Sexual Conflict and Secondary Sexual Traits in West Australian Riffle Bugs (Insecta: Hemiptera: Veliidae)

Home , Traumatic insemination

Sexual conflict and secondary sexual traits in West Australian Riffle Bugs (Insecta: Hemiptera: Veliidae). An overview of the relationship between morphology and behaviour

Paige Maroni (BSc) 32202683

This thesis is presented for the degree of Bachelor of Science Honours (Biological Science) School of Veterinary and Life Sciences Murdoch University 2017

Supervisors: Dr. Kate Bryant Dr. Nikolai Tatarnic

With support from: I declare that this thesis is my own account of my research and contains as its main content work which has not previously been submitted for a degree at any tertiary education institution.

Paige J. Maroni Abstract

How and why an animal chooses its mate is characteristic of morphological and behavioural adaptations developed to maximize reproductive success. This thesis compares patterns of sexual dimorphism (through micro-CT analyses) in three Veliidae species (Microvelia oceanica, Nesidovelia peramoena and N. herberti) and aims to explain these in the context of sexual conflict. Sexually antagonistic morphological and behavioural traits were also of interest and were documented through examinations of male-female reproductive interactions of N. peramoena via video recordings and micro-CT scan outputs of interlocked pairs, frozen in copula. Interlocked pairs were captured through a new technique for snap freezing surface dwelling insects, which was developed within this thesis. In all three species, females were found to be generally larger than males and macropters larger overall than apters, as is often the case with insects. All specimens examined were documented to express some form of a secondary sexual trait which are likely to be coevolving in an evolutionary arms race. In females, these resistance traits variously included connexival convergence, genital concealment and setal tufts. Females of N. peramoena showed strong pre-mating resistance behaviours evidenced by either fleeing or vigorously struggling during 96% of all observed intersexual interactions. In both

Nesidovelia species, males expressed persistence traits such as a ventral abdominal process on segment 8 and ventral abdominal segment 7 medial concavity. Through this study, these traits are now known to act as a pregenital clamp grasping to the females on the underside of her proctiger during the pre-mating struggle, as well as during copulation.

Interestingly, while macropterous males retain their secondary sexual traits, macropterous females were found to lack their putative resistance traits, suggesting either an energetic

iii or mechanical trade-off in development. This study is one of the first to provide evidence that sexual conflict is the likely driving force behind the behaviours and sexual dimorphism in the Veliidae. The array of different secondary sexual traits makes those in the Gerridae

(the current model group for sexual conflict and sexually antagonistic character studies) seem tame in comparison. Within these riffle bug species, unique coevolved male-female grasping and anti-grasping structures provide strong evidence for sexual antagonism in veliid mating that have not previously been documented in any surface dwelling insects.

Acknowledgments

To the following people I would like to express my sincere appreciation for your help in

making this project possible.

To Dr. Kate Bryant,

I wish to sincerely thank you Kate. Without your encouragement, support and feedback, none of this year would have been possible! To me, you are the perfect teacher and mentor and I have valued learning so much from you this year. I also wish to thank you for all of your time and understanding when things didn’t go to plan! I will always look back on my honours year and smile and that is hugely because of you!

To Dr. Nikolai Tatarnic,

Words cannot express how many thanks I have for you Nik! Thank you for introducing me to the crazy world that is bugs sex lives! Thank you for all of your time, your wisdom and insight into the processes that go on between male and female insects. Thank you also for all your help in every other aspect of this project from experimental design to field collection techniques…without you I would be lost! I look up to you and admire your drive and passion for science so thank you very much for all you have taught me over the past few years!

To Dr. Jeremy Shaw,

Thank you for all of your instruction, advice and help for using the Micro CT Xradia. Without your technical support none of the amazing images that have been produced through the software under your care within the CMCA facility would have been possible and the

v results produced throughout this project would not have been anywhere as incredible as I believe they are. Thank you very much Jeremy.

To Jacqueline Dyer & the Honours Chairs,

Thank you Jacqueline, and all other honours personnel who supported my application to conduct an honours year and more recently, aided me through the times when my health stood in the way of finishing the honours year on time. I cannot understate how thankful I am for all your support.

To Dr. Michael Calver,

Thank you very much for allowing me to kick start this amazing project in 2015 and it is because of you that I was able to utilize the micro-CT and image my specimens throughout this thesis so I truly am extremely thankful. Thank you also for your time and great feedback throughout this honours year.

To Dr. Bruno Buzatto,

Thank you for your great advice and time throughout this honours year. I am deeply thankful for your comments in regards to the betterment of my literature review.

To my Family and Friends,

A big thank you to my amazing support network! Mum, thank you for your huge amounts of support and my veliid colonies thank you for all the flies you caught for their meals! To my Dad and two sisters, Tessa and Meg, thank you for helping me through this very stressful year! All your advice and encouragement is so appreciated! Lastly, thank you to the two greatest, Claire and Danica for keeping me sane and laughing through the stressful times!

vi Table of Contents

Declaration of independence ii Abstract iii Acknowledgements V Table of Contents vii List of Tables x List of Figures xii List of Abbreviations xvii List of Appendices xviii 1.0 Chapter One - Introduction 1 1.1 General introduction 1 1.2 Sexual selection and sexual conflict in nature 2 1.3 The economy of mating in Insecta and resistance traits 5 1.4 Adaptations of persistence, resistance and harassment in Insecta 8 1.4.1 Male harassment 8 1.4.2 Secondary sexual morphologies 10 1.4.3 Grasping traits 11 1.4.4 Antigrasping traits and other resistance traits 14 1.4.5 Convenience polyandry 17 1.5 Sexually antagonistic coevolution 17 1.6 Water striders as a model taxon 21 1.7 This study 25 1.7.1 Project significance and overview 25 1.7.2 Study species 29 1.7.2.1 Veliidae, fam. Microvelia, gen. Nesidovelia, gen. 29 1.7.2.2 Species overview 30 1.7.3 Life cycle 31 1.8 Study approaches and objectives 32

vii

2.0 Chapter Two - Morphology 37 2.1 Introduction 37 2.2 Collection and maintenance of study species 37 2.2.1 Field locations 37 2.2.2 Laboratory colony 39 2.3 Comparative morphology 40 2.3.1 Micro-CT scanning 40 2.3.2 Digital image preparation 41 2.3.3 Putative secondary sexual traits 42 2.3.4 Size and shape 43 2.4 Reproductive behaviour and functional biology 46 2.4.1 Mating trials 46 2.4.2 Freezing engaged pairs 48 2.4.3 Videoed mating trials 50 2.4.4 Quantitative and qualitative measures 50 3.0 Chapter Three – Results 52 3.1 Comparative morphology and putative secondary sexual traits 52 3.1.1 Comparative morphology 52 3.1.1.1 Microvelia oceanica 52 3.1.1.2 Nesidovelia peramoena 56 3.1.1.3 Nesidovelia peramoena 59 3.1.2 Size and shape 63 3.1.2.1 Microvelia oceanica 67 3.1.2.2 Nesidovelia peramoena 68 3.1.2.3 Nesidovelia herberti 68 3.1.2.4 Abdominal segment 7 69 3.1.2.5 Female egg carrying capacity 69 3.2 Reproductive behaviour and functional biology 73 3.3 Micro-CT imaging of antagonistic structures 77

viii

4.0 Chapter Four – Discussion 81 4.1 Introduction 81 4.2 Putative secondary sexual traits 81 4.3 Size and shape 87 4.4 Female polymorphism 91 4.5 Reproductive behaviour and functional biology 93 4.6 Refinement of freezing methodology 97 4.7 Limitations and future works 98 5.0 Chapter Five – Conclusions 100 References 102 Appendices 119 Appendix 1 119 Appendix 2 122 Appendix 3 124 Appendix 4 126 Appendix 5 128 Appendix 6 133

List of Tables Table 1 Review of 12 published papers highlighting the scope of knowledge 28 presented on Australian Veliidae. This systematic review was performed in the late months of 2016 to the early months of 2017 and databases such as: Web of Science, Scopus, and JSTOR, were searched for relevant literature. Table 2 Distributions of the four species groups of Nesidovelia, as defined by 30 Andersen and Weir (2003).

Table 3 Collecting details for the three study species (Nesidovelia peramoena, 39 N. herberti and Microvelia oceanica). Table 4. Localities and number of individuals of each morph of the three study 44 species. Numbers in parentheses are the total numbers of adult animals collected for this study. Numeric values prior to the parentheses are total number of individuals scanned, measured or examined. Table 5 Quantitative and qualitative measurements taken from all micro-CT 46 scanned specimens. All measures were taken from micro-CT volumes using Drishti. Table 6 Quantitative and qualitative measurements taken from video footage 51 of N. peramoena male-female interactions.

Table 7 Summary of results presented in sections 3.1.1, 3.1.1.1, 3.1.1.2 and 63 3.1.1.3 illustrating secondary sexual traits in each sex and species. Table 8 Results of univariate ANOVAs adjusted using the Bonferroni 70 correction model. Means, standard deviations (in inverted brackets) and sample sizes (n) of the ten different species types by the ten proxy body measures. Species-morph codes are made of three letters. The first letter in the sequence identifies the species (O = M. oceanica, P = N. peramoena, H = N. herberti); the second letter refers to morphotype (A = apterous, M = macropterous); and the third letter in the sequence refers to sex (F = female, M = male). Table 9 Univariate ANOVA results showing how species-morphs differ 71 between species, morphs and the sexes for all collected specimens, based on ten body measures. Species-morph codes are made of three letters. The first letter in the sequence identifies the species (O = M. oceanica, P = N. peramoena, H = N. herberti); the second letter refers to morphotype (A = apterous, M = macropterous); and the third letter in the sequence refers to sex (F = female, M = male). Grey highlighted cells are statistically significant following Bonferroni adjustment (P <

0.005). All body measures in millimetres (mm), excluding egg count, which was counted numerically and egg volume (V = π r2 h). Table 10 Behavioural data taken from video footage of pre-mating behaviour 76 of apterous N. peramoena males and females.

List of Figures Figure 1 A) Distribution of Nesidovelia peramoena. 31 B) Distribution of Nesidovelia herberti. C) Distribution of Microvelia (Pacificovelia) oceanica. From Andersen and Weir (2003). Figure 2 A) Bow Bridge, Walpole (-34.9677, 116.9537). 38 B) Hamelin Station Reserve (-26.4268, 114.1959). C) Kalbarri National Park (-27.5555, 114.4545). D) Caladenia Lake, Perth (-32.0713, 115.8349). Original Image Source: Paige Maroni. Figure 3 Photograph of micro-CT specimen preparation. Specimens are placed 41 in a 1ml plastic transfer pipette (with the concave tip removed), filled with 100% ethanol, sealed at either end with blue tack. Specimens are separated with cotton tissue. Figure 4 Dorsal view of apterous male veliid (antenna and legs of right side 45 omitted). Letters depict body measures taken from micro-CT scanned specimens: blue letters indicate measures used in comparative analyses, all other measures are indicated in black. A) head width, B) head length, C) thorax length, D) thorax width, E) abdomen length, F) abdominal segment 1 width, G) abdominal segment 3 width, H) abdominal segment 7 width, I) femur length, J) tibia length, K) 1st tarsal segment length, L) 2nd tarsal segment length. (I-L taken on foreleg, mid leg, and hind leg). All measures are in millimetres. Modified from Andersen and Weir (2003).

Figure 5 A) 2mm (left) and 3mm (right) steel drill punches used for freezing 50 veliids. B) 2mm steel drill punch and ceramic petri dish used as mating arena for veliid snap-freezing experiment. C) Tip of 2mm drill punch that has been submerged in liquid nitrogen, prior to being used to freeze veliid pair. D) A frozen droplet that has been removed from the tip of a 2mm drill punch post encasing the veliids in a frozen droplet of water. This droplet encases two separate veliids. E) Can of ‘Chemical Technology Non-Flammable Freezing Spray’, used as alternative cooling source to liquid nitrogen. Figure 6 Micro-CT reconstructions of Microvelia oceanica apterous female. 53

xii

A) Dorsal view.

B) Postero-dorsal view of the abdominal tip.

C) Caudal view of the abdominal segment 7 and 8, showing deflected proctiger.

D) Lateral view. Scale bar 0.05mm, cn, connexiva; pc, proctiger.

Figure 7 Bar plot showing proportions of apterous females of M. oceanica 53 (n=14), N. peramoena (n=33) and N. herberti (n=42) exhibiting different degrees of connexival convergence: no convergence, partial convergence or complete convergence (the lateral borders of the abdomen are raised vertically and meet above the abdomen). Macropterous females of all three species did not display connexival convergence. Figure 8 Micro-CT reconstructions of Microvelia oceanica macropterous 54 female. A) Dorsal view. B) Lateral view. C) Caudal view of abdominal segment 7 and 8. D) Dorsal view of the abdominal tip. Scale bar 0.05mm. Figure 9 Micro-CT reconstructions of Microvelia oceanica apterous male. 55 A) Dorsal view. B) Lateral view. C) Lateral view of the femur, tibia, tarsi and grasping comb of the foreleg. D) Caudal view of abdomen, showing the genital segment. E) Lateral view of the abdominal segments 7 and 8 as well as showing the genital capsule. Scale bar 0.05mm, gc, grasping comb; py, pygophore; s8, segment 8. Figure 10 Micro-CT reconstructions of Microvelia oceanica macropterous male. 56 A) Dorsal view. B) Postero-dorsal view of the abdominal tip and wing structure. C) Caudal view of abdomen. D) Lateral view.

xiii

E) Lateral view of the abdominal segments 7, 8 and genital capsule. Scale bar 0.05mm. Figure 11 Micro-CT reconstructions of Nesidovelia peramoena apterous female 57 with partially convergent connexiva. A) Dorsal view. B) Postero-dorsal view of abdomen. C) Caudal view of the abdomen, showing deflected proctiger. D) Lateral view Scale bar 0.05mm. Figure 12 Micro-CT reconstructions of Nesidovelia peramoena macropterous 58 female. A) Dorsal view. B) Dorso-lateral view. C) Caudal view of abdomen. Scale bar 0.05mm. Figure 13 Micro-CT reconstructions of Nesidovelia peramoena apterous male. 59 A) Dorsal view. B) Postero-dorsal view of abdomen. C) Lateral view.

D) Caudal view abdomen, showing genital capsule and bifurcate abdominal process on segment 8. E) Lateral view of abdominal segments 7 and 8, genital capsule, and proctiger. F) Lateral view of the foreleg. G-H) Ventral views of the abdomen and genitalia showing the modification of abdominal segment 8 and the slight hollowing of segment 7. Scale bar 0.05mm, vp, segment 8 ventral process. Figure 14 Micro-CT reconstructions of Nesidovelia herberti apterous female. 60 A) Dorsal view. B) Caudal view of abdomen, showing converging connexiva, setal tufts, and deflected proctiger. C) Lateral view. D) Postero-dorsal view of abdomen. E) Lateral view of abdominal tip showing setal tufts (spray form).

xiv

F) Dorsolateral view of abdominal tip showing setal tufts (spray form). G) Lateral view of abdominal tip showing setal tufts (pointed-brush form). H) Dorso-lateral view of abdominal tip showing setal tufts (pointed- brush form) Scale bar 0.05mm, st, setal tufts. Figure 15 Bar plot of the proportion of different setal tuft types among apterous 61 female M. oceanica (n=14), N. peramoena (n=33) and N. herberti (n=42): no tufts, spray-shaped (setae spreading up and back) or pointed shaped tufts (setae condensed into pair of caudally directed paintbrush-like tufts). Figure 16 Micro-CT reconstructions of Nesidovelia herberti apterous male. 62 A) Dorsal view. B) Postero-dorsal view of abdomen. C) Lateral view of the foreleg, showing grasping comb. D) Caudal view of abdomen, showing genital capsule and bifurcate process on segment 8. E) Postero-ventral view of abdomen showing bifurcate process on segment 8. F) Lateral view of tip of abdomen. Scale bar 0.05mm. Figure 17 Principle Components Analysis biplot of first two principal 65 components (accounting for 87.998% of the total variance) for five body measures (body length, abdominal segment 1 and 7 widths, thorax width and length as based between the 10 species types. Figure 18. Principle Components Analysis biplot of first two principal 66 components (accounting for 94.669% of the total variance) for four body measures (body length, abdominal segment 1 width, thorax width and length as based between the 10 species types. Figure 19 Ethogram depicting the mating sequence of apterous N. peramoena, 75 based on 26 mating attempts/intersexual interactions observed in a laboratory colony. Numbers indicate how many times a given behaviour was recorded. Rectangles indicate male-initiated actions, ovals indicate female-initiated behaviours and dashed modified rectangles indicate an interaction between both sexes. Figure 20 Still images from video recordings of pre-copulatory struggles and 77 mating behaviour in N. peramoena. Images from 6 different pairs.

A-I) The male mounts the female without any observed pre- copulatory courtship. The female Immediately attempts to dislodge the male by thrashing and rolling via forward somersaults. The male is seen grasping with his legs and pregenital segments (segment 8 and the genital capsule are visibly extended and declivent). Once the female stops struggling the male rhythmically vibrates his hind legs before copulation begins, or else he is thrown from her abdomen. (Images A-C Nikolai Tatarnic 2017, WAM).

Figure 21 Micro-CT reconstructions of Nesidovelia peramoena apterous male- 79 female mating pairs. Images are of two different pairs (Pair 1: images A, C & E. Pair 2: images B, D & F). A) Detail of male grasping female thorax, with male grasping comb hooked behind the female’s pronotum. B) Lateral view of male-female pair connected by their genitalia, and the by the male abdominal ventral process on segment 8 grasping the underside of the female’s proctiger, pressing it into the caudal margin of segment 7. C-F) Lateral views of the connection between male-female mating pairs. Scale bar 0.50mm, gc, grasping comb; pc, proctiger; pr, pronotum; py, pygophore; s7, segment 7; s8, segment 8; vp, ventral process. Figure 22 Micro-CT reconstructions of Nesidovelia peramoena apterous male- 80 female mating pair 3. A-C) Lateral views of the connection points between the male and female. These images illustrate how the male’s ventral process connects to the underside of the female proctiger. The bifurcate prongs of the ventral process fit into paired depressions on the underside of the female’s procitger. D) Lateral view of a partial cross section of the female revealing the positioning of the male phallus inside the female. E-F) Underside view showing the male process fitting depressions in underside of female proctiger. Scale bar 50.00um, pc, proctiger; ph, male phallus; upc, underside of female proctiger; vp, ventral process

xvi

List of Abbreviations SST Secondary Sexual Trait SSD Sexual Size Dimorphism MAP Male Abdominal Process MANOVA Multivariate Analysis of Variance ANOVA Analysis of Variance PCA Principal Component Analysis OAF M. oceanica apterous females OAM M. oceanica apterous male OMF M. oceanica macropterous female OMM M. oceanica macropterous male PAF N. peramoena apterous female PAM N. peramoena apterous male PMF N. peramoena macropterous female HAF N. herberti apterous female HAM N. herberti apterous male HMF N. herberti macropterous female WAM Western Australian Museum GC Grasping comb Pn Pronotum VP Ventral Process S7 Segment 7 S8 Segment 8 Py Pygophore Pc Proctiger uPr Underside of Proctiger

xvii

List of Appendices Appendix 1 Defining features of the genus Microvelia Westwood in Australia 119 (Hemiptera: Heteroptera). Content from Andersen and Weir (2003).

Appendix 2 Defining features of the genus Nesidovelia (Cassis et al. 2017) in Australia 122 (Hemiptera: Heteroptera).

Appendix 3 Illustration of body structures in Nesidovelia queenslandiae. A) Dorsal view 124 of apterous ♂, antenna and legs of right side omitted; B) Left foreleg of apterous ♂. C) Ventral view of abdominal end of apterous ♂. D) Lateral view of genital segments of apterous ♂. E) Dorsal view of proctiger. F) Dorsal structure of apterous ♀, antenna and legs omitted. G) Lateral view of abdominal end of apterous ♀. H) Caudal view of abdominal end of apterous ♀. I) Dorsal view of macropterous ♂, antenna and legs omitted. pn, pronotum. Cn = connexivum; fw = forewing; gr = grasping comb of fore tibia; mt = metanotum; pn = pronotum; pr = proctiger; py = pygophore; s7 = abdominal segment 7; s8 = abdominal segment 8; t1 = abdominal tergite 1; t7 = abdominal tergite 7. tu = median tubercle on sternum 7; vp = ventral process on segment 8. From Andersen and Weir (2003). Appendix 4 Defining features of the subgenus, Microvelia (Pacificovelia). Content from 126 Andersen and Weir (2003). Appendix 5 Defining features of Nesidovelia peramoena, N. herberti and Microvelia 128 oceanica. Content from Andersen and Weir 2003; Cassis et al. 2017.

Appendix 6 6.1) Variations in connexival convergence, dorsal view. Scale bar 0.50 mm. 133 Black box highlights area of modification in A-C:

A) M. oceanica apterous male connexiva: widely separated and not strongly raised nor converging.

B) M. oceanica apterous female: converging partially but not meeting at midline.

C) N. herberti apterous female: converging and meeting at midline.

6.2) Proctiger orientation between female macropterous and apterous morphs. Scale bar 0.50mm. Black box highlights area of modification in A-B:

A) M. oceanica macropterous female proctiger deflected obscuring the gonocoxae.

B) M. oceanica apterous female proctiger is deflected obscuring the gonocoxae

6.3) Modifications of the caudal setal tufts on female abdominal laterotergite 7, lateral view. Scale bar 0.50mm. Black box highlights area of modification in A-B:

xviii

A) M. oceanica apterous female, tufts absent.

B) N. herberti apterous female, tufts present.

6.4) Modifications of the male abdominal sternite 8 lateral view. Scale bar 0.50mm. Black box highlights area of modification in A-B.

A) M. oceanica apterous male, no segment 8ventral process.

B) N. peramoena apterous male, segment 8 lateral process present.

xix

1.0 Chapter One - Introduction

1.1 General introduction

For a long time, insects have been considered simple, inflexible organisms, incapable of complicated phenotypic responses to changing environmental and/or social conditions

(Buzatto et al. 2014). This cannot be farther from the truth as this extremely diverse group of animals illustrates widespread adaptability (Chapman 2009; Wilson 2010) to all environmental conditions on earth with rapid genetically inherited phenotypic responses contributing to their success. Another factor influencing the success of insects is their extreme adaptability over evolutionary time, leading to a myriad of morphological, behavioural, physiological and developmental specializations (Whitman and

Ananthakrishnan 2009). Such rapid, dynamic responses to environmental and social conditions are central to insects’ remarkable flexibility, and have played a vital role in their evolutionary histories (Moczek 2010; Simpson et al. 2011). These traits are not restricted to adaptations driven by natural selection, but also to traits that function solely in interactions associated with mating and mate choice (for example, male-male competitive traits, mate choice signals, and even sexually coercive traits) (Buzatto et al. 2014).

Furthermore, mating sometimes involves intersexual conflict, with each sex striving to promote its own interests, often at a cost to its mate (Chapman et al. 2003; Arnqvist and

Rowe 2005; Tatarnic and Cassis 2010). This sexual conflict can result in the evolution of sexually antagonistic structures and behaviours, with adaptive advantages gained by one sex being soon after matched by counter adaptations in the other, generating an evolutionary arms race (Parker 1979; Alexander et al. 1997; Rice and Holland 1997;

Arnqvist and Rowe 2005). An insect mating system is no exception to this pattern and secondary sexual traits are the subject of the present study.

1.2 Sexual selection and sexual conflict in nature

Sexual selection was described by Charles Darwin (1871) as arising from “the advantages that certain individuals have over others of the same sex and species in exclusive relation to reproduction”. Sexual selection is often opposed by natural selection or is described as a “special case” of natural selection, as it gives rise to costly secondary sexual characters.

Extravagant examples of sexually selected characters include the horns of horned beetles or antelopes and the eye-catching plumes of many male birds (Darwin 1871; Møller 1998).

Sexual selection occurs via two processes, namely, competition among individuals of the chosen sex for access to individuals of the ‘choosy’ sex (intrasexual selection), usually male- male competition, and choice of attractive partners by individuals of the choosy sex

(intersexual selection), usually female choice (Darwin 1871; Bateman 1948; Trivers 1972;

Andersson 1994; Møller 1998; Shuker 2014). Sexual selection is an important evolutionary agent, with far-reaching consequences for behaviour, morphology, life history and mating system evolution, as well as the composition of biological communities, as sexual selection may influence speciation, as well as extinction (Andersson 1994).

Optimal mating behaviours (rate, time spent mating, location, time of day) differ between males and females such that their evolutionary interests differ. Such variation between the sexes’ optimal mating strategies are generally the result of costs, including differences in energy expended to provision offspring. For example, males normally invest less in gamete

2 production (i.e. production of many small, inexpensive sperm by males versus fewer, more costly eggs by females, i.e. anisogamy (Andersson 1994)), and parental care. This leads to differences between males and females in the factors that limit reproductive success.

Generally, males are limited by access to female gametes. Therefore, reproductive success increases for males with the number of mates, thus leading to competition between males for access to as many females as possible. This may result in overt male-male competition and/or benefits for males in advertising to females (Reynolds 1996). Contests between the sexes and more specifically, between rival males, can lead to the evolution of all sorts of male traits that improve a male’s chances of victory, including larger body size, aggression, fertilization efficiency, faster flying or running speeds, and/or increased energy reserves for sustaining prolonged contests (Arnqvist and Rowe 2013).

Generally, females are limited by their investment of time and energy into individual offspring and it benefits females to be careful in their choice of mates (Thornhill and Alcock

1983; Miller and Somjee 2014). This can lead to the evolution of exaggerated male traits which ultimately act as indicators of good genes, driven by female choice (Fisher 1915;

Williams 1966; termed the “good genes”, “indicator” or “handicap model” model of preference evolution; Zahavi 1975; Zahavi 1977). Thus, generally females are choosy and males are competitive. However role reversal is also known, such as in seahorses where the male invests heavily in parental care (Vincent 1994).

The male sex typically benefits from multiple matings in the sense that more matings will directly lead to more offspring (Arnold and Duvall 1994; Kingsolver et al. 2001; Arnqvist and Rowe 2013). Females may also benefit from multiple mating (polyandry), by accruing direct and indirect benefits that drive their fitness levels (Andersson 1994). Direct benefits include higher probabilities of fertilization (Walker 1980; Sheldon 1994), a lower probability of sexual harassment (Parker 1974), the benefit of parental care from additional males (Emlen 1978; Burke et al. 1989), a lower probability of infanticide (Hrdy 1977) and acquiring ejaculate nutrients (Thornhill 1976), food and courtship food (Wolf 1975).

Females may also enjoy indirect fitness benefits in the form of genetically more viable offspring (Møller 1988; Smith 1988), genetically more diverse offspring (Williams 1975), genetically more resistant offspring (Hamilton 1990), and genetically more attractive sons

(‘sexy sons’) (Birkhead and Møller 1992).

Sexual conflict, or sexual antagonism, arises when two sexes have differing reproductive optima, more specifically over the mode and frequency of mating, potentially leading to an evolutionary arms race between males and females (Parker 1979; Holland and Rice 1998;

Chapman et al. 2003; Arnqvist and Rowe 2005).

