Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Genetic assessment of taxonomic uncertainty and conservation units in painted turtles, with a focus on… Jensen, Evelyn Lise 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2013_fall_jensen_evelyn.pdf [ 2.91MB ]
Metadata
JSON: 24-1.0073968.json
JSON-LD: 24-1.0073968-ld.json
RDF/XML (Pretty): 24-1.0073968-rdf.xml
RDF/JSON: 24-1.0073968-rdf.json
Turtle: 24-1.0073968-turtle.txt
N-Triples: 24-1.0073968-rdf-ntriples.txt
Original Record: 24-1.0073968-source.json
Full Text
24-1.0073968-fulltext.txt
Citation
24-1.0073968.ris

Full Text

GENETIC ASSESSMENT OF TAXONOMIC UNCERTAINTY AND CONSERVATION UNITS IN PAINTED TURTLES, WITH A FOCUS ON THE AT-RISK CHRYSEMYS PICTA BELLII IN BRITISH COLUMBIA  by  Evelyn Lise Jensen  B.Sc. Honors, University of Alberta, 2011  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in  THE COLLEGE OF GRADUATE STUDIES  (Biology)  THE UNIVERSITY OF BRITISH COLUMBIA  (Okanagan)  June 2013  © Evelyn Lise Jensen, 2013  Abstract As biodiversity continues to be lost at an alarming rate, it has become increasingly important to resolve issues surrounding taxonomic uncertainty and how best to prioritize populations for conservation. Controversy can arise over whether populations merit conservation if their taxonomic status is unclear. Additionally, maintaining intraspecific genetic diversity is of particular importance for preserving evolutionary history and the potential for future adaptation. In order to effectively protect this diversity, species and units below the species level need to be defined. However, delineation of such units is subject to many challenges, with no one strategy applying universally across taxa. Using mitochondrial DNA sequence and microsatellite genotypic data, I examine population structure and demographic history of Chrysemys picta bellii (western painted turtles) in British Columbia, where it is a species-at-risk. I use this system to compare the application of evolutionarily significant unit and management unit criteria with Canadian designatable unit guidelines to determine appropriate conservation units. I find that that BC western painted turtles form a single evolutionarily significant unit, with each occupied site constituting a separate management unit. These findings contrast with the evidence for six discrete designatable units. Patterns of genetic variation in BC western painted turtles indicate that the conservation of each region is important to maintaining regional diversity and evolutionary novelty.  I also address the taxonomic uncertainty in Chrysemys, which is ambiguous due to questions regarding whether the four regional morphological variants warrant formal recognition. Despite using both mitochondrial and nuclear sequence data, I do not find genetic evidence to  ii  conclusively resolve the taxonomic uncertainty in Chrysemys, however, there are indications that the two currently recognized species Chrysemys dorsalis and Chrysemys picta may be warranted.  Overall, this research presents the first detailed population genetic study for this species, which can be used to directly inform conservation prioritization of western painted turtles in BC. Moreover, the range-wide phylogeographic analyses is the most through genetic investigation of Chrysemys taxonomy to date, yet future research involving additional nuclear markers is still warranted.  iii  Preface Several individuals have contributed to Chapters 2 and 3 and Appendix A of this thesis. Manuscript versions of these will be co-authored when submitted for publication. The co-authors and I shared in the identification and design of this research. I have been primarily responsible for collecting the samples and data, data analysis and manuscript preparation.  A version of Chapter 2 has been submitted for publication: Jensen EL, Govindarajulu P, Russello MA (2013). When the shoe doesn't fit: applying conservation unit concepts to western painted turtles at their northern periphery  I collected most of the samples and all of the data used in the manuscript. I conducted all analyses and drafted the manuscript. Michael Russello oversaw data collection and analyses, and helped draft the manuscript. Purnima Govindarajulu guided sample collection and helped draft the manuscript.  Some of the lab work associated with Appendix A was carried out by Jeanette Madsen under my supervision as part of her undergraduate honors project, although I collected additional data and analyzed all data independently of her study.  The data presented in this thesis was collected according to the animal care protocol of the University of British Columbia Research Ethics Board (Animal Care Certificate # A11-0163) and sampling permit # VI11-71744 from the BC Ministry of Environment.  iv  Table of Contents  Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents .......................................................................................................................... v List of Tables .............................................................................................................................. viii List of Figures ............................................................................................................................... ix Acknowledgements ....................................................................................................................... x 1  Chapter: INTRODUCTION .................................................................................................. 1 1.1  Conservation of biodiversity ....................................................................................................... 1  1.2  Species concepts .......................................................................................................................... 1  1.3  Intra-specific conservation units .................................................................................................. 4  1.3.1  The Evolutionarily Significant Unit ........................................................................................ 4  1.3.2  Policy based conservation units .............................................................................................. 7  1.4  Other issues in conservation ........................................................................................................ 8  1.5  The western painted turtle ......................................................................................................... 10  1.5.1  Chrysemys taxonomy ............................................................................................................ 10  1.5.2  Chrysemys picta bellii ........................................................................................................... 11  1.5.3  British Columbian western painted turtles ............................................................................ 11  1.6  2  Thesis objectives ....................................................................................................................... 14  Chapter: APPLYING CONSERVATION UNIT CONCEPTS TO WESTERN  PAINTED TURTLES AT THEIR NORTHERN PERIPHERY ............................................ 15 2.1  Background................................................................................................................................ 15  2.2  Materials and methods ............................................................................................................... 19  2.2.1  Sampling ............................................................................................................................... 19  2.2.2  Data collection ...................................................................................................................... 22  2.2.3  Haplotypic variation and network analysis ........................................................................... 23  2.2.4  Genotypic variation and population differentiation .............................................................. 24  2.2.5  Demographic history ............................................................................................................. 26  2.3  Results ....................................................................................................................................... 28  v  2.3.1  Haplotypic variation .............................................................................................................. 28  2.3.2  Genotypic variation and population differentiation .............................................................. 29  2.3.3  Demographic history ............................................................................................................. 34  2.4  3  Discussion.................................................................................................................................. 37  2.4.1  The extent and distribution of genetic variation .................................................................... 37  2.4.2  Demographic history of BC western painted turtles ............................................................. 39  2.4.3  Conservation unit concepts ................................................................................................... 40  2.4.4  Management implications ..................................................................................................... 43  Chapter: INVESTIGATING TAXONOMIC UNCERTAINTY IN PAINTED  TURTLES .................................................................................................................................... 45 3.1  Background................................................................................................................................ 45  3.2  Methods ..................................................................................................................................... 48  3.2.1  Sampling ............................................................................................................................... 48  3.2.2  Data collection ...................................................................................................................... 49  3.2.3  Population genetic analysis ................................................................................................... 50  3.2.4  Network and phylogenetic analysis....................................................................................... 51  3.3 3.3.1  Within and among subspecies variation ................................................................................ 53  3.3.2  Network and phylogenetic analysis....................................................................................... 55  3.4  4  Results ....................................................................................................................................... 53  Discussion.................................................................................................................................. 60  3.4.1  Uncertain taxonomy .............................................................................................................. 60  3.4.2  Discordance among results.................................................................................................... 63  3.4.3  Implications for Chrysemys paleogeographic history and conservation ............................... 65  3.4.4  Summary ............................................................................................................................... 67  Chapter: CONCLUSIONS ................................................................................................... 68 4.1  Research findings and significance ........................................................................................... 68  4.2  Limitation of this study and future work ................................................................................... 71  References .................................................................................................................................... 74 Appendices ................................................................................................................................... 88 Appendix A : Investigating the genetic consequences of introduced Chrysemys picta on endangered C. p. bellii in British Columbia ................................................................................................................... 88 A.1  Background ........................................................................................................................... 88  vi  A.2  Methods ................................................................................................................................. 91  A.3  Results ................................................................................................................................... 95  A.4  Discussion ........................................................................................................................... 100  Appendix B : Material supplemental to Chapter 2 ............................................................................... 103 B.1  Tables .................................................................................................................................. 103  B.2  Sites ..................................................................................................................................... 108  Appendix C : Material supplemental to Chapter 3 ............................................................................... 111 C.1  Tables .................................................................................................................................. 111  vii  List of Tables Table 2.1 Sampling locations and sample sizes of western painted turtle collected in BC, including estimates of within-site genetic variation and demographic history ............ 21 Table 2.2 Population pairwise matrix of θ values for BC western painted turtles ....................... 30 Table 2.3 Analysis of Molecular Variance (AMOVA) results showing the percentage of variation described by each hierarchical level for the following four grouping strategies: a) genetic clusters informed through the STRUCTURE analysis; b) current designatable units; c) faunal provinces; and d) ecological drainage units ....... 34 Table 2.4 Demographic history and effective population sizes of genetic clusters identified through the STRUCTURE analysis ............................................................................. 36 Table 3.1 Diversity indices from the analysis of the CR and PAX-P1 ......................................... 53 Table 3.2. Percentages of genetic variation in the CR and PAX-P1 intron that are partitioned among and within the morphotype groups determined through AMOVA .................. 54 Table 3.3 Pairwise comparisons of the fixation index, FST, based on the PAX-P1 intron above the diagonal and CR below the diagonal ........................................................... 54 Table A.1 Source and locality information for Chrysemys samples for which c-mos was sequenced ..................................................................................................................... 93 Table A.2 Summary of putative status based on morphology, sex and age class are indicated where known, mtDNA CR haplotype and alleles at two sites in the c-mos oncogene ............................................................................................................ 99 Table B.1 Details about the ecological drainage unit, faunal province (IM- Intermountain, RM- Rocky Mountain, PC- Pacific Coastal), designatable unit (IM-RM Intermountain-Rocky Mountain, PC Pacific Coastal) and coordinates for each site and the mitochondrial control region haplotypes identified at each .................... 103 Table B.2 Characteristics of the 10 microsatellite loci used in this study .................................. 105 Table B.3 Genbank accession numbers and haplotype for each sequence used in the haplotype analysis ...................................................................................................... 106 Table B.4 Migration rates among the 25 sites. ........................................................................... 107 Table C.1 Details about individuals used in each analysis ......................................................... 111 Table C.2 Identity and source of outgroups ................................................................................ 121  viii  List of Figures Figure 1.1 Map of the faunal provinces in Canada for amphibians and reptiles, which corresponds to the designatable units for Chrysemys picta bellii. ............................... 12 Figure 2.1 Map of study area indicating western painted turtle sampling locations within British Columbia .......................................................................................................... 20 Figure 2.2. Haplotype network based on partial mitochondrial control region (671 bp) of western painted turtles from British Columbia, Canada and across the US including Washington, Colorado, New Mexico, Minnesota, Kansas, Wisconsin, South Dakota, Nebraska, North Dakota, Montana, Illinois, Missouri and Iowa. ....... 28 Figure 2.3 Relationship between the pairwise genetic distance (θ) and geographic distance between sites showing a weak (r2= 0.194) but significant (P<0.001) trend of isolation by distance. .................................................................................................... 31 Figure 2.4 Genetic friction map computed for BC western painted turtles .................................. 31 Figure 2.5 STRUCTURE bar plots using a straight admixture model for a. all populations and b. reanalysis of substructure within two broader clusters ...................................... 33 Figure 3.1 Localities for Chrysemys individuals used in the population genetic analyses, including those based on both the mitochondrial control region (CR) and PAX-P1 intron (triangles), CR only (circles and triangles) and PAX-P1 intron only (triangles and squares). ................................................................................................. 50 Figure 3.2 Haplotype network for the CR .................................................................................... 56 Figure 3.3 Haplotype network for the PAX-P1 intron.................................................................. 57 Figure 3.4 Phylogenetic trees for A. the Bayesian analysis of the CR, B. Neighbour Joining of the CR with bootstrap support indicated, C. the Bayesian analysis of the PAX-P1 intron and D. the Bayesian analysis of the concatenated sequences ............. 59 Figure A.1 Map of Burnaby Lake. ................................................................................................ 89 Figure A.2 Morphology of native and non-native painted turtles. ............................................... 89 Figure A.3 Map of the localities of Chrysemys individuals for which the c-mos oncogene was sequenced. ............................................................................................................ 92 Figure A.4 Haplotype network reconstructed based on the mtDNA control region data ............. 97  ix  Acknowledgements This work has involved lot of hands to make it happen. Primarily I must thank my supervisor, Michael Russello, for his support, patience and enthusiasm. My committee members, Purnima Govindarajulu and Karen Hodges, have provided excellent guidance and feedback throughout. I must also thank Mike for providing me with such excellent lab mates. I cannot overstate how important Matt Lemay and Kelsey Robson have been for bouncing ideas off of and for emotional and intellectual support. Jeanette Madsen kindly allowed me to develop my mentoring skills and contributed greatly to the study at Burnaby Lake. I thank Anna Hall for her excellent canoe driving and turtle spotting skills. A genetic assessment is not possible without DNA samples and my study would have far fewer of them if it weren’t for the help of the Western Painted Turtle Working group. I would particularly like to thank Aimee Mitchell, Christian Engelstoft, David Stiles, Deanna Mactavish, Julie Steciw, Kristiina Ovaska, Michelle Evelyn, Nicole Basaraba, Melissa Tesche and Vanessa Kilburn for rolling up their sleeves and collecting samples. I would also like to thank Mark Hall, Kym Welstead and David Zehnder for helping me to identify sites. I would like to recognize the following people, organizations and municipalities for allowing me access to their land to sample, which really shows that they believe in the importance of research and protecting the wildlife in their area: the City of Burnaby, Metro Vancouver, Aldo Cogrossi at the Ruby Lake Resort, the Scout Island Nature Centre, the Regional District of the Central Okanagan, the Creston Valley Wildlife Management Area, Peter Miller, Charles Fipke and David Zehnder. Thank you to Dr. Bradly Shaffer and his lab at UCLA for their collaboration and kindly providing me with sequences and tissue samples from their archives. Sarah Larocque at Queen’s University Biological Station also provided me with valuable tissue samples. Also, thank you to Robert Thomson from U Hawaii for some of the outgroup tissue and sequences used in Chapter 3. Carrie Van Dorp, Hunter Le Blanc, Jeanette Madsen, Matt Lemay and Kelsey Robson each acted as volunteer field assistants; thanks for spending a day in the sunshine (or rain) on a lake with me. Wade Klaver provided excellent technical support when my computing needs exceeded my computing knowledge. This research was funded by the Habitat Conservation Trust Foundation. I was supported by a National Science and Engineering Research Council Postgraduate Scholarship and a University of British Columbia Graduate Fellowship. I am deeply grateful for the unyielding support my family has always given me. Karl, thanks for the tech support. Mom, Dad and Hunter, what can I say but thanks for everything. Last, but not least, I absolutely must thank the 605 turtles in BC that donated blood, shell and DNA to this study.  x  1 Chapter: INTRODUCTION 1.1  Conservation of biodiversity  The main goal of conservation biology is to preserve biodiversity, both in the number and variety of species present in an ecosystem, as well as the genetic diversity within those species (Frankham 2010). It is recognized that all three components of biological diversity (ecosystem, species and genetic) are important to protect (Convention on Biological Diversity 1992; McNeely et al. 1990). The maintenance of intraspecific genetic diversity is critical so that the process of evolution can continue without excessive constraint due to lack of natural variation (Frankel 1974; Mace et al. 2003). Explicitly incorporating genetic management into the protection of threatened species in the wild has emerged as one of the top priorities of conservation biology (Frankham 2010). However, the design and implementation of such strategies remains challenging due to several unresolved issues regarding the appropriate units for conservation. A fundamental issue is the continued lack of agreement over what constitutes a species. It is difficult to assess how many individuals there are in a species, and thereby determine conservation status (not at risk, threatened, endangered etc.), when it is unclear which groups of individuals have membership to that species (Agapow et al. 2004). Likewise, controversy persists regarding the designation of units below the species level for protecting intraspecific genetic diversity.  1.2  Species concepts  Since the time of Plato, philosophers and scientists have been trying to define groups of organisms, with the controversy over the reality and definition of species ongoing today 1  (Richards 2010; Wilkins 2009). Deciding upon a definition of species for conservation purposes is Frankham’s (2010) second top priority for conservation biology, followed by deciding upon a definition of a conservation unit. There are many different species definitions, which according to Lucklow (1995) fall into two categories: mechanistic, where species are seen as units undergoing evolution, and historic, where species are seen as endpoints of evolution. Sites and Marshall (2003) reviewed both non-tree-based and treebased methods for delineating species and concluded that it is not possible to generalize which method is best for more than one taxon.  There are two main species concepts upon which most conservation units are based. Under the biological species concept (BSC), intraspecific units that are evolutionarily important are possible, as species are defined as “ groups of interbreeding natural populations that are reproductively isolated from other such groups” (Mayr 1969). Indirect methods for determining a biological species involve looking for evidence of reciprocal monophyly of mitochondrial genes and concordance of gene-trees (Avise and Ball 1990). The main problem with the BSC is that the boundaries between species are largely untestable, especially in allopatry, and there are many cases where “species” can hybridize and produce viable offspring (Agapow et al. 2004). Also, it is arguable that methods for determining reciprocal monophyly are not valid below the species level, as phylogenetic analysis should be applied only to organisms related by a nested hierarchy, not those related by a pattern of ancestry and descent (see Goldstein et al. 2000 for a detailed discussion). Despite this controversy, the BSC remains one of the most popular and widely taught species concepts, and has many influential proponents (e.g. Avise and Ball 1990; Moritz 1994; Waples 1991).  