1 ENVIRONMENTAL EFFECTS ON SEXUAL SELECTION IN A WILD INSECT POPULATION OF LEPTOSCELIS TRICOLOR (HEMIPTERA: COREIDAE) By UMMAT SOMJEE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLM ENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2014
2 Â© 2014 Ummat Somjee
3 To my family
4 ACKNOWLEDGMENTS First, I would like to thank my family. I am forever grateful to my mother for her uncompromising support, my father for his encouragement and optimism , my brother for showing me en thusiasm and dedication , and mostly to my sister for courage and inspiration. I am deeply indebted to m y advisor, Christine Miller who p layed an instrumental role in guiding me throug h the rigors of graduate school and the highs and lows of experimental design and fieldwork . This work would not have been mad e possible without her infectious enthusiasm , encouragement and commitment . I have had the privilege to work with some dedicated individuals. Lourdes Hernandez, Claudio Monteza , Pablo Allen, Katherine Holmes, Angie Estrada , Santiago Meneses & Peter Marting all provided invaluable assistance in fieldwork and data collection. I would also like to thank John H. Christy, William Wcislo & Paula A. Trillo for stimulating discussions and suggestions. I would like to thank staff at STRI for their logistical and administrative support throughout the years, and especially for their emergency respo nse when I required it. I would also like to thank Dr. Allen J. Moore and Dr. Daniel A. Hahn for providing critical comments and suggestions. This work was supported by the Smithsonian Tropical Research Institute (STRI) , Panama Museum Research Grant to US , and National Science Foundation, Grant IOS 0926855 to CWM.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ... 4 ! LIST OF TABLES ................................ ................................ ................................ ............. 6 ! LIST OF FIGURES ................................ ................................ ................................ ........... 7 ! ABSTRACT ................................ ................................ ................................ ...................... 8 CHAPTER 1 NO EVIDENCE OF STRONG SEXUAL SELECTION ACROSS ENVIRONMENT S IN THE HELICONIA BUG, LEPTOSCELIS TRICOLOR (HEMIPTERA: COREIDAE ) ................................ ................................ ..................... 10 ! Introduction ................................ ................................ ................................ .............. 10 ! Methods ................................ ................................ ................................ ................... 12 ! Study S ystem ................................ ................................ ................................ .... 12 ! Data C ollection ................................ ................................ ................................ .. 13 ! Capture, Measurement and R ationale ................................ ............................... 13 ! Data A nalysis ................................ ................................ ................................ ..... 15 ! Results ................................ ................................ ................................ ..................... 16 ! Discussion ................................ ................................ ................................ ................ 16 ! 2 NATURAL ENVIRONMENTAL VARIAT ION CAUSES A REVERSAL IN EXPRESSION OF PRE AND POST COPULATORY SEXUAL TRAITS IN THE HELICONIA BUG, LEPTOSCELIS TRICOLOR (HEMIPTERA: COREIDAE) ......... 28 ! Introduction ................................ ................................ ................................ .............. 28 ! Methods ................................ ................................ ................................ ................... 30 ! Study Organism ................................ ................................ ................................ . 30 ! Insect R earing ................................ ................................ ................................ ... 32 ! Weapon Measurements & Testes W eights ................................ ....................... 32 ! Results ................................ ................................ ................................ ..................... 32 ! Discussion ................................ ................................ ................................ ................ 33 ! LIST OF REFERENCES ................................ ................................ ................................ 38 ! BIOGRAPHICAL SKETCH ................................ ................................ ............................. 44 !
6 LIST OF TABLES Table page 1 1 Male linear selection Ã”Early' and Ã”Late' season. ................................ ................... 26 ! 1 2 Male quadratic selection Ã”Early' and Ã”Late' season. ................................ ............. 26 ! 1 3 Female linear selection Ã”Early' and Ã”Late' season. ................................ ............... 26 ! 1 4 Female quadratic selection Ã”Early' and Ã”Late' season. ................................ ......... 26 ! 1 5 Male linear selection on H. mariae vs H. latispatha . ................................ ............ 27 ! 1 6 Male quadratic selection on H. mariae vs H. latispatha . ................................ ...... 27 ! 1 7 Female linear selection on H. mariae vs H. latispatha. ................................ ....... 27 ! 1 8 Female quadratic selection on H. mariae vs H. latispatha. ................................ . 27 !
7 LIST OF FIGURES Figure page 1 1 The number of insects found on each species of heliconia inflorescence. .......... 22 ! 1 2 Average size of male hind femurs Ã”Early' and Ã”Late' in the season.. ................... 23 ! 1 3 Patterns of selection in Leptoscelis tricolor over the season !!!!!!!! ... 24 ! 1 4 Patterns of selection in Leptoscelis tricolor across host plants. .......................... 25 ! 2 1 Variation in male hind femur morphology for Leptoscelis tricolor . ....................... 36 ! 2 2 Differential expression in weapons and testes depending on reari ng environment in Leptoscelis tricolor. ................................ ................................ ..... 37 !
8 A bstract of Thesis Presented to t he Graduate School of the University O f Florida i n Partial Fulfillment of the Requirements for the Degree o f Master o f Science EN VIRONMENTAL EFFECTS ON SEXUAL SELECTION IN A WILD INSECT POPULATION OF LEPTOSCELIS TRICOLOR (HEMIPTERA: COREIDAE ) By Ummat Somjee August 2014 Chair: Christine Miller Major: Entomology and Nematology Sexual selection is a potent evolutionary force drive n by competition for reproducti ve opportunities . Studies of sexual selection have revealed that the strength, form and even direction of selection can change across years in response to annual environmental variation. In many cases environments can vary su bstantially even within a breeding season, yet little is known about changing sexual selection patterns at these s horter time scales. I use the heliconia bug, Leptoscelis tricolor (Hemiptera: Coreidae), to examine the role of discrete environments on patte rns of sexual selection within a single breeding season. To examine the role of different environments on patterns of sexual selection in the wild I consider two perspectives. First I measure and describe sexual selection across environmenta l contexts; sec ond I explore the role of developmental environment on the expression of sexually selected traits. For my first chapter I measured sexual selection in a wild focal population of L. tricolor, over time and as environments change. Within my focal population , the
9 environment and demographic context of selection change consi derably over space and time. My findings are suggestive of weak patterns of selection over time as environments change. For my second chapter I examine expression of weapons and testes in L . tricolor across two common yet discrete environments. I find that L. tricolor that develop on one environment have increased expression of weapons and reduced expression of testes, while development on the second environment resulted in the opposite patt ern.