Under sexual conflict, males attempt to outcompete other males (Clutton-Brock and

Parker 1992; Kvarnemo and Ahnesjö 1996; Parker and Simmons 1996), and in so doing, might impose a much higher mating rate upon females than is optimal for her to achieve maximal fitness. In contrast, female fitness may be better served by foraging for food to nourish developing young, or by mating with only the best mates. In this manner, male and female interests are not aligned, and conflict ensues. An example of this in the form of

4 interlocus sexual conflict occurs in the fruit fly species, Drosophila melanogaster, as males release a seminal fluid that up-regulates females’ egg-laying rate and reduces their desire to re-mate. However, this shortens a female’s lifespan, thus reducing her fitness (see;

Chapman et al. 2003). This illustrates how sexually antagonistic adaptations, in both sexes

(in this example: males), bias the outcome of interactions toward the evolutionary interests of their bearer (Parker 1979; Rowe et al. 1994; Rice and Holland 1997; Gavrilets et al. 2001).

An alternative to this form of sexual conflict may manifest as a ‘tug of war’ between natural selection on both sexes and sexual selection on either males or females individually, whereby conflict arises over the expression of a shared trait influenced by the same genetic loci in both sexes, that is subject to sex-specific selection (intralocus sexual conflict;

Bonduriansky and Chenoweth 2009). An example here may be how males and females of the cricket Teleogryllus commodus both make similar dietary choices, despite the fact that the optimal diets for each sex diverge – resulting in neither sex being able to select its optimal diet composition (Bouduriansky and Chenoweth 2009).

1.3 The economy of mating in Insecta and resistance traits

Mating can be an expensive activity for both sexes, however in most cases, females experience greater costs. This occurs not just through greater energy expenditure (in gamete production and maternal care), but also through a female’s increased susceptibility to predation, and lost opportunities for feeding. Various studies have examined the costs of mating across taxa (see Daly 1978; Walker 1980; Arnqvist and Nilsson 2000). While mating costs are ubiquitous across all organisms, in insects they have been most extensively studied in the water striders (Gerridae). Female water striders experience a

5 variety of direct costs during mating, as females must physically carry males on their backs during copulation and post-copulatory guarding (Rowe et al. 1994; Arnqvist 1997a). When females are mounted, energy expenditure is increased by some 20%, and the speed at which females can skate on the water surface is greatly reduced (Arnqvist 1989a; Fairbairn

1993; Watson et al. 1998). This cost associated with carrying the males may be why females in some species allow smaller males to guard for longer following copulation (for example see Sih et al. 2002). The decreased speed and agility of copulating females increases predation rates on pairs and reduce foraging rates (Wilcox 1984; Arnqvist 1989a; Fairbairn

1993; Rowe 1992; Rowe 1994). These costs directly impact the water strider mating dynamics, as for example, mating rates have been shown to decrease in the presence of predators (Sih et al. 1990; Magnhagen 1991), and in some instances females will stop foraging and leave the water surface during copulatory engagements (Rowe 1992; Rowe et al. 1996).

Through struggles between persistent males and reluctant females, injuries or even death of a female may occur. As male dungflies (Sepsis synipsea) cling to vigorously shaking females during pre-mating struggles, studies have shown that their clasping forelegs cause injuries to the females’ wings (Blanckenhorn et al. 2002; Mühlhäuser and Blanckenhorn

2002; Leong et al. 1993). In many water strider species, pre-mating struggles also lead to increased energy expenditure (Watson et al. 1998) and increased predation rates (nearly fivefold) in both sexes (Rowe 1994). Females typically resist mating with males by somersaulting and rolling on the water surface in their attempts to dislodge unwanted suitors (Rowe et al. 1994; Arnqvist 1997a). These struggles can last from seconds to

6 minutes and females can be harassed multiple times a day (Rowe 1992; Arnqvist and Rowe

2013).

When the costs of mating are sufficiently high, females are expected to evolve mechanisms to resist unwanted matings, through behaviour, morphology, or both. The view that females resist males as a method of “screening” out the unfit males (female choice; West-

Eberhard 1983; Crump 1988; Eberhard 1996; Eberhard 2002; Wiley and Poston 1996;

Bisazza et al. 2000; Kokko et al. 2003; Pizzari and Snook 2003) is also referred to as

“selective resistance” by Eberhard (2002), resulting from the costs of mating with poor- quality males, and is touted as an alternative to sexual conflict. The costs females experience from resisting are referred to as “costs of choice” (Pomiankowski 1988) and it is known that in most insect groups, some form of overt and potentially costly female resistance is common (Arnqvist and Rowe 2013), generally in the form of either accepting/favouring preferred males or avoiding/resisting others. The costs females experience in relation to choice result from a physical interaction with the males attempting to mate, but instead result from the loss of energy directed toward gamete production and parental care (Shuker 2014). However this is not always the case, as it has been shown that resisting copulation attempts alone can be energetic and costly (Smuts and Smuts 1993; Clutton-Brock and Parker 1995; Gowaty and Buschhaus 1998). In many cases however, female choice is thought to be less likely than sexual conflict (although both can occur at the some time), as the direct costs of sexual conflict generally outweigh the indirect benefits of female choice (Shuker 2014).

Almost 20 years ago female resistance was documented in around 15 species by over 20 authors (reviewed by Jormalainen 1998), and since then such tactics have been documented in many more. These studies add to the growing body of evidence supporting the role of sexual conflict as a significant evolutionary driver in insect mating system evolution.

1.4 Adaptations of persistence, resistance and harassment in Insecta

1.4.1 Male harassment

Circumventing female pre-mating resistance behaviour is beneficial to males when encounters between males and females involve conflict over whether to mate or not. The evolution of pupal mating in insects is thought to be a persistence trait (Deinert et al. 1994;

Brower 1997). Male fruit flies (Drosophila melanogaster and D. simulans), as an example, patrol sites where adult females emerge from their pupae. There is little chance for a female to resist mating, as the males will copulate with the newly emerged (teneral) females before their wings unfold, whilst their bodies are still soft and transparent

(Markow 2000). It should also be known that this may in fact be a form of sperm competition, where males are avoiding intrasexual competition by getting to the female first. Similarly, mating in Opifex fuscus (mosquito) occurs as adults emerge on the water surface (Slooten and Lambert 1983). As the females are close to emergence from the pupae, males search for, grasp, guard, and then copulate with the newly emerged females.

During this time, the sex ratio is heavily male biased, with one study recording some 260 males to each female (Slooten and Lambert 1983).

Harassment is viewed as a persistence trait, evolved as a response mechanism to acts of resistance. As clashes between persistent and resistant individuals can be costly and occasionally result in mutilation or death, it is anticipated that given the high-stakes of a battle, any trait that increases the effectiveness of persistence will ultimately be favoured

(Arnqvist and Rowe 2013). The funnel-web spider Agelenopsis aperta, represents one dramatic example of where males have evolved a strategy to successfully prevent female resistance and/or conflict by anesthetizing them prior to mating (Singer et al. 2000). Males approach and spray females with a toxin that causes her to enter an “unconscious” state, whereby the male then proceeds to move, reposition and finally inseminate her whilst she remains in this unaware state.

The solitary bee (Anthrophore plumipes) is another well-studied case of where male harassment rates are very high (Alcock et al. 1977; Stone 1995). In these solitary bees, females forage, construct nests and produce and feed their offspring with no paternal investment beyond fertilization. Males harass these females on their foraging flights by vigorously chasing and pouncing on them, occasionally knocking them out of flight. During peak harassment times (usually midday), males harass females at rates of up to 11 pounces per minute resulting in high costs to females. Such male mate harassment can decrease female foraging returns by 50% during (male) optimal mating attempt time periods. Krupa and colleagues (1990) found that females of various insect species, including some bees, avoid male harassment by reducing their use of the most profitable resources in the presence of males.

One of the more common persistent traits exhibited is the development of grasping structures, that are employed throughout intersexual struggles (Arnqvist and Rowe 2013).

In response, female traits that increase their capability to resist or avoid mating attempts, such as anti-grasping structures and behaviours will be favoured.

1.4.2 Secondary sexual morphologies

Secondary sexual traits (SSTs), that is, sex-specific traits typically associated with some aspect of mating/reproductive behaviour but not strictly required for reproduction, often take the form of exaggerated morphologies (Emlen and Nijhout 2000). SST in insects include male weapons, such as spurs, horns, spines, and the thickening and/or elongation of legs, antennae, mandibles, and forceps; as well as ornaments, such as the expansion and/or presence of bright, colourful patches on males (Arnqvist and Rowe 2013). While many SST are associated with male-male competition, SSTs may also arise through the sexual pressures of sexual conflict. Such conflict SSTs can be expressed in both sexes.

In females, SSTs may function to impede male grasping and mating, such as the altered dorsal textures of female diving beetles (Green et al. 2013), and the elongate abdominal spines (Arnqvist 1992) and hidden genitals (Han and Jablonski 2009) of some water striders.

Producing and maintaining such persistence and resistance traits is inevitably costly (Emlen

2001; Jennions et al. 2001). Within a population, individuals of either sex will not pay for such costs to the same extent due to individual selective pressures; thus, such traits generally exhibit remarkable variation (Wilkinson and Taper 1999). Additionally, the extent

10 of phenotypic variation in exaggerated structures is anticipated to be greater in species under more intense sexual selection, generally because of changes in condition dependence rather than a change in genetic variation for the structures (Wilkinson and

Taper 1999). In other words, the extrinsic effect of condition (e.g. from food quality during development) helps maintain high trait variance, allowing for extreme expression of these traits. This is exemplified by the extremely long eyestalks of stalk-eyed flies (Wilkinson and

Taper 1999), the forceps of earwigs (Simmons and Tomkins 1996), and the enlarged mandibles of stag beetles (Knell et al. 2004).

1.4.3 Grasping traits

Darwin (1871) first noted that there is a seemingly infinite diversity of “organs of prehension” in males that clearly function to grasp the females prior to or during mating.

In some species these are unquestionably required for the act of mating, whereas in others, they only provide some slight advantage. Darwin believed that the advantages came from preventing a female’s escape, reducing their ability to be mated by other males, or in preventing takeovers by contesting males (Darwin 1871; Arnqvist and Rowe 2013). Female resistance to copulating was recognized as a potential force in moulding the evolution of grasping devices from as early as 1907 (Edwards 1907, cited by Richards 1927). For males, any such trait that enhances a male’s success in overcoming female resistance will be favoured. Grasping traits have been most extensively studied in insects. Examples include the work done on water striders (e.g. Gerris odontogaster (Anrqvist 1989b; Arnqvist 1992);

Rheumatobates rileyi (Westlake et al. 2000), diving beetles (Bergsten et al. 2001),

11 scorpionflies (Thornhill 1980; Thornhill 1984; Thornhill and Sauer 1991), and sagebush crickets (Sakaluk et al. 1995).

Modifications of the male pregenital segments for clasping females are very common among the insects (Darwin 1871; Eberhard 1985; Arnqvist 1997a). The earliest research conducted on whether or not these clasping traits functioned as antagonistic or persistence traits centred on Panorpa scorpionflies (Thornhill 1980; Thornhill 1984;

Thornhill and Sauer 1991). Females are described to vigorously resist as males attempt to grasp their forewings with their notal organ (a large muscular set of claspers on the abdominal segment) prior to mating. Through laboratory experiments, Thornhill (1980;

1984) disabled the male’s clasper functionality by covering them in beeswax, resulting in males who were unable to overcome female resistance, consequently resulting in no successful copulatory engagements, and confirming that these structures functioned to secure the female in a copulatory hold.

A similar clasping structure is exhibited on the terminal abdominal segments of male sagebush crickets (Cyphoderris strepitans), called the “gin trap” (Sakaluk et al.

1995). To copulate, a female will climb atop the male and feed on his fleshy wings (a form of nuptial gift), preoccupying her whilst he secures her in position with this gin trap and transfers his spermatophore. Sakaluk and colleagues (1995) stated that the gin trap may function to retain females in cases where a dismount was attempted prior to spermatophore transfer. Experimentally manipulating wing availability and functionality of the gin trap illustrated, as expected, that females tended to terminate matings early when wings were partially removed (female choice), and males with the gin trap enabled were

12 observed to prolong matings compared to males with a disabled gin trap. This indicated a male adaptation to female resistance due to mate choice.

Water strider species display numerous resistance and persistence adaptations utilised during male-female interactions. Consider the antennal claspers of the water strider, Rheumatobates rileyi, which are employed during precopulatory struggles

(Westlake et al. 2000; Arnqvist and Rowe 2013). As the antennae are only used during pre- copulatory struggles, and are released as soon as the male’s genitalia are inserted into the female, they function exclusively in sexual conflict and play no role in mate guarding or preventing takeovers (Arnqvist and Rowe 2013). Across the genus Rheumatobates, species exist where almost every appendage is apparently modified for grasping (Arnqvist 1997a;

Westlake et al. 2000). Grasping structures are not only restricted to limbs and antennae.

Arnqvist (1989b; Arnqvist 1992) found that in one species of water strider, Gerris odontogaster, males have a pair of abdominal clasping processes that are employed to grasp females during vigorous pre-mating struggles, whereby males with longer claspers withstood more somersaults, therefore achieving greater mating success.

Clasping structures on insect genitalia are also common, suggesting that intersexual conflict may be involved in the remarkable diversification of genital form. Animal genitalia show the immensity of sexual selection, with rapid evolution creating such diversification in many closely related species to the extent that some species can only be identified through their genitalia (Eberhard 1985; Eberhard 1996; Eberhard 2002; Arnqvist 1997a;

Arnqvist 1998).

1.4.4 Antigrasping traits and other resistance traits

Under certain circumstances, the evolution of grasping traits that increase the efficiency of persistence in one sex are expected to be matched by the counteradaptations in the opposite sex. In water striders, experiments have been conducted to verify the efficiency of male grasping devices (section 1.6) and have identified various resistance traits in females. In water striders, females of some species have altered genital and pregenital segments that function to increase their resistance capabilities against male mating attempts (Arnqvist and Rowe 1995; Arnqvist and Rowe 2002a; Andersen 1997; Arnqvist

1997a; Arnqvist and Rowe 2013). Such traits include elongated and upturned connexival spines (also referred to as setal tufts) that flank the genitalia, and a downward bending genital opening. By manipulating the length of female connexival spines, it was shown that females with more elongate connexival spines tended to mate at lower rates than those with less elongate spines, who were less able to resist males (Arnqvist and Rowe 1995;

Arnqvist 1997a). Female diving beetles are also described to possess traits that function to reduce the efficiency of male grasping devices (see section 1.5).

An atypical form of male harassment has evolved in certain robber flies of the genus Efferia

(Diptera: Asilidae). Males of Efferia actively search through vegetation for females (Dennis et al. 1986). When spotted and approached, females generally take to the wing to escape and are pursued by males. If captured during flight, females exhibit rapid, violent resistance behaviours that often successfully dislodge the mate. However, in a few species of this genus, more elaborate resistant behaviours are used. In these species, if grasped by a male, a female will exhibit thanatosis (playing dead) (Dennis and Lavigne 1976). Once the female

14 ceases to move, the male no longer acknowledges the motionless female as a potential mate, loses interest, and so releases the female. Similarly and more recently, this method of mate avoidance has also been discovered in female moorland hawker dragonflies

(Aeshna juncea) in the Swiss Alps. Females are described to “crash-dive” to the ground while being pursued by a male and lie motionless until her suitor flies away (Khelifa 2017).

Thanatosis is recognized as a widespread antipredator behaviour in insects, suggesting that females may have been preadapted to evolve this particular form of resistance (Dennis and

Lavigne 1976).

In the literature to date, it is known that sex-limited polymorphism and alternative mating tactics are common in males (e.g. with major and minor males employing alternative mating strategies), whereas in females, polymorphism and alternative strategies are rare and when they exist they seem to be associated with sexual conflict. The main clear cases have been recorded in diving beetles, damsel and dragonflies, some butterflies and the

African bat bug (Afrocimex constrictus: Reinhardt et al. 2007) where divergent female phenotypes have evolved under sexual conflict between males and females over the optimal number of copulations. In sexually dimorphic taxa where males search for and harass females using visual cues to determine potential mates from other males, females that act or look like a male should enjoy reduced harassment (‘mate mimicry’; Robertson

1985, Hinnekint 1987, Forbes 1991, Cordero 1992, Cordero et al. 1998, Andrés et al. 2002;

Fincke et al. 2005).

This phenomenon is well documented among insects (e.g. damselflies and butterflies) with many having a sex-limited female di- or polymorphism, where one morph

15 is distinct from males in colouration and behaviour (heteromorphs), and the other greatly resembles males in these distinct traits (andromorphs) (Johnson 1975; Robertson 1985;

Cook et al. 1994). This has been well studied in Coenagrionidae damselflies, where body colour is determined by a single bi- or triallelic autosomal locus with female-limited expression (Johnson 1964; Johnson 1966; Cordero 1990; Andrés and Cordero 1999).

The ‘learned mate recognition’ hypothesis reasons that females attempt to avoid male harassment by confusing males due to variation in female signals. This benefit comes at two costs: predation rates may be higher for the conspicuous andromorphs (see, Johnson

1975, Robertson 1985), and they may suffer some risk of remaining unmated (Robertson

1985; Cordero et al. 1998; Miller and Fincke 1999; Sherratt 2001). Both of these polymorphism systems are indicative of sexual conflict, with some females avoiding mate harassment at the expense of the males, thus males not achieving their ideal number of matings (Sirot and Brockmann 2001).

Female polymorphism has also been documented in diving beetles from the family

Dytiscidea, where males possess modified structures on their forelegs that act as suction cups (see section 1.5). In both examples, these female dimorphisms and polymorphisms function to resist male harassment and optimize (for the female) the number of successful copulations, as multiple phenotypes coexist at the same time within a population.

1.4.5 Convenience polyandry

Convenience polyandry refers to a final response of females or males to persistence traits in the other sex, where the animal simply gives up, thereby avoiding further costly harassment or costly resistance (Thornhill and Alcock 1983). A well-studied example of where this is the case is in the water strider genera Aquarius and Gerris. When harassment rates are experimentally elevated or the costs of mating are reduced or heightened, females tend to reduce their resistance to male mating attempts, or simply avoid the costs associated with resistance, therefore increasing mating rates (Arnqvist 1992; Rowe 1992;

Weigensberg and Fairbairn 1994; Lauer 1996; Ortigosa and Rowe 2003).

1.5 Sexually antagonistic coevolution

Evolutionary conflicts of interest between males and females are ever-present (Chapman and Partridge 1996; Rice 1996; Partridge and Hurst 1998; Rice 2000). Such conflict is suggested to fuel the arms race between persistence and resistance traits (Chapman and

Partridge 1996; Rice 1996; Arnqvist and Rowe 2013), during which time adaptations of one sex, which are harmful or costly for the opposite sex, select for counteradaptations in the other sex to mitigate costs imposed by such adaptations (sections 1.3; 1.4; Arnqvist and

Rowe 2002a). The resulting sexually antagonistic coevolution is becoming recognised as a crucial process of evolution, with the potential to shape various interactions between the sexes (Rice 1996; Gavrilets et al. 2001; Civetta and Clark 2000) and their gametes (Gavrilets

2000; Swanson et al. 2001), as well as to drive diversification (Arnqvist 1998), speciation and extinction rates (Rice 1998; Parker and Partridge 1998; Arnqvist et al. 2000).

Theory suggests that the net outcome of antagonistic male-female interactions should remain relatively unchanged. In theory, during an arms race, as one sex builds up arms, the opposite sex is expected to balance this with the build-up of arms itself, notwithstanding periods of escalation and de-escalation (Parker 1979; Parker 1983; Härdling 1999). Thus, sexually antagonistic coevolution is inherently difficult to detect, as the consequences of an arms race on sexual interactions may be virtually undetectable (Chapman and Partridge

1996; Rice 1996; Partridge and Hurst 1998; Rice 2000).

In keeping with this expectation of evolutionary balance in the level of arms between the sexes, one sex may briefly evolve a competitive advantage relative to the opposite sex

(Parker 1979; Parker 1983; Härdling 1999; Arnqvist and Rowe 2002a). In instances where male and female antagonistic traits are mismatched, the evolutionary costs of the arms race for interactions between the sexes may be exposed. An example of this may be a morphological and/or behavioural race that benefits males but not females to mate multiple times (Parker 1979), where it would be apparent that the advantage has shifted toward males, allowing high rates of mating within the species (Arnqvist and Rowe 2002a).

The counter situation would be where the advantage has shifted toward females. Testing these species-specific adaptations in regard to mating may uncover sexually antagonistic adaptations through the effect of the change in sexual interactions caused by evolutionary change in the relative levels of arms between males and females. Studies of diving beetles, water striders and bedbugs provide evidence of sexually antagonistic coevolution and illustrate how the sexes are working to maximize their own reproductive success (Parker

1979; Arnqvist and Rowe 2013). A problem in identifying such conflicts is that in practice it may be problematic to interpret when female resistance to male mating attempts is being used as a screening device to choose the most vigorous males (female choice), and/or when it is being used to avoid costly, unwanted matings (sexual conflict).

Within various species of diving beetles (Dytiscidae), females act to repel male mating attempts by repeatedly diving when an attempt to grasp them is made (Aiken 1992;

Bergsten et al. 2001). This vigorous resistance behaviour suggests that there is a cost associated with mating and/or multiple matings for females (Arnqvist and Rowe 2013).

Aiken and Khan (1992) described adhesive structures on the male tarsi that assist them in grasping females along their dorsal surface. These suction cups are used to capture and hold onto females during copula as well as during mate guarding, which proceeds after a long and exhausting pre-mating struggle, resulting in advantageous males increasing their mating rates (Wichard et al. 2002; Miller 2003; Green et al. 2013). Interestingly, several

Dytiscinae females have likewise evolved conspicuously altered dorsal surfaces that appear designed to be more difficult to grasp thus increasing female control during mating

(Arnqvist and Rowe 2013; Green et al. 2013).

From an originally smooth dorsal surface, females have evolved alternative phenotypes consisting of a ridged dorsum covered with elongate grooves, as well as increased setal density within these dorsal furrows, making it more difficult for the males’ suction cups to grasp and remain attached to the females (Bilto et al. 2008; Inoda et al.

2012). Male suction cups and female dorsal modification are phylogenetically correlated with each other in the genus Acilius; as dorsal sculpturing increases in females an increased

19 differentiation between large and small suction cups on male protarsi follows (Miller 2003;

Bergsten 2005; Bergsten and Miller 2007). Interestingly, even within populations of a single species both male and female phenotypes may exist, with a higher prevalence of sculptured females associated with a higher proportion of large plus small-cupped males.

A depletion of female dorsal sculpturing is additionally followed by a reversal of the male suction cup morphology (Bergsten and Miller 2007; Green et al. 2013). This phylogenetic pattern strongly suggests that sexually antagonistic coevolution between these traits occurs and was described as the first comparative evidence of an ongoing arms race produced as a by-product of sexual conflict over mating (see Bergsten et al. 2001; Bergsten and Miller 2007; Arnqvist and Rowe 2013; Buzatto et al. 2014).

Miller (2003), described the coevolution of these male and female morphological adaptations through a taxonomically broad study which included 52 species of diving beetle. The central aim was to indicate the order of evolution of persistence and resistance traits to determine whether they fit the expectations of an evolutionary arms race driven by sexual conflict over mating (Miller 2003; Arnqvist and Rowe 2013; Buzatto et al. 2014).

Through mapping traits of interest onto the phylogeny, Miller (2003) was able to illustrate an arms race over evolutionary time. Evidence indicated that males were the first to develop adhesive grasping structures, followed by females evolving anti-grasping traits, presumably as a method of countering the grasping traits possessed by males (Miller 2003;

Bergsten and Miller 2007; Arnqvist and Rowe 2013).

Sexually antagonistic coevolution is also observed in traumatically inseminating insects, such as bedbugs (Cimicidae; Schuh and Stys 1991), the plant bug genus Coridromius

(Tatarnic et al. 2006; Tatarnic and Cassis 2010) and various other heteropteran insects

(Stutt and Siva-Jothy 2001; Tatarnic and Cassis 2008; Tatarnic et al. 2014). Mating via traumatic insemination (internal insemination via puncturing the female’s cuticle and ejaculating into the body cavity, without involvement of the female’s genitalia) (Stutt and

Siva-Jothy 2001), illustrates how males may have evolved to overcome female resistance strategies or to bypass sperm competition (driver behind this phenomenon is unknown), and how females are counter adapted to mitigate the costs of this form of mating. As this mechanism of copulation bypasses the normal mechanics associated with copulation and sperm transport and causes overt damage, females experience more extensive costs (for example, excess leakage of blood and/or an increased risk of infection) (Stutt 1999). These systems provide one of the most interesting systems for assessing mating costs and sexual conflict, as females have evolved a “secondary” reproductive structures (paragenitalia: see

Davis 1965b; Carayon 1966), to mitigate the costs associated with unwanted male penetration and maximize their reproductive success (Ryckman and Ueshima 1964; Davis

1965a; Davis 1965b; Usinger 1966; Walpole 1988; Newberry 1989; Stutt and Siva-Jothy

2001; Morrow and Arnqvist 2003; Reinhardt et al. 2003).

1.6 Water striders as a model taxon

Sexual conflict over mating and has been widely documented in water striders (Gerridae), as evidenced produced by comparative research on the genus Gerris, indicates an arms race between the sexes (see Arnqvist 1989b; Arnqvist 1992; Arnqvist and Rowe 1995;

Arnqvist 1997a; Andersen 1997; Arnqvist and Rowe 2002a; Arnqvist and Rowe 2002b,

Rowe and Arnqvist 2002). In this group of semi-aquatic insects, females experience

21 significant costs associated with matings: males frequently harass females, and resistance by females can be vigorous (reviewed by Rowe et al. 1994; Arnqvist 1997a).

Under the arms race hypothesis, it is expected that where exaggerated grasping traits evolve, antigrasping traits will rapidly coevolve (Arnqvist and Rowe 2013). Through geometric morphometric methods, Göran Arnqvist and Locke Rowe (2002b) examined 16 species within this genus and found exactly that. Furthermore, they determined that the presence and degree of resistance and persistence traits present in different species were involved in determining mating rates, with occasional mismatches between the sexes

(Arnqvist and Rowe 2002a, Rowe and Arnqvist 2002).

The degree to which males and females are armed is strongly associated with mating duration and frequency in regards to copulatory success, both of which are highly variable

(Arnqvist and Rowe 2013). In most coevolutionary arms races, the sexes will be competing for control over mating, with antagonistic traits in one sex rapidly met with counteradaptations in the other, thus there may be no significant winner of the sexes in combat

(Parker 1979; Parker 1983; Rice and Holland 1997). As this pattern is common and studies generally observe only current interactions between the sexes, arms races may often be hidden from observation (Chapman and Partridge 1996, Rice 1996, Rice 2000, Härding et al. 2001, Arnqvist and Rowe 2002a, Arnqvist and Rowe 2002b). Arnqvist and Rowe (2002a) examined multiple species of water striders, focusing on those that deviated slightly from this balanced pattern of coevolution in order to uncover the conflict between sexes. They

22 anticipated that in those species where males were more armed than females, mating rates would be high. Likewise, where females were more armed than males, mating rates would be low (Arnqvist and Rowe 2002a; Arnqvist and Rowe 2013). Through comparisons of female mating activity in either male-biased settings (3:1) or an even sex ratio (1:1), evidence that mating rates increase with male armature was found, where the effects of male success in harassment and mating rates were dramatic (Arnqvist and Rowe 2002a;

Arnqvist and Rowe 2013).