2  The phylogenetic species concept (PSC) identifies species as “an irreducible cluster of organisms, within which there is a parental pattern of ancestry and descent, and which is diagnosably distinct from other such clusters” (Cracraft 1983). The methods for determining a phylogenetic species are based on finding the boundary between organisms that are related by a nested hierarchy and those that are related by a reticulating genealogy; all individuals within the reticulating genealogy are part of the phylogenetic species (Goldstein et al. 2000).This definition describes the minimal unit to which phylogenetic analysis can be legitimately applied (i.e. relationships between individuals are genealogical) that precludes the possibility for diagnosable, distinct, intraspecific units (Goldstein et al. 2000). If this species concept were accepted, then there would be no need for conservation units below the species level, as each species would be a single, cohesive evolutionary unit. This subdivision is the main fault that biologists have with the PSC: it describes too many entities. In a review by Agapow et al. (2004), application of the PSC resulted in 48% more species being named than other non-phylogenetic concepts. Frankham et al. (2012) caution that this over-splitting of taxa may result in an overall loss of biodiversity if small phylogenetic species of conservation concern suffer from inbreeding depression that could have been avoided through genetic rescue from other phylogenetic species. An advantage to the PSC is that species boundaries are hypotheses that are testable by the addition of more character data (Lucklow 1995).  Another controversy tied to species concepts is the subspecies problem. Since Linnaeus’ time, subspecific designation has been used to describe individuals that share some morphological trait or variant (Haig et al. 2006). However, subspecies is the least agreed  3  upon unit in taxonomy, as there is no established definition and the concept is applied inconsistently across taxa and kingdoms (Haig et al. 2006). The PSC does not recognize subspecies, as species are an irreducible, diagnosable cluster of organisms (Cracraft 1983). It is generally understood that distinct and biologically significant populations should be lent conservation protection, but the semantics of what they should be called (species, subspecies, or conservation unit) and how to define such populations remains unclear.  1.3  1.3.1  Intra-specific conservation units  The Evolutionarily Significant Unit  Conservation may involve the recognition of units below the species level to which recovery goals and efforts are applied. These units were typically based on the taxonomic level of subspecies, until Ryder (1986) published a paper describing the fundamental problems with basing conservation on taxonomic units that do not necessarily have an evolutionary basis. The concept of an evolutionarily significant unit (ESU) was proposed as a way to identify populations for conservation that were truly reflective of the evolutionary trajectory of the species (Ryder 1986), as opposed to using the subjective unit of subspecies. Zoo systems were being overwhelmed by the vast number of named subspecies apparently in need of ex situ conservation, while many of the subspecies were of uncertain validity. The ESU concept stemmed out of the need for zoo biologists to decide for themselves which lineages represent genetic variation significant to the present and future of the species and are most important to preserve within the limitations of ex situ conservation (Ryder 1986).  4  Ryder (1986) acknowledged that identifying ESUs would be difficult and recommended using a combination of natural history, phylogeography, and genetic analysis to delineate these intraspecific units. Since then, both the criteria for identifying and the definition of an ESU have undergone many interpretations by different authors. Waples (1991) defined a “species” under the U.S. Endangered Species Act as a population segment meeting two criteria: reproductive isolation from other conspecific units, and representing an important part of the evolutionary legacy of the species (either unique ecologically or genetically). His criteria for determining if a unit is evolutionarily important include: genetic distinctiveness, occupation of unique habitat, possession of unique adaptations to the environment and significant loss to the ecological or genetic diversity of the species if that unit became extinct. Dizon et al. (1992) wrote that ESUs must be evaluated based on their adaptive genetic uniqueness, and that behaviour, morphology and geographic location could be used as proxies for genetic adaptation. Unlike previous definitions, Vogler and Desalle (1994) advocated for the use of character data for cladistics analysis and population aggregation analysis to evaluate ESUs and turn them into testable hypotheses. The most broadly applied ESU definition to date is that of Moritz (1994). In this case, ESUs are those that are reciprocally monophyletic for mitochondrial haplotypes and show statistically significant divergence of allele frequencies at nuclear loci. This definition, although widely applied, is critiqued as being highly restrictive and incompatible with the phylogenetic species concept (Goldstein et al. 2000).  Other concepts attempt to add flexibility to the ESU concept. Frazer and Bernatchez (2001) attempted to reconcile the proposals of other authors by offering adaptive evolutionary  5  conservation (AEC) as a compromise, because they did not think that a single, rigid universal definition of an ESU will be appropriate in all situations. Under AEC, a conservation unit should be “a lineage demonstrating highly restricted gene flow from other such lineages within the higher organizational level of the species” (Fraser and Bernatchez 2001). They felt that any criteria, including adaptive divergence and historical isolation, that give evidence for lineage sorting as a result of restricted gene flow could be used to delineate biologically meaningful ESUs. In contrast, Crandall et al. (2000) argued against the ESU/not-ESU dichotomy of other authors’ definitions. They proposed using a continuum of classifications describing how exchangeable individuals are between populations, taking into account ecological, genetic, recent and historical exchangeability and scoring each category (Crandall et al. 2000)  A commonality among many of the versions of the ESU is the need for evidence of genetic divergence between units. There is debate, however, about the roles that neutral versus adaptive markers should play. Some authors maintain that neutral loci can be used to reliably estimate patterns of reproductive isolation and the demographic history of populations (Allendorf et al. 2010; Miller et al. 2010), which are the mechanisms that result in divergence. Other authors argue that patterns in neutral maker variation are not indicative of underlying adaptive genomic variation (Kohn et al. 2006; Ouborg et al. 2009; Reed and Frankham 2001; Sgro et al. 2011) and that units that appear differentiated at neutral loci are not necessarily significantly evolutionarily distinct. The term “partial-ESU” was put forth by De Guia and Saitoh (2006) to describe a unit that was designated based on either neutral or adaptive markers but not both. De Guia and Saitoh (2006) argued that both types of marker  6  need to be used to accurately recognize the status of a unit, and full ESU rank should be reserved for units supported by both types of evidence. Genomic information that would allow the identification of adaptive divergence is not currently available for most organisms of conservation concern, however, recent technological advances, such as the development of the double digest RAD sequencing pipeline (Peterson et al. 2012), may allow the development of adaptive genomic markers in non-model organisms. There are no clear guidelines about how such information could be used to delineate conservation units, however several authors have begun to make proposals (e.g. Bonin et al. 2007; Funk et al. 2012).  1.3.2  Policy based conservation units  In the face of rapid declines in many species and the lack of information on genetic and phylogenetic relatedness, conservation units are sometimes designated based on applied management imperatives. Although based on scientific principles, the designation of these conservation units is not necessarily based on species-specific genetic or phylogenetic information. The geopolitical unit (GU) was proposed to lend protection to populations within a particular political region, regardless of the status of conspecific populations in other political regions and independent of morphological, genetic or reproductive criteria (Karl and Bowen 1999). In Canada, the Species at Risk Act (SARA) lends protection to ‘distinct populations of wildlife’, which are identified by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) according to the designatable unit (DU) concept. COSEWIC (2008) may delineate DUs that are named subspecies or populations that are discrete and evolutionarily or ecologically significant without formal taxonomic distinction. Evidence of  7  population discreetness could include morphology, life history or behavioural traits and genetic evidence based on neutral genetic markers (COSEWIC 2004b). The significance of a population is assessed based on phylogenetic divergence, occupation of unique ecological settings resulting in local adaptation, evidence that it is the last surviving population within the species’ historical range (although it may be introduced elsewhere) and evidence that the loss of that population would result in a gap in the range of the species in Canada (COSEWIC 2004b). This Canadian policy is similar to that lending protection to distinct population segments under the Endangered Species Act in the United States of America, although there are differences in the way that the policies are applied (US Fish and Wildlife Service and National Oceanic and Atmospheric Administration 1996).  1.4  Other issues in conservation  The conservation of wide ranging species requires special consideration because, across a species’ distribution, populations at the margins of the range may represent unique pockets of genetic diversity and ecological characteristics (Gaston 2003). This pattern can be due to reduced gene flow with the core of the range and small population sizes resulting in elevated levels of differentiation and disparate selection pressures (Lesica and Allendorf 1995). Conserving and maintaining healthy peripheral populations may be important for the longterm persistence of species, as it has been found that the ranges of many endangered mammals have collapsed not into the core of their historical ranges, but in fact into their periphery (Lomolino and Channell 1995). In the northern hemisphere, the trend of collapse tends to be into the northwestern periphery of the range (Channell and Lomolino 2000). This  8  pattern goes against some of the most common recommendations for conservation, which put protecting core populations at a higher priority than peripheral ones (Griffith et al. 1989).  The current biodiversity crisis is affecting some taxa out of proportion to others. Turtles are one such group, with at least 47% of the 330 currently described terrestrial and freshwater species threatened with extinction (Rhodin et al. 2010). As of 2012, six out of 12 species of terrestrial and freshwater turtle with the northern periphery of their range in Canada are listed as threatened or endangered by COSEWIC (Kiester and Olson 2011) with a further two species of special concern and one species extirpated. The major causes of turtles’ decline globally are habitat destruction, road mortality, pollution and over-exploitation for human use in traditional medicine, as pets and for meat (Alacs et al. 2007; Gibbon et al. 2000; Kiester and Olson 2011). Climate change is also seen as a looming threat for the turtle species that have temperature-dependent sex determination (Gibbon et al. 2000; Telemeco et al. 2013).  Turtles are one of the most ancient extant lineages on the animal tree of life, but certain life history characteristics that may have helped them survive mass extinctions and global change in the past might now be working against them. Turtles are long lived with advanced age at sexual maturity often resulting in overlapping generations, which can have serious implications for inbreeding levels in small populations (Alacs et al. 2007). They also have mutation rates eight times slower than other vertebrates within the mitochondrial genome (Avise et al. 1992). With these factors in mind, it is imperative that turtle conservation  9  strategies are designed to preserve intraspecific genetic diversity because diversity, once lost, is not likely to regenerate quickly (Alacs et al. 2007).  1.5  1.5.1  The western painted turtle  Chrysemys taxonomy  Chrysemys (Schneider 1783) is a genus within the family Emydidae that is experiencing ongoing taxonomic revisions due to debate as regards to the status of four formally described morphotypes. Currently, the Society for the Study of Amphibians and Reptiles recognises two species: Chrysemys dorsalis and Chrysemys picta, the latter of which is composed of three subspecies: C. p. picta, C. p. bellii and C. p. marginata (Crother 2012). These designations are largely based on a study that looked at sequences of the mitochondrial control region (CR) of individuals across the United States, and showed C. dorsalis as a different evolutionary lineage from the other three morphotypes (Starkey et al. 2003). Other taxonomists disagree with the elevation of C. dorsalis to the species level, and argue that it does not have either geographic or reproductive isolation from the subspecies within C. picta (Ernst and Lovich 2009). Interbreeding and hybridization are common wherever the ranges of the morphotypes overlap (e.g. Weller et al. 2010) to the extent that some speculate that a continuously shared gene pool may exist over the entire range of Chrysemys (Ernst and Lovich 2009).  10  1.5.2  Chrysemys picta bellii  Chrysemys picta bellii (Gray 1831), commonly known as the western painted turtle, is the largest and most distinctively coloured subspecies. The range of C. p. bellii extends across the southernmost part of Canada west of the Great Lakes to British Columbia, and in the United States as far south as Missouri continuously and into Mexico as isolated populations (Fritz and Havas 2007). There are currently no accurate range maps available from reputable sources for Chrysemys.  1.5.3  British Columbian western painted turtles  In 2006, COSEWIC reviewed the status of C. p. bellii in Canada. Three DUs were identified: Pacific Coastal population (Endangered), Intermountain-Rocky Mountain population (Special Concern), and Prairie/Western Boreal-Canadian Shield population (Not at Risk) (Figure 1.1). In the absence of species specific information for herptofauna, COSEWIC recognizes eight distinct faunal provinces (Figure 1.1) and these faunal provinces are then used to delineate DUs for species, as was done for the western painted turtle. Both the Pacific Coastal and Intermountain-Rocky Mountain DUs occur in BC and encompass three faunal provinces (COSEWIC 2006). As these conservation units were delimited based on the generic faunal provinces, it is unclear whether these three DUs reflect distinct genetic units.  11  Figure 1.1 Map of the faunal provinces in Canada for amphibians and reptiles, which corresponds to the designatable units for Chrysemys picta bellii. Reproduced from COSEWIC (2006)  The endangered status of the Pacific Coastal population is particularly controversial because there is anecdotal evidence that at least some of these populations or some individuals within these populations have been introduced as a result of humans releasing turtles from elsewhere in their range (COSEWIC 2006). If this is the case, then their conservation value is questionable. Records of western painted turtles on Vancouver Island date back to the 1920s (COSEWIC 2006), so it is possible that the Pacific Coastal populations are native, but that turtles there have always been rare. Genetic comparison of the relatedness of Pacific Coastal western painted turtles to other populations could help determine whether they are native or have been introduced over time to that region.  12  C. p. bellii is one of only two species of freshwater turtle native to British Columbia. The other species, Emmys marmorata marmorata (Northern Pacific Pond Turtle), has been extirpated from Canada (COSEWIC 2004a). Red eared sliders (Trachemys scripta), an introduced species, are becoming increasingly common in ponds, lakes and streams in BC, with robust populations in the Fraser Valley and elsewhere in Western North America as a result of pets being released (Bunnell 2005). The presence of a non-native potential competitor is seen as a threat to western painted turtles in BC (COSEWIC 2006).  Of growing concern in the Lower Mainland of BC is the appearance of putative introduced individuals of the other subspecies of C. picta. Hybridization is a common occurrence where the ranges of the different subspecies naturally overlap (Weller et al. 2010); however, the presence of other subspecies in B.C. is due to human interference, likely via the release of unwanted pets. Interbreeding between native and introduced individuals would result in the introduction of alleles that could be detrimental to the local populations if those alleles are maladaptive to the local conditions (Templeton 1986). Also, genetic contamination could affect the conservation value of introgressed populations (Allendorf et al. 2001).  Due to clinal variation in morphology within the range of each subspecies (Ultsch et al. 2001), it is difficult to determine whether individuals are hybrids based solely on morphology. Hybridization resulting in intermediate morphology is common wherever the ranges of subspecies overlap (Weller et al. 2010). Inability to distinguish hybrids based on morphology means that genetic methods will need to be used to detect hybrids reliably and determine levels of introgression in a population.  13  1.6  Thesis objectives  The primary goal of this thesis is to carry out an explicit genetic assessment of western painted turtles across BC to reconstruct their population history, phylogenetic placement and conservation status. In chapter 2, I use mitochondrial DNA haplotypic and microsatellite genotypic data to assess the extent and distribution of genetic variation within and among BC western painted turtle populations. I further use this information to evaluate the current DUs and examine what the conservation priorities in BC would be under the conservation unit criteria of Moritz (1994) and COSEWIC (2008) and reconstruct demographic history. In chapter 3, I place BC western painted turtles in the context of the broader species distribution using mitochondrial and nuclear gene sequence data and examine the phylogenetic relationships among the four morphological types or subspecies within Chrysemys to further investigate taxonomic uncertainty and paleogeographic hypotheses within the group.  14  2 Chapter: APPLYING CONSERVATION UNIT CONCEPTS TO WESTERN PAINTED TURTLES AT THEIR NORTHERN PERIPHERY 2.1  Background  In the face of the on-going biological diversity crisis, an ever increasing number of species are considered threatened with extinction (Vié et al. 2009). As biodiversity continues to be lost, strategies for prioritizing populations for conservation have become increasingly important (Avise 2005). The maintenance of intraspecific genetic diversity is of particular importance for preserving both evolutionary history and the potential for future adaptation (Frankham 2010; Frankel 1974). In order to protect this diversity effectively, units below the species level need to be defined, and should be the level at which recovery goals and efforts are typically applied. Incorporating genetic management into the protection of threatened species in this way is one of the top priorities of conservation biology (Frankham 2010).  The delineation of units below the species level for conservation purposes is subject to many challenges, with no one strategy that applies universally across taxa and systems. Chief among these difficulties is deciding what criteria (genetic, geographic, political etc.) should be used to assess whether units are sufficiently diverged to warrant separate management and which genetic markers, if any, should be used in this determination. Subspecies constitute a controversial unit in taxonomy (Haig et al. 2006), and have been previously deemphasized as an appropriate unit for conservation by Ryder (1986) in favour of non-taxonomic methods. Since that time, a primary focus has been on delineating evolutionarily significant units (ESU), originally defined as real biological entities that possess a significant evolutionary  15  lineage (Ryder 1986). The procedures for delineating such units, however, have undergone many different interpretations by various authors since it was first proposed (Crandall et al. 2000; Dizon et al. 1992; Fraser and Bernatchez 2001; Moritz 1994; Vogler and Desalle 1994; Waples 1991).  The ESU sensu Moritz (1994) is the most widely applied version, principally because it explicitly uses the genetic criteria of reciprocal monophyly of mitochondrial DNA variation and significant allele frequency divergence at nuclear loci to define units for conservation. Units defined in this way consist of groups of populations that have been historically isolated from other such groups of populations. Within an ESU, populations that have diverged at allele frequencies are also recognized as significant for conservation as they are connected by low levels of gene flow and are functionally independent. These intra-ESU groups are referred to by Moritz (1994) as management units (MU) and are the level at which demographic and population monitoring should take place.  The ESU is not the only conservation unit concept, with many others having been defined based on their use of genetic and ecological data, type of genetic marker (nuclear versus organellar; neutral versus putatively adaptive loci) and level of emphasis along the spectrum from populations to species (Karl and Bowen 1999; Wood and Gross 2008). Among these other methods are a class of concepts that recognize that, in the face of rapid declines in many species and the lack of information on genetic and phylogenetic relatedness, conservation units must sometimes be designated based on management imperatives. In Canada, the Species at Risk Act (SARA) lends protection to ‘distinct populations of  16  wildlife’, which are identified by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) according to the designatable unit (DU) concept. COSEWIC may delineate DUs as named subspecies or populations that are discrete and evolutionarily or ecologically significant even if these units lack formal taxonomic distinction (COSEWIC 2008). Evidence for population discreteness may include genetics, geographic disjunction, morphology, life history and/or behaviour (COSEWIC 2008). The significance of a population is assessed based on phylogenetic divergence, occupation of unique ecological settings resulting in local adaptation, evidence that it is the last surviving population within the species’ historical range (although it may be introduced elsewhere), and/or evidence that the loss of that population would result in a gap in the range of the species in Canada (COSEWIC 2008). This Canadian policy is similar to that lending protection to distinct population segments under the Endangered Species Act in the United States of America, although there are important differences in the way that the policies are applied (US Fish and Wildlife Service and National Oceanic and Atmospheric Administration 1996).  The conservation of wide ranging species requires special consideration because, across a species’ distribution, populations at the margins of the range may represent unique pockets of genetic diversity and ecological characteristics (Gaston 2003). This pattern can be due to reduced gene flow with the core of the range and small population sizes resulting in elevated levels of differentiation and disparate selection pressures (Lesica and Allendorf 1995). Conserving and maintaining healthy peripheral populations may be important for the longterm persistence of species, as it has been found that the ranges of many endangered mammals have collapsed, not into the core of their historical ranges, but towards their  17  periphery (Lomolino and Channell 1995). In the northern hemisphere, the trend of collapse tends to be into the northwestern periphery of the range (Channell and Lomolino 2000). This pattern goes against some of the most common recommendations for conservation, which emphasize protection of core populations to a greater degree than peripheral ones (Griffith et al. 1989).  The painted turtle Chrysemys picta is a wide spread species, with a range that is extensive throughout the United States of America and the southernmost portions of Canada (Ernst and Lovich 2009). The western painted turtle (Chrysemys picta bellii) is one of four recognized subspecies, and is the only native member of the Emydidae extant in western Canada. In 2006, COSEWIC assessed the status of western painted turtles in Canada due to concerns that habitat loss was causing major population declines in parts of British Columba (BC). That process identified three designatable units (DU) of western painted turtle in Canada, two of which occur within the political borders of BC: the Pacific coastal population (Endangered) and the Intermountain-Rocky Mountain population (Special Concern) (COSEWIC 2006). The third DU is listed as Not at Risk, and encompasses the rest of the range of the western painted turtle within Canada, spanning from Alberta to Ontario. When population genetic evidence is not available, COSEWIC may delineate DUs based on generic faunal provinces, as was the case with western painted turtles in 2006. It is unclear whether these three DUs reflect distinct genetic units.  In this study, I used mitochondrial DNA haplotypic and microsatellite genotypic data to assess the extent and distribution of genetic variation within and among BC western painted  18  turtle populations. Observed patterns were used to reconstruct demographic history, infer population connectivity, and examine current designatable units in light of the genetic evidence to inform management actions at the northern periphery of the species’ range.  