10 CHAPTER 1 NO EVIDENCE OF STRONG SEXUAL SELECTION ACROSS ENVIRONMENTS IN THE HELICONIA BUG, LEPTOSCELIS TRICOLOR (HEMIPTERA: COREIDAE ) Introduction A central goal of evolutionary biology is to develop an understanding of selection, the primary mech anism of adaptive evolution in nature. A review of selection gradients spanning over a decade has revealed that selection is generally strong in nature, particularly for traits involved in sexual selection (Kingsolver et al. , 2001; Hereford, Hansen, & Houle, 2004) . Multiple and diverse lines of evidence suggest sexual selection is stronger that natural selection , indicating that evolutionary change by means of sexual selection should be occurring at a rapid rate ( discussed in Svensson and Gosden 2007). Many studies have now revealed rapid evolutionary change by means of natural selection ( reviewed in Hendry and K innison 1999, Kinnison & Hendry 2001). Yet, examples of rapid evolution by means of sexual selection are relatively scarce (Svensson and Gosden 2007). How can strong sexual selection gradients reported in natu re be resolved with the little evidence of rapi d evolutionary change by sexual selection? One hypothesis to explain this apparent paradox is that selection fluctuates, or varies in form, over space and time (Bell, 2010) . Fluctuating selection caus ed by changing environments over time may maintain or slow the erosion of genetic variation. Most studies of wild populations measure sexual selection as it occurs in a single environmental context at a particular time ( reviewed in Kingsolver et al. , 2001, 2012) . Such "snapshot" studies can be problematic because an implicit assumption is often made that the patterns detected are consistent and representative across environmental contexts.
11 Existing temporally replicated studies demonstrate that annual patterns of selection can vary drastically depending on demographic and environmental conditions at the time (Reznick et al. , 1997; Grant & Grant, 2002; Hamon et al. , 2005) . However, the majority of temporally replicated studies of sexual selection are conducted with organisms where breeding occurs synchronously during a relatively brief period of the year (Kingsolv er et al. , 2001) . For many organisms, breeding and sexual selection occur throughout large portions of the year, and environmental variation occurring within a season can yield changes in demography (Steele, Siepielski, & McPeek, 2011) , meta population dynamics (Kasumovic et al. , 2008) and the relative importance of different mechanisms of sexual selection (Hunt et al. , 2009) . Despite the potential of within season effects to change selection patterns, surprisingly few studies have examined within season changes in selecti on (Preziosi & Fairbairn, 2000; Kasumovic et al. , 2008; Kasumovic & Andrade, 2009; Pun zalan, Rodd, & Rowe, 2010; Steele et al. , 2011) . We examine patterns of sexual selection in the wild as environments change in a sexually dimorphic insect. Leptoscelis tricolor (Hemiptera: Coreidae) feed and mate on the inflorescences of heliconia host p lants. Males possess enlarged hind femurs, used as weapons to defend territories on discrete host plant inflorescences and females visit select male territories to mate, fee d, and lay eggs. Males perform pre copulatory courtship before attempting to mate a nd a female may or may not mate with a male present on a territory. Thus both male male competition and female mate choice occur on these discrete territories. As the wet season commences different species of h eliconia bloom at different times, each speci es differs in size, structure and quality (Miller & Emlen 2008). Because the competitive arena on which sexual selection occurs
12 chang es depending on the species of h eliconia inflorescence, L. tricolor provides an excellent opportunity to examine the effect s of changing environmental conditions on patterns of selection. I investigate the role discrete host plant species play in sexual selection in L. tricolor. I predict that as the species and quality of host resources change over time so too will the streng th and form of selection. The ability to defend critical resources that females need, has long been implicated in the evolution of specialized sexually selected weapons in males (Emlen & Oring, 1977; Emlen, 2008) . Specifical ly I predict on smaller more easily defended host plant species, there will be stronger selection for larger male body size and larger weapons . Methods Study S ystem Leptoscelis tricolor (Hemiptera: Coreidae) is a sexually dimorphic insect with overlapping generations found in Panama and Costa Rica. The life history of these insects is closely tied to the inflorescences of heliconia (Zinzerberales: Heliconiae) host plants. These insects feed on phloem, nectar, and fruit of h eliconia inflorescences. Male L. t ricolor defend i nflorescences as territories. Males enlarged hind femurs are used in competition with other males ; these sexually selected weapons are a characteristic feature of th e family of insects Coreidae (Miyatake, 1997, 2 002; Eberhard, 1998; Emlen, 2008) . When males encounter each other on an inflorescence they participate in aggressive displays or combat, using their hind femurs, and often the unsuccessful male hides or leaves the territory. Females fly among infloresce nces and are courted by males, but may or may not mate with a male present on an inflorescence (Miller, 2007; Miller & Emlen, 2010a,b) . Mating occurs on heliconia inflorescences and often lasts several hours, and females often fee d while mating. Courtship and mating occur on the
13 exposed surface of the host inflorescence and successful copulation is distinctly recognized as insects are facing in opposite directions and attached at the tips of the abdomen (Miller, 2007, 2008) . Further, non mating males are easily visible while defending the surface of the inflorescence. Thus, bot h mating and non mating individuals can be identified on their respective territories in natural conditions. Data C ollection I collected data on sexual selection from early May the to early October 2012 during the wet season in and around Gamboa, Panama. A ll data collection occurred within an overall area of approximately 25 km 2 . I established sites by visually identifying heliconia inflorescences and I surveyed these sites for insects every 2 4 days, or until inflorescences senesced and insects were no lon ger present at the site. Because insects move towards areas as new inflorescences were produced, new sites were continually established throughout the season. This sampling methodology allowed me to collect insects from newly blooming inflorescences of hos t species, thus reflecting the naturally changing species and abundances of host inflorescences available Capture, M easurement and R ationale I visually scanned inflorescences and identified mating or non mating insects. I caught insects by hand or with an insect net and I processed them in the following manner. I measured pronotum size as a proxy for body size, as this sclerotized region is a commonly used metric for insect size. I also measured the size of each hind femur at its widest part. I measured bot h these morphological features with a precision of 0.01mm using digital calipers (Miller & Emlen, 2010a) . The species of heliconia inflorescence where insects wher e caught was also recorded. Insects were individually marked on the pronotum with non toxic paint and a marker was used to write a unique
14 identification number on this paint ( Miller 2007) . All insects were marked, measured and released onto the same inflorescence within 5 minutes of capture. All insects were collected from defended territories (inflorescences) throughout the sea son. To estimate selection I used a sub sample of all insects caught, when a mating pair was found in a site, all mating and non mating individuals within a site were collected marked, measured and released. A site was specified as a 10 meter radius around the mating pair, as I observed that these insects rarely travel more than 10 meters in a single flight. Individuals found mating received a mating value of (1), while those that never mated received a mating value of (0). I ndividual insects within the sam e site are likely in direct competition for mating opportunities. Additionally, 19.6 % of insects where found with at least one other insect of the same sex on the same inflorescence, suggesting that competition for mating opportunities occurs . The number of insects collected from each spec ies of host plant inflorescence differed early in the season from later in the season due to the availability of different species of i nflorescence in the study sites. I thus considered two scales by which to measure sel ection. First, I considered a temporal scale in which I measured selection early and late in the sampling season. The second scale was environmental, where I measured selection throughout the season occurring on the two most common host inflorescences . H. mariae and H. latispatha , t he two host resources examined in this study differ in the timing of production of inflorescences, and also likely differ in the resources that they provide insects. L. tricolor raised on different species of heliconia show signi ficant differences in body and weapon size, showing that host plant environment during rearing can affect the expression of these traits (Miller & Emlen,
15 2010b) . Furthermore, l arger males with larger hind femurs have been found on H. latispatha than on H. mariae (Miller, 2007) . Data A nalysis To allow for a qualitative comparison of the strength and form of s election over time and across host plant species , I performed both parametric and non parametric analysis. To estimate the strength of selection on body size (pronotum width) and sexual trait size (hind femur width), I used simple linear regression analyse s to generate linear selection gradients (Lande and Arnold 1983). Fitness was measured as a binary score (mating = 1) and (not mating = 0) . Both pronotum width and hind femur width were z transformed to a mean = 0 and a standard deviation = 1 for all analy sis. To describe nonlinear selection acting on individual traits, I also calculated quadratic coefficients using linear and squared terms in the model (Brodie, Moore, & Janzen, 1995) , and doubled these coefficients to a chieve quadratic gradients (Stinchcombe et al. , 2008) . The use of parametric models have inherent limitations when describing selection, as these models may not reveal changes in the form of selection a nd important information may be missed. To reveal more complex selection patterns I used non parametric methods to visualize selection (Schluter 1988, Brodie et al. 1995). These analyses were performed by creating cubic splines for selection acting on sing le traits. All analys e s w ere performed using statistics the software R (Version 0.97.318 Â© 2009 2012 Rstudio, Inc.).
