926a

Haploblepharus edwardsii - absent - - Leigh-Sharpe 1926a

Hemigaleus microstoma 54.07 - 871.62 - White 2007

Hemipristis elongata 111.61 - 1505.83 - White 2007

Heterodontus japonicus - present - - Leigh-Sharpe 1926

225 Heterodontus portusjacksoni 77.21 present 651.90 - Leigh-Sharpe 1922b; Powter and Gladstone 2008 Hexanchus griseus 136.00 absent 2990.00 690.96 Field-collected; Leigh-Sharpe 1922b, Hexanchus nakamurai - absent - - Ebert et al. 2013

Hexanchus vitulus - absent - - Springer and Waller 1969

Iago omanensis - absent - - Compagno and Springer 1971

Isistius brasiliensis - present - - Hubbs T. Iwai, and K. Matsubara. 1967 Isurus oxyrinchus 325.67 present 2303.33 250.00 Field-collected; Clinton Duffy and Malcolm Francis, unpublished data Lamiopsis tephrodes - absent - - White et al. 2010

Lamna nasus - present - - Jensen et al. 2002

Megachasma pelagios - present - - Compagno 1990

Mitsukurina owstoni - absent - - Leigh-Sharpe 1924

Mustelus antarcticus 112.00 absent - 14.80 Jungersen 1899; Fitzpatrick et al. 2012 Mustelus canis - absent - - Leigh-Sharpe 1926

Mustelus lunulatus - absent - - Leigh-Sharpe 1926

Mustelus manazo 123.64 - 799.10 - Yamaguchi 1997

Mustelus mustelus - 0 - - Field-collected

Mustelus punctulatus 132.70 0 877.21 23.80 Saïdi et al. 2009

Mustelus widodoi 73.35 - 891.19 - White 2007

226 Negaprion brevirostris 222.22 absent 1902.73 257.95 Field-collected

Notorynchus cepedianus 154.97 - 1693.10 - Ebert 1996; Lucifora et al. 2005a

Orectolobus halei 293.81 present 1900.44 - Huveneers 2006; Corrigan et al. 2008 Orectolobus maculatus 209.30 - 1393.13 - Corrigan et al. 2008

Orectolobus ornatus 130.60 present 879.30 - Huveneers 2006; Corrigan et al. 2008 Oxynotus centrina - present - - Leigh-Sharpe 1926

Oxynotus japonicus - present - - Yano and Murofushi 1985

Poroderma africanum - absent - - Von Bonde 1948

Prionace glauca - absent - - Klimley 2013

Pristiophorus cirratus - present - - Leigh-Sharpe 1922b

Pristiophorus nancyae - present - - Weigmann et al. 2014

Pseudocarcharias kamoharai 114.36 - 932.15 12.07 Oliveira et al. 2010

Pseudotriakis microdon - absent - - Leigh-Sharpe 1926

Rhincodon typus - present - - White 1930

Rhizoprionodon acutus 128.87 - 983.90 - Capapé et al. 2006

Rhizoprionodon terraenovae 89.60 - 793.80 - Field-collected

Scroederichthys bivius - absent - - Leigh-Sharpe 1926a

Scyliorhinus cabofriensis - absent - - Soares et al. 2016

Scyliorhinus canicula 37.46 absent 452.52 12.63 Field-collected

227 Scyliorhinus capensis 84.63 absent 891.66 - Compagno 1988

Scyliorhinus comoroensis - absent - - Compagno 1988

Scyliorhinus haeckelii - absent - - Soares et al. 2016

Scyliorhinus stellaris 71.00 - 875.00 75.70 Field-collected

Scyliorhinus torazame - absent - - White 1937

Somniosus longus - present - - Yano et al. 2004

Somniosus microcephalus - present - - Lydersen et al. 2016

Sphyrna lewini 192.27 - 2336.32 53.18 Hazin et al. 2001

Sphyrna tiburo 86.33 absent 781.83 8.10 Field-collected; Fitzpatrick et al. 2012 Squalus acanthias 73.09 present 732.17 21.52 Field-collected

Squalus blainville 51.20 - 557.00 13.31 Field-collected; Muñoz-Chápuli and Ramos 1989 Squalus cubensis - present - - Muñoz-Chápuli and Ramos 1989

Squalus margaretsmithae - present - - Viana et al. 2017

Squalus megalops 41.33 present 501.67 4.63 Field-collected

Squalus mitsukurii 52.03 present 659.00 6.40 White 1937; Fischer et al. 2006

Squatina squatina - absent - - Jungersen 1899

Squatina varii - absent - - Vaz and de Carvalho 2018

Triakis semifasciata - absent - - Leigh-Sharpe 1924

Zameus ichiharai - present - - Yano and Tanaka 1984

228

Supporting Information References

Ainley, D. G., and A. P. Klimley. 1996. Great White Sharks : The Biology of Carcharodon Carcharias. Academic Press. Capapé, C., Y. Diatta, M. Diop, O. Guelorget, Y. Vergne, and J.P. Quignard. 2006. Reproduction in the milk shark, Rhizoprionodon acutus (Rüppell, 1837) (Chondrichthyes: Carcharhinidae), from the coast of Senegal (eastern tropical Atlantic). Acta Adriat. 47:111–126. Capapé, C., A. A. Seck, Y. Diatta, J. Zaouali, and F. Hemida. 2003. Distribution and reporductive biology of the spinner shark, Carcharhinus brevipinna (Müller and Henle, 1841)(Chondrichthyes: Carcharhinidae). Isr. J. Zool. 49:269–286. Carrier, J. C., H. L. Pratt, and L. K. Martin. 1994. Group reproductive behaviors in free-living nurse sharks, Ginglymostoma cirratum. Copeia 1994:646–656. Castro, J. I. 1996. Biology of the blacktip shark, Carcharhinus limbatus, off the Southeastern United States. Bull. Mar. Sci. 59:508–522. Castro, J. I. 1993. The biology of the finetooth shark, Carcharhinus isodon. Environ. Biol. Fishes 36:219–232. Chen, C.-T., K.-M. Liu, and Y.-C. Chang. 1997. Reproductive biology of the bigeye thresher shark, Alopias superciliosus (Lowe, 1839) (Chondrichthyes: Alopiidae), in the northwestern Pacific. Ichthyol. Res. 44:227–235. Chen, W. K., and K. M. Liu. 2006. Reproductive biology of whitespotted bamboo shark Chiloscyllium plagiosum in northern waters off Taiwan. Fish. Sci. 72:1215–1224. Clarke, M. W., P. L. Connolly, and J. J. Bracken. 2001. Aspects of reproduction of the deep water sharks Centroscymnus coelolepis and Centrophorus squamosus from west of Ireland and Scotland. J. Mar. Biol. Assoc. United Kingdom 81:1019–1029. Compagno, L. J. V., and S. Springer. 1971. Iago, a new genus of Carcharhinid sharks , with a redescription of I. omanensis. Fish. Bull. 69:615–626. Compagno, L. V. J. 1988. Scyliorhinus comoroensis sp. n., a new catshark from the

229 Comoro Islands, western Indian Ocean (Carcharhiniformes, Scyliorhinidae). Bull. du Muséum Natl. d’histoire Nat. Paris 10:603–625. Compagno, L. J. V. 1990. Relationships of the megamouth shark, Megachasma pelagios (Lamniformes: Megachasmidae), with comments on its feeding habits. Pp. 357–379 in H. L. Pratt Jr., S. H. Gruber, and T. Taniuchi, eds. Elasmobranchs as living resources: Advances in the Biology, Ecology, Systematics and the Status of the Fisheries. NOAA Technical Report 90. Corrigan, S., C. Huveneers, T. S. Schwartz, R. G. Harcourt, and L. B. Beheregaray. 2008. Genetic and reproductive evidence for two species of ornate wobbegong shark Orectolobus spp. on the Australian east coast. J. Fish Biol. 73:1662–1675. Ebert, D. A. 1996. Biology of the sevengill shark Notorynchus cepedianus (Peron, 1807) in the temperate coastal waters of southern Africa. South African J. Mar. Sci. 17:93–103. Ebert, D. A., L. J. V. Compagno, and P. D. Cowley. 2006. Reproductive biology of catsharks (Chondrichthyes: Scyliorhinidae) off the west coast of southern Africa. ICES J. Mar. Sci. 63:1053–1065. Ebert, D. A., W. T. White, and H. C. Ho. 2013. Redescription of Hexanchus nakamurai Teng 1962, (Chondrichthyes: Hexanchiformes: Hexanchidae), with designation of a neotype. Zootaxa 3752:20–34. Fahmi, and W. T. White. 2015. Atelomycterus erdmanni, a new species of catshark (Scyliorhinidae: Carcharhiniformes) from Indonesia. J. Ocean Sci. Found. 14:14–27. Fischer, A. F., D. P. Veras, F. H. V. Hazin, M. K. Broadhurst, G. H. Burgess, and P. G. V. Oliveira. 2006. Maturation of Squalus mitsukurii and Cirrhigaleus asper (Squalidae, Squaliformes) in the southwestern equatorial Atlantic Ocean. J. Appl. Ichthyol. 22:495–501. Fitzpatrick, J. L., R. M. Kempster, T. S. Daly-Engel, S. P. Collin, and J. P. Evans. 2012. Assessing the potential for post-copulatory sexual selection in elasmobranchs. J. Fish Biol. 80:1141–1158. Flammang, B. E., D. A. Ebert, and G. M. Cailliet. 2008. Reproductive biology of deep- sea catsharks (Chondrichthyes: Scyliorhinidae) in the eastern North Pacific.

