The Evolution and Maintenance of the Color Polymorphism in Plethodon Cinereus

Home , Apostatic selection, Panmixia, Polymorphism (biology), Sympatry, Timema cristinae, Western slimy salamander

The Evolution and Maintenance of the Color Polymorphism in Plethodon cinereus

(Caudata: Plethodontidae)

A dissertation presented to

the faculty of

the College of Arts and Sciences of Ohio University

In partial fulfillment

of the requirements for the degree

Doctor of Philosophy

Maggie M. Hantak

August 2019

This dissertation titled

The Evolution and Maintenance of the Color Polymorphism in Plethodon cinereus

(Caudata: Plethodontidae)

MAGGIE M. HANTAK

has been approved for

the Department of Biological Sciences

and the College of Arts and Sciences by

Shawn R. Kuchta

Associate Professor of Biological Sciences

Florenz Plassmann

Dean, College of Arts and Sciences 3

ABSTRACT

HANTAK, MAGGIE M., Ph.D., August 2019, Biological Sciences

The Evolution and Maintenance of the Color Polymorphism in Plethodon cinereus

(Caudata: Plethodontidae)

Director of Dissertation: Shawn R. Kuchta

Color polymorphism is the presence of two or more distinct, genetically determined color morphs within a single interbreeding population. An underexplored question in color polymorphic species pertains to how distinct phenotypes are maintained within and among populations. Little work has been done to examine geographic patterns in polymorphisms, with most studies focusing on a single population. Mechanisms that may maintain polymorphisms include negative frequency dependent selection, spatiotemporal variation in selection, and gene flow among populations. Investigating these mechanisms in multiple populations provides insight into the processes underlying the maintenance of genetic variation within and among populations.

The Eastern Red-backed Salamander, Plethodon cinereus, is widespread throughout northeastern North America, and has two common color morphs: striped and unstriped. Previous studies have suggested that the two color morphs of P. cinereus differ in many elements of their biology, including physiology, territoriality, and mating interactions. However, most studies focus on a single biological feature in a single population, so little is known about diversifying selection. In addition, the evolutionary processes that operate on the polymorphism in populations are not understood. 4

This dissertation examines multiple mechanisms that may be contributing to color morph maintenance in P. cinereus with replication over both space and time, including populations that vary from monomorphic to polymorphic. In chapter 1, I reviewed previous literature on co-adapted trait complexes in P. cinereus and provided an overview of the mechanisms that may be maintaining two color morphs. I suggested additional studies are necessary and I also conveyed the importance of studying the polymorphism in a geographic and phylogenetic context. In chapter 2, I investigated whether spatial and temporal variation in selection aids in color morph maintenance by examining whether the striped or unstriped morph was more camouflaged from the perspective of visual predators. I found that the unstriped morph was in general more conspicuous to potential predators, although the level of camouflage varied by population and season. In chapter 3, I aimed to identify the relationship between color morph frequency, genetic structure, landscape and environmental variables using a landscape genetic framework. The results suggested that a mix of gene flow, genetic drift, and selection interacted to maintain the striped/unstriped polymorphism. In chapter 4, I examined whether dietary partitioning characterized the two morphs over spatiotemporal scales. Across six populations and two seasons, there was spatial and temporal variation in diet. However, I found no evidence of diet differences between the morphs in polymorphic populations and no change in dietary breadth between polymorphic and monomorphic populations, demonstrating lack of dietary ecological release. Finally, in chapter 5, I investigated whether morphs assortatively mated by color and/or body size across geographically distinct populations that varied in color morph frequency. Across 5 the studied populations, I found evidence of random mating between color morphs, which may contribute to color morph maintenance within populations. In addition, I found geographic variation in size-assortative mating, which may have led to reproductive isolation among populations. Together, these studies provided information on how the morphs of P. cinereus are maintained and demonstrate the importance of studying color polymorphisms over spatial and temporal scales.

DEDICATION

For my family and friends, whose love and support made this dissertation possible.

ACKNOWLEDGMENTS

My advisor, Shawn Kuchta, was continuously kind and helpful during the duration of this work. I would not have made it to where I am today without his wonderful mentorship. I thank my committee: Willem Roosenburg, Harvey Ballard,

Kelly Williams, and Joseph Johnson for valuable feedback on this dissertation. I would especially like to thank Kelly Williams for providing statistical guidance on multiple chapters of this dissertation. My collaborators, Carl Anthony and Cari Hickerson were essential in contributing to the development and implementation of this research. Kyle

Brooks and Olivia Brooks were vital to completing many elements of this dissertation.

They helped collect an extraordinary amount of ecological data in the field, as well as substantially contribute to further data collection in the laboratory. I would like to thank my family, especially my parents, Debbie and Jim Hantak and my partner, Dan Paluh, for believing in and supporting me. Melissa Liotta, Alayna Tokash, and Kaili Boarman provided emotional support. Much traveling was necessary for the completion of this dissertation, and I would like to thank Carl Anthony and Cari Hickerson for providing housing, as well as my parents, and Old Woman Creek National Estuarine Research

Reserve (NERR). I worked in several incredible localities for the duration of my dissertation research: Chapin Forest Reservation, Case Western Reserve University

Squire Valleevue Farm, Cuyahoga Valley National Park, Manatoc Scout Reservation,

Edison Woods Metropark, Old Woman Creek NERR, East Harbor State Park, and the

Heineman property on South Bass Island. I would like to thank everyone who permitted me to work at these properties. Financial support for this dissertation was provided by the 8

Ohio Center for Ecology and Evolutionary Studies (OCEES), the National Science

Foundation, an Ohio University Student Enhancement Award, Ohio University Graduate

Student Senate Grants, the Society for the Study of Evolution Rosemary Grant Award, the Ohio University Graduate Student Research Fund, Old Woman Creek travel grants,

Ohio Biological Survey Small Grant awards, and the Society for the Study of

Amphibians and Reptiles Grants in Herpetology Award.

TABLE OF CONTENTS

Page

Abstract ...... 3 Dedication ...... 6 Acknowledgments...... 7 List of Tables ...... 11 List of Figures ...... 12 Chapter 1: A Review of Color Polymorphism and Co-adaptations in the Eastern Red- backed Salamadner (Plethodon cinereus) ...... 13 Introduction ...... 13 Chapter 2: Predator Perception Across Space and Time: Relative Camouflage in a Color Polymorphic Salamander ...... 25 Introduction ...... 25 Methods...... 28 Results ...... 32 Discussion ...... 37 Tables ...... 42 Figures...... 45 Chapter 3: Do Genetic Structure and Landscape Heterogeneity Impact Color Morph Frequency in a Polymorphic Salamander? ...... 48 Introduction ...... 48 Methods...... 51 Results ...... 56 Discussion ...... 60 Tables ...... 68 Figures...... 71 Chapter 4: A Spatiotemporal Assessment of Dietary Partitioning Between Color Morphs of a Terrestrial Salamander ...... 74 Introduction ...... 74 Methods...... 76 Results ...... 79 Discussion ...... 81 Tables ...... 89 10

Figures...... 92 Chapter 5: Does Random Mating Promote the Maintenance of a Common Salamander Color Pattern Polymorphism? ...... 95 Introduction ...... 95 Methods...... 98 Results ...... 101 Discussion ...... 102 Tables ...... 109 Figures...... 110 Chapter 6: Conclusions ...... 113 References ...... 117 Appendix A: Chapter 2 Supplemental Material...... 139 Appendix B: Chapter 3 Supplemental Material ...... 146 Appendix C: Chapter 4 Supplemental Material ...... 167 Appendix D: Chapter 5 Supplemental Material...... 172

LIST OF TABLES

Page

Table 2.1 Tukey’s HSD post hoc test P-values for dorsal P. cinereus ...... 42 Table 3.1 Resistance distance model selection with genetic distance as the dependent .. 68 Table 3.2 LCP model selection with genetic distance as the dependent ...... 69 Table 3.3 Resistance models with morph frequency as the dependent ...... 70 Table 4.1 Mean number and volume of prey consumed ...... 89 Table 4.2 PERMANOVA and ANOVA results for dietary analyses ...... 90 Table 4.3 Importance values for dominant prey categories ...... 91 Table 5.1 Results of the best-supported GLMs for color morph mate preference ...... 109

LIST OF FIGURES

Page

Figure 2.1. Reflectance of P. cinereus dorsal colors by season, population and morph .. 45 Figure 2.2. Dorsal contrast values for avian, snake, and mammalian visual systems ...... 46 Figure 2.3. Color space plots representing visual predator perspectives ...... 47 Figure 3.1. Sampling localities, morph frequencies, genetic structure ...... 71 Figure 3.2. Landscape surfaces with sampling localities ...... 72

Figure 3.3. Map of northern Ohio localities with FST values between pairs of sites ...... 73 Figure 4.1. Map of study sites on a percent canopy cover landscape surface in Ohio ..... 92 Figure 4.2. Nonmetric multidimensional scaling plots of diet variance ...... 93 Figure 4.3. Salamander color morph and leaf litter invertebrate breath ...... 94 Figure 5.1. Map of color morph frequencies in Ohio ...... 110 Figure 5.2. Distribution of mating pairs from three polymorphic sites ...... 111 Figure 5.3. Female and male body size regressions...... 112

CHAPTER 1: A REVIEW OF COLOR POLYMORPHISM AND CO-ADAPTATIONS

IN THE EASTERN RED-BACKED SALAMADNER (PLETHODON CINEREUS)

Introduction

Color polymorphism is the simultaneous occurrence of two or more discrete, genetically determined phenotypes within a population, with the rarest morph present at a higher frequency than can be maintained by recurrent mutation (Ford, 1945; Huxley

1955). Morphs are comprised of co-adapted sets of traits, and therefore differ in various elements of their biology in addition to color, including physiology, behavior, or other morphological characters (Sinervo & Svensson, 2002). For example, in the Side-blotched

Lizard (Uta stansburiana), males with distinct throat color patches also differ in mating behavior, territoriality, and testosterone levels (Sinervo & Lively, 1996; Corl et al.,

2010). These correlated trait complexes suggest that morphs occupy divergent fitness optima, allowing different individuals of a single species within a population to inhabit divergent ecological niches (West-Eberhard, 1986; Forsman et al., 2008; Calsbeek et al.,

2012). However, within populations morphs may compete over resources such as food, territories, or microhabitats (West-Eberhard, 1986; Svensson et al., 2001; Comendant et al., 2003; Anthony et al., 2008). If morphs compete, polymorphic populations that then become fixed for one morph should experience ecological release, which can lead to rapid evolution and species formation (West-Eberhard, 1986). With increased interest in how color polymorphism promotes species formation (Forsman et al., 2008; McKinnon

& Pierotti, 2010; Corl et al., 2010; Hugall & Stuart-Fox, 2012), it is necessary to study ecologically-relevant differences in form and behavior between color morphs, to identify 14 how niche evolution promotes speciation (West-Eberhard, 1986; Sinervo & Svensson,

2002; Corl et al., 2010; Calsbeek et al., 2012; Kuchta & Wake, 2016).

While color polymorphism is a classic problem in evolutionary biology (Cain &

Sheppard, 1945; Ford, 1945; Endler, 1978), little work has been done to examine geographic patterns in polymorphisms, with most studies instead focusing on single populations (Sinervo & Lively, 1996; Svensson et al., 2001; McLean & Stuart-Fox, 2014;

Hantak et al., 2016; but see Corl et al., 2010; Davis-Rabosky et al., 2016). Possible mechanisms to maintain polymorphisms over spatiotemporal scales include negative frequency dependent selection, variation in selection in space and time, and gene flow among populations (Gray & McKinnon, 2007; McLean & Stuart-Fox, 2014).

Investigating the contribution of these processes to creating patterns of population structure provides important insights into the relative influence of geographic variation and ecological interactions in adaptation and species formation (Endler, 1977; Corl et al.,

2010; Kuchta & Wake, 2016).

Here, I review previous research demonstrating correlated trait complexes in two common color morphs of the Eastern Red-backed Salamander (Plethodon cinereus).

There has been much work done examining these morph differences. However, many studies have found conflicting results, and therefore, a mechanistic understanding of the ecological differences between the morphs, and the evolutionary processes that maintain the polymorphism, remains lacking. I conclude there is a need for continued research in this field, but convey the importance of investigating polymorphic species in a geographic and phylogenetic context. 15

Color polymorphism in Plethodon

Plethodontidae is the most species-rich salamander family with 475 recognized species (AmphibiaWeb, 2019), which all lack lungs and generally require cool, moist conditions for cutaneous respiration (Spotila, 1972). Within Plethodontidae, Plethodon is the largest North American salamander genus with 56 described species (Wiens et al.,

2006). Plethodon are fully-terrestrial and occupy upland habitat under moist leaves, logs, and rocks to avoid desiccation (Petranka, 1998). All Plethodon species lack the characteristic amphibian aquatic larval stage and instead lay eggs underground that hatch into miniature versions of adults (Petranka, 1998). Plethodon home ranges are small (< 4 m2), but population sizes are extremely large (thousands/hectare), and in some areas in the eastern United States Plethodon are the most abundant terrestrial vertebrate

(Merchant, 1972; Burton & Likens, 1975; Kleeberger & Werner, 1982).

Ten species within Plethodon display a dorsal striped/unstriped color pattern polymorphism (Petranka, 1998; Highton, 2004). These species include P. shenandoah, P. dorsalis, P. serratus, P. ventralis, P. sherando, P. cinereus, P. angusticlavius, P. websteri, P. vehiculum, and P. dunni. The ‘striped’ color morph exhibits a red stripe overlaid on a black dorsum and the ‘unstriped’ morph is completely black in dorsal coloration. Closely related species that lack this polymorphism are typically fixed for a striped phenotype (P. welleri, P. idahoensis, P. vandykei, P. larselli, P. elongatus, P. stormi, P. asupak) or an unstriped phenotype (P. wehrlei, P. nettingi, P. richmondi, P. electromorphus, P. virginia, P. hoffmani, P. neomexicanus; Petranka, 1998; Fisher-Reid

& Wiens, 2013). We know little with regard to the polymorphism in most of these 16 species. Nearly all of our knowledge on the striped/unstriped dimorphism comes from studies on the Eastern Red-backed Salamander, Plethodon cinereus.

Plethodon cinereus is a model species for the biology of polymorphism. As one of the most abundant terrestrial vertebrates in the northeastern United States (Burton &

Likens, 1975; Mathis, 1991), P. cinereus has been used as a model organism for a variety of studies in ecology and evolution (reviewed in Anthony & Pfingsten, 2013; Jaeger et al., 2016). Several studies of P. cinereus have demonstrated that this salamander is territorial (Jaeger & Forester, 1993; Mathis et al., 1995; Reiter et al., 2014). Both males and females aggressively defend territories from conspecifics (Jaeger, 1981; Mathis,

1991) and heterospecifics (Deitloff et al., 2008), including large invertebrates (Gall et al.,

2003; Hickerson et al., 2004). Individuals of P. cinereus display site tenacity across seasons and years (Gergits & Jaeger, 1990; Gillette 2003; Anthony & Pfingsten, 2013), and when displaced from territories they exhibit homing behavior (Kleeberger & Werner,

1982; Ousterhout & Liebgold, 2010). Territories are established under cover objects to avoid desiccation and to acquire prey (Jaeger, 1980; Jaeger et al., 1981). Holding high quality territories is also beneficial for mate attraction (Mathis, 1991).

Plethodon cinereus has a widespread distribution within the northeastern United

States and is color polymorphic throughout portions of its range (Highton, 2004). The striped and unstriped color phenotypes have a genetic basis with no intermediate phenotypes (Highton, 1959, 1975). Previous research has demonstrated that the two morphs differ along multiple axes of niche variation, including physiology, territoriality, diet, and they mate assortatively by color in least one population (reviewed in Anthony & 17

Pfingsten, 2013). Altogether, these studies indicate that the morphs differ in more than color as selection has acted on suites of traits to produce covariant, differentially adapted character sets (Sinervo & Svensson, 2002).

Correlated trait complexes in P. cinereus

Physiology

Several studies have investigated temperature associations between morphs of P. cinereus. Of these, multiple investigations have demonstrated that the unstriped morph is more common in warmer and drier conditions, whereas the striped morph is associated with cooler and wetter habitats (Lotter & Scott, 1977; Moreno, 1989; Gibbs & Karraker,

2006; Anthony et al. 2008). There is also evidence that the unstriped morph retreats underground sooner with the onset of cold weather (Lotter & Scott, 1977; Moreno, 1989;

Anthony et al., 2008). In addition, in a metanalysis, Gibbs and Karraker (2006) demonstrated that the probability of finding striped morphs increased with altitude, latitude, and longitude.

More recent studies have demonstrated complexity with climatic trends, however.

Petruzzi et al. (2006) found that activity temperatures of P. cinereus differed from previous studies, with striped morph activity higher in warmer temperatures, although their results varied with location and season. Fisher-Reid et al. (2013) found that a parapatrically distributed population of striped and unstriped morphs on Long Island,

New York also exhibited complicated trends. In this study, monomorphic unstriped populations were correlated with colder, drier conditions, whereas, in polymorphic populations, unstriped morphs were more active on warmer days than striped morphs. 18

The authors suggested that these patterns may be due to coastal insulation of Long Island

(Fisher-Reid et al., 2013). Cosentino et al. (2017) found striped morph frequency was correlated with cooler temperatures and increased latitudes, but they also found that striped morph frequency was correlated with warmer regions that contained higher forest cover. Moore and Ouellet (2015) conducted a comprehensive literature search on P. cinereus occurrence data, paired with climatic and geographic variables, and found a weak relationship between morph frequency and temperature, suggesting that morph frequency is not affected by climate. Finally, Evans et al. (2018) was interested in whether alterations in climate affected color morph frequencies through time, but found no changes in the distribution of the morphs over a 40-year time period.

Two studies have shown that striped and unstriped morphs differ in their resting maintenance metabolic rate (Moreno, 1989; Petruzzi et al., 2006). In general, the unstriped morph has a lower metabolic rate, which authors postulate permit this morph to be more active in warmer, drier conditions (Moreno, 1989; Petruzzi et al., 2006). In an experimental study examining dehydration and rehydration rates, morphs did not differ in rate of water loss, however, striped morphs had higher rehydration rates than unstriped morphs (Smith et al., 2015). The authors suggested that rehydration rates may impact variation in morph microhabitat selection or activity times (Smith et al., 2015). Finally,

Davis and Milanovitch (2010) demonstrated that stress levels, measured by corticosterone, of unstriped morphs were higher than the stress levels of striped individuals. The authors suggest that these findings are in accordance with greater 19 predation pressure on unstriped morphs compared to striped morphs (discussed below;

Davis & Milanovitch, 2010).

Diet

Multiple studies have demonstrated that the striped and unstriped morphs differ in dietary composition. In one population in northeast Ohio, Anthony et al. (2008) and

Stuczka et al. (2016) found that striped morph diet was composed of more energetically profitable (i.e., soft bodied) prey. In addition, Anthony et al. (2008) found that striped morphs consumed a higher number of prey and had a more diverse diet than unstriped morphs. Conversely, Stuczka et al. (2016) found that unstriped morphs contained a more diverse diet than striped morphs. In the same polymorphic population, Paluh et al. (2015) examined whether morphs selectively foraged on ants within their natural territories. The authors found no morph-specific differences in ant species consumed, however, striped individuals consumed more ants, which were also more abundant in their territories. In this thesis (Chapter 4), I examined dietary composition and ecological release of morphs across multiple polymorphic and monomorphic populations. In contrast with other studies, I found no evidence of dietary partitioning between the morphs in polymorphic populations; instead, the morphs overlapped in prey type, number, and volume. In addition, there was no change in dietary breadth between polymorphic and monomorphic populations, and thus there was no signature of dietary release.

Disease

In an experimental study that examined morph-specific differences in the effects of chytrid fungus (Batrachochytrium dendrobatidis), Venesky et al. (2015) found that the 20 morphs differed in their disease prevalence, and the unstriped morph had a higher prevalence of infection.

Predation

Several studies have demonstrated that the morphs exhibit different responses to potential predators. When faced with a Common Garter Snake, Thamnophis sirtalis,

Venesky and Anthony (2007) found that striped morphs spent more time in an immobile defensive posture, whereas unstriped morphs put energy into escape behaviors. In a laboratory-based study, Otaibi et al. (2017) investigated morph differences in postautotomy tail movement. Tail autotomy is a common mechanism used by amphibians and reptiles to escape predation. In this study, the authors found that striped tails moved for a longer period of time compared to unstriped tails, which may afford the striped morph more time to escape predation (Otaibi et al., 2017). In addition, two studies have shown that morphs differ in their frequency of tail breakage, with unstriped morphs consistently having higher rates of tail autotomy, suggesting greater predation on this morph (Moreno,

1989; Venesky & Anthony, 2007). In contrast to these aforementioned studies, Grant et al. (2018) demonstrated some evidence that striped morphs were preyed upon more and had lower survival compared to unstriped morphs using mark-recapture data and a clay model study. The authors suggested that this may be due to the unstriped morph being more cryptic to avian predators. Hantak and Kuchta (2018; Chapter 2) investigated the role of spatial and temporal variation in selection by examining relative camouflage of morphs from the perspective of potential predators. With use of visual modeling, we found that, in general, the unstriped morph was more conspicuous to avian, mammalian, 21 and snake predators, although there was some spatial and temporal variation in the degree of morph camouflage.

Territoriality

Using both a field mark-recapture study and an experimental setup, Reiter et al.

(2014) examined whether the morphs differed in their levels of territoriality in one Ohio population. The authors found that striped morphs were dominant territory holders compared to unstriped morphs and expelled intruders from cover objects (Reiter et al.,

2014). Anthony et al. (2017) found that unstriped morphs occupy territories that have lower prey richness than striped morphs. In addition, when examining morph variation in fine-scale dispersal, Grant and Liebgold (2017) found striped morphs were more philopatric and displayed a lack of dispersal compared to unstriped morphs. Concordance among these studies suggests that the striped morph may consistently hold a territorial advantage over the unstriped morph.

Mating interactions

With field surveys and laboratory preference tests, Anthony et al. (2008) and

Acord et al. (2013) found that the morphs assortatively mate by color in one Ohio population. These studies also demonstrated a tendency for larger females to pair more often with striped males compared to unstriped males. The authors suggested this may be due to striped males holding higher-quality territories and consuming more profitable prey (Anthony et al., 2008; Reiter et al., 2014; Stuczka et al., 2016). For part of my thesis

(Chapter 5), I examined multiple polymorphic populations and found no evidence of color assortative mating or morph preference for a certain body size. These discrepancies 22 among studies indicate that there may be spatiotemporal variation in mate preference between P. cinereus color morphs. Finally, in a study solely investigating male mate preference in P. cinereus, Jaworski et al. (2018) demonstrated that color did not appear important during mate selection; instead, males preferred larger females that presumably produce larger clutch sizes.

Color morph maintenance

Although much research has investigated trait variation in P. cinereus, relatively few studies have considered why and how the two morphs are maintained over spatiotemporal scales. In a clay model replica study of the striped and unstriped morphs,

Fitzpatrick et al. (2009) found evidence that negative frequency-dependent selection may be maintaining the polymorphism. However, all tests were performed in an open field at one locality, and models contained a food reward on the underside. Another clay model study performed by Kraemer et al. (2016) found no evidence of negative frequency- dependent selection by mammalian predators. Instead the authors found mammalian predators attacked more conspicuous salamanders and avoided unfamiliar salamander colors. Lastly, a study by Hantak et al. (2019; Chapter 3) found that a combination of gene flow, genetic drift, and selection appear to play a role in the distribution of color morph frequencies, but the details of the selective pressures remain unknown.

Conclusions

A plethora of studies on the P. cinereus polymorphism have been instrumental in developing our understanding of the complex nature of this phenomenon. However, the ecological and evolutionary mechanisms maintaining the color polymorphism remain 23 unclear. It is well documented that selection pressures differ across both space and time

(Brodie & Brodie, 1990; Brodie et al., 2002; Thompson, 2005; Calsbeek et al., 2012;

Siepielski et al., 2009, 2013). With the widespread geographic distribution of the two morphs, it is possible that populations may have evolved in a complex spatial mosaic of ecological and evolutionary processes (Thompson & Cunningham, 2002).

To identify how and why morphs are maintained, future studies should examine trait variation in a geographic context because a single population at a single point in time only provides a snapshot of the ecological and evolutionary dynamics impacting a trait.

In addition, because P. cinereus is a small bodied, lungless salamander that generally requires cool, moist microhabitats, it is imperative to consider scale when examining subtle morph differences (Farallo & Miles, 2016). Coarse resolution landscape surfaces and global temperature data may not be reflective of the abiotic conditions these salamanders are experiencing. Thus, we recommend conducting ecological studies across multiple local populations when investigating trait divergence in P. cinereus.

Future work on the striped/unstriped polymorphism would also benefit from a phylogenetic/phylogeographic framework, as theoretical arguments and empirical evidence has indicated an association between polymorphism and speciation (West-

Eberhard, 1986; Corl et al., 2010; Hugall & Stuart-Fox, 2012). Within the genus

Plethodon, the striped/unstriped color polymorphism has either independently originated up to 10 times or has been lost in up to 14 species that have become fixed for a single morph (described above). We currently know little regrading co-adapted sets of traits in species other than P. cinereus (but see Fisher-Reid & Wiens, 2015). Future studies should 24 examine whether co-adapted traits, such as territoriality, metabolism, or microhabitat preference, have repeatedly evolved in striped and unstriped salamanders under similar selective contexts. In addition, Plethodon provides an ideal system to test whether color polymorphism is associated with increased rates of diversification, which has been suggested in other systems (Hugall & Stuart-Fox, 2012). If a morph becomes fixed in a population, there may be rapid phenotypic evolution towards specialization for the remaining morph, which can promote reproductive isolation across populations (West-

Eberhard, 1986; Corl et al., 2010). The existence of multiple morphs promoting speciation is important because polymorphisms are common in nature (reviewed in Gray

& McKinnon, 2007; McKinnon & Pierotti, 2010; McLean & Stuart-Fox, 2014). Future work on this enigmatic polymorphism will provide essential data for understanding the ecological and evolutionary basis for the evolution of color morphs and the relationship between polymorphism and diversity.

CHAPTER 2: PREDATOR PERCEPTION ACROSS SPACE AND TIME: RELATIVE

CAMOUFLAGE IN A COLOR POLYMORPHIC SALAMANDER

Introduction

Within populations, individuals that effectively blend into the background of their natural environment have a lower probability of being discovered by visual predators

(Cott, 1940; Ruxton, et al., 2004). However, camouflage is context dependent, and distinct colors can vary in their relative conspicuousness as a function of where and when predation occurs, as well as between the divergent visual systems of diverse predators

(Cain & Sheppard, 1954; Endler, 1980, 1990; Endler & Greenwood, 1988). Thus, coevolutionary dynamics between predators and prey often vary over space and time, promoting the evolution of phenotypic polymorphisms within and between populations

(Thompson, 2005; Bond, 2007; Klomp et al., 2014).

Color polymorphism is the presence of two or more distinct, genetically determined color morphs within a single interbreeding population (Ford, 1945; Huxley,

1955). How and why distinct morphs evolve is an active field of study, but visually mediated predation is often postulated to maintain many color polymorphisms (Punzalan et al., 2005; Fitzpatrick et al., 2009; Farallo & Forstner, 2012). One commonly hypothesized mechanism for the maintenance of color morphs within populations is a form of negative frequency-dependent selection known as apostatic selection, in which common morphs are preyed upon more frequently and rarer morphs are preyed upon less frequently as a consequence of the specific search image of a predator (Clarke, 1962,

1969; Greenwood, 1984; Allen et al., 1998; Bond, 2007). Similarly, neophobia, the fear 26 of unfamiliar stimuli, can result in negative frequency-dependent selection (Sherratt,

2011; Aubier & Sherratt, 2015; Crane & Ferrari, 2017). Variation in selection in space and time, combined with gene flow among populations, can also maintain a polymorphism. For example, alternative color morphs may be adapted to match different backgrounds (Endler, 1980; Bond & Kamil, 2006), and may differ in their relative camouflage as a function of habitat, season and light conditions (Endler, 1990).

Consequently, the processes that maintain color morphs are likely to occur in a geographic mosaic (Thompson, 2005; Calsbeek et al., 2012).

The Eastern Red-backed Salamander, Plethodon cinereus, has a widespread distribution in northeastern North American forests and is the most abundant terrestrial vertebrate in the northeastern United States (Burton & Likens, 1975). There are two common color morphs of P. cinereus, a striped morph and a lead (or ‘unstriped’) morph

(Highton, 1959). The striped morph has a red stripe running down the center of a black dorsum, whereas the lead morph lacks the red dorsal stripe. The morphs are distinct in dorsal coloration and were originally described as separate species (Highton, 1959).