16 Results Over a 6 month period I caught marked and measured a total of 1265 individual insects, 626 males and 629 females. I found tha t t he number of insects observed on a heliconia inflorescence at any particular time differed significantly depending on the species of inflorescence where insects were caught. Specifically, species of host plant with larger inflorescences generally held m ore insects (fig 1 1). I collected a total of 140 mating insects and 168 non mating within the same site and time period as a mating pair. I used these data to examine phenotypic distributions and patterns of selection. I found th at phenotypic distribution s differed with season for both males and females , with both males and females being smaller on average later in the season (fig 1 2). Non parametric visualizations of selec tion patterns suggest considerable variation exists in the direction and form of se lection acting on both males and females over time (fig 1 3) and across host plant species (fi g 1 4 ). F or a majority of these estimates I lack the statistical power to explicitly detect where there are differences in these selection patterns ( Table s 1 1 to 1 8 ). I did, however, detect significant disruptive selection on female body and leg size late in the season and on the host plant H. latispatha ( Table s 1 4, 1 7 & 1 8 ). Discussion Leptoscelis tricolor experiences substantial ecological and demographic ch anges within a single breeding season. Host plants change in quality and abundance (Miller & Emlen 2010), the numbers of insects found on single host plant inflorescences differ significantly across host plant species, and the phenotypic distributions of i nteracting males and females differ over time. I predicted that the changing environment of sexual
17 selection would result in strong differences in selection. However, I found no evidence for strong selection on body size or weapon size in either males or f emales. My results suggest two possibilities. The first is that there is actually no sexual selection acting during the time period studied. Indeed, selection is not always expected to occur, and strong sexual selection may only occur during specific times , or under particular conditions. Furthermore, selection inferred from the past may not be occurring in present populations. For example, male ornaments in the male jungle fowl appear to no longer have a large effect on mating success, suggesting that thes e traits have lost the functional importance to sexual selection they may have held in the past (Zuk et al. , 1990) . Additionally, for traits that are close to an optimum trait value it may be especially difficult to detect sele ction (Haller & Hendry, 2014) . The presence of male male competition and mate cho ice in L. tricolor suggests selection could be strong under the right circumstances. Although the large variation in sexual traits in L. tricolor suggest s that traits are not optimized and that selection if it is occurring, should be detec table . A more lik ely possibility is that selection is weak and fluctuates slightly across the ecological and social environments studied here . One reason there may be weak selection on male weapon size is that territories that males establish could differ in defensibility. The ability to defend critical resources has long been implicated in the evolution of specialized weapons of sexual selection (Emlen & Oring, 1977; Emlen, 2008) . A similar mating system to L. tricolor is found in the bamboo bug, Notobitus meleagris , where males defend bamboo shoots, a resource for female feed ing . In bamboo bugs , larger males were able to defend higher numbers of females on a bamboo shoot, however, as the size of the shoot increased its defensibility decrease d,
18 allowing rival males to invade these territories and mate with females (Miyatake, 2002) . For L . tricolor , each species of host plant inflorescence differs in size and likely provides differences in defensibility. Consistent with this hypothesis I found that species with larger host plant inflorescences held more insects , and specifically more males . Thus, investment in enlarged sexual traits may only be Ã”optimal' under very specific contexts where territories can be economically defended, conditions unlikely to be consistently encountered by individuals in the L. tricolor population over time. In L. tricolor , larger insects were found at the beginning of the wet season, while smaller insects were found near the end of the wet season for individuals competing for mates (fig 1 2). The phenotypic distribution thus changes considerably, as host plant spe cies availability changes throughout the season (Miller & Emlen, 2010a,b) . This change in the variance and mean of a trait distr ibution over space and time is a common feature of many natural systems, and such changes in demography may have large effects on the form and direction of selection measured (Kasumovic et al. , 2008; Steele et al. , 2011) . However, in many organisms population level estimates of demographic parameters may not be representative of what is occurring at a local scale, the scale in which ind ividuals actua lly interact ( e.g. Andrade, 2003) . In addition, population level estimates of selection averaged across changi ng environmental conditions, may confound our ability to tie changing ecological or demographic conditions to changes in selection (Kasumovic et al. , 2008) . In this study, I measured both population level phenotypic distributions and the phenotypic distributions of individuals in direct competition for mating opportuni ties. I found significant changes in phenotypes when I measured selection at an ecologically relevant scale (mating and
19 non mating individuals), suggesting that the social competitive context changes over time. Measuring selection within breeding seasons i s particularly important for organisms whose lifespan is short relative to their breeding season (Kasumovic et al. , 2008) . As seen in L. tricolor , the environmental context may change dramatically from one reproductive event to another. Indeed, measuring selection at ecologically relevant scales can yield unpredicted re sults as sexual selection may not operate consistently as conditions change (BussiÂre et al. , 2008) . These and other findings highli ght the importance of using the appropriate scale to measure selection relevant to the ecology of the organism (Carroll et al. , 2007; Cornwallis & Uller, 2010; Schoener, 2011) . Given the changes in the environmental context o f sexual selection, the number of insects competing on a host inflorescence and differences in phenotypic distributions, one would expect selection to be strong, and to vary wildly over space and time. Yet I do not detect strong selection. My result of wea k and dynamic selection may be indicative of a broader reality of selection in diverse and complex environments. Indeed, it may be precisely because of fluctuating environments that selection does have the opportunity to act consistently and incisively to produce and maintain optimal phenotypes (Bell 2010). T he result of selection not significantly different from zero is not unusual. In a review of 993 linear selection estimates from natural populations , Kingsolver (2001) found that approximately 75% of pub lished estimates of selection did not differ significantly from zero at a 95% significance level. In fact, this is likely a conservative estimate, as studies that report strong and significant selection are probably more likely to be published (Kingsolver & Pfennig, 2007) . Indeed, even studies
20 with large sample sizes may not be able to detect weak, yet biologically relevant patterns of selection (Hersch & Phillips, 2004) . My data are suggestive of weak selection patterns. For most organisms ecological variables may differ considerably over the course of reproducti ve season . Fluctuating e nvironments are especially relevant for L. tricolor which have multiple reproductive events within a breeding season, and where competition for reproduction may occur in very different environments. Changing selection pressures within a season may limit t he cumulative strength of selection on sexual traits in these insects, thus reducing the loss of additive genetic variance in sexual traits. Changes in selection pressures may also result in developmental plasticity where individuals respond to local envir onmental cues during development and invest differentially in particular sexual traits. To truly draw the link between ecological factors and selection patterns over time, further studies are needed to examine role phenotypic plasticity in context of the e nvironment to determine the effects of selection (Badyaev & QvarnstrÂšm, 2002; Preston et al. , 20 03; Steele et al. , 2011) . I explore the development of sexual traits as environments vary in the following chapter. In conclusion, temporal fluctuations in the size distribution of individual phenotypes, as well as changes in the availability of host plan ts provide a complex mating environment where one single trait, or a single combination of traits, may not provide a consistent advantage in changing competitive contexts. This dynamic environmental, demographic and social context of selection may cumulati vely lead to weak selection patterns on specialized sexual traits. The results of this study motivate
21 further studies that examine the potential role of discretely changing environments in driving selection patterns in nature.