230 Environ. Biol. Fishes 81:35–49. Galván-Tirado, C., F. Galván-Magaña, and R. I. Ochoa-Báez. 2015. Reproductive biology of the silky shark Carcharhinus falciformis in the southern Mexican Pacific. J. Mar. Biol. Assoc. United Kingdom 95:561–567. Giacomello, E., Menezes, G.M., Bergstad, O.A., 2013. An Integrated Approach for Studying Seamounts: CONDOR Observatory. Deep Sea Research Part II, vol.98 Part A, 1-6. Girard, M., and M. H. Du Buit. 1999. Reproductive biology of two deep-water sharks from the British Isles, Centroscymnus coelolepis and Centrophorus squamosus (Chondrichthyes: Squalidae). J. Mar. Biol. Assoc. United Kingdom 79:923–931. Hazin, F., A. Fischer, and M. Broadhurst. 2001. Aspects of reproductive biology of the scalloped hammerhead shark, Sphyrna lewini, off northeastern Brazil. Environ. Biol. Fishes 61:151–159. Hazin, F. H. V., F. M. Lucena, T. S. A. L. Souza, C. E. Boeckmann, M. K. Broadhurst, and R. C. Menni. 2000. Maturation of the night shark, Carcharhinus signatus, in the South-Western Equatorial Atlantic Ocean. Bull. Mar. Sci. 66:173–185. Hubbs T. Iwai, and K. Matsubara., C. L. 1967. External and internal characters, horizontal and vertical distribution, luminescence, and food of the dwarf pelagic shark, Euprotomicrus bispinatus. Bull. Scripps Inst. Oceanogr. 10:1–81. Huveneers, C. 2006. Redescription of two species of wobbegongs (Chondrichthyes: Orectolobidae) with elevation of Orectolobus halei Whitley 1940 to species level. Zootaxa 1284:29–51. Jensen, C. F., L. J. Natanson, H. L. Pratt Jr., N. E. Kohler, and S. E. Campagna. 2002. The reproductive biology of the porbeagle shark (Lamna nasus) in the western North Atlantic Ocean. Fish. Bull. 100:727–738. Jungersen, H. F. E. 1899. On the appendices genitales in the Greenland Shark Somniosus microcephalus (Bl. Schn.) and other selachians. Danish Ingolf- Expedition 2:1. Klimley, A. P. 2013. Courtship and Reproduction. in The Biology of Sharks and Rays. University of Chicago Press. Kyne, P. M., A. J. Courtney, and M. B. Bennett. 2011. Observations on the reproductive biology of three catsharks (Carcharhiniformes: Scyliorhinidae:

231 Asymbolus and Figaro) from the continental shelf of southern Queensland, Australia. J. Mar. Biol. Assoc. United Kingdom 91:1157–1164. Leigh-Sharpe, W. H. 1926. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands, together with a dissertation on the Cowper’s glands of Homo. Memoir XI. J. Morphol. 42:349–358. Leigh-Sharpe, W. H. 1921. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir II. J. Morphol. 35:359–381. Leigh-Sharpe, W. H. 1922a. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir IV. J. Morphol. 36:307–320. Wiley-Blackwell. Leigh-Sharpe, W. H. 1926a. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir IX. J. Morphol. 42:321–334. Leigh-Sharpe, W. H. 1924. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir VI. J. Morphol. 39:553–566. Leigh-Sharpe, W. H. 1926. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir VIII. J. Morphol. 42:307–320. Leigh-Sharpe, W. H. 1926b. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir XI. J. Morphol. 42:307–320. Leigh-Sharpe, W. H. 1922b. The comparative morphology of the secondary sexual characters of holocephali and elasmobranch fishes the claspers, clasper siphons, and clasper glands. Memoir V. J. Morphol. 36:221–243. Lucifora, L., R. Menni, and A. Escalante. 2005a. Reproduction, abundance and feeding habits of the broadnose sevengill shark Notorynchus cepedianus in north Patagonia, Argentina. Mar. Ecol. Prog. Ser. 289:237–244. Lucifora, L. O., R. C. Menni, and A. H. Escalante. 2005b. Reproduction and seasonal occurrence of the copper shark, Carcharhinus brachyurus, from north

232 Patagonia, Argentina. ICES J. Mar. Sci. 62:107–115. Lucifora, L. O., R. C. Menni, and A. H. Escalante. 2004. Reproductive biology of the school shark, Galeorhinus galeus, off Argentina: Support for a single south western Atlantic population with synchronized migratory movements. Environ. Biol. Fishes 71:199–209. Lydersen, C., A. T. Fisk, and K. M. Kovacs. 2016. A review of Greenland shark (Somniosus microcephalus) studies in the Kongsfjorden area, Svalbard Norway. Polar Biol. 39:2169–2178. Matthews, L. H. 1950. Reproduction in the basking shark, Cetorhinus maximus (Gunner). Philos. Trans. R. Soc. Lond. B. Biol. Sci. 234:247–316. Menezes, G.M., Giacomello, E., 2013. Spatial and temporal variability of demersal fishes at Condor seamount (Northeast Atlantic). Deep Sea Research Part II, vol.98 Part A, 101-113. Muñoz-Chápuli, R., and F. Ramos. 1989. Morphological comparison of Squalus blainvillei and S. megalops in the Eastern Atlantic, with notes on the genus. Japanese J. Ichthyol. 36:6–21. Oliveira, P., F. H. V. Hazin, F. Carvalho, M. Rego, R. Coelho, A. Piercy, and G. Burgess. 2010. Reproductive biology of the crocodile shark Pseudocarcharias kamoharai. J. Fish Biol. 76:1655–1670. Papastamatiou, Y. P., J. E. Caselle, A. M. Friedlander, and C. G. Lowe. 2009. Distribution, size frequency, and sex ratios of blacktip reef sharks Carcharhinus melanopterus at Palmyra Atoll: A predator-dominated ecosystem. J. Fish Biol. 75:647–654. Powter, D. M., and W. Gladstone. 2008. The reproductive biology and ecology of the Port Jackson shark Heterodontus portusjacksoni in the coastal waters of eastern Australia. J. Fish Biol. 72:2615–2633. Rey, J., R. Coelho, D. Lloris, B. Séret, and L. Gil de Sola. 2010. Distribution pattern of Galeus atlanticus in the Alborán Sea (south western Mediterranean) and some sexual character comparison with Galeus melastomus. Mar. Biol. Res. 6:364– 372. Saïdi, B., M. N. Bradaï, and A. Bouaïn. 2009. Reproductive biology and diet of Mustelus punctulatus (Risso, 1826) (Chondrichthyes: Triakidae) from the Gulf

233 of Gabès, central Mediterranean Sea. Sci. Mar. 73:249–258. Saïdi, B., M. N. Bradaï, A. Bouaïn, O. Guélorget, and C. Capapé. 2005. The reproductive biology of the sandbar shark, Carcharhinus plumbeus (Chondrichthyes: Carcharhinidae), from the Gulf of Gabès (southern Tunisia, central Mediterranean). Acta Adriat. 46:47–62. Sato, K., K. Nakaya, and A. L. Stewart. 1999. A new species of the deep‐water catshark genus Apristurus from New Zealand waters (Chondrichthyes: Scyliorhinidae). J. R. Soc. New Zeal. 29:325–335. Shirai, S., and K. Nakaya. 1990. A new squalid species of the genus Centroscyllium from the emperor seamount chain. Japanese J. Ichthyol. 36:391–398. Soares, K. D. A., U. L. Gomes, and M. R. De Carvalho. 2016. Taxonomic review of catsharks of the Scyliorhinus haeckelii group, with the description of a new species (Chondrichthyes: Carcharhiniformes: Scyliorhinidae). Zootaxa 4066:501-534. Springer, S., and R. A. Waller. 1969. Hexanchus vitulus, a new sixgill shark from the Bahamas. Bull. Mar. Sci. 19:159–174. Stevens, J. D., and K. J. McLoughlin. 1991. Distribution, size and sex composition, reproductive biology and diet of sharks from Northern Australia. Mar. Freshw. Res. 42:151–199. Tachikawa, H., and T. Taniuchi. 1987. Galeus longirostris, a new species of the sawtail catshark from Japan. Japanese J. Ichthyol. 33:352–359. Vaz, D. F. B., and M. R. de Carvalho. 2018. New Species of Squatina (Squatiniformes: Squatinidae) from Brazil, with Comments on the Taxonomy of Angel Sharks from the Central and Northwestern Atlantic. Copeia 106:144–160. Viana, S. T. de F. L., M. W. Lisher, and M. R. de Carvalho. 2017. Two new species of short-snouted dogfish sharks of the genus Squalus Linnaeus, 1758, from southern Africa (Chondrichthyes: Squaliformes: Squalidae). Mar. Biodiv. (online). https:// doi.org/10.1007/s12526-017-0673-8 Von Bonde, C. 1948. The development of the striped dogfish (Lui-Haai), Poroderma africanum (Gmelin). Trans. R. Soc. South Africa 31:465–474. Weigmann, S., M. F. W. Stehmann, and R. Thiel. 2014. Contribution to the taxonomy and distribution of Pristiophorus nancyae (Elasmobranchii:

234 Pristiophoriformes) from the deep western Indian Ocean. Mar. Biodivers. 44:189–202. Wetherbee, B. M., G. L. Crow, and C. G. Lowe. 1996. Biology of the Galapagos shark, Carcharhinus galapagensis, in Hawai’i. Environ. Biol. Fishes 45:299–310. Wetherbee, B. M., G. L. Crow, and C. G. Lowe. 1997. Distribution, reproduction and diet of the gray reef shark Carcharhinus amblyrhynchos in Hawaii. Mar. Ecol. Prog. Ser. 151:181–189. White, E. G. 1937. Interrelationships of the elasmobranchs with a key to the order Galea. Bull Am Mus Nat Hist 74:25–138. White, E. G. 1930. The whale shark, Rhineodon typus: Description of the skeletal parts and classification based on the Marathon specimen captured in 1923. Bull. Am. Museum Nat. Hist. 61:129–160. White, W. T. 2007. Aspects of the biology of carcharhiniform sharks in Indonesian waters. J. Mar. Biol. Assoc. United Kingdom 87:1269–1276. White, W. T., D. A. Ebert, G. J. P. Naylor, H. C. Ho, P. Clerkin, A. Veríssimo, and C. F. Cotton. 2013. Revision of the genus Centrophorus (squaliformes: Centrophoridae): Part 1-redescription of Centrophorus granulosus (Bloch & Schneider), a senior synonym of C. acus (Garman) and C. niaukang (Teng). Zootaxa 3752:35–72. White, W. T., P. R. Last, G. J. P. Naylor, and M. Harris. 2010. Resurrection and redescription of the Borneo Broadfin Shark Lamiopsis tephrodes (Fowler, 1905) (Carcharhiniformes: Carcharhinidae). Pp. 45–69 in P. R. Last, W. T. White, and J. J. Pogonoski, eds. Descriptions of New Sharks and Rays from Borneo. CSIRO Marine and Atmospheric Research Paper 032. Yamaguchi, A. 1997. Reproductive Biology of the Starspotted Dogfish Mustelus manazo from Tokyo Bay, Japan. Fish. Sci. 63:918–922. Yano, K. 1993. Reproductive biology of the slender smoothhound, Gollum attenuatus, collected from New Zealand waters. Pp. 59–71 in The reproduction and development of sharks, skates, rays and ratfishes. Springer, Dordrecht. Yano, K., and M. Murofushi. 1985. A new prickly dogfish, Oxynotus japonicus, from Japan. Japanese J. Ichthyol. 32:129–136. Yano, K., J. D. Stevens, and L. J. V. Compagno. 2004. A review of the systematics of

235 the sleeper shark genus Somniosus with redescriptions of Somniosus (Somniosus) antarcticus and Somniosus (Rhinoscymnus) longus (Squaliformes: Somniosidae). Ichthyol. Res. 51:360-373. Yano, K., and S. Tanaka. 1989. Hermaphroditism in the lantern shark Etmopterus unicolor (Squalidae, chondrichthyes). Japanese J. Ichthyol. 36:338–345. Yano, K., and S. Tanaka. 1984. Review of the deep sea squaloid shark genus Scymnodon of Japan, with a description of a new species. Japanese J. Ichthyol. 30:341–350. Yano, K., and S. Tanaka. 1988. Size at Maturity , Reproductive Cycle , Fecundity , and Depth Segregation of the Deep Sea Squaloid Sharks Centroscymnus owstoni and C . coelolepis in Suruga Bay , Japan. Nippon Suisan Gakkaishi 54:167–174.

236 Chapter 6:

General Conclusion

237 6. General Conclusion

6.1 Thesis Summary

In recent decades, our understanding of elasmobranch reproductive biology has increased exponentially. From the earliest systematic morphological descriptions and comparisons of the reproductive systems of sharks and rays over a century ago (Jungersen 1899; Leigh-Sharpe 1921, 1922, 1924, 1926), we have developed a comprehensive understanding of the behaviour, anatomy and physiology that characterise reproduction in this charismatic group of fishes, and the extraordinary diversity therein (Hamlett 2005). Technological advances have allowed researchers to move beyond the limitations of anatomical studies and behavioural observations by providing a window into previously obscure aspects of elasmobranch breeding biology. Accelerometers can be used to monitor mating behaviour remotely (Whitney et al. 2010), hormone analysis to study the reproductive cycle (Heupel et al. 1999) and molecular genetic techniques have yielded insights into mating systems (summarised in Fitzpatrick et al. 2012). However, despite these critical advances, prior to my thesis we knew virtually nothing about the selective processes that have contributed to variation in reproductive traits in sharks and rays (Fitzpatrick et al. 2012b). The state of the field in elasmobranchs stood in stark contrast to other taxonomic groups, where sexual selection and its effects on reproductive traits have been well-documented in the 40 years since Parker (1970) focused evolutionary biologists on the selective power of postcopulatory processes (Birkhead and Møller 1998). This constitutes a gap not only in our knowledge of shark and ray biology, but also in our understanding of post-copulatory sexual selection more generally. Assessing how sexual selection acts in this basal vertebrate group would be a valuable step toward evaluating the consistency of sexual selection across vertebrate tree of life.

238 In this thesis, I addressed this deficit by providing the first comparative investigations of how post-copulatory sexual selection influences the evolution of reproductive traits in elasmobranchs. In considering the evolution of reproductive traits across species, I show that some traits (i.e. body-size corrected testes mass and sperm morphology) are associated with sperm competition risk as predicted by theory, while others (i.e. testes organisation and clasper morphology) are not. These results broadly demonstrate that female multiple mating behaviour influences the evolution of male reproductive traits, though sperm competition does not appear to be a universal driver of diversification across the entirety of the male reproductive system. Specifically, my thesis addressed the following questions:

(1) Does the percentage of multiple paternity (sperm competition risk) and the number of sires per litter (sperm competition intensity) influence investment in testes across sharks and bony fishes?

In Chapter 2, I showed that body-size corrected testes mass is positively associated with the percentage of multiply sired broods/litters and the mean number of sires per litter across fishes generally, and sharks and bony fishes separately. Due to the positive relationship between testes size and the quantity of sperm produced (Marconato and Shapiro 1996; Schärer et al. 2004; Lüpold et al. 2009; Ramm and Stockley 2010; Rowe and Pruett- Jones 2011; Ramm and Schärer 2014), this suggests that selection favours an increase in sperm number when competition is high. Increases in sperm number in response to sperm competition are predicted where sperm competition conforms to the ‘raffle principle’ – that is, fertilisation success increases in direct proportion to the number of sperm ejaculated (Parker 1982). This positive association between testes mass and sperm competition risk and intensity validates the use of body size-corrected testes mass as an appropriate proxy for the level of sperm competition across fishes, which, though widespread (Montgomerie and Fitzpatrick 2009), had never previously been tested against genetic estimates of polyandry in this group.

239 Measures of genetic mating systems are likely to better reflect the true level of sperm competition males experience than, for example, social mating systems (e.g. Griffith et al. 2002; Sefc et al. 2008), and therefore constitute a more robust test of the validity of testes mass as a metric of sperm competition risk and intensity. This positive relationship is consistent with the pattern found in mammals (Ramm et al. 2005; Soulsbury 2012) and birds (Moller and Briskie 2005), indicating that increased investment in testes size represents a common evolutionary response to sperm competition across the vertebrate tree of life.

(2) Does sperm competition risk affect the proportion of sperm-producing tissue in sharks and rays?

While Chapter 2 demonstrated that male sharks invest in larger testes when sperm competition risk is high, likely due to the positive relationship between testes size and sperm production, (Marconato and Shapiro 1996; Schärer et al. 2004; Lüpold et al. 2009; Ramm and Stockley 2010; Rowe and Pruett-Jones 2011; Ramm and Schärer 2014), males may also increase sperm production through adaptations of the structural organisation of the testes (Lüpold et al. 2009). However, in Chapter 3 I have shown that neither the proportion of sperm-producing tissue nor the maximum size of the spermatocysts is associated with body-size corrected testes mass in sharks. This indicates that, unlike birds (Lüpold et al. 2009; Rowe and Pruett-Jones 2011) and rodents (Montoto et al. 2012), male sharks do not respond to sperm competition risk by increasing the proportion of sperm-producing tissue in the testes. However, I found evidence suggestive of a positive association between the proportion of sperm-producing tissue and body size-corrected testes mass in rays. Although this relationship remains speculative due to the small number of ray species that were assessed (n=4), it raises the intriguing possibility that sharks and rays may exhibit divergent patterns of selection on testes in response to sperm competition. I suggested that the differences in body plan between the two groups might

240 explain this inconsistency; the disc-shaped ray body type imposes spatial constraints on the reproductive organs (Musick and Ellis 2005), and therefore may limit investment with testes size. Thus, structural changes to testes organisation to increase sperm production in response to sperm competition risk could be limited to species in which testes size is constrained, such as birds (Lüpold et al. 2009; Rowe and Pruett-Jones 2011), rodents (Montoto et al. 2012; Firman et al. 2015) and rays, and absent in species where constraints are relaxed, such as sharks. To resolve this question more fully, samples from more ray species are needed.

(3) Does sperm competition risk influence the length of sperm components and variation in sperm morphology across the ejaculate in sharks and rays?

In Chapter 4, I showed that the sperm flagellum appears to be an important target of selection in elasmobranchs. In both sharks and rays, flagellum length was associated with body size-corrected testes mass. However, in sharks the association was positive, and in rays, negative. This inconsistency indicates that sperm competition imposes divergent patterns on selection on flagellum length between the two groups. The positive association detected in sharks could reflect a relationship between sperm length and speed, as has been shown in other fishes (Fitzpatrick et al. 2009). Alternatively, longer sperm may be preferentially stored by females (Miller and Pitnick 2002), or better at displacing the sperm of rival males (Lüpold et al. 2012). In rays, the negative association may indicate that males at higher risk of sperm competition invest in sperm number at the expense of length, provided sperm size and number trade-off against one another (Pizzari and Parker 2009). However, as the relationship between sperm morphology, performance, and trade-offs among ejaculate traits has yet to be evaluated in sharks and rays, the reasons for this striking divergence in responses to sperm competition are currently unclear. In addition, there was a negative relationship between variation in flagellum length and body size-corrected testes mass in sharks, which supports that prediction that selection imposed

241 by sperm competition should erode variation in sperm morphology (Birkhead et al. 2005). This is consistent with the relationships found in passerine birds (Calhim et al. 2007; Immler et al. 2008) and social insects (Fitzpatrick and Baer 2011). Interestingly these taxonomic groups all exhibit extended female sperm storage, which could impose additional selective pressures limiting variation in sperm morphology if certain sperm phenotypes are more likely to be stored (Fitzpatrick and Baer 2011). There was no relationship between head and midpiece length or variation in head and midpiece length and body-size corrected testes mass, which suggests that these components are not strongly influenced by sperm competition.