Many populations of P. cinereus are polymorphic, with varying morph frequencies, or are monomorphic for the striped morph; however, monomorphic lead populations are rare

(Moore & Ouellet, 2014, 2015; Hantak et al., 2015). The striped/lead dimorphism is common within the genus Plethodon, with at least ten species exhibiting this polymorphism (Highton, 2004).

Despite the number of studies on the striped and lead color morphs of P. cinereus

(reviewed in Anthony & Pfingsten, 2013), we know little with regard to the function of 27 the distinct dorsal colors (but see Fitzpatrick et al., 2009; Kraemer et al., 2016). Many elements of the behaviour and ecology of the morphs appear to differ, including metabolic rate, temperature associations, diet, stress levels, tail breakage rates, territoriality and disease prevalence (Lotter & Scott, 1977; Moreno, 1989; Venesky &

Anthony, 2007; Reiter et al., 2014; Paluh et al., 2015; Venesky et al., 2015). In addition, the morphs have been shown to mate assortatively by color in at least one population

(Anthony et al., 2008; Acord et al., 2013). These studies demonstrate that the striped morph is usually the competitive dominant. However, no studies on P. cinereus have examined multiple populations over multiple seasons and years, which is essential for quantifying whether differing selection across space and time contributes to the maintenance of the polymorphism.

A number of evolutionary processes could operate to maintain the polymorphism, such as social selection, sexual selection and predator–prey interactions. In this study, we tested hypotheses about how the color polymorphism in P. cinereus influences relative camouflage to potential predators. To investigate these questions, we used a spectrometer to quantify reflectance values of salamander body regions, as well as the distribution of colors on the forest floor. The degrees of relative camouflage of the two morphs were tested under divergent visual systems (avian, snake, mammalian), seasonal variation (fall vs. spring) and lighting conditions (direct sunlight vs. forest shade). First, we evaluated whether dorsal salamander colors and background (leaf litter and soil substrate) colors vary across seasons and populations. We predicted morph and soil colors would remain consistent, but leaf litter colors would vary by season and population. Second, we tested 28 whether different body regions (dorsum, venter and side) of the morphs were differentially camouflaged to diverse vertebrate predators. Salamander sides may be viewed by snakes and small mammals, and examining the venter may be of ecological relevance because individuals within Plethodon occasionally flip onto their backs when threatened (Brodie, 1977, pers. observ.). Finally, we quantified whether seasonal and spatial variation in background colors alters detectability of the morphs. We predicted striped morphs would be more camouflaged than lead morphs in the fall due to a greater abundance of newly fallen leaves with a high proportion of red colors, and that lead morphs would be more camouflaged against relatively decomposed spring leaf litter and soil substrates compared to striped morphs.

Methods

Salamander Sampling and reflectance measurements

Individuals of P. cinereus were collected from three populations in northern Ohio that vary in morph frequency: Squire Valleevue Farm (Squire; 100% striped), Manatoc

Scout Reservation, directly adjacent to Cuyahoga Valley National Park (CV; 80% striped) and the Heineman property on South Bass Island (SBI; >99% lead; Hantak et al.,

2015). In fall 2014 and spring 2015, 15 individuals/morph (120 total) were collected from each population. All salamanders were euthanized with tricaine methanesulfonate (MS-

222). Immediately following, ten reflectance measurements were recorded along the mid- sagittal section of the dorsum of each salamander, encompassing the area where a dorsal stripe could be present (Fig. 2.1). In addition, five reflectance measurements were recorded along the mid-sagittal region of the venter, as well as three measurements from 29 the side of each salamander (spring field season only). Spectral reflectance measurements were recorded with an Ocean Optics Jaz UV/Vis spectrometer (Model EL 200) with a

Jaz-PX Xenon light source and a QR400-ANGLE-UV reflectance probe. The light probe was fitted with a Mikopark CSH-45° holder to reduce specular reflectance, standardize reflectance measurements and exclude ambient light (Endler, 1990). Each reflectance spectrum was measured in 1 nanometer (nm) intervals from 300 to 700 nm. Dorsal, ventral and side reflectance measurements from each salamander were averaged and smoothed by body region for each individual.

At each of the three sites, 100 reflectance measurements were collected from spring leaf litter and fall leaf litter (600 measurements total) within 1 m of where a salamander was located, and from soil under cover objects in the spring where salamanders were found (5–10 soil measurements/salamander, 300 measurements total;

Fig. 2.1). These substrate types represent backgrounds on top of which P. cinereus could be viewed by a predator. All spring and fall leaf litter measurements were obtained 2 days following rainfall. Individuals of P. cinereus forage in leaf litter during moist conditions

(Jaeger, 1980); therefore, this represents an optimal time for visual predators to discover salamanders (Kuchta, 2005; Venesky & Anthony, 2007).

To describe color, we calculated mean brightness, hue and chroma (hereafter termed ‘color’) from reflectance measurements (Endler, 1990; Andersson & Prager,

2006; Kemp et al., 2015). Brightness is the total intensity of light that is reflected; hue roughly corresponds to the verbal definition of color and was calculated as the wavelength at peak reflectance; and chroma is a measure of color saturation and was 30 calculated as the relative difference between the maximum and minimum reflectance while taking into account mean brightness (Maia et al., 2013). These measures of color are predator independent and were used to quantify color variability. Color variables were obtained using the R (version 3.2.2, R Core Team, 2015) package pavo, version

0.5.2 (Maia et al., 2013).

Visual models

Visual models include reflectance values from the body regions of salamanders, background reflectance, predator spectral sensitivities and irradiance. To quantify how well potential predators discriminate between the morphs of P. cinereus given a set of background colors and ambient light conditions, we applied the visual model developed by Vorobyev et al. (1998). Color distances were obtained by calculating chromatic (ΔS) and achromatic (ΔL) contrasts, which correspond to color (hue and saturation) and brightness (luminance), respectively. These calculations estimate the contrast between an object and a background in units of just-noticeable differences (JNDs), where a value of 1 approximates the minimum difference between an object and the background that is detect- able to a given predator (Vorobyev et al., 1998). Due to variation in ambient light conditions, the distance between the viewer and target, or the length of time an object is viewed by a predator, a JND of 1 is not an absolute threshold, but is useful as an approximate criterion (Kemp et al., 2015).

To quantify the level of camouflage among color morphs and populations against each background type, the spectral sensitivities of the tetrachromatic blue tit (Parus caeruleus; Hart et al., 2000), trichromatic common garter snake (Thamnophis sirtalis; 31

Sillman et al., 1997) and the dichromatic thirteen-lined ground squirrel (Ictidomys tridecemlineatus; Jacobs et al., 1985) were used. The common garter snake is a natural predator of P. cinereus, but the blue tit and thirteen-lined ground squirrel are not.

However, the blue tit visual system is well characterized and is similar to corvids, which are common predators of North American salamanders (Murray et al., 2005; Kuchta et al., 2008). Similarly, the thirteen-lined ground squirrel has a visual system that resembles several mammalian mesopredators (Jacobs et al., 1985), which also prey on salamanders

(Brodie et al., 1979; Kuchta, 2005; Anthony & Pfingsten, 2013; Kraemer et al., 2016).

Finally, two irradiance measures were used in the visual models, forest shade and standard daylight (Endler, 1993). Contrast values did not differ with irradiance type; thus, we only report values calculated using forest shade irradiance. Full visual model and contrast value calculations can be found in Maia et al. (2013).

Statistical analyses

Salamander color variables as well as seasonal and population differences in substrate colors were compared using analyses of variance (ANOVAs; Supporting

Information, Table 2.S1) and multivariate analyses of variance (MANOVAs). We used a full factorial ANOVA to test for color morph, seasonal and population differences in chromatic and achromatic contrast, which were calculated using avian, snake and mammalian visual models against the three substrate types. We ran separate models for chromatic and achromatic contrasts, and for avian, snake and mammalian visual models.

Significance was assessed with Tukey’s HSD post hoc tests. Tetrahedral (four cones), trichromatic (three cones) and dichromatic (two cones) color space models, which are 32 measures of color (hue and saturation) overlap, were created to illustrate the visibility of morphs against background types given avian, snake and mammalian predators

(Goldsmith, 1990; Stevens et al., 2009). All statistical analyses were conducted in R version 3.2.2.

Results

Variation in salamander and background colors

Combined colorimetric variables of salamander dorsal measurements differed between seasons and populations (F3,112 = 4.9, P < 0.001; Fig. 2.1). In addition, the color of salamander sides differed among populations (F3,56 = 4.7, P < 0.001), and the color of ventral measurements differed by season and population (F3,112 = 2.6, P = 0.008).

Accordingly, salamanders from fall and spring and among populations were kept separate for subsequent analyses. Spring and fall leaf litter differed by season and population

(F2,114 = 20.2, P < 0.001; Fig. 2.1), and soil substrate differed among populations (F2,57 =

7.7, P < 0.001; Fig. 2.1). Thus, spring leaf litter, fall leaf litter and soil substrate from each population and season were also kept separate for subsequent analyses.

Salamander camouflage

From the perspective of avian and snake predators, the dorsum, side and venter of all salamanders were discriminable in chromatic and achromatic color space against all background types (JND > 1; Fig. 2.2; Supporting Information, Tables 2.S1–2.S3). From the perspective of a mammalian predator, the chromatic contrast of the dorsum of the lead morphs was discriminable against all background types (JND > 1), whereas striped morphs did not stand out against any background type (JND < 1; Fig. 2.2C). Dorsal 33 achromatic contrast was discriminable against all background types for both color morphs (Supporting Information, Fig. 2.S1C). Chromatic and achromatic contrasts of all salamander sides and venters were discernable against all background types (Supporting

Information, Figs 2.S2, 2.S3).

Dorsal contrasts: avian visual model

Dorsal contrasts using the avian visual model differed between background type

2 and population for both chromatic contrast (F6,168 = 24.57, P < 0.001, adj. R = 0.64) and

2 achromatic contrast (F6,168 = 25.24, P < 0.001, adj. R = 0.67). Pairwise comparisons revealed that striped individuals exhibited lower chromatic contrast compared to lead morphs against spring and fall leaf litter, but there was no difference in morph conspicuousness against soil substrate (Fig. 2.2A; Table 2.1A). Achromatic contrast varied by population (Supporting Information, Fig. 2.S1A; Table 2.1A). Tetrahedral color space plots, which demonstrate the visual overlap between morphs and background types at each population, illustrate the similarity between striped individuals and spring and fall leaf litter, whereas both morphs show similar overlap with the soil substrate (Fig. 2.3A).

Dorsal contrasts: snake visual model

Dorsal contrasts using the snake visual model differed between background type

2 and population for both chromatic contrast (F6,168 = 6.24, P < 0.001, adj. R = 0.32) and 2 achromatic contrast (F6,168 = 26.18, P < 0.001, adj. R = 0.63). Pairwise comparisons revealed that striped morphs exhibited lower chromatic contrast compared to lead morphs from SBI against spring leaf litter (Fig. 2.2B; Table 2.1B). Striped morphs from Squire were more camouflaged than striped and lead morphs from CV against fall leaf litter 34

(Fig. 2.2B; Table 2.1B). There was no difference in conspicuousness between morphs or populations against soil substrates (Fig. 2.2B; Table 2.1B). Dorsal achromatic contrast varied by background type and population (Supporting Information, Fig. 2.S1B; Table

2.1B). Trichromatic color space plots illustrate similar color overlap between striped and lead morphs and soil substrate (Fig. 2.3B). However, striped morph color overlaps more with spring leaf litter, and striped morphs from Squire overlap more with fall leaf litter, whereas lead morph colors overlap less with these background colors (Fig. 2.3B).

Dorsal contrasts: mammalian visual model

Dorsal contrasts using the mammalian visual model differed by background type

2 and population for both chromatic contrast (F6,168 = 14.70, P < 0.001, adj. R = 0.81) and 2 achromatic contrast (F6,168 = 21.03, P < 0.001, adj. R = 0.61). Pairwise comparisons revealed that striped morph dorsal chromatic contrast was lower than lead morph contrast

(Fig. 2.2C; Table 2.1C). Striped morphs were indistinguishable from the backgrounds they appeared against, but lead morphs were not (Fig. 2.2C). The achromatic contrast of striped and lead morphs varied by population (Supporting Information, Fig. 2.S1C; Table

2.1C). Dichromatic color plots demonstrate the high degree of color overlap between striped morphs and all background types, whereas lead morphs did not overlap with any background type (Fig. 2.3C).

Side and ventral contrasts: avian visual model

Chromatic contrast of salamander sides differed by background type and

2 population (F3,112 = 2.97, P = 0.035, adj. R = 0.69); however, achromatic contrast did not 2 differ (F3,112 = 1.64, P = 0.183, adj. R = 0.34). Pairwise comparisons show that against 35 soil substrate and spring leaf litter chromatic contrasts of lead morph sides from SBI were significantly less camouflaged than lead morph sides from CV and all striped morphs

(Supporting Information, Fig. 2.S2A, Table 2.S2A). No other comparisons of salamander sides differed.

The chromatic contrast of salamander venters differed by background type and

2 population (F6,168 = 83.64, P < 0.001, adj. R = 0.94), as did achromatic contrast (F6,168 =

50.19, P < 0.001, adj. R2 = 0.71). Overall, pairwise comparisons of the ventral chromatic and achromatic contrasts of lead morphs from SBI were less camouflaged against soil substrate and spring leaf litter (Supporting Information, Fig. 2.S3A, Table 2.S3A).

Conversely, against fall leaf litter, the venters of lead morphs from SBI were more camouflaged than lead morphs from CV and striped morphs (Supporting Information,

Fig. 2.S3A, B, Table 2.S3A).

Side and ventral contrasts: snake visual model

Chromatic contrasts of salamander sides differed by background type and

2 population (F3,112 = 3.57, P = 0.016, adj. R = 0.64); however, achromatic contrast did not 2 differ (F3,112 = 2.48, P = 0.065, adj. R = 0.46). Pairwise comparisons showed that against soil substrate the chromatic contrasts of lead morph sides from SBI were significantly less camouflaged than lead morphs from CV and all striped morphs (Supporting

Information, Fig. 2.S2C, Table 2.S2B). Lead morph sides from SBI were less camouflaged against spring leaf litter compared to lead and striped morphs from CV

(Supporting Information, Fig. 2.S2C, Table 2.S2B). Striped morphs from Squire were 36 less camouflaged than striped morphs from CV against spring leaf litter (Supporting

Information, Fig. 2.S2C, Table 2.S2B).

Chromatic contrasts of salamander venters differed by background type and

2 population (F6,168 = 47.54, P < 0.001, adj. R = 0.84), as did achromatic contrasts (F6,168 =

69.42, P < 0.001, adj. R2 = 0.81). In general, pairwise comparisons revealed that the ventral chromatic and achromatic contrasts of lead morphs from SBI were less camouflaged against soil substrate and spring leaf litter (Supporting Information, Fig.

2.S3C, D, Table 2.S3B). However, the venter of lead morphs from SBI was more camouflaged than lead morphs from CV and striped morphs against fall leaf litter

(Supporting Information, Fig. 2.S3C, D, Table 2.S3B).

Side and ventral contrasts: mammalian visual model

The chromatic contrasts of salamander sides differed by background type and

2 population (F3,112 = 4.51, P = 0.005, adj. R = 0.66), as did achromatic contrasts (F3,112 =

6.40, P < 0.001, adj. R2 = 0.63). In general, pairwise comparisons of salamander side chromatic and achromatic contrasts revealed that lead morphs from SBI were conspicuous against soil substrate and spring leaf litter, but there was no difference in chromatic contrast between lead morphs from CV and SBI against soil substrate

(Supporting Information, Fig. 2.S2E, F, Table 2.S2C).

2 Both chromatic (F6,168 = 75.22, P < 0.001, adj. R = 0.93) and achromatic (F6,168 =

92.00, P < 0.001, adj. R2 = 0.86) contrasts of salamander venters differed by background type and population. Pairwise comparisons of ventral chromatic and achromatic contrasts showed that lead morphs from SBI were less camouflaged against soil substrate and 37 spring leaf litter (Supporting Information, Fig. 2.S3E, F, Table 2.S3C). The chromatic and achromatic contrasts of the venter of lead morphs from SBI against fall leaf litter were more camouflaged than lead morphs from CV and striped morphs (Supporting

Information, Fig. 2.S3E, F, Table 2.S3C).

Discussion

Geographic variation is an intrinsic part of the evolutionary process, driven by differences in abiotic and biotic conditions, including co-evolving interactions among morphs (Thompson, 2005). However, despite steady interest in the biology of color polymorphisms, we still have a poor understanding of how spatial and temporal processes impact the evolutionary dynamics of color morphs within and among populations (Gray

& McKinnon, 2007; Corl et al., 2010; Hugall & Stuart-Fox, 2012), including how commonly morphs are exposed to divergent selection pressures (Gosden & Svensson,

2008; Calsbeek et al., 2012). With reflectance measurements of the polymorphic salamander, P. cinereus, and three background types, we examined whether spatial and temporal differences in coloration influenced visual perception of the morphs by avian, snake and mammalian predators. We found that salamander and background colors varied across populations and seasons. An unexpected finding in our study was that, in general, lead morphs were more conspicuous to visual predators. However, we also found there was spatial and temporal variation in the relative degree of morph camouflage.

In P. cinereus, the striped/lead polymorphism is genetic with a simple genetic architecture (Highton, 1959, 1975). However, individual colors differ within and among populations, which can be due to genetic differences, environmental differences or an 38 interaction between genes and the environment. Using reflectance measurements of P. cinereus, we found that the average coloration of populations varied between spring and fall. The striped morph expressed a brighter red coloration in the fall, whereas the lead morph was brighter in the spring. We did not anticipate these seasonal alterations in color; however, a study by Kraemer et al. (2012) demonstrated gradual color change in the striped and erythristic (entirely orange-red) morphs of P. cinereus in captivity. The authors suggested color change may have occurred due to diet, stress or natural seasonal changes; our study suggests that natural seasonal changes may have played a role.

Seasonal changes in color are a common biological phenomenon, and have been documented in many organisms, including insects (Tauber et al., 1986), birds (Delhey et al., 2006), mammals (Aldous, 1937; Caro, 2009), reptiles (Johnston, 1994; Boback &

Siefferman, 2010) and frogs (Wente & Phillips, 2003).

In our study of camouflage, we found that the striped morphs were more camouflaged in dorsal coloration compared to lead morphs to the avian, snake and mammalian predators. However, there was spatial variation in the degree of morph camouflage. For example, the avian and snake visual models were unable to discriminate between striped and lead morphs against soil substrate. In addition, striped morphs from

Squire were more camouflaged to snakes than striped morphs from CV or lead morphs against fall leaf litter.

Across seasons and populations, our results indicate that when visual predators are the agents of selection, striped morphs benefit from better camouflage relative to lead morphs, suggesting that relative camouflage may not play a role in the maintenance of 39 this polymorphism. This raises the question of why the striped morph does not go to fixation. It may be that predator-mediated selection does not play an important role in the maintenance of the color polymorphism, but rather the polymorphism is under, for example, strong social or sexual selection. Another possible explanation is that apostatic selection or neophobia plays a role in maintaining the polymorphism. For example, using clay model replicas of striped and lead morphs of Plethodon in an experimental setup,

Fitzpatrick et al. (2009) found that rare morphs benefitted from lower rates of avian predation. By contrast, another clay model study of predation on P. cinereus did not find evidence for apostatic selection by mammalian predators (Kraemer et al., 2016). Such incongruence may be a by-product of different experimental methods. For instance, the salamander replicas used by Fitzpatrick et al. (2009) contained a food reward on the underside, and were conducted in an open field, whereas Kraemer et al. (2016) deployed clay models without a reward in forested habitats. Additional studies within a geographic framework may aid our understanding of the role of predator-mediated selection in the maintenance of the striped/lead polymorphism.

In our study, much of the variation in contrast values among populations was due to differing substrate colors in different populations (Figs. 2.1, 2.3). Soil composition in

Ohio consists of 12 distinct series, which are defined by combinations of soil attributes.

Our CV and Squire sites are in soil series 8 (Mahoning-Canfield-Rittman-Chili), whereas

SBI is in soil series 1 (Hoytville-Nappanee-Paulding-Toledo; ODNR, 2017). Common tree species at the three study locations vary as well. Sugar maple (Acer saccharum) is common at all three sites, but American beech (Fagus grandifolia) and tulip poplar 40

(Liriodendron tulipefera) are common at CV and Squire. Red oak (Quercus rubra) is also common at CV, whereas shagbark hickory (Carya ovata) and red maple (Acer rubrum) are abundant at Squire. Conversely, common hackberry (Celtis occidentalis), American basswood (Tilia americana) and the invasive Amur honeysuckle (Lonicera maackii) dominate SBI. These different assemblages of tree species create spatial variation in leaf litter colors.

Our study is not the first to find that morph-specific predation risk is highly dependent on background colors (Forsman et al., 2011; Karpestam et al., 2014; Kraemer

& Adams, 2014). For instance, Karpestam et al. (2013) found predator perception of the divergent color morphs of the Pygmy Grasshopper (Tetrix subulata) was strongly dependent on whether the habitat was burnt, unburnt or intermediate. They also found a correlation between predator detection rates and color morph frequencies. In P. cinereus, striped morphs are typically at a higher frequency in polymorphic populations, and are more commonly fixed within populations (Moore & Ouellet, 2015), suggesting specific settings are required for the maintenance of the lead morph. For example, the fitness of lead morphs may be more dependent on behavioral or ecological attributes. A study by

Fisher-Reid et al. (2013) found that lead morphs from monomorphic populations on Long

Island, New York, have one more costal groove than striped morphs, and elongation in salamanders is associated with increased fossoriality (Wake, 1966). In addition, other factors, such as temperature, parasitism and competition between the morphs, have been shown to be correlated with morph frequencies in several taxa (Gibbs & Karraker, 2006;

McLean & Stuart-Fox, 2014). With the wide range of documented differences in ecology, 41 behaviour and coloration between the morphs of P. cinereus, it may be that multiple evolutionary processes interact to contribute to the maintenance of this polymorphism

(Merilaita, 2001; McLean & Stuart-Fox, 2014).

Conclusions

How color morphs are maintained within and among populations is a long- standing question in evolutionary biology (Ford, 1945; Huxley, 1955). Recent studies of geographic variation in color polymorphic species have aimed to elucidate the role of geography in divergence, including species formation (Corl et al., 2010; Ozgo, 2011;

Hugall & Stuart-Fox, 2012; McLean & Stuart-Fox, 2014). Our study demonstrates the importance of studying polymorphism in a geographic framework. A single population at a single point in time provides a snapshot of the ecological and evolutionary dynamics impacting a trait; however, many evolutionary interactions vary over space and time, and a consideration of larger spatial scales is often required to fully understand the evolutionary dynamics involved in trait evolution (Brodie et al., 2002; Thompson, 2005;

Gosden & Svensson, 2008; Siepielski et al., 2009; Kuchta & Wake, 2016). Future studies of predator-mediated selection, relative camouflage, social interactions and gene flow, preferably in a geographic framework and over multiple years, would be beneficial in deciphering the role of selection in the maintenance of the striped/lead color polymorphism in P. cinereus.

Tables

Table 2.1 (a). Tukey’s HSD post hoc test P-values for dorsal P. cinereus (A) avian, (B) mammalian, and (C) snake contrast values. Comparisons below the diagonal of each substrate type are chromatic contrasts, whereas values above the diagonal are achromatic contrasts. Significant comparisons are in bold. Fall Leaves

A CV Striped CV Lead Squire Striped SBI Lead

CV Striped -- 1.000 <0.001 0.713 CV Lead <0.001 -- <0.001 0.727

Squire Striped 0.106 <0.001 -- <0.001

SBI Lead <0.001 <0.001 <0.001 -- B CV Striped -- 0.003 <0.001 0.031 CV Lead <0.001 -- 1.000 1.000

Squire Striped 1.000 <0.001 -- 0.944

SBI Lead <0.001 <0.001 <0.001 -- C CV Striped -- 0.527 <0.001 1.000 CV Lead 1.000 -- 0.279 0.842

Squire Striped 0.006 <0.001 -- 0.001

SBI Lead 0.59 0.135 0.816 --

Table 2.1 (b). Spring Leaves

A CV Striped CV Lead Squire Striped SBI Lead

CV Striped -- <0.001 0.655 <0.001 CV Lead 0.002 -- <0.001 <0.001 Squire 1.000 0.032 -- 0.086 Striped SBI Lead <0.001 0.014 <0.001 -- B CV Striped -- <0.001 0.684 <0.001 CV Lead <0.001 -- <0.001 <0.001 Squire 1.000 <0.001 -- <0.001 Striped SBI Lead <0.001 0.039 <0.001 -- C CV Striped -- <0.001 0.726 <0.001 CV Lead 0.59 -- <0.001 <0.001 Squire 1.000 0.553 -- <0.001 Striped SBI Lead 0.001 0.512 0.001 --

Table 2.1 (c). Soil Substrate

A CV Striped CV Lead Squire Striped SBI Lead

CV Striped -- <0.001 0.98 0.134 CV Lead 1 -- <0.001 <0.001

Squire Striped 0.538 0.276 -- 0.883

SBI Lead 0.877 0.639 1 -- B CV Striped -- <0.001 0.237 0.971 CV Lead <0.001 -- 0.454 <0.001

Squire Striped 1 <0.001 -- 0.004

SBI Lead <0.001 0.972 <0.001 -- C CV Striped -- <0.001 0.675 0.998 CV Lead 1 -- <0.001 <0.001

Squire Striped 1 1 -- 0.996

SBI Lead 0.82 0.384 0.87 --

Figures

Figure 2.1. (A), representative lead and striped morphs of Plethodon cinereus. (B–D) spectral reflectance of P. cinereus dorsal colors by season, population and morph. (B) striped and lead morphs from CV, (C) striped morphs from Squire and (D) lead morphs from SBI. Lead morphs nearly overlap completely. (E–G) spectral reflectance of substrate colors by season and population. (E) CV, (F) Squire, (G) SBI. For all graphs, the center line represents the mean spectral reflectance, and the shading around the line represents the standard error. 46

Figure 2.2. Dorsal chromatic contrast values for the avian (A), snake (B) and mammalian (C) visual systems. Bars show means (±SE). Contrast values that lie below the grey horizontal line represent groups that are indistinguishable to the predator. Y-axes values vary by plot. 47

Figure 2.3. (A) tetrahedral color space plots representing the visual perspective of an avian predator. Hue is estimated from the angle of each point to the blue (short; s), green (medium; m), red (long; l) and UV (u) cone color channels (x-y-z axes), and saturation is measured as the distance from the achromatic origin (blue center star) to each individual point. (B) trichromatic color space plots representing the visual perspective of a snake predator. Hue is estimated by short (s) and medium (m) and long (l) cone color channels, and saturation is measured as the distance from the achromatic origin (blue center star). (C) dichromatic color space plots representing the visual perspective of a mammalian predator. Hue is estimated by short (s) and medium/long (m/l) cone color channels, and saturation is measured as the distance from the achromatic origin (blue center star). For (A–C), each plot represents a population, and each point represents the color of an individual salamander or an average of 5–10 substrate measurements to a visual predator.

CHAPTER 3: DO GENETIC STRUCTURE AND LANDSCAPE HETEROGENEITY

IMPACT COLOR MORPH FREQUENCY IN A POLYMORPHIC SALAMANDER?

Introduction

Landscape genetics combines landscape ecology, population genetics, and spatial statistics to quantify the relative contribution of environmental and landscape features on patterns of genetic connectivity (Manel et al., 2003; Manel & Holderegger, 2013).

Heterogeneity across a landscape can greatly impact population connectivity, particularly for dispersal-limited species (Spear et al., 2005; Wang, 2009, Peterman & Semlitsch,

2013). Population genetic structure may vary as a consequence of environmental preferences, mate selection, species interactions or the demographic history of populations (Slatkin, 1987; Storfer et al., 2007). In addition, phenotypic divergence can influence genetic structure and gene flow, which may result in reproductive isolation and speciation (Wang & Summers, 2010); however, the ubiquity of this relationship is unknown.

Color polymorphic species, where two or more discrete color morphs coexist within a population, represent an ideal system to investigate the ecological and evolutionary processes that contribute to population divergence (Ford, 1945). Although color polymorphism is a classic subject in evolutionary biology (Cain & Sheppard, 1954;

Endler, 1978), little attention has been given to the geographic context of polymorphisms, with most studies focusing on in-depth investigations of single populations (Sinervo &

Lively, 1996; Svensson et al., 2001; Hantak et al., 2016; but see Corl et al., 2010; Davis

Rabosky et al., 2016 for notable exceptions). Morphs represent contrasting character sets 49 that are the consequence of multivariate disruptive natural selection acting on the same genome (Sinervo & Svensson, 2002), and are important, in part because they allow a single species to occupy multiple niches within a single population (West-Eberhard,

1986, Forsman et al., 2008). Evolutionary processes that operate to maintain polymorphisms include overdominance, negative frequency-dependent selection, spatiotemporal variation in selection, and gene flow among populations (Allen, 1988;

Gray & McKinnon, 2007). Gene flow is noteworthy because it can maintain a polymorphism despite diversifying selection among populations (Slatkin, 1987;

Sandoval, 1994). In contrast, when gene flow is low, geographic distance, genetic drift, natural selection or a combination of these mechanisms can promote morph frequency divergence (including fixation) among populations (Sandoval, 1994).