22 Fig ure 1 1. The number of insects found on each species of heliconia inflorescence. The number of Leptoscelis tricolo r present on a single heliconia inflores c ence correlates positively with the average size of the inflorescence. Images of heliconia infloresce nces are presented in relative scale and depict the average size of an inflorescence within the population. Species of infloresc ence that are on average larger hold more insects .
23 Figure 1 2 . Average size of male hind femurs Ã”Early' and Ã”Late' i n the season . Phenotypic distributions of insects in areas and during times when insects were mating change over time. The graph shows male insects collected earlier in the year have larger hind femurs than those collected late r in the year. The trend of l arger traits earlier in the season and smaller traits later in the season is consistent for pronotum size and is consistent in both males and females .
24 Figure 1 3. Patterns of selection in Leptoscelis tricolor over the season . A) Selection on m ale pronotum and femur width across seasons. B) Selection on female pronotum and femur width across seasons.
25 Figure 1 4. Patterns of selection in Leptoscelis tricolor across host plants . A) Selection on male pronotum and femur width across host plants. B) Selection on female pronotum and femur width across host plants. Figures 1 3 & 1 4. Cubic splines (sensu Schluter 1988) allow for the non parametric visualization of selection acting on single traits on L. tricolor . Note the highly variable form and dir ection of these splines over time and across host plan t species .
26 Table 1 1. Male linea r selection Ã”Early' and Ã”Late' s eason . Season: " total N mating trait # Â± SE (lin.) p # Â± SE (Cor.) p Early 44 19 Pronotum 0.033 Â± 0.11 0.77 0.385 Â± 0.26 0 0.146 Femur 0.041 Â± 0.11 0.72 0.390 Â± 0.26 0 0.142 Late 75 36 Pronotum 0.196 Â± 0.14 0.18 0.352 Â± 0. 286 0.224 Femur 0.123 Â± 0.15 0.40 0.180 Â± 0.286 0.530 Table 1 2. Male quadrati c selection Ã”Early' and Ã”Late' s eason . Season : " total N mating trait $ Â± SE p Early 83 41 Pronotum 0 0.152 Â± 0.085 0.370 Femur 0 .022 Â± 0.093 0.904 L ate 63 27 Pronotum 0.018 Â± 0.110 0.934 Femur 0.271 Â± 0.112 0.234 Table 1 3 . Female linea r selection Ã”Early' and Ã”Late' s eason . Table 1 4. Female quadrati c selection Ã”Early' and Ã”Late' s eason . Season : " total N mating trait $ Â± SE p Early 88 43 Pronotum 0 .279 Â± 0.079 0. 082 Femur 0.127 Â± 0.061 0.297 Late 68 27 Pronotum 0.505 Â± 0.108 0.023* Femur 0.314 Â± 0.101 0.124 Table 1 1 to 1 4 . Selection gradients for males and fem ales during Ã”Early' and Ã”L ate' season time periods in L. tricolor . The table s depict the linear selection gradient (Â± SE), calculated from single linear regression of pro notum width and femur width. Correlational selection gradients, which take into accou nt the relative selection on each trait are also presented. The standard error (SE) and p value for linear coefficients were determined from logistic regression. Season: " total N mating trait # Â± SE (lin.) p # Â± SE (Cor.) p Early 88 43 Pronotum 0.093 Â± 0.110 0.404 0.206 Â± 0.141 0.148 Femur 0.053 Â± 0.110 0.634 0.181 Â± 0.141 0.202 Late 68 27 Pronotum 0.155 Â± 0.149 0.303 0.41 Â± 0.226 0.071 Femur 0.031 Â± 0.151 0.837 0.34 Â± 0.226 0.132
27 Table 1 5 . Male linear selection on H. mariae vs H. latispatha . Host : " total N mating trait # Â± SE (lin.) p # Â± SE (Cor.) p H. mariae 44 19 Pronotum 0.013 Â± 0.179 0.943 0.283 Â± 0.430 0.515 Femur 0.040 Â± 0.179 0.822 0.297 Â± 0.430 0.494 H. latispatha 75 36 Pronotum 0 .080 Â± 0.059 0.515 0.064 Â± 0.130 0.812 Femur 0.075 Â± 0.059 0.543 0.018 Â± 0.130 0.948 Table 1 6. Male quadrati c selection on H. mariae vs H. latispatha . Host : " total N mating trait $ Â± SE p H. mariae 44 19 Pr onotum D 0.136 Â± 0.129 0.869 Femur 0.004 Â± 0.150 0.988 H. latispatha 75 36 Pronotum 0.061 Â± 0.099 0.772 Femur 0.11 Â± 0.098 0.713 Table 1 7 . Female linear selection on H. mariae vs H. latispatha . Host : " t otal N mating trait # Â± SE (lin.) p # Â± SE (Cor.) p H. mariae 50 18 Pronotum 0.154 Â± 0.193 0.429 0.035 Â± 0.238 0.884 Femur 0.298 Â± 0.189 0.122 0.319 Â± 0.238 0.186 H. latispatha 84 38 Pronotum 0.206 Â± 0. 12 0.089 0.528 Â± 0.1 87 0.006** Femur 0.000 Â± 0.122 0.996 0.412 Â± 0.187 0.031* Table 1 8. Female quadratic selection on H. mariae vs H. latispatha . Host : " total N mating trait $ Â± SE p H. mariae 50 18 Pronotum 0.384 Â± 0.150 0.205 bvc Femur 0.136 Â± 0.115 0.558 H. latispatha 84 38 Pronotum 0.499 Â± 0.086 0.005** Femur 0.326 Â± 0.077 0.038* Tables 1 5 to 1 8 : Selection gradients for males (a) and females (b) occurring on H. mariae and H. latispat ha host plants in L. tricolor . The Tables depicts the linear selection gradient (Â± SE), calculated from single linear regression of pro notum width and femur width. Correlational selection gradients, which take into account the relative selection on each t rait are also presented. The standard error (SE) and p value for linear coefficients were determined from logistic regression .