(4) Do elasmobranch sperm components evolve at different rates?

In Chapter 4, I found that flagellum length exhibited a significantly higher rate of phenotypic diversification than head and midpiece length across all elasmobranchs. This could indicate that the flagellum is under stronger sexual selection than the head and midpiece, which is consistent with the finding that flagellum length is associated with sperm competition risk across both shark and ray species, but not head and midpiece length. Alternatively, the head and midpiece may be subject to stronger functional constraints on size than the flagellum. Both the head and midpiece are likely to interact more closely with the female reproductive tract and/or ovum during storage and fertilisation (Karr et al. 2009), which may impose stabilising selection on head and midpiece morphology and thereby limit diversification. However, we found no difference in the rates of diversification of flagellum length and head and midpiece length when sharks or rays were considered separately. It is unclear if this result reflects the reduced sample size when only sharks or only rays are included in the analysis, or if the opposing patterns of selection on flagellum length in sharks and rays drive diversification across elasmobranchs as a whole, but not in sharks or rays individually (Chapter 4).

242 (5) Where and how often do clasper appendages evolve across the shark phylogeny?

The ancestral state reconstruction I performed in Chapter 5 revealed multiple independent losses of clasper appendages (i.e. hooks, spines and spurs) throughout shark evolution, but no evolutionary gains. The presence or absence of clasper appendages tended to be phylogenetically clustered at the order level. This unidirectionality of clasper appendage evolution may suggest that evolving appendages from a non-appendage state may be difficult (for example, if it requires crossing a fitness valley). Alternatively, the presence or absence of clasper appendages could be associated with another physiological or life-history trait, or ecological change, that tends to evolve or progress in an asymmetrical manner.

(6) Does sperm competition risk influence clasper length and the presence or absence of clasper appendages across sharks?

In Chapter 5, I found no relationship between either clasper length or the presence of clasper appendages and body size-corrected testes mass. This suggests that, in contrast to other taxonomic groups (Ramm et al. 2005; Fitzpatrick et al. 2012a; Orr and Brennan 2016), sperm competition risk does not drive the evolution of longer or more ‘weaponised’ genitalia in sharks. The selective forces that have contributed to the high level of variation in clasper morphology across species therefore remain unclear. It is likely that other forms of post-copulatory selection could influence clasper evolution, such as cryptic female choice or sexual conflict. Cryptic female choice is likely to occur in sharks (Fitzpatrick et al. 2012b) and it has been suggested that genital appendages may be favoured if they stimulate the female (Stockley 2002). However, it is not clear why this function would be selected for in some clades, but selected against in others, resulting in the evolutionary loss of appendages. There is also some, albeit limited, evidence of sexually antagonistic coevolution in shark genital evolution (Matthews

243 1950), and sexual conflict is an important evolutionary driver of genital morphology in other taxa (e.g. Hotzy and Arnqvist 2009). It therefore seems plausible that that the presence of clasper appendages may be associated with more intense sexual conflict, though this remains to be tested.

In summary, my thesis demonstrates how polyandrous mating by females shapes the evolution of various male structures related to reproduction in elasmobranchs. When females are more likely to mate multiply or mate with a greater number of males in a given reproductive episode, males experience higher sperm competition risk and intensity respectively. Male sharks belonging to species in which the risk or intensity of sperm competition is high invest in larger testes relative to their body size (Chapter 2), which likely reflects selection acting to increase sperm production. However, sharks do not respond to sperm competition by increasing sperm production through changes to testes organisation (Chapter 3), although there is tentative evidence suggesting that rays have a higher proportion of sperm- producing tissue within the testes in species experiencing higher sperm competition risk (Chapter 3). In addition to sperm production, sperm competition influences sperm morphology in sharks and rays, albeit in opposite directions. While sharks at higher risk of sperm competition produce sperm with longer flagella, rays produce sperm with shorter flagella (Chapter 4). Flagellum length is also less variable across the ejaculate in sharks facing higher sperm competition risk, though not in rays (Chapter 4). The length of the head and midpiece is not associated with sperm competition risk in either sharks or rays, and flagellum length shows a higher rate of phenotypic diversification than head and midpiece length across all elasmobranchs. This could be interpreted either as evidence of stronger directional selection on the flagellum, or stronger functional constraints imposing stabilising selection on the head and midpiece (Chapter 4). Finally, the morphology of the claspers, which deliver sperm to the female, does not appear to be influenced by sperm competition in sharks (Chapter 5). Neither clasper length, nor the presence or absence of clasper appendages is associated with sperm competition risk, indicating that they do not function to increase performance in male-male competition

244 (Chapter 5), but may instead evolve as a result of cryptic female choice or sexual conflict.

6.2 Limitations

Although my thesis represents an important step forward in understanding how sexual selection acts on sharks, there are of course a number of limitations to my approaches. First, elasmobranchs are widely and sparsely distributed throughout the oceans, and many populations are in decline (Stein et al. 2018). Therefore, sampling is difficult and time-consuming. While I tried to maximise the efficiency of my field sampling by collecting samples from fisheries and fish markets, where sharks and rays are obviously more concentrated, the sample sizes in each if my thesis chapters are still low, both in terms of the number of species examined and the number of individuals sampled per species. A general assumption in comparative studies is that between-species variance in traits exceeds within- species variance. Provided this is the case, my results would yield insights into evolutionary responses to sperm competition in sharks and rays despite the low number of individuals sampled per species. However, to adequately validate this general assumption and to incorporate measurement error into the phylogenetic analyses, a greater number of individuals per species would be required. The opportunistic nature of my field sampling also meant that the taxonomic distribution of species could not be controlled. As a result, my datasets often span a very broad taxonomic range, making it challenging to investigate associations between sperm competition risk and reproductive traits at different taxonomic scales (i.e. family-specific responses). Relationships between reproductive traits and sperm competition risk may be obscured by contrasting patterns of selection in clades at lower taxonomic levels (e.g Immler and Birkhead 2007), which is likely to be especially important when there is significant divergence in biology, as is illustrated by the differences in evolutionary responses to sperm competition risk between sharks and rays in Chapter 3 and Chapter 4.

245 While relying on specimens caught in fisheries was necessary for both practical and ethical reasons, this approach also presented some challenges. In many cases I was not able to access fish immediately post-mortem, and they were often frozen before dissection. This was most problematic in Chapter 3, where freeze artefacts that altered the appearance of the spermatocysts precluded the use of histological samples from many individuals (although note that I am currently developing a novel method to quantify the amount of sperm producing tissue present in a histological section while accounting for variance in freeze artefacts among samples). For this reason, my dataset in Chapter 3 consisted almost entirely of species sampled from aquaria. It is very unlikely that any of the individuals sampled showed physiological adaptation to the captive environment, as the vast majority were wild-caught and therefore have no evolutionary history of captivity. However, it is not clear whether environmental factors associated with captivity could directly influence the development of reproductive traits in sharks or rays. For example, captive animals may receive a higher quality diet than their wild counterparts and therefore exhibit better body condition. Numerous reproductive traits, including testes size, can be influenced by male condition (Simmons and Kotiaho 2002; Schulte-Hostedde et al. 2005) and, although there is no evidence to suggest that testes organisation is condition-dependent in elasmobranchs, this possibility cannot be decisively excluded. However, as almost all species assessed in Chapter 3 came from aquaria, and the vast majority of these from the UK and Europe, which are subject to the same regulatory standards, it is unlikely that environmental conditions differed systematically across species to an extent that could obscure an association between testes organisation and sperm competition risk.

In addition, many shark species breed only during a distinct reproductive season (Wourms 1977; Stevens and McLoughlin 1991). This reproductive seasonality can present problems if certain reproductive traits vary throughout the year according to the breeding cycle. For example, testes are often much larger during the mating period/s, as sperm production increases (Parsons and Grier 1992). Therefore, I attempted to sample testes in the field only from males currently producing sperm, and therefore likely to be in breeding condition, or during the probable mating

246 season. However, the relationship between reproductive seasonality and testes size appears complex and variable across species (Parsons and Grier 1992). Controlling for this seasonality is further complicated by the fact that mating seasons and reproductive cycles are often not well-defined in shark and ray species, and may be environmentally dependent (Dobson and Dodd 1977). When compiling data from the literature, I preferentially collected testes mass data at peak gonadosomatic index (GSI) when a range of testes mass values throughout the year were presented, as peak values are likely to reflect sperm production levels close to mating, and therefore provide the best estimate of sperm competition risk. The effects of reproductive seasonality of testes organisation are less well-characterised than testes size, although seasonal cycles in spermatogenesis have been described in a number of shark species (Parsons and Grier 1992). In most species that have been assessed, it seems that spermatocysts at all stages of spermatogenesis are present throughout the year, with the exception of the bonnethead shark Sphyrna tiburo (Parsons and Grier 1992), although in some species there is a maximum accumulation of mature sperm over a period of months (Hanchet 1988). I restricted my analyses of testes organisation in Chapter 3 to only the areas of the testes in which spermatocysts containing the most mature sperm stages were present (stage 5-6), so any variation in the proportion of spermatocysts at different stages of development caused by seasonal effects are unlikely to have affected the results.

Due to the difficulties of sampling sharks and rays in the field, I was also unable to collect data on all reproductive and somatic traits measured for every individual and every species included in this thesis. For example, some species were too big to weigh, or individuals could not be dissected before they were sold. As a consequence, I had to combine data collected in the field with data compiled from published sources. This makes it challenging to form a holistic picture of how selection acts across the male reproductive anatomy in sharks and rays, as each chapter includes data from different species, and not all species have associated data for each trait being assessed. This is most evident in Chapter 2, where multiple paternity data was available for over 150 fish species, but I could only collect testes and body mass data for 36 species. There may also be additional error associated

247 with collecting data from the literature, for example due to population differences in body size or reproductive traits. To avoid this problem as far as possible, when combining data from different sources I tried to ensure that data collected from the literature came from studies of populations from the same geographical location where data was collected in the field.