2.2  2.2.1  Materials and methods  Sampling  Samples were collected from 475 individuals of western painted turtle from 25 sites within BC during 2011-2012 (Figure 2.1, Table 2.1, Table B.1). An additional 73 individuals were sampled from four more sites but later excluded from this study (see Appendix A and B). An individual site consisted of a single discrete pond or lake, except in two cases, where neighbouring lakes (Baird and Hidden Lake, <1.3 km apart) or lakes on a small island (Texada Island: Case, Emily, Capsheaf and Priest Lakes) were considered a single site (see Appendix B). These sites span the known geographic range of the species in BC (Figure 2.1). Turtles were caught using hoop traps baited with canned sardines or by hand using a dip net. At each site, a blood sample (~100µl) from six to 37 individuals (Table 2.1) was collected from either the dorsal coccygeal vein or brachial artery. Blood was stored in tubes containing a buffer solution (100 mM Tris-HCl, pH 8.0; 100 mM EDTA, pH 8.0; 10 mM NaCl; 0.5% SDS; Longmire et al. 1997) and stored at 4 ˚C until analysis. At some sites, tissue samples were excised from road-killed individuals and stored in 100% ethanol or shell samples were clipped from the margin of scutes on the upper plastron and stored in paper envelopes or dry tubes. All samples were collected in accordance with University of British Columbia Animal  19  Care Certificate # A11-0163 and sampling permit # VI11-71744 from the BC Ministry of Environment.  Figure 2.1 Map of study area indicating western painted turtle sampling locations within British Columbia Site acronyms follow Table 2.1. Colours and symbols indicate genetic clusters as follows: Sunshine coast-Gulf Island-Mid-Vancouver Island (SGV; orange circle), Sunshine coast (purple open square), South coast (blue triangle), Cariboo (pink star), Thompson-Okanagan (green diamond), Kootenay (yellow square). The heavy line represents the border between British Columbia, Canada and Washington, USA  20  Table 2.1 Sampling locations and sample sizes of western painted turtle collected in BC, including estimates of within-site genetic variation and demographic history  Site  Site Name  Genetic Cluster ThompsonOkanagan South coast  N  AR  He  Ho  PA  FIS  LR-R  BA  Baird Lake  23  5.22  0.69  0.72  0  -0.05(-0.16 - 0.01)  0.16  BE  Beaver Lake  BL  Burnell Lake  18  5.89  0.76  0.81  4  -0.07(-0.22 - 0.0)  0.25  ThompsonOkanagan Kootenay  20  6.22  0.75  0.66  0  0.13(0.02 - 0.18)  0.08  CV DO  Creston Valley Wildlife Management Area Dorothy Lake  15  6.67  0.75  0.77  2  -0.03(-0.15 - 0.01)  0.08  EL  Elizabeth Lake  Kootenay  14  5.67  0.76  0.84  0  -0.11 (-0.25 - -0.06)  0.12  Kootenay  18  6.67  0.76  0.71  0  0.07 (-0.05 - 0.13)  0.08  ER  Erie Lake  Kootenay  6  4.78  0.75  0.70  0  0.07 (-0.3 - 0.14)  0.16  JO  Johnson Lake  Kootenay  21  6.11  0.69  0.66  0  0.04 (-0.07 - 0.1)  0.15  KL  Klein Lake  15  2.89  0.51  0.61  0  -0.21 (-0.36 - -0.13)  0.46  LO  Loon Lake  Sunshine Coast Kootenay  25  7.00  0.76  0.72  2  0.06 (-0.03 - 0.09)  0.12  NI  Nicomen Slough  South coast  21  4.44  0.65  0.74  4  -0.16 (-0.26 - -0.11)  0.34  OF  ThompsonOkanagan SGV  21  6.56  0.75  0.74  0  0.0 (-0.1 - 0.04)  0.09  PA  Fipke Lake near Oyama Port Alberni  13  3.89  0.57  0.56  1  0.02 (-0.16 - 0.1)  0.27  RE  Redlich Pond  7  5.00  0.72  0.63  1  0.12 (-0.17 - 0.22)  0.02  RV  Revelstoke Marsh  20  6.44  0.77  0.76  2  0.01 (-0.1 - 0.05)  0.14  RO  Rosebud Lake  ThompsonOkanagan ThompsonOkanagan Kootenay  25  5.44  0.67  0.65  1  0.04 (-0.06 - 0.09)  0.2  RU  Sunshine Coast Cariboo  18  4.44  0.61  0.63  0  -0.03 (-0.16 - 0.05)  0.26  24  4.22  0.65  0.70  0  -0.07 (-0.18 - 0.01)  0.34  6.44  0.74  0.69  0  0.06 (-0.03 - 0.11)  0.11  19  5.56  0.71  0.71  0  -0.01 (-0.16 - 0.09)  0.13  SK  Skmana Lake  20  4.56  0.61  0.61  0  0.0 (-0.13 - 0.08)  0.2  SW  Swan Lake  ThompsonOkanagan ThompsonOkanagan ThompsonOkanagan South coast  37  SH  Ruby Lake Lagoon Scout Island Nature Centre Stephen Coyote Regional Park Shannon Lake  10  3.67  0.60  0.72  0  -0.22 (-0.47 - -0.17)  0.28  TE  Texada Island  SGV  17  3.78  0.51  0.45  0  0.13 (-0.06 - 0.23)  0.32  TI  Tie Lake  Kootenay  21  6.00  0.72  0.68  0  0.05 (-0.06 - 0.12)  0.13  WE  West Lake  SGV  27  3.44  0.55  0.56  0  -0.01 (-0.16 - 0.08)  0.45  SI SC  Total/ 47 5.20 0.68 0.68 17 -0.17 0.2 Average 5 N number of individuals, AR mean number of alleles, He expected heterozygosity, Ho observed heterozygosity, PA private alleles, FIS inbreeding coefficient with 95 % confidence interval, LR-R Lynch Ritland relatedness estimator  21  2.2.2  Data collection  DNA was extracted from blood samples using the Nucleospin QuickBlood kit (MachereyNagel) following manufacturer’s protocols. DNA was extracted from tissue samples using the Nucleospin Tissue kit (Macherey-Nagel) following manufacturer’s protocols, except in the case of shell samples, where the initial incubation with Proteinase K was carried out at 45˚C overnight (~16 hours).  For a representative sample from each site (1-3 individuals), a 671 base pair fragment of the mitochondrial DNA (mtDNA) control region (CR) was amplified as a single fragment using DES1 (Starkey et al. 2003) and Cp_CRExt primer (5’-GCTCTCGGATTTAGGGGTTT-3´). Polymerase chain reactions (PCR) were carried out on an ABI Veriti thermal cycler in 25 µl reactions containing: ~20-40 ng DNA, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 200 µM dNTPs, 7.5µg bovine serum albumin, 0.4µM each primer and 0.5 U Kapa Taq DNA Polymerase (Kapa Biosystems). Cycling conditions consisted of 95˚C for 2 minutes, 35 cycles of 95˚C for 30 seconds, 55˚C for 1 minute, 72˚C for 2 minutes, and a final extension at 72˚C for 10 minutes. The sequencing reactions were performed using ABI BigDye v3.1 Terminator chemistry and sequences were run on an Applied Biosystems 3130XL DNA automated sequencer. Sequences were visualized and edited using Sequencher 5.0 (Gene Codes Corporation).  Nuclear genotypic data were collected for all 475 samples at 10 microsatellite loci (Table B.2) (Gonçalves da Silva et al. 2009; Hauswaldt and Glenn 2003; King and Julian 2004; Pearse et al. 2001). All forward primers were 5’- tailed with an M13 sequence and used in  22  combination with an M13 primer of the same sequence 5’-labeled with one of four fluorescent dyes (6-FAM, VIC, NED, PET) to facilitate automated genotyping (Schuelke 2000). PCRs were carried out on an ABI Veriti thermal cycler in 12.5 µl reactions containing: ~20-40 ng DNA, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 200 µM dNTPs, 7.5µg bovine serum albumin, 0.04 µM of the M13-tailed forward primer, 0.4 µM each of the reverse primer and the M13 fluorescent dye-labeled primer, and 0.5 U of AmpliTaq Gold DNA polymerase (Applied Biosystems). Reaction conditions for all primer sets were optimized using a ‘touchdown’ cycling program that consisted of: 95˚C for 10 min; 35 cycles of 95˚C for 30 s, annealing for 30 s, and 72˚C for 45 s; and a final step of 72˚C for 7 min. We used two versions of the annealing step in the ‘touchdown’ program. The annealing temperature decreased 1˚C each cycle from 59˚C or 55˚C until it reached 51˚C or 45˚C, respectively, at which point the remaining cycles continued with a 51˚C or 45˚C annealing temperature. Loci were co-loaded and run on an Applied Biosystems 3130XL DNA automated sequencer. All alleles were scored using bins in the software Genemapper 4.0 (Applied Biosystems, Inc.). Bins were evaluated with the program Tandem (Matschiner and Salzburger 2009), and all allele calls were manually verified.  2.2.3  Haplotypic variation and network analysis  In addition to the mtDNA sequence data collected for my study in BC, exemplar sequences from across the range of the western painted turtle were taken from GenBank (Starkey et al. 2003) (see Table B.3 for accession numbers). The mtDNA sequences were aligned in Geneious (Biomatters Ltd.) using Geneious Aligner (default settings) and a haplotype  23  network was generated using statistical parsimony, as implemented in TCS (Clement et al. 2000).  2.2.4  Genotypic variation and population differentiation  The genotypic data set was examined for the presence of null alleles using MICROCHECKER (Van Oosterhout et al. 2004). To test whether any loci deviated from neutral expectations, we conducted FST outlier detection tests as implemented in LOSITAN (Antao et al. 2008). This method creates an expected distribution of the relationship between FST and expected heterozygosity (He) with neutral markers, and then uses that distribution to identify loci that have significantly higher or lower FST compared to neutral expectations. Outlier loci are candidates for exhibiting locus-specific effects, including due to selection. Deviation from Hardy-Weinberg equilibrium (HWE) was assessed using exact tests, as implemented in GENEPOP 3.3 (Raymond and Rousset 1995; Rousset 2008). Linkage disequilibrium (LD) was investigated for all pairs of loci using GENEPOP 3.3 (Raymond and Rousset 1995; Rousset 2008). Type I error rates for tests of linkage disequilibrium and departure from HWE were corrected for multiple comparisons using the sequential Bonferroni procedure (Rice 1989).  Allelic diversity, observed (Ho) and expected heterozygosity (He) were calculated at each locus for each site using ARLEQUIN (Excoffier et al. 2005). A list of private alleles was tabulated using GenAlEx (Peakall and Smouse 2006).  24  Levels of genetic differentiation among sites were estimated by pairwise site comparisons of θ (Weir and Cockerham 1984) as calculated in FSTAT (Goudet 2001). Multiple comparisons were corrected for using the false discovery rate correction (Benjamini and Hochberg 1995), as advocated by Narum (2006) for use in conservation genetic studies. A matrix of the geographic distance between sites was created using the Geographic Distance Matrix Generator (Ersts 2012). To test the relationship between genetic differentiation and geographic distance, a Mantel test was performed between the matrices of θ values and geographic distance using the Isolation by Distance Web Service (Jensen et al. 2005). To further examine isolation by distance patterns, a non-stationary genetic friction map was developed to display geographic areas with a relatively larger genetic difference per unit of geographic distance using a similarity matrix (1-θ), 4 simulated neighbours at a distance of 0.1 and 100 posterior replicates as implemented in LOCALDIFF (Duforet-Frebourg and Blum 2012). The directions and magnitudes of contemporary migration rates among sites were estimated using a non-equilibrium Bayesian method implemented in BAYESASS version 3 (Wilson and Rannala 2003). The program run length was 10,000,000 MCMC replicates after a burn-in period of 1,000,000, sampling the chain every 100 iterations. Consistency of the results was assured by running the program five times using different random seeds and monitoring the output as visualized in TRACER (Rambaut and Drummond 2007).  To determine the number of discrete genetic units within the data set, the Bayesian method of Pritchard et al. (2000) was used as implemented in STRUCTURE 2.3.4. Run length was set to 500,000 MCMC replicates after a burn-in period of 250,000 using correlated allele  25  frequencies under a straight admixture model as well as using the LOCPRIOR option, the latter of which uses sampling locations as prior information to assist in clustering. The most likely number of clusters was determined by varying the number of clusters (K) from 1 to 25 with ten iterations per value of K and calculating ΔK (Evanno et al. 2005) as implemented in STRUCTURE HARVESTER (Earl and vonHoldt 2011). To evaluate whether the genetic clusters identified through the STRUCTURE analysis conform to Hardy-Weinberg Equilibrium, an exact test for heterozygote deficit was implemented in GENEPOP (Rousset 2008; Raymond and Rousset 1995). Clusters that showed heterozygote deficit indicative of a potential Wahlund effect (Wahlund 1928) were separately analysed in STRUCTURE to determine if further substructure existed.  Analyses of molecular variance (AMOVAs) were performed using ARLEQUIN (Excoffier et al. 2005) for a range of grouping strategies including: 1) current DUs; 2) faunal provinces; 3) ecological drainage units of BC (EAU BC, accessed through iMap BC); and 4) clusters identified by STRUCTURE (see Table B.1 for further details). The faunal provinces were chosen as an a priori grouping strategy because they were used when delineating the DUs. The ecological drainage units were chosen as an a priori grouping strategy as they are an alternative strategy to the faunal provinces, both of which are biogeographically based.  2.2.5  Demographic history  The effective population size (Ne) of each STRUCTURE cluster was estimated using LDNe (Waples and Do 2008), which implements the bias correction of Waples (2006). Random  26  mating was assumed and alleles with frequencies lower than 0.02 were excluded from analyses.  Genetic signatures of demographic contraction were assessed using the heterozygote excess test and the mode-shift test, both implemented in BOTTLENECK 1.2.02 (Piry et al. 1999) and the M-ratio test using M_P_VAL.exe and critical_M.exe (Garza and Williamson 2001). For the heterozygote excess test, 1000 iterations were used with the Wilcoxon test under the two phase model with 90% stepwise mutations. For the M-ratio we calculated θ (θ=4Neµ) using a mutation rate (µ) of 1.5x10-4 and a pre-bottleneck Ne of 50, 500 and 1500, resulting in θ = 0.03, 0.3 and 0.9, respectively. Multiple values of θ were used to assess how robust conclusions were to permutation of that parameter. We used 3.5 base steps for multi-step mutations, and the amount of single step mutations, Ps, was 0.1.  Genetic signatures of post-glacial expansion into BC were investigated using two methods that infer population expansion based on microsatellite gene genealogies: a within-locus k test and an inter-locus g test (Reich et al. 1999), using the kgtests EXCEL macro (Bilgin 2007). The significance table in Reich et al. (1999) was used to assess the g values.  Relatedness and inbreeding coefficients were estimated among individuals sampled at a site. Pairwise relatedness was calculated within and among sites according to the method of Lynch and Ritland (1999) in COANCESTRY (Wang 2011). The inbreeding coefficient, FIS, was calculated for each sampling site as implemented in GENETIX (Belkhir et al. 2004).  27  2.3  Results  2.3.1  Haplotypic variation  The mtDNA CR fragment was sequenced in 72 individuals from across BC. One major (83%) and three low frequency (7%, 7%, 3%, respectively) haplotypes were identified, each differing by a single character from the major haplotype (Figure 2.2). Two of these haplotypes were unique to BC as determined by a comparative analysis with the 90 sequences from across the range of the western painted turtle published in Starkey et al (2003).  Figure 2.2. Haplotype network based on partial mitochondrial control region (671 bp) of western painted turtles from British Columbia, Canada and across the US including Washington, Colorado, New Mexico, Minnesota, Kansas, Wisconsin, South Dakota, Nebraska, North Dakota, Montana, Illinois, Missouri and Iowa. Each haplotype is represented by a circle and numerical identifier. The circles filled in colour are those present in BC, with an asterisk indicating haplotypes unique to BC. Inset pie chart represents the proportion of each haplotype recovered. Unsampled intermediate haplotypes are represented by a solid circle  28  2.3.2  Genotypic variation and population differentiation  The Cp2 locus had null alleles and was removed from all further analyses. Other loci (GmuD21, GmuD28, GmuD62, GT108, Terp2 and Terp7) were also found to have null alleles in a small proportion of sites (n=1-9 out of 25), with three sites having two loci exhibiting null alleles. None of the loci were found to deviate from neutrality as revealed by FST-based outlier tests (data not shown). Deviation from HWE was found at some loci at some sampling sites; subsequent analyses with and without the allele calls for those individuals at those loci produced highly similar results (data not shown). Non-random association of genotypes was found in only eight of the 900 pairwise tests for linkage disequilibrium (Gmu21/Gmu28 in NI, RO, SI, TE and WE, Gmu21/Gmu62 in BA and BE and Gmu62/Terp2 in RO). As these patterns were not consistent across sites, all downstream analyses were based on the nine retained microsatellite loci. The final dataset included 5.7% missing data.  Across the nine microsatellite loci, there was an average of 5.2 alleles in each sampling site, with 17 private alleles detected within 10 sites (Table 2.1). Overall Ho and He were 0.681 and 0.681, respectively (Table 2.1).  Significant pairwise population differentiation was detected across most comparisons (Table 2.2). A weak pattern of isolation by distance was detected (r2=0.194, P<0.001) based on Mantel tests. The genetic friction map (Figure 2.3) revealed a higher degree of genetic friction (defined here as genetic difference per unit of geographic distance) along the coast, which corresponds to sites sampled on islands and the nearby mainland.  29  Table 2.2 Population pairwise matrix of θ values for BC western painted turtles BA  BE  BL  CV  DO  EL  ER  JO  KL  LO  NI  OF  PA  RE  RV  RO  RU  SI  SC  SH  SK  SW  TE  BE  0.10*  BL  0.05*  0.10*  CV  0.09*  0.06*  0.07*  DO  0.08*  0.10*  0.05*  0.03*  EL  0.09*  0.11*  0.06*  0.04*  0.01*  ER  0.11*  0.10*  0.09*  0.05*  0.05*  0.05*  JO  0.15*  0.14*  0.10*  0.07*  0.03*  0.02*  0.10*  KL  0.19*  0.23*  0.20*  0.18*  0.18*  0.20*  0.22*  0.24*  LO  0.11*  0.11*  0.08*  0.03*  0.02*  0.03*  0.04*  0.07*  0.20*  NI  0.14*  0.12*  0.11*  0.14*  0.13*  0.15*  0.20*  0.17*  0.26*  0.17*  OF  0.04*  0.10*  0.02*  0.07*  0.07*  0.07*  0.09*  0.13*  0.17*  0.08*  0.09*  PA  0.13*  0.18*  0.08*  0.14*  0.10*  0.13*  0.21*  0.17*  0.18*  0.17*  0.19*  0.09*  RE  0.01*  0.09*  0.02*  0.07*  0.06*  0.08*  0.08*  0.14*  0.15*  0.08*  0.15*  0.02*  0.07*  RV  0.03  0.09*  0.04*  0.07*  0.06*  0.08*  0.05*  0.12*  0.17*  0.09*  0.11*  0.03*  0.11*  0.03*  RO  0.12*  0.10*  0.09*  0.05*  0.07*  0.07*  0.07*  0.12*  0.21*  0.10*  0.17*  0.09*  0.14*  0.07*  0.10*  RU  0.13*  0.18*  0.13*  0.12*  0.10*  0.13*  0.16*  0.15*  0.04*  0.14*  0.20*  0.12*  0.12*  0.10*  0.12*  0.16*  SI  0.15*  0.16*  0.09*  0.16*  0.15*  0.15*  0.18*  0.18*  0.29*  0.15*  0.17*  0.12*  0.22*  0.14*  0.12*  0.19*  0.24*  SC  0.04*  0.09*  0.03*  0.09*  0.07*  0.08*  0.08*  0.13*  0.17*  0.09*  0.13*  0.03*  0.09*  0.02*  0.03*  0.10*  0.12*  0.12*  SH  0.08*  0.10*  0.05*  0.10*  0.09*  0.08*  0.12*  0.11*  0.22*  0.11*  0.13*  0.06*  0.11*  0.03*  0.07*  0.11*  0.16*  0.09*  0.06*  SK  0.08*  0.16*  0.08*  0.12*  0.10*  0.11*  0.17*  0.14*  0.17*  0.15*  0.15*  0.09*  0.09*  0.07*  0.09*  0.16*  0.10*  0.21*  0.10*  0.11*  SW  0.10*  0.12  0.10*  0.15*  0.12*  0.13*  0.20  0.20*  0.34*  0.16*  0.22*  0.14*  0.20  0.12*  0.11*  0.14*  0.26*  0.23*  0.12*  0.16  0.17*  TE  0.18*  0.22*  0.15*  0.18*  0.18*  0.19*  0.26*  0.22*  0.19*  0.21*  0.23*  0.12*  0.15*  0.12*  0.18*  0.19*  0.11*  0.28*  0.17*  0.17*  0.15*  0.32*  TI  0.12*  0.13*  0.07*  0.06*  0.00  0.01*  0.07*  0.03*  0.22*  0.04*  0.16*  0.09*  0.15*  0.11*  0.10*  0.10*  0.13*  0.16*  0.10*  0.11*  0.12*  0.16*  0.20*  WE  0.15*  0.22*  0.14*  0.18*  0.17*  0.18*  0.25*  0.20*  0.22*  0.19*  0.25*  0.11*  0.15*  0.17*  0.16*  0.22*  0.13*  0.27*  0.12*  0.20*  0.16*  0.25*  0.17*  TI  0.18*  * indicates values that are statistically significant after correction for false discovery rate, Pcitical < 0.049.  30  Figure 2.3 Relationship between the pairwise genetic distance (θ) and geographic distance between sites showing a weak (r2= 0.194) but significant (P<0.001) trend of isolation by distance.  Figure 2.4 Genetic friction map computed for BC western painted turtles The black outline delineates the sampled area, diamonds are the locations of the sampling sites. Degree of genetic friction indicated by inset legend 31  Migration rates, calculated as the proportion of migrants per generation, ranged from 0.01 to 0.15, with a mean of 0.01 (Table B.4). The approach implemented in BAYESASS has been deemed reliable for estimating migration rates in situations where FST > 0.05 and migration rates are less than m = 0.1 (Faubet et al. 2007). Both of these assumptions hold for the vast majority of population pairs in this study (Table 2.2, Table B.4).  STRUCTURE analyses revealed three genetic clusters in BC, with the distribution of the ad hoc statistic ΔK modal at K=3 (ΔK = 433.26; Figure 2.4a). The three clusters were Sunshine Coast-Gulf Island-Vancouver Island, South Coast-Thompson-Okanagan-Cariboo and the Kootenay (Figure 2.4a). Although three clusters were identified as the most likely, the Sunshine Coast-Gulf Island-Vancouver Island and South Coast-Thompson-OkanaganCariboo clusters exhibited significant heterozygote deficit potentially due to unrecognized population subdivision (Wahlund 1928). STRUCTURE was run again separately for the clusters with heterozygote deficit using the admixture model, with the ad hoc statistic ΔK modal at K=3 (ΔK = 250.17; Figure 2.4b) for the Thompson-Okanagan, Cariboo and South coast cluster, and K=2 (ΔK = 227.37; Figure 2.4b) for the Sunshine Coast-Gulf IslandVancouver Island cluster.  32  Figure 2.5 STRUCTURE bar plots using a straight admixture model for a. all populations and b. reanalysis of substructure within two broader clusters Each colour represents an inferred genetic cluster. Each bar on the x-axis represents an individual, with the y-axis displaying the proportion of membership in each genetic cluster  33  Overall, the recovered clusters from the STRUCTURE analyses generally correspond to the following: Sunshine coast (KL, RU), Sunshine coast-Gulf Island-Mid-Vancouver Island (hence forth abbreviated to SGV; TE, WE, PA), South coast (BE, SW, NI), Cariboo (SI), Thompson-Okanagan (BA, BL, OF, RE, RV, SC, SH, SK) and Kootenay (ER, RO, CV, DO, EL, JO, LO, TI).  The series of AMOVA analyses revealed that among-population component of genetic variation was maximized by the six groups based on the geographically cohesive STRUCTURE clusters to a greater extent than by the 10 ecological drainage units, three faunal provinces or the current DUs (Table 2.3).  Table 2.3 Analysis of Molecular Variance (AMOVA) results showing the percentage of variation described by each hierarchical level for the following four grouping strategies: a) genetic clusters informed through the STRUCTURE analysis; b) current designatable units; c) faunal provinces; and d) ecological drainage units Among Sites within Groups  a. Genetic Clusters (n = 6)  Among Groups 7.37  5.74  Within Sites 86.89  b. Designatable Units (n = 2)  2.3  10.57  87.57  c. Faunal Provinces (n = 3)  3.63  9.97  86.39  d. Ecological Drainage Units (n = 10)  5.68  6.54  87.78  2.3.3  Demographic history  The estimated effective population size of each group ranged from 10.1 in the SGV to 109.7 in the Thompson-Okanagan (Table 2.4). The heterozygote excess and mode-shift tests were significant only in the Cariboo (Table 2.4). The M-ratios were significantly less than Mc for all values of θ (Table 2.4), indicating historical bottlenecks in all six groups. Historical  34  population expansion was only detected in the South coast group, and only by the k test (Table 2.4). The relatedness at the site level ranged from 0.02 to 0.46, with an average of 0.2 ± 0.02 SE (Table 2.1). FIS ranged from -0.22 to 0.13 with an average of 0.00 ± 0.02 SE..  35  Table 2.4 Demographic history and effective population sizes of genetic clusters identified through the STRUCTURE analysis  Sunshine Coast  SGV  South coast  Cariboo  Thompson-Okanagan  Kootenay  N  PA  Ne  Wilcoxon  Mode Shift  k test P-value  g test P-value  θ  M  MC  33  0  34.6 (18.4 - 96.5)  ns  Normal  0.46  0.85  0.03  0.58*  0.85  0.3  0.58*  0.82  0.9  0.58*  0.78  0.03  0.64*  0.85  0.3  0.64*  0.82  0.9  0.64*  0.78  0.03  0.59*  0.85  0.3  0.59*  0.82  0.9  0.59*  0.78  0.03  0.55*  0.85  0.3  0.55*  0.81  0.9  0.55*  0.77  0.03  0.77*  0.85  0.3  0.77*  0.82  0.9  0.77*  0.78  0.03  0.76*  0.85  0.3  0.76*  0.82  0.9  0.76*  0.78  57  49  24  167  145  1  10  0  4  11  10.1 (7.4 - 13.4)  11.3 (9.1 - 14.0)  10.8 (6.3 - 19.2)  109.7 (87.5 - 141.7)  67.5 (57.1 - 80.6)  ns  ns  HE*  ns  ns  Normal  Normal  Shifted  Normal  Normal  0.72  0.02*  0.89  0.89  0.72  0.93  0.83  1.05  0.51  0.67  N number of individuals, PA number of private alleles, NE effective population size with 95% confidence interval, ns not significant, HE heterozygote excess, M M-ratio, MC M-critical, * indicates values are those that are statistically significant at p = 0.05  36  2.4  2.4.1  Discussion  The extent and distribution of genetic variation  My analyses show that BC populations of western painted turtle contain unique genetic diversity compared to other populations in North America, as evidenced by my findings of novel CR haplotypes not previously detected within their North American range. Fossil evidence shows that Chrysemys was distributed as far north and west as Nebraska during the Miocene (Holman 1976; Holman and Sullivan 1981), but there is no reported fossil evidence in BC to suggest its presence in that area prior to the Wisconsinan glaciation. Consequently, Pleistocene epoch phenomena, such as the controversial Haida Gwaii glacial refugium hypothesized to span parts of Vancouver Island, likely did not play a role in the development of regional patterns of genetic variation in BC (Demboski et al. 1999; Byun et al. 1997; Byun et al. 1999). Rather, the star-like pattern of C. p. bellii haplotypes revealed in the network analysis (Figure 2.2) is consistent with an interpretation of post-glacial expansion to the north and west as suggested by Starkey et al (2003), continuing into BC.  