28 CHAPTER 2 NATURAL ENVIRONMENTAL VARIATION CAUSES A REVERSAL IN EXPRESSION OF PRE AND POST COPULATORY SEXUAL TRAITS IN TH E HELICONIA BUG, LEPTOSCELIS TRICOLOR (HEMIPTERA: COREIDAE) Introduction Sexual selection has led to striking examples of exaggerated morphologies and intense competitive behavior. It is no surprise that early studies of sexual selection focused primarily on dramatic traits displayed during pre copulatory male contests, with the underlying implication that copulation inevitably leads to fertilization (Darwin, 1871; Huxley, 1932) . It is now clear that copulation does not nec essarily lead to fertilization, and competition among males can continue after mating in the form of sperm competition (Parker, 1970) . Indeed, in many animal systems, traits that increase mating success and traits that increase fertilization success can both be important for overall reproductive success (Preston et al. , 2003; Trillo, 2008; Eberhard, 2009; Pischedda & Rice, 2012; Puniamoorthy, Blanckenhorn, & SchÂŠfer, 2012; Rahman, Kelley, & Evans, 2013) . Competition for access to receptive females is credited with the evolution o f enlarged weapons in males of many species (Darwin, 1871) . Th ese exaggerated structures are used in combat or as signals to other males, and individuals bearing larger weapons often outcompete those with smaller weapons , a clear form of pre copulatory sexual selection (Emlen, 2008) . However, males may also improve their fertilization success by investing relatively more in post copulatory sperm competition (Parker, 1970, 1998) . Theory predicts that increased allocation towards testes should occur in systems where sperm competition is common (Parker, 1998) . Following along these lines, empirical evidence across taxa has shown that investment in testes
29 correlates positively with levels of sperm competition (Gage, 1994; Harcourt, Purvist, & Liles, 1995; Byrne, Roberts, & Simmons, 2002) . Although expression of larger weapons and testes can be advantageous, the developmen t of these traits and their resulting size is often dependent on environmental factors. In fact, male sexually selected traits are among the most phenotypically plastic and environmentally sensitive traits (Griffith, Owens, & Burke, 1999; QvarnstrÂšm, 1999; Mocz ek & Emlen, 1999) . Many studies have shown that the expressio n of large secondary sexual characters comes at a cost to other life history traits such as immunity or dispersal ability (Hosken, 2001; Robinson et al. , 2006; Lewis, Price, & Wedell, 2008; Nakayama & Miyatake, 2010; Yamane et al. , 2010) . However, only a few studies have examined the e ffect of the environment on the relative investment in both male pre and post copulatory traits (Simmons & Emlen, 2006; Evans, 2010; Pischedda & Rice, 2012; Devigili et al. , 2012; Rahman et al. , 2013) . These existing studies were conducted in laboratory settings . My goal in this study was to examine the expression of weapons and testes as they occur in two distinct natur al environments. According to life history theory, differential allocation to fitness related traits should be more likely to occur in environments where resources are limited. Under these conditions, individuals cannot acquire sufficient resources to maxi mize multiple traits (Reznick, Nunney, & Tessier, 2000) . Similarly, predicti ons from sperm competition theory suggest that investment in somatic pre copulatory traits should come a t a cost to investment in traits involved in sperm competition (Parker 1998). Here I raise insects in two distinc t natural environments that differ in the resources that they provide. I test the hypothesis that the
30 environments in which a male insect devel ops mediates the relative expression of male weapons size and testes mass, elements of pre and post copulatory sexual selection in males. The specific aim of this study was to use a sexually dimorphic insect species to examine the influence of two distin ct rearing environments on: (1) male investment in body size and enlarged hind femurs (weapons of sexual selection); and (2) male investment in testes mass. I allowed insects to develop on wild host plants, providing a description of the natural variation in these sexual traits within the focal population. The two species of host plant selected as rearing environments in this study differ in the resources that they provide developing insects. Previous work has shown that i nsects that develop on one host pla nt species, Heliconia platystachys , produ ce larger bodies and weapons than insects that develop on the second host plant, H. mariae (Miller, 2007; Miller & Emlen, 2010b) . Consistent with predictions of sperm competition theory (Parker 1998 ), I predict that insects producing larger sexually selected weapons will invest less in test es mass. Methods Study Organism The heliconia bug Leptoscelis tricolor Westwood (Hemiptera: Coreidae) provides an opportunity to examine the effects of heterogeneous environm ents on male sexually selected traits. Enlarged hind legs are a characteristic fea ture of insects in the family Coreidae and in many species, males with larger hind femurs are more likely to win competitions (Mitchell, 1980; Miyatake, 1997; Eberhard, 1998; Procter, Moore, & Miller, 2012) . Leptoscelis tricolor are sexually dimorphic. Males exhib it enlarged hind femurs with spines, and use these weapons to squeeze their male opponents in combat , while
31 females have smaller hind femurs with few or no spines (Miller, 2008; Miller & Emlen, 2010a,b) . Mating occurs when a male taps his fore legs and waves his antennae in front of a fema le, after which the male mounts the female and att empts to establish contact with her genitalia. A female must open a genital tergite to allow for intromission, thus males are not always successful in mating. However, once intromission is achieved the male faces away from the female and the pair remains i n copula, often for extended periods of time . Male and female L. tricolor mate multiply and copulation often lasts multiple hours (Miller, 2007, 2008) . Sperm competition has been recorded in other Hemiptera (Rubenstein, 1989; Carroll, 1991) . Although prolonged copulation is often associated with mate guarding it has also been associated with increased tran sfer of sperm in multiple insect systems where sperm is predicted to compete numerically (Dickinson, 1986; Arnqvist & Danielsson, 1998; GarcÂ’a GonzÂ‡lez & Gomendio, 2004) . The life history of L. tricolor is tightly linke d to the inflorescences of heliconia plants (Zingiberales: Heliconiae) (Stiles, 1975) . Heliconia mariae and H. platystachys are abundant host plant species present in and around Gamboa, Panama. Leptoscelis tricolo r juveniles complete their entire development on a single inflorescence belonging to one species of host plant (Miller, 2007) . Once these insects undergo their final molt to adulthood their exoskeleton becomes sclerotized and they remain the same size throughout their adult life (Miller 2007). Previous research has established tha t insects maturing on H. platystachys when it is bloomin g and fruiting develop larger bodies and hind femurs than insects that mature on H. mariae plants at the same phenological stage (Miller & Emlen, 2010b) , suggesting that H. playtstachys inflorescence s are a higher quality resour ce.