6.3 Future directions

Expanding the taxonomic focus of research into elasmobranch reproductive biology would increase the potential for evaluating selective patterns at different taxonomic scales. As is clear throughout my thesis, rays are currently very understudied in comparison with sharks. For example, only 2 of 24 studies of the prevalence of multiple paternity within elasmobranch species identified in Chapter 2 assess ray species, despite the fact that rays (n = ~633 species) outnumber sharks (n = ~512 species). Indeed, in every chapter where both sharks and rays were considered (Chapters 3 and 4), far more sharks than rays were assessed, due in part to the greater availability of morphological data on sharks. While we identified intriguing evidence of possible differences in how sexual selection acts on reproductive traits between sharks and rays, further study of more ray species is needed to fully explore these hypotheses.

A fundamental limitation of comparative studies is that they are correlational in nature, and therefore I was unable to determine whether any of the associations I found reflect a causal relationship between sperm competition risk and the evolution of male reproductive traits. The only solution to this problem is to conduct intraspecific studies on sharks and rays which, though difficult, is certainly not impossible. Experimental evolution studies that manipulate the level of sperm competition risk males are exposed to and measure the effects on reproductive traits over successive generations would be highly impractical due to the very long generation times of elasmobranchs. However, many species breed well in captivity (Fitzpatrick et al. 2012b) and artificial inseminations of elasmobranchs have been

248 performed successfully, though relatively rarely (Luer et al. 2007). Intraspecific studies utilising artificial insemination techniques to perform competitive fertilisations under controlled conditions would shed fresh light on selective patterns on the ejaculate by helping to identify how sperm traits are associated with reproductive success, and explore the potential role of non-sperm ejaculate components such as seminal fluid in competitive dynamics.

While this thesis focuses on male-male competition post-mating and its effects on the male reproductive traits, reproduction obviously involves both sexes and sperm competition does not occur in isolation. Therefore, to fully understand how postcopulatory sexual selection shapes the evolution of the reproductive anatomy, it is necessary to examine male-female interactions such as cryptic female choice and sexual conflict. This could potentially resolve some of the inconsistencies identified in this thesis such as the divergent patterns of selection on flagellum length in sharks and rays, or provide an explanation for the variation shown in traits that are not influenced by sperm competition. Many features of shark and ray reproductive biology indicate that cryptic female choice is likely. Pre-copulatory mate choice is very limited, there is long-term sperm storage in specialised organs in many, if not all, species, and fertilisation is consequently delayed. Furthermore, the high degree of paternal skew in litters sired by multiple males (e.g. Schmidt et al. 2010; Farrell et al. 2014) strongly suggests that females may bias fertilisation of their ova towards certain males, but this has yet to be investigated. A good starting point would be to investigate how variation in the size and structure of the female sperm storage organ (the oviducal gland) influences selection on sperm size. Sperm morphology is associated with both the size of the female sperm storage organ (Miller and Pitnick 2002) and the duration of sperm storage (Kleven et al. 2009) in other taxonomic groups. In addition to sperm, male-female coevolutionary interactions are likely to shape selection on genitals (Brennan 2016), and future work should explore how the female reproductive tract morphology influences the evolution of clasper size, shape and weaponry, and vice versa. Examining these female effects is a critical next step in understanding the selective processes that shape the evolution of reproductive traits in sharks and rays.

249

Even if sexual selection does target reproductive traits, they may not be free to evolve if constraints or trade-offs prevent it. While I suggested one possible spatial constraint in rays that could explain differential evolutionary responses to selection on testes in elasmobranchs (Chapter 3), constraints on evolution are likely to be very common across animals. As the resources available for males to allocate to reproduction (and all other functions) are finite, trade-offs between reproductive traits are virtually inevitable, yet identifying them is challenging (Simmons et al. 2017). Future research on sexual selection should consider these pervasive, but often overlooked, evolutionary forces. In the broadest sense, it is necessary to integrate all selective processes, including the post-copulatory mechanisms of sperm competition, cryptic female choice and sexual conflict, but also pre- copulatory sexual selection, with the many evolutionary constraints and trade-offs that may limit responses to selection in order to work towards a holistic understanding of how reproductive traits evolve.

6.4 References

Birkhead, T. R., and A. P. Møller. 1998. Sperm Competition and Sexual Selection. 1st ed. Academic Press, San Diego, California. Birkhead, T. R., E. J. Pellat, P. Brekke, R. Yeates, and H. Castillo-Juarez. 2005. Genetic effects on sperm design in the zebra finch. Nature 434:383–387. Brennan, P. L. R. 2016. Studying Genital Coevolution to Understand Intromittent Organ Morphology. Integrative and Comparative Biology. 56:669–681 Calhim, S., S. Immler, and T. R. Birkhead. 2007. Postcopulatory sexual selection is associated with reduced variation in sperm morphology. PLoS One 2:e413. Dobson, S., and J. M. Dodd. 1977. The roles of temperature and photoperiod in the response of the testis of the dogfish, Scyliorhinus canicula to partial hypophysectomy (ventral lobectomy). Gen. Comp. Endocrinol. 32:114–115. Farrell, E. D., N. O’Sullivan, C. Sacchi, and S. Mariani. 2014. Multiple paternity in the starry smooth-hound shark Mustelus asterias (Carcharhiniformes: Triakidae). Biol. J. Linn. Soc. 111:119–125.

250 Firman, R. C., F. Garcia-Gonzalez, E. Thyer, S. Wheeler, Z. Yamin, M. Yuan, and L. W. Simmons. 2015. Evolutionary change in testes tissue composition among experimental populations of house mice. Evolution. 69:848–55. Fitzpatrick, J. L., M. Almbro, A. Gonzalez-Voyer, N. Kolm, and L. W. Simmons. 2012a. Male contest competition and the coevolution of weaponry and testes in pinnipeds. Evolution. 66:3595–3604. Fitzpatrick, J. L., and B. Baer. 2011. Polyandry reduces sperm length variation in social insects. Evolution. 65:3006–3012. Fitzpatrick, J. L., R. M. Kempster, T. S. Daly-Engel, S. P. Collin, and J. P. Evans. 2012b. Assessing the potential for post-copulatory sexual selection in elasmobranchs. J. Fish Biol. 80:1141–58. Fitzpatrick, J. L., R. Montgomerie, J. K. Desjardins, K. a Stiver, N. Kolm, and S. Balshine. 2009. Female promiscuity promotes the evolution of faster sperm in cichlid fishes. Proc. Natl. Acad. Sci. U. S. A. 106:1128–32. Griffith, S. C., I. P. F. Owens, and K. A. Thuman. 2002. Extra pair paternity in birds: a review of interspecific variation and adaptive function. Mol. Ecol. 11:2195– 2212. Hamlett, W. C. 2005. Reproductive Biology and Phylogeny of Chondrichthyes: Sharks, Batoids, and Chimaeras. CRC Press. Hanchet, S. 1988. Reproductive biology of Squalus acanthias from the east coast, South Island, New Zealand. New Zeal. J. Mar. Freshw. Res. 22:537–549. Heupel, M. R., J. M. Whittier, and M. B. Bennett. 1999. Plasma steroid hormone profiles and reproductive biology of the epaulette shark, Hemiscyllium ocellatum. Journal of Experimental Zoology. 284:586–594 Hotzy, C., and G. Arnqvist. 2009. Sperm Competition Favors Harmful Males in Seed Beetles. Curr. Biol. 19:404–407. Immler, S., and T. R. Birkhead. 2007. Sperm competition and sperm midpiece size: no consistent pattern in passerine birds. Proc. R. Soc. London B 274:561–568. Immler, S., S. Calhim, and T. R. Birkhead. 2008. Increased postcopulatory sexual selection reduces the intramale variation in sperm design. Evolution. 62:1538– 1543. Immler, S., S. Pitnick, G. A. Parker, K. L. Durrant, S. Lüpold, S. Calhim, and T. R.

251 Birkhead. 2011. Resolving variation in the reproductive tradeoff between sperm size and number. Proc. Natl. Acad. Sci. 108:5325–5330. Jungersen, H. F. E. 1899. On the appendices genitales in the Greenland Shark Somniosus microcephalus (Bl. Schn.) and other selachians. Danish Ingolf- Expedition 2:1. Karr, T. L., W. J. Swanson, and R. R. Snook. 2009. The evolutionary significance of variation in sperm-egg interactions. Pp. 305–365 in Sperm Biology. Elsevier. Kleven, O., F. Fossøy, T. Laskemoen, R. J. Robertson, G. Rudolfsen, and J. T. Lifjeld. 2009. Comparative evidence for the evolution of sperm swimming speed by sperm competition and female sperm storage duration in passerine birds. Evolution. 63:2466–2473. Leigh-Sharpe, W. H. 1921. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir II. J. Morphol. 35:359–381. Leigh-Sharpe, W. H. 1924. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir VI. J. Morphol. 39:553–566. Leigh-Sharpe, W. H. 1926. The comparative morphology of the secondary sexual characters of elasmobranch fishes. The claspers, clasper siphons, and clasper glands. Memoir VIII. J. Morphol. 42:307–320. Leigh-Sharpe, W. H. 1922. The comparative morphology of the secondary sexual characters of holocephali and elasmobranch fishes the claspers, clasper siphons, and clasper glands. Memoir V. J. Morphol. 36:221–243. Luer, C. A., C. J. Walsh, A. B. Bodine, and J. T. Wyffels. 2007. Normal embryonic development in the clearnose skate, Raja eglanteria, with experimental observations on artificial insemination. Pp. 133–149 in D. A. Ebert and J. A. Sulikowski, eds. Biology of Skates. Springer Netherlands, Dordrecht. Lüpold, S., G. M. Linz, J. W. Rivers, D. F. Westneat, and T. R. Birkhead. 2009. Sperm competition selects beyond relative testes size in birds. Evolution. 63:391–402. Lüpold, S., M. K. Manier, K. S. Berben, K. J. Smith, B. D. Daley, S. H. Buckley, J. M. Belote, and S. Pitnick. 2012. How multivariate ejaculate traits determine competitive fertilization success in Drosophila melanogaster. Curr. Biol.