Within BC, CR haplotype 1 is common across all regions, with three other haplotypes found in one (CR haplotype 4) or two sites (CR haplotypes 2, 3), respectively (Figure 2.2, Table B.1). This lack of regional structuring of mtDNA variation is in contrast to the signal detected from the microsatellite genotypic data. The six discrete groups delineated through the STRUCTURE analysis have distinct patterns of genetic variation, although in some instances the driving factor behind those regional patterns is unclear. Anthropogenic movement of turtles is common in modern times (E. Jensen, personal observation), and it is possible that translocation also occurred  37  historically. Such an occurrence may explain the genetic similarity among turtles from Texada Island and Port Alberni, and the lower mainland and southern Vancouver Island, which are geographically separated by deep ocean in both instances. However, there is no genetic evidence for a founder event at any of those sites based on heterozygote excess and mode-shift tests (data not shown), and natural colonization cannot be ruled out.  Pairwise comparisons found significant differences in microsatellite allele frequency distributions at the site level, with a weak trend of differentiation increasing with geographic distance between sites. However, the non-stationary genetic friction map (Figure 2.3) revealed a greater amount of genetic difference per unit geographic distance along the coast relative to other sampled areas. These trends were also apparent within my estimates of gene flow, with migration rates between most water bodies estimated at less than 0.1, which is the threshold for being considered demographically independent (Hastings 1993). The three comparisons that exhibited migration rates greater than 0.1 indicated unidirectional migration between populations that are still significantly differentiated (EL to JO, TI to JO, KL to RU), although θ values are low (Tables 2.2 and Table B.4).  Despite each site being genetically differentiated, the levels of within-site genetic diversity, as represented by He (ranging from 0.51 to 0.77), are comparable to the range reported in other species of freshwater turtle (Emydoidea blandingii 0.45 to 0.71, Mockford et al. 2007; Glyptemys muhlenbergii 0.57 to 0.70, Pittman et al. 2011). Although some of the sites have high mean pairwise relatedness (KL, NI, PA, RU, SI, SW, TE, WE), in most instances this is not accompanied by high levels of inbreeding, which was detected at only two sites (BL and TE;  38  Table 2.1). These results suggest that local population sizes are still sufficiently large to counteract the loss of variation due to drift.  2.4.2  Demographic history of BC western painted turtles  The range of the western painted turtle likely only extended into BC after ~12,000 years ago due to the Wisconsin glaciation, and potentially much later if modern expansion west of Chicago was not possible prior to 14,000 years ago, as suggested by Starkey et al. (2003). Here, we only detected a genetic signature of historical population expansion in one group (south coast, Table 2.4). Although this study has the appropriate sample size for k and g tests, it is recommended that at least 25 microsatellite loci be used, and that the mutation rates of loci are uniform. My panel of nine microsatellites included both di- and tetra-nucleotide repeat loci, which are known to have variable mutation rates (Chakraborty et al. 1997). Moreover, methods for detecting population expansion, including the k and g tests employed here as well as others (e.g. the imbalance index, Kimmel et al. 1998), have low statistical power when detecting events that occurred fewer than 1000 generations ago (King et al. 2000). The generation time of Chrysemys has not been calculated, but the IUCN listing suggests approximately 20 years (Van Dijk 2011). Given this estimate, post-glacial expansion occurred less than 700 generations ago and current methods are limited in their ability to detect that event. Consequently, the results of the k and g tests reported here should be interpreted with caution. Expansion into the north and western limit of the species’ range were likely accompanied by founder effects, detected in this study in all six groups as historical bottlenecks indicated by the M-ratio tests.  39  2.4.3  Conservation unit concepts  Overall, my results reflect the difficulties of designating units below the species level. The presence of novel mtDNA haplotypes suggest that BC contains unique genetic variation that may be important to conserve. However, it is not clear how this information can be used to prioritize populations within BC for conservation. For example, following the ESU definition based on reciprocal monophyly (Moritz 1994), all BC western painted turtles would be lumped into a single ESU. The major CR haplotype (number 2) is found across the province, including at sites containing other haplotypes (1, 3 and 4), thus not meeting the criterion of reciprocal monophyly. At a finer-level, defining a management unit as a group of populations that show significant divergence of allele frequencies at nuclear or mitochondrial loci (Moritz 1994) would result in each occupied site as a separate MU for western painted turtles (Table 2.2). From a management perspective, it is likely not practical to devise conservation strategies at the site-level on a province-wide scale.  The more flexible DU concept provides alternative units for this species. Based on the dual criteria of genetic distinctiveness (via STRUCTURE) and range disjunction, there are six discrete units of western painted turtle in BC: Sunshine coast, SGV, south coast, Cariboo, Thompson-Okanagan, and Kootenay. Assessing the significance of these units is more difficult, as the guidelines are largely value-based to allow policy makers the flexibility to decide which kind of intraspecific variation to recognize. The wording of the criteria leaves them poorly defined and without quantifiable measures. For example, one criterion for recognition is “loss of the unit would result in an extensive gap of the range of the species in Canada” (COSEWIC 2008). If the proposed Thompson-Okanagan unit of western painted turtles were to disappear,  40  the Kootenay and Cariboo units would be even more disjunct from one another, as would the Sunshine coast and Thompson-Okanagan if the south coast unit were to disappear. As there is little guidance on the scale required for an “extensive gap”, the significance of each unit based on this criterion is ambiguous. Similarly, although the Cariboo unit is potentially the most northern locality for this wide-ranging species, it is unclear whether this circumstance constitutes “a unique ecological setting”, another significance criterion by which DUs may be recognized.  Criteria for defining ESUs are by no means universally agreed upon, with some authors arguing that no single concept will be appropriate for all taxa (Fraser and Bernatchez 2001). Because turtle mitochondrial DNA evolves eight times slower than that of other vertebrates (Avise et al. 1992), Moritz’s condition of reciprocal monophyly of mitochondrial DNA haplotypes to define evolutionarily distinct lineages may not be suitable. Turtle populations may evolve along independent trajectories for hundreds of generations, spanning many thousand years before this criterion is met, long after demographic independence and separate management for conservation purposes is warranted. Moritz (2002) acknowledges that the evolutionary processes within a taxon should be considered and strategies based on relevant criteria should be applied, however, no guidance is given on how to incorporate this within the ESU framework.  The guidelines for defining DUs as implemented by COSEWIC are perhaps more robust to considering various organisms with widely different evolutionary histories and ecologies. The analyses here show that dividing the range of the western painted turtle in BC using the generic faunal provinces as the basis for delineation of the Pacific Coastal and Intermountain-Rocky Mountain DUs does a poor job of describing genetic variation, with almost four times as much  41  variation partitioned within units than between them (Table 2.3). In contrast, the ecological drainage units, which are based on the interaction between the distribution of freshwater fish species, and the physical characteristics and environmental processes that characterize the ecosystem (Ciruna et al. 2007), did represent the distribution of genetic variation within the province reasonably well. However, this may not be the case for other organisms and using genetic methods can be a robust and effective way to determine units that are both biologically meaningful and practical in scale for conservation purposes. Nevertheless, if a species specific assessment is not possible, the ecological drainage units may represent a better biogeographic grouping strategy than the faunal provinces for aquatic organisms in BC.  As the ultimate goal of identifying conservation units is to preserve genetic variation to allow future adaptation, some authors have argued that adaptive divergence should be used to inform such designations (e.g. Kohn et al. 2006; Ouborg et al. 2009; Reed and Frankham 2001; Sgro et al. 2011; Bonin et al. 2007; Funk et al. 2012). Despite this increasing awareness and the technological advances that allow for the detection and assaying of adaptive genetic variation within natural populations, there are still no generally accepted methods for incorporating this information into conservation prioritization (but see Bonin et al. 2007; Funk et al. 2012). Although beyond the scope of the current study, future research targeting adaptive genetic variation could help resolve uncertainties about the significance of the six discrete units proposed here for western painted turtles, especially given the large ecological variation observed across BC.  42  2.4.4  Management implications  My results demonstrate that there are six discrete units of western painted turtles in BC, corresponding to the Sunshine coast, SGV, south coast, Cariboo, Thompson-Okanagan and Kootenay (Figure 2.1). I suggest that these six units require further consideration to determine whether they meet the significance criteria to become DUs. Three of these units (SGV, south coast and Cariboo) are characterized by very small effective population sizes Ne (<15), evidence of significant historical bottlenecks, and embedded sites exhibiting signatures of recent bottlenecks (data not shown) indicating that these units warrant independent management. Across the province, each occupied water body is demographically independent and qualifies as a MU under Moritz (1994), suggesting that demographic and population monitoring should occur at the site level, especially in the small, isolated populations in SGV and the Cariboo.  Overall, the situation of the western painted turtle in BC clearly illustrates the difficulties of designating units below the species level. It is imperative, however, that units are delineated that will recognize and protect the novel genetic diversity detected at their northern range margin in BC. The main threats to BC western painted turtles cited in the 2006 COSEWIC status assessment were road mortality and habitat loss. These pressures have not diminished since its publication. Although road density is much higher in the other units, in the Cariboo at least 12 adult individuals were killed by vehicles on a single 200 meter stretch of highway in the summer of 2012 alone (unpublished data). Moreover, high human population density and associated development in the south coast unit will continue to negatively impact the quality of remaining turtle habitat in that area. Since historical times, 85% of wetlands in the south Okanagan and the Fraser River delta have been converted into agricultural lands (Boyle et al. 1997; B.C. Ministry  43  of Environment Lands and Parks 1993) and the remaining remnants are still under threat. Quality habitat and viable population sizes are needed for the western painted turtle to endure in BC. Based on my findings, the conservation value of each of the six discrete units should be recognized and management plans designed to maximize population persistence throughout the province.  44  3 Chapter: INVESTIGATING TAXONOMIC UNCERTAINTY IN PAINTED TURTLES 3.1  Background  Species are the fundamental units of biodiversity, and are an important concept in how humans view nature (Richards 2010). Despite the importance of species, there is ongoing controversy over species concepts and definitions, which has major implications for the number and composition of described species, which in turn affects how we understand ecological and evolutionary processes. Determining where the so-called “species boundary” lies and which groups of organisms are above it (separate species) or below it (lineages within the same species) is particularly difficult in recently diverged lineages (Shaffer and Thomson 2007). There are both theoretical challenges associated with deciding how much divergence is necessary for taxonomic distinction to be appropriate and practical challenges for disentangling relationships using genetic or morphological evidence.  A commonality among many species concepts is that species should be separately evolving metapopulation lineages (De Queiroz 2007); if one accepts this, then the challenge is to determine how to identify the boundaries between such lineages. The methodologies of phylogenetics provide a whole toolbox for determining cladogenetic relationships; however, such methods can only be legitimately applied to organisms that are in fact related by nested, hierarchical relationships (Goldstein et al. 2000). If organisms are instead related through patterns of ancestry and descent (i.e. reticulating genealogical relationships), phylogenetic analysis can impose hierarchical structure, thus producing spurious results (Goldstein et al.  45  2000). In such cases, network-based analyses which do not impose the criterion of bifurcation can be a powerful tool for visualizing relationships (Posada and Crandall 2001).  Concomitant with the use of an appropriate analytical framework, the choice of molecular marker is critical for accurately reconstructing evolutionary history among recently diverged taxa. Both mitochondrial DNA (mtDNA) and nuclear DNA (nuDNA) can be used for reconstructing evolutionary history and each has specific characteristics that can be used to approach different points along a divergence continuum. MtDNA evolves more rapidly than nuDNA, given its lower effective population size (¼ of nuDNA) and resulting accelerated rate of lineage sorting (Funk and Omland 2003). Moreover, mtDNA is maternally inherited, precluding significant levels of recombination characteristic of nuDNA. However, these attractive properties that make mtDNA useful for reconstructing recent events also contribute to higher rates of homoplasy or convergent mutations that can potentially obscure the signal of more distant events (Zink and Barrowclough 2008). Because it is uniparentally inherited, mtDNA can only trace the lineage of one of the sexes, which may be a biased portrayal of history if, for example, there was sex biased dispersal. The differences between mtDNA and nuDNA can result in the reconstruction of evolutionary history that is discordant across markers.  Wide ranging species that occupy diverse ecosystems can also be a challenge for taxonomy, as they may contain several lineages at various stages of divergence from one another. Across the species’ range, populations can experience disparate selection pressures resulting in local adaptation (Gaston 2003). Peripheral populations may also experience limited gene flow with the core of the range, which can result in divergence and ultimately peripartic speciation (Lesica and  46  Allendorf 1995). Recent genetic studies of such taxa have resulted in various outcomes from the description of several species with limited distributions (e.g. Rodriguez-Robles and De JesusEscobar 2000) to the finding of continuously shared gene pools across the ranges of several subspecies (e.g Burbrink et al. 2000).  Chrysemys (painted turtles, Schneider 1783), a genus within the family Emydidae, is a wide ranging group experiencing ongoing taxonomic debate with regards to the status of the four formally described morphological types. Currently, the Society for the Study of Amphibians and Reptiles recognizes two species: Chrysemys dorsalis and Chrysemys picta, the latter of which is composed of three subspecies: C. p. picta, C. p. bellii and C. p. marginata (Crother 2012). These designations are largely based on a phylogenetic study that used partial sequences of the mitochondrial control region (CR) of individuals across the United States, and showed C. dorsalis as a divergent evolutionary lineage from the other three morphotypes (Starkey et al. 2003). Other taxonomists disagree with the elevation of dorsalis to the species level, and argue that it does not have either geographic or reproductive isolation from the other morphotypes within Chrysemys (Ernst and Lovich 2009). Interbreeding and hybridization are common wherever the ranges of the morphotypes overlap (Weller et al. 2010) to the extent that some speculate that a continuously shared gene pool may exist over the entire range of Chrysemys (Ernst and Lovich 2009). For this reason, many taxonomists only tentatively accept C. dorsalis as a species and often still refer to Chrysemys as a monotypic genus with four subspecies (Fritz and Havas 2007). However, Starkey et al. (2003) did not find picta, marginata and bellii to be separate evolutionary lineages, and suggested that their subspecific taxonomic descriptions be dropped. Although there are morphological differences between picta, marginata and bellii,  47  those differences are based on morphometric averages that do not provide clear distinctions and cannot be used to diagnose individuals (Ultsch et al. 2001).  Here, I supplement the study of Starkey et al. (2003) with overlapping mtDNA CR sequence data from the northern range margin of painted turtles in Canada, as well as new range-wide sequence data from the PAX-P1 nuclear intron to further investigate taxonomic uncertainty and palaeogeography within the group. Moreover, I conducted comparative network and phylogenetic analyses to examine these different approaches and to explore the implications for providing insights at the species boundary.  3.2  3.2.1  Methods  Sampling  Blood samples were collected from painted turtles across British Columbia, as described in Chapter 2. Previously collected painted turtle tissue samples from across the USA and Ontario were obtained from the archives of Dr. Bradley Shaffer (University of California, Los Angeles) and the Queen’s University Biological Station, respectively. Due to the limitations of using archival samples, the number of samples of each morphotype available varied.  48  3.2.2  Data collection  A 665 base pair segment of the mitochondrial genome, including part of the control region (CR) was amplified as described in Chapter 2. A 644 base pair segment of the PAX-P1 intron was amplified as a single fragment using PAX.20F and PAX.21R primers (Kimball et al. 2009). PCR was carried out on an ABI Veriti thermal cycler in 25 µl reactions containing ~20-40 ng DNA, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 200 µM dNTPs, 7.5µg bovine serum albumin, 0.4µM each primer and 0.5 AmpliTaq Gold DNA polymerase (AB Biosystems). Cycling conditions consisted of 95˚C for 10 minutes, 35 cycles of 95 ˚C for 30 seconds, 55 ˚C for 1.15 minutes, 72 ˚C for 1.5 minutes, and a final extension at 72 ˚C for 10 minutes.  The sequencing reactions were performed using ABI BigDye v3.1 Terminator chemistry and sequences were run on an Applied Biosystems 3130XL DNA automated sequencer. Sequences were visualized and edited using Sequencher 5.0 (Gene Codes Corporation).  In addition to the CR and PAX-P1 sequence data collected here, exemplar sequences from across the range of the painted turtle were downloaded from GenBank and contributed by collaborators (see Appendix C, Table C.1). Outgroup taxa consisted of exemplar individuals of Platysternon megacephalum, Terrapene carolina, Emys marmorata, Pseudemys texana, Graptemys geographica and Trachemys scripta elegans (Table C.2). For all tree-based analyses Platysternon megacephalum was selected as the outgroup as it is sister to the family Emydidae (Guillon et al. 2012).  49  3.2.3  Population genetic analysis  The CR and PAX-P1 data sets were grouped into individuals from localities within the range of each of the four morphotypes; individuals potentially from hybrid zones were excluded. The resulting data set for the CR included 151 bellii, 49 marginata, 40 picta and 13 dorsalis sequences (Table C.1, Figure 3.1) and the data set for PAX-P1 included 77 bellii, 17 marginata, 9 picta and 4 dorsalis sequences (Table C.1, Figure 3.1).The sequences for CR and PAX-P1 were aligned separately in Geneious (Biomatters Ltd.) using MUSCLE (Edgar 2004) and default settings. A microsatellite and adjacent flanking region within the PAX-P1 contributed to alignment ambiguity and were subsequently removed.  Figure 3.1 Localities for Chrysemys individuals used in the population genetic analyses, including those based on both the mitochondrial control region (CR) and PAX-P1 intron (triangles), CR only (circles and triangles) and PAX-P1 intron only (triangles and squares). Polygon numbers indicate samples falling within the described distributions of bellii (1), dorsalis (2), marginata (3), and picta (4).  50  From the nuclear PAX-P1 sequences, haplotypes were reconstructed using the Bayesian inference method implemented in PHASE 2.1 (Stephens and Donnelly 2003; Stephens et al. 2001) using an input file generated by SeqPHASE (Flot 2010). PHASE was run five times using a run length of 100 iterations with a burn-in of 100, with a relaxed assumption of the stepwise mutation model. The haplotypes reconstructed from the run with the best average goodness-of-fit were used in further analyses.  Haplotypic and nucleotide diversity estimates for each morphotype were calculated from the CR and PAX-P1 haplotypes separately in ARLEQUIN (Excoffier et al. 2005). For each gene region, levels of genetic divergence among samples were calculated with the fixation index, FST, using the Tamura Nei model (Tamura and Nei 1993), as implemented in ARLEQUIN (Excoffier et al. 2005). Significance for the pairwise FST comparisons was assessed using 3000 permutations. Genetic structure among the four morphotypes was examined using analysis of molecular variance (AMOVA) using the Tamura Nei model and 3000 permutations (Tamura and Nei 1993) as implemented in ARLEQUIN (Excoffier et al. 2005). Sequence alignments for the CR and PAX-P1 were used to identify characters that were fixed within and differing among morphotypes using population aggregation analysis (Davis and Nixon 1992).  3.2.4  Network and phylogenetic analysis  Haplotype networks were constructed for the CR and PAX-P1 haplotype data sets using statistical parsimony as implemented in TCS (Clement et al. 2000). Best-fit models of nucleotide substitution were selected for each gene region using AIC criteria in MrModeltest2 (Nylander  51  2004). The HKY (Hasegawa et al. 1985) model with a proportion of the sites being invariable (I) with rate variation (G) was selected for both the CR and PAX-P1.  CR haplotype trees were reconstructed based on all available sequences including those from Starkey et al. (2003) as well as the new data collected here (total n=253). The resulting operational taxonomic units (OTUs) represent all haplotypes sampled within each morphotype (n=51). A gene tree was estimated using the Bayesian framework implemented in MrBayes (Ronquist and Huelsenbeck 2003) using the HKY+I+G model. MrBayes was run with four chains for 1,100,000 generations, sampling every 200 generations with P. megacephalum as the outgroup. The first 25% of the trees were discarded and a consensus tree with posterior probability support threshold of 25% was produced. A neighbour joining tree was created in PAUP* (Swofford 2003) using the HKY+I+G and 100 bootstrap replicates with P. megacephalum as the monophyletic outgroup.  Using a subset of individuals (n=76) that covered a representative sample of the range of each morphotype, a PAX-P1 gene tree was reconstructed using the Bayesian framework implemented in MrBayes and the same parameters as above. The CR and PAX-P1 sequences were concatenated for individuals with both sequences available (n=76) and a combined phylogeny was reconstructed in MrBayes using the HKY+I+G model and the same parameters as above.  52  3.3  3.3.1  Results  Within and among subspecies variation  A total of 47 CR haplotypes were recovered among the 253 individuals from across the range of the four morphotypes of Chrysemys. The number of CR haplotypes within a morphotype ranged from six (dorsalis) to 18 (marginata), with levels of haplotypic and nucleotide diversity ranging from 0.69 to 0.84 and 0.002 to 0.007, respectively (Table 3.1). Four CR haplotypes were shared among groups (two bellii and marginata, two marginata and picta), although no haplotypes were shared by more than two morphotypes. A total of 29 PAX-P1 haplotypes were recovered among the 107 Chrysemys individuals. The number of PAX-P1 haplotypes within a morphotype ranged from five (dorsalis) to 12 (bellii) (Table 3.1). Eight PAX-P1 haplotypes were shared among morphotypes. Haplotypic diversity ranged from 0.65 to 0.90 and nucleotide diversity ranged from 0.002 to 0.008 (Table 3.1).  Table 3.1 Diversity indices from the analysis of the CR and PAX-P1  N # of Haplotypes Haplotypic Diversity Nucleotide diversity bellii 151 17 0.69 ± 0.03 0.002 ± 0.001 CR marginata 49 19 0.84 ± 0.04 0.007 ± 0.004 picta 40 9 0.77 ± 0.05 0.004 ± 0.002 dorsalis 13 6 0.72 ± 0.13 0.002 ± 0.002 77 12 0.65 ± 0.04 0.004 ± 0.002 PAX-P1 bellii marginata 17 11 0.93 ± 0.02 0.007 ± 0.004 picta 9 9 0.90 ± 0.04 0.008 ± 0.005 dorsalis 4 5 0.72 ± 0.13 0.002 ± 0.002  53  Based on the CR, significantly greater amounts of genetic variation were structured among rather than within the samples of the four groups (p<0.0001, Table 3.2). A lower but still significant amount of variation was also detected among groups based on the PAX-P1 data (p<0.0001, Table 3.2). For both gene regions, when dorsalis was removed from the AMOVAs, the results were consistent (Table 3.2).  