32 Insect R earing All data w ere collected within an area of 25km 2 near Gamboa, Panama in July and August 2013. Insects were found as 4th or 5th instar nymphs on H. platystachys and H. mariae in the wild. Fine mesh bags were slipped over individual inflore scences where juveniles were found and only one juvenile was enclosed per bag. I returned to collect adults estimated to be on average 17 days old (+/ 5 days). Adults are sexually mature and prepared to mate by 12 days of age . These adults were immediatel y place d in individual deli cups with moist paper towels. Weapon Measurements & Testes W eights Within 7 hours of collection male pronotum width and hind femur width were measured using Mituto yo digital calipers (maximum accuracy 0.01 cm). Pronotum width is a common metric of body size in insects, and hind femurs are used as weapons in male male contests in this and other related species. Hind femur widths were measured on the third distal spine, which represents the widest part of the femur and the part t hat most likely contacts opponents during competitions (Miller, 2007; Miller & Emlen, 2010a,b) . Within 24 hours of removal from the field, males were dissected under saline solution ( 0.09% NaCl) and testes were extracted and placed on pre weighed pieces of aluminum foil. Testes samples were placed in a drying oven at 60 o (+/ 3) degrees Celsius. A fter a minimum of 5 days in the drying oven, testes samples were weighed to a precision of 0.01 mg within 6 minutes of e xtraction from the drying oven. Results I found that male femurs varied considerably in size (Fig ure 2 1). I found that males raised on Heliconia platystachys were larger than males raised on H. mariae (Body size: ANOVA, F 1,50 =4.25, p = 0.044; weapon size: F 1,50 = 5.46, p =0.0 24). However,
33 males raised on H. mariae had larger testes than males raised on H. platystachys (ANOVA, F 1,51 = 5.90 and p = 0.019; Figure 2 1). Some males reared on H. platystachys were estimated to be 12 days old (n = 11), while all other males raised on H . platystachys (n = 24) and all males raised on H. mariae (n=18) were older than 14 days old. To ensure that the observed difference in testes mass was not due to ag e differences, I analyzed the effect of age on testes size for males reared on H. platystac hys and found no evid ence that age influenced testes mass (F 1,32 = 1.13 and p = 0.295). Discussion Our study reveals that ecologically relevant variation in developmental environment modifies the phenotypic expression of both weapons and testes, traits that likely function in pre and post copulatory sexual selection in Leptoscelis tricolor . Consistent with previous work on this sexually dimorphic insect I found that the size of males and the size of weapons used in pre copulatory contests, hind femurs, diffe red depending on the host plant on which males developed (Miller & Emlen, 2010b) . In general, developmental environment can have drastic effects on the investment the weapons of sexual selection (Kasumovic & Brooks, 2011; Emlen et al., 2012). Other studies have shown strong effects of nutritional environment on testes and sperm production (Gage & Cook, 1994; Simmons & Kotiaho, 2002) . Here I examined the effects of rearing environment on both pre copulatory and post copulatory sexual tr aits in a single species and found L. tricolor reared on the host plant Heliconia platystachys develop on average larger body and weapon size but have smaller testes than insects reared on H. mariae. These results reveal an inverse relationship in the expr ession of weapons and testes in this population across two common environments.
34 The idea that individuals have a limited pool of resources is the basic premise that trade offs must occur between fitness related traits (Stearns 1987). Potential trade offs have been fo und between male sexually selected traits and other life history traits such as immunity ( reviewed in Lewis et al . 2008) , dispersal ability (Yamane et al. , 2010) , and predation avoidance (Nakayama & Miyatake, 2010) . Mix ed evidence exists that trade offs occur between pre and post copulatory traits (Simmons & Emlen, 2006; Yamane et al. , 2010; Devigili et al. , 2012; Rahman et al. , 2013) . However, most studies that examine the relationships between pre and post copulatory traits have been conducted under laboratory condi tions, where natural environmental contexts are often difficult to reconstruct. The two distinct host plant species explored in this study likely provide different nutritional resources to developing insects, and potentially influence the relative developm ent of these sexual traits (Miller, 2007, 2008; Miller & Emlen, 2010a) . Individual insects may differ in their ability to acquire resources from each of the two environments, and may also differ in their ability to allocate these r esources to either testes or weapons. This variation in the ability to acquire and allocate resources to fitness related traits is an alternative, that is not necessarily exclusive, to trade offs (van Noordwijk & de Jong, 1986; de Jong & van Noordwijk, 1992; Reznick et al. , 2000 ; Zera & Harshman, 2001) . Furthermore, pleiotropic variation might also lead to the observed inverse relationships in the expression of these two traits (Cheverud, 1996; Wagner, 1996; Klingenberg, Mebus, & Auffray, 2003) . Male traits that function in pre copulato ry sexual selection and those that function in post copulatory sexual selection are predicted to both contribute to reproductive success, and thus are likely candidates for
35 th is intra organism co evolution. However, in L. tricolor, the fitness consequences of investing more in weapons or testes, and the precision by which selection can act to affect relationships between these traits, has yet to be explored. My results demonstr ate that host plants have a significant effect on the expression of both pre and post copulatory traits in L. tricolor. Specifically, my finding that the expression of weapons and testes shows a negative phenotypic relationship across contexts highlights the importance of measuring pre and post copulatory sexual traits, particularly in the ecologically relevant context in which they occur. The observed pattern provides motivation to further explore the specific mechanism by which different environments af fect the expression of weapons and testes. The potential for pre and post copulatory traits to vary across environments offers exciting opportunities to investigate the effects of different environments on the expression and utilization of sexual traits .
36 Figure 2 1. Variation in male hind femur morphology for Leptoscelis tricolor . V isual outlines of male hind femurs showing the range of variati on in size of these sexually selected weapons in L. tricolor .
37 Figure 2 2. Differen tial expression in weapons and testes depending on rearing environment in Leptoscelis tricolor. Investment in femur size and testes mass across host plant species . Inflorescences belonging to Heliconia mariae (a) and H. platystachys (b). (i) Male L. tricol or raised on host plant H. mariae exhibited smaller hind femurs (mean Â± SE, 1.73 Â± 0.07cm, n =18) than did males raised on H. platystachys (1.92 Â± 0.05 cm, n = 35). The Insets show standardized representations of an average sized hind femur and testes fr om a male L. tricolor raised on H. mariae and a male raised on H. platystachys respectively. (ii) Males raised on H. mariae have h eavier testes (0.51 +/ 0.03 mg , n =18), than those raised on H. platystachys (0.44 Â± 0.04 mg, n= 34 ).
