Thesis Template

Home , Kochiura, Leviellus, Mecynogea, Neriene litigiosa, Paidiscura, Parasteatoda

Mate searching and choosiness are shaped by spatial structure and social information in western black widows

Catherine Elizabeth Scott

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Ecology and Evolutionary Biology University of Toronto

Mate searching and choosiness are shaped by spatial structure and social information in western black widows

Catherine Scott

Doctor of Philosophy

Ecology and Evolutionary Biology University of Toronto

2020 Abstract

The relative importance of different episodes of sexual selection for fitness varies with environmental and social context, so inferring the net effect of selection requires understanding how each mechanism operates in nature. Theory predicts that links between ecology, information availability, and mechanisms of sexual selection can have substantive effects on plasticity and behaviour, but these links are rarely demonstrated in the field. Using the western black widow,

Latrodectus hesperus, I investigated how ecological factors including demography, OSR, and spatial structure influence male and female reproductive behaviour in the field, and how these are linked by chemical information. I found that the OSR is extremely male biased for most of the mating season, and this imposes strong selection on males to find females before rivals. Mate searching males use social information, likely encoded by chemicals on silk draglines produced by rivals, to efficiently locate females even in the absence of female sex pheromone. In the face of intense scramble competition over access to receptive adult females, I found that males commonly guard and mate with subadult females using a recently-described alternative reproductive tactic: immature mating. Success at this tactic requires contests with rivals, which likely favours different traits than does pure scramble competition over adults, and may help to maintain extreme variation in male size. Moreover, because subadults apparently do not produce

ii volatile chemical cues that would allow males to locate them, the clumped spatial distribution of females in the field is a critical factor influencing the feasibility of immature mating; males only found subadults when there were signaling adults nearby. I also found that for females, proximity to conspecifics during development and early adulthood influences encounter rates with potential mates, such that some females risk remaining unmated in nature. Consequently, females adjust mate-choice decisions in response to chemical information about their local social environment; females in relatively isolated locations display decreased choosiness relative to females in close proximity to conspecifics. Together these studies provide insights into how mate competition and mate choice, key factors driving sexual selection, are related to the spatiotemporal distribution of conspecifics in a natural population.

iii

Acknowledgements

I sincerely thank the Tsawout First Nation for allowing me to do fieldwork on their beautiful lands. I have been so privileged to work with the black widows on the beach and T̸IX̱ EṈ since 2010 and am very grateful to the former lands managers and current band manager Eva Wilson for generously facilitating this work.

I also wish to acknowledge the land on which the University of Toronto operates. For thousands of years it has been the traditional land of the Huron-Wendat, the Seneca, and most recently, the Mississaugas of the Credit River. Today, it is still the home to many Indigenous people from across Turtle Island and I am grateful to have had the opportunity to work here.

A great many people have contributed to my research, supported me personally and professionally, and helped me in a wide variety of meaningful ways during my PhD. I want to express my deepest thanks to the following groups and individuals:

To Team Black Widow:

Catherine & Doug Antone, Joe Lapp, Robb Bennett, Roy Dunn, Sean Lambert, Betty Kipp, Dora Sardas, Kristen Cain, Christy Peterson, Christy Peterson, Raphael Royaute, Dawn Bazely, Woodrow Setzer, Pierre Robillard, John Barthelme, Nemo de Jong, Mike Boers & Tanya Stemberger, Sina Rastegar, Sarah Langer, Sidnee & John Scott, Stephen & Linda Lambert, Staffan Lindgren, Amanda Yee, Rob Higgins, Tonia Harris, Tanya Jones, Joe O’Franklin, Dezene Huber, Tracey Birch, Peggy Muddles, Regine & Gerhard Gries, Gwylim Blackburn & Samantha Vibert, Alex & Karla Antone, Gil Wizen, Gwen Pearson, Joan Andrade, Kate Compton, Peggy McCann, Peter Andrade, Rick Redus, Robyn Raban, Shelley Barkley, Stewart, Geoff Bennett, Kyle Cassidy, Colin & Heather McCann, Jonathan Meiburg, Lori Weidenhammer, Diana Davis, Ray Scanlon, Ashley Bradford, Ed Morris, Robert Cruickshank, Marc Rashinski, James Petruzzi, Joseph Peter McNamara, Ariel Ng, Robert Neylon, Auriel Fournier, Victoria Nations, Leah Ramsay, Tom Pearce, Chloe Gerak, Scott Severs, Angie Macias, Nick Spencer, Thomas Astle, Luna Nicolas Bradford Ley, Peter Midford, Laurel Ramseyer, Morgan Vis, Tom Pardue, Scott Schrage, Kelly Brenner, Karen Yukich, Charmaine Condy, Amy Parachnowitsch, Catherine Scott, Christine Rock, Jason Parker-Burlingham, Jonathan Kade, Joseph Peter

McNamara, Joshua Erikson, Juniper English, Nick Spencer, Robert Cruickshank, Sabrina Caine, Suran TheStorm, Richard Dashnau, Stephen Heard, Holly Fraser, Lynne Kelly, Roberta Chan, Kat Cruickshank, Meera Lee Sethi, Mike Hrabar, Tiffany Jacobs, Connie Larochelle, Willow English, David Steen, Michelle Reeve, Tone Killick, David Esopi, Antonia Guidotti, Elaine Wong, Lisa Wrede, Naomi Gonzales, Don Campbell, Matt Masterson, Paul Manning, Casey Peter, Dave Rich, Jessica Olin, Kate Rey, Katie Russell, Shari McDowell, Suzanne Spinelli, Christina Tran, Cindy Wu, Aaron Soley, Chris Garbutt, Greg Randolph, Lila Robinwood, Eric Damon Walters, The Spider Chick, and Steve Waycott. Thank you all so much for your generous contributions to our crowdfunding campaign, which made the 2017 field season possible and was critical to the success of this work.

To Maydianne Andrade, for being an incredibly supportive supervisor, wonderful teacher, and generous mentor.

To Darryl Gwynne and Andrew Mason, for providing valuable discussion and feedback on my work as members of my supervisory committee.

To Eileen Hebets, Joel Levine, and John Ratcliffe, for helpful comments and stimulating discussion as members of my PhD examining committee.

To Sean McCann, for being my partner and collaborator in all things. Among other things, thank you for spending many long nights with me watching spiders on the beach, and many long days driving back and forth across the continent, moving to (and from) Toronto and collecting black widows from BC to Texas.

To Mike Boers and Tanya Stemberger (and Jasmine the cat), for sharing their home with me for the final two years of my PhD, and for being my family in Toronto. I am so grateful for your friendship, for your help with data wrangling and other thesis-related issues large and small, and for all the ways you have supported me over the last several years.

To Claudia and Darren Copley (and Darwin and Wallace the cats), for sharing their home with us and our spiders, and providing logistical, moral, and libationary support during the 2017 field season.

To Peggy McCann, for sharing her home with us and our spiders during the 2016 field season. Thank you Peggy for being so supportive of me and Sean as we pursue our dreams, and for cheerfully spreading the word that spiders are our friends.

To John and Sidnee Scott, for loaning us a car for the 2017 field season, and countless other contributions that have supported my research directly and indirectly. Thank you Mum and Dad for everything, including your enthusiastic support of my single-minded passion for spiders.

To Sina Rastegar, for helping me to purchase a new laptop, without which I would not have been able to finish this thesis. Thank you for being such a kind and generous friend.

To Samantha Vibert and Gwylim Blackburn, for friendship, mentorship, and hospitality ever since I first started studying spiders. Thank you Sam and Gwylim especially for providing a restful and much-needed holiday destination each reading week during my PhD.

To Peggy Muddles and Gil Wizen, for friendship, moral support, spidering, delicious meals, fun conversations, cheesecake, and ice cream. Thank you for being such wonderful friends.

To the students in BIOD53 in winter 2017:

Joshua Carrière, Mariaelena Guarrasi, Jason Hoac, Zohal Kerami, Vicky Nguyen, Brittany Robinson, Roxanne Santos, Michael Swift, Thi Truong and Nicole Wong, for their enthusiastic contribution to data collection for Chapter 1. It was such a pleasure to work with you and to support your experience doing animal behaviour research in Sean’s class.

To the Andrade Lab:

Luciana Baruffaldi, Charmaine Condy, Sheena Fry, Monica Mowery, and Nishant Singh, for friendship, constructive feedback, and providing a truly supportive and positive lab community. Thank you for everything. I could not have asked for better labmates.

Ariella Kong, Archchana Rajmohan, Ramanja Pakirathan, and Yousef Safar, for their masterful work keeping things running efficiently as lab managers, and to all of the undergraduate lab assistants for their hard work and dedication to rearing spiders and keeping the lab clean and functioning. Your efforts are critical, and much appreciated. vi

To Sumaya Dano, Ajay David, Nimra Javaid, Dilakshan Srikanthan, and Amanda Yee, for their enthusiasm and dedication to their independent black widow research project and their assistance with spiders for my research. It was such a pleasure to advise you and to learn from you.

To Dan Peach, for friendship, support, and perpetual kindness and optimism.

To Ron Jasiuk and Ann Moum, for welcoming us into their home and enthusiastically encouraging us to search their property for black widows.

To Robb and Jenny Bennett, for their friendship and generous hospitality both in Victoria and at their cottage.

To Alex & Karla Antone, Bekka Brodie & Viorel Popescu, Auriel Fournier, Joe Lapp, Sarah Loboda, Chris MacQuarrie & Amanda Roe, Terry McGlynn, Antonia Musso & Andrew Cook, and Christy Pitto, for providing us with friendship and warm hospitality during trans-continental road trips for spider collecting and cross-country moves.

To Jay Cullen at UVic, for kindly allowing us to use his microbalance to weigh our spiders.

To Spider Twitter, Entomology Twitter, and Science Twitter in general, for all kinds of support, both personal and professional. I have met so many wonderful friends, colleagues, and mentors on twitter and I’m grateful for all of the discussions we’ve had and the things I’ve learned from you. Thank you for being such a welcoming and generous community.

I am also extremely grateful to the Entomological Society of Canada, the Toronto Entomologists’ Association, the American Arachnological Society, the International Society of Arachnology, and NSERC, for grants and awards that helped to support me during my PhD.

vii

Table of Contents

Acknowledgements ...... iv

Table of Contents ...... viii

List of Tables ...... xii

List of Figures ...... xiv

General Introduction ...... 1

Male black widows parasitize mate-searching effort of rivals to find females faster ...... 14

Abstract ...... 14

1.1 Introduction ...... 14

1.2 Materials and Methods ...... 18

1.2.1 Mating system and the intensity of sexual selection ...... 18

1.2.2 Field experiments on male mate searching under varying information availability...... 18

1.2.3 Laboratory tests of context-dependent male information use ...... 19

1.2.4 Statistical analyses ...... 20

1.3 Results ...... 21

1.3.1 Operational sex ratio, mate-searching success, and mortality in nature ...... 21

1.3.2 Mate-searching success, speed, and social information use in the field ...... 22

1.3.3 Context-dependent male responses to social information in the laboratory ...... 23

1.4 Discussion ...... 23

1.5 References ...... 27

Spatial clustering of females facilitates immature mating, an extreme reproductive tactic of male black widow spiders ...... 36

Abstract ...... 36

2.1 Introduction ...... 36

2.2 Methods and Results ...... 40 viii

2.2.1 Natural History...... 40

2.2.2 Observational Data: Methods ...... 41

2.2.3 Observational Data: Results ...... 45

2.2.4 Experimental Data: Methods ...... 47

2.2.5 Experimental Data: Results...... 49

2.3 Discussion ...... 49

2.4 References ...... 55

Widows as plastic wallflowers: female choosiness shifts with indicators of mate availability in a natural population of black widows ...... 69

Abstract ...... 69

3.1 Introduction ...... 69

3.2 Methods...... 72

3.2.1 Natural History and Field Site ...... 72

3.2.2 Field Experiment ...... 73

3.2.3 Mating Trials ...... 74

3.2.4 Statistical Analyses ...... 74

3.3 Results ...... 75

3.3.1 Field Experiment ...... 75

3.3.2 Mating Trials ...... 76

3.4 Discussion ...... 76

3.5 References ...... 79

A review of the mechanisms and functional roles of male silk use in spider courtship and mating ...... 90

Abstract ...... 90

4.1 Introduction ...... 90

4.1.1 Overview ...... 90

4.1.2 Properties of spider silk ...... 92 ix

4.1.3 Female silk and mating ...... 93

4.2 Male silk and mating ...... 95

4.2.1 Overview ...... 95

4.2.2 Fitness effects of silk use ...... 96

4.2.3 Mechanisms of effect ...... 97

4.3 Silk deposition on females’ webs or other silk structures...... 100

4.3.1 Overview and description of behaviors ...... 100

4.3.2 Proposed mechanisms and functions ...... 104

4.3.3 Current evidence and future directions ...... 106

4.4 Silk deposition on females ...... 108

4.4.1 Overview and description of behaviors ...... 108

4.4.2 Proposed mechanisms and functions ...... 109

4.4.3 Current evidence and future directions ...... 111

4.5 Silk associated with nuptial gifts ...... 113

4.5.1 Overview and descriptions of behaviors ...... 113

4.5.2 Proposed mechanisms and functions ...... 115

4.5.3 Current evidence and future directions ...... 117

4.6 Other examples of male silk use during mating interactions ...... 121

4.7 Conclusions and future directions ...... 122

4.7.1 Summary ...... 122

4.7.2 Functions and mechanisms of effect ...... 123

4.7.3 Improving our understanding of male silk use ...... 124

4.7.4 Concluding remarks ...... 127

4.8 References ...... 128

Conclusion ...... 170 x

Appendix 1 ...... 175

Appendix 2 ...... 190

List of Tables

Table 2-1 The observed percentages of Latrodectus hesperus males visiting only subadult females compared to the percentage expected based on the maximum relative abundance of subadult females during the early (n = 53 males observed) and late (n = 271 males observed) mating seasons...... 60

Table 3-1 Summary of the number of males arriving at cages of L. hesperus females before and after they moulted to maturity, during a 3-month field experiment...... 85

Table 3-2 Results of two logistic regression models assessing whether treatment (clustered or isolated), female age (in days since the moult to maturity), and the number of males who arrived at their web over the course of the field experiment affected two measures of L. hesperus female choosiness (copulation and pre-copulatory cannibalism) in mating trials...... 86

Table 4-1 Cross tabulation of potential functions (columns) and mechanisms (rows) of silk use by male spiders in courtship and mating...... 145

Table 4-2 Spider taxa in which males modify the female’s web or other silken structures by adding and/or removing silk (web reduction) ...... 147

Table 4-3 Spider taxa in which males deposit silk ‘bridal veils’ onto the female during courtship...... 155

Table 4-4 Spider taxa in which males present females with silk-associated nuptial gifts, including silk-wrapped prey, silk alone, or silk-lined burrows...... 159

Table 4-5 Other behaviors involving male silk deposition during courtship and mating...... 161

Table 4-6 Spider taxa in which there is behavioral evidence for male-produced sex pheromones. These families are also indicated in red in fig. 1 ...... 162

xii

Table 4-7 Spider taxa in which there is behavioral evidence for males responding to silk cues of conspecific males...... 164

xiii

List of Figures

Figure 1-1 Demographic data for a natural population of L. hesperus at a coastal field site on Vancouver Island, British Columbia, Canada (48°3501000 N, 123°2201700 W)...... 31

Figure 1-2 Patterns of arrival at female pheromone traps under differing wind conditions in field tests of mate searching by L. hesperus males...... 32

Figure 1-4 Distance from females and the size of L. hesperus males affect mate-searching success and speed in the field...... 34

Figure 2-2 Interactions between Latrodectus hesperus individuals on the webs of subadult females...... 63

Figure 2-4 Experimental design for testing male localization of females with different spatial distributions in the field...... 65

Figure 2-5 Minimum number of females of different age classes found by L. hesperus males in their lifetimes...... 66

Figure 2-6 Number of males found on webs of L. hesperus females who were followed from their penultimate instar through to maturity...... 67

xiv

Figure 3-2 Males arrived earlier and more males arrived in rapid succession at cages of clustered females compared to isolated females during our field experiment...... 88

Figure 3-3 Following the field experiment, females from the clustered treatment rejected males more often than isolated females in staged mating trials...... 89

Figure 4-1 Cladograms illustrating relationships between araneomorph spider families (based on Wheeler et al. 2016) and the occurrence of male silk and pheromone use...... 165

Figure 4-2 Examples of silk deposition onto females’ webs during courtship...... 166

Figure 4-3 Examples of silk ‘bridal veils’ applied to females’ legs and bodies during courtship...... 167

Figure 4-4 Examples of silk-wrapped nuptial gifts...... 168

Figure S1. Sample photograph of a marked male spider with a size reference used to measure the tibia-patella (t-p) lengths of the first legs...... 185

Figure S2. Distances moved by wild mate-searching males in the field. Histogram of the minimum distances moved by Latrodectus hesperus males who survived travel between their home web and that of a female and/or between webs of females between April and September 2016...... 186

Figure S3. Patterns of attraction to pheromone traps for experimentally released males in the field...... 187

Figure A2-1. Relationships between male morphology and speed during mate-searching experiments in the field...... 190 xv

Figure A2-2. Distributions of male leg length for field-collected and lab-reared males...... 191

xvi

General Introduction

Darwin (1871) developed his theory of sexual selection to explain the evolution of conspicuous male traits that seemed likely to decrease survival, and thus presented a problem for his more general theory of natural selection. Sexual selection is a type of natural selection that arises from variation in mating success caused by competition over mates and fertilizations (Andersson 1994). Darwin’s examples focused primarily on how contest competition favours male weaponry and how female choice favours elaborate male ornaments. We now recognize that the mechanisms of sexual selection acting prior to mating include contest competition, scramble competition, endurance rivalry, and mate choice (Andersson 1994). Additionally, analogues of contests and mate choice occur after copulation in the form of sperm competition (Parker 1970) and cryptic female choice (Lloyd 1979; Eberhard 1996). Andersson (1994) noted a dearth of research on scramble competition (in which competition among males manifests as a race to find females) relative to contest competition and mate choice, and this gap seems to persist today even though scramble competition mating systems are widespread (Herberstein et al. 2017). This lack of attention is likely because scramble competition is less obvious or more difficult to study, rather than being less important than contests and courtship displays (Andersson 1994). This is problematic because scramble competition is likely to be common not only in terrestrial arthropods (Thornhill & Alcock 1973; Herberstein et al. 2017), but also marine arthropods (Brockman 1990) and vertebrates including amphibians (Able 1999; Wells 2007); reptiles (Duvall et al. 1993); fishes (Taborsky 2001; Rios-Cardenas & Morris 2009) and mammals (e.g., Schwagmeyer & Woontner 1986; Lane et al. 2009; Marmet et al. 2012).

Mechanisms of sexual selection rarely act in isolation (Andersson 1994). Males might first scramble to find females, but subsequently engage in contests, courtship, and/or coercion in their efforts to secure fertilizations (Andersson & Iwasa 1996). Different male traits may be favoured in different contexts, and we risk making mistaken conclusions about the strength and direction of selection if we restrict study to single mechanisms of sexual selection. For instance, we may demonstrate female preferences in courtship interactions, or traits that confer a competitive advantage in combat, but fail to take into account that only those males that successfully locate a female or a breeding site will have the opportunity to engage in these

1 2 activities. In this example, mate-searching imposes a selective filter that may affect the range of trait values that are actually subject to selection via mate choice or contests. Similarly, since the relative importance of different episodes of selection varies as a function of the environmental and social context (Evans & Garcia-Gonzalez 2016), inferring the net effect of selection requires understanding how mechanisms of selection operate in the field.

Sexual selection is both a cause and consequence of mating systems, which are defined at the most basic level by how many mates are obtained by members of each sex (Kokko et al. 2014). Male fitness typically increases with each additional mating, but this is not usually the case for females (Bateman 1948). Asymmetry in parental investment underlies these differences in the fitness benefits of mating multiply and determines which sex is limiting, and thus expected to be choosy (Trivers 1972). The extent to which one sex limits the reproduction of the other sex can be quantified in the operational sex ratio (OSR), defined as in the ratio of adult males to fertilizable or receptive females (Emlen & Oring 1977). The OSR is a central idea in mating system theory as it is generally related to the strength of competition over mates (Emlen & Oring 1977, de Jong et al 2012) and the opportunity to be choosy about potential mates (Berglund 1995). The OSR is dependent not only on relative parental investment, but also on physiological and environmental factors that affect potential reproductive rates of each sex (Clutton-Brock & Parker 1992). Emlen & Oring (1977) noted that in addition to OSR, the extent to which the spatio-temporal distribution of receptive females permits male monopolization of multiple females (environmental potential for polygamy) is a key factor affecting mating systems.

Although there have been debates about how well OSR relates to the intensity of sexual selection, recent work suggests its utility in predicting competition (de Jong et al. 2012), and there has been development of theory for understanding how spatiotemporal factors relate to variance in mating success (Ims 1988; Shuster & Wade 2003). Indeed, empirical work has shown that the intensity of competition, and which sex competes, tracks variation in OSR over time and space (Kvarnemo & Ahnesjo 1996; Gwynne et al. 1998; Forsgren et al. 2004). Critically, assessing the strength of sexual selection in nature requires measuring OSR over appropriate spatial and temporal scales, and obtaining information about mating systems and processes that affect the degree of mate monopolization (Kvarnemo & Ahnesjo 1996; Klug et al. 2010). Thus, using OSR to predict how sexual selection will operate in a given population requires assessing phenology, and the timing and frequency of mating and sexual receptivity in a spatially explicit

3 framework (Ims 1998; Wade & Arnold 1980). This is particularly true in multivoltine species where these variables may shift over the season or over relatively small spatial scales (e.g. Gwynne et al. 1998). Moreover, estimates of the potential for sexual selection require measures of the proportion of the population that remains unmated (Krakauer et al. 2011).

Although well-used in studying competition dynamics among males, the OSR is less commonly discussed as a possible predictor of female choice (e.g., Hayes et al 2016, Head et al 2008). Instead, the focus is more typically on how male density, or the expected encounter rate with males (Henshaw 2018) may affect the expression of female preferences (ranked order of preferred males based on phenotypic traits) and choosiness (effort or risk expended to express a choice, Jennions & Petrie 1997; Kokko & Mappes 2005). Although encounter rates are likely to covary with density and/or OSR, spatial or temporal heterogeneity may lead to variation in mate encounter rates under a given population-level OSR (e.g., Rhainds 2019). Recent work suggests that OSR itself may not explain variation in female preferences (Hayes et al 2016). However, plasticity in female choosiness (Bailey & Zuk 2008) and preferences (Hebets 2003) has been shown in response to variation in the perceived availability, or traits of local males (West- Eberhard 2003; Pigliucci 2001; Kasumovic, Hall, & Brooks 2012). It is interesting to note that, depending on the mating system, a high density of males may lead to increased choosiness (e.g., many males from which to select), or decreased choosiness (e.g., if males harass unmated females sufficiently so that mating quickly and indiscriminately is the high-fitness option).

In changing environments, animals are at an advantage if they can acquire salient information about variables that affect the likely outcome of competition or the availability of mates, and adjust their behaviour in ways that increase reproductive success (West-Eberhard 1979; Allen et al 2008). Information may be acquired through direct detection (personal information) or indirectly by attending to the behaviour or cues produced by others (social information) (Danchin et al. 2004). The fitness benefit and feasibility of acquiring each type of information may be context-dependent (Gruter & Leadbeater 2014) and may depend on individual capacities or tendencies (Ibáñez et al. 2012; White et al 2017). Acquiring personal information (e.g., through exploration) may allow targeted decision-making, but can be costly (Galef 2009); when it is sufficiently risky, time-consuming, or unreliable, social information may be preferred (Rieucau & Giraldeau 2011; Grüter & Leadbeater 2014). Understanding contexts in which social information is used is important because it may affect evolutionary processes by

4 altering the intensity of selection. Sexual selection can favour the use of social information in assessing mates or competitors (Gibson & Höglund 1992; Valone & Templeton 2002), which alters the range of genotypes accruing high fitness via choice or competition. For example, when females copy the mate choices of others, or when they consistently express a common preference, the result is a strong reproductive advantage for relatively few preferred males (Wade & Pruett-Jones 1990). In contrast, when males race to find receptive females, social information may be used to avoid or reduce investment in females that are likely to have mated (Muñoz 2004; White 2004; Schneider et al. 2011), even if these females are otherwise preferred. This can lead to reduced variation in female mating rates and reduced inter-male competition over fertilization.

The assessment of chemical signals or cues facilitates the successful identification and localization of receptive mates across taxa (Karlson & Butenandt 1959; Svensson 1996; Johnston 2003; Belanger & Corkum 2009; Thiel & Breithaupt 2010; Thomas 2011; Andersson 1994; Johannson & Jones 2007), and can thus play a prominent role in scramble competition and choice. Odours have been recognized as examples of traits under sexual selection since Darwin (1871). Despite being the most ancient and common form of communication (Bradbury & Vehrencamp 1998; Wyatt 2003), chemical signals are less often studied than visual and acoustic signals in the context of sexual selection (Andersson 1994; Johannson & Jones 2007). As with scramble competition, this likely arises from the difficulty of detecting and measuring chemical communication rather than its relative importance. Across taxa, there is a general pattern of long- range mate attraction via female-produced sex pheromones, followed by close-range pheromones produced by males during courtship (Andersson 1994; Svensson 1996; Wyatt 2003). Chemical signaling by males can be important for mating success via both female choice (Johannson & Jones 2007) and contest competition (Gosling & McKay 1990; Wyatt 2003).

Chemical information may form a link between social context, demography and individual behavioural decisions in the context of competition and mate choice. For example, for species that produce chemical signals, there may be sustained information available regarding population structure, which varies on both spatial and temporal scales (Elias et al. 2011), and thus reflects variation in operational sex ratios. In turn, the OSR, mating system, and spatiotemporal distribution of mates will affect the intensity of competition and the potential for sexual selection (Emlen & Oring 1977; de Jong et al. 2012). At the same time, environmental

5 conditions and population structure impose constraints on the design and effectiveness of chemical signals (Alberts 1992; Endler 1992, 1993). To determine how chemical signals and cues mediate sexual selection, it is useful to consider the interplay between population structure, OSR, mating system, and signaling environment. Moreover, the interaction of these features with individual traits may determine intra-specific variation in reproductive tactics (Dominey 1984; Hutchings & Myers 1994; Gross 1996).

In this thesis I study how sexual selection is shaped by demographic factors and mediated by chemical information across environmental contexts. My work uses field and laboratory studies of the western black widow, Latrodectus hesperus (family Theridiidae), a spider with a scramble competition mating system. Theridiid spiders are useful for studying chemical signaling because they are generally solitary and females are sedentary, they are strictly cursorial, and they communicate on relatively small spatial scales (Uhl & Elias 2011; Uhl 2013). Web-building spiders are particularly tractable for studying chemical communication because their pheromones are associated with their silk (Gaskett 2007). Moreover, the widow spiders (~30 species in the genus Latrodectus), have been the focus of many sexual selection studies (reviewed in Andrade & MacLeod 2015), some of which concern the function and chemical structure of sex pheromones produced by both females and males (e.g., Ross & Smith 1979; Jerhot et al. 2010; Scott et al. 2015a; MacLeod & Andrade 2014; Baruffaldi & Andrade 2015).

Field site and study species

My field site is an area of coastal sand dunes above the high-tide line at Island View Beach on Vancouver Island, British Columbia (BC), Canada (48°35’10”N, 123°22’17”W), where driftwood logs, woody debris, and rocks provide microhabitats for western black widows (Latrodectus hesperus Chamberlin & Ivie, 1935). The population is dense, with 2-3 subadult or adult females per square metre of microhabitat in late summer (Salomon et al. 2010). Adult females are present all year, and produce egg sacs from May until July (the active season for oviposition and mating typically runs from April to October). Males and females can become sexually mature the same year they hatch or overwinter as juveniles and mature the following spring (Salomon et al. 2010).

Virgin female western black widows produce silk-bound sex pheromones that attract males at long range (Kasumovic & Andrade 2004; MacLeod & Andrade 2014) and elicit

6 courtship behaviour upon contact (Ross & Smith 1979; Scott et al. 2015a). Males can discriminate between females of different mating status and feeding history based on these volatile and contact pheromones (MacLeod & Andrade 2014; Baruffaldi & Andrade 2015). There is some behavioural evidence for male sex pheromones or chemical cues associated with silk draglines (Ross & Smith 1986), and work in a congener indicates that males produce volatile chemical cues that are detectable by other males (Kasumovic & Andrade 2006).

Once a male arrives at a female’s web he engages in an hours-long courtship display that involves vibratory signalling, silk deposition, and often web reduction (MacLeod 2013; Ross & Smith 1979; Scott et al. 2012; Vibert et al. 2014). Web reduction involves the male bundling up the female’s pheromone-emitting web and wrapping it with his own silk, which decreases female attractiveness or detectability to rivals (Watson 1986; Scott et al. 2015b). At copulation, males lose a small portion (apical sclerite) of each embolus (paired copulatory organs), and these can function as sperm plugs when correctly placed in the female’s spermathecae (paired sperm storage organs; Snow et al. 2006), leading to strong first male sperm precedence (Macleod 2013). Nevertheless, both males and females can mate multiply (Macleod 2013). Additional natural history information is presented in each chapter where it is most relevant.

Overview of thesis chapters*

* I use “I” below to describe activities undertaken as the primary author in each of these studies, but important contributions have been made by coauthors who are listed below the title of each chapter later in this thesis.

In chapter 1 I first examine the context of sexual competition in the field by documenting the phenology of a natural population of L. hesperus. I estimate the OSR as an indicator of the intensity of competition over mates, providing insight into the potential for sexual selection across the season. I then test predictions about the use of personal information (sex pheromones) or social information (chemical cues left by rivals) by mate-searching males as a function of environmental conditions and proximity to females. I estimate the distance over which males can detect information about the location of receptive females, which has implications for female mating rates and the nature and intensity of male competition. In field and laboratory experiments, I test the hypothesis that male mate searching success is mediated by a combination of factors including the spatial distribution females and competitors, environmental conditions, availability of personal and social information, and individual male morphology or condition.

This chapter has been published (Scott CE, McCann S, Andrade MC B. 2019 Proceedings of the Royal Society B: Biological Sciences 286: 20191470, doi:10.1098/rspb.2019.1470).

In chapter 2 I investigate how the spatial and temporal distribution of females affect male encounter rates with subadult and adult females and thus the feasibility of a newly- described alternative reproductive tactic, immature mating. I first use survey data to (i) assess the opportunity for immature mating in nature by comparing rates of male cohabitation with subadult and adult females and (ii) to assess ecological correlates of male cohabitation with subadults. I then experimentally test the hypothesis that proximity to signaling adult females facilitates immature mating in the absence of direct cues that would allow males to locate subadults.

In chapter 3 I assess plasticity in female mate choice decisions resulting from variation in social context experienced as a subadult and young adult. I use a field experiment to examine how spatial distribution (proximity to conspecific females) affects female encounter rates with potential mating partners. Subsequently, I staged mating trials to test the hypothesis that females adjust choosiness in response to previous experience with cues of mate availability and conspecific density.

In chapter 4 I review the current literature on silk use by males in the context of mating across spider taxa, using a framework of potential functions and mechanisms of effect. I present evidence that male silk has an important role in mate attraction, courtship, and mating and thus provides opportunities to test hypotheses about the evolution of male and female traits under sexual selection and/or sexual conflict. This chapter has been published (Scott CE, Anderson A, Andrade MCB. 2018 A review of the mechanisms and functional roles of male silk use in spider courtship and mating. Journal of Arachnology 46: 173-206. doi:10.1636/JoA-S-17-093.1).

References

Able, D. J. (1999). Scramble competition selects for greater tailfin size in male red-spotted newts (Amphibia: Salamandridae). Behavioral Ecology and Sociobiology, 46(6), 423-428. Alberts, A. C. (1992). Constraints on the design of chemical communication systems in terrestrial vertebrates. The American Naturalist, 139, S62-S89. Allen, R. M., Buckley, Y. M., & Marshall, D. J. (2008). Offspring size plasticity in response to intraspecific competition: an adaptive maternal effect across life-history stages. The American Naturalist, 171(2), 225-237. Andersson, M. B. (1994). Sexual Selection. Princeton, NJ: Princeton University Press. Andersson, M., & Iwasa, Y. (1996). Sexual selection. Trends in Ecology & Evolution, 11(2), 53- 58. Andrade, M. C. B., & MacLeod, E. C. (2015). Potential for CFC in black widows (genus Latrodectus): mechanisms and social context. In Peretti, A. V, & Aisenberg, A. (Eds.), Cryptic Female Choice in Arthropods (pp. 27-53). Switzerland: Springer International. Arnqvist, G., & Rowe, L. (2013). Sexual Conflict. Princeton, NJ: Princeton University Press. Bailey NW, Zuk M. 2008 Acoustic experience shapes female mate choice in field crickets. Proceedings of the Royal Society B: Biological Sciences 275, 2645–2650. Baruffaldi, L., & Andrade, M. C. B. (2015). Contact pheromones mediate male preference in black widow spiders: avoidance of hungry sexual cannibals? Animal Behaviour, 102, 25- 32. Bateman, A. J. (1948). Intra-sexual selection in Drosophila. Heredity, 2(3), 349-368. Belanger, R. M., & Corkum, L. D. (2009). Review of aquatic sex pheromones and chemical communication in anurans. Journal of Herpetology, 43(2), 184-191. Berglund, A. (1995). Many mates make male pipefish choosy. Behaviour, 132(3), 213-218. Bradbury, J. W., & Vehrencamp, S. L. (1998). Principles of Animal Communication. Sunderland, MA: Sinauer Associates. Brockmann, H. J. (1990). Mating behavior of horseshoe crabs, Limulus polyphemus. Behaviour, 114(1), 206-220. Clutton-Brock, T. H., & Parker, G. A. (1992). Potential reproductive rates and the operation of sexual selection. The Quarterly Review of Biology, 67(4), 437-456. Danchin, É., Giraldeau, L. A., Valone, T. J., & Wagner, R. H. (2004). Public information: from nosy neighbors to cultural evolution. Science, 305(5683), 487-491. Darwin, C. (1871). Sexual Selection and the Descent of Man. London, UK: Murray. de Jong, K., Forsgren, E., Sandvik, H., & Amundsen, T. (2012). Measuring mating competition correctly: available evidence supports operational sex ratio theory. Behavioral Ecology, 23(6), 1170-1177. Dominey, W. J. (1984). Alternative mating tactics and evolutionarily stable strategies. American Zoologist, 24(2), 385-396.

Duvall, D., Schuett, G. W., & Arnold, S. J. (1993). Ecology and evolution of snake mating systems. In: Seigel, R. A. & Collings, J. T. (Eds.), Snakes: Ecology and Behavior (pp. 165-200). New York, NY: McGraw Hill. Eberhard, W. (1996). Female control: Sexual Selection by Cryptic Female Choice. Princeton, NJ: Princeton University Press. Elias, D. O., Andrade, M. C., & Kasumovic, M. M. (2011). Dynamic population structure and the evolution of spider mating systems. Advances in Insect Physiology 41, 65-114. Emlen, S. T., & Oring, L. W. (1977). Ecology, sexual selection, and the evolution of mating systems. Science, 197, 215-223. Endler, J. A. (1992). Signals, signal conditions, and the direction of evolution. The American Naturalist, 139, S125-S153. Endler, J. A. (1993). Some general comments on the evolution and design of animal communication systems. Philosophical Transactions of the Royal Society B: Biological Sciences, 340, 215-225. Evans, J. P., & Garcia‐Gonzalez, F. (2016). The total opportunity for sexual selection and the integration of pre‐and post‐mating episodes of sexual selection in a complex world. Journal of Evolutionary Biology, 29(12), 2338-2361. Forsgren, E., Amundsen, T., Borg, Å. A., & Bjelvenmark, J. (2004). Unusually dynamic sex roles in a fish. Nature, 429, 551-554. Galef, B. G. (2009). Strategies for social learning: testing predictions from formal theory. Advances in the Study of Behavior, 39, 117-151. Gaskett, A. C. (2007). Spider sex pheromones: emission, reception, structures, and functions. Biological Reviews, 82(1), 27-48. Gibson, R. M., & Höglund, J. (1992). Copying and sexual selection. Trends in Ecology & Evolution, 7(7), 229-232. Gosling, L. M., & McKay, H. V. (1990). Competitor assessment by scent matching: an experimental test. Behavioral Ecology and Sociobiology, 26(6), 415-420. Gross, M. R. (1996). Alternative reproductive strategies and tactics: diversity within sexes. Trends in Ecology & Evolution, 11(2), 92-98. Grüter, C., & Leadbeater, E. (2014). Insights from insects about adaptive social information use. Trends in Ecology & Evolution, 29(3), 177-184. Gwynne, D. T., Bailey, W. J., & Annells, A. (1998). The sex in short supply for matings varies over small spatial scales in a katydid (Kawanaphila nartee, Orthoptera: Tettigoniidae). Behavioral Ecology and Sociobiology, 42(3), 157-162. Hayes, C. L., Callander, S., Booksmythe, I., Jennions, M. D., & Backwell, P. R. (2016). Mate choice and the operational sex ratio: an experimental test with robotic crabs. Journal of Evolutionary Biology, 29(7), 1455-1461. Head, M. L., Lindholm, A. K., & Brooks, R. (2008). Operational sex ratio and density do not affect directional selection on male sexual ornaments and behavior. Evolution: International Journal of Organic Evolution, 62(1), 135-144.

Hebets EA. 2003 Subadult experience influences adult mate choice in an arthropod: exposed female wolf spiders prefer males of a familiar phenotype. Proceedings of the National Academy of Sciences 100, 13390–13395. Henshaw, J. M. (2018). Finding the one: optimal choosiness under sequential mate choice. Journal of Evolutionary Biology, 31(8), 1193-1203. Herberstein, M. E., Painting, C. J., & Holwell, G. I. (2017). Scramble competition polygyny in terrestrial arthropods. Advances in the Study of Behavior. 49, 237-295. Hutchings, J. A., & Myers, R. A. (1994). The evolution of alternative mating strategies in variable environments. Evolutionary Ecology, 8(3), 256-268. Ibáñez, A., López, P., & Martín, J. (2012). Discrimination of conspecifics’ chemicals may allow Spanish terrapins to find better partners and avoid competitors. Animal behaviour, 83(4), 1107-1113. Ims, R. A. (1988). The potential for sexual selection in males: effect of sex ratio and spatiotemporal distribution of receptive females. Evolutionary Ecology, 2(4), 338-352. Jennions MD, Petrie M. 1997 Variation in mate choice and mating preferences: a review of causes and consequences. Biological Reviews 72, 283–327. Jerhot, E., Stoltz, J. A., Andrade, M. C., & Schulz, S. (2010). Acylated serine derivatives: a unique class of arthropod pheromones of the Australian redback spider, Latrodectus hasselti. Angewandte Chemie International Edition, 49(11), 2037-2040. Johansson, B. G., & Jones, T. M. (2007). The role of chemical communication in mate choice. Biological Reviews, 82(2), 265-289. Johnston, R. E. (2003). Chemical communication in rodents: from pheromones to individual recognition. Journal of Mammalogy, 84(4), 1141-1162. Karlson, P., & Butenandt, A. (1959). Pheromones (ectohormones) in insects. Annual Review of Entomology, 4(1), 39-58. Kasumovic, M. M., & Andrade, M. C. (2004). Discrimination of airborne pheromones by mate- searching male western black widow spiders (Latrodectus hesperus): species-and population-specific responses. Canadian Journal of Zoology, 82(7), 1027-1034. Kasumovic MM, Andrade MCB. 2006 Male development tracks rapidly shifting sexual versus natural selection pressures. Current Biology 16, R242–R243. Kasumovic MM, Hall MD, Brooks RC. 2012 The juvenile social environment introduces variation in the choice and expression of sexually selected traits. Ecology and Evolution 2, 1036–1047. Klug, H., Heuschele, J., Jennions, M. D., & Kokko, H. (2010). The mismeasurement of sexual selection. Journal of Evolutionary Biology, 23(3), 447-462. Kokko, H., Klug, H., & Jennions, M. D. (2014) Mating systems. In: Shuker, D. M. & Simmons, L. W. (Eds.) The Evolution of Insect Mating Systems. Oxford: Oxford University Press. Kokko H, Mappes J. 2005 Sexual selection when fertilization is not guaranteed. Evolution 59, 1876–1885.

Krakauer, A. H., Webster, M. S., Duval, E. H., Jones, A. G., & Shuster, S. M. (2011). The opportunity for sexual selection: not mismeasured, just misunderstood. Journal of Evolutionary Biology, 24(9), 2064-2071. Kvarnemo, C., & Ahnesjo, I. (1996). The dynamics of operational sex ratios and competition for mates. Trends in Ecology & Evolution, 11(10), 404-408. Lane, J. E., Boutin, S., Speakman, J. R., & Humphries, M. M. (2010). Energetic costs of male reproduction in a scramble competition mating system. Journal of Animal Ecology, 79(1), 27-34. Lloyd, J. E. (1979). Mating behavior and natural selection. The Florida Entomologist, 62(1), 17- 34. MacLeod, E. C. (2013). New insights in the evolutionary maintenance of male mate choice behaviour using the western black widow, Latrodectus hesperus. Doctoral dissertation, University of Toronto. MacLeod, E. C., & Andrade, M. C. B. (2014). Strong, convergent male mate choice along two preference axes in field populations of black widow spiders. Animal Behaviour, 89, 163- 169. Marmet, J., Pisanu, B., Chapuis, J. L., Jacob, G., & Baudry, E. (2012). Factors affecting male and female reproductive success in a chipmunk (Tamias sibiricus) with a scramble competition mating system. Behavioral Ecology and Sociobiology, 66(11), 1449-1457. Muñoz, A. (2004). Chemo-orientation using conspecific chemical cues in the stripe-necked terrapin (Mauremys leprosa). Journal of Chemical Ecology, 30(3), 519-530. Parker, G. A. (1970). Sperm competition and its evolutionary consequences in the insects. Biological Reviews, 45(4), 525-567. Pigliucci, M. 2001 Phenotypic Plasticity: Beyond Nature and Nurture. Baltimore, MD: John Hopkins University Press. Rieucau, G., & Giraldeau, L. A. (2011). Exploring the costs and benefits of social information use: an appraisal of current experimental evidence. Philosophical Transactions of the Royal Society B: Biological Sciences, 366, 949-957. Rios-Cardenas, O., & Morris, M. R. (2009). Mating systems and strategies of tropical fishes. Tropical Biology and Conservation Management, 8, 219-240. Rhainds, M. (2019). Ecology of female mating failure/lifelong virginity: a review of causal mechanisms in insects and arachnids. Entomologia Experimentalis et Applicata, 167(1), 73-84. Ross, K., & Smith, R. L. (1979). Aspects of the courtship behavior of the black widow spider, Latrodectus hesperus (Araneae: Theridiidae), with evidence for the existence of a contact sex pheromone. Journal of Arachnology, 69-77. Salomon, M., Vibert, S., & Bennett, R. G. (2010). Habitat use by western black widow spiders (Latrodectus hesperus) in coastal British Columbia: evidence of facultative group living. Canadian Journal of Zoology, 88(3), 334-346. Schneider, J. M., Lucass, C., Brandler, W., & Fromhage, L. (2011). Spider males adjust mate choice but not sperm allocation to cues of a rival. Ethology, 117(11), 970-978.

Schwagmeyer, P. L., & Woontner, S. J. (1986). Scramble competition polygyny in thirteen-lined ground squirrels: the relative contributions of overt conflict and competitive mate searching. Behavioral Ecology and Sociobiology, 19(5), 359-364. Scott, C., McCann, S., Gries, R., Khaskin, G., & Gries, G. (2015a). N-3-Methylbutanoyl-O- methylpropanoyl-l-serine methyl ester–pheromone component of western black widow females. Journal of Chemical Ecology, 41(5), 465-472. Scott, C., Kirk, D., McCann, S., & Gries, G. (2015b). Web reduction by courting male black widows renders pheromone-emitting females' webs less attractive to rival males. Animal Behaviour, 107, 71-78. Scott, C., Vibert, S., & Gries, G. (2012). Evidence that web reduction by western black widow males functions in sexual communication. The Canadian Entomologist, 144(5), 672-678. Shuster, S. M., & Wade, M. J. (2003). Mating Systems and Strategies. Princeton, NJ: Princeton University Press. Snow, L. S., Abdel‐Mesih, A., & Andrade, M. C. B. (2006). Broken Copulatory Organs are Low‐Cost Adaptations to Sperm Competition in Redback Spiders. Ethology, 112(4), 379- 389. Svensson, M. (1996). Sexual selection in moths: the role of chemical communication. Biological Reviews, 71(1), 113-135. Taborsky, M. (2001). The evolution of bourgeois, parasitic, and cooperative reproductive behaviors in fishes. Journal of Heredity, 92(2), 100-110. Thiel, M., & Breithaupt, T. (2010). Chemical communication in crustaceans: research challenges for the twenty-first century. In Breithaupt, T. & Thiel, M. (Eds.), Chemical Communication in Crustaceans (pp. 3-22). New York, NY: Springer. Thomas, M. L. (2011). Detection of female mating status using chemical signals and cues. Biological Reviews, 86(1), 1-13. Thornhill, R., & Alcock, J. (1983). The Evolution of Insect Mating Systems. Cambridge, MA: Harvard University Press. Trivers, R. L. 1972. Parental investment and sexual selection. In Campbell, B. (Ed.), Sexual Selection and the Descent of Man (pp. 136-179). London: Heinemann. Uhl, G. (2000). Female genital morphology and sperm priority patterns in spiders (Araneae). European Arachnology, 2000, 145-156. Uhl, G., and Elias, D. O. 2011. Communication, pp. 127–189, in M. E. Herberstein (ed.). Spider Behaviour: Flexibility and Versatility. Cambridge UK: Cambridge University Press. Valone, T. J., & Templeton, J. J. (2002). Public information for the assessment of quality: a widespread social phenomenon. Philosophical Transactions of the Royal Society B: Biological Sciences, 357, 1549-1557. Vibert, S., Scott, C., & Gries, G. (2016). Vibration transmission through sheet webs of hobo spiders (Eratigena agrestis) and tangle webs of western black widow spiders (Latrodectus hesperus). Journal of Comparative Physiology A, 202(11), 749-758.