Once again, studies of water striders; specifically Gerris odontogaster, beautifully disentangle the conflicting explanations of whether resistance/persistence traits are acting as a form of mate choice, and/or are being used to avoid costs associated with unwanted matings (sexual conflict) (Arnqvist 1992). Waters strider matings are rapid, chaotic engagements where males pounce on females and aggressively attempt to achieve genital coupling (Arnqvist 1992). Females attempt to dislodge males via backward somersaults and other sporadic movements, whilst the male attempts to withstand her resistance behaviours using various grasping mechanisms on the antennae, legs, abdomen or genitalia. Arnqvist (1992) assessed whether the female’s acts of resistance are a form of mate choice or an attempt to avoid the costs of excessive copulations, which lead to an increased risk of predation, energetic expenditures, and reduced foraging success. By manipulating sex ratios and density in a study population of G. odontogaster, he was able to show that females became less reluctant and mated more frequently as male density increased, which is consistent with predictions of sexual conflict (i.e. females modulate their resistance according to expected costs) but not what one would expect under female choice.

Insects owe a great deal of their success to their ability to fly. Nevertheless, secondary reduction and loss of the wings is relatively frequent in many groups of insects, including this family of semiaquatic and aquatic bugs (Menke 1979; Andersen 1982; Schuh and Slater 1995). Unlike the case in butterflies, wasps, and flies, flight does not form part of the everyday activity of most water bugs, but is probably required for dispersal between overwintering sites and breeding (Andersen and Weir 2004). In winged insects the development of wings and wing muscles is an energy-demanding process which has to compete with other life processes, especially the production of eggs in females. When flight is no longer necessary for survival there is a distinct trend toward flightlessness. This is most clearly demonstrated in marine water striders that live in an extremely stable environment. Adult sea skaters (Halobates), for example, always lack wings and wing muscles completely (Andersen and Weir 2004). Most semiaquatic bugs (Gerromorpha) living in fresh water habitats are wing polymorphic, with two or more distinct adult wing morphs. These insects live in habitats ranging from extremely stable (the sea, rivers and large lakes), intermediate (streams, ponds), to extremely unstable or temporary (streams, pools, and marshes that dry up during summer).

Taxonomic literature also reveals polymorphism in other gerromorph traits, most notably in secondary sexual traits in the Veliidae, where single species such as Microvelia childi and Microvelia woodwardi were described in two forms i.e. some with and some without black bristles or setal tufts (Andersen and Weir 2003), however the likely function or evolutionary reason for this polymorphism has not been determined within this group of insects. Species which display intrasexual polymorphism make interesting candidates for studying sexual conflict and SSTs, as the costs of repeated matings and the selective

24 pressures may differ between the two forms, influencing the development of resistance and persistence traits.

1.7 This study

1.7.1 Project significance and overview

As discussed above, there have been several studies of the role of male-female antagonistic interactions in shaping morphological and behavioural evolution in invertebrates. This research continuously adds to our understanding of how costs and benefits can shape the evolution of the sexes. Much of the research into sexual conflict has focused on only a few study systems, such as the Hemipteran family Gerridae, the water striders, which have been the focal taxa of an extensive amount of research involving sexual conflict, sexually antagonistic coevolution, resistance and persistence traits (mentioned in sections 1.3;

1.4.2; 1.4.3; 1.4.4; 1.4.5; 1.5; 1.6). Sister to the Gerridae is the family Veliidae, (also known as small water striders, ripple bugs, or riffle bugs). Like water striders, veliids live on the surface of the water, and presumably share similar selective pressures, including those which are thought to underpin sexual conflict in water striders (e.g. increased risk of predation when mating, restricted foraging and ovipositing sites: Eberhard 2006). Unlike the Gerridae the Veliidae are relatively understudied, and despite exhibiting considerable sexual dimorphism consistent with sexual conflict and sexually antagonistic coevolution,

(Cassis et al. 2017), including female-limited polymorphism in secondary sexual traits (N.

Tatarnic, pers. obs.), they have not been studied in the context of sexual conflict.

This thesis aims to describe biological adaptations utilised by Veliidae in direct response to sexual conflict. Taxonomic work has identified various potential grasping and anti-grasping structures for various species within the Veliidae family as they resemble modifications expressed by various gerrids (Table 1). However, to date the likely functions of these structures have not been tested. In a revision of the genus Nesidovelia, Cassis et al. (2017) noted that several male and female structures (e.g. setal tufts of hair, converged connexiva, concealed genitalia, male abdominal processes) that may function in sexual conflict.

In light of this literature, two Western Australian species from this genus (Nesidovelia), as well as one species from the genus Microvelia, also from the Veliidae, were chosen as study species. The use and function of male and female putative secondary sexual traits as identified in Cassis et al. (2017) will be compared between apterous and macropterous morphs of three species (Microvelia oceanica, Nesidovelia peramoena and Nesidovelia herberti) throughout mating interactions, as well as explore how size dimorphisms

(morphological size differentiations of sexually mature males and females: Fairbairn 1997) between sexes, morphs and species may relate to sexual and natural selection. This research forwards our knowledge of the Microvelia and Nesidovelia and provides an insight into the conflict ever-present between males and females in the race to optimize their reproductive success. Furthermore, this research has also developed a unique technique for snap freezing and imaging these animals during pre-mating struggles and during copulation.

The following sections review the current status of literature on the Veliidae, the contribution this thesis will make to the scholarly knowledge of this family and describe the groups and species within Veliidae that are the subjects of this thesis.

Table 1: Review of 12 published papers highlighting the scope of knowledge presented on Australian Veliidae. This systematic review was performed in the late months of 2016 to the early months of 2017 and databases such as: Web of Science, Scopus, and JSTOR, were searched for relevant literature.

Papers Title Coverage of Search terms used in published publication electronic databases Taxonomy Cassis et al. 2017 Phylogenetic reclassification and genitalia morphology Phylogenetic analyses of the Australasian Microveliinae is given, including a reanalysis of Andersen and ‘Veliidae’, ‘Microvelia’, of the small water strider genus Nesidovelia Adnersen Weir (2003) dataset and an analysis of a new dataset of 46 morphological characters and 42 taxa. ‘Australia’, ‘semi-aquatic & Weir and allied Microveliinae (Hemiptera:Veliidae) Analyses focused on pregenital, genital and postgenital structures of males and females, including bug’, taxonomy’, ‘biology’, putative male grasping and female anti-grasping structures. ‘evolution’, ‘sexual conflict’. Andersen and Weir The genus of Mircovelia Westwood in Australia The subgeneric classification of the genus Microvelia based on the results of a phylogenetic analysis using 2003 (Hemiptera: Heteroptera: Veliidae). maximum parsimony, describe three new subgenera and redescribed all previously known Australian species of the genus. Andersen and Weir New genera of Veliidae (Hemiptera: Heteroptera) from The generic classification of the subfamily as well as describe three new genera for species previously 2001 Australia, with notes on the generic classification of the classified in the genus Microvelia Westwood as well as three other new genera and nine new species. subfamily Microveliinae. Andersen and Weir The marine Haloveliinae (Hemiptera: Veliidae) of The marine Haloveliinae (Veliidae) of Australia, New Caledonia and southern New Guinea. They are 1999 Australia, New Caledonia and southern New Guinea. classified in two genera, Xenobates Esaki and Halovelia Bergroth. Descriptive notes are presented for five new species of Halovelia recorded from Australia. Polhemus 1982 Marine Hemiptera of the Northern Territory including This paper deals with marine and related Hemiptera found in the Darwin area, Northern Territory. Two the first fresh-water species of Halobates Eschscholtz subgenera are discussed: Hilliella China (genus Halobates) and Colpovelia subgen. n. (Gerridae, Veliida, Hermatobatidae and Corixidae). (genus HaloveliaBergroth). Eight species are discussed, 2 of them new: Halobates (Halobates) acherontis and Halovelia (Colpovelia) angulana. The world's first freshwater Halobates and the first Australian corixid from marine waters are reported. Malipatil 1980 Review of Australian Microvelia Westwood All known species of Microvelia occurring in Australia are reviewed, including Tasmania and Lord Howe (Hemiptera-Veliidae) with a description of 2 new Island. Two new species, M. distincta and M.fluvialis, and a new subspecies. M. fluvialis weiri, are species from Eastern Australia. described from eastern Australia. Available bio-ecological information is provided for five species and a key is given to species and subspecies. Biology Booth 2016 Things we do for genes Illustrates some of nature’s less kindly family interactions. ‘Veliidae’,‘Microvelia’, Jones et al. 2010 Extreme costs of male riding behaviour for juvenile Investigated the effect of male presence during juvenile development for the female Zeus bugs. ‘Australia’, ‘semi-aquatic females of the Zeus bug. bug’, taxonomy’, ‘biology’, Arnqvist et al. 2007 The extraordinary mating systems of Zeus bugs Provide natural history details of the mating behaviour for two Zeus bug species. ’evolution’, ‘sexual conflict’, (Heteroptera: Veliidae: Phoreticovelia sp.). ‘genialia’ ‘antagonistic’. Arnqvist et al. 2006 Sex-role reversed nuptial feeding reduces male By experimentally occluding the dorsal glands in females and varying food availability, it was shown that kleptoparasitism of females in Zeus bugs (Heteroptera; nuptial feeding by females reduces the extent to which the males kleptoparasitize their mates. Veliidae). Arnqvist et al. 2003 Insect behaviour: Reversal of sex roles in nuptial It is shown that females of the extraordinary insect Phoreticovelia disparate provide food for males feeding. during mating. This indicates that nuptial feeding may not be related to parental investment. Andersen 1999 The evolution of marine insects: phylogenetic, Indicates, through fossil records, that marine habitats were invaded by members of the families Veliidae ecological and geographical aspects of species diversity and Gerridae earlier that 20-30 and 45 million years before present, respectively. Suggests that in marine water striders. physiological and behavioural rather than morphological specializations were likely to have been the key innovations in the transition from limnic to marine habitats.

1.7.2 Study species

1.7.2.1 Veliidae, fam. Microvelia, gen. Nesidovelia, gen.

The Veliidae are a sister-family to the Gerridae. The veliid subfamily Microveliinae is distributed world-wide, and comprises 37 genera, with a total of around 350 described species that are primarily characterised by their unique one-segmented fore tarsi and two segmented middle and hind tarsi (Andersen and Weir 2003).

Microvelia Westwood (Appendix 1) is one of the largest genera of water striders and their allies (Hemiptera-Heteroptera, Gerromorpha) in the world with around 120 species

(Andersen 1982; Andersen and Weir 2001). Within Microvelia, the Australian clade previously identified as Microvelia (Austromicrovelia) by Andersen and Weir (2003) has been recently revised and formally identified as the genus Nesidovelia (Cassis et al. 2017).

All species within these two genera are small or very small predatory bugs that inhabit near-shore areas of standing or slow moving fresh water. Some species are often found in temporary habitats, such as puddles and pools filled with rainwater. Others occur specifically in rainwater pockets between the leaf axils of epiphytic plants (Bromeliaceae;

Polhemus and Polhemus 1991), burrows made by intertidal crabs (Polhemus and Houge

1972), and in tree holes (Laird 1956; Polhemus 1999). These insects are most commonly found in tropical and subtropical regions of the world and occur on many oceanic islands.

The first species of Microvelia recorded from Australia was M. australica (Bergroth 1916) and recently, fossil species of Microvelia have been described from 15-30 million-year-old

Dominican amber (Andersen 2000; Andersen 2001).

For this study, three species (N. peramoena, N. herberti and M. oceancia) were chosen based on their proximity to Perth, Western Australia (Andersen and Weir 2001; Andersen and Weir 2003).

Nesidovelia (Appendix 2) is comprised of 29 described species, all of which are endemic to

Australia. Most were originally placed in one of four species groups that are: the mjobergi- group, the angelesi-group, the peramoena-group, and the fluvialis-group (Andersen and

Weir 2003). For the purposes of this thesis, only specimens from the peramoena-group were chosen primarily due to their local availability (see: Table 2). Microvelia (Pacificovelia)

(Appendix 4), at present is comprised of 13 described very small to medium-sized species.

The point of distinction between the two studied genera is that within Microvelia

(Pacificovelia) the male’s abdominal venter is simple, rather than being modified as in

Nesidovelia (Appendix 3). It is predicted that these differences in male abdominal armature may translate into differences in mating and pre-mating behaviour between the two genera.

Table 2: Distributions of the four species groups of Nesidovelia, as defined by Andersen and Weir (2003).

Species group Locations recorded mjobergi-group QL, NSW, VIC, angelesi-group NT and Emma Gorge (Kimberly region of WA) peramoena-group SA, ACT, NSW, QL, SA, TAS, WA fluvalis-group NSW, QL

1.7.2.2 Species overview

Nesidovelia peramoena (Appendix 5) is the most common and widely distributed species in this genus and is recorded from all Australian states (Figure 1A; Andersen and Weir

2003). Sister to this species is N. herberti (Appendix 5), which is widely distributed in the

30 northern parts of Western Australia, Northern Territory and Queensland (Figure 1B;

Andersen and Weir 2003). Microvelia (Pacificovelia) oceanica (Appendix 5), the third species examined, is found within the Australian Capital Territory, New South Wales

(including Lord Howe Island), Norfolk Island, Northern Territory, Queensland, South

Australia, Tasmania, Victoria, and Western Australia (Figure 1C; Andersen and Weir 2003).

All three species are found in a variety of habitats, such as farm dams, roadside pools, edges of lakes, backwater pools of streams, stagnant pools in stream beds, along the edges of small ponds and lakes with and without rocks, and are sometimes able to colonise temporary habitats (Hale 1926; Malipatil 1980).

Figure 1: A) Distribution of Nesidovelia peramoena. B) Distribution of Nesidovelia herberti. C) Distribution of Microvelia (Pacificovelia) oceanica. From Andersen and Weir (2003).

1.7.3 Life cycle

Little is known of the biology of Australian Veliidae, with little attention paid toward their overwintering strategies, even though their manner of spending the winter influences their life-history traits dramatically. Work conducted on Velia caprai in the Czech Republic, found that these bugs can overwinter in both an adult and an egg stage, a particularly unique trait among semiaquatic bugs (Ditrich and Papáček 2009), as most gerromorphans are thought to overwinter as adults on land, sometimes far from water bodies, in forest leaf litter, close to roots of plants or under stones (Saulich and Musolin 2007). Most females of Velia caprai reach sexual maturity before overwintering, mate with mature males,

31 overwinter actively on the water surface and lay eggs during the winter (Ditrich and

Papáček 2009). These dormant eggs will then be induced to hatch by increasing temperatures as the season turns (Ditrich and Papáček 2009). However, experimental studies of the species Microvelia douglasi suggest that photoperiod may be the main trigger in triggering and ceasing reproductive diapause (Muraji and Nakasui 1989). Though collection records of Nesidovelia specimens exist from all seasons, it is suspected that they too are undergoing reproductive diapause during the winter months. For this reason, specimens for this study were only collected during the warmer months of the year to ensure their sexual maturity.

1.8 Study approaches and objectives

The overall objective of this project is firstly to investigate sexual conflict over mating in N. peramoena, N. herberti and M. oceanica, and test the function of putative antagonistic traits proposed in Cassis et al. (2017). More specifically, this project aims to document antagonistic behaviour between the males and females of these three species and test the function of the putative abdominal clamp in the genus Nesidovelia, using N. peramoena as the experimental organism. This project also aims to document and assess the prevalence and variation present between study species, between sexes, and between apterous and macropterous morphs (when available) in a range of morphological traits. In particular it is intended that the prevalence of these traits be documented in populations using more refined measuring techniques (micro-CT) than in Andersen and Weir (2003; microscopy).

Firstly, in relation to potential grasping and anti-grasping structures (putative secondary sexual antagonistic traits), with specific consideration of how wing morphs related to secondary sexual characters in both sexes. Secondly, in relation to a range of body and egg

32 measures (e.g. length and width measures of select body parts, female egg volume and count) to compare size relationships and egg measures between sexes, morphs and species.

In order to assess the morphological variations between male and female, winged and wingless forms, species, sex and morph representatives were scanned via micro-CT x-ray to be measured and categorised according to their external and internal characteristics.

Measurements across multiple representatives from each morph type were taken to typify the scaling relationships of structures by morph, and were examined to identify potential energetic trade-offs in different morphs (e.g. do winged morphs trade off sexual traits in exchange for flight?). Male and female structures that are considered as putative antagonistic structures (male segment 7 and 8 modifications, female connexival elevation and setal tufts; Appendix 6) were also examined and in consideration of their potential function with regards to mating and pre-mating behaviour.

As well as examination of individual specimens, paired specimens instantaneously frozen in copula will also be compared in order to identify where and how inter-sexual structures are connecting. Imaging connected pairs will help to identify how and when various male and female secondary sexual structures interact and thus allude to their functional significance. This aspect of the study follows the methodology of Arnqvist (1989b), and

Ingram et al. (2008). To assess the functional role of male abdominal processes (MAPs) in

Gerris odontogaster (Gerridae), Arnqvist (1989b) instantaneously froze a number of mating and struggling pairs in liquid nitrogen, and glued them together to preserve their positioning. Pairs were then dried and imaged using a scanning electron microscope (SEM).

Similarly, to study the morphological interactions between male and female sepsid flies (T. superba and T. lucida) in copula, pairs were frozen in liquid nitrogen as soon as the male mounted the female, placed in a -35°C freezer to allow the nitrogen to evaporate off, transferred into chilled 90% ethanol until they were prepared for SEM (Ingram et al. 2008).

As in the above studies, I froze mating pairs of N. peramoena in copula and imaged them using a micro-CT x-ray.

As in Bergsten and Miller’s (2007) study of diving beetles, video footage of interacting pairs, set up in plastic aquaria within laboratory conditions was also utilised to further discern how males and females interact and what structures may be used during pre-mating struggles as well as during mating. Contingent upon specimens available, behavioural comparisons were also made between the reproductive behaviour of the different morphs

(apterous and macropterous) of each species. This method, when paired with snap- freezing interlocked pairs, may aid in illustrating the mechanisms of a possible coevolutionary arms race within this family of insects (Veliidae).

Preliminary attempts to replicate the methods of Arnqvist (1989b) and Ingram and colleagues (2008) presented difficulties, with pairs separating during flash freezing, and thus a necessary first step of this project was to refine the methodology of freezing pairs using liquid nitrogen. This refinement will be useful in future studies of any very small to medium sized semi-aquatic insect, and therefore will represent a significant contribution to this field.

As mentioned previously, the overall aim of this thesis is to identify whether sexual conflict is present within the family of Veliidae, and to determine how male and female secondary sexual structures interact prior to (pre-mating struggle), during, and post copulation

(postmate guarding). The techniques mentioned (imaging, measuring, snap-freezing and videoing) were employed to address the specific aims of the project, which are as follows.

i. To document distinct putative secondary sexual traits (females connexival

convergence, female setal tufts presence, female proctiger orientation, the

male ventral abdominal processes on segment 8, the midline concavity of male

segment 7 and the size of the male forelimb grasping comb) between the three

study species (Microvelia oceanica, Nesidovelia peramoena, N. herberti).

ii. To assess the prevalence and polymorphism of these secondary sexual traits

across multiple individuals of each species type (sex, morph, species

combination: where sample size permits).

iii. To determine how the wing morphs relate to each secondary sexual character

in both sexes.

iv. Compare size and shape relationships and egg measures between sexes,

morphs and species.

v. Refine the methodology of Arnqvist (1989b) and Ingram et al. (2008), in order

to precisely freeze engaged male-female pairs prior to (pre-mating struggle),

during, and post copulation (post-mate guarding).

vi. Examine the uses of these antagonistic traits during pre-mating struggles and

during copulation (through video footage and micro-CT scans of snap-frozen

engaged pairs).

35 vii. To quantify resistance and persistence behaviours during pre-mating struggles

and during copulation.

2.0 Chapter Two – Methodology

2.1 Introduction

This chapter will describe the procedures, protocols and equipment used in order to accurately describe the biological adaptations utilised by Veliidae in direct response to sexual conflict, how these adaptations differ throughout the genera of Microvelia and

Nesidovelia and how these traits may interact with the opposite sex prior to, during and post-copulation. Males and females, including winged (when present) and apterous morphs of Nesidovelia peramoena, N. herberti and Microvelia oceanica, were examined.

These three species were all chosen due to the accessibility of populations from Perth,

Western Australia. All three were subject to the same methods of collection, storage, and imaging, while mating behaviour was observed and recorded for N. peramoena and M. oceanica.

2.2 Collection and maintenance of study species

2.2.1 Field locations

Nesidovelia peramoena, N. herberti and Microvelia oceanica were all collected in Western

Australia (Table 3) during the warmer seasons of spring, summer and autumn, in accordance with Department of Parks and Wildlife permits issued to the WA Museum (Reg

17: 08-000214-1). Specimens were collected during these months as this is when the insects were most active and likely to exhibit mating behaviour, as opposed to colder months, when adults may be present but reproductively inactive (see section 1.7.4).

Collecting localities were identified based in part on collection data provided by Andersen

and Weir (2003), which includes maps (Figures 1A, 1B, and 1C), field collection notes and

GPS coordinates.

Specimens were collected between dawn and dusk, and were found along the edges of still

eddies and pools along flowing streams, or within the reeds/leaf-litter of ponds where the

water was shallow and stagnant. Individuals were taken from the water’s surface using 50

mL plastic falcon vials, which were submerged adjacent to the insects, sucking them in. The

water was then drained and the bugs were transferred to dry containers lined with

absorbent cotton or paper towel. In order to minimise direct human contact this was done

by coaxing the bugs to jump off the falcon vial into the container or by transferring them

using an aspirator (made from transparent plastic hose with a filter placed 10cm from the

end to stop animals passing through). Males and females were separated immediately

after capture, and stored in separate containers. They were transported in an enclosure

without water to allow the insects’ hydrophobic hairs to dry and to prevent drowning,

while damp blotting paper was provided for the animals to drink from, rest on or hide

under.

Figure 2: A) Bow Bridge, Walpole (-34.9677, 116.9537). B) Hamelin Station Reserve (-26.4268, 114.1959). C) Kalbarri National Park (-27.5555, 114.4545). D) Caladenia Lake, Perth (-32.0713, 115.8349). Original Image Source: Paige Maroni.

Table 3: Collecting details for the three study species (Nesidovelia peramoena, N. herberti and Microvelia oceanica).

Nesidovelia peramoena Nesidovelia herberti Microvelia oceanica Location collected Bow River (Figure 3a) Hamelin Station (Figure Murchison River (Figure 3c) Chelodina wetland (Figure 3b) 3d) Latitude -34.9677 -26.4268 -27.5555 -32.0713 Longitude 116.9537 114.1959 114.4545 115.8349 Year collected 2016 – 2017 2016 2016 2015 - 2017 Month/s collected October – April August August January - April August - November Region of WA Walpole-Nornalup National Hamelin Station is a Kalbarri National Park is Located on the western Park. Located just over 202,000 hectare former located 485 kilometres third of the Swan Coastal 400km south of Perth. Bow sheep station(2) , situated north of Perth within the Plain, situated at the river is linked to an on 32 kilometres of coast mid-west region of WA. interface of two dune extensive river and line bordering the Shark systems within bushland wetland network(1). Bay World Heritage area. that was once part of an Now a conservation extensive eucalypt and reserve, it is located in the banksia woodland(6). mid-west region of WA (730km south of Perth). Fringing vegetation Corymbia calophylla, Acacia chartacea, Acacia Melaleuca rhaphiophylla, Platytheca galiodes, Eucalyptus jacksonii, galeata, Acacia roycei, Acacia saligna, Oxylobium Hypocalymma Eucalyptus guilfoylei, Eucalyptus fruticosa, lineare, narrow-leaved angustifolium, Astartea Eucalyptus diversicolor(3). Eucalyptus obtusiflora, oxylobium, Viminaria fasciculari, Acacia Triodia bromoides , Triodia juncea(4). pulchella, Euchilopsis, plurinervata, Atriplex Melaleuca raphiophylla, vesicaria. As well as weeds Melaleuca preissiana and including Lycium Banksia littorallis(6). ferocissimum, Tamarix aphylla, Prosopis spp., Argemone ochroleuca, Parkinsonia aculeata(3).

Water body Bow river flows from the Adjacent to the station’s The second largest river in A freshwater outcrop of Kent River Water Catch homestead is a bore water WA. It flows for about the water table that was Forest and is a permanent, lake, from which a small 820km from the southern artificially deepened(6). slow flowing river. creekflows for a few edge of the Robinson kilometres across the Ranges to the Indian Ocean property. at Kalbarri(5). 1) DPaW Walpole Nornalup National Park (2017) 2) Bush Heritage Hamelin (2017) 3) DPaW Nature Map (2017) 4) Chalmers and Wheeler (1997) 5) DPaW Kalbarri National Park (2017) 6) Dell and Bennett (1986)

2.2.2 Laboratory colony

Once in the laboratory, collected specimens were housed in water-filled plastic containers

with water pumps for aeration, as per the protocols of Andersen and Weir (2004). A

mixture of their in-situ water and cooled, boiled tap water was used. The enclosures

39 incorporated both water and exposed platforms for resting, made using styrofoam cuttings. The insects were fed Drosophila flies every second day, usually at a ratio of one fly to every three veliids.

2.3 Comparative morphology

2.3.1 Micro-CT scanning

In order to assess morphological variation and to compare secondary sexual traits between the three study species, between the sexes and between the winged and wingless morphs, specimens were imaged using X-ray microtomography (micro-CT). Scans were conducted using a Zeiss XRadia 520 Versa at the Centre for Microscopy, Characterisation and Analysis

(CMCA), located on the Crawley Campus of the University of Western Australia. Micro-CT scans allowed for the examination of fine details of each individual, which could not be accurately measured or examined through microscopy alone. Subsets of individuals from each sex of each species were scanned, including pairs of N. peramoena frozen in copula using liquid nitrogen. This included specimens that had died in transport, whilst in the laboratory, or during mating trials.

Dead specimens were first placed into 100% ethanol and were temporarily kept in a -15°C freezer. In order to improve scan contrast, specimens were stained by submersing them in

I2E (1% iodine in 100% ethanol) for 24 hours. They were then removed from the I2E and returned to 100% ethanol prior to scanning.

For scanning, specimens were placed into a modified 1ml plastic transfer pipette (with bulb removed) filled with 100% ethanol (Figure 3) and sealed at the base with blue tack. Multiple specimens were mounted in the pipette, separated from one another with cotton tissue.

Once the specimens were in place the top of the pipette was sealed with blue tack. Bubbles were removed by gently tapping the tube before mounting it on the XRadia specimen stage.

By taking and combining a series of images of a specimen rotated through 360, micro-CT produces high resolution, 3- dimensional images of small specimens. For each specimen two consecutive scans were undertaken. The first short scan

(~ 10-15 minutes, ~200 images) was used to determine optimal settings, including the centre (point of rotation) for each specimen scan, while the second longer scan (~120 Figure 3: Photograph of micro-CT specimen preparation. Specimens are minutes/specimen, 2000 images) produced the final datasets. placed into a 1ml plastic transfer pipette (with the concave tip Scans were run using the 4X magnification lens and no filter, removed), filled with 100% ethanol, sealed at either end with blue tack. Specimens are separated with cotton with a voltage of 70 kV and power of 5.7 Watts. tissue.

