University of New Hampshire University of New Hampshire Scholars' Repository

Master's Theses and Capstones Student Scholarship

Spring 2016

Cuticular hydrocarbons of the small carpenter bee Ceratina calcarata Robertson (Hymenoptera: Apidae: Xylocopinae)

Nicholas James Pizzi University of New Hampshire, Durham

Follow this and additional works at: https://scholars.unh.edu/thesis

Recommended Citation Pizzi, Nicholas James, "Cuticular hydrocarbons of the small carpenter bee Ceratina calcarata Robertson (Hymenoptera: Apidae: Xylocopinae)" (2016). Master's Theses and Capstones. 1080. https://scholars.unh.edu/thesis/1080

This Thesis is brought to you for free and open access by the Student Scholarship at University of New Hampshire Scholars' Repository. It has been accepted for inclusion in Master's Theses and Capstones by an authorized administrator of University of New Hampshire Scholars' Repository. For more information, please contact [email protected].

CUTICULAR HYDROCARBONS OF THE SMALL CARPENTER BEE Ceratina

calcarata ROBERTSON (HYMENOPTERA: APIDAE: XYLOCOPINAE)

BY

NICHOLAS JAMES PIZZI BS Zoology, University of New Hampshire, 2014

THESIS

Submitted to the University of New Hampshire

in Partial Fulfillment of

the Requirements for the Degree of

Master of Science

in

Zoology

May, 2016

ii

This thesis/dissertation has been examined and approved in partial fulfillment of the requirements for the degree of Masters of Science in Zoology by:

Sandra M. Rehan, Assistant Professor of Biology

Larry Harris, Professor of Zoology

Winsor Watson, Professor of Zoology

On April 19, 2016

Original approval signatures are on file with the University of New Hampshire Graduate School.

iii

ACKNOWLEDGEMENTS

I would like to thank my adviser Sandra Rehan for her guidance throughout my graduate school career. It was a pleasure working under your mentorship and I am truly grateful for this experience. Thanks for always challenging me and teaching me more than I could have ever imagined.

I would also like to thank all members of the UNH Bee Lab, past and present. Thank you

Wyatt Shell, Jake Withee, Sarah Lawson, Erika Tucker, Michael Steffen, and Selena Helmreich.

Special thanks to Sean Lombard, for without his assistance the completion of my second data chapter would not have been possible.

Thank you Richard Kondrat and Ron New for assistance with GC/MS analyses, and the

University of New Hampshire, New Hampshire Agricultural Experiment Station and the Tuttle

Foundation provided support for this research.

Thank you to my graduate committee members, Win Watson and Larry Harris. Your suggestions and feedback were most useful and greatly appreciated.

Finally, I would like to thank my family, especially my parents Paul and Lisa Pizzi, for always being there for me night and day. Thank you for your unconditional emotional and financial support.

iv

ABSTRACT

Cuticular hydrocarbons of the small carpenter bee Ceratina calcarata Robertson

(Hymenoptera: Apidae: Xylocopinae)

by

Nicholas J. Pizzi

University of New Hampshire, May, 2016

The formation and maintenance of eusocial insect groups, in which there are overlapping generations, cooperative brood care, and reproductive division of labor is a major evolutionary transition. To understand the origins of eusociality, simple societies must be studied.

Subsociality is the simplest form of social behavior and is defined as prolonged maternal care for offspring. Studies with subsocial species can provide powerful insights into the transition from basic to advanced social behaviors. In this thesis I use the subsocial small carpenter bee Ceratina calcarata Roberton (Hymenoptera: Xylocopinae) as a model organism. Specifically I study cuticular hydrocarbons (CHCs) of this species to examine two important evolutionary questions.

CHCs are long chain hydrocarbons present on the cuticle of insects and are used for signaling and communication purposes. In eusocial insects queen pheromones are CHCs that signal fertility and suppress worker reproduction. It is hypothesized that CHCs were first fertility signals in less social forms and subsequently coopted as queen pheromones in eusocial lineages.

I test this hypothesis and show supportive evidence for it, as my results suggest that C. calcarata, a subsocial species, may use CHCs to signal fertility. Second, the use of CHCs to recognize non- nestmates is essential to the fitness of eusocial colonies. My results suggest that C. calcarata v may possibly use CHCs as recognition cues, indicating that the use of CHCs for recognition of conspecifics is not a derived trait unique to eusocial lineages, but was probably conserved and present in less social forms. Therefore, this thesis contributes to our understanding of the factors favoring the formation and evolution of eusociality. I show that fertility signals, and the use of

CHCs for non-nestmate recognition are likely to be conserved traits.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... iii ABSTRACT ...... iv Chapter 1: General Introduction ...... 1 Part 1: Social Evolution in Bees ...... 1 Part 2: Behaviors of Bees ...... 6 Part 3: Cuticular Hydrocarbons and Chemical Communication ...... 8 Part 4: Thesis Research Aims ...... 11 Chapter 2: Characterization of cuticular hydrocarbons in a subsocial bee, Ceratina calcarata Robertson (HYMENOPTERA: APIDAE) ...... 13 INTRODUCTION ...... 13 METHODS ...... 15 Nest Collections and Dissections ...... 15 Morphological Measurements ...... 17 Cuticular Hydrocarbon Characterization ...... 17 Comparative Cuticular Hydrocarbon Analyses ...... 18 Statistical Analyses ...... 18 RESULTS ...... 19 Cuticular Hydrocarbon Profiles ...... 19 Morphological Measurements ...... 20 Comparative Cuticular Hydrocarbon Profiles ...... 20 DISCUSSION ...... 21 Reproductive Signaling and Ovarian Development ...... 22 CHCs vary as a function of age ...... 23 Comparative Analysis of Insect Cuticular Hydrocarbons ...... 25 Future Directions ...... 25 TABLES AND FIGURES ...... 27 Chapter 3: Cuticular hydrocarbons serve as recognition signals in a subsocial bee, Ceratina calcarata Robertson (HYMENOPTERA: XYLOCOPINAE) ...... 33 INTRODUCTION ...... 33 METHODS ...... 37 Bee Collections ...... 37 Nest Dissections and Behavioral Assay Preparations ...... 38

Preparation for Behavioral Assays ...... 38 Circle Tube Behavioral Assays ...... 39 Morphological and Physiological Measurements ...... 40 Statistical Analyses ...... 40 RESULTS ...... 42 Circle Tube Assays ...... 42 Morphological and Physiological Measurements ...... 43 DISCUSSION ...... 43 TABLES ...... 48 Chapter 4: General Conclusions and Future Directions ...... 51 REFERENCES ...... 53 APPENDIX A: Chapter 2 Supplementary Materials ...... 61 SUPPLEMENTARY TABLES ...... 61 SUPPLEMENTARY REFERENCES ...... 64 APPENDIX B: Chapter 3 Supplementary Materials ...... 70 SUPPLEMENTARY TABLES ...... 70 1

Chapter 1: General Introduction

This thesis is about chemical communication in the small carpenter bee Ceratina calcarata, with the ultimate goal of expanding our knowledge on the formation and maintenance of complex social groups in insects. This introduction is divided into several segments by topic with the purpose of providing the reader with sufficient background knowledge to make this thesis more palatable. The first part of this introduction provides essential background information on social evolution, defining derived terms which are likely unfamiliar to many outside this field of study. I also introduce my study organism, Ceratina calcarata, reviewing the phylogeny and nesting biology of this species. Second, I review the behaviors of bees across the social spectrum in the context of evolutionary biology. Third, I provide an introduction on insect cuticular hydrocarbons and chemical communication, and how this is relevant to behavioral studies and thus understanding the evolution of highly advanced social groups. Finally, after providing sufficient background information, the fourth subsection of this chapter explicitly states the research aims of my thesis.

Part 1: Social Evolution in Bees

There have been several major transitions in evolution, each sharing three major common features: first, individuals that were capable of reproducing as individuals before the transition can only reproduce as part of a larger unit after the transition, second, task specialization/division of labor, and third, there is a change in the way information is passed on to future generations.

Such major transitions include the evolution of prokaryotes into eukaryotes, protists to animals, plants, and fungi, and solitary individuals to social colonies with non-reproductive castes 2

(Szathmary and Maynard Smith, 1995). This thesis is only concerned with the latter transition, the evolution of eusociality.

Eusociality is a highly successful reproductive strategy, and is defined by overlapping generations, cooperative brood care, and a reproductive division of labor in which one individual reproduces while others forgo their own reproduction (Michener 1974; Wilson 1971). Only a select few lineages have achieved eusociality thus far: aphids (Stern and Foster 1997), ambrosia beetles (Kent and Simpson 1992), flatworms (Hechinger et al. 2010), snapping shrimp (Duffy

1996), gall forming thrips (Crespi 1992), termites (Thorne 1997), and naked mole rats (Jarvis

1981). However, eusociality is mostly represented by insects, namely order Hymenoptera. Even though most insect species are solitary, eusociality is such a successful reproductive strategy that the vast majority of the world’s insect biomass is that of eusocial species (Wilson 1971).

Eusociality can further be divided into two categories, primitively eusocial and advanced eusocial. In the primitively eusocial colonies, reproductive physiology and morphological characteristics such as body size differ much less so between castes than in advanced eusocial colonies. In the advanced eusocial colonies, castes have distinct division of labor, partitioning foraging behavior and reproductive effort among colony members. Advanced eusocial colonies are characterized by discrete body size variation, and morphological castes. For example, future reproductives, also termed gynes, in highly eusocial colonies do not have the structures used for manipulating and collecting pollen, and also do not display foraging behavior. When observing an advanced eusocial colony morphological differences in castes are unambiguous (Michener

1974). It is thought that these advanced eusocial lineages have reached an evolutionary “point of no return” (sensu Wilson and Hӧlldobler 2005), in which losses of eusociality and reversion back 3 less social or solitary life are not possible suggesting that only exceptionally rare circumstances favored such behavior to evolve (Wilson 2008).

To understand the origin of eusociality, comparative analyses spanning the social spectrum are necessary. While eusociality is the extreme of one side of the social spectrum, the other extreme is solitary nesting. In solitary species, individuals do not interact with conspecifics except for mating (Michener 1974). Conspecifics of solitary species may come within close proximity each other due to foraging effort, however, no degree of interaction occurs between them. Moreover, dense aggregations of nests commonly occur in ground nesting bees, and thus solitary individuals may frequently encounter other individuals which nest solitarily (Kocher and

Paxton 2014).

The most basic form of social behavior is subsociality, which is defined as prolonged maternal care for offspring. In subsocial species mothers are long-lived and nest loyal for their entire life. Mothers are the principle nest guards and interact with her offspring by progressively feeding them through adulthood. Additionally, siblings interact with each other within the pre- reproductive assemblage (Michener 1974).

Solitary and subsociality are likely necessary pre-requisites for eusociality to occur

(Michener 1985, 1990). In eusocial colonies, queens first establish nests solitarily. After laying eggs and provisioning brood, she then transitions into the subsocial phase of the colony cycle in which she progressively feeds and protects her larvae. When larvae develop into adult offspring her worker daughters take over the task of foraging, and the queen monopolizes reproduction creating a division of labor and thus transitioning into the eusocial phase of the life-cycle

(Michener 1974). Therefore eusociality can arise directly from solitary life and there need not be a series of intervening steps precluding eusociality (Michener 1985). This also suggests that 4 subsocial mother-offspring interactions are most likely necessary precursors for eusociality to occur (Michener 1990).

Bees are excellent model organisms for studying social evolution as they exhibit behaviors ranging the social spectrum (i.e., solitary to eusocial; Michener 1974). Therefore, comparative analyses among closely related bee species can provide powerful insights into the evolution of eusociality (Kocher and Paxton 2014). Further, bees are particularly interesting as eusociality has independently evolved four times, more than any other lineage. Eusociality has independently evolved in the bee family Halictidae twice, once in tribe Halictini and the second in Augochlorini (Brady et al. 2006; Gibbs et al. 2013). The other two origins were in family

Apidae, with origins occurring in subfamily Apinae and Xylocopinae (Rehan et al. 2012).

Although eusociality has evolved, subsequent losses have occurred (Danforth 2002) resulting in considerable variation in social behaviors among bees, great for comparative analyses (Kocher and Paxton 2014).

The subfamily Xylocopinae is divided into four tribes: Xylocopini, Manueliini,

Ceratinini, and Allodapini, the most basal tribe being Xylocopini (Rehan et al. 2012).

Subsequent losses of eusociality and reversions back to solitary or weakly social life have caused considerable variation in behaviors between closely related bee species in Xylocopinae.

Furthermore, variation in behaviors is even observed within species, with some individuals nesting solitarily, and others forming social colonies (Michener 1990). Tribes Xylocopini and

Ceratinini contain species ranging from solitary to primitively eusocial, while the Manueliini only contains three species of completely solitary individuals. Since subfamily Xylocopinae contains socially polymorphic, closely related species retaining the plasticity to switch to solitary 5 or social life, this subfamily is therefore an excellent candidate for comparative studies on the origins of eusociality.

Tribe Ceratinini is the most socially polymorphic tribe of Xylocopinae. In tribe Ceratinini there is only one genus, Ceratina, with 23 subgenera and hundreds of species (Terzo 2000).

Within this single genus species range the social spectrum from solitary behavior to eusociality

(Michener 1985). Most species are described as solitary, but occasionally multiple females form eusocial nests (Sakagami and Maeta 1995). Therefore Ceratina are excellent model organisms for investigating the transition from solitary to eusociality. In this thesis, I use the subsocial small carpenter bee Ceratina calcarata Robertson (Hymenoptera: Apidae: Xylocopinae). Since subsociality is a precursor for eusociality (Michener 1990) C. calcarata is an ideal model species for the study of social evolution.

As with all Xylocopinae, C. calcarata is a wood-nesting bee. In spring, newly dispersed

C. calcarata mothers establish nests solitarily by excavating the pith of the dead broken stem, essentially creating a burrow within the wood. A mother then forages for pollen, which she then takes back to the nest and kneads into a neatly formed pollen ball. She puts the first pollen ball at the furthest point in the back of the nest and deposits an egg onto the pollen mass, on which the egg will hatch and the larvae will feed on. After depositing an egg on the pollen ball, she scrapes the wall of the nest to collect pith from which she creates a brood cell partitioning the pollen ball and egg from the rest of the nest. She continues to lay eggs on masses of pollen she collected and build partitions between the brood cells in sequential order until her reproductive effort for the season is complete (Michener, 1990). The mother is long-lived, remaining in the nest with her developing offspring, being the primary nest guard protecting her offspring from predators and parasites. The mother also frequently breaks down brood cell partitions, cleaning feces and other 6 possibly contaminating debris from the brood cell, and then rebuilding the brood cell partition incorporating the debris and feces (Sakagami and Maeta, 1977).

Part 2: Behaviors of Bees

Breed et al. (1978) were the first to use the circle tube assay to elucidate behaviors in bees. Observing social behaviors of primitively eusocial bees was difficult because most species nest in burrows in the soil. Breed et al. (1978) used clear 10cm long plastic tubes with an inside diameter of 5mm to observe behaviors of the primitively eusocial L. (D) zephyrum. Remarkably, the behaviors among nest mates and non-nest mates observed in the circle tubes were similar to the behaviors bees display in their nests, indicating that using clear plastic tubing is an appropriate method of observing behaviors (Brothers and Michener 1974). Since then circle tube literature has greatly expanded to include bee species ranging the social spectrum from solitary to eusocial (Pabalan et al. 2000; Kukuk 1992; Rehan and Richards 2013; Richards and Packer

2010; McConnell Garner and Kukuk 1997).

Most behavioral studies have focused on obligate eusocial species within the family

Halictidae, reinforcing the hypothesis that adult females should show tolerant behaviors towards nestmates and aggressive behaviors towards non-nestmates. However, nestmate recognition may also have adaptive significance in solitary or weakly social species. For example, nest mate recognition is essential within pre-reproductive assemblages in subsocial small carpenter bees,

Xylocopinae. The mother needs to distinguish her own offspring from other individuals, and offspring within the nest must be capable of recognizing each other (Michener, 1990). Therefore, it is crucial to also understand behavioral interactions in subsocial species, since subsociality is a 7 likely intermediate between solitary and eusocial nesting, as adults have long life spans and nest loyalty, two precursors to eusocial behavior.

Rehan and Richards (2013) examined the role of reproductive status on aggression in the subsocial small carpenter bee, Ceratina calcarata (family Apidae). The levels of aggression in C. calcarata are context dependent; seasonal variation in aggression correlates with seasonal variation in reproductive status. Similar to two sweat bee species, L. figueresi and H. ligatus

(Wcislo 1997; Pabalan et al. 2000), C. calcarata females with large ovaries are more aggressive than adult females with small ovaries (Rehan and Richards 2013). Adult females were more aggressive towards non-nest mates than nest mates, and were most aggressive towards unfamiliar, reproductively active females. However, post-reproductive females, in which the ovaries were resorbed, were tolerant towards all unfamiliar individuals. In the mature brood phase of the nesting cycle, in which all adults live within the same nest, adult nest mates were tolerant of each other. These results are consistent with Michener’s (1990) suggestion that nest mate tolerance observed in eusocial colonies perhaps first evolved in connection with pre- reproduction assemblages.