Wade, M. J., & Arnold, S. J. (1980). The intensity of sexual selection in relation to male sexual behaviour, female choice, and sperm precedence. Animal Behaviour, 28(2), 446-461. Wade, M. J., & Pruett-Jones, S. G. (1990). Female copying increases the variance in male mating success. Proceedings of the National Academy of Sciences, 87(15), 5749-5753. Watson, P. J. (1986). Transmission of a female sex pheromone thwarted by males in the spider Linyphia litigiosa (Linyphiidae). Science, 233, 219-221. Weir, L. K., Grant, J. W., & Hutchings, J. A. (2011). The influence of operational sex ratio on the intensity of competition for mates. The American Naturalist, 177(2), 167-176. Wells, K. D. (2007). The Ecology and Behaviour of Amphibians. Chicago, IL: University of Chicago Press. West-Eberhard, M. J. (1979). Sexual selection, social competition, and evolution. Proceedings of the American Philosophical Society, 123(4), 222-234. West-Eberhard MJ. 2003 Developmental Plasticity and Evolution. Oxford, UK: Oxford University Press. White, D. J. (2004). Influences of social learning on mate-choice decisions. Animal Learning & Behavior, 32(1), 105-113. White, D. J., Davies, H. B., Agyapong, S., & Seegmiller, N. (2017). Nest prospecting brown- headed cowbirds ‘parasitize’ social information when the value of personal information is lacking. Proceedings of the Royal Society B, 284, 20171083. Wyatt, T. D. (2003). Pheromones and Animal Behaviour: Communication by Smell and Taste. Cambridge, UK: Cambridge University Press.

Male black widows parasitize mate-searching effort of rivals to find females faster

Catherine E Scott, Sean McCann, and Maydianne CB Andrade

Abstract

Mate-searching success is a critical precursor to mating, but there is a dearth of research on traits and tactics that confer a competitive advantage in finding potential mates. Theory and available empirical evidence suggest that males locate mates using mate-attraction signals produced by receptive females (personal information) and avoid inadvertently produced cues from rival males (social information) that indicate a female has probably already mated. Here, we show that western black widow males use both kinds of information to find females efficiently, parasitizing the searching effort of rivals in a way that guarantees competition over mating after reaching a female’s web. This tactic may be adaptive because female receptivity is transient, and we show that (i) mate searching is risky (88% mortality) and (ii) a strongly male-biased operational sex ratio (from 1.2:1 to more than 10:1) makes competition inevitable. Males with access to rivals’ silk trails moved at higher speeds than those with only personal information, and located females even when personal information was unreliable or absent. We show that following rivals can increase the potential for sexual selection on females as well as males and argue it may be more widespread in nature than is currently realized.

1.1 Introduction

Sexual selection arises when the reproductive success of one sex is limited by access to potential mates [1]. In most sexually reproducing species, males compete to fertilize the relatively limited number of eggs produced by females, and the form of competition depends on ecological factors including the distribution of potential mates in space and time [2]. In many taxa, females become sexually receptive at unpredictable spatial or temporal intervals and males increase their fitness through sequential mate acquisition rather than monopolization. The resulting mating system is ‘scramble competition polygyny’ [3–5]. If windows of female receptivity are brief, or the first male to mate fathers the majority of a female’s offspring, selection on scrambling males is particularly intense [5]. Scramble competition polygyny can broadly affect the evolution of

15 traits, including sexual dimorphism in development time and body size, and male speed, agility and sensitivity to female signals [4,5]. More mobile males encounter more potential mates [6], but also pay high costs [5], including energetic costs [7] and elevated mortality rates [8].

Mate-searching success is a critical precursor to mating, and only those males that successfully find females will be exposed to sexual selection via contest competition and/or female choice. Nevertheless, while studies of competition and choice are common [9], those investigating traits and tactics that confer a competitive advantage in scramble competition are scarce [10–12]. This is problematic since mate searching will often function as the first filter in the episodes of selection that determine reproductive success [13]. Moreover, mate searching may affect the importance of other components of sexual selection, particularly when males’ search tactics are tuned to the risk of competition over potential mates [14–17]. Overall, the dearth of studies of mate searching leaves a considerable deficit in our understanding of sexual selection in nature.

Mate-searching performance will often depend on how males use information that indicates the location and status of females. Males may use personal information, obtained directly via exploration or observation, and they may also acquire social information indirectly by observing or detecting the behaviour of other animals [18]. Integrating both types of information may increase the fitness payoff for behavioural decisions [19,20], although some species use social information only when it complements personal information [19,21]. In other species, social information may be preferred [22], particularly when acquiring individual information is risky [23], or when variable environmental conditions decrease the reliability of personal information [24]. For example, spatio-temporal variation in the environment can alter the transmission or detection of information [25,26], with differential effects on the availability of personal and social information. How and when animals use personal and social information to make mating-related decisions will depend on conditions including temperature/weather, physical structure of the habitat, and ecological variables such as predation risk and the spatiotemporal distribution of conspecifics. These factors will affect the fitness benefit and feasibility of acquiring and using social compared to personal information, so information use is expected to be context-dependent [22,27].

Understanding when social information is used in mate searching is important because the range of genotypes accruing high fitness via choice or competition can be altered relative to contexts where only personal information is used [28]. For example, females commonly use social information to copy the mate choices of other females [29], and the result can be a strong reproductive advantage for relatively few preferred males [30]. By contrast, male mate-choice copying is generally unexpected, because it will increase the risk of sperm competition, particularly in species with first-male sperm precedence [29,31]. Typically, males are expected to use social information (i.e. cues of the presence of rivals) to avoid or reduce investment in females that are likely to have already mated, even if these females are otherwise preferred [14– 17]. In contrast with copying, this can decrease inter-male competition over fertilization and reduce variation in female mating rates. Rare exceptions occur in cases of sex role reversal (e.g. pipefish [32]), in some unusual mating systems (e.g. sailfin mollies, where males are targeted by a gynogenetic congener [33]) and in cases where the presence of other males may be a marker of costly to ascertain female receptivity (e.g. meadow voles [34]). However, most studies showing male mate-choice copying were conducted in the laboratory (e.g. [34,35], but see [33]). It is not clear whether these patterns would manifest under complex environments in the field, or in a wider range of taxa. Thus, while the use of social information by males may be context- dependent, can change the direction or intensity of selection and affect evolutionary processes [36], it is largely unknown how mate-searching males use social information in nature.

We examined mate searching and information use by males in a species with intense competition over access to females [37]. By pairing insight from observational and experimental studies in the field with focused laboratory experiments, we provide a rare, replicated examination of information use in male mate searching in complex environmental conditions. We studied the western black widow spider (Latrodectus hesperus), a sexually dimorphic species with first-male sperm precedence mediated by sperm plugs [38]. In this species, males typically develop on solitary webs, mature before females and must search for sedentary mates over complex terrain [38]. In three congeners, mate searching leads to high mortality (greater than 80% [39–41]), imposing strong selection on efficient detection of and movement towards signalling females [6,42,43]. Black widow females signal receptivity using a short-lived, volatile sex pheromone and cease to be attractive or detectable during or immediately after mating [38,44]. Thus, males detecting airborne pheromones receive reliable personal information about

17 the presence and location of unmated females. Social information is also available because male spiders trail silk ‘safety lines’ behind them as they move through their habitat [45]. Contact with the silk of other males could thus provide chemical and/or tactile social information [46] about the presence and movement paths of rivals [15].

Here, we investigated the use of these forms of personal and social information during mate searching as a function of variables affecting costs and benefits of information use. We hypothesized that mate-searching males should make context-dependent decisions about whether to use social information, and about whether to avoid or follow competitors. The detection of personal information provided by female airborne pheromones is likely to be affected by environmental variables [25,26] that will not affect the availability of social cues based on physical contact with the silk of rivals. Moreover, the fitness effect of following or avoiding rivals should depend on the local competitive environment, and trade-offs between the risk of sperm competition (favouring the avoidance of rivals), and benefits of efficient searching (favouring use of all available information to find females quickly).

We report the results of a longitudinal field study showing that scramble competition polygyny is the most likely mating system at our field site (consistent with congeners [38]). We use census data to estimate (i) the mortality rate during mate searching as a measure of the potential intensity of selection [47] on mate-searching males and (ii) the operational sex ratio (OSR; the ratio of sexually active males to fertilizable females [3]) as a measure of the degree of competition to acquire mates across the season [48]. At high OSR, competition is inevitable [49], and rapid, efficient searching for females may be favoured. At low OSR, with fewer competitors, avoiding rivals may yield the highest payoffs.

We then use experimental field studies to show that male black widows detect and respond to both personal and social information in the context of mate searching. We confirm predictions of this inference via laboratory experiments that manipulate the availability of different information sources. In these experiments, we predicted (i) in the presence of both kinds of information, males would avoid females when male silk cues (social information) suggested the presence of a competitor [15], but that (ii) when personal information was not available, males should follow the draglines of rivals as using this social information would be

18 superior to a random search of the habitat. Thus, we expected to find plasticity in ‘mate-search copying’, despite strong first-male sperm precedence.

1.2 Materials and Methods

1.2.1 Mating system and the intensity of sexual selection

We studied a population of western black widows (Latrodectus hesperus Chamberlin & Ivie, 1935) inhabiting a roughly 20 × 400 m area of coastal sand dunes above the high-tide line on Vancouver Island, British Columbia, Canada (48°350 1000 N, 123°220 1700 W). At this site, driftwood logs, woody debris and rocks provide microhabitats, and these features combined with low-lying vegetation covering the dunes provide challenging terrain for male locomotion. The population is dense, with on average 2–3 subadult or adult females per square metre of microhabitat during the late summer mating season [50].

From April to September 2016, we surveyed the population (see electronic supplementary material, S1 for methodological details; each data point in figure 1 represents a survey date), tracked the movement of marked males among microhabitats and webs, and used these data to estimate mortality and distances travelled during mate searching. We also used demographic data from our surveys to calculate the ratio of adult males to final-instar (subadult) or sexually receptive adult females (OSR) over 2–10-day intervals across the late summer mating season. The OSR is a widely used metric predicting the direction and intensity of competition over mates [3,48,49,51].

1.2.2 Field experiments on male mate searching under varying information availability

We asked how distance from pheromone-emitting females and availability of social information affected male mate-searching success in two male-release field experiments (see electronic supplementary material, S1 for details about animal collection and husbandry). Released males moved over natural terrain (in an area without any naturally occurring females), searching for unmated, pheromone-emitting females and their webs contained in mesh cages (pheromone traps) [37,44]. We ran experiments on nights forecasted to have a southerly wind, which would place males upwind of pheromone traps (figure 2).

Prior to release, each male was marked with a unique colour code (electronic supplementary material, figure S1) using quick-drying modelling paint (Testor’s enamel ), which does not decrease survivorship [39]. Male size and condition affect mate-searching success and speed in spiders [42,43], so we measured males from photographs (size) and weighed males to calculate size-corrected mass (an index of condition, 2017 only, see electronic supplementary material, S1 for details).

Mate-searching experiments (7 September 2016; 11 September 2017) commenced after sunset (L. hesperus is nocturnal), when we placed 11 pheromone traps containing unmated females and their webs along a 20 m (2016; spaced 2 m apart) or 40 m (2017; spaced 4 m apart) east–west transect (approximately perpendicular to the forecasted wind direction; figure 2). We released uniquely marked males in groups (n ∼ 20 per group, see electronic supplementary material, table S1) at 10 m intervals along a 60 m north–south transect (2016: n = 117 males; 2017: n = 130 males), starting at the 60 m point (farthest from females). In this design, the probability of encountering silk draglines of other experimental males (social information) increases with distance from the line of females.

We continuously monitored male arrival at pheromone traps by walking along the line of traps and collecting males that were on or within 15 cm of each trap (2016), or that had crossed beyond the line of traps (2017; they were considered to have found the female nearest to the crossing point). The 2016 data were informative, but using this design, males would often remain still or slowly wander between two cages, likely because they could not localize the source of the pheromone when competing sources overlapped. We changed our design in 2017 by spacing females every 4 m instead of every 2 m. Both spacing patterns are realistic challenges for mate- searching males in nature, where females may share the same microhabitat (webs within 2 m [50]) or may be spaced across different microhabitats (mean nearest-neighbour distance=4.2m; C.E.S. 2016, unpublished data). After each experiment, we obtained hourly records of the average wind speed and direction from a nearby weather station.

1.2.3 Laboratory tests of context-dependent male information use

We tested the effects of personal compared to social information in the laboratory using an X- maze choice apparatus made of string which allowed males to choose or avoid the cues left by other males (figure 3a; see electronic supplementary material, S1 for experimental protocol).

This set-up reduced the mobility advantage of following silk relative to field conditions by allowing spiders to walk continuously on a small-diameter substrate [52].

In three laboratory experiments, we gave males the option to use or avoid the silk trails of a rival male (social information) when choosing a path of movement. We used naive males reared in the laboratory (n = 22 in each experiment). We first asked (figure 3b) whether males use social information when they simultaneously have access to personal information (fan- generated ‘wind’ blowing female pheromone towards the male), mimicking the conditions of the 2016 field experiment (figure 2c). Second (figure 3c), we tested male responses to social information in the absence of personal information but the presence of wind, which represents profitable conditions for seeking personal information (travelling upwind may result in the male encountering a pheromone plume in the field). Third (figure 3d), we asked if males would respond to the social information in the absence of personal information and wind, such that social information represents the only available cue of the location of a potential mate.

Two additional X-maze experiments examined whether males respond to the silk of conspecific male competitors in particular (with clear implications for sexual selection), or to spider silk in general (which may reflect general attraction to habitat suitable for spiders [18]). We tested male response to the silk of syntopic, confamilial Steatoda grossa (C. L. Koch, 1838) (Theridiidae) males collected from our field site (see electronic supplementary material, S1 for more details about experimental animals). First, in the presence of a pheromone-emitting conspecific female, male black widows (n = 32) were given a choice between heterospecific silk and a string-only control path (figure 3e). Next, following trials (n = 16) in which the test male avoided the heterospecific silk, a second test male was introduced to the apparatus and allowed to choose between the silk of the heterospecific male and the conspecific male (figure 3f ). These trials took place outdoors in Saanich, British Columbia, at a location with no naturally occurring L. hesperus.

1.2.4 Statistical analyses

Mate searching in the field. We ran general linear models in R (v. 3.2.0 [53]) to determine the effects of distance from females on (i) probability of arriving at a pheromone trap and (ii) average speed of males that successfully found females (time elapsed/ linear distance travelled). We included male size (2016, 2017) and body condition index (2017) in these models as

21 covariates. Models with probability of recapture as the response variable used a binomial distribution and logit link, and models with speed as the response variable used the normal distribution and identity link. We examined the residuals of each model to determine whether assumptions of normality and homoscedasticity were violated and we log-transformed male speed to meet model assumptions.

X-maze choice trials. We used binomial tests to determine whether males preferred one side of the X-maze, and multiple regression to determine whether the silk of another male affected average male speed (figure 3b). Using order (whether the male was first or second to enter the maze) as a predictor allowed us to test the effect of silk left by the first male on the second male’s speed, with male size and body condition as covariates. We subsequently ran separate models using leg length and body condition as predictors because they were negatively correlated for males used in this experiment (R2 = 0.36; F1,36 = 20.21; p < 0.0001).

1.3 Results 1.3.1 Operational sex ratio, mate-searching success, and mortality in nature

We recorded the location and developmental stage of 461 females, and marked and tracked 290 males that arrived at females’ webs throughout the active season for this population (April– September 2016). Our data show that males mature earlier than females (protrandry; figure 1), consistent with a scramble competition mating system [5]. Most sexual activity occurred in August, when males were most abundant. During this period, the OSR was consistently male biased (greater than or equal to 5 : 1; figure 1c), and a conservative estimate indicates fewer than 15 receptive females were available on most nights (figure 1b; Appendix 1). The number of mate-searching males decreased in September and the OSR was lower, but remained male-biased (1.2 : 1–7.5 : 1) until just before the end of the active season when few males remained and the OSR became slightly female-biased (figure 1c).

The risk of mortality during male movement between webs was approximately 88%. We estimated male survival in three ways: (i) 2/16 males (12.5%) first marked on their juvenile web were later found on females’ webs, (ii) 33/274 (12%) males first marked on a female’s web were later found on a second female’s web, and (iii) 4 of those 33 males (12%) were subsequently

22 found on a third female’s web. Most males were found on webs within 60 m (= more than 9000 body lengths for males of maximum size 6.5 mm) of their point of origin, but some moved more than 200 m (Appendix 1, figure S2).

1.3.2 Mate-searching success, speed, and social information use in the field

During the male-release study in 2016 (figure 2), the wind speed was high (mean speed 13 km h−1) and its direction relatively consistent (as forecast, figure 2c). Mate-searching success rates were high (62%), with males at all distances (up to 60 m) equally likely to find females (figure 4a; electronic supplementary material, tables S1 and S2). Successful males were distributed across the webs of females such that 73% of available females attracted at least one potential mate, with females most directly upwind from the male-release vector attracting the most males (figure 2a,c). This pattern is consistent with males using personal information (volatile pheromones) to find females over long distances. By contrast, during the 2017 replicate, the wind was weak (mean speed 4 km h−1) and its direction variable, so consistent personal information was not likely to be available (figure 2d) [54]. Overall searching success was much lower (26%) and decreased with distance such that no males released at 50 or 60 m were successful (figure 4b; Appendix 1, tables S1 and S2). Moreover, only 46% of females attracted at least one male, and there was strong coincidence of male attraction, with 82% of successful males arriving at the web of the single female directly in line with the release vector (and the wind direction during the first hour of the experiment; figure 2b,d). In this second replicate, males had access to personal information primarily during the first 2 h of the experiment, after which wind conditions would have limited transmission (figure 2d) [54]. Nevertheless, more than half of all males (53%) that found females arrived after the first 2 h, when personal information was unavailable or inconsistent. A third, modified version of this experiment showed comparable results under similarly weak and variable wind conditions (electronic supplementary material, figures S3 and S4 and table S4).

The maximum average speeds attained by males under poor wind conditions (less than

0.25 m min−1 in 2017) were much lower than those of males searching in strong, consistent wind

(maximum speeds of greater than 1.25 m min−1 in 2016; figure 4c,d). Surprisingly, males moved more quickly when released further away from females, particularly in 2016 when personal information was consistently available (figure 4c,d; Appendix 1, table S2). Since we released

23 males in sequence from 60 to 10 m from females, those males released from the furthest points were more likely to encounter silk and had access to longer pathways of rival silk (produced by males released slightly later, but closer to females). We frequently observed males moving along silk at a rapid pace, much more so than when traversing the ground or vegetation. We infer that the availability of social information and/or silk pathways facilitates efficient locomotion.

1.3.3 Context-dependent male responses to social information in the laboratory

When males were released in the X-maze in the presence of directional personal information ( pheromones and wind) and given the choice between a string-only path and string covered in conspecific male silk, 95% (n = 22) of males chose to follow the silk of the rival male (p = 0.00001; figure 3a,b). When only social information was available (silk and wind but no pheromones), all males (n = 22) readily moved upwind but showed no preference for following silk (59% chose the silk path; p = 0.53; figure 3c). In the absence of wind and pheromones, however, 82% (n = 22) of males followed silk trails left by conspecific rivals rather than the control path (p = 0.004; figure 3d ).

These responses were specific to conspecific rivals. Latrodectus hesperus males showed no response to the silk of S. grossa males relative to control strings (p = 0.5; figure 3e), and 87% (n = 15; one male did not make a choice) preferred to follow conspecific male silk when both were present (p = 0.007; figure 3f ).

In the laboratory, males with access to silk draglines moved towards females more quickly than those with only personal information (model with size as covariate: t = 1.9, p = 0.065; model with body condition as covariate: t = 2.3, p = 0.028; figure 3g,h; Appendix 1, table S3), indicating that social information coupled with personal information allows males to find females more rapidly, as in the field.

1.4 Discussion

Our results show that L. hesperus males follow the silk of male conspecifics during mate searching, that this response is specific to information left by potential sexual rivals and that males use social information to determine movement pathways if there is no possibility of acquiring personal information (e.g. no wind). Furthermore, social information combined with

24 personal information allows males to find females most efficiently. Contrary to our predictions, when males paid attention to social information, they consistently followed the silk of other males, with little evidence for variation in how the information was used.

Despite increasing the risk of contest and sperm competition, following the silk trails of rivals is likely to be adaptive for males in this population since it increases the speed and efficiency of finding females, but is not likely to significantly increase the risk of competition. The extremely male-biased OSR (greater than 10:1) during the peak mating period, along with very few receptive females available on any given night, means intense competition over females is inevitable. Moreover, few males (12%) survive the trip to find a female, so there is strong selection on males to succeed, even if they then have to compete with an accumulation of males for the few females that are sexually receptive on a given night. Although being first to mate is important for success in sperm competition in this species, being the first male to arrive at a female’s web is not critical. Courtship may last several hours before the first copulation occurs [38], so later-arriving males may still be successful at mating if they win in contest competition or are chosen after assessment by the female. In fact, later-arriving males may be more likely to mate if the courtship effort of rivals has already induced female receptivity by the time they arrive at the web [55].

Our results indicate that social information (chemical and/or structural cues) on the silk of earlier searchers can allow a male to find a female efficiently even when access to personal information (volatile sex pheromone) is minimized. This is consistent with the prediction that animals should use social information when personal information is out of date, unreliable or expensive to obtain [22–24,27]. Male black widows apparently move upwind to seek personal information (figure 3c), but when weak or absent wind makes detection of a pheromone source difficult or impossible, they can nonetheless find females by following the silk trails of rivals (figures 2b,d and 3d ). After initially detecting a female using personal information, social information produced by an early-arriving rival will allow a following male to rapidly catch up, even if wind conditions change. This may be of particular benefit to L. hesperus males because the signals of sexually receptive females are ephemeral. Courting males can interfere with transmission of the female’s signal by altering the web [37], rendering the female unattractive or undetectable during and after mating [38,44].

Social information increased male mate-searching efficiency when personal information was available simultaneously. Males that had access to conspecific silk moved more quickly; an effect that was suggested in the field and reproduced experimentally in the laboratory (figures 3g,h and 4c,d; Appendix 1, tables S1 and S2). Similarly, synergistic effects when combining social with personal information have been found for foraging ants and bumblebees [20,21] and for birds gauging brood parasitism risk [19]. The recency of social information is also likely to be important [56,57]; the benefit of following a rival should decrease with time since his trail was produced. Chemicals on draglines may provide reliable information precisely because the cues are transient; in other species, silk-borne chemicals are water soluble and washed away daily by dew [58]. Regardless, in the absence of any personal information about the location of receptive females, the expected payoff from following an old trail to an already-mated female and obtaining some share of paternity is likely to exceed the payoff from searching randomly for unmated females with the attendant risk of mortality or failure.

In scramble competition polygyny, males maximize their fitness via sequential mate acquisition rather than mate monopolization [12,34]. In these systems, more active males that move longer distances will encounter more females [4,5], but because mate searching is energetically costly [7] and increases predation risk [8], mating success should be higher for males that spend less time searching. Scramble competition can severely impair the evolution of choosiness [59] and mate-choice copying can be adaptive when mate choice is costly [28,30]. We expect that using social information to find females efficiently should also be adaptive in other taxa with scramble competition polygyny, particularly when the costs of searching are high. Since scramble competition is common across taxa, including arthropods [5], amphibians [6], reptiles [60], fishes [61] and mammals [34], this may be a common tactic that is currently unappreciated.

The use of inadvertent social information (produced as a byproduct of locomotion) by male black widows to locate females is best described as an example of local enhancement, where an observer’s attention is drawn to a location or resource (here a potential mate) by the activity of demonstrators [18,23]. Mate-choice copying requires animals to observe mating interactions directly and is typically discussed in terms of animals using public information (a subset of social information) to assess the quality of a potential mate (whose location is already known) [18,29]. Nevertheless, the consequence of social information use by L. hesperus males

(mate-search copying) is likely to be similar to that of mate-choice copying. Male movement towards a particular female will increase the rate at which subsequent males approach, and ultimately attempt to mate with that same female [29]. Widespread use of this tactic may result in some females remaining unmated, increasing the variance in female mating success ([28,30]; figure 2a,b) and altering the strength and direction of selection. This has implications for the evolution of both male and female traits and the potential for sexual selection on female signals [62].

Funding. Funding was provided by the Natural Sciences and Engineering Research Council of Canada (Discovery grant no. RGPIN-2017-06060) and the Canada Research Chairs Program (both to M.C.B.A.); C.E.S. was supported by an NSERC CGS-D during the study. The 2017 fieldwork was funded by the Toronto Entomologists’ Association Eberlie Grant (to C.E.S.), the Experiment.com Arachnid Challenge Grant and many generous individual donors to the ‘Team Black Widow’ crowdfunding campaign.

Acknowledgements. We thank the Tsawout First Nation for permission to do fieldwork on their lands. J. Carrière, M. Guarrasi, J. Hoac, Z. Kerami, V. Nguyen, B. Robinson, R. Santos, M. Swift, T. Truong and N. Wong collected the data for the first laboratory experiment. P. K. McCann and C. and D. Copley provided housing and other logistical support during fieldwork. We also thank M. Boers for help with data wrangling, J. T. Cullen for access to his microbalance and G. I. Holwell, G. S. Blackburn, D. T. Gwynne, A. C. Mason, L. Baruffaldi and Z. Luo for helpful comments on the manuscript. This study was completed in partial fulfilment of the requirements for a PhD in the Department of Ecology and Evolutionary Biology (University of Toronto).

1.5 References 1. Darwin C. 1871 The descent of man, and selection in relation to sex. London, UK: Murray. 2. Trivers RL. 1972 Parental investment and sexual selection. In BG Campbell (ed), Sexual selection & the descent of man (pp. 136 – 179). New York, NY: Aldine de Gruyter. 3. Emlen ST, Oring LW. 1977 Ecology, sexual selection, and the evolution of mating systems. Science 197, 215 – 223. 4. Thornhill R, Alcock J. 1983 The evolution of insect mating systems. Cambridge, MA: Harvard University Press. 5. Herberstein ME, Painting CJ, Holwell GI. 2017 Scramble competition polygyny in terrestrial arthropods. In Advances in the study of behavior (eds M Naguib, J Podos, LW Simmons, L Barrett, SD Healy, M Zuk), Vol. 49, pp. 237 – 295. Cambridge, MS: Elsevier. 6. Able DJ. 1999 Scramble competition selects for greater tailfin size in male red-spotted newts (Amphibia: Salamandridae). Behav. Ecol. Sociobiol. 46, 423 – 428. 7. Proctor HC. 1992 Effect of food deprivation on mate searching and spermatophore production in male water mites (Acari: Unionicolidae). Funct. Ecol. 6, 661– 665. 8. Gwynne D T. 1987 Sex-biased predation and the risky mate-locating behaviour of male tick- tock cicadas (Homoptera: Cicadidae). Anim. Behav. 35, 571 – 576. 9. Andersson MB. 1994 Sexual selection. Princeton, NJ: Princeton University Press. 10. Rank NE, Yturralde K, Dahlhoff EP. 2006 Role of contests in the scramble competition mating system of a leaf beetle. J. Insect Behav. 19, 699 – 716. 11. Barry KL, Holwell GI, Herberstein ME. 2011 A paternity advantage for speedy males? Sperm precedence patterns and female re-mating frequencies in a sexually cannibalistic praying mantid. Evol. Ecol. 25, 107 – 119. 12. Holwell GI, Allen, PJD, Goudie, F, Duckett, PE, Painting, CJ. 2016 Male density influences mate searching speed and copulation duration in millipedes (Polydesmida: Gigantowales chisholmi). Behav. Ecol. Sociobiol. 70, 1381 – 1388. 13. Cattelan, S, Evans, JP, Pilastro, A, Gasparini, C. 2016 The effect of sperm production and mate availability on patterns of alternative mating tactics in the guppy. Anim. Behav. 112, 105 – 110. 14. White DJ, Galef BG. 1999 Mate choice copying and conspecific cueing in Japanese quail, Coturnix coturnix japonica. Anim. Behav. 57, 465 – 473. 15. Schneider JM, Lucass C, Brandler W, Fromhage L. 2011 Spider males adjust mate choice but not sperm allocation to cues of a rival. Ethology 117, 970 – 978. 16. Ibáñez A, López P, Martín J. 2012 Discrimination of conspecifics’ chemicals may allow Spanish terrapins to find better partners and avoid competitors. Anim. Behav. 83, 1107 – 1113. 17. Mautz BS, Jennions MD. 2011 The effect of competitor presence and relative competitive ability on male mate choice. Behav. Ecol. 22, 769 – 775.

18. Valone TJ, Templeton JJ. 2002 Public information for the assessment of quality: a widespread social phenomenon. Phil. Trans. R. Soc. B 357, 1549 – 1557. 19. Thorogood R, Davies NB. 2016 Combining personal with social information facilitates host defences and explains why cuckoos should be secretive. Sci. Rep. 6, 19872. 20. Czaczkes TJ, Grüter C, Ellis L, Wood E, Ratnieks FL. 2012 Ant foraging on complex trails: route learning and the role of trail pheromones in Lasius niger. J. Exp. Biol. 216, 188 – 197. 21. Leadbeater E, Florent C. 2014 Foraging bumblebees do not rate social information above personal experience. Behav. Ecol. Sociobiol. 68, 1145 – 1150. 22. Rieucau G, Giraldeau LA. 2011 Exploring the costs and benefits of social information use: an appraisal of current experimental evidence. Phil. Trans. R Soc. B 366, 949 – 957. 23. Webster MM, Laland KN. 2008 Social learning strategies and predation risk: minnows copy only when using private information would be costly. Proc. R. Soc. B 275, 2869 – 2876. 24. Smolla M, Alem S, Chittka L, Shultz S. 2016 Copy-when-uncertain: bumblebees rely on social information when rewards are highly variable. Biol. Lett. 12, 20160188. 25. McNeil JN. 1991 Behavioral ecology of pheromone-mediated communication in moths and its importance in the use of pheromone traps. Ann. Rev. Entomol. 36, 407 – 430. 26. Pellegrino AC, Peñaflor MFGV, Nardi C, Bezner-Kerr W, Guglielmo CG, Bento JMS, McNeil JN. 2013 Weather forecasting by insects: modified sexual behaviour in response to atmospheric pressure changes. PloS One 8, e75004. 27. White DJ, Davies HB, Agyapong S, Seegmiller N. 2017 Nest prospecting brown-headed cowbirds ‘parasitize’ social information when the value of personal information is lacking. Proc. R. Soc. B 284, 20171083. 28. Gibson RM, Höglund J. 1992 Copying and sexual selection. Trends Ecol. Evol. 7, 229 – 232. 29. Witte K, Kniel N, Kureck IM. 2015 Mate-choice copying: status quo and where to go. Curr. Zool. 61, 1073 – 1081. 30. Wade MJ, Pruett-Jones SG. 1990 Female copying increases the variance in male mating success. Proc. Natl. Acad. Sci. USA 87, 5749 – 5753. 31. White DJ. 2004 Influences of social learning on mate-choice decisions. Anim. Learn. Behav. 32, 105 – 113. 32. Widemo MS. 2005 Male but not female pipefish copy mate choice. Behav. Ecol. 17, 255 – 259. 33. Witte K, Ryan MJ. 2002 Mate choice copying in the sailfin molly, Poecilia latipinna, in the wild. Anim. Behav. 63, 943 – 949. 34. Ferkin MH, Ferkin AC. 2017 The number of male conspecifics affects the odour preferences and the copulatory behaviour of male meadow voles, Microtus pennsylvanicus. Behaviour 154, 413 – 433. 35. Callander S, Backwell PR, Jennions MD. 2012 Context-dependent male mate choice: the effects of competitor presence and competitor size. Behav. Ecol. 23, 355 – 360.

36. Rodríguez RL, Rebar D, Fowler-Finn KD. 2013 The evolution and evolutionary consequences of social plasticity in mate preferences. Anim. Behav. 85, 1041 – 1047. 37. Scott C, Kirk D, McCann S, Gries G. 2015 Web reduction by courting male black widows renders pheromone-emitting females' webs less attractive to rival males. Anim. Behav. 107, 71 – 78. 38. Andrade MCB, MacLeod EC. Potential for CFC in black widows (genus Latrodectus): mechanisms and social context. In Cryptic female choice in arthropods—patterns, mechanisms and prospects (eds AV Peretti, A Aisenberg) pp. 27 – 53. Berlin: Springer. 39. Andrade MCB. 2003 Risky mate search and male self-sacrifice in redback spiders. Behav. Ecol. 14, 531 – 538. 40. Segev O, Ziv M, Lubin Y. 2003 The male mating system in a desert widow spider. J. Arachnol. 31, 379 – 393. 41. Segoli M, Harari AR, Lubin Y. 2006 Limited mating opportunities and male monogamy: a field study of white widow spiders, Latrodectus pallidus (Theridiidae). Anim. Behav. 72, 635 – 642. 42. Brandt Y, Andrade MCB. 2007 Testing the gravity hypothesis of sexual size dimorphism: are small males faster climbers? Functional Ecology 21, 379 – 385. 43. Foellmer MW, Fairbairn DJ. 2005 Selection on male size, leg length and condition during mate search in a sexually highly dimorphic orb-weaving spider. Oecologia 142, 653 – 662. 44. MacLeod EC, Andrade MCB. 2014 Strong, convergent male mate choice along two preference axes in field populations of black widow spiders. Anim. Behav. 89, 163 – 169. 45. Foelix R. 2011 Biology of Spiders. New York, NY: Oxford University Press 46. Scott CE, Anderson AG, Andrade MCB. 2018 A review of the mechanisms and functional roles of male silk use in spider courtship and mating. J. Arachnol. 46, 173 – 207. 47. Wade MJ, Arnold SJ. 1980 The intensity of sexual selection in relation to male sexual behaviour, female choice, and sperm precedence. Anim. Behav. 28, 446 – 461. 48. Janicke T, Morrow EH. 2018 Operational sex ratio predicts the opportunity and direction of sexual selection across animals. Ecol. Lett. 21, 384–391. 49. Weir LK, Grant JW, Hutchings JA. 2011 The influence of operational sex ratio on the intensity of competition for mates. Am. Nat. 177, 167 – 176. 50. Salomon M, Vibert S, Bennett RG. 2010 Habitat use by western black widow spiders (Latrodectus hesperus) in coastal British Columbia: evidence of facultative group living. Can. J. Zool. 88, 334 – 346. 51. de Jong K, Forsgren E, Sandvik H, Amundsen T. 2012 Measuring mating competition correctly: available evidence supports operational sex ratio theory. Behav. Ecol., 23, 1170 – 1177. 52. Prenter J, Pérez‐Staples D, Taylor PW. 2010 The effects of morphology and substrate diameter on climbing and locomotor performance in male spiders. Funct. Ecol. 24, 400 – 408.

53. R Core Team 2015 R: a language and environment for statistical computing (R Foundation for Statistical Computing, Vienna). Available at http://www.R-project.org 54. Bossert WH, Wilson EO. 1963 The analysis of olfactory communication among animals. J. Theor. Biol. 5, 443 – 469. 55. Stoltz JA, Andrade MCB. 2010 Female's courtship threshold allows intruding males to mate with reduced effort. Proc. R. Soc. B 277, 585 – 592. 56. Fletcher, RJ, Miller, CW. 2008 The type and timing of social information alters offspring production. Biol. Lett. 4, 482 – 485. 57. Loyau A, Blanchet S, Van Laere P, Clobert J, Danchin E. 2012 When not to copy: female fruit flies use sophisticated public information to avoid mated males. Sci. Rep. 2, 768. 58. Baruffaldi L, Costa FG, Rodríguez A, González A. 2010 Chemical communication in Schizocosa malitiosa: evidence of a female contact sex pheromone and persistence in the field. J. Chem. Ecol. 36, 759 – 767. 59. Dechaume-Moncharmont FX, Brom T, Cézilly F. 2016 Opportunity costs resulting from scramble competition within the choosy sex severely impair mate choosiness. Anim. Behav. 114, 249 – 260. 60. Duvall D, Schuett GW, Arnold SJ. 1993 Ecology and evolution of snake mating systems. In Snakes: Ecology and Behavior (eds RA Seigel, JT Collings), (pp. 165 – 200). New York, NY: McGraw Hill. 61. Rios-Cardenas O, Morris MR. 2009 Mating systems and strategies of tropical fishes. Trop. Biol. Conserv. 8, 219 – 240. 62. Edward DA, Chapman T. 2011 The evolution and significance of male mate choice. Trends Ecol. Evol. 26, 647 – 654. 63. Scott C, McCann S, Andrade M. Data from: Male black widows parasitize mate-searching effort of rivals to find females faster. Dryad Digital Repository. https://doi.org/10.5061/dryad.cv66j09

Figure 1-1 Demographic data for a natural population of L. hesperus at a coastal field site on Vancouver Island, British Columbia, Canada (48°3501000 N, 123°2201700 W). The population was assessed in regular nocturnal surveys (individual data points) and full censuses (arrows) over the 2016 late summer mating season. We recorded the number of (a) adult males, (b) sexually receptive females, and estimated (c) the OSR (ratio of males to receptive females; see electronic supplementary material, S1 for calculation details).

2016 2017 30

p a b

a 25

20 s

e 15 l

a 10

# 5 0 8404 8 16 808 16 WEWE distance from male release transect (m)

c N d N 2 1 1 5 5 k k 1 m 1 m 3 4 0 / 0 / k h k h m 2 m 5 1 5 1 k /h k /h m 5 m /h /h 4 6 3

hourly wind vector during experiment

Figure 1-2 Patterns of arrival at female pheromone traps under differing wind conditions in field tests of mate searching by L. hesperus males. (a,b) Number of males attracted by each caged female (black bars) relative to the male-release transect (vertical dashed lines). (c,d) Wind speed and direction during the experiment. Numbered arrows represent the wind velocity over each hour (spiders stopped arriving after 4 h in 2016 and 6 h in 2017).

Figure 1-3 Latrodectus hesperus males detect and follow the draglines of rival males in the laboratory on an X-maze choice arena, and the presence of this species-specific social information increases their speed when personal information is available simultaneously. (a) In the X-maze choice assay, the male released first (labelled ♂1) is attracted to a caged pheromone-emitting female (arrows indicate wind produced by electric fans), with choice of route determined by the male. A second male (♂2) is released after removal of the first male. The second male can choose to follow or avoid the silk of the first male. (b–f) The number of L. hesperus males choosing to follow silk of conspecific males (providing social information) or heterospecific (S. grossa) males over controls. The presence of a female (providing personal information) and/or a fan producing wind is indicated by a picture of a caged female and/or grey dashed arrows, respectively. Significant differences (binomial tests; p < 0.05) are indicated by asterisks (*). (g,h) The relationship between male phenotype (leg length and body condition index) and male speed after reaching the intersection of the X-shaped choice apparatus (decision point) during the experiment shown in (b). Grey-filled circles and solid lines: males with access to silk cues of a rival (= social information); empty circles and dotted lines: males without access to silk cues.

Figure 1-4 Distance from females and the size of L. hesperus males affect mate-searching success and speed in the field. Graphs show relationships between distance from pheromone-emitting females and (a,b) recapture rate, and (c,d) average speed of recaptured males. Lines are predicted fits (solid) and approximate 95% confidence intervals (grey dashed) from general linear models of the data. (a,c) Results from the 2016 replicate when the wind was strong and relatively consistent in direction (figure 1b). (b,d) Results from the 2017 replicate when the wind was relatively weak and highly variable in direction (figure 1d). (c,d) Speed was log-transformed for analysis with back-transformed predicted fits displayed on the raw data.

This chapter was published in Proceedings of the Royal Society B (Scott CE, McCann S, Andrade MCB. 2019 Male black widows parasitize mate-searching effort of rivals to find females faster. Proc. R. Soc. B 286: 20191470. http://dx.doi.org/10.1098/rspb.2019.1470) and is reprinted with permission of the journal.

Spatial clustering of females facilitates immature mating, an extreme reproductive tactic of male black widow spiders

Catherine E Scott, Sean McCann, and Maydianne CB Andrade

Abstract

The spatial distribution of females and the timing of their sexual receptivity are important factors shaping the form of male competition and the intensity of sexual selection on males. When competition is intense and receptive females are rare or difficult to find, selection may favour pre-copulatory mate guarding, particularly under strong first-male sperm precedence. We investigated the frequency and ecological correlates of male cohabitation with subadult females in a natural population of western black widows (Latrodectus hesperus). Immature mating, an extreme alternative reproductive tactic by which males can inseminate subadult females just prior to their moult to maturity was recently documented in the laboratory. Here we show that mate guarding and immature mating are important components of the natural mating system of male L. hesperus, and that despite the apparent absence of direct cues that would attract males to subadult females, spatial clumping of subadults with adults facilitates this tactic. Searching males only found subadult females when they were clustered near pheromone-emitting adults in a field experiment, and male presence increased with the number of conspecific females under shared microhabitats in surveys of a natural population. Since immature mating requires mate guarding, and constrains female choice relative to adult mating, this is likely to have important implications for sexual selection on males in nature.

2.1 Introduction

Sexual selection arises from variation in mating success resulting from competition over mates or fertilizations (Darwin 1871). The extent to which one sex limits the reproduction of the other can be quantified in the operational sex ratio (OSR), the ratio of adult males to fertilizable or receptive females (Emlen and Oring 1977). In many animal mating systems females are the sex in short supply, so males compete for mating opportunities and their sperm compete for access to eggs (Andersson 1994). The spatial distribution of females and the timing of their receptivity are key factors influencing the form of competition and the intensity of sexual selection on males

36 37

(Emlen and Oring 1977; Ims 1988; Clutton-Brock and Parker 1992). When female receptivity is asynchronous and transient, there may be strong selection on males to detect and find females just before or after their maturation, prior to rivals (Grafen and Ridley 1983; Harts and Kokko 2013). In such systems, male traits associated with efficient detection and localization of females will be favoured (Parker 1978; Thornhill and Alcock 1983; Herberstein et al. 2017). When this is the case, and the OSR is also male-biased, then males that encounter females who are not currently receptive may also engage in precopulatory mate guarding (Weir et al. 2011). Precopulatory mate guarding is expected if waiting for a female to become receptive and deterring other suitors leads to a higher expected fitness payoff than continuing to search for other females (Parker 1974a). This is likely to occur if guarding ensures paternity in systems where male-male competition is intense and receptive females are rare or hard to find (Prenter et al. 2003), particularly if there is first-male sperm precedence (Parker 1970; Elgar 1998; Simmons 2001).

When the first male to mate has an advantage in sperm competition, males are expected to preferentially seek unmated females (Dodson and Beck 1993; Elgar 1998; Simmons 2001). In a range of invertebrate taxa, males ensure their sperm has priority by guarding immature females and mating with them as soon as they become sexually mature (Thornhill and Alcock 1983; Ridley 1989; Schröder 2003). In some insects, males mate with newly-matured adult females either just prior to or during emergence from their pupal case (ants: Foitzik et al. 2002; Lepidoptera: Provost and Haeger 1967; Deinert et al. 1994; Estrada et al. 2010). Similarly, cohabitation with subadult females (one instar prior to adulthood) is common in spiders, with males copulating during or immediately following moulting (Elgar 1998; Uhl 2002; Tuni and Berger-Tal 2012; Uhl et al. 2015). In a more extreme version of these early-mating tactics, males of at least three spider species in the genus Latrodectus can inseminate immature females just prior to their moult to maturity by piercing the exoskeleton covering the developed genitalia (Biaggio et al. 2016; Baruffaldi and Andrade, in review). The frequency of this alternative to mating adult females will be limited by encounter rates with subadults, which may increase in high density populations, or if males are able to detect and locate females that are not yet mature. It is not clear whether the latter is possible. Theory predicts that immature females should not invest in mate attraction and should avoid signaling their presence to prevent costly harassment before they are ready to mate (Ridley 1989). However, in crustaceans and insects with

38 precopulatory mate guarding, males are able to detect chemical cues indicating that a female is about to moult (Jormalainan 1998; Weeks and Benvenuto 2008; Estrada et al. 2010) and in spiders there is some evidence for chemical cues produced by subadults (e.g. Ross and Smith 1979; Jackson 1986; Gaskett 2007) and recently-moulted adult females (e.g. Miyashita and Hayashi 1996).

Here we focus on how ecological context may affect the opportunity for males to engage in immature mating in a natural population of black widow spiders (Latrodectus hesperus). Male widow spiders (genus Latrodectus) engage in scramble competition over access to sexually receptive adult females (Andrade and MacLeod 2015; Scott et al. 2015a, Scott et al. 2019 [Chapter 1]). However, males may also mate with immature females, leading to similar paternity gains as mating with adults (Biaggio et al. 2016; Baruffaldi and Andrade 2017; Baruffaldi and Andrade in review). Initially reported in the otherwise monogynous Australian redbacks (L. hasselti) and brown widow spiders (L. geometricus), immature mating may allow multiple mating by males, which is otherwise prevented by the high frequency of sexual cannibalism when males mate adult virgin females in these species (Biaggio et al. 2016). This tactic has also been documented in the polygynous western black widow (L. hesperus; Baruffaldi and Andrade in review), in which sexual cannibalism is less common (Andrade and MacLeod 2015). In this species, there is strong first-male sperm precedence (MacLeod 2013) and intense competition for access to receptive adult females whose sexual signals are only availably briefly and asynchronously (Scott et al. 2015a; Scott et al. 2019 [Chapter 1]). In the population we studied, the OSR is strongly male biased and mate searching is risky, with only 12% of males surviving movements between webs (Scott et al. 2019 [Chapter 1]). These factors should favour mating with subadult females, provided males have the opportunity to interact with them. However, we do not yet know the frequency of immature mating in nature nor the factors that affect the feasibility of this tactic.