Here we examine how genetic differentiation, landscape features, gene flow and genetic drift impact color morph frequency among populations of the polymorphic eastern red-backed salamander, Plethodon cinereus. Within the salamander family

Plethodontidae, the genus Plethodon contains 10 species that display a striped/unstriped color polymorphism (Highton, 2004). The ‘striped’ morph possesses a red stripe overlaid on a black dorsum, whereas the ‘unstriped’ morph has a completely black dorsum

(Highton, 1962; Fig. 3.1a). The color morphs are divergent along multiple niche axes, including dietary composition, temperature associations, metabolic rate, territorial behavior and mating interactions (summarized in Anthony & Pfingsten, 2013; see also

Acord et al., 2013; Reiter et al., 2014; Paluh et al., 2015; Stuczka et al., 2016). It is likely that correlational selection has acted on suites of traits to produce the covariant, 50 differentially adapted character sets that define the morphs (Sinervo & Svensson, 2002).

Plethodon cinereus provides an ideal opportunity for studying polymorphism in a geographic framework as both morphs are common, population sizes are large, and dispersal is limited (Burton & Likens, 1975; Liebgold et al., 2011).

Our study is focused on a set of populations in post-glacial, northern Ohio that display unusually high variation in morph frequency, including populations that are monomorphic for both morphs (Fig. 3.1a). Using microsatellite loci, we quantified genetic structure among our sampled populations. Then, we tested whether morph frequency, geographic distance, or landscape variables better explain patterns of genetic variation in P. cinereus. The usual expectation, and our null hypothesis, is isolation by distance; however, genetic differentiation in amphibians is often associated more strongly with phenotypic variation than with geography (Funk et al., 2009; Wang & Summers,

2010). Geographic distance, genetic variation, and landscape variables can also directly influence geographic variation in polymorphisms (McLean et al., 2015); thus, we examined the contribution of these factors in influencing patterns of morph frequency variation in P. cinereus. Lastly, we tested whether gene flow and/or genetic drift impact the distribution of color morphs in northern Ohio. As morphs of P. cinereus compete over resources (Anthony et al., 2008, 2017), morphs may be adapted to their local environments, rendering populations differing in morph frequencies divergent in ecology and behavior. Accordingly, we hypothesized reduced gene flow between populations that are more divergent in morph frequency. 51

Methods

Population sampling and laboratory techniques

We collected tissue from small salamander tail tips (Cabe et al., 2007) from 648 individuals of Plethodon cinereus from 28 populations in northern Ohio between spring

2015 and fall 2016 (Fig. 3.1a). Tissue samples were immediately preserved in 95% ethanol. Genomic DNA was extracted using Qiagen DNeasy tissue kits following the manufacturer’s protocol. We amplified 10 microsatellite loci, with tri-pentanucleotide repeat motifs, developed for P. cinereus (Pc3, Pc7, Pc15– Pc17, Pc25, Pc28, Pc30, Pc34,

Pc37; Cameron et al. 2017) using PCR. Forward primers were modified with a 5′ florescent tag of 6-FAM, NED, HEX, PET or ATTO 565 (NHS Ester), and were multiplexed in an arrangement of 3–4 loci. PCR products were run on an ABI 3730 DNA

Analyzer using a LIZ 600 size standard at the Arizona State University CLAS DNA

Laboratory. Scoring and binning were done in Geneious ver. 9.1.8 (Kearse et al., 2012).

Population genetic analyses

We used MICRO-CHECKER ver. 2.2 (van Oosterhout et al., 2004) to check for scoring errors and used 10 000 Monte Carlo simulations to test for the presence of null alleles. The R package genepop (Rousset et al., 2008) was used to test for linkage disequilibrium among pairs of loci and to identify whether loci within populations conformed to Hardy–Weinberg equilibrium (HWE). Allele number (NA), observed heterozygosity (HO), and expected heterozygosity (HE) were calculated in GENALEX ver. 6.5 (Peakall & Smouse, 2012). Rarefied allelic richness (AR) and rarefied private allelic richness (pAR) were calculated in HP-RARE (Kalinowski, 2005). We used 52

GENODIVE ver. 2.0 (Meirmans & van Tienderen, 2004) to calculate pairwise FST, standardized FST (F’ST) and Jost’s D values (Dest; Jost, 2008) between localities.

We used STRUCTURE ver. 2.3 (Pritchard et al., 2000) to identify genotypic clusters from sampled localities. Structure was run from K = 1 to K = 28 populations, with a burn-in of 250,000 and 600,000 Markov chain Monte Carlo (MCMC) iterations after burn-in. Each K was replicated 10 times with random starting seeds. We used

STRUCTURE HARVESTER ver. 0.6.94 (Earl & vonHoldt, 2012) to obtain ∆K using the

Evanno method, which is based on the rate of change in the log probability of data between successive K-values (Evanno et al., 2005, Janes et al., 2017). To find optimal alignments from the 10 iterations for each cluster we used the program CLUMPP ver. 1.1

(Jakobsson & Rosenberg, 2007). Results from STRUCTURE and CLUMPP were visualized using DISTRUCT ver. 1.1 (Rosenberg, 2004). As the Evanno method recovers the basal level of genetic structure, this procedure was replicated iteratively within clusters to detect substructure (Janes et al., 2017). We also used a discriminant analysis of principal components (DAPC) to test for population structure using the R package adegenet (Jombart, 2008). In contrast with STRUCTURE, DAPC investigates patterns of population differentiation without assuming a model of evolution (Jombart et al., 2010).

We explored a range of K-values using K-means clustering and Bayesian information criteria (BIC). Cross-validation was used to determine the number of PC axes to retain

(Jombart et al., 2010).

Gene flow and genetic drift

Gene flow was estimated using MIGRATE-N ver. 3.6.11 in the CIPRES Science

Gateway ver. 3.3 (Beerli, 2008, Beerli & Palczewski, 2010). MIGRATE-N uses a coalescent framework to estimate gene flow across populations. We used a Brownian motion model with three independent runs and three replicates within each run. Runs consisted of one long chain of 10,000 generations sampled every 500 increments, with

5000 iterations excluded as burn-in. Four parallel, heated chains were used with temperatures of 1.0, 1.5, 3.0 and 1,000,000. Our large number of populations and individuals exceeded the computational limits of MIGRATE-N. Therefore, we ran

MIGRATE-N within one genetic cluster that was comprised of 19 populations that were highly variable in morph frequency; to improve speed we grouped geographically close populations with similar morph frequencies (3+5, 6+7, 10+11, 12+13, 17+18, 19+20).

Most combined sites were <10% divergent in morph frequency; the exception are sites

6+7, which differed by 30%, but we grouped these two populations because they are separated by only 3 km and had FST = 0.02. Population 21 was not included due to its high degree of admixture (Fig. 3.1b). We considered parameter estimates to be accurate if effective sample sizes were >1,000 (Converse et al., 2015).

To estimate the effect of genetic drift, we used the R package RAFM to obtain

Bayesian estimates of the admixture F model (AFM; Karhunen & Ovaskainen, 2012).

This approach assumes each of n contemporary local populations is derived from one or more evolutionarily independent lineages with a common ancestor T generations ago.

RAFM estimates one vector of parameters (alpha) and two matrices of parameters (kappa 54 and theta): alpha describes the amount of drift experienced by each of the n evolutionary lineages; kappa describes the proportional contribution of each respective lineage to each respective contemporary local population (i.e. a matrix of admixture coefficients); and theta is a pair-wise matrix of population-level coancestry coefficients. We ran RAFM with 100,000 MCMC iterations, sampled every tenth iteration, and excluded 50,000 generations as burn-in across all populations and within each cluster. We used median values across MCMC samples as best estimates of the elements within alpha and theta, and means as our best estimate for the elements of kappa.

Landscape resistance surfaces and analyses

To model dispersal in P. cinereus, we created landscape resistance surfaces based on features that are known to impact amphibian movement (Storfer et al., 2010). Digital elevation (DEM; 9.1×9.1m resolution) and tree canopy coverage (30×30m) were obtained from the U.S. Geological Survey National Map (USGS,

). Rivers, streams (30 × 30 m) and Lake Erie (9.1 × 9.1 m) were created in ARCGIS ver. 10.4.1. We classified rivers, streams, and Lake Erie as binary (presence/absence) surfaces and merged raster files to create a comprehensive

‘water-way’ surface. Environmental variables were used to create an ecological niche model (ENM) for P. cinereus. Occurrence localities (n = 795) for the ENM were obtained from online biodiversity databases (GBIF, iNaturalist, iDigBio, VertNet) and our sampling sites. Our ENM was built with the maximum entropy algorithm MAXENT ver.

3.4.1 (Phillips et al., 2006) using five uncorrelated (R≤0.70) BIOCLIM variables 55

(Supplementary material Appendix 1 Table 3.S1) at 30 arc-second (~1km) resolution

(; Hijmans et al., 2005).

To quantify the relationship between genetic distance and morph frequency with landscape resistance, we employed the R package ResistanceGA (Peterman, 2018).

ResistanceGA uses a genetic algorithm to optimize resistance surfaces, using either a series of transformations on continuous surfaces, or a classification of resistance values on categorical surfaces (Scrucca, 2013). In addition, ResistanceGA allows for resistance surface summation to create a ‘composite’ resistance surface, which we created from our

ENM, DEM, waterways, and canopy surfaces. Prior to running ResistanceGA, each resistance surface was re-sampled at 300 m resolution due to computational limitations resulting from our large number of populations and broad sampling extent (Peterman, pers. comm.; Fig. 3.2). Effective resistance between the 28 sampling localities, as well as within each genetic cluster, was measured using least cost path (LCP) distance and resistance distance (RD), as implemented in the ‘commuteDistance’ function within the R package gdistance (van Etten, 2017). LCP estimates the optimal route between two sites that minimizes the cost of moving through the landscape, whereas RD (circuit theory) assesses all possible pathways between two localities and generates a cumulative cost

(Adriaensen et al., 2003; McRae, 2006). All optimization runs were conducted twice

(Peterman 2018).

We calculated Euclidean distances between localities to generate a geographic distance matrix, and used the equation from McLean et al. (2015) to create a morph frequency distance matrix. To determine which resistance or distance matrix was most 56 closely correlated with genetic distance and morph frequency, we fit linear mixed models with maximum-likelihood population effects (MLPE) using the R package lme4 (Clarke et al., 2002, van Strien et al., 2012). Models were run separately for LCP and RD across all populations and within clusters. In the first set of models, linearized FST was the dependent variable, and scaled and centered effective resistance from each surface, morph frequency, and geographic distance were independent variables. To determine the factors directly affecting morph frequency variation, we created a second set of models where morph frequency was the dependent variable, and linearized FST, the scaled and centered effective resistance of each surface, and geographic distance were independent variables. We used simple, single predictor models due to a high degree of multicollinearity among surfaces (Peterman, 2018). Best-fit models were assessed and ranked using AICc with the R package AICcmodavg (Mazerolle, 2013).

Results

Population genetic analyses

Across the 28 populations, all microsatellite loci were polymorphic, with 2–18 alleles per locus (mean=4±0.82 SD). Observed heterozygosity ranged from 0.23 to 0.58

(mean=0.387±0.10) among populations (Supplementary material Appendix 1 Table

3.S2). Departures from HWE were found at four loci in eight sites after Bonferroni correction. MICRO-CHECKER did not indicate scoring errors, but null alleles were found at three loci in eight populations, except for Pc25, in which null alleles were found in 13 populations. No pairwise loci were in linkage disequilibrium across any populations. We retained all loci for further analyses because null alleles can have weak 57 effects on genetic structure (Chapuis & Estoup 2007). Deviations from HWE were not consistent across populations, and their elimination did not alter our results (not shown).

Genetic structure

STRUCTURE revealed a peak at ΔK = 2 (Fig. 3.1b, Supplementary material

Appendix 1 Fig. 3.S1). One cluster included sites from the center of the sampling range

(3–21); we call this the ‘Central Cluster’. Genetic admixture in the Central Cluster was low, except for site 21, and no substructure was detected within this cluster, including after removing site 21 (Supplementary material Appendix 1 Fig. 3.S2). Further tests of substructure with the inclusion of site 2 in the Central Cluster revealed no admixture

(Supplementary material Appendix 1 Fig. 3.S3). Morph frequencies vary greatly within the Central Cluster, ranging from 100% striped to 100% unstriped (Fig. 3.1). The second cluster combined localities from the western and eastern ends of our sampling. Tests of substructure recovered the eastern and western regions as distinct, with no evidence of admixture. We call these the ‘Western Cluster’ (sites 1–2) and the ‘Eastern Cluster’ (sites

22–28; Fig. 3.1b). Western Cluster sites are nearly monomorphic for the unstriped morph

(≥90%); whereas the Eastern Cluster is entirely fixed for the striped morph (Fig. 3.1).

Using BIC, 5–10 clusters were recovered using DAPC (Supplementary material

Appendix 1 Fig. 3.S4). We chose K=5 because it creates the most inclusive groups. The cross-validation test indicated that 15 principal components should be retained and we elected to include all discriminant functions. DAPC results were similar to the

STRUCTURE results: cluster 1=Western Cluster; clusters 2, 3, 5 (which widely overlap)=Central Cluster, plus one individual from site 2; cluster 4=Eastern Cluster, plus 58

13 individuals from site 21 and 1 individual each from sites 2 and 17. Thus, the

STRUCTURE and DAPC results largely correspond in recognizing three genetic clusters

(Supplementary material Appendix 1 Fig. 3.S4), with the exception of ambiguities stemming from population 21.

Genetic differentiation varied substantially among populations. FST ranged from 0 to 0.616, F’ST 0–0.835, and Dest 0–0.753 (Supplementary material Appendix 1 Table

3.S3, 3.S4). In general, variation was low within clusters (Fig. 3.3). In the Eastern

Cluster, the furthest separated sites had FST = 0.007 (sites 22 and 28; 41.2 km). In the

Central Cluster, the furthest separated sites had FST = 0.104 (sites 3 and 21; 109.8 km).

The two Western Cluster localities had FST = 0.28 (5.3 km), which is the highest intra- cluster comparison. Contrasts between clusters were high relative to intra-cluster comparisons (Fig. 3.3). For example, site 20 (Central Cluster) and site 22 (Eastern

Cluster), which are 7.1 km apart, had FST = 0.349. Similarly, site 2 (Western Cluster) and site 3 (Central Cluster), which are 17.2 km apart (including Sandusky Bay), had FST =

0.229. Site 1 in the Western Cluster and site 3 in the Central Cluster had FST =0.500 and are separated by 21.1km, though this includes Sandusky Bay and West Harbor.

Gene flow and genetic drift

Locus Pc16 was excluded in gene flow analyses because it led to greatly reduced

ESS values. In total, we recovered estimates with high consistency among three runs for

61 pairs of populations within the Central Cluster (Supplementary material Appendix 1

Table 3.S5). Gene flow rates are presented as the proportion of migrants (mh) between populations. Rates of gene flow were lowest between site 8 and 10+11 (m = 0.007), and 59 between site 3+5 and 4 (m=0.007). Gene flow was highest between site 17 + 18 and 19 +

20 (m = 0.115; Supplementary material Appendix 1 Fig. 3.S5, Table 3.S5).

Across all populations, RAFM revealed that low levels of coancestry between populations contributed to separate clusters, with increased genetic drift in lineages contributing to the Eastern and Western Clusters (Supplementary material Appendix 1

Fig. 3.S6). Within the Western Cluster, AFM found relatively strong genetic drift with little admixture between lineages (Supplementary material Appendix 1 Fig. 3.S7). Within the Central and Eastern Clusters, our sample sites exhibited relatively homogenous levels of coancestry; the degree of admixture varied among sites, with most sites having an admixture coefficient ≥0.50 for a particular lineage (Supplementary material Appendix 1

Fig. 3.S8, 3.S9). Overall, the AFM suggests that restriction of gene flow from the Central

Cluster and relatively strong genetic drift within the Eastern Cluster influenced differentiation between these groups.

Factors correlated with genetic variation

Across all sites, model selection using genetic distance as a response variable differed between RD and LCP. Under RD, ENM was the best-supported predictor variable, separated from other variables by ΔAICc = 34.8 (Table 3.1a). In the LCP analysis, the composite surface was best-supported and separated from the second ranked predictor (ENM) by ΔAICc = 66.7 (Table 3.2a). Within the Central Cluster, RD and LCP recovered DEM as the highest-ranking predictor, followed by canopy cover (ΔAICc =

14.6; RD; Table 3.1b) and ENM (ΔAICc = 31.7; LCP; Table 3.2b). Within the Eastern 60

Cluster, RD and LCP identified the composite surface as best-supported, however, ENM was also well supported (RD ΔAICc = 2.1, Table 3.1c; LCP ΔAICc = 0.2, Table 3.2c).

Factors correlated with morph frequency variation

Predictors of morph frequency variation were discordant between RD and LCP models. Under RD, waterways were the best-supported predictor variable across all populations and within the Central Cluster (Table 3.3a–b). Across all populations, DEM ranked second (ΔAICc = 12.98), and within the Central Cluster geographic distance ranked second (ΔAICc = 5.4; Table 3.3a–b). In the LCP analysis, across all populations and within the Central Cluster, geographic distance was top-ranked (Table 3c–d), followed by canopy cover across all populations (ΔAICc = 5.47). Within the Central

Cluster, waterways were the second most important predictor variable (ΔAICc = 0.16;

Table 3.3c–d).

Discussion

In this paper, we examined and compared factors that correlate with geographic variation in the striped/unstriped polymorphism in Plethodon cinereus. Across 28 sampling localities in northern Ohio, we found evidence for three genetic clusters

(Western, Central and Eastern; Fig. 3.1). The Western Cluster includes two sites (1–2), both of which have a high frequency of the unstriped morph (≥90%; Fig. 3.1a), a situation that is uncommon across the range of P. cinereus (Cosentino et al., 2017). The

Central Cluster includes localities 3–21 and exhibits a high variation in morph frequency among populations. The Eastern Cluster includes sites 22–28, all of which are fixed for the striped morph. 61

Across all populations, genetic distance was most correlated with our ENM using

RD. In contrast, LCP genetic distance was correlated with a combination of ecological and landscape features (our ‘composite model’). Other studies have also documented discrepancies between RD and LCP as measures of effective resistance (Avon & Bergès,

2016, McClure et al., 2016). In general, RD is more sensitive to raster surface aggregation, whereas LCP is more sensitive to the number of pixels representing a landscape and the Euclidean distance between populations (Marrotte & Bowman, 2017).

In our study, model discordance may be related to the computational complexity that results from combining genetic clusters. In the Central Cluster, RD and LCP were best fit by the DEM, whereas genetic variation within the Eastern Cluster was predicted by the composite model. Color morph frequency was not associated with genetic differentiation in P. cinereus. Using RD, waterways were the best predictor of morph frequency variation (Table 3.3a–b), while LCP found that geographic distance is the best predictor of morph frequency (Table 3.3c–d).

Correlates of genetic distance

In our study, elevation was the best predictor of genetic differentiation within the

Central Cluster. In northern Ohio, elevation and topographic diversity increase from west to east (Fig. 3.2, Table 3.1, 3.2). Elevation has been shown to influence gene flow in other amphibians as well (Funk et al., 2005, Lowe et al., 2006, Giordano et al., 2007). For example, Takahashi and Pauley (2010) found P. cinereus reach different adult body sizes and differentially allocate resources at low versus high elevation, suggesting elevation influences local adaptation, which may impact genetic divergence. We expected 62 environmental variables, rather than elevation, to correlate more strongly with genetic variation in P. cinereus, but our ENM may not have captured the fine-scale ecological features that affect niche use in plethodontid salamanders (Farallo & Miles, 2016).

Isolation by environment, where genetic distance is correlated with environmental variation, affects patterns of population structure in a wide variety of organisms (Wang &

Summers, 2010, Shafer & Wolf, 2013, Wang & Bradburd, 2014), and a recent review by

Sexton et al. (2014) found that in animals isolation by environment more commonly explains patterns of genetic variation than isolation by distance. Plethodontid salamanders lack lungs and depend on cutaneous respiration, and thus their dispersal behavior is heavily dependent upon cool, moist conditions (Spotila, 1972, Feder, 1983).

For example, Peterman et al. (2014) found inferred rates of water loss best described genetic differences between populations in the western slimy salamander, P. albagula.

While many factors influence population connectivity in P. cinereus, the salient result of our study is that all models found landscape resistance better represented patterns of genetic distance than geographic distance alone (Table 3.1, 3.2).

Correlates of morph frequency variation

Using RD, waterways had the greatest effect on morph frequency variation across all localities and within the Central Cluster (Fig. 3.1a, 3.2, Table 3.3a–b). Waterways have not been previously shown to affect variation in color morph frequency, but P. cinereus is a fully-terrestrial salamander that does not survive long when submerged in water, and thus waterways are likely important barriers. With LCP, we found that geographic distance was the most important factor influencing morph frequency 63 variation. Across the entire range of P. cinereus, as well as within Ohio, the distribution of the striped/unstriped polymorphism fits a mosaic pattern (Anthony & Pfingsten, 2013,

Moore & Ouellet, 2015), while in northern Ohio the striped/unstriped polymorphism appears to form a cline (Fig. 3.1a). Such a pattern is suggestive of secondary contact followed by lineage merger, or perhaps divergent selection between terminal portions of the range coupled with gene flow (McLean & Stuart-Fox, 2014). However, we instead found three discrete genetic clusters with limited or no merger where they meet: the

Eastern Cluster is fixed for the striped morph, while the Central Cluster is highly polymorphic; and the Western Cluster is nearly fixed for the unstriped morph. Thus, the broad clinal pattern in morph frequency across northern Ohio is not likely due to lineage merger or gene flow, but rather is the outcome of morph frequency variation in three genetic clusters.

Our discovery of three genetic clusters using microsatellites contrasts with a previous investigation of mtDNA variation by Radomski (2017). In that study, only two mtDNA clades were found in northern Ohio, and they form a secondary contact in the middle of our Central Cluster (Fig. 3.3). Discordance between mtDNA and nuclear markers is common and can have multiple causes (Toews & Brelsford, 2012). One possibility is that the two mtDNA clades correspond with our Eastern and Central genetic clusters, but mitochondrial introgression has shifted the mtDNA boundary westward (Fig.

3.3). Alternatively, male-biased dispersal could be the cause of discordance (Prugnolle & de Meeus, 2002), as males disperse over twice as far as females in P. cinereus (Liebgold et al., 2011). Under this scenario, the mtDNA contact zone could be located near the 64 original point of secondary contact, while the nuclear genome contact zone has shifted over time. More work is needed to distinguish between the competing explanations for cyto-nuclear discordance.

It is also possible that morph frequency variation is associated with environmental or genetic factors on a finer- scale than measured in this study. For instance, several studies have found the unstriped morph of P. cinereus is associated with warmer temperatures, while the striped morph is more associated with cooler temperatures

(reviewed by Cosentino et al., 2017). Similarly, a study by Fisher-Reid et al. (2013) on

Long Island, New York found that a parapatrically distributed population of striped and unstriped morphs was correlated with microclimatic preferences. Dispersal behavior may also impact morph frequencies. Grant and Liebgold (2017) documented morph-biased dispersal in P. cinereus and found that unstriped morphs dispersed farther than striped morphs. The authors postulated that unequal dispersal may lead to variation in morph frequency across the range of P. cinereus. Additional studies within a geographic context may help elucidate patterns of morph frequency variation in P. cinereus.

Maintenance of the color polymorphism

The correlation of color morph frequency with waterways and geographic distance, coupled with discordance between morph frequency and genetic distance, suggests that gene flow alone is not maintaining the polymorphism. If gene flow alone were maintaining the polymorphism, we would expect a correlation between morph frequency and genetic differentiation and higher rates of gene flow between populations with similar frequencies. Within the Central Cluster, low to moderate levels of gene flow 65 and low levels of genetic drift suggest that these evolutionary forces do not strongly influence morph frequency variation among sites. An exception to this generalization can be found in sites 3 and 4, which are relatively isolated sites within the Central Cluster; these sites have a relatively high frequency of the unstriped morph and were more influenced by genetic drift than other populations (Supplementary material Appendix 1

Fig. 3.S8). Among genetic clusters, genetic drift appears to have played a larger role in divergence, perhaps in part due to the Cuyahoga River acting as a barrier to gene flow between the Central and Eastern Clusters, and in part due to the isolation of the Western

Cluster sites, which historically were separated from Central Cluster sites by the Great

Black Swamp (Kaaz, 1955; Fig. 3.1, Supplementary material Appendix 1 Fig. 3.S6).

Our results suggest that the maintenance of the striped/ unstriped polymorphism is not heavily dependent on any one evolutionary process, but instead relies upon a balance between gene flow, genetic drift and selection. This mix of evolutionary processes has been shown to influence color morph variation in other systems as well (Gray &

McKinnon, 2007, Antoniazza et al., 2010, Muñoz et al., 2013). For instance, a balance between natural selection, gene flow, and genetic drift maintain the banded/unbanded color pattern polymorphism in the Lake Erie water snake Nerodia sipedon (King &

Lawson, 1995). Our understanding of how selection operates on the morphs of P. cinereus remains unclear; however, selection appears to be an important mechanism as geographic distance, gene flow, and genetic drift alone do not explain patterns of morph frequency variation across our study sites. Possible selective processes include variation in selection across space and time, negative frequency-dependent selection, and niche 66 partitioning within populations. Within the geographic range of the current study, Hantak and Kuchta (2018) found that the striped morph of P. cinereus was better camouflaged than the unstriped morph across multiple localities and seasons. If camouflage were the only trait under selection the polymorphism would not be maintained, suggesting other factors are at play. In another study, Fitzpatrick et al. (2009) used clay model replicas of the morphs to demonstrate that rare morphs have a survivorship advantage, suggesting negative frequency-dependent selection plays a role in the maintenance of the polymorphism. However, Kraemer et al. (2016) did not find evidence of negative frequency-dependent selection with mammalian predators. In addition, Grant et al. (2018) found no evidence of negative frequency-dependent selection in one population, although avian predation rates were higher and overall survivorship was lower in the striped morph. Finally, it is unclear how ecological niche divergence impacts color morph variation within and among populations.

In summary, using nuclear genetic markers we recovered three distinct genetic clusters in northern Ohio. Isolating barriers, such as inhospitable terrain and environmental conditions, appear to have promoted divergence, followed by the compounding consequence of increased genetic drift through a lack of gene flow. Past studies on the maintenance of the polymorphism in P. cinereus have produced a diversity of findings. Here, we found that waterway barriers and geographic distance, coupled with genetic drift, are correlated with divergence in morph frequency, while gene flow counteracts divergence. The mechanisms promoting variation in morph frequency among populations within the Central Cluster remain unclear, and it may be that there is 67 geographic variation in the evolutionary mechanisms maintaining the polymorphism.

Overall, our study suggests that a mix of gene flow, genetic drift, and selection interact to maintain the enigmatic striped/unstriped polymorphism.

Tables

Table 3.1. Resistance distance model selection from MLPE analysis with genetic distance as the dependent variable for (A) all sampling sites, (B) Central Cluster, (C) Eastern Cluster. ퟐ ퟐ Surface AICC AICC AICCWt 퐑퐦 퐑퐜 A: All sites ENM -497.45 0 1 0.51 0.71 Canopy cover -462.62 34.84 0 0.25 0.41 Waterways -460.29 37.17 0 0.25 0.39 Geographic distance -456.17 41.28 0 0.21 0.39 Composite -454.75 42.7 0 0.22 0.45 DEM -452.34 45.11 0 0.20 0.40 Morph frequency -384.75 112.7 0 0.06 0.28 B: Central Cluster DEM -807.44 0 1 0.53 0.78 Canopy cover -792.81 14.63 0 0.30 0.76 Waterways -789.39 18.05 0 0.03 0.69 ENM -789.26 18.18 0 0.03 0.69 Composite -788.37 19.07 0 0.03 0.69 Morph frequency -787.08 20.36 0 0.03 0.67 Geographic distance -785.56 21.88 0 0.02 0.68 C: Eastern Cluster Composite -116.39 0 0.74 0.68 0.68 ENM -114.25 2.14 0.26 0.65 0.65 Canopy cover -102.03 14.35 0 0.32 0.56 DEM -99.06 17.33 0 0.12 0.53 Waterways -98.29 18.09 0 0.07 0.54 Geographic distance -97.87 18.52 0 0.04 0.51 2 2 2 2 Rm = marginal R value; Rc = conditional R value.