Circle tube assays have been very informative in the context of social evolution. In summary, we have learned that eusocial species are typically highly aggressive towards non- nestmates while very tolerant towards nestmates (Breed et al. 1978). Conversely, solitary species are typically highly avoidant of non-nestmates (Richards and Packer 2010). Subsocial bees are intermediate in this regard, as they are both moderately aggressive, avoidant, and tolerant as well

(Rehan and Richards 2013). While this body of literature has expanded to include species ranging the social spectrum and aided in our understanding of the origins of eusociality, the proximate mechanisms were not addressed. In other words, why the bees behaved in a certain 8 manner was not examined. Next I describe the body of literature focused on the proximate mechanisms of insect communication, and connect back to the topic of social evolution.

Part 3: Cuticular Hydrocarbons and Chemical Communication

Animals have evolved advanced means of communication. Animals can communicate through visual (Hinz et al. 2013; Parr and de Waal 1999; Vokey et al. 2004), olfactory (Kraus et al. 2012; Kulahci et al. 2014; Gerlach et al. 2008), and acoustic cues (Searby and Jouventin

2003, 2004; Dorado-Correa et al. 2013). Animals may also utilize more than one method of communication (Hinz et al. 2013; Kulahci et al. 2014). However, while different forms of communication exist, chemical communication is ubiquitous. All living organisms, including single celled organisms such as bacteria, emit chemicals due to metabolism and are capable of responding to chemical stimuli (Wyatt 2014).

In insects, hydrocarbons present on the cuticle, called cuticular hydrocarbons (CHCs), are the primary chemical compounds used for communication purposes (Howard 1993). CHCs are long chain hydrocarbons, primarily alkanes, methyl-branched alkanes, and alkenes, which initially evolved for their anti-desiccation properties (Howard and Blomquist 1982). Alkanes contain the highest water proofing capacity, while alkenes provide the least desiccation resistance. Moreover, longer chained compounds are more efficient at preventing water loss than shorter compounds (Chung and Carroll 2015). Therefore, hydrocarbons are essential for maintaining water balance. In addition to maintaining water balance in the organism, some insects excrete hydrocarbons to strengthen the walls and waterproof their nests (Brooks et al.

1984; Kronenberg and Hefetz 1984). 9

While CHCs originally evolved for their anti-desiccation properties (Howard and

Blomquist 1982) they are primarily recognition signals in arthropods and play an essential role in mediating behavioral interactions (Howard and Blomquist 2005; Howard 1993). CHCs have since come to serve a variety of derived functions. CHCs can reveal age (Nunes et al. 2009;

Jackson and Bartelt 1986; Cuvillier-Hot et al. 2001) and signal both social and sexual experience

(Gershman and Rundle 2016; Oppelt and Heinze 2009; Pascoal et al. 2016). Most insect species are sexually dimorphic in CHC profiles and thus signal sex (dos Santos and Nascimento 2015;

Weiss et al. 2015; Thomas and Simmons 2008).

In addition to the aforementioned chemical cues, research on eusocial Hymenoptera

(ants, bees, and wasps) has shown that CHCs signal fertility and are correlated with ovarian development (Bonavita-Cougardan et al. 1991; Heinze et al. 2002; Liebig et al. 2000; Peeters et al. 1999; Ayasse et al. 1995). In the past five years several studies have utilized bioassays to identify several cuticular compounds that are queen pheromones regulating worker reproduction.

In these bioassays, certain compounds overexpressed by queens relative to workers were isolated and experimentally applied to workers, and shown to regulate reproduction by either preventing ovarian development or inducing ovary resorption (Smith et al. 2009; Holman et al. 2010, 2013;

Holman 2014; Van Oystaeyen et al. 2014), thereby showing direct evidence for pheromonal control of workers by queens.

Queen pheromones have been identified throughout the past several years by direct evidence, but their broad role in the evolution of eusociality has not been addressed until recently. Van Oystaeyen et al. (2014) compared CHCs across Hymenoptera and found that the overexpression of several saturated hydrocarbons in queens relative to workers was common across 64 eusocial species. Thus the current hypothesis is that fertility signals were present in 10 solitary ancestors and have been subsequently coopted as queen pheromones in eusocial species

(Oliveira et al. 2015; Van Oystaeyen et al. 2014). However, evidence confirming that CHCs signal fertility in subsocial species is necessary to fully support the hypothesis that queen pheromones evolved from pre-existing fertility signals in solitary ancestors. Moreover, no solitary or subsocial species were used in the comparative analysis and therefore there is currently no objective evidence suggesting that fertility signals precede the evolution of queen pheromones.

While bioassays using experimentally isolated compounds and have been informative, some studies have used simpler methods to understand how CHCs mediate behavioral interactions in insects. Studies in insects have used solvents such as pentane or hexane to wash individuals and remove their entire chemical profile. This simple but effective method has been very useful in studying recognition behavior in both solitary (Flores- Prado et al. 2008) and eusocial insects (Ruther et al. 1998; Bonavita-Cougourdan et al., 1987; Nowbahari et al., 1990).

When a test subject is presented to a dead solvent washed non-nestmate little aggression is observed. However, the tests subjects are highly aggressive towards control non-nestmates which have not been washed in solvent. Furthermore, aggressive behaviors can be induced by applied non-nestmate solvent washes to a nestmate (Ruther et al. 1998; Flores-Prado et al. 2008;

Bonavita-Cougourdan et al. 1987; Nowbahari et al. 1990).

In eusocial species the ability to identify non-nestmates is essential to the fitness of the colony. Guards monitor the nest entrance granting nest mates entrance and reject non-nest mates by initiating aggressive interactions such as biting and stinging (Breed 1998; Wilson 1976;

Buckle and Greenberg 1981). However, since the ability to recognize non-nestmates was reported in a solitary bee Manuelia postica, this suggests that eusocial traits may be present in 11 solitary species. Rehan and Richards (2013) have reported that C. calcarata is capable of identifying individuals as non-nestmates. To date no evidence exists that CHCs are used for non- nestmate discrimination in this species, or any subsocial bee species. The present hypothesis is that the use of CHCs as recognition signals precedes the origins of eusociality. However, although there is evidence supporting this in a solitary bee (Flores-Prado et al. 2008), this needs to be confirmed in a subsocial bee species since subsociality is a necessary pre-requisite for eusociality.

Part 4: Thesis Research Aims

Until now there have been no studies examining the CHCs of a subsocial bee species. In this thesis I present the first study to examine the CHCs of a subsocial, small carpenter bee,

Ceratina calcarata and how CHCs influence the outcome of behavioral interactions between non-nestmates. This thesis is divided into two main studies. The first is concerned with characterizing the CHCs of C. calcarata, while the second examines the role CHCs play in mediating behavioral interactions in this species. While each chapter has its own specific research aims the overall purpose of this thesis is to study the CHCs and behavior of a subsocial bee, with an ultimate objective of better understanding the major evolutionary transition from solitary to eusociality.

In Chapter 2, I characterize the CHCs of C. calcarata using Gas Chromatography coupled with Mass Spectrometry (GC/MS), a widely used method in CHC literature. I characterize the CHCs with the intention of testing the hypothesis that queen pheromones in eusocial species evolved from pre-existing fertility signals in solitary species. I analyze CHCs of pre-reproductive, reproductive, and post-reproductive females to determine how CHCs vary with reproductive status. 12

Chapter 3 has two main objectives. First, I aim to develop the first CHC removal assay using living insect specimens, as all previous experiments have used dead solvent washed individuals. Second, I test the hypothesis that the use of CHCs as recognition signals predates the origins of eusociality and is present in not only solitary but also subsocial bees. Therefore, I aim to provide the first empirical evidence that CHCs are used to recognize non-nest mates in a subsocial bee species, C. calcarata. 13

Chapter 2: Characterization of cuticular hydrocarbons in a subsocial bee, Ceratina calcarata Robertson (HYMENOPTERA: APIDAE)

INTRODUCTION

Cuticular hydrocarbons (CHCs) are long chain hydrocarbons, primarily alkanes and alkenes, and are present on the cuticle of insects. CHCs play an essential role in water balance in insects (Howard and Blomquist 1982). While CHCs originally evolved for their anti-desiccation properties (Howard and Blomquist 1982) they are primarily recognition signals in arthropods and play an essential role in mediating behavioral interactions (Howard and Blomquist 2005; Howard

1993). CHCs have evolved to serve a variety of derived functions such as signaling age (Nunes et al. 2009; Jackson and Bartelt 1986; Cuvillier-Hot et al. 2001), sex (dos Santos and Nascimento

2015; Weiss et al. 2015; Thomas and Simmons 2008), and social and sexual experience

(Gershman and Rundle 2016; Oppelt and Heinze 2009; Pascoal et al. 2016).

Since cuticular hydrocarbons mediate behavioral interactions, understanding their functions is therefore essential to our understanding of the evolution of eusociality (Van

Oystaeyen et al. 2014). Eusociality is a highly derived, extremely successful reproductive strategy defined by overlapping generations, cooperative brood care, and reproductive division of labor (Wilson 1971; Michener 1974). In eusocial insects, one individual (queen) monopolizes reproduction while workers carry out tasks to maintain a functional colony (Wilson 1971;

Michener 1974).

Nest mate recognition is key to the success of eusocial species (Hӧlldobler and Wilson

1990; Liang and Silverman 2000; Pradella et al. 2015; Esponda and Gordon 2015; Lorenzi et al.

2005). CHCs are used to distinguish queens from workers, and young versus old individuals 14

(Nunes et al. 2009). Additionally, there are different types of workers, each with a specialized task, and CHCs are used to communicate between worker sub-castes (Grüter and Keller 2016).

However, CHCs are used for recognition in solitary (Flores-Prado et al. 2008) species suggesting that social traits may have adaptive significance in solitary species (Flores-Prado et al. 2008;

Smith and Breed 2012).

Research on eusocial Hymenoptera (ants, bees, and wasps) has shown that CHCs signal fertility (Bonavita-Cougardan et al. 1991; Heinze et al. 2002; Liebig et al. 2000; Peeters et al.

1999; Ayasse et al. 1995) and direct evidence has identified several cuticular compounds that are queen pheromones regulating worker reproduction (Smith et al. 2009; Holman et al. 2010, 2013;

Holman 2014; Van Oystaeyen et al. 2014). While queen pheromones have been identified, their broad role in the evolution of eusociality has not been addressed until recently. Van Oystaeyen et al. (2014) compared CHCs across Hymenoptera and found that the overexpression of saturated hydrocarbons in queens relative to workers was common across 64 eusocial species. Thus, the current hypothesis is that fertility signals were present in solitary ancestors and have been subsequently coopted as queen pheromones in eusocial species (Oliveira et al. 2015; Van

Oystaeyen et al. 2014).

Much of the Hymenoptera CHC literature in the past thirty years has focused on eusocial and solitary species. While these studies have greatly contributed to our understanding of the evolution of eusociality, there is a severe lack of research on non-eusocial hymenopteran species.

To understand how eusociality evolved, simple societies must be well studied (Rehan and Toth

2015). Subsociality, defined as prolonged parental care, is the simplest form of social behavior and is a pre-requisite for eusociality (Michener 1974; Wilson 1971). Therefore elucidating the 15 function of CHCs in subsocial species is necessary to understand the transition from solitary to complex insect societies.

To fully support the hypothesis that queen pheromones evolved from pre-existing fertility signals in solitary ancestors, confirmation that CHCs signal fertility in subsocial species is necessary. However, not only have no studies to date examined CHCs of a subsocial species in the context of social evolution, CHCs of a subsocial species have never been characterized. Here we provide the first characterization of cuticular hydrocarbons in a subsocial bee, the small carpenter bee Ceratina calcarata Robertson (Hymenoptera: Apidae: Xylocopinae). The aims of this study were three-fold: first, to characterize the cuticular hydrocarbon profiles of C. calcarata; second, to examine the relationship between reproductive status and cuticular hydrocarbon profiles in this species; third, to identify if compounds signaling reproduction in C. calcarata are unique to Hymenoptera or common to other insect orders.

METHODS

Nest Collections and Dissections

Ceratina calcarata nests were collected from May through August 2014 from sumac trees,

Rhus typhina, in Durham, New Hampshire (43º08’02”N 70º55’35”W). Nests were collected between 6:00 and 8:00 AM to ensure the mother was still in the nest. To avoid damage to the contents of the nests and any offspring present, the dead broken stem was cut with garden shears at the junction with another branch. Masking tape was then applied to the nest entrance to contain the bees living inside. Collected nests were then brought back to the lab and stored in a

4°C cold room until dissection. 16

Nests were dissected with extreme care to avoid damaging the contents of the nest. The dead broken stems (nests) were split long-ways using a sharp non-serrated knife. Nest classification (described below) was determined and recorded. Nest occupants were captured and stored in chemically neutral tubes. Cuticular washes occurred immediately upon nest dissection to avoid any changes due to prolonged captivity.

Since we were primarily concerned with how CHCs vary as a function of reproductive status we collected pre-reproductive, actively reproductive, and post-reproductive Ceratina calcarata females. Rehan and Richards (2010a) previously defined different nest classifications for C. calcarata. In spring, overwintering females disperse from their natal nest and establish their own nest solitarily and their nests do not contain any feces or pollen and have clean nest walls. The females collected from these nests were classified as founding nest (FN). Therefore,

FN females have developed ovaries, but are not yet laying eggs. In early-mid summer females are fully reproductive and begin egg laying. Females collected from nests containing at least one pollen ball and egg were classified as active brood (AB). In mid-late summer when the brood cell closest to the nest entrance contains a larva or pupa, and the mother’s reproductive effort for the season is complete and she will not be laying any more eggs. Females collected from these nests were classified as full brood (FB). Finally, in late summer all offspring in the nest have finished developing into adults. The mother continues to provide maternal care being the principal nest guard and continues to forage and progressively feed her newly emerged offspring.

Adult brothers and sisters remain in the nest until they disperse the following spring, in which the cycle starts over. We collected only the callow pre-dispersal daughters (PD) from these nests. Using a NIKON H550S dissecting scope we distinguished mothers from PD females by measuring wing wear, described in the next section below. Mothers are easily distinguished from 17

PD females as they have moderately to critically torn wings along the apical margin whereas PD females had unworn wings with no nicks or tears.

Morphological Measurements

All morphological measurements were recorded after cuticular washes using a NIKON

H550S dissecting scope. Head width (i.e. greatest distance of head width, including eyes) was measured as an accurate metric of body size (Rehan and Richards 2010a). Bees were dissected to score ovarian development, recorded as the sum of the length (mm) of the three largest oocytes.

Cuticular Hydrocarbon Characterization

Adult bees were washed in 500µL of pentane for 45 minutes in chemically neutral glass vials (Flores-Prado et al. 2008). Cuticular hydrocarbon profiles were analyzed using Gas

Chromatography coupled with Mass Spectrometry (GC/MS). A Micromass AutoSpec with a

Hewlett-Packard 5890 series II gas chromatograph with helium (75 kPa) as a carrier gas was used to carry out the GC/MS analysis. The injector temperature was 250°C and the transfer line temperature was 300°C (Nunes et al. 2009a). The initial temperature of the oven was 50°C and held for one minute. Then, the temperature was raised 5°C/min to a final temperature of 300°C, which was held for four minutes. The total oven program was 55 minutes. Compounds were identified using three commercial libraries (Wiley 275, NIST 98, and Adams EO library 2205).

The mass spectrometer was operated at 70eV with an ion source temperature of 250ºC and scanned from mass 650 to 45 Da once per second (0.9 + 0.1 = 1 sec/scan). The detector voltage was 2600 V with a trap current of 164 Ua and solvent delay of 1.3 min. A 1µl sample of each 18 pentane wash was injected. The washes of five PD females, seven FN females, five AB females, and eight FB females were used for analysis.

Comparative Cuticular Hydrocarbon Analyses

The cuticular hydrocarbons identified in the present study were compared against a literature review on a variety of solitary, subsocial, primitively eusocial, and advanced eusocial species in

Arthropoda (Supplementary Table 1). Six major insect orders, Orthoptera, Isoptera, Coleoptera,

Hymenoptera, Lepidoptera, and Diptera, and one Araneae (Class: Arachnida) were included as outgroups. Within Hymenoptera, eleven families of ants, bees, and wasps were compared.

Within bees, CHC data from five different families, Halictidae, Colletidae, Andrenidae,

Megachilidae, and Apidae were compared.

Statistical Analyses

All statistical analyses were calculated using R Studio 3.2.2 (R Core Team 2014).