2.3.2 Digital image preparation

Using the XRadia software, the resulting dataset of stacked images was then converted into a DICOM (Digital Imaging and Communications in Medicine) dataset. DICOM is a standard for handling, storing, printing, and transmitting information in medical imaging. This was then rendered using the volume-rendering program Drishti 2.6.1. Drishti is a free, multi-

41 platform, open-source volume-exploration and presentation tool, produced by the

Australian National University (https://sf.anu.edu.au/Vizlab/drishti/). It is used to analyse these volumetric datasets and aids in visual exploration of the volume being studied. In this manner, accurate 3-dimensional renditions of each scanned specimen could be observed, measured and imaged.

2.3.3 Putative secondary sexual traits

From the 3-dimensional renderings of each specimen, male and female traits deemed to be secondary sexual traits (i.e. non-genital) by Cassis et al. (2017) were identified and categorised according to their presence, shape and position. In females such traits included the connexival elevation and degree of convergence (see: Appendix 6.1) above the midline of the body, the proctiger orientation (Appendix 6.2), and the presence of setal tufts

(connexival spines) intersegmentally between abdominal laterotergites 6 and 7 and/or medially or posteriorly on female abdominal laterotergite 7 (Appendix 6.3). Male secondary sexual traits observed included the ventral process on segment 8 (Appendix 6.4), the midline concavity of segment 7 and the size of their forelimb grasping comb (used to grasp the females during the process of mating). These traits were also compared between species and morphs). Cassis et al. (2017) suggested that all of these traits are likely to function in sexual conflict, however their functional significance has not been explicitly tested.

2.3.4 Size and shape

Morphometry was conducted in order to assess size and shape variation between species, sexes, and winged and wingless morphs. All scanned specimens were compared and measured using Drishti (Table 4). Landmarks were identified on each specimen scan, and using the "path" command, distances between these landmarks were accurately measured in millimetres. To assess variation, ten measures (Figure 4: measures in blue) were then subsampled from the entire dataset (Table 5) and were used to assess the overall shape and size differences between the sexes (males compared with females of both morphs) and/or between morphs of the same sex (female apterous compared with macropterous of the same species). Such proxy measures, taken to assess overall morphological variation, included the animal’s abdomen length, abdominal segments 1 and 7 width, thorax length, thorax width, foreleg femur length, foreleg tibia length, foreleg tarsal length and where possible, egg number and volume (V=πr2h) to assess potential differences in egg carrying capacity which may be tied to species or morph type within a species (Figure 4).

In order to test for dimorphism in mean trait values between the sexes and morphs of the three study species, data were first analysed via a two-way multivariate analysis of variance

(two-way MANOVA), using the statistical software package SPSS

(https://www.ibm.com/analytics/au/en/technology/spss/), which tested for any interactions between the groups. From all the measures taken (Table 5), ten proxy measures (Figure 4: measures in blue) were chosen to provide an estimate of the animals’ overall body size and shape, while for females’ egg measures were used to estimate female egg carrying capacity. With the output of these tests, a Principal Component Analysis was then run to visually examine the trends recovered through the MANOVA using the

statistical software package PAST (https://folk.uio.no/ohammer/past/). Following these

exploratory statistics, univariate ANOVAs, adjusted using the Bonferroni correction

method, were conducted between each of the ten species types (OAF: M. oceanica

apterous females, OAM: M. oceanica apterous male, OMF: M. oceanica macropterous

female, OMM: M. oceanica macropterous male, PAF: N. peramoena apterous female,

PAM: N. peramoena apterous male, HAF: N. herberti apterous female, HAM: N. herberti

apterous male, HMF: N. herberti macropterous female) and the ten proxy body measures

(Figure 4) in order to examine where the variation within and between species lies.

Table 4: Localities and number of individuals of each morph of the three study species. Numbers in parentheses are the total numbers of adult animals collected for this study. Numeric values prior to the parentheses are total number of individuals scanned, measured or examined. Species Locations Collected Apterous Macropterous Male Female Male Female Nesidovelia peramoena Bow River, Walpole, WA 23(41) 33(54) NA 3(3) Nesidovelia herberti Hamelin Station, WA 5(5) 37(39) NA 0(1) Kalbarri National Park, WA 4(4) 5(5) NA NA Microvelia oceanica Chelodina wetland, Perth, WA 13(19) 14(14) 6(6) 4(4)

Figure 4: Dorsal view of apterous male veliid (antenna and legs of right side omitted). Letters depict body measures taken from micro-CT scanned specimens: blue letters indicate measures used in comparative analyses, all other measures are indicated in black. A) head width, B) head length, C) thorax length, D) thorax width, E) abdomen length, F) abdominal segment 1 width, G) abdominal segment 3 width, H) abdominal segment 7 width, I) femur length, J) tibia length, K) 1st tarsal segment length, L) 2nd tarsal segment length. I-L) taken on foreleg, mid leg and hind leg). All measures are in millimetres. Modified from Andersen and Weir (2003).

Table 5: Quantitative and qualitative measurements taken from all micro-CT scanned specimens. All measures were taken from micro-CT volumes using Drishti.

Measurement Units Body length Numeric (mm) Body width Segment 1 Numeric (mm) Segment 3 Numeric (mm) Segment 7 Numeric (mm) Thorax length Numeric (mm) Thorax width Numeric (mm) Head length Numeric (mm) Head width Numeric (mm) Femur Foreleg Numeric (mm) Middle leg Numeric (mm) Hind leg Numeric (mm) Tibia Foreleg Numeric (mm) Middle leg Numeric (mm) Hind leg Numeric (mm) 1st Tarsal Foreleg Numeric (mm) Middle leg Numeric (mm) Hind leg Numeric (mm) 2nd Tarsal Middle leg Numeric (mm) Hind leg Numeric (mm) Female egg count Numeric (count) Female egg volume Numeric (V=πr2h) Female connexival convergence Categorised (widely separated; converging partially; converging and meeting at midline) Female setal tuft presence/shape Categorised (absent; present) Female proctiger position Categorised (level; deflected) Male abdominal segment 8 modification Categorised (absent; present) Male segment 8 modification shape Categorised (absent; paired lateral projections) Male grasping comb Categorised (absent; present) & Numeric (mm)

2.4 Reproductive behaviour and functional biology

2.4.1 Mating trials

To understand whether mating behaviour in the study species is similar to that documented in various Gerridae and gain an understanding of how their apparently antagonistic structures are likely to interact, mating trials were conducted using N. peramoena. Thirty minute trials were observed, with a subset digitally recorded, in an attempt to document the intersexual behaviour prior to (pre-mating struggles), during, and post-copulation (e.g. mate guarding).

All trials were structured the same, whether recorded digitally or not. Initially, trials were structured with a sex ratio of two males to one female. This male-biased trial design was modelled after Arnqvist and Rowe (2002a), who assessed the effects of male and female armament (clasping and anti-grasping morphologies) on the outcome of antagonistic mating interactions in 15 species of water strider (Hemiptera: Gerridae). In their study, if no mating attempts were observed, additional males and females were added sequentially to the mating arena at 30-minute intervals, in order to increase male-female encounter rates.

For each trial, males and females were first extracted from their respective enclosures using an aspirator, then transported to a Petri dish of water, where they were observed or videoed (section 2.4.3) for thirty-minute time blocks. After thirty minutes, one more male and female would be introduced into the same Perti dish as the other mating-trial subjects such that there were now 3 males to 2 females. This process of adding an additional male and an additional female was continually repeated until daylight hours expired. For the purposes of this study, mating trials were only attempted from dawn to dusk in order to maintain similar living conditions to wild populations. This addition of new potential mates was only conducted if no mating attempts occurred in the previous thirty-minute trial. If a mating attempt/struggle occurred, an attempt was made to freeze the couple for a subset of the trials only (see section 2.4.2).

Specimens subject to the trials each day were chosen at random from the collected animals and were only placed into the male-female environment once per trial-day. These animals were not placed back into their large single sex aquariums until the end of the day in order to not repeat-sample individuals taken from the enclosure on any given day.

2.4.2 Freezing engaged pairs

To illustrate how male and female veliids interact prior to mating (pre-mating struggle), during and post-copulation, engaged pairs needed to be instantaneously captured whilst intertwined to distinctively show where and how inter-sexual structures were connecting.

The underlying aim of conducting manually observed mating trials (section 2.4.1) was to snap-freeze engaged pairs in one of these three copulatory processes. Mating trials were set up as above, with males and females placed in a Petri dish and were observed for 30 minute blocks, with additional specimens added if no mating interactions were observed.

Attempts were then made to freeze struggling/copulating pairs using liquid nitrogen.

Arnqvist (1989b) instantaneously froze a number of mating and struggling pairs of Gerris odontogaster (Gerridae) in liquid nitrogen, and Ingram and colleagues (2008) froze male- female pairs of sepsid flies (T. superba and T. lucida) in liquid nitrogen as soon as males mounted the females. The protocols set by both authors were trialled for the purposes of this thesis, however both methods failed as the reaction between the nitrogen and H2O expelled the veliids in opposite directions.

As pouring, diffusing and dunking specimens in nitrogen all failed to secure the male and female pair together, the application of the liquid nitrogen to a secondary applicator was adopted to localise the freezing around the insect pairs while minimizing disturbance of the water surface. This approach involved trial and error with a range of potential applicators, before ultimately settling on a concave, hollow, drill punch (Figure 5A, 5B).

When a pair began to struggle/mate, this was held directly above the animals, forcing them under the water surface and freezing them rapidly within a droplet of ice (Figure 5C-D)

An alternative to liquid nitrogen that also proved successful, if only slightly less effective, was to use an aerosol freezing spray (Chemical Technology Non-Flammable Freezing Spray)

(Figure 5E), to cool the drill punch prior to applying it to the veliids. This was used as an alternative method for extended periods of time in the field. Earlier attempts at aiming this freezing spray at the bugs on the water surface proved too forceful, resulting in disturbance of engaged pairs. Both N2 and aerosol freezing methods were successfully trialled on N. peramoena specimens from Walpole.

Once the engaged pairs were successfully frozen, they were immediately transferred to a minus 30-degree centigrade freezer to allow the nitrogen to evaporate off, then submerged in chilled I2E (staining solution) for 24 hours. They were then rinsed, dried and glued in place to a cardboard triangle on one side. This was done to ensure that the pairs remained attached to one another. Pairs were imaged using a micro-CT scanned dry (as opposed to wet as in other specimens (see 2.3.1 and 2.3.2)). Through this methodology, the interaction of male and female structures was documented to assess functional morphology.

Figure 5: A) 2mm (left) and 3mm (right) steel drill punches for freezing veliids. B) 2mm steel drill punch and ceramic petri dish used as a mating arena for veliid snap-freezing experiment. C) Tip of 2mm drill punch that has been submerged in liquid nitrogen, prior to being used to freeze veliid pair. D) A frozen droplet that has been removed from the tip of a 2mm drill punch post encasing the veliids in a frozen droplet of water. This droplet encases two separate veliids. E) Can of ‘Chemical Technology Non-Flammable Freezing Spray’, used as an alternative cooling source to liquid nitrogen.

2.4.3 Videoed mating trials

Mating trials proceeded in an identical fashion to those conducted for the purpose of snap-

freezing pairs (section 2.4.1). Specimens were captured, transported, housed and chosen

in a controlled set of protocols that only varied in the sense that the trials were being filmed

from above using an iScopeStand (http://www.iscopestand.com/) and a Panasonic DMC-

TZ55 LUMIX DC VARIO rather than being manually observed. Video recordings were edited

using the free Apple Macintosh software iMovie (http://www.apple.com/au/imovie/).

2.4.4 Quantitative and qualitative measures

Mating struggles, matings and post-copulatory behaviours were examined for male and

female behaviours consistent with sexual conflict. The interaction of putative antagonistic

traits was determined by examining the footage frame by frame, combined with the

images taken through the micro-CT outputs. From the video footage, various interactions

were either timed or categorised (Table 6).

Table 6: Quantitative and qualitative measurements taken from video footage of N. peramoena male-female interactions. Measured Method of measurement Type of data Pre- mating struggle length Timed Quantitative Amount of attempts Count Quantitative Amount of flips Count Quantitative Direction of flips Categorised Qualitative First point of contact Categorised Qualitative Direction male faced initially after mounting Categorised Qualitative Points of contact during pre-mating struggle and copula. Categorised Qualitative Outcome of mating attempt Categorised Qualitative

3.0 Chapter Three – Results

3.1 Comparative morphology and putative secondary sexual traits

3.1.1 Comparative morphology

In order to examine the structural variation between the three study species, as well as to determine the variation within each species and/or population, as stated in sections 2.3.3 and 2.3.4. Multiple representatives (Table 4) of each sex and morph where possible for each species were micro-CT scanned, measured, imaged and their secondary sexual traits were classified accordingly (Table 7; Appendix 6). Multiple specimens were scanned to get an accurate estimate of morphological variation in each population.

3.1.1.1 Microvelia oceanica

Apterous and macropterous specimens of both sexes of the species M. oceanica were collected and imaged (Figures 6-10). All apterous females examined displayed raised and converging connexiva (Figure 6A-D), and a deflected proctiger, concealing the gonocoxae

(Figure 6C; Appendix 6.2) whilst the posterior connexival margins lacked setal tufts (Figure

6C-D; Appendix 6.3). Although the connexiva converge toward the midline of the body, they never appear to meet in the middle (Figure 6B-C) and the degree of convergence varied within the population (14% no convergence, 86% partially converged; Figure 7;

Appendix 6.1).

Cn Pc

Figure 6: Micro-CT reconstructions of Microvelia oceanica apterous female. A) Dorsal view. B) Postero-dorsal view of the abdominal tip. C) Caudal view of the abdominal segment 7 and 8, showing deflected proctiger. D) Lateral view. Scale bar 0.05mm, cn, connexiva; pc, proctiger.

Figure 7: Bar plot-showing proportions of apterous females of M. oceanica (n=14), N. peramoena (n=33) and N. herberti (n=42) exhibiting different degrees of connexival convergence: no convergence, partial convergence or complete convergence (the lateral borders of the abdomen are raised vertically and meet above the abdomen). Macropterous females of all three species did not display connexival convergence.

Macropterous females did not exhibit convergent connexiva, with the lateral margins of

the abdomen only slightly raised, never modified and not converging dorsomedially (Figure

8A-D). Additionally, there were no tufts of hair present bordering the genitalia. In

comparison to apterous females, the gonocoxae are relatively concealed (Figure 8C) by a

narrow and deflected proctiger (Appendix 6.2).

Figure 8: Micro-CT reconstructions of Microvelia oceanica macropterous female. A) Dorsal view. B) Lateral view. C) Caudal view of abdominal segment 7 and 8. D) Dorsal view of the abdominal tip. Scale bar 0.05mm.

In contrast to the apterous female, the apterous male’s connexiva do not converge and are

only marginally raised above the abdomen (Figure 9A-B). The genital capsule is simple,

large (about as long as wide) and unmodified, with both segment 8 and the proctiger

lacking lateral processes (Figure 9D-E; Appendix 6.4) (as compared to Nesidovelia, which

has a basally widened proctiger and one or two tooth-like processes arising from the

postero-ventral margin of segment 8: Andersen and Weir 2003). The grasping comb (Figure

9C) is short (about 0.2X tibial length).

Figure 9: Micro-CT reconstructions of Microvelia oceanica apterous male. A) Dorsal view. B) Lateral view. C) Lateral view of the femur, tibia, tarsi and grasping comb of the foreleg. D) Caudal view showing the genital capsule. E) Lateral view of the abdominal segments 7 and 8 as well as showing the genital capsule. Scale bar 0.05mm, gc, grasping comb; py, pygophore; s8, segment 8.

As with apterous males, macropterous males invariably displayed only marginally raised

connexiva (Figure 10A-D) as well as relatively large, but unmodified genitalia capsule (as in

apterous M. oceanica males). The genital capsule (pygophore) is simple, and similar to

apertous males; both segment 8 and the proctiger lack lateral processes (Figure 10 E;

Appendix 6.4). The grasping comb is short (0.2X tibial length).

Figure 10: Micro-CT reconstructions of Microvelia oceanica macropterous male. A) Dorsal view. B) Postero-dorsal view of the abdominal tip and wing structure. C) Caudal view of abdomen. D) Lateral view. E) Lateral view of the abdominal segments 7, 8 and genital capsule. Scale bar 0.05mm.

3.1.1.2 Nesidovelia peramoena

Figures 11-13 show apterous males and females, and macropterous females of N.

peramoena and Table 7 summarises the putative secondary sexual traits of the species. No

male macropterous N. peramoena specimens were collected during this study, as they

weren’t present in the sampled populations during the field season.

In sixty-percent of the apterous females examined, the connexiva were vertically raised,

converging upon each other towards the abdominal end, whilst in 40% of females there

was no convergence (Figure 7; Appendix 6.1). The posterior connexival margins lacked setal

tufts (Figure 11A-D). In this morph the genitalia are recessed within the terminal abdominal

segment, and covered by the deflected proctiger (Figure 11C), with no tufts of hair or

bristles present above (Appendix 6.3).

Figure 11: Micro-CT reconstructions of Nesidovelia peramoena apterous female with partially convergent connexiva. A) Dorsal view. B) Postero-dorsal view of abdomen. C) Caudal view of the abdomen, deflected proctiger. D) Lateral view. Scale bar 0.05mm.

In macropterous females, the connexiva are only slightly converging toward the abdominal

end, and also lack bristles or tufts of hair on the posterior corners (Figure 12A-C). Other

characters are as in apterous females, with the genitalia recessed within the terminal

abdominal segment and obscured by the deflected proctiger (Figure 12C).

Figure 12: Micro-CT reconstructions of Nesidovelia peramoena macropterous female. A) Dorsal view. B) Dorso-lateral view. C) Caudal view of abdomen. Scale bar 0.05mm.

In contrast to apterous females, in apterous males the connexiva are only slightly raised

and do not converge (Figure 13A-C). In contrast to M. oceanica, the grasping comb (Figure

13F) of the fore tibia is considerably longer, about 0.4 X fore tibial length. Males are also

armed ventrally with two pregenital prongs, paired with a shallow depression on segment

7, which is bordered laterally by paired patches of setal tufts (Figure 13E, G-H: see Cassis

et al. 2017). Although no winged males of N. peramoena were collected in this study, the

taxonomic literature indicates that the characters above are the same in macropterous

individuals (Andersen and Weir 2003).

Figure 13: Micro-CT reconstructions of Nesidovelia peramoena apterous male. A) Dorsal view. B) Postero-dorsal view of abdomen. C) Lateral view. D) Caudal view showing genital capsule and bifurcate abdominal process on segment 8. E) Lateral view of abdominal segments 7 and 8, genital capsule, and proctiger. F) Lateral view of the foreleg. G-H) Ventral views of the abdomen and genitalia showing the modification of abdominal segment 8 and the slight hollowing of segment 7. Scale bar 0.05mm, vp, segment 8 ventral process.

3.1.1.3 Nesidovelia herberti

In N. herberti apterous females, the connexiva are raised and converging, becoming

confluent above abdominal segment 7 (Figure 14A-B, D, E-H; Table 7). Most apterous

females also possess setal tufts at the intersegmental limits between laterotergites 5-6 and

6-7, and again on the posterior corners of laterotergites 7 the latter tufts extending both

vertically and horizontally (Figure 14A-H; Table 7). These setal tufts were either in the form

of two distinct paintbrush-like points (Figure 14G-H; Table 7; Appendix 6.3), or in an

expanding spray of bristles (Figure 14E-F; Table 7; Appendix 6.4). Six percent of all apterous

females examined lacked tufts, 75% had spray-like tufts, and 19% had pointed tufts (Figure

15).

As in N. peramoena the proctiger is deflected, concealing the gonocoxae, which are

recessed within the terminal abdominal segments (Figure 14B). Although multiple

macropterous females were not scanned over the course of this project, pinned specimens

from the Pilbara region housed in the WA Museum were examined; in these the proctiger

is as in apertous females, but the connexiva do not converge and there are no connexival

setal tufts.

Figure 14: Micro-CT reconstructions of Nesidovelia herberti apterous female. A) Dorsal view. B) Caudal view of abdomen, showing converging connexiva, setal tufts, and deflected proctiger. C) Lateral view. D) Postero-dorsal view of abdomen. E) Lateral view of abdominal tip showing the setal tufts (spray form). F) Dorso-lateral view of abdominal tip showing the setal tufts (spray form). G) Lateral view of abdominal tip showing setal tufts (pointed-brush form). H) Dorso-lateral view of abdominal tip showing setal tufts (pointed-brush shape). Scale bar 0.05mm, st, setal tufts.

0.8

t 0.6

r 0.4

P 0.2 0 None Spray Pointed Tuft Type

Figure 15: Bar plot of the proportion of different setal tuft types among apterous female M. oceanica (n=14), N. peramoena (n=33) and N. herberti (n=42): no tufts, spray-shaped (setae spreading up and back) or pointed shaped tufts (setae condensed into a pair of caudally directed paintbrush-like tufts).

As in M. oceanica and N. peramoena, the connexiva of apterous N. herberti males do not

converge above the abdomen (Figure 16A-B, D; Table 7) but are instead only marginally

elevated. The genital capsule is similar to N. peramoena, the ventral margin of abdominal

segment 8 is armed with a pair of pregenital prongs close to its ventral, posterior margin

(Figure 16D-F; Table 7). Each process is slender and slightly curved. Their grasping combs

(Figure 16C) of the fore tibia are about 0.5X fore tibial length and are unmodified. Again,

macropters were not collected, but pinned WA Museum specimens show identical

connexival, ventral abdominal and grasping comb traits to their apterous counterparts.

Figure 16: Micro-CT reconstructions of Nesidovelia herberti apterous male. A) Postero-dorsal view. B) Dorsal view of abdomen. C) Lateral view of the foreleg, showing grasping comb. D) Caudal view of abdomen, showing genital capsule and bifurcate process on segment 8. E) Postero-ventral view of abdomen showing bifurcate process on segment 8. F) Lateral view of tip of abdomen. Scale bar 0.05mm.

Table 7: Summary of results presented in sections 3.1.1, 3.1.1.1, 3.1.1.2 and 3.1.1.3 illustrating secondary sexual traits in each sex and species. Secondary Species Species sexual trait type M. oceanica N. peramoena N. herberti Apterous ♀ n = 14 n = 33 n = 25 Macropterous ♀ n = 4 n = 3 n = 2 Apterous ♂ n = 13 n = 23 n = 10 Macropterous ♂ n = 6 n = 0 n = 0 Connexiva Apterous ♀ 86% partially 60% partially Confluent above (Appenidx 6.1) converged. converged. abdominal sternum 7. 14% not converged. 40% not converged. Macropterous ♀ Slightly raised - non Slightly raised - non Slightly raised - non convergent. convergent. convergent. Apterous ♂ Slightly raised - non Slightly raised - non Slightly raised - non convergent. convergent. convergent. Macropterous ♂ Slightly raised - non Slightly raised - non Slightly raised - non convergent. convergent. convergent. Proctiger Apterous ♀ Deflected covering Deflected covering Deflected covering (Appenidx 6.2) gonocoxae. gonocoxae. gonocoxae. Macropterous ♀ Slightly deflected but Slightly deflected but Deflected covering not concealing the not concealing the gonocoxae. gonocoxae. gonocoxae. Apterous ♂ Not modified. Not modified. Basally widened. Macropterous ♂ Not modified. Not modified. Basally widened. Setal tufts Apterous ♀ None. None. 79% spray-like. (Appenidx 6.3) 21% pointed. Macropterous ♀ None. None. None.

Apterous ♂ None. None. None.

Macropterous ♂ None. None. None.

Male Apterous ♀ n/a n/a n/a abdominal Macropterous ♀ n/a n/a n/a sternite VIII Apterous ♂ Lack lateral processes. Possess a pair of Possess a pair of (Appenidx 6.4) pregenital prongs. pregenital prongs. Macropterous ♂ Lack lateral processes. Possess a pair of Possess a pair of pregenital prongs. pregenital prongs.

3.1.2 Size and shape

A two-way MANOVA revealed significant differences between species (F(38, 104)=9.555,

P=.000), sexes (F(19, 52)=8.540, P=.000) and morphs (F(19, 52)=31.595, P=.000) based on

the combination of dependent variables (body measures). In addition there was a

63 statistically significant interaction effect between species and sex (F(38, 104)=2.601,

P=.000) and species and morph (F(19, 52)=3.954, P=.000), but no significant interaction between sex and morph. This indicates that the effect of sex or morph on the dependent variables is not the same for each species.

Notably, the univariate tests found a statistically significant effect of the species-sex interaction model and the abdominal segment 7 width (F(1)=91.516, P=.000), indicating that the segment 7 widths of males and females are not varying to the same degree between the three species.

In light of these results, two principal component analyses (PCA) were conducted: one including body length, abdominal segment 1 and 7 width, thorax length and width of all ten species types (accounting for 87.998% of the total variance), and one that included these measures again but excluded abdominal segment 7 width (94.669% of the total variance) due to the results indicated from the MANOVA previously.

The PCA showed several distinct clusters based on species types (Figure 17). M. oceanica apterous males and females clustered together, whilst M. oceanica macropterous males and females were also quite distinct. N. peramoena and N. herberti apterous males and females formed one loose cluster with some separation between males and females.

However there was some overlap with both apterous and macropterous M. oceanica and macropterous specimens of N. peramoena.

The second PCA (with abdominal segment 7 width excluded; Figure 18) as a

measure between the ten species types, prevented the data from being skewed by the

small widths of apterous female segment 7, which is associated with their converging

connexiva. By removing this difference, two findings become evident. There is now no

distinguishable difference between the sexes of N. peramoena and N. herberti. It also

highlights that apterous specimens of M. oceanica, N. peramoena, N. herberti are clearly

distinct from their macropterous relatives.

Figure 17: Principal Component Analysis biplot of first two principal components (accounting for 87.998% of the total variance) for five body measures (body length, abdominal segment 1 and 7 widths, thorax width and length) as based between the 10 species types.

Figure 28: Principal Component Analysis biplot of first two principal components (accounting for 94.669% of the total variance) for four body measures (body length, abdominal segment 1 width, thorax width and length) as based between the 10 species types.

Univariate ANOVAs were conducted between the ten species types for each of the ten

body measures (Table 8; Table 9), and were adjusted in accordance with the Bonferroni

correction model. Overall females tend to have larger bodies than males (Table 8), however

this was not statistically significant within species (Table 9). Also, macropters tend to have

larger bodies than apters (Table 8; Table 9). This was statistically significant for thorax

width and length in M. oceanica and N. peramoena but non- significant for N. herberti (see

below).

3.1.2.1 Microvelia oceanica

M. oceanica apterous and macropterous morphs form distinct clusters in the PCA from each other and from the other species types. These groupings were supported by many significant differences between body measures of these morphs to other species types in the univariate ANOVAs (Table 9). Both of these M. oceanica morphs were significantly smaller in abdomen length, foreleg femur length and tibia length, than the other species types.

Within this species, the morphs were clearly distinct. Based on thorax width and length measures (often used as proxies for overall body size in insect studies), apterous males and females were found to be significantly smaller than macropterous males and females (P<0.005).