Both tactics available to widow males (adult mating and immature mating) require encountering a potential mating partner. The opportunity to pursue each mating tactic is likely to be related to the spatial and temporal distribution of females, and female detectability, often mediated by sex pheromone production in spiders (Elias et al 2011). Previous work shows that unmated adult females produce airborne sex pheromones, that males scramble to rapidly locate these signaling females, and likely mate soon after finding them (MacLeod and Andrade 2014;

Scott et al. 2015a, 2019 [Chapter 1]). The situation is different for males that encounter subadult females. First, since subadult females are not likely to actively produce sexual signals (Anava and Lubin 1993; Waner et al. 2018), it is unclear what determines encounter rates. Second, since mating is only possible during the few days preceding the female’s moult to maturity (Biaggio et l. 2016; Baruffaldi and Andrade 2015, Baruffaldi and Andrade in review), males that find immature females may cohabit with the female for several days before attempting mating. During this period they may need to deter challenges from rival males (mate guarding; Alcock 1991; Jormalainan 1998; Prenter et al. 2003; Bel-Venner and Venner 2006). Understanding the traits favoured by these alternative tactics and how they may shape sexual selection on male widows requires first obtaining information about the frequency and determinants of each tactic in nature.

Immature mating has only been reported in the laboratory, so our first goal was to determine whether it occurs in nature. We collected observational data on natural interactions between adult males and subadult and mature females across a full season in the field. Using these data, we sought to (1) estimate the frequency of opportunities for immature mating relative to adult mating, (2) obtain evidence for mate guarding, and (3) ask how these male tactics are related to demography, phenology, and the spatial distribution of potential mates. We also experimentally investigated (4) whether there is evidence for male attraction to subadult females via chemical cues, since chemical profiles of females may change as they undergo the physiological processes that precede moulting (Kittredge and Takahashi 1972; Jormalainan 1998; Weeks and Benvenuto 2008; Estrada et al. 2010). Although we tested this possibility, we hypothesized that a more critical source of information for males would be volatile chemical signals or cues produced by adult females that may be in close proximity to subadults (Kasumovic and Andrade 2004; MacLeod and Andrade 2014; Scott et al. 2015a). The clumped distribution of spiders at this site may allow males to encounter subadult females by moving towards pheromone-producing adult females. Once at an occupied microhabitat, males may be able to recognize subadult females using contact cues associated with their silk (Ross and Smith 1979; but see Anava and Lubin 1993). We tested a key prediction of this hypothesis in field experiments: that (5) males will be more likely to find subadults when they are clustered near adult females than when they are isolated. Together these data will provide a foundation for

40 understanding links between demography, spatial dynamics, chemical information, and the frequency of an extreme male mating tactic.

2.2 Methods and Results

2.2.1 Natural History

We studied a population of Latrodectus hesperus Chamberlin and Ivie, 1935 inhabiting the coastal sand dunes above the high-tide line at Island View Beach and T̸IX̱ EṈ (Cordova Spit) on the Saanich Peninsula of Vancouver Island, British Columbia, Canada (48°35’10”N, 123°22’17”W). Driftwood logs, other woody debris, and occasional rocks provide microhabitats for western black widows at this site (Fig. 1b). Here, these spiders typically spin silk retreats in sheltered areas under logs or other objects, and their three-dimensional capture webs extend onto the sand and nearby vegetation. The population is dense, with 2-3 subadult or adult females per square metre of microhabitat in late summer, such that most microhabitats are shared by multiple conspecific females and juveniles (Fig. 1c; see Salomon et al. 2010 for more details about this site and population). Adult females are present year-round, and produce egg sacs from May until July (the active season typically runs from April to October). Both males and females can become sexually mature the same year they emerge from egg sacs, or they can overwinter as subadults and mature the following spring (Salomon et al. 2010). Adult females can also overwinter and produce egg sacs the following spring and summer (females can live up to two years; Salomon et al. 2010).

Unmated female western black widows produce silk-bound sex pheromones that attract males at long range (Kasumovic and Andrade 2004; MacLeod and Andrade 2014; Scott et al 2019) and elicit male courtship behaviour upon contact (Ross and Smith 1979; Scott et al. 2015b). Males can discriminate between adult females of different mating status and feeding history based on these volatile and contact pheromones (MacLeod and Andrade 2014; Baruffaldi and Andrade 2015). At mating, males lose the tips (apical sclerites) of their paired copulatory organs (emboli), and these can function as sperm plugs when correctly placed in the female’s sperm storage organs (spermathecae; Snow et al. 2006, Macleod 2013). This results in strong first-male sperm precedence (Andrade and MacLeod 2015). Nevertheless, both males and females can mate multiply (Macleod 2013).

Recently, Baruffaldi and Andrade (in review) reported that L. hesperus males are capable of mating with subadult females just before their moult to maturity. The female’s genitalia and sperm storage organs are fully formed 2-4 days before moulting and the male is able to pierce the exoskeleton (presumably with his chelicerae) and successfully copulate with her. The stored sperm are retained and used to produce fertile egg sacs after the moult to maturity. The frequency of this behaviour in nature is unclear.

2.2.2 Observational Data: Methods 2.2.2.1 Data collection

2.2.2.1.1 General survey methods

From April to September 2016 we surveyed the population of Latrodectus hesperus inhabiting a ca. 20 x 200 m area of our study site on the Saanich Peninsula of Vancouver Island in British Columbia, Canada (48.586890, -123.371138). Over a series of consecutive days each month, we turned over every driftwood log and rock within this area to perform a systematic population census. We recorded the number, sex, and age class of spiders (determined by inspection; Kaston 1970) under each microhabitat, and all spiders that were within two moults of maturity were briefly collected, immobilized between a soft sponge and a mesh cloth, and marked with a unique colour code using quick-drying modelling paint (Testor’s enamel) before being returned to their webs (where they remained despite handling, see Salomon 2008). We also performed more frequent nocturnal surveys (each point in Fig. 1 represents a survey date) that involved non-intrusive inspection of the webs of all marked spiders to record presence/absence, evidence of moulting, and presence of males on webs of females. Spiders that had moulted shed their marks with their moult skins (which remained visible in the web). These spiders were marked again and returned to their webs. At the end of the season we assessed spatial characteristics of the population by mapping all microhabitats (logs and rocks) within the field site that were ever occupied by spiders, and calculating minimum distances between them.

2.2.2.1.2 Assessing mating status and maturity of cohabiting males and females

We collected data on male visits to females by recording male presence on the webs of subadult and adult females during surveys and censuses (distances traveled by searching males using this

42 dataset are reported in Scott et al. 2019 [Chapter 1]). Male maturity can be assessed by the presence of swelled, translucent pedipalps (subadult), or dark pedipalps with developed emboli (adult). Subadult females (in their penultimate instar) can be distinguished by their size and a prominent bulge at the anterior-ventral abdomen covering the developing genitalia. After their final moult, adult females can be distinguished by the sclerotized, open structure (epigynum) found in this same location (Kaston 1970).

Data on males included (i) males first marked as subadults who later moulted to maturity, were remarked, and then later found on the webs of one or more females, (ii) males first marked as adults on the web of a female, and then (in some cases) later found on the webs of one or more additional females. The former category includes males’ first mate searching attempt, while the latter likely includes males with previous mate-searching experience. Females in this dataset included (i) those first marked as subadults who later moulted to maturity and were remarked, (ii) females marked as subadults who disappeared (died or moved out of the study area), and (iii) females first marked as adults.

2.2.2.1.3 Timing and frequency of cohabitation and immature mating

We collected data on the timing of male visits for the subset of females for whom we had complete records (first marked as immatures in or prior to the penultimate instar, moulted to adulthood, and then remarked; n = 78). For these females, we looked for evidence of immature mating by noting individuals who had been marked as subadults (with no epigynal opening) who were later observed either (i) still marked (and thus not yet moulted), but with a visibly open epigynum or (ii) still marked and in copula, and who subsequently moulted to maturity.

2.2.2.1.4 Male encounter rates with adult vs subadult females

For males (n = 324), we recorded the number of females that they visited in their lifetimes as a function of female developmental stage (Fig. 3). For 78 of these females we obtained complete records (observations throughout the subadult and adult stages), and recorded the number of males they encountered before and after their moult to maturity. These should be considered minimum estimates of potential mates encountered, since it is possible that we did not observe every encounter. First, males cohabiting with subadult females are often hidden in the retreat with the female, so may not have been visible to us during our regular surveys. Second, we may

43 have missed some male visits to females’ webs on nights when there were no surveys. However, we often observed the same marked males present on subadults’ webs on several sequential nights, and movements between webs are unlikely to occur during the day (L. hesperus is nocturnal), suggesting that that we were unlikely to miss visits. Moreover, since in other Latrodectus species, males cohabit with adult and subadult females for similar lengths of time, there is unlikely to be a bias in observing each type of cohabitation (Andrade and Kasumovic 2005; Segoli et al. 2006).

2.2.2.1.5 Evidence for mate guarding

Mate guarding implies that cohabiting males will encounter rivals on the female’s web and engage in direct aggressive interactions during that time (e.g. Alcock 1991; Bel-Venner & Venner 2006). Due to the complex nature of female’s webs and the transient nature of these behavioural interactions (males are expected to depart after losing a fight; Grafen & Ridley 1983; Bel-Venner & Venner 2006), systematic quantification of these variables was challenging. However, we make a conservative estimate of the minimum frequency of guarding based on the proportion of females for which we have complete records (n = 78) seen with more than one male present in the web simultaneously at any point during the season. We also provide qualitative descriptions of inter-male aggressive interactions and male interactions with subadult females that were observed during our surveys.

2.2.2.2 Data analysis

2.2.2.2.1 Phenology and operational sex ratio

To describe phenology and operational sex ratio we divided females that were potentially available for mating into the following classes: (i) subadult (females at any point in their penultimate instar), (ii) recently matured (females within 7 days after the date we noted their moult to maturity), and (iii) other adult females (females more than 7 days post moult or females first found as adults) (Figure 3). Because males typically arrive at experimental cages containing unmated adult females within a few hours (Scott et al. 2015a; Scott et al. 2019 [Chapter 1]), seven days is likely a conservative overestimate of the time that recently matured females remain unmated and actively signalling at this site. For our estimates of cumulative abundance of spiders in each class (Fig. 1) if we were not able to confirm a previously marked spider’s presence or absence during regular surveys, we assumed that it remained alive until we confirmed its

44 presence or absence at the next population census. The result is that if these spiders died or left the population between censuses, we recorded them as doing so on the same date (just preceding the census), and thus the error in our cumulative abundance estimates is biased. However, it is biased in the same way for all classes of spiders and thus should not compromise comparisons between them.

We divided the mating season data into two periods for analysis, ‘early season’ (April 24 – June 28) and ‘late season’ (July 19 – September 30) since these correspond to two cohorts of males at this site (Figure 3). The early season males are those that emerge from egg sacs in the summer, overwinter in the penultimate instar and moult to maturity the following spring. The late season males include those that mature in the summer after emerging from egg sacs produced earlier the same year (Salomon et al. 2010).

2.2.2.2.2 Timing and frequency of cohabitation and immature mating

We used Chi-squared tests to assess whether males were found disproportionately with subadult or with adult females. We compared the observed number of males cohabiting with subadult and adult females to the number of females in each age class that were available during the corresponding period of the mating season (null expectation). For these analyses we combined all adult females because we were interested in testing the hypothesis that males were more likely to visit subadult females (compared to any other class of fertilizable female) than expected by chance.

2.2.2.2.3 Male encounter rates with adult vs subadult females

We used Chi-squared tests and Fisher’s exact tests to assess whether male patterns of cohabitation with subadult females changed over the season as this should affect the opportunity for immature mating. We compared: (i) the proportion of males we observed only with subadult females (never with adult females) in the early season relative to the late season, and (ii) the proportion of females visited only before maturity (never observed with a male while an adult) in the early season relative to the late season.

2.2.2.2.4 Effect of spatial distribution and demography on cohabitation with subadults

We used a generalized linear mixed model (glmm) to assess whether male cohabitation with subadult females varied as a function of distance to the nearest neighbouring microhabitat and the presence of cohabiting females of various age classes. This model had a binary response variable (one or more males present, or no males present) and the predictors were: the number of subadult females, the number of recently matured females (likely to be producing sex pheromone), and the number of older adult females (likely to be already mated). Approximate distance between each microhabitat and its nearest neighbor was also included to determine whether microhabitat location, in addition to female presence under individual microhabitats, predicted male presence. Because we included data for each log on each day of the active season, we included log identity and date as random effects in the model. All statistical tests were conducted using RStudio version 1.2.1335.

2.2.3 Observational Data: Results 2.2.3.1 Phenology and operational sex ratio

The relative abundance of recently matured and older adult females was generally consistent across the mating season (Fig. 3), indicating that the timing of female receptivity is highly asynchronous in this population. Both males and subadult females were more abundant during the late mating season than the early season (Fig. 3). In the early season, subadult females never comprised more than 8% of the total female population (i.e., all penultimate-instar and adult females). In the late season, up to 44.8% of females were subadults.

2.2.3.2 Timing and frequency of cohabitation and immature mating

Of 78 females that we followed from their penultimate instar through to adulthood, at least 60% of them were visited by at least one male (Fig. 6). Of the 47 females that were visited across both periods in the season, 79% (37) were visited only while immature, and at least 21% (10) of these mated before moulting to maturity. Five females (11%) were only visited after their moult to maturity and five females (11%) were visited both before and after their moult to maturity. Females were very likely to be visited only prior to maturity in the late season; of 41 visited by at least one male, 85% (35) of them were only visited as subadults. However, this was much less

46 common in the early season; of 6 females visited by at least one male, 33% (2) of were visited only as subadults (Fisher’s exact test, P = 0.0095).

2.2.3.3 Male encounter rates with adult vs subadult females

Of the 324 adult males observed throughout the season, 75% encountered at least one female, but few males were found with more than one female (Fig. 5). Of the 244 males who found at least one female, most (69%) were only ever found with subadult females, 27% found only adult females, and 4% found at least one of each (Fig. 5). Males were more likely to visit only subadult females in the late season (73% (155) of the 211 males that encountered at least one female) compared to the early season (42% (14) of 33 males; 2 = 11.5; df = 1; P = 0.0007). However, in both the early season and the late season, male visits to subadult females did not simply track their abundance relative to adult females. For males that found at least one female, the likelihood that they only ever encountered subadults was higher than predicted by chance (Table 1).

2.2.3.4 Effect of spatial distribution and demography on cohabitation with subadults

A glmm (Table 2) showed that the number of subadult females, recently matured females, and older adult females sharing a microhabitat were significant predictors of the presence of cohabiting males. Nearest neighbor distance (distance to the nearest microhabitat that was inhabited by conspecifics at any time during the mating season), did not significantly predict male presence. The odds ratios (Table 2) indicate that the presence of subadult or recently matured females increases the likelihood of male cohabition, much more so than does the presence of older adult females. For every additional subadult female under a microhabitat, the odds of male presence increased by 127%. For every additional recently matured female, the odds of male presence increased by 170%, and for every additional older mature female, the odds of male presence increase by 19%.

2.2.3.5 Evidence for mate guarding

Quantitative data. We examined cohabitation patterns for the 78 females for which complete records were available. In the early season we never observed more than one male simultaneously cohabiting with a female. In the late season, when population density was higher

(Fig. 3) cohabitation was more common. Forty-one subadult females had males on their web during at least one survey, and of these, 22% (9) had more than one male cohabiting simultaneously (two males: 6; three males: 3).

Qualitative data. Across all of our surveys (n = 198 subadult females total), we frequently witnessed multiple males cohabiting simultaneously on the webs of subadult females, although these are not included in the estimates above. Typically, these males remained still at the periphery of the web during our observations. However, we also witnessed pairs of males engaged in combat on ten occasions (Fig. 2a). Fighting males chased, attempted to bite, and flung silk at their opponents, behaviours that are distinct from those during sexual interactions with females. Courting males always approach females slowly, and do not open their chelicerae as they approach or attempt to bite their legs as we observed males doing during combat. Though courting males do wrap females with silk (the “bridal veil”; Ross & Smith 1986; Scott et al. 2018 [Chapter 4]), this fine silk is applied while the female is motionless and alternates with other courtship behaviours like abdomen vibration (Scott et al. 2012). Instead, fighting males flung silk at their active opponents in the same way they would during prey capture (widow spiders first subdue prey with sticky, glue-covered silk before biting it) or defence, though adult males of L. hesperus apparently lack the ability to make sticky silk (Vetter 1980). For two subadult females, we observed contests between a resident male with an intruding male on one night, followed by a fight with a new non-resident male on a subsequent night. Moreover, we recorded different males cohabiting with subadult females sequentially, suggesting that usurpation had taken place (six observations). We also observed a subadult female cannibalizing one of two males present simultaneously on her web; we infer that male combat resulted in the female attacking and killing one of the males (Fig. 2b). During monthly censuses we often found a single male in the retreat of a subadult female that was almost ready to moult (Fig. 2c).

2.2.4 Experimental Data: Methods

Our observational data suggested that males disproportionately cohabit with subadult females, that many females are only visited when immature (particularly in the late season), and that many males visit only subadult females, with the probability of such a visit increased by clustering near to other immature or young adult females. We designed an experiment to test

48 directly whether the spatial distribution of females affects male success at localizing subadult females.

2.2.4.1 Data collection

We ran a mate attraction experiment over two nights during the late season (August 29 and 30, 2016) at our field site to see whether web clumping affected mate attraction to the webs of subadult females. Latrodectus hesperus females were placed individually in mesh screen cages and allowed to build webs for one week, creating ‘pheromone traps’ (as in Macleod and Andrade 2014; see Scott et al. 2015a for details of cage design). At dusk, we placed fifteen blocks of pheromone traps at 20-m intervals along a 280-m transect, with each block containing a trio of unmated females at different stages (“young” adult females: 9-12 days post-maturity; “old” adult females: 2-5 months post-maturity; and subadult females in their penultimate instar). All females were lab-reared offspring of females collected from our field site in 2015 (see Scott et al. 2019 [Chapter 1] for rearing and collection details).

On the first night of the experiment, trios of cages were uniformly distributed (10-m spacing) and on the second they were clumped (1-m spacing; see Fig. 4). The location of the three types of females was randomized within each block. We checked cages every hour from sunset to sunrise, collecting any males within 10 cm of the cage perimeter. Because males can detect female pheromone from up to 60 m away (Scott et al. 2019 [Chapter 1]) and all nearby mate-searching males are typically caught within 12 hours during this kind of experiment (Scott et al. 2015a), we used two different areas of our field site (separated by 500 m) for the two trials.

2.2.4.2 Data analysis

We first ran a G-test of independence to ask whether there was a difference in the proportion of males visiting each class of female under the two experimental distributions. We then ran a second G-test of independence comparing the proportion of males visiting subadult females to the combined proportion visiting adults, to determine whether distribution has an effect on whether males find subadult females in particular.

2.2.5 Experimental Data: Results

When females were uniformly distributed (10 m between traps; Fig. 4a), males either strongly preferred unmated adult females or did not detect subadult females; we captured a total of 38 males and all of them were found with adult females (Table 5; Fig. 7). When females were clumped (1 m between traps; Fig. 4b), we captured 98 males in total, and seven (7%) of them were found with subadult females. There was a significant effect of distribution on the proportion of males visiting each class of female (G = 6.05, df = 2, P = 0.048) and males were more likely to find subadult females when they were clustered near unmated adult females (G = 5.0, df = 1, P = 0.025).

2.3 Discussion

Here we show that immature mating is an important part of the mating system of male Latrodectus hesperus, and that despite the apparent absence of direct cues that would attract males to subadult females, spatial clumping of subadults with adults facilitates this alternative mating tactic. In a natural population, we estimate that more than 2/3 of males encountered only subadults in their lifetime. The opportunity for immature mating varies across the mating season, with the frequency of males guarding subadult females increasing in the late season, and to a greater extent than predicted by their relative abundance alone (Fig. 3; Table 1). We observed direct evidence for immature mating in 21% of guarded females, but estimate the frequency of immature mating will more closely match the proportion of females only seen with cohabitants as subadults (e.g., Fig. 6). Thus, our data suggest that the opportunity for immature mating is high and that this tactic may even be more common than mating with adult females at our field site. Our data also indicate that the spatial distribution of females is particularly important in driving the feasibility of immature mating, as searching males only found subadult females when they were clustered near pheromone-emitting adults in our experiment (Fig. 7), and male presence increased with the number of conspecific females under shared microhabitats in our surveys (Table 2). Since immature mating requires mate guarding, and constrains female choice relative to adult mating (Biaggio et al. 2016; Baruffaldi and Andrade 2017), this is likely to have important implications for sexual selection on males in nature.

We have demonstrated that L. hesperus males in a natural population frequently engage in an alternative reproductive tactic—immature mating—that may select for different traits than

50 the conventional tactic of finding, courting, and mating with receptive adult females. We directly witnessed interactions between competing males on the webs of subadult females in the field and also saw evidence for usurpation of guarded females. Males engaged in combat on the periphery of webs, and resident males that managed to exclude rivals eventually cohabited near or in contact with females in their retreats (Fig. 2c), where they would have the opportunity to mate just prior to the female’s moult to maturity. Whereas scramble competition over access to receptive adult females favours male traits associated with speed and mobility (Parker 1978; Thornhill and Alcock 1983; Brandt and Andrade 2007; Kelly et al. 2008; Herberstein et al. 2017), guarding subadults should favour traits associated with resource holding potential, such as large size or high condition (Parker 1974b; Davies and Halliday 1978; Clutton-Brock and Albon 1979). Larger, heavier males may also be preferred by L. hesperus females (MacLeod 2015). Consequently, even if subadult mating constrains female choice relative to adult mating (Biaggio et al. 2016; Baruffaldi and Andrade 2017), it need not impose fitness costs by preventing females from mating males of preferred traits if these same traits confer an advantage in contest competition. Nevertheless, males who engage in guarding and subadult mating may not face the same selective pressures as those who scramble to mate with adult females. Searching for signaling females imposes a selective filter on male phenotypes: larger males are most likely to reach a female’s web, but of these males, the smallest ones are fastest, and may thus reach females’ webs, and mate, before larger rivals (Scott et al. 2019 [Chapter 1]). Male widow spiders lack specialized weapons but vary dramatically in body size and mass (i.e., male mass can vary by more than an order of magnitude within field populations: Andrade and MacLeod 2015; Andrade 2019). Further work is needed to evaluate which traits are favoured for males engaging in each tactic and across episodes of selection (beginning with searching), but we speculate that the persistence of both mating tactics may help to explain the maintenance of extreme variation in male size in Latrodectus spp. (Andrade and MacLeod 2015; Andrade 2019).

Encounter rates and the ability of males to detect subadult females are key factors that will affect the feasibility of guarding and mating with immature females in nature. Our demographic data (Fig. 3) indicate that the abundance of adult males tracks that of subadult females, but that subadult females consistently make up less than half of the total pool of potentially fertilizable females (i.e., subadults, unmated adults, and mated adults) in the population. Nevertheless, males were much more likely to be found on the webs of subadults

51 than expected if encounter rates were based on their relative abundance (Table 1). This suggests that males are not randomly encountering subadults, but actively seeking them out. The strongest predictor of male presence at microhabitats containing at least one subadult female was the presence of recently-matured females that are likely to be actively producing volatile sex pheromone (Table 2, Kasumovic and Andrade 2004; MacLeod and Andrade 2014; Scott et al. 2015a). Our experiment confirmed that males are able to find subadult females when they are in close proximity to signaling adults, whereas they never found isolated subadults (Fig. 7). Together, these data support the idea that males navigate to occupied microhabitats using chemical signals produced by adult females. This also raises the possibility that unattractive subadult females benefit from positioning themselves near actively signaling adults (Holdcraft et al. 2016; van Wijk et al. 2017). Since immature mating does not appear to impose fecundity or longevity costs on females (Baruffaldi and Andrade 2017), and producing sex pheromone may be physiologically expensive (Johansson et al. 2005; Harari et al. 2011) and/or increase predation risk (Magnhagen 1991; Zuk & Kolloru 1998; Wyatt 2003), subadult females may seek microhabitats containing adults to increase the likelihood of being guarded and mated prior to maturity. We have observed subadult females moving toward pheromone traps containing adult females in nature (CES and SM pers. obs.) but whether these females are seeking mating opportunities or simply suitable microhabitats requires further investigation.

The high rate of male cohabitation with subadult females that we report here is likely driven by the spatial distribution of spiders in this population. Our estimates (69% of males encountered only subadults; 79% of females visited only as subadults; Figs. 5 and 6) are higher than for related species, in which males are more often found on the webs of adult females. In L. revivensis and L. hasselti, 13% and 21% of females found with cohabiting males were subadults, respectively (Lubin et al. 1991; Biaggio et al. 2016) and only 35% of L. hasselti females collected in their penultimate instar were later confirmed to have been mated as immatures (Biaggio et al. 2016). Similarly, in L. pallidus, only 43% of females visited by males were subadults (Segoli et al. 2006). To our knowledge, the extent to which female spiders share microhabitats at our study site (Salomon et al. 2010) is unusual if not unique among species in this genus, and we hypothesize that this feature of their ecology facilitates guarding and mating with immature females. We expect that for other populations of Latrodectus hesperus with a less clumped female distribution, this tactic will be infrequent. Consistent with this, in a population

52 of L. hesperus in California where females do not share microhabitats (see MacLeod 2013), 95% of females collected as adults late in the season were unmated, indicating that a low rate of immature mating is the rule (S Fry, pers comm).

When the expected reproductive success resulting from mate guarding is higher than that resulting from mate searching, selection favours guarding (Parker 1974a). If a male encounters a subadult, and particularly if she is close to her moult to maturity, remaining with her may have a higher expected payoff than searching for adult females that are ready to mate. This seems likely to be true for males in this population (and in Latrodectus generally) for whom mate searching is extremely risky (>80% mortality during movements between microhabitats; Andrade 2003; Segev et al. 2003; Segoli et al. 2006; Scott et al. 2019 [Chapter 1]). The benefit of guarding a subadult should increase with proximity in time to her moult to maturity, and selection should favour the ability of males to detect this timing (Birkhead & Clarkson 1980; Jormalainan 1998; Estrada et al. 2010). We did not know the age of subadult females in our experiment and thus cannot assess whether females closer to their final moult were more likely to be detected and visited. Previous research suggests that L. hesperus males recognize and respond to contact cues on the silk of subadult females (Ross and Smith 1979 but see Anava and Lubin 1993). Direct contact with silk might thus provide males with the opportunity to assess the relative value of guarding a particular subadult female, as in jumping spiders (Salticidae), in which contact cues on the silk allow males to detect that a female is close to moulting (Jackson 1996).

However, our observations suggest that male recognition of subadults or discrimination of female status is not very fine-tuned in L. hesperus. We observed individual males guarding subadult females for as long as 15 days, and in one case two males guarded a female for a combined total of 20 days, with the first period of cohabitation beginning more than one month before her moult to maturity. We also occasionally found males cohabiting on the webs of females that were two moults prior to adulthood and even with particularly large subadult males (who can be similar in size to antepenultimate or small penultimate females at this site; CES pers. obs.). Thus, contact cues on silk may only allow males to recognize resident spiders as conspecifics, but not to effectively discriminate their developmental stage or perhaps even their sex. We speculate that males may depend primarily on vibrations to assess the size of the resident spider (Sivalinghem et al. 2010), and guard it if its mass is above some threshold. Further effort is needed to understand the extent to which males can use cues (whether chemical

53 or in other modalities) to assess the value of individual subadult females as potential mating partners.

Our data are consistent with the hypothesis that males use chemical cues produced by nearby adult females to locate subadult females. However, we cannot exclude the possibility that subadults become detectable at a distance (perhaps as a result of chemical changes associated with the moulting process; Kittredge and Takahashi 1972; Jormalainan 1998; Weeks and Benvenuto 2008; Estrada et al. 2010) at some point before maturity, since our experiments always provided males with a choice between actively signaling adult females and subadults. Our results and those of choice experiments in the immature-mating congeners L. hasselti and L. geometricus (Andrade and Kasumovic 2005; Waner et al. 2018) suggest that males either strongly prefer adult females to subadults or cannot detect subadults when using volatile chemical cues to locate potential mates. Whether or not males can detect subadult females at a distance will require further research, but even if they can, it is not surprising that males would prefer to approach adults over subadults when given a choice. The expected fitness payoff from mating with an adult who is ready to mate and produce egg sacs is almost certainly higher than that from mating with a subadult. Subadults must moult (with attendant increased vulnerability to predation; Tanaka 1984; Soluk 1990) and will only begin to produce egg sacs after a time lag if they survive. Furthermore, cohabiting with subadults is risky because they are predators who are likely to prioritize foraging over mating; we occasionally witnessed cannibalism of guarding males by subadults (Fig. 2b). However, the expression of male preferences is expected to be constrained by ecological factors including the costs of mate searching and the availability of potential mates (Bonduriansky 2001). In addition to suffering high mortality during mate searching, males in this population will not often have the opportunity to choose between unmated adults and subadults because actively signaling adult females are rare and the operational sex ratio is extremely male-biased (Scott et al. 2019 [Chapter 1]).

If signaling adult females are rare and subadults do not produce chemicals that allow males to detect them from a distance, it raises the question of how males locate subadults so effectively at this site. First, previously mated females may produce volatile chemical cues that are attractive to males. Searching L. hesperus males prefer unmated to mated females in choice experiments but do locate mated females at low rates (MacLeod and Andrade 2014). In a congener, females may continue to produce pheromones at low levels immediately after mating

(Baruffaldi and Andrade 2017) and increase pheromone production over time since mating (Perampaladas et al. 2008). Our model showed that male presence is significantly predicted by the number of older (likely mated) adult females under a microhabitat (Table 2). Second, in addition to chemical cues produced by adult females, males have access to inadvertent social information (Rieucau & Giraldeau 2011) produced by their rivals. Spiders leave silk draglines behind them as they move through the environment (Foelix 2011), and L. hesperus males recognize and follow conspecific draglines, using them to efficiently locate females even if they are no longer actively signaling or detectable via volatile cues (Scott et al. 2019 [Chapter 1]). We speculate that in the absence of any direct information indicating the location of an unmated females, males that encounter the draglines of rivals (who would have used volatile female signals, if available, to guide their movements) can follow them to locate a microhabitat containing at least one female. Even if she is already mated, once there, males are likely to encounter cohabiting subadults.

We conclude that the spatial distribution of females, in addition to other ecological factors such as OSR, affects the feasibility and frequency of immature mating as a reproductive tactic. Males are able to locate subadults by navigating to microhabitats inhabited by adult females and engage in contests over the opportunity to cohabit with subadults until they are fertilizable. The traits favoured under scramble competition for adult females vs. guarding subadults are likely to be different, and this may explain the maintenance of wide variation in male body size in this species. Future work to understand these traits and the fitness consequences of the subadult mating tactic compared to adult mating would be valuable.

Acknowledgements

We thank the Tsawout first Nation for allowing us to do fieldwork on their lands. We also thank PK McCann for providing housing during fieldwork. L Barrufaldi, C Condy, D Gwynne, and A Mason provided helpful comments on the paper. Funding was provided by the Natural Sciences and Engineering Research Council of Canada (Discovery grant no. RGPIN-2017-06060) and the Canada Research Chairs Program (both to MCBA); CES was supported by an NSERC CGS-D during the study.

2.4 References Alcock J. 1991. Adaptive mate-guarding by males of Ontholestes cingulatus (Coleoptera: Staphylinidae). Journal of Insect Behavior. 4(6): 763-771. Anava A, Lubin Y. 1993. Presence of gender cues in the web of a widow spider, Latrodectus revivensis, and a description of courtship behaviour. Bulletin of the British Arachnological Society. 9(4): 119-122. Andersson MB. 1994 Sexual selection. Princeton, NJ: Princeton University Press. Andrade MCB. 2003 Risky mate search and male self-sacrifice in redback spiders. Behavioral Ecology. 14: 531-538. Andrade MCB, Kasumovic MM. 2005. Terminal investment strategies and male mate choice: extreme tests of Bateman. Integrative and Comparative Biology. 45(5): 838-847. Andrade MCB, MacLeod EC. Potential for CFC in black widows (genus Latrodectus): mechanisms and social context. In: Cryptic female choice in arthropods—patterns, mechanisms and prospects (eds. Peretti AV, Aisenberg A.) Berlin: Springer. p. 27 – 53. Baruffaldi L, Andrade MCB. 2015. Contact pheromones mediate male preference in black widow spiders: avoidance of hungry sexual cannibals? Animal Behaviour. 102: 25-32. Baruffaldi L, Andrade MCB. 2017. Neutral fitness outcomes contradict inferences of sexual ‘coercion’derived from male’s damaging mating tactic in a widow spider. Scientific Reports. 71: 17322. Bel-Venner MC, Venner S. 2006. Mate-guarding strategies and male competitive ability in an orb-weaving spider: results from a field study. Animal Behaviour. 71(6): 1315-1322. Biaggio MD, Sandomirsky I, Lubin Y, Harari AR, Andrade MCB. 2016. Copulation with immature females increases male fitness in cannibalistic widow spiders. Biology Letters. 12(9): 20160516. Birkhead TR, Clarkson K. 1980. Mate selection and precopulatory guarding in Gammarus pulex. Zeitschrift für Tierpsychologie. 52(4): 365-380. Bonduriansky R. 2001. The evolution of male mate choice in insects: a synthesis of ideas and evidence. Biological Reviews. 76(3): 305-339. Brandt Y, Andrade MCB. 2007 Testing the gravity hypothesis of sexual size dimorphism: are small males faster climbers? Functional Ecology. 21: 379 – 385. Clutton-Brock TH, Parker GA. 1992. Potential reproductive rates and the operation of sexual selection. The Quarterly Review of Biology. 67(4): 437-456. Darwin C. 1871 The descent of man, and selection in relation to sex. London, UK: Murray. Davies NB, Halliday TR. 1978. Deep croaks and fighting assessment in toads Bufo bufo. Nature. 274(5672): 683-685. Deinert EI, Longino JT, Gilbert LE. 1994. Mate competition in butterflies. Nature. 370(6484): 23-24. Dodson GN, Beck MW. 1993. Pre-copulatory guarding of penultimate females by male crab spiders, Misumenoides formosipes. Animal Behaviour. 46(5): 951-959.

Elgar MA 1998. Sperm competition and sexual selection in spiders and other arachnids. In: Birkhead TR, Møller AP (eds) Sperm competition and sexual selection. Academic Press, London, pp 307–340. Elias DO, Andrade MC, Kasumovic MM. 2011. Dynamic population structure and the evolution of spider mating systems. In Advances in insect physiology (Vol. 41), pp. 65-114. Academic Press. Estrada C, Yildizhan S, Schulz S, Gilbert LE. 2009. Sex-specific chemical cues from immatures facilitate the evolution of mate guarding in Heliconius butterflies. Proceedings of the Royal Society B: Biological Sciences. 277(1680): 407-413. Emlen ST, Oring LW. 1977. Ecology, sexual selection, and the evolution of mating systems. Science. 197: 215-223. Foelix R. 2011 Biology of Spiders. New York, NY: Oxford University Press Foitzik S, Heinze J, Oberstadt B, Herbers JM. 2002. Mate guarding and alternative reproductive tactics in the ant Hypoponera opacior. Animal Behaviour, 63(3): 597-604. Gaskett AC. 2007. Spider sex pheromones: emission, reception, structures, and functions. Biological Reviews. 821: 27-48. Grafen A, Ridley M. 1983. A model of mate guarding. Journal of Theoretical Biology. 102(4): 549-567. Harari AR, Zahavi T, Thiéry D. 2011Fitness cost of pheromone production in signaling female moths. Evolution: International Journal of Organic Evolution. 65(6): 1572-1582. Harts AM, Kokko H. 2013. Understanding promiscuity: when is seeking additional mates better than guarding an already found one? Evolution. 6710: 2838-2848. Herberstein ME, Painting CJ, Holwell GI. 2017 Scramble competition polygyny in terrestrial arthropods. In M Naguib, J Podos, LW Simmons, L Barrett, SD Healy, M Zuk (eds), Advances in the study of behavior (Vol. 49, pp. 237-295. Cambridge, MS: Elsevier. Holdcraft R, Rodriguez-Saona C, Stelinski LL. 2016 Pheromone autodetection: evidence and implications. Insects. 7(2): 17. Ims RA. 1988. The potential for sexual selection in males: effect of sex ratio and spatiotemporal distribution of receptive females. Evolutionary Ecology. 2(4): 338-352. Jackson RR. 1986. Use of pheromones by males of Phidippus johnsoni (Araneae, Salticidae) to detect subadult females that are about to molt. The Journal of Arachnology. 141: 137- 139. Johansson, B. G., Jones, T. M., & Widemo, F. (2005). Cost of pheromone production in a lekking Drosophila. Animal Behaviour, 69(4), 851-858. Jormalainen V. 1998. Precopulatory mate guarding in crustaceans: male competitive strategy and intersexual conflict. The Quarterly Review of Biology. 73(3): 275-304. Kaston BJ. 1970. Comparative biology of American black widow spiders. Transactions of the San Diego Society of Natural History. 16: 33-82.

Kasumovic MM, Andrade MC. 2004. Discrimination of airborne pheromones by mate-searching male western black widow spiders (Latrodectus hesperus): species-and population- specific responses. Canadian Journal of Zoology. 82(7): 1027-1034. Kelly CD, Bussière LF, Gwynne DT. 2008. Sexual selection for male mobility in a giant insect with female-biased size dimorphism. The American Naturalist. 172(3): 417-423. Kittredge JS, Takahashi FT. 1972. The evolution of sex pheromone communication in the Arthropoda. Journal of Theoretical Biology. 35(3): 467-471. MacLeod EC. 2013. New insights in the evolutionary maintenance of male mate choice behaviour using the western black widow, Latrodectus hesperus. Doctoral dissertation, University of Toronto. MacLeod EC, Andrade MCB. 2014. Strong, convergent male mate choice along two preference axes in field populations of black widow spiders. Animal Behaviour. 89: 163 – 169. Magnhagen, C. (1991). Predation risk as a cost of reproduction. Trends in Ecology & Evolution, 6(6), 183-186. Miyashita T, Hayashi H. 1996. Volatile chemical cue elicits mating behavior of cohabiting males of Nephila clavata (Araneae, Tetragnathidae). Journal of Arachnology. 44(1): 9-15. Parker GA. 1970. Sperm competition and its evolutionary consequences in the insects. Biological Reviews. 45(4): 525-567. Parker GA. 1974a. Courtship persistence and female-guarding as male time investment strategies. Behaviour. 481(4): 157-183. Parker GA. 1974b. Assessment strategy and the evolution of fighting behaviour. Journal of Theoretical Biology. 471: 223-243. Parker GA. 1978. Evolution of competitive mate searching. Annual Review of Entomology. 231: 173-196. Perampaladas K, Stoltz JA, Andrade MCB. 2008. Mated redback spider females re‐advertise receptivity months after mating. Ethology. 114(6): 589-598. Prenter J, Elwood RW, Montgomery IW. 2003. Mate guarding, competition and variation in size in male orb-web spiders, Metellina segmentata: a field experiment. Animal Behaviour. 66(6): 1053-1058. Provost MW, Haeger JS. 1967. Mating and pupal attendance in Deinocerites cancer and comparisons with Opifex fuscus (Diptera: Culicidae). Annals of the Entomological Society of America. 60(3): 565-574. RStudio Team 2018. RStudio: Integrated Development for R. RStudio, Inc., Boston, MA URL http://www.rstudio.com/. Ridley M. 1989. The timing and frequency of mating in insects. Animal Behaviour. 37: 535-545. Rieucau G, Giraldeau LA. 2011. Exploring the costs and benefits of social information use: an appraisal of current experimental evidence. Philosophical Transactions of the Royal Society B. 366: 949 – 957.

Ross K, Smith RL. 1979. Aspects of the courtship behavior of the black widow spider, Latrodectus hesperus (Araneae: Theridiidae), with evidence for the existence of a contact sex pheromone. Journal of Arachnology. 7(1): 69-77. Salomon M. 2008. Facultative group living in the western black widow spider, Latrodectus hesperus: an evolutionary approach (Doctoral dissertation, Dept. of Biological Sciences- Simon Fraser University). Salomon M, Vibert S, Bennett RG. 2010 Habitat use by western black widow spiders (Latrodectus hesperus) in coastal British Columbia: evidence of facultative group living. Canadian Journal of Zoology. 88: 334-346. Scott CE, Anderson AG, Andrade MCB. 2018 A review of the mechanisms and functional roles of male silk use in spider courtship and mating. The Journal of Arachnology. 46(2): 173- 207. Scott C, Kirk D, McCann S, Gries G. 2015a Web reduction by courting male black widows renders pheromone-emitting females' webs less attractive to rival males. Animal Behaviour. 107: 71-78. Scott C, McCann S, Gries R, Khaskin G, Gries G. 2015b. N-3-methylbutanoyl-O- methylpropanoyl-L-serine methyl ester–pheromone component of western black widow females. Journal of Chemical Ecology. 41(5): 465-472. Scott CE, McCann S, Andrade MCB. 2019. Male black widows parasitize mate-searching effort of rivals to find females faster. Proceedings of the Royal Society B. 286: 20191470. Scott C, Vibert S, Gries G. 2012 Evidence that web reduction by western black widow males functions in sexual communication. The Canadian Entomologist. 144(5): 672-678. Schröder T. 2003. Precopulatory mate guarding and mating behaviour in the rotifer Epiphanes senta (Monogononta: Rotifera). Proceedings of the Royal Society of London. Series B: Biological Sciences. 270: 1965-1970. Segev O, Ziv M, Lubin Y. 2003 The male mating system in a desert widow spider. Journal of Arachnology. 31: 379-393. Segoli M, Harari AR, Lubin Y. 2006 Limited mating opportunities and male monogamy: a field study of white widow spiders, Latrodectus pallidus (Theridiidae). Animal Behaviour 72, 635 – 642. Simmons LW. 2001. Sperm competition and its evolutionary consequences in the insects. Princeton University Press, Princeton, NJ. Sivalinghem S, Kasumovic MM, Mason AC, Andrade MCB, Elias DO. 2010. Vibratory communication in the jumping spider Phidippus clarus: polyandry, male courtship signals, and mating success. Behavioral Ecology. 21(6): 1308-1314. Snow LS, Abdel‐Mesih A, Andrade MC. 2006. Broken copulatory organs are low‐cost adaptations to sperm competition in redback spiders. Ethology. 112(4): 379-389. Soluk DA. 1990. Postmolt susceptibility of Ephemerella larvae to predatory stoneflies: constraints on defensive armour. Oikos. 58(3): 336-342. Tanaka K. 1984. Rate of predation by a kleptoparasitic spider, Argyrodes fissifrons, upon a large host spider, Agelena limbata. Journal of Arachnology. 12: 363-367.

Thornhill R, Alcock J. 1983. The evolution of insect mating systems. Cambridge, MA: Harvard University Press. Tuni C, Berger-Tal R. 2012. Male preference and female cues: males assess female sexual maturity and mating status in a web-building spider. Behavioral Ecology. 23(3): 582-587. Uhl G. 2000. Female genital morphology and sperm priority patterns in spiders (Araneae). In 19th European Colloquium of Arachnology, 17-22 July (ed. S. Toft and N. Scharff), pp. 145–156. Aarhus University Press, Aarhus. Uhl G, Zimmer SM, Renner D, Schneider JM. 2015. Exploiting a moment of weakness: male spiders escape sexual cannibalism by copulating with moulting females. Scientific Reports. 5: 16928. van Wijk M, Heath J, Lievers R, Schal C, Groot AT. 2017Proximity of signallers can maintain sexual signal variation under stabilizing selection. Scientific reports, 7(1), 18101. Vetter RS. 1980 Defensive behavior of the black widow spider Latrodectus hesperus (Araneae: Theridiidae). Behavioral Ecology and Sociobiology. 7(3): 187-193. Waner S, Motro U, Lubin Y, Harari AR. 2018. Male mate choice in a sexually cannibalistic widow spider. Animal Behaviour. 137: 189-196. Weeks SC, Benvenuto C. 2008. Mate guarding in the androdioecious clam shrimp Eulimnadia texana: male assessment of hermaphrodite receptivity. Ethology. 1141: 64-74. Weir LK, Grant JW, Hutchings JA. 2011. The influence of operational sex ratio on the intensity of competition for mates. The American Naturalist. 1772: 167-176. Wyatt TD. 2003Pheromones and Animal Behaviour: Communication by Smell and Taste. Cambridge. Zuk M, Kolluru GR. 1998Exploitation of sexual signals by predators and parasitoids. The Quarterly Review of Biology. 73(4), 415-438.

Season Early Late

Observed Expected Observed Expected

Only visited subadult females 42% (14) 7.5% 73% (155) 44.8%

Ever visited adult females 58% (19) 92.5% 27% (56) 55.2%

Statistics 2 = 58.0; P < 0.0001 2 = 70.1; P < 0.0001

Table 2-2 Results of a generalized linear mixed effect model (glmm) assessing the relationship between Latrodectus hesperus male presence (binary response variable) under microhabitats and nearest neighbour distance and the presence of females at different developmental stages. Immature females (in their penultimate instar) likely do not signal but can be mated just before their molt to maturity, recently matured females are likely to be actively signalling, and mature females are likely to be already mated.