Shapiro-Wilkes tests were used to test for normal distribution of ovarian development, head width, and wing wear. All three physiological measurements were not normally distributed and non-parametric statistics were used. Spearman Rank Correlations of head width and ovarian development versus relative peak heights were employed to determine if cuticular hydrocarbons were correlated with body size and reproductive status, respectively.

Kruskal-Wallis tests were used to determine if there were significant differences in physiological measurements between PD, FN, AB, and FB females (Mant et al. 2005).

Additionally, Kruskal-Wallis tests were used to determine if there were significant differences in relative abundance of CHCs between PD, FN, AB, and FB females. Post-hoc HSD and LSD tests 19

(R package “agricolae”; de Mendeburu 2015), and Dunn’s tests (R package “CRAN”) were run after Kruskal-Wallis tests to determine the source of significant differences between groups.

Bonferroni corrections were calculated to control for multiple comparisons. There were fourteen chemical compounds detected and thus alpha was adjusted for multiple comparisons (Bonferroni corrected P = 0.05/14 = 0.00357). Boxplots were made for compounds that were differentially expressed across the reproductive classes. The box includes the middle two quartiles (middle

50% of the data), whereas the whiskers include the first and fourth quartiles. The line in the box plot shows the median and circles are outliers. Linear Discriminant Analysis (LDA) was used to determine if PD, FN, AB, and FB females differed in their cuticular hydrocarbons profiles.

RESULTS

Cuticular Hydrocarbon Profiles

A total of fourteen different hydrocarbons from four major classes were identified from cuticular extracts of Ceratina calcarata females. Alkanes and alkenes were the most abundant class of hydrocarbons, although three ethyl esters and one alcohol was identified (Table 1). The hydrocarbons ranged from 15 to 27 carbons in length with pentadecane as the shortest length and heptacosane was the longest length identified.

Nonadecane, heneicosane, pentacosane, ethylhexadecenoate, and farnesol concentrations were significantly different between females of different nest class stages (Table 1). PD females had significantly higher relative abundance of farnesol, nonadecane, and heneicosane than FN,

AB, and FB females (Fig. 1). AB females had significantly higher levels of pentacosane than FB and PD females but not significantly higher than FN females, and significantly greater amounts of ethylhexadecenoate than FN and FB females, but not significantly higher than PD females 20

(Fig. 2). Linear Discriminant Analysis of revealed that two linear discriminants explain 99.11% of the variance in CHC profiles, and shows three unique clusters (Fig. 3). PD females were clustered together, AB females were clustered together, and FN and FB females were clustered together.

Morphological Measurements

Active brood females had the largest ovarian development, and were significantly more developed than full brood and founding nest bees, but did not have significantly larger ovaries than PD females. Full brood females had the smallest ovaries and founding nest females had the second smallest ovaries (Kruskal-Wallis test: chi-squared = 16.30, df = 3, p-value = 0.001; Fig.

4A). There was a significant correlation between ovarian development and relative peak heights of nonadecane (Table S2). There were no other significant correlations between ovarian development and relative peak heights of the compounds identified in C. calcarata (Table

S2).There were no significant differences in head width between females of different nest class stages (Kruskal-Wallis test: chi-squared = 2.86, df = 3, p-value = 0.41; Fig. 4B). There was a significant positive correlation between head width and relative amounts of heneicosane (Table

S3). There were no other correlations between head width and relative peak heights of the compounds identified in C. calcarata females (Table S3).

Comparative Cuticular Hydrocarbon Profiles

The fourteen compounds found in C. calcarata were compared to a variety of solitary, subsocial, primitively eusocial, and advanced eusocial species in Arthropoda. Forty-three species in total were used for this comparative analysis (Table S1; Fig. 5). Four compounds, pentadecane, heptadecene, and nonadecene, and farnesol were unique to Hymenoptera. Ethyl 21 esters were also largely unique to Hymenoptera, except for one species, the American cockroach,

Periplaneta americana. All other compounds were found in two or more insect orders, indicating they are present in relatively distally related insect species. None of the compounds of C. calcarata were present in one species, the Australian field cricket, Teleogryllus oceanicus.

Although some compounds were taxon specific, no compounds were unique to solitary, subsocial, primitively eusocial, or advanced eusocial lineages.

DISCUSSION

Here we provide the first characterization of cuticular hydrocarbon profiles in a subsocial bee. There are three main findings from this study. First, we show that pentacosane is most abundant in reproductive females, indicating that it may signal reproductive status in this species

(Figs. 2 and 3). Second, we show that cuticular hydrocarbons may signal age, as young pre- dispersal females had higher concentrations of farnesol, nonadecane, and heneicosane (Fig. 1) than older females. Third, we show that two queen pheromones, heptacosane and pentacosane are present in C. calcarata and also many other solitary, subsocial, primitively eusocial, and advanced eusocial lineages (Fig. 5). We offer a broad evolutionary perspective, comparing empirical data from a subsocial bee with literature review across Hymenoptera and other arthropods to confirm that CHCs signaling reproduction in insects are well-established in solitary lineages and have subsequently been coopted as queen pheromones in eusocial species.

A total of seven alkanes ranging from C15-C27, three alkenes, three ethyl esters, and one alcohol were identified in this study (Table 1). We found heptadecane was the most abundant compound independent of reproductive status (Table 1). Breed (1998) found that heptadecane did not have an effect on recognition in the honey bee Apis mellifera when females were treated 22 with 100µL of this compound and presented to a sister. Consistent with previous reports that heptadecane likely is not used as a recognition signal we found heptadecane levels are stable across C. calcarata females of different reproductive status and therefore it is likely that heptadecane is not a signal compound in this species.

Reproductive Signaling and Ovarian Development

Previous studies have shown that pentacosane is a queen pheromone that controls worker reproduction in a eusocial bumble bee, Bombus terrestris (Van Oystaeyen et al. 2014; Holman

2014). Studies in ants (Pachycondyla inversa, Heinze et al. 2002; Harpegnathos saltator, Liebig et al. 2000; Dinoponera quadriceps, Peeters et al. 1999), bees (Bombus hypnorum, Ayasse et al.

1995), and wasps (Polistes dominulus, Bonavita-Cougardan et al. 1991) have shown that cuticular hydrocarbon profiles are correlated with ovarian development. Although we did not detect any strong correlations between ovarian development and the compounds identified in C. calcarata (Table S2), LDA of compounds show that reproductive females have unique chemical profiles suggesting the use of CHCs to signal reproductive status. Particularly, reproductive females produce the highest levels of the bumblebee queen pheromone pentacosane (Fig. 3;

Holman 2014). While pentacosane was not statistically greater in AB females than FN females

(Fig. 2), these results are still suggestive that C. calcarata females may signal fertility and that pentacosane is an honest signal of reproduction in this species. Moreover, since we found a possible reproductive signal in a subsocial bee, this supports the hypothesis that queen pheromones likely evolved from pre-existing fertility signals (Van Oystaeyen et al. 2014).

Pre-dispersal callow females had developed ovaries (Fig. 4A) despite being unmated non- reproductives. These results are consistent with Sakagami and Maeta (1984), which found that 23 newly emerged Ceratina japonica had fully developed ovaries and are capable of laying eggs.

Similarly, week old female bumble bees have fully developed ovaries (B. terrestris, Duchateau and Vulthuis 1989). Several studies have shown that in eusocial insects queens suppress worker

(daughter) reproduction via pheromonal control (e.g., pentacosane in B. terrestris; Van

Oystaeyen et al. 2014; Holman 2014). In C. calcarata, however, pre-dispersal females are about to enter diapause and have no opportunity for reproduction (Rehan and Richards 2010a).

Therefore, maternal overexpression of pentacosane for regulation of offspring reproduction is probably not necessary or adaptive in this subsocial bee. The effect of C. calcarata mother CHCs on maturing daughter ovarian development should be explored further to directly confirm whether or not mothers in subsocial species control daughter ovarian development.

CHCs vary as a function of age

We observed differences in cuticular hydrocarbon profiles as a function of age (Fig. 1 and 3). Pre-dispersal (PD) females are 8-10 months younger than FN, AB, and FB females

(Rehan and Richards 2010a). PD females had significantly higher concentrations of heneicosane than FN, AB, and FB females (Fig. 1; Table 1). These results are consistent with data from the mosquito, Anopheles gambiae, in which heneicosane concentrations were highest in younger individuals, and decreased with age (Polerstock et al. 2002). Although, heneicosane may also signal body size as we found a significant positive correlation between head width and relative peak heights of heneicosane (Table S3). In the European hornet, Vespa crabro, workers were aggressive towards dead nest mates that were experimentally treated with heneicosane (Ruther et al. 2002). However, PD females are typically not aggressive in C. calcarata (Rehan and Richards 24

2013), suggesting that heneicosane may not elicit the same behavioral response across all

Hymenoptera.

Two additional compounds nonadecane and farnesol were also characteristic of PD females (Fig. 1). PD females had significantly higher concentrations of nonadecane than FN,

AB, and FB females. Bioassays in the honey bee, A. mellifera, found bees experimentally treated with nonadecane received low rates of aggression (Dani et al. 2005). In C. calcarata PD females are tolerant of each other in the pre-reproductive assemblages (Rehan and Richards 2013; Rehan et al. 2014). Since PD females have significantly higher concentrations of nonadecane than females at other time points in the colony cycle and nonadecane induces low aggression rates in

A. mellifera, further investigation is warranted to determine if nonadecane helps maintains nest mate tolerance among C. calcarata PD females.

Farnesol was found almost exclusively in PD females. In the solitary ground nesting bee

Andrena nigroaenea, farnesol inhibits copulatory behavior in males, as males use farnesol to distinguish virgin from mated females (Schiestl and Ayasse 2000). Since farnesol was only found in PD C. calcarata daughters (virgin), perhaps it functions to prevent young male

(brothers) from mating with their sisters in the pre-reproductive assemblages. Ceratina calcarata mate in the spring (Rehan and Richards 2010a) and if farnesol inhibited male copulatory behavior in C. calcarata, then low levels would be expected in newly dispersed females.

Consistent with this idea, we found that spring FN females had significantly lower levels of farnesol than PD females. Furthermore, Rehan and Richards (2010a) reported that males are quiescent and do not interact with females during the summer reproductive phase. This would suggest that females need not produce farnesol during reproduction. Consistent with this notion, we found that AB and FB females had significantly reduced levels of farnesol (Table 1; Fig. 1). 25

Comparative Analysis of Insect Cuticular Hydrocarbons

The third aim of this study was to determine whether cuticular hydrocarbons signaling reproduction in C. calcarata are unique to closely related taxa within Hymenoptera, or if they are common throughout insects of varying degrees of sociality. Two alkanes, pentacosane and heptacosane, are highly conserved across insects and are also present in a subsocial spider (Fig.

5). Both of these hydrocarbons are known to be queen pheromones used to regulate worker reproduction in eusocial Hymenopteran species (Van Oystaeyen et al. 2014). Through our data and an expanded literature review we found that these compounds are indeed present in solitary and subsocial species and are not unique to eusocial lineages (Fig. 5). We also demonstrate that these two queen pheromones are not unique to Hymenoptera, and are common throughout six insect orders, and one arachnid species. Thus, empirical data from a subsocial bee in comparison with a literature review across Hymenoptera and other arthropods strongly suggests queen fertility signals suppressing worker reproduction were present in solitary and subsocial ancestors and have been subsequently coopted as queen pheromones.

Future Directions

Although C. calcarata is capable of nest mate recognition (Rehan and Richards 2013), a chemical basis of this behavior has not yet been verified in this species. Flores-Prado et al.

(2008) demonstrated a chemical basis of nest mate recognition in the closely related xylocopine bee, Manuelia postica, confirming that eusocial traits have adaptive significance in solitary bees as well. Since subsociality is a necessary pre-requisite for eusociality (Michener 1974), investigating whether C. calcarata recognize individuals via chemical profiles is a necessary 26 next step to our understanding of social evolution. Since both solitary (e.g. M. postica, Flores-

Prado et al. 2008) and social apid bees (e.g. A. mellifera, Breed 1998) are known to use pheromones to recognize individuals, we hypothesize that C. calcarata can discriminate individuals based on chemical profiles.

27

TABLES AND FIGURES

Table 1: The relative proportions of fourteen hydrocarbons found on the cuticles of Ceratina calcarata females at different stages of reproduction and life cycle.

Concentration (%) 2 PD (n=5) FN (n=7) AB (n=5) FB (n=8) X Alkanes C15 Pentadecane 11.91 ± 5.8 0.27 ± 0.27 3.00 ± 2.30 8.30 ± 4.00 5.63 C17 Heptadecane 29.00 ± 10.93 34.59 ± 12.84 16.36 ± 11.7 38.14 ±10.80 1.99 C19 Nonadecane 1.28 ± 0.89 0 0 0 8.33* C21 Heneicosane 8.14 ± 5.12 0.19 ± 0.50 1.69 ± 2.33 1.50 ± 3.4 13.10** C23 Tricosane 3.45 ± 2.43 3.55 ± 3.38 2.50 ± 1.41 1.11 ± 1.11 4.68 C25 Pentacosane 6.03 ± 3.13 16.63 ± 11.63 19.66 ± 13.70 1.12 ± 2.96 12.63** C27 Heptacosane 1.51 ± 0.74 14.26 ± 6.36 3.89 ± 2.16 2.53 ± 2.53 5.65

Alkenes C17:1 Heptadecene 4.15 ± 3.74 1.63 ± 1.47 6.32 ± 5.30 28.79 ± 12.10 4.16 C19:1 Nonadecene 0.44 ± 0.44 0.157 ± 0.157 0.59 ± 0.59 0 1.76 C23:1 Tricosene 24.28 ± 14.92 19.43 ± 5.57 21.43 ± 12.30 14.28 ± 7.92 1.52

Ethyl Esters Ethyl Et-C16 Hexadecanoate 1.20 ± 0.08 0 2.46 ± 2.07 0 6.79 Ethyl Et-C16:1 Hexadecenoate 3.10 ± 2.11 0 14.29 ± 9.82 0.17 ± 0.17 7.85* Ethyl 2.35 Et-C18:1 Octadecenoate 2.10 ± 1.76 1.49 ± 1.34 1.47 ± 0.911 1.02 ± 1.02

Alcohols

C15-OH Farnesol 3.41 ± 3.58 0 0.10 ± 0.226 0 10.34* Note: The number of bees washed is provided in parentheses. Chi-Squared values from Kruskal- Wallis non-parametric ANOVA. *P < 0.05, **P < 0.01.

Figure 1. A) Pre-dispersal (PD) Ceratina calcarata daughters contained significantly more farnesol than founding nest (FN), active brood (AB), and full brood (FB) females (Kruskal-Wallis test: chi-squared = 10.34, df = 3, p-value = 0.016). B) PD daughters had significantly more nonadecane than FN, AB, and FB females (Kruskal-Wallis chi-squared = 8.33, df = 3, p = 0.04). C) PD daughters had significantly more heneicosane than FN, AB, and FB females.

28

29

Figure 2. A) AB females have significantly higher relative abundances of ethylhexadecenoate than FN and FB females, but not PD females (Kruskal-Wallis chi-squared = 7.85, df = 3, p = 0.05). B) AB females have significantly higher amounts of pentacosane than PD and FB females, but not FN females (Kruskal-Wallis test: chi-squared = 12.63, df = 3, p-value = 0.006).

30

Figure 3. Linear Discriminate Analysis of cuticular hydrocarbon profiles of Ceratina calcarata females at different stages of reproductive development and life cycle. PD = pre-dispersal callow females, unmated and still in natal nest; FN = founding nest females, newly dispersed in spring; AB = actively reproducing females; FB = full brood females, which are post-reproductive.

31

Figure 4. A) Active brood (AB) Ceratina calcarata females have significantly larger ovaries than founding nest (FN) and full brood females (FB), but not pre-dispersal (PD) females (Kruskal-Wallis test: chi-squared = 16.30, df = 3, p-value = 0.001). B) There are no significant differences in head width between nest classes (Kruskal-Wallis test: chi-squared = 2.86, df = 3, p-value = 0.41).

Figure 5. A comparative analysis of arthropod cuticular hydrocarbons. The columns headings indicate the cuticular hydrocarbons found in C. calcarata from this present study, and the rows are 43 additional species for comparison (Table S1). Squares shaded in black show the presence of a particular hydrocarbon and blank squares indicate the absence of the compound. From left to right,

32

alkanes are columns 1-8, alkenes are columns 9-11, ethyl esters are columns 12-14, and alcohols are column 15. 33

Chapter 3: Cuticular hydrocarbons serve as recognition signals in a subsocial bee, Ceratina calcarata Robertson (HYMENOPTERA: XYLOCOPINAE)

INTRODUCTION

Animals have evolved highly derived strategies for signaling and communication. There are many functions to signals such as advertising mate quality (Martin and Lopez 2006), reproductive state or potential (Marco et al. 1998), advertising the location of food sources (von

Frisch 1965), warning conspecifics of a potential predator that may be nearby (Leavesley and

Magrath 2005). In any given communication network there is both a signaler and a receiver.