Table 3.2. LCP model selection from MLPE analysis with genetic distance as the dependent variable for (A) all sampling sites, (B) Central Cluster, (C) Eastern Cluster. ퟐ ퟐ Surface AICC AICC AICCWt 퐑퐦 퐑퐜 A: All sites Composite -608.79 0 1 0.62 0.71 ENM -542.08 66.71 0 0.49 0.67 DEM -471.53 137.26 0 0.27 0.43 Canopy cover -471.49 137.3 0 0.26 0.40 Waterways -457.79 151 0 0.21 0.40 Geographic distance -456.17 152.62 0 0.21 0.39 Morph frequency -384.75 224.04 0 0.06 0.28 B: Central Cluster DEM -821.77 0 1 0.50 0.76 ENM -790.05 31.72 0 0.12 0.77 Canopy cover -789.3 32.48 0 0.24 0.73 Morph frequency -787.08 34.69 0 0.03 0.67 Composite -786.34 35.43 0 0.02 0.68 Waterways -785.66 36.11 0 0.02 0.68 Geographic distance -785.56 36.21 0 0.02 0.68 C: Eastern Cluster Composite -112.81 0 0.52 0.63 0.63 ENM -112.64 0.17 0.48 0.62 0.62 DEM -100.43 12.38 0 0.21 0.59 Canopy cover -99.54 13.27 0 0.19 0.51 Waterways -98.08 14.73 0 0.05 0.51 Geographic distance -97.87 14.94 0 0.04 0.51 2 2 2 2 Rm = marginal R value; Rc = conditional R value.

Table 3.3. Results of MLPE analysis with morph frequency as the dependent variable for resistance distance (RD) model selection across (A) all sampling sites, (B) Central Cluster; LCP model selection across (C) all sampling sites, (D) Central Cluster. ퟐ ퟐ Surface AICC AICC AICCWt 퐑퐦 퐑퐜 A: RD All sites Waterways 92.23 0 1 0.41 0.57 DEM 105.21 12.98 0 0.36 0.52 Geographic distance 109.88 17.65 0 0.33 0.55 Canopy cover 125.57 33.35 0 0.33 0.57 Composite 159.08 66.85 0 0.25 0.57 ENM 213.1 120.87 0 0.29 0.70 FST 260.28 168.05 0 0.05 0.33

B: RD Central Cluster Waterways 27.55 0 0.85 0.20 0.52 Geographic distance 32.89 5.35 0.06 0.16 0.51 ENM 32.92 5.38 0.06 0.19 0.49 DEM 34.01 6.47 0.03 0.50 0.69 Composite 38.32 10.77 0 0.16 0.48 FST 53.12 25.57 0 0.15 0.42 Canopy cover 55.92 28.38 0 0.25 0.63

C: LCP All sites Geographic distance 109.88 0 0.94 0.33 0.55 Canopy cover 115.34 5.47 0.06 0.34 0.55 DEM 124.14 14.26 0 0.32 0.58 Waterways 124.3 14.42 0 0.30 0.54 Composite 209.48 99.6 0 0.24 0.53 ENM 229.06 119.18 0 0.17 0.54 FST 260.28 150.4 0 0.05 0.33

D: LCP Central Custer Geographic distance 32.89 0 0.4 0.16 0.51 Waterways 33.05 0.16 0.37 0.16 0.51 DEM 35.39 2.5 0.12 0.39 0.58 Composite 35.47 2.58 0.11 0.14 0.51 FST 53.12 20.23 0 0.15 0.42 Canopy cover 54.32 21.43 0 0.32 0.72 ENM 63.52 30.63 0 0.01 0.45 2 2 2 2 Rm = marginal R value; Rc = conditional R value. 71

Figures

Figure 3.1. A) Sampling localities with morph frequencies (gray = unstriped, red = striped morph) throughout Ohio. Important waterways are denoted: Cuyahoga River (CR), Black River (BR), Huron River (HR), and Sandusky Bay (SB). The unstriped morph is depicted in the bottom left and the striped morph in the bottom right. B) STRUCTURE results for all populations, and clusters found by looking at substructure: C = ‘Central Cluster’, W = ‘Western Cluster’, and E = ‘Eastern Cluster’. Sampling localities match A.

Figure 3.2. Landscape surfaces with sampling localities (blue circles) used in ResistanceGA. A) Percent tree canopy, B) Ecological Niche Model (ENM), C) Digital Elevation Model (DEM), D) waterways. Green represents high, yellow intermediate, and white low values for each surface.

Figure 3.3. Map of northern Ohio sampling localities (black = unstriped, white = striped morph) with FST values between pairs of sites. Population numbers are in the center of each pie and correspond with Fig. 1. The dashed, black lines denote breaks between the three microsatellite clusters: W = ‘Western Cluster’, C = ‘Central Cluster’, and E = ‘Eastern Cluster’. The solid, black line separates mtDNA clades (Radomski 2017).

CHAPTER 4: A SPATIOTEMPORAL ASSESSMENT OF DIETARY PARTITIONING

BETWEEN COLOR MORPHS OF A TERRESTRIAL SALAMANDER

Introduction

Color polymorphisms, where multiple distinct color phenotypes coexist within a population, are common in nature (Ford, 1945; Gray & McKinnon, 2007; McLean &

Stuart-Fox, 2014). Within a population, color polymorphic species, which are composed of co-adapted sets of traits are known as 'morphs' (Sinervo & Svensson, 2002; McKinnon

& Pierotti, 2010). Quantifying disparities between morphs elucidates the degree of ecological separation within a population and provides information on how selection might operate on morphs. On the other hand, a lack of partitioning of critical resources such as food or territories may result in strong competition between morphs, which under some circumstances can lead to the fixation of one morph within a population (West-

Eberhard, 1986; Svensson et al., 2001; Comendant et al., 2003). Polymorphism is important to understand because it informs us about the origin of ecologically-relevant differences in form and behavior, including species formation (Sinervo & Svensson,

2002).

Polymorphic species commonly have widespread geographic distributions, however, little work has focused on the coevolutionary interactions between morphs within a geographic context (but see Thompson, 2005; Davis-Rabosky et al., 2016;

Holmes et al., 2017; Evans et al., 2018). Selection and evolutionary change vary through space and time as a consequence of divergent ecological conditions, and suites of co- adapted traits may vary across the range of a species (Brodie et al., 2002; Thompson, 75

2005; Corl et al., 2010). Accordingly, a mechanistic understanding of the ecological differences between morphs may benefit from spatiotemporal replication.

Here we report on a study of geographic variation in the Eastern Red-backed

Salamander, Plethodon cinereus. There are two common morphs in this species: a striped morph with a red stripe overlaid on a black dorsum and an unstriped morph that is completely black in dorsal coloration (Highton, 1959; Moore & Ouellet, 2014). The morphs, which have a genetic basis with no intermediate phenotypes, are broadly distributed geographically (Highton, 1959; 1975; Moore & Ouellet 2015). The ecological and evolutionary dynamics involved in maintaining the polymorphism are not well understood, yet the morphs differ along several biological axes, including physiology, mating interactions, territoriality, and dietary composition (Lotter & Scott 1977; Moreno

1989; Gibbs & Karraker 2006; Acord et al., 2013; Anthony & Pfingsten, 2013; Reiter et al., 2014; Jaworski et al., 2018). However, most studies of morph divergence have been conducted without replication over space and time.

Diet has been hypothesized to be an important axis of niche differentiation separating many morphs in nature (Price, 1987; Schluter & McPhail, 1992; Karpestam &

Forsman, 2011). Individuals of P. cinereus defend territories that contain valuable resources, including prey (Mathis, 1990; 1991). While P. cinereus is a generalist predator

(Burton, 1976), studies have found that individuals with established territories, which are aggressively defended, consume prey that maximize their caloric yield (Jaeger et al.,

1981; Jaeger & Barnard, 1981). In a polymorphic population, an experimental study found that striped morphs are dominant territory holders compared to unstriped morphs 76 and will expel intruders from cover objects (Reiter et al., 2014). In addition, Anthony et al. (2017) found that unstriped morphs occupy territories that have lower prey richness than striped morphs. As such, previous studies have found that the diet of the striped morph is of higher quality, providing increased access to more profitable prey (Anthony et al., 2008; Stuczka et al., 2016; but see Paluh et al., 2015).

In the current study, we investigated dietary prey composition of the striped and unstriped morphs of P. cinereus across two seasons. We included six populations that vary in morph frequency, from 100% striped to polymorphic to >99% unstriped (Fig.

4.1). We hypothesized that morphs partition dietary resources within polymorphic populations, facilitating the maintenance of the polymorphism, but that dietary breadth in monomorphic sites would be wider. Following from previous studies (Mathis, 1990;

Anthony et al., 2008; Paluh et al., 2014; Reiter et al., 2014; Stuczka et al., 2016; Anthony et al., 2017), we also predicted that striped morphs would have access to and consume a higher diversity of prey that are more energetically profitable compared to unstriped morphs. For monomorphic populations, we hypothesized that individuals of both morphs would experience ecological release, resulting in a more variable diet.

Methods

Dietary and habitat data were collected from six populations in northern Ohio that vary in morph frequency: Squire Valleevue Farm (SVF; 100% striped); Chapin Forest

Reservation (CF; 100% striped); Manatoc Scout Reservation, directly adjacent to

Cuyahoga Valley National Park (CV; 80% striped); Edison Woods Reservation (EW;

45% striped); the Heineman property on South Bass Island (SBI; < 1% striped; Hantak et 77 al., 2015); and East Harbor State Park (EH; 8% striped; Fig. 4.1). All sites are comprised of unmanaged, secondary forest, but soil varies across populations: SVF, CF, and CV are in soil series 6 (Mahoning-Canfield-Rittman-Chili); EW is in soil series 2 (Conotton-

Conneaut-Allis); and SBI and EH are in soil series 1 (Hoytville-Nappanee-Paulding-

Toledo; ODNR, 2019).

Individual salamanders were collected under rocks and logs at each location. At the point of capture we recorded morph, sex, and snout-vent length (SVL) of each individual (Table 4.S1). We collected dietary prey items from 15 individuals/morph from each population in both the spring and fall of 2014, totaling 240 samples (Table 4.S1).

Dietary data were only collected from unstriped morphs in SBI and EH, because striped morphs in these populations are rare. Gastric lavage was used to effectively and non- lethally collect dietary contents from each salamander (Legler & Sullivan, 1979; Bondi et al., 2015; Hantak et al., 2016). Subsequently, individuals were replaced to their original point of capture. Dietary samples from each individual were stored in 70% ethanol.

To determine if salamanders non-randomly selected certain prey types from their environment, and if morphs differed in their selection of prey from the available pool of potential prey items, leaf litter was collected at each site from both seasons. Collection took place at three random locations adjacent to the area of salamander capture.

Invertebrates were obtained by placing leaf litter in Berlese funnels and Winkler extraction bags (Ivanov & Keiper, 2009) for 72 hours. Collected leaf litter invertebrates were stored in 70% ethanol. Dietary and leaf litter samples were collected on a single day 78 from each population and season, totaling 12 sampling dates (Table 4.S1). This sampling scheme aided in avoiding resampling and minimizing the effects of climatic variation.

For each stomach and leaf litter sample, invertebrates were sorted into morphospecies, counted, and identified to the lowest taxonomic category (usually order or class) with a dissecting stereo-microscope. Using an ocular micrometer, the length and width of each dietary invertebrate was recorded to the nearest 0.01 mm. Total dietary volume in each salamander was estimated using the equation for a prolate spheroid

(Dunham, 1983):

4휋 length width ( ) ( )2 3 2 2

To determine if the number of salamanders provided an accurate estimate of diet from each morph and population, we constructed species accumulation curves. We ran a permutational multivariate analysis of variance (PERMANOVA) to test for group mean differences in dietary composition across season, site, morph, sex, and SVL. Significance was assessed using the pairwiseAdonis package in R (Martinez, 2019), and to illustrate dietary variation we employed non-metric multidimensional scaling (nMDS).

PERMANOVA and nMDS were based on Jaccard dissimilarity indices (Oksanen et al.,

2016) and were run on morphospecies data. Total number and volume of dietary prey were log-transformed, and zero values were removed to fulfill assumptions of normality.

We used full factorial ANOVAs to test for season, site, and morph differences in the total number and volume of prey consumed. A test of homogeneity of multivariate group dispersions (PERMDISP) was used to examine variance in invertebrate breadth between leaf litter and salamander diet for each season, site, and morph. PERMDISP is a 79 multivariate equivalent to the Levene’s test and uses distance matrices to estimate the mean distance to group centroids (Anderson, 2006). We used Tukey’s HSD post hoc tests to examine pairwise differences. Dietary niche breadth for each individual was calculated with the Shannon diversity index for each season. We used a full factorial ANOVA to test for season, site, and color morph differences in dietary niche breadth. Statistical analyses were conducted in vegan (Oksanen et al., 2016) using R statistical software (R

Core Team, 2018).

To test if sympatric morphs in CV and EW differ in preferred dietary items, we calculated importance values (Ix) for each taxonomic prey category for each salamander using:

Ix = [(nx/N) + (vx/V) + (fx/F)]/3 where nx, vx, and fx represent the number, volume, and frequency (the number of individuals that consumed that prey item) of each prey category and N, V, and F denote the sum of each value across all stomachs from a population. Importance values range from 0 to 1, with higher values equaling more important prey (Powell et al., 1990;

Anderson & Mathis, 1999). Numbers of the six most important prey items were compared between morphs using chi-square tests within each season.

Results

We identified 2314 dietary prey items from 240 individuals of P. cinereus across spring and fall of 2014. Spring salamanders (n=120) consumed 1441 prey, whereas fall individuals (n=120) consumed 873 prey. Invertebrates were found from 21 taxonomic categories (class, order, or family) and totaled 51 morphospecies. Spring individuals 80 consumed a mean number of 11.91  14.67 (SD) prey, and fall salamanders consumed a mean of 7.16  5.39 prey. Species accumulation curves demonstrate that sample sizes from each morph and population provided an accurate assessment of diet (Fig. 4.S1). In the spring, the diet of P. cinereus largely consisted of Acari (mites; 29%), Collembola

(springtails; 23%), and Formicidae (ants; 15%). Fall salamander diet primarily consisted of Formicidae (31%), Acari (17%), and Collembola (12%). The mean number and volume of consumed prey from each site and season are listed in Table 4.1.

We found no effect of morph, sex, or SVL on dietary composition when accounting for season and site (Table 4.2). However, the interaction between season and site was significant (Table 4.2); therefore, we looked at population differences within each season and found significant variation within both the spring and fall (Table 4.2).

Dietary differences within each season are illustrated using nMDS with 95% confidence ellipses for population centroids (Fig. 4.2). Pairwise comparisons of p-values, which were adjusted for multiple comparisons, in the spring reveal that the western sites (SBI, EH, and EW) did not differ from each other, but each of these sites differed from the three eastern sites (CF, CV, SVF). In addition, SVF did not differ from CF or CV, and CV and

EW overlapped, but all other comparisons differed (Fig. 4.2a, Table 4.S2). In the fall,

SVF did not differ from EH, EW, CV, or CF, and EH did not differ from SBI, EW, or

CV, but all other comparisons significantly differed from each other (Fig. 4.2b, Table

4.S2). Across seasons and sites, morphs did not differ in the total number of prey consumed (Table 4.2). The total volume of consumed prey differed by season, site, and 81 morph (Table 4.2). However, within seasons, there was no interaction between site and morph in the spring or fall (Table 4.2).

In the spring we identified a total of 4464 leaf litter invertebrates, whereas in the fall we identified 3959 leaf litter invertebrates. Dietary and leaf litter invertebrates largely overlapped, and most comparisons did not differ (Table 4.S3). Salamander prey differed from leaf litter invertebrates at EW in the spring (F2,30 = 5.57, P = 0.009). Post hoc comparisons demonstrate the variance of invertebrates in the leaf litter is higher than in striped (P = 0.007) and unstriped (P = 0.014) stomachs, but diet did not differ between the morphs (P =0.861; Fig. 4.3a). In the spring, salamander diet and leaf litter invertebrates also differed at SBI (F1,15 = 5.00, P = 0.040), with higher invertebrate variance in the leaf litter compared to salamander diet (Fig. 4.3b, Table 4.S3). Estimates of niche breadth did not differ by site and morph in the spring or fall, but differed by site in the spring with EW having a lower niche breadth than CV (Fig. 4.S2, Table 4.2). The six most important prey for P. cinereus at CV and EW included Acari, Coleoptera,

Collembola, Hymenoptera, Lepidoptera larvae, and Oligochaeta (Table 4.3). Important prey did not differ between striped and unstriped morphs from CV in the spring (df = 5,

X2 = 1.0, P = 0.961), or the fall (df = 5, X2 = 10.7, P = 0.058). Striped and unstriped morphs from EW did not differ in important prey in the spring (df = 5, X2 = 2.6, P =

0.737), or fall (df = 5, X2 = 7.3, P = 0.199; Table 4.3).

Discussion

Discrete morphs are comprised of co-adapted sets of traits and differ in elements of their biology in addition to color, including physiology, behavior, and other 82 morphological characters (Sinervo & Svensson, 2002; McKinnon & Pierotti, 2010).

Morph traits at a single site are often thought to parallel other populations across the range of a species, however, it is well known that local species interactions and environmental conditions can result in geographically variable patterns of evolution

(Thompson, 2005; McLean & Stuart-Fox, 2014). In this paper, we examined the dietary composition of the striped and unstriped color morphs of P. cinereus across two seasons and six populations that varied in morph frequency (Fig. 4.1). Our data do not support the hypothesis that dietary partitioning broadly characterizes the two color morphs, as the diet of the morphs overlapped in prey type, number, and volume. Important dietary prey did not differ between morphs in the polymorphic EW site, or at CV in the spring; however, in the fall there was a trend for morphs at CV to differ in important prey.

Overlap in dietary and leaf litter invertebrates suggests individuals of P. cinereus are generalist predators; however, individuals at SBI and EW appear to select a subset of the prey available in the leaf litter. No change in dietary niche breadth between monomorphic and polymorphic populations suggests a lack of ecological release when a morph is lost.

Dietary composition varied across seasons and sites, demonstrating the importance of studying ecological attributes over spatial and temporal scales.

In the current study, while taking into consideration spatial and temporal effects, we predicted morphs of P. cinereus would differ in diet, and striped morphs would consume more energetically profitable prey compared to unstriped morphs. Instead, our results demonstrate that the diet of the two morphs overlap. These results were unexpected, as prior studies found the morphs differed in dietary composition (Anthony 83 et al., 2008; Paluh et al., 2015; Stuczka et al., 2016). However, findings across these earlier studies varied. For instance, Anthony et al. (2008) and Stuczka et al. (2016) found that striped morph diet was composed of higher quality prey, whereas Paluh et al. (2015) found that striped morphs consumed more ants than unstriped morphs, which are heavily- armored and less energetically profitable than many other insects (Jaeger, 1990); however, this study only examined variation in ant prey species, and it may be possible that striped morphs consumed other, higher quality prey. Discordant results were also found concerning which morph consumed a higher diversity of dietary prey, with

Anthony et al. (2008) finding that striped morph diet was more diverse, while Stuczka et al. (2016) found that the unstriped diet was comprised of a higher diversity of prey. Each of these studies took place in the same polymorphic population in northeast Ohio (our CV site). Differences between studies can be explained by discordant time scales, sampling techniques, and sample sizes. For instance, in our study and Anthony et al. (2008) all dietary samples were collected on a single day per site, whereas Stuczka et al. (2016) collected samples over the course of a year. Variation in dietary composition in the CV site, thus, likely reflects temporal changes in prey availability, which may alter the dynamics of the local social environment.

Our finding of dietary overlap between the morphs was unexpected, as they differ in many other elements of their biology. For instance, a number of studies have suggested that the striped morph is more common in cooler and wetter habitats than the unstriped morph (Lotter & Scott, 1977; Gibbs & Karraker, 2006, Cosentino et al., 2017; but see

Moore & Quellet, 2015), and that the unstriped morph withdraws into underground 84 retreats sooner with the onset of cold weather (Lotter & Scott, 1977; Moreno, 1989;

Anthony et al., 2008). Studies have also shown that the morphs differ in response to predators, including differences in tail breakage frequencies (Moreno, 1989; Venesky &

Anthony, 2007; Grant et al., 2018) and behavioral responses to predators (Venesky &

Anthony, 2007; Otaibi et al., 2017). One study found that the striped morph exhibited higher levels of territorial behavior and aggression compared to the unstriped morph

(Reiter et al., 2014). The unstriped morph has a lower metabolic rate, higher stress levels, and increased incidence of disease prevalence (Moreno, 1989; Petruzzi et al., 2006;

Fisher-Reid et al., 2013; Venesky et al., 2015). In addition, in one population in Ohio

(our CV site), males and females were found to pair assortatively by color (Anthony et al., 2008; Acord et al., 2013). These previous studies provide evidence that the morphs are comprised of correlated trait complexes; however, as with our results on dietary composition, re-analyzation of these findings within a geographic context may help elucidate whether morph trends are consistent over spatiotemporal scales.

A study by Hantak and Kuchta (2018) found that the striped morph was better camouflaged than the unstriped morph against multiple backgrounds across three populations and two seasons. It is possible that decreased detection of striped morphs by visual predators affords greater foraging opportunities. However, we found no evidence that striped morph diet was of higher quality than the unstriped morph. A similar prediction was made by Hantak et al. (2016), where the authors postulated that a rare erythristic (orange-red) morph of P. cinereus, which may be a mimic of the toxic Eastern

Newt (Notophthalmus viridescens; Kraemer & Adams, 2013), would have greater 85 foraging opportunities than sympatric striped morphs and, therefore, consume more profitable prey. However, Hantak et al. (2016) found no difference in dietary composition between sympatric erythristic and striped morphs. Plethodon cinereus is known to be a generalist predator (Burton, 1976; Maerz et al., 2005; Bondi et al., 2019), and therefore, if dietary resources are plentiful, competition may not occur. However, during certain times of the year, such as during periods of unfavorable weather conditions (e.g., extreme dry periods) or periods of high activity, competition may be more intense (Pianka, 1974).

It may be that our sampling took place during times of the year when competition for resources was relatively low. For example, our study was conducted in late spring and early fall, and thus did not coincide with peak mating season (October to April; Table

4.S1) (Anthony & Pfingsten, 2013), whereas previous studies, which found dietary divergence between the morphs, corresponded with the mating season of P. cinereus

(Anthony et al., 2008; Paluh et al., 2015; Stuczka et al., 2016). It is also possible that our sampling was biased towards more territorial individuals. Each salamander in this study was found under a cover object, which is presumably surrounded by high quality territory

(Jaeger, 1980; Gabor, 1995). Reiter et al. (2014) found that striped morphs were more territorial, but our sampling may have captured the few territorial unstriped individuals.

Future studies should investigate whether there is a territory-mediated difference in diet.

Important prey (Ix) also did not differ between morphs in our sampled polymorphic populations (CV and EW). However, there was a trend for important prey to differ between morphs in CV during the fall season (P = 0.058). Higher levels of sampling at CV in previous studies found that the diet of morphs differs seasonally 86

(Stuczka et al., 2016), in the number of consumed prey (Anthony et al., 2008), and in the available prey in territories (Paluh et al., 2015; Anthony et al., 2017). Thus, the current results corroborate previous findings at CV, although more subtle differences are not apparent in this study due our smaller sample sizes within populations. In our case, the logistical constraints of including multiple populations constrained our within-population sampling, meaning that only large effects had a high probability of detection. Such discrepancies between studies highlights the consideration of sample size and effect size when looking for differences between groups.

Across sites, non-random selection of leaf litter invertebrates by salamanders occurred at SBI and EW in the spring. At the polymorphic EW site, both morphs selected a subset of available prey, which could result in competition in this population. However, we found no morph differences in dietary niche breadth, including no change in dietary breadth between polymorphic and monomorphic populations (Fig. 4.S2). Thus, in contrast with our a priori predictions, there is no evidence of ecological release in diet in monomorphic populations. Ecological release arises when a species (or morph) gains access to resources that were previously unattainable or restricted (Dayan & Simberloff,

1994; Losos & de Queiroz, 1997; Bolnick et al., 2010). In polymorphic populations, some prey may be inaccessible to a subordinate morph when there is high competition for food or territories. This competition can result in ecological character displacement, which may impact divergence between polymorphic and monomorphic populations. For example, Adams (2000) demonstrated that individuals of Plethodon hoffmani and P. cinereus consume similar diets in allopatry, while in sympatry they deviate in diet. This 87 sympatric divergence was associated with a corresponding shift in body size and head shape. Morphological character displacement has been found between other salamander species (Adams et al., 2007; Adams, 2010; Deitloff et al., 2013), as well as in Anolis lizards (Schoener, 1970; Losos, 1990), stickleback fish (Schluter & McPhail, 1992), and

Darwin’s finches (Grant & Grant, 2006).

Seasonal and geographic variation in dietary composition was pervasive throughout our study. In the spring, we found sites from the western portion of our sampling (SBI, EH, EW) did not differ in dietary prey composition, whereas sites in the eastern portion of our sampling range (CF, SVF, CV) differed from most other sites (Fig.

4.2a). In the fall, we found that EH overlapped in dietary prey composition with SBI,

EW, CV, and SVF, while SVF overlapped with CF, CV, and EW (Fig. 4.2b). This lack of a clear pattern across seasons suggests that the temporal availability of leaf litter invertebrate communities varies substantially across our study sites (Scott & Epstein,

1987). Relatively few studies have examined dietary variation in salamanders over temporal scales (but see Denoël et al., 2006; Wheeler et al., 2007; Sebastiano et al., 2012;

Costa et al., 2015; Stuczka et al., 2016). It is possible that seasonal fluctuations impact the competitive environment. For instance, individuals in SBI and EW in the spring selected a subset of available leaf litter invertebrates, whereas in the fall we found no evidence of prey selection (Fig. 4.3). Our study sites were selected based on patterns of morph frequency variation and were distributed over a 120 km area that contains many other populations that vary in morph ratios (Fig. 4.1; Hantak et al., 2019). Thus, it is possible for site specific patterns to differ; however, our principal finding is that diet is 88 variable across seasons and populations, and thus we recommend caution when treating dietary differences and feeding ecology as species-level traits.

Tables

Table 4.1. Mean number (N) and volume (V) each with standard deviation (SD) of prey consumed from each season, site, and morph. Season Site Morph Mean N SD N Mean V SD V Spring SVF Striped 23.4 32.69 4.17 2.92 Spring CF Striped 17.27 10.66 14.63 14.71 Spring CV Striped 14.13 8.55 13.78 10.96 Spring CV Unstriped 8.93 6.97 11.5 9.24 Spring EW Striped 8.27 11.93 9.99 8.46 Spring EW Unstriped 9.07 8.86 16.24 11.48 Spring EH Unstriped 8.53 5.57 9.37 6.23 Spring SBI Unstriped 6.47 4.55 7.3 6.95 Fall SVF Striped 6.27 5.95 9.44 18.02 Fall CF Striped 11.4 9.69 22.24 22.93 Fall CV Striped 5.94 3.62 4.6 4.94 Fall CV Unstriped 7.4 3.83 5.99 5.29 Fall EW Striped 5.47 3.58 12.27 12.25 Fall EW Unstriped 6 4.16 8.36 9.48 Fall EH Unstriped 9.2 4.54 17.56 9.19 Fall SBI Unstriped 5.75 2.91 17.27 18.79

Table 4.2. Results of (A) PERMANOVA for the full model, as well as seasonal and site effects. Full factorial ANOVA results for (B) the total number of invertebrates consumed, (C) the total volume of consumed prey, and (C) dietary niche breadth for model variations of season, site, and morph effects. Significant results in bold. Model df F P A: PERMANOVA Season x Site x Morph x Sex x SVL 1, 242 1.25 0.228 Season x Site 5, 241 2.39 <0.001 Spring: Site 5, 119 3.64 <0.001 Fall: Site 5, 119 2.84 <0.001

B: Total Number (ANOVA) Season x Site x Morph 1, 217 2.03 0.156

C: Total Volume (ANOVA) Season x Site x Morph 1, 217 4.67 0.032 Spring: Site x Morph 1, 110 2.52 0.115 Fall: Site x Morph 1, 107 2.17 0.144

D: Niche Breadth (ANOVA) Spring: Site x Morph 1, 112 0.59 0.446 Spring: Site 5, 112 3.60 0.005 Fall: Site x Morph 1, 112 0.23 0.634 Fall: Site 5, 112 0.43 0.119

Table 4.3. (A) Spring and (B) fall importance values (Ix) for the six dominant prey categories in striped and unstriped morph stomachs within the CV and EW polymorphic sites. Importance values were calculated from the total number (nx), volume (vx), and frequency (fx), shown in parentheses for each prey type. A: Spring Prey taxon CV (striped) CV (unstriped) EW (striped) EW (unstriped) Acari 0.384 (58, 2.7, 13) 0.411 (39, 1.4, 14) 0.205 (42, 1.5, 4) 0.268 (54, 1.9, 6) Coleoptera 0.249 (17, 41.1, 7) 0.269 (10, 45.9, 7) 0.423 (25, 39.9, 12) 0.417 (31, 70.8, 11) Collembola 0.231 (19, 0.5, 9) 0.198 (17, 0.3, 7) 0.080 (5, 0.1, 3) 0.104 (6, 0.1, 4) Hymenoptera 0.384 (43, 30.7, 12) 0.332 (33, 25.6, 9) 0.250 (22, 16.0, 7) 0.270 (24, 24.6, 8) Lepidoptera 0.290 (8, 75.9, 7) 0.213 (5, 57.5, 4) 0.251 (5, 67.0, 4) 0.183 (5, 59.8, 4) Oligochaeta 0.032 (2, 4.1, 1) 0.059 (1, 17.6, 1) 0.000 (0, 0.0, 0) 0.141 (3, 49.2, 3) B: Fall Prey taxon CV (striped) CV (unstriped) EW (striped) EW (unstriped) Acari 0.217 (13, 1.0, 8) 0.250 (23, 0.8, 8) 0.179 (11, 0.3, 6) 0.310 (29, 0.8, 9) Coleoptera 0.148 (5, 5.8, 5) 0.252 (11, 11.7, 8) 0.340 (10, 79.6, 7) 0.317 (13, 42.7, 7) Collembola 0.274 (24, 0.6, 9) 0.239 (19, 1.2, 8) 0.106 (4, 24.7, 10) 0.078 (3, 0.1, 3) Hymenoptera 0.392 (27, 19.7, 10) 0.504 (37, 36.0, 12) 0.381 (28, 24.7, 10) 0.459 (27, 34.9, 12) Lepidoptera 0.000 (0, 0.0, 0) 0.131 (3, 15.8, 3) 0.036 (2, 3.1, 1) 0.130 (2, 29.3, 2) Oligochaeta 0.131 (3, 12.9, 3) 0.000 (0, 0.0, 0) 0.046 (1, 10.9, 1) 0.000 (0, 0.0, 0)

Figures

Figure 4.1. Map of study sites on a percent canopy cover landscape surface in northern Ohio. Color morph frequencies (black = unstriped, red = striped). From west to east: South Bass Island (SBI), East Harbor State Park (EH), Edison Woods (EW), Manatoc Scout Reservation near Cuyahoga Valley (CV), Squire Valleevue Farm (SVF), and Chapin Forest (CF).