Estimate (Odds ratio) SE Z P

Intercept -4.354 1.035 -4.303 <0.0001

Distance to nearest microhabitat 0.283 (1.33) 0.189 1.500 0.134

Immature females 0.817 (2.27) 0.129 6.338 <0.0001

Recently matured females 0.994 (2.70) 0.231 4.312 <0.0001

Mature females 0.176 (1.19) 0.067 2.618 0.0088

Figure 2-1 Latrodectus hesperus has a clumped distribution at our coastal field site, where driftwood logs above the high tide line provide the most common microhabitats, which are typically shared by multiple spiders. (a) Satellite image (Map data ©2015 Google) of our main field site; black dots indicate logs (>1 m from nearest neighbour) or clusters of logs (when two or more logs were 1m from each other) that were ever occupied by L. hesperus during our 2016 field season. (b) Photograph of field site depicting the sand dunes, low vegetation, and abundant driftwood logs. (c) Photograph of an individual log and two of the females (with visible paint marks on the legs indicating their unique IDs) who used it as a microhabitat.

Figure 2-2 Interactions between Latrodectus hesperus individuals on the webs of subadult females. (a) Males engaged in combat on the web of a subadult; the lower male has his jaws (chelicerae) spread. Fighting males attempt to bite their opponents and also to wrap one another with silk. (b) One of two cohabitating males is bitten by a subadult female (note the bulge on her abdomen, but with no opening) after being wrapped with silk. (c) A marked male guards a subadult female in her retreat. (d) A male copulates with a subadult female (though not marked, the white colouration and our subsequent observation of her moulting indicated that she was mated while still immature).

Figure 2-3 Cumulative abundance of Latrodectus hesperus males and females that were available for mating during two parts of the 2016 mating season at our field site: (a) early season (b) late season. Each point represents a survey date and arrows indicate the dates of full population censuses. ‘Subadult’ females were one moult away from maturity and ‘recently matured females’ were within one week of their moult; we conservatively assume that females remain receptive and actively signalling for one week following their moult to maturity. We do not include older adult females (at least one week post-maturity and thus likely to have already mated) in this figure. Their abundance remained relatively constant and was similar across both seasons (median [range]: 104 [81 – 136] in the early season and 108 [88 – 172] in the late season).

Figure 2-4 Experimental design for testing male localization of females with different spatial distributions in the field. Each square represents an unmated Latrodectus hesperus female inside a mesh cage (a pheromone trap). Black boxes represent “old” females (2 – 5 months post maturity), grey boxes represent “young” females (9 – 12 days post maturity), and white boxes represent subadult females in their penultimate instar. (a) Uniform distribution. (b) Clumped distribution.

Figure 2-5 Minimum number of females of different age classes found by L. hesperus males in their lifetimes. (a) Early season: April – June (n = 53 males). (b) Late season: July – September (n = 271 males). These data overestimate the percentage of males that found any female, because those that were not detected as juveniles and perished before finding a female were never observed (an estimated 88% of males die during moves between microhabitats; Scott et al. 2019 [Chapter 1]).

Figure 2-6 Number of males found on webs of L. hesperus females who were followed from their penultimate instar through to maturity. (a) Early season: April – June (n = 13 females). (b) Late season: July – September (n = 65 females). The zero classes may be overestimated; these values represent the minimum number of males that visited each female because we did not check their webs every night (see methods for more details).

Figure 2-7 Number of Latrodectus hesperus males captured at traps containing unmated females of different age classes in field experiments where females were arranged in either a uniform or clumped distribution (see Fig. 2 for details of experimental setup).

Widows as plastic wallflowers: female choosiness shifts with indicators of mate availability in a natural population of black widows

Catherine E Scott, Sean McCann, and Maydianne CB Andrade

Abstract

Female choice is an important force driving sexual selection, but because exercising choice can impose costs, including a risk of remaining unmated, females are expected to adjust their reproductive decisions to maximize fitness under a given context. The social environment experienced during development can provide reliable information about local demography and lead to adaptive plasticity in female preferences and choosiness. We investigated female choice in a field population of western black widow spider (Latrodectus hesperus). Females spent their final juvenile stage and early adulthood either 10-m (“isolated”) or 1-m (“clustered”) from the nearest naturally occurring conspecific female. Upon maturity, clustered females were visited by males earlier and by more males in rapid succession, indicating that proximity to conspecifics affords females greater opportunity to engage in mate choice and lowered risk of remaining unmated. Despite experiencing similar total numbers of male visitors during the experiment, isolated females displayed decreased choosiness compared to clustered females in staged mating trials: isolated females neither rejected males nor engaged in pre-copulatory cannibalism. These results demonstrate that natural variation in social environment in the field can drive shifts in female choosiness, and that female behavior forms a link between population density and intensity of sexual selection on males in the wild.

3.1 Introduction

Female choice is an important force driving sexual selection on males (Darwin 1871; Andersson 1994). Females vary in their mate choice decisions, and understanding this variation (which may be plastic within as well as among individual females; Ah-King & Gowaty 2016) is valuable because it will influence the strength and direction of sexual selection. Both intrinsic (e.g., age, body condition, mating experience) and environmental (e.g., predation risk, density, OSR) factors can cause variation in female mate choice (Jennions & Petrie 1997; Ah-King & Gowaty 2016; Kelly 2018). These factors may affect either the feasibility or fitness consequences of

69 70 exercising choice. Since mate choice can impose costs, including time or energy spent sampling, and increasing the risk of predation or remaining unmated, (Reynolds & Gross 1990; Milinski & Bakker 1992; Rowe 1994; Booksmythe et al. 2008), females should adjust reproductive decisions in a way that maximizes their fitness under a given context.

The social environment experienced by animals as juveniles and adults can affect mate choice, including both preferences (order in which prospective mates are ranked) and choosiness (effort or energy individuals are willing to spend on mate assessment) (Jennions & Petrie 1997). Attending to sexual signals (which may be in various modalities including acoustic, visual, and chemical), can provide individuals with information about the abundance or attractiveness of potential mates or competitors in their local environment (Kasumovic & Brooks 2011). Exposure to such social information during development can lead to phenotypic plasticity, including behavioural plasticity in the context of mate choice (West-Eberhard 2003; Pigliucci 2001; Kasumovic et al. 2012). For example, female wolf spiders express preferences for males with phenotypes that they have experienced as subadults (Hebets 2003) and female field crickets reared in silence are more responsive to male signals than females exposed to male signals during development (Bailey & Zuk 2008). Experience during development will be useful if social information provides reliable information about the selective environment an animal will experience upon adulthood (Lively 1986; Fawcett & Frankenhuis 2015); in changing environments, conditions experienced as a subadult or in early adulthood may be most relevant (Sachser et al. 2013; Nessler et al. 2009a,b; Cory & Schneider 2018a,b). In crickets, for instance, experience as adults had a more important effect on female choice than subadult experience (Swanger & Zuk 2015).

One cost of mate choice that is likely to be closely linked with the social environment is the risk of delays to reproduction or failure to mate at all when choosy females reject males (De Jong & Sabelis 1992; Kokko & Mappes 2005). Females are expected to be less choosy in low- density environments when encounter rates with males are likely to be low (Jennions & Petrie 1997; Bleu et al. 2012; Roff & Fairbairn 2015, Simmons and Kvarnemo, 2006). Demographic structure may vary on relatively small spatial and temporal scales (Gwynne et al. 1998; Kasumovic et al. 2008; Punzalan et al. 2010; Elias et al. 2011; Kasumovic & Brooks 2011), and in such situations plasticity in mate choice behavior is likely to be adaptive (Snell-Rood 2013). For example, an evolutionary history of a risk of mate limitation may lead to unmated females

71 adjusting choosiness as a function of local cues of male availability (e.g., Shelly and Bailey, 1992), and this may include accepting the first available mate (the adaptive ‘Wallflower effect’, Kokko & Mappes, 2005; Heubel et al 2008). Most evidence for mate choice plasticity comes from laboratory experiments (reviewed in Ah-King & Gowaty 2016). It is challenging to extrapolate from these to infer effects on sexual selection in the field, since complex social environments may lead to multiples cues experienced simultaneously in natural populations.

We focus here on how the social environment affects female choice in a field population of western black widow spiders (Latrodectus hesperus). Females are polyandrous (MacLeod 2013) and can exercise choice via mate rejection (including pre-copulatory cannibalism; Johnson et al. 2010), or allowing males to deposit sperm into only one of their paired sperm storage organs and thus retaining the opportunity to allocate some paternity to a subsequently-mating male (Andrade & MacLeod 2015). Males engage in scramble competition polygyny (Herberstein et al. 2017), and there is a male-biased OSR through most of the season (Scott et al. 2019 [Chapter 1]). However, there is evidence for a risk of remaining unmated in the field as 4% of females were not mated by the end of the 2016 mating season (CES unpublished data). Population density varies across the field site in that spiders may or may not cohabit with conspecifics. Work in a congener (L. hasselti) suggests that chemical cues produced by nearby conspecific males and females (Kasumovic & Andrade 2004; Scott et al. 2015a; Scott et al. 2018 [Chapter 4]) are available to Latrodectus females throughout their lives and may indicate local population density or proximity of conspecifics (Kasumovic & Andrade 2006; Kasumovic, Brooks, et al,. 2008; Stoltz et al. 2012).

We used a field experiment to ask whether female proximity to conspecifics affects mate encounter rates and whether the social environment (cues of conspecifics generally combined with male availability) experienced by females during subadulthood and early adulthood affects female choice. We placed caged subadult females under artificial microhabitats either nearby (“clustered” treatment) or far from (“isolated” treatment) conspecific-occupied natural microhabitats in the field. This ensured females completed development in the presence of natural cues of conspecific density. We predicted that females in the clustered treatment would be visited by more males, and would be visited sooner after maturity, compared to isolated females. Thus we expected that isolated females would perceive both a lower overall density of conspecifics and lower mate availability than clustered females. As a result, we predicted that

72 isolated females would be less likely to reject the first male that they encountered in staged mating trials (i.e., be less choosy) than would clustered females. This study is important because it tests for effects of density on female choosiness in a way that preserves natural exposure to complex cues (Dore et al 2018) throughout the female’s final developmental stage, which is the most likely critical period for female behavioural plasticity in Latrodectus (Andrade 2019).

3.2 Methods

3.2.1 Natural History and Field Site

We studied a population of Latrodectus hesperus Chamberlin and Ivie, 1935 inhabiting the coastal sand dunes above the high-tide line at Island View Beach and T̸IX̱ EṈ (Cordova Spit) on the Saanich Peninsula of Vancouver Island, British Columbia, Canada (48°35’10”N, 123°22’17”W). Driftwood logs, other woody debris, and occasional rocks provide microhabitats for western black widows at this site (Fig. 1). The spiders typically spin silk retreats in sheltered areas under logs, and their three-dimensional capture webs extend onto the sand and nearby vegetation. The population is dense, with 2-3 subadult or adult females per square metre of microhabitat in late summer, such that most microhabitats are shared by multiple conspecific females and juveniles (see Salomon et al. 2010 for more details about this site and population). However, there is also considerable variability in nearest neighbour distances between microhabitats (mean [range] = 4.2 [0.1–15.6] m; CES unpublished data) and some solitary females are found throughout the season.

Unmated female western black widows produce silk-bound sex pheromones that attract males at long range (Kasumovic and Andrade 2004; MacLeod and Andrade 2014; Scott et al 2019 [Chapter 1]) and elicit male courtship behaviour upon contact (Ross and Smith 1979; Scott et al. 2015b). Males apparently cannot detect subadult females based on volatile chemical cues (Chapter 2) and it is not clear how soon after maturity females begin producing sex pheromone. There is some behavioural evidence for male sex pheromones or chemical cues associated with silk draglines (Ross & Smith 1986; Scott et al. 2019 [Chapter 1]), and work in a congener indicates that males produce volatile chemical cues that are detectable by other males (Kasumovic & Andrade 2006).

Delays to mating are likely to be costly for Latrodectus females. There is evidence for a longevity cost of remaining unmated (e.g., in a congener; Stoltz et al. 2010). Moreover, since the availability of adult males drops off near the end of the season (Scott et a. 2019 [Chapter 1]), females unmated at that point would typically have to survive overwintering and attract a male in the subsequent season in order to reproduce.

3.2.2 Field Experiment

We placed 40 immature females (collected from our field site but outside of the study area) either close to (“clustered” treatment) or far from (“isolated” treatment) natural microhabitats and monitored them over approximately 90 days during the most active part of the mating season (mid-August to late September). We distributed 20 pairs of cages (“pheromone traps” as in MacLeod & Andrade 2014) each containing one immature female (who appeared to be in the penultimate instar based on inspection of their genitalia; Kaston 1970) and their webs at 10-m intervals along a 190-m transect. Each trap consisted of a mesh-sided cage (design modified from Scott et al. 2015a) underneath a wooden shed (two 30 × 30 cm pieces of plywood connected at right angles) that mimicked the natural driftwood microhabitats occupied by black widows at our field site (Fig. 1). Both cages and sheds were staked into the ground with bamboo skewers. Each pair of traps included one in an isolated location at least 10 m from the nearest occupied microhabitat (i.e. a log under which there was at least one adult or subadult conspecific female) and one clustered within 1 m of an occupied microhabitat.

We checked traps each morning and collected all males that had arrived the previous night (males tend to remain on a pheromone trap after their arrive; CES pers obs). We also checked daily for evidence that experimental females had moulted or mated. When a moult was present in the cage, we noted the date and determined whether the female had moulted to maturity (based on morphology of the external genitalia; Kaston 1970). After first moulting events, we determined that eight females were still in the penultimate instar (with closed genitalia), indicating that they had begun the experiment in the antepenultimate instar (isolated: 6 females; clustered: 2 females). If spiders did not reach maturity before the end of the experiment (whether they began in the antepenultimate or penultimate instar) they were excluded from our analysis (isolated: 5 females; clustered: 3 females). We also checked for immature mating (Baruffaldi & Andrade, in review; Chapter 2) because even though males could not enter

74 females’ cages, the mesh was large enough that it could be possible for a male to copulate with a female through the holes. Immature mating is evident when the normally exoskeleton-covered genitalia of an immature female is visibly open but no exuvium is present, and appearance of this opening is followed by a moult one or more days later (Biaggio et al. 2016). During the experiment, we fed all females one house cricket (Acheta domesticus) weekly.

3.2.3 Mating Trials

Following the field experiment, we brought all of the spiders indoors for mating trials. We introduced males into the cages of fifteen adult females from each treatment. The males were randomly selected from those collected at experimental cages near the end of the experiment. Trials began around 20:00 and proceeded with the lights on. We checked each cage periodically for copulation or cannibalism. We allowed males to remain in females’ cages all night, terminating trials at 09:00 the following morning. For females that we did not observe copulating, we determined copulation success a posteriori by dissecting their genitalia and checking for the presence of sperm plugs, which can only be deposited during copulation (see Andrade & MacLeod 2015). If we did not observe copulation prior to noting a cannibalized male, we confirmed that cannibalism was post-copulatory based on the presence of at least one sperm plug and concluded that it was pre-copulatory if there were no plugs present. We concluded that a female had rejected a male if no copulations occurred. All males courted persistently and thus failure to copulate is very unlikely to be the result of male choice. After the mating trials, all females were brought back to the laboratory in Scarborough and fed one cricket per week until the end of their natural lives. We recorded whether each female produced egg sacs and whether offspring emerged from them.

3.2.4 Statistical Analyses

We used Fisher’s exact tests to determine whether social environment treatment (isolated or clustered) was associated with whether or not any males visited focal females (i) before, and (ii) after they moulted to maturity. We also used a Wilcoxon rank sum test to ask whether treatment affected the total number of males that visited each female during the experiment.

We used a Welch’s t-test for unequal variances to determine whether males started to arrive earlier at the cages of females in the clustered treatment than the isolated treatment. We

75 first log-transformed the data (time of arrival of the first male, in days since the moult to maturity) to meet the assumption of normality. We excluded three females who were visited prior to maturity from this analysis. We also used a Wilcoxon rank sum test to ask whether the number of males that arrived over the two days following the arrival of the first male at a female’s web differed with density treatment. This analysis was included as a post-hoc test of the hypothesis that the opportunity for females to choose among males arriving in rapid succession differed between treatments. Males arriving on the same night or over two consecutive nights would provide a female with the opportunity to engage in simultaneous or sequential choice over a short period of time that would cause minimal delays to oviposition.

We used logistic regression with Firth’s bias-reduction method (brglm package in R; Kosmidis 2019) to determine whether social environment in the field, female age on the date of the mating trial (days since the moult to maturity), and the number of males who arrived at a female’s cage over the course of the experiment (which could provide information about the local availability of mating partners) affected female choosiness. We ran two separate models with treatment, female age, and number of males as predictors, and copulation (yes/no) and pre- copulatory cannibalism (yes/no) as binary response variables. We checked whether the rate of reproductive failure differed between treatments using a Fisher’s exact test. All analyses were performed in RStudio version 1.2.1335 (RStudio Team 2018).

3.3 Results

3.3.1 Field Experiment

Our treatments did not affect the number of males that visited females across the duration of the field experiment. As expected (Chapter 2), females were rarely visited by males before maturity, and this was equally rare in both treatments (Fisher’s exact test: P = 0.23; Table 1). Our social environment treatments also had no effect on whether females were visited by at least one male once they were adults (Fisher’s exact test: P = 0.75) nor did it affect the total number of males that visited each female over the entire experiment (Wilcoxon rank sum test: W = 214; P = 0.7; Table 1). One female from the isolated treatment was mated while she was still immature.

However, the timing of male arrival at females’ webs once they matured indicates that clustered females would have the opportunity to mate sooner and a greater opportunity to engage

76 in mate choice compared to isolated females. Males arrived earlier at the webs of clustered females (untransformed mean [range] = 8.1 [1–20] days after the moult to maturity; n = 11) than at webs of isolated females (12.8 [4 – 26] days after the moult to maturity; n = 8; t = -1.82, df = 16.9, P = 0.043; Fig. 2a). Moreover, the number of males that arrived within two days of the first male finding a female was also greater for clustered females (median = 2 males) than for isolated females (median = 1 male; W = 63.5; P = 0.033; Fig. 2b).

3.3.2 Mating Trials

Females from the clustered treatment were choosier than females from the isolated treatment (Fig. 3; Table 2). Females from the isolated treatment were more likely to copulate at least once than females from the clustered treatment (Fig. 3a; Table 2). Females in the clustered treatment also tended to be more cannibalistic than those in the isolated treatment; only females from the clustered treatment engaged in pre-copulatory cannibalism (Fig. 3b). We confirmed that female age (in days since the moult to maturity) did not differ between treatments (W = 241, P = 0.270) and did not affect female copulation or cannibalism behaviour (Table 2). Moreover, female choosiness was not affected by direct experience with male cues after adulthood (measured as the total number of males attracted to female’s cages in the field; Table 2). The rate of reproductive failure for females that copulated at least once also did not differ between treatments: 2/15 for isolated females vs. 4/10 for clustered females (Fisher’s exact test; P = 0.175).

3.4 Discussion

Our results demonstrate that the social environment experienced by L. hesperus females as subadults and young adults affects their mate choice decisions. Contrary to our predictions, the number of males who visited isolated and clustered females did not differ. Nevertheless, clustered females were visited on average five days earlier and were more likely to be visited by more than one male in rapid succession, which would afford them lowered risk of remaining unmated and greater opportunity to exercise choice compared to isolated females. Neither the number of males who visited females during the experiment nor female age (time spent as an unmated adult) affected female choice, but females from the clustered treatment rejected males more often than isolated females, suggesting that cues of general conspecific density, or proximity, drive variation in choosiness. While similar effects have been shown in other systems,

77 this has rarely been shown in a field population, with the complex cues and social contexts typical of nature (Ah-King & Gowaty 2016). These results show that female behaviour creates a link between population density in the wild and the intensity of sexual selection on males.

Our data are consistent with predictions that females should decrease choosiness in environments where there is a risk of costly delays to mating or of failure to mate at all (De Jong & Sabelis 1992; Jennions & Petrie 1997; Kokko & Mappes 2005). Arrival of males at isolated females’ webs was delayed by five days relative to clustered females’ webs. Such a delay may constitute a significant risk of remaining unmated, particularly close to the end of the mating season. At this time, the OSR shifts rapidly from highly male-biased to slightly female biased (Scott et al. 2019 [Chapter 1]), and thus encounter rates with males will decrease. Furthermore, although females can overwinter and mate or re-mate the following spring (Salomon et al. 2010), the risk of dying before the next mating season may be greater for unmated females because they suffer decreased longevity relative to mated females (Stoltz et al. 2010). For a female in an environment like that of our isolated treatment, mating indiscriminately with the first male she encounters is likely to be adaptive (Jennions & Petrie 1997; Bleu et al. 2012). This will provide fertility assurance but still leaves females with the option of “trading-up” if another, preferred male arrives later (Jennions & Petrie 2000; Pitcher et al. 2003; Fowler-Finn & Rodriguez 2011). Moreover, female widow spiders may also be able to bias paternity toward later-mating males via cryptic female choice (Andrade & MacLeod 2015). Conversely, females in close proximity to conspecifics (as in our clustered treatment) can expect to be visited by multiple males soon after maturity and face a low risk of remaining unmated. For these females, the benefits of exercising choice by rejecting non-preferred males may outweigh the costs (Reynolds & Gross 1990).

Sexual cannibalism may be a mechanism of exercising female choice (Prenter et al. 2006). Only females from the clustered treatment ever engaged in pre-copulatory cannibalism. Although we did not detect a statistically significant difference in cannibalism rates between our treatments, our results appear similar to those of Johnson (2005), who found that female fishing spiders exposed to cues of high male availability during development were more likely to engage in pre-copulatory cannibalism than females housed alone. Sexual cannibalism in L. hesperus is rarely observed in mating trials and typically manifests as pre-copulatory cannibalism when females are hungry (Johnson et al. 2011). This rarity may be exacerbated by the standard method

78 of rearing spiders in isolation from conspecific cues in the laboratory (e.g., Baruffaldi & Andrade 2015). Both clustered and isolated females engaged in post-copulatory cannibalism, which can allow females to prevent a second copulation (thus leaving one of the paired sperm storage organs empty and available for a second male; Andrade & MacLeod 2015), and possibly also provides direct fitness benefits (e.g., Johnson 2005; Welke & Schneider 2012; Schwarz et al. 2016; Anderson & Hebets 2018).

Consistent with our data from Chapter 2, subadult females apparently do not attract males via volatile cues. Once females matured, however, males found them rapidly (as early as one day after maturity), and particularly so if they were clustered near conspecifics. This suggests that that females may begin producing sex pheromone immediately upon maturity, unlike other spider species that delay sex pheromone production for several days after moulting (Watson 1986). Males arrived earlier at webs of females in the clustered treatment. Whether this was because they were near other signaling females (and thus produced a larger combined signal that allowed males to rapidly locate them), or because these females started signaling earlier (in response to signals of nearby conspecific females, as in moths; e.g., Lim & Greenfield 2006; Rehermen et al. 2016) requires further investigation.

We did not measure male phenotypes in this study and so the shape of females’ preference function is unclear for L. hesperus. Previous field experiments at our study site showed that larger males are more likely to find females, but that smaller males travel faster during mate searching (Scott et al. 2019 [Chapter 1]; Appendix 2). This suggests that relatively small males are likely to arrive first at a given female’s web in nature, and larger males later. If isolated females mate indiscriminately with the first male they encounter, this may help to maintain extreme variation in male size in this species (Andrade & MacLeod 2015; Andrade 2019). In addition to interacting with males as adults, L. hesperus females at our field site will often have the opportunity to interact with males while being guarded as subadults (Chapter 2) and these experiences could also shape their preferences and choosiness as adults (Hebets 2003; Dukas 2005; Johnson 2005; Hebets & Vink 2007). Further work is needed to understand female preferences and to measure the fitness consequences of female choice and choosiness in this species.

These results demonstrate plasticity in female mate choice decisions under complex conditions in the field. Our study allowed us to evaluate plasticity under natural levels and timing of mate availability, which is the most relevant context for understanding how selection acts in nature. However, our experimental design does not allow us to isolate subadult vs. adult experience or exposure to male vs. female cues. Further work is needed to determine the critical period for detecting and responding to social cues (Sachser et al. 2013; Fawcett & Frankenhuis 2015; Andrade 2109). Similarly, efforts to disentangle the effects of cues produced by males and females would benefit from direct manipulation of number and density of conspecifics that females experience. It will be particularly valuable to design experiments that allow us to determine if the differences in choosiness that we observed were in response to perceived local conspecific density per se (e.g., Pompilio 2016; Westerman et al. 2014) or simply proximity to signaling females.

Acknowledgements

We thank the Tsawout first Nation for allowing us to do fieldwork on their lands. We also thank C & D Copley for logistic support and N & S de Jong for assistance collecting immature spiders for the experiment. L Barrufaldi, D Gwynne, and A Mason provided helpful comments. Funding was provided by the Natural Sciences and Engineering Research Council of Canada (Discovery grant no. RGPIN-2017-06060) and the Canada Research Chairs Program (both to MCBA); CES was supported by an NSERC CGS-D during the study. The fieldwork was funded by the Toronto Entomologists’ Association Eberlie Grant (to CES), the Experiment.com Arachnid Challenge Grant, and many generous individual donors to the ‘Team Black Widow’ crowdfunding campaign.

3.5 References Ah‐King M, Gowaty PA. 2016 A conceptual review of mate choice: stochastic demography, within‐sex phenotypic plasticity, and individual flexibility. Ecology and Evolution 6, 4607–4642. Anderson AG, Hebets EA. 2018 Female nursery web spiders (Pisaurina mira) benefit from consuming their mate. Ethology 124, 475–482. Andersson MB. 1994 Sexual selection. Princeton, NJ: Princeton University Press. Andrade MCB, MacLeod EC. 2015 Potential for CFC in black widows (genus Latrodectus): mechanisms and social context. In Cryptic female choice in arthropods: patterns,

mechanisms and prospects (eds AV Peretti, A Aisenberg), pp. 27–53. Berlin, Germany: Springer. Andrade MCB. 2019 Sexual selection and social context: Web-building spiders as emerging models for adaptive plasticity. Advances in the Study of Behavior 51, 177–250. Baruffaldi L, Andrade MCB 2015. Contact pheromones mediate male preference in black widow spiders: avoidance of hungry sexual cannibals? Animal Behaviour 102, 25–32. Bailey NW, Zuk M. 2008 Acoustic experience shapes female mate choice in field crickets. Proceedings of the Royal Society B: Biological Sciences 275, 2645–2650. Biaggio MD, Sandomirsky I, Lubin Y, Harari AR, Andrade MCB. 2016 Copulation with immature females increases male fitness in cannibalistic widow spiders. Biology Letters 12, 20160516. Bleu J, Bessa-Gomes C, Laloi D. 2012 Evolution of female choosiness and mating frequency: effects of mating cost, density and sex ratio. Animal Behaviour 83, 131–136. Booksmythe I, Detto T, Backwell PR. 2008 Female fiddler crabs settle for less: the travel costs of mate choice. Animal Behaviour 76, 1775–1781. Cory AL, Schneider JM. 2018a Effects of social information on life history and mating tactics of males in the orb‐web spider Argiope bruennichi. Ecology and Evolution 8, 344–355. Cory AL, Schneider, J. M. 2018b Mate availability does not influence mating strategies in males of the sexually cannibalistic spider Argiope bruennichi. PeerJ 6, e5360. Darwin C. 1871 The descent of man, and selection in relation to sex. London, UK: Murray. De Jong MC, Sabelis MW. 1991 Limits to runaway sexual selection: the wallflower paradox. Journal of Evolutionary Biology 4, 637–655. Dore AA, McDowall L, Rouse J, Bretman A, Gage MJ, Chapman T. 2018 The role of complex cues in social and reproductive plasticity. Behavioral Ecology and Sociobiology 72, 124. Dukas R. 2005 Learning affects mate choice in female fruit flies. Behavioral Ecology 16, 800– 804. Elias DO, Andrade MCB, Kasumovic MM. 2011 Dynamic population structure and the evolution of spider mating systems. Advances in Insect Physiology 41, 65–114. Fawcett TW, Frankenhuis WE. 2015 Adaptive explanations for sensitive windows in development. Frontiers in Zoology 12, S3. Fowler‐Finn KD, Rodríguez RL. 2012 Experience‐mediated plasticity in mate preferences: mating assurance in a variable environment. Evolution: International Journal of Organic Evolution 66, 459–468. Gwynne DT, Bailey WJ, Annells A. 1998 The sex in short supply for matings varies over small spatial scales in a katydid (Kawanaphila nartee, Orthoptera: Tettigoniidae). Behavioral Ecology and Sociobiology 42, 157–162. Hebets EA. 2003 Subadult experience influences adult mate choice in an arthropod: exposed female wolf spiders prefer males of a familiar phenotype. Proceedings of the National Academy of Sciences 100, 13390–13395.

Hebets EA, Vink CJ. 2007 Experience leads to preference: experienced females prefer brush- legged males in a population of syntopic wolf spiders. Behavioral Ecology 18, 1010– 1020. Herberstein ME, Painting CJ, Holwell GI. 2017 Scramble competition polygyny in terrestrial arthropods. Advances in the Study of Behavior 49, 237–295. Heubel KU, Lindström K, Kokko H. 2008 Females increase current reproductive effort when future access to males is uncertain. Biology Letters 4, 224–227. Jennions MD, Petrie M. 1997 Variation in mate choice and mating preferences: a review of causes and consequences. Biological Reviews 72, 283–327. Jennions MD, Petrie M. 2000 Why do females mate multiply? A review of the genetic benefits. Biological Reviews 75, 21–64. Johnson JC. 2005 Cohabitation of juvenile females with mature males promotes sexual cannibalism in fishing spiders. Behavioral Ecology 16, 269–273. Johnson, JC, Trubl P, Blackmore V, Miles L. 2011 Male black widows court well-fed females more than starved females: silken cues indicate sexual cannibalism risk. Animal Behaviour 82, 383–390. Kaston BJ. 1970 Comparative biology of American black widow spiders. Transactions of the San Diego Society of Natural History 16, 33–82. Kasumovic MM, Andrade MC. 2004. Discrimination of airborne pheromones by mate-searching male western black widow spiders (Latrodectus hesperus): species-and population- specific responses. Canadian Journal of Zoology 82, 1027–1034. Kasumovic MM, Andrade MCB. 2006 Male development tracks rapidly shifting sexual versus natural selection pressures. Current Biology 16, R242–R243. Kasumovic MM, Brooks RC, Andrade MCB. 2009 Body condition but not dietary restriction prolongs lifespan in a semelparous capital breeder. Biology Letters 5, 636–638. Kasumovic MM, Brooks RC. 2011 It's all who you know: the evolution of socially cued anticipatory plasticity as a mating strategy. The Quarterly Review of Biology 86, 181– 197. Kasumovic MM, Bruce MJ, Andrade MC, Herberstein ME. 2008 Spatial and temporal demographic variation drives within‐season fluctuations in sexual selection. Evolution: International Journal of Organic Evolution 62, 2316–2325. Kasumovic MM, Hall MD, Brooks RC. 2012 The juvenile social environment introduces variation in the choice and expression of sexually selected traits. Ecology and Evolution 2, 1036–1047. Kelly CD. 2018 The causes and evolutionary consequences of variation in female mate choice in insects: the effects of individual state, genotypes and environments. Current Opinion in Insect Science 27, 1–8. Kokko H, Mappes J. 2005 Sexual selection when fertilization is not guaranteed. Evolution 59, 1876–1885. Kosmidis I. 2019 brglm: Bias Reduction in Binary-Response Generalized Linear Models. R package version 0.6.2. See https://cran.r-project.org/package=brglm

Lim H, Greenfield MD. 2006 Female pheromonal chorusing in an arctiid moth, Utetheisa ornatrix. Behavioral Ecology, 18, 165–173. Lively CM. 1986 Canalization versus developmental conversion in a spatially variable environment. The American Naturalist 128, 561–572. MacLeod EC. 2013. New insights in the evolutionary maintenance of male mate choice behaviour using the western black widow, Latrodectus hesperus. Doctoral dissertation, University of Toronto. MacLeod EC, Andrade MCB. 2014. Strong, convergent male mate choice along two preference axes in field populations of black widow spiders. Animal Behaviour 89, 163–169. Milinski M, Bakker TC. 1992 Costs influences sequential mate choice in sticklebacks, Gasterosteus aculeatus. Proceedings of the Royal Society B: Biological Sciences 250, 229–233. Nessler SH, Uhl G, Schneider JM. 2009a Sexual cannibalism facilitates genital damage in Argiope lobata (Araneae: Araneidae). Behavioral Ecology and Sociobiology 63, 355– 362. Nessler SH, Uhl G, Schneider JM. 2009b Scent of a woman–the effect of female presence on sexual cannibalism in an orb‐weaving spider (Araneae: Araneidae). Ethology 115, 633– 640. Pigliucci, M. 2001 Phenotypic plasticity: beyond nature and nurture. Baltimore, MD: John Hopkins University Press. Pitcher TE, Neff BD, Rodd FH, Rowe L. 2003 Multiple mating and sequential mate choice in guppies: females trade up. Proceedings of the Royal Society B: Biological Sciences 270, 1623–1629. Pompilio L, Franco MG, Chisari LB, Manrique G. 2016 Female choosiness and mating opportunities in the blood-sucking bug Rhodnius prolixus. Behaviour 153, 1863–1878. Prenter J, MacNeil C, Elwood RW. 2006 Sexual cannibalism and mate choice. Animal Behaviour 71, 481–490. Punzalan D, Rodd FH, Rowe L. 2010 Temporally variable multivariate sexual selection on sexually dimorphic traits in a wild insect population. The American Naturalist 175, 401– 414. Rehermann G, Altesor P, McNeil JN, González A. 2016 Conspecific females promote calling behavior in the noctuid moth, Pseudaletia adultera. Entomologia Experimentalis et Applicata 159, 362–369. Reynolds JD, Gross MR. 1990 Costs and benefits of female mate choice: is there a lek paradox?. The American Naturalist 136, 230–243. Roff DA, Fairbairn DJ. 2015 Bias in the heritability of preference and its potential impact on the evolution of mate choice. Heredity 114, 404–412. Ross K, Smith RL. 1979 Aspects of the courtship behavior of the black widow spider, Latrodectus hesperus (Araneae: Theridiidae), with evidence for the existence of a contact sex pheromone. Journal of Arachnology 7, 69–77.

Rowe L. 1994 The costs of mating and mate choice in water striders. Animal Behaviour 48, 1049–1056. RStudio Team 2018. RStudio: Integrated Development for R. Boston, MA: RStudio, Inc. See http://www.rstudio.com Sachser N, Kaiser S, Hennessy MB. 2013 Behavioural profiles are shaped by social experience: when, how and why. Philosophical Transactions of the Royal Society B: Biological Sciences 368, 20120344. Salomon M, Vibert S, Bennett RG. 2010 Habitat use by western black widow spiders (Latrodectus hesperus) in coastal British Columbia: evidence of facultative group living. Canadian Journal of Zoology 88, 334–346. Schwartz SK, Wagner WE, Hebets EA. 2016 Males can benefit from sexual cannibalism facilitated by self-sacrifice. Current Biology 26, 2794–2799. Scott CE, Anderson AG, Andrade MCB. 2018 A review of the mechanisms and functional roles of male silk use in spider courtship and mating. The Journal of Arachnology 46, 173– 207. Scott C, Kirk D, McCann S, Gries G. 2015a Web reduction by courting male black widows renders pheromone-emitting females' webs less attractive to rival males. Animal Behaviour 107, 71–78. Scott C, McCann S, Gries R, Khaskin G, Gries G. 2015b. N-3-methylbutanoyl-O- methylpropanoyl-L-serine methyl ester–pheromone component of western black widow females. Journal of Chemical Ecology 41, 465–472. Scott CE, McCann S, Andrade MCB. 2019. Male black widows parasitize mate-searching effort of rivals to find females faster. Proceedings of the Royal Society B: Biological Sciences 286, 20191470. Shelly TE, Bailey WJ. 1992 Experimental manipulation of mate choice by male katydids: the effect of female encounter rate. Behavioral Ecology and Sociobiology 30, 277–282. Simmons LW, Kvarnemo C. 2005 Costs of breeding and their effects on the direction of sexual selection. Proceedings of the Royal Society B: Biological Sciences 273, 465–470. Snell-Rood EC. 2013 An overview of the evolutionary causes and consequences of behavioural plasticity. Animal Behaviour 85, 1004–1011. Stoltz JA, Andrade MCB, Kasumovic MM. 2012 Developmental plasticity in metabolic rates reinforces morphological plasticity in response to social cues of sexual selection. Journal of Insect Physiology 58, 985–990. Stoltz JA, Hanna R, Andrade MCB. 2010 Longevity cost of remaining unmated under dietary restriction. Functional Ecology 24, 1270–1280. Swanger E, Zuk M. 2015 Cricket responses to sexual signals are influenced more by adult than juvenile experiences. Journal of Insect Behavior 28, 328–337. Watson PJ. 1986 Transmission of a female sex pheromone thwarted by males in the spider Linyphia litigiosa (Linyphiidae). Science 233, 219–221. Welke KW, Schneider JM. 2012 Sexual cannibalism benefits offspring survival. Animal Behaviour 83, 201–207.

West-Eberhard MJ. 2003 Developmental plasticity and evolution. Oxford, UK: Oxford University Press. Westerman EL, Drucker CB, Monteiro A. 2014 Male and female mating behavior is dependent on social context in the butterfly Bicyclus anynana. Journal of Insect Behavior 27, 478– 495.

Table 3-1 Summary of the number of males arriving at cages of L. hesperus females before and after they moulted to maturity, during a 3-month field experiment. Sample sizes include only spiders that matured by the end of the experiment.

Before maturity After maturity

Mean males Mean males Treatment n Total males Total males per female per female

Isolated 15 3 0.2 50 3.3

Clustered 17 0 0 64 3.8

Total 32 3 0.1 114 3.6

Response variable Predictor variable Estimate SE Z P

Copulation (yes/no)

density treatment 3.49 1.79 2.01 0.044

days since moult 0.08 0.07 1.10 0.272

number of males 0.09 0.13 0.71 0.481

Pre-copulatory cannibalism (yes/no)

density treatment -2.05 1.44 -1.42 0.155

days since moult -0.03 0.07 -0.42 0.674

number of males -0.002 0.11 -0.02 0.987

Figure 3-1 Photograph of a portion of our field site showing the configuration of a study examining the effect of proximity to conspecific females on experimental females during subadulthood and early adulthood. At this site, driftwood logs (seen in centre foreground and at right of main image) provide microhabitats for black widows. Pairs of females were placed at 10-m intervals along a 190-m transect. One female of each pair was isolated (at least 10 m from any naturally occurring females) and the other clustered (within 1 m of a microhabitat containing at least one other female). Each experimental female was housed in a screen cage and placed under an artificial microhabitat (plywood shed) designed to mimic a log (inset, with shed upturned).

Figure 3-2 Males arrived earlier and more males arrived in rapid succession at cages of clustered females compared to isolated females during our field experiment. Histograms show the day on which the first male (or males) arrived (a, c) and the number of males that arrived within two days after the date of the first male’s arrival (b, d) at cages of L. hesperus female who were either clustered near conspecific females (a, b) or isolated (c, d). Vertical red lines show medians.

Figure 3-3 Following the field experiment, females from the clustered treatment rejected males more often than isolated females in staged mating trials. (a) The number of females who copulated at least once was greater for isolated than clustered females. (b) Fewer isolated females cannibalized males, and only clustered females ever engaged in pre-copulatory cannibalism.

A review of the mechanisms and functional roles of male silk use in spider courtship and mating

Catherine E Scott, Alissa G Anderson, and Maydianne CB Andrade

Abstract

Spiders are well known for using chemical, vibratory, tactile, and visual signals within mating contexts. All spiders produce silk, and even in non-web building spiders, silk is intimately tied to courtship and mating. Silk produced by females provides a transmission channel for male vibratory courtship signals, while webs and draglines provide a substrate for female sex pheromones. Observations of male spiders producing silk during sexual interactions are also common across phylogenetically widespread taxa. However, the function of male-produced silk in mating has received very little study. Exploring the function of male silk use during mating will provide a deeper understanding of the complex mating systems of spiders and allow tests of hypotheses about the evolution of male and female traits under sexual selection and/or conflict. In this review, we outline functional hypotheses that may explain each of the following three main categories of silk deposition males exhibit during courtship and mating: (1) silk deposition on females’ webs or other silk structures, (2) silk deposition on females (‘bridal veils’) and (3) silk associated with nuptial gifts. We then summarize the current knowledge of silk use by male spiders within these three categories and the types of mechanisms that may lead to functional effects, and discuss areas where future work can be targeted.

4.1 Introduction

4.1.1 Overview

Mating in animals that are generally solitary, like spiders, necessarily involves a number of shifts in behavior to facilitate locating, approaching, and mating with the opposite sex (Elias et al. 2011; Schneider & Andrade 2011). These shifts provide interesting opportunities to test general aspects of theory related to communication (Hauser 1996; Bradbury & Vehrencamp 2011; Herberstein et al. 2014), mate choice (Bateson 1983), sexual conflict (Arnqvist & Rowe 2005), and sexual selection (Andersson 1994). In many species, males use a range of behavioral, morphological, sensory, and physiological traits when approaching females to seek matings

90 91

(Andersson 1994). These traits may enhance the success of the male through their effects on the behavior of potential mates or rivals. For example, females’ mating decisions may be based on the nature or intensity of male courtship displays or ornaments if these reflect desirable (female- fitness-enhancing) characteristics in a potential mate (Bateson 1983; Andersson 1994; Jennions & Petrie 1997; York & Baird, 2017). Moreover, in many taxa, courtship is a public event (Herberstein et al. 2002), vulnerable to interruption or interference by other males (e.g., Hibler & Houde 2006; Stoltz & Andrade 2009). Sexual selection on males to achieve matings is often intense and may lead to the evolution of remarkable adaptations to overcome competition or persuade females to mate (Andersson 1994). In addition, it is now clear that males may adjust investment in courtship as a function of the perceived fitness payoff (quality or risk) associated with approaching or mating with a given female (Johnson et al. 2011; Moskalik & Uetz 2011; Lane et al. 2015; McGhee et al. 2015; Rundus et al. 2015; Cross 2016; Rypstra et al. 2016). In cannibalistic spiders, there is an added dimension of risk to the male associated with approaching the wrong female (Herberstein et al. 2002; Johnson et al. 2011; Kralj-Fiser et al. 2016). The result is a rich interplay of male and female fitness interests that may be intertwined in different ways at different stages of courtship.

In general, spiders offer interesting opportunities for studying the ‘mating dance’ between the sexes in detail, since mating behavior varies considerably among taxa (Schneider & Andrade 2011), and courtship often includes multimodal communication (e.g., visual, vibratory, chemical, and tactile; Witt & Rovner 1982). Silk is a tangible, measurable, and manipulable medium that can convey information in all of these modalities and thus has been frequently studied in this context—but almost exclusively from the perspective of females. Thus, it is well known that female silk plays a central role in many aspects of communication and mating outcomes across spider taxa (Gaskett 2007; Elias & Mason 2010; Uhl & Elias 2011; Schulz 2013). In addition to its communicative role, for many spider species female silk is the substrate on which mating interactions occur (Foelix 2011) and this also has implications for its functional role.

What is less well known is the variety of ways in which male silk may mediate sexual interactions in spiders. In this review, we highlight accumulating evidence, from a variety of spider taxa, that male silk also has a significant role in mate attraction, courtship, and mating. We outline the ways in which male silk is used in these interactions and suggest tractable

92 approaches to testing a range of hypotheses to explain the evolution of male silk use in terms of sexual selection and/or sexual conflict. Finally, we identify, where possible, taxa where additional study may be particularly illuminating, both in terms of our understanding of silk use in spiders, and in terms of a more general understanding of male and female mating tactics.

We start by briefly summarizing some salient features of spider silk, and the well-known uses of silk by females during mating interactions. We then provide an overview of the main functional hypotheses for male silk use in mating, and the mechanisms that may lead to functional outcomes. We follow this with a description of the ways males use silk in mating, split into three broad categories (silk deposition on the female’s web or other silk structures, silk deposition on the female, and silk associated with nuptial gifts), with examples from a range of taxa. Each category ends with a qualitative evaluation of the hypotheses given the available data. Finally, we briefly discuss a few other ways males use silk in mating interactions (e.g., sperm webs).

4.1.2 Properties of spider silk

Spider silk consists of protein-based fibers, is energetically costly to produce (Peakall & Witt 1976; Prestwich 1977), is unique in its combination of strength and elasticity, and is one of the toughest known biological substances (Gosline et al. 1999; Rising et al. 2011). The physical properties of spider silk vary among taxa and among contexts within taxa (Craig 2003). Years of research show the biophysical properties of silk are strongly dependent on links between ecological context, evolutionary history (e.g., Wolff et al. 2017), and the physiology of the spider at the time of silk production (Blamires et al. 2017). This may explain why, despite recent progress (Rising et al. 2011; Hsia et al. 2012), attempts to develop industrial production methods to synthesize spider silk have been challenging (Kluge et al. 2008; Koeppel & Holland 2017). Features such as the silk tensile strength and elasticity depend on which glands are used to produce the silk (e.g., aciniform, ampullate, flagelliform, tubuliform, or piriform) and how it is extruded, which varies with use (e.g., egg sacs, structural web silk, capture silk, or drag lines; Vollrath & Knight 2001; Foelix 2011).

Spiders produce silk at every life history phase, and in most cases, leave silk draglines behind them as they move through their habitat (Foelix 2011). Thus, most spider behaviors have

93 the potential to create and leave behind information associated with silk in a variety of modalities. Variation in reflectance properties (Blackledge & Wenzel 2000; Barrantes et al. 2013) and color (e.g., Craig et al. 1996) of silk can have implications for visibility or attractiveness to prey, predators and conspecifics under a variety of light conditions. However, functional effects have primarily been investigated with regard to predator/prey dynamics (e.g., Craig & Barnard 1990; Blackledge 1998; Persons & Rypstra 2001; Rypstra & Buddle 2013; Bucher et al. 2014; Lai et al. 2017). The structure of silk makes it well-suited as a delivery vehicle for contact or airborne semiochemicals (Gaskett 2007; Schulz 2013; Henneken et al. 2017a) with interesting implications for function arising from variation in how long pheromones remain active after deposition, and whether rain (i.e., a polar solvent) can wash them away (e.g., water soluble pheromones in some wolf spiders: Dondale & Hegdekar 1973; Tietjen 1977; Baruffaldi et al. 2010; water-resistant pheromones in fishing spiders and some wolf spiders: Roland & Rovner 1983; Lizotte & Rovner 1989). Finally, silk also serves as a medium for vibratory communication and detection of vibratory cues, particularly in web-building species (Uhl & Elias 2011).