Signals evolve when, on average, there are fitness benefits to both the sender and the receiver of the signal (Johnston and Grafen 1993).

A widespread use of signals is for recognition. Animals utilize different methods to recognize others such as acoustic (Searby and Jouventin 2003, 2004; Dorado-Correa et al. 2013), visual (Hinz et al. 2013; Parr and de Waal 1999; Vokey et al. 2004), and olfactory cues (Kraus et al. 2012; Kulahci et al. 2014; Gerlach et al. 2008). These may be used for species recognition and to distinguish familiar from unfamiliar individuals. Moreover, recognition has been posited as a precursor for many kinds of social behavior (Fletcher and Michener 1987; Michener 1990).

The ability to recognize individuals as non-nestmates is essential to the success of eusocial species, in which there are overlapping generations, cooperative brood care, and a reproductive division of labor (Wilson 1971; Michener 1974; Breed 1998). In eusocial termites, bees, wasps, and ants, guards monitor the nest entrance granting nest mates entrance while rejecting non-nestmates by initiating aggressive interactions such as biting and stinging (Breed

1998; Wilson 1976; Buckle and Greenberg 1981). While most castes in eusocial species are 34 based on age (temporal castes) some species have physical castes, in which guards have specialized morphological features designed for colony defense (Wheeler 1991; Grüter et al.

2012; Hölldobler and Wilson 2009). In order to maximize the colony’s fitness, these guards have adapted advanced recognition capabilities with low acceptance failure rates (Hölldobler and

Wilson 2009).

Identifying individuals as non-nestmates is clearly advantageous in highly social insect species, but there are also adaptive advantages to recognition in less social forms. Solitary species do not interact with conspecifics except for mating (Michener 1974: Wilson 1971), but still possess recognition abilities (Flores-Prado et al. 2008; Wcislo 1997; Richards and Packer

2010). Although nesting solitarily, in tightly aggregated nesting sites individuals must be able to recognize and be tolerant towards their neighbors to avoid conflicts, and also distinguish their own nest from their neighbors’ (Kocher and Paxton 2014). Empirical data has shown evidence for non-nestmate recognition in solitary bees, confirming that eusocial traits may have adaptive significance in solitary species (Flores-Prado et al. 2008; Wcislo 1997).

Subsociality is the simplest form of social behavior and is defined as prolonged maternal care for offspring (Michener 1974; Wilson 1971). In subsocial species offspring must recognize and be tolerant towards their siblings, and mothers must be able to recognize her own offspring

(Michener 1990). Additionally, mothers are the principal nest guard and thus must be able to discriminate intruders from her offspring. Evidence supporting this has been shown in a subsocial bee in which mothers are capable of non-nestmate recognition (Rehan and Richards

2013).

Solitary behavior and subsociality are evolutionary antecedents of eusociality (Michener

1985, 1990; Rehan and Toth 2015). First, queens establish nests solitarily, and then transition 35 into the subsocial phase of the colony cycle in which she progressively feeds and protects her larvae. Then, worker daughters take over the task of foraging, and the queen monopolizes reproduction, creating a division of labor and thus transitioning into the eusocial phase of the life cycle (Michener 1974). Experiments studying solitary and subsocial behavior are essential to understanding how such an advanced reproductive strategy such as eusociality could have evolved.

Circle tube behavioral assays, developed by Breed et al. (1978) are a widely used and powerful method of studying recognition throughout the social spectrum, particularly in bees

(Pabalan et al. 2000; Kukuk 1992; Rehan and Richards 2013; Richards and Packer 2010;

McConnell Garner and Kukuk 1997). Eusocial bees are highly aggressive towards non-nestmates

(Breed et al. 1978; Pabalan et al. 2000), while solitary bees are extremely avoidant and display little aggression (Richards and Packer 2010; McConnell-Garner and Kukuk 1997). Subsocial species are also more aggressive towards non-nestmates, but also exhibit tolerance and avoidance of non-nestmates (Rehan and Richards 2013). These studies have shown that bees ranging from solitary to eusocial display recognition behavior. However, while these studies have greatly expanded our knowledge of social evolution, the proximate mechanisms of recognition were not explored in these studies.

In insects, cuticular hydrocarbons (CHCs) are recognition signals that play an important role in mediating behavioral interactions (Howard and Blomquist 2005; Howard 1993).

However, most of the CHC literature has focused on recognition in eusocial insect species

(Liang and Silverman 2000; Pradella et al. 2015; Esponda and Gordon 2015; Lorenzi et al.

2005). CHCs are used to distinguish queens from workers, and young versus old individuals 36

(Nunes et al. 2009). Additionally, there are different types of workers, each with a specialized task, and CHCs are used to communicate between worker sub-castes (Grüter and Keller 2016).

Experimentally manipulating CHCs has been an effective method of examining how

CHCs influence agonistic behaviors in both eusocial and solitary Hymenoptera (Ruther et al.

1998, 2002: Bonavita-Cougourdan et al., 1987; Flores-Prado et al. 2008; Dani et al. 2001, 2005;

Nowbahari et al., 1990). In both solitary and eusocial Hymenoptera, non-nestmates washed in solvent received little aggression while control (i.e., untreated) non-nestmates received significantly higher rates of aggression (Ruther et al. 1998; Bonavita-Cougourdan et al., 1987;

Nowbahari et al., 1990; Flores-Prado et al. 2008). This has been demonstrated in a solitary bee,

Manuelia postica, suggesting that CHCs are recognition signals in solitary species and thus confirming that eusocial traits may have adaptive significance in solitary species (Flores-Prado et al. 2008). However, in these studies only dead solvent washed individuals were used during the behavioral assays.

While these studies have contributed to our understanding of how CHCs mediate behavioral interactions in solitary and eusocial Hymenoptera, there have been no studies examining how CHCs affect the outcome of behavioral interactions in a subsocial species. Since subsociality is a necessary pre-requisite for eusociality (Michener 1990) understanding the link between CHCs and behavior in subsocial species is essential to our understanding of social evolution. CHCs are recognition signals in both solitary and eusocial bees (Flores-Prado et al.

2008; Pradella et al. 2015). Therefore, CHCs are likely used for recognition in subsocial bees, although this still has yet to be verified with empirical data. Here we conduct the first study to examine how CHCs affect behavioral interactions between non-nestmates of the small carpenter bee species Ceratina calcarata Robertson (Hymenoptera: Apidae: Xylocopinae). 37

The aims of this study were three-fold. First, we aimed to develop the first CHC removal assay using two live female C. calcarata non-nestmates, as to date experiments have only used dead solvent washed specimens (Bonavita-Cougourdan et al., 1987; Nowbahari et al., 1990;

Flores-Prado et al. 2008, Ruther et al. 1998). We provide a new method for removing CHCs in live bees and verify its effectiveness using circle tube behavioral assays. Second, we aimed to examine how live solvent washed non-nestmates interact compared to control bees which have been physically manipulated but not chemically treated. We predicted less aggression between solvent washed females than control females. Third, we conducted assays with control females versus solvent washed females, predicting that solvent washed females would be more aggressive towards control females, while control females would be less aggressive towards solvent washed females. Finally, by observing differential aggression based on the presence of absence of a CHC profile, we aimed to show the first evidence for a chemical basis of non- nestmate discrimination in a subsocial bee.

METHODS Bee Collections

Ceratina calcarata nests were collected in New Hampshire, U.S.A. (43º08º02 N 70 º55

º35 W) from May through August during the summers of 2014 and 2015. Since C. calcarata behavior varies by age and reproductive status (Rehan and Richards 2013) we collected nests from different parts of the colony cycle. Founding nests (FN) were new, and therefore do not contain any fecal matter or waste products of any kind from a previous nest owner. In addition, founding nests do not yet contain brood cells. Therefore, founding nest females were the youngest, and also pre-reproductive. Active brood (AB) nests contain at least one brood cell containing a pollen ball and egg, indicating these females are actively reproductive. When the 38 brood cell closest to the nest entrance contains a larva or a pupa the nest is defined as a full brood

(FB) nest, as the mother’s reproductive effort for the season is complete and she will not be laying any more eggs. Thus, full brood females are the oldest and also post-reproductive (Rehan and Richards 2010).

Nests were collected from sumac trees, Rhus typhina, between 6:00 and 8:00 AM to ensure the mother was still in the nest. To avoid damage to the contents of the nests and any offspring present, the dead broken stem was cut with garden shears at the junction with another branch. Masking tape was then applied to the nest entrance to contain the bees living inside.

Collected nests were then brought back to the lab and stored in a 4°C cold room until dissection.

The nest dissections and behavioral experiment preparations described in the next section were completed in less than two hours from field nest collections to avoid any changes in behavior due to captivity.

Nest Dissections and Behavioral Assay Preparations

Nests were dissected with extreme care to avoid damaging the contents of the nest. The dead broken stems (nests) were split long-ways using a sharp non-serrated knife. Nest type (e.g., founding nest, active brood, and full brood) was determined and recorded. Females were captured and stored in 2ml microcentrifuge tubes with an air hole poked in the top until behavioral assay preparations occurred. After nest dissections females were prepared for behavioral trials. Bees were chilled on ice during preparation.

Preparation for Behavioral Assays

We use the novel method of swabbing bees with the non-polar solvent pentane to remove

CHCs. Using an Eppendorf Research Plus P200 pipette, 40 µL of pentane (Sigma-Aldrich) was 39 pipetted onto a 2 mm3 piece of a cosmetic sponge. First, the dorsal side of the bee was swabbed for 15 seconds. After swabbing the dorsal side there was a 30 second pause before swabbing the ventral side as to avoid over exposing the bee to pentane. Since pentane is highly volatile, 40 µL pentane was reapplied to the sponge and the ventral side of the bee was swabbed for 15 seconds.

As a control, water swabbed females were used in behavioral trials, using the same method as described above, except 40 µL of water was pipetted onto the sponge. After swabbing, females were uniquely painted with a single spot on the dorsal surface of the thorax between the wings using a Sharpie enamel paint pen.

Circle Tube Behavioral Assays

There were three trial types for this experiment: pentane versus pentane (PP), water versus water (WW), and water versus pentane (WP) dyads. Behavioral assays were performed outside on clear and sunny days in direct sunlight between 10:30 AM and 3:00 PM. Trials were not performed on days in which it was raining or cloudy, as the bees require direct sunlight to engage in any activity. Furthermore, assays needed to be performed during the hottest part of the day

(i.e., mid-day).

The following behavioral assays are adopted and modified as described by Breed and colleagues (1978). The prepared adult females were inserted into opposite ends of a clear 30 cm long plastic tube with a 4 mm internal diameter. The diameter was large enough that two bees could pass each other, but also small enough that contact is necessary for passing to occur, and one bee was capable of blocking the other bee from passing. All behaviors are described in Table

1. Behavioral data were recorded in terms of both frequency and latency to first instance for the complete duration of each 20 minute trial (1200 seconds). 40

Four categories of behaviors are commonly observed during circle tube assays: aggression, avoidance, tolerance, and following (Table 1). Behaviors were only recorded when two bees came within one full body length of each other (Kukuk 1992; Packer 2005, Rehan and

Richards, 2013). All behaviors were binned into each of their respective categories before statistical analyses were performed.

Morphological and Physiological Measurements

After behavioral trials bees were stored at -80ºC until morphological and physiological measurements were recorded. We measured three morphological and physiological characteristics typically recorded in circle tube literature (Rehan and Richards 2013). First, head width (i.e. greatest distance of head width, including eyes) was measured as it is an accurate indication of body size (Rehan and Richards 2010). Second, wing wear was recorded as it is an honest predictor of foraging effort and age (Cartar 1992). Wing wear score was determined using a scale between zero and five. Females with new, unworn wings received a wing wear score of zero, females with moderately damaged wings received a wing wear score of three, and females with wings critically torn on the apical margins received a wing wear score of five. Third, using the same dissecting scope we scored ovarian development as the sum of the length of the three largest oocytes (Rehan and Richards 2013).

Statistical Analyses

All statistical analyses were calculated using R Studio 3.2.2 (R Core Team 2014).

Shapiro-Wilkes tests were used to test for normal distribution of ovarian development, head width, and wing wear. All three physiological measurements were not normally distributed and 41 non-parametric statistics were used. Since there are no multivariate statistical analyses for non- normally distributed data, as an alternative we use a series of non-parametric t-tests (Wilcox tests), ANOVA (Kruskal-Wallis tests), and correlations (Spearman Rank Correlation tests) to analyze these data.

We used Wilcox tests to compare the total behavioral interactions, aggression, avoidance, following, and tolerance between PP and WW trials. We also calculated the relative frequencies of behaviors by dividing the absolute frequency of behaviors by the total behavioral interactions.

Wilcox tests were used to compare the relative frequencies of behaviors between PP and WW trials. We performed Wilcoxon tests to compare head width, wing wear score, and ovarian development between PP and WW trials. We compared the absolute and relative behaviors between W and P bees in WP trials using Paired Wilcoxon tests. We also used Paired Wilcoxon tests to compare head width, wing wear, and ovarian development between W and P bees in WP trials.

Richards and Packers (2010) addressed statistical issues common in circle tube assay literature. When dealing with dyads, they argue that the behavior of one member of each dyad is affected by the other member. We calculated the differences in behavior between pairs in PP,

WP, and WW trials and used Kruskal-Wallis tests to compare the differences in behaviors between PP, WP, and WW trials. We also calculated the differences between behaviors and morphological measurements and used Spearman Rank Correlations to determine if there were correlations between behaviors of morphological measurements in PP and WW trials.

42

RESULTS Circle Tube Assays

There were a total of 88 PP trials, 82 WP trials, and 85 WW trials. Among founding nest females there were 29 PP trials, 28 WP trials, and 31 WW trials. In active brood females there were 29 PP trials, 24 WP trials, and 23 WW trials. In full brood females there were 30 PP trials,

30 WP trials, and 30 WW trials.

Overall, the mean frequency of total behavioral interactions in PP trials was 22.92 ± 1.05, which was significantly higher than WW trials in which the mean total encounters was 15.97 ±

1.03 (Table 2; p < 0.001). Aggression comprised 19.64 ± 1.57% (mean +/- SE) of the total behavioral interactions in PP trials and 20.26 ± 1.82% in WW trials. There was 18.58 ± 1.52% avoidance in PP trials and 22.66 ± 2.01% in WW trials. Following was the least observed behavior and only comprised 8.34 ± 0.83 % of total behaviors in PP trials and 12.41 ± 2.01 % in

WW trials. The most common behavior in both PP and WW trials was tolerance. There was significantly more tolerance in PP than WW trials, with 49.2 ± 2.29 % and 41.22 ± 2.39 %, respectively (Table S1; p = 0.0168).

There was significantly more avoidance in PP trials than WW trials, but only in the younger, pre-reproductive founding nest females (Table 2; p = 0.01). In post-reproductive and older full brood females there was significantly more tolerance in PP than WW trials (Table 2; p

< 0.01). Furthermore, in full brood females 63% of behaviors were tolerance in PP trials, and

40% were tolerance in WW trials (Table S1; p < 0.001). There was also significantly more relative avoidance and following in WW trials than PP trials in full brood females (Table S1).

In active brood WP trials, P bees displayed significantly more avoidance than W bees

(Table 3). There were no other significant differences in absolute frequency of behaviors between W and P bees overall and by nest class (Table 3). In active brood females W bees 43 displayed significantly more relative tolerance than P bees (Table S3). There were no other differences in the relative frequencies of behaviors between W and P bees overall and by nest class (Table S3). Kruskal-Wallis tests comparing the absolute differences in behaviors between dyads in PP, WP, and WW trials showed that in active brood trials W bees were significantly more avoidant than P bees in WP trials (Table S5). There were no other significant differences between PP, WP, and WW trials overall and by nest class (Table S5).

Morphological and Physiological Measurements

Overall, there were no other significant differences between head width, wing wear, or ovarian development between PP and WW trials (Table S2). Paired Wilcoxon tests showed there were no differences in head width, wing wear, or ovarian development between W and P bees in

WP trials (Table S4). There was a significant positive correlation between following and ovarian development in PP trials (Table S6), but no other correlations between behaviors and morphological measurements in PP or WW trials (Table S6).

DISCUSSION

The results of this study were two-fold. First, we have developed a novel CHC removal assay using live bees and demonstrate that it is both safe and effective. Second, we provide the first evidence that CHCs are signal molecules used for recognition and mediating behavioral interactions in a subsocial bee species, Ceratina calcarata. We show that solvent washed non- nestmates exhibit different rates of avoidance and tolerance than control females. However, this differential expression of behaviors does not only vary as a function of chemical treatment, but is also dependent on reproductive status. Since we show CHCs are used to identify non-nestmates 44 in a subsocial bee this suggests that the use of CHCs to recognize non-nestmates probably predated the evolution of eusociality.