Figure 4.2. Nonmetric multidimensional scaling (nMDS) plots with 95% confidence ellipses of diet variance based on morphospecies for each population: Chapin Forest (CF), Squire Valleevue Farm (SVF), Manatoc Scout Reservation near Cuyahoga Valley (CV), Edison Woods (EW), East Harbor State Park (EH), and South Bass Island (SBI) in the (A) spring and (B) fall. More overlap indicates greater dietary similarity.

Figure 4.3. (A) Edison Woods (EW) salamander color morph and leaf litter invertebrate breath in the spring season. (B) South Bass Island (SBI) salamander and leaf litter invertebrate breath in the spring season. 95

CHAPTER 5: DOES RANDOM MATING PROMOTE THE MAINTENANCE OF A

COMMON SALAMANDER COLOR PATTERN POLYMORPHISM?

Introduction

Evolutionary biologists are intrigued by phenotypic diversity in nature and seek to understand the mechanisms that contribute to the maintenance of such variation. Sexual selection driven by mate preference for certain traits can have a profound influence on intraspecific diversity (Andersson, 1994; Debelle et al., 2014). Assortative mating—the non-random mating of phenotypically similar individuals—can lead to the evolution of reproductive isolation between distinct forms (Dieckmann & Doebeli, 1999; Servedio &

Boughman, 2017), and within populations assortative mating may facilitate the evolution of sympatric divergence (Foote & Larkin, 1988; Elmer el al., 2009). Across phenotypically distinct populations, reproduction by immigrant individuals may be inhibited by selection against the immigrants’ divergent phenotype (Nosil et al., 2005;

Servedio 2016). Understanding mating interactions between phenotypically distinct groups provides information on the maintenance of phenotypic and genetic variation, and in some cases the origin of species.

Color polymorphic species, where two or more distinct, genetically determined color morphs exist within a single interbreeding population, provide a model to study the processes contributing to reproductive isolation (Ford, 1945; Gray & McKinnon, 2007).

Distinct morphs are comprised of alternative sets of co-adapted trait complexes (Sinervo

& Svensson, 2002); for example, morphs may differ in ecology, behavior, and morphology, as well as possessing discrete coloration (McKinnon & Pierotti, 2010). How 96 and why color morphs are maintained is a dynamic question in evolutionary biology and understanding the role of mate preference in polymorphic systems can be a useful step in identifying how this phenotypic variation is maintained (McLean & Stuart-Fox, 2014).

While random mating may contribute to color morph maintenance, assortative mating can result in reproductive isolation between morphs, which may provide a mechanism by which morphs evolve into separate species (Kirkpatrick, 2000; Jiggins et al., 2001).

Further, geographic trends in assortative mating remain relatively unexplored, even though population variation in mating preferences may promote reproductive isolation over spatial scales (Huyghe et al., 2010). For instance, if morphs assortatively mate by color in polymorphic populations, then immigration to a monomorphic site lacking the migrating color morph may result in a lack of genetic exchange. Size-assortative mating may also lead to divergence between color morphs. With positive assortment, the production of intermediate sized offspring is reduced, which may result in disruptive selection (Jiang et al., 2013). Morphic variation in assortative mating by size can thus lead to sympatric divergence, and can also be a premating isolating barrier across sites with disparate phenotypes.

The Eastern Red-backed Salamander, Plethodon cinereus, is widespread throughout northeastern North America, and displays a common striped/unstriped color polymorphism (Highton, 2004). The ‘striped’ color morph exhibits a red stripe overlaid on a black dorsum and the ‘unstriped’ morph is completely black in dorsal coloration

(Fig. 5.1). How the morphs are maintained over spatiotemporal scales is not well understood, however, previous studies have demonstrated that the two morphs are 97 differentiated along multiple axes of niche variation (Anthony & Pfingsten, 2013). For instance, the striped morph is more common in cooler and wetter habitats than the unstriped morph (Lotter & Scott, 1977; Gibbs & Karraker, 2006; but see Moore &

Ouellet, 2015), and the unstriped morph withdraws to underground retreats sooner with the onset of cold weather (Anthony et al., 2008). The unstriped morph also has a lower metabolic rate, which may permit it to be more active under warmer, drier conditions

(Moreno, 1989; Petruzzi et al., 2006; Fisher-Reid et al., 2013). The morphs appear to differ in response to predators, including differences in tail breakage frequencies

(Venesky & Anthony, 2007; Otaibi et al., 2017; Grant et al., 2018). The diet of the striped morph is more diverse and energetically profitable than the diet of the unstriped morph

(Anthony et al., 2008; Stuczka et al., 2016; but see Paluh et al., 2015), and the striped morph exhibits higher levels of territorial behavior (Reiter et al., 2014). Prior research has also found that the morphs mate assortatively by color (Anthony et al., 2008; Acord et al.,

2013). This is interesting because it may be a mechanism by which morphs evolve into separate species (Elmer et al., 2009).

Assortative mating between morphs of P. cinereus has been studied by Anthony et al. (2008) and Acord et al. (2013) using both field surveys and laboratory preference tests. Individuals of P. cinereus do not usually mate under laboratory conditions

(Highton, 1959; Acord et al., 2013), and thus laboratory breeding experiments are ineffective. In both studies, morphs in the field were found to assortatively associate (and likely mate; see below) by color in one population in Ohio. These studies also found larger females paired more often with striped males. The authors postulated that this 98 finding may be due to striped males holding higher-quality territories and consuming more profitable prey (Anthony et al., 2008; Reiter et al., 2014; Stuczka et al., 2016). In this study we re-examine these hypotheses in a geographic framework, across six populations that vary in color morph frequency from monomorphic striped to polymorphic to nearly monomorphic unstriped. In polymorphic populations, we predicted that morphs would assortatively mate by color and that larger females would be paired more often with striped males. Assortative mating by body size may also be important in this polymorphic system. In a monomorphic striped population of P. cinereus, Mathis

(1991a) found no evidence of size-assortative mating, although males found in mating pairs were larger than males found alone (1991a). However, larger body size is an indication of higher quality (i.e., larger territory and/or higher caloric diet) in salamanders

(Verrell, 1995; Eddy et al., 2016), and thus, we predict that larger males, who presumably hold better territories, would mate more often with larger females.

Methods

In spring of 2014, we placed 100 porcelain tiles (1 ft2), which serve as artificial cover objects (ACOs) for salamanders, in a grid at six localities in northern Ohio. ACOs provide repeatable, standardized, high quality territories for salamanders (Mathis, 1990;

Monti et al., 2000). Northern Ohio is ideal for studying this color polymorphism due to the discovery of many populations that vary in morph frequency, including populations that have a high frequency of the unstriped morph, which are extremely rare (Pfingsten &

Walker, 1978; Moore & Ouellet, 2015; Hantak et al., 2015). We selected sites based on variation in color morph frequency: Squire Valleevue Farm (SVF; 100% striped), Chapin 99

Forest Reservation (CF; 100% striped), Cuyahoga Valley National Park (CVNP; 80% striped), Edison Woods Reservation (EW; 45% striped), East Harbor State Park (EH; 8% striped), and the Heineman property on South Bass Island (SBI; < 1% striped; Fig. 5.1;

Hantak et al., 2015).

We surveyed each site once a week for mating pairs in spring and fall of 2014-

2016. Adult P. cinereus are territorial and will expel intruders from their cover objects, including unwanted individuals of the opposite sex (Mathis, 1990, 1991b). Male-female pairs within ~30 cm under the same cover object are usually mating pairs (Gillette et al.,

2000; Jaeger et al., 2002). As our ACOs were 30.48 cm2, we assumed male-female pairs found under an ACO together were a mating pair (Anthony et al., 2008). Sex was determined by the shape and size of the snout of male salamanders; adult male P. cinereus in reproductive condition have an enlarged, broad snout, whereas in the remaining portion of the year they possess a more rounded snout (Anthony & Pfingsten,

2013). In females, the presence of eggs visible through the side of the body is a clear indication that she is in reproductive condition; however, in some cases, eggs are not visible in reproductively viable individuals, thus the mating status of every female in this study was not perfectly known. We assumed every female with a reproductive male constituted a mating pair. When a mating pair was found, we recorded sex, color morph

(presence/absence of a dorsal stripe), snout-vent length (SVL), mass (g), and tail condition (partial presence/absence of tail). The latter trait was recorded to account for salamander condition (Wake & Dresner, 1967), as the tail contains fat reserves that are used in growth and reproduction (Maiorana, 1977; Jamison & Harris, 1992). Tail 100 autotomy may therefore impact the probability of obtaining a mate (Martín & Salvador,

1993). Lastly, a photograph was taken of the ventral side of each individual salamander, to serve as a mark and account for recaptures.

Each site was kept separate for statistical analyses as each population is geographically separated (12–122 km), variable in color morph frequency, and differs in levels of genetic variation (Hantak et al., 2019; Waldron et al., 2019). Due to a high degree of collinearity between SVL and mass at each population, we calculated body condition index (hereafter, body size) of each individual by performing an ordinary least squares (OLS) regression of SVL against mass and saving the residuals (Schulte-

Hostedde et al., 2005). Mating data collected from each site was analyzed using a linear model framework. In the initial set of models, we included year as a factor and Julian day as a covariate to test for any temporal variation. These variables were not significant in any analysis and were subsequently dropped from each model.

To examine whether mating pair ratios differed across polymorphic populations, we used a G-test of independence with the R package RVAideMemoire (Hervé, 2019). To test for patterns of color assortative mating in each of the polymorphic populations

(CVNP, EW, and EH), we used multiple logistic regression with a binomial error structure. In our first set of models, female color morph (striped or unstriped) was used as the response variable, and the predictor variables were male color morph, body size, male tail absence/presence, and the interaction between male color morph and body size. In the second set of models, male color morph was the response variable, and the predictor variables were female color morph, body size, female tail absence/presence, and the 101 interaction between male color morph and body size. Models were run with female and male predictors because both sexes invest energy into reproduction and are selective in regards to mating (Anthony & Pfingsten, 2013; Jaworski et al., 2018). We ran models with recaptures removed from the dataset and with male and female ID as random effects

(with recaptures); these results were not qualitatively different, and here we only present results with recaptures removed. We also analyzed patterns of assortative mating by color using Fisher’s exact tests. The importance of size-assortative mating in both monomorphic and polymorphic populations was tested by performing simple linear regressions on male and female body size at each site.

All analyses were performed using R software V3.5.1 (R Core Team 2018). We used the glm function in the stats package (R Core Team, 2018) to fit the generalized linear models (GLMs) and the glmer function in the package lme4 (Bates et al., 2015) to fit the generalized linear mixed-effects models (GLMMs; recaptures included). We assessed the significance of all combinations of factors, covariates, and interaction terms for the binomial GLMs by examining best-fit models and ranking them using AICc with the R package MuMIn (Barton, 2019). When models had similar support (ΔAIC < 2), we tested the significance of the model with the fewest number of parameters using a Wald’s test (Bolker et al., 2009).

Results

Across six populations and six field seasons, we observed 188 mating pairs. After removing recaptures our dataset included 159 mating pairs across the six study sites: 20

SVF; 15 CF; 50 CVNP; 32 EW; 25 EH; and 17 SBI. In polymorphic sites (CVNP, EW, 102 and EH), the distribution of color morph pairings differed among sites (G = 57.25; P <

0.001; Fig. 5.2), and a post hoc tests revealed that each site differed from the other (Table

5.S1).

We tested a total of seven models for female color morph mate preference and for male color morph mate preference from each of the three polymorphic populations (Table

5.S2). At CVNP, the interaction between color morph and body size was the best-fit model for both female and male preference (Table 5.1, 5.S1). The best-supported model for female color morph preference at EW was male color, while female body size was the highest-ranking model when male color morph was the predictor (Table 5.1, 5.S1).

Within EH, with female color morph as the predictor, male body size was the best-fit model, and with male color morph as the predictor, female color morph was the best supported model (Table 5.1, 5.S1). However, none of these best-fit models were significant (all P ≥ 0.247; Table 5.1). A Fisher’s exact test, which we ran with and without recaptures, gave qualitatively similar results (not shown).

Evidence of positive size-assortative mating was found at EH (R2 = 0.23, t = 2.61,

P = 0.016) and CF (R2 = 0.29, t = 2.29, P = 0.039; Fig. 5.3). Assortative mating by size was not found at CVNP (R2 = 0.03, t = 1.29, P = 0.20), EW (R2 = 0.02, t = 0.83, P =

0.415), SVF (R2 = 0.05, t = -0.97, P = 0.347), or SBI (R2 = 0.00, t = -0.03, P = 0.979; Fig.

5.3).

Discussion

Assortative mating may be a mechanism by which distinct forms diverge into separate species (Kirkpatrick & Ravigne, 2002; Bolnick & Kirkpatrick, 2012). On the 103 other hand, random mating can facilitate the maintenance of distinct forms within and among populations (Erlandsson & Rolán-Alvarez, 1998; Rolán-Alvarez et al., 2012). In this paper we examined six populations that varied in color morph frequency to determine whether P. cinereus assortatively mates by color or size, and whether these trends are consistent among populations. We found that morph ratios of mating pairs varied across polymorphic sites, but this variation was due to the frequency of each morph per population (Fig. 5.1, 5.2). There was no evidence of positive assortative mating by color in any of the three polymorphic populations. Females and males of both color morphs did not display an association with any particular color morph, body size, or tail presence/absence. Assortative mating by body size, with larger males found with larger females, was variable. We found positive assortative mating by size at EH and CF, but we did not see this trend in any of the other four populations.

Our findings are in contrast to previous studies on color assortative mating in P. cinereus, as Anthony et al. (2008) and Acord et al. (2013) found that morphs assortatively mate by color in one population in Ohio. In the current study, we re- examined this population, CVNP; however, we found no evidence of color assortative mating at this site or the other polymorphic sites (EW and EH). Discrepancies between our study and previous investigations may be due to differences in the number of discovered mating pairs. At CVNP, Anthony et al. (2008) uncovered 94 mating pairs and

Acord et al. (2013) found 112 pairs. Over the course of our study, we discovered 50 mating pairs at CVNP after accounting for recaptures. We tested the statistical power to detect mating patterns in our study using the R package pwr (Champely et al., 2018). In 104 our study, we had relatively low statistical power to detect trends ( = 0.05, ES = 0.003, power = 0.050), while Anthony et al. (2008) and Acord et al. (2013) had higher power (

= 0.05, ES = 0.185, power = 0.434;  = 0.05, ES = 0.184, power = 0.495, respectively).

Thus, much of the discrepancy between studies may be due to sample size alone. Indeed, when we combine results from our study and data from the two previous studies, we find that we have much stronger power to suggest that color assortative mating is an important evolutionary mechanism at CVNP (Fisher’s exact test, P = 0.016;  = 0.05, ES = 0.146, power = 0.644). Whether non-significant trends in the other polymorphic populations

(EW and EH) are actually representative of random mating is difficult to demonstrate based on sample size; however, both of these sites had greater power to detect effect sizes compared to CVNP (EW,  = 0.05, ES = 0.124, power = 0.108; EH,  = 0.05, ES =

0.161, power = 0.127). If our data represent true patterns, there is geographic variation in color assortative mating in P. cinereus. Plethodon cinereus has a large geographic distribution and many populations vary greatly in the frequency of the two morphs

(Moore & Ouellet, 2015; Hantak et al., 2019). As a result, selection pressures may vary geographically due to variation in coevolutionary dynamics. For example, with regard to color assortative mating, some populations may be coevolutionary hotspots, with morphs experiencing reciprocal selection, and other populations may be selection coldspots

(Thompson, 1999; Gomulkiewicz et al., 2000). Additional studies on color assortative mating in P. cinereus in a geographic framework will provide more information on how variation in selective pressures may play a role the distribution of color morphs. 105

Previous research on assortative mating in P. cinereus at CVNP also demonstrated that larger females were paired more often with striped males (Antony et al., 2008; Acord et al., 2013). The authors suggested that these results may be due to striped males consuming more energetically profitable prey and holding higher-quality territories (Antony et al., 2008; Reiter et al., 2014; Stuczka et al., 2016). As such, we also predicted that larger females would prefer striped males, but our data did not support this prediction; overall, there was no indication that that color morph had any impact on mate preference. We did, however, find positive assortative mating by body size in two populations, EH and CF. Previous studies have shown that female P. cinereus prefer males that hold high-quality territories and are larger in body size (Walls et al., 1989;

Mathis 1991a). Males have also been shown to display female preference, with males preferring larger females that have larger clutches (Jaworski et al., 2018).

Size-assortative mating in P. cinereus is noteworthy because it may not be as common as previously thought (Green, 2019), and the geographic variability found in this study warrants greater attention. Positive assortative mating by body size was found in

CF, a monomorphic striped population, and EH, a polymorphic population that has a higher frequency of the unstriped morph (92% unstriped). In P. cinereus, size-assortative mating may be a mechanism that drives reproductive isolation across populations. Hantak et al. (2019) found low levels of genetic differentiation between monomorphic striped populations in Eastern Ohio. For instance, SVF (site 25) and CF (site 27) are separated by

FST = 0.03. By contrast, sites in the western portion of Ohio, which typically contain a higher frequency of the unstriped morph, were more geographically and genetically 106 separated from other sites. For example, EH (site 2) is separated from adjacent populations by FST  0.25. The combination of these factors, including size-assortative mating, reduced gene flow, and increased genetic drift, may have important evolutionary consequences for the EH population.

Based on our findings, we suggest that random mating may aid in the maintenance of the color polymorphism in P. cinereus within some populations. Much of the reproductive biology of P. cinereus is enigmatic due to their unwillingness to mate in laboratory settings (Highton, 1959). However, we know that females of either color morph can produce mixed clutches containing both morphs (Highton, 1959, 1975).

Paternity is difficult to estimate in P. cinereus, as males typically do not remain with clutches, females can store sperm, and females often mate with multiple males (Sever,

1997; Liebgold et al., 2006). Further, Highton (1975) found that there is geographic variation in the genetic architecture of the traits that determine the polymorphism. Thus, there is much to still uncover regarding mating patterns and gene interactions in P. cinereus.

Other mechanisms may also be contributing to the maintenance of the striped/unstriped polymorphism, including spatial and temporal variation in selection, negative frequency dependent selection, gene flow among populations, or niche partitioning (Gray & McKinnon, 2007; McLean & Stuart-Fox, 2014). Hantak and Kuchta

(2018) tested the role of spatiotemporal variation in camouflage in maintaining the polymorphism. This study found that the striped morph was better camouflaged than the unstriped morph across populations, seasons, and predator types; thus, visual predation 107 does not appear to play a role in maintaining the polymorphism, although there was geographic variation in the levels of camouflage (Hantak & Kuchta, 2018). Fitzpatrick et al. (2009) documented negative frequency dependent selection (NFDS) by avian predators using clay model replicas of the P. cinereus morphs; however, this study was only conducted in one population and Kraemer et al. (2016) found no evidence of NFDS by mammalian predators. A recent study by Hantak et al. (2019) demonstrated that a combination of gene flow, genetic drift, and selection appear to interact to maintain the polymorphism, but the selective pressures remain unknown. Finally, as the morphs in P. cinereus appear to differ along several ecological axes, examining the importance of niche partitioning is critical, but has thus far been understudied in a geographic context.

Overall, in such a geographically widespread species (Petranka, 1998), it appears likely that there is spatial variation in the mechanisms that aid in the maintenance of the polymorphism.

This study demonstrates that random mating may be an important mechanism contributing to color morph maintenance in P. cinereus and contrasts previous investigations (Anthony et al., 2008; Acord et al., 2013). Several studies have demonstrated ecological differences between the morphs (reviewed in Anthony &

Pfingsten, 2013); however, these studies have largely taken place in single populations at single points in time. Re-examination of previous findings have resulted in ambiguity in which traits differ between morphs (Moore & Ouellet, 2015; Hantak et al., Chapter 4).

For instance, Moore and Ouellet (2015) tested the validity of using P. cinereus as an indicator of climate change after previous studies demonstrated morph specific 108 differences in climate preference (Gibbs & Karraker, 2006; Anthony et al., 2008), but, in contrast with previous studies, they found that climate does not influence variation in color morph frequency in P. cinereus. Further investigation of trait variation in P. cinereus may elucidate how morphs are being maintained over spatiotemporal scales, and if reproductive isolation is occurring.

109

Tables

Table 5.1. Results of the best-supported GLMs for male and female color morph preference across the three polymorphic sites: Cuyahoga Valley National Park (CVNP), Edison Woods Reservation (EW), and East Harbor State Park (EH). No models were significant. Site Sex (Response) Best-fit Predictors Wald P CVNP Female Male Color Morph * Body Size -0.98 0.327 CVNP Male Female Color Morph * Body Size -0.01 0.994 EW Female Male Color Morph 0.73 0.465 EW Male Female Body Size -1.12 0.265 EH Female Male Body Size 1.16 0.247 EH Male Female Color Morph -0.01 0.996

110

Figures

Figure 5.1. Map of color morph frequencies (black = unstriped, red = striped) in Ohio. Study sites: Squire Valleevue Farm (SVF), Chapin Forest Reservation (CF), Cuyahoga Valley National Park (CVNP), Edison Woods Reservation (EW), East Harbor State Park (EH), and South Bass Island (SBI). The photograph displays the unstriped (left) and striped (right) color morphs.

111

Figure 5.2. Distribution of mating pairs from the three polymorphic sites: Cuyahoga Valley National Park (CVNP), Edison Woods Reservation (EW), and East Harbor State Park (EH). Pie charts display color morph frequency at each site (gray = striped morphs; black = unstriped morph). 112

Figure 5.3. Female and male body size regressions from each population: Squire Valleevue Farm (SVF), Chapin Forest Reservation (CF), Cuyahoga Valley National Park (CVNP), Edison Woods Reservation (EW), East Harbor State Park (EH), and South Bass Island (SBI). Populations with an asterisk (*) next to their name display significant positive assortative mating by body size.

113

CHAPTER 6: CONCLUSIONS

An underexplored question in color polymorphic species is: why are morph frequencies variable among populations (McLean & Stuart-Fox, 2014)? Little work has been done to examine geographic patterns in polymorphisms, with most studies focusing on one trait in one or a few populations. Mechanisms that may maintain polymorphisms within and among populations include frequency dependent selection, genetic drift, and gene flow (Gray & McKinnon, 2007). Investigating these mechanisms in multiple populations provides insight into the processes underlying the maintenance of genetic variation within and among populations (Endler, 1977; Thompson, 2005).

The Eastern Red-backed Salamander, Plethodon cinereus, is widespread throughout northeastern North America and is color polymorphic throughout much of its range. Two color morphs are common: a ‘striped’ morph that has a red stripe overlaid on a black dorsum, and an ‘unstriped’ morph that is completely black. Previous studies have demonstrated that the two morphs are differentiated along multiple niche axes, but the ecological and evolutionary dynamics of the polymorphism remain poorly understood.

In chapter 1, I summarized the literature on the polymorphism in P. cinereus and describe the trends. The striped morph is better adapted to cooler, wetter habitats, while the unstriped morph, which has a lower metabolic rate, is better adapted to warmer, drier conditions. The diet of the striped morph is more diverse and energetically profitable than the diet of the unstriped morph, and the striped morph exhibits higher levels of territorial behavior. Finally, the morphs appear to differ in response to predators and mate assortatively by color. Though these cumulative studies identify general morph trends, 114 the majority of studies are of single populations within a single year. Future research should aim to study the polymorphism through space and time as variation in ecological conditions will alter coevolutionary dynamics.

In chapter 2, I investigated whether spatial and temporal variation in selection maintain the two color morphs. Color polymorphisms are often hypothesized to be maintained by interactions with visual predators, either via apostatic selection or by being differentially camouflaged in different habitats, seasons or populations. To identify if the striped or unstriped color morph of P. cinereus is more camouflaged, I collected reflectance measurements from salamanders and the distribution of background colors from three populations over two seasons. Given avian, snake, and mammalian visual models, I found striped morphs were better camouflaged than unstriped morphs against most background types; however, the level of camouflage was dependent on population and season. These results highlight the value of considering spatial and temporal dimensions when testing hypotheses regarding the origin and maintenance of polymorphisms.

Chapter 3 explored the relationship between color morph frequency, landscape heterogeneity, and genetic structure in P. cinereus. Landscape heterogeneity plays an important role in population structure and divergence, particularly for species with low vagility. In this study, I used a landscape genetic approach to identify how landscape and environmental variables affect genetic structure and color morph frequency in a polymorphic salamander, P. cinereus, in northern Ohio. Resistance distance and least cost path analyses were used to test whether genetic distance was more correlated with 115 morph frequency, elevation, canopy cover, waterways, ecological niche, or geographic distance. In addition, I examined whether landscape and environmental variables, genetic distance, or geographic distance were correlated with variation in morph frequency.

Genetic distance was most correlated with the ecological niche model, elevation, and a combination of landscape and environmental variables. In contrast, morph frequency variation was most correlated with waterways and geographic distance. These results suggest that a balance between gene flow, genetic drift, and selection may function to maintain the two color morphs.

In chapter 4, I examined if morphs partition prey resources over space (6 sites) and time (2 seasons). Based on previous studies, I predicted striped morph diet would consist of higher quality prey in polymorphic populations, whereas in monomorphic populations, I predicted both morphs would have a more variable diet due to ecological release from inter-morph competition. In the two polymorphic populations I examined, one showed no evidence of diet differences and the other mirrored differences reported from previous studies. There was no change in dietary breadth between polymorphic and monomorphic populations, and thus no signature of dietary release. These results show there is a high degree of overlap between dietary and leaf litter invertebrates, suggesting both morphs of P. cinereus are generalist predators. Finally, dietary composition varied across seasons and populations, which demonstrates the importance of examining morph traits over spatial and temporal scales.

Chapter 5 investigated whether morphs of P. cinereus assortatively mate by color and/or body size. Previous studies have demonstrated that the morphs assortatively mate 116 by color in one population in Ohio. In addition, these studies also found larger females paired more often with striped males. I re-examined these hypotheses in a geographic framework, across six populations that vary in color morph frequency from monomorphic striped to polymorphic to nearly monomorphic unstriped. I found that morphs do not assortatively mate by color, but instead demonstrate random mating, which may aid in color morph maintenance. I also found geographic variation in size-assortative mating, which may lead to reproductive isolation among some pairs of populations.