4.1.3 Female silk and mating

As a substrate for pheromones, transmission of vibratory signals, and the structure on which mating may take place, female silk is well known for its role within courtship and mating contexts in web-building spiders (Locket 1926; Foelix 2011). Webs or other silk structures (e.g., the silk associated with burrows) provide the stage for vibratory courtship displays by males in many spider taxa in which females are sedentary (Uhl & Elias 2011). Similarly, among cursorial spiders, female drag-line silk provides information to conspecific males (Bristowe & Locket 1926; Kaston 1936; Anderson & Morse 2001; Nelson et al. 2012; Rundus et al. 2015; Bell & Roberts 2017). Draglines can convey chemical (chemo-tactile; e.g., Nelson et al. 2012) or tactile (mechanical; e.g., Anderson & Morse 2001; Leonard & Morse 2006) information about the location and identity of the signaler.

4.1.3.1 Substrate for pheromones

Spiders are predatory and generally solitary, and thus face the challenge of attracting or finding mates. Behavioral evidence indicates that sex pheromones provide the solution to this problem in

94 many spiders (Gaskett 2007; Uhl & Elias 2011; Trabalon 2013). Indeed, since chemical signaling is the most ancient form of communication (Wyatt 2014), sex pheromone production is likely ubiquitous in spiders. Pheromones associated with female silk include those that release volatile, airborne chemicals, and those that require contact by the receiver (Gaskett 2007; Schulz 2013).

Airborne sex pheromones typically attract mates at long range and may also reveal information about the identity and quality of the signaler (Gaskett 2007; Uhl & Elias 2011; Uhl 2013). Volatile, attractive sex pheromones have been identified from the bodies and/or silk of females in only three species: Argiope bruennichi (Scopoli, 1772) (Araneidae; Chinta et al. 2010), Agelenopsis aperta (Gertsch, 1934) (Agelenidae; Papke et al. 2001), and Pholcus beijingensis Zhu & Song, 1999 (Pholcidae; Xiao et al. 2009). Although these chemically identified pheromones come from web-builders, cursorial spiders including lycosids and salticids also produce volatile sex attractants associated with their bodies and/or silk (e.g., Searcy et al. 1999; Nelson et al. 2012). Behavioral studies with Latrodectus Walckenaer, 1805 spp. (Theridiidae) indicate that volatile, silk-borne female pheromones allow males to discriminate between females of different age, mating status, body condition, and population of origin (Kasumovic & Andrade 2004; Andrade & Kasumovic 2005; MacLeod & Andrade 2014).

Silk-borne contact pheromones elicit male searching and courtship behavior in both cursorial and web-building spiders (e.g., Tietjen 1978; Suter & Renkes 1982; Taylor 1998), and may provide information about female identity, mating status, receptivity, diet, gravidity, and reproductive potential (Riechert & Singer 1995; Roberts & Uetz 2005; Baruffaldi & Costa 2010; Trabalon 2013; Henneken et al. 2015, 2017b). Contact sex pheromones have been identified from the silk of female spiders in four families: Linyphia triangularis (Clerck, 1757) (Linyphiidae; Schulz & Toft 1993), Latrodectus hasselti Thorell, 1870 and L. hesperus Chamberlin & Ivie, 1935 (Theridiidae; Jerhot et al. 2010; Scott et al. 2015a), Eratigena atrica (C.L. Koch, 1843) (Agelenidae; Prouvost et al. 1999), and Cupiennius salei (Keyserling, 1877) (Ctenidae; Papke et al. 2000).

4.1.3.2 Substrate for transmission of vibratory signals

Substrate-borne vibrations are extremely important for spiders (Barth 2002; Elias & Mason 2010), which are highly sensitive to vibrations detected via receptors on their legs (Barth 1982;

Foelix 2011). Spiders that build webs or snares or simply extend the silk lining of their burrow can essentially expand their field of sensory perception and create their own specialized signaling environments (Elias & Mason 2010; Krafft & Cookson 2012). The silk in these contexts transmits vibrations both from prey and courting males. The types of vibratory behaviors in spiders include percussion, stridulation, and tremulation, and these may transmit seismic and/or near-field airborne vibratory signals (reviewed in Uhl & Elias 2011). Vibratory courtship signals produced on webs have been recorded in only a small number of studies (Masters & Markl 1981; Masters 1984; Suter & Renkes 1984; Naftilan 1999; Wignall & Herberstein 2013a; Vibert et al. 2014). However, our understanding of the biomechanical properties of spider silk with respect to vibration transmission has expanded rapidly in recent years (primarily for orb-webs, e.g., Landolfa & Barth 1996; Watanabe 2000; Alam et al. 2007; Mortimer et al. 2014, 2015, 2016). For example, Mortimer and colleagues (2016) examined trade-offs between signal transmission and the structure of orb-webs; their work led them to conclude that silk tension and stiffness can affect vibration amplitude. This led them to the intriguing suggestion that females could construct webs to optimally balance multiple signal transmission functions (Mortimer et al. 2016).

4.1.3.3 Structural effects on courtship & mating activity

In web-building spiders, the female’s web and/or retreat is often the location of courtship and mating. Thus, the structure of the web or retreat may constrain the type of courtship, approach vector, or mobility of males. For example, in some species females rest with their genital opening in close proximity to dense silk sheets such that mating requires a postural change (e.g., Latrodectus; Andrade & MacLeod 2015). Similarly, mating by non-web-building spiders may take place inside the female’s burrow or silk retreat, where the movement of males and females is constrained (e.g., Phidippus C.L. Koch, 1846; Hoefler 2007). To our knowledge, there has been no investigation of female web or retreat structure in relation to mobility during mating.

4.2 Male silk and mating

4.2.1 Overview

Male spiders from diverse and distantly related families use silk during courtship and mating (Fig. 1). We review the main contexts in which males use silk in mating, with an emphasis on

96 what is known about the effects of silk on males and/or females (Table 1). We start with a brief overview of the general types of effects that are recurrent themes, which suggest a number of (non-exclusive) hypotheses about male silk use in mating. We then arrange existing data on male silk use during mating into three main categories: (I) silk addition to females’ webs (with or without web reduction; Table 2), (II) application of silk to female mating partners (‘bridal veils’; Table 3), and (III) presentation of silk associated with nuptial gifts (or silk itself) to females (Table 4). For each type of silk use, we end by considering the specific types of effects predicted by each hypothesis and suggest where additional study would be fruitful. We then briefly discuss other ways silk is used by males during mating interactions that do not fall into these categories (Table 5). Finally, we provide general conclusions and suggestions for future directions.

4.2.2 Fitness effects of silk use

There are a number of different hypotheses for the function of male silk use during mating interactions (Table 1). These are not mutually exclusive; as male silk use could have multiple functions in a given species. We consider these in terms of the way in which the behavior may increase the fitness of the silk-laying male, and the mechanism that leads to effects on fitness.

4.2.2.1 Fitness effects—current mating

Silk may increase a male’s fitness if it increases his mating success with a given female (see columns 1–3 in Table 1). We consider three ways males might use silk to increase their fitness during interactions with a potential mate. (1) Silk may increase the likelihood of copulation by increasing or accelerating female receptivity to mating (i.e., affect female preference). More receptive females may also copulate more quickly. Rapid copulation may decrease the risk of interference by other males or by predators, and/or reduce the energetic investment in courtship. (2) Silk use may increase male fitness if it increases sperm transfer via longer or more frequent copulations (total copulation duration is linked to paternity or fertilization success in some species; Andrade 1996; Elgar et al. 2000; Anderson & Hebets 2017). We predict these functions would have the most significant effects on male fitness in species in which females are very choosy, courtship is costly and possibly prolonged, where rival males commonly approach females that are being courted by other males, and/or where copulation frequency or duration is related to paternity or fertilization success. (3) Silk use might increase male fitness by reducing the risk of male injury or death. These effects may arise through decreased risk of sexual

97 cannibalism or attacks by rival males. Clearly, these effects would be most important in species where females frequently attack males during courtship or copulation, and/or where direct or escalated inter-male competition is coincident with courtship and mating attempts.

4.2.2.2 Fitness effects—decreased polyandry

Males may also benefit from silk use through a reduced risk of polyandry and thus a reduced risk of losing paternity to rival males (see columns 4–5 in Table 1). We divide this fitness benefit into two categories. (4) Silk use may decrease the risk of rival males courting or mating by interfering with the female’s attractiveness to rivals, causing ineffective courtship by rivals, or decreasing physical access to females for rival males. (5) Alternatively, male silk use may decrease the likelihood of females remating by decreasing the receptivity of mated females. These effects would be most significant in species where females typically have the opportunity for polyandry, where sperm of rival males mix or last-male sperm precedence is relatively common (so males that mate after the first male will secure some paternity).

4.2.3 Mechanisms of effect

In terms of mechanisms of effect (Table 1), fitness benefits of silk use can be derived indirectly through communication, where beneficial changes in the behavior of females or rival males arise from assessment of the information content of silk use. Information may be encoded in chemical, visual, tactile, or vibratory modalities. This may involve assessment of qualities of the male silk in itself (chemical, visual, and/or tactile modalities), or assessment of the performance of male behaviors involved in silk addition (vibratory and/or visual modalities). Finding evidence for fitness consequences of silk addition, or understanding the implications in terms of vibrations, may be less challenging than unravelling underlying mechanism(s) related to chemical or tactile signalling. Demonstrating the presence of pheromones on silk and differentiating between chemical and tactile cues requires carefully designed experiments (e.g., Anderson & Morse 2001). While identifying spider pheromones remains a challenge, other approaches include temporary or permanent ablation of the female’s tactile or chemosensory receptors (Zhang et al. 2011; Aisenberg et al. 2015).

Fitness benefits may also derive directly through physical effects of male silk addition to the female, her web, or a nuptial gift. This may arise if structural changes to the web affect

98 courtship and mating mobility, if silk acts as a physical constraint on movement, or if it functions to hide the contents of a nuptial gift (deception). In this category, we also include effects of silk addition on efficacy of communication that may arise through structural changes to the web.

4.2.3.1 Indirect effects—chemical signals/cues

Behavioral evidence suggests that, like female silk, silk produced by males can transmit chemical information as part of intersexual communication. Only one male spider pheromone has been chemically identified to date, an aphrodisiac isolated from whole-body extracts of Pholcus beijingensis (Pholcidae) males (Xiao et al. 2010). However, behavioral evidence demonstrates or supports the existence of silk-borne male sex pheromones in seven families (Fig. 1; Table 6). Like female pheromones, these putative male pheromones have diverse functions. Contact with male silk elicits courtship behavior in female L. hesperus (Theridiidae; Ross & Smith 1979). A pheromone from the male’s body and/or silk induces quiescence in female Agelenopsis aperta (Agelenidae; Becker et al. 2005). Tactile and/or chemical information on male silk facilitates orientation in female Tegenaria domestica (Clerck, 1757) and Coelotes terrestris (Wider, 1834) (Agelenidae; Roland 1983). Similarly, in the lycosid spider Pardosa milvina (Hentz, 1844) females discriminate between silk of courting and non-courting males, increasing their own silk production in response to contact with the male’s silk (Khan & Persons 2015). Airborne pheromones from the bodies and silk of Scytodes Latreille, 1804 sp. (Scytodidae) and Evarcha culicivora Wesolowska & Jackson, 2003 (Salticidae) males function in mate recognition and mate choice (Cross & Jackson 2009; Koh et al. 2009).

Pheromones on male silk may also be important for intrasexual communication (Table 7), including assessment of male-male competition. Airborne chemical cues from Latrodectus hasselti (Theridiidae) males and/or their silk provide information about the competitive environment and trigger shifts in development in other males (Kasumovic & Andrade 2006). Male Nephila senegalensis (Walckenaer, 1841) (Araneidae) use silk cues left behind by rival males to choose which females’ webs to visit. They avoid webs previously visited by another male, irrespective of the female’s quality (Schneider et al. 2011). Male courtship behavior is inhibited by a pheromone that can be extracted with methanol from the silk of Schizocosa ocreata (Hentz, 1844) (Lycosidae) males (Ayyagari & Tietjen 1987), and Frontinella communis

(Hentz, 1850) (Linyphiidae) males respond to compounds on male cuticle with aggressive behavior (Suter et al. 1987).

4.2.3.2 Indirect effects—visual signals/cues

For spider species with well-developed vision, male silk could play a role in visual signaling, or provide cues about the state of the male that produced the silk, and thus affect female receptivity or choice. For example, the density of silk male pisaurids and trechaleids use to wrap nuptial gifts (see section 4 below), affects the color of the gift (bright white to dark grey), and may provide information about the physiological state of the gift-giving male (Stalhandske 2002; Albo et al. 2011a; Trillo et al. 2014). White silk in itself may be highly visible, and so attract the attention of females, as has been shown in the crepuscular Paratrechalea ornata (Mello-Leitao, 1943) (Stalhandske 2002; Trillo et al. 2014).

4.2.3.3 Indirect effects—tactile cues

When silk comes in contact with females directly, or when females manipulate, touch, or move across silk laid down by males, that contact may provide information. For example, tactile cues on silk draglines allow male crab spiders to follow females, with recognition depending on mechanical characteristics of the silk (Anderson & Morse 2001).

4.2.3.4 Indirect effects—correlated effects of silk-laying behaviors

The behaviors associated with silk deposition may have functions independent of the silk itself. For many species, typical male abdominal movements have been described in association with silk application (see ‘abdomen waggle’ in Table 2), and these may produce vibrational or visual signals. In an analogous example, Vollrath (1979) showed that prey wrapping by web-building spiders creates a characteristic pattern of vibrations that are exploited by the kleptoparasite Argyrodes elevatus Taczanowski, 1873. Here we focus primarily on the way in which male silk itself may affect mating outcomes, but also outline behaviors that are reliably associated with silk deposition where relevant.

4.2.3.5 Direct effects—structural modifications

Males of many species modify the webs or other silk structures of females, and this may involve the use of male silk in various ways, including covering or wrapping females’ silk, or adding

100 new silk lines to existing structures. Modifications to web structure are traditionally described in terms of how they affect the behavior or possible movement (mobility) of the female and/ or rival males during a mating attempt. However, these structural changes may also affect the nature, directionality or efficacy of vibrational or chemical signals or cues. Thus, changes in signal transmission may be the primary mechanism by which silk use affects male fitness.

4.2.3.6 Direct effects—physical effects

Males produce a range of different types of silks, and comparable to the use of silk in prey- capture, male silk may be applied directly to the female in such a way that it restrains, impedes or slows the movement of females or even of rival males (physical constraints). Silken constructions may also support or adjust the posture of females in a way that facilitates genital coupling. Alternatively, silk wrapped around nuptial gifts may allow males to hide their contents when the gift is of low nutritional value (deceit).

4.3 Silk deposition on females’ webs or other silk structures

4.3.1 Overview and description of behaviors

In web-dwelling spiders, mating generally takes place on the female’s web or in her retreat. During courtship, males in several families representing the full range of web architectures lay silk on the female’s web, leading to modification of existing web structure (Figs. 1 & 2). Web modification with silk addition varies from the addition of a single line (a mating thread) to destruction of large areas of the female’s web and replacement with male silk (web reduction). Similarly, in burrow-dwellers like some mygalomorph and lycosid spiders, silk lines the burrow and may extend from its entrance, providing the substrate on which courtship occurs. Males of these taxa may also deposit silk onto the female’s silk during mating interactions, although modification of the overall architecture of the female’s silk structures has not been reported. Descriptions of silk-spinning behavior of males when courting on webs or on other silk structures have many similarities, and thus we consider them together in this section.

Among the Mygalomorphae, silk deposition by males during courtship has been reported for the web-building Dipluridae and Porrhothelidae (formerly Hexathelidae) and the burrow- dwelling Theraphosidae. In the diplurid spiders Thelechoris karschi (Simon, 1889) and Microhexura montigava Crosby & Bishop, 1925, both males and females spin silk as they move

101 about the web during courtship (Coyle 1985; Coyle & O’Shields 1990). Porrhothele antipodiana (Walckenaer, 1837) (Porrhothelidae) males spin silk during interactions with females on their webs (both before and after copulation), and also during interactions with other males. Silk spinning behavior in this species is accompanied by obvious lateral movements of the abdomen (Jackson & Pollard 1990). In the burrow-dwelling theraphosid spiders Grammostola vachoni Schiapelli & Gerschman, 1961 and Brachypelma klaasi (Schmidt & Krause, 1994), courting males lay down silk over the female’s silk around the burrow entrance (Yanez et al. 1999; Ferretti & Ferrero 2008).

Descriptions of silk deposition in some araneomorph spiders that build sheet webs are similar to those for the Mygalomorphae. Upon contact with the web of a virgin female, the crevice weaver Kukulcania hibernalis (Hentz, 1842) (Filistatidae) pulls swaths of silk threads from his spinnerets with his last pair of legs and deposits them on her web (Barrantes & Ramirez 2013). Less obvious silk deposition occurs during the courtship of the funnel weaver Eratigena agrestis (Walckenaer, 1802) (Agelenidae). Males deposit silk as they move around on the female’s web, periodically anchoring it to the sheet (C.E. Scott, pers. obs.). Lateral ‘abdomen wagging’ behavior is associated with silk deposition in E. agrestis, and this behavior is common during the courtship of several other agelenids, most notably the genus Agelenopsis C.L. Koch, 1837 (Table 2; Galasso 2012). This ‘wagging’ that occurs as males move around the web is usually accompanied by silk emission in these agelenids (S. Riechert, pers. comm.).

In orb-weavers (family Araneidae, including subfamily Nephilinae, formerly Nephilidae), courtship is grouped into three types (A–C), two of which involve male alteration of web architecture (Robinson & Robinson 1980). Type A courtship occurs on the female’s web and typically involves addition of silk near the hub. In type B courtship, the male cuts a hole in the web close to the hub and constructs a mating thread across it. Type C courtship does not involve any web cutting; the male constructs a mating thread that he attaches to the periphery of the web. In both type B and C courtship, the male engages in vibratory courtship on the mating thread (which may be multi-stranded; Table 2). Eventually the female joins the male on the mating thread, where copulation takes place. Typically, males of a given species use one type of courtship, but in Argiope argentata (Fabricius, 1775), for instance, males switch types depending on context (Robinson & Robinson 1980). Some males do type A, and others type B, or the same male might do both types on different days. If another male is already courting on the web (type

102

A or type B courtship), a second male will engage in type C courtship with a mating thread attached to the periphery of the web. Interestingly, when engaging in type B courtship, the male may add dragline silk to his mating thread after a courtship bout to which the female did not respond. Similarly, in Argiope picta L. Koch, 1871, males may enlarge the hole across which they spun their mating thread between bouts of unsuccessful courtship (Robinson & Robinson 1980).

In Isoxya tabulata (Thorell, 1859) and Scoloderus cordatus (Taczanowski, 1879) the mating thread is employed in a different way. The male situates himself such that the female walks onto a silk line still attached to his spinnerets, which he pays out as she tries to approach, resulting in a ‘treadmill effect’ (Robinson & Robinson 1980). The female attempts to walk along the line but makes no progress, rather she accumulates a bundle of the male’s silk below her cephalothorax (this silk may constitute a nuptial gift; see section 5).

Male cobweb weavers (Theridiidae) also commonly construct mating threads during courtship on the female’s web (Table 2). As in type C courtship of the orb-weavers, the male installs a silk line and then engages in vibratory courtship on it until the female eventually moves onto the thread, where copulation occurs (Knoflach 2004). In some species, the male reinforces the thread several times, or he constructs a larger area of threads referred to as a mating web, which is used similarly to mating threads (Knoflach 2004). In a few species, the male cuts some of the female’s threads, but in general, theridiids modify the web by adding their own silk without excising sections of the female’s web (Knoflach 2004). Exceptions include the widow and false widow spiders Latrodectus and Steatoda Sundevall, 1833, which engage in extensive web reduction behavior (discussed below) and males of the social theridiid spider Parasteatoda wau (Levi, Lubin & Robinson, 1982), which build courtship ‘arenas’ in their communal webs by cutting out small areas of the barrier web threads and laying down one or more of their own threads (Lubin 1986). Courtship occurs on these threads and they are considered functionally equivalent to the mating threads of araneid spiders (Lubin 1986).

Variations on the theme of mating threads and webs can also be found in cribellate web- dwellers (Table 2). In Fecenia Simon, 1887 sp. (Psechridae), which constructs an orb-web, the courting male cuts away most of the web, leaving only a single thread on which courtship and mating proceed (Robinson & Lubin 1979). Males of the meshweaver Dictyna arundinacea

103

(Linnaeus, 1758) (Dictynidae) cut a hole in the web and construct a ‘canopy’ of their own threads on which they mate (Locket 1926). In other dictynid spiders in the genera Dictyna Sundevall, 1833, Mallos O. Pickard-Cambridge, 1902, and Mexitlia Lehtinen, 1967, however, there are records of silk addition to the female’s web but no mention of males cutting the female’s silk (Jackson 1979).

Web reduction is a behavior involving extreme web modification with silk addition that has been recorded for sheet weavers (Linyphiidae) and some cobweb weavers (Theridiidae; Table 2). During web reduction, the male moves around the web cutting threads with his chelicerae, then he bundles dismantled sections of the web into thick ropes or balls and, in some cases, wraps them extensively with his own silk (first described by Van Helsdingen 1965 and later studied in detail by Watson 1986; Fig. 2b). The frequency at which this behavior occurs is variable within and among species, as is the extent to which the web is destroyed (Table 2). For instance, web reduction in Lepthyphantes leprosus (Ohlert, 1865) results in a web area decrease of 90% or more, but 55% of males do not engage in web reduction at all (Van Helsdingen 1965). In a field experiment, 69% of Neriene litigiosa (Keyserling, 1886) males reduced a large portion of the female’s web, but web reduction only occurred in 28% of laboratory trials (Watson 1986). In Latrodectus, removal of 50% or less of the female’s web is typical, with ~60–70% of males engaging in web reduction behavior (Anava & Lubin 1993; Scott et al. 2012). Steatoda grossa (C.L. Koch, 1838) males also engage in extensive web reduction, but unlike in Latrodectus, mating tends to take place on a rope or bridge-like section of the web that has been covered with male silk (Scott et al. 2017).

Previous workers have described the construction of a mating web in S. grossa (Gwinner- Hanke 1970; Knoflach 2004), but not the removal of large sections of the female’s capture web. Behavior much like web reduction has also been reported for some araneids that build both typical and irregular orb webs. In the process of constructing their mating threads, males of Micrathena sexspinosa (Hahn, 1822) and Mangora bimaculata (O. Pickard-Cambridge, 1889) engage in extensive web modification (Robinson & Robinson 1980). They break the threads of the viscid spiral on both sides of one radius, such that the web ends up looking like the missing sector webs of Zygiella sp. The resulting analog to the missing sector web’s signal thread is reinforced with up to 30 layers of the male’s silk until it is conspicuously thickened and white in color before being used as a mating thread (Robinson & Robinson 1980). The installation of the

104 mating thread in the irregular orb-webs of Kapogea sexnotata (Simon, 1895) also show parallels to web reduction in the Theridiidae and Linyphiidae. Males first cut away extensive portions of the lower snare of the female’s web and attach silk to the surface of the web periodically during ‘walkabouts.’ The male continues to cut away sections of the female’s web as he installs his mating thread, which he reinforces several times (Robinson & Robinson 1980).

4.3.2 Proposed mechanisms and functions

A number of hypotheses that focus on effect of the male’s silk on female receptivity or aggressive behavior have been proposed, mainly for species that engage in web reduction behavior (Table 2). However, these hypotheses may also apply to mating thread production or any behavior that may allow females to come in contact with male silk. Assuming that there are chemical signals or cues associated with the male silk produced during courtship, silk addition to the female’s web, retreat, or burrow entrance may function in several non-mutually exclusive ways. First, pheromones on male silk might increase or accelerate female receptivity, either by stimulating the female to mate (e.g., initiate receptive postures or behaviors), or by inducing catalepsis (which always precedes successful mating in some species) (Gering 1953; Robinson & Robinson 1973; Ross & Smith 1979, Anava & Lubin 1993). Second, silk addition could also decrease female aggression and the risk of injury to males. For example, in both Lepthyhphantes leprosus and Latrodectus hesperus, web reduction (accompanied by extensive silk deposition) is associated with fewer instances of female aggression (Van Helsdingen 1965; Scott et al. 2012). However, it is not clear whether this is because ‘shy’ or more receptive females tolerate web reduction while aggressive or unreceptive females prevent it, or whether chemical signals associated with male silk decrease female aggression or induce receptivity. In addition, for each of these proposed functions, if females come in contact with the male’s silk, it is also possible that behavioral changes are triggered by tactile (mechanical) cues on the silk rather than by chemical cues. Finally, chemicals (e.g., anti-aphrodisiacs) associated with the male’s silk may deter rival males or render the female’s silk unattractive (Yanez et al. 1999).

Clearly, silk addition to the web leads to structural alterations ranging from the addition of a single line (mating threads) to the major modification of web architecture that results from web reduction behavior (Table 2). Changing web architecture via web reduction and/or silk addition may generally function to improve the transmission of vibratory courtship signals

105

(Robinson & Robinson 1980; Berendonck 2003). By plucking or moving on an isolated mating thread rather than engaging in vibratory courtship on the female’s capture web, signal attenuation and degradation may be reduced. Similarly, constructing a mating web may allow a male to produce a transmission medium with properties that minimize courtship signal attenuation or degradation; these properties may differ from those that maximize capture efficiency for a hunting female. Isolating the female from extraneous vibrations, such as those produced by prey or other males arriving at the web, is a function proposed for web reduction behavior (Rovner 1968; Lubin 1986), but it could also apply to mating threads and webs. For instance, males in several orb-weaver species cut the mating threads of simultaneously courting rivals (Robinson & Robinson 1980).

Rather than improving transmission properties of the web, vibrations associated with cutting silk lines or adding silk could themselves transmit information to the female or attract her attention; that is, silk modification activities may in themselves be courtship signals (Forster 1995; Berendonck 2003). The materials and behaviors involved in silk addition may be energetically costly and provide the female with information about male quality (Anava & Lubin 1993; Harari et al. 2009). Silk is metabolically expensive to produce (Craig 2003), and adult males of web-building spiders apparently reduce or stop foraging after maturity (Foelix 2011), so they have limited energetic resources. When males invest considerable time and large amounts of silk during courtship, this could provide honest information about male nutritional status or vigor.

Fitness benefits of structural changes to the web may be less related to communication, and more related to restricting the mobility of a potentially dangerous female (Van Helsdingen 1965; Ross & Smith 1979; Breene & Sweet 1985). For example, the male may reduce the risk of cannibalism by altering the web in a way that restricts the female’s movements prior to or just after mating. For species that construct a mating thread (Table 2; Fig. 2a), a male may cut the silk line between himself and an aggressive female to remove the immediate risk of attack (Robinson & Lubin 1979; Robinson & Robinson 1980). Alternatively, the ‘treadmill’-type mating threads of some araneids may provide the male with some control over the female’s predatory response and thus decrease the likelihood of cannibalism (Robinson & Robinson 1980).

106

Similarly, structural modifications may reduce the likelihood of females mating with rivals and thus reduce the risk of losing paternity due to polyandry. This effect may also arise through effects on mobility, as altering the web may allow males to control the avenue of approach for a rival male attempting to court the female. If the female is on a mating thread or if web reduction has reduced the web’s surface area, then altering web structure reduces the area that must be defended from competitors (Van Helsdingen 1965; Ross & Smith 1979; Breene & Sweet 1985). Rather than affecting mobility, a similar benefit would arise if web reduction decreases the attraction of rivals in the first place by interfering with the release of the female’s airborne pheromones. Cutting out portions of the female’s web could reduce the surface area of pheromone-laden silk (Watson 1986), and wrapping bundles of the female’s silk with a layer of the male’s own silk may also block the release of pheromones (Scott et al. 2015b).

4.3.3 Current evidence and future directions

Male silk deposition onto webs or other silk structures has been hypothesized to increase female receptivity and the probability of mating, decrease the likelihood of polyandry, and/or decrease the risk of cannibalism by the female. These effects may arise via indirect means (communication, in a number of modalities) or directly through the structural changes to the web, but there is scant experimental evidence supporting these ideas to date (see Table 1).

Of the numerous examples of silk deposition and web modification described above, the only experimental work to determine functions and mechanisms has focused on web reduction in Neriene litigiosa (Watson 1986) and Latrodectus hesperus (Scott et al. 2015b). In both species, reduced webs are less attractive to males than intact webs, indicating that males that engage in web reduction decrease the probability of their long (often several hours) courtship displays being interrupted by rival males. The effect of web reduction is presumably long lasting because mated females rebuild their webs without pheromones, so web-reducing males also benefit by decreasing the probability of sperm competition (Watson 1986). For N. litigiosa, Watson (1986) argued that web reduction limits female silk pheromone emission by decreasing the exposed surface area of the female’s silk. Conversely, the results of a series of field experiments by Scott et al. (2015b) suggest that physical alteration of the web and male silk addition both play roles in the function of web reduction as a mate monopolization tactic. Further work is required to

107 determine the mechanism(s) by which web reduction decreases female attractiveness, and to what extent chemical cues affecting conspecifics (females or males) are involved.

The fitness benefits of web reduction for males are clear; decreasing female attractiveness limits sperm competition in two ways. First, it acts to quickly reduce the arrival rate of rival males, decreasing direct competition for access to females and increasing the likelihood of being first to copulate, which is important in spiders with long precopulatory courtship and first-male sperm precedence like Latrodectus and Neriene litigiosa (Watson 1991; Watson & Lighton 1994; Snow & Andrade 2005; MacLeod 2013). Second, mated females rebuild their webs without attractive pheromones (Watson 1986; MacLeod & Andrade 2014), decreasing the likelihood that males will face sperm competition from subsequently mating males. The fitness consequences for females, however, may be positive or negative. Females may benefit from the ability to quickly become unattractive after mating if they suffer costly harassment from subsequent males arriving at their webs. However, if females benefit from polyandry, web reduction may be costly; it may be a form of manipulation. Experimental studies are needed to explicitly test the fitness consequences to females to determine whether this is an example of cooperation or conflict between the sexes.

Other potential functions of silk addition with and without web modification remain to be experimentally investigated. A number of approaches would be valuable in future studies. First, tests of hypotheses related to vibrational signalling or signal transmission could harness Laser Doppler Vibrometry, which allows precise measurement of silk-borne vibrations with minimal loading of the web, unlike earlier methods based on accelerometers (Masters & Markl 1981; also see alternative methods in Vollrath 1979). Assessment of transmission properties of webs (e.g., Vibert et al. 2016) with and without male silk addition and the attendant web modifications (reduction and mating threads/webs) would be valuable for explicit tests of hypotheses associated with vibration transmission. For example, in recent work, Mortimer et al. (2015) combined laser vibrometry, electron microscopy, tensile testing, and behavioral assays to understand the function and biomechanical properties of the (predation-related) signal thread of Zygiella x-notata (Clerck, 1757). This study could be used as a model for exploring the characteristics of mating threads, and comparisons of signal threads used for predation and those used in mating may produce valuable insights. Second, once vibrations created during silk addition are characterized, synthesized vibrations can then be used in playback experiments to

108 gauge male and female responses (e.g., see Uhl & Elias 2011; Wignall & Herberstein 2013b; Vibert et al. 2014) and how male fitness is affected. Third, behavioral experiments in the laboratory or field that focus on whether the silk itself affects female attractiveness or male mating success could utilize the experimental addition of male silk (e.g., Scott et al. 2015b), or experimental blocking of male silk production by covering the spinnerets of courting males with wax or glue (e.g., Zhang et al. 2011).

4.4 Silk deposition on females

4.4.1 Overview and description of behaviors

Males may deposit silk directly on the female’s body during courtship or copulation—a widespread behavior reported in 16 families, including web-building and cursorial spiders (Figs. 1 & 2; Table 3). The term ‘bridal veil’ was coined by Bristowe (1931) as a descriptor for male silk-laying on females during mating in Xysticus cristatus (Clerck, 1757) and Pycnaxis krakatauensis (Bristowe, 1931) (Thomisidae). Other descriptions for bridal veil spinning behavior include ‘tying’, ‘mate-binding’, ‘silk-binding’, ‘copulatory silk-wrapping’, and ‘trussing’ (Table 3). We use ‘bridal veil’ and ‘veil’ to refer to this behavior because this is the original term and to avoid using terms that imply particular functions. There are several types of bridal veils, and species in which they have been reported vary in the context in which they are used, the part of the body on which the silk is applied, the predictability of silk-laying patterns, and the volume of silk used in the behavior (Table 3).

Extensive silk-wrapping behavior, often focused on the female’s legs, is seen across a number of families of cursorial spiders and some web-builders (Table 3; Fig. 3). Males deposit silk over the female’s first two pairs of legs and anchor the silk to the substrate in several crab spiders, including Xysticus C.L. Koch, 1835 spp., Pycnaxis krakatauensis, and Bassaniana versicolor (Keyserling, 1880) (Thomisidae; Bristowe 1931, 1958; Kaston 1936). Comparable veiling behavior has been described for the wolf spider Schizocosa malitiosa (Tullgren, 1905) (Lycosidae; Aisenberg et al. 2008) and the fishing spider Dolomedes triton (Walckenaer, 1837) (Pisauridae; Carico 1993). Similarly, males of Ctenus longipes Keyserling, 1891 (Ctenidae) concentrate silk deposition on the female’s forelegs and also spin silk over the palps, chelicerae, and eyes (Trillo 2016). Intriguingly, the female apparently eats the veil silk after copulation in this species (discussed below in section 5). In Oxyopes schenkeli Lessert, 1927 (Oxyopidae;

109

Preston-Mafham 1999) and in some cases, Pisaurina mira (Walckenaer, 1837) (Pisauridae; Bruce & Carico 1988; A.G. Anderson pers. obs.), both mates will hang from a dragline below a plant as the male deposits silk on the first two or three pairs of the female’s legs. Males systematically deposit a ring-like veil around the female’s legs as she stands on the ground in Homalonychus selenopoides Marx, 1891 and H. theologus Chamberlin, 1924 (Homalonychidae; Dominguez & Jimenez 2005; Alvarado-Castro & Jimenez 2011), or as the female hangs from her mating web in Nilus curtus (O. Pickard-Cambridge, 1876) (Pisauridae; Sierwald 1988). Males of Ancylometes bogotensis (Keyserling, 1877) (Ctenidae) wrap the distal segments of the female’s legs first with an outer ring of silk, and then add a second inner ring around the patellae (Merrett 1988). Complex, extensive veiling behavior has also been described for several orb-weaver genera (Araneidae; Table 3; Fig. 3a). The diminutive males move around on the dorsum of the female, spinning silk between the bases of her legs, over her cephalothorax, and between her cephalothorax and abdomen (e.g., Robinson & Robinson 1980; Gregoric et al. 2016).

Less extensive silk deposition on females has been described for species in the Agelenidae, Corrinidae, Dictynidae, Philodromidae, Tetragnathidae, Theridiidae and Zoropsidae (see Table 3 for details and references). In these taxa veiling behavior occurs on the female’s web and seems to be less ritualized or more variable than the types described above. In the Theridiidae, for example, there is variation in the occurrence of silk deposition behavior within and across species. Veiling took place in about 33% of courtship observations in Latrodectus revivensis Shulov, 1948 (Anava & Lubin 1993), in 50% of Steatoda bipunctata (Linnaeus, 1758) pairings (Knoflach 2004), and only occasionally in Steatoda grossa (Scott et al. 2017). Intriguingly, application of a bridal veil appears to be an obligate behavior in Nephila pilipes (Fabricius, 1793), as well as in Pisaurina mira (Bruce & Carico 1988; Kuntner et al. 2009; Anderson & Hebets 2016). Conversely, Cupiennius coccineus F.O. Pickard-Cambridge, 1901 (Ctenidae) males normally do not use veils in laboratory trials, but in an inter-species mating experiment, some male C. coccineus deposited silk on C. salei females, which are larger than conspecific females (Schmitt 1992).

4.4.2 Proposed mechanisms and functions

Bridal veils may increase female receptivity and the probability of mating. The veil may also act as a physical restraint that increases mating success or decreases the risk of sexual cannibalism.

110

Finally, it is also possible that application of a veil decreases the likelihood of polyandry. Typically, silk used in veils is not placed within the female’s field of view and is removed shortly after copulation, so visual cues are unlikely to play a functional role, but chemical or tactile information may be important to either the female or rival males. Below, although we focus on hypotheses that suggest mechanisms involving chemical cues, we note that all the proposed effects could also arise from female detection of tactile cues or signals, or perhaps even improved seismic signal transmission via direct contact with male silk.

Females may be more likely to mate with males that produce veils due to information in the silk itself (chemical or tactile modalities), due to mechanical stimulation of structures in the location of the veil (e.g., if silk is laid across particular sensory regions on the female’s body), or due to information in the activity associated with laying silk (tactile cues/signals). Silk-laying may increase female receptivity if the veil allows females to identify males as potential mates (e.g., rather than prey), or if the veil is instrumental in female choice among conspecifics. In terms of mating with rather than attacking the male, it has been proposed that pheromones on the male’s silk may lead to a general reduction in the female’s predatory or aggressive behaviors (e.g., Schmitt 1992; Dominguez & Jiminez 2005). In a more extreme proposal, chemicals in the veil could inhibit movements of the female so that she remains in a cataleptic state during copulation (Ross & Smith 1979; Aisenberg et al. 2008; Preston-Mafham 1999). Silk-borne pheromones could also provide the female with information about the male’s quality (Ross & Smith 1979; Anava & Lubin 1993) and thus increase her receptivity to mating with particular males. Both of these types of functional hypotheses are consistent with previous mechanistic arguments that bridal veils ‘stimulate’ the female or trigger physiological changes that prepare the female for mating (Robinson & Robinson 1973; Preston-Mafham 1999).

As has been proposed for web reduction (Scott et al. 2015b), the male’s silk could also function to deter rival males, possibly via pheromones (Aisenberg et al. 2008) that remain on the female’s body. ‘Antiaphrodisiacs’ (e.g., in butterflies; Estrada et al. 2011) may be particularly effective in species with first-male sperm precedence (such as many spiders), since this predicts the evolution of tactics that allow males to avoid previously-mated females (Parker 1970).

In many species, silk deposition by males seems to target the distal segments of the female’s legs (usually the first two or three pairs) and sometimes the pedipalps (see Table 3)

111

(Aisenberg et al. 2008). Spider chemoreceptors are concentrated on the distal segments of the legs and pedipalps (Trabalon 2013), thus the pattern of silk deposition supports the hypothesis of chemical information delivery to females. In a twist on this idea, Lopez (1987) argued that the female’s sensory hairs might be incapacitated by direct contact with the silk. This argument suggests that reduced predatory responses could be the result of silk-mediated impairment of the female’s sensory system (but see Zhang et al. 2011).

Independent of any signal function of veils, the application of silk to the female’s body could directly affect female positioning or mobility during mating interactions in ways that are beneficial for the male. Silken restraints may facilitate copulation by ensuring the female’s abdomen is supported in a posture that simplifies intromission. However, silk may also reduce female mobility, which could increase copulation duration and thus fertilization success (Anderson & Hebets 2017) or reduce the risk of injury or sexual cannibalism (Anderson & Hebets 2016). There has been some debate as to whether the veil is able to physically restrain the female. Most descriptions indicate that females are quickly and easily able to break free of their silken bonds (e.g., Ross & Smith 1979; Preston-Mafham 1999), making this interpretation seem unlikely for many species, but other authors argue that the brief moments of struggling free from the veil may provide the male with just enough time to escape from a potentially cannibalistic female (Breene & Sweet 1985; Bruce & Carico 1988; Anderson & Hebets 2016; Gregoric et al. 2016). The efficacy of the veil in interfering with female movement may depend on how this tactic is employed and to which body parts the veil is applied. It is worth noting that extensive binding of the female’s legs, common in some species (see Table 3), is also consistent with the idea of an effective restraint.

4.4.3 Current evidence and future directions

The function of the silk bridal veil has been investigated experimentally in only two studies, one with Nephila pilipes (Zhang et al. 2011) and another with Pisaurina mira (Anderson & Hebets 2016). Both studies found that the veil reduced the risk of sexual cannibalism and allowed males to obtain a second sperm transfer opportunity, and in P. mira, this led to higher fertilization success (Anderson & Hebets 2017). Zhang et al. (2011) ablated or occluded the female’s tactile and chemical receptors, revealing that tactile cues associated with tying behavior may be critical for this effect, with chemical cues playing a secondary role. Zhang et al. (2011) conclude that the

112 veil in N. pilipes reduces the risk of sexual cannibalism and allows males to overcome resistance of females to repeated copulations. While Anderson & Hebets (2016) did not directly test for chemical cues, their observations are consistent with the silk wrapping acting as a physical restraint, rather than effects mediated by chemical signals. Female P. mira attempt to free themselves from the silk wrapping (rather than showing reduced activity), and sexual cannibalism attempts occur whether or not the silk wrapping is present (Anderson & Hebets 2016).

For most species in which males apply silk to the female’s body during mating, the fitness consequences are unclear. The varied terms used to describe this behavior in the literature suggests that authors have inferred a range of possible functions from their observations. This is a fascinating phenomenon, and we suggest a number of different approaches could be fruitful for future study.

First, the phylogenetic distribution of the behavior is broad (Fig. 1; Table 3) and may suggest more than one evolutionary origin, so comparative analysis of the behavior and underlying physiology among taxa may be informative. For example, there are many species where extensive leg wrapping is typical, and physical restraint functions should be more likely in these species than in those where wrapping concentrates on the abdomen. We predict that leg- wrapping, but not abdomen-wrapping, will be more likely in taxa with a higher occurrence of sexual cannibalism.

Second, silk wrapping in the context of mating has been proposed to have its evolutionary origin in silk wrapping of prey (Lopez 1987; Schmitt 1992). Prey-wrapping has a similar underlying function, that is, reduced risk of injury from dangerous prey (Foelix 2011). This gives rise to a mechanistic prediction that bridal veils that function as physical restraints should be constructed from aciniform silk (the toughest type of silk, also used in prey capture; Craig 2003). Testing this supporting prediction may involve comparative analysis of silk structure (e.g., Parkhe et al. 1997; Hayashi et al. 2004), or analysis of the glandular origin of bridal veil silks.

Third, careful experimental designs that manipulate the male’s ability to produce the veil, or the female’s ability to detect it (e.g., Zhang et al. 2011; Anderson & Hebets 2016, 2017; and see Aisenberg et al. 2015) can be combined with assessments of female aggression, mating

113 outcomes (e.g., proxies for female choice), or the opportunity for polyandry (e.g., assessments of anti-aphrodisiac effects) to estimate effects on male fitness. Comparative approaches may be valuable here as well. If bridal veils are primarily related to female choice, then they should be more common in taxa with higher levels of inter-male competition over mates, or low overall mating rates.

Fourth, similar types of manipulations can be employed to assess which functions of the bridal veils are related to communication (rather than restraint), and which modalities are most important. Disentangling possible effects of tactile and chemical cues will be particularly interesting. For this work, examination of behavioral effects of extracts of bridal veil silk may also be informative. Moreover, given recent improvement of techniques for nerve recordings from spiders, there is the exciting potential to measure female responses to chemicals vs. tactile cues directly (Menda et al. 2014).

4.5 Silk associated with nuptial gifts

4.5.1 Overview and descriptions of behaviors

Nuptial gifts are material items transferred during mating that function as paternal effort (increasing male offspring number or success) or mating effort (increasing the likelihood of mating; reviewed in Vahed 1998, 2007; Gwynne 2008). Although rare in spiders, the types of gifts reported include the male’s body, glandular secretions from the male’s cephalothorax, and silk-wrapped prey (reviewed in Albo et al. 2013b). Here we will focus on silken nuptial gifts, in particular the wrapped-prey gifts reported in one theridiid, one tetragnathid and in several species in the closely related families Trechaleidae and Pisauridae. We also include silk produced by males and consumed by females (Theridiosomatidae, Lycosidae, and probably Araneidae provide examples of this phenomenon) and silk-lined burrows (provided by males in a sex-role reversed wolf spider) as examples of nuptial gifts. Our focus will be on the function of the silk associated with these nuptial gifts rather than the gifts themselves.

Silk-wrapped nuptial gifts have been well studied in both Pisaura mirabilis (Clerck, 1757) (Pisauridae; Fig. 4a) and Paratrechalea ornata (Mello-Leitao, 1943) (Trechaleidae). Female silk cues (probably sex pheromones) elicit courtship and gift construction in males of both P. ornata (Albo et al. 2009) and P. mirabilis (Albo et al. 2011a). However, female silk is

114 not required to elicit gift-wrapping by P. mirabilis males, who sometimes prepare nuptial gifts before they encounter a female or her draglines (Lang, 1996; Albo et al. 2011a). When a P. mirabilis or P. ornata male finds a female, he presents his gift by holding it in his chelicerae and raising his front legs in a characteristic display. If the female accepts his gift, she grasps it with her chelicerae and copulation ensues while she is feeding on the gift.