However, apterous and macropterous morphs of M. oceanica, for both sexes, did not significantly differ in mean body length, abdominal segment 1 width, foreleg tibia, femur or tarsal 1 length. For segment 7 measures, apterous males and females were significantly narrower than macropterous males. Apterous females were also significantly narrower than macropterous females (P<0.005) which can be attributed to their connexival convergence. Since apterous and macropterous females of M. oceanica were collected, additional measures were conducted on the female eggs. Total number of eggs per individual were tallied, and average egg volumes were calculated based on the micro-CT datasets: neither of these differed significantly between the morphs.

3.1.2.2 Nesidovelia peramoena

N. peramoena and N. herberti clustered together in the PCA and this grouping is supported by very few significant differences between these species based on the univariate ANOVAs

(Table 9). Within N. peramoena the morphs were also distinct. Apterous males and females were significantly smaller in their thorax length and width when compared to macropterous females (P<0.005), and presumably macropterous males (which were not collected in this study). However, across apterous and macropterous specimens of both sexes, there were no significant differences in body length, abdominal segment 1 width, foreleg tibia, tarsal 1 and femur length, nor in their egg volume and counts (females only).

Due to the females’ dorsal keel, apterous males and macropterous females abdominal segment 7 measures were significantly wider than apterous females (P<0.005) and for thorax width, apterous males and females were significantly smaller than macropterous females (P<0.005).

3.1.2.3 Nesidovelia herberti

As above, across morphs of both sexes, there were no significant differences in body length, abdominal segment 1 width, thorax length or width, foreleg tibial, tarsal 1 and femur length, nor in female egg counts or volumes. Segment 7 width was found to differ significantly, as apterous females were narrower (attributed to their convergent connexiva) than apterous males and macropterous females (P<0.005).

3.1.2.4 Abdominal segment 7

When comparing segment 7 widths, it was found that M. oceanica apterous females and males were significantly narrower than N. herberti apterous males, N. peramoena apterous males and N. peramoena macropterous females. Interestingly, M. oceanica macropterous females were significantly wider than apterous females of both N. herberti and N. peramoena, and M. oceanica apterous males were significantly wider than N. herberti apterous females, results which can all be attributed to the females’ connexival convergence. It was also found that N. peramoena apterous and macropterous females were significantly wider than N. herberti apterous females. This again is unsurprising, as N. herberti apterous females displayed complete connexival convergence over the abdomen whereas the connexival margins of N. peramoena females were only partially raised and never met in the middle. Also, N. peramoena, apterous males are significantly wider than

N. herberti apterous males.

3.1.2.5 Female egg carrying capacity

Finally, for fecund females, egg counts and the volumes of the total clutch were measured and also compared within and between species where possible. It was found that N. herberti apterous females carried significantly fewer eggs than N. peramoena apterous females, while there were no significant differences found between the egg volumes of the different species types (i.e. M. oceanica female apters and macropters did not differ significantly from the Nesidovelia species in these measures).

Table 8: Means, standard deviations (in inverted brackets) and sample sizes (n) of the ten different species types by the ten proxy body measures. Species-morph codes are made of three letters. The first letter in the sequence identifies the species (O = M. oceanica, P = N. peramoena, H = N. herberti); the second letter refers to morphotype (A = apterous, M = macropterous); and the third letter in the sequence refers to the sex (F = female, M = males).

Species Type Body measure (mm) OAF OAM OMF OMM PAF PAM PMF HAF HAM HMF Abdomen 1.074 (0.280) 1.015 (0.094) 1.132 (0.041) 0.978 (0.100) 1.423 (0.084) 1.312 (0.076) 1.253 (0.051) 1.503 (0.088) 1.378 (0.092) 1.660 (0.000) length n = 14 n = 13 n = 4 n = 6 n = 33 n = 23 n = 3 n = 25 n = 10 n = 2 Abdominal seg. 0.647 (0.039) 0.590 (0.051) 0.670 (0.027) 0.681 (0.067) 0.817 (0. 035) 0.785 (0.046) 0.906 (0.064) 0.851 (0.048) 0.863 (0.098) 0.860 (0.000) 1 width n = 14 n = 13 n = 4 n = 6 n = 33 n = 23 n = 3 n = 25 n = 10 n = 2 Abdominal seg. 0.297 (0.049) 0.360 (0.023) 0.480 (0.059) 0.555 (0.113) 0.330 (0.069) 0.518 (0.025) 0.523 (0.095) 0.249 (0.043) 0.532 (0.045) 0.440 (0.000) 7 width n = 14 n = 13 n = 4 n = 6 n = 32 n = 23 n = 3 n = 25 n = 10 n = 2 Thorax width 0.325 (0.066) 0.336 (0.076) 0.710 (0.018) 0.678 (0.109) 0.491 (0.057) 0.492 (0.053) 0.850 (0.121) 0.501 (0.040) 0.519 (0.054) 0.520 (0.000) n = 14 n = 13 n = 4 n = 6 n = 33 n = 23 n = 3 n = 25 n = 10 n = 2 Thorax length 0.602 (0.050) 0.529 (0.042) 0.890 (0.014) 0.838 (0.099) 0.790 (0.029) 0.746 (0.024) 1.000 (0.070) 0.820 (0.079) 0.804 (0.051) 0.950 (0.000) n = 14 n = 13 n = 4 n = 6 n = 33 n = 23 n = 3 n = 25 n = 10 n = 2 Tibia foreleg 0.317 (0.042) 0.275 (0.027) 0.306 (0.005) 0.285 (0.018) 0.439 (0.051) 0.450 (0.054) 0.370 (0.051) 0.463 (0.027) 0.510 (0.083) n/a length n = 13 n = 13 n = 3 n = 6 n = 22 n = 17 n = 3 n = 6 n = 7 Tarsal 1 foreleg 0.166 (0.027) 0.162 (0.013) 0.206 (0.005) 0.195 (0.016) 0.263 (0.146) 0.222 (0.031) 0.300 (0.174) 0.225 (0.028) 0.246 (0.035) n/a length n = 12 n = 13 n = 3 n = 6 n = 20 n = 16 n = 3 n = 6 n = 6 Femur foreleg 0.417 (0.044) 0.373 (0.027) 0.432 (0.018) 0.426 (0.030) 0.532 (0.058) 0.546 (0.052) 0.516 (0.070) 0.560 (0.048) 0.585 (0.035) n/a length n = 13 n = 13 n = 4 n = 6 n = 22 n = 18 n = 3 n = 9 n = 8 Egg count 2.000 (0.000) n/a n/a n/a 4.212 (1.473) n/a 3.333 (2.081) 2.240 (2.471) n/a n/a n = 4 n = 33 n = 3 n = 25 Egg volume 0.392 (0.007) n/a n/a n/a 0.156 (0.062) n/a 0.096 (0.070) 0.148 (0.091) n/a n/a n = 4 n = 33 n = 3 n = 16

3.2 Reproductive behaviour and functional biology

Based on video footage and observed intersexual mating interactions, Nesidovelia peramoena mating behaviour is similar to that observed in various water strider species (e.g.

Rheumatobates, (Rowe et al. 2006); Aquarius remigis (Weigensberg and Fairbairn 1994);

Gerris odontogaster (Arnqvist 1989b): for more examples see (Rowe et al. 1994; Arnqvist

1997a)), where sexual conflict has been well-established. Such interactions appear to be stereotyped and although there are variations between individual encounters, the outcome of the male-female interactions would always follow a cascade of behaviours (Figure 19).

Mating was always initiated by the males, whereby they would chase or attempt to mount females with no apparent pre-copulatory courtship. Of the 26 recorded mating attempts, 30% of all recorded females (8 females) would respond by kicking the male with their hind limbs prior to being mounted. In five of these, the male was dissuaded and no further interaction took place. Of the remaining 21-recorded intersexual interactions (Figure 19), 29% of females fled from males, 67% of females struggled vigorously while being mounted, and in one rare case the female did not resist at all (4%). This vigorous struggle was characteristic in that the females seem to raise heads-down and use their hind-legs to separate from the males, whereas gerrid females are known to raise heads-up and try to separate from the abdominal end (Arnqvist 1989b). It was also observed, however not recorded, that M. oceanica pairs in struggle roll from side to side and males were only ever observed grasping onto the female with their legs.

Of the fourteen females that resisted the male mating attempts, 57% (Figure 19) of them were successful in dislodging the male through an array or forward somersaults and sideward

73 rolls, combined with kicking the male with their hind limbs (Table 10; Figure 20A-I). Struggles lasted from 3 seconds to almost a minute (Table 10). Throughout these struggles, males appeared to grasp the females with their legs (grasping comb), as well as with their

“pregenital clamp”, formed by modifications of abdominal segments 7 and 8, which appeared to articulate like a pair of pliers when the apical abdominal segments were deflected downwards (Figure 20A-I). This was evident from the outset of struggles, with the male first extending his terminal abdominal segments beyond the tip of the female’s abdomen. He then deflected his apical abdominal segments downward approximately 90 degrees, appearing to clamp the tip of the female between the paired prongs on abdominal sternite 8 and the posterior margin of sternite 7. In all video recordings of the pre-mating struggle, the male could be seen flexing and relaxing his terminal abdominal segments, perhaps in an effort to secure themselves atop of the struggling females (Supplementary video – available on request). In the 6 cases where the female had ceased struggling (Figure 19), the male was then documented drawing his hind limbs in close to the female’s abdomen and vibrating his hind limbs vigorously whilst continuing to move his terminal abdominal segments. Eventually the male concluded moving his hind limbs and was then observed to slide back farther on the female’s dorsum, and was apparently able to insert his genitalia into the female genital opening to begin copulation.

Figure 19: Ethogram depicting the mating sequence of apterous N. peramoena, based on 26 mating attempts/intersexual interactions observed in a laboratory colony. Numbers indicate how many times a given behaviour was recorded. Rectangles75 indicate male-initiated actions, ovals indicate female-initiated behaviours and dashed modified rectangles indicate an interaction between both sexes. Table 10: Behavioural data taken from video footage of pre-mating behaviour of apterous N. peramoena males and females.

Measure Observed/quantified result n = 11 Struggle time (seconds). Minimum 3 Maximum 50.79 Average 17.28 Female dislodging methods. Forward flips, sideward rolls and kicking the male with her hind limbs. Females raise heads-down and use their hind limbs to separate. Number of flips during struggle. Minimum 2 Maximum 23 Average 13 Direction males approach females prior to initial contact. Front, side and behind. Points of contact between the sexes during the struggle. Abdominal segments 7 and 8 (genital capsule) and tips of all legs of male embracing onto the female. Outcomes of intersexual interactions (percentage of occurrence). Female’s fled from males advances. 29% Female’s resisted mounting attempts. 67% Females did not resist prior to copula. 4% Males were successful in copulating with females. 33%

Figure 20: Still images from video recordings of pre-copulatory struggles and mating behaviour in N. peramoena. Images from 6 different pairs. A-I) The male mounts the female without any observed pre- copulatory courtship. The female immediately attempts to dislodge the male by thrashing and rolling via forward somersaults. The male is seen grasping with his legs and pregenital segments (segment 8 and the genital capsule are visibly extended and declivent). Once the female stops struggling the male rhythmically vibrates his hind legs before copulation begins, or else he is thrown from her abdomen. (Images A-C Nikolai Tatarnic 2017, WAM).

3.3 Micro-CT imaging of antagonistic structures

In order to show how males grasp females during mating struggles and to confirm the function of the pregenital clamp and its interaction with the female, struggling male-female pairs of N. peramoena were snap frozen using liquid nitrogen (Figure 21A-F; Figure 22A-F). Of the countless attempts made directly aiming to freeze the veliids struggling or in copula, four pairs were successfully secured in position and mirco-CT scanned. In these four pairs, the proctiger was open and the pregenital clamp can be seen either grasping the underside of the proctiger

77 or nearly so (the latter attributed to the male being slightly dislodged during the freezing process). The results of this show that males use their grasping comb to grasp onto the females between their thorax and abdomen (Figure 21A), while at the same time clamping onto the lip of the female proctiger with their abdominal pregenital clamp, apparently using the prongs on sternite 8 to pry up the proctiger, which is usually deflected and concealing the gonocoxae. As the male appears to engage his pregenital clamp immediately upon mounting the female, we can assume that this act instigates the vigorous struggle, with the female resisting as the male attempts to hook on and pry open her proctiger. Interestingly this pregenital clamp remains in position during copula. This is shown in Figure 21B-F and Figure

22A-F, whereby the segment 8 process can be seen pressed on the underside of the female’s proctiger whilst the male’s genitalia are inserted in the female (Figure 22D). Closer examination of the female proctiger reveals that the underside is indeed modified seemingly to fit with the male’s segment 8 process, with paired concavities on either side of a midline ridge, into which each prong of the male process fits (Figure 22E-F). Additionally, the male aedeagus can been seen inside the female bursa, with the long thin male ductus seminis extending out the end, apparently entering into the female spermatheca, which is long, thin and coiled (Figure 22D; see also: Figure 4 in Cassis et al. 2017).

Following mating the male dismounted without struggle and there was no observed postcopulatory mate guarding.

♂ ♂

Pr →

♀

GC ♀

S7 →

S8 →

→

→ S7 Py

A) B)

♂ ♂

♀ Pc→

S7 → Pc

S8 → → S7

S7 S8 Py

♀ →

Py S7 C) D)

♂ ♀ ♂ Pc ♀

→ S7 Pc VP S7 VP

→ S8

Py → →

E) F) Py → S7 Figure 21: Micro-CT reconstructions of Nesidovelia peramoena apterous male-female mating pairs. Images are of two different pairs (Pair 1: images A, C & E. Pair 2: images B, D & F). A) Detail of male grasping female thorax, with male grasping comb behind the female’s pronotum. B) Lateral view of male-female pair connected by their genitalia, and by the male abdominal ventral process on segment 8 grasping the underside of the female’s proctiger, pressing it into the caudal margin of segment 7. C-F) Lateral views of the connection between male-female mating pairs. Scale bar 0.50mm, gc, grasping comb; pc, proctiger; pr, pronotum; py, pygophore; s7, segment 7; s8, segment 8; vp, ventral process. 79

Pc Pc

uPc

uPc VP

Pc VP Pc Ph A) uPc

VP uPc uPc VP

Figure 22: Micro-CT reconstructions of Nesidovelia peramoena apterous male-female mating pair 3. A-C) Lateral views of the connection points between the male and female. These images illustrate how the male’s ventral process connects to the underside of the female proctiger. The bifurcate prongs of the ventral process fit into paired depressions on the underside of the female’s procitger. D) Lateral view of a partial cross section of the female revealing the positioning of the male phallus inside the female. E-F) Underside view showing the male process fitting depressions in underside of female proctiger. Scale bar 50.00um, pc, proctiger; ph, male phallus; upc, underside of female proctiger; vp, ventral process

4.0 Chapter Four – Discussion

4.1 Introduction

The results of this comparative analysis supports the view that sexual conflict is driving a male and female coevolutionary arms race in the Veliidae, as has been found in the Gerridae (see

Arnqvist 1989b; Arnqvist 1992; Arnqvist and Rowe 1995; Arnqvist 1997a; Andersen 1997;

Arnqvist and Rowe 2002a; Arnqvist and Rowe 2002b, Rowe and Arnqvist 2002). In the following discussion, the significance of sexually dimorphic traits in the study species, in particular their roles in pre-mating struggles and copulation are explored, providing strong evidence for sexual conflict in these insects. These results could prompt future studies aimed at understanding the remarkable diversification of this families’ secondary sexual traits, genitalia and behaviour.

4.2 Putative secondary sexual traits

Inspections of the three study species external morphology and reference to the taxonomic literature allowed me to discern putative secondary sexual traits, their levels of variation and speculate on their uses. For the purposes of this study, female connexival elevation, female proctiger angle, setal tuft presence and shape, male ventral abdominal segment 8 process presence and shape, male ventral abdominal segment 7 medial concavity, and foreleg grasping comb size were all of interest. Such traits have been interpreted as having evolved to facilitate male grasping and female resistance during pre-mating struggles (Cassis et al.

2017), predicted as a direct outcome of sexual conflict (Arnqvist and Rowe 2005). Sexual conflict is ubiquitous in the Gerridae (e.g. Arnqvist 1992; Arnqvist & Rowe 1995; Rowe &

Arnqvist 2002; Arnqvist & Rowe 2002a), sister-clade to the veliids, and this study is the first to explicitly document how some of these morphological traits operate during intersexual interactions. By understanding the functional significance of these traits in N. peramoena (see section 4.5), it may be assumed that veliids are subject to similar selective pressures as the gerrids (for example, the Rheumatobates) as they have developed secondary sexual modifications that resemble those in the Gerridae. Across the Veliidae, there are many examples of sexually dimorphic structures – some similar to those in Microvelia and

Nesidovelia, and others different – which are best explained by sexual conflict, rather than mate choice. For example, species of Angilia, Stridulivelia, Velia, Veloidea, some Paravelia, and the Rhagoveliinae have their hind femora armed beneath with one or more rows of teeth

(Andersen 1982). This armature is usually stronger in the male than in the female (absent in some females, e.g. Velia caprai). More still, females of Microvelia prompta (Cheesman 1926) possess completely concealed genitalia, which are hidden by closed vertical connexival walls that meet at the dorsomedial plane above the abdomen, and metanotal and abdominal tufts of hair, all of which appear to impede access to the female’s genitalia (pers. obs.). This list is by no means extensive, but merely serves to highlight that Veliidae exhibit elaborate sexual dimorphism, much of which is best explained by sexual conflict.

Based on the evidence produced from the visual analysis of all three species, it can be said that abdominal and genital shape are dimorphic between both the sexes, and these traits, as well as thoracic structure, are dimorphic between morphs of the same species. For all three study species, the apterous females’ proctiger are deflected and conceal their gonocoxae. The connexiva (lateral margins of the abdomen) of the apterous females of both M. oceanica and

N. peramoena are raised and converge dorsomedially toward the tip of the abdomen.

However, these connexiva do not meet and touch at the midline. In N. herberti, however, the connexiva converge completely and meet at the dorsomedial plane, therefore almost completely concealing her genital region from above. The connexival convergence is thought to have evolved to help females make it more difficult for males to withstand the females’ resistance manoeuvres during a pre-mating struggle (Tatarnic et al. (in prep)).

N. herberti apterous females possess abdominal tufts slightly concealing their gonocoxae and effectively lengthening the dorsal surface of the abdomen, possibly making access to the genital opening more difficult for males and thereby aiding females in resisting costly male mating attempts, as has been shown with the female connexival spines in various water striders (Arnqvist and Rowe 2002a; Arnqvist and Rowe 2002b). Alternatively, they may have evolved as a cryptic female choice mechanism, providing a challenge that males must overcome to succeed in mating, thus allowing the female to assess mate quality before progressing with copulation. These tufts could be an example where the females are ahead in the evolutionary arms race in terms of sexual conflict, as apterous males did not appear to exhibit any additional external characteristics, such as abdominal lengthening or a deep ventral abdominal depression to accommodate the female tufts, as appears to be the case in

Nesidovelia fluvalis, Papuavelia siculifera and Tanyvelia missim (Cassis et al. 2017). Future work using experimental removal of setal tufts may aid in determining the capacity of females to resist males. Alternatively, in several species of Nesidovelia, apterous females are polymorphic for this trait, providing a natural system for comparing female resistance between tufted and tuftless morphs.

Veliids also often exhibit wing polymorphism, with some individuals winged and others wingless, which may have evolved because flight is an energetically expensive trait associated with a developmental cost (Roff 1986). Trade-offs arise when two or more body parts compete for a limited pool of resources to sustain their growth and differentiation (Moczek and Nijhout 2004). The evolution of flightlessness in so many insect orders, families and species suggests that there is a large cost in the development of wings, whether or not the animal chooses to fly (Roff 1986). The evolution of dispersal through flight will be a function of the habitat characteristics and may involve a trade-off between this and other traits. For example, in Drosophila melanogaster, both egg production and flight depend on the same energy reserves, hence reducing fecundity when flight is preferred resulting in obvious consequences for the fitness of the disperser (Roff 1986). Macropterous females of M. oceanica, N. peramoena and N. herberti do not exhibit connexival convergence, nor do N. herberti retain their setal tufts in their macropterous form. Along with the males of all three species, their connexiva are only slightly raised and never converge inwards above the abdomen. The fact that macropterous female M. oceanica and N. peramoena are larger in overall size than their apterous counterparts, yet lack the secondary sexual structures observed in this study (no setal tufts, no converging connexiva) may be the result of an energetic or mechanical trade-off between the development of flight muscles and possession of wings and the development and maintenance of resistance traits. For example, this lack of connexival convergence and setal tufts in macropterous females may indicate that they undergo a trade-off in resource allocation between sexually antagonistic traits and the capability of flight. Alternatively, rather than a trade-off in resources, it may simply be that the female secondary sexual abdominal modifications impede wing position (e.g. at rest, the wings could not fold over the abdomen), or that they would provide no resistance advantage

84 when concealed by the wings. Further, this loss in sexually antagonistic traits in females may reflect alternative mating tactics between the morphs, as macropters may opt to fly instead of fight (Fairbairn 1992). Though there are very few examples where females are polymorphic for secondary sexual traits and employ alternative mating tactics, thus far a majority of these have been associated with sexual conflict (Buzatto et al. 2014).

While wing polymorphism may have some impact, or even be partly driven by, sexual selection (and sexual conflict), in gerrids it is mostly thought to be an outcome of natural selection, and has been explained mainly in relation to different degrees of habitat permanence (Vepsäläinen 1974; Jarvinen and Vepsäläinen 1976; Vepsäläinen 1978; Andersen

1993). Although migration by flight permits individuals to escape from unfavourable environments and/or matings, it is also a hazardous process for aquatic insects occupying spatially isolated habitats, as is seen throughout gerrid and veliid populations. Therefore, flightlessness is often viewed as an adaptive specialization to maintain breeding populations in stable, or at the very least, predictable changing environments (Harrison 1980; Roff 1986).

This shows that coevolutionary arms races between the sexes are not inevitable and that other selective pressures, such as population dynamics or microhabitat structure, may prevent the evolution of secondary sexual traits.

The evolution of antagonistic sexual traits as described above for females (i.e. connexival elevation and convergence, setal tufts, and the downward deflection of the proctiger) is expected to exert selection on males for overcoming female resistance during pre-mating struggles. Through this study it has been found that all apterous and macropterous males of

N. peramoena and N. herberti possess a ventral abdominal process on segment 8. Water

85 striders demonstrate remarkable diversity in male genital and nongenital clasping mechanisms as well as female antigrasping structures in a number of genera (Arnqvist and

Rowe 2013). These structures are used for species identification and have been documented to be just as divergent as the genitalia in the Gerridae (Andersen 1991; Andersen and Weir

1994; Andersen and Weir 2004). In Halobates for example, there is great diversity in the shape and structure of the male grasping comb (Andersen and Cheng 2005). Similarly, in the

Rheumatobates, most of the morphological diversity in the genus is comprised of the diversity of their grasping traits as all male appendages (including the antennae) have been modified for holding onto females (Westlake et al.2000; Arnqvist and Rowe 200b; Rowe et al. 2006).

As mentioned, N. peramoena and N. herberti both have forked processes on segment

8 of the abdomen which articulates with the caudal margin of segment 7. This was initially documented by Andersen and Weir (2003), though they did not discuss its possible function.

Through freezing N. peramoena pairs in copula, my study has shown that when the terminal segments are bent 90 degrees, the processes on segment 8 pinch against the margin of segment 7 and the female proctiger. This assists the male in grasping the female, forcing access to her genitalia (similar to Gerris odontogaster in Arnqvist 1989b). Unlike females, macropterous males of N. peramoena and N. herberti do not lose their secondary sexual abdominal modifications when winged. The fact that males retain these traits when winged suggests that either the direct fitness benefit to males outweighs any potential trade-off to the development of wings, or else that the functioning of the male grasping apparatus is not affected by the presence of wings. Along with this abdominal process, the postero-ventral margin of the abdominal sternite 7 of both apterous and macropterous males from both

Nesidovelia species is medially depressed, with paired setal patches to either side of the midline. This is thought to interact with the process on segment 8 in clasping the female, while

86 the medial depression may also fit over the female’s keeled dorsum, thereby stabilizing the male onto the female throughout a pre-mating struggle. It is possible that males with deeper abdominal depressions may be more successful at withstanding female resistance behaviours.

4.3 Size and shape

My observations and analyses indicate that both sexes and both morphs of all three study species differ significantly in size and shape. Measurements based on micro-CT scanning and optical microscopy reveal clear dimorphism in various body measures such as thorax and abdominal length and width. Overall, females tend to have larger bodies than males, and macropters tend to have larger bodies than apters.

Apterous female representatives of all three species were longer or wider than males, however this trend was non-significant. In the PCAs, macropters cluster distinctly from apters.

Females typically tend to have larger bodies, as they must carry and provision their eggs, which are energetically and spatially much more costly than male sperm. This is consistent with what is seen in most insects, with body size strongly linked to reproductive investment

(Gilbert 1980; Evans 1982; Reiss 1989; Honěk 1993). However, one measure in which males exceeded apterous females is in the width of abdominal segment 7. This reflects the tapering of the female abdomen, which is associated with their connexival convergence.

In the ecology, life history and reproductive fitness of a species, body size is crucial and patterns of intraspecific variation in body size are expected to reflect patterns of adaptive divergence (Fairbairn 1997). Sexual size dimorphism (SSD) is a complex trait, potentially

87 influenced by various factors that affect one or both sexes, and varies greatly in direction and degree among species (Roff 1981; Arak 1988; Fairbairn 1997). From the data presented in relation to the veliids sizes, it is shown that apterous and macropterous females are larger than males of the same species and morph, indicating that female-biased SSD is present within these species (Fairbairn 1997) and selection associated with the differing reproductive roles of the two sexes is a likely factor determining this dimorphism (Darwin 1871; Fairbairn

1997). SSD varies across taxa, with female-biased SSD predominating among invertebrates

(Blackith 1957; Fairbairn 1990; Andersen 1994; Abouheif and Fairbairn 1997; Fairbairn 1997).

Studies on the evolution and adaptive significance of SSD have provided much of the evidence for the importance of selection on reproductive traits, particularly traits affiliated with sexual selection in males (Andersson 1994), fecundity in females (Forsman and Shire 1995), and patterns of parental investment (O’Neill 1985; Fairbairn 1997). Although multiple matings may be advantageous to males in terms of maximizing their fitness output (Parker 1979), parental investment and assessing their partners’ fitness may be more advantageous for females. Darwin hypothesized that the evolution of large body size in females was related to being able to produce more offspring, known as the “fecundity advantage”, assuming that clutch size increases with body size (Darwin 1871; Shine 1988).

Similarly, macropterous females were significantly larger than their macropterous male relatives, as well as their apterous relatives of both sexes. Interestingly, the macropterous females of all species were significantly larger than apterous females in both their thorax and abdominal segment 7 width. Although male macropterous N. herberti were not collected or measured, we expect them to show similar sex and morph scaling patterns as M. oceanica and N. peramoena. These shape differences between wing morphs are also consistent with their presumed specialisations - macropters being adapted for migration by

88 flight and apters for mate resistance and sexual conflict (Harrison 1980; Roff 1986; Fairbairn

1992). The dorsal longitudinal wing muscles are the largest muscles in the gerromorph body, and when fully developed occupy more than 60% of the volume of the mesothorax (Andersen

1973). Attachment of these and other flight muscles, as well as the two pairs of wings, is associated with changes in the shape and thickness of the cuticle structures of the thorax

(Matsuda 1960; Andersen 1973). Apterous M. oceanica, N. peramoena and N. herberti have neither wings nor flight muscles. It is, therefore, not surprising that the primary difference between wing morphs occurs in the size and shape of the thorax. The morphs also greatly differed in the abdominal measures however, as mentioned previously, the abdominal modifications of the apterous morphs are assumed to be incompatible with wings.