Alternative adaptive morphs, such as the striped and unstriped form of P. cinereus, represent important elements in the evolution of ecologically relevant diversity.

When two or more morphs exist within a population, diversifying selection will favor certain trait combinations over others (Schluter, 2000). However, it is well documented that selection pressures differ across both space and time (Calsbeek et al., 2012). And it is possible that populations, which vary in morph frequency, may evolve in a complex spatial mosaic of ecological and evolutionary processes due to these differences in selection (Thompson, 2005). Overall, my studies of P. cinereus provide essential information for understanding the ecological and evolutionary basis of the two forms and contribute to our understanding of the relationship between polymorphism and diversity.

117

REFERENCES

Acord M. A., Anthony C. D., Hickerson, C. M. 2013. Assortative mating in a polymorphic salamander. Copeia 4: 676–683.

Adams D. C. 2000. Divergence of trophic morphology and resource use among populations of Plethodon cinereus and P. hoffmani in Pennsylvania. The biology of plethodontid salamanders Springer US 383–394.

Adams D. C. 2010. Parallel evolution of character displacement driven by competitive selection in terrestrial salamanders. BMC Evolutionary Biology 10: 72.

Adams D. C., West M. E., Collyer M. L. 2007. Location-specific sympatric morphological divergence as a possible response to species interactions in West Virginia Plethodon salamander communities. Journal of Animal Ecology 76: 289–295.

Adriaensen F., Chardon J. P., De Blust G., Swinnen E., Villalba S., Gulinck H., Matthysen E. 2003. The application of ‘‘least- cost’’ modeling as a functional landscape model. Landscape Urban Plan 64: 233–247.

Aldous C. M. 1937. Notes on the life history of the snowshoe hare. Journal of Mammalogy 18: 46–57.

Allen J. A. 1988. Frequency-dependent selection by predators [and Discussion]. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 319: 485–503.

Allen J. A., Raison H. E., Weale M. E. 1998. The influence of density on frequency- dependent selection by wild birds feeding on artificial prey. Proceedings of the Royal Society B 265: 1031–1035.

Anderson M. J. 2006. Distance-based tests for homogeneity of multivariate dispersions. Biometrics 62: 245–253.

Anderson M. T., Mathis A. 1999. Diets of two sympatric neotropical salamanders, Bolitoglossa mexicana and B. rufescens, with notes on reproduction for B. rufescens. Journal of Herpetology 33: 601–607.

Andersson M. 1994. Sexual Selection. Princeton, New Jersey: Princeton University Press.

Andersson S., Prager M. 2006. Quantifying colors. In: Mcgraw KJ, Hill GE, eds. Bird coloration, Vol. I. Cambridge, MA: Harvard University Press, 41–89.

118

Anthony C. D., Pfingsten R. A. 2013. Eastern red-backed salamander, Plethodon cinereus. In: Pfingsten R. A., Davis J. G., Matson T. O., Lipps G. Jr., Wynn D., Armitage B. J., eds. Amphibians of Ohio, Vol. 17, No. 1, Ohio Biological Survey Bulletin New Series. Columbus, Ohio: Ohio Biological Survey, 335–360.

Anthony C. D., Venesky M. D., Hickerson C. A. 2008. Ecological separation in a polymorphic terrestrial salamander. The Journal of Animal Ecology 77: 646–653.

Anthony C. D., Jaworski K., Messner M., Hickerson C. M. 2017. Differences in prey availability within the territories of striped and unstriped Eastern Red-backed Salamanders (Plethodon cinereus). Herpetological Review 48: 509–514.

Antoniazza S., Burri R., Fumagalli L., Goudet J. R. M., Roulin A. 2010. Local adaptation maintains clinal variation in melanin-based coloration of European Barn Owls (Tyto alba). Evolution 85: 797–807.

Aubier T. G., Sherratt T. N. 2015. Diversity in Müllerian mimicry: the optimal predator sampling strategy explains both local and regional polymorphism in prey. Evolution 69: 2831–2845.

Austin R. M. Jr. 2000. Cutaneous microbial flora and antibiosis in Plethodon ventralis: Inferences for parental care in the Plethodontidae. In: Bruce R. C., Jaeger R. B., Houck L. D. (eds) The biology of the plethodontid salamanders. Kluwer Academic/Plenum, New York.

Avon C., Bergès L. 2016. Prioritization of habitat patches for landscape connectivity conservation differs between least-cost and resistance distances. Landscape Ecology 31: 1551–1565.

Barton K. 2019. MuMIN: multi-model inference. R package version 1.43.6. Available at: https://cran.r-project.org/web/packages/MuMIn/index.html.

Bates D., Maechler M., Bolker B., Walker S. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67: 1–48.

Beerli P. 2008. Migrate version 3.6.5: a maximum likelihood and Bayesian estimator of gene flow using the coalescent. Distributed over the internet at http://popgen.scs.edu/ migrate.html.

Beerli P., Palczewski M. 2010. Unified framework to evaluate panmixia and migration direction among multiple sampling locations. Genetics 185: 313–326.

Boback S. M., Siefferman L. M. 2010. Variation in color and color change in island and mainland boas (Boa constrictor). Journal of Herpetology 44: 506–515. 119

Bolker B. M., Brooks M. E., Clark C. J., Geange S. W., Poulsen J. R., Stevens M. H. H., White J.-S. S. 2009. Generalized linear mixed models: a practical guide for ecology and evolution. Trends in Ecology & Evolution 24: 127–135.

Bolnick D. I, Kirkpatrick M. 2012. The relationship between intraspecific assortative mating and reproductive isolation between divergent populations. Current Zoology 58: 484–492.

Bolnick D. I., Ingram T., Stutz W. E., Snowberg L. K., Lau O. L., Paull J. S. 2010. Ecological release from interspecific competition leads to decoupled changes in population and individual niche width. Proceedings of the Royal Society B 277: 1789– 1797.

Bond A. B. 2007. The evolution of color polymorphism: crypticity, searching images, and apostatic selection. Annual Review of Ecology, Evolution and Systematics 38: 489– 514.

Bond A. B., Kamil A. C. 2006. Spatial heterogeneity, predator cognition, and the evolution of color polymorphism in virtual prey. Proceedings of the National Academy of Sciences USA 103: 3214–3219.

Bondi C. A., Beier C.M., Fierke M. K. Ducey P. K. 2019. The role of feeding strategy in the tolerance of a terrestrial salamander (Plethodon cinereus) to biogeochemical changes in northern hardwood forests. Canadian Journal of Zoology 97: 281–293.

Bondi C. A., Green S., Beier C. M. 2015. Evaluation of gastric lavage method for diet analysis of the Eastern Red-backed Salamander (Plethodon cinereus). Herpetological Review 46: 196–199.

Brodie E. D. Jr. 1977. Salamander antipredator postures. Copeia 1977: 523–535.

Brodie E. D. III., Brodie E. D. Jr. 1990. Tetrodotoxin resistance in garter snakes: an evolutionary response of predators to dangerous prey. Evolution 44: 651–659.

Brodie E. D. Jr., Nowak R. T., Harvey W. R. 1979. The effectiveness of antipredator secretions and behavior of selected salamanders against shrews. Copeia 1979: 270–274.

Brodie E. D. Jr, Ridenhour B. J., Brodie E. D. III. 2002. The evolutionary response of predators to dangerous prey: hotspots and coldspots in the geographic mosaic of coevolution between garter snakes and newts. Evolution 56: 2067–2082.

120

Burton T. M. 1976. An analysis of the feeding ecology of salamanders (Amphibia, Urodela) of the Hubbard Brook Experimental Forest, New Hampshire. Journal of Herpetology 10: 187–204.

Burton T. M., Likens G. E. 1975. Salamander populations and biomass in the Hubbard Brook Experimental Forest, New Hampshire. Copeia 1975: 541–546.

Cabe P. R., Page R. B., Hanlon T. J., Aldrich M. E., Connors L. Marsh, D. M. 2007. Fine scale population differentiation and gene flow in a terrestrial salamander (Plethodon cinereus) living in continuous habitat. Heredity 98: 53–60.

Cain A. J., Sheppard P. M. 1954. Natural selection in Cepaea. Genetics 39: 89–116.

Calsbeek R., Gosden T., Kuchta S. R., Svensson E. I. 2012. Fluctuating selection and dynamic adaptive landscapes. In: Svensson EI, Calsbeek R, eds. The adaptive landscape in evolutionary biology. Oxford: Oxford University Press, 89–109.

Cameron A. C., Anderson J. J. Page R. B. 2017. Assessment of intra and interregional genetic variation in the Eastern Red-backed Salamander, Plethodon cinereus, via analysis of novel microsatellite markers. PLoS ONE 12: e0186866.

Caro T. 2009. Contrasting coloration in terrestrial mammals. Philosophical Transactions of the Royal Society B 364: 537–548.

Champely S., Ekstrom C., Dalgaard P., Gill J., Weibelzahl S., Anandkumar A., Ford C., Volcic R., De Rosario H. 2018. R package pwr. https://cran.r- project.org/web/packages/pwr/index.html.

Chapuis M. P., Estoup A. 2007. Microsatellite null alleles and estimation of population differentiation. Molecular Biology and Evolution 24: 621–631.

Clarke B. C. 1962. Balanced polymorphism and the diversity of sympatric species. In: Nichols D, ed. Taxonomy and geography. Oxford: Systematics Association, 47–70.

Clarke B. C. 1969. The evidence for apostatic selection. Heredity 24: 347–353.

Clarke R. T., Rothery P., Raybould A. F. 2002. Confidence limits for regression relationships between distance matrices: estimating gene flow with distance. Journal of Agricultural, Biological, and Environmental Statistics 7: 361–372.

Comendant T., Sinervo B., Svensson E.I., Wingfield J. 2003. Social competition, corticosterone and survival in female lizard morphs. Journal of Evolutionary Biology 16: 948– 955.

121

Converse P. E., Kuchta S. R., Roosenburg W. M., Henry P. F. P, Haramis G. M., King T. L. 2015. Spatiotemporal analysis of gene flow in Chesapeake Bay Diamondback Terrapins (Malaclemys terrapin). Molecular Ecology 24: 5864–5876.

Corl A., Davis A. R., Kuchta S. R., Sinervo B. 2010. Selective loss of polymorphic mating types is associated with rapid phenotype evolution during morphic speciation. Proceedings of the National Academy of Sciences USA 107: 4254–4259.

Cosentino B. J., Moore J. D., Karraker N. E., Ouellet M., Gibbs J. P. 2017. Evolutionary response to global change: Climate and land use interact to shape color polymorphism in a woodland salamander. Ecology and Evolution 7: 5426–5434.

Costa, A., Salvidio S., Posillico M., Matteucci G., De Cinti B., Romano A. 2015. Generalisation within specialization: inter-individual diet variation in the only specialized salamander in the world. Scientific Reports 5: 13260.

Cott H. B. 1940. Adaptive coloration in animals. London, UK: Methuen & Co. Ltd.

Crane A. L., Ferrari M. C. O. 2017. Patterns of predator neophobia: a meta-analytic review. Proceedings of the Royal Society B 284: 20170583.

Davis A. K., Milanovich J. R. 2010. Lead-phase and red-stripe color morphs of red- backed salamanders Plethodon cinereus differ in hematological stress indices: a consequence of differential predation pressure? Current Zoology 56: 238–243.

Davis-Rabosky A. R., Cox C. L., Rabosky D. L. 2016. Unlinked Mendelian inheritance of red and black pigmentation in snakes: Implications for Batesian mimicry. Evolution 70: 944–953.

Dayan T., Simberloff D. 1994. Character displacement, sexual dimorphism, and morphological variation among British and Irish mustelids. Ecology 75: 1063–1073.

Debelle A., Ritchie M. G., Snook R. R. 2014. Evolution of divergent female mating preference in response to experimental sexual selection. Evolution 68: 2524–2533.

Deitloff J., Adams D. C., Olechnowski B. F., Jaeger R. G. 2008. Interspecific aggression in Ohio Plethodon: implications for competition. Herpetologica 64: 180–188.

Deitloff, J. Peterson J. D., Adams D. C. 2013. Complex species interactions lead to unpredictable outcome in Plethodon. Herpetologica 69: 1-10.

Delhey K., Peters A., Johnsen A., Kempenaers B. 2006. Seasonal changes in blue tit crown color: do they signal individual quality? Behavioral Ecology 17: 790–798.

122

Denoël M., Whiteman H. H., Wissinger S. A. 2006. Temporal shift of diet in alternative cannibalistic morphs of the tiger salamander. Biological Journal of the Linnean Society 89: 373–382.

Dieckmann U., Doebeli M. 1999. On the origin of species by sympatric speciation. Nature 400: 354–357.

Dunham A. E. 1983. Realized niche overlap: Resource abundance and intensity of interspecific competition. Pp. 261–280, In Huey R. B., Pianka E. R., Shoener T. (Eds.). Lizard Ecology: Studies of a Model Organism. Harvard University Press, Cambridge, MA. 501 pp.

Earl D. A., vonHoldt B. M. 2012. Structure Harvester: a website and program for visualizing Structure output and implementing the Evanno method. Conservation Genetics Resources 4: 359–361.

Eddy S. L., Wilburn D. B., Chouinard A. J., Doty K. A., Kiemnec-Tyburczy K. M., Houck L. D. 2016. Male terrestrial salamanders demonstrate sequential mate choice based on female gravidity and size. Animal Behaviour 113: 23–29.

Elmer K. R., Lehtonen T. K., Meyer A. 2009. Color assortative mating contributes to sympatric divergence of neotropical cichlid fish. Evolution 63: 2750–2757.

Endler J. A. 1977. Geographic variation, speciation and clines. Princeton: Princeton University Press.

Endler J. A. 1978. A predator's view of animal color patterns. Evolutionary Biology 11: 319–364.

Endler J. A. 1980. Natural selection on color patterns in Poecilia reticulata. Evolution 34: 76–91.

Endler J. A. 1990. On the measurement and classification of colour in studies of animal colour patterns. Biological Journal of the Linnean Society 41: 315–352.

Endler J. A. 1993. The color of light in forests and its implications. Ecological Monographs 63: 1–27.

Endler J. A., Greenwood J. J. D. 1988. Frequency-dependent predation, crypsis and aposematic coloration. Philosophical Transactions of the Royal Society B 319: 505–523.

Erlandsson J., Rolán-Alvarez E. 1998. Sexual selection and assortative mating by size and their roles in the maintenance of a polymorphism in Swedish Littorina saxatilis populations. Hydrobiologia 378: 59–69. 123

Evanno G., Regnaut S., Goudet J. 2005. Detecting the number of clusters of individuals using the software structure: a simulation study. Molecular Ecology 14: 2611–2620.

Evans A. E., Forester B. R., Jockusch E. L., Urban M. C. 2018. Salamander morph frequencies do not evolve as predicted in response to 40 years of climate change. Ecography 41: 1687–1697.

Farallo V. R., Forstner M. R. 2012. Predation and the maintenance of color polymorphism in a habitat specialist squamate. PLoS One 7: e30316.

Farallo V. R., Miles D. B. 2016. The importance of microhabitat: A comparison of two microendemic species of Plethodon to the widespread P. cinereus. Copeia 104: 67–77.

Feder M. E. 1983. Integrating the ecology and physiology of plethodontid salamanders. Herpetologica 39: 291–310.

Fisher-Reid M. C., Engstrom T. N., Kuczynski C. A., Stephens P. R., Wiens J. J. 2013. Parapatric divergence of sympatric morphs in a salamander: incipient speciation on Long Island? Molecular Ecology 22: 4681–4694.

Fisher-Reid C. M., Wiens J. J. 2015. Is geographic variation within species related to macroevolutionary patterns between species? Journal of Evolutionary Biology 28: 1502– 1515.

Fitzpatrick B. M., Shook K., Izally R. 2009. Frequency-dependent selection by wild birds promotes polymorphism in model salamanders. BMC Ecology 2009: 9–12.

Foote C. J., Larkin P. A. 1988. The role of male choice in the assortative mating of anadromous and non-anadromous sockeye salmon (Oncorhynchus nerka). Behaviour 106: 43–62.

Ford E. B. 1945. Polymorphism. Biological Reviews 20: 73–88.

Forsman A., Ahnesjö J., Caesar S., Karisson M. 2008. A model of ecological and evolutionary consequences of color polymorphism. Ecology 89: 34–40.

Forsman A., Karlsson M., Wennersten L., Johansson J., Karpestam E. 2011. Rapid evolution of fire melanism in replicated populations of pygmy grasshoppers. Evolution 65:2530–2540.

Funk W. C., Blouin M. S., Corn P. S., Maxell B. A., Pilliod D. S., Amish S. Allendorf F. W. 2005. Population structure of Columbia spotted frogs (Rana luteiventris) is strongly affected by the landscape. Molecular Ecology 14: 483–496. 124

Funk W. C., Cannatella D. C. Ryan M. J. 2009. Genetic divergence is more tightly related to call variation than landscape features in the Amazonian frogs Physalaemus petersi and P. freibergi. Journal of Evolutionary Biology 22: 1839–1853.

Gabor C. R. 1995. Correlational test of Mathis’ hypothesis that bigger salamanders have better territories. Copeia 1995: 729–735.

Gergits W. F., Jaeger R. G. 1990. Field observations of the behavior of the red-backed salamander (Plethodon cinereus): courtship and agonistic interactions. Journal of Herpetology 24: 93–95.

Gibbs J. P., Karraker N. E. 2006. Effects of warming conditions in Eastern North American forests on red-backed salamander morphology. Conservation Biology 20: 913– 917.

Gillette J. R. 2003. Population ecology, social behavior, and intersexual differences in a natural population of red-backed salamanders: a long-term field study. Dissertation, University of Louisiana, Lafayette.

Gillette J. R., Jaeger R. G., Peterson M. G. 2000. Social monogamy in a territorial salamander. Animal Behaviour 59: 1241–1250.

Giordano A. R., Ridenhour B. J., Storfer A. 2007. The influence of altitude and topography on genetic structure in the long-toed salamander (Ambystoma macrodactulym). Molecular Ecology 16: 1625–1637.

Goldsmith T. H. 1990. Optimization, constraint, and history in the evolution of eyes. The Quarterly Review of Biology 65: 281–322.

Gosden T. P., Svensson E. I. 2008. Spatial and temporal dynamics in a sexual selection mosaic. Evolution 62: 845–856.

Grant A. H. Liebgold E. B. 2017. Color-biased dispersal inferred by fine-scale genetic spatial autocorrelation in a color polymorphic salamander. Journal of Heredity 108: 588– 593.

Grant A. H. Ransom T. S. Liebgold E. B. 2018. Differential survival and the effects of predation on a color polymorphic species, the red-backed salamander (Plethodon cinereus). Journal of Herpetology 52: 127–135.

Gray S. M., McKinnon J. S. 2007. Linking color polymorphism maintenance and speciation. Trends in Ecology & Evolution 22: 71–79.

125

Green, D. M. 2019. Rarity of size-assortative mating in animals: assessing the evidence with Anuran amphibians. American Naturalist 193: 279–295.

Greenwood J. J. D. 1984. The functional basis of frequency- dependent selection. Biological Journal of the Linnean Society 23: 177–199.

Hantak M. M., Brooks K. M., Hickerson C. M., Anthony C. D., Kuchta, S. R. In prep. A spatiotemporal assessment of dietary partitioning between color morphs of a terrestrial polymorphic salamander.

Hantak M. M., Kuchta S. R., Hickerson C. M., Anthony C. D. 2015. Plethodon cinereus: morphology. Herpetological Review 46: 409–410.

Hantak M. M., Kuchta S. R. 2018. Predator perception across space and time: relative camouflage in a colour polymorphic salamander. Biological Journal of the Linnean Society 123: 21–33.

Hantak M. M., Page R. B., Converse P. E., Anthony C. D., Hickerson C. M., Kuchta S. R. 2019. Do genetic structure and landscape heterogeneity impact color morph frequency in a polymorphic salamander? Ecography 42: 1383–1394.

Hantak M. M., Paluh D. J., Hickerson C. M. 2016. Comparison of the diets of sympatric erythristic and striped morphs of Plethodon cinereus (Eastern Red-backed Salamander). Northeastern Naturalist 23: 219–228.

Hart N. S., Partridge J. C., Cuthill I. C., Bennett A. T. 2000. Visual pigments, oil droplets, ocular media and cone photoreceptor distribution in two species of passerine bird: the blue tit (Parus caeruleus L.) and the blackbird (Turdus merula L.). Journal of Comparative Physiology A 186: 375–387.

Hervé M. 2019. RVAideMemoire: testing and plotting procedures for biostatistics. R package version 0.9. Available at: https://cran.r- project.org/web/packages/RVAideMemoire/index.html.

Highton R. 1959. The inheritance of the color phases of Plethodon cinereus. Copeia 1959: 33–37.

Highton, R. 1962. Revision of North American salamanders of the genus Plethodon. – Bull. Florida State Mus. 6: 236–367.

Highton R. 1975. Geographic variation in genetic dominance of the color morphs of the red-backed salamander, Plethodon cinereus. Genetics 80: 363–374.

126

Highton R. 2004. A new species of woodland salamander of the Plethodon cinereus group from the Blue Ridge Mountains of Virginia. Jeffersoniana 14: 1–22.

Highton R., Savage T. 1961. Functions of the brooding behavior in the female Red- backed Salamander, Plethodon cinereus. Copeia 1961: 95–98.

Hijmans R. J., Cameron S. E., Parra J. L., Jones P. G., Jarvis A. 2005. Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25: 1965–1978.

Holmes, I. A., Grundler M. R., Davis-Rabosky A. R. 2017. Predator perspective drives geographic variation in frequency-dependent polymorphism. American Naturalist 190: E78–E90.

Hugall A. F., Stuart-Fox D. 2012. Accelerated speciation in colour-polymorphic birds. Nature 485: 631–634.

Huxley J. S. 1955. Morphism in birds. Acta 6th International Ornithological Congress, Basel 1954: 309–328.

Huyghe K., Small M., Vanhooydonck B., Herrel A., Tadic Z., Van Damme R., Backeljau, T. 2010. Genetic divergence among sympatric colour morphs of the Dalmatian all lizard (Podarcis melisellensis). Genetica 138: 387–393.

Ivanov, K., Keiper J. 2009. Effectiveness and biases of winkler litter extraction and pitfall trapping for collecting ground-dwelling ants in northern temperate forests. Environmental Entomology 38: 1724–1736.

Jacobs G. H., Neitz J., Crognale M. 1985. Spectral sensitivity of ground squirrel cones measured with ERG flicker photometry. Journal of Comparative Physiology A 156: 503– 509.

Jaeger R. G. 1980. Fluctuations in prey availability and food limitation for a terrestrial salamander. Oecologia 44: 335–341.

Jaeger R. G. 1981. Dear enemy recognition and the costs of aggression between salamanders. American Naturalist 117: 962–974.

Jaeger R. G. 1990. Territorial salamanders evaluate size and chitinous content of arthropod prey. Behavioral Mechanisms of Food Selection (ed. Hughes R. N.), pp. 111– 126. NATO ASI Series, Subseries G. Ecological Sciences. Springer-Verlag, Heidelberg.

Jaeger R. G., Barnard D. E. 1981. Foraging tactics of a terrestrial salamander: choice of diet in structurally simple environments. American Naturalist 117: 639–664. 127

Jaeger R. G., Forester D. C. 1993. Social behavior of plethodontid salamanders. Herpetologica 49: 163–175.

Jaeger R. G., Gillette J. R., Cooper R. C. 2002. Sexual coercion in a territorial salamander: males punish socially polyandrous female partners. Animal Behaviour 63: 871–877.

Jaeger R. G., Gollman B., Anthony C. D., Gabor C. R., Kohn N. 2016. Behavioral ecology of the Eastern Red-backed Salamander: 50 years of research. Oxford University Press.

Jaeger R. G., Joseph R. G., Barnard D. E. 1981. Foraging tactics of a terrestrial salamander: sustained yield in territories. Animal Behaviour 29: 1100–1105.

Jaeger R. G., Rubin A. M. 1982. Foraging tactics of a terrestrial salamander: judging prey profitability. Journal of Animal Ecology 51: 167–176.

Jakobsson M., Rosenberg, N. A. 2007. CLUMPP: a cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics 23: 1801–1806.

Jamison J. A., Harris R. N. 1992. The priority of linear over volumetric caudal regeneration in the salamander Plethodon cinereus (Caudata: Plethodontidae). Copeia 1992: 235–237.

Janes J. K., Miller J. M., Dupuis J. R., Malenfant R. M., Gorrell J. C., Cullingham C. I., Andrew R. L. 2017. The K = 2 conundrum. Molecular Ecology 26: 3594–3602.

Jaworski K. E., Lattanzio M. S., Hickerson C. M., Anthony C. D. 2018. Male mate preference as an agent of fecundity selection in a polymorphic salamander. Ecology and Evolution 8: 8750–8750.

Jiang Y., Bolnick D. I., Kirkpatrick M. 2013. Assortative mating in animals. American Naturalist 181: E125–138.

Jiggins C. D., Naisbit R. E., Coe R. L., Mallet J. 2001. Reproductive isolation caused by color pattern mimicry. Nature 411: 302–305.

Johnston G. 1994. Seasonal changes in dorsal reflectance of two species of Australian elapid snakes. Journal of Herpetology 28: 109–112.

Jombart T. 2008. adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics 24: 1403–1405. 128

Jombart T., Devillard S., Balloux F. 2010. Discriminant analysis of principal components: a new method for the analysis of genetically structured populations. BMC Genetics 11: 94.

Jost L. 2008. GST and its relatives do not measure differentiation. Molecular Ecology 17: 4015–4026.

Kaaz M. R. 1955. The Black Swamp: a study in historical geography. Annals of the Association of American Geographers 35: 1–35.

Kalinowski S. T. 2005. HP-RARE 1.0: a computer program for performing rarefaction on measures of allelic richness. Molecular Ecology Notes 5: 187–189.

Karhunen M., Ovaskainen O. 2012. Estimating population-level coancestry coefficients by an admixture F model. Genetics 192: 609–617.

Karpestam E., Forsman A. 2011. Dietary differences among colour morphs of pygmy grasshoppers revealed by behavioral experiments and stable isotopes. Evolutionary Ecology Research 13: 461–477.

Karpestam E., Merilaita S., Forsman A. 2013. Detection experiments with humans implicate visual predation as a driver of colour polymorphism dynamics in pygmy grasshoppers. BMC Ecology 13: 17.

Karpestam E., Merilaita S., Forsman A. 2014. Natural levels of colour polymorphism reduce performance of visual predators searching for camouflaged prey. Biological Journal of the Linnean Society 112: 546–555.

Kearse M., Moir R., Wilson A., Stones-Havas S., Cheung M., Sturrock S., Buxton S., Cooper A., Markowitz S., Duran C., Thierer T., Ashton B., Meintjes P. Drummond A. 2012. Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28: 1647–1649.

Kemp D. J., Herberstein M. E., Fleishman L. J., Endler J. A., Bennett A. T., Dyer A. G., Hart N. S., Marshall J., Whiting M. J. 2015. An integrative framework for the appraisal of coloration in nature. The American Naturalist 185: 705–724.

King R. B., Lawson R. 1995. Color-pattern variation in Lake Erie water snakes: the role of gene flow. Evolution 49: 885–896.

Kirkpatrick M. 2000. Reinforcement and divergence under assortative mating. Proceedings of the Royal Society of London B 267: 1649–1655.

129

Kirkpatrick M., Ravigne V. 2002. Speciation by natural and sexual selection: models and experiments. American Naturalist 159: S22–S35.

Kleeberger S. R., Werner J. K. 1982. Home range and homing behavior of Plethodon cinereus in northern Michigan. Copeia 1982:409–415.

Klomp D. A., Stuart-Fox D., Das I., Ord T. J. 2014. Marked colour divergence in the gliding membranes of a tropical lizard mirrors population differences in the colour of falling leaves. Biology Letters 10: 20140776.

Kraemer A. C., Adams D. C. 2014. Predator perception of Batesian mimicry and conspicuousness in a salamander. Evolution 68: 1197–1206.

Kraemer A. C., Kissner J., Adams D. C. 2012. Morphological color-change in the red- backed salamander (Plethodon cinereus) while kept in captivity. Copeia 2012: 748–755.

Kraemer A. C., Serb J. M., Adams D. C. 2016. Both novelty and conspicuousness influence selection by mammalian predators on the colour pattern of Plethodon cinereus (Urodela: Plethodontidae). Biological Journal of the Linnean Society 118: 889–900.

Kuchta S. R. 2005. Experimental support for aposematic coloration in the salamander Ensatina eschscholtzii xanthoptica: implications for mimicry of Pacific newts. Copeia 2005: 265–271.

Kuchta S. R., Krakauer A. H., Sinervo B. 2008. Why does the yellow-eyed Ensatina have yellow eyes? Batesian mimicry of Pacific newts (genus Taricha) by the salamander Ensatina eschscholtzii xanthoptica. Evolution 62: 984–990.