Whereas nuptial gifts are the norm for Pisaura mirabilis and Paratrechalea ornata, in Metellina segmentata (Clerck, 1757) (Tetragnathidae) silk-wrapped prey items are used as an alternative mating tactic (Prenter et al. 1994b; reviewed in Neff & Svensson 2013; Fig. 4b). In this species, males guard females and normally wait until she has captured a prey item before initiating courtship. Once the female has captured and wrapped a prey item in silk, the male takes it from her, adds his own silk, and then incorporates the silk-wrapped prey item into his mating thread (Prenter et al. 1994a); he may also wrap the female in a light bridal veil as he does this (Bristowe 1929; Lopez 1987). Clearly, the prey in this situation is not a nuptial gift since the female captures it herself, although once the male steals it from her, he can prevent her from eating it if she does not mate with him (Schneider & Lubin 1998). In rare cases, however, when two males are present on a female’s web (in the field, 7% of females are guarded by two males simultaneously) the male captures the prey item himself and waits for the female to approach it before beginning courtship (Prenter et al. 1994b). In some cases, one male kills and wraps his rival male into a package with another prey item, using this silk-wrapped package to initiate courtship with the female (Prenter et al. 1994b; Fig. 4b).

In the kleptoparasitic and araneophagic spider Argyrodes elevatus (Theridiidae), two anecdotal reports of nuptial gifts are available. One A. elevatus male used a stolen prey item as a gift, and the other used the silk-wrapped carcass of a host spider (Cobbold & Su 2010; Uetz et al. 2010). In the case of the stolen prey item, the male was observed to present the gift to a female, wait until she began feeding on it, and then copulate with her (Uetz et al. 2010). Whether this functions as an alternative mating tactic or simply represents an occasional occurrence in this species remains to be seen.

Silk-wrapped prey gifts have most commonly been reported for pisaurids in the genera Pisaura Simon, 1886 (P. lama in addition to P. mirabilis), Perenethis L. Koch, 1878, Thaumasia Perty, 1833, and Tinus F.O. Pickard-Cambridge, 1901 (Table 4). In addition to Paratrechalea

115 ornata, two congeners and members of the genus Trechalea Thorell, 1869 also use silk-wrapped nuptial gifts (Table 4). The families Pisauridae and Trechaleidae are closely related members of the Lycosoidea (Wheeler et al. 2016; see Fig. 1) hinting at silk-wrapped nuptial gift-giving as a synapomorphy, however, spotty reports of silk-wrapped nuptial gifts in other species suggest that silk-wrapped nuptial gifts may have evolved more than once in spiders.

There a few examples of apparent nuptial gifts in which the male’s silk itself, rather than a prey item, constitutes the gift. In the ray spider Theridiosoma gemmosum (L. Koch, 1877) (Theridiosomatidae), males feed silk directly to the female between repeated copulations (Hajer & Rehakova 2011). This silk is considered a nutrient gift, because araneoids can recycle silk proteins by consuming silk (Craig 2003). Intriguingly, in one ctenid spider species where males deposit a bridal veil, the silk is apparently consumed after copulation. Trillo (2016) describes females of Ctenus longipes grooming the silk veil off of their legs and palps after mating and then bringing the silk to their mouthparts until it disappears. In the araneid spiders Scoloderus cordatus (Stowe 1978) and Isoxya tabulata (Robinson & Robinson 1980) males employ ‘treadmill’-type mating threads that they pay out as the female attempts to walk toward them on the thread. Robinson & Robinson (1980) note that during this process the female accumulates a conspicuous ball of silk under her cephalothorax, and though they do not mention it being consumed, it seems probable that females eat the silk as in the examples above.

Finally, the silk-lined burrows provided by males of the sex role reversed wolf spiders in the genus Allocosa Sundevall, 1833 are nuptial gifts (reviewed in Aisenberg 2014). In Allocosa senex (Mello-Leitao, 1945) and A. alticeps (Mello-Leitao, 1944), males construct silk-lined burrows in which females oviposit and brood their egg sacs, providing both female and offspring with protection from predators (Aisenberg 2014). Females prefer males that provide longer burrows (Aisenberg et al. 2007) and males lengthen their burrows after experiencing rejection by a female (Carballo et al. 2017).

4.5.2 Proposed mechanisms and functions

The most likely function for silk-wrapped nuptial gifts is to increase female receptivity and thus male mating success. Related to this may be a decreased risk of sexual cannibalism, which is a demonstrated function of gifts in Pisaura mirabilis (Toft & Albo 2016). Both of these are forms of mating effort. The silk wrapping of nuptial gifts may in general provide females with

116 information about males via visual, tactile, or chemical cues. Proposed mechanisms for such effects in pisaurids and trechaleids generally fall into two categories. In both cases, silk wrapping may function in several non-mutually exclusive ways that either conflict or align with the female’s interests. First, silk may have direct physical effects if it disguises gift contents and thus increases mating success even if prey items are insufficient (or missing). When silk is wrapped around a non-prey item, it may serve to hide the contents of a ‘worthless package’ (e.g., Ghislandi et al. 2017; Prokov & Semelbauer 2017), through visual obstruction and/or creating a barrier (physical or chemical) between the female and the contents. Males may thus deceive females into mating in the absence of a nutritious gift. However, recent work argues that such ‘worthless’ gifts may be most common in species where the nuptial gift has evolved to serve a signal function rather than a direct benefit (e.g., Albo et al. 2017; Pandulli-Alonso et al. 2017). Another possible physical effect of the silk wrapping is to allow males to maintain a firm grip on the gift to avoid it being stolen by the female (Andersen et al. 2008) or rival males (Nitzsche 2011).

Second, silk may have indirect effects through communication, whereby visual, tactile or chemical cues increase the likelihood of gift acceptance and mating by females. For example, the brightness of the silk wrapping around the gift or its chemo-tactile qualities may provide the female with information about the male’s quality, since silk and/or associated pheromones may provide an honest signal of male body condition (Stalhandske 2002; Albo et al. 2011a; Trillo et al. 2014). Moreover, the amount of silk a male can spin before or during mating may also provide information about body condition (Albo et al. 2011a; Klein et al. 2014). In this context, silk wrapping around prey may also provide a method for delivery of chemicals to female chemosensory organs.

Third, rather than providing information about the male, the silk wrapping may include cues that exploit female sensory biases by mimicking egg sacs (Stalhandske 2002), which females carry in their chelicerae in pisaurids and attached to the spinnerets in trechaleids (Carico 1993).

Fourth, when silk itself is a nuptial gift, it may also function as paternal effort. When the female consumes the silk as in ray spiders (Hajer & Rehakova 2011), at least one ctenid (Trillo 2016) and possibly some araneids that use ‘treadmill’-type mating threads (Stowe 1978;

117

Robinson & Robinson 1980), it may provide additional nutrients to females that are incorporated into the male’s offspring. However, the thick silk wrapping of the nuptial gift apparently does not itself provide a significant source of protein to the female in pisaurids (Nitzsche 1988 as cited by Nitzsche 2011). The silk burrows provided by Allocosa males clearly provide material benefits to females (safe places to oviposit and brood egg sacs) but may also represent mating effort, with males lengthening their burrows (requiring addition of costly silk) in response to rejection by females (Carballo et al. 2017).

In Argyrodes elevatus, silk-wrapped nuptial gifts may be an alternative mating tactic as in Metellina segmentata. Males of some spider species mate opportunistically with females engaged in feeding as a way to avoid sexual cannibalism (e.g., Austin & Anderson 1978; Fromhage & Schneider 2004), and the presentation of nuptial gifts may be a refinement of this mating strategy.

4.5.3 Current evidence and future directions

Experimental studies of the function of nuptial gifts in spiders are restricted to a few species. There is experimental evidence for several functions of nuptial gifts in both Paratrechalea ornata and Pisaura mirabilis. Nuptial gifts in P. mirabilis and P. ornata may have evolved by sexual selection through cryptic female choice for sperm storage (Albo & Costa 2010; Albo et al. 2013a). In both species, males that provide nuptial gifts to mates have longer copulations and transfer more sperm than males who do not provide gifts (Albo & Costa 2010; Albo et al. 2013a), and nuptial gifts are also correlated with accelerated oviposition in P. ornata (Albo & Costa 2010). We note here the interesting functional parallel with the bridal veil in Pisaurina mira (Anderson & Hebets 2016, 2017).

Nuptial gift silk may provide information via visual signals or cues. Paratrechalea ornata females accept smaller, brighter gifts more quickly than larger gifts that are darker in color—but the mechanism is not clear (Klein et al. 2014). Brighter gifts (painted white to match egg sacs) are more quickly accepted than unmanipulated silk-wrapped gifts, which in turn are more readily accepted than gifts painted brown (P. mirabilis, Stalhandske 2002). In some species, visual signals alone may be insufficient to elicit gift acceptance, but ether-extractable chemical compounds specific to nuptial gift silk elicit female acceptance of filter paper ‘gifts’ (Paratrechalea ornata; Brum et al. 2012). Moreover, females more often accepted gifts wrapped

118 by males than gifts wrapped with silk experimentally reeled from males’ spinnerets, suggesting that males control the type of silk they use or the compounds they add to the silk during gift construction (Brum et al. 2012). This suggests that pheromones on male silk stimulate females to accept gifts, thereby increasing mating success of males. In no-choice tests, females responded similarly to silk extracts and prey extracts, implying that the pheromone either has chemical similarities to prey cues and exploits the female’s foraging response, or comprises unrelated compounds that elicit the same response—the acceptance of and feeding on the gift (Brum et al. 2012).

The pheromone on silk-wrapped nuptial gifts may provide information about a male’s quality even if the gift itself does not necessarily honestly indicate prey capture ability (males can steal gifts from rivals or from the female herself; Prenter et al. 1994b; Nitzsche 2011). The extent of silk-wrapping during gift construction depends on male condition in P. mirabilis, with males in better condition adding more silk, and thus this may be an honest signal of male quality (Albo et al. 2011a). As expected if this is the case, P. mirabilis feed longer on gifts wrapped with more silk, and most males already carrying wrapped prey wrap it again after encountering a female (Lang 1996; Albo & Costa 2010), as do males that were previously rejected (Bilde et al. 2007), which appears to increase the attractiveness of the gift (Bilde et al. 2007; Brum et al. 2012). This suggests that visual, tactile and/or chemical cues associated with the male’s silk affect female acceptance of the gift (Bilde et al. 2007).

Similarly, wrapping low-quality gifts in pheromone-laden silk may be a strategy of males that minimizes the costs of providing a gift while maintaining its attractiveness. In some species females will not copulate unless males provide a gift (P. mirabilis, Prokop & Maxwell 2009; Albo et al. 2011b). Such effects may be limited however, as males that present a silk-wrapped gift containing a prey carcass or plant material instead of prey may obtain a short copulation, but, in P. mirabilis, it ends as soon as the female detects that there is no prey inside the silk (Albo et al. 2011b; Brum et al. 2012). Consistent with this, field studies of P. mirabilis found no evidence for ‘sham’ gifts concealed in silk, instead, 40% of males carried gifts, all of these were freshly killed arthropods (Prokop & Maxwell 2009), and gift size was correlated with male body size (Prokop & Semelbauer 2017). The idea that the gift is a sensory trap exploiting female maternal care behavior (Stalhandske 2002) was not supported in this species; experimental evidence

119 suggests that the gift exploits female foraging motivation instead (Bilde et al. 2007; Toft & Albo 2015).

In strong contrast to P. mirabilis, 70% of gifts carried by Paratrechalea ornata males are nutritionally worthless in nature. However, in P. ornata, female receptivity does not depend on hunger, as might be expected if females are permitting copulations with gift-bearing males because they are seeking food (Pandulli-Alonso et al. 2017). Albo et al. (2017) suggest that nuptial gifts may evolve initially due to direct benefits to females, but in some species, gifts may evolve a signal function (Bradbury & Vehrenkamp 2011). Thus P. mirabilis and P. ornata represent different points in the evolutionary ritualization of a direct benefit into a signal (Albo et al. 2017). Alternatively, the provision of worthless gifts may be maintained in a basically honest, direct-benefits system as long as the frequency of these deceptions remains sufficiently low (negative frequency dependence; Dawkins & Guilford 1991; Neff & Svensson 2013). If this is the case, then receiving deceitful gifts will be costly for females, but elevated discrimination would be even more costly than accepting worthless gifts at low frequency. Moreover, at equilibrium, the fitness of males using deceitful or honest tactics should be equal, as part of a mixed ESS (Evolutionarily Stable Strategy; Neff & Svensson 2013). Finally, a high frequency of worthless gifts may occur as a transient outcome of sexually-antagonistic coevolution (Ghislandi et al. 2014). In this case, receiving worthless gifts is costly for females, and the high frequency of male deceit would eventually lead to the evolution of increased female discrimination (Lindstedt & Mokkonen 2014).

In addition to other functions, silk-wrapping apparently affords P. mirabilis males greater control over their gifts by improving their grip, thus decreasing the risk of the female stealing the gift without mating. Moreover, the rounded shape of the wrapped gift facilitates access to the female’s genitalia for copulation, thus increasing their mating success (Andersen et al. 2008). The gift itself may also function as a ‘shield’ preventing cannibalism; cannibalism is six times more likely to occur when males do not provide the female with a gift (Toft & Albo 2016), although it is unclear whether silk wrapping is required for this effect.

Visual cues, chemical cues, and physical properties of the silk have all been implicated in the gift-giving systems of P. ornata and P. mirabilis. Given that the type of silk appears to be important (Brum et al. 2012), future studies could compare the chemical and biophysical

120 properties of silk used to wrap nuptial gifts to those of other silk types. This will facilitate consideration of the origin of gifts, and the identification of putative pheromones on the silk and study of their specific function(s). Whether chemical cues are important in other species that produce silk-wrapped nuptial gifts (or when silk alone acts as a gift) remains to be investigated. In the Allocosa species providing burrows as nuptial gifts, the silk lining alone (i.e., in the absence of the male’s body, which emits a volatile pheromone) is not sufficient to elicit female courtship behavior, but the potential role of the silk in female assessment of males has not been further tested (Aisenberg et al. 2010). Studies of the mechanisms by which nuptial gift silk influences female responses would benefit from experiments that systematically manipulate the possible cues presented in gift silk, and/or ablate the female’s sensory receptors and examine the effect on mating success, cannibalism risk, and female reproductive output (e.g., in the case of species where females consume the male’s silk).

Understanding the evolutionary trajectory of nuptial gift evolution in spiders will require a more explicitly comparative approach, with the addition of studies of more taxa that vary in the type of gift involved in mating. In the broadest sense, this may include males that wait until females are feeding before attempting to mate, those that steal prey from females and then present those same prey at mating, those that wrap nutritive prey to present to females, and those that frequently present non-nutritive, silk-wrapped items to females. A theoretical evolutionary sequence would predict co-occurrence of a number of features in species at different stages (e.g., Albo et al. 2017). In this case, silk wrapping may originally function to subdue prey or for easy manipulation of prey to facilitate gift giving as a direct benefit to females. Hungry females may be more likely to mate and accept these prey items, and most gifts carried by males should be nutritive. In such species, male honesty may be further augmented by the risk of cannibalism from hungry females who do not receive a gift. Later in the evolutionary sequence, silk wrapped packages may provide information to females (via visual, chemical or tactile cues), thus triggering receptive behavior regardless of female’s hunger, and in the absence of a risk of cannibalism. Under this scenario, these species should show a relatively high frequency of ‘worthless’ gifts, but features of the silk package itself would be correlated with male quality (e.g., Pandulli-Alonso et al. 2017). In contrast, species with a relatively high frequency of ‘worthless’ gifts may not be those in which the gift has evolved a signal function; rather this may be an exploitative behavior of males maintained through negative frequency dependence. If silk

121 wrapping serves a deceptive function, it is expected that receptivity of females will be linked to hunger, and gifts should deter sexual cannibalism. This can be tested in experiments that measure the fitness payoff to males bearing worthless gifts as a function of natural or manipulated variation in the relative frequency of the tactic.

4.6 Other examples of male silk use during mating interactions

The three types of silk use discussed above do not include all of the ways that male spiders can use silk during courtship and mating interactions. Below we briefly discuss some other kinds of silk use related to spider mating (see Table 5).

We have not considered sperm webs in this review because their production is rarely observed and described, and, to our knowledge, there have been no suggestions or investigations of functions other than the required one of charging the palps in preparation for sperm transfer (Foelix 2011). Indeed, in many spider taxa males charge their palps with sperm before they set off in search of mates, and thus the silk involved clearly has no effect on females during courtship (Foelix 2011). However, in those species where males build sperm webs on the female’s web and/or charge the palps in between mating bouts (e.g., in the Linyphiidae; Van Helsingden 1965; Watson & Lighton 1994), we cannot exclude the possibility that this silk plays some additional role.

Cursorial spiders trail dragline silk as they move around, periodically anchoring it to the substrate (Richman & Jackson 1992; Foelix 2011), and undoubtedly do so in close proximity to one another during mating interactions. In these taxa, where courtship occurs on substrates other than the female’s silk, male silk function could overlap substantially with those where males deposit silk on the female’s web or body. Explicit studies of male silk use in such contexts are rare, but the following example is illustrative. In the wolf spider Pardosa milvina (Lycosidae), the structure of male silk produced during courtship differs from typical dragline silk (e.g., in the number of attachment disks), and females respond to contact with courtship silk by spinning more of their own pheromone-laden silk (Khan & Persons 2015). Females increase their own silk production in response to males who court less intensively (i.e., males depositing less silk), suggesting that silk-bound pheromones and/or contact cues may mediate a two-way ‘conversation’ between the sexes (Havrilak et al. 2015). Additional studies analyzing the structural and/or chemical differences between silk deposited by males during courtship and

122 other contexts, as well as the behavioral responses of females to these different silk types (as in Khan & Persons 2015) would be very useful. Such studies may reveal that bi-directional communication mediated by silk is common across spider taxa.

The orb-weaver Manogea porracea (Araneidae) provides a unique example of male silk use that facilitates paternal care (Moura et al. 2017). After mating, the male builds his own capture web above that of the female and remains there until the end of the reproductive season. The female then hangs her egg sacs between the two webs and the male provides parental care by protecting the egg sacs from predators. Both parents provide protection, but females frequently die before spiderlings emerge, such that egg sacs attended by males are most common at the end of the season (Moura et al. 2017). This example has clear overlap with nuptial gifts that constitute paternal effort.

4.7 Conclusions and future directions

4.7.1 Summary

Here we reviewed evidence that male silk use during courtship and mating is taxonomically widespread, diverse in possible function and mechanism (Fig. 1; Table 1), and may play an important part in the mating dynamics in many spider species (Tables 2–7). The widespread occurrence of silk deposition by male spiders during courtship and mating (Fig. 1) suggests an important, but often neglected, function of male silk in behavioral interactions between males and females, and among competing males. Moreover, the evidence for ritualized silk use in both the Mygalomorphae and Araneomorphae and its prevalence across the phylogeny (Fig. 1) presents the intriguing possibility that functional roles for male silk use are plesiomorphic among spiders. Systematic use of silk in mating by males includes the addition of silk to females’ webs or other silk structures, silk deposition on females’ bodies, and the use of silk associated with nuptial gifts. In the former two types of silk use, the silk is invariably deposited in close proximity to the female, often in direct contact with her chemoreceptors or proprioreceptors, and in the latter case, female manipulation and/or consumption of gifts places male silk against the sensory receptors on her palps. Thus, simply considering these patterns of silk use suggests hypotheses regarding the role of silk in intersexual communication during mating. Perhaps not surprisingly, thus far, the bulk of experimental studies have focused on the potential importance of male-produced sex pheromones in such communication (Table 1). These studies suggest that

123 silk produced by males can play an important role in inter-sexual communication (see Gaskett 2007). However, chemical communication is just one possible mechanism by which silk use can affect the fitness of mating males (Table 1). Unfortunately, as is common in reviews of spider biology and behavior, our ability to make general inferences is limited because of the relatively narrow taxonomic range of the species that have been well-studied (Huber 2005; Schneider & Andrade 2011). Moreover, while in many spiders the role of the female’s silk is clear and relatively easy to measure, the role of male silk may be more challenging to untangle from correlated activities, even in those species that are relatively well-studied. For example, since silk deposition co-occurs with courtship or nuptial gift presentation, elegant experimentation is required to make clear inferences about the independent effects of the silk itself. Nonetheless, based on the available evidence, we conclude that male silk serves a number of important functions during courtship and mating, and these may be mediated through direct or indirect mechanisms (Table 1).

4.7.2 Functions and mechanisms of effect

Male silk use during mating may evolve or be maintained because it increases male success in the current mating or reduces the risk of losing paternity through polyandry (Table 1). The (scant) current evidence suggests an interesting pattern of segregation of benefits from different types of silk use. The data show that silk addition to the female’s web affects the likelihood of polyandry (and risk of paternity losses to sperm competition) but not the outcome of the mating attempt of the silk-laying male. In contrast, the data suggest that silk addition to the female’s body (bridal veils) and silk associated with nuptial gifts function exclusively to increase the likelihood of favorable outcomes in the current mating, but do not affect polyandry. It seems unlikely that this is a real division, however, given the small number of experimental studies. We consider two examples in which additional work might quickly remove this pattern. First, one of the two mechanisms for which there is currently no experimental support is the hypothesis that silk use could affect female mobility. However, the creation of mating threads in orb-weaving spiders has often been described anecdotally in terms of constraints on the movement of potentially cannibalistic mates, and this makes intuitive sense. Nevertheless, this does not appear in Table 1 because, to our knowledge, there are no experimental tests of this hypothesis, nor of any other way in which male silk use might reduce female mobility (e.g., web modifications that reduce the area of the web). Second, very few studies have examined long-term effects of

124 exposure to male silk on females. So, although in the short term, mate attraction or female receptivity to polyandry may not change, there could be longer-term effects that do confer benefits on silk-spinning males through decreased polyandry. This may be particularly likely if tactile or chemical cues trigger physiological (e.g., hormonal) changes in females that, over time, lead to changes in receptivity (e.g., in Drosophila; Wolfner 2002).

While it is possible that functional effects may arise through indirect effects (communication) or direct effects (physical or structural), the majority of studies to date have focused on indirect effects mediated by chemical communication (Tables 1, 5). There is experimental evidence that chemical communication is involved in all three types of silk use (web modification, veils, gifts). However, in most cases, these studies showed that chemicals are sufficient to elicit an effect but did not exclude other possible mechanisms that might also be operating simultaneously in nature. This is problematic since the hypotheses and mechanisms we suggest for male silk use may overlap, as males may acquire benefits in more than one way, context may determine which function has the strongest effect on male fitness, and more than one mechanism may operate simultaneously. Thus, it is unclear whether these results suggest the critical importance of indirect chemical information relative to other possible mechanisms of effect. Another possibility is that, since male silk is apparently pheromone-laden (Table 6), chemical communication effects may overlay other effects that also affect male fitness.

4.7.3 Improving our understanding of male silk use

To better understand male silk use in courtship and communication, the functional roles of both the silk itself and the behaviors associated with its deposition must be investigated. Preventing males from depositing silk during courtship by occluding their spinnerets with wax or glue is a good technique for investigating the function of male silk (e.g., Anderson & Hebets, 2016). Ablating female chemoreceptors may also be useful in determining the function and importance of chemical signals (e.g., Zhang et al. 2011; Aisenberg et al 2015). Testing the responses of males to the silk of rival males in the context of mate-searching and mate choice (e.g., Schneider et al. 2011) will allow us to determine the function of silk in intra-sexual communication. In species where behavioral evidence indicates the presence of a male silk-borne pheromone, pheromone identification should be pursued. Comparative pheromone analyses of male and female silk may be especially fruitful in those species in which the female pheromone is already

125 known. Recent evidence that silk gene expression and morphology of the spinning apparatus differ between males and females in Steatoda and Latrodectus (Correa-Garwhal et al. 2017) provide the opportunity to link silk structure with function in taxa for which sexual behavior and chemical communication is already well studied. Tichy et al. (2001) have obtained electrophysiological responses to volatile components from tarsal chemoreceptors in Cupiennius salei, and ‘electrolegograms’ have already been developed for whip spiders (Amblypygi; Hebets & Chapman 2000). As our knowledge of spider chemoreception improves, we should strive to develop an analog of the gas chromatographic-electroantennographic detection (GC-EAD) system previously invented for analyses of insect pheromone (Struble & Arn 1984; see Hebets & Chapman 2000). This technique would entail using a spider’s chemoreceptive appendage in place of an insect antenna as a sensor to determine the volatiles that elicit sensory responses. Such a technique would allow rapid screening for potential pheromones in extracts from spider silk or cuticle. Future studies should also attempt to determine the glandular origins of silks and associated pheromones that males produce during courtship and mating behavior. We still do not know where and how spiders synthesize pheromones, but comparative morphology and careful experimentation (e.g., assaying extracts of individual silk glands or body parts) should help us begin to address this major gap in our knowledge.

An intriguing suggestion that appears frequently in the literature is that males may use silk to manipulate females; that is, to partially or completely control female behavior (sensu Dawkins 1978) and thus mating outcomes. Here, ‘manipulation’ is a useful functional concept if the induced female behaviors are beneficial to males but decrease female fitness. In the context of mating, such an outcome may arise through an evolutionary history of sexual conflict (Arnqvist & Rowe 2005). This may be contrasted with communication, which increases the likelihood of a particular female behavior because the behavior is, on average, beneficial for the female as well as the male (Bradbury & Vehrencamp 2011). Manipulation is discussed frequently for the silk-wrapping around nuptial gifts, which can, and often does, conceal ‘worthless’ (nutritionless) items (Ghislandi et al 2017; Pandulli-Alonso et al. 2017). In some species such ‘worthless’ gifts are common and they may nonetheless increase mating success (Albo et al 2017; Pandulli-Alonso et al 2017). However, these deceptive gifts should be considered manipulative only if females mating with males carrying ‘worthless’ gifts have reduced reproductive fitness, and this has not been examined experimentally. Particularly if

126

‘worthless’ gifts are common, it may be that the gift itself is a ritualized representation of male quality (e.g., Albo et al. 2011a; Pandulli-Alonso et al. 2017). Studies of this aspect of male silk use may be particularly valuable, given that ritualization is thought to be the widespread basis of a wide range of signals (Bradbury & Vehrenkamp 2011), but there is very little empirical evidence for this phenomenon (e.g., Scott et al. 2010).

Another common discussion of manipulation arises in the context of silk-borne pheromones that may ‘induce receptivity’ in females, or otherwise change the outcome of the current mating (e.g., Becker et al. 2005). However, when chemical cues induce a behavioral change in females that increases mating success of the silk-laying male, this may also represent a normal, or necessary coordination of male and female behavior that is not maladaptive for females (e.g., Bradbury & Vehrencamp 2011). For example, a small proportion of female agelenids fail to recover from the state of catalepsis following mating (S. Riechert pers. comm.), an observation consistent with manipulation. However, Gehring (1953) suggests the complexity of the agelenid genitalia makes female immobility a necessity for copulation success. Showing a mechanism by which silk leads to negative effects on females does not necessarily demonstrate manipulation. Nevertheless, in general, the phenomena associated with male silk use during mating suggests intriguing questions regarding the role that sexual conflict plays in the evolution of male silk use during mating. Studies of fitness effects on males and females may advance our understanding of this interplay.

Unfortunately, we have insufficient data to analyze comparative patterns regarding male silk use, nor to test hypotheses about evolutionary sequences for current modes of silk use (e.g., nuptial gifts; Albo et al. 2017). We are limited because the bulk of our knowledge of spider mating behavior comes from extensive study of a small number of families including the Araneidae, Ctenidae, Linyphiidae, Lycosidae, Pholcidae, Pisauridae, Salticidae, and Theridiidae (Schneider & Andrade 2011; and see references in this review). Taxa for which we do not report male silk use are as likely to represent the absence of study as the absence of male silk use. Arguably, since males leave behind draglines when they move, and courtship and mate searching often involves extensive movement (Foelix 2011), male silk use during mating may be the rule rather than the exception, despite the limited literature now available. The more ritualized forms of silk use described here (silk deposition, bridal veils, nuptial gifts) may have arisen when more common uses that are found across spiders were coopted for mating. What is clear is that we

127 critically need more phylogenetic coverage in studies of mating, to test these and other hypotheses (Huber 2005; Schneider & Andrade 2011).

Although there are challenges with initiating studies with new species, there may be ways to offset the risk, while maximizing the likely payoff in terms of comparative analyses that increase our understanding. Studies of new species that document the prevalence of male silk use and conduct at least preliminary examinations of the functional importance would be valuable (e.g., by comparing behaviors and mating outcomes for males with and without occluded spinnerets; Zhang et al. 2011; or females with or without ablated sensory structures; Aisenberg et al. 2015). One approach that may be particularly useful would be to focus new efforts on representative species in little-known families within taxa that already have relatively extensive records of a variety of types of male silk use. Two examples are the superfamily Araneoidea, and the Oval Calamistrum clade (Fig. 1), each of which includes records of all three categories of male silk use. Choosing new species to study within these groups would benefit from strategic thinking. Among the web-building Araneoidea, it may be more feasible to stage laboratory matings of spiders that weave irregular webs rather than orb webs since the structural requirements for appropriate web frames may be less stringent (e.g., Nesticidae: scaffold-web weavers; Cyatholipilidae: sheet-web weavers), or, among the less well-known orb-weavers, those that build small webs may be more tractable for laboratory study (e.g., Mysmenidae and Anapidae). Another approach would be to focus on studying multiple species within families in which there are already records of all categories of male silk use (e.g., Theridiidae and Tetragnathidae). Either of these approaches would move us closer to valid comparative tests for understanding the evolution of silk use.

4.7.4 Concluding remarks

Overall, this review provides a functional and mechanistic framework for understanding the diversity of male silk use behaviors, and suggests fruitful approaches and taxa for study. Spiders are models for studies of sexual selection, and how choice, competition, and communication are affected by ecology, cannibalism, and sexual conflict more broadly (Herberstein et al. 2002; Schneider & Andrade 2011; Uhl & Elias, 2011; Kralj-Fiser et al. 2016). As in other fields, insight is limited by what we choose to study (Huber 2005). While technological limitations created challenges in the past, particularly to the study of silk, vibrations, or pheromones, a

128 number of novel approaches now make these studies more feasible (e.g., Hebets & Chapman 2000; Menda et al. 2014; Mortimer et al. 2015). Harnessing these techniques and expanding the range of taxa studied may lead to big advances in understanding. The strong evidence presented here for various effects of male silk in mating suggests that we currently have only part of the picture with respect to spider mating behavior in most taxa. Understanding how male-produced silk may influence, constrain, or manipulate interactions with females and with rival males could provide significant new insights into mating behavior, the evolution of traits related to mating, and fuel new tests of a wide range of theory in sexual selection and sexual conflict.

Acknowledgements

We thank Samantha Vibert for conversations inspiring the topic of the review and Rick Vetter for inviting us to write it. Anita Aisenberg, Gerhard Gries, and two anonymous reviewers provided helpful comments on earlier versions of the manuscript. We also thank Maria Hiles, Alan Lau, Sean McCann, Conall McCaughey, Ed Niewenhuys, and Shichang Zhang for the photographs. Funding was provided by a Natural Sciences and Engineering Research Council of Canada (NSERC) Alexander Graham Bell Canada Graduate Scholarship (to CES), and the Canada Research Chairs Program (to MCBA).

4.8 References Aisenberg, A. 2014. Adventurous females and demanding males: sex role reversal in a Neotropical spider. Pp. 163–182. In Sexual Selection: Perspectives and Models from the Neotropics. (R. Macedo & G. Machado, eds.). Academic Press, London, UK. Aisenberg, A., C. Viera & F.G. Costa. 2007. Daring females, devoted males, and reversed sexual size dimorphism in the sand-dwelling spider Allocosa brasiliensis (Araneae, Lycosidae). Behavioral Ecology and Sociobiology 62:29–35. Aisenberg, A., L. Baruffaldi & M. González. 2010. Behavioural evidence of male volatile pheromones in the sex-role reversed wolf spiders Allocosa brasiliensis and Allocosa alticeps. Naturwissenschaften 97:63–70. Aisenberg, A., N. Estramil, M. González, C.A. Toscano-Gadea & F.G. Costa. 2008. Silk release by copulating Schizocosa malitiosa males (Araneae, Lycosidae): a bridal veil. Journal of Arachnology 36:204–206. Aisenberg, A., G. Barrantes & W.G. Eberhard. 2015. Hairy kisses: tactile cheliceral courtship affects female mating decisions in Leucauge mariana (Araneae, Tetragnathidae). Behavioral Ecology and Sociobiology 69:313–323.

129

Alam, M.S., M.A. Wahab, & C.H. Jenkins. 2007. Mechanics in naturally compliant structures. Mechanics of Materials 39:145–160. Albo, M.J., S. Toft & T. Bilde. 2011. Condition dependence of male nuptial gift construction in the spider Pisaura mirabilis (Pisauridae). Journal of Ethology 29:473–479. Albo, M.J & F.G. Costa. 2010. Nuptial gift-giving behaviour and male mating effort in the Neotropical spider Paratrechalea ornata (Trechaleidae). Animal Behaviour 79:1031– 1036. Albo, M.J., S. Toft & T. Bilde. 2011a. Condition dependence of male nuptial gift construction in the spider Pisaura mirabilis (Pisauridae). Journal of Ethology 29:473–479. Albo, M.J., G. Winther, C. Tuni, S. Toft & T. Bilde. 2011b. Worthless donations: male deception and female counter play in a nuptial gift-giving spider. BMC Evolutionary Biology 11:329. Albo, M.J., L.E. Costa-Schmidt & F.G. Costa. 2009. To feed or to wrap? Female silk cues elicit male nuptial gift construction in a semiaquatic trechaleid spider. Journal of Zoolology 277:284–290. Albo, M.J., S. Toft & T. Bilde. 2013a. Sexual selection, ecology, and evolution of nuptial gifts in spiders, Pp. 183–200. In Sexual Selection: Perspectives and Models from the Neotropics. (R. Macedo & G. Machado, eds). Academic Press, London, UK. Albo, M.J., T. Bilde & G. Uhl. 2013b. Sperm storage mediated by cryptic female choice for nuptial gifts. Proceedings of the Royal Society B 280:20131735. Albo, M.J., N. Macías-Hernández, T. Bilde & S. Toft. 2017. Mutual benefit from exploitation of female foraging motivation may account for the early evolution of gifts in spiders. Animal Behaviour 129:9–14. Alvarado-Castro, J.A. & M.L. Jiménez. 2011. Reproductive behavior of Homalonychus selenopoides (Araneae: Homalonychidae). Journal of Arachnology 39:118–127. Anava, A. & Y. Lubin. 1993. Presence of gender cues in the web of a widow spider Latrodectus revivensis, and a description of courtship behaviour. Bulletin of the British Arachnological Society 9:119–122. Andersen, T., K. Bollerup, S. Toft & T. Bilde. 2008. Why do males of the spider Pisaura mirabilis wrap their nuptial gifts in silk: Female preference or male control? Ethology 114:775–781. Anderson, A.G. & E.A. Hebets. 2016. Benefits of size dimorphism and copulatory silk wrapping in the sexually cannibalistic nursery web spider, Pisaurina mira. Biology Letters 12:20150957. Anderson, A.G. & E.A. Hebets. 2017. Increased insertion number leads to increased sperm transfer and fertilization success in a nursery web spider. Animal Behaviour 132:121– 127. Anderson, J.T. & D.H. Morse. 2001. Pick-up lines: cues used by male crab spiders to find reproductive females. Behavioral Ecology 12:360–366. Andersson, M.B. 1994. Sexual Selection. Princeton University Press, Princeton, NJ.

130

Andrade, M.C.B. 1996. Sexual selection for male sacrifice in the Australian redback spider. Science 271:70–71. Andrade, M.C.B. & M.M. Kasumovic. 2005. Terminal investment strategies and male mate choice: extreme tests of Bateman. Integrative and Comparative Biology 45:838–847. Arnqvist, G. & L. Rowe. 2013. Sexual Conflict. Princeton University Press, Princeton, NJ. Austin, A.D., & D.T. Anderson. 1978. Reproduction and development of the spider Nephila edulis (Koch)(Araneidae: Araneae). Australian Journal of Zoology 26:501–518. Ayyagari, L.R. & W.J. Tietjen. 1987. Preliminary isolation of male-inhibitory pheromone of the spider Schizocosa ocreata (Araneae, Lycosidae). Journal of Chemical Ecology 13:237– 244. Barrantes, G. & M.J. Ramírez. 2013. Courtship, egg sac construction, and maternal care in Kukulcania hibernalis, with information on the courtship of Misionella mendensis (Araneae, Filistatidae). Arachnology 16:72–80. Barrantes, G., L. Sandoval, C. Sánchez-Quirós, P.P. Bitton & S.M. Doucet. 2013. Variation and possible function of egg sac coloration in spiders. Journal of Arachnology 41:342–348. Barth, F.G. 1982. Spiders and vibratory signals: sensory reception and behavioral significance, Pp. 67–120. In Spider Communication: Mechanisms and Ecological Significance. (P. Witt & J. Rovner, eds.). Princeton University Press, Princeton, NJ. Barth, F.G. 2002. A Spider’s World: Senses and Behavior. Springer, Heidelberg, Germany. Baruffaldi, L., F.G. Costa, A. Rodríguez & A. González. 2010. Chemical communication in Schizocosa malitiosa: evidence of a female contact sex pheromone and persistence in the field. Journal of Chemical Ecology 36:759–767. Bateson, P.P.G. 1983. Mate Choice. Cambridge University Press, Cambridge, UK. Becker, E., S. Riechert & F. Singer. 2005. Male induction of female quiescence/catalepsis during courtship in the spider, Agelenopsis aperta. Behaviour 142:57–70. Bell, R.D. & J.A. Roberts. 2017. Trail-following behavior by males of the wolf spider, Schizocosa ocreata (Hentz). Journal of Ethology 35:29–36. Benamú, M.A., N.E. Sánchez, C. Viera & A. González. 2012. Sexual behavior of Alpaida veniliae (Araneae: Araneidae). Revista de Biología Tropical 60:1259–1270. Benamú, M.A., N.E. Sánchez, C. Viera & A. González. 2015. Sexual cannibalism in the spider Alpaida veniliae (Keyserling 1865)(Araneae: Araneidae). Journal of Arachnology 43:72– 76. Berendonck, B. 2003. Reproductive strategies in Latrodectus revivensis (Araneae; Theridiidae): functional morphology and sexual cannibalism. PhD Thesis. Heinrich-Heine-Universität Düsseldorf. Berry, J.W. 1987. Notes on the life history and behavior of the communal spider Cyrtophora moluccensis (Doleschall)(Araneae, Araneidae) in Yap, Caroline Islands. Journal of Arachnology, 15:309–319.

131

Bilde, T., C. Tuni, R. Elsayed, S. Pekar & S. Toft. 2007. Nuptial gifts of male spiders: sensory exploitation of the female’s maternal care instinct or foraging motivation? Animal Behaviour 73:267–273. Blackledge, T.A. 1998. Signal conflict in spider webs driven by predators and prey. Proceedings of the Royal Society B 265:1991–1996. Blackledge, T.A. & J.W. Wenzel. 2000. The evolution of cryptic spider silk: a behavioral test. Behavioral Ecology 11:142–145. Blamires, S.J., T.A. Blackledge & I.M. Tso. 2017. Physicochemical property variation in spider silk: Ecology, evolution, and synthetic production. Annual Review of Entomology 62:443–460. Bourne, J.D. 1978. Observations on the sexual behaviour of Porrhomma egeria Simon (Araneae: Linyphiidae). Bulletin of the British Arachnological Society 4:221–225. Bradbury, J.W. & S.L. Vehrencamp. 2011. Principles of Animal Communication. Sinauer Associates, Sunderland, MA. Breene, R. & M. Sweet. 1985. Evidence of insemination of multiple females by the male black widow spider, Latrodectus mactans (Araneae, Theridiidae). Journal of Arachnology 13:331–335. Bristowe, W.S. 1929. The mating habits of spiders, with special reference to the problems surrounding sex dimorphism. Proceedings of the Zoological Society of London 1929:309–358. Bristowe, W.S. 1931. The mating habits of spiders: a second supplement, with the description of a new thomisid from Krakatau. Proceedings of the Zoological Society of London 1931:1401–1412. Bristowe, W.S. 1958. The World of Spiders. Collins, London, UK. Bristowe, W.S. & G.H. Lockett. 1926. The courtship of British lycosid spiders, and its probable significance. Proceedings of the Zoological Society of London 22:317–347. Bruce, J. & J. Carico. 1988. Silk use during mating in Pisaurina mira (Walckenaer) (Areneae, Pisauridae). Journal of Arachnology 16:1–4. Brum, P.E.D., L.E. Costa-Schmidt & A.M. de Araújo. 2011. It is a matter of taste: chemical signals mediate nuptial gift acceptance in a neotropical spider. Behavioral Ecology 23:442–447. Bucher, R., H. Binz, F. Menzel, & M.H. Entling. 2014. Effects of spider chemotactile cues on arthropod behavior. Journal of Insect Behavior 27:567–580. Bukowski, T.C., & T.E. Christenson. 2000. Determinants of mating frequency in the spiny orbweaving spider, Micrathena gracilis (Araneae: Araneidae). Journal of Insect Behavior 13:331–352. Bukowski, T.C., C.D. Linn, & T.E. Christenson. 2001. Copulation and sperm release in Gasteracantha cancriformis (Araneae: Araneidae): differential male behaviour based on female mating history. Animal Behaviour 62:887–895.

132

Carballo, M., F. Baldenegro, F. Bollatti, A.V. Peretti & A. Aisenberg. 2017. No pain, no gain: male plasticity in burrow digging according to female rejection in a sand-dwelling wolf spider. Behavioural Processes 140:174–180. Carico, J.E. 1993. Revision of the genus Trechalea Thorell (Araneae, Trechaleidae) with a review of the taxonomy of the Trechaleidae and Pisauridae of the western hemisphere. Journal of Arachnology 21:226–257. Chinta, S.P., S. Goller, J. Lux, S. Funke, G. Uhl & S. Schulz. 2010. The sex pheromone of the wasp spider Argiope bruennichi. Angewandte Chemie International Edition 49:2033– 2036. Cobbold, S.M. & Y.C. Su. 2010. The host becomes dinner: possible use of Cyclosa as a nuptial gift by Argyrodes in a colonial web. Journal of Arachnology 38:132–134. Costa-Schmidt, L.E., J.E. Carico & A.M de Araújo. 2008. Nuptial gifts and sexual behavior in two species of spider (Araneae, Trechaleidae, Paratrechalea). Naturwissenschaften 95:731–9. Coyle, F.A. 1985. Observations on the mating behaviour of the tiny mygalomorph spider, Microhexura montivaga Crosby & Bishop (Araneae, Dipluridae). Bulletin of the British Arachnological Society 6:328–330. Coyle, F.A. & T.C. O’Shields. 1990. Courtship and mating behavior of Thelechoris karschi (Araneae, Dipluridae), an African funnelweb spider. Journal of Arachnology 18:281–296. Craig, C.L. & G.D. Bernard. 1990. Insect attraction to ultraviolet‐reflecting spider webs and web decorations. Ecology 71:616–623. Craig, C.L., R.S. Weber & G.D. Bernard. 1996. Evolution of predator-prey systems: spider foraging plasticity in response to the visual ecology of prey. The American Naturalist 147:205–229. Craig, C.L. 2003. Spiderwebs and Silk: Tracing Evolution from Molecules to Genes to Phenotypes. Oxford University Press, Oxford, UK. Cross, F. R. 2016. Discrimination of draglines from potential mates by Evarcha culicivora, an East African jumping spider. New Zealand Journal of Zoology 43:84–95. Cross, F.R. & R.R. Jackson. 2009. Mate-odour identification by both sexes of Evarcha culicivora, an East African jumping spider. Behavioural Processes 81:74–79. Cross, F.R. & R.R. Jackson. 2013. The functioning of species-specific olfactory pheromones in the biology of a mosquito-eating jumping spider from East Africa. Journal of Insect Behavior. 26:131–148. Dawkins, M.S. & T. Guilford. 1991. The corruption of honest signalling. Animal Behaviour 41:865–873. Dawkins, R. 1978. Replicator selection and the extended phenotype. Zeitschrift für Tierpsychologie 47:61–76. Domínguez, K. & M.L. Jiménez. 2005. Mating and self-burying behavior of Homalonychus theologus Chamberlin (Araneae, Homalonychidae) in Baja California sur. Journal of Arachnology 33:167–174.