38 LIST OF REFERENCES Andrade MCB. 2003. Risky mate search and male self sacrifice in redback spiders. Behavioral Ecology 14: 531 Ã 538. Arnqvist G, Danielsson I. 1998. Postmating sexual selection%: the effects of male body size and r ecovery period on paternity and egg production rate in a water strider. Behavioral Ecology 10: 358 Ã 365. Badyaev A V, QvarnstrÂšm A. 2002. Putting sexual traits into the context of an organism: A life history perspective in studies of sexual selection. The Au k 119: 301 Ã 310. Bell G. 2010. Fluctuating selection: the perpetual renewal of adaptation in variable environments. Philosophical transactions of the Royal Society of London. Series B, Biological Sciences 365: 87 Ã 97. Brodie ED, Moore AJ, Janzen FJ. 1995. Vi sualizing and quantifying natural selection. Trends in Ecology and Evolution 10: 313 Ã 318. BussiÂre LF, Hunt J, StÂšlting KN, Jennions MD, Brooks R. 2008. Mate choice for genetic quality when environments vary: suggestions for empirical progress. Genetica 13 4: 69 Ã 78. Byrne PG, Roberts JD, Simmons LW. 2002. Sperm competition selects for increased testes mass in Australian frogs. Journal of Evolutionary Biology 15: 347 Ã 355. Carroll SP. 1991. The adaptive significance of mate guarding in the soapberry bug, Jader a haematoloma (Hemiptera: Rhopalidae). Journal of Insect Behavior 4: 509 Ã 530. Carroll SP, Hendry AP, Reznick DN, Fox CW. 2007. Evolution on ecological time scales. Functional Ecology 21: 387 Ã 393. Cheverud JM. 1996. Developmental integration and the evoluti on of pleiotropy. Integrative and Comparative Biology 36: 44 Ã 50. Cornwallis CK, Uller T. 2010. Towards an evolutionary ecology of sexual traits. Trends in Ecology and Evolution 25: 145 Ã 152. Darwin C. 1871. The descent of man, and selection in relation to s ex . London: John Murray. Devigili A, Kelley JL, Pilastro A, Evans JP. 2012. Expression of pre and postcopulatory traits under different dietary conditions in guppies. Behavioral Ecology 24: 740 Ã 749.
39 Dickinson JL. 1986. Prolonged mating in the milkweed lea f beetle Labidomera clivicollis clivicollis (Coleoptera: Chrysomelidae)%: a test of the "sperm loading" hypothesis. Behavioral Ecology and Sociobiology 18: 331 Ã 338. Eberhard WG. 1998. Sexual Behavior of Acanth ocephala declivis guatemalana (Hemiptera:Coreidae ) and the allometric scaling of their modified hind Leg s. Behavior 91: 863 Ã 871. Eberhard WG. 2009. Postcopulatory sexual selection: Darwin's omission and its consequences. Proceedings of the National Academy of Sciences 106: 10025 Ã 10032. Emlen DJ. 2008. The Evolution of Animal Weapons. Annual Review of Ecology , Evolution, and Systematics 39: 387 Ã 413. Emlen ST, Oring LW. 1977. Ecology, sexual selection and evolution of mating systems. Science 197: 215 Ã 223. Evans JP. 2010. Quantitative genetic evidence that males trade attractiveness for ejaculate quality in gupp ies. Proceedings of the Royal Society B: Biological Sciences 277: 3195 Ã 3 201. Gage MJG. 1994. Associations between body size, mating pattern , testis size and sperm lengths across butterflies. Proceedings of the Royal Society B: Biological Sciences 258: 247 Ã 254. Gage MJG, Cook PA. 1994. Sperm size or numbers%? Effects of nutritional stress upon eupyrene and apyrene sperm production strategies in t he moth Plodia interpunctella (Lepidoptera:Pyralidae ). Functional Ecology 8: 594 Ã 599. GarcÂ’a GonzÂ‡lez F, Gomendio M. 2004. Adjustment of copula duration and ejaculate size according to the risk of sperm competition in the golden egg bug ( Phyllomorpha laciniata ). Behavioral Ecology 15: 23 Ã 30. Grant PR, Grant BR. 2002. Unpredictable evolution in a 30 year study of Darwin's finches. Science 296: 707 Ã 711. Griffith SC, Owens IPF, Burke T. 1999. Environmental determination of a sexually selected trait. Nature 400: 358 Ã 360. Haller BC, Hendry AP. 2014. Solving the paradox of stasis: squashed stabilizing selection and the limits of detection. Evolution 68: 483 Ã 500. Hamon TR, Foote CJ, Url S, Salmon S, Nerka O. 2005. Concurrent Natural and Sexual Selection in Wild Male Sockeye Salmon , Oncorhynchus nerka . Evolution 59 : 1104 Ã 1118.
40 Harcourt AH, Purvist A, Liles L. 1995. Sperm competition: mating system , not b reeding season , affects testes size of primates. Functional Ecology 9: 468 Ã 476. Hereford J, Hansen TF, Houle D. 2004. Comparing strengths of directional selection: how strong is strong? Evolution 58: 2133 Ã 2143. Hersch EI, Phillips PC. 2004. Power and pote ntial bias in field studies of natural selection. Evolution 58: 479 Ã 4 85. Hosken DJ. 2001. Sex and death: microevolutionary trade offs between reproductive and immune investment in dung flies. Current Biology 11 : 379 Ã 3 80. Hunt J, Breuker CJ, Sadowski JA, Mo ore AJ. 2009. Male male competition, female mate choice and their interaction: determining total sexual selection. Journal of Evolutionary Biology 22: 13 Ã 26. Huxley JS. 1932. Problems of relative growth. London: Methuen. De Jong G, van Noordwijk AJ. 1992. Acquisition and allocation of Resources%: genetic ( CO ) variances , selection , and life histories. American Naturalist 139: 749 Ã 770. Kasumovic MM, Bruce MJ, Andrade MCB, Herberstein ME. 2008. Spatial and temporal demographic variation drives within seaso n fluctuations in sexual selection. Evolution 62: 2316 Ã 2325. Kasumovic MM, Andrade MCB. 2009. A change in competitive context reverses sexual selection on male size. Journal of Evolutionary B iology 22: 324 Ã 33. Kingsolver JG, Hoekstra HE, Hoekstra JM, Berri gan D, Vignieri SN, Hill CE, Hoang A, Gibert P, Beerli P. 2001. The strength of phenotypic selection in natural populations. The American Naturalist 157: 245 Ã 261. Kingsolver JG, Diamond SE, Siepielski AM, Carlson SM. 2012. Synthetic analyses of phenotypic selection in natural populations: lessons, limitations and future directions. Evolutionary Ecology 26: 1101 Ã 1118. Kingsolver JG, Pfennig DW. 2007. Patterns and Power of Phenotypic Selection in Nature. BioScience 57: 561 Ã 572. Klingenberg CP, Mebus K, Auffra y JC. 2003. Developmental integration in a complex morphological structure: how distinct are the modules in the mouse mandible? Evolution & Developmen t 5: 522 Ã 5 31. Lewis Z, Price TAR, Wedell N. 2008. Sperm competition, immunity, selfish genes and cancer. C ellular and Molecular Life Sciences 65: 3241 Ã 3254.