Table 3.2. Percentages of genetic variation in the CR and PAX-P1 intron that are partitioned among and within the morphotype groups determined through AMOVA  With all 4 CR dorsalis Removed PAX-P1 With all 4 dorsalis Removed  Among Populations Within Populations Significance 69.69 30.31 0.0000 62.46 37.54 0.0000 27.15 72.85 0.0000 26.56 73.44 0.0000  The CR fixation indices were significant for all pairwise comparisons, with FST being greatest between bellii and dorsalis (0.90) and least between picta and marginata (0.34) (Table 3.3). Fixation indices based on the PAX-P1 haplotypes were again greatest between bellii and dorsalis (0.41) and least between picta and marginata (0.05) (Table 3.3); the only non-significant comparison was between picta and dorsalis (FST= 0.08, p=0.11).  Table 3.3 Pairwise comparisons of the fixation index, FST, based on the PAX-P1 intron above the diagonal and CR below the diagonal  bellii marginata picta dorsalis 0.3** 0.27** 0.41** bellii 0.05* 0.1* marginata 0.63** 0.73** 0.34** 0.08 picta 0.9** 0.71** 0.82** dorsalis * P<0.05, ** P<0.001  54  Across the 665 base pairs of CR sequence, five characters were fixed and different between dorsalis and the other three morphotypes; there are a total of five fixed differences between dorsalis and marginata, eight between dorsalis and bellii and ten between dorsalis and picta. Only one fixed difference in the CR was found between bellii and picta, and none between bellii and marginata, or marginata and picta. In the PAX-P1 intron, no fixed character differences were found among morphotypes.  3.3.2  Network and phylogenetic analysis  The haplotype network recreated based on the CR shows dorsalis and bellii to be relatively distinct from the other morphotypes, with marginata not appearing as a cohesive entity and largely mixed with picta (Figure 3.2). The main feature of the PAX-P1 network is overarching patterns of reticulation, with none of the morphotypes separating from the others (Figure 3.3).  The Bayesian gene tree for the CR (Figure 3.4A) shows Chrysemys to be monophyletic; dorsalis forms a monophyletic clade with high support (posterior probability 98), that is nested within the greater Chrysemys clade. The neighbour joining gene tree for the CR (Figure 3.4B) shows Chrysemys to be monophyletic, with dorsalis forming a basal, monophyletic clade similar to the pattern reconstructed in Starkey et al. (2003).  55  Figure 3.2 Haplotype network for the CR Ovals indicate haplotypes in the sample, single colour ovals are those found in only one morphotype, dual coloured nodes are those detected in more than one morphotype. Frequency of the shared haplotypes are not indicated. Inferred but unsampled haplotypes are indicated as links in the network.  56  Figure 3.3 Haplotype network for the PAX-P1 intron Ovals indicate haplotypes in the sample, single colour ovals are those found in only one morphotype, dual coloured nodes are those detected in more than one morphotype. Frequency of the shared haplotypes are not indicated. Inferred but unsampled haplotypes are indicated as links in the network.  The Bayesian gene tree analysis for PAX-P1 (Figure 3.4C) returned a poorly resolved tree in which Chrysemys is not monophyletic, with no support for all but the most fine-scale clustering. The Bayesian tree based on the concatenated PAX-P1 and CR sequences (Figure 3.4D) shows Chrysemys to be monophyletic with high support for a basal clade of western New York and Ohio marginata. The concatenated tree places a monophyletic dorsalis within the greater Chrysemys, with picta and marginata individuals as sister taxa with low support.  57  A.  B.  58  C.  D.  Figure 3.4 Phylogenetic trees for A. the Bayesian analysis of the CR, B. Neighbour Joining of the CR with bootstrap support indicated, C. the Bayesian analysis of the PAX-P1 intron and D. the Bayesian analysis of the concatenated sequences Bayesian trees show clades with posterior probabilities greater than 25% but only values greater than 50% are indicated. Blue square is bellii, red diamond is dorsalis, green circle is marginata and purple triangle is picta.  59  3.4  3.4.1  Discussion  Uncertain taxonomy  Starkey et al. (2003) tentatively proposed that Chrysemys should be split into two species, Chrysemys picta and Chrysemys dorsalis, pending future analysis of nuclear genes. Here, I present data from a nuclear gene, as well as increased geographic sampling expanding into Canada, with the goal of shedding further light on the taxonomy of this group. The resolution of Chrysemys’ taxonomy is complicated by several factors. There are four morphotypes that have been described, however, even within the core of their ranges, the morphotypes are poorly differentiated based on morphometric characters (Ultsch et al. 2001). The ranges of the morphotypes naturally overlap to form hybrid zones (Mann 2007; Ultsch et al. 2001; Weller et al. 2010). Although there is environmental variation across the range of Chrysemys, each morphotype spans a large distance north-south, and clinal adaptations to environmental variation would be captured within each morphotype, likely not partitioned among them. The exception to this pattern is dorsalis, which has a distribution that is limited to the extreme southeast of the range of Chrysemys. Moreover, dorsalis is known to have a much shorter period of tolerance for anoxia than the other morphotypes (Ultsch et al. 1999; Ultsch et al. 1985), hinting at adaptation to long winters in picta, bellii and marginata, the three more northern morphotypes.  It is often lamented that the application of various species concepts results in the recognition of vastly different numbers and identities of taxa (Agapow et al. 2004; Isaac et al. 2004). The genus Chrysemys appears to be no different and the difficulties may indeed be exacerbated  60  by the above mentioned complications. Adherence to the biological species concept (BSC), which defines species boundaries based on the criteria of reproductive isolation (Mayr 1969) would support the continued description of Chrysemys as a monotypic genus with geographic variation. The existing subspecies (C. p. picta, C. p. dorsalis, C. p. bellii, and C. p. marginata) were described based on this geographic variation in morphology and do not have reproductive isolation. The taxonomic unit of subspecies can be controversial as there are no rigorous criteria for designating them, and so many taxonomists do not support the recognition of those units (Zink 2004).  The phylogenetic species concept (PSC, Cracraft 1983) endeavours to define the barrier between the reticulate relationships that occur within a population and the hierarchical relationships among divergent lineages; individuals that are related through genealogical patterns of ancestry and descent, as opposed to nested and divergent lineages are part of the same phylogenetic species. Operationally, delimitation is done based on the criteria of fixed differences between phylogenetic species that can be used for diagnosis, often identified through population aggregation analysis (Davis and Nixon 1992). Adherence to the PSC would support the description of C. dorsalis and C. picta (encompassing picta, bellii and marginata) based on the five fixed mtDNA CR characters that can be used to diagnose the membership of an individual to one of the phylogenetic species. A common critique of the PSC is that fixed differences can quickly evolve, especially when population sizes are small, resulting in too many entities being described (Agapow et al. 2004; Frankham et al. 2012). Additionally, having sufficient sampling that is representative of the whole range of a species is important to determine which characters are truly fixed and different. In this study I  61  intentionally included only individuals that were clearly assigned to a morphotype, avoiding geographically intermediate sampling locations. The inclusion of individuals from other areas, particularly in the regions where the ranges of the morphotypes overlap, could falsify the diagnostic ability of the characters reported here.  The genealogical species concept (GSC) defines species on the basis of forming monophyletic entities in cladistics analysis (Baum and Shaw 1995). The GSC is used by Starkey et al. (2003) when interpreting the two monophyletic entities reconstructed in their tree-based analyses, thus resulting in the elevation of dorsalis to the species level as sister to C. picta (encompassing picta, bellii and marginata subspecies). However, in the tree based analyses in this study (Figure 3.4), dorsalis does not consistently form a monophyletic group, sister to the rest of Chrysemys, making the outcome of adherence to the GSC unclear (see below).  Although the morphotypes picta, marginata and bellii are not diagnosable by either fixed genetic characters or monophyletic entities, there are trait frequency differences that differentiate them. This finding is contrary to the suggestion by Starkey et al. (2003) that a continuously shared gene pool exists within what they call C. picta. There are large and significant FST values between each of the morphotypes, and the AMOVA shows that there is more genetic variation among than within the morphotypes, even when the potentially more diverged dorsalis is excluded from the analysis. This result suggests that each morphotype may represent a local genealogical unit, in which most genetic recombination occurs. Neither the BSC or GSC lends such units formal distinction, however, Davis and Nixon (1992) do  62  support the acknowledgement of such local genealogical units under the PSC (but see Cracraft 1983; Eldredge and Cracraft 1980; Nelson and Platnick 1981; Nixon and Wheeler 1990).  3.4.2  Discordance among results  The addition of a nuclear gene does not appear to be of much assistance in resolving Chrysemys taxonomy. In contrast to the CR, there are no fixed differences in the PAX-P1 intron that can be used to diagnose any of the four morphotypes and Bayesian inference methods returned a poorly resolved gene tree. Whether this lack of informativeness is due to an absence of phylogenetic signal in PAX-P1 or discordant patterns between mtDNA and nuDNA is unclear. Other studies using nuclear markers to resolve relationships within the Emydidae found many nuclear genes show little informative variation in this group (Wiens et al. 2010). Alternatively, there is a large body of literature showing that mtDNA and nuDNA frequently return discordant patterns with various potential explanations (Fisher-Reid and Wiens 2011; Toews and Brelsford 2012). For this reason, many authors recommend using several nuclear genes as well as mtDNA in phylogenetic studies (Zink and Barrowclough 2008). My study was limited to the use of a single nuclear region because preliminary evaluations of other nuclear regions (c-mos, R-35 and GAPD, data not shown) did not find them to be promising candidates for being phylogenetically informative. I initially targeted the PAX-P1 intron as it is highly informative at or around the species boundary in other chelonian taxa (Spinks et al. 2012a; Spinks et al. 2012b, Garrick et al. submitted). In contrast to the relatively deeper signal typical of other nuclear introns, PAX-P1 has been shown to  63  exhibit levels of variation similar to the mtDNA CR, suggesting that it may complement the CR in reconstructing relatively recent population history.  There is also discordance in the relative relationships reconstructed by the various tree-based analyses (Figure 3.4). This result could be due to weak phylogenetic signal in the gene sequences or the inappropriate application of cladistic methods. As is cautioned by Goldstein et al. (2000), applying cladistic analysis when the underlying relationships among terminals are not-hierarchical may force a spurious hierarchical result. The various topologies produced by the different optimality criteria and highly mobile placement of dorsalis may be an indication that a hierarchical structure was improperly imposed upon the data set. Although there is little congruence across the tree-based analyses in this study regarding the placement of dorsalis, the results of the neighbour joining tree in this study (Figure 3.4 B) are consistent with that of Starkey et al. (2003), which position dorsalis as monophyletic and basal with respect to the rest of Chrysemys. However, the neighbour joining analysis method will always produce a fully resolved tree and may represent the worst case of inappropriately applying hierarchical methods. The Bayesian method allows for polytomies (unresolved branches) and only shows branching where there is sufficient support for that event. Thus, although the neighbour joining tree indicates the relative relationships among the terminals in more detail, those relationships may not have much support and are better displayed as polytomies, as in the Bayesian analysis, which does not force as many hierarchical nodes.  If relationships among the terminals in the analysis are non-hierarchical, then the network analyses should display the reticulating relationships more accurately, including allowing for  64  the presence of extant ancestral nodes, and the proper depictions of multifurcations and reticulation due to homoplasy (Posada and Crandall 2001). It is possible that all three of those circumstances occur within the CR data set, with the additional situation of potential recombination in PAX-P1, suggesting that the network-based analyses are more appropriate for reconstructing the evolutionary history of Chrysemys.  The CR network presented here (Figure 3.2) shows dorsalis to have six unique haplotypes that are separated from all other sampled haplotypes by at least eight character changes. Morphotype combinations picta-marginata, and marginata-bellii share haplotypes, with any haplotypes unique to a morphotype often interspersed in the network, particularly for marginata and picta. With additional sampling of individuals, it is possible that many of the haplotypes that are unique to a morphotype in this study would be found to be shared among morphotypes. The high degree of haplotype sharing and general lack of segregating, morphotype-specific haplotypes indicates that historical relationships among picta, bellii and marginata are more reticulating than hierarchical. The PAX-P1 network (Figure 3.3) is also highly reticulating, with no clear segregation by morphotype. In fact, reconstructed PAX-P1 haplotypes are shared among all combinations of morphotypes except dorsalis – marginata, further indicating reticulating rather than hierarchical relationships.  3.4.3  Implications for Chrysemys paleogeographic history and conservation  Taxonomy aside, there are several other interesting results from this study. Starkey et al. (2003) commented on how the recent postglacial expansion of bellii is reflected in the low genetic diversity detected within that subspecies. Here, I find bellii to have lower haplotypic 65  diversity than any of the other subspecies, particularly at the PAX-P1 intron, which supports some of the assertions of Starkey et al. (2003). They postulated that a possible period of extreme aridification in the Great Plains/Rocky Mountain region that was maximal 14000 years ago [proposed by Bartlein et al. (1998)] might have limited the range of Chrysemys to eastern North America. Due to the Wisconsinan glaciation, bellii could have arrived in the most north-western part of their range in BC after 12000 years ago, and potentially much later if expansion west of Chicago was not possible prior to 14000 years ago (Starkey et al. 2003). The relatively lower levels of genetic diversity detected in that part of the range are consistent with these paleogeographic reconstructions. Yet, in BC, there are unique CR (B.1 and B.2) and PAX-P1 haplotypes (B.4 and B.5), indicating that those peripheral populations may contain evolutionary novelty important for local adaptation.  Although Chrysemys picta and its currently recognized subspecies are listed globally as “least concern” by the IUCN (Van Dijk 2011), in BC, the western painted turtle is considered to be Endangered or of Special Concern in different areas (COSEWIC 2006). Controversy can arise when species are listed as of conservation concern when they are locally rare within a political jurisdiction, yet are globally common (Hunter and Hutchinson 1994). Evidence that BC populations have genetic diversity that is unique within the species suggests that conservation efforts in BC may be warranted (see Chapter 2).  66  3.4.4  Summary  It is clear that resolving the taxonomy of Chrysemys will depend upon which species concept is applied. No genetic evidence was found in the nuclear gene to support the elevation of dorsalis to the species level; however the number of fixed differences in the mitochondrial gene does indicate lineage separation. The mobile phylogenetic placement of dorsalis relative to the rest of Chrysemys is a source of uncertainty regarding the strength of evidence for dorsalis as a basal rather than derived lineage from Chrysemys. However, the additional data and analyses conducted here do not provide compelling evidence to refute the tentative designation of C. dorsalis and C. picta (encompassing picta, bellii and marginata) as separate species, as has been provisionally adopted by the Society for the Study of Amphibians and Reptiles (Crother 2012) and the Turtle Taxonomy Working Group (Rhodin et al. 2010). The existence of hybrid zones is problematic to the interpretation of the results of this study, as individuals from those regions were excluded from these analyses given lack of accompanying morphological identification to subspecies. Further studies including increased sampling of individuals from the entire range of Chrysemys and a larger sample size of dorsalis in particular, using a suite of nuclear markers will be necessary to shed further light on the taxonomy of this group. The recent publication of the genome of C. p. bellii (Abramyan et al. 2013) could help identify single nucleotide polymorphisms (SNP) and genes that may have neutral and adaptive differences to target in future studies.  67  4 Chapter: CONCLUSIONS 4.1  Research findings and significance  This is the first genetic assessment of Chrysemys picta bellii in British Columbia, and to my knowledge, the first detailed population genetic study for this group overall, representing a significant contribution towards informing the best management for this species-at-risk in BC. This research shows that BC western painted turtles contain both nuclear and mitochondrial genetic diversity that is unique within the species, despite BC constituting just a small portion of the species’ range. The insights gained by applying two different conservation unit criteria to western painted turtles in BC constitute a more thorough assessment of the genetic diversity within BC turtles than either concept alone. From the ESU approach, it is clear that BC turtles are closely linked evolutionarily to the western painted turtles in other parts of North America, which is consistent with their recent expansion into western North America and BC. The ESU method also indicates that processes at the site level are important to monitor to ensure population persistence, as each occupied site included in this study is differentiated and demographically independent from all others. The DU approach highlights the regional nature of genetic diversity patterns and forces reflection as to the ecological significance of differentiation at the regional level. Although the ESU and DU conservation unit concepts result in the recommendation of different units, they are complementary in the case of the western painted turtle in BC, indicating the existence of genetic diversity at the site, regional and range wide levels.  68  Anecdotal reports suggesting that western painted turtles may have been introduced to the coast from sites in the interior in the recent past (COSEWIC 2006) have been clouding conservation efforts in that region. I do not find genetic evidence to support such reports, and instead find that all of the coastal sites included in this study are part of genetic clusters that are unique to that area. The genetic clusters identified are geographically cohesive, despite it being difficult to infer the processes that would lead to the Sunshine Coast-Gulf IslandVancouver Island (SGV) grouping. The geographic cohesiveness of the clusters indicates that the presence of turtles in each region is due to natural processes, not recent human introductions. It is more difficult to determine whether the occupation of a particular site is natural or due to human movement of turtles within a region. There are some sites in the lower mainland that were excluded from the analyses in Chapter 2 due to very small population sizes (n=1-2). Preliminary results from assignment tests indicate that some of these turtles may have been translocated from the interior (data not shown), providing some evidence that sites occupied by only two or three turtles may be due to human activities. It is only because I now have a reference data set for genetic diversity across the regions of BC that such assignment tests are possible, which will become a valuable asset in the future to determine the origin of turtles if controversy over their source arises. In summary, there is no evidence to support the anecdotal reports that western painted turtles are not native to the coast; furthermore, although there are likely some individuals at sites in the lower mainland that originated from the interior, this does not appear to be a widespread occurrence.  Human-mediated introduction of non-native turtles is an additional concern for western painted turtle populations in urban settings. For example, there is ongoing monitoring of red  69  eared slider (Trachemys scripta elegans) populations, as they may pose a threat to the native western painted turtles through increased competition (Bunnell 2005). Here, I find that the introduction of painted turtles from other areas of North America to Burnaby Lake is also a serious threat to the native genetic diversity at that site since hybridization between native and non-native painted turtles has now been documented. It is unclear where the non-native painted turtles originated and how they came to be introduced at Burnaby Lake. For this reason monitoring for non-native painted turtles should be ongoing at other lakes in the lower mainland of conservation significance.  Despite carrying out the most thorough genetic investigation regarding Chrysemys taxonomy to date using both nuclear and mtDNA markers, I am unable to resolve the uncertainty regarding whether dorsalis should be considered a separate species from other painted turtles. The tentative conclusion of lineage separation between dorsalis and the rest of Chrysemys found by Starkey et al. (2003) based on mtDNA stands after the inclusion of samples in BC and Ontario in this study, although based on different lines of evidence. The tree-based analyses were inconclusive as to the placement of dorsalis, but the existence of five fixed character differences in the CR between dorsalis and the rest of Chrysemys serves as evidence of lineage separation. This result, however, should be interpreted with caution due to the possibility of overdiagnosis resulting from the lack of detailed morphological data associated with available range-wide samples. The nuclear intron, PAX-P1, does not show lineage separation among any of the morphotypes. Nevertheless, there is frequency based evidence that suggests that the morphotypes may exist as local genealogical units based on both the CR and PAX-P1. Whether the lineage separation in the CR between dorsalis and the  70  rest of Chrysemys is sufficient to warrant describing them as separate species remains unclear, as does whether the shallow, frequency based differentiation of the morphotypes picta, bellii and marginata should be recognized formally.  4.2  Limitation of this study and future work  A genetic assessment is only as good at the number and distribution of samples analyzed. Within BC, I sampled the vast majority of the known range of the western pained turtle, but there are a few key sites that would have contributed greatly to this study, if samples had been available. One such site is Naniamo, on Vancouver Island, which could help determine the boundary between the South Coast and the SGV genetic clusters. Samples from the lower Sunshine Coast could also help determine the boundary between the South Coast and Sunshine Coast genetic clusters on the main land. Moreover, there is very little known about whether occupied sites exist in the area between Kamloops and Williams Lake. Samples from any intervening populations, if present, would help determine the extent of the Cariboo genetic cluster. Although I do have samples from Burning Creek, near 150 Mile House, I was unable to incorporate them in the assessment in Chapter 2 due to poor sample quality (see Appendix B); however, further effort to genotype these individuals could be made.  For the samples from outside of BC, I must rely on second hand information from collaborators regarding the exact site of origin for each sample. Knowing the exact location for each sample is important, as it was the geographic location of sampling that was used in conjunction with a coarse range map to determine the morphotype/subspecies of each individual for use in the analyses in Chapter 3 and Appendix A. Because the ranges of the 71  morphotypes are not well documented, I was conservative when choosing individuals for the population aggregation analysis in Chapter 3 and only included individuals that were sampled well within the core of the range of the morphotype. Excluding individuals from near the overlapping parts of the morphotypes’ ranges does potentially overestimate the number of fixed differences between morphotypes. In order to include more individuals in those analyses, I would need detailed information about the morphology of each individual and whether the water body sampled in contains individuals from more than one morphotype. Including individuals from the entire range of Chrysemys and increased numbers of samples from each morphotype will be important for any future studies.  