The first aim of this study was to develop a CHC removal assay to examine behavioral differences using two live bees. We found that there were significantly more total behavioral interactions in solvent washed trials than control trials (Table 2), demonstrating that our method of CHC removal via pentane swabbing does not negatively affect the bees’ activity. Furthermore, our method of swabbing was effective in removing CHCs. Previous experiments in which individuals were washed with solvent have shown that test subjects are more aggressive towards untreated non-nestmates than solvent washed non-nestmates (Bonavita-Cougourdan et al. 1987;

Nowbahari et al. 1990; Flores-Prado et al. 2008; Ruther et al. 1998). Much like these experiments we also observed behavioral changes associated with treatment, suggesting that pentane swabbing does remove CHC profiles. Thus, we have developed a CHC removal assay using live test subjects. Moreover, this method is not exclusive to C. calcarata and can be applied to many other insect species.

We originally predicted greater aggression between control non-nestmates than solvent washed non-nestmates. While we did not observe greater aggression between control females we did observe significantly less tolerance. In the solitary bee Manuelia postica test subjects were equally as tolerant towards solvent washed and control non-nestmates (Flores-Prado et al. 2008).

However, our results are inconsistent with this as we found that solvent washed non-nestmates are significantly more tolerant to each other than control non-nestmates (Table 2 and S1). These results are consistent with Ruther et al. (1998) which also found that test subjects exhibited more

“neutral” behaviors towards solvent washed non-nestmates than control non-nestmates. 45

Although the most common behavior in this study was tolerance, C. calcarata females also exhibited moderate rates of aggression towards non-nestmates (Table S1). In solitary species aggression towards non-nestmates is rare (Richards and Packers 2010; Packer 2006), while eusocial species are highly aggressive towards non-nestmates (Breed et al. 1978). However, we found that C. calcarata is intermediate in this regard as about 20% of interactions were aggressive acts (Table S1). Rehan and Richards (2013) also found in circle tube assays that non- nestmates mostly display tolerance, but also exhibit moderate aggression. In C. calcarata, mothers are the principle nest guards and defend the nest. Therefore, aggression towards non- nestmates may be a defense against usurpation (Boesi and Polidori 2011). This suggests that non-nestmate aggression, a trait exhibited by eusocial species (Wilson 1976; Buckle and

Greenberg 1981; Wheeler 1991; Grüter et al. 2012), probably has adaptive advantages in subsocial species.

While individuals in each dyad were randomly paired we needed to confirm that morphological and physical traits were not significantly different. Therefore, we examined morphological measurements to verify whether the differences in behaviors between trial types were due to body size or reproductive status, or truly due to the chemical treatment. Consistent with Flores-Prado et al. (2008) we found that there were no differences in morphological measurements between individuals in dyads suggesting that behavioral differences between trial types were due to chemical treatment and not differences in morphological characteristics.

Some studies have reported that larger body size has also been associated with aggression (Boesi and Polidori 2011; Pabalan et al. 2000). However, in many studies body size differences are not associated with the frequency of agonistic interactions (McConnel-Garner and Kukuk 1997; 46

Richards and Packer 2010; Rehan and Richards 2013), consistent with our results for C. calcarata females (Table S6).

Studies in both solitary (Arneson and Wcislo 2003) and eusocial bees (Breed et al. 1978) have shown an effect of ovarian size on dominance behaviors, although results not supporting this have been reported (McConnell-Garner and Kukuk 1997). We did not find that ovarian development was correlated with aggression, but we did find a significant positive correlation between ovarian development and following behavior. Withee and Rehan (2016) found aggression and following to be positively correlated in C. calcarata. Furthermore we found that reproductive females are the most aggressive, also consistent with Rehan and Richards (2013).

Therefore these results are suggestive that dominance displays may be associated with larger ovaries and reproductive status. However, since we only observed this relationship between solvent washed females, this may suggest highly reproductive females possess an innate capacity to express dominance independent of the presence of chemical cues from other individuals.

In addition to differences in tolerance we also observed differences in avoidance associated with reproductive status. There was significantly more relative avoidance between post-reproductive control females than solvent washed females (Table S1). This suggests that when the chemical cues of a non-nestmate are present post-reproductive females tend to be more avoidant. Contrastingly, when chemical cues are removed avoidance significantly decreases

(Table S1). Therefore, there is an interesting inverse relationship between relative avoidance and tolerance in post-reproductive females (Table S1). When non-nestmate chemical cues are present there is more avoidance and less tolerance, but in the absence of chemical cues non-nestmates are significantly more tolerant towards each other. Some studies which have experimentally manipulated CHC profiles have only recorded aggressive behaviors (Dani et al. 2001, 2005) 47 while others have had more specific ethograms which included tolerance in addition to aggression (Nowbahari et al. 1990; Ruther et al. 1998; Flores-Prado et al. 2008). However, no studies so far have utilized ethograms including both avoidance and tolerance. Therefore, our results are likely the first to report changes in non-nestmate avoidance and tolerance induced by the presence or absence of a CHC profile. Since control non-nestmates are less tolerant towards each other than solvent washed non-nestmates, and control non-nestmates are more avoidant this suggests that C. calcarata uses CHCs as chemical cues for non-nestmate recognition.

This is the first study suggesting a possible chemical basis of recognition in a subsocial bees. Solitary and subsociality are both pre-requisites for eusociality (Michener 1990), and there is now data suggesting that CHCs are signal molecules used for recognition in both solitary

(Flores-Prado et al. 2008) and subsocial bees (this study). Therefore the use of CHCs in recognition is likely not a derived trait unique to eusocial lineages, but is instead a conserved trait and a probable necessary precursor for the evolution of eusociality.

48

TABLES

Table 1: The observed behaviors of Ceratina calcarata and their respective definitions during circle tube behavioral assays. Behavioral Category Behavior Definition Aggression Nudging One bee quickly applies force to the other bee using its head. Rehan and Richards (2013) noted that head-butting, lunging (Packer et al. 2003), and pushing (Peso and Richards 2010) are all synonymous with nudging. C-Posture One bee curves its body into a “C” shape, and points its stinger at the other bee. Biting One bee opens its mandibles and clamps down on a body part of the other bee.

Avoidance Reversal One bee switches directions when confronting another bee. Backing One bee walks backwards away from the other bee when they confront each other.

Tolerant Passing Both bees pass each other inside the circle tube. Direction and orientation of the passing.

Following Following One bee follows the other bee regardless of direction and/or orientation. Note: that true behavioral interactions were only recorded when both bees were at least one body length away from each other (Rehan and Richards 2013).

49

Table 2: Wilcox Tests comparing the absolute frequency of behaviors between PP and WW trials overall and within pre-reproductive (FN), reproductive (AB), and post-reproductive (FB) Ceratina calcarata females. PP WW All Females Mean ± S. E. Range Mean ± S. E. Range W Statistic/p-value Total Behavioral 20.92 ± 1.05 0-59 15.97 ± 1.03 0-59 W = 20434 Interactions p-value = 0.0003295*** Aggression 3.92 ± 0.40 0-33 3.33 ± 0.36 0-23 W = 18390 p-value = 0.1088 Avoidance 3.08 ± 0.24 0-18 2.43 ± 0.19 0-12 W = 18389 p-value = 0.1105 Following 1.77 ± 0.18 0-14 1.77 ± 0.17 0-10 W = 16394 p-value = 0.6842 Tolerance 12.16 ± 0.88 0-49 8.44 ± 0.82 0-47 W = 20585 p-value = 0.0001771*** Founding Aggression 3.62 ± 0.48 0-20 2.84 ± 0.51 0-15 W = 2421.5 Nest p-value = 0.06854 Avoidance 3.69 ± 0.47 0-18 2.38 ± 0.37 0-12 W = 2580.5 p-value = 0.01009** Following 1.78 ± 0.32 0-11 1.97 ± 0.32 0-10 W = 1998.5 p-value = 0.8091 Tolerance 12.02 ± 1.70 0-49 7.79 ± 1.11 0-32 W = 2398 p-value = 0.09394 Active Brood Aggression 5.71 ± 0.89 0-30 4.00 ± 0.70 0-18 W = 1784 p-value = 0.1905 Avoidance 3.42 ± 0.42 0-14 2.23 ± 0.31 0-10 W = 1836 p-value = 0.1053 Following 2.37 ± 0.38 0-14 1.96 ± 0.31 0-8 W = 1850 p-value = 0.09186 Tolerance 9.78 ± 1.01 0-32 9.69 ± 1.87 0-47 W = 1850 p-value = 0.09186 Full Brood Aggression 2.40 ± 0.60 0-33 3.34 ± 0.67 0-23 W = 1802 p-value = 0.4426 Avoidance 2.11 ± 0.31 0-13 2.66 ± 0.31 0-12 W = 1630 p-value = 0.1056 Following 1.13 ± 0.20 0-7 1.45 ± 0.24 0-8 W = 1773.5 p-value = 0.3457 Tolerance 14.76 ± 1.72 0-47 8.26 ± 1.39 0-46 W = 2587 p-value = 0.001694** Note: * indicates p<0.05, ** indicates p<0.01, and *** indicates p<0.001.

50

Table 3: Paired Wilcox tests comparing the absolute frequency of behaviors between W and P Ceratina calcarata females in WP trials, overall, and within pre-reproductive (FN), reproductive (AB), and post-reproductive (FB) females. W P All Females Mean ± S.E. Range Mean ± S.E. Range V-statistic/p-value (n=82) Aggression 3.52 ± 0.50 0-20 3.41 ± 0.59 0-34 V = 974.5 p-value = 0.5228 Avoidance 3.16 ± 0.35 0-13 2.43 ± 0.31 0-13 V = 1030.5 p-value = 0.109 Following 2.16 ± 0.30 0-9 1.62 ± 0.25 0-12 V = 816 p-value = 0.1877 Tolerance 12.24 ± 1.50 0-71 12.24 ± 1.56 0-72 V = 399 p-value = 0.3713 Founding Nest Aggression 2.93 ± 0.67 0-13 3.79 ± 1.06 0-23 V = 150.5 (n=28) p-value = 1 Avoidance 3.25 ± 0.58 0-13 2.5 ± 0.50 0-11 V = 194 p-value = 0.4013 Following 2.39 ± 0.52 0-9 2.21 ± 0.54 0-12 V = 141 p-value = 0.6476 Tolerance 11.36 ± 3.06 0-71 11 ± 3.11 0-72 V = 60.5 p-value = 0.0923 Active Brood Aggression 5.75 ± 1.17 0-20 3.92 ± 0.79 0-13 V = 75.5 (n=24) p-value = 0.2777 Avoidance 4.79 ± 0.80 0-12 3.08 ± 0.70 0-13 V = 72.5 p-value = 0.04708* Following 2.17 ± 0.53 0-8 1.46 ± 0.45 0-10 V = 83 p-value = 0.4205 Tolerance 16.7 ± 3.00 0-51 17.54 ± 3.06 0-53 V = 62 p-value = 0.5688 Full Brood Aggression 2.3 ± 0.68 0-15 2.67 ± 1.13 0-34 V = 122 (n=30) p-value = 0.8332 Avoidance 1.77 ± 0.33 0-7 1.87 ± 0.39 0-7 V = 161 p-value = 0.7557 Following 1.93 ± 0.51 0-9 1.2 ± 0.28 0-6 V = 60 p-value = 0.447 Tolerance 9.5 ± 1.60 0-27 9.17 ± 1.73 0-24 V = 60 p-value = 0.447 Note: * indicates p<0.05, ** indicates p<0.01, and *** indicates p<0.001.

51

Chapter 4: General Conclusions and Future Directions

Eusociality is an extremely derived and highly successful reproductive strategy (Wilson

1971). To fully understand the origins of eusociality, less social forms must be studied (Rehan and Toth 2015). Here I used the subsocial bee species Ceratina calcarata as a model organism

(Rehan and Richards 2010). In this thesis I contributed to our understanding of the transition from subsocial to eusocial life by addressing two main questions. First, were CHCs fertility signals present in solitary ancestors and then subsequently coopted as queen pheromones?

Second, does the use of CHCs for recognition of conspecifics precede eusociality?

The first major evolutionary hypothesis I tested was that CHCs were fertility signals and were subsequently coopted as queen pheromones in eusocial lineages (Van Oystaeyen et al.

2014). My results (Chapter 2) were suggestive that pentacosane may possibly be a fertility signal in a subsocial bee, C. calcarata. However, additional species need to be studied to better support our findings that fertility signals were coopted from solitary ancestors during the evolution of eusociality. While my results were suggestive that C. calcarata signal fertility by overexpressing pentacosane, further experimentation would be advantageous. The logical next step would be to perform behavioral bioassays by experimentally treating C. calcarata females with pentacosane and measuring the behavioral and physiological effects. In the bumblebee, Bombus terrestris, pentacosane is a queen pheromone that has been experimentally verified to confirm its role in regulation of worker reproduction (Holman 2014). Therefore, it would be interesting to observe the effects of pentacosane on C. calcarata ovarian development. Since C. calcarata is subsocial,

I propose that pentacosane would have negligible effects on ovarian development. My results from Chapter 2 also revealed that several other compounds, namely farnesol, ethylhexadecenoate, nonadecane, and heneicosane are also differentially expressed in C. 52 calcarata. Performing bioassays with these isolated compounds would also be of great interest for future research in this species to verify their causal roles on behavior and reproduction.

The second main evolutionary hypothesis I tested was that the use of CHCs in recognition is not a derived trait unique to eusocial lineages, but instead a conserved trait with adaptive advantages and fitness benefits in less social forms. In Chapter 3 of this thesis I report suggestive evidence that CHCs are used to identify non-nestmates. I found differential expression of avoidance and tolerance based on chemical treatment. However, in this experiment I focused on non-nestmates for behavioral trials. Future studies could examine if C. calcarata uses CHCs to discriminate non-nestmates from nestmates. Rehan and Richards (2013) showed that C. calcarata are capable of discriminating non-nestmates from nestmates. Therefore, we know that this species is capable of discriminating non-nestmates from nestmates (Rehan and Richards

2013), however a chemical basis for this has not been shown.

Another major finding from Chapter 3 is that my method of removing CHCs is both safe and effective. Swabbing females with pentane did not significantly decrease their overall activity levels. Further, since differential expression of behaviors was observed, this suggests that the pentane swabbing did indeed remove their CHC profiles.

In summary, this thesis contributed to our understanding of the evolution of eusociality, by examining the transition from subsocial to eusocial. I show two major findings regarding the evolution of eusociality. First, I show suggestive evidence that a subsocial bee may signal fertility, supporting the hypothesis that fertility signals preceded queen pheromones. Second, I show that a subsocial bee may use CHCs as recognition signals, suggesting that the use of CHCs in recognition is most likely not a derived trait exclusive to eusocial lineages.

53

REFERENCES

Andersson J, Borg-Karlson AK, Wiklund C (2000) Sexual cooperation and conflict in butterflies: a male-transferred anti-aphrodisiac reduces harassment of recently mated females. Proc R Soc B 267:1271-1275.

Andersson J, Borg-Karlson AK, Wiklund C (2003) Antiaphrodisiacs in pierid butterflies: a theme with variation! J Chem Ecol 29:1489-1499.

Ayasse M, Marlovits T, Tengö J, Taghizadeh T, Francke W (1995) Are there pheromonal dominance signals in the bumblebee Bombus hypnorum L (Hymenoptera, Apidae)? Apidologie 26:163-180.

Boesi R, Polidori C (2011) Nest membership determines the levels of aggression and cooperation between females of a supposedly communal digger wasp. Aggressive Behavior 37(5): 405-416

Bonavita-Cougourdan A, Clément JL, Lange C (1987) Nestmate recognition: the role of cuticular hydrocarbons in the ant Camponotus vagus Scop. J Entomol Sci 22: 1-10.

Bonavita-Cougourdan A, Theraulaz G, Bagnéres A-G, Roux M, Pratte M, Provost E, Clément J- L (1991) Cuticular hydrocarbons, social organization and ovarian development in a polistine wasp: Polistes dominulus Christ. Comp Biochem Physiol 100b(4): 667-680.

Brady SG, Sipes S, Pearson A, Danforth BN (2006) Recent and simultaneous origins of eusociality in halictid bees. The Royal Society 272(1594): 1643-1649.

Breed MD (1998) Recognition pheromones of the honey bee. BioScience 48: 463-470.

Breed MD, Silverman JM, Bell WJ (1978) Agonistic behavior, social interactions and behavioural specialization in a primitively eusocial bee. Insectes Sociaux 25: 351-364.

Brooks RW, Cane JH (1984) Origin and chemistry of the secreted nest entrance lining of Halictus hespersus (Hymenoptera: Hepoidae). Journal of the Kansas Entomological Society 57(1): 161-165.