This study found no differences between the apterous and macropterous egg numbers or the total volumes. This may be a true indicator of both morphs lifetime reproductive capabilities, however, this was a short-term measure and it does not provide data on whether the clutch sizes and average egg masses differ between the morphs in terms of overall lifetime reproductive success. Further studies to address this should look at the reproductive outputs over the animals’ lifetime, therefore providing insight into whether there is a trade-off between flight and fitness, or between the secondary sexual traits and fitness.

One would assume that selection favouring an increase in size of a given body component, such as the abdomen in females or the thorax in macropters, can be expected to be positively correlated with other body components, and thus an increase in overall body size. In relation to this, Fairbairn (1997) found the opposite in the gerrid, Aquarius remigis, whereby the larger morphs (females and macropters) tend to have relatively small genitals

89 and appendages due to a lack of positive genetic correlations between appendage and body components. This could also be the case with the veliid species in this study, and in order to assess this, future work will need to be conducted measuring both sexes genitalia and looking for patterns between these measures and their secondary sexual traits. Fairbairn (1997) also postulated that flight in females may be a resistance trait. For example, instead of engaging in a pre-mating struggle and therefore needing stronger appendages for throwing the male off her back, if the female chooses not to copulate with the attempting male, she may just fly away.

Apterous males and females dominated the sampled populations of all three species (at the time of collecting). This morph ratio further influences the idea that there may be a selective pressure against wing production simply as an indirect result of selection for earlier developmental requirements (i.e. costs of wing development). Macropterous individuals could also suffer higher mortality rates (for example, macropters may be less adept at skating on the water surface and avoiding predators, or the energetic input into wing development and flight muscle maintenance may add physiological stress). It is also possible that the presence of macropters varies seasonally, as has been found in various gerrids (e.g. see

Andersen 1993). The ratios collected of these populations may be the result of the season or a representation of the outcomes of multiple selective pressures balanced to maximize population fitness and stability.

4.4 Female polymorphism

Polymorphism in females was illustrated in all three species within their connexival convergence. In M. oceanica, 86% of females’ connexivas were partially converged and raised above the abdomen and in 14% of the specimens’ the connexiva were not raised, and they resembled the apterous male specimens. Similarly, in 60% of N. peramoena apterous females the connexiva were raised and partially converging, whilst in the remaining 40% the connexiva, similarly with M. oceanica, resembled the apterous male specimens, where the connexiva were slightly raised, however were not converging toward the midline. More interestingly, the apterous females of N. herberti collected showed polymorphism in their setal tuft arrangement. The females’ connexiva were all raised and converging, becoming confluent above sternum 7, however their setal tufts were observed in two separate forms.

Tufts in 21% of the females were compressed into a fine, paintbrush-like form and 79% had the tufts in the form of an expanding spray of bristles.

As these presumed sexually antagonistic traits (connexival convergence and setal tufts) are polymorphic, this suggests that there may be some kind of trade-off between their development, maintenance and/or other life history traits, such as allocation to egg production. Similar polymorphism is exhibited throughout the genus, as N. childi, N. woodwardi and N. canarvon all have both tufted and tuftless apterous females (Anderson and

Weir 2003). Comparative studies collectively suggest that conflicts between males and females over mating rates can have profound evolutionary consequences and may even be an engine of speciation and polymorphism (Rice and Holland 1997; Arnqvist et al. 2000).

Polymorphic changes can be observed as the animals invade new environments, as predation regimes are altered or as selective pressures become more important. Conflict between sexes

91 over control of copulation may drive the coevolution of elaborate genitalia or secondary sexual structures, such as colour or these tufts of hair (Arnqvist and Rowe 1995; Cordero and

Andrés 1996). Support for this hypothesis is limited to male adaptations that function to enhance male control over females in copulation. Evidence for morphological adaption in females is critical to this hypothesis yet lacking (Arnqvist and Rowe 1995). For this reason, the morphological adaptations specific to females of N. herberti (and other Nesidovelia species) are of particular interest. This polymorphic structure demonstrates the coevolutionary nature of sexual conflict and that females are indeed active participants in the evolutionary conflict over control of reproduction.

Although no differences in external body measures or in egg count and volume were found to be associated with the two types of tufts in N. herberti apterous females, differences may still be present. For example, relative ability to resist mating attempts might vary, and this may translate into changes in longevity thus altering lifetime reproductive success. More work is necessary, examining perhaps their methods of resistance against males, as well as conducting the same methods of snap-freezing copulating and struggling pairs in action, in order to be able to assess where these tufts interact with the males and what the males are doing in order to maneuver themselves around these structures. Evidence for female polymorphism is extremely limited, thus conducting work on these aspects of this families’ polymorphic traits will ultimately provide new evidence to an unknown area of the evolution of insect mating systems.

4.5 Reproductive behaviour and functional biology

Based on the recorded observations of N. peramoena pre-mating struggles and examination of snap-frozen mating pairs, the mating system of this species (and related taxa) can likely be characterised as one driven by sexual conflict, with female resistance (which occurred in 96% of all observed intersexual interactions, either in the form of fleeing or vigorous pre-mating struggles) and male coercive behaviours underpinning the development of the morphological traits of interest discussed in this study. Water strider matings (specifically Gerris odontogaster: Arnqvist 1992) are rapid and exaggerated, much like these riffle bugs. Males pounce on females and as the female attempts to dislodge the male via chaotic somersaults and other erratic movements, males attempt to withstand her resistance behaviours using various grasping mechanisms whilst aggressively trying to achieve genital coupling (Arnqvist

1992). Arnqvist (1992) documented that the resistance/persistence traits and behaviours of

Gerris odontogaster are consistent with predictions of sexual conflict, as they are being used to avoid costs associated with unwanted matings (i.e. females modulate their resistance according to expected costs) and not what one would expect under the rubric of female choice. In keeping with this, given the ecological similarity to the Gerridae, where sexual conflict is ubiquitous and has been associated with their water-surface lifestyle (Eberhard

2006), it could be predicted that sexual conflict occurs throughout the Veliidae family.

The findings of this study suggest that in response to male coercion, setal tufts that project caudally (as in N. herberti), a downwardly deflected proctiger, recessed and concealed genitalia and the female dorsal keel of N. peramonea, N. herberti and M. oceanica have arisen to impede male grasping by decreasing the physical proximity between the sexes and the

93 stability of the male’s grip during mounting. This in turn is interpreted to have driven the evolution of the male abdominal concavity, which would allow the male to maintain close contact with the female dorsum throughout pre-copulatory interactions, as well as during copula. Furthermore, the presence of these secondary sexual traits suggest, that male harassment is likely to be ancestral to both Microvelia and Nesidovelia and it can be assumed that male coercive behaviours led initially to female resistance, promoting this escalating arms race of antagonistic trait evolution. Female connexival convergence and male abdominal excavation are not exclusive to these study species, and are in fact found in various Veliidae

(for example: Papuavelia siculifera and Tanyvelia missim; Cassis et al. 2017), but their functions have not been experimentally determined. The various extensions of the tip of the female abdomen (e.g. setal tufts and extended connexiva) are, however, reminiscent of the male-blocking connexival spines of some water striders (Arnqvist and Rowe 2002a), and the hidden genitalia in Nesidovelia and Microvelia is similar to what is found in Gerris gracilicornis, which is thought to make it harder for males to access the female’s genitalia (Han and

Jablonski 2009). Other traits, such as the connexival convergence and dorsal setal tufts of some Nesidovelia, appear unique to the Veliidae. While the basic function of these traits can be interpreted with some confidence, there is still no evidence of their functional significance, thus future comparative studies in other genera where these traits also exist are required.

Throughout both sexes of the veliids, there are various other putative secondary sexual traits that are also likely to function in sexual conflict. In males, these include a vast array of ventral abdominal modifications and processes on segments 7 and 8 (Andersen and Weir 2003), which through this study are known to articulate and act as a pregenital clamp, as seen by

94 snap-freezing N. peramoena pairs with males grasping to the females on the underside of her proctiger during the pre-mating struggle, as well as during copula. More evidence to support this comes from the video footage of struggling pairs, whereby males occasionally appeared to maintain contact with the female solely by the abdomen. Other male traits include the forelimb grasping comb, which is believed to grip the females’ dorsum during mate struggles

(Andersen 1997) and various setal tufts, and swollen growths on the abdominal venter (Cassis et al. 2017).

This work represents the first study to explicitly show the function of the male ventral abdominal clamp, with the prongs on segment 8 acting to pry up the female’s downwardly deflecting proctiger and exposing her genital opening. At the same time, the clamping of the female proctiger between the segment 8 processes and the postero-ventral margin of segment 7 appears to help the male grasp the resisting female, as the pairs somersault and roll in a chaotic dance resulting in either the male being expelled from the female’s dorsum or copulation occurring. As the male clamp may begin opening the female genitalia right at the onset of the mating struggle (this was difficult to discern from the video recordings), it is possible that males attempt to initiate copulation immediately, rather than attempting to subdue the female prior to genital insertion. One can then assume that the varying degrees to which the female genitalia are hidden/recessed inside the final abdominal segment, along with the downwardly deflected proctiger, may be direct responses to the males clamping processes (in N. peramoena, however possibly in all species of veliids). Moreover, the apparent association between the underside of the female proctiger and the shape of the male segment 8 process begs for further research into assessing these structures in other

Nesidovelia species. If this structural association exists between other veliid species, it will then be important to understand which sex is driving the morphology of the other, as this can have important implications to our understanding of what selective pressures underpin the modifications of these traits (for example, female choice, sexual conflict, reproductive interference between sympatric species and so on). Most of the sexually dimorphic structures examined in this study have analogues in the Gerridae, where their functions have been experimentally determined.

Another potential avenue of research is to explore the evolution and function of the male and female genitalia itself. Recent work on Nesidovelia has shown that males and females of the genus exhibit complex genitalia, which are likely under postcopulatory sexual selection (Cassis et al. 2017) and Arnqvist and Rowe (2002a) indicated that genitalic traits were significantly more complex than nongenital traits. For example, in species exhibiting male and female antagonistic adaptations (as in N. peramoena and N. herberti), the female possesses an elongate, coiled spermatheca, which is matched by an elongate ductus seminis in the male.

In species lacking antagonistic adaptations, the spermatheca and ductus are short (Cassis et al. 2017). The use of micro-CT to examine specimens in copula has enabled one to observe the interactions of the male and female genitalia, and so gain insight into potential selective pressures. Indeed, the scans presented in this study appear to show that the male ductus enters into the spermatheca, rather than injecting sperm into the bursa. Whether this is another example of sperm competition or instead reflects female choice remains to be seen.

Within evolutionary arms races between the sexes, an overall balance in the development of resistance/persistence traits is expected. This is not always so straight forward as imbalances are also expected to occur where one sex may have evolved a greater degree of armature than the other (Parker 1979; Parker 1983; Härdling 1999; Arnqvist and Rowe 2002a; Arnqvist and Rowe 2002b). This thesis has outlined many internal and external structures, however future studies should also focus on those species where there is an apparent mismatch between the male and female secondary sexual traits. Perhaps in terms of the underside of female proctigers and their level of modification in regards to the same species male ventral abdominal processes or the level of internal genital complexity, as this may inform our understanding of the comparative value of the traits in question, by identifying species or populations where one sex or the other is “winning” the war. Evidence of this may include changes in mating rates, length of pre-mating struggles, and changes in relative fitness.

4.6 Refinement of freezing methodology

This study successfully developed a new method for freezing veliid couples that were in mating interactions with liquid nitrogen. As the study species are so slight in size and must be observed on the water surface, the interaction between the H2O and the N2 proved to be violent and disruptive to the animals, the resulting effervescence often forcing the couples apart. The method of introducing a frozen instrument to the intertwining couples rather than directly introducing the nitrogen to the animals and water, developed and perfected in this study, overcame this initial challenge.

4.7 Limitations and future works

Throughout this study, gaps in the collected specimens, i.e. the lack of sufficient macropterous morphs of both sexes across the species, introduced limits on the statistical analyses, as well as the morphological comparisons that could be made. In order to increase the robustness of this study, research should be conducted on multiple populations and across all seasons in an attempt to understand what is driving variations in the morph types, and tease apart environmental factors like seasonality from sexual selection, in particular sexual conflict.

Moreover, although morphology could be investigated in all three species, reproductive behaviour and functional use of secondary sexual traits in action was only observed during the pre-mating struggles and during copula of N. peramoena. This means that although the function of male ventral abdominal clamp could be shown for one species, it can only be speculated that this might also be true for other species within the group, in particular those with especially complex male abdominal processes. The strength of the results could also be enhanced by conducting mating trials on the macropterous morphs of these and other species, in order to identify differences in mating and mate struggle behaviour and success between macropterous and apterous individuals and compare how their morphology and behaviour intertwine during pre-, post- and during copulation. As mentioned previously, examining the resistance methods of females within species that express setal tufts or polymorphism within this trait would advance the knowledge of their function, strengthen these results, and would be ideal in order to further tease apart the evolutionary pressures that drive these traits expression.

Transporting study animals also proved to be a challenge. It became very clear that mating attempts and intersexual interactions were more frequent on the first and second days after collection from the natural habitat and introduction into the laboratory enclosures.

Unfortunately, attempts at freeing couples in the field were not successful, mainly due to inconsistency in sourcing liquid N2. This meant that attempts at freezing pairs had to be undertaken in the lab, which was a full day’s drive from the field site. The identification of stable populations much closer to the city would likely facilitate future research.

5.0 Chapter Five – Conclusion

This thesis compares patterns of sexual dimorphism in three Western Australian Veliidae species, and aims to explain these in the context of sexual conflict. Through recordings of male-female reproductive interactions and examinations of male-female pairs of N. peramoena frozen in copula, this study provides strong evidence for sexual antagonism in veliid mating, with the evolution of male and female antagonistic morphology and behaviour.

Males and females exhibit various secondary sexual traits, which are likely to be coevolving in an evolutionary arms race between the sexes. Females were found to be generally larger than males of the same species and macropters larger overall than apterous individuals, as is often the case in insects. Interestingly, while males retain their secondary sexual traits in the macropterous form, macropterous females were found to lack their resistance traits. This suggests that the fitness benefits males receive from their grasping structures are present in both macropters and apters. However for females, there may be either a resource allocation trade-off between wings and resistance structures (energetic trade-off), or else there is a mechanical trade-off such that these traits would not function as resistance structures once covered over by wings. This study is one of the first to provide evidence that sexual conflict is the likely driving force behind behaviour and morphology of male and female riffle bugs. The unique male grasping and female anti-grasping structures of these insects clearly beg for further research. Moreover, the internal genital coupling between mating pairs, as revealed through micro-CT imaging, suggests other aspects of sexual selection at play that are yet to be determined. The vast array of sexually antagonistic structures, both putative and functionally examined within both sexes of the Veliidae, make those expressed in the

100

Gerridae (the current model group for sexual conflict and sexually antagonisitc character studies) seem relatively tame in comparison.

101

References

Abouheif, E. and Fairbairn, D. J. (1997). A comparative analysis of allometry for sexual size dimorphism: assessing Rensch’s Rule. The American Naturalist 3, 540-562.

Aiken, R. B. (1992). The mating-behaviour of a boreal water beetle, dystiscus-alaskanus (coleopteran, Dystiscidae). Ethology, Ecology and Evolution 4, 245-254.

Aiken, R. B., and Khan, A. (1992). The adhesive strength of the palettes of males of a boreal water beetle, Dytiscus alakanus J. Balfour Browne (Coleoptera: Dytiscidae). Canadian Journal of Zoology 70, 1321-1324.

Alcock, J., Jones, C. E., and Buchmann, S. J. (1977). Male mating strategies in bee Centris pallida Fox (Anthophoridae: Hymenoptera). American Naturalist 111, 145-155.

Alexander, R. D., Marshall, D. C., Cooley, J. R., Choe, J. C., and Crespi, B. J. (1997). Evolutionary perspectives on insect mating. In ‘Mating Systems in Insects and Arachnids’. (Ed. J. C. Choe and B. J. Crespi). pp. 4–31. (Cambridge University Press: Cambridge.)

Andersen, N. M. (1973). Seasonal polymorphism and developmental changes in organs of flight and reproduction in bivoltine pondskaters (Hemitpera: Gerridae). Insect Systematics and Evolution 4, 1-20.

Andersen, N. M. (1982). The semiaquatic bugs (Hemiptera, Gerromorpha). Phylogeny, adaptations, biogeography, and classification. Entomonograph 3, 1-455.

Andersen, N. M. (1991). Marine insects: genital morphology, phylogeny and evolution of sea skaters, genus Halobates (Hemiptera, Gerridae). Zoological Journal of the Linnean Society 103, 21-60.

Andersen, N. M. (1993). The evolution of wing polymorphism in water striders (Gerridae): a phylogenetic approach. Oikos 67, 433-443.

Andersen, N. M. (1994). The evolution of sexual size dimorphism and mating systems in water striders (Hemiptera: Gerridae): a phylogenetic approach. Ecoscience 1, 208-214.

Andersen, N. M. (1997). A phylogenetic analysis of the evolution of sexual dimorphism and mating systems in water striders (Hemiptera: Gerridae). Biological Journal of the Linnean Society 61, 345-368.

Andersen, N. M. (1999). The evolution of marine insects: phylogenetic, ecological and geological aspects of species diversity in marine water striders. Ecography 22, 98-111.

Andersen, N. M. (2000). Microvelia polhemi n. sp. (Hemitpera: Veliidae) from Dominican amber: the first fossil record of a phytotelmic water strider. Journal of New York Entomological Society 107, 135-144.

102

Andersen, N. M. (2001). Fossil water striders in the Oligocene/Milocene Dominican amber (Hemiptera: Gerromorph). Insect Systematics & Evolution 31, 411-431.

Andersen, N. M., and Cheng, L. (2005). The marine insect Halobates (Heteroptera: Gerridae): biology, adaptations, distribution, and phylogeny. Evolution 66, 40-54.

Andersen, N. M., and Weir, T. A. (1999). The marine Haloveliinae (Hemiptera: Veliidae) of Australia, New Caledonia and southern New Guinea. Invertebrate Systematics 13, 309-350.

Andersen, N. M., and Weir, T. A. (2001). New genera of Veliidae (Hemiptera-Heteroptera) from Australia, with notes on the generic classification of the Microveliinea. Invertebrate Taxonomy 15, 217-258.

Andersen, N. M., and Weir, T. A. (2003). The genus of Microvelia Westwood in Australia (Hemiptera: Heteroptera: Veliidae). Invertebrate Systematics 17, 261-348.

Andersen, N. M., and Weir, T. A. (Ed.) (2004). ‘Australian water bugs: their biology and identification: (Hemiptera-Heteroptera, Gerromorpha & Nepomorpha).’ (CSIRO Publishing: Melbourne.)

Andersen, N. T., and Weir, T. A. (1994). The sea skaters, genus Halobates Eschscholtz (Hemiptera: Gerridae), of Australia: taxonomy, phylogeny and zoogeography. Invertebrate Systematics 8, 861-909.

Andersson, M. B. (Ed.) (1994). ‘Sexual Selection.’ (Princeton University Press: New Jersey.)

Andrés, J. A., and Cordero, A. (1999). The inheritance of female colour morphs in the damselfly Ceriagrion tenellum (Odonata, Coenagrionidae). Heredity 82, 328-335.

Andrés, J. A., Sanchez-Guillen, R. A., and Rivera, A. C. (2002). Evolution of female colour polymorphism in damselflies: Testing the hypothesis. Animal Behaviour 63, 677-685.

Arak, A. (1988). Sexual dimorphism in body size: a model and a test. Evolution 42, 820-825.

Arnold, S. J., and Duvall, D. (1994). Animal mating systems: A synthesis based on selection theory. American Naturalist 143, 317-348.

Arnqvist, G. (1989a). Multiple mating in water strider: Mutual benefits or intersexual conflict? Animal Behaviour 38, 709-756.

Arnqvist, G. (1989b). Sexual selection in a water strider: the function, mechanism of selection and heritability of male grasping apparatus. Oikos 56, 344-350.

Arnqvist, G. (1992). Pre-copulatory fighting in a water strider: inter-sexual conflict or mate assessment? Animal Behaviour 43, 559-567.

103

Arnqvist, G. (1997a). The evolution of water strider mating systems: Causes and consequences of sexual conflict. In ‘The Evolution of Mating Systems in Insects and Arachnids’. (Ed. J. C. Choe and B. J. Crespi.) pp. 146-163. (Cambridge University Press: Cambridge.)

Arnqvist, G. (1997b). The evolution of animal genitalia: Distinguishing between hypotheses by single species studies. Biological Journal of the Linnaean Society 60, 365-379.

Arnqvist, G. (1998). Comparative evidence for the evolution of genitalia by sexual selection. Nature 393, 784-786.

Arnqvist, G., and Nilsson, T. (2000). The evolution of polyandry: Multiple mating and female fitness in insects. Animal Behaviour 60, 145-164.

Arnqvist, G., and Rowe, L. (1995). Sexual conflict and arms races between the sexes: A morphological adaptation for control of mating in a female insect. Proceedings of the Royal Society of London Series B – Biological Sciences 261, 123-127.

Arnqvist, G., and Rowe, L. (2002a). Antagonistic coevolution between the sexes in a group of insects. Nature 415, 787-789.

Arnqvist, G., and Rowe, L. (2002b). Correlated evolution of male and female morphologies in water striders. Evolution 56, 936-947.

Arnqvist, G., and Rowe, L. (2005). ‘Sexual Conflict.’ (Princeton University Press: New Jersey.)

Arnqvist, G., and Rowe, L. (2013). Sexual Conflict Prior to Mating. In ‘Sexual Conflict’. pp. 92- 149. (Princeton University Press: New Jersey.)

Arnqvist, G., Edvardsson, M., Friberg, U., and Nilsson, T. (2000). Sexual conflict promotes speciation in insects. Proceedings of the National Academy of Sciences of the United States of America 97, 10460-10464.

Arnqvist, G., Jones, T. M., and Elgar, M. A. (2003). Insect behaviour: reversal of sex roles in nuptial feeding. Nature 424, 387-387.

Arnqvist, G., Jones, T. M., and Elgar, M. A. (2006). Sex-role reversed nuptial feeding reduces male kleptoparasitism of females in Zeus bugs (Heteroptera; Veliidae). Biology Letters 4, 491-493.

Arnqvist, G., Jones, T. M., and Elgar, M. A. (2007). The extraordinary mating system of Zeus bugs (Heteroptera: Veliidae: Phoreticovelia sp.). Australian Journal of Zoology 55, 131-137.

Bateman, A. J. (1948). Intra-sexual selection in Drosophila. Heredity 2, 349-368.

Bergroth, E. (1916). Heteropteroud Hemiptera collected by Professor W. Baldwin Spenser during the Horn Expedition into Central Australia. Proceedings of the Royal Society of Victoria 29, 19-39.

104

Bergsten, J. (2005). Taxonomy, phylogeny, and secondary sexual character evolution of diving beetles, focusing on the genus Acilius. Doctoral Dissertation, Umeå University, Department of Ecology and Environmental Sciences, Umeå, Sweden.

Bergsten, J., and Miller, K. B. (2007). Phylogeny of diving beetles reveals a coevolutionary arms race between the sexes. PLoS ONE 2, 502-522.

Bergsten, J., Töyra, A. and Nilsson, A. N. (2001). Intraspecific variation and intersexual correlation in secondary sexual characters of three diving beetles (Coleoptera: Dytiscidae). Biological Journal of the Linnean Society 73, 221-232.

Bilto, D. T., Thompson, A., and Foster, G. N. (2008). Inter- and intrasexual dimorphism in the diving beetle Hydroporus memnonius Nicolai (Coleoptera: Dytiscidae). Biological Journal of the Linnean Society 94, 685-697.

Birkhead, T. R., and Møller, A. P. (Ed.) (1992). ‘Sperm Competition in Birds: Evolutionary Causes and Consequences.’ (Academic Press: London.)

Bisazza, A., Manfredi, S., and Pilastro, A. (2000). Sexual competition, coercive mating and mate assessment in the one-sided livebearer, Jenynsia multidentata: Are they predictive of sexual dimorphism? Ethology 106, 961-978.

Blackith, R. E. (1957). Polymorphism in some Australian locusts and grasshoppers. Biometrics 13, 183-196.

Blanckenhorn, W. U., Hosken, D. J., Martin, O. Y., Reim, C., Teuschl, Y., and Ward, P. I., (2002). The costs of copulating in the dung fly Sepsis cynipsea. Behavioural Ecology 13, 353- 358.

Bonduriansky, R., and Chenoweth, S. F. (2009). Intralocus sexual conflict. Trends in Ecology & Evolution 24, 280-288.

Booth, C. (2016). The things we do for genes. Wildlife Australia 53, 32-32.

Brower, A. V. Z. (1997). The evolution of ecologically important characters in Heliconius butterflies (Lepidoptera: Nymphalidae): A cladistic review. Zoological Journal of the Linnean Society 199, 457-472.

Burke, T., Davies, N. B., Bruford, M. W., and Hatchwell, B. J. (1989). Parental care and mating behaviour of polyandrous dunnocks Prunella modularis related to paternity by DNA fingerprinting. Nature 338, 249-251.

Bush Heritage Hamelin. (2017). Bush Heritage Australia. https://www.bushheritage.org.au/places-we-protect/western-australia/hamelin

Buzatto, B. A., Tomkins, J. L., and Simmons, L. W. (2014). Alternative phenotypes withinmating strategies. In ‘The Evolution of Insect Mating Systems’. pp. 106-128. (Oxford University Press: United States of America.)

105

Carayon, J. (1966). Traumatic insemination and paragenital systems. In ‘Monograph of Cimicidae (Hemiptera, Heteroptera)’. (Ed. R. L. Usinger.) pp. 81-166. (Entomological Society of America, College Park: Maryland.)

Cassis, G., Hodgins, M., Weir, T. A., and Taranic, N. J. (2017). Phylogenetic reclassification and genitalic morphology of the small water strider genus Nesidovelia Andersen & Weir and allied Microveliinae (Hemiptera: Veliidae). Austral Entomology 56, 246-561.

Chalmers, L., and Wheeler, J. (1997). Native vegetation of freshwater rivers and creeks in south Western Australia. Waters and Rivers Commission (Waters and Rivers Commission: East Perth.)

Chapman, A. D. (2009). Numbers of living species of Australia and the World. Report for the Australian Biological Resources Study. (Department of the Environment, Water, Heritage and the Arts: Canberra.)

Chapman, T., and Partridge, L. (1996). Sexual conflict as fuel for evolution. Nature 381, 189- 190.

Chapman, T., Arnqvist, G., Bangham, J. and Rowe, L. (2003). Sexual conflict. Trends in Ecology & Evolution 18, 41-47.

Cheesman, L. E. (1926). XLIX. – Two new species of Veliidae from the SE Pacific. Journal of Natural History 18, 364-367.