252 22:1667–1672. Marconato, A., and D. Y. Shapiro. 1996. Sperm allocation, sperm production and fertilization rates in the bucktooth parrotfish. Anim. Behav. 52:971–980. Matthews, L. H. 1950. Reproduction in the basking shark, Cetorhinus maximus (Gunner). Philos. Trans. R. Soc. Lond. B. Biol. Sci. 234:247–316. Miller, G. T., and S. Pitnick. 2002. Sperm-female coevolution in Drosophila. Science. 298:1230–1233. Montgomerie, R., and J. L. Fitzpatrick. 2009. Testes, Sperm and Sperm Competition. Pp. 1–53 in B. G. M. Jamieson, ed. Reproductive Biology and Phylogeny of Fishes (Agnathans and Bony Fishes). Science Publishers. Montoto, L. G., L. Arregui, N. M. Sanchez, M. Gomendio, and E. R. S. Roldan. 2012. Postnatal testicular development in mouse species with different levels of sperm competition. Reproduction 143:333–346. Musick, J. A., and J. K. Ellis. 2005. Reproductive Evolution of Chondrichthyans. Pp. 45–79 in W. C. Hamlett, ed. Reproductive Biology and Phylogeny of Chondrichthyes: Sharks, Batoids and Chimaeras. Science Publishers, Inc., Plymouth, UK. Orr, T. J., and P. L. R. Brennan. 2016. All Features Great and Small - The Potential Roles of the Baculum and Penile Spines in Mammals. Pp. 635–643 in Integrative and Comparative Biology. Parker, G. A. 1970. Sperm competition and it’s evolutionary consequences in the insects. Biol. Rev. 45:525–567. Parker, G. A. 1982. Why are there so many tiny sperm ? Sperm competition and the maintenance of two sexes. J. Theor. Biol. 96:281–294. Parsons, G. R., and H. J. Grier. 1992. Seasonal changes in shark testicular structure and spermatogenesis. J. Exp. Zool. 261:173–184. Pizzari, T., and G. A. Parker. 2009. Sperm competition and sperm phenotype. Pp. 207–245 in Sperm Biology. Elsevier. Ramm, S. A., G. A. Parker, and P. Stockley. 2005. Sperm competition and the evolution of male reproductive anatomy in rodents. Proc. R. Soc. London B 272:949–55. Ramm, S. A., and L. Schärer. 2014. The evolutionary ecology of testicular function:

253 size isn’t everything. Biol. Rev. 89:874–888. Ramm, S. A., and P. Stockley. 2010. Sperm competition and sperm length influence the rate of mammalian spermatogenesis. Biol. Lett. 6:219–21. Rowe, M., and S. Pruett-Jones. 2011. Sperm Competition Selects for Sperm Quantity and Quality in the Australian Maluridae. PLoS One 6:e15720. Schärer, L., P. Ladurner, and R. Rieger. 2004. Bigger testes do work more: experimental evidence that testis size reflects testicular cell proliferation activity in the marine invertebrate, the free-living flatworm Macrostomum sp. Behav. Ecol. Sociobiol. 56:420–425. Schmidt, J. V., C. C. Chen, S. I. Sheikh, M. G. Meekan, B. M. Norman, and S. J. Joung. 2010. Paternity analysis in a litter of whale shark embryos. Endanger. Species Res. 12:117–124. Schulte-Hostedde, A. I., J. S. Millar, and G. J. Hickling. 2005. Condition dependence of testis size in small mammals. Evol. Ecol. Res. 7:143–149. Sefc, K. M., K. Mattersdorfer, C. Sturmbauer, and S. Koblmüller. 2008. High frequency of multiple paternity in broods of a socially monogamous cichlid fish with biparental nest defence. Mol. Ecol. 17:2531–2543. Simmons, L. W., and J. S. Kotiaho. 2002. Evolution of ejaculates: Patterns of phenotypic and genotypic variation and condition dependence in sperm competition traits. Evolution. 56:1622–1631. Simmons, L. W., S. Lüpold, and J. L. Fitzpatrick. 2017. Evolutionary Trade-Off between Secondary Sexual Traits and Ejaculates. Trends Ecol. Evol. 32:964- 976. Stein, R. W., C. G. Mull, T. S. Kuhn, N. C. Aschliman, L. N. K. Davidson, J. B. Joy, G. J. Smith, N. K. Dulvy, and A. O. Mooers. 2018. Global priorities for conserving the evolutionary history of sharks, rays and chimaeras. Nat. Ecol. Evol. 2:288–298. Stevens, J., and K. McLoughlin. 1991. Distribution, size and sex composition, reproductive biology and diet of sharks from Northern Australia. Mar. Freshw. Res. 42:151–199. Stockley, P. 2002. Sperm competition risk and male genital anatomy: comparative evidence for reduced duration of female sexual receptivity in primates with penile spines. Evol. Ecol. 123–137.

254 Whitney, N. M., H. L. Pratt, T. C. Pratt, and J. C. Carrier. 2010. Identifying shark mating behaviour using three-dimensional acceleration loggers. Endanger. Species Res. 10:71–82. Wourms, J. P. 1977. Reproduction and development in chondrichthyan fishes. Integr. Comp. Biol. 17:379–410.

255 Appendix 1: Sperm Competition

Authors:

Amy Rowley John L. Fitzpatrick

In: Encyclopaedia of Evolutionary Biology (R. Kilman, ed), pp. 245–248. Academic Press.

256 Sperm Competition AG Rowley and JL Fitzpatrick, University of Manchester, Manchester, UK

Glossary Promiscuity A mating system in which both males and Monogamy A mating system in which both males and females mate with multiple partners during each females mate with only one individual during a reproductive period. reproductive episode, most commonly a socially paired Sperm competition Competition to fertilize egg(s) partner. between ejaculates of different males that overlap spatially Polyandry A mating system in which a female mates with and temporally in the fertilization environment. multiple males during each reproductive period. Sperm quality A composite measure of the ability of a Polygyny A mating system in which a male mates male’s sperm to successfully fertilize ova relative to other with multiple females during each reproductive males in the population, as determined by sperm period. morphology, number, and performance.

Introduction influences the evolution of animal behaviors, anatomy, and physiology. Darwin (1871) recognized that males who are better able to Among species, female mating behavior varies widely and attract mates or out-compete rival males would leave more this has important implications for the risk of sperm com- offspring to future generations, and would therefore be petition experienced by males (Figure 1). In species with favored by evolution. Perhaps not surprisingly then, the pur- monogamous females, who only mate with one male, the risk suit of reproductive opportunities has led to the evolution of of sperm competition is low (or absent) and in these species some of the most striking features in the animal kingdom. sperm competition is unlikely to represent an important se- Males produce conspicuous ornaments (e.g., the peacocks tail) lective force shaping reproductive traits. However, genetic that are used to attract more mates than rivals or arm them- monogamy is exceedingly rare in animals. More often, socially selves with costly weapons (e.g., a stag’s antlers) that are used monogamous females mate with, and produce offspring by, to outcompete rivals for access to mates. However, for Darwin males outside of their social pairing. Furthermore, in many and evolutionary biologists who followed for the next century, species female promiscuity is common, as females either seek competition among males ended at mating. We now know out multiple mating opportunities or have little control over that this is not the case. Molecular evidence have revealed that the number of males attempting to fertilize their eggs (as is females frequently mate with multiple males during a single often the case in externally fertilizing species). Therefore, the reproductive period and different males commonly sire eggs prevalence of female multiple mating means that males of the from a single clutch (Birkhead and Møller, 1998). Such female vast majority of species will experience sperm competition, promiscuity means that competition between males continues albeit to varying extents, and researchers have capitalized on after mating in the form of sperm competition, the contest this variation in female promiscuity and animal mating sys- between sperm from rival males to fertilize an egg(s). The tems to gain a better understanding of how sperm competition evolutionary significance of sperm competition was first rec- shapes reproductive traits. ognized by Parker (1970), sparking a paradigm shift that As with females, males too have evolved a suite of resulted in an explosion of interest in the importance of behaviors that influence the risk of sperm competition sperm competition in shaping the evolution of reproductive (Figure 1). Males can actively guard their mates to prevent behaviors and phenotypes. subsequent matings by rival males and thus reduce their risk of The pervasive influence of sperm competition has played a sperm competition. In an effort to prevent sperm competition vital role in the evolution and extraordinary diversification of males can even take the extreme action of detaching and reproductive traits. Adaptation to the intense selective pres- lodging his genitalia inside the female to physically block sures imposed by sperm competition has generated an as- access to the female’s reproductive tract (Fromhage, 2006). In tonishing variety of reproductive phenotypes, from male contrast, male reproductive behaviors can also increase sperm genitals that resemble medieval weapons (Crudgington and competition risk. For example, socially subordinate males who Siva-Jothy, 2000), to testes so large that they account for more are unable to successfully attract females often adopt ‘sneak- than 10% of adult male body mass (Montgomerie and Fitz- ing’ behaviors (Figure 2), attempting to surreptitiously fertilize patrick, 2009), to gigantic sperm measuring a remarkable 6 cm eggs without the knowledge of socially dominant males in length (20 times longer than the male who produced (Gross, 1996). Therefore, despite the best efforts of males to them!) (Pitnick et al., 1995). Here we explore how sperm reduce their risk of sperm competition, both male and female competition generates the tremendous diversity in repro- reproductive behaviors ensure that sperm competition persists. ductive traits by describing the conditions under which sperm For internally fertilizing species, male genitalia, which de- competition occurs and examining how sperm competition liver sperm to the female’s reproductive tract, are at the front

Encyclopedia of Evolutionary Biology, Volume 4 doi:10.1016/B978-0-12-800049-6.00158-X 245 246 Sperm Competition

Genetic Multiple Promiscuous monogamy mating mating Female behaviors Sperm competition Low High Male

behaviors Mate Incomplete No mate monopolization monopolization monopolization

Figure 1 Summary of variation observed in male and female reproductive behaviors and how this variation inﬂuences the risk of sperm competition experienced by males. Male and female behaviors can work synergistically or antagonistically to inﬂuence sperm competition risk. Solid lines between males and females represent matings, while dashed lines represent thwarted mating attempts.