Kuchta S. R., Wake D. B. 2016. Wherefore and whither the ring species? Copeia 104: 189–201.

Legler J. M., Sullivan L. J. 1979. The application of stomach-flushing to lizards and anurans. Herpetologica 35: 107–110.

Liebgold E. B., Brodie E. D. III., Cabe P. R. 2011. Female philopatry and male-biased dispersal in a direct-developing salamander, Plethodon cinereus. Molecular Ecology 20: 249–257.

Liebgold E. B., Cabe P. R., Jaeger R. G., Leberg P. L. 2006. Multiple paternity in a salamander with socially monogamous behavior. Molecular Ecology 15: 4153–4160.

Losos J. B. 1990. A phylogenetic analysis of character dis placement in Caribbean Anolis lizards. Evolution 44: 1189–1203.

130

Losos J. B., de Queiroz K. 1997. Evolutionary consequences of ecological release in Caribbean Anolis lizards. Biological Journal of the Linnean Society 61: 459–483.

Lotter F., Scott N. J. Jr. 1977. Correlation between climate and distribution of the color morphs of the salamander Plethodon cinereus. Copeia 1977: 681–690.

Lowe W. H., Likens G. E., McPeek M. A., Buso D. C. 2006. Linking direct and indirect data on dispersal: isolation by slope in a headwater stream salamander. Ecology 87: 334– 339.

Maerz J. C., Karuzas J. M., Madison D. M., Blossey B. (2005). Introduced invertebrates are important prey for a generalist predator. Diversity and Distributions 11: 82–90.

Maia R., Eliason C. M., Bitton P. P., Doucet S. M., Shawkey M. D. 2013. pavo: an R package for the analysis, visualization and organization of spectral data. Methods in Ecology and Evolution 4: 906–913.

Maiorana V. C. 1977. Tail autonomy, functional conflicts and their resolution by a salamander. Nature 265: 533–535.

Manel S., Holderegger R. 2013. Ten years of landscape genetics. Trends in Ecology & Evolution 28: 614–621.

Manel S., Schwartz M. K., Luikart G., Taberlet P. 2003. Landscape genetics: combining landscape ecology and population genetics. Trends in Ecology & Evolution 18: 189–197.

Marrotte R. R., Bowman J. 2017. The relationship between least-cost and resistance distance. PLoS One 12: e0174212.

Martín J., Salvador A. 1993. Tail loss reduces mating success in the Iberian rock-lizard, Lacerta monticola. Behavioral Ecology and Sociobiology 32: 185–189.

Martinez A. P. 2019. pairwiseAdonis: Pairwise multilevel comparison using adonis. R package version 0.3

Mathis A. 1990. Territoriality in a terrestrial salamander: the influence of resource quality and body size. Behaviour 112: 162–175.

Mathis A. 1991a. Large male advantage for access to females: evidence of male-male competition and female discrimination in a territorial salamander. Behavioral Ecology and Sociobiology 29: 133–138.

Mathis A. 1991. Territories of male and female terrestrial salamanders: costs, benefits, and intersexual spatial associations. Oecologia 86: 433–440. 131

Mathis A., Jaeger R. G., Keen W. H., Ducey P. K., Walls S. C., Buchanan B. W. 1995. Aggression and territoriality by salamanders and a comparison with the territorial behavior of frogs. In: Amphibian Biology, Vol. 2 (Heatwole H., Sullivan B. K., eds.), pp. 633-676. Surrey Beatty and Sons, New South Wales, Australia.

Mazerolle M. J. 2013. Package ‘AICcmodavg’. The R Project for Statistical Computing, Vienna, Austria.

McClure M. L., Hansen A. J., Inman R. M. 2016. Connecting models to movements: testing connectivity model predictions against empirical migration and dispersal data. Landscape Ecology 31: 1419–1432.

Mckinnon J. S., Pierotti M. R. 2010. Colour polymorphism and correlated characters: genetic mechanisms and evolution. Molecular Ecology 19: 5101–5125.

McLean C. A, Stuart-Fox D. 2014. Geographic variation in animal colour polymorphisms and its role in speciation. Biological Reviews 89: 860–873.

McLean C. A., Stuart-Fox D., Moussalli A. 2015. Environment, but not genetic divergence, influences geographic variation in colour morph frequencies in a lizard. BMC Evolutionary Biology 15: 156.

McRae B. 2006. Isolation by resistance. Evolution 60: 1551–1561.

Meirmans P. G., van Tienderen P. H. 2004. Genotype and Genodive: two programs for the analysis of genetic diversity of asexual organisms. Molecular Ecology Notes 4: 792– 794.

Merchant H. 1972. Estimated population size and home range of the salamanders Plethodon jordani and Plethodon glutinosus. Journal of the Washington Academy of Science 62: 248–257.

Merilaita S. 2001. Habitat heterogeneity, predation and gene flow: colour polymorphism in the isopod, Idotea baltica. Evolutionary Ecology 15: 103–116.

Monti L., Hunter M. Jr., Witham J. 2000. An evaluation of the artificial cover object (ACO) method for monitoring population of the Redback Salamander Plethodon cinereus. Journal of Herpetology 34: 624–629.

Moore J. D., Ouellet M. 2014. A review of colour phenotypes of the eastern red-backed salamander, Plethodon cinereus, in North America. Canadian Field Naturalist 128: 250– 259.

132

Moore J. D., Ouellet M. 2015. Questioning the use of an amphibian colour morph as an indicator of climate change. Global Change Biology 21: 566–571.

Moreno G. 1989. Behavioral and physiological differentiation between the color morphs of the salamander, Plethodon cinereus. Journal of Herpetology 23: 335–341.

Muñoz M. M., Crawford N. G., McGreevy T. J. Jr., Messana N. J., Tarvin R. D., Revell L. J., Zandvliet R. M., Hopwood J. M., Mock E., Schneider A. L., Schneider C. J. 2013. Divergence in coloration and ecological speciation in the Anolis marmoratus species complex. Molecular Ecology 22: 2668–2682.

Murray M. P., Pearl C. A., Bury R. B. 2005. Apparent predation by gray jays, Perisoreus Canadensis, on long- toed salamanders, Ambystoma macrodactylum, in the Oregon cascade range. Canadian Field Naturalist 119: 291–292.

Nosil P., Vines T. H., Funk D. J. 2005. Reproductive isolation caused by natural selection against immigrants from divergent habitats. Evolution 59: 705–719.

Ohio Department of Natural Resources (ODNR). 2017. Soil regions. ODNR Department of Agriculture. Available at: www.agri.ohio.gov/divs/SWC/SWC.aspx. Accessed 6 March 2017.

Ohio Department of Natural Resources (ODNR). 2019. Soil regions. ODNR Department of Agriculture. Available at: www.agri.ohio.gov. Accessed 2 May 2019.

Oksanen J., Blanchet F. G., Kindt R., Legendre P., Minchin P. R., O’Hara R. B., Simpson G. L., Solymos P., Stevens M. H. H., Wagner H. 2016. vegan: community ecology package. R package version 2.5-4. https://cran.r-project.org/package=vegan.

Otaibi B. W., Johnson Q. K., Cosentino B. J. (2017). Postautotomy tail movement differs between colour morphs of the red-backed salamander (Plethodon cinereus). Amphibia- Reptilia 38: 395–399.

Ousterhout B. H., Liebgold E. B. 2010. Dispersal versus site tenacity of adult and juvenile Red-backed Salamanders (Plethodon cinereus). Herpetologica 66: 269–275.

Ozgo M. 2011. Rapid evolution in unstable habitats: a success story of the polymorphic land snail Cepaea nemoralis (Gastropoda: Pulmonata). Biological Journal of the Linnean Society 102: 251–262.

Paluh D. J., Eddy C., Ivanov K., Hickerson C. M., Anthony C. D. 2015. Selective foraging on ants by a terrestrial polymorphic salamander. American Midland Naturalist 174: 265–277.

133

Peakall R., Smouse P. E. 2012. GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research – an update. Bioinformatics 28: 2537–2539.

Peterman W. E. 2018. ResistanceGA: an R package for the optimization of resistance surfaces using genetic algorithms. Methods in Ecology and Evolution 49: 340–10.

Peterman W. E., Semlitsch, R. D. 2013. Fine-scale habitat associations of a terrestrial salamander: the role of environmental gradients and implications for population dynamics. PLoS One 8: e62184.

Peterman W. E., Connette G. M., Semlitsch R. D., Eggert L. S. 2014. Ecological resistance surfaces predict fine-scale genetic differentiation in a terrestrial woodland salamander. Molecular Ecology 23: 2402–2413.

Petranka J. W. 1998. Salamanders of the United States and Canada. Smithsonian Press, Washington.

Petruzzi E. E., Niewiarowski P. H., Moore F. B. G. 2006. The role of thermal niche selection in maintenance of a colour polymorphism in redback salamanders (Plethodon cinereus). Frontiers in Zoology 5: 3–10.

Pfingsten R. A., Walker C. F. 1978. Some nearly all black populations of Plethodon cinereus (Amphibia, Urodela, Plethodontidae) in Northern Ohio. Journal of Herpetology 12: 163–167.

Phillips S. J., Anderson R. P., Schapire R. E. 2006. Maximum entropy modeling of species geographic distributions. Ecological Modelling 190: 231–259.

Pianka E. R. 1974. Niche overlap and diffuse competition. Proceedings of the National Academy of Sciences 71: 2141–2145.

Powell R., Parmerlee J. S. Jr., Rice M. A., Smith D. D. 1990. Ecological observations of Hemidactylus brookii haitianus Meerwarth (Sauria: Gekkonidae) from Hispaniola. Caribbean Journal of Science 26: 67–70.

Price T. 1987. Diet variation in a population of Darwin’s Finches. Ecology 68: 1015– 1028.

Pritchard J. K., Stephens M., Donnelly P. 2000. Inference of population structure using multilocus genotype data. Genetics 155: 945–959.

Prugnolle F., de Meeus T. 2002. Inferring sex-biased dispersal from population genetic tools: a review. Heredity 88: 161–165.

134

Punzalan D., Rodd F. H., Hughes K. A. 2005. Perceptual processes and the maintenance of polymorphism through frequency- dependent predation. Evolutionary Ecology 19: 303–320.

R Core Team. 2015. R: A language and environment for statistical computing, R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.

R Core Team. 2018. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.

Radomski T. P. 2017. Biogeography and climatic niche evolution in the eastern red- backed salamander. MS thesis, Ohio University, USA.

Reiter M. K., Anthony C. D., Hickerson C. M. 2014. Terrestrial behavior and ecological divergence in a polymorphic salamander. Copeia 2014: 481–488.

Rolán-Alvarez E., Saura M., Diz A. P., Rivas M. J., Alvarez M., Cortés, B. 2012. Can sexual selection and disassortative mating contribute to the maintenance of a shell color polymorphism in an intertidal marine snail? Current Zoology 58: 463–474.

Rosenberg N. A. 2004. Distruct: a program for the graphical display of population structure. Molecular Ecology Notes 4: 137–138.

Rousset F. 2008. Genepop’007: a complete re-implementation of the genepop software for Windows and Linux. Molecular Ecology Resources 8: 103–106.

Ruxton G. D., Sherratt T. N., Speed M. P. 2004. Avoiding attack: the evolutionary ecology of crypsis, aposematism, and mimicry. Oxford, UK: Oxford University Press.

Sandoval C. P. 1994. The effects of the relative geographic scales of gene flow and selection on morph frequencies in the walking-stick Timema cristinae. Evolution 48: 1866–1879.

Sayler A. 1966. The reproductive ecology of the Red-backed Salamander, Plethodon cinereus, in Maryland. Copeia 1966: 183–193.

Schluter D., McPhail J. D. 1992. Ecological character displacement and speciation in sticklebacks. American Naturalist 140: 85–108.

Schoener T. W. 1970. Size patterns in West Indian Anolis lizards. II. Correlations with the size of particular sympatric species-displacement and convergence. American Naturalist 104: 155–174.

135

Schulte-Hostedde A. I., Zinner B., Miller J. S., Hickling G. J. 2005 Restitution of mass- size residuals: validating body condition indices. Ecology 86: 155–163.

Scott J. A., Epstein M. E. 1987. Factors affecting phenology in a temperature insect community. American Midland Naturalist 117: 103–118.

Scrucca L. 2013. GA: a package for genetic algorithms in R. Journal of Statistical Software 53: 1–37.

Sebastiano S., Antonio R., Fabrizio O., Daria O., Roberta M. 2012. Different season, different strategies: feeding ecology of two syntopic forest-dwelling salamanders. Acta Oecologica 43: 42–50.

Servedio M. R. 2016. Geography, assortative mating, and the effects of sexual selection on speciation with gene flow. Evolutionary Applications 9: 91–102.

Servedio M. R., Boughman J. W. 2017. The role of sexual selection in local adaptation and speciation. Annual Review of Ecology, Evolution, and Systematics 48: 85–109.

Sever D. M. 1997. Sperm storage in the spermatheca of the Red-backed Salamander, Plethodon cinereus (Amphibia: Plethodontidae). Journal of Morphology 234: 131–146.

Sexton J. P., Hangartner S. B., Hoffmann A. A. 2014. Genetic isolation by environment or distance: which pattern of gene flow is most common? Evolution 68: 1–15.

Shafer A. B. A., Wolf, J. B. W. 2013. Widespread evidence for incipient ecological speciation: a meta-analysis of isolation-by-ecology. Ecology Letters 16: 940–950.

Sherratt T. N. 2011. The optimal sampling strategy for unfamiliar prey. Evolution 65: 2014–2025.

Siepielski A. M., DiBattista J. D., Carlson S. M. 2009. It’s about time: the temporal dynamics of phenotypic selection in the wild. Ecology Letters 12: 1261–1276.

Siepielski A. M., Gotanda K., Morrissey M., Diamond S., DiBattista J., Carlson S. M. 2013. Spatial patterns of directional phenotypic selection. Ecology Letters 16: 1382– 1392.

Sillman A. J., Govardovskii V. I., Röhlich P., Southard J. A., Loew E. R. 1997. The photoreceptors and visual pigments of the garter snake (Thamnophis sirtalis): a microspectrophotometric, scanning electron microscopic and immunocytochemical study. Journal of Comparative Physiology A 181: 89–101.

136

Sinervo B., Lively C. M. 1996. The rock-paper-scissors game and the evolution of alternative male strategies. Nature 380: 240–243.

Sinervo B., Svensson E. 2002. Correlational selection and the evolution of genomic architecture. Heredity 89: 329–338.

Slatkin M. 1987. Gene flow and the geographic structure of natural populations. Science 236: 787–792.

Spear S. F., Peterson C. R., Matocq M. D., Storfer A. 2005. Landscape genetics of the blotched tiger salamander (Ambystoma tigrinum melanostictum). Molecular Ecology 14: 2553–2564.

Spotila J. R. 1972. Role of temperature and water in the ecology of lungless salamanders. Ecological Monographs 42: 95–125.

Stevens M., Stoddard M. C., Higham J. P. 2009. Studying primate color: towards visual system-dependent methods. International Journal of Primatology 30: 893–917.

Storfer A., Murphy M. A., Evans J. S., Goldberg C. S., Robinson S., Spear S. F., Dezzani R., Delmelle E., Vierling L., Waits L. P. 2007. Putting the ‘landscape’ in landscape genetics. Heredity 98: 128–142.

Storfer A., Murphy M. A., Spear S. F., Holderegger R., Waits L. P. 2010. Landscape genetics: where are we now? Molecular Ecology 19: 3496–3514.

Stuczka A., Hickerson C. M., Anthony C. D. 2016. Niche partitioning along the diet axis in a colour polymorphic population of Eastern Red-backed Salamanders, Plethodon cinereus. Amphibia- Reptilia 37: 283–290.

Svensson E., Sinervo B., Comendant T. 2001. Density-Dependent Competition and Selection on Immune Function in Genetic Lizard Morphs. Proceedings of the National Academy of Sciences of the United States of America 98: 12561–12565.

Takahashi, M. K., Pauley, T. K. 2010. Resource allocation and life history traits of Plethodon cinereus at different elevations. American Midland Naturalist 163: 87–94.

Tauber M. J., Tauber C. A., Masaki S. 1986. Seasonal adaptations of insects. Oxford: Oxford University Press.

Thompson J. N. 2005. Coevolution: the geographic mosaic of coevolutionary arms races. Current Biology 15: 992–994.

137

Thompson J. N. 2005. The geographic mosaic of coevolution. Chicago, IL: University of Chicago Press.

Thompson J. N., Cunningham B. M. 2002. Geographic structure and dynamics of coevolutionary selection. Nature 417: 735–738.

Toews D. P., Brelsford A. 2012. The biogeography of mitochondrial and nuclear discordance in animals. Molecular Ecology 21: 3907–3930. van Etten J. 2017. R package gdistance: distances and routes on geographical grids. Journal of Statistical Software 76: 1–21. van Oosterhout C., Hutchinson W. F., Wills D. P. M., Shipley P. 2004. micro-checker: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4: 535–538. van Strien M. J., Keller D., Holderegger R. 2012. A new analytical approach to landscape genetic modelling: least-cost transect analysis and linear mixed models. Molecular Ecology 21: 4010–4023.

Venesky M.D., Anthony C. D. 2007. Antipredator adaptations and predator avoidance by two color morphs of the eastern red-backed salamander, Plethodon cinereus. Herpetologica 63: 450–458.

Venesky M. D., Hess A., DeMarchi J. A., Weil A., Murone J., Hickerson C. A., Anthony C. D. 2015. Morph-specific differences in disease prevalence and pathogen-induced mortality in a terrestrial polymorphic salamander. Journal of Zoology 295: 279–285.

Verrell P. A. 1995. Males choose larger females as mates in the salamander Desmognathus santeelah. Ethology 99: 162–171.

Vorobyev M., Osorio D., Bennett A. T., Marshall N. J., Cuthill I. C. 1998. Tetrachromacy, oil droplets and bird plumage colours. Journal of Comparative Physiology A 183: 621–633.

Wake D. B. 1966. Comparative osteology and evolution of the lungless salamanders, family Plethodontidae. Memoirs Southern California Academy of Sciences 4: 1–111.

Wake D. B., Dresner I. G. 1967. Functional morphology and evolution of tail autotomy in salamanders. Journal of Morphology 22: 265–306.

Waldron B. P., Kuchta S. R., Hantak M. M., Hickerson C. M., Anthony C. D. 2019. Genetic analysis of a cryptic contact zone between mitochondrial clades of the Eastern Red-backed Salamander, Plethodon cinereus. Journal of Herpetology 53: 144–153. 138

Walls S. C., Mathis A., Jaeger R. G., Gergits W. F. 1989. Male salamanders with high quality diets have faeces attractive to females. Animal Behavior 38: 546–548.

Wang I. J. 2009. Fine-scale population structure in a desert amphibian: landscape genetics of the black toad (Bufo exsul). Molecular Ecology 18: 3847–3856.

Wang I. J., Bradburd G. S. 2014. Isolation by environment. Molecular Ecology 23: 5649– 5662.

Wang I. J., Summers K. 2010. Genetic structure is correlated with phenotypic divergence rather than geographic isolation in the highly polymorphic strawberry poison-dart frog. Molecular Ecology 19: 447–458.

Wente W. H., Phillips J. B. 2003. Fixed green and brown color morphs and a novel color- changing morph of the Pacific tree frog Hyla regilla. The American Naturalist 162: 461– 473.

West-Eberhard M. J. 1986. Alternative adaptations, speciation and phylogeny (a review). Proceedings of the National Academy of Sciences of the United States of America 83: 1388–1392.

Wheeler C. A., Karraker N. E., Welsh H. H. Jr., Ollivier L. M. 2007. Diet of the Del Norte Salamander (Plethodon elongatus): differences by age, gender, and season. Northwestern Naturalist 88: 85–94.

Wiens J. J., Engstrom T. N., Chippindale P. T. 2006. Rapid diversification, incomplete isolation, and the “speciation clock” in North American salamanders (genus Plethodon): testing the hybrid swarm hypothesis of rapid radiation. Evolution 60: 2585–2603.

139

APPENDIX A: CHAPTER 2 SUPPLEMENTAL MATERIAL

Table 2.S1. Univariate analyses of variance (ANOVAs) examining seasonal and population effects on the coloration of P. cinereus. Significant effects are in bold. F df P-value Dorsal Brightness 6.13 3,112 <0.001 Chroma 8.00 3,112 <0.001 Hue 2.18 3,112 0.095 Ventral Brightness 3.94 3,112 0.010 Chroma 2.10 3,112 0.104 Hue 1.44 3,112 0.236 Side Brightness 8.54 3,112 <0.001 Chroma 6.00 3,112 0.001 Hue 3.53 3,112 0.021

140

Table 2.S2. Tukey’s HSD post hoc test P-values for side P. cinereus (A) avian, (B) snake, and (C) mammalian contrast values. Comparisons below the diagonal of each substrate type are chromatic contrasts, whereas values above the diagonal are achromatic contrasts. Significant comparisons are in bold. Spring Leaves Soil Substrate CV CV Squire SBI CV CV Squire SBI Striped Lead Striped Lead Striped Lead Striped Lead A CV Striped -- 0.005 1.000 0.012 -- 0.970 1.000 0.054 CV Lead 0.515 -- 0.013 1.000 0.478 -- 0.964 0.461 Squire Striped 0.090 0.985 -- 0.029 1.000 0.204 -- 0.049 SBI Lead <0.001 <0.001 <0.001 -- <0.001 0.049 <0.001 -- B CV Striped -- 0.033 1.000 <0.001 -- 1.000 0.887 0.009 CV Lead 0.972 -- 0.018 0.020 0.944 -- 0.920 0.007 Squire Striped 0.016 0.218 -- <0.001 0.964 1.000 -- <0.001 SBI Lead <0.001 0.001 0.617 -- <0.001 <0.001 <0.001 -- C CV Striped -- 0.156 0.990 <0.001 -- 0.851 0.100 0.001 CV Lead 0.192 -- 0.016 <0.001 0.192 -- 0.850 <0.001 Squire Striped 0.003 0.813 -- <0.001 0.999 0.650 -- <0.001 SBI Lead <0.001 <0.001 <0.001 -- <0.001 0.551 0.009 --

141

Table 2.S3. Tukey’s HSD post hoc test P-values for ventral P. cinereus (A) avian, (B) snake, and (C) mammalian contrast values. Comparisons below the diagonal of each substrate type are chromatic contrasts, whereas values above the diagonal are achromatic contrasts. Significant comparisons are in bold. Fall Leaves Spring Leaves CV CV Squire SBI CV CV Squire SBI Striped Lead Striped Lead Striped Lead Striped Lead A CV Striped -- 0.946 0.047 <0.001 -- 0.863 1.000 <0.001 CV Lead 0.051 -- 0.781 <0.001 0.912 -- 0.690 <0.001 Squire Striped 0.984 <0.001 -- <0.001 0.107 0.951 -- <0.001 SBI Lead <0.001 <0.001 <0.001 -- <0.001 <0.001 <0.001 -- B CV Striped -- 0.989 0.061 <0.001 -- 1.000 0.983 <0.001 CV Lead 0.577 -- 0.652 <0.001 0.981 -- 1.000 <0.001 Squire Striped 0.037 <0.001 -- <0.001 0.146 0.891 -- <0.001 SBI Lead <0.001 <0.001 <0.001 -- <0.001 <0.001 0.053 -- C CV Striped -- 0.999 0.060 <0.001 -- 1.000 0.152 <0.001 CV Lead 0.018 -- 0.456 <0.001 0.718 -- 0.181 <0.001 Squire Striped 0.997 <0.001 -- <0.001 0.550 1.000 -- <0.001 SBI Lead <0.001 <0.001 <0.001 -- <0.001 <0.001 <0.001 --

142

Table 2.S3. continued Soil Substrate CV CV Squire SBI Striped Lead Striped Lead A CV Striped -- 0.990 1.000 0.001 CV Lead 0.899 -- <0.001 <0.001 Squire Striped 0.991 0.199 -- <0.001 SBI Lead <0.001 <0.001 <0.001 -- B CV Striped -- 0.973 1.000 <0.001 CV Lead 0.983 -- 0.999 <0.001 Squire Striped 1.000 0.998 -- <0.001 SBI Lead <0.001 <0.001 <0.001 -- C CV Striped -- 0.948 1.000 <0.001 CV Lead 0.718 -- 0.999 <0.001 Squire Striped 0.941 0.033 -- <0.001 SBI Lead <0.001 <0.001 <0.001 --

143

Fig. 2.S1. Dorsal achromatic contrast values for the avian (A), snake (B), and mammalian (C) visual systems. Bars show means (± SE). Contrast values that lie below the gray horizontal line represent groups that are indistinguishable to the predator. Y-axes values vary by plot.

144

Fig. 2.S2. Side chromatic (A) and achromatic (B) contrast values for the avian visual system. Side chromatic (C) and achromatic (D) contrast values for the snake visual system. Side chromatic (E) and achromatic (F) contrast values for the mammalian visual system. Bars show means (± SE). Contrast values that lie below the gray horizontal line represent groups that are indistinguishable to the predator. Y-axes values vary by plot.

145

Fig. 2.S3. Ventral chromatic (A) and achromatic (B) contrast values for the avian visual system. Ventral chromatic (C) and achromatic (D) contrast values for the snake visual system. Ventral chromatic (E) and achromatic (F) contrast values for the mammalian visual system. Bars show means (± SE). Contrast values that lie below the gray horizontal line represent groups that are indistinguishable to the predator. Y-axes values vary by plot.

146

APPENDIX B: CHAPTER 3 SUPPLEMENTAL MATERIAL

Table 2.S1. Five uncorrelated (R  0.7) climatic variables. Code Bioclimatic variable BIO1 Annual Mean Temperature BIO2 Mean Diurnal Range (Mean of monthly (max temp - min temp)) BIO8 Mean Temperature of Wettest Quarter BIO12 Annual Precipitation BIO18 Precipitation of Warmest Quarter

147

Table 3.S2. Locality information (site numbers match Fig. 1), sample size (N), and population genetic summary statistics: expected heterozygosity (HE), observed heterozygosity (HO), rarefied allelic richness (AR), and rarefied private allelic richness (pAR). Site Latitude Longitude N HE HO AR pAR 1 41.58404 -82.84023 19 0.33 0.26 2.29 0.18 2 41.54521 -82.8148 31 0.51 0.43 2.93 0.03 3 41.389167 -82.82485 33 0.39 0.38 2.31 0.00 4 41.41289 -82.60341 15 0.39 0.39 2.46 0.01 5 41.28728 -82.64306 17 0.41 0.40 2.71 0.05 6 41.322641 -82.498534 30 0.31 0.27 2.31 0.01 7 41.34070 -82.48906 31 0.32 0.31 2.43 0.08 8 41.33131 -82.35005 17 0.35 0.38 2.53 0.13 9 41.38110 -82.32103 27 0.27 0.26 2.18 0.05 10 41.40607 -82.23343 28 0.30 0.29 2.38 0.08 11 41.42552 -82.23048 18 0.32 0.31 2.24 0.03 12 41.45887 -82.10595 29 0.30 0.30 2.27 0.06 13 41.46008 -82.09644 14 0.30 0.30 2.33 0.14 14 41.41602 -82.10184 26 0.26 0.23 2.21 0.05 15 41.37416 -82.10906 14 0.37 0.34 2.72 0.05 16 41.48644 -81.93446 30 0.37 0.39 2.63 0.14 17 41.42073 -81.85917 33 0.41 0.41 2.88 0.05 18 41.39006 -81.69121 30 0.36 0.34 2.71 0.06 19 41.22628 -81.71442 27 0.36 0.34 2.67 0.06 20 41.31902 -81.61631 25 0.36 0.33 2.74 0.06 21 41.229616 -81.518825 34 0.55 0.58 3.46 0.07 22 41.37557 -81.57355 17 0.56 0.51 3.17 0.06 23 41.49361 -81.593533 15 0.54 0.57 2.91 0.08 24 41.42324 -81.4207 24 0.52 0.50 3.00 0.09 25 41.496453 -81.417458 12 0.49 0.53 2.88 0.02 26 41.456753 -81.324957 12 0.55 0.46 3.28 0.17 27 41.597586 -81.356861 23 0.53 0.52 2.98 0.06 28 41.61772 -81.20503 17 0.53 0.49 3.02 0.04

148

Table 3.S3. Pairwise estimates of FST (below diagonal) and associated p-values based on 999 permutations (above diagonal) based on 10 microsatellite loci across 28 populations. Population numbers correspond with Fig. 1.