Equally important to the geographic sampling of a study such as this one is the sampling across the genome. The tests for linkage among microsatellite loci indicate that each of the nine loci are inherited separately and represent independent samples of the nuclear genome. Also, the microsatellite loci do not show patterns that deviate from neutral expectations, as would be caused by linkage to loci under natural selection. The inclusion of additional microsatellite loci would have enriched the analyses, however, nine loci is within the range typically used in studies such as this (Coleman et al. 2013; Echelle et al. 2009). The CR is a commonly used mitochondrial marker for turtles, which has been found to provide better phylogenetic resolution than other mitochondrial regions (Lamb et al. 1994). As mtDNA is inherited without recombination, phylogenetic patterns should be consistent across the entire mitochondrial genome and the use of multiple regions is not necessary provided that the region used is sufficiently variable. As discussed in Chapter 3, additional nuclear markers will be needed to further address the taxonomic uncertainty in Chrysemys. I initially screened  72  four nuclear introns for use in this study, however, RNA fingerprint protein 35 intron 1 (Fujita et al. 2004) and Glyceraldehyde-3-phosphate dehydrogenase (GAPD) XI intron (Friesen et al. 1997) were found to be less variable in a pilot sample than the c-mos oncogene and PAX-P1 introns. The c-mos oncogene was subsequently sequenced for a larger number of individuals for use in the Burnaby Lake study (Appendix A), but was found to have weak phylogenetic signal (data not shown). Although the phylogenetic signal also appears to be weak in PAX-P1, that marker has been successfully used in other chelonian taxa (Spinks et al. 2012a; Spinks et al. 2012b, Garrick et al. submitted), and was a reasonable candidate for use in this study. In depth screening of additional informative nuclear regions is warranted and required for future studies to be successful in resolving the taxonomy and relationships within Chrysemys.  The recent publication of the genome of a C. p. bellii individual (Abramyan et al. 2013) could allow the development of molecular tools to investigate adaptive differences among painted turtles. Genetic markers that are linked to adaptive differences can be used both to clarify taxonomy and delineate and prioritize conservation units (Bonin et al. 2007; Funk et al. 2012; Ouborg et al. 2009; Reed and Frankham 2001). Information regarding the potential adaptive differences among the six genetic clusters identified in Chapter 2 could resolve the uncertainty about whether the clusters meet the significance criteria of the DU concept. Similarly, information regarding adaptive differences among the four morphotypes of Chrysemys could be used to clarify whether any of the morphotypes are sufficiently differentiated to merit recognition as a separate evolutionary entity.  73  References Abramyan J, Badenhorst D, Biggar KK, Borchert GM, Botka CW, Bowden RM, Braun EL, Bronikowski AM, Bruneau BG, Buck LT, Capel B, Castoe TA, Czerwinski M, Delehaunty KD, Edwards SV, Fronick CC, Fujita MK, Fulton L, Graves TA, Green RE, Haerty W, Hariharan R, Hillier LH, Holloway AK, Janes D, Janzen FJ, Kandoth C, Kong L, de Koning J, Li Y, Literman R, Mardis ER, McGaugh SE, Minx P, Mork L, M OL, Paitz RT, Pollock DD, Ponting CP, Radhakrishnan S, Raney BJ, Richman JM, St John J, Schwartz T, Sethuraman A, Shaffer B, Shedlock AM, Spinks PQ, Storey KB, Thane N, Thomson RC, Valenzuela N, Vinar T, Warren DE, Warren WC, Wilson RK, Zimmerman LM, Hernandez O, Amemiya CT (2013) The western painted turtle genome, a model for the evolution of extreme physiological adaptations in a slowly evolving lineage. Genome Biol 14 (3):R28 Agapow PM, Bininda-Emonds OR, Crandall KA, Gittleman JL, Mace GM, Marshall JC, Purvis A (2004) The impact of species concept on biodiversity studies. Q Rev Biol 79:161-179 Alacs EA, Janzen FJ, Scribner KT (2007) Genetic issues in freshwater turtle and tortoise conservation. Chelonian Res Monogr 4:107-123 Allendorf FW, Hohenlohe PA, Luikart G (2010) Genomics and the future of conservation genetics. Nat Rev Genet 11 (10):697-709 Allendorf FW, Leary RF, Spruell P, Wenburg JK (2001) The problems with hybrids: Setting conservation guidelines. Trends Ecol Evol 16 (11):613-622 Antao T, Lopes A, Lopes RJ, Beja-Pereira A, Luikart G (2008) LOSITAN: A workbench to detect molecular adaptation based on a F-st-outlier method. BMC Bioinformatics 9:323 Avise JC (2005) Phylogenetic units and currencies above and below the species level. In: Purvis A, Gittleman J, Brooks T (eds) Phylogeny and Conservation. Cambridge University Press, Cambridge, UK, pp 76-101 Avise JC, Ball RM (1990) Principles of genealogical concordance in species concepts and biological taxonomy. In: Futuyma D, Antonovics J (eds) Oxford Surveys in Evolutionary Biology vol 7. Oxford University Press, New York, pp 45-67 Avise JC, Bowen BW, Lamb T, Meylan AB, Bermingham E (1992) Mitochondrial DNA evolution at a turtle's pace: Evidence for low genetic variability and reduced microevolutionary rate in the Testudines. Mol Biol Evol 9 (3):457-473  74  B.C. Ministry of Environment Lands and Parks (1993) State of the environment report for British Columbia. B.C. Ministry of Environment, Lands and Parks, Victoria, BC and Environment Canada, Communications, Pacific and Yukon Region, North Vancouver, BC. 127pp. Available at: www.env.gov.bc.ca/soe/ Barley AJ, Spinks PQ, Thomson RC, Shaffer HB (2010) Fourteen nuclear genes provide phylogenetic resolution for difficult nodes in the turtle tree of life. Mol Phylogenet Evol 55 (3):1189-1194 Bartlein PJ, Anderson KH, Anderson PM, Edwards ME, Mock CJ, Thompson RS, Webb RS, Webb Iii T, Whitlock C (1998) Paleoclimate simulations for North America over the past 21,000 years: features of the simulated climate and comparisons with paleoenvironmental data. Q Sci Rev 17 (6–7):549-585 Baum DA, Shaw KL (1995) Genealogical perspectives on the species problem. In: Hoch PC, Stephenson AG (eds) Experimental and molecular approaches to plant biosystematics, vol 53. Monographs in Systematic Botany. Missouri Botanical Garden Press, St. Louis (MO), pp 289-303 Belkhir K, Borsa P, Chikhi L, Raufaste N, Bonhomme F (2004) GENETIX 4.05, logiciel sous Windows TM pour la génétique des populations. Laboratoire Génome, Populations, Interactions, CNRS UMR 5000, Université de Montpellier II, Montpellier (France) Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: A practical and powerful approach to multiple testing. J Roy Statist Soc Ser B 57 (1):289-300 Biedrzycka A, Solarz W, Okarma H (2012) Hybridization between native and introduced species of deer in Eastern Europe. J Mamm 93 (5):1331-1341 Bilgin R (2007) Kgtests: A simple Excel Macro program to detect signatures of population expansion using microsatellites. Mol Ecol Note 7 (3):416-417 Bonin A, Nicole F, Pompanon F, Miaud C, Taberlet P (2007) Population adaptive index: A new method to help measure intraspecific genetic diversity and prioritize populations for conservation. Conserv Biol 21 (3):697-708 Boyle CA, Lavkulich L, Schreier H, Kiss E (1997) Changes in land cover and subsequent effects on lower Fraser basin ecosystems from 1827 to 1990. J Environ Manag 21 (2):185-196 Bunnell CG (2005) Field survey of Red-eared Sliders (Trachemys scripta elegans) in the lower Fraser River Valley, British Columbia, in 2005. Wildlife Afield 2:51-57  75  Burbrink FT, Lawson R, Slowinski JB (2000) Mitochondrial DNA phylogeography of the polytypic North American rat snake (Elaphe obsoleta): A critique of the subspecies concept. Evolution 54 (6):2107-2118 Byun AS, Koop B, Reimchen TE (1999) Coastal refugia and postglacial recolonization routes: A reply to Demboski, Stone, and Cook. Evolution 53 (6):2013-2015 Byun SA, Koop BF, Reimchen TE (1997) North American black bear mtDNA phylogeography: Implications for morphology and the Haida Gwaii glacial refugium controversy. Evolution 51 (5):1647-1653 Chakraborty R, Kimmel M, Stivers DN, Davison LJ, Deka R (1997) Relative mutation rates at di-, tri-, and tetranucleotide microsatellite loci. Proc Natl Acad Sci USA 94 (3):1041-1046 Channell R, Lomolino MV (2000) Trajectories to extinction: Spatial dynamics of the contraction of geographical ranges. J Biogeogr 27 (1):169-179 Ciruna KA, Butterfield B, McPhail JD, BC Ministry of Environment (2007) Ecological aquatic units of British Columbia. Nature Conservancy of Canada, Toronto, Ontario Clement M, Posada D, Crandall KA (2000) TCS: A computer program to estimate gene genealogies. Mol Ecol 9 (10):1657-1659 Coleman RA, Weeks AR, Hoffmann AA (2013) Balancing genetic uniqueness and genetic variation in determining conservation and translocation strategies: a comprehensive case study of threatened dwarf galaxias, Galaxiella pusilla (Mack) (Pisces: Galaxiidae). Mol Ecol 22 (7):1820-1835 Convention on Biological Diversity (1992). United Nations Environment Programme COSEWIC (2004a) Canadian Species at Risk, November 2004.Committee on the Status of Endangered Wildlife in Canada.:58pp. COSEWIC (2004b) Guidelines for recognizing Designatable Units. COSEWIC (2006) COSEWIC assessment and status report on the Western Painted Turtle Chrysemys picta bellii (Pacific Coast population, Intermountain-Rocky Mountain population and Prairie/Western Boreal - Canadian Shield population) in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa. vii + 40 pp. (www.sararegistry.gc.ca/status/status_e.cfm)  76  COSEWIC (2008) Guidelines for recognizing disignatable units below the species level (approved by COSEWIC November 2008). Available from http://www.cosewic.gc.ca/eng/sct2/sct2_5_e.cfm Cracraft J (1983) Species concepts and speciation analysis. In: Johnston RF (ed) Current Ornithology, vol 1. Current Ornithology. Plenum Press, New York (NY), pp 159-187 Crandall KA, Bininda-Emonds ORP, Mace GM, Wayne RK (2000) Considering evolutionary processes in conservation biology. Trends Ecol Evol 15:290-295 Crother BI (2012) Scientific and standard English names of amphibians and reptiles of North America north of Mexico, with comments regarding confidence in our understanding. Herpetological Circular 39 Davis JI, Nixon KC (1992) Populations, genetic variation, and the delimitation of phylogenetic species. Syst Biol 41 (4):421-435 de Guia APO, Saitoh T (2006) The gap between the concept and definitions in the Evolutionarily Significant Unit: the need to integrate neutral genetic variation and adaptive variation. Ecol Res 22 (4):604-612 De Queiroz K (2007) Species concepts and species delimitation. Syst Biol 56 (6):879-886 Demboski JR, Stone KD, Cook JA (1999) Further perspectives on the Haida Gwaii glacial refugium. Evolution 53 (6):2008-2012 Dizon AE, Lockyer C, Perrin WF, Demaster DP, Sisson J (1992) Rethinking the stock concept: A phylogeographic approach. Conserv Biol 6:24-36 Duforet-Frebourg N, Blum MGB (2012) Non-stationary patterns of isolation-by-distance: Inferring measures of genetic friction. arXiv:1209.5242 Earl DA, vonHoldt BM (2011) STRUCTURE HARVESTER: A website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv Genet Res 4 (2):359-361 Echelle AA, Hackler JC, Lack JB, Ballard SR, Roman J, Fox SF, Leslie DM, Den Bussche RA (2009) Conservation genetics of the alligator snapping turtle: cytonuclear evidence of range-wide bottleneck effects and unusually pronounced geographic structure. Conserv Genet 11 (4):1375-1387 Edgar RC (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res 32 (5):1792-1797 77  Eldredge N, Cracraft J (1980) Phylogenetic patterns and the evolutionary process: Methods and theory in comparative biology. Columbia University Press, New York (NY) Ernst CH, Lovich JE (2009) Turtles of the United States and Canada, Second Edition. Johns Hopkins University Press, Baltimore (MD) Ersts PJ (2012) Geographic Distance Matrix Generator(version 1.2.3). American Museum of Natural History, Center for Biodiversity and Conservation. Available from http://biodiversityinformatics.amnh.org/open_source/gdmg Accessed on 2012-10-17 Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software STRUCTURE: A simulation study. Mol Ecol 14 (8):2611-2620 Excoffier L, Laval G, Schneider S (2005) Arlequin ver. 3.0: An integrated software package for population genetics data analysis. Evol Bioinform Online 1:47-50 Faubet P, Waples RS, Gaggiotti OE (2007) Evaluating the performance of a multilocus Bayesian method for the estimation of migration rates. Mol Ecol 16 (6):1149-1166 Fisher-Reid MC, Wiens JJ (2011) What are the consequences of combining nuclear and mitochondrial data for phylogenetic analysis? Lessons from Plethodon salamanders and 13 other vertebrate clades. BMC Evol Biol 11:300 Flot JF (2010) seqphase: a web tool for interconverting phase input/output files and fasta sequence alignments. Mol Ecol Resour 10 (1):162-166 Frankel OH (1974) Genetic conservation - our evolutionary responsibility. Genetics 78 (1):53-65 Frankham R (2010) Challenges and opportunities of genetic approaches to biological conservation. Biol Conserv 143 (9):1919-1927 Frankham R, Ballou JD, Dudash MR, Eldridge MDB, Fenster CB, Lacy RC, Mendelson JR, Porton IJ, Ralls K, Ryder OA (2012) Implications of different species concepts for conserving biodiversity. Biol Conserv 153:25-31 Frankham R, Ballou JD, Eldridge MDB, Lacy RC, Ralls K, Dudash MR, Fenster CB (2011) Predicting the Probability of Outbreeding Depression. Conserv Biol 25 (3):465-475 Fraser DJ, Bernatchez L (2001) Adaptive evolutionary conservation: Towards a unified concept for defining conservation units. Mol Ecol 10:2741-2752  78  Friesen VL, Congdon BC, Walsh HE, Birt TP (1997) Intron variation in marbled murrelets detected using analyses of single stranded confomational polymorphisms. Mol Ecol 6:1047-1058 Fritz U, Havas P (2007) Checklist of chelonians of the world. Vert Zool 57 (2):149-368 Fujita MK, Engstrom TN, Starkey DE, Shaffer HB (2004) Turtle phylogeny: Insights from a novel nuclear intron. Mol Phylogenet Evol 31:305-318 Funk DJ, Omland KE (2003) Species-level paraphyly and polyphyly: Frequency, causes, and consequences, with insights from animal mitochondrial DNA. Annu Rev Ecol Evol Syst 34:397-423 Funk WC, McKay JK, Hohenlohe PA, Allendorf FW (2012) Harnessing genomics for delineating conservation units. Trends Ecol Evol 27 (9):489-496 Garza JC, Williamson EG (2001) Detection of reduction in population size using data from microsatellite loci. Mol Ecol 10:305-318 Gaston KJ (2003) The structure and dynamics of geographic ranges. Oxford University Press, Oxford Gibbon JW, Scott DE, Ryan TJ, Buhlmann KA, Tuberville TD, Metts BS, Greene JL, Mills T, Leiden Y, Poppy S, Winne CT (2000) The global decline of reptiles, déjà vu amphibians. BioScience 50 (8):653 Goldstein PZ, Desalle R, Amato G, Vogler AP (2000) Conservation genetics at the species boundary. Conserv Biol 14:120-131 Gonçalves da Silva AG, Williams KE, Kirk SL, Bishop CA, Hodges KE, Russello MA (2009) Isolation and characterization of microsatellite loci in two species-at-risk in British Columbia: Great Basin spadefoot (Spea intermontana) and Western painted turtle (Chrysemys picta bellii). Conserv Genet Res 2 (1):37-40 Gonzalez-Trujillo R, Rodriguez D, Gonzalez-Romero A, Forstner MRJ, Densmore LD, Reynoso VH (2012) Testing for hybridization and assessing genetic diversity in Morelet's crocodile (Crocodylus moreletii) populations from central Veracruz. Conserv Genet 13 (6):1677-1683 Goudet J (2001) FSTAT, a program to estimate and test gene diversities and fixation indices (version 2.9.3). Available from http://www.unil.ch/izea/softwares/fstat.html.  79  Griffith B, Scott JM, Carpenter JW, Reed C (1989) Translocation as a species conservation tool: Status and strategy. Science 245 (4917):477-480 Guillon J-M, Guery L, Hulin V, Girondot M (2012) A large phylogeny of turtles (Testudines) using molecular data. Contrib Zool 81 (3):147-158 Haig SM, Beever EA, Chambers SM, Draheim HM, Dugger BD, Dunham S, Elliott-Smith E, Fontaine JB, Kesler DC, Knaus BJ, Lopes IF, Loschl P, Mullins TD, Sheffield LM (2006) Taxonomic considerations in listing subspecies under the U.S. Endangered Species Act. Conserv Biol 20 (6):1584-1594 Hasegawa M, Kishino H, Yano TA (1985) Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. J Mol Ecol 22 (2):160-174 Hastings A (1993) Complex interactions between dispersal and dynamics - lessons from coupled logistic equations. Ecology 74 (5):1362-1372 Hauswaldt JS, Glenn TC (2003) Microsatellite DNA loci from the Diamondback terrapin (Malaclemys terrapin). Mol Ecol Note 3 (2):174-176 Holman JA (1976) The herpetofauna of the lower Valentine formation north-central Nebraska. Herpetologica 32:262-268 Holman JA, Sullivan RM (1981) A small herpetofauna from the type section of the Valentine formation (Miocene, Barstovian), Cherry-County, Nebraska. J Paleontol 55 (1):138144 Hunter ML, Hutchinson A (1994) The virtues and shortcomings of parochialism Conserving species that are locally rare, but globally common. Conserv Biol 8 (4):1163-1165 Huxel GR (1999) Rapid displacement of native species by invasive species: Effects of hybridization. Biol Conserv 89 (2):143-152 Isaac NJ, Mallet J, Mace GM (2004) Taxonomic inflation: Its influence on macroecology and conservation. Trends Ecol Evol 19 (9):464-469 Jensen JL, Bohonak AJ, Kelley ST (2005) Isolation by distance web service. BMC Genet 6:13 Karl SA, Bowen BW (1999) Evolutionary significant units versus geopolitical taxonomy: Molecular systematics of an endangered sea turtle (genus Chelonia). Conserv Biol 13:990-999 80  Kiester AR, Olson DH (2011) Prime time for turtle conservation. Hepr Rev 42 (2):198-204 Kimball RT, Braun EL, Barker FK, Bowie RC, Braun MJ, Chojnowski JL, Hackett SJ, Han KL, Harshman J, Heimer-Torres V, Holznagel W, Huddleston CJ, Marks BD, Miglia KJ, Moore WS, Reddy S, Sheldon FH, Smith JV, Witt CC, Yuri T (2009) A welltested set of primers to amplify regions spread across the avian genome. Mol Phylogenet Evol 50 (3):654-660 Kimmel M, Chakraborty R, King JP, Bamshad M, Watkins WS, Jorde LB (1998) Signatures of population expansion in microsatellite repeat data. Genetics 148 (4):1921-1930 King JP, Kimmel M, Chakraborty R (2000) A power analysis of microsatellite-based statistics for inferring past population growth. Mol Biol Evol 17 (12):1859-1868 King TL, Julian SE (2004) Conservation of microsatellite DNA flanking sequence across 13 Emydid genera assayed with novel bog turtle (Glyptemys muhlenbergii) loci. Conserv Genet 5:719-725 Kohn MH, Murphy WJ, Ostrander EA, Wayne RK (2006) Genomics and conservation genetics. Trends Ecol Evol 21 (11):629-637 Lamb T, Lydeard C, Walker RB, Gibbons JW (1994) Molecular Systematics of Map Turtles (Graptemys): A Comparison of Mitochondrial Restriction Site Versus Sequence Data. Syst Biol 43 (4):543-559 Lesica P, Allendorf F (1995) When are peripheral populations valuable for conservation? Conserv Biol 9 (4):753-760 Lomolino MV, Channell R (1995) Splendid isolation: Patterns of geographic range collapse in endangered mammals. J Mamm 76 (2):335-347 Longmire J, Maltbie M, Baker R (1997) Use of 'lysis buffer' in DNA isolation and its implication for museum collections. Occas Pap Mus Texas Tech Univ (163):1-3 Lucklow M (1995) Species concepts: Assumptions, methods and applications. Syst Bot 20 (589-605):589 Lynch M, Ritland K (1999) Estimation of pairwise relatedness with molecular markers. Genetics 152 (4):1753-1766 Mace GM, Gittleman JL, Purvis A (2003) Preserving the tree of life. Science 300 (5626):1707-1709  81  Mann MR (2007) A taxonomic study of the morphological variation and intergradation of Chrysemys picta (Schneider) (Emydidae, Testudines) in West Virginia. Marshall University, Huntington (WV) Matschiner M, Salzburger W (2009) TANDEM: Integrating automated allele binning into genetics and genomics workflows. Bioinforma 25 (15):1982-1983 Mayr E (1969) The biological meaning of species. Biol J Linn Soc 1 (3):311-320 McNeely JA, Miller KR, Reid WV, Mittermeier RA, Werner TB (1990) Conserving the world's biological diversity. IUCN, Gland, Mikulicek P, Horak A, Zavadil V, Kautman J, Pialek J (2012) Hybridization between three crested newt species (Triturus cristatus superspecies) in the Czech Republic and Slovakia: Comparison of nuclear markers and mitochondrial DNA. Folia Zool 61 (34):202-218 Miller HC, Allendorf F, Daugherty CH (2010) Genetic diversity and differentiation at MHC genes in island populations of tuatara (Sphenodon spp.). Mol Ecol 19 (18):3894-3908 Mockford S, Herman T, Snyder M, Wright J (2007) Conservation genetics of Blanding’s turtle and its application in the identification of evolutionarily significant units. Conserv Genet 8 (1):209-219 Moritz C (1994) Defining 'evolutionarily significant units' for conservation. Trends Ecol Evol 9:373-374 Moritz C (2002) Strategies to protect biological diversity and the evolutionary processes that sustain it. Syst Biol 51 (2):238-254 Narum SR (2006) Beyond Bonferroni: Less conservative analyses for conservation genetics. Conserv Genet 7 (5):783-787 Nelson G, Platnick N (1981) Systematics and biogeography: Cladistics and vicariance. Columbia University Press, New York (NY) Nixon KC, Wheeler QD (1990) An amplification of the phylogenetic species concept. Cladistics 6 (3):211-223 Nylander JAA (2004) MrModeltest v2. Program distributed by the author. Evolutionary Biology Centre, Uppsala University.  82  Ouborg JN, Angeloni F, Vergeer P (2009) An essay on the necessity and feasibility of conservation genomics. Conserv Genet 11 (2):643-653 Parham JF, Feldman CR, Boore JL (2006) The complete mitochondrial genome of the enigmatic bigheaded turtle (Platysternon): Description of unusual genomic features and the reconciliation of phylogenetic hypotheses based on mitochondrial and nuclear DNA. BMC Evol Biol 6:11 Peakall R, Smouse PE (2006) GenAlEx 6: Genetic analysis in Excel. Population genetic software for teaching and research. Mol Ecol Note 6:288-295 Pearse DE, Janzen FJ, Avise JC (2001) Genetic markers substantiate long-term storage and utilization of sperm by female painted turtles. Heredity 86 (3):378-384 Peterson BK, Weber JN, Kay EH, Fisher HS, Hoekstra HE (2012) Double digest RADseq: An inexpensive method for de novo SNP discovery and genotyping in model and non-model species. PloS one 7 (5):e37135 Piry S, Luikart G, Cornuet J-M (1999) BOTTLENECK: A computer program for detecting recent reductions in the effective population size using allele freqency data. J Hered 90 (4):502-503 Pittman SE, King TL, Faurby S, Dorcas ME (2011) Demographic and genetic status of an isolated population of bog turtles (Glyptemys muhlenbergii): Implications for managing small populations of long-lived animals. Conserv Genet 12 (6):1589-1601 Posada D, Crandall KA (2001) Intraspecific gene genealogies: Trees grafting into networks. Trends Ecol Evol 16 (1):37-45 Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155:945-959 Queller DC, Goodnight KF (1989) Estimating relatedness using genetic markers. Evolution 43 (2):258-275 Rambaut A, Drummond A (2007) Tracer v1.4, Available from http://beast.bio.ed.ac.uk/Tracer. Raymond M, Rousset F (1995) GENEPOP (version1.2): Population genetics software for exact tests and ecumenicism. J Hered 86:248-249 Reed DH, Frankham R (2001) How closely correlated are molecular and quantitative measures of genetic variation? A meta-analysis. Evolution (55):1095 83  Reich DE, Feldman MW, Goldstein DB (1999) Statistical properties of two tests that use multilocus data sets to detect population expansions. Mol Biol Evol 16 (4):453-466 Rhodin A, van Dijk PP, Iverson J, Shaffer B (2010) Turtles of the World, 2010 Update: Annotated Checklist of Taxonomy, Synonymy, Distribution, and Conservation Status.000.085-000.164 Rice WR (1989) Analyzing tables of statistical tests. Evolution 43 (1):223-225 Richards R (2010) The species problem: A philosophical analysis. University Press, Cambridge Rodriguez-Robles JA, De Jesus-Escobar JM (2000) Molecular systematics of new world gopher, bull, and pinesnakes (Pituophis : Colubridae), a transcontinental species complex. Mol Phylogenet Evol 14 (1):35-50 Ronquist F, Huelsenbeck JP (2003) MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinforma 19:1572-1574 Rousset F (2008) Genepop'007: A complete reimplementation of the Genepop software for Windows and Linux. Mol Ecol Resour 8:103-106 Ryder OA (1986) Species conservation and systematics: The dilemma of subspecies. Trends Ecol Evol 1:9-10 Rymer JM, Simberloff D (1996) Extinction by hybridization and introgression. Annual Review of Ecology, Evolution, and Systematics 27:83-19 Saint KM, Austin CC, Donnellan SC, Hurchinson MN (1998) C-mos, a nuclear marker useful for Squamate phylogenetic analysis. Mol Phylogenet Evol 10:259-263 Schuelke M (2000) An economic method for the fluorescent labeling of PCR fragments. Nature Biotechnol (18):233-234 Sgro CM, Lowe AJ, Hoffmann AA (2011) Building evolutionary resilience for conserving biodiversity under climate change. Evol Appl 4:326-337 Shaffer HB, Thomson RC (2007) Delimiting species in recent radiations. Syst Biol 56 (6):896-906 Sites JW, Marshall JC (2003) Delimiting species: A Renaissance issue in systematic biology. Trends Ecol Evol 18 (9):462-470  84  Spinks PQ, Shaffer HB (2005) Range-wide molecular analysis of the western pond turtle (Emys marmorata): Cryptic variation, isolation by distance, and their conservation implications. Mol Ecol 14 (7):2047-2064 Spinks PQ, Thomson RC, Hughes B, Moxley B, Brown R, Diesmos A, Shaffer HB (2012a) Cryptic variation and the tragedy of unrecognized taxa: the case of international trade in the spiny turtle Heosemys spinosa (Testudines: Geoemydidae). Zool J Linn SocLond 164 (4):811-824 Spinks PQ, Thomson RC, Pauly GB, Newman CE, Mount G, Shaffer HB (2013) Misleading phylogenetic inferences based on single-exemplar sampling in the turtle genus Pseudemys. Mol Phylogenet Evol (0) Spinks PQ, Thomson RC, Zhang Y, Che J, Wu Y, Bradley Shaffer H (2012b) Species boundaries and phylogenetic relationships in the critically endangered Asian box turtle genus Cuora. Mol Phylogenet Evol 63 (3):656-667 Starkey DE, Shaffer HB, Burke RL, Forstner MRJ, Iverson JB, Janzen FJ, Rhodin AGJ, Ultsch GR (2003) Molecular systematics, phylogeography, and the effects of pleistocene glaciation in the painted turtle (Chrysemys picta) complex. Evolution 57:119 Stephens M, Donnelly P (2003) A comparison of bayesian methods for haplotype reconstruction from population genotype data. Am J Hum Genet 73 (5):1162-1169 Stephens M, Smith NJ, Donnelly P (2001) A new statistical method for haplotype reconstruction from population data. Am J Hum Genet 68 (4):978-989 Stuart BL, Parham JF (2007) Recent hybrid origin of three rare Chinese turtles. Conserv Genet 8 (1):169-175 Swofford DL (2003) PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). vol Version 4. . Sinauer Associates, Sunderland (MA) Tamura K, Nei M (1993) Estimation of the number of nucleotide substitutions in the control region of mitochondrial-DNA in humans and chimpanzees. Mol Biol Evol 10 (3):512-526 Telemeco RS, Abbott KC, Janzen FJ (2013) Modeling the effects of climate change-induced shifts in reproductive phenology on temperature-dependent traits. Am Nat 181 (5):637-648  85  Templeton AR (1986) Coadaptation and outbreeding depression. In: ME S (ed) Conservation Biology: The Science of Scarcity and Diversity. Sinauer Associates, Sunderland, MA, pp 105-116 Toews DP, Brelsford A (2012) The biogeography of mitochondrial and nuclear discordance in animals. Mol Ecol 21 (16):3907-3930 Ultsch GR, Carwile ME, Crocker CE, Jackson DC (1999) The physiology of hibernation among painted turtles: The eastern painted turtle Chrysemys picta picta. Physiol Biochem Zool 72 (4):493-501 Ultsch GR, Hanley RW, Bauman TR (1985) Responses to anoxia during simulated hibernation in northern and southern painted turtles. Ecology 66 (2):388-395 Ultsch GR, Ward GM, LeBerte CM, Kuhajda BR, Stewart ER (2001) Intergradation and origins of subspecies of the turtle Chrysemys picta: Morphological comparisons. Can J Zool 79 (3):485-498 US Fish and Wildlife Service and National Oceanic and Atmospheric Administration (1996) Policy regarding the recognition of distinct vertebrate population segments under the Endangered Species Act. Federal Register 26 (61):4722–4725. Van Dijk PP (2011) Chrysemys picta In: IUCN 2012. IUCN Red List of Threatened Species. Version 2012.2. www.iucnredlist.org Downloaded on 15 March 2013. Van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P (2004) MICRO-CHECKER: Software for identifying and correcting genotyping errors in microsatellite data. Mol Ecol Note 4 (3):535-538 Vié JC, Hilton-Taylor C, Stuart SN (2009) Wildlife in a changing world – An analysis of the 2008 IUCN Red List of Threatened Species™. IUCN, Gland, Switzerland Vilaca ST, Vargas SM, Lara-Ruiz P, Molfetti E, Reis EC, Lobo-Hajdu G, Soares LS, Santos FR (2012) Nuclear markers reveal a complex introgression pattern among marine turtle species on the Brazilian coast. Mol Ecol 21 (17):4300-4312 Vogler AP, Desalle R (1994) Diagnosing units of conservation management. Conserv Biol 8:354-363 Wahlund S (1928) Composition of populations and correlation appearances viewed in relation to the studies of inheritance. Hereditas 11:65-106  86  Wang J (2011) COANCESTRY: A program for simulating, estimating and analysing relatedness and inbreeding coefficients. Mol Ecol Resour 11 (1):141-145 Waples RS (1991) Pacific salmon, Oncorhynchus spp., and the definition of “species” under the Endangered Species Act. Mar Fish Rev (53):11-22 Waples RS (2006) A bias correction for estimates of effective population size based on linkage disequilibrium at unlinked gene loci. Conserv Genet 7 (2):167-184 Waples RS, Do C (2008) LDNE: A program for estimating effective population size from data on linkage disequilibrium. Mol Ecol Resour 8 (4):753-756 Weir BS, Cockerham CC (1984) Estimating f-statistics for the analysis of populationstructure. Evolution 38 (6):1358-1370 Weller WF, Hecnar SJ, Hecnar DR, Casper GS, Dawson FN (2010) Quantitative assessment of intergradation between two subspecies of painted turtles, Chrysemys picta bellii and C. p. marginata, in the Algoma district of west central Ontario, Canada. Herp Conserv Biol 5:166-173 Wiens JJ, Kuczynski CA, Stephens PR (2010) Discordant mitochondrial and nuclear gene phylogenies in emydid turtles: Implications for speciation and conservation. Biol J Linn Soc 99 (2):445-461 Wilkins JS (2009) Species: The history of an idea. University of California Press, Berkeley, CA Wilson GA, Rannala B (2003) Bayesian inference of recent migration rates using multilocus genotypes. Genetics 163 (3):1177-1191 Wood CC, Gross MR (2008) Elemental conservation units: Communicating extinction risk without dictating targets for protection. Conserv Biol 22 (1):36-47 Zink RM (2004) The role of subspecies in obscuring avian biological diversity and misleading conservation policy. Proc R Soc Lond B Biol Sci 271 (1539):561-564 Zink RM, Barrowclough GF (2008) Mitochondrial DNA under siege in avian phylogeography. Mol Ecol 17 (9):2107-2121  87  Appendices Appendix A : Investigating the genetic consequences of introduced Chrysemys picta on endangered C. p. bellii in British Columbia  A.1 Background  The introduction and establishment of non-native individuals can impose severe impacts to native fauna, especially if interbreeding with the resident form occurs, resulting in hybridization and genetic introgression (Rymer and Simberloff 1996; Huxel 1999). A common source of non-native individuals is often the release of unwanted pets as in the case of turtles (Stuart and Parham 2007). Populations with genetic contamination through introgression may be compromised from a conservation standpoint, so the use of genetic methods to detect non-native and hybrid individuals can be important to inform management (Allendorf et al. 2001).  The western painted turtle (Chrysemys picta bellii) is listed as Endangered in the Pacific Coastal designatable unit in British Columbia (COSEWIC 2006). It is the only extant native pond turtle found in Western Canada and for that reason, holds special ecological and cultural significance (COSEWIC 2006). In the lower mainland of BC, Burnaby Lake (Figure A.1) is an important stronghold of the species and is one of only two sites in that region where recruitment has been documented (A. Mitchell, personal communication). In 2010, volunteers watching for nesting on a newly created artificial nesting beach observed six females nesting that displayed colour patterns on their plastrons that are atypical of the native  88  C. p. bellii (Figure A.2a and A.2b), and more similar to non-native conspecific subspecies found in other parts of the wide range of this species (A. Mitchell, personal communication).  Figure A.1 Map of Burnaby Lake. Oval encompasses the area at the east end of the lake where 23 turtles were sampled, including at the artificial nesting beach. Square encompasses the area at the west end of the lake where two turtles were sampled  Figure A.2 Morphology of native and non-native painted turtles. a. Turtle with non-native morphology sampled at Burnaby Lake, b. Example of a typical western painted turtle, c. Juvenile sampled at Burnaby Lake with western painted turtle morphology and a non-native control region haplotype  89  The distribution of C. picta is continuous across the North American continent, with geographic variation recognized in the form of four subspecies (C .p. picta, C. p. marginata, C. p. dorsalis, and C. p. bellii) with overlapping ranges (Ernst and Lovich 2009). Hybridization is common where the ranges overlap (Weller et al. 2010), which is cause for concern if non-native conspecifics are now present in Burnaby Lake. The introduction of alleles from elsewhere could be detrimental to the local populations if those alleles are maladaptive to the local conditions, resulting in a decrease in fitness due to outbreeding depression (Allendorf et al. 2001). Also, genetic contamination could affect the conservation value of introgressed populations (Allendorf et al. 2001).  Previous phylogeographic analyses did not find sufficient differentiation at the mitochondrial DNA (mtDNA) control region (CR) to warrant continued recognition of picta, marginata and bellii as separate evolutionary lineages (Starkey et al. (2003), however these range-wide haplotypes are informative for determining coarse geographic origin (see Chapter 3). However, mtDNA is uni-parentally inherited and thus limited to reconstructing matrilines. To complement insights from mtDNA, nuclear genotypic data have proved useful for investigating the presence and direction of hybridization, and degree of introgression in a wide range of species (Mikulicek et al. 2012; Gonzalez-Trujillo et al. 2012; Biedrzycka et al. 2012; Vilaca et al. 2012). Here, I used mitochondrial and nuclear DNA sequence data to investigate whether individuals with atypical C. p. bellii morphology in Burnaby Lake are non-native, and to reconstruct their likely origin. Further, I used nuclear microsatellite genotypic data to estimate relatedness among introduced individuals in order to infer whether their presence is due to release from a single or multiple sources.  90  A.2 Methods  Site and sampling  Burnaby Lake is a large post-glacial lake in an urban setting within the city of Burnaby in the lower mainland of BC (Figure A.1 and A.3). It is a popular lake for recreational activities and is easily accessible on all sides via a walking trail that circles its perimeter. Blood samples (~100µl) were collected from the dorsal coccygeal vein from 24 adults (plastron length >10 cm) and one juvenile (plastron length <7 cm) painted turtle at Burnaby Lake. Twelve samples were collected from females following nesting at the artificially created nesting beach at the east end of Burnaby Lake; the other 13 turtles were collected by dip netting (Figure A.1). Only two samples were collected from the west end of Burnaby Lake (Figure A.1). Blood was stored in tubes containing a buffer solution (100 mM Tris-HCl, pH 8.0; 100 mM EDTA, pH 8.0; 10 mM NaCl; 0.5% SDS; Longmire et al. 1997) and refrigerated until analysis. Of the adult turtles, 12 individuals qualitatively displayed typical C. p. bellii morphology whereas the other 12 displayed morphology which, to varying degrees, more closely resembled other subspecies of C. picta (Figure A.2, E. Jensen, personal observation). Samples were collected from turtles (n=26) from 12 other lakes in BC as part of a larger study assessing genetic diversity of C. p. bellii in BC (see Chapter 2) and were used in this study as references for native BC genotypes (Figure A.3, Table A.1). Tissue samples from across North America (n=46) were obtained from the archives of Dr. Bradley Shaffer (University of California, Los Angeles) and the Queen’s University Biological Station to be used as references for the other subspecies (Figure A.3, Table A.1).  91  Figure A.3 Map of the localities of Chrysemys individuals for which the c-mos oncogene was sequenced. The star indicates the location of Burnaby Lake. Blue circles are C. p. bellii, red squares are C. p. dorsalis, green diamonds are C. p. marginata and purple triangles are C. p. picta.  92  Table A.1 Source and locality information for  Source  Chrysemys samples for which c-mos was  Voucher ID RE.2  sequenced PS present study Voucher ID HBS27284  Source Shaffer  State/ Province AL  County /Region Marshall  HBS27288  Shaffer  AL  Marshall  Alak.5.12  PS  BC  Lower Mainland  Alak.51  PS  BC  Lower Mainland  Burn.21  PS  BC  Lower Mainland  Burn.212  PS  BC  Lower Mainland  Burn.213  PS  BC  Lower Mainland  Burn.214  PS  BC  Lower Mainland  Burn.2140  PS  BC  Lower Mainland  Burn.215  PS  BC  Lower Mainland  Burn.217  PS  BC  Lower Mainland  Burn.218  PS  BC  Lower Mainland  Burn.219  PS  BC  Lower Mainland  Burn.22  PS  BC  Lower Mainland  Burn.223  PS  BC  Lower Mainland  Burn.224  PS  BC  Lower Mainland  Burn.225  PS  BC  Lower Mainland  Burn.226  PS  BC  Lower Mainland  Burn.227  PS  BC  Lower Mainland  Burn.23  PS  BC  Lower Mainland  Burn.24  PS  BC  Lower Mainland  Burn.25  PS  BC  Lower Mainland  Burn.26  PS  BC  Lower Mainland  Burn.27  PS  BC  Lower Mainland  Burn.28  PS  BC  Lower Mainland  Burn.29  PS  BC  Lower Mainland  Burn.32  PS  BC  Lower Mainland  Burn.35  PS  BC  Lower Mainland  Burn.56  PS  BC  Lower Mainland  GBHNR.1  PS  BC  Lower Mainland  SL.33  PS  BC  Lower Mainland  SL.37  PS  BC  Lower Mainland  OF.26  PS  BC  Okanagan  OF.27  PS  BC  Okanagan  OF.29  PS  BC  Okanagan  OF.30  PS  BC  Okanagan  RE.1  PS  BC  Okanagan  PS  State/ Province BC  County /Region Okanagan  RE.3  PS  BC  Okanagan  RE.4  PS  BC  Okanagan  Cran.51  PS  BC  Sunshine Coast  WE.318  PS  BC  Sunshine Coast  WE.34  PS  BC  Sunshine Coast  TI.113  PS  BC  Texada Island  TI.23  PS  BC  Texada Island  RV.32  PS  BC  Thompson  RV.85  PS  BC  Thompson  RV.98  PS  BC  Thompson  BE.136  PS  BC  Vancouver Island  BE.137  PS  BC  Vancouver Island  BE.17  PS  BC  Vancouver Island  PA.24  PS  BC  Vancouver Island  PA.28  PS  BC  Vancouver Island  HBS27533  Shaffer  CO  Douglas  HBS28620  Shaffer  CO  LaPlata  HBS26213  Shaffer  GA  Jackson  HBS28303  Shaffer  ID  Boundary  HBS23169  Shaffer  IL  Alexander  HBS23170  Shaffer  IL  Alexander  HBS28134  Shaffer  KS  Sherman  HBS31533  Shaffer  LA  St Martin  HBS26058  Shaffer  MA  Middlesex  HBS33050  Shaffer  MA  Worcester  HBS28035  Shaffer  ME  Kennebec  HBS28041  Shaffer  ME  Kennebec  HBS23616  Shaffer  MN  Houston  HBS28277  Shaffer  MT  Lewis and Clark  HBS27748  Shaffer  ND  Rolette  HBS28049  Shaffer  NH  Salem  HBS28026  Shaffer  NH  Sullivan  HBS31658  Shaffer  NJ  Sussex  HBS28644  Shaffer  NM  Apache  HBS28625  Shaffer  NM  San Juan  HBS28001  Shaffer  NY  Clinton  HBS27581  Shaffer  NY  Sufflolk  HBS27179  Shaffer  OH  Shelby  93  Data collection and analysis  DNA was extracted from blood samples using the Nucleospin QuickBlood kit (MachereyNagel) following manufacturer’s protocols. DNA was extracted from tissue samples using the Nucleospin Tissue kit (Macherey-Nagel) following manufacturer’s protocols.  Mitochondrial sequences  A 671 base pair segment of the mitochondrial genome (mtDNA), including part of the control region (CR) was amplified as a single fragment using the methods described in Chapter 2. Exemplar CR sequences from across North America were taken from the Starkey et al. (2003) popset in GenBank (accession number JQ963656). All the CR sequences were aligned in Geneious (Biomatters Ltd.) using Geneious Aligner and a haplotype network was generated using statistical parsimony, as implemented in TCS (Clement et al. 2000).  Nuclear sequences  A 519 base pair segment of the nuclear oncogene c-mos (Saint et al. 1998) was sequenced for all Burnaby Lake individuals (n=25) and a subset of BC individuals (n=26) as well as for a representative sample (n=45) of individuals from across North America (Table A.1). The CMOSG77 primer from Saint et al. (1998) was used in conjunction with Cp_CMOSG78 (5AGGGTGATGTCAAAGGAGTAGATGTC-3´; this study) in 25 µl reactions containing ~20-  40 ng DNA, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 200 µM dNTPs, 7.5µg bovine serum albumin, 0.4µM each primer and 0.5 U KAPA Taq DNA Polymerase. Cycling  94  conditions consisted of 94˚C for 2 minutes, 35 cycles of 95 ˚C for 30 seconds, 55 ˚C for 30 seconds, 72 ˚C for 45 seconds, and a final extension at 72 ˚C for 7 minutes. The sequencing reactions were performed using ABI BigDye v3.1 Terminator chemistry and sequences were run on an Applied Biosystems 3130XL DNA automated sequencer. Sequences were visualized and edited using Sequencher 5.0 (Gene Codes Corporation). The c-mos sequences were aligned using Geneious Aligner and examined for polymorphisms.  Microsatellites  The Burnaby Lake individuals were genotyped at ten microsatellite loci (Table B.2) (Gonçalves da Silva et al. 2009; Hauswaldt and Glenn 2003; King and Julian 2004; Pearse et al. 2001) using the methods described in Chapter 2. A high level of missing data were found at locus GT124 (48%), which was subsequently excluded from further analysis. Locus Cp2 was also excluded for the reasons discussed in Chapter 2. The pairwise relatedness (Queller and Goodnight 1989) among Burnaby Lake individuals was estimated in GenAlEx (Peakall and Smouse 2006).  A.3 Results  Mitochondrial haplotypes  Four mtDNA CR haplotypes were detected across the representative sampling from BC, each different by only a single base pair change (Figure A.4). All individuals identified as nonnative based on morphology were found to have CR haplotypes common to eastern North  95  America and differing from BC native haplotypes by at least four character changes. Of those 12 individuals, 11 shared a single haplotype (haplotype S.3, Figure A.4). The other morphologically non-native individual had a different haplotype (haplotype S.4, Figure A.4). Haplotype S.4 is found in C. p. bellii individuals, however only those sampled at the easternmost portion of the range in Wisconsin, Illinois and Minnesota. The only juvenile turtle to be sampled at Burnaby Lake, which was morphologically identified as native (Figure A.2c), had the more common of the non-native CR haplotypes, S.3.  96  Figure A.4 Haplotype network reconstructed based on the mtDNA control region data Ovals indicate haplotypes in the sample. Single colour ovals are those found in only one subspecies, dual coloured nodes are those detected in more than one subspecies. Frequencies of the shared haplotypes are not indicated. Inferred but unsampled nodes are indicated as links in the network. Stars indicate haplotypes that were sampled in native BC individuals; arrows indicate haplotypes that were sampled in the morphologically non-native turtles in Burnaby Lake  97  Nuclear genotypes  Most of the non-native individuals (75%) had private alleles at two positions in the c-mos sequence, c-mos 205 and c-mos 436 (Table A.2) relative to other turtles in BC. Where all individuals from other sites in BC and all C. p. bellii from the rest of their range were monomorphic at these two positions, nine of the non-native individuals were polymorphic for both the native allele and a foreign one. Using the samples from across North America as a reference, the non-native alleles for those positions were found only in individuals belonging to C. p. dorsalis and C. p. marginata from the USA (Table A.2).  Microsatellites  The relatedness among the morphologically atypical individuals was low (mean -0.118) as estimated from the microsatellite genotypic data. One potential parent-offspring or fullsibling pair (individuals 2-12 and 2-17) was identified, with a pairwise relatedness of 0.513. In addition, the one sampled juvenile turtle with native morphology and a non-native CR haplotype did exhibit a pairwise relatedness value (0.377) and multi-locus allele transmission pattern (data not shown) consistent with a parent-offspring relationship with one of the putatively non-native adults (individual 2-1), suggesting a hybrid origin.  98  Table A.2 Summary of putative status based on morphology, sex and age class are indicated where known, mtDNA CR haplotype and alleles at two sites in the c-mos oncogene  UK unknown Locality  Individual  Sex  Putative status  CR Haplotype  c-mos205  c-mos436  BC other (n=26) Burnaby Lake  -  -  bellii  B.1, B.2, B.3, S.1  G  C  2-2  F  bellii  B.2  G  C  2-3  F  bellii  B.2  G  C  2-5  F  bellii  B.2  G  C  2-6  F  bellii  B.2  G  C  3-2  UK  bellii  B.2  G  C  2-140  M  bellii  B.1  G  C  2-19  UK  bellii  B.1  G  C  2-27  M  bellii  B.1  G  C  2-7  F  bellii  B.1  G  C  2-8  F  bellii  B.1  G  C  3-5  UK  bellii  B.1  G  C  5-6  M  bellii  B.1  G  C  2-23  UK  Atypical  S.4  R  Y  2-1  F  Atypical  S.3  R  Y  2-12  F  Atypical  S.3  R  Y  2-13  F  Atypical  S.3  G  C  2-14  F  Atypical  S.3  R  Y  2-15  F  Atypical  S.3  G  C  2-17  UK  Atypical  S.3  R  Y  2-18  M  Atypical  S.3  R  Y  2-24  UK  Atypical  S.3  R  Y  2-25  Juvenile  bellii  S.3  R  Y  2-26  UK  Atypical  S.3  R  Y  2-4  F  Atypical  S.3  G  C  2-9  F  Atypical  S.3  USA  Ontario  G  C  dorsalis (n=5)  R  Y  marginata (n=3)  R  Y  bellii (n=20)  G  C  picta (n=9)  G  C  marginata (n=8)  G  C  99  A.4 Discussion  All morphologically atypical individuals are likely non-native in origin based on the genetic evidence, exhibiting a CR haplotype as well as c-mos alleles that are more representative of painted turtles from eastern North America. I was unable to infer the subspecies or locality of origin based on the genetic data, except that the recovered haplotypes and genotypes were not otherwise detected across the western range of C. p. bellii. The majority of non-native individuals are genetically unrelated, potentially indicating multiple introduction events from varied sources. Their presence in Burnaby Lake is possibly due to the release of unwanted pets as reported for other species of turtle elsewhere in the lower mainland (Bunnell 2005), although the mode of introduction at Burnaby Lake is currently unknown.  The non-native individuals have been observed nesting at Burnaby Lake, but it was previously unknown whether the offspring were progeny of two non-natives or whether they were hybrids. Here, I show genetic evidence of hybridization between the introduced nonnatives and the local population of C. p. bellii. The only juvenile turtle that was sampled is a hybrid exhibiting high relatedness and an allele transmission pattern consistent with parentoffspring relationship with individual 2-1, indicating that some level of genetic introgression is present in this population. There is no reason to believe that hybrid offspring will be sterile or otherwise non-viable, as C. p. belli and the other subspecies form hybrid swarms where their ranges overlap (Weller et al. 2010). Yet, hybrid offspring may be less fit if they possess genes that are not adapted to their local environment indicative of outbreeding depression (Allendorf et al. 2001). Chrysemys subspecies do naturally hybridize where their ranges overlap, however, individuals involved in the crosses originate from the same geographical 100  area and ecological conditions. Since the non-native individuals at Burnaby Lake likely originated from eastern North America, they may possess gene variants maladapted in BC. Frankham et al. (2011) provide a decision tree for predicting whether outbreeding depression is likely to occur when populations interbreed. In the case of Burnaby Lake, this decision tree would recommend that the non-native and native individuals be kept from interbreeding since outbreeding depression is predicted to occur, as the populations from which the native and non-native turtles originated have been isolated without gene flow for at least 500 years.  The instance of hybridization detected in this study resulted in an individual that had native morphology, which indicates that genetic testing may be required to identify hybrid individuals. Fully half of the individuals sampled in this study were found to be non-native, indicating that, at the east end of Burnaby Lake at least, non-native individuals make up a high proportion of all painted turtles. No non-native turtles were observed or sampled at the west end of the lake, although turtle density appears to be very low in that area (E. Jensen, personal observation). This study is limited by the number of samples that were available, and the demographic groups represented in the current sample. Further study, ideally including a large number of juvenile and hatchling individuals, is necessary to determine the extent of introgression in the Burnaby Lake population of western painted turtles. Here, I present evidence that hybridization has occurred, but without more samples from additional generations of turtles, it is difficult to estimate the proportion of hybrid individuals and the level of introgression in the population. The development of additional nuclear genetic markers that have the power to diagnose non-native individuals would be needed to positively identify hybrid individuals with native mothers and non-native fathers. The  101  development of such markers is challenging because, although C. p. bellii, C. p. marginata, C. p. picta and C. p. dorsalis are recognized as subspecies, no fixed differences at nuclear markers have been identified so far and there is debate as to whether each subspecies actually exists as an evolutionary lineage separate from the others (see Chapter 3).  If the view is taken that the presence of non-native alleles is undesirable in the Burnaby Lake population of C. p. bellii, then intensive management over a long period of time will be required to preserve the native gene pool. Initially, the removal of non-native and confirmed hybrid individuals would be required, with genetic testing of individuals for non-native alleles on-going for years or potentially decades afterwards given the potential for sperm storage in this species (Pearse et al. 2001). In order to confirm whether individuals are hybrids, genetic testing will be required as the hybrid detected in this study was morphologically identified as native. Even if an individual displays an atypical morphology, it may simply be displaying natural and native variation rather than a hybrid phenotype, as there is considerable morphological variation among native western painted turtles, particularly in plastron patterning, dorsal stripes and carapace colouration and patterning (E. Jensen, personal observation).  Additional introduction events of non-native individuals could occur in the future as the source of the non-native individuals at Burnaby Lake has not yet been discovered or ameliorated. Monitoring of lakes across the BC lower mainland, including Burnaby Lake, should be on-going to identify sources of introduction events and their potentially impacts on the viability and genetic integrity of this listed species within the region.  