(b)

(a) (c)

Figure 2 Sperm competition and sneaky matings. (a) Comparison of a large guarding male and a smaller sneaking male in plainﬁn midshipman (Porichthys notatus), a species with two alternative male reproductive tactics. Both males are reproductively mature. Guarding males court females, defend territories, and provide parental care for developing embryos, while sneaking males attempt to fertilize eggs surreptitiously and then leave the costly parental care to guarding males. Because sneaking males exclusively release sperm in the presence of a rival male they experience a higher sperm competition risk compared with guarding males. Testis size of a (b) guarding and (c) sneaking male demonstrating increased investment in testes in sneaking males, which represents a characteristic response to increased sperm competition risk. Photo credit: John Fitzpatrick. line of competition among rival males. Male genital morph- in spines that remove almost all previously deposited sperm of ology is extraordinarily diverse (Figure 3), even among closely rival males from the female reproductive tract, thereby virtually related species, and this diversity is attributable in large part to eliminating sperm competition (Waage, 1979). When looking the selective pressures imposed by sperm competition across species, genitalia have more elaborate sexual ‘weap- (Hosken and Stockley, 2004). Male genitalia exhibit an as- onry,’ including adaptations for sperm removal or displace- tonishing variety of adaptations to maximize their chances of ment, in species where the risk of sperm competition is high success in sperm competition. For example, male genitalia can compared with closely related monogamous species (Stockley, displace sperm from previous matings, as is the case in the 2002). Similarly, populations of seed beetles (Callosobruchus damselﬂy Calopteryx maculata, where male genitalia are covered maculatus) evolved under experimentally enforced monogamy Sperm Competition 247

evolution of sperm number and sperm quality, both of which predict male fertilization success during sperm competition, albeit to varying degrees (Simmons and Fitzpatrick, 2012). The number of sperm competing to fertilize an egg(s) can dramatically influence a male’s competitive fertilization success. In many species, sperm competition is thought to follow a ‘raffle principle,’ where success in the raffle (in this case fertilizing eggs) is related to the number of ‘tickets’ (in this case (a) sperm) a male holds (Parker, 1982). Under the raffle principle, all sperm have an equal chance of fertilizing eggs and therefore the probability of fertilization during sperm competition increases with the number of sperm transferred. Therefore, males are expected to invest more in sperm number when the risk of sperm competition is elevated (Figure 2; Parker, 1982). The testes are the site of sperm production, and therefore an important target of selection for sperm number. Indeed, increases in testes size (in relation to body size) represent one (b) of the most robust responses to sperm competition (Figure 2). The impact of sperm competition on testes size is especially well documented among primates (Harcourt et al., 1981). For example, gorillas (Gorilla gorilla) experience a very low risk of sperm competition, as a single dominant male controls a harem of females and mates with them (almost) exclusively, and have remarkably small testes for their body size. In contrast, the closely related chimpanzee (Pan troglodyte) is highly promiscuous, facilitating very high levels of sperm com- (c) petition, and has testes four times larger than those of a gorilla, despite weighing a quarter of a gorilla’s body mass (Harcourt fi Figure 3 Diversity in genital morphology among cartilaginous shes et al., 1981). A similar pattern of increasing investment in (sharks, skates, rays, sawfish, and chimaeras). Clasper (paired testes in response to increase sperm competition risk is ob- genitalia in cartilaginous fish) morphology differs dramatically in size and shape among (a) Australian ghostshark, Callorhinchus milii served across a much broader range of primates (Harcourt (Photo credit: Eduardo Garza Gisholt), (b) common skate, Dipturus et al., 1981) and indeed across a wide range of other taxo- batis (Photo credit: Amy Rowley), which shows paired claspers with nomic groups, including birds, fish, amphibians, and reptiles the tail extending beyond the claspers and outside of the margins of (Simmons and Fitzpatrick, 2012). the photo, and (c) New Zealand lanternshark, Etmopterus baxteri Sperm competition not only selects for increases in sperm (Photo credit: Eduardo Garza Gisholt). Note the genital hooks and number but also influences the way males allocate their sperm spurs observed at the terminal portion of claspers in (a) and (b). during mating (Wedell et al., 2002). While the costs of producing an individual sperm may be negligible, the ejaculate show reduced genital spine length relative to polygamous as a whole can be costly to produce (Dewsbury, 1982). Con- populations (Cayetano et al.,2011). sequently, to maximize their reproductive success males are Sperm competition also influences genital size and shape. expected to strategically allocate their sperm during mating in For example, in house mice (Mus domesticus), males main- response to cues of sperm competition (Wedell et al., 2002; tained under breeding regimes where sperm competition oc- Parker and Pizzari, 2010). Males in many species indeed show curred for 27 generations had thicker penis bones (baculum) strategic patterns of sperm allocation, but only under specific than males maintained under enforced monogamy (Simmons conditions. There is now clear evidence that males allocate and Firman, 2013). Across rodent and carnivores, the baculum more sperm when mating in the presence of a single rival male is longer in species where sperm competition is prevalent compared with matings where no rival males were present (Ramm, 2007). There is good reason for genital morphology (Delbarco-Trillo, 2011; Kelly and Jennions, 2011). to respond to sperm competition risk, as several studies reveal However, sperm number is not the sole determinant of that genital morphology predicts male reproductive success reproductive success during sperm competition, and under a during sperm competition (Simmons et al., 2009; Stockley broad range of conditions sperm quality (i.e., sperm morph- et al., 2013). ology and performance) plays an important role in deter- Following the release of sperm, either from male genitalia mining male fertility. Unique among cells, sperm must survive inside the female’s reproductive tract or into the external en- and travel outside the body in order to fulfill their function of vironment in the case of internal and external fertilizing spe- fertilizing ova. Thus, as sperm moves toward the egg, various cies, respectively, the sperm themselves are the primary aspects of sperm quality, including sperm motility and combatants in male–male competition. Consequently, sperm swimming speed, will experience intense selection (Simmons are under intense selection, as those sperm traits that provide and Fitzpatrick, 2012). For example, in domestic fowl (Gallus an advantage during sperm competition will be favored by domesticus) and Atlantic salmon (Salmo salar), when females selection. In particular, sperm competition influences the are artificially inseminated with an equal number of sperm 248 Sperm Competition from two males, males with greater relative sperm velocity sire of sperm from rival males. In some social insect species, sperm more offspring (Birkhead et al., 1999; Gage et al., 2004). Thus, perform better in the presence of the male’s own seminal fluid in most species (but, for an exception see Dziminski et al., than seminal fluid from a different, competing male (Den Boer 2009; Fitzpatrick et al., 2012), sperm competition selects for et al., 2010). However, the influence of seminal fluid on rival increased sperm swimming speed (Simmons and Fitzpatrick, male sperm performance was not observed in monogamous 2012). social insect species, where sperm competition risk is low, Selection on sperm quality is predicted to impact on both indicating that seminal fluid function evolves in response to sperm size and speed due to an assumed link between sperm sperm competition risk (Den Boer et al., 2010). Similarly, in swimming speed and sperm morphology, with longer flagella the grass goby (Zosterisessor ophiocephalus), sperm of males expected to generate greater propulsive force and allow sperm to using sneaky behaviors when attempting to fertilize eggs, and swim faster (Gomendio and Roldan, 1991). While the under- thus experiencing a higher risk of sperm competition, swim lying relationship between sperm morphology and swimming faster in the presence of the seminal fluid from a rival male, speed is far from clear (Humphries et al.,2008; Fitzpatrick while sperm from males that use conventional courtship to et al.,2010; Simpson et al.,2014), numerous studies have woo females showed reduced fertilization rate when exposed evaluated how sperm size responds to varying levels of sperm to the seminal fluid of sneaky male (Locatello et al., 2013). competition risk. In almost every taxonomic group studied to Thus seminal fluid has an important function during com- date (including mammals, birds, fish, reptiles, amphibians, and petitive matings and as such both the accessory gland and the insects) there is evidence that sperm size increases with sperm proteins they produce evolve rapidly in response to sperm competition risk (Simmons and Fitzpatrick, 2012). Moreover, competition (Ramm et al., 2005; Linklater et al., 2007; Crud- a handful of recent studies have demonstrated that sperm gington et al., 2009). competition also selects for faster swimming sperm across In the four decades following Parker’s (1970) recognition species, and this appears to be due to a positive relationship of the evolutionary significance of sperm competition, enor- between sperm size and speed in these groups (Gomendio and mous advances have been made toward developing a com- Roldan, 1991; Fitzpatrick et al.,2009; Lüpold et al.,2009). prehensive understanding of the evolutionary dynamics that However, gaining a robust understanding of how sperm govern sperm competition and its pervasive influence on male competition influences sperm size and speed has remained a reproductive behavior, anatomy, and physiology. However, contentious issue, as results contrary to the general pattern the study of sperm competition is still relatively young and outlined above abound. Increasingly, efforts to understand many novel and exciting discoveries undoubtedly remain as how selection shapes sperm size and speed are focusing on advances in analytical tools and genomic approaches allow an better understanding the relationship and potential trade-off ever widening set of questions to be addressed. Moreover, ef- between sperm number and sperm size (Immler et al., 2011). forts to move the focus of sperm competition studies from the Although this work remains limited in scope, what is becoming researcher’s microscope to the female’s reproductive tract increasingly clear is that sperm number, size, and speed are all (Manier et al., 2010), the site of sperm competition in internal important targets of selection by sperm competition, and which fertilizing species, promise to revolutionize our understanding of these sperm traits are favored depends on the mechanism of of how sperm competition proceeds under ‘real world’ con- sperm competition operating in a species (Immler et al., 2011). ditions. In this review, we focused our attention exclusively on For example, sperm are produced in greater numbers at the evolutionary responses observed in males, paying particular expense of size in large species with dilute female reproductive attention to how sperm competition influences the evolution tracts where sperm competition adheres to the raffle principle. of male behaviors, genitalia, sperm, and seminal fluid. How- Conversely, where sperm displacement is the primary mech- ever, any discussion of sperm competition inevitably runs into anism of sperm competition, as in many insects, selection acts the related topics of cryptic female choice, where females can to increase sperm size at the expense of producing greater influence the outcome of sperm competition by biasing fer- sperm numbers (Immler et al., 2011). tility in favor of preferred males, and sexual conflict, where the Increasingly, researchers are recognizing the importance of evolutionary interests between the sexes differ. While these the non-sperm component of the ejaculate – seminal fluid – in related topics are outside the immediate scope of this entry, we mediating male reproductive success during sperm com- acknowledge their impact on shaping reproductive traits and petition (Chapman, 2001). Sperm are released from males in encourage interested readers to follow up on these topics. the company of seminal fluid, a medium rich in proteins and other molecules produced by the accessory glands. These proteins exert a powerful influence on sperm competitive dy- See also: Mating Systems in Flowering Plants. Polyandry and namics. First, seminal fluid proteins act on female physiology Female Postcopulatory Choice. Sexual Selection, Theory of and behavior, which has knock-on effects for the risk of sperm competition faced by a male’s ejaculate. For example, seminal proteins can reduce the risk of sperm competition experienced by an already inseminated ejaculate by reducing female re- References ceptivity to future matings (Chapman, 2001), aiding the displacement of rival males’ sperm (Harshman and Prout, 1994), Birkhead, T.R., Martínez, J.G., Burke, T., Froman, D.P., 1999. Sperm mobility determines the outcome of sperm competition in the domestic fowl. Proceedings and contributing to the production of mating plugs that block of Royal Society London B 266, 1759–1764. access to the female’s reproductive tract (Chapman, 2001). Birkhead, T.R., Møller, A.P., 1998. Sperm Competition and Sexual Selection, first Second, seminal proteins can act to diminish the performance ed. San Diego, CA: Academic Press. Sperm Competition 249