133

Dondale, C.D. & B.M. Hegdekar. 1973. The contact sex pheromone of Pardosa lapidicina Emerton (Araneida: Lycosidae). Canadian Journal of Zoology 51:400–401. Dondale, C.D., J.H. Redner, P. Paquin & H.W. Levi. 2003. Orb-Weaving Spiders of Canada and Alaska The Insects and Arachnids of Canada Series, Part 23: Araneae: Uloboridae, Tetragnathidae, Araneidae, Theridiosomatidae. NRC Research Press, Ottawa, ON, Canada. Elgar, M.A. & D.R. Nash. 1988. Sexual cannibalism in the garden spider Araneus diadematus. Animal Behaviour 36:1511–1517. Elgar, M.A. 1991. Sexual cannibalism, size dimorphism, and courtship behavior in orb‐weaving spiders (Araneidae). Evolution 45:444–448. Elgar, M.A., J.M. Schneider & M.E. Herberstein. 2000. Female control of paternity in the sexually cannibalistic spider Argiope keyserlingi. Proceedings of the Royal Society B 267:2439–2443. Elias, D.O. & A.C. Mason. 2011. Signaling in variable environments: substrate-borne signaling mechanisms and communication behavior in spiders. In The Use of Vibrations in Communication: Properties, Mechanisms and Function Across Taxa. (C.E. O'Connell- Rodwell, ed.). Research Signpost, Kerala, India. Elias D.O., M.C.B. Andrade & M.M. Kasumovic. 2011. Dynamic population structure and the evolution of spider mating systems. Advances in Insect Physiology. 41:65–114. Engelhardt, W. 1964. Die mitteleuropäischen arten der gattung Trochosa CL Koch, 1848 (Araneae, Lycosidae). Morphologie, chemotaxonomie, biologie, autökologie. Zoomorphology 54:219–392. Estrada, C., S. Schulz, S. Yildizhan & L.E. Gilbert. 2011. Sexual selection drives the evolution of antiaphrodisiac pheromones in butterflies. Evolution 65:2843–2854. Fischer, M.L., A. Čokl, E.N. Ramires, E. Marques-da-Silva, C. Delay, J.D. Fontana et al. 2009. Sound is involved in multimodal communication of Loxosceles intermedia Mello-Leitão, 1934 (Araneae; Sicariidae). Behavioural Processes 82:236–243. Foelix, R.F. 2011. Biology of Spiders. Oxford University Press, Oxford, UK. Forster, L. 1992. The stereotyped behavior of sexual cannibalism in Latrodectus hasselti Thorell (Araneae, Theridiidae), the Australian redback spider. Australian Journal of Zoololgy 40:1–11. Forster, L. 1995. The behavioral ecology of Latrodectus hasselti (Thorell), the Australian redback spider (Araneae: Theridiidae): A review. Records of the Western Australia Museum Supplement 52:13–24. Fromhage, L. & J.M. Schneider. 2004. Safer sex with feeding females: sexual conflict in a cannibalistic spider. Behavioral Ecology 16:377–382. Galasso, A.B. 2012. Comparative analysis of courtship in Agelenopsis funnel-web spiders (Araneae, Agelenidae) with an emphasis on potential isolating mechanisms. PhD thesis. University of Tennessee, Knoxville. Gaskett, A.C. 2007. Spider sex pheromones: emission, reception, structures, and functions. Biological Reviews 82:27–48.

134

Gerhardt, U. 1924. Weitere Studien über die Biologie der Spinnen. Archiv für Naturgeschichte 90:85–192. Gerhardt, U. 1933. Neue untersuchungen zur sexualbiologie der spinnen, insbesondere an arten der mittelmeerländer und der tropen. Zeitschrift für Morphologie und Ökologie der Tiere 27:1–75. Gering, R. 1953. Structure and function of the genitalia in some American agelenid spiders. Smithsonian Miscellaneous Collections 121:1–84. Ghislandi, P.G., M.J. Albo, C. Tuni & T. Bilde. 2014. Evolution of deceit by worthless donations in a nuptial gift-giving spider. Current Zoology 60:43–51. Ghislandi, P.G., M. Beyer, P. Velado & C. Tuni. 2017. Silk wrapping of nuptial gifts aids cheating behaviour in male spiders. Behavioral Ecology 28:744–749. Gosline, J.M., P.A. Guerette, C.S. Ortlepp & K.N. Savage. 1999. The mechanical design of spider silks: from fibroin sequence to mechanical function. Journal of Experimental Biology 202:3295–3303. Gregorič, M., K. Šuen, R.C. Cheng, S. Kralj-Fišer & M. Kuntner. 2016. Spider behaviors include oral sexual encounters. Scientific Reports 6:25128. Gunnarsson, B., G. Uhl & K. Wallin. 2004. Variable female mating positions and offspring sex ratio in the spider Pityohyphantes phrygianus (Araneae: Linyphiidae). Journal of Insect Behavior 17:129–144. Gwinner‐Hanke, H. 1970. Zum Verhalten zweier stridulierender Spinnen Steatoda bipunctata Linné und Teutana grossa Koch (Theridiidae, Araneae), unter besonderer Berücksichtigung des Fortpflanzungsverhaltens. Ethology 27:649–678. Gwynne, D.T. 2008. Sexual conflict over nuptial gifts in insects. Annual Review of Entomology 53:83–101. Hajer, J. & D. Řeháková. 2011. Mating behavior of Theridiosoma gemmosum (Araneae: Theridiosomatidae): The unusual role of the male dragline silk. Archives of Biological Sciences 63:199–208. Harari, A.R., M. Ziv & Y. Lubin. 2009. Conflict or cooperation in the courtship display of the white widow spider, Latrodectus pallidus. Journal of Arachnology 37:254–260. Hauser, M.D. 1996. The Evolution of Communication. MIT Press, Cambridge, MA. Hayashi, C.Y., T.A. Blackledge & R.V. Lewis. 2004. Molecular and mechanical characterization of aciniform silk: uniformity of iterated sequence modules in a novel member of the spider silk fibroin gene family. Molecular Biology and Evolution 21:1950–1959. Hebets, E.A. & R.F. Chapman. 2000. Electrophysiological studies of olfaction in the whip spider Phrynus parvulus (Arachnida, Amblypygi). Journal of Insect Physiology 46:1441–1448. Henneken, J., T.M. Jones, J.Q. Goodger, D.A. Dias, A. Walter & M.A. Elgar. 2015. Diet influences female signal reliability for male mate choice. Animal Behaviour 108:215– 221. Henneken, J., J.Q. Goodger, T.M. Jones & M.A. Elgar. 2017a. The potential role of web-based putrescine as a prey-attracting allomone. Animal Behaviour 129:205–210.

135

Henneken, J., Goodger, J.Q., Jones, T.M., & M.A. Elgar. 2017b. Variation in the web-based chemical cues of Argiope keyserlingi. Journal of Insect Physiology. 101:15–21. Herberstein, M.E., J.M. Schneider & M.A. Elgar. 2002. Costs of courtship and mating in a sexually cannibalistic orb-web spider: female mating strategies and their consequences for males. Behavioral Ecology and Sociobiology 51:440–446. Herberstein, M.E., A.E. Wignall, E.A. Hebets & J.M. Schneider. 2014. Dangerous mating systems: signal complexity, signal content and neural capacity in spiders. Neuroscience & Biobehavioral Reviews 46:509–518. Herms, W., S. Bailey & B. McIvor. 1935. The black widow spider. University of California Agricultural Experiment Station Bulletin 591:1:30. Hibler, T.L. & A.E. Houde. 2006. The effect of visual obstructions on the sexual behaviour of guppies: the importance of privacy. Animal Behaviour 72:959–964. Hsia, Y., E. Gnesa, R. Pacheco, K. Kohler, F. Jeffery & C. Vierra. 2012. Synthetic spider silk production on a laboratory scale. Journal of Visualized Experiments: JoVE 65:4191. Hoefler, C.D. 2007. Male mate choice and size-assortative pairing in a jumping spider, Phidippus clarus. Animal Behaviour 73:943–954. Huber, B.A. 2005. Sexual selection research on spiders: progress and biases. Biological Reviews 80:363–385. Itakura, Y. 1993. The life history and nuptial feeding of a nursery web spider, Pisaura lama. Insectarium 30:88–93. Itakura, Y. 1998. Discovery of nuptial feeding in the spider, Perenethis fascigera (Araneae: Pisauridae ). Acta Arachnologica 47:173–175. Jackson, R.R. 1979. Comparative studies of Dictyna and Mallos (Araneae, Dictynidae) II. The relationship between courtship, mating, aggression and cannibalism in species with differing types of social organization. Revue Arachnologique 2:103–132. Jackson, R.R. & A.M. Macnab. 1989. Display, mating, and predatory behaviour of the jumping spider Plexippus paykulli (Araneae: Salticidae). New Zealand Journal of Zoology 16:151–168. Jackson, R.R. & S.D. Pollard. 1990. Intraspecific interactions and the function of courtship in mygalomorph spiders: a study of Porrhothele antipodiana (Araneae: Hexathelidae) and a literature review. New Zealand Journal of Zoology 17:499–526. Jackson, R.R. & B.A. Poulsen. 1990. Predatory versatility and intraspecific interactions of Supunna picta (Araneae: Clubionidae). New Zealand Journal of Zoology 17:169–184. Jennions, M.D. & M. Petrie. 1997. Variation in mate choice and mating preferences: a review of causes and consequences. Biological Reviews 72:283–327. Jerhot, E., J.A. Stoltz, M.C.B. Andrade & S. Schulz. 2010. Acylated serine derivatives: a unique class of arthropod pheromones of the Australian redback spider, Latrodectus hasselti. Angewandte Chemie International Edition 49:2037–2040. Johnson, J.C., P. Trubl, V. Blackmore & L. Miles. 2011. Male black widows court well-fed females more than starved females: silken cues indicate sexual cannibalism risk. Animal Behaviour 82:383–390.

136

Kaston, B.J. 1936. The senses involved in the courtship of some vagabond spiders. Entomologica Americana 16:97–159. Kaston, B. 1970. Comparative biology of American black widow spiders. Transactions of the San Diego Society of Natural History 16:33–82. Kasumovic, M.M. & M.C.B. Andrade. 2004. Discrimination of airborne pheromones by mate- searching male western black widow spiders (Latrodectus hesperus): species-and population-specific responses. Canadian Journal of Zoology 82: 1027–1034. Kasumovic, M.M. & M.C.B. Andrade. 2006. Male development tracks rapidly shifting sexual versus natural selection pressures. Current Biology 16:R242–R243 Khan, R. & M.H. Persons. 2015. Female Pardosa milvina wolf spiders increase silk advertisements when in the presence of silk from courting males. Journal of Arachnology 43:168–173. Klein, A.L., M.C. Trillo, F.G. Costa & M.J. Albo. 2014. Nuptial gift size, mating duration and remating success in a Neotropical spider. Ethology Ecology & Evolution 26:29–39. Kluge, J.A., O. Rabotyagova, G.G. Leisk & D.L. Kaplan. 2008. Spider silks and their applications. Trends in Biotechnology 26:244–251. Knoflach, B. 2004. Diversity in the copulatory behaviour of combfooted spiders (Araneae, Theridiidae). Denisia 12:161–256. Knoflach, B., & A. van Harten. 2000. Palpal loss, single palp copulation and obligatory mate consumption in Tidarren cuneolatum (Tullgren, 1910)(Araneae, Theridiidae). Journal of Natural History 34:1639–1659. Knoflach, B. & A. van Harten. 2002. The genus Latrodectus (Araneae: Theridiidae) from mainland Yemen, the Socotra Archipelago and adjacent countries. Fauna of Arabia 19:321–362. Koeppel, A. & C. Holland. 2017. Progress and trends in artificial silk spinning: a systematic review. ACS Biomaterials Science and Engineering 3:226–237. Koh, T.H., W.K. Seah, L.M.Y. Yap & D. Li. 2009. Pheromone-based female mate choice and its effect on reproductive investment in a spitting spider. Behavioral Ecology and Sociobiology 63:923–930. Krafft, B. & L.J. Cookson. 2012. The role of silk in the behaviour and sociality of spiders. Psyche 2012:529564. Kralj-Fišer, S., M. Gregorič, T. Lokovšek, T. Čelik & M. Kuntner. 2013. A glimpse into the sexual biology of the “zygiellid” spider genus Leviellus. Journal of Arachnology 41:387– 391. Kralj-Fišer, S., K. Čandek, T. Lokovšek, T. Čelik, R.C. Cheng, M.A. Elgar & M. Kuntner. 2016. Mate choice and sexual size dimorphism, not personality, explain female aggression and sexual cannibalism in raft spiders. Animal Behaviour 111: 49–55. Kuntner, M., S. Kralj-Fišer, J.M. Schneider & D. Li. 2009. Mate plugging via genital mutilation in nephilid spiders: an evolutionary hypothesis. Journal of Zoology 277:257–266.

137

Laborda, A. & M. Simo. 2015. Description of the female of Eutichurus ibiuna Bonaldo, 1994 (Araneae: Eutichuridae) with notes on natural history and sexual behavior. Zootaxa 4021:591–596. Lai, C.W., S. Zhang, D. Piorkowski, C.P. Liao & I.M. Tso. 2017. A trap and a lure: dual function of a nocturnal animal construction. Animal Behaviour 130:159–164. Landolfa, M.A. & F.G. Barth. 1996. Vibrations in the orb web of the spider Nephila clavipes: cues for discrimination and orientation. Journal of Comparative Physiology A 179:493– 508. Lane, S.M., J.H. Solino, C. Mitchell, J.D. Blount, K. Okada, J. Hunt & C.M. House. 2015. Rival male chemical cues evoke changes in male pre-and post-copulatory investment in a flour beetle. Behavioral Ecology 26:1021–1029. Lang, A. 1996. Silk investment in gifts by males of the nuptial feeding spider Pisaura mirabilis (Araneae: Pisauridae). Behaviour 133:697–716. Lapinski, W. & M. Tschapka. 2009: Erstnachweis von Brautgeschenken bei Trechalea sp. (Trechaleidae, Araneae) in Costa Rica. Arachne 14:4–13. Leonard, A.S. & D.H. Morse. 2006. Line-following preferences of male crab spiders, Misumena vatia. Animal Behaviour 71:717–724. Lindstedt, C. & M. Mokkonen. 2014. The evolutionary strategy of deception. Current Zoology 60:1–5. Lizotte, R. & J.S. Rovner. 1989. Water-resistant sex pheromones in lycosid spiders from a tropical wet forest. Journal of Arachnology 17:122–125. Locket, G.H. 1926. Observations on the mating habits of some web-spinning spiders, with some corroborative notes by W.S. Bristowe. Proceedings of the Zoological Society of London 1926:1125–1146. Locket, G.H. 1927. XII.—On the mating habits of some spiders in the family Theridiidae. Annals and Magazine of Natural History 20:91–99. Lopez, A. 1987. Glandular aspects of sexual biology, Pp. 121–132. In Ecophysiology of Spiders. (W. Nentwig, ed.). Springer, Heidelberg, Germany. Lubin, Y. D. 1986. Courtship and alternative mating tactics in a social spider. Journal of Arachnology 14:239–257. MacLeod, E.C. & M.C.B. Andrade. 2014. Strong, convergent male mate choice along two preference axes in field populations of black widow spiders. Animal Behaviour 89:163– 169. Mallis, R.E. & K.B. Miller. 2017. Natural history and courtship behavior in Tengella perfuga Dahl, 1901 (Araneae: Zoropsidae). Journal of Arachnology 45:166–176. Masters, W.M. 1984. Vibrations in the orbwebs of Nuctenea sclopetaria (Araneidae). Behavioral Ecology and Sociobiology 15:207–215. Masters, W.M. & H. Markl. 1981. Vibration signal transmission in spider orb webs. Science 213:363–365.

138

McClintock, W.J. & G.N. Dodson. 1999. Notes on Cyclosa insulana (Araneae, Araneidae) of Papua New Guinea. Journal of Arachnology 27:685–688. McGhee, K.E., S. Feng, S. Leasure & A.M. Bell. 2015. A female's past experience with predators affects male courtship and the care her offspring will receive from their father. Proceedings of the Royal Society B 282:20151840. Menda, G., P.S. Shamble, E.I. Nitzany, J.R. Golden & R.R. Hoy. 2014. Visual perception in the brain of a jumping spider. Current Biology 24:2580–2585. Merrett, P. 1988. Notes on the biology of the neotropical pisaurid, Ancylometes bogotensis (Keyserling) (Araneae: Pisauridae). Bulletin of the British Arachnological Society 7:197– 201. Mortimer, B., S.D. Gordon, C. Holland, C.R. Siviour, F. Vollrath & J.F. Windmill. 2014. The speed of sound in silk: linking material performance to biological function. Advanced Materials 26:5179–5183. Mortimer, B., C. Holland, J.F. Windmill & F. Vollrath. 2015. Unpicking the signal thread of the sector web spider Zygiella x-notata. Journal of The Royal Society Interface 12:20150633. Mortimer, B., A. Soler, C.R. Siviour, R. Zaera & F. Vollrath. 2016. Tuning the instrument: sonic properties in the spider's web. Journal of The Royal Society Interface 13:20160341. Moskalik, B., & G.W. Uetz. 2011. Experience with chemotactile cues indicating female feeding history impacts male courtship investment in the wolf spider Schizocosa ocreata. Behavioral Ecology and Sociobiology 65:2175-2181. Moura, R.R., J. Vasconcellos-Neto & M. de Oliveira Gonzaga. 2017. Extended male care in Manogea porracea (Araneae: Araneidae): the exceptional case of a spider with amphisexual care. Animal Behaviour 123:1–9. Naftilan, S.A. 1999. Transmission of vibrations in funnel and sheet spider webs. International Journal of Biological Macromolecules 24:289–293. Neff, B.D. & E.I. Svensson. 2013. Polyandry and alternative mating tactics. Philosophical Transactions of the Royal Society B 368:20120045. Nitzsche, R.O.M. 1988. “Brautgeschenk” und umspinnen der beute bei Pisaura mirabilis, Dolomedes fimbriatus und Thaumasia uncata (Arachnida, Araneida, Pisauridae). Verhandlungen des Naturwissenschaftlichen Vereins Hamburg 20:353–393. Nitzsche, R.O.M. 2011. Courtship, mating and agonistic behaviour in Pisaura mirabilis (Clerck, 1757). Arachnology 15:93–120. Nelson, X.J., C.M. Warui & R.R. Jackson. 2012. Widespread reliance on olfactory sex and species identification by lyssomanine and spartaeine jumping spiders. Biological Journal of the Linnean Society 107:664–677. Pandulli-Alonso, I., A. Quaglia & M.J. Albo 2017. Females of a gift-giving spider do not trade sex for food gifts: a consequence of male deception? BMC Evolutionary Biology 17:112. Papke, M., S. Schulz, H. Tichy, E. Gingl & R. Ehn. 2000. Identification of a new sex pheromone from the silk dragline of the tropical wandering spider Cupiennius salei. Angewandte Chemie International Edition 39:4339–4341.

139

Papke, M.D., S.E. Riechert & S. Schulz. 2001. An airborne female pheromone associated with male attraction and courtship in a desert spider. Animal Behaviour 61:877–886. Parker, G.A. 1970. Sperm competition and its evolutionary consequences in the insects. Biological Reviews 45:525–567. Parkhe, A.D., S.K. Seeley, K. Gardner, L. Thompson & R.V. Lewis. 1997. Structural studies of spider silk proteins in the fiber. Journal of Molecular Recognition 10:1–6. Peakall, D.B. & P.N. Witt. 1976. The energy budget of an orb web-building spider. Comparative Biochemistry and Physiology Part A: Physiology 54:187–190. Peaslee, J.E. & W.B. Peck. 1983. The biology of Octonoba octonarius (Muma) (Araneae: Uloboridae). Journal of Arachnology 11:51–67. Persons, M.H. & A.L. Rypstra. 2001. Wolf spiders show graded antipredator behavior in the presence of chemical cues from different sized predators. Journal of Chemical Ecology 27:2493–2504. Persons, M.H., S.E. Walker & A.L. Rypstra. 2002. Fitness costs and benefits of antipredator behavior mediated by chemotactile cues in the wolf spider Pardosa milvina (Araneae: Lycosidae). Behavioral Ecology 13:386–392. Platnick, N. 1971. The evolution of courtship behaviour in spiders. Bulletin of the British Arachnological Society 2:40–47. Prenter, J., R.W. Elwood & W.I. Montgomery. 1994a. Alternative mating behaviour in Metellina segmentata (Clerck). Newsletter of the British Arachnological Society 70:10–11. Prenter, J., R. Elwood & S. Colgan. 1994b. The influence of prey size and female reproductive state on the courtship of the autumn spider, Metellina segmentata: a field experiment. Animal Behaviour 47:449–456. Preston-Mafham, K.G. 1999. Notes on bridal veil construction in Oxyopes schenkeli Lessert, 1927 (Araneae: Oxyopidae) in Uganda. Bulletin of the British Arachnological Society 11:150–152. Prestwich, K.N. 1977. The energetics of web-building in spiders. Comparative Biochemistry and Physiology Part A: Physiology 57:321–326. Prokop, P. & M.R. Maxwell. 2009. Female feeding regime and polyandry in the nuptially feeding nursery web spider, Pisaura mirabilis. Naturwissenschaften 96:259–265. Prokop, P. & M. Semelbauer. 2017. Biometrical and behavioural associations with offering nuptial gifts by males in the spider Pisaura mirabilis. Animal Behaviour 129:189–196. Richman, D.B. & R.R Jackson. 1992. A review of the ethology of jumping spiders (Araneae, Salticidae). Bulletin of the British Arachnological Society 9:33-37. Rinaldi, I.M., & A.A. Stropa. 1998. Sexual behaviour in Loxosceles gaucho Gertsch (Araneae, Sicariidae). Bulletin of the British Arachnological Society 11:57–61. Rising, A., M. Widhe & J. Johansson. 2011. Spider silk proteins: recent advances in recombinant production, structure–function relationships and biomedical applications. Cellular and Molecular Life Sciences 68:169–184.

140

Roberts, J.A. & G.W. Uetz. 2005. Information content of female chemical signals in the wolf spider, Schizocosa ocreata: male discrimination of reproductive state and receptivity. Animal Behaviour 70:217–223. Robinson, M.H. & B. Robinson. 1973: Ecology and behavior of the giant wood spider Nephila maculata (Fabricius) in New Guinea. Smithsonian Contributions to Zoology 149:1–176. Robinson, M.H. & B. Robinson. 1980. Comparative studies of the courtship and mating behavior of tropical araneid spiders. Pacific Insects Monograph 36:1–218. Robinson M.H. & Y.D. Lubin. 1979. Specialists and generalists: the ecology and behavior of some web-building spiders from Papua New Guinea. Pacific Insects 21:133–164. Roland, C. 1984. Chemical signals bound to the silk in spider communication (Arachnida, Araneae). Journal of Arachnology 309–314. Roland, C. & J.S. Rovner. 1983. Chemical and vibratory communication in the aquatic pisaurid spider Dolomedes triton. Journal of Arachnology 11:77–85. Ross, K. & R.L. Smith. 1979. Aspects of the courtship behavior of the black widow spider, Latrodectus hesperus (Araneae: Theridiidae), with evidence for the existence of a contact sex pheromone. Journal of Arachnology 7:69–77. Rovner, J. 1968. Territoriality in the sheet-web spider Linyphia triangularis (Clerck) (Araneae, Linyphiidae). Zeitschrift für Tierpsychologie 25:232–242. Rundus, A.S., R. Biemuller, K. DeLong, T. Fitzgerald & S. Nyandwi. 2015. Age-related plasticity in male mate choice decisions by Schizocosa retrorsa wolf spiders. Animal Behaviour 107:233–238. Rypstra, A.L., C. Wieg, S.E. Walker & M.H. Persons. 2003. Mutual mate assessment in wolf spiders: differences in the cues used by males and females. Ethology 109:315–325. Rypstra, A.L. & C.M. Buddle. 2013. Spider silk reduces insect herbivory. Biology Letters 9: 20120948. Rypstra, A.L., S.E. Walker & M.H. Persons. 2015. Cautious versus desperado males: predation risk affects courtship intensity but not female choice in a wolf spider. Behavioral Ecology 27:876-885. Schmitt, A. 1992. Conjectures on the origins and functions of a bridal veil spun by the males of Cupiennius coccineus (Araneae, Ctenidae). Journal of Arachnology 20:67–68. Schneider, J.M. & M.C.B. Andrade. 2011. Mating behaviour and sexual selection. In Spider Behaviour: Flexibility and Versatility. (M.E. Herberstein, ed.). Cambridge University Press, Cambridge, UK. Schneider, J.M. & Y. Lubin. 1998. Intersexual conflict in spiders. Oikos 83:496–506. Schneider, J.M., C. Lucass, W. Brandler & L. Fromhage. 2011. Spider males adjust mate choice but not sperm allocation to cues of a rival. Ethology 117:970–978. Schulz, S. 2013. Spider pheromones–a structural perspective. Journal of Chemical Ecology 39:1–14. Schulz, S. & S. Toft. 1993. Identification of a sex pheromone from a spider. Science 260: 1635– 1635.

141

Scott, C., S. Vibert & G. Gries. 2012. Evidence that web reduction by western black widow males functions in sexual communication. The Canadian Entomologist 144:672–678. Scott, C., McCann, S., Gries, R., Khaskin, G., & Gries, G. 2015a. N-3-Methylbutanoyl-O- methylpropanoyl-L-serine methyl ester–pheromone component of western black widow females. Journal of Chemical Ecology 41:465–472. Scott, C., D. Kirk, S. McCann & G. Gries. 2015b. Web reduction by courting male black widows renders pheromone-emitting females' webs less attractive to rival males. Animal Behaviour 107:71–78. Scott, C., Gerak, C., McCann, S., & G. Gries. 2017. The role of silk in courtship and chemical communication of the false widow spider, Steatoda grossa (Araneae: Theridiidae). Journal of Ethology https://link.springer.com/article/10.1007/s10164-017-0539-3 Scott, J.L., A.Y. Kawahara, J.H. Skevington, S.H. Yen, A. Sami, M.L. Smith & J.E. Yack. 2010. The evolutionary origins of ritualized acoustic signals in caterpillars. Nature Communications 1:1–9. Segoli, M., R. Arieli, P. Sierwald, A.R. Harari & Y. Lubin. 2008. Sexual cannibalism in the brown widow spider (Latrodectus geometricus). Ethology 114:279–286. Shulov, A. 1940. On the biology of two Latrodectus species in Palestine. Proceedings of the Linnean Society of London 152:309–328. Sierwald, P. 1988. Notes on the behavior of Thalassius spinosissimus (Arachnida: Araneae: Pisauridae). Psyche 95:243–252. Silva, E.L.C. 2005. New data on the distribution of Trechalea cezariana Mello-Leitão, 1931 in Southern Brazil. Acta Biologica Leopoldensia 27:127–128. Silva, E.L.C. & A.A. Lise. 2009. New record of nuptial gift observed in Trechalea amazonica (Araneae, Lycosoidea , Trechaleidae). Revista Peruana de Biología 16:119–120. Singer, F., S.E. Riechert, H. Xu, A.W. Morris, E. Becker, J.A. Hale et al. 2000. Analysis of courtship success in the funnel-web spider Agelenopsis aperta. Behaviour 137:93–117. Smithers, R.H.N. 1944. Contribution to our knowledge of the genus Latrodectus (Araneae) in South Africa. Annals of the South African Museum 36:263–312. Stålhandske, P. & B. Gunnarsson. 1996. Courtship behaviour in the spider Pityohyphantes phrygianus (Linyphiidae Araneae): do females discriminate injured males? Revue Suisse de Zoologie Proceedings of the XIIIth Congress of Arachnology: 617–625. Stålhandske, P. 2001. Nuptial gift in the spider Pisaura mirabilis maintained by sexual selection. Behavioral Ecology 12:691–697. Starr, C.K. 1988. Sexual behaviour in Dictyna volucripes (Araneae, Dictynidae). Journal of Arachnology 16:321–330. Stoltz, J.A. & M.C.B. Andrade. 2009. Female's courtship threshold allows intruding males to mate with reduced effort. Proceedings of the Royal Society B 277:585–592. Stowe, M. K. 1978. Observations of two nocturnal orbweavers that build specialized webs: Scoloderus cordatus and Wixia ectypa (Araneae: Araneidae). Journal of Arachnology 6:141–146.

142

Struble, D.L. & H. Arn. 1984. Combined gas chromatography and electroantennogram recording of insect olfactory responses. Pp. 161–178. In Techniques in Pheromone Research. (H.E. Hummel & T.A. Miller, eds.). Springer, New York, NY. Suter, R.B. & A.J. Hirscheimer. 1986. Multiple web-borne pheromones in a spider Frontinella pyramitela (Araneae: Linyphiidae). Animal Behaviour 34:748–753. Suter, R. B. & G. Renkes. 1982. Linyphid spider courtship: releaser and attractant functions of a contact sex pheromone. Animal Behaviour 30:714–718. Suter, R.B. & G. Renkes. 1984. The courtship of Frontinella pyramitela (Araneae, Linyphiidae): patterns, vibrations and functions. Journal of Arachnology 12:37–54. Suter, R.B., C.M. Shane & A.J. Hirscheimer. 1987. Communication by cuticular pheromones in a linyphiid spider. Journal of Arachnology 15:157–162. Taylor, P.W. 1998. Dragline-mediated mate-searching in Trite planiceps (Araneae, Salticidae). Journal of Arachnology 26:330–334. Tichy, H., E. Gingl, R. Ehn, M. Papke & S. Schulz. 2001. Female sex pheromone of a wandering spider (Cupiennius salei): identification and sensory reception. Journal of Comparative Physiology A 187:75–78. Tietjen, W.J. 1977. Dragline-following by male lycosid spiders. Psyche 84:165–178. Tietjen, W.J. 1978. Tests for olfactory communication in four species of wolf spiders (Araneae, Lycosidae). Journal of Arachnology 6:197–206. Trabalon, M. 2013. Chemical communication and contact cuticular compounds in spiders. Pp. 125–140. In Spider ecophysiology. (W. Nentwig, ed.) Springer, Heidelberg. Toft, S. & M.J. Albo 2015. Optimal numbers of matings: the conditional balance between benefits and costs of mating for females of a nuptial gift‐giving spider. Journal of Evolutionary Biology 28:457–467. Toft, S. & M.J. Albo 2016. The shield effect: nuptial gifts protect males against pre-copulatory sexual cannibalism. Biology Letters 12:20151082. Trillo, M.C. 2016. Comportamiento sexual y deposición de velo nupcial en la araña errante Ctenus longipes (Ctenidae). Unpublished thesis. Universidad de la República Uruguay. Trillo, M.C., V. Melo-González & M.J. Albo 2014. Silk wrapping of nuptial gifts as visual signal for female attraction in a crepuscular spider. Naturwissenschaften 101:123–130. Uetz, G.W., A. McCrate & C.S. Hieber. 2010. Stealing for love? Apparent nuptial gift behavior in a kleptoparasitic spider. Journal of Arachnology 38:128–131. Uhl, G. 2013. Spider olfaction: attracting, detecting, luring and avoiding. Pp. 141–157. In Spider Ecophysiology (W. Nentwig, ed.) Springer, Heidelberg. Uhl, G. & D.O. Elias. 2011. Communication. Pp. 127–189. In Spider Behaviour: Flexibility and Versatility. (M.E. Herberstein, ed.). Cambridge University Press, Cambridge, UK. Vahed, K. 1998. The function of nuptial feeding in insects: a review of empirical studies. Biological Reviews 73:44–78. Vahed, K. 2007. All that glisters is not gold: Sensory bias, sexual conflict and nuptial feeding in insects and spiders. Ethology 113:105–127.

143

Van Helsdingen, P.J. 1965. Sexual behaviour of Lepthyphantes leprosus (Ohlert) (Araneida, Linyphiidae) with notes on the function of the genital organs. Zoologische Mededelingen 41:15–44. Vibert, S., C. Scott & G. Gries. 2014. A meal or a male: the ‘whispers’ of black widow males do not trigger a predatory response in females. Frontiers in Zoology 11:4. Vibert, S., C. Scott & G. Gries. 2016. Vibration transmission through sheet webs of hobo spiders (Eratigena agrestis) and tangle webs of western black widow spiders (Latrodectus hesperus). Journal of Comparative Physiology A 202:749–758. Vollrath, F. 1979. Vibrations: their signal function for a spider kleptoparasite. Science 205:1149– 1151. Vollrath, F. & D.P. Knight. 2001. Liquid crystalline spinning of spider silk. Nature 410:541–548. Watanabe, T. 2000. Web tuning of an orb-web spider, Octonoba sybotides, regulates prey- catching behaviour. Proceedings of the Royal Society B 267:565–569. Watson, P. 1986. Transmission of a female sex pheromone thwarted by males in the spider Linyphia litigiosa (Linyphiidae). Science 233:219–221. Weldingh, D.L., S. Toft & O.N. Larsen. 2011. Mating duration and sperm precedence in the spider Linyphia triangularis. Journal of Ethology 29:143–152. Wheeler, W.C., J.A. Coddington, L.M. Crowley, D. Dimitrov, P.A. Goloboff, C.E. Griswold et al. 2016. The spider tree of life: phylogeny of Araneae based on target‐gene analyses from an extensive taxon sampling. Cladistics 2016:1–43. Whitehouse, M. & R.R. Jackson. 1994. Intraspecific interactions of Argyrodes antipodiana, a kleptoparasitic spider from New Zealand. New Zealand Journal of Zoology 21:243–168. Wignall, A.E. & M.E. Herberstein. 2013a. The influence of vibratory courtship on female mating behaviour in orb-web spiders (Argiope keyserlingi, Karsch 1878). PLoS ONE 8:e53057. Wignall, A.E. & M.E. Herberstein. 2013b. Male courtship vibrations delay predatory behaviour in female spiders. Scientific Reports 3:3557. Wignall, A.E., D.J. Kemp & M.E. Herberstein. 2014. Extreme short-term repeatability of male courtship performance in a tropical orb-web spider. Behavioral Ecology 25:1083–1088. Willey Robertson, M. & P.H. Adler. 1994. Mating behavior of Florinda coccinea (Hentz) (Araneae: Linyphiidae). Journal of Insect Behavior 7:313–326. Wolff, J.O., M. Řezáč, T. Krejčí & S.N. Gorb. 2017. Hunting with sticky tape: functional shift in silk glands of araneophagous ground spiders (Gnaphosidae). Journal of Experimental Biology 220:2250–2259. Wolfner, M.F. 2002. The gifts that keep on giving: physiological functions and evolutionary dynamics of male seminal proteins in Drosophila. Heredity 88:85–93. Wyatt, T. 2014. Pheromones and Animal Behavior, 2nd ed. Cambridge University Press, Cambridge, UK. Xiao, Y., J. Zhang & S. Li. 2009. A two-component female-produced pheromone of the spider Pholcus beijingensis. Journal of Chemical Ecology 35:769–778.

144

Xiao, Y.H., J.X. Zhang & S.Q. Li. 2010. Male-specific (Z)-9-tricosene stimulates female mating behaviour in the spider Pholcus beijingensis. Proceedings of the Royal Society B 277:3009–3018. Yáñez, M., A. Locht & R. Macías-Ordóñez. 1999. Courtship and mating behavior of Brachypelma klaasi (Araneae, Theraphosidae). Journal of Arachnology 27:165–170. York, J.R. & T.A. Baird. 2017. Sexual selection on male collared lizard (Crotaphytus collaris) display behaviour enhances offspring survivorship. Biological Journal of the Linnean Society 122:176–183. Zhang, S., M. Kuntner & D. Li. 2011. Mate binding: male adaptation to sexual conflict in the golden orb-web spider (Nephilidae: Nephila pilipes). Animal Behaviour 82:1299–1304.

Table 4-1 Cross tabulation of potential functions (columns) and mechanisms (rows) of silk use by male spiders in courtship and mating. Symbols in filled cells indicate functions and their mechanisms for which there is experimental evidence. Symbols are as in Fig. 1 (web indicates silk addition to or reduction of the female’s web; spider indicates silk deposition on the female’s body; insect indicates silk-associated nuptial gift). ✗ indicates experimental evidence against a specific function/mechanism and ‘‘nt’’ indicates that, to our knowledge, no test has been conducted.

Function → Increased success: current mating Decreased polyandry

Increased female Increased sperm Decreased risk of Decreased risk of rival Decreased female Mechanisms receptivity/probability transfer (copulation injury or death males courting/mating receptivity to remating of mating # or duration)

INDIRECT:

Communication

Chemical nt (1) (2) (2) (3,4)

Visual nt nt (5-7)

Tactile nt nt nt (2) (2)

Vibratory (correlated nt nt nt nt nt effect)

DIRECT:

Structural effects

Mobility nt nt nt nt nt

145 146

Signal Transmission nt nt nt nt (3,4)

DIRECT:

Physical effects

Physical constraints ✗ (8) nt nt (8) (9)

Deceit nt nt nt (10,11) (10,11)

(1) Paratrechalea ornata (Trechaleidae); Brum et al. 2012 (silk extract alone is sufficient to elicit female gift acceptance) (2) Nephila pilipes (Araneidae); Zhang et al. 2011 (chemical and tactile effects of binding on cannibalism and sperm transfer duration) (3) Neriene litigiosa (Linyphiidae); Watson 1986 (female pheromone emission/attractiveness decreased by web reduction and females remain unattractive after mating) (4) Latrodectus hesperus (Theridiidae); Scott et al. 2015b (female pheromone emission/attractiveness decreased by web reduction and females remain unattractive after mating) (5) Pisaura mirabilis (Pisauridae); Stålhandske 2002 (females accept brighter gifts more quickly) (6) Paratrechalea ornata (Trechaleidae); Trillo et al. 2014 (males with white painted chelicerae had higher mating success than those without, in absence of prey item) (7) Paratrechalea ornata (Trechaleidae); Klein et al. 2014 (small bright gifts accepted more quickly than large dark ones, but does not exclude chemical/tactile cues) (8) Pisaurina mira (Pisauridae); Anderson & Hebets 2017 (male silk does not affect copulation success but males that mate-bind achieve increased sperm transfer and fertilization success) (9) Pisaurina mira (Pisauridae); Anderson & Hebets 2016 (silk wrapping physically restrains females, reducing male’s risk of cannibalism following/during sperm transfer and increasing the number of insertions achieved) (10) Pisaura mirabilis (Pisauridae); Albo et al. 2011 (males presenting worthless gifts achieve similar copulation success to those with prey gifts, and significantly higher copulation success than males without gifts; males with worthless gifts achieve longer copulation duration than those without gifts, but shorter duration than those with prey gifts) (11) Paratrechalea ornata (Trechaleidae); Albo et al. 2014 (males with worthless gifts achieve copulations while males without gifts do not; copulation duration— correlated with sperm transfer amount—is similar for males with worthless gifts and prey gifts)

147 Table 4-2 Spider taxa in which males modify the female’s web or other silken structures by adding and/or removing silk (web reduction). Y = yes; N = no; P = probable; n = number of mating interactions observed. Where data are available, the specific behavior is described in brackets, as is the percentage of males that engage in that behaviour, with a superscript indicating the reference specific to these data where necessary.

Taxon Addition of silk Web reduction Citations Agelenidae Agelenopsis actuosa (Gertsch & Ivie, 1936) P (abdomen waggle) N Galasso 2012 Agelenopsis aperta (Gertsch, 1934) Y (abdomen waggle) N Singer et al. 2000; Galasso 2012; S. Reichert pers. comm. Agelenopsis aleenae Chamberlin & Ivie, 1935 P (abdomen waggle) N Galasso 2012 Agelenopsis emertoni Chamberlin & Ivie, 1935 P (abdomen waggle) N Galasso 2012 Agelenopsis kastoni Chamberlin & Ivie, 1941 P (abdomen waggle) N Galasso 2012 Agelenopsis naevia (Walckenaer, 1841) P (abdomen waggle) N Galasso 2012 Agelenopsis oklahoma (Gertsch, 1936) P (abdomen waggle) N Galasso 2012 Agelenopsis pennsylvanica (C. L. Koch, 1843) P (abdomen waggle) N Galasso 2012 Agelenopsis potteri (Blackwall, 1846) P (abdomen waggle) N Galasso 2012 Agelenopsis spatula Chamberlin & Ivie, 1935 P (abdomen waggle) N Galasso 2012 Agelenopsis utahana (Chamberlin & Ivie, P (abdomen waggle) N Galasso 2012 1933) Barronopsis texana (Gertsch, 1934) P (abdomen waggle) N Galasso 2012 Eratigena agrestis (Walckenaer, 1802) Y (depositing silk) N Vibert et al. 2014 Araneidae Aetrocantha falkensteini Karsch, 1879 Y (mating thread) N Robinson & Robinson 1980 Alpaida veniliae (Keyserling, 1865) Y (mating thread) N Benamú et al. 2012, 2015

148

Araneus diadematus Clerck, 1757 Y (mating thread) N Elgar & Nash 1988 Araneus quadratus Clerck, 1757 Y (mating thread) N Elgar 1991 Argiope aemula (Walckenaer, 1841) Y (‘miniweb’ within web & N Robinson & Robinson 1980 silk at hub) Argiope aetherea (Walckenaer, 1841) Y (mating thread) Y (small hole) Robinson & Robinson 1980 Y (dragline silk on web & Y (small hole) or N Robinson & Robinson 1980 Argiope argentata (Fabricius, 1775) female’s dragline or multistranded mating thread inside or outside web) Argiope aurantia Lucas, 1833 Y (miniweb within web & Y (small hole) Robinson & Robinson 1980 silk on web) Argiope aurocincta Pocock, 1898 Y (mating thread & silk dep Y (small hole) Robinson & Robinson 1980 during walkabouts) Argiope bruennichi (Scopoli, 1772) Y (mating thread) Y (small hole) Elgar 1991 Argiope cuspidata Thorell, 1859 Y (mating thread & silk dep Y (small hole) Robinson & Robinson 1980 during walkabouts) Argiope flavipalpis (Lucas, 1858) Y (mating thread) Y (small hole) Robinson & Robinson 1980 Argiope florida Chamberlin & Ivie, 1944 Y (mating thread; silk dep Y (small hole) Robinson & Robinson 1980 during walkabouts) Argiope keyserlingi Karsch, 1878 Y (mating thread) Y (small hole) or N Herberstein et al. 2002 Argiope ocyaloides L. Koch, 1871 Y (mating thread) Y (small hole) Robinson & Robinson 1980 Argiope picta L. Koch, 1871 Y (multistranded mating Y (large hole; male may Robinson & Robinson 1980; thread) increase size between Elgar 1991 courtship bouts) Argiope radon Levi, 1983 Y (mating thread) Y (small hole) Robinson & Robinson 1980; Wignall et al. 2014

149

Argiope reinwardti (Doleschall, 1859) Y (extensive silk dep. at hub) N Robinson & Robinson 1980 Argiope submaronica Strand, 1916 Y (mating thread & silk dep. Y (small hole) Robinson & Robinson 1980 on web) Argiope submaronica Strand, 1916 Y (mating thread) Y (small hole) Robinson & Robinson 1980 (as Argiope savignyi) Cyclosa caroli (Hentz, 1850) Y (mating thread) N Robinson & Robinson 1980 Cyclosa insulana (Costa, 1843) Y (multistranded mating N Robinson & Robinson 1980 thread) Cyclosa insulana (Costa, 1843) Y (silk laid down onto Y (cuts threads of rival McClintock & Dodson 1999 guylines of web; n = 4) males, web size reduced by 40%; n = 1) Cyrtophora moluccensis (Doleschall, 1857) Y (mating thread) N Berry 1987 Eriophora fuliginea (C. L. Koch, 1838) Y (mating thread) N Robinson & Robinson 1980 Eriophora transmarina (Keyserling, 1965) Y (mating thread) Y (small hole) Elgar 1991 (as Eriophora transmarinus) Gasteracantha cencriformis (Linnaeus, 1758) Y (mating thread) N Robinson & Robinson 1980; Bukowski et al. 2001 Gasteracantha curvispina (Guérin, 1837) Y (mating thread) N Robinson & Robinson 1980 Gea C. L. Koch, 1843 sp. Y (mating thread) N Robinson & Robinson 1980 Herennia multipuncta (Doleschall, 1859) Y (dragline silk) N Robinson & Robinson 1980 (as Herennia ornatissima) Isoxya cicatricosa (C. L. Koch, 1844) Y (mating thread) N Robinson & Robinson 1980 (as Isoxya cicatrosa) Isoxya tabulata (Thorell, 1859) Y (‘treadmill’-type mating Y (cuts some web Robinson & Robinson 1980 thread) elements) Kapogea sexnotata (Simon, 1895) Y (multistranded mating Y (cuts away extensive Robinson & Robinson 1980 thread & silk dep. on web) portion of lower snare) (as Cyrtophora nympha) Leviellus thorelli (Ausserer, 1871) Y (mating thread) N Kralj-Fišer et al. 2013

150

Mangora bimaculata (O. Pickard-Cambridge, Y (converts radius to mating Y (removes viscid spiral Robinson & Robinson 1980 1889) thread; n = 3) elements on either side of mating thread; n = 3) Mecynogea lemniscata (Walckenaer, 1841) Y (mating thread & silk N Robinson & Robinson 1980 deposition on snare) Micrathena clypeata (Walckenaer, 1805) Y (mating thread) N Robinson & Robinson 1980 Micrathena duodecimspinosa (O. Pickard- Y (mating thread) N Robinson & Robinson 1980 Cambridge, 1890) Micrathena gracilis (Walckenaer, 1805) Y (mating thread) N Bukowski & Christenson 2000 Micrathena sagittata (Walckenaer, 1841) Y (mating thread) N Robinson & Robinson 1980 Micrathena schreibersi (Perty, 1833) Y (mating thread) N Robinson & Robinson 1980 Micrathena sexspinosa (Hahn, 1822) Y (converts one radius to Y (removes viscid spiral Robinson & Robinson 1980 thick, multistranded mating elements on either side of thread) mating thread radius) Nephila edulis (Labillardière, 1799) Y (dragline silk) N Robinson & Robinson 1980 Nephila pilipes (Fabricius, 1793) Y (dragline silk; extensive, N Robinson & Robinson 1980 incl. on females’ dragline) Nephila pilipes (Fabricius, 1793) Y (dragline silk) N Robinson & Robinson 1980 (as Nephila maculata) Nephila clavipes (Linnaeus, 1767) Y (dragline silk) N Robinson & Robinson 1980 Nephilengys malabarensis (Walckenaer, 1841) Y (dragline silk) N Robinson & Robinson 1980 Scoloderus cordatus (Taczanowski, 1879) Y (“treadmill”-type mating N Stowe 1978 thread) Thelacantha brevispina (Doleschall, 1857) Y (mating thread) N Robinson & Robinson 1980 (as Gasterocantha brevispina) Zilla spp. C. L. Koch, 1834 Y (multistranded mating N Robinson & Robinson 1980 thread) Zygiella x-notata (Clerck, 1757) Y (mating thread) N Blanke 1986 as cited by Dondale et al. 2003