Brothers DJ, Michener CD (1974) Interactions in colonies of primitively social bees. III Ethometry of division of labour in Lasioglossum zephyrum. Journal of Comparative Physiology 90:129-168.

Buckle GR, Greenberg L (1981) Nestmate recognition in sweat bees (Lasioglossum zephyrum): Does an individual recognize its own odour or only odours of its nestmates? Anim Behav 29: 802-809.

Cartar RV (1992) Morphological senescence and longevity: an experiment relating wing wear and life span in foraging wild bumble bees. J Anim Ecol 61:225-231.

54

Chao TH, Ehrman L, Permaul A, Vincent R, Sattaur L, Brandt D (2010) Male-specific cuticular compounds of the six Drosophila paulistorum semispecies: structural identification and mating effect. J Chem Ecol 36:933-942.

Chung H, Carroll SB (2015) Wax, sex and the origin of species: Dual roles of insect cuticular hydrocarbons in adaptation and mating." Bioessays 37(7): 822-830.

Crespi BJ (1992) Eusociality in Australian gall thrips. Nature 359: 724-726.

Cuvillier-Hot V, Cobb M, Malosse C, Peeters C (2001) Sex, age and ovarian activity affect cuticular hydrocarbons in Diacamma ceylonense, a queenless ant. J Insect Physiol 47(4-5): 485- 493.

Danforth BN (2002) Evolution of sociality in a primitively eusocial lineage of bees. Proceedings of the National Academy of Science 99: 286-290.

Dani FR, Jones GR, Destri S, Spencer SH, Turillazzi S (2001) Deciphering the recognition signature within the cuticular chemical profile of paper wasps. Anim Behav 62(1): 165-171.

Dani FR, Jones GR, Corsi S, Beard R, Pradella D, Turillazzi S (2005) Nestmate recognition cues in the honey bee: differential importance of cuticular alkanes and alkenes. Chem Senses 30(6):477-489. de Mendiburu F (2015). agricolae: Statistical Procedures for Agricultural Research. R package version 1.2-2. http://CRAN.R-project.org/package=agricolae

Dorado-Correa AM, Goerlitz HR, Siemers BM (2013) Interspecific acoustic recognition in two European bat communities. Frontiers in Physiology 4: 192. dos Santos AB, do Nascimento FB (2015) Cuticular hydrocarbons of orchid bee males: interspecific and chemotaxonomy variation. PLOS one. 10(12): e0145070.

Duchateau MJ, Velthuis HHW (1989) Ovarian development and egg laying in workers of Bombus terrestris. Entomologia Experamentalis et Applicata 51:199-213.

Duffy J E (1996) Eusociality in a coral-reef shrimp. Nature 381: 512-514.

Esponda F, Gordon DM (2015) Distributed nestmate recognition in ants. Proc R Soc B 282: 20142838.

Flores-Prado L, Aguilera-Olivares D, Niemeyer HM (2008) Nest-mate recognition in Manuelia postica (Apidae: Xylocopinae), an eusocial trait is present in a solitary bee. Proc R Soc B 275:285-291.

Gerlach G, Hodgins-Davis A, Avolio C, Schunter C (2008) Kin recognition in zebrafish: a 24- hour window for olfactory imprinting. Proc R Soc Lond B 275: 2165-2170. 55

Gershman SN, Rundle HD (2016) Level up: The expression of male sexually selected cuticular hydrocarbons is mediated by sexual experience. Anim Behav 112:169-177.

Gibbs J, Dumesh S, Packer L, Danforth BN (2013) Revision and reclassification of Lasioglossum (Evylaeus), L. (Hemihalictus) and L. (Sphecodogastra) in eastern North America (Hymenoptera: Apoidea: Halictidae). Zootaxa 3672: 1–117.

Grüter C, Keller L (2016) Inter-caste communication in social insects. Current Opinion in Neurobiology 38: 6-11.

Grüter C, Menezes C, Imperatriz-Fonseca VL, Ratnieks FLW (2012) A morphologically specialized soldier caste improves colony defense in a neotropical eusocial bee. Proc Nat Acad Sci USA 109: 1182-1186.

Happ GM (1969) Multiple sex pheromones of the mealworm beetle, Tenebrio molitor L. Nature 222:180-181.

Heinze J, Stengl B, Sledge MF (2002) Worker rank, reproductive status and cuticular hydrocarbon signature in the ant, Pachycondyla cf. inversa. Behave col Sociobiol 52:59-65.

Hetchinger RF, Wood AC, Kuris AM (2011) Social organization in a flatworm: trematode parasites form soldier and reproductive castes. The Royal Society 278(1706): 656-665.

Hinz C, Kobbenbring S, Kress S, Sigman L, Müller A, Gerlach G (2013) Kin recognition in zebrafish, Danio rerio, is based on imprinting on olfactory and visual stimuli. Anim Behav 85: 925-930.

Hölldobler B, Wilson EO (1990) The Ants. Harvard University Press, Cambridge, Mass.

Hölldobler B, Wilson, EO (2009) The superorganism: the beauty, elegance, and strangeness of insect societies. WW Norton & Company.

Holman L (2014) Bumblebee size polymorphism and worker response to queen pheromone. PeerJ 2: e604.

Holman L, Jǿrgensen CG, Nielen J, d’Ettore P (2010) Identification of an ant queen pheromone regulating worker sterility. Proc R Soc B 282(1814).

Holman L, Lanfear R, d’Ettore P (2013) The evolution of queen pheromones in the ant genus Lasius. J Evol Biol 26(7):1549-1558.

Howard RW (1993) Cuticular hydrocarbons and chemical communication. In: Stanley DW, Nelson DR (ed) Insect Lipids: Chemistry, Biochemistry, and Biology.

56

Howard RW, Blomquist GJ (1982) Chemical ecology and biochemistry of insect hydrocarbons. Annu Rev Entomol 27:149-172.

Howard RW, Blomquist GJ (2005) Ecological, behavioral, and biochemical aspects of insect hydrocarbons. Annu Rev Entomol 50:371-393.

Jackson LL, Bartelt RJ (1986) Cuticular hydrocarbons of Drosophila virilis: comparison by age and sex. Insect Biochem 16(2):433-439.

Jarvis JU (1981) Eusociality in a mammal: cooperative breeding in naked mole-rat colonies. Science 212: 571-573.

Johnstone RA, Grafen A (1993) Dishonesty and the handicap principle. Animal Behaviour 46(4): 759-764.

Kaurse ET, Krüger O, Kohlmeier P, Caspers BA (2012) Olfactory kin recognition in a songbird. Biology Letters 8: 327-329.

Kent DS, Simpson JA (1992) Eusociality in the beetle Austroplatypus incompertus (Coleoptera: Curculionidae). Naturwissenschaften 79(2): 86-87.

Kocher SD, Paxton RJ (2014) Comparative methods offer powerful insights into social evolution in bees. Apidologie 45: 289-305.

Kronenberg S, Hefetz A (1984) Comparative analysis of dufour’s gland secretions of two carpenter bees (Xylcopinae: Anthophoridae) with different nesting habits. Comparative Biochemistry and Physiology 79B(3): 421-425.

Kukuk PF (1992) Social Interactions and familiarity in a communal halictine bee Lasioglossum (Chilalictus) hemichalceum. Ethology 91: 291-300.

Kulahci IG, Drea CM, Rubenstein DI, Ghazanfar AA (2014) Individual recognition through olfactory–auditory matching in lemurs. Proc R Soc Lond B 281: 20140071.

Leavesley AJ, Magrath RD (2005) Communicating about danger: urgency alarm calling in a bird. Animal Behavior 70(2): 365-373.

Liang D, Silverman J (2000) “You are what you eat”: diet modifies cuticular hydrocarbons and nest mate recognition in the Argentine ant, Linepithema humile. Naturwissenschaften 87:412-416.

Liebig J, Peeters C, Oldham NJ, Markstädter C, Hӧlldobler B (2000) Are variations in cuticular hydrocarbons of queens and workers a reliable signal of fertility in the ant Harpegnathos saltator? Proc Natl Acad Sci USA 97:4124-4131.

57

Lorenzi MC, Sledge MF, Laiolo P, Sturlini E, Turillazzi S (2004) Cuticular hydrocarbon dynamics in young Polistes dominulus (Hymenoptera: Vespidae) and their role of linear hydrocarbons in nestmate recognition systems. J Insect Physiol 50(10):935-941.

Mant J, Brändli C, Vareecken NH, Schulz CM, Francke W, Schiestl FP (2005) Cuticular hydrocarbons as sex pheromones of the bee Colletes cunicularis and the key to its mimicry by the sexually deceptive orchid, Ophrys exaltata I. J Chem Ecol 31:1765-1878.

Marco A, Chivers DP, Keisecker JM, Blaustein AR (1998) Mate choice by chemical cues in Western Redback (Plethodon vehiculum) and Dunn’s (P. dunni) salamanders. Ethology 104(9): 781-788.

Martin J, Lopez P (2006) Links between male quality, male chemical signals, and female mate choice in Iberian Rock lizards. Functional Ecology 20(6): 1087-1096.

McConnell-Garner J, Kukuk PF (1997) Behavioral Interactions of two solitary, halictine bees with comparisons among solitary, communal, and eusocial species. Ethology 103: 19-32.

Michener CD (1974) Social Behavior of the Bees. Harvard University Press.

Michener CD (1985) From solitary to eusocial: need there be a series of intervening species? Fortschritte der Zoologie 31: 293-305.

Michener CD (1990) Castes in xylocopine bees. In: Social Insects: an Evolutionary Approach to Castes and Reproduction (Ed. by W. Engels), pp. 123-146. New York: Springer-Verlag

Nowbahari E, Lenoir A, Clément JL, Lange C, Bagneres AG, Joulie C (1990) Individual, geographical and experimental variation of cuticular hydrocarbons of the ant Cataglyphis cursor (Hymenoptera: Formicidae): Their use in nest and subspecies recognition. Biochem System and Ecol 18: 63-73.

Nunes TM, Turatti IC, Mateus S, Nascimento FS, Lopes NP, Zucchi R (2009) Cuticular hydrocarbons in the stingless bee Schwarziana quadripunctata (Hymenoptera: Apidae: Meliponini): differences between colonies, castes, and age. Genet Mol Res 8:589-595.

Oliveira RC, Oi CA, do Nascimento MMC et al. (2015) The origin and evolution of queen and fertility signals in corbiculate bees. BMC Evol Biol 15:254.

Oppelt A, Heinze J (2009) Mating is associated with immediate changes in the hydrocarbon profile of Leptothorax gredleri and queens. J Insect Physiol 55:624-628.

Pabalan N, Davey KG, Packer L (2000) Escalation of aggressive interactions during staged encounters in Halictus ligatus (Hymenoptera: Halictidae), with a comparison of circle tube behaviors with other halictine species. J Insect Behavior 13: 627-650.

58

Packer L (2005) The influence of marking upon bee behavior in circle tube experiments with a methodological comparison among studies. Insectes Sociaux 52: 139-146.

Packer L (2006) Use of artificial arenas to predict the social organization of halictine bees: data for fourteen species from Chile. Insectes Sociaux 53: 307-315.

Parr LA, de Waal F (1999) Visual kin recognition in chimpanzees. Nature 399: 647-648.

Pascoal S, Mendrok M, Mitchell C, Wilson AJ, Hunt J, Baily NW (2016) Sexual selection and population divergence I: The influence of socially flexible cuticular hydrocarbon expression in male field crickets (Teleogryllus oceanicus) Evolution 70(1):82-97.

Peeters C, Monnin T, Malosse C (1999) Cuticular hydrocarbons correlated with reproductive status in a queenless ant. Proc R Soc Lond B 266:1323-1327.

Peso M, Richards MH (2010) Knowing who's who: nestmate recognition in the facultatively social carpenter bee, Xylocopa virginica. Animal Behaviour 79(3): 563-570.

Peso M, Elgar MA, Barron AB (2014) Pheromonal control: reconciling physiological mechanism with signaling theory. Biol Rev. 90(2):942-959.

Polerstock, AR, Eigenbrode SD, Klowden MJ (2002) Mating alters the cuticular hydrocarbons of females Anopheles gambiae sensu stricto and Aedes aegypti (Diptera: Culicidae). J Med Entomol 39:545-552.

Pradella D, Martin SJ, Dani FR (2015) Using errors by guard honeybees (Apis mellifera) to gain new insights into nestmate recognition signals. Chem Senses 40(9):649-653.

Rehan SM, Leys R, Schwarz MP (2012) A mid-Cretaceous origin of society in xylocopine bees with only two origins of true worker castes indicates severe barriers to eusociality. PLOS one 7(4): 1-8.

Rehan SM, Richards MH (2010a) Nesting biology and subsociality in Ceratina calcarata (Hymenoptera: Apidae). Can Entomol 142:65-74.

Rehan SM, Richards MH (2010b) The influence of maternal quality on brood sex allocation in the small carpenter bee, Ceratina calcarata. Ethology 116:876-887.

Rehan SM, Richards MH (2013) Reproductive aggression and nestmate recognition in a subsocial bee. Anim Behav 85:1-9.

Rehan SM, Richards MH, Adams M, Schwarz MP (2014) The cost and benefits of sociality in a facultatively social bee. Anim Behav 97:77-85.

Rehan SM, Toth AL (2015) Climbing the social ladder: molecular evolution of sociality. Trends in Ecology and Evolution 30: 426-433. 59

Richards MH, Packer L (2010) Social Behaviours in solitary bees: Interactions among individuals in Xeralictus bicuspidariae Snelling (Hymenoptera: Halictidae: Rophitinae). J Hym Res 19: 66-76.

Ruther JS, Sieben S, Schricker B (1998) Role of cuticular lipids in nestmate recognition of the European hornet Vespa crabro L. (Hymenoptera: Vespidae). Insectes Sociaux 45: 169-179.

Ruther JS, Sieben S, Schricker B (2002) Nestmate recognition in social wasps: manipulation of hydrocarbon profiles induces aggression in the European hornet. Naturwissenschaften 89:111- 114.

Saïd I, Malosse C, Durier V, Colette R (2015) Intraspecific signals inducing aggregation in Periplaneta americana (Insecta: Dictyoptera). Environmental Entomology 44(3):713-723.

Sakagami SF, Maeta Y (1977) Some presumably presocial habits of Japanese Ceratina bees, with notes on various social types in Hymenoptera. Insect Soc 24: 319–343.

Sakagami SF, Maeta Y (1984) Multifemale nests and rudimentary castes in the normally solitary bee Ceratina japonica (Hymenoptera: Xylocopinae). J Kan Entomol Soc 57:639-656.

Sakagami SF, Maeta Y (1995) Task allocation in artificially induced colonies of a basically solitary bee Ceratina (Ceratinidia) okinawana, with a comparison of sociality between Ceratina and Xylocopa (Hymenoptera, Anthophoridae: Xylocopinae). Japanese Journal of Entomology 63(1): 115-150.

Schiestl FP, Ayasse M (2000) Post-mating odor in females of the solitary bee, Andrena nigroaenea (Apoidea, Andrenidae), inhibits male mating behavior. Behav Ecol Sociobiol 48:303-307.

Searby A, Jouventin P (2003) Mother-lamb acoustic recognition in sheep: a frequency coding. Proc R Soc Lond B 270: 1765-1771.

Searby A, Jouventin P, Aubin T (2004) Acoustic recognition in macaroni penguins: an original signature system. Anim Behav 67: 615-625.

Smith AA, Hӧlldobler B, Liebig J (2009) Cuticular hydrocarbons reliably identify cheaters and allow enforcement of altruism in a social insect. Curr Biol 19:78-81.

Smith BH, Breed MD (2012) The chemical basis for nestmate recognition and mate discrimination in social insects. In: Chemical Ecology of Insects 2. Cardé RT, Bell WJ (ed).

Stern DL, Foster WA (1997) The evolution of sociality in aphids: a clone’s eye view. The evolution of social behavior in insects and arachnids. Ch. 7, pp. 150-165. The University of Cambridge Press.

60

Szathmary E, Maynard-Smith J (1995) The major evolutionary transitions. Nature 374: 227–232.

Terzo, M. (2000) Classification phylogénétique des Cératines du monde et monographie des espèces de la région ouest-palé-arctique et de l'Asie Centrale (Hymenoptera, Apoidea, Xylocopinae: Ceratina Latreille). Ph.D. dissertation, Universite de Mons-Hainaut, 293 pp.

Thomas ML, Simmons LW (2008) Sexual dimorphism in cuticular hydrocarbons of the Australian field cricket Teleogryllus oceanicus (Orthoptera: Gryllidae). J Insect Physiol 54:1081-108

Thorne BL (1997) Evolution of eusociality in termites. Annual Review of Ecology and Systematics 28: 27-54.

Van Oystaeyen A, Oliveira RC, Holman L et al. (2014) Conserved class of queen pheromones stops social insect workers from reproducing. Science 343:287-290.