102  Appendix B : Material supplemental to Chapter 2  B.1 Tables Table B.1 Details about the ecological drainage unit, faunal province (IM- Intermountain, RM- Rocky Mountain, PC- Pacific Coastal), designatable unit (IMRM Intermountain-Rocky Mountain, PC Pacific Coastal) and coordinates for each site and the mitochondrial control region haplotypes identified at each Abbreviation  Sampling Site  Faunal Province IM  Designatable Unit IM-RM  Latitude  Longitude  Baird Lake  Ecological Drainage Unit #10 Thompson  50.5692  -118.7928  mtDNA Haplotypes 2  BA BE  Beaver Lake  #31 Vancouver Island  PC  PC  48.5138  -123.3934  1  BL  Burnell Lake  #5 Okanagan  IM  IM-RM  49.208  -119.612  1  CV  # 4 Lower Kootenay  RM  IM-RM  49.123  -116.6299  1  DO  Creston Valley Wildlife Management Area Dorothy Lake  #7 Upper Columbia  RM  IM-RM  50.4984  -116.0257  1  EL  Elizabeth Lake  #8 Upper Kootenay  RM  IM-RM  49.4978  -115.793  1, 3  ER  Erie Lake  RM  IM-RM  49.1895  -117.3479  1  JO  Johnson Lake  #1 Columbia - Arrow Lakes #8 Upper Kootenay  RM  IM-RM  49.9562  -115.7647  3  KL  Klein Lake  #29 South Coastal  PC  PC  49.7294  -123.9699  1  LO  Loon Lake  #8 Upper Kootenay  RM  IM-RM  49.1132  -115.1062  1  NI  Nicomen Slough  #24 Lower Fraser  PC  PC  49.183  -122.114  1  OF  Fipke Lake near Oyama  #5 Okanagan  IM  IM-RM  50.1323  -119.3558  1  PA  Port Alberni  #31 Vancouver Island  PC  PC  49.2426  -124.827  1  RE  Redlich Pond  #5 Okanagan  IM  IM-RM  49.8945  -119.4616  1  RV  Revelstoke Marsh  IM  IM-RM  50.9612  -118.1805  1, 2  RO  Rosebud Lake  RM  IM-RM  49.048  -117.2678  4  RU  Ruby Lake Lagoon  #1 Columbia - Arrow Lakes #1 Columbia - Arrow Lakes #29 South Coastal  PC  IM-RM  49.724  -123.9917  1  SI  Scout Island Nature Centre  #9 Middle Fraser  IM  IM-RM  52.1186  -122.1164  1  SC  Stephen Coyote Regional Park  #5 Okanagan  IM  IM-RM  49.9579  -119.4394  1  103  Abbreviation  Sampling Site  Faunal Province IM  Designatable Unit IM-RM  Latitude  Longitude  Shannon Lake  Ecological Drainage Unit #5 Okanagan  49.8563  -119.612  mtDNA Haplotypes 1  SH SK  Skmana Lake  #10 Thompson  IM  IM-RM  50.8788  -119.7292  1  SW  Swan Lake  #31 Vancouver Island  PC  PC  48.4636  -123.3733  1  TE  Texada Island  #29 South Coastal  PC  PC  49.7462  -124.545  1  TI  Tie Lake  #8 Upper Kootenay  RM  IM-RM  49.4148  -115.3095  1  WE  West Lake  #29 South Coastal  PC  PC  49.7322  -124.0851  1  104  Table B.2 Characteristics of the 10 microsatellite loci used in this study Locus  Primer Sequences (5’-3’)  Repeat Unit  CpGT108  CCTAGAAAGTAAGAACCAATTTCAG  PCR Method TD59-51  M13 Label VIC  Source  (CA)4CT(CA)11  Allele Size Range (bp) 230-336  (GT)31(GC)5  193-245  TD59-51  PET  Goncalves da Silva et al. (2009)  GT  212-246  TD59-51  FAM  Pearse et al. (2001)  GT  154-192  TD55-45  NED  Pearse et al. (2001)  AGAT  172-244  TD59-51  PET  Hauswaldt and Glenn (2003)  CAAA  291-323  TD55-45  FAM  Hauswaldt and Glenn (2003)  AGAT  117-165  TD59-51  VIC  Hauswaldt and Glenn (2003)  ATCT  166-226  TD55-45  VIC  King and Julian (2004)  ATCT  234-286  TD59-51  NED  King and Julian (2004)  ATCT  171-215  TD59-51  PET  King and Julian (2004)  Goncalves da Silva et al. (2009)  CCACCAACAGAAGGAAGTTAGTG CpGT124  TCGGGGAGCACACTATACC CTCAGCCCCAAAATGAAC  Cp2  CTCTAAGGGTTGCACTTCTCAAA GAGGTGGCATCAAAACATCAT  Cp3  ATCTTTAAGTCTGTGAACTTCAGGG CTGTCTCATGCAAAGCTGGTAG  TerpSH2  TGGCCAGCAGGAGTAATG CTATTAGGGCAGAGACGAG  TerpSH3  TCCCCCAATGCACAC CTGCCCAATCCATTTAGA  TerpSH7  CACACACACTGTATTTTGATA CTATGCCCTTTCTAGTTTG  GmuD21  GCAGTTAGGCATTACTCAACATC AGGGTATGAATACAGGGGTGTC  GmuD28  AGCTGTTTGTCATCATACACTCTC TGGCCCTCATGTTTTATAAGTG  GmuD62  GGTGGTATAGAAAATCCTAAAATGG GTGCAAACTGTCTGGAAATAGG  105  Table B.3 Genbank accession  HBS23513  JQ963686.1  3  HBS26917  JQ963759.1  11  numbers and haplotype for  HBS23517  JQ963687.1  3  HBS26918  JQ963760.1  11  HBS23533  JQ963688.1  3  HBS26919  JQ963761.1  11  HBS23534  JQ963689.1  3  HBS26920  JQ963762.1  11  HBS23536  JQ963690.1  3  HBS28625  JQ963864.1  13  HBS23540  JQ963691.1  9  HBS28644  JQ963865.1  12  HBS23545  JQ963692.1  3  HBS28654  JQ963866.1  12  HBS23548  JQ963693.1  3  HBS28177  JQ963848.1  3  HBS23590  JQ963694.1  3  HBS28185  JQ963849.1  3  HBS23599  JQ963695.1  3  HBS28205  JQ963850.1  3  HBS23602  JQ963696.1  3  HBS28256  JQ963851.1  3  HBS23610  JQ963697.1  3  HBS28263  JQ963852.1  3  HBS23616  JQ963698.1  10  HBS27429  JQ963793.1  6  HBS23625  JQ963699.1  3  HBS27448  JQ963794.1  1  HBS27866  JQ963827.1  3  HBS27461  JQ963795.1  1  HBS28515  JQ963857.1  3  HBS27466  JQ963796.1  1  HBS27370  JQ963789.1  3  HBS27476  JQ963797.1  5  HBS27373  JQ963791.1  3  HBS27486  JQ963798.1  1  HBS27760  JQ963825.1  3  HBS27495  JQ963799.1  1  HBS28273  JQ963853.1  3  HBS27502  JQ963800.1  1  HBS28277  JQ963854.1  1  HBS27513  JQ963801.1  1  HBS28296  JQ963855.1  1  HBS29249  JQ963868.1  1  HBS27748  JQ963821.1  3  HBS29255  JQ963869.1  1  HBS27752  JQ963822.1  3  HBS23630  JQ963700.1  3  HBS27753  JQ963823.1  3  HBS23652  JQ963702.1  3  HBS27754  JQ963824.1  3  HBS23654  JQ963703.1  3  HBS27770  JQ963826.1  3  HBS23670  JQ963706.1  3  HBS27044  JQ963768.1  3  HBS23674  JQ963708.1  3  HBS27045  JQ963769.1  3  HBS23675  JQ963709.1  3  HBS27047  JQ963770.1  3  HBS26939  JQ963767.1  3  HBS27048  JQ963771.1  3  HBS28668  JQ963867.1  3  HBS26328  JQ963754.1  11  HBS26330  JQ963755.1  11  each sequence used in the haplotype analysis  Voucher HBS27532  Accession # JQ963802.1  Haplotype 3  HBS27533  JQ963803.1  3  HBS28092  JQ963838.1  3  HBS28097  JQ963839.1  3  HBS28101  JQ963840.1  3  HBS28134  JQ963844.1  3  HBS28138  JQ963845.1  3  HBS28146  JQ963846.1  3  HBS28171  JQ963847.1  3  HBS28603  JQ963861.1  3  HBS28610  JQ963862.1  14  HBS28620  JQ963863.1  14  HBS28535  JQ963859.1  3  HBS28303  JQ963856.1  1  HBS26294  JQ963749.1  3  HBS26298  JQ963751.1  3  HBS28109  JQ963841.1  7  HBS28111  JQ963842.1  3  HBS28113  JQ963843.1  3  HBS23506  JQ963679.1  3  HBS23507  JQ963680.1  3  HBS23508  JQ963681.1  3  HBS23509  JQ963682.1  3  HBS23510  JQ963683.1  8  HBS23511  JQ963684.1  3  HBS23512  JQ963685.1  8  106  Table B.4 Migration rates among the 25 sites. The direction of migration is from the site below the diagonal to the site above the diagonal (e.g. migration from BE to BA is 0.02) BA BA  BE  BL  CV  DO  EL  ER  JO  KL  LO  NI  OF  PA  RE  RV  RO  RU  SI  SC  SH  SK  SW  TE  TI  WE  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.07  0.01  0.01  0.01  0.04  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.09  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.01  0.01  0.01  0.12  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.04  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.14  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.01  0.01  0.03  0.03  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.04  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  BE  0.02  BL  0.01  0.01  CV  0.01  0.01  0.01  DO  0.01  0.01  0.01  0.01  EL  0.01  0.01  0.01  0.01  0.01  ER  0.01  0.01  0.02  0.01  0.03  0.01  JO  0.01  0.01  0.01  0.01  0.01  0.02  0.01  KL  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  LO  0.01  0.01  0.01  0.01  0.03  0.01  0.01  0.07  0.01  NI  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  OF  0.03  0.01  0.04  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  PA  0.01  0.01  0.03  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  RE  0.05  0.01  0.03  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  RV  0.08  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  RO  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  RU  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  SI  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  SC  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  SH  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  SK  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.01  0.01  0.01  SW  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.02  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  TE  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  TI  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.15  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  WE  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01  0.01 0.01  107  B.2 Sites  Some of the sites where I sampled require further comment. Below are explanations of why I combined samples from separate water bodies into a single site for analysis and why some sites where samples were collected were excluded from the analyses in Chapter 2.  Baird and Hidden Lake  As indicated in Chapter 2, this site actually consists of two discrete water bodies: Baird Lake and Hidden Lake. These lakes are only 1.7 km apart as the bird flies, and there is a possibility of a stream connecting the two. In 2011 I sampled at Baird Lake but collected only seven samples. In 2012 I returned but did not observe many turtles and managed to catch only two more. Sampling at Hidden Lake in 2012 was more successful, resulting in the collection of 11 blood samples.  The finding in Chapter 2 that even nearby sites are demographically independent would seem to suggest that combining samples from Baird and Hidden Lakes into a single population for analysis is unjustified; however, it is highly doubtful that the major findings of my research would be altered by combining the sites for analysis purposes. No significant heterozygote deficit (i.e. a Wahlund effect, Wahlund 1928) was found to suggest that unrecognized population subdivision was present when the sites were combined.  108  Texada Island  As indicated in Chapter 2, this site actually consists of several discrete water bodies: Capsheaf Lake (n=3), Emily Lake (n=13), Priest Lake (n=1) and Case Lake (n=2). Considerations are similar as described for Baird and Hidden Lake above.  Miller Pond  This site near Clearwater (latitude 51.6554, longitude -120.0247) was sampled in 2012. The site consisted of a pond on private land that had a population of at least 150 western painted turtles. After collecting blood samples from 20 individuals we interviewed the land owner and were told that his family, which had lived on that property for four generations, had introduced turtles to that pond in the 1960’s. He recalled that the turtles had originated from Stump Lake, near Kamloops, and had since spread from his property to nearby lakes, such as Dutch Lake in the town of Clearwater where western painted turtles had not previously been. Due to this anecdotal evidence that the turtles may have been introduced, this site was excluded from the analyses presented in Chapter 2. Preliminary site level analyses were performed, however, which did not indicate any of the genetic signatures typically associated with a recent founder event, calling into question the anecdotal evidence of a recent introduction.  109  Manly Oxbow  This site consists of a series of sloughs along a secondary road near Grand Forks (latitude 49.0067, longitude -118.3632). I sampled 20 individuals in 2012. Preliminary analyses indicated heterozygote deficit at this site, which is an indication that individuals that were introduced to that site were included in this sample; for this reason Manly Oxbow was excluded from the analyses in Chapter 2.  Bruning Creek  In 2012, Julie Steciw collected 12 road killed individuals from a 200 m stretch of Highway 97, near 150 Mile House (latitude 52.091504, longitude -121.920879). DNA was extracted from the tissue samples; however, it was of poor quality. I attempted to genotype the ten microsatellite loci for those individuals, but I had low confidence in the alleles called and the site was excluded from the analyses in Chapter 2.  110  Appendix C : Material supplemental to Chapter 3  C.1 Tables  Table C.1 Details about individuals used in each analysis X or haplotype number, included in analysis; blank, not included in analysis Voucher #  State  County  SI.01  Putative ssp. bellii  BC  Cariboo  CR Haplotype B.2  SI.12  bellii  BC  Cariboo  CV.01  bellii  BC  CV.12  bellii  CV.17 DO.01  CR Accession #  PAX-P1 Haplotype 1 S.2  PAX-P1 Haplotype 2  B.2  S.3  S.2  Kootenay  B.2  B.2  S.2  BC  Kootenay  B.2  S.3  S.2  bellii  BC  Kootenay  B.2  B.3  bellii  BC  Kootenay  B.2  B.4  DO.06  bellii  BC  Kootenay  B.2  B.4  EL.01  bellii  BC  Kootenay  S.1  S.2  EL.12  bellii  BC  Kootenay  B.2  S.2  ER.01  bellii  BC  Kootenay  B.2  B.2  ER.06  bellii  BC  Kootenay  B.2  B.2  GR.01  bellii  BC  Kootenay  B.2  B.5  GR.02  bellii  BC  Kootenay  B.2  S.2  JO.01  bellii  BC  Kootenay  S.1  B.4  JO.38  bellii  BC  Kootenay  S.1  S.2  LO.01  bellii  BC  Kootenay  B.2  B.4  LO.12  bellii  BC  Kootenay  B.2  B.4  LO.29  bellii  BC  Kootenay  B.2  S.2  Concatenated Tree  CR Tree  PAX Tree  X  X  X  X X  S.2 X  X  X  X  S.2  111  Voucher #  State  County  RO.01  Putative ssp. bellii  BC  Kootenay  CR Haplotype B.2  RO.12  bellii  BC  Kootenay  B.3  RO.34  bellii  BC  Kootenay  NI.01  bellii  BC  NI.29  bellii  NI.35  bellii  SL.37  CR Accession #  PAX-P1 Haplotype 1  PAX-P1 Haplotype 2  Concatenated Tree  CR Tree  PAX Tree  S.2  X  X  X  B.3  B.3  X  X  Lower Mainland  B.2  S.2  BC  Lower Mainland  B.2  S.3  BC  Lower Mainland  B.2  S.2  bellii  BC  Lower Mainland  S.1  B.4  X  X  BA.01  bellii  BC  Okanagan  B.1  S.2  X  BA.10  bellii  BC  Okanagan  B.1  S.2  BA.801  bellii  BC  Okanagan  B.1  S.3  S.2  BU.01  bellii  BC  Okanagan  B.2  S.3  S.2  BU.10  bellii  BC  Okanagan  B.2  S.3  S.2  BU.19  bellii  BC  Okanagan  B.2  S.2  OF.27  bellii  BC  Okanagan  B.2  S.3  S.2  OF.29  bellii  BC  Okanagan  B.2  S.3  S.2  OF.30  bellii  BC  Okanagan  B.2  B.2  RE.01  bellii  BC  Okanagan  B.2  B.2  S.2  RE.03  bellii  BC  Okanagan  B.2  S.3  S.2  RE.04  bellii  BC  Okanagan  B.2  S.2  KL.34.01  bellii  BC  Sunshine Coast  B.2  S.2  KL.34.12  bellii  BC  Sunshine Coast  B.2  B.2  KL.34.16  bellii  BC  Sunshine Coast  B.2  S.2  TI.113  bellii  BC  Sunshine Coast  B.2  S.3  B.2  TI.23  bellii  BC  Sunshine Coast  B.2  TI.31  bellii  BC  Sunshine Coast  B.2  S.3  S.2  WE.3137  bellii  BC  Sunshine Coast  B.2  S.2  WE.318  bellii  BC  Sunshine Coast  B.2  B.2  S.2  WE.34  bellii  BC  Sunshine Coast  B.2  B.2  S.2  S.2  S.2  X  X  X  X  X  X  112  Voucher #  State  County  RV.32  Putative ssp. bellii  BC  Thompson  CR Haplotype B.2  CR Accession #  PAX-P1 Haplotype 1 S.2  RV.85  bellii  BC  Thompson  B.1  S.2  RV.98  bellii  BC  Thompson  B.2  S.2  SK.01  bellii  BC  Thompson  B.2  S.2  SK.12  bellii  BC  Thompson  B.2  S.3  SK.25  bellii  BC  Thompson  B.2  S.2  BE.136  bellii  BC  Vancouver Island  B.2  S.2  BE.137  bellii  BC  Vancouver Island  B.2  S.3  BE.17  bellii  BC  Vancouver Island  B.2  S.2  PA.24  bellii  BC  Vancouver Island  B.2  S.2  PA.28  bellii  BC  Vancouver Island  B.2  S.2  PA.36  bellii  BC  Vancouver Island  B.2  S.2  HBS27532  bellii  CO  Douglas  S.1  JQ963802.1  HBS27533  bellii  CO  Douglas  S.1  JQ963803.1  HBS28603  bellii  CO  La Plata  S.1  JQ963861.1  HBS28610  bellii  CO  La Plata  B.13  JQ963862.1  B.3  HBS28620  bellii  CO  La Plata  B.13  JQ963863.1  B.3  X  HBS28097  bellii  CO  Morgan  S.1  JQ963839.1  HBS28101  bellii  CO  Morgan  S.1  JQ963840.1  HBS28531  bellii  IA  Hamilton  B.12  B.6  X  HBS28535  bellii  IA  Hamilton  S.1  JQ963859.1  HBS28303  bellii  ID  Boundary  B.2  JQ963856.1  S.2  X  HBS26298  bellii  IL  Carroll  S.1  JQ963751.1  HBS26301  bellii  IL  Carroll  B.7  JQ963752.1  HBS26322  bellii  IL  Carroll  S.4  JQ963753.1  HBS28109  bellii  KS  Berton  B.11  JQ963841.1  S.2  HBS28111  bellii  KS  Berton  S.1  JQ963842.1  B.7  HBS28113  bellii  KS  Berton  S.1  JQ963843.1  S.2  PAX-P1 Haplotype 2  Concatenated Tree  CR Tree  X  PAX Tree  X  S.2 X  X  X  S.2  B.6  X  X  X  X X X  X  X X  X X S.1  X  X  X X  113  Voucher #  State  County  HBS28134  Putative ssp. bellii  Sherman  CR Haplotype S.1  CR Accession # JQ963844.1  KS  HBS23534  bellii  MN  Atikin  S.1  JQ963689.1  HBS23590  bellii  MN  Brown  S.1  JQ963694.1  HBS23599  bellii  MN  Brown  S.1  JQ963695.1  HBS23536  bellii  MN  Crow Wing  S.1  JQ963690.1  HBS23540  bellii  MN  Crow Wing  B.5  JQ963691.1  HBS23616  bellii  MN  Houston  B.6  JQ963698.1  HBS23625  bellii  MN  Houston  S.1  JQ963699.1  HBS23517  bellii  MN  Itasca  S.1  JQ963687.1  HBS23533  bellii  MN  Itasca  S.1  JQ963688.1  HBS23545  bellii  MN  Nicollet  S.1  JQ963692.1  HBS23548  bellii  MN  Nicollet  S.1  JQ963693.1  HBS23602  bellii  MN  Nicollet  S.1  JQ963696.1  HBS23610  bellii  MN  Nicollet  S.1  JQ963697.1  HBS23506  bellii  MN  St Louis  S.1  JQ963679.1  HBS23507  bellii  MN  St Louis  S.1  JQ963680.1  HBS23508  bellii  MN  St Louis  S.1  JQ963681.1  HBS23509  bellii  MN  St Louis  S.1  JQ963682.1  HBS23510  bellii  MN  St Louis  B.4  JQ963683.1  HBS23511  bellii  MN  St Louis  S.1  JQ963684.1  HBS23512  bellii  MN  St Louis  B.4  JQ963685.1  HBS23513  bellii  MN  St Louis  S.1  JQ963686.1  HBS27866  bellii  MN  Wabasha  S.1  JQ963827.1  HBS28515  bellii  MN  Wabasha  S.1  JQ963857.1  HBS28518  bellii  MN  Wabasha  S.4  JQ963858.1  HBS27370  bellii  MO  Boone  S.1  JQ963789.1  HBS27372  bellii  MO  Boone  B.7  JQ963790.1  HBS27373  bellii  MO  Boone  S.1  JQ963791.1  PAX-P1 Haplotype 1 S.2  PAX-P1 Haplotype 2 B.6  Concatenated Tree X  CR Tree  X  X  PAX Tree X  X  X  B.6  B.9  X  X  X  114  Voucher #  State  County  HBS27376  Putative ssp. bellii  Boone  CR Haplotype B.7  CR Accession # JQ963792.1  MO  HBS28273  bellii  MT  Big Horn  S.1  JQ963853.1  HBS28277  bellii  MT  Lewis and Clark  B.2  JQ963854.1  HBS28092  bellii  MT  Lincoln  S.1  JQ963838.1  HBS28296  bellii  MT  Sanders  B.2  JQ963855.1  HBS27760  bellii  MT  Sheridan  S.1  JQ963825.1  HBS27770  bellii  ND  Kidder  S.1  JQ963826.1  HBS27748  bellii  ND  Rolette  S.1  JQ963821.1  HBS27752  bellii  ND  Rolette  S.1  JQ963822.1  HBS27753  bellii  ND  Rolette  S.1  JQ963823.1  HBS27754  bellii  ND  Rolette  S.1  JQ963824.1  HBS28668  bellii  NE  Cherry  S.1  JQ963867.1  HBS27044  bellii  NE  Sand Hills  S.1  JQ963768.1  HBS27045  bellii  NE  Sand Hills  S.1  JQ963769.1  HBS27047  bellii  NE  Sand Hills  S.1  JQ963770.1  HBS27048  bellii  NE  Sand Hills  S.1  JQ963771.1  HBS28644  bellii  NM  Apache  B.15  JQ963865.1  HBS28654  bellii  NM  Apache  B.15  JQ963866.1  HBS28625  bellii  NM  San Juan  B.14  JQ963864.1  HBS26917  bellii  NM  Sierra  B.8  JQ963759.1  HBS26918  bellii  NM  Sierra  B.8  JQ963760.1  HBS26919  bellii  NM  Sierra  B.8  JQ963761.1  HBS26920  bellii  NM  Sierra  B.8  JQ963762.1  HBS26328  bellii  NM  Socorro  B.8  JQ963754.1  HBS26330  bellii  NM  Socorro  B.8  JQ963755.1  HBS28177  bellii  SD  Bennet  S.1  JQ963848.1  HBS28185  bellii  SD  Bennet  S.1  JQ963849.1  HBS28205  bellii  SD  Bennet  S.1  JQ963850.1  PAX-P1 Haplotype 1 S.2  PAX-P1 Haplotype 2 B.6  S.2  S.2  B.6  B.1 S.3  B.2  Concatenated Tree X  CR Tree  PAX Tree X  X  X  X  X  X  X  X  X  X  X  X S.2  X  X  115  Voucher #  State  County  HBS28138  Putative ssp. bellii  Charles Mix  CR Haplotype S.1  CR Accession # JQ963845.1  SD  HBS28146  bellii  SD  HBS28171  bellii  HBS28256  PAX-P1 Haplotype 1  Charles Mix  S.1  JQ963846.1  SD  Charles Mix  S.1  JQ963847.1  bellii  SD  Meade  S.1  JQ963851.1  HBS28263  bellii  SD  Meade  S.1  JQ963852.1  HBS29249  bellii  WA  Clark  B.2  JQ963868.1  B.2  HBS27461  bellii  WA  Grant  B.2  JQ963795.1  S.2  HBS27466  bellii  WA  Grant  B.2  JQ963796.1  HBS27476  bellii  WA  Grant  B.10  JQ963797.1  HBS27495  bellii  WA  Okanogan  B.2  JQ963799.1  HBS27486  bellii  WA  Okawogan  B.2  JQ963798.1  HBS29255  bellii  WA  Skamawia  B.2  JQ963869.1  HBS27502  bellii  WA  Spokane  B.2  JQ963800.1  HBS27513  bellii  WA  Spokane  B.2  JQ963801.1  HBS27429  bellii  WA  Yakima  B.9  JQ963793.1  HBS27448  bellii  WA  Yakima  B.2  JQ963794.1  HBS23630  bellii  WI  Chippewa  S.1  JQ963700.1  HBS23635  bellii  WI  Chippewa  S.4  JQ963701.1  HBS23652  bellii  WI  Chippewa  S.1  JQ963702.1  HBS26245  bellii  WI  La Crosse  S.4  JQ963742.1  HBS26939  bellii  WI  La Crosse  S.1  JQ963767.1  HBS23656  bellii  WI  Lincoln  JQ963704.1  HBS27284  dorsalis  AL  Marshall  JQ963787.1  D.1  HBS23299  dorsalis  AR  Lonoke  D.2  JQ963670.1  S.1  HBS23319  dorsalis  AR  Lonoke  D.2  JQ963671.1  HBS23324  dorsalis  AR  Lonoke  D.2  JQ963672.1  HBS23169  dorsalis  IL  Alexander  D.1  JQ963656.1  HBS23176  dorsalis  IL  Alexander  D.6  JQ963663.1  PAX-P1 Haplotype 2  Concatenated Tree  CR Tree  PAX Tree  B.2  X  X  X  X  X B.9  B.2 X  X  D.3  D.2  X  X  X  X  X  X  X  X  X  X  X  X  X  116  Voucher #  State  County  HBS26210  Putative ssp. dorsalis  Concordial  CR Haplotype D.3  CR Accession # JQ963733.1  LA  HBS26211  dorsalis  LA  HBS26212  dorsalis  HBS26921  PAX-P1 Haplotype 1  PAX-P1 Haplotype 2  Concatenated Tree X  CR Tree X  PAX Tree  Concordial  D.2  JQ963734.1  LA  Concordial  D.4  JQ963735.1  X  X  X  dorsalis  LA  Concordial  D.2  JQ963763.1  HBS27641  dorsalis  LA  Concordial  D.5  JQ963816.1  HBS27651  dorsalis  LA  Concordial  D.1  JQ963817.1  HBS31496  dorsalis  LA  Iberville  D.2  JQ963870.1  HBS31533  dorsalis  LA  St Martin  D.2  JQ963872.1  HBS28542  marginata  IN  Boone  S.1  JQ963860.1  X  HBS23366  marginata  IN  Kosciusko  S.4  JQ963673.1  X  HBS23395  marginata  IN  Kosciusko  S.4  JQ963674.1  HBS31531  marginata  IN  Marion  S.4  JQ963871.1  HBS27134  marginata  MI  Cheboygan  S.4  JQ963775.1  HBS27135  marginata  MI  Cheboygan  S.4  JQ963776.1  HBS27158  marginata  MI  Lenawee  S.4  JQ963780.1  HBS27160  marginata  MI  Lenawee  M.11  JQ963781.1  HBS27161  marginata  MI  Lenawee  M.12  JQ963782.1  HBS27141  marginata  MI  Mecosta  S.4  JQ963777.1  HBS27147  marginata  MI  Mecosta  S.4  JQ963778.1  HBS27151  marginata  MI  Mecosta  S.4  JQ963779.1  HBS27109  marginata  MI  Schoolcraft  S.4  JQ963772.1  HBS27117  marginata  MI  Schoolcraft  S.4  JQ963773.1  HBS27124  marginata  MI  Schoolcraft  M.10  JQ963774.1  HBS26252  marginata  MI  Shiawassee  S.4  JQ963743.1  HBS26256  marginata  MI  Shiawassee  S.4  JQ963744.1  HBS26260  marginata  MI  Shiawassee  S.4  JQ963745.1  HBS26261  marginata  MI  Shiawassee  S.4  JQ963746.1  HBS26267  marginata  MI  Shiawassee  S.4  JQ963747.1  X  X  S.6  X  S.3  X  S.3  X  X  X  X  X  X  X  X  X  117  Voucher #  State  County  HBS26293  Putative ssp. marginata  Shiawassee  CR Haplotype S.4  CR Accession # JQ963748.1  MI  HBS26353  marginata  MI  HBS28001  marginata  HBS26092  PAX-P1 Haplotype 1  PAX-P1 Haplotype 2  Concatenated Tree  CR Tree  PAX Tree  Shiawassee  M.9  JQ963756.1  NY  Clinton  S.2  JQ963828.1  marginata  NY  Madison  M.4  JQ963726.1  HBS26065  marginata  NY  Oneida  M.3  JQ963718.1  HBS26068  marginata  NY  Oneida  M.4  JQ963719.1  HBS26069  marginata  NY  Oneida  M.2  JQ963720.1  HBS26081  marginata  NY  Onondaga  M.2  JQ963721.1  HBS26083  marginata  NY  Onondaga  M.2  JQ963722.1  HBS26086  marginata  NY  Oswego  M.3  JQ963723.1  HBS26090  marginata  NY  Oswego  M.5  JQ963724.1  HBS26091  marginata  NY  Oswego  M.2  JQ963725.1  HBS26061  marginata  NY  Rochester  M.2  JQ963717.1  HBS27221  marginata  OH  Pike  M.14  JQ963785.1  HBS27227  marginata  OH  Pike  M.2  JQ963786.1  HBS27179  marginata  OH  Shelby  S.4  JQ963783.1  HBS27185  marginata  OH  Shelby  M.13  JQ963784.1  X  X  X  Ont.3  marginata  Ont  Frontenac  M.1  M.3  X  X  X  Ont.4  marginata  Ont  Frontenac  S.2  M.4  X  X  X  Ont.5  marginata  Ont  Frontenac  S.2  S.5  Ont.6  marginata  Ont  Frontenac  S.2  Ont.7  marginata  Ont  Frontenac  S.2  Ont.8  marginata  Ont  Frontenac  S.2  M.5  HBS26108  marginata  TN  Anderson  M.6  JQ963729.1  HBS26112  marginata  TN  Anderson  M.7  JQ963730.1  S.5  X  X  X  HBS26113  marginata  TN  Anderson  M.8  JQ963731.1  S.7  X  X  X  HBS26114  marginata  TN  Anderson  M.7  JQ963732.1  HBS28015  marginata  VT  Caledonia  M.15  JQ963829.1  S.8  X  X  X  X X  X X  S.4  X  X  X  S.8  X  X  X  X  X  X  X  X  X  S.5  M.1  M.2  X  M.3  X  X  X  M.5  X  X  M.1  X  X X  S.3  118  Voucher #  State  County  HBS28016  Putative ssp. marginata  Caledonia  CR Haplotype S.3  CR Accession # JQ963830.1  VT  HBS26213  picta  GA  HBS26214  picta  HBS26215  PAX-P1 Haplotype 1 M.6  PAX-P1 Haplotype 2 M.4  Concatenated Tree X  CR Tree X  Jackson  P.1  JQ963736.1  GA  Jackson  P.1  JQ963737.1  picta  GA  Jackson  P.1  JQ963738.1  HBS26216  picta  GA  Jasper  P.1  JQ963739.1  HBS26217  picta  GA  Jasper  P.1  JQ963740.1  HBS26218  picta  GA  Jasper  P.1  JQ963741.1  HBS26356  picta  GA  Jasper  P.1  JQ963757.1  HBS27686  picta  GA  Macon  P.1  JQ963820.1  HBS27654  picta  GA  Monroe  P.1  JQ963818.1  HBS27675  picta  GA  Monroe  P.3  HBS27676  picta  GA  Monroe  P.3  JQ963819.1  X  HBS26038  picta  MA  Middlesex  P.4  JQ963713.1  X  HBS26040  picta  MA  Middlesex  S.3  JQ963714.1  HBS26042  picta  MA  Middlesex  P.5  JQ963715.1  HBS26058  picta  MA  Middlesex  S.2  JQ963716.1  HBS33050  picta  MA  Worcester  S.2  JQ963877.1  HBS33053  picta  MA  Worcester  S.2  JQ963878.1  HBS32992  picta  ME  Hancock  S.2  JQ963875.1  HBS33030  picta  ME  Hancock  S.2  JQ963876.1  HBS28035  picta  ME  Kennebec  S.2  HBS28041  picta  ME  Kennebec  S.2  HBS28044  picta  ME  Kennebec  S.2  JQ963835.1  HBS28049  picta  NH  Salem  S.2  JQ963836.1  HBS28051  picta  NH  Salem  S.3  JQ963837.1  HBS31658  picta  NJ  Sussex  S.2  JQ963873.1  HBS31678  picta  NJ  Sussex  S.2  JQ963874.1  HBS27575  picta  NY  Sufflolk  S.3  JQ963804.1  PAX Tree X  X S.4  X  X  S.5  X  X  P.1  X  X  S.6  X  S.2  X  X  JQ963833.1  X  X  JQ963834.1  X  X  X  X  X  X  P.2  S.3  X  X  119  Voucher #  State  County  HBS27577  Putative ssp. picta  CR Accession #  Sufflolk  CR Haplotype S.2  NY  HBS27578  picta  NY  HBS27579  picta  HBS27580  PAX-P1 Haplotype 1 S.5  PAX-P1 Haplotype 2  Concatenated Tree X  CR Tree  PAX Tree  Sufflolk  S.3  JQ963805.1  NY  Sufflolk  S.2  JQ963806.1  picta  NY  Sufflolk  P.6  JQ963807.1  HBS27581  picta  NY  Sufflolk  S.2  JQ963808.1  HBS100115  picta  PA  Bucks  HBS26358  picta  RI  Kent  S.2  JQ963758.1  HBS26923  picta  RI  Kent  S.3  JQ963764.1  HBS26924  picta  RI  Kent  S.2  JQ963765.1  HBS26925  picta  RI  Kent  P.2  JQ963766.1  S.5  HBS27583  picta  VA  Fairfax  S.3  JQ963809.1  S.7  HBS27585  picta  VA  Fairfax  P.7  JQ963810.1  HBS27590  picta  VA  Fairfax  P.7  JQ963811.1  HBS33304  picta  VA  Montgomery  S.3  JQ963879.1  HBS23676  MI  Dickinson  JQ963710.1  X  X  HBS23690  MI  Dickinson  JQ963712.1  X  X  HBS28030  NH  Sullivan  JQ963832.1  X  X  X  X S.6  X  X  S.4  X  X  X  X  X  X X S.8  X  X X  X X  X  X  120  Table C.2 Identity and source of outgroups Taxon  Voucher ID  E. marmorata  HBS39753  E. marmorata  HBS39816  P. megacephalum P. megacephalum  HBS16255  T. s .elegans  HSB23001  T. carolina  HSB27240  P. texana  RCT2  G. geographica  RCT94  CR Accession #  CR Source  AY904894  Spinks and Shaffer (2005)  DQ256377  Parham et al. (2006)  PAX Accession #  PAX Source  GU085659  Barley et al. (2010)  GU085663.1  Spinks et al. (2013)  P.Q. Spinks, Personal Communication  P.Q. Spinks, Personal Communication  KC687248  Spinks et al. (2013)  P.Q. Spinks, Personal Communication  KC687245  Spinks et al. (2013)  P.Q. Spinks, Personal Communication, Tissue from Robert Thompson P.Q. Spinks, Personal Communication, Tissue from Robert Thompson  P.Q. Spinks, Personal Communication, Tissue from Robert Thompson  121  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073968/manifest

Comment

Related Items