Cayetano, L., Maklakov, A., Brooks, R.C., Bonduriansky, R., 2011. Evolution of male and Kelly, C.D., Jennions, M.D., 2011. Sexual selection and sperm quantity: Meta- female genitalia following release from sexual selection. Evolution 65, 2171–2183. analyses of strategic ejaculation. Biological Reviews 86, 863–884. Chapman, T., 2001. Seminal fluid-mediated fitness traits in Drosophila. Heredity 87, Linklater, J.R., Wertheim, B., Wigby, S., Chapman, T., 2007. Ejaculate depletion 511–521. patterns evolve in response to experimental manipulation of sex ratio in Crudgington, H.S., Fellows, S., Badcock, N.S., Snook, R.R., 2009. Experimental Drosophila melanogaster. Evolution 61, 2027–2034. manipulation of sexual selection promotes greater male mating capacity but does Locatello, L., Poli, F., Rasotto, M.B., 2013. Tactic-specific differences in seminal not alter sperm investment. Evolution 63, 926–938. fluid influence sperm performance. Proceedings of the Royal Society of London Crudgington, H.S., Siva-Jothy, M.T., 2000. Genital damage, kicking and early death. B 280, 20122891. Nature 407, 855–856. Lüpold, S., Calhim, S., Immler, S., Birkhead, T.R., 2009. Sperm morphology and Darwin, C., 1871. The Descent of Man and Selection in Relation to Sex, first ed. sperm velocity in passerine birds. Proceedings of the Royal Society B 276, London: John Murray. 1175–1181. Delbarco-Trillo, J., 2011. Adjustment of sperm allocation under high risk of sperm Manier, M.K., Belote, J.M., Berben, K.S., et al., 2010. Resolving mechanisms of competition across taxa: A meta-analysis. Journal of Evolutionary Biology 24, competitive fertilization success in Drosophila melanogaster. Science 328, 354–357. 1706–1714. Montgomerie, R., Fitzpatrick, J.L., 2009. Reproductive Biology and Phylogeny of Den Boer, S.P., Baer, B., Boomsma, J.J., 2010. Seminal fluid mediates ejaculate Fishes, first ed. Enfield, New Hampshire: Science Publishers. competition in social insects. Science 327, 1506–1509. Parker, G.A., 1970. Sperm competition and it’s evolutionary consequences in the Dewsbury, D.A., 1982. Ejaculate cost and male choice. American Naturalist 119, insects. Biological Review 45, 525–567. 601–610. Parker, G.A., 1982. Experimental evidence for the evolution of the mammalian Dziminski, M.A., Roberts, J.D., Beveridge, M., Simmons, L.W., 2009. Sperm baculum by sexual selection. Journal of Theoretical Biology 96, 281–294. competitiveness in frogs: Slow and steady wins the race. Proceedings of Royal Parker, G.A., Pizzari, T., 2010. Sperm competition and ejaculate economics. Society London B 276, 3955–3961. Biological Review 85, 897–934. Fitzpatrick, J.L., Garcia-Gonzalez, F., Evans, J.P., 2010. Linking sperm length and Pitnick, S., Spicer, G.S., Markow, T.A., 1995. How long is a giant sperm? Nature velocity: The importance of intramale variation. Biology Letters 6, 797–799. 375, 109. Fitzpatrick, J.L., Montgomerie, R., Desjardins, J.K., et al., 2009. Female promiscuity Ramm, S., 2007. Sexual selection and genital evolution in mammals: A phylogenetic promotes the evolution of faster sperm in cichlid fishes. Proceedings of the analysis of baculum length. The American Naturalist 169, 360–369. National Academy of Sciences of the United States of America 106, 1128–1132. Ramm, S.A., Parker, G.A., Stockley, P., 2005. Sperm competition and the evolution Fitzpatrick, J.L., Simmons, L.W., Evans, J.P., 2012. Complex patterns of multivariate of male reproductive anatomy in rodents. Proceedings of the Royal Society B selection on the ejaculate of a broadcast spawning marine invertebrate. Evolution 272, 949–955. 66, 2451–2460. Simmons, L.W., House, C.M., Hunt, J., García-González, F., 2009. Evolutionary Fromhage, L., 2006. Emasculation to plug up females: The significance of pedipalp response to sexual selection in male genital morphology. Current Biology 19, damage in Nephila fenestrata. Behavioral Ecology 17, 353–357. 1442–1446. Gage, M.J.G., Macfarlane, C.P., Yeates, S., et al., 2004. Spermatozoal traits and Simmons, L.W., Firman, R.C., 2013. Experimental evidence for the evolution of the sperm competition in Atlantic Salmon. Current Biology 14, 44–47. mammalian baculum by sexual selection. Evolution 68, 276–283. Gomendio, M., Roldan, E.R., 1991. Sperm competition influences sperm size in Simmons, L.W., Fitzpatrick, J.L., 2012. Sperm wars and the evolution of male mammals. Proceedings of Royal Society London B 243, 181–185. fertility. Reproduction 144, 519–534. Gross, M.R., 1996. Alternative reproductive tactics: Diversity within sexes. Trends in Simpson, J.L., Humphries, S., Evans, J.P., Simmons, L.W., Fitzpatrick, J.L., 2014. Ecology and Evolution 11, 92–98. Relationships between sperm length and speed differ among three internally and Harcourt, A.H., Harvey, P.H., Larson, S.G., Short, R.V., 1981. Testis weight, body three externally fertilizing species. Evolution 68, 92–104. weight and breeding system in primates. Nature 293, 55–56. Stockley, P., 2002. Sperm competition risk and male genital anatomy: Comparative Harshman, L.G., Prout, T., 1994. Sperm displacement without sperm transfer in evidence for reduced duration of female sexual receptivity in primates with penile Drosophila melanogaster. Evolution 48, 758–766. spines. Evolutionary Ecology 16, 123–137. Hosken, D.J., Stockley, P., 2004. Sexual selection and genital evolution. Trends in Stockley, P., Ramm, S., Sherborne, A.L., et al., 2013. Baculum morphology predicts Ecology and Evolution 19, 87–93. reproductive success of male house mice under sexual selection. BMC Biology Humphries, S., Evans, J.P., Simmons, L.W., 2008. Sperm competition: Linking form 11, 66. to function. BMC Evolutionary Biology 8, 319. Waage, J.K., 1979. Dual function of the damselfly penis: Sperm removal and Immler, S., Pitnick, S., Parker, G.A., et al., 2011. Resolving variation in the transfer. Science 203, 916–918. reproductive tradeoff between sperm size and number. Proceedings of the Wedell, N., Gage, M.J.G., Parker, G.A., 2002. Sperm competition, male prudence National Academy of Sciences of the United States of America 108, 5325–5330. and sperm-limited females. Trends in Ecology and Evolution 17, 313–320.