Vokey JR, et al. (2004) Visual kin recognition and family resemblance in chimpanzees (Pan troglodytes). J Comp Physiol 118: 194-199. von Frisch, K. Tanzsprache und Orientierung der Bienen. Heidelberg: Springer Verlag. 566 pp.

Wcislo WT (1997) Social interactions and behavioral context in a largely solitary bee Lasioglossum (Dialictus) figueresi (Hymenoptera: Halictidae). Insectes Sociaux 44: 199-208.

Weiss I, Hofferberth J, Ruther J, Stӧkl J (2015) Varying importance of cuticular hydrocarbons and iridoids in the species-specific mate recognition pheromones of three closely related Leptopilina species. Ecol and Evol 3:1-12.

Wheeler DA (1991) The developmental basis of worker caste polymorphism in ants. The American Naturalist 138: 1218-1238.

Wilson EO (1971) The Insect Societies. Oxford University Press.

Wilson EO (1976) The Organization of Colony Defense in the Ant Pheidole dentata Mayr (Hymenoptera: Formicidae). Behav Ecol and Sociobiol 1: 63-81.

Wilson EO (2008) One giant leap: how insects achieved altruism and colonial life. BioScience 58(1): 17-25.

Wilson EO, Hӧlldobler B (2005) Eusociality: origin and consequences. Proceedings of the National Academy of Science U.S.A. 102(38): 13367-13371.

Withee JR, Rehan SM (2016) Cumulative effects of body size and social experience on aggressive behavior in a subsocial bee. Behaviour. In review.

Wyatt TD (2014). Pheromones and animal behavior: chemical signals and signatures. Cambridge University Press. 61

APPENDIX A: Chapter 2 Supplementary Materials

SUPPLEMENTARY TABLES

Table S1. References for comparative analysis of arthropod cuticular hydrocarbons (Fig. 5). Order Family Species Reference Araneae Eresidae Stegodyphus lineatus Grinsted et al. 2011 Orthoptera Gryllidae Teleogryllus oceanicus Thomas and Simmons 2008 Blattodea Blattidae Periplaneta fuliginous Saïd et al. 2005; Jackson 1970 Periplaneta japonica Jackson 1972 Periplaneta bunnea Saïd et al. 2005; Jackson 1970 Periplaneta americana Saïd et al. 2005; Saïd et al. 2015 Coleoptera Chrysomelidae Leptinotarsa decemlineata Yocum et al. 2011; Nelson et al. 2002 Callosobruchus maculatus Howard 2001; Baker and Nelson 1981 Curculionidae Hypothenemus hampei Howard and Infante 1996 Lepidoptera Culicidae Anopheles gambiae Caputo et al. 2005; Polerstock et al. 2002 Gelechiidae Sitotroga cerealellae Howard 2001 Diptera Drosophilidae Drosophila melanogaster Everaerts et al. 2010; Ferveur 1997; Antony and Jallon 1982 Hymenoptera Pteromalidae Pteromalus cerealellae Howard 2001 Eurytomaidae Eurytoma amygdali Krokos et al. 2001 Chrysididae Cephalonomia hyalinipennis Howard and Pérez-Lachaund 2002 Cephalonomia tarsalis Howard 1998 Formicidae Pogonomyrymex barbatus Wagner et al. 2001; Wagner et al. 1998; Johnson and Gibbs 2004; Wagner et al. 2000 Linepithema humile de Biseau et al. 2004; Cavill and Houghton 1973; Brophy et al. 1983 Harpegnathos saltator Liebig et al. 2000 Diacamma ceylonense Cuvillier-Hot et al. 2001 Dinoponera quadriceps Monnin et al. 1997 Vespidae Polistes metricus Toth et al. 2014; Layton et al. 1994; Espelie et al. 1990 Polistes dominulus Sledge et al. 2001; Dani et al. 2001; Bonavita- Cougourdan et al. 1991 Sphecidae Eurcerceris conata Clarke et al. 2001 Eurcerceris rubripes Clarke et al. 2001 Halictidae Lasioglossum malchurum Soro et al. 2011; Ayasse et al. 1999 Lasioglossum zephyrum Smith et al. 1985 Colletidae Colletes cunicularius Mant et al. 2005; Vereecken et al. 2007 Andrenidae Andrena nigroaenaea Schiestl et al. 2000; Schiestl and Ayasse 2000 Megachilidae Osmia lignaria Gudeot et al. 2006; Buckner et al. 2009 Apidae Ceratina calcarata Manuelia postica Flores-Prado et al. 2008 Amegilla dawsoni Simmons et al. 2003 Apis mellifera Schmitt et al. 2007; Abou-Shaara 2014; Blomquist et al. 1980 Bombus terrestris Sramkova et al. 2008; Krieger et al. 2006 Bombus hypnorum Ayasse et al. 1995 Scaptotrigona bipunctata Jungnickel et al. 2004 Schwarziana quadripunctata Nunes et al. 2009B Frieseomelitta varia Nunes et al. 2008; Nunes et al. 2009A Lestrimelitta limao Nunes et al. 2008 Melipona scutellaris Kerr et al. 2004; Pianaro et al. 2007 Melipona bicolor Abdalla et al. 2003 Melipona quadrifasciata Borge et al. 2012 Melipona asilvai Nascimento and Nascimento 2012

62

Table S2. Spearman Rank Correlation results for relative peak heights versus ovarian development (mm) in Ceratina calcarata females.

Compound Spearman rank correlation results Alkanes C15 Pentadecane S = 2916.8; p-value = 0.3032; rho = 0.201

C17 Heptadecane S = 4638.7; p-value = 0.1655; rho = -0.269

C19 Nonadecane S = 2290.9; p-value = 0.05056; rho= 0.373*

C21 Heneicosane S = 2724.7; p-value = 0.1916; rho= 0.254

C23 Tricosane S = 3084.3, p-value = 0.4282; rho = 0.155

C25 Pentacosane S = 4085.4; p-value = 0.5496; rho = -0.118

C27 Heptacosane S = 2803.7; p-value = 0.2334; rho= 0.232

Alkenes C17:1 Heptadecene S = 3044.7; p-value = 0.3964; rho= 0.166

C19:1 Nonadecene S = 2302.1; p-value = 0.05263; rho = 0.369

C23:1 Tricosene S = 3140.4; p-value = 0.4756; rho = 0.140

Ethyl Esters Et-C16 Ethyl Hexadecanoate S = 3774.6; p-value = 0.8676; rho = -0.032

Et-C16:1 Ethyl Hexadecenoate S = 3881.6; p-value = 0.7528; rho = -0.062

Et-C18:1 Ethyl Octadecenoate S = 3433.1; p-value = 0.7599; rho= 0.060

Alcohols C15-OH Farnesol S = 3336.9; p-value = 0.6606; rho= 0.086

63

Table S3. Spearman rank correlations of relative peak heights versus head width of Ceratina calcarata females. Compound Spearman rank correlation results Alkanes C15 Pentadecane S = 1838.6; p-value = 0.007156; rho = 0.496

C17 Heptadecane S = 2983.3; p-value = 0.3498; rho = 0.183

C19 Nonadecane S = 3318.1; p-value = 0.6417; rho = 0.091

C21 Heneicosane S = 1875.4; p-value = 0.02612; rho = 0.427*

C23 Tricosane S = 3581.2; p-value = 0.9199; rho = 0.019

C25 Pentacosane S = 4634.8; p-value = 0.1672; rho = -0.268

C27 Heptacosane S = 2949.6; p-value = 0.3257; rho = 0.192

Alkenes C17:1 Heptadecene S = 2784.1; p-value = 0.2225; rho = 0.238

C19:1 Nonadecene S = 3366.6; p-value = 0.6907; rho = 0.078

C23:1 Tricosene S = 3938.4; p-value = 0.6939; rho = -0.077

Ethyl Esters

Et-C16 Ethyl Hexadecanoate S = 4020.1; p-value = 0.6119; rho = -0.100

Et-C16:1 Ethyl Hexadecenoate S = 5000.2; p-value = 0.053; rho = -0.368

Et-C18:1 Ethyl Octadecenoate S = 4189; p-value = 0.4572; rho = -0.146

Alcohols C15-OH Farnesol S = 3333.7; p-value = 0.6574; rho = 0.087

64

SUPPLEMENTARY REFERENCES

Abdalla FC, Jones GR, Morgan ED, da Cruz-Landim C (2003) Comparative study of the cuticular hydrocarbon composition of Melipona bicolor Lepeletier, 1836 (Hymenoptera, Meliponini) workers and queens. Genet Mol Res 2:191-199

Abou-Shaara HF (2015) Thermal tolerance characteristics of two honey bee races. J Agr Urban Entomol 31:1-8. doi:10.3954/JAUE14-05.1

Antony C, Jallon JM (1982) The chemical basis for sex recognition in Drosophila melanogaster. J Insect Physiol 28(10): 873-880. doi:10.1016/0022-1910(82)90101-9

Ayasse M, Marlovits T, Tengö J, Taghizadeh T, Francke W (1995) Are there Pheromonal dominance signals in the bumblebee Bombus hypnorum L (Hymenoptera, Apidae)? Apidologie 26:163-180.

Ayasse M, Engels W, Lübke G, Taghizadeh T, Francke W (1999) Mating expenditures reduced via female sex pheromone modulation in the primitively eusocial Halictid bee, Lasioglossum (Evylaeus) malachurum (Hymenoptera: Halictidae). Behav Ecol Sociobiol 45:95-106

Baker JE, Nelson DR (1981) Cuticular hydrocarbons of adults of the cowpea weevil, Callosobruchus maculatus. J Chem Ecol 7(1):175-182. doi:10.1007/BF00988645

Blomquist GJ, Chu AJ, Remaley S (1980) Biosynthesis of wax in the honeybee Apis mellifera L. Insect Biochem 10:313-321

Bonavita-Cougourdan A, Theraulaz G, Bagnéres A-G, Roux M, Pratte M, Provost E, Clément J- L (1991) Cuticular hydrocarbons, social organization and ovarian development in a Polistine wasp: Polistes dominulus Christ. Comp Biochem Physiol 100b(4): 667-680. doi:10.1016/0305- 0491(91)90272-F

Borges AA, Ferreira-Caliman MJ, Nascimento FS, Campos LAO, Tavares MG (2012) Characterization of cuticular hydrocarbons of diploid and haploid males, workers and queens of the stingless bee Melipona quadrifasciata. Insect Soc 59: 479-486. doi 10.1007/s00040-012- 0242-x

Brophy JJ, Cavill GWK, Davies NW, Gilbert RD, Philip RP, Plant WD (1983) Hydrocarbon constituents of three species of dolichoderine ants. Insect Biochem 13(4):381-389. doi:10.1016/0020-1790(83)90021-5

Buckner JS, Pitts-Singer TL, Guédot C, Hagen MM, Fatland CL, Kemp WP (2009) Cuticular lipids of female solitary bees, Osmia lignaria Say and Megachile rotundata (F.) (Hymenoptera: Megachilidae). Comp Biochem Physiol, Part B 153:200-205 doi:10.1016/j.cbpb.2009.02.018

65

Caputo B, Dani FR, Horne GL et al. (2005) Identification and composition of cuticular hydrocarbons of the major Afrotropical malaria vector Anopheles gambiae s.s. (Diptera: Culicidae): analysis of sexual dimorphism and age-related changes. J Mass Spectrom 40(12):1595-1604

Cavill GWK, Houghton E (1973) Hydrocarbon constituents of the Argentine ant, Iridomyrmex humilis. Australian J Chem 26(5):1131-1135

Clarke SR, Dani FR, Jones GR, Morgan ED, Schmidt JO (2001) (Z)-3-hexenyl-3- hydroxybutanoate: a male specific compound in three North American decorator wasps Eucerceris rubripes, E. conata and E. tricolor. J Chem Ecol 27(7):1437-1447. doi: 10.1023/A:1010373427774

Cuvillier-Hot V, Cobb M, Malosse C, Peeters C (2001) Sex, age and ovarian activity affect cuticular hydrocarbons in Diacamma ceylonense, a queenless ant. J Insect Physiol 47(4-5): 485- 493. doi:10.1016/S0022-1910(00)00137-2

Dani FR, Jones GR, Destri S, Spencer SH, Turillazzi S (2001) Deciphering the recognition signature within the cuticular chemical profile of paper wasps. Anim Behav 62(1):165-171. doi:10.1006/anbe.2001.1714

De Biseau J-C, Passera L, Daloze D, Aron S (2004) Ovarian activity correlates with extreme changes in cuticular hydrocarbon profile in the highly polygynous ant, Linepithema humile. J Insect Physiol 50(7):585-593. doi:10.1016/j.jinsphys.2004.04.005

Espelie KE, Wenzel JW, Chang G (1990) Surface lipids of social wasp Polistes metricus Say and its nest and nest pedicel and their relation to nestmate recognition. J Chem Ecol 16(7): 2229- 2241

Everaerts C, Farine J-P, Cobb M, Ferveur J-F (2010) Drosophila cuticular hydrocarbons revisited: mating status alters cuticular profiles. PLoS One 5(3): e9607. doi:10.1371/journal.pone.0009607

Ferreira-Calimann MJ, Zucchi R, Nascimento FS (2012) Cuticular hydrocarbons discriminate distinct colonies of Melipona marginata (Hymenoptera: Apinae: Meliponini). Sociobiol 59:1-12

Ferveur J-F (1996) The Pheromonal role of cuticular hydrocarbons in Drosophila melanogaster. BioEssays 19(4):353-358

Flores-Prado L, Aguilera-Olivares D, Niemeyer HM (2008) Nest-mate recognition in Manuelia postica (Apidae: Xylocopinae), an eusocial trait is present in a solitary bee. Proc R Soc B 275:285-291. doi:10.1098/rspb.2007.1151

Grinsted L, Bilde T, d’Ettore P (2011) Cuticular hydrocarbons as potential kin recognition cues in a subsocial spider. Behav Ecol 22(6):1187-1194. doi:10.1093/beheco/arr105

66

Guedot C, Pitts-Singer TL, Buckner JS, Bosch J, Kemp WP (2006) Olfactory cues and nest recognition in the solitary bee Osmia lignaria. Physiol Entomol 31:110-119. doi:10.1111/j.1365- 3032.2005.00490.x

Howard RW (1998) Ontogenetic, reproductive, and nutritional effects on the cuticular hydrocarbons of the host-specific ectoparasitoid Cephalonomia tarsalis (Hymenoptera: Bethylidae). Ann Entomol Soc Am 91(1): 101-112

Howard RW (2001) Cuticular hydrocarbons of adult Pteromalus cerealellae (H menoptera: Pteromalidae) and two larval hosts, Angoumois Grain Moth (Lepidoptera: Gelechiidae) and Cowpea weevil (Coleptera: Bruchidae). Ann Entomol Soc Am 94(1):152-158. doi:10.1603/0013- 8746(2001)094[0152:CHOAPC]2.0.CO;2

Howard RW, Infante F (1996) Cuticular hydrocarbons of the host-specific ectoparasitoid Cephalonomia stephandoeris (Hymenoptera: Bethylidae) and its host the coffee berry borer (Coleoptera: Scolytidae). Ann Entomol Soc Am 89(5):700-709. doi:10.1093/aesa/89.5.700

Howard RW, Pérez-Lachaund G (2002) Cuticular hydrocarbons of the ectoparasitic wasp Cephalonomia hyalinipennis (Hymenoptera: Bethylidae) and alternative host, the stored product pest Caulophilus oryzae (Coleoptera: Curculionidae). Arch Insect Biochem Physiol 50(2):75-84. doi: 10.1002/arch.10034

Jackson, L (1970) Cuticular lipids of insects: II. Hydrocarbons of the cockroaches Periplaneta australasiae, Periplaneta brunnea and Periplaneta fuliginosa. Lipids 5(1): 38-41. doi:10.1007/BF02531092

Jackson L (1972) Cuticular lipids of insects – IV. Hydrocarbons of the cockroaches Periplaneta japonica and Periplaneta americana compared to other cockroach hydrocarbons. Comp Biochem Physiol Part B 41(2):331-336. doi:10.1016/0305-0491(72)90035-1

Johnson RA, Gibbs AG (2004) Effect of mating stage on water balance, cuticular hydrocarbons and metabolism in the desert harvester ant, Pogonomyrmex barbatus. J Insect Physiol 50:943- 953. doi:10.1016/j.jinsphys.2004.07.006

Jungnickel H, da Costa AJS, Tentschert J et al. (2004) Chemical basis for inter-colonial aggression in the stingless bee Scaptotrigona bipunctata. J Insect Physiol 50:761-766. doi:10.1016/j.jinsphys.2004.05.011

Kerr WE, Jungnickel H, David Morgan E (2004) Workers of the stingless bee Melipona scutellaris are more similar to males than to queens in their cuticular compounds. Apidologie 35:611-618. doi: 10.1051/apido:2004052

Krieger GM, Duchateau M-H, Van Doorn A, Ibarra F, Francke W, Ayasse M (2006) Identification of queen sex pheromone components of the bumblebee Bombus terrestris. J Chem Ecol 32:453-471. doi:10.1007/s10886-005-9013-8

67

Krokos FD, Konstantopoulou M, Mazomenos BE (2001) Alkadienes and alkenes, sex pheromone components of the almond seed wasp Eurytoma amygdali. J Chem Ecol 27(11):2169 – 2181.