Civetta, A., and Clark, A. G. (2000). Correlated effects of sperm competition and post mating female mortality. Proceedings of the National Academy of Sciences of the United States of America 97, 13162-13165.

Clutton-Brock, T. H., and Parker, G. A. (1992). Potential reproductive rates and the operation of sexual selection. Quarterly Review of Biology 67, 437-456.

Clutton-Brock, T. H., and Parker, G. A. (1995). Sexual coercion in animal societies. Animal Behaviour 49, 1345-1365.

Cook, S. E., Vernon, J. G., Bateson, M., and Guilford, T. (1994). Mate choice in the polymorphic African swallowtail butterfly, Papilio dardanus: male-like females may avoid sexual harassment. Animal Behaviour 47, 389-397.

Cordero, A. (1990). The inheritance of female polymorphism in the damselfly Ischnura graellsii (Rambur) (Odonata, Coenagrionidea). Heredity 64, 341-346.

Cordero, A. (1992). Density-dependent mating success and colour polymorphism in females of the damselfly Ischnura graellsii (Odonata, Coenagrionidae). Journal of Animal Ecology 61, 769-780.

106

Cordero, A., and Andrés, J. A. (1996). Colour polymorphism in odonates: females that mimics males. Journal of the British Dragonfly Society 12, 50-60.

Cordero, A., Carbone, S. S., and Utzeri, C. (1998). Mating opportunities and mating costs are reduced in androchrome female damselflies, Ischnura elegans (Odonata). Animal Behaviour 55, 185-197.

Crump, M. L. (1988). Aggression in harleguin frogs: Male-male competition and a possible conflict of interest between the sexes. Animal Behaviour 36, 1064-1077.

Daly, M. (1978). The costs of mating American Naturalist 112, 771-774.

Darwin, C. (Ed.) (1871). ‘The Descent of Man and Selection in Relation to Sex.’ (J. Murray: London.)

Davis, N. T. (1965a). Studies of the reproductive physiology of the Cimicidae (Hemiptera). II. Artificial insemination and the function of the seminal fluid. Journal of Insect Physiology 11, 355-366.

Davis, N. T. (1965b). Studies on the reproductive physiology of the Cimicidae (Hemiptera). III. The seminal stimulus. Journal of Insect Physiology 11, 1199-1211.

Deinert, E. I., Longino, J. J., and Gilbert, L. E. (1994). Mate competition in butterflies. Nature 370, 23-24.

Dell, B., and Bennet, I. J. (Ed.) (1986). ‘The flora of Murdoch University: A guide to the native plants on campus.’ (Murdoch University: Perth.)

Dennis, D. S., and Lavigne, R. J. (1976). Ethology of Efferia varipes with comments on species coexistence (Diptera: Asilidae). Journal of the Kansas Entomological Socitey 49, 48-62.

Dennis, D., Lavigne, R. J., and Bullington, S. W. (1986). Ethology of Efferia cressoni with a review of the comparative ethology of the genus (Diptera: Asilidae). Proceedings of the Entomological Society of Washington 88, 42-55.

Ditrich, T., and Papáček, M. (2009). Effective strategy of the overwintering of semiaquatic bugs: overwintering of Velia caprai (Heteroptera: Gerromorpha: Veliidae). Journal of Natural History 43, 529-543.

DPaW Kalbarri National Park. (2017). Department of Parks and Wildlife. https://parks.dpaw.wa.gov.au/park/kalbarri

DPaW Nature Map. (2017). Department of Parks and Wildlife (Nature Map: Matting Western Australia’s biodiversity). https://naturemap.dpaw.wa.gov.au/

DPaW Walpole Nornalup National Park. (2017). Department of Parks and Wildlife. https://parks.dpaw.wa.gov.au/park/walpole-nornalup

107

Eberhard, W. (2006). Sexually antagonistic coevolution in insects is associated with only limited morphological diversity. Journal of Evolutionary Biology 19, 657–681.

Eberhard, W. G. (2002). The function of female resistance behaviour: Intromission by male coercion vs. female cooperation in sepsid flies (Diptera: Sepsidae). Revista de Biologia Tropical 50, 485-505.

Eberhard, W. G. (Ed.) (1985). ‘Sexual Selection and Animal Genitalia.’ (Harvard University Press: Cambridge.)

Eberhard, W. G. (Ed.) (1996). ‘Female Control: Sexual Selection by Cryptic Female Choice.’ (Princeton University Press: Princeton, New Jersey.)

Emlen, D. J. (2001). Costs and the diversification of exaggerated animal structures. Science 291, 1534-1536.

Emlen, D. J., and Nijhout, H. F. (2000). The development and evolution of exaggerated morphologies in insects. Annual Review of Entomology 45, 661-708.

Emlen, S. T. (1978). Cooperative breeding. In ‘Behavioural Ecology’. (Eds. J. R. Krebs and N. B. Davies.) pp. 245-281. (Blackwell: Oxford.)

Evans, E. W. (1982). Consequences of body size for fecundity in the predatory stinkbug, Podisus maculiventris (Hemiptera: Pentatomidae). Annals of the Entomological Society of America 75, 418-420.

Fairbairn, D. J. (1990). Factors influencing sexual size dimorphism in temperate water striders. The American Naturalist 136, 61-86.

Fairbairn, D. J. (1992). The origins of allometry: size and shape polymorphism in the common waterstrider, Gerris remiges (Heteroptera, Gerridae). Biological Journal of the Linnean Society 45, 167-186.

Fairbairn, D. J. (1993). The costs of loading associated with mate-carrying in the water strider Aquarius remigis. Behavioral Ecology 4, 224-231.

Fairbairn, D. J. (1997). Allometry for sexual size dimorphism: pattern and process in the coevolution of body size in males and females. Annual review of ecology and systematics 28, 659-687.

Fincke, O. M., Jödicke, R., Paulson, D., and Schultz, T. D. (2005). The evolution and frequency of female color morphs in Holarctic Odonata: why are male-like morphs typically the minority? International Journal of Odonatology 8, 183-212.

Fisher, R. A. (1915). The evolution of sexual preference. Eugenics Review 7, 184-192.

Forbes, M. R. L. (1991). Female morphs of the damselfly Enallagma boreale selys (Odonata: Coenagrionidae): A benefit for androchromatypes. Canadian Journal of Zoology 69, 1969- 1970.

108

Forsman, A., and Shine, R. (1995). Sexual size dimorphism in relation to frequency of reproduction in turtles (Testudines: Emydidae). Copeia 1995, 727-729.

Gavrilets, S. (2000). Rapid evolution of reproductive barriers driven by sexual conflict. Nature 403, 886-889.

Gavrilets, S., Arnqvist, G., and Friberg, U. (2001). The evolution of female mate choice by sexual conflict. Proceedings of the Royal Society London B 268, 531-539.

Gilbert, N. (1980). Comparative dynamics of a single-host aphid. I. The evidence. The Journal of Animal Ecology 49, 331-369.

Gowaty, P. A., and Buschhaus, N. (1998). Ultimate causation of aggressive and forced copulation in birds: female resistance, the CODE hypothesis, and social monogamy. American Zoologist 38, 207-225.

Green, K. K., Kovalev, A., Svensson, E. I., and Gorb, S. N. (2013). Male clasping ability, female polymorphism and sexual conflict: fine-scale elytral morphology as a sexually antagonistic adaptation in female diving beetles. Journal of Royal Society 86, 1-10.

Hale, H. M. (1926). Studies in Australian aquatic Hemiptera, No. VII. Records of the South Australian Museum 3, 23-28.

Hamilton, W. D. (1990). Mate choice near and far. Animal Zoology 30, 341-352.

Han, C. S., and Jablonski, P. G. (2009). Female genital concealment promotes intimate male courtship in a water strider. PLoS One 4, 5793.

Härding, R., Smith, H. G., Jormalainen, V., and Tuomi, J. (2001). Resolution of evolutionary conflicts: Costly behaviours enforce the evolution of cost-free competition. Evolutionary Ecology Research 3, 829-844.

Härdling, R. (1999). Arms races, conflict costs and evolutionary dynamics. Journal of Theoretical Biology 196, 163-167.

Harrison, R. G. (1980). Dispersal polymorphisms in insects. Annual Review of Ecology, Evolution and Systematics 11, 95-118.

Hinnekint, B. O. N. (1987). Population dynamics of Ischnura e. elegans (Vander Linden) (Insecta: Odonata) with special reference to morphological colour changes, female polymorphism, multiannual cycles and their influence of behaviour. Hydrobiologia 146, 3- 31.

Holland, B., and Rice, W. R. (1998). Chase-away sexual selection: antagonisitc seduction versus resistance. Evolution 52, 1-7.

Honěk, A. (1993). Intraspecific variation in body size and fecundity in insects: a general relationship. Okios 66, 483-492.

109

Hrdy, S. B. (1977). ‘The Langurs of Abu.’ (Harvard University Press: Cambridge, MA.)

Ingram, K. K., Laamanen, T., Puniamoorthy, N., and Meier, R. (2008). Lack of morphological cooevolution between male forelegs and female wings in Themira (Sepsidae: Diptera: Insecta). Biological Linnean Society 93, 227-238.

Inoda, T., Hardling, R., and Kubota, S. (2012). The inheritance of intrasexual dimorphism in female diving beetles (Coleoptera: Dystiscidae). Zoological Science 29: 505-509.

Järvinen, O., and Vepsäläinen, K. (1976). Wing dimorphism as an adaptive advantage strategy in water-striders (Gerris). Hereditas 84, 61-68.

Jennions, M. D., Møller A. P., and Petrie, M. (2001). Sexually selected traits and adult survival: A meta-analysis. Quarterly Review of Biology 76, 3-36.

Johnson, C. (1964). The inheritance of female dimorphism in damselfly Ischnura damula. Genetics 49, 513-519.

Johnson, C. (1966). Genetics of female dimorphism in Ischnura demorsa. Heredity 21, 453- 459.

Johnson, C. (1975). Polymorphism and natural selection in ischnuran damselflies. Evolutionary Theory 1, 81-90.

Jones, T. M., Elgar, M. A., and Arnqvist, G. (2010). Extreme cost of male riding behaviour for juvenile females of the Zeus bug. Animal Behaviour 79, 11-16.

Jormalainen, V. (1998). Precopulatory mate guarding in crustaceans: Male competitive strategy and intersexual conflict. Quarterly Review of Biology 73, 275-304.

Khelifa, R. (2017). Faking death to avoid male coercion: extreme sexual conflict resolution in a dragonfly. Ecology 98, 1724-1726.

Kingsolver, J. G., Hoekstra, H. E., Hoekstra, J. M., Berrigan, D., Vignieri, S. N., Hill, C. E., Hoang, P., Gilbert, P., and Beerli, P. (2001). The strength of phenotypic selection in natural populations. American Naturalist 157, 245-261.

Knell, R. J., Pomfret, J. C., and Tomkins, J. L. (2004). The limits of elaboration: curved allometries reveal the constraints on mandible size in stag beetles. Proceedings of the Royal Society of London B 271, 523-528.

Kokko, H., Brooks, R., Jennions, M. D., and Morely, J. (2003). The evolution of mate choice and mating biases. Proceedings of the Royal Society of London Series B – Biological Sciences 270, 653-664.

Krupa, J. J., Leopold, W. R., and Sih, A. (1990). Avoidance of male giant water striders by females. Behaviour 115, 247-253.

110

Kvarnemo, C., and Ahnesjö, I. (1996). The dynamics of operational sex ratios and competition for mates. Trends in Ecology & Evolution 11, 404-408.

Laird, M. (1956). Studies of mosquitos and freshwater ecology in the South Pacific. Bulletin of the Royal Society of New Zealand 6, 1-213.

Lauer, M. J. (1996). Effect of sperm depletion and starvation on female mating behaviour in the water strider, Aquarius remigis. Behavioural Ecology and Sociobiology 38, 89-96.

Leong, K. L. H., Frey, D., Hamaoka, D., and Honma, K. (1993). Wing damage in overwintering populations of monarch butterfly at two California sites. Annals of the Entomological Society of America 86, 728-733.

Lundbald, O. M. (1933). Zur Kenntnis der aquatilen und semi-aquatilen Hemipteren von Sumatra, Java und Bali. Archive for Hydrobiology 12, 1-195.

Magnhagen, C. (1991). Predation risk as a cost of reproduction. Trends in Ecology and Evolution 6, 183-185.

Malipatil, M. B. (1980). Review of Australian Microvelia Westwood (Hemiptera: Veliidae) with the description of two new species from eastern Australia. Australian Journal of Marine and Freshwater Research 31, 85-108.

Markow, T. A. (2000). Forced matings in natural populations of Drosophila. American Naturaliast 156, 100-103.

Matsuda, R. (1960). Morphology, evolution and a classification of the Gerridae (Hemiptera- Heteroptera). The University of Kansas Science Bulletin 41, 25-632.

Menke, A. S. (1979). The semiaquatic and aquatic Hemiptera of California (Heteroptera: Hemiptera). Bulletin of the California Insect Survey 21, 1-166.

Miller, C. W., and Somjee, U. (2014). Male-male competition. In ‘Oxford Bibliographies in Evolutionary Biology’. (Ed. J. Losos.) pp. 208-234. (Oxford University Press: New York.)

Miller, K. B. (2003). The phylogeny of diving beetles (Coleoptera: Dytiscidae) and the evolution of sexual conflict. Biological Journal of the Linnean Society 79, 359-388.

Miller, N. M., and Fincke, O. M. (1999). Cues for mate recognition and the effect of prior experience on mate recognition in Enallagma damselflies. Journal of Insect Behaviour 12, 801-814.

Moczek, A. P. (2010). Phenotypic plasticity and diversity of insects. Philosophical Transactions of the Royal Society of London B 365, 593-603.

Moczek, A. P. and Nijhout, H. F. (2004). Trade-offs during the development of primary and secondary sexual traits in a horned beetle. The American Naturalist 163, 184-191.

111

Møller, A. P. (1988). Female choice selects for male sexual tail ornaments in the monogamous swallow. Nature 322, 640-642.

Møller, A. P. (1998). Sperm Competition and Sexual Selection. In ‘Sperm Competition and Sexual Selection’. (Ed. T. R. Birkhead and A. P. Møller.) pp. 55-90. (Cambridge University Press: Cambridge.)

Morrow, E. H., and Arnqvist, G. (2003). Costly traumatic insemination and a female counteradaptation in bed bugs. Proceedings of the Royal Society of London Series B – Biological Sciences 270, 2377-2381.

Mühlhäuser, C., and Blanckenhorn, W. U. (2002). The costs of avoiding matings in the dung fly Sepsis cynipsea. Behavioural Ecology 13, 359-365.

Muraji, M. M., and Nakasui, F. (1989). Change in photoperiodic sensitivity during hibernation in a semi-aquatic bug, Microvelia douglasi (Heteroptera: Veliidae). Applied Entomology and Zoology 24, 450-457.

Newberry, K. (1989). The effects of domestic infestation of Cimex lectularius bedbugs of interspecific mating with Cimex hemipterus. Medical and Veterinary Entomology 3, 407-414.

O’Neill, K. M. (1985). Egg size, prey size, and sexual size dimorphism in digger wasps (Hymenoptera: Sphecidae). Canadian Journal of Zoology 63, 2187-2193.

Ortigosa, A., and Rowe, L. (2003). The role of mating history and male size in determining mating behaviours and sexual conflict in a water strider. Animal Behaviour 65, 851-858.

Parker, G. A. (1974). Courtship persistence and female guarding as male time investment strategies. Behaviour 48, 157-184.

Parker, G. A. (1979). Sexual selection and sexual conflict. In ‘Sexual Selection and Reproductive Competition in Insects’. (Ed. M. S. Blum and N. A. Blum.) pp. 123-166. (Academic Press: New York.)

Parker, G. A. (1983). Arms race in evolution: An ESS to the opponent-independent costs game. Journal of Theoretical Biology 101, 141-166.

Parker, G. A., and Partridge, L. (1998). Sexual conflict and speciation. Philosophical Transactions of the Royal Society of London B 353, 261-274.

Parker, G. A., and Simmons, L. W. (1996). Parental investment and the control of sexual selection: Predicting the direction of sexual competition. Proceedings of the Royal Society of London Series B – Biological Sciences 263, 315-321.

Patridge, L., and Hurst, L. D. (1998). Sex and conflict. Science 281, 2003-2008.

Pizzari, T., and Snook, R. R. (2003). Perspective: Sexual conflict and sexual selection: chasing away paradigm shifts. Evolution 57,1223-1236.

112

Polhemus, J. T. (1982). Marine Hemiptera of the northern territory, including the freshwater species of Halobates eschscholtz (Gerridae. Veliidae, Hermatobatidae and Corixidae). Australian Journal of Entomology 21, 5-11.

Polhemus, J. T. (1999). Two new species of Microvelia from treeholes, with notes on other container-inhabiting veliid speces (Heteroptera: Veliidae). Journal of the New York Entomological Society 107, 31-37.

Polhemus, J. T., and Hogue, C. L. (1972). Two new Microvelia from crabholes in Costa Rica (Hemiptera: Veliidae). Contributions to Science, Natural History Museum, Los Angeles 224, 1- 6.

Polhemus, J. T., and Polhemus, D. A. (1991). A review of the veliid fauna of bromeliads, with a key and description of a new species (Heteroptera: Veliidae). Journal of the New York Entomological Society 99, 204-216.

Pomiankowski, A. (1988). The evolution of female mate preference for male genetic quality. Oxford Survey of Evolutionary Biology 5, 136-184.

Reinhardt, K., Harney, E., Naylor, R., Gorb, S., and Siva-Jothy, M. T. (2007). Female‐limited polymorphism in the copulatory organ of a traumatically inseminating insect. The American Naturalist 170, 931-935.

Reinhardt, K., Naylor, R., and Siva-Jothy, M. T. (2003). Reducing a cost of traumatic insemination: Female bedbugs evolved a unique organ. Proceedings of the Royal Society of London Series B – Biological Sciences 270, 2371-2375.

Reiss, M. J. (Ed.) (1989). ‘The allometry of growth and reproduction.’ (Cambridge University Press: Cambridge, New York.)

Reynolds, J. D. (1996). Animal breeding systems. Trends in Ecology & Evolution 2, 68-72.

Rice, W. R. (1996). Sexually antagonistic male adaptation triggered by experimental arrest of female evolution. Nature 381, 232-234.

Rice, W. R. (1998). Species and Speciation. In ‘Endless Forms’. (Ed. D. J. Howard and S. H. Berlocher.) pp. 261-270. (Oxford University Press: Oxford.)

Rice, W. R. (2000). Dangerous liaisons. Proceedings of the National Academy of Sciences of the Untied States of America 97, 12953-12955.

Rice, W. R., and Holland, B. (1997). The enemies within: intergenomic conflict, interlocus contest evolution (ICE), and the intraspecific Red Queen. Behavioural Ecology and Sociobiology 41, 1-10.

Richards, O. W. (1927). Sexual selection and allied problems in the insects. Biological Reviews 2, 298-364.

113

Robertson, H. M. (1985). Female dimorphism and mating behaviour in a damselfly, Ischnura ramburi: Females mimicking males. Animal Behaviour 33, 805-809.

Roff, D. (1981). On being the right size. American Naturalist 1, 405-422.

Roff, D. A. (1986). The evolution of wing dimorphism in insects. Evolution 40, 1009-1020.

Rowe, L. (1992). Convenience polyandry in a water strider: foraging conflicts and female control of copulation frequency and guarding duration. Animal Behaviour 44, 189-202.

Rowe, L. (1994). The costs of mating and mate choice. Animal Behaviour 48, 1049-1056.

Rowe, L., and Arnqvist, G. (2002). Sexually antagonistic coevolution in a mating system: Combining experimental and comparative approaches to address evolutionary processes. Evolution 56, 754-767.

Rowe, L., Arnqvist, G., Sih, A., and Krupa, J. (1994). Sexual conflict and the evolutionary ecology of mating patterns: Water striders as a model system. Trends in Ecology & Evolution 9, 289-293.

Rowe, L., Krupa, J. J., and Sih, A. (1996). An experimental test of condition dependent mating behaviour and habitat choice by water striders in the wild. Behavioural Ecology 7, 474-479.

Rowe, L., Westlake, K. P., and Currie, D. C. (2006). Functional significance of elaborate secondary sexual traits and their evolution in the water strider genus Rheumatobates. The Canadian Entomologist 138, 568-577.

Ryckman, R. E., and Ueshima, N. (1964). Biosystematics of the Hesperocimex complex (Hemiptera: Cimicidae) and avian hosts. Annals of the Entomological Society of America 57: 624-638.

Sakaluk, S. K., Bangert, P. J., Eggert, A. K., Gack, C., and Swanson, L. V. (1995). The gintrap as a device for facilitating coercive mating in sagebush crickets. Proceedings of the Royal Society of London Series B – Biological Sciences 261, 65-71.

Saulich, A. H., and Musolin. (Ed.) (2007). ‘Seasonal Development of Aquatic and Semiaquatic True Bugs (Heteroptera).’ (St Petersburg University Press: Russia).

Schuh, R. T, and Stys, P. (1991). Phylogenetic analysis of cimicomorphan family relationship (Heteroptera). Journal of the New York Entomological Society 99, 298-350.

Schuh, R. T., and Slater, J. A. (Ed.) (1995). ‘True bugs of the world (Hemiptera: Heteroptera): classification of natural history. ‘ (Cornell University Publishing: Ithaca and London).

Sheldon, B. C. (1994). Male phenotype, fertility, and the pursuit of extra-pair copulations by female birds. Proceedings of the Royal Society of London B 257, 25-30.

114

Sherratt, T. N. (2001). The evolution of female-limited polymorphisms in damselflies: A signal detection model. Ecology letters 4, 22-29.

Shine, R. (1988). The evolution of large body size in females: a critique of Darwin’s “fecundity advantage” model. The American Naturalist 131, 124-131.

Shuker, D. (2014). Sexual selection theory. In ‘The Evolution of Insect Mating Systems’. (Eds. D. M. Shuker and L. W. Simmons.) pp. 20-41. (Oxford University Press: United States of America.)

Sih, A., Krupa, J. and Travers, S. (1990). An experimental study on the effects of predation risk and feeding regime on the mating behaviour of the water strider. American Naturalist 135, 284-290.

Sih, A., Lauer, M., and Krupa, J. J. (2002). Path analysis and the relative importance of male female conflict, female choice and male-male competition in water striders. Animal Behaviour 63, 1079-1089.

Simmons, L. W., and Tomkins, J. L. (1996). Sexual selection and the allometry of earwing forceps. Evolutionary Ecology 10, 97-104.

Simpson, S. J., Sword, G. A., Lo, N. (2011). Polyphenism in insects. Current Biology 21, 738- 739.

Singer, F., Riechert, S. E., Xu, H. F., Morris, A. W., Becker, E., Hale, J. A., and Noureddine, M. A. (2000). Analysis of courtship success in the funnel-web spider Agelenopsis aperta. Behaviour 137, 93-117.

Sirot, L. K., and Brockmann, H. J. (2001). Costs of sexual interactions to females in Rambur’s forktail damselfly, Ischnura ramburi (Zygoptera: Coenagrionidae). Animal Behaviour 61, 415- 424.

Slooten, E., and Lambert, D. M. (1983). Evolutionary studies of the New Zealand coastal mosquito Opifex fuscus (Hutton). I. Mating behaviour. Behaviour 84, 157-172.

Smith, S. M. (1988). Extra-pair copulations in black-capped chickadees: The role of the female. Behaviour 107, 15-23.

Smuts, B. B., and Smuts, R. W. (1993). Male-aggression and sexual coercion of females in nonhuman-primates and other mammals – evidence and theoretical implications. Advances in the Study of Behaviour 22, 1-63.

Stone, G. N. (1995). Female foraging responses to sexual harassment in the solitary bee Anthophore plumipes. Animal Behaviour 50, 405-412.

Stutt, A. D. (1999). Reproductive strategies and sexual conflict in the bed bug. Ph.D. Thesis, Univeristy of Sheffield, England.

115

Stutt, A. D., and Siva-Jothy, M. T. (2001). Traumatic insemination and sexual conflict in the bedbug Cimex lectularius. Proceedings of the National Academy of Sciences of the United States of America 98, 5683-5687.

Swanson, W. J., Yang, Z., Wolfner, M. F., and Aquadro, C. F. (2001). Positive Darwinian selection drives the evolution of several female reprodutice proteins in mammals. Proceedings of the National Academy of Sciences of the United States of America 98, 2509- 2514.

Tatarnic, N. J., and Cassis, G. (2008). Revision of the plant bug genus Coridromius Signoret (Insecta: Heteroptera: Miridae). Bulletin of the American Museum of Natural History 315, 1- 95.

Tatarnic, N. J., and Cassis, G. (2010). Sexual coevolution in the traumatically inseminating plant bug genus Coridromius. Journal of Evolutionary Biology 23, 1321-1326.

Tatarnic, N. J., Cassis, G., and Hochuli, D. T. (2006). Traumatic insemination in the plant bug genus Coridromius Signoret (Heteroptera: Miridae). Biology Letters 2, 58-61.

Tatarnic, N. J., Cassis, G., and Siva-Jothy, M. T. (2014). Traumatic insemination in terrestrial arthropods. Annual Review of Entomology 59, 245-261.

Tatarnic, N. J., Hodgins, M., Weir, T. A., and Cassis, G. (in prep). Sexual conflict and the coevolution of sexually antagonistic armature in the ripple bug genus Nesidovelia (Vellidae).

Thornhill, R. (1976). Sexual selection and nuptial feeding behaviour in Bittacus apicalis (Insecta: Mecoptera). American Naturalist 110, 529-548.

Thornhill, R. (1980). Rape in panorpa scorpionflies and a general rape hypothesis. Animal Behaviour 28, 52-59.

Thornhill, R. (1984). Alternative hypotheses for traits believed to have evolved by sperm competition. In ‘Sperm Competition and the Evolution of Animal Mating Systems’. (Ed. R. L. Smith.) pp. 367-383. (Academic: Orlando, Fla.)

Thornhill, R., and Alcock, J. (Eds.) (1983). ‘The Evolution of Insect Mating Systems.’ (Harvard University Press: Cambridge University.)

Thornhill, R., and Sauer, K. P. (1991). The notal organ of the scorpionfly (Panorpa vulgaris): An adaptation to coerce mating duration. Behavioural Ecology 2, 156-164.

Trivers, R. L. (1972). Parental investment and sexual selection. In ‘Sexual Selection and the Descent of Man’. (Ed. B. Campbell.) pp. 136-179. (Aladine: Chicago.)

Usinger, R. L. (Ed.) (1966). ‘Monograph of Cimicidae (Hemiptera, Heteroptera).’ (Entomological Society of America: College Park, Md.)

116

Vepsäläinen, K. (1974). The wing lengths, reproductive stages and habitatas of Hungarian Gerris Fabr. species (Heteroptera, Gerridae). Annales Academiae Scientiarum Fennicae 202, 1-18.

Vepsäläinen, K. (1978). Wing dimorphism and diapause in Gerris: determination and adaptive significance. In ‘Evolution of insect migration and diapause’. (Ed. H. Dingle.) pp. 218-253. (Springer: New York.)

Vincent, A. C. J. (1994). Seahorses exhibit conventional sex roles in mating competition despite male pregnancy. Behaviour 128, 135-151.