41 Miller CW. 2007. Maternal effects and sexual selection in the heliconia bug, L eptoscelis tricolor (Hemiptera: Coreidae). Ph.D. dissertation, Division of Biological Sciences, University of Montana, Missoula. Miller CW. 2008. Seasonal effects on offspring reproductive traits through maternal oviposition behavior. Behavioral Ecology 19: 1297 Ã 1304. Miller CW, Emlen DJ. 2010a. Across and Within Population Differences in the Size and Scaling Relationship of a Sex ually Selected Trait in Leptoscelis tricolor (Hemiptera: Coreidae). Annals of the Entomological Society of America 103: 209 Ã 215. Miller CW, Emlen DJ. 2010b. Dynamic effects of oviposition site on offspring sexually selected traits and scaling relationships . Evolutionary Ecology 24: 375 Ã 390. Mitchell PL. 1980. Combat and territorial defense of Acanthocephala femorata ( Hemiptera%: Coreidae ). Entomological Society of America 73: 404 Ã 408. Miyatake T. 1997. Functional morphology of the hind legs as weapons for male contests in Leptoglossus australis (Heteroptera:Coreidae ). Journal of Insect Behavior 10: 727 Ã 735. Miyatake T. 2002. Multi Male Mating Aggregation in Notobitus meleagris ( Hemiptera%: Coreidae ). Ecology and Population Biology 95: 340 Ã 344. Moczek AP, Emlen DJ. 1999. Proximate determination of male horn dimorphism in the beetle Onthophagus taurus (Coleoptera: Scarab aeidae). Journal of Evolutionary Biology 12: 27 Ã 37. Nakayama S, Miyatake T. 2010. Genetic trade off between abilities to avoid attack and to mate: a cost of tonic immobility. Biology Letters 6: 18 Ã 20. Van Noordwijk AJ, de Jong G. 1986. Acquisition and allo cation of resource s %: Their influence on variation in life history tactics. American Naturalist 128: 137 Ã 142. Parker GA. 1970. Sperm competition and its evolutionary consequences in the insects. Biological Reviews 45: 525 Ã 567. Parker GA. 1998. Sperm competition and the evol ution of ejaculates: towards a theory base. Sperm Competition and Sexual Selection (T.R. Birkhead & A.P. MÂ¿ller, eds), pp. 3 Ã 54. Academic Press, New York. Pischedda A, Rice WR. 2012. Partitioning sexual selection into its mating success and fertilization s uccess components. Proceedings of the National Academy of Sciences 109: 2049 Ã 2053.
42 Preston BT, Stevenson IR, Pemberton JM, Coltman DW, Wilson K. 2003. Overt and covert competition in a promiscuous mammal: the importance of weaponry and testes size to male reproductive success. Proceedings of the Royal Society B: Biological Sciences 270: 633 Ã 6 40. Preziosi RF, Fairbairn DJ. 2000. Lifetime selection on adult body size and components of body size in a waterstrider: opposing selection and maintenance of sexual size dimorphism. Evolution 54: 558 Ã 566. Procter DS, Moore a J, Miller CW. 2012. The form of sexual selection arising from male male competition depends on the presence of females in the social environment. Journal of Evolutionary B iology 25: 803 Ã 8 12. Punia moorthy N, Blanckenhorn WU, SchÂŠfer MA. 2012. Differential investment in pre vs. post copulatory sexual selection reinforces a cross continental reversal of sexual size dimorphism in Sepsis punctum (Diptera: Sepsidae). Journal of Evolutionary Biology 25: 2 253 Ã 22 63. Punzalan D, Rodd FH, Rowe L. 2010. Temporally variable multivariate sexual selection on sexually dimorphic traits in a wild insect population. The American Naturalist 175: 401 Ã 414. QvarnstrÂšm A. 1999. Genotype by environment interactions in the d etermination of the size of a secondary sexual charact er in the collared flycatcher ( Ficedula albicollis ). Evolution 53: 1564 Ã 1572. Rahman MM, Kelley JL, Evans JP. 2013. Condition dependent expression of pre and postcopulatory sexual traits in guppies. Ec ology and Evolution 3: 2197 Ã 2 213. Reznick DN, Shaw FH, Rodd FH, Shaw RG. 1997. Evaluation of the rate of evolution in natural populations of guppies ( Poecilia reticulata ). Science 275: 1934 Ã 1937. Reznick D, Nunney L, Tessier A. 2000. Big houses, big cars, superfleas and the costs of reproduction. Trends in Ecology and Evolution 15: 421 Ã 425. Robinson MR, Pilkington JG, Clutton Brock TH, Pemberton JM, Kruuk LEB. 2006. Live fast, die young: trade offs between fitness components and sexually antagonistic select ion on weaponry in Soay sheep. Evolution 60: 2168 Ã 21 81. Rubenstein DI. 1989. Sperm competition in the water strider, Gerris remigis . Animal Behaviour 38: 631 Ã 636. Schoener TW. 2011. The newest synthesis: understanding the interplay of evolutionary and ecol ogical dynamics. Science 331: 426 Ã 429. Simmons LW, Emlen DJ. 2006. Evolutionary trade off between weapons and testes. Proceedings of the National Academy of Sciences 103: 16346 Ã 163 51.
43 Simmons LW, Kotiaho JS. 2002. Evolution of ejaculates: patterns of pheno typic and genotypic variation and condition dependence in sperm competition traits. Evolution 56: 1622 Ã 16 31. Steele DB, Siepielski AM, McPeek MA. 2011. Sexual selection and temporal phenotypic variation in a damselfly population. Journal of Evolutionary Bi ology 24: 1517 Ã 15 32. Stiles FG. 1975. Ecology, flowering phenology, and hummingbird pollination of some Costa Rican heliconian species. Ecology 56: 285 Ã 301. Stinchcombe JR, Agrawal AF, Hohenlohe PA, Arnold SJ, Blows MW. 2008. Estimating nonlinear selection gradients using quadratic regression coefficients: double or nothing? Evolution 62: 2435 Ã 40. Trillo PA. 2008. Pre and post copulatory sexual selection in the tortoise beetle. Ph.D. dissertation, Division of Biological Sciences, University of Montana, Mis soula. Wagner GP. 1996. Homologues, Natural Kinds and the Evolution of Modularity. Integrative and Comparative Biology 36: 36 Ã 43. Yamane T, Okada K, Nakayama S, Miyatake T. 2010. Dispersal and ejaculatory strategies associated with exaggeration of weapon i n an armed beetle. Proceedings of the Royal Society B: Biological Sciences 277: 1705 Ã 10. Zera AJ, Harshman LG. 2001. The physiology of life history trade offs in animals. Annual Review of Ecology and Systematics 32: 95 Ã 126. Zuk M, Thornhill R, Ligon JD, Jo hnson K, Austad S, Ligon SH, Thornhill NW, Costin C. 1990. The role of male ornaments and courtship behavior in female mate choice of red jungle f owl. The American Naturalist 136: 459 Ã 473.
44 BIOGRAPHICAL SKETCH Ummat Somjee was born in 1985 in Nairobi, Kenya. He grew up with his family in the suburbs of Nairobi city and f rom a very young age took advantage of the opportunit i es to explore and experience the remarkable diversity of cultures, wildlife and ecosystems of East Africa. He enjoyed spending time outdoors observing animals , a quality that has persisted throughout his life . He attended Simon Fraser University in British Columbia, Canada, where he participated in a diverse range of field and laboratory research activities in cluding work in Canada's Pacific Coast and in Alaska. He enjoys reading , hiking and climbing in the outdoors . He received his Master of Science from the University of Florida in the summer of 2014.