Kross J, Schmitt T, Strohm E (2009) Low level of cuticular hydrocarbons in a parasitoid of a solitary digger wasp and its potential for concealment. Entomol Soc Jap 12(1):9-16.

Layton JM, Camann MA, Espelie KE (1994) Cuticular lipid profiles of queens, workers, and males of social wasp Polistes metricus Say are colony-specific. J Chem Ecol 20(9):2307-2321.

Liebig J, Peeters C, Oldham NJ, Markstädter C, Hӧlldobler B (2000) Are variations in cuticular hydrocarbons of queens and workers a reliable signal of fertility in the ant Harpegnathos saltator? Proc Natl Acad Sci USA 97:4124-4131.

Mant J, Brändli C, Vareecken NH, Schulz CM, Francke W, Schiestl FP (2005) Cuticular hydrocarbons as sex pheromones of the bee Colletes cunicularis and the key to its mimicry by the sexually deceptive orchid, Ophrys exaltata I. J Chem Ecol 31:1765-1878.

Monnin T, Malosse C, Peeters C (1997) Solid-phase microextraction and cuticular hydrocarbon differences related to reproductive activity in queenless ants Dinoponera quadriceps. J Chem Ecol 24(3):473-490.

Nascimento DL Nascimento FS (2012) Acceptance threshold hypothesis is supported by chemical similarity of cuticular hydrocarbons in a stingless bee, Melipona asilvai. J Chem Ecol 38(11):1432-1440.

Nelson DR, Adams TS, Fatland CL (2002) Hydrocarbons in the surface wax of eggs and adults of the Colorado potato beetle, Leptinotarsa decemlineata. Comp Biochem Physiol Part B 134(3):447-466.

Nunes TM, Nascimento FS, Turatti IC, Lopes NP, Zucchi R (2008) Nestmate recognition in a stingless bee: does the similarity of chemical cues determine guard acceptance? Anim Behav 75:1165-1171.

Nunes TM, Turatti IC, Lopes NP, Zucchi R (2009a) Chemical signals in the stingless bee, Frieseomelitta varia, indicate caste, gender, age, and reproductive status. J Chem Ecol 35:1172- 1180.

Nunes TM, Turatti IC, Mateus S, Nascimento FS, Lopes NP, Zucchi R (2009b) Cuticular hydrocarbons in the stingless bee Schwarziana quadripunctata (Hymenoptera: Apidae: Meliponini): differences between colonies, castes, and age. Genet Mol Res 8:589-595.

Pianaro A, Flach A, Petricio EFLRA, Nogueria-Neto P, Marsaioli AJ (2007) Chemical changes associated with the invasion of a Melipona scutellaris colony by Melipona rufiventris workers. J Chem Ecol 33:971:984.

68

Polerstock, AR, Eigenbrode SD, Klowden MJ (2002) Mating alters the cuticular hydrocarbons of females Anopheles gambiae sensu stricto and Aedes aegypti (Diptera: Culicidae). J Med Entomol 39:545-552.

Saïd I, Malosse C, Durier V, Colette R (2015) Intraspecific signals inducing aggregation in Periplaneta americana (Insecta: Dictyoptera). Environmental Entomology 44(3):713-723.

Saïd I, Castagliola G, Leoncini I, Colette R (2005) Cuticular hydrocarbon profiles and aggregation in four Periplaneta species (Insecta: Dictyoptera). J Insect Physiol 51(9):995-1003.

Schiestl FP, Ayasse M, Paulus HF, Löfstedt C, Hansson BS, Ibarra F, Francke W (2000) Sex pheromone mimicry in the early spider orchid (Ophrys sphegodes): patterns of hydrocarbons as the key mechanism for pollination by sexual deception. J Comp Physiol 186:567-574.

Schiestl FP, Ayasse M (2000) Post-mating odor in females of the solitary bee, Andrena nigroaenea (Apoidea, Andrenidae), inhibits male mating behavior. Behav Ecol Sociobiol 48:303-307.

Schmitt T, Herzner G, Weckerle B, Schrieier P, Strohm E (2007) Volatiles of foraging honeybees Apis mellifera (Hymenoptera: Apidae) and their potential role as semiochemicals. Apidologie 38:164-170.

Simmons LW, Alcock J, Reeder A (2003) The role of cuticular hydrocarbons in male attraction and repulsion by female Dawson’s burrowing bee, Amegilla dawsoni. Anim Behav 66:677-685.

Sledge MF, Boscaro F, Turillazzi S (2001) Cuticular hydrocarbons and reproductive status in the social wasp Polistes dominulus. Behav Ecol Sociobiol 49:401-409.

Smith BH, Carlson RG, Frazier J (1985) Identification and bioassay of macrocyclic lactone sex pheromone of the Halictine bee Lasioglossum zephyrum. J Chem Ecol 11(10):1447-1456.

Soro A, Ayasse M, Zobel MU, Paxton RJ (2011) Kin discriminators in the eusocial sweat bee Lasioglossum malachurum: the reliability of cuticular and Dufour’s gland odours. Behav Ecol Sociobiol 65:641-653. doi:10.2307/41414553

Sramkova A, Schulz C, Twele R, Francke W, Ayasse M (2008) Fertility signals in the bumblebee Bombus terrestris (Hymenoptera: Apidae). Naturwissenschaften 95:515-522.

Thomas ML, Simmons LW (2008) Sexual dimorphism in cuticular hydrocarbons of the Australian field cricket Teleogryllus oceanicus (Orthoptera: Gryllidae). J Insect Physiol 54:1081- 1089.

Toth AL, Tooker JF, Radhakrishnan S, Minard R, Henshaw MT, Grozinger CM (2014) Shared genes related to aggression, rather than chemical communication, are associated with reproductive dominance in paper wasps (Polistes metricus). BMC Genomics 15: 75.

69

Vereecken NJ, Mant J, Schiestl FP (2007) Population differentiation in female sex pheromone and male preferences in a solitary bee. Behav Ecol Sociobiol 61:811-821.

Wagner D, Brown MJF, Broun P, Cuevas W, Moses LE, Chao DL, Gordon DM (1998) Task- related differences in the cuticular hydrocarbons composition of harvester ants, Pogonomyrmex barbatus. J Chem Ecol 24(12):2021-2037.

Wagner D, Tissot M, Gordon D (2001) Task-related environment alters the cuticular hydrocarbon composition of harvester ants. J Chem Ecol 27(9):1805-1819.

Wagner D, Tissot M, Cuevas W, Gordon DM (2000) Harvester ants utilize cuticular hydrocarbons in nest mate recognition. J Chem Ecol 26: 2245-2257.

Yocum GD, Buckner JS, Fatland CL (2011) A comparison of internal and external lipids of nondiapausing and diapausing phase adult Colorado potato beetles, Leptinotarsa decemlineata. Comp Biochem Physiol Part B 159(3):163-170.

70

APPENDIX B: Chapter 3 Supplementary Materials

SUPPLEMENTARY TABLES

Supp. Table 1: Wilcoxon tests comparing the relative frequency of behaviors between PP and WW trials, and how they vary within pre-reproductive (FN), reproductive (AB), and post- reproductive (FB) C. calcarata females. PP WW Mean ± S.E. W Statistic/p-value All Females Aggression/Total 19.64 ± 1.57 20.26 ± 1.82 W = 17140 p-value = 0.4799 Avoidance/Total 18.58 ± 1.52 22.66 ± 2.01 W = 15948 p-value = 0.4038 Following/Total 8.34 ± 0.834 12.41 ± 1.26 W = 15045 p-value = 0.07475 Tolerance/Total 49.20 ± 2.29 41.22 ± 2.39 W = 19213 p-value = 0.01677* Founding Aggression/Total 21.52 ± 2.84 20.93 ± 3.62 W = 2227.5 Nest p-value = 0.1683 Avoidance/Total 21.46 ± 2.58 17.75 ± 3.14 W = 2392.5 p-value = 0.09858 Following/Total 8.87 ± 1.64 12.34 ± 2.32 W = 1906 p-value = 0.4847 Tolerance/Total 43.68 ± 3.94 43.89 ± 4.35 W = 2077 p-value = 0.8897 Active Aggression/Total 24.81 ± 2.64 20.78 ± 2.78 W = 1708.5 Brood p-value = 0.4159 Avoidance/Total 18.69 ± 2.54 20.69 ± 3.44 W = 1541 p-value = 0.7691 Following/Total 11.31 ± 1.59 14.32 ± 2.39 W = 1488.5 p-value = 0.5471 Tolerance/Total 41.70 ± 3.42 42.12 ± 4.17 W = 1627.5 p-value = 0.8432 Full Brood Aggression/Total 12.37 ± 2.43 19.14 ± 3.07 W = 1687.5 p-value = 0.1794 Avoidance/Total 15.50 ± 2.76 28.36 ± 3.83 W = 1416 p-value = 0.007627** Following/Total 4.74 ± 0.82 11.29 ± 2.05 W = 1563.5 p-value = 0.04167* Tolerance/Total 62.63 ± 4.08 39.60 ± 4.10 W = 2716.5 p-value = 0.0001576*** Note: * indicates p<0.05, ** indicates p<0.01, and *** indicates p<0.001.

71

Supp. Table 2: Paired Wilcoxon tests comparing head width, wing wear, and ovarian development between PP and WW trials. Comparisons were made overall and by nest class.

Overall FN AB FB HW (mm) W = 16242 W = 1708 W = 1530 W = 1885 p-value = 0.153 p-value = 0.2268 p-value = 0.3403 p-value = 0.03288* WW Score W = 15922 W = 1900.5 W = 1566 W = 1819.5 p-value = 0.1305 p-value = 0.8325 p-value = 0.1574 p-value = 0.3219 Ovarian W = 16798 W = 1962.5 W = 1500 W = 1719 Development (mm) p-value = 0.6308 p-value = 0.6871 p-value = 0.9785 p-value = 0.6718 72

Supp. Table 3: Paired Wilcoxon tests comparing the relative frequencies of behaviors between W and P bees in WP trials, overall, and within pre-reproductive (FN), reproductive (AB), and post-reproductive (FB) C. calcarata females.

Mean ± S.E. V Statistic/p-value All Females W P (n=82) Aggression/Total 16.35 ± 2.19 19.11 ± 2.57 V = 1320 p-value = 0.5031 Avoidance/Total 23.17 ± 2.87 19.83 ± 2.88 V = 1005 p-value = 0.05787 Following/Total 10.16 ± 1.45 9.25 ± 1.31 V = 1017 p-value = 0.7193 Tolerance/Total 45.44 ± 3.49 48.15 ± 3.93 V = 1295 p-value = 0.1468 Founding Nest Aggression/Total 16.65 ± 4.07 22.74 ± 5.30 V = 185 (n=28) p-value = 0.5539 Avoidance/Total 28.38 ± 5.11 23.91 ± 5.59 V = 108 p-value = 0.1462 Following/Total 12.58 ± 2.68 13.38 ± 2.65 V = 141 p-value = 0.6494 Tolerance/Total 38.82 ± 5.75 39.97 ± 6.37 V = 137 p-value = 0.4654 Active Brood Aggression/Total 17.40 ± 3.80 16.34 ± 3.20 V = 121 (n=24) p-value = 0.871 Avoidance/Total 21.82 ± 4.77 17.23 ± 4.85 V = 98.5 p-value = 0.2355 Following/Total 7.30 ± 2.04 7.51 ± 2.53 V = 98.5 p-value = 0.8227 Tolerance/Total 49.32 ± 5.56 58.92 ± 6.02 V = 182 p-value = 0.02179* Full Brood Aggression/Total 15.24 ± 3.61 17.94 ± 4.34 V = 141.5 (n=30) p-value = 0.6378 Avoidance/Total 19.39 ± 4.92 18.11 ± 4.54 V = 148 p-value = 0.7064 Following/Total 10.18 ± 2.59 6.79 ± 1.46 V = 115.5 p-value = 0.5033 Tolerance/Total 48.52 ± 6.49 47.16 ± 7.36 V = 129.5 p-value = 0.8077 Note: * indicates p<0.05, ** indicates p<0.01, and *** indicates p<0.001.

73

Supp. Table 4: Paired Wilcoxon tests comparing head width, wing wear, and ovarian development between W and P bees in WP trials. Comparisons were made overall and by nest class.

Overall FN AB FB HW (mm) V = 1051.5 V = 137 V = 114.5 V = 103.5 p-value = 0.3524 p-value = 0.7209 p-value = 0.7088 p-value = 0.3008 WW Score V = 666.5 V = 56.5 V = 40.5 V = 124 p-value = 0.1652 p-value = 0.8631 p-value = 0.7488 p-value = 0.09424 Ovarian V = 1701 V = 201.5 V = 211 V = 159 Development (mm) p-value = 0.1457 p-value = 0.5172 p-value = 0.08392 p-value = 0.9357

Supp. Table 5: Kruskal-wallis tests comparing the absolute differences in behavior between dyads in PP, WP, and WW trials. All Females FN AB FB Aggression chi-squared = 0.15277 chi-squared = 5.3257 chi-squared = 0.46123 chi-squared = 3.0562 p-value = 0.9265 p-value = 0.06975 p-value = 0.794 p-value = 0.2169

Avoidance chi-squared = 3.8284 chi-squared = 5.6098 chi-squared = 6.0052 chi-squared = 2.2055 p-value = 0.1475 p-value = 0.06051 p-value = 0.04966* p-value = 0.332

Following chi-squared = 2.2267 chi-squared = 1.9031 chi-squared = 1.6206 chi-squared = 3.0718 p-value = 0.3284 p-value = 0.3861 p-value = 0.4447 p-value = 0.2153

Tolerance chi-squared = 4.0495 chi-squared = 0.015094 chi-squared = 1.1689 chi-squared = 9.5206 p-value = 0.132 p-value = 0.9925 p-value = 0.5574 p-value = 0.008563** Note: df=2 for all tests, * indicates p<0.05, ** indicates p<0.01.

74

Supp. Table 6: Spearman Rank Correlation statistics for the differences in morphological measurements and absolute frequencies of behaviors between both members of a dyad in PP and WW trial types. PP Trials Diff HW (mm) Diff WW Score Diff Ovarian Development (mm) Diff Aggression S = 78510 S = 53537 S = 64924 p-value = 0.3184 p-value = 0.3958 p-value = 0.5143 rho = -0.1167851 rho = 0.1023255 rho = 0.07647097 Diff Avoidance S = 83166 S = 62610 S = 79000 p-value = 0.116 p-value = 0.68 p-value = 0.2901 rho = -0.1830177 rho = -0.04979784 rho = -0.1237569 Diff Following S = 66794 S = 62900 S = 52004 p-value = 0.6709 p-value = 0.6507 p-value = 0.02413 rho = 0.04986979 rho = -0.05466794 rho = 0.2602532 Diff Tolerance S = 66135 S = 69611 S = 62028 p-value = 0.6136 p-value = 0.1635 p-value = 0.3147 rho = 0.059245 rho = -0.1671812 rho = 0.1176715 Diff Head Width X S = 55012 S = 59967 (mm) p-value = 0.52 p-value = 0.2082 rho = 0.07760716 rho = 0.1469862 Diff Wing Wear X X S = 54168 p-value = 0.4466 rho = 0.09175821 Diff Ovarian X X X Development (mm)

WW Trials Diff Aggression S = 83005 S = 64274 S = 60721 p-value = 0.1207 p-value = 0.6837 p-value = 0.2437 rho= -0.1807264 rho = 0.04815079 rho = 0.1362596 Diff Avoidance S = 61458 S = 79256 S = 63183 p-value = 0.2823 p-value = 0.1388 p-value = 0.3875 rho = 0.1257703 rho = -0.1737224 rho = 0.101237 Diff Following S = 57432 S = 62587 S = 70486 p-value = 0.116 p-value = 0.5358 p-value = 0.982 rho = 0.1830496 rho = 0.07312241 rho = -0.002642734 Diff Tolerance S = 65169 S = 56933 S = 62221 p-value = 0.5338 p-value = 0.182 p-value = 0.3262 rho = 0.07298069 rho = 0.1568643 rho = 0.114926 Diff Head Width X S = 64009 S = 67787 (mm) p-value = 0.6595 p-value = 0.7608 rho = 0.05206405 rho = 0.03574453 Diff Wing Wear X X S = 58605 p-value = 0.2619 rho = 0.1321015 Diff Ovarian X X X Development (mm)