151

Corrinidae Nyssus coloripes Walckenaer, 1805 Y (zigzags of silk laid down N Jackson & Poulsen 1990 (as onto female’s web) Supunna picta) Dictynidae Dictyna arundinacea (Linnaeus, 1758) Y (small ‘canopy’) Y (small hole) Locket 1926; Bristowe 1958 Dictyna tridentata Bishop & Ruderman, 1946 Y N Jackson 1979 Dictyna volucripes Keyserling, 1881 Y N Starr 1988 Mallos gregalis (Simon, 1909) Y N Jackson 1979 Mexitlia trivittata (Banks, 1901) Y N Jackson 1979 (as Mallos trivittatus) Dipluridae Microhexura montivaga Crosby & Bishop, Y (sometimes apply silk to N Coyle 1985 1925 female’s web) Thelechoris striatipes (Simon, 1889) Y (spin silk while courting; N Coyle & O’Shields 1990 (as 4%; n = 45) Thelechoris karschi) Filistatidae Kukulcania hibernalis (Hentz, 1842) Y (83%; n = 6) N Barrantes & Ramirez 2013 Linyphiidae Florinda coccinea (Hentz, 1850) Y Y (part of web; 75%; n = Willey Robertson & Adler 20) 1994 Lepthyphantes leprosus (Ohlert, 1865) Y Y (90-100% of web; van Helsdingen 1965 45%; n = 29)

Linyphia triangularis (Clerck, 1757) Y Y (part or all of web2; 2Rovner 1968 68%; n = 603) 3Weldingh et al. 2011 Neriene litigiosa (Keyserling, 1886) Y Y (large portions of web; Watson 1986 (as Linyphia 28%; n = 50) litigiosa)

152

Pityohyphantes phrygianus (C. L. Koch, 1836) unknown Y (web reduced to a 4Stålhandske & Gunnarsson small wad; n = 18)4 1996; Gunnarsson et al. 2004 Porrhomma egeria Simon, 1884 Y (adds threads to web) N Bourne 1978

Porrhothelidae

Porrhothele antipodiana (Walckenaer, 1837) Y (spins silk on web before & N Jackson & Pollard 1990 after copulation; 47%; n = 186) Psechridae Fecenia Simon, 1887 sp. N Y (most of web, leaving Robinson & Lubin 1979 single thread; n = 1) Sicariidae Loxosceles gaucho, Gertsch, 1967 N Y (often cut some threads Rinaldi & Stropa 1998 of web) Tetragnathidae Metellina segmentata (Clerck, 1757) Y (mating thread & wrapping Y (small section of web Prenter et al 1994b; Bristowe silk around prey item) cut out) 1929 Theraphosidae Brachypelma klaasi (Schmidt & Krause, 1994) Y (deposits silk around N Yáñez et al. 1999 female’s burrow & over her silk) Grammostola vachoni Schiapelli & Y (lays silk over female’s N Ferretti & Ferrero 2008 (as Gerschman, 1961 silk) Grammostola schulzei) Theridiidae Argyrodes antipodianus O. Pickard- Y Y Whitehouse & Jackson 1994 Cambridge, 1880 (as Argyrodes antipodiana) Argyrodes argyrodes (Walckenaer, 1841) Y (web-spinning; 14%; n = 7) Y Knoflach 2004 Dipoena melanogaster (C. L. Koch, 1837) Y (mating web) N Knoflach 2004

153

Echinotheridion gibberosum (Kulczyński, Y (mating web) N Knoflach 2004 1899) Enoplognatha afrodite Hippa & Oksala, 1983 Y (mating web; n = 4) N Knoflach 2004 Enoplognatha diversa (Blackwall, 1859) Y (mating web; n = 3) N Knoflach 2004 Enoplognatha latimana Hippa & Oksala, 1982 Y (mating web; n = 1) N Knoflach 2004 Enoplognatha macrochelis Levy & Amitai, Y (mating web; n = 5) N Knoflach 2004 1981 Enoplognatha ovata (Clerck, 1757) Y (mating web; n = 5) N Knoflach 2004 Enoplognatha quadripunctata Simon, 1884 Y (mating web; n = 2) N Knoflach 2004 Enoplognatha thoracica (Hahn, 1833) Y (mating web; n = 1) N Knoflach 2004 Kochiura aulica (C. L. Koch, 1838) Y (mating thread; n = 6) Y (hole cut for mating Knoflach 2004 (as Anelosimus thread) aulicus) Latrodectus dahli Levi, 1959 Y (50%; n = 2) N Knoflach & van Harten 2002 Latrodectus geometricus C. L. Koch, 1841 Y Y (‘less commonly’) Segoli et al. 2008 Latrodectus hasselti Thorell, 1870 Y Y Forster 1992, 1995

Latrodectus hesperus Chamberlin & Ivie, 1935 Y Y (up to 50% of web; Ross & Smith 1979; 1Scott et 58%; n = 121) al. 2012 Latrodectus mactans (Fabricius, 1775) Y Y Breene & Sweet 1985 Latrodectus pallidus O. Pickard-Cambridge, Y Y Harari et al. 2009 1872 Latrodectus revivensis Shulov, 1948 Y (69% of males) Y (up to 50% of barrier Anava & Lubin 1993 web) Paidiscura Archer, 1950 sp. Y (mating web) Knoflach 2004 Paidiscura pallens (Blackwell, 1834) Y Y Locket 1927 (as Theridion pallens) Parasteatoda tepidarorium (C. L. Koch, 1841) Y (web-spinning; n = 3) N Knoflach 2004 (as Achaearanea tepidariorum)

154

Parasteatoda wau (Levi, Lubin & Robinson, Y (small mating arena) Y (small area reduced) Lubin 1986 (as Achaearanea 1982) wau) Steatoda bipunctata (Linnaeus, 1758) Y (mating web; n = 3) Y (removed threads) Knoflach 2004 Steatoda castanea (Clerck, 1757) Y (mating web; n = 1) Knoflach 2004 Steatoda grossa (C. L. Koch, 1838) Y (all males added silk to Y (>50% of web; 74%; n Scott et al. 2017 female’s web; n = 23) = 23)

Y (mating web &/or web- N 2Knoflach 2004; Gwinner- spinning; n = 42) Hanke 1970 (as Teutana grossa) Steatoda paykulliana (Walckenaer, 1806) Y (silk-throwing; 66%; n = 3) N Knoflach 2004 Steatoda triangulosa (Walckenaer, 1802) Y (mating thread; n = 4) Y (removed threads; Knoflach 2004 50%; n = 4) Simitidion simile (C. L. Koch, 1836) Y N Locket 1927 (as Theridion simile) Theridion varians Hahn, 1833 Y Y Locket 1927 Tidarren argo Knoflach & van Harten, 2001 Y (mating web) N Knoflach 2004 Tidarren cuneolatum (Tullgren, 1910) Y (multistranded mating N Knoflach & van Harten 2000 thread) Uloboridae Octonoba sinensis (Simon, 1880) Y (mating thread) N Peaslee & Peck 1983 (as Octonoba octonarius) Uloborus Latreille, 1806 sp. Y (mating thread) N Bristowe 1958

155 Table 4-3 Spider taxa in which males deposit silk ‘bridal veils’ onto the female during courtship.

Taxon Context Type of ‘veil’ Reference

Agelenidae Eratigena agrestis (Walckenaer, 1802) female’s web some silk on legs & carapace S. Vibert unpublished data Araneidae Argiope aemula (Walckenaer, 1841) female’s web silk on carapace, legs & abdomen Robinson & Robinson 1980 (extensive) Argiope aurantia Lucas, 1833 female’s web draglines attached to abdomen Robinson & Robinson 1980 Argiope picta L. Koch, 1871 female’s web some silk on legs Robinson & Robinson 1980 Argiope Audouin, 1826 spp. female’s web silk on legs Robinson & Robinson 1980 Caerostris darwini Kuntner & Agnarsson, female’s web silk on legs & body (extensive) Gregorič et al. 2016 2010 Herennia multipuncta (Doleschall, 1859) female’s web silk on & around abdomen Robinson & Robinson 1980 (as Herennia ornatissima) Nephila pilipes female’s web silk between legs, between base of Robinson & Robinson 1980 (as abdomen & dorsal surface of Nephila maculata) cephalothorax (extensive) Nephila pilipes (Fabricius, 1793) female's web silk on carapace, legs & abdomen; Robinson & Robinson 1980; Kuntner connected to web (extensive) et al. 2009; Zhang et al. 2011 Corrinidae Nyssus coloripes Walckenaer, 1805 female’s web zigzags of silk placed on female’s Jackson & Poulsen 1990 (as Supunna body as male walks over her picta) Ctenidae Ancylometes bogotensis (Keyserling, substrate silk rings around front tibiae & Merrett 1988 1877) patellae (extensive) Ctenus longipes Keyserling, 1891 substrate silk on forelegs, pedipalps, chelicerae, Trillo 2016 & eyes (later consumed)

156

Cupiennius coccineus F. O. Pickard- substrate some silk on legs Schmitt 1992 Cambridge, 1901 Dictynidae Dictyna volucripes Keyserling, 1881 female's web some silk on female Starr 1988

Eutichuridae Eutichurus ibiuna Bonaldo, 1994 substrate legs I, II & palps tied to substrate Laborda & Simo 2015

Homalonychidae Homalonychus selenopoides Marx, 1891 substrate silk ring around legs Alvarado-Castro & Jiménez 2011 Homalonychus theologus Chamberlin, substrate silk ring around legs Domínguez & Jiménez 2005 1924 Lycosidae Schizocosa malitiosa (Tullgren, 1905) substrate legs I & II tied to substrate & silk Aisenberg et al. 2008 near mouthparts Oxyopidae Oxyopes schenkeli Lessert, 1927 hanging on silk spun around legs I, II, & III Preston-Mafham 1999 dragline Philodromidae Tibellus oblongus (Walckenaer, 1802) substrate some silk on female Kaston 1936; Preston-Mafham 1999 Tibellus Simon, 1875 sp. substrate some silk on female Platnick 1971

Pisauridae Dolomedes triton (Walckenaer, 1837) substrate legs I & II tied to substrate Carico 1993 Pisaurina mira (Walckenaer, 1837) substrate or silk spun around legs I & II Bruce & Carico 1988; Anderson & hanging on (extensive) Hebets 2016 dragline Nilus curtus (O. Pickard-Cambridge, female’s mating silk ring around patellae Sierwald 1988 (as Thalassius 1876) web spinosissimus)

157

Tetragnathidae Metellina segmentata (Clerck, 1757) female's web female wrapped with fine silk Bristowe 1929; Lopez 1987

Theridiidae Euryopis episinoides (Walckenaer, 1847) female's web some silk on female Knoflach 2004 Latrodectus geometricus C. L. Koch, female's web some silk on legs & body Knoflach & van Harten 2002; 1841 Segoli et al. 2008 Latrodectus hasselti Thorell, 1870 female's web some silk on legs & body Forster 1992 Latrodectus hesperus Chamberlin & Ivie, female's web some silk on legs & body Ross & Smith 1979; Kaston 1970; 1935 Herms et al. 1935; Scott et al. 2012 Latrodectus indistinctus O. Pickard- female's web some silk on legs & body Smithers 1944 Cambridge, 1904 Latrodectus mactans (Fabricius, 1775) female's web some silk on legs & body Breene & Sweet 1985 Latrodectus pallidus O. Pickard- female's web some silk on legs & body Shulov 1940 Cambridge, 1872 Latrodectus revivensis Shulov, 1948 female's web some silk on legs & body Anava & Lubin 1993 Latrodectus tredecimguttatus (Rossi, female's web some silk on legs & body Shulov 1940 1790) Steatoda bipunctata (Linnaeus, 1758) female’s web some silk on female Knoflach 2004 Steatoda grossa (C. L. Koch, 1838) female's web some silk on legs & body Scott et al. 2017 Steatoda paykulliana (Walckenaer, 1806) female’s web some silk on legs & body Knoflach 2004 Steatoda triangulosa (Walckenaer, 1802) female’s web some silk on female Knoflach 2004

Thomisidae Bassaniana versicolor (Keyserling, 1880) substrate female tied to substrate Kaston 1936 (as Coriarachne versicolor) Xysticus cristatus (Clerck, 1757) substrate legs I & II tied to substrate Bristowe 1931; Bristowe 1958

158

Xysticus lanio C. L. Koch 1835 substrate legs I & II tied to substrate Gerhardt 1924 as cited by Bristowe 1926 Xysticus audax (Schrank, 1803) substrate female tied to substrate Thomas 1930 as cited by Kaston 1936 (as Xysticus pini) Xysticus striatipes L. Koch, 1870 substrate female tied to substrate Sytschewskaja 1935 as cited by Kaston 1936 Xysticus triguttatus Keyserling, 1880 substrate female tied to substrate Kaston 1936 Xysticus tristrami (O. Pickard-Cambridge, substrate female tied to substrate Gerhardt 1933 as cited by Kaston 1872) 1936 Pycnaxis krakatauensis (Bristowe, 1931) substrate legs I & II tied to substrate Bristowe 1931 (as Xysticus krakatauensis) Uloboridae Uloborus Latreille, 1806 sp. female's web not described Gerhardt 1933 (as cited by Berendonck 2003) Zoropsidae Tengella perfuga Dahl, 1901 female’s web some silk on legs & carapace Mallis & Miller 2017

159 Table 4-4 Spider taxa in which males present females with silk-associated nuptial gifts, including silk-wrapped prey, silk alone, or silk-lined burrows.

Taxon Type of gift Reference

Araneidae Isoxya tabulata (Thorell, 1859) mating thread silk (probably) Robinson & Robinson 1980 Scoloderus cordatus (Taczanowski, 1879) mating thread silk (probably) Stowe 1978 Ctenidae Ctenus longipes Keyserling, 1891 bridal veil silk Trillo 2016 Lycosidae Allocosa alticeps (Mello-Leitão, 1944) silk-lined burrow Aisenberg et al. 2010 Allocosa senex (Mello-Leitão, 1945) silk-lined burrow Aisenberg et al. 2007 (as Allocosa brasiliensis); Caraballo et al. 2017 Pisauridae Pisaura lama Bösenberg & Strand, 1906 silk-wrapped prey Itakura 1993 (as cited by Costa-Schmidt et al. 2008) Pisaura mirabilis (Clerck, 1757) silk-wrapped prey Bristowe & Lockett 1926; Bristowe 1958 Perenethis fascigera (Bösenberg & Strand, 1906) silk-wrapped prey Itakura 1998 Thaumasia argenteonotata (Simon, 1898) silk-wrapped prey Nitzsche 1987 (as cited by Nitzsche 2011) Tinus peregrinus (Bishop, 1924) silk-wrapped prey J. Carico pers. comm. in Nitzsche 2011

Tetragnathidae Metellina segmentata (Clerck, 1757) silk-wrapped prey or rival male Prenter et al. 1994a

Theridiidae Argyrodes elevatus Taczanowski, 1873 spider lightly wrapped in silk Cobbold & Su 2010

160

stolen silk-wrapped prey Uetz et al. 2010

Theridiosomatidae Theridiosoma gemmosum (L. Koch, 1877) silk Hajer & Řeháková 2011

Trechaleidae Paratrechalea azul Carico, 2005 silk-wrapped prey Costa-Schmidt et al. 2008 Paratrechalea galianoe Carico, 2005 silk-wrapped prey Costa-Schmidt et al. 2008 Paratrechalea ornata (Mello-Leitão, 1943) silk-wrapped prey Costa-Schmidt et al. 2008 Trechalea amazonica F. O. Pickard-Cambridge, 1903 silk-wrapped prey Silva & Lise 2009 Trechalea bucculenta (Simon, 1898) silk-wrapped prey Silva 2005 (as cited by Silva & Lise 2009) Trechalea Thorell, 1869 sp. silk-wrapped prey Lapinski & Tschapka, 2009 (as cited by Nitzsche 2011)

161 Table 4-5 Other behaviors involving male silk deposition during courtship and mating. Note that silk deposition on the substrate is likely widespread in cursorial spiders, but is rarely explicitly mentioned in descriptions of courtship behavior.

Taxon Behavior Reference

Araneidae

Manogea porracea (C. L. Koch, 1838) Male builds web above female’s & protects egg sacs from Moura et al. 2017 predators Ctenidae Ctenus longipes Keyserling, 1891 Male deposits silk on the substrate prior to mounting the female Trillo 2016 (82% of males; n = 11 matings) Lycosidae

Pardosa milvina (Hentz, 1844) Male deposits silk on substrate in response to female silk cues Khan & Persons 2015 Salticidae Plexippus paykulli (Audouin, 1826) Male spins silk as he walks around outside/near female’s nest Jackson & Macnab 1989

162 Table 4-6 Spider taxa in which there is behavioral evidence for male-produced sex pheromones. These families are also indicated in red in fig. 1

Taxon Source Type Female response Reference

Agelenidae Agelenopsis aperta (Gertsch, 1934) body airborne quiescence/catalepsis Becker et al. 2005 Coelotes terrestris (Wider, 1834) silk contact orientation Roland 1984 Tegenaria domestica (Clerck, 1757) silk contact orientation Roland 1984

Lycosidae Allocosa alticeps (Mello-Leitão, 1944) body airborne courtship Aisenberg et al. 2010 Allocosa brasiliensis (Petrunkevitch, 1910) body airborne courtship Aisenberg et al. 2010 Pardosa milvina (Hentz, 1844) silk contact increased silk production Khan & Persons 2015 Trochosa C. L. Koch, 1847 sp. silk contact mate recognition Engelhardt 1964 (as cited by Uhl & Elias 2011) Pholcidae Pholcus beijingensis Zhu & Song, 1999 body airborne stimulates mating behaviour Xiao et al. 2010

Salticidae Evarcha culicivora Wesolowska & Jackson, body & silk airborne + courtship & attraction/mate Cross & Jackson 2013 2003 contact recognition Scytodidae Scytodes Latreille, 1804 sp. body & silk airborne mate choice Koh et al. 2009

Theridiidae Latrodectus hesperus Chamberlin & Ivie, 1935 silk contact courtship Ross & Smith 1979

Sicariidae

163

Loxosceles intermedia body unknown mate recognition & avoiding Fischer et al. 2009 cannibalism

164 Table 4-7 Spider taxa in which there is behavioral evidence for males responding to silk cues of conspecific males.

Taxon Source Type Male response Citations

Araneidae Nephila senegalensis (Walckenaer, silk contact avoidance/mate choice Schneider et al. 2011 1841) Linyphiidae Frontinella communis (Hentz, 1850) silk contact positive geotaxis Suter & Hirscheimer 1986 (as Frontinella pyramitela) cuticle contact aggressive behavior Suter et al. 1987

Lycosidae Pardosa amentata (Clerck, 1757) silk contact increased silk production Richter & Kraan 1970 Rabidosa rabida (Walckenaer, 1837) body airborne reduces exploratory behavior Tietjen 1978 (as Lycosa rabida) Schizocosa ocreata (Hentz, 1844) silk airborne & contact inhibits courtship Ayyagari & Tietjen 1987

Theridiidae Latrodectus hasselti (Thorell, 1870) body &/or silk airborne shift in development Kasumovic & Andrade 2006

Figure 4-1 Cladograms illustrating relationships between araneomorph spider families (based on Wheeler et al. 2016) and the occurrence of male silk and pheromone use. (a) Overview of the order Araneae. (b) Families in clade Synspermiata. (c) Families in clade Araneoidea. (d) Families in the marronoid clade. (e) Families in the Oval Calamistrum clade. (f) Families in clade Dionycha. Red type or symbols next to a clade (see legend) indicates that there is evidence for a given type of male silk use or the presence of male pheromones in at least one species in that clade (see Tables 2–5 for lists of species and references). Note that in the Mygalomorphae (families not shown on the figure) there are records of male silk deposition on the female’s web or silk for species in the following three families: Dipluridae, Porrhothelidae, and Theraphosidae.

165 166

Figure 4-2 Examples of silk deposition onto females’ webs during courtship. (a) Araneus diadematus (Araneidae) male and female hanging from the male’s mating thread, attached to the periphery of the female’s web (photo: Maria Hiles). (b) Web reduction with silk addition by a Latrodectus hesperus (Theridiidae) male. The male has dismantled part of the capture web (which would have filled the lower half of the photograph before he began web reduction behavior) and is wrapping it with his own silk (photo: Sean McCann).

167

Figure 4-3 Examples of silk ‘bridal veils’ applied to females’ legs and bodies during courtship. (a) Nephila pilipes (Araneidae) male depositing silk onto the female’s carapace, legs, and abdomen (photo: Shichang Zhang). (b) Xysticus cristatus (Thomisidae) female with silk on her forelegs and abdomen as she feeds on a prey item—note that the male is underneath her abdomen (photo: Ed Niewenhuys). (c) Latrodectus hesperus (“texanus” morph, formerly Latrodectus mactans texanus; Theridiidae) male depositing silk onto the female’s legs (photo: Sean McCann). (d) Pisaurina mira (Pisauridae) male wrapping a female’s legs with silk prior to sperm transfer (Photo: Alissa Anderson).

168

Figure 4-4 Examples of silk-wrapped nuptial gifts. (a) Female (right) Pisaura mirabilis (Pisauridae) accepts a silk- wrapped gift from a male (photo: Alan Lau). (b) A male (right) Metellina segmentata (Tetragnathidae) has wrapped a rival male in silk as a nuptial gift for the female (photo: Conall McCaughey).

169

This chapter was published in the Journal of Arachnology (Scott, C. E., Anderson, A. G., & Andrade, M. C. 2018. A review of the mechanisms and functional roles of male silk use in spider courtship and mating. The Journal of Arachnology. 46(2): 173-207. https://doi.org/10.1636/JoA- S-17-093.1) and is reprinted with permission of the journal.

Conclusion

The three data chapters (1-3) of this thesis comprise observational and experimental data that provide insights into how mate competition and mate choice are related to the spatial and temporal distribution of conspecifics, including potential mates and competitors. I have shown that chemical signals and cues link social context, demography, and behaviour in ways that allow

L. hesperus males and females plastically adjust reproductive decisions and tactics according to environmental conditions.

I found that mate-searching L. hesperus males use both personal and social information in the race to locate sexually receptive adult females (Chapter 1). Males can ‘parasitize’ the mate searching effort of rivals to rapidly locate females in an environment where mate searching is risky (88% of males perish without finding a female) and receptive females are rare (the OSR is extremely male-biased for most of the mating season). Access to the silk of rival males allowed searching males to move at higher speeds and to locate females even when personal information

(female sex pheromone) was unavailable. As a consequence of this social information use

(‘mate-search-copying’), some females may remain unmated in nature, altering the strength and direction of selection, with implications for the evolution of both male and female traits.

In the face of intense scramble competition over access to adult females whose receptivity is asynchronous and transient, selection may favour traits that allow males to find and mate with females before rivals, and even before females reach sexual maturity. I confirmed that immature mating—a tactic by which L. hesperus males can inseminate subadult females just prior to maturity (Barrufaldi & Andrade, in review)—occurs in nature (Chapter 2). Despite the absence of volatile chemical cues that would allow males to locate subadult females, the spatial and temporal distribution of females at my field site facilitates immature mating. Since this tactic

170 171 requires mate guarding, which selects for different male traits than pure scramble competition, in addition to constraining female choice relative to adult mating, this is likely to have important implications for sexual selection on males in nature.

Despite the high frequency of mate guarding and immature mating that we recorded in nature, L. hesperus females face a risk of remaining unmated. I showed that females exposed to natural levels of variation in density (proximity to the nearest conspecific female) experience differences in encounter rates with potential mates (Chapter 3). The first male visitor to isolated females was delayed by 5 days compared to clustered females, imposing a risk of costly delays to mating or failure to mate at all. Isolated females subsequently displayed decreased choosiness

(in terms of mate rejection and pre-copulatory cannibalism) relative to clustered females in mating trials. Female mate choice decisions thus form a link between ecological factors like population density and sexual selection on males.

Across these studies, I found that chemical cues provide an important source of information for L. hesperus males and females in the context of reproductive behaviour and decision-making. Female sex pheromone provides personal information that allows males to rapidly locate females from up to 60 m away (Chapter 1). This pheromone also provides a source of social information that allows males to locate subadult females who share microhabitats with signaling adults and thus facilitates mate guarding and immature mating (Chapter 2). At the same time, chemically signaling females provide social information about local density and mate availability to nearby females during development, resulting in female plasticity in choosiness

(Chapter 3). Although I cannot exclude the possibility that tactile cues associated with the silk are involved, it is very likely that chemicals on male silk (see Chapter 4) also provide an important source of information to conspecifics. Mate-searching males recognize the silk of

172 conspecifics rivals and use it as a source of social information that allows them to find receptive females efficiently even under conditions where personal information (female sex pheromone) is unavailable (Chapter 1). These draglines may also facilitate mate guarding and immature mating by allowing males to follow rivals to occupied microhabitats that contain subadults (Chapter 2).

Finally, male silk likely also contributes to the social information detected by females during development and adulthood that provides a reliable indicator of the local availability of potential mates (Chapter 3).

Plasticity in reproductive behaviour also emerges as a common thread through this thesis.

L. hesperus males choose to avoid or follow silk trails of rivals depending on environmental conditions (Chapter 1). Males also engage in one or both of two alternative reproductive tactics during their lifetimes: mating with immature females or with adults (Chapter 2). Similarly, females are plastic in their mate choice decisions (Chapter 3).

This work is grounded in detailed natural history observations and longitudinal data on a wild population. This allowed me to begin to understand how demographic structure influences the operation of sexual selection in nature, using experiments that tested hypotheses about reproductive behaviour under relevant ecological contexts. I showed that the OSR varies across a single season (Chapters 1 & 2) and that this results in a scramble competition mating system where there is likely to be strong selection on males to arrive at females’ webs before rivals.

Spatial structure and is an important factor that influences male mate-searching success

(Chapters 1 & 2) and female encounter rates with males as subadults and adults (Chapters 2 and

3) and will strongly influence the feasibility and fitness consequences of both male mating tactics and female choice.

173

In summary, this thesis represents an examination of the intersection of chemical information, the spatiotemporal distribution of mates and competitors, and mating tactics in a genus of spiders (Latrodectus) that is rapidly developing as model system for linking our understanding of mechanism, mating behaviour and ecology (e.g., Andrade 1996; Andrade 2003;

Andrade & Kasumovic 2005; Segoli et al. 2006, Pruitt et al. 2011; DiRienzo et al. 2013; Andrade

& MacLeod 2015; Baruffaldi & Andrade 2015; MacLeod & Andrade 2015; Scott et al. 2015b).

Latrodectus hesperus is a fascinating species in which to explore hypotheses related to mechanisms and fitness implications of mate searching and mate choice, including the critical importance of chemical signals and cues as personal and social information sources during development, mate searching, and mate assessment. Mate searching is poorly studied in general despite being the critical first step in sexual selection for a wide range of taxa, particularly in terrestrial arthropods (Thornhill & Alcock 1983; Herberstein et al. 2017). Moreover, while mate localization using sex pheromones is relatively well understood in insects and some aquatic arthropods, few studies examine how these function in nature in other taxa. Overall, these studies provide important, but rare, empirical data related to hypotheses about how competition and female choice, key factors that drive sexual selection and shape mating systems, are related to the spatiotemporal distribution of conspecifics and chemical information.

References

Andrade, M. C. B. (1996). Sexual selection for male sacrifice in the Australian redback spider. Science, 271(5245), 70-72. Andrade, M. C. B. (2003). Risky mate search and male self-sacrifice in redback spiders. Behavioral Ecology, 14(4), 531-538. Andrade, M. C., & Kasumovic, M. M. (2005). Terminal investment strategies and male mate choice: extreme tests of Bateman. Integrative and Comparative Biology, 45(5), 838-847.

174

Andrade, M. C. B., & MacLeod, E. C. (2015). Potential for CFC in black widows (genus Latrodectus): mechanisms and social context. In Peretti, A. V, & Aisenberg, A. (Eds.), Cryptic Female Choice in Arthropods (pp. 27-53). Switzerland: Springer International. Baruffaldi, L., & Andrade, M. C. B. (2015). Contact pheromones mediate male preference in black widow spiders: avoidance of hungry sexual cannibals? Animal Behaviour, 102, 25- 32. DiRienzo, N., Pruitt, J. N., & Hedrick, A. V. (2013). The combined behavioural tendencies of predator and prey mediate the outcome of their interaction. Animal Behaviour, 86(2), 317-322. Herberstein, M. E., Painting, C. J., & Holwell, G. I. (2017). Scramble competition polygyny in terrestrial arthropods. Advances in the Study of Behavior. 49, 237-295. MacLeod, E. C., & Andrade, M. C. B. (2014). Strong, convergent male mate choice along two preference axes in field populations of black widow spiders. Animal Behaviour, 89, 163- 169. Pruitt, J. N., DiRienzo, N., Kralj-Fišer, S., Johnson, J. C., & Sih, A. (2011). Individual-and condition-dependent effects on habitat choice and choosiness. Behavioral Ecology and Sociobiology, 65(10), 1987-1995. Scott, C., Kirk, D., McCann, S., & Gries, G. (2015). Web reduction by courting male black widows renders pheromone-emitting females' webs less attractive to rival males. Animal Behaviour, 107, 71-78. Segoli, M., Harari, A. R., & Lubin, Y. (2006). Limited mating opportunities and male monogamy: a field study of white widow spiders, Latrodectus pallidus (Theridiidae). Animal Behaviour, 72(3), 635-642. Thornhill, R., & Alcock, J. (1983). The Evolution of Insect Mating Systems. Cambridge, MA: Harvard University Press.

175

Appendix 1

Supplementary Material:

Male black widows parasitize mate-searching effort of rivals to find females faster

CONTENTS:

1. Supplementary Methods (pp. 1-5)

2. Supplementary Tables S1-S4 (pp. 6-7)

3. Supplementary Figures S1-S4 (pp. 8-12)

176

1. Supplementary Methods

(a) General field survey methods

From April to September 2016 we surveyed the population of Latrodectus hesperus inhabiting a ca. 20 x 200 m area of our study site. On a single day each month, we turned over every driftwood log and rock within this area to perform a systematic population census. The number, sex, and age class of spiders (determined by inspection; Kaston 1970) under each microhabitat was recorded, and all spiders that were within two moults of maturity were briefly collected, immobilized between a soft sponge and a mesh cloth, and marked with a unique colour code using quick-drying modelling paint (Testor’s enamel; Fig. S1) before being returned to their webs (where they remained, see Salomon 2008). We also performed more frequent nocturnal surveys (Fig. 1) that involved non-intrusive inspection of the webs of all marked spiders to record presence/absence, evidence of moulting, and presence of males or other unmarked spiders on webs of females. Spiders that had moulted were marked again and returned to their webs.

(b) Estimating the abundance of sexually receptive females.

Unmated L. hesperus females produce an airborne sex pheromone that attracts males at long range (Kasumovic & Andrade 2004; MacLeod & Andrade 2014). The exact timing of pheromone production has not been studied for this species but it likely begins shortly after the moult to maturity and definitely within the first week following maturity (CES unpublished data). Females become unattractive or difficult to detect once males arrive on the web and engage in web reduction behaviour (destroying large sections of the web, bundling them up, and wrapping them with their own silk; Watson 1986; Scott et al. 2015) and once mated, females rebuild their webs without attractive pheromones (MacLeod & Andrade 2014). At our field site, pheromone traps containing unmated females typically attract males within a few hours (Scott et al. 2015), thus it is likely that most females are visited by males and mated shortly after they mature and begin pheromone production (if not before; in other Latrodectus species, males may guard subadult females and inseminate them just prior to their moult to maturity (Biaggio et al. 2016; Baruffaldi & Andrade 2017). We did not usually know the exact date on which females moulted to maturity, so for each female we assumed that they were recently matured and available for mating for six days prior to the date on which we noted their moult, and for seven

177 days after, for a total of 14 days. This is likely a conservative overestimate of the length of time that recently matured females remain unmated and actively signalling at this site.

We found and marked a total of 66 males during their penultimate instar and followed 16 through to maturity (penultimate males that disappeared either died or matured and left their webs to look for females). Penultimate and adult males are much smaller and more cryptic than their female counterparts, and can use dense vegetation as microhabitats in which they are practically impossible to detect (unlike females, which almost exclusively build webs under driftwood logs at our field site; Salomon et al. 2010). Males are easier to detect when they have webs under logs or rocks, and when they are on the webs of females. We marked a total of 274 males that were found as adults cohabiting with females or juveniles. However, when we flipped over logs during monthly censuses, we often found adult males inside of females’ retreats, where they would not be visible during nightly surveys. Once we found and marked a male for the first time, we assumed that he remained alive until we confirmed his presence or absence at the next monthly census. We assumed that absent males that were not re-sighted elsewhere had died during movement between webs. These assumptions are reasonable given that the difficulty detecting males means that we likely underestimated their abundance overall, and because males from this population typically live more than one month after maturing (CES unpublished data).

(d) Tracking male movements in the field.

We mapped all microhabitats (logs and rocks) within the field site and calculated minimum distances between them. Thus, when we found previously marked males on females’ webs, we could infer the minimum distance they had traveled. We could also estimate the mortality risk associated with mate-searching in the field by (1) tracking the number of males marked on their juvenile web that were later re-sighted on females’ webs (indicating that they survived a first mate-searching event), and (2) tracking the number of males who were sighted on two or more different females’ webs (indicating that they survived two or more moves between webs). These estimates assume that males that were not re-sighted on a female’s web died during mate searching. Sexual cannibalism is rare in this population and readily apparent when it occurs because the exoskeleton of the consumed male remains identifiable in or under the female’s web for at least several days. When we found males dead in the webs of females we included them in

178 our data set as having successfully found a female. Thus, sexual cannibalism was not likely to cause us to underestimate mortality during mate searching.

(e) Experimental animal capture and maintenance

All Latrodectus hesperus spiders used in experiments were field-collected during the summer they were used in experiments (2016 or 2017) or were the lab-reared, outbred offspring of females collected from the field in 2015 or 2016. Lab-reared spiders were raised using standard methods (Baruffaldi & Andrade 2015) and the diets gauged to their body size. Juveniles were fed vinegar flies (Drosophila melanogaster), antepenultimate males and similarly sized juvenile females were fed cricket nymphs (Acheta domesticus or Gryllodes sigillatus), and subadult females were fed adult crickets. Adult males were given water ad libitum, but were not fed (adult males do not normally hunt in nature; Foelix 2011).

Black widow males used in field experiments (both years) were collected using pheromone traps made from screen cages with an unmated female and her web inside (see Scott et al. 2015 for cage design and MacLeod & Andrade 2014 for live collection technique) or using males collected from webs of naturally occurring females in an area adjacent to our main study site (2017 only). For field experiments in 2016, females were reared in the laboratory at the University of Toronto Scarborough and then transported to the field site. In 2017, we collected females from the field in their last two juvenile instars, and reared them to maturity indoors, under conditions similar to those in the laboratory, to ensure that they remained unmated.

Steatoda grossa males were collected from our field site in 2017 as penultimate instars or adults and identified using Levi (1957). They were then kept indoors under similar conditions to lab-reared L. hesperus males. At this site, S. grossa is found in the same microhabitats as L. hesperus (i.e., under rocks and driftwood logs), but is much less common.

(f) Measuring male body size and condition

Before field or lab experiments, we measured each male by placing him in a clear plastic bag next to a size reference (see Fig. S1) and measured the tibia-patella length of both first legs (a standard size index for spiders) to the nearest 0.1 mm using ImageJ (Rasband 2015). When possible, we also weighed males to the nearest 0.1 mg using a microbalance (males used in field

179 experiments in 2017 were weighed after marking). We calculated size-corrected mass (a body condition index) as the residuals of a linear regression of log(mass) against tibia-patella length (Jakob et al 1996).

(g) X-maze choice experiments

The choice test setup consisted of an X-shaped maze (see Fig. 2) made from cotton yarn (ca. 2.5 mm diameter) strung tightly between vertical bamboo posts topped with alligator clips (2.8 cm long). An unmated female in a mesh-sided cage (l × w × h = 9 × 9 × 11.5 cm) or an empty cage was placed 10 cm from one end of the maze, equidistant from the two end posts. For trials with wind two small fans (Travelon 3-speed folding fan, set on medium speed) behind the cage blew air (and pheromone, when a female was in the cage) toward the starting posts. The wind speed experienced by males was 0.3 m/s at the starting posts and 1.0 m/s at the end posts. The X-maze apparatus was surrounded by a shallow moat (65 × 210 cm total area) to restrict male movements and vertical walls made out of foam board (l × h = 220 × 51 cm) along the long sides of setup prevented any cross-wind.

(h) X-maze experimental protocol

For all X-maze experiments (Fig. 2a), a stimulus male (either Latrodectus hesperus or Steatoda grossa) provided a silk trail as follows. We first introduced the stimulus male on one of the two starting posts, and once he climbed to the top he would proceed upwind (fans were always on while stimulus males traversed the maze) along the string until he reached one of the ending posts, after which we recaptured him. Next, we introduced a test male onto the opposite starting post and he would traverse the maze, facing a choice between silk and no silk once he reached the intersection of the X. We alternated the side of the entry post for stimulus males between replicates. For the first experiment only, we recorded the time both stimulus and test males took to travel the distance between the intersection of the X and an end post. In four cases test males fell into the moat (because they were traversing a sagging silk line) before reaching an end post so we were not able to calculate their time, but because these were more than halfway to a female when they fell, we included them the set of males that made a choice. In the final experiment we omitted one replicate in which the test male did not touch the conspecific silk at the intersection because he was already traversing a sagging heterospecific silk line (males in this

180 experiment were always introduced on the post where the heterospecific stimulus male had started).

(i) Sample sizes and animals used in X-maze choice tests

The sample size for the first three experiments was 22 test males. We added replicates to the fourth experiment (heterospecific silk vs. no silk) until 16 males had chosen the non-silk path (total n = 32) so that we would could run 16 replicates of the final experiment (con- vs. heterospecific silk). In all experiments, males were never used as test males more than once. In the first experiment only, each male was used as both a stimulus and a test male, in random order. The same 22 males were each used twice as stimulus males in the second and third experiments, which were run on consecutive days.

181

2. Supplementary Tables

Table S1. Sample sizes and recapture rates for two mate-searching experiments in the field

Year 10 m 20 m 30 m 40 m 50 m 60 m total

2016 released: 19 19 20 18 22 19 117

recaptured: 12 11 16 11 15 8 73

2017 released: 21 22 22 22 22 21 130

recaptured: 16 9 7 2 0 0 34

182

Table S2. Results of general linear models assessing the effects of distance from females on recapture rate (binomial distribution and logit link) and speed (normal distribution and identity link) in a field experiment where males searched for females over 10-60 m. See Table S1 for sample sizes.

Estimate SE Z P R2

2016: recapture rate ~ distance + leg length

Intercept -5.015 2.318 -2.163 0.031

Distance -0.008 0.012 -0.714 0.476

Leg length 1.069 0.414 2.582 0.010

2016: log(speed) ~ distance + leg length 0.336

Intercept 0.257 0.801 0.321 0.749

Distance 0.021 0.004 5.164 <0.001

Leg length -0.344 0.138 -2.490 0.015

2017: recapture rate ~ distance + leg length + body condition index

Intercept 3.460 3.265 1.060 0.289

Distance -0.119 0.023 -5.200 <0.001

Leg length -0.189 0.571 -0.337 0.741

Condition index -0.655 1.451 -0.452 0.651

2017: log(speed) ~ distance + leg length + body condition index 0.388

Intercept -0.146 0.771 -0.190 0.850

Distance 0.011 0.006 1.996 0.055

Leg length -0.431 0.138 -3.123 0.004

Condition index 0.582 0.306 1.904 0.067

183

Table S3. Results of general linear models* assessing the effects of social information availability on speed in a laboratory mate-searching experiment where first males had access to only female chemical cues and second males had access to both female chemical cues and silk cues from a rival (n = 22 pairs of males).

Estimate SE t P R2

Model 1: speed ~ silk + leg length 0.258

Intercept -0.291 0.663 -0.440 0.663

Silk 0.240 0.126 1.907 0.065

Leg length 0.369 0.123 2.997 0.005

Model 2: speed ~ silk + body condition index 0.279

Intercept 1.272 0.152 8.364 <0.001

Silk 0.289 0.126 2.298 0.028

Body condition index -0.906 0.283 -3.197 0.003

*We ran two separate models because there was a significant negative correlation between leg length and size-corrected mass for males used in this experiment (R2 = 0.36; F1,36 = 20.21; P <0.0001).

184

Table S4. Results of general linear models assessing the effects of male size and distance from females on recapture rate (binomial distribution and logit link) and speed (normal distribution and identity link) in a field experiment where males searched for females over 25 m (n = 25 males), 50 m (n = 24), 75 m (n = 26), or 100 m (n = 19). Details of experimental design are shown in Fig. S4.

Estimate SE Z p R2

2016 (100m): recapture rate ~ distance + leg length

Intercept 3.368 3.471 0.970 0.332

Distance -0.078 0.016 -4.786 <0.0001

Leg length -0.052 0.600 -0.087 0.930

2016 (100m): speed ~ distance + leg length 0.198

Intercept 0.650 0.236 2.758 0.011

Distance 0.004 0.002 1.804 0.083

Leg length -0.078 0.041 -1.898 0.069

185

2. Supplementary Figures

Figure S1. Sample photograph of a marked male spider with a size reference used to measure the tibia- patella (t-p) lengths of the first legs. Note the paint marks on the femora of the legs and on the abdomen; unique combinations of colour and leg numbers were used to identify individual males in the field.

186

187

Figure S3. Patterns of attraction to pheromone traps for experimentally released males in the field. We ran this additional mate searching experiment in 2016, with methods modified from those described in the main text. On 9 September we placed four pheromone traps in two pairs (with pairs separated by 2 m) at opposite ends of a 10-m transect approximately perpendicular to the forecasted wind direction. We released males in groups at 100 m, 75 m, 50 m, and then 25 m from the point midway between the two pairs of cages. (a) Number of Latrodectus hesperus males captured at each pheromone-emitting female’s cage during a second field experiment in 2016. Dotted line represents the location of the male release transect and the locations of pheromone traps (mesh cages containing females) are shown on the (vertical) x-axis. Males used in this experiment had already been used in the 2016 mate-searching experiment discussed in the main text. (b) Wind velocity (speed and direction as recorded hourly at a nearby weather station) during the experiment. Numbered arrows represent the wind velocity over each hour that spiders continued to arrive at traps.

188

Figure S4. Relationships between distance from pheromone-emitting females and (a) recapture rate, and (b) average speed of recaptured males of Latrodectus hesperus in a mate-searching experiment in the field. Data are shown as points with predicted fits (solid lines) and approximate 95% confidence intervals (dashed grey lines) from general linear models overlaid.

189

References

Biaggio MD, Sandomirsky I, Lubin Y, Harari AR, Andrade MCB. 2016 Copulation with immature females increases male fitness in cannibalistic widow spiders. Biol. Lett. 12, 20160516. Baruffaldi L, Andrade MCB. 2015 Contact pheromones mediate male preference in black widow spiders: avoidance of hungry sexual cannibals? Anim. Behav. 102, 25 – 32. Baruffaldi L, Andrade MCB. 2017 Neutral fitness outcomes contradict inferences of sexual ‘coercion’ derived from male’s damaging mating tactic in a widow spider. Sci. Rep. 7, 17322. Foelix R. 2011 Biology of Spiders. New York, NY: Oxford University Press Jakob EM, Marshall SD, Uetz GW. 1996 Estimating fitness: a comparison of body condition indices. Oikos 77, 61 – 67. Kaston BJ. 1970 Comparative biology of American black widow spiders. San Diego Soc. Natur. Hist. Trans. 16, 33 – 82. Kasumovic MM, Andrade MCB. 2004 Discrimination of airborne pheromones by mate- searching male western black widow spiders (Latrodectus hesperus): species-and population-specific responses. Can. J. Zool. 82, 1027–1034. Levi HW 1957 The spider genera Cristulina and Steatoda in North America, Central America, and the West Indies (Araneae, Theridiidae). Bull. Mus. Comp. Zool. 117, 367 – 424 MacLeod EC, Andrade MCB. 2014 Strong, convergent male mate choice along two preference axes in field populations of black widow spiders. Anim. Behav. 89, 163–169. Rasband WS. 2015 ImageJ 1.51 (National Institutes of Health, Bethesda). Available at https://imagej.net Salomon M. 2008 Facultative group living in the western black widow spider, Latrodectus hesperus: an evolutionary approach. Doctoral dissertation, Simon Fraser University. Salomon M, Vibert S, Bennett RG. 2010 Habitat use by western black widow spiders (Latrodectus hesperus) in coastal British Columbia: evidence of facultative group living. Can. J. Zool. 88, 334–346. Scott C, Kirk D, McCann S, Gries G. 2015 Web reduction by courting male black widows renders pheromone-emitting females' webs less attractive to rival males. Anim. Behav. 107, 71–78. Watson PJ. 1986 Transmission of a female sex pheromone thwarted by males in the spider Linyphia litigiosa (Linyphiidae). Science 233, 219–221.

Appendix 2

Supplementary Figures (Chapter 1)

Figure A2-1. Relationships between male morphology and speed during mate-searching experiments in the field. Graphs show relationships between male morphology and (a,b) recapture rate, and (c,d,e) average speed of recaptured males. Speed was log-transformed for analysis with back-transformed predicted fits displayed on the raw data. Lines are predicted fits (solid) and approximate 95% confidence intervals (grey dashed) from general linear models of the data. (a,c) Results from the 2016 replicate when the wind was strong and relatively consistent in direction. (b,d,e) Results from the 2017 replicate when the wind was relatively weak and highly variable in direction (figure 1d).

190 191

Figure A2-2. Distributions of male leg length for field-collected and lab-reared males. (a, b) Box plots and histograms of tibia-patella lengths for Latrodectus hesperus males collected from the field and used in mate searching experiments in 2016 and 2017, respectively. (c) Box plot and histogram of tibia-patella lengths for lab-reared males, including those used in laboratory choice tests. Vertical red lines indicate means, which did not differ for field-caught males in 2016 and 2017 (t = 0.97; df = 233.9; p = 0.33). Mean tibia-patella length was significantly shorter for lab-reared males (c) than for field collected males in either year (2016: t = 5.96; df = 204.2; p < 0.0001; 2017: t = 5.47; df = 190.1; p <0.0001).