Walker, W. F. (1980). Sperm utilization strategies in nonsocial insects. American Naturalist 115, 780-799.

Walpole, D. E. (1988). Cross-mating studies between two species of bedbugs (Hemiptera: Cimicidae) with a description of a marker of inter-specific mating. South African Journal of Science 84, 215-216.

Watson, P. J., Arnqvist, G. and Stallmann, R. R. (1998). Sexual conflict and the energetic costs of mating and mate choice in water striders. American Naturalist 151, 46-58.

Weigensberg, I., and Fairbairn, D. J. (1994). Conflicts of interest between the sexes: A study of mating interactions in a semiaquatic bug. Animal Behaviour 48, 893-901.

Weigensberg, I., and Fairbairn, D. J. (1994). Conflicts of interest between the sexes: a study of mating interactions in a semiaquatic bug. Animal Behaviour 48, 893-901.

West-Eberhard, M. J. (1983). Sexual selection, social competition and speciation. Quarterly Review of Biology 58, 155-183.

Westlake, K. P., Rowe, L., and Currie, D. C. (2000). Phylogeny of the water strider genus Rheumatobates (Heteroptera: Gerridae). Systematic Ecology 25, 125-145.

Whitman, D. W., and Ananthakrishnan, N. T. (Ed.) (2009). ‘Phenotypic Plasticity of Insects: Mechanisms and Consequences.’ (Science Publishers: Enfield.)

Wichard, W., Arens, W.m and Eisenbeis, G. (Ed.) (2002). ‘Biological atlas of aquatic insects.’ (Apollo Books: Denmark.)

Wilcox, R. S. (1984). Male copulatory guarding enhances female foraging in a water strider. Behavioral Ecology and Sociobiology 15, 171-174.

Wiley, R. H., and Poston, J. (1996). Perspective: Indirect mate choice, competition for mates, and coevolution of the sexes. Evolution 50, 1371-1381.

Wilkinson, G. S., and Taper, M. (1999). Evolution of genetic variation for condition- dependent traits in stalk-eyed flies. Proceedings of the Royal Society of London B 266, 1685- 1690.

117

Williams, G. C. (Ed.) (1966). ‘Adaptation and Natural Selection; A Crtique of Some Current Evolutionary Thought.’ (Princeton University Press: Princeton, N.J.)

Williams, G. C. (Ed.) (1975). ‘Sex and Evolution.’ (Princeton University Press: Princeton, N.J.)

Wilson, E. O. (2010). Threats to Global Diversity. United Nations Education, Scientific and Cultural Organization Media Services. (UNESCO: Harvard University.)

Wolf, L. L. (1975). ‘Prostitution’ behaviour in a tropical hummingbird. Condor 77, 140-144.

Zahavi, A. (1975). Mate selection: Selection for a handicap. Journal of Theoretical Biology 53, 205-214.

Zahavi, A. (1977). Cost of honesty: (Further remarks on handicap principle). Journal of Theoretical Biology 67, 603-605.

118

Appendix 1

Taxonomic characteristics of the genus Microvelia Westwood

119

Defining features of the genus Microvelia Westwood in Australia (Hemiptera: Heteroptera). Content from Andersen and Weir (2003).

Microvelia Westwood , genus. Characteristic Description Size 1.0-3.4cm (♂, ♀). Apterous (Appendix 3A,3F), Macropterous (Appendix 3I)

Body shape Shape is variable, however usually oval or suboval.

Body colour Colour is chiefly black or brownish, with yellowish, brown markings. Pilosity length is variable but usually is relatively short. Head Head Head shorter than wide, moderately to strongly deflected in front of eyes, not produced posteriorly; head with distinct shiny, median furrow and a pair of pseudocular pits at base. Eyes Eyes relatively large, semiglobular, situated at base of the head. Antennae Antennal tubercules small, situated close to eye margins and barely visible from above; antennae relatively long, 0.4-0.6X total length of insect; segment 1 always shorter than head, surpassing tip of head with less than half of its own length; segment 4 longer than segments 2 and 3, fusiform. Bucculae Ventral lobes of the head (bucculae) highest in the middle, distinctly removed from the anterior margin of prothorax. Rostrum Rostrum relatively slender, its apex reaching middle of mesosternum. Thorax Pronotum Pronotum of apterous form usually longer than head Appendix 3A, 3F), usually with large, transverse pale marking or paired spots anteriorly; anterior margin slightly concave, bordered by dark punctures; pronotal lobe with scattered dark punctures. Pronotum of macropterous form large, pentagonal, with distinctly raised humeral angles (Appendix 3I); anterior collar never formed.

Mesonotum Mesonotum usually completely covered by the pronotal lobe; mesonotum only exposed laterally (Appendix 3A, 3F).

Sutures Ventral sutures of thorax and abdomen fairly distinct; metathoracic scent glands present, with evaporative channels extending laterally from the median scent gland orifice, running close to hind margin of metasternum; small evaporative areas and hair tufts on metacetabula. Legs Middle legs distinctly longer than forelegs, but distinctly shorter than hind legs. Fore femora moderately thickened, but otherwise not modified; fore tibia of male usually with a grasping comb on inner surface before apex (Appendix 3B). Middle femur slender, middle tibia of male with or without a short grasping comb; middle tarsus relatively short, 0.55-0.7x length of tibia; middle tibiae with long row of curved hairs on inner surface; first tarsal segment distinctly shorter than second segment. Hind tibia much longer than middle tibia; first segment of hind tarsus distinctly shorter than second segment. Claws of middle and hind legs long, slender and falcate, with bristle- like aroliae. Wings Of macropterous form long, reaching abdominal end when folded; fore wings usually dark with whitish stripes and spots; venation forming four closed cells (Appendix 3I); basal parts of veins pilose. Abdomen Abdomen Relatively long, sides usually evenly rounded.

Laterotergites Laterotergites 1 small, triangular, connexiva (= laterotergites 2-7, Appendix 3A, 2F) broad, obliquely raised throughout (♂) or sometimes almost vertical and converging to meet each other posteriorly (♀).

120

Appendix 1 continue: Characteristic Description Abdomen Venter Structure of abdominal venter of male variable, simple or modified to some extent, depressed with hair tufts, tubercules and so on.

Genital segment Are of variable shape and relative length, but always protruding from pregenital

abdomen (Appendix 3A,3C); segment 8 sometimes modified ventrally (Appendix 3D); pygophore suboval (Appendix 3c,2d); parameres variable in size and shape, and relatively small, symmetrical; but sometimes large and more or less asymmetrica ; proctiger suboval, sometimes widened basally, with lateral processes (Appendix 3E). Tergum 8 of female often prolonged above proctiger, sternum 7 more or less tubular (Appendix 3F, 3G, 3H), usually covering bases of gonocoxae from beneath; proctiger either button-shaped and protruding or suboval and deflected (Appendix 3H). Distinguishing characteristics Small species with a total length rarely exceeding 3.0mm. The pronotum of the apterous form is always longer than the head, is produced posteriorly to shield the mesonotum and median part of the metanotum. The pretarsus is simple, with claws and aroliae never altered to form a ‘swimming fan’. The missle tarsus is not prolonged and is shorter than 0.8x middle tibia; and tarsal segment 1 is shorter than segment 2.

121

Appendix 2

Taxonomic characteristics of the genus Nesidovelia

122

Defining features of the genus Nesidovelia (Cassis et al. 2017) in Australia (Hemiptera: Heteroptera).

Nesidovelia, genus. Characteristic Description Size Medium-sized to large veliid species, length 1.9 – 3.4 mm. Body shape Body elongate oval (♂) to suboval ( ♀). Body colour Body mostly brown to black, often with paler brown markings on pronotal calli, connexiva and abdominal terga, sometimes intermixed with bright blue markings on abdominal terga. Macropterous morph dark, usually with two broad stripes basally and three to four spots in distal part, whitish. Pilosity Covered with short, opporessed pilosity. Head Head Triangular, transverse, produced in front of eyes, longitudinal sulcus along midline.

Eyes Eyes relatively large, semiglobular, situated at base of the head. Long and slender, sub equal to half the length of body; antennal segment 4 usually Antennae not distinctly thicker than segment 3; first antennal segment arcuate.

Thorax Pronotum Pentagonal and distinctly longer than head, with strongly projecting triangular posterior lobe in macropterous morphs, covering anterior segments of abdominal sterna; subhemispherical in apterous morphs, covering mesonotum and median part of metanotum, all abdominal terga exposed.

Legs Leg colour is mostly yellowish-brown, sometimes with dark brown highlighting of femora and tibiae; tarsi dark brown. The male foretibial grasping comb present, 0.4–0.75× length of segment (Appendix 3B); mesotibiae with a row of sparse, thin, elongate setae on ventral surface, at most about 2× width of segment.

Wings Fore wings dark, usually with two broad stripes in basal parts and 3-4 spots in distal part, whitish. Abdomen Abdomen Relatively long, sides usually evenly rounded. Female connexiva usually vertically raised and converging posteriorly (Appendix Laterotergites 3F). Venter of males always modified; segment VIII with one or two plate-like processes Venter arising from its ventral, posterior margin; segment 7 often with denticle patches, sometimes raised on one or two tubercules; female connexiva usually vertically raised and converging posteriorly, sometimes touching above midline; proctiger suboval and weakly to sharply deflected. Genital segment Male genitalia: parameres symmetrical, small, cone-shaped; proctiger widened basally, with lateral projections, usually pointed; ductus seminis at least 1.5× length of phallotheca, often strongly looped within aedeagus; aedeagal sclerites present, usually two to three sclerites; aedeagal lobal sclerite sometimes present, when present mostly sickle shaped; secondary gonopore apical.

Female genitalia: spermatheca very long and tightly coiled, mostly between five and six coils.

Distinguishing The presence of lateral processes on the male proctiger. characteristics The presence of a ventromedial process on male abdominal segment 8. The concealment of the female genitalia within the terminal abdominal segments. In females, the raised and converging connexiva.

123

Appendix 3

Schematic diagram of the body structures of Veliids

124

Illustration of body structures in Nesidovelia queenslandiae. A) Dorsal view of apterous ♂, antenna and legs of right side omitted; B) Left foreleg of apterous ♂. C) Ventral view of abdominal end of apterous ♂. D) Lateral view of genital segments of apterous ♂. E) Dorsal view of proctiger. F) Dorsal structure of apterous ♀, antenna and legs omitted. G) Lateral view of abdominal end of apterous ♀. H) Caudal view of abdominal end of apterous ♀. I) Dorsal view of macropterous ♂, antenna and legs omitted. pn, pronotum. Cn = connexivum; fw = forewing; gr = grasping comb of fore tibia; mt = metanotum; pn = pronotum; pr = proctiger; py = pygophore; s7 = abdominal segment 7; s8 = abdominal segment 8; t1 = abdominal tergite 1; t7 = abdominal tergite 7. tu = median tubercle on sternum 7; vp = ventral process on segment 8. From Andersen and Weir (2003).

125

Appendix 4

Taxonomic characteristics of the subgenera, Microvelia (Pacificovelia)

126

Defining features of the subgenus, Microvelia (Pacificovelia). Content from Andersen and Weir (2003).

Microvelia (Pacificovelia), subgen. Characteristic Description Size Very small to medium-sized veliidspecies, length 1.0-2.5 mm. Body shape Body elongate oval (♂) to suboval (♀).

Body colour Most of body covered with short, oppressed pilosity; sides of thorax and abdomen with longer, semierect hairs (♀). Head Antennae Antennae slender, 0.4-0.55x total length of body; antennal segment 4 distinctly thicker than segment 3.

Thorax Pronotum Apterous form: Pronotum subrectangular, with a transverse brownish yellow band in anterior part; its posterior margin usually bordered with brownish yellow or yellow. Macroapterous form: Pronotum form pentagonal in outline. Legs Male fore tibia comb less than ¼ tibial length; grasping comb not present on middle tibia. Wings Fore wings usually dark with broad stripes in basal parts and 3-4 spots or stripes in distal parts. Abdomen Laterotergites Female connexiva vertically raised, usually converging and sometimes meeting above abdominal end, with or without marginal tufts of hairs.

Abdominal venter simple (♂).

Venter Male: Genital segment (♂) relatively large, almost symmetrical; parameres large, slight asymmetrically developed (right paramere Genital segment largest); proctiger elongate oval, with a lateral process on each side. Female: proctiger deflected to cover gonocoxae.

127

Appendix 5

Taxonomic characteristics of the three study species: N. peramoena, N. herberti & M. oceanica

128

Defining features of Nesidovelia peramoena, N. herberti and Microvelia oceanica. Content from Andersen and Weir 2003; Cassis et al. 2017.

Nesidovelia peramoena Nesidovelia herberti Microvelia (Pacificovelia) oceanica Characteristic Description Description Description Size Apterous (♂) length 2.2-2.75, width 0.88-1.00; Apterous (♂) length 2.1-2.25, width 0.8-0.85; Apterous (♂) length 1.45-1.7, width 0.55-0.65; macropterous (♂), length 2.5-2.6, width 1.05- macropterous (♂) length 2.35-2.4 width 1.0; macropterous (♂) length 2.1-2.2, width 0.85-0.9; 1.15, apterous (♀) length 2.4-3.0, width 1.0-1.1; apterous (♀) length 2.4-2.5, width 0.9-1.0; apterous (♀) length 1.55-2.2, width 0.65-0.9, macropterous (♀), length 2.65-2.7, width 1.15- macropterous (♀) length 2.55-2.7, width 1.05-1.1. macropterous (♀) length 2.15-2.5, width 0.9-1.0. 1.2. Body shape Apterous ♂: Body elongate oval, total length 2.7x Apterous ♂: Total length about 2.6x greatest Apterous ♂: body elongate oval, total length greatest width across thorax. Most of body width across thorax. Macropterous ♂: same as about 2.7x greatest width across thorax. Most of covered with pilosity of short, semierect hairs. apterous (♂).Apterous ♀: Total length about 2.7x body covered with short, oppressed or semierect Macropterous ♂: same as apterous (♂). Apterous greatest wdith across thorax. Macropterous ♀: pilosity. Macropterous ♂: same as apterous (♂). ♀: Body elongate oval, total length above 2.5x same as apterous (♀). Apterous ♀: Body suboval, total length about greatest width across thorax (2.36: 0.98). Most of 2.45x greatest width across thorax. Most of body body covered with pilosity of short, semierect covered with short, oppressed pilosity; sides of hairs. Macropterous ♀: same as apterous (♀). thorax and abdomen with longer, semierect hairs.. Macropterous ♀: same as apterous (♀).

Body colour Dark brown above with lateral patches of silvery Dark brown above with lateral patched of silvery Dark brownish to blackish above with small lateral pubescence on anterior pronotum and abdominal pubescence on anterior pronotum and abdominal patches of silvery pubescence on head, and terga. Pronotum with two rectangular, usually terga. Pronotum with a transverse band in anterior pronotum. Pronotum with a transverse, distinctly separated spots in anterior third, anterior third, yellowish brown; this band may be brownish yellow band in anterior part; entire yellowish beown. Abdominal dorsum dark brown, dissolved into two separate spots in some posterior margin of pronotum usually yellowish connexival margins yellowish brown (♀); lateral specimens. Abdominal dorsum dark brown, brown or orange; pronotum of macropterous parts of abdominal terga 2 and 3 as well as terga connexival margins yellowish brown; lateral parts form with darker median line. Abdominal dorsum 6-7 (♀) with greyish or blueish tinge. Antennae of abdominal terga 2 and 3 as well as terga 6-7 (♀) dark with greyish frosting; middle of terga 5-7 and legs chiefly brownish yellow, ventral and with greyish of bluish tinge. Antennae and legs blackish, shiny; abdominal terga of apterous distal parts of femora and tibia darker. Fore wings chiefly brownish yellow, ventral and distal parts of female with large yellowish or orange spots, at dark brown with one or two stripes in basal parts femora and tibia darker. Fore wings dark brown least on connexiva. Antennae brownish. Legs and 3-5 spots or stripes in distal parts, whitish. with one or two stripes in basal part and 3-5 spots chiefly brownish, coxae, trochanters, and basal or stripes in distal part, whitish. parts of femora, yellowish. Fore wings dark brownish to brown, with broad stripes in basal parts and 3-4 elongate spots in distal parts, whitish; pale markings often confluent, obscured or absent.

129

Appendix 5 continue: Nesidovelia peramoena Nesidovelia herberti Microvelia (Pacificovelia) oceanica Characteristic Description Description Description Head Head Apterous ♂: Head width about 1.3x median length Apterous ♂: Head width about 1.4x median length Apterous ♂: Head width 1.6x head length (0.55:0.43). (0.56:0.40). (0.46:0.29). Macropterous ♂: As seen in apterous (♂). Macropterous ♂: As seen in apterous (♂). Macropterous ♂: As seen in apterous (♂). Apterous ♀: Head width across eyes about 1.45x Apterous ♀: Head width across eyes about 1.35x Apterous ♀: Head width across eyes about 1.5x

head length (0.60:0.41). head length (length 0.58:0.43). head length (0.50: 0.33). Macropterous ♀: As seen in apterous (♀). Macropterous ♀: As seen in apterous (♀). Macropterous ♀: As seen in macropterous (♂) and apterous (♀).

Antennae Apterous ♂: Antennae about half total body Apterous ♂: Lengths of antennal segments 1-4: Apterous ♂: Antennae relatively short, about 0.4x length of insect; lengths of antennal segments 1- 0.30, 0.25, 0.31, 0.39. total length of insect; lengths of antennal 4: 0.26, 0.24, 0.28, 0.38. Macropterous ♂: As seen in apterous (♂). segments 1-4: 0.15, 0.13, 0.15 and 0.29; antennal Macropterous ♂: As seen in apterous (♂). Apterous ♀: Lengths of antennal segments 1-4: segment 4 distinctly thicker than segment 3. Apterous ♀: Antennae about half total length of 0.26, 0.24, 0.29, 0.41. Macropterous ♂: As seen in apterous (♂). insect; lengths of antennal segments 1-4: 0.29, Macropterous ♀: As seen in apterous (♀). Apterous ♀: Antennae about 0.4x total length of 0.24, 0.29, 0.41. insect; lengths of antennal segments 1-4: 0.17, Macropterous ♀: As seen in apterous (♀). 0.13, 0.16, 0.31. Macropterous ♀: As seen in macropterous (♂) and apterous (♀).

Thorax Pronotum Apterous ♂: Pronotum subrectangular, subequal Apterous ♂: Pronotum length about 1.1x head Apterous ♂: Pronotum subrectangular, about 1.5x in length to head length (0.41). length (0.45). head length (0.33); anterior margin slightly Macropterous ♂: Pronotum pentagonal, shorter Macropterous ♂: Pronotum shorter than wide concave, posterior margin straight in middle, than wide across humeral angles (0.94: 1.15). across humeral angles (0.83:0.98). exposing narrow strips of mesonotum laterally. Apterous ♀: Pronotum about 1.2x head length Apterous ♀: Pronotum subequal in length to head Macropterous ♂: Pronotum pentagonal in outline, (0.50). length (0.45). shorter than wide across humeral angles Macropterous ♀: Pronotum as in macropterous Macropterous ♀: Pronotum as in macropterous (0.58:0.70). (♂). (♂). Apterous ♀: Pronotum subequal in length to head (0.33). Macropterous ♀: Pronotum as in macropterous (♂).

Legs Apterous ♂: Measurements of leg segments Apterous ♂: Measurements of leg segments Apterous ♂: Measurements of leg segments (femur, tibia, 1st tarsal, 2nd tarsal): forleg: 0.53, (femus, tibia, 1st tarsal, 2nd tarsal): fore leg 0.58, (femus, tibia, 1st tarsal, 2nd tarsal): for leg: 0.40, 0.46, 0.26; middle leg: 0.68, 0.59, 0.15, 0.23; hind 0.48, 0.26; middle leg: 0.74, 0.69, 0.18, 0.25; hind 0.29, 0.19; middle leg: 0.45, 0.36, 0.10, 0.15: hind leg 0.73, 0.84, 0.19,0.23; grasping comb of fore… leg: 0.78, 0.88, 0.18, 0.26; grasping comb of… leg 0.51, 0.58, 0.11, 0.15. Fore femur slightly…

130

Appendix 5 continue: Nesidovelia peramoena Nesidovelia herberti Microvelia (Pacificovelia) oceanica Characteristic Description Description Description Legs …tibia (0.19) about 0.4x fore tibial length …foretibia (0.25) about 0.5x fore tibial length. …thickened in middle; grasping comb of fore tibia (Lundbald 1933). Macropterous ♂: As seen in apterous (♂). short, about 0.2x tibial length (0.06); hind femur Macropterous ♂: As seen in apterous (♂). Apterous ♀: Measurements of leg segments as thick as middle femus. Apterous ♀: Measurements of leg segments (femus, tibia, 1st tarsal, 2nd tarsal): for leg: 0.59, Macropterous ♂: As seen in apterous (♂). (femur, tibia, 1st tarsal, 2nd tarsal): foreleg: 0.58, 0.44, 0.28: middle leg: 0.73, 0.65, 0.20, 0.25; hind Apterous ♀: Measurements of leg segments 0.49, 0.26; middle leg: 0.75, 0.69, 0.19, 0.25: hind leg: 0.81, 0.93, 0.21, 0.28. (femus=r, tibia, 1st tarsal, 2nd tarsal): foreleg: 0.44, leg: 0.88, 0.96, 0.21, 0.28. Macropterous ♀: As seen in apterous (♀). 0.30, 0.21: middle leg: 0.50, 0.43, 0.13, 0.18; hind Macropterous ♀: As seen in apterous (♀). leg: 0.56, 0.64, 0.14, 0.16. Macropterous ♀: As seen in macropterous (♂) and apterous (♀).

Wings Macropterous ♂: Fore wings dark brown with one Macropterous ♂: Fore wings dark brown with one Macropterous ♂: Fore wings with long, semierect or two stripes in basal parts and 3-5 spots or or two stripes in basal part and 3-5 spots or hairs along their anterior margins. stripes in distal parts, whitish. stripes in distal part, whitish. Macropterous ♀: As seen in macropterous (♂). Macropterous ♀: As seen in macropterous (♂). Macropterous ♀: As seen in macropterous (♂).

Abdomen Abdomen Apterous ♀: Abdominal dorsum in lateral view Apterous ♀: Abdominal dorsum in lateral view Apterous ♀: Abdominal dorsum in lateral view curved in anterior part, almost straight in almost straight throughout. curved downwards towards abdominal end. posterior part. Macropterous ♀: As seen in apterous (♀). Macropterous ♀: As seen in macropterous (♂) Macropterous ♀: As seen in apterous (♀). and apterous (♀).

Apterous ♀: Connexiva vertically raised Laterotergites Apterous ♀: Connexiva almost vertically raised Apterous ♀: Connexiva almost vertically rasied, throughout, converging posteriorly, but rarely anteriorly, converging upon and sometimes converging upon and reaching each other above meeting each other at abdominal end; connexival reaching each other towards abdominal end; sternum 7; connexival margins mot modified margins not modified; posterior corners of connexival margins not modified except for except occasional tufts of black hairs at the laterotergites 7 rounded, with a few long bristles. elongate, shiny pads (‘wart-like structures’ of intersegmental limits between laterotergites 5-6 Macropterous ♀: Connexiva only slightly Malipatil 1980) on laterotergites 7; posterior and 6-7; posterior corners of laterotergies 7 converging posteriorly. corners angular, with or without long, black angular, with conspicuous tufts of long, black bristles. bristles.

Macropterous ♀: Connexiva only slightly Macropterous ♀: Connexiva only slightly converging towards abdominal end, posterior converging towards abdominal end, posterior corners without black bristles. corners without tufts of black bristles.

131

Appendix 5 continue: Nesidovelia peramoena Nesidovelia herberti Microvelia (Pacificovelia) oceanica Characteristic Description Description Description Venter Apterous ♂: Abdominal venter slightly modified. Apterous ♂: Abdominal venter slightly modified; Apterous ♂: Abdominal venter simple; sternum 7 Sterna 4-7 with faint, median impression, sternum sterna 4-7 with faint median impression; sternum shorter than preceeding two sterna together 7 slightly shorter than preceeding two sterna 7 slightly shorter than preceeding two sterna (0.19: 0.21), posterior margin broadly concave. together (0.23-0.29), with anterior, transverse pile together (0.26:0.30), with anterior, transverse pile Macropterous ♂: As seen in apterous (♂). of brownish hairs; posterior margin slightly raised. of brownish hairs, posterior margin simple. Apterous ♀: Sternum 7 subequal in length to the Macropterous ♂: As seen in apterous (♂). Macropterous ♂: As seen in apterous (♂). preceding two sterna together (0.31:0.35), Apterous ♀: Sternum 7 subequal in length to Apterous ♀: Sternum 7 a little longer than the posterior margin straight. preceeding two sterna together (0.44: 0.39); preceeding two sterna together (0.46:0.43); Macropterous ♀: As seen in macropterous (♂) posterior margin slightly produced, with a small posterior margin almost straight, with a small and apterous (♀). process on each side. process on each side. Macropterous ♀: As seen in apterous (♀). Macropterous ♀: As seen in apterous (♀).

Genital Apterous ♂: Genital segments relatively large, Apterous ♂: Genital segments relatively large; Apterous ♂: Genital segments large, about as long segment segment 8 with a pair of triangular processes segment 8 with a pair of processes close to its as wide (0.26); segment 8 simple; pygophore close to its ventral, posterior margin, partly ventral, posterior margin, each process slender suboval; proctiger elongate oval with a pair of obscured by brushes of pale hairs; parameres and slightly curved: parameres small, cone- slender lateral processes in middle; parameres small, cone-shaped (Lundbald 1933; Malipatil shaped: proctiger basally widened, with a short almost symmetrical, right paramere large, 1980); proctiger basally widened, with a short pointed process on each side. falciform, with broad basal part of pointed apex. pointed process on each side Macropterous ♂: As seen in apterous (♂). Macropterous ♂: As seen in apterous (♂). Macropterous ♂: As seen in apterous (♂). Apterous ♀: Genital segments partly concealed; proctiger relatively narrow, deflected. Macropterous ♀: As seen in macropterous (♂) and apterous (♀).

132

Appendix 6

The variations in morphology deemed as antagonistic trait

133

Appendix 6.1) Variations in connexival convergence, dorsal view. Scale bar 0.50 mm. Black box highlights area of modification in A-C: A) M. oceanica apterous male connexiva: widely separated and not strongly raised nor converging. B) M. oceanica apterous female: converging partially but not meeting at midline. C) N. herberti apterous female: converging and meeting at midline.

Appendix 6.2: Proctiger orientation between female macropterous and apterous morphs. Scale bar 0.50mm. Black box highlights area of modification in A-B: A) M. oceanica macropterous female proctiger deflected obscuring the gonocoxae. B) M. oceanica apterous female proctiger is deflected obscuring the gonocoxae.

134

Appendix 6.3) Modifications of the caudal setal tufts on female abdominal laterotergite 7, lateral view. Scale bar 0.50mm. Black box highlights area of modification in A-B: A) M. oceanica apterous female, tufts absent. B) N. herberti apterous female, tufts present.

Appendix 6.4) Modifications of the male abdominal sternite 8 lateral view. Scale bar 0.50mm. Black box highlights area of modification in A-B. A) M. oceanica apterous male, no segment 8ventral process. B) N. peramoena apterous male, segment 8 lateral process present.

135