Pop 1 2 3 4 5 6 7 8 1 -- 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2 0.28 -- 0.00 0.00 0.00 0.00 0.00 0.00 3 0.50 0.23 -- 0.00 0.00 0.00 0.00 0.00 4 0.53 0.25 0.16 -- 0.00 0.00 0.00 0.00 5 0.50 0.22 0.06 0.08 -- 0.00 0.00 0.01 6 0.58 0.33 0.10 0.17 0.04 -- 0.00 0.04 7 0.58 0.35 0.09 0.14 0.06 0.02 -- 0.00 8 0.56 0.29 0.10 0.14 0.03 0.02 0.03 -- 9 0.61 0.35 0.09 0.21 0.06 0.05 0.06 0.04 10 0.60 0.35 0.10 0.17 0.08 0.03 0.01 0.05 11 0.57 0.31 0.06 0.13 0.03 0.01 0.02 0.03 12 0.60 0.34 0.07 0.17 0.05 0.05 0.04 0.04 13 0.61 0.33 0.11 0.16 0.08 0.11 0.07 0.06 14 0.62 0.36 0.13 0.19 0.07 0.04 0.04 0.03 15 0.53 0.27 0.06 0.14 0.02 0.01 0.04 0.01 16 0.54 0.30 0.08 0.16 0.04 0.05 0.05 0.02 17 0.49 0.26 0.08 0.16 0.04 0.06 0.07 0.02 18 0.54 0.28 0.08 0.16 0.03 0.03 0.06 0.01 19 0.56 0.31 0.08 0.14 0.04 0.01 0.02 0.02 20 0.55 0.28 0.10 0.13 0.03 0.04 0.06 0.03 21 0.37 0.17 0.10 0.14 0.07 0.12 0.13 0.09 22 0.38 0.26 0.33 0.31 0.30 0.41 0.40 0.35 23 0.41 0.27 0.32 0.30 0.29 0.40 0.38 0.34 24 0.41 0.30 0.36 0.34 0.33 0.43 0.42 0.38 25 0.41 0.29 0.40 0.36 0.36 0.48 0.46 0.42 26 0.38 0.26 0.35 0.33 0.30 0.43 0.42 0.37 27 0.40 0.28 0.32 0.30 0.28 0.39 0.37 0.33 28 0.42 0.29 0.35 0.33 0.30 0.42 0.41 0.36

149

Table 3.S3. Continued Pop 9 10 11 12 13 14 15 16 1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5 0.00 0.00 0.01 0.00 0.00 0.00 0.09 0.00 6 0.00 0.01 0.13 0.00 0.00 0.00 0.21 0.00 7 0.00 0.08 0.02 0.00 0.00 0.00 0.00 0.00 8 0.00 0.00 0.01 0.00 0.00 0.01 0.24 0.01 9 -- 0.00 0.00 0.00 0.00 0.00 0.02 0.00 10 0.09 -- 0.00 0.00 0.00 0.00 0.00 0.00 11 0.06 0.03 -- 0.00 0.00 0.00 0.17 0.00 12 0.05 0.06 0.05 -- 0.01 0.00 0.00 0.00 13 0.09 0.09 0.09 0.03 -- 0.00 0.00 0.00 14 0.03 0.08 0.06 0.04 0.06 -- 0.00 0.00 15 0.03 0.04 0.01 0.04 0.07 0.04 -- 0.03 16 0.06 0.07 0.04 0.03 0.03 0.05 0.02 -- 17 0.07 0.07 0.06 0.06 0.06 0.05 0.01 0.02 18 0.03 0.08 0.03 0.05 0.09 0.04 0.00 0.04 19 0.04 0.01 0.02 0.04 0.05 0.05 0.00 0.03 20 0.04 0.09 0.04 0.07 0.08 0.04 0.01 0.06 21 0.13 0.14 0.11 0.13 0.13 0.13 0.06 0.08 22 0.41 0.42 0.37 0.40 0.36 0.41 0.32 0.33 23 0.40 0.41 0.36 0.38 0.34 0.40 0.32 0.32 24 0.43 0.44 0.40 0.42 0.39 0.43 0.35 0.36 25 0.49 0.49 0.45 0.47 0.44 0.49 0.39 0.40 26 0.43 0.44 0.39 0.42 0.39 0.44 0.33 0.35 27 0.39 0.40 0.36 0.37 0.34 0.38 0.30 0.31 28 0.42 0.43 0.39 0.41 0.38 0.42 0.33 0.34

150

Table 3.S3. Continued Pop 17 18 19 20 21 22 23 24 1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 6 0.00 0.00 0.09 0.00 0.00 0.00 0.00 0.00 7 0.00 0.00 0.03 0.00 0.00 0.00 0.00 0.00 8 0.01 0.04 0.06 0.00 0.00 0.00 0.00 0.00 9 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 10 0.00 0.00 0.05 0.00 0.00 0.00 0.00 0.00 11 0.00 0.01 0.02 0.00 0.00 0.00 0.00 0.00 12 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 13 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 14 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 15 0.20 0.61 0.30 0.21 0.00 0.00 0.00 0.00 16 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 17 -- 0.00 0.00 0.00 0.00 0.00 0.00 0.00 18 0.02 -- 0.00 0.11 0.00 0.00 0.00 0.00 19 0.04 0.03 -- 0.00 0.00 0.00 0.00 0.00 20 0.04 0.01 0.03 -- 0.00 0.00 0.00 0.00 21 0.04 0.09 0.11 0.09 -- 0.00 0.00 0.00 22 0.29 0.37 0.37 0.35 0.14 -- 0.04 0.44 23 0.29 0.36 0.36 0.34 0.14 0.02 -- 0.05 24 0.32 0.39 0.40 0.37 0.16 0.00 0.02 -- 25 0.35 0.43 0.44 0.42 0.18 0.01 0.02 0.00 26 0.30 0.38 0.39 0.37 0.14 0.00 0.04 0.01 27 0.27 0.35 0.35 0.33 0.13 0.01 0.03 0.01 28 0.29 0.37 0.38 0.36 0.14 0.01 0.03 0.01

151

Table 3.S3. Continued Pop 25 26 27 28 1 0.00 0.00 0.00 0.00 2 0.00 0.00 0.00 0.00 3 0.00 0.00 0.00 0.00 4 0.00 0.00 0.00 0.00 5 0.00 0.00 0.00 0.00 6 0.00 0.00 0.00 0.00 7 0.00 0.00 0.00 0.00 8 0.00 0.00 0.00 0.00 9 0.00 0.00 0.00 0.00 10 0.00 0.00 0.00 0.00 11 0.00 0.00 0.00 0.00 12 0.00 0.00 0.00 0.00 13 0.00 0.00 0.00 0.00 14 0.00 0.00 0.00 0.00 15 0.00 0.00 0.00 0.00 16 0.00 0.00 0.00 0.00 17 0.00 0.00 0.00 0.00 18 0.00 0.00 0.00 0.00 19 0.00 0.00 0.00 0.00 20 0.00 0.00 0.00 0.00 21 0.00 0.00 0.00 0.00 22 0.29 0.52 0.14 0.22 23 0.06 0.01 0.01 0.01 24 0.43 0.81 0.05 0.85 25 -- 0.39 0.02 0.03 26 0.00 -- 0.09 0.57 27 0.03 0.02 -- 0.06 28 0.03 0.00 0.01 --

152

Table 3.S4. Pairwise estimates of standardized FST (F’ST below diagonal) and Jost’s D (Dest; above diagonal) based on 10 microsatellite loci across 28 populations. Population numbers correspond with Fig. 1. Pop 1 2 3 4 5 6 7 8 1 -- 0.33 0.61 0.70 0.64 0.66 0.69 0.69 2 0.41 -- 0.25 0.30 0.27 0.36 0.39 0.35 3 0.75 0.41 -- 0.12 0.04 0.06 0.06 0.07 4 0.77 0.45 0.24 -- 0.06 0.11 0.09 0.10 5 0.73 0.42 0.09 0.13 -- 0.02 0.04 0.02 6 0.80 0.55 0.14 0.23 0.05 -- 0.01 0.01 7 0.82 0.59 0.14 0.21 0.09 0.03 -- 0.02 8 0.79 0.52 0.14 0.21 0.04 0.01 0.04 -- 9 0.83 0.58 0.13 0.30 0.10 0.06 0.08 0.05 10 0.82 0.59 0.14 0.23 0.11 0.03 0.01 0.05 11 0.76 0.53 0.08 0.17 0.03 0.01 0.02 0.02 12 0.84 0.57 0.11 0.25 0.08 0.06 0.05 0.05 13 0.83 0.58 0.16 0.24 0.13 0.14 0.10 0.08 14 0.83 0.59 0.18 0.26 0.10 0.05 0.06 0.03 15 0.74 0.50 0.09 0.22 0.03 0.00 0.06 0.00 16 0.80 0.54 0.13 0.24 0.06 0.07 0.07 0.03 17 0.77 0.48 0.13 0.27 0.07 0.08 0.10 0.03 18 0.78 0.49 0.13 0.25 0.05 0.03 0.09 0.02 19 0.82 0.55 0.13 0.23 0.07 0.01 0.02 0.02 20 0.80 0.50 0.15 0.18 0.04 0.05 0.09 0.02 21 0.66 0.37 0.19 0.27 0.13 0.20 0.23 0.16 22 0.61 0.56 0.62 0.61 0.60 0.70 0.69 0.66 23 0.63 0.56 0.57 0.55 0.57 0.66 0.64 0.61 24 0.60 0.60 0.63 0.62 0.60 0.70 0.70 0.66 25 0.52 0.57 0.68 0.64 0.66 0.76 0.75 0.72 26 0.54 0.56 0.63 0.63 0.60 0.71 0.72 0.67 27 0.65 0.59 0.58 0.57 0.54 0.65 0.64 0.61 28 0.61 0.60 0.63 0.63 0.58 0.69 0.69 0.64

153

Table 3.S4. Continued. Pop 9 10 11 12 13 14 15 16 1 0.70 0.69 0.66 0.71 0.75 0.70 0.66 0.68 2 0.37 0.40 0.35 0.37 0.40 0.38 0.33 0.36 3 0.05 0.06 0.04 0.04 0.07 0.07 0.04 0.06 4 0.13 0.11 0.09 0.11 0.11 0.11 0.11 0.12 5 0.04 0.05 0.02 0.03 0.06 0.04 0.01 0.03 6 0.02 0.01 0.01 0.03 0.06 0.02 0.00 0.03 7 0.03 0.00 0.01 0.02 0.04 0.02 0.02 0.03 8 0.02 0.02 0.02 0.02 0.03 0.01 0.00 0.01 9 -- 0.04 0.03 0.02 0.04 0.01 0.01 0.03 10 0.11 -- 0.02 0.03 0.04 0.03 0.02 0.04 11 0.06 0.03 -- 0.02 0.05 0.03 0.01 0.03 12 0.06 0.07 0.05 -- 0.01 0.02 0.02 0.01 13 0.12 0.12 0.10 0.03 -- 0.02 0.04 0.02 14 0.05 0.10 0.06 0.05 0.07 -- 0.02 0.02 15 0.04 0.05 0.00 0.06 0.11 0.06 -- 0.01 16 0.08 0.09 0.05 0.03 0.05 0.06 0.03 -- 17 0.10 0.11 0.08 0.09 0.10 0.08 0.01 0.03 18 0.05 0.11 0.03 0.08 0.13 0.05 0.01 0.06 19 0.06 0.01 0.02 0.05 0.08 0.07 0.01 0.05 20 0.05 0.12 0.04 0.08 0.10 0.05 0.00 0.07 21 0.22 0.24 0.19 0.22 0.25 0.21 0.12 0.15 22 0.70 0.72 0.67 0.68 0.67 0.69 0.63 0.61 23 0.65 0.67 0.61 0.62 0.60 0.64 0.59 0.57 24 0.69 0.72 0.67 0.68 0.66 0.68 0.63 0.62 25 0.75 0.77 0.73 0.74 0.72 0.74 0.69 0.67 26 0.71 0.74 0.68 0.71 0.70 0.71 0.64 0.63 27 0.64 0.67 0.62 0.62 0.60 0.63 0.57 0.55 28 0.69 0.71 0.66 0.67 0.66 0.68 0.61 0.60

154

Table 3.S4. Continued. Pop 17 18 19 20 21 22 23 24 1 0.63 0.64 0.69 0.69 0.54 0.55 0.58 0.59 2 0.31 0.31 0.37 0.33 0.25 0.42 0.42 0.47 3 0.06 0.05 0.06 0.07 0.11 0.44 0.40 0.47 4 0.14 0.12 0.11 0.09 0.16 0.44 0.39 0.47 5 0.03 0.02 0.03 0.02 0.07 0.43 0.40 0.45 6 0.03 0.01 0.01 0.02 0.11 0.51 0.47 0.54 7 0.04 0.03 0.01 0.04 0.12 0.49 0.43 0.52 8 0.02 0.01 0.01 0.02 0.09 0.49 0.44 0.51 9 0.04 0.02 0.02 0.02 0.11 0.48 0.43 0.50 10 0.05 0.04 0.01 0.05 0.13 0.53 0.47 0.55 11 0.04 0.01 0.01 0.02 0.11 0.50 0.45 0.53 12 0.04 0.03 0.02 0.03 0.12 0.48 0.42 0.50 13 0.04 0.05 0.03 0.05 0.14 0.48 0.41 0.50 14 0.03 0.02 0.02 0.02 0.11 0.48 0.42 0.49 15 0.00 0.00 0.00 0.00 0.07 0.45 0.42 0.48 16 0.02 0.03 0.02 0.03 0.08 0.43 0.39 0.45 17 -- 0.02 0.03 0.03 0.04 0.39 0.36 0.41 18 0.04 -- 0.02 0.00 0.08 0.48 0.44 0.50 19 0.07 0.05 -- 0.02 0.11 0.50 0.45 0.53 20 0.06 0.01 0.05 -- 0.09 0.46 0.42 0.49 21 0.07 0.16 0.20 0.16 -- 0.21 0.22 0.23 22 0.57 0.67 0.69 0.64 0.31 -- 0.03 0.00 23 0.54 0.63 0.64 0.59 0.32 0.04 -- 0.02 24 0.57 0.68 0.70 0.64 0.33 0.02 0.01 -- 25 0.63 0.73 0.75 0.70 0.38 0.01 0.01 0.04 26 0.58 0.68 0.71 0.65 0.32 0.01 0.09 0.05 27 0.51 0.62 0.63 0.58 0.29 0.02 0.05 0.01 28 0.54 0.66 0.68 0.63 0.30 0.01 0.06 0.04

155

Table 3.S4. Continued. Pop 25 26 27 28 1 0.52 0.53 0.56 0.60 2 0.44 0.42 0.44 0.46 3 0.51 0.45 0.41 0.45 4 0.49 0.47 0.41 0.46 5 0.50 0.43 0.37 0.41 6 0.58 0.52 0.46 0.50 7 0.56 0.52 0.44 0.49 8 0.56 0.49 0.43 0.47 9 0.55 0.49 0.43 0.47 10 0.60 0.54 0.48 0.51 11 0.57 0.51 0.45 0.49 12 0.54 0.50 0.42 0.47 13 0.54 0.51 0.41 0.47 14 0.54 0.49 0.42 0.47 15 0.53 0.46 0.40 0.44 16 0.49 0.45 0.37 0.42 17 0.45 0.39 0.34 0.37 18 0.55 0.49 0.43 0.46 19 0.58 0.52 0.44 0.49 20 0.54 0.48 0.41 0.46 21 0.27 0.21 0.19 0.20 22 0.01 0.00 0.01 0.01 23 0.03 0.06 0.03 0.04 24 0.00 0.01 0.02 0.01 25 -- 0.00 0.04 0.03 26 0.03 -- 0.02 0.01 27 0.04 0.03 -- 0.02 28 0.03 0.02 0.02 --

156

Table 3.S5. Mean, consistent, rates of gene flow determined by MIGRATE-N between pairs of populations. The left population is the source population, and the right population is the recipient population. Gene flow (population) direction Mean rate 95% CI 15 → 10/11 0.022 0.007, 0.037 3/5 → 10/11 0.016 -0.007, 0.039 12/13 → 10/11 0.011 0.002, 0.020 8 → 10/11 0.007 -0.008, 0.023 4 → 10/11 0.019 -0.005, 0.043 19/20 → 14 0.050 0.010, 0.091 16 → 14 0.016 -0.003, 0.036 9 → 14 0.053 0.039, 0.067 8 → 14 0.012 -0.006, 0.031 14 → 19/20 0.080 0.032, 0.128 16 → 19/20 0.097 0.078, 0.115 17/18 → 19/20 0.115 0.085, 0.145 19/20 → 15 0.024 0.016, 0.032 9 → 15 0.015 -0.003, 0.034 10/11 → 3/5 0.032 0.008, 0.057 19/20 → 3/5 0.034 0.022, 0.045 6/7 → 3/5 0.042 0.021, 0.063 12/13 → 3/5 0.037 -0.004, 0.078 16 → 3/5 0.023 -0.002, 0.049 9 → 3/5 0.031 -0.020, 0.082 10/11 → 6/7 0.037 -0.007, 0.080 14 → 6/7 0.025 -0.018, 0.067 19/20 → 6/7 0.022 0.008, 0.036 15 → 6/7 0.024 -0.010, 0.057 3/5 → 6/7 0.026 -0.005, 0.058 12/13 → 6/7 0.027 -0.001, 0.055 16 → 6/7 0.026 0.001, 0.052 8 → 6/7 0.022 -0.031, 0.074 4 → 6/7 0.015 0.003, 0.027 8 → 12/13 0.047 0.022, 0.071 4 → 12/13 0.046 -0.002, 0.093 3/5 → 16 0.050 0.022, 0.077 6/7 → 16 0.063 0.034, 0.091 9 → 16 0.035 0.020, 0.049 17/18 → 16 0.039 0.016, 0.061 8 → 16 0.028 -0.006, 0.061 4 → 16 0.059 0.014, 0.104 15 → 9 0.023 0.000, 0.046 157

Table 3.S5. Continued 3/5 → 9 0.036 0.010, 0.062 6/7 → 9 0.034 -0.014, 0.082 12/13 → 9 0.019 -0.005, 0.043 8 → 9 0.026 -0.018, 0.069 4 → 9 0.020 -0.010, 0.051 14 → 17/18 0.020 0.010, 0.029 19/20 → 17/18 0.021 -0.005, 0.048 3/5 → 17/18 0.036 0.007, 0.065 6/7 → 17/18 0.027 -0.009, 0.062 12/13 → 17/18 0.033 0.004, 0.063 16 → 17/18 0.012 -0.011, 0.035 8 → 17/18 0.034 -0.001, 0.070 4 → 17/18 0.034 -0.004, 0.072 3/5 → 8 0.024 -0.009, 0.057 16 → 8 0.029 0.009, 0.048 9 → 8 0.047 0.023, 0.070 4 → 8 0.029 -0.005, 0.063 10/11 → 4 0.036 0.004, 0.069 19/20 → 4 0.023 -0.002, 0.049 3/5 → 4 0.007 -0.004, 0.018 6/7 → 4 0.020 -0.023, 0.063 9 → 4 0.022 -0.023, 0.067 17/18 → 4 0.029 -0.004, 0.063

158

Fig. 3.S1. Evanno method K plot from initial STRUCTURE run with all 28 populations.

159

Fig. 3.S2. STRUCTURE run on populations in the Central Cluster after the removal of site 21 (see text).

160

Fig. 3.S3. STRUCTURE run on populations in the Central Cluster with the addition of site 2 (see text).

161

Fig. 3.S4. DAPC for 15 retained principle component axes and 4 discriminant functions. We interpret the data as having 5 clusters. Cluster 1 = sites 1-2; Clusters 2, 3, 5 = sites 3- 21, and 1 individual from site 2; Cluster 4 = sites 22-28, and 13 individuals from site 21, and one individual from sites 2 and 17. Colors correspond with STRUCTURE plots.

162

Fig. 3.S5. Gene flow rates within the Central Cluster (except population 21) in northern Ohio. Population numbers are in the center of each pie and correspond with Fig. 1. Circles indicate combined populations in the MIGRATE-N analysis.

163

Fig. 3.S6. A) Across all populations the admixture F-model (AFM) identified 28 linages (L1-L28) from the putative ancestral population contributing to the 28 sampled sites (bar plot). Kappa estimates indicate a high degree of admixture within distinct genetic clusters (bar plot). Alpha parameter estimates varied largely among lineages (0.51-22.14), with sharp differences among clusters – low alpha estimates in the Eastern Cluster (0.51-3.42) and Western Cluster (0.57-1.92) and high estimates in the Central Cluster (4.99-22.14; alpha values in parentheses next to each lineage name). B) Theta parameter estimates indicate high population-level similarity within the Western Cluster, whereas all sites in the Eastern Cluster are highly similar, and the Central Cluster sites are also homogenous.

164

Fig. 3.S7. A) In the Western Cluster the AFM identified 2 lineages (L1-L2) from the reputed ancestral population contributing to the 2 sites in this cluster, and kappa estimates indicate a high degree differentiation between sites (bar plot). Alpha parameter estimates are relatively low in this cluster (2.18-3.13). B) Theta estimates show a high degree of differentiation among sites, while individuals within sites are highly similar.

165

Fig. 3.S8. A) In the Central Cluster the AFM identified 19 lineages (L3-L21) from the hypothetical ancestral population contributing to the 19 sites in this cluster, and the kappa estimates are relatively homogenous (bar plot). Alpha estimates are variable, but relatively high within lineages (3.66-47.61); however, population 21 has a low alpha estimate (0.97), suggesting that this population is highly admixed between the Central and Eastern Clusters (also see Fig. 1, S6). B) Theta parameter estimates indicate coancestry coefficients across sites are homogenous.

166

Fig. 3.S9. A) AFM results from the Eastern Cluster identified 7 lineages (L22-L28) from the putative ancestral population contributing to the 7 sites in this cluster, and kappa values indicate a high degree of admixture (bar plot). Alpha parameter estimates are all relatively high within this cluster (7.92-17.91). B) Theta estimates are not indicative of high variation among sites.

167

APPENDIX C: CHAPTER 4 SUPPLEMENTAL MATERIAL

Table 4.S1. Dates of stomach contents and leaf litter collections, temperature during sampling, percent humidity, and snout-vent length (SVL) range of salamanders per site. Site Date Temperature (C) Humidity (%) SVL range (mm) SVF 5/8/14 22.2 52 32.21 - 40.78 CV 5/12/14 20.6 82 32.79 - 41.93 CF 5/13/14 24.4 65 32.83 - 45.98 EW 5/27/14 26.1 64 33.22 - 45.62 SBI 5/28/14 18.9 84 32.33 - 46.48 EH 6/3/14 27.7 52 32.29 - 39.33 SVF 10/3/14 17.8 89 32.47 - 39.96 CV 10/4/14 9.4 74 32.04 - 43.47 CF 10/4/14 7.8 69 32.05 - 42.74 EW 10/3/14 18.3 74 32.66 - 45.88 SBI 10/5/14 10.0 62 34.60 - 47.26 EH 10/5/14 10.0 62 34.21 - 45.78

168

Table 4.S2. Pairwise comparisons of adjusted p-values for population differentiation in dietary composition in the spring (below the diagonal) and fall (above the diagonal). Chapin Forest (CF), Squire Valleevue Farm (SVF), Manatoc Scout Reservation near Cuyahoga Valley (CV), Edison Woods (EW), East Harbor State Park (EH), and South Bass Island (SBI) CF SVF CV EW EH SBI CF – 0.225 0.015 0.015 0.030 0.015 SVF 0.210 – 1.000 0.765 1.000 0.015 CV 0.030 0.090 – 0.015 0.120 0.015 EW 0.030 0.030 0.120 – 1.000 0.015 EH 0.015 0.015 0.030 1.000 – 1.000 SBI 0.015 0.015 0.015 1.000 1.000 –

169

Table 4.S3. Results of PERMDISP to test for differences in salamander versus leaf litter invertebrate breadth. Significant results (P < 0.05) are in bold. Site Season df F P SVF Spring 1,16 0.1 0.872 CV Spring 2,30 1.1 0.340 CF Spring 1,16 2.8 0.114 EW Spring 2,30 5.6 0.009 SBI Spring 1,16 5.0 0.040 EH Spring 1,16 4.1 0.060 SVF Fall 1,16 1.5 0.235 CV Fall 2,30 2.2 0.128 CF Fall 1,16 1.3 0.267 EW Fall 2,30 1.2 0.327 SBI Fall 1,16 2.1 0.169 EH Fall 1,16 1.4 0.262

170

Fig. 4.S1. Species accumulation curves for dietary prey for each morph and population over the (A) spring and (B) fall field seasons. Striped morphs include Squire Valleevue Farm (SVF), Chapin Forest (CF), Manatoc Scout Reservation near Cuyahoga Valley (CV; S), and Edison Woods (EW; S). Unstriped morphs include Manatoc Scout Reservation near Cuyahoga Valley (CV; U), Edison Woods (EW; U), East Harbor State Park (EH), and South Bass Island (SBI).

171

Fig. 4.S2. Mean dietary niche breadth for each morph and population for the spring season, measured by the Shannon diversity index. Red represents striped morphs (S), black represents unstriped morphs (U). Error bars denote standard error of the mean.

172

APPENDIX D: CHAPTER 5 SUPPLEMENTAL MATERIAL

Table 5.S1. Pairwise post hoc test P-values for polymorphic site differences in color morph mating ratios. Cuyahoga Valley National Park (CVNP), Edison Woods Reservation (EW), and East Harbor State Park (EH). Significant comparisons are in bold. CVNP EW EW 0.002 - EH <0.001 0.001

173

Table 5.S2. Model selection for female and male color morph preference at each of the three polymorphic populations: (A & B) Cuyahoga Valley National Park (CVNP), (C & D) Edison Woods Reservation (EW), and (E & F) East Harbor State Park (EH). Models are ranked from best to least-supported within each category according to AICc score. (A) CVNP: female color morph vs. male predictors df LL AICc ΔAICc AICcWt Color morph * body size 4 -23.455 55.8 0 0.256 Tail presence/absence 2 -25.965 56.2 0.39 0.211 Body size 2 -26.181 56.6 0.82 0.17 Color morph 2 -26.345 56.9 1.15 0.144 Body size + tail presence/absence 3 -25.85 58.2 2.42 0.076 Color morph * body size + tail presence/absence 5 -23.454 58.3 2.47 0.074 Color morph + tail presence/absence 3 -25.964 58.5 2.65 0.068 (B) CVNP: male color morph vs. female predictors Color morph * body size 4 -17.003 42.9 0 0.721 Color morph * body size + tail presence/absence 5 -16.872 45.1 2.21 0.238 Body size 2 -23.395 51 8.15 0.012 Tail presence/absence 2 -23.53 51.3 8.42 0.011 Color morph 2 -23.57 51.4 8.5 0.01 Body size + tail presence/absence 3 -23.313 53.1 10.25 0.004 Color morph + tail presence/absence 3 -23.53 53.6 10.69 0.003 (C) EW: female color morph vs. male predictors Color morph 2 -21.849 48.1 0 0.309 Body size 2 -22.09 48.6 0.48 0.243 Tail presence/absence 2 -22.097 48.6 0.5 0.241 Color morph + tail presence/absence 3 -21.841 50.5 2.43 0.092 Body size + tail presence/absence 3 -22.053 51 2.85 0.074 Color morph * body size 4 -21.575 52.6 4.52 0.032 Color morph * body size + tail presence/absence 5 -21.572 55.5 7.34 0.008

174

Table 5.S2. Continued (D) EW: male color morph vs. female predictors df LL AICc ΔAICc AICcWt Body size 2 -21.466 47.3 0 0.345 Color morph 2 -21.849 48.1 0.77 0.235 Tail presence/absence 2 -22.097 48.6 1.26 0.184 Body size + tail presence/absence 3 -21.424 49.7 2.36 0.106 Color morph + tail presence/absence 3 -21.789 50.4 3.09 0.074 Color morph * body size 4 -20.973 51.4 4.08 0.045 Color morph * body size + tail presence/absence 5 -20.878 54.1 6.72 0.012 (E) EH: female color morph vs. male predictors Body size 2 -8.329 21.2 0 0.355 Color morph 2 -8.612 21.8 0.57 0.268 Tail presence/absence 2 -9.097 22.7 1.54 0.165 Body size + tail presence/absence 3 -8.31 23.8 2.56 0.099 Color morph + tail presence/absence 3 -8.593 24.3 3.13 0.074 Color morph * body size 4 -8.002 26 4.8 0.032 Color morph * body size + tail presence/absence 5 -7.999 29.2 7.95 0.007 (F) EH: male color morph vs. female predictors Color morph 2 -10.431 25.4 0 0.373 Tail presence/absence 2 -10.99 26.5 1.12 0.213 Body size 2 -10.991 26.5 1.12 0.213 Color morph + tail presence/absence 3 -10.425 28 2.58 0.102 Body size + tail presence/absence 3 -10.99 29.1 3.71 0.058 Color morph * body size 4 -10.42 30.8 5.43 0.025 Color morph * body size + tail presence/absence 5 -9.273 31.7 6.3 0.016

! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

Thesis and Dissertation Services ! !